S: Hexacene, Heptacene, and Derivatives

SYNTHESIS AND STUDY OF HIGHER POLY(ACENE)S: HEXACENE, HEPTACENE, AND DERIVATIVES

Rajib Mondal

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 2007

Committee:

Douglas C. Neckers, Advisor

Daniel M. Pavuk Graduate Faculty Representative

Thomas H. Kinstle

Michael A. J. Rodgers

Rajib Mondal

ABSTRACT

Douglas C. Neckers, Advisor

Poly(acene)s, linear poly(benzenoid) hydrocarbons, consist of an aromatic linear array. The largest whose synthesis has been authenticated is hexacene, C26H16. However, all reported syntheses of hexacene are difficult to repeat. Synthesis of higher acenes (seven member and higher) have challenged chemists for a long time. Heptacene has been elusive by the attempted classical synthetic routes because such procedures instantly yield an array of dimers. Recently, pentacene and its derivatives have been shown to be excellent candidates with enhanced π-stacking ability for application in OFET and in other electronic devices. Hexacene and heptacene can be considered potential molecules for opto-electronic applications.

A classical synthetic procedure to produce heptacene was followed first, which involved Meerwein-Ponndorf-Verley (MPV) reduction of corresponding quinone. Product appeared to be a mixture of dimers of heptacene. In order to minimize dimerization processes, several reactions to substitute at the carbonyl group of quinone with a bulkier group were attempted. However, none of these reactions was successful. The insolubility of the precursor dione seems to be the primary reason for the failure of these reactions.

To enhance the solubility and stability of heptacene and precursors, substituted heptacenes retaining the polyacene backbone were designed. Symmetric quinones were considered as the key synthons. While many reduction methods failed to yield the final product from substituted quinones, the borane-THF complex reduced 6,8,15,17- tetraarylheptacene-7,16-quinones to the 7,16-dihydro derivatives. An alternative approach using coupling between in-situ generated dibenzyne and naphthofuran also failed to yield any heptacene core.

Dihydroheptacene derivatives emit in the region of 420 – 428 nm in several solvents (ΦF = 0.15 – 0.21 in CH2Cl2) and in the solid state (ΦF = 0.37 – 0.44). These compounds have good solubility in common organic solvents, are reasonably stable, and retain color purity even after annealing for 24 hours at 110 oC. Though their dilute solutions showed blue emission (λmax ~ 420 nm), they showed excimer emission (λmax ~ 480 and 510 nm) at higher concentration. The OLED devices containing 6,8,15,17- tetraphenyldihydroheptacene showed green emission (λmax ~ 515 and 550 nm) that is even further red shifted than the emission of excimer. This indicates that an inter-ion pair, electromer, is responsible for the electroluminescence. Pump-probe experiments of dihydroheptacenes revealed that the S1 state shows a broad absorption (~ 500-650 nm) in dichloromethane with a lifetime of ~ 0.23–0.33 ns.

Another synthetic strategy employed was photochemical expulsion of two molecules of carbon monoxide from α-diketones of ethano polyacenes. Photo-precursors iv

of hexacene and heptacene were synthesized. The Strating-Zwanenburg photodecarbonylation of these photoprecursors in a poly(methyl methacrylate) matrix yielded the target hexacene and heptacene, respectively. The semi-rigid ploymer matrix enabled retention of highly reactive hexacene and heptacene through the prevention of thermal dimerization and oxidation. Heptacene was also generated in inert gas matrices at low temperature. Uv-vis-NIR absorption and IR spectra of heptacene were recorded in argon matrix at 10 K. When heptacene was generated in nitrogen matrix, it was stable up to 34 K. However, it was stable up to ~50 K, when generated in argon matrix.

Steady state photolysis, nanosecond laser flash photolysis, and femtosecond pump-probe experiments of α-diketone precursors of acenes were carried out to understand the mechanism of the Strating-Zwanenburg photodecarbonylation. It appears that both the singlet and triplet states of the diketones are involved in the decarbonylation process. These compounds have a small singlet-triplet energy gap (~ 4 kcal/mol). The lifetimes of the singlet excited states are in the range of 20-218 ps and decrease as the number of the benzenoid ring increases in the molecule. The triplet states are short lived (> 370 ps < 7 ns) and do not appear during the nanosecond experiments. It seems that the decarbonylation occurs within 7 ns. During the LFP experiment of heptacene precursor, the triplet state of the photoproducts, i.e., heptacene (λmax = 580 nm, τ ~ 11 μs), was also observed.

Rapid oxidation of heptacene occurs when a polymer film containing heptacene is exposed to air and this could be easily monitored by following gradual disappearance of its absorption in the visible region. The rates of disappearance of heptacene in different polymer films were observed to follow pseudo first order kinetics. Interestingly, those rates measured in the films of polystyrene (7.25 × 10-4 s-1), poly(ethyl methacrylate) (4.27 × 10-4 s-1), poly(methyl methacrylate) (1.60 × 10-4 s-1), and poly(vinyl chloride) (1.03 × 10-4 s-1) were found to correlate well with their oxygen permeability values. This indicates that the high reactivity of heptacene towards molecular oxygen can be used to determine the oxygen permeability of polymers. v

This dissertation is dedicated to my parents (Mr. Bibekananda Mondal & Mrs. Sona Rani Mondal), my uncle (Dr. Rabindranath Mandal), and my respected teacher (Mr. Ranjit K. Das)………… vi

ACKNOWLEDGMENTS

I wish to take this unique opportunity to express my deep sense of indebtedness, gratitude, and respect to my advisor, Dr. Douglas C. Neckers. He initiated me into this very fascinating and challenging field of organic electronics. His constant support, encouragement and timely interventions are responsible for making this thesis a reality. I enjoyed the full freedom in working under his guidance. I feel honored to acknowledge his seminal role behind this dissertation.

It is my pleasure to acknowledge Dr. Bipin K. Shah for encouraging, reviewing, and criticizing almost all the research presented in this dissertation. I am thankful to Dr. Holger F. Bettinger at Ruhr University at Bochum for making this project exciting with a wonderful collaboration. Thanks are due to Dr. Brigitte Wex (Lebanese American University), Dr. Bilal R. Kaafarani (American University of Beirut), Dr. Albert N. Okhrimenko, and Ravi M. Adhikari for their help with several experiments and fruitful discussions.

Dr. Thomas H. Kinstle - one of my committee members, my organic teacher, and my landlord – has been always very helpful. I’ve asked him almost anything ranging from his favorite truck to answers to various scientific problems. Thanking him might not be the right way to acknowledge him. I’d like to thank Dr. Michael A. J. Rodgers for serving in my committee and for his inspiration to maintain the good work.

All present and former DCN group members, namely Sujeewa, Thilini, Hannah, Kelechi, Koushik, Sunil, Andrey, Dmitry, Priya, Jiang, and others, who have been very helpful, collaborative, and friendly inside and outside the lab during these past few years. Besides them, thanks are due to my friends in BG (Jaydeep, Padmanava, Ramesh, Siri, Aritra, Mallar, Mithun, Neeraj, Upali, Saptarshi, Madhumita), Ujjal, Amit, Sanjukta, Aparajita, and Debbani for sharing all joys and sorrows with me.

I wish to thank Nora, Alita, Midge, Karen, Mary for all the administrative help in the department, and Craig, Doug, Larry, Chen, Romanowicz for taking care of all the vii

technical issues regarding the research. Help of Center for International Program is highly appreciated. Financial support from the McMaster Endowment is greatly acknowledged for providing me the research fellowship.

Finally, I would like to give my deep appreciation to my family members: my parents – Bibekananda Mondal and Sona Rani Mondal, my brother and sister – Rakesh and Moumita, my uncle – Rabindranath Mandal, and my beloved nephew and niece – Toton and Titli. Their emotional support, love, and encouragement have inspired me to complete this thesis.

I thank each and every person associated directly or indirectly with the success of this endeavor.

viii

TABLE OF CONTENTS

Page

CHAPTER 1. POLY(ACENE)S: PROPERTIES AND APPLICATIONS ...... 1

1.1 Introduction…………………………………………………………………… 1

1.2 Acenes and Erich Clar ...... 1

1.3 Structure, Reactivity and Properties of Acenes ...... 2

1.4 Principles of Photophysical Processes ...... 8

1.5 Organic Electronic Devices ...... 10

1.5.1 Organic Light-Emitting Devices ...... 11

1.5.2 Organic Field Effect Transistors ...... 12

1.5.3 Photovoltaic Devices ...... 14

1.6 Objectives and Scopes of the Project ...... 15

1.7 References ...... 16

CHAPTER 2. SYNTHESIS OF POLY(ACENE)S ...... 19

2.1 Introduction ...... 19

2.2 General Synthetic Approaches to poly(acene)s ...... 20

2.2.1 Synthesis of Pentacene ...... 21

2.2.2 Synthesis of Hexacene ...... 22

2.2.3 Synthesis of Heptacene ...... 23

2.3 Attempted Synthesis of Heptacene ...... 24

2.3.1 Results and Discussion ...... 24

2.3.2 Experimental Section ...... 27

2.4 Conclusion ...... 29 ix

2.5 References ...... 29

CHAPTER 3. SYNTHETIC APPROACHES TO SUBSTITUTED HEPTACENES ...... 31

3.1 Introduction ...... 31

3.2 Results and Discussion ...... 32

3.2.1 Synthetic Approaches to Substituted Heptacene ...... 32

3.2.2 X-ray Crystal Structure of Dihydroheptacenes ...... 36

3.2.3 Dibenzyne Approach ...... 37

3.3 Experimental Section ...... 37

3.4 Conclusion ...... 47

3.5 References ...... 47

CHAPTER 4. PHYSICAL PROPERTIES OF DIHYDROHEPTACENES ...... 49

4.1 Introduction ...... 49

4.2 Results and Discussion ...... 51

4.2.1 Photoluminescence in Solution ...... 51

4.2.2 Photoluminescence in the Solid State ...... 56

4.2.3 Electrochemical Properties ...... 58

4.2.4 Electroluminescence and Device Characteristics ...... 60

4.2.5 Transient Spectroscopy ...... 62

4.3 Experimental Section ...... 64

4.3.1 Fluorescence Quantum Yields ( ΦF) ...... 64

4.3.2 Fluorescence Lifetime (τF) Measurement ...... 65

4.3.3 Fabrication and Characterization of OLEDs...... 65

4.3.4 Electrochemical Measurements ...... 66 x

4.3.5 Ultrafast Spectrometry ...... 67

4.3.6 Geometry Optimization ...... 68

4.4 Conclusions ...... 68

4.5 References ...... 68

CHAPTER 5. PHOTOCHEMICAL SYNTHESIS OF HEXACENE AND HEPTACENE . 71

5.1 Introduction ...... 71

5.2 Results and Discussion ...... 72

5.2.1 Photochemical Synthesis of Hexacene ...... 72

5.2.2 Photochemical Synthesis of Heptacene ...... 78

5.2.3 Photochemical Synthesis of Anthracene ...... 81

5.2.4 Photogeneration and Thermal Stability of Heptacene in

Inert Gas Matrices ...... 82

5.3 Experimental Section ...... 86

5.3.1 Synthesis of Hexacene Precursor ...... 87

5.3.2 Synthesis of Heptacene Precursor ...... 90

5.3.3 Synthesis of Anthracene Precursor ...... 92

5.3.4 Sample Preparation ...... 94

5.4 Conclusion ...... 95

5.5 References ...... 95

CHAPTER 6. MECHANISM OF PHOTO-DECARBONYLATION ...... 97

6.1 Introduction ...... 97

6.2 Results and Discussion ...... 99

6.2.1 Photophysical Properties of the α-Diketones ...... 99 xi

6.2.2 Steady-state Photolysis of the α-Diketones ...... 100

6.2.3 Nanosecond Laser Flash Photolysis (LFP) of the α-Diketones ...... 103

6.2.4 Femtosecond Pump-probe Spectrometry of the α-Diketones ...... 105

6.2.5 Proposed Mechanism of Photodecarbonylation ...... 108

6.3 Experimental Section ...... 111

6.4 Conclusion ...... 112

6.5 References ...... 112

CHAPTER 7. DETERMINATION OF OXYGEN PERMEABILITY

USING HEPTACENE ...... 114

7.1 Introduction ...... 114

7.2 Results and Discussion ...... 116

7.3 Experimental Section ...... 121

7.3.1 Materials and Instruments ...... 121

7.3.2 Sample Preparation ...... 122

7.3.3 Determination of Rate Constant (kd) of

Disappearance of Heptacene ...... 123

7.4 Conclusion ...... 123

7.5 References ...... 124

APPENDIX A ...... 125

APPENDIX B ...... 127

APPENDIX C ...... 132

xii

LIST OF TABLES

Page

Table 1.1. Experimental and Computed Adiabatic S0 - T1 and Vertical S0 - S1 Energies

of Polyacenes……………………………………………………………………………... 4

Table 1.2. Physical Properties of Acenes [Ionization Potential (IP), Electron Affinity

(EA), Absorption maxima (Amax), Extinction co-efficient (ε), Fluorescence maxima (λF),

Fluorescence quantum yield (ΦF), and Fluorescence lifetime (τF)]………………………. 7

Table 4.1. Photophysical Properties of 4.1 – 4.3 in CH2Cl2……………………………… 52

Table 4.2. Photophysical Properties of 4.1 in Different Solvents………………………... 54

Table 4.3. Photophysical Properties of 4.1-4.3 Measured in the Solid State…………….. 57

Table 4.4. Electrochemical Properties and HOMO-LUMO Energy Gaps (EH-L) of 4.1-

4.3…………………………………………………………………………………………. 59

Table 4.5. Vibrational relaxation time (τ1) and lifetimes of the S1 state (τ2) of 4.1-4.3

recorded in dichloromethane……………………………………………………………… 64

Table 5.1. Theoretical and experimental S0-S1 values of acenes………………………… 81

Table 6.1. Photophysical properties of α-diketones (6.2-6.4)…………………………… 100

Table 7.1. Rate constant of disappearance (kd) and half-life (t1/2) of 7.1 in different

polymer films with varying oxygen permeability (Pm)…………………………………… 118

xiii

LIST OF SCHEMES

Page

Scheme 2.1. Retrosynthesis of poly(acene)……………………………………...... 20

Scheme 2.2. Bergman cyclization approach…………………………………………….. 21

Scheme 2.3. Synthesis of pentacene……………………………………………………… 21

Scheme 2.4. Modern synthesis of pentacene……………………………………………... 21

Scheme 2.5. Synthesis of hexacene………………………………………………………. 22

Scheme 2.6. Synthesis of hexacene………………………………………………………. 22

Scheme 2.7. Synthesis of heptacene……………………………………………………… 23

Scheme 2.8. Attempted synthesis of heptacene…………………………………………... 24

Scheme 3.1. Synthesis of lactols………………………………………………………….. 33

Scheme 3.2. Synthesis of dihydroheptacens……………………………………………... 33

Scheme 3.3. Synthesis of 3.11……………………………………………………………. 35

Scheme 3.4. Synthesis of 3.12……………………………………………………………. 38

Scheme 5.1. Synthesis of photoprecursor of hexacene (5.3)……………………………... 72

Scheme 5.2. Photochemical synthesis of hexacene (5.1)………………………………… 73

Scheme 5.3. Oxidation of 5.1…………………………………………………………….. 74

Scheme 5.4. Synthesis of heptacene (5.2)………………………………………………... 78

Scheme 5.5. Synthesis of phtotprecursor of anthracene (5.16)…………………………... 81

Scheme 5.6. Synthesis of phtotprecursor of anthracene (5.17)…………………………... 82

Scheme 6.1 Photochemical synthesis of poly(acene)s………………………………….. 97

Scheme 6.2 Photochemistry of bicyclo[2.2.2]octane-2,3-dione………………………… 98 xiv

Scheme 6.3 Photochemistry of α-diketone of anthracene………………………………. 98

Scheme 6.4. Proposed mechanism of decarbonylation…………………………………... 109

Scheme 7.1. Photochemical synthesis of heptacene……………………………………… 115

Scheme 7.2. Formation of endoperoxides of heptacene………………………………….. 120

LIST OF FIGURES

Page

Figure 1.1 A portrait of Erich Clar……………………………………………………….. 2

Figure 1.2 (a) Valence bond theory proposed structures of poly(acene)s; and (b) two

fully delocalized non alternating ribbons joined by relatively long bonds……………….. 3

Figure 1.3 Clar’s Sextet concept…………………………………………………………. 4

Figure 1.4 Normalized Absorption spectra of anthracene – heptacene (spectra of

antharcene to pentacene recorded in toluene, spectra of hexacene and heptacene recorded

in a polymer matrix)…………………………………………………………….. 6

Figure 1.5 Herringbone (left) and π-stacking (right) arrangements of pentacene……….. 8

Figure 1.6 Jablonski diagram…………………………………………………………….. 9

Figure 1.7 A schematic diagram representing the major applications of poly(acene)s….. 11

Figure 1.8 Structure and operation of OLEDs…………………………………………... 11

Figure 1.9 Structures of some acene derivatives used for OLEDs………………………. 12

Figure 1.10 Device configurations of OTFT devices……………………………………. 13

Figure 1.11 A schematic diagram showing the best performances of OFETs of different classes of compounds as an active layer vs. time plot……………………………………. 14

Figure 2.1 Structures of poly(acene)s…………………………………………………… 19

1 Figure 2.2 H-NMR of crude heptacene dimers in DMSO-d6…………………………... 25

Figure 2.3 DIP-MS of crude heptacene dimmers………………………………………... 26

Chart 3.1 Heptacene Derivatives……………………………………………………….. 31

Figure 3.1 1H-NMR spectrum of 3.6c at different temperature…………………………. 35 xvi

Figure 3.2 X-ray crystal structure of 3.1: (a) with a labeling scheme of atoms and (b) a view showing the angle between the two anthracene planes……………………………... 36

Figure 3.3 1H-NMR spectra of 3.12a and 3.12b………………………………………… 38

Figure 4.1 Chemical Structures of 7,16-dihydroheptacene derivatives (4.1-4.3)………... 50

Figure 4.2 Normalized fluorescence spectra of 4.1 (black), 4.2 (red), and 4.3 (green) in

CH2Cl2…………………………………………………………………………………….. 51

Figure 4.3 Fluorescence decay of 4.3 monitored at λmax = 424 nm in CH2Cl2; excitation

wavelength = 372 nm……………………………………………………………………... 53

Figure 4.4 Normalized fluorescence spectra of 4.1 in different solvents………………... 53

Figure 4.5 Normalized emission spectra of compound 4.1 (a) red: PL spectra in dichloromethane, 5.5×10-4 M; (b) blue: PL spectra in dichloromethane, 5.5×10-3 M; and

Figure 4.6 Molecular packing of X-ray crystal of 4.1: a view showing π-π interaction

56 between two anthracene moieties (distance = 4.277 Ȧ)…………………………………...

Figure 4.7 Normalized solid state fluorescence spectra of 4.1 (black), 4.2 (red), and 4.3

(green) recorded in thin films of PMMA…………………………………………………. 57

Figure 4.8 Normalized solid state fluorescence spectra recorded from thin films of

PMMA containing 4.1: (i) pristine, (ii) after exposing the film for 7 days at ambient condition, and (iii) after heating the film at 110 °C for 24 hours and cooling it down to room temperature…………………………………………………………………………. 58

Figure 4.9 (a) HOMO and (b) LUMO of the geometry optimized structure of 4.1 xvii

[B3LYP/6-31G(d)]………………………………………………………………………... 60

Figure 4.10 EL spectra of the device with the structure of ITO/PEDOT/PVK:PBD,

4.1/LiF/Al/Ag: (Device 1, 2, 3, and 4 containing 1%, 5%, 10%, 0% of 4.1, respectively). 61

Figure 4.11 Current density–Voltage (I–V) curves for devices 1, 2, 3, and 4…………… 62

Figure 4.12 Voltage–Luminance (V–L) curves for devices 1, 2, 3, and 4……………….. 62

Figure 4.13 Transient absorption spectra obtained from ultrafast pump-probe experiments of 4.1 in dichloromethane (1 × 10-4 M), recorded 0.20 ps (blue), 0.90 ps

(green), 4.15 ps (red), and 129 ps (black) after the laser pulse (excitation at 340 nm). …... 63

Figure 5.1 Normalized Absorption (red) , fluorescence (blue), and phosphorescene

(green) spectra of 5.3 (absorption and fluorescence spectra recorded in toluene at room

temperature and phosphorescence spectrum recorded in methanol/ethanol (1:4) matrix at

77 K)……………………………………………………………………………………… 73

Figure 5.2 Absorption spectra recorded during and after irradiation of 5.3 in degassed

toluene (inset: enlarged portion from 525 to 725 nm)……………………………………. 75

1 Figure 5.3 H-NMR spectra recorded at different times of irradiation of a CDCl3 solution

of 11 purged with oxygen (light source: a 395 nm UV-LED array)…………….. 76

Figure 5.4 Absorption spectra recorded during and after irradiation of 5.3 in a PMMA

film……………………………………………………………………………………...... 77

Figure 5.5 Absorption spectra recorded during and after irradiation of 5.12 in a PMMA

film, (inset – enlarged portion from 600 to 850 nm)……………………………………... 79

Figure 5.6 (A) Difference spectrum in the range of 3300 – 400 cm–1 obtained after

irradiation (high pressure mercury lamp, 350-450 nm) at 15 K containing 5.12

(sublimation at 220–225 °C). (B) IR spectrum computed for 5.12 at the B3LYP/6-31G* xviii

level of theory. (C) IR spectrum computed for 5.2 at the UB3LYP/6-31G* level of

theory……………………………………………………………………………………… 83

Figure 5.7 UV-vis spectra obtained after deposition of 5.12 (dark blue trace) and

subsequent irradiation after successively doubled time intervals. Arrows pointing down

indicate bands which decrease during irradiation, while upward pointing arrows indicate

bands with increase during irradiation. Isosbestic points are marked with small-headed

arrows. Inset: image showing photogenerated 5.12 (green) in Ar matrix at 10 K………... 84

Figure 5.8 Annealing of a heptacene (black) containing argon matrix. Here the cryostat is switched off at t = 0, and the spectra are recorded in 2 min intervals during warm up.

Until t = 16 min no significant change is observed (spectra omitted for clarity), at t = 18 min (T = 50 K), the base line shifts and at t = 20 min (T = 53 K), the heptacene

absorptions have already disappeared…………………………………………………….. 85

Figure 5.9 Monitoring of the spectral changes of the p- and α-bands of heptacene during

slow evaporation of a nitrogen matrix at 34 K. Top: Vis spectrum at the beginning of the

annealing. Bottom: Vis spectrum after 30 min annealing to 34 K. The matrix host is

sublimed away after this time. Middle: Spectra measured at 1.5 min intervals between t =

0 and t = 30 demonstrate the slow disappearance of heptacene…………………………… 86

Chart 6.1 Structure of α-diketones (6.1-6.4)……………………………………………. 99

Figure 6.1 Steady state photolysis of 6.2 in toluene. Inset: plot of ln(At-Aα) vs. time of

irradiation; absorbance monitored at 460 nm……………………………………………... 101

Figure 6.2 Energy differences of the ground state optimized structures of 6.1, 6.2, their isomers (6.1a and 6.2a), and photodecarbonylated products; 1,3-cyclohexadiene and anthracene, respectively…………………………………………………………………... 102 xix

Figure 6.3 Absorption spectra obtained from the nanosecond laser flash photolysis of 6.2

in toluene, recorded (a) 40 ns (black), (b) 100 ns (red), and (c) 500 ns (green) after the

laser pulse (λex = 460 nm)…………………………………………………………….. 103

Figure 6.4 Transient absorption spectra obtained from the nanosecond laser flash photolysis of 6.4 in an argon saturated dry toluene, recorded (a) 3 μs (black), (b) 10 μs

(red), and (c) 50 μs (green) after the laser pulse (λex = 460 nm). Inset: kinetic trace

monitored at 580 nm……………………………………………………………………… 105

Figure 6.5 (a) Absorption difference spectra obtained from the pump-probe spectrometry

of 6.4 in toluene recorded after 0.5 ps (green), 17 ps (red), and 105 ps (black) (130 fs

excitation pulse at 475 nm, pulse energy = 5 μJ), inset: kinetic trace monitored at 620

nm; (b) Kinetic trace monitored at 500 nm, showing decay profile in a wider time

window………………………………………………………………………... 106

Figure 6.6 Absorption difference spectra obtained from the pump-probe spectrometry of

6.2-6.4 in toluene(130 fs excitation pulse at 400 nm, pulse energy = 5 μJ); (a) spectrum

of 6.4 taken after 1 ps of the laser pulse; (b) spectrum of 6.3 taken after 1.6 ps of the laser pulse; and (c) spectrum of 6.2 taken after 50 ps of the laser pulse.…………...... 108

Figure 7.1 The absorption spectra of a PMMA film containing 7.1 recorded at different

times………………………………………………………………………………………... 117

Figure 7.2 Decomposition profiles of 7.1 in films of different polymers as monitored at

760 nm; (a) blue: polystyrene, (b) red: poly(ethyl methacrylate), (c) green: poly(methyl

methacrylate), and (d) orange: poly(vinyl chloride)……………………………………….. 117

Figure 7.3 Plot of the rates of disappearance (kd) of 7.1 vs. oxygen permeability (Pm)….. 119

Figure 7.4 1H-NMR spectra of 7.2 recorded before irradiation (bottom,) and after 15 xx

minutes of irradiation (top) in CDCl3 purged with oxygen. (The peaks appeared in the

6.0-6.3 ppm region in the top spectrum are due to the protons at the C-atoms that are attached with the oxygen bridge in endoperoxides)……………………………………….. 120

CHAPTER 1. POLY(ACENE)S: PROPERTIES AND APPLICATIONS

1.1 Introduction

Two-thirds of all known chemical species are aromatic.1 They are the basic building units of graphite2 and carbon nano-tubes.3 These include the well studied polyaromatic hydrocarbons

(PAHs), compounds with condensed benzene rings. Poly(acene)s, defined as linear

poly(benzenoid) hydrocarbons, are among the most important class of PAHs. The name “acene”

was coined by Erich Clar and is based on one of the most common acenes, anthracene.4

Poly(acenes) areπ-functional materials which have various applications in molecular electronics because of such interesting properties as (i) the wide range of band gaps (HOMO-LUMO gap),

(ii) tunable colors, and (iii) packing morphology. The properties are easily manipulated by substitution on the aromatic skeleton, or by increasing the number of aromatic rings. Smaller members of the series, including anthracene, tetracene, pentacene, and their derivatives, are used extensively in organic field effect transistors (OFETs),5 organic light emitting diodes (OLEDs),6 liquid crystals with enhanced electron transport,7 single crystal semiconductors,8 and photovoltaic devices.9 It has also been predicted that larger poly(acene)s should behave as one

dimensional organic conductors, or zero band gap semiconductors.10 Thus, the importance and

study of poly(acene)s has grown significantly over the last few decades.11

1.2 Acenes and Erich Clar

Erich Clar is generally considered the father of modern PAH chemistry (Figure 1.1). He was born in Czech-Sudetenland, and moved to Glasgow, Scotland in 1946 where he was professor in the Department of Chemistry, University of Glasgow from 1953 to 1972, and died in 1987. He

2 was awarded the first "Polycyclic Aromatic Hydrocarbon Research Award" at the 11th

International Symposium on PAHs, 24 September, 1987, after his death.12 Clar’s contribution

toward acene chemistry paved the way for today’s various applications.13

Figure 1.1 A portrait of Erich Clar (taken from the website of University of Glasgow).12

1.3 Structure, Reactivity and Properties of Acenes

The poly(acene)s can be represented by several limiting valence bond structures, including the undistorted, cis-distorted, or trans-distorted (Figure 1.2a) forms.14 Electronic properties depend

on the preferred structure. Valence-bond theory suggests all three structures are energetically

similar.15 The stability of a poly(acene) depends on a number of factors and calculations favoring

the stability of all three structures are reported.1 The cis-distorted form is more stable than the

trans-form, while long range Coulomb interactions were considered. Non interacting models

conclude the trans-form the most stable form while according to MP2/6-311G**, the undistorted

structure is preferred.15 According to Houk et al., poly(acene)s consist of two fully delocalized

non-alternating ribbons joined by relatively longer bonds (Figure 1.2b).14 The energy levels are

discrete in case of a particular finite acene and the HOMO-LUMO gap ( ΔE) decreases for higher

3 acenes. The spacing between levels, ΔE, near the Fermi surface for a chain of N monomeric acenes is given by the following equation:

ΔE ~ W / N2

where W ≈ 10 eV, the π-band width.10 It is clear from the above equation that with an increasing

number of rings, the band gap narrows. The vertical transition energies from benzene through

octacene decrease significantly (Table 1.1).

(a)

syn cis trans

(b)

Figure 1.2 (a) Valence bond theory proposed structures of poly(acene)s; and (b) two fully delocalized non alternating ribbons joined by relatively long bonds (B3LYP, ref. 14).

There are at least three series of aligned benzenoid compounds, including poly(acene)s, phenacenes with zig-zag type condensation, and helicenes with ortho-condensation. Compounds of the latter series are found to be quite stable in comparison isomers in the former series.16

Increasing the number of rings in poly(acene)s not only decreases the band gap, but also

4 increases the proton and electron affinities, and reduces the ionization potential.17 This, in fact, reduces the stability of this class of compounds.

Table 1.1 Experimental and Computed Adiabatic S0 - T1 and Vertical S0 – S1 Energies of Polyacenes14 Experimental UB3LYP/6-31G* Experimental TDDFT/6-31G* N S0 – T1 S0 – T1 S0 – S1 S0 – S1 (kcal/mol) (kcal/mol) (eV) (nm) (eV) 1 85 84 4.84[256] 5.54 2 61 59 3.97[312] 4.46 3 42 42 3.35[370] 3.28 4 30 26 2.61[475] 2.49 5 20 16 2.14[582] 1.95 6 12 11 1.79[695] 1.54 7 - 5.2 - 1.24 8 - 1.5 - 1.00 9 - - - 0.80

Their unstable nature can be explained by Clar’s sextet concept which approaches the

matter qualitatively.13 In case of poly(acene)s, whether its benzene or pentacene or heptacene or

one of the larger analogs, there is only one π-electron sextet (Figure 1.3). Thus, for the higher

acenes one sextet is shared over a larger number of rings and the larger assemblies become

increasingly less stable. According to molecular orbital theory, this is explained through the

sequential loss of benzenoid character (aromaticity).18

......

Figure 1.3 Clar’s Sextet concept.

Computational study by Houk et al. predicted that, after a certain number of rings, the

poly(acene) ground state becomes open-shell, where all the valence shell electrons do not

5 participate in chemical bonding.19 However, the ground states will still remain singlet states

because of their disjointed biradical nature. The singlet-triplet gap decreases from hexacene to

heptacene and octacene. The energy of the gap stays almost constant (5-6 kcal/mol) as one

increases in size from octacene, and the triplet state is always of higher energy than open-shell

singlet. The value is 5.7 kcal/mol is for decacene. The soliton’s effective length in the π-

conjugated oligoene system is about 14 carbon atoms according to the Su-Schrieffer-Heeger

model.20 It is feasible for heptacene (15 π-conjugated carbon atoms) to produce two parallel

oligoene solitons. In the case of decacene, it was shown, theoretically at least, that the two singly

occupied orbitals are localized on the two ribbons.1,19

Concerning the reactivity of poly(acene)s, as one moves to larger molecular weight

members of the series, the compounds act, in a reaction illustrated first with anthracene, as

reactive dienes in Diels-Alder (DA) addition reactions. The aromaticity increases from the edges of the molecule to the center acene rings. The magnitude of HOMO coefficients increase inwards

while the ring at the center becomes more reactive. For example, the central rings of pentacene and anthracene have activation energies (Ea) towards the DA addition with acetylene of 24.0 and

21 29.4 kcal/mol, respectively while the Ea for the rings at the edge of pentacene is 32.7 kcal/mol.

Recently, the stability of acenes toward nucleophilic and electrophilic additions as well as

DA addition with oxygen were addressed computationally.22-23 The reactivity of the acenes with

water and HCl increases from benzene to hexacene and then remains constant due to the

biradical character of the ground state of higher acenes. These reactions are calculated to be

exothermic with the activation barrier (Ea) for HCl addition being lower (~27 kcal/mol) than that

for the addition of water. The addition of HCl to benzene has Ea of 44 kcal/mol, whereas it is

only 16-18 kcal/mol for pentacene-nonacene. The Ea value for oxygen addition to acenes,

6 particularly singlet oxygen, falls in the same region as that of HCl addition. This value for benzene, anthracene, and pentacene is about 48, 29, and 20 kcal/mol, respectively with singlet oxygen. Interestingly, both triplet and singlet oxygen react with acenes and lead to the same products, endoperoxides. Concerted24 as well as biradical stepwise (starting from anthracene)22 mechanisms for the addition of oxygen have been suggested.

The electronic absorption (S0 – Sn) bands of the PAHs were classified by Clar into three

types based on their intensity, vibrational structure, and frequency shifts. They are: (i) α-bands –

2 3 1 1 4 ε ~ 10 -10 , complicated vibrational structure, A – Lb transitions; (ii) p-bands – ε ~ 10 , very

1 1 5 regular vibrational structure, A – La transitions; and (iii) β-bands – ε ~ 10 , shorter wavelength

1 1 1 1 25 and less vibrational structure, A – Bb, – Cb, and – Ba transitions.

Figure 1.4 Normalized absorption spectra of anthracene – heptacene (spectra of antharcene to pentacene recorded in toluene, spectra of hexacene and heptacene recorded in a polymer matrix).

Normalized UV-vis absorption spectra for anthracene through heptacene are shown in

Figure 1.4. The spectra of anthracene to pentacene were recorded in toluene, whereas the spectra

7 of hexacene and heptacene were recorded after they were generated photochemically in a polymer matrix. These are discussed in detail in Chapter 5. A systematic red shift (~100 nm) for the addition of each ring is observed, as expected from the theory. Physical properties for the known acenes (benzene to hexacene) are summarized in Table 1.2.

Table 1.2 Physical Properties of Acenes [Ionization Potential (IP), Electron Affinity (EA), Absorption maxima (Amax), Extinction co-efficient (ε), Stoke’s Shift(Δυ), Fluorescence quantum yield (ΦF), and Fluorescence lifetime (τF); ref 1, 25, 26, and 35].

IP EA Amax εmax Δυ τF Acene -1 -1 -1 ΦF (eV) (eV) (nm) (M cm ) cm (ns) 254 250 9.24378 ± Benzene -1.12 ± 0.03 204 8800 4780 0.04 62 0.00007 184 68000 301 270 275 5600 Naphthalene 8.144 ± 0.001 -0.19 ± 0.03 220 117000 3140 0.21 175 190 10000 167 30000 374 8500 251 220000 Anthracene 7.439 ± 0.006 0.530 ± 0.005 4100 0.27 4.7 221 11400 186 32000 472 14000 275 410000 Tetracene 6.97 ± 0.05 1.04 ± 0.04 3700 0.17 8.3 220 10600 210 31500 585 5900 430 710 Pentacene 6.63 ± 0.05 1.35 ± 0.04 3400 0.08 7.5 302 148000 330 8900 565 Hexacene 6.36-6.44 - 614 - 3000 - 1.5 672

Acenes usually adopt one of two common packing motifs: (i) the “herringbone” arrangement in which aromatic edge-to-face interactions dominate, and (ii) the coplanar arrangement, wherein π-electron rich faces stack on each other to form two dimensional electronic coupling (Figure 1.5).28

Figure 1.5 Herringbone (left) and π-stacking (right) arrangements of pentacene.

1.4 Principles of Photophysical Processes25,29,30

A photophysical process is defined as the physical process resulting from the electronic

excitation of a molecule or a system of molecules by photons. The molecular electronic states

and the transitions involved between them can be illustrated from the Jablonski diagram, shown

in Figure 1.6. In this diagram, the electronic states are arranged vertically by energy and grouped horizontally by its spin multiplicity. Radiative transitions involve absorption, fluorescence, and

phosphorescence and are indicated by straight arrows. The nonradiative transitions are internal

conversion (IC), intersystem crossing (ISC), and vibrational relaxation. They are indicated by

wavy arrows in the diagram. The singlet states along with their vibrational levels constitute the

singlet manifold and similarly the triplet states with their vibrational levels constitute the triplet

manifold. The zero point vibrational levels are indicated by the thicker lines in the diagram.

The radiative excitation transition, in which a photon is annihilated, is called absorption.

The molecule is excited from a lower to higher electronic state. The S1 Å S0 or Sn Å S0 transitions are spin allowed and usually appear in the absorption spectra. The T1 Å S0 or Tn Å

S0 transitions are spin forbidden, but can be observed spectrometrically by using long light paths,

intense light sources, or perturbation methods. The Sn Å S1 or Tn Å T1 absorptions are usually

observed using time-resolved flash photolysis. First, S1 is populated by an intense light source,

such as a pulsed laser, and the transient absorption of S1 can be observed during its lifetime.

Similarly, when T1 is populated via intersystem crossing (vide infra) from S1, the transient

absorption of T1 can be observed by time-resolved spectrometry. Such experiments yield the

optical absorption spectra of the transient states, their lifetimes, and rates of any reactions.

Figure 1.6 Jablonski diagram

The opposite of the absorption is luminescence, and can be defined as the radiative de- excitation transition, in which the molecule is de-excited from the higher electronic state to the lower electronic state by the emission of photon. There are two types of luminescent processes.

When a transition occurs between states with the same multiplicity, the process is called fluorescence. Whereas the process involving a transition between states of different multiplicity is called phosphorescence. Usually, S0 Å S1 transition is fluorescence and S0 Å T1 transition is

phosphorescence. The latter process is a spin forbidden, and is of longer duration (millisecond)

than fluorescence (nanosecond). Emission from higher excited states (S2 or T2) is also known, but very rare.

Nonradiative transitions arise through several different mechanisms. Relaxation of the excited state to its lowest vibrational level is called vibrational relaxation. This process involves

the dissipation of energy from the molecule to its surroundings, usually a solvent. A second type

of nonradiative transition is internal conversion (IC), which occurs when a vibrational state of an

electronically excited state can couple to a vibrational state of a lower electronic state. A third

type is the intersystem crossing (ISC); this is a transition to a state with a different spin

multiplicity, such as T1 Å S1. ISC is facilitated in molecules with large spin-orbit coupling, and

is responsible for efficient phosphorescence.

1.5 Organic Electronic Devices

Poly(acene)s up to pentacene are used extensively in two major kinds of organic electronic

devices: (i) organic light-emitting diodes (OLEDs); and (ii) organic field-effect transistors

(OFETs) or organic thin-film transistors (OTFTs).5-11 Smaller members of the series, mainly

anthracene, tetracene, and their derivatives show emission with relatively higher quantum

efficiencies and are useful for OLEDs. Larger acenes, such as pentacene and derivatives are useful for OFETs because of their smaller band gap as well as suitable packing behavior. They

also find applications in other electronic devices, such as photovoltaic solar energy collectors and

p- or n-type semiconductors in p-n hetero-junctions. These organic semiconductors have many advantages over their inorganic semiconductor counterparts: as they have structural flexibility, low temperature solution processing, and low cost.

Figure 1.7 A schematic diagram representing the major applications of poly(acene)s.

1.5.1 Organic Light-Emitting Devices

OLEDs have attracted considerable attention given their great potential application in flat panel

displays.6 Significant progress has been achieved in the development of full color OLED based

displays. The basic principle of an OLED is simple. When a layer of emitting material is placed

between a cathode and an anode, light is emitted due to the recombination of holes and electrons.

However, many modifications are needed in order to enhance the performance of current devices.6 A schematic diagram is presented in the Figure 1.8, showing the structure and

operation of OLEDs.

HTL emissive layer

- - LUMO e e LUMO anode

HOMO

h+ h+ HOMO EMISSION h+ transparent substrate cathode

Figure 1.8 Structure and operation of OLEDs

Many acene derivatives are successfully used in OLEDs, wherein the acene core is the chromophore and responsible for emission. The major problem with such compounds is the undesired emission with reduced efficiency in the solid state or in the devices due to formation of aggregates. Usually bulkier groups introduced to the central rings of the acenes overcome this problem.28 Representative acene derivatives (1.1-1.4), that have been successfully used in

OLEDs with various degrees of success, are shown in Figure 1.9. Anthracene derivatives (1.1

and 1.2) are blue emitters;31 tetracene and pentacene (1.3 and 1.4, respectively) are red emitters.32

R' R'

R R R Ph Ph

Ph Ph 1.1 1.2 R R R 1.4a R = CH3, R' = H 1.3a R = H, R' = t-BuPh 1.4b R = H, R' = t-butyl 1.3b R = H, R' = Si(i-Pr) R' 3 other such derivatives R'

Figure 1.9 Structures of some acene derivatives used for OLEDs.

1.5.2 Organic Field Effect Transistors

OFETs based on organic polymers, oligomers, and monomers are an alternative to traditional inorganic Si, Ge, and GaAs-based semiconductors. However, the relatively low charge carrier mobility (μ) of organic semiconductors along with a too low switching speed (Ion/Ioff) have

become an issue in their practical application. However, their processing advantages and

structural flexibility make them competitive in several existing thin film transistor applications

such as in active matrix flat panel displays based on LCDs or OLEDs and electronic paper

displays. Several novel applications have been proposed, in sensors, smart cards, and in radio-

frequency identification tags.1

There are two common type of device configurations used in OTFTs, (i) top-contact and

(ii) bottom-contact configurations (Figure 1.10). In the top-contact devices, the source and drain electrodes are placed onto the organic semiconducting layer. Whereas, the semiconducting layer is deposited onto the source and drain electrodes in the bottom-contact devices. The important device parameters are field effect mobility (µ), current modulation (Ion/Ioff), and threshold voltage

33 (VT). In case of p-type semiconductors (majority careers are holes), if the gate electrode is biased positively with respect to the source electrode, they operate in the depletion mode. Thus, the channel region is depleted of careers and producing the high channel resistance. This is called the off state. On the other hand, if the gate is negatively biased with respect to the source, they operate in accumulation mode and results low channel resistance. This is called the on state. The

ratio of the currents in the on state and off state is called the current modulation and usually

referred as Ion/Ioff ratio. Electrode polarity is reversed for the n-type semiconductors.

Figure 1.10 Device configurations of OTFT devices.

The most widely used organic compounds in semiconductors are pentacenes, poly(thiophenes), thiophene oligomers, and phthalocyanines. The highest field effect mobility known so far for an organic molecule is of pentacene (2.4 cm2V-1; on/off ration greater than 106 at 15 V of operating voltage). This value is more than the industrial bench mark of 1 cm2V-1, set

14 by an amorphous hydrogenated silicon based field effect transistor.33 The plot (Figure 1.11)

shows the maximum filed effect mobility values for a particular class of compounds as reported in the literature for different years. Recent research has been concentrated on pentacene and its derivatives with the aim to enhance the efficiency of OFET devices. Alkoxy-substituted silylethynylated pentacene derivatives have shown enhanced π-stacking abilities as well as

increased stability over unsubstituted analogs.34,35 Application of these compounds is not limited

to OFETs. With the synthesis of perfluoropentacene, workers have also explored their use as

high performance p-n junctions.36

Figure 1.11 A schematic diagram showing the best performances of OFETs of different classes

of compounds as an active layer vs. time plot.

1.5.3 Photovoltaic Devices

The use of poly(acene)s in photovoltaic cells is not common. However, only recently,

polycrystalline pentacene is reported to have been used in photovoltaic cells as a donor in

37 conjunction with C60 as the acceptor. A heterojunction with an electron acceptor promotes the

dissociation of excitons and allows broader coverage of the solar spectrum by compensating for

the spectral region where the absorption of pentacene is low. Efficient light harvesting occurs

throughout the visible spectrum with the peak external quantum efficiency of 58±4% at 670 nm,

the maximum absorption of pentacene.37 Pentacene was also used to build heterojunction with

38 perylene, Al, CdS0.6Se0.4, Si, and as hole conductor in Grätzel-type cells. However, these show

minor photovoltaic responses. In order to improve efficiency, pentacene derivatives with

enhanced molecular packing have also evaluated in this application.9

1.6 Objectives and Scopes of the Project

Recently, pentacene and its derivatives have been shown to be excellent candidates for various

electronic applications. However, the higher polyacenes (derivatives of hexacene and heptacene)

remain almost unexplored. This is mainly because they have yet to be unequivocally synthesized

and subsequently investigated. Hexacene and heptacene can be considered as the next potential

molecules for opto-electronic applications. Though the synthesis of hexacene is known, the

synthetic procedures reported are difficult to reproduce. The synthesis of heptacene was not

known until 2006 and has been a challenge for more than 50 years!

Our initial objectives were to design and synthesize stable hexacene, heptacene, and

certain derivatives. Several new synthetic strategies were employed to achieve the synthesis of

the target compounds. Most of the approaches failed to yield any heptacene core. However, a

new class of compounds, dihydroheptacenes, was synthesized following the reduction of

6,8,15,17-tetrasubstituted heptacene quinone using borane-tetrahydrofuran complex. These

showed interesting photophysical, electrochemical, and electroluminescent properties. OLED

devices containing tetraphenyldihydroheptacene exhibit green emission (λmax ~ 515 and 550 nm)

that is even further red shifted than the emission of an excimer of tetraphenyldihydroheptacene.

This indicates an inter-ion pair, electromer, is responsible for the electroluminescence.

We devised a unique photochemical route to produce heptacene in the first unequivocal

synthesis reported. All syntheses reported before were proven wrong. Hexacene was also

synthesized using this photochemical route and its stability was revisited. Whether the same

route can be utilized to synthesize even higher acenes, such as octacene and nonacene, is to be

explored.

Steady state photolysis, nanosecond laser flash photolysis, and femtosecond pump-probe

experiments on the α-diketone precursors of acenes reveal that both the singlet and triplet states

of the diketones are involved in the decarbonylation process and occur within 7 ns. However, not

all of the transients likely to be involved in this photoreaction could be clearly distinguished and

IR transient spectroscopy must next be employed to assign the intermediates.

The high reactivity of heptacene towards molecular oxygen has been used to determine

the oxygen permeability of polymers. However, study was limited to a few soluble polymers. We

expect that this can be extended to more useful polymers, such as polyethylene terepthalate.

1.7 References

1. Bendikov, M.; Wudl, F.; Perepichka, D. F. Chem. Rev. 2004, 104, 4891. 2. Chung, D. D. L. J. Mater. Sci. 2002, 37, 1475. 3. Iijima, S.; Ichihashi, T. Nature 1993, 363, 603. 4. Briggs, J. B.; Miller, G. P. C. R. Chimie 2006, 9, 916. 5. (a) Dimitrakopoulos, C. D.; Malenfant, P. R. L. Adv. Mater. 2002, 14, 99; (b) Ito, K.; Suzuki, T.; Sakamoto, Y.; Kubota, D.; Inoue, Y.; Sato, F.; Tokito, S. Angew. Chem. Int. Ed. Engl. 2003, 42, 1159. 6. (a) Wolak, M. A.; Jang, B.; Palilis, L. C.; Kafafi, Z. H. J. Phys. Chem. B 2004, 108, 5492; (b) Odom, S. A.; Parkin, S. R.; Anthony, J. E. Org. Lett. 2003, 5, 4245.

7. Shiyanovskaya, I.; Singer, K. D.; Percec, V.; Bera, T. K.; Miura, Y.; Glodde, M. Proc. SPIE-Int. Soc. Optical Eng. 2003, 4991, 242. 8. Twieg, R. J.; Getmanenko, Y.; Lu, Z.; Semyonov, A. N.; Huang, S.; He, P.; Seed, A.; Kiryanov, A.; Ellman, B.; Nene, S. Proc. SPIE-Int. Soc. Optical Eng. 2003, 4991, 212. 9. Lloyd, M. T.; Mayer, A. C.; Tayi, A. S.; Bowen, A. M.; Kasen, T. G.; Herman, D. J.; Mourey, D. A.; Anthony, J. E.; Malliaras, G. G. Org. Elect. 2006, 7, 243. 10. Kivelson, S.; Chapman, O. L. Phys. Rev. B 1983; 28, 7236. 11. (a) Harvey, R. G. Polycyclic Aromatic Hydrocarbons; Wiley-VCH: New York, 1997; (b) Anthony, J. E. Chem. Rev. 2006, 106, 5028. 12. http://www.chem.gla.ac.uk/dept/history/clar.htm 13. Clar, E. Polycyclic Hydrocarbons; Academic Press: London and New York, 1964; Vols. 1, 2. 14. Houk, K. N.; Lee, P. S.; Nendel, M. J. Org. Chem. 2001, 66, 5517. 15. (a) Garcia–Bach, M. A.; Peñaranda, A.; Klein, D. J. Phys. Rev. B 1992, 45, 10891; (b) Raghu, C.; Pati, A.; Ramasesha, S. Phys. Rev. B 2002, 65, 155204-1. 16. Portella, G.; Poater, J.; Bofill, J. M.; Alemany, P.; Solá, M. J. Org. Chem., 2005, 70, 2509. 17. (a) Rienstra-Kiracofe, J. C.; Barden, C. J.; Brown, S. T.; Schaefer, H. F., J. Phys. Chem. A. 2001, 105, 524; (b) Deleuze, M. S.; Claes, L.; Kryachko, E. S.; François, J.-P. J. Chem. Phys. 2003, 119, 3106. 18. Suresh, C. H.; Gadre, S. R. J. Org. Chem. 1999, 64, 2505. 19. Bendikov, M.; Duong, H. M.; Starkey, K.; Houk, K. N.; Carter, E. A.; Wudl, F. J. Am. Chem. Soc. 2004, 126, 7416. 20. Heeger, A. J.; Kivelson, S.; Schrieffer, J. R.; Su, W.-P. Rev. Mod. Phys. 1988, 60, 781. 21. Schleyer, P. v. R.; Manoharan, M.; Jiao, H.; Stahl, F. Org. Lett. 2001, 3, 3643. 22. Reddy, A. R.; Bendikov, M. Chem. Comm. 2006, 1179 23. Reddy, A. R.; Fridman-Marueli, G.; Bendikov, M. J. Org. Chem. 2007, 72, 51. 24. Chien, S. -H.; Cheng, M. -F.; Lau, K. -C.; Li, W, -K. J. Phys. Chem. A 2005, 109, 7509. 25. Birks, J. B. Photophysics of Armatic Molecules, Wiley-Interscience: New York, 1970. 26. Murov, S. L.; Carmichael, I.; Hug, G. L. Handbook of Photochemistry, 2nd ed. Marcel Dekker, Inc. New York, 1993. 27. Nijegorodov, N.; Ramachandran, V.; Winkoun, D. P. Spectrochimica Acta Part A 1997, 53, 1813. 28. Anthony, J. E. Chem. Rev. 2006, 106, 5028. 29. Turro, N. J. Modern Molecular Photochemistry; University Science Books; California, 1991. 30. Lakowicz, J. R. Principles of Fluorescence Spectroscopy; 2nd Ed. Kluwer/Plenum; New York, 1999. 31. Tao, S.; Hong, Z.; Peng, Z.; Ju, W.; Zhang, X.; Wang, P.; Wu, S.; Lee, S. Chem. Phys. Lett. 2004, 397, 1; (b) Ni,S. Y.;Wang, X. R.; Wu, Y. Z.; Chen, H. Y.; Zhu, W. Q.; Jiang, X. Y.; Zhang, Z. L. Appl. Phys. Lett. 2004, 85, 878. 32. (a) Odom, S. A.; Parkin, S. R.; Anthony, J. E. Org. Lett. 2003, 5, 4245; (b) Wolak, M. A.; jang, B.-B.; Palilis, L. C.; Kafafi, Z. H. J. Phys. Chem. B 2004, 108, 5492.

33. Dimitrakopoulos, C. D.; Malenfant, R. L. Adv. Mater. 2002, 14, 99. 34. Payne, M. M.; Delcamp, J. H.; Parkin, S. R.; Anthony, J. E. Org. Lett. 2004, 6, 1609. 35. Haddon, R. C.; Chi, X.; Itkis, M. E.; Anthony, J. E.; Eaton, D. L.; Siegrist, T.; Mattheus, C. C.; Palstra, T. T. M. J. Phys. Chem. B. 2002, 106, 8288. 36. Sakamoto, Y.; Suzuki, T.; Kobayashi, M.; Gao, Y.; Fukai, Y.; Inoue, Y.; Sato, F.; Tokito, S. J. Am. Chem. Soc., 2004, 126, 8138. 37. Yoo, S.; Domercq, B.; Kippelen, B. Appl. Phys. Lett. 2004, 85, 5427. 38. Mayer, A. C.; Lloyd, M. T.; Herman, D. J.; Kasen, T. G.; Malliaras, G. G. Appl. Phys. Lett. 2004, 85, 6272.

CHAPTER 2. SYNTHESIS OF POLY(ACENE)S

2.1 Introduction

The smallest member of acene family, benzene, was first discovered in 1825.1 Recently, the degree of aromaticity and stability of the oligoacenes have been a subject of debate.2 The lower

homologs, such as benzene, naphthalene, and anthracene, can be extracted from coal tar.

However, higher homologs can only be achieved via multi step synthesis. A list of poly(acene)s

has been shown in Figure 2.1. Surprisingly, poly(acene)s only up to pentacene (2.1) are known in

the series. Although hexacene (2.2) is a known compound,3 its reported synthetic schemes are

difficult to repeat. An extensive study of 2.2 is not available in the literature because of these

synthetic difficulties .4 The synthesis of heptacene (2.3) was not known when this thesis was started, and we succeeded in its first unequivocal synthesis.5 Octacene and nonacene are not yet

known.

Figure 2.1. Structures of poly(acene)s

In this chapter, general synthetic approaches to poly(acene)s, mainly syntheses of compound 2.1, 2.2, and 2.3 will be discussed briefly (from the literature). An attempted synthesis of 2.3 following usual routes will be analyzed. The alternative successful syntheses of compound

2.2 and 2.3 are described in detail in Chapter 5.

2.2 General Synthetic Approaches to Poly(acene)s

Though there are many syntheses reported for angular condensed aromatic hydrocarbons there is no general approach for synthesizing linear condensed ring systems. Actually, the common ring closure approaches tend to proceed in the angular position to form phenacene or helicene.

Usually poly(acene)s have been synthesized with reduction of the corresponding quinones offering the dihydro skeleton. The general synthetic routes to quinones are mainly either Diels-

Alder reactions or aldol type condensations. A retrosynthetic scheme is shown in Scheme 2.1.

Recently, a Bergman cyclization approach was employed by Anthony et al.,6 but only

poly(acenes) no larger than the tetracenes can be synthesized using this route (Scheme 2.2).

Scheme 2.1. Retrosynthesis of poly(acene).

Poly(acene)

n reduction

O O O OHC + + OHC n O O O Diels-Alder reaction aldol type condensation quinone

Scheme 2.2. Bergman Cyclization Approach

X X X

2.2.1 Synthesis of Pentacene

Erich Clar attempted to synthesize almost all the possible poly(acene)s larger than pentacene. He

used the Friedel-Crafts reaction to synthesize 4,6-dibenzoyl-1,3-demethylbenzene from m-

xylophenone.7 The former was then heated with copper to convert it into dihydropentacene,

which was then dehydrogenated by passing its vapor over copper at 350-400 oC (Scheme 2.3). In

a modern synthesis, pentacene was prepared by reduction of pentacenequinone (Scheme 2.4).8

The latter can be easily synthesized by base catalyzed aldol-type condensation between 1,4- cyclohexadione and phthaldialdehyde.

Scheme 2.3. Synthesis of pentacene

O O

Cu Cu Δ 350-400 oC

Scheme 2.4. Modern synthesis of pentacene

O O CHO aq. KOH + cyclohexanol Al, HgCl2 CHO O O

2.2.2 Synthesis of Hexacene

The synthesis of hexacene (2.2) was reported independently by Clar9 and Marschalk10 in 1939.

Clar used copper mediated dehydrogenation of dihydrohexacene, whereas Marschalk used Pd/C

catalyzed hydrogenation in trichlorobenzene (Scheme 2.5). Compound 2.2 was also synthesized

following Scheme 2.6 with an overall yield of 15%.11 This synthesis starts with a Diels-Alder

reaction between octahydro-1,4-anthraquinone and 2,3-dimethylenedecalin to form hydrogenated

hexacene quinone. Formation of hydrogenated hexacene via its -SEt derivative from hexacene

quinone, followed by Pd/C catalyzed dehydrogenation leads to the formation of 2.2 in overall

four steps. Another of Clar’s four step syntheses of 2.2, starts from 4-(2-carboxybenzoyl)phthalic

anhydride and tetralin, but produces hexacene in very poor yield.

Scheme 2.5. Synthesis of hexacene

Pd/C or Cu

Scheme 2.6. Synthesis of hexacene

O O EtS SEt

+ EtSH ZnCl2 SEt O O EtS Ni

Pd/C

2.2.3 Synthesis of Heptacene

The original synthesis of heptacene (2.3) was controversial from the outset and heptacene was commonly considered to be the limiting acene with respect to stability.2 The synthetic history of

2.3 is checkered. Reported in 1942 by Clar, he claimed 2.3 formed from reducing heptacenequinone,13 but later withdrew this.3 Marschalk also failed to reproduce the Clar

synthesis.14 Compound 2.3 was subsequently reported by Bailey (Scheme 2.7), and, though the

elemental analysis was correct, this report was also shown later to be wrong.11 Compound 2.3

has been elusive following the attempted classical synthetic routes because such procedures

instantly yield an array of dimers. The last report of the synthesis 2.3 of which we are aware was

reported in a Ph. D. dissertation15 from UCLA wherein the NMR spectra of the dimers is first

presented.

Scheme 2.7. Synthesis of heptacene

O O EtS SEt

+ EtSH ZnCl2 SEt O O EtS Ni

Pd/C

Anthony and co-workers succeeded in 2005 in the synthesis of a heptacene derivative16 by introducing bulky substituents for kinetic stabilization of the acene π system, thought they noticed their heptacene derivative decomposed within weeks in solution. Finally in 2006, we used the photobisdecarbonylation of a bridged α-diketone to generate unsubstituted heptacene in a polymer matrix at room temperature (Chapter 5).17

2.3 Attempted Synthesis of Heptacene

2.3.1 Results and Discussion

As described above, the synthesis of heptacene remained controversial until 2006. We first attempted to synthesize heptacene following a classical route (Scheme 2.8) that started with a coupling reaction between α,α,α’,α’-tetra-bromo-o-xylene (2.4) and maleic anhydride followed by hydrolysis to yield 2,3-naphthalenedicarboxylic acid (2.5). Compound 2.5 was reduced to

2,3-bis(hydroxymethyl)naphthalene (2.6) using borane-tetrahydrofuran complex. Swern oxidation was carried out to reoxidize 2.6 to the dialdehyde, 2,3-naphthalenedicarboxaldehyde

(2.7). Heptacene-7,16-quinone (2.8) was synthesized via a base catalyzed aldol-type condensation between 2.8 and 1,4-cyclohexadione. Meerwein-Ponndorf-Verley (MPV) reduction of quinone 2.8 leads to the brown-dusty final product.

Scheme 2.8. Attempted synthesis of heptacenea

O CHBr2 CO H CH OH CHO (a) 2 (b)2 (c) + O CHBr2 CO2H CH2OH CHO 2.4 O 2.5 2.6 2.7 (d)

(e)

2.3 O 2.8

a Reagents and conditions: (a) NaI, DMF, 90 oC, 4 h; (b) BH3.THF, THF, 0 - 25 oC, 12 h; (c) (COCl)2, DMSO, CH2Cl2, NEt3, 3 h; (d) aq. KOH (5%), r.t., 10 min; (e) cyclohexanol, Al-metal, HgCl2, CCl4, reflux, 12 h.

1 The H-NMR spectra (Figure 2.2) of the brownish product in DMSO-d6 revealed several

protons in the aliphatic region, especially near ~5.4 ppm. Purification through usual laboratory

chromatographic techniques failed due to the lack of solubility of the residue in various organic

solvents. Vacuum sublimation was also unsatisfactory so nothing could be separated from the mixture. In 1986, Fang recorded 1H-NMR spectra of the isolated compound formed via his

synthesis of heptacene.15 The NMR of Fang’s compound appears to be similar to Figure 2.2. He

proposed that the isolated compound was a mixture of dimers of heptacene. Dimerization could

occur in several ways. The most probable path is between the two central rings.

1 Figure 2.2. H-NMR of crude heptacene dimers in DMSO-d6.

There is no evidence of dimer or oligomer in the DIP mass spectrum (Figure 2.3). The product seems to be the monomer. We surmise that at high temperature in the mass spectrometer, dimers/oligomers convert to monomer through a reverse dimerization reaction. The base peak at

26 m/e 189 indicates the stability of doubly-charged ion. Again, the molecular ion peak is M+2, m/e

380 which is likely the result of hydrogen transfer occurring in the ionization chamber. It can also be attributed to the dihydroheptacenes formed due to the over reduction. The product shows none of the expected long wavelength absorption of heptacene and the synthesis thus becomes questionable.

Figure 2.3. DIP-MS of crude heptacene dimers

Several reactions have been attempted in an effort to afford derivatives of heptacene in which the π-backbone is retained. Dimerization processes can be minimized by substituting the ring with bulky groups. However, none of the methods was successful in yielding the product of addition to the carbonyl group in 7,16-heptacenequinone (2.8). The insolubility of the precursor dione is the primary reason for the failure of these reactions.

2.3.2 Experimental Section

General Procedures. The starting materials α,α,α’,α’-tetra-bromo-o-xylene (2.4) and maleic anhydride were purchased from Lancaster and Aldrich, respectively, and used as received.

Solvents were either spectroscopic grade or purified by distillation and dried before use using proper drying reagents. IR spectra were recorded on Thermo Nicolet IR 200 spectrometer. NMR spectra were recorded from BRUKER Avance 300 MHz NMR spectrometer using TMS as internal standard. Mass spectra were recorded either on a HP-5880A GC-MS or SHIMADZU

GC-17A mass spectrometer. Progress of organic reactions was monitored either by TLC or GC.

Synthesis:

2,3-Naphthalenedicarboxylic acid (2.5). A mixture of 2.4 (8.44 g, 0.02 mol), maleic anhydride

(6.0 g, 0.06 mol) and sodium iodide (20.0 g, 0.134 mol) was added to 70 mL dry DMF. The mixture was then allowed to reflux (85-90 oC) under aspirator vacuum for 4 h. After cooling, the

tarry material formed was transferred to 350 mL water in which 10 g of sodium bisulfite was

dissolved. The pale yellow solid thus formed was filtered and washed with cold water. To

hydrolyze the anhydride, the yellow solid was dissolved in dil. NaOH and decolorized with active charcoal. Finally, acidification of the resulting solution gave 2.5, which was filtered and

dried under vacuum (63%, mp 239-240 oC); (lit.18 240 oC). IR (Neat) 1670, 2800-3100 (broad)

-1 1 cm ; H NMR (300 MHz, CD3OD) δ 8.18 (s, 2 H), 7.88 (dd, 2 H, J = 8, 8 Hz), 7.5 (dd, 2 H, J =

8, 8 Hz); mass spectrum, m/e (%) M+ 198 (45), 154 (50), 126 (100).

2,3-Bis(hydroxymethyl)naphthalene (2.6). Compound 2.5 (1.08 g, 5 mmol) in 5 mL of

THF was placed in a dry round bottom flask that was flushed with Ar. A 1M solution of

BH3.THF/THF (13 mL, 13 mmol) was added drop wise to the stirred mixture that was kept at 0

oC (ice bath with little NaCl). The gelatinous reaction mixture was allowed to stir at 0 oC for 1 h,

28 then brought to r.t. and left for 12 h. Following the addition of 1:1 (v/v) aq. THF (5 mL) formed a clear solution. This was saturated with K2CO3 to partition the phases. The aq. phase was

extracted several times with THF. The combined organic extracts were dried and evaporated to

yield a white solid. The residue was washed with 1:2 ether/hexane giving almost pure 2.6 (0.75

g) with 80% yield. mp 158-160 oC (lit.19 155-160 oC). IR (Neat) 3168 (broad) cm-1; 1H NMR

(300 MHz, acetone-d6) δ 7.91 (s, 2 H), 7.85 (dd, 2 H, J = 8, 8 Hz), 7.5 (dd, 2 H, J = 8, 8 Hz), 4.9

(d, 4 H), 4.45 (s, 2 H); mass spectrum, m/e (%) M+ 188 (45), 170 (100), 141 (50), 115 (40).

2,3-Naphthalenedicarboxaldehyde (2.7). To a cooled (-78 oC, dry ice/acetone) solution

of oxalyl chloride (0.9 mL, 0.01 mol) in 15 mL of dry dichloromethane (DCM), a mixture of dry

dimethyl sulfoxide (DMSO, 1.5 mL, 0.02 mol) and DCM (5.0 mL) was added drop wise under

Ar. After 5 min, a solution of the diol 2.6 in 5 mL of THF containing 0.4 mL of DMSO was

added drop wise at -78 oC. The resulting slurry was vigorously stirred for 1 h. Triethylamine

(5.64 mL, 0.04 mol) was added and the reaction mixture was allowed to reach room temperature

whereupon stirring was continued for another 1.5 h. The reaction mixture was poured into water

and extracted with ether. After drying and evaporating of the organic solvents, the resulting semisolid compound was recrystallized from ethyl acetate/hexane mixture to yield 0.47 g (70%)

o 18 o -1 1 of 2.7, mp 129-130 C (lit. mp 131-132 C). IR (Neat) 1680 cm ; H NMR (300 MHz, CDCl3) δ

10.7 (s, 2 H) 8.5 (s, 2 H), 8.05 (dd, 2 H, J = 8, 8 Hz), 7.75 (dd, 2 H, J = 8, 8 Hz); mass spectrum, m/e (%) M+ 184 (65), 155 (95), 127 (100).

7,16-Heptacenequinone (2.8).20 Cyclohexane-1,4-dione (0.11g, 0.95 mmol) and 2.7

(0.35 g, 1.9 mmol) were dissolved in 20 mL of ethanol. Upon addition of few drops of 5% aq.

KOH, an immediate brown precipitate was formed. Reaction was complete upon adding a few

more drops of 5% aq. KOH and heating the reaction mixture to ~50oC for a short time. The

29 precipitate was centrifuged, washed with ethanol and dried under vacuum and afforded 0.2887 g

(74%) of 2.8. mp - >360 oC. Mass spectrum, m/e (%) M+ 408 (100), 380 (30), 350 (15).

Heptacene dimer. A literature reported procedure was adopted for the reduction of 2.8.21

Aluminum (0.25 g) was suspended in 7.5 mL of cyclohexanol containing mercuric chloride

(0.005 g) and carbon tetrachloride (0.05 mL). The mixture was allowed to warm slowly to mild reflux and heating was continued overnight under an Ar atmosphere. 7,16-Heptacenedione

(0.2887 g, 0.71 mmol) was added and the resulting reaction mixture refluxed under Ar for another 48 h. The reaction mixture was cooled to room temperature and then centrifuged to collect the precipitate, which was washed with warm cyclohexanol, 10% HCl and ethanol.

Finally, drying under vacuum gave 0.15 g of the product mixture. m/e (%) M+ 380 (80), 378

(85), 189 (100).

2.4 Conclusion

The preparation of heptacene by classical routes failed due to the high reactivity of acene

backbone which leads to the immediate formation of dimer or oligomer. Pure heptacene is likely

not isolable under normal conditions following classical routes. Approaches to the

functionalization of 7,16-heptacenequinone turned out to be difficult due to its lower solubility in

the common organic solvents.

2.5 References

1. Schleyer, P. v. R. Chem. Rev. 2001, 101, 1115. 2. Bendikov, M.; Wudl, F.; Perepichka, D. F. Chem. Rev. 2004, 104, 4891. 3. (a) Angliker, H.; Rommel, E.; Wirz, J. Chem. Phys. Lett. 1982, 87, 208; (b) Bailey, W. J.; Liao, C. -W. J. Am. Chem. Soc. 1955, 77, 992; (c) Satchell, M. P.; Stacey, B. E. J. Chem. Soc. C: Organic. 1971, 3, 468; (d) Nijegorodov, N.; Ramachandran, V.; Winkoun, D. P. Spectrochimica Acta Part A 1997, 53, 1813.

4. Mondal, R.; Adhikari, R. M.; Shah, B. K.; Neckers, D. C. Org. Lett. 2007, 9, 2505. 5. Mondal. R.; Shah, B. K.; Neckers, D. C. J. Am. Chem. Soc. 2006, 128, 9612. 6. Bowles, D. M.; Anthony, J. E. Org. Lett. 2000, 2, 85. 7. (a) Clar, E.; John, F. Chem. Ber. 1929, 62, 3027; (b) Clar, E.; John, F. Chem. Ber. 1930, 63, 2987; (c) Clar, E.; John, F. Chem. Ber. 1931, 64, 2194. 8. Goodings, E. P.; Mitchard, D. A.; Owen, G. J. Chem. Soc., Perkin Trans. 1 1972, 11, 1310. 9. Clar, E. Chem. Ber. 1939, 72B, 2137. 10. Marschalk, C. Bull. Soc. Chim. 1939, 6, 1112. 11. Bailey, W. J.; Liaio, C. –W. J. Am. Chem. Soc. 1955, 77, 992. 12. Clar, E. Chem. Ber. 1939, 75B, 1283. 13. Clar, E. Ber. 1942, 75B, 1330. 14. Boggiano, B., Clar, E. J. Chem. Soc. 1957, 2681. 15. Fang, T. Heptacene, Octacene, Nonacene, Supercene and Related Polymers. Ph.D.Thesis, University of California, Los Angeles, CA, 1986. 16. Payne, M. M.; Parkin, S. R.; Anthony, J. E. J. Am. Chem. Soc. 2005, 127, 8028. 17. Mondal. R.; Shah, B. K.; Neckers, D. C. J. Am. Chem. Soc. 2006, 128, 9612. 18. Carlson, R. G.; Srinivasachar, K.; Givens, R. S.; Matuszewski, B. K. J. Org. Chem., 1986, 51, 3978. 19. Wilcox, C. F.; Weber, K. A. Jr. J. Org. Chem., 1986, 51, 1088. 20. Ried, D. D. W.; Anthofer, D-C. F. Angew. Chem., 1953, 65, 601. 21. Laquindanum, J. G.; Katz, H. E.; Lovinger, A. J. J. Am. Chem. Soc., 1998, 120, 664.

CHAPTER 3. SYNTHETIC APPROACHES TO SUBSTITUTED

HEPTACENES*

3.1 Introduction

The higher reactivity of heptacene towards Diels-Alder addition (dimerization and oxidation) as

well as the lower solubility of synthetic precursors makes the synthesis of heptacene challenging.

Anthony et al. succeeded in synthesizing a substituted heptacene1 by introducing bulky substituents for kinetic stabilization of the acene π system. In order to enhance the solubility and stability of heptacene and precursors, a new class of substituted heptacenes (Chart 3.1) was designed and synthesis was attempted. The 6, 8, 15, and 17 positions are substituted with several bulkier groups in these compounds.

Chart 3.1. Heptacene Derivatives.

Symmetric quinones were considered key synthons. While many reduction methods

failed to yield the final product from substituted quinones, the borane-THF complex reduced

6,8,15,17-tetraarylheptacene-7,16-quinones to the 7,16-dihydro derivatives (Scheme 3.2). The heptacene core is difficult from reduction of quinone. In order to eliminate the reduction of quinone at the last step, an alternative dibenzyne approach was designed. This requires one to initially form diepoxide derivatives of heptacenes but the approach also failed to yield the heptacene core.

3.2 Results and Discussion

3.2.1 Synthetic Approaches to Substituted Heptacene

The synthetic precursor quinone (3.10a-c) can be prepared from the furans (3.8 or similar) and p- benzoquinone.2 Cava’s synthetic procedure3 was used to synthesize lactols 3.7a-d, the precursor

of furans, from 2,3-naphthalic anhydride 3.4 (Scheme 3.1). These lactols easily undergo

dehydration under acidic conditions to yield 1,3-diarylnaphtho[2,3-c]-furans. However, alkyl

substituted furans such as 1,3-di-tert-butylnaphtho[2,3-c]-furan are unstable and hard to prepare

via this route. Thus, 3.8 was synthesized by dehydration of 3.7a in glacial acetic acid (Scheme

3.2), and isolated in the form of a red solid by direct recrystallization from the reaction mixture.

The Diels Alder reaction of 3.8 with p-benzoquinone followed by acid catalyzed dehydration of

3.9 yielded 3.10a.2,4

Compounds 3.7b and 3.7c produced the corresponding furan (analogous to 3.8, not

shown in the scheme) in glacial acetic acid as indicated by the appearance of the bright red

colored solution. However, isolation of these products via recrystallization was unsuccessful

because of their higher solubility in the solvent employed (glacial acetic acid). Thus, 3.10b and

3.10c were directly produced under optimized acidic conditions involving in situ Diels-Alder

33 cycloaddition and dehydration steps (route d in Scheme 3.2). These quinones show enhanced solubility in the common organic solvents compared to unsubstituted heptacene quinone.

Scheme 3.1. Synthesis of lactols a

R1 R2 R3 R O R 2 O 3 HO Ar' R3 R1 O R O O (a) 2 (b) (c) R CO H 1 O 2 O Ar 3.4 3.5a. R1, R2, R3 = H 3.6a. R1, R2, R3 = H 3.7a. Ar = Ar' = phenyl 3.5b. R1 = t-butyl, R2, R3 = H 3.6b. R1 = t-butyl, R2, R3 = H 3.7b. Ar = phenyl, 3.5c. R1, R2, R3 = methyl 3.6c. R1, R2, R3 = methyl Ar' = p-tert-butylphenyl 3.7c. Ar = Ar' = p-tert-butylphenyl 3.7d. Ar = Ar' = mesityl

a Reagents and conditions: (a) AlCl3, 90 – 95 °C, 15 h; (b) NaBH4, aq. NaOH, rt, 5 days; (c) Mg, Ar’Br, ether, THF, I2 (catalyst), 0 °C, 5 h.

Scheme 3.2. Synthesis of dihydroheptacens a

Ar' O Ar' HO Ar' Ar' (a) (b) O O O O

3.7a-c Ar Ar Ar O Ar 3.8 Ar = Ar' = phenyl 3.9 Ar = Ar' = phenyl (c) (d)

R1 O R2 R1 R2 (e)

R3 O R4 R3 R4

3.10a R1, R2, R3, R4 = phenyl 3.1 R1, R2, R3, R4 = phenyl 3.10b R1, R4 = phenyl; 3.2 R1, R4 = phenyl; R2, R3 = p-tert-butylphenyl R2, R3 = p-tert-butylphenyl 3.10c R1, R2, R3, R4 = p-tert-butylphenyl 3.3 R1, R2, R3, R4 = p-tert-butylphenyl

a Reagents and conditions: (a) glacial AcOH, 50 oC, 15 min. (b) p-benzoquinone, benzene, reflux, overnight. (c) p-toluenesulfonic acid, benzene, reflux, overnight. (d) p-benzoquinone, p- o toluenesulfonic acid, benzene, reflux, overnight. (e) BH3.THF, THF, 50 – 60 C, 2 days.

Furans, homologs to 3.8 substituted with aliphatic groups such as tert-butyl, are very unstable and couldn’t be synthesized under our conditions. The reaction between 3.8 and p- benzoquinone proceeds in an endo, exo sequence as indicated from the 1H-NMR spectra of 3.9.4

The Meerwein-Ponndorf-Verley (MPV) reduction of the diones 3.10a-c failed.5 The approach of

the alkoxide ion towards the carbonyl center is hindered by the aryl groups. However, use of an

excess of borane-tetrahydrofuran (BH3.THF) complex over reduces 3.10a-c to the corresponding

dihydro compounds 3.1-3.3. Controlling the reaction with a lesser quantity of the reducing agent

produces a product in which one of the carbonyl groups remains unreacted and the other is

transformed to the alcohol.

During this synthetic scheme, when two molecules of furan produced in situ from 3.7b

underwent cycloaddition with p-benzoqionone, two regioisomers of 3.10b were produced due to

the non-selective Diels Alder reaction - one symmetrical isomer having both phenyl groups on

one side of anthracene moieties, while the other unsymmetrical isomer having one phenyl group

and one p-tert-butylphenyl group on one side of the anthracene moieties. These isomers were

directly subjected to reduction (step e), yielding a mixture of two isomers of 3.2, which could not

be separated. 1H-NMR showed the ratio of the two isomers of 3.2 to be approximately 5:1.

3.7d, which has two mesityl groups, failed to produce the corresponding furan in acidic solution. This may be because the resulting furan would be sufficiently sterically crowded so that its formation was prevented. Instead, 3.7d underwent acid catalyzed ring opening and, following oxidation by p-benzoquinone, produced diketone 3.11 (Scheme 3.3). A similar oxidative product

of furan 3.8 was afforded when 3.7a was oxidized with chromium trioxide in pyridine.3

Scheme 3.3. Synthesis of 3.11a

HO Ar' O O O Ar 3.7d 3.11

a Reagents and conditions: (a) p-benzoquinone, p-toluenesulfonic acid, benzene, reflux,

overnight.

Figure 3.1. 1H-NMR spectrum of 3.6c at different temperature.

The 1H-NMR spectrum of mesityl lactone 3.6c is unusual in comparison to similar derivatives. The spectrum is temperature dependent (Figure 3.1). Peaks at 1.7, 2.25, and 2.65 ppm can be assigned to the methyl protons of mesityl group. Proton at 3 position of lactone

shows peak at 6.75 ppm. As shown in the figure, the half width of the peaks at 1.7, 2.65, and

6.75 become ~40 Hz from ~18 Hz upon increasing the temperature from 25 oC to 40 oC. This can

be explained by hindered rotation of the mesitylene group around the lactone moiety,6 which

leads to the unequal environment to the protons of two methyl groups at 2’ and 6’ position. The

environment of the lactone proton at the 3 position is also affected by this restricted rotation.

(a)

(b)

Figure 3.2. X-ray crystal structure of 3.1: (a) with a labeling scheme of atoms and (b) a view

showing the angle between the two anthracene planes.

3.2.2 X-ray Crystal Structure of Dihydroheptacenes

Compounds 3.1-3.3 normally exist in amorphous solid form at room temperature. However, very

slow recrystallization from hexane/dichloromethane yielded block like crystals of 3.1 suited for

single crystal x-ray diffraction study (Figure 3.2). Crystals of 3.1 are monoclinic, space group

P2(1). The anthracene moieties are twisted and the angle between the two anthracene planes is

48.11º. The four phenyl groups are almost orthogonal to the planes of anthracene moieties. The angles between the phenyl groups and the anthracene planes are observed to be 78.01, 69.52º,

86.85º, and 67.66º.

3.2.3 Dibenzyne Approach

1,2,4,5-Tetrabromobenzene is equivalent to the dibenzyne or bisaryne system and used earlier to synthesize ortho-condensed rings from both sides.7 Treatment of 1,2,4,5-tetrabromobenzene with

n-BuLi in presence of furan 3.8 yielded the bis-adducts (3.12) with 15% overall yield. In fact, the

reaction produced a mixture of trans (3.12a) and cis (3.12b) isomers. These were separated by

silica-gel column chromatography with an isolated 2.2 to 1 ratio of tras to cis. In 1H-NMR

spectra, peaks at 7.3 ppm for cis isomer and 7.37 ppm for trans isomer are clearly

distinguishable (Figure 6.3). They were assigned to the cis and trans based on the polarity as

indicated by thin layer chromatogram. One might expect higher polarity for cis isomer due to the

higher dipole moment of the molecule compared to the trans isomer. Further attempts for

8 deoxygenation, mainly by low-valent titanium (TiCl3) in n-BuLi was not successful to yield any

heptacene core.

3.3 Experimental Section

Solvents and reagents were used as received from commercial suppliers. Standard grade silica

gel (60 Å, 32-63 μm) and silica gel plates (200 μm) were purchased from Sorbent Technologies.

Reactions that required anhydrous conditions were carried out under argon in oven-dried

glassware. Organic solvents were either spectroscopic grade or purified by distillation and dried

before use using proper drying reagents.

Scheme 3.4. Synthesis of 3.12

Ph Ph

O O trans Ph Br Br n-BuLi Ph Ph 3.12a O + + Ph Ph trans Br Br cis 3.8 Ph O O cis

Ph Ph 3.12b

Figure 3.3. 1H-NMR spectra of 3.12a and 3.12b.

Mass spectra were recorded on Shimadzu GCMS-QP5050A instrument equipped with a

direct probe (ionization 70 eV). Matrix assisted laser desorption ionization (MALDI) spectra

were obtained using Bruker Daltonic Omniflex® instrument (N2 laser, 337 nm). A Bruker

spectrometer (working frequency 300.0 MHz for H1) was used to record NMR spectra.

Absorption and fluorescence spectra were recorded on a Shimadzu UV-2401 spectrophotometer

and a Fluorolog®-3 spectrometer, respectively. All measurements were carried out at room

temperature unless otherwise specified.

Synthesis.

2,3-Naphthalic anhydride (3.4). Compound 3.4 was synthesized following a literature

procedure reported for the synthesis of 2,3-naphthalenedicarboxylic acid, except for that the

hydrolysis step was not carried out.9 The starting materials were α,α,α’,α’-tetrabromo-o-xylene

and maleic anhydride. The unhydrolyzed product mostly contained the desired anhydride (3.4).

Ketoacids (3.5a-c). The general procedure followed to prepare ketoacids (3.5a-c) is as

follows: a suspension of 3.4 (2.0 g, 10 mmol) was made in 25 ml of benzene (for 3.5a) or tert- butylbenzene (for 3.5b), or mesitylene (for 3.5c). Aluminum chloride (2.77 g, 20 mmol) was

added to it at room temperature. The mixture was stirred overnight (15 hrs) at 85-90 oC. To the tarry reaction mixture, an excess of 6M aq. HCl was added. An excess ethyl acetate was also added. Additional aq. HCl and ethyl acetate were added until the solid dissolved. The organic layer was separated and washed with water and extracted thoroughly with saturated sodium bicarbonate solution. Acidification of the bicarbonate solution with conc. HCl, produced a white precipitate, which was filtered and then recrystallized from methanol/hexane mixture.

3-Benzoyl-2-naphthoic acid (3.5a). Yield = 87%, mp 208 - 209 oC (lit.3 209.5 oC). IR (Neat)

1467, 1678, 2500-3100 (broad) cm-1; 1H NMR (300 MHz, MeOD) δ 7.45 (t, 2 H, J = 7.5 Hz),

7.55 (t, 1 H, J = 7.5 Hz), 7.65 (m, 2 H), 7.72 (d, 2 H, J = 7.2 Hz), 7.9 (s, 1 H), 8.0 (s, 1 H), 8.1

(m, 1 H), 8.62 (s, 1 H); 13C NMR (300 MHz, MeOD) δ 129.04, 129.21, 129.41, 129.65, 130.24,

130.32, 130.60, 132.65, 134.22, 135.63, 139, 169, 199; mass spectrum (GC-MS), m/e (%) M+

232 (100), 155 (100), 127 (80), 105 (50), 77(95).

3-(p-tert-Butylphenyl)-2-naphthoic acid (3.5b). Yield = 60%. mp 239 - 240 oC. IR (Neat) 1452,

1678, 2500-3100 (broad) cm-1; 1H NMR (300 MHz, MeOD) δ 1.35 (S, 9 H), 7.5 (d, 2 H), 7.65

(m, 4 H), 7.85 (s, 1 H), 7.95 (m, 1 H), 8.1 (m, 1 H), 8.65 (s, 1 H); 13C NMR (300 MHz, MeOD) δ

31.84, 36.36, 126.88, 129.30, 129.38, 129.49, 129.73, 130.54, 130.60, 131.04, 133.09, 134.53,

135.96, 136.77, 139.72, 158.56, 169.59; mass spectrum (DIP-MS), m/e (%) M+ 332(12),

317(10), 299(15), 273(10), 231(45), 216(35), 198(45), 172(50), 161(40), 155(80), 126(100).

3-Mesitoyl-2-naphthoic acid (3.5c). Yield = 66%. mp 221 - 222 oC. IR (Neat) 1658, 1692,

-1 1 2400-3300 (broad) cm ; H NMR (300 MHz, CDCl3) δ 2.22 (s, 6 H), 2.36 (s, 3 H), 6.94 (s, 2 H),

7.6 (m, 2 H), 7.8 (d, 1 H, J = 8.2 Hz), 7.87 (s, 1 H), 7.97 (d, 1 H, J= 8.2 Hz), 8.42 (s. 1 H); mass spectrum (DIP), m/e M+ 318 (55), 299 (80), 285 (70), 273 (80), 258 (50), 229 (30), 215 (25), 199

(55), 147 (54), 133 (50), 115 (100).

Lactones (3.6a-c). These lactones were synthesized by the two general procedures described below. Route a3 produced higher yield of 3.6a and 3.6b, while this pathway produced

only little of 3.6c. Alternative route b10 was employed to obtain a higher yield of 3.6c.

Route a (synthesis of 3.6a and 3.6b). Compound 3.5a or 3.5b (7 mmol) was suspended

in 40 mL of water. Sodium hydroxide (2M) was added to enhance the solubility of the acid (3.5a

or 3.5b). Sodium borohydride (1.29 g, 34 mmol) was added and the solution was stirred for 70

hrs at room temperature. The reaction mixture was subsequently neutralized (pH ~7.5) with 3M

HCl. Additional sodium borohydride (0.26 g, 6.8 mmol) was added and the mixture was stirred

for another 46 hrs. Finally, the solution was acidified with 6M HCl and extracted with

dichloromethane. The organic extract was washed with dilute HCl and water. After evaporating

41 the solvent, a solid residue was obtained. 3.6a was purified by recrystallization from benzene, while 3.6b was purified through a filter column (silica gel) using dichloromethane as the eluent.

1,3-Dihydro-3-phenylnaphtho[2,3-c]furan-1-one (3.6a). Yield = 78%. mp 153 - 154 oC (lit.3

o -1 1 153-155 C). IR (Neat) 1750 cm ; H NMR (300 MHz, CDCl3) δ 6.6 (s, 1 H), 7.4 (m, 5 H), 7.6

13 (m, 2 H0, 7.75 (s, 1 H), 7.9 (m, 1 H), 8.1 (m, 1 H), 8.55 (s, 1 H); C NMR (300 MHz, CDCl3) δ

83, 122, 123.5, 127.05, 127.1, 127.14, 128.34, 129.01, 129.04, 129.30, 129.95, 133.34, 136.45,

137.33, 143.53, 170; mass spectrum(GC-MS), m/e (%) M+ 260 (80), 231 (10), 215 (40), 155

(100).

1,3-Dihydro-3-tert-butylphenylnaphtho[2,3-c]furan-1-one (3.6b). Yield = 78%. mp 235 – 236

o -1 1 C. IR (Neat) 1755 cm ; H NMR (300 MHz, CDCl3) δ 1.36 (s, 9 H), 6.55 (s, 1 H), 7.25 (m, 4H,

CDCl3), 7.4 (d, 2 H, J = 8.4 Hz), 7.6 (m, 2 H), 7.75 (s, 1 H), 7.9 (m, 1 H), 8.1 (m, 1 H), 8.54 (s, 1

13 H); C NMR (300 MHz, CDCl3) δ 31.28, 34.73, 82.76, 122, 123.77, 125.94, 126.95, 127.07,

128.35, 128.96, 129.95, 133.34, 134.24, 136.44, 143.62, 152.52, 170.35; mass spectrum, m/e (%)

M+ 316 (32), 301(48), 259 (100), 215 (42), 183 (30), 161 (50), 155 (75).

Route b (synthesis of 1,3-Dihydro-3-mesitoyl[2,3-c]furan-1-one, 3.6c). Compound

3.5c (2.35 g, 7.4 mmol), zinc dust (3.42 g, 52 mmol), water (7 mL), and glacial acetic acid (25 mL) were taken into a round bottom flask fitted with a condenser. The reaction mixture was refluxed for 2 hrs and then allowed to cool to room temperature. Dilution with water yielded a white precipitate, which was filtered and subjected to filter column (silica gel, dichloromethane)

o 1 to obtain 3.6c (1.5 gm, yield - 67%). mp 144 -145 C. H NMR (300 MHz, CDCl3) δ 1.7 (s,

broad, 3 H), 2.25 (s, 3 H), 2.65 (s, broad, 3 H), 6.75 (s, broad, 1 H), 7.0 (s, 2 H), 7.6 (m, 3 H),

13 7.8 (m, 1 H), 8.05 (s, 1 H), 8.55 (s, 1 H); C NMR (300 MHz, CDCl3) δ 20.9, 79.9, 120.7, 124.7,

126.8, 126.9, 128.2, 129.0, 129.1, 130.0, 133.1, 136.4, 139.0, 143.4, 170.8; mass spectrum (GC-

MS), m/e M+ 302 (35), 281 (10), 257 (10), 243 (20), 228 (20), 207 (50), 147 (100).

Lactols (3.7a-d). Dry THF (25 mL) was added to a round bottom flask containing 3.6a

(7.3 mmol) under argon. The container was placed on an ice bath (-5 – 0 oC) and stirred for few

minutes. A solution of the phenylmagnesium bromide in ether (29.1 mmol) was added drop wise

over 20 min. Stirring was continued for another 0.5 hr at 0 oC. The reaction mixture was poured

onto a saturated solution of ammonium chloride. The organic layer was extracted with ether. The

resulting solution was dried over sodium sulfate and evaporated to get an oily product (crude

3.7a). Crude 3.7b-d were synthesized similarly, reacting 3.6b-d with the corresponding Grignard reagents. The Grignard reagents were freshly prepared following a general procedure. Since bromomesitylene reacts slowly with magnesium in ethyl ether, a modified procedure was employed to synthesize the mesityl Grignard reagent.11

Crude 3.7a was purified by washing with hexanes. Crude 3.7b-d were passed through a pad of alumina with dichloromethane. Evaporating the solvent produced solid white products.

1,3-Dihydro-1,3-diphenylnaphtho[2,3-c]furan-1-ol (3.7a). Yield = 82%. mp 158-160 oC (lit.3

155-160 oC). mass spectrum (DIP-MS), m/e (%) M+ 338 (70), 320 (100).

1,3-Dihydro-1-(p-tert-butylphenyl)-3-phenylnaphtho[2,3-c]furan-1-ol (3.7b). Yield = 56%.

mass spectrum (DIP-MS), m/e (%) M+ 259 (70), 289 (15), 319 (15), 337 (100), 361 (35), 376

(25), 394 (13).

1,3-Dihydro-1,3-bis-(p-tert-butylphenyl)-naphtho[2,3-c]furan-1-ol (3.7c). Yield = 62%. mass

spectrum (DIP-MS), m/e (%) M+ 450 (5), 432 (100), 417 (60), 393 (15). 377 (20).

1,3-Dihydro-1,3-dimesitoylnaphtho[2,3-c]furan-1-ol (3.7d). Yield = 86%, mass spectrum

(DIP-MS), m/e M+ 422 (25), 404 (15), 389 (5), 301 (25), 287 (100), 273 (10), 256 (30).

1,3-Diphenylnaphtho[2,3-c]furan (3.8). A mixture of 3.7a (2.43 g, 7.1 mmol) and glacial acetic acid (25 mL) was heated on a hot plate to a moderate temperature (~ 60 oC) and

stirred with a glass rod until all the white lactol was converted to the red furan. The reaction

mixture was then cooled to ~10 oC in an ice-water bath. This, upon filtration and washing with

petroleum ether, produced 1.66 g of 3.8 (yield - 72%). mp 146-147 oC (lit. 148-154 oC). mass spectrum (DIP-MS), m/e (%) M+ 320 (100), 289 (30), 259(40).

6,8,15,17-Tetraphenyl-6,17:8,15-dioxido-6,6a,7,7a,8,15,15a,16,16a,17-decahydro-

7,16-heptacene quinone (3.9).19,20 To a suspension of furan (3.8) (1.16g, 3.62 mmol) in 30 mL

of benzene, p-benzoquinone (0.196g, 1.81 mmol) was added and the reaction mixture was stirred

for 0.5 h at room temperature. The white solid precipitate of 13 was filtered and washed with

1 hexane. The yield was 86% (1.17 g). H NMR (300 MHz, CDCl3) δ 3.4 (s, 2 H), 4.4 (s, 2 H),

6.72 (d, 4 H, J = 7.8 Hz), 7.15 (t, 4 H, J = 7.8 Hz), 7.25-7.35 (m, 4 H), 7.45 (dd, 2 H, J = 6,6 Hz),

7.56 (s, 2 H), 7.6-7.7 (m, 8 H), 7.75 (dd, 2 H, J = 6, 6 Hz), 7.9 (m, 4 H); 13C NMR (300 MHz,

CDCl3) δ 58, 64, 89, 92, 118, 122, 126, 126.69, 126.79, 127.97, 128.19, 128.25, 128.37, 128.64,

129.18, 132.51, 132.90, 134.12, 136.59, 143.71, 144.66, 204; mass spectrum (MALDI), m/e M+

750.39, 749.37, 732.38, 731.37.

6,8,15,17-Tetraaryl-7,16-quinone (3.10a-c). Compound 3.10a was synthesized according to a literature procedure starting from 3.8.4 Compounds 3.10b and 3.10c, however,

were prepared by a modified method during which both Diels Alder reaction and the subsequent

dehydration steps were carried out in situ.

p-Toluene sulfonic acid (1.52 g, 8 mmol) and p-benzoquinone (0.108 g, 1 mmol) were

added to a suspension of 3.7b or 3.7c (2 mmol) in 35 mL of dry benzene, and the reaction

mixture was refluxed for 15 hrs on an oil bath. Subsequently, the mixture was allowed to cool to

44 room temperature and was diluted with benzene. The organic extract was washed with saturated sodium bicarbonate and brine solution. Finally, the organic layer was concentrated on a rotary evaporator and subjected to column chromatography (silica gel). Elution was started with 20%

(vol.) of dichloromethane in hexane and ended with 50% (vol.) of dichloromethane in hexane.

Yellowish-brown product was obtained after evaporating the solvent.

6,15-bis-(p-tert-Butylphenyl)-8,17-diphenyl-7,16-quinone (3.10b). Yield = 41 %, 1H

NMR (300 MHz, CDCl3) δ 1.45 (s, 18 H), 7.3 – 7.5 (m, 22 H), 7.7 – 7.9 (m, 4 H), 8.15 (s, 2 H),

8.25 (s, 2 H); mass spectrum (MALDI), m/e M+ 827.42, 826.43, 825.43, 824.41, 823.42; HRMS

(FAB) m/z = 825.3736 (M + H+), calcd m/z = 825.373256.

6,8,15,17-tetrakis-(p-tert-Butylphenyl)-7,16-quinone (3.10c). Yield = 40%, 1H NMR

(300 MHz, CDCl3) δ 1.464 (s, 36 H), 7.29 (d, 8 H, J = 8 Hz), 7.4 (dd, 4 H, J = 6.4, 6.4 Hz), 7.47

(d, 8 H, J = 8.4 Hz), 7.8 (dd, 4 H, J = 6.4, 6.4 Hz), 8.018 (s, 4 H); mass spectrum (MALDI) m/e

M+ 936.77; HRMS (FAB) m/z = 937.4988 (M + H+), calcd m/z = 937.498457.

2,3-Dimesitoylnaphthalene (3.11). Compound 3.7d leads to 3.11 under conditions used for the synthesis of 3.10b or 3.10c form 3.7b or 3.7c, respectively. Yield = 60%. 1H NMR (300

MHz, CDCl3) δ 2.0 (s, 6 H), 2.05 (s, 3 H), 2.22 (s, 3 H), 2.32 (s, 3 H), 2.7 (s, 3 H), 6.85 (s, 2 H),

6.97 (s, 1 H), 7.1 (s, 1 H), 7.5 – 7.7 (m, 2 H), 7.8 – 8.0 (m, 4 H); mass spectrum (DIP), m/e M+

420(70), 405(60), 389(80), 299(75), 285(55), 273(100); HRMS (FAB) m/z = 421.2166 (M +

H+), calcd m/z = 421.216755.

7,16-Dihydro-6,8,15,17-tetraphenylheptacene (3.1). To a suspension of 3.10a (0.22 g,

0.3 mmol) in 10 mL THF cooled to 0oC, 1M solution of borane.THF (0.9 mL, 0.9 mmol) was

added drop wise under an argon-atmosphere. The mixture was allowed to stir for an hour at room

temperature. The temperature was increased very slowly to 50 oC and the mixture was stirred for

45 another 12 hrs. The reaction mixture was allowed to cool to room temperature and an additional amount of 1 M borane.THF solution in THF (0.9 mL, 0.9 mmol) was added. The temperature was slowly increased again to 50 oC and stirring was continued for 12 hrs. After cooling to room

temperature, the reaction mixture was quenched with methanol, the product concentrated on a

rotary evaporator, and subjected to column chromatography (silica gel). Elution was started with

5% (vol.) of dichloromethane in hexane and ended with 20% (vol.) of dichloromethane in

hexane. Evaporation of the solvent produced pure 3.1 in the form of a yellowish white solid

1 (0.08 gm, yield - 38%). H NMR (300 MHz, CDCl3) δ 3.9 (s, 4 H), 7.24-7.32 (m, 18 H, CDCl3),

13 7.40 (m, 12 H), 7.74 (dd, 4 H, J = 6, 6 Hz), 7.96 (s, 4 H); C NMR (300 MHz, CDCl3) δ 32,

124.97, 125.31, 126.97, 128.19, 128.32, 130.49, 130.75, 131, 133.5, 135.8, 139; mass spectrum

(DIP-MS), m/z M+ (%) 685 (60), 684 (100), 607 (50), 530 (30); mass spectrum (MALDI), m/e

M+ 684.51, 685.51, 686.51; HRMS (FAB) m/z = 685.2899 (M + H+), calcd m/z = 685.289526.

7,16-Dihydro-6,15-bis-(p-tert-butylphenyl)-8,17-diphenylheptacene (3.2). Compound

3.2 was obtained from 3.10b (yield -30%), using the same procedure described for the synthesis

1 of 3.1. H NMR (300 MHz, CDCl3) δ 1.4 (s, 3 H), 1.45 (s, 15 H ), 3.9 (s, 3 H), 4.1 (s, 1 H) 7.15

(m, 3 H), 7.3 (m, 8 H), 7.42 (m, 10 H), 7.75 (m, 5 H), 7.95 (s, 2 H), 8 (s, 2 H); 13C NMR (300

MHz, CDCl3) δ 31.8, 33.7, 35, 125, 125.3, 128.2, 130.2, 130.9, 131.1, 134, 135.5., 136, 139.7;

mass spectrum (MALDI), m/e M+ 797.68, 796.68; HRMS (FAB) m/z = 797.4148 (M + H+),

calcd m/z = 797.414727.

7,16-Dihydro-6,8,15,17-tetrakis-(p-tert-butylphenyl)heptacene (3.3). Compound 3.3 was obtained from 3.10c (yield - 33%), using the same procedure described for the synthesis of 1

o o 1 (instead of 50 C, the temperature was raised to 60 C in this case). H NMR (300 MHz, CDCl3)

δ 1.45 (s, 36 H), 4.1 (s, 4 H), 7.3 (m, 12 H), 7.45 (d, 8 H, J = 8 Hz), 7.7 (dd, 4 H, J = 6, 6 Hz), 7.8

13 (s, 4 H); C NMR (300 MHz, CDCl3) δ 32, 34, 35, 124.6, 125, 125.2, 128.2, 130, 130.5, 131.5,

133.5, 135.8, 136.2; mass spectrum (MALDI), m/e M+ 908.66; HRMS (FAB) m/z = 909.5402

(M + H+), calcd m/z = 909.539928.

6,8,15,17-Tetraphenylheptacene-6,17:8,15-diepoxide (3.12). Compound 3.8 (100 mg,

0.31 mmol) and 1,2,4,5-tetrabromobenzene (270 mg, 0.68 mmol) were dissolved in dry toluene

(10 mL) and cooled to –78 °C with stirring under Ar. n-BuLi (1.6 M in hexanes, 1.5 mmol) was

diluted in 2 mL of dry hexanes and added dropwise to the cold solution. The reaction mixture

was allowed to warm to room temperature and it was stirred overnight. The reacting mixture was

then cooled again to –78 °C, and another portion of 1.6 M n-BuLi (3.0 mmol) was added

dropwise. After warming to room temperature and stirring overnight again, the reaction was quenched with 1 mL of methanol. Dichloromethane (10 mL) was added, and the resulting suspension was washed twice with saturated NaCl. The organic extracts were dried over Na2SO4.

Evaporation of the solvent, followed by silica-gel column using 20% of dichloromethane in hexanes yielded 0.151g (10.5%) of trans 3.12a and 0.07g (4.5%) of cis 3.12b isomers.

1 trans-isomer (3.12a). H NMR (300 MHz, CDCl3) 7.37 (dd, 4 H), 7.415 (s, 2 H), 7.51 (m, 4 H),

7.61 (m, 16 H), 7.92 (s, 4 H), 7.95 (d, 2 H); mass spectrum (MALDI-MS) m/z 715.56 (100),

(FAB) 377 (45), 433 (100), 489 (50), 698 (5), 715 (10); HRMS (FAB) m/z 715.26370 (M+),

calcd m/z 715.26371.

1 cis-isomer (3.12b). H NMR (300 MHz, CDCl3) 7.3 (dd, 4 H), 7.405 (s, 2 H), 7.5 (m, 4 H), 7.6

(m, 16 H), 7.91 (s, 4 H), 7.94 (d, 2 H); mass spectrum (MALDI-MS) 715.53 (100), HRMS

(FAB) m/z 715.26377 (M+), calcd m/z 715.26371.

X-ray Crystallography. Data collection was performed at 150 K with Mo Kα using a

Bruker AXS SMART platform diffractometer.12 Intensity data were collected using three

47 different φ settings and 0.3° increment ω scans, 2θ < 56.58º, which corresponds to more than a

hemisphere of data. The SAINT12 program was used for data integration, and corrections for

absorption and decay were carried out using the SADBS program.13 The X-ray structures was

determined by direct method,14 and refinement was done by full matrix least squares13 on F2 using all 7698 unique data. The refinement included anisotropic thermal parameters for all

2 hydrogen atoms. The final refinement converged to wR2 = 0.1541 (for F , all data) and R1 =

0.0737 [F, 1608 reflections with I > 2σ(I)].

3.4 Conclusion

Two synthetic approaches were attempted to yield the heptacene core. Though neither was successful, the reduction of substituted heptacene quinones 3.10a-c yielded a new class of compounds, substituted dihydroheptacenes (3.1-3.3). A mesityl substituted lactone (3.6c) shows temperature dependence in 1H-NMR due to hindered rotation. The dibenzyne approach yielded

cis- and trans- diepoxides, which are distinguishable in 1H-NMR.

3.5 References

* Partially adopted from: Mondal, R.; Shah, B. K.; Neckers, D. C. J. Org. Chem. 2006, 71, 4085.

1. Payne, M. M.; Parkin, S. R.; Anthony, J. E. J. Am. Chem. Soc. 2005, 127, 8028. 2. Miller, G. P.; Briggs, J. Tetrahedron Lett. 2004, 45, 477-481 3. Cava, M. P.; VanMeter, J. P. J. Org. Chem. 1969, 34, 538-545. 4. Miller, G. P.; Briggs, J. Org. Lett. 2003, 5, 4203. 5. Goodings, E. P.; Mitchard, D. A.; Owen, G. J. Chem. Soc., Perkin Trans. 1 1972, 11, 1310. 6. Chandross, E. A.; Sheley, C. F. J. Am. Chem. Soc. 1968, 90, 4345 – 4354. 7. Lu, J.; Ho, D. M.; Vogelaar, N. J.; Kraml, C. M.; Pascal, R. A., Jr. J. Am. Chem. Soc. 2004, 126, 11168- 11169. 8. Hart, H.; Nwokogu, G. J. Org. Chem. 1981, 46, 1251-1255. 9. Carlson, R. G.; Srinivasachar, K.; Givens, R. S.; Matuszewski, B. K. J. Org. Chem. 1986, 51, 3978.

10. Hauser, C. R.; Tetenbaum, M. T.; Hoffenberg, D. S. J. Org. Chem. 1958, 23, 861. 11. Hawkins, R. T.; Lennarz, W. J.; Snyder, H. R. J. Am. Chem. Soc. 1960, 82, 3053. 12. Bruker SMART (Ver. 5.05) and SAINT-Plus (Ver. 7.08); Bruker AXS Inc.: Madison, WI, 1999. 13. Sheldrick, G. M. SADABS, Program for the Empirical Absorption Correction of Area Detector Data; University of Göttingen: Göttingen, Germany, 1996. 14. Sheldrick, G. M. SHELXTL (ver. 6.12); Bruker AXS, Inc.: Madison, WI, 2000.

CHAPTER 4. PHYSICAL PROPERTIES OF DIHYDROHEPTACENES*

4.1 Introduction

Getting rid of the “light bulb” is the dream of every homeowner. In addition to incredible

inefficiencies, incandescent bulbs are, seemingly, always ‘burned out’. Even fluorescent bulbs

are less than 70% energy efficient. The world awaits the white light organic light emitting diode

(OLED). Due to this, the development of new electroluminescent materials, potentially for

OLED applications, has become an active area of research in the past few decades.1 In order to achieve the full color display three primary colors; red, green, and blue, are required. The stability and efficiency of materials emitting green and red have reached commercial levels,2 whereas the stability and efficiency of blue light emitting materials remain a challenge.3

Anthracene is one of the earliest reported luminescent materials and emits violet both in the solid state and in solution. Derivatives like 9,10-di-2-naphthylanthracene and 2-tert-butyl-

9,10-di-2-naphthylanthracene doped with aggregation-resistant 2,5,8,11-tetra-tert-bytylperylene

(TBP) have been used successfully as blue emitters in commercial OLED products.4 On the basis

of the efficiency reported and stability of anthracene derivatives, a new class of emitters [7,16-

dihydro-6,8,15,17-tetraphenylheptacene (4.1), 7,16-dihydro-6,15-bis-(p-tert-butylphenyl)-8,17- diphenylheptacene (4.2), and 7,16-dihydro-6,8,15,17-tetrakis-(p-tert-butylphenyl)heptacene

(4.3)] is designed, synthesized, and extensively studied (Chart 4.1). They are dihydroheptacene derivatives in which two anthracene moieties are attached with each other via two methylene bridges.

R1 R2

R3 R4

4.1 R1, R2, R3, R4 = phenyl 4.2 R1, R4 = phenyl; R2, R3 = p-tert-butylphenyl 4.3 R , R , R , R = p-tert-butylphenyl 1 2 3 4

Figure 4.1. Chemical Structures of 7,16-dihydroheptacene derivatives (4.1 – 4.3).

Photophysical properties of these compounds in solution and in a poly(methylmethacrylate), [PMMA], matrix were studied. Compounds 4.1-4.3 show blue emission (λem = 420 – 428 nm) in dilute solution (ΦF = 0.15-0.21) and in the PMMA matrix (ΦF

5 0 7 -1 0 8 -1 = 0.37-0.44). Their radiative (k R ~6 – 8 x 10 s ) and nonradiative (k NR ~3 – 3.6 x 10 s ) rate constants of deactivation from the singlet surface were found to be similar to those of anthracene

0 8 - 0 8 -1 (k R ~2 x 10 s1 and k NR ~6 x 10 s ). These compounds also show high stability and retain color purity after aging under ambient condition and annealing at 110 °C for more than 24 hours.5

The electrochemical characteristics of 4.1-4.3 and performance of 4.1 in OLED devices

were also studied. The device containing 4.1 as active emitting layer shows green emission (510

and 550 nm) with a maximum external quantum efficiency of 0.26%. The characteristics of these

devices are unusual in that the emission does not come from the molecular units even at a dopant

concentration as low as 1% or from the excimer6 or exciplex formation.7 It rather comes from the

electromer. This prompted us to investigate and understand the nature of light emitting species

involved in photoluminescence (PL) and electroluminescence (EL) processes of this class of

material. Detailed transient spectroscopic studies of 4.1-4.3 are also discussed.

4.2 Results and Discussion

4.2.1 Photoluminescence in Solution

The absorption spectra of 4.1-4.3 were similar in shape, showing a π - π* band at 372 nm (ε372 =

4 -1 -1 1.3 x 10 M cm for 4.1 in CH2Cl2). Dilute solutions of these compounds showed blue emission

in several solvents such as hexane, dichloromethane, acetonitrile, toluene and methanol. The fluorescence spectra of 4.1-4.3 recorded in CH2Cl2 are shown in Figure 4.1. The absorption maxima (Amax), emission maxima (λmax), quantum yields (ΦF), and lifetimes (τF) of fluorescence

of 1-3 measured in dichloromethane are presented in Table 4.1. The λmax of 4.2 (426 nm) and 4.3

(428 nm) are red shifted only slightly from that of 4.1 (424 nm), indicating almost no difference

in delocalization of π electrons in these compounds. This indicates that be it a phenyl group or a

substituted phenyl group such as the p-tert-butylphenyl group, it lies almost orthogonal to the anthracene planes and there is no change in the geometry of the molecules.

1.0 1 2 0.8 3

0.6

0.4 Intensity

0.2

0.0

400 450 500 550 600 Wavelength (nm)

Figure 4.2. Normalized fluorescence spectra of 4.1 (black), 4.2 (red), and 4.3 (green) in CH2Cl2.

The ΦF values of 4.1-4.3 were measured exciting the molecules at 372 nm, and were in

the range of 0.15 – 0.21 in CH2Cl2 relative to the ΦF of 9,10-diphenylanthracene (0.90 in cyclohexane).8 Fluorescence decays of 4.1-4.3 were fitted with monoexponential functions, indicating emission from the singlet excited state (S1) in each case. The τF values were found to be ~2.5 ns in CH2Cl2 (argon degassed solution).

a Table 4.1. Photophysical Properties of 4.1 – 4.3 in CH2Cl2.

2 0 -1 0 -1 Compound Amax (nm) λmax (nm) ΦF τF (ns) χ k R (s ) k NR (s )

4.1 354, 372, 392 402, 424 0.21 ± 0.01 2.58 1.02 8.1 x 107 3.1 x 108

4.2 354, 372, 392 404, 426 0.16 ± 0.01 2.67 1.30 6 x 107 3.14 x 108

4.3 353, 373, 393 409, 428 0.15 ± 0.01 2.35 1.27 6.4 x 107 3.61 x 108

a Excitation wavelength = 372 nm for ΦF and τF. ΦF values are relative to that of 9,10- diphenylanthracene (0.90 in cyclohexane). τF values are measured from argon saturated solutions and decay was monitored at the corresponding λmax.

No phosphorescence was observed for 4.1, even in a frozen matrix of 20% (v/v) methanol

in ethanol at 77 K. This indicates ~80% of the excited molecules decay without radiation in

0 0 solution. Rate constants of radiative (k R) and nonradiative (k NR) decay of the fluorescence of

0 4.1 – 4.3 were calculated from the ΦF and τF values measured from CH2Cl2 solutions. The k R of

7 -1 deactivation of the S1 state (ΦF/τF) were found in the range of 6 - 8 x 10 s . The nonradiative

0 8 pathway of deactivation of the S1 state was observed to be 4 – 6 times faster (k NR ~3 – 3.6 x 10

-1 0 8 s ) than the radiative pathway. The observation is similar to the case of anthracene (k R ~2 x 10

- 0 8 -1 13 0 7 - 0 8 -1 14 s 1 and k NR ~6 x 10 s ) and anthanthrene (k R ~8 x 10 s1 and k NR ~2.5 x 10 s ).

0 Furthermore, the calculated k NR values of 4.1-4.3 may represent the rate of S1 Æ T1 intersystem

crossing (ISC), because the rate of S1 Æ T1 ISC for rigid aromatic molecules is shown to be in

6 8 -1 9-10 the same range (~10 – 10 s ) and internal conversion from S1 cannot compete with it.

Figure 4.3. Fluorescence decay of 4.3 monitored at λmax = 424 nm in CH2Cl2; excitation wavelength = 372 nm.

Hexane 1.0 Toluene CH2Cl2 MeCN 0.8 MeOH

0.6

Intensity 0.4

0.2

0.0

400 450 500 550 600 Wavelength (nm)

Figure 4.4. Normalized fluorescence spectra of 4.1 in different solvents.

A negligible solvatochromic effect was observed on the fluorescence of 4.1 (Figure 4.4).

The polarity of the solvent also showed no systematic effect on its fluorescence quantum yield or

lifetime (Table 4.2). The ΦF of 4.1 in toluene (0.41) was the highest among the solvents

employed and is close to the absolute quantum yield value (ΦF = 0.44) measured in the solid state (vide infra). It is also noted that the τF of 4.1 was found to be much higher in acetonitrile

(6.87 ns) than in other solvents.

Table 4.2. Photophysical Properties of 4.1 in Different Solvents

Solvent Amax (nm) λmax (nm) τF (ns) ΦF

Hexane 350, 368, 388 396, 420 3.80 0.28

Toluene 353, 371, 391 402, 424 3.24 0.41

Dichloromethane 354, 372, 392 402, 424 2.58 0.21

Methanol 351, 368, 388 398, 420 3.68 0.30

Acetonitrile 351, 369, 389 400, 421 6.87 0.27

The spectroscopic properties of 4.1-4.3 are comparable with those of anthracene,

indicating that absorption and emission are due to the anthracene moieties present in the

compounds. Photoluminescences of 4.1-4.3 were observed to be concentration dependent. For

example, additional peaks centered at 480 and 510 nm become predominant in the fluorescence

spectra of 4.1 (Figure 4.5) at higher concentration (1.1×10-3 M or higher in dichloromethane),

while these peaks were absent in the fluorescence spectra recorded at lower concentration

-4 (5.5×10 M solution; λem = 402 and 424 nm). Since no change in absorption was found in these

concentration ranges, the red shift of about 80-85 nm in the emission spectra can be considered due to an excimer. The latter is formed by the interaction of an excited molecule with an

55 unexcited counterpart.11 Moreover, self quenching occurs in these compounds and emission

becomes very weak at very high concentration (>1.0×10-2 M).

Figure 4.5. Normalized emission spectra of compound 4.1 (a) red: PL spectra in dichloromethane, 5.5×10-4 M; (b) blue: PL spectra in dichloromethane, 5.5×10-3 M; and (c) black: EL spectra of device 2 (5% of 4.1 doped in 1:1 PVK:PBD).

The X-ray crystal structure (Figure 4.6) indicates that the molecule of 4.1 is V-shaped

with an angle of 131.51o between the two anthracene arms. Crystal molecular packing shows that

the adjacent planar anthracene moieties of two molecules are parallel and closely spaced. The

possibility of such a close proximity between two molecules seems the reason for the excimer formation at higher concentration. The distance between the two parallel anthracene moieties was found to be about 4.28 Å.

Figure 4.6. Molecular packing of X-ray crystal of 4.1: a view showing π-π interaction between two anthracene moieties (distance = 4.277 Å).

4.2.2 Photoluminescence in the Solid State

The solid state emission spectra of 4.1-4.3 were recorded in thin films of PMMA (Figure 4.7)

and the quantum yields measured using an integrating sphere (Table 4.3).12 The solid state

emission of these compounds (λmax = 420 – 426 nm) are similar to those recorded in CH2Cl2 solution. However, it appears that solid state emission (ΦF = 0.37 - 0.44) is more efficient than

the solution emission in each case. Anthanthrene and its derivatives show similar trend.13 This observation can be explained based on the involvement of a T2 state in the relaxation pathway,

similar to the case of anthracene and anthanthrene.13 There is a relatively larger energy gap

14 between S1 (3.1 eV) and T1 (1.8 eV) states of anthracene. It has been suggested that the presence of the T2 state which is almost isoenergetic with the S1 state provides a facile pathway

for intersystem crossing for anthracene in solution, causing the molecule to lose majority of its

10 excited state energy in dark processes via the T2 and T1 states. A similar situation may pertain

with the 7,16-dihydroheptacene derivatives in solution. However, in the solid state the

configuration of the excited states of these molecules may be such that the T2 state is slightly

higher in energy than the S1 state minimizing the S1-T2 intersystem crossing process. In such a

case, energy transfer from the S1 state to the T1 state would not be efficient due to the reasonably

57 high energy difference. This would cause enhanced fluorescence in the solid state as is observed experimentally.

1.0 1 2 0.8 3

0.6

Intensity 0.4

0.2

0.0

400 450 500 550 Wavelength (nm)

Figure 4.7. Normalized solid state fluorescence spectra of 4.1 (black), 4.2 (red), and 4.3 (green) recorded in thin films of PMMA.

Table 4.3. Photophysical Properties of 4.1-4.3 Measured in the Solid State. a

Compound λmax (nm) ΦF

4.1 400, 420 0.44

4.2 404, 424 0.39

4.3 404, 426 0.37

a Thin films of PMMA were used as matrix; excitation wavelength = 372 nm; the ΦF were measured using an integrating sphere (errors within 15%).

The effects of aging and annealing on the solid state emission spectra of 4.1-4.3 were evaluated from thin films of PMMA containing the samples. The thin films were exposed to ambient light for 7 days and emission recorded. Similarly, the emissions were also recorded after heating the thin films at 110 oC for 24 hours and cooling them down to room temperature. The

58 emission spectra of 4.1-4.3 remained unchanged after aging or annealing (shown for 4.1 in

Figure 4.8), indicating a higher stability of these compounds in the solid state in comparison to several other blue emitters such as fluorene and derivatives.15-16

Figure 4.8. Normalized solid state fluorescence spectra recorded from thin films of PMMA containing 4.1: (i) pristine, (ii) after exposing the film for 7 days at ambient condition, and (iii) after heating the film at 110 °C for 24 hours and cooling it down to room temperature.

4.2.3 Electrochemical Properties

The cyclic voltammetry (CV) graphs of 4.1-4.3 were measured with a positive scan from 0 to 1.7

V and negative scan from 0 to -2.0 V with the compounds in 0.1 M tetrabutyl ammonium hexafluorophosphate in anhydrous dichloromethane. The CV measurements revealed three

oxidation potentials (E1/2) of 4.1 at +1.07 V, +1.25 V, and +1.54 V without any reduction

potential. The E1/2 peaks of 4.2 and 4.3 are similar to that of 4.1 (Table 4.4). It is well known that

anthracene (+1.16 and +1.50 V) and 9,10-diphenylanthracene (+1.22 and +1.60 V) show two reversible monoelectronic oxidation potentials.17,18 The first oxidation peak (~1.20 V) is due to the generation of a radical cation, which undergoes second monoelectronic oxidation to form a

59 dication at higher potentials (~1.50 V). Similar formation of monocation and dication seems responsible for the oxidation peaks of 4.1-4.3. The observation of three distinctive oxidation potentials in the case of 4.1-4.3 may be due to the two spatially separated anthracene rings, which are not electrochemically equivalent.

The HOMO energy levels of 4.1-4.3 were calculated from the CV measurements and by comparison of their oxidation potentials with that of ferrocene (4.8 eV below the vacuum level).19 LUMO levels were estimated from the onset of the absorption spectra.20 The HOMO-

LUMO energy gaps (EH-L) of 4.1-4.3 were found to be in the range of 3.01-3.06 eV (Table 4.4).

The structure of 4.1 was optimized using density functional theory (DFT) method [B3LYP/6-

31G(d)]. The HOMO of 4.1 appears to be localized on the anthracene moieties, while the LUMO

is more distributed in the molecule and has a significant contribution from the central methylene

bridge (Figure 4.9). A similar diffused LUMO was also calculated for fused pyrazines that are

21 structurally similar to dihydroheptacenes. The DFT calculated EH-L value for 4.1 (3.43 eV) is

close to the electrochemically estimated value (3.06 eV).

Table 4.4. Electrochemical Properties and HOMO-LUMO Energy Gaps (EH-L) of 4.1-4.3

a a a b Compound E1/2(1) E1/2(2) HOMO LUMO EH-L (V) (V) (eV) (eV) (eV) 4.1 +1.07, +1.25 +1.54 -5.87 -2.81 3.06 (3.43)c 4.2 +1.04, +1.22 +1.50 -5.84 -2.80 3.04

4.3 +1.04, +1.24 +1.53 -5.84 -2.83 3.01 aDetermined from cyclic voltammetry (solvent- dichloromethane). bEstimated from the onset of the absorption spectra. cCalculated using density functional theory (DFT) method [B3LYP/6- 31G(d)].

(a) (b)

Figure 4.9. (a) HOMO and (b) LUMO of the geometry optimized structure of 4.1 [B3LYP/6- 31G(d)].

4.2.4 Electroluminescence and Device Characteristics

OLED devices consisting ITO/PEDOT (~12 nm)/emissive layer (~102 nm)/LiF(~1 nm)/Al (30 nm)/Ag (~150 nm) layers were constructed. A 1:1 mixture of poly(N-vinylcarbazole), (PVK, a hole-transporting material) and 2-tert-butylphenyl-5-biphenyl-1,3,4-oxadiazole (PBD, an electron-transporting material) was used as the host material, while different concentrations of

4.1 were used as emissive dopants. Devices 1, 2, 3, and 4 contained 1%, 5%, 10%, and 0% of

4.1, respectively. The EL spectrum of the device containing 5% 4.1 is shown in Figure 4.5.

The device 4 (undopped) exhibited very weak blue emission (peak centered at 440 nm) originating from the host materials (Figure 4.10).22 The device 1 showed intense peaks centered

at 515 nm and 550 nm, with smaller peaks centered at 404 and 424 nm. The latter peaks are

residual molecular emission from 4.1 and emission from the host materials. The intensity of

those lower wavelength peaks gradually diminishes as the weight percent of the dopant increased

from 1% to 10%. Interestingly, the higher wavelength peaks (515 and 550 nm) are not due to the

excimer of 4.1, because they are considerably red shifted than the emission of the excimer (480

and 510 nm). This shift may be explained on the basis of the formation of electromer of 4.1,

61 which was observed in all devices (device 1 through 3), irrespective of the amount of the dopant used.23 An electromer represents an intermolecular electron-hole entity, which can be

represented as Em+/Em− (Em = emitter).

Figure 4.10. EL spectra of the device with the structure of ITO/PEDOT/PVK:PBD, 4.1/LiF/Al/Ag: (Device 1, 2, 3, and 4 containing 1%, 5%, 10%, 0% of 4.1, respectively).

While excimer formation is facilitated by the short inter-ion distance and efficient π

orbital overlap, electromer formation is believed to depend on the inter-ion separation and steric

hindrance present in the molecule.24-26 It has been suggested that large molecules with small

chromophoric units facilitate formation of efficient electromers, as the charge separation on

localized excited states become major pathways for electromeric emission.24-25,27 Nevertheless, anthracene is also known to exhibit electromeric emission at 540 nm.27 Electromer emission was

observed when 20% of anthracene was doped in neutral polycarbonate binder. It is interesting to

note that electromeric emission was observed even from the device that contained only 1% of

4.1.

At higher doping concentration (device 3 containing 10% of 4.), the electromeric emission became very weak. This may be due to the high concentration quenching, where the emitted light can be absorbed by neighboring molecules or the excitons are de-excited via energy transfer mechanisms. In the case of 4.1, a dopant concentration of 1 to 5% seemed to give the optimal result.

The current-voltage (I–V) and voltage-luminescence (V–L) curves for devices 1-4 are shown in Figures 4.11 and 4.12, respectively. The maximum brightness of 190 cd/m2 was observed at a current density of 34 mA/cm2 at operating voltage of 19 V for device 1. The

external quantum efficiency was found to be maximum for device 2 (0.26%), while that was

lower for device 1 (0.22%) and 3 (0.19%). The turn-on voltage was also higher for device 3,

while that was lower for device 1. This indicates that overall device performances decreased with

an increase in the doping concentration of 4.1.

Figure 4.11. Current density–Voltage (I– Figure 4.12. Voltage–Luminance (V–L) V) curves for devices 1, 2, 3, and 4. curves for devices 1, 2, 3, and 4.

4.2.5 Transient Spectroscopy

Transient spectroscopic studies were performed to understand the photophysical behavior of 4.1-

4.3. The SnÅS1 absorption spectra obtained by ultrafast pump-probe experiments (instrument

63 response function ~ 130 fs) of 4.1 are presented in Figure 4.13. Excitation of a degassed solution of 4.1 (1 × 10-4 M in dichloromethane) at 340 nm produced a broad transient absorption in the

500-650 nm region, which is assigned to the S1 state of 4.1 because of its similarity with the

26 reported S1 absorption of anthracene. Similar spectra were recorded in the case of 4.2 and 4.3

in dichloromethane.

Time profiles monitored at 570 nm for 4.1 showed bi-exponential decay. Similar bi-

exponential decays were observed in the case of 4.2 and 4.3 monitored at 580 nm. These

lifetimes (τ1 and τ2) are summarized in Table 4.5. The shorter decay times (τ1) can be assigned to

28 the reorganization of the aromatic ring systems or vibrational relaxation on the S1 surface.

Longer lifetimes calculated to be in the range of 0.23–0.33 ns can be assigned to the S1 state of

4.1-4.3. There are no differences among 4.1-4.3 in terms of the absorption and the lifetimes of their S1 states.

Figure 4.13. Transient absorption spectra obtained from ultrafast pump-probe experiments of 4.1 in dichloromethane (1 × 10-4 M), recorded 0.20 ps (blue), 0.90 ps (green), 4.15 ps (red), and 129 ps (black) after the laser pulse (excitation at 340 nm). Inset: decay profile monitored at 570 nm.

Table 4.5. Vibrational relaxation times (τ1) and lifetimes of the S1 state (τ2) of 4.1-4.3 recorded in dichloromethane

Compound τ1 τ2 (ps) (ps) 4.1 21±3 257±17

4.2 9±1 229±10

4.3 16±1 330±18

It is also noted that about 40% of the S1 molecules of 4.1 do not come back to the S1 state

after they are excited to a certain Sn level as indicated by decay profile not coming back to the

zero level (see the decay profile provided in the inset of Figure 4.13). This may suggest that

about 40% of the excited S1 state decomposes. Although the experiments were performed under argon, a small amount of diffused oxygen may be reacting with the excited S1 state causing its

decomposition. It is also known that a T2 state is isoenergetic to the S1 state in the case of

anthracene and similar other polyaromatic hydrocarbons.13 Thus, it may also be possible that a

fraction of the excited S1 molecules intersystem cross into the T2 state, thus finding an alternate channel to relax than to come to the S1 state.

4.3 Experimental Section

4.3.1 Fluorescence Quantum Yields (ΦF)

Fluorescence quantum yields in solution were measured following a general method using 9,10-

-5 diphenylanthracene (ΦF = 0.9 in cyclohexane) as the standard. Diluted solutions of 4.1-4.3 (10 -

10-7 M) in appropriate solvents were used for recording the fluorescence spectra. Sample

solutions were added to quartz cuvettes and degassed for ~15 minutes. The degassed solution

had an absorbance of 0.06-0.09 at 372 nm. The fluorescence spectra of each of the sample

65 solutions were recorded 3-4 times and an average value of integrated areas of fluorescence used for the calculation of ΦF. The refractive indices of solvents at the sodium D line were used.

12 Values of ΦF in the solid state were measured following a literature method. A CH2Cl2 solution of sample was mixed with a concentrated solution of PMMA in acetonitrile in such a way that the overall concentration of the sample was ~10-4 M. The resulting solution was cast as

thin films on quartz plates and then dried in an oven at ~ 100 °C for about an hour. The plate was

inserted into an integrating sphere and the required spectra recorded. For 4.1-4.3, λex= 372 nm. It is well known that for compounds showing an overlap of the absorption and the emission spectra

(a small Stokes shift), the use of an integrating sphere results in a substantial lost of emission due to reabsorption of the emitted light. A method employed earlier was used to minimize the impact

13 of this on the calculation of the ΦF.

4.3.2 Fluorescence Lifetime (τF) Measurement

-4 -6 Solutions of 4.1-4.3 (10 -10 M in CH2Cl2) showing absorbance 0.07-0.1 at 372 nm were placed in quartz cuvettes. Fluorescence decay profiles of argon-degassed (~15 minutes) solutions were recorded using a single photon counting spectrofluorimeter. Decays were monitored at the emission maximum of the corresponding compounds. In-built software allowed the fitting of the decay spectra (χ2 = 1-1.5) and yielded the fluorescence lifetimes.

4.3.3 Fabrication and Characterization of OLEDs

The OLED devices were fabricated on ITO-coated glass substrates with a nominal sheet

resistance of 40 Ω/sq which had been ultrasonicated in acetone, methanol, and 2-propanol, dried

in a stream of nitrogen, and then plasma-etched for 60 s. Poly(3,4-ethylenedioxythiophene)

(PEDOT):PSS Al4083 (12±2 nm) was deposited by spin-casting (spin rate: 6000 rpm, 30 s; and

then thermal curing for 10 min at 190 °C). The emissive layer (102±3 nm) consisted of the host

PVK:PBD (1:1) matrix (poly(N-vinylcarbazole), 2-tert-butylphenyl-5-biphenyl-1,3,4-oxadiazole;

12 mg/mL) and dopant 4.1 at different concentrations (device 1 – 1%, device 2 – 5%, device 3 –

10%, and device 4 – 0%). This layer was also formed by spin-casting (spin rate: 1000 rpm; acceleration: 3000 rpm/s; 60 s). Subsequently, layers of LiF (~1 nm), aluminum (~30 nm), and silver (~150 nm) were deposited thermally. The operating vacuum system had a pressure of about 10-6 Torr.

Current-voltage and light output characteristics of the devices were measured in forward

bias. Device emission was measured using a silicon photodetector at a fixed distance from the sample. The response of the detector had been calibrated in this setup where the total power emitted in the forward direction was measured with a NIST traceable integrating sphere. In this case, the external quantum efficiency represents the ratio of the photons measured in the forward direction and the total charge injected in the device. All device measurements were done in a nitrogen glove box.

4.3.4 Electrochemical Measurements

Cyclic voltammetry measurements were carried out with an Electrochemical Workstation using a three-electrode cell assembly (platinum working electrode) at room temperature. The dichloromethane solutions of the samples containing 0.1 M of the recrystallized supporting electrolyte tetrabutyl-ammonium hexafluorophosphate were used. All potentials are referred against Ag/AgNO3 as the reference electrode, which was calibrated against the

ferrocene/ferrocenium (Fc) redox system. The Fc couple potential was determined to be +0.161

vs Ag/Ag+. The energy level of the ferrocene/ferrocenium (Fc) redox system is estimated to be

4.8 eV below the vacuum level, determined from -4.6 eV for the standard electrode potential (E)

of normal hydrogen electrode (NHE) on the zero vacuum level scale and a value of 0.2 V for Fc

67 vs NHE.

4.3.5 Ultrafast Spectrometry

The system for the ultrafast transient absorption spectrometric experiments consisted of a

Ti:sapphire laser (Spectra-Physics, Hurricane), the output of which was typically 1 mJ/pulse

(pulse width ~ 130 fs) at a repetition rate of 1 kHz. The hurricane output was 800 nm. An optical parametric amplifier (OPA-800C, Spectra-Physics) was used to obtain the 340 nm excitation wavelength. A total of 92% of the fundamental laser output was used to generate the required excitation wavelength whereas 8% of the output was used for white light generation. A 3 mm thick sapphire plate (Crystal Systems, Inc., HEML UX grade) was used for continuum generation.

Prior to generating the probe continuum, the amplified fundamental was passed to a delay line (Newport) that provided an experimental time window of 1.4 ns. The energy of the probe pulses was <1 μJ/cm2 at the sample. The pump beam was typically arranged to be 5 μJ/pulse with spot size of 1-2 mm diameter at the sample. The angle between pump and probe beam was

5-7°. The sample cell had an optical path of 2 mm. Both beams were coupled into 200 μm fiberoptic cables after the sample cell and thereafter input into a CCD spectrograph (Ocean

Optics, S2000 UV-vis) for time resolved spectral information (425-800 nm). Typically, 5000 excitation pulses were averaged to obtain the transient spectrum at a particular delay time. The

CCD spectrograph, the delay line, and the shutters were driven by a computer-controlled system.

In-house LabView (National Instruments) software allowed automatic spectral acquisition over a series of delay line settings. Kinetic traces at appropriate wavelengths were assembled from the accumulated spectral data. Sample solutions were prepared to have a

68 absorption of 0.8-1.0 at the excitation wavelength in the 2 mm cell and were used without deaeration. All measurements were carried out at room temperature, 22±2 °C.

4.3.6 Geometry Optimization

The Gaussian 03 program package was used for DFT calculations to optimize the geometries and calculate the HOMO and LUMO of 4.1.

4.4 Conclusions

7,16-Dihydroheptacenes containing phenyl and tert-butylphenyl groups at the 6,8,15, and 17 positions were studied as emitting materials for OLEDs. Each emits in the region of 420 – 428 nm in several solvents (ΦF = 0.15 – 0.21 in CH2Cl2) and in the solid state (ΦF = 0.37 – 0.44).

These compounds have good solubility in common organic solvents, are reasonably stable, and

retain color purity even after annealing for 24 hours at 110 oC. Though their dilute solutions

showed blue emission (λmax ~ 420 nm), but at higher concentration they showed excimer

emission (λmax ~ 480 and 510 nm). Interestingly, the OLED devices containing 4.1 showed green

emission (λmax ~ 515 and 550 nm) that is even further red shifted than the emission of excimer.

This indicates that an inter-ion pair, electromer, is responsible for the electroluminescence.

Pump-probe experiments of 4.1-4.3 revealed that the S1 state shows a broad absorption (~ 500-

650 nm) in dichloromethane with a lifetime of ~ 0.23–0.33 ns.

4.5 References

* Partially adopted from: (i) Mondal, R.; Shah, B. K.; Neckers, D. C. J. Org. Chem. 2006, 71, 4085; (ii) Mondal, R.; Wex, B.; Shah, B. K.; Kaafarani, B. R.; Danilov, E. O.; Jabbour, G. E.; Neckers, D. C. Org. Elect. 2007, Submitted for publication.

1. (a) Shinar, J. Organic Light-Emitting Devices, Springer-Verlag, New York, 2003; (b) Müllen, K.; Scherf, U. Organic Light-Emitting Devices. Synthesis, Properties and Applications, WILEY-VCH Verlag GmbH

& Co. KGaA, Weinheim, 2006; (c) Mitschke, U.; Bäuerle, P. J. Mater. Chem. 2002, 10, 1471; (d) Friend, R. H.; Gymer, R. W.; Holmes, A. B.; Burroughes, J. H.; Marks, R. N.; Taliani, C.; Bradley, D. D. C.; Dos Santos, D. A.; Brédas, J. L.; Lögdlund, M.; Salaneck, W. R. Nature 1999, 397, 121; (e) Segura, J. L.; Martin, N. J. Mater. Chem. 2000, 10, 2403; (f) Li, Z. H.; Wong, M. S.; Tao, Y.; Lu, J. Chem. Eur. J. 2005, 11, 3285; (g) Zheng, S.; Shi, J. Chem. Mater. 2001, 13, 4405. 2. Service, R. F. Science, 2005, 310, 1762. 3. Kim, D. Y.; Cho, H. M.; Kim, C. Y. Prog. Polym. Sci. 2000, 25, 1089. 4. Shi, J.; Tang, C. W. Appl. Phys. Lett. 2002, 80, 3201. 5. Mondal, R.; Shah, B. K.; Neckers, D.C. J. Org. Chem. 2006, 71, 4085. 6. Williams, E. L.; Haavisto, K.; Li, J.; Jabbour, G. E. Adv. Mater. 2007, 19, 197. 7. Wang, J. -F.; Kawabe, Y.; Shaheen, S. E.; Morrell, M. M.; Jabbour, G. E.; Lee, P. I.; Anderson, J.; Armstrong, N. R.; Kippelen, B.; Mash, E.; Peyghambarian, N. Adv. Mater. 1998, 10, 230. 8. Sciano, J. C. Handbook of Organic Photochemistry; CRC Press, Inc.: Boca Raton, Florida, 1989, vol. 1, p231. 9. Shah, B. K.; Neckers, D. C.; Shi, J.; Forsythe, E. W.; Morton, D. J. Phys. Chem. A 2005, 109, 7677. 10. Turro, N. J. Modern Molecular Photochemistry; University Science Books; California, 1991; p186-187. 11. Lai, K. -Y.; Chu, T. -M.; Hong, F. C. -N.; Elangovan, A.; Kao, K. -M.; Yang, S. –W.; Ho, T. -I. Surf. Coat. Technol. 2006, 200, 3283. 12. de Mello, J. C.; Wittmann, H. F.; Friend, R. H. Adv. Mater. 1997, 9, 230. 13. Shah, B. K.; Neckers, D. C.; Shi, J.; Forsythe, E. W.; Morton, D. Chem. Mater. 2006, 18, 603. 14. Schwob, H. P.; Williams, D. F. J. Chem. Phys. 1973, 58, 1542. 15. List, E. J. W.; Guentner, R.; de Freitas, P. S.; Scherf, U. Adv. Mater. 2002, 14, 374. 16. Chan, K. L.; McKiernan, M. J.; Towns, C. R.; Holmes, A. B. J. Am. Chem. Soc. 2005, 127, 7662. 17. Doménech, A.; Casades, I.; Garcia, H. J. Org. Chem.1999, 64, 3731. 18. Visco, R. E.; Chandross, E. A. J. Am. Chem. Soc. 1964, 86, 5350. 19. Pommerehne, J.; Vestweber, H.; Guss, W.; Mahrt, R.F.; Bässler, H.; Porsch, M.; Daub, J. Adv. Mater. 1995, 7, 551. 20. Thomas, K. R. J.; Lin, J. T.; Tao, Y. -T.; Ko, C. -W. J. Am. Chem. Soc. 2001, 123, 9404. 21. Kobayashi, T.; Kobayashi, S. Eur. J. Org. Chem. 2002, 2066. 22. Luszczynska, B.; Dobruchowska, E.; Glowacki, I.; Ulanski, J.; Jaiser, F.; Yang, X.; Neher, D.; Danel, A. J. Appl. Phys. 2006, 99, 024505. 23. Kalinowski, J.; Giro, G.; Cocchi, M.; Fattori, V.; Di Marco, P. Appl. Phys. Lett. 2000, 76, 2352. 24. Giro, G.; Cocchi, M.; Fattori, V.; Gardret, G.; Ruani, G.; Murgia, M.; Cavallini, M.; Biscarini, F.; Zamboni, R.; Loontjens, T.; Thies, J.; Leigh, D. A.; Morales, A. F.; Mahrt, R. F. Synth. Metals 2001, 122, 27. 25. Giro, G.; Cocchi, M.; Fattori, V.; Gardret, G.; Ruani, G.; Cavallini, M.; Biscarini, F.; Zamboni, R.; Loontjens, T.; Thies, J.; Leigh, D. A.; Morales, A. F.; Mahrt, R. F. Synth. Metals 2001, 122, 63. 26. Sepiol, J. J. Lumin. 1986, 36, 115.

27. Kalinowski, J.; Giro, G.; Cocchi, M.; Fattori, V.; Zamboni, R. Chem. Phys. 2002, 277, 387. 28. Anders, J.; Byrne, H. J.; Poplawski, J.; Roth, S.; Björholm, T.; Sommer-Larsen, P.; Schaumburg, K. Synth. Metals 1993, 55-57, 4820.

CHAPTER 5. PHOTOCHEMICAL SYNTHESIS OF HEXACENE AND

HEPTACENE*

5.1 Introduction

Poly(acene)s consist of an aromatic linear array. Lower molecular weight poly(acene)s (up to

pentacene) are usually synthesized by reduction of the corresponding quinones. A similar

approach used for higher poly(acene)s proved to be unsuccessful since the quinones were over-

reduced to hydrogenated acenes.1 Other methods of synthesis also failed for higher poly(acene)s, presumably because of their higher reactivity towards Diels-Alder additions which ultimately result in formation of either dimers or oxygen adducts (endoperoxides).2

The synthesis of heptacene has been challenging for more than 50 years!3 We used

Strating-Zwanenberg photodecarbonylation to synthesize both hexacene (5.1) and heptacene

(5.2), the latter for the first time. This decarbonylation, first discovered in 1969, was employed by the Strating group while seeking the dimer of carbon monoxide.4 An impressive attribute of

this reaction is that it is clean producing a poly(acene) following the expulsion of carbon

monoxide – either in a molecular form or a dimeric form- from the precursor α-diketones.

Pentacene was also synthesized using the Strating-Zwanenberg photochemical route.5

A semirigid polymer matrix was employed to reduce the reactivity of target poly(acene)s.

In this chapter, synthesis of hexacene and heptacene from their photoprecursors will be discussed. Heptacene was also produced in inert gas matrices, and its UV-vis-NIR absorption, as well as, IR spectra recorded. The thermal stability of heptacene in argon and nitrogen was evaluated and is discussed.

5.2 Results and Discussion

5.2.1 Photochemical Synthesis of Hexacene

The dione precursor of 5.1, 6,15-dihydro-6,15-ethanohexacene-17,18-dione (5.3), was prepared in 10-27% overall yield depending on the starting materials, either 1,2-dibromobenzene (5.10) or

2,3-dibromonaphthalene (5.11) and bicyclo[2,2,2]oct-2,3,5,6,7-pentaene (5.4; Scheme 5.1).

Compound 5.7 was synthesized following two consecutive cycloaddition reactions. Treatment of one equiv of 5.10 with n-BuLi at -60 °C (benzyne), in the presence of 5.4 produces 5.5 (70%), which was further reacted with one equivalent of 5.7 in presence of n-BuLi (naphthyne) to yield

80% of 5.7 (overall 56%). Similarly, 5.4 can be reacted first with 5.11 and then with 5.10 to produce 5.7 via compound 5.6 (60%), but with overall low yield of 5.7 (21%). Aromatization by refluxing 5.7 with chloranil for 2 h gives 5.8 (92%), which was converted to 5.9 (92%). Modified

Swern oxidation of 5.9 produced 5.3 (57%).

Scheme 5.1. Synthesis of photoprecursor of hexacene (5.3)a

(b) (a)

5.5 (b) 5.7 5.4 (a)

O HO (e) (d)

5.3 5.9 5.8

a Reagents and conditions: (a) 1,2-Dibromobenzene, n-BuLi, toluene, -60 °C, 3 h; (b) 2,3- Dibromonaphthalene, n-BuLi, toluene, -60 °C, 3 h; (c) chloranil, toluene, reflux, 2 h; (c) OsO4, NMO, acetone, r.t., 48 h; (d) dimethyl sulfoxide, trifluoroacetic anhydride, CH2Cl2, -78 °C, 3 h. using a catalytic amount of OsO4 in the presence of 4-methylmorpholin-N-oxide (NMO).

Scheme 5.2. Photochemical synthesis of hexacene (5.1)

395 nm 5.3

5.1

The Strating-Zwanenburg photodecarbonylation of 5.3 was carried out both in solution

and in a polymer matrix (Scheme 5.2). A solution of 5.3 in toluene showed an n-π* band at 465

nm (ε = 850), with an additional π-π* transitions at 327 (5000), 345 (5500), and 367 (4200) nm

(Figure 5.1). This is similar to the absorption of 5.3 in the

Figure 5.1. Normalized Absorption (red) , fluorescence (blue), and phosphorescene (green) spectra of 5.3 (absorption and fluorescence spectra recorded in toluene at room temperature and phosphorescence spectrum recorded in methanol/ethanol (1:4) matrix at 77 K).

poly(methyl methacrylate), PMMA, matrix. A weak fluorescence (λem - 525 nm, ΦF = 0.003)

was observed when the solution of 5.3 in toluene was excited at 460 nm. It also showed

phosphorescence (λem - 570 nm) at 77 K in a frozen matrix of methanol/ethanol (1:4). An array

74 of UV-LEDs (395 ± 25 nm) was used to selectively excite the n-π* transition and to prevent any

subsequent photochemical side reactions.

A solution of 5.3 in toluene (4.0 x 10-4 M) was degassed using freeze-pump-thaw.

Photolysis of this solution produced initially a new structured absorption band in the 550 – 700

nm region with maxima at 574, 612, 672 nm (Figure 5.2). The new structured bands can be

assigned to the π-π* transition of 5.1 in good agreement with the absorption spectrum of 5.1 recorded in 1,2,4-trichlorobenzene.6 However, the growth of this absorption continued only for a

few minutes, after which its intensity diminished with subsequent irradiation. MALDI-MS

analysis of the over irradiated (40 minutes) solution indicated the formation of dimer of 5.1 as

well as oxygen adducts. The latter formed through the reaction of 5.1 with residual oxygen in

solution. Thus, after 5.1 is generated in solution, it rapidly undergoes dimerization and oxidation

indicating it to be extremely reactive in solution. Compound 5.1 undergoes dimerization even

when it is generated at a very low concentration (< 10-4 M). These results are surprising given

that 5.1 was reported to have been sold commercially in the past!

Scheme 5.3. Oxidation of 5.1

O O O2 5.1

and other endoperoxides

Figure 5.2. Absorption spectra recorded during and after irradiation of 5.3 in degassed toluene (inset: enlarged portion from 525 to 725 nm).

When similar experiments were carried out with solutions saturated with oxygen,

immediate formation of endoperoxides of 5.1 was observed (Scheme 5.3). This was indicated by

MALDI-MS as well as by 1H-NMR spectra.

These results imply that when 5.1 is formed in solution, it immediately reacts with

oxygen.7 MALDI-MS analysis indicated the addition of one as well as two molecules of oxygen.

The protons of the bridge C atoms of 5.3 appear at 5.36 ppm in the 1H-NMR spectrum. Two

additional peaks appeared between 6 and 6.1 ppm in the 1H-NMR of an irradiated solution of 5.3 in CDCl3 purged with oxygen. The intensity of these peaks increased with the time of irradiation

(Figure 5.3). These are assigned to the protons at the carbon atoms attached to the oxygen bridge

in the endoperoxides. The many peaks in the 6-6.1 region also indicate that oxygen bridges form

at different positions of 5.1 (for example, the 6-15 and/or 8-13 positions).

1 Figure 5.3. H-NMR spectra recorded at different times of irradiation of a CDCl3 solution of 5.3 purged with oxygen (light source: a 395 nm UV-LED array).

The photodecarbonylation of 5.3 was carried out in a poly(methyl methacrylate)

[PMMA] (Scheme 5.3). A semi-rigid polymer matrix enables retention of reactive 5.1 by preventing its dimerization. Polymer matrix isolation also enables minimal contact of 5.1 with environmental oxygen reducing the rate of oxidation. Thus, when a PMMA film (thickness ~ 0.5 mm) containing 5.3 (3.5 x 10-3 M) was irradiated using the UV-LED array, this produced 5.1 as the film turned green and showed structured absorption in the region of 550 -700 nm (565, 614, and 672 nm) similar to that observed in solution (Figure 5.4). Quick dissolution of the irradiated

77 film in dichloromethane and subsequent analysis by MALDI mass spectra confirmed formation of 5.1.

Figure 5.4. Absorption spectra recorded during and after irradiation of 5.3 in a PMMA film.

The intensity of the 672 nm absorption band of 5.1 gradually increased through the first

60 min of irradiation. But further irradiation resulted in no additional growth. In fact, the

intensity of the 672 nm band started decreasing, irrespective of irradiation, indicating slow

decomposition of 5.1 even in PMMA matrix. Compound 5.1 could be retained in the matrix over

night under ambient conditions. These results attest to the unstable nature of 5.1 even under

conditions manipulated so as to maintain its isolation; that is, in the solid PMMA matrix.

Compound 5.1 seems to be consumed by oxidation with a small amount of molecular oxygen

that diffuses into the matrix.

5.2.2 Photochemical Synthesis of Heptacene

The dione precursor of 5.2, 7,16-dihydro-7,16-ethanoheptacene-19,20-dione (5.12), was prepared in 18% overall yield from 2,3-dibromonaphthalene (5.11) and bicyclo[2,2,2]oct-2,3,5,6,7-

pentaene (5.4) (Scheme 5.4). Treatment of two equivalents of 5.11 with n-BuLi at -60 oC

{naphthyne}, in the presence of 5.4 produces 5.13 (53%). Aromatization by refluxing 5.13 with

chloranil for 2 hrs gives 5.14 (81%), which was converted to 5.15 (83%), using a catalytic

amount of OsO4 in the presence of 4-methylmorpholin-N-oxide (NMO). Modified Swern

oxidation of 5.15 produced 5.12 (51%). As described previously with hexacene, the Strating-

Zwanenburg photodecarbonylation can be conveniently used to generate the target heptacene 5.2

in the PMMA matrix from 5.12 by

Scheme 5.4. Synthesis of heptacene (5.2)a

Br + (a) Br 5.4 5.11 5.13

(b) OH

HO (c)

5.15 5.14

(d) O

O (e)

5.12 5.2

a o Reagents and conditions: (a) n-BuLi, toluene, -60 C, 3 h; (b) chloranil, toluene, 2 h; (c) OsO4, o NMO, acetone, r. t., 48 h; (d) dimethylsulfoxide, trifluoroacetic anhydride, CH2Cl2, -78 C, 3 h; (e) hυ (395 nm), PMMA matrix.

79 irradiating with the array of UV-LEDs light source (395 nm ± 25 nm) to prevent any subsequent photochemical side reactions.

A solution of 5.12 in toluene showed an n-π* band at 466 nm (ε = 1430), with an additional π-π* transition at 368 nm (ε = 6640) and 347 nm (ε = 8850). This is similar to the observations in the PMMA matrix. A thin PMMA film containing 5.12 (~3.0×10-3 ML-1) was prepared (thickness ~0.5 mm, o.d. = 0.5 at 395 nm). When irradiated using the UV-LED array, a new structured absorption band extending from 600 to 825 nm and having a maximum at ~760 nm appeared in the film. This is assigned to the π-π* transition of 5.2 (Figure 5.5). The absorption of 5.2 recorded in the solid matrix is blue shifted relative to that of 7,16-

8 bis(tris(trimethylsilyl)silylethynyl)heptacene (λmax = 810 nm) recorded in hexane. Quick

dissolution of the irradiated film in CH2Cl2 and subsequent analysis by matrix assisted laser

desorption ionization mass spectra (MALDI-MS) confirmed formation of 5.2.

Figure 5.5. Absorption spectra recorded during and after irradiation of 5.12 in a PMMA film, (inset – enlarged portion from 600 to 850 nm).

The intensity of the 760 nm absorption band of 5.2 gradually increased through the first

60 min of irradiation. But further irradiation achieved no additional enhancement. In fact, the intensity of the 760 nm band started decreasing after 60 min irrespective of irradiation, and the band completely vanished in about 4 hours. This attests to the unstable nature of 5.2 even under conditions manipulated so as to maintain its isolation; i.e. in the solid PMMA matrix. Changes in the intensities of the bands at 420 and 450 nm are concomitant with that of the 760 nm band and assigned to 5.2 (Figure 5.5). The weak bands in the 500-600 nm region kept increasing with time and are likely due to pentacene derivatives that result when 5.2 dimerizes or reacts with oxygen at the 5 and 18 positions. The possibility of formation of similar tetracene and anthracene derivatives cannot be ruled out. Nevertheless, irradiation of an oxygen saturated toluene solution of 5.12 produced oxygen adducts of 5.2 as observed by the MALDI-MS analyses. This indicates that similar to 5.1, as soon as 5.2 forms in the solution, it is trapped by oxygen due to its extremely high reactivity as a Diels-Alder diene.9

Formation of oxygen adducts of 5.2 in solution was further verified by 1H-NMR. The protons of the bridged C-atoms of 5.12 appear at 5.4 ppm. Three additional peaks appeared

1 between 6.0-6.3 ppm in the H-NMR of an irradiated solution of 5.12 in CDCl3 purged with

oxygen due to the formation of endoperoxides (details in Chapter 7).

The S0-S1 energy gap of 5.2 (1.50 eV) as assessed from the edge (825 nm) of the

absorption band correlates well with the calculated values (Table 5.1).10-12 In fact, it is closer to

the theoretical value calculated for an open shell (OS) ground state. The S0-S1 energy gap of 5.2

also compares well with those for pentacene and 5.1 obtained experimentally.10-12 For example,

there is a difference of 0.35 eV between the S0-S1 energy gaps of pentacene and hexacene.

Similarly, the difference between the S0-S1 energy gaps of hexacene and 5.2 is 0.29 eV.

Table 5.1. Theoretical and experimental S0-S1 values of acenes

† Experimental TDDFT/6-31G* S0(OS)-S1

Compound S0-S1 S0-S1

nm eV eV eV

Pentacene 582 2.14 1.95 -

Hexacene 695 1.79 1.54 1.57

Heptacene 825 1.50 1.24 1.36

† S0(OS)-S1 values were calculated considering an open shell (OS) ground state using the stabilization energy from ref. 12.

5.2.3 Photochemical Synthesis of Anthracene

9,10-Dihydro-9,10-ethanoanthracene-11,12-dione13 (5.16), the Strating-Zwanenburg photo-

precursor of anthracene (similar to 5.3 and 5.12), was synthesized according to Scheme 5.5 and

studied for comparison. A solution of 5.16 in toluene showed an n-π* transition at 461 nm (ε =

310). Photolysis of 5.16 in toluene produced anthracene (5.17) with a quantum efficiency of 0.02

(ferrioxalate actinometry, details in Chapter 6). Photolysis of 5.16 in the PMMA film was

equally facile producing 5.17 (Scheme 5.6).

Scheme 5.5. Synthesis of phtotprecursor of anthracene (5.16)a

O O OH O O HO O O (a) (b) (c) + O O 5.19 5.16 5.17 5.18 a Reagents and conditions: (a) vinylene carbonate, toluene, 180 oC, 3 days, 90%; (b) NaOH, o Dioxane, reflux, 2 h, 100%; (c) DMSO, trifluooacetic anydride, CH2Cl2, -78 C, 3 h, 65%.

The only difference between 5.16 and 5.3/5.12 is that the former produced the corresponding acene efficiently when irradiated in solution. The later precursors also produce the corresponding acene, but they undergo further reaction, such as dimerization or oxidation due to their reactive in nature.

Scheme 5.6. Synthesis of phtotprecursor of anthracene (5.17)

O 395 nm

5.16 5.17

5.2.4 Photogeneration and Thermal Stability of Heptacene in Inert Gas Matrices

Isolation of the α-diketone precursor 5.12 in inert gas matrices is achieved by heating a sample

to approximately 225° at a pressure of approximately 10–5 mbar. The IR spectrum of 5.12

obtained differs somewhat from that obtained in KBr in the υ(C=O) stretching region where a

number of bands are observed between 1700 – 1400 cm-1. These bands are due to matrix site

effects and disappear with different rates upon irradiation with a high-pressure mercury lamp

(385 < λ < 450 nm). Under these irradiation conditions all the bands of 5.12 decrease, and

prolonged photolysis results in complete disappearance of 5.12. Concomitantly, a set of new

bands at 1309, 901, 734 and 602 cm–1 as well as a broad signal in the CO stretching region (2138 cm–1, 2137 cm–1, and 2133 cm–1 due to monomeric CO and aggregated CO) are growing in

(Figure 5.6).14 Bands with the arrows pointing downwards are decreasing; whereas those with the arrows pointing upwards are growing during the irradiation. The reaction is clean and the only organic photoproduct formed in this reaction is assigned to heptacene 5.2 based on the

83 comparison of the measured IR spectrum with computational data obtained at the UB3LYP/6-

31G* level.

Figure 5.6. (A) Difference spectrum in the range of 3300 – 400 cm–1 obtained after irradiation (high pressure mercury lamp, 350-450 nm) at 15 K containing 5.12 (sublimation at 220–225 °C). (B) IR spectrum computed for 5.12 at the B3LYP/6-31G* level of theory. (C) IR spectrum computed for 5.2 at the UB3LYP/6-31G* level of theory.

Following the photochemical decomposition of 5.12 by UV/vis spectroscopy in various

matrix host gases (N2, Ar, Xe) at 10 K is particularly revealing. Upon irradiation (385 < λ < 450

nm in Ar at 10 K), the absorption bands of 5.12 are decreasing while those of 5.2 are increasing

continuously with six isosbestic points at 225 nm, 295 nm, 359 nm, 408 nm, 450 nm, and 504

nm (Figure 5.7). The presence of isobestic points proofs that the photodecarbonylation proceeds

without any detectable intermediate in frozen inert gases at 10 K. The electronic absorption

spectrum of 5.2 is in good agreement with computational analysis at the RICC2/TZVP level of

84 theory.15 The electronic absorption spectrum of 5.2 is characterized by the very intense β-band

1 1 (2 B2u ← X A1g) with its most intense feature at 326 nm and its origin at 338 nm. The weak p-

1 1 band (1 B1u ← X A1g) is extending from 769 to 559 nm and is characterized by three groups of

signals having their strongest absorptions at 728 nm, 656 nm and 601 nm. The p-system of 5.2 differs from all lower acenes in so far as strongest band in the p-system is not the 0-0 transition at 769 nm, but a vibrationally excited transition at 728 nm.15 The weak feature at 769 nm is also

observed after annealing the matrix from 10 K to 30 K. The β-band was not observed in the

PMMA matrix.

Figure 5.7. UV-vis spectra obtained after deposition of 5.12 (dark blue trace) and subsequent irradiation after successively doubled time intervals. Arrows pointing down indicate bands which decrease during irradiation, while upward pointing arrows indicate bands with increase during irradiation. Isosbestic points are marked with small-headed arrows. Inset: image showing photogenerated 5.12 (green) in Ar matrix at 10 K.

The thermal reactivity of 5.2 was investigated by carefully evaporating the matrix host gas (Ar and N2) and monitoring the changes in the UV-vis spectrum. In the Ar matrix, the peaks

of 5.2 started deforming at 50 K and completely vanished at 53 K within 20 min (Figure 5.8).

Annealing the N2 matrix to 34 K resulted in slow sublimation of the host, as evidenced by the increased pressure to 10–4 mbar in the matrix isolation apparatus. During evaporation, the

spectral features of the heptacene decrease, and have disappeared by the time the host gas is

completely evaporated as indicated by a pressure drop back to 10–6 mbar (Figure 5.9). These

experiments indicate that the life time of heptacene is very short at 34 K, if it is not protected

from dimerization or oligomerization reactions by the matrix.

Figure 5.8. Annealing of a heptacene (black) containing argon matrix. Here the cryostat is switched off at t = 0, and the spectra are recorded in 2 min intervals during warm up. Until t = 16 min no significant change is observed (spectra omitted for clarity), at t = 18 min (T = 50 K), the base line shifts and at t = 20 min (T = 53 K), the heptacene absorptions have already disappeared.

Figure 5.9. Monitoring of the spectral changes of the p- and α-bands of heptacene during slow evaporation of a nitrogen matrix at 34 K. Top: Vis spectrum at the beginning of the annealing. Bottom: Vis spectrum after 30 min annealing to 34 K. The matrix host is sublimed away after this time. Middle: Spectra measured at 1.5 min intervals between t = 0 and t = 30 demonstrate the slow disappearance of heptacene.

5.3 Experimental Section

Solvents and reagents were used as received. Standard grade silica gel (60 Å, 32-63 μm) and silica gel plates (200 μm) were purchased from Sorbent Technologies. Reactions that required anhydrous conditions were carried out under argon in oven-dried glassware. Organic solvents were either spectroscopic grade or purified by distillation and dried before use using proper drying reagents.

Mass spectra were recorded on Shimadzu GCMS-QP5050A instrument equipped with a direct probe (ionization 70 eV). Matrix assisted laser desorption ionization (MALDI) spectra were obtained using Bruker Daltonic Omniflex® instrument (N2 laser, 337 nm). A Bruker spectrometer (working frequency 300.0 MHz for 1H) was used to record NMR spectra.

Absorption and fluorescence spectra were recorded on a Shimadzu UV-2401 spectrophotometer

87 and a Fluorolog®-3 spectrometer, respectively. All measurements were carried out at room temperature unless otherwise specified.

5.3.1 Synthesis of Hexacene Precursor

Bicyclo[2,2,2]oct-2,3,5,6,7-pentaene (5.4). Compound 5.4 was prepared following a literature

16 o 1 procedure. Mp 97 – 98 C; H NMR (300 MHz, CDCl3) 3.851 (m, 2H), 4.9 (s, 4H), 5.25 (s,

13 4H), 6.36 (m, 2H); C NMR (300 MHz, CDCl3) 53.15, 104.13, 132.34, 144.27; mass spectrum

(GC-MS) m/z M+ 104(80), 115(70), 141(100), 156(96).

5,6,9,10-Tetrahydro-6,9-dietheno-7,8-dimethyleneanthracene (5.5). n-BuLi (2.5 M in

hexane) (0.68 mL, 1.7 mmol) diluted in 15 mL of dry hexane was added drop-wise under an

argon atmosphere to a suspension of 5.4 (0.265 g, 1.7 mmol) and 1,2-dibromobenzene (5.10;

0.398 g, 1.7 mmol) in 20 mL of dry toluene cooled to -50 – -60oC. The mixture was allowed to

stir for three hours at -50 – -60oC. Then the temperature was allowed to slowly increase to r. t.

and the excess of n-BuLi quenched by methanol. The filtrate was concentrated on a rotary

evaporator and subjected to column chromatography [silica gel, 10% (vol.) DCM in hexane] to

1 yield 5.5 (0.275 g, yield 70%). H NMR (300 MHz, CDCl3) 3.545 (s, 4H), 3.96 (m, 2H), 4.859

13 (s, 2 H), 5.076 (s, 2 H), 6.490 (m, 2 H) 7.1 (s, 4 H); C NMR (300 MHz, CDCl3) 31.63, 52.34,

101.54, 125.93, 128.74, 133.69, 133.88, 133.95, 144.02; mass spectrum (GC-MS) m/z M+

104(70), 178(60), 202(50), 217(80), 232(100).

6,7,10,11-Tetrahydro-7,10-dietheno-8,9-dimethylenetetracene (5.6). n-BuLi (2.5 M in

hexane) (3.6 mL, 9 mmol) was added drop-wise under argon to a suspension of 2,3-

dibromonaphthalene (5.11; 1.79 g, 6.25 mmol) and 5.4 (0.48 g, 3.07 mmol) in 125 mL of dry

88 toluene cooled to -50 – -60oC. The mixture was allowed to stir for three hours at -50 – -60oC,

then the temperature was allowed to very slowly increase to r. t. and the excess of n-BuLi

quenched by methanol. The product was concentrated on a rotary evaporator. GC/MS analysis of

the product showed it to be a mixture of mono-adduct 5.6 and bi-adduct (6,7,8,15,16,17-

hexahydro-7,16-diethenoheptacene). The latter was separated from the former by column

chromatography (silica gel). Elution was started with 2% (vol.) of DCM in hexane and ended

with 50% (vol.) of DCM in hexane. Compound 5.6 eluted first, followed by elution of the bi-

adduct. The separated solvent fractions were evaporated to isolate 5.6 (0.30 g, yield 35%) and bi-

o 1 adduct (0.67 g, yield 53%). Mp 206 - 207 C; H NMR (300 MHz, CDCl3) 3.75 (s, 4H), 4.014

(m, 2H), 4.89 (s, 2 H), 5.1 (s, 2 H), 6.52 (m, 2 H) 7.36 (m, 2 H), 7.61 (s, 2 H), 7.71 (m, 2 H); 13C

NMR (300 MHz, CDCl3) 31.70, 52.47, 101.66, 125.15, 126.79, 127.06, 132.22, 132.87, 133.77,

134.03, 144.05; mass spectrum (DIP-MS) m/z M+ 133(60), 178(85), 228(65), 252(50), 267(75),

282(100); HRMS (EI) m/z 282.1409 (M+), calcd m/z 282.1409.

5,6,7,14,15,16-Hexahydro-6,15-diethenohexacene (5.7). Compound 5.7 was synthesized

following a similar procedure to that described above for the synthesis of 5.5 or 5.6. A

stoichiometric mixture of 5.5 and 5.11 or 5.6 and 5.10 in presence of n-BuLi (1 eq.) produced

o 1 5.7, with 80% and 60% yield, respectively. Mp 223-225 C; H NMR (300 MHz, CDCl3) 3.6 (s,

4 H), 3.8 (s, 4 H), 4.35 (m, 2 H), 6.85 (m, 2 H), 7.1 (s, 4 H), 7.35 (dd, 2 H), 7.6 (s, 2 H), 7.75 (dd,

13 2 H); C NMR (300 MHz, CDCl3) 33, 54.36, 125.06, 125.91, 126.71, 127.07, 128.76, 132.25,

133.31, 134.26, 139.37, 140.31; mass spectrum (DIP-MS) m/z M+ 178 (80), 191 (50), 202 (20),

215 (45), 228 (40), 253 (30), 315 (15), 328 (20), 341 (10), 358 (100); HRMS (EI) m/z 358.1730

(M+), calcd m/z 358.1722.

6,15-Dihydro-6,15-diethenohexacene (5.8). Compound 5.7 (0.151 g, 042 mmol) and chloranil (0.227 g, 0.924 mmol) were taken into a round bottom flask containing 20 mL of dry toluene and refluxed for 2 hrs. The reaction mixture was concentrated on a rotary evaporator and then dissolved in DCM and washed with 3M sodium hydroxide (aq.) solution. The organic layer was evaporated and subjected to column chromatography [silica gel, 20% (vol.) DCM in hexane]

o 1 to yield 5.8 (0.14 g, yield 92%). Mp > 300 C (decomp.); H NMR (300 MHz, CDCl3) 5.3 (m, 2

H), 7.05 (m, 2H), 7.40 (m, 4 H) 7.75 (m, 4 H), 7.8 (s, 2 H), 7.95 (m, 2 H), 8.25 (s, 2 H); 13C

NMR (300 MHz, CDCl3) 50, 120.83, 121.36, 124.91, 125.57, 127.47, 127.97, 130.50, 131.67,

131.85, 137.85, 141.21, 141.75; mass spectrum (DIP-MS) m/z M+ 176 (80), 212 (10), 248 (12),

328 (10), 354 (100); HRMS (EI) m/z 354.1412 (M+), calcd m/z 354.1409.

6,15-Dihydro-17,18-dihydroxy-6,15-ethanohexacene (5.9). To a solution of 4-

methylmorpholine N-oxide (0.14 g, 1 mmol) in 75 mL of acetone a 2.5 % (w) solution of

osmium tetroxide in t-butanol (0.25 mL) was added under an argon atmosphere and stirred for 10

min at r. t. Then, a suspension of 5.8 (0.14 g, 0.4 mmol) in 50 mL of acetone was added and the

mixture continued to stir for 48 hrs. Sodium dithionite (0.2 g) was added to the reaction mixture,

which was stirred for another 20 min to yield a heterogeneous solution. It was then filtered

through a pad of cerite, followed by washing with acetone. The filtrate was then concentrated to

about 20 mL, resulting in an off-white precipitate, which was filtered and subjected to column

chromatography (silica-gel, 40% acetone/hexane, v/v). After evaporating the solvent, pure 5.9

(0.14 g, yield 92%) was isolated as white solid. Mp > 300oC (decomp.); 1H NMR (300 MHz,

CDCl3) 4.30 (s, 2 H) 4.68 (s, 2 H), 7.43 (m, 4 H), 7.9 (m, 8 H), 8.36 (m, 2 H); mass spectrum

(DIP-MS) m/z M+ 127 (100), 141 (95), 169 (65), 185 (60), 185 (58), 228 (100), 328 (50); HRMS

(TOF-ES+) m/z 411.1355 (M+Na) +, calcd m/z 411.1361.

6,15-Dihydro-6,15-ethanohexacene-17,18-dione (5.10). Trifluoroacetic anhydride (2.1 mL) was added drop wise to a stirring solution of dry dimethyl sulfoxide (DMSO, 2.2 mL) in 25 mL of DCM kept at –78 oC under an argon atmosphere. After 15 min of stirring, a solution of 5.9

(0.15 g, 0.39 mmol) in 2.2 mL of DMSO in 20 mL of DCM was added very slowly. Stirring was

continued for 90 minutes of the resulting mixture was continued for 90 min, followed by drop-

wise addition of triethylamine (2.5 mL). After stirring for another 90 min at –78 oC under inert

atmosphere, the reaction mixture was allowed to warm to r. t. The resulting bright yellow

solution was then extracted with DCM, followed by washing the organic extract with water.

Further purification was achieved by column chromatography (silica gel, DCM) to obtain pure

1 5.10 (0.085 g, yield 57%) as bright yellow solid. H NMR (300 MHz, CDCl3) 5.36 (s, 2 H), 7.51

(dd, 4 H), 7.85 (dd, 2 H), 7.98 (m, 4 H), 8.1 (s, 2 H), 8.428 (s, 2 H); 13C NMR (300 MHz,

CDCl3) 60.93, 125.43, 125.49, 126.08, 126.56, 127.07, 127.91, 128.17, 131.20, 131.34, 131.94,

132.25, 133.71, 185.49; mass spectrum (EI) m/z M+ 108 (37), 124 (16), 149 (50), 164 (24), 170

(11), 263 (13), 328 (100), 329 (33), 384 (5); HRMS (EI) m/z 384.1150 (M+), calcd 384.1150.

5.3.2 Synthesis of Heptacene Precursor

2,3-Dibromonaphthalene (5.11). It was prepared following a literature procedure.17 Mp 136 oC;

1 13 H NMR (300 MHz, CDCl3) 7.51 (m, 2H), 7.72 (m, 2H), 8.15 (s, 2H); C NMR (300 MHz,

+ CDCl3) 121.96, 126.86, 127.18, 132.25, 133.06; mass spectrum (GC-MS) m/z M 126(95),

205(30), 20(30), 284(50), 286(100), 288(50).

6,7,8,15,16,17-Hexahydro-7,16-diethenoheptacene (5.13). n-BuLi (2.5 M in hexane)

(3.6 mL, 9 mmol) was added drop-wise under an argon atmosphere to a suspension of 5.11 (1.79 g, 6.25 mmol) and 5.4 (0.48 g, 3.07 mmol) in 125 mL of dry toluene cooled to -50 – -60 oC. The

91 mixture was allowed to stir for three hour at -50 – -60oC. Then the temperature was allowed to very slowly increase to r. t. and the excess of n-butyllithium was quenched by methanol. The product was concentrated on a rotary evaporator. GC/MS analysis of the product showed it to be a mixture of mono-adduct (5.6) and bi-adduct (5.13). The latter was separated from the former by column chromatography (silica gel). Elution was started with 2% (vol.) of dichloromethane in hexane and ended with 50% (vol.) of dichloromethane in hexane. The mono-adduct eluted first, which was followed by elution of the bi-adduct (5.13). The separated solvent fraction was

o 1 evaporated to isolate 5.13 (0.67 g, yield 53%). Mp 268 - 269 C; H NMR (300 MHz, CDCl3),

3.81 (s, 8 H), 4.411 (m, 2 H), 6.91 (m, 2 H), 7.35 (m, 4 H), 7.61 (s, 4 H), 7.71 (m, 4 H); 13C

NMR (300 MHz, CDCl3) 33.21, 54.47, 125.05, 126.77, 127.04, 132.24, 133.27, 139.44, 140.04;

mass spectrum (DIP-MS) m/z M+ 378(10), 389(10), 408(100); HRMS (EI) m/z 408.1879 (M+),

calcd m/z 408.1878.

7,16-Dihydro-7,16-diethenoheptacene (5.14). 5.13 (1.13 g, 2.77 mmol) and chloranil

(1.35 g, 5.49 mmol) were taken in a round bottom flask containing 110 mL of dry toluene and

refluxed for 2 hrs. The reaction mixture was kept for overnight at r. t., when a part of the product

(0.46 g) precipitated out. It was separated by filtration followed by washing with hexane. The

filtrate was concentrated on a rotary evaporator and subjected to column chromatography [silica

gel, 20% (vol.) dichloromethane in hexane] to yield 5.14 (0.45 g, yield 81%). Mp > 300oC

1 (decomp.); H NMR (300 MHz, CDCl3) 5.32 (m, 2 H), 7.04 (m, 2H), 7.40 (m, 4 H) 7.873 (s, 4

13 H), 7.927 (m, 4 H), 8.281 (s, 4 H); C NMR (300 MHz, CDCl3) 49.83, 120.97, 124.94, 125.62,

127.98, 130.59, 131.68, 137.51, 140.87; mass spectrum (DIP-MS) m/z M+ 201(80), 378(15),

404(100); HRMS (EI) m/z 404.1556 (M+), calcd m/z 404.1565.

7,16-Dihydro-19,20-dihydroxy-7,16-ethanoheptacene (5.15). To a solution of 4- methylmorpholine N-oxide (0.336 g, 2.5 mmol) in 200 mL of acetone, 2.5 % (w) solution of osmium tetroxide (0.5 mL, 0.05 mmol) in t-butanol was added under an argon atmosphere and stirred for 10 min at r. t. Then, a suspension of 5.14 (0.4 g, 1.0 mmol) in 100 mL of acetone was added and the mixture continued to stir for 48 hrs. Sodium dithionite (0.5 g) was added to the reaction mixture, which was stirred for another 20 min to yield a heterogeneous solution. It was then filtered through a pad celite, followed by washing with acetone. The filtrate was then concentrated to about 100 mL, which resulted in an off-white precipitate. The precipitate was filtered and subjected to column chromatography (silica-gel). Impurities were washed with 5-

50% (vol.) of chloroform in hexane and finally 5.15 was eluted with chloroform. After evaporating the solvent, pure 5.15 (0.35 g, yield 83%) was isolated as white solid. Mp > 300oC

1 (decomp.); H NMR (300 MHz, CDCl3) 2.33 (s, broad, 2 H, quenched with D2O), 4.33 (s, 2 H)

13 4.70 (s, 2 H), 7.45 (m, 4 H), 8.00 (m, 8 H), 8.38 (m, 4 H); C NMR (300 MHz, Me2SO-d6)

51.47, 67.88, 123.06, 124.42, 125.56, 125.66, 125.80, 126.03, 128.37, 131.10, 131.24, 131.36,

131.55, 137.64, 138.39; mass spectrum (DIP-MS) m/z M+ 189 (68), 378 (100), 438 (6); HRMS

(EI) m/z 438.1627 (M+), calcd m/z 438.1620.

7,16-Dihydro-7,16-ethanoheptacene-19,20-dione (5.12). Trifluoroacetic anhydride (4.2 mL) was added drop wise to a stirring solution of dry-dimethyl sulfoxide (DMSO, 2.4 mL) in 50 mL of dichloromethane kept at –78 oC under an argon atmosphere. After 15 min of stirring, a

solution of 5.15 (0.30 g, 0.68 mmol) in 2.4 mL of DMSO in 30 mL of dichloromethane was

added very slowly. The resulting mixture was continued to stir for 90 min, followed by drop-

wise addition of triethylamine (5 mL). After stirring for another 90 min at –78 oC under inert

atmosphere, the reaction mixture was allowed to warm to r. t. The resulting bright yellow

93 solution was then extracted with dichloromethane, followed by washing the organic extract with water. Further purification was achieved by column chromatography (silica gel, dichloromethane) to obtain pure 5.12 (0.155 g, yield 51%) as bright yellow solid. 1H NMR (300

13 MHz, CDCl3) 5.398 (s, 2 H), 7.492 (dd, 4 H), 8.00 (dd, 4 H), 8.117 (s, 4 H), 8.435 (s, 4 H); C

NMR (300 MHz, CDCl3) 61.14, 125.54, 126.08, 126.57, 128.17, 131.19, 131.37, 132.24, 185.80;

mass spectrum (DIP-MS) m/z M+ 189 (69), 190 (14), 378 (100), 379 (33), 434 (3); HRMS (EI)

m/z 434.1324 (M+), calcd 434.1307.

5.3.3 Synthesis of Anthracene Precursor

9,10-Dihydro-11,12-dihydroxy-9,10-ethanoanthracene (5.19). Anthracene (5.17; 1.05 g, 6

mmol) and vinylene carbonate (0.5 mL, 9.5 mmol) were taken in a 30 mL steel-bomb reactor

containing dry toluene (25 mL). The mixture was heated at 180 oC and stirred for 3 days. After

cooling to room temperature, the reactor was opened very carefully and the gel-type material

produced evaporated under reduced pressure in a rotary evaporator to obtain a white solid. The

solid was further purified by column chromatography (silica gel). The residual reactants were

washed with 20% (v) of dichloromethane in hexane. Finally, the product was eluted with

dichloromethane. Evaporation of the solvent yielded 5.18 (1.41 g, yield 90%). IR 1791 cm-1; 1H

NMR (300 MHz, CDCl3) 4.7 (d, J = 1.5 Hz, 2H), 4.88 (m, 2H), 7.25 (m, 5 H, CDCl3), 7.385 (m,

4 H); mass spectrum (GC-MS) m/z M+ 178(100), 264(2).

Compound 5.18 (2.0 g, 7.56 mmol) was taken in 1,4-dioxane (90 mL) in a round bottom flask, to which 30 mL of 4N NaOH (aq.) was added. The reaction mixture was refluxed for 2 hrs under an argon atmosphere. After cooling the reaction mixture to r. t., the product was extracted with chloroform. The organic extract was then dried over Na2SO4. Removal of the solvent

-1 1 yielded 5.19 (1.80 g, yield 100%). IR 3300-3500 cm (broad), H NMR (300 MHz, CDCl3) 2.091

(s, 2 H), 4.067 (s, 2 H), 4.423 (s, 2 H), 7.19 (m, 4 H), 7.35 (m, 4 H), mass spectrum (DIP-MS)

178(100), HRMS (EI) m/z 238.0994 (M+), calcd m/z 238.0994.

9,10-Dihydro-9,10-ethanoanthracene-11,12-dione (5.16). A synthetic procedure similar

to that used for 5.3 was employed to synthesize 5.16 from 5.18. The yield was 65%. 1H NMR

13 (300 MHz, CDCl3) 5.003 (s, 2 H), 7.38 (dd, 4H), 7.48 (dd, 4 H); C NMR (300 MHz, CDCl3)

60.02, 126.32, 129.40, 134.87, 183.71; mass spectrum (GC-MS) m/z M+ 89 (15), 178 (100), 234

(2); HRMS (EI) m/z 234.0683 (M+), calcd m/z 234.0681.

5.3.4 Sample Preparation

Irradiation in solution. A dilute solution of 5.16 (1.5 x 10-3 ML-1) in toluene was prepared in 4

mL quartz cuvette and irradiated at 395 nm using a UV-LED lamp while UV-vis spectra were

recorded every 5-10 min. Similarly, a dilute solution of 5.3 or 5.12 (~1.0 ×10-3 M. L-1) in toluene

was prepared and made air-free by the freeze-pump-thaw method. This solution was then used

for irradiation.

Preparation of PMMA films. A saturated solution of the appropriate photoprecursor

(5.3 or 5.12 or 5.16) in dichloromethane was mixed with a solution of PMMA in acetonitrile. A

few drops of the mixture was placed on a quartz disc and dried overnight under air. Films

showing absorbance 0.2 – 0.4 were used for irradiation. Attempts were made to avoid light as

much as possible while preparing the thin films,.

MALDI of hexacene and heptacene from the thin film. A thin PMMA film of 5.3 or

5.12 was irradiated for 30 min. A portion of the film was scratched from the quartz substrate and

95 dissolved in degassed dichloromethane. The MALDI mass spectra were immediately recorded using a dithranol matrix.

5.4 Conclusion

In summary, hexacene (5.1) and heptacene (5.2) were synthesized using Strating-Zwanenberg photodecarbonylation. Compound 5.1 undergoes dimerization in solution even when generated at very low concentrations (< 10-4 M). Generation of 5.1 or 5.2 in oxygen saturated solution

exclusively leads to the formation of endoperoxides. Irradiation of the diketone precursors in the

PMMA matrix at 395 nm produced 5.1 and 5.2, which showed the desired long wavelength

absorption. Compound 5.1 and 5.2 were found to be stable in the PMMA matrix up to 12 and 4

hours, respectively. Compound 5.2 was also generated in inert gas matrices at low temperature.

Uv-vis-NIR absorption and IR spectra of heptacene were recorded in argon matrix at 10 K.

Several isosbestic points indicate that this reaction proceeds without any detectable intermediates

even at temperatures as low as 10 K. When heptacene was generated in nitrogen matrix, it was

stable up to 34 K. However, it was stable up to ~50 K, when generated in argon matrix. It was

also shown that 5.2 is the first acene for which the vibrational progression of the p-band differs

from those of the well-studied smaller acenes.

5.5 References

* Partially adopted from: (i) Mondal, R.; Shah, B. K.; Neckers, D. C. J. Am. Chem. Soc. 2006, 128, 9612; (ii) Bettinger, H. F.; Mondal, R.; Neckers, D. C. Angew. Chem. Int. Ed. 2007, Manuscript submitted; (iii) Mondal, R.; Adhikari, R. M.; Shah, B. K.; Neckers, D. C. Org. Lett. 2007, 9, 2505.

1. (a) Clar, E. Polycyclic Hydrocarbons; Academic Press: London and New York, 1964; Vols. 1, 2. (b) Harvey R. G. Polycyclic Aromatic Hydrocarbons; Wiley-VCH: New York, 1997; (c) Bailey, W. J.; Liao,

C. -W. J. Am. Chem. Soc. 1955, 77, 992; (d) Satchell, M. P.; Stacey, B. E. J. Chem. Soc. C: Organic. 1971, 3, 468. 2. Fang, T. Heptacene, Octacene, Nonacene, Supercene and Related Polymers. Ph.D.Thesis, University of California, Los Angeles, CA, 1986. 3. Mondal. R.; Shah, B. K.; Neckers, D. C. J. Am. Chem. Soc. 2006, 128, 9612. 4. Strating, J.; Zwanenburg, B.; Wagenaar, A.; Udding, A. C. Tetrahedron Lett. 1969, 10, 125. 5. (a) Uno, H.; Yamashita, Y.; Kikuchi, M.; Watanabe, H.; Yamada, H.; Okujima, T.; Ogawa, T.; Ono, N. Tet. Lett. 2005, 46, 1981. (b) Yamada, H.; Yamashita, Y.; Kikuchi, M.; Watanabe, H.; Okujima, T.; Ogawa, T.; Ohara, K.; Ono, N. Chem. Eur. J. 2005, 11, 6212. 6. Angliker, H.; Rommel, E.; Wirz, J. Chem. Phys. Lett. 1982, 87, 208. 7. Reddy, A. R.; Bendikov, M. Chem. Commun. 2006, 1179. 8. Payne, M. M.; Parkin, S. R.; Anthony, J. E. J. Am. Chem. Soc. 2005, 127, 8028. 9. Schleyer, P. v. R.; Manoharan, M.; Jiao, H.; Stahl, F. Org. Lett. 2001, 3, 3643. 10. Bendikov, M.; Duong, H. M.; Starkey, K.; Houk, K. N.; Carter, E. A.; Wudl, F. J. Am. Chem. Soc. 2004, 126, 7416. 11. Houk, K. N.; Lee, P. S.; Nendel, M. J. Org. Chem. 2001, 66, 5517. 12. Poater, J.; Bofill, J. M.; Alemany, P.; Solà, M. J. Phys. Chem. A 2005, 109, 10629. 13. Udding, A. C. Doctoral thesis, University of Groningen, Groningen, Netherlands, 1963. 14. Givan, A.; Loewenschuss, A.; Nielsen, C. J. J. Chem. Soc., Faraday Trans. 1996, 92, 4927. 15. Bettinger, H. F.; Mondal, R.; Neckers, D. C. in preparation. 16. Gabioud, R.; Vogel, P. Tetrahedron 1980, 36, 149. 17. Bowles, D. M.; Anthony, J. E. Org. Lett. 2000, 2, 85.

CHAPTER 6. MECHANISM OF PHOTO-DECARBONYLATION*

6.1 Introduction

As mentioned earlier, the soluble α-diketone photoprecursors of hexacene and heptacene undergo Strating-Zwanenburg photodecarbonylation (Scheme 6.1) yielding the corresponding poly(acene)s.1,2,3 This photodecarbonylation was recently shown to be an efficient method to

synthesize poly(acene)s including pentacene.4 The detailed mechanism of the

photodecarbonylation, however, has yet to be delineated.

Scheme 6.1 Photochemical synthesis of poly(acene)s

O O

hυ

m n m n

The photodecarbonylation of α-diketones is an unusual reaction driven, in the aromatic precursor case, by the stability of the photoproduct. In most instances, as in the cases of the aliphatic diketones biacetyl, camphor quinone, and benzil, photoreduction via electron transfer is generally followed by proton-transfer to the radical anion.5 In only one case of which we are

aware, that of photolysis of the homoallylic conjugated bicyclic α-diketone bicyclo[2.2.2]octane-

2,3-dione (6.1), does bisdecarbonylation result, in this case, producing 1,3-cyclohexadiene.6,7

Allylic rearrangement forming 6.1a, the isomer of 6.1 (a cyclobutadione derivative) competes with bisdecarbonylation (Scheme 6.2).6,7 Cyclobutadione 6.1a also forms 1,3-cyclohexadiene by

bisdecarbonylation. Rearrangement was suggested to occur from the singlet manifold of 6.1,

98 while decarbonylation was presumed to result from the triplet manifold of both 6.1 and 6.1a.7

Similar rearrangement prior to photodecarbonylation of any α-diketone that is a precursor to an

‘aromatic’ acene seems unlikely because the process would involve loss of aromaticity. For example, isomerization of the α-diketone of anthracene (6.2) into its allylic rearranged isomer

6.2a seems highly unlikely (Scheme 6.3). Bisdecarbonylation may be the only possibility in such cases.

Scheme 6.2 Photochemistry of bicyclo[2.2.2]octane-2,3-dione

O O

O 6.1a O hν

6.1

Scheme 6.3 Photochemistry of α-diketone of anthracene

O O O O

hν

6.2 6.2a

Time resolved nanosecond laser flash photolysis (LFP) and femtosecond pump-probe

UV-vis spectrometric techniques along with steady state photolysis were used to try and understand the mechanism of the Strating-Zwanenburg photodecarbonylation. The detailed results obtained for the relaxation and photodecarbonylation of the excited states of 9,10-

99 dihydro-9,10-ethanoanthracene-11,12-dione (6.2), 6,15-dihydro-6,15-ethanohexacene-17,18- dione (6.3), and 7,16-dihydro-7,16-ethanoheptacene-19,20-dione (6.4) are presented and compared with that of 6.1 (Chart 6.1).

O O O O O O O O

6.1 6.2 6.3 6.4

Chart 6.1. Structure of α-diketones (6.1-6.4).

6.2 Results and Discussion

6.2.1 Photophysical Properties of the α-Diketones

The absorption spectra of 6.2-6.4 recorded in toluene show an n-π* band attributed to the

dicarbonyl functionality centered around 461-466 nm. Compound 6.3 and 6.4 show an additional

π-π* band in the 330 - 370 nm region attributable to the anthracene moieties. Table 6.1

summarizes the photophysical properties measured in toluene. A weak fluorescence (λF = 525 -

530 nm) was observed from 6.2-6.4 (excitation at 450 nm). Due to the low quantum efficiency of

8 fluorescence (ΦF = 0.001 - 0.004), the Stickler-Berg equation was used to estimate singlet

lifetimes (τs), which are found to be very short (τs ~59 - 453 ps). Interestingly, the τs value decreases with an increase in the number of rings in the molecule.

The phosphorescence spectra of 6.2-6.4 (λPh ~ 570 nm) recorded at 77 K in a frozen

matrix of methanol/ethanol (1:4) are in good agreement with that of other α-diketones, such as

9 benzil, furil, and camphor quinone. Similarly, the triplet energies (ET) deduced from the

phosphorescence spectra (53.5 - 54.5 kcal/mol) are found to be in good agreement with the ET

100

10 values of other α-diketones (ET ~58 kcal/mol). The singlet energies of 6.2-6.4 (0,0 energy, ES)

have been calculated from the intersection of the absorption and fluorescence spectra and are

found to be 57.7 - 58.8 kcal/mol.

Table 6.1. Photophysical properties of α-diketones (6.2-6.4)

E A (nm) [ε λ λ S E Comp. max F Φ b,c τ (ps)d τ (ps)e Ph (kcal/mol) T (M-1cm-1)]a (nm)a,b F s s (nm)f (kcal/mol)h g 295 [1530], 6.2 527 0.001 453 218.5±4.5 565 58.0 53.5 461 [310] 327 [5000], 345 [5500], 6.3 525 0.003 83 29.4±3.1 570 57.7 53.9 367 [4200], 465 [850] 347 [8850], 6.4 368 [6640], 530 0.004 59 19.1±1.1 570 58.8 54.5 466 [1430] a b Absorption and fluorescence spectra recorded in toluene; Excitation wavelength = 460 nm for λF and ΦF; c d ΦF values are relative to that of tetracene (0.17 in cyclohexane); τs values are estimated from the Strickler- e f Berg equation; τs values are measured from the ultrafast pump-probe experiments; Phosphorescence spectrum recorded in methanol/ethanol (1:4) matrix at 77 K; gSinglet energies calculated from 0-0 energy; hTriplet energies estimated from the phosphorescence spectra.

6.2.2 Steady-state Photolysis of the α-Diketones

Anthracene is formed exclusively when a dilute solution of 6.2 in toluene is irradiated (Figure

6.1). A similar photolysis of 6.3 in solution produces hexacene, but the latter vanishes quickly due to its susceptibility towards oxidation and dimerization.1 6.4 is even more illusive, and

formation of heptacene in solution is instantly followed by oxidation or dimerization.2

Nevertheless, hexacenes and heptacenes can be synthesized by irradiating 6.3 and 6.4, respectively, and retained for a considerable period of time (4-12 hrs) in polymer matrices.1,2,11

Steady state photolysis of 6.2-6.4 is a clean reaction in that no other products are observed except for the corresponding acene.

101

Figure 6.1. Steady state photolysis of 6.2 in toluene. Inset: plot of ln(At-Aα) vs. time of irradiation; absorbance monitored at 460 nm.

The quantum yield of photodecarbonylation (ΦR) of 6.2 was measured using potassium

12 ferrioxalate actinometry. The ΦR of 6.2 in an oxygen saturated solution (0.020) was similar to that recorded from an argon saturated solution (0.021). On the other hand, when benzophenone

7,13 (ET = 69 kcal/mol) was used as a triplet sensitizer, a 2.5 fold increase in the ΦR (0.050) was

observed. This indicates involvement of the triplet state of 6.2 in the decarbonylation process.

Surprisingly, the ΦR values of 6.2 are considerably lower than that of the photodecarbonylation

of 6.1 sensitized with benzophenone (ΦR ~ 0.9), although the decarbonylation is followed by aromatization in the case of 6.2. The reason for this may be the formation of anthracene (ET = 47

kcal/mol), which might possibly quench the triplet state of 6.2.13 Interestingly the rate of

decarbonylation of an oxygen saturated solution of 6.2 in toluene shows no significant difference

102 from that observed using an argon saturated solution (Figure 6.1, Inset), despite the fact that photodecarbonylation proceeds from the triplet manifold. The observation of the unusually low lifetime of the triplet state of 6.2 (vide infra) is consistent with this. A similar instance of short lived triplet states (τ ~ 300 ps) not being affected by oxygen is reported for the naphthopyran.14

The ΦR values of 6.3 and 6.4 could not be calculated because the disappearance of 6.3

and of 6.4 is difficult to monitor due to the overlap of their absorption spectra with those of the

photoproducts which also absorb in the 400-500 nm region.1-2 The quantum yield of formation of

products in longer wavelength regions (600-800 nm) could also not be measured. The products of 6.3 and 6.4, i.e., hexacene and heptacene, respectively, are too reactive for their concentrations to be monitored precisely in solution at room temperature.

11.38 kcal/mol

38.84 kcal/mol 32.40 kcal/mol 6.2a

10.45 kcal/mol

6.1a

6.2 6.1

Figure 6.2. Energy differences of the ground state optimized structures of 6.1, 6.2, their isomers (6.1a and 6.2a), and photodecarbonylated products; 1,3-cyclohexadiene and anthracene, respectively.

103

The photochemistry of 6.1 presumably involves two competing reactions: (i) bisdecarbonylation occurring from the triplet state, and (ii) isomerization via allylic rearrangement which occurs from the singlet state.7 As mentioned earlier, isomerization seems

unlikely in the case of 6.2-6.4, because this would require the loss of aromaticity (Scheme 6.3).

The ground state energy difference between 6.1 and 6.1a was calculated to be 10.45 kcal/mol,

while that between 6.2 and 6.2a was 38.84 kcal/mol suggesting that conversion of 6.2 into 6.2a

is kinetically unfavorable (Figure 6.2). The ground states of 6.1, 6.2, and their corresponding

isomers (6.1a and 6.2a) were optimized using Gaussian03 program [B3LYP/6-31G(d,p)].15

6.2.3 Nanosecond Laser Flash Photolysis (LFP) of the α-Diketones

The absorption spectra obtained from the nanosecond laser flash photolysis of 6.2 in dry toluene

at room temperature (excitation wavelength - 460 nm, pulse width – 7 ns) are presented in

Figure 6.3. Absorption spectra obtained from the nanosecond laser flash photolysis of 6.2 in toluene, recorded (a) 40 ns (black), (b) 100 ns (red), and (c) 500 ns (green) after the laser pulse (λex = 460 nm).

104

Figure 6.3. Immediate bleaching in the 460 nm region indicates disappearance of 6.2 and occurs simultaneously with formation of anthracene which is observed in the 350 nm region. Thus, the photodecarbonylation occurs within 7 ns, the width of the laser pulse. No absorption due to any other transient was observed.

Formation of heptacene (650-800 nm) was observed when similar experiments were carried out with 6.4 in toluene, though heptacene is unstable11 and not observed during

photolysis in toluene at room temperature. However, heptacene appears to be stable enough to

appear during LFP experiments and lives through the data accumulation time window of the

instrument (1.4 ms).

An additional transient exhibiting a sharp peak at 580 nm together with a broad peak

around 460 nm was observed from the argon saturated solution of 6.4 (Figure 6.4). This transient

completely disappears in the presence of oxygen and can be assigned to the T-T absorption of

heptacene. The analogous diketone precursor of pentacene is known to produce a similar transient at 510 nm (τ ~ 48.48±0.15 μs) ascribable to the T-T absorption of pentacene.4 It

appears that heptacene produced within the laser pulse is further excited by the same laser pulse,

ultimately generating its triplet state. This indicates that the singlet state of heptacene is an

extremely short lived species, and intersystem crossing into the triplet state occurs well within 7

ns.

The lifetime of the T-T absorption of heptacene was calculated by monitoring the kinetic

traces at 580 nm which can be fitted satisfactorily with a monoexponential decay function with a

decay time of 10.62 ± 0.13 μs. Kinetic traces at 460 nm were similar to those at 580 nm,

confirming that they belong to the same transient species. The lifetime of the triplet state of

105

heptacene is quite similar to that of the triplet state of hexacene (τT ~10 μs), which is also known

to absorb at 550 nm.16

Figure 6.4. Transient absorption spectra obtained from the nanosecond laser flash photolysis of 6.4 in an argon saturated dry toluene, recorded (a) 3 μs (black), (b) 10 μs (red), and (c) 50 μs (green) after the laser pulse (λex = 460 nm). Inset: kinetic trace monitored at 580 nm.

6.2.4 Femtosecond Pump-probe Spectrometry of the α-Diketones

A 475 nm excitation wavelength (pulse width ~ 130 fs) was chosen to selectively pump the n-π* band of diketones 6.2-6.4 in toluene. A broad transient absorption from 475-650 nm was observed for 6.4 (Figure 6.5). Kinetic traces monitored at various wavelengths follow bi- exponential decay, revealing a total of three lifetimes. The shortest decay lifetime is 1.47±0.19 ps, the second is around 20 ps, and the longest lifetime is 370.4±46.9 ps. The second lifetime is close to the singlet state lifetime (τ = 59 ps) estimated using the Strickler-Berg equation. Thus, this transient absorption spectrum can be assigned to the absorption of the S1 state of 6.4. At the

early times of decay, there is no significant change in the spectral shape. But the shortest lifetime

106 component (1.47±0.19 ps) exhibits wavelength dependence, being smaller at the red edge of the transient spectrum and increasing towards the blue edge. This is indicative of a vibrational cooling process17 taking place in the singlet state manifold. The longest decay time (370.4±46.9 ps) might be due to either the T1 states of 6.4 or some biradical species, which form simultaneously with the formation of the S1 state (vide infra). This is unlikely to be associated

with the S1 state of 6.4 because the lifetimes of the singlet excited states of α-diketones are

generally short.18

(a) (b)

Figure 6.5. (a) Absorption difference spectra obtained from the pump-probe spectrometry of 6.4 in toluene recorded after 0.5 ps (green), 17 ps (red), and 105 ps (black) (130 fs excitation pulse at 475 nm, pulse energy = 5 μJ), inset: kinetic trace monitored at 620 nm; (b) Kinetic trace monitored at 500 nm, showing decay profile in a wider time window.

The 475 nm excitation of 6.2 and 6.3 in toluene results in transient spectra similar to that of 6.4. These also follow second order decay kinetics with three lifetimes. Vibrational relaxation

in the S1 manifold of 6.2 and 6.3 is completed in less than 5 and 2 ps, respectively. Lifetimes of

the vibrationally relaxed S1 state of 2 and 3 were found to be 218.5±4.5 ps and 29.4±3.1 ps,

107 respectively. Similar to the trend shown by singlet lifetimes estimated from the Strickler-Berg equation, these lifetimes decrease with an increase in the number of rings in the molecule.

The transients with the longest lifetime (>350 ps < 7 ns) for 6.2 and 6.3 appear as an offset of the baseline, possibly due to the low signal intensity. Another possible reason for the masked transient signal could be interference of the excitation beam (475 nm). Data were also acquired in the wider time window by exciting the molecules at 400 nm in anticipation of revealing the origin of the longer lived component. The 400 nm excitation produced the S1 absorptions (525-700 nm) along with the long-lived transients (>350 ps < 7 ns) for 6.2-6.4 in the region of 460-505 nm (Figure 6.6). Kinetic traces belonging to the long lived transient could be extracted despite the significant overlap of this band with the S1 absorption. These transients

showed a red shift in the series from 6.2 (460 nm) to 6.3 (495 nm) and 6.4 (505 nm).

As indicated earlier, these transients might be assigned to either the T1 states of 6.2-6.4 or

some biradical species. The triplet state of certain diketones, for example of 1,2-

aceanthrylenedione, reportedly absorbs at 465 nm.19 Thus, the observed longer lived transients

are most likely the T1 states of 6.2-6.4, the formation of which can be explained by rapid intersystem crossing from the S1 state. The small singlet-triplet energy gap of 6.2-6.4 (ES and ET difference ~ 4 kcal/mol) may be the reason for the fast intersystem crossing. As discussed earlier, the unusually short lived triplet state of 6.2 also explains why the quantum yield of its decarbonylation, which takes places from the triplet manifold, was not observed to be affected by oxygen.

108

Figure 6.6. Absorption difference spectra obtained from the pump-probe spectrometry of 6.2-6.4 in toluene(130 fs excitation pulse at 400 nm, pulse energy = 5 μJ); (a) spectrum of 6.4 taken after 1 ps of the laser pulse; (b) spectrum of 6.3 taken after 1.6 ps of the laser pulse; and (c) spectrum of 6.2 taken after 50 ps of the laser pulse.

6.2.5 Proposed Mechanism of Photodecarbonylation

The pump-probe experiments in combination with those obtained from nanosecond LFP clearly indicate that the decarbonylation process is initiated within several hundreds of picoseconds from the triplet manifold and is completed within the pulse duration of the nanosecond excitation laser beam (7 ns). The proposed mechanism of the photodecarbnylation of 6.2-6.4 is presented in

Scheme 6.4. Like 6.1, these diketones undergo decarbonylation from the triplet manifold.

However, the major difference between 6.1 and 6.2-6.4 appears to be that the former undergoes isomerization from the singlet manifold while the latter likely undergoes decarbonylation from the singlet manifold as well. Furthermore, it seems most likely that the bond between a carbonyl carbon atom and a methylene carbon gets broken first forming the radical 6.5a. This hypothesis

109 is supported by the facile isomerization of 6.1 from the singlet manifold, during which the carbonyl carbon-methylene carbon bond requires that it be exclusively broken. Regardless, there is no irrefutable evidence to rule out the possibility of a first bond dissociation between the carbonyl carbon atoms. Decarbonylation might also proceed through the radical 6.5b. However, it is for sure that the decarbonylation process is a fast process completing within 7 ns.

Scheme 6.4. Proposed mechanism of decarbonylation

hν 1 ~1.5 - 5.4 ps 1 ISC 3 α-diketone α-diketone * α-diketone * α-diketone * vibrational fast relaxation τ ~ 20 - 218 ps (τ > 370 ps)

O O O O < 7ns + m n m n m n m n Poly(acene) 6.6 6.5a 6.5b

We are unable to comment on whether the radicals 6.5a and 6.5b lose the carbonyl

functionalities in either a concerted or in as step-wise reaction to finally produce the

corresponding poly(acene) through the biradical 6.6. Whether the dimer of carbon monoxide or

two molecules of carbon monoxide are expulsed during the decarbonylation is still an open

question. Spectral recognition of several of the intermediates, especially radicals, cannot be

achieved using any experiment the observation of which is via UV or visible methodology,

primarily due to either the significant spectral overlap of such species or their absorptions

appearing in the 1.4 ns – 7.0 ns time regime.

110

6.3 Experimental Section

Chemicals and solvents required were obtained from commercial suppliers. The solvents used were spectrophotometric grade, and distilled over sodium as required. The syntheses of 6.2-6.4 are described in Chapter 5.

Absorption and fluorescence spectra were recorded on a Shimadzu UV-2401 spectrophotometer and a Fluorolog®-3 spectrometer, respectively. All measurements were carried out at room temperature (22 ± 2 oC) unless otherwise specified. A 395 nm (±25 nm) UV-

LED lamp was used for the steady state photolysis. The quantum yield of photodecarbonylation

(ΦR) of 6.2 in toluene was calculated by first estimating the light intensity of the irradiation using

a potassium ferrioxalate actinometry.12

Fluorescence quantum yields (ΦF) in toluene were measured following a general method

20 -4 using tetracene (ΦF = 0.17 in cyclohexane) as the standard. Dilute solutions of 6.2-6.4 (~10

M) were used for recording the fluorescence spectra. Sample solutions were added to quartz cuvettes and degassed for 15 min. The degassed solution had an absorbance of 0.15-0.3 at the excitation wavelength (450 nm). The refractive indices of solvents at the sodium D line were used.

The singlet lifetime of 6.2-6.4 in toluene was estimated using the Strickler-Berg equation8 which allows calculation of the radiative lifetime of fluorescence from absorption and fluorescence spectra:

3 n −1 ε dν 12.8810τν==k ∗∗∗−−93f ∗ rad radn f ∫ ν a

where,

111

- nf and na are the mean refractive indices of the solvent over the fluorescence band and the

21 absorption band (for liquid toluene at 20 °C nf = n520nm = 1.5036 and na = n461nm = 1.5125);

- ε is the molar decadic absorption coefficient;

- ν is frequency, cm-1;

−3 −1 −3 - the term 〈ν f 〉 is the inverse of the expectation value of ν f , calculated from the integrals of

22 the fluorescence intensity I f (ν ) as follows:

I f (ν )dν I f (ν )dν 〈v−3 〉 −1 = ∫ = ∫ f I f (ν )dν ⎛ 1 ⎞ I f (ν )d⎜− ⎟ ∫ ν 3 ∫ ⎝ 2ν 2 ⎠

Nanosecond laser flash photolysis (LFP) studies were performed on a kinetic spectrometric detection system previously described.23 The excitation pulse (460 nm, 0.5-3 mJ/pulse) was generated with nonlinear parametric oscillator of a Q-switched Nd:YAG laser.

The excitation pulse width was ~7 ns. Transient produced were followed temporally and spectrally by a computer controlled kinetic spectrophotometer. The sample solutions showing an absorbance in the range of 0.4-0.6 at the excitation wavelength of 460 nm in 1 cm2 quartz cuvettes were used and degassed continuously with argon while inert atmosphere was needed.

All experiments were carried out using a 1 cm flow cell.

The laser system and apparatus used for the femtosecond transient absorption spectroscopy has been described elsewhere.24 Improvements that enhanced signal-to-noise characteristics were reported previously.25 A Spectra-Physics Hurricane/Evolution Ti:sapphire combination generated 800 nm pulses of 130 fs duration at a rate of 1 kHz. In the current experiments, the excitation wavelength of 400 nm and 475 were used. The 400 nm was derived from the second harmonic of the Ti:sapphire fundamental using a Super Tripler instrument (CSK

112

Optronics). The 475 nm was generated using with optical parametric amplifier (OPA, Spectra-

Physics OPA 800). The excitation energy was 5 μj/pulse. In-house LabVIEW (National

Instruments) software allowed automatic spectral acquisition over a series of delay line settings.

Kinetic traces at appropriate wavelengths were assembled from the accumulated spectral data.

Sample solutions were prepared to have an absorbance of 0.6-0.8 at 460 nm in 1 cm2 cuvette.

The experiments were carried out in the 2 mm flow cell and were used without deaeration.

6.4 Conclusion

Steady state flash photolysis, nanosecond LFP, and femtosecond pump-probe experiments of 6.2-6.4 (α-diketone precursors of acenes) were carried out to understand the mechanism of the Strating-Zwanenburg photodecarbonylation. It appears that both the singlet and triplet states of the diketones are involved in the decarbonylation process. Compounds 6.2-

6.4 show weak fluorescence (λF ~ 525-530 nm, ΦF ~ 0.1-0.4%) and phosphorescence (λPh ~ 565-

570 nm) and have a small singlet-triplet energy gap (~ 4 kcal/mol). The lifetimes of the singlet excited states of 6.2 (τs ~ 218 ps), 6.3 (τs ~ 29 ps), and 6.4 (τs ~ 20 ps) decrease as the number of the benzenoid ring increases in the molecule. The triplet states of 6.2-6.4 (> 370 ps < 7 ns) are also short lived and do not appear during the nanosecond experiments. Instant appearance of the corresponding poly(acene)s during nanosecond LFP indicates the decarbonylation to occur within the laser pulse width (7 ns). During the LFP experiment of 6.4, the triplet state of the photoproducts, i.e., heptacene (λmax = 580 nm, τ ~ 11 μs), was also observed.

6.5 References

* Partially adopted from: Mondal, R.; Okhrimenko, A. N.; Shah, B. K.; Neckers, D. C. J. Am. Chem.Soc. 2007, submitted.

113

1. Mondal, R.; Adhikari, R. M.; Shah, B. K.; Neckers, D. C. Org. Lett. 2007, 9, 2505. 2. Mondal. R.; Shah, B. K.; Neckers, D. C. J. Am. Chem. Soc. 2006, 128, 9612. 3. Strating, J.; Zwanenburg, B.; Wagenaar, A.; Udding, A. C. Tetrahedron Lett. 1969, 10, 125. 4. Uno, H.; Yamashita, Y.; Kikuchi, M.; Watanabe, H.; Yamada, H.; Okujima, T.; Ogawa, T.; Ono, N. Tetrahedron Lett. 2005, 46, 1981. 5. (a) Gream, G. E.; Paice, J. C.; Ramsay, C. C. R. Aust. J. Chem. 1969, 22, 1229; (b) Turro, N. J.; Lee, T.-J. J. Am. Chem. Soc. 1969, 91, 5651; (c) Rubbin, M. B.; Gutman, A. L. J. Org. Chem. 1986, 51, 2511; (d) Park, J. W.; Kim, E. K.; Park, K. K. Bull. Korean Chem. Soc. 2002, 23, 1229. 6. (a) Rubin, M.B. J. Am. Chem. Soc. 1981, 103, 7791; (b) Hassoon, S.; Rubin, M.B.; Speiser, S. J. Photochem. 1984, 26, 295. 7. Rubin, M.B.; Kapon, M. J. Photochem. Photobiol. A Chem. 1999, 124, 41. 8. Strickler, S. J.; Berg, R. A. J. Chem. Phys. 1962, 37, 814. 9. Kuboyama, A.; Yabe, S. Bull. Chem. Soc. Jap. 1967, 40, 2475. 10. Turro, N. J.; Lee, T.-J. J. Am. Chem. Soc. 1969, 91, 5651. 11. Mondal, R.; Shah, B. K.; Neckers, D. C. J. Photochem. Photobiol. A Chem. 2007, published on web (DOI:10.1016/j.jphotochem.2007.05.002). 12. Murov, S. L.; Carmichael, I.; Hug, G. L. Handbook of Photochemistry, 2nd ed. Marcel Dekker, Inc. New York, 1993, 299. 13. Turro, N. J. Modern Molecular Photochemistry, The Benjamin/Cummings Publishing Co., Inc. Menlo Park, CA, 1978. 14. Hobley, J.; Malatesta, V.; Hatanaka, K.; Kajimoto, S.; Williams, S. L.; Fukumura, H. Phys. Chem. Chem. Phys. 2002, 4, 180. 15. Gaussian03, Inc. 16. Angliker, H.; Rommel, E.; Wirz, J. Chem. Phys. Lett. 1982, 87, 208. 17. Okhrimenko, A. N.; Gusev, A. V.; Rodgers, M. A. J. J. Phys. Chem. A 2005, 109, 7653. 18. Singh, A.; Scott, A. R.; Sopchyshyn, F. J. Phys. Chem. 1969, 73, 2633. 19. (a) Serra, A. C. S.; de Lucas, N. C.; Netto-Ferrira, J. C. J. Braz. Chem. Soc. 2004, 15, 481; (b) de Lucas, N. C.; Correa, R. J.; Albuquerque, A. C. C.; Firme, C. L.; Garden, S. J.; Bertoti, A. R.; Netto-Ferreira, J. C. J. Phys. Chem. A. 2007, 111, 1117. 20. Birks, J. B. Photophysics of Aromatic Molecules, Wiley-Interscience, New York, 1970, 704. 21. Samoc, A. J. Appl. Phys. 2003, 94, 6167 22. Tauber, M. J.; Mathies, R. A. J. Phys. Chem. A 2001, 105, 10952. 23. Ford, W. E.; Rodgers, M. A. J. J. Phys. Chem. 1994, 98, 3822. 24. Nikolaitchik, A. V.; Korth, O.; Rodgers, M. A. J. J. Phys. Chem. A 1999, 103, 7587. 25. Pelliccioli, A. P.; Henbest, K.; Kwag, G.; Carvagno, T. R.; Kenney, M. E.; Rodgers, M. A. J. J. Phys. Chem. A 2001, 105, 1757.

114

CHAPTER 7. DETERMINATION OF OXYGEN PERMEABILITY USING

HEPTACENE*

7.1 Introduction

The oxygen permeability (Pm) of biocompatible and other polymers used in the packaging industry, as surface protective coatings, and in drug delivery membrane systems is an important physical parameter.1-3 Permeability of a polymer film depends on several factors including density, crystallinity, orientation and cross-linking characteristics of the polymer as well as on

4 the conditions of measurement. Determining the Pm of a polymer film usually requires measurement of a change in gas pressure across the film.5,6 The pressure difference is caused by the gas transferring through the membrane from a high-pressure side to a low-pressure side and measured in a specific apparatus. Although the principle is simple, these measurements are tedious and require a long time and expensive instrumentation.

Polarographic methods and membrane covered electrode techniques are used to measure

7,8 the Pm. Optical spectroscopy techniques, such as fluorescence, phosphorescence, and triplet- triplet absorption quenching of organic compounds have also been used for this purpose with varying degree of success.3,9-16 However, these spectroscopic methods require the generation of the excited states of the sensing molecule and depend upon the measurement of the dynamics and behavior of those excited states. Fluorescence methods are also found to be sensitive to the presence of even small amount of impurities in the polymer.

In this chapter, we demonstrate the use of a UV-visible spectroscopic technique to estimate the Pm of polymer films in a simple, fast, and reliable manner. Recently, we reported the

115 synthesis of a highly reactive poly(acene), heptacene (7.1), using photodecarbonylation of its α- diketone precursor (7.2) (Scheme 7.1).17

Scheme 7.1. Photochemical synthesis of heptacene

O O 395 nm

7.2 7.1

Heptacene (7.1) was generated in a poly(methyl methacrylate) matrix and the photodecarbonylation achieved with a ultraviolet light-emitting diode (UV-LED) light source

(λmax - 395 nm) that almost selectively addresses the n-π* absorption band of the dione 7.2. The photodecarbonylation is a clean process exclusively leading to the formation of 7.1, which shows absorption bands in the visible and near infra-red region (600-825 nm). However, 7.1 is extremely sensitive to oxidation and is easily converted to oxidized products (endoperoxides) by molecular oxygen when the latter diffuses into the polymer matrix as visibly demonstrated by the gradual disappearance of its absorption.17

We report that quenching of the absorption of 7.1 due to its oxidation in polymer matrices follows a pseudo first order kinetics and is dependent on the Pm of the polymer. The lower the Pm of the polymer, the lower was the rate of quenching of 7.1, and vice versa. This has enabled us to develop a technique that utilizes 7.1 as a marker to estimate the Pm of polymer films. The technique is based on a simple concept- once 7.1 is generated in a thin film of a polymer, disappearance of its absorption in the visible region can be monitored as a function of the oxygen

116 permeability of the polymer. Since this method is based on the measurement of the change in the absorption of 7.1 embedded in the polymer matrix, it requires a simple measurement setup and avoids fallacies and measurement difficulties associated with excited state chemistries.

7.2 Results and Discussion

In order to demonstrate the applicability of the concept, five commercially available polymers- low-density polyethylene (LDPE), polystyrene (PS), poly(ethyl methacrylate) (PEMA), poly(methyl methacrylate) (PMMA), and poly(vinyl chloride) (PVC) were chosen, These

-13 -13 2 -1 -1 5 polymer show different Pm values ranging from 0.034x10 to 2.2x10 cm .s .Pa . Films of

PS, PEMA, PMMA, PVC, and LDPE of uniform thickness (0.50±0.02 mm) containing 7.2 were formed. The prepared polymer films were irradiated to generate 7.1. The relative amount of 7.1 generated could be estimated from the intensity of its absorption spectra.

When the films were left in air under limited ambient light, the depletion of 7.1 as a function of time was recorded by monitoring the absorption intensity at 760 nm. The gradual disappearance of the absorption of 7.1 recorded at different times in the case of the PMMA film is shown in Figure 7.1. The rates of decrease in the intensity at 760 nm were observed to be different for each polymer film. For example, the absorbance of 7.1 was found to completely disappear in about 115, 145, 380, and 435 minutes in the case of PS, PEMA, PMMA, and PVC, respectively. Interestingly, the order in which 7.1 underwent oxidation in these polymers correlates well with the order of their Pm and the depletion of 7.1 is directly related with the amount of oxygen that diffuses into the films. The decrease in the absorption of 7.1 followed mono-exponential decay in each case (Figure 7.2). This can be explained by a pseudo-first order reaction rate law.

117

Figure 7.1. The absorption spectra of a PMMA film containing 7.1 recorded at different times.

Figure 7.2. Decomposition profiles of 7.1 in films of different polymers as monitored at 760 nm; (a) blue: polystyrene, (b) red: poly(ethyl methacrylate), (c) green: poly(methyl methacrylate), and (d) orange: poly(vinyl chloride).

The rate of decomposition of 7.1 (kd) and its half-life (t1/2) in PS, PEMA, PMMA, and

PVC films were calculated from the exponential fitting function (Table 7.1). The kd value

118

increases and half-life of 7.1 in the polymer films decreases with the decreasing Pm values of the polymers. A linear behavior was observed between kd and Pm (Figure 7.3). This linear

-13 -13 relationship allows one to estimate the Pm of a polymer in the range of 0.034 × 10 to 2.2 × 10

2 -1 -1 cm .s .Pa , if its kd can be experimentally evaluated using this technique.

Table 7.1. Rate constant of disappearance (kd) and half-life (t1/2) of 7.1 in different polymer films with varying oxygen permeability (Pm)

a Film thickness Pm kd t1/2 Polymer mm 10-13 × cm2.s-1.Pa-1 s-1 Min

LDPE 0.52 2.2 - <10

PS 0.51 1.9 7.25 × 10-4 16

PEMA 0.48 0.889 4.27 × 10-4 27

PMMA 0.48 0.116 1.60 × 10-4 73

PVC 0.50 0.034 1.03 × 10-4 112

a The Pm data were taken from ref. 4.

A reasonable concentration of 7.1 could not be produced in LDPE film because of its higher Pm value preventing accurate monitoring of the reaction of 7.1 with oxygen in this polymer. In the case of PS, an equilibrium condition involving formation and disappearance of

7.1, when produced over a longer time with a lower flux of photons, was observed. Thus, the technique may be most useful for polymers that show medium to low oxygen permeability. A limitation of the technique is that the polymer to be evaluated needs to be transparent >390 nm to facilitate formation of 7.1 in the film and to be compatible with UV-vis spectroscopic measurements in the visible region.

119

Figure 7.3. Plot of the rates of disappearance (kd) of 7.1 vs. oxygen permeability (Pm).

Addition of molecular oxygen to 7.1, preferably at the central ring, is responsible for its rapid disappearance.18 Poly(acene)s are known to be highly reactive towards Diels-Alder addition.18 Pentacene, for example, undergoes oxidation at the central ring to form an endoperoxide.19 Both concerted and stepwise processes have been postulated for formation of endoperoxides from acenes,20,21 and the reactivity is the highest at the central ring but gradually decreases as one moves outward. For example, the energy barrier (Ea) of DA addition of acetylene to the central ring of pentacene is 24 kcal/mol, whereas the barriers is 32.7 kcal/mol for the edge ring of pentacene.18 Nevertheless, endoperoxides were also formed due to oxidation at the peripheral rings in the case of 7.1 (Scheme 7.2).

120

Scheme 7.2. Formation of endoperoxides of heptacene

O O

O2 + 7.1 O O O O

and similar other adducts

Figure 7.4. 1H-NMR spectra of 7.2 recorded before irradiation (bottom,) and after 15 minutes of irradiation (top) in CDCl3 purged with oxygen. (The peaks appeared in the 6.0-6.3 ppm region in the top spectrum are due to the protons at the C-atoms that are attached with the oxygen bridge in endoperoxides).

Formation of oxygen adducts of 7.1 was verified by 1H-NMR and MALDI-MS experiments. The protons of the bridged C-atoms of 7.2 appear at 5.4 ppm. When a CDCl3 solution of 7.2 purged with oxygen is irradiated, formation of 7.1 is instantly followed by its oxidation. The 1H-NMR of such an irradiated solution showed three additional peaks between

6.0-6.3 ppm (Figure 7.4) indicating endoperoxides of 7.1 as shown in Scheme 7.2. The peaks can be assigned to the protons at the C-atoms attached at the oxygen bridge. The reason for many

121 proton peaks in the 6.0-6.3 ppm region comes from formation of bridges at different positions.

The formation of endoperoxides containing two bridges is also possible. The MALDI-MS analysis of the same solution also suggested the formation of oxygen adducts of 7.1, containing one as well as two molecules of oxygen.

Whether the oxidation of 7.1 is caused solely by molecular oxygen diffusing into the polymer or some other species is also contributing to its disappearance deserves consideration.

Similar experiments carried out in the dark confirm that the disappearance of 7.1 was not light induced. There is a possibility that a small amount of moisture may permeate into the polymer films. However, it is highly unlikely that this will have any effect because permeation of moisture into the polymer film in comparison to that of oxygen can be considered negligible.

Even if this is the case, moisture is likely to have no chemical effect on 7.1. The calculated energy barrier for the reaction between 7.1 and water (43 kcal/mol) is much higher than that of the reaction between 7.1 and oxygen (<20 kcal/mol) and poly(acene)s do not generally react with water..22 Thus, it can be safely said that the depletion of 7.1 is due to its oxidation. This is substantiated by the fact that there exists a direct correlation between the rate of decomposition of 7.1 and Pm of polymer films.

7.3 Experimental Section

7.3.1 Materials and Instruments

Five commercially available polymers - low-density polyethylene (LDPE), polystyrene (PS), poly(ethyl methacrylate) (PEMA), poly(methyl methacrylate) (PMMA), and poly(vinyl chloride)

(PVC) were chosen for the study. These polymers have different Pm values. Linear LDPE (1,4-

o cis: 20-45%, 1,4-trans: 45-65%, vinyl: 13-30%; density = 0.918 g/mL at 25 C), PEMA [Mw =

122

340000, Mw/Mn ~ 2.7], PMMA [Mw = 120000], and PVC [Mw = 43000, Mw/Mn ~ 1.95] were purchased from Sigma-Aldrich and PS [Mw = 100000, Mw/Mn ~ 1.06] was purchased from

Polysciences Inc. All polymers were used as received without further purification.

Absorption spectra of the films were recorded on Cary 50 UV-Vis spectrophotometer.

Matrix assisted laser desorption ionization (MALDI) spectra were obtained using Bruker

Daltonic Omniflex® instrument (N2 laser, 337 nm). A Bruker spectrometer (working frequency

300.0 MHz for 1H) was used to record NMR spectra. Photolysis of 7.2 was carried out using an ultraviolet light-emitting diode (UV-LED) light source (λmax - 395 nm) from UV Process Supply,

Inc. All measurements were carried out at room temperature (22±2 oC) unless otherwise specified. The synthesis of 7.2 is described elsewhere.17

7.3.2 Sample Preparation

A saturated solution of photoprecursor (7.2) in dichloromethane was mixed with a solution of the target polymers (except LDPE) in dichloromethane. Thin films were cast by carefully spreading a few drops of polymer solutions of 7.2 of identical concentration on quartz discs. The films were dried overnight at room temperature until all solvent had evaporated and the thickness

(0.50±0.02 mm) measured using a digital caliper. When similarly prepared films were put in the oven at 65-70 oC for 6 hours, they showed no significant loss in weight, indicating that the amount of the solvent in the films before oven drying was negligible. The weight losses were

5%, 1%, 1.5%, and 2% for PS, PEMA, PMMA, and PVC, respectively. Films of LDPE embedded with 7.2 were cast thermally because of the poor solubility of LDPE in low-boiling organic solvents.

123

7.3.3 Determination of Rate Constant (kd) of Disappearance of Heptacene

The prepared polymer films were irradiated using a UV-LED array (395±25 nm) for 20 minutes to generate 7.1. The relative amount of 7.1 generated was estimated from the intensity of its absorption spectra. The films were then left in the air under limited ambient light and the spectra of films of each polymer recorded at regular intervals until the absorption of 7.1 had completely disappeared. Decomposition of 7.1 as a function of time was recorded by monitoring the absorption intensity at 760 nm.

The oxidation of 7.1 by the diffused molecular oxygen into the polymer matrix can be represented as follows:

7.1 + O2 Æ Products

The rate law of the above reaction is written as: d[7.1]/dt = - k’ [7.1] [O2], where k’ is the second order rate constant. In any particular polymer film, the concentration of diffused oxygen will depend on the oxygen permeability of the film. Oxygen diffusion depends on the oxygen permeability of the polymer and can be considered constant with time. Thus, k’ [O2] can be considered to remain constant for a particular polymer and the rate law becomes pseudo first order: d[7.1]/dt = kd [7.1], where kd is the pseudo or effective first-order rate constant and equals

-kd.t to k’ [O2]. Solution of this rate law will be [7.1] = [7.1]o e . Without going into the details of

Fick’s and Henry’s law of diffusion, we estimated kd from the mono-exponential decay of 7.1 in polymer.

7.4 Conclusion

Rapid oxidation of 7.1 occurs when a polymer film containing 7.1 is exposed to air and this could be easily monitored by following gradual disappearance of its absorption in the visible

124 region. The rates of disappearance of 7.1 in different polymer films were observed to follow pseudo first order kinetics. Interestingly, those rates measured in the films of PS (7.25 × 10-4 s-1),

PEMA (4.27 × 10-4 s-1), PMMA (1.60 × 10-4 s-1), and PVC (1.03 × 10-4 s-1) were found to correlate well with their oxygen permeability values. This indicates that the high reactivity of 7.1 towards molecular oxygen can be used to determine the oxygen permeability of polymers.

7.5 References

*Adopted from: Mondal, R.; Shah, B. K.; Neckers, D. C. J. Photochem. Photobiol. A Chem. 2007, in press.

1. Wichterlová, J.; Wichterle, K.; Michálek, J. Polymer 2005, 46, 9974. 2. Galić, K.; Ciković, N. Polym. Test. 2001, 20, 599. 3. Chu, D. Y.; Thomas, J. K. Macromolecules 1988, 21, 2094. 4. Pauly, S. Permeability and Diffusion Data. In Polymer Handbook; Brandrup, J.; Immergut, E. H., Eds., John Wiley & Sons, Inc.: New York, 1989; p VI/435. 5. Tuwiner, S. B. Diffusion and Membrane Technology; Reinhold: New York, 1962; p 232. 6. Celina, M., Gillen, K. T. Macromolecules 2005, 38, 2754. 7. Aiba, S.; Ohashi, M.; Huang, S.-Y. Ind. Eng. Chem. Fundam. 1968, 7, 497. 8. Mancy, K. H.; Okun, D. A.; Reilley, C. N. J. Elctroanal. Chem. 1962, 4, 65. 9. Cox, M. E.; Dunn, B. Appl. Opt. 1985, 24, 2114. 10. Cox, M. E. J. Polym. Sc. Polym. Chem. 1986, 24, 621. 11. Hormats, E. I.; Unterleitner, F. C. J. Phys. Chem. 1965, 69, 3677. 12. Oster, G.; Geacintov, N.; Khan, A. U. Nature 1962, 196, 1089. 13. MacCallum, J. R.; Rudkin, A. L. Eur. Polym. J. 1978, 14, 655. 14. Yekta, A.; Masoumi, Z.; Winnik, M. A. Can. J. Chem. 1995, 73, 2021. 15. Rharbi, Y.; Yekta, A.; Winnik, M. A. Anal. Chem. 1999, 71, 5045. 16. Jayarajah, C. N.; Yekta, A.; Manners, I.; Inc., M. A. Macromolecules 2000, 33, 5693. 17. Mondal, R.; Shah, B. K.; Neckers, D. C. J. Am. Chem. Soc. 2006, 128, 9612. 18. Schleyer, P. v. R.; Manoharan, M.; Jiao, H.; Stahl, F. Org. Lett. 2001, 3, 3643. 19. Yamada, H.; Yamashita, Y.; Kikuchi, M.; Watanabe, H.; Okujima, T.; Ogawa, T.; Ohara, K.; Ono, N. Chem.- Eur. J. 2005, 11, 6212. 20. Chien, S. -H.; Cheng, M. -F.; Lau, K. -C.; Li, W, -K. J. Phys. Chem. A 2005, 109, 7509. 21. Reddy, A. R.; Bendikov, M. Chem. Comm. 2006, 1179 22. Reddy, A. R.; Fridman-Marueli, G.; Bendikov, M. J. Org. Chem. 2007, 72, 51.

125

APPENDIX A

LIST OF ABBREVIATIONS: tert Tertiary ISC Intersystem Crossing fs Femtosecond ps Picosecond ns Nanosecond μs Microsecond s or sec Second min Minute h or hr. Hour nm Nanometer M Milli mL Milliliter L Liter mm Millimeter cm Centimeter g Gram μ Micro MHz MegaHertz W Watt Ȧ Angstrom v or V Volume kcal Kilo Calorie Conc. Concentration mol Mole mmol Millimol M Moles per liter Mp or mp Melting Point oC Degree Celsius K Kelvin D Debye Wt Weight ORTEP Oak ridge thermal ellipsoid plot TLC Thin Layer Chromatography GC Gas Chromatography MS Mass Spectrometry NMR Nuclear Magnetic Resonance s (for NMR) Singlet d (for NMR) Doublet m (for NMR) Multiplet UV or Uv Ultra Violet IR Infra Red

126 hν Photon THF Tetrahydrofuran DCM Dichloromethane NMO 4-Methylmorpholin-N-oxide DMSO Dimethyl sulfoxide DMF Dimethyl formamide Ph Phenyl Ar Aryl Me Methyl LFP Laser Flash Photolysis T Triplet S Singlet

127

APPENDIX B

Figure B1. Molecular packing in the crystal structure of 3.1.

Figure B2. MALDI-MS of hexacene (left) and heptacene (right) from the thin films of PMMA.

128

Figure B3. MALDI mass spectrum recorded Figure B4. MALDI mass spectrum recorded after irradiating 5.3 for 40 min in degassed after irradiating 5.3 for 40 min in oxygen toluene. saturated CDCl3.

Figure B5. Normalized Absorption, fluorescence, and phosphorescene spectra of 6.2 (left) and 6.4. (absorption and fluorescence spectra recorded in toluene at room temperature and phosphorescence spectra recorded in methanol/ethanol (1:4) matrix at 77 K).

129

Figure B6. 1H NMR spectra of 3.1

Figure B7. 1H NMR spectra of 3.2

1 Figure B8. H NMR spectra of 3.1

130

Figure B9. 1H-NMR spectra of compound 5.3.

Figure B10. 13C-NMR spectra of compound 5.3.

131

Figure B11. 1H-NMR spectrum of 5.12.

Figure B12. 13C-NMR spectrum of 5.12.

132

APPENDIX C

0.006

0.004 A Δ 0.002

0.000

020406080 Time (μs) Figure C1. Nanosecond flash photolysis of 6.4 in toluene, argon-saturated (λex = 460 nm): kinetic traces monitored at 500 nm (τ= 13.34 +/-0.27 μs) showing the lifetime of the triplet state of heptacene.

3 μs 0.0010 10 μs 50 μs

0.0005 A

Δ 0.0000

-0.0005

-0.0010 500 600 700 Wavelength (nm)

Figure C2. Absorption spectra obtained from the nanosecond laser flash photolyis of 6.4 in toluene (oxygen saturated), recorded (a) 3 μs (black), (b) 10 μs (red), and (c) 50 μs (green) after the laser pulse (λex = 460 nm); the absorption in the 650-800 region is that of heptacene.

133

Figure C3. Absorption difference spectra in the visible region of diketone 6.2 in toluene acquired following a 130 fs excitation pulse at 475 nm (pulse energy = 5 μJ). Inset: kinetic trace monitored at 550 nm.

0.006

0.005 τ 2= 29.4 ± 3.1 ps 0.004

0.003 A Δ 0.002

0.001

0.000

0 50 100 150 200 250 300 Time (ps) Figure C4. Absorption difference spectra in the visible region of diketone 6.3 in toluene acquired following a 130 fs excitation pulse at 475 nm (pulse energy = 5 μJ) (left). Kinetic traces monitored at 620 nm (right).

134

Figure C5. Absorption difference spectra in the visible region of diketone 6.2 in toluene acquired following a 130 fs excitation pulse at 400 nm (pulse energy = 5 μJ). Inset: kinetic trace monitored at 550 nm.

Figure C6. Absorption difference spectra in the visible region of diketone 6.3 in toluene acquired following a 130 fs excitation pulse at 400 nm (pulse energy = 5 μJ). Inset: kinetic trace monitored at 620 nm.

135

Figure C7. Absorption difference spectra in the visible region of diketone 6.4 in toluene acquired following a 130 fs excitation pulse at 400 nm (pulse energy = 5 μJ). Inset: kinetic trace monitored at 620 nm.