Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Disparate symmetries in head-to-tail Schiff-base macrocycles Chen, Zhengyu 2017

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

Item Metadata

Download

Media
24-ubc_2017_may_chen_zhengyu.pdf [ 8.43MB ]
Metadata
JSON: 24-1.0342813.json
JSON-LD: 24-1.0342813-ld.json
RDF/XML (Pretty): 24-1.0342813-rdf.xml
RDF/JSON: 24-1.0342813-rdf.json
Turtle: 24-1.0342813-turtle.txt
N-Triples: 24-1.0342813-rdf-ntriples.txt
Original Record: 24-1.0342813-source.json
Full Text
24-1.0342813-fulltext.txt
Citation
24-1.0342813.ris

Full Text

DISPARATE SYMMETRIES IN HEAD-TO-TAIL SCHIFF-BASE MACROCYCLES   by  Zhengyu Chen  B.Sc., Nanjing University, 2010  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Chemistry)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  February 2017  © Zhengyu Chen, 2017 ii  Abstract  Two new Schiff-base macrocycles (called campestarenes) with 5-fold symmetry were prepared with bulky triphenylsilyl and triisopropylsilyl substituents.  A single crystal structure of one campestarene showed an almost flat conformation of campestarenes are in their extreme enol-imine form. Tautomerization within the campestarene between the enol-imine and keto-enamine form was investigated by variable-temperature NMR and UV-vis spectroscopy. It was found that the molecule displays strong solvent- and temperature-dependent tautomerization that leads to large changes in color.  These results were supported by computational investigations that showed the possibility of tautomerization between keto-enamine form and enol-imine form. The experimental studies showed the relative permittivity of solvents has a large influence on the relative stability of different campestarene tautomers in solution. Some methyl/phenyl substituted campestarene precursors were prepared for macrocyclization of methyl/phenyl substituted campestarenes. Different synthetic conditions were tested to facilitate the formation of campestarens. One phenyl substituted campestarene was synthesized under acid condition. Two new Pt3 Schiff-base macrocycles with 3-fold symmetry were synthesized following a head-to-tail approach. Computational investigations showed an almost flat conformation of Pt3 macrocycles. Aggregation of Pt3 macrocycles in solid state was studied by MALDI-TOF, TEM and PXRD. It was found that Pt3 macrocycles displays nanotubular structures due to aggregation. Aggregation of Pt3 macrocycles in solution was investigated by variable-temperature 1D NMR and 2D NMR. The experimental results showed aggregation of Pt3 macrocycles at both high and low temperatures. iii  Preface  All the work in this thesis was carried out under the guidance of my supervisor, Prof. Mark J. MacLachlan. Chapter 2 involved collaboration with Prof. Francesco Lelj of the Dipartimento di Chimica, Università della Basilicata, Italy, who performed all of the computational studies described throughout this chapter. X-ray diffraction analyses were performed by Dr. Nicholas G. White of the Department of Chemistry, University of British Columbia. Peixi Wang, of the Department of Chemistry, University of British Columbia, collected the TEM images reported in Chapter 4.   Chapter 2: Portions of this chapter have been submitted for publication: Zhengyu Chen, Samuel Guieu, Nicholas G. White, Francesco Lelj and Mark J. MacLachlan, “The rich tautomeric behavior of campestarenes” Chem. Eur. J. 2016, in press. Prof. Francesco Lelj conducted all computational experiments that are discussed in this chapter. Dr. Nicholas G. White collected and analyzed the XRD data. I performed the other experiments reported in chapter 2.  Chapter 3: Fabian Beckmann conducted the phenyl substituted campestarenes experiments with Ti catalysts.  Chapter 4: Portions of chapter 4 will be submitted for publication: Zhengyu Chen and Mark J. MacLachlan, “Self-Assembly of Extended Head-to-Tail Pt3 Macrocycles into Nanotubes” (2016). Brian Sahli of the Department of Chemistry, University of British Columbia, conducted the iv  computational studies on the conformation of Pt3 macrocycles. Peixi Wang, of the Department of Chemistry, University of British Columbia, collected the TEM images reported in this chapter.  v  Table of Contents  Abstract .......................................................................................................................................... ii Preface ........................................................................................................................................... iii Table of Contents ...........................................................................................................................v List of Figures .................................................................................................................................x List of Symbols .......................................................................................................................... xxii List of Abbreviations ............................................................................................................... xxiv Acknowledgements ................................................................................................................. xxvii Dedication ............................................................................................................................... xxviii Chapter 1: Introduction ................................................................................................................1 1.1 Supramolecular Chemistry.............................................................................................. 1 1.1.1 Background ................................................................................................................. 1 1.1.2 Macrocycles ................................................................................................................ 3 1.1.3 Shape-Persistent Macrocycles .................................................................................... 7 1.2 Synthesis of Macrocycles ............................................................................................. 11 1.2.1 Intramolecular Ring Closure of Linear Oligomers ................................................... 11 1.2.2 Template Mediated Macrocycle Synthesis ............................................................... 13 1.2.3 One-Pot Synthesis of Macrocycles ........................................................................... 17 1.3 Head-to-Tail Macrocycles ............................................................................................ 24 1.3.1 Metal-Free Head-to-Tail Macrocycles ...................................................................... 24 1.3.1.1 Intramolecular Closure towards Head-to-Tail Macrocycles ............................. 24 1.3.1.2 Inspiration for One-Pot Synthesis of Head-to-Tail Macrocycles ..................... 26 vi  1.3.1.3 Hydrogen-Bonding-Assisted One-Pot Synthesis of Head-to-Tail Macrocycles ..   ........................................................................................................................... 29 1.3.1.4 One-Pot Synthesis of Head-to-Tail Macrocycles without Hydrogen Bonding 43 1.3.2 Head-to-Tail Metallomacrocycles ............................................................................ 45 1.4 Schiff-Base Macrocycles .............................................................................................. 52 1.4.1 Schiff-Base Chemistry--Salens and Salphens ........................................................... 52 1.4.2 Schiff-Base Macrocycles .......................................................................................... 53 1.5 Design of Schiff-Base Macrocycles.............................................................................. 56 1.5.1 Geometry of Macrocycles ......................................................................................... 56 1.5.2 [n + n] Schiff-Base Macrocycles .............................................................................. 57 1.6 Goals and Scope ............................................................................................................ 59 Chapter 2: The Rich Tautomeric Behavior of Campestarenes ...............................................63 2.1 Introduction ................................................................................................................... 63 2.1.1 Campestarene ............................................................................................................ 63 2.1.2 Tautomerization ........................................................................................................ 65 2.2 Results and Discussion ................................................................................................. 66 2.2.1 Synthesis and Characterization ................................................................................. 66 2.2.2 Solid-State Structure of Campestarene 1e ................................................................ 75 2.2.3 Solvent-Dependent Tautomerization ........................................................................ 81 2.2.4 Photophysical Behavior ............................................................................................ 92 2.2.5 Two or More Tautomers? ......................................................................................... 95 2.2.6 Variable Temperature Studies................................................................................. 100 2.3 Conclusions ................................................................................................................. 117 vii  2.4 Experimental ............................................................................................................... 118 2.4.1 Materials ................................................................................................................. 118 2.4.2 Equipment ............................................................................................................... 118 2.4.3 Procedure and Experimental Data .......................................................................... 119 2.4.4 Details of Crystallography ...................................................................................... 129 2.4.5 Computational Methods .......................................................................................... 131 Chapter 3: Synthesis of Methyl/Phenyl Substituted Campestarenes ....................................133 3.1 Introduction ................................................................................................................. 133 3.2 Results and Discussion ............................................................................................... 135 3.2.1 Synthesis of Methyl Substituted Campestarene ...................................................... 136 3.2.2 Synthesis of Phenyl Substituted Campestarenes..................................................... 140 3.3 Conclusion .................................................................................................................. 149 3.4 Experimental ............................................................................................................... 150 3.4.1 Materials ................................................................................................................. 150 3.4.2 Equipment ............................................................................................................... 150 3.4.3 Procedure and Experimental Data .......................................................................... 151 Chapter 4: Self-Assembly of Extended Head-to-Tail Triangular Pt3 Macrocycles into Nanotubes ...................................................................................................................................159 4.1 Introduction ................................................................................................................. 159 4.1.1 Pt4 Head-to-Tail Schiff-Base Macrocycles ............................................................. 160 4.1.2 Self-Assembly of Pt4 Macrocycles into Nanotubes ................................................ 161 4.2 Results and Discussion ............................................................................................... 162 4.2.1 Synthesis and Characterization ............................................................................... 162 viii  4.2.2 Self-Assembly of Pt3 Macrocycles in Solid State ................................................... 170 4.2.3 Dilution Experiments .............................................................................................. 173 4.2.4 Variable Temperature Experiments ........................................................................ 178 4.2.5 Diffusion Ordered Spectroscopy (DOSY) .............................................................. 183 4.2.6 Nuclear Overhauser Effect NMR Spectroscopy (NOESY) .................................... 188 4.3 Conclusions ................................................................................................................. 196 4.4 Experimental ............................................................................................................... 197 4.4.1 Materials ................................................................................................................. 197 4.4.2 Equipment ............................................................................................................... 197 4.4.3 Procedure and Experimental Data .......................................................................... 198 Chapter 5: Conclusions and Future Directions.......................................................................209 5.1 Overview ..................................................................................................................... 209 5.2 Future Directions ........................................................................................................ 210 5.2.1 More Phenyl Derived Campestarenes ..................................................................... 210 5.2.2 Campestarenes with Liquid Crystal Properties ....................................................... 212 5.2.3 Extended Campestarenes ........................................................................................ 214 5.2.4 Pt(II) Containing Macrocycles with Different Geometry ....................................... 215 5.3 Experimental ............................................................................................................... 216 5.3.1 Materials ................................................................................................................. 216 5.3.2 Equipment ............................................................................................................... 217 5.3.3 Procedure and Experimental Data .......................................................................... 217 References ...................................................................................................................................222   ix  List of Tables  Table 2.1  Isolated yields of campestarenes 1. .............................................................................. 71 Table 2.2  C-O chemical shift collected in various solvents......................................................... 85 Table 2.3  Chemical shifts and 1JNH, 2JNH, 3JHCNH coupling constants for 1e-15N5 in DMSO-d6, DMF-d7, CD2Cl2, CDCl3, toluene-d8, and benzene-d6................................................................ 89 Table 2.4  Possible values of coupling constants 1JNH, 2JNCH and 3JHCNH for different local patterns of tautomerization. ........................................................................................................ 107 Table 2.5  Percentage of keto-enamine character of N/O pairs in campestarene-15N5 (1e-15N5) in DMSO-d6, DMF-d7, CD2Cl2, CDCl3, toluene-d8, and benzene-d6 at room temperature. ........... 112 Table 2.6  Selected crystallographic data for structure of 1e. ..................................................... 130 Table 3.1  Reaction conditions for the synthesis of campestarene 1j from amine compound 2j.138 Table 3.2  reaction conditions for the synthesis of campestarene 1l from amine 2l. .................. 145 Table 4.1  Diffusion coeffiecient of 5a and 5c measured by DOSY at 298 K. ........................... 185 Table 4.2  Diffusion coeffiecient of 5a and 5c measured by DOSY at 250K. ............................ 187 Table 4.3  Relative integral and estimated distance data from NOESY NMR study of macrocycle 5a at 240 K. ................................................................................................................................. 192 Table 4.4  Relative integral and estimated distance data from NOESY NMR study of macrocycle 5a at 298K. .................................................................................................................................. 195  x  List of Figures  Figure 1.1  Structures of dibenzo-18-crown-6 (1); cryptand [2. 2. 2] (2); and spherands (3). ....... 2 Figure 1.2  The four different conformers of calix[4]arenes. ......................................................... 4 Figure 1.3  p-tert-butylcalix[4]arene with bulky groups on phenolic oxygens (4) and p-tert-butylcalix[4]arene with complexation of sodium (5). ..................................................................... 5 Figure 1.4  Structures of α, β, γ– Cyclodextrins. ............................................................................ 6 Figure 1.5  Hydrogen-bonding-induced stacking of peptide-based macrocycles. Only a small segment of a nanotube is shown. .................................................................................................... 7 Figure 1.6  Structures of (a) porphyrin and (b) complicated macrocycle composed of multiple porphyrins. ...................................................................................................................................... 9 Figure 1.7  Synthesis of macrocycle 6 by (a) Stephens-Castro coupling in 4.6% yield, (b) Sonogashira coupling in 75% yield and (c) one-pot alkyne metathesis in 68% yield. ................. 10 Figure 1.8  Structures of monomeric complex ZnL (left), macrocyclic dimer with template (ZnL)2dpy (middle) and macrocyclic trimer with template (ZnL)3tpyt (right)........................... 14 Figure 1.9  Structures of a tridentate template (7) and hexameric wheel of porphyrins (8). ........ 15 Figure 1.10  Synthesis of the 12-porphyrin nano-ring with templates. ........................................ 16 Figure 1.11  Schematic representations of phenylene-ethynylene macrocycles and the building blocks from which they are constructed. ...................................................................................... 18 Figure 1.12  A mixture of macrocycles with various geometries from a self-assembly process. 20 Figure 1.13  Macrocycles synthesized from one reaction with 2,2′:6′,2″-terpyridine building blocks (9). ................................................................................................................................................. 21 Figure 1.14  Synthesis of hexagonal wreaths from ligands 10 and 11. ........................................ 21 xi  Figure 1.15  Structures of hexagonal wreaths 12 and 13. ............................................................. 22 Figure 1.16  Structure of a [2+2] amide macrocycle formed by flexible organic building blocks (left) and [3+3] Schiff-base macrocycles with rigid framework (right). ...................................... 23 Figure 1.17  Structure of gramicidin S.......................................................................................... 25 Figure 1.18  Synthesis of a peptide macrocycle from linear precursor. ....................................... 26 Figure 1.19  Synthesis of a peptoid macrocycle from linear precursor. ....................................... 26 Figure 1.20  Structures of oligoamide macrocycle (14) and oligohydrazide macrocycle (15) reported by Gong and coworkers. ................................................................................................. 28 Figure 1.21  Chemical structure (a) and Crystal structure of macrocyclic amide pentamer: (b) top view, and (c) side view both with methoxy methyl groups in CPK representations. ................... 29 Figure 1.22  Structures of macrocyclic amide pentamers 16, 17, 18 and 19 which can be used to bind metal ions with a radius of ~1.5 Å. ....................................................................................... 30 Figure 1.23  Regio- and chemo-selective demethylation of methoxybenzene-based macrocyclic pentamers. ..................................................................................................................................... 31 Figure 1.24  (a) Structures of fluoropentamers 20. (b) Top and side views of crystal structure of 20a, illustrating the formation of interplanar H-bonds of 2.50 Å in length. Dotted cycles in (b) indicate the amide bonds that are twisted out of the plane to form stronger intermolecular H-bonds that enhance the interplanar aggregations. .................................................................................... 31 Figure 1.25  (a) Chemical structures of pentamers 21a and 21b. (b) Top and side views of computationally optimized structures for 21a and 21b, with the exterior side chains replaced by methyl groups at the B3LYP/6-31G level, illustrating five-fold-symmetric planarity in 21a and 21b. ............................................................................................................................................... 32 xii  Figure 1.26  Computationally optimized structures of 1D columnar aggregates possibly formed by (a, b) [21aKBr]n and (c, d) [21bKBr]n at the B3LYP/6-31G level under periodic boundary conditions. The top-down views illustrate two possible packing modes and their relative energies. Side views, with the exterior side chains removed, illustrate the interplanar distances that dictate the strength of ionic interactions. In the CPK models, K+ = 1.38 Å and Br- = 1.95 Å. ................ 33 Figure 1.27  Structures of hybrid pentamers employed by Zeng and coworkers to probe the potential for selective recognition of metal ions by a family of macrocyclic hybrid pentamers with modularly tunable interior properties. The ion-extraction profiles determined under the identical conditions for the three well-known ligands, i.e., 18-crown-6, dibenzo-21-crown-7, and kryptonfix-222, were included for comparison purpose. The ions whose extractability lies within 80% of the most extractable ones are additionally highlighted below the most extractable ones in the cavity. ...................................................................................................................................... 35 Figure 1.28  One-pot preparation of circular (a) alkoxybenzene pentamers (22) and (b) pyridone pentamers (23) from their respective monomers. ......................................................................... 36 Figure 1.29  Synthesis of macrocyclic pentamers from monomers and oligomers. ..................... 38 Figure 1.30  One-pot preparation of circular (a) alkoxybenzene pentamers from monomers and (b) pyridone pentamers from dimers. ................................................................................................. 39 Figure 1.31  One-pot synthesis of cyanostar 25. ........................................................................... 41 Figure 1.32  (a) Sandwiches of two bowl-shaped cyanostar macrocycles 25 result in a mixture of four possible stereoisomers. (b) When the cyanostar macrocycles are viewed from the tops of their bowls stereoisomers with either M chirality or P chirality can be defined. .................................. 42 xiii  Figure 1.33 (a) Cartoon representation of [3]rotaxane 26-TBA+ and (b) X-ray crystal structure of [3]rotaxane 26-TBA+ (solvent molecules, TBA+ and protons are removed for clarity). The M-P isomer is shown. (TBA: tetrabutylammonium) ............................................................................ 42 Figure 1.34  Rocking pathways between all-out conformers (pink P) and their enantiomers (blue M) are established through corresponding transition states (yellow). The free energies were calculated by DFT with implicit solvation.................................................................................... 43 Figure 1.35  Head-to-tail cyclotrimerization of an unsymmetrical diene leading to a C3-symmetric macrocycle instead of AB-ring system of taxane. ........................................................................ 44 Figure 1.36  Thiazoline macrocycles (left and middle) and thiazoline-thiazole hybrid macrocycles (right) prepared via head-to-tail approach by Fukuyama and coworkers. .................................... 45 Figure 1.37  Structures of a repeating unit of cyclic trimer (left) and cyclic heterometallic trimer 27 (right). Phenyl groups are omitted for clarity. ......................................................................... 46 Figure 1.38  Assembly of Zn4 cyclic tetramer in a non-coordinating solvent. Solubilizing groups of the space-filling tetramer model (right) have been omitted for clarity. .................................... 47 Figure 1.39  Structures of tetranuclear Mn(II) complexes 284+. The counter ions are (ClO4)-. ... 47 Figure 1.40  The monomer-oligomer conversion of Cu(II) complexes controlled by pH. ........... 48 Figure 1.41  Structure of trinuclear Pd(II) head-to-tail macrocycle 29. ....................................... 49 Figure 1.42  Structures of 4-fold (left) symmetric macrocycle 30 and 6-fold (right) symmetric molecular “paddlewheel” 31. ........................................................................................................ 50 Figure 1.43  Structure of tetranuclear Pt(II)-containing macrocycle 32, Pt4(N,C,N-NPA)4(CH3)4. NPA = N-(2’-pyridyl)-7-azaindole. .............................................................................................. 51 Figure 1.44  Synthesis of salen and salphen compounds. ............................................................. 53 xiv  Figure 1.45  Salen-like compounds for asymmetric epoxidation (33) and electroluminescence (34)........................................................................................................................................................ 53 Figure 1.46  Chemical structure of a dinuclear Robson macrocycle complex 35. ....................... 54 Figure 1.47  Structures of dinuclear complexes 36 and 37. .......................................................... 54 Figure 1.48  Structures of [3+3] Schiff-base macrocycles. .......................................................... 55 Figure 1.49  Comparison between two [3+3] macrocycles. ......................................................... 57 Figure 1.50  Structures of Schiff-base macrocycles with 2-, 3-, 6-fold symmetry. ...................... 58 Figure 1.51  Prediction of macrocycle geometry from the angles between functional groups. ... 59 Figure 1.52  One-pot synthesis of Campestarene (left) and its dimerization behavior (right). .... 60 Figure 1.53  Tautomerization between the enol-imine (left) and the keto-enamine (right) forms. The NH tautomer may be formally depicted as a zwitterionic iminium-phenolate or a neutral keto-enamine in the extreme cases. (The notation Ha and Hb will be used in the discussion.) ............. 60 Figure 1.54  Synthesis of head-to-tail Pt4 macrocycles and their MALDI-TOF mass spectrum which shows aggregation. ............................................................................................................. 61 Figure 2.1  (a) Prediction of [3+3] macrocycle geometry from precursors; (b) Predicted 6-fold geometry and actual 5-fold geometry of campestarene. ............................................................... 64 Figure 2.2  Tautomerization between the enol-imine (left) and the keto-enamine (right) forms. The NH tautomer may be formally depicted as a zwitterionic iminium-phenolate or a neutral keto-enamine in the extreme cases. (The notation Ha and Hb will be used in the following discussion.)....................................................................................................................................................... 65 Figure 2.3  a) Chemical structures of campestarene 1f and 1g; b) Possible orientation of the keto-enamine dimer of 1f. ..................................................................................................................... 69 Figure 2.4  1H NMR spectrum of macrocycle 1d in CD2Cl2 (400 MHz, room temperature). ...... 70 xv  Figure 2.5  1H NMR spectrum of 1e in CDCl3 (400 MHz, room temperature). ........................... 72 Figure 2.6  MALDI-TOF mass spectrum of macrocycle 1e. Matrix: trans-2-[3-(4-t-butyl-phenyl)-2-methyl-2-propenylidene] malononitrile (DCTB). ..................................................................... 73 Figure 2.7 (a) Chemical structure and Mass spectrum of campestarene 1h; (b) Chemical structure and Mass spectrum of campestarene 1h’. ..................................................................................... 74 Figure 2.8   (a) IR spectra of macrocycle 1e (black) and 1e-15N5 (red); (b) comparison of peaks at ~1600 cm-1 of macrocycle 1e (black) and 1e-15N5 (red). .............................................................. 76 Figure 2.9  Views of the molecular structure of 1e as determined by SCXRD. a) View of individual macrocycle; b) view of crystallographic unit cell, showing herringbone packing; c) side-on view, showing mean plane between inner phenylene carbon atoms.  For clarity, only the major component of the disordered TIPS group is shown, and most hydrogen atoms are omitted; in image c, the TIPS groups are also omitted. ............................................................................................. 78 Figure 2.10  Bond lengths of 1f in the two limiting tautomeric forms: (a) all enol-imine form and (b) all keto-enamine form, computed at the 6-311g(d,p)/M06/benzene level of theory. .............. 80 Figure 2.11  1H NMR spectra (room temperature, 400 MHz) of campestarene 1e in (a) DMSO-d6, (b) DMF-d7, (c) CD2Cl2, (d) CDCl3, (e) toluene-d8, (f) benzene-d6. (* indicates a residual solvent peak due to DMF-d7). ................................................................................................................... 81 Figure 2.12  13C{1H} NMR data of 1e in C6D6, toluene-d8, and CD2Cl2. .................................... 83 Figure 2.13  1H-13C HMBC data for 1e in CDCl3 and DMF-d7. .................................................. 84 Figure 2.14  1H NMR spectra (room temperature, 400 MHz) of campestarene-15N5 (1e-15N5)  in (a) DMSO-d6, (b) DMF-d7, (c)  CD2Cl2, (d) CDCl3, (e) toluene-d8, (f) benzene-d6. (* indicates a residual solvent peak due to DMF-d7). ......................................................................................... 86 xvi  Figure 2.15  Section of the 1H-15N HMQC NMR spectrum of campestarene-15N5 (1e-15N5) in CD2Cl2........................................................................................................................................... 88 Figure 2.16  Calculated charge distributions of the enol-imine and the keto-enamine tautomer of campestarene 1f. Atoms are colored according to the values of their net atomic charges computed according to the Merz-Kollman procedure, fitting the electrostatic potential computed at the 6-311+g(2d,2p)/PBE1PBE//Vacuum/6-311g(d,p)/PBE1PBE/Vacuum........................................... 90 Figure 2.17  (a) UV-Vis spectra of macrocycle 1e in toluene, benzene, CHCl3, CH2Cl2, DMF and DMSO (this is the order from bottom to top at 550 nm); (b) photographs of the same solutions (all 10-5 mol L-1). ................................................................................................................................. 92 Figure 2.18  Comparison between the experimental spectra of 1e in benzene and DMSO (brown and green lines) with the TD-DFT calculations in the 1g keto-enamine and enol-imine tautomers (green and brown vertical arrows). ............................................................................................... 93 Figure 2.19  Effect of the peripheral substitution on the most intense transitions in case of 1f (red) 1g (green) in the two limiting tautomeric forms (a) keto-enamine and (b) enol-imine. TD-DFT calculations at the 6-311g(d,p)/M06/DMF. The lines indicate the calculated spectra while the curves are the experimental data. .................................................................................................. 94 Figure 2.20  Solvent effect on the TD-DFT 6-311g(d,p)/M06 computed spectra of (a) keto-enamine and (b) enol-imine of 1g in benzene, DMSO and DMF. .............................................................. 95 Figure 2.21  Relative stability of different tautomers of 1f in benzene (red) and DMSO (green). The energy reference is related to the most stable isomer in the same solvent. Full geometry optimization at the 6-311g(d,p)/ M06/solvent level of theory. ..................................................... 97 Figure 2.22  Molecular energies (ΔE, green) and internal Gibbs free energies (ΔG(298), black) for the eight tautomers computed at the 6-311g(d,p)/ M06 /DMF level of theory for 1g. ................. 98 xvii  Figure 2.23  Molecular Energies (E, green) and Gibbs free energies (G(298), red) for eight tautomers of  1f and their transition state for sequential interconversion computed at the 6-311g(d,p)/mPW2PLYP/DMF//6-311g(d,p)/M06/DMF level of theory (dashed lines are added just to allow easy following of the series of values; they do not imply linear changes.). ............ 99 Figure 2.24  Variable-temperature 1H NMR spectra of 1e in  toluene-d8, only the resonances near 17 (Ha) and 9 (Hb) ppm are shown. ............................................................................................. 101 Figure 2.25  Variable-temperature 1H NMR spectra of 1e in (a) CDCl3, (b) toluene-d8, (c) CD2Cl2 and (d) DMF-d7. Only the resonances near 17 (Ha) and 9 (Hb) ppm are shown. ........................ 102 Figure 2.26  Variable-temperature 1H NMR spectra of 1e-15N5 in (a) CD2Cl2 and (b) DMF-d7.103 Figure 2.27  (a) Variable-temperature 1H-15N HMQC NMR spectra of 1e-15N5 in CD2Cl2, the projection is at 267 K; (b) Variable-temperature 1H-15N HSQC NMR spectra of 1e-15N5 in CD2Cl2, the projection is at 267 K. ........................................................................................................... 105 Figure 2.28  (a) Variable-temperature 1H-15N HMQC NMR spectra of 1e-15N5 in DMF-d7, the projection is at 293 K; (b) Variable-temperature 1H-15N HSQC NMR spectra of 1e-15N5 in DMF-d7, the projection is at 293 K. ...................................................................................................... 106 Figure 2.29  TD-DFT computed values of the 1JNH, 2JNCH and 3JHCNH coupling constants in case of (a) 11010 15N tautomer, (b) 11111 15N tautomer and (c) 11100 15N tautomer at the GIAO 6-311g(d,p)/cc-QZVP/M06/DCM level of theory. Red numbers, short and long arcs identify 1JNH, 2JNCH and 3JHCNH respectively. .................................................................................................... 108 Figure 2.30  (a) Non-linear curve fitting based on the measured 1JNH in variable-temperature 1H NMR spectra of 1e-15N5 in DMF-d7 (red dashed and dot; experimental blue square) and 3JHCNH values (black dashed and dot; experimental green circles) assuming only the keto-enamine and xviii  enol-imine forms. R2=0.9999; b) non-linear curve fitting based on the measured 1JNH values in variable-temperature 1H NMR spectra of 1e-15N5 in CD2Cl2, R2=0.999. ................................... 110 Figure 2.31  Variable-temperature UV-vis spectroscopy of 1e in (a) toluene (183 - 293 K); (b) chloroform (213 - 323 K); (c) dichloromethane (223 - 293 K) and (d) DMF (213 - 298 K). The arrows indicate the direction of change in the spectra with decreasing T. ................................. 113 Figure 2.32  UV-vis spectra of campestarene 1e and intensities of Gaussian functions used for the decomposition. The same set is used along the whole set of temperatures. Plot and triangles: 213 K (green) and 298 K (red). Inset: Trend of the most relevant Gaussian intensities as function of temperature. ................................................................................................................................ 114 Figure 2.33  Comparison of the experimental UV-vis spectra of 1e in DMF at 213 K (green dashed), 258 K (red continuous), and 213 K (blue continuous) and the TD-DFT computed transition energies (only transitions with oscillator strength larger than 0.05 are reported) at the 6-311G(d,p)/M06/DMF level of theory for the 11111 (green triangles), 00000 (blue triangles), 11110 (yellow squares), 11100 (red triangles), 10101 (aqua circles), and 11000 (black rhombuses) tautomeric forms of 1g. ............................................................................................................... 116 Figure 3.1  1H NMR spectra of campestarene 1e in toluene-d8, (a) when temperature was increased from 303 K to 377 K; (b) when temperature was decreased from 377 K to 315 K (400 MHz). 134 Figure 3.2  Structures of dimer, trimer and tetramer synthesized from 2j.................................. 138 Figure 3.3  ESI-MS of mixture from reaction conducted in toluene with the presence of PTSAH2O and molecular sieve 4Å  (Experiment 6 in Table 3.1). ............................................................... 140 Figure 3.4  ESI-MS spectrum of mixture from reaction conducted in toluene with the presence of PTSAH2O. .................................................................................................................................. 141 xix  Figure 3.5  ESI-MS spectrum of reaction mixture from reaction conducted under neat condition...................................................................................................................................................... 142 Figure 3.6  ESI-MS spectrum of reaction mixture from reaction conducted in toluene with the presence of PTSAH2O (Experiment 3 in Table 3.2) after (a) 7 days; (b) 14 days; (c) 28 days; (d) 38 days. ....................................................................................................................................... 147 Figure 3.7  1H NMR spectrum of campestarene 1l in CDCl3 (room temperature, 400 MHz). Only the spectral region above 6 ppm is shown. ................................................................................. 148 Figure 3.8  ESI-MS spectrum of campestarene 1l. ..................................................................... 149 Figure 4.1  Structures of the target extended Pt4 head-to-tail macrocycles 5a’ and 5b’. ........... 162 Figure 4.2  MALDI-TOF MS of macrocycle 5a.  The peak at m/z = 2755.7 Da corresponded to the protonated Pt3 macrocycle, [5a+H]+. Matrix: trans-2-[3-(4-t-butyl-phenyl)-2-methyl-2-propenylidene] malononitrile (DCTB). ...................................................................................... 167 Figure 4.3  MALDI-TOF MS of macrocycle 5b.  The peak at m/z = 3055.9 Da corresponds to the protonated Pt3 macrocycle, [5b+H]+. Matrix: trans-2-[3-(4-t-butyl-phenyl)-2-methyl-2-propenylidene] malononitrile (DCTB). ...................................................................................... 168 Figure 4.4  Chemical structures of Pt3 macrocycle (a) 5a and (b) 5b. ........................................ 169 Figure 4.5  (a) Chemical structure of Pt3 macrocycle with R = H; (b) top view and (c) side view of DFT optimized geometry of macrocycle with R = H, computed at the 6-31g(d) level of theory...................................................................................................................................................... 170 Figure 4.6  Powder X-ray diffraction pattern of macrocycle 5a. ................................................ 171 Figure 4.7  TEM images of macrocycle 5a. The repeating distance between adjacent stripes is approximately 2.6 nm, and the inset shows the proposed organization of 5a in the solid state. 172 xx  Figure 4.8  (a) Top view and (c) side view of 2 Pt3 macrocycles stacking with the same orientation; (b) Top view and (d) side view of 2 Pt3 macrocycles stacking with opposite orientation. For clarity, hydrogen atoms and R groups were omitted. All macrocycles have the same chemical structure. Blue and red colors are used to distinguish macrocycles with different orientations. ................ 173 Figure 4.9  Aromatic region of 1H NMR spectrum of macrocycle 5a in CDCl3 at room temperature (400MHz, c = 7.0 ×10-3 mol/L). ................................................................................................ 174 Figure 4.10  1H NMR spectra of macrocycle 5a in CDCl3 at different concentrations (room temperature, 400 MHz). .............................................................................................................. 175 Figure 4.11  (a) UV-vis spectra of macrocycle 5a in CHCl3 at different concentration (room temperature); (b) linear fitting for absorption of different 5a solution at 328 nm (black), 442 nm (red) and 465 nm (blue). The R2 value for 328 nm, 442 nm and 465 nm data are 0.99992, 0.99992, 0.99993 respectively. .................................................................................................................. 176 Figure 4.12  (a) The absorption spectrum of macrocycle 5a in chloroform at room temperature; (b) photos of macrocycle 5a solution under visible/UV light; (c) excitation spectrum of macrocycle 5a solution, emission light wavelength is 635 nm; (d) emission spectrum of macrocycle 5a solution, excitation light wavelength is 450 nm. c = 2.5 × 10-6 mol/L. ..................................................... 177 Figure 4.13  Variable temperature 1H NMR spectra of macrocycle 5a in CDCl3 (400 MHz). Only the aromatic region is shown. ..................................................................................................... 179 Figure 4.14  1H NMR spectra of model compound 5c in CDCl3 at different concentrations (room temperature, 400 MHz). .............................................................................................................. 181 Figure 4.15  Variable temperature 1H NMR spectra of model compound 5c in CDCl3 (400 MHz). Only the aromatic region is shown. ............................................................................................ 181 xxi  Figure 4.16  1H-13C HSQC spectrum of 5c in CDCl3. Insert is the structure of 5c. The peak at 9.2 ppm was assigned to a pyridyl proton (room temperature, 400 MHz). ...................................... 182 Figure 4.17  Gaussian fit to diffusion peak intensity at 298K using a non-linear fit. ................. 184 Figure 4.18  Gaussian fit to diffusion peak intensity at 250 K using a non-linear fit. ................ 187 Figure 4.19  1H-13C HSQC and 1H-13C HMBC NMR of macrocycle 5a at 240 K. ................... 189 Figure 4.20  1H NMR spectrum of macrocycle 5a with peak assignment at 240 K (400 MHz). Only aromatic region is shown. ........................................................................................................... 190 Figure 4.21  NOESY NMR spectrum of macrocycle 5a at 240K. ............................................. 191 Figure 4.22  Top view of two stacked macrocycle 5a. Only protons a and h from different macrocycles are shown in green and light blue color. The other protons are removed for clarity...................................................................................................................................................... 193 Figure 4.23  NOESY NMR spectrum of macrocycle 5a at 298K. ............................................. 194  xxii  List of Symbols  Symbol  Description ̶   covalent bond (CH2O)n  paraformaldehyde °   degrees °C   degrees Celsius Å   Ångstrom AcOH   acetic acid bpy   bipyridyl Bu   butyl cm-1   wavenumber D   diffusion coefficient d   doublet, days equiv.   equivalents Et   ethyl EtOH   ethanol g   grams h   hours iPr   isopropyl K   Kelvin kB   Boltzmans constant m   multiplet xxiii  m   meta Me   methyl MeOH   methanol mg   milligram MHz   megahertz mL   milliliter mmol   millimoles nBu   normal-butyl p   para Ph   phenyl pH   power of hydrogen Pr   propyl rs   radius of the molecule s   singlet sept   septet T   temperature t   triplet tBu   tert-butyl tpyt   tripyridinyl-triazine δ   chemical shift (in ppm) η   viscosity λ   wavelength ν   frequency xxiv  List of Abbreviations  Abbreviation  Definition 1D   one dimensional 2D   two dimensional 3D   three dimensional BOP benzotriazole-1-yl-oxy-tris-(dimethylamino)-phosphonium hexafluorophosphate DCM   dichloromethane DCTB   trans-2-[3-(4-t-butyl-phenyl)-2-methyl-2-propenylidene] malononitrile DFT   density functional theory DIAD   diisopropyl azodicarboxylate DLS   dynamic light scattering DMF   dimethylformamide DMSO   dimethyl sulfoxide DNA   deoxyribonucleic acid DOSY   diffusion-ordered spectroscopy EA   elemental analysis ESI   electrospray ionization FT   Fourier transform HBTU   2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate HCl   hydrochloric acid HMBC  heteronuclear multiple-bond correlation spectroscopy xxv  HMQC  heteronuclear Multiple Quantum Coherence HR   high resolution HSQC   heteronuclear single-quantum correlation spectroscopy IR   infrared MALDI  matrix-assisted laser desorption/ionization mCPBA  m-Chloroperoxybenzoic acid MP   melting point MS   mass spectrometry MW   molecular weight NMR   nuclear magnetic resonance NOE   nuclear Overhauser effect NOESY  nuclear Overhauser effect spectroscopy NPA   N-(2’-pyridyl)-7-azaindole PTSA   p-toluenesulfonic acid PXRD   powder X-ray diffraction RT   room temperature SCRF   self-consistent reaction field SCXRD  single-crystal X-ray diffraction STM   scanning tunneling microscope TD   time-dependent TEM   transmission electron microscopy THF   tetrahydrofuran TIPS   triisopropylsilyl xxvi  TOF  time of flight UV  ultraviolet VC  variable concentration vis  visible VT  variable temperature XRD  X-ray diffraction  xxvii  Acknowledgements  I am deeply indebted to my outstanding supervisor, Dr. Mark MacLachlan. His wisdom, humor and enthusiasm have made my PhD studies an enjoyable and educational experience. He has always been an inspiration and an encouragement for me. It would have been impossible for me to finish compiling this thesis without his help. I am grateful to the rest of the MacLachlan group members, both post and present, that have helped me tremendously through many frustrating moments of my PhD studies. Their support kept me in good spirits when I ran into troubles. Specifically, I would like to thank Dr. Nicholas White, who is a great and patient crystallographer. Thanks to Samuel Guieu, Brian Sahli, Hessam Mehr, Peixi Wang, and Clement Cheung, for their knowledge and technical support to my project. Thanks to the UBC NMR staff, especially Maria E., for providing a lot of help and advice on 2D NMR experiments. Many thanks also to the UBC MS staff, for their great job on analyzing compounds. In addition, powder X-ray Diffraction was obtained by Anita Lam. I owe much gratitude to Dr. Francesco Lelj, for his patient and thoughtful computational study on campestarenes.  Lastly, I would like to thank my parents for their unconditional support, especially during the frustrating moments of my PhD study.  xxviii  Dedication  This dissertation is dedicated to my mother and father, for their unconditional support, especially during the frustrating moments of my PhD study.  1  Chapter 1: Introduction  1.1 Supramolecular Chemistry  1.1.1 Background While working at Dupont in the 1960s, Charles Pedersen isolated a mysterious by-product from a reaction to prepare of bis[2-(o-hydroxyphenoxy)ethyl] ether. His experiment involved reacting an aqueous 1-butanol bis(2-chloroethyl) ether with the sodium salt of 2-(o-hydroxyphenoxy)tetrahydropyran that was contaminated with some catechol. This white compound itself was not soluble in methanol. However, he noticed that the solubility of this by-product could be highly enhanced by adding sodium salts. After additional studies, Pedersen determined the structure of this by-product as a macrocyclic polyether, which is now known as a crown ether (Figure 1.1). The enhanced solubility is due to complexing with alkali metal cations such as Na+ and K+ inside the cavity with the oxygen atoms donating electron density. Following this work, Pedersen synthesized many similar macrocyclic polyether compounds.1 Shortly after Pedersen’s discovery of crown ethers, Jean-Marie Lehn prepared a compound that has a three-dimensional structure designed to bind to various metal ions. Since the three dimensional binding cavity provides a more comprehensive binding environment for ions, this new compound has a better binding ability than crown ethers. 2  Lehn named this family of compounds as cryptands (Figure 1.1). Upon binding to a metal ion, the ligand is called a cryptate. Crystal structures of the originally synthesized crown ethers and cryptands show that these cyclic molecules do not have pre-formed binding sites before ion complexation. 2  Following Pedersen and Lehn’s work, Donald J. Cram synthesized spherands (Figure 1.1) to improve the binding ability. Different from crown ethers and cryptands, which do not have oxygen atoms rigidly anchored for binding cations, spherands have a pre-organized cavity of oxygen atoms for binding. This rigid binding pocket made spherands the first system to be pre-organized for metal binding, before the complexation process.3  Figure 1.1  Structures of dibenzo-18-crown-6 (1); cryptand [2. 2. 2] (2); and spherands (3).  In 1987, Charles Pedersen, Jean-Marie Lehn and Donald J. Cram were jointly awarded the Nobel Prize in Chemistry for their development and use of molecules with structure-specific interactions of high selectivity. From that moment on, the field of supramolecular chemistry started to draw more attention from chemists. Jean-Marie Lehn first introduced the term supramolecular chemistry in 1978.4 He defined supramolecular chemistry as a field of chemistry beyond the molecule. While conventional chemistry is focused on forming and breaking covalent bonds, supramolecular chemistry emphasizes on non-covalent bonds, including hydrogen bonding, metal coordination, hydrophobic forces, van der Waals forces, - interactions and electrostatic effects.  3  With the improvement of the understanding of natural structures, more people noticed the importance of non-covalent interactions, For example, in a DNA chain.  Two DNA strands, which are known as polynucleotides, coil around each other to form a double helix. The main driving force is the formation of hydrogen bonding between the nitrogenous bases of the two separated strands. Another well-known example of supramolecular chemistry in nature is protein folding. While the primary structure of protein relies on covalent bonds in a polypeptide chain, formation of more complicated structures is assisted by non-covalent bonds such as hydrogen bonding, hydrophobic forces, and metal coordination. These non-covalent interactions can bring molecules together to form larger species with more complicated structures.  In its infancy, supramolecular chemistry mostly involved host-guest chemistry. However, over the last four decades, this field of chemistry has become a much broader field spanning diverse topics such as molecular self-assembly, 5  template-directed synthesis, 6  and molecular machinery7.  1.1.2 Macrocycles Considering the vast quantity of different systems involved in modern supramolecular chemistry, a comprehensive summary of this field will not be covered in its entirety in this thesis. However, there is a fundamental structure contributing to most of these systems and is recurring in this thesis--macrocycles.  A macrocycle is a large cyclic molecule, often containing heteroatoms or functional groups that allow for further modification and applications. Previously mentioned compounds, including crown ethers, cryptands and spherands, are typical examples of macrocycles. The oxygen atoms 4  inside macrocycles can bind to metal ions forming complexes. Depending on cavity size and number of electron donors, macrocycles can show different affinities to various metal ions. Calixarenes are a family of macrocycles that consist of several phenol moieties linked by methylene groups (Figure 1.2). As nomenclature for naming such a complex system, a number is usually placed within a square bracket to indicate the number of phenolic residues in the calixarene, for instance, calix[4]arene contains 4 phenolic residues, whereas calix[6]arene possesses 6 repeating units. Calixarenes are well known in host-guest chemistry since the hydrophobic cavity can bind to small ions or molecules.  Figure 1.2  The four different conformers of calix[4]arenes.  Depending on the relative orientation of the phenolic residues in a calixarene, different conformers can be formed. For calix[4]arenes, there are “cone”, “partial cone”, “1,2-alternate” and “1,3-alternate” conformations (Figure 1.2). 5  Among all four conformers, the cone-shaped calix[4]arene is  relatively more stable than the others as a result of hydrogen bonding. However, the energy barrier to rotation of the phenolic groups is low enough to allow for rapid interconversion of the different conformers.8 Moreover, by coordinating to metal ions9 or inducing bulky substituent groups onto phenolic oxygens (Figure 1.3), macrocycles can be locked to a certain conformation.10   Figure 1.3  p-tert-butylcalix[4]arene with bulky groups on phenolic oxygens (4) and p-tert-butylcalix[4]arene with complexation of sodium (5).  Cyclodextrins are a family of compounds composed of glycopyranose units. They can be produced from ordinary starch by enzymatic conversion with a set of easily available enzymes. They usually contain 6, 7 or 8 sugar units, and are named as α-, β-, γ- cyclodextrin respectively (Figure 1.4).  Cyclodextrins have a hydrophilic exterior and hydrophobic interior, making it possible for cyclodextrins to uptake organic molecules in aqueous media. 11  The formation of inclusion compounds of cyclodextrins can greatly modify both physical and chemical properties of the guest molecules.  6  Cyclodextrins have been used for building rotaxanes 12  and making supramolecular polymers. 13  By utilizing the host-guest interaction between cyclodextrins and other stimuli-responsive chemicals, researchers managed to prepare supramolecular polymers that form chemical-responsive14 or photoswitchable supramolecular hydrogels.15  Figure 1.4  Structures of α, β, γ– Cyclodextrins.  7  In 1993, Ghadiri and coworkers reported peptide-based macrocycles.16 These macrocycles can aggregate as a result of intermolecular hydrogen bonding between the amide NH and the carbonyl groups belonging to adjacent macrocycles (Figure 1.5). This aggregation leads to self-assembled nanotubes, which have the potential for mimicking ion channels. Since one of the highlights of supramolecular chemistry is to mimic biological systems, nanotubes formed from self-assembly of macrocycles have drawn a lot of attention in the field of chemistry.  Figure 1.5  Hydrogen-bonding-induced stacking of peptide-based macrocycles. Only a small segment of a nanotube is shown.  1.1.3 Shape-Persistent Macrocycles Shape-persistent macrocycles are macrocycles characterized by rigid frameworks that can hold a specific conformation. With rigid moieties, these macrocycles can be designed for binding guest molecules with certain dimensions. This preorganization can improve both binding selectivity and affinity. In addition, since shape-persistent macrocycles can hold their internal 8  voids in the absence of guest molecules, they can be used to create porous structures after self-assembly. In order to achieve the rigidity in shape-persistent macrocycles, conjugated organic structures, such as aromatic rings and alkynyl groups, are often incorporated into macrocycles. As a result of the incorporation of rigid conjugated organic structures, shape-persistent macrocycles often tend to stack to form layered structures resulting from - interactions.  One of the best known shape-persistent macrocycles is porphyrin, a small macrocycle composed of four pyrrole rings connected by methine bridges (Figure 1.6). The fully conjugated system gives porphyrin not only light harvesting properties, but also rigidity to hold its shape. Many porphyrins are naturally occurring, such as in heme groups associated with transport of oxygen in hemoglobin, or in chlorophyll associated with photosynthesis. The rigid aromatic structure bestows porphyrin versatility as optoelectronic or catalytic units. Moreover, such rigidity even allows porphyrin to be used as a building block for more complicated structures while still maintaining a flat conformation. Recently, some developments were made in the synthesis and applications of macrocycles in which porphyrins are key components of the repeat units (Figure 1.6).17 9   Figure 1.6  Structures of (a) porphyrin and (b) complicated macrocycle composed of multiple porphyrins.  Macrocycles consisting of phenylene and ethynylene units are well studied as shape-persistent macrocycles. The synthesis of macrocycle 6 was first reported in 1974 by Neunhoeffer and coworkers (Figure 1.7).18 However, the yield of this ring-closing reaction was only 4.6%. Owing to the low yield of synthesis, this macrocycle was not well studied until another step-wise synthetic route was reported in 1992.19  This improved macrocyclization step has a yield of 75%. However, the synthesis of this macrocyclization precursor is cumbersome, as it requires multiple steps. Twelve years later, a one-pot synthesis using precipitation-driven alkyne metathesis was reported by Moore and coworkers.20 Thanks to the development of synthetic tools such as various coupling reactions, synthesis of shape-persistent macrocycles can be achieved with relative ease.   10   Figure 1.7  Synthesis of macrocycle 6 by (a) Stephens-Castro coupling in 4.6% yield, (b) Sonogashira coupling in 75% yield and (c) one-pot alkyne metathesis in 68% yield.  Proceeding with Moore and coworkers’ report on the improved synthesis of macrocycle 6 and its ability to be used for self-assembly, interest in designing and synthesizing new shape-persistent macrocycles boomed. These new shape-persistent macrocycles have proven to be ideal candidates for one-dimensional self-assembly through - interactions, which is an effective 11  approach to synthesize 1D nanostructures. After investigating self-assembly of planar macrocycles for 2 decades, researchers not only managed to synthesize numerous shape-persistent macrocycles, but also explored different techniques to achieve self-assembly,21 including solution-based self-assembly, sol-gel processing and surface-supported self-assembly. Breakthroughs in both synthesis and self-assembly contributed to this rapidly developing field.  1.2 Synthesis of Macrocycles After decades of study in macrocycle synthesis, chemists have been able to prepare diverse macrocycles in various ways. Among the different possible strategies, the majority can be classified into three types: (1) intramolecular ring closure of linear oligomers; (2) templated macrocycle synthesis; and (3) one-step cyclization between two or more monomeric residues. Each of these methods has unique benefits and drawbacks in terms of minimizing steps and maximizing yield of the desired product.  1.2.1 Intramolecular Ring Closure of Linear Oligomers Intramolecular ring closure of linear oligomers is widely used in nature, such as plasmids (cyclic DNA) in bacteria and α-amanitin (cyclic peptide) found in several species of the amanita genus of mushrooms. In both cases, the cyclic compound has a complicated structure composed of different residues lacking a simple repeating pattern. To produce such products, step-wise chain growing processes are necessary before the final ring closure step.    12  Scheme 1.1  Synthesis of cyclic oligomers. a) iPr3SiCCMgBr, [Ni(acac)2], THF. b) Bu4NF, CH2Cl2 and separation. c) 1,8-diiodoanthracene, [Pd(PPh3)4], CuI, NEt3/THF. d) Bu4NF, THF, then [Pd(PPh3)4], CuI, NEt3. acac=acetylacetanato.  However, for a macrocycle comprised of repeating units, such a general strategy can be extremely tedious and as a result, the synthetic procedures often hamper experimental exploration of such materials. In order to prepare a macrocycle with N repeating units, at least N-1 intermolecular bond-forming reactions and one intramolecular ring closure reaction is required (see Scheme 1.1 for an example).22 Due to lack of efficiency, this strategy is not widely used in macrocycle synthesis.  13  1.2.2 Template Mediated Macrocycle Synthesis Without pre-formed cyclization precursors, cyclization reactions may not yield desired products effectively. Although the rigidity of the precursor backbone greatly improved the macrocycle yield, the efficiency of the cyclization remained only modest in some cases. Additional measures have to be taken to further optimize the macrocyclization step. Thus, a template methodology was explored, first by Sanders and Anderson, to increase the yield of desired macrocycles. In Sanders and Anderson’s work, they first found that attempts at obtaining cyclic Glaser coupling products from monomeric porphyrin complex ZnL using conventional procedures only resulted in formation of insoluble material. However, when 4,4'-bipyridine (bpy) was used as template, a cyclic dimer (ZnL)2bpy was formed with bpy binding to both Zn(II) centers. Interestingly, if 2,4,6-tri(pyridin-4-yl)-1,3,5-triazine (tpyt) was used to template the same reaction, a cyclic trimeric architecture  (ZnL)3tpyt was produced with a yield of 50%. Similarly, the tpyt was bound to all three Zn(II) centers in the same macrocycle (Figure 1.8). 23 In both templated reactions, a template molecule first binds to multiple monomeric porphyrin complexes ZnL. As a result, monomers are oriented at positions suitable for macrocyclization; formation of linear oligomers of macrocycles with undesired geometry is inhibited. After further exploring the insight of templated macrocyclization, the authors found that the tpyt ligand clearly enhances formation of (ZnL)3tpyt, whereas bpy has a different effect, templating the formation of (ZnL)2bpy.  By judicious use of templates, the outcome of this reaction can be directed down either of two different pathways with good yields (70% for (ZnL)2bpy). 14   Figure 1.8  Structures of monomeric complex ZnL (left), macrocyclic dimer with template (ZnL)2dpy (middle) and macrocyclic trimer with template (ZnL)3tpyt (right).  While Sanders and Anderson used tpyt as template after they first synthesized macrocyclic trimer (ZnL)3 by another method, recently chemists started to find more suitable templates to prepare new macrocycles. A successful example of finding a proper template by computational study was reported by Lindsey and coworkers in 1999.24 A hexameric wheel of porphyrins was successfully synthesized by using a calculated tridentate template (Figure 1.9). To synthesize this macrocycle, they first constructed the structural model of the target cyclic molecule and measured its cavity size. In one example, the distances from the center of the porphyrin plane (at the metal center) to the center of the cavity of the cyclic hexamer are 15.2 and 19.3 Å for the p,p′- and m,m′- diaryl substituted porphyrins, respectively.  They also considered the size of metal ions in the center of each porphyrin. Since the template should provide a complementary fit, the arm length of this tridentate ligand can be calculated. As a result, ligand (7) was prepared for templating. By using this well-designed template, they 15  successfully synthesized their target molecule (8) in a one-step coupling reaction. In the absence of a template, this reaction would have generated a low yield of the desired product, at best.  Figure 1.9  Structures of a tridentate template (7) and hexameric wheel of porphyrins (8).  Since then, chemists have been exploring templated macrocyclization due to its simplicity and has since made a lot of breakthroughs. Templated macrocyclization has been used to prepare diverse macrocycles with different geometries as a result. Recently, Anderson synthesized a 12-porphyrin nano-ring by Vernier templating with a template having 6 binding sites (Figure 1.10).25 16   Figure 1.10  Synthesis of the 12-porphyrin nano-ring with templates.  After two decades of study, templation has become a very powerful tool in macrocycle chemistry. At the molecular scale, templates ensure that out of a myriad of possible chemical reactions, only the one that leads to desired product is promoted. It not only ensures the selective and effective synthesis of target macrocycles, but also offers a chance to avoid the otherwise tedious step-wise syntheses. However, macrocyclization through templation has its own drawbacks. First, although templated macrocyclization often produces product more effectively, it often requires one more 17  step after macrocyclization to remove the template. In some extreme cases, templates cannot be removed without breaking the macrocyclic structures. Besides, while cyclization with templates usually requires much easier synthetic procedures than conventional ring closure strategy, occasionally the formation of the template itself suffers from troublesome and low-yielding synthesis.  1.2.3 One-Pot Synthesis of Macrocycles Among all three kinds of synthetic methods for constructing cyclic structures, one-pot synthesis of macrocycles using repeated monomer units has an evident advantage over the other two methods: the starting materials are readily accessible and the target molecule is generated in a single step. One can easily produce macrocycles without the troublesome multi-step synthesis of either linear oligomers or templates. One family of well-studied macrocycles prepared through a one-pot reaction is arylene ethynylene macrocycles. As various kinds of transition metal-catalyzed cross coupling reactions, such as the Sonogashira reaction, Glaser couplings, Heck reaction, Negishi reaction and McMurry coupling, have been utilized to generate -conjugated linkage, there are multitudes of versatile approaches toward modular construction of arylene ethynylene macrocycles.  Although one-pot synthesis of macrocycles can be achieved by these coupling approaches, all of these approaches suffer from a common problem--low yield, largely in part that in a one-pot macrocyclization, the desired macrocycle must compete against a broad, statistical distribution of side-products including all linear or cyclic oligomers of different chain lengths. To improve upon the yield of a desired cyclic product, geometrical shape design must be considered. By using combinations of a small set of building blocks: ortho-, meta- and para-phenylenes (having bond 18  angles of 60°, 120° and 180°, respectively), chemists can design macrocycles with various geometries (Figure 1.11).26 The rigidity of each arylene ethynylene fragment not only results in the overall shape-persistence, but also makes the formation of the desired product more thermodynamically favorable by carrying some geometric information of the starting material to the desired product.   Figure 1.11  Schematic representations of phenylene-ethynylene macrocycles and the building blocks from which they are constructed.  Unfortunately, despite the availability of a variety of cross-coupling synthetic approaches, the efficient preparation of macrocycles is challenged by the disadvantage that these reactions are kinetically controlled. As a result, oligomers grown by coupling reactions may overshoot the length required by the target macrocycles; as these coupling reaction are not reversible, overgrown oligomers cannot contribute to the formation of desired product. In order to avoid the problem of “overshooting”, reversible reactions should be introduced to one-pot macrocyclization. With this 19  methodology, the overgrown linear or cyclic oligomers can be shortened to the desired chain length. Reversible reactions that are usually used in one-pot synthesis of macrocycles involve imine formation and coordination chemistry.  As a result, the geometric information of desired macrocycles carried by rigid monomers contribute greatly to the yield of thermodynamically favored product.  Although introducing reversible reactions to one-pot macrocyclization makes it possible to correct undesired bond formation, sometimes formation of macrocycles with undesired geometries may still occur. In general, the success in precise control of macrocycle geometry relies on the angles of the functional moieties of the monomer. If metal ions are involved in the macrocyclization process, the result may also depend on the preferred coordination geometry of the metal ions. However, macrocycles with unpredicted geometry often appear due to the flexibility of organic compounds. Even with rigid organic compounds, geometry of macrocycles may not be absolutely certain, for example, Lin and coworkers reported that macrocycles with 3-, 4-, 5-, 6-, 7-, and 8-fold geometries can be observed simultaneously as the product of a self-assembly process (Figure 1.12).27  The poor selectivity of macrocycle geometries not only leads to low yield of the desired product, but also requires extra steps to isolate the desired product from a complex mixture. Since those unpredicted macrocycles often have similar properties to the desired product, this purification may be cumbersome. In order to avoid these problems, rational design of macrocycles that are prepared with high selectivity is crucial. Although a rationally designed structure already has “structure-directed” contribution to macrocyclization, applying more restriction generally augments the existing structure-directed process leading to the macrocycles.  20   Figure 1.12  A mixture of macrocycles with various geometries from a self-assembly process.  In recent studies, Newkome and coworkers investigated forming macrocycles with ditopic 2,2′:6′,2″-terpyridine (tpy) building blocks (9) as ligands.28 The rigid ligand has a 120° angle which is designed to form hexameric macrocycles after coordination with Zn(II). However, they found unexpected pentameric, heptameric, octameric, nonameric, and decameric macrocycles as well as the desired hexameric macrocycle after the macrocyclization (Figure 1.13). This result once again proved that rigid organic linkers are sometimes more flexible than we expect.  Following this work, Li and coworkers reported a method to improve the selectivity of macrocycle geometry in 2014. Instead of utilizing molecular templates, their strategy was to increase the number of coordination sites within the macrocycle (shown in Figure 1.14).29 To achieve this, ligands 10 and 11 were synthesized. These multitopic ligands provide more noncovalent interactions and significantly increased the density of coordination sites in macrocycles 12 and 13. With increased density of coordination sites, the rigidity of the organic framework was reinforced to achieve better control over macrocycle geometry. 21   Figure 1.13  Macrocycles synthesized from one reaction with 2,2′:6′,2″-terpyridine building blocks (9).   Figure 1.14  Synthesis of hexagonal wreaths from ligands 10 and 11.  22  By employing multitopic terpyridine ligands, they were able to put more geometric constraints into macrocycles to help form discrete and thermodynamically stable structures. Those multitopic ligands not only increase the rigidity of macrocycles, but also prevent the formation of various undesired cyclic products. As a result, hexagonal wreaths (12 and 13) were the only macrocyclic products they isolated from the reaction (Figure 1.15). Hence, precise control over the geometry of the macrocycle is achieved without extraneous templates.  Figure 1.15  Structures of hexagonal wreaths 12 and 13.  Besides coordination interactions, intramolecular hydrogen bonding can also facilitate the formation of cyclic structures, such as cyclic amides.30  Studies show that the intramolecular hydrogen bonding between amide N-H and O help amides assume a folded conformation (Figure 1.16). The folded conformation can pre-organize the system for an efficient macrocyclization. As a result, even with flexible organic structures, cyclic amides can preferentially form over linear polymers. If combined with rigid structures, intramolecular hydrogen bonding interactions can facilitate selective synthesis of cyclic structures via one-pot synthesis. 23   Figure 1.16  Structure of a [2+2] amide macrocycle formed by flexible organic building blocks (left) and [3+3] Schiff-base macrocycles with rigid framework (right).  Similarly, intramolecular hydrogen bonding can contribute to the formation of cyclic Schiff-bases as well as cyclic amides. However, Schiff-bases do not contain a hydrogen atom that is directly connected to a highly electronegative atom, as a result Schiff-bases would not have hydrogen bonding. This problem can be mitigated if another functional group such as OH or NH2 is anchored close to the imine N atom. The first successful synthesis of shape-persistent Schiff-base macrocycles with intramolecular hydrogen bonding was reported by Nabeshima and coworkers in 2001 (Figure 1.16).80 For macrocycles prepared through amide/imine synthesis, usually two different organic components are required: one contains two carboxylic acid/aldehyde groups and another contains two amine groups. Since one-pot synthesis relies on self-assembly of monomer units, highly symmetric monomer units are usually used to simplify the self-assembly process. By using two kinds of highly symmetric AA and BB type monomers, chemists do not need to worry about 24  orientations of each monomer unit. These resulting macrocycles always contain two kinds of monomers with a ratio of 1:1. To clarify the symmetry of those macrocycles, they are usually named by the number of monomers in each macrocycle. For example, in Figure 1.16, the Schiff-base macrocycles composed of three diamines and three dialdehydes are called [3+3] Schiff-base macrocycles, while the amide macrocycle is named as [2+2] amide macrocycle. This [n+n] construction strategy is widely used in macrocycle chemistry.  1.3 Head-to-Tail Macrocycles  1.3.1 Metal-Free Head-to-Tail Macrocycles 1.3.1.1 Intramolecular Closure towards Head-to-Tail Macrocycles Although this [n + n] approach can lead to macrocycles with various geometry and different sizes, it has its own drawback. This approach requires equal amounts of diamines and dialdehydes; therefore, it can only be used to synthesize macrocycles with an even number of moieties. To overcome this drawback, another strategy needs to be applied. This new strategy is called head-to-tail cyclization. One of the well-studied compounds used for head-to-tail cyclization are peptides. Although linear peptides are the most commonly observed in nature, cyclic peptides play an important role in medicine and biology. Since the discovery of gramicidin S (Figure 1.17) by Russian microbiologist Georyyi Frantsevitch Gause and his wife Maria Brazhnikova in 1942,31  cyclic peptides have offered themselves as promising medicinal candidates that pre-organize their amino acid sequences into rigid conformations which are more resistant to enzymatic degradation in comparison with their linear variants.32 Moreover, some cyclic peptides have shown increased 25  receptor selectivity and bioavailability. The demand for cyclic peptides required more investigation in the chemical synthesis. In general, the synthesis of cyclic peptides is more challenging than that of linear peptides. Since the head-to-tail cyclization of peptides is an intramolecular reaction, it is usually carried out in highly dilute solution to inhibit intermolecular reactions. However, even in dilute solution, linear products can still form. Furthermore, macrocycles with different sizes may also form in one reaction. The formation of these by-products not only decreases the yield of the desired cyclic product, but also complicates purification.   Figure 1.17  Structure of gramicidin S.  After decades of effort, chemists have made a lot of improvement in optimizing the synthetic conditions of head-to-tail cyclization, including the nature of linear peptide precursors, coupling reagents, concentration of the reaction solution, temperature and so on.33 Despite the fact that cyclic peptides with various sizes can be synthesized with these improvements, most of those peptide macrocycles are prepared from linear peptide precursors. Similarly, head-to-tail peptoid macrocycles are often synthesized from linear peptoid precursors 34 (Figure 1.18 and Figure 1.19). By using this strategy, chemists have better control of the sizes of cyclic products and the nature of linear peptide precursors. However, this strategy relies on tedious step-by-step synthesis of 26  linear precursors. Moreover, the macrocyclization reaction usually has relatively low yield even under highly dilute conditions.  Figure 1.18  Synthesis of a peptide macrocycle from linear precursor.   Figure 1.19  Synthesis of a peptoid macrocycle from linear precursor.  1.3.1.2 Inspiration for One-Pot Synthesis of Head-to-Tail Macrocycles One-pot reactions are attractive to avoid time consuming syntheses. Different from the traditional one-pot reaction with low selectivity, more features have to be applied to find a new approach. To improve the selectivity of head-to-tail cyclization reactions, chemists choose rigid structures, such as aromatic rings, as macrocycle backbones. 27  A pioneering study on hydrogen-bonding-assisted macrocyclization was performed by Hunter and coworkers in 1994. 35  They synthesized macrocycles in 80~90% yield by using intramolecular hydrogen bonding to direct the cyclization process. These intramolecular hydrogen bonding interactions orient and pre-organize the linear intermediates into crescents so that the reactive functional groups at either end of the intermediates are brought into close proximity. As a result, the cyclization process occurs successfully. Although Hunter and coworkers explored the significance of using intramolecular hydrogen bonding to direct cyclization, their work did not draw much attention until later. In 2000, Gong and coworkers reported a new class of folding oligomers, known as crescent oligoamides.36 Although the products in this paper are still crescent (linear) oligomers, they show a tendency to fold into cyclic structures. In 2001, Gong and coworkers systemically investigated the significance of a stable three-center hydrogen bond in a partially rigidified structure.37 In this paper, they first reported a positive cooperativity in an intramolecular three-center H-bond. When combined with proper structural scaffolds, it can potentially serve as a reliable basic folding unit for designing a variety of unnatural folded structures.  Following those reports, Gong and coworkers applied their discoveries to macrocycle chemistry. A highly efficient, one-step macrocyclization was reported. By using this one-step macrocyclization, both oligoamide and oligohydrazide macrocycles were synthesized selectively (Figure 1.20). 38  These AB-type macrocycles bearing different exterior side chains can be produced in high isolated yields via hydrogen-bonding-assisted one-pot macrocyclization reactions in about one day under mild conditions. In the absence of the three-center intramolecular hydrogen bonding network, the yield for the respective cyclic hexamers (14) was found to be extremely low even in the presence of template or in highly dilute conditions.39 28   Figure 1.20  Structures of oligoamide macrocycle (14) and oligohydrazide macrocycle (15) reported by Gong and coworkers.  For this type of acyclic oligomer with hydrogen bonding interactions, a crystal structure showed that it required about 6.5 building blocks to form a helical turn.40 This suggests that cyclic hexamers would have much less strain compared to smaller or larger macrocycles. Only in the case of the acyclic hexamer precursor can a folded backbone of suitable length bring the amino and acid chloride end groups into close proximity, resulting in a rapid intramolecular cyclization with minimized strain. Although those macrocycles are not prepared using a head-to-tail approach, the idea of combining intramolecular hydrogen bonding and rigid macrocycle backbones really provided a possibility of preparing oligoamide macrocycles by a head-to-tail approach.  29  1.3.1.3 Hydrogen-Bonding-Assisted One-Pot Synthesis of Head-to-Tail Macrocycles Inspired by the careful analysis of the crystal structures reported by Gong and coworkers, Zeng and coworkers decided to use a non-symmetric precursor instead of two different, but both symmetric, precursors for the amide macrocyclization. This resulting non-symmetric amide is essential to allow chain extension to proceed in a linear manner through to the pentamer followed ultimately by cycle formation; in contrast, the reaction between two different but symmetric building blocks can only access cycles with even numbers. In 2008, Zeng and colleagues first reported an unusual 5-fold symmetric amide macrocycle (Figure 1.21). 41  A crystal structure clearly suggests the formation of the novel macrocyclic pentamer. Although the angle between amine and carboxylic acid from the same residue is about 60°, they reasoned that an oligomeric backbone incorporating an inward-pointing, continuous H-bonding network can cyclize a rigidified crescent pentamer into a macrocycle. Thereafter, Zeng and coworkers extensively explored such 5-fold symmetric macrocycles.  Figure 1.21  Chemical structure (a) and Crystal structure of macrocyclic amide pentamer: (b) top view, and (c) side view both with methoxy methyl groups in CPK representations.  In 2010, Zeng and coworkers demonstrated a novel strategy for the modular construction of diverse macrocyclic pentamers.42  By using a stepwise synthetic approach, they were able to 30  introduce different functionalities in both the interior and exterior of macrocycles (Figure 1.22). As a consequence, a precise pinpoint modification of macrocycles was attained. If treated with tetrabutylammonium hydroxide, the interior hydroxyl groups could be deprotonated. The resulting anionic forms of those macrocyclic pentamers demonstrated a high affinity for metal cations with a radius below ~1.5 Å.   Figure 1.22  Structures of macrocyclic amide pentamers 16, 17, 18 and 19 which can be used to bind metal ions with a radius of ~1.5 Å.  Later work also shows that, even for methoxybenzene-based macrocycles, anionic forms of the macrocycle could be prepared by treating it with tetrabutylammonium chloride salt (Figure 1.23). 43  This highly efficient chemo-selective monodemethylation and regio-selective double demethylation can eliminate up to two out of five methyl groups in the ring.  In 2011, Zeng and coworkers synthesized macrocyclic fluoropentamers 20 (Figure 1.24) by a stepwise method.44 In these fluoropentamers, their circular pentameric backbone is reinforced by intramolecular C-F…H-N hydrogen bonding interactions. The planar conformation allows 31  enhanced interplanar - interactions. In addition, studies show that every pentamer forms two intermolecular C=O…H-N hydrogen bonding interactions in dimers.  Figure 1.23  Regio- and chemo-selective demethylation of methoxybenzene-based macrocyclic pentamers.   Figure 1.24  (a) Structures of fluoropentamers 20. (b) Top and side views of crystal structure of 20a, illustrating the formation of interplanar H-bonds of 2.50 Å in length. Dotted cycles in (b) indicate the amide bonds that are twisted out of the plane to form stronger intermolecular H-bonds that enhance the interplanar aggregations.   While macrocycle 20a exhibits very poor solubility in organic solvents due to the enhanced intermolecular interactions, the other two planar macrocyclic fluoropentamers, 20b and 20c, 32  demonstrate better solubility because of modified hydrocarbon chains. Both of them were found to gelate organic solvents, largely derived from their strong tendency to form 1D stacked fibrillar structures stabilized by both interplanar hydrogen bonding-interactions and π-π stacking forces.44 Later in the same year, Zeng and coworkers prepared another 5-fold symmetric macrocyclic aromatic pentamer by a stepwise synthetic strategy using HBTU (2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate)-mediated amide coupling reactions.45  In this work, they prepared macrocyclic pyridone pentamers 21 (Figure 1.25). Both carbonyl oxygens and amide protons point inward to form a continuous intramolecular hydrogen bonding network. The conjugated macrocycle backbone becomes increasingly curved and eventually forms a pentagonal shape. Computational studies show that these macrocycles have a hydrophilic oxygen-containing cavity of 2.85 Å, which is nearly identical to the average coordination bond distance between K+ ions and covalently bound oxygen atoms. This calculation suggests a potential application of macrocycle 21 to bind to alkali metal ions.  Figure 1.25  (a) Chemical structures of pentamers 21a and 21b. (b) Top and side views of computationally optimized structures for 21a and 21b, with the exterior side chains replaced by methyl groups at the B3LYP/6-31G level, illustrating five-fold-symmetric planarity in 21a and 21b. 33  Experimental results further proved that macrocyclic pentamer 21 is capable of high-affinity recognition of metal ions. In addition, the resulting ion-containing macrocycles can self-assemble into 1D columns, and can further associate into ion-pair-induced fibers (Figure 1.26). Depending on the exterior chain anchored on the macrocycles, the thermodynamically favored packing mode could change from twist packing of pentamer 21a to eclipsed packing of pentamer 21b. This self-assembly can be tuned by alkali metal ions or their halide salts.  Figure 1.26  Computationally optimized structures of 1D columnar aggregates possibly formed by (a, b) [21aKBr]n and (c, d) [21bKBr]n at the B3LYP/6-31G level under periodic boundary conditions. The top-down views illustrate two possible packing modes and their relative energies. Side views, with the exterior side chains removed, illustrate the interplanar distances that dictate the strength of ionic interactions. In the CPK models, K+ = 1.38 Å and Br- = 1.95 Å. 34  Thus far, Zeng and colleagues successfully synthesized alkoxybenzene-based, fluorobenzene-based and pyridone-based macrocyclic pentamers via a step-by-step approach. Their work proved that they can tune both the interior and exterior of each residue of macrocyclic pentamers since macrocycles are built stepwise. This allows them to adjust not only the cavity properties to affect ion recognition, but also the macrocycle self-assembly process, such as packing mode.  Following all these works, Zeng explored combining different macrocycle backbones to further tune the cavity properties of macrocycles in a more precise and subtle way.46 Thanks to the stepwise synthetic approach, they could use different monomeric residues during the synthesis of one macrocycle. As a result, they can control both the monomeric component and the residue sequence of macrocycles.  Figure 1.27 shows the structures of hybrid macrocyclic pentamers employed to tune ion recognition properties by Zeng and coworkers. In this work, they used hydrogen bonding rigidified macrocyclic pentamers to effectively and convergently pre-organize different electronic traits (e.g., O and F atoms) and influence steric/ hydrophobic (e.g., OCH3 and OH groups) factors into adjustable cavities, thereby permitting a systematic modification of ion recognition property.46 By combining hydrogen-bonding-assisted macrocyclization with the head-to-tail approach, Zeng and coworkers proved that they can synthesize planar 5-fold symmetric shape-persistent macrocycles with different backbones, different interior and exterior properties. These macrocyclic pentamers not only demonstrate unique geometric properties, but also show various potential applications. 35   Figure 1.27  Structures of hybrid pentamers employed by Zeng and coworkers to probe the potential for selective recognition of metal ions by a family of macrocyclic hybrid pentamers with modularly tunable interior properties. The ion-extraction profiles determined under the identical conditions for the three well-known ligands, i.e., 18-crown-6, dibenzo-21-crown-7, and kryptonfix-222, were included for comparison purpose. The ions whose extractability lies within 80% of the most extractable ones are additionally highlighted below the most extractable ones in the cavity.  However, despite of the precise control over the properties of those amide pentamers, all those 5-fold symmetric macrocycles suffer from tediously long synthetic routes. For example, pentamers 21a and 21b were both made by a stepwise construction strategy requiring15-16 steps with an overall yield of 1-2%. Pentamers 20b and 20c were prepared after about 15 steps also with an overall yield of 1-2%. Such time-consuming synthetic routes with very low overall yield greatly limited further application of those macrocycles as well as their practical synthetic utilities. To overcome this low-yielding synthetic bottleneck, Zeng and coworkers turned to one-pot, multi-molecular macrocyclization to efficiently generate macrocyclic pentamers. In 2011, they 36  published a paper about the one-pot synthesis of pentagon-shaped amide macrocycles with alkoxybenzenes as backbones.47 By using one-pot, multimolecular macrocyclization with POCl3 as coupling reagent, they synthesized various alkoxybenzene pentamers with decent yield (Figure 1.28(a)). For instance, pentamer 16, which was synthesized after 15-16 steps with a yield of 1-2%, can be prepared in 12 hours with 46% yield by a one-pot reaction. Notably, 5-fold symmetric macrocycles were selectively generated in this one step reaction.  Figure 1.28  One-pot preparation of circular (a) alkoxybenzene pentamers and (b) pyridone pentamers from their respective monomers.  Later in the same year, they reported another one-pot macrocyclization that can be used to generate 5-fold symmetric pyridone macrocycles.48 In this work, they successfully used BOP-37  mediated one-pot macrocyclization (BOP = benzotriazole-1-yl-oxy-tris-(dimethylamino)-phosphonium hexafluorophosphate) to prepare various macrocyclic pyridone pentamers (Figure 1.28(b)). For example, pentamer 21b was generated in a one-pot reaction after 30 hours with a yield of 18%. Zeng and coworkers believe that the success of the one-pot macrocyclization is facilitated and guided by internally placed intramolecular hydrogen bonds to allow for the highly selective formation of 5-fold symmetric macrocycles. Compared to the previous lengthy stepwise synthetic strategy, the one-pot macrocyclization is much more cost-effective and time-saving. It established a better access to diverse macrocyclic amide pentamers.  Since this one-pot macrocyclization allows for the highly selective formation of 5-fold symmetric macrocycles, rather than 4-fold or 6-fold symmetric macrocycles, macrocyclic pentamers with different periphery can be prepared by reacting monomers with higher oligomers bearing different exterior side chains (Figure 1.29).49 This result suggests more than a success in the synthesis of macrocycles with different periphery via one-pot macrocyclization. It implies that the addition of monomeric residues onto oligomers is much faster than the reaction between two oligomers. This hypothesis was then supported by kinetic simulation of reaction rates.50 Their findings indicate that the mechanism of this one-pot macrocyclization is predominantly conducted through a chain-growth process. With the synthetic improvement brought by one-pot macrocyclization, macrocyclic amide pentamers can be generated in a less time consuming way with better yield. Thereafter Zeng and coworkers were able to explore further applications. In 2013, they reported macrocyclic pyridone pentamers as a highly efficient organocatalyst for the direct arylation of unactivated arenes.51 In the presence of KOtBu, this novel macrocycle allows for the efficient construction of biaryls via 38  direct arylation of unactivated arenes with iodoarenes and bromoarenes. This transition-metal-free organocatalyst demonstrates excellent yields for broad substrates with as low as 2 mol% catalyst loading. In 2014, pentamer 21b was found to exhibit size-dependent patterned recognition.52 It can extract larger ions such as Cs+, Ba2+, Au+, K+ and Rb+ preferentially over 18 smaller metal ions from the aqueous phase into chloroform phase.  Figure 1.29  Synthesis of macrocyclic pentamers from monomers and oligomers.  Although alkoxybenzene-based and pyridone-based macrocyclic pentamers can be synthesized through a one-pot reaction in the presence of their own coupling reagents, Zeng hasn’t been able to find a proper coupling reagent suitable to facilitate formation of fluorobenzene-based macrocycles. BOP or POCl3 only works efficiently with their own monomeric residues. Based on such observations, Zeng and coworkers hypothesized that every type of monomer building block destined to form the most stable circular pentamer requires its own unique "cognate" macrocyclization reagents that are "orthogonal" to each other and function well only with their own specific set of "cognate" monomer units. While Zeng and coworkers were searching for a “cognate” macrocyclization reagent required by fluorobenzene-based monomer units, they found trimethylaluminum as a very surprising 39  macrocyclization reagent. Ignoring the fact that macrocyclic pentamers are more energetically favored than hexamers, trimethylaluminum helps selectively form more strained macrocyclic hexamers via one-pot macrocyclization of alkoxybenzene-based monomers (Figure 1.30(a)).53     Figure 1.30  One-pot preparation of circular (a) alkoxybenzene pentamers (22) from monomers and (b) pyridone pentamers (23) from dimers.  Just as POCl3 only functions well with alkoxybenzene monomers, and not with the other monomers, AlMe3 only works against alkoxybenzene monomers, but not against pyridone monomers. So far, they haven’t found a coupling reagent leading pyridone monomers to form macrocyclic hexamers. To solve this problem, Zeng’s group presented a cyclization from pyridone dimers (Figure 1.30(b)).54 Starting from the pyridone dimers, macrocyclic hexamers 23a and 23b, 40  both could be produced with yields of 31−56% by using BOP, but not AlMe3 or POCl3. Resulting strained macrocycles 23a and 23b exhibit highly selective recognition of Cu2+ ions. This property has been tested in reducing Cu2+ content in artificial seawater. Zeng’s group has clearly shown hydrogen-bonded rigidified aromatic structures as efficient building blocks in head-to-tail macrocycles. Via stepwise synthesis, a novel family of 5-fold symmetric amide macrocycles can be prepared with the presence of intramolecular hydrogen bonding. Not only each individual interior and exterior can be modified, but also each residual backbone can be tuned to build hybrid macrocyclic pentamers. If conducted with proper coupling reagents, time-saving one-pot strategies can be used to facilitate access to hydrogen-bonding-assisted macrocyclization. In addition, this highly selective head-to-tail approach does not limit macrocycle geometry to 5-fold symmetry. All in all, by applying a head-to-tail approach to hydrogen-bonding-assisted macrocyclization, Zeng and coworkers recently expanded the one-pot hydrogen-bonding-assisted head-to-tail macrocyclization strategy, which is a great addition to the macrocyclization toolbox. As a recently emerging concept, one-pot hydrogen-bonding-assisted head-to-tail macrocyclization strategy is not limited to amide macrocycles. In 2013, Flood and coworkers reported a high-yielding, one-pot and multigram-scale synthesis of C5-symmetric cyanostar macrocycles which takes advantage of Knoevenagel condensations (Figure 1.31).55 The idea of C5-symmetric cyanostar macrocycles arose during examination of a hexameric design. Also inspired by Gong’s work, Flood and coworkers found the preferred turning angles would probably facilitate a pentameric C5-symmetric structure. Synthesis of this cyanostar macrocycle was attained by treatment of 24 with cesium carbonate as the catalytic base. The carbonate base or the deprotonated benzylic anion may help the macrocyclization by assisting intramolecular self-41  organization. For the first time, electropositive cyano-stilbene-based CH groups were used to hydrogen bond with anions inside the cyanostar cavity. With hydrogen-bonding-assisted self-organization, cyanostar 25 was obtained with a yield of 81%. The easy access to the cyanostar framework provided a foundation on which to further investigate their interesting properties.  Figure 1.31  One-pot synthesis of cyanostar 25.  Unlike the semi-planar structures of Zeng’s amide macrocyclic pentamers, cyanostars 25 are shaped like shallow bowls (Figure 1.32(a)). Unexpectedly, however, the dimers meet at the seams created by their inner rims rather than being nested together. In addition, following the suggestion by Szumna,56 two stereoisomers (M/P) were defined (Figure 1.32(b)), and it is believed that cyanostar dimers racemize between the M and P stereoisomers in solution. The crystal structure also revealed that the two macrocycles are rotationally offset from each other, which was attributed to a combination of electrostatic complementarity and steric effects involving the t-butyl groups. Interestingly, this dimeric structure remains after cyanostar 25 binds to large anions such as BF4-, ClO4- and PF6-. Those large anions are usually weakly coordinating 42  anions and require positive charges to be captured. They can be bound to cyanostar by size-selective recognition to form sandwich complexes between two cyanostar macrocycles.   Figure 1.32  (a) Sandwiches of two bowl-shaped cyanostar macrocycles 25 result in a mixture of four possible stereoisomers. (b) When the cyanostar macrocycles are viewed from the tops of their bowls stereoisomers with either M chirality or P chirality can be defined.   Figure 1.33 (a) Cartoon representation of [3]rotaxane 26-TBA+ and (b) X-ray crystal structure of [3]rotaxane 26-TBA+ (solvent molecules, TBA+ and protons are removed for clarity). The M-P isomer is shown. (TBA: tetrabutylammonium)  Taking advantage of this selectivity to large anions, Flood and coworkers prepared a dialkylphosphate[3]rotaxane 26 (Figure 1.33). Additionally, the anion-induced dimerization of 43  cyanostars into 2D self-assembled crystals and into 3D crystalline solids were studied by STM and XRD respectively.57  While cyanostars can interconvert between multiple conformations rapidly (Figure 1.34), the noticeable dynamic motions of single cyanostars originate entirely from the olefinic units, while the aromatic units remain static. Flood and coworkers believed that a single cyanostar molecule is both shape-persistent and flexible at the same time.58 And this coexistence of shape persistence and flexibility is unperturbed even with a diglyme guest molecules bound to form a sandwich complex.  Figure 1.34  Rocking pathways between all-out conformers (pink P) and their enantiomers (blue M) are established through corresponding transition states (yellow). The free energies were calculated by DFT with implicit solvation.  1.3.1.4 One-Pot Synthesis of Head-to-Tail Macrocycles without Hydrogen Bonding Both Zeng’s macrocyclic amide pentamers and Flood’s cyanostars are prepared through a hydrogen-bonding-assisted head-to-tail macrocyclization. Not limited to this type of assisted self-assembly, templating without intramolecular hydrogen bonding, one-pot head-to-tail 44  macrocyclization with diverse building blocks is still available. For instance, in 2005, Srikrishna’s group reported a C3-symmetric macrocycle by head-to-tail cyclotrimerization of an unsymmetrical diene (Figure 1.35).59 This highly regio- and stereoselective metathesis reaction generated a 24-membered macrocycle with a hydrophilic cavity as a surprising product, instead of the expected AB-ring system of taxane. The macrocycle can accommodate one molecule of water as a guest via hydrogen bonding with the three inward-pointing hydroxyl groups. This macrocycle shows potential for being used as a chiral host molecule.   Figure 1.35  Head-to-tail cyclotrimerization of an unsymmetrical diene leading to a C3-symmetric macrocycle instead of AB-ring system of taxane.  In 2006, Fukuyama and coworkers produced another chiral macrocycle with a thiazoline framework (Figure 1.36).60 This efficient construction can lead to cyclic oligomers with various sizes. These chiral cyclic oligothiazolines showed molecular recognition and anti-cancer properties in the following works.61 In 2007, the same research group reported the successful synthesis of thiazoline-thiazole hybrid macrocycles via head-to-tail macrocyclization. Metal coordination studies revealed their potential as highly selective receptors for heavy metal ions.62 However, the syntheses of both thiazoline and hybrid macrocycles suffer from competing reactions of macrocycles with different geometries.  45   Figure 1.36  Thiazoline macrocycles (left and middle) and thiazoline-thiazole hybrid macrocycles (right) prepared via head-to-tail approach by Fukuyama and coworkers.  1.3.2 Head-to-Tail Metallomacrocycles In the field of macrocycle chemistry, there is another type of head-to-tail macrocycle: metal-containing head-to-tail macrocycles. The basic synthetic strategy is the linking of more or less rigid organic ligands with suitably disposed metal ions. Depending on the stereochemical preferences of metal centers and the shape of organic linkers, cyclic structures can be achieved by self-assembly reactions. Metal-mediated self-assembly has been recognized as an excellent strategy to build up well-defined molecular architectures with an isolated, functional nanospace. To date, a large number of self-assembled macrocyclic molecules have been reported. Internal spaces of these molecules have their unique shape and volume, and thereby provide a variety of functions such as molecular recognition and specific chemical reactions. The head-to-tail approach occupies an important place in metal-mediated macrocycles. In 1995, Lechosław Latos-Grażyński and coworkers reported a novel cyclic gallium(III) porphyrin trimer 25.63 In the trimer, each gallium ion is located in the center of a porphyrin, and the gallium(III) complex has a head-to-tail trimeric structure with the pyrrolic-alkoxide groups 46  forming bridges from one macrocycle to the metal in the adjacent macrocycle. Extension of this work showed that with different metal ions such as iron(III) and manganese(III), cyclic trimers with similar structures could be formed as well. 64  Interestingly, experiments supported the formation of heterometallic (gallium(III), iron(III), manganese(III)) cyclic trimers (Figure 1.37).   Figure 1.37  Structures of a repeating unit of cyclic trimer (left) and cyclic heterometallic trimer 27 (right). Phenyl groups are omitted for clarity.  With modified organic linkages, porphyrin could also be applied to producing cyclic molecules of different geometry, for example, a 4-fold symmetric macrocycle with a porphyrin framework was published in 2008.65 The cyclic self-assembled tetramer was produced from an asymmetric meso-ethynylpyridyl-functionalized Zn(II)-porphyrin (Figure 1.38). The change of functional groups from 2-methoxy in 27 to 5-ethynylpyridyl led to the geometric conversion from 3-fold to 4-fold symmetry. With four porphyrin structures in the macrocycle, this tetramer demonstrated femtosecond transient absorption and rapid energy transfer (3.8 ps-1) between porphyrin subunits.65 47   Figure 1.38  Assembly of Zn4 cyclic tetramer in a non-coordinating solvent. Solubilizing groups of the space-filling tetramer model (right) have been omitted for clarity.   Figure 1.39  Structures of tetranuclear Mn(II) complexes 284+. The counter ions are (ClO4)-.  Similarly, head-to-tail metallomacrocycles can be prepared even with non-macrocyclic ligands. For example, the synthesis of complex 28 (Figure 1.39) represented a new route to the formation of tetranuclear Mn-complexes without the use of any small auxiliary ligands.66 In this case, metal ions were located in the binding sites of less rigid and non-macrocyclic ligands and the mononuclear complexes further self-assembed into tetranuclear metallomacrocycles. 48  This new route to prepare head-to-tail metallomacrocycles was not limited to metals with octahedral coordination geometry. Metal ions including Cu(II), Pd(II) and Pt(II) can also self-assemble into metallomacrocyclic complexes as well. In 1999, Tuchagues and coworkers reported a monomer-oligomer interconversion of Cu(II) complexes.67 While in monomer state, each copper ion has a square planar geometry in its individual complex. Several copper monomeric complexes can induce a self-assembly reaction via formation of metal-imidazolate nitrogen coordination bonds between adjacent units to form an oligomer. As a result, a 4-fold or 6-fold symmetric metallomacrocycle is formed.   Figure 1.40  The monomer-oligomer conversion of Cu(II) complexes controlled by pH.  Since this process involved the donor ability of imidazole and that the donor ability of imidazole’s nitrogen atom can be concealed by a proton at low pH, the monomer-oligomer 49  conversion can be thus controlled by varying pH. The size and geometry of the self-assembled oligomeric structures could be tuned through the design of the ligand, evident in the respective tetranuclear and hexanuclear metallomacrocyclic structures in Figure 1.40. The steric interaction played an important role in tuning the geometry of the macrocycles. Pd(II) and Pt(II), which also have square planar coordination geometries, are playing more important roles in modern macrocycle chemistry. This is due to their great potential in host-guest chemistry and catalysis. The current interests of Pd and Pt macrocycles and molecular cages arose from Fujita 68  and Stang’s 69  work. The majority of previously reported Pd or Pt molecular assemblies were ionic, but several examples of neutral Pt(II) molecular macrocycles were known as well.  In 2001, Williams and coworkers reported a trinuclear Pd(II) head-to-tail macrocycle (Figure 1.41).70 The preparation of this 3-fold symmetric macrocycle was conducted by cyclometalation of 1,3-bis(1-alkylbenzimidazol-2-yl)benzene with palladium acetate. After the formation of mononuclear complexes, trimerization of monomers led to the trimeric complex 29.   Figure 1.41  Structure of trinuclear Pd(II) head-to-tail macrocycle 29.  In 2006, Marchetti and coworkers reported the synthesis of a 4-fold symmetric Pd(II)-containing macrocycle 30 (Figure 1.42). 71  In this macrocycle, each palladium atom was 50  coordinated to the three nitrogen atoms of the anionic ligand, while the fourth coordination position was occupied by the amidato oxygen atom of an adjacent unit.  Interestingly, with slight modification of the R groups, a conversion from 4-fold to 6-fold symmetry was observed.72 This unexpected hexameric structure, rather than previously reported tetramers, was possibly attributed to the more rigid nature of the directly attached phenyl substituents as compared to the ethylphenyl side chains in the tetramers. Due to their direct attachment to the amide N groups, the phenyl groups were less flexible in hexamers. The internal hole formed by the hexameric array also appears to provide the perfect amount of space to accommodate the six phenyl groups, with an approximate wheel radius of 5.2 Å.  In 31, the twelve phenyl rings were divided into two groups of six, with one set pointed inward, forming the spokes of the wheel, and the other set pointed outward, forming the paddles. Each set of phenyl rings, including the phenyl spokes and paddles, showed alternating up-down orientations.  Figure 1.42  Structures of 4-fold (left) symmetric macrocycle 30 and 6-fold (right) symmetric molecular “paddlewheel” 31.  51  A similar synthetic approach has been applied to platinum. In 2007, Wang’s group demonstrated that NPA (N-(2’-pyridyl)-7-azaindole) was a very effective ligand in directing intramolecular C-H activation on a Pt(II) center. As a key step, the C-H bond cleavage contributed to the self-assembly of a cyclic organoplatinum Pt4 structure 32 (Figure 1.43).73 X-ray diffraction showed a large distortion of the 7-azaindolyl N atom from square planar geometry at the Pt center in order to minimize the steric interactions between adjacent pyridyl groups. As a result, the internal cavity in 32 has a distinct tetrahedral shape, as defined by the four interior pyridyl groups.   Figure 1.43  Structure of tetranuclear Pt(II)-containing macrocycle 32, Pt4(N,C,N-NPA)4(CH3)4. NPA = N-(2’-pyridyl)-7-azaindole.  Polynuclear metallocyclic and cage-like complexes with high symmetry and architectural beauty have attracted great interest in the field of supramolecular chemistry. Not only can these often complex molecular frameworks mimic certain natural structures, but they can also provide a wide variety of potential ‘‘working’’ functionalities. As a result, these intriguing assemblies have applications in many different fields such as bioinorganic chemistry and materials chemistry. And as a powerful tool, head-to-tail synthetic approaches are playing very important roles in this field.  52  1.4 Schiff-Base Macrocycles  1.4.1 Schiff-Base Chemistry--Salens and Salphens Schiff-bases are named after Germany chemist Hugo Schiff for his discovery of Schiff bases. They are functional groups with the structure R2C=NR’, where R’ is not a hydrogen atom. Schiff bases are usually produced by a condensation reaction between an amine and an aldehyde. The ease and reversibility of condensation make it a useful tool for synthesis, including the ring closing step in macrocycle formation. The undemanding synthetic conditions tolerate the presence of other more sensitive functional groups, which can be used for coordination or other applications. Due to the reversibility of condensation, this reaction usually leads to the most thermodynamically stable compound instead of a mixture of various species. One important Schiff-base structure is N, N’-bis(salicylidene)ethylenediamine, also known as salen. Salen and its derivatives are usually prepared by the Schiff-base condensation of salicylaldehyde derivatives with ethylenediamine (Figure 1.44). Similarly, salphen is synthesized from salicylaldehyde and phenylenediamine. Both salens and salphens are famous for their ability to coordinate to transition metals to form complexes, often some compounds exhibiting useful properties, for example, Jacobsen’s catalyst 33 with manganese is used to enantioselectively transform prochiral alkenes into epoxides.74 Salen-like compound 34 also contains a N2O2 pocket for coordination. With a zinc ion bound in the pocket, compound 34 shows electroluminescence in the blue region of the spectrum when it is incorporated into a layered device (Figure 1.45). 75 53   Figure 1.44  Synthesis of salen and salphen compounds.   Figure 1.45  Salen-like compounds for asymmetric epoxidation (33) and electroluminescence (34).  1.4.2 Schiff-Base Macrocycles Considering the advantages of N2O2 pockets in coordination chemistry and the attractive nature of Schiff-base condensation in macrocyclization, many chemists have synthesized Schiff-base macrocycles with salen or salphen moieties. The most famous example is the Robson macrocycles, which have two N2O2 pockets to incorporate two metal ions in very close proximity (Figure 1.46).76  54   Figure 1.46  Chemical structure of a dinuclear Robson macrocycle complex 35.  Different metals have been successfully incorporated into Robson macrocycles, and they can show some unique properties such as antiferromagnetic exchange interactions77 of macrocycle complex 36 and catalytic asymmetric cyclopropanation of styrene78 with macrocycle complex 37 (Figure 1.47).  Figure 1.47  Structures of dinuclear complexes 36 and 37.  In 1995, Reinhoudt and coworkers first synthesized a type of shape-persistent Schiff-base macrocycle 38 by using Ba2+ as template (Figure 1.48).79 Unfortunately, the Ba2+ ions could not 55  be removed after formation of the macrocycles. At that point, it was believed that templating is necessary to prevent linear oligomers from forming.  That idea was proven to be wrong several years later, when macrocycle 39 was synthesized in the absence of templates.80 Instead, chiral diamines were used in macrocyclization process. Although macrocycle 39 lacked functional groups for coordination or other applications, the formation of this macrocycle was still a breakthrough since it suggested macrocyclization is possible even without metal templation.  Figure 1.48  Structures of [3+3] Schiff-base macrocycles.  In 2001, Nabeshima and coworkers isolated a [3+3] shape-persistent Schiff-base macrocycle 40 prepared without metal templation.81 The interior hydroxyl groups formed hydrogen bonds with the imine nitrogen atoms. This noncovalent interaction promoted the formation of macrocycles without a template. However, the macrocycle took two weeks to prepare and had very poor solubility in organic solvents. In order to further study this type of macrocycle, increasing the solubility is crucial. To do so, our group introduced long alkoxy chains to the rigid aromatic rings.82 56  The resulting macrocycles with increased solubility can be used as templates for metal cluster formation. In the presence of excess zinc acetate, macrocycle 41 can complex three Zn2+ ions in its three salphen pockets and four more Zn2+ are bridged by acetate groups.83   1.5 Design of Schiff-Base Macrocycles  1.5.1 Geometry of Macrocycles To be used for host-guest chemistry, a macrocycle must have at least one cavity in which a guest molecule can fit. Both the size and shape of this cavity play important roles in binding guest molecules selectively. Similar to the situation in coordination chemistry, a macrocycle needs to provide a binding site with the proper size and coordination geometry in order to bind to specific metal ions. Thus synthesizing macrocycles with various sizes and geometries is crucial for further study and applications. As the applications of supramolecular chemistry become more diversified, precise control over the size and geometry of macrocycles is becoming more important in order to address topics like molecular machines and sensing. To obtain macrocycles with a certain geometry, there are two major problems to be solved. The first one is to avoid or suppress the formation of linear products. This problem is usually solved by conducting reactions at very low concentration to inhibit the chance of intermolecular reactions. Another obstacle is the presence of competing reactions leading to macrocycles with unexpected geometries. More efforts are required in the design of molecules to conquer this problem. Even with the same geometry, cyclic molecules of different sizes may demonstrate dramatically different properties. Usually, sizes of macrocycles can be tuned by using organic 57  building blocks with similar shape but different lengths. For example, larger [3+3] Schiff-base macrocycle 42 can be synthesized by replacing catechol moieties in macrocycle 41 with phenyleneethynylene dialdehyde (Figure 1.49).84 Compared to tuning macrocycle sizes, tuning macrocycle geometry is usually more difficult since a change in macrocycle geometry usually requires organic moieties with different shape, which implies a new design.  Figure 1.49  Comparison between two [3+3] macrocycles.  1.5.2 [n + n] Schiff-Base Macrocycles Since Nabeshima and coworkers isolated the [3+3] Schiff-base macrocycle 40 (Figure 1.48),81 Schiff-base macrocycles with salen or salphen building blocks have drawn a lot of attention from chemists. The OH group adjacent to imine not only provided a hydrogen atom for intramolecular hydrogen bonding, but also formed a binding pocket together with imine groups.  58  Following this work, our group synthesized other shape-persistent [n+n] Schiff-base macrocycles with different geometries,85 such as 2-, 3-, and 6-fold symmetric macrocycles (Figure 1.50). Although those shape persistent macrocycles have different geometries, they share the same design strategy. All of them were synthesized by condensation between rigid diamines and dialdehydes. In every case, OH groups are anchored adjacent to aldehyde groups so that the synthesized salen or salphen moieties can pre-organize the structure for an efficient macrocyclization step.   Figure 1.50  Structures of Schiff-base macrocycles with 2-, 3-, 6-fold symmetry.  Because of the extra constraints from hydrogen bonding, the structure-directed information carried by diamines and dialdehyde can be successfully transferred into cyclic products. The geometry of macrocycles can be easily predicted by the angles between two amine groups and angles between two aldehyde groups (Figure 1.51). For example, in the 3-fold symmetric macrocycle in Figure 1.50, there is a 60° angle between amine groups and a 180° angle between aldehyde groups. A trigonal shape was predicted and experimental results showed the formation of the predicted structures with high selectivity. 59   Figure 1.51  Prediction of macrocycle geometry from the angles between functional groups.   1.6 Goals and Scope My thesis is concerned with the synthesis, structures, and behaviors of two different types of macrocycles made by head-to-tail assembly, and both previously reported by our group. In 2010, Guieu et al. reported the one-pot reaction to form campestarenes,86 a novel family of Schiff-base macrocycles with 5-fold symmetry. A 6-nitrosalicylaldehyde derivative was first reduced with sodium dithionite to give 6-aminosalicylaldehyde, which spontaneously condenses into the macrocycle with 5-fold symmetry. This was very surprising, since the researchers expected to form a molecule with 6-fold symmetry based on the geometry of the precursors. In the previous study, our group found that campestarenes can undergo dimerization and aggregation in solution (Figure 1.52). This behavior not only limits their solubility in organic solvents, but also prevents campestarenes from being further studied. Theoretically, campestarenes may undergo a tautomerization process between enol-imine form and keto-enamine form (Figure 1.53). This behavior is interesting to study, but the tautomerization behavior may only be revealed in the absence of dimerization. 60   Figure 1.52  One-pot synthesis of Campestarene (left) and its dimerization behavior (right).  The research goal of this project was to synthesize campestarenes with increased solubility and no dimerization behavior by incorporating bulky substituents. Once that is achieved, then the study of the tautomerization behavior of campestarenes would follow. The results will be discussed in Chapter 2.   Figure 1.53  Tautomerization between the enol-imine (left) and the keto-enamine (right) forms. The NH tautomer may be formally depicted as a zwitterionic iminium-phenolate or a neutral keto-enamine in the extreme cases. (The notation Ha and Hb will be used in the discussion.)  During the study of campestarenes, it was found that regular campestarenes suffered from poor stability to weak acids, weak bases and relatively high temperatures (323K). In order to study the coordination chemistry with campestarenes, the stability of those macrocycles need to be improved. Due to more steric hindrance in ketones than in aldehydes, imine synthesized from ketones are more kinetically stable. To solve the stability issue of regular campestarenes, one of my side projects was to synthesize methyl or phenyl substituted campestarenes. Chapter 3 61  describes the synthesis of methyl or phenyl substituted campestarenes to show the attempts to prepare more stable campestarenes for further study and applications. Another goal of the research concerned head-to-tail metallomacrocycles. In 2010, Frischmann and MacLachlan reported head-to-tail self-assembled Pt4 macrocycles.87  An interesting property of these Pt4 macrocycles was their capacity to self-assemble into columnar structures in solution and in the solid state. However, the macrocycles had a relatively small pore that, as a result, yielded nanotubes with small channels when they aggregated (Figure 1.54).  Figure 1.54  Synthesis of head-to-tail Pt4 macrocycles and their MALDI-TOF mass spectrum which shows aggregation.  Thus, the goal of this project was to adjust Pt4 metallomacrocycles to yield larger pores. Just as the Schiff-base macrocycles discussed in section 1.5.1 could be extended by changing the precursor, we anticipated that extending the length of the pyridylsalicylaldehyde used to make the Pt4 rings would lead to larger macrocycles that give large-pore nanotubes when stacked. Also, the 62  additional rigid groups may result in a higher stacking number without dramatically diminishing solubility. Although our strategy led to an unexpected change in geometry, the resulting macrocycle did undergo stacking as anticipated. The results of this investigation will be discussed in Chapter 4. Overall, the objectives of the research projects have been met. This research helped to better understand the previously unknown tautomerization behavior in campestarenes and also has led to the discovery of new triangular metallomacrocycles. These developments opened the door to future investigations on head-to-tail macrocycles based on Schiff-base chemistry and metal coordination.   63  Chapter 2: The Rich Tautomeric Behavior of Campestarenes†  2.1 Introduction  The Schiff-base condensation of aldehydes with amines is an effective route to imines. By preparing a molecule with two aldehyde functional groups and one with two amine functionalities, Schiff-base condensation can yield polymers, oligomers or macrocycles. For the synthesis of Schiff-base macrocycles, the precursors should carry enough information for cyclization; by using rigid organic molecules with proper geometric information, chemist could realize control over macrocycle geometry (Figure 2.1(a)). Different from the typical synthetic approach to Schiff-base macrocycles, this work is about head-to-tail Schiff-base macrocycles.  2.1.1 Campestarene In 2011, Guieu and MacLachlan reported a shape-persistent Schiff-base macrocycle composed of one kind of precursor. Despite the 120°angle in the building blocks, which was designed for hexameric structure, a pentameric macrocycle was the sole product of the one-pot reaction. This 5-fold symmetric macrocycle was named as “campestarene”. Intramolecular                                                  † A version of this chapter has been submitted for publication: Chen, Z.; Guieu, S.; White, N. G.; Lelj, F.; MacLachlan, M. J. “The Rich Tautomeric Behavior of Campestarenes” Chem. Eur. J. 2016, 22, 17657-17672.  64  hydrogen bonding provided extra angle distortion leading to a geometry change from 6-fold symmetry to 5-fold symmetry (Figure 2.1(b)). Macrocycles with 5-fold symmetry are relatively rare, but some notably beautiful examples were reported by Flood,88 Zeng,89 Moore,90 Gong,91 and others.92 Also, there were other examples where macrocycles with 5-fold symmetry have been prepared in a lengthy, stepwise procedure, or are obtained as a minor by-product.93   Figure 2.1  (a) Prediction of [3+3] macrocycle geometry from precursors; (b) Predicted 6-fold geometry and actual 5-fold geometry of campestarene.  While these one-pot reactions have high yields, the resulting campestarenes were very difficult to study due to their poor solubility in organic solvents. Previous NMR experiments showed that those macrocycles undergo a dimerization progress, which may be responsible for the 65  poor solubility. Therefore, the primary goal of this work was to develop new campestarenes with improved solubility for further studies, such as tautomerization.  2.1.2 Tautomerization Tautomerization is a well-known example of proton transfer, characterized by the rearrangement of a labile proton and a double-bond,94  which can happen both in intra- and intermolecular fashions,95 and be catalyzed by acid or base.96 In the particular case of the enol-imine motif shown in Figure 2.2, a strong intramolecular hydrogen bond is involved in keeping the molecule nearly planar, and this conformation facilitates intramolecular proton transfer. The enol-imine form is generally the most stable at room temperature,97 even if it is in rapid equilibrium with the keto-enamine form. Only in a few cases has the keto-enamine been reported as the major isomer in salicylaldimines, 98  but it is known to be more stable in other polyaromatic o-hydroxyimine compounds.99 The equilibrium between the enol-imine and the keto-enamine forms, together with the resonance contribution of a zwitterionic intermediate, explains the unusual stability of this imine, and the locked conformation (trans for the imine, cis for the enamine) of this motif. This fixed geometry has been used to selectively synthesize some Schiff-base macrocycles.100   Figure 2.2  Tautomerization between the enol-imine (left) and the keto-enamine (right) forms. The NH tautomer may be formally depicted as a zwitterionic iminium-phenolate or a neutral keto-enamine in the extreme cases. (The notation Ha and Hb will be used in the following discussion.)  66  This chapter focuses on the synthesis, purification, characterization and computational studies of shape-persistent campestarenes. By choosing triisopropylsilyl-campestarene as an example, insight into the enol-imine and keto-enamine tautomerization was provided.   2.2 Results and Discussion  2.2.1 Synthesis and Characterization Although the typical synthetic strategy of Schiff-base macrocycles, which involves a condensation reaction between a compound with diformyl groups and a diamine, turned out to be successful and effective, this method could only be used to prepare macrocycles comprised of an even number of residues. To expand the Schiff-base macrocycle field, another strategy needed to be developed. In some cases, Schiff-base macrocycles had been synthesized using a single building block bearing both the amine and the formyl groups. As the amine normally reacted with the aldehyde to yield an imine, it was necessary to protect one of these groups in order to isolate and purify the precursor. The amine could be masked with a Boc group,101 or as a nitro group,102 and in other cases the aldehyde had been masked as an acetal or aminal group.103   Scheme 2.1 Synthesis of 4c-e.  67  Here in this work, the use of precursors 2a-e bearing a nitro group was chosen. Precursors 2a-e could be synthesized as shown in Scheme 2.2. Phenols 4a-e with desired R groups could be purchased directly (4a-b) or synthesized from 4-bromo-phenol (4c-e). (Scheme 2.1)  Compounds 3a-e were synthesized through a formylation reaction selectively with MgCl2. Nitration was conducted by reacting formylation products with PTSAH2O and KNO3 in acetic acid. The resulting nitro compounds 1a-e were the macrocycle precursors.   Scheme 2.2  General synthetic route of campestarenes 1.  A reduction reaction was required to be performed before macrocyclization in order to expose the masked group for reaction. Sodium hydrosulfite was chosen as the reducing agent to reduce 68  2a-e into amines, which then cyclized to offer campestarenes 1a-e. This strategy avoided the use of protecting groups, inert atmospheres, and transition metals to reduce the nitro group. Previously reported campestarene 1a was insoluble in most solvents. Actually, its poor solubility in organic solvents facilitated its purification step. After the condensation reaction, a simple filtration was performed to collect the precipitate, followed by washing with organic solvents. The solid that remained was campestarene 1a.  Adding to the issue of the poor solubility of 1a, its tendency to dimerize caused a lot of trouble in further studies regarding tautomerization. Notably, computational studies showed that the dimer of campestarenes could take two different relative orientations: one in which the rotation around the C5 axis of the monomer was able to generate an inversion center between the two molecules and another one that can engender a mirror plane. The former belongs to the D5 and the latter to the C5h point group. The possible orientation of the keto-enamine dimer of macrocycle 1f is shown in Figure 2.3. Considering that there were two different possible face-to-face dimers and there were different possible tautomers of both the monomers and the dimers, a simpler system was sought to allow us to explore the structure and tautomeric behavior of campestarenes. Both triphenylsilyl and triisopropylsilyl (TIPS) groups seemed to be good options as they were bulkier than tert-butyl substituents and may also enhance the solubility of the macrocycle and prevent aggregation as well. 69   Figure 2.3  a) Chemical structures of campestarene 1f and 1g; b) Possible orientation of the keto-enamine dimer of 1f.  With longer or bulkier substituents anchored on phenyl rings, campestarenes did demonstrate increased solubility. Unlike 1a, campestarenes 1c-e are soluble in organic solvents. Especially 1e, synthesized with bulky triisopropylsilyl groups, presented good solubility in a wide-range of organic solvents such as benzene, chloroform and N,N-dimethylformamide. Besides, NMR experiments showed no dimerization for campestarene 1c-e. Bulky groups were believed as effective separators to prevent the macrocycles from aggregating.  70   Figure 2.4  1H NMR spectrum of macrocycle 1d in CD2Cl2 (400 MHz, room temperature).  Experiments showed that triphenylsilyl groups were effective separators to prevent aggregation as well. Although macrocycle 1d did not dimerize in CD2Cl2 (Figure 2.4), its poor solubility in polar solvents such as DMF and DMSO prevented further study of macrocycle 1d in these solvents.  Moreover, since precipitation worked as one of the driving forces for macrocyclization, the yield of macrocyclization decreased with the increase in solubility (Table 2.1). Also with improved solubility, filtration no longer worked as a purification step. Thus, column chromatography was applied to isolate campestarene 1e from the reaction mixture. As a family of Schiff-base macrocycles, campestarenes are not stable under acidic conditions. Thus, neutral alumina, as opposed to silica, which is weakly acidic due to the surface –OH groups, was used as stationary phase for purification.     71  Table 2.1  Isolated yields of campestarenes 1. Compound R Yield 1a tert-Bu 99% 1b (1,1-dimethyl)-propyl 88% 1c (1,1,3,3-tetramethyl)-butyl 70% 1d Triphenylsilyl 38% 1e Triisopropylsilyl 38%  The 1H NMR spectrum (Figure 2.5) of 1e is indicative of high symmetry with one imine (δ = 8.9 ppm) and two singlet aromatic resonances (δ = 7.5, 7.7 ppm). Interestingly, a signal for the OH groups was found at δ = 16.7 ppm, which implied that the significant downfield shift arose from strong hydrogen bonding to the nearby imine. Figure 2.6 shows the MALDI-TOF mass spectrum of 1e. The most dominant peaks correspond to [1e+H+], [1e+Na+], and [1e+K+].  Peaks corresponding to dimers and trimers plus alkali metal were also evident in the mass spectrum.   72   Figure 2.5  1H NMR spectrum of 1e in CDCl3 (400 MHz, room temperature).  Further attempts to increase the size of bulky substituent groups resulted in surprising geometry changes. Precursor 2h with tris(4-(tert-butyl)phenyl)methyl group was synthesized by following the general synthetic route. However, the macrocyclization step ended up with a much lower yield. The signal at 2649.7 in mass spectrum is consistent with theoretical data (2649.7) of campestarene [1h+H]+.(Figure 2.7(a))    73   Figure 2.6  MALDI-TOF mass spectrum of macrocycle 1e. Matrix: trans-2-[3-(4-t-butyl-phenyl)-2-methyl-2-propenylidene] malononitrile (DCTB).  Interestingly, if the reaction was conducted in THF/H2O instead of MeOH/H2O, no macrocycle 1h can be detected by mass spectrometry. Figure 2.7(b) shows the mass spectrum of product produced in THF/H2O. The major peak at m/z = 1590.1 was assigned to an unexpected 6-fold symmetric macrocycle [1h’+2H]2+, which should situate at m/z = 1590.0. Formation of macrocycle 1h’ might be caused by strong steric interaction between adjacent substituent groups. By forming the cyclic hexamer 1h’, macrocycles could reduce steric interaction by taking a chair conformation. However, both reactions with 2h only offered macrocycles at a very low yield, where most of the products were linear oligomers. Evidently, the increased steric interaction between substituent groups played a negative role in the macrocyclization. 74   Figure 2.7 (a) Chemical structure and Mass spectrum of campestarene 1h; (b) Chemical structure and Mass spectrum of campestarene 1h’.  Notably, such a macrocycle with unexpected geometry was only observed for compound 2h. Macrocyclization with 2a-e exclusively produced 5-fold symmetric macrocycles 1a-e. Due to 75  these factors, campestarene 1e was finally chosen for further studies since its substituent groups (TIPS) imparted adequate steric bulk to prevent aggregation but not too large to hinder macrocyclization.  2.2.2 Solid-State Structure of Campestarene 1e The solid-state structure of campestarene 1e was investigated by infrared (IR) spectroscopy and single-crystal X-ray diffraction (SCXRD). Frequencies reported for the C=N stretching vibration in compounds of the type X-C(-Y)=N-Z, where X, Y and Z may be hydrogen, alkyl or aryl, occur in the region 1680-1600 cm-1. Factors affecting the position of the C=N stretching absorption within this band include the physical state of the compound, the nature of the substituent groups, conjugation with either carbon or nitrogen, or both, and hydrogen bonding.104  For comparison, 15N-labelled campestarene 1e-15N5 was prepared using the same synthetic route shown in Scheme 2.2. 15N was introduced by using K15NO3 in the nitration step. The IR spectrum of 1e showed a C=N stretching mode at 1615 cm-1, whereas the peak in question was observed at 1600 cm-1 in the isotopomer 1e-15N5 while the other peaks remained at the same position. A pure C=N simple harmonic oscillator was expected to show a 1.5% decrease vibrational frequency when 14N is replaced with 15N atoms. The measured decrease of 1.0% is close to the calculated value. Both IR spectra are shown in Figure 2.8. In continuation, a computational study was conducted to calculate the theoretical vibrational frequencies of campestarene 1g at the DFT 6-311g(d,p)/M06 level of theory. The computed values were 1619 and 1601 cm-1 for the 14N5 and 15N5 isotopomers, respectively, which are in very good agreement with the experimental values and the observed isotopic shift. In addition, in the range around 1540 cm-1, where a NH deformation vibration was expected, neither 1e nor 1e-15N5 showed 76  any band. Analysis of the computed modes at the same level of theory indicated that the keto-enamine form should show a very intense absorption peak in the range 1543-1537 cm-1 irrespective of the dielectric properties in the environment. Together, the computation and experimental data analysis suggests that campestarene 1e exists in the solid state in the enol-imine form rather than in the keto-enamine form.105   Figure 2.8   (a) IR spectra of macrocycle 1e (black) and 1e-15N5 (red); (b) comparison of peaks at ~1600 cm-1 of macrocycle 1e (black) and 1e-15N5 (red).  To better understand the behavior of campestarenes in the solid state, more computational simulations were performed. Chart 2.1 illustrates the 3 types of tautomers that could be present in a 5-fold symmetric campestarene in the extreme cases (see Figures 2.5 for the nomenclature Ha and Hb): 1. Ha can be covalently bound to oxygen and H-bonded to nitrogen with a ring of 6 atoms (enol-imine form). 77  2. Ha can be covalently bound to oxygen and H-bonded to nitrogen with a ring of 5 atoms (enol-imine form). 3. Ha can be covalently bound to nitrogen and H-bonded to oxygen with a ring of 5 or 6 atoms (keto-enamine form). Hybrid (PBEPBE1), meta-hybrid (M06) DFT computations of 1f and 1g at the 6-311g(d,p)/vacuum suggest that the 1e-6r should be the most stable. In particular the 5r form of 1f is less stable by 79 kJ mol-1 at the 6-311g(d,p)/PBE1PBE1/vacuum and 83 kJ mol-1 at 6-311g(d,p)/M06/vacuum level of theories. However, all three isomeric structures tended to stay in planar conformations.  Chart 2.1  5-fold symmetric possible limiting isomeric structures of campestarene 1f.   Left: the enol-imine form where the O-H---N bonding forms 6-membered rings (1f-6r).  Middle: the enol-imine form where the O-H---N bonding forms 5-membered ring (1f-5r). Right: the keto-enamine form with N-H bonds.  78  There are only seven molecules containing the triisopropysilylphenyl motif in the Cambridge Structural Database (one of which contains the TIPS-phenyl groups coordinated to chromium tricarbonyl moieties), suggesting that these groups typically give poorly crystalline products.106 Nevertheless, after numerous attempts with various macrocycles, we were able to obtain single crystals of 1e by vapor diffusion of pentane into a dichloromethane solution of the molecule. Although the crystals we obtained were poorly-diffracting, leading to relatively low quality diffraction data, the overall structure of the macrocycle can be unambiguously determined (Figure 2.9).   Figure 2.9  Views of the molecular structure of 1e as determined by SCXRD. a) View of individual macrocycle; b) view of crystallographic unit cell, showing herringbone packing; c) side-on view, showing mean plane between inner phenylene carbon atoms.  For clarity, only the major component of the disordered TIPS group is shown, and most hydrogen atoms are omitted; in image c, the TIPS groups are also omitted. 79  Importantly, the macrocycle is very close to being perfectly planar, with a very slight bowing inwards of the phenyl rings: the 25 ring atoms show a mean deviation of 0.10 Å from a mean plane defined by the five inner phenylene atoms, and the 45 macrocycle atoms, excluding the TIPS group show a mean deviation of 0.19 Å from this mean plane. The C–H to N bond lengths range from 1.210(10)–1.253(11) Å (mean = 1.23 Å) and phenol C–O bond lengths range from 1.353(7)–1.365(7) Å (mean = 1.36 Å), entirely consistent with C–N double bonds and C–O single bonds (typical C-N double bonds are about 1.25 Å and C(sp2)-O single bonds are about 1.36 Å while typical C-N single bonds are about 1.47 Å and C-O double bonds are 1.21 Å). Again, this implies that 1e exists in the enol-imine form in the solid state, in accordance with the IR data; it was notable that this form persists even at 90 K (the temperature at which the crystal structure data were collected). The central oxygen atoms defined a regular pentagon, with 3.3 Å long sides. All these results supported our computational studies in conformation of tautomers. While the macrocycle had the potential for π stacking, as observed in the alkyl-substituted campestarenes, the bulky TIPS groups inhibited this organization in the solid state. Instead, 1e packed with a herringbone structure with negligible intermolecular interactions between the cores of the campestarenes. Calculations of the enol-imine form of 1g at the 6-311g(d,p)/M06/benzene and 6-311g(d,p)/mPW2PLYP/benzene107 level of theories indicated that the distortion from planarity observed by SCXRD was not because of packing forces, but was instead an intrinsic property of the molecule. In fact, the presence of the 6r H-bonding represented a further constraint to the overall ring-geometry of the molecule hence removing planarity. It was worth mentioning that the computed very low vibrational frequencies involved different distortion modes from planarity and 80  suggested that the cost of having longer H-bonds and planarity in the molecule is higher than removing the planarity and getting shorter, stronger H-bonds.   Figure 2.10  Bond lengths of 1f in the two limiting tautomeric forms: (a) all enol-imine form and (b) all keto-enamine form, computed at the 6-311g(d,p)/M06/benzene level of theory.  The computed geometries of 1f and 1g agree fairly well with the experimental data, though suggesting a tiny (less than one tenth of an Ångstrom) elongation of one side of the pentagon traced out by the O atoms and definitive C-N double and single C-O bond lengths. As for the bond lengths in case of the enol-imine tautomer, the carbon-carbon bonds starting from the phenolic carbon atoms within the aromatic moieties are longer than the other aromatic carbon-carbon bonds and are around 1.415 Å. Because of the asymmetry of the hydrogen bond, the two carbon-carbon bonds are not the same and the C-C(OH) bond “inside” the 6-membered ring involved in hydrogen bonding is slightly longer: 1.418 Å. This pattern is preserved in the case of the keto-enamine 81  tautomer (Figure 2.10 for computed geometrical parameters). The lower stability of keto-enamine form might be due to the substantially weaker hydrogen bond in this tautomer, which was computed as 1.743 Å for the keto-enamine tautomer vs. 1.586 Å in the enol-imine tautomer.  2.2.3 Solvent-Dependent Tautomerization With the goal of investigating the tautomeric behavior of campestarenes in solvents, first penta(4-(triisopropylsilyl))campestarene 1e was dissolved in various deuterated organic solvents for NMR experiments. Figure 2.11 shows its 1H NMR spectra in those deuterated solvents, which varied form non-polar solvents such as benzene-d6 and toluene-d8 to polar solvents including DMF-d7 and DMSO-d6.   Figure 2.11  1H NMR spectra (room temperature, 400 MHz) of campestarene 1e in (a) DMSO-d6, (b) DMF-d7, (c) CD2Cl2, (d) CDCl3, (e) toluene-d8, (f) benzene-d6. (* indicates a residual solvent peak due to DMF-d7).  82  In all of those solvents, only a single set of resonances was observed, consistent with the campestarene’s 5-fold symmetry in solution. Notably, in polar solvents (CDCl3, CD2Cl2 and DMF-d7), the signal for Ha around 17 ppm always split into a doublet signal due to J-coupling with Hb.  Due to the low solubility of 1e in DMSO, the 1H NMR signals had low intensity and the J-coupling between Ha and Hb could not be resolved. This 3JHH value varied from ~7 Hz in DMF-d7 to ~2 Hz in CDCl3. (To remove any ambiguity about which J-coupling is being discussed, this 3JHH will be referred to as 3JHCNH hereafter.) This 3JHCNH should not be observed in the extreme enol-imine tautomer. The observation of 3JHCNH coupling suggested the existence of keto-enamine tautomer in those solvents.  In literature,108 a keto-enamine form of a salicylaldimine compound showed a ~13 Hz 3JHCNH coupling constant. The 1H NMR spectrum of 1e in DMF-d7 showed only one doublet for Ha and one doublet for Hb, which corresponds to the signal of the monomer of the keto-enamine tautomer. Weak J-coupling between Ha and Hb was observed in the 1H NMR spectra of 1e in CD2Cl2 (3JHCNH = 4 Hz) and CDCl3 (3JHCNH = 2 Hz). However, in non-polar solvents such as C6D6 and toluene-d8, no coupling between Ha and Hb was observed. The peaks around 17 ppm and 9 ppm were both singlets in both solvents, indicating that the proton Ha resides mostly on the phenolic oxygen, which in turn corresponds to the enol-imine form.  Not unique to 1H NMR spectra, 13C{1H} NMR shifts for salicylaldimines are known to be sensitive to the tautomeric form.109  Figure 2.12 shows the 13C{1H} NMR spectra of campestarene 1e in C6D6, toluene-d8, and CD2Cl2. 83   Figure 2.12  13C{1H} NMR data of 1e in C6D6, toluene-d8, and CD2Cl2.  However, the 13C{1H} NMR data in the other solvents studied could not be obtained owing to poor solubility of 1e in said solvents. Instead, 13C{1H} NMR data for 1e was measured indirectly by HMBC spectroscopy in DMF-d7 and CDCl3 (Figure 2.13). 84   Figure 2.13  1H-13C HMBC data for 1e in CDCl3 and DMF-d7.  In Table 2.2, the relative permittivity and the C-O chemical shift collected in each solvent is shown. The relative permittivity of a solvent is a relative measure of its chemical polarity. For 85  example, a polar solvent like water has a relative permittivity of 80.10 at 20 °C, while a non-polar solvent such as n-hexane has a relative permittivity of 1.89 at 20 °C. This property is also known as dielectric constant.   Table 2.2  C-O chemical shift collected in various solvents. Solvent Relative permittivity (20 °C) C-O chemical shift/δ DMSO-d6 46.70 n.a. DMF-d7 36.70 166 dichloromethane-d2 8.93 163 chloroform-d 4.81 162 toluene-d8 2.38 161 benzene-d6 2.27 161  In benzene-d6 and toluene-d8, a signal belonging to CO had a chemical shift of 161 ppm, while in CD2Cl2, the signal belonging to CO shifted to 163 ppm. In solvents of even higher dielectric constant like DMF-d7, the peak further shifted downfield to 166 ppm. The CO resonance of a typical enol-imine form of salicylaldimine is observed at ~155 ppm, while in the keto-enamine form it could shift downfield to ~180 ppm. The observed trend for macrocycle 1e showed that there was a strong solvent dependence on the enol-imine/keto-enamine equilibrium, and that in solvents with higher dielectric constant, the keto-enamine form was favored. Based on the data presented so far, the principal tautomer of 1e present in solution at room temperature strongly depends on the solvent. From the 13C and 1H NMR data, the macrocycle appears to be primarily in the keto-enamine form in high dielectric solvents such as DMSO and DMF, and in the enol-imine form in low dielectric solvents, such as benzene and toluene.  In order to further confirm the position of the hydrogen Ha in campestarene 1e, 15N-labelled penta(4-(triisopropylsilyl))campestarene 1e-15N5 was used for 1H NMR experiments. Figure 2.14 86  shows the 1H NMR spectrum in different solvents of the 15N-labelled analog of campestarene 1e, 1e-15N5.   Figure 2.14  1H NMR spectra (room temperature, 400 MHz) of campestarene-15N5 (1e-15N5)  in (a) DMSO-d6, (b) DMF-d7, (c)  CD2Cl2, (d) CDCl3, (e) toluene-d8, (f) benzene-d6. (* indicates a residual solvent peak due to DMF-d7).  Previous literature indicated that the keto-enamine form of a salicylaldimine compound would show a 1JNH coupling constant of about 89 Hz,110 whereas the purely enol-imine form, where the H atom is not covalently bonded to the N atom, shows no 1JNH coupling.  As shown in Figure 2.14 (a), the signal around 17 ppm split into a doublet due to the new 1JNH coupling between Ha and 15N. This observed coupling constant of about 72 Hz indicates a large amount of campestarene 1e-15N5 exists in the keto-enamine form. When the 1H NMR spectrum of 1e-15N5 was measured in DMF-d7, the signal around 17 ppm (doublet for 1e, Figure 2.11) was split into a doublet of doublets due to J-coupling with nitrogen-15 (1JNH = 48 Hz) and J-coupling with Hb (3JHCNH = 7 Hz). Also, the signal for Hb at ~9.5 ppm was split into a doublet of doublets 87  (3JHCNH = 7 Hz; 2JNH = 2 Hz), showing coupling to the 15N nucleus and Hb. This result clearly indicates the presence of a chemical bond between the nitrogen and Ha, which is stronger than a simple hydrogen bond: the campestarene is thus adopting a structure dominated by the keto-enamine isomer rather than the enol-imine isomer. With the decrease of solvent relative permittivity, 1JNH decreased from 1JNH = 72 Hz in DMSO-d6 to 1JNH = 11 Hz in CDCl3. When non-polar solvents were used, only a very small coupling constant (~2 Hz) to 15N can be observed. This small coupling in benzene and toluene indicates that the proton Ha resides mostly on the phenolic oxygen, corresponding to the enol-imine form, but still retains some keto-enamine character at the same time. The identity of the tautomeric form of 1e in solution is clearly very complicated. Besides this 1JNH coupling between hydrogen Ha and 15N atom, another new coupling, 2JNH coupling between hydrogen Hb and nitrogen 15N, might emerge. However, this 2JNH coupling did not change much with solvent permittivity. It only varied from ~2 Hz in DMF-d7 to ~3 Hz in benzene-d6. The assignment of the tautomers was assisted by 2D NMR spectroscopy. 1H-15N-HMQC spectra were recorded to confirm the coupling between the nitrogen and the different protons in each solvent (except DMSO-d6, where solubility hindered data acquisition); a section of a representative spectrum in CD2Cl2 is shown in Figure 2.15. The very simple coupling data indicates that the campestarene has high symmetry. In benzene-d6 and toluene-d8, 1e-15N has a 15N chemical shift of 277 and 276 ppm, respectively. With the increase of the solvent dielectric constant, the 15N signal shifted upfield. The 15N chemical shift of macrocycle 1e is 263, 245 and 213 ppm in CDCl3, CD2Cl2, and DMF-88  d7, respectively. This dramatic upfield shift clearly suggests different tautomers of campestarene 1e-15N in different organic solvents.  Figure 2.15  Section of the 1H-15N HMQC NMR spectrum of campestarene-15N5 (1e-15N5) in CD2Cl2.  In addition, the 1H-15N HSQC method was used to record the coupling between directly connected 1H and 15N atoms in order to obtain direct evidence of the Ha location change in different solvents. In low dielectric constant solvents like benzene-d6, toluene-d8 and CDCl3, no signal could be obtained at room temperature. Under the same experimental conditions, however, coupling between Ha and 15N was easily observed in solvents with higher dielectric constant such as CD2Cl2 and DMF-d7. This result suggests that most of macrocycle 1e is in the enol-imine form in benzene-89  d6, toluene-d8 and CDCl3, while the keto-enamine form plays a more important role in CD2Cl2 and DMF-d7 at room temperature.  Table 2.3  Chemical shifts and 1JNH, 2JNH, 3JHCNH coupling constants for 1e-15N5 in DMSO-d6, DMF-d7, CD2Cl2, CDCl3, toluene-d8, and benzene-d6. Solvent Relative permittivity OH or NH CH 15N HSQC {1H,15N} Observed? Proposed predominant species DMSO-d6 46.70 17.07 ppm 1JNH = 71.6 Hz 3JHCNH = NAa 9.42 ppm 2JNH = NAa 3JHCNH = NAa NAa YES Keto-enamine DMF-d7 36.70 17.43 ppm 1JNH = 47.6 Hz 3JHCNH = 7.3 Hz 9.73 ppm 2JNH = 1.9 Hz 3JHCNH = 7.3 Hz 213 ppm YES Keto-enamine CD2Cl2 8.93 17.19 ppm 1JNH = 23.9 Hz 3JHCNH = 3.8 Hz 8.93 ppm 2JNH = 2.7 Hz 3JHCNH = 3.8 Hz 245 ppm YES Enol-imine & Keto-enamine CDCl3 4.81 16.77 ppm 1JNH = 11.4 Hz 3JHCNH = 2.2 Hz 8.93 ppm 2JNH = 2.5 Hz 3JHCNH = 2.2 Hz 263 ppm NO Enol-imine  & Keto-enamine Toluene-d8 2.38 16.70 ppm 1JNH = 2.7 Hzb 3JHCNH = 0 Hzc 8.92 ppm 2JNH = 3.4 Hzb 3JHCNH = 0 Hzc 276 ppm NO Enol-imine Benzene-d6 2.27 16.98 ppm 1JNH = 2.2 Hzb 3JHCNH = 0 Hzc 8.88 ppm 2JNH = 3.1 Hzb 3JHCNH = 0 Hzc 277 ppm NO Enol-imine a It is difficult to get accurate data due to the poor solubility of the macrocycle in DMSO-d6. b In the proposed structure, the H is not directly connected to the 15N. 1JNH is still used for consistency since there is a dynamic equilibrium. This data is approximate since the coupling is not very obvious. c No coupling can be observed in the 1H NMR spectrum of the 14N macrocycle.  90  The data for campestarene 1e (13C NMR shifts, 15N NMR shifts, J-coupling) are consistent with the keto-enamine as the dominant form in DMSO-d6 and DMF-d7, the enol-imine form dominant in benzene-d6 and toluene-d8, and an intermediate form in CD2Cl2 and CDCl3. Table 2.3 summarizes the NMR data for campestarene 1e in different solvents.  Figure 2.16 shows calculated charge distributions of the enol-imine and the keto-enamine tautomer of campestarene 1f. Compared to the enol-imine form, in each repeating unit, the keto-enamine tautomer has a larger dipole moment. To stabilize the keto-enamine tautomers, solvent with larger polarity was favored.  Figure 2.16  Calculated charge distributions of the enol-imine and the keto-enamine tautomer of campestarene 1f. Atoms are colored according to the values of their net atomic charges computed according to the Merz-Kollman procedure, fitting the electrostatic potential computed at the 6-311+g(2d,2p)/PBE1PBE//Vacuum/6-311g(d,p)/PBE1PBE/Vacuum.  91  As a measurement of solvent polarity, solvent dielectric constant is well-known to affect the enol-imine / keto-enamine equilibrium of Schiff-base compounds, with the equilibrium shifting toward the more polar NH form (keto-enamine) in solvents with higher dielectric constant.111 This has been attributed to both the increased dipole moment of the NH form as well as contributions from local solvent molecules.112  On the other hand, it is worth noting that the most stable tautomer is not the one with the largest molecular dipole moment, but the one which has the largest number of shifted hydrogen atoms, even though for symmetry reasons it lacks any significant dipole moment. Local dipole moment, rather than macrocycle dipole moment, apparently plays a crucial role in affecting the enol-imine/keto-enamine equilibrium. The keto-enamine macrocycle is almost neutral in the exterior part of the molecule (mainly CH aromatic and aliphatic moieties) with a “core” characterized by a strong alternating charge distribution. Oxygen and nitrogen atoms bear negative charges, and Ha has a strong positive character. Some carbon atoms of the "phenylene" rings also exhibit positive or negative charges. On the other hand, the enol-imine tautomer does not exhibit such a charge distribution. The oxygen and Ha have smaller charges, and the aryl carbons and protons are almost neutral. The enhanced stability of the keto-enamine in more polar solvents can be traced back to the charge distribution of the molecule, where the larger local charges facilitate more intense interaction with the solvent (keto-enamine form), whereas the enol-imine form shows smaller charge separation, which is better suited for less polar solvents.  92  2.2.4 Photophysical Behavior Campestarenes are intensely colored and they show strong solvatochromism. 113  When dissolved in polar organic solvents including DMF and DMSO, campestarene 1e yields a solution very dark purple in color. The color turns to red-brown when the solvent is switched to DCM, and futher into yellow-brown when 1e is dissolved in benzene. Figure 2.17 shows a photograph of solutions of campestarene 1e in various solvents, and the corresponding UV-vis spectra. Notably, in most solvents (apart from DMSO), there are always two peaks in the UV-vis spectra, at ~400 and ~550 nm. The ratio of these two peaks vary from solvent to solvent. The peak at ~400 nm decreases in high dielectric constant solvents, whereas the peak at ~550 nm increases dramatically in DMF and DMSO. Combined with previous NMR data, the peak at ~400 nm can be assigned to enol-imine form while the peak at 550 nm is predominantly from the keto-enamine tautomer.  Figure 2.17  (a) UV-Vis spectra of macrocycle 1e in toluene, benzene, CHCl3, CH2Cl2, DMF and DMSO (this is the order from bottom to top at 550 nm); (b) photographs of the same solutions (all 10-5 mol L-1).  93  To get a better understanding of the optical spectra of the campestarenes, the electronic transitions were calculated with time-dependent DFT (TD-DFT) calculations. The first 40 electronic transitions of the keto-enamine and enol-imine isomers of hypothetical campestarenes 1f and 1g were computed in benzene and DMSO (Figure 2.18). Although the keto-enamine computed transitions appear slightly hyperchromic, the first clear observation is that the enol-imine tautomer does not absorb at energies below 22000 cm-1 (455 nm) and that the most intense absorption for the keto-enamine tautomer is below 20000 cm-1 (500 nm).  Figure 2.18  Comparison between the experimental spectra of 1e in benzene and DMSO (brown and green lines) with the TD-DFT calculations in the 1g keto-enamine and enol-imine tautomers (green and brown vertical arrows).  Addition of larger substituents to the aromatic rings of hypothetical campestarene 1f-g should result in small shifts in the transitions (see Figure 2.19 for a comparison between 1f and 1g).  This 94  supports the idea that T and L bands are due to transitions of the keto-enamine tautomer, whereas the other tautomer does not show any transition in this range of energies. On the other hand, band H, and to a minor extent band M, owes its intensity mainly to the enol-imine tautomer, with the keto-enamine tautomer showing transitions with less relevant computed intensity.  Figure 2.19  Effect of the peripheral substitution on the most intense transitions in case of 1f (red) 1g (green) in the two limiting tautomeric forms (a) keto-enamine and (b) enol-imine. TD-DFT calculations at the 6-311g(d,p)/M06/DMF. The lines indicate the calculated spectra while the curves are the experimental data.  Overall, this analysis supports the predominance of the keto-enamine tautomer in DMSO, and the predominance of the enol-imine form in benzene and toluene, in both cases with a small amount of the other tautomer noticeable in the UV-Vis spectrum. TD-DFT computations of the spectra of the keto-enamine and the enol-imine tautomers in benzene and DMSO, respectively, do not show significant differences in the intensities and frequencies of the transitions for each tautomer (Figure 2.20). This indicated that the solvent does not modify the pattern and composition of the excited states, which is one of the main mechanisms of solvatochromism, but has the effect of shifting the equilibrium among the different possible tautomers that have different absorption spectra. 95   Figure 2.20  Solvent effect on the TD-DFT 6-311g(d,p)/M06 computed spectra of (a) keto-enamine and (b) enol-imine of 1g in benzene, DMSO and DMF.  2.2.5 Two or More Tautomers? In the solid state, the experimental data suggest the presence of only the enol-imine form, whereas in solution it is in equilibrium with its keto-enamine tautomer; the equilibrium depends on the solvent used (shown in Table 2.3). Up to now, the discussion regarding the macrocycle was as though it were the enol-imine form, the keto-enamine form, or an equilibrium of these two forms. However, apart from those two extreme tautomers, there are a substantial number of intermediate isomers that are possible. This possibility is now discussed. If one assumes that the same nitrogen atom cannot be involved at the same time within two hydrogen bonds, and that every OH must be involved in at least one hydrogen bond, fifteen isomers are possible for hypothetical campestarene 1f. All the enol-imine tautomers with OH---N 6r and 5r and all the keto-enamine tautomer are shown in Chart 2.1. Starting from the enol-imine tautomer, six other isomers can be generated by isomerizing one, two, three or four imine bonds to the keto-enamine form. The isomers with OH---N bond forming only 6-membered rings are shown in 96  Chart 2.2. To differentiate among the different tautomers, the presence of an amine form is identified by 1 and the imine form by 0 as shown in the same Chart.   Chart 2.2  Isomers of campestarene with hydrogen bonds forming 6-membered rings. The numbers below each macrocycle show the relative arrangements of NH (1) and OH (0) tautomeric forms around the ring.  An additional seven isomers with 5-membered hydrogen-bonding rings can be generated in the same way. However, as shown above, the computed DFT energy of the full enol-imine 5r is much higher than that of 6r. Thus, it is reasonable to focus the theoretical study on only the structures with 6-membered hydrogen-bonding rings as typically observed in salicylaldimine structures.  It is worth noting that the enol-imine form of campestarene 1f in vacuo is more stable than the keto-enamine form by 17 kJ mol-1. Taking into account the influence of the solvent without 97  altering the structure obtained by the calculations in the gas phase, the relative energies obtained for the enol-imine and the keto-enamine tautomers are qualitatively in good agreement with the experimental results: the keto-enamine is more stable in DMSO and less stable in benzene, indicating that the solvent stabilization does not depend on the relaxation of the geometry in a given solvent but is related to the interaction between the molecule and the solvent reaction field. Relaxing each geometry in the appropriate solvent appears to enhance the effect of solvent polarity on the relative stability of the whole set of tautomers of 1f as reported in Figure 2.21 independent of the level of theory used.  Figure 2.21  Relative stability of different tautomers of 1f in benzene (red) and DMSO (green). The energy reference is related to the most stable isomer in the same solvent. Full geometry optimization at the 6-311g(d,p)/ M06/solvent level of theory. 98  To probe the effect of peripheral substitution on the stability of different tautomers of campestarene 1e, further computational studies on campestarene 1g were performed. Compared to 1f, campestarene 1g should give a chance to get a closer model for 1e. No significant differences could be identified on the general trend both in terms of the solvent effect changing from benzene to DMSO and DMF and within the different tautomers compared to 1f. In the case of 1g, there is a small destabilization in molecular energies of all the tautomers compared to the 11111 one. Vibrational contributions, on the other hand, stabilized the whole set of structures and, to a larger extent, the fully (00000)-enol-imine and (11000) tautomers (Figure 2.22).  Figure 2.22  Molecular energies (ΔE, green) and internal Gibbs free energies (ΔG(298), black) for the eight tautomers computed at the 6-311g(d,p)/ M06 /DMF level of theory for 1g.  99  Close results were found also in the case of 1f free energies in DMF though inclusion of correlation effects by means of the double hybrid mPW2PLY xc-functional, which stabilized the full (00000)-enol-imine form compared to the intermediate ones still leaving the full (11111)-keto-enamine form as the most stable by 15.7 kJ mol-1.  Figure 2.23  Molecular Energies (E, green) and Gibbs free energies (G(298), red) for eight tautomers of  1f and their transition state for sequential interconversion computed at the 6-311g(d,p)/mPW2PLYP/DMF//6-311g(d,p)/M06/DMF level of theory (dashed lines are added just to allow easy following of the series of values; they do not imply linear changes.).  Vibrational corrections still afford the same pattern in the relative stability, but introduced a further stabilization that "smoothed" the trend and further reduced the molecular and Gibbs free 100  energies of the whole set of tautomers and transition states. The keto-enamine remained as the most stable tautomer. To probe the rate of conversion among the different tautomers, some studies on the energies of the transition state (TS) were conducted. As shown in Figure 2.23, the relative energy of each transition state is low, indicating that the interconversion among the different tautomers should be very fast even without the presence of specific interaction with the solvent molecules (but only due to self-consistent reaction field (SCRF) polarization). The enol-imine molecule should be less stable by 6.7 kJ mol-1 and the (11111)-keto-enamine is the most stable and most abundant in polar solvent, though small amounts of other tautomers can also be present. Because these calculations and characterization results suggest a more complex system of equilibria among the different tautomers, temperature-dependent studies were undertaken by both NMR and UV-vis spectroscopies.  2.2.6 Variable Temperature Studies Before performing variable temperature experiments of campestarene 1e in various solvents, first the appropriate temperature range for the following studies was determined. Toluene-d8 was chosen for this test due to toluene’s wide temperature range as a liquid. A solution of campestarene 1e in toluene-d8 was cooled to -80 °C. 1H NMR spectra were collected every 30 min from -80 °C to +90 °C. Variable temperature 1H NMR spectra of 1e in toluene-d8 are presented in Figure 2.24. At low temperature, no obvious new peaks show up in the spectra. However, when the temperature rose above 50 °C, a large portion of macrocycles started decomposing rapidly, resulting in the growth of new impurity peaks. This indicated that the decomposition of campestarene was accelerated at relatively high temperature. Based on that, variable temperature measurements were 101  only conducted at lower temperature (below 50 °C) to avoid decomposition. In addition, considering both benzene and DMSO have high freezing points, variable temperature studies in toluene, dichloromethane, chloroform and dimethylformamide were also conducted.  Figure 2.24  Variable-temperature 1H NMR spectra of 1e in  toluene-d8, only the resonances near 17 (Ha) and 9 (Hb) ppm are shown.  Figure 2.25 shows the results of variable-temperature 1H NMR studies of 1e in toluene-d8, CDCl3, CD2Cl2 and DMF-d7.114 In CDCl3 at room temperature, the 1H NMR spectrum of 1e appears mostly consistent with the enol-imine form. That is, only a very small J-coupling (about 2 Hz) is observed between the OH and the CH (imine) protons. As the temperature is lowered, however, there is a substantial downfield shift of the OH resonance and an upfield shift of the imine CH resonance, which shows obvious splitting and J-coupling at low temperature (< ~273 K). Thus, as the temperature is lowered, there is a gradual shift in the keto-enamine / enol-imine equilibrium toward the keto-enamine form. This is consistent with previous studies of 102  tautomerization of Schiff-bases, which always show an increase in the amount of the keto-enamine (NH) form at low temperature.115   Figure 2.25  Variable-temperature 1H NMR spectra of 1e in (a) CDCl3, (b) toluene-d8, (c) CD2Cl2 and (d) DMF-d7. Only the resonances near 17 (Ha) and 9 (Hb) ppm are shown.  In toluene-d8, the OH and imine CH protons of 1e show similar shifts at low temperature as observed in CDCl3, but the NH-CH coupling is not apparent even at 193 K. Although the low temperature shifts the equilibrium toward the keto-enamine form, it was not observed in this non-polar solvent. This absence of 3JHCNH coupling over the entire temperature range suggests that even at low temperature, the enol-imine form is still dominant in toluene-d8. In CD2Cl2 and DMF-d7, the chemical shifts of the peaks show temperature dependence as well. Moreover, at room temperature, an obvious 3JHCNH coupling could be observed in both solvents, indicating a large population of the keto-enamine form existing at room temperature. With the decrease of temperature, the population of the keto-enamine tautomer increases, resulting in an increase of the observed coupling constant. This result is consistent with an increase in the 103  population of the keto-enamine form upon reduction of temperature. Compared to macrocycles in CD2Cl2, the sample in DMF-d7 shows a larger 3JHCNH coupling at room temperature. The temperature-dependent equilibrium was also probed by examining the NMR spectra of 1e-15N5 as a function of temperature in different solvents.  As expected, in CD2Cl2 and DMF-d7 (Figure 2.26), both 1JNH and 3JHCNH coupling constants increase as the temperature decreases, indicating the keto-enamine form is more favored at lower temperature.   Figure 2.26  Variable-temperature 1H NMR spectra of 1e-15N5 in (a) CD2Cl2 and (b) DMF-d7.   104  Besides the variable temperature 1H NMR spectra, the chemical shifts of the 15N atoms in 1e-15N5 is another way to show the tautomeric equilibrium. However, 15N NMR usually takes a relatively long time. Hence, performing variable-temperature 15N NMR of 1e-15N5 was practically difficult due to macrocycle decomposition. As a result, to investigate the chemical shifts of the 15N atoms in 1e-15N5, both variable-temperature 1H-15N HMQC NMR spectroscopy and variable-temperature 1H-15N HSQC NMR spectroscopy were carried out. Both kinds of variable-temperature 2D NMR spectra showed a dramatic 15N chemical shift change from 240 ppm at 267 K to 190 ppm at 193 K (shown in Figure 2.27 and Figure 2.28). This change of chemical shift resulted from the tautomerization between the enol-imine form and keto-enamine form. Noticing that there was always only one signal in the 1H-15N HSQC NMR spectra over the entire temperature range, it was believed that there is always a significant amount of campestarenes existing in keto-enamine form to give rise to this signal of protons directly connected to 15N. Although at relatively high temperature there were less keto-enamine forms in CD2Cl2 and DMF-d7 compared to those at low temperature, this population of keto-enamine tautomers was not negligible even at relatively high temperatures.  105   Figure 2.27  (a) Variable-temperature 1H-15N HMQC NMR spectra of 1e-15N5 in CD2Cl2, the projection is at 267 K; (b) Variable-temperature 1H-15N HSQC NMR spectra of 1e-15N5 in CD2Cl2, the projection is at 267 K. 106    Figure 2.28  (a) Variable-temperature 1H-15N HMQC NMR spectra of 1e-15N5 in DMF-d7, the projection is at 293 K; (b) Variable-temperature 1H-15N HSQC NMR spectra of 1e-15N5 in DMF-d7, the projection is at 293 K.  107  If we assume that: (1) the coupling constants 1JNH, 2JNCH and 3JHCNH of the different tautomers is mainly sensitive to the structure of the two flanking OH or O substitution on the "phenylene" rings; and (2) the observed value for a given tautomer is the average of the values of each molecular site in case of fast hydrogen atom exchange among the different sites as suggested from the values of the computed transition state (TS) energies, then, it can be easily checked that the possible combinations of substitution on the two nearest neighbor "phenylene" rings taking into account the circular topology of the molecule are reduced to the values reported in Table 2.4.  Table 2.4  Possible values of coupling constants 1JNH, 2JNCH and 3JHCNH for different local patterns of tautomerization. Nearest neighbour 1JNH 2JNCH 3JHCNH O|NH|OH A2 B2 G2 OH|N|OH A1 = 0 B1 G1 = 0 OH|N|O A3 B3 G3 O|NH|O A4 B4 G4  TD-DFT computational studies were applied to calculate the values of the 1JNH, 2JNCH and 3JHCNH coupling constants in different 15N tautomers. The computed values of the 1JNH, 2JNCH and 3JHCNH coupling constants in the case of the tautomers (11111), (11100) and (11010)116 (Figure 2.29) indicate that the assumption was sound.   108    Figure 2.29  TD-DFT computed values of the 1JNH, 2JNCH and 3JHCNH coupling constants in case of (a) 11010 15N tautomer, (b) 11111 15N tautomer and (c) 11100 15N tautomer at the GIAO 6-311g(d,p)/cc-QZVP/M06/DCM level of theory. Red numbers, short and long arcs identify 1JNH, 2JNCH and 3JHCNH respectively.  For the whole set of the 8 tautomers, the temperature independent,117 site-averaged isotropic Jij value discussed above can be written in matrix form as:  The set of parameters can be further reduced assuming  A4  A2 ,  B4  B2 ,   B3  B1 and  G4  G2 as suggested by the TD-DFT computed values and A1 = A3 = 0.0 ; G1 = G3 = 0.0  owing to the lack of hydrogen atoms on the N atoms. This reduced to 109   The  < 𝑋𝑐𝑎𝑙𝑐 > (𝑇) value to be compared with the observed one could be computed for the ensemble by averaging over all the tautomers at each temperature according to the equation:  This model and computational results suggest that, aside from strictly experimental difficulties, it is not possible to distinguish between the following two pairs of tautomers: (11000, 10100) and (11100, 11010). With the goal of getting relative abundance of each tautomer in DMF-d7, the equation below was then used to fit the experimental data by nonlinear least squares118: 𝐽𝑚𝑒𝑎𝑠𝑢𝑟𝑒 =𝐽𝑒𝑛𝑜𝑙 + 𝐽𝑘𝑒𝑡𝑜 ∙ 𝑒𝑆𝑅 −𝐻𝑅𝑇1 + 𝑒𝑆𝑅 −𝐻𝑅𝑇 Among all three different couplings, 1JNH had the largest value while 2JNCH had the smallest value. To reduce the relative error caused by data measurement, 1JNH was prioritized for data fitting. Moreover, since the 3JHCNH coupling constants collected from DMF-d7 were relatively large, one set of these data was also used to fit into the equation above. With the data collected from variable-temperature experiments of 1e-15N5 and assuming only the two limiting tautomers and as reference the energy of the full keto-enamine form, a fit of the 1H NMR experimental data in DMF-d7 was obtained (Figure 2.30).  110  The fitted absolute values of 1JNH = 82.80 ± 0.77  Hz and 3JHCNH = 12.64 ± 0.17 Hz  are consistent with reference data of 82 and 12.5 Hz, respectively (note that owing to the negative gyromagnetic ratio of 15N, all of the coupling constants are actually negative). The associated thermodynamic parameters for 1e are ΔH = -13.5 ± 4.9 kJ/mol and ΔS = -42.7 ± 1.5 J/mol.  Figure 2.30  (a) Non-linear curve fitting based on the measured 1JNH in variable-temperature 1H NMR spectra of 1e-15N5 in DMF-d7 (red dashed and dot; experimental blue square) and 3JHCNH values (black dashed and dot; experimental green circles) assuming only the keto-enamine and enol-imine forms. R2=0.9999; b) non-linear curve fitting based on the measured 1JNH values in variable-temperature 1H NMR spectra of 1e-15N5 in CD2Cl2, R2=0.999.  The same set of parameters and adding those of the (11110) tautomer in the model gave comparable fitting of the experimental data. Although the parameters showed larger errors (ΔH(00000) 33 ± 13 kJ/mol, ΔS(00000)  -0.12 ± 0.05 kJ/mol , ΔH(11110) 22.8 ± 5.9 kJ/mol, ΔS(11110)  -0.89 ± 0.29 kJ/mol  with absolute values of 1JNH  78.19 ± 0.75 Hz and 3JHCNH  11.93 ± 0.13 Hz),  in any case the model suggests that more than two tautomers might be present even in DMF. 111  Similar non-linear curve fitting led to thermodynamic parameters in CD2Cl2 with ΔH = -8.0 ± 0.2 kJ/mol and ΔS = -37.5 ± 0.6 J/mol (calculation using 1JNH data from NMR). Unfortunately, the measured 3JHCNH of campestarene 1e-15N5 in CD2Cl2 was so small that fitting with such data would amplify the relative error. Moreover, this calculation could not be applied to CDCl3 or toluene-d8 since the measured coupling constants in those solvents are compatibly smaller, leading to a much bigger error. The model above was appropriate when high quality data was available, but as a first approximation, the system could be characterized according to the percentage of the N/O pair independent of the particular quantities of the various possible tautomeric forms (e.g., (11111), (11110), (11100), etc.) of the campestarene that were present and where the N/O pair was located. That is, if all campestarenes’ repeating units were in their keto-enamine forms, the measured 1JNH coupling should equal to 82.80 Hz, while 1JNH = 0 Hz when only enol-imine repeating units exist.  The percentage of keto-enamine repeating units can be calculated based on the following equation: 𝐽𝑚𝑒𝑎𝑠𝑢𝑟𝑒 = 𝐽𝑒𝑛𝑜𝑙 ∙[𝑒𝑛𝑜𝑙][𝑒𝑛𝑜𝑙] + [𝑘𝑒𝑡𝑜]+ 𝐽𝑘𝑒𝑡𝑜 ∙[𝑘𝑒𝑡𝑜][𝑒𝑛𝑜𝑙] + [𝑘𝑒𝑡𝑜] Since 𝐽𝑒𝑛𝑜𝑙 should have a value of 0 Hz, the equation can be simplified into: 𝐽𝑚𝑒𝑎𝑠𝑢𝑟𝑒 = 𝐽𝑘𝑒𝑡𝑜 ∙[𝑘𝑒𝑡𝑜][𝑒𝑛𝑜𝑙] + [𝑘𝑒𝑡𝑜] For instance, in DMSO-d6, at room temperature the measured 1JNH equaled to 71.6 Hz, the percentage of N/O pairs in the keto-enamine form is roughly 71.6/82.8 = 86%. This calculation yielded the percentages of keto-enamine form irrespective to the molecule in which they were located (Table 2.5). For example, a value of 86% indicats that at any instant in solution, approximately 86% of the N atoms are bonded to H atoms, while 14% are not. 112  Considering the results obtained from the NMR study and computational suggestions, it seemed reasonable to attribute the absorbance around 24500 cm-1 (400 nm) mainly to the (00000)-enol-imine tautomer, and the absorbance around 18000 cm-1 (550 nm) mainly to the (11111)-keto-enamine tautomer.  Table 2.5  Percentage of keto-enamine character of N/O pairs in campestarene-15N5 (1e-15N5) in DMSO-d6, DMF-d7, CD2Cl2, CDCl3, toluene-d8, and benzene-d6 at room temperature. Solvent Relative permittivity Percentage of Keto-enamine at room temperature DMSO-d6 46.70 86% DMF-d7 36.70 57% CD2Cl2 8.93 29% CDCl3 4.81 14% Toluene-d8 2.38 3.3% Benzene-d6 2.27 2.7%  I performed a variable temperature UV-vis study of 1e in toluene, CHCl3, CH2Cl2 and DMF.  All variable temperature UV-vis spectra are shown in Figure 2.31. In toluene (Figure 2.31 (a)), when the temperature had changed by 110 K, both peaks at 400 nm and 550 nm showed very small intensity changes. This minor change of the spectrum of 1e in toluene between -90 °C and room temperature is consistent with a stable enol-imine form at all temperatures. 113    Figure 2.31  Variable-temperature UV-vis spectroscopy of 1e in (a) toluene (183 - 293 K); (b) chloroform (213 - 323 K); (c) dichloromethane (223 - 293 K) and (d) DMF (213 - 298 K). The arrows indicate the direction of change in the spectra with decreasing T.  On the other hand, the spectra of 1e in CHCl3 and CH2Cl2 show a dramatic change over this temperature range. Previously mentioned 1H NMR spectroscopy had shown that in both CHCl3 and CH2Cl2, a mixture of the keto-enamine and enol-imine forms are present at room temperature. 114  A dramatic change of UV-vis spectra over a range of temperatures implies a dramatic change in the ratio of two tautomers. The peak at 550 nm increased at lower temperature at the expense of the peak near 400 nm. This suggests that the ratio of keto-enamine to enol-imine form increases at lower temperature; this is known to be the case for other salicylaldimines, where the keto-enamine form is most prevalent at low temperatures. In DMF, where the keto-enamine form was more favored than enol-imine form, similar trends are observed as the temperature varied.  The analysis by Gaussian functions decomposition (Figure 2.32) of the set of the spectra of 1e in DMF for the whole set of temperatures allowed a more quantitative identification of the changes. These were compared with TD-DFT absorption spectra calculations for the whole set of tautomers with the model 1g whose structure is similar to 1e.   Figure 2.32  UV-vis spectra of campestarene 1e and intensities of Gaussian functions used for the decomposition. The same set is used along the whole set of temperatures. Plot and triangles: 213 K (green) and 298 K (red). Inset: Trend of the most relevant Gaussian intensities as function of temperature.  115  The main feature going from 213 to 297 K is the almost general lower contribution (green triangles to red triangles) of all the Gaussian intensities in the range 14500-24500 cm-1 (690-400 nm) and the increase of their contributions in the range 24600-27000 cm-1 (410-370 nm). In the higher energy region above 27000 cm-1 (below 370 nm), the reduction of the intensity was more related to a widening than to a change of the Gaussian intensity. As already mentioned above, the TD-DFT calculations of 1g electronic excitations at the 6-311g(d,p)/M06/DMF level of theory suggested that the two limiting tautomeric forms had a definite non-overlapping range of absorptions. Besides the lack of isosbestic points in the VT-absorption spectra, the 15N VT-NMR data and the computed spectra of each of the tautomeric forms suggest that the UV-vis spectra might have contributions from more than just the two limiting forms. The trend of the L band in the range 20000-17000 cm-1 (500-600 nm) hints that there is more than one transition contributing to its shape, whereas only one intense peak was predicted for the (11111) tautomer, whose intensity arose from two nearly degenerate transitions since the molecule has close to C5h symmetry, and one far weaker transition. Furthermore the two most relevant Gaussian contributions around 18500 (540 nm) and 19900 cm-1 (500 nm) do not have the same dε/dT (Figure 2.32 inset), indicating that the two bands might receive contributions from different phenomena. For this reason the attention was focused on the contribution to the spectrum also by those tautomers whose computed free energies are within the range of energies of the two limiting keto-enamine and enol-imine forms. Observing that the energy of the most intense transition of the enol-imine tautomer was accurately reproduced but that of the keto-enamine tautomer was hyperchromically shifted by 1200 cm-1 (0.15 eV), the computed transition energies of each of the 1g tautomers had been 116  bathochromically shifted assuming that each N-H induced an error in the calculation of transition energies of one fifth of this amount (less than 0.02 eV).  Figure 2.33  Comparison of the experimental UV-vis spectra of 1e in DMF at 213 K (green dashed), 258 K (red continuous), and 213 K (blue continuous) and the TD-DFT computed transition energies (only transitions with oscillator strength larger than 0.05 are reported) at the 6-311G(d,p)/M06/DMF level of theory for the 11111 (green triangles), 00000 (blue triangles), 11110 (yellow squares), 11100 (red triangles), 10101 (aqua circles), and 11000 (black rhombuses) tautomeric forms of 1g.  At low temperatures, the (11111) and the (11110) forms (green triangles and yellow squares of Figure 2.33) prevail and their absorptions determine the shape of band-L and the shoulder of band-T near 19900 cm-1 (500 nm). Upon increasing the temperature, other tautomers start to be more populated (11100 - red triangles, 10101 - aqua circles, the 00000 full enol-imine - blue 117  triangles, and perhaps small contributions also from the 11000 tautomer - black rhombuses). The 00000-form gave the largest contribution to band-M as described above and to the sides of band-H, leading to widening of the band; the intensity did not decrease because the contribution of intermediate tautomeric forms. Furthermore, a number of low intensity transitions contributed to the overall almost constant intensity of the spectrum.  2.3 Conclusions  Campestarenes are a new family of Schiff-base macrocycles exhibiting unique properties. Described in this chapter are their convenient selective synthesis and characterization. They have been prepared using precursors containing different peripheral groups, including bulky triarylsilyl substituents, which prevent aggregation of the molecules. The structures and tautomeric equilibria of campestarenes have been analyzed by a combination of single crystal X-ray diffraction, IR spectroscopy, variable-temperature NMR (1H, 13C, 15N) and UV-Vis experiments, and DFT and TD-DFT calculations on model compounds.  Similar to other salicylaldimines, each repeating unit of campestarenes could have either enol-imine form and keto-enamine form. As macrocycles composed of 5 repeating units, each of which can take either enol-imine or keto-enamine form, campestarenes could exist as many possible tautomers.  Studies in this chapter showed that the predominance of one tautomer over the other is dictated by the solvent. In the solid state, only the enol-imine tautomer was observed and it also prevailed in non-polar solvents such as benzene and toluene. When the solvent was changed to a more polar one, such as chloroform or dichloromethane, the keto-enamine began to appear alongside with other keto-enamine/enol-imine forms. The relative quantities of each are 118  temperature dependent. In highly polar solvents, such as DMF and DMSO, the largest contribution is given by the keto-enamine form of the campestarene while intermediate tautomeric forms are responsible of the temperature dependence of the UV-vis spectra. Computational models of all these species helped to rationalize this unique behavior, and gave insight into the structure of campestarenes in solution. The rich tautomeric behavior of campestarenes combined with their easy access make this new family of five-fold macrocycles suitable to form liquid crystalline phases, proton channels, and metallic complexes. Further investigations of the applications of campestarenes are underway.  2.4 Experimental  2.4.1 Materials All reactions were carried out under air unless otherwise stated. Tetrahydrofuran (THF) was distilled from sodium/benzophenone under nitrogen. Triethylamine was purged with nitrogen gas and dried over 4Å molecular sieves before use. All reagents were used as received unless otherwise stated.  2.4.2 Equipment 300 MHz 1H and 75.5 MHz 13C{1H} NMR spectra were recorded on a Bruker AV-300 spectrometer. 400 MHz 1H and 100.6 MHz 13C{1H} NMR spectra were recorded on a Bruker AV-400 spectrometer. Mass spectra and elemental analyses were obtained at the UBC Microanalytical Services Laboratory. Matrix assisted laser desorption/ionization (MALDI) mass spectra were obtained on a Bruker Biflex IV time-of-flight (TOF) mass spectrometer equipped with a MALDI 119  ion source. Electrospray ionization (ESI) mass spectra were obtained on a Bruker Esquire-LC ion trap mass spectrometer equipped with an electrospray ion source. High-resolution electrospray ionization (HR-ESI) mass spectra were obtained on a Micromass LCT time-of-flight (TOF) mass spectrometer equipped with an electrospray ion source. Samples for both ESI and HR-ESI were analyzed in methanol, methanol / dimethyl sulfoxide mixture or methanol / methylene chloride mixtures at 1 μM. Gramicidin S, Rifampicin, and Erythromycin were used as the references for HR-ESI. Electron Impact (EI) mass spectra were recorded on a Kratos MS-50 double focusing sector mass spectrometer equipped with an EI ion source. Elemental analyses were obtained on a Carlo Erba Elemental Analyzer EA 1108. Melting points were obtained on a Fisher-John’s melting point apparatus. IR spectra were obtained using a Thermo Scientific Nicolet 6700 FT-IR spectrometer equipped with a diamond Attenuated Total Reflectance. UV-vis spectra were obtained (ca. 1 x 10-6 M) on a Varian Cary 5000 UV-vis-near-IR spectrophotometer using a 1 cm quartz cuvette.  2.4.3 Procedure and Experimental Data Synthesis of 4-(Triphenylsilyl)phenol (4d): To a solution of 4-bromophenol  (1 eq., 4.00 g, 23.1 mmol) in dry THF (80 mL) at -78 ºC under a nitrogen atmosphere was added n-BuLi (1.6 mol L-1 in hexanes, 3 eq., 43 mL, 69 mmol), and the solution was stirred at RT for 1 h. A solution of triphenylchlorosilane (1 eq., 6.82 g, 23.1 mmol) in anhydrous THF (20 mL) was added slowly at -78 ºC. The solution was allowed to warm slowly to room temperature, then heated to reflux. After heating at reflux for 12 h, 1 M HCl (150 mL) was added at RT. The aqueous phase was extracted with CH2Cl2 (3 x 20 mL). The combined organic phases were dried over Na2SO4, filtered, and then concentrated under reduced pressure. Silica gel flash column chromatography (eluent: 120  CH2Cl2/hexanes, 1/1) of the residue gave the compound as a white solid (5.84 g, 16.6 mmol, 72% yield).  Data for 4-(Triphenylsilyl)phenol (4d): 1H NMR (400 MHz, CDCl3, 25 ºC): δ = 7.57 (dd, 3JHH = 8.0 Hz, 4JHH = 1.6 Hz, 6H, Ar), 7.41-7.47 (m, 5H, Ar), 7.35-7.40 (m, 6H, Ar), 6.86 (d, 3JHH = 8.4 Hz, 2H, Ar), 4.87 (br, 1H, OH) ppm. 13C{1H} NMR (100.6 MHz, CDCl3, 25 ºC): δ = 157.0, 138.3, 136.5, 134.7, 129.7, 128.0, 125.4, 115.3 ppm. IR (neat): v = 3542, 1597, 1578, 1500, 1481, 1426, 1408, 1263, 1163, 1106, 829, 742, 699 cm-1. ESI-MS (MeOH) m/z = 351.0 [4d-H]-. MP = 228-232 ºC. Elemental Analysis: Calc’d for C24H20OSi : C 81.78%, H 5.72%; Found: C 81.70%, H 5.72%.  Synthesis of 4-(Triisopropylsilyl)phenol (4e): To a solution of 4-bromophenol (1 eq., 2.00 g, 11.6 mmol) in dry THF (40 mL) at -78 ºC under a nitrogen atmosphere was added n-BuLi (1.6 mol L-1 in hexanes, 3 eq., 22 mL, 35 mmol), and the solution was stirred at RT for 1 h. Triisopropylchlorosilane (1 eq., 2.45 mL, 11.6 mmol) was added slowly at -78 ºC. The reaction mixture was warmed slowly to room temperature, then heated to reflux. After heating at reflux for 12 h, 1 M HCl (40 mL) was added at RT. The aqueous phase was extracted with CH2Cl2 (3 x 20 mL). The combined organic phases were dried over Na2SO4, filtered, and then concentrated under reduced pressure. Silica gel flash column chromatography (eluent: CH2Cl2/hexanes, 1/1) of the residue gave the compound as a white solid (2.20 g, 8.80 mmol, 76% yield).  Data for 4-(Triisopropylsilyl)phenol (4e): 1H NMR (400 MHz, CDCl3, 25 ºC): δ = 7.38 (d, 3JHH = 8.7 Hz, 2H, Ar), 6.85 (d, 3JHH = 8.7 Hz, 2H, Ar), 4.82 (s, 1H, OH), 1.37 (sept, 3JHH = 7.6 Hz, 3H, 121  CH), 1.07 (d, 3JHH = 7.6 Hz, 18H, CH3) ppm. 13C{1H} NMR (100.6 MHz, CDCl3, 25 ºC): δ = 156.1, 137.0, 125.9, 114.9, 18.7, 11.0 ppm. IR (neat): v = 3348, 2942, 2864, 1599, 1583, 1503, 1461, 1260, 1225, 1181, 1102, 989, 881, 816, 679, 661 cm-1. ESI-MS (MeOH) m/z = 249.1 [4e-H]-. HRMS (ESI-TOF): Calc’d for C15H25OSi : 249.1675; Found: 249.1674 (-0.4 ppm). MP = 91-93 ºC. Elemental Analysis: Calc’d for C15H26OSi : C 71.93%, H 10.46%; Found: C 71.82%, H 10.68%.  General procedure for the formylation reaction: Triethylamine (2 eq.) was added dropwise to a solution of p-substituted-phenol (1 eq.), MgCl2 (2 eq.) and (CH2O)n (2.2 eq.) in dry THF (ca. 50 mL for 10 mmol of the p-substituted-phenol) under a nitrogen atmosphere. After heating at reflux for 12 h, dilute HCl was added at RT until the precipitate dissolved. Most of the THF was removed by rotary evaporation, then the aqueous phase was extracted with CH2Cl2 (3 x 25 mL). The combined organic phases were dried over MgSO4, filtered, then concentrated under reduced pressure. Silica gel flash column chromatography (eluent: CH2Cl2) of the residue gave the compound.  Synthesis of 2-Hydroxy-5-(1,1,3,3-tetramethyl-butyl)benzaldehyde (3c): This compound was prepared by the general procedure starting with p-(1,1,3,3-tetramethyl-butyl)-phenol (12.0 g, 58.2 mmol).  Isolated as a white solid in 94% yield (12.9 g, 54.9 mmol).  Data for 2-Hydroxy-5-(1,1,3,3-tetramethyl-butyl)benzaldehyde (3c): 1H NMR (400 MHz, CDCl3, 25 ºC): δ = 10.88 (s, 1H, OH), 9.90 (s, 1H, CHO), 7.58 (dd, 3JHH = 8.8Hz, 4JHH = 2.8Hz, 1H, Ar), 7.50 (d, 4JHH = 2.8Hz, 1H, Ar), 6.94 (d, 3JHH = 8.8Hz, 1H, Ar), 1.74 (s, 2H, CH2), 1.38 (s, 6H, CH3), 0.74 (s, 9H, CH3) ppm. 13C{1H} NMR (100.6 MHz, CDCl3, 25 ºC): δ = 197.0, 159.6, 122  142.0, 135.6, 130.6, 120.1, 117.1, 56.8, 38.1, 32.5, 32.0, 31.6 ppm. IR (neat): v = 2955, 2895, 1643, 1621, 1587, 1484, 1377, 1364, 1325, 1284, 1253, 1231, 1189, 1157, 897, 842, 831, 776, 742, 722, 656 cm-1. ESI-MS (MeOH) m/z : 233.2 [3c-H]-. MP = 39-42 ºC. Elemental Analysis: Calc’d for C15H22O2 : C 76.88%, H 9.46%; Found: C 76.81%, H 9.48%.  Synthesis of 2-Hydroxy-5-(triphenylsilyl)benzaldehyde (3d): This compound was prepared by the general procedure starting with p-(triphenylsilyl)-phenol (8.55 g, 24.3 mmol).  Isolated as a white solid in 61% yield (5.64 g, 14.8 mmol).  Data for 2-Hydroxy-5-(triphenylsilyl)benzaldehyde (3d): 1H NMR (400 MHz, CDCl3, 25 ºC): δ = 11.20 (s, 1H, OH), 9.82 (s, 1H, CHO), 7.71-7.73 (m, 2H, Ar), 7.64 (m, 6H, Ar), 7.45-7.47 (m, 3H, Ar), 7.38-7.42 (m, 6H, Ar), 7.03 (d, 3JHH = 8.8 Hz, 1H, Ar) ppm. 13C{1H} NMR (100.6 MHz, CDCl3, 25 ºC): δ = 197.2, 163.0, 144.6, 142.6, 136.4, 133.7, 130.0, 128.2, 125.3, 120.8, 117.6 ppm. IR (neat): v = 3176, 2940, 2887, 2863, 1682, 1649, 1607, 1574, 1463, 1391, 1366, 1280, 1232, 1180, 1132, 1094, 1074, 1014, 990, 908, 881, 828, 769, 737, 727, 694, 676, 650 cm-1. ESI-MS (MeOH) m/z = 378.9 [3d-H]-. MP = 140-143 ºC. Elemental Analysis: Calc’d for C25H20O2Si : C 78.91%, H 5.30%; Found: C 78.84%, H, 5.35%.  Synthesis of 2-Hydroxy-5-(triisopropylsilyl)benzaldehyde (3e): This compound was prepared by the general procedure starting with p-(triisopropylsilyl)-phenol (1.0 g, 4.0 mmol).  Isolated as a white solid in 51% yield (570 mg, 2.1 mmol).  123  Data for 2-Hydroxy-5-(triisopropylsilyl)benzaldehyde (3e): 1H NMR (400 MHz, CDCl3, 25 ºC): δ = 11.07 (s, 1H, OH), 9.93 (s, 1H, CHO), 7.66 (d, 4JHH = 2 Hz, 1H, Ar), 7.64 (dd, 4JHH = 2 Hz, 3JHH = 7.6 Hz, 1H, Ar), 7.01 (d, 3JHH = 7.6 Hz, 1H, Ar), 1.41 (sept, 3JHH = 7.6 Hz, 3H, CH), 1.08 (d, 3JHH = 7.6 Hz, 18H, CH3) ppm. 13C{1H} NMR (100.6 MHz, CDCl3, 25 ºC): δ = 197.2, 162.2, 143.8, 141.2, 125.6, 120.8, 117.1, 18.6, 10.9 ppm. IR (neat): v = 3176, 2941, 2864, 1649, 1608, 1575, 1463, 1366, 1281, 1232, 1181, 1094, 1014, 991, 881, 828, 767, 693, 676 cm-1. ESI-MS (MeOH) m/z = 277.1 [3e-H]-. HRMS (ESI-TOF) Calc’d for C16H25O2Si : 277.1624; Found: 277.1625 (0.4 ppm). MP 52-54 ºC. Elemental Analysis: Calc’d for C16H26O2Si : C 69.01%, H 9.41%; Found: C, 69.14%, H, 9.55%.  Synthesis of 2-Hydroxy-3-nitro-5-(1,1,3,3-tetramethyl-butyl)benzaldehyde (2c): To a solution of 5-(1,1,3,3-tetramethyl-butyl)-2-hydroxybenzaldehyde 3c (1 eq., 5.24 g, 22.4 mmol) in glacial acetic acid (100 mL) at 0 ºC were slowly added p-toluenesulfonic acid (1.2 eq., 5.10 g, 26.8 mmol) and potassium nitrate (1.1 eq., 2.49 g, 24.6 mmol). After stirring at RT for 12 h, water (100 mL) was added and the aqueous phase was extracted with CH2Cl2 (3 x 50 mL). The combined organic phases were dried over MgSO4, filtered, and then concentrated under reduced pressure. Silica gel flash column chromatography (eluent: CH2Cl2/hexanes, 1/1) of the residue gave the compound as a yellow oil that slowly solidified to give a yellow solid (4.38 g, 15.7 mmol, 70% yield).  Data for 2-Hydroxy-3-nitro-5-(1,1,3,3-tetramethyl-butyl)benzaldehyde (2c): 1H NMR (400 MHz, CDCl3, 25 ºC): δ = 11.24 (s, 1H, OH), 10.42 (s, 1H, CHO), 8.32 (d, 4JHH = 2.8Hz, 1H, Ar), 8.13 (d, 4JHH = 2.8Hz, 1H, Ar), 1.77 (s, 2H, CH2), 1.40 (s, 6H, CH3), 0.74 (s, 9H, CH3) ppm. 13C{1H} NMR (100.6 MHz, CDCl3, 25 ºC): δ = 189.5, 154.6, 143.1, 135.0, 134.8, 128.6, 124.9, 56.5, 38.6, 124  32.5, 32.0, 31.3 ppm. IR (neat): v = 2956, 1694, 1666, 1620, 1593, 1532, 1466, 1417, 1366, 1302, 1257, 1223, 1174, 1141, 962, 925, 768, 731, 695 cm-1. ESI-MS (MeOH) m/z = 278.1 [2c-H]-. Elemental Analysis: Calc’d for C15H21NO4 : C 64.50%, H 7.58%, N 5.01%; Found: C 64.43%, H 7.60%, N 5.01%.   Synthesis of 2-Hydroxy-3-nitro-5-(triphenylsilyl)benzaldehyde (2d): To a suspension of 7 g claycop in CHCl3 (130 mL) were added 3d (5.64 g, 14.82mmol) and acetic anhydride (22.2 mL). After stirring at RT for 6 h, the mixture was filtered through celite to collect the filtrate. Water (200 mL) was added and the mixture was stirred for 2 h. The organic phase was collected and the aqueous phase was extracted with CH2Cl2 (3 x 30 mL). The combined organic phases were dried over Na2SO4, filtered, and then concentrated under reduced pressure. Silica gel flash column chromatography (eluent: CH2Cl2/hexanes, 1:2) of the residue gave the compound as a yellow solid (2.78 g, 6.53 mmol, 44% yield). To prepare claycop, 4.03 g montmorillonite K10 was added to a solution of Cu(NO3)2·2.5H2O (4.50 g in 75 mL acetone), the mixture was evaporated to dryness. Then the residue was dried under vacuum. Reference: J. Org. Chem., 1995, 60, 3445-3447  Data for 2-Hydroxy-3-nitro-5-(triphenylsilyl)benzaldehyde (2d): 1H NMR (400 MHz, CDCl3, 25 ºC): δ = 11.70 (s, 1H, OH), 10.30 (s, 1H, CHO), 8.47 (d, 4JHH = 1.6Hz, 1H, Ar), 8.23 (d, 4JHH = 1.6Hz, 1H, Ar), 7.54-7.57 (m, 6H, Ar), 7.48-7.54 (m, 3H, Ar), 7.41-7.46 (m, 6H, Ar) ppm. 13C{1H} NMR (100.6 MHz, CDCl3, 25 ºC): δ = 190.9, 157.3, 144.9, 138.8, 136.3, 135.9, 132.2, 130.5, 128.5, 127.4, 124.7 ppm.  IR (neat): v = 3068, 1667, 1575, 1532, 1427, 1296, 1277, 1229, 1163, 1110, 958, 895, 736, 707, 697 cm-1. ESI-MS (MeOH) m/z = 424.3 [2d-H]-. MP = 162-164 ºC. 125  Elemental Analysis: Calc’d for C25H19NO4Si : C 70.57%, H 4.50%, N 3.29%; Found: C 70.46%, H 4.58%, N 3.30%.  Synthesis of 2-Hydroxy-3-nitro-5-(triisopropylsilyl)benzaldehyde (2e): To a solution of 5-(triisopropylsilyl)-2-hydroxybenzaldehyde 3e (1 eq., 500 mg, 1.80 mmol) in glacial acetic acid (5 mL) at 0 ºC were slowly added p-toluenesulfonic acid (1.2 eq., 410 mg, 2.16 mmol) and potassium nitrate (1.1 eq., 200 mg, 1.98 mmol). After stirring at RT for 12 h, water (20 mL) was added and the aqueous phase was extracted with CH2Cl2 (3 x 20 mL). The combined organic phases were dried over MgSO4, filtered, and then concentrated under reduced pressure. Silica gel flash column chromatography (eluent: CH2Cl2/hexanes, 1/1) of the residue gave the compound as a yellow oil that slowly solidified to give a yellow solid (431 mg, 1.33 mmol, 74% yield).  Data for 2-Hydroxy-3-nitro-5-(triisopropylsilyl)benzaldehyde (2e): 1H NMR (400 MHz, CDCl3, 25 ºC): δ = 11.49 (s, 1H, OH), 10.42 (s, 1H, CHO), 8.40 (d, 4JHH = 1.6 Hz, 1H, Ar), 8.15 (d, 4JHH = 1.6 Hz, 1H, Ar), 1.45 (sept, 3JHH = 7.6 Hz, 3H, CH), 1.09 (d, 3JHH = 7.6 Hz, 18H, CH3) ppm. 13C{1H} NMR (100.6 MHz, CDCl3, 25 ºC): δ = 190.2, 156.9, 143.7, 137.6, 135.4, 127.6, 124.8, 18.5, 10.7 ppm. IR (neat): v = 2940, 2888, 2964, 1660, 1578, 1524, 1459, 1355, 1259, 1234, 1193, 1183, 1123, 1017, 994, 958, 895, 881, 769, 672 cm-1. ESI-MS (MeOH) m/z = 322.1 [2e-H]-. HRMS (ESI-TOF) Calc’d for C16H24NO4Si : 322.1475; Found: 322.1469 (-1.9 ppm). MP = 106-107 ºC. Elemental Analysis: Calc’d for C16H25NO4Si : C 59.41, H 7.79, N 4.33; Found: C 59.25, H 7.86, N 4.33.   126  Synthesis of 2-Hydroxy-3-nitro(15N)-5-(triisopropylsilyl)benzaldehyde (2e-15N): To a solution of 5-(triisopropylsilyl)-2-hydroxybenzaldehyde 3e (1 eq., 1.11 g, 4.01 mmol) in glacial acetic acid (20 mL) at 0 ºC were slowly added p-toluenesulfonic acid (1.2 eq., 915 mg, 4.81 mmol, mg) and K15NO3 (1.1 eq., 450 mg, 4.41 mmol). After stirring at RT for 24 h, water (50 mL) was added and the aqueous phase was extracted with CH2Cl2 (3 x 50 mL). The combined organic phases were dried over MgSO4, filtered, and then concentrated under reduced pressure. Silica gel flash column chromatography (eluent: CH2Cl2/hexanes, 1/1) of the residue gave the compound as a yellow oil that slowly solidified to give a white solid (1.01 mg, 3.13 mmol, 78% yield).  Data for 2-Hydroxy-3-nitro(15N)-5-(triisopropylsilyl)benzaldehyde (2e-15N): 1H NMR (400 MHz, CDCl3, 25 ºC): δ = 11.49 (s, 1H, OH), 10.43 (s, 1H, CHO), 8.40 (dd, 4JHH = 2 Hz, 3JNH = 2 Hz, 1H, Ar), 8.15 (d, 4JHH = 2 Hz, 1H, Ar), 1.45 (sept, 3JHH = 7.6 Hz, 3H, CH), 1.09 (d, 3JHH = 7.6 Hz, 18H, CH3) ppm. 13C{1H} NMR (100.6 MHz, CDCl3, 25 ºC): δ = 190.2, 156.9, 143.7, 137.6, 135.4 (d, 1JCN = 14.6 Hz), 127.6, 124.8, 18.5, 10.7 ppm. 15N NMR (CDCl3, 25 ºC): δ = 371 ppm. IR (neat): v = 2940, 2888, 2864, 1663, 1579, 1499, 1459, 1367, 1336, 1257, 1234, 1194, 1018, 994, 957, 893, 881, 764, 671 cm-1. ESI-MS (MeOH) m/z = 323.1 [(2e-15N) -H]-. HRMS (ESI-TOF) Calc’d for C16H2415NO4Si : 323.1445; Found: 322.1447 (0.6 ppm). MP = 106-107 ºC. Elemental Analysis: Calc’d for C16H2515NO4Si : C 59.23%, H 7.77%, N 4.62%; Found: C 59.33%, H 7.90%, N 4.35%.   General procedure for the synthesis of Campestarenes: Sodium dithionite (Na2S2O4, 6 eq.) was added to a solution of the nitrosalicylaldehyde (1 eq.) in MeOH and water (40/3 in volume, about 10 mL for 100 mg of nitrosalicylaldehyde), and heated at reflux for 12 h under air. After cooling 127  down to RT, water was added, followed by extraction with CH2Cl2. The organic phase was dried with Na2SO4. Then the solvent was removed by rotary evaporation. Flash column chromatography using neutral alumina (eluent: CH2Cl2/EtOH) gave the macrocycle as a dark purple or dark red brown solid.   Synthesis of Campestarene (1c) : This compound was prepared by the general procedure starting with 2c (0.72 mmol, 200 mg). Isolated as a dark purple solid in 35% yield (58 mg, 0.050 mmol).   Data for Campestarene (1c) : 1H NMR (400 MHz, CD2Cl2, 25 ºC): δ = 16.65 (s, 5H, OH), 9.04 (s, 5H, imine), 7.72 (s, 5H, Ar), 7.49 (s, 5H, Ar), 1.86 (s, 10H, CH2), 1.51 (s, 30H, CH3), 0.85 (s, 45H, CH3) ppm. 13C{1H} NMR (100.6 MHz, CD2Cl2, 25 ºC): δ = 158.1, 155.1, 139.5, 132.7, 128.7, 120.4, 116.7, 57.4, 38.7, 32.9, 32.2, 32.0 ppm. MALDI-TOF MS (DCTB matrix) m/z = 1156.9 [1c+H]+, 1178.9 [1c +Na]+. MP decomposition before melting.   Synthesis of Campestarene (1d): This compound was prepared by the general procedure starting with 2d (0.47 mmol, 200 mg).  Isolated as a dark purple solid in 29% yield (51 mg, 0.027 mmol).   Data for Campestarene (1d) : 1H NMR (400 MHz, CD2Cl2, 25 ºC): δ = 17.34 (d, 3JHH = 8 Hz, 5H, OH), 8.49 (d, 3JHH = 8 Hz, 5H, imine), 7.66 (d, 4JHH = 0.8 Hz, 1H, Ar), 7.57 (dd, 3JHH=8 Hz, 4JHH = 1.2 Hz, 30H, Ar), 7.40-7.46 (m, 15H, Ar), 7.34-7.40 (m, 35H, Ar) ppm. 13C{1H} NMR (100.6 MHz, CD2Cl2, 25 ºC): δ = 166, 153, 141, 136, 134, 132, 130, 128, 125, 118 ppm (one signal missing). IR (neat): v = 3068, 1609, 1509, 1427, 1229, 1108, 910, 741, 697 cm-1. MALDI-TOF 128  MS (DCTB matrix) m/z = 1887.7 [1d +H]+, 1909.6 [1d +Na]+, 1925.6 [1d +K]+. MP decomposition before melting.   Synthesis of Campestarene (1e) : This compound was prepared by the general procedure starting with 2e (0.62 mmol, 200 mg). Isolated as a dark purple solid in 38% yield (65 mg, 0.05 mmol).   Data for Campestarene (1e) : 1H NMR (400 MHz, CD2Cl2, 25 ºC): δ = 17.18 (d, 3JHH = 4 Hz, 5H, OH), 8.94 (d, 3JHH = 4 Hz, 5H, imine), 7.74 (s, 5H, Ar), 7.56 (s, 5H, Ar), 1.52 (sept, 3JHH = 7.2 Hz, 15H, CH), 1.18 (d, 3JHH = 7.2 Hz, 90H, CH3) ppm. 13C{1H} NMR (100.6 MHz, CD2Cl2, 25 ºC): δ = 162.7, 154.5, 139.1, 132.5, 124.6, 121.4, 120.3, 19.0, 11.5 ppm. IR (neat): v = 2943, 2890, 2864, 1613, 1512, 1462, 1282, 1229, 1116, 1015, 995, 917, 882, 676, 644 cm-1. MALDI-TOF MS (DCTB matrix) m/z = 1377.5 [1e+H]+, 1399.5 [1e +Na]+, 1415.6 [1e +K]+. MP decomposition before melting.   Synthesis of Campestarene (1e-15N5): This compound was prepared by the general procedure starting with 2e-15N (0.62 mmol, 200 mg). Isolated as a dark purple solid in 36% yield (61 mg, 0.044 mmol).   Data for Campestarene (1e-15N5) : 1H NMR (400 MHz, CD2Cl2, 25 ºC): δ = 17.18 (dd, 1JNH = 64 Hz, 3JHH = 4 Hz, 5H, OH), 8.94 (dd, 2JNH = 3.2 Hz, 3JHH = 4 Hz, 5H, imine), 7.74 (s, 5H, Ar), 7.56 (s, 5H, Ar), 1.52 (sept, 3JHH = 7.2 Hz, 15H, CH), 1.18 (d, 3JHH = 7.2 Hz, 90H, CH3) ppm. 13C{1H} NMR (100.6 MHz, CD2Cl2, 25 ºC): δ = 163, 155, 139, 133, 125, 121, 120, 19, 11 ppm. IR (neat): v = 2943, 2890, 2865, 1603, 1461, 1280, 1228, 1116, 1027, 1015, 913, 882, 711, 675, 644 cm-1. 129  ESI-MS (MeOH) m/z = 1404.9 [(1e-15N5) +Na]+. HRMS (ESI-TOF) Calc’d for C80H12615N5O5Si5: 1381.8457; Found: 1381.8446 (-0.8 ppm). MP decomposition before melting.  2.4.4 Details of Crystallography Crystallography was performed by Dr. Nicholas G. White, but details are included here for completeness.  A single crystal of campestarene 1e was obtained by vapor diffusion of pentane into a dichloromethane solution of the molecule. Single crystal X-ray data were collected on a Bruker APEX DUO diffractometer using graphite monochromated Mo Kα radiation ( = 0.71073 Å) at 90 K. Despite using long exposure times (240 s/º), reflections were only visible to approximately 0.85 Å resolution. Raw frame data (including data reduction, interframe scaling, unit cell refinement and absorption corrections) were processed using APEX2. The structure was solved using SUPERFLIP and refined using full-matrix least-squares on F2 within the CRYSTALS suite. All non-hydrogen atoms were refined with anisotropic displacement parameters. Hydrogen atoms on the macrocyclic ring were visible in the Fourier difference map and were initially refined with restraints on bond lengths and angles, after which the positions were used as the basis for a riding model.  The TIPS hydrogen atoms were inserted at calculated positions, while the phenol O-H protons are inserted at idealized hydrogen bonding positions; these positions were then used as the basis for a riding model. The TIPS groups exhibited high thermal motion: in one case, it was possible to model this as positional disorder over two sites; in other cases, positional disorder could not be modelled, and so these atoms had large thermal ellipsoids.  Residual electron density was located around the 130  macrocycle framework - it appeared that this may be due to positional disorder of the entire macrocycle, although the data were not of sufficient quality to allow modelling.  Some diffuse electron density was present, which could not be modeled sensibly, and so PLATON-SQUEEZE was used to include this electron density in the refinement.       Generally, the structure was of relatively low quality, and it was necessary to add restraints necessary to bond lengths and bond angles as well as thermal and vibrational ellipsoid parameters in order to achieve a chemically sensible refinement.  Despite these difficulties, the structure and connectivity, as well as the overall crystal packing of the molecule could be unambiguously determined.  Table 2.6  Selected crystallographic data for structure of 1e. Parameter Value Radiation type Mo Temperature (K) 90 Chemical formula C80H125N5O5Si5a   Formula weight 1377.33a a (Å) 16.5631(9) b (Å) 14.8911(7) c (Å) 38.301(2)  (º) 90 β (º) 90.0043(14) γ (º) 90 Unit cell volume (Å3) 9446.8(9) Crystal system monoclinic Space group P21/c Z 4 Reflections (all) 35380 Reflections (unique) 17133 Rint 0.027 131  R1 [I > 2(I)] 0.181 wR2(F2)      (all data) 0.445  2.4.5 Computational Methods Computational studies were performed by Dr. Francesco Lelj, but details are included for completeness. All of the reported computations were performed by using the Gaussian09 rev. D.01 software.119 All studies were performed by applying the Density Functional Theory (DFT).120 SCF and structure optimization convergence criteria are the default ones. Solvation effects were included by the means of the IEFPCM method121 implemented in the standard software without any change with respect to the default parameters. UV-vis spectra were computed using the TD-DFT linear response theory122 and using the 6-311G(d,p) basis (standard basis sets in Gaussian 09) as in case of geometry optimization and M06 xc functional. Different xc functionals, hybrid PBE1PBE,123 double hybrid mPW2PLYP,124 and meta-hybrid M06 xc functional125 were used in some computations. In all the computations, the integration grid was augmented in comparison to the default one (“int=ultrafine” key), as suggested for limiting errors originated by small integration-grids in energy computations with meta-GGA xc functionals.126  The vibration frequencies discussed in the text were obtained by scaling the computed harmonic frequencies by a 0.950 factor. This factor produced a satisfactory match between experimental and computational IR features in the 1400-1800 cm-1 region of the spectrum. This value was identical to the one proposed for M06 and the 6-31+G(d,p) basis set127 and was in line with the ones suggested for the same xc functional with polarization-consistent basis sets (between 0.963 and 0.965)128 and, more recently, for the 6-31G(d) basis set (0.959).129  132  Spin-Spin indirect NMR coupling constants parameters were computed130 with the GIAO method131 using the Local Dense Basis Set approach132 where larger basis set was employed on the nuclei interested by the calculation. In our case the cc-pQVZ133 basis set on the N, C and H atoms were involved in the coupling whereas on the remaining atoms the 6-311G(d,p) basis sets were used. Atomic charges had been computed according to the Merz-Kollman procedure. 134 Experimental UV-vis spectra and their first derivative were fitted using the same number of Gaussian functions in the whole set of temperatures using an in house LSQ program.   133  Chapter 3: Synthesis of Methyl/Phenyl Substituted Campestarenes  3.1 Introduction  In Chapter 2, a convenient and efficient way to synthesize campestarenes with various substituent groups through a one-pot reaction was demonstrated. All of the reported campestarenes and campestarene precursors shared the same general structure shown in Scheme 3.1.   Scheme 3.1  General synthesis of campestarenes from 2-hydroxy-3-nitrobenzaldehyde precursors.  All these campestarenes were prepared from 2-hydroxy-3-nitrobenzaldehyde compounds. After a reduction reaction, the generated amine group underwent a condensation reaction with an aldehyde group spontaneously, eventually yielding a 5-fold symmetric Schiff-base macrocycle. One of the advantages of using 2-hydroxy-3-nitrobenzaldehyde compounds as starting materials in campestarene syntheses is that the condensation reactions between amines and aldehydes are easily accessible. In addition, the equilibrium reaction allows the system to produce the most thermodynamically stable product--the campestarenes in this instance. As a result, condensation 134  reactions between amines and aldehydes have been widely used in Schiff-base macrocycle chemistry.135 Although this established synthetic strategy is convenient and efficient, the resulting regular campestarenes suffer issues with stability. Since the imine condensation reaction is reversible, campestarenes decompose slowly by hydrolysis. Although campestarenes decompose slowly in solid state, this decomposition process is accelerated when macrocycles are dissolved in organic solvents. Furthermore, this process can be further accelerated by increasing the temperature. As shown in Figure 3.1, as a solution of campestarene 1e in toluene-d8 was heated above 50 °C, a large portion of macrocycles started decomposing rapidly, resulting in new peaks corresponding to impurities. Unfortunately, when this sample was cooled back to room temperature, the new peaks did not disappear. This indicates that regular campestarene 1e is not stable at relatively high temperatures (above 50 °C) despite being in a neutral environment.   Figure 3.1  1H NMR spectra of campestarene 1e in toluene-d8, (a) when temperature was increased from 303 K to 377 K; (b) when temperature was decreased from 377 K to 315 K (400 MHz).   135  Since the hydrolysis of Schiff base can be catalyzed with acids,136 it is not surprising that campestarenes are not stable under acidic conditions. For example, campestarenes dissolved in slightly acidic CDCl3 tend to show formation of oligomers within several hours. Moreover, experiments with transition metal salts such as copper(II) acetate and zinc(II) acetate led to decomposition of campestarenes. Weakly basic conditions created by salt such as Na2CO3 and Na2SO3 also facilitate the decomposition of campestarenes. The decomposition of campestarene in the presence of bases can be explained by the deprotonation of phenolic protons in campestarenes. Since campestarenes are stabilized by intramolecular hydrogen bonding, removal of protons involved in such intramolecular hydrogen bonds would lead to the loss of stability. The stability issue of regular campestarenes not only caused difficulties in the studies of campestarenes, but also limits the application of these 5-fold symmetric macrocycles, such as in coordination chemistry 137  and liquid crystals. 138  In this chapter, the syntheses of different campestarene derivatives from aryl or methyl ketone compounds are explored.  3.2 Results and Discussion  With the aim of preparing more stable campestarenes, the synthesis of campestarene derivatives from ketone compounds was explored. The general strategy is shown in Scheme 3.2. Due to the steric hindrance caused by introducing acetyl or phenyl groups, imine hydrolysis should slow down, resulting in more stable macrocycles. This replacement might also introduce some different properties to campestarenes.   136  Scheme 3.2  General synthetic route to campestarene derivatives from ketone compounds.   3.2.1 Synthesis of Methyl Substituted Campestarene The acetyl campestarene precursor 3j was synthesized as shown in Scheme 3.3. First, the phenol compound was treated with base and acetyl chloride to yield the phenyl acetate. The acetophenone framework was prepared through Fries rearrangement of the acetate ester. Following nitration with PTSAH2O and KNO3, the macrocycle precursor 3j was isolated as a yellow solid. Following the previous procedure for campestarene synthesis (see Scheme 2.2), acetyl precursor 3j was treated with Na2S2O4 for reduction. Different from reactions with regular campestarene precursors, this reaction yielded the amine monomer quantitatively instead of condensation products. As the replacement of the aldehyde with a ketone group in the macrocycle precursor might increase the stability of campestarene, the reaction rate of imine formation would decrease as well due to both steric hindrance and electronic effects. As a result, macrocycles did not for spontaneously from the amine compound. Actually, only the resulting amine compound 2j could be isolated from the reduction reaction mixture. 137  Scheme 3.3 Synthesis of acetyl derived campestarene precursor 2j.  Although many literature reports show that imine synthesis between ketones and amines can be conducted by simply heating the amine and ketone to reflux in toluene139 or ethanol,140 refluxing 2j in toluene or EtOH did not produce macrocycles. Only a small amount of a dimer (Figure 3.2) was observed after 48 h at reflux. According to some literature, the use of drying reagents can improve imine synthesis. With the purpose of driving equilibrium to the imine side, drying reagent helped to remove excess water. Considering that campestarenes are not stable under strongly acidic or basic conditions, activated neutral 4Å molecular sieves were used to remove excess water. 141 Moreover, since regular imine synthesis between ketone and amine is much slower than the reaction between an aldehyde and an amine, a catalyst is required to accelerate the macrocyclization process. Both acid and base were tested to optimize the macrocyclization. To find appropriate conditions for preparing campestarene 1j, various reaction conditions were explored (Table 3.1).   138   Figure 3.2  Structures of dimer, trimer and tetramer synthesized from 2j.  Table 3.1  Reaction conditions for the synthesis of campestarene 1j from amine compound 2j. Number Solvent Catalyst Drying reagent T/oC Products# 1 Ethanol piperidine NA 55 Linear trimer 2 Ethanol PTSAH2O NA 55 Linear tetramer 3 Ethanol piperidine Molecular sieve 4Å 55 Linear trimer 4 Ethanol PTSAH2O Molecular sieve 4Å 55 Linear tetramer 5 1-Butanol PTSAH2O Molecular sieve 4Å reflux Linear tetramer 6 Toluene PTSAH2O Molecular sieve 4Å  reflux campestarene 7 NA NA NA 200 Insoluble solid #: only products with relatively high molecular weight are reported. Products were observed by ESI-MS spectroscopy. Experiments 1-6 were stirred for 14 days, Experiment 7 was stopped after 1h.  139  Condensation reactions were first conducted in ethanol with either piperidine 142  or PTSAH2O143 as catalyst. Based on the results of Experiments 1-4 in Table 3.1, I found that PTSAH2O is a better catalyst than piperidine. Hence, only PTSAH2O was used as catalyst hereafter. Although no campestarene could be found in these reactions, formation of linear oligomers indicated that imine formation was successfully accelarated under such conditions. To further increase the reaction rate of imine synthesis, solvents with higher boiling points were used in place of ethanol. Although switching to solvents with higher boiling points did quicken the color change of the reaction mixture, the chain growth of linear oligomers in 1-butanol stopped at the tetramer. Neither the linear pentamer nor campestarene could be found after 14 days of reaction. In contrast, the reaction performed at the same temperature in toluene showed formation of campestarenes in the reaction mixture (Figure 3.3). The peak at m/z = 899.3 indicated the formation of campestarene 1j [1j+H]+. Unfortunately, only a small amount of campestarene was found in the mixture along with mostly linear oligomers. Due to low yield of campestarene 1j, no macrocycle could be isolated from the mixture. In addition, another imine synthesis was conducted under neat conditions by melting precursor 2j at 200 oC.144 Although this reaction process was much faster than previous conditions, only a red insoluble solid was obtained after the reaction. Under these neat conditions at high temperature, water was removed from reaction mixture as the imine was synthesized. The absence of water prevents the hydrolysis of imines, which is important to obtain the thermodynamically stable product. Instead, linear oligomers formed as kinetically controlled products.  140   Figure 3.3  ESI-MS of mixture from reaction conducted in toluene with the presence of PTSAH2O and molecular sieve 4Å  (Experiment 6 in Table 3.1).  3.2.2 Synthesis of Phenyl Substituted Campestarenes Similar to acetyl derived campestarenes, phenyl derived campestarenes are expected to have better stability due to the steric effects of phenyl groups. Meanwhile, the introduction of phenyl rings would further extend the conjugation system of campestarene, which may affect campestarenes’ solvatochromism properties.145 The first phenyl derived campestarene precursor 3k was synthesized as shown in Scheme 3.4. 4-Methylphenol was treated with base and benzoyl chloride to yield phenyl benzoate. The benzophenone framework was obtained through Fries rearrangement of this benzoate ester. Nitration was conducted by reacting the formylation products with PTSAH2O and KNO3 in acetic acid. The resulting nitro compound 3k was the macrocycle precursor. A reduction reaction was required to yield 2k before macrocyclization. Similar to acetyl derived campestarene precursor 3j, 2k did not undergo a rapid condensation reaction after 141  reduction. As a result, different conditions were tested in order to prepare campestarene 1k from amine compound 2k.   Scheme 3.4  Synthesis of phenyl derived campestarene precursor 2k.   Figure 3.4  ESI-MS spectrum of mixture from reaction conducted in toluene with the presence of PTSAH2O. 142  Based on previous observations of 1j synthesis, PTSAH2O was chosen as catalyst. Both 1-butanol and toluene were chosen as solvents so the imine synthesis could be conducted at higher temperature to accelerate the process. After 5 days of reflux in toluene, ESI-MS spectrum showed the formation of a small quantity of campestarene 1k (Figure 3.4). The signal at 1046.9 was assigned to campestarene 1k [1k+H]+. However, the yield of campestarene was so low that no macrocycle could be isolated from the reaction mixture. Notably, even without 4Å molecular sieves, the reaction yielded campestarene in a very low yield.   Figure 3.5  ESI-MS spectrum of reaction mixture from reaction conducted under neat condition.  143  In addition, a neat reaction was conducted by melting amine 2k at 160 ºC. A dark red solid was obtained as product. The soluble portion (dichloromethane/MeOH) was characterized with ESI-MS (Figure 3.5). In the mass spectrum, a peak at 877.7 is attributed to linear tetramers [tetramer + Na]+ (877.4) while a peak at 1086.9 is attributed to a linear pentamer [pentamer + Na]+ (1086.4). The most intense signal at 1311.9 is assigned to a linear hexamer [hexamer + K]+ (1311.5). Neither the cyclic pentamer nor cyclic hexamer was observed in the MS spectrum. MS analysis at lower m/z ranges showed only linear products and no multiply charged macrocycles. Compared to the reaction conducted in solution, the neat condition was not an appropriate condition for campestarene synthesis. One of the reason could be a lack of water at such high temperatures.  The reversibility of imine synthesis allows for error checking, or, in other words, proofreading.146 This is very important for the formation of macrocycles, when compared with kinetically controlled approaches.147 Thus, imine synthesis became irreversible in the absence of water, and these conditions did not allow the most thermodynamically stable compound to form.  Although both campestarenes 1j and 1k were not isolated from their respective reactions, both of them showed the potential to be prepared as products. However, during the experiments, both acetyl and aryl derived products (linear oligomers and macrocycles) demonstrated poor solubility in most organic solvents. The poor solubility might not only restrain the formation of macrocycles, but also inhibit their potential applications. To take the ketone derived campestarene one step further, the introduction of bulky substituent groups into these macrocycle precursors was considered. The introduction of bulky groups should not only help to increase the solubility of ketone derived campestarenes, but also prevent the possible aggregation of macrocycles. Based on 144  previous experience with regular campestarenes, the 1,1,3,3-tetramethylbutyl group was chosen to be installed onto the macrocycle precursor 3l (Scheme 3.2).  However, previously both acetyl and aryl precursors 3j and 3k were synthesized through Fries rearrangement of esters. The existence of 1,1,3,3-tetramethylbutyl groups on the phenolic ring would interrupt the Fries rearrangement with AlCl3 due to the dealkylation of 1,1,3,3-tetramethylbutyl groups through a reverse Friedel-Crafts reaction. To avoid this dealkylation reaction, another synthetic route was used for the synthesis of campestarene precursor 3l (Scheme 3.5). Instead of the Fries rearrangement of benzoate esters, the benzophenone framework was prepared through a copper-catalyzed ortho-acylation of 4-(1,1,3,3-tetramethylbutyl)phenol. 148 Nitration was performed by reacting ketone products with PTSAH2O and KNO3 in acetic acid. A reduction reaction was conducted to yield amine compound 2l before macrocyclization.  Scheme 3.5  Synthesis of phenyl derived campestarene precursor 2l.  145  With 1,1,3,3-tetramethylbutyl group introduced, amine 2l demonstrated better solubility in organic solvents. Various condensation conditions were performed with 2l to synthesize campestarene (Table 3.2). Among these reaction conditions, most yielded linear oligomers rather than macrocycles. Although either titanium(IV) chloride149 or titanium(IV) isopropoxide150 could catalyse the imine condensation very effectively, reactions conducted in the presence of titanium(IV) chloride or titanium(IV) isopropoxide only resulted in linear oligomers. The optimized reaction conditions for the synthesis phenyl derived campestarene 1l are shown in Scheme 3.6.   Table 3.2  reaction conditions for the synthesis of campestarene 1l from amine 2l. # Solvent Catalyst T/ ºC Time Products# 1 p-xylene PTSAH2O reflux 14 d Linear oligomers 2 Toluene PTSAH2O reflux 14 d Linear oligomers 3 1-Butanol PTSAH2O reflux 14 d campestarene 4 1-Butanol PTSAH2O microwave 30 min Linear oligomers 5 Toluene Titanium(IV) chloride RT 4 d Linear oligomers 6 Toluene Titanium(IV) isopropoxide 80 16 h Linear oligomers 7 Toluene Titanium(IV) isopropoxide RT 4 d Linear oligomers #: only products with relatively high molecular weight are reported. Products were observed by ESI-MS spectroscopy.   During the macrocyclization reaction, formation of the mixture was monitored with ESI-MS. After 14 days of reflux, the peaks at 1538.9, 1560.8 and 1576.8 appeared in the mass spectrum of reaction mixture. They can be assigned to campestarene 1l [M + H]+ , [M + Na]+ and [M + K]+ 146  respectively (Figure 3.6 (a)). With the increase of reaction duration, the campestarene signals increased their relative intensity (Figure 3.6 (b)), which indicated the increased yield of campestarene 1l with increased reaction time.   Scheme 3.6  Synthesis of campestarene 1l from amine compound 2l.  147   Figure 3.6  ESI-MS spectrum of reaction mixture from reaction conducted in toluene with the presence of PTSAH2O (Experiment 3 in Table 3.2) after (a) 7 days; (b) 14 days; (c) 28 days; (d) 38 days.  148  After 38 days, the reaction was stopped and campestarene was isolated as a purple solid with a yield of approximately 15%. This was the first isolated campestarene synthesized from ketone compounds. Figure 3.7 shows the 1H NMR spectrum of campestarene 1l. The simplicity of the spectrum confirms that the molecule has high symmetry. Notably, the phenolic proton in 1l shifted downfield to about 16.7 ppm due to intramolecular hydrogen bonding. This is consistent with what is observed with regular campestarenes. Only one set of peaks could be observed at relatively high concentration, indicating no dimerization at room temperature.151 The mass spectrum of purified 1l is shown in Figure 3.8. Campestarene 1l tends to bind to sodium and potassium cations, which is similar to regular campestarenes as well.  Figure 3.7  1H NMR spectrum of campestarene 1l in CDCl3 (room temperature, 400 MHz). Only the spectral region above 6 ppm is shown.   149   Figure 3.8  ESI-MS spectrum of campestarene 1l.  3.3 Conclusion  In order to study the potential applications of campestarenes in coordination chemistry and in liquid crystals, the disadvantage of the low stability of campestarenes must be overcome. With the purpose of improving the stability of campestarenes, acetyl and aryl derived campestarene precursors were synthesized. Different from regular campestarene precursors, which undergo macrocyclization spontaneously after reduction reaction, the formation of acetyl and aryl derived campestarenes required specific conditions to proceed. Various reaction conditions were conducted in order to synthesize campestarene derivatives. Although both acetyl and aryl campestarenes could be prepared with the presence of PTSAH2O as catalyst, many of the reactions suffered from low yield so that isolation became very challenging and impractical. 150  With the introduction of bulky 1,1,3,3-tetramethylbutyl groups, phenyl derived campestarene precursor 2l demonstrated improved solubility in organic solvents. With the presence of PTSAH2O as catalyst, campestarene 1l was isolated as a purple solid with an isolated yield of about 15%. This was the first isolated ketone derived campestarene. However, the current optimized reaction conditions for campestarene 1l took about 5 weeks. This low reaction rate is the result from the relatively low reactivity of ketones due to both steric hindrance and electronic effects. Despite the selective formation of phenyl derived campestarenes, this time-consuming process limits the practical application of this synthetic route. More investigation is needed to improve the reaction rate.   3.4 Experimental  3.4.1 Materials All reactions were carried out under air unless otherwise stated. Tetrahydrofuran (THF) was distilled from sodium/benzophenone under nitrogen. Triethylamine was purged with nitrogen gas and dried over 4Å molecular sieves before use. All reagents were used as received unless otherwise stated.  3.4.2 Equipment 300 MHz 1H and 75.5 MHz 13C{1H} NMR spectra were recorded on a Bruker AV-300 spectrometer. 400 MHz 1H and 100.6 MHz 13C{1H} NMR spectra were recorded on a Bruker AV-400 spectrometer. Mass spectra and elemental analyses were obtained at the UBC Microanalytical Services Laboratory. Electrospray ionization (ESI) mass spectra were obtained on a Bruker 151  Esquire-LC ion trap mass spectrometer equipped with an electrospray ion source. Samples for both ESI and HR-ESI were analyzed in methanol or methanol / methylene chloride mixtures at 1 μM. Elemental analyses were obtained on a Carlo Erba Elemental Analyzer EA 1108. Melting points were obtained on a Fisher-John’s melting point apparatus. IR spectra were obtained using a Thermo Scientific Nicolet 6700 FT-IR spectrometer equipped with a diamond Attenuated Total Reflectance.   3.4.3 Procedure and Experimental Data Synthesis of 1-(2-hydroxy-5-propylphenyl)ethan-1-one (4j): To a solution of 4-propylphenol (1 eq., 2.95 g, 21.7 mmol) in acetone (40 mL) at room temperature was added triethylamine (2 eq., 6.04 mL, 43.3 mmol) and acetyl chloride (1 eq., 1.55 mL, 21.7 mmol). The solution was then refluxed for 4 h. The reaction mixture was treated with 2 % HCl aqueous solution after cooling. The aqueous phase was extracted with CH2Cl2 (3 x 40 mL). The combined organic phases were dried over Na2SO4, filtered, and then concentrated under reduced pressure. The residue was directly used for the next step without further purification. The obtained phenyl acetate derivative was reacted with AlCl3 (1.5 eq., 4.33 g, 32.5 mmol) in a flask placed in an oil bath kept under temperature control at 120 °C. After 1 h reaction, the mixture was poured into 1M ice-cold HCl solution. The aqueous phase was extracted with CH2Cl2 (3 x 40 mL). The combined organic phases were dried over Na2SO4, filtered, and then concentrated under reduced pressure. Silica gel flash column chromatography (eluent: CH2Cl2/hexanes, 1/1) of the residue yielded the compound as a white solid (2.80 g, 15.7 mmol, 73% yield).  152  Data for 1-(2-hydroxy-5-propylphenyl)ethan-1-one (4j): 1H NMR (400 MHz, CDCl3, 25 ºC): δ = 12.11(s, 1H, OH), 7.51 (d, J = 2.0 Hz, 1H, Ar), 7.30 (dd, J = 8.4 Hz, 2.0 Hz, 1H, Ar), 6.91 (d, J = 8.4 Hz, 1H, Ar), 2.64 (s, 3H, CH3), 2.56 (t, J = 7.6 Hz, 2H, CH2), 1.63 (m, 2H, CH2), 0.96 (t, J = 7.6 Hz, 3H, CH3) ppm. ESI-MS (MeOH) m/z = 177.1 [4j-H]-. Elemental Analysis: Calc’d for C11H14O2 : C 74.13%, H 7.92%; Found: C 74.09%, H 7.88%.  Synthesis of 1-(2-hydroxy-3-nitro-5-propylphenyl)ethan-1-one (3j): To a solution of  1-(2-hydroxy-5-propylphenyl)ethan-1-one 4j (1 eq., 2.80 g, 15.7 mmol) in glacial acetic acid (50 mL) at 0 ºC were slowly added p-toluenesulfonic acid (1.2 eq., 3.59 g, 18.9 mmol) and potassium nitrate (1.1 eq., 1.76 g, 17.3 mmol). After stirring at RT for 12 h, water (100 mL) was added and the aqueous phase was extracted with CH2Cl2 (3 x 50 mL). The combined organic phases were dried over Na2SO4, filtered, and then concentrated under reduced pressure. Silica gel flash column chromatography (eluent: CH2Cl2/hexanes, 1/1) of the residue yielded the compound as a yellow oil that slowly solidified to yield a yellow solid (2.33 g, 10.4 mmol, 66% yield).  Data for 1-(2-hydroxy-3-nitro-5-propylphenyl)ethan-1-one (3j): 1H NMR (400 MHz, CDCl3, 25 ºC): δ = 12.81 (s, 1H, OH), 8.02 (d, J = 2.4 Hz, 1H, Ar), 7.85 (d, J = 2.4 Hz, 1H, Ar), 2.72 (s, 3H, CH3), 2.63 (t, J = 7.6 Hz, 2H, CH2), 1.67 (m, 2H, CH2), 0.97 (t, J = 7.6 Hz, 3H, CH3) ppm. 13C{1H} NMR (100.6 MHz, CDCl3, 25 ºC): δ = 202.5, 154.2, 137.3, 136.5, 133.2, 131.0, 123.6, 36.6, 28.4, 24.2, 13.5 ppm. IR (neat): v = 3078, 2960, 2931, 2868, 1653, 1581, 1521, 1460, 1359, 1340, 1268, 1222, 1172, 1104, 1085, 978, 854, 806, 771, 660 cm-1. ESI-MS (MeOH) m/z = 224.2 [3j+H]+. Elemental Analysis: Calc’d for C11H13NO4 : C 59.19%, H 5.87%, N 6.27%; Found: C 59.09%, H 5.88%, N 6.42%.  153  General procedure for the synthesis of amine precursors: Sodium dithionite (Na2S2O4, 6 eq.) was added to a solution of the nitro compounds (1 eq.) in MeOH and water (40/3 in volume, about 10 mL for 100 mg of nitro compounds), and heated at reflux for 12 h under air. After cooling down to RT, water was added, followed by extraction with CH2Cl2. The organic phase was dried with Na2SO4. Then the solvent was removed by rotary evaporation. The resulting amine compounds are used for the next step without further purification.  Synthesis of 1-(3-amino-2-hydroxy-5-propylphenyl)ethan-1-one (2j): This compound was prepared by the general procedure starting with 1-(2-hydroxy-3-nitro-5-propylphenyl)ethan-1-one 3j  (2.24 mmol, 500 mg).  Data for 1-(3-amino-2-hydroxy-5-propylphenyl)ethan-1-one (2j): 1H NMR (400 MHz, CDCl3, 25 ºC): δ = 12.32 (s, 1H, OH), 6.95 (d, J = 2.0 Hz, 1H, Ar), 6.77 (d, J = 2.0 Hz, 1H, Ar), 5.31 (S, 2H, NH2), 2.61 (s, 3H, CH3), 2.48 (t, J = 7.6 Hz, 2H, CH2), 1.61 (m, 2H, CH2), 0.94 (t, J = 7.6 Hz, 3H, CH3) ppm. ESI-MS (MeOH) m/z = 194.4 [2j+H]+.  Synthesis of (2-hydroxy-5-methylphenyl)(phenyl)methanone (4k): To a solution of 4-methylphenol (1 eq., 3.72 g, 34.4  mmol) in acetone (40 mL) at room temperature was added trimethylamine (2 eq., 9.60 mL, 68.9 mmol) and benzoyl chloride (1 eq., 4.0 mL, 34.4 mmol). Then the solution was refluxed for 4 h. The reaction mixture was treated with 2 % HCl aqueous solution after cooling. The aqueous phase was extracted with CH2Cl2 (3 x 40 mL). The combined organic phases were dried over Na2SO4, filtered, and then concentrated under reduced pressure. The residue was directly used for the next step without further purification. 154  The obtained p-tolyl benzoate was reacted with AlCl3 (1.5 eq., 6.88 g, 51.6 mmol) in a flask placed in an oil bath kept under temperature control at 120 °C. After 1 h reaction, the mixture was poured into 1M ice-cold HCl solution. The aqueous phase was extracted with CH2Cl2 (3 x 40 mL). The combined organic phases were dried over Na2SO4, filtered, and then concentrated under reduced pressure. Silica gel flash column chromatography (eluent: CH2Cl2/hexanes, 1/1) of the residue yielded the compound as a yellow solid (6.50 g, 30.6 mmol, 89% yield).  Data for (2-hydroxy-5-methylphenyl)(phenyl)methanone (4k): 1H NMR (400 MHz, CDCl3, 25 ºC): δ = 11.87 (s, 1H, OH), 7.66-7.71 (m, 2H, Ar), 7.61 (tt, J = 7.6 Hz, 1.6 Hz, 1H, Ar), 7.50-7.55 (m, 2H, Ar), 7.32-7.39 (m, 2H, Ar), 7.00 (d, J = 8.4 Hz, 1H, Ar), 2.27 (s, 1H, Ar) ppm. 13C{1H} NMR (100.6 MHz, CDCl3, 25 ºC): δ = 201.5,  161.1, 138.0, 137.3, 133.2, 131.8, 129.1, 128.3, 127.7, 118.8, 118.1, 20.4 ppm.  IR (neat): v = 3057, 2914, 1627, 1601, 1484, 1445, 1334, 1292, 1246, 1226, 959, 824, 787, 756, 733, 700, 658cm-1. ESI-MS (MeOH) m/z = 213.2 [3j+H]+. Elemental Analysis: Calc’d for C14H12O2 : C 79.23%, H 5.70%; Found: C 78.99%, H 5.72%.   Synthesis of (2-hydroxy-5-methyl-3-nitrophenyl)(phenyl)methanone (3k): To a solution of  (2-hydroxy-5-methylphenyl)(phenyl)methanone 4k (1 eq., 2.55 g, 12.0 mmol) in glacial acetic acid (50 mL) at 0 ºC were slowly added p-toluenesulfonic acid (1.2 eq., 2.74 g, 14.4 mmol) and potassium nitrate (1.1 eq., 1.35 g, 13.2 mmol). After stirring at RT for 12 h, water (100 mL) was added and the aqueous phase was extracted with CH2Cl2 (3 x 50 mL). The combined organic phases were dried over Na2SO4, filtered, and then concentrated under reduced pressure. Silica gel flash column chromatography (eluent: CH2Cl2/hexanes, 1/1) of the residue yielded the compound as a yellow oil that slowly solidified to give a yellow solid (2.45 g, 9.5 mmol, 79% yield). 155  Data for (2-hydroxy-5-methyl-3-nitrophenyl)(phenyl)methanone (3k): 1H NMR (400 MHz, CDCl3, 25 ºC): δ=  11.36 (s, 1H, OH), 8.05 (dd, J = 2.4 Hz, 0.8Hz, 1H, Ar), 7.76-7.80 (m, 2H, Ar), 7.62 (tt, J = 7.6 Hz, 1.6 Hz, 1H, Ar), 7.59 (d, J = 2.4 Hz, 1H, Ar), 7.47-7.53 (m, 2H, Ar), 2.38 (s, 3H, CH3) ppm. 13C{1H} NMR (100.6 MHz, CDCl3, 25 ºC): δ = 195.9, 152.0, 138.4, 136.7, 134.9, 133.4, 129.5, 129.3, 128.5, 128.4, 127.7, 20.2 ppm. IR (neat): v = 3266, 3073, 1661, 1594, 1534, 1453, 1408, 1307, 1281, 1246, 1215, 1171, 1125, 991, 934, 812, 767, 705, 656 cm-1. ESI-MS (MeOH) m/z = 258.2 [3k+H]+. Elemental Analysis: Calc’d for C14H11NO4 : C 65.37%, H 4.31%, N 5.45%; Found: C 65.07%, H 4.32%, N 5.53%.   Synthesis of (3-amino-2-hydroxy-5-methylphenyl)(phenyl)methanone (2k): This compound was prepared by the general procedure starting with (2-hydroxy-5-methyl-3-nitrophenyl)(phenyl)methanone 3k  (1.94 mmol, 500 mg).  Data for (3-amino-2-hydroxy-5-methylphenyl)(phenyl)methanone (2k): 1H NMR (400 MHz, CDCl3, 25 ºC): δ = 12.03 (s, 1H, OH), 7.66-7.71 (m, 2H, Ar), 7.55-7.62 (m, 1H, Ar), 7.46-7.54 (m, 2H, Ar), 6.75-6.80 (m, 2H, Ar), 2.19 (s, 3H, CH3) ppm. ESI-MS (MeOH) m/z = 228.4 [2k+H]+.  Synthesis of (2-hydroxy-5-(2,4,4-trimethylpentan-2-yl)phenyl)(phenyl)methanone (4l): To a solution of 4-(1,1,3,3-tetramethylbutyl)phenol (1.3 eq. 15.804 g, 76.60 mmol) in toluene (180 mL) at room temperature were slowly added benzaldehyde (1eq., 6 .0 mL, 58.89 mmol),  tripotassium phosphate (2.2 eq., 27.506 g, 129.58 mmol), copper(II) chloride (0.05 eq., 397.6 mg, 2.96 mmol) and triphenylphosphine (0.075 eq., 1.158 g, 4.42 mmol). The mixture was stirred at 110 oC for 18 h before cooling. Water (100 mL) was added and the aqueous phase was extracted with CH2Cl2 (3 156  x 50 mL). The combined organic phases were dried over Na2SO4, filtered, and then concentrated under reduced pressure.  Silica gel flash column chromatography (eluent: CH2Cl2/hexanes, 1/1) of the residue yielded the compound as a yellow oil (4.53g, 14.58 mmol, 25%).   Data for (2-hydroxy-5-(2,4,4-trimethylpentan-2-yl)phenyl)(phenyl)methanone (4l): 1H NMR (400 MHz, CDCl3, 25 ºC): δ = 11.81 (s, 1H, OH), 7.67-7.72 (m, 2H, Ar), 7.59-7.65 (m, 1H, Ar), 7.49-7.59 (m, 2H, Ar), 7.02 (d, J = 12.0 Hz, 1H, Ar), 1.67 (s, 2H, CH3), 1.29 (s, 6H, CH3), 0.74(s, 9H, CH3) ppm. 13C{1H} NMR (100.6 MHz, CDCl3, 25 ºC): δ = 202.3, 161.5, 141.0, 138.8, 135.4, 132.5, 131.2, 129.9, 128.9, 119.0, 118.1, 57.1, 38.5, 32.8, 32.2, 31.8 ppm. ESI-MS (MeOH) m/z = 311.3 [4l+H]+. Elemental Analysis: Calc’d for C21H26O2 : C 81.25%, H 8.44%; Found: C 81.20%, H 8.39%.   Synthesis of (2-hydroxy-3-nitro-5-(2,4,4-trimethylpentan-2-yl)phenyl)(phenyl)methanone (3l): To a solution of  (2-hydroxy-5-(2,4,4-trimethylpentan-2-yl)phenyl)(phenyl)methanone 4l (1 eq., 7.41 g, 23.9 mmol) in glacial acetic acid (150 mL) at 0 ºC were slowly added p-toluenesulfonic acid (1.2 eq., 5.45 g, 28.7 mmol) and potassium nitrate (1.1 eq., 2.68 g, 26.3 mmol). After stirring at RT for 12 h, water (150 mL) was added and the aqueous phase was extracted with CH2Cl2 (3 x 100 mL). The combined organic phases were dried over Na2SO4, filtered, and then concentrated under reduced pressure. Silica gel flash column chromatography (eluent: diethyl ether/hexanes, 1/12) of the residue yielded the compound as a yellow oil that slowly solidified to give a yellow solid (6.02 g, 16.95 mmol, 71% yield).  157  Data for (2-hydroxy-3-nitro-5-(2,4,4-trimethylpentan-2-yl)phenyl)(phenyl)methanone (3l): 1H NMR (400 MHz, CDCl3, 25 ºC): δ = 11.33 (s, 1H, OH), 8.25 (d, J = 2.8 Hz, 1H), 7.77-7.81 (m, 3H, Ar), 7.64 (tt, J = 7.6 Hz, 1.6Hz, 1H, Ar), 7.49-7.54 (m, 2H, Ar), 1.74 (s, 2H, CH3), 1.39 (s, 6H, CH3), 0.78(s, 9H, CH3) ppm. 13C{1H} NMR (100.6 MHz, CDCl3, 25 ºC): δ = 195.9, 151.8, 142.2, 136.9, 135.9, 134.7, 133.5, 129.6, 128.6, 127.6, 125.6, 56.4, 38.4, 32.4, 31.9, 31.2 ppm. IR (neat): v = 3191, 2952, 2872, 1683, 1597, 1533, 1461, 1448, 1407, 1314, 1285, 1251, 1229, 1174, 982, 777, 734, 710, 658 cm-1. ESI-MS (MeOH) m/z = 356.3 [3l+H]+. Elemental Analysis: Calc’d for C21H25NO4 : C 70.96%, H 7.09%, N 3.94%; Found: C 70.79%, H 7.19%, N 3.90%.   Synthesis of (3-amino-2-hydroxy-5-(2,4,4-trimethylpentan-2-yl)phenyl)(phenyl)methanone (2l): This compound was prepared by the general procedure starting with (2-hydroxy-3-nitro-5-(2,4,4-trimethylpentan-2-yl)phenyl)(phenyl)methanone 3l  (0.84 mmol, 300 mg).  Data for (3-amino-2-hydroxy-5-(2,4,4-trimethylpentan-2-yl)phenyl)(phenyl)methanone (2l): 1H NMR (400 MHz, CDCl3, 25 ºC): δ = 11.96 (s, 1H, OH), 7.67-7.72 (m, 2H, Ar), 7.60 (tt, J = 7.6 Hz, 1.6 Hz, 1H, Ar), 7.47-7.53 (m, 2H, Ar), 7.01 (d, J = 2.4 Hz, 1H, Ar), 6.96 (d, J = 2.4 Hz, 1H, Ar), 3.89 (br, 1H, Ar), 1.64 (s, 2H, CH2), 1.25 (s, 6H, CH3), 0.76 (s, 9H, CH3) ppm. ESI-MS (MeOH) m/z = 326.3 [2l+H]+.  Synthesis of campestarene 1l: To a solution of (3-amino-2-hydroxy-5-(2,4,4-trimethylpentan-2-yl)phenyl)(phenyl)methanone 2l (1 eq., 260 mg, 0.84 mmol) in 1-butanol (50 mL) was added p-toluenesulfonic acid (0.2 eq., 30 mg, 0.17 mmol). The reaction mixture was stirred at 140 oC for 5 weeks before cooling. The reaction mixture was concentrated under reduced pressure. The residue 158  was dissolved in CH2Cl2 (20 mL). The organic phase was washed with distilled water (3 x 20 mL), dried with Na2SO4. Then the solvent was removed by rotary evaporation. Flash column chromatography using neutral alumina (eluent: CH2Cl2/ethyl acetate) yielded the macrocycle as a dark brown solid.  Data for campestarene 1l: 1H NMR (400 MHz, CDCl3, 25 ºC): δ = 16.64 (s, 1H, OH), 7.36-7.52 (m, 5H, Ar), 6.80 (d, J = 1.6 Hz, 1H, Ar), 6.54 (d, J = 1.6 Hz, 1H, Ar), 1.21 (s, 2H, CH2), 0.83 (s, 6H, CH3), 0.53 (s, 9H, CH3) ppm. ESI-MS (MeOH) m/z = 1537.9 [1l+H]+, 1559.9 [1l+Na]+.    159  Chapter 4: Self-Assembly of Extended Head-to-Tail Triangular Pt3 Macrocycles into Nanotubes†  4.1 Introduction  Supramolecular chemistry, where intermolecular interactions are used to organize structures, has emerged as a powerful paradigm in the synthesis of complex substances.152 Taking advantage of electrostatic interactions, hydrogen bonding, van der Waals forces, donor-acceptor interactions and metal-ligand bonding, researchers have constructed fascinating structures. Disc-shaped molecules comprised primarily of conjugated systems arre often coded for the assembly of one-dimensional fibers or nanotubes that have potential applications in solar cells, 153  biological sensors,154 supramolecular polymers,155 and nanowires in electronic devices.156  The combination of electron-rich pyridyl donors and square-planar Pt(II) has been widely used in metallomacrocycle chemistry.157 Despite various fascinating structures emerging from this strategy, this method has a drawback--most of the resulting metallomacrocycles are highly charged.158 The high charge might prevent macrocycles from self-association, which is crucial for the formation of nanotubes.                                                    † A version of this chapter is under preparation for publication: Chen, Z.; MacLachlan, M. J. “Self-Assembly of Extended Head-to-Tail Triangular Pt3 Macrocycles into Nanotubes” 160  4.1.1 Pt4 Head-to-Tail Schiff-Base Macrocycles Neutral Schiff-base platinum(II) complexes with trans-N2O2 donors have been previously reported.159 And it has been demonstrated that similar complexes are sterically unencumbered for one-dimensional assembly.160 In 2010, our group reported a new family of platinum-containing macrocycles that was constructed from a head-to-tail assembly.161 In particular, a monomeric building block with an unsaturated metal center (a Lewis acid) and a Lewis base (a pyridyl group) at the other end of the molecule was designed. Scheme 4.1 shows the structure of the Pt4 macrocycles that were formed either from preforming the platinum salicylaldehyde complex, or by an in situ approach where the ligand was assembled in the presence of the metal salt. In each case, only the Pt4 macrocycle was observed by mass spectrometry, with no evidence of forming either smaller or larger rings.  Scheme 4.1  Synthesis of Pt4 head-to-tail macrocycle.   161  4.1.2 Self-Assembly of Pt4 Macrocycles into Nanotubes These macrocycles could self-assemble into columnar structures by - stacking. With short alkyl chains, the macrocycles appeared to form extended stacks and with branched alkyl substituents, it was possible to observe liquid crystallinity. Most interestingly, by using large substituents, the aggregates of Pt4 macrocycles could be isolated in the solid state. The aggregation number could be four or six depending on the size of the substituent groups. 162  From the crystallography of the Pt4 macrocycles, the Pt4 macrocycles had pores large enough to host ethanol molecules, but not larger guests. In addition, the study showed that Pt4 macrocycles could only stack up to hexamers. Longer stacks were not observed in the solid state.  Sizes of macrocycles can be tuned by using organic building blocks with similar shape but different lengths.163 Thus, the organic linker was extended in order to make larger Pt4 macrocycles (Figure 4.1), so that they would have better guest-accessible channels for exploring their supramolecular chemistry. We believed that the geometry of the ligand as shown in Figure 4.1 was robust, and that extending the distance between the platinum salicylaldehyde groups while maintaining the geometry of the structure would lead to larger Pt4 macrocycles (as discussed in Section 1.5.1). Insertion of ethynyl groups between two aromatic rings should extend the conjugation system while keep the framework’s rigidity. In addition, since previous studies showed that larger substituent groups had a negative impact on stacking, extended macrocycles might demonstrate a larger stacking number as a result of reduced steric effect. This chapter focuses on the synthesis and characterization of extended Pt3 head-to-tail Schiff-base macrocycles. Due to issues with solubility, only one of the extended Pt3 macrocycles was chosen for self-assembly studies. Insight into its self-assembly behavior is provided.  162   Figure 4.1  Structures of the target extended Pt4 head-to-tail macrocycles 5a’ and 5b’.  4.2 Results and Discussion  4.2.1 Synthesis and Characterization In order to make the extended Pt(II)-pyridyl type Schiff-base macrocycle, we first synthesized the amine building blocks 11 for both extended macrocycles 5a’ and 5b’. The synthesis route to amine 11 is shown in Scheme 4.2.  Bulky substituent groups R were synthesized through a Grignard reaction. Then this R group was introduced onto a phenol ring. Nitration was conducted by reacting phenol compound 13 with PTSAH2O and KNO3 in acetic acid. The resulting nitro compound 12 could be isolated as a yellow solid.  A reduction reaction was required to be performed before imine synthesis. Hydrazine hydrate was chosen as reducing agent to reduce 12 into amine 11 in the presence of palladium on carbon. 163  Since amine 11 oxidized slowly in the air, it was immediately used for imine synthesis. The bulky substituents R were selected to inhibit aggregation of the macrocycles and improve their solubility so that they could be studied in solution.  Scheme 4.2  Synthesis of amine building block 11 for extended Pt-pyridyl type macrocycles.  The extended pyridyl-salicylaldehyde compounds were constructed by Sonogashira coupling of acetylene derivatives to aryl halides. For both pyridyl-salicylaldehyde compounds, there was more than one possible synthetic route, provided that multiple steps of the Sonogashira coupling reactions could be conducted in different sequences. In both cases, more than one route was tested (Scheme 4.3 and Scheme 4.4).  For example, to prepare pyridyl-salicylaldehyde 7a, the sequence of introducing acetylene group to pyridyl compound or salicylaldehyde compound might lead to different yield. In each case, using either bromo- or iodo-substituted compounds might produce different results as well. The comparison of 4 different synthetic routes to 7a is listed in Scheme 4.3. Using iodo-substituted 164  compounds offered higher yield. Also, the coupling reaction between 3-ethynylpyridine 9a and aryl halides demonstrated much better yield. As a result, synthetic route (b) was chosen as the optimized synthetic route over the other three.   Scheme 4.3  Comparison of different synthetic routes to pyridyl-salicylaldehyde 7a.  In the case of preparing 7b, another issue arose. Since iodo-substituted compounds usually show better reactivity than bromo-substituted compounds in Sonogashira coupling reactions,164 reacting 3-ethynylpyridine 9a with 1-bromo-4-iodobenzene at 1:1 ratio gave very good yield and selectivity. No double coupling product was found. However, the resulting product demonstrated very poor reactivity in the following coupling reaction (Scheme 4.4 (a)).  Another synthetic route involved using a more reactive symmetric compound--1,4-diiodobenzene. To reduce the overreaction leading to symmetric by-product, excess 1,4-diiodobenzene (4 eq.) was used in this reaction. Although reacting 3-ethynylpyridine with excess 1,4-diiodobenzene still resulted in lower yield due to overreaction, the following coupling step 165  showed higher yield, thus resulting in a higher overall yield in comparison (Scheme 4.4 (b)). Moreover, excess 1,4-diiodobenzene could be recovered from the reaction mixture. The optimized synthetic routes for both pyridyl-salicylaldehyde compounds were chosen based on the comparison of the results from different synthetic routes. After the synthesis of both amine 11 and pyridyl-salicylaldehyde 7a-b, the imine synthesis was conducted in ethanol for 30 minutes yielding imines 6a-b as precipitates. The optimized synthesis of both imine 7a and 7b are shown in Scheme 4.5 and Scheme 4.6, respectively.  Scheme 4.4  Comparison of two different synthetic routes to pyridyl-salicylaldehyde 7b.   166  Scheme 4.5  Synthetic route of imine 6a.   Scheme 4.6  Synthetic route of imine 6b.  These new ligands 6a-b have similar overall geometry to the ligands used to prepare the Pt4 macrocycles (Scheme 4.1) in terms of the orientation of the pyridyl N with respect to the 167  salicylaldehyde. Consequently, it was expected that reaction of these new ligands with Pt(II) salts would afford Pt4 macrocycles 5a’ and 5b’ (Figure 4.1). Reaction of 6a with K2PtCl4 led to a yellow powder. Matrix-assisted laser deposition ionization time-of-flight mass spectrometry (MALDI-TOF MS) of the product revealed that the product was not the intended Pt4 macrocycle 5a’ (MW = 3670.4 g/mol), but rather a Pt3 macrocycle 5a (MW = 2754.1 g/mol) (Figure 4.2). As well, the MALDI-TOF MS showed aggregates of the Pt3 macrocycle, suggesting that the macrocycles have a strong tendency to stack (as previously observed with the Pt4 macrocycles).  More importantly, MALDI-TOF MS showed larger stacks than previously observed for Pt4 macrocycles. This suggests that the stacking of the Pt3 macrocycle 5a is not limited to hexamers or tetramers. Long range stacking can be achieved with this new macrocycle.  Figure 4.2  MALDI-TOF MS of macrocycle 5a.  The peak at m/z = 2755.7 Da corresponded to the protonated Pt3 macrocycle, [5a+H]+. Matrix: trans-2-[3-(4-t-butyl-phenyl)-2-methyl-2-propenylidene] malononitrile (DCTB). 168  Another Pt3 macrocycle 5b was prepared by a similar approach from 6b, and MALDI-TOF MS verified that it was also the Pt3 macrocycle that predominated in the product. However, macrocycles with other geometries could also be observed in MALDI-TOF mass spectrum as minor products (Figure 4.3). Moreover, no strong stacking could be observed in this case. Notably, macrocycle 5b showed much weaker signals in MALDI-TOF MS compared to macrocycle 5a. Because macrocycle 5b demonstrated very poor solubility in organic solvents, macrocycle 5a was chosen for the following self-assembly studies. The actual chemical structure of macrocycles 5a and 5b are shown in Figure 4.4.   Figure 4.3  MALDI-TOF MS of macrocycle 5b.  The peak at m/z = 3055.9 Da corresponds to the protonated Pt3 macrocycle, [5b+H]+. Matrix: trans-2-[3-(4-t-butyl-phenyl)-2-methyl-2-propenylidene] malononitrile (DCTB).  169   Figure 4.4  Chemical structures of Pt3 macrocycle (a) 5a and (b) 5b.  It came as a surprise that the Pt3 rings were exclusively obtained with the expanded ligands 6a and 6b. The change of geometry can be explained by two factors. First, the insertion of acetylene groups introduced flexibility to the organic linkers. A study of molecular systems of 1,2-diphenylacetylene showed that, in the solid state, the C-C triple bonds in 1,2-diphenylacetylene can be bent by up to 6°. 165 The additional flexibility could contribute to formation of macrocycles with unexpected geometry. 166  Second, in our previous organic linker containing pyridyl-substituted salicylaldehyde, the ground-state conformation of the ligand had a torsion between the two aromatic rings. This non-planarity may favor the square-shaped macrocycle, whereas in 5a and 5b, the ligands can adopt a conformation where the pyridyl and salicylaldimine components are coplanar. Thus the flat conformation of the macrocycle might favor the triangular macrocycles.  A computational study was conducted to investigate the macrocycle conformation. To simplify the system, a Pt3 macrocycle with R = H was used as an example (Figure 4.5). In the optimized conformation, all three repeating units showed similar bond angles and twisting angles. 170  Overall, the optimized conformation of macrocycle was almost planar. It bent to one side very slightly to take a shallow bowl shape conformation.  All C-C triple bonds in organic linkers showed a bend angle of about 6°. The extra flexibility introduced by acetylene groups contributed to the formation of the 3-fold symmetric macrocycle. In addition, in the simulated structure, two aromatic rings belonging to the same organic linker almost stayed in the same plane to maximize conjugation. Actually, all aromatic rings from the same Pt3 macrocycle were almost coplanar. The flattened conformation of three organic linkers with flexible acetylene groups allowed the triangular macrocycle to form.   Figure 4.5  (a) Chemical structure of Pt3 macrocycle with R = H; (b) top view and (c) side view of DFT optimized geometry of macrocycle with R = H, computed at the 6-31g(d) level of theory.   4.2.2 Self-Assembly of Pt3 Macrocycles in Solid State The MALDI-TOF MS data for macrocycle 5a suggested that it aggregates strongly in the solid state. More experiments were performed to investigate its self-assembly. Unfortunately, attempts to grow a single crystal of macrocycle 5a were unsuccessful. Powder X-ray diffraction 171  (PXRD) of macrocycle 5a showed no sharp diffraction peaks beyond 5°-2θ, indicating that the material was mostly amorphous (Figure 4.6). However, there was a peak evident at 3.75°-2θ (23.5 Å d-spacing) that indicated higher order organization within the substance. We felt that this could arise from columnar stacking of the macrocycles in the solid-state. Since the typical pattern of columnar stacks of disc-like molecules are usually a hexagonal lattice, 167  the diameter of macrocycles can be calculated from the 23.5 Å d-spacing as 2.71 nm, which is consistent with estimated diameter of macrocycle 5a.  Figure 4.6  Powder X-ray diffraction pattern of macrocycle 5a.  To corroborate this interpretation, transmission electron microscopy (TEM) studies of macrocycle 5a were undertaken. Macrocycle 5a was suspended in ethanol, and the suspension was then transferred onto a TEM copper grid and placed in the oven at 333 K to dry. We were pleased to observe stripes with an average separation of about 2.6 nm clearly evident in the TEM images (Figure 4.7). This value is close to the diameter estimated of a stacked macrocycle nanotube and 172  previous PXRD result. Thus, we believe that the TEM images show that the macrocycles have stacked one on top of the other into a nanotubular structure, which is organized into a hexagonal lattice. This organization is reminiscent of the Pt4 macrocycles, which also stacked into columns in the solid state. The proposed model of the solid-state organization of 5a is shown in the inset of Figure 4.7.   Figure 4.7  TEM images of macrocycle 5a. The repeating distance between adjacent stripes is approximately 2.6 nm, and the inset shows the proposed organization of 5a in the solid state.  Although TEM images showed organized stripes in solid state, the actual stacking structure might be disordered (Figure 4.8). When a head-to-tail macrocycle stacks on top of another one, the second macrocycle could stack with either the same or opposite orientation as the first molecule. And since each time when another macrocycle stacked on an existing self-assembled nanotube, it could take on two different orientations, this orientation issue could cause exponentially more disorganization within a stack of macrocycles. This might be the reason for the lack of sharp peaks in the PXRD pattern of macrocycle 5a.  173   Figure 4.8  (a) Top view and (c) side view of 2 Pt3 macrocycles stacking with the same orientation; (b) Top view and (d) side view of 2 Pt3 macrocycles stacking with opposite orientation. For clarity, hydrogen atoms and R groups were omitted. All macrocycles have the same chemical structure. Blue and red colors are used to distinguish macrocycles with different orientations.   4.2.3 Dilution Experiments Although it was shown that Pt3 macrocycle 5a self-assembles in solid state, its behavior in solution remained unknown. I therefore conducted a study of its behavior in solution by NMR spectroscopy. Due to the macrocycle’s poor solubility in most organic solvents, NMR experiments were only conducted in CDCl3. The 1H NMR spectrum of macrocycle 5a in CDCl3 showed a set of broad peaks (Figure 4.9). Although the overlapping of several broad aromatic peaks at about 7.3~7.4 ppm made it hard to assign peaks, the 1H NMR spectrum unambiguously suggested a highly symmetrical structure associated with macrocycle 5a. In addition, the broadening of proton peaks in 1H NMR studies of macrocycles is often a good indication of aggregation,168 since large 174  aggregates have shorter T2 relaxation times and macrocycles exchanging between different aggregation environments on the NMR timescale are both effects that could lead to broadening.   Figure 4.9  Aromatic region of 1H NMR spectrum of macrocycle 5a in CDCl3 at room temperature (400MHz, c = 7.0 ×10-3 mol/L).  Stacking of aromatic compounds usually results in upfield shifts in the nuclear magnetic resonance spectra.169 To further probe the self-association behavior of macrocycle 5a, a dilution experiment in CDCl3 was carried out (Figure 4.10). Different from what was expected, there was no chemical shift change observed in the dilution experiment. Instead, a new set of peaks appeared when the concentration of 5a decreased. 175   Figure 4.10  1H NMR spectra of macrocycle 5a in CDCl3 at different concentrations (room temperature, 400 MHz).  At higher concentration (7.3 × 10-3 mol/L), the macrocycle only shows one set of major peaks that correspond reasonably well with those expected for macrocycle 5a. After being diluted by 16 times (to 4.5 × 10-4 mol/L), a new set of sharper peaks was observed, together with the original peaks. As the solution was further diluted, this new set of peaks grew at the expense of the original set of broad peaks. At very low concentrations (1.1 × 10-4 mol/L), the original broad signals disappeared and only the new, relatively sharp peaks were present. This dilution experiment not only confirmed the aggregation of macrocycle 5a in CDCl3 at room temperature, but also suggested that the aggregation process was not very fast compared to the NMR time scale since both aggregates and monomer could be simultaneously resolved in the spectra. 176   Figure 4.11  (a) UV-vis spectra of macrocycle 5a in CHCl3 at different concentration (room temperature); (b) linear fitting for absorption of different 5a solution at 328 nm (black), 442 nm (red) and 465 nm (blue). The R2 value for 328 nm, 442 nm and 465 nm data are 0.99992, 0.99992, 0.99993 respectively.  Another dilution experiment was conducted by UV-vis spectroscopy (Figure 4.11 (a)). Compared to the previous dilution experiment performed by NMR, this second experiment was carried out at a lower concentration due to the intense absorption of the molecule in UV-vis. When the concentration of solution decreased, the absorption pattern stayed the same. The absorption values at 328 nm, 442 nm and 465 nm were collected for linear data fitting (Figure 4.11 (b)). They all showed very good linearity. This linearity suggests that the self-association of macrocycle 5a was negligible at such low concentrations. And at high concentration (c = ~7.0 × 10-3 mol/L), the predominant form of macrocycle 5a in CHCl3 was the aggregate. As a Pt-containing Schiff-base complex, macrocycle 5a has the potential to show fluorescence.170 Considering no fluorescence could be observed from a solution of macrocycle 5a in chloroform at any concentration, a solution of 5a in chloroform was purged with argon gas for 177  30 mins to remove dissolved oxygen, which is well known to be an efficient quencher of the singlet and triplet states of organic compounds.171 The fluorescence spectrum of Pt3 macrocycle 5a is shown in Figure 4.12.   Figure 4.12  (a) The absorption spectrum of macrocycle 5a in chloroform at room temperature; (b) photos of macrocycle 5a solution under visible/UV light; (c) excitation spectrum of macrocycle 5a solution, emission light wavelength is 635 nm; (d) emission spectrum of macrocycle 5a solution, excitation light wavelength is 450 nm. c = 2.5 × 10-6 mol/L.  The excitation spectrum is similar to the absorption spectrum. The fluorescence spectra clearly showed that macrocycle 5a emitted light of 635 nm, which results in the red colored 178  luminescence in Figure 4.12(b). Unfortunately, previous dilution experiments showed no aggregation at such low concentrations. So the fluorescence spectra represented the behavior of macrocycle itself, not fluorescence of an aggregate. Due to its high absorption at high concentrations, fluorescence could not be applied to study much more concentrated solutions. As a result, we could not quantify aggregation by studying aggregation-induced quenching.172  4.2.4 Variable Temperature Experiments To further study this aggregation property, we performed a variable-temperature NMR study of macrocycle 5a (Figure 4.13) in CDCl3. Usually, stacking is more favored at lower temperatures. Some results have shown that lowering the temperature might lead to upfield shifts of aromatic signals caused by stronger stacking. And since larger aggregates usually have shorter relaxation time T2, the broadening of peaks in 1H NMR spectra upon lowering the temperature was expected.  Different from the prediction, the decrease of temperature sharpened the signals rather than broadening them. Moreover, although large chemical shift changes of many peaks were observed, those shifts did not follow the same trend. While some peaks were shifting upfield by up to 1.5 ppm, some other aromatic peaks showed downfield shifts up to 1.0 ppm. Although those shifts are not consistent with the usually observed upfield shifts in stacking of aromatic macrocycles,173 macrocycles showing both upfield and downfield shifts at the same time due to aggregation have been reported.16,17 179   Figure 4.13  Variable temperature 1H NMR spectra of macrocycle 5a in CDCl3 (400 MHz). Only the aromatic region is shown.  To help the understanding of the behavior of Pt3 macrocycles in solution and to investigate these unusual chemical shifts, a model compound 5c was synthesized. Scheme 4.6 shows the synthetic route to model compound 5c. Compound 5c was designed to have a similar structure with macrocycle 5a. To prevent cyclization, the pyridyl group was replaced by a phenyl ring.  When model compound 5c was dissolved in CDCl3, it showed a set of sharp peaks in the 1H NMR spectrum. Dilution experiments with 5c were conducted and are shown in Figure 4.14. Different from 5a, which demonstrated a conversion between 2 sets of peaks during dilution, 180  model compound 5c only provided one set of peaks in the 1H NMR spectrum at different concentrations. While many peaks stayed untouched during the entire dilution experiment, only very minimal chemical shift changes could be observed for some aromatic peaks. As well, the chemical shifts of these aromatic peaks only changed by 0.01-0.03 ppm when the sample was diluted form 1.3 × 10-2 mol/L to 2.0 × 10-3 mol/L. Considering that the self-association of shape-persistent macrocycles usually causes a shift of about 0.3-0.5 ppm in this concentration range,174 it is feasible to think that model compound 5c shows no self-association in this concentration range. Based on their different behaviors in dilution experiments, macrocycle 5a demonstrated self-association behavior which could not be achieved with model compound 5c.  Scheme 4.6  Synthesis of model compound 5c.                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                      181   Figure 4.14  1H NMR spectra of model compound 5c in CDCl3 at different concentrations (room temperature, 400 MHz).    Figure 4.15  Variable temperature 1H NMR spectra of model compound 5c in CDCl3 (400 MHz). Only the aromatic region is shown. 182  Following the dilution experiment of model compound 5c, variable-temperature NMR experiments with compound 5c (Figure 4.15) were performed. As the temperature was lowered, most peaks remained in their same positions. Only the peak situated around 9.2 ppm showed a small shift of about 0.08 ppm. With the help of 2D NMR experiments (COSY, 1H-13C HSQC and 1H-13C HMBC), this peak is assigned to the ortho-H atoms of coordinated pyridine (Figure 4.16). This small shift might be due to the rotation of pyridine ligand.   Figure 4.16  1H-13C HSQC spectrum of 5c in CDCl3. Insert is the structure of 5c. The peak at 9.2 ppm was assigned to a pyridyl proton (room temperature, 400 MHz).  Based on computational studies of 5c, pyridine prefers to be coplanar with the benzene rings. When the temperature was lower, the population of coplanar pyridine may increase, causing the chemical shift change. While macrocycle 5a showed dramatic shifts in variable-temperature NMR 183  experiments, the chemical shift change for model compound 5c was negligible. Compared with results from dilution experiments for both compounds, we reason this unusual shift in variable-temperature NMR involved the self-assembly process of 5a.   4.2.5 Diffusion Ordered Spectroscopy (DOSY) To further investigate the self-assembly behavior of macrocycle 5a in chloroform, a dynamic light scattering (DLS) experiment was conducted to measure the size of macrocycle 5a. Unfortunately, the result was not reproducible. Another experiment performed was to measure the boiling-point elevation of macrocycle solution. A 10 g/L solution of 5a in chloroform was prepared. Then the boiling point of this solution and pure chloroform were measured for comparison. Due to the low solubility and high molecular weight of macrocycle 5a, the theoretical boiling-point elevation was only about 0.008 °C. This experiment was also unsuccessful due to the lack of a precise measurement of the boiling point.  In order to determine the size of aggregates in solution at different temperature, diffusion ordered spectroscopy (DOSY) was conducted. Diffusion NMR experiments are usually used to resolve different compounds spectroscopically in a mixture based on their differing diffusion coefficients, which depends on both the size and shape of the molecules (aggregate). Aggregation of macrocycles can also be studied by DOSY.175 The Stokes-Einstein equation gives the relation between diffusion coefficient D and the radius of the molecule (aggregate):  k: Boltzmann constant; T: the temperature; η is the viscosity of the liquid; rs the (hydrodynamic) radius of the molecule (aggregate). 184  To help estimate the size of macrocycle 5a and its aggregate, model compound 5c was used as reference. Therefore, both macrocycle 5a and model compound 5c were dissolved in CDCl3 to prepare the sample for DOSY experiments. Figure 4.17 and Table 4.1 show the results from DOSY experiments of both macrocycle 5a and model compound 5c at 298 K. The diffusion coefficient D of both 5a and 5c were measured. For both compounds, more than one peak was chosen to measure the diffusion coefficient. Due to overlapping of some aromatic peaks, not all proton signals were used to measure diffusion coefficient D. The reported D values are the average D value of several protons belonging to the same compound.   Figure 4.17  Gaussian fit to diffusion peak intensity at 298K using a non-linear fit.  185  Table 4.1  Diffusion coeffiecient of 5a and 5c measured by DOSY at 298 K. Number Chemical shift (ppm) Peak assignment Diffusion coefficient (m2/s) 1 9.22 model compound 5c 7.84×10-10 2 8.75 macrocycle 5a 4.94×10-10 3 8.51 macrocycle 5a 5.12×10-10 4 8.17 model compound 5c 7.83×10-10 5 6.62 macrocycle 5a 5.22×10-10 6 6.23 macrocycle 5a 5.19×10-10  Based on these results, the radius of macrocycle 5a could be calculated by:  Since the dynamic viscosity of chloroform at 298 K was 0.5425 mPa·s, the hydrodynamic radius 𝑟5𝑎298𝐾 could be calculated directly from this equation as 0.79 nm. Compared to the radius of single macrocycle 5a, this value looks too small. However, the hydrodynamic radius calculated through the Stokes-Einstein equation is more accurate when the molecule has a spherical shape. It is known that applying this equation to a non-spherical molecule usually leads to large errors. Actually chemists have already reported several modifications of the Stokes-Einstein equation that take the molecule's shape into account by adding shape correction factors.176  Without correction, the radius calculated from the equation is usually consistent with the van der Waals radius. For example, the hydrodynamic radius of 9,10-diphenylanthracene measured by DOSY was about 0.42 nm, comparing well with the measured mean van der Waals radius of 0.41 nm. The directly measured dimensions of a flat 9,10-diphenylanthracene conformation is much 186  larger than its hydrodynamic radius. So, the 0.79 nm hydrodynamic radius was a reasonable estimate for macrocycle 5a. However, since macrocycle 5a (aggregate) did not have a spherical shape, this hydrodynamic radius does not help to piece together the aggregation dynamics. Additionally, the DOSY spectra are frequently affected by various sources of errors like temperature, viscosity and concentration effects. So it becomes clear that DOSY result would be more accurate under identical conditions.177 To overcome the complications of these effects and to enable tabulated diffusion coefficients, an internal standard is usually used in DOSY experiments. The reference compounds provide more resilient diffusion coefficients for any changees in the NMR sample.178 Since the previous method was based on an assumption that the viscosity of the solution was close to the viscosity of the neat solvent, it was decided to estimate the size of macrocycle 5a (aggregate) by introducing model compound 5c as an internal reference in order to eliminate the influence from viscosity changes. The relative diffusivity is defined as the ratio of diffusion coefficient of analyte to that of reference. This approach reduces the impact of viscosity and temperature and provides more robust data.179  At 298 K, model compound 5c, which was used as an internal reference, had a diffusion coefficient 𝐷5𝑐298𝐾  of 7.84×10-10 m2/s and the macrocycle 5a had a diffusion efficient 𝐷5𝑎298𝐾  of 5.12×10-10 m2/s. So the ratio between the radiuses of two compounds 𝑟5𝑎 𝑟5𝑐⁄  at 298 K could be calculated as 1.53. Different from previous calculation, this value was calculated without involving actual solvent viscosity. 187   Figure 4.18  Gaussian fit to diffusion peak intensity at 250 K using a non-linear fit.  Table 4.2  Diffusion coeffiecient of 5a and 5c measured by DOSY at 250K. Number Chemical shift (ppm) Peak assignment Diffusion coefficient (m2/s) 1 9.14 model compound 5c 4.32×10-10 2 8.84 macrocycle 5a 3.29×10-10 3 8.61 macrocycle 5a 3.14×10-10 4 8.19 model compound 5c 4.25×10-10 5 7.99 model compound 5c 4.28×10-10 6 6.23 macrocycle 5a 3.55×10-10 7 5.60 macrocycle 5a 3.63×10-10  188  The same sample of 5a and 5c was cooled to 250 K in order to measure the diffusion coefficients of both compounds by DOSY NMR at lower temperature (Figure 4.18 and Table 4.2). At 250K, 𝐷5𝑐250𝐾  was 4.28×10-10 m2/s, and 𝐷5𝑎250𝐾  was 3.40×10-10 m2/s, which yielded the ratio between the radii of the two compounds 𝑟5𝑎 𝑟5𝑐⁄  at 250 K of 1.26.  Since dilution and variable temperature experiments revealed that there was no aggregation of 5c at that concentration, it is reasonable to assume that the size of compound 5c did not change much at different temperatures. Compared to the ratio 𝑟5𝑎 𝑟5𝑐⁄  at 298K of 1.53, the measured size of macrocycle 5a decreased as the temperature was lowered. The temperature decrease led to a significant and consistent decrease of the D values of the macrocycles probed with all their four signals at the downfield area.180 These observations support the decreased size of macrocycle aggregates at lower temperature. Similar results could be drawn by calculating radii at different temperatures without considering viscosity error.  Moreover, since the 1H NMR spectrum of macrocycle 5a at low temperature was still relatively broad and different from its 1H NMR spectrum at very low concentrations, this implies that 5a remains stacked even at low temperatures. To support this, NOESY experiments were conducted at both room temperature and low temperature.  4.2.6 Nuclear Overhauser Effect NMR Spectroscopy (NOESY) Another experiment conducted to prove the aggregation of macrocycle 5a was nuclear overhauser effect NMR spectroscopy (NOESY). This technique is used to identify through-space relationships between protons that are close to each other in space but not directly bonded. Quantitative measurements of the intensities of the NOE cross peaks allow the determination of internuclear distances, making it possible to study the stereochemistry in 189  solutions so as to construct a three-dimensional conformation for proteins.181 Recently, it has been used for studying macrocycles182 or crescents.183 NOESY requires a dipolar interaction of spins (the Nuclear Overhauser Effect, NOE) for correlation of protons. Since the dipolar interaction of spins decreases dramatically as the separation increases, the NOE signals are usually only observed within a distance of ~5 Å, with dipolar interactions over longer distances demonstrating weaker signals.  Figure 4.19  1H-13C HSQC and 1H-13C HMBC NMR of macrocycle 5a at 240 K. 190  To study the distances between certain protons in macrocycle 5a, the 1H NMR peaks need to be assigned to individual protons. Since several peaks of 5a overlapped at room temperature, which caused difficulties in peak assignment, it was decided to study the macrocycle’s behavior at lower temperatures. With the aid of 1H-13C HSQC and 1H-13C HMBC NMR (Figure 4.19), all peaks were assigned to macrocycle 5a (Figure 4.20) at 240 K.   Figure 4.20  1H NMR spectrum of macrocycle 5a with peak assignment at 240 K (400 MHz). Only aromatic region is shown.  The 2D NOESY NMR spectrum of 5a at 240 K is shown in Figure 4.21. Those signals observed by NOESY NMR suggest that the protons with those coupling signals are close to each other in space. In Figure 4.21, we observed strong coupling signals between proton a and aromatic protons from substituents R. Similarly, proton b also coupled with substituents R through space. Those signals were almost as strong as the coupling signals between h and R. The closest distances between proton a and substituents R on a single molecule of 5a was estimated at about 7~8 Å. Such strong coupling signals are unusual for such long distances. 191   Figure 4.21  NOESY NMR spectrum of macrocycle 5a at 240K.  The intensity of the coupling signal in NOESY NMR increases with diminishing distance between two protons. The quantitative relation is shown in the equation below:  rij: the distance between proton i and proton j; rref : the distance between two protons used as reference; aij : the integral of coupling between proton i and proton j; aref : the integral of coupling between two protons used as reference. By using the previously calculated conformation of the macrocycle (see Figure 4.5), the expected distances between some protons could be measured (Table 4.2). Since the distance between two types of protons have three different values in one macrocycle due to the 3-fold symmetry, only the closest value is listed in the table.  192   Table 4.3  Relative integral and estimated distance data from NOESY NMR study of macrocycle 5a at 240 K. Proton 1 Proton 2 Relative Integral Relative Distance Estimated Closest Distance# / Å Expected Closest Distance* / Å b c 1.00 1.00 2.51 2.51 a b 1.02 1.00 2.51 2.51 h g 1.67 0.92 2.30 2.24 h k 2.40 0.86 2.17 2.05 a g 0.12 1.42 3.57 5.10 a h 0.11 1.44 3.63 7.20 a k 0.21 1.30 3.26 8.58 b g 0.09 1.49 3.73 7.55 b h 0.11 1.44 3.62 8.93$ b k 0.15 1.37 3.43 9.22$ *: expected distance was measured from one single macrocycle with optimized geometry (see Figure 4.5); #: estimated distance was calculated from relative distance with assuming rbc = 2.51 Å; $: the closest distance was measured between two protons from adjacent organic linkers within one single macrocycle.  With protons b and c as reference protons, the distance between the many protons could be estimated from their integral intensities (Table 4.3). For example, the estimated distance between proton h and proton g was about 2.36 Å, which was consistent with the expected value (2.24 Å). Notably, the distance between proton a and proton h was about 3.34 Å, which was very close to 193  the distance between protons a and g (3.35 Å). Besides, the distance between protons a and k (3.21 Å) was much smaller than expected intramolecular value (8.58 Å). Similar unusually strong dipolar interactions such as abg,abh and abk should not be observed within a single macrocycle. Considering that macrocycle 5a has a very rigid framework and was impossible to fold, these surprisingly strong coupling signals are thought to be arise from the stacking of macrocycle 5a. After being stacked on another macrocycle, proton a from one macrocycle could stay close to proton h from another macrocycle, resulting in a much stronger signal (Figure 4.22). Protons a and h from the same macrocycle (labelled with the same color) are too far apart to offer NOESY signals. The observed NOEs are due to the dipolar interaction between protons from different macrocycles (interaction between green protons and light blue protons).  Figure 4.22  Top view of two stacked macrocycle 5a. Only protons a and h from different macrocycles are shown in green and light blue color. The other protons are removed for clarity.  This stacking behavior also explained the couplings between protons a/b and aromatic protons from substituents R, which are far apart in a single macrocycle. In addition, since the 194  distances rag, rah and rak showed similarly high values, it is likely that one macrocycle stacks on top of another macrocycle with proton a located close to proton g, h and k from another macrocycle. Moreover, the similar values of the distances rag, rah and rak are not consistent with a fixed conformation of aggregates, which implies that the relative twisting angle between two stacked adjacent macrocycles varies within a stack.184 Notably, the unexpectedly short distances such as rag, rah are between 3.3-3.8 Å, consistent with typical - stacking distance.185  Figure 4.23  NOESY NMR spectrum of macrocycle 5a at 298K.  With the purpose of comparing the aggregation behavior at room temperature with the behavior at low temperatures, NOESY NMR of macrocycle 5a at room temperature was performed as well. However, due to overlapping peaks at room temperature, only some signals in its 1H NMR spectrum could be assigned; peak overlapping also caused trouble in the assignment of coupling signals between 7.2 ppm and 7.5 ppm in the 2D NOESY NMR spectrum. Hence, less data could 195  be used for distance estimation in this case. Figure 4.23 shows the NOESY NMR spectrum of macrocycle 5a at room temperature. Despite coupling of signals between 7.2 ppm and 7.5 ppm, calculation was conducted to estimate distances between some protons (Table 4.4).  Table 4.4  Relative integral and estimated distance data from NOESY NMR study of macrocycle 5a at 298K. Proton 1 Proton 2 Relative Integral Relative Distance Estimated Closest Distance# / Å Expected Closest Distance* / Å b c 1.00 1.00 2.51 2.51 i j 2.10 0.88 2.22 2.49 b i 0.25 1.26 3.16 5.91$ b j 0.16 1.42 3.55 8.39$ *: expected distance was measured from one single macrocycle with optimized geometry (see Figure 4.5); #: estimated distance was calculated from relative distance with assuming rbc = 2.51 Å; $: the closest distance was measured between two protons from adjacent organic linkers.  Although only limited coupling signals were observed at room temperature, some protons that are far away from each other in the molecule contribute to signals in the NOESY NMR spectrum. For instance, signals abi and abj indicate separations of about 3.5 Å, which suggests the existence of stacking at room temperature. Moreover, although accurate distances between two macrocycles could not be determined by this method, we think two stacking models at different temperatures might have similar separations since the closest NOE signals in both cases were about 3.5 Å. Notably, the closest protons at 240 K are rah and rak, while rbi and rci have the shortest separation distances at 298 K. This difference might be caused by the change of the relative 196  twisting angle while stacking. As the relative position of two stacked macrocycles changed, some protons might become more shielded by pointing towards the aromatic rings of another macrocycle, while others become less shielded by moving away from the aromatic rings of another macrocycle. These changes of the shielding effect contributed to both upfield shifts and downfield shifts of proton signals in the variable temperature NMR spectra of macrocycle 5a.  NOESY experiments showed that Pt3 macrocycle 5a displays aggegation in solution over wide temperature range. Both the larger stack size from DOSY experiments and broadening of 1H NMR signals at higher temperature indicate that macrocyle 5a have larger size at relatively higher temperature. This trend indicates that self-assembly of Pt3 macrocycles is entropy-driven, which is similar to Pt4 macrocycles. Since aggregation of Pt3 macrocycles causes the decrease of particle numbers, this self-assmebly process is believed to be solvent-driven process.  In addition, the assembly of Pt3 macrocycles should be considered as a dynamic process. Several different dynamic modes such as intra-columnar exchange of the macrocycles, intra-columnar rotation of macrocycles, change of distance between macrocycles, macrocycle slipping and macrocycle tilting could happen. As a result, considering the columnar structure as a dynamic mode is reasonable. Stacking behavior studied by using NMR spectroscopy should be considered as an average behavior over NMR time scale.   4.3 Conclusions  Head-to-tail Pt3 Schiff-base macrocycles were synthesized by expanding the organic linkers of Pt4 macrocycles with ethynyl groups. This chapter described the highly selective synthesis and characterization of 3-fold symmetric macrocycles. Expanding organic linkers by introducing 197  acetylene groups led to the changes in both size and geometry of macrocycles. The geometry change from 4-fold to 3-fold reduced the steric interaction between bulky substituent groups. As a result, Pt3 macrocycles demonstrated comparably enhanced stacking. The self-assembly of Pt macrocycles in the solid state was studied with MALDI-TOF, PXRD and TEM. Experiments showed that multiple Pt3 macrocycles stack into nanotubes in solid state.  In addition, UV-vis and NMR spectroscopy were applied to study macrocycles’ self-association in solution. Aggregates of Pt3 macrocycles showed a decrease in size at lower temperatures, indicating that this self-assembly is an enthropy-driven process. The conformation and aggregation of Pt3 macrocycles were analyzed by a combination of computation, UV-Vis experiments, variable-temperature NMR, DOSY and NOESY NMR spectroscopy. The self-assembly of Pt3 macrocycles might provide a convenient and efficient way to form nanotubes.   4.4 Experimental  4.4.1 Materials All reactions were carried out under air unless otherwise stated. Tetrahydrofuran (THF) was distilled from sodium/benzophenone under nitrogen. Diisopropylamine was distilled from NaOH under nitrogen. Acetonitrile, methanol, and triethylamine were purged with nitrogen gas and dried over molecular sieves before use. All reagents were used as received unless otherwise stated.  4.4.2 Equipment 300 MHz 1H and 75.5 MHz 13C{1H} NMR spectra were recorded on a Bruker AV-300 spectrometer. 400 MHz 1H and 100.6 MHz 13C{1H} NMR spectra were recorded on a Bruker AV-198  400 spectrometer. Mass spectra and elemental analyses were obtained at the UBC Microanalytical Services Laboratory. Matrix assisted laser desorption/ionization (MALDI) mass spectra were obtained on a Bruker Biflex IV time-of-flight (TOF) mass spectrometer equipped with a MALDI ion source. Electrospray ionization (ESI) mass spectra were obtained on a Bruker Esquire-LC ion trap mass spectrometer equipped with an electrospray ion source. High-resolution electrospray ionization (HR-ESI) mass spectra were obtained on a Micromass LCT time-of-flight (TOF) mass spectrometer equipped with an electrospray ion source. Gramicidin S, Rifampicin, and Erythromycin were used as the references for HR-ESI. Elemental analyses were obtained on a Carlo Erba Elemental Analyzer EA 1108. Melting points were obtained on a Fisher-John’s melting point apparatus. IR spectra were obtained using a Thermo Scientific Nicolet 6700 FT-IR spectrometer equipped with a diamond Attenuated Total Reflectance. UV-vis spectra were obtained in chloroform on a Varian Cary 5000 UV-vis-near-IR spectrophotometer using a 1 cm quartz cuvette. For TEM studies, macrocycle 5a was suspended in ethanol, and the suspension was then transferred onto a copper TEM grid and placed in the oven at 333 K to dry. TEM was performed on a Hitachi H7600 transmission electron microscope.  4.4.3 Procedure and Experimental Data Compounds 3-ethynylpyridine (9a)186 and 5-iodosalicylaldehyde (8a)187 were synthesized by following literature procedures. Compounds 3-((4-iodophenyl)ethynyl)pyridine (9b)188  and 5-Ethynyl-2-hydroxybenzaldehyde (8b)189 were synthesized according to literature procedures with modifications.  199  Synthesis of 2-nitro-4-(tris(4-(tert-butyl)phenyl) methyl)phenol (12): 4-(Tris(4-(tert-butyl)phenyl)methyl)phenol (1.00 g, 1.98 mmol) and p-toluenesulfonic acid monohydrate (377 mg, 1.98 mmol) were stirred in 40 mL acetic acid. Potassium nitrate (200 mg, 1.98 mmol) was added into the flask slowly while the flask was kept in an ice-water bath. After addition, the ice-water bath was removed and the mixture was stirred at room temperature for 16 h. Water was used to quench the reaction, followed by extraction with DCM. The organic phase was washed with saturated aqueous NaHCO3 and water to remove acid. Then it was dried over anhydrous MgSO4. The solvent was removed under reduced pressure, and the crude product was further purified by column chromatography with DCM/hexanes (40:60). The product had an Rf of approximately 0.7 and was isolated as a yellow solid (0.82 g, 1.49 mmol, 75%).  Data for 2-nitro-4-(tris(4-(tert-butyl)phenyl)methyl)phenol (12): 1H NMR (400 MHz, CDCl3): δ = 10.60 (s, 1H), 8.05 (d, J = 2.4 Hz, 1H), 7.43 (dd, J = 8.9, 2.4 Hz, 1H), 7.26-7.30 (m, 6H), 7.05-7.09 (m, 6H), 7.03 (d, J = 9.2 Hz, 1H), 1.32 (s, 27H) ppm. 13C{1H} NMR (400 MHz, CDCl3): δ = 153.6, 149.2, 143.0, 141.6, 140.5, 132.8, 130.6, 126.0, 124.7, 118.7, 63.1, 34.5, 31.5 ppm. IR: v = 2960, 2902, 2867, 1626, 1584, 1537, 1507, 1479, 1324, 1268, 1247, 1188, 1174, 1109, 1018, 842, 824, 689 cm-1. ESI-MS (MeOH): m/z = 548.4 [M-H]- (theoretical: 548.3). TOF HRMS ESI Calc’d for C37H43NO3Na: 572.3141. Found: 572.3139 (-0.3 ppm). m.p. 225-228 ºC. Anal. Calc’d for C37H43NO3: C, 80.84; H, 7.88; N, 2.55. Found: C, 81.12; H, 7.81; N, 2.53.  Synthesis of 2-amino-4-(tris(4-(tert-butyl)phenyl)methyl)phenol (11): 2-Nitro-4-(tris(4-(tert-butyl)phenyl)methyl)phenol (0.27 g, 0.48 mmol) and palladium on carbon were stirred in 15 mL THF. Hydrazine hydrate (0.10 mL, 2.1 mmol) was added via a micropipette. The mixture was 200  refluxed for 3 h. After cooling, the mixture was filtered through celite to remove the catalyst.  The solvent was removed under reduced pressure. The crude product was eluted through silica gel plug with 5% ethyl acetate in DCM. Product was isolated as a white solid (0.23 g, 0.44 mmol, 92%). It became a brown solid after exposure to air for several days. Stability is decreased in solution state.  Data for 2-amino-4-(tris(4-(tert-butyl)phenyl)methyl)phenol (11): 1H NMR (400 MHz, CDCl3): δ = 7.20-7.24 (m, 6H), 7.08-7.12 (m, 6H), 6.59-6.63 (m, 2H), 6.49 (dd, J = 8.3, 2.4 Hz, 1H), 4.74 (br, s, 1H), 3.48 (br, s, 2H), 1.31 (s, 27H) ppm. ESI-MS (MeOH): m/z = 520.5 [M - H]-(theoretical: 520.4).  Synthesis of 2-hydroxy-5-(pyridin-3-ylethynyl) benzaldehyde (7a): 3-Ethynylpyridine (900 mg, 8.73 mmol), 2-hydroxy-5-iodobenzaldehyde (2.16 g, 8.73 mmol), cis-Bis(triphenylphosphine)platinum(II) dichloride (346 mg, 0.438 mmol) and copper(I) iodide (83 mg, 0.44 mmol) were suspended in 60 mL dry diisopropylamine under nitrogen. The mixture was stirred at room temperature for 16 h. The reaction was quenched by adding H2O followed by extraction with dichloromethane. Organic phase was dried with anhydrous MgSO4. The solvent was removed under reduced pressure. The brown solid obtained was loaded onto a silica gel column and eluted with 10:1 DCM/ethyl acetate. The product had an Rf of approximately 0.3 and was isolated as a yellow solid (890 mg, 3.98 mmol, 46%).   Data for 2-hydroxy-5-(pyridin-3-ylethynyl) benzaldehyde (7a): 1H NMR (400 MHz, CDCl3): δ = 11.16 (s, 1H), 9.90 (s, 1H), 8.75-8.76 (m, 1H), 8.55 (dd, J = 4.8, 1.3 Hz, 1H), 7.79-7.81 (m, 1H), 7.77-7.78 (m, 1H), 7.67 (dd, J = 8.3, 2.2 Hz, 1H), 7.29 (m, 1H), 7.00 (d, J = 8.3 Hz, 1H) ppm. 201  13C{1H} NMR (400 MHz, CDCl3): δ = 196.1, 161.9, 152.3, 148.8, 139.9, 138.5, 137.2, 123.2, 120.7, 120.3, 118.5, 114.6, 91.0, 85.7 ppm. IR: v = 3025, 2883, 2214, 1653, 1590, 1483, 1410, 1367, 1281, 1187, 1133, 1019, 922, 843, 796, 744, 699 cm-1. ESI-MS (MeOH): m/z = 224.4 [M+H]+ (theoretical: 224.1). TOF HRMS ESI Calc’d for C14H10NO2: 224.0712. Found: 224.0716 (1.8 ppm). m.p. 113-114 ºC. Anal. Calc’d for C14H9NO2: C, 75.33; H, 4.06; N, 6.27. Found: C, 75.11; H, 4.10; N, 6.11.  Synthesis of 2-((2-hydroxy-5-(pyridin-3-ylethynyl)benzylidene)amino)- 4-(tris(4-(tert-butyl) phenyl)methyl)phenol (6a): A suspension of 2-amino-4-(tris(4-(tert-butyl)phenyl)methyl)phenol (232 mg, 0.445 mmol) and 2-hydroxy-5-(pyridin-3-ylethynyl)benzaldehyde (104 mg, 0.466 mmol) in ethanol (15 mL) was heated at reflux for 15 min. The precipitate was collected by filtration after cooling to produce a light orange solid. The product was purified by recrystallization from ethanol to yield 280 mg imine (0.386 mmol, 87%).  Data for 2-((2-hydroxy-5-(pyridin-3-ylethynyl)benzylidene)amino)-4-(tris (4-(tert-butyl)phenyl) methyl)phenol (6a): 1H NMR (400 MHz, CDCl3): δ = 12.75 (br, s, 1H), 8.78 (s, 1H), 8.59 (d, J = 3.9 Hz, 1H), 8.39 (s, 1H), 7.83 (d, J = 7.8 Hz, 1H), 7.58 (dd, J = 8.3, 2.0 Hz, 1H), 7.54 (d, J = 2.4 Hz, 1H), 7.34-7.37 (m, 1H), 7.30-7.33 (m, 6H), 7.15-7.18 (m, 6H), 7.13 (dd, J = 8.8, 2.4 Hz, 1H), 7.06 (d, J = 8.3 Hz, 1H), 6.99 (d, J = 2.4 Hz, 1H), 6.94 (d, J = 8.8 Hz, 1H), 5.69 (br, s, 1H), 1.36 (s, 27H) ppm. 13C{1H} NMR (400 MHz, CDCl3): δ = 162.4, 161.2, 148.8, 148.5, 147.9, 143.8, 140.8, 138.4, 136.6, 136.0, 134.0, 132.0, 130.8, 124.4, 121.6, 119.5, 118.0, 114.9, 113.8, 91.9, 85.1, 63.4, 34.5, 31.5, 29.8 ppm (two signals missing). IR: v = 2954,  2904, 2869, 2219, 1624, 1565, 1503, 1491, 1299, 1268, 1117, 1018, 841, 822, 809, 704 cm-1. ESI-MS (MeOH): 202  m/z = 725.6 [M + H]+ (theoretical: 725.4). TOF HRMS ESI Calc’d for C51H53N2O2: 725.4107. Found: 725.4094 (-1.8 ppm). m.p. 298-300 ºC. Anal. Calc’d for C51H52N2O2: C, 84.49; H, 7.23; N, 3.86. Found: C, 84.25; H, 7.16; N, 3.89.  Synthesis of macrocycle (5a): A suspension of K2PtCl4 (58 mg, 0.14 mmol) in DMSO (8 mL) was sparged with nitrogen and then heated to 100 ºC until all of the salt dissolved. A separate Schlenk flask was charged with imine (0.10 g, 0.14 mmol), and K2CO3 (48 mg, 0.35 mmol) 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 then heated at 120 °C for 6 h. The yellow brown suspension was cooled to room temperature followed by centrifugation. After decanting the supernatant solution, three cycles of washing, centrifuging, and decanting were performed, once with water and twice with MeOH. Upon drying the precipitate, macrocycle was isolated as a yellow powder (49 mg, 0.053 mmol, 39%).  Data for macrocycle (5a): 1H NMR (400 MHz, CDCl3, 25 ºC): δ = 8.74 (br, s, 1H), 8.57 (br, s, 1H), 7.64 (br, s, 1H), 7.15-7.42 (m, 16H), 7.13 (br, s, 1H), 6.89 (d, J = 8.0 Hz, 1H), 6.61 (d, J = 8.0 Hz, 1H), 6.23 (br, s, 1H), 1.39 (s, 27H) ppm. IR: v = 3083, 3031, 2958, 2903, 2869, 2206, 1596, 1571, 1492, 1415, 1380, 1362, 1316, 1269, 1195, 1135, 1109, 1064, 1017, 916, 818, 708, 688 cm-1. MALDI-TOF MS: m/z = 2755.7 [M+H]+, 5511.6 [2M+H]+, 8268.2 [3M+H]+, 11026.4 [4M+H]+, 13785.4 [5M+H]+, 16545.2 [6M+H]+, 19305.9 [7M+H]+, 22064.5 [8M+H]+, 24824.7 [9M+H]+, 27578.7 [10M+H]+, 30344.2 [11M+H]+, 33117.0 [12M+H]+, 35863.9 [13M+H]+. m.p. above 360 ºC.  203  Synthesis of 3-((4-iodophenyl)ethynyl)pyridine (9b): 3-Ethynylpyridine (1.45 g, 14.0 mmol), 1,4-diiodobenzene (18.5 g, 56.2 mmol), cis-bis(triphenylphosphine)platinum(II) dichloride (493 mg, 0.624 mmol) and copper(I) iodide (134 mg, 0.704 mmol) were suspended in 50 mL dry diisopropylamine under nitrogen. The mixture was concentrated under reduced pressure. The remaining solid was dissolved 40 mL dichloromethane. The solution was filtered through celite, then the solvent was removed under reduced pressure. The yellow solid obtained was loaded onto a silica gel column and eluted with 9:1 DCM/ethyl acetate. The product has an Rf of approximately 0.3 and was isolated as a yellow solid (2.46 g, 8.14 mmol, 58%).   Data for 3-((4-iodophenyl)ethynyl)pyridine (9b): 1H NMR (400 MHz, CD2Cl2): δ = 8.73 (dd, J = 2.0, 0.8 Hz, 1H), 8.54 (dd, J = 4.8, 1.6 Hz, 1H), 7.81 (dt, J = 8.0, 2.0×(2) Hz, 1H), 7.72-7.75 (m, 2H), 7.27-7.32 (m, 3H) ppm. 13C{1H} NMR (400 MHz, CD2Cl2): δ = 152.7, 149.5, 138.9, 138.3, 133.6, 123.6, 120.7, 95.3, 91.9, 87.9 ppm (one signal missing). IR: v = 1615, 1405, 1391, 1298, 1270, 1150, 1139, 1021, 968, 818 cm-1. ESI-MS (MeOH): m/z = 306.1 [M + H]+ (theoretical: 306.0).   Synthesis of 5-ethynyl-2-hydroxybenzaldehyde (8b): 5-Bromo-2-hydroxybenzaldehyde (4.65 g, 23.1 mmol), ethynyltrimethylsilane (3.73 mL, 26.2 mmol), cis-bis(triphenylphosphine)platinum(II) dichloride (324 mg, 0.410 mmol) and copper(I) iodide (132 mg, 0.693 mmol) were suspended in 150 mL dry diisopropylamine under nitrogen. The mixture was refluxed for 4 h. After cooling, the mixture was concentrated under reduced pressure, then the remaining solid was dissolved in 150 mL pentane and filtered through celite. Filtrate was concentrated under reduced pressure to yield a dark brown solid. Then the solid was dissolved in a solution of MeOH (15 mL)/THF (30 mL) of 204  KOH (2.19 g, 39.1 mmol). The mixture was stirred at room temperature for 12 h. Solvent was removed under reduced pressure. The residue was dissolved in 50 mL chloroform. The organic solution was washed with 0.1 M HCl aqueous solution and then with distilled H2O. The organic phase was dried with anhydrous Na2SO4. The solvent was removed under reduced pressure. The brown solid obtained was loaded onto a silica gel column and eluted with 1:1 DCM/hexanes. The product has an Rf of approximately 0.6 and was isolated as a white solid (1.34 g, 9.25 mmol, 40%).   Data for 5-ethynyl-2-hydroxybenzaldehyde (8b): 1H NMR (400 MHz, CDCl3): δ = 11.14 (s, 1H), 9.88 (s, 1H), 7.73 (d, J = 2.0 Hz, 1H), 7.63 (dd, J = 8.4, 2.0 Hz, 1H), 6.97 (d, J = 8.4 Hz, 1H), 3.05 (s, 1H) ppm. 13C{1H} NMR (400 MHz, CDCl3): δ = 196.0, 161.7, 140.2, 137.5, 120.4, 118.1, 114.0, 81.8, 76.8 ppm. IR: v = 3272, 1659, 1579, 1476, 1376, 1289, 1200, 1139, 934, 910, 846, 764, 741, 664 cm-1. ESI-MS (MeOH): m/z = 145.2 [M - H]- (theoretical: 145.0).   Synthesis of 2-hydroxy-5-((4-(pyridin-3-ylethynyl)phenyl)ethynyl) benzaldehyde (7b): 3-((4-Iodophenyl)ethynyl)pyridine (1.0 g, 3.4 mmol), 5-ethynyl-2-hydroxybenzaldehyde (0.50 g, 3.4 mmol), cis-bis(triphenylphosphine)platinum(II) dichloride (120 mg, 0.15 mmol) and copper(I) iodide (65 mg, 0.34 mmol) were suspended in 60 mL dry diisopropyl amine under nitrogen. The mixture was stirred at 35 ºC for 15 h. The mixture was concentrated by rotary evaporation. The residue was purified by column chromatography with 20:1 DCM/ethyl acetate. The product was isolated as a yellow solid (0.81 g, 2.5 mmol, 73%) with an Rf of approximately 0.5.  Data for 2-hydroxy-5-((4-(pyridin-3-ylethynyl)phenyl)ethynyl) benzaldehyde (7b): 1H NMR (400 MHz, CDCl3): δ = 11.15 (s, 1H), 9.91 (s, 1H), 8.79 (br, s, 1H), 8.58 (br, s, 1H), 7.82 (dt, J = 205  8.0, 1.5×(2) Hz, 1H), 7.77 (d, J = 2.0 Hz, 1H), 7.68 (dd, J = 8.5, 2.0 Hz, 1H), 7.49-7.56 (m, 4H), 7.31 (dd, J = 7.9, 4.8 Hz, 1H), 7.01 (d, J = 8.5 Hz, 1H) ppm. 13C{1H} NMR (400 MHz, CDCl3): δ = 196.2, 161.7, 152.4, 148.9, 139.9, 138.6, 137.1, 131.8, 131.6, 123.5, 123.3, 122.6, 120.7, 120.4, 118.4, 115.0, 92.4, 89.9, 88.7, 88.0 ppm. IR: v = 3040, 2206, 1654, 1581, 1509, 1471, 1407, 1325, 1279, 1186, 1020, 830, 803, 750, 699, 684 cm-1. ESI-MS (MeOH): m/z = 324.3 [M + H]+ (theoretical: 324.1). TOF HRMS ESI Calc’d for C22H14NO2: m/z = 324.1025. Found: 324.1027 (0.6 ppm). m.p. decomposes at 181 ºC. Anal. Calc’d for C22H13NO2: C, 81.72; H, 4.05; N, 4.33. Found: C, 81.28; H, 3.98; N, 4.25.  Synthesis of 2-((2-hydroxy-5-((4-(pyridin-3-ylethynyl)phenyl)ethynyl)benzylidene) amino)- 4-(tris(4-(tert-butyl)phenyl)methyl)phenol (6b): A suspension of 2-amino-4-(tris(4-(tert-butyl)phenyl)methyl)phenol (310 mg, 0.594 mmol) and 2-hydroxy-5-((4-(pyridin-3-ylethynyl)phenyl)ethynyl)benzaldehyde (193 mg, 0.597 mmol) in ethanol (35 mL) was heated at reflux for 1 h. The precipitate was collected by filtration after cooling to produce a red solid. The product was purified by recrystallization from ethanol to yield 393 mg imine (0.477 mmol, 79%).  Data of 2-((2-hydroxy-5-((4-(pyridin-3-ylethynyl)phenyl)ethynyl) benzylidene) amino)-4-(tRis(4- (tert-butyl)phenyl)methyl)phenol (6b): 1H NMR (400 MHz, CDCl3): δ = 12.67 (br, s, 1H), 8.78 (br, s, 1H), 8.57 (br, s, 1H), 8.36 (s, 1H), 7.83 (d, J = 7.8 Hz, 1H), 7.47-7.56 (m, 6H), 7.30-7.34 (m, 1H), 7.26-7.30 (m, 6H), 7.11-7.15 (m, 6H), 7.09 (dd, J = 8.8, 2.4 Hz, 1H), 7.01 (d, J = 8.8 Hz, 1H), 6.96 (d, J = 2.4 Hz, 1H), 6.91 (d, J = 8.3 Hz, 1H), 5.66 (br, s, 1H), 1.33 (s, 27H) ppm. IR: v = 3034, 2958, 2904, 2869, 2206, 1621, 1504, 1362, 1294, 1271, 1225, 1157, 1110, 1017, 818, 808, 723, 703 cm-1.  ESI-MS (MeOH): m/z = 825.7 [M + H]+ (theoretical: 825.4). TOF 206  HRMS ESI Calc’d for C59H57N2O2: m/z = 825.4420. Found: 825.4418 (-0.2 ppm). m.p. 306-307 ºC. Anal. Calc’d for C59H56N2O2: C, 85.89; H, 6.84; N, 3.40. Found: C, 85.69; H, 6.75; N, 3.38.  Synthesis of macrocycle (5b): A suspension of K2PtCl4 (146 mg, 0.351 mmol) in DMSO (20 mL) was sparged with nitrogen and then heated to 100 ºC until all of the salt dissolved. A separate Schlenk flask was charged with imine (290 mg, 0.351 mmol), and K2CO3 (97 mg, 0.70 mmol) 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 at 120 ºC for 6 h. The yellow brown suspension was cooled to room temperature followed by centrifugation. After decanting the supernatant solution, three cycles of washing, centrifuging, and decanting were performed, once with water and twice with MeOH. Upon drying the precipitate, macrocycle was isolated as a yellow powder (136 mg, 0.0445 mmol, 38%).  Data for macrocycle (5b): IR: v = 3083, 3031, 2960, 2904, 2869, 2201, 1596, 1508, 1489, 1409, 1363, 1314, 1269, 1197, 1175, 1135, 1018, 918, 820, 709, 685 cm-1. MALDI-TOF MS: m/z = 3055.88 [M+H]+. m.p. above 360 ºC.  Synthesis of 2-hydroxy-5-(phenylethynyl) benzaldehyde (7c): 5-Bromo-2-hydroxybenzaldehyde (1.01 g, 5.00 mmol), ethynylbenzene (1.02 g, 10.0 mmol), cis-bis(triphenylphosphine)platinum(II) dichloride (175 mg, 0.221 mmol) and copper(I) iodide (48 mg, 0.25 mmol) were suspended in a mixture of 10 mL dry diisopropylamine and 10 mL dry THF under nitrogen. The mixture was stirred at 75 ºC for 16 h. After cooling, the reaction was quenched by adding HCl/H2O followed by extraction with diethyl ether. The organic phase was dried with 207  anhydrous Na2SO4. The solvent was removed under reduced pressure, and the black solid obtained was loaded onto a silica gel column and eluted with 19:1 hexanes/ethyl acetate. The product has an Rf of approximately 0.4 and was isolated as a yellow solid (0.470 g, 2.11 mmol, 42%).  Data for 2-hydroxy-5-(pyridin-3-ylethynyl) benzaldehyde (7c): 1H NMR (400 MHz, CDCl3): δ = 11.13 (s, 1H), 9.91 (s, 1H), 7.78 (d, J = 1.8 Hz, 1H), 7.69 (dd, J = 8.8, 2.0 Hz, 1H), 7.51-7.54 (m, 2H), 7.35-7.39 (m, 3H), 7.01 (d, J = 8.8 Hz, 1H) ppm. 13C{1H} NMR (400 MHz, CDCl3): δ = 196.2, 161.4, 139.9, 136.9, 131.6, 128.5, 122.9, 120.6, 118.2, 115.3, 88.9, 87.6 ppm (one signal missing). IR: v =  1653, 1578, 1491, 1379, 1317, 1280, 1178, 1117, 834, 745, 685 cm-1. ESI-MS (MeOH): m/z = 221.2 [M-H]- (theoretical: 221.1). m.p. 80-81 ºC. Anal. Calc’d for C15H10O2: C, 81.07; H, 4.54. Found: C, 80.97; H, 4.57.  Synthesis of 2-((2-hydroxy-5-(phenylethynyl)benzylidene)amino)-4-(tris(4-(tert-butyl)phenyl)methyl)phenol (6c): A suspension of 2-amino-4-(tris(4-(tert-butyl)phenyl)methyl)phenol (782 mg, 1.51 mmol) and 2-hydroxy-5-(pyridin-3-ylethynyl) benzaldehyde (336 mg, 1.51 mmol) in ethanol (90 mL) was heated at reflux for 30 min. The precipitate was collected by filtration after cooling to produce a light orange solid. The product was purified by recrystallization from ethanol to yield 873 mg imine (1.21 mmol, 80%).  Data for 2-((2-hydroxy-5-(phenylethynyl)benzylidene)amino)-4-(tris(4-(tert-butyl)phenyl)methyl)phenol (6c):  1H NMR (400 MHz, CDCl3): δ = 12.58 (br, s, 1H), 8.36 (s, 1H), 7.78 (d, J = 8.0 Hz, 1H), 7.54 (dd, J = 8.4, 2.0 Hz, 1H), 7.48-7.52 (m, 3H), 7.33-7.39 (m, 3H), 7.26-7.30 (m, 6H), 7.11-7.15 (m, 7H), 7.09 (dd, J = 8.4, 2.4 Hz, 1H), 7.01 (d, J = 8.8 Hz, 1H), 6.95 (d, J = 2.0 Hz, 1H), 6.90 (d, J = 8.4 Hz, 1H), 5.60 (br, s, 1H), 1.33 (s, 27H) ppm. IR: v = 2961, 208  2901, 2865, 2212, 1624, 1506, 1362, 1291, 1158, 1107, 839, 816, 753, 690 cm-1. ESI-MS (MeOH): m/z = 724.6 [M + H]+ (theoretical: 724.4). m.p. 297-299 ºC. Anal. Calc’d for C52H53NO2: C, 86.27; H, 7.38; N, 1.93. Found: C, 86.24; H, 7.42; N, 1.91. Synthesis of model compound (5c): A suspension of K2PtCl4 (52 mg, 0.12 mmol) in DMSO (6 mL) was sparged with nitrogen and then heated to 100 ºC until all of the salt dissolved. A separate Schlenk flask was charged with imine (90 mg, 0.12 mmol), and K2CO3 (34 mg, 0.25 mmol) 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 at 150 ºC for 4 h. The yellow brown suspension was cooled to room temperature followed by centrifugation. After decanting the supernatant solution, three cycles of washing, centrifuging, and decanting were performed, once with water and twice with MeOH. Upon drying the precipitate, macrocycle was isolated as a yellow powder (28 mg, 0.028 mmol, 23 %).  Data for model compound (5c): 1H NMR (400 MHz, CDCl3, 25 ºC): δ = 9.19 (m, 2H), 8.11 (s, 1H), 7.95 (tt, J = 8.0×(2), 1.8×(2) Hz, 1H), 7.63 (d, J = 2.0 Hz, 1H), 7.55 (dd, J = 8.8, 2.0 Hz, 1H), 7.46-7.53 (m, 5H), 7.31-7.36 (m, 3H), 7.28-7.31 (m, 7H), 7.18-7.21 (m, 6H), 7.15-7.18 (m, 1H), 6.96-7.01 (m, 2H), 1.35 (s, 1H) ppm. 13C{1H} NMR (400 MHz, CDCl3): δ = 165.7, 162.0, 149.9, 148.5, 144.1, 142.4, 138.8, 138.4, 137.5, 135.5, 135.1, 132.8, 131.3, 130.8, 128.3, 127.7, 125.2, 124.1, 123.8, 122.2, 121.8, 117.2, 116.3, 111.0, 89.2, 87.8, 63.1, 34.4, 31.4 ppm. IR: v = 2963, 2902, 2867, 2212, 1671, 1606, 1595, 1494, 1484, 1375, 1314, 1271, 1177, 1018, 922, 840, 822, 757, 687 cm-1. ESI-MS (MeOH): m/z = 996.5 [M + H]+ (theoretical: 996.4). m.p. above 360 ºC. Anal. Calc’d for C57H56N2O2Pt: C, 68.73; H, 5.67; N, 2.81. Found: C, 69.03; H, 5.71; N, 2.83.  209  Chapter 5: Conclusions and Future Directions  5.1 Overview  In this thesis, the synthesis and characterization of a variety of new Schiff-base macrocycles are discussed. Different from the typical synthesis of Schiff-base macrocycles190 involving condensation reactions between symmetric diamines and symmetric dialdehydes, these new Schiff-base macrocycles were prepared through the head-to-tail approach.191 As a result, 5-fold symmetric macrocycles (campestarenes) and 3-fold symmetric Pt3 macrocycles were synthesized from asymmetric macrocycle precursors.  Different substituent groups were introduced to campestarenes to prevent aggregation so that their tautomeric behavior could be studied. Benefitting from bulky triisopropylsilyl groups, campestarene 1e demonstrated no aggregation but increased solubility in various solvents. Numerous studies were conducted to understand campestarenes’ rich tautomeric behavior in solution. Studies showed that the predominance of one tautomer over the others is dictated by both the relative permittivity of solvent and temperature. Different from its rich tautomeric behavior in solution, campestarenes in the solid state only display extreme enol-imine tautomers. However, regular campestarenes suffer from issues with stability, which inhibited their further study and potential applications. With the purpose of improving the stability of regular campestarenes, attempts were made to synthesize campestarenes from ketone compounds. Despite a time-consuming macrocyclization process, one phenyl derived campestarene 1l was synthesized and isolated.  210  In addition, in order to synthesize macrocycles with larger guest-accessible channels, I set out to explore the synthesis and self-assembly of extended Pt-containing Schiff-base macrocycles.192 With the goal of preparing expanded Pt4 macrocycles, ethynyl spaced pyridyl salicylaldehydes were synthesized through Sonogashira coupling reactions. Macrocyclization with expanded pyridyl salicylaldehydes resulted in Pt3 macrocycles. Compared to Pt4 macrocycles, Pt3 macrocycles 5a demonstrated improved aggregation behavior. Self-assembly of Pt3 macrocycles 5a led to the formation of nanotubes in solid, which was comfired by MALDI-TOF, TEM and PXRD. And variable-temperature 1H NMR, DOSY, NOESY studies were conducted to investigate aggregation of macrocycle 5a in solution. NOESY showed that macrocycle 5a displays aggregation over wide temperature range. DOSY experiments showed that Pt3 macrocycles have larger size at higher temperature, indicating an entropy-driven self-assembly.  5.2 Future Directions  5.2.1 More Phenyl Derived Campestarenes The MacLachlan group has been working to explore Schiff-base macrocycles in coordination chemistry.193 With a unique 5-fold symmetry, campestarenes might demonstrate some interesting results in forming complexes or metal clusters.194 However, regular campestarenes suffer from issues with decomposition. Although phenyl derived campestarenes 1l has been synthesized and isolated, the time-consuming macrocyclization process limits its practical application. To study campestarenes in coordination chemistry, more efficient synthesis to ketone-derived campestarenes is required to be explored. Apart from preparing campestarene 1l from amine 2l under different conditions, synthesis of ketone-derived campestarenes with different structures, 211  such as campestarene 1m, 1n and 1o, were also under investigation. The proposed synthesis of campestarenes and precursors are shown in Scheme 5.1. These electron-donating or electron-withdrawing groups might not only affect the macrocyclization of precursors, but also affect properties of campestarenes.  Scheme 5.1  Proposed synthesis of ketone derived campestarenes 1m, 1n and 1o.  Another approach to improve the synthesis of phenyl-derived campestarenes is template mediated synthesis.195 By adding a templete that only selectively binds to campestarenes, the formation of macrocyclic product may increase.196 Especially in the formation of cyclic imines whose size match to templating metal ions, high yields of macrocyclic imine-metal ion complexes may be obtained.197   212  5.2.2 Campestarenes with Liquid Crystal Properties Since first reported in 1977, discotic liquid crystals have attracted considerable attention from chemists.198 Although many organic compounds demonstrate this property, columnar mesophases based on macrocycles199 are particularly interesting due to their potential to self-assemble into nanotubes.200  Attempts to synthesize campestarenes with liquid crystalline properties have already been made. Long alkoxy chains were introduced to campestarenes 14 to impart potential liquid crystalline properties to these 5-fold symmetric macrocycles. Instead of having one alkoxy group in the precursor, precursor 15 bore three alkoxy chains. It was also the first campestarene bearing more than one substituent group on each repeating unit. The synthesis of campestarene precursor 15 from commercially available gallic acid proved successful (Scheme 5.2). However, more research is required to find effective and convenient macrocyclization conditions.             213  Scheme 5.2 Synthesis of trisubstituted campestarene 14.  In addition, another attempt to synthesize campestarenes with liquid crystalline properties could be performed by replacing substituent groups R in regular campestarenes with a long alkoxy chain. Scheme 5.3 shows the proposed synthesis of campestarene 20 which only has one substituent group in each repeating unit.     214  Scheme 5.3  Proposed synthetic route of campestarene 21.   5.2.3 Extended Campestarenes As macrocycles are usually involved in host-guest chemistry,201 both the size and shape of macrocyclic cavity plays important roles in their binding affinity.202 Similarly, in coordination chemistry, a macrocycle needs to provide a binding site with the proper size and coordination geometry in order to host specific metal ions. Thus synthesizing macrocycles with various sizes and geometries is crucial for further study and applications.     215  Scheme 5.4  Proposed synthesis of expanded campestarene 21.   With the goal of tuning the size of campestarenes, ethynyl spaced campestarene precursor 22 could be synthesized using Sonogashira coupling reactions (Scheme 5.4). However, since the ethynyl groups are sometimes more flexible than expected, formation of macrocycles with different geometry may occur with these extended macrocycle precursors.203  5.2.4 Pt(II) Containing Macrocycles with Different Geometry Macrocycles can be potential candidates for building ion channels. 204  With the goal of synthesizing macrocycles with better guest-accessible channels for exploring their supramolecular chemistry, we set out to prepare Pt(II) containing macrocycles with larger dimensions. Apart from extending organic linkers, macrocycles with larger cavities could be achieved by producing macrocycles with different geometries.205 With the replacement of 3-pyridyl groups with 4-pyridyl 216  groups, a conversion of geometry from 3-fold to 6-fold symmetry may be expected (Scheme 5.5). This geometry change would result in a much larger macrocycle that could be used to construct larger guest-accessible channels after self-assembling into nanotubes.  Scheme 5.5  Proposed synthesis of hexagonal Pt(II) containing Schiff-base macrocycles.   5.3 Experimental  5.3.1 Materials All reactions were carried out under air unless otherwise stated. Tetrahydrofuran (THF) was distilled from sodium/benzophenone under nitrogen. Triethylamine was purged with nitrogen gas 217  and dried over 4Å molecular sieves before use. All reagents were used as received unless otherwise stated.  5.3.2 Equipment 400 MHz 1H and 100.6 MHz 13C{1H} NMR spectra were recorded on a Bruker AV-400 spectrometer. Mass spectra and elemental analyses were obtained at the UBC Microanalytical Services Laboratory. Electrospray ionization (ESI) mass spectra were obtained on a Bruker Esquire-LC ion trap mass spectrometer equipped with an electrospray ion source. Samples for both ESI and HR-ESI were analyzed in methanol or methanol / methylene chloride mixtures at 1 μM. Elemental analyses were obtained on a Carlo Erba Elemental Analyzer EA 1108. IR spectra were obtained using a Thermo Scientific Nicolet 6700 FT-IR spectrometer equipped with a diamond attenuated total reflectance.   5.3.3 Procedure and Experimental Data Synthesis of methyl 3,4,5-tris(dodecyloxy)benzoate (19): To a solution of gallic acid (30.00 g, 176 mmol) in methanol (500 mL) was added concentrated H2SO4 (20 mL). The reaction mixture was stirred at 80 oC for 12 h. The mixture was concentrated under reduced pressure. Water was added to the residue, then the aqueous phase was extracted with ethyl acetate (3 x 100 mL). The combined organic phases were dried over Na2SO4, filtered, and then concentrated under reduced pressure. The crude methyl 3,4,5-trihydroxybenzoate was recrystallized from water. The product was directly used for the next step. To a solution of crude methyl 3,4,5-trihydroxybenzoate (1 eq., 4.00 g, 21.7 mmol) in acetone (150 mL) were added potassium carbonate (9.9 eq., 29.7 g, 215 mmol), potassium iodide (0.200 218  g, 1.20 mmol) and 1-bromododecane (3.3 eq., 17.21 mL, 71.7 mmol). After 24 h reflux, the mixture was evaporated to dryness under reduced pressure. The residue was partitioned between CH2Cl2 and water. The organic phase was dried over Na2SO4, filtered, and then concentrated under reduced pressure. The crude product was purified by trituration in ethanol and was obtained as a white solid (6.73 g, 9.77 mmol, 45% yield).  Data for methyl 3,4,5-tris(dodecyloxy)benzoate (19): 1H NMR (400 MHz, CDCl3, 25 ºC): δ = 7.26 (s, 2H, Ar), 3.99-4.05 (m, 6H, CH2), 3.89 (s, 3H, CH3), 1.70-1.86 (m, 6H, CH2), 1.43-1.52 (m, 6H, CH2), 1.23-1.40 (m, 48H, CH2),  0.86-0.91 (m, 9H, CH3) ppm. 13C{1H} NMR (100.6 MHz, CDCl3, 25 ºC): δ = 166.9, 152.8, 142.4, 124.6, 108.0, 73.4, 69.2, 52.1, 31.9, 30.3, 29.5-29.8 (m), 29.2-29.5 (m), 26.0-26.1 (m), 22.7, 14.1 ppm.  IR (neat): v = 2916, 2848, 1716, 1589, 1504, 1467, 1430, 1336, 1220, 1125, 1012, 962, 762, 721 cm-1. ESI-MS (MeOH): m/z = 689.7 [19+H]+. Elemental Analysis: Calc’d for C44H80O5 : C 76.69%, H 11.70%; Found: C 76.29%, H 12.09%.   Synthesis of 2-(3,4,5-tris(dodecyloxy)phenyl)propan-2-ol (18): To a suspension of Mg (2.2 eq., 0.39 g, 16 mmol) in diethyl ether (50 mL) under nitrogen atmosphere was added MeI (2.2 eq., 1.0 mL, 16 mmol). Reacion was initiated by heating and then the reaction was controlled with an iced water bath. After Mg disappeared, to the Grignard reagent solution was added a solution of methyl 3,4,5-tris(dodecyloxy)benzoate 19 (1 eq., 5.0 g, 7.3 mmol) in diethyl ether (50 mL). The reaction mixture was refluxed for 12 h. NH4Cl aqueous solution was added to quench the reaction. Then the aqueous phase was extrated with diethyl ether (3 x 50 mL). The combined organic phases were dried over Na2SO4, filtered, and then concentrated under reduced pressure. Silica gel flash column 219  chromatography (eluent: CH2Cl2) of the residue gave the compound as a pale yellow oil (4.0 g, 5.7 mmol, 79% yield).  Data for 2-(3,4,5-tris(dodecyloxy)phenyl)propan-2-ol (18): 1H NMR (400 MHz, CDCl3, 25 ºC): δ = 6.68 (s, 2H, Ar), 3.91-4.02 (m, 6H, CH2), 1.69-1.86 (m, 6H, CH2), 1.59 (s, 6H, CH3),  1.41-1.53 (m, 6H, CH2), 1.23-1.40 (m, 48H, CH2),  0.85-0.92 (m, 9H, CH3) ppm.  ESI-MS (MeOH): m/z = 689.6 [18+H]+.  Synthesis of 3,4,5-tris(dodecyloxy)phenol (17): A solution of BF3·Et2O (10 eq., 3.58 mL, 29.0 mmol) and NaBO3·4H2O (2 eq., 0.890 g, 5.78 mmol) in THF (40 mL) was stirred at 0 oC under nitrogen atmosphere for 30 min. To this mixture was added a solution of 2-(3,4,5-tris(dodecyloxy)phenyl)propan-2-ol 18 (1 eq., 2.00 g, 2.90 mmol) in THF (10 mL). The reaction mixture was stirred at room temperature for 12 h. A saturated aqueous sodium sulfite solution (20 mL) was added together with water (200 mL). The aqueous phase was extracted with diethyl ether (3 x 30 mL). The combined organic phases were dried over Na2SO4, filtered, and then concentrated under reduced pressure. Silica gel flash column chromatography (eluent: ethyl acetate/hexanes, 1/5) of the residue yielded the compound as a light yellow solid (1.38 g, 2.13 mmol, 74% yield).  Data for 3,4,5-tris(dodecyloxy)phenol (17): 1H NMR (400 MHz, CDCl3, 25 ºC): δ 6.02 (s, 2H, Ar), 3.83-3.89 (m, 6H, CH2), 1.69-1.80 (m, 6H, CH2), 1.39-1.50 (m, 6H, CH2), 1.24-1.38 (m, 48H, CH2),  0.86-0.92 (m, 9H, CH3) ppm. 13C{1H} NMR (100.6 MHz, CDCl3, 25 ºC): δ = 153.5, 152.0, 131.6, 94.2, 73.7, 68.9, 31.9, 30.2, 29.6-29.8 (m), 29.2-29.5 (m), 26.0-26.1 (m), 22.7, 14.1 ppm. IR (neat): v = 3300, 2917, 2849, 1597, 1507, 1464, 1455, 1388, 1229, 1120, 1021, 805, 722 cm-1. 220  ESI-MS (MeOH): m/z = 647.7 [17+H]+. Elemental Analysis: Calc’d for C42H78O4 : C 77.96%, H 12.15%; Found: C 77.57%, H 12.43%.  Synthesis of 2,3,4-tris(dodecyloxy)-6-hydroxybenzaldehyde (16): Triethylamine (2 eq., 0.18 mL, 1.2 mmol) was added dropwise to a solution of 3,4,5-tris(dodecyloxy)phenol 17 (1 eq., 0.40 g, 0.62 mmol), MgCl2 (2 eq., 12 mg, 1.2 mmol) and (CH2O)n (2.2 eq., 40 mg, 1.4 mmol) in dry THF (6  mL) under a nitrogen atmosphere. After heating at reflux for 12 h, dilute HCl was added at RT until the precipitate dissolved. Most of the THF was removed by rotary evaporation, then the aqueous phase was extracted with CH2Cl2 (3 x 10 mL). The combined organic phases were dried over Na2SO4, filtered, then concentrated under reduced pressure. Silica gel flash column chromatography (eluent: CH2Cl2/hexanes, 1/1) of the residue yielded the compound as a pale orange solid (0.37 g, 5.5 mmol, 89% yield).  Data for 2,3,4-tris(dodecyloxy)-6-hydroxybenzaldehyde (16): 1H NMR (400 MHz, CDCl3, 25 ºC): δ = 12.09 (s, 1H, OH), 10.06 (s, 1H, CHO), 6.16 (s, 1H, Ar), 4.21 (t, J = 6.8 Hz, 2H, CH2), 4.00 (t, J = 6.4 Hz, 2H, CH2), 3.87 (t, J = 6.4 Hz, 2H, CH2), 1.70-1.88 (m, 6H, CH2), 1.41-1.52 (m, 6H, CH2), 1.25-1.38 (m, 48H, CH2),  0.86-0.92 (m, 9H, CH3) ppm. 13C{1H} NMR (100.6 MHz, CDCl3, 25 ºC): δ = 192.7, 161.8, 161.0, 155.1, 133.2, 108.7, 95.6, 74.8, 73.9, 69.0, 31.9, 30.2, 30.1, 29.5-29.8 (m), 29.2-29.5 (m), 28.9, 25.9-26.2 (m), 22.7, 14.1 ppm.  IR (neat): v = 2917, 2849, 1637, 1464, 1364, 1314, 1297, 1239, 1170, 1109, 796, 723 cm-1. ESI-MS (MeOH): m/z = 675.7 [16+H]+. Elemental Analysis: Calc’d for C43H78O5 : C 76.50%, H 11.65%; Found: C 76.67%, H 12.07%.  221  Synthesis of 2,3,4-tris(dodecyloxy)-6-hydroxy-5-nitrobenzaldehyde (15): To a solution of 2,3,4-tris(dodecyloxy)-6-hydroxybenzaldehyde 16 (1 eq., 0.40 g, 0.59 mmol) in glacial acetic acid (12 mL) and CH2Cl2 (6 mL) at 0 ºC was slowly added a solution (2 mL) of nitric acid (0.20 mL) in glacial acetic acid (6 mL). After stirring at RT for 12 h, the reaction mixture was concentrated under reduced pressure. Water was added and the aqueous phase was extracted with ethyl acetate (3 x 10 mL). The combined organic phases were dried over Na2SO4, filtered, and then concentrated under reduced pressure. Silica gel flash column chromatography (eluent: ethyl acetate/hexanes, 1/10) of the residue gave the compound as an orange oil (62 mg, 8.6 mmol, 15% yield).  Data for 2,3,4-tris(dodecyloxy)-6-hydroxy-5-nitrobenzaldehyde (15): 1H NMR (400 MHz, CDCl3, 25 ºC): δ = 12.25 (s, 1H, OH), 10.15 (s, 1H, CHO),  4.25-4.31 (m, 4H, CH2), 3.90 (t, J = 6.8 Hz, 2H, CH2), 1.69-1.83 (m, 6H, CH2), 1.21-1.49 (m, 54H, CH2), 0.86-0.92 (m, 9H, CH3) ppm. 13C{1H} NMR (100.6 MHz, CDCl3, 25 ºC): δ = 193.5, 158.1, 154.2, 151.6, 136.5, 136.5, 130.5, 110.4 ppm. IR (neat): v = 2923, 2853, 1482, 1476, 1466, 1253, 1171, 1131, 1096, 988, 966, 920, 939, 693  cm-1. ESI-MS (MeOH): m/z = 720.7 [15+H]+.   222  References 1 (a) Pedersen, C. J. J. Am. Chem. Soc. 1967, 26, 7017-7036. (b) Pedersen, C. J. Angew. Chem. Int. Ed. Engl. 1988, 27, 1021-1027. 2 Lehn, J. –M. Angew. Chem. Int. Ed. Engl. 1988, 27, 89-112. 3 Cram, D. J. Angew. Chem. Int. Ed. Engl. 1988, 27, 1009-1020. 4 Lehn, J. –M. Acc. Chem. Res. 1978, 11, 49-57. 5 (a) Bong, D. T.; Clark, T. D.; Granja, J. R.; Ghadiri, M. R. Angew. Chem. Int. Ed. 2001, 40, 988-1011. (b) Ghadiri, M. R.; Granja, J. R.; Buehler, L. K. Nature 1994, 369, 301-304. (c) Xin, P. Y.; Zhang, L.; Su, P.; Hou, J. L.; Li, Z. T. Chem. Commun. 2015, 51, 4819-4822. (d) Stepien, M.; Donnio, B.; Sessler, J. L. Angew. Chem. Int. Ed. 2007, 46, 1431-1435. 6 Leung, K. C.-F.; Arico, F.; Cantrill, S. J.; Stoddart, J. F. J. Am. Chem. Soc. 2005, 127, 5808-5810. 7 (a) Anelli, P. L.; Spencer, N.; Stoddart, J. F. J. Am. Chem. Soc. 1991, 113, 5131-5133. (b) Harada, A. Acc. Chem. Res. 2001, 34, 456-464. (c)  8 Gutsche, C. D.; Levine, J. A. J. Am. Chem. Soc. 1982, 104, 2653-2655. 9 Arnaud-Neu, F.; Collins, E. M.; Deasy, M.; Ferguson, G.; Harris, S. J.; Kaitner, B.; Lough, A. J.; McKervey, M. A.; Marques, E.; Ruhl, B. L.; Schwing-Weill, M. J.; Seward, E. M. J. Am. Chem. Soc. 1989, 111, 8681-8691. 10 Ikeda, A.; Shinkai, S. J. Chem. Soc. Perkin Trans. I 1993, 22, 2671-2673.  11 (a) Aoyagi, T.; Nakamura, A.; Ikeda, H.; Mihara, H.; Ueno, A. Anal. Chem. 1997, 69, 659-663. (b) Ueno, A.; Suzuki, I.; Osa, T. Chem. Lett. 1989, 18, 1059-1062. 12 (a) Harada, A. Acc. Chem. Res. 2001, 34, 456-464. (b) Murakami, H.; Kawabuchi, A.; Kotoo, K.; Kunitake, M.; Nakashima, N. J. Am. Chem. Soc. 1997, 119, 7605-7606.  223   13 Harada, A.; Takashima, Y.; Yamaguchi, H. Chem. Soc. Rev. 2009, 38, 875-882. 14 Deng, W.; Yamaguchi, H.; Takashima, Y.; Harada, A. Angew. Chem. Int. Ed. 2007, 46, 5144-5147. 15 Tamesue, S.; Takashima, Y.; Yamaguchi, H.; Shinkai, S.; Harada, A. Angew. Chem. Int. Ed. 2010, 49, 7461-7464. 16 Ghadiri, M. R.; Granja, J. R.; Milligan, R. A.; McRee, D. E.; Khazanovich, N. Nature 1993, 366, 324-327.  17 Varnavski, O.; Raymond, J. E.; Yoon, Z. S.; Yotsutuji, T.; Ogawa, K.; Kobuke, Y.; Goodson, T. J. Phys. Chem. C 2014, 118, 28474-28481. 18 Staab, H. A.; Neunhoeffer, K. Synthesis 1974, 6, 424. 19 Moore, J. S.; Zhang, J. Angew. Chem. Int. Ed. 1992, 31, 922-924. 20 Zhang, W.; Moore, J. S. J. Am. Chem. Soc. 2004, 126, 12796. 21 Zang, L.; Che, Y.; Moore, J. S. Acc. Chem. Res. 2008, 41, 1596-1608. 22 Toyota, S.; Goichi, M.; Kotani, M. Angew. Chem. Int. Ed. 2004, 43, 2248-2251. 23 (a) Anderson, H. L.; Sanders, J. K. M. J. CHEM. SOC. CHEM. COMMUN. 1989, 29, 1714-1715. (b) Anderson, H. L.; Sanders, J. K. M. Angew. Chem. Int.  Ed.  Engl. 1990, 29, 1400-1403.  24 Li, J.; Ambroise, A.; Yang, S. I.; Diers, J. R.; Seth, J.; Wack, C. R.; Bocian, D. F.; Holten, D.; Lindsey, J. S. J. Am. Chem. Soc. 1999, 121, 8927-8940. 25 O’Sullivan M. C.; Sprafke J. K.; Kondratuk D. V.; Rinfray C.; Claridge T. D.; Saywell A.; Blunt, M. O.; O’Shea, J. N.; Beton, P. H.; Malfois, M.; Anderson, H. L. Nature 2011, 469, 72-75. 26 Zhao, D.; Moore, J. S. Chem. Commun. 2003, 7, 807-818. 27 Lin, W.; Jiang, H. J. Am. Chem. Soc. 2003, 125, 8084-8085. 28 (a) Chan, Y.-T.; Li, X.; Yu, J.; Carri, G. A.; Moorefield, C. N.; Newkome, G. R.; Wesdemiotis, 224   C. J. Am. Chem. Soc. 2011, 133, 11967-11976; (b) Wang, J.-L.; Li, X.; Lu, X.; Chan, Y.-T.; Moorefield, C. N.; Wesdemiotis, C.; Newkome, G. R. Chem. Eur. J. 2011, 17, 4830-4838; (c) Li, X.; Chan, Y.- T.; Casiano-Maldonad, M.; Yu, J.; Carri, G. A.; Newkome, G. R.; Wesdemiotis, C. Anal. Chem. 2011, 83, 6667-6674. 29 Wang, M.; Wang, C.; Hao, X. Q.; Liu, J.; Li, X.; Xu, C.; Lopez, A.; Sun, L.; Song, M. P.; Yang, H. B.; Li, X. J. Am. Chem. Soc. 2014, 136, 6664-6671. 30 (a) Becerril, J.; Bolte, M.; Burguete, M. I.; Galindo, F.; Garcia-Espana, E.; Luis, S. V.; Miravet, J. F. J. Am. Chem. Soc. 2003, 125, 6677-6686; (b) Chmielewski, M. J.; Jurczak, J. Chem. Eur. J. 2006, 12, 7652 – 7667. 31 Gause, G. F.; Brazhnikove, M. G. Nature. 1944, 154, 703. 32 Tyndall, J. D. A.; Nall, T.; Fairlie, D. P. Chem. Rev. 2005, 105, 973–999.  33 (a) Gao, X. M.; Ye, Y. H.; Bernd, M.; Kutscher, B.; J. Peptide. Sci. 2002, 8, 418-430. (b) Ye, Y. H.; Gao, X. M.; Liu, M.; Tang, Y. C.; Tian, G. L. Lett. Pept. Sci. 2003, 10, 571-579. (c) Chen, X.; Tirado, M.; Steill, J. D.; Oomens, J.; Polfer, N. C. J. Mass. Spectrom. 2011, 46, 1011–1015. (d) Tirado, M.; Polfer, N. C. Angew. Chem. Int. Ed. 2012, 51, 6436–6438. (e) Fujita, Y.; Fujita, S.; Okada, Y.; Chida, K. Org. Lett. 2013, 15, 1155-1157. (f) Rohrbacher, F.; Deniau, G.; Luther, A.; Bode, J. W. Chem. Sci. 2015, 6, 4889-4896.  34 (a) Shin, S. B. Y.; Yoo, B.; Todaro, L. J.; Kirshenbaum, K. J. Am. Chem. Soc. 2007, 129, 3218-3225. (b) Culf, A. S.; Cuperlovic-Culf, M.; Leger, D. A.; Decken, A. Org. Lett. 2014, 16, 2780-2783. 35 Carver, F. J.; Hunter, C. A.; Shannon, R. J. J. Chem. Soc., Chem. Commun. 1994, 10, 1277-1280. 36 Zhu, J.; Parra, R. D.; Zeng, H.; Skrzypczak-Jankun, E.; Zeng, X. C.; Gong, B. J. Am. Chem. Soc. 2000, 122, 4219-4220. 225   37 Parra, R. D.; Zeng, H.; Zhu, J.; Zheng, C.; Zeng, X. C.; Gong, B. Chem. Eur. J. 2001, 7, 4352-4357. 38 (a) Yuan, L.; Feng, W.; Yamoto, K.; Sanford, A. R.; Xu, D.; Guo, H,; Gong, B. J. Am. Chem. Soc. 2004, 126, 11120-11121. (b) Ferguson, J. S.; Yamoto, K.; Liu, R.; He, L.; Zeng, X. C.; Gong, B. Angew. Chem. Int. Ed. 2009, 48, 3150 –3154. 39 Kim, Y. H.; Calabrese, J.; McEwen, C. J. Am. Chem. Soc., 1996, 118, 1545–1546. 40 Gong, B.; Zeng, H. Q.; Zhu, J.; Yua, L.; Han, Y.; Cheng, S.; Furukawa, M.; Parra, R. D.; Kovalevsky, A. Y.; Mills, J. L.; Skrzypczak-Jankun, E.; Martinovic, S.; Smith, R. D.; Zheng, C.; Szyperski, T.; Zeng, X. C. Proc. Natl. Acad. Sci. USA 2002, 99, 11583–11588. 41 Qin, B.; Chen, X. Y.; Fang, X.; Shu, Y. Y.; Yip, Y. K.; Yan, Y.; Pan, S. Y.; Ong, W. Q.; Ren, C. L.; Su, H. B.; Zeng, H. Q. Org. Lett. 2008, 10, 5127-5130. 42 Qin, B.; Ren, C. L.; Ye, R. J.; Sun, C.; Chiad, K.; Chen, X. Y.; Li, Z.; Xue, F.; Su, H. B.; Chass, G. A.; Zeng, H. Q. J. Am. Chem. Soc. 2010, 132, 9564–9566. 43 Qin, B.; Jiang, L. Y.; Shen, S.; Sun, C.; Yuan, W. X.; Li, S. F. Y.; Zeng, H. Q. Org. Lett. 2011, 13, 6212-6215. 44 Ren, C. L.; Xu, S. Y.; Xu, J.; Chen, X. Y.; Zeng, H. Q. Org. Lett. 2011, 15, 3840-3843. 45 Ren, C. L.; Maurizot, V.; Zhao, H. Q.; Shen, J.; Zhou, F.; Ong, W. Q.; Du, Z. Y.; Zhang, K.; Su, H. B.; Zeng, H. Q. J. Am. Chem. Soc. 2011, 133, 13930–13933. 46 Liu, Y.; Shen, J.; Sun, C.; Ren, C. L.; Zeng, H. Q. J. Am. Chem. Soc. 2015, 137, 12055−12063. 47 Qin, B.; Ong, W. Q.; Ye, R. J.; Du, Z. Y.; Chen, X. Y.; Yan, Y.; Zhang, K.; Su, H. B.; Zeng, H. Q. Chem. Commun. 2011, 47, 5419–5421. 48 Du, Z. Y.; Ren, C. L.; Ye, R. J.; Shen, J.; Maurizot, V.; Lu, Y. J.; Wang, J.; Zeng, H. Q. Chem. Commun. 2011, 47, 12488–12490. 226   49 Qin, B.; Sun, C.; Liu, Y.; Shen, J.; Ye, R. J.; Zhu, J.; Duan, X. F.; Zeng, H. Q. Org. Lett. 2011, 13, 2270-2273.  50 Qin, B.; Shen, S.; Sun, C.; Du, Z. Y.; Zhang, K.; Zeng, H. Q. Chem. Asian J. 2011, 6, 3298–3305. 51 Zhao, H. Q.; Shen, J.; Guo, J. J.; Ye, R. J.; Zeng, H. Q. Chem. Commun. 2013, 49, 2323-2325. 52 Shen, J.; Ma, W. L.; Yu, L.; Li, J. B.; Tao, H. C.; Zhang, K.; Zeng, H. Q. Chem. Commun. 2014, 50, 12730-12733. 53 Fu, H. L.; Chang, H.; Shen, J.; Yu, L.; Qin, B.; Zhang, K.; Zeng, H. Q. Chem. Commun. 2014, 50, 3582-3584. 54 Ren, C. L.; Shen, J.; Zeng, H. Q. Org. Lett. 2015, 17, 5946-5949. 55 Lee, S.; Chen, C. –H.; Flood, A. H. Nat. Chem. 2013, 5, 704-710. 56 Szumna, A. Chem. Soc. Rev. 2010, 39, 4274-4285. 57 Hirsch, B. E.; Lee, S.; Qiao, B.; Chen, C. –H.; McDonald, K. P.; Tait, S. L.; Flood, A. H. Chem. Commun. 2014, 50, 9827-9830. 58 Liu, Y.; Singharoy, A.; Mayne, C. G.; Sengupta, A.; Raghavachari, K.; Schulten, K.; Flood, A. H. J. Am. Chem. Soc. 2016, 138, 4843−4851. 59 Dethe, D. H.; Srikrishna, A. Tetrahedron Lett. 2005, 46, 3381-3383. 60 Han, F. S.; Osajima, H.; Cheung, M.; Tokuyama, H.; Fukuyama, T. Chem. Commun.  2006, 16, 1757–1759. 61 Han, F. S.; Osajima, H.; Cheung, M.; Tokuyama, H.; Fukuyama, T. Chem. Eur. J. 2007, 13, 3026–3038. 62 Han, F. S.; Tokuyama, H.; Fukuyama, T. Chem. Commun. 2007, 33, 3444–3446. 63 Wojaczyński, J.; Latos-Grażyński, L. Inorg. Chem., 1995, 34, 1054-1062. 227   64 Wojaczyński, J.; Latos-Grażyński, L. Olmstead, M. M.; Balch, A. L. Inorg. Chem., 1997, 36, 4548-4554. 65 Jensen, R. A.; Kelley, R. F.; Lee, S. J.; Wasielewski, M. R.; Hupp, J. T.; Tiede, D. M. Chem. Commun., 2008, 16, 1886–1888. 66 Berggren, G.; Thapper, A.; Huang, P.; Kurz, P.; Eriksson, L.; Styring, S.; Anderlund, M. F. Dalton Trans., 2009, 45, 10044–10054. 67 Matsumoto, N.; Motoda, Y.; Matsuo, T.; Nakashima, T.; Re, N.; Dahan, F.; Tuchagues, J. -P. Inorg. Chem. 1999, 38, 1165-1173. 68 Fujita, M.; Yazaki, J.; Ogura, J. J. Am. Chem. Soc., 1990, 112, 5645-5647. 69 Stang, P. J.; Olenyuk, B. Acc. Chem. Res. 1997, 30, 502-518. 70 Carina, C. F.; Williams, A. F. Inorg. Chem., 2001, 40, 1826-1832. 71 Dell’Amico, D. B.; Calderazzo, F.; Colo, F. D.; Guglielmetti, G.; Labella, L.; Marchetti, F. Inorg. Chim. Acta., 2006, 359, 127-135. 72 Wang, Q. -Q.; Day, V. W.; Bowman-James, K. Chem. Commun., 2013, 49, 8042-8044. 73 Zhao, S. -B.; Wang, R. -Y.; Wang, S. J. Am. Chem. Soc. 2007, 129, 3092-3093. 74 Tokunage, M.; Larrow, J. F.; Kakiuchi, F.; Jacobsen, E. N. Science 1997, 277, 936-938. 75 Sano, T.; Nishio, Y.; Hamada, Y.; Takahashi, H.; Usuki, T.; Shibata, K. J. Mater. Chem. 2000, 10, 157-161. 76 Adams, H.; Bailey. N. A.; Bertrand, P.; Rodriguez de Barbarin, C. O.; Fenton, D. E.; Gou, S. J. Chem. Soc., Dalton Trans. 1995, 2, 275-279. 77 Dziembowska, T.; Guskos, N.; Tupek, J.; Szymczak, R.; Likodimos, V.; Glenis, S.; Lin, C. L.; Wabia, M.; Jagodzinska, E.; Fabrycy, E. Mater. Res. Bull. 1999, 34, 943-954. 78 Gao, J.; Gou, S.; Hu, D.; Chantrapromma, S.; Fun, H.-K.; Meng, Q. Inorg. Chem. 2001, 40, 228   1712-1715. 79 Huck, W. T. S.; van Veggel, F. C. J. M.; Reinhoudt, D. N. Recl. Trav. Chim. Pays-Bas. 1995, 114, 273-276. 80 Gawroński, J.; Kolbon, H.; Kwit, M.; Katrusiak, A. J. Org. Chem. 2000, 65, 5768-5773. 81 Akine, S.; Taniguchi, T.; Nabeshima, T. Tetrahedron Lett. 2001, 42, 8861-8864. 82 Gallant, A. J.; MacLachlan, M. J. Angew. Chem. Int. Ed. 2003, 42, 5307-5310. 83 Gallant, A. J; Chong, J. H.; MacLachlan, M. J. J. Inorg. Chem. 2006, 45, 5248-5250. 84 Ma, C.; Lo, A.; Abdolmaleki, A.; MacLachlan, M. J. J. Org. Lett. 2004, 6, 3841-3844. 85 (a) Frischmann, P. D.; Jiang, J.; Hui, J. K. -H.; Grzybowski, J. J.; MacLachlan, M. J. Org. Lett. 2008, 10, 1255-1258. (b) Ma, C.; Lo, A.; Abdolmaleki, A.; MacLachlan, M. J. Org. Lett. 2004, 6, 3841-3844. (c) Hui, J. K. -H.; MacLachlan, M. J. Chem. Commun. 2006, 23, 2480-2482. 86 Guieu, S.; Crane, A. K.; MacLachlan, M. J. Chem. Commun., 2011, 47, 1169-1171. 87 Frischmann, P. D.; Guieu, S.; Tabeshi, R.; MacLachlan, M. J. J. Am. Chem. Soc. 2010, 132, 7668-7675. 88 Lee, S.; Chen, C.-H.; Flood, A. H. Nat. Chem. 2013, 5, 704-710. 89 (a) Qin, B.; Ren, C.; Ye, R.; Sun, C.; Chiad, K.; Chen, X.; Li, Z.; Xue, F.; Su, H.; Chass, G. A.; Zeng, H. J. Am. Chem. Soc. 2010, 132, 9564-9566; (b) Qin, B.; Ong, W. Q.; Ye, R.; Du, Z.; Chen, X.; Yan, Y.; Zhang, K.; Su, H.; Zeng, H. Chem. Commun. 2011, 47, 5419-5421; (c) Du, Z.; Ren, C.; Ye, R.; Shen, J.; Maurizot, V.; Lu, Y.; Wang, J.; Zeng, H. Chem. Commun. 2011, 47, 12488-11490; (d) Qin, B.; Sun, C.; Liu, Y.; Shen, J.; Ye, R.; Zhu, J.; Duan, X.-F.; Zeng, H. Org. Lett. 2011, 13, 2270-2273. 90 Zhang, J.; Pesak, D. J.; Ludwick, J. L.; Moore, J. S. J. Am. Chem. Soc. 1994, 116, 4227-4239. 91 Ferguson, J. S.; Yamato, K.; Liu, R.; He, L.; Zeng, X. C.; Gong, B. Angew. Chemie Int. Ed. 2009, 229   48, 3150-3154. 92 Cragg, P. J.; Sharma, K. Chem. Soc. Rev. 2012, 41, 597-607. 93 (a) Qin, B.; Chen, X.; Fang, X.; Shu, Y.; Yip, Y. K.; Yan, Y.; Pan, S.; Ong, W. Q.; Ren, C.; Su, H.; Zeng, H. Org. Lett. 2008, 10, 5127-5130; (b) Pedersen, C. J. J. Am. Chem. Soc. 1967, 89, 7017-7036; (c) Bowsher, B. R.; Rest, A. J. Inorganica Chim. Acta 1980, 45, L5-L6. 94 Toullec, J. Adv. Phys. Org. Chem. 1982, 18, 1-77. 95 Kereselidze, J. A.; Zarqua, T. S.; Kikalishvili, T. J.; Churgulia, E. J.; Makaridze, M. S. Russ. Chem. Rev. 2002, 71, 993-1003. 96 (a) Pollack, S. K.; Hehre, W. J. J. Am. Chem. Soc. 1977, 99, 4845-4846; (b) Zimmerman, H. E. Acc. Chem. Res. 1987, 20, 263-268; (c) Toullec, J.; Dubois, J. E. J. Am. Chem. Soc. 1974, 96, 3524-3532. 97 (a) Charette, J. J.; De Hoffmann, E. J. Org. Chem. 1979, 44, 2256-2262; (b) Koll, A. Int. J. Mol. Sci. 2003, 4, 434-444. 98 (a) Chong, J. H.; Sauer, M.; Patrick, B. O.; MacLachlan, M. J. Org. Lett. 2003, 5, 3823-3826; (b) Sauer, M.; Yeung, C.; Chong, J. H.; Patrick, B. O.; MacLachlan, M. J. J. Org. Chem. 2006, 71, 775-788; (c) Yelamaggad, C. V.; Achalkumar, A. S.; Shankar Rao, D. S.; Prasad, S. K. J. Am. Chem. Soc. 2004, 126, 6506-6507; (d) Freiherr von Richthofen, C.-G.; Stammler, A.; Bögge, H.; Glaser, T. J. Org. Chem. 2012, 77, 1435-1448; (e) Vieweger, M.; Jiang, X.; Lim, Y.-K.; Jo, J.; Lee, D.; Dragnea, B. J. Phys. Chem. A 2011, 115, 13298-13308. 99 (a) Gallant, A. J.; Yun, M.; Sauer, M.; Yeung, C. S.; MacLachlan, M. J. Org. Lett. 2005, 7, 4827-4830; (b) Yelamaggad, C. V.; Prabhu, R.; Rao, D. S. S.; Prasad, S. K. Tetrahedron Lett. 2010, 51, 4579-4583. 230   100 (a) Akine, S.; Taniguchi, T.; Nabeshima, T. Tetrahedron Lett. 2001, 42, 8861-8864; (b) Gallant, A. J.; MacLachlan, M. J. Angew. Chemie Int. Ed. 2003, 42, 5307-5310; (c) Ma, C. T. L.; MacLachlan, M. J. Angew. Chemie Int. Ed. 2005, 44, 4178-4182; (d) Gallant, A. J.; Hui, J. K.-H.; Zahariev, F. E.; Wang, Y. A.; MacLachlan, M. J. J. Org. Chem. 2005, 70, 7936-7946. 101 Xu, X.-N.; Wang, L.; Wang, G.-T.; Lin, J.-B.; Li, G.-Y.; Jiang, X.-K.; Li, Z.-T. Chem. - A Eur. J. 2009, 15, 5763-5774. 102 Korich, A. L.; Hughes, T. S. Org. Lett. 2008, 10, 5405-5408. 103 (a) Hartley, C. S.; Moore, J. S. J. Am. Chem. Soc. 2007, 129, 11682-11683; (b) Fekner, T.; Gallucci, J.; Chan, M. K. Org. Lett. 2004, 6, 989-992. 104 Fabian, J.; Legrand, M.; Poirier, P. Bull. Soc. Chim. Fr. 1956, 23, 1499-1509. 105 Krishnankutty, K.; Ummathur, M. B.; Sayudevi, P. J. Argentine Chem. Soc. 2008, 96, 13-21. 106 (a) Allen, F. H. Acta Crystallogr. Sect. B Struct. Sci. 2002, 58, 380–388; (b) CSD Version 5.37 (November 2015 release plus two updates) was searched for phenyl rings containing TIPS substituents, and six such structures were found. This analysis excludes a further three structures containing a TIPS group bound to a larger aromatic . 107 Benzene was used to mimic the possibly average low dielectric constant of the solid phase. 108 Minkin, V. I.; Tsukanov, A. V.; Dubonosov, A. D.; Bren, V. A. J. Mol. Struct. 2011, 998, 179-191. 109 Alarcón, S. H.; Olivieri, A. C.; Sanz, D.; Claramunt, R. M.; Elguero, J. J. Mol. Struct. 2004, 705, 1-9. 110 Dudek, G. O.; Dudek, E. P. J. Am. Chem. Soc. 1966, 88, 2407-2412. 111 Antonov, L.; Fabian, W. M. F.; Nedeltcheva, D.; Kamounah, F. S. J. Chem. Soc. Perkin Trans. 2 2000, 6, 1173-1179. 231   112 (a) Beyramabadi, S. A.; Morsali, A. Int. J. Phys. Sci. 2011, 6, 1780–1788; (b) Tezer, N.; Karakus, N. J. Mol. Model. 2009, 15, 223-232. 113 Marini, A.; Muñoz-Losa, A.; Biancardi, A.; Mennucci, B. J. Phys. Chem. B 2010, 114, 17128-17135. 114 Toluene, rather than benzene, was selected for this study as toluene has a substantially greater temperature range for variable temperature studies. The room temperature UV-visible and 1H NMR spectra of 1h in toluene were nearly identical to those in benzene. 115 Dudek, G. O.; Dudek, E. P. J. Am. Chem. Soc. 1964, 86, 4283-4287. 116  The choice of these tautomers allows to probe the largest possible number of local tautomerization patterns without computing the whole set of tautomers. 117  It is assumed that vibrational contributions are negligible compared to other sources of uncertainity and that the effect of temperature is brought about by population change of the different tautomers. 118  A program has been developed (Table S6) using GNU Octave 4.0 from https://www.gnu.org/software/octave/ using the routines available in the optimization package. 119 Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., J.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Keith, T.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; 232   Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. G. Inc., Wallingford CT 2013. 120 (a) Kohn, W.; Sham, L. J. Phys. Rev. 1965, 140, A1133-A1138; (b) Pople, J. A.; Gill, P. M. W.; Johnson, B. G. Chem. Phys. Lett. 1992, 199, 557-560. 121 (a) Tomasi, J.; Mennucci, B.; Cancès, E. J. Mol. Struct. THEOCHEM 1999, 464, 211-226; (b) Tomasi, J.; Mennucci, B.; Cammi, R. Chem. Rev. 2005, 105, 2999-3094; (c) Scalmani, G.; Frisch, M. J. J. Chem. Phys. 2010, 132, 114110. 122 (a) Bauernschmitt, R.; Ahlrichs, R. Chem. Phys. Lett. 1996, 256, 454-464; (b) Casida, M. E.; Jamorski, C.; Casida, K. C.; Salahub, D. R. J. Chem. Phys. 1998, 108, 4439-4449. 123 Adamo, C.; Barone, V. J. Chem. Phys. 1998, 108, 664-675. 124 Schwabe, T.; Grimme, S. Phys. Chem. Chem. Phys. 2006, 8, 4398-4401. 125 Zhao, Y.; Truhlar, D. G. Theor. Chem. Acc. 2008, 120, 215-241. 126 Wheeler, S. E.; Houk, K. N. J. Chem. Theory Comput. 2010, 6, 395-404. 127 Alecu, I. M.; Zheng, J.; Zhao, Y.; Truhlar, D. G. J. Chem. Theory Comput. 2010, 6, 2872-2887. 128 Laury, M. L.; Carlson, M. J.; Wilson, A. K. J. Comput. Chem. 2012, 33, 2380-2387. 129 Kesharwani, M. K.; Brauer, B.; Martin, J. M. L. J. Phys. Chem. A 2015, 119, 1701-1714. 130 (a) Helgaker, T.; Watson, M.; Handy, N. C. J. Chem. Phys. 2000, 113, 9402-9409; (b) Deng, W.; Cheeseman, J. R.; Frisch, M. J. J. Chem. Theory Comput. 2006, 2, 1028-1037. 131 London, F. J. Phys. le Radium 1937, 8, 397-409. 132 Chesnut, D. B.; Moore, K. D. J. Comput. Chem. 1989, 10, 648-659. 133 Dunning, T. H. J. Chem. Phys. 1989, 90, 1007-1023. 233   134 Singh, U. C.; Kollman, P. A. J. Comput. Chem. 1984, 5, 129-145. 135 (a) Baret, P.; Moneta, W.; Pierre, J. L. Acta Cryst. 1988, 44, 1254-1256; (b) Brooker, S. Coord. Chem. Rev. 2001, 222, 33-56; (c) Gao, J.; Reibenspies, J. H.; Zingaro, R. A.; Woolley, R.; Martell, A. E.; Clearfield, A. Inorg. Chem. 2005, 44, 232-241; (d) Balić, T.; Matasović, B.; Marković, B.; Šter, A.; Štivojević, M.; Matković-Čalogović, D. J. Incl. Phenom. Macrocycl. Chem. 2016, 85, 217-226; (e) Balić, T.; Matasović, B.; Jaźwiński, J.; Matković-Čalogović, D. Microporous Mesoporous Mater. 2016, 226, 53-60. 136 Dash, A. C.; Dash, B.; Praharai, S. J. Chem. Soc., Dalton Trans. 1981, 10, 2063-2069. 137 (a) Gallant, A. J.; Chong, J. H.; MacLachlan, M. J. Inorg. Chem. 2006, 45, 5248-5250; (b) Frischmann, P. D.; Gallant, A. J.; Chong, J. H.; MacLachlan, M. J. Inorg. Chem. 2008, 47, 101-112. 138 (a) Li, X.; Li, B.; Chen, L.; Hu, J.; Wen, C.; Zheng, Q.; Wu, L.; Zeng, H.; Gong, B.; Yuan, L. Angew. Chem. Int. Ed. 2015, 54, 11147-11152; (b) Jiang, J.; Dong, R. Y.; MacLachlan, M. J. Chem. Commun. 2015, 51, 16205-16208. 139 (a) Lai, P. -S.; Dubland, J. A.; Sarwar, M. G.; Chudzinski, M. G. Taylor, M. S. Tetrahedron 2001, 67, 7586-7592; (b) Ross, N. A.; MacGregor, R. R.; Bartsch, R. A. Tetrahedron 2004, 60, 2035-2041; (c) Shirai, S. -Y.; Nara, H.; Kayaki, Y.; Ikariya, T. Organometallics 2009, 28, 802-809; (d) Shi, Z.; Suri, M.; Glorius, F. Angew. Chem., Int. Ed. 2013, 52, 4892-4896. 140 (a) Samec, J. S. M.; Éll, A. H.; Åberg, J. B.; Privalov, T.; Eriksson, L.; Bäackvall, J. -E. J. Am. Chem. Soc. 2006, 128, 14293-14305; (b) Lefranc, J.; Tetlow, D. J.; Donnard, M.; Minassi, A.; Gálvez, E.; Clayden, J. Org. Lett. 2011, 13, 296-299; (c) Tait, M.; Donnard, M.; Minassi, A.; Lefranc, J.; Bechi, B.; Carbone, G.; O’Brien, P.; Clayden, J. Org. Lett. 2013, 15, 34-37; (d) Wakchaure, V. N.; Kaib, P. S. J.; Leutzsch, M.; List, B. Angew. Chem., Int. Ed. 2015, 54, 11852-234   11856. 141 Gautier, F. -M.; Jones, S.; Martin, S. J. Org. Biomol. Chem. 2009, 7, 229-231. 142 Jiang, J.; MacLachlan, M. J. Org. Lett. 2010, 12, 1020-1023. 143 Barluenga, J.; Jiménez-Aquino, A.; Aznar, F.; Valdés, C. J. Am. Chem. Soc. 2009, 131, 4031-4041. 144  Frischmann, P. D.; Jiang, J.; Hui, J. K. -H.; Grzybowski, J.; MacLachlan, M. J. Org. Lett. 2008, 10, 1255-1258. 145 Marini, A.; Muñoz-Losa, A.; Biancardi, A.; Mennucci, B. J. Phys. Chem. B 2010, 114, 17128-17135. 146 Fischmann, S.; Lüning, U. Isr. J. Chem. 2013, 53, 87-96. 147 Carnall, J. M.; Waudby, C. A.; Belenguer, A. M.; Stuart, M. C.; Peyralans, J. J.; Otto, S. Science 2010, 327, 1502-1506. 148 Hu, J.; Adogla, E. A.; Ju, Y.; Fan, D.; Wang, Q. Chem. Commun. 2012, 48, 11256-11258.  149 (a) Moretti, I.; Torre, G. Synthesis 1970, 1970, 141; (b) Cobas, A.; Guitian, E.; Castedo, L. J. Org. Chem. 1993, 58, 3113-3117; (c) He, R.; Jin, X.; Chen, H.; Huang, Z. -T.; Zheng, Q. -Y.; Wang, C. J. Am. Chem. Soc., 2014, 136, 6558-6561; (d) Wakchaure, V. N.; Kaib, P. S. J.; Leutzsch, M.; List, B. Angew. Chem. Int. Ed. 2015, 54, 11852-11856; (e) Chambers, S. J.; Coulthard, G.; Unsworth, W. P.; O’Brien, P.; Taylor, R. J. K. Chem. Eur. J. 2016, 22, 6496-6500. 150 (a) Blacker, A. J.; Roy, M.; Hariharan, S.; Headley, C.; Upare, A.; Jagtap, A.; Wankhede, K.; Mishra, S. K.; Dube, D.; Bhise, S.; Vishwasrao, S.; Kadam, N. Org. Process Res. Dev. 2011, 15, 331-338; (b) Ickes, A. R.; Ensign, S. C.; Gupta, A. K.; Hull, K. L. J. Am. Chem. Soc. 2014, 136, 11256-11259. 235   151 Guieu, S.; Crane, A. K.; MacLachlan, M. J. Chem. Commun. 2011, 47, 1169-1171. 152 Lehn, J. M. Science 1993, 260, 1762-1763. 153 Wong, W. W. H.; Ma, C.-Q.; Pisula, W.; Yan, C.; Feng, X.; Jones, D. J.; Müllen, K.; Janssen, R. A. J.; Bäuerle, P.; Holmes, A. B. Chem. Mater. 2010, 22, 457-466. 154 Yin, M.; Shen, J.; Pisula, W.; Liang, M.; Zhi, L.; Müllen, K. J. Am. Chem. Soc. 2009, 131, 14618-14619 155 van Hameren, R.; van Buul, A. M.; Catriciano, M. A.; Villari, V.; Micali, N.; Schön, P.; Speller, S.; Scolaro, L. M.; Rowan, A. E.; Elemans, J. A. A. W.; Nolte, R. J. M. Nano Lett. 2008, 8, 253-259. 156 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. 157 Northrop, B. H.; Zheng, Y.-R.; Chi, K.-W.; Stang, P. J. Acc. Chem. Res. 2009, 42, 1554-1563 158  (a) Stang, P. J.; Cao, D. H. J. Am. Chem. Soc. 1994, 116, 4981-4982; (b) Northrop, B. H.; Yang, H.-B.; Stang, P. J. Chem. Commun. 2008, 5896-5908; (c) Lee, J.; Ghosh, K.; Stang, P. J. J. Am. Chem. Soc. 2009, 131, 12028-12029. 159  Motschi, H.; Nussbaumer, C.; Pregosin, P. S.; Bachechi, F.; Mura, P.; Zambonelli, L. Helv. Chim. Acta 1980, 63, 2071-2086. 160  (a) Crowley, J. D.; Goshe, A. J.; Bosnich, B. Chem. Commun. 2003, 392-393; (b) Crowley, J. D.; Steele, I. M.; Bosnich, B. Inorg. Chem. 2005, 44, 2989-2991; (c) Crowley, J. D.; Steele, I. M.; Bosnich, B. Eur. J. Inorg. Chem. 2005, 3907-3917.  161 Frischmann, P. D.; Guieu, S.; Tabeshi, R.; MacLachlan, M. J.  J. Am. Chem. Soc. 2010, 132, 7668-7675. 162 Frischmann, P. D.; Sahli, B. J.; Guieu, S.; Patrick, B. O.; MacLachlan, M. J. Chem. Eur. J. 2012, 236   18, 13712-13721. 163 Ma, C.; Lo, A.; Abdolmaleki, A.; MacLachlan, M. J. J. Org. Lett. 2004, 6, 3841-3844. 164 Chinchilla, R.; Nájera, C. Chem. Rev. 2007, 107, 874-922. 165 Thomas, R.; Lakshmi, S.; Pati, S. K.; Kulkarni, G. U. J. Phys. Chem. B 2006, 110, 24674-24677. 166 Lin, W.; Jiang, H. J. Am. Chem. Soc. 2003, 125, 8084-8085. 167 Kline, M. A.; Wei, X.; Horner, I. J.; Liu, R.; Chen, S.; Chen, S.; Yung, K. Y.; Yamato, K.; Cai, Z.; Bright, F. V.; Zeng, X. C.; Gong, B. Chem. Sci. 2015, 6, 152-157. 168 Wei, X.; Zhang, G.; Shen, Y.; Zhong, Y.; Liu, R.; Yang, N.; Al-mkhaizim, F. Y.; Kline, M. A.; He, L.; Li, M.; Lu, Z. -L.; Shao, Z.; Gong, B. J. Am. Chem. Soc. 2016, 138, 2749-2754. 169 (a) Chan, S. I.; Schweizer, M. P.; Ts’o, P. O. P.; Helmkamp, G. K. J. Am. Chem. Soc. 1964, 86, 4182-4188; (b) Zhang, J.; Moore, J. S. J. Am. Chem. Soc. 1992, 114, 9701-9702; (c) Zhao, D.; Moore, J. S. J. Org. Chem. 2002, 67, 3548-3554. 170 (a) Che, C. -M.; Kwok, C. -C.; Lai, S. -W.; Rausch, A. F.; Finkenzeller, W. J.; Zhu, N.; Yersin, H. Chem. Eur. J. 2010, 16, 233-247; (b) Zhang, B.; Li, Y.; Sun, W. Eur. J. Inorg. Chem. 2011, 2011, 4964-4969. 171 (a) Birks, J. B. Photophysics of Aromatic Molecules, New York: John Wiley, 1970; (b) Kawaoka, K.; Khan, A. U.; Kearns, D. R. J. Chem. Phys. 1967, 46, 1842-1853; (c) Gijzeman, O. L. J.; Kaufman, F.; Porter, G. Faraday Trans, II 1973, 69, 708-720. 172 (a) M. Belletête, J. Bouchard, M. Leclerc and G. Durocher, Macromolecules 2005, 38, 880-887; (b) Borisov, S. M.; Wolfbeis, O. S. Chem. Rev. 2008, 108, 423-461. 173 Shetty, A. S.; Zhang, J.; Moore, J. S. J. Am. Chem. Soc. 1996, 118, 1019-1027. 174 Tobe, Y.; Utsumi, N.; Kawabata, K.; Nagano, A.; Adachi, K.; Araki, S.; Sonoda, M.; Hirose, K.; Naemura, K. J. Am. Chem. Soc. 2002, 124, 5350-5364. 237   175 Basílio, N.; Garcia-Rio, L.; Martín-Pastor, M. Langmuir 2012, 28, 2404-2414. 176 (a) Chen, H. C.; Chen, S. H. J. Phys. Chem. 1984, 88, 5118-5121; (b) Evans, R.; Deng, Z.; Rogerson, A. K.; McLachlan, A. S.; Richards, J. J.; Nilsson, M.; Morris, G. A. Angew. Chem., Int. Ed. 2013, 52, 3199-3202. 177 Neufeld, R.; Stalke, D. Chem. Sci. 2015, 6, 3354-3364. 178 (a) Cabrita, E. J.; Berger, S.; Magn. Reson. Chem. 2001, 39, 142-148; (b) Yao, S.; Howlett, G.; Norton, R.; J. Biomol. NMR 2000, 16, 109-119. 179 (a) Durand, E.; Clemancey, M.; Lancelin, J.-M.; Verstraete, J.; Espinat, D.; Quoineaud, A.-A.  J. Phys. Chem. C 2009, 113, 16266-16276; (b) Jones, J.; Wilkins, D.; Smith, L.; Dobson, C. J. Biomol. NMR 1997, 10, 199-203. 180 Zhu, Y. -Y.; Li, C.; Li, G. -Y.; Jiang, X. -K.; Li, Z. -T. J. Org. Chem. 2008, 73, 1745-1751. 181 (a) Zwahlen, C.; Legault, P.; Vincent, S. J. F.; Greenblatt, J.; Konrat, R.; Kay, L. E. J. Am. Chem. Soc. 1997, 119, 6711-6721; (b) Herrmann, T.; Guntert, P.; Wuthrich, K. J. Mol. Biol. 2002, 319, 209-227. 182 (a) Mizuno, Y.; Aida, T.; Yamaguchi, K. J. Am. Chem. Soc. 2000, 122, 5278-5285; (b) Syakaev, V. V.; Shalaeva, Y. V.; Kazakova, E. K.; Morozova, Y. E.; Makarova, N. A.; Konovalov, A. I.  Colloid J. 2012, 74, 346-355; (c) Dong, S.; Yuan, J.; Huang, F. Chem. Sci. 2014, 5, 247-252. 183 Zhu, J.; Parra, R. D.; Zeng, H.; Skrzypczak-Jankun, E.; Zeng, X. C.; Gong, B. J. Am. Chem. Soc. 2000, 122, 4219-4220. 184 (a) Sandström, J. Dynamic NMR Spectroscopy Academic: London, 1982, (b) Gong, H. -Y.; Rambo, B. M.; Karnas, E.; Lynch, V. M.; Sessler, J. L. Nat. Chem. 2010, 2,  406-409. 185 Janiak, C. J. Chem. Soc., Dalton Trans. 2000, 3885-3896. 186 Holmes, B.T.; Pennington, W. T.; Hanks, T. W. Synth. Commun. 2003, 33, 2447-2461. 238   187 Djukic, B.; Poddutoori, P. K.; Dube, P. A.; Seda, T.; Jenkins, H. A.; Lemaire, M. T. Inorg. Chem. 2009, 48, 6109-6116. 188 Kaae, B. H.; Harpsøe, K.; Kvist, T.; Mathiesen, J. M.; Mølck, C.; Gloriam, D.; Jimenez, H. N.; Uberti, M. A.; Nielsen, S. M.; Nielsen, B.; Bräuner-Osborne, H.; Sauerberg, P.; Clausen, R. P.; Madsen, U. ChemMedChem 2012, 3, 440-451. 189 Li, J.; Wu, Y.; Song, F.; Wei, G.; Cheng, Y.; Zhu, C. J. Mater. Chem. 2012, 22, 478-482. 190 (a) Gawroński, J.; Kolbon, H.; Kwit, M.; Katrusiak, A. J. Org. Chem. 2000, 65, 5768-5773; (b) Akine, S.; Taniguchi, T.; Nabeshima, T. Tetrahedron Lett. 2001, 42, 8861-8864; (c) Gallant, A. J.; MacLachlan, M. J. Angew. Chem. Int. Ed. 2003, 42, 5307-5310; (d) Frischmann, P. D.; Jiang, J.; Hui, J. K. -H.; Grzybowski, J. J.; MacLachlan, M. J. Org. Lett. 2008, 10, 1255-1258. (e) Ma, C.; Lo, A.; Abdolmaleki, A.; MacLachlan, M. J. Org. Lett. 2004, 6, 3841-3844. (f) Hui, J. K. -H.; MacLachlan, M. J. Chem. Commun. 2006, 23, 2480-2482. 191 (a) Qin, B.; Chen, X. Y.; Fang, X.; Shu, Y. Y.; Yip, Y. K.; Yan, Y.; Pan, S. Y.; Ong, W. Q.; Ren, C. L.; Su, H. B.; Zeng, H. Q. Org. Lett. 2008, 10, 5127-5130; (b) Qin, B.; Ren, C. L.; Ye, R. J.; Sun, C.; Chiad, K.; Chen, X. Y.; Li, Z.; Xue, F.; Su, H. B.; Chass, G. A.; Zeng, H. Q. J. Am. Chem. Soc. 2010, 132, 9564-9566; (c) Qin, B.; Jiang, L. Y.; Shen, S.; Sun, C.; Yuan, W. X.; Li, S. F. Y.; Zeng, H. Q. Org. Lett. 2011, 13, 6212-6215; (d) Ren, C. L.; Xu, S. Y.; Xu, J.; Chen, X. Y.; Zeng, H. Q. Org. Lett. 2011, 15, 3840-3843; (e) Ren, C. L.; Maurizot, V.; Zhao, H. Q.; Shen, J.; Zhou, F.; Ong, W. Q.; Du, Z. Y.; Zhang, K.; Su, H. B.; Zeng, H. Q. J. Am. Chem. Soc. 2011, 133, 13930-13933; (f) Fu, H. L.; Chang, H.; Shen, J.; Yu, L.; Qin, B.; Zhang, K.; Zeng, H. Q. Chem. Commun. 2014, 50, 3582-3584; (g) Ren, C. L.; Shen, J.; Zeng, H. Q. Org. Lett. 2015, 17, 5946-5949; (h) Lee, S.; Chen, C. -H.; Flood, A. H. Nat. Chem. 2013, 5, 704-710; (i) Dethe, D. H.; Srikrishna, A. Tetrahedron Lett. 2005, 46, 3381-3383. 239   192 Frischmann, P. D.; Guieu, S.; Tabeshi, R.; MacLachlan, M. J. J. Am. Chem. Soc. 2010, 132, 7668-7675. 193 (a) Gallant, A. J; Chong, J. H.; MacLachlan, M. J. J. Inorg. Chem. 2006, 45, 5248-5250; (b) Frischmann, P. D.; MacLachlan, M. J. Chem. Commun. 2007, 43, 4480-4482; (c) Frischmann, P. D.; Gallant, A. J.; Chong, J. H.; MacLachlan, M. J. Inorg. Chem. 2008, 47, 101-112. 194 (a) Gallant, A. J.; Chong, J. H.; MacLachlan, M. J. Inorg. Chem. 2006, 45, 5248-5250; (b) Frischmann, P. D.; Gallant, A. J.; Chong, J. H.; MacLachlan, M. J. Inorg. Chem. 2008, 47, 101-112. 195 Carnall, J. M.; Waudby, C. A.; Belenguer, A. M.; Stuart, M. C.; Peyralans, J. J.; Otto, S. Science 2010, 327, 1502-1506. 196 Fischmann, S.; Lüning, U. Isr. J. Chem. 2013, 53, 87-96. 197 (a) Curtis, N. F.; House, D. A. Chem. & Ind. 1961, 42, 1708-1709; (b) J. D. Curry, D. H. Busch, J. Am. Chem. Soc. 1964, 86, 592-594; (c) Cook, D. H.; Fenton, D. E.; Drew, M. G. B.; McFall, S. G.; Nelson, S. M.; J. Chem. Soc., Dalton Trans. 1977, 446-449. 198 Chandrasekhar, S.; Sadashiva, B. K.; Surwh, K. A. Pramana 1977, 9, 471-480. 199 (a) Dalcanale, E.; Du Vosel, A.; Levelut, A. M. Liq. Cryst. 1991, 10, 185-198; (b) Bonsignore, S.; Cometti, G.; Dalcanale, E.; Du Vosel, A. Liq. Cryst. 1990, 8, 639-649; (c) Kawano, S. -I.; Ishida, Y.; Tanaka, K. J. Am. Chem. Soc. 2015, 137, 2295-2302; (d) Li, X.; Li, B.; Chen, L.; Hu, J.; Wen, C.; Zheng, Q.; Wu, L.; Zeng, H. Gong, B.; Yuan, L. Angew. Chem. Int. Ed. 2015, 54, 11147-11152; (e) Kawano, S. -I.; Hamazaki, T.; Suzuki, A.; Kurahashi, K.; Tanaka, K. Chem. Eur. J. 2016, 22, 1-11. 200 Zhang, J.; Moore, J. S. J. Am. Chem. Soc. 1994, 116, 2655-2656. 201 (a) Yue, N. L. S.; Eisler, D. J.; Jennings, M. C.; Puddephatt, R. J. Inorg. Chem. 2004, 43, 7671-240   7681; (b) Danjo, H.; Hirata, K.; Yoshigai, S.; Azumaya, I.; Yamaguchi, K. J. Am. Chem. Soc. 2009, 131, 1638-1639; (c) Dong, S.; Zheng, B.; Xu, D.; Yan, X.; Zhang, M.; Huang, F. A. Adv. Mater. 2012, 24, 3191-3195; (d) Xue, M.; Yang, Y.; Chi, X.; Zhang, Z.; Huang, F. Acc. Chem. Res. 2012, 45, 1294-1308. 202 (a) Qin, B.; Ren, C. L.; Ye, R. J.; Sun, C.; Chiad, K.; Chen, X. Y.; Li, Z.; Xue, F.; Su, H. B.; Chass, G. A.; Zeng, H. Q. J. Am. Chem. Soc. 2010, 132, 9564-9566; (b) Qiao, B.; Anderson, J. R.; Pink, M.; Flood, A. H. Chem. Commun. 2016, 52, 8683-8686.  203 (a) Lin, W.; Jiang, H. J. Am. Chem. Soc. 2003, 125, 8084-8085; (b) Chan, Y.-T.; Li, X.; Yu, J.; Carri, G. A.; Moorefield, C. N.; Newkome, G. R.; Wesdemiotis, C. J. Am. Chem. Soc. 2011, 133, 11967-11976. 204 (a) Xin, P.; Zhang, L.; Su, P.; Hou, J. -L.; Li, Z. -T. Chem. Commun. 2015, 51, 4819-4822; (b) Si, W.; Xin, P.; Li, Z. -T.; Hou, J. -L. Acc. Chem. Res. 2015, 48, 1612-1619; (c) Wei, X.; Zhang, G.; Shen, Y.; Zhong, Y.; Liu, R.; Yang, N.; Al-mkhaizim, F. Y.; Kline, M. A.; He, L.; Li, M.; Lu, Z. -L.; Shao, Z.; Gong, B. J. Am. Chem. Soc. 2016, 138, 2749-2754. 205 Zhao, D.; Moore, J. S. Chem. Commun. 2003, 7, 807-818. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0342813/manifest

Comment

Related Items