Functionalized Π-Extended Porphyrins

MIAMI UNIVERSITY The Graduate School

Certificate for Approving the Dissertation

We hereby approve the Dissertation of Lin Jiang

Candidate for the Degree: Doctor of Philosophy

______

Hong Wang, Advisor

______

Richard Taylor, Committee Chair

______

Michael Novak, Reader

______

Shouzhong Zou, Reader

______

Elisabeth Widom, Graduate School Representative ABSTRACT

FUNCTIONALIZED -EXTENDED PORPHYRINS

by Lin Jiang

-Extended porphyrins are of broad interest due to their unique combination of photophysical, optoelectronic and physicochemical properties, and their potential applications in many areas. Chapter one gives general introduction to porphyrins. The description of available synthetic methods for -extended porphyrins is also included. In chapter two, the development of a novel synthetic methodology to synthesize opp-dibenzoporphyrins is described. A number of porphyrins with a variety of functional groups on the fused benzene rings were synthesized using this method These porphyrins were evaluated as light harvesters in dye-sensitized solar-cells (DSSCs), displaying moderate solar to electricity conversion efficiencies. Our study shows that incorporation of conjugated carboxylic acid linkers on the porphyrin to further extend the conjugate system considerably broadened and red-shifted the absorption bands of the porphyrin leading to higher conversion efficiency. Chapter three is focused on the synthesis and characterization of the first examples of triphenylene-fused porphyrins through a Pd-catalyzed cascade reaction followed by oxidative ring closure. Further extension of the porphyrin π-system to fuse one dibenzo-[fg,op]-tetracene has also been achieved using a similar approach. These π-extended porphyrins displayed much broadened and bathochromic shifted UV-vis absorptions, as compared with their unfused precursors. In chapter four, the synthesis and characterization of highly water soluble ionic tetrabenzoporphyrins (TBPs) are presented. The UV-Vis spectra of these porphyrins displayed significantly red-shifted and broadened Soret bands. Unusually intense Q bands were observed for cationic TBPs. These interesting electronic properties have never been observed in other water soluble porphyrins. These porphyrins are

substituted with eight ionic groups which are highly water soluble, and serve as potential PDT candidates. Chapter five introduces a synthetic method to prepare the pentacene-fused diporphyrin and quinone-fused dinaphtho[2,3]porphyrins. The pentacene-fused diporphyrin is remarkably more stable than the relevant pentacene and heptacene derivatives, indicating that these novel materials hold great potential for the applications in various areas. Unsymmetrical cross-conjugated quinone-fused dinaphtho[2,3]-porphyrins and triporphyrin have also been obtained using a similar approach. All these diporphyrins and triporphyrin displayed remarkably perturbed broad absorption bands spreading over the whole visible region of the spectrum, suggesting strong electronic interactions among these multichromophoric systems.

FUNCTIONALIZED Π-EXTENDED PORPHYRINS

A DISSERTATION

Submitted to the Faculty of Miami University in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Chemistry and Biochemistry

Lin Jiang Miami University Oxford, OH 2013

Dissertation Director: Hong Wang

Table of Contents Chapter 1: Introduction ...... 1 1.1 Introduction to Porphyrins ...... 1 1.1.1 General Structure of Porphyrins ...... 1 1.1.2 Nomenclature of Porphyrins ...... 2 1.1.3 General Synthetic Methods of Porphyrins ...... 3 1.1.4 Electronic Properties of Porphyrins ...... 4 1.2 Introduction to -Extended Porphyrins ...... 5 1.2.1 Template Condensation Method ...... 6 1.2.2 Retro-Diels-Alder Method ...... 7 1.2.3 Oxidative Aromatization Method ...... 9 1.2.4 Olefin Ring-Closure Metathesis Method ...... 12 1.2.5 Thermal Electrocyclization Method ...... 13 Chapter 2: A Concise Approach to the Synthesis of opp-Dibenzoporphyrins and Application towards DSSCs ...... 17 2.1 Dyes for Dye-Senstized Solar Cells (DSSCs) ...... 18 2.2 Porphyrins as Light Harvester for DSSCs ...... 18 2.3 Results and Discussion ...... 21 2.3.1 Synthesis ...... 21 2.3.2 Optical and Electrochemical Properties ...... 22 2.3.3 DFT Calculations ...... 24 2.3.4 Cyclic Voltammetry ...... 27

2.3.5 Photovoltaic Properties of Porphyrin-Sensitized TiO2 Solar Cells ...... 29 2.4 Conclusion ...... 32 2.5 Experimental Section ...... 32 Chapter 3：Triphenylene-Fused Porphyrins ...... 42 3.1 Introduction ...... 43 3.2 Result and Discussion ...... 44 3.3 Conclusion ...... 51 3.4 Experimental Part ...... 53 Chapter 4: Water-Soluble Ionic Benzoporphyrins ...... 106 4.1 Introduction to Ionic Water Soluble Porphyrins ...... 107

4.2 Synthesis of Starting TBPs ...... 108 4.3 UV-Vis Absorption and Fluorescence Spectroscopy ...... 112 4.4 Structure Optimization Calculations ...... 114 4.5 Application of Water Soluble Porphyrins towards Photodynamic Therapy (PDT) ...... 118 4.6 Synthesis of Water Soluble TBPs ...... 120 4.7 Characterization of Water Soluble TBPs ...... 121 4.7.1 Mass Spectrometry ...... 121 4.7.2 1H NMR Characterization ...... 122 4.7.3 UV-Vis Absorption and Fluorescence Spectroscopes ...... 123 4.8 Conclusion ...... 124 4.9 Experimental Part ...... 127 4.10 Beer’s Law Experiment ...... 142 Chapter 5: Pentacene-Fused Diporphyrins ...... 159 5.1 Introduction ...... 159 5.2 Results and Discussion of Symmetrical Pentacene-Fused Diporphyrins ...... 161 5.2.1 Synthesis of Diformylbenzoporphyrin 3a-c ...... 161 5.2.2 Synthesis of Pentacene-Fused Diporphyrins ...... 164 5.2.3 Absorption and Emission Spectra ...... 166 5.2.4 Stability Study of Pentacene-Fused Diporphyrin 1 ...... 168

5.2.5 Investigation of Direct Reduction of 2a with LiAlH4 ...... 168 5.3 Results and Discussion of Unsymmetrical Diporphyrins and Triporphyrins ...... 170 5.3.1 Synthesis of Unsymmetrical Diporphyrins ...... 170 5.3.2 Synthesis of Triporphyrins ...... 173 5.4 Experimental Section ...... 177

iii

List of Tables Table 2.1. Calculated molecular orbital energy levels of 5a-5e...... 26

Table 2.2. The oxidation-reduction potentials (V) of porphyrins...... 28

Table 2.3. Concentration study of porphyrin 5b...... 30

Table 2.4. Photovoltaic performance of DSSCs based on the opp-dibenzoporphyrins (5a-5e) ...... 31

List of Figures Figure 1.1 Structure of porphine, the simplest porphyrin ...... 1

Figure 1.2 Tautomerism in tetraarylporphyrins ...... 2

Figure 1.3 Fischer’s numbering system ...... 3

Figure 1.4 IUPAC nomenclature of porphyrin macrocycle ...... 3

Figure 1.5 UV-Visible spectrum of TPP...... 5

Figure 1.6 Two selective examples of -extended porphyrins ...... 6

Figure 2.1 Different types of porphyrin dyes studied for DSSCs ...... 21

Figure 2.3 Steady-state fluorescence spectra of 5a-5e in methanol...... 23

Figure 2.4 Molecular geometry of 5a-5e calculated at the B3LYP/6-31G(d) level of theory...... 25

Figure 2.5 Molecular orbitals of 5a-5e calculated at the B3LYP/6-31G(d) level of theory...... 26

Figure 2.6 Cyclic voltammograms of porphyrins in DMF containing 0.1M TBAP...... 28

Figure 2.7 UV-vis absorption data of 5a and 5c. Before: before adsorption on TiO2 surface; after: after adsorption on TiO2 surface...... 31

Figure 3.1 X-ray crystal structure of 5a: top, edge view; bottom, top view...... 46

Figure 3.2 UV-Vis absorption spectra of 4e, 5e, 5c, 9 and 13 in CH2Cl2...... 50

Figure 3.3 Optimized molecular structures of 5e, 5c, 9 and 13 (DFT B3LYP-631Gdp)...... 51

Figure 4.1 Two representative cationic (left) and anionic (right) water soluble porphyrins...... 107

Figure 4.2 X-ray crystal structure of 4a with 35% thermal ellipsoids: left, edge view; right, top view...... 112

Figure 4.3 (a) Normalized absorption spectra of porphyrins 4a-e in CH2Cl2 solution; (b) normalized absorption spectra of porphyrins 4a-e upon treatment with TFA in CH2Cl2 solution; (c) fluorescent spectrum of 4d ...... 114

Figure 4.4 Optimized molecular structure of 4a-4d, protonated 4a and 2a...... 116

Figure 4.5 The calculated HOMOs and LUMOs of 4a-4d and protonated 4a ...... 117

Figure 4.6 Structures of the porphyrin based photosensitisers used for clinical or preclinical PDT

...... 119

Figure 4.7 LDI-TOF mass spectrum of 7c...... 122

1 Figure 4.8 H NMR in d4-methanol of compound 7c...... 123

Figure 4.9 (a) Normalized absorption spectra of porphyrins 4c-d and 7c-d (4c and 4d were tested in CH2Cl2 solution; 7c and 7d were tested in methanol solution); (b) Fluorescent spectra of 7c and

7d in methanol...... 124

Figure 5.1 The synthesis of diformylbenzoporphyrin from different pathways...... 164

Figure 5.2 Synthesis of pentacene-fused diporphyrin 1...... 165

Figure 5.3 X-ray crystal structure of 2a: top, edge view; bottom, top view...... 165

Figure 5.4 (a) Normalized UV-vis absorption spectra of porphyrins 2a-c and 1 in CH2Cl2 solution; inset, fluorescent spectrum of 2b (excitation wavelength: 428 nm). (b) UV-vis absorption spectra of porphyrin 1 in benzene recorded at different times...... 168

Figure 5.5 Reduction of 2a...... 169

Figure 5.6 Investigation of the aldol reaction of 3a...... 171

Figure 5.7 Synthesis of two unsymmetrical diporphyrins 15a and 15b...... 173

Figure 5.8 UV-visible spectra of different cross-conjugated diporphyrins (2a, 15a and 15b) in toluene...... 173

Figure 5.9 Fluorescence spectra of 2b and 15b...... 173

Figure 5.10 Synthesis of triporphyrin 17...... 174

Figure 5.11 UV-visible spectra of 2a and 17 in toluene...... 175

List of Schemes Scheme 1.2 Synthesis of TBP through template condensation method ...... 7

Scheme 1.3 First synthesis of TBPs through a retro-Diels-Alder method ...... 8

Scheme 1.4 Lim’s approach to synthesize the tetraphenylporphyrin precursor...... 8

Scheme 1.5 Synthesis of TBPs from oxidative aromatization approach ...... 9

Scheme 1.6 Synthesis of non-benzo-substituted TBPs through the oxidative aromatization method

...... 10

Scheme 1.7 Two pathways to synthesize the meso-disubstituted TBPs ...... 11

Scheme 1.9 Olefin ring-closure metathesis method to synthesize opp-dibenzoporphyrin ...... 13

Scheme 1.10 Synthesis of opp-dibenzoporphyrin through thermal electrocyclizaiton method ...... 14

Scheme 2.1 Synthesis of opp-dibenzoporphyrins ...... 22

Scheme 3.1 Synthesis of opp-dibenzoporphyrin via Pd0-catalyzed cascade reaction ...... 44

Scheme 3.2 Synthesis of mono triphenylene-fused porphyrins ...... 45

Scheme 3.3 Synthesis of bis-triphenylene-fused porphyrins ...... 48

Scheme 3.4 Synthesis of dibenzo[fg,op]tetracene-fused porphyrin ...... 49

Scheme 4.1 The synthesis of ionic monobenzoporphyrins via a cascade reaction ...... 109

Scheme 4.2 The synthesis of ionic tetrabenzoporphyrins 4a-e, 6, and cationic opp-dibenzoporphyrin 5...... 111

Scheme 4.3 Synthesis of novel water soluble TBPs ...... 121

vii

Abbreviation Table I. Reagents and Solvents

TPP Tetraphenyl porphyrin TBPs Tetrabenzoporphyrins DBU 1,8-Diazabicycloundec-7-ene DDQ 2,3-Dichloro-5,6-dicyano-benzoquinone DCM Dichloromethane TFA Trifluoroacetic acid DMM Dimethyl malonate THF Tetrahedrofuran DCB Dichlorobenzene MeOH Methanol

CHCl3 Chloroform NBS N-Bromosuccinimide DMF N, N-Dimethylformamide

Ni(OAc)2 Nickel acetate

Zn(OAc)2 Zinc acetate

Cu(OAc)2 Copper acetate

K2CO3 Potassium carbonate KOH Potassium hydroxide

Na2CO3 Sodium carbonate

H2SO4 Sulfuric acid HCl Hydrochloric acid

NaBH4 Sodium borohydride t-BuOK Potassium tert-butoxide MCPBA meta-Chloroperoxybenzoic acid

Et3N Triethylamine

viii

TsOH p-Toluenesulfonic acid or tosylic acid HOAc Acetic acid

BF3•Et2O Boron trifluoride diethyl etherate DMP Dess–Martin periodinane MeI Methyl iodide

LiAlH4 Lithium aluminium hydride

Pd(OAc)2 Palladium acetate

PPh3 Triphenylphosphine DIBAL-H Diisobutylaluminium hydride

Pd[P(t-Bu)3]2 Bis(tri-tert-butylphosphine)palladium(0)

SnCl2 Tin(II) chloride

II. Analytical Procedures

NMR Nuclear magnetic resonance TLC Thin layer chromatography UV-visible Ultraviolet visible spectroscopy MALDI-TOF Matrix-assisted laser desorption ionization - time of fly mass spectrometry IR Infrared spectroscopy FL Fluorescence spectroscopy ESI-MS Electrospray ionization - mass spectrometry CV Cyclic voltammetry DFT Density functional theory

III. NMR Data Interpretation s Singlet d Doublet

ix m Multiplet t Triplet br Broad dd Doublet of doublet TMS Trimethylsilane ppm Part per million

IV. Others

DSSC Dye sensitized solar cell PDT Photodynamic therapy OLED Organic light-emitting diode QDs Quantum dots HOMO Highest occupied molecular orbital LUMO Lowest unoccupied molecular orbital MO Molecular orbital PAHs Polycyclic aromatic hydrocarbons CT Charge transfer Equiv. Equivalent Conc. Concentration R.T. Room temperature OFETs Organic field-effect transistors IUPAC The International Union of Pure and Applied Chemistry

Acknowledgement I would like to express my sincere gratitude to my supervisor, Dr. Hong Wang, for her guidance, encouragement, and support all through my Ph.D study. All the opportunities she has provided and her confidence in my ability played an important role for me to complete this dissertation and changed me from an engineer to an organic synthetic chemist. I feel I have learned a lot and I am thankful that I had the opportunity to be part of Wang’s group in the past five years.

I would like to thank Dr. Hong Wang, Dr. Richard Taylor, Dr. Michael Novak, Dr. Shouzhong Zou and Dr. Elisabeth Widom for their time being on my committee.

I would like to express my appreciation to all the members of Dr. Wang’s research group including Dr. Zhenghu Xu, Mr. Philias Daka (soon to be Dr. Philias Daka), Mr. Rohit Deshpande, Mr. Yongming Deng (soon to be Dr. Yongming Deng), Dr. Lu Liu, Mr. Alex Matus, Ms. Laura Sirk, Mr. Ross Zaenglein, Mr. Ryan Sarkisian and Ms. Erika Csatary for their help and friendship. You made my stay here interesting, informative, and enjoyable. I would like to say “thank you” to everyone who helped and encouraged me over the five years in Oxford.

I would also like to thank Miami University and the Ohio Board of Regent for providing financial support for the past five years.

My deepest gratitude goes to my parents, Mr. Zhenmin Jiang and Ms. Dezhen Lin, for their enormous love, support and encouragement throughout my life.

My last, but not least, appreciation goes to my husband, Mr. Bo Wang (soon to be Dr. Bo Wang). Without his unconditional support and love, and continuously giving me strength and encouragement, my Doctoral degree was not possible.

Chapter 1: Introduction

1.1 Introduction to Porphyrins

Porphyrins are a group of heterocyclic macrocycles consisting of four modified pyrrole subunits linked via methine bridges (=CH-). They have played very important roles in nature. One of the most famous examples is hemoglobin, the protein in red blood cells that carry oxygen in the blood. Another well-known porphyrin in leaves is a green pigment called chlorophyll. The name porphyrin comes from a Greek word for purple, due to their intense absorption bands in the visible region and deeply colored nature. Besides their natural role, porphyrins are currently employed as platforms for study of theoretical principles and applications in a wide variety of fields including chemistry, biology, medicine, geology, material science and so on. Herein, I would like to focus on introducing the structure and nomenclature of porphyrins, the general methods to synthesize them, as well as their basic electronic properties.

1.1.1 General Structure of Porphyrins

Figure 1.1 Structure of porphine, the simplest porphyrin

Unsubstituted porphyrin, which is also known as the parent porphyrin, named

1 porphine. (Figure 1.1) It is a heterocyclic macrocycle containing twenty carbon atoms and four nitrogen atoms. There are two delocalization pathways in porphyrins as shown in Figure 1.2.1 They obey Hückel's rule, possessing 4n+2 π electrons (n=4 for the shortest cyclic path in dashed line) delocalized over the macrocycle. Thus porphyrin macrocycles are highly aromatic conjugated systems.

Figure 1.2 Tautomerism in tetraarylporphyrins

1.1.2 Nomenclature of Porphyrins

Hans Fisher was the first to give the porphyrin macrocycle a systematic numbering system. In Fischer’s numbering system (Figure 1.3), carbon atoms connected to the nitrogen atoms on the pyrrolic rings are named as -carbons with no Arabic numerals. Carbon atoms next to the -carbons on the pyrrolic rings which are open to substitutions are called  carbons and numbered from 1 to 8 consecutively. The carbon atoms on the methine bridges are called meso-carbons and numbered by Greek lower case letter, - . The four pyrrolic rings are labeled as A, B, C and D. (Figure 1.3)

Figure 1.3 Fischer’s numbering system

Fischer’s numbering system is straightforward, but it is only suitable for simple porphyrins. A more systematic IUPAC nomenclature was introduced in 1979 in order to name more complicated porphyrin molecules. In the IUPAC nomenclature, the parent macrocycle is called porphyrin. All the carbon and nitrogen atoms in the porphyrin macrocycle are numbered as demonstrated in Figure 1.4.

Figure 1.4 IUPAC nomenclature of porphyrin macrocycle

1.1.3 General Synthetic Methods of Porphyrins

A number of different synthetic approaches to constructing porphyrins have been developed over the years. One of the most commonly used synthetic methods is through condensation of a pyrrole and an aldehyde followed by oxidation of the

3 intermediate porphyrinogen. This method was initially developed by Paul Rothemund.2 (Scheme 1.1) This approach is straightforward and more suitable for larger scales. Tetraphenylporphyrin (TPP) is the simplest porphyrin synthesized via this method. (Scheme 1.1)

Scheme 1.1 Synthetic pathway of TPP

1.1.4 Electronic Properties of Porphyrins

Porphyrins typically have very intense absorption bands in the visible region with deep color which is the most prominent and diagnostic feature for identifying various porphyrins. For a simple porphyrin, there are two distinct regions in its UV-Visible absorption spectrum. The intense and sharp absorption band found in ultraviolet to purple region (390-425 nm) is called the Soret band or B band. This Soret band is the most intense band and is characteristic of porphyrins. The molar absorption coefficient () is approximately 4 x 105 M-1cm-1. In the visible region, normally between 480 nm to 700 nm, there are several weaker and broadened bands called Q bands. The molar absorption coeffeicients () of the Q bands are approximately 103 to 104 M-1cm-1 (Figure 1.5). The number of the Q bands varies due to the metallation or substitution of the porphyrins. For example, four weaker Q bands are often observed in a free base porphyrin (no metal inserted in the porphyrin inside core). For a metallated porphyrin, the number of the Q bands largely depends on the central metal. For example, for a metalloporphyrin having a square planar coordination sphere, two

Q bands will be detected. Further coordination with extra ligands normally leads to red shift of the absorption bands. On the other hand, electrophilic groups directly attached at the  positions of the porphyrin periphery can lead to a red shifted Soret band and Q bands. These phenomena will be discussed in detail in the next chapter.

Figure 1.5 UV-Visible spectrum of TPP

1.2 Introduction to -Extended Porphyrins

Porphyrins with externally conjugated -bonding substituent(s) or with fused aromatic rings on the porphyrin periphery are commonly known as -extended porphyrins(Figure 1.6). Recently, -extended porphyrins, in which aromatic rings are fused through the , ’–positions, are attractive synthetic targets due to their unique combination of photophysical and optoelectronic properties.3 The UV-Visible absorption spectra of largely -extended porphyrins can be significantly red-shifted and broadened up to 450-550nm for the Soret band and enhanced Q bands are also often observed. These features make them attractive in many areas, such as dye-sensititized solar cells (DSSCs)4, photodynamic therapy (PDT),5 organic light-emitting diodes (OLEDs)6, etc. Our interest in benzoporphyrins lies in their

5 ability to shift the excitation spectra deep into the red region Porphyrins with absorptions in the red to infrared regions promise practical applications in various areas.

Figure 1.6 Two selective examples of -extended porphyrins

Interest in -extended porphyrins has been dramatically increased over the years; however, tetrabenzoporphyrins (TBPs) have remained the major focus of the research in the past decade. The reason is that, compared to classical porphyrins, the organic synthesis of -extended porphyrins is much more difficult. In particular, functionalized tetrabenzoporphyrins have been little explored. Another big problem is the low solubility of these porphyrins arising from π−π stacking effect. Introduction of bulky meso aryl rings can help to increase the solubility mildly. In the following section, I would like to introduce the major synthetic pathways for TBPs.

1.2.1 Template Condensation Method

TBPs were first synthesized by Helberger et al. in 1938,7 and later in 1940–1950 by Linstead et al.8 These methods were developed by the same authors who developed the synthetic methods of phthalocyanines.9 (Scheme 1.2)

Scheme 1.2 Synthesis of TBP through template condensation method There are several major drawbacks in this original method. First, the starting isoindole derivatives are not so stable making the preparation and functionalization step very difficult. Second, extremely harsh conditions such as heating at 375oC have to be used in this method which results in low yields and large number of side products. Third, the formed TBPs have very poor solubility in common organic solvents, leading to tedious laborious purifications. A number of modifications such as changing starting materials to phthalimides and phthalimidines were processed to improve the reaction conditions, unfortunately, they suffered similar disadvantages.

1.2.2 Retro-Diels-Alder Method

Unsubstituted isoindoles are well-known unstable and always involve tremendous problems in synthesizing TBPs. A new type of synthetic approach has been developed to preparing masked isoindole moiety in order to enhance their stability. Ono and co-workers 10 introduced a retro-Diels–Alder reaction to give a relatively stable isoindole in essentially quantitative yield, which can then be applied to the synthesis of TBPs. (Scheme 1.3)

Scheme 1.3 First synthesis of TBPs through a retro-Diels-Alder method

In this method, the authors introduced this masked isoindole 3 as a new building block for the construction of the corresponding porphyrin 5. Compound 1 was prepared through a Diels-Alder reaction of cyclohexadiene and (E)-((2-nitrovinyl)sulfonyl)benzene. Then through the Barton-Zard pyrrole synthesis process, with the treatment of 1 with ethyl isocyanoacetate in the presence of DBU, the reaction proceeded via elimination of sulfinic acid, addition of ethyl isocyanoacetate, cyclization and elimination of nitrous acid to give the desired pyrrole 2. Deethoxycarbonylation in the presence of potassium hydroxide and ethane-1,2-diol gave the ideal precursor 3. Reaction of this masked isoindole 3 with aromatic aldehydes followed by oxidation gave the corresponding porphyrin 4 in 20–30% yield. After the thermal retro-Diels–Alder extrusion of ethylene by heating at 200 oC, the tetraphenylporphyrin 5 was successfully synthesized. After the first synthesis, a lot of effort has been made to functionalize different meso-substituted -extended porphyrins.11 Recently, Lim and co-worker discovered a new approach to synthesizing the TBP precursor, which is more efficient than Ono’s first one.12 (Scheme 1.4)

Scheme 1.4 Lim’s approach to synthesize the tetraphenylporphyrin precursor.

Compared with the original method, this approach showed obvious advantages in synthesizing polyfunctionalized TBPs. Nevertheless, the synthesis of bicyclic fused pyrroles is still very complicated and always involves problems related with price, availability, hazardous compounds such as -sulfonylnitroethylene and the lack of versatility of the building blocks. Another drawback of this method is the harsh conditions required for the aromatization step which can lead to the loss of required porphyrins.

1.2.3 Oxidative Aromatization Method

In 2001, a new method based on an oxidative aromatization using DDQ was put forward by Cheprakov and co-workers as an alternative way to synthesize -extended porphyrins.13 This method allows for the introduction of substituents on both benzo- and meso-phenyl-rings and employs readily available, inexpensive starting materials. (Scheme 1.5)

Scheme 1.5 Synthesis of TBPs from oxidative aromatization approach In this method, inexpensive sulfolene 1 was chosen as the key starting material.

Then 1 was converted to 3-phenylsulfonyl-3-sulfolene 2 in 86% yield. Diels-Alder adduct 3 was obtained from 2 and dimethyl maleate (DMM). Sulfone 3 reacted with tert-butyl isocyanoacetate yielding cyclohexanopyrrole 4 in 85% yield, which was then deprotected and decarboxylated by TFA in DCM in 35% yield under argon. The pyrrole derivative 5 was introduced into the Lindsey condensation conditions,14 which produced porphyrins 6a–c in 25–35% yields. The formed porphyrins 6a–c were finally converted to their metal complexes and aromatized by refluxing in the presence of excess DDQ giving MAr4TBPs (M = Ni, Cu, Zn) 7a–c in high yields ranging from 90% to 95%. It is worth mentioning that free-base porphyrins could not be aromatized under such conditions, most likely due to the formation of dications under these conditions, which are apparently not oxidized by DDQ. This synthesis can offer gram quantities of substituted Ar4TBPs in a single preparation. Later on, the non-benzo-substituted derivatives was also synthesized through this method.15 (Scheme 1.6) However, those types of TBPs suffered low solubility, thus subsequent studies of these porphyrins were not carried out.

Scheme 1.6 Synthesis of non-benzo-substituted TBPs through the oxidative aromatization method Moreover, meso-disubstituted TBPs have been synthesized based on [2 + 2] condensation of dipyrromethanes followed by oxidative aromatization.16 Two pathways were investigated in this paper: the tetrahydroisoindole pathway and the dihydroisoindole pathway. In the tetrahydroisoindole pathway, precursor 5, 15-diaryltetracyclohexenoporphyrins 4a-e was assembled from cyclohexeno-fused

10 meso-unsubstituted dipyrromethanes 3 and aromatic aldehydes. In this case, the authors found that aromatization by tetracyclone was more effective than DDQ but failed in the cases of porphyrins with electron-withdrawing substituents on the meso-aryl rings such as 5b and 5d. The dihydroisoindole pathway was found to be much better than the first one, and the aromatization step was much more efficient than the first pathway. In this case, meso-aryl porphyrins 10a-e containing both electron-donating and electron-withdrawing substituents were successfully synthesized. (Scheme 1.7)

Scheme 1.7 Two pathways to synthesize the meso-disubstituted TBPs

Recently, in order to overcome the existing drawbacks associated with the dehydrogenation problem and low yield, Krautler and co-workers reported a one-pot

11 cycloaddition-aromatization approach to synthesize the “black” TBPs through the direct cycloaddition of sulfoleneporphyrins and benzoquinone.17 (Scheme 1.8)

Scheme 1.8 One-pot cycloaddition-aromatization approach

To conclude, the cycloaddition-aromatization method offers a relatively cost-effective way, and is a versatile strategy to prepare different functionalized TBPs. However, the major drawback of this method is the aromatization step. This step involves a total loss of sixteen electrons at one time which could be a very difficult task. In addition, this procedure requires lengthy laborious purifications and the overall yields are very low. Larger scale synthesis has turned out to be difficult.

1.2.4 Olefin Ring-Closure Metathesis Method

Smith and co-workers introduced an olefin ring-closure metathesis method to synthesize mono-, di-, and tri-benzoporphyrins in 2006.18 For example, in the synthesis of dibenzoporphyrin 5, the method started with Suzuki reaction of readily available vicinal tetrabromoporphyrins 2; cyclization of the Suzuki product by way of olefin metathesis using the Grubb’s second generation catalyst and subsequent oxidation with DDQ led to the formation of 5. (Scheme 1.9)

Scheme 1.9 Olefin ring-closure metathesis method to synthesize opp-dibenzoporphyrin Different from all the methods mentioned above, this method was carried out using readily available meso-aryl porphyrins to avoid total synthetic routes starting from unstable isoindole or functionalized pyrrole. At the same time, this approach overcame problems of the regioselective synthesis of mono-, di- and tribenzoporphyrins. However, direct functionalization on the fused benzene rings still remains a challenge.

1.2.5 Thermal Electrocyclization Method

In 2008, Cavaleiro and co-workers introduced a thermal electrocyclization method which is another approach starting from relatively easy accessible tetraarylporphyrins.19 Take opp-dibenzoporphyrin 4 as an example, β, β’-diformylporphyrin 2 served as the key precursor, which can be converted into dibutadienylporphyrin 3 through a Wittig reaction. Further conversion of 3 into opp-dibenzoporphyrin 4 can be achieved through thermal electrocyclization reactions. (Scheme 1.10)

Scheme 1.10 Synthesis of opp-dibenzoporphyrin through thermal electrocyclizaiton method One of the major advantages of this method is that mono- and dibutadienylporphyrins can be prepared conveniently from the corresponding β-formyl derivatives. Thus monobenzo-, adjacent dibenzo-, and opp- dibenzoporphyrins can be selectively synthesized. It is worth mentioning that monobutadienylporphyrin can be converted to the corresponding benzoporphyrin either through Diels–Alder reaction in moderate yields (68%) or through the thermal electrocylcization reaction in good yield (78%). However, the formylation reaction of 1 gives a complex mixture which requires laborious separation and purification. Another drawback is the lack of flexibility for further functionalization of the fused benzene rings. To conclude, -extended porphyrins have broad potential applications in both natural and applied science. These porphyrins tend to have strong -stacking interactions due to the extended -system, which is attractive in electronic devices and material science. In the past decade, several improvements on synthetic methods have been achieved. Despite that, obtaining -extended porphyrins through more efficient procedures is still a big challenge. Lengthy synthetic steps that involved pyrrole chemistry and marked lack of functional groups on the porphyrins periphery, especially at the β-positions are the two major drawbacks for most of the methods discussed above. In order to make

14 breakthroughs in the synthesis of -extended porphyrins, we decided to do the direct functionalization of the porphyrin periphery starting from inexpensive and relatively available starting materials, such as meso-tetraarylporphyrins that will eventually reduce the number of steps required to get the desired compounds. Thus new methodology development to fuse one or more aromatic rings on porphyrin periphery is the main objective in this research. Based on this idea, we developed a concise and versatile method based on a Pd0 catalyzed cascade reaction to prepare functionalized benzoporphyrins. Further investigation of this novel methodology allowed us to synthesize the first examples of triphenylenoporphyrins, new class of water soluble ionic tetrabenzoporphyrins and novel pentacene-fused porphyrin oligomers. The realization of these methods solves the long lasting lengthy synthetic problems and opens a door and makes it possible to conveniently introduce a wide variety of functional groups to the fused aromatic rings. At the same time, applications of these novel -extended porphyrins were explored in various areas ranging from molecular electronics to biomedicine.

Reference: 1. Crossley, M. J.; Harding, M. M.; Sternhell, S., Journal of the American Chemical Society 1986, 108 (13), 3608-3613. 2. Rothemund, P., J. Am. Chem. Soc. 1936, 58 (4), 625-627; Rothemund, P., Formation of Porphyrins from Pyrrole and Aldehydes. Undation, C. F. K., Ed. J. Am. Chem. Soc., 1935. 3. Baluschev, S.; Yakutkin, V.; Miteva, T.; Avlasevich, Y.; Chernov, S.; Aleshchenkov, S.; Nelles, G.; Cheprakov, A.; Yasuda, A.; Mullen, K.; Wegner, G., Angewandte Chemie-International Edition 2007, 46 (40), 7693-7696; Borek, C.; Hanson, K.; Djurovich, P.; Thompson, M.; Aznavour, K.; Bau, R.; Sun, Y.; Forrest, S.; Brooks, J.; Michalski, L.; Brown, J., Angewandte Chemie-International Edition 2007, 46 (7), 1109-1112; Ongayi, O.; Gottumukkala, V.; Fronczek, F.; Vicente, M., Bioorganic & Medicinal Chemistry Letters 2005, 15 (6), 1665-1668; Mack, J.; Bunya, M.; Shimizu, Y.; Uoyama, H.; Komobuchi, N.; Okujima, T.; Uno, H.; Ito, S.; Stillman, M.; Ono, N.; Kobayashi, N., Chemistry-a European Journal 2008, 14 (16), 5001-5020. 4. Eu, S.; Hayashi, S.; Urneyama, T.; Matano, Y.; Araki, Y.; Imahori, H., Journal of Physical Chemistry C 2008, 112 (11), 4396-4405; Kira, A.; Matsubara, Y.; Iijima, H.; Umeyama, T.; Matano, Y.; Ito, S.; Niemi, M.; Tkachenko, N.; Lemmetyinen, H.; Imahori, H., Journal of Physical Chemistry C

2010, 114 (25), 11293-11304. 5. Bonnett, R., Chemical Society Reviews 1995, 24 (1), 19-33. 6. Perez-Morales, M.; de Miguel, G.; Bolink, H.; Martin-Romero, M.; Camacho, L., Journal of Materials Chemistry 2009, 19 (24), 4255-4260. 7. Helberger, J. H.; Reday, A. V.; Hever, D. B., Justus Liebigs Ann. Chem. 1938, 533, 197. 8. Barrett, P. A.; Linstead, R. P.; Rundall, F. G.; Tuey, G. A. P., J. Chem. Soc. 1940, 1079; Linstead, R. P.; Weiss, F. T. J., Chem. Soc. Rev. 1950, 2975. 9. Carvalho, C. M. B.; Brocksom, T. J.; de Oliveira, K. T., Chemical Society Reviews 2013, 42 (8), 3302-3317. 10. Ito, S.; Murashima, T.; Uno, H.; Ono, N., Chemical Communications 1998, (16), 1661-1662; Ito, S.; Watanabe, H.; Uno, H.; Murashima, T.; Ono, N.; Tsai, Y.; Compton, R., Tetrahedron Letters 2001, 42 (4), 707-710. 11. Finikova, O.; Chernov, S.; Cheprakov, A.; Filatov, M.; Vinogradov, S.; Beletskaya, I., Doklady Chemistry 2003, 391 (4-6), 222-224; Shen, Z.; Uno, H.; Shimizu, Y.; Ono, N., Organic & Biomolecular Chemistry 2004, 2 (23), 3442-3447; Yamada, H.; Kushibe, K.; Okujima, T.; Uno, H.; Ono, N., Chemical Communications 2006, (4), 383-385; Inokuma, Y.; Matsunari, T.; Ono, N.; Uno, H.; Osuka, A., Angewandte Chemie-International Edition 2005, 44 (12), 1856-1860. 12. Jeong, S.; Min, B.; Cho, S.; Lee, C.; Park, B.; An, K.; Lim, J., Journal of Organic Chemistry 2012, 77 (18), 8329-8331. 13. Finikova, O.; Cheprakov, A.; Beletskaya, I.; Vinogradov, S., Chemical Communications 2001, (03), 261-262. 14. Lindsey, J. S.; MacCrum, K. A.; Tyhonas, J. S.; Chuang, Y. Y., The Journal of Organic Chemistry 1994, 59 (3), 579-587. 15. Finikova, O.; Cheprakov, A.; Beletskaya, I.; Carroll, P.; Vinogradov, S., Journal of Organic Chemistry 2004, 69 (2), 522-535. 16. Filatov, M.; Lebedev, A.; Vinogradov, S.; Cheprakov, A., Journal of Organic Chemistry 2008, 73 (11), 4175-4185. 17. Banala, S.; Ruhl, T.; Wurst, K.; Krautler, B., Angewandte Chemie-International Edition 2009, 48 (14), 2442-2442; Krautler, B.; Sheehan, C.; Rieder, A., Helvetica Chimica Acta 2000, 83 (3), 583-591. 18. Jiao, L.; Hao, E.; Fronczek, F.; Vicente, M.; Smith, K., Chemical Communications 2006, (37), 3900-3902. 19. Silva, A. M. G.; de Oliveira, K. T.; Faustino, M. A. F.; Neves, M. G. P. M. S.; Tomé, A. C.; Silva, A. M. S.; Cavaleiro, J. A. S.; Brandão, P.; Felix, V., European Journal of Organic Chemistry 2008, 2008 (4), 704-712.

Chapter 2: A Concise Approach to the Synthesis of opp-Dibenzoporphyrins and Application towards DSSCs

Abstract A novel two-step synthetic method starting from free-base tetraarylporphyrins for the synthesis of opp-dibenzoporphyrins was established. This method can be used to introduce a variety of functional groups to the fused benzene rings. opp-Dibenzoporphyrins bearing carboxylic acid linker groups on the benzene rings fused to the porphyrin periphery through the ’-positions were synthesized, and were examined for the first time as sensitizers for dye-sensitized solar cells (DSSCs). All the porphyrins displayed moderate solar to electricity conversion efficiencies of 1.54-3.14%. One opp-dibenzoporphyrin with conjugated carboxylic acid linkers displayed the highest conversion efficiency and exceptionally high Jsc value. Cyclic voltammetry study of these porphyrins revealed that the fusion of two aromatic benzene rings to the porphyrin periphery decreases the HOMO-LUMO gap; and the incorporation of the conjugated carboxylic acid linker groups decreases the HOMO-LUMO gap even further. Incorporation of the conjugated carboxylic acid linkers on the porphyrin considerably broadened and red-shifted the absorption bands of the porphyrin leading to higher conversion efficiency. Papers published from the work in this chapter:

1) R. Deshpande, L. Jiang, G. Schmidt, J. Rakovan, X. Wang, K. Wheeler, and H. Wang.

Organic Letters, 2009, 11(19), 4251-4253.

2) R. Deshpande, B. Wang, L. Dai, L. Jiang, C. S. Hartley, S. Zou, H. Wang and L. Kerr.

Chemistry – An Asian Journal. 2012, 11, 2662-2669.

2.1 Dyes for Dye-Senstized Solar Cells (DSSCs)

Due to the increasing global demand on energy, environmental friendly and renewable energy sources have attracted special attention in recent years. Among all the available techniques, DSSCs have emerged as a promising photovoltaic technology, offering a cheap and efficient way to solve the energy problem.1 DSSCs have four main components: dye (sensitizer), nanocrystalline semiconductor

- - photoanode (TiO2 is the most often used), redox electrolyte (liquid I /I3 is the most used), and counter electrode (Pt is often used), all of which are currently undergoing rapid development. Three different types of sensitizers have attracted significant amount of attention: inorganic dyes, quantum dots (QDs) and organic dyes. Among all these different dyes, ruthenium polypyridyl dyes are the most promising ones,

- - currently the best match for TiO2 based DSSCs using liquid I /I3 as the electrolyte, displaying an overall photovoltaic cell energy conversion efficiency of 11.5 %.2, 3

4 -1 -1 However, their low molar absorption coefficient (max ~ 1.4 x 10 M cm ) and high expenses together with environmental concerns, limit their practical applications. CdS is one of the most commonly used QDs sensitizers due to its adjustable particle size and better UV stability as well as their reduced dark current.4 But the current efficiencies of CdS sensitized solar cells are still very low.5 Besides the fast development of the other three major components of DSSCs,3, 6 a variety of organic dyes have been explored to widen the applications of DSSCs. Due to their relatively high molar absorption coefficients, low to modest cost, as well as the relatively easier synthetic methods and modifications, organic dyes such as indolines,7 coumarins,8 phthalocyanines,9 hemicyanines10 and porphyrins11 have been investigated for DSSCs, and their conversion efficiencies range between 2 to 9%.

2.2 Porphyrins as Light Harvester for DSSCs

In the past decade, porphyrins have emerged as attractive candidates for light

18 harvesters for DSSCs, given their primary light harvesting role in nature. Compared with traditional mononuclear and polynuclear inorganic dyes and quantum dots dyes based on heavy metals such as Ru, Os, Pt, Re and Cd, porphyrin dyes have incomparable advantages. First, porphyrin based chromophores play an important role in capturing solar energy and converting it efficiently to chemical energy in nature.12 Second, porphyrins usually have a very intense Soret band with high molar absorption

5 -1 -1 coefficient (max ~ 4 x 10 M cm ) at 400-500 nm and multiple moderate Q bands at 550-750 nm. This property makes them promising candidates as light harvesters for DSSCs. Third, in general, porphyrins are highly thermo- and chemo-stable. Fourth, the optical, photophysical and electrochemical properties of porphyrins can be easily tuned through displacing the central metal to a different one or functionalizing the porphyrin periphery. Despite these advantages, the overall performance of porphyrins in DSSCs is still poor comparing with the ruthenium dyes. In order to be usable in DSSCs, a linker (usually a carboxylic acid group) must be attached at the porphyrin periphery to connect the dyes with the semiconductor. Previous study focused on the porphyrins with the linker group attached at the meso position(s), which always exhibited relatively low solar to electricity conversion efficiencies (< 3%)13. (Figure 2.1 (a)) It was discovered in 2007 that when the linker group was moved to the -position(s) of porphyrin, the solar to electricity conversion efficiency (7.1%) was significantly increased.14 (Figure 2.1 (b)) Recently, a new class of porphyrins which have donor and acceptor groups at the meso-positions through a -conjugated bridge has been developed. Through incorporating two octyloxy groups and combining the porphyrin dye with another donor-bridge-acceptor dye that has an absorption spectrum complementary to the porphyrin further improved the power conversion efficiency of the photovoltaic device to 12.3%.15 (Figure 2.1 (c)) This is the best efficiency so far for DSSCs. Despite the recent development of porphyrin dyes in DSSCs, only a few -functionalized -extended porphyrins have been reported mainly due to their

19 synthetic complexity and lack of methods for the functionalization of their periphery.16 Recently, Imahori group showed a class of quinoxaline-fused -extended porphyrins containing carboxylic acid linkers at the -pyrrolic position, leading to a maximum cell efficiency of 6.3%.17 (Figure 2.1 (d)) Nevertheless, -extended porphyrins appear to be the best candidates for further exploration of porphyrin dyes for DSSCs. Due to their extension of -conjugation, the Soret band of these porphyrins can be bathochromically shifted up to 450-550 nm, and enhancement of Q bands is often observed. These features make this class of molecules extremely attractive as dye sensitizers for DSSCs. Most of the available procedures to synthesize -extended porphyrins require harsh conditions and/or lengthy process. In particular, the lack of functional groups in these methods makes the design and synthesis of functionalized extended porphyrins challenging. Recently, we have developed a concise method to prepare functionalized benzoporphyrins. In this method, an alkene reacts with a , ’-dibromoporphyrin through a three-step cascade reaction involving a vicinal two-fold Heck reaction, 6- electrocyclization, and subsequent aromatization.18 The wide variety of commercially available alkenes and the easy access to bromoporphyrins make it possible to access relatively large number of functionalized benzoporphyrins. Herein, we reported the synthesis and characterization of a series of novel opp-dibenzoporphyrins, and their evaluation as light harvesters in DSSCs.

Figure 2.1 Different types of porphyrin dyes studied for DSSCs

2.3 Results and Discussion

2.3.1 Synthesis

The synthesis of the opp-dibenzoporphyrin dyes is idllustrated in Scheme 2.1. We have reported the synthetic methods for 5a and 5b.18 It is well accepted that porphyrin aggregation can affect solar cell efficiency. So for the comparison purposes, opp-benzoporphyrins with increasly bulky substituents, e.g. p-methyl, p-iPr, and p-tert-butyl groups, on the meso-phenyl rings were prepared. The opp-benzoporphyrin

21 bearing p-methoxy group on the meso-phenyl rings was also prepared for the study of electronic effects. To further extend the  system of these porphyrins, a conjugated alkene, (E)-ethyl penta-2,4-dienoate, was synthesized. Using a similar synthetic method, tetrabromoporphyrins 2a-2d were readily obtained through the treatment of corresponding free base porphyrin 1a-1d with NBS in refluxing chloroform in high yields (85-95%). Tetrabromoporphyrins 2a-2d reacted with methyl acrylate or (E)-ethyl penta-2,4-dienoate through the Pd0 catalyzed cascade reaction to afford opp-dibenzoporphyrins 3a-3e in 45-60% yields. Metal insertion with Zn(OAc)2 converted free base porphyrins 3a-3e to ZnII porphyrins 4a-4e in high yields (85-95%). Hydrolysis of 4a-4e using KOH in refluxing iPrOH afforded DSSC suitable ZnII opp-dibenboporphyrins 5a-5e in 65-78% yields. The structures of all the new compounds have been verified by 1H and 13C NMR spectroscopy, UV-Vis spectroscopy, and LDI-TOF mass spectrometry. 19

Scheme 2.1 Synthesis of opp-dibenzoporphyrins

2.3.2 Optical and Electrochemical Properties

Figure 2.2 displays the UV-visible absorption spectra of compounds 5a-5e in

22 methanol. The Soret band of 5a-5d is centered at 458nm, 20-30nm red-shifted relative to their respective parent porphyrins 1a-1d upon fusion of two benzene rings on the porphyrin. The Soret band of 5e is further red-shifted by 11 nm relative to 5a-5d through the incorporation of conjugated carboxylic acid linkers on the porphyrin. The Q bands of 5e are also red-shifted and broadened. The absorption of 5e is expected to have more overlap with the solar spectrum, and should be more beneficial for harnessing solar energy.

1.00 5a 5b 0.75 5c 5d

0.50 5e Absorption 0.25

0.00 375 425 475 525 575 625 675 725 Wavelength (nm)

Figure 2.2. Normalized UV-visible absorption spectra of 5a-5e in methanol. Figure 2.3 displays the steady-state fluorescence spectra of 5a-5e in methanol. The solutions of these porphyrins were excited at the wavelength given the strongest Soret absorption. Two emission bands were observed centering at 638nm and 703 nm for compounds 5a-5d. Corresponding to its red-shifting and broadening of the absorption, the two emission bands of 5e were also broadened and red-shifted to 643nm and 708nm, respectively.

600 5a 500 5b 400 5c 5d 300

5e Emission 200

100

0 550 600 650 700 750 800 850 Wavelength (nm) Figure 2.3 Steady-state fluorescence spectra of 5a-5e in methanol.

2.3.3 DFT Calculations

In order to gain insight into the optical and electrochemical properties of the DSSC-suitable porphyrins 5a-5e, DFT calculations were performed. From the optimized geometry (Figure 2.4), all the porphyrins (5a-5e) adopt slightly saddled conformations. The electronic absorption of porphyrins including both Soret band and Q bands arises from -* transitions. The frontier orbitals responsible for the transitions in the parent porphin are two orbitals (a1u and a2u) and two degenerate * 20 orbitals (egx and egy) in the Gouterman four-orbital model. Analogous MOs are predicted for the current, less symmetrical (effectively C2v) structures (Figure 2.5). The calculated energy levels (Table 2.1) suggest that the LUMO and LUMO+1 remain very nearly degenerate for all the porphyrins 5a-5e, owing to the symmetric structure of these porphyrins. The energy levels of the frontier orbitals of 5a, 5b and 5c are almost identical, which can be explained by similar alkyl substituents (methyl, iPr, and tBu, respectively) at the para-positions of the meso-phenyl groups. When the substituents at the para-position of the meso-phenyl ring are replaced with methoxy groups (5d), both the HOMO and the LUMO energy levels move up slightly by ~0.06 eV relative to those of 5a, while the HOMO-LUMO energy gap (~2.67 ev) remains unchanged. The LUMO and LUMO+1 of 5e are essentially degenerate, but the HOMO−1 and HOMO are further apart than those of 5a-5c. Comparison of 5e with 5a suggests that incorporation of conjugated carboxylic acid linkers leads to a decreased HOMO-LUMO energy gap (by 0.13 ev), which can be ascribed to the significantly lower-energy LUMO as the HOMO is almost unchanged. These effects are consistent with the red-shifting and broadening in the absorption and emission of 5e, upon the introduction of conjugated linker groups to the fused benzene rings of the porphyrin. The HOMO of 5a-5e involves heavy participation of the fused benzene rings. While the LUMO of 5a-5d remains mainly on the porphyrin core, the LUMO of 5e engages both the fused benzene rings and the conjugated linker groups.

Figure 2.4 Molecular geometry of 5a-5e calculated at the B3LYP/6-31G(d) level of theory.

Figure 2.5 Molecular orbitals of 5a-5e calculated at the B3LYP/6-31G(d) level of theory.

Table 2.1. Calculated molecular orbital energy levels of 5a-5e.

Compounds HOMO-1 HOMO LUMO LUMO+1

5a -5.25 -5.13 -2.46 -2.43

5b -5.27 -5.14 -2.47 -2.44

5c -5.26 -5.13 -2.46 -2.43

5d -5.12 -5.07 -2.40 -2.38

5e -5.34 -5.11 -2.57 -2.53

2.3.4 Cyclic Voltammetry

Cyclic voltammetry (CV) was used to determine the redox potentials of these porphyrins (Figure 2.6). Both of the first oxidation potentials of the dibenzoporphyrins 5a and 5b shift negatively by 100 mV, as compared with their respective parent porphyrins, indicating the increase of the HOMO energy level by 0.1 eV upon fusion of two benzene rings on the porphyrin periphery. The LUMO energy level remains almost unchanged (decreases only about 0.03 eV) (Figure 2.6, (a) and (b)). The reduced HOMO-LUMO energy gap corresponds to their red-shifted electronic absorption. Due to the low solubility of 5d and 5e in DMF, the CV of 4d and 4e were measured instead, assuming no significant difference between the carboxylic acid group and the methyl ester group exists for the electronic properties. For comparison, the CV of 4a was also measured. The potentials are collected in Table 2.2. The peak potential for the first reduction (around -1.0 V) is nearly unchanged for 4a, 4d and 4e. However, the difference comes in the addition of second electron; 4e is more readily reduced than 4a and 4d. For the first oxidation (the peak around +1.0 V), 4e is about 50 mV more easily oxidized than both 4a and 4d. These CV data suggest that 4e has the smallest HOMO-LUMO energy gap; 4a and 4d have similar HOMO-LUMO energy gap. Incorporation of conjugated linker groups decreases the HOMO-LUMO energy gap, consistent with the red shifts of both the Soret and Q bands in the electronic absorption spectra. These data also agree well with the DFT calculations.

Figure 2.6 Cyclic voltammograms of porphyrins in DMF containing 0.1M TBAP. Scan rate: 0.1

Vs-1. (a) 1b and 5b; (b) 1a and 5a; (c) 4a, 4d and 4e.

Table 2.2. The oxidation-reduction potentials (V) of porphyrins.

V vs Pt Compounds -1 to -2 0 to -1 0 to +1

4a -1.40 -1.00 +1.06

4d -1.40 -1.00 +1.04

4e -1.30 -1.00 +1.01

2.3.5 Photovoltaic Properties of Porphyrin-Sensitized TiO2 Solar Cells

The amount of dye absorbed on TiO2 surface influences the cell efficiency. In order to obtain an optimized solar to electricity conversion, a concentration study was carried out using 5b. The TiO2 electrode was immersed in ethanol solutions of the porphyrin (2b) for 12 hr. The solar to electricity conversion efficiency () was calculated as: = JscVocFF. As illustrated in Table 2.3, the best solar to electricity conversion efficiency () was obtained at 0.2 mM. At lower concentration, the conversion efficiency () was also lower, which may be due to the insufficient absorption of the dye; when higher concentrations were used, a decreasing trend of conversion efficiency () was clearly observed. The decreasing conversion efficiency () is likely attributed to the more serious aggregation of porphyrin at higher concentrations. On the other hand, the excess dye molecules block the subsequent layers of dye next to TiO2 surface to be exposed to the redox species, and reduce the dye regeneration, resulting in lower solar to electricity conversion efficiency in DSSC.

The optimized concentration (0.2 mM, ethanol) was then used for modifying the

TiO2 electrode with other porphyrins (5a, 5c-5e). The data for the solar cell performance of 5a-5e are organized in Table 2.4. N719 dye ([RuL2(NCS)2]:2TBA,

L=2,2'-bipyridyl-4,4'-dicarboxylic acid, TBA=tetra-n-butylammonium) on TiO2 was used as a reference. 5a, 5b and 5c were prepared to understand how the porphyrin aggregation affects the solar energy to electricity conversion efficiency of DSSC. 5c bears the most bulky group on the porphyrin meso-phenyl rings, and are expected to have least aggregation as compared with 5a and 5b. It is generally understood that the less the aggregation, the higher the solar energy conversion. Due to similar alkyl substitutions at the meso-phenyl rings of the porphyrins, the optical and electrochemical properties of 5a, 5b and 5c are almost identical. Contrary to the generally understanding, 5a, which is expected to have the highest aggregation effect, displayed the highest solar to electricity conversion efficiency (= 2.93%). The lower

29 efficiency of 5c is attributed to its much lower Jsc. UV-Vis absorption spectra of the dye solutions before and after immersion of TiO2 electrode are shown in Figure 2.7 5a displays a much bigger absorption difference between before and after dye adsorption relative to that of 5c, implying larger amount of 5a was adsorbed on TiO2 surface than 5c under similar conditions (0.2 mM, 12 hr). It is likely that the lower Jsc of 5b and 5c was resulted from the lesser amount of porphyrin dyes present on the TiO2 surface, and the aggregation effect did not play a role in determining the solar conversion efficiencies of these porphyrins under the above mentioned conditions.

The solar to electricity conversion efficiency of 5d is 2.22%, smaller than that of 5a. This can be explained by a slight increase in the HOMO-LUMO gap of 5a due to the introduction of a strong electron-donating methoxy groups at the para-positions of the meso-phenyl rings. The highest solar conversion efficiency was obtained with 5e. 5e possesses a more extended  system than 5a-5d. The UV-Vis spectra of 5e showed much more broadened and red-shifted absorption bands. This feature is expected to bring broader overlap with solar spectrum, and thus enhance solar energy conversion. It is notable that the Jsc (12.87) value of 5e is exceptionally high, comparable to that of the best performing porphyrin dye in DSSC.21 This suggests that attaching conjugated linkers can considerably change the electronic profile of the porphyrin dye, and thus significantly enhance the light harvesting capability of porphyrin dye.

Table 2.3. Concentration study of porphyrin 5b.

Concentration(mM) Jsc (mA/cm2) Voc (V) FF Efficiency (%)

0.1 3.61 0.5 0.45 0.82

0.2 6.26 0.54 0.47 1.59

0.3 4.27 0.54 0.46 1.06

0.4 3.41 0.53 -0.46 0.82

0.5 1.28 1.05 0.41 0.54

Table 2.4. Photovoltaic performance of DSSCs based on the opp-dibenzoporphyrins (5a-5e)

Dye Jsc (mA/cm2) Voc (V) FF Efficiency (%)

N719 19.6 0.72 0.57 7.98

5a 8.86 0.57 0.58 2.93

5b 5.05 0.54 0.56 1.54

5c 5.52 0.58 0.57 1.81

5d 7.48 0.57 0.53 2.22

5e 12.87 0.53 0.46 3.14

2 5c before 5c after 5a before 5a after

1 Absorbance (AU) Absorbance

0 350 400 450 500 550 600 650 700 750 800 850

Wavelength (nm)

Figure 2.7 UV-vis absorption data of 5a and 5c. Before: before adsorption on TiO2 surface; after:

after adsorption on TiO2 surface.

2.4 Conclusion

Novel opp-dibenzoporphyrins have been synthesized through Pd0 catalyzed cascade reaction. These opp-dibenzoporphyrins were evaluated for the first time as light harvester for dye-sensitized solar cells. These porphyrins displayed moderate solar to electricity conversion efficiencies of 1.54-3.14%. Our study shows that installing a more bulky group on the porphyrin meso-phenyl rings did not help enhance the solar energy conversion, which is likely due to decreased adsorption of porphyrin dye on the TiO2 surface. Incorporation of conjugated carboxylic acid linkers on the porphyrin considerably broadened and red-shifted the absorption bands of the porphyrin leading to a very high Jsc value (12.89), comparable to the best of porphyrin dyes for DSSC. The electronic and optical properties have been measured using UV-Vis, steady-state fluorescence spectrometry and cyclic voltammetry, and these data correlates well with DFT calculations.

2.5 Experimental Section

General All reported synthetic work was performed at the Departement of Chemistry and Biochemistry of Miami University. All solvents were analytical reagent grade unless otherwise stated and were obtained either from Sigma-Aldrich or ACROS. Analytical TLCs were performed on Silicycle UltraPure Silica Gel 60 F254 TLC plates. Preparative column chromatography was performed on silica gel (40-60m), which was purchased from Silicycle. 1H and 13C experiments were conducted on a Bruker Avance 500 NMR spectrometer. All samples

1 were prepared in either CDCl3 or CD3OD. The chemical shifts for H NMR were referenced to CDCl3 at 7.24 ppm, and to CD3OD at 4.78 and 3.35 ppm. The chemical

13 shifts for C NMR were referenced to CDCl3 at 77 ppm and CD3OD at 49.3 ppm. UV-Visible spectra were recorded on an Agilent 8453 UV-Visible spectrometer in

CH2Cl2 or MeOH. Mass spectra were obtained on Bruker LDI-TOF mass spectrometer. Melting points were measured on an Electrothermal MEL-TEMP apparatus and were uncorrected. DFT calculations. Gaussian 03 (Rev. D.02) calculations were carried out on Miami University's Redhawk computer cluster. Following geometry optimizations, vibrational frequency analysis was used to ensure that all stationary points were energy minima. Geometry optimization and electronic structure calculations of the porphyrins were performed by using the B3LYP functional and 6-31G(d) basis set. Molecular orbitals were visualized using Molekel 5.4.0.8. (Ugo Varetto, Molekel 5.4.0.8; Swiss National Supercomputing Centre: Manno)

Electrochemistry. Cyclic voltammograms were recorded with a three-electrode cell using a CHI 750 electrochemical analyzer. A Au coil was used as the working electrode, a Pt coil and a Pt disk served as the counter electrode and the quasi-reference electrode, respectively.

Photovoltaic measurement. The porphyrin molecules were sensitized on TiO2 for solar cell efficiency testing. TiO2 film method used in this experiment is based on

Nazeeruddin et al 1993. Commercial TiO2 (P25, Degussa AG, Germany), acetylacetone and Triton X-100 were used to make TiO2 paste. A doctor blade method was used to spread the paste onto SnO2: F (FTO) glasses. Samples were then annealed in 450C for 30 min.

Synthesis Synthesis of tetrabromoporphyrins 2. In a 250 ml round bottom flask, a mixture of porphyrin 1 (0.4 g, 0.5 mmol) and NBS (0.6 g, 3.3 mmol) in dry chloroform (120 ml) was refluxed overnight. The progress of the reaction was monitored by TLC using

CHCl3/cyclohexane (1:1) as eluent. The reaction mixture was then cooled down to room temperature and filtered through a short plug of silica (eluting with CH2Cl2).

The filtered was evaporated to dryness and recrystallized from CH2Cl2/MeOH to give 2.

o 1 o 2b: Dark brown powder (93%); mp. >310 C; H-NMR (500 MHz, CDCl3, 25 C, TMS): δ= 8.71 (s, 4 H; β-H), 8.08 (d, 3J(H,H)=8.0 Hz, 8H; o-Ph–H), 7.62 (d, 3J(H,H)=8.0 Hz, 8H; m-Ph–H), 3.23 (m, 4 H; isopropyl CH), 1.51 (d , 3J(H,H)=7.0

13 o Hz, 24 H; isopropyl CH3), -2.76 ppm (s, 2 H; inner H); C-NMR (125 MHz, 25 C, TMS): δ= 24.37, 34.19, 120.41, 124.69, 125.65, 135.68, 138.10, 140.39, 148.42,

149.62 ppm; UV/Vis (CH2Cl2): λmax (log )= 442 (5.66), 542 (4.89), 692 (4.85), 628 (4.83), 634 nm (4.80); MS (LDI-TOF): m/z calculated for C56H50Br4N4: 1098.640; found: 1098.542.

o 1 o 2c: Dark brown powder (85%); mp> 310 C; H-NMR (500 MHz, CDCl3, 25 C, TMS): δ= 8.72 (s, 4 H; β-H), 8.08 (d, 3J(H,H)=8.0 Hz, 8 H;o-Ph–H), 7.65 (d, 3J(H,H)=8.0 Hz 4H; m-Ph–H), 1.57 (s, 36 H; tert-butyl-H), -2.76 ppm (s, 2 H; inner

H); UV/Vis (CH2Cl2): λmax (log )= 432 (5.66), 533 (4.56), 673 (4.35), 604 (4.19), 625 nm (4.15); The spectroscopic data were in agreement with the literature.[11]

Synthesis of free base opp-dibenzoporphyrins 3a-3e: Tetrabromoporphyrin 2 (0.045 mmol), palladium acetate (5.12 mg, 0.023 mmol), triphenylphosphine (15.54 mg,

0.059 mmol) and K2CO3 (26.16 mg, 0.19 mmol) were added to a Schlenk flask and dried under vacuum. The vacuum was released under argon to allow the addition of DMF (7 mL), dry toluene (9 mL) and methyl acrylate or (E)-methyl penta-2,4-dienoate (1.35 mmol). The mixture was then degassed via four freeze-pump-thaw cycles before the flask was purged with argon again. The Schlenk flask was sealed and heated to reflux for 72 h. The solvent was removed. The residue was subjected to preparative column chromatography. The band containing the desired porphyrin was collected and recrystallized from CHCl3/MeOH.

o 1 o 3a: Reddish brown crystal (60%); mp > 330 C; H-NMR (500 MHz, CDCl3, 25 C,

TMS): δ= 8.85 (s, 4 H; -H), 8.05 (d, 3J(H,H)=8.0 Hz, 8 H; o-Ph–H), 7.65 (d, 3J(H,H)=7.5 Hz, 8H; m-Ph–H), 7.37 (s, 4H; fused-benzene-H), 3.90 (s, 12H;

13 o -OCH3 ), 2.76 (s, 12H; -CH3), -2.66 ppm (s, 2H; inner H); C-NMR (125 MHz, 25 C, TMS): δ=21.62, 29.71, 52.48, 118.19, 125.76, 127.90, 128.60, 129.06, 138.63, 138.73, 139.16, 142.70, 148.73, 168.43 ppm; IR: ν˜= 3001, 2962, 2919, 2850, 1712, 1419,

-1 1358, 1259, 1219, 1091, 1009, 901, 791, 701 cm ; UV/Vis (CH2Cl2): λmax (log )= 443 (6.02), 534 (5.21), 568 (5.19), 610 (5.16) 628 nm (5.15); MS (LDI-TOF): m/z calculated for C64H50N4O8: 1002.363; found: 1002.253.

o 1 o 3b: Reddish brown crystal (55%); mp > 320 C; H-NMR (500 MHz, CDCl3, 25 C, TMS): δ= 8.82 (s, 4H; -H), 8.10 (d, 3J(H,H)=7.5 Hz, 8H; o-Ph–H), 7.68 (d,

3 J(H,H)=7.5 Hz, 8H; m-Ph–H), 7.39 (s, 4H; fused-benzene-H), 3.85 (s, 12H; -OCH3 ),

3 3.30 (m, 4H, isopropyl CH), 1.58 (d, J(H,H)=7.0 Hz, 24H; isopropyl CH3), -2.60

13 o ppm (s, 2H; inner H); C-NMR (125 MHz, CDCl3, 25 C, TMS): δ= 24.43, 34.35, 52.46, 119.03, 125.35,125.96, 127.93, 129.10, 133.89, 138.95, 139.24, 142.55, 148.72 149.85, 168.57 ppm; IR: ν˜= 2961, 2919, 2850, 1735, 1459, 1375, 1259, 1089, 864,

-1 691 cm .UV/Vis (CH2Cl2): λmax (log )= 444 (6.43), 534 (5.43), 568 (5.32), 611 nm (5.29); MS (LDI-TOF): m/z calculated for C72H66N4O8: 1114.488, found: 1114.455.

o 1 o 3c: Reddish brown crystal (48%); mp > 330 C; H-NMR (500 MHz, CDCl3, 25 C, TMS): δ= 8.83 (s, 4 H; -H), 8.11 (d, 3J(H,H)=8.0 Hz, 8 H; o-Ph–H), 7.83 (d, 3J(H,H)=8.0 Hz, 8H; m-Ph–H), 7.39 (s, 4H; fused-benzene-H), 3.84 (s, 12H;

13 -OCH3 ), 1.64 (s, 36H; tert-butyl-H), -2.57 ppm (s, 2H; inner H); C-NMR (125 MHz,

o CDCl3, 25 C, TMS): δ= 31.74, 35.07, 52.48, 118.97, 124.80, 125.26, 127.93, 129.10, 133.68, 138.53, 139.22, 142.49, 148.53, 152.21, 168.43 ppm; IR: ν˜= 3001, 2962, 2919, 2850, 1712, 1419, 1358, 1259, 1219, 1091, 1009, 901, 791, 701 cm-1; UV/Vis

(CH2Cl2): λmax (log )= 440 (5.64), 530 (5.21), 566 nm (5.20); MS (LDI-TOF): m/z calculated for C76H74N4O8: 1170.551, found: 1170.455.

o 1 o 3d: Brown solid (52%); mp > 330 C; H-NMR (500 MHz, CDCl3, 25 C, TMS): δ= 8.87 (s, 4H; -H), 8.06 (d, 3J(H,H)=8.0 Hz, 8H; o-Ph–H), 7.40 (s, 4H;

3 fused-benzene-H), 7.35 (d, J(H,H)=8.0 Hz, 8H; m-Ph–H), 4.13 (s, 12H; -OCH3 ),

13 o 3.90 (s, 12H; -OCH3), -2.64 ppm (s, 2H; inner H); C-NMR (125 MHz, CDCl3, 25 C, TMS): δ= 52.48, 55.74, 113.48, 118.63, 125.74, 127.89, 129.05, 134.03, 134.81, 139.41, 142.61, 149.05, 160.43, 168.43 ppm; IR: ν˜= 3001, 2962, 2919, 2850, 1712,

-1 1419, 1358, 1259, 1219, 1091, 1009, 901, 791, 701 cm ; UV/Vis (CH2Cl2): λmax (log )= 444 (6.02), 534 (5.00), 610 (4.71) 634 nm (4.53); MS (LDI-TOF): m/z calculated for C64H50N4O12: 1066.343, found: 1066.293.

o 1 o 3e: Brown solid (45%); mp > 330 C; H-NMR (500 MHz, CDCl3, 25 C, TMS): δ= 8.84 (s, 4H; -H), 8.13 (d, 3J(H,H)=15.5 Hz, 4H; alkenyl protons), 8.05 (d, 3J(H,H)=7.5 Hz, 8H; o-Ph-H), 7.68 (d, 3J(H,H)=7.5 Hz, 8H; m-Ph-H), 7.29 (s, 4H; fused-benzene-H), 6.05 (d, 3J(H,H)=15.5 Hz, 4H; alkenyl protons), 3.89 (s, 12H;

-OCH3), 2.81 (s, 12H; -CH3), -2.61 ppm (s, 2H; inner H); UV/Vis (CH2Cl2): λmax (log )= 461 (6.08), 488 (5.99), 580 (4.31) 618 (4.16), 675 nm (3.83); MS (LDI-TOF): m/z calculated for C72H58N4O8: 1107.250, found MS: 1108.002.

Synthesis of ZnII opp-dibenzoporphyrins 4a-4e: A mixture of porphyrin 3 (0.042 mmol) and Zn(OAc)2 (6.0 equiv.) in CHCl3/MeOH (18 ml/ 6 ml) was heated under reflux for 1 h. Upon completion of the reaction (monitored by TLC), the solvent was evaporated and the residue was subjected to column chromatography using CH2Cl2 as eluent. The fraction containing the desired compound was collected and recrystallized from CH2Cl2/MeOH.

o 1 o 4a: Greenish brown solid (90%); mp > 330 C; H-NMR (500 MHz, CDCl3, 25 C, TMS): δ= 8.85 (s, 4H; -H), 7.95 (d, 3J(H,H)=7.5 Hz, 8H; o-Ph–H), 7.62 (d,

3 J(H,H)=8.0 Hz, 8H; m-Ph–H), 7.41 (s, 4H; fused-benzene-H), 3.72 (s, 12H; -OCH3),

13 o 2.76 ppm (s, 12H; CH3 ); C-NMR (125 MHz, CDCl3, 25 C, TMS): δ= 22.60, 53.46,

119.98, 126.75, 128.90, 129.73, 130.05, 134.58, 139.62, 139.72, 140.15, 143.63, 149.71, 169.42 ppm; IR: ν˜= 3001, 2956, 2921, 2850, 1710, 1427, 1336, 1253, 998,

-1 791, 764, 670 cm ; UV/Vis (CH2Cl2): λmax (log )= 450 (5.88), 581 (5.30), 627 nm (5.26); MS (LDI-TOF): m/z calculated for C64H48N4O8Zn: 1064.402, found: 1064.139.

o 1 o 4b: Greenish brown solid (93%); mp > 330 C; H-NMR (500 MHz, CDCl3, 25 C, TMS): δ= 8.81 (s, 4H; -H), 8.02 (d, 3J(H,H)=7.5 Hz, 8H; o-Ph–H), 7.63 (d,

3 J(H,H)=8.0 Hz, 8H; m-Ph–H), 7.36 (s, 4H; fused-benzene-H), 3.59 (s, 12H; -OCH3),

3 3.28 (m, 4H; isopropyl CH), 1.56 ppm (d , J(H,H)=7.0 Hz, 24H; isopropyl CH3);

13 o C-NMR (125 MHz, CDCl3, 25 C, TMS): δ= 24.44, 34.29, 52.28, 119.74, 125.60, 127.99, 131.56, 133.35, 139.96, 140.37, 143.89, 149.32, 150.62, 168.32 ppm; IR: ν˜= 3001, 2956, 2919, 2850, 1719, 1429, 1350, 1250, 1105, 1018, 972, 837, 791, 764, 693

-1 cm ; UV/Vis (CH2Cl2): λmax (log )= 451 (6.25), 582 (5.36), 622 nm (5.22); MS (LDI-TOF): m/z calculated for C72H64N4O8Zn: 1176.402, found: 1176.366.

o 1 o 4c: Greenish brown solid (95%); > 330 C; H-NMR (500 MHz, CDCl3, 25 C, TMS): δ= 8.80 (s, 4H; -H), 8.02 (d, 3J(H,H)=7.5 Hz, 8H; o-Ph–H), 7.76 (d, 3J(H,H)=8.0 Hz,

8H; m-Ph–H), 7.26 (s, 4H; fused-benzene-H), 3.49 (s, 12H; -OCH3 ), 1.60 ppm (s,

13 o 36H; tert-butyl-H); C-NMR (125 MHz, CDCl3, 25 C, TMS): δ= 31.77, 34.99, 52.28, 119.58, 124.38, 125.39, 127.71, 131.54, 133.18, 139.61, 140.30, 143.75, 150.63, 151.57, 168.26 ppm; IR: ν˜= 3001, 2962, 2919, 2850, 1712, 1419, 1358, 1259, 1219,

-1 1091, 1009, 901, 791, 701 cm . UV/Vis (CH2Cl2): λmax (log )= 451 (6.02), 581 (4.71), 556 nm (4.60); MS (LDI-TOF): m/z calculated for C76H72N4O8Zn: 1232.464, found: 1232.479.

o 1 o 4d: Greenish solid (85%); mp > 330 C; H-NMR (500 MHz, CDCl3, 25 C, TMS): δ= 8.83 (s, 4H; -H), 7.96 (d, 3J(H,H)=6.5 Hz, 8H; o-Ph–H), 7.45 (s, 4H;

3 fused-benzene-H), 7.29 (d, J(H,H)=6.5 Hz, 8H; m-Ph–H), 4.09 (s, 12H; -OCH3 ),

13 o 3.77 ppm (s, 12H; -OCH3); C-NMR (125 MHz, CDCl3, 25 C, TMS): δ= 52.91, 56.24, 113.98, 119.14, 126.24, 128.40, 129.55, 134.54, 135.32, 139.92, 143.11, 149.55, 160.94, 168.88 ppm; IR: ν˜= 3001, 2962, 2919, 2850, 1712, 1419, 1358, 1259,

-1 1219, 1091, 1009, 901, 791, 701 cm ; UV/Vis (CH2Cl2): λmax (log )= 452 (6.20), 580 (5.50), 625 nm (5.42); MS (LDI-TOF): m/z calculated for C64H48N4O12Zn: 1128.256, found, m/z, 1128.075.

o 1 o 4e: Greenish brown solid (84%); mp > 330 C; H-NMR (500 MHz, CDCl3, 25 C,

3 TMS): δ= 8.85 (s, 4H; -H), 7.97 (d, J(H,H)=7.5 Hz, 8H; o-Ph-H), 7.82 (d, 3J(H,H)=15.5 Hz, 8H; alkenyl protons), 7.63 (d, 3J(H,H)=7.5 Hz, 8H; m-Ph-H) 7.29 (s, 4H; fused-benzene-H), 5.91 (d, 3J(H,H)=15.5 Hz, 4H; alkenyl protons), 3.76 (s,

12H; -OCH3), 2.78 ppm (s, 12H; -CH3); UV/Vis (CH2Cl2): λmax (log )= 471 (6.03), 596 (4.51), 631 (5.25) 636 nm (4.24); MS (LDI-TOF): m/z calculated for C72H56N4O8Zn: 1170.620, found: 1170.230.

Synthesis of porphyrin 5a-e: A mixture of porphyrins 4 (0.033 mmol) and KOH

(1.05 mmol) in isopropanol (15ml) /H2O (7 ml) was heated under reflux overnight. Upon completion of the reaction (monitored by TLC), the solvent was removed and the residue was purified using silica gel column. The column was first eluted with

CH2Cl2: MeOH (50:1) to separate the un-fully hydrolyzed products, and then with methanol to collect the desired product. The solvent was removed under reduced pressure and the residue was recrystallized from MeOH/CH2Cl2.

o 1 o 5a: Greenish brown solid (78%); mp > 330 C; H-NMR (500 MHz, CDCl3, 25 C, TMS): δ= 8.65 (s, 4H; -H), 7.95 (d, 3J(H,H)=7.0 Hz, 8H; o-Ph–H), 7.58 (d, 3J(H,H)=7.0 Hz, 8H; m-Ph–H), 8.16 (s, 4H; fused-benzene-H), 2.70 ppm (s, 12H;

-CH3); IR: ν˜= 3300-2800(broad),3127, 3040, 2809, 1706, 1540, 1440, 1401, 1337,

-1 1284, 1121, 992, 793, 759, 693 cm ; UV/Vis (CH2Cl2): λmax (log )= 457 (5.71), 589 (4.91), 627 nm (4.80); MS (LDI-TOF): m/z calculated for C60H40N4O8Zn:

1010.392, found: 1011.0

o 1 o 5b: Greenish brown solid (78%); mp > 330 C; H-NMR (500 MHz, CDCl3, 25 C, TMS): δ= 8.60 (s, 4H; -H), 7.92 (d, 3J(H,H)=7.5 Hz, 8H; o-Ph–H), 7.54 (d, 3J(H,H)=7.5 Hz, 8H; m-Ph–H), 7.96 (s, 4H; fused-benzene-H), 3.18 (m, 4H;

3 13 isopropyl CH)), 1.48 ppm (d , J(H,H)=6.5 Hz, 24H; isopropyl CH3); C-NMR (125

o MHz, CDCl3, 25 C, TMS): δ= 24.92, 35.66, 120.66, 126.69, 129.12, 132.01, 132.24, 134.40, 142.05, 145.55, 150.46, 151.95, 174.13 ppm; IR: ν˜= 3300-2800 (broad), 3119, 3033, 2809, 1698, 1544, 1440, 1396, 1367, 1290, 1119, 1054, 990, 790, 759,

-1 693 cm ; UV/Vis (CH2Cl2): λmax (log )= 457 (5.91), 590 (4.96), 634 nm (4.76); MS (LDI-TOF): m/z calculated for C68H56N4O8Zn: 1120.339, found: 1120.366.

o 1 o 5c: Greenish brown solid (75%); > 330 C; H-NMR (500 MHz, CDCl3, 25 C, TMS): δ= 8.59 (s, 4H; -H), 7.95 (d, 3J(H,H)=7.5 Hz, 8H; o-Ph–H), 7.74 (d, 3J(H,H)=8.0 Hz, 8H; m-Ph–H), 7.52 (s, 4H; fused-benzene-H), 1.60 ppm (s, 36H; tert-butyl-H); IR: ν˜= 3001, 2962, 2919, 2850, 1712, 1419, 1358, 1259, 1219, 1091, 1009, 901, 791, 701

-1 cm ; UV/Vis (CH2Cl2): λmax (log )= 458 (6.04), 590 (5.39), 562 nm (5.35); MS (LDI-TOF): m/z calculated for C72H64N4O8Zn: 1178.681, found: 1179.1.

o 1 o 5d: Greenish solid (70%); mp > 330 C; H-NMR (500 MHz, CDCl3, 25 C, TMS): δ= 8.59 (S, 4H; -H), 7.94 (d, 3J(H,H)=6.5 Hz, 8H; o-Ph–H), 7.90 (s, 4H; fused-benzene-H), 7.27 (d, 3J(H,H)=6.5 Hz, 8H; m-Ph–H), 4.05 ppm (s, 12H;

-OCH3 ); IR: ν˜= 3001, 2962, 2919, 2850, 1712, 1419, 1358, 1259, 1219, 1091,

-1 1009, 901, 791, 701 cm . UV/Vis (CH2Cl2): λmax (log )= 459 (6.05), 590 (5.38), 626 nm (5.30); MS (LDI-TOF): m/z calculated for C60H40N4O12Zn: 1074.359, found: 1075.075.

o 1 o 5e: Greenish solid (60%); mp > 330 C; H-NMR (500 MHz, CDCl3, 25 C, TMS): δ= 8.72 (s, 4H; -H), 8.47 (d, 3J(H,H)=15.5 Hz, 4H; alkenyl protons), 7.93 (d, 3J(H,H)=7.5 Hz, 8H; o-Ph-H), 7.65 (d, 3J(H,H)=7.5 Hz, 8H; m-Ph-H ), 7.39 (s, 4H;

39 fused-benzene-H), 6.09 (d, 3J(H,H)=15.5 Hz, 4H; alkenyl protons), 2.76 ppm (s, 12H;

-CH3); UV/Vis (CH2Cl2): λmax (log )= 471 (5.95), 596 (5.13), 638 (4.86); MS (LDI-TOF): m/z calculated for C68H48N4O8Zn: 1114.510, found: 1115.018.

Reference: 1. Hamann, T. W.; Jensen, R. A.; Martinson, A. B. F.; Van Ryswyk, H.; Hupp, J. T., Energy & Environmental Science 2008, 1 (1), 66-78; Robertson, N., Angewandte Chemie International Edition 2008, 47 (6), 1012-1014; Robertson, N., Angewandte Chemie International Edition 2006, 45 (15), 2338-2345; Martinson, A. B. F.; Hamann, T. W.; Pellin, M. J.; Hupp, J. T., Chemistry – A European Journal 2008, 14 (15), 4458-4467. 2. Nazeeruddin, M. K.; Kay, A.; Rodicio, I.; Humphry-Baker, R.; Mueller, E.; Liska, P.; Vlachopoulos, N.; Graetzel, M., Journal of the American Chemical Society 1993, 115 (14), 6382-6390. 3. Grätzel, M., Journal of Photochemistry and Photobiology A: Chemistry 2004, 164 (1–3), 3-14. 4. Yu, W. W.; Qu, L.; Guo, W.; Peng, X., Chemistry of Materials 2003, 15 (14), 2854-2860; Wang, B.; Kerr, L., Journal of Solid State Electrochemistry 2012, 16 (3), 1091-1097. 5. Chang, C.-H.; Lee, Y.-L., Applied Physics Letters 2007, 91 (5), 053503-3; Wang, B.; Kerr, L., 35th Ieee Photovoltaic Specialists Conference 2010, 1819-1822. 6. Wang, B.; Kerr, L., Solar Energy Materials and Solar Cells 2011, 95 (8), 2531-2535; Niitsoo, O.; Sarkar, S. K.; Pejoux, C.; Rühle, S.; Cahen, D.; Hodes, G., Journal of Photochemistry and Photobiology A: Chemistry 2006, 181 (2–3), 306-313; Lin, S.-C.; Lee, Y.-L.; Chang, C.-H.; Shen, Y.-J.; Yang, Y.-M., Applied Physics Letters 2007, 90 (14), 143517-3. 7. Horiuchi, T.; Miura, H.; Sumioka, K.; Uchida, S., Journal of the American Chemical Society 2004, 126 (39), 12218-12219. 8. Hara, K.; Wang, Z.-S.; Sato, T.; Furube, A.; Katoh, R.; Sugihara, H.; Dan-oh, Y.; Kasada, C.; Shinpo, A.; Suga, S., The Journal of Physical Chemistry B 2005, 109 (32), 15476-15482. 9. Durrant, J. R.; Haque, S. A.; Palomares, E., Chemical Communications 2006, 0 (31), 3279-3289; Komori, T.; Amao, Y., Journal of porphyrins and phthalocyanines (Wiley) 7 (2), 131. 10. Chen, Y.-S.; Li, C.; Zeng, Z.-H.; Wang, W.-B.; Wang, X.-S.; Zhang, B.-W., Journal of Materials Chemistry 2005, 15 (16), 1654-1661. 11. Wang, Q.; Campbell, W. M.; Bonfantani, E. E.; Jolley, K. W.; Officer, D. L.; Walsh, P. J.; Gordon, K.; Humphry-Baker, R.; Nazeeruddin, M. K.; Grätzel, M., The Journal of Physical Chemistry B 2005, 109 (32), 15397-15409. 12. Imahori, H.; Mori, Y.; Matano, Y., Journal of Photochemistry and Photobiology C: Photochemistry Reviews 2003, 4 (1), 51-83. 13. Campbell, W. M.; Burrell, A. K.; Officer, D. L.; Jolley, K. W., Coordination Chemistry Reviews 2004, 248 (13–14), 1363-1379. 14. Campbell, W. M.; Jolley, K. W.; Wagner, P.; Wagner, K.; Walsh, P. J.; Gordon, K. C.;

Schmidt-Mende, L.; Nazeeruddin, M. K.; Wang, Q.; Grätzel, M.; Officer, D. L., The Journal of Physical Chemistry C 2007, 111 (32), 11760-11762. 15. Yella, A.; Lee, H.-W.; Tsao, H. N.; Yi, C.; Chandiran, A. K.; Nazeeruddin, M. K.; Diau, E. W.-G.; Yeh, C.-Y.; Zakeeruddin, S. M.; Grätzel, M., Science 2011, 334 (6056), 629-634. 16. Hayashi, S.; Tanaka, M.; Hayashi, H.; Eu, S.; Umeyama, T.; Matano, Y.; Araki, Y.; Imahori, H., The Journal of Physical Chemistry C 2008, 112 (39), 15576-15585; Imahori, H.; Umeyama, T.; Ito, S., Accounts of Chemical Research 2009, 42 (11), 1809-1818. 17. Kira, A.; Matsubara, Y.; Iijima, H.; Umeyama, T.; Matano, Y.; Ito, S.; Niemi, M.; Tkachenko, N.; Lemmetyinen, H.; Imahori, H., Journal of Physical Chemistry C 2010, 114 (25), 11293-11304. 18. Deshpande, R.; Jiang, L.; Schmidt, G.; Rakovan, J.; Wang, X.; Wheeler, K.; Wang, H., Organic Letters 2009, 11 (19), 4251-4253. 19. Deshpande, R.; Wang, B.; Dai, L.; Jiang, L.; Hartley, C. S.; Zou, S.; Wang, H.; Kerr, L., Chemistry – An Asian Journal 2012, 7 (11), 2662-2669. 20. Anderson, H. L., Chem. Communs 1999, (23), 2323-2330; Ceulemans, A.; Oldenhof, W.; Gorllerwalrand, C.; Vanquickenborne, L. G., J. Am. Chem. Soc 1986, 108 (6), 1155-1163. 21. Lee, C.-W.; Lu, H.-P.; Lan, C.-M.; Huang, Y.-L.; Liang, Y.-R.; Yen, W.-N.; Liu, Y.-C.; Lin, Y.-S.; Diau, E. W.-G.; Yeh, C.-Y., Chemistry – A European Journal 2009, 15 (6), 1403-1412; Lo, C.-F.; Hsu, S.-J.; Wang, C.-L.; Cheng, Y.-H.; Lu, H.-P.; Diau, E. W.-G.; Lin, C.-Y., The Journal of Physical Chemistry C 2010, 114 (27), 12018-12023; Lin, C.-Y.; Lo, C.-F.; Luo, L.; Lu, H.-P.; Hung, C.-S.; Diau, E. W.-G., The Journal of Physical Chemistry C 2008, 113 (2), 755-764; Lee, C. Y.; She, C.; Jeong, N. C.; Hupp, J. T., Chemical Communications 2010, 46 (33), 6090-6092; Hsieh, C.-P.; Lu, H.-P.; Chiu, C.-L.; Lee, C.-W.; Chuang, S.-H.; Mai, C.-L.; Yen, W.-N.; Hsu, S.-J.; Diau, E. W.-G.; Yeh, C.-Y., Journal of Materials Chemistry 2010, 20 (6), 1127-1134.

Chapter 3：Triphenylene-Fused Porphyrins

Abstract

Triphenylene has been successfully fused to porphyrin periphery through a convenient Pd0-catalyzed cascade reaction followed by oxidative ring closure reaction. Bis-triphenylene-fused porphyrins and a dibenzo[fg,op]tetracene-fused porphyrin have also been obtained using a similar approach. These -extended porphyrins exhibited much broadened and bathochromic shifted UV-Visible absorptions, as compared with their unfused precursors.

Papers published from this chapter’s work:

1) L. Jiang, J. T. Engle, L. Sirk, C.S. Hartley, C. J. Ziegler, and H. Wang. Organic Letters,

2011, 13(12), 3020-3023.

3.1 Introduction

Porphyrins fused with external aromatic rings show promise in a broad range of applications in various areas, including near-infrared dyes,1 molecular devices,2 organic light-emitting diodes,3 and biosensors.4 In particular, porphyrins fused with polycyclic aromatic hydrocarbons (PAHs) hold great potential in the fast growing areas of molecular electronics and nanotechnology as heteroatom containing molecular graphene mimics.5, 6 As such, the development of new methodologies to fuse aromatic rings to the porphyrin periphery has remained an intense area of interest. The oxidative ring closure reaction has been proven to be an effective approach to construct polycyclic aromatic hydrocarbons.7 Recently, this approach has also been utilized in porphyrins. Porphyrins fused with anthracene,8 pyrene,9 azulene,10 and other porphyrins6, 11 have been reported. In these π-expanded porphyrins, an aromatic fragment is designed to match the geometry of the porphyrin, such that the fusion occurs at the adjacent porphyrin meso- and β-positions. In this work, we use a different strategy. The polycyclic aromatic hydrocarbons are introduced at the porphyrin β, β’- positions through a convenient oxidative coupling reaction. In this chapter, I will discuss the synthesis and characterization of the first examples of triphenylene- and dibenzo[fg,op]tetracene-fused porphyrins. Despite recent progress,12 most of the syntheses of β, β’-fused extended porphyrins require harsh conditions and/or are lengthy. Recently, we developed a concise method to prepare functionalized benzoporphyrins (Scheme 3.1).13 In this method, an alkene reacts with a β, β’ –dibromoporphyrin through a three-step cascade reaction involving a vicinal 2-fold Heck reaction, 6-π electrocyclization, and subsequent aromatization. This method made a new entry to access a number of functionalized benzoporphyrins. We envisioned that two vicinal aromatic rings could be introduced to the fused benzene ring(s) via this approach, and more benzene rings

43 could then be fused to the porphyrin through the well-established oxidative ring-closure reaction.

Scheme 3.1 Synthesis of opp-dibenzoporphyrin via Pd0-catalyzed cascade reaction

3.2 Result and Discussion

In order to investigate the viability of this approach, we prepared 2,3-dibromo-12-nitroporphyrin 3a-3d (Scheme 3.2). Thus, 3a-3d reacted with 3-methoxystyrene in the presence of a palladium catalyst to give aryl-substituted nitrobenzoporphyrin 4a-4d in 40-50% yields. Treatment of 4a-4d with excess of

FeCl3 in a mixture of nitromethane and dichloromethane (DCM) lead to the formation of the triphenylene fused porphyrin 5a-5d in 35-40% yields. Another compound (6a-6d) was also isolated in 7-12% yield. When Zn(II) porphyrins (4b and 4d) underwent oxidative coupling reactions, exclusive demetalation occurred, resulting in the formation of free base triphenylenoporphyrins (5b, 5d, 6b and 6d). The structure of 5a-5d was determined by 1H and 13C NMR spectroscopy, and LDI-TOF MS, and was further confirmed by a X-ray crystal structure of 5a (Figure 3.1). The crystal structure of 5a reveals that the , -pyrrole-fused triphenylene moiety retains its planarity, while the porphyrin core adopts a slightly saddled conformation.

Scheme 3.2 Synthesis of mono triphenylene-fused porphyrins

Figure 3.1 X-ray crystal structure of 5a: top, edge view; bottom, top view.

Initially, the identity of 6a-6d could not be determined due to the structural complexity. 5 and 6 carry exactly the same molecular weight, and the UV-Visible absorptions of these two compounds are also similar. However, the 1H NMR spectra of these two compounds display different patterns of proton shifts. We speculated that two types of oxidative couplings occurred to porphyrin 4a-4d. To further investigate the nature of this oxidative coupling reaction, we prepared a symmetrical benzoporphyrin 4e from a simpler dibromoporphyrin 3e. 1H NMR spectra of 6e, which exhibited much simpler proton shift pattern as compared with those of 6a-6d, suggest that 6e was resulted from an unsymmetrical ring closure of the two adjacent methoxyphenyl substituents (Scheme 3.2). A X-ray crystal structure of 6e further confirmed our speculation (see Experimental part). The single crystal X-ray diffraction data were colleceted and analyzed by Dr. Christopher J. Ziegler’s group at

University of Akron. This type of unsymmetrical oxidative ring closure is not favored due to the expected steric hindrance generated by the methoxy group and the ring proton on the neighbouring benzene ring, and has never been reported previously to our knowledge. We believe the observed unsymmetrical ring closure is caused by the competition arose from another steric hindrance introduced by the alkyl groups on the meso-phenyl rings and the methoxy group on the fused benzene ring. Having established the feasibility of this approach to fusing one triphenylene to porphyrins, we prepared porphyrins fused with two triphenylenes to further extend the porphyrin -system, starting from tetrabromoporphyrin 7 (Scheme 3.3). Similar to the mono-fused systems, unsymmetrical ring closure product was also isolated (10) in 14% yield. The symmetrical 9 was obtained in 48% yield. Other possible unsymmetrical ring closure products were not detected.

Scheme 3.3 Synthesis of bis-triphenylene-fused porphyrins

The geometry of triphenylene fused porphyrins 5 provides an ideal template to conveniently fuse more aromatic rings, if proper substituents are installed on the fused rings. Thus, we prepared monobenzoporphyrin 11 bearing two methoxy groups on the phenyl substituents (Scheme 3.4). 11 was treated with 1,2-dimethoxybenzene in the presence of FeCl3 as the oxidant, leading to the formation of the desired dibenzo[fg,op]tetracene-fused porphyrin 13 in 28% yield. Further investigation of this reaction suggests that, as expected, this one pot reaction involves two steps: first, formation of the triphenylene-fused 12; second, fusion of 1,2-dimethoxybenzene to the triphenylene moiety of 12.

Scheme 3.4 Synthesis of dibenzo[fg,op]tetracene-fused porphyrin

UV-Visible absorption spectra of 4e, 5e, 5c, 9 and 13 are shown in Figure 3.2.

The C2v symmetrical monobenzoporphyrin 4e showed an intense Soret band at 434 nm; upon fusion of the triphenylene moiety (5e), the Soret band was broadened and red-shifted to 447 nm; further extension of the -system (13) pushed the Soret band to 457 nm. When two triphenylenes were fused (9), the UV-Visible spectra displayed an interesting shouldered Soret band (absorbed at 469 nm with a shoulder at 443 nm), a rare observation for Ni(II) porphyrins with a D2h symmetry. All the triphenylene-fused porphyrins exhibited broadened, red-shifted, and more intense Q bands, typical of -extended porphyrins.

1.00

4e 0.75 5e 5c 0.50 13

Absorption 9 0.25

0.00 400 450 500 550 600 650 700 750 800 Wavelength

Figure 3.2 UV-Vis absorption spectra of 4e, 5e, 5c, 9 and 13 in CH2Cl2.

Density function calculations (DFT at B3LYP-631Gdp) were conducted by Dr. Scott Hartley of Miami University for 5e, 5c, 13 and 9 to better understand their structures (Figure 3.3). Similar to the crystal structure (5a), the , -pyrrole-fused triphenylene moiety in 5c is essentially planar in the calculated structure; the porphyrin core is, however, a little more distorted from planarity as compared with the crystal structure, adopting a conformation somewhere between a saddled and a ruffled distortion. In the absence of a nitro group at the porphyrin - position, 5e shows a less distorted structure with a ruffled conformation. In the more -extended 13, the fused polycyclic aromatic hydrocarbon slightly deviates from planarity, likely due to the full occupation of the methoxy groups at the bay positions, and the porphyrin core again adopts a ruffled conformation. It should be noted that, while both of the triphenylene moieties in the bis-triphenylene-fused porphyrin 9 retain their planarity, the porphyrin core displays a ruffled conformation with significant distortion from planarity. As a result, the two planar triphenylene moieties do not lie in the same plane. This distortion causing the loss of its D2h symmetry, explains, at least partially, for its observed shouldered and broadened Soret band.

Figure 3.3 Optimized molecular structures of 5e, 5c, 9 and 13 (DFT B3LYP-631Gdp).

3.3 Conclusion

In summary, we have synthesized the first examples of triphenylene-fused porphyrins through a Pd0-catalyzed cascade reaction followed by oxidative ring closure. Further extension of the porphyrin π -system to fuse one dibenzo[fg,op]tetracene has also been achieved using a similar approach. Unusual unsymmetrical ring closure products were obtained, likely resulting from two competing steric effects. These π-extended porphyrins displayed much broadened and bathochromic shifted UV-visible absorptions, as compared with their unfused precursors. Structural optimization suggests that the two planar triphenylene moieties of the bistriphenylene- fused porphyrin (9) do not lie in the same plane due to significant distortion of the porphyrin core, breaking the expected D2h symmetry.

Reference： 1. Schwab, P. F. H.; Levin, M. D.; Michl, J., Chemical Reviews 1999, 99 (7), 1863-1934; Tsuda, A.; Osuka, A., Science 2001, 293 (5527), 79-82. 2. Linke-Schaetzel, M.; Anson, C. E.; Powell, A. K.; Buth, G.; Palomares, E.; Durrant, J. D.;

Balaban, T. S.; Lehn, J.-M., Chemistry – A European Journal 2006, 12 (7), 1931-1940; Xu, H.; Wang, Y.; Yu, G.; Xu, W.; Song, Y.; Zhang, D.; Liu, Y.; Zhu, D., Chemical Physics Letters 2005, 414 (4–6), 369-373; Liu, Z.; Yasseri, A. A.; Lindsey, J. S.; Bocian, D. F., Science 2003, 302 (5650), 1543-1545. 3. Borek, C.; Hanson, K.; Djurovich, P.; Thompson, M.; Aznavour, K.; Bau, R.; Sun, Y.; Forrest, S.; Brooks, J.; Michalski, L.; Brown, J., Angewandte Chemie-International Edition 2007, 46 (7), 1109-1112; Baluschev, S.; Yakutkin, V.; Miteva, T.; Avlasevich, Y.; Chernov, S.; Aleshchenkov, S.; Nelles, G.; Cheprakov, A.; Yasuda, A.; Mullen, K.; Wegner, G., Angewandte Chemie-International Edition 2007, 46 (40), 7693-7696. 4. Akhigbe, J.; Zeller, M.; Bruckner,̈ C., Organic Letters 2011, 13 (6), 1322-1325; Chen, W.; Ding, Y.; Akhigbe, J.; Brückner, C.; Li, C. M.; Lei, Y., Biosensors and Bioelectronics 2010, 26 (2), 504-510. 5. Matsumoto, S.; Qu, S.; Kobayashi, T.; Kanehiro, M.; Akazome, M.; Ogura, K., heterocycles 2010, 80, 645-656; Yamada, H.; Kuzuhara, D.; Ohkubo, K.; Takahashi, T.; Okujima, T.; Uno, H.; Ono, N.; Fukuzumi, S., Journal of Materials Chemistry 2010, 20 (15), 3011-3024; Tanaka, T.; Nakamura, Y.; Osuka, A., Chemistry-a European Journal 2008, 14 (1), 204-211. 6. Nakamura, Y.; Jang, S.; Tanaka, T.; Aratani, N.; Lim, J.; Kim, K.; Kim, D.; Osuka, A., Chemistry-a European Journal 2008, 14 (27), 8279-8289. 7. Müller, M.; Kübel, C.; Müllen, K., Chemistry – A European Journal 1998, 4 (11), 2099-2109; Ashenhurst, J. A., Chemical Society Reviews 2010, 39 (2), 540-548. 8. Davis, N. K. S.; Pawlicki, M.; Anderson, H. L., Organic Letters 2008, 10 (18), 3945-3947; Davis, N.; Thompson, A.; Anderson, H., Journal of the American Chemical Society 2011, 133 (1), 30-31. 9. Diev, V.; Hanson, K.; Zimmerman, J.; Forrest, S.; Thompson, M., Angewandte Chemie-International Edition 2010, 49 (32), 5523-5526. 10. Kurotobi, K.; Kim, K. S.; Noh, S. B.; Kim, D.; Osuka, A., Angewandte Chemie International Edition 2006, 45 (24), 3944-3947. 11. Kamo, M.; Tsuda, A.; Nakamura, Y.; Aratani, N.; Furukawa, K.; Kato, T.; Osuka, A., Organic Letters 2003, 5 (12), 2079-2082. 12. Ito, S.; Murashima, T.; Uno, H.; Ono, N., Chemical Communications 1998, (16), 1661-1662; Filatov, M.; Lebedev, A.; Vinogradov, S.; Cheprakov, A., Journal of Organic Chemistry 2008, 73 (11), 4175-4185; Vicente, M.; Jaquinod, L.; Khoury, R.; Madrona, A.; Smith, K., Tetrahedron Letters 1999, 40 (50), 8763-8766; Jiao, L.; Hao, E.; Fronczek, F.; Vicente, M.; Smith, K., Chemical Communications 2006, (37), 3900-3902; Silva, A. M. G.; de Oliveira, K. T.; Faustino, M. A. F.; Neves, M. G. P. M. S.; Tomé, A. C.; Silva, A. M. S.; Cavaleiro, J. A. S.; Brandão, P.; Felix, V., European Journal of Organic Chemistry 2008, 2008 (4), 704-712; Lee, S. H.; Smith, K. M., Tetrahedron Letters 2005, 46 (12), 2009-2013. 13. Deshpande, R.; Jiang, L.; Schmidt, G.; Rakovan, J.; Wang, X.; Wheeler, K.; Wang, H., Organic Letters 2009, 11 (19), 4251-4253. 14. Jaquinod, L.; Khoury, R. G.; Shea, K. M.; Smith, K. M., Tetrahedron 1999, 55 (46), 13151-13158.

3.4 Experimental Part

General All solvents were Analytical Reagent grade unless otherwise stated and were obtained either from Sigma-Aldrich or ACROS. Analytical TLC’s were performed on Silicycle UltraPure Silica Gel 60 F254 TLC plates. Preparative column chromatography was performed on silica gel (40-60m), which was purchased from Silicycle. 1H and 13C experiments were conducted on a Bruker Avance 500 spectrometer. All samples were prepared either in CDCl3 and chemical shifts were referenced to CHCl3 at 7.24ppm

1 13 for H NMR and referenced to the CDCl3 at 77 ppm for C-NMR or in DMSO-d6 and chemical shift was referenced at 2.50 ppm. UV-Visible spectra were recorded on an Agilent 8453 UV-Visible spectrometer in CH2Cl2. Mass spectra were obtained on Bruker LDI-TOF mass spectrometer. M.P.’s were measured on an Electrothermal MEL-TEMP apparatus and were uncorrected. General procedure for the synthesis of substituted monobenzoporphyrins Metallated dibromoporphyrins 3a, 3b, 3c, 3d, 3e were prepared according to published procedures.14

Dibromoarylporphyrins (0.045 mmol), palladium acetate (0.011 mmol), triphenylphosphine (0.030 mmol) and K2CO3 (0.10 mmol) were added to a schlenk tube and dried under vacuum. The vacuum was released under argon to allow the addition of dry DMF (10 mL) and dry xylene (10 mL) and 3-methoxystyrene (25-fold excess) or 3, 4-dimethoxystyrene. The mixture was then degassed via four freeze-pump-thaw cycles before the vessel was purged with argon again. The schlenk flask was sealed and heated to reflux for 60h. After 60 h, the mixture was diluted with

CHCl3 and washed with water for 3 times. The organic layer was removed under vacuum. The residue was subjected to preparative column chromatography. The bands containing the desired porphyrins were collected and recrystallized from

CHCl3/MeOH. This procedure was used to prepare 4a, 4b, 4c, 4d, 4e, and 11.

o 3a: mp > 320 C. UV-Vis λmax (CH2Cl2)/nm 441 (log ε 5.48), 555 (4.27), 601 (4.38);

1 H-NMR (500 MHz, CDCl3, Me4Si) δ 8.87 (1 H, s, β-H), 8.44-8.61 (4H, m, β-H),

13 7.70-7.85 (8H, m), 7.41-7.49 (8H, m), 2.58-2.62 (12H, m, CH3-H); C-NMR (500

MHz, CDCl3, Me4Si) δ 21.45, 21.57, 21.61, 29.71, 118.43, 119.18, 119.26, 123.10, 125.37, 125.76, 127.97, 128.07, 128.17, 128.20, 128.44, 133.06, 133.51, 133.63, 133.68, 133.92, 133.96, 134.11, 135.95, 136.08, 136.10, 137.12, 137.74, 138.39, 138.42, 138.53, 138.57, 143.33, 145.70, 146.21, 146.42, 150.74; Calculated Mass, 930.31, Found MS (LDI-TOF), m/z 929.82.

o 3b: mp > 320 C. UV-Vis λmax (CH2Cl2)/nm 436 (log ε 5.49), 567 (4.35), 616 (4.41);

1 H-NMR (500 MHz, DMSO-D6, Me4Si) δ 8.92 (1 H, s, β-H), 8.59-8.72 (4H, m, β-H),

7.87-8.05 (8H, m), 7.46-7.60 (8H, m), 2.65 (12H, m, CH3-H), the solubility in DMSO is too low to do a 13C-NMR. Calculated Mass, 937.00, Found MS (LDI-TOF), m/z 936.88.

o 3c: mp > 320 C. UV-Vis λmax (CH2Cl2)/nm 441 (log ε 5.53), 550 (4.27), 598 (4.31);

1 H-NMR (500 MHz, CDCl3, Me4Si) δ 8.90 (1H, s, β-H) 8.62 (2H, d, J = 5Hz, β-H), 8.50 (2H, d, J = 5Hz, β-H), 7.86-7.89 (4H, m), 7.73-7.75 (4H, m), 7.46-7.55 (8H, m),

13 3.13-3.19 (4H, m, CH-H), 1.43-1.45 (24H, m, CH3-H); C-NMR (500 MHz, CDCl3,

Me4Si) δ 24.10， 24.15， 24.28， 33.97， 34.12，34.13, 118.53， 119.25， 119.33， 123.23， 125.33, 125.43, 125.46, 125.56, 125.58, 125.68, 128.55, 133.14, 133.54, 133.76, 133.93, 134.02, 134.11, 134.18, 136.27, 136.34, 136.36, 137.16, 137.85, 143.41, 145.75, 146.36, 149.38, 149.56；Calculated Mass, 1042.52, Found MS

(LDI-TOF), m/z 1042.92.

o 3d: mp > 320 C. UV-Vis λmax (CH2Cl2)/nm 441 (log ε 5.37), 575 (4.20), 624 (4.29);

1 H-NMR (500 MHz, DMSO-D6, Me4Si) δ 8.95 (1H, m, β-H), 8.64-8.73 (4H, m, β-H), 7.89-8.09 (8H, m), 7.52-7.67 (8H, m), 3.18-3.25 (4H, m, CH-H), 1.42-1.48 (24H, m,

13 CH3-H); the solubility in DMSO is too low to obtain a C-NMR spectrum; Calculated Mass, 1049.21, Found MS (LDI-TOF), m/z 1049.30.

o 3e: mp > 320 C. UV-Vis λmax (CH2Cl2)/nm 428 (log ε 5.43), 542 (4.23), 886 (4.10);

1 H-NMR (500 MHz, CDCl3, Me4Si) δ 8.66-8.78 (6H, m, β-H), 7.92 (4H, d, J = 8.0 Hz), 7.67 (4H, d, J = 8.0 Hz), 7.50-7.56 (8H, m), 3.18-3.23 (4H, m, CH-H), 1.48-1.51

13 (24H, m, CH3-H); C-NMR (500 MHz, CDCl3, Me4Si) δ 24.19, 24.32, 34.07, 34.11, 117.91, 119.53, 124.04, 125.16, 125.21, 132.31, 132.64, 133.56, 133.76, 133.84, 135.78, 137.12, 137.49, 143.31, 143.34, 143.44, 148.51, 149.10; Calculated Mass, 997.52, Found MS (LDI-TOF), m/z 997.35.

o 4a: Yield 47%, mp > 320 C. UV-Vis λmax (CH2Cl2)/nm 452 (log ε 5.50), 557 (4.62),

1 598 (4.54); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.83 (1 H, s, β-H), 8.49-8.63 (4H, m, β-H), 7.77-7.84 (8H, m, o-Ph–H), 7.39-7.53 (8 H, m, m-Ph–H), 7.04-7.13 (4 H, m),

13 6.58-6.75 (6H, m), 3.63(6 H, d,-OCH3 ), 2.57-2.61(12H, m, (CH3)-H); C-NMR (500

MHz, CDCl3, Me4Si) δ 21.35, 21.45, 29.71, 55.05, 112.51, 115.61, 122.53, 123.14, 126.61, 126.82, 127.18, 127.97, 128.18, 128.67, 129.01, 131.08, 131.71, 132.47, 133.60, 134.46, 136.24, 137.33, 138.19, 138.67, 139.28, 139.72, 140.61, 140.82, 142.87, 146.75, 149.18, 159.09; Calculated Mass, 1034.82, Found MS (LDI-TOF), m/z 1033.29.

o 4b: Yield 50%, mp > 320 C. UV-Vis λmax (CH2Cl2)/nm 449 (log ε 5.43), 570 (4.50),

1 616 (4.32); H-NMR (500 MHz, CDCl3, Me4Si) δ 9.02 (1 H, s, β-H), 8.72-8.87 (4H, m, β-H), 7.98-8.04 (8H, m, o-Ph–H), 7.45-7.58 (8 H, m, m-Ph–H), 7.14-7.20 (4 H, m),

6.61-6.76 (6H, m), 3.62(6 H, s,-OCH3 ), 2.63-2.67(12H, m, (CH3)-H); The solubility is very bad; Calculated Mass, 1041.52, Found MS (LDI-TOF), m/z 1042.53.

o 4c: Yield: 40%, mp > 320 C. UV-Vis λmax (CH2Cl2)/nm 449 (log ε 5.43), 570 (4.50),

1 616 (4.32); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.86 (1 H, s, β-H), 8.52-8.86 (4H, m, β-H), 7.55-7.90 (8H, m, o-Ph–H), 7.46-7.54 (8 H, m, m-Ph–H), 7.08-7.13 (4 H, m), 6.63-6.71 (4H, m), 6.49-6.50(2 H, d, J = 6 Hz ), 6.55 (6H, d, J = 6.5 Hz, OMe-H),

13 3.12-3.16 (4H, m, (CH)-H), 1.35-1.48 (24H, m, CH3-H); C-NMR (500 MHz, CDCl3,

Me4Si) δ 24.10, 24.11, 24.14, 24.19, 33.97, 34.12, 34.16, 54.95, 76.78, 77.03, 77.28, 112.42, 115.54, 115.65, 115.69, 116.18,119.50, 122.23, 122.25, 123.42, 125.32, 125.48, 126.24, 126.28, 126.32, 126.48, 127.23, 128.85, 131.14, 131.83, 132.43, 132.57, 132.97, 133.39, 133.60, 133.76, 134.49, 135.64, 135.55, 136.55, 136.58, 137.43, 137.53, 138.60, 138.74, 139.28, 139.73, 140.04, 140.76, 140.98, 142.99, 143.01, 143.35, 146.75, 147.03, 149.06, 149.14, 149.32, 149.60, 158.87; Calculated Mass, 1147.03, Found MS (LDI-TOF), m/z 1146.61.

o 4d: Yield: 41%, mp > 320 C. UV-Vis λmax (CH2Cl2)/nm 449 (log ε 5.35), 569 (4.42),

1 614 (4.19); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.98 (1 H, s, β-H), 8.76-8.81 (4H, m, β-H), 8.61-8.62 (6H, m), 7.50-7.59 (6 H, m), 7.31-7.33 (2 H, d, J = 11.5 Hz), 7.11-7.13 (2H, m), 6.69-6.73 (4H, m), 6.55 (2H, s), 6.34-6.35 (2H, m), 3.57(6 H,

13 s,-OCH3 ), 3.16-3.25 (4H, m, (CH)-H), 1.41-1.54 (24H, m, (CH3)-H); C-NMR (500

MHz, CDCl3, Me4Si) δ 24.19, 24.25, 24.30, 33.99, 34.13, 34.16, 54.96, 112.29, 115.15, 115.80, 117.47, 117.80, 120.76, 121.73, 122.42, 124.79, 125.78, 127.08, 127.15, 127.74, 128.85, 131.15, 132.93, 133.23, 133.38, 133.78, 134.59, 134.85, 137.61, 138.86, 138.94, 139.29, 139.73, 139.79, 140.38, 141.57, 143.41, 147.37, 147.90, 148.44, 148.55, 148.71, 149.14, 149.90, 150.17, 153.84, 153.93, 158.85; Calculated Mass, 1153.73, Found MS (LDI-TOF), m/z 1154.63.

o 4e: Yield: 42%, mp > 320 C. UV-Vis λmax (CH2Cl2)/nm 434 (log ε 5.39), 547 (4.32),

1 585 (4.11); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.86-8.72 (6 H, m, β-H), 7.91-7.93 (4H, d, J = 8.0, o-Ph-H), 7.87-7.89 (4H, d, J = 8.0, o-Ph-H), 7.54-7.56 (4H, d, J = 8.0, m-Ph-H), 7.51-7.52 (4H, d, J = 8.0, m-Ph-H), 7.21 (2H, s, the protons of the fused benzene ring), 7.10-7.13 (2H, m), 6.70-6.72 (4H, m), 6.54 (2H, s), 3.57 (6H, s,

13 OMe-H), 3.11-3.21 (4H, m, CH-H), 1.38-1.48 (24H, d, J = 6.9, CH3-H); C-NMR

(500 MHz, CDCl3, Me4Si) δ 24.16, 24.24, 34.09, 34.16, 54.95, 112.33, 115.24, 115.76, 120.18, 122.37, 125.07, 126.07, 126.19, 128.80, 130.96, 131.32, 132.51, 132.75, 133.65, 137.91, 137.93, 138.00, 138.32, 139.69, 140.78, 141.79, 143.38, 144.11, 148.32, 149.22, 158.82; Calculated Mass, 1100.45, Found MS (LDI-TOF), m/z 1100.51.

o 11: Yield: 35%, mp > 320 C. UV-Vis λmax (CH2Cl2)/nm 435 (log ε 5.38), 548 (4.26),

1 587 (4.02); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.68-8.72 (6H, m, β-H), 7.87-7.93 (8H, m), 7.50-7.54 (8H, m), 7.19 (2H, s), 6.70-6.77 (4H, s), 6.44-6.45 (2H, m), 3.86 (6H, s, OMe-H), 3.51 (6H, s, OMe-H), 3.10-3.20 (4H, m, CH-H), 1.38-1.48 (24H, m,

13 CH3-H); C-NMR (500 MHz, CDCl3, Me4Si) δ 149.12, 148.32, 147.98, 147.64, 144.12, 141.79, 140.76, 139.51, 138.45, 138.07, 137.90, 137.75, 135.03, 133.64, 132.81, 132.49, 131.31, 130.89, 126.02, 125.07, 121.80, 120.19, 115.13, 114.16, 110.98, 55.94, 55.55, 34.24, 34.08, 24.27, 24.23; Calculated Mass, 1160.48, Found MS (LDI-TOF), m/z 1160.50. General procedure for the synthesis of substituted opp-dibenzoporphyrins Metallated tetrabromoporphyrins 7 were prepared according to published

57 procedures.13

Tetrabromoarylporphyrin 7 (0.045 mmol), palladium acetate (0.023 mmol), triphenylphosphine (0.059 mmol) and K2CO3 (0.19 mmol) were added to schlenk tube and dried under vacuum. The vacuum was released under argon to allow the addition of dry DMF (10 mL) and dry xylene (10 mL) and m-methoxystyrene (50-fold excess). The mixture was then degassed via four freeze-pump-thaw cycles before the vessel was purged with argon again. The schlenk flask was sealed and heated to reflux for

72 h. After 72 h, the mixture was diluted with CHCl3 and washed with water. The organic layer was removed under vacuum. The residue was subjected to preparative column chromatography. The bands containing desired porphyrins were collected and recrystallized from CHCl3/MeOH. This procedure was used to prepare 8.

o 8: Yield: 38%, mp > 320 C. UV-Vis λmax (CH2Cl2)/nm 452 (log ε 5.46), 576 (4.58),

1 625 (4.44); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.57-8.71 (4H, m, β-H), 7.86-7.89 (5H, m), 7.53-7.58 (10H, m), 7.09-7.19 (8 H, m), 6.52-6.98 (10H, m), 3.55-3.71 (12H,

13 m, OMe), 3.11-3.22 (4H, m, CH), 1.36-1.43 (24H,m); C-NMR (500 MHz, CDCl3,

Me4Si) δ 24.16, 34.18, 54.95, 112.13, 112.31, 115.38, 116.50, 118.25, 122.38, 125.87, 125.95, 126.08, 128.78, 128.98, 129.13, 132.70, 132.87, 137.62, 138.00, 138.15, 143.38, 143.44, 149.26, 158.82, 159.51; Calculated Mass, 1364.34, Found MS

(LDI-TOF), m/z 1364.35. General procedure for the oxidative ring closure reactions

Starting porphyrins 4a-4e (0.17 mmol) were dissolved in dry dichloromethane (50 mL) under an argon atmosphere. A solution of ferric chloride (1.7 mmol) in nitromethane (8 mL) was added to the solution and the mixture was stirred for 2.5 hours at room temperature. During the reaction, a constant stream of argon was passed through the mixture. The reaction was monitored by UV-Vis (12-18 nm red-shifted) and TLC. To the resulting mixture was added methanol (5 mL) followed by water (80 mL) and dichloromethane (40 mL). The dichloromethane layer was separated and evaporated to produce a solid. The residue was subjected to preparative column chromatography. The bands containing the desired porphyrins were collected and recrystallized from

CHCl3/MeOH. This procedure was used to prepare 5a, 6a, 5b, 6b, 5c, 6c, 5d, 6d, 5e, and 6e.

o 5a: Yield: 40%, mp > 320 C. UV-Vis λmax (CH2Cl2)/nm 463 (log ε 5.56), 562 (4.70),

1 609 (4.66); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.79 (1 H, s, β-H), 8.58 (1H, d, J = 5 Hz, β-H), 8.51 (1H, d, J = 5 Hz, β-H), 8.41-8.46 (3H, m), 8.28-8.30 (2H, d, J = 9 Hz), 7.82-7.92 ( 8H, m), 7.62-7.65 (4H, m), 7.34-7.48 (7H, m), 7.13-7.16 (2H, m),

3.98 (3H, d, OMe-H), 3.95 (3H, d, OMe-H), 2.76 (6H, s, CH3), 2.57-2.61 (6H, d,

CH3); The solubility for this porphyrin is very low; Calculated Mass, 1033.31, Found MS (LDI-TOF), m/z 1032.28.

o 6a: Yield: 9%; mp > 320 C. UV-Vis λmax (CH2Cl2)/nm 462 (log ε 5.69), 564 (4.69),

1 609 (4.65); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.78-8.79 (1H, m, β-H), 8.43-8.62 (5H, m), 8.31-8.32 (1H, m), 7.80-7.92 (8H, m), 7.58-7.69 (5H, m), 7.39-7.45 (7H, m),

7.14-7.23 (2H, m), 3.98-4.07 (6H, m, OMe-H), 2.77-2.82 (6H, m, CH3-H), 2.56-2.60

(6H, m, CH3-H); Calculated Mass, 1033.31, Found MS (LDI-TOF), m/z 1033.52

(-NO2).

o 5b: Yield: 40%, mp > 320 C. UV-Vis λmax (CH2Cl2)/nm 465 (log ε 5.41), 547 (4.45),

1 609 (4.30); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.93 (1 H, s, β-H), 8.81-8.83 (2H, m, β-H), 8.57-8.60 (2H, m, β-H), 8.12-8.35 (11H, m), 7.26-7.73 (11H, m), 6.84-6.92

(2H, m), 3.96-3.98 (6H, d, OMe-H), 2.66-2.84 (12H, m, CH3), -2.23 - -2.17 (2H, d, free base H); The solubility for this porphyrin is too low to do a 13C-NMR; Calculated Mass, 976.13, Found MS (LDI-TOF), m/z 976.42.

o 6b: Yield: 10%, mp > 320 C. UV-Vis λmax (CH2Cl2)/nm 463 (log ε 5.52), 546 (4.54),

1 609 (4.39); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.72-8.96 (5 H, m, β-H), 8.42-8.55 (2H, m), 8.13-8.24 (8H, m), 7.73-7.83 (5H, m), 7.50-7.62 (7H, m), 7.20-7.21 (2H, m),

4.10-4.12 (6H, d, OMe-H), 2.90-2.91 (5H, CH3-H), 2.69-2.74 (7H, CH3-H), -2.05--2.02(2H, m, free base H); Calculated Mass, 976.27, Found MS (LDI-TOF), m/z 976.42.

o 5c: Yield: 35%, mp > 320 C. UV-Vis λmax (CH2Cl2)/nm 464 (log ε 5.43), 564

1 (4.49), 613 (4.47); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.82 (1 H, s, β-H), 8.39-8.60 (6H, m), 8.08-8.10 (2H, m), 7.88-7.93 (8H, m), 7.68-7.69 (4H, m), 7.53-7.54 ( 4H, m), 7.46-7.48 (2H, m), 6.97-7.03 (2H, m), 3.55-3.86 (6H, m, OMe-H), 3.14-3.35 (4H, m,

13 CH-H), 1.43-1.55 (24H, m, CH3); C-NMR (500 MHz, CDCl3, Me4Si) δ 23.83, 24.17, 33.82, 33.97, 114.86, 115.66, 119.64, 123.83, 125.50, 126.57, 127.80, 127.90, 130.58, 132.95, 133.34, 133.50, 133.87, 136.59, 136.73, 137.40, 137.54, 138.11, 138.50, 140.82, 141.62, 143.48, 149.08, 149.20, 149.46; Calculated Mass, 1145.02, Found MS

(LDI-TOF), m/z 1143.55.

o 6c: Yield: 7%; mp > 320 C. UV-Vis λmax (CH2Cl2)/nm 462 (log ε 5.45), 567 (4.51),

1 613 (4.48); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.80 (1H, s, -H), 8.43-8.61 (6H, m), 7.87-7.98 (8H, m), 7.69-7.73 (4H, m), 7.38-7.55 (8H, m), 7.15-7.17 (2H, m), 3.95-4.06 (6H, m), 3.14-3.35 (4H, m), 1.43-1.55 (24H, m) Calculated Mass, 1143.42, Found MS (LDI-TOF), m/z 1143.67.

o 5d: Yield: 36%, mp > 320 C. UV-Vis λmax (CH2Cl2)/nm 467 (log ε 5.38), 549 (4.40),

1 610 (4.21); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.97 (1 H, s, β-H), 8.80 (2H, m, β-H), 8.46-8.50 (2H, m, β-H), 8.09-8.28 (10H, m),7.61-7.75 (8H, m), 6.89-7.10 (4H, m), 6.49-6.51 (1H, m), 6.31-6.32 (1H, m), 3.86 (3H, s, OMe-H), 3.79 (3H, s, OMe-H),

3.22-3.41 (4H, m, CH-H), 1.52-1.57 (24H, m, CH3-H), -2.22 (1H, s, free base H),

13 -2.30 (1H, s, free base H); C-NMR (500 MHz, CDCl3, Me4Si) δ 23.86, 24.21, 24.28, 33.87, 34.02, 34.20, 55.13, 55.25, 108.39, 108.56, 112.22, 112.33, 112.63, 116.11, 116.72, 119.67, 119.83, 120.84, 122.76, 122.89, 123.06, 123.18, 124.60, 125.39, 125.47, 126.18, 126.21, 127.06, 127.76, 127.84, 128.00, 129.03, 129.66, 130.24, 130.33, 130.45, 134.09, 134.35, 135.31, 135.58, 136.91, 138.27, 138.40, 138.48, 139.17, 139.31, 140.10, 142.16, 142.28, 142.61, 145.38, 149.20, 149.36, 149.41, 151.68, 152.82; Calculated Mass, 1088.34, Found MS (LDI-TOF), m/z 1088.48.

o 6d: Yield: 8%, mp > 320 C. UV-Vis λmax (CH2Cl2)/nm 464 (log ε 5.37), 547 (4.33),

1 609 (4.14); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.84-8.92 (3 H, m, β-H), 8.65-8.66 (2H, m, β-H), 8.52-8.56 (2H, m), 8.14-8.24 (8H, m), 7.77-7.79 (4H, m), 7.28-7.64 (7H, m), 7.10-7.16 (3H, m), 4.04 (3H, s, OMe-H), 3.99 (3H, s, OMe-H), 3.21-3.42 (4H, m,

CH-H), 1.48-1.54 (24H, m, CH3-H), -2.03--1.97(2H, m, free base H); Calculated Mass, 1088.34, Found MS (LDI-TOF), m/z 1088.54.

o 5e: Yield: 50%, mp > 320 C. UV-Vis λmax (CH2Cl2)/nm 447 (log ε 5.41), 557 (4.40),

1 598 (4.23); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.62-8.66 (8 H, m, β-H + d), 8.32 (2H, d, J = 9.0 Hz, b), 7.90-7.97 (8H, d, J = 7.5 Hz, o-Ph-H), 7.69-7.71 (4H, d, J = 7.5 Hz, m-Ph-H), 7.50-7.51 (4H, d, J = 8 Hz, m-Ph-H), 7.43 (2H, s, c), 7.12-7.14 (2H, m,

61 a), 3.93 (6H, s, OMe-H), 3.14-3.20 (4H, m, CH-H), 1.46-1.57 (24H, d, J = 6.5 Hz,

13 CH3-H); C-NMR (500 MHz, CDCl3, Me4Si) δ 23.84, 24.20, 26.91, 29.69, 31.58, 33.85, 34.05, 55.41, 109.36, 113.10, 114.60, 119.36, 120.47, 124.11, 124.23, 125.06, 126.36, 127.56, 130.74, 131.04, 131.21, 132.52, 133.02, 133.61, 137.83, 138.29, 148.31, 149.14; Calculated Mass, 1100.02, Found MS (LDI-TOF), m/z 1099.64.

o 6e: Yield: 12%, mp > 320 C. UV-Vis λmax (CH2Cl2)/nm 445 (log ε 5.38), 555 (4.38),

1 597 (4.21); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.62-8.65 (6H, m), 7.88-7.99 (7H, m), 7.69-7.72 (5H, m), 7.38-7.62 (7H, m), 7.03-7.18 (3H, m), 3.97-4.06 (4H, m,

OMe-H), 3.14-3.47 (4H, m, CH-H), 1.45-1.55 (24H, m, CH3-H); Calculated Mass, 1100.02, Found MS (LDI-TOF), m/z 1100.26.

Porphyrin 8 (0.35 mmol) was dissolved in dry dichloromethane (60 mL) under an argon atmosphere. A solution of ferric chloride (7 mmol) in nitromethane (10 mL) was added to the solution and the mixture was stirred for 2.5 hours at room temperature to form 12 first. During the reaction, a constant stream of argon was passed through the mixture. The reaction was monitored by UV-Vis (18 nm red-shifted, 452 nm to 470 nm) and TLC. To the resulting mixture was added methanol (10 mL) followed by water (100 mL) and dichloromethane (60 mL). The dichloromethane layer was separated and evaporated to produce a solid. The residue

62 was subjected to preparative column chromatography. The bands containing the desired porphyrins were collected and recrystallized from CHCl3/MeOH. This procedure was used to prepare 9 and 10.

o 9: Yield: 48%, mp > 320 C. UV-Vis λmax (CH2Cl2)/nm 469 (log ε 5.39)，595 (4.47),

1 655 (4.32); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.58-8.61 (8H, m), 8.27-8.29 (4H, m), 7.99 (8H, d, J = 8.0 Hz, o-Ph-H), 7.69 (8H, d, J = 8.0, m-Ph-H), 7.237 (4H, s), 7.09-7.11 (4H, m), 3.92 (12H, s, OMe-H), 3.31-3.38 (4H, m, CH-H), 1.55 (24H, d, J =

13 7.0 Hz, CH3-H); C-NMR (500 MHz, CDCl3, Me4Si) δ 23.87, 29.71, 33.89, 55.37, 109.09, 110.03, 113.11, 116.37, 118.75, 124.05, 124.10, 126.33, 127.22, 130.73, 131.21, 133.06, 138.00, 149.16, 158.10; Calculated Mass, 1360.31, Found MS (LDI-TOF), m/z 1360.58.

o 10: Yield: 14%, mp > 320 C. UV-Vis λmax (CH2Cl2)/nm 470 (log ε 5.62), 593 (4.73),

1 657 (4.60); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.59-8.64 (6H, m), 8.31-8.33 (2H, m), 7.98-7.99 (8H, m), 7.69-7.71 (9H, m), 7.37-7.62 (6H, m), 7.12-7.14 (5H, m),

3.94-4.05 (12H, m, OMe-H), 3.32-3.39 (4H, m, CH-H), 1.54-1.55 (24H, CH3-H);

13 C-NMR (500 MHz, CDCl3, Me4Si) δ 22.71, 23.88, 24.48, 29.38, 29.72, 33.90, 34.28, 55.27, 55.37, 55.98, 108.89, 109.12, 110.03, 112.10, 113.10, 116.07, 116.49, 120.86, 124.04, 126.08, 126.34, 126.42, 130.67, 131.21, 132.37, 133.02, 134.98, 138.04, 149.17, 149.25, 158.09; Calculated Mass, 1360.31, Found MS (LDI-TOF), m/z 1360.56.

Porphyrin 11 (0.17 mmol) was dissolved in dry dichloromethane (50 mL) under an argon atmosphere. A solution of ferric chloride (1.7 mmol) in nitromethane (8 mL) was added to the solution and the mixture was stirred for 2.5 hours at room temperature. During the reaction, a constant stream of argon was passed through the mixture. The reaction was monitored by UV-Vis (12-18 nm red-shifted) and TLC. Then dimethoxybenzene (0.34 mmol) was added to the solution and the mixture was stirred for one day. The reaction was monitored by TLC. To the resulting mixture was added methanol (5 mL) followed by water (80 mL) and dichloromethane (40 mL). The dichloromethane layer was separated and evaporated to produce a solid. The residue was subjected to preparative column chromatography. The band containing desired porphyrins were collected and recrystallized from CHCl3/MeOH. This

64 procedure was used to prepare 13.

o 13: Yield: 28%, mp > 320 C. UV-Vis λmax (CH2Cl2)/nm 457 (log ε 5.39), 569 (4.43),

1 613 (4.34); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.35-8.74 (6H, m, β-H ), 7.91-7.92 (6H, m), 7.75 (2H, m), 7.42-7.61 (10H, m), 7.00-7.02 (2H, m), 6.40-6.69 (1H, m), 6.39 (1H, m), 4.01-4.11 (12H, m, OMe-H), 3.83-3.85 (3H, m, OMe-H), 3.59 (3H, m, OMe-H), 3.14-3.21 (3H, m, CH-H), 2.90-2.91 (1H, m, CH-H), 1.24-1.28 (24H, m,

CH3-H); Calculated Mass, 1294.20, Found MS (LDI-TOF), m/z 1294.30. X-ray crystal structures Crystals of compounds 5a and 6a (Scheme S1) suitable for X-ray diffraction analysis were obtained by slow evaporation from CHCl3/EtOH. Crystallographic data have been deposited with the Cambridge Crystallographic Data Centre (CCDC 819346 & 819347). Copies of the data can be obtained free of charge on application to the CCDC, 12 Union Road, Cambridge CB21EZ, UK (fax: (+44)-1223-336-033; e-mail: [email protected]. Experimental details for compound 5a: A black needle-like specimen of

C92H80N4NiO4, approximate dimensions 0.05 mm x 0.06 mm x 0.28 mm, was used for the X-ray crystallographic analysis. The X-ray intensity data were measured. The total exposure time was 43.69 hours. The frames were integrated with the Bruker SAINT software package using a narrow-frame algorithm. The integration of the data using a triclinic unit cell yielded a total of 21358 reflections to a maximum θ angle of 65.52° (0.85 Å resolution), of which 7836 were independent (average redundancy

2.726, completeness = 86.9%, Rint = 6.46%, Rsig = 7.20%) and 5684 (72.54%) were greater than 2σ(F2). The final cell constants of a = 13.2069(7) Å, b = 14.6494(7) Å, c = 15.7541(9) Å, α =72.654(4)°, β = 76.898(4)°, γ = 64.857(4)°, volume = 2616.0(2) Å3, are based upon the refinement of the XYZ-centroids of 8823 reflections above 20 σ(I) with 7.444° < 2θ < 129.4°. Data were corrected for absorption effects using the multi-scan method (SADABS). The ratio of minimum to maximum apparent transmission was 0.681. The calculated minimum and maximum transmission

65 coefficients (based on crystal size) are 0.5510 and 0.8928. The structure was solved and refined using the Bruker SHELXTL Software Package, using the space group P -1, with Z = 2 for the formula unit, C67H48Cl3N5NiO4. The final anisotropic full-matrix least-squares refinement on F2 with 762 variables converged at R1 =8.99%, for the observed data and wR2 = 24.62% for all data. The goodness-of-fit was 1.598. The largest peak in the final difference electron density synthesis was 1.185 e-/Å3 and the largest hole was -1.115 e-/Å3 with an RMS deviation of 0.113 e-/Å3. On the basis of the final model, the calculated density was 1.463 g/cm3 and F(000), 1192 e-. Tables S1 and S2 contain the crystal parameter, data collection and structure refinement data for 5a. Figure S1 shows the structure of the compound. Experimental details for compound 6a: A lustrous black block-like specimen of C66H47N5NiO4, approximate dimensions 0.16 mm x 0.18 mm x 0.20 mm, was used for the X-ray crystallographic analysis. The X-ray intensity data were measured. The total exposure time was 43.65 hours. The frames were integrated with the Bruker SAINT software package using a narrow-frame algorithm. The integration of the data using a triclinic unit cell yielded a total of 19965 reflections to a maximum θ angle of 63.42° (0.86 Å resolution), of which 7149 were independent (average redundancy 2.793, completeness = 82.0%, Rint = 4.71%, Rsig = 5.67%) and 4358 (60.96%) were greater than 2σ(F2). The final cell constants of a = 13.960(2) Å, b = 14.0133(11) Å, c = 16.0894(13) Å, α =112.417(5)°, β = 107.325(7)°, γ = 97.545(7)°, volume = 2668.6(5) Å3, are based upon the refinement of the XYZ-centroids of 158 reflections above 20 σ(I) with 10.97° < 2θ < 76.63°. Data were corrected for absorption effects using the multi-scan method (SADABS). The ratio of minimum to maximum apparent transmission was 0.824. The calculated minimum and maximum transmission coefficients (based on crystal size) are 0.8320 and 0.8639. The structure was solved and refined using the Bruker SHELXTL Software Package,

66 using the space group P -1, with Z = 2 for the formula unit, C66H47N5NiO4. The final anisotropic full-matrix least-squares refinement on F2 with 722 variables converged at R1 = 8.70%, for the observed data and wR2 = 26.99% for all data. The goodness-of-fit was 1.639. The largest peak in the final difference electron density synthesis was 0.524 e-/Å3 and the largest hole was -0.408 e-/Å3 with an RMS deviation of 0.075 e-/Å3. On the basis of the final model, the calculated density was 1.285 g/cm3 and F(000), 1076 e-. Tables S3 and S4 contain the crystal parameter, data collection and structure refinement data for 6a. Figure S2 shows the structure of the compound.

Table S1. Sample and crystal data for 5a.

Identification code 5a

Chemical formula C67H48Cl3N5NiO4

Formula weight 1152.16

Temperature 100(2) K

Wavelength 1.54178 Å

Crystal size 0.05 x 0.06 x 0.28 mm

Crystal habit black needle

Crystal system triclinic

Space group P -1

Unit cell dimensions a = 13.2069(7) Å α = 72.654(4)°

b = 14.6494(7) Å β = 76.898(4)°

c = 15.7541(9) Å γ = 64.857(4)°

Volume 2616.0(2) Å3

Z 2

Density (calculated) 1.463 Mg/cm3

Absorption coefficient 2.417 mm-1

F(000) 1192

Table S2. Data collection and structure refinement for 5a.

Theta range for data 3.72 to 65.52° collection

-15<=h<=15, -17<=k<=16, Index ranges -14<=l<=17

Reflections collected 21358

Independent 7836 [R(int) = 0.0646] reflections

Coverage of independent 86.9% reflections

Absorption correction multi-scan

Max. and min. 0.8928 and 0.5510 transmission

Structure solution direct methods technique

Structure solution SHELXS-97 (Sheldrick, 2008) program

Refinement method Full-matrix least-squares on F2

Refinement program SHELXL-97 (Sheldrick, 2008)

2 2 2 Function minimized Σ w(Fo - Fc )

Data / restraints / 7836 / 0 / 762 parameters

Goodness-of-fit on F2 1.598

5684 data; R1 = 0.0899, wR2 = Final R indices I>2σ(I) 0.2263

R1 = 0.1192, wR2 = all data 0.2462

2 2 2 w=1/[σ (Fo )+(0.1000P) +0.0000P] Weighting scheme 2 2 where P=(Fo +2Fc )/3

Largest diff. peak and 1.185 and -1.115 eÅ-3 hole

R.M.S. deviation from 0.113 eÅ-3 mean

Table S3. Sample and crystal data for 6a.

Identification code 6a

Chemical formula C66H47N5NiO4

Formula weight 1032.80

Temperature 100(2) K

Wavelength 1.54178 Å

Crystal size 0.16 x 0.18 x 0.20 mm

Crystal habit lustrous black block

Crystal system triclinic

Space group P -1

Unit cell dimensions a = 13.960(2) Å α = 112.417(5)°

b = 14.0133(11) Å β = 107.325(7)°

c = 16.0894(13) Å γ = 97.545(7)°

Volume 2668.6(5) Å3

Z 2

Density (calculated) 1.285 Mg/cm3

Absorption coefficient 0.960 mm-1

F(000) 1076

Table S4. Data collection and structure refinement for 6a.

Theta range for data 4.81 to 63.42° collection

Index ranges -8<=h<=15, -15<=k<=14, -17<=l<=18

Reflections collected 19965

Independent 7149 [R(int) = 0.0471] reflections

Coverage of independent 82.0% reflections

Absorption correction multi-scan

Max. and min. 0.8639 and 0.8320 transmission

Structure solution direct methods technique

Structure solution SHELXS-97 (Sheldrick, 2008) program

Refinement method Full-matrix least-squares on F2

Refinement program SHELXL-97 (Sheldrick, 2008)

2 2 2 Function minimized Σ w(Fo - Fc )

Data / restraints / 7149 / 0 / 722 parameters

Goodness-of-fit on F2 1.639

4358 data; R1 = 0.0870, wR2 = Final R indices I>2σ(I) 0.2524

R1 = 0.1267, wR2 = all data 0.2699

2 2 2 w=1/[σ (Fo )+(0.1000P) +0.0000P] Weighting scheme 2 2 where P=(Fo +2Fc )/3

Largest diff. peak and 0.524 and -0.408 eÅ-3 hole

R.M.S. deviation from 0.075 eÅ-3 mean

Figure S1: The structure of 5a with 35% thermal ellipsoids. Hydrogen atoms have been omitted for clarity.

Figure S2: The structure of 6a with 35% thermal ellipsoids. Hydrogen atoms have been omitted for clarity.

Spectroscopy data：

100

101

102

103

104

105

Chapter 4: Water-Soluble Ionic Benzoporphyrins

Abstract

Novel ionic water-soluble tetrabenzoporphyrins have been successfully synthesized via a cascade reaction based on the Heck reaction. The UV-Vis spectra of these porphyrins displayed red-shifted and broadened Soret bands, and significantly enhanced and red-shifted Q bands up to 650-750nm. These porphyrins bearing eight ionic groups at the porphyrin -positions are highly water soluble.

Papers published from this chapter’s work:

1) L. Jiang, R. A. Zaenglein, J. T. Engle, C. Mittal, C.S. Hartley, C. J. Ziegler, and H. Wang.

Chem. Commun., 2012, 48, 6927-6929.

106

4.1 Introduction to Ionic Water Soluble Porphyrins

Water soluble porphyrins, in particular ionic porphyrins, are very useful in a wide range of applications requiring aqueous media, such as photodynamic therapy (PDT),1 as DNA cleavage catalysts,2 biosensors,3 and organic light-emitting diodes (OLEDs).4 Water soluble porphyrins are almost exclusively obtained through the introduction of hydrophilic groups, including pyridinium5, 6 and sulfonate,6-8 at the meso-positions of the porphyrins. The most representative water soluble porphyrins are the cationic meso-tetrakis(N-methylpyridinium-4-yl)porphyrin iodide, [(TMPyP)]I(4); and the anionic meso-tetra[4-sulfonatophenyl] porphyrin (TPPS4−).(Figure 4.1) In fact, a large number of published works were based on these two porphyrins. Water soluble porphyrins obtained through the introduction of hydrophilic groups at the -positions are very rare.9 However, functional groups installed at the -positions are expected to have more profound influence on the electro- and photophysical properties of the porphyrins.

Figure 4.1 Two representative cationic (left) and anionic (right) water soluble porphyrins.

 -Extended porphyrins, in which aromatic rings are fused to the porphyrin periphery through the , ’-positions, are attractive synthetic targets due to their unique set of photophysical and electrochemical properties.10, 11 As compared with the

107 parent porphyrins, -extended porphyrins often show significantly red-shifted absorption bands, a much desired feature for various applications. Significantly enhanced Q bands have also been observed.10 However, due to limited synthetic methods, particularly the marked lack of functionalization methods, only a small number of -extended porphyrins have been made available. To our knowledge, there are only two examples of water soluble extended porphyrins in the literature, in which either carboxylic acid groups or hydroxyl groups are attached at the porphyrin periphery.12, 13

4.2 Synthesis of Starting TBPs

Ionic water soluble extended porphyrins appear to be very attractive given their huge potential in various applications. Very recently, we have developed a concise and versatile method to synthesize functionalized benzoporphyrins.14, 15 In this method, an alkene reacts with a , ’-dibromoporphyrin through a three-step cascade reaction involving a vicinal two-fold Heck reaction, six-electrocyclization, and subsequent aromatization. We envisioned that through the utilization of vinyl pyridine or sodium styrenesulfonate in this method, up to eight cationic (pyridinium) or anionic groups (sulfonate) could be attached at the fused benzene rings. Herein, we report the synthesis and characterization of the first examples of water soluble cationic and anionic tetrabenzoporphyrins. In order to investigate the feasibility of this approach, we first tried the Heck-based cascade reaction on Ni(II) dibromoporphyrin 1 with 4-vinyl-pyridine, 3-vinylpyridine and sodium 4-styrene sulfonate. To our delight, all these reactions went smoothly to give the desired substituted monobenzoporphyrins 2a-c (Scheme 4.1). The structures of 2a-c were determined by 1H and 13C NMR spectroscopy as well as LDI-TOF mass spectrometry.

108

Scheme 4.1 The synthesis of ionic monobenzoporphyrins via a cascade reaction

We then proceeded to further investigate the reaction of vinyl pyridines with octabromoporphyrin 3a and 3b (Scheme 4.2). The desired tetrabenzoporphyrins 4a and 4b were isolated in 51-54% yields. It is surprising that the yields did not decrease with increased reaction complexity. Free base tetrabenzoporphyrin 4c was obtained by demetallation of 4b using a mixture of TFA and sulfuric acid. Reinsertion of Zn(II) using Zn(OAc)2 gave the Zn(II) tetrabenzoporphyrin 4d. 4d can also be prepared from Zn(II) octabromoporphyrin 3c using the Heck-based cascade reaction in 46% yield. The structures of 4a-e were determined by 1H and 13C NMR spectroscopy, and LDI-TOF mass spectrometry, and were further confirmed with the X-ray structure elucidation of 4a (Figure. 4.2). The single X-ray diffraction data were collected and analyzed by Dr. Christopher J. Ziegler’s group at University of Akron. The X-ray crystal structure reveals a saddle conformation for the Ni(II) tetrabenzoporphyrin 4a. It is notable that the central Ni(II) assumes an octahedral coordination with two axial pyridine ligands which came from the solvent used to grow the crystal. Computational geometries of the porphyrins and images of their frontier molecular orbitals are given in the Supporting Information. Unlike most other tetrabenzoporphyrins, 4a-e are highly soluble in common organic solvents such as chloroform, dichloromethane, pyridine and THF. They are also highly soluble in methanol. For comparison purpose, we also prepared an opp-dibenzoporphyrin 5. (Scheme 4.2) We also tried the Heck-based cascade reaction of sodium 4-styrene sulfonate with

109 octabromoporphyrin 3a in order to obtain the anionic tetrabenzoporphyrin 6 (Scheme 4.2). After the reaction was completed, the reaction mixture went through a short silica plug using methanol/DCM as the eluent, and was then directly converted to its acidic form (6) using acidic ion exchange resin (Dowex 50Wx8). 6 was then purified by silica column chromatography using MeOH/DCM as the eluent, followed by size-exclusion chromatography (Sephadex LH-20). ESI mass spectrometry revealed the existence of 6. UV-Vis spectra of 6 showed well defined Soret band at 466nm, similar to the Soret bands of 4a and 4e, and Q bands at 630 and 646 nm. However, 1H NMR of 6 displayed broadened and undefined proton shifts. This is likely due to the combined effects arising from porphyrin aggregation, intermolecular hydrogen bonding of the sulfonic acid groups, the dynamic ring flipping of the macrocycle owning to the crowding on the porphyrin periphery as well as the hindered ring rotation of phenyl substituents. This phenomenon has been observed for other peripherally crowded porphyrins.15, 16 6 is highly soluble in water, methanol and

DMSO, and does not dissolve in DCM, CHCl3, and acetonitrile.

110

Scheme 4.2 The synthesis of ionic tetrabenzoporphyrins 4a-e, 6, and cationic opp-dibenzo-

porphyrin 5.

111

Figure 4.2 X-ray crystal structure of 4a with 35% thermal ellipsoids: left, edge view; right, top view. Hydrogen atoms have been removed for clarity.

4.3 UV-Vis Absorption and Fluorescence Spectroscopy

UV-vis absorption spectra of 4a-e are shown in Figure 4.3 (a). As compared with their unfused parent porphyrins, the Soret bands of these porphyrins are significantly broadened and red-shifted by ~50 nm upon fusion of four benzene rings on the porphyrin periphery. The Q-bands of these porphyrins are also red-shifted and much enhanced. It is notable that the Q bands of Ni porphyrins 4a and 4e are much more intense than those of the corresponding Cu(II), Zn(II) and free base porphyrins. However, such enhancement of Q bands was not observed for Ni(II) monobenzoporphyrin 3a and opp-dibenzoporphyrin 5 (see Experimental part).

112

Treatment of 4a-e with trifluoroacetic acid (TFA) resulted in further shifting of the Soret bands and Q bands. The red-shifting of the absorption bands for 4-pyridyl substituted porphyrins (4a-d) is much more significant than the 3-pyridyl substituted porphyrin 4e, likely due to the exceptionally strong electron-withdrawing nature of the protonated 4-pyridyl groups. The Soret bands of 4a-d are further red-shifted by 33-37 nm to 503-525nm, and the Q bands close to near IR region (up to 708 nm for 4d) (Figure 4.3 (b)). It is remarkable that the Q band of protonated 4a absorbs at 684 nm with unusually high intensity (the ISoret/IQ ratio is 1.6), a rare observation for monomeric porphyrins,5, 7 and a phenomenon not known for Ni(II) porphyrins. We speculate that the unusual enhancement and red-shifting of the Q band observed for 4a may relate to the central metal (Ni(II)), the nonplanar distortion of the porphyrin macrocycle brought by the fusion of four benzene rings, and the strong electron-withdrawing groups at the fused rings. This feature is very attractive for applications in photodynamic therapy. Fluorescence spectroscopy was performed for

4d, displaying near IR emissions (max = 703 nm, 756 nm) (Figure 4.3, (c)).

(a)

1.00 4a 4b 0.75 4c 4d 4e

0.50 Absorption 0.25

0.00 400 450 500 550 600 650 700 750 800 Wavelength (nm)

113

(b)

1.00 4a + TFA 4e + TFA 0.75 4b + TFA 4d + TFA

0.50 4c + TFA Absorption 0.25

0.00 500 600 700 800 Wavelength (nm)

(c)

250

200

150

Intensity 100

0 650 750 850 Wavelength (nm)

4.4 Structure Optimization Calculations

DFT calculations: Gaussian 03 (Rev. D.02) calculations were carried out on Miami University's Redhawk computer cluster. Following geometry optimizations, vibrational frequency analysis was used to ensure that all stationary points were energy minima. Geometry optimization and electronic structure calculations of the porphyrins were performed by using the B3LYP functional and 6-31G(d,p) basis set. Molecular orbitals were visualized using Molekel 5.4.0. 17

114

Density functional calculations (B3LYP/6-31G(d,p)) were conducted for 4a-e by Dr. Scott Hartley of Miami University. Similar to the crystal structure of 4a, the calculated structure of 4a also adopts a saddle conformation. All the other tetrabenzoporphyrins 4b-4c assume a similar conformation (Figure 4.4). The electronic absorption of porphyrins including both Soret band and Q bands arises from -* transitions. The frontier orbitals responsible for the transitions in the parent porphin are two orbitals (a1u and a2u) and two degenerate * orbitals (egx and egy) in the Gouterman four-orbital model.18 Figure 4.5 illustrates the calculated HOMOs and LUMOs for compounds 4a-4d, and protonated 4a. It is interesting to note that, while all the HOMOs and LUMOs of 4a-4d do not clearly involve the participation of the pyridyl substitutes on the fused benzene rings, the HOMOs and LUMOs of protonated 4a heavily involve those substitutes.

115

Figure 4.4 Optimized molecular structure of 4a-4d, protonated 4a and 2a.

116

Figure 4.5 The calculated HOMOs and LUMOs of 4a-4d and protonated 4a

117

4.5 Application of Water Soluble Porphyrins towards Photodynamic Therapy (PDT)

Photodynamic therapy (PDT) has been established as a significant method in medicine due to their less harmful side effects to normal cells compared to traditional tumor treatments such as chemotherapy or radiotherapy. There are three major components in modern PDT: a proper photosensitizer, an appropriate visible light source and reactive tissue oxygen species. Upon irradiation, the excited state of the photosensitizer generates highly reactive singlet oxygen and other cytotoxic reactive oxygen species which can induce localized cell death and ultimately tissue apoptosis or necrosis. 19 PDT has been successfully applied in various cancers including brain tumors20, esophagus21, early oral and laryngeal cancers22 as well as breast carcinoma.23 The most efficient photosensitizers are those which have strong absorption bands at the infrared range of the visible spectrum because both absorption and scattering of light by tissue decrease as the wavelength increases. On the other hand, an extension of the -system leads the decrease of the oxidation potential, then the photosensitizer becomes less stable kinetically, thus subject to photobleaching.1, 24 Based on these two factors, the ideal photosensitizers should be activated at wavelengths between 700 to 850 nm for maximum light penetration through tissues

25 with minimum light scattering. Porphyrin based photosensitizers have arisen as the most promising sensitizers for PDT due to their selectivity and affinity for tumor cells.26, 27 Traditional porphyrin photosensitizers have several disadvantages including poor light absorption at the ideal wavelength range, high hydrophilicity, and serious aggregation in aqueous media, as well as the cutaneous phototoxicity side effect. New generations of porphyrin photosensitizers have been successfully synthesized such as Tookad,28 Photofrin,29 Lutrin,30 Purlytin,31 in order to improve the efficacy of PDT, and to solve the existing problems with traditional porphyrin photosensitizers and improving the

118 efficacy of PDT. (Figure 4.6) These new porphyrin photosensitizers are chemically purer and can absorb light at longer wavelength. As a result, they have shown improved activity and decreased side effects.32 However, the major disadvantages still exist. These porphyrins still have poor solubility in water, high dark toxicity and cutaneous phototoxicity side effect, as well as limited depth of penetration in tumor cells. In addition, the lengthy synthetic steps involved in the preparation of these porphyrins and the lack of functional groups on these porphyrins have also hindered the fully investigation of their applications.

Figure 4.6 Structures of the porphyrin based photosensitisers used for clinical or preclinical PDT

 etrabenzoporphyins (TBPs), which have fused aromatic rings at the  pyrrolic positions, are attractive synthetic targets due to their unique combination of photophysical and electrochemical properties. In general, these compounds have a UV-Vis absorption spectrum similar to porphyrins. However, depending on the length of the extended π-systems attached on the porphyrin periphery, TBPs often show significantly red-shifted and broadened absorptions, a much desired feature for various applications. The Q bands of extended porphyrins are often red-shifted to 600-750 nm in the red region of the visible spectrum. These properties, together with the fact that they can generate singlet oxygen efficiently make benzoporphyrins and their derivatives promising candidates in therapeutic applications. However, due to the poor solubility and limited synthetic methods, especially the lack of functionalization methods, only a small number of TBPs and their derivatives have been made available. A new type of water-soluble benzoporphyrin having eight

119 carboxylic acid groups at the fused benzene rings was successfully synthesized by Sugimoto and co-workers.13 Unfortunately, the application of this novel TBP was hindered by the solubility issue (mainly because of the lack of meso-substitutions and highly planar structure) coupled with long synthetic steps. Introduction of aryl groups tothe meso-positions of TBPs were proved to increase the solubility of the benzoporphyrinsin organic solvents partially because of the resulted nonplanar structure.33 To increase the water solubility of TBPs, introduction of sulfonic acid,34 nido-carboranyl,26 glucosyl and polyaminol groups33 have been introduced to the meso-aryl groups. All these above mentioned strategies enhanced the water solubility and showed promising results in PDT. However, functional groups installed at the -positions are expected to have more profound influence on the properties of the TBPs. In this section, I describe the synthesis of two novel class of highly water soluble tetrabenzoporphyrins, Ar4[T(MPy)8BP]I8, bearing eight pyridinium-iodide units, along with full experimental data (1H NMR, LDI-TOF MS, absorption and fluorescence spectroscopes).

4.6 Synthesis of Water Soluble TBPs

Water soluble Zn and free base octa-(N-methylpyridinium-4-yl)tetrabenzo- tetraaryl porphyrins 7c and 7d have been synthesized successfully, as shown in Scheme 4.3. 25 mg 4c and 4d were dissolved in 1 mL of MeI and 3 mL of methanol and stirred at room temperature for two days. The color changed from dark green to rosy/brownish. Upon removal of solvents and recrystallized through chloroform and methanol, water soluble porphyrins 7c and 7d were synthesized in 95% and 96% yield separately. 7c and 7d are highly soluble in water and the aqueous solution and stable for more than one week at room temperature when shielded from light. Precipitation was not detected by visual observation and the UV-vis spectra were consistent with the observation.

120

Scheme 4.3 Synthesis of novel water soluble TBPs

4.7 Characterization of Water Soluble TBPs

4.7.1 Mass Spectrometry

The measurement of the Mass Spectra of porphyrin 7c and 7d was carried out using laser desorption ionization time-of-flight (LDI-TOF) technique. Both compounds gave a series of peaks in the 1600-1720 m/z region, corresponding to the loss of one to eight methyl groups consecutively from the pyridium rings of the parent TBP, thus confirming the formation of 7c and 7d. (Figure 4.7, 7c)

121

Figure 4.7 LDI-TOF mass spectrum of 7c.

4.7.2 1H NMR Characterization

1H NMR spectra of 7c and 7d were recorded on a 500 MHz NMR spectrometer

1 (Bruker) in CD3OD. H NMR spectrum of compound 7c (Fig. 4.8) is composed of four separate parts: a) the proton shifts between 7.5 and 8.9 ppm account for 56 aromatic protons of the fused benzene and pyridine rings on the porphyrin periphery; b) the proton shifts at around 4.4 ppm are from the 24 protons of the cationic methyl groups attached to the pyridine rings; c) the 4 isopropyl protons on the meso-fused aromatic rings can be found as a multiplet at 3.4ppm; d) the proton shifts at 1.5 ppm are assigned to the 24 methyl protons on the meso-phenyl rings. The ratio of integration of the four parts equals to14: 6: 1: 6, suggesting that all the eight pyridine rings on the porphyrins were methylated.

122

1 Figure 4.8 H NMR in d4-methanol of compound 7c.

4.7.3 UV-Vis Absorption and Fluorescence Spectroscopes

UV-vis absorption spectra of 4c-d and 7c-d are shown in Figure 4.9 (a). In general, the Soret bands of these tetrabenzoporphyrins are significantly broadened and red-shifted to 484-530 nm due to the fusion of four benzene rings on the porphyrin periphery. The Q-bands of these porphyrins are also enhanced and red-shifted to red to near IR region. Furthermore, fully methylation of these porphyrins broadened and red-shifted both the Soret bands and the Q bands by another 30 to 40 nm. Fluorescence spectroscopes were performed for 7c and 7d. Both 7c and 7d displayed near IR emissions. (Fig. 4.9 (b)) Compared with other known water soluble TBPs,13, 33 this novel class of porphyrins (Ar4[T(MPy)8BP]I8) have shown more promising UV-vis and fluorescence data, indicating that functionalization of porphyrins at the -positions has the potential to solve long existing solubility issue and limited depth of penetration problems.

123

(a) (b)

1.00 7d 4d 100 7c 0.75 7d 4c

0.50 7c

Absorption 50

0.25

0.00 0 400 450 500 550 600 650 700 750 800 650 700 750 800 850 900 Wavelength (nm) Wavelength (nm)

Figure 4.9 (a) Normalized absorption spectra of porphyrins 4c-d and 7c-d (4c and 4d were tested

in CH2Cl2 solution; 7c and 7d were tested in methanol solution); (b) Fluorescent spectra of 7c and

7d in methanol.

4.8 Conclusion

In this study, we described a concise synthetic method affording highly water soluble Ar4[T(MPy)8BP]I8 type tetrabenzoporphyrins. With eight ionic groups at the -positions of the porphyrin periphery, these compounds are highly soluble in water and methanol. The UV-vis spectra of these porphyrins displayed significantly red-shifted and broadened Soret bands and Q-bands, covering almost the whole visible region of the spectrum. 7c and 7d exhibited near IR fluorescence in the range of 700 to 870 nm, the most desired wavelength for penetrating depth. These interesting data have never been discovered in other types of water soluble porpohyrins. It is expected that other novel photophysical and electronic properties will be discovered in these porphyrins by displacing the central metal with a different metal such as Pt(II), La(III) etc. This method can be used to introduce a variety of counter ions on the eight pyridine rings. These porphyrins hold great potential for applications in various areas.

Reference： 1. Bonnett, R., Chemical Society Reviews 1995, 24 (1), 19-33.

124

2. Tabata, M.; Nakajima, K.; Nyarko, E., Journal of Inorganic Biochemistry 2000, 78 (4), 383-389. 3. Zhang, X.-a.; Lovejoy, K. S.; Jasanoff, A.; Lippard, S. J., Proceedings of the National Academy of Sciences 2007, 104 (26), 10780-10785. 4. Perez-Morales, M.; de Miguel, G.; Bolink, H.; Martin-Romero, M.; Camacho, L., Journal of Materials Chemistry 2009, 19 (24), 4255-4260. 5. Villanueva, A.; Caggiari, L.; Jori, G.; Milanesi, C., Journal of Photochemistry and Photobiology B: Biology 1994, 23 (1), 49-56. 6. Lieb, D.; Zahl, A.; Shubina, T. E.; Ivanović-Burmazović, I., Journal of the American Chemical Society 2010, 132 (21), 7282-7284. 7. Hasbrouck, R. W.; Anderson, A. D., The Journal of Organic Chemistry 1973, 38 (11), 2103-2106. 8. Ohno, O.; Kaizu, Y.; Kobayashi, H., The Journal of Chemical Physics 1993, 99 (5), 4128-4139; Villari, V.; Mazzaglia, A.; Trapani, M.; Castriciano, M. A.; de Luca, G.; Romeo, A.; Scolaro, L. M.; Micali, N., The Journal of Physical Chemistry C 2011, 115 (13), 5435-5439. 9. Aoyagi, K.; Haga, T.; Toi, H.; Aoyama, Y.; Mizutani, T.; Ogoshi, H., Bull. Chem. Soc. Jpn. 1997, 70; Finikova, O.; Galkin, A.; Rozhkov, V.; Cordero, M.; Hägerhäll, C.; Vinogradov, S., Journal of the American Chemical Society 2003, 125 (16), 4882-4893. 10. Baluschev, S.; Yakutkin, V.; Miteva, T.; Avlasevich, Y.; Chernov, S.; Aleshchenkov, S.; Nelles, G.; Cheprakov, A.; Yasuda, A.; Mullen, K.; Wegner, G., Angewandte Chemie-International Edition 2007, 46 (40), 7693-7696; Borek, C.; Hanson, K.; Djurovich, P.; Thompson, M.; Aznavour, K.; Bau, R.; Sun, Y.; Forrest, S.; Brooks, J.; Michalski, L.; Brown, J., Angewandte Chemie-International Edition 2007, 46 (7), 1109-1112. 11. Ongayi, O.; Gottumukkala, V.; Fronczek, F.; Vicente, M., Bioorganic & Medicinal Chemistry Letters 2005, 15 (6), 1665-1668; Mack, J.; Bunya, M.; Shimizu, Y.; Uoyama, H.; Komobuchi, N.; Okujima, T.; Uno, H.; Ito, S.; Stillman, M.; Ono, N.; Kobayashi, N., Chemistry-a European Journal 2008, 14 (16), 5001-5020; Finikova, O.; Cheprakov, A.; Beletskaya, I.; Vinogradov, S., Chemical Communications 2001, (03), 261-262; Lash, T. D., Journal of Porphyrins and Phthalocyanines 2001, 5 (3), 267-288. 12. Filatov, M. A.; Cheprakov, A. V., Tetrahedron 2011, 67 (19), 3559-3566. 13. Murashima, T.; Tsujimoto, S.; Yamada, T.; Miyazawa, T.; Uno, H.; Ono, N.; Sugimoto, N., Tetrahedron Letters 2005, 46 (1), 113-116. 14. Deshpande, R.; Jiang, L.; Schmidt, G.; Rakovan, J.; Wang, X.; Wheeler, K.; Wang, H., Organic Letters 2009, 11 (19), 4251-4253. 15. Jiang, L.; Engle, J.; Sirk, L.; Hartley, C.; Ziegler, C.; Wang, H., Organic Letters 2011, 13 (12), 3020-3023. 16. Jentzen, W.; Song, X.-Z.; Shelnutt, J. A., The Journal of Physical Chemistry B 1997, 101 (9), 1684-1699. 17. Varetto, U. Molekel 5.4.0.8, Swiss National Supercomputing Centre: Manno (Switzerland). 18. Ceulemans, A.; Oldenhof, W.; Gorller-Walrand, C.; Vanquickenborne, L. G., Journal of the American Chemical Society 1986, 108 (6), 1155-1163; Gouterman, M., Journal of Chemical Physics 1959, 30 (5), 23. 19. Kessel, D., Photodiagnosis and Photodynamic Therapy 2004, 1 (1), 3-7; Dougherty, T.; Gomer, C.; Henderson, B.; Jori, G.; Kessel, D.; Korbelik, M.; Moan, J.; Peng, Q., Journal of the National Cancer

125

Institute 1998, 90 (12), 889-905; Moan, J.; Peng, Q., Anticancer Research 2003, 23 (5A), 3591-3600. 20. Powers, S.; Cush, S.; Walstad, D.; Kwock, L., Neurosurgery 1991, 29 (5), 688-696. 21. Marcon, N., Seminars in Oncology 1994, 21 (6), 20-23. 22. Biel, M., Photochemistry and Photobiology 2007, 83 (5), 1063-1068. 23. Allison, R.; Mang, T.; Hewson, G.; Snider, W.; Dougherty, D., Cancer 2001, 91 (1), 1-8. 24. Balaz, M.; Collins, H.; Dahlstedt, E.; Anderson, H., Organic & Biomolecular Chemistry 2009, 7 (5), 874-888. 25. O'Connor, A.; Gallagher, W.; Byrne, A., Photochemistry and Photobiology 2009, 85 (5), 1053-1074. 26. Gottumukkala, V.; Ongayi, O.; Baker, D.; Lomax, L.; Vicente, M., Bioorganic & Medicinal Chemistry 2006, 14 (6), 1871-1879. 27. Thomas, A. P.; Saneesh Babu, P. S.; Asha Nair, S.; Ramakrishnan, S.; Ramaiah, D.; Chandrashekar, T. K.; Srinivasan, A.; Radhakrishna Pillai, M., Journal of Medicinal Chemistry 2012, 55 (11), 5110-5120; Schmitt, F.; Barry, N. P. E.; Juillerat-Jeanneret, L.; Therrien, B., Bioorganic & Medicinal Chemistry Letters 2012, 22 (1), 178-180; Orlandi, V. T.; Caruso, E.; Banfi, S.; Barbieri, P., Photochemistry and Photobiology 2012, 88 (3), 557-564; Dosselli, R.; Tampieri, C.; Ruiz-González, R.; De Munari, S.; Ragàs, X.; Sánchez-García, D.; Agut, M.; Nonell, S.; Reddi, E.; Gobbo, M., Journal of Medicinal Chemistry 2012, 56 (3), 1052-1063. 28. Weersink, R.; Bogaards, A.; Gertner, M.; Davidson, S.; Zhang, K.; Netchev, G.; Trachtenberg, J.; Wilson, B., Journal of Photochemistry and Photobiology B-Biology 2005, 79 (3), 211-222. 29. Karotki, A.; Khurana, M.; Lepock, J.; Wilson, B., Photochemistry and Photobiology 2006, 82 (2), 443-452. 30. Hsi, R.; Kapatkin, A.; Strandberg, J.; Zhu, T.; Vulcan, T.; Solonenko, M.; Rodriguez, C.; Chang, J.; Saunders, M.; Mason, N.; Hahn, S., Clinical Cancer Research 2001, 7 (3), 651-660. 31. Josefsen, L. B. a. B., Ross. W., Metal-Based Drugs 2008, 2008, 1-24. 32. Koudinova, N.; Pinthus, J.; Brandis, A.; Brenner, O.; Bendel, P.; Ramon, J.; Eshhar, Z.; Scherz, A.; Salomon, Y., International Journal of Cancer 2003, 104 (6), 782-789; Liang, X.; Li, X.; Yue, X.; Dai, Z., Angewandte Chemie-International Edition 2011, 50 (49), 11622-11627. 33. Menard, F.; Sol, V.; Ringot, C.; Granet, R.; Alves, S.; Le Morvan, C.; Queneau, Y.; Ono, N.; Krausz, P., Bioorganic & Medicinal Chemistry 2009, 17 (22), 7647-7657. 34. Vinogradov, S.; Wilson, D., Journal of the Chemical Society-Perkin Transactions 2 1995, (1), 103-111.

126

4.9 Experimental Part

General All solvents were analytical reagent grade unless otherwise stated and were obtained either from Sigma-Aldrich or ACROS. Analytical TLC’s were performed on Silicycle UltraPure Silica Gel 60 F254 TLC plates. Preparative column chromatography was performed on silica gel (1000m), which was purchased from Silicycle. 1H and 13C experiments were conducted on a Bruker Avance 500MHz spectrometer. All samples were prepared in CDCl3 and chemical shifts were referenced to CDCl3 at 7.24ppm for

1 13 H NMR and referenced to CDCl3 at 77 ppm for C-NMR unless otherwise stated. UV-Visible spectra were recorded on an Agilent 8453 UV-Visible spectrometer in

CH2Cl2. Mass spectra were obtained on Bruker LDI-TOF mass spectrometer and Bruker ESQUIRE~LCMS. M.P.’s were measured on an Electrothermal MEL-TEMP apparatus and were uncorrected. General procedure for the Heck coupling reaction of metalated dibromoporphyrins

Dibromoarylporphyrin 1 (0.045 mmol), palladium acetate (0.012 mmol), triphenylphosphine (0.030 mmol) and K2CO3 (0.09 mmol) were added to Schlenk tube and dried under vacuum. The vacuum was released under argon to allow the

127 addition of dry DMF (10 mL) and dry xylene (10 mL) and 4-vinylpyridine or 3-vinylpyridine or sodium 4-phenyl sulfonate (25-fold excess). The mixture was then degassed via four freeze-pump-thaw cycles before the vessel was purged with argon again. The Schlenk flask was sealed and heated to reflux for 72h. After 72 h, the mixture was diluted with CHCl3 and washed with water. The organic layer was removed under vacuum. The residue was subjected to silica column chromatography. The band containing the desired porphyrin was collected and recrystallized from

CHCl3 and methanol. This procedure was used to prepare 2a, 2b. For 2c, after refluxing for 72h, the solvent was removed under vacuum and the compound was redissolved in isopropanol and passed through a short silica gel plug. Solvent was removed again and the mixture was run through a saphadex resin column in methanol to get rid of excess 4-styrenesulfonic acid sodium salt hydrate. The ideal product 2c was isolated on a silica column (DCM/MeOH). Recrystallization was performed using methanol and DCM, in which DCM served as the poorer solvent. In order to obtain a better 1H NMR, excess of ionic exchange resin (Dowex 50Wx8) was used to convert the product to the acidic form 2c’.

o 2a: mp > 320 C. Yield: 56%. UV-Vis λmax (CH2Cl2)/nm 434 (log ε 5.52), 545 (4.31),

1 577 (4.12); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.68-8.71 (6 H, m, β-H), 8.46-8.48 (4H, m, o-pyridine-H), 7.85-7.90 (8H, m, o-Ph-H on meso-phenyl ring), 7.50-7.56 (8H, m, m-Ph-H on meso-phenyl ring), 7.21 (2H, s, fused benzene-H), 6.95 (4H, m, m-pyridine-H), 3.13-3.19 (4H, m, isopropyl(CH)-H), 1.41-1.47 (24H, m, isopropyl

13 (CH3)-H) ; C-NMR (500 MHz, CDCl3, Me4Si) δ 24.21, 24.29, 34.07, 34.15, 115.50, 120.49, 124.88, 125.12, 126.10, 126.47, 131.24, 131.79, 132.77, 132.80, 133.64, 134.44, 137.11, 137.66, 138.15, 140.35, 141.20, 142.14, 143.99, 148.47; Calculated Mass, 1042.42, Found MS (LDI-TOF), m/z 1042.52.

o 2b: mp > 320 C. Yield: 50%. UV-Vis λmax (CH2Cl2)/nm 433 (log ε 5.51), 538 (4.32),

1 577 (4.10); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.68-8.70 (6 H, m, β-H), 8.43-8.44 (2H, m), 8.26 (2H, m), 7.85-7.91 (8H, m, o-Ph-H on meso-phenyl ring), 7.50-7.55

128

(8H, m, m-Ph-H on meso-phenyl ring), 7.29-7.30 (2H, m), 7.099-7.11 (4H, m),

3.11-3.20 (4H, m, isopropyl(CH)-H), 1.36-1.48 (24H, m, isopropyl (CH3)-H) ;

13 C-NMR (500 MHz, CDCl3, Me4Si) δ 24.21, 24.26, 34.08, 34.17, 115.43, 120.38, 122.80, 125.09, 126.12, 126.54, 131.10, 131.62, 132.73, 133.64, 137.09, 137.19, 137.37, 137.75 138.20, 140.17, 141.06, 142.01, 143.96, 147.63, 148.41, 149.54, 150.73; Calculated Mass, 1042.42, Found MS (LDI-TOF), m/z 1043.18.

o 2c’: mp > 320 C. Yield: 42%. UV-Vis λmax (CH2Cl2)/nm 430 (log ε 5.51), 540 (4.60),

1 575 (4.47); H-NMR (500 MHz, CDCl3 with two drops of CD3OD, Me4Si) δ 8.56-8.70 (6 H, m, β-H), 7.80-8.10 (8H, m), 7.45-7.63 (12H, m), 6.80-7.15 (6H, m),

3.11-3.15 (4H, m, isopropyl(CH)-H), 1.33-1.46 (24H, m, isopropyl (CH3)-H); Calculated Mass, 1200.35, Found MS (LDI-TOF), m/z 1200.32. (For 2c’, five proton

NMRs were attached, of which CDCl3 with two drops of CD3OD gave the best results.

Other solvent systems including only CDCl3, only MeOD, and one or three drops of

MeOD mixed with CDCl3 gave worse results.) General procedure for the octabromination reactions

Ni-arylporphyrin (1 mmol) or Cu-arylporphyrin (1 mmol) and NBS (14 mmol) were added into a round bottom flask and dissolved in dry 1, 2-dichloroethane. The mixture was then refluxed for 2 hours. The organic layer was removed under vacuum. The

129 residue was subjected to column chromatography. The bands containing the desired porphyrins were collected and recrystallized from CHCl3/MeOH. This procedure was used to prepare 3a and 3b. 3c was prepared by demetalation of 3b through treatment of concentrated H2SO4 and TFA, followed by re-insertion of Zn using Zn(OAc)2.

o 3a: mp > 320 C. yield: 54%. UV-Vis λmax (CH2Cl2)/nm 451 (log ε 5.54), 564 (4.96),

1 598 (4.62); H-NMR (500 MHz, CDCl3, Me4Si) δ 7.79 (8H, d, J = 8.0 Hz, o-Ph-H), 7.51 (8H, d, J = 8.0 Hz, m-Ph-H), 3.13-3.19 (4H, m, isopropyl (CH)-H), 1.44 (24H, d,

13 J = 6.5 Hz, isopropyl (CH3)-H); C-NMR (500 MHz, CDCl3, Me4Si) δ 24.33, 24.36, 34.20, 34.25, 120.12, 126.22, 126.29, 126.99, 133.51, 134.97, 143.60, 150.99; Calculated Mass, 1470.90, Found MS (LDI-TOF), m/z 1470.91.

o 3b: mp > 320 C. Yield: 55%. UV-Vis λmax (CH2Cl2)/nm 461 (log ε 5.56), 579 (4.82), 624 (4.50); Calculated Mass, 1475.75, Found MS (LDI-TOF), m/z 1475.11.

o 3c: mp > 320 C. Yield: 72% in two steps. UV-Vis λmax (CH2Cl2)/nm 463 (log ε 5.55),

1 597 (4.88), 654 (4.58); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.00 (8H, m, o-Ph-H), 7.58 (8H, m, m-Ph-H), 3.22 (4H, m, isopropyl (CH)-H), 1.49 (24H, m, isopropyl

13 (CH3)-H); C-NMR (500 MHz, CDCl3, Me4Si) δ 24.49, 34.28, 109.88, 111.42, 111.67, 113.33, 113.57, 121.08, 124.62, 126.04, 136.49, 145.51, 150.61; Calculated Mass, 1477.59, Found MS (LDI-TOF), m/z 1476.70. General procedure for the Heck coupling reaction of metalated octabromoporphyrins

130

Octabromoarylporphyrins (0.045 mmol), palladium acetate (0.040 mmol), triphenylphosphine (0.120 mmol) and K2CO3 (0.36 mmol) were added to Schlenk tube and dried under vacuum. The vacuum was released under argon to allow the addition of dry DMF (10 mL) and dry xylene (10 mL) and relevant alkene (60-fold excess). The mixture was then degassed via four freeze-pump-thaw cycles before the vessel was purged with argon again. The Schlenk flask was sealed and heated to

131 reflux for 72h. After 72 h, the mixture was diluted with CHCl3 and washed with water. The organic layer was removed under vacuum. The residue was subjected to silica column chromatography. The bands containing the desired porphyrins were collected. This procedure was used to prepare 4a, 4b and 4e. 4c was obtained after demetalation of 4b by treating with concentrated H2SO4 and TFA. 4d was obtained by re-insertion of Zn using Zn(OAc)2. Water solubility was tested for 4a. Porphyrin 4a (20 mg) was dissolved in methanol and added excess of TFA (0.1 ml). After the mixture was stirred for 30 mins, the solvent was removed. The protonated porphyrin was recrystallized from DCM/MeOH (DCM acts as the bad solvent). The solid porphyrin was tested for solubility in water. Both protonation and methylation were performed for 4c. The protonated 4c was obtained through the treatment of 4c in methanol with excess of TFA for 5 hours. The solvent was removed. The desired product was recrystallized from DCM and methanol. The protonated product can be dissolved in methanol and water, but not in

1 chloroform or DCM. A HNMR was taken in CD3OD. Methylation was also performed on 4c. Excess of MeI was added into the solution of 4c in methanol, and the resulting mixture was stirred for 1.5 days. The solvent was removed under reduced pressure. The product was recrystallized from DCM and methanol. The methylated product can be dissolved in water (concentration ≥ 15mM) and methanol, but not in

1 chloroform or DCM. HNMR was taken in CD3OD. For 6, after the reaction was completed, the reaction mixture went through a short silica plug using methanol/DCM as the eluent, and was then directly converted to its acidic form 6 using acidic ion exchange resin (Dowex 50Wx8). 6 was then purified by silica column chromatography using MeOH/DCM as the eluent, followed by size-exclusion chromatography (Sephadex LH-20). Discussion of 6: ESI mass spectrometry revealed the existence of 6. UV-Vis spectra of 6 showed well defined Soret band at 466 nm, similar to the Soret bands of Ni(II) porphyrin 4a and 4e. However, 1H NMR of 6 displayed broadened and undefined

132 proton shifts. This is likely due to the combined effects arising from porphyrin aggregation, intermolecular hydrogen bonding of the sulfonic acid groups, the dynamic ring flipping of the macrocycle owning to the crowding on the porphyrin periphery as well as the hindered ring rotation of phenyl substituents. This phenomenon has been observed for other peripherally crowded porphyrins.9b 6 is highly soluble in water, methanol and DMSO, and does not dissolve in DCM, CHCl3, and acetonitrile. 1H NMR measurement of 6 was carried out in deuterated methanol, DMSO and water. All the spectra showed broadened shifts, although the broadening varies to different extent in different solvents. For comparison, we also took a 1H NMR of 4c (with pyridyl substituents) in deuterated methanol. As expected, the 1H NMR shifts of 4c were also significantly broadened due to the hydrogen-bonding with the solvent. In all these solvents, hydrogen bonding can be easily formed between the solvent molecules and the sulfonic groups. For comparison, we also took a 1H NMR of 4c (with pyridyl substituents) in deuterated methanol. As expected, the 1H NMR shifts of 4c were also significantly broadened due to the hydrogen-bonding with the solvent. Another factor may make the situation of 6 much more complicated than 4a-e

133 is that 6 can form "intramolecular" hydrogen bond through a solvent molecule or intermolecular hydrogen bond with itself (see Figure 1), but 4a-e are less likely to form such kind of hydrogen bonds. We believe that the formation of hydrogen bond restricts the rotation of the aromatic substituents on the fused benzene rings. The restricted rotation will also slow down the dynamic ring flipping of the macrocycle. In addition, 6 is likely to self-assemble leading to more complicated situation to obtain well-defined 1HNMR. Esterification of 6: 20 mg of 6 was treated with 5 mL of thionyl chloride and one drop of dry DMF. The mixture was allowed to stir for four hours and dried under vacuum. Then 10 mL of dry methanol and three drops of dry triethylamine were added and the mixture was stirred overnight. After drying under vacuum, DCM was added and the solution was sonicated for 20 min then passed through a filter to isolate the precipitant. The organic filtrate was then washed with water in order to remove inorganic residues. However, it turned out that the ester of 6 was hydrolyzed back to sulfonic acid by the acequeous workup. Amidation of 6 was then performed. A 50 mL round-bottomed flask containing 6 was equipped with a magnetic stirring bar and an addition funnel fitted at

134 the top with an argon balloon. The flask was cooled to 0oC in an ice-water bath and 2 mL of thionyl chloride and one drop of dry DMF were added dropwise through the addition funnel. The reaction mixture was warmed to room temperature and stirred for 1.5 hours after addition of thionyl chloride and DMF. Then the reaction flask is fitted with a distillation head and the excess of thionyl chloride was distilled off at reduced pressure. The flask containing residue was dried for one hour, 5 mL of dry DCM was then added into the flask under argon. Then 1.5 mL of dry hexylamine was added dropwise into the flask. The reaction mixture was stirred for 2 hours. The solvent was removed, and the excess hexylamine was washed off with hexane. The product was isolated and purified through a preparative TLC plate. 1H NMR and LDI-TOF were done in chloroform.

o 4a: mp > 320 C. Yield: 54%. UV-Vis λmax (CH2Cl2)/nm 465 (log ε 5.93), 608 (4.81),

1 659 (5.57); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.42 (16 H, d, J = 4.5 Hz, o-pyridine-H), 8.01 (8H, d, J = 7.5 Hz, o-Ph-H), 7.65 (8H, d, J = 7.5Hz, m-Ph-H), 7.22 (8H, s, fused-benzene-H), 6.88 (16H, d, J = 5.0 Hz, m-pyridine-H), 3.12-3.18

13 (4H, m, isopropyl(CH)-H), 1.36 (24H, d, J = 7.0 Hz, isopropyl (CH3)-H) ; C-NMR

(500 MHz, CDCl3, Me4Si) δ 24.32, 34.39, 108.56, 116.27, 124.60, 126.16, 127.39, 133.12, 134.42, 137.69, 138.63, 138.88, 148.87, 149.45, 151.16; Calculated Mass, 1654.63, Found MS (LDI-TOF), m/z 1654.83.

o 4b: mp > 320 C. Yield: 51%. UV-Vis λmax (CH2Cl2)/nm 473 (log ε 5.82), 612 (4.78), 662 (5.30); Calculated Mass, 1661.49, Found MS (LDI-TOF), m/z 1661.32.

o 4c: mp > 320 C. Yield: 90%. UV-Vis λmax (CH2Cl2)/nm 484 (log ε 5.80), 605 (4.60),

1 658 (5.08), 708 (4.49); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.43 (16 H, m, o-pyridine-H), 8.30 (8H, d, J = 7.0 Hz, o-Ph-H), 7.69 (8H, d, J = 7.0 Hz, m-Ph-H), 7.24 (8H, s, fused-benzene-H), 6.90 (16H, m, m-pyridine-H), 3.16-3.21 (4H, m, isopropyl(CH)-H), 1.36 (24H, d, J = 7.0 Hz, isopropyl (CH3)-H), -1.01 (2H, s, free

13 base H); C-NMR (500 MHz, CDCl3, Me4Si) δ 24.32, 34.37, 116.56, 124.58, 127.29, 134.16, 135.56, 139.13, 148.88, 149.52, 149.65, 151.25; Calculated Mass, 1598.71,

135

1 Found MS (LDI-TOF), m/z 1598.29. Protonated 4c: H-NMR (500 MHz, CD3OD) δ 8.68 (16 H, d, J = 5.5 Hz), 8.50 (8 H, d, J = 7.5 Hz), 7.89 (8 H, d, J = 8.0 Hz), 7.56-7.60 (24H, m), 3.34 (4H, m, isopropyl(CH)-H), 1.40 (24H, d, J = 7.0 Hz,

1 isopropyl (CH3)-H). Methylated 4c: H-NMR (500 MHz, CD3OD) δ 8.82-8.85 (16H, m), 8.52-8.76 (8H, m), 7.89-7.93 (8H, m), 7.75-7.76 (18H, m), 7.50-7.58 (6H, m),

4.40-4.42 (24H, s, -CH3), 3.34 (4H, m, isopropyl(CH)-H), 1.46 (24H, d, J = 6.5 Hz, isopropyl (CH3)-H).

o 4d: mp > 320 C. Yield: 88%. UV-Vis λmax (CH2Cl2)/nm 492 (log ε 5.90), 630 (4.89),

676 (5.19); FL (CH2Cl2, excited by 480nm) 707.50 (183.290), 750.0 (202.980);

1 H-NMR: No good 1H NMR was got from 3d by using CDCl3, MeOD or d5-pyridine.

The only interpretable NMR was got from combination of CDCl3 and d5-pyridine

(500 MHz, CDCl3 and d5-pyridine with a ratio of 2: 1, Me4Si) δ the singlet peaks at

8.71, 7.83 and 7.33 are from d5-pyridine solvents with the integration ratio of 2 : 1 : 2. δ 8.61 (16 H, m, o-pyridine-H), 8.46 (8H, d, J = 7.5 Hz, o-Ph-H), 7.83 (8H, d, J = 7.5 Hz, m-Ph-H), 7.55 (8H, s, fused-benzene-H), 7.10 (16H, m, m-pyridine-H), 4.32

(water peak from d5-pyridine), 3.31-3.32 (4H, m, isopropyl(CH)-H), 1.36 (24H, d, J =

7.0 Hz, isopropyl (CH3)-H); Calculated Mass, 1663.33, Found MS (LDI-TOF), m/z 1662.67.

o 4e: mp > 320 C. Yield: 49%. UV-Vis λmax (CH2Cl2)/nm 463 (log ε 5.95), 605 (4.80),

1 657 (5.55); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.42 (8 H, m), 8.26 (8H, m), 8.04 (8H, d, J = 8.0 Hz, o-Ph-H), 7.64 (8H, d, J = 8.0 Hz, m-Ph-H), 7.28-7.29 (8H, m), 7.28 (4H, m), 7.09-7.11 (12H, m), 3.12-3.15 (4H, m, isopropyl(CH)-H), 1.32 (24H, d, J =

13 7.0 Hz, isopropyl (CH3)-H) ; C-NMR (500 MHz, CDCl3, Me4Si) δ 24.32, 34.42, 116.08, 122.75, 126.25, 127.37, 128.51, 133.09, 133.74, 136.90, 137.04, 137.91, 138.54, 138.65, 147.81, 150.62, 151.13; Calculated Mass, 1656.64, Found MS (LDI-TOF), m/z 1656.04. 6: mp > 320oC. Yield: 15%. UV-Vis λmax (MeOH)/nm 466 (log ε 5.68), 630 (5.29), 646 (5.22); No good 1H NMR was got from any solvent. Found MS (ESI) (after

136 treated with excess of triethylamine(TEA)) from negative ion polarity 325.2 ([M-7H]7-), 380.9 ([M-6H]6-), 457.4 ([M-5H]5-); from positive ion polarity 102.1 (TEA+). General procedure for the Heck coupling reaction of metalated tetrabromoporphyrins

Tetrabromoarylporphyrins (0.045 mmol), palladium acetate (0.023 mmol), triphenylphosphine (0.058 mmol) and K2CO3 (0.17 mmol) were added to Schlenk tube and dried under vacuum. The vacuum was released under argon to allow the addition of dry DMF (10 mL) and dry xylene (10 mL) and 4-vinylpyridine (35-fold excess). The mixture was then degassed via four freeze-pump-thaw cycles before the vessel was purged with argon again. The Schlenk flask was sealed and heated to reflux for 72h. After 72 h, the mixture was diluted with CHCl3 and washed with water. The organic layer was removed under vacuum. The residue was subjected to silica column chromatography. The band containing the desired porphyrin was collected. This procedure was used to prepare 5.

o 5: mp > 320 C. Yield: 48%. UV-Vis λmax (CH2Cl2)/nm 454 (log ε 5.67), 579 (4.68),

1 618 (4.60); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.74 (4H, s, -H), 8.42 (8H, m, o-pyridine), 7.86 (8H, d, J = 8.0 Hz, o-Ph-H), 7.56 (8H, d, J = 8.0 Hz, m-Ph-H), 7.17 (4H, s, fused-benzene-H), 6.91(8H, m, m-pyridine-H), 3.13-3.18 (4H, m,

13 isopropyl(CH)-H), 1.41 (24H, d, J = 7.0 Hz, isopropyl (CH3)-H); C-NMR (500 MHz,

CDCl3, Me4Si) δ 24.28, 29.70, 34.17, 113.78, 116.91, 118.08, 120.15, 124.81, 126.18,

137

131.61, 132.66, 134.39, 136.52, 137.59, 139.96, 142.38, 149.05, 149.46, 149.66; Calculated Mass, 1246.49, Found MS (LDI-TOF), m/z 1246.09. General procedure for the methylation reaction of tetrabenzoporphyrins

To a solution of porphyrin 4c (25 mg, 16 mol) in methanol (3 mL) was added an excess of iodomethane (1 mL, 16 mmol). The mixture was stirred vigorously at room temperature for two days. The solvent was evaporated under vacuum. Recrystallization from methanol and chloroform gave brownish crystals of the title porphyrin 7c (41mg, 96%). To a solution of porphyrin 4d (25 mg, 15 mol) in methanol (3 mL) was added an excess of iodomethane (1 mL, 16 mmol). The mixture was stirred vigorously at room temperature for two days. The solvent was evaporated under vacuum. Recrystallization from methanol and chloroform gave brownish crystals of the title porphyrin 7d (40mg, 95%).

1 7c: UV-Vis λmax (methanol)/nm 524 (log ε 5.29), 694 (4.20); H-NMR (500 MHz,

CD3OD, Me4Si) δ 8.82-8.85 (16H, m), 8.52-8.76 (8H, m), 7.89-7.93 (8H, m),

7.75-7.76 (18H, m), 7.50-7.58 (6H, m), 4.40-4.42 (24H, s, -CH3), 3.34 (4H, m, isopropyl(CH)-H), 1.46 (24H, d, J = 6.5 Hz, isopropyl (CH3)-H). Calculated Mass,

8+ 7+ 6+ 2735, Found MS (LDI-TOF), m/z 1720 (M) , 1704 (M-CH3) , 1690 (M-2CH3) ,

5+ 4+ 3+ 2+ 1676 (M-3CH3) , 1662 (M-4CH3) , 1647 (M-5CH3) , 1632 (M-6CH3) , 1614

138

+ (M-7CH3) , 1600 (M-8CH3).

1 7d: UV-Vis λmax (methanol)/nm 530 (log ε 5.30), 653 (3.98), 698 (4.23); H-NMR

(500 MHz, CD3OD, Me4Si) δ 8.82-8.83 (16H, m), 7.87-8.31 (16H, m), 7.73-7.74

(16H, m), 7.32-7.46 (8H, m), 4.35-4.39 (24H, s, -CH3), 3.39 (4H, m, isopropyl(CH)-H), 1.42 (24H, d, J = 6.5 Hz, isopropyl (CH3)-H). Calculated Mass,

6+ 5+ 2799, Found MS (LDI-TOF), m/z 1692 (M-Zn-2CH3) , 1678 (M-Zn-3CH3) , 1663

4+ 3+ 2+ + (M-Zn-4CH3) , 1650 (M-Zn-5CH3) , 1635 (M-Zn-6CH3) , 1619 (M-Zn-7CH3) ,

1603 (M-Zn-8CH3). 4.9 X-ray Crystal Structures

Crystal of compound 4a suitable for X-ray diffraction analysis was obtained by slow evaporation from CHCl3/pyridine in a ratio of 1:30. Crystallographic data have been deposited with the Cambridge Crystallographic Data Centre (CCDC 865552). Copies of the data can be obtained free of charge on application to the CCDC, 12 Union Road, Cambridge CB21EZ, UK (fax: (+44)-1223-336-033; e-mail: [email protected]. Crystal data for 4a:

139

140

NSD data for 4a:

NSD result generated from file Hong.pdb at Sun Mar 25 00:02:54 MDT 2012

Summary of the NSD (in A):

NSD result generated from file untitled.pdb at Sun Mar 25 18:30:19 MDT 2012

Summary of the NSD (in A):

141

basis Dip dip B2g B1g Eu(x) Eu(y) A1g A2g min. 0.0936 0.0345 -0.0304 0.0152 -0.0136 -0.0173 0.0841 0.0062 ext. 0.1363 0.0271 -0.0302 0.0150 -0.0132 -0.0171 0.0915 0.0048 -0.0122 -0.0116 0.0152 0.0102 -0.0592 -0.0749 0.0121 -0.0050 0.0019 0.0000 0.1428 -0.0037 0.0043 -0.0036 0.0103 0.0086 0.0570 -0.0012 0.0033 0.0110 0.0074 0.0032 0.0146 -0.0182 0.0010 0.0146 -0.0072 0.0047 -0.0173 0.0014 -0.0018 0.0060 0.0008 -0.0014 -0.0010 -0.0052 0.0000 -0.0016 -0.0057 comp. 0.2092 0.0000 0.0354 0.0272 0.0266 0.0233 0.1859 0.0772 basis Doop doop B2u B1u A2u Eg(x) Eg(y) A1u min. 2.0650 0.0278 -2.0109 0.4647 -0.0214 -0.0452 0.0072 -0.0449 ext. 2.0713 0.0051 -2.0079 0.4647 -0.0205 -0.0463 0.0075 -0.0449 0.1542 -0.0346 0.0219 -0.0225 0.0039 0.0079 0.0017 0.0041 -0.0012 -0.0274 -0.0048 0.0078 -0.0029 0.0030 0.0101 comp. 2.0715 0.0000 2.0168 0.4660 0.0306 0.0579 0.0142 0.0456

4.10 Beer’s Law Experiment

Compound 4a was chosen to perform the Beer’s law experiment. 1mg of 4a was dissolved in 10 mL of DCM. 0.1 mL of this solution was dissolved in 1 mL, 1.5 mL, 2 mL, 2.5 mL, 3 mL and 4 mL respectively. The absorption was measured for these solutions using a UV-vis spectrophotometer. Graph 1 displays the absorption vs. the concentration. The trendline equation is y = 0.2738x + 0.2959 with an R2 value equal to 0.9781. So the porphyrin is monomeric.

142

Beer's law experiment on 4a

2.5

2 y = 0.2738x + 0.2959 R2 = 0.9781 1.5

1 Absorption

0.5

0 0 1 2 3 4 5 6 7 concentration (*10^-6 mol/L)

Graph 1. Beer’s law experiment on 4a.

Spectroscopy Data

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

Chapter 5: Pentacene-Fused Diporphyrins

Abstract

Constructing conjugated multichromophores in one molecule is very challenging. The successful development of a Pd0-catalyzed cascade reaction of acrolein and dibromoporphyrins allowed us to prepare a pentacene-fused diporphyrin and quinone-fused dinaphtho[2,3]porphyrins. A highly atom-economy multicomponent reaction fusing two rings tandemly to the porphyrin periphery in one operation was also discovered in this work. Both the cross-conjugated quinone-fused dinaphtho[2,3]porphyrins and the linearly conjugated pentacene-fused diporphyrin have been synthesized. Furthermore, unsymmetrical cross-conjugated quinone-fused dinaphtho[2,3]porphyrins and a triporphyrin have also been obtained using a similar approach. All these diporphyrins and triporphyrin displayed remarkably perturbed broad absorption bands spreading over the whole visible region of the spectrum, suggesting strong electronic interactions among these multichromophoric systems. The pentacene-fused diporphyrin is remarkably more stable than the relevant pentacene and heptacene derivatives, indicating that these novel materials hold potential for the applications in various areas.

5.1 Introduction

Acenes consisting of linearly fused benzene rings have been of great interest due to their unique electronic properties associated with their -bond topology as well as

159 their close relation to graphene.1, 2 Organic electronic materials based on acenes show promise to make revolutionary transformations in a range of applications including photovoltaics, organic field-effect transistors (OFETs), light emitting devices (OLEDs), etc. For examples, organic semiconductors derived from pentacene have achieved record high thin-film transistor mobility;3 more recently, pentacene-doped p-terphenyl has been shown to be effective as the amplifier medium for a room-temperature maser.4 Increasing the length of the acenes are expected to lead to rapid evolution of electronic properties due to their predicted decreasing band gap, lower reorganization energy, and increasing charge carrier mobility. However, longer acenes suffer low stability and poor solubility, causing the synthesis and characterization of these materials extremely difficult. Much effort has been devoted to the synthesis, stabilization/functionalization and characterization of acenes longer than pentacene in recent years, and significant progress has been made on hexacene, heptacene, octacene and even nonacene.5 We are interested in increasing the length of the acenes by linearly fusing a pentacene or a longer acene to another aromatic chromophore such as porphyrin. The fusion of an acene with two or more porphyrins will tremendously extend the -conjugation, and will likely produce electronically highly interactive multiporphyrinic systems. On the other hand, -extended porphyrins, in which aromatic rings are fused to the porphyrin periphery through the , ’-positions, are attractive synthetic targets due to their unique set of photophysical and electrochemical properties.6-8 In particular, -extended porphyrins often show significantly red-shifted and broadened absorption bands, a much desired feature for various applications. Given the rich photophysical and optoelectronic properties of porphyrins and the unique set of electronic properties of acenes, such fused -systems are enticing. The availability of such systems will open a door to the exploration of novel organic materials with unprecedented electronic and photophysical properties.

160

5.2 Results and Discussion of Symmetrical Pentacene-Fused Diporphyrins

Will a pentacene-fused diporphyrin be stable enough to be prepared? Pentacene is the largest sufficiently stable acene for device studies. Unsubstituted pentacene decomposes slowly in the air in solid state and in minutes in solution. When a pentacene is fused to two porphyrins (1, Figure 5.2), two pyrrole rings are linearly fused to a pentacene making it into a heteroheptacene. Unsubstituted heptacene is very unstable and cannot be formed in solution. On the other hand, the decomposition pathways for acenes mainly involve either photoinduced endoperoxide formation followed by subsequent oxidation to the corresponding diketone, or “butterfly” dimerization.9 We speculate that the steric bulkiness brought with the fused porphyrins might be possible to slow down the dimerization pathway, and the ease? to-stack nature of porphyrins might be able to force the pentacene-fused diporphyrin to adopt face-to-face -stacking and thus retard the oxidation pathway. Based on these factors, we envisioned that pentacene-fused porphyrins might be stable enough to be prepared.

5.2.1 Synthesis of Diformylbenzoporphyrin 3a-c

A common approach to the synthesis of larger acenes is through nucleophilic functionalization of a relevant quinone precursor followed by reduction.10-12 In order to obtain the pentacene-fused porphyrins, quinone-fused diporphyrin 2 (Figure 5.2) must be prepared from a diformylbenzoporphyrin 3 (Figure 5.1). Diformylbenzo- porphyrin 3 is thus a critical precursor to make the pentacene-fused diporphyrin 1. Functionalization of the -positions of benzoporphyrins is very challenging due to the limited synthetic methods, particularly the lack of functionalization methods.6 We decided to take advantage of a concise and versatile method developed in our laboratory recently for the synthesis of functionalized benzoporphyrins.13 In this

161 method an alkene reacts with a , ’-dibromoporphyrin through a three-step cascade reaction involving a vicinal two-fold Heck reaction, six- electrocyclization, and subsequent aromatization. This methodology allows efficient synthesis of a number of novel functionalized benzoporphyrins. Based on this method, the direct Heck arylation of acrolein appears attractive (Figure 5.1). However, acrolein is reactive and will easily polymerize at the reaction temperature, the result of the Heck product is generally low14. Initially, we tried the Heck-based cascade reaction of nickel dibromoporphyrin 4a with acrolein using the conditions originally developed for this reaction. As expected, the reaction turned out to be very messy with no major product identified15. We then attempted reactions using different solvents and bases, resulting in either recovery of the starting porphyrin or a complex mixture. When the palladium catalyst changed from Pd(OAc)2 to bis(tri-tert-butylphosphine) palladium(0), a clean reaction occurred giving one major product in 35% yield. This product, however, turned out not to be the desired diformylbenzoporphyrin 3a. 1H and 13C NMR spectroscopy analysis, and LDI-TOF mass spectrometry suggest a multicomponent cascade/or multistep reaction happened leading to the formation of 1, 3-cyclopentanedione-fused benzoporphyrin 5. The structure of 5 was further confirmed by X-ray crystallography (Figure 5.1). This unexpected reaction, in which a six-membered aromatic ring and a five-membered ring were fused to the porphyrin periphery sequentially in one operation, is highly atom-economic; in particular, the simultaneous introduction of two ketone groups to the ring in this reaction offers the opportunity for further functionalization of the molecules. This reaction has potential applications in natural product synthesis, pharmaceutical industry, and organic materials science. To our knowledge, such kind of reaction has never been reported in the literature. Although the mechanism of this reaction remains unclear, our study has indicated that the color of the bis(tri-tert-butylphosphine) palladium(0), which is supposed to be off-white, must turn red; the presence of DMF (N, N-dimethylformaldehyde) is also necessary for the reaction to occur. Further

162 investigation of this reaction is underway in our laboratory. In order to further investigate the possibility of acrolein and bis(tri-tert-butylphosphine) palladium(0) to synthesize the desired diformylabenzoporphyrin, we screened the reaction conditions by combining different bases and solvents. We also purchased the bis(tri-tert-butylphosphine) palladium(0) catalyst from a different vendor, which was received as an off-white powder. When the solvent was changed to toluene only, this reaction can be separated into two steps involving an effective vicinal two-fold Heck reaction followed by oxidation of the

Heck product (6a) using tetrachloro-p-benzoquinone (p-chloranil) to give 3a in 50% yield in two steps. The same conditions also worked well for zinc and free base dibromoporphyrin 4b and 4c to give diformylbenzoporphyrins 3b and 3c, respectively. Alternatively, 3a could also be obtained from a benzoporphyrin diester 7a, which was easily prepared through the Heck-based cascade reaction. Reduction of 7a to 8a using di(tert-butyl)aluminium hydride (DIBAL-H) followed by oxidation with Dess– Martin periodinane produced diformylbenzoporphyrin 3a in 29% yield in three steps.

163

Figure 5.1 The synthesis of diformylbenzoporphyrin from different pathways.

5.2.2 Synthesis of Pentacene-Fused Diporphyrins

Having the diformylbenzoporphyrins (3a-c) in hands, we proceeded to prepare the quinone precursor 2 (Figure 5.2). Treatment of 3a-c with 1,4-cyclohexanedione in the presence of KOH afforded 2a-c in 55-70% yield. While 2a is highly soluble in common organic solvents including chloroform, dichloromethane, pyridine and THF, the solubility of 2b and 2c are relatively low due to aggregation at higher concentration.

164

Figure 5.2 Synthesis of pentacene-fused diporphyrin 1.

A single crystal suitable for X-ray analysis was obtained by vapor diffusion of cyclohexane into a solution of 2a in toluene. The two porphyrin macrocycles of 2a (Figure 5.3) are significantly distorted from planarity assuming ruffled conformations, while the quinone fusion component between the two porphyrin rings is essentially planar.

Figure 5.3 X-ray crystal structure of 2a: top, edge view; bottom, top view.

We decided to first attempt the preparation of 6, 13-disubstituted pentacene-fused diporphyrins, as 6, 13-disubsituted pentacenes are reportedly more stable than the

165 unsubstituted pentacene.1 When 2a was treated with ethynyltriisopropylsilane, however, no reaction occurred with major recovery of the starting 2a, likely due to the steric hindrance brought by both the meso-aryl groups of the porphyrin and the bulky ethynyltriisopropylsilane; replacing ethynyltriisopropylsilane with smaller phenyllithium resulted in quantitative conversion of 2a to addition products 9a; the isomeric mixture of 9a was not further isolated. Treatment of the isomeric mixture of

9a with SnCl2 in the presence of HCl under strictly air-free condition afforded a solid. This solid was not soluble in the reaction mixture (DCM/water, v/v, 10/3). This solid was then purified by washing with oxygen-free water and methanol. LDI-Tof mass spectrometry of this solid showed a single peak at 2051.079, indicating the existence

1 of pentacene-fused diporphyrin 1. H NMR (C6D6) of this compound revealed only one set of 1H NMR signals in which two singlet peaks corresponding to the protons attached to the pentacene moiety can be found at 7.08 and 8.74 ppm. These peaks are shifted from 7.54 and 8.61 ppm in the parent 2a due to the reduction of the quinone to the acene (see Experimental Part). In addition, the integrated numbers of the aromatic protons correlates well with 1, further confirming the identification of 1. The analysis of the COSY spectroscopy of this compound is also consistent with that of its 1H NMR spectroscopy (see Experimental Part). Based on these data, we believe that pentacene-fused diporphyrin 1 was obtained.

5.2.3 Absorption and Emission Spectra

The UV-Vis absorption spectra of 2a-c and 1 are shown in Figure 5.4 (a). Quinone-fused porphyrins 2a-c displayed unusual characteristics of absorption bands with no well-defined Soret band and Q bands typical of metallated porphyrins, indicating strong interaction between the two fused porphyrin chromophores. The absorption bands of 2a-c are significantly broadened compared to the diformylbenzoporphyrin precursors 3a-c, covering a large range of wavelength from 350 to 650 nm on the spectrum. The metals inside the porphyrin rings also played a

166 role in the optical properties of 2a-c, resulting in different shapes of the broad absorption bands. Compared to its cross-conjugated precursor 2a, the absorption spectrum of the linearly conjugated diporphyrin 1 showed different characteristics. Five absorption bands at 436 nm, 545 nm, 617 nm, 730 nm and 812 nm were observed on the spectrum, noting the near-IR absorption at 730 and 812 nm. The Soret band, although much broadened, is better defined than that of 2a. While 2a, 2c and 1 were not fluorescent, 2b showed emission bands at 664 and 721 nm (Figure 5.4, (a), inset), in sharp contrast with the quinone-fused zinc porphyrins reported recently by the Kräutler group which were not fluorescent.7

167

Figure 5.4 (a) Normalized UV-vis absorption spectra of porphyrins 2a-c and 1 in CH2Cl2 solution;

inset, fluorescent spectrum of 2b (excitation wavelength: 428 nm). (b) UV-vis absorption spectra

of porphyrin 1 in benzene recorded at different times.

5.2.4 Stability Study of Pentacene-Fused Diporphyrin 1

While 1 is stable in the air in solid state for more than 3 months if shielded from ambient light, it decomposes in solution much more quickly. We performed stability study of 1 by tracking the UV-Vis absorption change in benzene and in dichloromethane (Figure 5.4 (b)). As it turned out, pentacene-fused diporphyrin 1 is significantly more stable than heptacene derivatives, being comparable to the more stable 1, 13-aryl-disubstituted pentacene derivatives.16 The half-life time of 1 in benzene solution is about 400 min in benzene and the half-life time of 1 in dichloromethane is much shorter, about 75 min. Since the pentacene component is 6,13-disubstituted, the decomposition pathway is likely limited to oxidation only. As shown in Figure 5.4, as 1 was decomposing, the absorption bands at 545 nm, 617 nm, 730 nm and 812 nm were changing; at the same time, the Soret band remained almost unchanged. When the bands at 730 nm and 812 nm were completely disappeared, the absorption spectrum looked like a normal porphyrin, consistent with the breakage of the conjugation between the two porphyrin monomers upon oxidation of the pentacene component. Corresponding to the change of the UV-Vis absorption spectrum, changes in the 1H NMR and LDI-TOF spectra of 1 were also observed (see

Experimental part).

5.2.5 Investigation of Direct Reduction of 2a with LiAlH4

Further investigation of direct reduction by LiAlH4 was carried out on 2a through a published procedure.12 (Figure 5.5) First, the two carbonyl groups of 2a are reduced to alcohols. After acidification, water is removed from the formed diol to afford 10a-b.

168

10a was characterized through 1H NMR spectroscopy and LDI-TOF mass spectrometry. The loss of a carbonyl group and the addition of an alcohol group in the IR spectrum further confirmed the presence of 10a. The second reduction step was carried out on the crude mixture of 10a and 10b to give 11. 11 is soluble in chloroform and benzene, 11 can decompose rapidly in solution and air to form a butterfly dimer 12. (see experimental part for details)

Figure 5.5 Reduction of 2a.

In benzene, the UV-vis spectrum of 11 did not change when exposed to light and the airin several days. On TLC (toluene : cyclohexane, v:v, = 1 : 2), there were two inseparable bands running together with a brownish color (11-1) on the leading edge and a greenish color (11-2) on the bottom edge of the band. In benzene the two compounds are very stable. No change was observed in LDI-TOF MS or UV-vis spectra after the solutions were exposed to the atmosphere and light for one day.

While in CH2Cl2 the color changed from greenish to brownish very quickly, approximately in five minutes, upon exposure to light and air. The TLC of this sample showed only the brownish band, which gave the same Rf value as 11-1, indicating the

169 conversion of 11-2 to 11-1 Based on our knowledge of pentacene and its derivatives, we assign this brown layer to a butterfly porphyrin dimer 12. (See Experimental part in details)

5.3 Results and Discussion of Unsymmetrical Diporphyrins and Triporphyrins

In the last section, fully cross-conjugated porphyrin dimers 2a-c were synthesized. 2a-c displayed remarkably broadened absorption bands spreading over the whole visible region of the spectrum, suggesting strong electronic interactions in these multichromophoric systems. So we plan to make unsymmetrical porphyrin dimers, in which one porphyrin serves as a donor and the other one serves as an acceptor; further extending the -conjugated system to make a porphyrin trimer can further change the electronic interactions within the multichromophoric systems, leading to even more interesting electronic properties.

5.3.1 Synthesis of Unsymmetrical Diporphyrins

In order to make unsymmetrical diporphyrins, 13a is the key synthetic intermediate. (Figure 5.6) Unfortunately 13a is very unstable and can be oxidized to 13b quickly when exposed to the air.11, 17 Even in the solid state13a is still not very stable and can undergo oxidation reaction slowly. In order to investigate the feasibility of this approach, we first tried the aldol reaction of 3a with large excess of 1,4-cyclohexanedinone (20 equiv.). (Figure 5.6) We found that the amounts of based usd (aqueous KOH) and the use of anaerobic conditions are very important in this aldol reaction. The followings are the results from several representative conditions investigated. First, when 0.3 equivalent of KOH aqueous was used in this reaction without exclusion of air, 13b was the only major product. Second, when 0.3 equivalent of aqueous KOH was used under argon protection, 13a and 13c were the

170 major products. Based on the LDI-TOF mass spectra, the MW of 13c is 16 numbers higher than that of 13a, so it is very likely the aldol addition products before dehydration. This is further confirmed by the second aldol reaction of 13c with 3b leading to the formation of the unsymmetrical dimer 15a. (Figure 5.7) Third, when three equivalents of aqueous KOH were used under argon protection, the reaction color changed from brown to green immediately. The resulting mixture can be dissolved in methanol and water easily, but not in DCM or chloroform. Based on this phenomenon and the LDI-TOF MS, it is likely that the third condition produced a complex mixture with the self-oxidation and reduction products of 13a. In the end, we chose the second procedure for further investigation.

Figure 5.6 Investigation of the aldol reaction of 3a.

Having obtained 13a and 13c, we then pursued the second aldol reaction of 3b and 14 to get the unsymmetrical diporphyrins 15a and 15b. (Figure 5.27) It is worth

171 to mention that when porphyrin 14 was used as the donor, the nitro group could be removed under the executed reaction conditions to give the symmetrical diporphyrin 2a as a minor product. Both 15a and 15b diporphyrins displayed similar perturbed broad absorption bands spreading over the whole visible region of the spectrum. (Figure 5.8) Unsymmetrical 15b displayed red-shifted emission bands relative to that of the symmetrical 2b at near IR region. (Figure 5.9)

172

Figure 5.7 Synthesis of two unsymmetrical diporphyrins 15a and 15b.

1.00 2a 15a 15b 0.75

0.50 Absorption

0.25

0.00 350 375 400 425 450 475 500 525 550 575 600 625 650 675 700 Wavelength (nm)

Figure 5.8 UV-visible spectra of different cross-conjugated diporphyrins (2a, 15a and 15b) in

toluene.

500 450 2b (ex @ 430nm) 400 350 15b (ex @ 421nm) 300 250

Intensity 200 150 100 50 0 575 600 625 650 675 700 725 750 775 800

Wavelength (nm)

Figure 5.9 Fluorescence spectra of 2b and 15b.

5.3.2 Synthesis of Triporphyrins

The successfully synthesis of the key starting materials 13a and 13c provides an ideal template to conveniently fuse more porphyrin rings on it, if proper substituents

173 are installed. Thus, tetraformylbenzoporphyrin 16 was synthesized. When we tried the Pd(0) catalyzed cascade reaction of the corresponding tetrabromoporphyrin, the reaction worked but the yield was very low (around 10%). We then tried an alternative method based on a reduction-reoxidation pathway in order to obtain larger amount of 16. Thus, triporphyrin 17 was prepared through the aldol reaction. (Figure 5.10) 1HNMR spectra and LDI-TOF MS confirmed the formation of triporphyrin 17. (see Experimental section for details) UV-visible spectrum of 17 displayed further red-shifted absorption bands relative to that of diporphyrin 2a. (Figure 5.11)

Figure 5.10 Synthesis of triporphyrin 17.

174

1.00 2a 17 0.75

0.50 Absorption 0.25

0.00 400 450 500 550 600 650 700 750 Wavelength (nm)

Figure 5.11 UV-visible spectra of 2a and 17 in toluene.

Reference： 1. Zade, S. S.; Bendikov, M., Journal of Physical Organic Chemistry 25 (6), 452-461; Sun, Z.; Wu, J. S., Australian Journal of Chemistry 64 (5), 519-528; Anthony, J. E., Chem. Rev. 2006, 106 (12), 5028-5048; Anthony, J. E., Angew. Chem. Int. Ed 2008, 47 (3), 452-483; Anthony, J. E.; Facchetti, A.; Heeney, M.; Marder, S. R.; Zhan, X. W., Adv. Mater. 22 (34), 3876-3892. 2. Bendikov, M.; Wudl, F.; Perepichka, D. F., Chem. Rev. 2004, 104 (11), 4891-4945. 3. Okamoto, T.; Bao, Z. A., J. Am. Chem. Soc 2007, 129 (34), 10308-+. 4. Oxborrow, M.; Breeze, J. D.; Alford, N. M., Nature 488 (7411), 353-357. 5. Payne, M. M.; Parkin, S. R.; Anthony, J. E., J. Am. Chem. Soc 2005, 127 (22), 8028-8029; Zade, S. S.; Bendikov, M., Angew. Chem. Int. Ed 49 (24), 4012-4015; Mondal, R.; Tonshoff, C.; Khon, D.; Neckers, D. C.; Bettinger, H. F., J. Am. Chem. Soc 2009, 131 (40), 14281-14289; Watanabe, M.; Chang, Y. J.; Liu, S. W.; Chao, T. H.; Goto, K.; Islam, M. M.; Yuan, C. H.; Tao, Y. T.; Shinmyozu, T.; Chow, T. J., Nature Chemistry 4 (7), 574-578; Qu, H. M.; Chi, C. Y., Org. lett 12 (15), 3360-3363; Kaur, I.; Jazdzyk, M.; Stein, N. N.; Prusevich, P.; Miller, G. P., J. Am. Chem. Soc 132 (4), 1261-+; Kaur, I.; Stein, N. N.; Kopreski, R. P.; Miller, G. P., J. Am. Chem. Soc 2009, 131 (10), 3424-+; Chun, D.; Cheng, Y.; Wudl, F., Angew. Chem. Int. Ed 2008, 47 (44), 8380-8385; Mondal, R.; Shah, B. K.; Neckers, D. C., J. Am. Chem. Soc 2006, 128 (30), 9612-9613; Xiao, J. C.; Duong, H. M.; Liu, Y.; Shi, W. X.; Ji, L.; Li, G.; Li, S. Z.; Liu, X. W.; Ma, J.; Wudl, F.; Zhang, Q. C., Angew. Chem. Int. Ed 51 (25), 6094-6098; Tonshoff, C.; Bettinger, H. F., Angew. Chem. Int. Ed 49 (24), 4125-4128. 6. Cheprakov, A. V.; Filatov, M. A., J. Porphyrins&Phthalocyanines 2009, 13 (3), 291-303. 7. Banala, S.; Ruhl, T.; Wurst, K.; Krautler, B., Angew. Chem. Int. Ed 2009, 48 (3), 599-603. 8. Banala, S.; Huber, R. G.; Muller, T.; Fechtel, M.; Liedl, K. R.; Krautler, B., Chem. Communs 48

175

(36), 4359-4361; Akita, M.; Hiroto, S.; Shinokubo, H., Angew. Chem. Int. Ed 51 (12), 2894-2897; Baluschev, S.; Yakutkin, V.; Miteva, T.; Avlasevich, Y.; Chernov, S.; Aleshchenkov, S.; Nelles, G.; Cheprakov, A.; Yasuda, A.; Mullen, K.; Wegner, G., Angew. Chem. Int. Ed 2007, 46 (40), 7693-7696; Borek, C.; Hanson, K.; Djurovich, P. I.; Thompson, M. E.; Aznavour, K.; Bau, R.; Sun, Y. R.; Forrest, S. R.; Brooks, J.; Michalski, L.; Brown, J., Angew. Chem. Int. Ed 2007, 46 (7), 1109-1112; Gottumukkala, V.; Ongayi, O.; Baker, D. G.; Lomax, L. G.; Vicente, M. G. H., Bioorg. Med. Chem. 2006, 14 (6), 1871-1879; Mack, J.; Bunya, M.; Shimizu, Y.; Uoyama, H.; Komobuchi, N.; Okujima, T.; Uno, H.; Ito, S.; Stillman, M. J.; Ono, N.; Kobayashi, N., Chem. Eur. J. 2008, 14 (16), 5001-5020; Yamada, H.; Kuzuhara, D.; Takahashi, T.; Shirnizu, Y.; Uota, K.; Okujima, T.; Uno, H.; Ono, N., Org. lett 2008, 10 (14), 2947-2950; Finikova, O. S.; Aleshchenkov, S. E.; Brinas, R. P.; Cheprakov, A. V.; Carroll, P. J.; Vinogradov, S. A., J. Org. Chem 2005, 70 (12), 4617-4628; Lash, T. D., J. Porphyrins&Phthalocyanines 2001, 5 (3), 267-288. 9. Berg, O.; Chronister, E. L.; Yamashita, T.; Scott, G. W.; Sweet, R. M.; Calabrese, J., J. Phys. Chem . A 1999, 103 (14), 2451-2459. 10. Payne, M.; Odom, S.; Parkin, S.; Anthony, J., Organic Letters 2004, 6 (19), 3325-3328. 11. Swartz, C.; Parkin, S.; Bullock, J.; Anthony, J.; Mayer, A.; Malliaras, G., Organic Letters 2005, 7 (15), 3163-3166. 12. Vets, N.; Smet, M.; Dehaen, W., Tetrahedron Letters 2004, 45 (39), 7287-7289. 13. Deshpande, R.; Jiang, L.; Schmidt, G.; Rakovan, J.; Wang, X.; Wheeler, K.; Wang, H., Organic Letters 2009, 11 (19), 4251-4253. 14. Battistuzzi, G.; Cacchi, S.; Fabrizi, G., Org. lett 2003, 5 (5), 777-780; Noel, S.; Djakovitch, L.; Pinel, C., Tetrahedron Letters 2006, 47 (23), 3839-3842; Pan, K.; Noel, S.; Pinel, C.; Djakovitch, L., Journal of Organometallic Chemistry 2008, 693 (17), 2863-2868. 15. Deshpande, R.; Jiang, L.; Schmidt, G.; Rakovan, J.; Wang, X. P.; Wheeler, K.; Wang, H., Org. lett 2009, 11 (19), 4251-4253. 16. Kaur, I.; Jia, W. L.; Kopreski, R. P.; Selvarasah, S.; Dokmeci, M. R.; Pramanik, C.; McGruer, N. E.; Miller, G. P., J. Am. Chem. Soc 2008, 130 (48), 16274-16286. 17. Ashworth, G.; Berry, D.; Smith, D. C. C., Journal of the Chemical Society, Perkin Transactions 1 1979, 0 (0), 2995-3000.

176

5.4 Experimental Section

General All reagents, unless otherwise stated, were obtained either from Sigma-Aldrich or ACROS. Analytical thin layer chromatography (TLC) was performed on Silicycle UltraPure Silica Gel 60 F254 TLC plates. Preparative column chromatography was performed on silica gel (1000m), which was purchased from Silicycle. 1H and 13C experiments were conducted on a Bruker Avance 500MHz spectrometer. All samples were prepared in CDCl3 and chemical shifts were referenced to CDCl3 at 7.24ppm for

1 13 H NMR and referenced to CDCl3 at 77 ppm for C-NMR unless otherwise stated. UV-Visible spectra were recorded on an Agilent 8453 UV-Visible spectrometer in

CH2Cl2. Mass spectra were obtained on Bruker Matrix-assisted laser desorption/ionization time-of-flight (LDI-TOF) mass spectrometer. M.P.’s were measured on an Electrothermal MEL-TEMP apparatus and were uncorrected.

Synthesis of 5 via a cascade reaction.

Synthesis of 5: Dibromoarylporphyrin 4a (0.045 mmol), bis(tri-turt-butylphosphine) palladium(0) (purchased from ACROS and received as red particles) (0.012 mmol), and K2CO3 (0.09 mmol) were added to a Schlenk flask and dried under vacuum. The vacuum was released under argon to allow the addition of dry DMF (10 mL) and dry xylene (10 mL). The mixture was then degassed via four freeze-pump-thaw cycles and purged with argon again. 25-fold excess of acrolein (1.125 mmol) was then added

177 into the Schlenk flask under argon. The Schlenk flask was sealed and heated to reflux for 72h. After 72 h, the mixture was diluted with CHCl3 and washed with water. The organic layer was removed under vacuum. The residue was subjected to silica column chromatography (CHCl3 and cyclohexane in 1:1 ratio as the solvent system). The band containing the desired porphyrin was collected and recrystallized from CHCl3 and methanol. A suitable crystal was obtained by slow evaporation in chloroform.

o Characterization of 5: mp > 320 C. Yield: 35%. UV-Vis λmax (CH2Cl2)/nm 446 (log ε

1 5.36), 552 (4.35), 587 (4.17); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.70-8.82 (6 H, m, β-H), 7.81-7.89 (8H, two sets of d, J = 8 Hz, o-Ph-H on meso-phenyl ring), 7.50-7.63 (8H, two sets of d, J = 8 Hz, m-Ph-H on meso-phenyl ring), 7.27 (2H, s, fused benzene-H), 3.24-3.29 (4H, m, isopropyl(CH)-H on meso-phenyl ring),

3.15-3.20 (4H, m, CH and CH3 on the fused five membered ring), 1.46-1.59 (24H,

13 two sets of d, J = 7 Hz, isopropyl (CH3)-H) ; C-NMR (500 MHz, CDCl3, Me4Si) δ 24.20, 24.41, 34.08, 34.48, 116.58, 119.53, 120.34, 125.18, 126.37, 127.79, 131.95, 132.24, 132.65, 132.93, 133.64, 136.65, 137.44, 137.49, 141.83, 142.45, 143.40, 143.49, 148.55, 150.24, 168.66; Calculated Mass, 971.85, Found MS (LDI-TOF), m/z 971.54.

178

1 H NMR spectrum of 5 (500 MHz in CDCl3)

13 C NMR spectrum of 5 (500 MHz in CDCl3)

179

LDI-TOF MS of 5.

Synthetic Steps Involved in the Preparation of 3a through Methyl Acrolate Approach.

Synthesis of 7a: Dibromoarylporphyrin 4a (0.045 mmol), palladium acetate (0.012

180 mmol), triphenylphosphine (0.030 mmol) and K2CO3 (0.09 mmol) were added to a Schlenk flask and dried under vacuum. The vacuum was released under argon to allow the addition of dry DMF (10 mL) and dry xylene (10 mL). The mixture was then degassed via four freeze-pump-thaw cycles before the vessel was purged with argon again. 25-fold excess of methyl acrolate (1.125 mmol) was then added into the flask under argon. The Schlenk flask was sealed and heated to reflux for 72h. After 72 h, the mixture was diluted with CHCl3 and washed with water. The organic layer was removed under vacuum. The residue was subjected to silica column chromatography

(CHCl3 and cyclohexane in 1:1 ratio as the solvent system). The band containing the desired porphyrin was collected and recrystallized from CHCl3 and methanol.

o Characterization of 7a: mp > 320 C. Yield: 55%. UV-Vis λmax (CH2Cl2)/nm 435 (log

1 ε 5.46), 547 (4.35), 579 (4.06); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.71-8.76 (6 H, m, β-H), 7.83-7.91 (8H, two sets of d, J = 8 Hz, o-Ph-H on meso-phenyl ring), 7.51-7.61 (8H, two sets of d, J = 8 Hz, m-Ph-H on meso-phenyl ring), 7.45 (2H, s, fused benzene-H), 3.86 (6H, s, COOMe-H) 3.15-3.27 (4H, m, isopropyl(CH)-H),

13 1.47-1.54 (24H, two sets of d, J = 7 Hz, isopropyl (CH3)-H) ; C-NMR (500 MHz,

CDCl3, Me4Si) δ 24.22, 24.38, 26.95, 34.08, 34.28, 52.49, 116.07, 120.41, 125.15, 126.14, 128.21, 131.60, 131.95, 132.74, 132.82, 133.66, 136.61, 137.62, 141.09, 141.46, 142.26, 143.82, 148.49, 149.61, 168.47; Calculated Mass, 1005.86, Found MS (LDI-TOF), m/z 1005.65.

181

1 H NMR spectrum of 7a (500 MHz in CDCl3)

13 C NMR spectrum of 7a (500 MHz in CDCl3)

182

LDI-TOF MS of 7a.

Synthesis of 8a: 7a (0.3 mmol) was dissolved in dry CH2Cl2. The temperature of the solution was lowered to -78oC. DIBAL-H (1M solution in hexane) (1.8 mmol) was added to the solution dropwise with an argon inlet. The resulting mixture was stirred at -78oC for 1 hour. Ice water was added to the mixture slowly and the organic layer was extracted by CH2Cl2. The organic layer was separated and dried. The residue was purified by silica column chromatography (CH2Cl2 as the solvent system).

o Characterization of 8a: mp > 320 C. Yield: 70%. UV-Vis λmax (CH2Cl2)/nm 430 (log

1 ε 5.38), 544 (4.22), 582 (3.95); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.67-8.75 (6 H, m, β-H), 7.79-7.90 (8H, two sets of d, J = 8 Hz, o-Ph-H on meso-phenyl ring), 7.49-7.57 (8H, two sets of d, J = 8 Hz, m-Ph-H on meso-phenyl ring), 6.90 (2H, s, fused benzene-H), 4.63 (4H, s, CH2-H), 3.15-3.23 (4H, m, isopropyl(CH)-H),

13 1.46-1.51 (24H, two sets of d, J = 7 Hz, isopropyl (CH3)-H) ; C-NMR (500 MHz,

CDCl3, Me4Si) δ 24.21, 24.49, 34.07, 34.30, 65.35, 115.25, 120.18, 125.06, 125.42, 126.04, 131.01, 131.48, 132.57, 132.77, 133.63, 136.41, 137.65, 137.81, 140.43, 140.91, 141.85, 142.26, 143.79, 148.35, 149.45; Calculated Mass, 949.84, Found MS (LDI-TOF), m/z 948.51.

183

1 H NMR spectrum of 8a (500 MHz in CDCl3).

13 C NMR spectrum of 8a (500 MHz in CDCl3).

184

LDI-TOF MS of 8a.

Synthesis of 3a: 8a (0.2 mmol) was dissolved in dry CH2Cl2, and then Dess-Martin periodinane (0.4 mmol) was added to the solution. The resulting mixture was stirred at room temperature for half an hour. Saturated sodium thiosulfate aqueous solution was then added to the mixture. The organic layer was separated and dried. The residue was subjected to silica column chromatography (CHCl3 and cyclohexane in a 2:3 ratio served as the solvent system). Characterization of 3a: mp > 320oC. Yield: 75%.

1 UV-Vis λmax (CH2Cl2)/nm 450 (log ε 5.32), 555 (4.36), 5.92 (4.22); H-NMR (500

MHz, CDCl3, Me4Si) δ 10.33 (2H, s, CHO-H), 8.71-8.83 (6 H, m, β-H), 7.83-7.89 (8H, two sets of d, J = 8 Hz, o-Ph-H on meso-phenyl ring), 7.63 (4H, d, J = 8 Hz, m-pH-H on meso-phenyl ring), 7.55 (2H, s, fused benzene-H), 7.52 (4H, d, J = 8 Hz, m-Ph-H on meso-phenyl ring), 3.15-3.28 (4H, m, isopropyl(CH)-H), 1.46-1.57 (24H,

13 two sets of d, J = 7 Hz, isopropyl (CH3)-H) ; C-NMR (500 MHz, CDCl3, Me4Si) δ 24.20, 24.23, 24.41, 34.06, 34.11, 34.37, 34.45, 116.76, 120.51, 125.17, 125.26, 126.32, 128.67, 128.75, 132.72, 133.65, 136.37, 137.39, 137.44, 142.04, 142.13, 142.63, 143.62, 148.61, 150.15, 192.48, 193.05; Calculated Mass, 945.81, Found MS (LDI-TOF), m/z 944.99.

185

1 H NMR spectrum of 3a (500 MHz in CDCl3).

13 C NMR spectrum of 3a (500 MHz in CDCl3).

186

LDI-TOF MS of 3a.

One Pot Synthesis of 3a-c through Acrolein Approach.

Synthesis of 3a-c: Dibromoarylporphyrin 4a-c (0.045 mmol), bis(tri-turt-butylphosphine) palladium(0) (purchased from Strem Chemicals Inc. and received as off-white powder) (0.045 mmol) and K2CO3 (0.09 mmol) were added to a Schlenk flask and dried under vacuum. The vacuum was released under argon to allow the addition of dry DMF (10 mL) and dry xylene (10 mL). The mixture was then degassed via four freeze-pump-thaw cycles before the vessel was purged with argon again. 20-fold excess of acrolein (0.90 mmol) was then added into the Schlenk

187 flask under argon. The Schlenk flask was sealed and heated to reflux for 48h. After

48h, the mixture was diluted with CHCl3 and washed with water. The organic layer was removed under vacuum without further purification. The residue was then redissolved in xylene and treated with tetrachloro-p-benzoquinone (0.5 mmol), the mixture was heated to reflux overnight. The solvent was removed under vacuum. The residue was subjected to silica column chromatography (CH2Cl2 and cyclohexane in a 2:3 ratio served as the solvent system). The band containing the desired porphyrin was collected and recrystallized from CHCl3 and methanol. The yield of 3a for two steps is

o 50%. Characterization of 3b: mp > 320 C. Yield: 42%. UV-Vis λmax (CH2Cl2)/nm

1 453 (log ε 5.34), 569 (4.37), 605 (3.95); H-NMR (500 MHz, CDCl3, Me4Si) δ 10.39 (2H, s, CHO-H), 8.89-8.96 (6 H, m, β-H), 8.05-8.11 (8H, two sets of d, J = 8 Hz, o-Ph-H on meso-phenyl ring), 7.75 (2H, s, fused benzene-H), 7.58-7.70 (8H, d, J = 8 Hz, m-Ph-H on meso-phenyl ring), 3.22-3.38 (4H, m, isopropyl(CH)-H), 1.53-1.64

13 (24H, two sets of d, J = 7 Hz, isopropyl (CH3)-H) ; C-NMR (500 MHz, CDCl3,

Me4Si) δ 24.31, 24.52, 34.11, 34.46, 119.07, 122.91, 124.70, 125.81, 126.64, 127.53, 128.34, 129.65, 131.83, 132.09, 132.42, 132.61, 133.26, 134.48, 139.82, 140.04, 141.77, 143.95, 148.18, 149.81, 150.03, 150.66, 151.53, 192.90, 192.94; Calculated Mass, 950.35, Found MS (LDI-TOF), m/z 950.13. Characterization of 3c: mp >

o 320 C. Yield: 46%. UV-Vis λmax (CH2Cl2)/nm 453 (log ε 5.34), 569 (4.37), 605

1 (3.95); H-NMR (500 MHz, CDCl3, Me4Si) δ 10.35 (2H, s, CHO-H), 8.72-8.93 (6 H, m, β-H), 8.08-8.13 (8H, two sets of d, J = 8 Hz, o-Ph-H on meso-phenyl ring), 7.60-7.71 (8H, d, J = 8 Hz, m-Ph-H on meso-phenyl ring), 7.55 (2H, s, fused benzene-H), 3.22-3.36 (4H, m, isopropyl(CH)-H), 1.52-1.62 (24H, two sets of d, J = 7

13 Hz, isopropyl (CH3)-H), 2.61 (2H, s, free base H) ; C-NMR (500 MHz, CDCl3,

Me4Si) δ 24.28, 24.49, 34.13, 34.47, 118.59, 121.47, 124.94, 126.00, 128.14, 128.44, 128.97, 133.23, 133.76, 134.36，134.73, 138.85, 139.05, 139.26，144.18， 148.51， 150.18， 192.92; Calculated Mass, 889.13, Found MS (LDI-TOF), m/z 889.46.

188

1 H NMR spectrum of 3b (500 MHz in CDCl3).

13 C NMR spectrum of 3b (500 MHz in CDCl3).

189

LDI-TOF MS of 3b.

1 H NMR spectrum of 3c (500 MHz in CDCl3).

190

13 C NMR spectrum of 3c (500 MHz in CDCl3).

LDI-TOF MS of 3c.

191

Synthesis of Aldol Porphyrin Dimer 2a-c. Synthesis of 2a-c: 3a-c (20 mg, 0.02 mmol) and 1, 4-cyclohexanedione (1.2 mg, 0.01 mmol) were dissolved in tetrahydrofuran (3 mL) in a round bottomed flask. 0.3 mL of

0.02 mol/L KOH/H2O was then added to the flask. The mixture was stirred at room temperature overnight. The solvent was removed under vacuum. The reaction mixture was diluted with CHCl3 and washed with water. The organic layer was removed under vacuum. The residue was then subjected to silica column chromatography (CH2Cl2 and cyclohexane in 1:2 ratio as the solvent system). The band containing the desired porphyrin was collected and characterized. 2a can dissolve in CH2Cl2 very easily. For 2b and 2c, the solubility is poor and aggregation occured while running the column.

o Characterization of 2a: mp > 320 C. Yield: 60%. UV-Vis λmax (CH2Cl2)/nm 429 (log

1 ε 5.05), 577 (4.53), 612 (4.63); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.59-8.69 (16 H, m), 7.81-7.87 (16H, two sets of d, J = 7.5 Hz), 7.63 (8H, d, J = 7 Hz), 7.54 (4H, s),

192

7.47 (8H, d, J = 8 Hz), 3.34-3.38 (4H, m, isopropyl(CH)-H), 3.11-3.17 (4H, m, isopropyl (CH)-H), 1.68 (24H, d, J = 7 Hz, isopropyl (CH3)-H), 1.44 (24H, d, J = 7 Hz, isopropyl (CH3)-H) ; Calculated Mass, 1931.69, Found MS (LDI-TOF), m/z 1931.80.

o Characterization of 2b: mp > 320 C. Yield: 45%. UV-Vis λmax (CH2Cl2)/nm 433 (log ε 5.04), 586 (4.43), 619 (4.31); Calculated Mass, 1945.06, Found MS (LDI-TOF), m/z

o 1945.03. Characterization of 2c: mp > 320 C. Yield: 50%. UV-Vis λmax (CH2Cl2)/nm 431 (log ε 5.05), 493 (4.87), 535 (4.63), 582 (4.51), 612 (4.41); 1H-NMR (500 MHz,

CDCl3, Me4Si) δ 8.79-8.90 (12 H, m), 8.64-8.70 (4H, m), 8.10-8.14 (12H, m), 8.00 (2H, s), 7.80-7.81 (8H, m), 7.59-7.61 (10H, m), 7.45 (2H, s), 7.03 (2H, s), 3.43-3.47

(8H, m, isopropyl(CH)-H), 1.80-1.82 (24H, m, isopropyl (CH3)-H), 1.69-1.72 (24H, m, isopropyl (CH3)-H); Calculated Mass, 1818.33, Found MS (LDI-TOF), m/z 1818.12.

1 H NMR spectrum of 2a (500 MHz in CDCl3).

193

LDI-TOF MS of 2a.

LDI-TOF MS of 2b.

194

1 H NMR spectrum of 2c (500 MHz in CDCl3).

LDI-TOF MS of 2c.

195

UV-vis spectra

1.00

2a 0.75 2b

0.50 2c Absorption 0.25

0.00 350 400 450 500 550 600 650 700 750 Wavelength (nm)

Fluorescence spectra

200 2b (428ex)

150

100 Intensity

0 600 700 800 900 Wavelength (nm)

UV-vis Spectra of 2a-c and Fluorescence Spectrum of 2b (exited at 428nm).

196

Synthesis of 1.

Synthesis of 1: 2a (18 mg, 0.009 mmol) was added to a Schlenk flask and dried under vacuum. The vacuum was released under argon to allow the addition of dry THF (10 mL). The solution was then degassed via three freeze-pump-thaw cycles before the vessel was purged with argon again. 40-fold excess of phenyl lithium (0.18 mmol) was then added into the Schlenk flask in an ice bath and stirred for half an hour. The solution was removed from the ice bath and stirred at room temperature overnight.

The solvent was dried quickly under argon atmosphere. Then dry CH2Cl2 (10 mL) was added to the Schlenk flask and argon was bubbled for 15 minutes. Tin chloride (0.058 g, 0.27 mmol) was dissolved in 3 mL (3M) HCl, had argon bubbled for 15 minutes and was then transferred into the Schlenk flask under argon. The reaction mixture was stirred at room temperature for one hour. The mixture was dried under vacuum. The residue was washed twice with distilled water (argon bubbled for 5 minutes) and three times with methanol (argon bubbled for 5 minutes) under argon. A recrystallization was performed in CH2Cl2 and methanol and the product 1 was then dried under vacuum while shielded from light. Characterization of 1: mp > 320oC.

Yield: 62%. UV-Vis λmax (CH2Cl2)/nm 436 (log ε 5.07), 545 (4.68), 617 (4.35), 730

1 (4.00), 812 (3.94); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.65-8.74 (16 H, m, β-H and 4 singlet protons from the fused pentacene part), 7.97-7.99 (16H, m, 16 protons from meso-phenyl rings), 7.66-7.83 (8H, m, 8 protons from phenyl rings attached on the

197 pentacene part), 7.37-7.42 (18H, m, 16 protons from meso-phenyl rings and 2 protons from the two phenyl rings attached on pentacene part), 7.08 (4H, s, 4 protons from the fused pentacene part), 3.00-3.18 (8H, m, isopropyl(CH)-H), 1.38-1.54 (48H, two sets of d, J = 7 Hz, isopropyl (CH3)-H); Calculated Mass, 2050.82, Found MS (LDI- TOF), m/z 2051.08.

1H NMR spectrum of 1 (500 MHz in d-benzene).

198

COSY spectrum of 1 (500 MHz in d-benzene).

LDI-TOF MS of 1. 1 is stable as a solid when shielded from light. However 1 is not stable in a solution when exposed to light and air. If exposed to air and light for one day in benzene,

199 theLDI-TOF MS and 1H NMR spectra are as follows:

1H NMR spectrum ((after 1 was exposed to the air in solution, 500 MHz in d3-chloroform)

LDI-TOF MS (after 1 was exposed to the air in solution)

Stability study of this compound: freshly made solid 1 was dissolved in benzene, and exposed to ambient air and light, the absorption bands at the range of 500nm to 900nm decreased as time

200 passed by as shown in the UV-vis spectra. UV-vis Spectra of 1

1.00 1 fresh 1 15min 0.75 1 50min 1 90min 0.50 1 190min

Absorption 1 240min 0.25 1 400min 1 7140min 0.00 400 500 600 700 800 900 Wavelength (nm)

UV-vis Spectra of 1. (Stability study, 1 was dissolved in benzene) UV-vis Spectra

1.00 1 in benzene 1 in DCM 0.75

0.50 Absorption 0.25

0.00 400 500 600 700 800 900 Wavelength (nm)

UV-vis Spectra of 1 in different solvents.

201

Synthesis of 10a and 10b: 2a (20 mg, 0.01 mmol) was dissolved in dry tetrahydrofuran (10 mL) in a round bottomed flask, and cooled in an ice water bath under argon atmosphere. LiAlH4 (0.2 mmol) was then added to the flask dropwise. The whole mixture was then refluxed for 30 min. The mixture was cooled to room temperature and then cooled in an ice bath as HCl (6M, 10 mL) was added. The mixture was then refluxed for an additional 3 hours. The residue was separated with

CH2Cl2 and water. The organic layer was collected and dried under vacuum. The reaction mixture was then run on a preparative TLC plate (CHCl3 and cyclohexane 7:5 ratio served as the solvent system). One layer 10a was separated and fully characterized by 1H NMR, LDI-TOF, UV-vis and IR spectra. The other layer could not be fully characterized but has the same LDI-TOF peak as 10a. So we infer that tautomerization occurred resulting in the formation of 10b. Characterization of 10a:

o mp > 320 C. Yield: 35%. UV-Vis λmax (CH2Cl2)/nm 450 (log ε 5.10), 573 (4.46), 611

1 (4.55); H-NMR (500 MHz, CDCl3, Me4Si) δ 8.55-8.70 (12 H, m, -H), 7.84-7.89 (15H, m), 7.41-7.72 (21H, m), 6.92-7.15 (5H, m), 5.14 (1H, s, OH-H), 3.38-3.41 (4H, m, isopropyl(CH)-H), 3.15-3.17 (4H, m, isopropyl (CH)-H), 1.70 (24H, d, J = 7 Hz, isopropyl (CH3)-H), 1.44-1.46 (24H, m, isopropyl (CH3)-H) ; Calculated Mass,

202

1915.75, Found MS (LDI-TOF), m/z 1916.23.

1 H NMR spectrum of 10a (500 MHz in CDCl3).

LDI-TOF spectrum of 10a. Synthesis of 11 and 12: Two different procedures were made to synthesize 11 and 12. First procedure: The whole process was shielded from light by aluminum foil. The

203 mixture of 10a and 10b (20 mg, 0.01 mmol) was added to a Schlenk flask and dried under vacuum. The vacuum was released under argon to allow the addition of dry THF (6 mL). The solution was then degassed via three freeze-pump-thaw cycles before the vessel was purged with argon again. LiAlH4 (0.2 mmol) was then added into the Schlenk flask. The whole mixture was then refluxed for 30 min. The mixture was cooled to room temperature and HCl (6M, 10 mL, bubbled argon for 10 minutes) was added under cooling with ice. The mixture was then refluxed for another 3 hours. After cooling to room temperature, the solvent was withdrawn via a syringe. The reaction mixture was then washed twice with NaHCO3(aq), twice with distilled water and once with methanol under argon (all the solvents used here were bubbled with argon for 10 minutes). Then the residue was dried under vacuum.

The residue containing 11 is soluble in normal organic solvents such as CH2Cl2, chloroform, THF, benzene and has a greenish color. The crude 1H NMR was taken immediately,(Figure 5.8) as this compound is unstable, in solution.

Crude 1H NMR spectrum of the residue (500 MHz in d-benzne).

In benzene, the UV-vis spectrum doesn’t change if exposed to light and the atmosphere for several days. On TLC (toluene : cyclohexane = 1 : 2), there are two inseparable bands running

204 together with a brownish color (11-1) on the leading edge and a greenish color (11-2) on the bottom edge of the band. In benzene the two compounds are very stable, seeing as there is no change in LDI-TOF or UV-vis spectra after they were exposed to the atmosphere and light for one day.

LDI-TOF spectrum of 11-1.

LDI-TOF spectrum of 11-2.

205

LDI-TOF spectrum of 11-1 after exposed to the air and light for one day.

LDI-TOF spectrum of 11-2 after exposed to the air and light for one day.

206

UV-vis Spectra

1.00 11-1 fresh 11-1 1 day 0.75

0.50 Absorption 0.25

0.00 400 500 600 700 800 Wavelength (nm)

UV-vis Spectra and stability study of 11-1. UV-vis Spectra

1.00 11-2 11-2 1 day 0.75

0.50 Absorption 0.25

0.00 400 500 600 700 800 Wavelength (nm)

UV-vis Spectra and stability study of 11-2.

While in CH2Cl2 the color changed from greenish to brownish very quickly, approximately five minutes, when exposed to light and air. The TLC of this sample has only the brownish band left, which gives the same Rf value as 11-1. We infer that this brown layer is a butterfly porphyrin dimer 12. Characterization of 12: Yield: 25%. UV-Vis λmax (CH2Cl2)/nm 451 (log ε 5.10), 567

(4.24), 610 (4.17); Calculated Mass, 3803.41, Found MS (LDI-TOF), m/z 1901.29 (half of the

M.W.) and 3806.79.

207

LDI-TOF spectrum of 12.

100 100 90 95 80 90 70 85 T% T% 60 2a 80 50 12 75 40 10 70 30 65 20 60 4000 3500 3000 2500 2000 1500 1000 wavelength nm

IR spectra of 2a, 10 and 12.

208

UV-vis spectra

1.00 2a

0.75 12 2b 0.50 2c

10a Absorption 0.25

0.00 350 400 450 500 550 600 650 700 750

Wavelength (nm) UV-vis spectra of 2a-c, 10a and 12. Second procedure: This procedure was identical to the first except that 12M HCl was used instead of 6M HCl. Under this condition, the porphyrin can dissolve in the mixture of THF/HCl. The solution is a green color, due to the dissolved porphyrin. As opposed to the first procedure, which yields a clean solution with solid on the sides of the flask. This means the porphyrin has been in the THF solution for 4 hours by the time the reaction is complete. This yields side products despite being shielded from light and protected under argon. The mixture LDI showed the following peaks: M-2Ni (1790), M-2Ni+[O](1807), M-Ni+4*[O](1909), M-Ni+6*[O](1938), M+7*[O](2011), M+7*[O]+THF(2083), M+7*[O]+2*THF(2156). The crude 1H NMR is not very informative.

209

LDI-TOF spectrum of the residue

Crude 1H NMR spectrum of the residue (500 MHz in d-benzne)

Then the residue was separated using a preparative TLC plate (CHCl3 and cyclohexane in a 5:8 ratio served as the solvent system). The TLC was very messy, and four major layers were separated from the plate which were labeled as 11-1’,

210

11-2’, 11-3’ and 11-4’ (in order of decreasing Rf). For each layer, LDI-TOF and UV-vis Spectrum were taken and compared. By comparing TLC, LDI-TOF and UV-vis spectra, we infer that 11-1’ is the same as 12; and 11-2’ is primarily the starting material 2a. 11-3’ and 11-4’ are mixtures of several photoinduced oxidation products.

LDI-TOF spectrum of 11-1’. UV-vis Spectra

1.00 11-1'

0.75 12

0.50 Absorption 0.25

0.00 400 500 600 700 Wavelength (nm)

UV-vis spectra of 11-1’ and 12.

211

LDI-TOF spectrum of 11-2’.

UV-vis Spectra

1.00 11-2¡¯

0.75 2a

0.50 Absorption 0.25

0.00 400 500 600 700 800 Wavelength (nm)

UV-vis spectra of 11-2’ and 2a.

212

LDI-TOF spectrum of 11-3’.

LDI-TOF spectrum of 11-4’.

213

Synthesis of 13 via the aldol reaction. Synthesis of 13: Diformylarylporphyrin 3a (0.03 mmol), 1,4-cyclohexandione (0.6 mmol), and KOH aqueous (0.009 mmol) were added to a Schlenk flask and dried under vacuum. The vacuum was released under argon to allow the addition of dry THF (4 mL). The mixture was then degassed via four freeze-pump-thaw cycles and purged with argon again. The Schlenk flask was sealed and stirred at room temperature for 24h. After 24 h, the mixture was removed and washed once with DCM and water. The residue was subjected to silica column chromatography quickly. Three layers were collected as 13a-c.

1 Characterization of 13a: Yield: 20%. H-NMR (500 MHz, CDCl3, Me4Si) δ 8.66-8.71 (6 H, m, β-H), 8.40 (2H, s, aromatic proton on directly fused benzene ring on porphyrin), 7.81-7.88 (8H, two sets of d, J = 8.0 Hz, o-Ph-H on meso-phenyl ring), 7.65 (4H, d, J = 8.0 Hz, m-Ph-H), 7.49-7.51 (6H, m, m-Ph-H on meso-phenyl ring and aromatic protons on the second fused benzene ring), 3.14-3.34 (4H, m, isopropyl(CH)-H on meso-phenyl ring), 3.12 (4H, s, CH2-H), 1.63 (12H, d, J = 7.0 Hz,

214

13 isopropyl (CH3)-H), 1.46 (12H, d, J = 7.0 Hz, isopropyl (CH3)-H) ; C-NMR (500

MHz, CDCl3, Me4Si) δ 24.19, 24.56, 29.71, 34.07, 34.43, 37.76, 114.70, 121.12, 125.14, 125.75, 126.61, 130.25, 130.41, 130.70, 131.36, 131.68, 132.62, 132.92, 133.51, 137.54, 137.72, 138.12, 140.64, 140.72, 141.91, 144.23, 148.48, 149.99, 195.88; Calculated Mass, 1020.39, Found MS (LDI-TOF), m/z 1020.45.

1 Characterization of 13b: Yield: 8%. H-NMR (500 MHz, CDCl3, Me4Si) δ 8.62-8.73 (6 H, m, β-H), 8.40 (2H, s, aromatic proton on directly fused benzene ring on porphyrin), 7.82-7.88 (8H, two sets of d, J = 8.0 Hz, o-Ph-H on meso-phenyl ring), 7.49-7.67 (6H, m, m-Ph-H on meso-phenyl ring and aromatic protons on the second fused benzene ring), 7.02 (2H, s, two protons on fused quinine ring), 3.15-3.35 (4H, m, isopropyl(CH)-H on meso-phenyl ring), 3.12 (4H, s, CH2-H), 1.64 (12H, d, J = 7.0 Hz,

13 isopropyl (CH3)-H), 1.46 (12H, d, J = 7.0 Hz, isopropyl (CH3)-H) ; C-NMR (500

MHz, CDCl3, Me4Si) δ 24.20, 24.58, 27.33, 29.71, 34.07, 34.44, 114.86, 121.08, 125.15, 126.12, 126.60, 127.75, 130.68, 130.81, 131.24, 131.44, 132.65, 133.52, 137.52, 137.67, 138.11, 140.29, 140.77, 141.99, 144.21, 148.50, 150.02, 184.44; Calculated Mass, 1018.38, Found MS (LDI-TOF), m/z 1018.76.

215

1H NMR spectrum of 13a

13C NMR spectrum of 13a

LDI-TOF MSof 13a

216

1H NMR of 13b

13C NMR of 13b

217

LDI-TOF MS of 13b

Synthesis of 14.

Synthesis of 19: Nitro-dibromoarylporphyrin 18 (0.045 mmol), palladium acetate

218

(0.012 mmol), triphenylphosphine (0.030 mmol) and K2CO3 (0.09 mmol) were added to a Schlenk flask and dried under vacuum. The vacuum was released under argon to allow the addition of dry DMF (10 mL) and dry xylene (10 mL). The mixture was then degassed via four freeze-pump-thaw cycles before the vessel was purged with argon again. 25-fold excess of methyl acrolate (1.125 mmol) was then added into the flask under argon. The Schlenk flask was sealed and heated to reflux for 72h. After 72 h, the mixture was diluted with CHCl3 and washed with water. The organic layer was removed under vacuum. The residue was subjected to silica column chromatography

(CHCl3 and cyclohexane in 1:1 ratio as the solvent system). The band containing the desired porphyrin was collected and recrystallized from CHCl3 and methanol.

1 Characterization of 19: Yield: 50%. H-NMR (500 MHz, CDCl3, Me4Si) δ 8.86 (1H, s, β-H), 8.51-8.64 (4H, m, β-H), 7.76-7.86 (8H, m, o-Ph-H on meso-phenyl ring), 7.44-7.58 (8H, m, m-Ph-H on meso-phenyl ring), 7.33-7.34 (2H, m, aromatic protons on fused benzene-H), 3.81-3.82 (6H, s, COOMe-H) 3.11-3.22 (4H, m,

13 isopropyl(CH)-H), 1.45-1.50 (24H, m, isopropyl (CH3)-H) ; C-NMR (500 MHz,

CDCl3, Me4Si) δ 24.11, 24.15, 24.33, 33.98, 34.12, 34.27, 52.54, 116.39, 117.02, 119.87, 123.65, 125.09, 125.22, 125.36, 125.54, 126.33, 126.38, 127.78, 128.91, 129.00, 131.71, 132.29, 132.58, 132.71, 132.97, 133.43, 133.78, 134.66, 136.04, 136.25, 136.29, 136.79, 136.87, 138.49, 139.12, 140.81, 141.32, 141.51, 143.80, 146.38, 146.58, 149.27, 149.30, 149.91, 150.00, 168.10, 168.13.

219

1H NMR of 19

13C NMR of 19

Synthesis of 20: 19 (0.3 mmol) was dissolved in dry CH2Cl2. The temperature of the solution was lowered to -78oC. DIBAL-H (1M solution in hexane) (1.8 mmol) was

220 added to the solution dropwise with an argon inlet. The resulting mixture was stirred at -78oC for 1 hour. Ice water was added to the mixture slowly and the organic layer was extracted by CH2Cl2. The organic layer was separated and dried. The residue was purified by silica column chromatography (CH2Cl2 as the solvent system).

1 Characterization of 20: Yield: 69%. H-NMR (500 MHz, CDCl3, Me4Si) δ 8.51-8.85 (5H, m, β-H), 7.79-7.89 (8H, m, o-Ph-H on meso-phenyl ring), 7.44-7.78 (8H, m, m-Ph-H on meso-phenyl ring), 6.85-6.86 (2H, m, aromatic protons on fused benzene-H), 4.61-4.64 (4H, m, OH-H), 3.12-3.23 (4H, m, isopropyl(CH)-H),

13 1.45-1.52 (24H, m, isopropyl (CH3)-H) ; C-NMR (500 MHz, CDCl3, Me4Si) δ 24.23, 24.50, 34.47, 115.22, 116.22, 120.03, 123.44, 125.44, 126.44, 133.66, 135.66, 137.26, 138.06, 139.47, 140.67, 141.87, 144.08, 146.48, 148.28, 150.09; Calculated Mass,

994.84, Found MS (LDI-TOF), m/z 949.40(M-NO2).

1H NMR of 20

221

13C NMR of 20

LDI-TOF MS of 20

Synthesis of 14: 20 (0.2 mmol) was dissolved in dry CH2Cl2, and then Dess-Martin

222 periodinane (0.4 mmol) was added to the solution. The resulting mixture was stirred at room temperature for half an hour. Saturated sodium thiosulfate aqueous solution was then added to the mixture. The organic layer was separated and dried. The residue was subjected to silica column chromatography (CHCl3 and cyclohexane in a 2:3 ratio served as the solvent system). Characterization of 14: Yield: 76%. 1H-NMR (500

MHz, CDCl3, Me4Si) δ 10.30-10.33 (2H, m, CHO-H), 8.57-8.91 (5H, m, β-H), 7.80-7.89 (8H, m, o-Ph-H on meso-phenyl ring), 7.24-7.63 (10H, m, m-pH-H on meso-phenyl ring and two protons on fused benzene ring), 3.13-3.28 (4H, m,

13 isopropyl(CH)-H), 1.46-1.56 (24H, m, isopropyl (CH3)-H) ; C-NMR (500 MHz,

CDCl3, Me4Si) δ 24.11, 24.16, 24.20, 24.37, 24.41, 26.93, 33.99, 34.08, 34.12, 34.40, 109.52, 116.75, 117.11, 117.66, 120.06, 120.50, 123.73, 125.20, 125.41, 125.60, 126.31, 126.51, 126.57, 132.56, 132.71, 132.96, 133.44, 133.64, 133.78, 136.05, 136.15, 136.35, 136.37, 136.68, 136.73, 137.40, 137.44, 138.14, 142.04, 142.12, 142.56, 142.63, 143.62, 144.18, 146.06, 146.24, 148.61, 149.43, 150.14, 150.29, 150.51, 150.54, 192.38, 192.75; Calculated Mass, 990.81, Found MS (LDI-TOF), m/z

944.57 (M-NO2).

1H NMR of 14

223

13C NMR of 14

LDI-TOF spectrum of 14

224

Synthesis of 15a: A mixture of 13a and 13c (0.03 mmol), 1,4-cyclohexandione (0.03 mmol), KOH aqueous (0.009 mmol), and THF (4 mL) were added to a round bottom flask and bubbled with Argon. The flask was then sealed and stirred at room temperature for 24h. After 24 h, the mixture was removed and washed once with DCM and water. The residue was subjected to silica column chromatography quickly. Two layers were collected as 15a (yield: 40%) and 2a (yield: 12%). Characterization

1 of 15a: H-NMR (500 MHz, CDCl3, Me4Si) δ 8.48-8.79 (15H, m, β-H and aromatic protons on directly fused benzene ring on porphyrins), 7.43-7.87 (36H, m, all the left aromatic protons), 3.10-3.38 (8H, m, isopropyl(CH)-H), 1.41-1.69 (48H, m, isopropyl

(CH3)-H) ; Calculated Mass, 1976.69, Found MS (LDI-TOF), m/z 1976.23.

225

1H NMR of 15a

LDI-TOF MS of 15a

226

Synthesis of 15b: A mixture of 13a and 13c (0.03 mmol), 1,4-cyclohexandione (0.03 mmol), KOH aqueous (0.009 mmol), and THF (4 mL) were added to a round bottom flask and bubbled with Argon. The flask was then sealed and stirred at room temperature for 24h. After 24 h, the mixture was removed and washed once with DCM and water. The residue was subjected to silica column chromatography quickly and 15b was collected. The solubility of 15b was not very good in DCM or chloroform mainly due to the aggregation. Characterization of 15b: Calculated Mass, 1938.37, Found MS (LDI-TOF), m/z 1938.09.

227

LDI-TOF MS of 15b

Synthesis of 16.

228

Synthesis of 22: 21 (0.3 mmol) was dissolved in dry CH2Cl2. The temperature of the solution was lowered to -78oC. DIBAL-H (1M solution in hexane) (3.6 mmol) was added to the solution dropwise with an argon inlet. The resulting mixture was stirred at -78oC for 1 hour. Ice water was added to the mixture slowly and the organic layer was extracted by CH2Cl2. The organic layer was separated and dried. The residue was purified by silica column chromatography (CH2Cl2 as the solvent system).

1 Characterization of 22: Yield: 60%. H-NMR (500 MHz, CDCl3, Me4Si) δ 8.75 (4H, s, β-H), 7.80 (8H, d, J = 7.5 Hz, o-Ph-H on meso-phenyl ring), 7.56 (8H, d, J = 8.0 Hz, m-Ph-H on meso-phenyl ring), 6.88 (4H, s, aromatic protons on fused benzene-H),

4.63 (8H, s, CH2-H), 3.20-3.23 (4H, m, isopropyl(CH)-H), 1.50 (24H, m, isopropyl

13 (CH3)-H) ; C-NMR (500 MHz, CDCl3, Me4Si) δ 24.51, 34.27, 34.33, 65.38, 112.38, 116.51, 125.03, 125.27, 126.04, 131.12, 131.21, 132.69, 136.17, 136.72, 137.97, 140.02, 141.91, 149.51; Calculated Mass, 1058.43, Found MS (LDI-TOF), m/z 1058.59.

1H NMR of 22

229

13C NMR of 22

LDI-TOF MS of 22

Synthesis of 16: 22 (0.2 mmol) was dissolved in dry CH2Cl2, and then Dess-Martin periodinane (0.8 mmol) was added to the solution. The resulting mixture was stirred at room temperature for half an hour. Saturated sodium thiosulfate aqueous solution was then added to the mixture. The organic layer was separated and dried. The residue was

230 subjected to silica column chromatography (CHCl3 and cyclohexane in a 2:3 ratio served as the solvent system). Characterization of 16: Yield: 58%. 1H-NMR (500

MHz, CDCl3, Me4Si) δ 10.31 (4H, s, CHO-H), 8.82 (4H, s, β-H), 7.81 (8H, d, J = 8.0 Hz, o-Ph-H on meso-phenyl ring), 7.63 (8H, s, J = 8.0 Hz, m-pH-H on meso-phenyl ring), 7.50 (4H, s, aromatic protons on fused benzene rings), 3.23-3.29

13 (4H, m, isopropyl(CH)-H), 1.54 (24H, d, J = 7.0 Hz, isopropyl (CH3)-H) ; C-NMR

(500 MHz, CDCl3, Me4Si) δ 24.37, 34.39, 41.99, 108.57, 118.19, 126.48, 128.56, 132.36, 132.56, 136.49, 136.64, 141.91, 142.99, 150.44, 192.52, 192.57; Calculated Mass, 1051.89, Found MS (LDI-TOF), m/z 1051.24.

1H NMR of 16

231

13C NMR of 16

LDI-TOF MS of 16

232

Synthesis of 17: 13a (0.04 mmol), 16 (0.02 mmol), KOH aqueous (0.01 mmol), and THF (4 mL) were added to a round bottom flask and bubbled with Argon. The flask was then sealed and stirred at room temperature for 24h. After 24 h, the mixture was removed and washed once with DCM and water. The residue was subjected to silica column chromatography quickly. 17 was then collected. Characterization of 17: yield:

1 30%. H-NMR (500 MHz, CDCl3, Me4Si) δ 8.62-8.71 (16H, m, β-H), 7.24-7.88 (64H, m, aromatic protons), 3.05-3.47 (12H, m, isopropyl(CH)-H), 1.22-1.71 (72H, m, isopropyl (CH3)-H) ; Calculated Mass, 3019.11, Found MS (LDI-TOF), m/z 3019.77.

233

1H NMR of 17

LDI-TOF MS of 17

234