Mechanistic Studies of Cation Photogeneration

Loyola University Chicago Loyola eCommons

Dissertations Theses and Dissertations

1998

Mary Frances Clifton Loyola University Chicago

Follow this and additional works at: https://ecommons.luc.edu/luc_diss

Part of the Chemistry Commons

Recommended Citation Clifton, Mary Frances, "Mechanistic Studies of Cation Photogeneration" (1998). Dissertations. 3732. https://ecommons.luc.edu/luc_diss/3732

This Dissertation is brought to you for free and open access by the Theses and Dissertations at Loyola eCommons. It has been accepted for inclusion in Dissertations by an authorized administrator of Loyola eCommons. For more information, please contact [email protected].

MECHANISTIC STUDIES OF CATION PHOTOGENERATION

A DISSERTATION SUBMITTED TO

THE FACULTY OF THE GRADUATE SCHOOL

IN CANDIDACY FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY

DEPARTMENT OF CHEMISTRY

MARY FRANCES CLIFTON

CHICAGO, ILLINOIS

Special thanks are extended to Dr. Mary K. Boyd, Dr. David S. Crumrine, Dr. Kurt

W. Field, and Dr. Gerry K. Noren. I am also indebted to the Department of Education for

a Graduate Assistantship in Areas of National Need (GAANN) and the Arthur J. Schmitt

Foundation for a dissertation fellowship.

111 TABLE OF CONTENTS

ACKNOWLEDGMENTS ...... iii

LIST OF TABLES ...... vii

LIST OF FIGURES ...... Vlll

LIST OF ABBREVIATIONS ...... xi

ABSTRACT ...... xiii

Chapter

I. INTRODUCTION ...... I A. Carbocations: Definitio°' Applications and Photogeneration . . I 1. Definition and Applications ...... I 2. Photoinitiated Polymerization ...... 2 a. Photoinitiated Radical Polymerization ...... 3 b. Photoinitiated Charge-Transfer Polymerization . 6 c. Photoinitiated Cation-Radical Polymerization . I 0 d. Photoinitiated Simultaneous Radical and Cation-Radical Polymerization ...... 14 3 . Photoinitiators ...... 15 a. Free Radical Initiators ...... 15 b. Cationic Photoinitiators ...... 21 4. Methods for Photogeneration of Carbocations ...... 22 a. Photoheterolysis of Carbon-Heteroatom Bonds ...... 25 b. Carbon-Carbon Bond Cleavage ...... 27 c. Heterolysis ofPhotogenerated Radical Cations ...... 28 d. Oxidation of Photochemically Generated Radicals ...... 30 e. Protonation of Photochemically Generated Carbenes ...... 31 f Photoprotonation of Styrenes, Aromatics and Phenylacetylenes ...... 3 2 B. Photochemistry of Carbonyl Compounds ...... 33 1. Norrish Type I Reactions ...... 38 2. Norrish Type Il Reactions ...... 40 3. Intermolecular Hydrogen Abstraction ...... 43

IV 4. Paterno-Biichi Reaction ...... 45 C. Photochemistry of the Carbon-Carbon Double Bond ...... 45 D. Earlier Work On The Thermal and Photogeneration of Xanthyl Cations ...... 49

II. RES ULTS ...... 56 A. 9-Xanthylacetone and 9-Xanthylacetophenone ...... 56 1. Syntheses ...... 56 2. Preparative Photolyses ...... 57 3. Acid Catalysis ...... 62 4. Triplet Sensitization and Quenching Studies ...... 63 5. Laser Flash Photolysis Studies ...... 67 6. Photoproduct Studies ...... 77 a. 9-Xanthylidene Acetophenone ...... 77 i. Syntheses ...... 77 ii. Quantum Yields ...... 78 iii. Synthesis of Proposed Tertiary Benzylic Alcohol ...... 79 a. Method 1 ...... 79 b. Method 2 ...... 81 7. Fluorescence ...... 83 B. 2-Methyl-3-(9-xanthyl)propene ...... 83 1. Syntheses ...... 83 2. Preparative Photolyses ...... 84 3. Acid Catalysis ...... 85 4. Triplet Sensitization ...... 93 5. Triplet Quenching ...... 94 6. Fluorescence ...... 95

III. DISCUSSION ...... 96 A. 9-Xanthylacetone and 9-Xanthylacetophenone ...... 96 1. Acid Catalysis ...... 96 2. M~chanism of the Photogeneration of the Xanthyl Cation from 9-Xanthylacetone and 9-Xanthylacetophenone ...... 96 3. Triplet Sensitization and Quenching Results ...... 98 4. Product Quantum Yields ...... 100 5. Photoproduct Studies ...... 103 a. Mechanism ofFormation of9-Xanthylidene Acetophenone ...... 103 B. 2-Methyl-3-(9-xanthyl)propene ...... 111

IV. CONCLUSIONS ...... 118

v V. EXPERIMENTAL ...... 121 A. Materials ...... 121 B. General Methods ...... 121 C. General Photolysis Procedure ...... 122 D. Triplet Sensitized Photolysis: General Method ...... 123 E. Triplet Quenching ...... 124 F. Laser Flash Photolysis ...... 124 G. Syntheses ...... 125 1. Ethyl 2-(9-Xanthyl)-3-oxobutanoate 3 ...... 125 2. 9-Xanthylacetone 1 ...... 125 3. Ethyl 2-(9-xanthyl)-3-phenyloxopropanoate 4 ..... 126 4. 9-Xanthylacetophenone 2 ...... 126 5. 2-Methyl-3-(9-xanthyl)propene 14 ...... 126 6. 9-Phenylethynylxanthenol 6 ...... 127 7. 9-Xanthylidene acetophenone 5 ...... 128 8. Attempted Synthesis of Tertiary Benzylic Alcohol to: 9-Hydroxy-(9-xanthyl)acetophenone ...... 128 a. Method 1 ...... 128 b. Method 2 ...... 129 9. Product Quantum Yield ...... 130

VI. SPECTRA ...... 133

APPENDIX ...... 152

REFERENCES ...... 180

VITA ...... 188

VI LIST OF TABLES

1. GC Area Percent Analyses of Products Obtained from the Attempted Hydration of 9-Xanthylidene Acetophenone 5 ...... 80

2. GC Area Percent Analyses Obtained from Products of the Irradiation of 9-Xanthylacetophenone 2 ...... 82

3. Product Quantum Yields ...... 102

4. Disproportionation/Combination Ratios kA ...... 106

vii LIST OF FIGURES

Figure Page

1. Examples of cation-radical photoinitiators: aryldiazonium salts, quaternary ammonium salts and onium salts ...... 11

2. Examples ofType I free radical photoinitiators: a,a'-dialkoxyacetophenones, a-alkoxydeoxybenzoins, dibenzoyl disulfides, azocompounds such as phenylazo-4-diphenylsulfone and dibenzoyl methanes ...... 16

3. Examples of Type II free radical photoinitiators: benzils, benzophenones, fluorenones, xanthones and camphorquinone ...... 19

4. Examples of cationic photoinitiators: triarylsulfonium salts, diaryliodonium salts, Bronsted acids and Lewis acids ...... 23

5. Effect of solvent polarity on wavelength shifts ...... 36

6. Xanthyl alcohols and xanthyl ketone used to thermally generate the xanthyl cation: 2-(9-xanthyl)ethanol a, 1-(9-xanthyl)-2-methyl-2-propanol b, 1-(9-xanthyl)-2-propanol c, 1-(9-xanthyl)-2-phenyl-2-propanol d and 9-xanthylacetone 1. The trityl alcohol, 4,4,4-triphenyl-2-methyl- 2-butanol e, demonstrated no cation formation...... 50

7. Absorption spectra measured following irradiation of9-xanthylacetone 1 in

30% H2S04 (with 33% acetonitrile cosolvent) with an excitation wavelength of300 nm. Spectra were measured at 1 min irradiation time intervals and show growth of the xanthyl cation peaks ...... 58

8. Absorption spectra measured following irradiation of xanthyl cation solutions

produced from photolysis of9-xanthylacetone 1in30% H2S04 (with 33% acetonitrile cosolvent) with an excitation wavelength of300 nm. Spectra were measured at 4 min irradiation time intervals and show absorption peaks corresponding to the xanthyl cation decreasing with the concomitant formation of xanthone...... 60

Vlll LIST OF FIGURES (Continued) Figure Page

9. Plot of absorption intensity ofxanthyl cation (at 374 nm) versus percent

aqueous H2S04 (acid concentration not corrected for cosolvent). The xanthyl cation was produced from irradiation of 9-xanthylacetone 1 with irradiation times of 1 min ( 1. ), 2 min ( • ), and 4 min ( •). These times correspond to cation conversions of 10%, 32%, and 60% respectively...... 64

10. Transient spectra obtained following photolysis of9-xanthylacetophenone 2 in

90:8:2 CH3CN:H20:H2S04. Solutions were purged with nitrogen and show decay of the initially formed xanthyl radical...... 68

11. Transient spectra obtained following photolysis of 9-xanthylacetophenone 2 in

90:8:2 CH3CN:H20:H2S04. Substrate solutions were purged with air and show clean conversion of the xanthyl radical (A.max=345 nm) to the xanthyl cation (A.max=374 nm) ...... 71

12. Transient absorption waveforms of9-xanthylacetophenone 2 (solid) and 9-

xanthylacetone 1 (dotted) at 370 nm in 90:8:2 CH3CN:H20:H2S04 purged with air...... 73

13. Transient absorption waveforms for 9-xanthylacetophenone 2 (370 nm) in

90:8:2 CH3CN:H20:H2S04 purged with nitrogen and purged with air. . 75

14. Absorption spectra measured following irradiation of2-methyl-3-(9-xanthyl)

propene 14 in 30% H2S04 (with 33% acetonitrile cosolvent) with an excitation wavelength of 254 nm. Spectra were measured at 2 min irradiation time intervals and show growth of the xanthyl cation peak. . 86

15. Plot of absorption intensity ofxanthyl cation (at 374 nm) versus 1-50% aqueous

H2S04 (acid concentration not corrected for cosolvent). The xanthyl cation was produced from irradiation of2-methyl-3-(9-xanthyl)propene 14 after 0 min(•), 4 min ( • ), and 4 min minus 0 min ( 1. )...... 89

16. Plot of absorption intensity ofxanthyl cation (at 374 nm) versus 0-20% aqueous

17. GC/MS (EI) ofXMP photoproduct tR 42 ...... 155

lX LIST OF FIGURES (Continued)

Figure Page

18. Possible structures considered for XMP photoproduct tR 42. 157

19. Possible structures considered for XMP photoproduct tR 42. 159

20. FT/IR ofXMP photoproduct tR 42 ...... 161

21. 13C NMR (75 MHz) ofXMP photoproduct tR 42 ...... 163

22. 13C NMR (100 MHz) ofXMP photoproduct tR 42...... 165

23. 1H NMR (400 MHz) ofXMP photoproduct tR 42 ...... 168

24. 1HNMR (300 MHz) ofXMP photoproduct tR 42 ...... 170

25. 1H NMR (300 MHz) RESOLV of XMP photoproduct tR 42...... 172

26. GC/MS High Resolution EI ofXMP photoproduct tR 42 ...... 174

27. GC/MS High Resolution EI ofXMP photoproduct tR 42 ...... 176

28. GC/MS High Resolution EI Peak matching results...... 178

x LIST OF ABBREVIATIONS

APT attached proton test

Bu butyl

c Celsius

Et ethyl

kcal/mol kilocalories per mole

disproportionation/combination ratio

wavelength

MALDI-TOF matrix assisted laser desorption ionization-time of flight

Me methyl

µs microsecond mm minutes

M Molarity

MS mass spectroscopy nm nanometer ns nanosecond p para

Pr propyl

X1 LIST OF ABBREVIATIONS (Continued) s seconds t tertiary

TFA trifluoroacetic acid tR retention time

UV ultraviolet v volts

VlS visible

Xll ABSTRACT

9-Xanthylacetone and 9-xanthylacetophenone have been sythesized by a two step procedure. Initially, P-ketoesters were prepared via acid catalyzed reaction of xanthydrol with either ethyl acetoacetate or ethyl benzoylacetate. Base hydrolysis yielded the corresponding ketones. 2-Methyl-3-(9-xanthyl)propene has been synthesized via a three step procedure. Initially, 9-xanthylacetone was formed via the two step procedure described above. Wittig reaction of the ketone yielded 2-methyl-3-(9-xanthyl)propene.

Irradiation of 9-xanthylacetone, 9-xanthylacetophenone, and 2-methyl-3-(9- xanthyl)propene in dilute acidic aqueous solution results in facile formation of the xanthyl cation. This cation photogeneration is not subject to acid catalysis over the acid range of 5-

30% H2S04 in the case of the xanthyl ketones or 1-20% H2 S04 in the case of the xanthyl propene. In contrast, the thermal reaction requires considerably stronger acidic media, and its rate shows a strong dependence on acid concentration.

Laser flash photolysis studies of the ketones demonstrate that the excited-state mechanism proceeds through an initial homolytic carbon-carbon P-bond cleavage to generate the xanthyl radical. Although carbonyl compounds are known to undergo photochemically induced a-cleavage or Norrish Type I reactions, P-cleavage is rare and only typically occurs with compounds containing relatively weak Ca-Cp bonds. The xanthyl cation is produced from the xanthyl radical in a subsequent step, with oxygen acting as the oxidizing agent. The

Xlll CHAPTER I

INTRODUCTION

A. Carbocations: Definition, Applications and Photogeneration

1. Definition and Applications

Carbocations are organic compounds possessing a carbon atom with a positive charge.

The terms carbonium and carbenium ions have also been used to refer to what is now called

carbocation. McManus1 and others initially defined a carbonium ion as an organic cation

containing a positively charged carbon with a sextet of electrons surrounding the charged carbon. The parent member is the methyl cation, with three hydrogens attached to the carbon. In conflict with this definition, Sefcik2 and coworkers found MS evidence for the existence of CHs +, the methanonium ion. Olah3 and coworkers also found evidence for a carbon atom with the formal covalency of five rather than three. Thus, Olah referred to pentacoordinated carbocations, the parent being CHs +, as "nonclassical" carbonium ions. The term "carbonium" originates from .onium ions, such as ammonium and sulfonium. The description of the carbonium ion though, does not fit the onium ion electron octet definition.

Onium ions increase their covalency by one on formation from their precursors, whereas carbonium ions decrease their covalency by one, in going from four in their precursor to three.

4 2 Kennedy and Marechal defined the term "carbenium ion" as a trivalent trigonal sp -

I 2 hybridized carbocation ("classical ions"), the parent being CH3 +. Carbenium ions were considered as derivatives of carbenes with the addition of a proton or R + to give the trivalent

ion R3C+. "Carbenium ion", a more systematic term for carbocations, is generated from oxenium and sulfenium ion nomenclature. The latter pair of ions also contains an electron sextet and thus are isoelectronic with carbonium ions. Carbenium ions and carbonium ions are considered to be carbocations. Today the term carbocations, analogous to carbanions, is widely accepted. 5

A carbocation's positive charge denotes the absence of a pair of electrons and usually

5 6 indicates a high degree of reactivity at that carbon atom. • This high degree of reactivity makes carbocations important intermediates in many synthetic and commercial applications.

These applications include use in the dye, pharmaceutical, and polymer industry. In the polymer industry, carbocations are used as curing ( crosslinking) agents in adhesives and

7 8 coatings for paper, wood, and automobile parts. ' Carbocations are also used as initiators

9 10 in polymerization reactions. •

2. Photoinitiated Polymerization

A photoinitiator is defined as a chemical compound or chemical system which absorbs light, initiates polymerization, and then is physically changed or destroyed in the process. 11

Photoinitiation of polymerization is the result of direct photoproduction of an active intermediate (radical, cation) which can interact with a monomer molecule producing initiation, propagation and ultimately chain termination. In some cases, the ability to generate both radical and cation, depending on the reaction conditions, may have applications in 3

concurrent radical/cation photoinitiator systems. 12 Photoinitiated polymers result in several

industrially important applications including printed circuits, surface coatings, curable printing

13 14 inks, adhesives, relief printing plates, and resin reinforced structures. •

There are several requirements for the successful design of a photoinitiator. 11 These

include a chemically active excited state (rather than undergoing a physical decay process),

efficiency in initiating polymerization, solubility in the monomer system, longterm thermal

stability, little or no byproducts, low toxicity, and the properties of being odorless, colorless and inexpensive. 11 Photoinitiated polymerization can be divided into four major classes: photoinitiated radical polymerization, charge transfer polymerization, cation-radical polymerization, and simultaneous radical-cation polymerization. Each class will be described in detail, including mechanisms and specific examples.

a.) Photoinitiated Radical Polymerization

Photoinitiated radical polymerization can proceed via three major mechanisms, each resulting in the generation of free radicals. The first mechanism involves the initial irradiation

of photoinitiator I to form an excited singlet (S1) or triplet (T1 ) state. The initiator I* undergoes intramolecular homolytic cleavage into two radicals, which can then initiate free radical polymerization as shown below (Scheme 1 ). A second mechanism involves the

Scheme 1

I+ hv I* 4 excited state initiator I* (Scheme 2). The excited singlet or triplet initiator can abstract a hydrogen atom intermolecularly from donor molecules (RH). The resulting free radical R·

Scheme 2

I*+ RH - ·IH + R •

can then initiate polymerization. A third mechanism involves the formation of a charge- transfer complex. This charge-transfer complex dissociates into free radicals, which go on to initiate free radical polymerization (Scheme 3).

Scheme 3

I*+ AH - [l... .. AH)* - • IH +A • CT COMPLEX

Photoinitiation of free radical polymerization in an ideal case occurs via two steps: initial formation offree radicals from a photoinitiator as descnbed above, followed by addition of the free radical to a monomer_ molecule to begin chain initiation. However, most photoinitiated radical polymerizations are more complicated due to several factors.

Monomers present could be capable of quenching the excited state of the photoinitiator through reducing the number of free radicals or by forming nonradical products. Radical recombination also can occur, ifthe radicals recombine more readily with each other than with the monomer molecules involved in chain initiation.

A specific type of free radical polymerization is chain addition polymerization of vinyl 5 monomers by a free radical mechanism. The steps involve initiatio~ propagation and termination (via recombination or disproportionation). A generalized mechanism is shown in Scheme 4 below.

Scheme 4

INITIATION:

PROPAGATION:

TERMINATION:

via recombination reaction:

via disproportionation reaction: -

Several vinyl monomers, including styrene, methyl methacrylate and acrylonitrile will photopolymerize through free radical polymerization when exposed to UV irradiation.

Additional photosensitive monomers which undergo UV-induced polymerization include: 2- ethylhexyl acrylate, 2-hydroxyethyl methacrylate, vinylcinnamate and pentaerythritol tetramethacrylate. 11 The initiator radical is formed from direct photolysis of the.monomer, 6

from impurities present in the monomer, or from charge-transfer complexes formed between

the monomers and oxygen. Most vinyl monomers though require the aid of a photoinitiator

to generate free radicals for initiation of polymerization. A specific example of photoinitiated

free radical vinyl polymerization (Scheme 5) is the irradiation of styrene to yield dimers,

trimers, oligomers and polystyrene. 15

Scheme 5

hv +

b.) Photoinitiated Charge-Transfer Polymerization

A second major class of photoinitiated polymerization requires the interaction of two chemically different molecules in a ground state charge-transfer complex (CT). One molecule acts as the electron acceptor (A), and the second chemically distinct molecule, acts as the donor (D). Initially, the charge-transfer complex undergoes excitation to the excited singlet or triplet state. Electron transfer occurs from the donor molecule to the acceptor molecule and results in the formation of radical cations and radical anions as shown in Scheme 6. A 7

Scheme 6

D +A = (D-A)(CT COMPLEX)

(D-A) ~ (D-A)*

or A~A*

A*+ D - (D-A)* + - + - 0 0 (D-A)*- (Ds ---As ) - D8 °+ As"

specific example is the photoinitiated charge-transfer cationic polymerization of N- vinylcarbazole, using chloranil as an electron acceptor. 16 The radical cation, formed from electron donating monomers, can then proceed to initiate either cationic, radical, or simultaneous radical/cationic polymerization (Scheme 7).

Scheme 7

+ • CH=CH 0 I 2 + Cl~CI hv ----~~Yi\TCI Cl~CI ~ Cl~CI 0 CTCOMPLEX O

Several additional monomers are involved in charge-transfer cationic polymerization, including styrene (electron donor D) with pyromellitic dianhydride17 (electron acceptor A),

2-vinylnaphthalene (D) with diethyl fumarate18 (A), and cyclohexene (D) with N- 8 ethylmaleimide19 (A).

In addition to a mechanism involving the formation of a charge-transfer complex, photoinitiated charge-transfer polymerization may also proceed through a mechanism in which an exciplex is formed. An exciplex is an excited charge-transfer complex existing in either the singlet or triplet state. Two chemically distinct molecules associate via an excited donor (D) and ground state acceptor (A), or an excited acceptor (A) and ground state donor

(D), to generate an excited state complex. An excited charge-transfer complex (D+ A-)• or

(A+ D)* is produced in the formation of the exciplex (Scheme 8). Exciplexes are distinguished from charge-transfer complexes through the consideration of the donor ionization energy and the acceptor electron affinity. Ifthis difference is very small, the ground state charge-transfer complex is formed, rather than the exciplex.

Scheme 8

D* +A - {D·A+)*- {D+A-)*- D*A - DA*={DA)*

A*+ D - {KD+)*- {A+D-)*- A*D - AD*={DA)*

The photoinitiated polymerization of methyl methacrylate involves the exciplex

formation between the excited triplet state (T1) of poly(vinyl benzophenone) and indole-3- ylacetic acid. 20 The indole radical initiates polymerization, whereas the polymeric ketal radical acts as a crosslinker as shown in Scheme 9. Photoreducible dyes are another example of 9

Scheme 9

EXCIPLEX

+ 10 exciplex formation resulting in free radical initiation of polymerization. These dyes have found application in photoimaging processes. 21

c.) Photoinitiated Cation-Radical Polymerization

Some common photoinitiators involved m cation-radical polymerization are aryldiazonium salts, 22 quaternary ammonium salts, 23 and onium salts 24 (Figure 1). These photoinitiators are involved in polymerization ofthe following monomers: 1,2-epoxypropane, poly(vinyl alcohol) and glycidyl methacrylate ( aryldiazonium salts); methyl methacrylate, acrylonitrile, and styrene (quaternary ammonium salts); and in the case of onium salts any cationically polymerizable monomer such as olefins, epoxides and lactones.

The mechanisms by which photoinitiated cation-radical polymerization may proceed vary, depending on the nature of the photoinitiator. Photoinitiation by diaryliodonium salts proceeds by two processes. The major process is shown in Scheme 10 below. Upon irradiation the diaryliodonium salt homolytically cleaves to generate the radical-cation, radical

Scheme 10

hv Ari·++ Ar. + x-

Arr++ SH 11

Figure 1

Examples of cation-radical photoinitiators: aryldiazonium salts, quaternary ammonium salts and onium salts. 12

CH3 o-~+-CH2-0 BF4- CH3 13

and anionic fragments. 24 The radical-cation reacts with solvent molecules (SH) and then rapidly deprotonates to form the aryliodo compound and tetrafluoroborate. Laser flash photolysis studies confirmed a homolytic cleavage mechanism. 25

In certain cation-radical photoinitiated polymerizations, the monomer and solvent act as the photoinitiators. A specific example (Scheme 11) involves maleic anhydride in dioxane. 26 Irradiation of maleic anhydride initially results in the formation of excited triplet radical-cations and radical-anions. The radical-cations abstract a hydrogen from the solvent

Scheme 11

INITIATION:

hv 3[ + --o~o +

PROPAGATION:

• + + + (°) + (°} 0¥0 0 0¥0 0

+ . 0¥0 + 0¥0 o.J:.~o o2)~o

TERMINATION:

2 (°} + 0 14 dioxane to generate dioxane radicals and maleic anhydride cations. These cations act to propagate the polymer chain. Most of the dioxane radicals act to terminate the growing chain.

d.) Photoinitiated Simultaneous Radical and Cation-Radical Polymerization

The fourth major class of photoinitiated polymerizations is photoinitiated simultaneous radical and cation-radical polymerization. Upon photolysis of certain photoinitiators, for example, azobisisobutyronitrile or benzoin, an initiating free radical R· is formed which can yield both cations and anions, depending on the solvent. These photoinitiators are successful at initiating polymerization even if two different monomers, one free radically polymerized and one cationically polymerized, are chosen. Both can be polymerized simultaneously to yield an interpenetrating polymer network. Polymerization of glycidyl acrylate, 27 which contains two different functional groups, is accomplished using a triaryl sulfonium salt as the photoinitiator. Both the vinyl and the epoxy groups are polymerized through free radical and cationic mechanisms (Scheme 12).

Scheme 12

hv 15

3. Photoinitiators

An extensive list of potential photoinitiators described in the literature28 include:

halogens and halogenated organics, triarylsulfonium salts, peroxides, polycyclic hydrocarbons,

azocompounds, metal complexes, polymeric photoinitiators, and ketones. Aromatic ketones

such as benzophenone, acetophenone, di- and trichloroacetophenone, dialkoxyacetophenone, fluorenone, and thioxanthone are among the photoinitiators that have commercial

significance. Types of photoinitiators and specific examples of each will be discussed in detail below.

Two major types of photoinitiators are currently of interest. These include radical initiators and cationic initiators. Both photophysical and photochemical properties of photoinitiators are important in controlling their reactivity. A photoinitiator should possess the following properties:29 1) high absorptivity in the region of activation, 2) high quantum yield for free radical formation (in the case of free radical photoinitiators), 3) solubility in the resin system, 4) storage stability, 5) the characteristics of being odorless, non-yellowing, nontoxic, and inexpensive and 6) ease in handling.

a.) Free radical initiators

There are two major categories of free radical photoinitiators. Type I initiators are based on either the benzoyl functionality ( a-alkoxydeoxybenzoins, a,a ' - dialkoxyacetophenones, dibenzoyl methanes, dibenzoyl disulfides) or the azo group

(phenylazo-4-diphenylsulfone), as shown in Figure 2. Type II initiators include benzophenones, camphorquinone, fluorenones, xanthones, thioxanthones and ·benzils, as 16

Figure 2

Examples of Type I free radical photoinitiators: ex, ex' -dialkoxyacetophenones, cx alkoxydeoxybenzoins, dibenzoyl disulfides, azo compounds such as phenylazo-4- diphenylsulfone, and dibenzoyl methanes. 17

R O OAlk 111 C-C-H & I OAlk

R 0 OAlk R ~c-b___lfX' ~ ,~ H

R 'Q-c-s-s-c-0O O R

O-N=N-S02-0-0

R O RI O R c-c-cII II-a & I - R 18

shown in Figure 3. The R groups of Type I and Type II photoinitiators include: alkyl, aryl,

alkylaryl, alkoxy, halogen, alkylsulfonates, acrylates, tertiary amines and others.

The photoinitiators of the Type I class undergo a photofragmentation in the excited

state to yield free radicals. Type I photoinitiators may also be considered "unimolecular'',

referring to the process by which radical production occurs. In the case ofbenzophenone,

the benzoyl and benzyl radicals are formed, which then go on to initiate free radical

polymeriz.ation. Bond scission may take place at either the a or p position, depending on the

nature of the photoinitiator. Benzoin ethers, as do most ketone photoinitiators, more commonly fragment at the a-position (Norrish Type I cleavage). These ethers may be the most extensively studied and utilized photoinitiators. 13

Type Il class or "bimolecular'' photoinitiators, upon excitation and intersystem crossing to the triplet state, abstract a hydrogen atom from the monomer, polymer or solvent to generate a ketyl radical. This ketyl radical can combine with another ketyl radical to form a pinacol. The other radical species then acts as the initiator or crosslinker.

A few Type I unimolecular photoinitiators do exhibit P-cleavage. One example is a chlorinated acetophenones.30 The main fragmentation involves P-cleavage of a C-Cl bond to generate the Cl radical. The chlorine radical can then abstract hydrogen atoms to generate hydrogen chloride. Since a strong acid is generated, this photoinitiator could be useful in acid curing of urea-formaldehyde or melamine based resins.

In some photoinitiators initiation efficiency can be enhanced through the use of tertiary amine synergists such as ethyl-p-dimethylaminobenzoate. 29 These tertiary amines form an excited state complex (exciplex) with the triplet excited state ketone. The ·exciplex is 19

Figure 3

Examples of Type II free radical photoinitiators: benzils, benzophenones, fluorenones, xanthones and camphorquinone. 20

R '()-c-c-00 0 R

R 'Q-c--1(3O R

R~R ~

R O R ~

CH3'k_~~3 ~o 0 21

stabilized via charge transfer of the amine donor to the ketone acceptor. Tertiary amine co- synergists may also function as scavengers of oxygen precursor peroxy radicals, thus preventing the quenching of the excited triplet state of many ketone photoinitiators.

Type II bimolecular photoinitiators, such as the thioxanthone derivatives, 2- chlorothioxanthone and 2-isopropylthioxanthone, require an amine synergist. The synergist,

2-dimethylaminobenzoate, 31 also acts as a solvent for the thioxanthone derivatives. The

combined amine-thioxanthone derivatives act as effective photoinitiators in pigmented (Ti02) organic systems such as printing inks, varnishes and paints. A few thioxanthone derivatives,

I -substituted chloro- and bromo-4-n-propoxythioxanthone, have exhibited photoinitiation in the absence of an amine co-synergist. 32 They are believed to undergo very efficient direct dehalogenation of the C-Cl bond.

b.) Cationic photoinitiators

The understanding and development of cationic photoinitiators is not as extensive as that of radical photoinitiators and has only recently come to the forefront. For many years it was assumed that initiating cations could not be generated by photochemical means, since it was believed that photons in the near UV and visible range did not have sufficient energy to ionize isolated organic molecules.33 Methods were developed to photochemically generate initiating cations through the formation of a charge-transfer complex. This charge-transfer complex, after excitation, cleaves into initiating ions.

The major classes of cationic photoinitiators include: 1) Bronsted acids (H+BF 4• and

H+AsF6"), 2) Lewis acids (BF3 and PF5), 3) triarylsulfonium salts, 4) diaryliodonium salts, and 22

5) triarylselenonium salts. 34 Examples of cationic photoinitiators are given in Figure 4.

A problem associated with the initial development of aryldiazonium initiator salts was

nitrogen gas evolution upon irradiation. Bubbles released in cured films resulted in

deformation of the coating. More stable diaryliodonium, triarylsulfonium and selenonium

salts have been developed in response to these problems. Scheme 13 shows the use of the cationic photoinitiator triarylsulfonium salt, and the polymeriz.ation of a monomer (M). 29 Free radicals are also produced and this cationic-free radical mixture is considered to be a "hybrid initiator." These hybrid initiators may be effective in acrylate-epoxy systems. Cationic initiators are a more recent development for the polymerization of epoxy and vinyl-ether based polymers. Due to their expense and toxicity, cationic initiators remain less widely used than their radical initiator counterparts.

Scheme 13

HMn +x- + Impurity Polymer

4. Methods for Photogeneration of Carbocations

Thermal methods for the generation of carbocations often involve harsh reaction conditions utilizing strong acids, which may result in unwanted side products. The use of 23

Figure 4

Examples of cationic photoinitiators: triarylsulfonium salts, diaryliodonium salts,

Bronsted acids and Lewis acids. 24

O-s-O-Os·PFs-9 ~ 1 ~

~s•ssF - CH~ 'cH:----<::'>-cN 25 photochemical or excited state methods allows the reaction to be carried out under milder and more selective conditions. The photochemical route for the generation of carbocations also addresses environmental issues such as the control of air-polluting organic compounds, disposal of toxic waste and the reduction of energy required for many types of reactions.

Photochemically initiated carbon-carbon bond cleavage to generate carbocations may also be used for the selective breakdown of fossil fuels, such as coal, providing a more energy efficient method than current thermal routes. 35 Understanding the photogeneration of carbocations will also provide fundamental insight into excited state reactions. Six major methods for the photogeneration of carbocations have been recently reviewed36 and are discussed below.

a.) Photoheterolysis of Carbon-Heteroatom Bonds

The most prominent method for the photogeneration of carbocations is

37 38 photoheterolysis of carbon-heteroatom bonds. • In addition to generating a carbocation, the corresponding heteroatom anionic leaving group is also formed as shown in Scheme 14.

Scheme 14

I hv -c-x -c+ + X· I I

3 3 44 45 Carbocations generated through photoheterolysis include: di- and triarylmethyl, 9-4 xanthyl • and fluorenyl46-4 8 cations. Heteroatom anionic leaving groups include halide, acetate, hydroxide, tosylate, 4-cyanophenoxide, cyanide and phosphonium chloride. 26

Minto and Das44 demonstrated both heterolytic and homolytic bond cleavage in

solutions of 9-phenylxanthen-9-ol. The relative efficiency of each process was shown to depend on the nature of the solvent. Polar hydroxylic solvents gave mainly the 9- phenylxanthenium catio~ whereas in nonpolar solvents, photolysis is dominated by homolytic cleavage to the 9-phenylxanthenyl radical (Scheme 15).

Scheme 15

+ •OH

~ + H,o•-

1 2

Homolysis and heterolysis of the carbon-heteroatom bond of para-substituted diphenylmethyl halides, acetates and ethers, 39 produced radicals and cations in photolyzed acetonitrile solutions. Cation/radical ratios appear to be affected by not only the nature of the substituent on the benzene ring, but also by the nature of the leaving group (Scheme 16). 27

Scheme 16

Ar"'-. CH+ + Cl_ Ar'/ hv 248 nm

b.) Carbon-Carbon Bond Cleavage

The photolysis of covalent compounds composed of stable anions and cations was observed using picosecond and nanosecond laser techniques. 49 Pienta and coworkers prepared coordination complexes from the triphenylcyclopropenyl cation and malonitrilo, acetonitrilo and fluorenyl anions (Scheme 17). Initially, a photon is absorbed by either the

Scheme 17

+ ~ )r ~ Ph Ph 28

stilbene chromophore of the triphenylcyclopropenyl portion, or the aryl chromophore of the

starting carbanion. The singlet excited state initiates an electron transfer process resulting in

a radical ion pair. This radical ion pair can either undergo back electron transfer to form the triplet excited state of the triphenylcyclopropenyl portion or they can undergo homolytic carbon-carbon bond cleavage of the central bond to give both the triphenylcyclopropenyl cation and its corresponding anion.

Wan and Muralidharan50 observed photochemical heterolytic carbon-carbon bond cleavage in the photo-retro-aldol type reactions of several nitroaromatic compounds. The proposed mechanism involves a triplet excited state in the primary photochemical step, generating a nitrobenzyl carbanion and its corresponding carbocation fragment. Carbon- carbon photoheterolysis, producing carbocation and carbanion intermediates, was also suggested by Wan and Muralidharan51 on the basis of product analysis for nitrobenzyl acetals

(Scheme 18). Later laser flash photolysis studies52 confirmed the formation of the p- nitrobenzyl anion and the oxocarbocation.

Scheme 18

c.) Heterolysis of Photogenerated Radical Cations

Radical cations may be photochemically generated through photoinduced electron transfer, photoionization, or electron transfer to a photolytically generated oxidant such as 29

the sulfate radical anion. Carbon-carbon bond cleavage of these radical cations results in a

carbocation and a radical fragment. One specific example includes the biphotonic ionization

and resulting cleavage of 2,3-dimethyl-2,3-diphenylbutane (bicumene). 53 The radical cation

intermediate undergoes carbon-carbon bond cleavage to form the cumyl radical and cumyl

cation (Scheme 19). A mild preparation of alkyl radicals was reported by Mella et al. 54 via

Scheme 19

o-tt-01·-~- o-<· + )---0

photoinduced electron transfer from 2-alkyl and 2,2-dialkyldioxolane to singlet excited benzene-1,2,4,5-tetracarbonitrile. A subsequent fragmentation of the donor radical cation yields alkyl radicals and dialkoxy carbocations.

55 56 Akaba et al. • has studied the photoinduced electron transfer of several diaryl and aralkyl ketones and aldehydes. In the presence of the sensitizer, 2,4,6-triphenylpyrylium tetrafluoroborate, these carbonyl compounds can undergo carbon-carbon bond cleavage via electron-transfer quenching of the tetrafluoroborate salt. The resulting cation radicals

(Scheme 20) cleave to form arylmethyl cations and benzoyl or formyl radicals. 30

Scheme 20 . -'-

CH3-0--C+~ ¢ CH3

Laser flash photolysis studies gave the first direct evidence for the formation of arylmethyl cations through carbon-carbon bond cleavage of a photogenerated cation radical of a carbonyl compound. 56

d.) Oxidation of Photochemically Generated Radicals

Photochemically generated radicals can be oxidized thermally or photochemically to 31 form their corresponding carbocations. Aqueous solutions of triphenylacetic acid were photolyzed, using an excimer laser, to initially produce the triphenylmethyl radical and the hydrated electron. 57 Depletion of the triphenylmethyl radical is evident upon a second pulse, leading to the fonnation of the triphenylmethyl cation. The photochemical conversion of the radical into the cation appears to be highly (75-85%) efficient. In aqueous solutions the trityl cation forms its corresponding alcohol (Scheme 21 ).

Scheme 21

hv P~C· 248 or308 nm

P~C+ +

The diphenylmethyl radical was photogenerated through laser flash photolysis of 1, 1- diphenylacetone. Several electron-transfer quenchers such as carbon tetrachloride, m dicyanobenzene or benzyl bromide were examined and the resulting diphenylmethyl cation yields determined. 58

e.) Protonation of Photochemically Generated Carbenes

Kirmse, Steenken, and Kilian observed the generation of diaryl carbocations through laser flash photolysis of diaryl carbenes in protic solvents. 59 Earlier work concerning the

formation of diaryl carbocations through heterolytic photocleavage of Ar2CH-X provided data

39 42 used to identify transients derived from diaryl carbene precursors. ' Laser flash- photolysis 32

of 4,4'-dimethoxydiphenyldiazomethane in acetonitrile-water gave rise to the same transient

absorption (Amax 500 nm) as that obtained by photoheterolysis of (Ar)2 CHOAc. The data

supports protonation of singlet diaryl carbocations. Kirmse and coworkers also demonstrated

the formation of allylic cations through protonation of photochemically generated vinyl

carbenes (Scheme 22). 60

Scheme 22

hv Me OH

f) Photoprotonation of Styrenes, Aromatics and Phenylacetylenes

Styrenes produce Markovnikov-oriented hydration products when irradiated in

61 62 aqueous acids (Scheme 23). ' Additional aromatic alkenes, alkynes and allenes have also

Scheme 23

OH hv 1 I ArCH=CH2 (ArCH = CH:i) * Ar-CH-CH3

given Markovnikov products through intra- and intermolecular proton transfer upon photohydration in dilute aqueous acid. 63 McClelland and coworkers (Scheme 24) observed through laser flash photolyses, the formation of cyclohexadienyl cations from irradiation of mesitylene in HFIP (1,1,1,3,3,3-hexafluoro-2-propanol).64 33

Scheme 24

Me Me Me~Mey hv Me~Mey HOCH(CF3)2 Me Me Me

Carbocations have been defined, their applications as photoinitiators discussed and methods ofphotogeneration described. Next, the general reactions of the photochemistry of carbonyl compounds will be given.

B. Photochemistry of Carbonyl Compounds

Ketones, aldehydes, and carboxylic acids have been studied to examine their photochemical reactivity. The majority of work concerns aldehydes and ketones, since their longest wavelength absorptions are in a readily accessible region, having a Amax of~ 280-300 nm. 65 In contrast, the less studied carboxylic acids and their derivatives have a Amax of 200-

220 nm. 65 There are two types of electronic transitions possible for the carbonyl group, these

65 66 being n-1t· or 1t-1t· . • The first type of transition, n-1t·, corresponds to excitation of a nonbonding ( n) electron on oxygen into an antibonding 1t • orbital of the carbonyl group.

These n-1t· transitions are symmetry forbidden, their intensities weak and extinction coefficients low. 66 A second means of electronically exciting the carbonyl group is through a 1t -TI• transition. This involves promotion of an electron from the highest occupied molecular orbital (HOMO) of the 1t system to the original lowest unoccupied molecular orbital (LUMO). These 1t-1t· transitions are typically symmetry allowed, their intensities 34 strong and extinction coefficients high.

Electronic excitation results in a change in electronic distribution around the carbonyl group. Polarization in the carbonyl group accounts for its ready reaction at the carbonyl carbon in the ground state. In the excited state, an unpaired electron exists in both a 1t • antibonding orbital and a p-type orbital on oxygen, resulting in oxygen being electron deficient and carbon being electron rich and exhibiting a nucleophilic character. Triplet carbonyl chromophores (n,1t*} have a diradical character in their chemical and physical characteristics. This is a result of electron spin repulsion and results in the diminished effect of any zwitterionic tendencies of the carbon-oxygen bond. Triplet carbonyl chromophores

(n,1t*} exhibit photochemical reactions similar to the alkoxy group such as ex-cleavage and hydrogen abstraction. Since the energy difference between the singlet and triplet (n,1t*} states of aliphatic ketones is low and intersystem crossing possible, reactions may occur from either the singlet or triplet state, or a mixture of the two. In aryl ketones, enones and similar conjugated systems, the (n,1t*} and (1t,1t•) excited singlet states are lower in energy and at longer wavelengths. Due to conjugation in these systems, intersystem crossing occurs at a higher rate, and thus we are concerned mainly with the triplet excited states (n,1t*} and (1t,1t).

The lower energy state oftens depends on a combination of solvent effects and the nature of the substituent. The n,1t· and 1t,1t· transitions are affected to various degrees as the nature of the solvent is changed. 66 In protic solvents, the carbonyl group can become protonated and excitation ofthe nonbonding electron ( n, 1t *) is more difficult, thus a characteristic shift in the

A.max in the absorption spectrum to lower wavelengths and higher energy when compared to spectra in nonprotic solvents. The 1t, 1t • excitation is much less sensitive to the nature of the 35

solvent.

In the 1t-1t· transition, electron density at the carbonyl group is increased. Electron

donor substituents in conjugation with the carbonyl group reinforce this effect, stabilizing the

1t-1t· triplet state and destabilizing the n-1t· triplet state. In contrast, there is electron movement away from the oxygen in an n-1t· transition, and this will become stabilized relative to the (1t,1tJ triplet state with the addition of electron accepting groups. Increasing solvent polarity causes a hypsochromic shift to shorter wavelengths of (n,1t*} absorptions and a bathochromic shift to longer wavelengths of (1t,1t•) absorptions (Figure 5). Thus, polar solvents stabilize the 1t, 1t • triplet relative to the n, 1t • triplet and can cause an inversion in states, since reaction will follow the lowest energy state. The nature of the excited state determines, for example, in enones the preference for reaction at the carbonyl group (n,1t*} or the ethene moiety (1t,1tJ and for aryl ketones the efficiency of carbonyl hydrogen abstraction and addition processes. Aryl ketones also have a high efficiency for intersystem crossing, long wavelength absorptions (-320 nm) and small differences in energy between singlet and triplet excited states. Due to these characteristics they are often used as triplet sensitizers.

The most prominent photoreactions of ketones67 include the 1) Norrish Type I reaction: cleavage of the C-C bond (a to the carbonyl carbon), the 2) Norrish Type II reaction: intramolecular abstraction of a hydrogen atom by the carbonyl oxygen atom, 3) intermolecular hydrogen atom abstraction and 4) the Paterno-BU.chi reaction: cycloaddition to an electron-rich or electron-poor olefin. Each will be described in detail below. 36

Figure 5

Effect of solvent polarity on wavelength shifts. 37

So----- Nonpolar Polar solvent solvent

LIB1

1. Norrish Type I Reactions

Norrish Type I reactions, 65 or a-cleavage reactions, were first described in 1907 by

Ciamician and Silber. This reaction involves the excitation of the carbonyl group and results in the cleavage of a carbon-carbon bond adjacent, or a to, the carbonyl group (Scheme 25).

Scheme 25

0 0 II hv II R-C-R' R-C· + ·R'

Norrish Type I reactions tend to preferentially cleave the bond toward the more substituted carbon. This is true in the case of aliphatic acyclic ketones, resulting in an alkyl and acyl radical. In aliphatic aldehydes though, the carbon-hydrogen bond cleaves preferentially.

Initially, radical pairs are produced from homolytic cleavage in Norrish Type I reactions. These radicals can go on to react in a variety of ways including intermolecular radical-radical recombination or disproportionation of the same type of radical or different types of radicals. Initiation of polymerization of various monomers may also consume free radicals. The nature of the photoproducts often depends on the conditions used in the irradiation. For example, the major photoproducts in the gas phase photolysis of acetone are methane, ketene, carbon monoxide, and ethane (Scheme 26). 66 In a condensed medium the 39

Scheme 26

0 0 II hv II CH3-C-CH3 CH3-C• + ·CH3 / ~ CH3-CH3 I CH2=C=O + CH4 ·CH3 + CO

radical fragments remain in close proximity and produce mainly dimers and polymers. An

example of a Type I a-cleavage is demonstrated in the photoreaction of diisopropyl ketone.

The radical fragment abstracts a hydrogen from the other radical of the initial radical pair

formed. This hydrogen abstraction results in the formation of either a ketene and a saturated

fragment, or an aldehyde and an alkene. The ketene can undergo secondary reactions

resulting in, for example, a carboxylic acid and an ester in solvents such as water and alcohol.

Norrish Type I cleavage of di-t-butyl ketone (Scheme 27)65 results in the following

photoproducts: recombination to regenerate di-t-butyl ketone, cleavage of the initial radical

pair to produce t-butyl radical and carbon monoxide, combination to produce the t-butyl

radical coupling product, 2,2 '3,3' tetramethylbutane, and disproportionation to produce 2- methylpropane and 2-methylpropene. 40

Scheme 27

313 nm, pentane recombination

disproportionation Jcleavage and combination

2. Norrish Type II Reactions

Another common photoreaction of carbonyl compounds is an intramolecular hydrogen

abstraction reaction referred to as the Norrish Type II reaction. 66.68 Initially, a cyclic transition

state is formed, as the intramolecular hydrogen abstraction by the carbonyl oxygen atom

occurs. Then homolytic cleavage of one of the carbon-carbon bonds adjacent to a carbon

atom attached to the carbonyl carbon occurs, resulting in a ketone and its corresponding

alkene. A specific example of a Norrish Type II reaction would be the fragmentation of2-

pentanone. Upon irradiation the fragmentation results in an enol precursor, followed by the formation of acetone and ethylene (Scheme 28). Norrish Type II reactions can occur when 41

Scheme 28

0 II hv CH3-C-CH2CH2CH3

0 II CH3-C-CH3 + CH2=CH2 ---

excited aldehydes or ketones possessing a y hydrogen undergo intramolecular hydrogen abstraction to form a six-membered transition state. The resulting 1,4-biradical can either cyclize to form the substituted cyclobutanol product or cleave to give an alkene and an enol.

The enol goes on to form the corresponding ketone.

Intramolecular hydrogen abstraction is usually favored from the y carbon in ketones and aldehydes in which the molecular structure allows a close approach between the excited carbonyl oxygen and a hydrogen attached to an sp3 carbon. The hydrogen abstraction initially results in a diradical (1,4-diradical in the case of y hydrogen abstraction), and is followed by cyclization and cleavage reactions. A general example is shown in Scheme 29. 65 42

Scheme 29

OH R1 ?i H, ,R1 hv cyclization 2 R'.. c"--/c'R2 R-tl-R

Quantum yields for Nonish Type II reaction from both singlet and triplet (n, 1t*} states of pentan-2-one, hexan-2-one, 5-methylhexan-2-one were determined by Wagner.69

Conclusions drawn from these studies of intramolecular y hydrogen abstraction of ketones show that the reactivity of both singlet and triplet states increase with a decrease in y C-H bond strength, with reactivity values greatest for tertiary y C-H. Molecular geometry of the ground state also affects the efficiency of hydrogen abstraction. 70

Intramolecular hydrogen abstraction can occur from not only the y position (when there is a y hydrogen substituent), but also from p, a, and €-hydrogens. P-Hydrogen abstraction is rare, but has been observed in the syntheses of cyclopropanols. 5-Hydrogen abstraction, generating a 1,5 diradical, is important in syntheses in which there are no y hydrogens. 71 Abstraction of €-hydrogens is also rare and occurs in compounds containing only €-hydrogens or those compounds in which the molecule's conformation brings the E, rather than the y hydrogen, in close proximity to the carbonyl group. 43

3. Intermolecular Hydrogen Abstraction

Intermolecular hydrogen abstraction or photoreduction involves initial transfer of a hydrogen atom from a donor molecule (solvent, reagent or ground state molecule of the starting material) to the carbonyl oxygen of the excited state ketone. The triplet {n,1t*) state is usually the excited state involved in the hydrogen abstraction process, since the singlet excited state is often too short lived. Secondary radical reactions then follow initial hydrogen abstraction. The reaction of acetone in cyclohexene shown in Scheme 30 demonstrates the

Scheme 30 0 -hv -

photoproducts produced upon intermolecular photoreduction. 72 The nature of the resulting photoproducts depends on the concentrations of reagents, structure of the reducing agent, identity of the initial radical precursor, and temperature. Photoreduction, in cases such as fluorenones and biphenyl ketones, is inefficient, due to the high electron density on oxygen in the excited {1t,1t•) triplet state.

The triplet {n, 1t•) state of carbonyl compounds has a radical-like structure 73 and readily abstracts hydrogen from donor molecules. The rates of hydrogen abstraction are related to the C-H bond strength of the donor molecule. Thus, the weaker the C-H bond, the greater the rate coefficient for hydrogen abstraction and the shorter the triplet lifetime.

Another example of hydrogen abstraction is shown in the irradiation ofbenzophenone in 2- propanoL resulting in pinacol formation (Scheme 31). 74 Laser flash photolysis studies 75 show 44

Scheme 31

Ph:zC=O hv . . Ph2COH + (CH3)2COH . Pti:zC=O + (CH3hCOH ---

Ph Ph . I I 2(P~COH) -- Ph-C-C-Ph I I OHOH

that an increase in the ease of hydrogen abstraction resulted in an enhanced disappearance of ketone and the absorption intensity of the transient.

Since photoreduction involves a triplet (n,1t*) state, the efficiency of the process is affected by change in solvent and by the type and position of the substituent, as seen in aromatic carbonyl compounds. A comparison of hydrogen abstraction efficiencies of acetophenone (small energy difference between n, 1t • and 1t, 1t • triplet state) and benzophenone

(large energy difference between n,1t· and 1t,1t· triplet states) was performed by Porter and

Wtlkinson. 75 In nonpolar cyclohexane abstraction efficiencies are comparable, since the lower triplet state in both cases is (n,1t*). As the C-H bond strength is lowered, as in 2-propanol, the rate of hydrogen abstraction increases for both acetophenone and benzophenone.

Acetophenone though, acts more slowly than benzophenone, since the polar properties of the solvent promote an increasing contribution to the lowest-lying energy state. Bertzophenone 45 exhibits a large energy difference between the (n,TI*) and (1t,1t*} triplet states, and thus does not show a tendency for state inversions. Solvent polarity exhibits a marked effect on the hydrogen abstraction abilities of xanthone. 76 Laser flash photolysis studies demonstrate a solvent-induced inversion ofthe reactive triplet (n,1t*) and the unreactive triplet (1t,1t*} states of xanthone.

4. Patemo-Biichi Reaction

The Patemo-Biichi reaction involves the reaction of a photoexcited carbonyl

66 77 78 chromophore and an alkene. • ' This reaction can be considered a reversible [2 + 2] photocycloaddition reaction and results in the formation of a four-membered cyclic ether or oxetane (Scheme 32). The n,1t· carbonyl group adds to the ground state alkene from either

Scheme 32

R R' R2C=O + R'CH=CHR' Rt:!: R'

the S1 or T1 state. A biradical precursor to the oxetane has been confirmed by picosecond spectroscopy.

C. Photochemistry of the Carbon-Carbon Double Bond

Simple olefins and dienes exhibit important differences in the geometry of their ground and excited states, resulting in variations in their singlet and triplet excited state energies.

Ethylene has its lowest triplet energy when the angle of twist between the two, sp2 carbon 46

atoms is 90 degrees, relative to the ground state and is referred to as the p (perpendicular)

geometry.79 Thus, upon initial irradiation, the planar ground state geometry of an olefin

changes to an orthogonal excited state geometry. 80 This twisted geometry in the triplet

excited state of olefins, plays an important part in their ability to undergo cis-trans

isomerization. Geometrical factors of alkenes and polyenes often determine which reaction

of a series of energetically feasible competing processes actually takes place. 81 The broad

scope of C=C photochemistry is evidenced by the many photochemical rearrangements and

:fragmentations possible for alkenes and polyenes. This is in part due to their generally high excess energies after absorption of light and the large number of deactivational processes available to the excited state compound.

In compounds containing only the C=C chromophore, the allowed 1t-1t· transition occurs from the highest occupied 1t molecular orbital to the lowest unoccupied 1t• orbital.

Absorption is intense, with absorption maximum of 170-180 nm in simple nonconjugated alkenes. 82 Ethylene shows an intense band at 145-190 nm, with a weak tail to approximately

207 nm. This tail extends to longer wavelengths as alkyl substitution occurs. Ethylene 1t-1t· triplet absorption is weak and extends to 3 50 nm. There is a large energy difference between

1 1 the planar singlet and the planar triplet states (150 kcal moi- vs 80 kcal mol- ) and the rate constant for intersystem crossing is very low. Thus, direct irradiation of alkenes results in photoproducts derived from the singlet (1t,1t) state, but not the triplet (1t,1t•) state. This explains the fact that triplet state reactions of alkenes can be studied only by triplet sensitizers as a source of electronic energy.

The most frequently observed photoprocess for alkenes is the photochemical 47

83 interconversion of cis and trans isomers. Electronic repulsion of the triplet (1t, 1t *) state, coupled with a weakened o bond results in an orthogonal conformation and explains the predominance of cis-trans isomeriz.ation. In order to generate the triplet ( 1t,1t *) excited state of either the cis or trans isomer, and subsequent rapid relaxation to an orthogonal species, intersystem crossing must occur. Due to a large energy difference between the singlet and triplet excited states, various carbonyl sensitizers such as acetone or acetophenone, or aromatic hydrocarbon sensitizers such as benzene, toluene, or xylene can be employed to bring about isomerization.

Simple monoenes can undergo a number of photochemical reactions. These include:

1) cleavage of the double bond, 2) a-cleavage, 3) P-cleavage with ring opening and substituent migration (1,2 shifts and 1,3 shifts), 4) intramolecular hydrogen transfer, and 5) intramolecular C-C addition.

Alkenes such as methylpropene undergo mainly cleavage of the C-H or C-C bond,84 with only a small percent double bond cleavage and subsequent carbene formation. Cleavage of the double bond does occur in polyhalogenated alkenes (tetrafluoroethylene) and ketenes.

These reactions result in the formation of carbenes which can be trapped or used as a synthetic route to dimerization (Scheme 33).

Scheme 33

In contrast to the cleavage of the double bond in alkenes, photochemical cleavage of 48

ex-bonds in simple alkenes is common. Ethylene, 85 difluoroethylene, 86 I-butene, 87 and I -

iodocyclohexene, 88 readily undergo ex-cleavage (Scheme 34). More complicated compounds

Scheme 34

H H hv H>==< H

such as that shown in Scheme 35, also undergo photochemical ex-cleavage. 89

Scheme 35

hv +

P-Cleavage is rare in carbonyl photochemistry, but is fairly common in alkene photochemistry. P-Cleavage can result in ring opening reactions such as the conversion of cyclobutenes to cyclobutadienes. 90 Fragmentations or photochemical cycloreversions, such as that shown in Scheme 36, can result in both a P- and y-bond being cleaved. The P- cleavage of alkenes can also involve 1,2 and 1,3 ( allyl) shifts. Allyl chloride and allyl bromide, upon irradiation in acetone, gives the corresponding cyclopropylhalide in 10-17% yield. 91 49

Scheme 36 , hv ~ II + Ph ~Ph

The photoreaction of nonconjugated dienes, having their 1t-systems separated by an sp3 hybridized carbon atom (i.e. 1,4-pentadienes) and of3-phenyl-alkenes in which one of the double bonds is replaced by a benzene ring is called the di-1t-methane rearrangement. When p,y-unsaturated ketones are involved, the reaction is referred to as the oxa-di-1t-methane rearrangement. In the case of 3-phenyl-1-propene, irradiation results in a 1,2 shift and formation of the phenylcyclopropane (Scheme 37). 1,4 and 1,3-biradicals can be involved

Scheme 37 . c)_hv___ &-· 0

in some di-1t-methane rearrangements, with the specific reaction path followed determined by the most stable biradical intermediate. Often the rearrangement follows a concerted path.

D. Earlier Work On The Thermal and Photogeneration ofXanthyl Cations

Deno and Sacher demonstrated the thermal generation of the xanthyl cation from several xanthyl alcohols (Figure 6): 2-(9-xanthyl) ethanol a, 1-(9-xanthyl)-2-propanol b, 1-(9- 50

Figure 6

Xanthyl alcohols and xanthyl ketone used to thermally generate the xanthyl cation: 2-

(9-xanthyl)ethanol a, 1-(9-xanthyl)-2-methyl-2-propanol b, 1-(9-xanthyl)-2-propanol c, 1-(9- xanthyl)-2-phenyl-2-propanol d and 9-xanthylacetone 1. The trityl alcohol, 4,4,4-triphenyl-2- methyl-2-butanol e, demonstrated no cation formation. 51

~OH)(CfiJh

~OH)(CH3)Ph

e 52 xanthyl)-2-methyl-2-propanol c, 1-(9-xanthyl)-2-phenyl-2-propanol d and the xanthyl ketone,

9-xanthylacetone 1. 92 These reactions were performed in moderately to strongly ( 40-70%) concentrated aqueous acidic solutions. The formation of the xanthyl cation was monitored using UV/vis spectroscopy and appeared as a peak at 375 nm absorbance. The corresponding thermally generated alkenes, ethylene and propene, were observed using a standard procedure

93 to titrate for gases (Br2 in acetic acid, KI and titration with sodium thiosulfate). The formation of acetone from 9-xanthylacetone was shown through the formation of its 2,4- dinitrophenylhydrazone derivative. In addition to the xanthyl compounds, an analogous series

of triphenylmethyl derivatives were also examined in 0-96% H2SO 4 for thermal generation of the triphenylmethyl cation. Formation ofthe cation was not observed (UV/vis spectroscopy)

even when subjected to I00°C in 75-96% H2S04• No cation was thermally generated from

4,4,4-triphenyl-2-methyl-2-butanol e and other triphenylmethyl derivatives, presumably due to the lesser stability of the triphenylmethyl cation, when compared to its xanthyl cation counterpart. Deno and Sacher proposed a Grob-type fragmentation94 for cation formation from the xanthyl alcohols, proceeding through protonation of the hydroxyl oxygen with subsequent heterolytic C-C bond. cleavage, to generate the xanthyl cation and the corresponding alkene (Scheme 38). 53

Scheme 38

In the case of the xanthyl ketones, Trudell and Keeffe95 demonstrated the thermal generation of xanthyl cations from 9-xanthylacetone and 9-xanthylacetophenone via a proposed Al mechanism. They described initial reversible protonation of the carbonyl oxygen followed by heterolytic bond cleavage to form the xanthyl cation and its corresponding ketone. The possibility of a concerted mechanism could not be ruled out (Scheme 39).

Scheme 39

1: R=Me 2:R=Ph j

H ~ uo~ 54

Mechanistic studies have also been performed by Trudell and Keeffe95 on the thermal generation ofthe xanthyl cation from 2-methyl-3-(9-xanthyl)propene. The xanthyl cation was thermally formed under mildly to strongly acidic conditions from the xanthyl propene. They proposed the first step shown in Scheme 40, to be a rate-controlling protonation followed by a rapid fragmentation to form the xanthyl cation and 2-methylpropene. Thermal formation of the xanthyl cation did not appear to be first order, but rather involves a complex stepwise mechanism. They suggest that in changing the structure from the carbon-oxygen double bond to a carbon-carbon double bond, the mechanism has a greater chance of being concerted.

Protonation of the oxygen is rapidly reversible, whereas on the carbon it is not.

Scheme40

H ~ + UON

Relatively few instances of carbocation photogeneration involving carbon-carbon bond cleavage exist, and these have been discussed in an earlier section (I.A.4.b.). The 55 development of novel methods for carbon-carbon bond cleavage are desirable, especially in areas such as the fossil fuel industry. The area of photoinitiated polymerization, with the development of new and more efficient photoinitiators, is another significant application of carbocation photochemistry. Carbocations have also been shown to be important intermediates in many organic reactions. Thus, we are particularly interested in novel methods to photogenerate carbocations, specifically the xanthyl cation, requiring less severe

(heat or acid) conditions than those necessitated in the ground state. Mechanisms unique to the photogeneration of xanthyl cations have also been investigated and will be described herein. CHAPTER II

RESULTS

A. 9-Xanthylacetone and 9-Xanthylacetophenone

1. Syntheses

9-Xanthylacetone 1 and 9-xanthylacetophenone 2 were prepared by a two-step

95 96 procedure. ' Acid-catalyzed reaction of xanthydrol with ethyl acetoacetate or ethyl benzoylacetate gave the P-keto esters ethyl 2-(9-xanthyl)-3-oxobutanoate 3 and ethyl 2-(9- xanthyl)-3-phenyloxopropanoate 4, respectively (Scheme 41 ). Base hydrolysis of 3 yielded ketone 1, while hydrolysis of 4 gave ketone 2. The P-ketoesters, 3 and 4, and the ketones,

1 and 2, were characterized via melting point, 1H NMR and 13C NMR (1 and 2 only).

Scheme 41

0 0 II II ~ CH3CH20CCH2CR UaV H~~

1: R =Me 2: R =Ph

56 57

2. Preparative Photolyses

Solutions were prepared of 1 and 2 ( 4 x 104 M) in aqueous acidic solution ( 5-3 0%

H2S04) with acetonitrile cosolvent for solubility reasons (aqueous/acetonitrile 2: 1 for 1 and

1:2 for 2). The sulfuric acid percentages have not been corrected for the cosolvent (see

experimental section for details). Irradiations were performed at 3 00 or 254 nm in a Rayonet

RPR-100 photoreactor. The reaction progress was monitored by UV/vis absorption

spectroscopy, since the absorption characteristics of the substrates and the xanthyl cation are

quite different. Solutions of 1 and 2 were irradiated for 1-4 min. A spectrum corresponding

to the xanthyl cation (absorption maximum of 3 74 nm) was observed to grow in cleanly over

the entire acid range examined, with a decrease in the absorption peaks corresponding to 1

and 2. Clean isosbestic points were observed for growth of the xanthyl cation absorption

spectra for irradiation of both 1 and 2 in each acid concentration studied. Figure 7 illustrates

a series of absorption spectra obtained from irradiation of 1 in 30% H2S04. The reaction is

obviously very fast, since the spectra were obtained at 1 min irradiation times using only two

lamps in the Rayonet reactor. A control dark reaction, also monitored by absorption

spectroscopy, was used to ensure that no thermal reaction occurred in these acidic solutions

over the time scale of the irradiations.

Continued irradiation resulted in a decrease in xanthyl cation absorbance, with the concomitant increase in the absorption spectrum of a peak with absorption maximum of 340 nm. Formation of this photoproduct was observed upon irradiation of the xanthyl cation solutions produced from photolyses of 1 and 2 over the entire acid range of 5-30% aqueous sulfuric acid. Figure 8 illustrates absorption spectra obtained following further irradiation of 58

Figure 7

Absorption spectra measured following irradiation of 9-xanthylacetone 1 in 30%

H2S04 (with 33% acetonitrile cosolvent) with an excitation wavelength of300 nm. Spectra were measured at I min irradiation time intervals and show growth of the xanthyl cation peaks. 59

0 0 in

0 0 ""1'

z t C!> z UJ ..J UJ > < 0 % 0 (T')

0 0 C\I

0 0 0 0 0 0 0 0 0 0 0 0 0 v m C\I lO 0 m ID ~ (T') .-1 0 . 0 0 0 0 0 0 + + + + +

3::lN'V'8t!OS8'V' 60

Figure 8

Absorption spectra measured following irradiation of xanthyl cation solutions

produced from photolysis of 9-xanthylacetone 1 in 30% H2S04 (with 33% acetonitrile cosolvent) with an excitation wavelength of 300 nm. Spectra were measured at 4 min irradiation time intervals and show absorption peaks corresponding to the xanthyl cation decreasing with the concomitant formation of xanthone. 61

0 0 LO

0 0 "q'

::c.... z(!) L!.J _J UJ > 0 < 0 x CT)

0 0 N

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 CD lD ~ N 0 .... 0 0 0 0 0 + + + + +

3:lN"V8tiOS8'1 62

the xanthyl cation produced from irradiation of 1 in 30% H2S04. This secondary

photoproduct was identified as xanthone by comparison to the absorption spectrum of an

authentic sample of xanthone. Further confirmation of xanthone as the secondary

photoproduct came from its isolation following exhaustive irradiation of 1. A yield of75%

for the formation of xanthone was obtained from GC analysis. Aqueous workup and product

isolation followed by GC analysis also identified acetophenone and bixanthyl as minor

photoproducts (yields of approximately 2-3 % were observed for both photoproducts) from

the irradiation ofl in 5% H2S04. Later irradiation ofl at higher concentrations showed the

formation of an additional new photoproduct, 9-xanthylidene acetophenone. Its isolation and

identification will be described in detail in Section 11.A.6.a.i.

The initially formed xanthyl cation was stable indefinitely in the aqueous acidic

solutions without further irradiation, according to absorption spectra. A control experiment

was performed to determine the source of the xanthone. Xanthydrol was dissolved in the

same acidic media (5-300/o aqueous H2S04, with acetonitrile cosolvent). UV/vis spectroscopy

showed that the xanthyl cation was thermally generated, although the xanthyl cation was in

partial equilibrium with xanthydrol _in these dilute acid concentrations. Irradiation of the

argon-purged cation/alcohol solutions at 254 nm also yielded xanthone. No xanthone was

produced in a control dark experiment.

3. Acid Catalysis

Earlier studies established that formation of the xanthyl cation from the thermal

9 95 cleavage of 1 and 2 was strongly dependent on the acid concentration. 2. As described in 63

the previous section (Il.A2.), the xanthyl cation is produced from irradiation of 1 and 2 over

a wide range of acid concentrations. In order to determine whether this photochemical

generation of the xanthyl cation is subject to acid catalysis, we irradiated 1 in solutions of

varying acidity (5-300/o aqueous H2S04, with acetonitrile cosolvent). Aliquots were removed

after irradiation times of I min, 2 min, and 4 min. The relative amounts of xanthyl cation

produced in the varying acid solutions were obtained from the absorption intensity at 3 74 nm.

However, under these dilute acid concentrations, the xanthyl cation is in partial equilibrium

with its corresponding alcohoL xanthydrol. 97 In order to account for this, sufficient acid was

added to each aliquot in order to shift the cation-alcohol equilibrium completely to the cation

side prior to analysis by UV/vis spectroscopy.

The cation absorbance at 374 nm was identical for each acid solution at a particular

photolysis time, showing that the amount of xanthyl cation formed is independent of acid concentration (Figure 9). The amount of xanthyl cation generated in the photochemical experiments varied from I 0% after I min irradiation to 60% after 4 min irrradiation time.

Carrying out the irradiations to higher conversion was not attempted due to a secondary photochemical reaction producing xanthone. Control dark experiments ensured no thermal generation of the cation from 1 under either the experimental or analysis conditions.

4. Triplet Sensitization and Quenching Studies

In order to determine the multiplicity of the reactive state of 1 and 2, sensitization experiments were carried out with benzophenone as the triplet sensitizer. Sufficient

benzophenone was added to an aqueous acidic (5% H2S04) solution of 1 and 2 to ensure that 64

Figure 9

Plot of absorption intensity of xanthyl cation (at 374 run) versus percent aqueous

H2S04 (acid concentration not corrected for cosolvent). The xanthyl cation was produced from irradiation of9-xanthylacetone 1 with irradiation times of 1 min (.t.), 2 min(•), and 4 min ( • ). These times correspond to cation conversions of 10%, 32%, and 60% respectively. 65

--,..--.,.--.---..---..___.---..----.------~~~L[) r'")

• •

I.[) N

~ 0 0 Cf) • • N :r:N

~ t1) ~ 3 "3 0 • ~

• t1)

L.-J"--1__..__,,.__._.--'-_..._.__._--'-_,__.__..__.___,_-'--~...__.o . I""). N . 0 0 0 0 0 0 e:>uoq..Josqv 66 the sensitizer absorbed at least 90% of the incident light (A.max=300 run). Acetonitrile was

again used as cosolvent for solubility puposes. The high concentration and strong absorbance of benzophenone did not allow us to use absorption spectroscopy to analyze the amount of xanthyl cation formed. Instead, the photolysis solutions were analyzed by GC for both reactant disappearance and product formation. GC analysis of the reaction mixture following neutralization and extraction showed that the amounts of both 1 and 2 decreased, with the formation of the same photoproducts obtained from the direct irradiation. Reactions of both

1 and 2 in the presence of the sensitizer suggest the involvement of the triplet excited state for these compounds.

Triplet quenching experiments were performed on 1 and 2 using potassium sorbate or oxygen as quenchers. The effect ofthese triplet quenchers on the reaction was determined by comparison ofthe absorption intensity at 374 nm (corresponding to xanthyl cation) for the quenched versus unquenched irradiations. For both quenchers, solutions of 1 or 2 were

prepared in aqueous acidic (5% H2S04) acetonitrile. As previously described (Section II. A.

3. ), sufficient acid was added to the photolyzed solutions to shift the cation/alcohol equilibrium completely over to the cation side prior to analysis. For the experiments using oxygen as a triplet quencher, solutions of 1 and 2 were purged with argon or oxygen for 30 min. For identical irradiation times, the solutions purged with oxygen gave absorption intensities approximately 35% lower at 374 nm, compared to the argon-purged solutions.

Irradiation at 300 run of 1 and 2 in the presence of potassium sorbate ( 10-3 M) similarly gave lower absorbance values for the amount of cation formed, when compared to irradiated solutions lacking the diene quencher. Higher concentrations of potassium sorbate 67 could not be used, since the diene would then absorb the incident light (A.ex=300nm). The oxygen and potassium sorbate quenching experiments showed that less cation was formed in the presence of the triplet quenchers, providing further support for the triplet state mechanism.

5. Laser Flash Photolysis Studies

Laser flash photolysis studies were performed on 1 and 2 in aqueous acidic acetonitrile using a XeCl excimer laser 0.·ex=308 nm). The acetonitrile: &O: acid ratio was 90:8:2 by volume. The larger amount of acetonitrile was necessary for substrate solubilities at the higher concentrations required for laser studies. The transient spectra obtained following laser flash photolysis of either 1 or 2 in oxygen or nitrogen purged solutions show formation of an intermediate with an absorption maximum of 345 nm, within the rise time of the laser pulse (10 ns). The spectrum ofthe primary photoproduct is shown in Figure 10, as the trace obtained 10 µs after laser excitation of2. The xanthyl radical was identified as the primary

44 8 photoproduct by comparison to literature reports .9 and by independently generating the radical from irradiation of xanthene i~ the presence of tert-butyl peroxide. Reference to the literature spectra eliminated the 9-xanthylidene carbene as a possible intermediate, since its absorption spectrum exhibits broad maxima at 374, 383, and 393 nm. 99

In nitrogen-saturated solutions, the absorbance corresponding to the xanthyl radical decays slowly (Figure 10), whereas in oxygen-saturated solutions the xanthyl radical absorption spectrum decay is accompanied by the growth of a new spectrum with an absorption maximum at 370 nm. This secondary photoproduct was identified as the xanthyl 68

Figure 10

Transient spectra obtained following photolysis of9-xanthylacetophenone 2 in 90:8:2

CH3CN:H20:H2S04. Solutions were purged with nitrogen and show the decay of the initially formed xanthyl radical. 69

0 ~

ti) ti) VJ VJ 0 VJ :::2 :::2 ::l :::2 ::l 0 0 0 0 ~ 0 0 0 C> C> ~ ~ N ff) v 0 C"t + 1 + y t .q-

0 ..__,]' ~ til) d 0 ~ 00 v ('fl ~ ~ 0 \0 ('fl

~ ('fl

0 N 0 ('') 0 0 0 ~ N ...-4 0 ci ci ·a·o uJt:>a 70 cation, by comparison to its characteristic UV/vis spectrum (Figure 7). Both 1 and 2 show the initial formation of the xanthyl radical with subsequent conversion to the xanthyl cation, with clean isosbestic points. Figure 11 illustrates this clean conversion of the xanthyl radical to the xanthyl cation for irradiation of2. Growth of the 370 nm absorbance, corresponding to xanthyl cation formation, is a pseudo-first-order process for both 1 and 2 (Figure 12).

Least squares analysis gave rate constants for growth of the cation of 1.5 x 104 s-1 for 1 and

2.1 x 104 s-1 for 2.

In solutions that were rigorously purged with nitrogen, no absorptions due to the xanthyl cation were detected, and only the decay of the xanthyl radical was observed. Figure

13 compares the behavior of oxygen- and nitrogen-purged samples of 2 by monitoring the absorption at 3 70 nm. This wavelength is the absorption maximum for the xanthyl cation.

The xanthyl radical also absorbs at this wavelength, but with less intensity. In oxygen saturated solution there is an initial jump in the absorption intensity resulting from formation of the radical within the duration of the laser pulse, followed by a slower pseudo-first-order growth resulting from conversion of the radical into the more strongly absorbing cation. In nitrogen-purged solutions the same initial jump due to radical formation is detected; however, in this case the absorption decays in a second-order process as the radical recombines and reacts to form stable products other than the cation. These results indicate that the presence of oxygen is necessary for xanthyl cation formation. The magnitude of the initial jump differs for 1 and 2 (Figure 12), which can be attributed to differences in quantum efficiency for formation of the radical.

To examine the role of acid on radical and cation formation, 1 and 2 were irradiated 71

Figure 11

Transient spectra obtained following photolysis of9-xanthylacetophenone 2 in 90:8:2

CH3CN:H20:H2SO4 . Substrate solutions were purged with air and show clean conversion of the xanthyl radical o.. max=345 run) to the xanthyl cation (A.max=374 run). 72

0 \0 ~

0 Cl) Cl) Cl) Cl) Cl) ::i ::i ::i ::i ~ ::i 0 0 0 0 0 0 0 0 _.. _.. N M ~ 0 N + + t y t ~ 0 ~ ~ '-' ~I)c: 0 u 00 u M - ~ ~ 0 \0 M

~ C'f"l

0 N _.. OC'f"l \C? "'! ~. C'f"l <'! 0 0 0 d 0 0

·a·o u11~a 73

Figure 12

Transient absorption waveforms of 9-xanthylacetophenone 2 (solid) and 9-

xanthylacetone 1 (dotted) at 370 run in 90:8:2 CH3CN:H20:H2S04 purged with air. 74

I • •) 1• ' \. 8 ' ,, - " '• -,.. _, __ ~---~------~------

------..----.-----...---.....------..... ----..------~§ qN" 75

Figure 13

Transient absorption waveforms for 9-xanthylacetophenone 2 (370 nm) in 90:8:2

CH3CN:H20:H2S04 purged with nitrogen and purged with air. 76

8 C"I

0 ~ s-

0 :!:

0 ~

8 ~ Cf.) - ::l ~ ~ ~

•0

~ Nq 77

in 100% acetonitrile and in a solution of acetonitrile/water (90: 10 by volume). Transient

spectra obtained following laser flash photolysis show that the xanthyl radical is produced

from both substrates within the rise time of the laser pulse, but no peaks corresponding to

xanthyl cation formation were observed to grow in. Apparently the presence of acid is

required to observe efficient cation formation. Preparative irradiation of 2 in acetonitrile

resulted in formation of xanthone.

6. Photoproduct Studies

a. 9-Xanthylidene Acetophenone

i. Syntheses

We had previously observed through preparative photolyses and GC analyses the

formation of xanthone, acetophenone and bixanthyl as photoproducts resulting from the

irradiation of 2. 100 Careful examination of photoproducts obtained following irradiation of

2 at higher concentrations showed the formation of an additional new photoproduct which we isolated and identified. Preparative photolyses of 2 in 5% aqueous acidic solution, using

acetonitrile as a cosolvent (aqueous/acetonitrile 1:2), were performed at 254 nm in a Rayonet

RPR Photoreactor with irradiation times of 5-30 min. The resulting photoproducts were separated via silica gel column chromatography and identified through GC analyses. Analysis of the new photoproduct by 1H and 13C NMR (including APT), FT-IR and GC-MS suggested

9-xanthylidene acetophenone 5 as a possible structure. The structure of the new photoproduct was confirmed through independent synthesis of 5, by the procedure shown in

Scheme 42. 101 Xanthone was reacted with lithium phenylacetylide to yield 9- 78 phenylethynylxanthenol 6. 9-Phenylethynylxanthenol 6 was dehydrated using concentrated

HCl to yield 9-xanthylidene acetophenone 5.

Scheme 42

c=c-u 0 ~ THF 6 + UaJlJ

ii. Quantum Yields

Photogeneration of the xanthyl cation and subsequent formation of photoproducts from the photolysis of 1 and 2 appeared to be an efficient process, requiring only 2 Rayonet

102 105 lamps for 1-4 min irradiation at 300 nm or 254 nm. We used chemical actinometry - to measure quantum yields for formation of photoproducts. V alerophenone was chosen as the 79 chemical actinometer for our system based on its similar structure and wavelength, as described extensively by Wagner et al. 106 Using a reported quantum yield of 0.40 for acetophenone formation from photolysis of valerophenone in cyclohexane, 107 along with GC response factors for valerophenone, acetophenone, 9-xanthylacetophenone and 9- xanthylidene acetophenone, the quantum yield for 9-xanthylidene acetophenone formation from the photolysis of2 (2x104 M) was found to be 0.84 ± 0.06.

iii. Synthesis of Proposed Tertiary Benzylic Alcohol IO: Precursor of 9-Xanthylidene Acetophenone 5 a.) Method 1. In order to determine the mechanism of formation of 9-xanthylidene acetophenone 5, two methods to synthesize a proposed tertiary benzylic alcohol 10 precursor

were explored. Method 1 (Section V.G.8.a.) attempted to hydrate 9-xanthylidene acetophenone 5 at the 9-position using aqueous sulfuric acid (0.10-14.5 wt%). Several products, as shown in the Table I below, were identified: 9-xanthylacetophenone 2, the starting material 9-xanthylidene acetophenone 5, 9-hydroxyxanthene 7, xanthone 8, acetophenone 9, and an unidentified peak, having a GC retention time of75 minutes (tR75). 80

Table 1

GC AREA PERCENT ANALYSES OF PRODUCTS OBTAINED FROM THE

ATTEMPTED HYDRATION OF 9-XANTHYLIDENE ACETOPHENONE 5

2a 5b 7c 8d 9e Run# % aq. H,S04 tR 75f 1 0.60 4 45 1 46 0 0 2 0.10 0 78 1 20 0 0 3 1.50 3 56 1 35 0 1

4 9.70 0 16 22 34 10 15 5 9.70 14 0 7 76 7 6

6 9.70 1 28 28 29 8 5 7 14.5 0 6 4 67 0 2 8 5.0 4 71 9 8 2 0 9 5.0 0 42 14 27 6 14 39-xanthylacetophenone; b9-xanthylidene acetophenone; c9-hydroxyxanthene; ~anthone; eacetophenone; i\inidentified peak

GC/MS analyses were perfonned on the peak representing the unidentified compound and indicated a molecular ion M+• peak of 374 (see Section VI.). The molecular weight of the proposed tertiary benzylic alcohol 10 is 316 and thus excludes the structure 10 as that of the unidentified peak shown in Table I. Several other compounds were excluded as possible structures from their molecular ion peaks. These included: bixanthyl 11 (mw 362), 9- hydroxybixanthyl 12 (mw 378), and xanthopinacol 13 (mw 394). A simple fragmentation or rearrangement from one of these structures could not give a molecular ion peak of 3 7 4.

Either the alcohol 10 is not fonned, or in the presence of aqueous acid the- alcohol is 81

dehydrated, reforming the 9-xanthylidene acetophenone starting material. Alternatively, if

alcohol 10 is formed in the synthesis, it may decompose by an alternate route, such as on the

GC column or upon injection.

12 13

In order to exclude the possibility of formation of 10 on the column, 9-xanthylidene acetophenone 5 was placed in acetonitrile and injected under the same GC conditions as the products isolated above. Only starting material, 9-xanthylidene acetophenone 5, was observed. Since 10 was not observed, we concluded that it was not formed from 5 on the GC column.

b.) Method 2. Irradiation ofxanthylacetophenone in neutral solution was then explored as a route to the formation of the proposed tertiary benzylic alcohol 10 intermediate (see Section

V.G.8.b.). GC analyses, shown in Table II, indicated the following photoproducts (area% uncorrected for response factors): 9-xanthylacetophenone 2 (starting material}, 9- xanthylidene acetophenone 5, xanthone 8, and acetophenone 9. Three unidentified photoproducts were also observed at retention times of tR=40 minutes ( 1-8%), tR=43 minutes

(4-35%), and tR=44 minutes (1-13%). 82

Table 2

GC AREA PERCENT ANALYSES OBTAINED FROM PRODUCTS OF THE

IRRADIATION OF 9-XANTHYLACETOPHENONE 2

Run# T= (min) 2a 5b sc 9d tR 40e tR 43f tR 44g 1 1 72.4 5.8 9.1 1.4 4.6 4.0 6.6 2 1 88.5 5.1 0.6 0.6 2.2 35.0 3.0 3 5 58.6 5.2 21.5 3.4 4.2 0 7.2 4 5 56.7 1.0 4.6 0 1.6 0 1.8

5 10 52.1 0 33.3 7.2 0 0 7.3 6 10 48.1 6.4 32.5 1.1 5.1 0 6.9 7 25 0 0 73.4 15.6 0 0 10.9 8 25 0 1.8 77.6 0 7.9 0 12.7 39-xanthylacetophenone; b9-xanthylidene acetophenone; cxanthone; dacetophenone; e4unidentified peaks

GC/MS analyses were then performed on each of the three unidentified peaks (see

Section VI.). The peak having a retention time oftR=40 showed a molecular ion M+. peak

of 312 (or M-4, considering the mw of the proposed benzylic alcohol 7 to be 316). This

would be highly unlikely though, since this would require the loss of four hydrogens.

Typically, benzylic alcohols exhibit a strong parent peak and prominent M-1, M-2, and M-3

peaks, but not M-4. 108 Although tR=40 was not identified, it was excluded as the possible tertiary benzylic alcohol 10 due to its M+• peak and fragmentation pattern. Upon evaluation of their molecular ion peak and fragmentation pattern the two peaks corresponding to retention times of 43 (M+.=259) and 44 (M+. =249) minutes were not identified but were 83 excluded as the proposed benzylic alcohol intermediate 10 (M+-=316). The ethers, 15 and 16, were also considered as possible photoproducts for the compounds with retention times tR=40 or tR=43 or tR=44 minutes, but were excluded due to their molecular weight (302) and the dissimilar molecular weights and fragmentation patterns of the unidentified photoproducts.

H-C-CH39 I oQO O-CH2-0 o(D 15 16

7. Fluorescence

Fluorescence spectra were obtained for solutions of 9-xanthylacetone 1 O·ex =270, 285 nm) and 9-xanthylacetophenone 2 O"ex =290 nm) in fluorescence grade acetonitrile (purged with dry argon) and aqueous acidic acetonitrile. No emission was observed for these solutions (240-290 nm), suggesting .a short lived fluorescent state and adding support for a triplet excited state mechanism.

B. 2-Methyl-3-(9-xanthyl)propene

I. Syntheses

95 109 2-Methyl-3-(9-xanthyl)propene 14 was prepared by the three step procedure • shown in Scheme 43. Acid catalyzed reaction ofxanthydrol with ethylacetoacetate gave the 84

P-ketoester, ethyl 2-(9-xanthyl)-3-oxobutanoate 3. 100 P-Ketoester 3 was then base hydrolyzed to yield 9-xanthylacetone 1. 100 The Wittig reactionuo,m of 9-xanthylacetone 1 with the prepared ylide methyltriphenyl phosphonium bromide and sodium amide yielded 2- methyl-3-(9-xanthyl) propene 14. The propene 14 was characterized via melting point,

GC/MS, 1H NMR and 13C ~as described in Section V.G.5.

Scheme 43

0 II cc~ I 0 0 II II 0)01CH2CH3 ~ C~CH20CCH2CCH3 Uo)V H+ 6. I 3

l5%NaOH,6

CH2 0 II II ~c-c~ Ph3P+CH38( NaNH2 cQO~

14 1

2. Preparative Photolyses

Solutions were prepared of 14 (4 x 10-4 M) in aqueous acidic solution (0.5-50%

H2S04) with acetonitrile cosolvent for solubility reasons (aqueous/acetonitrile 1:2). The sulfuric acid percentages have not been corrected for the cosolvent. Irradiations were performed at 300 or 254 nm in a Rayonet RPR-100 photoreactor. The reaction progress was 85 monitored by UV/vis absorption spectroscopy, since the absorption characteristics of the substrates and the xanthyl cation are quite different. Solutions of 14 were irradiated for 1-8 min. A spectrum corresponding to the xanthyl cation (absorption maximum of374 nm) was observed to grow in cleanly over the entire acid range examined, with a decrease in absorption peaks corresponding to 14 (Figure 14). A control dark reaction, also monitored by absorption spectroscopy, showed a minor amount of thermal reaction occurring in these acidic solutions over the time scale of the irradiations.

Continued irradiation resulted in a decrease in xanthyl cation absorbance with the concomitant increase in the absorption spectrum of a peak with absorption maximum of 340 nm. This secondary photoproduct was identified as xanthone by comparison to the absorption spectrum of an authentic sample of xanthone. Aqueous workup and product isolation followed by GC/MS analysis also identified the following photoproducts: 9- xanthylidene 2-methylpropene (mw 234 or two mass units less than the parent), M+ propene

(corresponding to mw 279, or M 236 + mw 43), and an unidentified peak with a retention time of 42 min. Refer to the appendix for spectroscopic details and structural analyses for the unidentified photoproduct. A minor.amount of the radical coupling product, bixanthyl, was also observed.

3. Acid Catalysis

Earlier studies by Trudell and Keeffe established that formation of the xanthyl cation from the thermal cleavage of propene 14 was strongly acid dependent95 (Scheme 40). In order to determine whether the photochemical generation of the xanthyl cation from 14 is 86

Figure 14

Absorption spectra measured following irradiation of 2-methyl-3-(9-xanthyl)propene

14 in 300/o H2S04 (with 33% acetonitrile cosolvent) with an excitation wavelength of254 nm.

Spectra were measured at 2 min irradiation time intervals and show growth of the xanthyl cation peaks. 87

0 0 ID

I 0 0 In ~co i ~t c: I~ ~ CX) co .q N z £~ 0 U=U· 0 }f • u

0 0 N ...... 0 0 0 0 N ~ IO m 0 en ... 0 m ID '11 ~ 0 • • "• • . • "N N ... .. 0 0 + + + + +

3~NYSHOSBV. 88 similarly dependent upon the acid concentration, we irradiated 14 in solutions of varying acidity. Aliquots were removed after varying irradiation times, and the relative amounts of the xanthyl cation produced were obtained from the absorption intensity at 3 7 4 nm. The xanthyl cation is in partial equilibrium with its corresponding alcohol under these dilute acid

97 concentrations . Sufficient acid was then added to shift the equilibrium completely to the cation side prior to analysis by UV/vis spectroscopy. Solutions containing more than 30%

H2S04 appeared to be dominated by the thermal generation of the xanthyl cation, with absorbance values for the xanthyl cation increasing as the percent acid increased (Figure 15).

For irradiated solutions containing 0-20% H2S04 (Figure 16), the amount of cation formed, as indicated by the absorbance at 374 nm, was independent of acid concentration. Thus, a different mechanism occurred in the photochemical generation of the xanthyl cation from 14, in contrast to the thermal mechanism (Scheme 40) reported by Trudell and Keeffe. 95

Solutions were also irradiated and monitored for xanthyl cation formation using only

distilled acetonitrile (over CaH2) or distilled acetonitrile/boiled deionized water. The unexpected absorbance noted at 374 nm, after T=O min irradiation, indicated that these solutions were, in fact, not completely devoid of acid, or that acid is not needed for the reaction. These results suggest some degree of thermal reaction competing with the photochemical reaction. After T=4 min irradiation (254 nm, 2 lamps), an increase in absorption intensity at 3 74 nm was noted for both the solutions containing distilled acetonitrile and distilled acetonitrile/boiled deionized water. Solutions containing only acetonitrile exhibited a proportionately greater increase in absorption intensity than the solutions containing both acetonitrile and deionized water. 89

Figure 15

Plot of absorption intensity of xanthyl cation (at 374 run) versus 1-50% aqueous

cfi. 0 LO

'U ·o cfi. <( 0 C\J 0 "i:: :::J ~ :::J (J) cfi. cfi. 0 .,_ ~ ~

co co v C\J 0 0 0 0 0 38NV8tlOS8V 91

Figure 16

Plot of absorption intensity of xanthyl cation (at 374 run) versus 0-20% aqueous

0 N

"'O ~ ·o ~ <( 0 .i:: ::J ~ en::J ~ 0 ~ ~ 3:

L--~~~~~~~~---+--=~----+-~~ 0 LO ~ V ~ M ~ N ~ ~ LO 0 0 V 0 M 0 N 0 ~ 0 0 0 0 0 0 0 38N\f8tl0S8\f 93

Analogous studies using 9-xanthylacetone 1 (5-30% H2S04 and irradiation times of

1, 2 and 4 min) showed no increase in absorbance for the xanthyl cation (374 nm) at a

particular irradiation time over the acid range studied, 100 as described in detail in a previous

section (see Section II.A.3.). For aqueous acidic solutions of 9-xanthylacetone 1,

photogeneration of the xanthyl cation was independent of the H2S04 concentration over the

acid range studied. Similarly, the lack of acid catalysis over the acid range 1-30% in the

photochemical generation of the xanthyl cation from propene 14 indicates a different

mechanism operating in the excited state.

4. Triplet Sensitization

In order to determine the multiplicity of the reactive state, benzophenone was

examined as a potential triplet sensitizer. Sufficient benzophenone (Sxl0-4M to Sx10"3M) was

added to the aqueous acidic (1 % H2S04) acetonitrile solutions of 14 to insure that at least

90% of the incident light (.Amax 300 nm) was absorbed by the sensitizer. The high

concentration and strong absorbance of benzophenone did not allow the direct analysis of the xanthyl cation via UV/vis spectros.copy. Irradiated solutions were instead analyzed for

reactant disappearance and product formation. GC analysis of the benzophenone-containing

solutions following neutralization and extraction showed a decrease in area percent

photoproducts and no new photoproducts when compared with similar solutions obtained

from irradiation of 14 not containing the triplet sensitizer. Reaction of 14 in the presence of

benzophenone suggests the involvement of the triplet excited state. Similar results using benzophenone were observed with both 9-xanthylacetone 1 and 9-xanthylacetophenone 2, 94

with GC analyses showing a decrease in the starting materials and the formation of the same

photoproducts obtained from direct irradiation (see Section 11.A.4.). This again substantiates

the involvement of the triplet excited state. 100

5. Triplet Quenching

Triplet quenching was then examined using the known triplet quenchers, oxygen or

potassium sorbate. Aqueous acidic acetonitrile solutions of 14 were purged for 30 minutes

with either oxygen or argon and then irradiated as described earlier (see Section 11.A.4.).

Sufficient aqueous H2S04 was then added to an aliquot of irradiated solution in order to shift

the alcohoVcation equilibrium completely to the cation side. The UV/vis absorbance at 374

nm, corresponding to the absorption maximum of the xanthyl cation, showed approximately

I 0% lower absorption intensity for solutions purged with oxygen, when compared to the

argon-purged solutions.

Aqueous acidic solutions either containing the diene quencher potassium sorbate or lacking the quencher were irradiated, and the absorbance of the xanthyl cation (374 nm) compared. Solutions with equivalent irradiation times showed approximately 50% lower absorption intensity for solutions containing potassium sorbate quencher when compared to those lacking the diene quencher. This provides further support for a triplet excited state mechanism for the photogeneration of the xanthyl cation from xanthylpropene 14.

Prior experiments, using oxygen as a triplet quencher in solutions of9-xanthylacetone

1and9-xanthylacetophenone 2, showed absorption intensities approximately 35% lower at

374 nm when compared to argon-purged solutions. Irradiated solutions of these two ketones 95 containing the triplet quencher potassium sorbate also gave lower absorption intensities than solutions lacking the diene quencher. 100

6. Fluorescence

Fluorescence spectra o.cx=240, 250, 280, and 290 nm) were obtained for solutions of xanthylpropene 14 (4. 7 x 10-3 M and 4. 7 x 104 M) in fluorescence grade acetonitrile (purged with dry argon) and aqueous acidic acetonitrile. No emission was observed for these solutions, suggesting a short lived fluorescent state. This adds support for a triplet excited state mechanism. CHAPTER III

DISCUSSION

A. 9-Xanthylacetone and 9-Xanthylacetophenone

1. Acid Catalysis

The results reported above demonstrate that the xanthyl cation can be generated from

1 and 2 photochemically under less acidic conditions than required for the thermal reaction.

Formation of the xanthyl cation in the ground-state reaction is strongly acid dependent.92.95

In contrast, preparative irradiations of 1 and 2 in aqueous acidic acetonitrile showed no acid

dependence on the amount ofxanthyl cation formed, over an acid range of 5-30% H 2S04.

No acid catalysis was observed in the photochemical reaction for conversions of I 0-60%

cation. Acid concentrations higher than 30% H 2S04 could not be studied in the irradiations due to competing thermal reactions.

2. Mechanism of the Photogeneration of the Xanthyl Cation from 1and2

The lack of acid catalysis in the photogeneration of the xanthyl cation, combined with the observation from laser flash photolysis experiments that the xanthyl radical is the primary photoproduct, indicate a different mechanism operating in the excited state. We propose the two-step mechanism for the cation photogeneration illustrated in Scheme 44. The first step

96 97 involves initial homolytic carbon-carbon bond cleavage of 1 or 2 to generate the xanthyl radical and the corresponding cx-acyl radical. This bond cleavage process would be independent of the acid concentration and is in accord with the experimental finding. The radical coupling product bixanthyl was isolated from preparative irradiations of 2 and supports the intermediacy of the xanthyl radical in the mechanism. Acetophenone was also confirmed as the product from photolysis of2. Its formation can be attributed to hydrogen abstraction by the precursor cx-acyl radical.

Scheme 44

I 3* 0 hv casCR 0

1: R=Me 2:R=Ph

H H 0 0 II II ~ ~ CH3CR + + •CH2CR uov u"v0

The second step of the proposed mechanism for cation photogeneration involves oxidation ofthe xanthyl radical to generate the xanthyl cation. Laser flash photolysis studies show that the initially formed xanthyl radical is cleanly converted to the xanthyl cation in the 98 presence of oxygen. Oxygen has a reduction potential of -0.05 V vs NHE, 112 while the oxidation potential of the xanthyl radical is -0.002 V vs SCE (or 0.24 V vs NHE). 35 Thus, oxygen can act as the oxidizing agent to produce the xanthyl cation, since the electron transfer process is calculated to be exothermic by 0 .19 V.

Oxidation of the xanthyl radical to the xanthyl cation may occur via a one-step

( outersphere) electron transfer pathway or by a two-step (innersphere) process in which oxygen adds to the xanthyl radical followed by heterolytic carbon-oxygen bond cleavage. 113

Both the innersphere and outersphere mechanisms would generate xanthyl cation and superoxide. Our experiments are incapable of distinguishing between these two mechanisms regarding the role of oxygen.

The laser flash photolysis experiments demonstrate that no cation is formed in rigorously deoxygenated solutions. Rather, the observed decay of the radical absorption in the absence of oxygen is likely due to secondary chemical reactions of the xanthyl radical, such as the formation of the radical coupling product, bixanthyl. We were unable to fully inhibit xanthyl cation formation in the preparative irradiations solely by purging the photolysis solutions with argon. Apparently, .small amounts of adventitious oxygen are sufficient to oxidize the xanthyl radical, due to the low substrate concentrations ( 10-5 M) used in the preparative irradiations.

3. Triplet Sensitization and Quenching Results

The initial bond cleavage is proposed to proceed through an excited triplet state of 1 and 2. Triplet sensitization experiments using benzophenone showed that both 1 and 2 99 underwent the same bond cleavage process in the presence of the sensitizer as in the direct irradiation. Triplet quenching experiments using potassium sorbate resulted in a decreased efficiency of formation of the xanthyl cation produced from 1and2. Both the sensitization and triplet quenching experiments implicate a reactive triplet state in the mechanism. Oxygen was also observed to act as a quenching agent in the preparative irradiations. Irradiation of

1 and 2 in oxygen-purged solutions resulted in less xanthyl cation produced compared to the amount of cation produced in argon-purged solutions. While oxygen is presumed to act as the oxidizing agent to generate the xanthyl cation, it may also act as a triplet quencher. 114

Detection ofthe xanthyl cation following laser flash photolysis of 1 and 2 requires the presence of aqueous acid. Irradiation of 1 or 2 in 100% acetonitrile with oxygen or nitrogen purging resulted in formation of the xanthyl radical, as detected by transient absorption spectroscopy. No xanthyl cation peaks were observed to grow in. The absorption spectrum corresponding to the xanthyl radical decayed over time, presumably through secondary radical reactions. The xanthone photoproduct isolated from preparative irradiation of 2 in acetonitrile supports the idea that radical chemistry dominates under these conditions, presumably due to a slower rate constant for cation formation in the less polar solvent.

Laser flash photolysis of 1 and 2 in solutions of acetonitrile/water (90: 10) similarly showed formation of the xanthyl radical, although no cation peaks were detected in either oxygen- or nitrogen-purged solutions. In the deoxygenated solutions the xanthyl radical is not oxidized, and likely undergoes the secondary radical reactions previously described. In the presence of oxygen, the radical presumably is oxidized to the xanthyl cation. This cation would then be rapidly trapped by water to give xanthydrol. McClelland and co-workers 100 generated the xanthyl cation via laser flash photolysis of xanthydrol in 1: 4 acetonitrile/water.

A rate constant of2.3 x 104 s·1 was measured for reaction of the cation with solvent. 45 The presence of acid is thus required to stabilize the cation for transient detection in our experiments.

Carbonyl compounds are well known to undergo photochemically induced a-cleavage

(Norrish Type I) reactions to give two radical fragments. 114 Cleavage of the P-bond is considerably rarer and occurs with compounds containing relatively weak Ca-C~ bonds, such as cyclopropyl ketones115 and y ,&-unsaturated ketones.116 a,p-Epoxy ketones similarly undergo photochemically induced P-cleavage, due to the weak Ca-0~ bond. 117 Bromoacetyl chloride undergoes P-cleavage of the C-Br bond as well as a-cleavage of the C-Cl bond. 118

In the case of 1 and 2, the facile P-homolytic bond cleavage is attributed to the stability of the xanthyl radical. Alternatively, the reaction can also be viewed as a benzylic-type cleavage, as opposed to P-cleavage.

4. Product Quantum Yields

Photogeneration of the xanthyl cation and subsequent formation of photoproducts from the photolysis of 1 and 2 appeared to be an efficient process, requiring only two Rayonet lamps for 1-4 min irradiation at 300 nm or 254nm. In order to measure product quantum yields, valerophenone was used as a chemical actinometer. Using a quantum yield of0.40 for acetophenone formation from photolysis of valerophenone in cyclohexane (A.ex =254nm),107 the quantum yield for 9-xanthylidene acetophenone 5 formation from the photolysis of 2 in

5% aqueous H2SO/acetonitrile (1 :2) solutions was found to be 0.84±0.06. 101

A literature search of reported quantum yields for benzylic cleavage and homolytic carbon-carbon bond cleavage gave values for comparison. Table 3 summarizes the results of several representative quantum yields reported in the literature. Noh, Lei and Turro119 report a quantum yield of0.84 ± 0.06 for product formation of 1,2-diphenylethane in benzene 102

TABLE3

PRODUCT QUANTUM YIELDS

COMPOUND A.run QUANTUM REF.

YIELD 0 H CH2-~-0 This (XJb - 2543 0.84±0.06 Work 0

0 0-CH2_ g_CH2-0 313b 0.84±0.06 119 0.70±0.10 120

d---cH2-Lc~-O 313b 0.31±0.02 119

0 C1-t:3-o- CH2_ g_CH2-0-CH3 313b 0.71±0.11 120

0 II CH3-C-CH3 250-320c 1.0 122

0 II 290-320c CH3-CH2-C-CH2·-CH3 0.60 123

0 313b 0.37 125 o-~-CH2-CH2-CH2-CH3 254e 0.40 107 313 3 0.85 125 313d 0.90 125

3aCetonitrile, 11benzene, cneat, d2-methyl-2-propanol, ecyclohexane

from irradiation of dibenzyl ketone through homolytic carbon-carbon bond cleavage. As 103 methyl groups are added to the a-phenyl group, the quantum yield for formation of diarylethanes decreases to 0.31 ± 0.02 for o-tolyl-methyl benzyl ketone, and 0.11 ± 0.01 for mesitylmethyl benzyl ketone. Product quantum yields for the formation of diarylethanes in the irradiation of other dibenzyl ketones have been reported earlier by Robbins and

Eastman. 120 As a comparison, selected product quantum yields for representative

106 101 11 125 25 ketones • • 9- are also shown in Table 3. Wagner reported quantum yields for valerophenone (acetophenone formation via Norrish Type II reaction) of0.37 (in benzene) to 0.90 (in I-butyl alcohol). Therefore, in comparison to product quantum yields of dialkyl ketones, the photolysis of 2 to yield 9-xanthylidene acetophenone 5 appears to be very efficient (0.84 ± 0.06) and is in reasonable agreement with similar quantum yields for homolytic carbon-carbon benzylic cleavage reported in the literature.

5. Photoproduct Studies a. Mechanism ofFormation of9-Xanthylidene Acetophenone

In addition to the photoproducts xanthone, acetophenone and bixanthyl, 9- xanthylidene acetophenone 5 was. also observed in GC analyses following preparative photolyses of9-xanthylacetophenone 2. We propose the mechanism in Scheme 45 to account for the formation of 5 from irradiation of 2 in aqueous acidic acetonitrile. Irradiation of 2 in aqueous acidic acetonitrile results in the formation of the xanthyl radical la and a-acyl radical

2a through P-homolytic cleavage. Unreacted starting material 2 is also present in solution.

The xanthyl radical la can then abstract either a methylene hydrogen from 2 to generate the

9-xanthylacetophenone methine radical 3a, or it abstracts the benzylic hydrogen from 9- 104 xanthylacetophenone 2 to generate the 9-xanthylacetophenone radical 4a. A second hydrogen abstraction from 3a or 4a by either the xanthyl radical la or the a-acyl radical 2a results in the formation of the stable 9-xanthylidene acetophenone 5.

Scheme 45

hv ~ CH3CN I H20, H + Uo,,V f3-cleavage 1a 2a

2 H· abstraction

0 H CH-C-0 H· abstraction ~- Disproportionation UoYJ

5 3a 4a

Alternatively, hydrogen abstraction from 2, 3a or 4a might also occur by cyanomethyl radicals. These cyanomethyl radicals would be generated from hydrogen abstraction of the acetonitrile cosolvent by xanthyl radical la. Earlier studies by Wan and Krogh126 demonstrated abstraction ofH· from acetonitrile in aqueous acidic solutions by the 9-fluorenyl radical. Similarly, H· abstraction by the xanthyl radical from acetonitrile might occur to form cyanomethyl radicals in our system. The mechanism shown in Scheme 45 would follow a stepwise path, with initial formation of the short lived radical precursors 3a and/or 4a 105 followed by the formation of the stable 9-xanthylidene acetophenone 5. These hydrogen abstractions, resulting in an alkene photoproduct, may be considered a disproportionation

121 130 type reaction. Examples of disproportionation reactions are given in Table 4. - We noted the formation of 1-2% xanthene in only a few instances of 9-xanthylacetophenone 2 preparative photolyses. The formation of bixanthyl from radical recombination of the xanthyl radical was also noted. Independent irradiation of aqueous acidic acetonitrile solutions of xanthene, using similar concentrations, resulted in the formation of bixanthyl. As the concentration of 9-xanthylacetophenone 2 starting material in preparative photolyses was increased, only a small percentage of xanthone and acetophenone was observed, with the majority of the photoproduct being 9-xanthylidene acetophenone 5. In some cases, 9- xanthylidene acetophenone 5 was the sole photoproduct of 9-xanthylacetophenone 2 photolyses.

Additional mechanisms for 9-xanthylidene acetophenone 5 formation were also considered. One possible mechanism (Scheme 46) involves initial carbon-hydrogen benzylic cleavage to generate the stable 9-xanthylacetophenone radical 4a. The benzylic hydrogen of

2 may be abstracted by cyanomethyl radicals generated in the aqueous sulfuric acid/acetonitrile solvent. Literature precedence for benzylic cleavage ofH· includes the work of Camaioni and Franz. 131 Their work also describes the formation of cyanomethyl radicals

from the attack of the sulfate radical anion (S04-·) on acetonitrile. The sulfate radical anion was generated from irradiated solutions containing peroxydisulfate. Generation of the sulfate radical anion is unlikely though in our media. An alternate method to generate cranomethyl radicals was described by Wan and Krogh126 as indicated earlier in this section. They TABLE 4: DISPROPORTIONATION/COMBINATION RATIOS (kikc)

RADICALS DISPROPORTIONATION PRODUCTS kikc REF. 2 (CH3)2CH· H I 3 CH2=C 1.2 127 \ CH3

2 (CH3)3C· CH3 I CH2=C 7.2 127b \ CH3 2Q. 0 1.1 128c

Ph 0-c~ \ 2 9· C=CH2 0.054 129d CH3 I CH3 t-Bu9-CH3 t-Bu 2 9· t-Bu~ 0.21 130e CH3 t-Bu C==CH2 I CH3

adecalin, 30°C, 254 nm; bn-pentane, 30°C, 254 nm; ccyclohexane, 33 °C, 254 nm; d20°C, photolysis; eC6D6, 60°C ...... 0 O'I 107 observed abstraction ofH· from acetonitrile in aqueous acidic solutions by the 9-fluorenyl radical. Similarly, H- abstraction from acetonitrile by the xanthyl radical la (see Scheme 45;

Scheme 46

0 CHi-C--0 hv c¢O -

10 Sa

0 H /C11--o~ 'c - ~ Va~

also possibly generated under the same reaction conditions in Scheme 46) or radical 4a 108

(Scheme 46) might occur to form cyanomethyl radicals. The resulting cyanomethyl radicals can abstract a hydrogen from 2 to generate radical 4a. The 9-xanthylacetophenone radical

4a may be oxidized, with oxygen acting as the oxidizing agent to produce the 9- xanthylacetophenone cation Sa. Oxidation potentials might also be considered for the 9- xanthylacetophenone radical 4a. 132 If the oxidation potential of radical 4a is similar to that ofthe xanthyl radical la, 4a could be readily oxidized by oxygen in solution to the cation Sa.

We were unable to locate redox potentials for 9-xanthylacetophenone radical 4a or 9- xanthylacetophenone cation Sa in the literature. Under aqueous acidic conditions the tertiary benzylic alcohol IO can be formed via nucleophilic trapping of cation Sa, as shown in Scheme

46. This tertiary benzylic alcohol 10 would readily dehydrate to form 9-xanthylidene acetophenone S.

The 9-xanthylacetophenone radical 4a might be formed by initial carbon-hydrogen benzylic cleavage as shown in Scheme 46. Instead of then being oxidized to the 9- xanthylacetophenone cation Sa, 4a could undergo a disproportionation, as shown earlier in

Scheme 45, to directly form the 9-xanthylidene acetophenone S.

In order to investigate the likelihood of the mechanism proposed in Scheme 46, synthesis of the tertiary benzylic alcohol precursor 10 from 9-xanthylidene acetophenone S was attempted. 9-Xanthylidene acetophenone S was heated in the presence of varying

percentages of aqueous H2S04, using acetonitrile as a cosolvent. Isolation and GC analyses were then performed on the reaction mixtures. In addition to unreacted 9-xanthylidene acetophenone S, several products were identified including 9-hydroxyxanthene, xanthone, 9- xanthylacetophenone, acetophenone, and an unidentified peak. GC/MS analyses were 109 performed on the unidentified peak and indicated the molecular ion M+ peak as 374 (mw of proposed alcohol 316). Either the alcohol 10 is not formed, or in the presence of aqueous acid the alcohol is dehydrated, reforming the 9-xanthylidene acetophenone 5 starting material.

Irradiation of 9-xanthylacetophenone in neutral solution using acetonitrile as a co solvent was then explored as an alternate route to the formation of the proposed alcohol intermediate. GC analyses indicated the following photoproducts: acetophenone, xanthone,

9-xanthylidene acetophenone, and 9-xanthylacetophenone starting material. Three additional photoproducts were also observed, but upon evaluation of their molecular ion peak and fragmentation pattern, they were excluded as the proposed benzylic alcohol intermediate 10

(see Section 11.A.6.a.iii.b.).

Mechanisms other than those shown in Scheme 45 or Scheme 46 were also considered. 9-Xanthylidene acetophenone 5 might be formed through initial intramolecular y or P-hydrogen abstraction133 to form a five-membered 17 or three-membered 18 cyclic

alcohol precursor. In the case of y-hydrogen abstraction, this is unlikely since it would involve abstraction of an aromatic hydrogen. In the case of P-hydrogen abstraction of 18, benzylic hydrogen abstraction is likely; however P-hydrogen abstraction is rare. Formation of 9-xanthylidene acetophenone from the 9-xanthylacetophenone methine radical 3a via 110 oxidation to the 9-xanthylacetophenone cation with subsequent deprotonation to form the carbene134 was also considered (Scheme 47). The resulting carbene could rearrange,135 with migration of a hydrogen, forming 9-xanthylidene acetophenone 5.

Scheme 47

0 0 H CH-c-O H~H-C-0 ~- [O] ~- VoV UoV

0 ~-".. "-0'' UoV

Both mechanisms shown in Scheme 45 and Scheme 46 may be operative. Scheme 46 though relies on an initial C-H benzylic cleavage to generate radical 4a. There would have to be some initial P-homolytic carbon-carbon bond cleavage to generate the xanthyl radical la. The xanthyl radical la could then act as the initial H· abstractor, or radical la could abstract H · from acetonitrile to form cyanomethyl radicals. The xanthyl radical la or cyanomethyl radicals could then abstract the benzylic hydrogen to form radical 4a. Although 111 the tertiary benzylic alcohol 10 was not isolated, this does not preclude the existence of this short lived intermediate or the possibility of the generation of9-xanthylidene acetophenone

5 via the alcohol 10. Scheme 45 is the more reasonable route, since it relies upon the formation ofthe xanthyl radical la and its subsequent H· abstraction of either 2, 3a, 4a,and/or acetonitrile. Earlier laser flash photolysis studies and photoproduct analysis of 2 demonstrated the photogeneration of la from 2 under identical conditions. 100 A series of stepwise hydrogen abstractions would then result in the stable 9-xanthylidene acetophenone

5 photoproduct.

B. 2-Methyl-3-(9-xanthyl)propene

The results recorded earlier demonstrated the photogeneration of the xanthyl cation from 2-methyl-3-(9-xanthyl)propene 14 under less acidic conditions than required for the thermal reaction. Preparative irradiations of 14 in aqueous acidic acetonitrile showed no acid

dependence on the amount ofxanthyl cation formed, over the acid range of 1-20% H2S04.

Competing thermal reactions were apparent at acid concentrations higher than 20%. Based on acid catalysis and photoproduct studies, coupled with results observed for 1 and 2, we propose the two step mechanism shown in Scheme 49 for the photogeneration of the xanthyl cation from 2-methyl-3-(9-xanthyl)propene 14. The first step involves initial homolytic carbon-carbon bond cleavage to generate the xanthyl radical and the 2-methylpropene radical.

This step would be independent of acid concentration over the acid range of 1-20% aqueous

H2S04. The radical coupling product, bixanthyl, was also isolated in preparative photolyses of 14 and supports the intermediacy of the xanthyl radical. In a second step the xanthyl 112

radical is oxidized to the xanthyl cation with oxygen as the oxidizing agent.

Scheme 48

CH II 2

ttoCCH:H3C: H20,W 0 14

H H uov~ [O] uov~

An analogous mechanism has also been proposed for the photogeneration of the xanthyl cation from 9-xanthylacetone 1 and 9-xanthylacetophenone 2 (Scheme 44). Lack of

acid catalysis coupled with results from laser flash photolyses and preparative photolyses added support to the mechanism of Scheme 44. Similarly, the first step involves P-homolytic carbon-carbon bond cleavage to generate the xanthyl radical and the corresponding a-acyl radical. The xanthyl radical (photogenerated from 1, 2 and now 14) is then oxidized to the xanthyl cation as shown in both Scheme 44 and Scheme 48. Oxygen can also act as the oxidizing agent in the case of the photogeneration of the xanthyl cation from 14.

Photogeneration of the xanthyl cation from 14 results from an initial homolytic 113

carbon-carbon bond cleavage. This initial cleavage is proposed to proceed through an excited

triplet state of 14. Triplet sensitization experiments using benzophenone (a known triplet

sensitizer) showed that 14 underwent the same bond cleavage process in the presence of the

sensitizer as in direct irradiation. The addition oftriplet quencher, potassium sorbate, showed

a decreased efficiency in the photogeneration of the xanthyl cation from 14. Oxygen, another

known triplet quencher, acted as a quenching agent in preparative irradiations of 14. Results

of triplet sensitization and quenching experiments indicate a reactive triplet state in the

mechanism.

In addition to the photoproducts, xanthone and bixanthyl, 2-methyl-3-(9-

xanthylidene )propene 19 was also observed via GC/MS analyses following preparative

photolyses of irradiated solutions of 2-methyl-3-(9-xanthyl)propene 14. This new

photoproduct 19, two mass units less than the parent, is analogous to the 9-xanthylidene

acetophenone 5 photoproduct observed in the irradiation of 9-xanthylacetophenone 2.

Scheme 49 accounts for the formation of 2-methyl-3-(9-xanthylidene)propene 19 from

irradiation of propene 14 in aqueous acidic acetonitrile. Initial irradiation results in the

formation of the xanthyl radical la and an a-methyl-propenyl radical 6a through P-homolytic

cleavage. These radical photoproducts would be present in the aqueous acidic acetonitrile

solution, as shown in Scheme 50. The xanthyl radical la can then proceed to abstract either

a methylene hydrogen to generate the 2-methyl-3-(9-xanthyl)propene methine radical 7a, or

abstract the benzylic (9-H) hydrogen to generate the 2-methyl-3-(9-xanthyl)propene radical

Sa. In a disproportionation step, a second hydrogen abstraction by either the xanthyl radical

la or the a-methyl-propenyl radical 6a results in the formation of the stable photoproduct 9- 114

xanthylidene-2-methyl-3-(9-xanthylidene)propene 19.

Scheme 49

~Hi ~ a())cc:H,c~; H,OW 0 Ua~

14 1a 6a

14 H· abstraction

~Hi H /C-CH3 CH2 'c c())CCH,• II C(:o Disproportionation 0 19 7a 8a

Numerous examples of radical-radical disproportionation are demonstrated in the literature as shown in Table 4 (see section III. A.5.a.). The formation ofxanthene was not observed from the irradiation of aqueous acidic acetonitrile solutions ofxanthylpropene 14, but some bixanthyl was observed. This suggests hydrogen abstraction is possible from both the a-methyl-propenyl radical 6a and the 9-xanthyl radical la, even though some of the xanthyl radical la would be consumed through radical coupling to generate bixanthyl. In turn, hydrogen abstraction by the a-methyl-propenyl radical 6a would generate the gas, 2- methylpropene, which could not be isolated and identified under our experimental conditions.

Preparative photolyses of aqueous acidic acetonitrile solutions of xanthene using similar 115

concentrations and irradiation times resulted in the formation of bixanthyl and a small

percentage of xanthone.

Additional, but less likely mechanisms were also considered as shown in Scheme 50.

Initially the stable 2-methyl-3-(9-xanthyl)propene radical Sa may be formed from carbon

hydrogen benzylic cleavage. The benzylic hydrogen may be abstracted by cyanomethyl

radicals in the aqueous acidic acetonitrile solvent, analogous to the mechanism shown in

Scheme 46. If the 2-methyl-3-(9-xanthyl)propene radical Sa is formed, it could then be

oxidized, using oxygen as the oxidizing agent, to the 2-methyl-3-(9-xanthyl)propene cation

9a. This makes the assumption that the oxidation potential of the 2-methyl-3-(9- xanthyl)propene radical Sa is similar in value to the oxidation potential of the xanthyl radical

la. Under aqueous acidic conditions a tertiary benzylic alcohol 20 might be formed through acid catalyzed hydration of the 2-methyl-3-(9-xanthyl)propene cation 9a, with subsequent dehydration to form 2-methyl-3-(9-xanthylidene)propene 19 (two mass units less than the parent). 116

Scheme 50

In analogous photoproduct studies of the irradiation of aqueous acidic acetonitrile solutions of 9-xanthylacetophenone 2, as described in an earlier section (111.A.5.a.), a photoproduct with a mass unit of two less than the parent was also suggested (Scheme 46). 117

This photoproduct, 9-xanthylidene acetophenone 5, was isolated from preparative photolyses of9-xanthylacetophenone 2. Identification was confirmed through independent synthesis and

1H NMR and 13 C NMR analyses. Attempts to sythesize and isolate the tertiary benzylic alcohol precursor 10 of9-xanthylidene acetophenone 5 were unsuccessful and not attempted for 2-methyl-3-(9-xanthylidene)propene 19. CHAPTER IV

CONCLUSIONS

9-Xanthylacetone, 9-xanthylacetophenone and 2-methyl-3-(9-xanthyl)propene have been synthesized and their photochemistry studied. The xanthyl cation can be generated photochemically from the xanthyl ketones and the xanthyl alkene under less acidic conditions than required for the thermal reaction. This cation photogeneration is not subject to acid

catalysis over the acid range of 5-30% H2S04 in the case of the xanthyl ketones or 1-20%

H2S04 in the case ofthe xanthyl propene. In contrast, formation of the xanthyl cation in the ground-state reaction requires considerably stronger acidic media and shows a strong dependence on acid concentration.

The lack of acid catalysis in the photogeneration of the xanthyl cation, combined with the observation from laser flash photolysis experiments of the xanthyl ketones that the xanthyl radical is the primary photoproduct, suggest a different mechanism operating in the excited state. A two step mechanism is proposed for cation photogeneration from the xanthyl ketones and xanthyl alkene. The first step involves initial homolytic carbon-carbon bond cleavage to generate the xanthyl radical and the corresponding a-acyl radical. The isolation of the radical coupling product bixanthyl supports the intermediacy of the xanthyl radical in the mechanism. The second step of the proposed mechanism involves oxidation of the xanthyl radical to generate the xanthyl cation, as demonstrated by laser flash photolysis

118 119 studies showing clean conversion of the xanthyl radical to the xanthyl cation in the presence of oxygen. Redox potentials support the indication that oxygen is the oxidizing agent.

Carbonyl compounds readily undergo photochemically induced a-cleavage, but P-homolytic bond cleavage is rare. The mechanism proposed suggests a new photochemical route for carbocation formation involving P-homolytic carbon-carbon bond cleavage.

Preparative photolyses of the xanthyl ketones and xanthyl alkene were further examined. Irradiation of9-xanthylacetophenone also produced acetophenone as a photolysis product. Its fonnation was attributed to hydrogen abstraction by the precursor a-acyl radical.

In the case of 9-xanthylacetophenone, a new photoproduct, 9-xanthylidene acetophenone, was isolated and identified. An analogous photoproduct from 9-xanthylacetone and 2-methyl-

3-(9-xanthyl)propene was also suggested.

A mechanism for the formation of 9-xanthylidene acetophenone from 9- xanthylacetophenone involving initial formation of the xanthyl radical was proposed.

Irradiation of 9-xanthylacetophenone in aqueous acidic acetonitrile results in the formation of the xanthyl radical and the a-acyl radical through P-homolytic carbon-carbon bond cleavage. The xanthyl radical then abstracts either a methylene hydrogen or benzylic hydrogen from 9-xanthylacetophenone. A second hydrogen abstraction or radical-radical disproportionation, results in the stable 9-xanthylidene acetophenone.

Product quantum yields were determined for the formation of 9-xanthylidene acetophenone from 9-xanthylacetophenone, using valerophenone as a chemical actinometer.

The quantum yield for 9-xanthylidene acetophenone formation is in agreement with literature values for quantum yields for homolytic carbon-carbon benzylic cleavage and suggests an 120 efficient process. CHAPTER V

EXPERIMENTAL

A. Materials

Ethyl acetoacetate was obtained from Fisher and distilled prior to use. Methylene chloride, methanol and concentrated sulfuric acid were obtained from Fisher and used as received. Ethyl benzoylacetate (tech 90%), 9-hydroxyxanthene, trifluoroacetic acid (TFA), acetonitrile (reagent grade), diethyl ether, petroleum ether, benzene, lithium phenylacetylide, the prepared ylide of methyltriphenyl phosphonium bromide and deuterated chloroform were obtained from Aldrich and used as received. In some experiments acetonitrile and benzene were distilled over calcium hydride. Spectroscopic grade acetonitrile used in fluorescence measurements was obtained from Mallinckrodt. Merck silica gel, grade 60, 230-400 mesh, was used for flash column chromatography. All 1H NMR and 13C NMR spectra were taken

in CDC13, and all absorption spectra were taken in aqueous acidic solutions using acetonitrile as a cosolvent.

B. General Methods

NMR spectra were obtained on either a Varian VXR300 or a Varian VXR 4008 MHz instrument with 1H NMR chemical shifts reported in parts per million (ppm) from TMS

121 122

13 internal standard. C NMR chemical shifts(&) were reported using 77.0 ppm (CDC13) as a reference peak. UV and visible absorption spectra were measured on a Hewlett-Packard

8452A diode array spectrophotometer. Fluorescence measurements were obtained on a

Photon Technology International LS-100 spectrophotometer. Melting points were obtained on a Mel-Temp apparatus and are uncorrected. GC analyses were performed using a

Hewlett-Packard 5890 series II chromatograph with FID detector and fitted with an HP 1 capillary column (crosslinked methyl silicone gum) measuring 25 m x 0.2 mm x 0.33 µm film thickness. The gas chromatograph was equipped with a Hewlett-Packard 3396 series II integrator. Temperature programming starting at 120 °C for 6 min, increasing at 3 °C/min, and holding at 250 °C for 10 min was used. Injector and detector temperatures were held at

275 °C. Sulfuric acid densities were determined at 25 °C with an Anton-Paar DMA 48 digital precision density meter. Sulfuric acid solutions were prepared from distilled water and reagent grade acid. Solution densities were determined and converted to weight percent acid from literature tables. 136

C. General Photolysis Procedure

Stock solutions of 9-xanthylacetone 1, 9-xanthylacetophenone 2 and 2-methyl-3-(9- xanthyl)propene 14 were prepared at 4 x 104 Min acetonitrile. Solutions for irradiation were prepared by combining 5 mL of these stock solutions with acetonitrile (29 mL for 1, 66 mL

for 2 and 14) and aqueous sulfuric acid (5-300/o H2S04, 66mL for 1 and 29 mL for 2 and 14).

Solutions were purged with dry argon for 30 min using a fritted glass rod and then transferred to 20 mL or 200 mL quartz tubes. The solutions were irradiated at an excitation wavelength 123 of 254 nm or 300 nm using from 2-6 lamps in a Rayonet photoreactor. A merry-go-round device (in the case of20 mL cuvettes) provided equivalent irradiation to each solution. The total time of irradiation was varied, with aliquots being removed at one minute intervals.

Samples wrapped in foil served as dark controls. The solutions were either directly analyzed by UV/vis spectroscopy or the photoproducts were initially isolated through aqueous workup

(2 x 20 mL 1 M NaOH, 1 x 20 mL deionized H20), extraction (4 x 20 mL methylene

chloride), and drying of pooled organic layers over anhydrous MgSO4 • The solutions were then gravity filtered, and solvent removed in vacuo followed by GC analysis.

D. Triplet Sensitized Photolysis: General Method

Benzophenone was used as the triplet sensitizer in the photolysis of 1 and 2.

Solutions (100 mL total volume) were prepared containing benzophenone (4 x 10-4 M), 1, 2

5 or 14 (4 x 10- M), dodecane (5 mg), and aqueous acidic (5% H2S04) acetonitrile, with the same relative amounts of aqueous acid to acetonitrile as described in the Section V. C. The concentration of benzophenone was chosen in order that benzophenone absorb at least 90% of the incident light. Dodecane was added to each solution to serve as an internal standard for GC response factors. Solutions were transferred to 200 mL quartz cuvettes and photolyzed (as described above) for 10 min using two lamps of 254 nm wavelength. Dark reactions of the solutions were also tested. Direct irradiation control experiments of 1, 2 and

14, as well as benzophenone, were conducted. In order to isolate unreacted starting materials and possible photoproducts from the aqueous acidic solution, the photolyzed solutions were washed with 1 M NaOH, extracted using methylene chloride, and dried over anhydrous 124

MgSO 4. The solutions were gravity filtered and solvent removed in vacuo. GC analyses were

then performed as described above.

E. Triplet Quenching

The potassium salt of2,4-hexadienoic acid (potassium sorbate) was used as a triplet

3 quencher at a concentration of 10- M in aqueous acidic (5% H2S04) acetonitrile

(acid/acetonitrile 1:9 by volume). 1, 2 or 14 was added to the stirred solution to give a

concentration of 10-3 M, and the solutions were purged with dry argon for 30 min. UV/vis spectra of the solutions were recorded with and without addition of quencher, before and after purging. The solutions were then photolyzed for 0, 4, or 8 min using two lamps of 300 nm wavelength. The irradiated solutions were analyzed by UV/vis spectra following the addition of sufficient sulfuric acid to shift the xanthyl cation/xanthydrol equilibrium completely to the cation side.

Oxygen was also used as a triplet quencher of 1, 2 or 14 in aqueous acidic (5%

H2S04) acetonitrile (aqueous/acetonitrile 2: 1 for 1 and 1:2 for 2 and 14). Solutions of 1, 2 or 14 (2 x 10-5 M) were purged using a flitted rod with either dry argon or oxygen for 30 min.

These solutions were then irradiated for 0 or 4 min using two lamps of 254 nm. Sulfuric acid was added to the irradiated solutions prior to analysis by UV/vis spectroscopy, as described above.

F. Laser Flash Photolysis

These experiments were performed using a Questek 2120 excimer laser as the 125

excitation source. All of the spectra and kinetic runs reported here were done with XeCl

reagent gas. The XeCl excimer laser provides UV pulses at 308 nm with a duration of 6-10

ns and a pulse energy of30-50 ml Transient UV/vis absorption signals were monitored using

a CW 500 W Xe lamp beam which was passed through the sample perpendicular to the excitation beam. Single wavelength transient waveforms were digitized using a LeCroy 9420

350 MHz digital oscilloscope and transferred to an IBM PS 2/286 computer for storage and spectral analysis.

Sample solutions were placed in an all glass flow cell137 which was sealed with a rubber serum cap and purged with nitrogen (or oxygen where noted) for I 0-15 min prior to the experiment. Sample concentrations were adjusted such that their optical densities were

I. 0-1. 5 at the excitation wavelength.

G. Syntheses

95 96 1. Ethyl 2-(9-Xanthyl)-3-oxobutanoate 3 •

Ethyl acetoacetate (1.54 mL, 12.1 mmol) was alk:ylated with 9-hydroxyxanthene (2.0 g, 1O .1 mmol) using TF A (3 drops) as a catalyst. The reaction mixture was heated at reflux

(90°C-100°C) for 2 h. The crude P-keto ester was recrystallized from 95% ethanol giving a white solid: mp 86-87 °C (lit. 95 mp 87-88 °C); 1H NMR 0 7.31 (m, SH), 4.89 (d, IH), 4.05

(q, 2H), 3.83 (d, lH), 1.95 (s, 3H), 1.13 (t, 3H).

95 96 2. 9-Xanthylacetone 1 '

P-Keto ester 3 (0.20g,0.65 mmol) was hydrolyzed using 5% NaOH (5.2 mL) by I26 heating at reflux (96°C-I00°C) for 5 h. Acidification (5x3 mL IM HCl, acidic by pH paper) followed by recrystalliz.ation from 95% EtOH gave a white solid: mp 93-94 °C (lit. 95 mp 93-

94 °C); 1HNMR l> 7.24 (m, 8H), 4.70 (t, IH),2.87 (d, 2H),2.04 (s, 3H); 13C NMR l> 206.24

(q), I52. I2 (q), I28.52, I27.60, I25. I9, I23.4I, I I6.53, 54.43, 34.54, 31. I6.

95 96 3. Ethyl 2-(9-xanthyl)-3-phenyloxopropanoate 4 •

Ethyl benzoylacetate (2.08 mL, I2. I mmol), 9-hydroxyxanthene (2.0 g, IO. I mmol) and TFA (2 drops) were heated at reflux (I 00 ° C) for 5 h. The crude P-keto ester was

purified using silica gel column chromatography (CH2Cl2 eluant). Recrystallization from 95%

EtOH gave a white solid: mp 74-75 °C (lit.95 mp 81-85 °C); 1HNMR l> 7.73 (d, 2H), 7.24

(m, l lH), 5.05 (d, IH), 4.60 (d, lH), 3.91 (q, 2H), 0.99 (t, 3H).

95 96 4. 9-Xanthylacetophenone 2 •

P-Keto ester 4 (0.25 g, 0.67 mmol) was hydrolyzed using 5% NaOH (5.4 mL) by heating at reflux (102°C) for 5 h. Acidification (5x2 mL IM HCl or until acidic by pH paper) and extraction with methylene chloride, followed by recrystalliz.ation from 95% EtOH, yielded a white solid: mp 82-85 °C (lit. 95 mp 82-85 °C); 1H NMR l> 7.26 (m, 13H), 4.85 (t, lH), 3.35

(d, 2H); 13C NMR l> I97.94 (q), I52.36 (q), 133.14, 128.84, I28.54, 128. I l, I27.88, 125.55,

I23.49, 116.58, 49.74, 34.70.

95 109 5. 2-Methyl-3-(9-xanthyl)propene 14 •

A commercial (Aldrich) mixture of the prepared ylide of methyltriphenyl phosphonium 127 bromide and sodium amide (0.301 g, 0.72 mmol where 1 g of the mixture - 2.4 mmol of methyltriphenylphosphonium bromide) was heated at 60°C for 1 hour in 10-15 mL of dry distilled benzene under dry argon. A solution of9-xanthylacetone (85 mg, 3.57 mmol) in dry benzene (3 mL) was then added dropwise over 15 min to the stirred solution. The solution was then heated to reflux (80°C). Aliquots (-25 µL) of the reaction mixture were removed

at -1 hr intervals, quenched with ice/H20 and extracted with CH2Cl2. Reaction progress was

followed by TLC (CH2Cl2 eluant, silica plates) and was complete after 5-6 hrs. The solution

was then cooled to RT, quenched with ice/H20, washed with aqueous saturated NH 4 Cl followed by water, the organic layer collected, dried over anhydrous MgS04 and gravity filtered. Solvent was removed by rotary evaporation. The resulting bright yellow solid was

95 1 recrystallized from ethanol to yield white needles (mp: 78-79°C, lit : 80-81°C). H NMR a

7.23 (m,8H), 4.72 (s,lH), 4.33 (s,lH), 4.10 (t,lH), 2.38 (d,2H), 1.67 (s,3H). 13C NMR a

152.17 (q), 141.90 (q), 128.79 (q), 127.49, 125.40, 116.27, 114.28, 49.52, 38.32, 22.77.

6. 9-Phenylethynylxanthenol 6101

Xanthone (0.8240 g, 5 mmol) was reacted with lithium phenylacetylide (10 mL of

1. OM solution in THF, 1Ommol) in THF (25mL) at 0 °C over 4h. Reaction progress was

followed by TLC (CH2Cl2 eluant, silica plates), until all xanthone had reacted. The solution

was neutralized through dropwise addition of 0.01 M NH4Cl (25 mL) over 30 min. The

crude light brown oil was extracted with diethyl ether, dried over anhydrous MgS04, gravity filtered, and the solvent removed by rotary evaporation. The resulting light brown solid was

101 recrystallized from petroleum ether to give a white solid (mp: 122-123 °C, lit : 12 l-122°C). 128

1H NMR 08.05 (d,2H), 7.37 (m,13H), 2.75 (s,lH).

7. 9-Xanthylidene acetophenone 5101

9-Phenylethynylxanthenol 6 (0.1088g, 0.365 mmol) was dissolved in a methanol

(20mL):H20(2mL) mixture. Concentrated HCl (2 drops) was added and the solution was

allowed to stir for 30 min at RT. The solution was then heated at 50 °C for 1 h followed by

heating at reflux temperature for 2 h. The reaction progress was followed by TLC ( CH2Cl2

eluant, silica plates), until all 9-phenylethynylxanthenol had reacted. After cooling to RT, the

resulting yellow solution was washed with 1.0M aqueous NaHC03 and extracted with diethyl

ether. The yellow-orange organic layer was dried with MgS04, followed by filtration and

solvent evaporation to yield a yellow oil. The crude oil was purified by silica gel

101 chromatography (CH2Cl2 eluant) to yield a light brown solid (mp: 121-123 °C, lit mp:119-

123 °C). 1HNMRo 7.44 (m,14H); 13C NMR o 194.32 (q), 152.46 (q), 151.66 (q), 138.64,

137.71, 133.12, 131.51, 130.85, 129.92, 129.05, 128.91, 128.77, 124.39, 124.07, 123.05,

122.96, 119.66, 117.66, 117.57, 117.21, 117.09; MS (EI, 70eV) m/z (relative intensity) 298

(M+·, 41), 297(100), 269(12), 221(6_7), 165(50), 134(27), 77(48), 51(25). FT-IR (CDC13):

1 1653 (C=O) cm· .

8. Attempted Synthesis of Tertiary Benzylic Alcohol to: 9-Hydroxy-( 9-xanthyl)acetophenone

a. Method 1

In one procedure we attempted to hydrate 5 in aqueous acidic solution. Initially 9- xanthylidene acetophenone 5 (23 mg, 7. 7 x 10-5 mol) was placed in an acetonitrile-deionized I29 water (7mL:3mL) solution. Aqueous sulfuric acid was added dropwise via pipette to the solution to give a final acid concentration of 0. I 0- I 4. 5 wt % and allowed to stir from I - I 0 days at RT. The reaction was followed by TLC using methylene chloride as an eluant. Some solutions were initially heated to 60-80°C and allowed to stir I-2 days. The concentration of starting materials was also varied from 7.7 x I0"5 mol to I.5 x I04 mol. Unreacted starting materials and possible products were isolated by washing the reaction mixture with IM

NaOH, followed by extraction with methylene chloride, and the organic layers were dried

over anhydrous MgSO4 . The solutions were gravity filtered and solvent removed in vacuo.

GC analyses were then performed (area % uncorrected for response factors) in order to identify the resulting products and remaining starting material.

b. Method 2

Aqueous solutions of xanthylacetophenone usmg acetonitrile as a cosolvent

(H20:acetonitrile 1:2) were irradiated in a Rayonet Photoreactor (254 nm, 2 lamps) for 0-25 minutes. The concentration ofxanthylacetophenone was varied from Ix I04 M to 2 x I04 M to 5 x I 04 M. The neutral solutions were initially irradiated using deionized water as the aqueous portion. Later experiments included boiling the deionized water and base washing all glassware used in the photolyses and workup in order to ensure no acid present to give a neutral solution (pH 7). Unreacted starting material and possible photoproducts were isolated in most cases by washing the reaction mixtures with IM NaOH, extraction using methylene

chloride and drying the organic layers over anhydrous MgSO4 . In some cases, washing with deionized water was substitututed for washing with IM NaOH. All solutions were gravity 130

filtered and solvent removed in vacuo. GC analyses were then performed; the resulting

products and starting material were identified.

9. Product Quantum Yield

Product quantum yields for the photolysis of 2 were determined. Quantum yield or quantum efficiency can be defined as the number of molecules of product formed per photon absorbed. Initially, UV/vis spectra were measured for both valerophenone in cyclohexane and

9-xanthylacetophenone in 5% aqueous H2SO/acetonitrile solutions. Concentrations for valerophenone and 2 were chosen such that the absorbances measured at 254 run were greater than 2 (a concentration high enough to isolate all photoproducts and to ensure complete absorbance of all incident light)138 and roughly equivalent (2.5 for valerophenone and 2.7 for

2). Solutions ofvalerophenone in cyclohexane were irradiated for 1-4 min using 2 Rayonet lamps (254 run). Direct GC analyses of the solutions showed 15% conversion of valerophenone to acetophenone after 3 minutes of irradiation. In order to eliminate the possibility of secondary photoreaction, percent conversion of reactant was limited to less than

200/o. Using dodecane as an internal standard, GC response factors were determined in order to account for variations in detector response to the various analytes (starting materials and photoproducts).

Product quantum yields were determined by initially identifying each component of the photoproduct mix, by comparison of previously determined GC retention times of known compounds. Area counts for each component were then adjusted using an average ( 4 or more values) response factor calculated for each component and the internal standard 131

dodecane. Response factors were determined using known concentrations of each previously

identified component and the internal standard dodecane, using ratios of mol/L/count compound to mo1/Ucount dodecane. Adjusted total area counts were then used to determine an adjusted area percent for each component. These adjusted area percentages, coupled with the known product quantum yield of0.40 for acetophenone formation from valerophenone

(actinometer), gave the unknown product quantum yield.

Aqueous acidic acetonitrile solutions of 2 were irradiated for 1-5 minutes using 2

Rayonet lamps (254 run). A solution ofvalerophenone in cyclohexane was irradiated at the same time with the solution containing 2. The irradiated valerophenone solution was analyzed directly by GC, as descnbed above. Irradiated solutions of 2 were first washed with aqueous base and extracted with methylene chloride. The solution was then examined by GC to determine percent conversion to photoproducts. After 3 min irradiation, 9-xanthylidene acetophenone (34% ± 3%) was the sole photoproduct, according to GC analysis. Under varying conditions (concentrations of starting material, irradiation times, and choices of solvent) both xanthone and bixanthyl were also identified as photoproducts. Several limiting factors were noted: I) equivalent irradiation times (2 and valerophenone actinometer were irradiated in parallel and thus required equivalent irradiation times), 2) similar absorbance values at 254 run,and 3) concentrations high enough to isolate and identify photoproducts upon GC analyses. Thus, in the case of the irradiation of 2 it was necessary to have a conversion greater than 20% (34%). We did not see additional photoproducts upon irradiation of 2 (only 9-xanthylidene acetophenone 5); we made the assumption that at the low concentrations utilized, absorbance of the resulting photoproduct 5 was not significant. 132

9-Xanthylidene acetophenone 5 was also irradiated under similar concentrations and times, with no photoproducts noted. Increasing percent conversion of reactant to photoproducts in the case of valerophenone (> 15%) resulted in the additional unidentified photoproducts, possibly cyclobutanols as observed earlier by Wagner. 106 VI. SPECTRA

133 134

,.. ;..

;... r- ..... L L

,..~ -:-

~ - -"'- .sc: u"' :- >-ftS ~ ftS )( -L... a, 0 -a:= z~ - J: 135

z a. a.

-... I i 0 « ::!:z ..,-0 136

-~ cu gc: cu

-a0 i) I ~ -0 0:: ~z .....J: 137

-N -G) c i 11 Iu 0 a:: :Ez 0 ...fl) I ..... -r-r;r' I I I I .-.-~ ...... -.-.- ..... r· I I .'~~~'!'*#WMl*1 ll(·~·.rnll'"1-..::- ..... 0 II CCH3 I H CHC02CH2CH3 '-=:: 1H NMR of Ethyl 2-(9-xanthyl)-3-oxobutanoate, (3) o5b0 -!,JJ 00 139

-.... ~m 0 c: m Q. ....0 Q. ~ 0 ~

f...,... N ~ 'O a:: z~ ....J: 140

r! rll ~y .. O=O 'o -.r:.'

-IO .,~

J• f 0 a:: :Ez -J:

;- !- ....

,.. ~ I ., ..,. 141

o:u9 )J J:

'°- -0 c 0c a0 s0 •u ! 0 "O { )(• 0, 0 -a: :E z tk'.... 142

j I 5?_p ! ~.013 ~ Ii I. irj i! -la s -.,, i gc G ~ Q. 0 'G u ca cG .,,G

I I Ica ~ • I -.: • I -0 :ea: z

·wzs· -= . I i : ~

..- ..: • =J

: ~ : i ~ - ..! - - 143

f:_! 0 ----- 'F yf ~ .. Ill ~ 0 I !..

L. I

... I ,..i ,_~ j- L. t I 144

.l .., :x: ~ ,.. 0 :x: I O=O I,.. fl :x: • 0

-• -oi c: u Q.. 0 ...Q.. ->. ~ca :...,.. C")

2I N 0 -a: 2z • -:x:

• 145

-·I

rr r 1 L...... I... f ::- .

;.. ., -~ . -... - ; .. . . . 0 •- ...... -. . •I :.! .. • ... - ... . r~ -0 - . - a: t :E . z ~ L.. - -

--: l.....- . -. ~ .-: - . . - . -i i!. "' ~ 149

140000 91

120000

100000

80000

60000

40000 65 104 123 206 20000 56

1 1 1 0 1 i lf•tp I 'jl; ...... I 1 .. I I,,, I I, ... I •• I I Ill I I #a I 1•1 I I I I I ,u .. I I I I I I 1 1 I I I I I I I I I I I I I I I I I I I I z-->

GC/MS (El) of Photoproduct lft 40 -~ undance 5.7 129

60000·

50000... 5

40000

30000

112 20000 83 147 10000 101 9 0 II I ...,~iY!m,,.,~t...ra. I'··~ IJ, I l'~~~ f I I I I I' I I I ~~ • mI .,,;_g I I~~!.. z--> 60 80 100 1 0 1 0 160 1 0 200 220 240 260

GC/MS (El) of Photoproduct tR 43 -~ -...l unaance l.t9 25000

20000

15000 I 55

10000 ~ I 167 I I 5000 61 83

0 - - I - - - - I - - - - I - - . - I - - - I - I - - ' z--> 60 80 100 120 140 160 180 200' 220 240

GC/MS (El) of Photoproduct tR 44 -~ 00 undance I 5 180

160

140 219 120

100

80 181 60 109

40 233

20 71 94 I I I '~1.L 0 ' ' ' o, •• "' ,' " "I' .' •' .. ' " ,o ' ' ' ' "I '• ' ' ~; ~ "'"• ~ ci ~' ' ;~ z--> CH2 II H, ,.C-CH2 -C-CH3 ~ GC/MS (El) of 9-Xanthytldene-2-methytpropene UaV .i:i. '° a nee 149 500·1 57

400

300

200 -f II 71 167

100 83 104.13

1 1 11 9 0 \ I 1• f' I 1 \"'t p I "11 I 11 11 I p I 11 I 1 I I I I I I 1 I I I I 11 I I I I I o I I I I I I I I I I I I I I I I I I I I I 11 I I z--> 60 80 100 120 140 160 180 200 220 240 260 280

GC/MS (El) of Photoproduct lft 27 VI 0 I lf 9 45000

40000

35000

30000

25000

20000

15000 167 10000 I 55 500:w

I z--> 60 80 180 200 220 240 260

GC/MS (El) of Photoproduct lft 42 -U\ APPENDIX

IDENTIFICATION OF PHOTOPRODUCT TR 42 OBTAINED FROM

2-METHYL-3-(9-XANTHYL)PROPENE 14

152 153

Earlier results (Section II.B.2.) described the fonnation of five photoproducts from

the irradiation of aqueous acidic acetonitrile solutions of2-methyl-3-(9-xanthyl)propene 14

(ta=24), as shown in the scheme below. These were identified through aqueous workup and

product isolation followed by GC/MS analysis as xanthone (tR=23), bixanthyl (~=52), 9-

xanthylidene 2-methylpropene (tR=29), 2-methyl-1-(9-xanthyl)hexane (~ =27) and an

unidentified peak with a retention time of 4 2 min.

Scheme 51

~~ ~cc~ hv U0~ CH3CN/ H2o,H+ 14

The following data for tR 42 photoproduct, a white crystalline product, was

obtained: 1H NMR. o 7.63-7. 71 (m), 7.52-7.59 (m), and 7.43-7.50 (m); integration ratio:

93:57:97. 13C NMR (75 MHz) o 133.11 (q), 132.02, 131.89, 131.80, 131.76, 128.43,

128.27. 13C NMR (100 MHz) o 133.06 {q), 132.12, 132.02, 131.87, 131.83, 128.52,

128.36; Ff/JR 3057.10, 2363.33, 2335.02, 1601.10, 1483.95, 1437.18, 1192. 79, 1119.47,

1 720.28, 693.64 cm· . GC/MS EI 70 eV: base peak 277 (100), 278 (38), 279 (15), possible parent peak 280 (<1); MALDI-TOF: possible parent peak 280. 154

Initial GC/MS EI 70 eV analysis indicated a highest rn/e peak of 257. 9-

Phenylxanthene 21 or 9-phenylxanthen-9-ol 22 was initially proposed as the possible

structure for ta 42, but eliminated by comparison of 1H NMR and subsequent GC/MS data

which showed a 277 base peak (Figure 17). Dimerization or radical-radical recombination

of photoproducts of 14 upon photolysis was then considered as a possible route to the

formation of tR 42. Several structures, including 23-26 (Figure 18) were proposed and 13C

NMR shifts calculated utilizing the ACD software program. 139 Structure 27 (Figure 18) was

also considered. Upon examination ofFr/IR. data ofta 42 (absence ofC=O stretch and OH

band), 1H NMR. (no OH) and 13 C NMR. (especially absence ofC=O, and the absence ofa

peak at -155 corresponding to quaternary carbons of the C-0-C xanthyl backbone), it was concluded that the structure of tR 42 did not contain an oxygen. Several structures, 28-35140

(Figure 19), were then proposed for ta 42 utilizing 13C NMR shifts as reported in the Aldrich

Library of Spectra. This was based on the absence of oxygen and nitrogen (absence of nitrile stretch in FT/IR as shown in Figure 20, absence of 13C peak at 115-125). Nitrogen would be present if acetonitrile solvent had reacted with starting material 14 upon photolysis.

The 13C NMR data (Figures 21 and 22) suggested that the compound contained six monoprotonated sp2 carbons and one quaternary carbon. There are at least seven different carbons, and it has been established that there are more than seven total carbons from

GC/MS data. There appears to be no quaternary carbons next to the oxygen of the xanthyl backbone, suggesting a backbone similar to tluorene. The short acquisition time used for the 13C NMR data may have saturated quaternary carbons that are present; thus, the lack of Figure 17

GC/MS (EI) ofXMP photoproduct tR.

155 Abundance Average of 42.430 to 43.436 min.: MFC42B.D 277 40000

35000

30000

25000

20000

15000 77 183 199 10000 I511 152 5000· 7 95 0 lm/z--> 60 80 100 120 140 160 180 200 220 240 260 280

..... V\ °' Figure 18

Possible structures considered for XMP photoproduct tR 42.

157 158

mw258 mwZ74

21 22

1H NMR 5 5.28 (s, 1H), 7.00-7.30 (m, 13H)

~... c H C'H mw386 o¢o 0 24 25

13C NMR: 156.18 13C NMR: 155.44 -No peri-Hs 137.16 139.51 -No C=O 129.44 136.94 -No C (155) 125.05 136.44 123.43 130.51 121.72 129.15 125.80 125.12 121.38 119.85

mw394 mw378

13C NMR: 151.30 128.81 128.32 124.19 121.06 114.30 74.28 Figure 19

Possible structures considered for XMP photoproduct ta 42.

159 160

H 0 c=c' mw 180 00 ,C=C, mw 180 H H H' b trans-Stilbene (28) cis-Stilbene (29) 1 , 1-0iphenytethytene (30)

13C NMR: 137.26 13C NMR: 137.16 13C NMR: 149.95 128.63 130.17 141.38 128.58 128.79 128.18 127.52 126.99 127.62 126.44 114.23

H () 'c=~ 00C=C Vb mw256 Ob mw332

Triphenylethytene (31) Tetraphenylethytene (32)

13c NMR: 143.34 128.10 13C NMR: 143.61 142.51 127.86 140.85 140.29 127.52 131.22 137.30 127.40 127.54 129.46 127.30 126.30 128.53 126.65

0-0' mw178

bis-Auorenylidene (33) Phenanthrene (34) Ftuoranthrene (35)

13C NMR: 140.51 13C NMR: 131.9 13C NMR: 139.38 139.43 130.1 136.90 131.35 128.3 132.34 129.01 126.6 129.93 126.03 126.3 127.83 125.52 126.3 127.43 119.12 122.2 126.53, 121.42 119.92 Figure 20

FT/IR ofXMP photoproduct tR 42.

161 162

8 IO

0 0 ....0

0c ....IO

0 co A 0 CD ... 0 .J:l- 8 N E :J c: CD ~(EtS% > 0 as 0 Ill ~ N

0 0 0 (")

0 0 0 ~ 0 0 C'i "#. 1--mccoE---mcoCD Figure 21

13C NMR (75 MHz) ofXMP photoproduct tR 42.

163 164

0 Figure 22

13C NMR. (100 MHz) ofXlv.IP photoproduct ta 42.

165 166

:::::::1.------.J.....u.-::r- 167 quaternary carbons could not be conclusively established, but is unlikely. COSY analysis will be performed to determine connectivities and resolve peaks.

The 1H NMR data showed the total number of protons to be 10-12. Results ofboth

300 MHz and 400 MHz 1H NMR data (Figure 23) are complex and may not be considered to show first order splitting. One analysis of the 300 MHz data (Figures 24 and 25) shows for the first range o 7.63-7. 71 a doublet of doublets and a doublet of triplets with coupling constants J 12 Hz and 1.5 Hz. This analysis holds true for both 300 :MHz and 400 :MHz data. There appears to be no simple solution to the ranges o 7.52-7.59 and 7.43-7.50.

There is no sufficient explanation for these last two ranges. There may be a mixture of two similar isomers or a structure that we have not yet been able to identify. A 12 Hz coupling constant sounds large, but the alternative is no ortho coupling at all (a typical ortho coupling constant range would be 6-10 Hz). Later EI and MALDI-TOF data suggested the parent peak to be 280 or above. The molecular weight ofthe parent peak may be even higher than

280 it: for example, the tR 42 photoproduct is photoionizing within the MS instrument.

GC/MS High resolution EI spectra were obtained (Figures 26-27), with possible parent

peaks of 277, 278, 279 and 280 giving the following molecular formulas: C7H 170 11 ,

C7H 110 1., C 14H 15 0 6 and C 14 H 16 0 6 (Figure 28) respectively. (Mass spectrometry was provided by the Washington University Mass Spectrometry Resource, an NIH Research

Resource, Grant No. P41RR0954). Figure 23

1H NMR (400 MHz) ofXMP photoproduct ta 42.

168 169 ..l

•...

,..... Figure 24

1H NMR (300 MHz) ofXMP photoproduct tR 42.

170 171

-·--..., .. _ .a..:::::::'.- "·--

0 I ;.:. 5 ;0 I i at·-=',.. __ ., ~ l I i- I I : ,. - a .. ! ~ ~

I .-i I I .. ..: ! I .. oi

,.. ! ! - :

~ • I ~ .. i' ..,...... : Figure 25

1H NMR (300 MHz) RESOLV ofXMP photoproduct tR 42.

172 173

M.lllD_.

Ll"nllB- ... .,._ •1·au- •·oa-··--'-=

··--a·jiiji-

··1a1- ··ZLD-

·-•·11u-

··---- •·na- Figure 26

GC/MS High Resolution EI ofXMP photoproduct tR 42.

174 ent: :l:l_,n W1n lOOPIPM Acq: ZAB-E EI+ Magnet BpM: 69 Bpi: 30:31363 TICi 33167109 Plags :ACC Pile Te~c:Clifton/Bayd, sample id I FR42-Crystal; HIGH RESOLUTION EI1 POSITIVE MODE 100, 277.f786 1.es5; 90 1.6851

80 1. 485· 70 l.28Si 278. 0:842 60 1.1851

50 8.9E4.

7 .184.

5.384. 3.684, 279.0890 280.9834 1.884. 274.9942 276.01626 I I I .... v.,...... -.- I 281.98419 282.991'7? O.OEO: 21s. 216 211· 219. 21~ 2ao1 2a1 2a2 283 mt-z Pile:396'1 Ident:22_3il Win lOOPPM FILT(BXCL,ALL) Acq:l:S-JUN-1997' 18:34:58 +4.:23 Cal:SERCAL062397G: ZAB-B EI+ Magnet BpM: 277 Bpi: 17?995 TIC: 3;);67109 Plagis::ACC File Text1::Clifton/Boyd, sample id I FR42-Crystal; HIGH RESOLUTION EI; POSITIVE MODB 100~ 217.-0786 l .8E5: l.6ES

l.4ES 1. 2ES:

l. lES: 8.9E41

7.1E41

30. .5. 3E41 77 .. ()401 20 183 .0154 3.6E41

.. CI009 l.8E41 l0~13 1 0 5 1 1~· ~ 11h 11 Ill ~,---i:-- - 2l3.99"II·i----111 3.J.6.9812 368.9798' 418.9797 O.OEO. o_ Lili. djlih.. 1.. 1 11~1,""1,"~"'i'Y I! 1,"li • tl•,n,l.t 11·~.LJ lp1• gii~•,U~ ~1llL1,L ~I, 1,i l"11t,l.l ,1 111, 'I·,•, 1, 'r, I,"' i.,u., ,"·, •J·+-.. 60 90 l o 1 o 1 o 160 1ao 2 o 220 2!0 2$0 2Ao 30o· 3~0 340• 360 3Ao1 400 42:0 ' 0 m/.z ------·-·------' I -...I -v. Figure 27

GC/MS High Resolution EI ofXMP photoproduct tR 42.

176 ··--- .. in ~vv...-c.j,·~ ..,.,.....,~ ,.,..~'-....,., z:AB-E EI+ Magnet BpM:27.7 Bpl:201436 TIC:3218409 F]ags:ACC flile Text::Cliftoru/Boyd,, sarnp:le id # FR42-Crystal; HIGH RESOLUTION EI;: POSITIVE MODE ]00i. 277 .. 0786; 2.0E5

90 1.8E5

1.6E5

1.4E5

278.0845 1.2E5 1. OES

8.1E4

6.0E4

279.0898 4. OE4 2.0E4 280.0958 274.9917 276. 0644, 281. 98!83 O.OEO 2~4 275 276 277 278 279 280 281 282 m/z flile:396l Ident:l.7_25 Win lOIOPPM FILT(EXCL,ALL) Acq:25-JON-1997 18:34:58 +3:30 Cal:SERCAL062397G ZAB-E EI+ Magnet BpM:277 BpI:201436 TIC:32184091 Flags:ACC File Text:Clifton/Boyd,, samp:le id I FR42-Crystail; HIGH RESOLUTI

90 1.8E5

BO 1.6E5

70 1.4ES

60 1.2E5

50~ 1. OES

40 8.1E4

30 6.0E4 20 ~ 1B3.03S7 4. OE4 10 2 .OE4 O.OEO 0 m/z ----j ...... -....J -....J Figure 28

GC/MS High Resolution EI Peak matching results.

178 179 ---

.00... "° ...\GO ..

.... II)......

...NU ...... NU .....

• l "°• • D• Do I'- ID•""I· N I .ii .a u u &. 0 0 0 • 11 at 0 . . • H 0 •- H 0 • 0 11~ 0 • ... • .. I 0 I 0 • • ..0 • • .. ...• • •0 ...• • •"'0 I . r :c• ..r .... i DI E rn N

.00...... 00 ...... •.. ...NU ...... NU "" ... l .. • ID• Do IO• ID"'I·• N I .a .au u G G II 11 ... 0 II ...... 0 • H 0 • H. 0 • 0 • G • .. • .. 0 0 ! I N • •...... • .. 0 •0 ii... r;..... ii... •.... • N rn N REFERENCES

(I) McManus, S. P. Organic Reactive Intermediates, Academic Press: New York, 1973.

(2) Sefcik, M. D.; Henis, J.M. S.; Gaspar, P. P. J. Chem. Phys. 1974, 61, 4321.

(3) Traynham, J. G. J. Chem. Educ. 1986, 63, 930.

(4) Kennedy, J.P.; Marechal, E. Carbocationic Polymerization, Wiley: New York, 1982.

(5) Vogel, P. Carbocation Chemistry, Elsevier: Amsterdam, 1985.

(6) Olah, G. A.; Schleyer, P. von R. Carbonium Jons, Wiley: New York, 1970; Vols. I V.

(7) Allen, N. S. Trends in Polymer Science 1993, 1, No. 7, 213.

(8) Verbanic, C. Chemical Business 1993, 15, No. 3, 13.

(9) Pappas, S. P. CRC Handbook of Organic Photochemistry, CRC Press: Boca Raton, 1985; Vol. II.

(IO) Lissi, E. A.; Encinas, M. V. CRC Photochemistry and Photophysics, CRC Press: Boca Raton, 1991; Vol. IV.

(11) Rabek, J. F. Mechanisms ofPhotophysical Processes and Photochemical Reactions in Polymers, Wiley: New York, 1987.

(12) Timpe, H.J. Pure Appl. Chem. 1988, 60, 1033.

( 13) Phillips, R. Sources and Applications of Ultraviolet Radiation, Academic Press: New York, 1983.

(14) Kelly, J.M.; McArdle, C. B.; de F. Maunder, M. J. Photochemistry and Polymeric Systems, Royal Society of Chemistry: Cambridge, 1993.

(15) Kaufinann, H.F. Makromol. Chem. 1979, 180, 2649.

180 181

(16) Tada, K.; Shirota, Y; Mikawa, H.J. Polym. Sci. Al 1973, 11, 2961.

(17) Suzuki, M.; Yamamoto, Y.; Irie, M.; Haysahi, K. J. Macromol. Sci. Chem. 1976, 10, 1607.

(18) Williamson, M.A.; Smith, J. D. B.; Castle, P. M.; Kaufman, R. N. J. Polym. Sci. Al 1982, 20, 1875.

(19) Wippelder, E.; Heusinger, H.J. Polym. Sci. Al 1978, 16, 1779.

(20) Ledwith, A.; Bosley, J. A.; Purbick, M. J. Oil Colour Chem. Assoc. 1978, 61, 95.

(21) Cha, Y. S.; Tsunooka, M.; Tanaka, M. J. Polym. Sci. Al 1984, 22, 2973.

(22) Petterson, R. C.; Dimaggio, A.; Herberts, A. L.; Haley, T. J.; Mykytka, J.P.; Sarkar, I. M. J. Org. Chem. 1971, 36, 631.

(23) Ko, M.; Sato, T.; Otsu, T. J. Polym. Sci. Al 1974, 12, 2943.

(24) Pappas, S. P. UV Curing: Science and Technology, Stanford Technology Marketing Corp.: Norwalk, CT., 1978, Vol. 1.

(25) Pappas, S. P.; Pappas, B. C.; Gatechair, L. R.; Schnabel, W. J. Polym. Sci. Al 1984, 22, 69.

(26) Nagaro, I.; Nishihara, K.; Sakota, N. J. Polym. Sci. Al 1974, 12, 785.

(27) Crivello, J. V. Polym. Eng. Sci. 1983, 23, 953.

(28) Saegusa, T. Topics in Cu"ent Chemistry, Springer-Verlag: New York, 19, Vol. 100.

(29) Allen, N. S. J. Photochem. Photobiol. A: Chemistry, 1996, 100, 101.

(30) Heine, H.-G.; Rosenkranz, H.J.; Rudolph, H. Angew. Chem. Int. Ed 1972, 11, 974.

(31) Davis, M. J.; Doherty, J.; Godfrey, A. A.; Green, P. N.; Young, J. R. A.; Parrish, M. A. J. Oil Colour Chem. Ass. 1978, 61, 256.

(32) Allen, N. S.; Mallon, D.; Timms, A. W.; Green, W. A.; Catalina, F.; Corrales, T.; Navaratnam, S.; Parson, B. J. J. Chem. Soc., Faraday Trans. 1993, 90, 83.

(33) Tazuke, S. High Polym. 1977, 29, 681. 182

(34) Crivello, J. V. Adv. Polym. Sci. 1984, 62, 1.

(35) Arnett, E. M.; Amamath, K.; Harvey, N. G.; Cheng, J.-P. J. Am. Chem. Soc. 1990, 112, 344.

(36) Boyd, M. K. ''Photochemistry and Photophysics ofCarbocations,, in Molecular and Supramolecular Photochemistry, Eds. V. Ramamurthy and K. Schanze, New York: Marcel Dekker, Inc., 1997, 147-186.

(37) McClelland, R. A. Tetrahedron 1996, 52, 6823.

(38) Das, P. K. Chem. Rev. 1993, 93, 119.

(39) Bartl, J.; Steenken, S.; Mayr, H.; McClelland, R. A. J. Am. Chem. Soc. 1990, 112, 6918.

(40) McClelland, R. A.; Kanagasabapathy, V. M.; Banait, N. S; Steenken, S. J. Am. Chem. Soc. 1989, 111, 3966.

(41) Mathivanan, N.; McClelland, R. A.; Steenken, S. J. Am. Chem. Soc. 1990, 112, 8454.

(42) McClelland, R. A.; Kanagasabapathy, V. M.; Steenken, S. J. Am. Chem. Soc. 1988, 110, 6913.

(43) Chateauneuf, J. J. Chem. Soc., Chem. Comm.1991, 1437.

(44) Minto, R. E.; Das, P.K. J. Am. Chem. Soc. 1989, 111, 8858.

(45) McClelland, R. A.; Banait, N.; Steenken, S. J. Am. Chem. Soc. 1989, 111, 2929.

(46) Krogh, E.; Wan, P. Tetrahedron Lett. 1986, 823.

(47) Krogh, E.; Wan, P. Can. J. Chem. 1990, 68, 1725.

(48) McClelland, R. A.; Mathivanan, N.; Steenken, S. J. Am. Chem. Soc. 1990, 112, 4857.

(49) Pienta, N. J.; Kessler, R. J.; Peters, K. S.; O'Driscoll, E. D.; Arnett, E.M.; Molter, K. E. J. Am. Chem. Soc. 1991, 113, 3773.

(50) Wan, P.; Muralidharan, S. J. Am. Chem. Soc. 1988, 110, 4336. 183

(51) Wan, P.; Muralidharan, S. Can. J. Chem. 1986, 64, 1949.

(52) Steenken, S.; McClelland, R. A. J. Am. Chem. Soc. 1989, 111, 4967.

(53) Faria, J. L.; Steenken, S. J. Phys. Chem. 1992, 96, 10869.

(54) Mella, M.; Fasani, E.; Albini, A. J. Org. Chem. 1992, 57, 3051.

(55) Akaba, R.; Kamata, M.; Sakuragi, H.; Tokumaru, K. Tetrahedron Lett. 1992, 33, 8105.

(56) Akaba, R.; Niimura, Y.; Fukushima, T.; Kawai, Y.; Tajima, T.; Kuragami, T.; Negishi, A.; Kamata, M.; Sakuragi, H.; Tokumaru, K. J. Am. Chem. Soc. 1992, 114, 4460.

(57) Faria, J. L.; Steenken, S. J. Am. Chem. Soc. 1990, 112, 1277.

(58) Arnold, B. R.; Scaiano, J. C.; McGimpsey, W. G. J. Am. Chem. Soc. 1992, 114, 9978.

(59) Kirmse, W.; Kilian, J.; Steenken, S. J. Am. Chem. Soc. 1990, 112, 6399.

(60) Kirmse, W.; Strehlke, I. K.; Steenken, S. J. Am. Chem. Soc. 1995, 117, 7007.

(61) McClelland, R. A. Tetrahedron 1996, 52, 6823.

(62) Wan, P.; Yates, K. J. Am. Chem. Soc. 1982, 104, 2509.

(63) Arnaut, L. G.; Formosinho, S. J. J. Photochem. Photobiol. A: Chem. 1992, 69, 41.

(64) Mathivanan, N.; Cozens, F.; McClelland, R. A.; Steenken, S. J. Am. Chem. Soc. 1992, 114,2198.

(65) Gilbert, A.; Baggott, J. Essentials ofMolecular Photochemistry; CRC Press: Ann Arbor, 1991.

( 66) Kagan, J. Organic Photochemistry: Principles and Applications; Academic Press: New York, 1993.

(67) Formosinho, S. J.;Arnaut, L. G. Advances in Photochemistry; Wiley: New York, 1991; Vol. 16. 184

(68) Wagner, P. J. Topics in Current Chemistry; Springer-Verlag: New York, 1976; Vol. 66.

(69) Wagner, P. J. Acc. Chem. Res. 1971, 4, 168.

(70) Wagner, P. J. Acc. Chem. Res. 1983, 16, 461.

(71) Wagner, P. J. Acc. Chem. Res. 1989, 22, 83.

(72) Borrell, P.; Sedlar, J. Trans. Faraday Soc. 1970, 66, 1670.

(73) Scaiano, J. C. J. Photochem. 1973, 2, 81.

(74) Filipescu, N.; Minn, F. L. J. Am. Chem. Soc. 1968, 90, 1544.

(75) Porter, G.; Wilkinson, F. Trans. Faraday Soc. 1963, 59, 1689.

(76) Scaiano, J. C. J. Am. Chem. Soc. 1980, 102, 7747.

(77) Cowan, D. O.; Drisko, R. L. Elements of Organic Photochemistry; Plenum Press: New York, 1976.

(78) KopeckY, J. Organic Photochemistry: A Visual Approach; VCH: New York, 1992.

(79) Carey, F. A.; Sunberg, R. J. Advanced Organic Chemistry, Part A; Plenum Press: New York, 1977.

(80) Swenton, J. S. J. Chem. Ed 1969, 46, 7.

(81) Kaup, G. Angew. Chem. Int. Ed Engl. 1978, 17, 150.

(82) Barltrop, J. A.; Coyle, J. D. Excited States in Organic Chemistry; Wiley: New York, 1975.

(83) Padwa, A. Organic Photochemistry; Marcel Dekker: New York, 1979; Vol. 4.

(84) Borrell, P.; Cashmore, P. Trans. Faraday Soc. 1969, 65, 1595.

(85) Hara, H.; Tanaka, I. Bull. Chem. Soc. Jpn. 1974, 47, 1543.

(86) Guillory, W. A.; Andrews, G. H.J. Chem.Phys. 1975, 62, 3208.

(87) Collin, G. J. Can. J. Chem. 1973, 51, 2853. 185

(88) McNeely, S. A.; Kropp, P. J. J. Am. Chem. Soc. 1976, 98, 4319.

(89) Kropp, P. J.; Fravel, H. G.; Fields, T. R. J. Am. Chem. Soc. 1976, 98, 840.

(90) Kaupp, G.; Stark, M. Angew. Chem. Int. Ed Engl. 1977, 16, 552.

(91) Cristo!, S. J.; Lee, G. A.; Noreen, A. L. J. Am. Chem. Soc. 1973, 95, 7067.

(92) Deno, N. C.; Sacher, E. J. Am. Chem. Soc. 1965, 87, 5120.

(93) Miller, S. A.; Pearman, F. H. Analyst 1950, 75, 492.

(94) Grob, C. A. Angew. Chem. Int. Ed. Engl. 1969, 8, 535.

(95) (a) Trudell, M. B. M.Sc. Thesis, San Francisco State University, 1988; (b) Keeffe, J. R. Personal communication.

(96) Fosse, R. Bull. Soc. Chim. Fr. 1906, 35, 105.

(97) Boyd, M. K. Ph.D. Thesis, University of Toronto, 1988.

(98) Samanta, A.; Gopidas, K. R.; Das, P. K. J. Phys. Chem. 1993, 97, 1583.

(99) Lapin, S. C.; Schuster, G. B. J. Am. Chem. Soc. 1983, 107, 4243.

(100) Clifton, M. F.; Fenick, D. J.; Gasper, S. M.; Falvey, D. E.; Boyd, M. K. J. Org. Chem. 1994, 59, 8023.

(101) Reid, W.; Schonherr, J. Chem. Ber. 1960, 93, 1870.

(102) Poole, C. F.; Poole, S. K.; Chromatography Today, Elsevier: New York; 1991.

(103) McNair, H. M.; Bonelli, E. J. Basic Gas Chromatography, Consolidated Printers: Berkeley; 1969.

(104) Schomburg, G. Gas Chromatography. A Practical Course, VCH: New York; 1990.

(105) Bunce, N. J. Photochem. 198115, 1.

(106) (a) Wagner, P. J.; Kemppainen, A. E. J. Am. Chem. Soc. 1972, 94, 7495. (b) Wagner, P. J.; Kelso, P.A.; Kemppainen, A. E.; Zepp, R. G. J. Am. Chem. Soc. 1972, 94, 7500. 186

(107) Fleming, S. A. Personal Communication, 1995.

(108) Silverstein, R. M.; Bassler, G. C.; Morrill, T. C. Spectrometric Identification Of Organic Compounds; Wiley: New York, 1991.

{l 09) Nader, R. Dissertation zur Erlangung der Doktorgrades der Mathematisch Naturwissenschaftlichen Fakultat der Gerog-August, Universitat zu Gottingen, Gottingen 1978.

(110) Pommer, H. Angew. Chem. Int. &I. Engl. 1977, 16, 423.

(111) Maryanoff, B. E.; Reitz, A. B. Chem. Rev. 1989, 89, 863.

(112) Oberley, L. W. Superoxide Dismutase; CRC Press Inc.: Boca Raton, 1982.

(113) Eberson, L. Electron Transfer Reactions in Organic Chemistry; Springer-Verlag: Berlin, 1987.

(114) Turro, N. J. Modem Molecular Photochemistry; Benjamin/Cummings: Menlo Park, 1978.

(115) (a) Hess, L. D.; Jacobson, J. L.; Schaffuer, K.; Pitts, J. N. J. Am. Chem. Soc. 1967, 89, 3684. (b) Dauben, W. G.; Shaffer, G. W.; Deviny, E. J. J. Am. Chem. Soc. 1970, 92, 6273.

(116) (a) Padwa, A. Org. Photochem. 1967, 1, 91. (b) Johnson, C. K.; Doming, B.; Reusch, W. J. Am. Chem. Soc. 1963, 85, 3894.

(117) Crandall, J. K.; Mayer, C. F.; Arrington, J.P.; Watkins, R. J. J. Org. Chem. 1974, 39, 248.

(118) Person, M. D.; Kash, P. W.; Butler, L. J. J. Chem. Phys. 1992, 97, 355.

(119) Noh, T.; Lei, X-G; Turro, N.J. J. Am. Chem. Soc. 1993, 115, 3105.

(120) Robbins, W. K.; Eastman, R. H.J. Am. Chem. Soc. 1970, 92, 6076.

(121) Coulson, D.R.; Yang, N. C. J. Am. Chem. Soc. 1966, 88, 4511.

(122) Calvert, J. G.; Pitts, J. N. Photochemistry; Wiley: New York, 1966.

(123) Kutschke, K. O.; Wijnen, M. H. J.; Steacie, E. W. R. J. Am. Chem. Soc. 1952, 74, 714. 187

(124) Whiteway, S. G.; Masson, C.R. J. Am. Chem. Soc. 1955, 77, 1508.

(125) Wagner, P. J. J. Am. Chem. Soc. 1967, 89, 5898.

(126) Wan, P.; Krogh, E. J. Am. Chem. Soc. 1989, 111, 4887.

(127) Sheldon, R. A.; Koehl, J. K. J. Am. Chem. Soc. 1970, 92, 4395.

(128) Cramer, W. A. J. Phys. Chem. 1967, 71, 1171.

(129) Nelsen, S. F.; Bartlett, P. D. J. Am. Chem. Soc. 1966, 88, 137.

(130) Kopecky, K. R.; Yeung, M.-Y. Can. J. Chem. 1988, 66, 374.

(131) Camaioni, D. M.; Franz, J. A. J. Org. Chem. 1984, 49, 1607.

(132) Okamato, A.; Snow, M. S.; Arnold, D.R. Tetrahedron 1986, 42, 6175.

(133) Padwa, A. Organic Photochemistry, Marcel Dekker: New York, 1991; Vol. 11.

(134) Olofson, R. A.; Walinsky, S. W.; Marino, J.P.; Jernow, J. L. J. Am. Chem. Soc. 1968, 90, 6554.

(135) March, J. Advanced Organic Chemistry, Wiley: New York; 1992.

(136) International Critical Tables; McGraw-Hill: New York, 1930;Vol. 3, p 35.

(137) Winkler, J. R.; Netzel, T. L.; Creutz, C.; Sutin, N. J. Am. Chem. Soc. 1987, 109, 2381.

(138) Jackson, R. L.; Lishan, D. G. J. Phys. Chem. 1984, 88, 5986.

(139) ACD/Labs, Advanced Chemistry Development Inc., Ontario, Canada, 1997.

(140) Pouchert, C. J.; Behnke, J. The Aldrich Library of 13C and 1H FT NMR Spectra, Aldrich Chemical Company: Milwaukee, 1993; Vol. 1, 2. VITA

The author, Mary Frances Clifton, was born in Chicago, Illinois. She received a B.S.

degree from the University of Illinois at Champaign-Urbana. After graduation Fran was

employed for nine years as a Polymer Chemist at DeSoto, Inc. She left industry to complete

her M.S. in Chemistry under the direction ofDr. Kurt Field at Bradley University. In August,

1992 Fran enrolled at Loyola University of Chicago working under the direction of Dr. Mary

K. Boyd. While a graduate student at Loyola, Fran received both the Graduate Assistantship in Areas of National Need (GAANN) and the Arthur J. Schmitt Dissertation Fellowship.

188 APPROVAL SHEET

The dissertation submitted by Mary Frances Clifton has been read and approved by the following committee:

Dr. Mary K. Boyd, Ph.D., Director Associate Professor, Chemistry Loyola University Chicago

Dr. David S. Crumrine, Ph.D. Professor, Chemistry Loyola University Chicago

Dr. Willetta Greene-Johnson, Ph.D. Assistant Professor, Chemistry Loyola University Chicago

Dr. Duarte Mota de Freitas, Ph.D. Professor, Chemistry Loyola University Chicago

The final copies have been examined by the director of the dissertation and the signature which appears below verifies the fact that any necessary changes have been incorporated and that the dissertation is now given final approval by the committee with reference to content and form.

The dissertation is therefore accepted in partial fulfillment of the requirements for the degree of Doctor of Philosophy.

Date Research Director