1 RATIONAL DESIGN OF ONE DIMENSIONAL COLUMNAR SELF ASSEMBLED NANOSTRUCTURE S FOR APPLICATIONS AS ORGANIC SEMICONDUCTIVE MATERIALS By BENJAMIN M. SCHULZE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2013
2 2013 Benjamin M. Schulze
3 To science: without it, we are ignorant, with it, we have to write 300 page dissertations
4 A CKNOWLEDGMENTS mother, without whose dedicated patience, unconditional love, and everlasti ng support I would be a disastrous mess. While neither is scientifically inclined and I sometimes became annoyed at the repetitive explanations, they relentlessly inquired about my research and classes (especially my father). I now realize that they simp ly, yet genuinely, cared about me, even while battling through cancer and cancer scares, something not quantifiable by my scientific mind. For that and for them, I am most sincerely and eternally thankful. I cannot write any further without thanking my ad visor, Dr. Ron Castellano, whose capacity for sarcastic wit is only surpassed by his tireless pursuit of knowledge. It is the combination of these two traits that make him not only a great PI, but also a great friend. Wearing his emotions on his sleeve, it was easy to tell when the mood was ** molding me into the scientist I am now, which was NOT easy, I am definitely thankful. I need to thank my brothers, Eric and Pat. Eric, the first PhD of our family, provided the initial impetus to even join grad school. While he frustrates me with his need for empirical data, he always pushes me to be a bett er scientist. Pat, thank you for not being Eric. Monica, you have stuck with me through thick and thin, hell and high water, and our journey has only just begun. To the best listener (without having a clue of what I a
5 talking about 98.4% of the time), my best friend, and my love, thank you. I am not I would also like to thank my time tested best friend, my dog, Edison. While your antics are both adorable and annoying, your companionship will never be forgotten during these p ast four years. Lastly, and perhaps most importantly, I would like to thank the people most directly responsible for keeping me sane during this time, my labmates. I need to thank Dr. Matt Baker specifically for getting me to think like a physical organ ic chemist, being a work out partner, and being my own personal Dr. Phil. Thanks to Mike Meese for introducing me to the world of soccer and microbrews. I would to thank Ashton Bartley whose southern accent and cooking always brightened my day (I hope yo u realize, you are one of the closest things I have ever had to a little sister). Also, thanks to my I ever wanted to know about New York sports franchises). Reggie (Raghi da Bou Zerdan), thanks for your companionship and all help you have given throughout the years. Others worthy of thanks are Dr. Pam Cohn, Dr. Yan Li, and anyone else that has helped me throughout this endeavor Further thanks are due to the Air Force for funding my journey and the Department of Chemistry at the US Air Force Academy for all the support they have given along the way. Also, thanks to Research Corporation for Scien ce Advancement for further funding.
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 9 LIST OF FIGURES ................................ ................................ ................................ ........ 10 LIST OF ABBREVIATIONS ................................ ................................ ........................... 16 ABSTRACT ................................ ................................ ................................ ................... 19 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 21 Organic Semiconductors ................................ ................................ ......................... 21 One Dimens ional (1D) Columnar Nanostructures ................................ ................... 23 Self Assembly and 1D Nanostructures ................................ ................................ ... 27 Strategies and Interactions ................................ ................................ ..................... 28 Self Assembly and Discotic Liquid Crystals ................................ ............................ 33 Self assembled 1D Systems in Device Contexts ................................ .................... 35 Scope and Outline of Dissertation ................................ ................................ .......... 38 2 DONOR ACCEPTOR COLUMNAR ASSEMBLI ES OF BENZOTRIFURAN: CHARGE TRANSFER (CT) COMPLEXATION AND MESOMORPHISM ............... 41 Background and Previous Work of the Benzotrifuranone Scaffold ......................... 41 Synthesis of Donor and Acceptor Compounds ................................ ....................... 45 Purification of Acylated Target Compounds ................................ ............................ 47 Synthesis of Mellitic Triimides ................................ ................................ ................. 49 Spectroscopic Analysis of CT Complexation of 1:1 Mixtures ................................ .. 50 Characterization of Thermotropic Mesomorphism of 1:1 Mixtures .......................... 55 X ray Diffraction Analysis ................................ ................................ ........................ 63 Summary ................................ ................................ ................................ ................ 64 Experimental ................................ ................................ ................................ ........... 65 Materials and General Methods ................................ ................................ ....... 65 Synthetic Procedures ................................ ................................ ....................... 67 Characteriza tion of CT Complex by Ultraviolet Visible (UV Vis) Spectrometry ................................ ................................ ................................ 76 Characterization of Mixtures by Polarizing Optical Microscopy (POM) ............. 78 Differential Scanning Calorimetry (DSC) Analysis ................................ ............ 79 Thermogravimetric Analysis (TGA) ................................ ................................ .. 80 Analysis of Spin Cast Film by X ray Diffra ction (XRD) ................................ ..... 80
7 3 IMPROVED PERFORMANCE, BULK ORGANIZATION, AND CHARGE TRANSPORT OF A SMALL MOLECULE OPV DEVICE THROUGH HYDROGEN BONDED 1D COLUMNAR AGGREGATES ................................ ...... 82 Background ................................ ................................ ................................ ............. 82 General Approach and Experimental Design ................................ .......................... 86 Synthesis of Hydrogen Bonding and Chromo phore Modules ................................ 91 Synthesis of Phthalocyanine Module ................................ ................................ 91 Synthesis of Branched Quaterthiophene Module ................................ ............. 95 Synthesis of Phthalhydrazide Module and Subsequent Functionalization ........ 97 Alternative Strategies and Reactions for Linking Phthalhydrazide and Chromophore Modules ................................ ................................ ................ 100 Structure Property Relationship Findings ................................ ............................. 105 Intrinsic Properties of Neat Materials ................................ .............................. 105 Bulk and Solution Based Aggregation Behavior ................................ ............. 106 Device Based Performance and Characterization Results ................................ ... 113 Summary ................................ ................................ ................................ .............. 117 Experimental ................................ ................................ ................................ ......... 118 Materials and Gene ral Methods ................................ ................................ ..... 118 Synthetic Procedures ................................ ................................ ..................... 122 4 SIZE AND SHAPE APPROXIMATION OF SUPRAMOLECULAR ASSEMBLIES THROUGH VARIABLE TEMPERATURE DIFFUSION ORDERED SPECTROSCOPY (DOSY NMR) ................................ ................................ ......... 139 Background ................................ ................................ ................................ ........... 139 The Modified Stokes Einstein Equation ................................ .......................... 140 Using DOSY NMR to Characterize Supramolecular Assemblies ................... 143 Experimental ................................ ................................ ................................ ......... 146 Equipment, Materia ls, and Sample Preparation ................................ ............. 146 Data Collection Procedure ................................ ................................ .............. 147 Results and Discussion ................................ ................................ ......................... 148 Data Analysis and Example Fitting Procedure ................................ ............... 148 Fitting Results ................................ ................................ ................................ 150 Summary and Future Work ................................ ................................ ................... 153 Miscellaneous Experimental Information ................................ .............................. 155 Synthetic Procedures ................................ ................................ ..................... 155 Viscosity Temperature Fitting ................................ ................................ ......... 158 Di ffusion Versus Temperature Data with Calculated Viscosities .................... 159 5 CONCLUSION ................................ ................................ ................................ ...... 162 Donor Acceptor Columnar Assemblies of Benzotrifuran: Charge Transfer Complexation and Mesomorphism ................................ ................................ .... 163 Improved Performance, Bulk Organization, and Charge Transport of a Small Molecule OPV Device through Rational Design of Hydrogen Bonded 1D Columnar Aggregates ................................ ................................ ........................ 165
8 Size and Shape Approximation of Supramolecular Assemblies of One Dimensional Columnar Nanostructures through Variable Temperature Diffusion Ordered Spectros copy ................................ ................................ ........ 168 APPENDIX A NUCLEAR MAGNETIC RESONANCE (NMR) SPECTRA ................................ .... 172 B MISCELLANEOUS DATA ................................ ................................ ..................... 220 C SYNTHETI C PROCEDURES ................................ ................................ ............... 245 LIST OF REFERENCES ................................ ................................ ............................. 273 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 282
9 LIST OF TABLES Table page 2 1 Various cyclization reaction conditions and results as performed and reported by previous Castellano group members (Yan Li and Matt Baker) ....................... 47 3 1 Optical, electrochemical, and computational characteristics for compounds 3 5a 3 6a and 3 7 ................................ ................................ .............................. 106 3 2 Characteristics of BHJ OPV dev ices containing 3 6a 3 7 and 3 5a as the donor and C 60 as the acceptor in 3:2 weight ratio ................................ ............. 114 4 1 Fitted molecular size and shape parameters for unaggregated 4 1a,b solutions (approximately 27 mM in dimethylsulfoxide d 6 or DMSO d 6 ) ............. 150 4 2 Summarized results of fitting parameters for 4 1a,b studied in toluene d 8 ....... 152 4 3 Diffusion versus temperature data with calculated viscosities for 4 1a in toluene d 8 ................................ ................................ ................................ ......... 160 4 4 Diffusion versus temperature data with calculated viscosities for 4 1a in DMSO d 6 ................................ ................................ ................................ .......... 160 4 5 Diffusion versus temperature data with calculated viscosities for 4 1b in toluene d 8 ................................ ................................ ................................ ......... 160 4 6 Diffusion versus temperature data with calculated viscosities for 4 1b in DMSO d 6 ................................ ................................ ................................ .......... 161 B 1 Electrochemical potentials, electrochemical energy gaps, and corresponding highest occupied molecular orbital (HOMO)/lowest unoccupied molecular orbital (LUMO) energies for 3 5a 3 6a and 3 7 ................................ ............... 224 B 2 Manual indexing of 3 5a powder X ray diffraction pattern ................................ 230 B 3 Miscellaneous computation energies and parameters ................................ ...... 236 B 4 HOMO and LUMO wavefunctions for 3 5a 3 6a and 3 7 ................................ 236
10 LIST OF FIGURES Figure page 1 1 Cartoon representation of self assembly of disc shaped molecules ................... 25 1 2 Chemical structures for hexabenzocoronenes ( 1 1 ), pentacene ( 1 2 ), perylene bisimides ( 1 3 ), and naphthalene bisimides ( 1 4 ) ................................ 27 1 3 Cartoon schematic illustrating the electron density of a symmetric delocalized syste m ................................ ................................ ................................ ............ 30 1 4 Chemical structure of 1,3,5 trisbenzenecarboxamide ( 1 5 ) motif ....................... 32 1 5 Chemical structure of cyanurate ( 1 6a ) and melamine ( 1 6b ) derivatives, aggregated hexameric rosette structure ( (1 6a 1 6b) 3 ), oligothiophene ( 1 7 ), and perylene bisimide ( 1 8 ) with amide functionality ................................ .......... 33 1 6 Chemical structures of hexadecylthiotriphenylene ( 1 9 ), epindolidione ( 1 10 ), quinacridone ( 1 11 ), tetra cene ( 1 12 ), and pentacene ( 1 13 ) ............................. 37 1 7 Cartoon representation of donor and acceptor material ................................ ..... 37 2 1 Previous work performed with benzotrifuranone (BTF) ................................ ...... 43 2 2 Chemical structure of 2,4,7 trinitrofluore n 9 one ( 2 3 ) and mellitic triimides (TIMs, 2 4 ) ................................ ................................ ................................ .......... 44 2 3 Cartoon representation of possible donor acceptor columnar aggregation ........ 44 2 4 Synthetic scheme for access to BTF ( 2 1 ) and related derivatives ..................... 46 2 5 Reaction scheme for the transformation of mellitic acid ( 2 13 ) to alkylated mellitic triimides ( 2 14a,b,c ) ................................ ................................ ............... 49 2 6 Images of colorless and colored 10 mM n heptane solutions ............................. 51 2 7 Absorbance and fitting graphs of 1:1 mixture of 2 2c and 2 14c ........................ 54 2 8 Photophysical characteristics and linear fitting of Benesi Hildebrand treatment ................................ ................................ ................................ ............ 55 2 9 Differential scanning calorimetry (DSC) thermograms ................................ ........ 58 2 10 Observed phase transition diagram for 1:1 mixture of 2 2c and 2 14c wit h associated temperatures ................................ ................................ .................... 60 2 11 Polarized optical microscopy (POM) images of 1:1 bulk mixture of 2 2c and 2 14c taken at 100x tot al magnification between crossed polarizers .................. 61
11 2 12 POM images of 1:1 bulk mixture of 2 2c and 2 14c taken at 72 C between crossed polarizers ................................ ................................ .............................. 62 2 13 POM image of 1:1 bulk mixture of 2 12c and 2 14c taken at 40 C at 100x total magnification between crossed pol arizers ................................ .................. 63 3 1 Cartoon representation of bulk heterojunction (BHJ) organic photovoltaic (OPV) device and experimental device efficiencies ................................ ............ 85 3 2 Schematic representation for creating improved OPV device performance ....... 87 3 3 Examples of hydrogen bonding (H bonding) units ................................ .............. 88 3 4 Chemical structures of potential donor chromophore units ................................ 90 3 5 Computationally derived images of the trimeric aggregated tautomer of 3 5a and C 60 ................................ ................................ ................................ ............... 91 3 6 Targeted system consisting of phthalhydraz ide and phthalocyanine ( 3 1a,b ) and subsequent reactions involved in its preparation ................................ ......... 94 3 7 Chemical structures H 3 5 a,b ) and H 3 6a,b and 3 7 ) phthalhydrazide branched quaterthiophene ................................ ......... 95 3 8 Reaction scheme detailing the formation of the Suzuki precursor ( 3 13 ) ........... 97 3 9 Reaction scheme of the phthalhydrazide unit ................................ ..................... 98 3 10 Reaction scheme detailing the S uzuki cross coupling and failed deprotection ... 99 3 11 Reaction scheme detailing the synthesis of the phthalhydrazide module ......... 100 3 12 Reaction scheme showing Sonogashira cross coupling ................................ ... 102 3 13 Reaction scheme depicting the ethynyl linking strategy ................................ ... 103 3 14 Reactio n scheme showing 1,3 dipolar cycloaddition linking strategy ............... 104 3 15 Absorption spectra for H 3 5a and H 3 6a and 3 7 ................................ ................................ ................................ .................... 106 3 16 Bulk characteristics of 3 5a 3 6a and 3 7 ................................ ....................... 107 3 17 Various Fourier trans form infrared (FT IR) spectra for 3 5a 3 6a and 3 7 ...... 109 3 18 X ray diffraction (XRD) patterns ................................ ................................ ........ 111 3 19 Variable temperature proton nuclear magnetic resonance ( 1 H NMR) spectra of a 10 mM toluene d 8 solution of 3 5b ................................ ............................. 113
12 3 20 BHJ OPV device characteristics ................................ ................................ ....... 115 3 21 Electrical properties and thin film morphology of BHJ OPV s ............................ 117 4 1 Chemical structures of 4 1a,b ribbon like aggregate of 4 1a and four different tautomeric forms of 4 1a all capable of H bonded self assembly ....... 140 4 2 Shapes of diffusing entities ................................ ................................ ............... 142 4 3 Computationally derived model (AMBER*) of discrete trimeric aggregate of 4 1a ................................ ................................ ................................ ..................... 145 4 4 Diffusion data for 4 1b ................................ ................................ ...................... 149 4 5 Diffusion data for 4 1a,b ................................ ................................ ................... 152 4 7 Cubic fitting of viscosity versus temperature data for toluene found in the literature. 137 ................................ ................................ ................................ ....... 158 4 8 Cubic fitting of viscosity versus temperature data for DMSO found in the literature. 136 ................................ ................................ ................................ ....... 159 A 1 Proton nuclear magnetic resonance ( 1 H NMR) spectrum (CDCl 3 300 MHz) of 2 2a ................................ ................................ ................................ .................. 173 A 2 1 H NMR spectrum (CDCl 3 300 MHz) of 2 2b ................................ ................... 174 A 3 1 H NMR spectrum (CDCl 3 300 MHz) of 2 2c ................................ ................... 175 A 4 1 H NMR spectrum (CDCl 3 300 MHz) of 2 9 ................................ ..................... 176 A 5 1 H NMR spectrum (CDCl 3 300 MHz) of 2 11 ................................ ................... 177 A 6 1 H NMR spectrum (CDCl 3 300 MHz) of 2 12a ................................ ................. 178 A 7 1 H NMR spectrum (CDCl 3 300 MHz) of 2 12b ................................ ................. 179 A 8 1 H NMR spectrum (CDCl 3 300 MHz) of 2 12c ................................ ................. 180 A 9 1 H NMR spectrum (DMSO d 6 300 MHz) of 2 14a ................................ ............ 181 A 10 1 H NMR spectrum (CDCl 3 300 MHz) of 2 14b ................................ ................. 182 A 11 1 H NMR spectrum (CDCl 3 300 MHz) of 2 14c ................................ ................. 183 A 12 Carbon nuclear magnetic resonance ( 13 C NMR) (CD Cl 3 75 MHz) spectrum of 2 2b ................................ ................................ ................................ .............. 184 A 13 13 C NMR (CDCl 3 75 MHz) spectrum of 2 12b ................................ .................. 185
13 A 14 13 C NMR (DMSO d 6 126 MHz) spectrum of 2 14a ................................ .......... 186 A 15 13 C NMR (CDCl 3 75 MHz) spectrum of 2 14b ................................ .................. 187 A 16 13 C NMR (CDCl 3 75 MHz) spectrum of 2 14c ................................ .................. 188 A 17 1 H NMR spectrum (CDCl 3 300 MHz) of 1:1 mixture ( 2 2c and 2 14c ) after aging ................................ ................................ ................................ ................. 189 A 18 1 H NMR spectrum (CDCl 3 300 MHz) of 1:1 mixture ( 2 2c and 2 14c ) used for DSC study ................................ ................................ ................................ ... 190 A 19 1 H NMR spectrum (CDCl 3 300 MHz) of 3 3e ................................ ................... 191 A 20 13 C NMR (CDCl 3 75 MHz) spectrum of 3 3e ................................ .................... 192 A 21 1 H NMR spectrum (CDCl 3 300 MHz) of 3 3f ................................ .................... 193 A 22 1 H NMR spectrum (CDCl 3 300 MHz) of 3 2 ................................ ..................... 194 A 23 13 C NMR (CDCl 3 75 MHz) spectrum of 3 2 ................................ ...................... 195 A 24 1 H NMR spectrum (CDCl 3 300 MHz) of 3 18 ................................ ................... 196 A 25 13 C NMR (CDCl 3 75 MHz) spectrum of 3 18 ................................ .................... 197 A 26 1 H NMR spectrum (CDCl 3 300 MHz) of 3 19 ................................ ................... 198 A 27 13 C NMR (CDCl 3 75 MHz) spectrum of 3 19 ................................ .................... 199 A 28 1 H NMR spectrum (CDCl 3 300 MHz) of 3 20 ................................ ................... 200 A 29 13 C NMR (CDCl 3 75 MHz) spectrum of 3 20 ................................ .................... 201 A 30 1 H NMR spectrum (CDCl 3 300 MHz) of 3 23 ................................ ................... 202 A 31 1 H NMR spectrum (CDCl 3 300 MHz) of 3 24a ................................ ................. 203 A 32 1 H NMR spectrum (CDCl 3 300 MHz) of 3 24b ................................ ................. 204 A 33 13 C NMR (CDCl 3 75 MHz) spectrum of 3 24a and 3 24b ................................ 205 A 34 1 H NMR spectrum (CDCl 3 300 MHz) of 3 28 ................................ ................... 206 A 35 13 C NMR (CDCl 3 75 MHz) spectrum of 3 28 ................................ .................... 207 A 37 1 H NMR spectrum (CDCl 3 300 MHz) of 3 29b ................................ ................. 209 A 38 1 H NMR spectrum (CDCl 3 300 MHz) of 3 30 ................................ ................... 2 10
14 A 39 13 C NMR (CDCl 3 75 MHz) spectrum of 3 30 ................................ .................... 211 A 40 1 H NMR spectrum (CDCl 3 300 MHz) of 3 31 ................................ ................... 212 A 41 13 C NMR (CDCl 3 75 MHz) spectrum of 3 31 ................................ .................... 213 A 42 1 H NMR spectrum (CDCl 3 300 MHz) of 3 32a ................................ ................. 214 A 43 1 H NMR spectrum (CDCl 3 500 MHz) of 3 32b ................................ ................. 215 A 44 13 C NMR (CDCl 3 125 MHz) spectrum of 3 32b ................................ ................ 216 A 45 1 H NMR spectrum (CDCl 3 500 MHz) of 3 34a ................................ ................. 217 A 46 1 H NMR spectrum (CDCl 3 500 MHz) of 3 34b ................................ ................. 218 A 47 1 H NMR spectrum (CDCl 3 500 MHz) of 4 4 ................................ ..................... 219 B 1 POM image taken of 2 2c on cooling with crossed polarizers on at 100x total magnification ................................ ................................ ................................ .... 220 B 2 POM image taken on heating at 100x total magnification with crossed polarizers on of 2 2c ................................ ................................ ......................... 221 B 3 POM images of 2 14c taken on heating with crossed polarizers on at 100x total magnification ................................ ................................ ............................. 222 B 4 POM image of a 1:1 mixture of 2 2a and 2 14b taken at room temperature at 100x total magnification ................................ ................................ .................... 222 B 5 X ray diffractogram of a 1:1 mixture of 2 2c and 2 14c ................................ .... 223 B 6 Electrochemical data for 3 6a ................................ ................................ ........... 225 B 7 Electrochemical data for 3 5a ................................ ................................ ........... 225 B 8 Electrochemical data for 3 7 ................................ ................................ ............. 225 B 9 Thermogravimetric analysis (TGA) scan of 3 6a ................................ .............. 226 B 10 TGA scan of 3 5a ................................ ................................ ............................. 227 B 11 DSC trace of 3 6a with peak labels ................................ ................................ .. 228 B 12 DSC trace of 3 5a with peak labels ................................ ................................ .. 229 B 13 Expanded FT IR spectrum for 3 6a ................................ ................................ .. 231 B 14 Expanded FT IR spectrum of 3 7 ................................ ................................ ..... 232
15 B 15 Expanded FT IR spectrum of 3 5a ................................ ................................ ... 232 B 16 Atomic force microscopy (AFM) data ................................ ................................ 233 B 17 The overall light absorption efficiency of the organi c photovoltaic cells containing approximately 40 nm thick donor:C 60 (3:2 by weight) photoactive layers ................................ ................................ ................................ ................ 234 B 18 Curve fitting for t he determination of the charge collection length at short circuit ................................ ................................ ................................ ................ 235 B 19 Screenshot of Microsoft Excel spreadsheet used to fit 4 1b in dimethylsulfoxide d 6 (DMSO d 6 )using the prolate spheroid model ................... 242 C 1 Proton nuclear magnetic resonance ( 1 H NMR) spectrum of compound 3 12a (300 MHz, chloroform d or CDCl 3 ) ................................ ................................ .... 253 C 2 1 H NMR spectrum of compound 3 14a (300 MHz, CDCl 3 ) ............................... 254 C 3 Carbon nuclear magnetic resonance ( 13 C NMR) spectrum of compound 3 14a (75 MHz, CDCl 3 ) ................................ ................................ ........................ 255 C 4 1 H NMR spectrum of compound 3 16a (300 MHz, CDCl 3 ) ............................... 256 C 5 13 C NMR spectrum of compound 3 16a (75 MHz, CDCl 3 ) ................................ 257 C 6 1 H NMR spectrum of compound 3 6a (300 MHz, CDCl 3 ) ................................ 258 C 7 1 H NMR spectrum of compound 3 6a (500 MHz, toluene d 8 room temperature) ................................ ................................ ................................ ..... 259 C 8 13 C NMR spectrum of compound 3 6a (75 MHz, CDCl 3 ) ................................ .. 260 C 9 Expanded 13 C NMR spectrum of compound 3 6a (75 MHz, CDCl 3 ) of the aryl region ................................ ................................ ................................ ............... 261 C 10 1 H NMR spectrum of compound 3 5a (300 MHz, DMSO d 6 ) ............................ 262 C 11 13 C NMR spectrum of compound 3 5a (75 MHz, DMSO d 6 ) ............................. 263 C 12 1 H NMR spectrum of compound 3 27 (300 MHz, CDCl 3 ) ................................ 264 C 13 13 C NMR spectrum of compound 3 27 (75 MHz, CDCl 3 ) ................................ .. 265 C 14 1 H NMR spectrum of compound 3 7 (500 MHz, DMSO d 6 ) .............................. 266 C 15 13 C NMR spectrum of compound 3 7 (125 MHz, DMSO d 6 ) ............................. 267
16 LIST OF ABBREVIATIONS 1D One dimension(al) 1 H NMR Proton nuclear magnetic resonance 13 C NMR Carbon nuclear magnetic resonance 2D Two dimension(al) AFM Atomic force microscopy AM1.5G Ai r mass 1.5 global AMOLED Active matrix organic light emitting diode APCI Atmospheric pressure chemical ionization BCP Bathocuproine BHJ Bulk heterojunction BTF Benzotrifuranone CI Chemical ionization CT Charge transfer CV Cyclic voltammetry DART Direct ana lysis in real time DCM Dichloromethane (methylene chloride) DFT Density functional theory DLC Discotic liquid crystal DLS Dynamic light scattering DMEDA dimethylethylenediamine DMF dimethylformamide DMSO Dimethylsulfoxide DOSY NMR Diffusion ordered nuclear magnetic resonance spectroscopy DPV Differential pulse voltammetry
17 DSC Differential scanning calorimetry EI Electron ionization EPR Electron paramagnetic resonance EQE External quantum efficiency ESI Electrospray ionization FT IR Fourier transform infrared H B ONDING Hydrogen bonding HOMO Highest occupied molecular orbital HPLC High performance liquid chromatography ITO Indium tin oxide LUMO Lowest unoccupied molecular orbital M BTF Methyl benzotrifuranone M TRF Methyl benzotrifuran MO Molecular orbital MS Mass spectrometry NMR Nuclear magnetic resonance spectroscopy OFET Organic field effect transistor OLED Organic light emitting diode OPV Organic photovoltaic PCE Power conversion efficiency PGSE Pulsed gradient spin echo POM Polari zed optical microscopy PPA Polyphosphoric acid TD DFT Time dependent density functional theory TBAF Tri n butylammonium fluoride
18 TFA Trifluoroacetic acid TGA Thermogravimetric analysis THF Tetrahydrofuran TIM Mellitic triimides TLC Thin layer chromatography TMS ACETYLENE Trimethylsilylacetylene (ethynyltrimethylsilane) TOF Time of flight TRF Benzotrifuran UV Vis Ultraviolet visible spectroscopy XRD X ray diffraction
19 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy RATIONAL DESIGN OF ONE DIMENSIONAL COLUMNAR SELF ASSEMBLED NANOSTRUCTURE S FOR APPLICATIONS AS ORGANIC SEMICONDUCTIVE MATERIALS By Benjamin M. Schulze August 2013 Chair: Ronald K. Castellano Major: Chemistry Special classes of organic semiconductive molecules capable of aggregating into one dimensional columnar nanostructures have shown remarkable charge transport characteristics and great promise in devices like organic field effect transistors and organic ph otovoltaics. Driving aggregation toward these self assembled architectures bonds is well known, yet few examples exist of their fabrication into devices. This work detai ls three facets of study regarding one dimensional columnar nanostructures. First, acylated derivatives of benzotrifuranone were incorporated into a bicomponent donor acceptor system with complementary electron deficient derivatized mellitic triimides. Benesi Hildebrand analysis performed on the orange 1:1 equimolar n heptane bicomponent solution yielded a K a sc = 4.33 x 10 2 M 1 ( = 3.56 kcal/mol) for the charge transfer complex. Signatures of liquid crystalline behavior of bulk 1:1 equimolar mixtures were borne out through dendritic fan and mosaic textures. Second, a rational design strategy was developed and implemented to impr ove morphology and performance of small molecule bulk heterojunction organic
20 photovoltaics. To this, hydrogen compounds were utilized in structure property performance relationship study. Hydrogen bond d irected assembly manifested through spectral broadening and carbonyl peak shifts in infrared spectroscopy (features persistent in blends with C 60 ), red shifted thin film absorption, and improved decomposition and melt transition temperatures resulted in significantly different bulk behavior. Nuclear magnetic resonance spectroscopic studies of hydrogen bonded assemblies showed shifted and broadened N H/O stacking, and a high tautomerization activation energy. Impr oved electrical properties and threefold increase in power conversion efficiency in vacuum deposited devices implicate improved morphology through hydrogen bonding. Third, to more directly probe the hydrogen bond directed assemblies as mentioned above, a novel variable temperature diffusion ordered nuclear magnetic resonance spectroscopy technique was employed to approximate the size, shape and molecularity of the supramolecular nanostructures through fitting of the modified Stokes Einstein equation. Dimensionality of the supramolecular monomer in dimethy l sulfoxide d 6 congener. Supramolecular assembly parameters in toluene d 8 where bimodal behavior stacked aggregates consisting of an average of 9 hydrogen bonded trimeric discs and disassembled stacks of free trimeric discs in the second mode
21 CHAPTER 1 INTRODUCTION Organic Semiconductors the science behind them is anything but. Silicon, having a variety of semiconducting properties throughout its different bulk phases, has dominated the advanced electronics market. Lately, however, carbon based semiconductive materials have been making a large impact both in laboratory and consumer device settings, such as cell phone displays like the Samsung Galaxy S series active matrix organic light emitting diode ( AMOLED ) screen s, solar cells like the PowerPlastic offered by Konarka, and televisions such as the organic light emitting diode ( OLED ) models offered by LG. While some of the technologies are still in their consumer application infancy (OLED televisions) and some are quite mature (OLED cell phone displays), carbon based or organic semiconductors have been studied for over 30 years due to their numerous advantages over traditional silicon based semiconductors 1 17 First, while the overall price of silicon based electronics has dropped over time, the price of the processed and purified bulk material has essentially remained unchanged The price drop can be attributed to the miniaturization of man ufacturing processes 18 This is highlighted in the applications of silicon: high quality semiconductors require polycrystalline silicon whereas display technologies utilize amorphous silicon, of which there are several purity requirements Most industria l processes to create the usable forms of silicon involve melting ( the melt ing point of silicon is 1414 C) and recrystallizing the material upon cooling. Further, deposition of the silicon onto substrates can also be temperature intensive, making large a rea applications even
22 more costly. 10 Last, the intrinsic properties of these materials are not appreciably tunable that is, it is difficult chemically or synthetically modify the bulk material to induce or produce desired properties. On the other hand, or ganic semiconductors carry none of these pitfalls. 1,5,7 11,13 16 To begin, organic semiconductive devices are fabricated with organic molecules conjugation. Synthetic tailorability of these organic molecules can be easily realized to bring about numerous desired qualities such as solubility or band gap manipulation. Creating tailored systems can be synthetically facile especially when aryl cross coupling reactions, such as Suzuki, Stille, and Heck couplings, are utilized. Solubility can be readily addressed as well, which lends these systems to large area appl ications via either vacuum or solution based deposition. Either fabrication method represents far less harsh processing conditions than what is required for silicon, and therefore, they remain a less expensive alternative. These qualities of organic semic onductors speak to the attractiveness of the field and hence the explosion of research the field has witnessed The concept of c harge transport the movement of charge carriers through a material, is central to the development of organic semiconductive mat erials and is measured experimental ly via charge carrier mobility, (or H for hole mobility and E for electron mobility specifically) 4,6,12 Carrier mobility measurements can be performed through a variety of bulk material techniques such as space char ge limited current experiments, 4 and the essence is to capture the speed of the charge carriers. Faster carriers generally results in higher performance devices. Additionally, specialized p or n type semiconducting materials can be created wherein the m olecule itself is rationally
23 designed to sustain transport of either holes (p type) or electrons (n type). In general, organic devices seek to emulate the transport properties of amorphous silicon ( 1 10 cm 2 /Vs ). 17 Facile charge transport can occur i conjugated electron systems. Relatively facile intermolecular c harge transport has been c onjugation in conjunction with orbital overlap is critical to the success and function of these materials ; 4 without it, transport theoretically occurs inefficiently via electron hopping 12 The importance of these two requirements was high lighted computationally by Br das and co conjugated organic was roughly equal to the intermolecular centroid to centroid distance) to laterally slipped stacks, the resulting expected charge transport characteristics of the system dropped systematically 6 Maximization of this type of specific molecular ordering necessitates control of assembly and organization and leads to an ideal supramolecular structure suited for applications in charge transport. One Dimensional ( 1D ) Columnar Nanostructures conjugated molecules is their inherent planarity derived from a network of sp 2 hybridized carbons whose overall structure can resemble a rod or a disc; disc systems will be the primary focus herein (F igure 1 1 ). The aggregate association energy between molecules and causes them to self assemble to varying degrees and with varying alignments. There exist two primary alignments in stacked
24 aggregates, cofa cially aligned stacks (Figure 1 1 1 1 b). Both aggregates can organize further i nto different supramolecular assemblies such as nanofibers and nanotubes (Figure 1 1 c d respectively ) 9,14,19 21 In this dissertation, or more molecul es whose association, either with itself or a guest, is in state of thermodynamic equilibrium, which can be affected by various factors such as solvent, pH, temperature, etc. Such architectures, collectively known as one dimensional (1D) columnar nanostru ctures, promote molecular orbital (MO) overlap between molecular species. The two assemblies shown in Figure 1 1 illustrate possible columnar hexagonal (Figure 1 1 c middle ) and columnar rectangular (Figure 1 1d middle ) assemblies. While both have been s hown to exhibit high charge transport capabilities, the columnar hexagonal arrangement is desired due to complete MO overlap. system is the culprit for the herringbone arrangement, and strategies for overcoming that specific pitfall of columnar nanostructures will be discussed later.
25 Figure 1 1 Cartoon representation of self assembly of disc shaped molecules into (a) cofacial stacks or (b) slipped stacks ; further organization of those aggregates into a (c) columnar hexagonal arrangement leading to nanothread formation and (d) columnar rectangular arra ngement leading to nanotubes In terms of device relevant contexts such as organic photovoltaics (OPVs) or organic field effect transistors (OFETs), 1D columnar structures could transport charge effectively albeit under slightly different device specific conditions. OP Vs will be discussed later in more detail, but the simple, overarching goal is to create unidirectional current from light. Briefly, state of the art OPV devices utilize a bulk conjugated molecules with semicon ducting properties are mixed with electron accepting molecules (usually fullerene derivatives), (a) (b) (c) (d)
26 which produces a nanoscale phase separated material. 1,7,15,16,22 If the phase separation were to include 1D nanocolumns (of an appropriate diameter) spanning t he BHJ layer, the charge carriers could efficiently migrate to their respective electrodes through the column. OFET devices, based on thin film transistor architecture, utilize an external electric field to produce bidirectional current in the semiconduct or layer. 13 Again, if well aligned 1D columns could be employed, the devices could bear performance similar to silicon. In fact, 1D carrier mobility of discotic molecules outperformed non single crystalline conjugated polymers when the discs were organ ized into 1D columnar nano structures. 23 Unsurprisingly molecules synthesized for 1D columnar nanostructures, like conjugated core 20 These molecules ( 1 1 Figure 1 1 ) have been functionalized quite extensively around the periphery, and subsequently studied for ordering and charge carrier mobility, which ranged from 10 7 to 10 1 cm 2 /Vs. Derivatized versions of 1 1 can form nanotubes or nanorods, but tend to aggregate in the herri ngbone structure, and recently, nanochanneled templates were successfully used to suppress it 24 Linear acene derivatives, namely pentacene ( 1 2 Figure 1 1 ), have remained the highest performing organic molecules in device settings, reaching a peak mo bility of 40 cm 2 /Vs in single crystalline form 11,20,22 One drawback of pentacene based devices is the relative instability of the molecule due to sensitivity to oxidation 25 Perylene and naphthalene bisimides ( 1 3 and 1 4 respectively, Figure 1 1 ) are also common scaffolds for 1D columnar structures, but tend to be utilized as n type semiconductors due to their inherent electron deficiency. As a result, these derivatives have been used as the
27 acceptor material in OPVs and exhibit electr on mobility around 10 2 cm 2 /Vs. Other well studied 1D assembly conjugated motifs include porphyrins, phthalocyanines, rubrenes, poly and oligothiphenes, and pyrenes 20 Figure 1 2 Chemical structures for hexabenzocoronenes ( 1 1 ), pentacene ( 1 2 ), perylene bisimides ( 1 3 ), and naphthalene bisimides ( 1 4 ) Self Assembly and 1D Nanostructures Armed with knowledge of the physical nature of these structures and what types of molecules are used, the outstanding grand challenge is enforce ment of 1D columnar aggregation to device relevant specifications 26 Top down methods, i.e., methods in whic h the finished material is manipulated to provide the highest performance, such as mechanical electric and magnetic induction, and liquid crystalline alignment techniques have all been explored for polymeric and small molecule systems 27 These metho dologies, while effective, can involve complex machinery, elaborately patterned substrates, or auxiliary molecules that must be washed away. Consequently, nature 1 1 1 2 1 3 1 4
28 has provided a simpler and more elegant solution that requires none of these techniques. Natu up self assembly approach, a method in which the lowest level materials are biosynthetically appended with all the necessary functionality to achieve high performance, to enforce specific structural aggregation. Deoxyribonucle assembly. Moreover, DNA is an especially relevant example because it can be loosely thought of as a semiconductive 1D nanostructure. The combination of complementary hydrogen bonding (H and vertical structural blue printing favorable for charge transport. Barton and co workers have exhaustively researched the charge transport capabilities of DNA and have proposed a mechan ism by which the double helix structure is validated against defects via charge transport 28 In other words, a defect within DNA is detected when a charge carrier is sent at one end of the sequence and not received at the terminal end. Not only does t his example illustrate how H to create well defined 1D semiconductive nanostructures, but it also highlights the interdependence between ordering and successful charge transport. The question now becomes can we har ness these same intermolecular interactions to intelligently design and create organic semiconductive materials as well as nature does? Strategies and Interactions The toolkit of self assembly involves utilization of non covalent interactions like H 29 Other non covalent interaction s like halogen bonding 30 and CH 31 have been successfully applied in self assembl ing organic semiconducting materials, but the scope
29 of this dissertation will focus on primarily on H Accordingly, understanding the physical nature of these interactions is necessary for their successful functional implementation. Briefly, the strength of non covalent interactions is characterized by a Gibbs free energy change of the association. 32 In many cases, an association co nstant is measured, e.g. in solution via nuclear magnetic resonance ( NMR ) titration or dilution studies of a sample, and the free energy change can be calculated from: ( 1 1) where K as c is the experimentally determined association c onstant, R is the universal gas constant, and T is the temperature. Gas phase calculations are often used to gauge relative strengths of these interactions and are used as predictors of self assembly behavior. However, the extent to which these relative association strengths can be applied to practical observations remains limited to overall trends. Numerous extrinsic factors can influence the strength of these associations such as solvent or temperature, so practical assessment of K as c often ca rries the caveat of specific experimental conditions. Intrinsic factors such as electronegativity and electron withdrawing effects can also influence association strengths. Through rational application of these factors, H bonding interactions, various levels of aggregation complexity can be achieved from self assembling materials. system, which creates a highest occupied molecular orbital (HO MO ) with discrete regions of appreciable negative electrostatic character (Figure 1 3 a, red region). By consequence, this build up of electrostatic energy creates local regions of opposite character ( Figure 1 3 a, blue region). Local elec trostatic differences can allow
30 delocalized molecules wherein the regions of complementary electrostatics will associate ( shown in Figure 1 3 b). 32 The resulting free energy change can range from zero to 3 kcal/mol, which corresponds to a relatively conjugated molecules pervade the field of organic semiconductive materials, this interaction is often the seed interaction for self assembly. Indeed, in OPV device setti ngs, it is often the only self assembly driving force. However, as Figure 1 3 conjugated molecules sharing similar electrostatics will not stack cofacially, which has previously been demonstrated as desired for optimal charge transport. Cofacial alignment can be in teracting species exhibit a centroid to centroid distance that surfaces (Figure 1 3b,c shows poor cofacial alignment highlighted by the nonequal centroid to centroid and interplanar distances). Figure 1 3 Cartoon schematic illustrating the electron density of a symmetric delocalized system (a) and top (b) and side (c) views of how electrostatics dictate molecular alignment in a stacked system Conse system have been developed to overcome this organizational pitfall. By synthetically append ing electron withdrawing groups such as cyano, nitro, and imide functionalities, the effect results in a system and subsequently lowering (a) (b) (c)
31 of the energy of the HOMO and lowest unoccupied molecular orbital (LUMO). Conversely, appending elec tron donating groups such as al kyloxy and alkylamino system and raising the energy of the frontier orbitals. If the HOMO of the electron rich species (the ron deficient species (the 33 35 This unique interaction first proposed by Mulliken, is derived from orbital mixing between each species and subsequent varying degrees of charge transfer from the donor to the acceptor to create a donor acceptor charge transfer (CT) complex As a result of the orbital mixing, the band gap of the complex becomes relatively small as evidenced by CT complex s olution s exhibiting color 32 With careful design of the two systems, cofacially aligned species can be obtained, which results in the high utility of these electron poor/electron assembly of 1D columnar nanostructures 3 6 The two previously discussed interactions are relatively weak non covalent interactions, which result in gas phase calculation based free energy changes less than BHJ solely relying on those to drive long range self assembly appears to be inefficient as most BHJs exhibit nanoscale aggregation that is largely random. 37 In contrast, hydrogen bonding ( H bonding ) interactions, when intelligently implemented, can have a solution based free energy c hange greater than 16 kcal/mol 38 They consist of a C oulombic interaction between a polar donor bond and an acceptor atom. In practice, the majority of H bonds consist of N H or O H as the donor s and nitrogen or oxygen atoms with free
32 electron pairs as the acceptor s While not the only case, 39 t he benzene 1,3,5 tricarboxamide motif ( 1 5 Figure 1 4 ) is a particularly relevant example of how H bonding can enhance ordering in organic semiconductive systems to produce favorable 1D nanostructures. 4 0 The resulting stability is derived from intermolecular persistent alignment via the amide bond (Figure 1 4a b ). In one case, a core composed of 1 5 was appended with triphenylene pendants, and the resulting charge carrier mobility was the highest recor ded at the time for triphenylene 2 /Vs). 41 Figure 1 4 Chemical structure of 1,3,5 trisbenzenecarboxamide ( 1 5 ) motif with (a) side and (b) top views of the intermolecular H bonding Rational design of molecules using moieties capable of self assembly can lead to high utility supramolecular structures materials. This notion can be elucidated by t he comparison of two key examples of molecular design. Building off of the successful development of discrete hexameric 42 researchers developed high performance bicomponent system consisting of perylene functionalized cyanurates ( 1 6a ) and melamines ( 1 6b ) that showed remarkable self assembly capabilities 43 The benefits of this rationally implemented design were borne out in the carrier mobility performance of the bulk material, which was 0.20 cm 2 /Vs In contrast, oligothiophenes ( 1 7 Figure 1 5 ) mixed with perylene bisimides ( 1 8 Figure 1 5 ) both appended with amide bond functionality did not form well organized aggregates 44 In this instance, 1 5 (a) (b)
33 amide bonds were thought to help organize the two materials into morphologies beneficial for OPV applications. However, cursory implementation of H bonding produced a small photocurrent. These two examples highlight t he necessity of rational design in high performing 1D self assembled systems for organic semiconductors. Figure 1 5 Chemical structure of cyanurate ( 1 6 a ) and melamine ( 1 6b ) derivatives aggregated hexameric rosette structure ( ( 1 6 a 1 6b) 3 ), oligothiophene ( 1 7 ), and perylene bisimide ( 1 8 ) with amide functionality Self Assembly and Discotic Liquid Crystals The aforementioned non covalent interactions have been instrumental in the development of 1D nanocolumnar assemblies, and one consequence of complementary 1 6 a ( 1 6 a 1 6b) 3 1 7 1 8 1 6 b
34 interactions is the formation of liquid crystalline or mesomorphic materials. The link interactions, and 1D columnar structures is surface is large enough, then the interactions can be strong enough to spontaneously induce ordering over larger length scales even in the bulk liquid phase 45 49 In the bulk solid phase, further aggregation can occur to form the columnar architectures that have been previously shown. However, synthetic organic chemists surfaced molecules often exhibit low solubility and high melting points. To this, DLCs typically contain long primary alkyl alkyloxy, or alkyl ester chain functionalities that allow for mesophase temperature range tuning and solution based processability 48 Further, the alkyl chains can contribute to better cofacial alignment due to favorable bulk phase inter chain van der Waals attractions. 47 Devices fabricated with DLCs could have several unique features, namely the long range ordering inherent within the bulk material. Indeed, several DLCs have had their charge transport capabilities measured, and the charge carrier mobility results are among the highest recorded for non single crystalline organic materials 23,45 DLCs also carry other advantages such as self healing capability and ease of processing. The rational modification of DLC molecules has also been explored, usually with desired property enhancements in mind. The general results of employing H bonding contacts in conjugated cores were improved mesophase stability and stricter enforcement of cofacial alignment 39,50 53 and c harge transport performance o f H bonded DLCs have been noteworthy 39,52 even in heteromeric bicomponent systems. 43 In addition to H bonding, Ringsdorf and co workers pioneered
35 the introduction of electron rich/electron into polymeric systems leading to induce d mesomorphism where none previously existed. 54 Further effects of these donor acceptor DLC systems are borne out in mesophase stability throughout broader temperature ranges and better cofacial alignment than either single component While these benefits make donor acceptor systems attractive for semiconductive applications, 33 the exact or dering of the system is not completely understood. 55 conjugated species, which may or may not be detrimental to performance and showcases the need for more research of these material s in device based settings. Self assembled 1D Systems in Device Contexts Strategies for 1D self assembly coupled with functional molecules appear to be a recipe for successful development of new semiconductive technologies. However, remarkably few devices have been constructed utilizing the afore mentioned techniques 20 As a result, a large roadblock remains in determining how 1D nanostructured materials perform in macroscopic settings. Structure property function relationships based on device implemen tations can build a knowledge base upon which bottom up design strategies can be devised. To clarify, it is well understood which molecular features contribute to functional self assembled materials; it is less clear as to whether or not thos e features contribute to improved performance in devices. Understanding the nature of the assemblies in device based settings can be quite challenging. In laboratory settings, numerous techniques can be utilized to probe structural characteristics like so lution based association constants of self assembling molecules, but it is unclear as to how this behavior would translate to a bulk material, much less blends of materials as is required for a n OPV BHJ. Bulk methods like X ray
36 diffraction ( XRD ) can be useful in determining morphologies, but again, spectral interpretation can become difficult when C 60 a material with its own self assembly characteristics, interacts with the target molecules. To highlight this point, hexadecylthiotriphenylene ( 1 9 Figure 1 6 ), a 1D columnar nanostructure forming discotic molecule, was utilized in a processing comparison study in which it s nanofibers were prepared via two different methods 56 Compound 1 9 (theoretical mobility as high as 1 cm 2 /Vs) was thought to be an ideal candidate for OFET devices. However, despite showing encouraging nanoscale morphological results (the molecule formed 1D columnar fibers as expected), the devices performed unexceptionally ( 6 10 7 cm 2 /Vs). The authors concluded that mechanical and field induction techniques mentioned previously were likely more effective methods, and left the scientific community with no greater understanding on the self assembly within the device Attempts at controlling aggregation through rational design where H bonding was utilized involved have seen limited study. One recent remarkable OFET example invoked the acene motif (known to exhibit high carrier mobilities) and utilized elegantly oriented H bonding contacts to create high performance mate rials in device settings 57 In fact, the H bonded molecules epindolidione and quinacridone ( 1 10 and 1 11 respectively, Figure 1 6 ) performed as well as or better than their all carbon covalently bound congeners tetracene and pentacene ( 1 12 and 1 13 respectively, Figure 1 6 ), and demonstrated exceptional performance reliability. Most importantly, the authors examined the morphology of the device via atomic force microscopy ( AFM ) and noted that the aggregates resembled nanorods, and further commen ted that the larger aggregates resulted in better performance.
37 Figure 1 6 Chemical structures of hexadecylthiotriphenylene ( 1 9 ), epindolidione ( 1 10 ), quinacridone ( 1 11 ), tetracene ( 1 12 ), and pentacene ( 1 13 ) The nature of self assembly within OPV contexts has also only been modestly explored. BHJ morphologies are largely random (Figure 1 7a ), so the nanoscale morphologies that form within them can be randomly oriented I de separated aggregates are viewed as detrimental to efficiency due to the lack of clear migration pathways for charge carriers to electrodes. 1,7,15,16,22 An idealized BHJ morphology (Figure 1 7b ) would inclu de phase separated aggregates that spanned the entire BHJ from anode to cathode as such, the proper implementation of well ordered self assembling 1D columnar nanostructures rationally designed from the bottom up could mimic this theoretical high efficienc y morphology. Figure 1 7 Cartoon representation of donor and acceptor material blue and orange respectively, (a) in a randomly oriented active layer morphology in BHJ OPV device and (b) in an idealized highly ordered morphology of BHJ OPV One strategy to achieve this ideal morphology, and therefore improve performance, would be to incorporate supramolecular structu re guiding H bond s into the 1 9 1 10 1 11 1 12 1 13 (a) (b)
38 BHJ material. The rational implementation of H bonding in molecular design can create modestly explored in device settings. 9,58 Examples exis t in both polymeric and small molecule device settings but have only seen limited success in terms of performance improvements. Modified fullerenes complemented with H bonding motifs attached to polymer chains have exhibited favorable device improvements such as long term device reliability and nanoscale phase separation. 59 61 Pendant melamine units attached to a conjugated bisuracil units demonstrated improved performance, but i t i s unclear as to whether the species. 62,63 H bonding in small molecule based OPVs has only been modestly explored as well. While not fully investigated, substituting carbon for nit rogen in benzothiadiazole moieties has been favorably implicated for improve d aggregation due to H bonding 64 H bonding contacts of unsubstituted isoindigo based devices were thought to decrease p erformance due to disrupt ed phase separation. 65 Absent ar e systematic structure property performance relationships that fully explore the nature of the BHJ morphology when H bonding self assembly design strategies have been rationally applied. Scope and Outline of Dissertation This document explores the rationa l design of self assembling materials for high performing organic semiconductive devices. Electron rich/electron interactions and H bonding have been implemented to drive 1D columnar aggregation in two separate projects. Investigation s have prim arily centered on elucidating aggregation behavior in solution and in bulk, and subsequently incorporating the new
39 materials into devices where possible. Device fabrication is necessary for establishing structure property performance relationships. Chapte r 2 details the synthetic functionalization of a C 3h symmetric scaffold benzotrifuranone or BTF whose aromatic core could be expanded through one step functionalization Previous stud ies derivatives had revealed nonideal assembly for organic semiconductive applications, so other derivatives were synthesized and mixed with similarly shaped yet electron deficient mellitic triimides. Previous work with mellitic triimides led to the hypothesis that the mixture of the m with the BTF derived derivatives could lead to a donor acceptor complex that exhibited charge transfer characteristics and mesomorphism. E quimolar mixtures of size matched compounds yielded CT complexation, which was studied via UV Vis solvent titration and dilution study Further evidence of electron rich/electron poor interactions were examined through the b ulk organization of 1:1 mixtures Differential scanning calorimetry (DSC) and polarized optical microscopy (POM) were utilized on both the neat materi als and the mixtures to assess thermal characteristics and mesomorphism. Chapter 3 explores the bottom up methodology of self conjugated chromophores for improving blend morphology in BHJ OPV devices. D esign of the branched quaterthiophene c hromophore included functionaliz ation with a phthalhydrazide motif capable of self complementary aggregation Both the neat material and blended films revealed enhanced aggregation. When fabricated into BHJ OPV devices, a power co n version efficiency (PCE ) increase of threefold attributed to
40 enhanced electrical properties resulted when comparing the H bonded chromophore to its non H bonding congeners. conjugated H bonding chromophores previously mentioned in Chapter 3 through variable temperature diffusion ordered nuclear magnetic resonance spectroscopy (DOSY NMR). Diffusion measurements were performed in both H bond promoting and suppressing solvents and were compared against the measurements of the non H b onding congener. Fitting routines performed on plots of diffusion coefficients versus temperature based on the modified Stokes Einstein equation revealed molecular and aggregate shape parameters consistent with low level modeled structures. Further insig ht was gained through comparison of diffusion measurements and shape parameters between H bond active and inactive congeners. Chapter 5 summarizes the overall effort of examining 1D columnar ordering within self assembled systems and the notion that ratio nal design of these nanostructures can lead to interesting physical characteristics and organic semiconducting device improvements. The conclusion also discusses on going and future research endeavors.
41 CHAPTER 2 DONOR ACCEPTOR COLUMNAR ASSEM BLIES OF BENZOTRIFURAN: CHARGE TRANSFER (CT) COMPLEXATION AND MESOMORPHISM Background and Previous Work of the Benzotrifuranone Scaffold Recent work performed by the Castellano group at the University of Florida examined the novel one step derivitization of a C 3h symmetric heterocycle benzotrifuranone ( 2 1 BTF, Figure 2 1 ) to a symmetric planar aromatic benzotrifuran ( 2 2 TRF, Figure 2 1 ) 66 This one step methodology also proved successful to convert modified BTFs (e.g., methyl BTF or m BTF, Figure 2 1 a). BTF or m BTF can be reacted with acid chlorides or chloroformates in the presence of triethylamine at 0 C in 18 h to form TRFs and m TRFs. Previously, TRFs, which have been shown t o have mesomorphic properties, were synthesized under harsh conditions and suffered extremely low yields. 67 With the new methodology presented in the literature, TRFs could be synthesized in higher yields, with different acylating agents, and with modified BT Fs. 68 The crystal structure of a m TRF derivative (Figure 2 1b) showed face to face centroid to centroid distance o f 4.61 . 68 High level density functional theory ( DFT ) Recent work performed by the Castellano group at the University of Florida examined the novel one step derivitization of a C 3h symmetric heterocycle benzotrifuranone ( 2 1 BTF, Figure 2 1 ) to a symmetric planar aromatic benzotrifuran ( 2 2 TRF, Figure 2 1 ). 66 This one step methodology also proved successful to convert modified BTFs (e.g., methyl BTF or m BTF, Figure 2 1 a). BTF or m BTF can be reacted with acid chlorides or chloroformates in t he presence of triethylamine at 0 C in 18 h to form TRFs and m TRFs. Previously, TRFs, which have been shown t o have
42 mesomorphic properties, were synthesized under harsh conditions and suffered extremely low yields. 67 With the new methodology presented in the literature, TRFs could be synthesized in higher yields, with different acylating agents, and with modified BT Fs. 68 The crystal structure of a m TRF derivative (Figure 2 1b) showed face to face with a spacing of 3.35 and a centroid to centroid distance of 4.61 . 68 High level DFT calculations (together with collaborators at Oak Ridge National Laboratory) suggested that this arrangement was a consequence of repulsive interactions between the re latively electron rich aromatic cores. Accordingly, preliminary experiments utilizing polarized optical microscopy (POM) and X ray diffraction (XRD ) gave little evidence that the molecules organized into columnar arrangements in solution or the bulk. This initial work also revealed potential molecular stability issues with the TRFs and m TRFs worthy of additional investigation. This chapter investigates the scope of the synthetic methodology presented in Figure 2 1a, the stability of the TRFs, and th e self assembly behavior of TRFs in solution and in bulk. Can well stacked assemblies from the rapidly accessible TRF core be obtained?
43 Figure 2 1 Previous work performed with benzotrifuranone (BTF) showing (a) g eneralized reaction scheme of acylation of BTF to give benzotrifurans ( TRFs ) and (b) crystal structure showing slipped stack arrangement of a m ethyl TRF derivative As previously stacking can be promoted by introduction of complementary electron systems, researchers have paired donor aromatics with 2,4,7 trinitrofluoren 9 one ( 2 3 Figure 2 2 ) to induce well ordered ass emblies and CT complexation 54 However, since 2 3 has been classified as a Class 1.1D explosive material and is no longer commercially available, this strategy has become less popular today. McMenimen and co workers recently showed that C 3 symmetric mellitic triimides ( 2 4 TIM, Figure 2 2 ), whose first reduction potential is similar to 2 3 could serve as alternative acceptors 69 Notably, they showed that 2 4 can induce mesomorphism with comparably sized and C 3 symmetric donor molecules and were able to obtain a single co crystal suitable for X ray diffraction analysis. Later, Reczek and co workers expanded on this work with a small library of donor molecules and provided insight into the liquid crystalline behavior donor acceptor materials utilizing 2 4 can afford 70 (a) (b) 2 1 2 2
44 Figure 2 2 Chemical structure of 2,4,7 trinitrofluoren 9 one ( 2 3 ) and mellitic triimides ( TIMs, 2 4 ) Given the McMenimen precedent and subsequent detailing of liquid crystalline behavior, a rational design strategy to engineer well ordered TRF stacked assemblies (Figure 2 3) was envisioned wherein TRF or m TRF derivatives would be paired with similarly sized TIM molecules. It was expected that such a system could possibly form a CT complex in solution and exhibit liquid crystallinity in the bulk. Described are solution and solid phase investigations of 1:1 TRF TIM donor acceptor assemblies along these lines Figure 2 3 Cartoon representation of possible donor acceptor columnar aggregation of TRF and TIM 2 3 2 4
45 Synthesis of Donor and Acceptor Compounds Previous work has been published on the synthetic methodology and d eriva tization of BTF ( 2 1 Figure 2 4) and m BTF ( 2 11 Figure 2 4) by previous members of the Castellano group 66,71 However, a new procedure for cyclization was developed which saw vastly improved yields of 2 11 Figure 2 4 shows the syntheses of 2 1 and 2 11 in their entirety. Briefly, 1,3,5 trimethoxybenzene ( 2 5 ) was bromo methylated under acidic conditions with paraformaldehyde on a multi gram scale, and 2 6 was isolated in low yield after flash chromatography. Several attempts at optimizing this step ha ve been made with no success. Benzylic S N 2 substitution afforded the cyano functionalized 2 7 in high yield. In order to obtain 2 1 compound 2 7 was simultaneously hydrolyzed and deprotected with concentrated hydrobromic acid at reflux moderately yieldi ng 2 8 After cyclization of this product with polyphosphoric acid (PPA), 2 1 was achieved in low yield. To synthesize 2 11 2 7 was deprotonated with position to afford 2 9 as a mixture of dias tereomers (1:3 ratio of syn to anti ) in good yield. Hydrolysis and deprotection afforded 2 10 in good yield. Cyclization was performed with poor yield under the same conditions as 2 1 to yield 2 11 after flash chromatography.
46 Figure 2 4 Synthetic scheme for access to BTF ( 2 1 ) and related derivatives m ethyl BTF ( m BTF 2 11 ), and acylated versions ( TRF, 2 2a,b,c and m TRF, 2 12a,b,c ) Abysmal yields and the high demand of 2 1 and 2 11 material warranted further investigation into the final cyclization step. Several other protocols had been previously explored by the Castellano group, but the best results could only produce moderate yields of 2 1 ( Table 2 1 ), and the PPA methodolog y emerged as the primary cyclization methodology. Because of this, no alternate methods of cyclization were explored for 2 11 For the conversion of 2 10 to 2 11 the barrier to achieving high yields could be the ring strain associated with forming a fiv e membered ring and the steric hindrance of the methyl group. However, the added methyl group does make the compound soluble in a wider variety of organic solvents, which became apparent when the new cyclization
47 methodology was first utilized. Conseque ntly, 2 11 could be synthesized from 2 10 through usage of a Soxhlet apparatus and a substoichiometric amount of p toluenesulfonic acid in toluene over 48 h in 73% yield (a 53% boost from previous attempts). 2 8 under the same conditions, did not achieve improved results (<10% yield, impure) from the published method, presumably due to low solubility Table 2 1 Various cyclization reaction conditions and results as performed and reported by previous Castellano group members (Y an Li and Matt Baker ) Entry Conditions and Reagents Temperature ( C) Time (h) Yield 1 POCl 3 Reflux 1 <10% ( 2 1 ) 2 POCl 3 Reflux 4 <10% ( 2 1 ) 3 PPA 110 4 20% ( 2 1 ) 4 PPA 130 overnight 40% ( 2 1 ) 5 PPA 110 overnight 50% ( 2 1 ) 6 TFA, toluene Reflux overnight none 7 4 mol sieves 120 unknown trace ( 2 1 ) 8 P 2 O 5 toluene Reflux 5 11% ( 2 1 ) 9 Oven 160 24 trace ( 2 1 ) 10 Soxhlet Reflux 48 73% ( 2 1 1 ) Purification of Acylated Target Compounds Previous work showed that 2 1 could be acylated by acetyl chloride at 0 C under inert atmosphere in the presence of triethylamine. Generalization of the procedure expanded the reagent scope to include acylation of 2 1 with other acid chlorides as well as the acylation of 2 11 Comp eting reactions between O and C acylation as well as purification difficulties led to low and moderate yields. Post work up crude reaction mixtures often required two flash chromatography columns of different solvent systems. Indeed, previous work had shown yields to be much higher for similar reactions (for example, 2 2c was reported at 44%), but purification was less strenuous in that work.
48 For this research, high purity material was desired in further experimentation, so more extensive purification was employed. To probe target spot resolution, a variety of solvent systems (ethyl acetate in hexanes, dichloromethane in hexanes, 1,4 dioxane in hexanes, acetonitrile in hexanes, etc.) were tested using thin layer chromatography (TLC). Ethyl acetate and hexanes gave the best resoluti on, but required a low concentration of the former (typically less than 2.5% ethyl acetate in hexanes). Product co elution necessitated more TLC investigation and yielded interesting results. 25% dichloromethane (DCM) in hexanes, which originally showed poor target spot resolution, performed highly on a second column and gave pure product. In the case of 2 1 2c purification was arduous. After performing the acylation reaction, appropriate work up, and purification, the colorless, waxy product was spectro scopically confirmed as pure. However, after approximately two weeks on the lab bench, the compound turned yellow. Proton nuclear magnetic resonance spectroscopy ( 1 H NMR ) revealed no significant spectroscopic changes, but TLC revealed an impurity. Reactions conditions were modified to include purified 2 1 (purification of 2 1 could be difficult due to its low solubility) and multiple chromatographic separations were performed, but f rustratingly, the impurity persisted through four separate trials On the last attempt, after the target spot eluted from a second column, TLC performed on the freshly col 5% ethyl acetate in he xanes instead of DCM in hexanes) revealed a spot above the target that had persisted through the second round of flash chromatography. Two dimensional TLC was performed to assess the stability of 2 12c on silica, and it showed no
49 degradation. Having exhausted most purification options, it was decided to move away from this target compound. Synthesis of Mellitic Triimides Synthesis of the mellitic triimides ( 2 14a,b,c Figure 2 5) was performed according to a modified version of the literature procedure. 72 Synthesis of the TIMs ( 2 14a,b,c Figure 2 5) was performed according to a modified version of the li terature procedure. 72 Mellitic acid ( 2 13 ) was dissolved in an appropriate solvent, then combined with three equivalents of a primary amine (details provided in the Experimental section). Results of these reactions after heating in a drying oven for 96 h and subsequent purification via flash column chromatography are highlighted in Figure 2 5. Careful examination of the 1 H NMR spectra of the series of compounds shows broadening of the methylene triplet peak suggestive of slowed bond rotation, suggest ive of CH O H bonding interactions Figure 2 5 Reaction scheme for the transformation of mellitic acid ( 2 13 ) to alkylated mellitic triimides ( 2 14a,b,c ) Attempts at crystallization and co crystallization of 2 2 and 2 14 for single crystal X ray diffraction study were unsuccessful. Since long alkyl chains tend to solvate molecules better (indeed, 2 2c 2 12c and 2 14c mostly stayed in solution), short al kyl
50 chain derivatives (for example, 2 2 a and 2 12 a ), which have already been studied in crystalline form were utilized in most crystallization experiments In general, the neat solutions form ed e, 2 2 b was dissolved in a small amount of warm ethyl acetate, then pentane was diffused into that of the solvent, the structure would collapse to the flat surface of the v ial. Attempts at vapor diffusion with 2 14 b proved unfruitful utilizing vapor diffusion of ethyl acetate into N N dimethylformamide (DMF). Again, a cottonball formed that would collapse upon removal of solvent. Slow evaporation techniques could not be employed due to derivatives of 2 14 being largely insoluble in any solvent less polar than DMF. Additionally, derivatives of 2 2 were far more soluble than their accept or counterparts, which made co crystallization extremely difficult. Consequently, the results of those experiments resembled recrystallization purification techniques (i.e., when the crystalline material was spectroscopically analyzed, it was solely compr ised of 2 14 ). Spectroscopic Analysis of CT Complexation of 1:1 Mixtures Several studies have shown that the combination of donors and acceptors produces visible light color changes in solution due to the formation of a charge transfer ( CT ) complex. 70,73 78 Figure 2 6 (below) shows the color produced from the formation of a 2 2 c / 2 14 c 1:1 CT complex Neat 10 mM 1:1 solutions of 2 2c and 2 14 c in n heptane (Figure 2 6 a,b) were colorless. However, upon mixing, the solution becomes orange (F igure 2 6c) with max = 434 nm and onset = 549 nm ( E g = 2.26 eV) DCM solutions with 1:1 mixtures of the same concentration remained colorless ; further discussion of this solvatochromic behavior comes below
51 Figure 2 6 Images of colorless and colored 10 mM n heptane solutions of (a) 2 2c (b) 2 14c and (c) the mixture of 2 2c and 2 14c (the concentration of each individual component is 5 mM ) CT complexation was characterized by utilizing two different methodologies (detailed description s can be found in the Experimental Section ). The first method involved a modified solvent titration wherein a fixed concentration (10 mM total ) of 1:1 mixture was subjected to various solvent ratios (Figur e 2 7 a ) of n heptane and DCM As the percentage of n heptane decreased the CT absorbance band ( CT = 434 nm) decreased in intensity but a literature search yielded no usable modeling equations to describe the behavior More generally, the absorbance at CT exponentially decayed in good correlation to increasing calculated dielectric constant of the solvent mixture (Figure 2 7b). 79 T his observation indicates that the complex is likely the result of a ( i.e., noncovalent) interaction likely formed by partial charge transfer from 2 2c to 2 14c from full charge transfer, polar solvents such as DCM would stabilize the separation of charge and enhance the overall stability of the complex, not cause it to dissociate. 73 In other words, this experiment highlights the competition between dissolution of the CT complex as a single entity and the individual components; DCM molecularly disso lves (a) (b) (c)
52 the individual components because of the relatively weak association formed via partial charge transfer 73 The CT absorbance also decrease d with increasing dilution but with different trend behavior (see Figure 2 8a ). The modified Benesi Hildebrand equation could be used to determine the molar absorptivity ( ) and the equilibrium constant ( K CT ) of the CT complex: 73,77,78,80 ( 2 1 ) where [A] 0 is the initial acceptor concentration, [D] 0 i s the initial donor concentration, A is the absorbance, and l is the path length in centimeters. Figure 2 8b shows the plot of on the y axis and on the x axis K CT and were derived from a linear fitting whose slope and y intercept are equal to 1/ and 1/ K CT respectively giving K CT = 4 33 x 10 2 M 1 and = 2 34 x 10 3 M 1 cm 1 with R 2 = 0.948. Furthermore, the calculated Gibbs free energy change of the complex ( G equation 1 1, R = 1.986 cal/K mol and T = 295 K ) was 3 .5 6 kcal/mol, which falls in the range of expected values of electron rich/electron For reference, values of 0. 35 kcal/mol have been determined for a toluene tetracyanoethylene complex 73 and 4.23 kcal/mol for a viologen linked pyrene tryptophan pincer type complex. 81 One unanticipated feature of the 1:1 mixture was the apparent bichromism in solution a nd bulk. An abrupt color change of the 1:1 bulk (i.e., solventless) mixture from dark orange to purple over a period of one week was observed, which translated to bright orange to vivid purple in solution ( max = 550 nm, DCM) 1 H NMR s pectroscopy ( Ap pendix A ) confirmed no structural changes had occurred to either compound, which suggested that the molecules had not decomposed, at least to the extent of NMR
53 detection A modified s olvent titration of the aged 1:1 mixture elicited solvent dependent bichromism ( Figure 2 8c ) with no clear isosbestic point that could be attenuated by addition of DCM. To clarify, a 1:1 solution of 2 2c and 2 14c in n heptane exhibited an orange color, but as DCM was added to that soluti on the color changed from orange to purple due solely due to the break up of the orange species and overall persistence of the purple species Rationalization of this behavior stems from the difference in solvent polarity which affects which aggregate dom inates. Close inspection of the 100% n heptane spectrum shows a small shoulder on the CT absorption peak with a max value similar to the purple species. Combined with t he lack of isosbestic behavior the evidence seems to indicate that the purple species may be present in the orange solution, but the absorbance at 550 nm does not dominate the spectrum. It seems logical, then, that the solvatochromic effect at 434 nm is comparable between the fresh and aged specie s in that the effect stems from the interdependence of solvent polarity and CT complex stability Nonpolar solvents enforce the CT complex association whereas polar solvents do not (i.e., they molecularly dissolve the individual components instead of the complex as a whole ). The question then becomes, what is the composition of the purple species? The onset absorption wavelength ( onset = 687 nm, E g = 1.80 eV ) denotes a lower energy band gap relative to the orange species ( o nset = 549 nm, E g = 2.26 eV) which indicates a n overall lower energy aggregate persistent in both polar and nonpolar solvents. A CT complex band gap of 2.16 eV was approximated by subtracting E HOMO of 2 2 ( 5.82 eV, determined through density functional theory ( DFT ) calculations
54 performed at B3LYP/6 311+G* level of theory) from E LUMO of 2 14 ( 3.66 eV). While Mulliken first proposed using electron affinities and first ionization potentials (a method still in practice today) to predict E g of CT complexes, 35 other researchers have been shown the HOMO LUMO method to be an accurate predictor of onset of CT complexes. 70 Hence, the max observed at 434 nm is verified as the absorbance band ( CT ) of the 1:1 CT complex of 2 2c and 2 14c According to rationale previously discussed, t he persistence of the purple species in a polar solvent could denote a kinetically formed aggregate consist ing of full transfer of an electron from 2 2c to 2 14c Electron paramagnetic resonance (EPR) studies on CT complexes have proven the existence of full charge transfe r, and such technique could be useful in understanding the nature of this new absorbance. 82 Figure 2 7 Absorbance and fitting graphs of 1:1 mixture of 2 2c and 2 14c for (a) m odified solvent titration with decreasing percentage of n heptane in DCM and (b) exponential decay of absorbance at CT with respect to the calculated dielectric constant of the solve nt mixture (a) (b)
55 Figure 2 8 Photophysical characteristics and linear fitting of Benesi Hildebrand treatment ; UV Vis absorbance data of 1:1 mixtures of 2 2c and 2 14c recorded at (fixed concentration, 10 mM total ) via ( a ) an n heptane dilution study and (c ) a modified solvent titration of an aged sample using n heptane in DCM ( b ) Linear fitting of Benesi Hildebrand treatment of UV Vis absorbance data recorded in ( a ) Characterization of Thermotropic Mes omorphism of 1:1 Mixtures To further probe the effect of ordering induced by the addition of electron deficient 2 14 derivatives to bulk samples of 2 2 1:1 bulk mixtures 2 2c / 2 14c and 2 10c / 2 14c were subjected to variable temperature polarized optical microscopy (POM) to investigate their thermotropic behavior. As previously mentioned, a consequence of employing electron rich/electron shaped molecules can be the induction of liquid crystalline behavior. The long range alignment of the discs manifests physically in the bulk as a birefringent material when placed in between (a) (b) (c)
56 crossed polarizers of a polarized optical microscope. Often, the birefringent patterns (ca molecular arrangements or liquid crystalline mesophases Predominant within discotic liquid crystals ( DLCs ) are columnar mesophases such as hexagonal columnar (denoted as Col h Chapter 1 Figure 1 1c) and rectangular columnar (denoted as Col r Chapter 1, Figure 1 1d ), which are identified microscopically textures 50,70 Previous POM analysis on neat 2 2c has shown that it can exhibit birefringence under 68 Since mesophases are temperature dependent, DSC is performed in accordance with POM analysis to elucidate mesophase transition temperatures and enthalpies. First, mesomorphic analysis was performed on indiv idual components 2 2c and 2 14c (details shown in Experimental) via DSC and POM. The thermogram for 2 2c (Figu re 2 9a ) shows three features on heating: a glass transition ( T g ) at 24 C (onset), a cold crystallization at 11 C ( 12.9 J/g 2.44 kcal/mol ), and a crystalline to isotropic transition at 41 C ( 58.5 J/g 11.1 kcal/mol ). The features seen on cooling are a broad exothermic peak at 1 C ( 19.9 J/g 3.77 kcal/mol ), which corresponds to an isotropic to crystalline ( seen as by POM ) transition and the mirror of the T g at 28 C (onset) POM analysis of 2 2c verified each of these features ( Appendix B ) and 2 14c (Figure 2 9b ) shows a major endothermic peak at 106 C ( 53.0 J/g 10.0 kcal/mol ) and a minor peak at 26 C ( 2.99 J/g 0.565 kcal/mol ) on heating, which were both mirrored on cooling with slight hysteresis POM study of 2 14c revealed the major peak
57 corresponds to a crystalline to isotropic transition, while the minor peak corresponds to a crystal to crystal transition ( Appendix B ). Next, the 1:1 mixture of 2 2c and 2 14c was analyzed by DSC with various heating and cooling rates ( Figure 2 9c ). The initial heating trace shares some transitions ( e.g., glass transition, cold crystallization, and crystalline to isotropic) with 2 2c indicative of a mutiphasic mixture. Exothermic peak broadening on cooling, which can be common in both neat 52 and donor acceptor 70 DLCs, was observed at 13 C ( 1.42 1.66 kcal/mol). Typically, the broadening effect can be mitigated by varying the ramp rate of cooling or heating; however, the viscous and sticky nature of the mixed substance made preparation of subsequent DSC samples quite difficult. Further, while not an uncommon feature, assigning a mesophase transition temperature to such a broadened peak would remain largely subjective. As such, since thermogram features were i nconclusive or did not resemble conventional bulk thermal behavior, mesophase transition temperatures could not be determined by this method.
58 Figure 2 9 D ifferential scanning calorimetry (DSC) thermogram s on heating (upper traces) and cooling (lower traces) with exothermic peaks down of (a) 2 2c (b) 2 14c and (c) 1:1 mixture 2 2c / 2 14c However, physical changes in organization were better understood through POM, and the phase transitions which were correlated to POM images, are summarized in Figure 2 10 Importantly, in cases of analogous heterocyclic DLCs, mesogenic features can be seen in reverse order of calamitic (rod like) liquid crystals (i.e., DLCs can display crystalline to nematic to columnar to isotropi c transitions on heating) with the possibility of re entrant phases 46,83 Increasing alkyl side chain interactions between aggregated columns is thought to be responsible for this counterintuitive behavior. Given this, the 2 2c / 2 14c DLC system war ranted study on
59 both heating and cooling. Briefly, after a two week aging period at room temperature the system appeared to exhibit both needle like crystalline and dendritic fan like liquid crystal birefringent patterns which initially seemed like the crystallites were suspended in the liquid crystalline material Assignment of the birefringent texture patterns in the absence of extensive XRD data (preliminary XRD data will be discussed later) was based on comparison with the literature 50,70 However inspection of the aged birefringent sample showed no mono domains similar to the crystalline textures of both individual materials. Additionally, DLC samples on a glass slide covered with a glass cover slip are often sheared to ensure that the sam ple is indeed liquid, a test which the 2 2c / 2 14c system passed. Consequently, the initial hypothesis concerning the nature of the biphasic mixture was revised to exhibiting a mixture of different liquid crystalline textures The biphasic aged mixture consisted of both dendritic fan and mosaic like texture s (Figure 2 1 1a ) that, as the system is heated to 45 C, partially transition to isotropic (Figure 2 11c ) Only the dendritic fan texture whose transition to isotropic could be monitored (Figure 2 1 1 b ), disappears at this temperature, but as the sample is heated to 75 C, the mosaic like texture becomes isotropic as well. Upon slow cooling from 75 C to 72 C, the mixture exhibits a fan like liquid crystalline phase that slowly grows in from isotropy ( Figure 2 12a,b ). Of the liquid crystal spherulites imaged from the same sample, individual fan blades were found to exhibit different birefringent patterns, which indicates different degrees of ordering. Figure 2 1 2a shows fan blades with unifo rm and relatively intense birefringence suggestive of a relatively high ordered species, whereas the fan blades shown in Figure 2 12b display varying degrees of
60 birefringence. Heating the sample beyond 105 C results in isotropy, but upon slow cooling, an other liquid crystalline phase resembling a mosaic texture appears slowly at 102 C (Figure 2 12c) The aforementioned biphasic room temperature mixture could then be rationalized as the kinetic combination of the two mesophases Thermotropic liquid crysta lline analysis via POM was also performed on a 1:1 mixture of 2 12c and 2 14c (Figure 2 13 at 40 C) and persisted upon heating to 86 C when the mixture became isotropic. Schlieren textures (Germ mesophase (N Col ) mesophase and are envisioned as discs organizing into tumbling columns 46 While these findings were initially promising, compound 2 12c was observed to contain an unknown impurity ( vide supra ) that could not be removed. These complications ultimately resulted in the termination of this area of study. Figure 2 10 Observed phase transition diagram for 1:1 mixture of 2 2c and 2 14c with associated temperatures
61 Figure 2 11 P olarized optical microscopy ( POM ) images of 1:1 bulk mixture of 2 2c and 2 14c taken at 100x total magnification between crossed polarizers at (b) 41 C, and (c) 75 C (a) ( b ) ( c )
62 Figure 2 12 POM image s of 1:1 bulk mixture of 2 2c and 2 14c taken at 72 C between crossed polarizers showing (a) higher ordered fan blades at 100x total magnification, (b) lower ordered fan blades at 200x total magnification, and (c) POM images taken at 102 C at 100x total magnification between crossed polarizers (a) ( b ) ( c )
63 Figure 2 13 POM image of 1:1 bulk mixture of 2 1 2c and 2 14c taken at 40 C at 100x total magnification between crossed polarizers X ray Diffraction Analysis With the help of the Department of Engineering and the Major Analytical Instrumentation Center (MA I C, Dr. Valentin Craciun), XRD was performed at room temperature on a spin cast sample of 2 2c and 2 14 c as an equimolar mixture ( Appendix B for spectra). The only prominent feature that could be elucidated was a peak at d = 29 , which could correspond to an intercolumnar distance ( the 100 reflection ) or the distance between stacks. Additionally, since XRD measures structural periodicity pe rpendicular to the substrate, this finding implies edge on or homogeneous alignment of the columns with respect to the substrate. Indeed, if the columns were zero birefri be the side chain halo. Absent from these spectra is the interplanar distance between discs (001), which is
64 aligned sy stem, this periodicity would not be observed due to its parallel alignment with the substrate. Summary Well ordered columnar assemblies were engineered from mixtures of acylated derivatives of benzotrifuranone ( 2 1 ) and mellitic triimides ( 2 14 ). Synthesis of derivatives was straightforward; however, purification remained difficult for target compounds 2 2 and 2 10 often requiring multiple instances of column chromatography. Co crystallization attempts were unsuccessful presumably due to differing crystallization kinetics ( 2 14 crystallizes faster than 2 2 or 2 10 ). Evidence of these assemblies was observed throug h solution based and bulk characterization of 1:1 mixtures. When equimolar n heptane solutions of 2 2c and 2 14c were mixed, the solutions changed from colorless to orange, indicative of a donor acceptor CT complex. The association constant ( K CT 4 33 x 10 2 M 1 ) of the complex was determined through Benesi Hildebrand analysis along with the molar absorptivity ( 2 34 x 10 3 M 1 cm 1 ) and the Gibbs free energy change ( G 3 .5 6 kcal/mol ). Upon aging of the bulk mixture, a new absorbance was observed which could indicate full charge transfer from donor to acceptor. EPR experiments could elucidate the nature of the complex if that were the case. 82 The bulk 1:1 mixture of 2 2c and 2 14c exhibited thermal behavior not indicative of its individual components. DSC, while informative for the individual components, was inconclusive in regard to the thermal transitions of the 1:1 mixture. The mesophase transitions were better understood through POM study of both the individual components and the mixture. Initially, a bicomponent mixture was present at room temperature of an aged sample, and further study elucidated that mixture to correspond
65 to a mixture of the two mesophases. The thermotropic liquid crystalline behavior is described as transitioning from fans (Col h and Col r room temperature to 75 C) to mosaic (75 C to 105 C) to isotropic (105 C and above). XRD of 1:1 mixture at room temperature revealed a possible intercolumnar distance of 29 Two dimensional XRD techniques could yield more infor mation on the liquid crystalline organizational parameters. Without compounds of high purity, charge carrier mobility measurements from 2 2 or 2 10 or 1:1 mixtures thereof with 2 14 could not be realized. Ideally, thin films of both individual components and 1:1 mixtures of would be fabricated, and carrier measurements would be performed to elucidate what improvements in performance, if any, could be observed. OFET device fabrication using the same method would also be desired. Such a device should be ana lyzed for performance and morphology so that the desperately needed links between molecular design, self assembly, and performance could be assessed. Experimental Materials and General Methods Reagents and solvents were purchased from commercial sources an d used without further purification unless otherwise specified. Tetrahydrofuran ( THF ) ether, DCM and DMF were degassed in 20 L drums and passed through two sequential purification columns (activated alumina; molecular sieves for DMF) und er a positive argon atmosphere. TLC was performed on SiO 2 60 F254 aluminum plates with visualization by UV light or staining. Flash column chromatography was performed using Purasil SiO 2 g points (m.p.) were
66 determined on a Mel temp electrothermal melting point apparatus. 300 (75) and 500 (125) MHz 1 H ( 13 C) NMR spectra were recorded on Varian Mercury 300 or Gemini 300 or Inova 500 spectrometers. Chemical shifts ( ) are given in parts per million (ppm) relative to tetramethylsilane ( TMS ) and referenced to residual protonated solvent ( chloroform d CDCl 3 : dimethylsulfoxide d 6 DMSO d 6 re s (singlet), d (doublet), t (triplet), q (quartet), quin (quintet), hp (heptet), b (broad), and m (multiplet). UV Vis absorption spectra were obtained using a Cary 100 Bio spectrophotometer as described in detail later. Electrospray ionization time o f flight ( ESI TOF ) and direct analysis in real time time of flight ( DART TOF ) mass spectrometry ( MS ) spectra were recorded on an Agilent 6210 TOF spectrometer. Chemical ionization mass spectrometry ( CI MS ) spectra were recorded on a Thermo Trace GC DSQ (single quadrupole) spectrometer. Elemental analyses were performed on a Carlo Erba 1106 instrument. BTF ( 2 1 ) and its precursors ( 2 6 through 2 8 ) were all synthesized and characterized according to literature procedures and characterization information. 71 TRFs ( 2 2a,b, c ) and m TRFs ( 2 12 a ,b, c ) were also synthesized via a similar, yet modified version of a previously published procedure. 66 Lastly, TIMs ( 2 1 4 a ,b, c ) were new compounds yet were similar to previously reported compounds and were synthesized via a modified version of a literatur e procedure. 72
67 Synthetic Procedures General procedure for the preparation of benzotrifurans ( 2 2 a ,b,c) and methyl benzotrifurans (2 12 a ,b, c). These compounds were prepared according to a modified version of the synthesis f ound in the literature 66 2 1 or 2 11 (0.407 mmol) was placed in a 50 mL three arm round bottom flask with a stir bar under inert conditions. Tetrahydrofuran (15 mL) was added, the flask was chilled to 0 C, and 2 1 or 2 11 was dissolved with stirring. Keeping the flask chilled, the appropriate acid chloride was added ( 2 2 a 2 12 a : acetyl chloride; 2 2 b 2 12 b : octanoyl chloride; 2 2 c 2 12 c : lauroyl chloride, 3.5 equiv) and stirred for 10 min. Then, triethylamine (3.5 equ iv) was added dropwise to the solution. The solution was then stirred for 12 h and allowed to warm gradually to rt. The solution was worked up by pouring the mixture into deionized ( DI ) water (300 mL), then extracted with ethyl acetate (300 mL). The ext ract was then washed with DI water (100 mL), then brine (200 mL). Flash chromatography was utilized to achieve higher purity compounds using the conditions specified.
68 Benzo[1,2 b:3,4 b':5,6 b'']trifuran 2,5,8 triyl triacetate ( 2 2 a) This compound was prepared according to the general procedure for the preparation of benzotrifurans. It was purified via flash column chromatography with 20% ethyl acetate in hexanes. Compound 2 2 a was isolated as a white powder in 68% yield (0.158 g): 1 H NMR (CDCl 3 1 H NMR data matches that reported in the literature. 66 Benzo[1,2 b:3,4 b':5,6 b'']trifuran 2,5,8 triyl trioctanoate ( 2 2 b) This compound was prepared according to the general procedure for the preparation of benzotrifurans. It was purified via flash column chromatography with 25% dichloromethane in hexanes. Compound 2 2 b was isolated as a brownish red oil, 49% yield (0.258 g); 1 H NMR (CDCl 3 J = 7.5 Hz), 1.79 (quin, 2H, J = 7.4 Hz), 1.50 1.32 (m, 8H), 0.90 (t, 3H, J = 6.6 Hz); 13 C NMR (CDCl 3 31.7, 29.0, 24.7, 22.7, 14.2 (note: there are two carbon peaks missing from the list); HRMS (ESI TOF) calculated 647.3191 for C 36 H 48 O 9 Na (M+Na) + found 647.3185.
69 Benzo[1,2 b:3,4 b':5,6 b'']trifuran 2,5,8 triyl tridodecanoate ( 2 2 c) This compound was prepared according to the general procedure for the preparation of benzotrifurans. It was purified via flash column chromatography fir st with 2.5% ethyl acetate in hexanes, then with 25% dichloromethane in hexanes. Compound 2 2 c was isolated as a white powder in 20% yield (0.100 g): 1 H NMR (CDCl 3 J = 7.5 Hz), 1.79 (quin, 2H, J = 7.33 Hz), 1.51 1.24 (m, 18H), 0.88 (t, 3H, J = 6.5 Hz); the 1 H NMR data matches that reported in the thesis of Yan Li. 68 2,2',2'' (2,4,6 T rimethoxybenzene 1,3,5 triyl)tripropanenitrile ( 2 9 ) Compound 2 7 (6.93 g, 24.3 mmol) was placed in three armed 250 mL round bottom flask with stirring under inert conditions. Dry N N dimethylformamide (100 mL) was added to the flask, 2 7 was dissolved, and the solution was chilled to 0 C. Sodium hydride (60% in mine ral oil, 3.36 g, 80.2 mmol) was added in portions to the mixture, which caused the solution to turn yellow and evolve gas. After bubbles stopped evolving, methyl iodide (5.0 mL, 80 mmol) was slowly added to the solution, which caused the solution to turn colorless and turbid. The mixture was gradually warmed to room temperature over 12 h with stirring. The product was worked up by pouring the mixture into DI water (500 mL) and extracted using ethyl acetate (1 L). The extract was washed with DI water (2 L) and brine (300 mL) and dried over Na 2 SO 4 Flash column chromatography with 25% ethyl acetate in hexanes afforded compound 2 9 (a mixture of two diastereomers) as white crystals in 84% yield (6.7 g). 1 H NMR (CDCl 3 4.21 (m, 3H), 4.04 (s, 3H), 4 .00
70 (s, 3H), 1.71 1.62 (m, 9H); the 1 H NMR data matches that reported in the thesis of Yan Li 68 3,6,9 T rimethylbenzo[1,2 b:3,4 b':5,6 b'']trifuran 2,5,8(3H,6H,9H) trione, methyl benzotrifuranone, methyl BTF m BTF ( 2 11 ). Compound 2 9 (6.7 g, 21 mmol) and concentrated hydrobromic acid (48% by weight in water, 70 mL, 610 mmol) were placed in a 250 mL round bottom flask and heated to reflux with stirring over 12 h. Saponification was performed by dissolving sodium hydroxide (35 g, 880 mm ol) in DI water (approximately 100 mL), then adding that solution to the flask until the mixture reached a pH of approximately 10 (as determined by litmus paper). The solution was heated to 70 C for 1 h with stirring, then cooled to room temperature, the n chilled to 0 C. The mixture was then acidified by slow addition of hydrochloric acid (12.1 N, 80 mL) until the mixture reached a pH of approximately 2. During acidification, the solution changed from brown to yellow. The product was extracted with et hyl acetate (600 mL) and dried over MgSO 4 After concentration in vacuo, 7.0 g (> 90% yield) of white crystals of 2 10 were isolated. The crystals (1.34 g, 3.91 mmol), p toluenesulfonic acid monohydrate (0.146 g, 0.768 mmol), and toluene (20 mL) were add ed to a 25 mL round bottom flask affixed with a Soxhlet apparatus and reflux condenser. The apparatus was wrapped in successive layers of aluminum foil, fiberglass insulation, and then more aluminum foil. The Soxhlet thimble contained 4 molecular sieves (1.5 g). The solution was then heated to reflux for 2 days. After cooling to room temperature, the
71 solution was dissolved in ethyl acetate (150 mL), washed with NaHCO 3 (150 mL), brine (75 mL), dried over Na 2 SO 4 and concentrated in vacuo. Flash column c hromatography using 20% ethyl acetate in hexanes yielded the product 2 11 a white powder in 73% yield (0.80 g) as a mixture of diastereomers. 1 H NMR (CDCl 3 3H, J = 7.50 Hz); the 1 H NMR data matches that reported in the thesis of Yan Li 68 3,6,9 T rimethylbenzo[1,2 b:3,4 b':5,6 b'']trifuran 2,5,8 triyl triacetate ( 2 12 a) This compound was prepared according to the general procedure for the preparation of benzotrifurans. It was purified via recrystallization by vapor diffusion of pentane into a concentrated solution of the product and ethyl acetate, which yielded 2 12 a in 24% (0.03 g). 1 H NMR (CDCl 3 1 H NMR data matches that reported in the thesis of Yan Li. 68
72 3,6,9 T rimethylbenzo[1,2 b:3,4 b':5,6 b'']trifuran 2,5,8 triyl trioctanoate ( 2 12 b) This compound was prepared according to the general procedure for the preparation of benzotrifurans. It was purified via flash column chromatography with 30% dichloromethane in hexanes. Compound 2 12 b was isolated as a white powder in 23% yield (0.80 g): 1 H NM R (CDCl 3 J = 7.45 Hz), 2.30 (s, 3H), 1.81 (quin, 2H, J = 7.34 Hz), 1.33 1.50 (m, 8H), 0.91 (t, 3H, J = 6.37 Hz); 13 C NMR (CDCl 3 148.3, 140.3, 111.2, 97.6, 33.9, 31.9, 29.2, 29.1, 25.0, 22.8, 14.3 (note: there are two carbon pea ks missing from the list); HRMS (APCI TOF) calculated 667.3841 for C 39 H 55 O 9 (M+H) + found 667.3868. 3,6,9 T rimethylbenzo[1,2 b:3,4 b':5,6 b'']trifuran 2,5,8 triyl tridodecanoate ( 2 12 c) This compound was prepared according to the general procedure for the preparation of benzotrifurans. It was purified via flash column chromatography first with 5% ethyl acetate in hexanes, then with 25% dichloromethane in hexanes. Compound 2 12 c was isolated as a colorless oil, 38% (0.166 g): 1 H NMR (CDCl 3 J = 7.42 Hz), 2.29 (s, 3H), 1.80 (quin, 2H, J = 7.26 Hz), 1.27 1.50 (m 16H), 0.88 (t, 3H, J = 6.37 Hz); the 1 H NMR data matches that reported in the thesis of Yan Li. 68
73 General procedure for the preparation of mellitic triimides (2 14 a ,b, c) These compounds were prepared via a modified version of the literature procedure 72 Mellitic acid (1.0 equiv) was added to a 25 mL round bottom flask with DI water (8 mL) and dissolved with stirring. Then, the appropriate primary amine (3.0 equiv) was added to the flask along with acetone (4 mL). The mixture was stirred for 25 min, then the sol ution was concentrated in vacuo. The mixture was then placed into the drying oven set to approximately 140 C for 4 days. The product was removed, and after cooling, was purified by flash column chromatography. If the product was still impure after colu mn chromatography, then the combined fractions were washed with NaHCO 3 (100 mL), DI water (100 mL), and brine (100 mL), and dried over Na 2 SO 4 The resulting solution was then concentrated in vacuo. 2,5,8 T rimethyl 1H dipy rrolo[3,4 e:3',4' g]isoindole 1,3,4,6,7,9(2H,5H,8H) hexaone ( 2 14 a) This compound was prepared according to the general procedure for the preparation of mellitic triimides (0.145 g, 0.420 mmol of mellitic acid and 0.11 mL, 1.27 mmol of methylamine, 40% b y weight). DI water (7 mL) was used without acetone. The
74 crude compound was purified via flash column chromatography with 33% ethyl acetate in hexanes. Compound 2 14 a was isolated as a white powder in 40% yield (0.140 g). 1 H NMR (CDCl 3 13 C NMR (CDCl 3 TOF) calculated 350.0384 for C 15 H 9 N 3 O 6 Na (M+Na) + found 350.0398 (confirmed by C 30 H 18 N 6 O 12 Na (2M+Na) + calculated 677.0875 and found 677.0878). 2,5,8 T riheptyl 1H dipyrrolo[3,4 e:3',4' g]isoindole 1,3,4,6,7,9(2H,5H,8H) hexaone ( 2 14 b) This compound was prepared according to the general procedure for the preparation of mellitic triimides (0.192 g, 0.561 mmol of mellitic acid and 0.253 mL, 1.68 mmol of h eptylamine). DI water (2 mL) was used to dissolve mellitic acid and no acetone was used. The crude compound was purified via flash column chromatography with 100% dichloromethane. Compound 2 14 b was isolated as a white powder in 45% yield (0.150 g). 1 H NMR (CDCl 3 b t, 2H, J = 7.2 Hz), 1.70 (m, 2H), 1.24 1.31 (m, 8H), 0.85 (t, 3H, J = 6.5 Hz); 13 C NMR (CDCl 3 28.2, 26.8, 22.6, 14.1; HRMS (ESI TOF) calculated 580.3381 for C 33 H 46 N 3 O 6 (M+H) + found 580.3374 (conf irmed by C 33 H 45 N 3 O 6 Na (M+Na) + calculated 602.3201, found 602.3205).
75 2,5,8 T ridodecyl 1H dipyrrolo[3,4 e:3',4' g]isoindole 1 ,3,4,6,7,9(2H,5H,8H) hexaone (2 14 c) This compound was prepared according to the general proced ure for the preparation of mellitic triimides (0.409 g, 1.20 mmol of mellitic acid and 0.710 g, 3.82 mmol of dodecylamine). DI water (5 mL) and acetone (20 mL) were used. The crude compound was purified via flash column chromatography with 20% ethyl acet ate in hexanes. Compound 2 14 c was isolated as a white powder, in 40% yield (0.140 g): 1 H NMR (CDCl 3 b t, 2H, J = 7.13 Hz), 1.70 (m, 2H), 1.16 1.29 (m, 18H), 0.86 (t, 3H, J = 6.59 Hz); 13 C NMR (CDCl 3 29.5, 29.4, 29.4, 29.1, 28.1, 26.8, 22.6, 14.1; HRMS (ESI TOF) calculated for C 48 H 76 N 3 O 6 (M+H) + 790.5731, found 790.5729.
76 Characterization of CT Complex by U ltraviolet Vis ible ( UV Vis ) Spectrometry Materials and equipment. HPLC grade dichloromethane (DCM, Fischer Scientific, submicron filtered, 240 nm cutoff) and n heptane (Fischer Scientific, submicron filtered, 198 nm cutoff) were both used as purchased. Plastic syringes were purchased through National Scientific Company (1 mL) and used with Precision G lide #1710, Hamilton Co.) were used with the provided syringe tips. All solvent volume measurements were made after all air bubbles were removed from the syringe. An Infrasil Spectrophotometric 1 mm cuvette was used (Starna Cells, Inc.) in a Cary 100 Bio spectrophotometer. Sonication (Fischer Scientific, FS30H) was used to help dissolve samples as well as gentle heating by a heat gun (MHT Products, Inc., Model 750VT). Cell preparation. The cell was initially rinsed three times with DI water, then three times with acetone, then three times with DCM, and dried by flowing nitrogen. Then, the cell was wiped with a Kimwipe and blasted with nitrogen again to remove any particulates. The in side of the cell was then rinsed with n heptane three times and dried by flowing nitrogen. At this point, the cell was ready for sample addition and spectrum collection. Sample preparation for modified titration experiment. Compounds 2 2 c (7.9 mg, 0.010 mmol) and 2 14 c (7.9 mg, 0.010 mmol) were weighed out in separate, clean vials, dissolved in DCM and then combined together in one vial. The mixture was evaporated at atmospheric pressure, then placed on high vacuum for 10 minutes. Then, the solution was prepared by adding 1 mL of solvent by either plastic or glass syringe. For example, to obtain the 95:5 n heptane/DCM solution, 0.9 mL of n heptane
77 was drawn into a plastic syringe and deposited into the vial containing the mixture. Then, n heptane was measured using the glass syringe and added to the added to the mixture. After covering the solution, it was subjected to sonication and/or gentle heati ng as need to dissolve the compounds by visual inspection. Then, enough sample was added via pipette to the spectrophotometric cell to reach the upper neck of the cell, and the spectrum was collected. To obtain a blank sample for baseline correction, a s ample of 50:50 DCM/ n heptane was prepared. Sample p reparation for dilution experiment. Compounds 2 2c (7.9 mg, 0.010 mmol) and 2 14c (7.9 mg, 0.010 mmol) were weighed out in separate, clean vials, dissolved in DCM and then combined together in one vial. That mixture was then evaporated down at atmospheric pressure, then placed on high vacuum for 10 minutes. Then, the appropriate amount of n heptane was measured to create the specific concentration and added to the mixture, which was sonicated and/or gent ly heated until the mixture had dissolved by visual inspection. For example, to create a 5 mM solution, 2.0 mL was added to the vial. Then, enough sample was added via pipette to the cell to reach the upper neck of the cell, and the spectrum was collecte d. To obtain a blank sample for baseline correction, a sample of 100% n heptane was used. Post collection procedures. After collection, the mixture was returned to the original vial with three to four rinses of the cell with DCM (each rinse was collected in the original vial as well). The cell was then rinsed with n heptane three times and dried under flowing nitrogen. The solution in the vial was evaporated at atmospheric pressure until all solvent appeared gone, then placed on high vacuum for 10 20 min utes. After
78 that period of time, the proce dure was repeated as necessary resulting in approximately 30 45 minutes total between UV Vis scans Characterization of Mixtures by Polarizing Optical Microscopy (POM) Slide preparation. Fisherfinest Premium Micr oscope slides (3 inches by 1 inch by 1 mm) were cleaned by rigorous rinsing with acetone, then wiped while still wet with Kimwipes in the same direction (each wiping stroke was performed longways along the slide) until no residue remained. The same proced ure was applied to the slide covers (Fisherfinest Premium Cover Glass, 25 by 25 by 1 mm). The slide was then blasted with flowing nitrogen to remove any particulates and gently heated with a heat gun when ready to receive a sample. Sample preparation. Co mpounds 2 2c (7.9 mg, 0.010 mmol) and 2 14c (7.9 mg, 0.010 mmol) were weighed out in separate, clean vials, dissolved in DCM and then combined together in one vial. The remaining vial was rinsed with DCM, and that was also combined with mixture. That mix ture was then evaporated down at atmospheric pressure, then placed on high vacuum for 10 minutes. After removal from vacuum, the sample was gently heated with a heat gun until melted. Then, a cleaned, heated metal spatula was used to collect a small amou nt of the heated mixture (approximately 5 mg) and to smear the sample onto the heated slide. Immediately, the slide cover was placed on top of the sample. POM characterization techniques. The slide was then placed a Leica DMLP POM with Linkam TS93 heatin g stage and a Leica DFC295 camera with Leica Application Suite 3.3.1 software. Several different rates of heating and cooling were utilized for each sample (0.1, 0.2, 1, 2, 5, 10, 20 C/min) in order to elucidate any
79 mesophasic behavior. Temperature scan ning ranges varied from 50C to 150C. When crystals or liquid crystals, the slide cover would be sheered to test for liquidity. Differential Scanning Calorimetry (DSC) Analysis Sample preparation of single components. Either 2 2c or 2 14c was placed on high vacuum for 10 minutes. TGA was performed on approximately 2 mg of the sample to find the temperature of 5% weight loss. Then, approximately 2 mg of either 2 2c or 2 14c was scooped into an aluminum hermetic pan and sealed with a hermet ic lid (TA) where the mass of each component (the pan, lid, and substance) was recorded. Sample preparation of 1:1 mixture. Compounds 2 2c (7.9 mg, 0.010 mmol) and 2 14c (7.9 mg, 0.010 mmol) were weighed out in separate, clean vials, dissolved in DCM and then combined together in one vial. The remaining vial was rinsed with DCM, and that was also combined with mixture. That mixture was then evaporated down at atmospheric pressure, then placed on high vacuum for 10 minutes (which was sufficient to remove all solvent as revealed by TGA). TGA was performed on approximately 2 mg of the sample to find the temperature of 5% weight loss. Then, approximately 2 mg of the mixture was scooped into an aluminum hermetic pan and sealed with a hermetic lid (TA) where the mass of each component was recorded. Sample analysis. The sample was then loaded into TA Instruments DSCQ1000 0620 V9.9 by robotic arm and analyzed against an empty reference pan using Universal Analysis 2000 4.4A software. The analysis temperature ra nge spanned from 80C to about 10C below the temperature of 5% weight loss (as observed by TGA). Ramp rates were either 2.5, 5, or 10C/min, and each sample underwent a minimum of two heating and cooling cycles. For example, for 2 2c the sample was lo aded, then cooled to 80C and held at that temperature for five minutes. Then, it was heated at a rate of
80 5C/min until it reached 80C, where it was held for 5 minutes, then cooled back down to 80C at the same rate. That heat and cool cycle was repea ted two more times. Thermogravimetric Analysis (TGA) Sample and equipment preparation. Single components were prepared for analysis by placement on high vacuum for 10 minutes prior. For the 1:1 mixture, 2 2c (7.9 mg, 0.010 mmol) and 2 14c (7.9 mg, 0.010 m mol) were weighed out in separate, clean vials, dissolved in DCM and then combined together in one vial. The remaining vial was rinsed with DCM, and that was also combined with mixture. That mixture was then evaporated down at atmospheric pressure, then placed on high vacuum for 10 minutes. The platinum TGA pan was heated directly in a Bunsen burner flame until red hot. If residue still existed, the pan was allowed to cool to room temperature, then the residue was rinsed with DI water. Then the pan was heated in the same fashion again. Sample analysis. After taring a clean platinum pan, the analysis was performed by placing approximately 2 mg in the pan, placing the sample in the autoloader of the TA Instruments TGA Q5000 0121, and taking the sample fr om room temperature to 400C at a rate of 20C/m in under a nitrogen atmosphere. Analysis of Spin Cast Film by X ray Diffraction ( XRD ) Sample preparation. 2 2c (7.9 mg, 0.010 mmol) and 2 14c (7.9 mg, 0.010 mmol) were weighed out in separate, clean vials, dissolved in DCM and then combined together in one vial. The remaining vial was rinsed with DCM, and that was also combined with mixture. That mixture was then evaporated down at atmospheric pressure, then placed on high vacuum for 10 minutes. After removing from vacuum, the sample was dissolved in approximately 1 mL of DCM and drawn into a syringe. A slide square was cut, then cleaned using successive rinses of DI water, ethanol, then
81 acetone. The square was then wiped with a Kimwipe and placed onto the spin caster. Two films were cast, one at 3000 rpm, the other at 1000 rpm. The samples were then placed onto the POM heating stage (Linkam TS93) and annealed at 40C overnight. Analysis of spin cast films. After examining th e films by POM, it was determined that the 3000 rpm film would be used in the SAXS experiment due to a more uniform texture. In collaboration with the Department of Materials Science and Engineering at the University of Florida, Dr. Valentin Craciun perfo rmed the analysis
82 CHAPTER 3 IMPROVED PERFORMANCE, BULK ORGANIZATION, AND CHARGE TRANSPORT OF A SMALL MOLECULE OPV DEVICE THROUGH HYDROGEN BONDED 1D COLUMNAR AGGREG ATES Background Organic photovoltaics (OPVs) have stood at the forefront of rapid technological advancements for the past decade because they represent a potentially clean, renewable, synthetically tunable, low cost, and aesthetically pleasing mode of energy generation. 1,7,15,16,84 The highest performing OPV devices utilize a BHJ active layer ( vide supra ), a device architecture that has essentially become the standard for polymer and small molecule based devices. 22 However, power conversion efficienci es (PCEs, current) have not met the performance level of traditional silicon based solar panels, which typically exhibit approximately 18% PCE. While 18% PCE is no t necessarily fundamental to the commercialization of OPVs, overall improvements are. Figure 3 1h further highlights the disparity of OPV PCEs (red squares) relative to other photovoltaic technologies. 85 To examine this problem, a discussion of the photo physical and electrical properties of BHJ based OPV devices is necessary. In a BHJ OPV device ( Figure 3 1) conjugated chromophore) is responsible for absorbing visible light, so much research has been devoted to the synthesis of molecules that absorb light over much of the solar spectrum. Photons of energy equal to the energy difference between the highest occupied molecular orbital ( HOMO ) and the lowest occupied molecular orbital ( LUMO ) called the band gap, are ab sorbed by the donor material, and an exciton is generated (Figure 3 1a) Energies describing the HOMO, LUMO, and band gap are typically gained from
83 characterization or computation methods wherein the donor material exists in its molecular, unaggregated st ate. In device contexts, these values can change due to the formation of bands (i.e., valence and conduction bands), but are generally accepted in the OPV community to be the overall descriptor of energy levels for a device. The exciton must then diffuse to the molecular interface of the donor and acceptor materials (Figure 3 1b) At the interface, a charge transfer event occurs (Figure 3 1c) which results in charge separation and the subsequent formation of charge carriers (electrons and holes Figure 3 1d ). These carriers must then reach their respective electrodes (indium tin oxide, ITO or aluminum) to produce usable electrical current (Figure 3 1e) The success of the BHJ architecture is derived from the nanoscale morphology that is produced, which should ideally feature domain sizes (10 20 nm) compatible with the exciton diffusion length for organic molecules (5 10 nm ) 1 However, as previously mentioned, the nanoscale aggregation of BHJ active layers remains random. While excitons can readily d iffuse to the donor acceptor interface, PCEs can and do suffer from inefficient charge transport. The formation of de (Figure 3 1 f,g respectively) local morphologies prevents charge carriers from reaching their respective electr odes and leads to charge recombination. Improvements in the nanoscale morphology are typically gained from solvent or thermal annealing or the use of additives 86 89 These top down approaches ultimately result in PCE improvements generally attribut ed to improved BHJ morphology. Absent are methods that employ bottom up strategies targeted at improving morphologies, and it was this approach that this research aimed to explore.
84 In order to understand the relationship between design components employed in a bottom up approach and their effects, a structure property relationship study must be developed. Generally, this study is affected by comparing some extrinsic aspect (e.g., power conversion efficiency of an OPV) of a family of molecules that are int rinsically similar but bear some differing functionalization. Frequently, in OPV research, thes e studies are aimed at improving chromophore performance 1,15 and less targeted toward improving bulk molecular organization. 1,7 Specifically, this project sought to prove that OPV device performance increases throughout a family of intrinsically similar c hromophores were a direct result of their respective self assembly.
85 Figure 3 1 Cartoon representation of bulk heterojunction ( BHJ ) organic photovoltaic ( OPV ) device and experimental device efficiencies (where blue and orange represent the donor and acceptor material respectively) depicting the proce ss of generating a photocurrent : (a) photoexcitation, (b) exciton diffusion, (c d) charge transfer and separation, (e) charge migration; and (f) isla nd and (g) cul de sac local BHJ morphologies ; (h) best research photovoltaic cell efficiencies with OPVs at the bottom (red squares)
86 General Approach and Experimental Design The bottom up approach of improving OPV PCE detailed herein involves rational des conjugated chromophore with the capability to form robust assemblies favorable for efficient charge transport when co deposited into a BHJ with C 60 Accordingly, assemblies producing one dimensional ( 1D ) columnar nanostructures were targeted as possible candidates as they have been shown to exhibit favorable charge transport characteristics ( vide supra ). T he essence of the strategy ( Figure 3 2 ) involves appending a chromophore ( D Figure 3 2 a) with a hydrogen bonding ( H bonding ) self assembly unit ( HB Figure 3 2 a) in such a fashion that will produce some bi tri or multi molecular plana r discotic aggregate (Figure 3 2 b). Using H bonding as a means for providing BHJ organization was discussed previously in Chapter 1 The inherent planarity of the aggregate should allow for possible a ssembly into columns (Figure 3 2 c), which can potentially form a favorable 1D aggregate structure whose intercolumnar voids will be filled with acceptor material when co depo sited on a substrate (Figure 3 2 d).
87 Figure 3 2 Schematic representation for creating improved OPV device performance showing (a) molecular design and (b d) self assembly strategy Among many candidates capable of cyclic homomeric assembly (Figure 3 3a e ) the phthalhydrazide H bonding motif was initially targeted due to self complementary H bonding contacts (Figure 3 3e ) Previously, tautomeric forms (Figure 3 3f) of phthalhydrazides were shown to form trimeric aggregates (Figure 3 3g ) 90 92 which, when appropriately functionalized, could form liquid crystalline columnar assemblies. 90 Additionally, synthetic routes to phthalhydrazides contain intermediates that were envisioned as non H bonding congeners or H phthalhydrazide, necessary for structure property relationship study. Further, high yielding coupling reactions were envisioned as the synthetic key to creating the fully realized phthalhydrazide chromophore assembly. Venerable reactions like Suzuki, Sonogashira, and Stille cross couplings have such wide utility that examin ing the modularity of this strategy becomes a logical outcome area of this project as well. (a) (b) (c) (d)
88 Essentially, a library of H bonding and chromophore modules could be built that when mixed and matched could provide valuable structure property performance inform ation for the effective implementation of this strategy. Figure 3 3 Examples of hydrogen bonding ( H bonding ) units: (a) modified guanine, (b) GC hybrid, (c) modified isophthalic acid, (d) modified melamine, and (e) phthalhydrazide whose (f) tautomerization and (g) self assembly into trimeric discs is shown conjugated chromopho re that can absorb a broad range of light and have a narrow band gap These features of OPVs remain a great advantage over traditional silicon based solar cells, which have exceptional charge transport characteristics but poor absorption profiles. Common oligomeric chromophore units like phthalocyanine, oligothiophene, boron subphthalocyanines, and diketopyrrolopyrole are shown in Figure 3 4. Polymeric chromophore units were excluded from this study due to their inherent polydispersity, which would make a definitive structure property relationship difficult to derive as well as their incompatibility with vacuum deposition (a) (b) (c) (d) (e) (f) (g)
89 Of those chromophores, phthalocyanine and oligothiophene based derivatives were chosen for study. Phthalocyanines (Figure 3 3 a ) have been synthetically functionalized, 93 96 shown to aggregate into 1D columnar assemblies in presence of C 60 97 and appended with H bonding motifs. 9,98 Furthermore, our collaborator (Prof. Jiangeng Xue, Department of Engineering, University of Florida) has extensive experience in fabricating and understanding the physical nature of OPV BHJ devices employing phthalocyanines. 2,99,100 Accordingly, the wealth of knowledge surrounding this chromophore unit lead to its logical choice as a module worthy of study within the design strategy. Oligothiophene s also have been extensively used in BHJ OPVs and synthetical ly manipulated to create a number of easily accessible derivatives (branched and linear) relevant to this study. The branched quaterthiophene unit (Figure 3 4d) was employed due to its apparent expedient synthesis and previous use as a repeat unit in a dendrimeric chromophore system whose use in a solution processed BHJ OPV elicited a 1.24% PCE 101 Functionalization at the R position would allow for alkyl chain tailoring to accommodate requirements for thin film preparation via vacuum deposition or solu tionsolution processing. Therefore, with the design of two possible HB D systems (a phthalhydrazide branched quaterthiophene and a phthalhydrazide phthalocyanine) in hand, work toward preparing these target s could begin.
90 Figure 3 4 Chemical structures of potential donor chromophore units (a) phthalocyanine, (b) boron subphthalocyanine, (c) diketopyrrolopyrrole, and (d) oligothiophene Low level gas phase mode ling was performed on assemblies of the phthalhydrazide tautomer (Figure 3 3f) coupled to the branched quaterthiophene (Figure 3 4d, R groups approximated as methyl groups for simplicity) using the AMBER* forcefield (as implemented in MacroModel v. 9.1) to initially probe the self assembly behavior, to ensure that columnar aggregation was possible, and to examine the dimensionality of the aggregates ( Figure 3 8 for the results). To form the trimeric aggregate, the H bonding contacts were initially conf ormationally locked, and the trimer was minimized. Subsequent removal of the conformational locks, and re minimization iteratively duplicating the trimer and minimizing the new ly generated aggregate ( Experimental for further computational details). Then, pre minimized C 60 molecules were added and the entire model was minimized again. While some of the thiophenes within the chromophore unit show some deviation from planarit y, the overall results stacked aggregation of this HB D system blended with C 60 (a) (b) (c) (d)
91 Figure 3 5 Computationally derived images of the trimeric aggr egated tautomer of 3 5a and C 60 (blue) viewed from (a) top down and (b) side on showing 1D columnar organization favorable for charge transport; these structures were modeled using the AMBER* forcefield Synthesis of H ydrogen B onding and Chromophore Modules Synthesis of Phthalocyanine Module Phthalhydrazide phthalocyanine HB D systems targeted for study are shown in Figure 3 5 ( 3 1a,b ). Literature preparations and purification of phthalocyanine derivatives initially seemed relatively facile using a Lewis acid (e.g., zinc (II) acetate) to annulate phthalonitrile; 93,95,96,98,102 105 however, these strategies utilize functionalized phthalonitriles such as 3 3a to form 3 4a (a mixture of regioisomers), which allow for facile purification and handling. t Butyl groups could not only disrupt planarity and stacking, but also remove the option of vacuum deposition due to increased molecular weight. Further, the incorporation of an aryl halide into the phthalocyanine structu re was necessary for future cross coupling reactions, and could be achieved by incorporating an equivalent of halogenated phthalonitrile into the annulation reaction mixture. This then manifests as a final reaction mixture of non mono and multi haloge nated products, which would require purification. Accordingly, (a) (b)
92 a balance between solubility maintaining molecular planarity and processability would be needed to realize a phthalocyanine derivative. The preparation of a maximally planar derivative ca pable of being vacuum deposited was attempted first. 4 Iodophthalonitrile ( 3 2 Figure 3 6 Phthalhydrazide phthalocyanine HB D systems targeted for study are shown in Figure 3 5 ( 3 1a,b ). Literature preparations and purification of phthalocyanine derivatives initially seemed relatively facile using a Lewis acid (e.g., zinc (II) acetate) to annulate phthalonitrile; 93,95,96,98,102 105 however, these strategies utilize functionalized phthalonitriles such as 3 3a to form 3 4a (a mixture of regioisomers ), which allow for facile purification and handling. t Butyl groups could not only disrupt planarity and stacking, but also remove the option of vacuum deposition due to increased molecular weight. Further, the incorporation of an aryl halide into the phthalocyanine structure was necessary for future cross coupling reactions, and could be achieved by incorporating an equivalent of halogenated phthalonitrile into the annulation reaction mixture. This then manifests as a final reaction m ixture of un mono and multi halogenated products, which would require purification. Accordingly, a balance between solubility maintaining molecular planarity and processability would be needed to realize a phthalocyanine derivative. The preparation of a maximally planar derivative capable of being vacuum deposited was attempted first. 4 Iodophthalonitrile ( 3 2 Figure 3 5 ) and phthalonitrile ( 3 3 b Figure 3 6 ) were dissolved in dimethylformamide ( DMF ) in a 1:3 stoichiometric ra tio with zinc (II) acetate and heated to reflux. M ono functionalized zinc (II) phthalocyanine product ( 3 4 b Figure 3 6 ) was verified through mass spectrometry
93 of the crude product, but isolation of pure 3 4 b through column chromatography and recrystalli zation was not possible presumably due to poor solubility Subsequent borylation or cross coupling attempts of crude 3 4b with phthalhydrazide (whose synthesis is forthcoming) also proved unsuccessful. Sacrificing the opportunity for vacuum deposition but maintaining planarity and increasing solubility, an alkoxy version was attempted using 3 3c and 3 2 under similar conditions, and initially appeared spectroscopically promising, but proved difficult to cultivate in high yields and purify by conventional methods Eventually, several different tweaks in synthetic procedures were employed like changing solvents, equivalents of Lewis acid, changing the type of Lewis acid and using microwave based technique s, all of which proved fruitless The difficult synthesis as well as the emergence of fully synthesized and purified 3 1 a or 3 1b r esulted in termination of this HB D system.
94 Figure 3 6 Tar geted system consisting of phthalhydrazide and phthalocyanine ( 3 1a,b ) and subsequent reactions involved in its preparation
95 Synthesis of Branched Quaterthiophene Module T arget molecules conceived using the phthalhydrazide b ranched quaterthiophene HB D system can be seen in Figure 3 6 where 3 5a,b represent the H ( 3 5at,bt represent one tautomeric form of 3 5a,b ) and 3 6a,b and 3 7 represent the H Figure 3 7 Chemical structure s H 3 5a,b ) and H 3 6a,b and 3 7 ) phthalhydrazide branched quaterthiophene HB D system The synthe sis of 3 5a,b 3 6a,b and 3 7 came through a collaborative effort with Dr. Jing Zhang ( 3 5a,b and 3 6a,b ) and Dr. Davita Watkins ( 3 7 ). While I have performed aspects of this synthesis (e.g., the Kumada coupling lithiation borylation, and Suzuki cross coupling) I have not completed it in its entirety. For simplicity synthetic efforts outside of the main scheme as performed by myself will be detailed. Synthetic details of the primary route will not be reported as part of this chapter and instead as Appendix C
96 T he branched quaterthiophene moiety with both methyl and hexyl alkyl groups ( 3 16a,b ) was prepared by a modified iterative approach found in the literature involving thiophene bromination and Kumada cross coupling reactions ( see Figure 3 7 ) 101 Briefly, commercially available 3 8a or previously prepared 3 8b was brominated with N bromosuccinimide in chloroform to give compound 3 9a,b Next, commercially available 3 10 was treated with magnesium turnings under anhydrous conditions to yield the Gr ignard reagent that reacted with 3 9a,b to yield the Kumada coupled product 3 11a,b Subsequent bromination and Kumada coupling with metallated 3 1 3 afforded 3 1 4a,b in good yield. Lithiation and borylation of 3 1 4a,b with 3 1 5 provided the corresponding boronic ester and Suzuki coupling precursor 3 1 6a,b
97 Figure 3 8 Reaction scheme detailing the formation of the Suzuki precursor ( 3 13 ); yields are shown for R = Me only Synthesis of Phtha lhydrazide Module and Subsequent Functionalization Originally, the phthalhydrazide module synthesis was attempted according to a procedure found in the literature ( Figure 3 9 ) wherein 4 nitrophthalic acid ( 3 17 ) was esterified ( 3 18 ), hydrogenated ( 3 19 ), and reacted using Sandmeyer conditions to give the halogenated phthalate diethyl ester ( 3 20 ). 106 Conversion of 3 2 0 to 3 2 1 could not be affected under various conditions using combinations of hydrazine and different solvents, temperatures, and reactions times. Instead, the phthalhydrazide unit could be achieved in one step through functionalization of 3 22 (Figure 3 9) with hydrazine under acidic conditions to give 3 23 which was then protected with p methoxybenzyl chloride
98 to yield a mixture of regioisomers 3 24 a and 3 2 4 b 107 These regioisomers could be separated via column chromatography, but their individual yield i s not reported. Figure 3 9 Reaction scheme of the phthalhydrazide unit detailing the (a) original and (b) subsequent synthesis Non H bonding congeners necessary for the structure property relationship study could be made by Suzuki cross coupling of 3 16b and 3 24a to yield 3 2 5 (Figure 3 10 ) 101 but unfortunately, deprotection of the coupled product and subsequent purification to afford 3 5b could not be achieved. Conditions such as trifluoroacetic acid (TFA) at 70 C for 1 h, 107 TFA at room temperature for 8 h in DCM, and ammonium (a) (b)
99 cerium (IV) nitrate in acetonitrile and water all proved unsu ccessful. Treatment of 3 1 4b with TFA confirmed the chromophore could not withstand the deprotection conditions due to degradation, leading us to abandon this synthetic route. Figure 3 10 R eaction scheme detailing the Suzuki cross coupling and failed deprotection to achieve the target compound ( 3 5b ) Optimized preparation of the phthalhydra zide module ( Figure 3 11 ) began with acid catalyzed esterification of commercially available 3 22 to yield 3 26 The N substituted version ( 3 27 ) was derived from acid catalyzed ring expansion of 3 22 and subsequent cyclization with dimethylhydrazine. Both compounds could be cross coupled with the Suzuki precursor ( 3 16a,b for 3 26 and 3 16a for 3 27 ) under previously established conditions to afford both non H bonding congeners 3 6a,b and 3 7 Treatment of 3 6a,b with hydrazine monohydrate according to a modified literature
100 procedure resulted in the target compound 3 5a,b in low yield 108 Figure 3 11 Reaction scheme detailing the synthesis of the phthalhydrazide module including the N substituted derivative ( 3 26 3 27 ), both non H bonding congeners ( 3 6a,b 3 7 ), and the final H bonding chromophore ( 3 5a,b ) Alternative Strategies and Reactions for Linking Phthalhydrazide and Chromophore Modules As previously mentioned, the scope of reactions that can be utilized to link H bonding and chromophore modules, or more specifically 3 1 4a,b and 3 26 is quite vast. The reaction methodology developed above gave reliable and predictable yields of 3
101 5a,b but several other linking options were explored. Ethynyl bridged target compounds were envisioned through successive Sonogashira cross couplings (as seen in 3 29 ) derivatives of ei ther 3 1 4a,b or 3 26 Both strategies were applied to realize targets similar to 3 5a,b but were not carried out fully due to time constraints. Ethynyl bridged derivatives were prepared according to a modified procedure found in the literature (summarize d in Figure 3 12 ). 107 Attempts at 3 28 (F igure 3 12 ) began with a Sonogashira cross coupling of 3 24 a with trimethylsilylacetylene (TMS acetylene), which gave 3 2 8 in reasonable yield. After deprotection with tri n butyl ammonium fluoride (confirmed spectroscopically this technique was used universally for deprotection of TMS acetylene containing compounds ), 3 2 8 was subjected to a second Sonogashira cross coupling with 3 4 b with no success due to the fact that 3 4 b could never be isolated as a pure precursor.
102 Figure 3 12 Reaction scheme showing Sonogashira cross coupling with 3 2 4 a and trimethylsilylacetylene to give 3 2 8 and subsequent attempt at second cross coupling with 3 4b Ind ependent synthesis of 3 1 4a,b and 3 26 fragments allowed for application of the Sonogashira linking strategy such that an ethynyl bridged version of 3 5a,b could be envisioned (summarized in Figure 3 13 ). First, 3 1 4a,b underwent facile bromination with NBS to give 3 30 a,b Either 3 30 a or 3 26 could then be cross coupled with TMS acetylene to give TMS protected derivatives 3 3 1 and 3 3 2 (Figure 3 13 ). Deprotection and subsequent cross coupling of 3 3 2 with 3 30 a,b yie lded a non H bonding precursor, 3 3 3 a,b To improve yields of the final step, microwave assisted treatment of 3 3 3a,b was performed, and while crude results seemed spectroscopically promising, the target compound s 3 3 4 a,b were not isolated. The proton nu clear magnetic resonance ( 1 H NMR ) spectrum showed two additional protons in the Csp 2 H region suggesting the ethynyl linkage had succumbed to reduction with hydrazine. Susceptibility to side reactions of the ethynyl linkage was an initial worry for this s trategy, but the reaction
103 modules and could be easily employed on other HB D systems in the future. Figure 3 13 Reaction scheme depicting the ethynyl linking strategy and the conversion of 3 1 4a,b and 3 26 fragments to 3 34 a,b T he susceptibility of the ethynyl bridge necessitated the development of alternate linking methods. Recent literature preparations of th iophenes connected through 1,3 from terminally
104 functionalized with azides and acetylenes presented an efficient method to link 3 3 2 with 3 30 a,b 109 This one pot procedure which made use of ample 3 32 was also envisioned as providing a linkage capable of withstanding the hydrazine cyclization conditions. To put this strategy into practice, compounds 3 3 2 (after deprotection) and 3 30 a,b were mixed with a stoichiometric amount of sodium azide and catalytic amounts of copper (I) iodide, dimethylethylenediamine (DMEDA), and sodium ascorbate and heated overnight to afford the clicked products ( 3 3 5 a,b ) in low yield ( Figure 3 1 4 ). T hose products were further subjected to various conditions including treatment with hydrazine under microwave assisted conditions to affect the final clicked phthalhydrazide branched quaterthiophene target compound ( 3 3 5a,b ). Again, purification of 3 3 5 a, b proved difficult, but this strategy can now be used in linking other HB D targets. Figure 3 14 Reaction scheme showing 1,3 dipolar cycloaddition linking strategy affecting 3 35 a,b from 3 30 a,b and attempts at creating the final phthalhydrazide functionalized compound 3 35a,b
105 Structure Property Relationship Findings Intrinsic Properties of Neat Materials The structure property investigation first sought to establish that the intrinsic opti cal and electrochemical properties of 3 5a 3 6a and 3 7 were minimally different. To this end, the three derivatives were studied in dilute solution via ultraviolet visible ( UV Vis ) spectroscopy and differential pulse voltammetry (DPV). As summarized i n Table 3 1 and shown in Figure 3 1 5 tetrahydrofuran ( THF ) solution s of each possess very similar absorption profiles with max = 395 nm and onset = 464 nm. The absorbance spectra are also essent ially identical in DMF solution ( Appendix B ) In both cases, intermolecular H bonding was not expected. D ensity functional theory ( DFT B3LYP/6 31G* performed by Dr. Ron Castellano) calculations predict similar estimated HOMO and LUMO energies and wavefunctions ( Table 3 1 ). Further, time dependent density functional theory ( TD DFT ) calculations in the gas phase are fully consistent with the solution phase optical data, as the theoretically determined S 0 to S 1 transition energies vary by only 0.1 eV across the family (Appendix B). DPV measureme nts (in DMF) show that all three compounds have nearly identical oxidation potentials (versus Fc/Fc + ); these values have been determined from the oxidation onsets and converted to HOMO energies for comparison (Table 3 1 ). The reduction potentials (and cor responding LUMO energies) for 3 6a and 3 7 are also comparable; poorly reproducible electrochemical behavior for 3 5a at negative potentials as reflected in the data was troubling Both aggregation and redox chemical reactivity were speculated as the root cause, but no definitive proof of either was sought due to the prevailing similarities that the other methods afforded. The conclusion from
106 the above data is that the H properties in the molecularly dissolved state Table 3 1 Optical, electrochemical, and computational characteristics for compounds 3 5a 3 6a and 3 7 ; this table represents the collective work of Raghida Bou Zerdan (DPV), Dr. Davita Watkins (DPV and UV Vis for 3 7 ), Dr. Ron Castellano ( DFT calculations), and myself (UV Vis for 3 5a 3 6a and DPV ) Absorbance (THF) Differential Pulse Voltammetry (DM F) DFT Calculations (B3LYP/6 31G*) max (nm) onset (nm) E g (eV) E HOMO (eV) E LUMO (eV) E g (eV) E HOMO (eV) E LUMO (eV) E g (eV) 3 6a 393 462 2.68 5.61 (0.01) 3.25 (0.06) 2.37 (0.05) 5.16 2.02 3.14 3 7 397 468 2.65 5.63 (0.03) 3.30 (0.04) 2.31 (0.04) 5.16 2.06 3.10 3 5a 394 463 2.68 5.63 (0.03) 2.90 (0.03) 2.72 (0.01) 5.20 2.18 3.02 Figure 3 15 A bsorption spectra for H 3 5a and H 3 6a and 3 7 (100 M in THF ) Bulk and Solution Based Aggregation Behavior Differences between the 3 5a 3 6a and 3 7 are immediately apparent upon aggregation; the powdered bulk material of 3 6a is yellow while 3 5a varies from deep
107 orange to red (Figure 3 16 b,c ). Spectroscopic examination of solid state thin films (40 nm) prepared via vacuum thermal evaporation reveale d a red shifted absorption for 3 5a compared to the 3 6a and 3 7 films by 15 to 20 nm (Figure 3 1 6 a ). Similar behavior in other conjugated systems has been ordering in the film 110 112 Planarity di fferences between the phthalic dimethyl ester portion of 3 6a and the N,N dimethylphthalhydrazide portion of 3 7 ( t he computed geometries in Appendix B ) likely account for the slightly red shifted film absorption between the two. Given the identical planarity of 3 7 and 3 5a phthalic ring systems the further red shift for the latter suggests stronger aggregation as mediated by its constituent H bonding unit. We expect ed this to lead to better molecular stacking and therefore enh anced carrier transport. Figure 3 16 Bulk characteristics of 3 5a 3 6a and 3 7 : (a) UV Vis absorption spectr a of vacuum deposited thin films (40 nm) of 3 5a 3 6a and 3 7 and bulk samples of (b) 3 6a and (c) 3 5a exhibiting different colors (a) (b) (c) (a) (b) (c)
108 The signatures of H bonding in the bulk are identified directly from Fourier transform infrared ( FT IR ) spectra of vacuum deposited films as well as indirectly through bulk thermal behavior reported by thermal gravimetric analysis (TGA) and differential scanning calorimetry (DSC). FT IR spectral peak broadening in neat films of 3 5a especially when compared to the spectra for 3 6a and 3 7 was a preliminary indicator of H bonding in the bulk. Digging deeper, u naggregated and unfunctionalized p hthalhydrazide in its tautomeric form (Figure 3 3f) has a calculated carbonyl stretch ( C=O ) at 1701 cm 1 which shifts to 1662 cm 1 in an H bonded cobalt water complex 91 As Figure 3 17b shows, t his is in excellent agreement with the neat film IR spectrum of 3 5a which has C=O = 1659 cm 1 Further, Figure 3 17c,d displays nearly identical spectra l features of blended (3:2 weight ratio with C 60 ) 3 5a compared to neat 3 5a de monstrating the persistence of H bonding in device relevant contexts
109 Figure 3 17 Various Fourier transform infrared (FT IR) spectra for 3 5a 3 6a and 3 7 : (a) Full and (b) zoomed in FT IR spectra of thin films of neat 3 5a 3 6a and 3 7 and (c) full and (d) zoomed in FT IR spectra of thin films of neat 3 5a and 3 5a :C 60 blend (3:2 weight ratio ) Indirect evidence for H bonded aggregation comes through TGA, where the decomposition temperature for 3 5a is approximately 70 C higher than 3 6a Consistently, DSC ( Appendix B) reveal s a high melting transition (332 C) for 3 5a relative to 3 6a (118 C) as well as additional exothermic transitions (298 C, 1 .51 J/g and 198 C, 14.4 J/g) upon cooling from the melt. Although not investigated in detail yet, these transitions are suggestive of liquid crystalline behavior, characterized earlier for appropriately substituted phthalhydrazide based molecules capable of forming discotic mesophases 90 (a) (b) (c) (d) (a) (b) (c) (d)
110 X ray diffraction ( XRD ) patterns obtained from neat powdered and vacuum deposited thin film samples of 3 5a and 3 6a can be seen in Figure 3 1 8 Interestingly, while both congeners exhibit crystalline powder patter n s (Figure 3 1 8 a), albeit with different peak characteristics, only the H bond active 3 5a exhibits crystallinity when deposited onto silicon wafers (Figure 3 18 b). Manual indexing of the 3 5a powder spectrum elucidated a hexagonal two dimensional l attice structure with a = 26.3 0.7 ( Appendix B) If 3 5a is envisioned as aggregating laterally as an H bonded trimeric disc and vertically as columns, which then form a two dimensional hexagonal lattice of columns, then t his lattice parameter could be correlated to the intercolumnar distance with a column radius approximately equal to a The radius of the column could also be approximated by the end to end molecular length of 3 5a 18 as determined by low level gas phase computation ( r Fig ure 3 18c) C omparison of these two values suggests a two dimensional packing scheme represented in Figure 3 1 8 c While initially promising, software based indexing of the diffractograms could not be computed due to low intensity and broadened peaks. La st, given that this specific XRD technique measures distances normal to the substrate, the majority of intense peaks being denotes predominant ly home o tropic alignment of bulk 3 5a which could lead to better charge transport in device settings.
111 Figure 3 18 X ray diffraction ( XRD ) patterns obtained for 3 5a and 3 6a as (a) powder samples and (b) vacuum deposited thin films (approximately 2 silicon wafers ) ; (c) proposed two dimensional packing pattern of trimeric 3 5a T o more directly probe the ability of 3 5a to participate in H bond directed assembly, a series of s olution based 1 H NMR studies were performed with the soluble (a) (b) (c) (a) (b) (c)
112 hexyl derivative 3 5b A v ariable temperature 1 H NMR experiment (Figure 3 14 ) performed in a low dielectric solvent that supports H bonding ( toluene d 8 ), was particularly insightful Broadening and upfield shifting of the thiophene protons (in the 6.5 8.5 ppm region ) as the temperature is lowered are consistent with aggregation through stack ing. In contrast, 3 6 b gives a sharp 1 H NMR spectrum at room temperature in toluene d 8 ( Ap pendix B ). Simultaneous aggregation of 3 5 b by H bonding is indicated by the two b roadened peaks at = 12.78 and 13.75 ppm (at 27 C). These correspond to the N H and O H protons of the lactim lactam tautomer 3 5bt NMR timescale) and shifted downfield as a consequence of H bonded assembly. T he N H/O H peaks coalesce at approximately 85 C ( T c ) ( G c = 17 18 kcal/ mol based on the standard treatment for exchanging nuclei that are not spin spin coupled 113 ) but remain significantly deshielded, consistent with intermolecular H bonding persisting even at this temperature. These results are fully consistent with phthalhydrazide derivatives known to form stable H bonded trimeric disc like aggregate s in solution 90
113 Figure 3 19 Variable temperature proton nuclear magnetic resonance ( 1 H NMR ) spectra of a 10 mM toluene d 8 solution of 3 5 b Device Based Performance and Characterization Results With the optical and electronic properties of 3 6a 3 7 and 3 5a established in the molecular and aggregated states, final studies aimed at prob ing the consequences of phthalhydrazide self assembly on the performance of bulk BHJ OPVs were investigated through collaboration with Prof Jiangeng Xue P hotovoltaic devices were fabricated with the structure shown in Figure 3 20 a Both H congeners ( 3 5a 3 6a and 3 7 ) were vacuum evaporated in a blend with C 60 at a weight ratio of 3:2 sandwiched between an indium tin oxide/molybdenum oxide ( ITO/MoO x ) anode and a bathocuproine/aluminum ( BCP/Al ) cathode. Current voltage characteristics for the devices in the dark and under 1 sun AM1.5G simulated solar illuminat ion from a Xe arc lamp are shown in Figure 3 20 b As summarized in Table 3 2 the 3 5a :C 60 device show s as much as three fold enhancement in PCE (PCEs are adjusted to account for spectral mismatch between Xe arc lamp and the sun, which is
114 relative to the other two devices, primarily due to improved short circuit current density ( J SC ) and fill factor (FF) F itting the voltage dependence of the photocurrent to a charge extraction model provide s a measure of how far an average charge carrier can travel before it recombines 114,115 Applied to current voltage curves in Figure 3 20 b, the models revealed a short circuit charge collection length of 1 5 nm, 1 8 nm, and 4 2 nm is obtained for the 3 6a 3 7 and 3 5a devices, respectively T he se result s suggest improved charge transport in the 3 5a based device. Table 3 2 Characteristics of BHJ OPV devices containing 3 6a 3 7 and 3 5a as the donor and C 60 as the acceptor in 3:2 weight ratio Device V OC (V) J SC (mA/cm 2 ) FF (%) cm 2 ) PCE (%) Corrected PCE (%) 3 6a 0.95 0.95 30 2.5 0.27 0.35 3 7 0.99 1.28 28 4.1 0.36 0.46 3 5a 0.88 2.47 45 2.1 0.97 1.2
115 Figure 3 20 BHJ OPV device characteristics: (a) s chematic representation of BHJ OPV device architecture (where 3 XX corresponds to either 3 6a 3 7 or 3 5a ), and (b) current voltage curves for devices in the dark and under AM 1.5G illumination External quantum efficiency (EQE) under the short circuit condition, which is a measure of how efficiently photocurrent is produced at each incident wavelength, is s hown for the three devices in Figure 3 21 a The EQE spectrum of the 3 5 a based device shows a red shifted leading edge at around = 500 nm by approximately 15 nm, compared with those of the other two devices consistent with the absorption of the molecules in neat films (Figure 3 1 6 a ). This further indicates that the H bon ding induced aggregation of the 3 5a molecules is preserved in the blend with C 60 Furthermore, t he peak EQE for 3 5a :C 60 of 4 7 % at = 423 nm nearly doubles that of the control devices (23% at = 410 nm for 3 6a ; 30% at = 414 nm for 3 7 ), in strong contrast to the approximate 10% difference in the device absorptance at this wavelength ( SI for details ). The long wavelength tails ( > 550 nm ) of the EQE spectra (a) (b) (a) (b)
1 16 shown in Figure 3 1 7 a are solely due to absorption by C 60. With the same C 60 content in these devices and therefore the same light absorption efficiency, the difference in EQE at > 550 nm could be directly correlated to the charge collection efficiency. The significantly longer charge collection length in the 3 5a device dem onstrates the improved charge transport in the BHJ, attributed to the H bonding enabled molecular assembly Another interesting feature of H bonded assembly present in device relevant 3 5a :C 60 blended films was the presence of fine structure as revealed i n atomic force microscopy (AFM) tapping mode phase images shown in Figure 3 21 b. AFM phase images differ from height images (which were unremarkable for the 3 5a :C 60 films) in that they detect material contrast. The image shown in Figure 3 21 b showed a relatively uniform and smooth surface determined by the height image, yet displayed interesting threaded structure when material density was studied While bearing some visual similarity to other 1D columnar nanostructures, 9,20,24,29 the composition of the structure is inconclusive T he thickness of the threaded domains and of the spatially adjacent domains is on the order of approximately 10 nm, which is close to the ideal s ize of phase separated domains for efficient BHJ OPVs. 1
117 Figure 3 21 Electrical properties and thin film morphology of BHJ OPVs: (a) EQE spectra for BHJ OPV devices blended with C 60 and (b) AFM tapping mode phase image of 3 5a :C 60 blend showing fine structure Summary The first systematic structure property relationship study employing a rational, bottom up approach to improve the PCE of small molecule BHJ OPVs has prov ided s ignificant insight into the consequences of dedicated H bonding interactions between molecular donors on molecular assembly, absorption, charge collection, and performance in OPV devices. While the H bonding enabled 3 5a shares nearly identical opti cal and electronic properties with H bonding deficient 3 6a and 3 7 in the molecularly dissolved state, H bond promoted aggregation significantly differentiates the molecules The effects of which were borne out through red shifted absorption, broadened N H / O H stretches, and a shifted C=O stretch in thin films. Solution based study of 3 5 b in low dielectric solvent yielded further evidence of H stacked aggregation and evidence of a trimeric aggregate. These self assembly (a) (b) (a) (b)
118 features are ultimately transferred to C 60 blends where they result in a three fold increase in device PCE due to improved charge transport Although considerable work remains to characterize the assembly structure of the H bonded species in donor/acceptor blends it appears that properly installed H bonding units can have a favorable effect on the BHJ morphology. Future studies will exploit the modularity of the design to study to what extent improvements can be realized across different classes of donor chromop hores. Experimental Materials and General Methods Synthesis. Reagents and solvents were purchased from commercial sources and used without further purification unless otherwise specified. THF, ether, dichloromethane, and DMF were degassed in 20 L drums a nd passed through two sequential purification columns (activated alumina; molecular sieves for DMF) under a positive argon atmosphere. Thin layer chromatography (TLC) was performed on SiO 2 60 F254 aluminum plates with visualization by UV light or staining Flash column chromatography was performed using Purasil SiO 2 Full synthetic details for 3 5a,b 3 6a,b and 3 7 are located in Appendix C. Thin film deposition and device fabrication. Vacuum thermal evaporation was used to deposit neat or blended thin films of the three methyl terminated branched quaterthiophene derivatives, 3 5a 3 6a and 3 7 Organic photovoltaic devices were fabricated on commercial indium tin oxide (ITO) coated glass substrates with a sheet resistance of ~ 15 /square. The substrates were sequentially sonicated for 15 minutes in detergent, water, acetone and isopropanol be fore UV ozone treatment for an additional 15 minutes. Treated substrates were then passed into a vacuum chamber
119 pumped down to 10 6 Torr, and organic layers were sequentially evaporated through a shadow mask with thicknesses monitored by a quartz crystal monitor. The layer structure for all devices reported here was glass/ITO/MoO 3 (8nm)/donor:C 60 /BCP (8nm)/Al (100nm), where BCP is bathocuproine. The donor: C 60 bulk heterojunctions were approximately 40 nm thick, and were co deposited using two quartz crys tal monitors to independently control the evaporation rates of the donor and the C 60 acceptor The cross bar geometry was used to define an active area of 4 mm 2 for the organic photovoltaic cells. After deposition, devices were encapsulated with a UV cura ble epoxy layer to prevent degradation from exposure to ambient oxygen and water. Nuclear Magnetic Resonance (NMR). 300 (75) and 500 (125) MHz 1 H ( 13 C) NMR spectra were recorded on Varian Mercury 300 or Gemini 300 or Inova 500 spectrometers. Deuterated ( c hloroform d or CDCl 3 toluene d 8 and dimethylsulfoxide d 6 or DMSO d 6 ) solvents were purchased from Cambridge Isotope Laboratories, Inc. and used without further purification. Chemical shifts ( ) are given in parts per million (ppm) relative to TMS and referenced to residual protonated solvent ( CDCl 3 : H 7.26 ppm, C 77.16 ppm; DMSO d 6 : H 2.50 ppm, C 39.52 ppm). Abbreviations used are s (singlet), d (doublet), t (triplet), q (quartet), q uin (quintet), hp (heptet), b (broad), and m (multiplet). UV Vis Absorption (solution). S pectra were obtained using a Cary 100 Bio spectrophotometer and an Infrasil Spectrophotometric 1 mm cuvette (Starna Cells, Inc.) UV Vis Absorption (thin film). M easurements were carried out with a calibrated Newport 818 UV Si photodiode illuminated by a Newport Oriel Apex
120 illuminator and monochromator system chopped at 400 Hz. The signal was detected using a Stanford Research Systems SR830 DSP lock in amplifier. Mass Spectrometry. Electrospray ionization ( ESI ) atmospheric pressure chemical ionization ( APCI ) and direct analysis in real time ( DART ) time of flight ( TOF ) mass spectrometry ( MS ) spectra were recorded on an Agilent 6210 TOF spectrometer with MassHunter software Chemical ionization ( CI ) or electron ionization MS (70 eV) spectra were recorded on a Thermo Scientific DSQ MS after sample introduction via gas chromatography ( GC ) or direct injection ( DIP ) with data processing on Xcalibur software (accurate masses are calculated with the CernoBioscience MassWorks software) Elemental Analysis. Elemental analyses were performed on a Carlo Erba 1106 instrument. FT IR Spectroscopy Samples were prepared via vacuum deposition (10 6 Torr) the salt plate was placed approximately 1 cm away from the organic source position in order to maximize the solid angle of deposition captured. F ilm thicknesses were on the order of 2.5 performed on a Bruker 80v FT IR Spectrometer with OPUS v6.5 software or a Perkin Elmer Spectrum One spectrometer with Spectrum v5.0.1 software. Cyclic Voltammetry and Differential Pulse Voltamme try (CV/DPV). Measurements were performed on a Voltammetric Analyzer potentiostat/galvanostat under the control of BAS CV 50W software from Bioanalytical Systems ( single compartment three electrode cell under argon blanket with a platinum flag as the counter electrode, a silver wire reference electrode, and a 0.02 cm 2 platinum disk as the
121 working electrode ) using 3.5 mM solutions of analyte in DMF, tetra n butyl ammonium hexafluorophosphate [ ( n Bu) 4 N PF 6 ] as the electrolyte, a ferrocene standard, and a scan rate of 50 mV/s. Thermogravimetric Analysis (TGA) Measurements were performed on TA Instruments TGA Q5000 0121 (platinum pan, room temperature to 600 C, ramp rate = 20 C/min under nitrogen atmosphe re) and analyzed on Universal Analysis 2000 4.4A software. Differential Scanning Calorimetry (DSC) Scans were performed on TA Instruments DSCQ1000 0620 V9.9 (sealed aluminum pan, empty aluminum reference pan, ramp rate = 20 C/min, two heating and cool ing cycles) and analyzed on Universal Analysis 2000 4.4A software. Computational Details Starting geometries were obtained from a conformational search ( 10 000 iterations max, 0.01 convergence threshold, max 10 conformers reported, energy window = defa ult) and geometry minimization (MM3* force field) using the MacroModel v9.1 application within the Maestro v7.0.110 interface. The final ground state geometries and orbital energies were obtained from DFT calculations at the B3LYP/6 31G* level as implemen ted in Gaussian 03 116 accessed through the UF High Performance Computing Center Frequency calculations were performed at the same computational lev el, and no imaginary frequencies were found Molecular orbital plots were made using Gabedit v2.3.9 117 fro m the Gaussian output files The resultant DFT geometries were used for TD DFT calculations (TD(Singlets,NStates=6) using Gaussian 03 to determine the S 0 to S 1 electronic transition energies and oscillator strengths ( f )
122 Photovoltaic Device Performance Characterization. Devices were characterized under illumination from a 150W Xe arc lamp solar simulator with a KG1 filter, calibrated to 1 sun intensity using the AM1.5G spectrum. External quantum efficiency m easurements were carried out with a calibrated Newport 818 UV Si photodiode illuminated by a Newport Oriel Apex illuminator and monochromator system chopped at 400Hz. The signal was detected using a Stanford Research Systems SR830 DSP lock in amplifier. Synthetic Procedures 4 Iodophthalonitrile (3 2). According to a modified literature procedure, commercially available 4 aminophthalonitrile (0.502 g, 3.50 mmol) and p toluenesulfonic acid (2.00 g, 10.5 mmol) were dissolved in acetonitrile (15 mL) in a 100 mL round bottom flask with stirring. The mixture was chilled to 0 C for 15 min. A separate solution of sodium nitrite ( 0.480 g, 7.00 mmol) and potassium iodide (1.49 g, 8.74 mmol) dissolved in water (21 mL) was also made. The sodium nitrit e/potassium iodide solution was added to the reaction vessel slowly in 2 mL portions at 0 C. After 1 h, the mixture was diluted with water (50 mL), saturated NaHCO 3 until pH reached approximately 9, and Na 2 S 2 O 3 solution (1.89 g Na 2 S 2 O 3 6 mL water). The organics were extracted with ethyl acetate (150 mL), dried over Na 2 SO 4 and concentrated in vacuo. Purification via flash column chromatography (20% ethyl acetate in hexanes) afforded 3 2 in 74% yield (0.66 g): 1 H NMR ( CDCl 3 ) 8.16 (dd, J = 1.7, 0.4 Hz, 1H), 8.10 (dd, J = 8.2, 1.7 Hz, 1H), 7.50 (dd, J
123 = 8.2, 0.4 Hz, 1H); 13 C NMR (CDCl 3 The 1 H NMR data match es that in the literature. 94 1,2 Bis(octyloxy)benzene (3 3e ). According to a modified literature preparation 118 commercially available 3 3 d (4.99 g, 45.3 mmol), 1 bromooctane (6 mL), potassium carbonate (25 g, 181 mmol), potassium iodide (0.376 g 2.27 mmol), ethyl acetate (20 mL), and aliquout 334 (0.5 mL) were added to a 100 mL round bottom flask. The mixture was stirred and heated to reflux for 48 h. Then, the mixture was poured into water (500 mL), extracted with ethyl acetate (500 mL), drie d over MgSO 4 decolorized with activated charcoal, and concentrated in vacuo. Purification via flash column chromatography (100% hexanes to elute excess 1 bromooctane, then 10% ethyl acetate in hexanes) afforded 3 3 e in 67% yield (10.3 g): 1 H NMR ( CDCl 3 ) 6.89 (d, J = 1.1 Hz, 4 H), 4.00 (td, J = 6.6, 1.1 Hz 4 H), 1.82 ( quin J = 6.7 Hz, 4H), 1.56 1.20 (m, 2 2 H), 0.90 (t, J = 6.3 Hz, 6H) ; 13 C NMR ( CDCl 3 ) 149.4, 121.1, 114.2, 69.4, 32.0, 29.6, 29.5, 29.4, 26.2, 22.8, 14.2. The 1 H NMR data match es that found in the literature 118 1,2 D ibromo 4,5 bis(octyloxy)benzene (3 3 f ). According to a literature procedure 118 compound 3 3 e (10.3 g, 30.5 mmol) was dissolved in DCM (20 mL) and chilled to 0 C. Bromine (3.14 mL, 61.1 mmol) was added to the solution dropwise with stirring. Hydrogen bromide gas was cannulated into a saturated NaHCO 3 solution. After 3 h, the vessel was purged with argon gas, and the mixture was diluted with DCM (300 mL).
124 The organics wer e washed with 10% Na 2 S 2 O 3 solution (450 mL), 10% NaOH solution (175 mL), water (300 mL), and brine (150 mL). The solutions was dried over MgSO 4 and concentrated in vacuo. The product 3 3 f was afforded in 87% yield (13.1 g): 1 H NMR (CDCl 3 ) 7.05 (s, 2H), 3.93 (t, J = 6.6 Hz, 4H), 1.79 (quin J = 7.9, 7.1, 6.6 Hz, 4 H), 1.44 ( quin J = 7.1 Hz, 4H), 1.39 1.20 (m, 18 H), 0.88 (t, J = 7.1 Hz, 6H); 13 C NMR (CDCl 3 ) 149.2, 118.1, 114.8, 69.7, 31.9, 29.4, 29.4, 29.4, 29.3, 29.2, 26.1, 22.8, 14.2. The 1 H data match es that found in the literature. 118 4,5 Bis(octyloxy)phthalonitrile (3 3 c ). According to a modified version of a literature procedure 119 compound 3 3 f (1.00 g, 2.02 mmol), zinc (II) cyanide (0.286 g, 2.43 mmol), and DMF were added to a flame dried two arm 50 mL round bottom flask backfilled with argon. That solution was degassed for 30 min, then tetrakis(triphenylphosphine)palladium was added, and the mixture was heated to 120 C. After 16 h, the mixture was poured into water (400 mL) and extracted with ethyl acetate (400 mL). The organics were collected and washed with 30% NH 4 OH solution (100 mL), water (400 mL), and brine (100 mL). The solution was then dried over MgSO 4 and concentrated in vacuo. Purification via flash column chromatography (40% DCM in hexanes) afforded the product 3 3 c in 60% yield (0.463 g): 1 H NMR ( CDCl 3 ) 7.11 (d, J = 0.9 Hz, 2H), 4.04 (t, J = 6.4 Hz, 4H), 1.84 ( quin J = 6.6 Hz, 4H), 1.56 1.14 (m, 22 H), 0.87 (t, J = 6.5 Hz, 6 H). The 1 H NMR data matches that found in the literature 119
125 2 Iodozinc (II) phthalocyanine (3 4 b ). According to a modified literature procedure 104 commerically available 3 3b (0.151 g, 1.18 mmol), 3 2 (0.100 g, 0.394 mmol), and anhydrous zinc (II) acetate (0.289 g, 1.58 mmol) were added to a 25 mL round bottom flask. DMF (10 mL) was added to the mixture, and it was refluxed for 3 h with stirring du ring which the reaction changed from colorless to turquoise. The mixture was concentrated in vacuo, and after several attempts at purification, a pure isolated product could not be obtained. A turquoise solid, obtained in 63% crude yield (0.175 g), was a nalyzed via HRMS (ESI TOF): calculated 701.9750 for C 32 H 15 N 8 IZn (M) + found 701.9746. Diethyl 4 nitrophthalate (3 18 ). Compound 3 18 was prepared according to a procedure found in the literature 106 Commerically available 3 17 (0.200 g, 0.948 mmol) was dissolved in ethanol (200 proof, 15 mL) in a 25 mL round bottom flask with stirring. Sulfuric acid (18 M, 1.1 mL, 1.5 mmol) was added, and the solution was refluxed for 23 h. The mixture was diluted with ether (1 50 mL) and washed with saturated NaHCO 3 solution (100 mL). The organics were then washed with Na 2 SO 4 and concentrated in vacuo. Compound 3 18 a colorless oil, was obtained in 62% yield (0.158 g) without
126 further purification: 1 H NMR (CDCl 3 ): 8.54 (d, J = 2.3 Hz, 1H), 8.32 (dd, J = 8.4, 2.3 Hz, 1H), 7.79 (d, J = 8.4 Hz, 1H), 4.37 ( d q J = 7.1, 1.6 Hz, 4 H), 1.35 (d t J = 7.2, 4.7 Hz, 6H); 13 C NMR ( CDCl 3 ) 166.2, 165.0, 148.7, 138.3, 133.0, 130.0, 125.9, 124.3, 62.5, 14.0, 14.0. The 1 H NMR data matches that found in the literature 120,121 Diethyl 4 aminophthalate (3 19 ). According to a modified literature procedure 122 Compound 3 18 (0.1575 g, 0.5892 mmol) was added to a three arm 25 mL round bottom flask an d was evacuated and backfilled with argon three times. Ethanol (200 proof, 10 mL) was added, and the solution was stirred with gentle heating. Palladium on carbon (10% w/w, 0.0312 g, 0.0295 mmol) was added in two portions: the first portion was added to the mixture, then hydrazine monohydrate (0.060 mL, 1.2 mmol) was added, then the second portion of palladium on carbon was added. After stirring for 1 h at 50 C, the mixture was filtered over Celite and concentrated in vacuo. Water (100 mL) was added t o the mixture, and a white precipitate formed, which was filtered. Compound 3 19 a white powder, was obtained in 76% yield without further purification: 1 H NMR (CDCl 3 ): 7.71 (d, J = 8.4 Hz, 1H), 6.71 (d, J = 2.4 Hz, 1H), 6.66 (dd, J = 8.4, 2.4 Hz, 1H), 4.31 (dq, J = 19.0, 7.1 Hz, 4H), 4.14 (s, 2H), 1.34 (q, J = 7.1 Hz, 6H); 13 C NMR (CDCl 3 ): 169.4, 166.4, 150.0, 136.7, 132.0, 118.7, 115.0, 113.5, 61.7, 6 1.0, 14.3, 14.2. The 1 H data matches that found in the literature 121
127 Diethyl 4 iodophthalate (3 2 0) According to a modified literature procedure 123 compound 3 19 (0.346 g, 1.46 mmol) was dissolved in acetonitrile (10 mL) in a 25 mL round bottom flask. p Toluenesulfonic acid (0.842 g, 4.38 mmol) was dissolved in the mixture, which was then chilled to 0 C for 15 min. A separate solution of sodium nitrite (0.208 g, 2.92 mmol) and potassium iodide (0.606 g, 3.65 mmol) was made with water (10 mL) and was added dropwise to the mixture of 3 19 and p toluenesulfonic acid. The solution turned brown, and after 2.5 h of stirring at 0 C, water (50 mL) was added to the mixture. Saturated NaHCO 3 was added until the solution reached ph of approximately 9. At that point, a Na 2 S 2 O 3 solution (1.91 g in 6 mL of water) was added and the solution turned white. The organics were extracted with ethyl acetate (125 mL), dried over Na 2 SO 4 and concentrated in vacuo. Purification via flash column chromatography (10% ethyl acetate in hexanes ) afforded 3 20 in 60% yield (0.307 g): 1 H NMR (CDCl 3 ): 8.02 (d, J = 1.8 Hz, 1H), 7.86 (dd, J = 8.2, 1.8 Hz, 1H), 7.46 (d, J = 8.1 Hz, 1H), 4.35 ( d q J = 7.1, 3.9 Hz, 4 H), 1.36 ( dt J = 7.1, 3.5 Hz, 6H); 13 C NMR (CDCl 3 ): 166.9, 166.3, 140.0, 137.7, 134.2, 131.3 130.5, 97.5, 62.1, 62.0, 14. 2 ; HRMS (DART TOF) calculated 348.9931 for C 12 H 14 IO 4 (M+H) + found 348.9936.
128 6 B romo 2,3 dihydrophthalazine 1,4 dione (3 23) Commercially available 3 22 (1.0244 g, 4.532 mmol) and glacial acetic acid (5 mL) were added to a 25 mL round bottom flask with stirring and refluxed for 1 h. After removal from heat and cooling to rt, hydrazine monohydrate (0.2344 mL, 4.780 mmol) was added and a precipitate immedi ately formed. The mixture was refluxed for 30 min, then dissolved in 5% NaOH solution (110 mL). The mixture was then acidified to neutral pH as determined by Litmus paper with glacial acetic acid causing the product to crash out. The product was filtere d and washed with water and methanol. The result was 3 23 a white powder, in 46% yield (0.500 g): 1 H NMR ( DMSO d 6 ) 11.66 (s, 2 H), 8.15 (s, 1H), 8.00 (q, J = 8.5 Hz, 2H); 13 C NMR ( DMSO d 6 ) 154.4, 153.6, 135.7, 128.8, 127.7, 127.6, 126.5, 126.2 The 1 H and 13 C NMR data match that found in the literature 107 6 B romo 2,3 bis(4 methoxybenzyl) 2,3 di hydrophthalazine 1,4 dione (3 24 a) and 6 bromo 1,4 bis((4 methoxybenzyl)oxy)phthalazine (3 2 4 b). Compound 3 23 (0.257 g, 1.066 mmol) was added to a 25 mL round bottom flask, which was subsequently evacuated and backfilled with argon three times. DMF (10 mL) was added and the solution was stirred at 0 C. Sodium hydride (60% in mineral oil, 0.090 g, 2.3 mmol) was added to the mixture, and it was stirred for an additional 30 min at 0 C. Then, the mixture was warmed to rt over 30 min, then heated at approximately 60 C for 1 h. after which p methoxybenzyl chloride (PMB Cl, 0.303 mL, 2.23 mmol) was added. After he ating overnight (18 h) at 60 C, the mixture was poured into water (100 mL) and
129 extracted with ethyl acetate (150 mL). The organics were dried over MgSO 4 and concentrated in vacuo. Purification via flash column chromatography (10% ethyl acetate in hexan es) afforded a mixture pure 3 2 4 a and 3 24 b a colorless oil, in 73% yield (0.373 g): 3 24 a 1 H NMR (CDCl 3 ): 8.54 (dd, J = 1.6, 0.9 Hz, 1H), 7.83 7.79 (m, 2H), 7.41 (d, J = 8.8 Hz, 2H), 7.36 (d, J = 8.8 Hz, 2H), 6.90 (d, J = 8.8 Hz, 2H), 6.85 (d, J = 8.8 Hz, 2H), 5.26 (s, 2H), 5.22 (s, 2H), 3.83 (s, 3H), 3.78 (s, 3H) ; 3 24b 1 H NMR (CDCl 3 ): 8.24 (d, J = 8.5 Hz, 1H), 8.08 (d, J = 1.9 Hz, 1H), 7.83 (ddd, J = 8.5, 1.9, 0.8 Hz, 1H), 7.40 (d, J = 8.4 Hz, 2H), 7.37 (d, J = 8.4 Hz, 2H), 6.91 (d, J = 8.3 H z, 2H), 6.85 (d, J = 8.3 Hz, 2H), 5.26 (s, 2H), 5.21 (s, 2H), 3.83 (s, 3H), 3.79 (s, 3H); 3 24 a and 3 24 b 13 C NMR (CDCl 3 135.1, 130.5, 130.29, 130.26, 130.2, 130.14, 130.09, 130.04, 130.02, 129.4, 129.1, 129.0 128.2, 128.1, 127.9, 127.8, 126.7, 126.2, 126.0, 125.3, 123.3, 114.0, 113.92, 113.88, 113.83, 68.6, 68.5, 55.23, 55.18. The 1 H and 13 C NMR data match that found in the literature 107 D imethyl 4 bromophthalate (3 26 ). Commercially available 3 22 (2.44 g, 8.872 mmol) and methanol (20 mL) were added to a 25 mL round bottom flask with stirring. Gentle heating was applied to dissolve 3 22 Then, sulfuric acid (18 M, 0.945 mL, 17.74 mmol) was adde d, and the solution was refluxed overnight (approximately 18 h). The mixture was poured into saturated NaHCO 3 solution (100 mL), and the organics were extracted with ethyl acetate (300 mL). The extract was washed with more saturated NaHCO 3 solution (100 mL), dried over MgSO 4 and concentrated in vacuo. The product was 3 26 a colorless oil, in 94% yield (2.28 g): 1 H NMR (CDCl 3 7.8 4 (d, J = 1.7 Hz, 1H),
130 7.6 9 (dd, J = 8.2, 1.8 Hz, 1H), 7.63 (d, J = 8.3 Hz, 1H), 3.91 (d, J = 3.8 Hz, 6H) ppm ; 13 C NMR (CDCl 3 ): 167.5, 167.3, 134. 6 134.5, 132. 4 131.1, 130.8, 126.3, 53. 5 53.3 ppm. The 1 H and 13 C NMR data match that found in the literature 124 2,3 B is(4 methoxybenzyl) 6 ((trimethylsilyl)ethynyl) 2,3 dihydrophth a lazine 1,4 dione (3 28 ). Compound 3 24 a (0.5065 g, 1.053 mmol), triethylamine (0.732 mL, 5.265 mmol), copper (I) iodide (0.040 g, 0.21 mmol), trime thylsilylacetylene (0.16 mL, 1.2 mmol), and THF (15 mL) were added to a flame dried three arm 100 mL round bottom flask backfilled with argon and degassed with argon for 45 min. Then, b is(triphenylphosphine)palladium (II) dichloride (0.0739, 0.105 mmol) w as added to the mixture, which turned from slightly turbid to dark reddish brown in approximately 1 min. The reaction was monitored via thin layer chromatography (25% ethyl acetate in hexanes), and after heating at 50 C overnight (18 h), more copper (I) iodide (0.040 g, 0.21 mmol), trimethylsilylacetylene (0.16 mL, 1.2 mmol), and b is(triphenylphosphine)palladium (II) dichloride (0.0739, 0.105 mmol) were added. After 1 h, the solution was poured into water (150 mL) and extracted with ethyl acetate (200 mL ). The organics we re washed over MgSO 4 filtered over Celite and concentrated in vacuo. Purification via flash column chromatography afforded 3 2 8 a light yellow powder, in 59% yield (0.310 g): 1 H NMR (CDCl 3 ): 8.47 (d, J = 1.1 Hz, 1H), 7.86 (dd, J = 8.3, 0.5 Hz, 1H), 7.74 (dd, J = 8.3, 1.6 Hz, 1H), 7.39 (dd, J = 14.3, 8.7 Hz, 4H), 6.88 (dd, J = 14.7, 8.7 Hz, 4H), 5.26 (s, 2H), 5.23 (s, 2H), 3.83 (s, 3H), 3.78 (s, 3H), 0.27 (s,
131 7H) ; 13 C NMR (CDCl 3 159.7, 159.2, 149.3, 135.4, 130.9, 1304, 130.2, 129 .4, 129.2, 128.5, 127.0, 124.0, 123.6, 114.0, 113.9, 98.5, 55.3, 53.5, 0.13 ; HRMS ( DART TOF) calculated 499.2048 for C 29 H 31 N 2 O 4 Si (M+H) + found 499 .2054. 6 E thynyl 2,3 bis(4 methoxybenzyl) 2,3 dihydrophthalazine 1,4 dione Compound 3 2 8 (0.0613 g, 0.123 mmol) was dissolved in DCM (5 mL) in a small vial. To that solution, tri n butylammonium fluoride (1.0 M in THF, 0.25 mL, 0.25 mmol) was added, and the mixture was swi rled. After 5 min, thin layer chromatography (30% ethyl acetate in hexanes) showed complete conversion of the starting material, so the mixture was washed with water (5 mL), and the mixture was dried under high vacuum. Purification via flash column chromatography (30% ethyl acetate in hexanes) afforded 6 ethynyl 2,3 bis(4 methoxybenzyl) 2,3 dihydrophthalazine 1,4 dione in greater than 50% yield (greater than 0.026 g): 1 H NMR ( CDCl 3 ) 8.50 (d, J = 1.4 Hz, 1H), 7.90 (d, J = 8.3 Hz, 1H), 7.78 (dd, J = 8.3, 1.7 Hz, 1H), 7.41 (d, J = 8.7 Hz, 2H), 7.37 (d, J = 8.6 Hz, 2H), 6.90 (d, J = 8.7 Hz, 2H), 6.85 (d, J = 8.7 Hz, 2H), 5.27 (s, 2H), 5.22 (s, 2H), 3.83 (s, 3H), 3.79 (s, 3H ). 5' (5 B ro mothiophen 2 yl) 5,5'' dihexyl 2,2':3',2'' terthiophene (3 29b) DMF (approximately 10 mL; alternately, CHCl 3 was also used) was added to a 50 mL round bottom flask and chilled to 0 C. Compound 3 14b (0.891 g, 1.786 mmol) was added to
132 the flask with stirring, then N bromosuccinimide (NBS, 0.3115 g, 1.750 mmol) added. The mixture was covered with aluminum foil and left to warm to rt for 72 h. The mixture was poured into water (250 mL) and extracted w ith DCM (350 mL), dried over Na 2 SO 4 and concentrated in vacuo. Purification via flash column chromatography (100% hexanes) afforded 3 29b in 59% yield (0.607 g): 1 H NMR (CDCl 3 ): 7.15 (s, 1H), 7.00 (d, J = 2.1 Hz, 1H), 7.00 (d, J = 1.8 Hz, 1H), 6.95 (d, J = 1.1 Hz, 1H), 6.94 (d, J = 0.6 Hz, 1H), 6.75 6.70 (m, 2 H), 2.83 (q, J = 8.3 Hz, 4 H), 1.72 (dq, J = 15.0, 7.1 Hz, 4H), 1.51 1.31 (m, 12 H), 0.96 (t, J = 6.4 Hz, 6 H) ; HRMS ( DART TOF) calculated 579.0702 for C 28 H 34 BrS 4 (M+H) + found 579.0709 T rimethyl((5'' methyl 5' (5 methylthiophen 2 yl) [2,2':4',2'' terthiophen] 5 yl)ethynyl)silane (3 29). Compound 3 29a (0.310 g, 0.709 mmol), triethylamine (0.50 mL, 3.5 mmol), and THF (10 mL) were added to a flame dried, two arm round bottom flask backfilled with argon. This mixture was degassed for 30 min, then copper (I) iodide (0.0269 g, 0.142 mmol) b is(triphenylphosphine)palladium (II) dichloride (0.0 497 0. 0709 mmol) and trimethylsilylacetylene (0.20 mL, 1.4 mmol) were added in that order. After 15 min of stirring, more trimethylsilylacetylene (0.20 mL, 1.4 mmol) was added. After 18 h, the entire reaction mixture was concentrated in vacuo. Purification via flash c olumn chromatography (100% hexanes) afforded 3 30 in 78% yield (0.253 g) as a yellow oil: 1 H NMR (CDCl 3 ): 7.15 (s, 1H), 7.13 (d, J = 3.8 Hz, 1H), 7.00 (d, J = 3.8 Hz, 1H), 6.94 (d, J = 3.5 Hz, 1H), 6.89 (d, J = 3.5 Hz, 1H), 6.71 6.63 (m, 4H), 2.49 (d, J =
133 0.8 Hz, 3H), 2.47 (d, J = 0.8 Hz, 3H), 0.27 (s, 8H) ; 13 C NMR (CDCl 3 ): 141.8, 140.5, 138.2, 134.6, 134.4, 133.6, 132.4, 132.2, 131.3, 127.9, 126.9, 126.8, 125.6, 125.5, 123.5 122.2, 100.4, 97.5, 15.6, 0.0 D imethyl 4 ((trimethylsilyl)ethynyl)phthalate (3 30). Compound 3 26 (0.6033 g, 2.210 mmol), triethylamine (1.5 mL, 11 mmol), trimethylsilylacetylene (0.63 mL, 4.4 mmol), and THF (10 mL) were added to a flame dried three arm 25 mL round bottom flask backfilled with argon. That solution was degassed for 30 min, then b is(triphenylphosphine)palladium (II) dichloride (0.155 g, 0.221 mmol) and copper (I) iodide (0.084 g, 0.44 mmol) were added. After 15 min and 2 h, additional aliquots of trimethylsilylacetylene (0.63 mL, 4.4 mmol each) were added. After 4 h, the entire reaction mixture was concentrated in vacuo. Purification via fl ash column chromatography (40% DCM in hexanes) afforded 3 3 1 in 85% yield (0.545 g): 1 H NMR (CDCl 3 ): 7.77 (dd, J = 1.6, 0.5 Hz, 1H), 7.67 (dd, J = 8.0, 0.5 Hz, 1H), 7.57 (dd, J = 8.0, 1.6 Hz, 1H), 3.89 (s, 3H), 3.88 (s, 3H), 0.24 (s, 9H) ; 13 C NMR (CDCl 3 ) 167.5, 167.4, 134.1, 132.5, 132.2, 131.0, 129.1, 126.6, 102.9, 98.4, 52.9, 52.8, 0.13 ; HRMS (ESI TOF) calculated 291.1045 for C 15 H 19 O 4 Si (M+H) + found 2 91.1047.
134 D imethyl 4 ((5'' methyl 5' (5 methylthiophen 2 yl) [2,2':4',2'' terthiophen] 5 yl)ethynyl)phthalate ( 3 3 2a ). Compound 3 3 1 (0.386 g, 1.33 mmol) was dissolved in DCM (5 mL) in a small vial. To that solution, tri n butylammonium fluoride (1.0 M in THF, 1.5 mL, 1.5 mmol) was added, and the mixture was swirled. After 5 min, the mixture was washed with water (5 mL), and the mixture was dried under high vacuum. Deprotected 3 3 1 (added as a THF solution), triethylamine (1.85 mL, 13.3 mmol), 3 29a (R = Me, 0.611 g, 1.395 mmol), and THF (10 mL) were added to a flame dried three arm 25 mL round bottom flask backfilled with argon. After 45 min of degassing with argon, b is(triphenylphosphine)palladium (II) dichloride (0.0933 g, 0.133 mmol) and copper (I) iodide (0.0505 g, 0.266 mmol) were added. The mixture was stirred for 18 h at rt. Purification via flash column chromatography (50% DCM in hexanes) afforded 3 32a in 35 % yield (0. 270 g): 1 H NMR (CDCl 3 ): 1 H NMR (300 MHz, Chloroform d (d, J = 1.6 Hz, 1H), 7.75 (d, J = 8.0 Hz, 1H), 7.63 (dd, J = 8.2, 1.5 Hz, 1H), 7.22 (d, J = 3.6 Hz, 1H), 7.18 (s, 1H), 7.07 (d, J = 3.9 Hz, 1H), 6.95 (d, J = 3.5 Hz, 1H), 6.90 (d, J = 3.5 Hz, 1H), 6.67 (dq, J = 3.2, 1.1 H z, 2H), 3.93 (s, 3H), 3.92 (s, 3 H), 2.49 ( d, J = 1.0 Hz, 3H), 2.47 (d, J = 0.9 Hz, 3H ). D imethyl 4 ((5'' hexyl 5' (5 hexylthiophen 2 yl) [2,2':4',2'' terthiophen] 5 yl)ethynyl )phthalate (3 32b ). Compound 3 3 1 (0.133 g, 0.458 mmol) was dissolved in DCM (5 mL) in a small vial. To that solution, tri n butylammonium fluoride (1.0 M in THF, 0.92 mL, 0.92 mmol) was added, and the mixture was swirled. After 5 min, the
135 mixture was washed with water (5 mL), and the mixture was dried under high vacuum. Deprotected 3 3 1 (added as a THF solution) triethylamine ( 0.30 mL, 2.3 mmol) 3 29b (R = Hex, 0.286 g, 0.504 mmol), and THF (10 mL) were added to a flame dried three arm 25 mL round bottom flask backfilled with argon. After 30 min of degassing with argon, b is(triphenylphosphine)palladium (II) dichloride (0.0161 g, 0.0229 mmol) and copper (I) iodide (0.0087 g, 0.046 mmol) were added. The mixture was stirred and heated at 50 C for 18 h. Thin layer chromatography (5% ethyl acetate in hexanes) still showed deprotected 3 31 was present, so additional aliquouts of b is(triphenylphosphine)palladium (II) dichloride (0.0161 g, 0.0229 mmol) and copper (I) iodide (0.0087 g, 0.046 mmol) were added. After 2 h, the entire mixture was concentrated in vacuo. Purification via flash column chromatography (50% DCM in hexanes) afforded 3 32b in 28% yield (0.091 g) : 1 H NMR (CDCl 3 ): 7.83 (d, J = 1.7 Hz, 1H), 7.64 (dd, J = 8.1, 1.7 Hz, 1H), 7.2 2 (d, J = 3.8 Hz, 1H), 7.19 (s, 1H), 7.08 (d, J = 3.8 Hz, 1H), 6.96 (d, J = 3.6 Hz, 1H), 6.90 (d, J = 3.5 Hz, 1H), 6.68 (td, J = 2.2, 1.1 Hz, 1H), 3.93 (s, 2H), 3.92 (s, 3H), 2.78 (q, J = 7.1 Hz, 4 H), 1.75 1.59 (m, 4H), 1.45 1.22 (m, 12H), 0.89 (td, J = 7.4, 6.0, 2.3 Hz, 6H) ; 13 C NMR (CDCl 3 ) 167.6, 167.4, 148.0, 146.8, 139.5, 134.3, 134.1, 134.0, 133.4, 132.8, 132.6, 132.0, 131.9, 131.6, 130.7, 129.4, 127.7, 127.2, 126.7, 126.5, 124.4, 124.3, 123.8, 121.1, 99.9, 92.7, 86.3, 77.4, 77.2, 76.9, 63.7, 53. 6, 53.0, 52.9, 31.7, 31.7, 31.7, 31.7, 31.6, 30.3, 30.3, 28.9, 28. 9, 25.4, 22.7, 22.7, 14.2, 14.2
136 D imethyl 4 (1 (5'' methyl 5' (5 methylthiophen 2 yl) [2,2':4',2'' terthiophen] 5 yl) 1H 1,2,3 triazo l 4 yl)phthalate (3 34a) Compound 3 3 1 (0.0387 g, 0.133 mmol) was dissolved in DCM (5 mL) in a small vial. To that solution, tri n butylammonium fluoride (1.0 M in THF, 0.14 mL, 0.14 mmol) was added, and the mixture wa s swirled. After 5 min, the mixture was washed with water (5 mL), and the mixture was dried under high vacuum. Deprotected 3 3 1 (added as an ethanol solution), compound 3 29a ( 0.0573 g, 0.131 mmol), sodium azide (0.0105 g, 0.162 mmol), dimethylethyl enediamine (1 drop, approximately 0.05 mL, 0. 5 mmol), sodium ascorbate (0.0022 g, 0.011 mmol), and copper (I) iodide (0.0030 g, 0.016 mmol) were added to a 7:3 ethanol/water mixture (4 mL) with stirring and heated to reflux. The reaction was monitored thr ough thin layer chromatography (25% ethyl acetate in hexanes), and after 4 h, additional aliquots of copper (I) iodide (0.0022 g, 0.012 mmol) and dimethylethylenediamine (1 drop, approximately 0.05 mL, 0.5 mmol) were added. After 48 h at reflux, the reaction mixture was poured into water ( 150 mL) and extracted with DCM (200 mL). The organics were washed with aqueous ammonium hydroxide (30%, 50 mL), water (200 mL), and brine (100 mL), dried over Na 2 SO 4 and concentrated in vacuo. Purification via flash column chromatography (50% to 75% DCM in hexanes to 100% DCM) afforded 3 3 4a in 25% yield (0.0204 g): 1 H NMR (CDCl 3 8.21 (s, 1H), 8.18 (d, J = 1.8 Hz, 1H), 8.10 (dd, J = 8.0, 1.8 Hz, 1H), 7.86 ( d, J = 8.1 Hz, 1H), 7.20 (d, J = 4.0 Hz, 1H), 7.20 (s, 1H), 7.09 (d,
137 J = 4.0 Hz, 1H), 6.96 (d, J = 3.5 Hz, 1H), 6.90 (d, J = 3.5 Hz, 1H), 6.68 (dd, J = 3.5, 1.1 Hz, 3H), 3.95 (s, 3H), 3.93 (s, 3H), 2.49 (d, J = 0.8 Hz, 3H), 2.47 (d, J = 0.7 Hz, 3H ). D imethyl 4 (1 (5'' hexyl 5' (5 hexylthiophen 2 yl) [2,2':4',2'' terthiophen] 5 yl) 1H 1, 2,3 triazol 4 yl)phthalate (3 34b ). Compound 3 3 1 (0. 180 g, 0 618 mmol) was dissolved in DCM (5 mL) in a small vial. To that solution, tri n butylammonium fluoride (1.0 M in THF, 0. 70 mL, 0. 70 mmol) was added, and the mixture was swirled. After 5 min, the mixture was washed with water (5 mL), and the mixture was d ried under high vacuum. S odium azide (0.0 246 g, 0. 378 mmol), s odium ascorbate (0.00 37 g, 0.01 9 mmol), and copper (I) iodide (0.003 6 g, 0.01 9 mmol) were added to a two arm round bottom flask and evacuated and backfilled with argon three times Then compou nd 3 29b (0.109 g, 0.189 mmol, added as a DMF solution) and dimethylethylenediamine ( 2 drop s approximately 0. 1 mL, 1.0 mmol) were added and stirred at 8 0 C for 20 min. Then, deprotected 3 3 1 (added as an DMF solution) was added, and the mixture was stirred and heated at 8 0 C for 30 min. The reaction was monitored through thin layer chromatography ( 80 % DCM in hexanes), and additional aliquots of sodium ascorbate (0.0041 g, 0.021 mmol) and dimethylet hylenediamine (2 drops, approximately 0.1 mL, 1.0 mmol) were added. After 3 h, an additional aliquot of copper (I) iodide (0.00 90 g, 0.0 47 mmol) was added. After 60 h at 80 C the reaction mixture was poured into water ( 200 mL) and extracted with ethyl acetate (200 mL). The organics were washed
138 with aq ueous ammonium hydroxide (30%, 10 0 mL), water (200 mL), and brine (100 mL), dried over Mg SO 4 and concentrated in vacuo. Purification via flash column chromatography (50% to 75% DCM in hexanes) afforded 3 3 4b in 43 % yield (0.0 611 g): 1 H NMR (CDCl 3 ): 8.21 (s, 1H), 8.18 (d, J = 1.7 Hz, 1H), 8.10 (dd, J = 8.1, 1.7 Hz, 1H), 7.86 (d, J = 8.1 Hz, 1H), 7.20 (d, J = 4.2 Hz, 2H), 7.08 (d, J = 4.0 Hz, 1H), 6.97 (d, J = 3.5 Hz, 1H), 6.91 (d, J = 3.5 Hz, 1H), 6.68 (d, J = 3.5 Hz, 2H), 3.95 (s, 3H), 3.93 (s, 2H), 2.95 2.59 (m, 4H), 1.83 1.50 (m, 4H), 1.48 1.14 (m, 6H), 0.89 (t, J = 6.7 Hz, 6H )
139 CHAPTER 4 SIZE AND SHAPE APPROXIMATION OF SUPRAMOLECULAR ASSEMBLIES THROUGH VARIABLE TEMPERATURE DIFFUSION ORDERED SPECTROSCOPY (DOSY NMR) Background A significant challenge for all supramolecular chemists is the definitive characterization of the constituency and resulting structure of aggregates Determining a ssociation equilibria which can be influenced by temperature, solvent, pH, etc., has become quite commonplace and can provide valuable information about supramolecu lar assembly However, measurements involving aggregates beyond dimeric assembly such as 4 1 a become increasingly complicated especially when tautomeric equilibria (Figure 4 1) must be considered To clarify, Chapter 3 largely portrays the aggregation of 4 1 a into hydrogen bonded ( H bonded ) trimeric discs, but ribbon like aggregates (Figure 4 1) can also be envisioned wherein the resulting assembly would resemble a supramolecular polymer. To expand upon the previous spectroscopic work of 4 1a and to f urther understand the supramolecular assembly of the phthalhydrazide branched quaterthiophene system, we turned to diffusion measurements to provide this insight.
140 Figure 4 1 Chemical structures of 4 1a,b r ibbon like aggregate of 4 1 a and four different tautomeric forms of 4 1 a all capable of H bonded self assembly T he Modified Stokes Einstein Equation Albert Einstein is credited with many of the equations governing translational molecular diffusion. To begin, diffusion ( D ) is related to molecular size through the Einstein Smoluchowski equation: 125 ( 4 1) where D is the diffusion coefficient, k B is th e Boltzmann constant, T is the temperature, and f is the hydrodynamic frictional coefficient. Essentially, diffusion in a solvent is inversely related to the frictional forces felt on the diffusing entity. For a hard (i.e.,
141 solvent cannot penetrate) sphe rical molecule, frictional forces on the diffusing entity are approximated by ( 4 2) where r h is the hydrodynamic radius and is the viscosity of the solvent. The substitution of E quation 4 2 into E quation 4 1 give s the well known Stokes Einstein equation: ( 4 3) This equation has been used in numerous instances to derive hydrodynamic radii from several different diffusion measurement techniques and can be very useful for determining molecular aggregation. 125 128 However, the limitations of this equation lie in its assumptions E quation 4 3 describes the diffusion of a system whose shape is approximated by a sphere and whose size is much larger than the solvent. Such an equation cannot accurat ely describe the diffusion of a much smaller, non spherical system like 4 1a and its self assembled aggregates. Therefore, the expression for f was modified to ( 4 4) where c is a size correlation factor between r h of the diffusing species and the van der Waals radius ( r vdW ) and f s is a system derived shape friction correction factor pioneered by Perrin. 127 In essence, c corrects for the difference in friction felt on the diffusing entity as it relates to the size of the solvent. In other words, a molecule with r vdW equal to the size of the solvent (typically < 5 ) has less surface area available for friction than a
142 molecule with r vdW = 100 that relationship is expressed through c which is defined as ( 4 5) where r h is the hydrodynamic radius of the diffusing species and r vdW is the van der Waals radius of the solvent. Frictional di fferences also manifes shape Perrin envisioned diffusing entiti es as similar to prolate (cigar shaped) or oblate (pancake shaped) spheroids and derived formulas for the correction factor parameterized by the polar ( a ) and equitorial ( b ) axes of the sphe roid (see Figure 4 2) 127 Figure 4 2 Shapes of diffusing entities as portrayed by (a) spheres as in the Stokes Einstein equation and (b) prolate or (c) oblate spheroid s as in the modified Stokes Einstein equation Specifically, the friction correction factor, f s is defined as (prolate, 4 6 )
143 (oblate, 4 7 ) where a and b are defined as previously mentioned 127 Substituting E quation 4 4 into E quation 4 1 gives the modified Stokes Einstein equation: ( 4 8 ) which is essentially a function of r h a and b An important consequence of using these correctional factors is that r h no longer represents a physically measureable distance. Instead, it represents the hydrodynamic radius of a sphere with equivalent volume to that of the spheroid used to calculate f s 127 Using DOSY NMR to Characterize Supramolecular Assemblies A pulsed gradient spin echo (PGSE) pulse sequence and a fitting algorithm are both necessary to generate a diffusion coefficient from nuclear magnetic resonance (NMR) spectroscopy. Briefly, a from a gradient pulse. After a short period, another gradient of the same strength is applied, and the resulting change in magnetization is measured and correlated to the diffusion of the nucleus ( more technical explanations can be found in the literature 125 127 ). Arrayed spectra are produced by varying the gradient strength, from which a plot of intensity versus gradient strength is generated. Different pulse sequences and data collection software utilize different fitting equations, but the end result yields the diffusion coefficient of that specific nucleus. Ideally, all nuclei within a diffusing entity should have nearly identical diffusion coefficients.
144 DOSY experiments hav e two primary applications: characterization of supramolecular complexes and mixtures. 129 While the latter can be helpful in characterizing the outcome of a reaction, it is the former that we hoped to exploit. Carefully designed DOSY NMR experiments have previously been employed for molecularity, size, shape and K a sc determination of supramolecular complexes. 130 140 Analysis of diffusion coefficients gained from DOSY experiments ranges from trends 138 to molecular weight approximations 132 134,139,140 to size and shape modeling 130 Typically unseen in the literature are variable temperature DOSY studies due to the possibility of convection currents. Pulse sequences with convection correction, while sacrificing signal to noise, have been developed to allow variable temperature DOSY NMR spectra to be explored. 141,142 Detailed in this chapter is a novel protocol developed using the modified Stokes Einstein equation (equation 4 8 ) and variable temperature diffusion ordered nuclear magnetic resonance spec troscopy (DOSY NMR) to determine the size, shape, and molecularity of self assembled 4 1a aggregates. 127 Previous studies on phthalhydrazide derivatives similar to 4 1a performed in H bond promoting solvents such as toluene have shown evidence of cyclic, disc like trimer formation as determined by comparison of high performance liquid chromatography ( HPLC ) retention times against a covalently bound analog. 90 Our goal was to directly probe the expected columnar assemblies of 4 1a in toluene by measuring diffusion as a function temperature. We hypothesized that the diffusion behavior of the aggregates would mimic the aggregation behavior seen in variable temperature NMR experiments (see Chapter 3 ) ;
145 stacked aggregates o f H bonded trimers of 4 1a would be predominant up to a temperature of approximately 65 C whose diffusion would be measured. At higher stacked aggregates would break up, leaving the trimeric discs as the majority species whose diffu sion would also be measured ( Figure 4 2) Subsequent fitting of the diffusion versus temperature curve to the modified Stokes Einstein equation would yield physical parameters of the aggregates present and provide evidence for the existence of a discrete trimeric aggregate Figure 4 3 Computationally derived model (AMBER*) of discrete trimeric aggregate of 4 1a showing the approximate molecular dimensions of a supramolecular monomeric unit
146 protocol in practice. First, knowing that 4 1a and 4 1b shared similar size and shape characteristics when 4 1a is in its monomeric form, measuring diffusion versus temperature i n an H bond suppressive solvent (e.g., dimethylsulfoxide or DMSO) for both compounds should yield the same size and shape parameters. Further, since 1 H NMR previously showed 4 1b to remain unaggregated in toluene, size and shape parameters derived from di ffusion measurements should mirror those seen in DMSO for both 4 1a and 4 1b The combination of these control experiments and the fitting routine should elucidate the size and shape of aggregates of 4 1a in toluene and disassembled monomers in DMSO. Expe rimental Equipment, Materials, and Sample Preparation Spectra were collected on a INOVA 500 MHz spectrometer equipped with a 5 mm Varian indirect detection, triple resonance ( 1 H 19 F)/ 13 C/( 31 P 15 N) PFG probe and Vnmrj v3.51 software using a gradient compen pulse sequence with convection correction. Temperature was varied using a FTS Systems TC 84 temperature controller. Deuterated (toluene d 8 and DMSO d 6 ) solvents were purchased from Cambridge Isotope Laboratories, Inc and used without further purification. Temperature based viscosities were derived from fitted curves of reported viscosity versus temperature data 143,144 Solvent radii, r vdW ( toluene d 8 : 2.81 DMSO d 6 : 2.54 ) were calculated by creating a low level model (MM2 minimization in Chem 3D Pro), computing a van der Waals volume in Accelrys Discovery Studio v3.5.0, and back calculating out the radius of a sphere of equivalent volume. 20 mM and 27 mM solutions of 4 1a samples were prep ared by dissolving 13.2 mg in 1.0 mL of toluene d 8
147 and 13.2 mg in 0.75 mL of DMSO d 6 respectively. 20 mM and 2 4 mM solutions of 4 1b were prepared by dissolving 13.8 mg in 1.0 mL of toluene d 8 and 13.8 mg in 0.75 mL of DMSO d 6 respectively (synthetic details for 4 1a,b can be found in Appendix C) An aliquot of each was placed into a Wilmad NMR tube and sealed with paraffin. When not in use, the NMR tubes containing the samples were stored in a dessicator. Data Collection Procedure First a normal 1 H NMR spectrum was taken in order to determine the correct parameters (acquisition time, acquisition delay, spectral width, etc.) for obtaining an accurate one dimensional ( 1D ) spectrum at room temperature. Then, those spectral parameters were transferred to an empty experimental window. In that same window, the DOSY pulse sequence (DgcsteSL_cc) was loaded so that the optimal parameters for collecting 1D spectra were adopted for the DOSY sequence. Upon the loading of the sequence, the lower and upper limits of gradient strength were set at 1000 and 31000 respectively. Both diffusion gradient length (default = 2.0 ms) and diffusion delay (default = 100.0 ms) were adjusted such that the peaks heights in final spectrum in the array were ~10 30% of the peak heights in the first spectrum. Either seven (7) or fifteen (15) spectra were used to determine the diffusion coefficients of the samples if seven were used, a software calculation correction was used. Once all spectra were collected, they were batch processed and baseline corrected. Then, diffusion coefficients were calculated based on peak integrals of the arrayed decay spectra Where possible, the solvent diffusion coefficient was also included as a point of comparison. This data was c ollected as a two dimensional DOSY ( 2D DOSY ) spectrum where chemical shift (x axis) was plotted against diffusion coefficient (y axis), and the temperature was recorded. An important consequence of using the peak
148 integrals as the fitting parameter (and us ing DOSY NMR in general) is that while each chemical shift peak should have an identical diffusion coefficient, the reality is measured diffusion coefficients vary slightly from peak to peak. Therefore, the overall diffusion coefficient of the compound wa s selected from the chemical shift with the smallest calculated error from decay fitting. After the necessary data was compiled, the temperature of the instrument was raised by using the temperature controller in increments of roughly 10 C The next spectra would be collected at least an hour later to minimize convection. At the new temperature, the diffusion parameters were changed to again match the previously stated requirements, and the spectra were collected using the same methodology. Results and Discussion Data Analysis and Example Fitting Procedure To begin, 4 1b was dissolved in DMSO d 6 (27 mM) and six diffusion measurements were recorded (21.9, 29.9, 40.8, 53.3, 64.4, and 76.3 C, Figure 4 2a ). If equation 4 8 is organized such that each temperature measurement is divided by the corresponding viscosity of DMSO according to ( 4 9) the resulting line will have a slope inversely related to c f s and r h (Figure 4 2b) whi ch was 1.2 3 x 10 15 m 3 /kg As previously stated, the correctional factors c and f s are derived from r h and the spheroid shape parameters, a and b respectively Moreover, a and b also depend on the type of spheroid selected to model the behavior of the d iffusing entity. The prolate model will result in f s < 1 whereas the oblate model will result in f s > 1 ; essentially, diffusion behavior fitted via the prolate model cannot be fitted
149 using the oblate model, giving an initial estimate of the shape of the molecule For 4 1b in DMSO, fitting using the prolate model yielded values of a = 13.4 b = 7.4 and r h = 9.0 ultimately resulting in f s = 0.690 and c = 5.76 Figure 4 4 Diffusion data for 4 1b : (a) d iffusion versus temperature plot and (b) linearized diffusion versus temperature plot To ensure that these values accurately reflect the actual size of the diffusing entity, the values of a and b must be reconciled to match r h Since the ratio of a to b directly determines f s (i.e., the aspect ratio) proportional changes to both will not affect the fitted value. This fact allows the calculated volume of a spheroid of parameters a and b to be manipulated such that f s is not a ffected Further, it allows the calculated volume of the spheroid to be manipulated to match a spherical value derived from r h For example, if the fitting of 4 1b in DMSO resulted in a = 10.7 and b = 5.9 , then the resulting prolate spheroid volum e would be 1560 3 That volume converted to the radius of a sphere results in a value of 7.2 , which does not agree with the derived r h of 9.0 . However, if a and b are both increased by 20%, the values of a = 13.4 and b = 7.4 are obtained which yield a prolate spheroid of nearly the same volume as sphere with r = 9.0 (the derived r = 9.01 ). (a) (b)
150 Fitting Results Through an iterative methodology involving the data collection procedure and fitting routine mentioned above, size and shape parameters were first collected for DMSO solutions of 4 1a and 4 1b ( Table 4 1). Both could be fit using nearly identical prolate spheroid parameters and identical hydrodynamic radii. Further, the molecular dimensions show good ag reement with modeled values (Figure 4 2) of monomeric unaggregated species. These results indicate that DMSO d 6 does indeed suppress intermolecular H bonding of 4 1a stacking of both 4 1a and 4 1b within the range of concentrations studied Table 4 1 Fitted molecular size and shape parameters for unaggregated 4 1a,b solutions (approximately 27 mM in dimethylsulfoxide d 6 or DMSO d 6 ) Molecule Model a () b () r h () f s c 4 1a Prolate 15.5 6.60 9.00 0.776 5.76 4 1b Prolate 13.4 7.40 9.00 0.690 5.76 Next, compounds 4 1a and 4 1b were dissolved in toluene d 8 and the same data collection procedure and fitting routine were performed the results of which are summarized in Table 4 2 As expected, evidence of aggregation was borne out in a much lower diffusion coefficient for 4 1a ( 1.03 x 10 10 m 2 /s) at room temperature than for 4 1b (5.9 x 10 10 m 2 /s) Linearization of the diffusion versus temperature data for 4 1b and subsequent fitting le d to a slope of 1.1 8 x 10 15 m 3 /kg. When this value was subjected to the prolate spheroid fitting routine, values of a = 14.3 , b = 7.0 , and r h = 9.0 could be derived, which are similar to 4 1a and 4 1b in DMSO d 6 These data all support the conclusion that 4 1b i s monomeric in both solvents, and that 4 1a is monomeric in DMSO d 6
151 Interestingly, when subjected to linearization, 4 1a exhibited a distinct slope change at approximately 60 C highlighted by the intersection of the red and blue fitting lines in Figure 4 aggregation to occur between 55 75 C ( Chapter 3, specifically the discussion for Figure 3 19); coupled with this finding, the result could indicate that this temperature corresponds to a transition between stacked aggregates and discrete trimer discs. Cons equently, e ach mode of the bimodal line was fitted separately and since the discs are postulated as being the predominant species the oblate model was used for parameterization. Values of a = 24 , b = 29 , and r h = 27.4 for the first mode and a = 3.8 , b = 26.5 , and r h = 14 for the second mode taken together indicate that an oblate aggregate with a semimajor radius (29 ) nearly equal to the end to end length of 4 1a (26 , Figure 4 2) experi ences a significant size decrease only in the semiminor axis direction upon heating (first mode: a = 24 , second mode: a = 3.8 ). Cylindrical fitting of the bimodal data (i.e., treatment of the diffusing entity as a cylinder) produces similar results, w here the dimensions of the cylinder are length L and diameter d 145 The results of this model (shown in Table 4 2) also show only a decrease in the length of cylinders as the temperature increases (first mode: L = 38 , second mode: L = 5 ), while the diameter remains roughly the same.
152 Figure 4 5 Diffusion data for 4 1a,b : r aw and linearized plots with fits for (a,b) 4 1b and (c,d) 4 1a Table 4 2 Summarized results of fitting parameters for 4 1a,b studied in toluene d 8 ; a denotes the first fitted mode (temperature range: 21.8 52.6 C); b denotes the second fitted mode (temperature range: 58.8 92.2 C ) Molecule Model a () b () L () d () r h () f s c 4 1a a Oblate 24 .0 29 .0 --27.4 1.003 5. 66 4 1a a Cylinder --38 52 26.6 1.030 5.97 4 1a b Oblate 3.8 0 26.5 --14 .0 1.325 5.89 4 1a b Cylinder --5 .0 54 14 .0 1.325 5.89 4 1b Prolate 1 4.3 7.0 0 --9.00 0. 729 5. 71 With size and shape parameters in hand, molecularity calculations can be performed by comparing volumes generated via r h Even though r h represents the volume of a sphere, that spherical volume is equivalent to the volume of the fitted spheroid or cylinder (vide supra). Monomeric 4 1a and 4 1b have r h = 9.0 and are (a) (b) (c) (d)
153 designated a molecularity of 1. Both fitting routines of 4 1a in toluen e d 8 (oblate: r h = 27.4 , cylindrical: r h = 26.6 ) yield an average molecularity of 27 for the first mode which corresponds to an average supramolecular assembly of 9 trimeric discs or [( 4 1a ) 3 ] 9 Identical derived r h values ( r h = 14 ) from both fitting routines yield an average molecularity of 3.8 for the second mode, which, when coupled with the derived shape parameters a and b (or L and d ) and molecular dimensionality gained from modeling, indicates the presence of a discotic trimeri c aggregate or ( 4 1a ) 3 Importantly, these molecularity calculations are reported as averages because they are fitted over a temperature range. To clarify, temperatures in the cooler region of the first mode could produce assemblies consisting of 10 or m ore trimeric discs, such as [( 4 1a ) 3 ] 10 Conversely, temperatures in the warmer region of the first mode could produce assemblies consisting of 8 or fewer trimeric discs, such as [( 4 1a ) 3 ] 2 Nevertheless, this method of supramolecular size and shape appr oximation has provided valuable insight into the assembly characteristics of 4 1a that were not previously known. Summary and Future Work To explore the dimensionality of supramolecular aggregates of 4 1a a novel method of size and shape approximation involving variable temperature diffusion measurements was developed. Diffusion coefficients, collected via DOSY NMR, were plotted against temperature, linearized, and fitted to the modified Stokes Einstein equation using prolate spheroid, oblate spheroid, or cylindrical models. Compound 4 1a in its monomeric, unaggregated form exhibited molecular dimensions similar to its H 4 1b When examined in its aggregated form, 4 1a showed b stacked aggregates of trimeric discs, which was in good agreement with previous
154 variable temperature NMR spectroscopic experimentation. The second mode of aggregation was predominantly a discotic trimeric aggregate, which also agreed with previous findings and the literature 90 Verification of this methodology through comparison of unaggregated 4 1a and 4 1b was critically important, but verification of the discotic trimer could also be desired. This could be achieved through creating a covalently bound size similar analog of 4 1a such as the functionalized truxene 4 5 and performing the same diffusion measurements on it as well. Initial work was started on the s ynthesis of 4 5 (Figure 4 5) ; evidence for preparation of the target was provided through high resolution mass spectrometry, but the material could not be isolated through conventional techniques. Verification of this methodology could also be performed using the same variable temperature conditions with dynamic light scattering (DLS). However, DLS treatment of raw correlation data follows a cumulant fitting procedure, a data analysis technique only viable for monodisperse samples and monomodal correlati on data. Initial variable temperature studies with 20 mM toluene solutions of 4 1a showed bimodal correlation data, which ultimately caused the fitting routine to fail. Work is currently underway using a fitting algorithm (known as CONTIN) capable of han dling polydisperse samples, which should help process the DLS raw correlation data and subsequently help validate the DOSY NMR methodology.
155 Figure 4 6 Synthetic scheme for functionalized truxen e 4 5 Miscellaneous Experimental Information Synthetic Procedures 5 (5'' H exyl 5' (5 hexylthiophen 2 yl) [2,2':4',2'' terthiophen] 5 yl) 2,3 dihydro 1H inden 1 one (4 4). This compound was prepared according to Suzuki co uplings described in Appendix C. Commercially available 4 3 (0.0 253 g, 0. 120 mmol), 4 2 (prepared according to the procedure in Appendix C, 0.0 786 g, 0. 126 mmol),
156 tri t butylphosphonium tetrafluoroborate (0.0069 g, 0.024 mmol), and toluene (10 mL) were added to a two arm round bottom flask back filled with argon. Then, tribasic potassium phosphate solution (0.0 18 mL, 2.0 M) was added, and the mixture was degassed with argon for 1 h. Next, tris( dibenylideneacetone ) dipalladium (0) ( 0.0 11 g, 0.0 12 mmol) was added and solution immediately turned purple. After 3 h of stirring at rt the mixture was poured into water (200 mL) and extracted with DCM (150 mL). The organics were washed with brine, dried over MgSO 4 and concentrated in vac uo. Purification via flash column chromatography afforded 4 4 in 15% yield (0.0111 g): 1 H NMR ( CDCl 3 ) 7.76 (d, J = 8.0 Hz, 1H), 7.67 (s, 1H), 7.61 (dd, J = 8.0, 1.5 Hz, 1H), 7.37 (d, J = 3.8 Hz, 1H), 7.23 (s, 1H), 7.18 (d, J = 3.8 Hz, 1H), 6.96 (d, J = 3.5 Hz, 1H), 6.91 (d, J = 3.5 Hz, 1 H), 6.71 6.66 (m, 2 H), 3.17 (t, J = 6.1 Hz, 2 H), 2.79 (dt, J = 9.6, 7.6 Hz, 4H), 2.75 2.71 (m, 2H), 1.67 (h, J = 7.5 Hz, 4H ), 1.43 1.22 (m, 12H), 0.90 ( d t J = 6.9, 1.2 Hz, 6 H ); HRMS (DART TOF) calculated 629.2035 for C 37 H 41 OS 4 (M+H) + found 629.2048.
157 3,8,13 T ris(5'' hexyl 5' (5 hexylthiophen 2 yl) [2,2':4',2'' terthiophen] 5 yl) 10,15 dihydro 5H diindeno[1,2 a:1',2' c]fluorine (4 5). Compound 4 4 (0.007 g, 0.01 mmol) an d o dichlorobenzene (5 mL) were added to a two arm 50 mL round bottom flask that was evacuated and back filled with argon. Upon addition of boron trichloride (1 M in THF, 0.12 mL), the solution changed from yellow to purple. The mixture was heated to 50 C and stirred overnight (16 h). Thin layer chromatography (25% ethyl acetate in hexanes) revealed complete consumption of starting material, so the mixture was poured into water (200 mL) and extracted with DCM (100 mL). The organics were dried over MgSO4 and concentrated in vacuo. Purification via flash column chromatography (multiple solvent systems) did not afford a pure product: HRMS (MALDI DTL) calculated 1829.5483 for C 111 H 11 3 S 12 (M H) + found 1829.5524 (confirmed via isotope pattern).
158 Viscos ity Temperature Fitting Figure 4 7 Cubic fitting of viscosity versus temperature data for toluene found in the literature. 144
159 Figure 4 8 Cubic fitting of viscosity versus temperature data for DMSO found in the literature 143 Diffusion Versus Temperature Data with Calculated Viscosities Note: the data presented below is extracted from hard copy print outs that will be stored with the hard copy of this dissertation.
160 Table 4 3 Diffusion versus temperature data with calculated viscosities for 4 1a in toluene d 8 Temperature ( C) Temperature (K) Viscosity (Pa s) (K/Pa s) D (m 2 /s) 21.8 294.95 0.000574 513858.5 1.03E 10 25.1 298.25 0.000552 539861 1.11E 10 31.6 304.75 0.000513 594021.6 1.24E 10 38.8 311.95 0.000474 658545.3 1.42E 10 46.5 319.65 0.000436 732692.6 1.6 0 E 10 52.6 325.75 0.00041 0 795054 1.8 0 E 10 58.8 331.95 0.000385 861551.5 2.01E 10 66.1 339.25 0.00036 0 943650.2 2.28E 10 67.9 341.05 0.000354 964500.6 2.44E 10 73.2 346.35 0.000337 1027250 2.58E 10 76.7 349.85 0.000327 1069792 2.81E 10 78.8 351.95 0.000321 1095740 2.87E 10 85.1 358.25 0.000305 1175553 3.25E 10 92.2 365.35 0.000288 1269341 3.65E 10 Table 4 4 Diffusion versus temperature data with calculated viscosities for 4 1a in DMSO d 6 Temperature ( C) Temperature (K) Viscosity (Pa s) (K/Pa s) D (m 2 /s) 22.1 295.25 0.002076 142229 1.3 0 E 10 29.1 302.25 0.001826 165531.8 1.56E 10 37 .0 310.15 0.001584 195801.4 1.87E 10 45.9 319.05 0.001357 235059.1 2.35E 10 54.3 327.45 0.001183 276874 2.76E 10 64 .0 337.15 0.001022 330011 3.29E 10 Table 4 5 Diffusion versus temperature data with calculated viscosities for 4 1 b in toluene d 8 Temperature ( C) Temperature (K) Viscosity (Pa s) (K/Pa s) D (m 2 /s) 22.1 295.25 0.000572 516180.9 5.9 E 10 30.8 303.95 0.000518 587145.2 6.8 E 10 39.6 312.75 0.00047 0 666004.6 7.8 E 10 48.4 321.55 0.000428 751781.2 8.8 E 10 58.2 331.35 0.000388 854983.6 10 .0 E 10 67.8 340.95 0.000354 963336.1 11.2 E 10 77 .0 350.15 0.000326 1073479 12.5 E 10 85.5 358.65 0.000304 1180724 13.8 E 10
161 Table 4 6 Diffusion versus temperature data with calculated viscosities for 4 1 b in DMSO d 6 Temperature ( C) Temperature (K) Viscosity (Pa s) (K/Pa s) D (m 2 /s) 21.9 295.05 0.002084 141609.9 1.57 E 10 29.9 303.05 0.0018 00 168402.3 1.93 E 10 40.8 313.95 0.001482 211899.6 2.4 5E 10 53.3 326.45 0.001202 271672.1 3.1 1E 10 64.4 337.55 0.001016 332293.3 3.9 0E 10 76.3 349.45 0.000868 402493.1 4.81 E 10
162 CHAPTER 5 CONCLUSION This work has explored the synthesis optoelectronic properties, and supramolecular assemblies of rationally designed conjugated molecules capable of forming one dimensional ( 1D ) columnar nanostructures for organic semiconductive applications. First, it is well appreciated by t he community that 1D columnar assemblies are attractive for these applications because of their favorable charge transport surface orbital interactions. These architectures can be stabilized through different mode s of self assembly namely hydrogen bonding ( H bonding ) An interesting outcome of self conjugation is liquid crystalline behavior, an organizational property shown to be favorable for high performance organic semiconductive applications. For all of the self assembly approaches discussed, a community wide problem remains in that the structure property performance relationships of these rationally designed materials are not well unders tood. Fundamental understanding of how molecular structure influences bulk supramolecular structure through device based study has been only modestly explored. Instances were shown where molecules exhibited exceptional charge transport characteristics, b ut that expected performance was not observed in device based settings. In this vein, this work sought to rationally design molecules capable of self assembly into 1D columnar nanostructures, then to study physical nature of these structures in their neat form and in device based settings when possible.
163 Donor Acceptor Columnar Assemblies of Benzotrifuran: Charge Transfer Complexation and Mesomorphism Previous work on 2 1 showed that acylated derivatives could be achieved in one step, but the bulk organizat ion of those derivatives was not optimal for charge transport. This work details the rational design of a bicomponent donor acceptor system wherein acylated 2 1 derivatives, such as 2 2a,b,c were envisioned as the electron rich donor component, and melli tic triimide derivatives, such as 2 14a,b,c were envisioned as complementary electron deficient acceptor component. While obtaining acylated derivatives 2 2a,b,c and 2 10a,b,c was synthetically facile, purification was not. In a few cases, isolating the derivatives was not possible through conventional methods. A new methodology involving a Soxhlet apparatus was developed for the conversion of 2 10 to 2 11 which saw dramatically improved yields. Derivatives 2 14a,b,c were obtained according to the lite rature, although co crystallization attempts proved unsuccessful. Upon mixing equimolar n heptane solutions of 2 2c and 2 14c the solution changed from colorless to orange indicative of charge transfer complexation. The association constant of the complex was derived by a ultraviolet visible ( UV Vis ) dilution study and Benesi Hildebrand analysis and determined to be 4.33 x 10 2 M 1 which corresponds to a Gibbs free energy change of 3.56 kcal/mol. Further, the orange colored solution could be attenuated through addition of dichloromethane ( DCM ) indicating the dissolution of the charge transfer ( CT ) complex. Upon two week agi ng of the sample, the same treatment of dissolving the sample in n heptane and subsequent addition of DCM resulted in a color change from orange to purple. Characterized through a modified UV Vis solvent titration, the purple species was observed in 100% n
164 heptane solution and was persistent even in DCM. Coupled with previous observation, these results could suggest the formation of a kinetically trapped aggregate with full transfer of charge from 2 2c to 2 14c To understand the thermotropic behavior of 1:1 2 2c / 2 14c equimolar bulk mixtures, the individual components were studied first via differential scanning calorimetry ( DSC ) and polarized optical microscopy ( POM ) DSC of both individual components only experienced crystalline to isotropic transitio ns which were confirmed via POM. However, when mixed, DSC observations remained inconclusive. When POM was utilized, the room temperature birefringence was initially attributed to a biphasic mixture of both crystalline and liquid crystalline material. After further study and noting that the mixture shared no features with either individual component, a clearer understanding of phase transitions could be ascertained: the mixture of birefringent material at room temperature is the result of two different liquid crystalline textures, dendritic fans and mosaic like. From room temperature to 75 C, the texture is a mixture of highly ordered (Col h ) and lower ordered (Col r ) fans. From 75 C to 105 C, the mixture exhibits a higher degree of order with mosaic like structures. A lesser ordered Schlieren texture was observed from room temperature to 86 C in a 1:1 mixture of 2 10c and 2 14c Both features of the 1:1 2 2c / 2 14c mixture are consistent with a material having favorable organic semiconductive properties; however, due to the possible impurities arising from competitive C and O acylation further study could not be carried out. Utilization of a protection deprotecti on scheme on the carbon of 2 1 could be employed to overcome this pitfall. Essentially, in the same fashion that 2 7 is
165 methylated, a bulky, acid stable silane based protective group would be installed. The bulky silane, envisioned as sterically blocki ng C acylation, would subsequently be removed after the acylated product is achieved. Once a pure product is obtained, fabrication of a 1:1 2 2c / 2 14c system into an organic field effect transistor ( OFET ) device could shed light on the charge transport properties of the material. In depth study into the morphology of the device through atomic force microscopy ( AFM ) and X ray diffraction ( XRD ) would also be beneficial in developing a relationship between donor acceptor complexes and their use in organic semiconductive devices. Additionally, the self healing capability of the device, a common attribute of liquid crystalline devices 45 49 could be investigated by measuring e material to the desired mesophase and measuring charge transport again. Testing t his unique behavior derived from a bicomponent system could open valuable avenues of research in organic electronics. Improved Performance, Bulk Organization, and Charge Transport of a Small Molecule OPV Device through Rational Design of Hydrogen Bonded 1D Columnar Aggregates As previously mentioned, the link between bottom up rational design of organic semiconductive materials and performance has been modestly explored. Therefore, a design strategy involving appending an H bonding motif, like a phthalhydrazide, capable of forming robust self assembled columnar nanostructures favorable for charge transport to a chromophore, like a branched quaterthiophene, was developed. This rationally designed molecule was envisioned as being vacuum deposited onto a substrate, forming self assembled nanostructures, and being co deposited with an acceptor like C 60 in a bulk heterojunction ( BHJ ) for use in an organic photovoltaic ( OPV )
166 dev ice. An important consequence of this design is the modularity of this approach, which could be exploited by exchanging H bonding or chromophore units. Lastly, to derive a relationship between structure and performance, other intrinsically similar deriva tives incapable of H bond directed assembly were envisioned as being studied as well. Synthesis of phthalocyanine phthalhydrazide ( 3 1a,b ) HB D system began with attempts at creating a mono functionalized zinc (II) phthalocyanine. Unfortunately, only mass spectrometr y could elucidate the existence of 3 4b and subsequent purification and functionalization attempts were unsuccessful. Concomitant synthesis of the branched quaterthiophene phthalhydrazide ( 3 5a,b ) system performed by Dr. Jing Zhang was achieved by first synthesizing the chromophore through iterative brominations and Kumada couplings. Deprotection of Suzuki c oupl ed 3 25 could not be achieved, so a revised synthesis was entertained wherein the precursor 3 6a,b (a non H bonding congener ) was coupl ed under Suzuki conditions and cyclized with hydrazine to give 3 5a,b An alternate non H bonding congener was synthesized by Dr. Davita Watkins through Suzuki coupling of 3 27 with 3 16a to give 3 7 After optical, electrochemical, and computational stud y proved 3 5a 3 6a and 3 7 to be intrinsically similar, the aggregation behavior of the compounds was studied in bulk and in solution. Fourier transform infrared ( FT IR ) spectroscopy revealed signatures of H bonding for 3 5a in broadened peaks and shifted carbonyl stretches that were persistent even when blended with C 60 Thin film UV Vis revealed red shifted absorption for 3 5a versus both 3 6a and 3 7 indicative of improved molecular ordering.
167 Thermogravimetric analysis ( TGA ) showed higher decomposition temperature for 3 5a than 3 6a and DSC revealed a higher melting temperature for 3 5a than the non H bonding congeners with the possibility of H bond driven liquid crystalline behavior. Manual indexing of powder XRD showed a possible two dimensional ( 2D ) hexagonal packing of 3 5a which suggests the presence of 1D columnar nanostructures driven by H bonding. Thin film XRD showed several crystalline peaks for 3 5a whereas 3 6a was amorphous. Solution based variable temperat ure proton nuclear magnetic resonance ( 1 H NMR ) characterization of 3 5b showed broadened and downfield shifted N H and O H peaks indicative of H bonding, which were persistent to high temperature. The coalescence temperature and derived activation energy for proton exchange (85 C, 17 18 kcal/mol) are unusually high for amide tautomeric systems suggestive of the stacking was also noted for 3 5b due aryl peak broadening, which was not seen for 3 6b in dicating cooperation between H stacking. Finally, when blended with C 60 and studied as BHJ OPV devices as fabricated by collaboration with Prof Jiangeng Xue, 3 5a exhibited a threefold increase in performance over the H ners. While further studies are being performed to directly ascertain the ordering in the BHJ, improvements in external quantum efficiency (EQE) and charge collection length from devices containing 3 5a against 3 6a and 3 7 directly implicate enhanced electrical properties as the source of the performance boost. Combined with previous observations, bulk organization can also be indirectly responsible for the performance enhancement as well.
168 Currently, this example represents one of only a handful of structure property relationships showing the benefits of bottom up rational design of self assembled chromophores for use in BHJ OPVs. Work is being performed to explore the modularity aspect of the design as well as directly probing the BHJs of 3 5a 3 6b and 3 7 to determine the organizational details of the devices. While preliminary in nature, the results showcase a promising method toward specific fashion. diketopyr r olopyr r ole derivatives band ga p donor acceptor backbone 146,147 PCEs as high as 12 18 % can be envisioned a number considered to put OPVs on par with silicon based PVs in terms of commercial viability Additionally, this work describes the implementation of the design for a monotopic H bonding motif capable of homomeric assembly. Symmetrical functionalization of a chromophore to create bis ditopic phthalhydrazide molecules could lead to homomeric networked assemblies capable of charge transport. Expansion of the design to include heter omeric hexameric rosette assemblies of melamine and barbituric acid derivatives would further help develop the link between molecular structure, expected self assembly, and device based performance. Needless to say, this design has many areas that can be explored to help develop a greater understanding of self assembled organic semiconductive materials. Size and Shape Approximation of Supramolecular Assemblies of One Dimensional Columnar Nanostructures through Variable Temperature Diffusion Ordered Spectro scopy Due to the difficulties and limitations involved in assessing the structure of bulk assemblies of 4 1a a novel solution based diffusion ordered nuclear magnetic
169 resonance ( DOSY NMR ) technique was utilized as an alternative. Specifically, DOSY NMR e mploys a pulse sequence that measures and fits the decay of chemical shift peaks to yield a diffusion coefficient. That measurement is inversely related to the size of the molecule through the modified Stokes Einstein equation, which takes into account b oth size and shape parameters to accurately describe frictional forces felt by a diffusing entity. Those size and shape parameters were then developed into a fitting routine that could model the size and shape of a diffusing aggregate as temperature was i ncreased. This methodology was applied to 4 1a and 4 1b such that first unaggregated behavior was observed by recording diffusion versus temperature of both in dimethylsulfoxide d 6 ( DMSO d 6 ), a solvent considered to suppress intermolecular H bonding interactions. As expected, both 4 1a and 4 1b shared similar size and shape characteristics when the data was fitted to a prolate spheroid model. In toluene d 8 4 1b exhibited nearly identical si ze and shape characteristics as in DMSO d 6 as expected due to its inability to H bond. On the other hand, 4 1a in toluene d 8 exhibited drastically different diffusion behavior. Linearization of the diffusion versus temperature data revealed bimodal linea r character. Each mode was fitted separately, and the results stacked aggregates of H bonded trimers when fitted through either oblate spheroid or cylindrical models. T he second mode was determined to have size, shape, and molecularity characteristics of free H bonded trimers, which was the first direct observation of discrete discotic H bonded trimers formed by the phthalhydrazide motif.
170 Future work involves the validat ion of this technique through diffusion measurements of a covalently bound congener for the trimeric assembly of 4 1a 4 5 Comparison of diffusion of the two could elucidate the nature of the assembly seen in the second mode of aggregation of 4 1a in tol uene d 8 Additionally, dynamic light scattering ( DLS ) measurements were attempted on 4 1a in toluene, but instrumentation limitations did not allow for proper data treatment, and alternative data processing techniques are being investigated. This work rep resents a novel method for approximating the size, shape, and molecularity for supramole cular assemblies of a phthalhydrazide system that has been difficult to assess. Lateral aggregation through H bonding and vertical aggregation with an additional dynamic tautomeric equilibrium combine to make traditional 1 H NMR dilution studies quite difficult to perform and analyze. DOSY NMR measurements, which avoid tautomeric complications, simplify the system into diffusing entities, which opens the scope of this methodology to other dynamic supramolecular assemblies. Validation of the proposed donor acceptor CT complex (or any CT complex) from Chapter 2 could be performed via this technique. More information could be learned about the assembly of the purple species by executing this method in both polar and nonpolar solvents. Furthermore, this could be a critical component for elucidating modes of aggr egation of future H bonded chromophore systems from Chapter 3 While these applications are specific to 1D columnar aggregates, this method has potential to become a valuable tool for any supramolecular chemist. For example, further insight into th
171 transition from concentric spheres to prolate spheroids could be gained from this technique. 148 Additionally, morphological changes of protein substrate binding dynamics, typically studied through x ray crystallography, could be probed through this technique. Indeed, the utility of this methodology can be very far reaching and can broadly impact multiple scientific communities.
172 APPENDIX A N UCLEAR M AGNETIC R ESONANCE ( NMR ) SPECTRA
173 Figure A 1 Proton nuclear magnetic resonance ( 1 H NMR ) spectrum (CDCl 3 300 MHz) of 2 2a
174 Figure A 2 1 H NMR spectrum (CDCl 3 300 MHz) of 2 2 b
175 Figure A 3 1 H NMR spectrum (CDCl 3 300 MHz) of 2 2 c
176 Figure A 4 1 H NMR spectrum (CDCl 3 300 MHz) of 2 9
177 Figure A 5 1 H NMR spectrum (CDCl 3 300 MHz) of 2 11
178 Figure A 6 1 H NMR spectrum (CDCl 3 300 MHz) of 2 12 a
179 Figure A 7 1 H NMR spectrum (CDCl 3 300 MHz) of 2 12 b
180 Figure A 8 1 H NMR spectrum (CDCl 3 300 MHz) of 2 12 c
181 Figure A 9 1 H NMR spectrum (DMSO d 6 300 MHz) of 2 14 a
182 Figure A 10 1 H NMR spectrum (CDCl 3 300 MHz) of 2 14 b
183 Figure A 11 1 H NMR spectrum (CDCl 3 300 MHz) of 2 14 c
184 Figure A 12 Carbon nuclear magnetic resonance ( 13 C NMR ) (CDCl 3 75 MHz) spectrum of 2 2 b
185 Figure A 13 13 C NMR (CDCl 3 75 MHz) spectrum of 2 12 b
186 Figure A 14 13 C NMR (DMSO d 6 126 MHz) spectrum of 2 14 a
187 Figure A 15 13 C NMR (CDCl 3 75 MHz) spectrum of 2 14 b
188 Figure A 16 13 C NMR (CDCl 3 75 MHz) spectrum of 2 14 c
189 Figure A 17 1 H NMR spectrum (CDCl 3 300 MHz) of 1:1 mixture ( 2 2 c and 2 14 c ) after aging
190 Figure A 18 1 H NMR spectrum (CDCl 3 300 MHz) of 1:1 mixture ( 2 2c and 2 14 c ) used for DSC study
191 Figure A 19 1 H NMR spectrum (CDCl 3 300 MHz) of 3 3e
192 Figure A 20 13 C NMR (CDCl 3 75 MHz) spectrum of 3 3e
193 Figure A 21 1 H NMR spectrum (CDCl 3 300 MHz) of 3 3f
194 Figure A 22 1 H NMR spectrum (CDCl 3 300 MHz) of 3 2
195 Figure A 23 13 C NMR (CDCl 3 75 MHz) spectrum of 3 2
196 Figure A 24 1 H NMR spectrum (CDCl 3 300 MHz) of 3 18
197 Figure A 25 13 C NMR (CDCl 3 75 MHz) spectrum of 3 18
198 Figure A 26 1 H NMR spectrum (CDCl 3 300 MHz) of 3 19
199 Figure A 27 13 C NMR (CDCl 3 75 MHz) spectrum of 3 19
200 Figure A 28 1 H NMR spectrum (CDCl 3 300 MHz) of 3 20
20 1 Figure A 29 13 C NMR (CDCl 3 75 MHz) spectrum of 3 20
202 Figure A 30 1 H NMR spectrum (CDCl 3 300 MHz) of 3 23
203 Figure A 31 1 H NMR spectrum (CDCl 3 300 MHz) of 3 24a
204 Figure A 32 1 H NMR spectrum (CDCl 3 300 MHz) of 3 24b
205 Figure A 33 13 C NMR (CDCl 3 75 MHz) spectrum of 3 24a and 3 24b
206 Figure A 34 1 H NMR spectrum (CDCl 3 300 MHz) of 3 28
207 Figure A 35 13 C NMR (CDCl 3 75 MHz) spectrum of 3 28
208 Figure A 36 1 H NMR spectrum (CDCl 3 300 MHz) of deprotected 3 28
209 Figure A 37 1 H NMR spectrum (CDCl 3 300 MHz) of 3 29b
210 Figure A 38 1 H NMR spectrum (CDCl 3 300 MHz) of 3 30
211 Figure A 39 13 C NMR (CDCl 3 75 MHz) spectrum of 3 30
212 Figure A 40 1 H NMR spectrum (CDCl 3 300 MHz) of 3 31
213 Figure A 41 13 C NMR (CDCl 3 75 MHz) spectrum of 3 31
214 Figure A 42 1 H NMR spectrum (CDCl 3 300 MHz) of 3 32a
215 Figure A 43 1 H NMR spectrum (CDCl 3 5 00 MHz) of 3 32 b
216 Figure A 44 13 C NMR (CDCl 3 12 5 MHz) spectrum of 3 32b
217 Figure A 45 1 H NMR spectrum (CDCl 3 5 00 MHz) of 3 34a
218 Figure A 46 1 H NMR spectru m (CDCl 3 5 00 MHz) of 3 34 b
219 Figure A 47 1 H NMR spectrum (CDCl 3 500 MHz) of 4 4
220 APPENDIX B MISCELLANEOUS DATA Figure B 1 POM image taken of 2 2c on cooling with crossed polarizers on at 100x total magnification at (a) 8 C and (b) 34 C showing thread like crystal (this is not present on DSC thermogram ) (a) ( b )
221 Figure B 2 POM image taken on heating at 100x total magnification with crossed polarizers on of 2 2c (a) before the glass transition, (b) after the glass transition with partial healing of fractures, and (c) after the glass transition with nearly full healing of the fractures (a) ( b ) ( c )
222 Figure B 3 POM images of 2 14c taken on heating with crossed polarizers on at 100x total magnification showing (a) black streaks emanating from spherulites before the crystal to crystal transition and (b) the disappearance of the black streaks after the crystal to crystal transition Figure B 4 POM image of a 1:1 mixture of 2 2a and 2 14b taken at room temperature at 10 0 x total magnification (a) without and (b) with crossed polarizers showing the transition from orange to purple (a) ( b )
223 Figure B 5 X ray diffractogram of a 1:1 mixture of 2 2c and 2 14c dissolved in DCM, spin coated on a glass substrate, and annealed at 40 C overnight
224 Table B 1 Electrochemical potentials, electrochemical energy gaps, and corresponding highest occupied molecular orbital ( HOMO ) / lowest unoccupied molecular orbital ( LUMO ) energies for 3 5a 3 6a and 3 7 based on DPV peak onsets. Representative DPV and CV data are shown on the following pages E ox onset (V) c E red onset (V) c E g (V) c E HOMO (eV) d E LUMO (eV) d 3 6a 0.511 (0.011) 1.86 (0.05) 2.37 (0.05) 5.61 (0.01) 3.25 (0.06) 0.524 1.84 2.36 5.62 3.29 0.506 1.91 2.42 5.61 3.19 0.504 1.82 2.32 5.60 3.28 3 7 b 0.532 (0.028) 1.80 (0.04) 2.31 (0.04) 5.63 (0.03) 3.30 (0.04) 0.512 1.77 2.28 5.61 3.33 0.551 1.83 2.34 5.61 3.27 3 5a a ,e 0.525 (0.027) 5.63 (0.03) 0.513 5.61 0.556 5.66 0.505 5.61 a Values reported are the average of three DPV runs from two different sample preparations. b Values reported are the average of two DPV runs from one sample preparation. c Oxidation ( E ox onset ) and reduction ( E red onset ) potentials are reported vs. Fc/Fc + d HOMO and LUMO levels were calculated from E ox onset and E red onset respectively, considering that Fc/Fc + is 5.1 eV relative to vacuum 149 e T he compound is not readily soluble in DMF and additionally has not given reproducible data at negative potentials.
225 Figure B 6 Electrochemical data for 3 6a : (a) DPV and (b) CV Figure B 7 Electrochemical data for 3 5a : (a) DPV and (b) CV Figure B 8 Electrochemical data for 3 7 : (a) DPV and (b) CV (a) (b) (a) (b) (a) (b)
226 Figure B 9 Thermogravimetric analysis ( TGA ) scan of 3 6a
227 Figure B 10 TGA scan of 3 5a
228 Figure B 11 DSC trace of 3 6a with peak labels
229 Figure B 12 DSC trace of 3 5a with peak labels; in this particular experiment, two cycles of heating and subsequent cooling were performed, however, for clarity, only the second run is shown
230 Table B 2 Manual indexing of 3 5a powder X ray diffraction pattern according to ; t he average value for the hexagonal lattice parameter a is 26.3 ( 0.7 ) 2 I ntensity d 1/d 2 h k s=(h 2 +hk+k 2 ) (1/d 2 )/s a 3.84012 998.746 23.04763 0.001883 1 0 1 0.001883 26.61311 6.80017 323.7741 13.02042 0.005899 1 1 3 0.001966 26.04083 7.97308 312.3991 11.10744 0.008105 2 0 4 0.002026 25.65154 10.64971 1054.718 8.321034 0.014443 2 1 7 0.002063 25.42118 11.05514 262.5491 8.016769 0.01556 3 0 9 0.001729 27.7709 13.48761 524.7754 6.575942 0.023125 2 2 12 0.001927 26.30377 13.82458 317.2979 6.416405 0.024289 3 1 13 0.001868 26.71363 17.17168 428.1162 5.172536 0.037376 3 2 19 0.001967 26.03453
231 Figure B 13 Expanded FT IR spectrum for 3 6a
232 Figure B 14 Expanded FT IR spectrum of 3 7 Figure B 15 Expanded FT IR spectrum of 3 5a
233 Figure B 16 Atomic force microscopy ( AFM ) data: (a,c,e) height and (b,d,f) phase images of (a,b) 3 5a (c,d) 3 6a and (e,f) 3 7 (a) (b) (c) (d) (e) (f)
234 Figure B 17 The overall light absorption efficiency of the organic photovoltaic cells containing approximately 40 nm thick donor:C 60 (3:2 by weight) photoactive layers
235 Figure B 18 Curve fitting for the determination of the charge collection length at short circuit (symbols: experimental data; solid lines: fitted data) for (a) 3 5a (b) 3 6a and (c) 3 7 The current density voltage characteristics of the bulk heterojunction organic photovoltaic cells were fitting according to a charge collection model 114,115 using where and J Dark is the current density in the dark, is the maximum photocurrent achieved at high reverse bias, L C is the charge collection length ( L C0 is the charge collection length at 0V or short circuit), V BI is the build in voltage, and d m is the mixed layer thickness (a) (b) (c)
236 Table B 3 Miscellaneous computation energies and parameters; HOMO and LUMO energies, S 0 to S 1 vertical transition energy ( E 01 ), wavelength ( 01 ), and oscillator strength ( f ) determined with density functional theory ( DFT ) and time dependent DFT at the B3LYP/6 31G(d) level of theory HOMO (eV) LUMO (eV) E HL (eV) E 01 (eV) 01 (nm) f 3 5a (lactam tautomer) 5.20 2.18 3.02 2.73 454 0.91 3 5a (lactim lactam 1) 5.21 2.07 3.14 2.84 437 1.01 3 5a (lactim lactam 2) 5.14 2.07 3.07 2.78 446 1.02 3 6a 5.16 2.02 3.14 2.84 437 0.98 3 7 5.16 2.06 3.10 2.80 443 0.98 Table B 4 HOMO and LUMO wavefunctions for 3 5a 3 6a and 3 7 from DFT (B3LYP/6 31G(d)) calculations 3 5a (lactam tautomer) 3 6a 3 7 LUMO HOMO Cartesian Coordinates and Energies for Optimized Structures 3 5a (lactam tautomer) Energy: 2854.29973519 a.u. XYZ file generated by gabedit : coordinates in Angstrom C 6.2511580000 0.5698260000 0.2612830000 C 7.2026640000 0.3679160000 0.1742210000 C 6.7697930000 1.6225120000 0.6225420000 C 5.4201390000 1.9344510000 0.6248250000 C 4.4500640000 1.00 82850000 0.1711750000 C 4.8946400000 0.2470800000 0.2715180000 C 8.6431880000 0.0359110000 0.2113530000 N 8.9831320000 1.1814400000 0.3607080000 O 9.51260000 00 0.7570680000 0.6857100000 C 6.6739080000 1.9008420000 0.7647690000 N 8.0204740000 2.1850100000 0.6159590000 O 5.9196830000 2.7250890000 1.2661800000 C 3.0325660000 1.3655430000 0.1693750000 C 2.4618000000 2.6205740000 0.2257130000 C 1.0477440000 2.6108780000 0.1954870000 C 0.5034880000 1.3430630000 0. 1176550000
237 S 1.7783920000 0.1400870000 0.0832850000 C 0.8871690000 0.9538290000 0.0662530000 C 1.4310530000 0.3103880000 0.0269640000 C 2.8580510000 0.346912 0000 0.0198910000 C 3.4181640000 0.9250040000 0.0448170000 S 2.1656690000 2.1536180000 0.0713640000 C 4.8024280000 1.3620410000 0.0162110000 C 5.3807930000 2.4241310000 0.6436700000 C 6.7423700000 2.6397780000 0.2929180000 C 7.2182010000 1.7547130000 0.6401110000 S 5.9690440000 0.6287550000 1.1099290000 C 3.5882880000 1.6220540000 0.0310720000 C 3.3943160000 2.7038530000 0.7925000000 C 4.2081410000 3.8266240000 0.4611090000 C 5.0190290000 3.6184510000 0.6234 170000 S 4.7818590000 2.0085530000 1.2596400000 C 6.0025360000 4.5647250000 1.2456440000 C 8.5975570000 1.6717450000 1.2229320000 H 7.5126550000 2.3305520000 0.9748230000 H 5.1001450000 2.9017460000 0.9988410000 H 4.2021540000 0.9936820000 0.6473000000 H 9.8573210000 1.5588110000 0.0098180000 H 8.3554630000 2.8976030000 1.2558540000 H 3.0443150000 3.5344430000 0.2607110000 H 0.4439700000 3.5117380000 0.2242960000 H 0.8341360000 1.2159430000 0.0436970000 H 4.8440610000 3.0183820000 1.3758360000 H 7.3582440000 3.4207940000 0.7275510000 H 2.6976680000 2.6867610000 1.6243060000 H 4.1970940000 4.7610300000 1.01363 50000 H 5.7634150000 4.7745540000 2.2954670000 H 7.0259660000 4.1710280000 1.2155010000 H 5.9933050000 5.5159280000 0.7040590000 H 9.2231900000 2.4672950000 0.8061570000 H 8.5901390000 1.7871310000 2.3137350000 H 9.0797850000 0.7123620000 0.9986500000 3 5a ( lactim lactam 1 ) Energy: 2854.30217968 a.u. XYZ file generated by gabedit : coordinates in Angstrom C 6.2711780000 0.5740620000 0.2345710000 C 7.2127870000 0.3803470000 0.2008650000 C 6.7693480000 1.6467600000 0.6056430000 C 5.4207040000 1.95264 90000 0.5796380000 C 4.4602640000 1.0086760000 0.1346900000 C 4.9070500000 0.2550370000 0.2702820000 C 8.6487310000 0.0487100000 0.2346470000 N 8.9200890000 1.2389590000 0.1943130000 O 9.5461440000 0.8042970000 0.5973200000 C 6.7898400000 1.8660050000 0.6416330000 N 8.0402000000 2.1914550000 0.6267380000 O 5.8869410000 2.7860800000 1.0643970000 C 3.0417920000 1.3669120000 0.1051150000 C 2.4746280000 2.6234920000 0.0699770000 C 1.0598450000 2.6149920000 0.033 8550000 C 0.5120810000 1.3470400000 0.0444260000 S 1.7847050000 0.1415440000 0.1087440000 C 0.8796780000 0.9587120000 0.0142780000 C 1.4257750000 0.303943000 0 0.0395330000 C 2.8526260000 0.3389870000 0.0175950000 C 3.4112750000 0.9321680000 0.0436910000
238 S 2.1566770000 2.1589560000 0.0726620000 C 4.7960640000 1.3700580000 0.0188180000 C 5.3554170000 2.4343200000 0.6917050000 C 6.7259450000 2.6501410000 0.3782540000 C 7.2282810000 1.7626280000 0.5383810000 S 5 .9936490000 0.6338360000 1.0390380000 C 3.5866230000 1.6127260000 0.0196670000 C 3.4422310000 2.6622900000 0.8932110000 C 4.2468770000 3.7928970000 0.5649130 000 C 5.0008480000 3.6228380000 0.5662890000 S 4.7196370000 2.0405640000 1.2517800000 C 5.9586840000 4.5874400000 1.2003900000 C 8.6234020000 1.6786540000 1.0821860000 H 7.5069460000 2.3654450000 0.9474410000 H 5.0876560000 2.9258820000 0.9257960000 H 4.2066430000 0.9988190000 0.6344450000 H 9.8897550000 1.5279610000 0.1963530000 H 6.3909040000 3.5860710000 1.2978820000 H 3.0607650000 3.5351980000 0.0381960000 H 0.4587830000 3.5170320000 0.0115920000 H 0.8298300000 1.2100480000 0.0612170000 H 4.7984410000 3.0304700000 1.4070260000 H 7.3287890000 3.4330970000 0.8274180000 H 2.7883860000 2.6159720000 1.758009 0000 H 4.2709740000 4.7049820000 1.1532190000 H 5.6664800000 4.8422290000 2.2265800000 H 6.9786370000 4.1858050000 1.2408230000 H 5.9867080000 5.5154030000 0.6204740000 H 9.2364340000 2.4766850000 0.6516730000 H 8.6462760000 1.7892520000 2.1732800000 H 9.1001900000 0.7208120000 0.8403200000 3 5a ( lactim lactam 2 ) Energy : 2854.30241471 a.u. XYZ file generated by gabedit : coordinates in Angstrom C 6.2481360000 0.6036260000 0.2607380000 C 7.2042030000 0.3385660000 0.1704040000 C 6.7701890000 1.6078690000 0.5949340000 C 5.4220340000 1.9150960000 0.5870860000 C 4.4491220000 0.9812460000 0.1466210000 C 4.8908270000 0.2795430000 0.2739320000 C 8.590 9280000 0.0702570000 0.1517230000 N 9.0086640000 1.2320790000 0.2342220000 O 9.5079710000 0.8382920000 0.5704180000 C 6.6840190000 1.9423190000 0.7124600000 N 8.0536700000 2.1191050000 0.6487870000 O 5.9407190000 2.8344360000 1.1103720000 C 3.0330580000 1.3449630000 0.1386990000 C 2.4673760000 2.6032560000 0.1626980000 C 1.0529420000 2.5984020000 0.1339260000 C 0.5036230000 1.3314620000 0.0900430000 S 1.7740240000 0.1227130000 0.0893120000 C 0.8887630000 0.9 466630000 0.0486100000 C 1.4367210000 0.3152920000 0.0020340000 C 2.8640670000 0.3477700000 0.0054360000 C 3.4202140000 0.9250350000 0.0528510000 S 2.1637550 000 2.1497390000 0.0851420000 C 4.8038710000 1.3668960000 0.0097890000 C 5.3758680000 2.4155960000 0.6958480000 C 6.7388580000 2.6424110000 0.3572830000 C 7.2221070000 1.7793730000 0.5923370000
239 S 5.9782450000 0.6611810000 1.0938110000 C 3.5973050000 1.6210510000 0.0064830000 C 3.4060630000 2.6964350000 0 .8261520000 C 4.2220930000 3.8201820000 0.5037610000 C 5.0320920000 3.6193260000 0.5827950000 S 4.7913340000 2.0152440000 1.2322090000 C 6.0170810000 4.56880 20000 1.1977430000 C 8.6048620000 1.7125510000 1.1691660000 H 7.4992290000 2.3340710000 0.9375500000 H 5.1005640000 2.8880520000 0.9448780000 H 4.1990480000 1.0338360000 0.6357080000 H 10.3792550000 0.4094000000 0.4972070000 H 8.4134710000 3.0173240000 0.9449380000 H 3.0536660000 3.5153920000 0.1709090000 H 0.4528680000 3.5022520000 0.1359950000 H 0.8420960000 1.2224930000 0.0064910000 H 4.8334420000 2.9922960000 1.4377850000 H 7.3503400000 3.4152490000 0.812 3070000 H 2.7096670000 2.6739820000 1.6580350000 H 4.2131190000 4.7500200000 1.0639880000 H 5.7767350000 4.7892010000 2.2451400000 H 7.0395440000 4.172194000 0 1.1731050000 H 6.0112570000 5.5148830000 0.6472000000 H 9.2258470000 2.5011160000 0.7326010000 H 8.6030400000 1.8504800000 2.2573790000 H 9.0888790000 0.7501520000 0.9622320000 3 6a Energy: 2973.88435529 a.u. XYZ file generated by gabedit : coordinates in Angstrom C 5.7074210000 0.2669950000 0.0371450000 C 6.6248050000 0.8017310000 0. 0668000000 C 6.1290590000 2.1116060000 0.0399200000 C 4.7675550000 2.3651490000 0.0513910000 C 3.8418030000 1.3075680000 0.1418640000 C 4.3431410000 0.004563 0000 0.0909110000 H 6.8386390000 2.9310210000 0.0903360000 H 4.4153000000 3.3917230000 0.0440830000 H 3.6692170000 0.8531260000 0.1555410000 C 8.1114500000 0.6548900000 0.0530600000 O 8.8781370000 1.4406870000 0.5722480000 C 6.0805380000 1.7100920000 0.1997390000 O 5.5524260000 2.6230290000 0.3995860000 C 2.4109020000 1.5796160000 0.2757310000 C 1.7957190000 2.7324300000 0.7177520000 C 0.3822990000 2.6596440000 0.7359330000 C 0.1142120000 1.4450540000 0.3055 070000 S 1.2017200000 0.3795960000 0.1478450000 H 2.3473320000 3.6018340000 1.0579850000 H 0.2552770000 3.4658340000 1.0828460000 C 1.4894320000 1.0130760000 0.1928220000 C 1.9908950000 0.2658040000 0.1077130000 C 3.4159910000 0.3441790000 0.0500790000 C 4.0148390000 0.9096830000 0.0838020000 S 2.8016180000 2.1740970000 0.1692600000 H 1.3659870000 1.1513920000 0.1469620000 C 5.4095770000 1.3072940000 0.0107530000 C 6.0435240000 2.3270670000 0.6642440000 C 7. 3980560000 2.5147500000 0.2716130000 C 7.8128010000 1.6505740000 0.7087860000 S 6.5130120000 0.5804310000 1.1721710000
240 H 5.5514830000 2.9101820000 1.43570200 00 H 8.0528690000 3.2608650000 0.7106750000 C 4.1067570000 1.6404670000 0.0195340000 C 3.8411630000 2.7098580000 0.8004130000 C 4.6331600000 3.8600410000 0.5132780000 C 5.4989850000 3.6852460000 0.5340630000 S 5.3425930000 2.0733770000 1.1895820000 H 3.1079940000 2.6648690000 1.5991470000 H 4.5671040000 4.7896950000 1.0700280000 C 6.4800430000 4.6662000000 1.1042970000 H 7.5135470000 4.3077140000 1.0219780000 H 6.4104170000 5.6151160000 0.5630720000 H 6.2880 680000 4.8704250000 2.1648370000 C 9.1672980000 1.5478610000 1.3443930000 H 9.8322510000 2.3077430000 0.9219210000 H 9.1249620000 1.7046200000 2.4292430000 H 9.6266070000 0.5664780000 1.1733900000 O 8.5148920000 0.4039000000 0.6822070000 O 7.0055050000 1.8840710000 1.1664920000 C 9.9364080000 0.5970330000 0.7472400000 H 10.3443760000 0.7822660000 0.2502500000 H 10.4237460000 0.2843660000 1.1718350000 H 10.0832040000 1.4650010000 1.3906170000 C 7.4056730000 3.24 54510000 1.3897470000 H 7.8575080000 3.6642160000 0.4863550000 H 6.5447880000 3.8567280000 1.6724280000 H 8.1325760000 3.2059400000 2.2014370000 3 6a Energy: 2932. 92279230 a.u. XYZ file generated by gabedit : coordinates in Angstrom C 5.7339660000 0.1787620000 0.1332910000 C 6.6310280000 0.8408490000 0.2076630000 C 6.1363420000 2.1183110000 0.5062150000 C 4.7749510000 2.3656430000 0.4584450000 C 3.8555380000 1.3490930000 0.1026350000 C 4.3626280000 0.0759580000 0.1936490000 H 6.8414870000 2 .8919560000 0.7903240000 H 4.4046120000 3.3518610000 0.7198780000 H 3.7120280000 0.7372030000 0.4984480000 C 8.0760360000 0.5739160000 0.3368430000 N 8.49136 00000 0.7057930000 0.0006100000 O 8.8748540000 1.4098990000 0.7574140000 C 6.2263140000 1.5210210000 0.5174390000 N 7.5930250000 1.7195840000 0.4150850000 O 5.4854740000 2.4043370000 0.9457630000 C 2.4228850000 1.6383200000 0.0473070000 C 1.7973050000 2.8623650000 0.0698910000 C 0.3850140000 2.7846310000 0.1068770000 C 0.1016290000 1.4950560000 0.0143820000 S 1.2252750000 0.3580410000 0.1305770000 H 2.3399090000 3.7968130000 0.1607140000 H 0.2582190000 3.651 4740000 0.2159020000 C 1.4736160000 1.0400090000 0.0229000000 C 1.9577140000 0.2481420000 0.0585760000 C 3.3812320000 0.3529130000 0.0178510000 C 3.999875000 0 0.8902890000 0.0386080000 S 2.8064490000 2.1769770000 0.0450620000 H 1.3175690000 1.1234950000 0.0824160000 C 5.4050420000 1.2615470000 0.0236830000 C 6.0172940000 2.2748140000 0.7277200000
241 C 7.3965010000 2.4328120000 0.4162310000 C 7.8524430000 1.5518610000 0.5303950000 S 6.5620190000 0.5031620000 1.0633300000 H 5.4915250000 2.8740650000 1.4637580000 H 8.0383470000 3.1699440000 0.8883560000 C 4.0481880000 1.6618460000 0.0016300000 C 3.8116470000 2.7327 960000 0.8251930000 C 4.5642170000 3.8946150000 0.4840450000 C 5.3702380000 3.7269860000 0.6110720000 S 5.2052770000 2.1073270000 1.2451110000 H 3.1276310000 2.6812050000 1.6660180000 H 4.5138010000 4.8273420000 1.0372590000 C 6.2980120000 4.7207250000 1.2448200000 H 6.0398190000 4.9137590000 2.2933430000 H 7.3408580000 4.3810390000 1.2218580000 H 6.2436180000 5.6722610000 0.7064490000 C 9.2407180000 1.4164430000 1.0814210000 H 9.8940010000 2.1673680000 0.62 60140000 H 9.2674300000 1.5629560000 2.1681940000 H 9.6687720000 0.4280140000 0.8738960000 C 8.1877440000 2.8308150000 1.1542010000 H 7.3561100000 3.41662300 00 1.5417950000 H 8.7989090000 2.4541140000 1.9827170000 H 8.8008110000 3.4622280000 0.5055810000 C 9.7834540000 1.1680220000 0.5001330000 H 9.6438300000 1.9250970000 1.2810770000 H 10.4005190000 1.5807500000 0.3023350000 H 10.2765540000 0.2914680000 0.9168070000
242 Figure B 19 Screenshot of Microsoft Excel spreadsheet used to fit 4 1b in dimethylsulfoxide d 6 ( DMSO d 6 ) using the prolate spheroid model
243 The red boxes are the only values changed for the entire spreadsheet (all are in , which are subsequently converted to m) value converted to m tio a:b are references to the values for a and b and the formula tudio 3.5.0 obtained from a MM2 minimized model of DMSO made in Chem 3D Pro van der Waals volume equal to the volume of a sphere; embedded in the cell is the formula rh and the formula C) and diffusion c oefficients (in m 2 /s); temp was converted to K by adding 273.15 to the original temperature; the embedded formula ( Figure 4 7) where T is the temperature in K Einstein equation with refe rences to Kb, T (temperature in K), c, fs, the embedded formula
244 the spheroid generated by a and b; it contains a reference to the value of spheroid volume and has the embedded formula hydrodynamic radius and the equivalent sphere radius necessary to adjust a and b such that the vol ume generated by them matches the volume generated by rh.
245 APPENDIX C SYNTHETIC PROCEDURES 2 Bromo 5 methylthiophene (3 9a ). In the absence of light, 2 methylthiophene ( 3 8a ) (1.00 mL, 10.3 mmol) was added to a solution of N bromosuccinimide (2.07 g, 11.4 mmol) in chloroform/acetic acid (10 mL of a 1:1 solution). The resulting solution was stirred at 0 C for 1 h. The mixture was then allowed to warm to room temperature and s tirred for an additional 12 h. The reaction was quenched with aqueous NaOH. The organic layer was separated, washed with water, and dried over MgSO 4 The product was distilled under reduced pressure and obtained as a pale yellow oil in 75% yield (1.4 g): 1 H NMR (CDCl 3 500 MHz): 6.82 (d, J = 3.6 Hz, 1H), 6.51 (d, J = 3.6, 1H), 2.42 (s, 3H) ppm; 13 C NMR (CDCl 3 ): 141.5, 129.7, 125.6, 108.7, 15.6 ppm. The 1 H and 13 C NMR data match that found in the literature 150 5,5 '' Dime thyl 2,2':3',2'' terthiophene (3 11a ). Under argon, 3 9a (3.97 g, 22.4 mmol) was added dropwise to a suspension of iodine and magnesium turnings (0.63 g, 26 mmol) in dry diethyl ether (20 mL) to form the Grignard reagent. The resulting solution was heated to reflux for 1 h. After cooling to room temperature, the Grignard reagent was the n slowly added to a mixture of 3 10 (2.85 mL, 25.2 mmol) and [1,3 bis(diphenylphosphino)propane]dichloronickel(II) (Ni(dppp) 2 Cl 2, 160 mg, 0.30 mmol) in diethyl ether (100 mL) at 0 C under argon. The resulting mixture was heated to reflux
246 for 24 h and then quenched with 1 M HCl (10 mL). The mixture was extracted with diethyl ether (300 mL). The organic layers were combined, washed with water, dried over MgSO 4 and conc entrated under reduced pressure. The product was purified by flash column chromatography (petroleum ether) and obtained as a green oil in 90% yield (5.52 g): 1 H NMR (CDCl 3 300 MHz): 7.22 (d, J = 5.2 Hz, 1H), 7.13 (d, J = 5.2 Hz, 1H), 6.95 (d, J = 3.2 Hz 1H), 6.88 (d, J = 3.2 Hz, 1H), 6.68 (dd, J = 2.4 Hz, J = 2.4 Hz, 2H), 2.49 (s, 6H) ppm. The 1 H NMR data match es that found in the literature 151 5' Bromo 5,5'' dime thyl 2,2':3',2'' terthiophene (3 12a ). In the absence of light, 3 11a (1.45 g, 5.25 mmol) was added to a solution of N bromosuccinimide (1.01 g, 5.67 mmol) in chloroform, and the resulting solution was stirred at 0 C for 1 h. The mixture was allowed to warm to room temperature and stirred for 12 h. The mixture was then warmed to 30 C and allowed to react for an additional 24 h. The reaction was quenched with aqueous NaOH. The organic layer was separated, washed with water, dried over MgSO 4 and concentrated under reduced pressure. The product was then purified by flash column chromatography (hexanes) and obtained as a green oil in 88% yield (1.64 g): 1 H NMR (CDCl 3 ): 7.07 (s, 1H), 6.89 (d, J = 3.4 Hz, 1H), 6.82 (d, J = 3.4 Hz, 1H), 6.66 (d, J = 2.5 Hz, 1H), 6.63 (d, J = 2.4 Hz, 1H), 2.46 (s, 3 H), 2.45 (s, 3H) ppm; 13 C NMR (CDCl 3 500 MHz, determined via gHMBC): 142.6, 141.7, 140.1, 138.9, 133.6, 132.0, 131.0, 128.1, 126.2, 125.8, 125.6, 105.0, 16.5, 15.4 ppm; HRMS (MALDI TOF) calculated 354.9100 for C 14 H 10 S 3 Br (M H) + found 354.9106.
247 5,5'' Dimethyl 5' (thiophen 2 yl) 2,2',3',2'' terthiophene (3 14a ). Under argon, 3 13 (10.3 g, 63.3 mmol) was added dropwise to a suspension of iodine and magnesium (1.52 g, 63.3 mmol) in dry diethyl ether (20 mL) to form the Grignard reagent. The resulting solution was heated to reflux for 1 h. After cooling to room temperature, the Grign ard reagent was slowly added to a mixture of 3 12a (13 g, 37 mmol) and [1,3 bis(diphenylphosphino)propane]dichloronickel(II) (Ni(dppp) 2 Cl 2, 198 mg, 0.36mmol) in dry diethyl ether (100 mL) at 0 C under argon. The resulting mixture was heated to reflux for 24 h and quenched with dilute 1 M HCl (10 mL). The mixture was extracted with diethyl ether (300 mL). The organic layers were combined, washed with water, dried over MgSO 4 and concentrated under reduced pressure. The product was then purified by flash col umn chromatography (hexanes) and obtained as a yellow liquid in 92% yield (12 g): 1 H NMR (CDCl 3 ): 7.40 (s, 1H), 7.35 (m, 2H), 7.17 (m, 3H), 6.86 (m, 2H), 2.65 (dd, J = 7.7, 1.1Hz, 6H) ppm; 13 C NMR (CDCl 3 ): 141.3, 140.0, 136.6, 135.2, 134.8, 132.4, 132. 3, 130.3, 127.9, 127.8, 126.7, 126.1, 125.5, 125.4, 124.6, 123.9, 15.31, 15.28 ppm; HRMS (APCI TOF) calculated 359.0051 for C 18 H 14 S 4 (M+H) + found 359.0065.
248 4,4,5,5 Tetramethyl 2 (5'' methyl 5' (5 methylthiophen 2 yl) [2 ,2':4',2'' terthioph en] 5 yl) 1,3,2 dioxaborolane (3 16a ). Under argon, n butyllithium in hexane (2.5 M, 4.4 mL, 11 mmol) was added to a solution of 5,5'' dimethyl 5' (thiophen 2 yl) 2,2',3',2'' terthiophene 3 14a (3.9 g, 10 mmol) in dry THF (150 mL) at 7 8 o C and the mixture was stirred at this temperature for 2 h. Compound 3 15 (2.2 mL, 11 mmol) was added, and the reaction mixture was warmed to room temperature and stirred for an additional 12 h. The reaction was then quenched with brine (50 mL) and the p roduct was extracted with diethyl ether. The organic layers were combined, washed with water, dried over MgSO 4 and concentrated under reduced pressure. The crude product was purified by gradient flash column chromatography (0 20% dichloromethane in hexane) to give the product as a green liquid in 79% yield (3.8 g): 1 H NMR (CDCl 3 ): 7.58 (d, J = 4.0 Hz, 1H), 7.28 (s, 1H), 6.99 (d, J = 3.2 Hz, 1H), 6.93 (d, J = 3.3 Hz, 1H), 6.70 (d, J = 2.5 Hz, 2H), 2.52 (d, J = 4.0 Hz, 6H), 1.39 (s, 12H) ppm; 13 C NMR (CDCl 3 ): 143.1, 141.3, 140.0, 137.7, 134.8, 134.5, 132.2, 132.0, 130.8, 127.6, 126.5, 126.4, 125.3, 125.1, 124.8, 83.9, 24.5, 15.2, 15.1 ppm; HRMS (APCI TOF) calculated 485.0908 for C 24 H 25 BO 2 S 4 (M+H) + found 485.0908.
249 Dimethyl 4 (5'' methyl 5' (5 methylthiophen 2 yl) [2,2':4',2'' terthiophen] 5 yl)phthalate ( 3 6a ). Under argon, d egassed toluene (15 mL) was added to a suspension of 3 16a (1.5 g, 3.1 mmol), potassium phosphate tribasic (K 3 PO 4, 2 M aqueous solution, 10 mL), tris(dibenzylideneacetone)dipalladium(0) ( Pd 2 (dba) 3, 21 mg), tri tert butylphosphonium tetrafluoroborate (( t Bu) 3 P HBF 4, 27 mg) and 3 26 (1.3 g 4.5 mmol). The solution was heated to reflux for 24 h. After cooling to room temperature, the mixture was poured into water and extracted with methylene chloride. The organic layer s were combined, washed with water, dried over MgSO 4 and concentrated under reduced pressure. The product was purified by flash column chromatography (petroleum ether/CH 2 Cl 2 1:1) and obtained as a yellow solid in 61% yield (1.04 g): 1 H NMR (CDCl 3 ): 7.8 6 (d, J = 1.7 Hz, 1H), 7.82 (d, J = 8.1 Hz, 1H), 7.72 (dd, J = 8.1, 1.8 Hz, 1H), 7.35 (d, J = 3.9 Hz, 1H), 7.20 (s, 1H), 7.17 (d, J = 3.9 Hz, 1H), 6.96 (d, J = 3.6 Hz, 1H), 6.91 (d, J = 3.5 Hz, 1H), 6.68 (d, J = 3.2 Hz, 2H), 3.95 (d, J = 9.9 Hz, 6H), 2.49 (d, J = 5.0 Hz, 6H) ppm; 13 C NMR (CDCl 3 ): 168.2, 167.1, 141.6, 140.4, 140.3, 137.9, 137.1, 134.6, 134.5, 133.7, 132.4, 132.2, 131.1, 130.1, 129.2, 127.9, 127.0, 126.9, 126.6, 125.8, 125.6, 125.4, 125.1, 124.9, 52.8, 52.6, 15.42, 15.38 ppm; HRMS (ESI TOF) calculated 551.0474 for C 28 H 22 O 4 S 4 (M+H) + found 551.0491; elemental analysis calculated C: 61.07; H: 4.03 and found C: 61.13; H: 3.74.
250 6 (5'' Methyl 5' (5 methylthiophen 2 yl) [2,2':4',2'' terthiophen] 5 yl) 2,3 dihydrophthalazine 1,4 dione ( 3 5a ). Under argon, anhydrous hydrazine (2.42 mL, 50 mmol) was added to a solution of 3 6a (0.55 g, 1.0 mmol) in DMF (40 mL). The reaction mixture was heated at 80 C for 24 h. The mixture was then cool ed to 0 C and ethanol (40 mL) was added. The yellow precipitate formed was isolated by filtration. The precipitate was recrystallized from DMF ethanol (1:1) to yield the product as a dark orange solid in 31% yield (400 mg): 1 H NMR (DMSO d 6 ): 11.61 (s, 2 H), 8.22 (d, J = 8.8 Hz, 2H), 8.09 (d, J = 8.4 Hz, 1H), 7.81 (d, J = 3.9 Hz, 1H), 7.54 (s, 1H), 7.48 (d, J = 3.8 Hz, 1H), 7.09 (d, J = 3.5 Hz, 1H), 7.04 (d, J = 3.5 Hz, 1H), 6.81 (m, 2H), 2.44 (d, J = 5.2 Hz, 6H) ppm; 13 C NMR (DMSO d 6 ): 162.3, 141.8, 140 .7, 140.2, 136.8, 136.7, 134.2, 133.5, 132.6, 130.8, 129.6, 129.1, 128.7, 127.4, 127.2, 126.5, 126.2, 126.1, 125.7, 120.6, 15.0, 14.9 ppm; HRMS (DART TOF) calculated 519.0324 for C 26 H 18 N 2 O 2 S 4 (M+H) + found 519.0321; elemental analysis calculated N: 5.40; C : 60.21; H: 3.50 and found N: 5.50; C: 59.95; H: 3.76.
251 6 Bromo 2,3 dimethyl 2,3 dihydrophthalazine 1,4 dione ( 3 27 ). To a solution of 3 22 (1.03 g, 4.54 mmol) in glacial acetic acid (50 mL) was added N N dimethylhydrazine dihydrochloride (1.13 g, 9.86 mmol) and the reaction mixture heated at reflux for 18 h. The reaction mixture was then cooled, and poured into deionized water. The mixture was extracted with methylene chloride (300 mL). The organic layers were combined, washed with water, dried over MgSO 4 and concentrated under reduced pressure. The product was purified by flash column chromatography (methylene chloride and ethyl acetate, 1:1) and obtained as a white solid in 71% yield (0.86 g): 1 H NMR (C DCl 3 ): 8.45 (s, 1H), 8.19 (d, J = 8.2 Hz, 1H), 7.87 (d, J = 8.2 Hz, 1H), 3.74 (s, 6H) ppm; 13 C NMR (CDCl 3 ): 157.3, 156.6, 136.7, 130.6, 130.3, 129.5, 128.7, 127.7, 33.5, 33.4 ppm; HRMS (DIP CI DSQ) calculated 268.9926 for C 10 H 10 O 2 N 2 Br (M+H) + found 268 .9931. 6 (5'' Methyl 5' (5 methylthiophen 2 yl) [2,2':4',2'' terthiophen] 5 yl) 2,3 dimethyl 2,3 dihydrophthalazine 1,4 dione ( 3 7 ). Under argon, degassed toluene (15 mL) was added to a suspension of 3 16a (0.8 g, 2 mmol), potassium phosphate tribasic (K 3 PO 4, 2 M aqueous solution, 5 mL), tris(dibenzylideneacetone)dipalladium(0) ( Pd 2 (dba) 3, 11 mg),
252 tri tert butylphosphonium tetrafluoroborate (( t Bu) 3 P HBF 4, 15 mg) and 3 27 (0.83 g, 3.1 mmol). The solution was heated to reflux for 24 h. After cooling to room temperature, the mixture was poured into water and extracted with methylene chloride. The organic layers were combined, washed with water, dried over MgSO 4 and concentrated under red uced pressure. The product was purified by flash column chromatography (ethyl acetate/methylene chloride, 3:7) followed by recrystallization from chloroform ethanol (1:1) to obtain the product as an orange solid in 30% yield (0.262 g): 1 H NMR (DMSO d 6 ): 8.32 (d, J = 1.7 Hz, 1H), 8.23 (dd, J = 8.1,1.7 Hz, 1H), 8.19 (d, J = 8.1 Hz, 1H), 7.85 (d, J = 3.9 Hz, 1H), 7.58 (s, 1H), 7.50 (d, J = 3.9 Hz, 1H), 7.11 (d, J = 3.6 Hz, 1H), 7.07 (d, J = 3.5 Hz, 1H), 6.83 (d, J = 3.2 Hz, 1H), 6.80 (d, J = 3.2 Hz, 1H), 3.6 9 (s, 3H), 3.67 (s, 3H), 2.46 (s, 3H); 2.44 (s, 3H) ppm; 13 C NMR (DMSO d 6 ): 156.5, 156.5, 142.3, 140.7, 140.7, 138.0, 137.4, 134.6, 134.0, 133.1, 131.3, 130.1, 130.0, 129.8, 129.2, 128.7, 128.0, 127.9, 127.6, 127.1, 126.7, 126.6, 126.2, 122.8, 33.4, 33.2, 15.5, 15.4 ppm; HRMS (DART TOF) calculated 547.0637 for C 28 H 23 N 2 O 2 S 4 (M+H) + found 547.0642; elemental analysis calculated N: 5.12; C: 61.51; H: 4.06 and found N: 4.99; C: 61.53; H: 3.90.
253 Figure C 1 Proton nuclear magnetic resonance ( 1 H NMR ) spectrum of compound 3 12a (300 MHz, chloroform d or CDCl 3 )
254 Figure C 2 1 H NMR spectrum of compound 3 14a (300 MHz, CDCl 3 )
255 Figure C 3 Carbon nuclear magnetic resonance ( 13 C NMR ) spectrum of compound 3 14a (75 MHz, CDCl 3 )
256 Figure C 4 1 H NMR spectrum of compound 3 16a (300 MHz, CDCl 3 )
257 Figure C 5 13 C NMR spectrum of compound 3 16a (75 MHz, CDCl 3 )
258 Figure C 6 1 H NMR spectrum of compound 3 6a (300 MHz, CDCl 3 )
259 Figure C 7 1 H NMR spectrum of compound 3 6a (500 MHz, toluene d 8 room temperature )
260 Figure C 8 13 C NMR spectrum of compound 3 6a (75 MHz, CDCl 3 )
261 Figure C 9 Expanded 13 C NMR spectrum of compound 3 6a (75 MHz, CDCl 3 ) of the aryl region
262 Figure C 10 1 H NMR spectrum of compound 3 5a (300 MHz, DMSO d 6 )
263 Figure C 11 13 C NMR spectrum of compound 3 5a (75 MHz, DMSO d 6 )
264 Figure C 12 1 H NMR spectrum of compound 3 27 (300 MHz, CDCl 3 )
265 Figure C 13 13 C NMR spectrum of compound 3 27 (75 MHz, CDCl 3 )
266 Figure C 14 1 H NMR spectrum of compound 3 7 (500 MHz, DMSO d 6 )
267 Figure C 15 13 C NMR spectrum of compound 3 7 (125 MHz, DMSO d 6 )
268 2 hexylthiophene ( 3 8b ). Under argon, n butyllithium ( 40 mL, 2.5 M in hexanes) was added dropwise to a stirred solution of thiophene ( 9.45 g, 0.113 mol) in dry THF ( 150 mL) at 78 C for 1 h Bromohexane ( 16.5 g 0.100 mol) was then added dropwise at 78 C. The mixture was then allowed to warm to room temperature and stirred overnight. A saturated solution of ammonium chloride (40 mL) was added, and the reaction mixture was extracted with diethyl ether. The combined organics were dried over MgSO 4 and concentrated in vacuo. The product was isolated via distillation on high vacuum as an oil in 57% yield (9.6 g ). 2 bromo 5 hexylthiophene ( 3 9b ). In the absence of light, 3 8b (9.6 g, 571 mmol) was added to a solution of N bromosuccinimide (10.68 g, 600 mmol) in DMF (50 mL). The resulting solution was stirred at 0 o C for 1 h. The mixture was then allowed to warm to room temperature and stirred for an additional 12 hrs. The mixt ure was poured into water and extracted with DCM. The organics were collected, dried over MgSO 4 and concentrated in vacuo. The product was isolated via distillation on high vacuum as an oil in 75% yield (10.58 g )
269 5,5'' dihexyl 2,2':3',2'' terthiophene ( 3 11b ). This compound was prepared according to the procedure used on compound 3 11a with 3 9b (6.2 g, 25.1 mmol), iodine ( 1 granule ), magnesium (1 g, 41.7 mmol), dry diethyl ether (100 mL), 3 10 (2.42 g, 10.0 mmol), and [1,3 Bis(diphenylphosphino)propane]dichloronickel(II) (Ni(dppp) 2 Cl 2, 0.5 g, 0.92 mmol). The product was then purified by flash column chromatography (hexanes) and obt ained as a pale yellow oil in 84% yield (3.5 g ) 5' bromo 5,5'' dihexyl 2,2':3',2'' terthiophene ( 3 12b ). This compound was prepared according to the procedure used on compound 3 12a with 3 11b (3.5 g, 8.4 mmol) and N bromosuccinimide (1.51 g, 8.5 mmol). The product was then purified by flash column chromatography (hexanes) and obtained as an oil in 55% yield (2.3 g): 1 H NMR (CDCl 3 ): 7.10 (s, 1H), 6.91 (d, J = 3.5 Hz, 1H), 6.84 (d, J = 3.5 Hz, 1H), 6.68 (dt, J = 3.6, 0.9 Hz, 1H), 6.65 ( dt J = 3.6, 0.9 Hz, 1H), 2.77 ( dt J = 8.0, 5.6, 4H), 1.78 1.56 (m, 4H), 1.46 1.2 1 (m, 12H), 0.96 0.83 (m, 6H) 5,5'' dihexyl 5' (thiophen 2 yl) 2,2',3',2'' terthiophene ( 3 14b ). This compound was prepared according to the procedure used on compound 3 14a with 3 12b (12.50 g, 25.2 mmol), iodine (1 granule ), magnesium (0.78 g, 32.8 mmol), dry diethyl ether (20
270 mL), 3 13 (5.33 g, 32.8 mmol), and [1,3 Bis(diphenylphosphino)propane]dichloronickel(II) (Ni(dppp) 2 Cl 2, 136 mg, 0.25 mmol). The product was then purified by flash column chromatography (hexanes) and o btained as a yellow liquid in 87% yield. (10.96 g) 1 H NMR (CDCl 3 J = 4.9, 3.5 Hz, 1H), 7.00 (d, J = 3.4 Hz, 1H), 6.95 (d, J = 3.5 Hz, 1H), 6.72 (d, J = 3.6 Hz, 2H), 2.85 (m, 4H), 1.73 (m, 4H), 1.44 (m, 12H), 0.92 (t, 6H); 13 C NM R (CDCl 3 146.9, 137.4, 135.7, 135.1, 132.9, 132.7, 131.1, 128.5, 128.0, 127.0, 126.8, 125.2, 124.8, 124.7, 124.4, 32.2, 32.18, 32.1, 30.8, 30.7, 29.4, 23.2, 14.7; HRMS (ESI) calculated 498.1538 for C 2 8 H 3 4 S 4 (M ) + found 499.1531. 2 (5'' hexyl 5' (5 hexylthiophen 2 yl) [2,2':4',2'' terthiophen] 5 yl) 4,4,5,5 tetramethyl 1,3,2 dioxaborolane ( 3 16b ). This compound was prepared according to the procedure used on compound 3 16a with 3 14b (2.4 g, 4.8 mmol), n butyllithium in hexane (2.5 M, 2.12 mL, 5.3 mmol), and 3 15 (1.08 mL, 5.3 mmol). The crude product was purified by gradient flash column chromatography (0 20% dichloromethane in hexane) to give product as a green liquid in 51% yield. (1.51 g). 1 H NMR (CDCl 3 (dd, J = 3.7, 1.4 Hz, 1H), 7.28 (m, 2H), 6.99 (dd, J = 3.7, 1.3 Hz, 1H), 6.93 (dd, J = 3.6, 1.3 Hz, 1H), 6.70 (d, J = 3.8 Hz, 2H), 2.84 (m, 4H), 1.73 (m, 4H), 1.45 (m, 24H), 0.95 (t, 6H); 13 C NMR (CDCl 3 143.6, 138.1, 135.1, 134.6, 132.6, 132.2, 131.3, 127.7, 126.9, 126.5, 125.1, 124.4, 124.2, 84.3, 31.7, 31.7, 30.3, 30.29, 28.9, 24.9, 22.8, 14.3; HRMS (ESI TOF) calculated 625.2475 for C 3 4 H 4 6 BO 2 S 4 (M+H ) + found 625.2501.
271 Dimethyl 4 (5'' hexyl 5' (5 hexylthiophen 2 yl) [2,2':4',2'' terthiophen] 5 yl)phthalate ( 3 6b ). This compound was prepared according to the procedure used on compound 3 6a with 3 16b (0.624 g, 1 mmol), potassium phosphate (tribasic, K 3 PO 4, 2 M aqueous solution, 10 ml), tris (dibenzylideneacetone)dipalladium(0) ( Pd 2 (dba) 3, 7 mg, 0.010 mmol), tri tert butylphosphonium tetrafluoroborate (( t Bu) 3 P HBF 4, 9 mg, 0.024 mmol), compound 3 26 (0.409 g 1.5 mmol), and deg assed toluene (15 mL). The product was purified by flash column chromatography (petroleum ether/CH 2 Cl 2 1:1) and obtained as a yellow solid in 61% yield. (0.418 g); 1 H NMR (CDCl 3 J = 1.8 Hz, 1H), 7.72 (d, J = 7.9 Hz, 1H), 7.60 (dd, J = 8.0, 1 .8 Hz, 1H), 7.22 (d, J = 3.7 Hz, 1H), 7.17 (s, 1H), 7.05 (d, J = 4.1 Hz, 1H), 6.94 (d, J = 3.6 Hz, 1H), 6.90 (d, J = 3.6 Hz, 1H), 6.67 (dd, J = 9.1, 3.6 Hz, 2H), 3.94 (d, J = 9.9 Hz, 6H), 2.80 (m, 4H), 1.69 (m, 4H), 1.40 (m, 12H), 0.92 (t, 6H); 13 C NMR (CD Cl 3 137.0, 134.4, 134.3, 133.7, 132.3, 132.0, 131.2, 130.0, 129.1, 127.6, 126.9, 126.5, 125.7, 125.0, 124.8, 124.3, 124.1, 52.7, 52.5, 31.6, 31.6, 31.6, 31.6, 31.5, 30.2, 28.8, 22.7, 14.1; HRMS (APCI TOF) calcu lated 690.1960 for C 3 8 H 4 2 O 4 S 4 (M) + found 690.1945.
272 6 (5'' hexyl 5' (5 hexylthiophen 2 yl) [2,2':4',2'' terthiophen] 5 yl) 2,3 dihydrophthalazine 1,4 dione ( 3 5b ). This compound was prepared according to the procedure used on compound 3 5a with 3 6b (0.50 g, 0.72 mmol), anhydrous hydrazine (1.75 mL, 36 mmol), and DMF (40 ml). The precipitate was recrystallized from DMF ethanol (20 ml/20 ml) to obtain the product as o range solid in 36% yield. (169 mg); 1 H NMR (DMSO d 6 J = 4.1 Hz, 1H), 7.30 (s, 1H), 7.23 (d, J = 4.1 Hz, 1H), 6.84 (d, J = 3.6 Hz, 1H), 6.80 (d, J = 3.6 Hz, 1H), 6.57 (dd, J = 11.8, 3.6 Hz, 2H), 2.28 (m, 4 H), 1.35 (m, 4H), 1.08 (m, 12H), 0.63 (t, 6H); 13 C NMR (DMSO d 6 129.8, 129.1, 127.9, 127.8, 127.2, 127.1, 126.9, 126.8, 125.7, 125.2, 31.8, 31.7, 31.6, 31.6, 30.0, 29.99, 28.7, 28.70, 22.7, 14.6; HRMS (APCI TOF) calculated 659.1889 for C 3 6 H 3 9 N 2 O 2 S 4 (M+H) + found 659.1890; CHN calculated N: 4.247; C: 65.553; H: 5.766 and found N: 4.199; C: 65.760; H: 5.845.
273 LIST OF REFERENCES (1) Mishra, A.; Buerle, P. Angew. Chem. Int. Ed. 2012 51 2020. (2) Xue, J. In Organic Electronics ; 2010; pp. 359. (3) Kalowekamo, J.; Baker, E. Solar Energy 2009 83 1224. (4) Coropceanu, V.; Cornil, J.; da Silva Filho, D. A.; Olivier, Y.; Silbey, R.; Bredas, J. L. Chem. Rev. 2007 107 926. (5) Li, H.; Giri, G.; Tok, J. B. H.; Bao, Z. MRS Bull 2013 38 34. (6) Brdas, J. L.; Beljonne, D.; Coropceanu, V.; Cornil, J. Chem. Rev. 2004 104 4971. (7) Lin, Y.; Li, Y.; Zhan, X. Chem. Soc. Rev. 2012 41 4245. (8) Tovar, J. D. Acc. Chem. Res. 2013 DOI: 10.1021/ar3002969 (9) D.; Schenning, A. P. H. J. Chem. Mater. 2011 23 310. (10) Arias, A. C.; MacKenzie, J. D.; McCulloch, I.; Rivnay, J.; Salleo, A. Chem. Rev. 2010 110 3. (11) Dong, H.; Wang, C.; Hu, W. Chem. Comm. 201 0 46 5211. (12) Li, L.; Kosina, H. In Organic Electronics ; 2010; Vol. 223, pp. 301. (13) Braga, D.; Horowitz, G. Adv. Mater. 2009 21 1473. (14) Hoeben, F. J. M.; Jonkheijm, P.; Meijer, E. W.; Schenning, A. P. H. J. Chem. Rev. 2005 105 1491. (15) Myers, J. D.; Xue, J. Polym. Rev. 2012 52 1. (16) Xue, J. Polym. Rev. 2010 50 411. (17) Saeki, A.; Koizumi, Y.; Aida, T.; Seki, S. Acc. Chem. Res. 2012 45 1193. (18) Silicon: Evolution and Future of a Technology ; Siffert, P.; Krimmel, E. F., Eds.; Springer: Berlin, 2004. (19) Klosterman, J. K.; Yamauchi, Y.; Fujita, M. Chem. Soc. Rev. 2009 38 1714. (20) Yamamoto, Y. Sci. Tech. Adv. Mater. 2012 13 033001.
274 (21) Kim, F. S.; Ren, G.; Jenekhe, S. A Chem. Mater. 2011 23 682. (22) Zheng, Y.; Xue, J. Polym. Rev. 2010 50 420. (23) Warman, J. M.; Haas, M. P. De; Dicker, G.; Grozema, F. C.; Piris, J.; Debije, M. G. Chem. Mater. 2004 16 4600. (24) Duran, H.; Hartmann Azanza, B.; Steinhart, M.; Gehrig, D.; Laquai, F.; Feng, X.; Mll en, K.; Butt, H. J.; Floudas, G. ACS Nano 2012 6 9359. (25) Hasegawa, T.; Takeya, J. Sci. Tech. Adv. Mater. 2009 10 024314. (26) Zang, L.; Che, Y.; Moore, J. S. Acc. Chem. Res. 2008 41 1596. (27) Su, B.; Wu, Y.; Jiang, L. Chem. Soc. Rev. 2012 41 7832. (28) Sontz, P. A.; Mui, T. P.; Fuss, J. O.; Tainer, J. A.; Barton, J. K. Proceedings of the National Academy of Sciences 2012 109 1856. (29) Toksoz, S.; Acar, H.; Guler, M. O. Soft Matter 2010 6 5839. (30) Gutzler, R.; Fu, C.; Dadvand, A.; Hua, Y.; MacLeod, J. M.; Rosei, F.; Perepichka, D. F. Nanoscale 2012 4 5965. (31) Benito Hernndez, A.; Pandey, U. K.; Cavero, E.; Termine, R.; Garcia Frutos, E. M.; Serrano, J. L.; Golemme, A.; Gomez Lor, B. Chem. Mate r. 2013 25 117. (32) Anslyn, E. V.; Dougherty, D. A. Modern Physical Organic Chemistry ; University Science Books: Sausalito, 2006. (33) Horiuchi, S.; Hasegawa, T.; Tokura, Y. J. Phys. Soc. Jpn. 2006 75 051016. (34) McGlynn, S. P. Chem. Rev. 1958 58 1113. (35) Mulliken, R. S. J. Phys. Chem. 1952 56 801. (36) Wang, C.; Dong, H.; Hu, W.; Liu, Y.; Zhu, D. Chem. Rev. 2012 112 2208. (37) Walker, B.; Kim, C.; Nguyen, T. Q. Chem. Mater. 2011 23 470. (38) Blight, B. A.; Hunter, C. A.; Leigh, D. A.; McNab, H.; Thomson, P. I. T. Nature Chemistry 2011 3 244. (39) Gearba, R. I.; Lehmann, M.; Levin, J.; Ivanov, D. a.; Koch, M. H. J.; Barber, J.; Debije, M. G.; Piris, J.; Geerts, Y. H. Adv. Mater. 2003 15 1 614.
275 (40) Cantekin, S.; de Greef, T. F. A ; Palmans, A. R. A. Chem. Soc. Rev. 2012 41 6125. (41) Paraschiv, I.; Giesbers, M.; van Lagen, B.; Grozema, F. C.; Abellon, R. D.; Siebbeles, L. D. A.; Marcelis, A. T. M.; Zuilhof, H.; Sudholter, E. J. R. Chem. Mater. 2006 18 968. (42) Fenniri, H.; Mathivanan, P.; Vidale, K. L.; Sherman, D. M.; Hallenga, K.; Wood, K. V; Stowell, J. G. J. Am. Chem. Soc. 2001 123 3854. (43) Yagai, S.; Usui, M.; Seki, T.; Murayama, H.; Kikkawa, Y.; Uemura, S.; Karatsu, T.; Kit amura, A.; Asano, A.; Seki, S. J. Am. Chem. Soc. 2012 134 7983. (44) Sugiyasu, K.; Kawano, S.; Fujita, N.; Shinkai, S. Chem. Mater. 2008 20 2863. (45) Kaafarani, B. R. Chem. Mater. 2011 23 378. (46) Bisoyi, H. K.; Kumar, S. Chem. Soc. Rev. 2010 39 264. (47) Kumar, S. Chem. Soc. Rev. 2006 35 83. (48) Sergeyev, S.; Pisula, W.; Geerts, Y. H. Chem. Soc. Rev. 2007 36 1902. (49) Chandrasekhar, S.; Ranganath, G. S. Rep. Prog. Phys. 1990 53 57. (50) Yagai, S.; Goto, Y.; Lin, X.; Karatsu, T.; K itamura, A.; Kuzuhara, D.; Yamada, H.; Kikkawa, Y.; Saeki, A.; Seki, S. Angew. Chem. Int. Ed. 2012 51 6643. (51) Ryu, M. H.; Choi, J. W.; Kim, H. J.; Park, N.; Cho, B. K. Angew. Chem. Int. Ed. 2011 50 5737. (52) Demenev, A.; Eichhorn, S. H.; Taerum, T.; Perepichka, D. F.; Patwardhan, S.; Grozema, F. C.; Siebbeles, L. D. A.; Klenkler, R. Chem. Mater. 2010 22 1420. (53) Miao, J.; Zhu, L. Chem. Mater. 2010 22 197. (54) Ringsdorf, H.; Wustefeld, R.; Zerta, E.; Ebert, M.; Wendorff, J. H. Angew. Chem. Int. Ed. 1989 28 914. (55) Haverkate, L.; Zbiri, M. J. Phys. Chem. B 2012 116 13098. (56) Savage, R. C.; Mativetsky, J. M.; Orgiu, E.; Palma, M.; Gbabode, G.; Geerts, Y. H.; Samor, P. J. Mater. Chem. 2011 21 206.
276 (57) Vla J.; White, M. S.; Monkowius, U.; Romanazzi, G.; Suranna, G. P.; Mastrorilli, P.; Sekitani, T.; Bauer, S.; Someya, T.; Torsi, L.; Sariciftci, N. S. Adv. Mater. 2013 25 1563. (58) Bassani, D. M.; Jonusauskaite, L.; Lavie Cambot, A.; McClenaghan, N. D.; Pozzo, J. L.; Ray, D.; Vives, G. Coord. Chem. Rev. 2010 254 2429. (59) Hsu, S.; Chen, C.; Cheng, Y.; Wei, K. J. Polym. Sci. Part A: Polym. Chem. 2011 49 603. (60) Yao, K.; Chen, L.; Li, F. J. Phys. Chem. C 2011 116 714. (61) Lin, Y.; Lim, J. A.; Wei, Q.; Mannsfeld, S. C. B.; Briseno, A. L.; Watkins, J. J. Chem. Mater. 2012 24 622. (62) Patra, D.; Ramesh, M.; Sahu, D.; Padhy, H.; Chu, C. W.; Wei, K. H.; Lin, H. C. J. Polym. Sci. Part A: Polym. Chem. 2012 50 967. (63) Patra, D.; Ramesh, M.; Sahu, D.; Padhy, H.; Chu, C. W.; Wei, K. H.; Lin, H. C. Polymer 2012 53 1219. (64) Steinberger, S.; Mishra, A.; Reinold, E.; Levichkov, J.; Uhrich, C.; Pfeiffer, M.; Buerle, P. Che. Comm. 2011 47 1982. (65) Yassin A.; Leriche, P.; Allain, M.; Roncali, J. New. J. Chem. 2013 37 502. (66) Li, Y.; Lampkins, A. J.; Baker, M. B.; Sumpter, B. G.; Huang, J.; Abboud, K. A.; Castellano, R. K. Org. Lett. 2009 11 4314. (67) Destrade, C.; Tinh, N. H.; Gasparoux, H.; Maml ok, L. Liquid Crystals 1987 2 229. (68) Li, Y. Functional Synthetic Architectures Derived from Phloroglucinol, University of Florida, 2010, pp. 1. (69) Park, L. Y.; Hamilton, D. G.; McGehee, E. A.; McMenimen, K. A. J. Am. Chem. Soc. 2003 125 10586. (70) Leight, K. R.; Esarey, B. E.; Murray, A. E.; Reczek, J. J. Chem. Mater. 2012 24 3318. (71) Li, H.; Homan, E. A.; Lampkins, A. J.; Ghiviriga, I.; Castellano, R. K. Org. Lett. 2005 7 443. (72) Rose, K. G.; Jaber, D. A.; Gondo, C. A.; Hamilton, D. G. J. Org. Chem. 2008 3950.
277 (73) McKim, W. D.; Ray, J.; Arnold, B. R. J. Molec. Struct. 2013 1033 131. (74) Das, A.; Molla, M. R.; Maity, B.; Koley, D.; Ghosh, S. Chemistry (Weinheim an der Bergstr asse, Germany) 2012 18 9849. (75) Alvey, P. M.; Reczek, J. J.; Lynch, V.; Iverson, B. L. J. Org. Chem. 2010 7682. (76) Babu, P.; Sangeetha, N. M.; Vijaykumar, P.; Maitra, U.; Rissanen, K.; Raju, a R. Chem. Eur. J. 2003 9 1922. (77) Abu Eittah, R.; Al Sugeir, F. Canad. J. Chem. 1976 54 3705. (78) Kuroda, H.; Amano, T.; Ikemoto, I.; Akamatu, H. J. Am. Chem. Soc. 1967 89 6056. (79) Wang, P.; Anderko, A. Fluid Phase Equilibria 2001 186 103. (80) Hossan, A. S. M.; Abou Melha, H. M.; Refat, M. S. Spectrochim Acta, Part A 2011 79 583. (81) Hariharan, M.; Karunakaran, S. C.; Ramaiah, D. Org. Lett. 2007 9 417. (82) Alghanmi, R. M.; Al Attas, A. S.; Habeeb, M. M. J. Molec. Struct. 2013 1034 223. (83) Warmerdam, T. W.; Nolte, R. J. M.; Drenth, W.; Van Miltenburg, J. C.; Frenkel, D.; Zijlstra, R. J. J. Liquid Crystals 1988 3 1087. (84) Heliatek GmbH Heliatek www.heliatek.com (accessed May 31, 2013). (85) Best Research Cell Efficiencies http://www.nre l.gov/ncpv/images/efficiency_chart.jpg (accessed Jul 22, 2013). (86) Dennler, G.; Scharber, M. C.; Brabec, C. J. Adv. Mater. 2009 21 1323. (87) Kang, M.; Park, H. J.; Ahn, S. H.; Xu, T.; Guo, L. J. IEEE J. Sel. Top. Quantum Electron. 2010 16 1807. (8 8) Pandey, R.; Holmes, R. J. Appl. Phys. Lett. 2012 100 083303. (89) Pivrikas, A.; Neugebauer, H.; Sariciftci, N. S. Solar Energy 2011 85 1226. (90) Surez, M.; Lehn, J. M.; Zimmerman, S. C.; Skoulios, A.; Heinrich, B. J. Am. Chem. Soc. 1998 120 9526. (91) Morzyk Ociepa, B. J. Molec. Struct. 2007 846 74.
278 (92) Mourran, A.; Ziener, U.; Mller, M.; Suarez, M.; Lehn, J. M. Langmuir 2006 22 7579. (93) Maya, E. M.; Vzquez, P.; Torres, T. Chem. Eur. J. 1999 5 2004. (94) Leznoff, C. C.; Terekho v, D. S.; McArthur, C. R.; Vigh, S.; Li, J. Canad. J. Chem. 1995 73 435. (95) Garca Frutos, E. M.; de la Torre, G.; Vzquez, P.; Shirk, J. S.; Torres, T. Eur J. Org. Chem. 2009 2009 3212. (96) Ali, H.; Lier, J. E. Van Tetrahedron Lett. 2009 50 33 7. (97) Bottari, G.; de la Torre, G.; Guldi, D. M.; Torres, T. Chem. Rev. 2010 110 6768. (98) Seitz, W.; Jimnez, A. J.; Carbonell, E.; Grimm, B.; Rodrguez Morgade, M. S.; Guldi, D. M.; Torres, T. Chem. Comm. 2010 46 127. (99) Zhao, W.; Mudrick, J. P.; Zheng, Y.; Hammond, W. T.; Yang, Y.; Xue, J. Org. Electron. 2012 13 129. (100) Zheng, Y.; Bekele, R.; Ouyang, J.; Xue, J.; Paper, I. IEEE J. Sel. Top. Quantum Electron. 2010 16 1544. (101) Ma, C. Q.; Mena Osteritz, E.; Debae rdemaeker, T.; Wienk, M. M.; Janssen, R. a J.; Buerle, P. Angew. Chem. Int. Ed. 2007 46 1679. (102) Ali, H.; Ait Mohand, S.; Gosselin, S.; van Lier, J. E.; Gurin, B. J. Org. Chem. 2011 76 1887. (103) Gonzlez Rodrguez, D.; Claessens, C. G.; Torres T.; Liu, S.; Echegoyen, L.; Vila, N.; Nonell, S. Chem. Eur. J. 2005 11 3881. (104) Kudrevich, S. V. J. Chem. Soc. Perkin Trans 1 1994 2767. (105) Zanotti, G.; Angelini, N.; Paoletti, A. M.; Pennesi, G.; Rossi, G.; Bonapasta, A. A.; Mattioli, G.; Di Carlo, A.; Brown, T. M.; Lembo, A.; Reale, A. J. Chem. Soc. Dalton Trans. 2011 40 38. (106) Guha, S. K.; Obora, Y.; Ishihara, D.; Matsubara, H.; Ryu, I.; Ishii, Y. Adv. Synth. Cat. 2008 350 1323. (107) Han, J.; Jose, J.; Mei, E.; Burgess, K. Angew. Chem. Int. Ed. 2007 46 1684. (108) Aly, A. A. Org. Biomolc. Chem. 2003 1 756.
279 (109) Potratz, S.; Mishra, A.; Buerle, P. Beilstein J. Org. Chem. 2012 8 683. (110) Fitzner, R.; Reinold, E.; Mishra, A.; Mena Osteritz, E.; Ziehlke, H.; Krner, C.; Le o, K.; Riede, M.; Weil, M.; Tsaryova, O.; Wei, A.; Uhrich, C.; Pfeiffer, M.; Buerle, P. Adv. Funct. Mater. 2011 21 897. (111) Grell, M.; Bradley, D. D. C.; Ungar, G.; Hill, J.; Whitehead, K. S. Macromolecules 1999 32 5810. (112) Steinberger, S.; Mi shra, A.; Reinold, E.; Levichkov, J.; Uhrich, C.; Pfeiffer, M.; Buerle, P. Che m Comm. 2011 47 1982. (113) Friebolin, H. Basic One and Two Dimensional NMR Spectroscopy ; 5th ed.; Wiley VCH: Weinheim, 2010. (114) Xue, J.; Rand, B. P.; Uchida, S.; Forrest, S. R. Adv. Mater. 2005 17 66. (115) Rand, B. P.; Xue, J.; Uchida, S.; Forrest, S. R. J. Appl. Phys. 2005 98 124902. (116) Frisch, M. J. T.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Mo ntgomery, Jr., J. A.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J .; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W. ; Gonzalez, C.; Pople, J. A. Gaussian 03, Revision E.01 2004 (117) Allouche, A. R. J. Comp. Chem. 2011 32 174. (118) Sleven, J.; Cardinaels, T.; Grller walrand, C.; Binnemans, K. ARKIVOC 2003 2003 68. (119) Gonidec, M.; Luis, F.; Vlchez, A.; Esqu ena, J.; Amabilino, D. B.; Veciana, J. Angew. Chem. Int. Ed. 2010 49 1623. (120) Williams, F. J.; Relles, H. M.; Manello, J. S.; Donahue, P. E. J. Org. Chem. 1977 43 3419.
280 (121) Yanaihara, N.; Kato, I.; Nagasawa, S.; Kodaira, T. Immunoassay for Phthalic Acid Esters. 6,399,318, 2002. (122) Bavin, P. M. G. Org. Synth. 1960 40 5. (123) Krasnokutskaya, E.; Semenischeva, N.; Filimonov, V.; Knochel, P. Synthesis 2007 2007 81. (124) Jadhav, R.; Sharma, R. WO Patent 2,011,080,718 2011 389. (125) Cohen, Y.; Avram, L.; Frish, L. Angew. Chem. Int. Ed. 2005 44 520. (126) Pons, M.; Millet, O. Prog. Nucl. Magn. Reson. Spectrosc. 2001 38 267. (127) Macchioni, A.; Ciancaleoni, G.; Zuccaccia, C.; Zuccaccia, D. Chem. Soc. Rev. 2008 37 479. (128) E dward, J. T. J. Chem. Ed. 1970 47 261. (129) Morris, K. F.; Jr., Johnson, C. S. J. Am. Chem. Soc. 1992 114 3139. (130) Allouche, L.; Marquis, A.; Lehn, J. M. Chem. Eur. J. 2006 12 7520. (131) Consoli, G. M. L.; Granata, G.; Garozzo, D.; Mecca, T.; Geraci, C. Tetrahedron Lett. 2007 48 7974. (132) J. Molec. Struct. 2012 1013 26. (133) Pellizzaro, M. L.; Fisher, J .; Wilson, A. J. RSC Advances 2013 3 3103. (134) Wittenberg, J. B.; Zavalij, P. Y.; Isaacs, L. Angew. Chem. Int. Ed. 2013 52 3690. (135) Heisel, K. A Am. J. Anal. Chem. 2012 03 401. (136) Avram, L.; Cohen, Y. J. Org. Chem. 2002 67 2639. (137) Mayzel, O.; Cohen, Y. Chem. Comm. 1994 1901. (138) Eisenberg, D.; Quimby, J. M.; Scott, L. T.; Shenhar, R. J. Phys. Org. Chem. 2013 26 124. (139) Kotz, I. a; Gerber, W. J.; Wu, Y. S.; Koch, K. R. J. Chem. Soc. Dalton Trans. 2013 42 3791.
281 (140) Xia, B.; Zheng, B.; Han, C.; Dong, S.; Zhang, M.; Hu, B.; Yu, Y.; Huang, F. Polym. Chem. 2013 4 2019. (141) Nilsson, M.; Morris, G. A J. Magn. Reson. 2005 177 203. (142) Jerschow, A.; Muller, N. J. Magn. Reson. 1997 125 372. (143) Fick, E. Indust rial Solvents Handbook ; Noyes Data Corp: New Jersey, 1998. (144) Santos, F. J. V. J. Phys. Chem. Ref. Data 2006 35 1. (145) Hansen, S. J. Chem. Phys. 2004 121 9111. (146) Walker, B.; Tamayo, A. B.; Dang, X. D.; Zalar, P.; Seo, J. H.; Garcia, A.; Tantiwiwat, M.; Nguyen, T. Q. Adv. Funct. Mater. 2009 19 3063. (147) Liu, X.; Sun, Y.; Perez, L. a; Wen, W.; Toney, M. F.; Heeger, A. J.; Bazan, G. C. J. Am. Chem. Soc. 2012 134 20609. (148) Jang, S. G.; Audus, D. J.; Klinger, D.; Krogstad, D. V; Kim, B. J.; Cameron, A.; Kim, S. W.; Delaney, K. T.; Hur, S. M.; Killops, K. L.; Fredrickson, G. H.; Kramer, E. J.; Hawker, C. J. J. Am. Chem. Soc. 2013 DOI: 10.1021/ja4019447 (149) Beaujuge, P. M.; Tsao, H. N.; Hansen, M. R.; Amb, C. M.; Risko, C.; Subbiah, J.; Choudhury, K. R.; Mavrinskiy, A.; Pisula, W.; Brdas, J. L.; So, F.; Mllen, K.; Reynolds, J. R. J. Am. Chem. Soc. 2012 134 8944. (150) Marsella, M. J.; Piao, G.; Tham, F. S. Synthesis 2002 9 1133. (151) Saini, G.; Lucas, N. T.; Jacob, J. Tetrahedron Lett. 2010 51 2956.
282 BIOGRAPHICAL SKETCH Captain Benjamin M. Schulze (USAF) was born in the land of cheese, Milwaukee, WI to where he still is fiercely loyal, especially in regard to his favorite football team, the Green Bay Packers. After graduating from the US Air Force Academy in 2007 with a record as the tallest chemistry major ever, Capt Schulze began his scientific career studying experimental jet fuel a t Wright Patterson Air Force Base, Ohio. After two years of nagging, the Air Force released Capt Schulze to obtain a PhD in chemistry at the University of Florida beginning in 2009. After four years under the guidance of Prof Ron Castellano, five fitne ss tests, and two club volleyball national championship losses, Capt Patterson in 2013 to join a Research and Development unit, and following on to realize one of his long term goals to mold the minds of the youth of America by becoming an instructor at the US Air Force Academy. (Disclaimer: the views expressed in this dissertation are those of the author and do not reflect the official policy or position of the United States Air Force, Department of Defense, or the U.S. Government.)