BY ALKYL BENZENES . Thesis for the Degree‘offPh_.D.* ' MICHIGANSTATE—umvmsm _ ' RICHARD ALAN mm ‘ 1971 ‘ EVIDENCE FOR CHARGE TRANSFER IN ,THE1 _: ; PHOTOREDUCT-ION om- TRIFLUOROACETOPHENONE ‘i i :2 ABSTRACT EVIDENCE FOR CHARGE TRANSFER IN THE PHOTOREDUCTION 0F o-TRIFLUOROACETOPHENONE BY ALKYL BENZENES By Richard Alan Leavitt A comparison of the photoreactivities of acetophenone and trifluoroacetophenone triplets toward alkyl benzenes reveals differences in the mechanisms by which they abstract hydrogen atoms. The results, detailed below, suggest that the strongly electron-withdrawing trifluoro- methyl group makes the trifluoroacetophenone triplets considerably more electron-deficient than the acetophenone triplets, so that they rapidly undergo charge transfer interactions with electron-rich alkyl benzenes prior to actual hydrogen atom abstractions. (1) Quantum yields of (substituted) bibenzyl formation were measured as a function of hydrogen donor concentration with both trifluoroaceto- phenone and acetophenone. The rate constants of ketone triplet-donor interaction (kr)’ derived from the quantum yields, for acetophenone were consistent with previous investigations, while the kr values for trifluoro- acetophenone were 16-l40 times as large as the acetophenone values (e.g., Richard Alan Leavitt 5 l 73 and 1.2 x lO M' sec"1 respectively, with toluene). However, the quantum yields of bibenzyl formation were much lower with trifluoro- acetophenone suggesting a new source of inefficiency occurring after the initial triplet interaction. (2) Trifluoroacetophenone's kr values were quite sensitive to substituents on the aromatic ring of the donor. A linear Hammett correlation of log kr versus 0; (p = -l.80) was found with a series of p-substituted toluenes (p-Cl-, F-, CH -, CH30-, CN-, and H-) indicating that the primary trifluoroacetophenone triplet process involves the generation of a positively charged alkyl benzene. (3) Toluene and toluene-d3 were found to have identical kr values with trifluoroacetophenone which suggests that the initial triplet reaction is not hydrogen abstraction. However, the large isotope effect (3.3) observed in the maximum quantum yields indicates that the inefficiency of product fOrmation may be due to reverse charge transfer. A similar effect was found with cumene and cumene-a-d, although a small isotope effect (1.25) in the kr values may indicate some direct hydrogen atom abstraction in competition with charge transfer. Actually some hydrogen abstraction must occur since cyclohexane does photoreduce 6 l trifluoroacetophenone, although with a smaller rate constant (3.7 x 10 M' sec-1) than fOund for alkyl benzenes in general. Richard Alan Leavitt (4) p-Di-tfbutylbenzene, which cannot undergo direct hydrogen abstraction, nevertheless quenches trifluoroacetophenone triplets with a rate constant (1.3 x 107M'15ec'1) comparable to the kr values found for toluene and p-xylene. 'Therefore triplet trifluoroacetophenone must undergo similar interactions with the three substrates. (5) Trifluoroacetophenone, unlike acetophenone, does not show selectivity toward C-H bond strength and, for example, interacts faster with toluene than with cumene. EVIDENCE FOR CHARGE TRANSFER IN THE PHOTOREDUCTION OF a-TRIFLUOROACETOPHENONE BY ALKYL BENZENES By Richard Alan Leavitt A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1971 To Susie who came into my life at the beginning of these studies and to Kristine Marie who came to life at the conclusion. ii ACKNOWLEDGEMENTS The author never ceased to be amazed at the way Professor Peter J. Wagner, from a distance of thirty feet, could predict rate constants from vpc peaks that hadn't even come down to the base line yet. His never ending desire to be at the front of important developments characterizes his attitude toward both research and the welfare of his students. I sincerely appreciate having had the opportunity to work with him. I wish to thank the Department of Chemistry for providing the overall excellent facilities for conducting research and for financial assistance. Special thanks to NSF for research assistantships administered by Professor Wagner. Finally, I'm sure I would never have survived if it hadn't been fOr the many hours of baseball, football, tennis, ping-pong, darts and parties... and chemistry enjoyed with my fellow graduate students. TABLE OF CONTENTS Page INTRODUCTION .......................... l A. DEVELOPMENT OF MECHANISM ............... l 1. Early Work .................... l 2. Identity of Reactive State ............ 3 3. n,n* and n,n* Triplets .............. 3 4. General Mechanistic Scheme and Kinetic Expressions .................... 6 B. PREVIOUS STUDY OF TRIFLUOROACETOPHENONE ........ 9 C. CHARGE TRANSFER .................... 12 D. RESEARCH OBJECTIVES .................. 14 RESULTS ............................ l6 A. QUANTUM YIELD DETERMINATIONS ............. 16 B. KETONE TRIPLET QUENCHING ................ l6 l. With Naphthalene ................. l6 2. With p-Di-tggtrbutylbenzene ............ l7 C. INTERSYSTEM CROSSING ................. l7 DISCUSSION ........................... 27 A. CHARGE TRANSFER MECHANISM ............... 27 l. Hammett o-p Relationship ............. 29 iv TABLE OF CONTENTS (Continued) Page 2. Charge Transfer Quenching ............. 3l 3. Sensitivity to C-H Bond Strength ......... 32 4. Deuterium Isotope Effects ............. 33 5. Solvent Effects .................. 34 6. Non-aromatic Donor ................ 35 7. Yields and Quantum Yields ............. 35 B. WHICH TRIPLET REACTS? ................. 37 C. SUMMARY ........................ 39 D. FURTHER EXPERIMENTS .................. 39 EXPERIMENTAL .......................... 41 PART I. MATERIALS AND PROCEDURES .............. 4l A. PREPARATION AND PURIFICATION OF MATERIALS ....... 4l l. Ketones ...................... 4l a. a-Trifluoroacetophenone ............ 4l b. Valerophenone ................. 42 c. Acetophenone ................. 42 2. Solvents ..................... 42 a. Benzene .................... 42 b. Acetonitrile ................. 42 3. Hydrogen (Deuterium) Donors ............ 42 a. Toluene .................... 42 TABLE OF CONTENTS (Continued) Page b. Toluene-d3 .................. 43 c. p-Chlorotoluene ............... 43 d. p-Fluorotoluene ............... 43 e. p-Methoxytoluene ............... 43 f. p-Cyanotoluene ................ 43 g. p-Xylene ................... 43 h. m-Xylene ................... 43 i. o-Xylene ................... 43 j. Mesitylene .................. 43 k. Cyclohexane ................. 43 l. Cumene .................... 43 m. Cumene-a-d .................. 44 Quenchers .................... 44 a. §i§;piperylene ................ 44 b. Naphthalene ................. 45 c. p-Di-tg357butylbenzene ............ 45 Photoproducts for Identification and Standardization ................. 45 a. Bibenzyl ................... 45 b. Bicumyl ................... 45 c. Bi—p-xylyl .................. 45 vi lxll I'll l TABLE OF CONTENTS (Continued) Fkge d. Bicyclohexyl .................. 45 6. Internal Standards ................. 45 a. Dodecane .................... 45 b. Tetradecane .................. 45 c. Pentadecane .................. 45 d. Hexadecane ................... 45 e. Octadecane ................... 46 f. Nonadecane ................... 46 g. Eicosane .................... 46 7. Other Materials .................. 46 a. a-Methylstyrene ................ 46 b. a-Chlorocumene ................. 46 c. l,2-Dimethoxyethane .............. 46 C. PHOTOLYSIS PROCEDURE .................. 46 1. Preparation of Samples ............... 46 2. Degassing ..................... 47 3. Irradiation .................... 48 D. ANALYSIS OF PHOTOLYSATE ................ 48 1. Instruments .................... 48 2. Product Identification ............... 49 3. Standardization .................. 50 vii TABLE OF CONTENTS (Continued) Page E. ACTINOMETRY ...................... 5] EXPERIMENTAL .......................... 53 PART II. KINETIC DATA .................... 53 A. QUANTUM YIELD OF BIBENZYL FORMATION AS A FUNCTION OF HYDROGEN DONOR CONCENTRATION ............. 53 B. QUENCHING STUDIES ................... 60 C. DETERMINATION OF INTERSYSTEM CROSSING QUANTUM YIELD. . 62 LIST OF REFERENCES ....................... 63 viii LIST OF TABLES TABLE Page 1 Kinetic Data for the Photoreduction of Acetophenone (ACP) and Trifluoroacetophenone (TFA) by p—Xylene in Benzene ....................... ll 2 Least Squares Slope and Intercept Values from 633'] versus [SH(D)] Plots for Trifluoroacetophenone (TFA) (Figures 1-4) and Acetophenone (ACP) (Figure 5) . . . . l8 3 Kinetic Data for Photoreduction of 0.1 M Trifluoro- acetophenone by Hydrogen (Deuterium) Donating Substrates in Benzene ................. l9 4 Kinetic Data for Photoreduction of 0.1 M Acetophenone by Hydrogen Donating Substrates in Benzene ....... 20 5 Triplet Quenching Parameters for Trifluoroacetophenone (TFA) and Acetophenone (ACP) .............. 21 6 Quantum Yield of (Substituted) Bibenzyl Formation for O.lOOO M Trifluoroacetophenone in Benzene with Indicated SH, C", and Conditions ............ 53 Part A. pyChlorotoluene ............ , . . . . 53 Part B. pfFluorotoluene ................ 54 Part C. prMethoxytoluene ............... 54 Part D. Mesitylene .................. 54 Part E. m;Xylgng_ ................... 55 ix LIST OF TABLES (Continued) TABLE Page Part F. 9;Xylggg_ ................... 55 Part G. Toluene ................... 55 Part H. Toluene-d3 .................. 56 Part I. Cumgng_ .................... 56 Part J. Cumene-a-d .................. 56 Part K. Cyclohexane . . . . . ............ 57 Part L. pryanotoluene ................ 57 7 Quantum Yield of (Substituted) Bibenzyl FOrmation for 0.1000 M Acetophenone in Benzene with Indicated SH, Cn, and Conditions .................... 57 Part A. Toluene ................... 57 Part B. Toluene-d3 .................. 58 Part C. Cumene .................... 58 8 Photolysate Hydrogen-DonorbContaining Product Distribution ..................... 59 9 0.1000 M Trifluoroacetophenone - 0.5000 M Toluene System in Benzene. Quenching with p-Di-t; Butylbenzene ..................... 60 10 0.1000 M Trifluoroacetophenone - 0.5000 M p-Xylene System in Benzene. Quenching with p-Di-t; Butylbenzene ..................... 60 TABLE 11 12 l3 14 LIST OF TABLES (Continued) Page 0.1000 M Acetophenone - 0.5000 M Toluene System in Benzene. Quenching with p-Di-thutylbenzene ...... 61 0.1000 M Trifluoroacetophenone - 1.0000 M p-Xylene System. Quenching with Naphthalene in Benzene. . . . . 61 0.1000 M Trifluoroacetophenone - 1.0000 M p-Xylene System. Quenching with Naphthalene in Acetonitrile . . 6l Triplet Sensitized gjsrtrans Isomerization of Piperylene ....................... 62 xi FIGURE LIST OF FIGURES Reciprocal quantum yield plot for trifluoroacetophenone with cumene and cumene-a-d ............... Reciprocal quantum yield plot fbr trifluoroacetophenone with toluene and toluene-d3 .............. Reciprocal quantum yield plot for trifluoroacetophenone with m-xylene, o-xylene, mesitylene, and p-methoxytoluene .................... Reciprocal quantum yield plot for trifluoroacetophenone with p-cyanotoluene, cyclohexane, p-chlorotoluene, and p-fluorotoluene .................... Reciprocal quantum yield plot for acetophenone with cumene, toluene, and toluene-d3 ............ Interaction of trifluoroacetophenone triplets with p-substituted toluenes ................. Vpc traces of photolysates of trifluoroacetophenone with (A), the lower boiling hydrogen donors, and (B), the higher boiling hydrogen donors on COL-2. (1) benzene, trifluoroacetophenone and hydrogen donor; (2) bibenzyl; (3) cross-coupling product; (4) pinacol (d,l and meso) ..................... xii Page 22 23 24 25 26 3O 49 PREFACE The work presented in this thesis represents the fruition and expansion of ideas put forward in the author's Master's thesis (Michigan State University, 1969). Necessarily some duplication of content is unavoidable in order to give an orderly account on the subject material. Many of the results and conclusions presented in this thesis are complementary to and indeed dependent upon those appearing in the Master's thesis. Thus, a summary of this previous work is quite necessary and will be included in the introductory section. xiii INTRODUCTION This dissertation deals with the phenomenological events occuring after the absorption of light by alkyl aryl ketones in the presence of substrates formally capable of donating hydrogen atoms to such ketones. Specifically, the liquid phase intermolecular photoreduction of aceto- phenone and a-trifluoroacetophenone by various substituted alkyl benzenes has been examined in detail and new mechanistic conclusions drawn. A. DEVELOPMENT OF MECHANISM. 1. Early Work. Although photoreduction has always played an essential role in nature in the utilization of CO2 by plants, products from even "simple" systems were not characterized until 1900, when Ciamician and Silber1 first identified benzpinacol and acetone as the photOproducts from the action of sunlight on a solution of benzophenone in isopropyl alcohol. That particular system has been the subject of an enormous number of 00 ©—:© Sunlight \\_ /, + (CH3)2C0 (l) + @@ papers and is still being actively investigated.2 Many other substrates have also been used as the hydrogen source, including, alkanes,3 amines,1+ alkyl benzenes,5 and tributylstannane,6 in the photoreduction of aldehydes and ketones. The current understanding of the mechanism(s) of photoreduction is based on a number of important observations and postulations. During the thirties several researchers7'9 suggested that the photoactivated species responsible for the reaction was a biradical form of the carbonyl. Abstraction of hydrogen from the alcohol by this biradical would then yield two new radicals which subsequently could combine to give the observed products. R CHOH h . . 2 . . Ar2C0 —L—> ArZC—O 9 ArZC-OH + RZC—OH (2) At first it was thought that products such as acetone arose from dis- proportionation of carbinol radicals7a8 but Hirshberg9 and later Pittslo found that optically active alcohols were not racemized when photolized in the presence of aryl ketones. Instead, carbinol radicals are apparently oxidized by reacting with ground state ketones, which in the case of the_ benzophenone-isopropyl alcohol system, is consistent with the observation that the quantum yield of ketone disappearance can approach 2, depending on the conditions used.11,12 Ar co + RZCOH > Ar COH + RCO (3) 2. Identity of Reactive State. Identification of the reactive excited state multiplicity responsible for photoreduction has come from both chemical and spectroscopic evidence. In 1955, Backstrom and Sandros13 suggested that the earlier proposed biradical state could be considered equivalent to an excited triplet state and later found that the phosphorescence emission of biacetyl, which had been previously identified as occurring from the excited triplet state,1“ could be sensitized by benzophenone.15 Terenin and Ermolaev16 observed that in EPA glass at 77°K benzophenone sensitized the phosphorescence of naphthalene while simultaneously quenching its own phosphorescence. Also, oxygen}0 paramagnetic metal chelates,17 naphthalene,13'20 conjugated dienes,21 and various hydrogen donors22 have all been found to quench the reactive state of benzophenone in liquid solutions. Hammond and co-workers confirmed that the reactive state was a triplet by measuring the excited state lifetimes of benzophenone in several solvent systems?”23 and finding them much too long to correspond to excited singlet states. Finally, although conclusive evidence for organic triplets had existed since 1945,ll+b it wasn't until 1963 that triplet states of ketones such as benzophenonez“ and 2-acetonaphthone25 were observed directly by flash photolysis and their lifetimes measured. 3. n,n* and n,n* Triplets. For aryl ketones the fact that two types of triplet states exist has caused considerable controversy in the interpretation of some results. The distinction between the two triplet states can be made by observation of their phosphorescence emission spectra at 77°K. n,n* triplets characteristically have much shorter emission lifetimes than n,n* triplets (on the order of several msec for n,n* and >50 msec for n,1r*)11926 and n,n* triplet emission is much more solvent and substi tuent dependent?“28 n,n* triplets formally arise via the promotion of a non-bonding oxygen electron to an antibonding n* orbital. The net result is a * C=Qj ”1‘" > céo (4) decrease in dipole moment relative to that in the ground state and a lengthened carbon-oxygen bond. Thus, with a lone non-bonding electron remaining, the oxygen becomes much more electrophilic relative to the ground state and resembles an oxy radical. In fact, Walling and Gibian,29 Cohen and Baumgarten,30 and Padwa30 have found that the behavior of benzophenone, whose lowest triplet is clearly n,n* in character,31 parallels very closely that of tertrbutoxy radicals32 in hydrogen abstraction from several donors. n,n* triplets, on the other hand, involve the promotion of an electron from the n-system to an.anti-bonding n* orbital. Lamola33 has demonstrated that the n,n* triplet state of phenyl alkyl ketones corresponds to the lowest triplet state of benzene (3La)' Thus, unlike the n,n* triplet, the n,n* triplet has most of its excitation localized in the aromatic ring, and therefore, has an electron-rich rather than an electron- deficient carbonyl oxygen.3” 2:..- A V One would expect this state to be nucleophilic and thus behave much differently toward hydrogen abstraction than the n,n* triplet and indeed such is the case. Various naphthyl, p-phenyl and p-amino ketones, - possessing lowest n,n* triplets, show little or no tendency toward pinacolization in the presence of most hydrogen donor56a11:27 except amines.37'39 The significance of the latter will be presented later. The controversy arises from situations where the n,n* and n,n* triplet levels are in energetic proximity. The n,n* triplet of unsubstituted acetophenone lies only a few kcal/mole above the n,n* triplet.35’36 p-Methyl- or 2,3-dimethyl-substitution inverts the levels and a significant decrease in triplet reactivity is observed. Yang attributes this to a mixed state arising from vibronic coupling of TI”1* and n,n* triplets; the more n,n* character present the less the reactivity. The same effect is observed for the type II reaction of phenyl alkyl ketones. Electron-donating substituents or high solvent polarity put the n,n* triplet slightly below the n,n* triplet and a substantial decrease in reactivity is reported.“0 Wagner and co-workers“1’“2 suggest that a thermal equilibrium of the two states would still allow the upper n,n* triplet to be the reactive state. Indeed, they have found that the reactivities of p-methoxyphenyl ketones are subject to the same inductive effects of y and 6 substituents and have the same dependence on y C-H bond strength as unsubstituted phenyl ketones possessing n,n* lgwg§t_triplets. In the same context, the principal ketone studied in this thesis, a-trifluoroacetophenone, possesses a n,n* lowest triplet,"3 and yet reacts much faster with alkyl benzenes than does acetophenone. Further aspects of this situation will be considered in the discussion section. 4. General Mechanistic Scheme and Kinetic Expressions. One advantage liquid-phase photochemistry has over the gas-phase is that several kinetic simplifications can usually be made. Absorption of light by a ketone may produce several excited singlets with many vibrational levels. In solution, decay to the lowest excited singlet is ‘2 sec). Thus, it may be assumed that all almost instantaneous (<10' singlet reactions occur from the lowest singlet. For aryl ketones in general, the rate of intersystem crossing (isc) from singlet to triplet 11 s manifold is much faster (10 ec'l)”“ than either fluorescence or radiationless decay. Thus, the excited triplet state is reached essentially with unit quantum efficiency."5 Finally, the rate of phosphorescence is usually much slower than the rates of other triplet processes and can be ignored. The following mechanistic scheme can be written for the photoreduction of ketone, A, by substrate, SH, in the presence of quencher, Q. Scheme I mess; .8162 A0 + ho -—————{;> A'* -¥E¥L{;> A3* Ia (6) A3* -————{;> A0 kd[A3*] (7) A + SH -—————{;> AH- + s- kr[SH][A3*] (8) 113* + Q ——> A, + 03* kqto][A3*J (9) AH + AH- -—————{;> AHAH (10) AH- + S- ___> AHS (1 l) -—————{;> 35 (12) U) + U) Under steady state conditions d[A3*]/dt = 0 and the only directly measurable kinetic parameter is the quantum yield. Using the definition of'Wagner,"6 the quantum yield for a particular photOprocess i is given by the following expression: P (13) i = ¢ES¢R i where ¢ES represents the probabiiity that absorption of light will produce the requisite excited state; d’R is the probability that the excited state will undergo the primary photoreaction necessary for process i; P1. is the probability that any metastable ground state intermediate will proceed to stable product, thus completing process i, rather than forming by-products or reverting to the ground state of the reactant. Thus, since ¢A3* = l for aryl ketones kr[SH] 4’ss 2 Ysspss (krISH]+ kd) ('4) Note that the factor YSS’ the chemical yield of SS product, defined as [SS]/{2[SS] + [AHS]}, must be included since the measured quantum yield is based on only one of the products containing 5- radicals. The term 2[SS] appears in the denominator since two ketone molecules are required for each SS molecule produced. The ratio kd/kr can be determined by inverting equation (14). k _ -l d Ass ‘ Yss Pss (‘ * ‘E;Ishj’° ('5) A plot of reciprocal quantum yield versus reciprocal substrate concentration -1 -1 is linear with an intercept equal to YSS PSS and a slope/intercept value equal to kd/kr' When a triplet quencher is used, equation (14) becomes: kr[SH] YSSPSS ( krfsfij+ kd + kq[Q] ) (l6) ¢ss Dividing equation (14) by equation (16) gives the familiar Stern-Volmer ¢§s/¢SS = 1 + kqiolr (17) 0 SS absence of quencher and r, the lifetime of the triplet state in the relationship where o represents the quantum yield of SS formation in the absence of quencher, is the reciprocal of the sum of the rates of all the reactions undergone by the triplet. _ —l r - (kr[SH] + kd) (18) A plot of relative quantum yield versus quencher concentration with constant substrate concentration is linear and has a slope equal to qu and intercepts at l. The reciprocal of equation (18) is also useful since a plot of 1'] versus substrate concentration is linear with an intercept equal to kd and a slope equal to kr' B. PREVIOUS STUDY OF TRIFLUOROACETOPHENONE. The substitution of fluorine for hydrogen at the a position of aceto- phenone results in a change in both absorption and emission spectra. The trifluoroacetophenone absorption spectrum is quite similar to but shifted ~12 nm bathoChromatically in the 230-300 nm region from that of aceto— phenone. The phosphorescence emission decay of trifluoroacetophenone at 77°K is exponential with a 57-msec lifetime in hydrocarbon glasses (0-0 band at 70.9 kcal) and a ZOO-msec lifetime in ethanol glass (0-0 band at 70.0 kcal). This implies that the La n,n* triplet lies slightly below the n,n* triplet, although both are undoubtedly populated at room temperature."3 In the author's Masters thesis the photochemical behavior of a-trifluoro- acetophenone was compared to that of acetophenone using p-xylene as the 10 hydrogen donating substrate. Basically three experiments were carried out with both ketones: 1) mass balance; 2) triplet quenching; and 3) quantum yield determination. The results of that study which have a direct bearing on the results appearing in the next section are summarized below. With the mass balance experiment it was shown that in both cases the pinacol, bibenzyl, and cross-coupling products account for all the reacted ketone. Unambiguous identifications of the photoproducts were made with authentic samples and various spectral methods. For each of five different p-xylene concentrations Stern-Volmer quenching of the triplet state with naphthalene at 366 nm was done to obtain values of 1 according to equation (17). Plots of reciprocal lifetime versus p-xylene concentration according to the inverse of equation (18) then gave the values of kd and kr which appear in Table l. The absolute quantum yield of bixylyl was determined as a function of p-xylene concentration at 313 nm. Plots of inverse quantum yield versus inverse p-xylene concentration according to equation (15) gave values of kd/kr and oggx which appear in Table l. The results of this study were quite interesting in two regards. While the reactivity of triplet acetophenone toward p-xylene agreed well with that predicted by previous investigations,29:35:“7 the rate constant, kr’ for trifluoroacetophenone triplet concentration with p-xylene was two orders of magnitude larger than that for acetophenone. Moreover, the quantum yield of bixylyl formation (433x) extrapolated to infinite p-xylene concentration was much lower for trifluoroacetophenone than for acetophenone. 11 TABLE 1. Kinetic Data for the Photoreduction of Acetophenone (ACP) and Trifluoroacetophenone (TFA) by p-Xylene in Benzene. b d e -1 -la -1 c -1 max Ketone kr’M sec kd,sec kd/kr kd,sec $88 7 6 6 TFA 9.7 x 10 7.0 x 10 0.091 8.8 x 10 0.040 5 5 5 ACP 7.0 x 10 4.9 x 10 1.01 7.1 x 10 0.102 aSlope of reciprocal lifetime plot. bIntercept of reciprocal lifetime plot. cSlope/intercept of reciprocal quantum yield plot. dCalculated from the slope of the reciprocal quantum yield plot and the value of kr from the reciprocal lifetime plot. eExtrapolated quantum yield of bixylyl at infinite p-xylene concentration. These results thus suggested that either the radicals produced in the trifluoroacetophenone-p-xylene system were considerably different from those of the acetophenone-p-xylene system or that in the trifluoro- acetophenone-p-xylene system the photoreduction proceeded either partially or totally by another mechanism. One of the suggestions for an alternate mechanism made at that time was for a charge transfer interaction between triplet trifluoroacetophenone and p-xylene in competition with direct radical formation. Subsequent decay of this state to give either radicals or ground state reactants could then account for the inefficiency of product formation. * Ph'zcr + ———>k" (If 19 3 PhCCF3 ( ) CH CH3 3 L 12 C. CHARGE TRANSFER. The concept of charge transfer in photoreactions has previously been proposed for reductions involving amines.27’“3'5° In many cases ketones considered unreactive toward most hydrogen donors are easily reduced by amines; even normally reactive ketones have enhanced rates of interaction approaching the diffusion-controlled limit. For example, while the photoreduction of p-aminobenzophenone by 2-propanol is very inefficient, the bimolecular rate constant, kr’ for interaction with triethylamine is approximately 4 x 107M'15ec'].38 An extremely fast rate constant 10 l (1.6 x 10 M' sec-1) has been reported for the photoreduction of fluorenone (n,n* lowest triplet) by l,4-bis(dimethylamino)benzene.51 Cohen has proposed that the great reactivity of amines may be associated with charge transfer or electron transfer to the ketone triplet proceeding actual proton transfer. [Arzé——o' RCH2N+RZJ (20) k k_r Ar COH + RCHNR NR 2 2 Ar2C=O + RCH 2 2 In concert with this notion Cohen has found that t-butylamine quenches 7 l the phosphorescence of benzophenone with a rate constant of 7 x 10 sec- which is only slightly less than amines possessing a-hydrogens and l3 considerably larger than 2-propanol or benzhydrol.52 In a recent study of fluorenone quenching by a series of p-substituted dimethyl anilines, Cohen and Parsons51 found a linear relationship between log kr and op+ values which supports the argument for development of positive charge at the nitrogen atom. In addition, reactions of similar systems show little sensitivity to C-H bond strength, no racemization of optically active amine, and no or very small deuterium isotope effects.38"*9 The charge transfer mechanism has also been previously invoked to explain n,n*] fluorescence quenching by amines.5'*’56 Furthermore, plots of o'] versus [amine]'] have been fOund to curve sharply upward at high amine concentrations indicating singlet quenching.38"'7’56 The type II reaction, expressed in general form below, shows close OH RI JCT—>1; \\\\\\f£§ //J[\\ _+ fl/IR parallels with intermolecular photoreduction when amines are involved.“9a50 Again systems with low lying n,n* triplets are found to have similar reactivities to those with n,n* triplets. Wagner and Kemppainenso in a recent study of y-dimethylaminobutyrophenone and of amine quenching of valerophenone in various solvents have found evidence that the primary triplet ketone reaction in the presence of amines 14 is not the normal type II y-hydrogen abstraction process.26 However, since they did not observe an expected rate enhancement in acetonitrile relative to that in benzene57 they proposed the formation of a charge transfer complex between amine and ketone rather than actual electron transfer. Subsequent competitive decay of this state to the ground state and 1,4 * I.\i\ ' A; > U Ph ° biradicals would then account for the observed inefficiencies of product formation despite shorter triplet lifetimes. 0- RESEARCH OBJECTIVES. After the discovery that trifluoroacetophenone had unusual triplet state reactivity toward p-xylene a set of experiments was designed to test the validity of a charge transfer mechanism. (1) With a charge transfer mechanism the rate constant, kr’ for triplet interaction with aromatic hydrogen donors should be sensitive to both electron donating and electron withdrawing substituents on the aromatic ring. Thus, the kinetics of photoreduction for trifluoroacetophenone with several p-substituted toluenes were examined. (2) kr for a charge transfer mechanism should be insensitive to deuterium substitution of the extractable hydrogens. Thus, comparisons with toluene-d3 and cumene-a-d were made. (3) A charge transfer mechanism would demand that even alkylbenzenes without abstractable hydrogens (a-H) be able to quench the trifluoroaceto- 15 phenone triplets. Thus, quenching with p-di-trbutylbenzene was examined. (4) A comparison of trifluoroacetophenone reactivity with non- aromatic hydrogen donors was needed to set an upper limit on "normal" reduction by radical abstraction. Cyclohexane was chosen for this purpose. (5) Actual electron transfer should be sensitive to solvent polarity. Thus, a comparison of reactivity in benzene and acetonitrile was made. RESULTS A. QUANTUM YIELD DETERMINATIONS. Absolute quantum yields of (substituted) bibenzyl formation were determined for both acetophenone and trifluoroacetophenone as a function of hydrogen (deuterium) donating substrate concentration. Degassed benzene solutions containing 0.1 M ketone, internal standard (typically n-hexadecane, 5 0.004 M) and various concentrations of hydrogen donor (typically 2, l, 0.75, and 0.5 M) were irradiated in parallel at 313 nm. After irradiation each tube was analyzed for bibenzyl product, cross-coupling product, and internal standard by vpc. Valerophenone actinometry, described in the experimental section, was used throughout to monitor the light absorption. In all cases except for the acetOphenone-toluene-d3 system, plots of ¢BB-] versus [SH(D)]-1 were linear and are depicted in Figures 1-5. Slopes drawn in these figures represent least squares fits of the data. Numerical slope and intercept values of these plots appear in Table 2. kd/kr and kr values derived from these data and the kd values from Table l, are given in Tables 3 and 4. B. KETONE TRIPLET QUENCHING. 1. With Naphthalene. A comparison of trifluoroacetophenone triplet lifetimes in benzene and acetonitrile was made by using the Stern-Volmer relationship (equation 16 17 (17)) and the known kq values fbr naphthalene in these solvents.59 Degassed solutions containing 0.1 M ketone, 1.0 M p-xylene, 0.001 M C20 internal standard, and 0.00, 0.01, 0.02, 0.03, or 0.04 M naphthalene were irradiated in parallel at 366 nm to < 4% ketone conversion. Vpc analysis of bibenzyl and C20 peak areas permitted calculation of o The 0 BB/¢BB' Stern-Volmer plots were linear with qu values (slopes) as indicated in Table 5. 2. With p-Di-tert-butylbenzene. The quenching efficiency of p-di-tertybutylbenzene in the presence of trifluoroacetophenone and acetophenone was assessed using the Stern- Volmer relationship and known values of r with toluene and p-xylene. Degassed benzene solutions containing 0.1 M ketone, internal standard (5 0.004 M), constant concentration of hydrogen donor, and various concentrations of p-di-tgrt:butylbenzene were irradiated in parallel at 313 nm. Vpc analysis of bibenzyl or bixylyl and internal standard peaks permitted calculation of ogB/CBB. The Stern-Volmer plots were linear with qu values as indicated in Table 5. C. INTERSYSTEM CROSSING. A comparison of the amount of cis-piperylene isomerized to trans-piperylene by trifluoroacetophenone and acetophenone will give the for trifluoroacetophenone since ¢ 4 SC for acetOphenone is unity.“5 isc i Degassed benzene solutions containing 0.1 M ketone and 0.2 M gisfpi- perylene were irradiated in parallel at 313 nm to ~10% formation of trans-piperylene. 18 The data appearing in Table 14 yield a a. value for 15¢ trifluoroacetophenone of 0.95 which is considered to be within experimental error of 1.0, and henceforth, will be taken as unity in all calculations involving it. 1 TABLE 2. Least Squares Slope and Intercept Values from ¢BB- versus [SH(D)] Plots for Trifluoroacetophenone (TFA) (Figures 1-4) and Acetophenone (ACP) (Figure 5).a Ketone Donor Slopeb Interceptb TFA p-Methoxytoluene 0.53 10 TFA p-Methyltoluene 2.3 25 TFA p-Fluorotoluene 11 23 TFA Toluene 22 19 TFA p-Chlorotoluene 16 19 TFA p-Cyanotoluene 2340 123 TFA o-Methyltoluene 2.9 26 TFA m-Methyltoluene 3.3 1 .1 33 t 1 TFA Toluene—d3 80 66 TFA Mesitylene 1.4 23 TFA Cumene 28 14 TFA Cumene-a—d 63 25 TFA Cyclohexane 1060 434 ACP p-Methyltoluene 9.9 9.8 ACP Cumene 14 5.4 ACP Toluene 45 7.7 aTo check the reproducibility of these data, two completely separate determinations were made for the TFA-meMethyltoluene system. The values given represent the averages of these two determinations. bValues are estimated to be t 3% per footnote a. 19 TABLE 3. Kinetic Data for Photoreduction of 0.1 M Trifluoroacetophenone by Hydrogen (Deuterium) Donating Substrates in Benzene. a maxb c 7 -l -l +e Donor YBB ¢BB kd/kr kr,10 M sec op p-Methoxytoluene 0.14 0.098 0.053 17 -0.78 p-Methyltoluene 0.26 0.040 0.091 9.7 -0.31 p-Fluorotoluene 0.24 0.044 0.48 1.8 -0.07 Toluene 0.24 0.053 1.2 0.73 0.00 p-Chlorotoluene 0.23 0.053 0.84 1.0 +0.11 p-Cyanotoluene 0.03 0.008 19 0.046 +0.65 o-Methyltoluene 0.24 0.038 0.11 8.0 m-Methyltoluene 0.24 0.030 0.10 t .005 8.8 Toluene-d3 0.24 0.015 1.2 0.73 Mesitylene 0.30 0.044 0.061 14 Cumene 0.25 0.073 2.0 0.44 Cumene-a-d 0.25 0.039 2.5 0.35 Cyclohexanef 0.089 0.002 2.4 0.37 aYBB, as defined in the introduction, is calculated from the [(BB)/(CC)]N values appearing in Table 8 using the following relationship: YBB = [(BB)/(CC)]N/{2[(BB)/(CC)]N + 1}. bExtrapolated quantum yield of (substituted) bibenzyl at infinite hydrogen (deuterium) donor concentration. CSlope/intercept values from Table 2. Values are estimated to be t 5% based on that found for the TFA-m-Methyltoluene system. dCalculated from kd/kr values assuming kd = 8.8 x lofisec'1 (from Table 1). Values estimated to be reproducible t 10%. eReference (58). f gTakes into account cyclohexylbenzene. BB is bicyclohexyl. TABLE 4. Kinetic Data for Photoreduction of 0.1 M Acetophenone by Hydrogen Donating Substrates in Benzene. kr’ 10 d 5 1 1 M' sec- 7.0 2.8 1.2 e Donor YBB p-Xylene 0.30 Cumene 0.28 Toluene 0.27 Toluene-d3 0.27 a,b,c Same as Table 3. 7.1 x105 sec'1 (from Table l). e¢ Calculated from kd/kr values assuming kd = 8-1 vs. [50]”1 was not linear. .xpco mmpwswpmm mm umugmmmg on cpaosm use memNcmnpauan-uLmu-mcua mo cowpmeucmocoo 21 mco xpco mcwm: mums mew; muopa Lm5~o>-cgmumu .Ammv mucmgwmmmo .z ooom.on .z oooo.Fm m-o_ x m._ mop x F.m uoe.o mcmNcmm mcm~cmanpzaum-ea-a nmcmapoe au< mlop x m.~ No_ x m._ No.F memNcmm mcm~cma_zp:num-wo-a amcmspoh Fom Lmnocmac Locos mcopmx — — F- .Amu ground-state th—CH2-<:::::> ‘3'99 PhLCF3 + reactants CF3 L.— __ CC k . dif disp kBB CHZCH2 4 have been reported for the triplet type II (equation (21)) reactions of nonanophenone,66 5-decanone67 and 2-hexanone.68 There is, however, a significant isotope effect on the quantum yields of bibenzyl fonnation. Table 3 shows that the maximum quantum yield with toluene-d3 is 3.5 times smaller than that for toluene. This is an indication that after the charge-transfer complex is formed there is competition between proton transfer (kpt) and back electron transfer (k-r)° Unfortunately, it was not possible to make a direct kinetic comparison with the acetophenone-toluene-d3 system since the 033'] versus [SD].1 plot was not linear but curved upward, presumably because unimolecular radical scavanging reactions (by solvent or residual oxygen) begin to compete with coupling at low steady-state radical concentrations. 34 The individual quantum yields, however, were significantly lower than the corresponding ones with toluene. The fact that trifluoroacetophenone triplets interact more slowly with cumene than with toluene suggests that the photoreduction with cumene may proceed in part by direct hydrogen abstraction and should thus exhibit a kinetic isotope effect. From Table 3 the kr value for cumene-a-d is ~20% less than the kr value for cumene itself. The possible i 10% experimental error in these values, however, makes the magnitude of the isotope effect difficult to assess. If one does use these kr values and assumes an isot0pe effect of 5 for direct hydrogen atom abstraction, then one calculates that only 25% of the reaction with cumene involves direct hydrogen abstraction with a rate constant of 1.1 x 106M']sec']. Again a significant isotope effect (1.9) is found in the intercepts of plots according to equation (15). 5. Solvent Effects. Solvent effects were examined only briefly in this study but do deserve some comment. As shown in Table 5, the trifluoroacetophenone triplet lifetime in the presence of l M p—xylene is 10 x 10'95ec in 9sec in acetonitrile. It's unlikely that 6 benzene while only 5.5 x 10' kd in acetonitrile cuuld be any larger that the 8.8 x 10 sec.1 observed in benzene since in benzene kd is already unusually large for reasons speculated on earlier. Thus kr must increase by a factor of two in acetonitrile which, although not a large factor, is in the right direction to lend further support f0r the formation of charged species as the primary photochemical process of triplet trifluoroacetophenone. 35 6. Non—aromatic Donor. Cyclohexane was the only non-aromatic hydrogen donating substrate utilized in this investigation and was needed to establish an upper limit f0r the amount of direct hydrogen abstraction by trifluoroacetophenone 6 1sec-1 triplets. Its rate constant of 3.7 x 10 M' made it second only to p-cyanotoluene as the least reactive donor measured and is significant in two regards. First, triplet trifluoroacetophenone is about an order of magnitude more reactive toward it than is triplet acetophenone.29’69 Second, the relative smallness of this rate constant emphasizes the difference in the behavior of aromatic hydrogen donors toward trifluoro- acetophenone triplets. Benzophenone and acetophenone have been reported to be 2-3 times more reactive with cyclohexane than with toluene; and yet for trifluoro- acetophenone the factor is reversed, toluene being twice as reactive as cyclohexane. Since the primary triplet reaction with cyclohexane must be hydrogen abstraction, this behavior is again a good indication that‘ the primary trifluoroacetophenone triplet interaction with alkyl benzenes is not direct hydrogen abstraction. However, this study and results presented for cumene point out the fact that a small amount of direct hydrogen abstraction can compete with charge transfer. The amount proceeding via direct hydrogen abstraction is most important with the less reactive donors and with donors of secondary and tertiary hydrogen atoms. 7. Yields and Quantum Yields. Quantum yields of bibenzyl formation are considerably lower for trifluoroacetophenone than for acetophenone. Theoretically, the maximum 36 quantum yield would be 0.5 if every triplet ketone molecule produced one benzyl radical and that radical then coupled only with a like radical. However, maximum quantum yields are considerably less than 0.5 because benzyl radicals cross-couple with trifluoromethyl phenyl carbinol radicals. The maximum quantum yield of bibenzyl f0rmation is equivalent to the YBB values appearing in Tables 3 and 4. If one assumes little difference in rates of radical recombination, then a statistical recombination of radicals would produce two molecules of cross-coupling product for every one bibenzyl molecule, resulting in YBB = 0.25. Examination of the YBB values for trifluoroacetophenone in Table 3 shows the majority to be 0.25 t 0.01 indicating close to statistical recombination of radicals. In the case of acetophenone the YBB values (0.27-0.30) from Table 4 show a somewhat higher proportion of like radical coupling. These values for both ketones would further suggest that, after their formation, the radicals have time to diffuse out of the solvent cage before recombination, since otherwise cross-coupling products would dominate. The observed quantum yields of bibenzyl formation extrapolated to infinite donor concentration were much lower than the 0.25 predicted by YBB values for a statistical product distribution. For trifluoro- acetophenone, the highest quantum yield was 0.098 with p-methoxytoluene, most of the others being in the 0.03-0.07 range. Quantum yields were substantially higher for acetophenone, being, for example, 0.18 with cumene. 37 The efficiency with which bibenzyl is formed after the primary triplet state process can be calculated by the following expression. p33 = oggleBB (23) For trifluoroacetophenone these values are all less than 0.30 with the exception of p-methoxytoluene (PBB = 0.70). Each of the corresponding values for acetophenone is more than twice as large as the for trifluoro- acetophenone. However, they are still significantly less than one (the largest value was 0.64 with cumene) and must indicate some disproportionation of the radicals. If one assumes similar amounts of radical dispropor- tionation with trifluoroacetophenone, then the much smaller PBB values would indicate that a large portion of the charge transfer interactions do not result in radical f0rmation but instead in reversion to the ground state of the reactants via reverse electron transfer. This would be in agreement with the previously mentioned isotope effects on 033x. B. WHICH TRIPLET REACTS? Trifluoroacetophenone is an unusual ketone in that it possesses both a n,n* lowest triplet and enhanced triplet state reactivity. Although ketones with n,n* lowest triplets are generally quite unreactive in photo- reduction processes, two types of exceptions apparently exist. (1) Ketones that have a n,n* triplet only a few kcal above the n,n* triplet show moderate reactivities which can be attributed to either reactions from an equilibrium concentration of the upper triplet (n,n*) or vibronic coupling of the two states, producing a n,n* lowest triplet with some n,n* character. 38 Regardless of how it is stated, the point is that the n,n* state still directs the abstraction process. The high photoreactivity of trifluoro- acetophenone relative to acetophenone with cyclohexane can best be attributed to the fact that the trifluoromethyl group creates a more electron-defficient, and thus more reactive, n,n* triplet state. (2) Ketones with n,n* lowest triplets show enhanced reactivities in the presence of substrates capable of electron donation. Amine donors apparently have sufficiently low ionization potentials that they can give up electrons in preference to hydrogen atoms. 0n the other hand, trifluoroacetophenone triplets, in contrast to ketone triplets in general, apparently have a sufficiently high electron affinity that they can abstract electronic charge from such weak donors as alkyl benzenes. ‘ To speculate on which state is responsible for this charge transfer to difficult at this point and could easily be a combination of the two states with kr being given by the following expression, _ n n kr — xnkCT + xnkCT + XnkH (24) where Xn and X1T are the equilibrium concentrations of n,«* and n,n* triplets; k T are rate constants for charge transfer; and kH is the C rate constant for hydrogen atom abstraction. 39 C. SUMMARY. The involvement of charge transfer in the photoreduction of trifluoroacetophenone by alkyl benzenes donors is well supported. Triplet reactivities showed marked sensitivity toward substituents on the aromatic ring of the donor, yielding a large negative 9 value from a Hammett relationship; a donor without abstractable hydrogens was found to interact with the triplets as efficiently as those donors with abstractable hydrogens; the rate constants of triplet-donor interaction were not reduced by deuterium substitution; and the reactivities did not display sensitivities to C-H bond strength, as is characteristic of hydrogen atom abstraction processes. However, direct hydrogen abstraction may be a minor competing factor since cyclohexane does photoreduce trifluoro- acetophenone, although with a considerably slower rate than the aromatic donors. 0. FURTHER EXPERIMENTS. The following experiments are just a sample of the many kinds of things that could be done with systems of a similar nature. (1) Since the question of which triplet reacts has not been answered, future research could be directed toward elucidation of the triplet state reSponsible for charge transfer. The spacing of trifluoroacetophenone n,n* and n,n* triplets is much too close for studies of this nature. However, with appropriate substitution on the aromatic ring it should be possible to create situations where either the n,n* or the n,n* triptet is much lower than the other. 40 (2) In view of the exceedingly rapid donating capabilities of substituted amines and the rapid accepting properties of trifluoro- acetophenone it should be possible to observe charge transfer interactions of the ketone singlet state. (3) Other a-substituted acetophenones could be examined to ascertain how strongly electron-withdrawing the group need be to induce charge transfer. EXPERIMENTAL PART I. MATERIALS AND PROCEDURES. A. PREPARATION AND PURIFICATION OF MATERIALS. Purity is of critical importance in determining photochemical rate constants, since even small amounts of quencher can have large effects on slow rate constants. For example, for a hydrogen donor (k=lx106] 9' r impurities, the intercept of the plot according to equation (15) becomes M' sec-1) containing 0.1% quenching (kq = 5 x 10 M' sec") 9 ""t —7—(5xm)( )1 (> Y P 1 + 0.001 25 (l x 10 ) or 6Y"]P'1 instead of Y"]P'1 when no quencher is present. Thus, kr (observed) would be six times the kr (actual) with no quencher present. Therefore, all compounds used in the photolyses described in this thesis were carefully scrutinized by vpc after purification to insure against such occurrences. l. Ketones. a. a-Trifluoroacetgphenone (Columbia Organic Chemicals) as received contained ~10% impurities. Purification by spinning band distillation at atmospheric pressure gave a center cut that was >99.9% trifluoroacetophenone. The results of a photolysis check for quenching impurities by varying the concentration of ketone with a constant amount 41 42 of hydrogen donor indicated ng_quenching impurities present (qu = 103 and 101 for 0.1 and 0.2 M trifluoroacetophenone respectively, in the presence of 0.5 M p-xylene). b. Valeroghenone (Aldrich Chemical Company) for actinometry purposes, was distilled under reduced pressure, passed through alumina, and redistilled. c. Acetophenone (Matheson Coleman and Bell) was distilled under reduced pressure and the center cut retained. 2. Solvents. a. Benzene (Fisher Scientific Company, 99% thiophene free) was stirred over concentrated H2504 several times (24 hour periods) until the acid layer no longer turned yellow. The benzene was then successively washed with l M Na0H, distilled water, and saturated NaCl solution, followed by drying over anhydrous M9504 and fractional distillation from P205. Only the center cut (~80%) was retained. b. Acetonitrile (Fisher Scientific Company) was distilled from KMnO4 and Na2C03 according to the procedure of O'Donnell, Ayres and Mann.70 The center cut of a final fractional distillation was retained. 3. Hydrogen (Deuterium) Donors. a. Toluene (Fisher Scientific Company) purification was analogous to that of benzene except that the bottle of toluene was kept in cold water while stirring with concentrated H2504 to minimize sulfonation. 43 b. Toluene-d3 (Merck Sharp and Dohme of Canada Limited, 5 g, 99 mole % D) was used as received. c. p:Chlorotoluene (Eastman Organic Chemicals) purification was analogous to that of benzene. d. p-Fluorotoluene (Matheson Coleman and Bell) purification was analogous to that of benzene. e. p3Methoxytoluene (Columbia Organic Chemicals) was purified by successively washing with l M Na0H, distilled water, and saturated NaCl solution, and dried over anhydrous MgSO4. The center cut was retained from a fractional distillation over sodium. f. p-Cyanotoluene (Eastman Organic Chemicals) was fractionally distilled from P205 and the center fraction retained. g. g-Xylene (Aldrich Chemical Company) purification was analogous to that of toluene. h. m-Xylene (Eastman Organic Chemicals) purification was analogous to that of toluene. i. o-Xylene (Aldrich Chemical Company, 99+ %) purification was analogous to that of toluene. j. Mesitylene (Matheson Coleman and Bell) purification was analogous to that of toluene except that it was fractionally distilled under reduced pressure. k. Cyclohexane (Fisher Scientific Company) purification was analogous to that of benzene. l. Cumene (Eastman Organic Chemicals) purification was analogous to that of toluene. 44 m. Cumene-a-d was prepared by reducing a-chlorocumene with a 1:1 molar ratio of LiAlD4 (International Chemical and Nuclear, 99 atom % D) and AlCl 4.66 g (0.111 mole) LiAlD4 were covered with 100 ml 3. of glyme (distilled from P205) in a 500 mfl 3-necked flask equipped with a mechanical stirrer, condenser and an addition funnel containing 34 g (0.21 mole) a-chlorocumene in 75 ml of glyme. A slury of 14.82 g (0.111 mole) A1Cl3 in 100 ml of glyme was slowly added to the LiAlD4 solution. The mixture was brought to reflux befOre adding the a-chloro- cumene over a one hour period. After an additional hour of refluxing and subsequent cooling, 100 ml of wet ether was added to destroy any unreacted LiAlD4-A1Cl3. The resultant solution was poured into 100 ml of 10% H2504 and the ether layer separated. The aqueous layer was extracted with three 50 ml portions of ether which were subsequently combined with the original ether layer. The ether was successively washed with saturated NaHCO3 solution, cold distilled water, and saturated NaCl solution, and dried over anhydrous M9504. The ethers were removed by distillation on a spinning band column to minimize loss of cumene-a-d. The subsequent purification procedure, analogous to that of toluene, removed the major side product, a-methylstyrene. Distillation on a spinning band column afforded a center cut that was >99.98 % cumene-a-d, with no detectable a-methylstyrene. Mass spectral analysis showed the cumene-a-d to be 96.4% d]. 4. Quenchers. a. §i§:piperylene (Aldrich Chemical Company) was passed through alumina followed by distillation. 45 b. Naphthalene (Matheson Coleman and Bell) was purified by three recrystallizations from ethanol. c. p-Di-tgptrbutylbenzene obtained from student preparations (Friedel Crafts alkylation of benzene) was purified by three recrystallizations from methanol, mp 77-78°C. 5. Photoproducts for Identification and Standardization. a. Bibenzyl (Aldrich Chemical Company) was recrystallized twice from methanol. b. Bicumyl (Columbia Organic Chemicals) was recrystallized twice from methanol. c. Biep-xylyl (Aldrich Chemical Company) was recrystallized twice from methanol. d. Bicyclohexyl was prepared using the method of Wilds and McCormack,71 which is a coupling reaction of cyclohexyl bromide induced by C0612. 6. Internal Standards. a. Dodecane (Aldrich Chemical Company) purification was analogous to that of benzene with a final distillation under reduced pressure. b. Tetradecane (Columbia Organic Chemicals) purification was analogous to that of dodecane. c. Pentadecane (Columbia Organic Chemicals) purification was analogous to that of dodecane. d. Hexadecane (Aldrich Chemical Company) purification was analogous to that of dodecane. 46 e. Octadecane (Aldrich Chemical Company) purification was analogous to that of benzene with a final recrystallization from ethanol. f. Nonadecane (Chemical Samples Company) was used without further purification. g. Eicosane (Matheson Coleman and Bell) was used without further purification. 7. Other Materials. a. a-Methylstyrene (Aldrich Chemical Company) was used as received in the preparation of a-chlorocumene. b. a-Chlorocumene was easily prepared by adding gaseous HCl to stirring a-methylstyrene cooled to 0°C. A manometer attached to the reaction flask indicated when the reaction was completed. After aspiration to remove excess HCl and drying over anhydrous MgSO4, the liquid was fractionally distilled under reduced pressure. The center cut was retained for the preparation of cumene-a-d. c. l,2-Dimethoxyethane (Mallinckrodt Chemical Works) was fractionally distilled from P205. C. PHOTOLYSIS PROCEDURE. 1. Preparation of Samples. Class A volumetric flasks and pipettes were used exclusively to make up solutions for photolysis. Typically, two stock solutions, ketone- internal standard and hydrogen donor, were prepared. The appropriate amount of internal standard was weighed into a 10 ml volumetric flask and filled to the mark with solvent. 1 ml was then pipetted into a 5 ml __“T! .l“ E; g: 47 volumetric flask containing the appropriate amount of weighed ketone and diluted to the mark with solvent. Likewise, the hydrogen donor was weighed into a 10 ml volumetric flask and diluted to the mark with solvent. Into each of four 10 ml volumetric flasks was placed 1 ml of the ketone-internal standard stock solution, 1, 1.5, 2, or 4 ml of hydrogen donor stock solution, and solvent to fill to the mark. From each of these solutions three exactly 2.8 ml portions were withdrawn via a 5 m1 syringe and injected into 13 x 100 mm pyrex culture tubes which had been drawn into small capillaries about 2 cm from the open end to facilitate sealing after degassing. On occasions when the hydrogen donor was in limited supply (cumene-a-d and toluene-d3) two 2.4 ml portions drawn from 5 ml volumetric flasks were injected into drawn tubes containing eleven 4 mm glass beaks. This amount of glass beads was found to keep the level of liquid high enough in the photolysis tubes to receive all the light and yet not interfere with the photolysis. 2. Degassing. In order to remove dissolved oxygen, sample tubes were attached to a vacuum line over no. 00 one-holed rubber stoppers on individual stopcocks. The solutions were slowly frozen above liquid nitrogen and then immersed before opening to the vacuum. A minimum vacuum of 0.005 torr was attained before closing the stopcocks and allowing the tubes to thaw. After the fourth freezing and evacuation the tubes were sealed off with a torch. “if; [I'll 1|,Il.‘ I .II 48 3. Irradiation. Sample tubes were irradiated in parallel on a rotating merry-go- round apparatus72 immersed in a water bath at 25°C, to insure that the same amount of light impinged upon each sample. For quantum yield determinations a 450 watt Hanovia medium pressure mercury lamp housed in a water cooled quartz immersion well was used. The 300-320 nm region was isolated with a 1 cm path of 0.002 M potassium chromate in a 1% aqueous solution of potassium carbonate. For naphthalene quenching experiments the 366 nm region was isolated with a set of corning no.7083 filter combinations. 0. ANALYSIS OF PHOTOLYSATE. 1. Instruments. All analyses for photoproducts were made on the following two vpc's which used flame ionization detectors. VPC-l: Aerograph Hy-Fi model 6000 with 550 oven and 328 programmer. Leeds and Northrup Speedomax H recorder equipped with model 207 Disc integrator. VPC-2: Varian Aerograph 1200. Leeds and Northrup Speedomax W recorder equipped with model 224 Disc integrator. A variety of vpc columns were utilized as designated below. COL-l: 6' x 1/8" Aluminum containing 4% QF-l, 1.2% Carbowax 20 M on 60/80 chromosorb G. COL-2: 6' x 1/8" stainless steel containing 5% SE-30, a/w DEGS on chromosorb W. *mmc '15 E. I. Q 49 COL-3: 25' x 1/8" aluminum containing 25% l,2,3-Tris(2-cyanoethoxy) propane on 60/80 chromosorb P. COL-4: 6' x 1/8" stainless steel containing 3% SE-30 on 100/120 varaport 30. COL-5: 1.5' x 1/8" aluminum containing 15.5% diisodectylphthalate on 60/80 chromosorb P. COL-6: 8' x 1/8" aluminum containing 4% QF-l, 1% carbowax 20 M on 60/80 chromosorb G. 2. Product Identification. Photoproducts were identified primarily by their vpc retention times and comparisons to authentic samples. The photolysate of trifluoroacetophenone with any of the hydrogen donors used always contained the three products shown below. (A) ’ l (a) ” . J UL (1) (21(3) (4) (l) (4) (3) (2) FIGURE 7. Vpc traces of photolysates of trifluoroacetophenone with (A), the lower boiling hydrogen donors, and (B),the higher boiling hydrogen donors on COL-2. (l) benzene, trifluoroacetophenone and hydrogen donor; (2) bibenzyl; (3) cross-coupling product; (4) pinacol (d,1 and meso). 50 Since the pinacol was common to all, it was relatively easy to identify the other two products in each case. With the lower boiling hydrogen donors (p-fluorotoluene and cyclohexane) the hydrogen donor coupling product came off first followed by the cross-coupling and the double pinacol peaks. With the higher boiling hydrogen donors (p-chloro-, p-methoxy-, and p-cyanotoluene, cumene, cumene-a-d, and mesitylene) the pinacols had the shortest retention time followed by the cross-coupling and bibenzyl peaks respectively. Only the photolysates from the xylenes and unsubstituted toluene had to be analyzed on a QF-l column (COL-l) because the three peaks overlapped on the SE-30 column (COL-2). 3. Standardization. To determine the concentration of a photoproduct, PP, one need only compare its vpc peak area to that of an internal standard, IS. [PP] = SF [IS] :2 (26) where [X] and (X) represent the concentration and vpc peak area respectively, of X. SF, the vpc standardization factor which compensates for the different molar responses for each compound, is determined by weighing out known amounts of PP and IS and measuring the relative vpc peak area ratio. _ PP Std IS Std 27 SF _ ITSistd PP std ( ) Bicumyl, bibenzyl, bixylyl, and bicyclohexyl were standardized directly with the appropriate hmernal standard using the same conditions Erma: used for the corresponding photolysate analysis. 51 Since bimesityl, di-p- chloro-, di-p-fluoro-, and di—p-methoxybibenzyl were not readily available, they were standardized with the appropriate internal standard indirectly using the fol (toluene) lowing relationship. (xltoluene thus, where, -t 1 _ (bibenzyl) di-x-bibenz l (28) i-x-bibenzyl) 1benzy [di-x-bibenzyl] = SF [IS d"x;bib°"z ' (29) SF = ,(toluene std [x-toluene]st (k-toluenelstd [toluene]Std Actual SF's used appear with the tables of quantum yields found in the following sec tion. E. ACTINOMETRY. d (IS)std (bibenzyl s bibenz 1 std 19< IISIstd >00) All sample tubes for which quantum yields were measured had ketone concentrations sufficient to absorb >99.9% of the incident light. Acetophenone formation from the type II photoelimination of valerophenone was used to monitor the light absorption. Actinometer tubes containing 0.1000 M valerophenone and 0.0050 M C14 internal standard in benzene were prepared as described before and irradiated in parallel with sample tubes. 52 Photolysate analyses were made on VPC-2, COL-6. The quantum yield of acet0phenone formation under these conditions was taken to be 0.33.“0 In cases of low quantum yields, several sets of actinometer tubes had to be used in series. §j§;Piperylene-acet0phenone actinometry was used to measure the quantum yield of intersystem crossing for trifluoroacetophenone. Tubes containing 0.2000 M gj§;piperylene and either 0.1000 M trifluoroacetophenone or acetophenone in benzene were irradiated in parallel at 313 nm. .4..-“ F - 1 The triplet state of acetophenone, formed quantitatively (61.5c from the singlet, is completely quenched by gisrpiperylene; the excited =1) piperylene then decays to both gig; and trans-piperylene in a known ratio.“S By use of the following relationship, the amount of excited piperylene . .555 . 3 [gjgfp1p.]oln.555_% trans. = [p1p.* 1' (3]) triplets produced can be calculated for both acetphenone (ACP) and trifluoroacetophenone (TFA). Thus, 3 __ PIP* TFA (¢isc) TFA ‘ PIP* ACP (32) Photolysate analyses were made on VPC-l; COL-3. EXPERIMENTAL PART II. KINETIC DATA A. QUANTUM YIELD OF BIBENZYL FORMATION AS A FUNCTION OF HYDROGEN DONOR CONCENTRATION. General Comments: BB refers to the (substituted) bibenzyl formed from the indicated hydrogen donor, SH. Cn represents the internal standard. Values of (BB)/(Cn) are averages of two or more tubes. Ia values were obtained from valerophenone actinometry as described previously and are averages of at least two tubes per period. Many quantum yield determinations required several sets of actinometer tubes irradiated in series. TABLE 6. Quantum Yield of (Substituted) Bibenzyl Formation for 0.1000 M Trifluoroacetophenone in Benzene with Indicated SH, C", and Conditions. Part A. p_-Chlorotoluene.a -5 -1 [SH], M (BB)/(C20) [BB], 10 M Ia’ E1 088 2.0000 1.694 253 0.0675 0.0375 1.0000 1.296 194 0.0675 0.0287 0.7500 1.120 167 0.0675 0.0248 0.5000 0.890 133 0.0675 0.0197 a0.0010 M 020, SF = 1.494. vpc-1; COL-2. 53 54 a Part B. p;Fluorotoluene. -5 -1 [SH], M (BB)/(C]4) [BB], 10 M Ia, El 438 2.0000 1.512 152 0.0431 0.0354 1.0000 1.283 129 0.0431 0.0300 0.7500 1.449 115 0.0431 0.0268 0 5000 0.964 97 0.0431 0.0226 a0.0010 M 0 SF = 1.008. VPC-l; COL-2. Part C. prMethoxytoluene.a -5 -1 [SH], M (BB)/(C]9) [BB], 10 M Ia, El 033 0.5000 1.164 174 0.0195 0.0891 0.2000 1.838 274 0.0346 0.0791 0 1000 1.468 219 0.0346 0.0631 0.0750 1.333 199 0.0346 0.0573 0.0500 1.131 168 0.0346 0.0487 a0.0010 M C SF = 1.49. vpc-1; COL-2. . a Mes1tylene. -5 -1 [SH], M (BB)/(C]9) [BB], 10 M Ia, E1 088 0 2000 0.655 145 0.0436 0.0333 0.1000 0.540 120 0.0436 0.0275 0.0750 0.484 107 0.0436 0.0247 0.0500 0.388 86 0.0436 0 0197 a0.0020 M 0 SF = 1 11. vpc-1; COL-2. 55 Part E. EL-Xylene.a -5 -1 [SH], M (BB)/(C]9) [88], 10 M Ia, E1 033 2.0000 1.438 192 0.0669 0.0288 1.0000 1.376 184 0.0669 0 0275 0 7500 1.337 179 0 0669 0.0267 0.5000 1.263 169 0.0669 0.0253 0 2000 1.887 252 0.1241 0.0203 0 1000 1.415 189 0.1241 0.0152 0.0750 1.217 163 0 1241 0.0131 0.0500 0.936 125 0.1241 0.0101 a0.0010 M C19, SF = 1.337. vpc-1; COL-l. Part F. p;Xylene.a -5 -1 [SH], M (BB)/(019) [BB], 10 M Ia, E1 OBB 0 2000 0.439 235 0.0971 0.0242 0 1000 0.329 176 0 0971 0.0181 0.0750 0.274 147 0 0971 0.0151 0.0500 0.214 114 0 0971 0.0118 a0.0040 M C19, SF = 1.337. vpc-1; COL-l. Part G. Toluene.a [SH] M (BB)/(C ) [BB] 10’5 M I E1"1 6 ’ 15 ’ a’ BB 2.0000 0.594 200 0.0579 0.0345 1.0000 0.841 275 0.1150 0.0239 0.5000 0.700 236 0.1150 0.0205 0.5000 0.548 185 0 1150 0.0161 a0.0025 M C SF = 1.35. vpc-1; COL-l. 19’ 56 Part H. Toluene-d3 a -5 -1 [SD], M (BB)/(C]9) [BB], 10 M Ia’ El <1>BB 2.0000 0.620 209 0.2265 0.00923 1.0000 0.458 155 0.2265 0.00684 0 7500 0.495 167 0.2818 0.00593 0.5000 0.436 147 0.3374 0.00436 a0.0025 M 019, SF = .35. vpc-1; COL-l. Part I. Cumene.a -6 -1 [SH], M (BB)/(C]5) [BB], 10 M Ia, El 038 1.0000 0.2844 972 0.0412 0.0236 0.5000 0.1693 578 0.0412 0.0143 0.5000 0.3591 1227 0.0839 0 3750 0.2766 945 0.0839 0.0113 0.2500 0.2901 991 0.1259 0.0079 a0.0040 M 015, SF = 0.854. vpc-2; COL-4. Part J. Cumene-a-d.a -6 -1 [SD], M (BB)/(C]5) [BB], 10 M a, El OBB 1.0000 0.2683 917 0.0830 0.01119 1.0000 0.2782 950 0 0839 0.5000 0.2445 835 0.1259 0.00663 0.3750 0.2532 865 0.1680 0.00515 0.2500 0.2639 902 0.2507 0.00360 a0.0040 M 015, SF = 0.854. vpc-2; COL-4. 57 Part K. Cyclohexane. [SH] M (BB)/(C )b [88] 10'6 M E1" 4 ’ 12 ’ a’ BB 8.2482 0.192 103 0.0463 0.00222 2.0000 0.185 99 0.0917 0.00108 1.0000 0.173 93 0.1369 0.00068 0.7500 0.188 101 0.1816 0.00055 0.5000 0.172 92 0.2265 0.00041 a0.0005 M C12. SF = 1.07. VPC-l; COL-2. bBB is bicyclohexyl. Part L. p_-Cyanotoluene.a [SH]. M (cup/(CC)b 2.000 0.415 1.000 0.760 0.750 1.022 0.500 1.633 a0.0020 M C18' VPC-l; COL-2. bCC refers to the cross-coupling product, which was measured because of BB decomposition on the column. TABLE 7. Quantum Yield of (Substituted) Bibenzyl Formation for 0.1000 M Acetophenone in Benzene with Indicated SH, Cn’ and Conditions. Part A. Toluene.a ‘ -5 -1 [SH], M (BB)/(C]9) [BB], 10 M Ia’ E1 033 2.0000 0.577 188 0.0574 0.0328 1.0000 0.689 225 0.1155 0.0195 0.7500 0.766 250 0.1724 0.0145 0.5000 0.543 177 0.1724 0.0103 a0.0025 M C SF = 1.305. vpc-1; COL-l. 19’ 58 Part B. Toluene-d3.a [S0] M (BB)/(C ) [BB] 10‘5 M E1‘1 6 ’ 19 ’ a’ BB 2.0000 0.547_ 181 0.2262 0.00798 1.0000 0.533 176 0.4959 0.00354 0.7500 0.433 143 0.6059 0.00236 0.5000 0.250 82 0.6059 0.00136 a0.0025 M 019, SF = 1.32. vpc-1; COL-l. Part C. Cumene.a [SH] M (BB)/(C ) [BB] 10'13 M E1“ 6 3 19 ’ a’ 88 6.4130 0.345 911 0.0686 0.1328 2.0000 0.219 578 0.0686 0 081] 2.0000 0.412 1088 0.1396 1.0000 0.284 750 0.1396 0.0537 0.7500 0.348 919 0.2173 0.0423 0.5000 0.306 807 0 2173 0 0362 0.5000 0.387 1022 0.2890 a0.0025 M C SF = 1.056. vpc-1; COL-1. 19’ 59 TABLE 8. Photolysate Hydrogen-Donor-Containing Product Distribution. Ketonea Donor (BB)/(CC)b [(BB)/(CC)]NC TFA p-Fluorotoluene 0.458 0.475 TFA p-Chlorotoluene 0.402 0.417 TFA p-Methoxytoluene 0.189 0.196 TFA p-Cyanotoluene 0.032d 0.031 TFA Cumene 0.559 0.512 TFA Cumene-a-d 0.524 0.480 TFA p-Xylene 0.545 0.528 TFA m-Xylene 0.478 0.464 TFA o-Xylene 0.490 0.475 TFA Mesitylene 0.824 0.756 TFA Toluene 0.447 0.462 TFA Toluene-a-d3 0.429 0.444 TFA Cyclohexane 0.111 0.085e ACP p-Xylene 0.786 0.761 ACP Cumene 0.683 0.626 ACP Toluene 0.588 0.609 ACP Toluene-a-d3 0.582 0.603 aTFA = trifluoroacetophenone, ACP = acet0phenone. prc area ratios; BB = (substituted) bibenzyl, CC = cross-coupling product. Average of several determinations. cNormalized area ratio which takes into account number of carbon and oxygen atoms in both products. Assumes response linear with number of carbons and that a single C—O bond reduces response of that carbon by 1/2. e.g., correction factor for TFA with p-fluorotoluene would be 14 5/14. dEstimate. eCyclohexyl— benzene also produced. 60 B. QUENCHING STUDIES. Relative quantum yields of (substituted) bibenzyl formation were measured as a function of quencher concentration for use in Stern-Volmer quenching plots. Studies with naphthalene and p-di-t butylbenzene used 366 nm and 313 nm light respectively. Concentrations of bibenzyl formation 3 were kept in the 5 x 10'4 - 2 x 10' M region to insure linearity. TABLE 9. 0.1000 M Trifluoroacetophenone - 0.5000 M Toluene System in Benzene. Quenching with p-Di-trButylbenzene.a [Quencher], M (Bibenzyl)/(C]7)b ¢°BBI¢BB 0.0000 0.260 1.00 0.0500 0.203 1.28 0.5000 0.174 1.49 0.7500 0.145 1.79 1.0000 0.126 2.06 b a0.0040 M C17 internal standard. VPC-2; COL-4. Vpc area ratio. TABLE 10. 0.1000 M Trifluoroacetophenone - 0.5000 M p-Xylene System in Benzene. Quenching with p-Di-t-Butylbenzene.a . b o [Quencher], M (Bibenzy1)/(C]8) 6 33/638 0.0000 1.400 1.00 0.6000 1.215 1.15 b a0.00125 M C18 internal standard. VPC-l; COL-l. Vpc area ratio. 61 TABLE 11. 0.1000 M Acetophenone - 0.5000 M Toluene System in Benzene. Quenching with p-Di-trButylbenzene.a b o [Quencher], M (Bibenzyl)/(C]6) 6 BB/6BB 0.0000 0.371 1.00 1.0000 0.266 1.40 b a0.0040 M C16 internal standard. VPC-2; COL-4. Vpc area ratio. TABLE 12. 0.1000 M Trifluoroacetophenone - 1.0000 M p-Xylene System. Quenching with Naphthalene in Benzene.a . b o [Quencher], M (B1xyly1)/(C20) 6 BB/6BB 0.0000 0.714 1.00 0.0100 0.472 1.51 0.0200 0.355 2.01 0.0300 0.288 2.48 b a0.0010 M C20 internal standard. VPC-l; COL-l. Vpc area ratio. TABLE 13. 0.1000 M Trifluoroacetophenone - 1.0000 M p-Xylene System. Quenching with Naphthalene in Acetonitrile.a b o [Quencher], M (Bixylyl)/(C20) 6 BBl6BB 0.0000 0.898 1.00 0.0100 0.578 1.55 0.0200 0.430 2.09 0.0300 0.342 2.63 a0.0010 M C20 internal standard. VPC—l; COL-l. prc area ratio. 62 C. DETERMINATION OF INTERSYSTEM CROSSING QUANTUM YIELD. TABLE 14. Triplet Sensitized gj§;trans Isomerization of Piperylene.a bi Ketone Sample 1 Sample 2 Sample 3 Average % trans % trans % trans % trans Acetophenone 0.0940 0.0932 0.0949 0.0940 Trifluoro- acetophenone 0.0895 0.0894 0.0912 0.0900 a'0.2000 M initial gig-piperylene. vpc-1; COL-3. b0.1000 M. —l O soooumcn-bwm 10. 11. 12. 13. 14. 15. 16. LIST OF REFERENCES . Ciamician and P. Silber, Ber,, 66, 2911 (1900); 66, 1530 (1901). . A. Weiner, J. Amer. Chem. Soc., 26, 425 (1971). . Padwa, Tetrahedron Lett., 3465 (1964). . G. Cohen and R. J. Baumgarten, J. Amer. Chem. Soc., 61, 2996 (1965). . S. Hammond and P. A. Leermakers, ibid., 66, 207 (1962). . Bodenstein, Z. Physik, Chem., 6kg, 151 (1931). . L. J. Backstrom, ibid., 666, 99 (1934). G S A S G. S. Hammond, W. P. Baker, and W. M. Moore, ibid,, 66, 2795 (1961). G M H C . Weizmann, E. Bergmann, and Y. Hirshberg, J. Amer. Chem. Soc., 66, 1530 (1938). J. N. Pitts, Jr., R. L. Letsinger, R. P. Taylor, J. M. Patterson, G. Rektenwald, and R. B. Martin, 1610,, 6L, 1068 (1959). J. N. Pitts, Jr., H. W. Johnson, and T. Kuwana, J. Phys. Chem., 66, 2456 (1962). A. Beckett and G. Porter, Trans. FaradaypSoc., 66, 2039 (1963). H. L. J. Backstrom and K. Sandros, J. Chem. Phys., 66, 2197 (1955). a) G. N. Lewis and M. Kasha, J. Amer. Chem. Soc., 66, 2100 (1944); b) G. N. Lewis and M. Calvin, ibid,, 61, 1232 (1945). H. L. J. Backstrom and K. Sandros, Acta Chem. Scand., 16, 48 (1960). A. Terenin and V. Ermolaev, Trans. Faraday Soc., 66, 1043 (1956). 63 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 64 W. M. Moore, G. S. Hammond, and R. P. Foss, J. Chem. Phys., 6%, 1594 (1960). G. Porter and R. Wilkinson, Trans. Faraday Soc., 61, 1686 (1961). G. S. Hammond and P. A. Leermakers, J. Phys. Chem., 66, 1148 (1962). W. M. Moore and M. Ketchum, J. Amer. Chem. Soc., 66, 1368 (1962). G. S. Hammond, N. J. Turro, and P. A. Leermakers, ibid., 66, 2396 (1961); J. Phys. Chem., 66, 1144 (1962). H. L. J. Backstrom and K. Sandros, Acta. Chem. Scand., L6’ 822 (1958). a) G. S. Hammond and W. M. Moore, J. Amer. Chem. Soc., 6L, 6334 (1959); b) W. M. Moore, G. S. Hammond, and R. P. Foss, ibid., 811’ 2789 (1961 ). . A. Bell and H. Linschitz, 1910,, 66, 528 (1963). . J. Wagner, Accounts Chem. Research, 6, 168 (1971). J W. A. Bryce and G. H. J. Wells, Can. J. Chem., Ql, 2722 (1963). p G . Porter and P. Suppan, Trans. Faraday Soc., 61, 1664 (1965); 3375 (1966). . G. Schultz, C. D. DeBorer, W. G. Herkstroeter, and R. H. Schlessinger, . Amer. Chem. Soc. , 26, 6086 (1970). . Padwa, Tetrahedron Lett. ,3465 (1964). . Tsubomura and S. Tanaka, Chem. Phys. Letters, 1, 309 (1967). . Walling and M. J. Mintz, J. Amer. Chem. Soc. , 66,1515 (1967). . A. Lamola, J. Chem. Phys. , 61, 4810 (1967). . Bryce-Smith, Pure App1.Chem., 16, 47 (1968). 8% A J C. Walling and M. J. Gibian, ibid. ,61, 3361 (1965). A H C A D R. M. Hochstrasser, Accounts Chem. Research, ,266 (1968). 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 65 N. C. Yang and R. L. Dusenbery, J. Amer. Chem. Soc., 22, 5899 (1968); Molecular Photochem., l, 159 (1969). . G. Cohen and J. I. Cohen, J. Phys. Chem., 22, 3782 (1968). G. Cohen and J. B. Guttenplan, Tetrahedron Lett., 5353 (1968). A. Davis, J. Amer. Chem. Soc., 22, 2264 (1969). . J. Wagner and A. E. Kemppainen, ibid., 22, 5898 (1968). '0 ‘U G) U) (A O o . J. Wagner, A. E. Kemppainen, and H. N. Schott, 1919,, 22, 5280 (1970). P. J. Wagner, M. J. May, A. Haug, and D. R. Graber, 1919,, 22, 5269 (1970). P. J. Wagner and R. A. Leavitt, 1919,, 22, 5806 (1970). F. Wilkinson and J. T. Dubosis, J. Chem. Phys., 22, 377 (1963). A. A. Lamola and G. S. Hammond, 1919,, 22, 2129 (1965). P. J. Wagner in "Creation and Detection of the Excited State", Vol. I, Part A, A. A. Lamola, Ed., Marcel Dekker, Inc., New York, N. Y., 1971, p. 175. S. G. Cohen and 8. Green, J. Amer. Chem. Soc., 22, 6824 (1969). S. G. Cohen and H. M. Chao, 1919,, 22, 165 (1968). a) A. Padwa, W. Eisenhardt, R. Gruber, and D. Pashayan, 1919,, 22, 1857 (1969); b) A. Padwa and W. Eisenhardt, 1999,, 22, 1400 (1971). P. J. Wagner and A. E. Kemppainen, 1919,, 22, 3085 (1969). . G. Cohen and G. Parsons, 1919,, 22, 7603 (1970). . G. Cohen and A. D. Litt, Tetrahedron Lett., 837 (1970). S S H. Leonhard and A. Weller, Ber. Bunseges Physik. Chem., 62, 791 (1963). N. Mataga and K. Ezumi, Bull. Chem. Soc. Japan, 22, 1355 (1967). 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 66 W. R. Ware and H. P. Richter, J. Chem. Phys., 22, 1595 (1968). J. Guttenplan and S. G. Cohen, Tetrahedron Lett., 2125 (1969). E. M. Kosower and M. Mohanmad, J. Amer. Chem. Soc., 22, 3271 (1968). L. M. Stock and H. C. Brown, Advances in Physical Organic Chemistry, 2, 89 (1963). P. J. Wagner and I. Kochevar, J. Amer. Chem. Soc., 22, 2232 (1968). K. M. Johnson and G. H. Williams, J. Chem. Soc., 1146 (1960). C. Walling and B. B. Jacknow, J. Amer. Chem. Soc., 22, 6133 (1960). W. D. K. Clark, A. D. Litt, and C. Steel, 1919,, 22, 5413 (1969). J. Saltiel, H. C. Curtis, L. Metts, J. W. Miley, J. Winterle, and M. Wrighton, 1919,, 22, 410 (1970). W. A. Pryor, Free Radicals, McGraw-Hill Book Company, New York, 1966, p. 154. E. I. Heiba, R. M. Dessau, and W. J. Koehl, Jr., J. Amer. Chem. Soc., '22, 6830 (1969). P. A. Kelso, Ph.D. Thesis, Michigan State University, 1971. A. Padwa and W. Bergmark, Tetrahedron Lett., 5795 (1968). J. A. Barltr0p and J. D. Coyle, ibid., 3235 (1968). F. D. Lewis, ibid., 1373 (1970). J. F. O'Donnell, J. T. Ayres, and C. K. Mann, Anal. Chem., 22, 1161 (1965). A. L. Wilds and W. B. McCormack, J. Org. Chem., 22, 45 (1949). F. G. Moses, R. S. H. Liu, and B. M. Monroe, Molecular Photochem., 2, 245 (1969). S ATE 1111 1293 “(1111111111111111111.“ 0301 7266 5 "11111 .RL-:-I_ ..a_ d“-A-——~ H- 2 - _ I