We have investigated a wide series of benzo-fused thiophene compounds for ... photophysical characterization was carried out on the series of benzo-fused.
PHOTOCHEMICAL, PHOTOPHYSICAL, AND ELECTRONIC PROPERTIES OF FUSED RING SYSTEMS WITH ALTERNATING BENZENE AND THIOPHENE UNITS
Brigitte Wex
A Dissertation Submitted to the Graduate College of Bowling Green State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY December 2005 Committee: Douglas C. Neckers, Advisor Lewis P. Fulcher Graduate Faculty Representative Michael A. J. Rodgers Thomas H. Kinstle
© 2005 Brigitte Wex All Rights Reserved
iii
ABSTRACT
Douglas C. Neckers, Advisor
We have synthesized a series of polycyclic benzo-fused thiophene compounds. Specifically, we have achieved the isomer-pure synthesis of two new pentacyclic representatives of TBBT that is thieno[2,3-f:5,4-f’]bis[1]benzothiophene (anti) and thieno[3,2-f:4,5f’]bis[1]benzothiophene (syn). Using benzo[1,2-b:4,5-b’]dithiophene, we have shown a successful photoaddition to dimethyl acetylenedicarboxylate and have discovered a new photoaddition between benzo[1,2b:4,5-b’]dithiophene and a series of 1,4-diarylbutadiynes. This photoreaction led to acetylenesubstituted cyclobutene derivatives in regioselective manner. When embedded into a field-effect transistor, the new TBBT materials exhibited p-type semiconducting properties. X-ray diffraction studies revealed a favorable edge-on packing of both compounds on the substrate. We observed competitive field-effect mobilities and on/off current ratios, yet a distinct difference among the isomers. We have investigated a wide series of benzo-fused thiophene compounds for both, holevibrational and electron-vibrational couplings. Specifically, we compared acenes, acenedithiophenes, and phenacenes to both TBBT isomers. Photoelectron measurements and density functional theory calculations showed an electronic structure of the radical-cation state of the TBBT isomers very different from that in anthradithiophene and pentacene and a close resemblance to the corresponding electronic structure of pentaphene. The calculations further showed that the nature of the hole-vibrational interactions in TBBT was very different from
iv
those found in acenedithiophenes and oligoacenes, as a significant coupling in TBBT between holes and low-energy vibrational modes was found. A complete photophysical characterization was carried out on the series of benzo-fused thiophenes. We found that the fusion of one (two) thiophene(s) to benzothiophene (dibenzothiophene) caused a redshift in absorption and emission. The phosphorescence lifetimes and quantum yields were larger for the syn series of compounds compared to the anti series, yet no comparable trend could be found for fluorescence lifetimes. In fact, thieno[2,3-f:5,4f’]bis[1]benzothiophene showed the longest fluorescence and triplet lifetime in the entire series with 4.6 nanoseconds and 128 microseconds, respectively. Using time-resolved absorption spectroscopy, we characterized the lowest energy singlet and triplet excited state in the series. Finally, low temperature UV-vis absorption spectroscopy was combined with quantum chemical computations to demonstrate a reorganization of the molecular orbitals in two isomers of benzodithiophene.
v
This dissertation is dedicated to my parents, Albert and Gerda Wex to my sisters Angelika, Claudia and Martina to my friends to Bilal R. Kaafarani for their devotion, faith and support throughout this process.
vi
ACKNOWLEDGMENTS I would like to voice my sincere gratitude to Dr. Douglas C. Neckers, for his guidance throughout my work leading up to this disseration. Thank you for opening my eyes to research and for letting me enjoy the process of becoming a researcher. Thank you for the opportunity to participate in grant writing, the quantum yield review, as well as for your support of my work at local, national, and international conferences. My appreciation is extended to Dr. Thomas H. Kinstle, Dr. Michael A. J. Rodgers, and Dr. Lewis P. Fulcher for their service in my dissertation committee. Furthermore, I am indebted to the McMaster Endowment, the International Society for Optical Engineering, the Department of Chemistry, Graduate Student Senate, and the Professional Development Fund of the Graduate College at Bowling Green State University for their financial support. I would like to thank all of my collaborators for taking the journey of scientific discovery with me. My gratitude to Dr. Veaceslav Coropceanu, Dr. Ohyun Kwon and Dr. J.-L. Brédas (Georgia Institute of Technology, Atlanta, GA) for theoretical calculations. A special thanks to Dr. Raoul Schröder, Dr. Leszek A. Majewski and Dr. Martin Grell for making my OFET devices a reality. Thank you Dr. Jeanette A. Krause Bauer (U. Cincinnati, OH) and Dr. Kristin Kirschbaum (U. Toledo, OH) for all your efforts in X-ray crystal structure analysis and Dr. Pannee Burckel for PXRD and thin-film XRD analysis. My appreciation to Dr. Eugene Danilov for his support in the theoretical and experimental intricacies using the laser systems at the Ohio Laboratory for Kinetic Spectrometry at BGSU. I also wish to acknowledge the members of Spectra Group Limited, Dr. Oleg Grinevich for assistance with AFM and Dr. John Malpert for guidance in difficult synthetic procedures. Many thanks to the members of Dr. Neckers' research group.
vii
I wish to acknowledge the knowledgeable and skilled help of the wonderful BGSU Ogg Science Library staff, thanks to Mary Keil and Ed Weilant in particular. Thank you Melinda Horst (Summer REU) for synthesizing a whole battery of compounds in the few weeks working with me in 2004. My deepest gratitude to Larry Ahl for the precision and efficiency in making all these little devices that made my life in the lab so much easier, to Doug Martin for getting my stirrers back in working order, and Craig Bedra for keeping my computer(s) healthy and free of viruses, horses, and the like. My thanks to Alita Frater, Nora Cassidy, Midge Wittmer, and Lisa Rood for their help with the everyday student issues. I would like to thank my parents and three sisters for their love and support during these years of higher education. Thank you your words of wisdom and your help in realizing this dissertation. Last but not least, I would like to thank my soul mate, best friend, collaborator, and supporter Dr. Bilal R. Kaafarani (American University of Beirut) for continuous encouragement and support. Thank you for all your help, inspiration, and most importantly thank you for the trust you placed in my abilities.
viii TABLE OF CONTENTS Page CHAPTER I. BACKGROUND........................................................................................
1
1.1 GENERAL INTRODUCTION TO THE UNIQUE MOLECULE THIOPHENE ....................
1
1.2 TOPIC OF THIS DISSERTATION RESEARCH AND DESCRIPTION OF CHAPTERS ........
5
1.3 REFERENCES ....................................................................................................
6
CHAPTER II. SYNTHESIS OF EMBEDDED S-CONTAINING HETEROCYCLES .....
7
2.1 INTRODUCTION .................................................................................................
7
2.1.1 RING-CLOSURE AT THE THIOPHENE RING ...........................................
8
2.1.1 RING-CLOSURE AT THE ARENE MOIETY ..............................................
9
2.2 SCOPE OF OUR RESEARCH AND ORGANIZATION OF CHAPTER .............................
10
2.3 SYNTHESIS OF TARGET COMPOUND 2.6 .............................................................
11
2.4 SYNTHESIS OF TARGET COMPOUND 2.5 .............................................................
13
2.5 ISOMERIC PURITY .............................................................................................
16
2.6 GENERAL EXPERIMENTAL DETAILS ...................................................................
17
2.6.1 SYNTHESIS OF BENZO[1,2-B:4,5-B’]DITHIOPHENE (2.3) .......................
18
2.6.2 SYNTHESIS OF BENZO[1,2-B:5,4-B’]DITHIOPHENE (2.4) .......................
20
2.6.3 SYNTHESIS OF THIENO[2,3-F:5,4-F’]BIS[1]BENZOTHIOPHENE (2.5) ......
22
2.6.4 SYNTHESIS OF THIENO[3,2-F:4,5-F’]BIS[1]BENZOTHIOPHENE (2.6) ......
24
2.7 REFERENCES ....................................................................................................
26
CHAPTER III. PHOTOADDITION REACTIONS OF ACETYLENE AND BUTADIYNE DERIVATIVES TO BENZODITHIOPHENE ..................................................................
30
3.1 INTRODUCTION .................................................................................................
30
ix 3.2 PHOTOCYCLOADDITION OF DIMETHYL ACETYLENEDICARBOXYLATE TO BENZODITHIOPHENES .............................................................................
32
3.3 PHOTOCYCLOADDITION OF BUTADIYNES TO BENZODITHIOPHENES.....................
35
3.4 CONCLUSIONS ..................................................................................................
40
3.5 X-ray Crystallography of 3.734,35 ......................................................................
41
3.6 EXPERIMENTAL SECTION .................................................................................
48
3.7 REFERENCES ....................................................................................................
53
CHAPTER IV. PREPARATION OF ORGANIC FIELD-EFFECT TRANSISTORS (OFET) USING THIENO[F,F’]BIS[1]BENZOTHIOPHENE (SYN, ANTI) ......................
57
4.1 INTRODUCTION .................................................................................................
57
4.2 DESIGN STRATEGIES FOR NEW MATERIALS .......................................................
59
4.3 X-RAY CRYSTALLOGRAPHY OF 2.5 AND 2.6.......................................................
61
4.4 THIN-FILM X-RAY DIFFRACTION ANALYSIS OF MATERIALS 2.5 AND 2.6 ............
63
4.5 CYCLIC VOLTAMMETRY ...................................................................................
66
4.6 THERMOGRAVIMETRIC ANALYSIS .....................................................................
69
4.7 GAS-PHASE PHOTOELECTRON SPECTROSCOPY ...................................................
70
4.8 MATERIALS CHARACTERIZATION IN OFET DEVICES .........................................
73
4.8.1 DEVICE CONSTRUCTION .....................................................................
73
4.8.2 TOPOGRAPHICAL IMAGES OF THIN-FILMS ...........................................
74
4.8.3 DEVICE CHARACTERIZATION ..............................................................
74
4.9 CONCLUSIONS ..................................................................................................
79
4.10 EXPERIMENTAL PARAMETERS FOR X-RAY CRYSTAL STRUCTURE ANALYSIS ....
79
4.11 EXPERIMENTAL DETAILS AND DATA TABLES FOR XRD ANALYSIS ..................
83
x 4.12 REFERENCES ..................................................................................................
97
CHAPTER V. VIBRONIC COUPLING IN ORGANIC SEMICONDUCTORS: THE CASE OF FUSED POLYCYCLIC BENZENE-THIOPHENE STRUCTURES ........................... 102 5.1 INTRODUCTION ................................................................................................. 102 5.2 EXPERIMENTAL SECTION .................................................................................. 104 5.3 THEORETICAL METHODOLOGY ......................................................................... 104 5.4 RESULTS AND DISCUSSION ................................................................................ 107 5.4.1 PHOTOELECTRON SPECTROSCOPY ....................................................... 107 5.4.2 FUSED THIOPHENE VERSUS FUSED BENZENE: S VERSUS C=C.............. 107 5.4.3 GEOMETRY ........................................................................................ 110 5.4.4 REORGANIZATION ENERGY ................................................................ 111 5.5 CONCLUSIONS .................................................................................................. 121 5.6 REFERENCES .................................................................................................... 123 CHAPTER VI. PHOTOPHYSICAL INVESTIGATION OF COMPOUNDS WITH ALTERNTATING BENZENE AND THIOPHENE RINGS ............................................. 126 6.1 INTRODUCTION ................................................................................................. 126 6.2 PROPERTIES OF THE SINGLET GROUND AND EXCITED STATE .............................. 127 6.2.1 ABSORPTION SPECTRA........................................................................ 127 6.2.2 EMISSION FROM SINGLET STATE ......................................................... 131 6.2.3 FLUORESCENCE LIFETIMES ................................................................. 132 6.3 FEMTOSECOND TRANSIENT ABSORPTION SPECTROSCOPY .................................. 133 6.4 PROPERTIES OF THE TRIPLET EXCITED STATE .................................................... 139 6.4.1 TRIPLET PROPERTIES AS OBSERVED BY PHOSPHORESCENCE ................ 141
xi 6.5 NANOSECOND TRANSIENT ABSORPTION SPECTROSCOPY ................................... 143 6.6 CONCLUSIONS .................................................................................................. 145 6.7 GENERAL EXPERIMENTAL PARAMETERS ........................................................... 146 6.7.1 ABSORPTION SPECTRA ....................................................................... 146 6.7.2 FLUORESCENCE SPECTRA ................................................................... 146 6.7.3 PHOSPHORESCENCE SPECTRA ............................................................. 147 6.7.4 TIME-CORRELATED SINGLE PHOTON COUNTING ................................. 147 6.7.5 TRANSIENT SPECTROSCOPY ................................................................ 147 6.7.5.1 Femtosecond Transient Absorption Spectroscopy ............... 148 6.7.5.2 Nanosecond Transient Absorption Spectroscopy................. 149 6.8 REFERENCES .................................................................................................... 151 CHAPTER VII. COMBINED THEORETICAL AND SPECTROSCOPIC STUDY OF THE INTERESTING ELECTRONIC STRUCTURAL AND SPECTRAL PROPERTIES OF THE TWO BENZO[B,B’]DITHIOPHENE ISOMERS.............................................................. 154 7.1 INTRODUCTION ................................................................................................. 154 7.2 EXPERIMENTAL ................................................................................................ 154 7.3 UV-VIS ABSORPTION SPECTRA OF 2.3 AND 2.4 AT 77K...................................... 155 7.4 THEORETICAL INVESTIGATIONS AND DISCUSSION OF EXPERIMENTAL DATA ....... 156 7.5 CONCLUSIONS .................................................................................................. 160 7.6 REFERENCES .................................................................................................... 160 APPENDIX I. SUPPORTING MATERIAL FOR CHAPTER V...................................... 163 APPENDIX II. SUPPORTING MATERIAL FOR CHAPTER VI ................................... 177 APPENDIX III. LIST OF SELECTED ABBREVIATIONS AND ACRONYMS ............ 189
xii LIST OF FIGURES Figure
Page
1.1 Chemical structures and bond lengths of thiophene (1.1) and benzene (1.2)...........
1
1.2 Photoelectron spectrum of thiophene .....................................................................
2
1.3 Thiophene-containing contaminants of fossile fuels...............................................
3
1.4 Structures of thiophenes as used in daily lives .......................................................
4
2.1 Condensed ring systems 2.1–2.6 containing benzenes and thiophenes....................
11
2.2 Comparison of 1H NMR of 2.5 (upper) and 2.6 (lower) .........................................
17
3.1 Thiophene to acene conversion..............................................................................
31
3.2 Molecular structure of benzodithiophenes..............................................................
31
3.3
1
H NMR spectrum of photoproduct 3.3 in CDCl3...................................................
33
3.4 Gas chromatogram (upper) and mass spectrum (lower) of photoproduct ................
34
3.5 Proposed pyrolysis product of 3.3..........................................................................
34
3.6
1
H NMR spectrum of irradiated mixture of 2.4 and 3.1 ..........................................
35
3.7 Possible photoreaction products between 2.3 and 3.6.............................................
36
3.8 ORTEP representation of 3.7 at 173K (ellipsoids at 50% probability); minor conformer of disordered phenyl ring omitted for clarity...............................
37
3.9 Effect of concentration on product formation (SM=starting material 2.3, PA=photoadduct 3.7); ratio of SM/PM corresponds to the integrated absorption of an a-proton of 2.3 and the lower-field proton of the cyclobutene ring of 3.7......
38
3.10 Absorption spectra of 2.3, 3.6, and 3.7 in ethanol ..................................................
39
3.11 1H NMR spectrum of dimethylcyclobuta[b]thieno[2,3-f][1]benzothiophene5a,6[7aH]dicarboxylate (3.3) .................................................................................
49
xiii 3.12 APT (upper) and 13C (lower) NMR spectra of dimethyl cyclobuta[b]thieno[2,3-f][1]benzothiophene-5a,6[7aH]dicarboxylate (3.3) ............
50
3.13 1H NMR spectrum of 6-phenyl-7-(phenylethynyl)-5a,7adihydrocyclobuta[b]thieno[2,3-f][1]benzothiophene (3.7)...................................... 3.14
13
51
C NMR spectrum of 6-phenyl-7-(phenylethynyl)-5a,7a-
dihydrocyclobuta[b]thieno[2,3-f][1]benzothiophene (3.7)......................................
52
4.1 Lilienfeld's device for controlling electric current ..................................................
57
4.2 Schematic overview of a top-contact (left) and bottom-contact (right) design for OFET ..........................................................
58
4.3 Molecular structure of new semiconductors...........................................................
60
4.4 Solid-state packing of 2.5 as viewed along the c-axis. All hydrogens are omitted ..
61
4.5 Crystal packing as observed along the a-axis. All hydrogens are omitted...............
62
4.6 Molecular structure of 4.6 and 4.7 .........................................................................
62
4.7 Edge-to-face packing as observed for the solid-state of 2.6....................................
63
4.8 X-ray diffraction pattern of 2.5 (1) and 2.6 (2)... ....................................................
65
4.9 Model of the alignment of the molecules 2.5 along the device plane = bc molecular plane... ..................................................................................................
65
4.10 Cyclic voltammogram of 2.5 .................................................................................
67
4.11 Differential pulse voltammogram of 2.5 ................................................................
67
4.12 Cyclic voltammogram of 2.6 .................................................................................
68
4.13 Differential pulse voltammogram of 2.6 ................................................................
68
4.14 Thermogravimetric analysis curve of 2.5 ...............................................................
69
4.15 Thermogravimetric analysis curve of 2.6 ...............................................................
70
xiv 4.16 Molecular structure for anthradithiophene (4.8) .....................................................
71
4.17 Photoelectron spectra of first ionization of 2.5 (bottom) and 2.6 (top)....................
72
4.18 AFM topographical images of thin-films of 2.5 (left) and 2.6 (right) on p-Si/SiO2/OTS evaporated at Tsub = 75 °C..............................................................
74
4.19 Output characteristic of compound 2.5 on a 100 nm (left) and 2.6 on a 300 nm (right) SiO2 gate insulator......................................................
75
4.20 Transfer characteristics of compound 2.5 on a 100 nm (left) and 2.6 on a 300 nm (right) SiO2 gate insulator......................................................
76
4.21 Mobility plot of 2.5 (left) and 2.6 (right)................................................................
77
4.22 Ortep representation of compound 2.5 ...................................................................
81
5.1 Molecular structures of the compounds investigated in this work: benzo[1,2-b:4,5b’]dithiophene (2.3), benzo[1,2-b:5,4-b’]dithiophene (2.4), thieno[2,3-f:5,4f’]bis[1]benzothiophene (2.5), thieno[3,2-f:4,5-f’]bis[1]benzothiophene (2.6), naphtho[2,3-b:6,7-b’]dithiophene (5.1), naphtho[2,3-b:7,6-b’]dithiophene (5.2), anthra[2,3-b:7,8-b’]dithiophene (4.8), anthra[2,3-b:8,7-b’]dithiophene (5.3), pentaphene (5.4), phenanthro[3,2-b:6,7-b’]dithiophene (5.5), phenanthro[2,3-b:7,6-b’]dithiophene (5.6), dinaphtho[2,3-b:2’,3’-d] thiophene (5.7) ...................................................................................................... 103 5.2 Sketch of the potential energy surfaces related to electron (hole) transfer, showing the related transitions, the normal mode displacement DQ, and the relaxation energies lr................................................................................. 106 5.3 Gas-phase photoelectron spectra of the BDT isomers 2.3 (top), 2.4 (middle) and anthracene (bottom) ........................................................................................ 109
xv 5.4 Gas-phase photoelectron spectra of the TBBT isomers (left), ADT (right, top), and pentacene (right, bottom) ................................................................................ 110 5.5 Huang-Rhys factors related to the electron and hole transfer, derived from the B3LYP/6-31G** calculations of the relaxation energy, l(r A1) ........................... 116 5.6 High-resolution close-up and B3LYP/6-31G** simulation of the vibrational structure of the first ionization of BDT isomers ..................................................... 117 5.7 Electron- and hole-vibrational couplings in TBBT isomers derived from the B3LYP/6-31G** calculations of l(rA1) .............................................................. 120 5.8 Illustration of the bonding/antibonding pattern of the HOMO and LUMO orbitals of TBBT isomers and the atomic displacements associated with the vibrational modes yielding the largest Huang-Rhys factors .............................. 121 6.1 Molecular structures of compounds investigated in this work. [1]Benzothiophene (2.1); dibenzothiophene (2.2); benzo[1,2-b:4,5-b’]dithiophene (2.3), benzo[1,2-b:5,4-b’]dithiophene (2.4), thieno[2,3-f:5,4f’]bis[1]benzothiophene (2.5); thieno[3,2-f:4,5-f’]bis[1]benzothiophene (2.6) ........ 127 6.2 Absorption spectra of 2.5 and 2.6 in dichloromethane............................................ 128 6.3 Normalized absorption spectra of 2.1, 2.3, 2.4 (top) and 2.2, 2.5, 2.6 (bottom)....... 129 6.4 Fluorescence emission spectra of 2.1, 2.3 and 2.5 (left) and 2.1, 2.2, 2.4 and 2.6 (right).............................................................................. 132 6.5 Time profile for the decay of fluorescence of 2.5 in solution of methylcyclohexane in presence of oxygen. Depicted are the instrument response function (left) and the native data (–) together with a super-imposed fit (–) and a residual plot (bottom). ........................................................................ 133
xvi 6.6 Fs transient absorption spectra of 2.1 (left) and 2.2 (right) ..................................... 134 6.7 Fs transient absorption spectra of 2.3 (left) and 2.4 (right) ..................................... 135 6.8 Fs transient absorption spectra of 2.5 (left) and 2.6 (right) ..................................... 135 6.9 Decay of singlet state S1 of 2.1 at 580 nm (left) and rise of triplet state T1 at 426 nm (right) ............................................................................................... 136 6.10 Decay of singlet state at 630 nm and rise of T1 at 388 nm for 2.2 ........................... 136 6.11 Decay of singlet state of 2.3 at 400 nm (left) and observed stimulated emission 350 nm (right)......................................................................................... 137 6.12 Rise and decay for 2.3 at 487 nm and 750 nm, respectively ................................... 137 6.13 Decay of 2.4 signals at 755 nm (right) ................................................................... 138 6.14 Decay of singlet state (780 nm, left) and rise of triplet state (513 nm, right) of 2.5. 138 6.15 Decay of singlet state of 2.6 at 785 nm and rise of triplet state as observed at 530 nm ........................................................................................... 139 6.16 Phophorescence spectra of 2.129 (left) and 2.230 (right) in EPA at 77K ................... 141 6.17 Phosphorescence spectra of 2.3 (left) and 2.4 (right) in EPA at 77K ...................... 142 6.18 Phosphorescence spectra of 2.5 (left) and 2.6 (right) in EPA at 77K ...................... 142 6.19 Triplet energy as a function of singlet energies for 2.1-2.6..................................... 143 6.20 Transient absorption spectrum of 2.1 (left) and 2.2 (right) in MCH with lmax=430 nm and 390 nm, respectively........................................................... 144 6.21 Transient absorption spectrum of 2.3 (left, lmax=490 nm) and 2.4 (right) in MCH.. 144 6.22 Transient absorption spectrum of 2.5 (lmax=520 nm) and 2.6 (right) in MCH ......... 145 7.1 Molecular formula of benzodithiophenes (2.3) and (2.4)........................................ 154
xvii 7.2 Low temperature absorption spectra of 2.3 in EPA (ethyl ether, isopentane, ethanol) at 77K. Left: Absorption maxima are 249 nm (40,161 cm-1), 258 nm (38,760 cm-1) and a shoulder at 254 nm (39,730 cm-1). Right: Absorption maxima are 290 nm (34,483 cm-1), 304 nm (32,895 cm-1), 321 nm (31,153 cm-1), 335 nm (29,851 cm-1). Minor peaks are located at 327 nm (30,581 cm-1) and 331 nm (30,211 cm-1) ....................................................................................................... 155 7.3 Low temperature absorption spectra of 2.4 in EPA (ethyl ether, isopentane, ethanol) at 77K. Left: Maxima of absorption are located at 248 nm (40,323 cm-1) and 257 nm (38,911 cm-1). Right: Maxima of absorption are located at 296 nm (33,784 cm-1), 303 nm (33,003 cm-1), 315 nm (31,746 cm-1), 321 nm (31,153 cm-1), 328 nm (30,488 cm-1). ........................................................ 156
xviii LIST OF SCHEMES Scheme
Page
2.1 Electrophilic ring-closure ......................................................................................
8
2.2 Electrophilic cyclizations of terminal acetylenes....................................................
8
2.3 Acid-catalyzed cyclization.....................................................................................
9
2.4 Desulfurization of thianthrene ...............................................................................
9
2.5 Synthesis of anthradithiophene ..............................................................................
10
2.6 Schematic representation of the Bradsher reaction.................................................
10
2.7 Synthesis of a,a'-bis(3-bromo-2-thienyl)-2,5-thiophenedimethanol (2.10) ............
11
2.8 Reduction of dicarbinol (2.10) ...............................................................................
12
2.9 Attempt I for synthesis of 2.5 ................................................................................
14
2.10 Attempt II for synthesis of 2.5 ...............................................................................
14
2.11 Attempt III for synthesis of 2.5..............................................................................
15
2.12 Synthetic scheme for preparation of 2.5.................................................................
16
2.13 Synthesis of benzo[1,2-b:4,5-b']dithiophene (2.3)..................................................
18
2.14 Synthetic scheme of benzo[1,2-b:5,4-b']dithiophene (2.4)......................................
20
3.1 Photoaddition of DMAD (3.1) to benzo[1,2-b:4,5-b’]dithiophene (2.3) .................
32
3.2 Photoaddition of 3.1 to 2.4.....................................................................................
35
3.3 Photoaddition of butadiynes (3.6, 3.11–3.13) to 2.3 ...............................................
40
4.1 Mechanism of endoperoxide formation..................................................................
59
4.2 Cycloadditions to anthracene (4.1), naphtho(b)thiophene (4.4) and benzothiophene (4.5).......................................................................................
60
4.3 Formula description of photoelectric effect............................................................
71
xix LIST OF TABLES Table
Page
1.1 Atomic contributions to molecular orbitals of thiophene........................................
2
2.1 Optimization of reaction conditions for double-lithiation .......................................
13
3.1 Optimization of reaction conditions.......................................................................
33
3.2 Comparison of 1H NMR data of photoproducts......................................................
40
3.3 Crystal data and structure refinement.....................................................................
42
3.4 Bond lengths [Å] and angles [º] .............................................................................
43
3.5 Selected torsion angles [º]......................................................................................
46
3.6 Least squares planes ..............................................................................................
47
4.1 Redox potentials (vs. Fc) of compounds 2.5 and 2.6 in comparison to 4.7..............
68
4.2 Thermal parameters of compounds 2.5 and 2.6 compared to pentacene (4.7).. .......
70
4.3 Vertical adiabatic ionization energy (IP)................................................................
72
4.4 Performance of 2.5 and 2.6 in a top-contact OFET device .....................................
76
4.5 Crystal data and structure refinement for 2.559 .......................................................
80
4.6 Bond lengths [Å] and angles [º] for 2.5..................................................................
81
4.7 Experimental parameters for thin-film XRD acquisition ........................................
84
4.8 Thin-film XRD data and calculated d-spacings of 2.5............................................
86
4.9 Powder pattern simulated from 2.5 single-crystal data (CCDC254385.cif) with Cu K-alpha1 radiation (1.5406 Å)..................................................................
87
4.10 Thin-film XRD data and calculated d-spacings of 2.6............................................
94
4.11 Indexed thin-film diffraction data of 2.6 ................................................................
95
xx 5.1 B3LYP/6-31G** estimates of the reorganization energies l (eV) related to electron transfer (ET) and hole transfer (HT) in acenedithiophenes and oligoacenes ..................................................................................................... 112 5.2 B3LYP/6-31G** estimates of frequency, w (cm-1), Huang-Rhys factors, S, and relaxation energies, lrel (eV), for hole transfer in anti BDT (2.3) ................... 114 5.3 B3LYP/6-31G** estimates of frequency, w (cm-1), Huang-Rhys factors, S, and relaxation energies, lrel (eV), for hole transfer in syn BDT (2.4) ...................... 115 5.4 B3LYP/6-31G** estimates of the reorganization energies l (eV) related to electron transfer (ET) and hole transfer (HT) in pentaphene derivatives............. 118 6.1 Properties of the singlet excited state ..................................................................... 130 6.2 Properties of the triplet excited state ...................................................................... 140 7.1 Relative energies of anti (2.3) and syn (2.4) benzodithiophenes ............................. 156 7.2 HF and DFT optimized geometries........................................................................ 157 7.3 Optimized geometries of ground, 1st and 2nd excited states at the HF/6-31G** level ........................................................................................ 158 7.4 Energies and pictures of frontier molecular orbitals of benzodithiophene isomers 2.3 and 2.4 ................................................................... 159 7.5 Comparison of UV-vis absorption bands in experiment and theory (TDDFT)........ 160
1 CHAPTER I. BACKGROUND
1.1 GENERAL INTRODUCTION TO THE UNIQUE MOLECULE THIOPHENE Thiophene (1.1) is a five-membered, stable aromatic compound, Figure 1.1. When compared to the aromatic compound benzene (1.2), one ethene group (H-C=C-H) is replaced by the chalcogen sulfur. In this arrangement, sulfur has two lone pairs of electrons, one of which is part of the aromatic sextet, while the other one exists as an sp2 hybrid orbital in the plane of the ring. The mesomeric contribution into the ring by this double bond is minimal due to the high electronegativity of the sulfur atom. Overall, a negative inductive effect causes a polarization toward sulfur and results in a dipole moment of 0.52 D. In comparison to carbon, sulfur possesses a larger bonding radius and d orbitals, the bond angle around sulfur are therefore larger in comparison to a carbocycle and the overall stability of thiophene is comparable to benzene. The bond lengths found in thiophene are indicated in Figure 1.1. The interesting reactivities that result out of these properties are explored and applied in Chapter 2 of this dissertation. 1.42 Å H
H
1.37 Å H
S
1.40 Å
H 1.71 Å
1.1
1.2
Figure 1.1. Chemical structures and bond lengths of thiophene (1.1) and benzene (1.2). The molecular orbital composition of thiophenes was extensively studied using computational and experimental techniques. As the symmetry decreases from benzene (D6h) to thiophene (C2v), a splitting of the degenerate e1g π orbitals of benzene at 9.2 eV into 1a2 and 3b1 in thiophene is observed. Using photoelectron spectroscopy, the energy level and atomic
2 contributions of these molecular orbitals (MOs) can be directly determined as depicted in Figure 1.2 and Table 1.1. Therein, the band (A) at 8.87 eV is assigned to the π3 (1a2) MO, band (B) at 9.52 eV is assigned to the π2 (2b1) MO and band (C) at 12.1 eV is assigned to the π1 (1b1) MO. Furthermore, the shape of the peaks in the spectrum allows a quantification of atomic contribution to the MO (Figure 1.2).1 Narrow bands (B) are an indication for MO with a high pp contribution of S. The time-scale of the photoionization is 10-15 to 10-18 s, i.e. only FranckCondon transitions to discrete vibrational levels of the ion state are observed in the photoelectron spectrum (Born-Oppenheimer approximation).
Figure 1.2. Photoelectron spectrum of thiophene. Table 1.1. Atomic contributions to molecular orbitals of thiophene. Molecular Orbital 1a2 (p) 3b1(p) 2b1(p)
Atom
Thiophene
%C %S %C %S %C %S
100 0 43.6 56.4 85.2 14.8
Thiophene occurs naturally in crude oil and was first discovered as a contaminant of coal
3 tar benzene by Victor Meyer in 1882.2 In crude oil, thiophenes are part of the polycyclic aromatic sulfur-containing hydrocarbons (PASH) fraction. This fraction varies substantially with geographic origin and makes up 10 - 30 % of the aromatic fraction of crude oil. The aromatic fraction in turn constitutes ~28.2 % of all crude oils including tar.3,4 As such, sulfur-containing compounds account for a substantial contaminant in fossil fuels. They are responsible for SOx emissions upon combustion (Figure 1.3) and are closely regulated in the United States by the Environmental Protection Agency.5 The commercial production of thiophenes is carried out via the dehydrogenation of butane with sulfur as the dehydrogenating agent, followed by cyclization with sulfur.
Figure 1.3. Thiophene-containing contaminants of fossil fuels.5 Animal biosynthesis does not include aromatic heterocycles, while plants of the family asteraceae can produce aromatic thiophenes. A prominent member of this family is the Marigold. For humans, thiophenes enter everyday live in the form of medications. Examples include Evista (raloxifen), a selective estrogen receptor modulator6 and the drug Plavix (clopidogrel), a
4 platelet aggregation inhibitor.7 Thiophenes are further used as herbicides, such as Outlook, which controls grasses and small-seeded broadleaf weeds.8 The tetrahydro-derivative of thiophene is extensively used as lubricant additives such as 3-(decyloxy)tetrahydrothiophene-1,1-dioxide (Figure 1.4). NH O O O
H
OCH3
OCH3
H3C
N
HO S
Evista
OH
S
O Cl
Plavix
O C10H23
N S
Outlook
O
S
O
lubricant additive
Figure 1.4. Structures of thiophenes as used in daily lives. With the discovery of the fourth generation polymers,9 i.e. polymers with semiconducting and metallic polymers, interest in thiophene took a radical turn. In recent years, polythiophenes (polymers of thiophenes) have received much attention in the context of the combination of electronic and optical properties of semiconductors with processability and mechanical plasticity of polymers.10 Materials such as poly(3,4-ethylenedioxythiophene) have entered the market and are widely used for antistatic, electric, and electronic applications.11 Highly doped, metallic polythiophenes are used in batteries, electrochromic windows, electromagnetic shields, and antistatic coatings. Semiconducting thiophenes have thereby found applications in optoelectronic devices, such as organic field-effect transistors (OFET), organic light-emitting diodes (OLED), photodetectors, and photovoltaic cells.10 Both, the high degree of freedom among the thiophene units in the polymer, and the relative disorder of the polymeric materials add significant complexity to the study of the observed electronic processes. A strategy to remove the degrees of freedom between the
5 thiophene rings is to embed the rings into a framework with aromatic (i.e. benzene) rings. Such rigid oligomers can the serve as model compounds for polymeric materials as they posses well characterized structure, order and purity.
1.2 TOPIC OF THIS DISSERTATION RESEARCH AND DESCRIPTION OF CHAPTERS The strategies described in 1.1 have been applied in this dissertation research. We have designed, synthesized, and characterized (both theoretically and experimentally) well-defined, isomer-pure embedded thiophenes. Embedded thereby means that the thiophene rings are fused into an aromatic framework. These materials vary in orientation of the thiophene rings along the long axis of the molecule. Chapter 2 portrays the isomer-pure synthesis of these new thienobisbenzothiophene (TBBT) materials. In Chapter 3, we describe the interesting photochemical properties of these materials. Specifically, we explored the photochemical [2p +2p] cycloaddition of dimethyl acetylenedicarboxylate to two isomers of benzo[b:b’]dithiophene. Furthermore, we discovered a new [2p+2p] photocycloaddition reaction in that we applied irradiated a series of butadiynes in presence of two isomers of benzo[b:b’]dithiophene. The semiconducting properties of our new materials are reported in Chapter 4, as the materials are integrated into organic field-effect transistors (OFET). Therein, the field-effect mobility of these two isomers of TBBT was studied. Both isomers exhibited substantial mobilities up to 0.12 cm2/Vs, yet the value varied significantly with isomer structure. Chapter 5 is therefore a combined experimental (high resolution photoelectron spectroscopy) and first-principles quantum-mechanical study of the intramolecular vibrational couplings in these compounds. Interestingly, there is a marked difference between the hole-vibrational couplings for the anti and syn isomers. Chapter 6 is a comprehensive photophysical study of the new
6 materials, where steady-state and time-resolved spectroscopy in nanosecond and femtosecond time regime was applied to characterize the singlet and triplet excited states of the series of new, embedded thiophenes. Finally, Chapter 7 is a short study of the molecular orbital arrangement in two benzodithiophene isomers (anti, syn).
1.3 REFERENCES (1)
Rabalais, I. W.; Werme, L. O.; Bergmark, T.; Karlsson, L.; Siegbahn, K. Intern. J. Mass. Spectrom. Ion Phys. 1972, 9, 185.
(2)
Joule, J. A.; Mills, K. Heterocyclic Chemistry; 4th ed.; Blackwell Science Ltd: Oxford, 2000.
(3)
Andersson, J. T.; Obinger, S.; Traulsen, F. In PittCon New Orleans, 1998.
(4)
Tissot, B. P.; Welte, D. H. Petroleum Formation and Occurrence: A New Approach to Oil and Gas Exploration; Springer-Verlag: New York, 1978.
(5)
Kemsley, J. Chem. Eng. News 2003, 81, 40-41.
(6)
Evista Home Page. http://www.evista.com/index.jsp (accessed October 2004).
(7)
Information about Plavix Home Page. http://www.plavix.com (accessed August 2004)
(8)
U.S. Agricultural Products Home Page. http://www.agproducts.basf.com/Products/Products.asp (accessed August 2004).
(9)
Heeger, A. J. Angew. Chem. Int. Ed. 2001, 40, 2591-2611.
(10)
Handbook of Oligo- and Polythiophenes; Fichou, D., Ed.; Wiley-VCH: Weinheim, 1999.
(11)
Kirchmeyer, S.; Reuter, K. J. Mater. Chem. 2005, 15, 2077-2088.
7 CHAPTER II. SYNTHESIS OF EMBEDDED S-CONTAINING HETEROCYCLES
2.1 INTRODUCTION Benzene, discovered in 1825 by Faraday, and represented by the aromatic sextet as dreamed up by Kékulé in 1865,1 is the lowest molecular weight oligoacene. Upon extension of the number of repeat units, acenes acquire interesting electronic properties and are currently under thorough investigation for applications in OFET and OLED.2 Thiophene is a fivemembered heterocycle that has an analogous aromatic sextet resembling benzene. Therein one ethene unit (H-C=C-H) of the p-system is replaced with a sulfur atom that provides a lone pair of electrons to complete the sextet. Among the heterocycles pyrrole, thiophene and furan, thiophene has the highest degree of aromaticity. Due to their favorable electronic properties, oligo- and polythiophenes are applied in a plethora of opto-electronic devices.3,4 Our research targets are hybrids of the two materials, that is, compounds with alternating thiophene and benzene units. We specifically target a class of compounds, where the b-bonds of the thiophenes are fused with the a- and d-bonds of the phenyl units. In these, the thiophene units are forced into a planar, rigid structure. An isomer-pure synthesis of these materials is challenging and several synthetic strategies for smaller ring systems are outlined in the following paragraphs. In 1954, Hartough and Meisel reviewed the properties of condensed thiophene systems including sources, physical constants, and syntheses such as reaction conditions and percent yields.5 Synthetic approaches to the preparation of materials composed of alternating thiophene and benzene units include ring closure at the thiophene ring (2.1.1) and at the benzenoid unit (2.1.2) as outlined below. A comprehensive review of the dibenzothiophenes as a group was
8 prepared by Andrews.6 The nomenclature of condensed ring systems was reviewed in the 1998 IUPAC report on fused and bridged fused ring systems.7 Condensed thiophenes have been the target of intense research. Groups such as Wynberg, Bechgaard and Yamada applied photochemical cyclization of dithienylethene and applied this procedure to synthesize a wide variety of compounds,8-10 wherein the fusion along the a- and c-bond of the benzene moiety leads to spiraled molecules. With as little as 5 rings, these molecules become optically active and are called heterohelicenes. 2.1.1 RING-CLOSURE AT THE THIOPHENE RING The thiophene ring in benzothiophene can be constructed via intramolecular electrophilic attack of a ketone or acid of thioarene onto an aromatic ring, Scheme 2.1.11,12
Scheme 2.1. Electrophilic ring-closure. O
S O
S
Starting from commercially available o-iodothioanisol, Yue and Larock prepared terminal acetylenes and explored ring fusion between S and C catalyzed by electrophiles such as I2, Br2, NBS, p-O2NC6H4SCl, and PhSeCl. These reactions yield products in round 90%, Scheme 2.2.13
Scheme 2.2. Electrophilic cyclizations of terminal acetylenes. SR
E+
S R
R
E
9
A thiophene ring between two phenyls can be generated by acid-catalyzed coupling of aromatic methyl sulfoxides followed by base-catalyzed demethylation reaction (Scheme 2.3).14 This procedure is only mildly regiospecific and a double-cyclization to create a heptacyclic system lead to several regioisomers that can be observed by IR spectroscopy.15
Scheme 2.3. Acid-catalyzed cyclization. SCH3
H2O2
Me S
O
1. TFA
HOAc
S
2. H2O/Py
Carbon-sulfur bonds are smoothly cleaved using alkali metals16 and/or Raney nickel in heterogeneous media. Soluble Ni(0) complexes can be used to desulfurize and contract rings in heterocyclic compounds. Overall, this reaction can be used to effect transformation of thianthrene to dibenzothiophene in 50-70% yield (Scheme 2.4).17
Scheme 2.4. Desulfurization of thianthrene. S
2(COD)2Ni*bipy
S
THF / 50 oC
S
2.1.2 RING-CLOSURE AT THE ARENE MOIETY Using an Aldol condensation as key step, Laquindanum et al.18 prepared various anthradithiophene, Scheme 2.5. These and related compounds with even more extended acene units could be synthesized only as mixtures of syn and anti isomers.
10 Scheme 2.5. Synthesis of anthradithiophene. O CHO
1. [OH-]
S
+ S
2. Al, HgCl2
CHO
S
O
+ S
Our approach is the ring-closure at the benzenoid unit via acid-catalyzed cyclodehydration (Bradsher reaction), Scheme 2.6.
Scheme 2.6. Schematic representation of the Bradsher reaction. CHO
H+
2.2 SCOPE OF OUR RESEARCH AND ORGANIZATION OF CHAPTER The family of compounds we were interested is depicted in Figure 2.1. Therein, [1]benzothiophene (2.1) and dibenzothiophene (2.2) were commercially available. Benzo[1,2b:4,5-b’]dithiophene (2.3) and benzo[1,2-b:5,4-b’]dithiophene (2.4) were synthesized according to modified, previously reported procedures, in which a single Bradsher reaction was employed as the key step. We developed new schemes for the preparation of the extended, pentacyclic ring systems thieno[2,3-f:5,4-f’]bis[1]benzothiophene (2.5) and thieno[3,2-f:4,5f’]bis[1]benzothiophene (2.6). These syntheses are outlined below in chronological order of their synthesis; that is the synthesis of compound 2.6 was followed by the synthesis of compound 2.5.
11
S
S
2.1
S
S
2.5
2.3
S
2.2
S
S
S
S
S
S
S
2.4
2.6
Figure 2.1. Condensed ring systems 2.1-2.6 containing benzenes and thiophenes.
2.3 SYNTHESIS OF TARGET COMPOUND 2.6 Synthesis of 2.6 was achieved efficiently in four steps without column chromatography. Our approach took advantage of the effective regioselectivity of halogen-lithium exchange using commercially available thiophenes and their halogenated thiophene counterparts.19 Bromo- and iodothiophenes, when substituted in both a- and b-positions, exchange readily with alkyllithiums to yield the respective lithiothiophenes.20 Thiophene was converted into 2,5thiophenedicarboxaldehyde using hydrogen-lithium exchange followed by formylation using DMF as previously described.21 The step-by-step synthetic approach for 2.6 is discussed below.
Scheme 2.7. Synthesis of a,a'-bis(3-bromo-2-thienyl)-2,5-thiophenedimethanol (2.10). Br S
2.7
Br
Br
nBuLi
S
2.8
Li
OHC S CHO
2.9
Br
Br
S
S HO
S
OH
2.10
Selective bromine-lithium exchange at the a-position of 2,3-dibromothiophene (2.7) was achieved using nBuLi.22 To this solution was added a solution of thiophene-2,5-dicarboxaldehyde
12 (2.9) and the resulting dicarbinol (2.10) was obtained in synthetic purity as confirmed by 1H NMR.23
Scheme 2.8. Reduction of dicarbinol 2.10. Br
Br
S
S S
HO
OH
2.10
Br
NaCNBH3
Br
S ZnI2
S S
2.11
Due to the instability of dicarbinol 2.10, traditional reducing agents of diarylcarbinol, such as LiAlH4/AlCl3,24,25 NaBH4/TFA,26 and even TMSCl/NaI27 did not produce the desired product 2.11 in satisfactory yields. NaCNBH3/ZnI2 a mild and effective reagent known for the reduction of aryl carbonyl compounds as well as benzylic, allylic and tertiary alcohols28 yielded pure 2.11 in 80%. t
BuLi is recommended for lithiation at the b-positions of bromothiophene.29 In case of
2.11, however, this reagent was too harsh and lead to decomposition of the compound. As depicted in Table 2.1, careful monitoring of the lithiation conditions (time and temperature) was necessary to optimize the formation of dialdehyde 2.12. With optimized conditions, a simple precipitation of the product from CHCl3/hexanes served for purification.
13 Table 2.1. Optimization of reaction conditions for double-lithiation. Br
Br
S
S
S
2. DMF
S
R2
R1
1. nBuLi
S S
2.11 Lithiation time
R1=R2 =Br (2.11) --
R1=H, R2=H
R1=H, R2=Br
R1=H, R2=CHO
R1=Br, R2=CHO
19 %
12 %
18 %
13 %
R1=R2 =CHO (2.12) 38 %
20 min
--
5%
4%
16 %
14 %
61 %
10 min
--
7%
--
18 %
--
75 %
1h
The acid-catalyzed cyclodehydration was traditionally carried out using HBr or polyphosphoric acid as catalysts; however HBr produces deformylated, and brominated side products and the handling of highly viscous polyphosphoric acid is awkward. To afford the effective cyclization of dialdehyde 2.12, we applied the convenient to use and easy to remove solid-state catalyst Amberlyst-15.30,31 Following azeotropic distillation in benzene using a DeanStark trap, we obtained product 2.6 in 46% yield after sublimation and recrystallization. The overall yield in the synthesis of 2.6 was 34%.32
2.4 SYNTHESIS OF TARGET COMPOUND 2.5 The synthesis of 2.5 was attempted via several different pathways until a viable approach could be found. The crucial steps and the various attempts are discussed below. In attempt I, the starting materials were tetrabromothiophene (2.13) and 3-thiophenecarboxaldehyde (2.14). Double-lithiation reactions at positions 2 and 5 were followed by treatment with the aldehyde 2.14. The molecular ion peak of the resulting product 2,5-bis(3-thienylmethyl)-3,4-
14 dibromothiophene (2.15) was observed by mass spectrometry, though no clean conversion could be achieved. This approach was further complicated by the relative difficulty of substituting the bromine substituents at positions 3 and 4 of the central thiophene ring. We therefore updated our pathway to attempt II.
Scheme 2.9. Attempt I for synthesis of 2.5. Br Br
Br Br
S
S 1. nBuLi 2. CHO HO
2.13
S
Br
Br S
S
S
S
OH
2.15
S
2.5
2.14
In 1975, Ayres reported the conversion of 2,5-dibromo-3,4-diiodothiophene (2.16) to 2,5dibromothiophene-3,4-dicarboxaldehyde,33 taking advantage of the higher reactivity of iodine versus bromine in halogen-lithium exchange. In this case, however, a clean double-lithiation of 2,5-dibromo-3,4-diiodothiophene could not be achieved. In a control reaction, we observed a 15:5:1 mixture of products formylated at (3,4):(2,3):(2,5) positions, which were inseparable by column chromatography.
Scheme 2.10. Attempt II for synthesis of 2.5. I Br
I S
1. nBuLi, - 78 ºC Br
2. S
2.16
CHO
2.17
OH HO
S Br
S
S
S
Br
2.18
S
S
2.5
In a third attempt, the excess halogens of the center ring were moved to the outer rings of the construct. In this case, our synthesis started with 2-iodothiophene-3-carboxaldehyde (2.19) and 2,5-dibromothiophene (2.20). While we were successful in the preparation of
15 2,5-dilithiothiophene (2.21), a pitfall was the faster lithium iodine exchange compared to the reaction of the lithium salt (2.21) with the electrophilic aldehyde (2.19).
Scheme 2.11. Attempt III for synthesis of 2.5. Br
S
Br
2.20
Li CHO I
S
2.19
S
Li
S
2.21 HO
I
I S
2.22
S
OH
S
S
S
2.5
A successful preparation of 2.5 could only be achieved with the synthetic scheme outlined below. Key to this approach was the development of a viable synthesis for 3,4thiophenedicarboxaldehyde. Previous synthetic reports of 2.23 were unsatisfactory. Nucleophilic aromatic substitution of cyanide in 3,4-dibromothiophene followed by reduction and hydrolysis using LiAlH4 leads to a complex mixture of isomers.34-37 Robba published a single-pot synthesis of thiophene-3,4-dicarboxaldehyde starting from 3,4-diiodothiophene. This starting material however, is not readily available and the synthesis is quite involved.38-40 Successful preparation of 2.23 from 3,4-dibromothiophene was achieved in a one-pot procedure by stepwise replacement of the halogens. An initial bromine-lithium exchange was carried out using nBuLi. The resulting lithium salt was quenched with a stoichiometric amount of DMF. Without further hydrolysis, the second bromine-lithium exchange was carried out using the stronger base tBuLi. The organolithium salt, the kinetic product, was quickly quenched with DMF to avoid rearrangements. This approach yielded at least four products as observed by GC-MS and TLC. A
16 simple sublimation followed by a single recrystallization from cyclohexane resulted in 2.23 in 34% yield. The application of the synthetic scheme as outlined in Scheme 2.12 together with the experimental parameters developed for compound 2.6, target molecule 2.5 could be successfully synthesized in 6% overall yield.41
Scheme 2.12. Synthetic scheme for preparation of 2.5. Br S
2.7
Br
OH HO
S 1. OHC 2.
S
nBuLi
S
S
NaCNBH3 Br
CHO S
S
Br
Br
ZnI2
S
Br
2.25
2.24
2.23
2.25
1. nBuLi 2. DMF
S
S CHO
S
S
S Amberlyst-15
OHC
2.26
S
2.5
2.5 ISOMERIC PURITY The isomeric purity of 2.5 and 2.6 was observed by 1H NMR (Figure 2.2). Therein, the terminal protons in the anti isomer (2.5) showed an AX spin system, whereas the protons of the syn isomer (2.6) showed an AB spin system.
17
Figure 2.2. Comparison of 1H NMR of 2.5 (upper) and 2.6 (lower).
2.6 GENERAL EXPERIMENTAL DETAILS Diethyl ether (ether) and tetrahydrofuran (THF) were freshly distilled from sodium/benzophenone ketyl radical prior to use. All organolithium reactions were carried out under inert atmosphere (argon) on a bath containing slurry of dry ice/acetone unless otherwise indicated. Organolithium reagents were purchased as solutions (2.5 M nBuLi in hexanes; 1.7 M t
BuLi in pentane). nBuLi was titrated prior to use with N-pivaloyl-o-toluidine.42 In general, the
lithiation process was monitored by gas chromatography wherein a small aliquot of reaction mixture was quenched with D2O and analyzed. DMF was kept overnight over BaO and distilled under vacuum from alumina (neutral, activated) prior to use.43 Alternatively, DMF was freshly distilled under vacuum from P2O5.44 All melting points are uncorrected. Column chromatography was carried out using silica gel (60Å, 32 – 63 mm, standard grade). MgSO4 was of anhydrous grade. NMR spectra were recorded at 300 and 400 MHz for 1H NMR, as well as at 100 and 75 MHz for 13C NMR. Chemical shifts are reported in parts per million (ppm), referenced relative to
18 residual non-deuterated solvents as indicated. Elemental analysis was performed at Atlantic Microlab, Atlanta, GA.
2.6.1 SYNTHESIS OF BENZO[1,2-B:4,5-B’]DITHIOPHENE (2.3) Benzo[1,2-b:4,5-b’]dithiophene (3)45-48 was synthesized according to modified, previously reported synthetic procedures, Scheme 2.13.49
Scheme 2.13. Synthesis of benzo[1,2-b:4,5-b']dithiophene (2.3). Br S
2.7
Br
Br
1. nBuLi 2.
S
S CHO
Br
LiAlH4/AlCl3 S
OH
2.28
2.27
S
S
2.14
2.28
CHO
1. nBuLi 2. DMF
S
S
PPA S
S
2.29
2.3
Synthesis of 3-Bromo-(3-thienyl)-2-thiophenemethanol (2.27). To a chilled solution of 2,3dibromothiophene (2.7, 1g, 4.13 mmol, 0.46 mL) in ether nBuLi (1.82 mL, 4.54 mmol) was added dropwise via syringe. After stirring for 10 min, 3-thiophenecarboxaldehyde (2.14) was added and the reaction stirred and gradually warmed to room temperature over night. The reaction was quenched with 25 mL H2O; the organic layer was separated and was washed with water until neutral. The aqueous layer was back-extracted with ether and the combined organic fractions dried over MgSO4. Evaporation of the solvent yielded 0.81 g (71%) of 2.2749 as a yellow oil. 1H NMR (CDCl3, 400 MHz): d 3.14 (1H, d, J = 3.5 Hz), 6.11 (1H, d, J = 3.5 Hz),
19 6.87 (1H, d, J = 5.2 Hz), 7.03 (1H, m), 7.16 (1H, d, J = 5.2 Hz), 7.22 (2H, m). 13C NMR (CDCl3, 100 MHz): d 68.1, 108.4, 122.4, 125.6, 126.4, 126.7, 130.1, 142.3, 143.3. 3-Bromo-2-(3-thienylmethyl)thiophene (2.28). Anhydrous AlCl3 (1.55 g, 11.62 mmol) was dissolved in ether and then slowly added to a chilled solution of LiAlH4 (0.22 g, 5.81 mmol) in ether at 0 °C. After warming to room temperature, carbinol 2.27 was added dropwise to promote gentle reflux. After 30 min, the reaction was chilled to 0 °C and quenched by careful addition of 15 mL ethyl acetate. The reaction mixture was poured onto ice-cold 1 M HCl and vigorously shaken. The organic phase was washed with 1 M HCl, saturated NaHCO3, and water. After drying over MgSO4, solvent was evaporated to yield 0.624 g (83%) of 2.28.47 1H NMR (CDCl3, 400 MHz): d 4.12 (2H, s), 6.92 (1H, d, J = 5.4 Hz), 6.98 (1H, dd, J1 = 4.9 Hz, J2=1.1 Hz), 7.04 (1H, m), 7.13 (1H, d, J = 5.4 Hz), 7.26 (1H, dd, J1 = 4.9 Hz, J2 = 3.0 Hz). 13C NMR (CDCl3, 100 MHz): d 29.9, 109.1, 121.9, 123.8, 125.9, 128.2, 129.9, 138.4, 139.2. 2-(3-Thienylmethyl)-3-thiophenecarboxaldehyde (2.29). nBuLi (0.184 g, 2.86 mmol, 1.14 mL) was chilled in ether to –78 °C. A solution of 2.28 in ether was added dropwise and stirred for 10 min, while the solution turned yellow. This solution was then transferred via cannula to a wellstirred mixture of DMF (0.35 g, 4.78 mmol, 0.37 mL) in ether wherein a color change to deep blue occurred. After stirring at room temperature for 12 hours, the reaction was quenched with ice-water. The organic fraction was washed with 1 M HCl, saturated NaHCO3, and water. After drying over MgSO4, the solvent was removed to yield 0.408 g (81%) of 2.2948 as a pale pink liquid. 1H NMR (CDCl3, 400 MHz): d 4.53 (2H, s), 6.97 (1H, dd, J1 = 4.9 Hz, J2 = 1.31 Hz), 7.05 (1H, m), 7.11 (1H, dd, J1 = 5.3 Hz, J2 = 0.5 Hz), 7.27 (1H, dd, J1 = 4.9 Hz, J2 = 3.0 Hz), 7.40 (1H, d, J = 5.3 Hz), 10.05 (1H, s). 13C NMR (CDCl3, 100 MHz): d 28.8, 122.2, 123.7, 126.2, 127.7, 127.8, 136.6, 139.2, 155.1, 184.6.
20 Benzo[1,2-b:4,5-b’]dithiophene (2.3). A solution of 2.29 in 2 mL benzene was added into polyphosphoric acid (50 g). The slurry was kept on a water bath at 75 °C for 30 min. After the addition of crushed ice in water, the mixture was extracted into ether. The organic phase was then washed with water until neutral. After drying over MgSO4, the solvent was removed. The addition of some ethanol facilitated removal of benzene. Recrystallization using ethanol yielded 0.36 g (35%) of 2.347 as a white, crystalline solid. Alternative purification methods included washing through silica using hexanes as elution solvent, or sublimation at 110 °C at 9.0x10-3 torr. Mp. 194-195 °C (Lit.50-52 197.5-198 °C). 1H NMR (CDCl3, 400 MHz): d 7.34 (2H, d, J = 5.5 Hz), 7.45 (2H, d, J = 5.5 Hz), 8.31 (2H, s). 13C NMR (CDCl3, 100 MHz): d 116.8, 122.9, 127.2, 137.5, 137.1.
2.6.2 SYNTHESIS OF BENZO[1,2-B:5,4-B’]DITHIOPHENE (2.4) Benzo[1,2-b:5,4-b’]dithiophene was synthesized according to modified synthetic procedures as indicated in Scheme 2.14.53
Scheme 2.14. Synthetic scheme of benzo[1,2-b:5,4-b']dithiophene (2.4). Br Br
S
2.7
2.31
Br
1. nBuLi S
2.
S
S
OH
S CHO
2.17
2.30
S
PPA S
2.32
S
2.31
CHO
1. nBuLi 2. DMF
Br
LiAlH4/AlCl3
S
S
2.4
21 3-Bromo-(2-thienyl)-2-thiophenemethanol (2.30). 2,3-Dibromothiophene 2.7 (20.57 g, 85.0 mmol) was dissolved in ether and chilled to –78 °C. A solution of nBuLi (93.5 mmol, 37.4 mL) was added and the reaction stirred for 10 min. Then, 2-thiophenecarboxaldehyde 2.17 (9.53 g, 85.0 mmol, 7.94 mL) was dropwise added and the reaction stirred overnight whereupon it slowly warmed to room temperature. A workup similar to compound 2.27 resulted in 19.47 g (83%) of 2.30 as a yellow oil. 1H NMR (CDCl3, 400 MHz): d 2.70 (1H, d, J = 3.6 Hz), 6.37 (1H, d, J = 3.6 Hz), 6.93 (1H, d, J = 5.2 Hz), 6.95 (1H, dd, J1 = 5.0 Hz, J2 = 3.5 Hz), 7.02 (1H, m), 7.27 (2H, m). 13
C NMR (CDCl3, 100 MHz): d 68.0, 108.4, 125.2, 125.5, 125.7, 126.7, 130.0, 141.6, 145.5.
3-Bromo-2-(2-thienylmethyl)thiophene (2.31). The reduction of 2.30 (19.47 g, 70.73 mmol) was carried out using AlCl3 (37.72 g, 282.93 mmol) and LiAlH4 (5.37 g, 141.46 mmol) as described for 2.28 and yielded 13.97 g (76%) of 2.31 as a yellow liquid.51 1H NMR (CDCl3, 400 MHz): d 4.30 (2H, s), 6.92 (3H, m), 7.15 (2H, m): d 13C NMR (CDCl3, 100 MHz) d 29.4, 109.2, 124.2, 124.3, 125.6, 126.9, 129.9, 137.8, 141.4. 2-(2-Thienylmethyl)-3-thiophenecarboxaldehyde (2.32). The formylation of 2.31 (13.97 g, 53.90 mmol) was carried out as described before using nBuLi (64.68 mmol, 25.87 mL) and DMF (107.80 mmol, 8.35 mL). This procedure yielded 10.3251 (10.90 g) in 97%. 1H NMR (CDCl3, 400 MHz): d 4.72 (2H, s), 6.90 (1H, m), 6.93 (1H, dd, J = 5.2 Hz, J = 3.5 Hz), 7.14 (1H, d, J = 5.3 Hz), 7.19 (1H, dd, J1 =5.2 Hz, J2 = 1.3 Hz), 7.40 (1H, d, J = 5.3 Hz), 10.1 (1H, s). 13C NMR (CDCl3, 100 MHz): d 28.4, 123.8, 124.7, 126.0, 127.0, 128.0, 136.6, 141.3, 154.5, 184.6. Benzo[1,2-b:5,4-b']dithiophene (2.4). This improved procedure was carried out by Mindy Horst (Undergraduate research student, 2004). Aldehyde 2.32 (3.48 g, 16.7 mmol) was dissolved in benzene and refluxed using a Dean-Stark trap for 9.5 hours over Amberlyst-15 (0.5 g). After cooling, the beads were filtered and washed with dichloromethane. The resulting brown solid
22 was sublimed at 140 °C under vacuum (9x10-3 torr) and recrystallized from hexanes to yield 2.20 g (69%) of an off-white solid. Mp. 182-184 °C (Lit. 187-188 °C). 1H NMR (CDCl3, 400 MHz): d 7.38 (2H, d, J = 5.5 Hz), 7.43 (2H, d, J = 5.5 Hz), 8.26 (1H, s), 8.36 (1H, s). 13C NMR (CDCl3, 100 MHz): d 115.7, 117.7, 123.6, 126.2, 137.2, 137.6.
2.6.3 SYNTHESIS OF THIENO[2,3-F:5,4-F’]BIS[1]BENZOTHIOPHENE (2.5) 3,4-Thiophenedicarboxaldehyde (2.23). 3,4-Dibromothiophene (8.75 g, 4.0 mL, 36.17 mmol) was dissolved in 100 mL of ether and the solution purged with Ar. After cooling to –78 °C, a solution of nBuLi (15.91 mL, 39.78 mmol) was added dropwise over a period of 5 min. DMF (3.07 mL, 2.91 g, 39.78 mmol) was then added and the reaction stirred for 1 h. tBuLi (57.88 mL, 86.83 mmol) was chilled to –78 °C and then rapidly added. After 5 min, DMF (6.70 mL, 6.34 g, 86.83 mmol) was added and the reaction was stirred overnight. The reaction mixture was then poured onto 2 N HCl at –20 °C and warmed to room temperature. After separating the organic layer, the aqueous layer was extracted with copious amounts of ether. The combined organic layers were washed with brine and water and subsequently dried over MgSO4. After evaporation of the solvent, the orange-brown solid was sublimed at 110 °C under vacuum (ca. 28 in. Hg). The yellowish-white solid was dissolved in hot cyclohexane and the supernatant decanted. This process was repeated until only a dark-brown residue remained in the flask. The cyclohexane fractions were combined, reheated, and left for crystallization in a freezer (–20 °C). This process yielded 1.72 g (34%) of a yellowish-white solid, mp 76–77 °C (Lit. 78–80 °C). 1H NMR (CDCl3, 300 MHz): d 8.23 (2H, s), 10.31 (2H, s). 13C NMR (CDCl3, 75 MHz): d 137.6, 140.3, 185.8. 3,4-Bis(3-bromo-2-thienylmethyl)thiophene (2.25). 2,3-Dibromothiophene (2.7, 10.00 g, 41.33 mmol) was dissolved in ether/THF (5:1, 200 mL). At –78 °C, nBuLi (20.0 mL, 47.1 mmol) was
23 added dropwise and the solution was stirred for 5 min. A chilled solution of 2.23 (3.0 g, 21.4 mmol) in THF (160 mL) was subsequently added and the reaction was allowed to warm to room temperature. After addition of water, the aqueous layer was extracted with ether and the combined organic layers washed with water and dried over MgSO4. Evaporation of the solvent yielded 2.24 as a yellow liquid of 2.24, which was dissolved in 1,2-dichloroethane. ZnI2 (20.50 g, 64.21 mmol) and NaCNBH3 (20.18 g, 321.06 mmol) were added and the reaction was stirred overnight. The reaction mixture was subsequently filtered through Celite, washed with aqueous saturated NH4Cl, dried over MgSO4, and evaporated. Column chromatography using hexanes yielded 3.99 g (43%) of a yellow oil. 1H NMR (CDCl3, 300 MHz): d 4.02 (4H, s), 6.93 (2H, d, J = 5.1 Hz), 6.99 (2H, s), 7.14 (2H, d, J = 5.1 Hz). 13C NMR (CDCl3, 75 MHz): d 28.8, 109.3, 123.6, 124.1, 130.0, 137.6, 137.9. Anal. Calcd for C14H10Br2S3: C, 38.72; H, 2.32. Found: C, 38.69; H, 2.53. 2,2’-[3,4-Thiophenediylbis(methylene)]bis[3-thiophenecarboxaldehyde] (2.26). nBuLi (8.27 mL, 19.86 mmol) was chilled to –78 °C in ether. Compound 2.25 (3.92 g, 9.03 mmol) in ether was added dropwise. Ten minutes after the addition, DMF (1.53 mL, 1.45 g, 19.84 mmol) was added and the reaction was stirred for 2 h. The mixture was poured into a separatory funnel containing ice-cold saturated NH4Cl. After separating the organic layer, the aqueous layer was extracted with ether and the combined organic layers washed with water and dried over MgSO4. Recrystallization of 2.26 from ether yielded a white solid, 1.62 g (54%), mp 89–90 °C. An analytical sample was prepared using column chromatography (4:1 hexanes/ethyl acetate as elution solvent), mp 93–94 °C. 1H NMR (CDCl3, 300 MHz): d 4.45 (4H, s), 7.04 (2H, s), 7.12 (2H, d, J = 5.4 Hz), 7.39 (2H, d, J = 5.4 Hz), 10.05 (2H, s). 13C NMR (CDCl3, 75 MHz): d 27.8, 123.9, 124.4, 128.2, 136.7, 138.0, 154.3, 184.7. Anal. Calcd for C16H12O2S3: C, 57.80; H, 3.64.
24 Found: C, 57.60; H, 3.79. Thieno[2,3-f:5,4-f’]bis[1]benzothiophene (2.5). Compound 2.26 (1.63 g, 4.90 mmol) was dissolved in dry benzene and Amberlyst-15 (0.5 g) was added. The reaction was refluxed for 36 h, while water was removed by means of a Dean-Stark trap. After cooling, dichloromethane was added to dissolve the precipitate and the mixture was filtered through a cotton plug. The filtrate was washed with saturated NH4Cl and then dried over MgSO4. Evaporation of the solvent furnished a brown solid. Purification by column chromatography on silica gel using hexanes as elution solvent yielded a pale white solid. Recrystallization from CHCl3 yielded 0.35 g (24%) of pale-white crystals, mp 237–238 °C. 1H NMR (CDCl3, 300 MHz): d 7.39 (2H, d, J = 5.4 Hz), 7.54 (2H, d, J = 5.4 Hz), 8.24 (2H, s), 8.68 (2H, s). 13C NMR (CDCl3, 75 MHz): d 115.0, 117.0, 123.1, 127.8, 132.7, 136.9, 137.1, 139.8. Anal. Calcd for C16H8S3: C, 64.83; H, 2.72. Found: C, 64.54; H, 2.95.
2.6.4 SYNTHESIS OF THIENO[3,2-F:4,5-F’]BIS[1]BENZOTHIOPHENE (2.6) 2,5-Thiophenedicarboxyaldehyde (2.9). The title compound was synthesized according to a literature procedure.21 Mp 113-114 ºC (Lit.54 113 ºC); 1H NMR (CDCl3, 400 MHz): d 7.84 (2H, s), 10.04 (2H, s). 13C NMR (CDCl3, 100 MHz) d 135.7, 149.8, 184.0; GC–MS m/z (%): 140 (100)[M+], 139 (72) [M+-1], 111 (25) [M+-29]. a,a'-Bis(3-bromo-2-thienyl)-2,5-thiophenedimethanol (2.10). 2,3-Dibromothiophene (2.7, 8.12 g, 33.6 mmol) was added to a 300 mL solution of 5:1 ether/THF and cooled to –78 °C. n
BuLi (14.8 mL, 36.96 mmol) was added dropwise while maintaining the temperature below –60
°C. After 15 min, 2.9 (2.35 g, 16.7 mmol) dissolved in 150 mL of THF was added slowly and the mixture stirred for 1 h. Aqueous workup and drying over MgSO4 yielded a yellow oil which was
25 used without further purification. 1H NMR (THF-d8, 400 MHz): d 3.05 (2H, s), 6.31 (2H, s), 6.86 (2H, d, J = 2.8 Hz), 6.93 (2H, dd, J = 5.2 Hz, J = 0.4 Hz), 7.27 (2H, d, J = 5.2 Hz). MS (EI, 70 eV) m/z (%): 468 (12) [M + 2], 466 (21) [M+], 464 (12) [M – 2], 448 (18) [M – H2O], 432 (50), 369 (20), 272 (48), 227 (42), 191 (100), 178 (34), 134 (29), 111 (48). 2,5-Bis(3-bromo-2-thienylmethyl)thiophene (2.11). To a solution of 2.10 (7.824 g, 16.7 mmol) in dichloroethane were added ZnI2 (16.11 g, 53.37 mmol) and NaCNBH3 (14.72 g, 234.94 mmol). The reaction mixture was stirred at room temperature overnight. Then the mixture was filtered through Celite. The filtrate was washed with saturated NH4Cl and water. After drying over MgSO4 and removal of solvent under reduced pressure, the resulting orange oil was passed through a short-bed of silica using hexanes as elution solvent. This yielded 13.46 g of yellow oil (80% over two steps). 1H NMR (CDCl3, 400 MHz): d 4.22 (4H, s), 6.71 (2H, s), 6.92 (2H, d, J = 5.2 Hz), 7.15 (2H, d, J = 5.2 Hz). 13C NMR (CDCl3, 100 MHz): d 140.5, 137.6, 129.9, 125.4, 124.2, 109.3, 29.7. GC-MS (EI, 70 eV) m/z (%): 434 (66) [M+], 355 (51) [M – Br], 259 (100), 177 (69), 137 (54), 45 (68). Anal. Calcd for C14H10Br2S3: C, 38.72; H, 2.32. Found: C, 38.88; H, 2.34. 2,2'-[2,5-Thiophenediylbis(methylene)]bis[3-thiophenecarboxaldehyde] (2.12). nBuLi (2.5 M, 2.6 mL, 6.59 mmol) was cooled to –78 °C in 90 mL of ether. Compound 2.10 (1.3 g, 2.99 mmol) dissolved in 15 mL of ether was added dropwise and the mixture stirred for 10 min. DMF (0.5 mL, 6.23 mmol) was added and the reaction mixture stirred for 1 h. Quenching with saturated NH4Cl and workup with water and brine yielded 0.939 g (94%) of 2.12. Recrystallization from CH2Cl2/hexanes provided pale pink needles, mp 89–91 °C. 1H NMR (CDCl3, 400 MHz): d 4.64 (4H, s), 6.73 (2H, s), 7.14 (2H, d, J = 5.6 Hz), 7.40 (2H, d, J = 5.6 Hz), 10.05 (2H, s). 13C NMR (CDCl3, 100 MHz): d 184.6, 154.1, 140.8, 136.5, 127.9, 125.8, 123.9, 28.5. MS (EI, 70 eV) m/z
26 (%): 332 (45) [M], 303 (13) [M – CHO], 207 (100), 179 (34), 124 (32), 97 (33). Anal. Calcd for C16H12O2S3: C, 57.80; H, 3.64. Found: C, 57.47; H, 3.53. Thieno[3,2-f:4,5-f’]bis[1]benzothiophene (2.6). Dialdehyde 2.12 (0.919 g, 2.77 mmol) was dissolved in 50 mL of benzene, Amberlyst-15 (1.3 g) was added, and the reaction was refluxed overnight using a Dean-Stark trap. The color changed from pink to pale beige, and the product started to precipitate as a white solid. The solid was dissolved in dichloromethane, the Amberlyst-15 was removed by filtration, and the reaction mixture washed with water. Drying over MgSO4 and removal of the solvent yielded 0.717 g (88%) of beige solid. After sublimation (190 °C, 2.0 x 10–2 torr) and recrystallization from ethanol, 0.412 g of white crystals were obtained in 46% yield, mp 272–273 °C. 1H NMR (CDCl3, 400 MHz): d 7.48 (4H, s), 8.28 (2H, s), 8.64 (2H, s). 13C NMR (CDCl3, 75 MHz): d 139.4, 137.4, 136.7, 133.3, 126.2, 123.8, 116.0, 115.7. MS (EI, 70 eV) m/z (%): 298 (15) [M + 2], 296 (100) [M], 148 (24). Anal. Calcd for C16H8S3: C, 64.83; H, 2.72. Found: C, 64.92; H, 2.81.
2.7 REFERENCES (1)
Schleyer, P. v. Chem. Rev. 2001, 101, 1115-1117.
(2)
Benedikov, M.; Wudl, F. Chem. Rev. 2004, 104, 4891-4945.
(3)
Fichou, D. Handbook of Oligo- and Polythiophenes; Wiley-VCH: Weinheim, 1999.
(4)
Kirchmeyer, S.; Reuter, K. J. Mater. Chem. 2005, 15, 2077-2088.
(5)
Hartough, H. D.; Meisel, S. L. Compounds with Condensed Thiophene Rings; Interscience Publishers, Inc.: New York, 1954.
(6)
Andrews, M. D. In Science of Synthesis: Houben-Weyl Methods of Molecular Transformations; Thomas, E. J., Ed.; Thieme: Stuttgart, 2001; Vol. 10, p 211-263.
27 (7)
Moss, G. P. Pure Appl. Chem. 1998, 70, 143-216.
(8)
Yamada, K.; Ogashiwa, S.; Tanaka, H.; Nakagawa, H.; Kawazura, H. Chem. Lett. 1981, 343-346.
(9)
Wynberg, H. Acc. Chem. Res. 1971, 4, 65-73.
(10)
Larsen, J.; Bechgaard, K. Acta Chem. Scand. 1996, 50, 77-82.
(11)
Tilak, B. D. Tetrahedron 1960, 9, 76-95.
(12)
Clark, P. D.; Kirk, A.; Yee, J. G. K. J. Org. Chem. 1995, 60, 1936-1938.
(13)
Yue, D.; Larock, R. C. J. Org. Chem. 2002, 67, 1905-1909.
(14)
Hori, M. K., T.; Shimizu, H.; Miyagaki, M.; Takagi, T. Chem. Pharm. Bull. 1974, 22, 2030-2041.
(15)
Sirringhaus, H.; Friend, R. H.; Wang, C.; Leuninger, J.; Müllen, K. J. Mater. Chem. 1999, 9, 2095-2101.
(16)
Gilman, H.; Swayampati, D. R. J. Am. Chem. Soc. 1955, 77, 3387-3389.
(17)
Eisch, J., Kyoung, R. I. M. J. Organomet. Chem. 1977, 139, C51-C55.
(18)
Laquindanum, J. G.; Katz, H. E.; Lovinger, A. J. J. Am. Chem. Soc. 1998, 120, 664-672.
(19)
Clayden, J. Organolithiums: Selectivity for Synthesis; Elsevier Science Ltd.: Oxford, U.K., 2002.
(20)
Joule, J. A.; Mills, K. Heterocyclic Chemistry; 4 ed.; Blackwell Science Ltd.: Oxford, U.K., 2000.
(21)
Feringa, B. L.; Hulst, R.; Rikers, R.; Brandsma, L. Synthesis 1988, 316-318.
(22)
Spagnolo, P.; Zanirato, P. J. Chem. Soc. Perkin Trans. 1 1996, 963-964.
(23)
Compound 2.10 decomposes within several hours. It could not be purified and no 13C NMR spectrum or elemental analysis obtained.
28 (24)
Yamada, M.; Ikemote, I.; Kuroda, H. Bull. Chem. Soc. Jpn. 1988, 61, 1057-1062.
(25)
Aubry, J.-M.; Pierlot, C.; Rigaudy, J.; Schmidt, R. Acc. Chem. Res. 2003, 36, 668-675.
(26)
Nutaitis, C. F.; Patragnoni, R.; Goodkin, G.; Neighbour, B.; Obaza-Nutaitis, J. Org. Prep. Proced. Int. 1991, 23, 403-411.
(27)
Nenajdenko, V. G.; Baraznenok, I. L.; Balenkova, E. S. J. Org. Chem. 1998, 63, 61326136.
(28)
Lau, C. K.; Dufresne, C.; Belanger, P. C.; Pietre, S.; Scheigetz, J. J. Org. Chem. 1986, 51, 3038-3043.
(29)
Mitsumori, T.; Inoue, K.; Koga, N.; Iwamura, H. J. Am. Chem. Soc. 1995, 117, 24672478.
(30)
Amberlyst-15 consists of sulfonated polystyrene beads.
(31)
Baxendale, I. R.; Storer, R. I.; Ley, S. V. In Polymeric Materials in Organic Synthesis and Catalysis; Buchmeiser, M. R., Ed.; Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim, Germany, 2003, p 53-135.
(32)
Wex, B.; Kaafarani, B. R.; Neckers, D. C. J. Org. Chem. 2004, 69, 2197-2199.
(33)
Ayres, B. E.; Longworth, S. W.; McOmie, J. F. W. Tetrahedron 1975, 31, 1755-1760.
(34)
Trofimenko, S. J. Org. Chem. 1964, 29, 3046-3049.
(35)
Dubus, P.; Decroix, B.; Morel, J.; Pastour, P. Bull. Soc. Chim. Fr. 1976, 628-634.
(36)
Inaoka, S.; Collard, D. M. J. Mater. Chem. 1999, 9, 1718-1725.
(37)
Langeveld-Voss, B. M. W.; Janssen, R. A. J.; Meijer, E. W. J. Mol. Struct. 2000, 521.
(38)
Steinkopf, W.; Schmitt, H. F.; Fiedler, H. Liebigs Ann. Chem. 1937, 527, 237-263.
(39)
Robba, M.; Moreau, R. C.; Roques, B. C. R. Hebd. Seances Acad. Sc. Fr. 1964, 259, 3568-3570.
29 (40)
Boberg, R.; Jachiewicz, A.; Garming, A. Phosphorus Sulfur Silicon Rel. Elem. 1992, 72, 1-12.
(41)
Wex, B.; Kaafarani, B. R.; Kirschbaum, K.; Neckers, D. C. J. Org. Chem. 2005, 70, 4502-4505.
(42)
Suffert, J. J. Org. Chem. 1989, 54, 510-512.
(43)
Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory Chemicals; 3rd ed.; Butterworth-Heinemann Ltd.: Oxford, U. K., 1988.
(44)
Burfield, D. R.; Smithers, R. H. J. Org. Chem. 1978, 43, 3966-3968.
(45)
Rao, D. S.; Tilak, B. D. J. Sci. Ind. Res. 1957, 16B, 65-68.
(46)
Aggarwal, N.; MacDowell, D. W. H. Org. Prep. Proced. Int. 1979, 11, 247-249.
(47)
Beimling, P.; Kossmehl, G. Chem. Ber. 1986, 119, 3198-3203.
(48)
Laquindanum, J. G.; Katz, H. E.; Lovinger, A. J.; Dodabalapur, A. Adv. Mater. 1997, 9, 36-39.
(49)
Kraak, A. W., A. K.; Jordens, P.; Wynberg, H. Tetrahedron 1968, 24, 3381-3398.
(50)
Rao, D. S.; Tilak, B. D. J. Sci. Ind. Res. 1954, 13B, 829-834.
(51)
Wynberg, H.; De Wit, J.; Sinnige, H. J. M. J. Org. Chem. 1970, 35, 711-715.
(52)
Ahmed, M.; Ashby, J.; Ayad, M.; Meth-Cohn, O. J. Chem. Soc. Perkin Trans. 1 1973, 1099-1103.
(53)
MacDowell, D. W. H.; Wisowaty, J. C. J. Org. Chem. 1971, 36, 4004-4012.
(54)
Stetter, H.; Rajh, B. Chem. Ber. 1976, 109, 534-540.
30 CHAPTER III. PHOTOADDITION REACTIONS OF ACETYLENE AND BUTADIYNE DERIVATIVES TO BENZODITHIOPHENE
3.1 INTRODUCTION In his classic review written in 1971 about thiophene, Wynberg1 wrote, “To many chemists thiophene is merely one member of the inexhaustible supply of heterocyclic compounds. In spite of being known for 100 + years, it has never found large scale industrial use save in some semi-synthetic penicillins.” Today, due mostly to their efficient electrochemical polymerization, however, thiophene building blocks are playing major roles in the development of new materials ranging from conducting polymers2,3 and organic semiconductors4 to materials for light-emitting diodes.5 In this chapter, we investigate fused thiophenes in their synthetic applications as synthons for arenes. Generally, a thermal [4p+2p] or a photochemical [2p+2p] either followed by pyrolysis can potentially lead to new aromatic units. Thiophenes are prone to undergo pericyclic reactions upon irradiation. The most wellknown example is the photochromic electrocyclic ring opening and closing of 1,2-bis(3thienyl)perfluorocyclopentene.6,7 The photochemistry of thiophene8 remains largely unexplored. For instance, thiophene uniquely undergoes photochemical cycloaddition reactions with acetylenes.9 This reaction can be used as a convenient pathway in the transformation of a thiophene to a benzene moiety and serve as a photochemical approach in the synthesis of oligo[n]acenes,10 polycyclic aromatic hydrocarbons of the Clar type10 and molecular subunits of graphite11 (Figure 3.1).
31
*
S
*
*
*
hv,
n
n+1
Figure 3.1. Thiophene to acene conversion. Intermolecular photoaddition reactions of acetylenes to heteroaromatic compounds such as furan, thiophene, and pyrrole previously have been reported.12 The photochemical reaction of thiophene to dimethyl acetylenedicarboxylate (DMAD) affords a [4p+2p] Diels-Alder adduct that undergoes instant extrusion of sulfur yielding dimethyl phthalate as the sole reaction product.9 [1]Benzothiophene (2.1) undergoes cycloaddition reactions with DMAD, methyl propiolate, and methyl phenylpropiolate affording adducts that differ from the expected simple [2p + 2p] photoadduct.13,14 The primary photoadduct of the reaction with DMAD undergoes a secondary photorearrangement under the reaction conditions as verified by the work of Ditto and co-workers.14,15 Interestingly, this rearranged photoadduct is thermally unstable and yields dimethyl naphthalene-1,2-dicarboxylate upon pyrolysis.14 In this chapter, we report on the photocycloaddition of benzodithiophenes 2.3 and 2.4 (Figure 3.2) to DMAD and various 1,4diarylbutadiynes.
S S
S
2.3
S
2.4
Figure 3.2. Molecular structure of benzodithiophenes.
32 3.2 PHOTOCYCLOADDITION OF DIMETHYL ACETYLENEDICARBOXYLATE TO BENZODITHIOPHENES We have investigated the photochemical [2p + 2p] cycloaddition of benzo[1,2-b:4,5b’]dithiophene (2.3) to DMAD (3.1), Scheme 3.1. Irradiation of 2.3 and 3.1 in benzene at 300 nm under argon for 16 hours yielded one product as observed by TLC. When the irradiation was carried out in toluene, no product was observed. Photolysis of only 2.3 in benzene did not alter the 1H NMR of 2.3. The photolysis of 3.1 under these conditions yielded one new signal in NMR that is attributed to the self-dimerization product. Expected products, Scheme 3.1, were the direct [2p + 2p] photoproduct 3.2 or the rearranged product of secondary photolysis (3.3). The hydrogens of the cyclobutene unit can distinguish these products. Two hydrogens attached to sp3-hybridized carbons are present in product 3.2, whereas two hydrogens are attached to one sp3 and one sp2-hybridized carbon in product 3.3.
Scheme 3.1. Photoaddition of DMAD (3.1) to benzo[1,2-b:4,5-b’]dithiophene (2.3). S
COOMe S
hv
+ S COOMe
2.3 1
3.1
S
H COOMe
S H
hv
COOMe
S H
COOMe
3.2
COOMe
H
3.3
H NMR of the purified photoproduct exhibited two absorptions at a chemical shift of
4.89 and 6.99 ppm indicative for product 3.3 (Figure 3.3).
33
Figure 3.3. 1H NMR spectrum of photoproduct 3.3 in CDCl3. Product 3.3 was isolated in 16% yield. Unreacted starting material 2.3 was recovered in 53%. The optimization of the reaction conditions revealed that a 1:1 mixture of 2.3 and 3.1 yielded the highest conversion rate (Table 3.1). Table 3.1. Optimization of reaction conditions. Ratio of 2.3/3.1
Ratio of Integrals
Conversion
(P/SM’)a
[%]
1:1
1/2.6
43
1:5
1/5.7
25
1:10
1/3.3
37
5:1
1/6.2
24
10:1
1/6.4
23
a) Integrated signals after irradiation were the proton of the cyclobutene ring (P) and the two protons on the benzene ring of 2.3 (SM’). The conversion was calculated using the following formula: conversion=(P*2/SM)*100, where SM=(SM’+P*2).
34 Figure 3.4 depicts the chromatogram of 3.3 as observed by GCMS. The observed peak of 3.3 with m/z = 300 [M+ - 32] is attributed to the thermal rearrangement product dimethyl naphtho[2,3-b]thiophene-5,6-dicarboxylate (3.4), in agreement with previous literature reports for similar systems.14,16
Figure 3.4. Gas chromatogram (upper) and mass spectrum (lower) of photoproduct. S
COOMe
COOMe COOMe
D
COOMe
S
S
3.3
3.4
Mol. Wt.: 332.40
Mol. Wt.: 300.33
Figure 3.5. Proposed pyrolysis product of 3.3. The photoreaction of 2.4 with 3.1 yielded similar 1H NMR signals at 3.29 and 3.36 ppm for the two protons of the methyl ester groups and at 4.44 and 6.45 ppm for the protons on the cyclobutene ring (Scheme 3.2, Figure 3.6).
35 Scheme 3.2. Photoaddition of 3.1 to 2.4. + S
COOMe hv
H
COOMe
S COOMe
2.4
H
3.1
S
S
COOMe
3.5
Figure 3.6. 1H NMR spectrum of irradiated mixture of 2.4 and 3.1.
3.3 PHOTOCYCLOADDITION OF BUTADIYNES TO BENZODITHIOPHENES Photocycloaddition of conjugated poly-ynes to alkenes yields cyclobutene adducts and minor quantities of bicyclopropyl adducts.17-19 1,4-Diphenyl-1,3-butadiyne (3.6) undergoes photoaddition to various alkenes such as 2,3-dimethylbut-2-ene,20-22 dimethyl fumarate,23,24 acrylonitrile, and ethyl vinyl ether25 affording regioselective and site-selective cyclobutene derivatives, cyclopropyl, and oxirane adducts. The proposed reaction pathways involve a cumulene-type triplet excited state, a polar triplet structure of the 1:1 cyclobutene adduct, and carbene intermediates in the case of cyclopropyl and oxirane adducts.20-24 We investigated the [2p+2p] photocycloaddition reaction of 2.3 to 3.6. Irradiation of 2.3 and 3.6 using 300 nm light under argon could potentially lead to four different products, two primary (3.7 and 3.9) and two secondary (3.8 and 3.10) products as indicated in Figure 3.7.
36 S S hv S
S S
Benzene
3.7
l = 300 nm
S
3.8
H
S S
2.3
S
hv
3.6
S H
3.9
3.10
Figure 3.7. Possible photoreaction products of 2.3 and 3.6. Two photoproducts were observed by TLC and 1H NMR. The two isomers were formed as a 5:1 mixture based on 1H NMR. One photoproduct could be isolated by column chromatography, (4%), while 96% of 2.3 was recovered. The second photoproduct could not be isolated. It is important to note that if photolyzed separately, 2.3 and 3.615 do not undergo photoadditions or decomposition as observed by 1H NMR. Both, unreacted 2.3 and 3.6 can be recovered after photolysis. The photoproduct was characterized by X-ray crystal structure analysis. Figure 3.8 depicts an Ortep representation of the isolated photoproduct. The photoaddition is regioselective as the X-ray crystal structure revealed in that the photoaddition reaction proceeded to yield only syn adduct. Based on the X-ray structure, the molecular structure was assigned to product 3.7. The second photoproduct of the reaction between 3.6 and 2.3 was assigned to be 3.8 as the secondary photolysis product of 3.7.
37
Figure 3.8. ORTEP representation of 3.7 at 173K (ellipsoids at 50% probability); minor conformer of disordered phenyl ring omitted for clarity. The UV-vis absorption spectrum of a mixture of 2.3 and 3.6 is a superposition of the individual spectra; we can therefore rule out a ground state complex in the reaction pathway.26,27 The molar decade extinction coefficient of 3.6 is ~3-fold higher than 2.3 at 300 nm.28 The most efficient product formation occurred when compound 2.3 was used in excess (Figure 3.9). The 0-0 band of 2.3 was assigned at 489 nm and corresponds to a triplet energy of 59 kcal/mol.29 The triplet energy of 3.6 was 58 kcal/mol.9 The yield of the photoproduct as observed by NMR was slightly increased in the presence of iodoethane, while the presence of piperylene (ET = 57 kcal/mol; triplet quencher) decreased the yield of photoproduct suggesting the photoreaction proceeds via the triplet excited state of 2.3.30
38
Figure 3.9. Effect of concentration on product formation (SM=starting material 2.3, PA=photoadduct 3.7); Ratio of SM/PM corresponds to the integrated absorption of an a-proton of 2.3 and the lower-field proton of the cyclobutene ring of 3.7. The absorption spectrum of 3.7 (Figure 3.10) exhibited a broad, structure-less absorption band around 300 – 350 nm. This band is similar to the charge-transfer absorption band observed in the primary photoproduct of 3.1 and 2.1.15
39
Figure 3.10. Absorption spectra of 2.3, 3.6, and 3.7 in ethanol. Beside 3.6, we have investigated the photoaddition of a series of electron-poor and electron-rich butadiynes (3.11–3.13). These reactions yielded two major products identified as the cyclobutene analogs of 3.7 and 3.8 as confirmed by 1H NMR spectroscopy, Table 3.1. The photoreaction of 2.3 to 1,4-bis(4’-fluorophenyl)buta-1,3-diyne (3.11)31, 1,4-bis(2’pyridyl)butadiyne (3.12)32 were successful. Only the electron rich diyne, 1,4-bis(4’dodecyloxy)buta-1,3-diyne (3.13),33 did not undergo photoaddition to 2.3. All results are summarized in Scheme 3.3, Table 3.2.
40 Scheme 3.3. Photoaddition of butadiynes (3.6, 3.11–3.13) to 2.3. X= S
S
hv
+ S
S
H1 R
S
+ H4
R
3.X
6
H
11
F
R
S
H2
2.3
R=
R
R
H3
N
12 R
13
OC12H25
Table 3.2. Comparison of 1H NMR data of photoproducts. 1
a b
H NMR signals
Diyne
H1a
H2a
H3a
H4a
3.6
4.54 d 4.4
4.72 d 4.4
4.64 s
5.91 s
3.11
4.52 d 4.4
4.63 d 4.4
4.62 s
5.81 s
3.12
4.47 d 4.4
5.12 d 4.4
b
b
3.13
--
--
--
--
Chemical shift d (ppm), multiplicity (d=doublet, s=singlet), coupling constant J (Hz) No corresponding peaks were observed. 3.4 CONCLUSIONS We have shown that the successful photoaddition of dimethyl acetylenedicarboxylate
(3.1) to benzodithiophene (2.3) occured in 16% isolated yield. The photoproduct underwent thermolysis in the gas chromatograph (loss of sulfur). We have discovered a new reaction, the photoaddition of benzo[1,2-b:4,5-b’]dithiophene (2.3) to 1,4-diarylbutadiynes (3.6, 3.11, 3.12). This photoreaction lead in regioselective manner
41 to new, acetylene-substituted cyclobutene derivatives as observed by 1H NMR and X-ray crystal structure analysis.
3.5 X-RAY CRYSTALLOGRAPHY OF 3.734,35 Clusters of colorless, needle-like crystals were obtained from chloroform-pentane solutions. Due to the weakly diffracting nature of the sample and the small crystal size, it was necessary to collect data using synchrotron radiation rather than a lab source. For X-ray examination and data collection, a suitable crystal with the approximate dimensions 0.14 x 0.04 x 0.01mm was mounted on a cryo-loop with paratone-N and immediately transferred to the goniostat bathed in a cold stream. Intensity data were collected at 173K on a Kappa goniostat equipped with a Proteum300 detector at Beamline 11.3.1 at the Advanced Light Source (Lawrence Berkeley National Laboratory) using synchrotron radiation tuned to 14KeV, l=0.88500Å. The detector was set at a distance of 6.6 cm from the crystal. A dataset covering the following scan arcs were collected: 2 q=0º, w=360, 240º, f=0, 90 and 55º. The data frames were processed using the program SAINT.36 However, the final two blocks of data where f=55º were unusable due to severe icing problems encountered during data collection. However, the data that could be integrated were sufficient to give a completeness of 98% to 0.85Å resolution. The data were corrected for decay, Lorentz and polarization effects as well as absorption and beam corrections based on the multiscan technique. The structure was solved by a combination of direct methods SHELXTL v6.1 and the difference Fourier technique and was refined by full-matrix least squares on F2. Non-hydrogen atoms were refined with anisotropic displacement parameters. Weights were assigned as
42 w-1=[s2(Fo2)+ (0.0831P)2 + 0.7535P] where P=0.33333Fo2+0.66667Fc2. All hydrogen atoms were either located directly or calculated based on geometric criteria and treated with a riding model. The isotropic temperature factors for the hydrogen atoms were set as 1.2*Ueq of the adjacent atom. The molecule crystallizes as a hydrate. The water oxygens occupy partial sites (O1:60%; O2A:25%; O2B:15%). The displacement parameter for O2B was set equivalent to the betterbehaved O2A. The water hydrogens were not located or calculated. The disordered phenyl carbons were set at an occupancy ratio of 65:35. The refinement converged with crystallographic agreement factors of R1=5.87%, wR2=16.11% for 2736 reflections with I>2sigma(I) (R1=7.43%, wR2=17.54% for all data) and 310 variable parameters. The crystal data and structure refinement are listed in Table 3.3. Bond lengths and angles are listed in Table 3.4. Selected torsion angles are listed in Table 3.5 and least square planes are listed in Table 3.6. The following tables: atomic coordinates and equivalent isotropic displacement parameters, anisotropic displacement parameters, hydrogen coordinates and isotropic displacement parameters, and least squares planes, can be found at the Cambridge Crystallographic Data Center (CCDC #211794).
Table 3.3. Crystal data and structure refinement. Empirical formula
C26H16S2*H2O
Formula weight
410.52
Temperature
173(2) K
Wavelength
0.88560 Å
Crystal system
Triclinic
Space group
P-1
43 Unit cell dimensions:
a = 4.3109(11) Å
alpha=88.420(14)º
b = 13.779(4) Å
beta=86.848(16)º
c = 17.624(4) Å
gamma=85.523(16)º
Volume, Z
1041.8(5) Å3, 2
Density (calculated)
1.309 Mg/m3
Absorption coefficient
0.270 mm-1
F(000)
428
Crystal size
0.14 x 0.04 x 0.01 mm
Theta range for data collection
3.46 to 31.39º
Limiting indices
-5 < h < 5, -16 < k < 16, -20 < l < 20
Reflections collected
8021
Independent reflections
3449 (Rint = 0.0534)
Max. and min. transmission
0.9973 and 0.9632
Refinement method
Full-matrix least-squares on F2
Data / restraints / parameters
3449 / 0 / 310
Goodness-of-fit on F2
1.067
Final R indices [I>2sigma(I)]
R1 = 0.0587, wR2 = 0.1611
R indices (all data)
R1 = 0.0743, wR2 = 0.1754
Largest diff. peak and hole
0.788 and -0.407 eÅ-3
Table 3.4. Bond lengths [Å] and angles [º]. S(1)-C(1)
1.768(3)
S(1)-C(12)
1.828(3)
S(2)-C(6)
1.733(3)
S(2)-C(5)
1.739(4)
C(1)-C(2)
1.381(4)
C(1)-C(8)
1.414(4)
44 C(2)-C(3)
1.412(5)
C(3)-C(6)
1.411(5)
C(3)-C(4)
1.444(5)
C(4)-C(5)
1.343(6)
C(6)-C(7)
1.403(4)
C(7)-C(8)
1.383(4)
C(8)-C(9)
1.504(4)
C(9)-C(10)
1.536(4)
C(9)-C(12)
1.577(4)
C(10)-C(11)
1.357(4)
C(10)-C(13)
1.419(4)
C(11)-C(31)
1.456(4)
C(11)-C(12)
1.513(4)
C(13)-C(14)
1.202(4)
C(14)-C(21)
1.451(5)
C(21)-C(26B)
1.32(2)
C(21)-C(22B)
1.369(19)
C(21)-C(26A)
1.389(15)
C(21)-C(22A)
1.406(10)
C(22A)-C(23A)
1.41(2)
C(23A)-C(24)
1.42(2)
C(25A)-C(24)
1.352(17)
C(25A)-C(26A)
1.36(2)
C(22B)-C(23B)
1.39(3)
C(23B)-C(24)
1.25(3)
C(25B)-C(24)
1.36(2)
C(25B)-C(26B)
1.42(3)
C(31)-C(32)
1.398(4)
C(31)-C(36)
1.400(4)
C(32)-C(33)
1.389(5)
C(33)-C(34)
1.387(5)
C(34)-C(35)
1.387(5)
C(35)-C(36)
1.379(5)
C(1)-S(1)-C(12)
92.93(14)
C(6)-S(2)-C(5)
91.05(18)
C(2)-C(1)-C(8)
121.7(3)
C(2)-C(1)-S(1)
124.0(2)
C(8)-C(1)-S(1)
114.4(2)
C(1)-C(2)-C(3)
118.4(3)
C(6)-C(3)-C(2)
119.7(3)
C(6)-C(3)-C(4)
111.3(3)
C(2)-C(3)-C(4)
129.1(3)
C(5)-C(4)-C(3)
112.9(3)
45 C(4)-C(5)-S(2)
113.3(3)
C(7)-C(6)-C(3)
121.4(3)
C(7)-C(6)-S(2)
127.1(3)
C(3)-C(6)-S(2)
111.5(2)
C(8)-C(7)-C(6)
118.3(3)
C(7)-C(8)-C(1)
120.5(3)
C(7)-C(8)-C(9)
125.2(3)
C(1)-C(8)-C(9)
114.2(3)
C(8)-C(9)-C(10)
114.3(2)
C(8)-C(9)-C(12)
109.1(2)
C(10)-C(9)-C(12)
85.0(2)
C(11)-C(10)-C(13)
137.0(3)
C(11)-C(10)-C(9)
94.2(2)
C(13)-C(10)-C(9)
128.7(3)
C(10)-C(11)-C(31)
135.3(3)
C(10)-C(11)-C(12)
94.0(2)
C(31)-C(11)-C(12)
130.7(3)
C(11)-C(12)-C(9)
86.7(2)
C(11)-C(12)-S(1)
117.7(2)
C(9)-C(12)-S(1)
108.8(2)
C(14)-C(13)-C(10)
175.2(3)
C(13)-C(14)-C(21)
177.3(4)
C(26B)-C(21)-C(22B)
116.8(14)
C(26A)-C(21)-C(22A)
119.5(8)
C(26B)-C(21)-C(14)
120.8(10)
C(22B)-C(21)-C(14)
121.5(9)
C(26A)-C(21)-C(14)
122.2(7)
C(22A)-C(21)-C(14)
118.3(5)
C(21)-C(22A)-C(23A)
119.9(12)
C(22A)-C(23A)-C(24)
117.8(14)
C(24)-C(25A)-C(26A)
121.3(13)
C(25A)-C(26A)-C(21)
120.5(14)
C(21)-C(22B)-C(23B)
120.0(18)
C(24)-C(23B)-C(22B)
124(2)
C(24)-C(25B)-C(26B)
120.9(18)
C(21)-C(26B)-C(25B)
119.6(19)
C(23B)-C(24)-C(25B)
117.6(15)
C(25A)-C(24)-C(23A)
120.9(10)
C(32)-C(31)-C(36)
118.3(3)
C(32)-C(31)-C(11)
121.5(3)
C(36)-C(31)-C(11)
120.1(3)
C(33)-C(32)-C(31)
120.1(3)
C(34)-C(33)-C(32)
121.1(3)
C(35)-C(34)-C(33)
118.9(3)
C(36)-C(35)-C(34)
120.5(3)
C(35)-C(36)-C(31)
121.1(3)
46
Table 3.5. Selected torsion angles [º]. C(12)-S(1)-C(1)-C(2)
-174.5(3)
C(12)-S(1)-C(1)-C(8)
6.6(2)
C(8)-C(1)-C(2)-C(3)
0.2(4)
S(1)-C(1)-C(2)-C(3)
-178.6(2)
C(1)-C(2)-C(3)-C(6)
-0.8(4)
C(1)-C(2)-C(3)-C(4)
179.2(3)
C(6)-C(3)-C(4)-C(5)
-0.3(4)
C(2)-C(3)-C(4)-C(5)
179.6(3)
C(3)-C(4)-C(5)-S(2)
-0.3(4)
C(6)-S(2)-C(5)-C(4)
0.6(3)
C(2)-C(3)-C(6)-C(7)
1.0(4)
C(4)-C(3)-C(6)-C(7)
-179.1(3)
C(2)-C(3)-C(6)-S(2)
-179.2(2)
C(4)-C(3)-C(6)-S(2)
0.8(3)
C(5)-S(2)-C(6)-C(7)
179.0(3)
C(5)-S(2)-C(6)-C(3)
-0.8(2)
C(3)-C(6)-C(7)-C(8)
-0.5(4)
S(2)-C(6)-C(7)-C(8)
179.7(2)
C(6)-C(7)-C(8)-C(1)
-0.1(4)
C(6)-C(7)-C(8)-C(9)
-176.9(3)
C(2)-C(1)-C(8)-C(7)
0.3(4)
S(1)-C(1)-C(8)-C(7)
179.2(2)
C(2)-C(1)-C(8)-C(9)
177.3(3)
S(1)-C(1)-C(8)-C(9)
-3.8(3)
C(7)-C(8)-C(9)-C(10)
81.9(3)
C(1)-C(8)-C(9)-C(10)
-95.0(3)
C(7)-C(8)-C(9)-C(12)
175.0(3)
C(1)-C(8)-C(9)-C(12)
-1.9(3)
C(8)-C(9)-C(10)-C(11)
106.2(3)
C(12)-C(9)-C(10)-C(11)
-2.4(2)
C(8)-C(9)-C(10)-C(13)
-72.7(4)
C(12)-C(9)-C(10)-C(13)
178.7(3)
C(13)-C(10)-C(11)-C(31)
1.8(6)
C(9)-C(10)-C(11)-C(31)
-176.9(3)
C(13)-C(10)-C(11)-C(12)
-178.7(3)
C(9)-C(10)-C(11)-C(12)
2.5(2)
C(10)-C(11)-C(12)-C(9)
-2.5(2)
C(31)-C(11)-C(12)-C(9)
177.0(3)
C(10)-C(11)-C(12)-S(1)
-112.0(2)
C(31)-C(11)-C(12)-S(1)
67.5(4)
C(8)-C(9)-C(12)-C(11)
-111.8(2)
C(10)-C(9)-C(12)-C(11)
2.2(2)
47 C(8)-C(9)-C(12)-S(1)
6.4(3)
C(10)-C(9)-C(12)-S(1)
C(1)-S(1)-C(12)-C(11)
89.1(2)
C(1)-S(1)-C(12)-C(9)
C(11)-C(10)-C(13)-C(14)
176(4)
C(9)-C(10)-C(13)-C(14)
C(10)-C(13)-C(14)-C(21)
-56(10)
C(13)-C(14)-C(21)-C(26B)
-145(8)
C(13)-C(14)-C(21)-C(22B)
24(8)
C(13)-C(14)-C(21)-C(26A)
-113(8)
C(13)-C(14)-C(21)-C(22A)
66(8)
C(10)-C(11)-C(31)-C(32)
-0.5(5)
C(12)-C(11)-C(31)-C(32)
-179.8(3)
C(10)-C(11)-C(31)-C(36)
178.4(3)
C(12)-C(11)-C(31)-C(36)
-0.8(4)
C(11)-C(31)-C(32)-C(33)
179.9(3)
C(11)-C(31)-C(36)-C(35)
-179.5(3)
Table 3.6. Least squares planes. Least squares planes
Deviation (Å)
S2-C3-C4-C5-C6 (1)
0.004
C1-C2-C3-C6-C7-C8 (2)
0.003
Dihedral(º) 0.8(2)
(2) 3.4(2) S1-C1-C8-C9-C12 (3)
0.039
(3) 65.1(1) C9-C10-C11-C12 (4)
0.016
(4) 67.4(1) (12)
0.038
where (12) = S1-C1-C2-C3-C4-C5-S2-C6-C7-C8-C9-C12
120.36(19) -7.3(2) -6(4)
48 (4) 8.6(4) C21-C22A-C23A-C24-C25A-C26A (5)
0.011
(4) 30.3(7) C21-C22B-C23B-C24-C25B-C26B (6)
0.041
(4) 2.7(2) C31-C32-C33-C34-C35-C36 (7)
0.003
(5) 9.9(4) (7) (6) 27.8(7) (7) 3.6 EXPERIMENTAL SECTION All photochemical reactions were carried out in NMR tubes (f 5mm; cut-off 300nm) purged with dry argon. Photoreactions were carried out for 16 hrs in a Rayonet photochemical reactor using 16 lamps (300 nm, 21 Watts/each). Spectroscopy-grade benzene was dried by azeotropic distillation and stored over molecular sieves under argon. Benzo[1,2-b:4,5-b’]dithiophene (2.3). See Chapter 2. Dimethyl cyclobuta[b]thieno[2,3-f][1]benzothiophene-5a,6[7aH]-dicarboxylate (3.3). Yield: 16%. GC–MS m/z (%): 300 (92) [M – 32], 268 (100), 254 (45), 182 (53), 170 (36), 138 (31).1H NMR (400 MHz, CDCl3) d 3.78 (3H, s), 3.85 (3H, s), 4.89 (1H, s), 6.90 (1H, d, J = 1.2 Hz), 7.20 (1H, d, J = 5.4 Hz), 7.40 (1H, d, J = 5.4 Hz), 7.58 (1H, s) 7.70 (1H, d, J = 0.8 Hz), 13C NMR (100 MHz, CDCl3) d 170.2, 165.1, 160.8, 146.6, 140.8, 139.2, 138.5, 137.2, 130.4, 127.8, 123.2,
49 119.0, 117.4, 59.4, 53.6, 52.2. Anal. Calcd for C16H12S2O4: C, 57.81; H, 3.64; S, 19.29. Found: C, 57.62; H, 3.76; S, 19.44. Mp = 154–155 °C.
Figure 3.11. 1H NMR spectrum of dimethyl cyclobuta[b]thieno[2,3-f][1]benzothiophene5a,6[7aH]dicarboxylate (3.3).
50
Figure 3.12. APT (upper) and 13C (lower) NMR spectra of dimethyl cyclobuta[b]thieno[2,3f][1]benzothiophene-5a,6[7aH]dicarboxylate (3.3). Diphenylbuta-1,3-diyne (3.6). Mp. 86-87 °C (Lit.37 87 °C), purchased and used as received. 6-Phenyl-7-(phenylethynyl)-5a,7a-dihydrocyclobuta[b]thieno[2,3-f][1]benzothiophene (3.7). Yield: 4%. 1H NMR (400 MHz, CDCl3) d 4.97 (1H, d, J = 4.4 Hz), 5.15 (1H, d J = 4.4 Hz), 7.16 (1H, d, J = 5.5 Hz), 7.33-7.42 (7H, m), 7.51-7.54 (3H, m), 7.71-7.73 (2H, m), 7.85 (1H, d, J = 0.8 Hz); 13C NMR (100 MHz, CDCl3) d 184.6, 140.5, 140.2, 137.0, 133.6, 132.4, 132.2, 129.4, 129.2, 128.9, 128.8, 127.1, 126.6, 123.2, 123.0, 122.8, 119.1, 117.7, 97.4, 83.9, 57.4, 49.0. Crystallization from CHCl3/pentane gave white needles, mp 202 °C. MS. m/z (%): 392 (12) [M], 358 (3), 290 (11), 202 (6), 190 (100). Anal. Calcd for C26H16S2*1H2O: C, 76.06; H, 4.42. Found: C, 76.02; H, 4.08; HRMW (C26H16S2): Calcd. Mass: 392.069644 Found Mass: 392.069408.
51
Figure 3.13. 1H NMR spectrum of 6-phenyl-7-(phenylethynyl)-5a,7adihydrocyclobuta[b]thieno[2,3-f][1]benzothiophene (3.7).
52
Figure 3.14. 13C NMR spectrum of 6-phenyl-7-(phenylethynyl)-5a,7adihydrocyclobuta[b]thieno[2,3-f][1]benzothiophene (3.7). Bis(4’-fluorophenyl)buta-1,3-diyne (3.11). The title compound was synthesized according to a literature procedure.38,39 Mp. 190-191 °C. 1H NMR (400 MHz, CDCl3): d 7.01-7.06 (4H, t), 7.497.53 (4H, dd, J1=8.8 Hz, J2=2.4 Hz). 13C NMR (100 Hz, CDCl3) d 73.85, 80.74, 116.11-116.33, 118.12-118.15, 134.81-134.89, 164.63. Bis(2’-pyridyl)butadiyne (3.12). The title compound was synthesized according to a literature procedure.31 Mp. 118-119 °C. 1H NMR (400 MHz, CDCl3): d 7.28-7.32 (2H, ddd), 7.53-7.56 (2H, dd), 7.67-7.71 (2H, td), 8.62-8.63 (2H, dd). 124.05, 128.67, 136.46, 142.18, 150.68.
13
C NMR (100 Hz, CDCl3) d 73.43, 81.20,
53 Bis(4’-dodecyloxy)buta-1,3-diyne (3.13). The title compound was synthesized according to a literature procedure.32 Mp. 79-85 °C. 1H NMR (400 MHz, CDCl3): d 0.89-0.92 (12H, t), 1.331.34 (16H, m), 1.43-1.49 (8H, p), 1.77-1.85 (8H, m), 3.96-3.98 (4H, t), 3.99-4.02 (4H, t), 6.786.80 (2H, d, J= 8.4Hz), 7.00-7.01(2H, d, J= 1.6Hz), 7.01-7.10(2H, dd, J1= 8.4Hz, J2= 1.6 Hz). 13
C NMR (100 Hz, CDCl3) d 14.33, 22.91, 25.96, 29.41, 29.43, 31.87, 69.41, 69.56, 72.95, 81.92,
113.37, 114.10, 117.48, 126.42, 148.98, 150.80.
3.7 REFERENCES (1)
Wynberg, H. Acc. Chem. Res. 1971, 4, 65-73.
(2)
Pomerantz, M.; Wang, J. P.; Seong, S.; Starkey, K. P.; Nguyen, L.; Marynick, D. S. Macromolecules 1994, 27, 7478-7485.
(3)
Schopf, G.; Kossmehl, G. Polythiophenes - Electrically Conductive Polymers; Springer: Berlin, Germany, 1997; Vol. 129.
(4)
Handbook of Oligo- and Polythiophene; Fichou, D., Ed.; Wiley-VCH: Weinheim, Germany, 1999.
(5)
Inganaes, O.; Berggren, M.; Andersson, M. R.; Gustafsson, G.; Hjertberg, T.; Wennerstroem, O.; Dyreklev, P.; Granstroem, M. Synth. Met. 1995, 71, 2121-2124.
(6)
Higashiguchi, K.; Matsuda, K.; Yamada, T.; Kawai, T.; Irie, M. Chem. Lett. 2000, 13581359.
(7)
Higashiguchi, K.; Matsuda, K.; Asano, Y.; Murakami, A.; Nakamura, S.; Irie, M. Eur. J. Org. Chem. 2004, 91-97.
(8)
Wynberg, H.; Kellogg, R. M.; van Driel, H.; Beekhuis, G. E. J. Am. Chem. Soc. 1966, 88, 5047-5048.
54 (9)
Kuhn, H. J.; Gollnick, K. Chem. Ber. 1973, 106, 674-696.
(10)
Geerts, Y.; Klärner, G.; Müllen, K. In Electronic Materials: The Oligomer Approach; Müllen, K., Wegner, G., Eds.; Wiley-VCH: Weinheim, Germany, 1998, p 1-103.
(11)
Campbell, S. J.; Kelly, K. C.; Peacock, T. E. Aust. J. Chem. 1989, 42, 479-488.
(12)
Reinhoudt, D. N. Adv. Heterocycl. Chem. 1977, 21, 253-321.
(13)
Dopper, J. H.; Neckers, D. C. J. Org. Chem. 1971, 36, 3755-3762.
(14)
Neckers, D. C.; Dopper, J. H.; Wynberg, H. Tetrahedron Lett. 1969, 2913-2916.
(15)
Ditto, S. R.; Davis, P. D.; Neckers, D. C. Tetrahedron Lett. 1981, 22, 521-521.
(16)
Hofmann, H.; Westernacher, H.; Haberstroh, H.-J. Chem. Ber. 1969, 102, 2595-2602.
(17)
Serve, M. P.; Rosenberg, H. M. J. Org. Chem. 1970, 35, 1237-1237.
(18)
Arnold, D. R.; Chang, Y. C. J. Heterocycl. Chem. 1971, 8, 1097-1098.
(19)
Owsley, D. C.; Bloomfield, J. J. J. Am. Chem. Soc. 1971, 93, 782-784.
(20)
Shim, S. C.; Lee, T. S.; Lee, S. J. J. Org. Chem. 1990, 55, 4544-4549.
(21)
Shim, S. C.; Kim, S. S. Bull. Korean Chem. Soc. 1985, 6, 153-157.
(22)
Chung, C. B.; Kim, G. S.; Kwon, J. H.; Shim, S. C. Bull. Korean Chem. Soc. 1993, 14, 506-510.
(23)
Lee, S. J.; Shim, S. C. Tetrahedron Lett. 1990, 31, 6197-6200.
(24)
Shim, S. C. K., S. S. Tetrahedron Lett. 1985, 26, 765-766.
(25)
Shim, S. C.; Lee, S. J.; Kwon, J. H. Chem. Lett. 1991, 1767-1770.
(26)
Ryashentseva, M. A.; Belanova, E. P.; Minachev, K. M.; Polosin, V. M.; Bodganov, V. S. Bull. Acad. Sci. USSR Div. Chem. Sci. 1988, 37, 2579.
(27)
Nagano, Y.; Ikoma, T.; Akiyama, K.; Tero-Kubota, S. J. Chem. Phys. 2001, 114, 17751784.
55 (28)
Phosphorescence emission spectra (excitation wavelength = 302 nm) were recorded in ethyl iodide/ethanol/isopentane (1/2/1) at 77 K (liquid nitrogen).
(29)
Nagano, Y.; Ikoma, T.; Akiyama, K.; Tero-Kubota, S. J. Phys. Chem. 1998, 102, 57695774.
(30)
Baughman, R. H. J. Appl. Phys. 1972, 43, 4362-4370.
(31)
Rodriguez, J. G.; Martin-Villamil, R.; Cano, F. H.; Fonseca, I. J. Chem. Soc., Perkin Trans. 1 1997, 709-714.
(32)
Kitamura, T.; Lee, C. H.; Taniguchi, Y.; Fujiwara, Y. J. Am. Chem. Soc. 1997, 119.
(33)
Birks, J. B. Photophysics of Aromatic Molecules; Wiley: New York, 1970.
(34)
Samples for crystallographic analysis at the synchrotron were submitted through the SCrAPS-West (Service Crystallography at Advanced Photon Source) program. Crystallographic data was collected at the Small-Crystal Crystallography Beamline 11.3.1 at the Advanced Light Source (ALS) that was developed by Albert Thompson of the Experimental Systems Group of the ALS. The ALS is supported by the U.S. Department of Energy, Office of Basic Energy Sciences under contract DE-AC0376SF00098.
(35)
SMART v5.625 and SAINT v6.06 programs were used for data collection and data processing, respectively. Bruker Advanced X-ray Solutions, Inc., Madison, WI. SADABS was used for the application of semi-empirical absorption and beam corrections. G. M. Sheldrick, University of Göttingen, Germany. SHELXTL v6.1 was used for the structure solution and generation of figures and tables. G. M. Sheldrick, University of Göttingen, Germany and Bruker Advanced X-ray Solutions, Inc., Madison, WI. Neutral-atom scattering factors were used as stored in this package.
56 (36)
SMART v5.625 and SAINT v6.06 programs were used for data collection and data processing, respectively. SADABS was used for the application of semi-empirical absorption and beam corrections. SHELXTL v6.1 was used for the structure solution and generation of figures and tables. Neutral-atom scattering factors were used as stored in this package.
(37)
Coates, G. W.; Dunn, A. R.; Henling, L. M.; Dougherty, D. A.; Grubbs, R. H. Angew. Chem. Int. Ed. Engl. 1997, 36, 248-251.
(38)
Salah, B. Aust. J. Chem. 1976, 29, 531.
(39)
Bevan, W. I. J. Chem. Soc. Dalton Trans. 1974, 2305-2309.
57 CHAPTER IV. PREPARATION OF ORGANIC FIELD-EFFECT TRANSISTORS (OFET) USING THIENO[F,F’]BIS[1]BENZOTHIOPHENE (SYN, ANTI)
4.1 INTRODUCTION A field-effect transistor (FET) is a device, wherein an applied electric field controls the conductivity of a semiconducting material. Generally, the active layer of such a device is a single crystal inorganic semiconductor, such as silicon. A subclass of the FET is the thin-film transistor (TFT), which is composed of a sandwich structure of deposited dielectric, semiconducting layer, and metal electrodes. TFTs are key components of modern flat-panel displays. The invention of the TFT dates back to conceptual patents of the 1930s by Julius E. Lilienfeld,1 who described a device with solid layers of different conductivity, such as cooper sulfide, aluminum oxide and aluminum, as depicted in Figure 4.1. O. Heil2 then followed with a definition of the term “gap” to describe p-type semiconductors. William B. Shockley3 finally branded the terminology of “source” (“drain”) as the positive (negative) terminal of a hole channel device.4 A practical realization of TFTs came with the seminal work of P.K. Weimar at RCA, who built a top-gate TFT based on a thin-film of polycrystalline cadmium sulfide,5 leading ultimately to today’s high-tech device industry based on low temperature Si-TFT and the LCD and emerging OLED market.
Figure 4.1. Lilienfeld's device for controlling electric current. The first OFETs go back to the 1980s.6-10 Figure 4.2 depicts two schematic drawings of the bottom gate OFET devices in top-contact and a bottom-contact geometry. The relative youth
58 of this field is reflected in the fact that device characterizations were just recently standardized when the Institute of Electrical and Electronic Engineers released Standard 1620™: Standard Test Methods for Characterization of Organic Transistors and Materials in 2004.11 The development of high performance organic semiconductors for OFETs is the stepping stone toward flexible plastic electronics,12,13 where research has primarily focused on the development and optimization of devices based on pentacene and its derivatives. Crucial features of organic semiconductors are high field-effect mobility, high on-off current ratios and low threshold voltages. Pentacene, for instance, exhibits a field-effect mobility of 35 cm2/Vs when used as ultra-high purity single crystals at room temperature.14
V=Vs
V=VD
Source
Drain
Organic Semiconductor Gate Insulator Gate
V=VG
V=Vs
V=VD
Organic Semiconductor Source Drain Gate Insulator Gate
V=VG
Figure 4.2. Schematic overview of a top-contact (left) and bottom-contact (right) design for OFET. Similar to benzene in aromaticity, yet superior in terms of intrinsic conductivity, solubility and richness of synthetic versatility, thiophenes and their oligo- and polymeric derivatives have long been used as organic semiconductors.15-19 Low-molecular weight oligomers thereby display the advantages of potentially high molecular order (defect-free molecules), high chemical purity and high structural order in the solid state. Recent advances
59 include new semiconducting organic materials such as hybrid acene-thiophenes,20,21 embedded thiophenes,22 and acenedithiophenes.23,24 Therein, the preparation of embedded thiophenes has posed synthetic challenges to achieve isomeric purity (see Chapter 2).25
4.2 DESIGN STRATEGIES FOR NEW MATERIALS We designed two new materials as potential new p-type organic semiconductors composed of alternating thiophene and benzene units. The design of these materials addresses the crucial problem of oxidative stability, a major drawback in oligoacene materials. It is known that large polycyclic aromatics undergo cycloaddition reactions in solution when exposed to visible light in presence of oxygen.26 Pentacene, for instance, is rapidly photooxidized.27 The mechanism of the photoreaction28 is depicted in Scheme 4.1. G, S, and T represent the ground, singlet and triplet excited state, respectively. M and M* represent the ground and excited state of the molecule. M(G)
hv ⎯⎯ → M * (S) ⎯ ⎯→ M * (T)
M * (T) + O2 (G,
3
∑
M(G) + O2 (S,1∆ g )
− g
) ⎯ ⎯→ M(G) + O2 (S,1∆ g ) ⎯ ⎯→ MO2
Scheme 4.1. Mechanism of endoperoxide formation. These endoperoxides can then undergo cycloreversion or cleavage. The latter leads effectively to degradation of the material.29 The photooxidation of polycyclic aromatics has been studied in the solid state.30 Photooxidation occurs primarily on the surface. Dependent on the crystal packing, photooxidation can also occur in the bulk of the material. In extended oligoacenes, reactivity increases toward the center of the molecule. Using the cycloaddition of benzyne as an analogous model reaction, it can be seen that anthracene undergoes a [4π + 2π]
60 cycloaddition reaction at positions 5 and 10 yielding a triptycene.31 Replacing just one of the terminal benzene rings with a thiophene unit effectively shuts down this reaction pathway (Scheme 4.2).32,33 Scheme 4.2. Cycloadditions to anthracene (4.1), naphtho(b)thiophene (4.4) and benzothiophene (4.5).
4.2
4.1
4.3 4.2
S
no reaction
4.4 S
4.2
S
no reaction
4.5 The design of our target molecules thieno[2,3-f:5,4-f’]bis[1]benzothiophene (2.5) and thieno[3,2-f:4,5-f’]bis[1]benzothiophene (2.6) depicted in Figure 4.3 included terminal thiophene rings to increase solubility and a central thiophene ring to stabilize the material against oxidative decomposition. The synthesis of 2.5 and 2.6 was described in Chapter 2. S
S S
S S
S
2.5
2.6
Figure 4.3. Molecular structure of new semiconductors.
61 4.3 X-RAY CRYSTALLOGRAPHY OF 2.5 AND 2.6 Compound 2.5 crystallized in an orthorhombic space group Pmn21 with two molecules in the unit cell. The central atom S2 resides on the crystallographic mirror plane that relates the two halves of the molecule. The distance between the central S atoms is 3.89 Å; whereas the shortest intermolecular S-S distance is 3.49 Å and, therefore, shorter than the sum of the van der Waals radii. Molecules of 2.5 are nearly planar with the highest deviation of atoms from the best plane through the entire molecule of 0.174(2) Å. The crystals of 2.5 exhibit a herringbone-packing pattern with favorable molecular overlap along the c-axis of the unit cell (Figure 4.4). The alternating π-stacked columns are tilted at an angle of 50.24° (viewed down a-axis, Figure 4.5). In comparison, this tilt-angle is 51.9° in pentacene.34
Figure 4.4. Solid-state packing of 2.5 as viewed along the c-axis. All hydrogens are omitted.
62
Figure 4.5. Crystal packing as observed along the a-axis. All hydrogens are omitted. Tight packing in the solid state increases the material’s stability toward oxidation in that it decreases the quantity of oxygen diffusing into the bulk material.30 The effective volume occupation is expressed as the Kitaigorodskii packing index (KPI).35,36 Compound 2.5 has a packing coefficient of 0.75 compared to thieno[3,2-e:4,5-e’]bis[1]benzothiophene (4.6)37 with 0.72; whereas pentacene (4.7) shows with a KPI of 0.76,37 the highest percent of filled space (Figure 4.6). S
S
S
4.6
4.7
Figure 4.6. Molecular structure of 4.6 and 4.7. Crystal structure analysis of 2.6 was attempted. The unit cell parameters were a=7.796 Å, b=5.966 Å, and c=27.337 Å with alpha=90.03°, beta=97.42°, gamma=90.05° (V=1261;
63 T=150K). The acquired data could not be refined to a satisfactory value. The preliminary data indicated an edge-to-face stacking of the molecules in the solid state as depicted in Figure 4.6.
Figure 4.7. Edge-to-face packing as observed for the solid-state of 2.6.
4.4 THIN-FILM X-RAY DIFFRACTION ANALYSIS OF MATERIALS 2.5 AND 2.6 The sharp and few diffraction peaks of both 2.5 and 2.6 indicate a high crystallinity of the thin-films as deposited onto an OTS treated p-Si wafer. In this symmetric reflection mode, the diffraction vector is perpendicular to the film plane and therefore only probes planes which lie parallel to that plane.38 Using the powder pattern simulation from single-crystal structure of 2.5, which is orthorhombic, space group Pmn21, with a=25.68 Å, b=6.08 Å, c=3.89 Å, α=β=γ =90º, the refection peaks were indexed as (h00) reflections where h is an even number (due to symmetry restriction of screw axis). The room temperature thin-film X-ray diffractogram of 2.5 shows reflections up to third order, where the primary diffraction peak (200) is at 2θ = 6.84° (dspacing of 12.91 Å) and the second and third order peak at 2θ = 13.66° (d-spacing of 6.47 Å) and 2θ = 20.52° (d-spacing of 4.32 Å), respectively. The presence of (h00) planes and absence of
64 any other reflection indicate that crystals in the film are oriented with h00 plane or the bc plane parallel to the substrate surface. The observed d-spacing of 12.91 Å corresponds closely to the length of the molecule as obtained from single-crystal analysis (12.70 Å) and therefore confirms that the molecules orient themselves with the long axis perpendicular to the device surface (Figure 4.8, Figure 4.9). For 2.6, the X-ray diffractogram consists of a series of sharply resolved peak assignable to multiple (h00) reflections up to the 8th order, suggesting a highly ordered crystal packing of the thin-film. The primary diffraction peak is at 2θ = 6.64° (d-spacing of 13.29 Å) with higher order peaks at 2θ = 13.17° (d-spacing of 6.72 Å), 2θ = 19.73° (d-spacing of 4.50 Å), 2θ = 26.36° (d-spacing of 3.38 Å), 2θ = 33.09° (d-spacing of 2.71 Å), 2θ = 39.94° (d-spacing of 2.26 Å), 2θ = 46.94° (d-spacing of 1.93 Å), and 2θ = 54.11° (d-spacing of 1.70 Å). These reflections were indexed as (h00) where h =2, 4, 6, 8, 10, 12, 14, and 16, respectively, of an orthorhombic unit cell (see below) with a= 27.08 Å, b=6.80 Å, c=4.78 Å, α=β=γ =90º. No single crystal data for structure and molecular packing was available. We have therefore assumed that the observed d-spacing of the first order diffraction indicates molecules that are perpendicular to the substrate surface. In both cases, a definite proof of preferred orientation would require either transmission XRD analysis or synchrotron grazing incidence angle X-ray diffraction studies with in-plane and out-of-plane scattering geometry.
65
Figure 4.8. X-ray diffraction pattern of 2.5 (1) and 2.6 (2).
Figure 4.9. Model of the alignment of the molecules 2.5 along the device plane = bc molecular plane.
66 4.5 CYCLIC VOLTAMMETRY Electrochemical measurements were carried out using an Epsilon Electrochemical Workstation from Bioanalytical Systems in cyclic voltammetry and differential pulse voltammetry mode. Dichloromethane (DCM) was freshly distilled from CaH prior to use. A standard electrochemical cell with embedded platinum auxiliary electrode, glassy carbon working electrode and non-aqueous reference electrode was used to carry out the experiments. The electrolyte tetra-nbutylammonium hexafluorophosphate (TBAPF) was recrystallized from ethanol and dried under a vacuum for 16 hours prior to use. A 0.1 M solution of TBAPF in DCM was used as supporting electrolyte. Ag/AgNO3 (0.01 M in 0.1 M TBAPF in acetonitrile) served as non-aqueous reference electrode. All experiments were carried out at a scan rate of 100 mV/s. Ferrocene (Fc) was used as internal standard with a half-wave potential of 0.276 V under these reaction conditions. All potentials are reported vs. Fc. The potential of Fc vs. SCE is 0.46 V.39 Differential pulse voltammetry was used to determine the oxidation potentials. In differential pulse voltammetry the half-wave potential is given by Equation 4.1, wherein ∆E is the pulse amplitude. E1/ 2 = E p +
∆E (Equation 4.1) 2
Compound 2.5 exhibited a quasi-reversible oxidation at 0.874 V, while 2.6 exhibited a reversible oxidation at a half wave potential of 0.844 V vs. Fc, respectively. Compound 2.5 formed an intensely colored film on the electrode surface, and the gradual appearance of lower potential waves upon repeated cyclization indicated electropolymerization of the radical cations that have been generated.
67
Figure 4.10. Cyclic voltammogram of 2.5.
Figure 4.11. Differential pulse voltammogram of 2.5.
68
Figure 4.12. Cyclic voltammogram of 2.6.
Figure 4.13. Differential pulse voltammogram of 2.6. Table 4.1. Redox potentials (vs. Fc) of compounds 2.5 and 2.6 in comparison to 4.7. 2.5 2.6 4.7 E1/2(Ox, V)
0.874
0.844
0.2240
69
4.6 THERMOGRAVIMETRIC ANALYSIS Thermogravimetric analysis (TGA) was carried out in collaboration with Mike Zemo and Bowen Greil using a TGA/SDTA851e in the laboratory of Mettler Toledo in Columbus, OH. The stability of 2.5 and 2.6 was determined under air and nitrogen atmospheres at a heating rate of 10 °C per min. Under nitrogen, the onset temperature for decomposition was highest for 2.6 reaching 400 °C. Both isomers exhibited comparable stability under air and started to decompose around 340 °C (Table 4.2). This constitutes an increase of 20 °C to the reported decomposition temperature of 319 °C of 4.7.20
Figure 4.14. Thermogravimetric analysis curve of 2.5.
70
Figure 4.15. Thermogravimetric analysis curve of 2.6. Table 4.2. Thermal parameters of compounds 2.5 and 2.6 compared to pentacene (4.7). T on (N2, ºC) % (dec.) T on (air, ºC) % (dec.) 2.5
375.04
99.82
344.23
94.46
2.6
400.91
N/A
341.58
86.41
4.7
339
N/A
319
N/A
4.7 GAS-PHASE PHOTOELECTRON SPECTROSCOPY This technique is based on the photoelectric effect discovered by Hertz in 1887,41 demonstrated by Thompson42 and explained by Einstein.43 The energy of light with sufficient energy and matter leads to ionization events (Scheme 4.3).
71 Scheme 4.3. Formula description of photoelectric effect. hv + M → M + + e− + leftover energy E e − = E hv − ionization energy The time-scale of the photo-ionization event is 10-15 to 10-18 s. At this time-scale only Franck-Condon transitions to discrete vibrational levels of the ion state are observed under application of the Born-Oppenheimer approximation. The second assumption is that molecules reside in their ground state in the lowest vibrational level vo. We carried out gas-phase photoelectron spectroscopy for compounds 2.5 and 2.6. These results were then compared to the values of pentacene 4.7 and anthradithiophene 4.8 (Figure 4.16).44 S S
4.8 Figure 4.16. Molecular structure for anthradithiophene (4.8). The gas-phase photoelectron spectrum of 2.5 and 2.6 (Figure 4.17, Table 4.3) were collected using the instrument and experimental procedures reported in more detail elsewhere.45 Both isomers sublimed at 190-240 ºC. There was no evidence of contaminants present in the gas phase during data collection. The instrument resolution during data collection was better than 35 meV (measured using the full-width-at-half-maximum for the 2P3/2 ionization of Ar).
72
Figure 4.17. Photoelectron spectra of first ionization of 2.5 (bottom) and 2.6 (top). Table 4.3. Vertical adiabatic ionization energy (IP). Compound
IP [eV]
2.5
7.43
2.6
7.36
4.746
6.58
4.847
6.69
A polarization energy of typically 1.5 – 2.0 eV must be taken into account when adjusting gas-phase IP values to solid-state values.48 For example, the measured solid-state ionization energy of 4.7 is 4.85 eV.49 The overall goal is to match the ionization potential of the semiconductor closely to that of the gate dielectric and the contact electrodes. Typical values for work functions of commonly used metals are 4.6-4.91 eV for p-Si, 4.2 eV for Al, 5.2 eV for Au,
73 5.7 for Pt.50 It becomes apparent that the ionization potential of both 2.5 and 2.6 is at or above the level of Au.
4.8 MATERIALS CHARACTERIZATION IN OFET DEVICES 4.8.1. DEVICE CONSTRUCTION The semiconductors 2.5 and 2.6 were evaporated onto heavily p-doped silicon wafers, either with 100 nm or 300 nm of thermally grown SiO2. To improve the morphology of the materials, the silicon dioxide surface was coated with octadecyltrichlorosilane (OTS). Coating was carried out in a chilled cyclohexane/OTS solution into which the silicon substrates were dipped. Thereafter, the substrates were washed in pure cyclohexane to wash away any OTS that had not reacted with the hydroxy groups of the oxide. Compounds were evaporated onto substrates heated to 75 °C in a vacuum of 10-6 Torr. The substrate temperature, in conjunction with a slow evaporation rate of 1 Å/s, was chosen to ensure best growth conditions of the compounds on the surfaces. In fact, no field-effect current was observed for 2.5 without heating the substrates, and only a very reduced mobility was measured for 2.6, thus highlighting the importance of the deposition parameters. Subsequently, gold, defining the source/drain electrodes, was evaporated on top of the semiconductor through shadow masks. The channel was 25 µm wide and 2000 µm long for a width to length ratio of 80. After the evaporation, the devices were stored and measurements made under an atmosphere of nitrogen. The gate and the drain voltages were applied using two Keithley 2400 source measurement units, which also measured the drain current as well as the leakage current of the transistors.
74 4.8.2 TOPOGRAPHICAL IMAGES OF THIN-FILMS AFM topographical images were acquired on a Burleigh Vista-100 Scanning Probe Microscope in AC (non-contact) mode at Spectra Group Ltd, Millbury, OH. Under the chosen deposition conditions, both semiconductors formed granular surfaces, and compound 2.5 exhibited grains that were elongated up to µm size and spatially separated. In contrast, the grains of 2.6 were near-regular in shape and had an average size of 0.4 µm (Figure 4.18). The observed low mobility of 2.5 might be the result of these observed grain barriers, and an optimization of the deposition conditions might lead to improved mobility.
Figure 4.18. AFM topographical images of thin-films of 2.5 (left) and 2.6 (right) on pSi/SiO2/OTS evaporated at Tsub = 75 °C. 4.8.3 DEVICE CHARACTERIZATION The output characteristic (drain current versus drain voltage, while keeping the gate voltage constant) of transistors using 2.5 and 2.6 as the semiconductor is shown in Figure 4.19. It is clearly visible that the currents of 2.5 are about five times higher than 2.6. However, at the same time, a strong hysteresis is seen in the output characteristic of 2.6. During the sweep of the drain voltage, the gate voltage is held constant. It appears that the current is diminished with increasing drain current when the gate voltage is not varied. This effect is typically called gate stress and can be caused by many things. The effect is not normally, however, as clearly visible in the output characteristic. It appears that a significant amount of charge carriers in the so-called
75 “accumulation layer” stop responding to the drift field introduced by the drain voltage – potentially through a trapping process. To discuss this phenomenon in more detail, the transistor characteristics need to be looked at from the perspective of the transfer characteristic.
-3,000
-700
- 65 V -600 -2,500
- 150 V
Source-Drain Current, ISD [nA]
Source-Drain Current, ISD [nA]
-500 - 60 V
-2,000
-1,500
- 55 V
- 140 V
-400
-300
- 130 V
-1,000 -200
-500
- 50 V
-100 - 120 V
- 45 V 0
- 110 V
0
0
-5
-10
-15
-20
-25
Source-Drain Voltage, VSD [V]
-30
-35
-40
0
-5
-10
-15
-20
-25
-30
-35
-40
Drain Voltage [ V]
Figure 4.19. Output characteristic of compound 2.5 on a 100 nm (left) and 2.6 on a 300 nm (right) SiO2 gate insulator.
-45
76 1x10-4
1x10 -5
1x10-5
1x10
Source-Drain Current, ISD [A]
Source-Drain Current, ISD [A]
1x10 -6
-7
1x10 -8
1x10-6
1x10-7
1x10-8
1x10 -9 -9
1x10
-10
1x10
-80
1x10 -70
-60
-50
-30 -40 -20 Gate Voltage, VG [V]
-10
0
10
-10
20
-160
-140
-120
-100
-60 -80 -40 Gate Voltage, VG [V]
-20
0
20
40
Figure 4.20. Transfer characteristics of compound 2.5 on a 100 nm (left) and 2.6 on a 300 nm (right) SiO2 gate insulator. Table 4.4. Performance of 2.5 and 2.6 in a top-contact OFET device. Compound
Ci
S
Mobility
[nF/cm2]
[V/dec]
[cm2/Vs]
2.5
32.1
2
0.011
4 x 104
-42.0
2.6
11.3
15
0.12
1.6 x 105
-114.6
Ion/Ioff
VT [V]
The transfer characteristic shows the drain current responding to a change in the gate voltage. Due to the faster change of the gate voltage, it is easier to measure meaningful data that is not affected as much by the gate stress. The transfer characteristics for 2.5 and 2.6 are shown in Figure 4.20. From the transfer characteristic, the on/off ratio, the threshold voltage, and also the mobility are collected (Table 4.5). In comparison, anthradithiophene, which could be synthesized only as a mixture of syn and anti isomers, exhibits a highest observed FET mobility of 0.09 cm2/Vs in a bottom geometry (Tsub = 85 °C).23 Both semiconductors exhibit a good on/off
77 ratio of 104-105. The threshold voltage and the mobility are extracted from the data in Figure 4.20 by fitting the current with Equation 4.2.51
I SD = µ
W ε 0k (VG − VT )2 2l ti
(Equation 4.2)
Therein, ISD is the current between source and drain, W is the width of the channel, l is the length of the channel, ε0*k is the dielectric constant of the gate insulator, ti is the thickness of the insulator, VG is the applied gate voltage, and VT is the gate threshold voltage. The mobility plots are depicted in Figure 4.21. 100
300
90
250
80
70
] 1/2
Sqr(ISD ) [(nA)
Sqr(ISD ) [(nA)
1/2
]
200
60
50
150
40 100 30
20 50
10
0
0
0
10
20
30
40 VG = VSD
50
60
70
80
0
20
40
60
80
100
120
140
160
180
VG = VSD
Figure 4.21. Mobility plot of 2.5 (left) and 2.6 (right). During the device sweeping, a phenomenon called gate stress is observed that is associated with the increase of the threshold voltage to higher absolute voltages.52 In physical terms it means that the flat band voltage has changed, due to one or more of the following occurrences:
200
78 •
(Semi-) permanent polarization of the gate insulator53
•
Charge formation at the semiconductor/insulator interface53
•
Formation of bipolarons in polymer semiconductors54,55
•
Shift of the injection barrier, and/or
•
Irreversible doping/oxidation of the semiconductor. Inverting the gate voltage of the transistor can usually reverse all but the last two of the
aforementioned causes. In the transistors of 2.5 and 2.6, however, the threshold voltage stayed the same after the initial ramping, even when the gate voltage is inversed to “de-stress the gate”. Also, it can be seen that the threshold voltage increases even further if the current is ramped higher. Figure 4.20 (right) depicts a transistor using compound 2.6 on 300 nm of a silicon oxide gate insulator, which allows gate voltage sweeps up to -160 V. It can be seen that the shift of the threshold voltage is limited only by the final sweep voltage. When the maximum voltage is no longer being increased, the threshold voltage remains constant as well. Since the shift described is irreversible, this would initially point towards irreversible oxidation. It should be noted, however, that conversely to the oxidation of polythiophene that increases the amount of charge carriers (and thus the off current), the off-currents for both compounds 2.5 and 2.6 stay low even after repeated cycling. It, therefore, appears that the observed effect cannot be described through oxidation of the compounds. As a final option, the shift of the injection barrier, or rather, the shift of the flat band voltage remains the only explanation. The exact mechanism is not yet understood, but the most likely candidates for the mechanism are charge carrier trapping at the interface of the compounds with the gold electrode or with the gate insulator. Gold is a most inert metal, and although it can influence the electronic structure of nearby organic molecules, no change in this influence due to
79 device sweeping is known. It is, however, known that organic molecules can have slightly altered performance characteristics depending on the gate insulator that they are deposited upon. Usually, this trapping can be at least partially reversed by the application of high counter voltages, which does not seem to be the case here. Further research into the physical and chemical nature of the two compounds presented here and possible gate insulators is therefore of high interest.
4.9 CONCLUSIONS In conclusion, we have successfully developed two new p-type organic semiconducting materials 2.5 and 2.6. Both feature competitive field-effect mobilities and on/off ratios. Due to the lower threshold voltage, material 2.5 is the more promising candidate for further research. Suitable substitutions on the terminal thiophene units can be used out to raise the ionization potential, while optimization of the deposition conditions might improve the thin-film properties and thereby the effective mobility.
4.10 EXPERIMENTAL PARAMETERS FOR X-RAY CRYSTAL STRUCTURE ANALYSIS X-ray crystal structure analysis was carried out on both compounds.56 For compound 2.5, a pale-yellow, plate-like crystal was analyzed. Data collection was carried out with Mo-Kα radiation using a Bruker platform diffractometer equipped with a SMART6000 CCD detector. Intensity data was collected using 0.3º increment ω scans up to 2θ < 70.4º (2θ < 50.2º with a completeness of 0.99). Data integration was carried out with SAINT,57 and corrections for absorption and decay were applied using SADABS.58 The structure was solved by direct methods and refinement was performed by full matrix least squares on F2 using all 1418 unique
80 data. The final refinement included anisotropic displacement parameters for non-hydrogen atoms and isotropic displacement parameters for all hydrogen atoms. The refinement converged to wR2 = 0.0694 (for F2, all data) and R1 = 0.0263 (F, 1372 reflections with I >2 σ(I )).
Table 4.5. Crystal data and structure refinement for 2.5.59 _____________________________________________________________ Empirical formula
C16 H8 S3
Formula weight
296.40
Temperature
120(2) K
Wavelength
0.71073 Å
Crystal system, space group
orthorhombic, Pmn21
Unit cell dimensions
a = 25.6814(5) Å
alpha = 90 deg.
b = 6.07740(10) Å
beta = 90 deg.
c = 3.89110(10) Å
gamma = 90 deg.
Volume
607.31(2) Å3
Z, Calculated density
2, 1.621 Mg/m3
Absorption coefficient
0.588 mm-1
F(000)
304
Crystal size
0.30 x 0.20 x 0.08 mm
Theta range for data collection
3.17 to 35.17 deg.
Limiting indices
-40750 425, 575, 780 k l l ts ns 0.35 0.9 0.25 >1.0 >1.0 0.57 Properties determined in methylcyclohexane unless otherwise indicated; sh = shoulder; EtOH = ethanol; CH = cyclohexane; Ref. = reference; a The underlined values represent the maxima in the absorption spectra, the second number represents the longest wavelength absorption band; b 0-0 Energies obtained from the crossing point of normalized absorption and fluorescence spectra; c Ref. 23; d Ref. 10; e Ref. 24; f Standard: Ff (2-aminopyridine) = 0.6 in 0.1 N H2SO4 and Ff (anthracene) = 0.27, values ±15%; g Standard: Ff (9-bromoanthracene) = 0.02 and Ff (anthracene) = 0.27 in ethanol; values ±5%; h Ref. 25; i Lifetime faster than instrument resolution; j kf=Ff/tf and kNR=[1-Ff]/tf; k Value determined by fs transient absorption spectroscopy; l Lifetime longer than instrument resolution. E0-0 (S1ÆSo)b
nm
130
131 6.2.2 EMISSION FROM SINGLET STATE The fluorescence emission spectra of all compounds were acquired in MCH and are depicted in Figure 6.4. Compounds 2.1, 2.3, and 2.4 show a negligible Stokes' loss while compounds 2.2, 2.5, and 2.6 show a Stokes’ loss of 1-2 nm. This observation indicates a minimal loss of vibrational energy after excitation into a high vibrational level of the first singlet excited state and, therefore, a minimal displacement of the potential energy surfaces of singlet ground and first singlet excited state. The trend observed for absorption is also observed for fluorescence emission. The fusion of one thiophene onto 2.1 introduces a bathochromic shift for anti and syn of 38 and 30 nm, respectively. When introducing two thiophene fusions onto 2.2, a bathochromic shift of 54 (38) nm is observed for the anti (syn) orientation, and the same is true for the fusion of a benzothiophene unit onto 2.3 and 2.4, where a bathochromic shift is observed of 45 and 37 nm, respectively. The energy of the singlet state was calculated from the point of intersect between the absorption and fluorescence spectra. The spectra showed well-defined 0-0 transitions and vibronic progressions with 0.13 – 0.19 eV energy spacing. These modes are characteristic for C=C stretching modes of conjugated systems. The exception in the series is 2.4, where a second vibrational mode of 0.09 eV is detected. These values decrease consistently with the extension of the ring systems from di- to tri- to pentacyclic. Compounds 2.1, 2.3, 2.4, and 2.6 have comparable low fluorescence quantum yields below 2%. Exceptions are 2.2 and 2.5 with quantum yields of 8 and 14%, respectively.
1x10
5
9x10
4
8x10
4
7x10
4
6x10
4
5x10
4
4x10
4
3x10
4
2x10
4
1x10
4
2.1 2.3 2.5
Normalized Fluorescence Emission (a. u.)
Normalized Fluorescence Emission (a. u.)
132
0
1x10
5
9x10
4
8x10
4
7x10
4
6x10
4
5x10
4
4x10
4
3x10
4
2x10
4
1x10
4
2.1 2.2 2.4 2.6
0 280
300
320
340
360
380
400
420
Wavelength (nm)
440
460
480
280
300
320
340
360
380
400
420
440
460
480
Wavelength (nm)
Figure 6.4. Fluorescence emission spectra of 2.1, 2.3 and 2.5 (left) and 2.1, 2.2, 2.4 and 2.6 (right). 6.2.3 FLUORESCENCE LIFETIMES The fluorescence lifetime (τf) of 2.1 was reported as 0.28 ns.23 This lifetime, as well as the fluorescence lifetimes of 2.2, 2.3, 2.4, and 2.6, were faster than the time-resolution of our instrument for time-correlated single photon counting (TCSPC). The values listed in Table 6.1 therefore must be treated as estimates. The decay plot of 2.5 is depicted in Figure 6.5 and the lifetime was accurately determined as 4.6 ns. From these values, we calculated the radiative and nonradiative rates as indicated in Table 6.1. In contrast to our results, Nijegorodov and Mabbs24 reported a fluorescence lifetime of 2.6 ns and a radiative rate kf of 3.0 x 107 s-1 for 2.2 in cyclohexane.
133 10000
Counts
1000
100
Residuals
10
4 2 0 -2 -4
2
χ R=1.276
10
13
15
18
20 23 Time (ns)
25
28
30
Figure 6.5. Time profile for the decay of fluorescence of 2.5 in solution of methylcyclohexane in presence of oxygen. Depicted are the instrument response function (left) and the native data (—) together with a super-imposed fit (—) and a residual plot (bottom).
6.3 FEMTOSECOND TRANSIENT ABSORPTION SPECTROSCOPY Femtosecond (fs) transient absorption spectra of all compounds are depicted in the following figures. These figures depict the spectral evolution of processes observed as a function of delay time after excitation. The fs time regime is fast enough to resolve processes that occur between initial excited state and relaxed, lowest excited state as well as decay processes of the lowest excited states. For compounds 2.3 and 2.5 we observed a negative value of the optical density difference at the blue-edge of our detection window. In both cases, we attribute this
134 observation to stimulated emission due to the resemblance with steady-state fluorescence emission and the match in decay time of the negative transient absorption signal compared to fluorescence decay time as observed by TCSPC. For most compounds, we observe absorption signals with multiple, overlapping decay processes, some of which in the picosecond timedomain. Due to a lack of computational data for possible vibrational modes and/or electronic excited states, the interpretations of these fast signals are outside the scope of this investigation. To the best of our knowledge, this is the first report of the fs transient absorption signals for the entire series of compounds. The following figures represent optical density difference spectra versus wavelength and their evolution as a function of delay time (0 – 1500 ps). Figure 6.7 depicts the transient absorption spectra of 2.1 and 2.2. For 2.1, the broad signal around 550 nm decays with a lifetime of 350 ps, while the signal at 430 nm grows in with a rise time of 260 ps (Figure 6.10). The two signals are connected by a clear isosbestic point around 450 nm. The signal at 430 nm is assigned to the triplet state (see discussion in section 6.6 below). Its precursor, observed at 550 nm, therefore was assigned as the singlet state S1 of 2.1. 0.018
10 ps 200 ps 400 ps 1300 ps
0.016 0.014
61 ps 612 ps 1434 ps
0.055 0.050 0.045
0.012 0.040 0.035
0.008
∆ O.D.
∆ O. D.
0.010
0.006
0.030 0.025 0.020
0.004
0.015
0.002 0.010
0.000
0.005
-0.002
0.000
400
450
500
550
600
Wavelength (nm)
650
700
750
400
450
500
550
600
650
Wavelength (nm)
Figure 6.6. Fs transient absorption spectra of 2.1 (left) and 2.2 (right).
700
750
800
135
0.050
50 ps 100 ps 304 ps 1204 ps
0.045 0.040 0.035
0.018 0.016 0.014
0.030
0.012
∆ O.D.
∆ O.D.
53 ps 150 ps 610 ps 1224 ps
0.020
0.025 0.020 0.015
0.010 0.008 0.006
0.010
0.004
0.005
0.002
0.000
0.000
-0.005 350
400
450
500
550
600
650
700
750
400
800
450
500
550
600
650
700
750
Wavelength (nm)
Wavelength (nm)
Figure 6.7. Fs transient absorption spectra of 2.3 (left) and 2.4 (right). 0.07 0.06
49 ps 300 ps 750 ps 1055 ps 1436 ps
0.05
0.045 0.040 0.035 0.030
∆ O. D.
0.04
∆ O. D.
25 ps 100 ps 505 ps 1386 ps
0.050
0.03 0.02
0.025 0.020 0.015
0.01
0.010
0.00 0.005
-0.01
0.000
400
450
500
550
600
Wavelength (nm)
650
700
750
400
450
500
550
600
650
700
750
800
Wavelength (nm)
Figure 6.8. Fs transient absorption spectra of 2.5 (left) and 2.6 (right). The signals of 2.2-2.6 were assigned using the same assumptions and the data are summarized in Table 6.1. The decay of the singlet state population for 2.2 is 0.9 ns, which coincides with the fluorescence decay time as reported by Berlman.25 The plots in Figure 6.10 clearly exhibit that faster components are mixed into the decay and rise of the singlet and triplet state. It is evident from the ground state absorption spectra that an excitation with laser light of 267 nm leads to population of higher order excited states, such as S2 or above. The fast decaying
136 signals observed during the first several picoseconds of data acquisition can be ascribed to internal conversions from these higher energy excited states to the lowest energy excited state S1.
0.010
0.018 Data: 2.1 at 582nm Model: ExpDec1
0.008
Chi^2/DoF R^2 y0 A1 t1
0.00052 0.00681 354.75417
0.014 0.012
±0.00012 ±0.00011 ±14.81702
0.010
∆ O.D.
∆ O.D.
0.006
0.016
= 5.2652E-7 = 0.90019
0.004
0.002
0.008
Data: 2.1 at 426 nm Model: ExpDec1
0.006
Chi^2/DoF R^2 y0 A1 t1
0.004 0.000
= 1.2685E-6 = 0.65553
0.01443 -0.00454 262.54851
±0.00012 ±0.00012 ±20.83228
0.002 0.000
-0.002 0
200
400
600
800
1000
0
200
Time (ps)
400
600
800
1000
Time (ps)
Figure 6.9. Decay of singlet state S1 of 2.1 at 580 nm (left) and rise of triplet state T1 at 426 nm (right).
Data: Data1_B Model: ExpDec1
0.06
0.05
Chi^2/DoF = 9.3851E-6 R^2 = 0.92007
0.05
y0 A1 t1
0.04
0.00129 0.04907 869.61531
±0.00124 ±0.00113 ±37.9859
0.04
∆ O.D.
∆ O.D.
0.03 0.03 0.02 0.01 0.00
0.02
Data: Data5_B388 Model: ExpDec1
0.01
Chi^2/DoF = 6.2932E-6 R^2 = 0.89288
0.00
y0 A1 t1
0.05041 -0.03567 1060.25077
±0.00127 ±0.00123 ±57.73185
-0.01 -0.01 0
200
400
600
800
Time (ps)
1000
1200
1400
0
200
400
600
800
1000
1200
1400
Time (ps)
Figure 6.10. Decay of singlet state at 630 nm and rise of T1 at 388 nm for 2.2. The absorption bands observed for 2.3 are composed of multiple processes. Unlike for 2.1 and 2.2, there is no clear separation of the signals for S1 and T1 states. When probing the decay signal of the negative transient absorption signal at 350 nm (stimulated emission) and the
137 decay at 400 nm, a time constant of 260 ps is observed for both processes (Figure 6.11). This number is of equal value as the estimated fluorescence lifetime from TCSPC. At wavelengths 487 nm and 750 nm, the observed time constant for growth and decay is 182 ps, respectively (Figure 6.12). The kinetic traces at 400 nm and 487 nm exhibit fast components.
0.008
0.006
0.004
∆ O.D.
0.002 Data: BDTA350NM_B Model: ExpDec1
0.000
Chi^2/DoF = 3.6432E-7 R^2 = 0.95635
-0.002
y0 A1 t1
-0.004
0.00416 -0.00763 259.2141
±0.00013 ±0.00013 ±13.11609
-0.006 0
100 200 300 400 500 600 700 800 900 1000 1100 1200
Time (ps)
Figure 6.11. Decay of singlet state of 2.3 at 400 nm (left) and observed stimulated emission 350 nm (right).
0.05
0.0030 0.0025
Data: BDTA750NM_B Model: ExpDec1
0.04 0.0020
∆ O.D.
∆ O.D.
Chi^2/DoF = 8.9626E-8 R^2 = 0.7431
0.0015
0.03 Data: 2.3 at 487 nm Model: ExpDec2
0.02
Chi^2/DoF R^2 y0 A1 t1 A2 t2
0.01
= 2.8616E-6 = 0.94744
0.03945 -0.00962 16.09197 -0.01593 182.92762
y0 A1 t1
0.0010
0.00003 0.00154 185.59144
±0.00005 ±0.00007 ±24.40797
0.0005 0.0000
±0.0003 ±0.00121 ±4.69103 ±0.00098 ±20.84917
-0.0005
0.00
-0.0010 0
200
400
600
Time (ps)
800
1000
1200
0
200
400
600
800
1000
1200
Time (ps)
Figure 6.12. Rise and decay for 2.3 at 487 nm and 750 nm, respectively. For 2.4, no clean profile for a rising triplet state could be extracted due to the weak nature of this signal as observed for both fs and ns transient absorption spectroscopy. The decay at 755 nm (Figure 6.13) is composed of two components: 26 ps and 1020 ps. The longer component is
138 assigned to the decay of the single excited state and is consistent with the value of ~1.12 ns as estimated by TCSPC. Data: A755NM_B Model: ExpDec2
0.008
Chi^2/DoF = 1.9285E-7 R^2 = 0.88153 y0 A1 t1 A2 t2
∆ O.D.
0.006
0.00097 0.0007 26.30149 0.00477 1020.29142
±0.00035 ±0.00011 ±9.42548 ±0.00029 ±135.54943
0.004
0.002
0.000
0
200
400
600
800
1000
1200
1400
Time (ps)
Figure 6.13. Decay of 2.4 signals at 755 nm. The rise and decay profiles for 2.5 as observed at 513 nm and 780 nm (Figure 6.14) could not be fit due to the long time constants. The profiles are only shown for clarity. For 2.6, the decay of the singlet and the rise of the triplet state are observed with time constants of 0.57 ns, similar to the value observed by TCSPC (Figure 6.15).
0.06
513 nm
0.08
780 nm
0.07 0.05 0.06 0.05
∆ O.D.
∆ O.D.
0.04
0.03
0.04 0.03
0.02 0.02 0.01
0.01 0.00
0.00 0
200
400
600
800
Time (ps)
1000
1200
1400
1600
0
200
400
600
800
1000
1200
1400
1600
Time (ps)
Figure 6.14. Decay of singlet state (780 nm, left) and rise of triplet state (513 nm, right) of 2.5.
139
0.030
Chi^2/DoF = 3.5075E-6 R^2 = 0.9114
0.025
y0 A1 t1
0.00382 0.02088 576.12424
0.04
±0.0004 ±0.00037 ±23.96167
0.03
∆ O.D.
0.020
∆ O.D.
0.05
Data: A785NM_B Model: ExpDec1
0.015
Data: A530NM_B Model: ExpDec2
0.02
Chi^2/DoF = 1.8049E-6 R^2 = 0.8997
0.010 y0 A1 t1 A2 t2
0.01 0.005
0.03736 0.01741 2.60264 -0.01438 565.69666
±0.00029 ±0.0039 ±0.41579 ±0.00026 ±25.72704
0.00 0.000 0
200
400
600
800
1000
1200
1400
0
Time (ps)
200
400
600
800
1000
1200
1400
Time (ps)
Figure 6.15. Decay of singlet state of 2.6 at 785 nm (left) and rise of triplet state as observed at 530 nm (right).
6.4 PROPERTIES OF THE TRIPLET EXCITED STATE The properties of the excited triplet state are summarized in Table 6.2. Due to the high cracking incidence of organic glasses made from MCH under the experimental conditions, all steady-state phosphorescence emission and lifetime data was acquired in EPA (ethyl ether/isopentane/ethyl alcohol with 5/5/2) at 77K while the transient data was acquired in MCH to avoid formation of radical cations upon laser excitation.
Table 6.2. Properties of the triplet excited state.
l max (EPA)
ET(onset) a
unit
2.1
2.2
2.3
2.4
2.5
2.6
nm
414, 423, 431,
409, 424, 437,
496 (sh), 506,
474 (sh), 484,
509, 523, 542,
453 (sh), 471,
442, 452, 461,
447, 455, 470,
526, 540, 550.
498, 512, 522.
556, 569.
487, 497, 510.
473, 484
480, 489
410
405
478
464
494
(3.02 eV)
(3.06 eV)
(2.59 eV)
(2.67 eV)
(2.51 eV)
(2.77 eV)
0.42b
0.97b
0.02
0.10
0.04
0.56
nm
F p (77K,
447
EPA) t p (77K, EPA)
s
0.32c,d
1.3e
0.11
0.27
0.21
0.38
l max,T (MCH)
nm
430
390f
490
410, 500
520
390, 520
t Tg (MCH)
ms
17.3
47.9
21.8
5.0
128
13.6
Laser power
mJ/pulse
500
100
100
200
30
500
kq (O2, MCH)
109 M-1s
6.78
1.57
2.10
3.30
0.97
3.98
EPA = ethyl ether, isopentane, ethyl alcohol (5/5/2); MCH = methylcyclohexane; Ref = reference a Value obtained from the onset of phosphorescence band; b Ref. 32; c Ref. 7; d Ref. 26; e Ref. 27; f Ref. 11; g Data acquired after thorough purging with argon (30 min). As the lifetime is strongly dependent on purging, only the longest determined lifetimes are reported.
140
141 6.4.1 TRIPLET PROPERTIES AS OBSERVED BY PHOSPHORESCENCE Phosphorescence emission spectra of 2.1 to 2.6 in EPA at 77K are depicted in Figure 6.16, Figure 6.17, and Figure 6.18. All compounds exhibit spectra with rich vibrational structure. A redshift in phosphorescence emission is observed for 2.3 and 2.4 (2.5 and 2.6) when compared to 2.1 (2.2). The lifetimes and quantum yields are consistently larger for the syn compounds 2.4 and 2.6 when compared to the anti compounds 2.3 and 2.5, respectively. The syn compounds retain the properties of their parent compounds 2.1 and 2.2 the closest. The radiative lifetimes are in the range of 0.1 to 1.3 s. These lifetimes and the large S1-T1 gap are indicators of a π,π* triplet state.28 2.1 Normalized Emission Intensity (a.u.)
Normalized Emission Intensity (a.u.)
2.2
400
420
440
460
480
500
520
Wavelength (nm)
540
560
580
600
400
420
440
460
480
500
520
540
560
580
Wavelength (nm)
Figure 6.16. Phosphorescence spectra of 2.129 (left) and 2.230 (right) in EPA at 77K.
600
142
2.4
Normalized Emission Intensity (a.u.)
Normalized Emission Intensity (a.u.)
2.3
460
480
500
520
540
560
580
600
620
640
660
440
460
480
Wavelength (nm)
500
520
540
560
580
600
Wavelength (nm)
Figure 6.17. Phosphorescence spectra of 2.3 (left) and 2.4 (right) in EPA at 77K. 2.6
Normalized Emission Intensity (a.u.)
Normalized Emission Intensity (a. u.)
2.5
480
500
520
540
560
580
Wavelength (nm)
600
620
640
440
460
480
500
520
540
560
580
600
620
640
Wavelength (nm)
Figure 6.18. Phosphorescence spectra of 2.5 (left) and 2.6 (right) in EPA at 77K. The triplet energies were calculated from the onset of the phosphorescence spectra. The relaxed S1-T1 singlet triplet energy gap is ~1.1 eV for compounds 2.1, 2.3, and 2.4 while compounds 2.2, 2.5, and 2.6 show a relaxed S1-T1 gap of ~0.7 eV. A plot of the T1 energy versus the S1 energy revealed a linear relationship for each group. The observed linear fit resulted in slopes of near unity indicative that the S1 and T1 are equally modified by the orientation of the thiophene unit (Figure 6.19). This trend no longer held true, when considering the fusion of a benzothiophene unit onto 2.3 (2.4) to receive 2.5 (2.6), particularly when plotted together with thiophene (Es=5.1 eV and ET=3.44)4 as the root molecule in the series.
143
3.1
2.2
3.0
2.1
Slope = 1.01 (R = 0.99)
E (T1) / eV
2.9
2.8
2.6
Slope = 0.93 (R=0.99)
2.7
2.4 2.6
2.3 2.5
2.5 3.2
3.3
3.4
3.5
3.6
3.7
3.8
3.9
4.0
4.1
4.2
E(S1) / eV
Figure 6.19. Triplet energy as a function of singlet energies for 2.1-2.6.
6.5 NANOSECOND TRANSIENT ABSORPTION SPECTROSCOPY The nanosecond (ns) transient absorption spectra were acquired in MCH under ambient conditions. The solutions were checked for degradation using UV-vis spectrometry before and after laser excitation. The spectra were acquired at a step-size of 10 nm under excitation from the fourth harmonic of a Q-switched Nd:YAG laser. As was observed in fs transient absorption spectroscopy, all compounds except 2.4 and 2.6 exhibited a clear maximum transient absorption band. Compound 2.1 exhibited a transient absorption band maximum around 430 nm, while the maximum of 2.2 is blueshifted to 390 nm. The measured lifetime of the triplet state is critically dependent on laser power and method of degassing.12 Table 6.2 lists the laser power as determined in a setup after the empty sample holder. The laser power was decreased until the samples exhibited mono-exponential decay for the signal of interest. For 2.1, the transient signal
144 of the triplet at 430 nm was 17.3 µs, a 6-fold increase compared to the value reported by Seixas de Melo and coworkers23 as a result of lower laser power and more extensive purging with argon. All triplet lifetimes reported in Table 6.2 represent the highest values for τT detected after 30 min of purging with argon. The lifetime of the triplet state of 2.2 was reported before.11,12 Our spectrum matches very well with a previously reported spectrum in cyclohexane.11 The spectra for 2.3-2.6 are depicted in Figure 6.21 and Figure 6.22 and are summarized in Table 6.2.
0.018
0.18
2.1 in MCH
0.016
0.16
0.014
0.14
0.012
2.2 in MCH
0.12
∆ O.D.
∆ O.D.
0.010 0.008 0.006
0.10 0.08 0.06
0.004
0.04
0.002
0.02
0.000
0.00
-0.002 360
380
400
420
440
460
480
500
520
540
560
360
380
400
420
Wavelength (nm)
440
460
480
500
520
540
Wavelength (nm)
Figure 6.20. Transient absorption spectrum of 2.1 (left) and 2.2 (right) in MCH with λmax=430 nm and 390 nm, respectively.
0.06
0.020
2.3 in MCH
0.05
0.018
0.04
0.016
∆ O.D.
∆ O.D.
2.4 in MCH
0.03
0.014
0.02 0.012 0.01 0.010 0.00 0.008 380
400
420
440
460
480
500
Time (microseconds)
520
540
340
360
380
400
420
440
460
480
500
520
540
Wavelength (nm)
Figure 6.21. Transient absorption spectrum of 2.3 (left, λmax=490 nm) and 2.4 (right) in MCH.
145
2.5 in MCH
0.14
2.6 in MCH
0.018 0.12 0.016 0.014
0.08
∆ O.D.
∆ O.D.
0.10
0.06
0.012 0.010 0.008
0.04
0.006 0.02 0.004 0.00 380
400
420
440
460
480
500
520
540
560
580
600
350
Wavelength (nm)
400
450
500
550
600
650
700
750
800
Wavelength (nm)
Figure 6.22. Transient absorption spectrum of 2.5 (λmax=520 nm, left) and 2.6 (right) in MCH. The signals observed were assigned to triplet states due to the fact that the signals were quenched with cis,cis-1,3-cyclooctadiene and with oxygen. The bimolecular quenching rate of oxygen was determined using Stern-Volmer analysis. (All quenching plots and calculations can be found in Appendix II). The bimolecular quenching rates with oxygen are listed in Table 6.2 and range from 0.97 to 6.78 x 109 M-1s, values comparable with those for other polycyclic aromatic compounds.
6.6 CONCLUSIONS We have characterized the photophysical properties of a series of benzo-fused polycyclic heterocarbons 2.1-2.6 with well-defined embedded thiophene rings. We found that the fusion of one (two) thiophene(s) to 2.1 (2.2) causes redshifts in absorption and both fluorescence and phosphorescence emission bands. The phosphorescence lifetimes and quantum yields are larger for the syn compounds compared to the anti (2.3, 2.5) series, yet no comparable trend could be found for fluorescence lifetimes. In fact, compound 2.5 shows the longest fluorescence and triplet lifetimes in the entire series with 4.6 ns and 128 µs, respectively.
146 Using time-resolved transient absorption spectroscopy we characterized the excited states of 2.1-2.6. The signals assigned to the triplet states served as roadmap and were observed consistently in fs time-domain as rising absorption bands and in ns time-domain as decaying absorption bands, which are quenched with triplet quenchers (oxygen and cis,cis-1,3cyclooctadiene). Further studies need to be carried out, combined with quantum mechanical calculations, to elucidate the fast processes as observed by fs transient absorption spectroscopy as well as to study the interesting fact that 2.5 is fluorescent with 14%, while 2.6 phosphoresces with a quantum yield of 56% with the only difference being the arrangement of the thiophene units.
6.7 GENERAL EXPERIMENTAL PARAMETERS 6.7.1 ABSORPTION SPECTRA The UV-vis absorption spectra were acquired on a Shimadzu UV-visible spectrophotometer (UV-2401 PC and Multispec-1501. All spectra were acquired in quartz UV cells and are corrected for background absorption from UV-cell and solvent. 6.7.2 FLUORESCENCE SPECTRA Fluorescence emission spectra were acquired in ethanol and in methylcyclohexane on a SPEX Fluorolog3-11 using a 450-W xenon short arc and sample detection in a 90º geometry. Fluorescence quantum yields were determined in ethanol using anthracene [φf = 0.27] and 9bromoanthracene [φf = 0.02] as standards.31
147 6.7.3 PHOSPHORESCENCE SPECTRA Phosphorescence spectra were acquired using a SPEX 1934D3 Phosphorimeter on a Fluorolog 3 spectrometer from Horiba Yvon (Horiba Group) utilizing a UV Xenon flash lamp as light source guided through an excitation monochromator. Emission detection was carried out in a 90° geometry after an emission monochromator. Emission spectra were acquired through a dewar flask and a quartz NMR tube in a frozen matrix (liquid nitrogen) of EPA that is diethyl ether/isopentane/ethyl alcohol in a ratio of 5/5/2. Biphenyl [φph = 0.24] 32 was used as the phosphorescence standard in quantum yield determination. 6.7.4 TIME-CORRELATED SINGLE PHOTON COUNTING We thank Dr. Denis Kozlov for help in data acquisition and for several discussions. Time-correlated single photon counting was carried out on a single photon counting spectrofluorimeter from Edinburgh Analytical Instruments (FL/FS 900) at ambient temperature in air for all measurements. A nanosecond flashlamp operating under an atmosphere of H2 gas (0.50-0.55 bar, 0.7 nm fwhm, and 40 kHz repetition rate) was used as excitation source, which was passed through a monochromator. The samples were prepared in methylcyclohexane and data acquired in aerated solutions. Analysis of the TCSPC data was achieved by iterative convolution of the luminescence decay profile with the instrument response function using software provided by Edinburgh Instruments. 6.7.5 TRANSIENT SPECTROSCOPY Both, femtosecond (fs) and nanosecond (ns) transient experiments were carried out at the Ohio Laboratory for Kinetic Spectrometry in collaboration with Dr. Evgeny Danilov. For both type of experiments, the sample solutions were prepared to have an absorbance of 0.5-1.0 at the excitation wavelength (267 nm). The solutions were filtered through a Teflon filter (17 mm, 0.45
148 µm) and checked before and after laser excitation for decomposition using UV-vis spectrometry. All measurements were carried out at ambient temperature. 6.7.5.1 Femtosecond Transient Absorption Spectroscopy The apparatus for femtosecond time-resolved experiments used a Spectra-Physics Hurricane system as the laser source. This system comprised a seed laser (Mai Tai, a cw diode pumped and mode-locked Ti:sapphire pulsed laser), a pump laser (Evolution, a diode-pumped Qswitched Nd:YLF laser), a stretcher, a Ti:sapphire regenerative amplifier, and a compressor. The output of the system consisted of pulses at 800 nm, 1 mJ, and ca. 100 fs (FWHM) at a repetition rate of 1 kHz. The output from the Hurricane was split (85% and 15%) in two beams. The larger one of these was frequency converted to the excitation wavelength of choice by coupling it into a third-harmonic generator (for 267 nm excitation) or into an optical parametric amplifier (SpectraPhysics OPA 800C for tunable wavelengths in the region 320-700 nm). The pump beam passed through an optical chopper (DigiRad C-980) rotating at a frequency of 100 Hz, and was focused, with the spot size of about 1 mm, into the sample cell where it was overlapped with the probe beam at an angle of ca. 5°; the relative polarizations of the pump and probe beams were set at the magic angle. The pump power was ~5 mW at 267 nm. The other (probe) beam passed through a computer-controlled delay line (Newport Corp. MTL 250 PP 1 250 mm linear positioning stage) that provided an experimental time window of about 1.6 ns with a step resolution of 6.6 fs, then it was focused into a 3 mm thick CaF2 plate to generate a white light continuum (effective useful range, 340-820 nm) and then it was focused into the sample. The energy of the probe pulse was
