UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Quantum yield studies of the photolyses of various tetrahydro-1,4-naphthoquinones 1975

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
UBC_1975_A6_7 L68.pdf
UBC_1975_A6_7 L68.pdf
UBC_1975_A6_7 L68.pdf [ 5.11MB ]
UBC_1975_A6_7 L68.pdf
Metadata
JSON: 1.0061086.json
JSON-LD: 1.0061086+ld.json
RDF/XML (Pretty): 1.0061086.xml
RDF/JSON: 1.0061086+rdf.json
Turtle: 1.0061086+rdf-turtle.txt
N-Triples: 1.0061086+rdf-ntriples.txt
Citation
1.0061086.ris

Full Text

QUANTUM YIELD STUDIES OF THE PHOTOLYSES OF VARIOUS TETRAHYDRO-l,4-NAPHTHOQUINONES BY JOHN PETER LOUWERENS B.Sc, U.B.C, Vancouver 1970 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in the Department of CHEMISTRY We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August, 1975 In present ing th is thes is in p a r t i a l fu l f i lment of the requirements for an advanced degree at the Un ivers i ty of B r i t i s h Columbia, I agree that the L ibrary sha l l make it f r ee ly ava i l ab le for reference and study. I fur ther agree that permission for extensive copying of th is thes is for s c h o l a r l y purposes may be granted by the Head of my Department or by h is representa t ives . It is understood that copying or pub l i ca t ion of th is thes is fo r f i n a n c i a l gain sha l l not be allowed without my wri t ten permission. Depa rtment The Univers i ty of B r i t i s h Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 - i i - ABSTRACT. The 366 nm photolysis of 2,3,4a,6,7,8a-hexamethyl-4a3,5,8,8a6- tetrahydro-l,4-naphthoquinone ( 2_7_ ) i n benzene yielded 28_ with a quantum y i e l d of 0.066 + 0.003 and 29_ with a quantum y i e l d of 0.089 + 0.003. The formation of 28_ was suggested to occur v i a a C Q hydrogen abstraction by C. oxygen i n a f i v e membered t r a n s i t i o n state ( 3-hydrogen abstraction ), and the formation of 29 was believed to occur v i a a C Q hydrogen abstraction by C„ carbon i n a o j six membered t r a n s i t i o n state ( y-hydrogen abstraction ). A Stern-Volmer analysis ( effect of t r i p l e t quencher concentration on quantum yiel d ) showed that .28 was formed from a singlet excited state, whereas 29_ was formed from a t r i p l e t excited state. The 366 nm photolysis of 6,7-dimethyl-4a(3,5,8,8a(3-tetrahydro- 1,4-naphthoquinone (10 ) i n benzene yielded _12 with a quantum y i e l d of 0.0080 + 0.0008 and 1_3 with a quantum y i e l d of 0.0164 + 0.0012. The 366 nm photolysis of 10_ i n tert-butanol yielded 11_ with a quantum y i e l d of 0.0081 + 0.0008. A l l three photoproducts of 10 were suggested to occur via' 3-hydrogen abstraction by oxygen. Stern-Volmer analyses showed that both photoproducts 12_ and 1J arise v i a a t r i p l e t and a singlet excited state, whereas 11_ only arises v i a a singlet state. A mechanism i s proposed explaining why products similar to 11 and - i i i - 12 are not observed in the photolysis of 2_7. This mechanism is bas on conformational control of the biradical intermediate by the bridgehead substituents. The effect of methyl substituents on the chromophore of 2_7 are implicated in shifting down the energy of the ( TT,TT* ) t r i p l e t . It is argued that 29_ arises via yhydrogen abstraction by this ( IT,IT* t r i p l e t . 0 11 - iv - TABLE OF CONTENTS Page INTRODUCTION 1 A. General 1 B. Photochemistry of Various Tetrahydro-l,4-naphthoquinones 5 C. Objectives of the Present Research 16 RESULTS AND DISCUSSION 17 A. Quantum Yield Studies of 2,3,4ae,6,7,8a3-Hexamethyl-4a, 5,8,8a-tetrahydro-l,4-naphthoquinone ( 27_ ) in Benzene 17 1. Synthesis and photolysis 17 2. Unquenched Quantum Yield Measurements 18 3. Photolysis of 27 using Piperylene as Quencher .... 19 4. Photolysis of 2_7 using 1,3-Cyclohexadiene as Quencher 20 5. Photolysis of 27_ using trans-Stilbene as Quencher. 24 6. Photolysis of 2_7 using Oxygen as Quencher 28 7. Photolysis of 27: Effect of Changing the Concentration of 27_ 29 B. Interpretation of the Results of Quenching on the Photolysis of 2_7 31 C. Quantum Yield Studies of the Photolysis of 6,7-Dimethyl- 4aB,5,8,8a6-tetrahydronaphthoquinone 1JD in Benzene ... 38 1. Synthesis and Photolysis 38 2. Unquenched Quantum Yield Measurements 39 - v - Page 3. Photolysis of 1_0 using Piperylene as Quencher .... 40 4. Photolysis of 1£ using 1,3-Cyclohexadiene as Quencher 40 5. Photolysis of 10̂  using trans-Stilbene as Quencher. 46 6. Photolysis of 1_0: Effect of Changing the Concentration of 10_ on Quantum Yields 46 D. Quantum Yield Studies of the Photolysis of 10 in tert-Butanol 46 1. Photolysis 46 2. Unquenched Quantum Yield Measurements 48 3. Photolysis of 10_ using 1,3-Cyclohexadiene as Quencher in tert-Butanol 48 4. Photolysis of 10_ in tert-Butanol: Effect of Changing the Concentration of 10 on the Quantum Yield of Formation of 11 51 E. Interpretation of the Results of Quenching the Photolysis of 10 51 F. Conclusions 58 APPARATUS 61 EXPERIMENTAL 76 APPENDIX 109 BIBLIOGRAPHY H4 - v i - LIST OF GRAPHS, TABLES, SCHEMES AND FIGURES Page Graph 1 22 2 23 3 25 4 26 5 30 6 41 7 42 8 44 9 45 10 47 11 49 12 50 13 69 14 73 15 82 16 84 17 95 18 97 19 99 Table 1 18 2 39 3 48 4 70 5 74 6 75 7 81, 83 8 83, 85 9 86 10 87, 88 11 89 12 90 13 91 14 94, 96 15 96 16 98 17 100 18 i 101 19 102 20 103 21 104 22 105 23 105, 106 - v i i - Page Scheme 1 7 2 .' 7 3 10 4 11 5 12 6 31 7 36 8 52 9 54 10 55 11 56 12 57 F i g . 1 3 2 62 3 64 4 65 5 109 6 HO 7 I l l - v i i i - ACKNOWLEDGEMENT I wish to express my sincere gratitude to Dr. J.R. Scheffer. His cheerful approach to science has been throughout the years a pleasant encouragement to a l l who have had the good fortune to work with him. His continuous interest and h e l p f u l suggestions have made chemistry an exciting f i e l d for many of us. I wish to thank as w e l l a l l the fellow students who made our laboratory such an enjoyable place to work, namely, Dr. K.S. Bhandari, A l i c e Dzakpasu, Barry Jennings, Dr. R.E. Gayler, May Ngan, Dr. R.A. Wostradowski and l a s t but not least Theoharis Zakouras. A deep gratitude i s directed to my wife Charlene who has spent many days and often long hours preparing diagrams and typing t h i s thesis. I wish to also thank Dr. A. Rosenthal for allowing me to make steady use of his Varian 1520 B gas chromatograph, without which my work could not have been completed. My gratitude i s directed as wel l to a l l the technicians especially Joe Shim, of the e l e c t r i c a l department, who have helped keep the research project i n motion. To my Mother fo r her love and wisd and to Charlene f o r everything - 1 - INTRODUCTION A. General Until about twenty years ago, photochemistry was largely a branch of physical chemistry^ Organic chemists depended largely on the Bunsen burner or its equivalent and on the use of catalysts to make or break bonds. The development of the gas chromatograph and new spectroscopic tools allowed organic chemists to study photoreactions which often yielded small quantities of material. Since then the interest in photochemistry has expanded enormously. Organic chemists have discovered reaction pathways which were previously unknown and which afforded in some cases, simple methods for making compounds which were hitherto difficult to prepare. Conventional organic photochemistry relies on the use of light of wavelength in the range of 200 - 400 nm. Molecules that are able to absorb this light contain IT bonds and are excited by 140 to 70 kcal/ mole. Since only TT electron systems are able to absorb in this wavelength range, t;he excitation occurs in specific areas of the molecule called chromophores. There are several types of chromophores, more notably: 1) carbon-carbon double bonds and conjugated polyenes which are able to promote upon excitation a Tr-electron to a TT* antibonding orbital (designated TT - TT*) ; 2) carbonyl containing compounds which can have TT - TT* - 2 - excitation, usually from absorption of 200 - 250 nm light, and have a high extinction coefficient (VL0,000); and 3) n - TT* excitation ( promotion of an electron from a nonbonding orbital on oxygen to the TT* orbital of the carbonyl group) in the range of 270 - 400 nm, depending on sbstitution and the degree of conjugation. The latter are forbidden transitions and have low extinction coefficients (/v/100). The chromophores of the molecules studied in this work were conjugated carbonyl groups. The ground state of nearly any molecule has a l l electron spins paired, has a multiplicity of one , and is thus called a singlet. Absorption of light excites a molecule to an excited singlet state or S£ (spin must nearly always be conserved in an electronic transition). The molecule in an excited singlet state has several avenues open for deactivation. The more important of these are reaction to give products, deactivation to the ground state, and a 4 process known as intersystem crossing. This involves spin inversion and results in an excited triplet state. It is clear that for every singlet excited state there wil l be a corresponding excited triplet state. According to Hund's first rule"* this wi l l have a lower energy than the corresponding excited singlet state. The excited triplet state has also the possibility of reacting to give products and of deactivating to the singlet ground state. Figure 1 represents the different pathways available for deactivation of an electronically excited molecule. It is of great importance to realize that for a triplet state to deactivate to the singlet ground state, a spin inversion must accompany the loss of energy. The selection rules ^ for electronic - 3 - transitions formally forbid this type of transition and i t becomes possible only by the mixing of the states due to molecular perturbations. Figure 1. Jablonski Diagram. Horizontal lines represent vibrational levels. Solid arrows represent absorption or emission of light, wavy arrows represent nonradiative deactivation pathways, and the dotted line represents intersystem crossing. This results in a much longer lived triplet state relative to the -6 —12 singlet state. The singlet has a lifetime of 10 to 10 sec -3 -9 whereas the triplet can have a lifetime of 10 to 10 sec and sometimes can be as long lived as a second or more. One of the most useful and fundamental quantities in the study of photochemical reaction mechanisms is the quantum yield ( $ ). Its value and the influence of the experimental variables upon i t , give important information as to the nature of the reaction. The quantum yield for disappearance of reactant can be defined as the number of molecules - 4 - reacting per photon absorbed. One can have a quantum yield for formation of product, a quantum yield for fluorescence, phosphorescence, intersystem crossing and so on. In general, quantum yields vary between zero and unity. In some cases, however, the quatum yield for disappearance of reactant can exceed unity i f a reactive intermediate consumes starting material. In the case of the photon acting only as a "catalyst" to promote the initiation step of chain reactions, the quatum yield can be extremely large ( e.g. the photolysis of Bv^ in the formation of HBr from H 2 and Br 2 ). If one considers the photoreaction of a compound X , whose triplet state results in product formation,: the simplest scheme for such a reaction is . . -Rate X > x* 1 X * 1 __2 _> x*3 of I X*3 — > X k [ X*3 ] X* —^2 > P k 2 [ X* ] X* 3 + Q —^3 > X + Q*3 k 3 [ X * 3 ] [ Q ] a= efficiency of intersystem crossing; f= efficiency of light absorption; 1= light intensity; Q= triplet quencher. 3 The rate of formation of triplet (X* ) is given by d[ X*3 ]/ dt = afl - [ X*3 ]( k x + k 2 + k 3[ Q ] ) and the steady state approximation gives the expression afl = [ X*3 ]( k x + k 2 + k 3[ Q ] ) The quantum yield of product ( P ) formation in the presence of quencher is * p = k 2 t X* 3 ]/ f l = ak2/ ( k x + k 2 + k 3[ Q ] ) in the absence of quencher i t is $ = akj ( k, + k ) o L i Z - 5 - The r a t i o of these quantum yi e l d s i s the Stern-Volmer expression If one plots * / $p vs. [ Q ] one w i l l obtain a straight l i n e ( i f the mechanism i s correct ) with a slope k^/ ( k^ + k^ ), i . e . the r a t i o of quenching rate constant to the sum of the rates of a l l t r i p l e t deactivation processes ). If the quenching i s d i f f u s i o n controlled . then an approximate value can be assigned to k^ depending on the solvent used by applying the s i m p l i f i e d Debye expression^ k ( d i f f u s i o n controlled ) = 8RT/ 3000n( l i t e r mole~ 1sec~" L ) where n i s the v i s c o s i t y of the solvent i n poise. One can then obtain a value for the t r i p l e t l i f e t i m e T = 1/ ( k^ + kj )• B. Photochemistry of Various Tetrahydro-l,4-naphthoquinones. The study by Cookson et a l ^ on the photoreaction of compound 1̂  revealed a di f f e r e n t behavior than expected from comparison to s i m i l a r structures _3 which gave r i s e to only cage products ji * Q/ $ p = 1 + k 3[ Q ]/ ( k L + k 2 ) + tar ( i ) 3 (.n- 0.1.2.) 4 - 6 - This unusual behavior by 1_ intrigued our research group who set out to reinvestigate the reaction."'""'"Photolysis of compound 1̂  using a f i l ter transmitting light of X> 340 nm led to the discovery of two new products 5_ and j5. 5 6 benzene 1 : 7 tert-butanol 5 : 1 The proposed mechanism for the formation of these products is presented in Scheme 1_. This involved the novel hydrogen abstraction by the carbonyl oxygen through a five-membered transition state to give a bis-al lyl ic biradical _7. Bonding at different termini of the a l ly l ic radicals leads to the formation of enols 9_ and 8_, which upon ketonization yield the observed products 5_ and _6. The successful investigation of compound 1̂  led our research team to study the effect of substituents on the photochemistry of the tetrahydronaphthoquinone ring. In the case of the 6,7-dimethyl-4a3,5,8, 12 8a$-tetrahydro-l,4-naphthoquinone 10, aside from the products analogous to 5̂  and 6̂ , a new product 13 was observed (Scheme 2). This new product formed from the photolysis of 10 in benzene, can also be formed through a bis-al lyl ic biradical intermediate as suggested for the unsubstituted case. The photolysis of the enone-alcohol 13_ in tert-butanol gave rise to ene-dione 11, and photolysis in benzene afforded ene-dione \2_> showing the same solvent dependence as the photolysis of 10_. This observation gave added strength to the argument of a common intermediate such as 1A_ in both photoreactions. hv 3j8-bondine , . , , —1 — enol or 11 1^ A - h n n ^ ^ n n ^ enol of 12 1,6-bondlng >- 13 ( * . * • • » • or ) - 8 - The formation of ene-dione VL from the enone-alcohol 1̂  is formally a [3,3]-suprafacial sigmatropic rearrangement, however, this 13 process is not allowed photochemically to be concerted. The formation of ene-dione 12_ from the alcohol 1_3 on the other hand is formally a [l,3]-suprafacial sigmatropic rearrangement. This process 13 is allowed by the Woodward Hoffmann rules. Nevertheless, the study 14 by Cargill et al on compound 17_ suggested that the 1,3 shift of the y-carbon in the ct,$-unsaturated ketone could occur in a nonconcerted mode. Both the thermolysis of the alcohol 13_ and of the ene-dione 12 led to the exclusive formation of ene-dione 11. It is probably the presence of a more substituted double bond which makes dione 11 thetjodynamically more stable. ( 5 ) 13 11 12 The alcohol 13_ can give rise to product 1_1 in an allowed [3,3]- suprafacial sigmatropic rearrangement. However, the thermal reaction - 9 - of dione 12_ to give 11_ may not occur in a concerted manner because 13 this would involve a forbidden [l,3]-suprafacial rearrangement. Experiments using tert-butanol-O-d as a solvent for the photolysis of compound 10_ gave rise to the exo -d eut era ted ene-dione 1_9_. The same product was obtained by the base-catalyzed deuterium exchange of ene- dione 11. Studies by Thomas'''"* , and Werstiuk"'"̂  on base catalyzed deuteration of several methyl substituted bicyclo[2.2.l]heptanones also resulted in the preferential deuteration of the exo position on the carbon l adjacent to the carbonyl cf. 20. These experiments proved the int of the enol 18_ in the photochemical reaction. 0 0 ( 6 ) 10 18 19 0 - 10 - Finally a tetradeuterated form 21_ of quinone 10_ was photolyzed both in benzene and in tert-butanol. Scheme 3 O D 23 24 25 The experiment was designed to test the hypothesis of 8-hydrogen abstraction by oxygen and enol formation. Note that in tert-butanol, the enol deuterium can exchange with the solvent, resulting in a compound with only 3.0 D. This was in fact observed. In benzene, 60% D was found at C^, cf•, 26. This was explained on the basis that the remainder had exchanged for hydrogen due to some moisture present in the benzene. The investigation of the effect of substituents on the photolysis of tetrahydro-l,4-naphthoquinones led to the study of the hexamethyl substituted compound 2 7 I t was anticipated that this compound might 18 react in a manner different from its less substituted analogues. The X> 340 nm photolysis of 27 led to three products (Scheme 4 ). Relative Ratios benzene 0.5 1.0 t̂ BuOH 1.1 1.0 CH3CN 4.0 1.0 MeOH 13 1 dioxane-H00 30 1  - 13 - The structure of the enone-alcohol 28̂  was identified by X-ray analysis. The formation of both products 28_ and 30^ can be rationalized by invoking the intermediacy of a bis-al lyl ic biradical species 31̂  ( scheme 5 ) formed via 8-hydrogen abstraction by oxygen. Similarily, the formation of the ene-dione _29 can be thought of as arising from intermediate 32. 36 This intermediate can be formed from biradical 31^ by a shift of the hydroxyl hydrogen. The intermediate thus formed can close, giving 19 the enol 214_ which upon ketonization, gives product 29• However, a second mechanism (B) can be visualized. This involves a yhydrogen abstraction by a 3 carbon atom on the quinone ring to give intermediate 35. Collapse of the biradical yields the ene-dione .29 directly, without involving an enol. Hydrogen abstraction by a 3 enone carbon atom 20 has been observed by Herz and Nair in the photolysis of 36, and by ( 8 ) - 14 - In the case of cyclopentenone 3_7, this was suggested to occur through a six-membered transition state resulting in the formation of biradical 41_. Closure affords the bicyclic structure 40. Structures 38 and 39_ arise from jtl by a second hydrogen abstraction, but this time by the carbon a to the carbonyl. The reaction was believed to arise via a triplet state based on the evidence of the effect of quenchers and 22 sensitizers. Similarly, Nakanishi, et. a l . found that taxinines such as 42 underwent photoinduced H-abstraction by an ct-enone carbon atom to yield structures like 43 . Once again, sensitization studies suggested that the reaction proceeded via a triplet state. ( 9 ) The photolysis of the tetrahydronaphthoquinone 2_7 in tert-butanol-0-d and in 1:1 dioxane/deuterium oxide showed no incorporation of deuterium in the ene-dione 29. In the latter more polar solvent, ene-dione 3J) was formed containing exactly one deuterium per molecule in the 4 exo position as expected. This evidence thus gave support to the mechanism for ene-dione 2_9_ formation not involving an enol intermediate. Thus, It is likely that the ene-dione 2_9 arises by a mechanism involving y-hydrogen abstraction by the enone carbon (path B), whereas enone-alcohol 28 and ene-dione 3J) arise through in i t i a l B-hydrogen abstraction by oxygen. - 15 - The y-hydrogen abstraction by carbon ( i .e . , transfer of hydrogen from C,. to C^) may be facilitated in the photolysis of compound 27_ by 23 the effect of the bridgehead methyl groups. An X-ray study of 27 revealed that these methyls, to remain staggered, hold rings A and B in close proximity. This arrangement places the abstracted "down" hydrogen at close to the abstracting p-orbital at carbon 2. 2.80A CH 27 Thermolysis of the alcohol 28_ resulted in the conversion to ene-dione 30 and to quinone 27. The former reaction can formally be considered as an allowed [ 3,3 ] - suprafacial sigmatropic rearrangement. 17 24 The latter has been suggested to occur through an oxy-retro-ene reaction (see arrows in equation ( 10 ). C3 3J H ^ V ^ 28 27 ( 10 ) Thermolysis of the ene-dione 29_ resulted in the formation of the naphthoquinone 44. Its mechanism was postulated to be a retro-ene 2 A reac t W (s ee arrows in equation 11 ). - 16 - ( 11 ) The driving force for the reaction is probably in part due to relief of strain of the cyclobutanone ring and the formation of a highly conjugated chromophore. C. Objectives of the Present Research The main objective of this work was to try to elucidate the electronic states involved in the photolysis of the tetrahydr©naph- thoquinones 10 and 27. It was hoped that this knowledge might shed further light on the exact nature of the mechanism for the formation of a l l the observed photoproducts of 10 and 27. In the case of the photolysis of compounds 10 and 27, literature analogy suggested that a singlet state might form the enone alcohols 13 and 28̂  and the ene-diones 11, 12, and 30_ while a triplet state might lead to the ene-dione 2£. Thus Agosta has recently found that B-hydrogen 25 abstraction by oxygen typically occurs from a singlet state whereas y-hydrogen abstraction by a carbon 8 to a carbonyl is commonly a 21 triplet process. In the case of triplet reactions i t was hoped that diffusion controlled quenching might be achieved in order to obtain values for the rates of reaction. - 17 - RESULTS & DISCUSSION A. Quantum Yi e l d Studies of 2,3,4a6,6,7,8af3-Hexamethyl-4a,5,8,8a- tetrahydro-1,4-naphthoquinone(27) i n Benzene. 1. Synthesis and Photolysis. 26 The method of Ansell et a l was followed for the preparation of 27. This material was photolyzed on a large scale ( 2.5 gm/400ml of benzene) using a f i l t e r transparent to X> 340 nm. 0 ( 12 ) The two products formed were isolated by column chromatography to 19 y i e l d 54% of the ene-dione 29 and 27% of the enone-alcohol 2_8. Glpc response curves were obtained for the response of the flame io n i z a t i o n detector to each of the two photoproducts compared to biphenyl used as internal standard. ( 13 ) 27 28 29 - 18 - 2. Unquenched Quantum Yield Measurements. A series of 366 nm photolyses of 0.015 M degassed solutions of 27 in benzene were carried out. A 15 ml portion of the photolysate was combined with a 2 ml of a stock solution of biphenyl ( internal standard) in benzene. Glpc analysis determined the amount of each of the photoproducts formed. The measurement of the amount of ferrous ion produced in the actinometer reference cells yielded information on the amount of light absorbed by the test solution. The unquenched yield ( $Q ) w a s determined for each photoproduct and listed in Table 1 . % Conversion of 27 TABLE 1 Quantum Yield' of Formation of ene-dione 29 Quantum Yield of Formation of alcohol 28 1.4 1.2 1.2 3.6 6.9 13.3 13.3 14.3 0.086 0.084 0.088 0.088 0.094 0.093 0.093 0.089 0.069 0.066 0.066 0.064 0.070 0.066 0.066 0.060 The mean quantum yield for the formation of ene-dione 2_9 was calculated to be 0.089 + 0.003 , and for the formation of enone-alcohol 28 the mean quantum yield was 0.066 + 6.003. The percent conversion of 27 to photoproducts was kept low since both photoproducts are able to absorb 366 nm light. At high conversion percentages, one would thus expect the quantum yield for formation of - 19 - photoproducts to decrease. In addition i t i s conceivable ( although unlikely) that photoproduct 28_ could undergo secondary photolysis to afford ene-dione 29_. In this case the quantum y i e l d of formation of 29 should increase with time. To test whether the observed quantum yi e l d s were due to primary processes, the length of i r r a d i a t i o n was varied ( Table i l ). A ten-fold v a r i a t i o n f a i l e d to have s i g n i f i c a n t effect on the quantum y i e l d of formation of 28_ and _29_. This indicated that complications due to photoproduct absorption and/or secondary photoreactions were i n s i g n i f i c a n t under the conditions employed. 3. Photolysis of 21_ using Piperylene as Quencher. To study the effect of t r i p l e t energy quenchers on the photoreaction of 27_ i t was necessary to locate f i r s t of a l l , the 97 position of the t r i p l e t energy of 27. Barltrop and co-workers were able to observe the phosphorescence spectrum of 45_ and calculated from the position of the 0-0 ban<$, E = 57.8 + 1.2 kcal/mole. 45 . Piperylene ( a 1:1.89 mixture of c i s - and trans-1,3-pentadiene ) was used to help locate the t r i p l e t energy of 27. Piperylene has an average t r i p l e t energy of 58.1 kcal/mole ( E T = 56.9 kcal/mole, 29 Ê , =58.8 kcal/mole ), but can quench systems as low as trans 56.9 kcal/mole. A 0.0154 M solution of 27_ with piperylene ( 0.597 M ) i n benzene - 20 - was photolyzed at 366 nm. This resulted in some quenching of the formation of ene-dione 29 $ = 0.025 ( $Q = 0.089 + 0.003 ) and no appreciable effect on the formation of the alcohol 28_, $ = 0.064 ( $Q = 0.066 + 0.003 ). This indeed suggested that the triplet energy of 27 should l ie above 57 kcal/mole. 4. Photolysis of 27_ Using 1,3-Cyclohexadiene as Quencher. To undertake meaningful studies of the effect of quenchers on photoreactions i t is desirable to know whether quenching wi l l occur at a diffusion controlled rate. If this rate is not approached, then the quenching efficiency may be very low and l i t t l e information may be obtained as to the nature of the photoreaction. If one considers the case of a fast triplet reaction, then the lower the quenching efficiency of the triplet quencher becomes, the closer the reaction wil l appear to proceed through a singlet state. 30 Porter and Wilkinson have suggested that bimolecular triplet energy transfer, exothermic by more than 3 kcal/mole, is diffusion controlled. 29 For this reason, 1,3-cyclohexadiene (ET = 53.0 kcal/mole) was chosen as a triplet quencher for the photoreaction of 27. A series of 366 nm photolyses were conducted on 0.015 M degassed solutions of 27 in benzene with varying amounts of 1,3-cyclohexadiene. The formation of dimers j>6_ and 47_ of 1,3-cyclohexadiene was observed by glpc. Dimer formation is normally associated with triplet energy transfer to the 31 quencher. Hammond, et a l , reported the formation of three major product dimers 46̂ , 47, and 48_ for the triplet sensitized reaction of 1,3-cyclohexadiene. - 21 - =0 sensitizer ( 14 ) The quenching results (graphs 1 and 2 ) indicate that the formation of the ene-dione 29̂  proceeds through a triplet excited state whereas the formation of the enone-alcohol either via a singlet excited state or via a very short-lived triplet. It is important to note the non-linear effect of changing quencher concentration at concentration levels > 0.1 M on the quantum yield for formation of ene-dione 29_. This type of positive curvature 32 was observed as well by Wagner in the quenching of Y _ m e t nylvalerophone by 2,4-hexadiene-l-ol. Wagner suggested that such an effect indicates quenching is occuring at a rate greater than diffusion controlled. His argument was that at quencher concentrations higher than ^ 0.1 M a significant number of excited state molecules wi l l have, the instant they are formed, a quencher molecule as nearest neighbour. If exothermic energy transfer to the quencher is 100% efficient, that portion of the excited molecules "born" with quencher molecules as nearest neighbours wil l be quenched immediately. These molecules wi l l thus never enter into normal competition between photoreaction and diffusion controlled quenching. The equation suggested for such a energy wi l l be transfered to the quencher during an encounter and u = the fraction of donor molecules which have at least one quencher situation was: * where a= the probability that Graph 1 - 22 - I—(•)—I STERN-VOLMER PLOT FOR THE QUENCHING OF THE FORMATION OF PHOTOPRODUCT 28. H 6 H H D H f—OH H H !—oh - o n h - O - i l HQ—I l - O H H O H l - O - H h - O - h - O H h - O H h - O H , Graph 2 - 23 - molecule as nearest neighbour. When trip l e t energy transfer i s truly diffusion controlled then a = 1. 33 The least squares slope for the quenching curve of ene-dione 29 formation at quencher concentration below 0.08 M was 158 + 70 M ^ ( 99.9 % confidence limit ) , and for the quenching of the alcohol 2]S_ the slope was -0.03 + 10 M - 1. 5. Photolysis of 27_ Using trans-Stilbene as Quencher. trans-Stilbene (E^ = 49 kcal/mole ) was also used as a t r i p l e t quencher for the photoreaction of 27. The purpose of these experiments was to test the hypothesis that 1,3-cyclohexadiene quenching was diffusion controlled, a conclusion that may be drawn i f the quenching curves are the same for both quenchers. For example, Zimmerman3^ found that the rate of quenching of t r i p l e t excited 4,4-diphenylcyclohexenone ( E T = 69 kcal/mole ) by naphthalene ( E = 61 kcal/mole ) did not seem to be diffusion controlled. Only when 2,5-dimethyl-2,4-hexadiene ( Ê , = 58 kcal/mole ) was used did the rate of quenching appear to be diffusion controlled. This conclusion was strengthened by the observation that the quenching rate was not increased when 1,3-cyclohexadiene ( F̂ j, = 53 kcal/mole ) was used as a quencher. 35 trans-Stilbene isomerizes to cis-stilbene upon t r i p l e t excitation. The two isomers are easily detected and well separated in the glpc columns used in this work ( 10' x 1/8" and 3' x 1/8" columns packed with 20 % DEGS on 60/80 Chromosorb W ). trans-Stilbene does absorb some light at the 366 nm excitation wavelength used for photolysis ( e 0.2 ) and so does cis-stilbene ( e 0.6 ). However, photolyses were performed on solutions containing only trans-stilbene in benzene. - 25 - Graph 3 STERN-VOLMER PLOT FOR THE QUENCHING OF THE ALCOHOL PHOTOPRODUCT 28. I—OH h-QH h-OH h<5 I i - e - H [ trans-STILBENE ] ( M ) - 27 - The results indicated that the conversion of trans-stilbene to cis-stilbene by direct absorption of light represented about one fifth of the total of conversion of trans-stilbene to cis-stilbene in the sensitized experiments. Once again the data obtained for the quenching of 27_ by trans- stilbene was plotted in graphs 3 and 4 as the ratio $q / $ versus trans-stilbene concentration. These Stern-Volmer plots show a slight increase in quenching rates by trans-stilbene as compared to 1,3-cyclohexadiene. The slope for the least squares plot ( graph 4 ) of quenching of ene-dione 29̂  was calculated to be 186 + 86 M - 1 ( 99.5 % confidence limit). The slope for the alcohol quenching plot ( graph 3 ) was calculated by least squares to be 0.92 + 5.1 M _ 1 ( 99.9 % confidence limits ). The variation in slopes between the plots for quenching by 1,3-cyclohexadiene and trans-stilbene of the quantum yield of formation of 28_ and 2£ may reflect the effect of absorption of light by trans-stilbene. This conclusion was reached by considering that the formation of enone-alcohol 28_ was totally unquenched by 1,3-cyclohexadiene quencher ( graph ) when triplet energy transfer from Z7 to the quencher was indicated by quencher dimer formation and when the formation of ene-dione 29_.was 97 % quenched. Thus i f 28_ is unquenchable by using 1,3-cyclohexadiene as quencher i t should also be unquenchable when using trans-stilbene. Even so, the slopes of quenching of ene-dione 29̂  by both trans-stilbene ( 186 + 86 M _ 1 ) and 1,3-cyclohexadiene ( 158 + 70 M - 1 ) are within experimental errors. Since the triplet energy separation for the two quenchers is 4 kcal, the comparable slopes suggest that quenching by 1,3-cyclohexadiene is controlled by diffusion. /Q^^4^^i^d ' - 28 - 28 Barltrop in his studies of the photoreaction of 1,4-quinones ( including 45 ) with olefins, used trans-stilbene as one of the olefins. The experiment showed that the 1,4-quinone was deactivated by energy transfer, but no photoreaction between the 1,4-quinone and trans-stilbene was observed. trans-Stilbene would thus not be expected to react chemically with 27_. The fact that the quenching results of 27 are the same for D O t n trans-stilbene and 1,3-cyclohexadiene rules out chemical quenching for either. The Stern-Volmer equation for the plot * / * versus [Q] i s : 0 * - T " - 1 + k qt[ Q ] and the slope is k t ( T - triplet state lifetime ). The bimolecular diffusion controlled rate constant for benzene is 1.0 x 101® M~* sec - 1, ' and assuming this value for the quenching rate constant ( k^) then the triplet lifetime can be calculated from the slope quenching by 1,3-cyclohexadiene, T - 1.58 x 10 ' sec. 6. Photolysis of 27 Using Oxygen as Quencher. Oxygen was used as a quencher in diagnostic tests for the participation of triplet states in the photolysis of 27. Oxygen is a very efficient quencher of triplet states. The purpose of its use was to determine i f the formation of the alcohol photoproduct 28 could be quenched at a l l . Two experiments were performed on 0.015 M solutions of 27 in benzene which were degassed and then repressurized to one atmosphere with oxygen. The concentration of oxygen was 36 calculated to be ca. 0.01 M. The formation of ene-dione 29 was quenched as expected, * - 0.0014 - 29 - ( * = 0.089 ). However, i t is important to note that the amount of quenching was greater than expected by the diffusion controlled rate ( by comparison to 1,3-cyclohexadiene and to trans-stilbene ). It is known that oxygen can succesfully quench the formation of compounds 37 that arise from free-radical intermediates by chemical reaction. It is possible that this may be the reason for the exaggerated quenching effect by oxygen on the formation of ene-dione 29. The formation of the enone-alcohol _28_ was also quenched but to a small extent, $ = 0.043 ( $q = 0.066 ). Again, i t is entirely possible that this quenching does not reflect only quenching of a triplet intermediate but also reaction with a biradical intermediate which is removed by oxygen. The results for oxygen quenching are not conclusive. They reaffirm nevertheless, the observation that the ene-dione 29_ arises from a triplet state. 7. Photolysis of 27_ : Effect of Changing the Concentration of 27. These experiments were designed to determine to what extent i f o p any, the formation of "excimers" (excited dimers ) were responsible for the low quantum yields of formation of the photoproducts of 27. The term "excimer" is used to describe the excited complex formed as a consequence of the interaction of an excited and ground-state molecule. This complex is stable only in the excited state. After deexcitation the two partners repel each other as ground state monomers. The phenomenon of self-quenching via excimer formation has 39 recently been demonstrated in several systems, and the mechanism associated with it is shown below: - 30 - EFFECT OF CHANGING THE CONCENTRATION OF 27 ON THE 9 OF FORMATION OF PHOTOPRODUCTS 28 AND 29. X - 31 - hv + X -> X + X ( deexcitation by fluorescence or by radiationless decay ) ( excimer ) If self-quenching occurs, then a change in the concentration of the photoreactant would result in an inversely proportional change in quantum yield. The results of such experiments are plotted in graph 5 as $q / $ versus concentration of photoreactant 27. Clearly no significant change in the quantum yields of formation of either photoproduct is observed, thus self-quenching is insignificant in this system under the conditions employed. B. Interpretation of the Results of Quenching on the Photolysis of 27. The photolysis of 27 yields both an enone-alcohol 2fS_ and an ene-dione 19 29. The mechanism suggested by Gayler was: (a) g-Hydrogen abstraction of a CQ hydrogen by the adjacent carbonyl oxygen. The o biradical thus formed can bond C. to C, to yield 28. (b) y-Hydrogen I o — abstraction of a Cg hydrogen by C^. Bonding C 2 to Cg of the biradical yields 29. 29 - 32 - The formation of structure types of 28_ was also observed i n the 19 40 -photolysifi-.o£- 49 , 50 , 51, 52, and 5_3. o o On the other hand, the formation of photoproduct type 19_ i s unusual i n that i t i s only observed i n the photol y s i s of tetrahydronaphthoquinones bearing methyl groups at the bridgehead and at carbon atoms 2 and 3, i . e . 27 and 53_. I t i s p o s s i b l e that the formation of the a l c o h o l 2J3 occurs by e x c i t a t i o n of 27_ to an excited s i n g l e t by an n - ir* t r a n s i t i o n . These n - IT* excited systems have been shown to have r e a c t i v i t y s i m i l a r to alkoxy f r e e r a d i c a l s and thus hydrogen a b s t r a c t i o n by oxygen i s a favourable process f o r these states. Studies by C a r g i l l arid coworkers^ on cyclopentenones showed 3 that the lowest t r i p l e t s t ate of 54_ and _56_ i s an ( n,ir* ) state whereas the lowest t r i p l e t s t ate of 55_ and 57_, which have methyl S4- S5 ST(o 51 - 33 - 3 groups on the enone chromophore, is a ( TT,TT *) state. It is thus possible to consider that the lowest triplet level of 2_7 is also 3 a ( TT,TT*) due to the effect of methyl substituents on the chromophore. 27 Barltrop suggested that 2+2 cycloaddition of 45, which has a 3 similar chromophore to 27, originated from ( TT,TT* ) . This was indicated by its lack of hydrogen abstraction from propan-2-ol and by solvent shifts in the uv absorption spectra, ( TT,TT* triplets 42 of carbonyl groups do not abstract hydrogens). One can thus consider the hydrogen abstraction by carbon process in 2_7 which leads to the formation of _29_ to occur via a 3 ( TT,TT*) excited state after n — > TT * absorption by 27. The 3 ( n,Tr* ) state is probably less likely to be populated than the 3 lower energy ( TT,TT*) state of 27. This may be a reason why no triplet products arising from hydrogen abstraction by oxygen are observed. Notably the results are in agreement with the results of earlier investigations. The y~hydrogen abstraction by a 3-enone carbon atom 21 22 was found to be a triplet process. ' Furthermore, Schaffner has shown that hydrogen abstraction by the S-carbon of an a, 8-unsaturated ketone is typical of ( T r , 7 r * ) t r ip le t s^ X-ray studies of various tetrahydronaphthoquinones with and without bridgehead methyl groups indicated that the two rings are 23 tucked close together. Thus from a proximity standpoint i t is possible for a l l such compounds to undergo y-hydrogen abstraction by carbon, since the C„ hydrogen is close to the ir -orb i ta l of the o and C.j carbon atoms. The fact that this is not observed in a l l 3 cases suggests that a ( IT,TT* ) state may indeed be the factor necessary for such a hydrogen abstraction. - 34 - 44 Photolysis of and 58_ does not lead to the formation of the analogue of dione 29. 58 Studies on the position of the triplet energy levels of 1,4-naphthoquinone and 1,4-anthraquinone'^-' revealed that the former 3 has an ( n ,Tr*) as lowest triplet level while that for the latter 3 is a ( TT,Tr* ) . From this standpoint one might argue that 5_8_ should be able to abstract hydrogen by a carbon atom to give the analogue to 29. The fact that this does not occur may reflect the need to break the aromaticity of the system which is an unfavourable process. Considering the formation of alcohol photoproduct 28, Agosta 25 et a l has shown that 8-hydrogen abstraction by oxygen in a-methylene ketones occurs via a singlet state ( equation 15 ). ( 15 ) - 35 - This Is In accord with the present work, Photoproduct 23 has been shown"'-'' to arise via a B-hydrogen abstraction by oxygen and is unquenchable when t r i p l e t energy quenching is demonstrated. The 366 nm irradiation of 2_7 seems most probably to excite the ( n — > TT* ) absorption band of this compound. Uv absorption spectra of 27 in solvents of varying polarity: A^: n-hexane, B^: dichloro methane ( same concentration as A^ ), C^: methanol ( same concentration as A^ ), A^: n-hexane, B^: dichloro methane ( same concentration as A^), C 0: methanol ( same concentration as A^ ). The slight blue shift of the uv absorption band of 27_ centered at 363 nm ( hexane ) upon changing solvents to those of greater polarity, - 36 - and the very low extinction coefficient ( Eg66 nm ^ ^ suggests that this band represents an n > TT* absorption. The band at 285 nm ( hexane ) could be a T - IT* absorption suffering a red-shift in methanol ca. 291 nm, however, the extinction coefficient is very low, e 430 ( hexane). IOD nm The simplest scheme that accounts for a l l the experimental evidence for the photolysis of 27_ is presented in scheme 7: Scheme 7 : o <— Sx s •<- o BR 28 isc -> T, S = -0 S l = T l = BR = BR 29 singlet ground state of 27. singlet excited state of 27. triplet excited state of 2_7. biradical intermediate ( BR 4 BR' ) -> S -> S The quantum yield for the formation of enone-alcphol 2_8 is formation 28 = [ k2 1 < k i sc + k l + I k4 1 ( k3 + k 4 ) ] where the first term represents the probability that the excited singlet state gives rise to the biradical intermediate, and the second term the probability that this intermediate gives rise to photoproduct. - 37 - There are thus several pathways open for the system to deactivate without giving rise to the enone-alcohol photoproduct 2_8. Clearly in this system k^ g c ( the rate constant for intersystem crossing) is of the same order of magnitude as V.^ since the second photoproduct 29_ arises from the triplet state with a quantum yield similar to that of 28. However It is not necessary that k^ should be large to account for the low quantum yields. It is more likely that the low quantum yields are due to the expected facile collapse of the biradical to ground state 46 photoreactant 27. Thus the following conditions may hold: k. > k , » k 1 and k = k J 3 4 1 2 isc The formation of the ene-dione 2j9 can also be accounted for in this manner. The quantum yield formation of 2_9 i s : formation 29 " 1 O K 6 7 ( K 5 + V 1 1 k 8 7 ( K 7 + k 8 ) 1 where a = the intersystem crossing efficiency, and ( ak,. )/(k_ + k,) o 5 o represents the quantum yield for formation of the triplet biradical, and kg / ( k^ + kg) the probability that this biradical collapses to product 29. In the presence of quencher the quantum yield becomes: $29 = [ a k 6 1 ( k5 + k6 + k q C Q ] > ^ [ kg / ( k ? + kg ) ] and the Stern-Volmer equation that results i s : *29 7 $29 = 1 + k q [ Q ] / ( k 5 + k6 > The slope of the Stern-Volmer plot for the quenching of the formation of ene-dione 29 by 1,3-cyclohexadiene was 158 + 70 M 1 and the triplet —8 lifetime was calculated to be ( 1.6 + 0.7 )10 s e c , cf. section A 3. This lifetime is dependent on two independent rate constants k^ - 38 - ( rate constant for a l l triplet decay to the ground state ) and k^ ( rate constant for biradical formation). As before i t is likely that the low quantum yield of formation of the ene-dione is due to an efficient collapse process of the triplet biradical to the ground state. An attempt was made to obtain an independent measurement of the triplet lifetime of 21_ by phosphorescencehowever, the compound did not phosphoresce , and no further information could be obtained. Considering the distinct possibility that k >> k then the rate constant k, can be approximated from the lifetime of the triplet b state x = 1 / ( k c + V., ) to be k, = 6 x 107 sec 3 0 o C. Quantum Yield Studies of the Photolysis of 6,7-Dimethyl-4ag,5,8,8a8- tetrahydronaphthoquinone JLQ. in Benzene. 1. Synthesis and Photolysis. The procedure of Mandelbaum and Cais ^ was followed for the synthesis of 10_ ( eq. 16 ). ( 16) A solution was prepared containing 1.5 g of this material in 400 ml of benzene, and photolyzed for 21 hrs using light of wavelength longer than 340 nm. The two products formed JL2 and 1_3 were isolated by column chromatography to yield 30 % of dione Y2_ and 25 % of alcohol 13. - 39 - 10 hv X > 340 nm 12 13 ( 17 ) Glpc response curves were obtained for each product using 1,4-naphthoquinone as an internal standard. 2. Unquenched Quantum Yield Measurements. These experiments were designed to obtain the quantum yield of formation of the dione 12_ and the alcohol 13. A series of 0.02 M degassed solutions of 10 in benzene were photolyzed at 366 nm for a period of less than 5 hours. The percent conversion was kept low to avoid secondary photoreactions. To test this, the length of photolysis was varied. There was no change within experimental error in the quantum yield of formation of each photoproduct. TABLE 2 % Conversion of 10 Quantum yield of formation of 13 Quantum yield of formation of 12 0.18 0.51 2.7 2.3 0.0149 0.0169 0.0180 0.0158 0.0073 0.0091 0.0085 0.0070 - 40 - The mean quantum yield of formation for the dione 12 was calculated to be 0.0080 + 0.0008 and for the alcohol 13 i t was 0.0164 + 0.0012. 3. Photolysis of 1£ Using Piperylene as Quencher. These tests were performed solely to acertain that the triplet energy of the chromophore of 10_ was >_ 57 kcal/mole. Since the triplet energy of _27_ was credited to be >_ 57 kcal/mole i t 41 was unlikely to expect that the chromophore bearing no methyl groups should have a lower triplet energy than that bearing two methyl groups. A 0.02 M solution of 10 in benzene and containing piperylene at a concentration of 1.27 M was degassed and photolyzed at 366 nm for 4.5 hrs. Glpc analysis of the solution showed that both photoproducts were quenched, * , 1 0 = 0.0040, *_ fc. = 0.0104. formation J_2_ formation 13_ 4. Photolysis of 10_ Using 1,3-Cyclohexadiene as Quencher. 1,3-Cyclohexadiene ( Ê , = 53 kcal/mole )?was used as a quencher, first because piperylene quenching showed that the triplet energy of 10 was above 57 kcal/mole, and second because quenching of 27_ by 1,3-cyclohexadiene appeared to occur at a diffusion controlled rate. The quenching studies were performed on 0.02 M solutions of 10 in benzene, degassed and photolyzed at 366 nm. As was the case for the 1,3-cyclohexadiene quenched photolysis of 27, formation of quencher dimers j46 and 47_ was detected. The results are plotted in graphs 6 and 7 . It is directly apparent that more that one excited state is responsible for the formation of 12_ and 13. The modified Stern-Volmer equation that can be 49 used to describe such a process i s : Graph - 41 - STERN-VOLMER PLOT FOR THE QUENCHING OF THE FORMATION OF PHOTOPRODUCT 12. Graph 7 - 42 - STERN-VOLMER PLOT FOR THE QUENCHING OF THE FORMATION OF PHOTOPRODUCT 13- - A3 - *?.(l+^L\/(l+ H"! ( 1 8 ) V kf+ kd j \ kr +kd where $ = quantum yield at infinite quencher concentration. It is possible to obtain from this Stern-Volmer plot an approximate upper limit value for the lifetime of the triplet state of 10. If one considers equation 18 then it is easy to see that the divisor approaches one i f the quencher concentration approaches zero. Thus at this limit one gets the familiar Stern-Volmer equation: — = 1 + ( k [Q] ) / ( k + k ) $ q r d where = T ( lifetime of the triplet state ). k + k. r d The data for quencher concentrations below 0.11 M are presented in graphs 8 and 9 . The least squares slope for the plot of the dione 12_ was calculated to be 4.8 + 6.2 M - 1 ( 99.9 % confidence limits). For the alcohol 13_ the least squares slope was calculated to be 3.0 + 6.2 M ^ ( 99.9 % confidence limits ). These two slopes are very similar in value as they should be, considering that the quenching is occuring for the same triplet state intermediate. Taking the average value of 4 M ^ for the slope, and assuming a diffusion controlled quenching rate ( cf. section A5 ) of 1.0 x 10"^ M * sec ^ in benzene, then T <_ 4-x 10 ^ sec. Thus the triplet state is very short lived. Stern-Volmer plots give at the limiting value [Q] > °° the ratio $q / $ s ( $ g = quantum yield of product formation from the singlet state). Furthermore <J>q = + $g and thus ( quantum yield of product formation from the triplet state) can be determined. In the case of quenching of the formation of dione 12 * o / * g >   - 46 - 2.08 at [Q] = 1.4 M, * = 0.0038 and thus 48% of 12 is obtained via a s singlet intermediate. In the quenching plot of formation of alchol 13^ $ / $ > 1.65 at [Q] = 1.4 M, $ = 0.0099 and thus 60% of 13 arises o s x s — via the singlet state. 5. Photolysis of 10 Using trans-Stilbene as Quencher. trans-Stilbene was used again as a diagnostic test for the assumption that 1,3-cyclohexadiene quenching of 10_ was diffusion controlled and to show that chemical quenching was not occuring. The highest concentration level of trans-stilbene used was less than 0.17 M because of the absorption capacity of the quencher for 366 nm light ( e = 0.2 ). The results are plotted with the data of 1,3-cyclohexadiene quenching at low concentrations in graphs 8 and 9 . The least squares slopes for the points were: (a) for the quenching'of the formation of dione 12, 6.6+11.2 M _ 1 , and (b) for the alcohol 13, 3.9 + M - 1 . Here again i t seems reasonable to assume that 1,3-cyclohexadiene quenching is likely to be diffusion controlled. 6. Photolysis of 10: Effect of Changing the Concentration of 10 on Quantum Yields. The investigation of the effect of changing the concentration of 10 on the quantum yields of formation of photoproducts L2 and 13 showed ( graph 10) that an eight-fold increase in concentration of 10_ had no marked effect. Thus as argued earlier, ( cf. section A ) this was considered to be sufficient evidence to rule out self-quenching as a source for deactivation of excited states. D. Quantum Yield Studies of the Photolysis of 1Q. in £er_t.-Butanol. 1. Photolysis. A solution containing 1.00 g of 10 in 400 ml of 80:20 mixture of tert- butanol and benzene was degassed and photolyzed using light of Graph 10 - 47 - EFFECT OF CHANGING THE CONCENTRATION OF 10 ON THE <!>0F FORMATION OF PHOTOPRODUCTS 12 AND 13. 0 0 o o 4> Q<3 cm •18 0< <1 0 •8 - 48 - X > 340 nm for 20 hrs. One product 11 was obtained in 79 % yield. A glpc calibration curve was obtained for the response of the flame ionization detector to 11_ compared to 1,4-naphthoquinone ( internal standard ). 2. Unquenched Quantum Yield Measurements. The quantum yield of formation of 11^ in the absence of quencher was determined. The solutions were 0.02 M in 10 in 95:5 tert-butanol- benzene. These were compared to runs made in neat tert-butanol. No appreciable effect was observed by the introduction of 5 % benzene. This benzene allowed for easier handling of the solvent since tert-butanol freezes at 2 5 . 5 ° . Once again, changing the percent conversion of 10 to 11 had no effect on the quantum yield of formation of 11. TABLE 3 Solvent used % Conversion Quantum Yield of formation of 11 tert-Butanol-Benzene (95:5) 0.11 0.0084 tert-Butanol-Benzene " 0.22 0.0091 tert-Butanol 1.2 0.0095 tert-Butanol 0.77 0.0069 tert-Butanol-Benzene (95:5) 0.68 0.0075 tert-Butanol-Benzene " 0.74 0.0076 tert-Butanol-Benzene " 0.85 0.0079 The mean value was calculated to be $ = 0.0081 + 0.0008 o — 3. Photolysis of 10 using 1,3-Cyclohexadiene as Quencher in tert-Butanol. The results for the quenching of 0.02 M solutions of 10 by 1,3-cyclohexadiene are presented graphically ( graph 11 ). There appears to be no measurable quenching of the formation of 11. It is - 4 9 - Graph 11 STERN-VOLMER PLOT FOR THE QUENCHING OF THE FORMATION OF PHOTOPRODUCT 11. i-o-H i—CM en UK w 55 W W O • J o u I Graph 12 - 50 - EFFECT OF CHANGING THE CONCENTRATION OF 10 ON THE OF FORMATION OF ENE-DIONE 11 . r-OH H G M H-GM h-oH r-O-fl 00 3 ~1 r-en - 51 - expected, however, that this is not a reflection of the inefficiency of 1,3-cyclohexadiene as a quencher since the triplet energy of 10_ w i l l , i f it shifts in energy, become larger when using the more polar solvent tert-butanol considering that the lowest triplet energy of 10_ is ( n - IT in character. Unfortunately trans-stilbene is not very soluble in tert-butanol and could not be used as a quencher in this system. Oxygen was used as a quencher. The results showed some quenching but were rendered difficult to interpret by the appearance of a new unidentified product, probably from reaction of oxygen with one of the excited states and/or biradical intermediates involved. 4. Photolysis of 10 in tert-Butanol : Effect of Changing the Concentration of 1_0 on the Quantum Yield of Formation of 11. Within experimental error there was no significant effect on the quantum yield of formation of JL1 when the concentration of 10_ was varied eight-fold, ( cf. graph 12 ). Thus again, no self-quenching was indicated. E. Interpretation of the Results of Quenching the Photolysis of 10. The mechanism for the photolysis of 1_0 in both benzene and tert - butanol has been suggested to occur via 3-hydrogen abstraction by oxygen to give the bis-allylic biradical 14. This process as mentioned earlier, ( cf. section B ), has been documented to occur via a singlet 25 state. The work here presented throws light on the possibility that 3-hydrogen abstraction by oxygen may also proceed via a triplet state. The long wavelength uv absorption spectrum of 10_ in solvents of different polarity shows quite clearly a blue-shift when going from nonpolar to polar solvents of the band centered at 365 nm ( hexane ), - 52 - 355 run ( methanol). This band is thus probably an n-TT* transition 50 ZOO Z S O 300 3 S O 4oo Uv absorption spectra of 10 in solvents of varying polarity: A^: n-hexane, C^: methanol ( same concentration as A^ ), k^i n-hexane, ll^'' dichloro methane (twice the concentration of k^ ) C^i methanol ( same concentration as ). 450 k scheme for the observed photoreactions of 10 could be: Scheme 8: k_ k S « - - T, (n-n*)< — o 1 13 10 S 11 o< BR benzene 12 12 + 13 hv ^-butanol k. Sx (n-TT*) ^ 1 k 7 BR 1 >g 12 + 13 * = 0.0042 * = 0.0065 « = 0.0038 * = 0.0099 o o o o -> Zwitterion 11 $ = 0.0081 o - 53 - Again, as in the case of 27, the biradical collapsing to ground state reactant could account for the low quantum yields observed, k̂ >>k̂  and k ^ » k g . No phosphorescence measurements were attempted on 10^, and thus no independent measurement of the triplet lifetime was obtained. The solvent effect is of course very important here. In both benzene and tert-butanol, a l l three products are observed ( cf_. appendix). However, the product ratios are enormously affected by the solvents. In benzene, 11. is only formed as a trace material. In tert-butanol on the other hand, both jL2 and 3 ^ are very minor products. An explanation^ for such an effect is that the solvent tert-butanol somehow is able to stabilize and localize the unpaired electron at CQ resulting from o g-hydrogen abstraction by oxygen. This localizing effect could then result in preferential collapse of the diradical to yield 11_ through - Cg bonding. This argument would also apply in the case of a Zwitterion intermediate. The structures proposed for such a solvation by a polar solvent like tert-butanol of the diradical intermediate or of a zwitterion intermediate are shown below ( structures 60 and 61 ) . 0 0 60 61 The expected kinetics are thus summarized: benzene: k„ *v* k isc > k 4 tert-butanol: k, > k and k. > k. * k_ 4 isc 2 - 54 - The intermediacy of two different biradicals in the formation of 12 and 13_'is suggested rather than a common species ( eg. a singlet biradical) by the shapes of the Stern-Volmer plots for JL2 and 13_ then $ both plots ( o vs [Q] ) would be identical. $ Scheme 9 : 12 + 13 The Stern-Volmer plots for the formation of 3.2 and 13_ reflects the efficiency of collapse of the singlet and triplet excited state of the photoreactant to biradical. Since the two photoproducts arise from the same biradical intermediate the Stern-Volmer plots for both 12 and 13_ should be superimpossable. In actuality the Stern-Volmer plots for 12_ and 13 ( graphs 6 and 7 respectively ) do not overlap at higher quencher concentrations where triplet quenching is almost complete. As was discussed earlier, the suggested reason why the triplet state of 27_ leads to yhydrogen abstraction by carbon to yield 29_ was because the lowest triplet for 27_ was believed to be ( TT ,IT* ) in character. Whereas in 10_ the lowest triplet state is most likely ( n,ir* ) in character and favours 0-hydrogen abstraction by oxygen to yield the observed products. - 55 - It is entirely possible that the bridgehead substituents are also responsible for directing the fate of the photoreactions of 10 and 27, i .e . conformational control of the reaction pathway. This type of control on the mechanism of certain photoreactions was investigated 52 by Alexander in the photolysis of cyclobutaryl ketones and by Agosta' in the photolysis of 62. Scheme 10: Agosta was able to direct the fate of the biradical 63_ by placing substituents at key positions around the ring, thus obtaining either an aldehyde 64_ or a ketene 65_. He suggested that for conformational control argument to apply, the lifetime of the biradical had to be large enough to allow conformational relaxation of the biradical to compete successfully with the hydrogen transfer which leads to products ( scheme 10). The presence of a tert-butyl group at the bridgehead, compound 66, - 56 - led to the exclusive formation of a ketene due to the steric effect of the bulky tert-butyl group preventing the conformer type 63a from forming. On the other hand.location of methyl or methoxy groups at other positions around the ring, compound 67_, or of no substituents at a l l led to the exclusive formation of aldehyde. In this case, again the more stable biradical conformer seems to be favoured. R = CH3 or OCH, 66 67 Similarly 27 may follow a reaction pathway which reflects conformational control due to the effect of methyl substituents at the bridgehead positions. The energy barrier for free rotation of eclipsing methyl groups in n-butane is 4.4 - 6.1 kcal / mole compared to 3 kcal / mole for e t h a n e . T h i s is a 1.4 - 3.1 kcal / mole energy difference for this system. One can see that free rotation about the bridgehead bond of 27, which is not directly comparable to free rotation of n-butane, is going to be hindered nevertheless to a larger extent than free rotation about the bridgehead bond of 10, which has no bridgehead substituents. - 57 - Scheme 12: R = H only The conformational requirement for the formation of both dione 11 and 12_ from 10_ is a " ring-flip " of the biradical formed after 3-hydrogen abstraction by oxygen ( cf_. scheme 12). This " ring-flip " involves the rotation of the bridgehead bond with concomitant eclipsing of the bridgehead groups. In the case of 27 the methyl substituents at the bridgehead may well suffice to sterically hinder such bond rotation. If this occurs, then the formation of structure types 11 and L2 are not possible for compound 27, and indeed these are not observed. On the other hand the formation of the enone alcohol 13_ can be achieved without the " ring-flip " of the biradical ( cf. scheme 12). In the same manner, the biradical formed after 3-hydrogen abstraction by oxygen in 27_ can collapse directly without " ring-flip " to yield the observed enone-alcohol photoproduct 28. Finally, the formation of dione 2_9_ can be understood without invoking the bridgehead bond rotation ( cf_. scheme 11). The carbon atom abstracts the Cfl hydrogen closest to i t and the biradical thus formed can readily collapse by bonding the C_ and CQ carbon atoms z o to yield the observed photoproduct 29. F. Conclusions. The main observation of this work was that the photolysis of compound 27_ led to the formation of products _28_ and _29_ via totally different states. Photoproduct 28̂  arose from a singlet excited state and product 29̂  from a triplet excited state. This discovery added weight to the suggested mechanisms for the formation of each product. Product 28_ was thought to proceed via 3-hydrogen abstraction by oxygen to yield a biradical which upon 25 collapse formed 28. Literature research revealed that this type of hydrogen abstraction had been attributed to singlet states. On the other hand, product 2j9_ was thought to be produced via y~bydrogen abstraction by carbon. Again, earlier researchers^'^Iiave observed that this process occurs via a triplet excited state. It is interesting to note that the photolysis of 10_ in benzene to yield 12 and 13 proceeds apparently via both a singlet and a triplet excited state. The mechanism suggested for the formation of 12 and 13_ was through a 3-hydrogen abstraction by oxygen process, known to occur from only a singlet excited state. Furthermore, the photolysis of 10_ in tert-butanol yields JUL only via a singlet excited state. The mechanism here again is a 3-hydrogen abstraction by oxygen. No reason could be given why the solvent used should have such an effect on the rate of intersystem crossing of the excited state of 10, although changing solvents from benzene to tert-butanol for the photolysis of 27 is known to decrease the rate of intersystem crossing ( Ratio of - 59 - 28 0 .5 ~ in benzene is ^ — , in tert-butanol i t is - J ^ J ) . Finally and not least, an explanation was presented for the observation that 2_7_ did not form upon photolysis, structures of the type of 11_ and 12 . The explanation involved the conformational control due to bridgehead substituents, of the biradical intermediate produced from 8-hydrogen abstraction by oxygen. The argument presented was that to obtain products similar to 11. a n d 12_ bridgehead bond rotation would have to occur. This process was sterically impeded by the bridgehead substituents. The only products observed for 27 are those not requiring such a bond rotation. Triplet sensitization studies would certainly help to disprove any triplet intermediacy in the formation of alcohol 2B_ from 27. Likewise in the photolysis of 1Q_ * n tert-butanol, triplet sensitization would show whether the product 11^ observed is truly obtained from only a singlet state. The main problem of using sensitizers is that a l l the common triplet sensitizers,having a triplet energy high enough to sensitize either 10 or 2_7,absorb light in the same region as 10 and 7J_, There is however, a solution to this problem. Sensitization can also be performed by thermolyzing tetramethyldioxetane 68_ ( equation 19 ) . a C« 3 CH '3 68 - 60 - The thermal ( 6 8 ° ) decomposition of 68̂  yields in 50 % a triplet excited and a ground state acetone molecule^"* leaving 0.5 % of the cases for a decomposition to a singlet excited and a ground state molecule. Another interesting experiment would be to influence the rate of intersystem crossing in the photolysis of JLO and 27_ in benzene. It is known that the presence of heavy atoms in the solvent or in the molecule itself can increase the rate of spin inversion of the excited state of that molecule via spin-orbit coupling of the heavy atom 56 nucleus with the electronic system. By increasing the rate of intersystem crossing from the excited singlet state to the excited triplet state of 27_ there should be an increase in the rate of product formation 29/28. In the photolysis of 3-0_ the yield of product 1_2 and 13_ arising from the triplet excited state wil l increase in the presence of heavy atoms. The effect can be measured from the Stern-Volmer plot ( _o_ ^ g ^ ). The limiting quantum yield $m at high quencher concentration wil l reveal that fraction of excited 10 molecules s t i l l converting to 12_ and L3 through the singlet state. j - 61 - APPARATUS A. The Quantum Yield Apparatus: The U.B.C. Blue Box The system used to determine a l l the quantum yields reported in this thesis stems from a similar unit used by Zimmerman,̂ 7 "The Wisconsin Black Box". Our System is designed for small scale photolysis ( 27 ml cells were used ) whereas the unit used by Zimmerman allowed the photolysis of 100 ml or more of solution. The light source used was a Bausch and Lomb SP-200 housing fitted with an Osram HBO 200 (200 watts) super high pressure mercury lamp. A Bausch and Lomb 200-700 nm monochromator with 1200 grooves/mm grating was used to select out the desired wavelength, the band of light allowed through the monochromator being no larger than 5 nm. The monochromator was fitted with a variable focal length quartz-fluoride condenser lens having a leaf type diaphragm. The photolysis cells used were a matched pair of 10cm x 2cm quartz cylinders and one 5cm x 2cm quartz cylinder (all three from Hellma )• The outer surfaces of the cells , except the end windows, were carefully 58 59 silvered , and then coated with black epoxy paint . The cells were arranged such that one 10 cm test solution cell and the 5 cm reference cel l were in line with the light source and monochromator, roue*. I J Figure 1'.' Quantum Yield Apparatus - " U.B.C. Blue Box" 1- Osram HBO 200 super high pressure mercury lamp 7- 2" x 2" x 1/16" quartz plate 2- Aluminum plate used to block light during lamp 8- light baffles warm up ^_ c m x 2 cm quartz ce l l ( test solution cell) 3- Monochromator entrance s l i t 5.36 mm , « r n -,-,/<? * ,-.\ 10- 5 cm x 2 cm quartz cel l ( 5 cm reference cell) 4- 1200 grooves/mm diffraction grating i i - i r . o , , „ c 0 11- 10 cm x 2 cm quartz cel l ( 10 cm reference cell) 5- Monochromator exit s l i t 3.00 mm 6- Achromatic quartz-fluorite condenser - 63 - and 16 cm from the condenser lens. At this distance the light beam comes to a focus 5cm into the 10 cm ce l l . The second 10 cm ce l l ( reference cel l ) was placed at right angles to the light beam. A 2" x 2" x 1/6" quartz plate was placed in the light path at a 45° angle in such a way that some of the incident light was reflected into the 10 cm reference ce l l . Each of the 10 cm cells was equidistant from the quartz plate. A box completely enclosed the apparatus, including the condenser lens but not the monochromator or light source. The box was designed to have three compartments: (a) containing the condenser lens, quartz reflecting plate and light baffles to eliminate stray light entering the cells, (b) containing the 10 cm test and 5 cm reference cells on a movable stage, (c) containing the 10 cm reference ce l l , also on a movable stage. Magnetic stirring motors were placed in compartments (b) and (c) to stir the solutions in the rear quarter of the 10 cm cells. The magnetic bars used were 1 x 0.2 cm and their speed of rotation was controlled by variable resistors located outside the photolysis box. The solutions to be photolyzed were thoroughly degassed before photolysis by the freeze-pump-thaw method. For this purpose a 25 ml round bottomed flask was adapted to the 10 cm quartz ce l l as shown in figure ( 3 ) . The solution to be photolyzed ( 26.8 ml ) was introduced into the 25 ml round bottomed flask ( f ig . 3 ). A l l the joints were greased at their outer extremity by Apiezon N grease ( such that no grease would find its way into the solution ). The apparatus was then assembled. The round bottomed flask was immersed in liquid nitrogen for 15 minutes, vacuum (0.05 mm Hg) was applied in the cel l and then purged four times with argon. The liquid nitrogen was removed and - 64 - the solution allowed to thaw. The solution was then frozen again and the cycle repeated. Four such cycles were considered sufficient to remove the oxygen from the ce l l . Figure 3 . Cell system for degassed solutions (a) 10 cm x 2 cm quartz cylinder (b) 25 ml Pyrex round bottom flask (c) connection for vacuum manifold - 65 - Figure 4 represents the emission spectrum of the mercury light source used for quantum yield measurements. The very intense 366 nm mercury resonance line was selected for a l l the work described. /OO Zooo 4Qoo 6°°° Angstrom Units dOoa /otooo Figure 4. Emission spectrum of the Osram HBO 200 super high pressure mercury lamp. Adapted from J.G. Calvert and J.N. Pitts, Jr, "Photochemistry", John Wiley & Sons, Inc., New York (1967),p.704. - 66 - B. Actlnometry The basic requirement of quantum yield measurements is knowledge of the amount of light that entered and was absorbed in the reaction solution. This can be determined in several ways: (a) Measure with the actinometer the light intensity before and after photolysis, then average those values to determine the incident light on the reaction solution. (b) Monitor the amount of light entering the reaction solution in the test ce l l by using a quartz plate beam splitter which reflects a known amount of light into an actinometer solution in the 10 cm reference ce l l . (c) Carry out the procedure described in (b) but also measure with actinometer in both test cel l and 10 cm reference ce l l , the light intensity before and after photolysis to determine the splitting ratio of light by the beam splitter. (d) One can perform any of the above methods using calibrated phototubes instead of actinometers. The work described herein was done using procedure (b). The beam splitting caused by the quartz plate was determined and then assumed to remain constant. Phototubes afford simplicity of operation once calibrated but they cost more than actinometers. The actinometer used for a l l the work was potassium ferrioxalate.^ This is a very sensitive actinometer and its usefulness ranges from 250 - 509 nm.^ The actinometer reacts photolytically as follows:^ 67 - [ F e m ( C 2 0 4 ) 3 ] - 3 - ^ > C 2 0 4 ~ + [ F e n ( ]~2 ( 2 0 ) C 2 0 4 " + [ Fe I I ] C ( C 2 0 4 ) 3 ] - 3 — * • ( C 20 4) 2 + [ F e i n ( C 2 0 4 ) 3 ] " 2 ( 21 ) [ Fe I ] C I ( C 2 0 4 ) 3 ] - 2 * [ Fe I T ( C ^ ] " 2 + 2C02 ( ^ } The amount of ferrous ion produced can be measured when an aliquot of the photolysis mixture is combined with 1,10-phenanthroline and the absorbance measured at 510 nm. A l l actinometry work was done under Kodak OB safelights and a ruby red lamp. Potassium ferrioxalate [ K3Fe( C 2 0 4 ) 3 .3 H20 ] was prepared by the method of Hatchard and Parker.^ A solution of 600 ml of 1.5 M potassium oxalate ( reagent grade) was mixed vigorously with 200 ml of 1.5 M ferric chloride ( reagent grade). Beautiful green crystals were obtained. These were recrystallized three times from warm water and dried in an oven at 50° for 48 hours. Safelights were used during recrystallization. The solution used for actinometry was 9.82 gm (0.0999 moles/liter) of potassium ferrioxalate in 200 ml of 0.1N H2S04. The solution was kept sealed and in darkness. 61 A modification by Kurien involved adding acetate buffer and 1,10-phenanthroline solution to the solution containing the actinometer before photolysis. This simplified the procedure of analyzing the actinometer after photolysis. The solution could simply be diluted a proper amount and measured directly in the spectrophotometer. However, i t was found that the results were lower in value and less consistent than those obtained by the method of Parker. It could be that due to the fairly substantial amount of light absorbed during photolysis, the - 68 - ferrioxalate became somewhat depleted and that the 1,10-phenanthroline may have absorbed some of the incident light. Kurien's method was not adopted. The solution showed l i t t l e ferrous ion formation in the first four days after preparation. The solutions used were never older than this. In preparation for photolysis a 27.0 ml aliquot of solution was introduced into the 10 cm reference cel l and a 13.5 ml aliquot introduced into the 5 cm reference ce l l . Both cells were then placed in the photolysis box. There seemed to be no need to degas the solution since the results were comparable for degassed and non-degassed solutions. After photolysis the solutions were diluted sufficiently with 0.1N ^SO^ t o S-*-ve a n absorbance in the range of 0.3 to 0.7. A portion of this diluted solution was then combined with acetate buffer (600 ml lN-sodium acetate and 360 ml IN ^SO^, to 1 l iter) and 1,10- phenanthroline monohydrate in water (0.1 %) in a ratio 5:3:2 respectively. The solution was stirred and allowed to stand for one hour, then i t was analyzed in a Cary Model 15 spectrophotometer at 510 nm. C. Calibration of the Cary Model 15 Spectrophotometer response to I | Fe -Phenanthroline complex. A stock solution containing 1.1203 gms (4.030 mmol) of FeS0^,.7H20 made up to 250 ml with 0.1N ^SO^ w a s P r e P a r ed . A 25 ml aliquot was diluted to 1 l i ter with 0.1N H„S0. . A series of eleven 20 ml solutions 2 4 were prepared by combining from 0 to 5.0 ml of the diluted ferrous sulfate solution with 4 ml of 0.1% 1,10-phenanthroline monohydrate in water, 6 ml of acetate buffer and topped with 0.1N J^SO^. The solutions were well mixed, allowed to stand for at least one hour and then analyzed at 510 nm on the Cary 15. Fe CONCENTRATION ( moles/ml ) - 70 - TABLE 4 Concentration of Fe Absorbance 510 (nm) 0.110 0.222 0.333 0.446 0.546 0.664 0.775 0.897 0.986 1.106 The data is plotted on graph 13. The least squares slope for the points is 1.099 x 10^ ml/mole and the standard error for the slope is 1.3 x 10^ ml/mole. The callibration of the machine was carried out at six month intervals, however, the variation in the slope never exceeded the experimental error . D. Determination of the Percentage Splitting Caused by the Quartz Plate Beam Splitter. It was necessary to know the amount of light actually reflected by the Quartz plate in the quantum yield apparatus. For this purpose a series of experiments were performed in which a l l the cells were f i l led with 0.0999 M K^Fe ( c 2 ° 4 ) 3 « The mercury lamp was turned on but the beam block (2) Fig. was left in place for thirty minutes to assure that the lamp had warmed up and the arc stabilized before photolysis was permitted. (mole/ml) 1.007 x 10" 2.015 x 10" 3.022 x 10" 4.030 x 10" 5.037 x 10" 6.044 x 10' 7.052 x 10' 8.059 x 10' 9.067 x 10" 10.07 x 10' A ten minute 366 nm photolysis was performed. The solutions were stirred during photolysis and for another fifteen minutes after. The solutions were diluted with O.lN H2SO^ except for the 5 cm reference solutio^and the required amount of 1,10-phenanthroline solution and buffer was added, stirred, allowed to stand for one hour, and measured on the Cary 15 at 510 nm. The results were: 7.76% ,7.88 , 7.72 , 7.92 , 7.57 , 7.62 , 7.84 , 7.80 , 8.07 , 7.75 , 7.99 . The mean value was 7.81 % and the standard error 0.14 %. E. Formula for Calculating the Number of Quanta Absorbed by the Test Solution. For each occasion that actinometer solutions were measured for ferrous ion content, the ferrous ion content of the unphotolyzed stock solution was measured. No. of Einsteins absorbed by the solution photolyzed in the test ce l l [ V x ( d ^ - AQ) 11.8 ] - [ V, ( d ^ - AQ) ] = _ I x e „++ x 10 x $ ++ Fe Fe The first term represents the amount of light incident on the test solution ce l l and the second term represents the amount of light not absorbed by the test solution. Vj= volume of the 10 cm reference ce l l (27.0 ml) d̂ = dilution factor for absorbance measurement of the 10 cm reference cel l actinometer solution A =̂ absorbance of the 10 cm reference cel l actinometer solution. Aft= absorbance of the unphotolyzed solution. - 72 - 11.8 = factor derived from the % light reflected into the 10 cm reference cel l [light entering test cel l _ (light entering 10cm ref.cell)(1-0.0781) 0.0781 V" = volume of the 5 cm reference cel l (13.5 ml) d^ = dilution factor for absorbance measurement of the 5 cm reference cel l actinometer solution = absorbance of the 5 cm reference cel l actinometer solution I = path length of the cel l used in the spectrophotometer ( 1cm) £ _ , + + = extinction coefficient of ferrous ion at 510 nm Fe ( 1.099 x 104 M _ 1 cm"1 ) $ 4+ Fe = quantum yield for the formation of ferrous ion at 366 nm ( 1.15 is the value adopted here. This value was found 60 64 by Hatchard and Parker, however Lee and Selinger found a value of 1.20 ) F. Cary 15 Response Calibration Curve to Benzophenone Concentration at 342 nm. To determine the accuracy of the quantum yield apparatus, a series of experiments were designed to establish the quantum yield of 0.1M benzophenone with 0.1M benzohydrol in benzene. To this end a calibration curve of response of the Cary 15 spectrophotometer to benzophenone concentration was required. Benzophenone has a X at 342 nm, and i t max was this absorption that the Cary 15 was callibrated to. Two stock solutions were made: (a) 0.1998 M Benzhydrol (Aldrich reagent, twice recrystallized from ethanol, mp 65.5 - 6 6 . 0 ° ) to 250ml in benzene ; (b) 0.2201 M Benzophenone (Aldrich reagent, twice dist i l led, mp 46.5 - 47.0° ) to 100 ml in benzene. To each of six 50 ml volumetric CONCENTRATION OF BENZOPHENONE ( M ) flasks, 25 ml of the benzhydrol solution ( 0.0999 M ) and 0 to 25 ml of the benzophenone solution was added, then topped with benzene (spectro grade). Then 3 ml aliquots of these solutions were diluted with 45 ml of benzene and then analyzed in the uv machine. TABLE 5 Concentration of Benzophenone (M) in the diluted 48 ml solution 6.872 x 10 5.502 x 10 4.127 x 10 2.751 x 10" 1.372 x 10 0 -3 -3 -3 -3 Absorbance (342 ml) o.919 0.730 0.543 0.362 0.180 0 The points are plotted on graph 14. The best f i t to these points was. a least squares slope of 132M "*cm ^ This of course, represents the extinction coefficient of benzophenone at X 342. max G. Quantum Yield of 0.1 M Benzophenone with 0.1 M Benzhydrol in Benzene. Four experiments were made. One of these four involved using 25 ml of the 0.1998 M solution of benzhydrol and 25 ml of the 0.2201 M benzophenone solution from the calibration runs ( previous section ). The other three experiments used two new stock solutions: 100 ml of 0.2000 M benzhydrol in benzene ; 100 ml of 0.2000 M benzophenone in benzene. - 75 - These were combined i n equal parts as we l l . The 26.8 ml of each of these solutions were placed i n the c e l l , degassed and then photolyzed for f i v e hours at 366 nm. TABLE 6 Absorbance Mole Benzophenone Unphotolyzed Photolyzed Reacted Solution Solution (m mol) 0.922 0.813 0.837 0.711 0.841 0.726 0.839 0.728 The mean value for i s 0.69. This value i s 65 by Hammond et a l 0.67 0.351 0.406 0.371 0.358 Light Quantum Absorbed Yi e l d for mEinstein Benzophenone disappearance 0.492 0.71 0.580 0.70 0.535 0.69 0.542 0.66 the quantum y i e l d of benzophenone disappearance i n good agreement with the values obtained. 66 and by Moore et a l 0.68. - 76 - EXPERIMENTAL A. General Infrared (ir) spectra were recorded on a Perkin-Elmer Model 137 spectrometer, using sodium chloride cells . Nuclear magnetic resonance ( mar ) spectra were recorded on the Varian Model T-̂ 60, HA-100, and XL-100 by Ms. Philis Watson and Mr. William Lee of this department. TMS was used as an internal standard in a l l cases. Melting points were determined on a Fisher-Johns melting point block and are a l l uncorrected. Ultraviolet ( uv ) spectra and visible measurements were recorded on a Cary Model 15 recording spectrophotometer. I | The 510 nm absorption of the a-phenanthroline-Fe complex was measured with wavelength control approaching 510 nm from higher wavelength. The gas liquid partition chromatography (glpc) other than quantum yield measurements were done on a Varian Aerograph Model 90P and Varian Aerograph Autoprep Model 700. Both were connected to Honeywell Electronik 15 strip chart recorders. For a l l glpc operations involved in the measurement of quantum yields, a Varian Aerograph Model 1520B with a flame ionization detector was used. The carrier gas was helium. Pure grade air, hydrogen and oxygen were used to combust the materials isolated by glpc. The oxygen was fed into the hydrogen line at a point close to the detector, using a Y-tube. Each line was fitted with one-way valves opening under a pressure of 1 psi. The use of - 77 - oxygen^ improved the s e n s i t i v i t y of the instrument almost threefold. The glpc was connected to a Honeywell Electronik 15 s t r i p chart recorder with a 1 mvolt f u l l scale s e n s i t i v i t y and f i t t e d with a Disc Chart Integrator Model 201-B. The columns used for a l l glpc measurements of the quantum yiel d s were: (a) 10' x 1/8" stainless s t e e l packed with 20% DEGS on 60/80 Chromosorb W (this material was c a r e f u l l y f l u i d i z e d and then sieved to obtain a homogeneous support. The column was packed as a straight pipe under 45 p s i pressure. The column was then bent to the desired shape after packing.). The column temperature was kept at 135° and the injector and detector at 190°. The c a r r i e r gas flowed at 30 ml/ minute ; (b) 3' x 1/8" stainless s t e e l packed with 20% DEGS on 60/80 Chromosorb W i n the same manner as (a). The column operated at 145° and the in j e c t o r and detector at 190°. The He c a r r i e r gas flowed at 30ml/ minute. Column (a) was used for the i s o l a t i o n of the photoproducts of tetrahydronaphthoquinone 2_7 and column (b) for the i s o l a t i o n of the photoproducts of tetrahydronaphthoquinone 10. Biphenyl (Aldrich reagent grade, twice r e c r y s t a l l i z e d from ethanol, mp 67.8 - 68.0°) and 1,4-naphthoquinone ( K & K Labs., Inc., r e c r y s t a l l i z e d from pertoleum ether (68°), decolorized with carbon, r e c r y s t a l l i z e d again, mp 122-123°) were used as i n t e r n a l standards for the glpc measurements. Internal standard was added after photolysis. A 15 ml portion of the photolysis mixture was combined with 2 ml of i n t e r n a l standard stock solution. This mixture was then immediately analyzed by glpc. The two dif f e r e n t i n t e r n a l standards used were selected for t h e i r glpc retention time, such that t h e i r peaks would not overlap with any other peak expected and yet have a retention time close to that - 78 - of the photoproducts studied. Pipets and volumetric flasks were used for a l l the measurements of volume. For a l l quantum yield measurements, spectro grade benzene was used. For those runs using tert-butanol, reagent grade material was used. A 95:5 mixture of tert-butanol-benzene was dried through a column packed with molecular sieves Linde Type 4A 1" x 16" mesh. - 79 - Preparation of 2,3,4af3>6,7,8aB-Hexamethyl-4a,5,8,8a-tetrahydro -1,4-naphthoquinone (27). 26 The method proposed by Ansell et al was followed. A slurry of 3.200gm (19.5 mmol) of duroqulnone [( prepared from durene according to 68 the method of Smith , yield 81%, recrystallized from petroleum ether ( 68°C) , mp 112 - 112.5° )], 4.0 gm (48mmol) of 2,3-dimethyl-l,3- butadiene (Aldrich, 98%),and a few crystals of hydroquinone were sealed in a Pyrex tube and heated for 27 hrs. at 197°C. The resulting pale yellow solution crystallized on cooling. The material was recrystallized from petroleum ether (68°) to give 4.593 gm (18.6 mmol, 96% ) of faint yellow crystals of the desired quinone. The material was recrystallized four more times from petroleum ether (68°) which 26 produced a material with a mp 114.5-115.5° ( l i t . 115-117° ) ; i r (CCl^) 5.98 (C=0)u ; nmr (CCl^) T 7.3-8.4 (m,4,methylene), 8.0 (s,6,C. and C„ methyl), 8.4 (s,6,C, and C, methyl), 8.9 ( s,6, bridgehead 2 j o I 4 methyls); uv (hexane) ( ^1-20 x 10 ), 275 - 340 nm, broad shoulder ( e „ o r 430), 345 - 475 nm, broad featureless absorption 2o5 nm ( e366 6 5 ) ' 17 Large Scale Photolysis of 27_ in Benzene. A solution of 2.5 gm ( 10.1 mmol) of 27_ in 300 ml benzene (reagent, distilled) was irradiated through a Corning glass f i l ter No. 7380 (transmitting light of X>340 nm) using a 450 W Hanovia type L medium pressure mercury lamp fitted in a water cooled quartz jacket. Glpc (20% DEGS on 60/80 Chromosorb W as solid support in a 5" x 1/4" column) was used to follow the progress of the photolysis. The solution was - 80 - degassed for 30 minutes with argon before photolysis, and a positive argon pressure maintained during photolysis. Irradiation for 4.3 hrs was sufficient to convert ca. 95% of the starting material to two products. The photolysis mixture was concentrated to a yellow o i l , and the compounds separated by column chromatography [15" x 1" column charged with 125 gm of Silica Gel (less than 0.08 mm) from E. Merck AG, 10% ethyl acetate/benzene as eluant was used and the passage of material assisted by a positive pressure of 5 - 10 psi nitrogen] 1,3,4,6,8,9-Hexamethyltricyclo[4.4.0.0' ]dec-8-ene-2,5-dione 29 was isolated as a pale yellow o i l . Two Kugelrohr distillations at 80° and 0.02 mm Hg gave a colorless o i l ( 1.34 gm, 5.4 mmol, 54% yield); i r (CCl^) 5.67 y ( C=0,4 membered ring ), 5.85 u ( C=0,6 membered ring ); nmr (CCl^) x 7.57 ( q,l,J=7.5 Hz, C^ methine ), 7.90 - 8.10 (m,3,C^Q methine and methylene), 8.25 - 8.40 (m,6, vinyl methyls), 8.78 (s,3,methyl), 8.95 (s,3,methyl), 8.95 (d,3,J=7.5 Hz, methyl), 9.03 (s,3,methyl); uv X (CC1.) 256 nm ( e 6.3 x 102 ), 295 nm J max 4 ( e 110 ). The spectral data is identical to that reported by Gayler, et. a l . 1 ^ 5 9 1,3,4,6,8,9-Hexamethyl-5-hydroxytricyclo[4.4.0.0 * ]deca-3,7-dien- 2-one 2J3 was isolated as a colorless o i l which crystallized readily. Recrystallization from petroleum ether (68°) yielded 0.67 gm (2.7 mmol, 27% yield) of the alcohol, mp 101-102° (litl 7mp 101-102°) ; i r (CC14) 2.69 p (OH), 5.98 u (C=0); nmr (CC14) T 4.62 (m,l,vinyl), 7.79 ( broad s , l , OH), 8.12-8.16 (m,3,C3 or C 4 methyl), 8.20 - 8.26 (m,6,C3 or C 4 methyl and C methyl), 8.43 (d,l,J=12.5 Hz, one of C i n methylenes), 8.92 (s,3,methyl), 9.03 (d,l,J=12.5 Hz one of C 1 Q methylenes), 9.14 (s,3,methyl), 9.20 (s,3.methyl); uv X (CC1.) 257 nm (e 7.4 x 103), max 4 320 nm (E 46). - 81 - GLPC Response to Photoproducts 29_ and 28_ Calibration Curve. (a) Two stock solutions were prepared. One contained 76.2 mg of ene-dione 29_ in 100 ml of benzene, the other contained 100.7 mg of biphenyl ( internal standard ) in 100 ml of benzene. These two solutions were mixed in predetermined proportions and diluted with benzene to 25 ml to yield the first four entries in Table 7. At this point the ene-dione 29. stock solution was diluted by one half with benzene. Combination of this diluted solution with the internal standard solution in exactly measured proportions and diluted to 25 ml with benzene yielded the rest of the entries of i Table . A l l the solutions were analyzed three times by glpc ( 10' x 1/8", 20% DEGS, column a ). Peak area ratios [(Peak area of ene-dione)/(Peak area of ene-dione +Peak area of Internal Standard)] were measured for each analysis. The three values obtained for each solution were then averaged and plotted against, the true weight ratios ( Graph 15). TABLE 7 Weight of ene-dione Weight of Biphenyl Averaged Peak Area 29 in 25 ml ( tag) in 25 ml ( mg. ) Ratio ( ene-dione _2£ ) (ene-dione 29 + IS ) 15.20 1.01 0.902 7.60 1.01 0.829 5.45 1.01 0.759 3.80 1.01 0.703 3.17 1.01 0.655 3.05 1.01 0.658 Graph 15 GLPC RESPONSE CALIBRATION CURVE FOR PHOTOPRODUCT 2 9 . i H C <U CM bi  (U c + o •H T3 1 on  <U •H ti <D en e- EA  TI O AR  RA  - 83 - ( TABLE 7 continued ) Weight of ene-dione Weight of Biphenyl Averaged Peak Area 29 Ratio 2.67 1.01 0.617 2.27 1.01 0.582 2.18 1.01 0.578 1.80 1.01 0.520 1.36 1.01 0.452 1.25 1.01 0.455 0.900 1.01 0.356 0.638 1.01 0.294 0.450 1.01 0.208 0.325 1.01 0.156 (b) Another stock solution was prepared, this time 74.6 mg of the enone-alcohol 28̂ was diluted to 100 ml with benzene. The internal standard solution used was the same as in section (a), ( 100.7 mg biphenyl in 100 ml of benzene ). Once again the first four entries in Table 8 represent combinations in particular proportions of these two solutions. The rest of the entries represent combinations of the internal standard solution with a two times diluted enone-alcohol stock solution. A l l final solutions were made up to 25 ml with benzene. Graph 16 represents the calibration curve of the enone- alcohol 28. TABLE 8 Weight of enone- Weight of biphenyl alcohol 28 in 25 ml in 25 ml ( mg ) (mg) Averaged Peak Area Ratio ( alcohol 28 ) ( alcohol 28 + IS ) 14.91 7.45 4.99 1.01 1.01 1.01 0.910 0.842 0.775  - 85 - ( TABLE 8 continued ) Weight of enone- Weight of biphenyl Averaged Peak Area Ratio alcohol 28 3.74 1.01 0.730 3.12 1.01 0.675 2.99 1.01 0.679 2.49 1.01 0.645 2.13 1.01 0.600 1.23 1.01 0.464 0.625 1.01 0.306 0.318 1.01 0.166 (c) Finally a standard solution of 30.5 mg of ene-dione 29, 29.9 mgm of enone-alcohol 2J[, and 25 ml of a solution of 38.7 mg of biphenyl in 100 ml of benzene was diluted with benzene 250 ml. The weight ratio of the ene-dione was 0.759 and the alcohol 0.756. This solution was used to check the response of the gc detector by injecting after every analysis of photolysis mixture. The solution was at a l l times kept in the refrigerator to minimize any reactions. A new solution was prepared after every four weeks. Quantum Yield Determinations of the Photolysis of 2,3,43 8,6,7, 8af3-Hexamethyl-4a,5 ,8,8a-tetrahydro-l,4-naphthoquinone 27_ in Benzene. (a) Unquenched experiments: A series of experiments were performed to determine the quantum yield of formation of photoproducts 28_ and 2_9_ in benzene. -2 These consisted of introducing 26.8 ml of a ca. 1.55 x 10 M solution of 27 in benzene into the 25 ml round bottomed flask attached to the cel l ( see apparatus page 64 ). The solution was then degassed and - 86 - then photolyzed at 366 nm for a period sufficient to allow no more than a 15% total conversion, ca. 4 hrs. After photolysis, a 15 ml aliquot of the solution was mixed with 2 ml of a stock solution of 30.3 mg biphenyl ( internal standard ) in 100 ml of benzene. The resulting solution was then analyzed by glpc. Table 9 gives the results for the quantum yield of formation of the ene-dione 29_ and the alcohol 28. TABLE 9 Naphthoquinone 27_ Light % Conversion Quantum Yield Quantum Yield Concentration ( M ) mEinsteins of formation of formation of ene-dione of alcohol 2ji_ ** 29* 0.0161 0.0177 0.0174 0.0154 0.0156 0.0153 0.0153 0.0153 0.0392 0.0389 0.0351 0.0978 0.175 0.343 0.340 0.393 1.4 1.2 1.2 3.6 6.9 13.3 13.3 14.3 0.086 0.084 0.088 0.088 0.094 0.093 0.093 0.089 0.069 0.066 0.066 0.064 0.070 0.066 0.066 0.060 * The mean value for the quantum yield is 0.089 + .003. * * The mean value here is 0.066 + .003. ( The errors expressed are standard errors for the data available.) (b) Photolysis of 27 using Piperylene as Quencher. cis & trans-1,3-Pentadiene (piperylene, K & K Labs, practical grade) was disti l led (explosion hazard) from lithium aluminum hydride ^ behind a safety shield ( bp 42.2° ). A solution of 113.4 mgm ( 0.0154 M) of 27 with 1.8 ml ( 0.597 M ) piperylene in 30 ml of benzene was prepared. As before, 26.8 ml of the solution was degassed and then - 87 - photolyzed at 366nm for ca_. 4 hrs. The result was that there was some quenching of the formation of ene-dione 29_, $ = 0.025, and no quenching of the alcohol 28, $ = 0.064. (c) Photolysis using 1,3-cyclohexadiene as quencher. A set of ca. 0.015 M solutions of compound 27_ in benzene with varying concentrations of 1,3-cyclohexadiene ( Aldrich 99%, twice fractionally dist i l led, bp 8 0 . 2 ° ) in benzene were thoroughly degassed and photolyzed at 366 nm for approximately 2 - 4 hrs. depending on the age of the lamp. TABLE 10 Naphthoquinone 1,3-cyclohexadiene Light Quantum Yield Quantum Yield 27 Concentration Concentration mEinsteins of formation of formation ( M ) ( M ) ene-dione 29 alcohol 28 0.0178 3.94 X i o " 4 0.0348 0.081 0.060 0.0166 3.94 X i o ' 4 0.0353 0.085 0.063 0.0193 3.94 X i o ' 4 0.0333 0.093 0.066 0.0164 7.88 X i o " 4 0.0299 0.080 0.074 0.0154 7.88 X i o " 4 0.114 0.078 0.070 0.0165 1.58 X 10"3 0.0391 0.069 0.061 0.0155 1.58 X 10"3 0.110 0.077 0.077 0.0154 1.58 X i o " 3 0.0960 0.071 0.069 0.0164 3.15 X 10"3 0.0375 0.062 0.061 0.0155 3.15 X 10~3 0.116 0.063 0.067 0.0153 4.73 X i o " 3 0.0574 0.052 0.066 0.0153 4.73 X i o " 3 0.103 0.054 0.093 0.0153 4.73 X i o ' 3 0.100 0.056 0.070 0.0156 5.52 X i o " 3 0.0544 0.051 ' 0.065 0.0169 6.30 X i o " 3 0.0372 0.045 0.062 0.0153 6.30 X 10"3 0.408 0.046 0.065 0.0174 9.86 X 10"3 0.0334 0.041 0.064 - 88 - ( TABLE lOcontinued ) Naphthoquinone 1,3-cyclohexadiene Light Quantum Yield Quantum Yield 27 Concentration Concentration mEinsteins of formation of formation ( M ) ( M ) ene-dione 2_9 alcohol 2_8 0.0153 9.86 x I O - 3 0.0963 0.045 0.073 0.0154 9.86 x 10~3 0.273 0.048 0.069 0.0154 1.02 x 10"2 0.109 0.034 0.062 0.0153 1.26 x 10~2 0.321 0.030 0.065 0.0153 1.58 x 10~2 0.110 0.028 0.059 0.0153 1.89 x 10~2 0.137 0.019 0.060 0.0155 2.05 x 10"2 0.115 0.022 0.061 0.0153 2.52 x 10~2 0.329 0.017 0.061 0.0154 3.15 x 10"2 0.116 0.016 0.086 0.0155 4.10 x 10"2 0.107 0.012 0.065 0.0154 4.41 x 10~2 0.160 0.011 0.058 0.0153 5.68 x 10~2 0.133 0.0092 0.076 0.0153 7.57 x 10~2 0.172 0.0059 0.075 0.0154 8.20 x I O - 2 0.142 0.0062 0.063 0.0154 9.15 x 10~2 0.121 0.0052 0.080 0.0154 1.10 x 10"1 0.164 0.0047 0.075 0.0153 1.26 x 10"1 0.130 0.0039 0.063 0.0153 1.42 x I O - 1 0.188 0.0031 0.066 0.0154 1.58 x I O - 1 0.117 0.0022 0.065 0.0154 1.58 x I O " 1 0.124 0.0029 0.068 For each different quencher concentration, an average ratio was obtained for the ratio $ / $ ( * = unquenched quantum yield, $ with o o quencher). These results were then plotted as * / * versus 1,3- cyclohexadiene concentration ( see Graphs 1 &2 ). For the ene-dione 29 quenching graph 2 } a least squares treatment was made of a l l the points up to and including that for 0.0441 M 1,3-cyclohexadiene to obtain the best straight line through these points. The slope was calculated to be 158 M \ the standard error for the slope was 19 M 1 and the error expected for a 99.9 % confidence limit ( 26 degrees of - 89 - freedom ) for the points was + 70 M \ For the quenching of alcohol 28_, a least squares treatment was made as well. The slope was calculated to be -0.03 M _ 1 and the 99.9 % confidence limit ( 26 degrees of freedom ) of the slope was + 10 M \ ( the standard error for the slope was 2.8 M ). (d) Photolysis of 27. using trans-stilbene as quencher. trans-Stilbene ( Aldrich 98 % ) was twice recrystallized from ethanol, mp 123.5 - 124.0. As before, 0.0155 M solutions of compound 27_ were photolyzed at 366 nm with varying amounts of quencher. The solution was first thoroughly degassed by the freeze- pump-thaw method. It is important to realize that trans-stilbene does absorb some of the light at 366 nm that enters the test solution ce l l , ( 0.2 ) and that cis-stilbene (the product from the excited state jobnm of trans-stilbene ) does too ( e„^- 0.6 ). The data for the joonm quenching by trans-stilbene is presented in Table 11 . TABLE 11 Naphthoquinone trans-stilbene Light Quantum Yield Quantum Yield 27 Concentration Concentration mEinstein of formation of formation ( M ) ( M ) ene-dione 29 alcohol 28 0.0153 5.22 X 10"A 0.426 0.074 0.066 0.0153 1.04 X 10~3 0.409 0.069 0.067 0.0154 2.08 X 10"3 0.416 0.055 0.067 0.0153 1.05 X i o " 2 0.418 0.033 0.061 0.0154 2.11 X i o " 2 0.450 0.017 0.063 0.0153 4.19 X i o " 2 0.408 0.013 0.060 0.0157 6.35 X i o " 2 0.423 0.0065 0.060 0.0154 1.05 X i o " 1 0.420 0.0042 0.059 - 90 - The data is plotted as $ / $ versus trans-stilbene concentration in o graph 3 and graph 4 . Least squares treatment yielded the following results: (a) ene-dione _29 quenching ( graph 4) : slope 186 M 1 , standard error 20 M \ 99.9 % confidence limit on the slope ( six degrees of freedom ) + 119 M 1 s 99.5 % confidence limit on the slope + 86 M (b) alcohol 28_ quenching ( graph 3 ): slope 0.92 M - 1 , standard error 0.85 M _ 1 , 99.9 % confidence limit on the slope ( six degrees of freedom ) + 5.1 M " 1 . (e) Photolysis of 27_ using Oxygen as Quencher : The 0.0154 M solution of 27 in benzene was degassed twice and then flushed with the oxygen ( Matheson, Ultra High Purity, 99.95 % ) and pressurized to 760 mm Hg. The solution was allowed to thaw. The pressure in the cel l was checked again and adjusted to 760 mm Hg. The gauge used was a simple U-tube half f i l led with mercury, open at one end, and connected to the vacuum system at the other end. The latter had a pinch-clamp attached so that the manometer was only open to the system two periods of about 4 seconds. The concentration of oxygen in the solution was ca. 0.01 M. TABLE 12 Naphthoquinone Oxygen Light Quantum Yield Quantum Yield 27 Concentration Concentration mEinsteins of formation of formation ( M ) ( M ) ene-dione 29 alcohol 28 0.0155 @0.01M 0.395 0.0015 0.044 0.0153 00.01*1 0.408 0.0013 0.042 - 91 - (f) Study of the Effect of Changing Quinone 22 Concentration on the Quantum Yields of Photoproducts 29 and 28. Naphthoquinone 27 Concentration ( M ) TABLE 13 % Conversion Light Quantum Yield Quantum Yield mEinsteins of formation of formation ene-dione 29 alcohol 28 0.00769 0.0153 0.0229 0.0306 0.0383 0.0458 0.0611 20.7 13.0 9.1 6.8 5.6 4.6 3.3 0.266 0.343 0.364 0.373 0.371 0.360 0.354 0.093 0.090 0.086 0.084 0.088 0.088 0.087 0.066 0.064 0.066 0.064 0.065 0.069 0.063 The procedure in these experiments was the same as in the previous runs, however no quencher was added. The data is presented also in graph 5. Synthesis of 6,7-Dimethyl-4a B,5,8,8a6-tetrahydro-l,4-naphthoquinone 10. 48 Following the procedure of Mandelbaum and Cais , a mixture of 5.50 g of p-benzoquinone ( 50 mmol, Eastman, practical grade, recrystallized from petroleum ether ( 68° ), decolorized with charcoal, and recrystallized twice more, mp 112.5 - 113.0° ) and 9.45 g of 2,3-dimethyl-l,3-butadiene ( 116 mmol, Aldrich 98 % ) was heated to 60° and stirred for one hour. The diene was removed and the residual solid was recrystallized from petroleum ether 68° and from ethanol to give 8.58 g (45 mmol, 91 % yield ) of pale yellow needles. The material was recrystallized three more times - 92 - 48 from petroleum ether to y i e l d needles of mp 114.5 - 115.0° ( reported mp 115 - 117°); i r (CHC13) 5.90 y ( C=0 ); nmr ( CC14) x 3.5 ( s,2,C2 and v i n y l ), 6.9 ( t , 2, J= 3 Hz, Q and C Q a methines), ( m, 4, C, and C 0 methylenes ), 8.4 ( s, 6, v i n y l methyls ); o o uv ( n-hexane ) A 221 rm ( £ 8720 ), 298 nm ( e 123 ), shoulder 365 nm ( e 60 ). Large Scale Photolysis of 6 ,7-Dimethyl-4a 8,5 ,8,8ag-tetrahydro- 12 1,4-naphthoquinone 10 i n Benzene. Compound 10 "(' 1.500 gm, 7.89 mmol ) was dissolved i n 400 ml of benzene ( reagent grade, d i s t i l l e d ). The solution a f t e r degassing by argon bubbling for 30 minutes, was photolyzed for 21 hrs with a 450 W medium pressure mercury Hanovia Type L lamp. A Corning 7380 f i l t e r allowed only wavelength longer than 340 nm to enter the solution. The photolysis was followed by glpc, using a 5' x 1/4" column packed with 20 % DEGS on 60/80 Chromosorb W ( column temp 150 °C, detector and in j e c t o r temperature 200 °C, helium c a r r i e r gas at 60 ml/minute). Two products appeared. The s t a r t i n g material was heat l a b i l e and could not be detected. The photolysis was stopped when the photoproduct peaks on the gc did not increase i n s i z e any longer. The two photoproducts were separated by column chromatography using 120 gm of S i l i c a Gel ( less than 0.08 mm ) E Merck AG i n a 15" x 1" column and chloroform as eluant. The two photoproducts overlaped i n two of the thirteen fractions i n which they eluted,and these two fractions were discarded. After the chloroform was removed, - 93 - crystals formed for each compound. These were recrystallized from ether-petroleum ether ( 30 - 60° ) to yield: (a) 450 mg ( 2.37 mmol, 30% ) of white crystals of white crystals of 3 9 1 2 8,9-dimethyltricyclo[4.4.0.0 ' ]dec-7-ene-2,5-dione 12, mp 77-78° ( l i t . mp 77-78° ); i r (CHC13) 5.69, 5.81 y ( C=0 ); nmr (CDC13> x 8.63 (s,3,Cn methyl ), 8.13 ( d,3, J= 2 Hz, CQ methyl ), 4.48 ( m,l, vinyl ); uv (methanol) X 292 nm ( e 220 ), shoulder 310 nm . max ( e 200 ). (b) 378 mg (1.99 mmol, 25% ) of white crystals of 8,9-dimethyl- 5-hydroxytricyclo[ 4.4.0.05,9]dec-3,7-diene-2-7;one ;13_V mp ,?r3^94f'""(:iis6..12 mp 93-94°) ; i r (CC14) 2.8 ( weak, OH ), 5.90 y ( C=0 ); nmr (CC14) x 8.90 ( s,3,C 9 methyl ), 8.50 ( d,2,J=5 Hz, C 1 Q methylene),8.20 ( d,3,J= 2 Hz, C_ methyl), 7.76 ( m,2, OH and methine), 6.98 ( d , l , o J=3 Hz, C 6 methine), 4.38 ( m,l , C ? vinyl), 4.15 ( d,l,J=10 Hz, C 3 vinyl), 3.35 ( d,l,J=10 Hz, C. vinyl); uv (methanol) X 242 nm ( e 4000 ), 4 max shoulder 330 nm ( e 30 ). Large Scale Photolysis of 6,7-Dimethyl-4aB,5,8,8aB-tetrahydro- 1,4-naphthoquinone 10 in tert-Butanol. Compound 10_ ( 1.00 g, 5.26 mmol) was dissolved in about 400 ml of an 80:20 mixture of tert-butanol and benzene. The solution was photolyzed X > 340 nm for 20 hrs after degassing with argon. The crude photolysate was disti l led in a Kugelrohr apparatus at 90° and 0.01 mm Hg. The disti l late crystallized on cooling to give beautiful white crystals, 0.789 g (4.15 mmol, 79% ) of 8,9-dimethyltricyclo- 3 7 [4.4.0.0 ' ]dec-8-ene-2,5-dione 11 was obtained this way. - 94 - Recrystallization from petroleum ether (68°) afforded a compound 12 melting at 85-85.5° ( l i t . mp 84-85° ); i r (CHC13) 5.69, 5.81 u (C=0); nmr (CDC13) x 8.63 (s,3, C g - methyl ), 8.13 (d,3,J= 2 Hz, C g methyl), 4.48 ( m,l, vinyl ); uv (methanol) X 292 nm ( e 220 ), shoulder ' j ' ' - ' max 310 nm ( e 200 ). GLPC Response to Photoproducts 12, 13. and 11 Calibration Curve. (a) A stock solution of 90.6 mg of photoproduct 13_ in 100 ml of benzene was prepared. Another stock solution was made up with 103.2 mg of 1,4-naphthoquinone (internal standard) in 100 ml of benzene. These were mixed in predetermined ratios and diluted to 25 ml with benzene in volumetric flasks. Each solution thus made was injected three different times (4 ul injections, peaks separated using the 3' x 1/8 " column of 20% DEGS on 60/80 Chromosorb W, column b). For each run a weight ratio of peak size [(peak area, product 13) / (peak area product JL3_.+ peak area internal standard)] was obtained. The average of the three was then plotted against true weight ratio of product 13_ [ (weight product 13) / (weight product 13 + weight internal standard)], (graph 17). The data thus obtained is presented in Table 14. TABLE 14 Weight of Product Weight of Averaged Peak Area Ratio 13 in 25 ml. 1,4-naphthoquinone alcohol 13 in 25 ml. " , _ , , „ . , . „ alcohol 13 + IS 14.45 mg 2.75 mg 0.676 11.32 2.75 0.616 10.20 2.75 0.594 Graph 17 - 95 - GLPC RESPONSE CALIBRATION CURVE FOR PHOTOPRODUCT 1 3 . - 96 - ( TABLE 14 continued ) Weight of Product Weight of 1,4- Averaged Peak Area Ratio 13 naphthoquinone 9.05 2.57 0.563 7.92 2.57 0-529 6.80 2.57 0.495 5.65 2.57 0.442 4.52 2.57 0.388 2.26 2.57 0.242 1.08 2.57 0.100 (b) A third stock solution was prepared containing 67.6 mg of photoproduct 12 in 100 ml of benzene. Varying aliquots were mixed with the internal standard stock solution and diluted to 25 ml. The solutions were analyzed as described in section (a) above. The results are plotted on graph 18 , and presented in Table 15 . TABLE 15 Weight of Product Weight of 1,4- Averaged Peak Area Ratio 12 in 25 ml. naphthoquinone ( dione 12_ ) in 25 ml. (dione 12 + IS ) 10.15 mg 2.57 mg 0.659 8.45 2.57 0.608 7.60 2.57 0.582 6.75 2.57 0.555 5.92 2.57 0.524 5.07 2.57 0.485 4.22 2.57 0.439 3.38 2.57 0.383 1.69 2.57 0.230 0.861 2.57 0.125  - 98 - (c) A standard solution was prepared containing 30.4 mg of the alcohol 13, 30.4 mg of the dione 12 and 20.2 mg of 1,4-naphthoquinone ( internal standard ) and made up to 100 ml with benzene ( weight ratio alcohol 13 = .601, weight ratio dione 12 = .601 ). This solution was kept refrigerated at a l l times and only kept for two weeks at which time a new solution was prepared. This solution was injected into the gc after every glpc analysis of the photolysis mixture of 10^ in benzene. (d) A stock solution of 94.0 mg of photoproduct 11 in 100 ml of benzene was prepared. A stock solution containing 140.4 mg of 1,4-naphthoquinone in 50 ml of benzene was also made. Once again, the solutions were combined in varying ratios, diluted to 25 ml in benzene, and analyzed by glpc in the same manner as described in section (a). Graph 19 represents the standard curve for the data in Table 16 . TABLE 16 Averaged Peak Area Ratio (dione 11^ ) (dione 11 + IS> 14.10 mg * 5.62 0.706 11.28 5.62 0.659 9.40 5.62 0.618 7.52 5.62 0.568 5.64 5.62 0.496 4.70 5.62 0.457 3.76 5.62 0.397 2.82 5.62 0.332 1.88 5.62 0.253 0.94 5.62 0.150 Weight of Product 11 in 25 ml Weight of 1,4- Naphthoquinone in 25 ml  - 100 - (e) Finally, a standard solution of 37.9 mg of dione 11 and 19.9 mg of 1,4-naphthoquinone ( internal standard ) in 100ml of benzene ( weight ratio = 0.656 ) was prepared. This solution was kept cold, and used to check the response of the gc detector to the photolysis mixture of 10 in tert-butanol. This standard solution was kept only for two weeks before a new solution was prepared. Quantum Yield Determinations of the Photolysis of 6,7-Dimethyl -4a3,5,8,8ag-tetrahydro-l,4-naphthoquinone JL0_ in Benzene. (a) Quantum Yield for the Formation of Alcohol 1_3_ and Dione 12 — Unquenched Photolysis. The solutions of ca. 0.02 M of compound H) in benzene were degassed and photolyzed for approximately 4.5 hrs at 366 nm. After photolysis a 15 ml aliquot of the photolysate was mixed with 2 ml of a stock solution of 1,4-naphthoquinone ( internal standard ) and this mixture was then injected into the gc for analysis. Each photolysis mixture was injected into the gc twice interspaced by an injection of the standard solution of alcohol 13_ , dione 12, and 1,4-naphthoquinone ( described earlier ) and another injection of the standard solution at the end. Table 17 gives the quantum yield of formation of both products 12_ and 13 in benzene. TABLE 17 Naphthoquinone 10 Light % Quantum Yield Quantum Yield Concentration (M) mEinsteins Conversion of formation of formatioi * ** alcohol 13_ dione _12_ 0.0199 0.043 0.18 0.0149 0.0073 0.0198 0.102 0.51 0.0169 0.0091 0.0200 0.532 2.7 0.0180 0.0085 0.0201 0.548 2.3 0.0158 0.0070 - 101 - * The mean quantum y i e l d i s 0.0164 + 0.0012 ** The mean quantum y i e l d i s 0.0080 + 0.0008 ( The errors represent standard errors for the data given.) (b) Photolysis using Piperylene as Quencher. Piperylene (3.4 ml, 33.8 mmol, to make a 1.27 M sol u t i o n ) , p u r i f i e d _2 by d i s t i l l a t i o n from LAH, was added to a solution of 113.1 mg (1.98x10 M) of 10 i n benzene to make up 30 ml of solution. The 26.8 ml of the solution was degassed and photolyzed at 366 nm. The quantum y i e l d for the alcohol 13 was $ = 0.0104 and for the dione 12 $ = 0.0040. (c) Photolysis using 1,3-Cyclohexadiene as Quencher. Benzene solutions (26.8 ml) ca. 0.02 M i n 10 containing various concentrations of p u r i f i e d 1,3-cyclohexadiene were degassed and photolyzed for ca. 5 hrs at 366 nm. TABLE 18 Naphthoquinone 10 1,3-Cyclohexadiene Light Quantum Y i e l d Quantum Y i e l d Concentration ( M ) Concentration (M) mEinsteins of formation of formation alcohol 13 dione 12 0.0199 3.94 X IO" 3 0.561 0.0162 0.0076 0.0198 9.86 X IO" 3 0.525 0.0156 0.0078 0.0198 1.97 X i o ' 2 0.533 0.0153 0.0072 0.0198 2.96 X i o " 2 0.490 0.0152 0.0070 0.0199 3.94 X i o ' 2 . 0.564 0.0145 0.0066 0.0198 5.91 X i o " 2 0.514 0.0137 0.0059 0.0198 9.86 X i o " 2 0.543 0.0128 0.0055 0.0198 1.97 X 10"1 0.462 0.0115 0.0044 0.0199 3.94 X i o " 1 0.559 0.0110 0.0042 0.0198 7.88 X 10"1 0.529 0.0100 0.0038 0.0198 1.18 0.508 0.0102 0.0040 - 102 - A least squares treatment was made of a l l points up to and including that _2 for a 1,3-cyclohexadiene concentration of 9.86 x 10 M for both photoproducts. 33 For the alcohol 13_ an F test of the points gave an F = 9.8 ( F ^ ^ ^= 6.61, F.^ ^ ^= 16.3 ) indicating that the points do very likely make a straight line. The slope for this line was then calculated to be 3.0 M \ the standard error for the slope was 0.9 M 1 and the error expected for 99.9 % confidence limit ( 5 degrees of freedom ) is + 6.2 M ^ The data for the dione 12 gave an F test of F= 28.4 ( F ^ ^ ^ = 16.3 ), thus this data does seem to l ie in a straight line. The slope of this line was calculated to be 4.8 M \ the standard error for the slope was 0.9 M 1 and the error expected for 99.9 % confidence limit is + 6.2 M _ 1 . (d) Photolysis using trans-Stilbene as Quencher. Table 19 presents the data obtained from photolysis of 26.8 ml of a solution of ca.113 mg of compound 10 in 30 ml of benzene plus varying amounts of purified trans-stilbene. The degassed solution was photolyzed at 366 nm for approximately 5 hrs. TABLE 19 Naphthoquinone trans-Stilbene Light Quantum Yield Quantum Yield 10 concentration concentration m Einsteins of formation of formation M ) ( M ) alcohol 13 dione 0.0199 5.21 x 10~3 0.496 0.0159 0.0078 0.0198 1.05 x 10"2 0.538 0.0161 0.0073 0.0199 1.57 x 10~2 0.511 0.0152 0 0)074 0.0198 2.26 x 10"2 0.515 0.0150 0.0067 0.0198 4.32 x 10"2 0.547 0.0138 0.0062 0.0198 1.04 x 10"1 0.508 0.0122 0.0048 0.0198 1.67 x 10"1 0.522 0.0115 0.0044 - 103 - The data is presented in a Stern-Volmer plot in graph 9 for the alcohol 13_ and graph 8 for the dione 12. The slope for the graph of the alcohol 13_ was determined by the least squares method for a l l points except for the trans-stilbene concentration of 0.167 M. An F test resulted in an F = 39 ( F ^ ^ ^= 21.2 ), indicating that there is a linear relationship of the points. Slope = 3.9 M \ the standard error was 0.6 M 1 and the 99.9 % confidence limit for the slope was (for 5 degrees of freedom) +5.2 M _ 1 . A least squares treatment was made for a l l points of ene-dione , 12 quenching, except for a quencher concentration of 0.167 M. An F test of the points resulted in F = 26.1 ( F ^ ^ 1 ^ =̂ 21.2 ), thus the points again appear to form a straight line. The best line had a calculated slope of 6.6 M \ standard error 1.3 M \ and the 99.9 % confidence limit of the slope was + 11.2 M \ (e) Photolysis of 10 in Benzene : Effect of Variation of Compound 10_ Concentration on Quantum Yield. TABLE 20 Naphthoquinone 10_ % Conversion Light Quantum Yield Quantum Yield Concentration (M) mEinstein of formation of formation alcohol 13 ene-dione 12 0.00996 0.0198 0.0296 0.0396 0.0494 0.0593 0.0792 5.0 2.2 1.7 1.3 0.96 0.82 0.60 0.535 0.518 0.531 0.540 0.522 0.529 0.528 0.0168 0.0153 0.0170 0.0164 0.0154 0.0160 0.0165 0.0080 0.0072 0.0076 0.0089 0.0081 0.0085 0.0073 - 104 - Quantum Yield Determinations of the Photolysis of 6,7-Dimethyl -4a 3,5,8,8a0-tetrahydro-l,4-naphthoquinone 10 in tert-Butanol. (a) Quantum Yield for the formation of dione 11 — Unquenched Experiments. A series of 0.02 M solutions of 10_ were thoroughly degassed by the freeze-pump-thaw method and photolyzed again at 366 nm. TABLE 21 Naphthoquinone 10 Solvent Light % Conversion Quantum Yield Concentration (M) used mEinstein of formation dione 11 ** 0.0198 _t-BuOH/Benzene* 0.072 0.11 0.0084 0.0199 * _t-BuOH/Benzene 0.127 0.22 0.0091 0.0198 tert-BuOH 0.690 1.2 0.0095 0.0198 tert-BuOH 0.58 5 0.77 0.0069 0.0198 t-BuOH/Benzene* 0.48 0 0.68 0.0075 0.0199 t-BuOH/Benzene 0.512 0.74 0.0076 0.0198 * _t-BuOH/Benzene 0.5 67 0.85 0.0079 * A 95:5 r a t i o of tert-Butanol/benzene was used for the runs. **Mean value $ = .0081 + 0.0008. The error i s the standard o — error for t h i s data. (b) Photolysis of 10 i n tert-Butanol Using 1,3-Cyclohexadiene as Quencher. Photolysis at 366 nm of degassed 0.02 M solutions of 10 in ( 95:5 ) tert-butanol-benzene with varying concentrations of 1,3-cyclohexadiene were performed. Again 15 ml aliquots of the photolysate were combined with 2 ml of standard 1,4-naphthoquinone ( in t e r n a l standard ) solution and these mixtures analyzed by glpc. Table 22 represents the data obtained from these experiments. Naphthoquinone 10_ Concentration (M) - 105 - TABLE 22 1,3-Cyclohexadiene Light Concentration (M) mEinstein Quantum Yield of formation ene-dione 11 0.0198 3.94 x 10~3 0.566 0.0081 0.0198 7.88 x 10~3 0.525 0.0079 0.0199 1.58 x 10~2 0.470 0.0091 0.0198 3.15 x 10~2 0.418 0.0082 0.0199 7.88 x 10~2 0.518 0.0080 0.0199 2.37 x 10 - 1 0.433 0.0079 0.0199 3.94 x 10 _ 1 0.627 0.0074 0.0198 3.94 x 10 - 1 0.518 0.0090 0.0199 7.10 x 10 - 1 0.526 0.0077 0.0198 11.04 x 10 - 1 0.522 0.0080 The Stern-Volmer plot of this data is graph n . The slope was calculated from a least squares treatment to be 0.047 M \ the standard error 0.16 M 1 and the 99.9 % confidence limit ( 8 degrees of freedom ) 0.8 M (c) Photolysis of 10 in tert-Butanol: Effect of Variation of 10 Concentration on Quantum Yield The solvent used was ( 95:5 ) tert-Butanol-Benzene (dried), Photolysis was at 366 nm for ca.5 hrs (depending on lamp age). TABLE 23 Naphthoquinone 10 % Conversion Light Concentration (M) mEinstein Quantum Yield of formation ene-dione 11 0 .0099 1.5 0 . 0 1 9 8 0 . 8 0 0 .0297 0 . 5 3 0 . 0 3 9 6 0 . 4 4 0.522 .0074 0.554 0.0076 0.508 0.0083 0.567 0.0081 - 106 - ( TABLE 23 continued ) Naphthoquinone 10 % Conversion Light Concentration (M) mEinstein Quantum Yield of formation ene-dione 11 0.0496 0.37 0.547 0.0089 0.0594 0.28 0.524 0.0084 0.0791 0.19 0.516 0.0077 Photolysis of 1,3-Cyclohexadiene in Benzene with Benzophenone. A solution of 2.521 gm ( 31.5 mmol ) of 1,3-cyclohexadiene (Aldrich 98 %, twice fractionally dist i l led ) and 0.920 gm ( 5.1 mmol ) of benzophenone ( Aldrich, reagent , twice disti l led ) in 50 ml of benzene was stirred & purged with argon for 15 minutes and then photolyzed for 27.0 hrs using a 450 W medium pressure mercury Hanovia Type L lamp, using a Corning 7380 f i l ter to block light of wavelength shorter than 340 nm. Argon bubbling and stirring was continued during the photolysis. After photolysis the benzene was removed and the remaining clear o i l was analyzed by glpc, using a 5' x 1/4" column packed with OV-1 on 60/80 Chromosorb W ( column temperature 120° , injector and detector kept at 170° , helium carrier gas flow at 30 ml/min.). Several peaks of short retention time ( less than six minutes ) were observed; three major peaks were also detected. One of these, t ^ = 30 minuteSjWas due to benzophenone ( determined by co-injection of a fresh solution of benzophenone in benzene ). The other two peaks,t 44 = 18 minutes, and t „ 45 = 21 minutes, were obtained in a 4:1 ratio r e s p e c t i v e l y . However, the major peak had a shoulder at t „ 46 = 17 minutes. R — - 107 - The isolation of the dimers by column chromatography, using 10 % silver nitrate-alumina (Al^O^ Woelm, neutral, activity grade I), 70 described by Hammond was attempted but failed to give a satisfactory separation. Separation of the photolysis mixture was also tried with a 15' x 1/8" column of Apiezon L on 40/60 Firebrick ( column 160° , injector and detector 180° , helium carrier gas flow 20 ml/minute). Once again, two major peaks were obtained: t 44= 72 minutes with shoulder peak at t_,46 = 76 minutes; and t_, 45 = 91 minutes. The ratios of R K the two peaks were again 4:1 respectively, ( 44 + 4̂> / 45 ). The two peaks ( 44_ and 45 ) were collected by glpc from the OV-1 column. However, the major component 44, was collected, starting from a retention time above 17.5 minutes, thus avoiding a significant portion of the material 46_ representing the shoulder peak, 2 7 The major product, cis,anti,cis-tricyclo[6.4.0.0 ' Jdodeca -3,11-diene 44̂  had the following nmr (CCl^) T 4.22 (m,4,vinyl),7.64 (m, methines), 7.96 (m,4,C^ and methylenes), 8.45 (m,4,Cg and methylenes). 2 7 The minor product, cis,syn, cis-tricyclo[6.4.0.0 * ]dodeca-3,ll -diene 45, had the following nmr (CCl^) x 4.30 (m,4, vinyl) , 7.26 (m,4,methines), 8.21 (m,8,C^,Cg,Cg,C^g methylenes). These two nmr spectra were identical to those published by 70 Hammond and coworkers. These two dimers of 1,3-cyclohexadiene were injected separately into the gc, using the 10' x 1/8" and the 3' x 1/8" 20% DEGS on 60/80 Chromosorb W columns used for quantum yield determinations, operating at a reduced temperature of 8 0 ° . The retention times of - 108 - the dimers corresponded with those of peaks appearing only when 1,3-cyclohexadiene was used as quencher. Coinjection of dilute samples of 4_4 and 45_ in benzene, with photolysis mixtures of 10 and 27_ ( quenched with 1,3-cyclohexadiene), confirmed the assignment of these new peaks. - 109 - APPENDIX Glpc recorder trace of the photolysis products of 27_ in benzene: A. Internal standard biphenyl ( R = 22 min. ), B. Photoproduct ene-dione 2_9 ( R = 52 min. ), C. Photoreactant 27_ D. Photoproduct alcohol 28_ ( R̂ = 66 min. ). - 110 - A B C Glpc recorder trace of the photolysis products of 10 In benzene A. Internal standard 1,4-naphthoquinone ( Rc= 26 min. ), B. Photoreactant 10_ ( R = 34 min., thermally decomposes ), C. Photoproduct alcohol 13_ ( R = 41 min. ), D. Photoproduct ene-dione 1]L_ E. Photoproduct ene-dione 12 ( R = 59 min. ). - I l l - Glpc recorder trace of the photolysis products of 10 in tert-butanoli A. B. C. D. E. Internal standard 1,4-naphthoquinone ( R = 19 min. ), Photoreactant 10 ( R = 28 min., thermally decomposes ), Photoproduct alcohol 13, Photoproduct ene-dione LI ( R = 42 min. ), Photoproduct ene-dione 12.   - 114 - BIBLIOGRAPHY 1. D. Bryce-Smith, "Photochemistry", vol. 1, The Chemical Society, ( London, 1970 ), p. 1. 2. D.C. Neckers, "Mechanistic Organic Photochemistry", Reinhold Publishing Corp., ( New York, 1967. ) ;..p. 28. 3. Multiplicity is given by the equation 2|s|+l, where S is the net spin of the electrons. The net spin is zero and the multiplicity one ( singlet ) i f a l l spins are paired. However, the net spin is one and the multiplicity three ( triplet ) i f the spins of two electrons are the same. This does not violate Pauli's exclussion principle when i t occurs in an excited state. 4. P.A. Leermakers, "Techniques of Organic Chemistry", vol XIV, Interscience Publishers, ( Toronto, 1969 ), p. 3. 5. J.G. Calvert and J.N. Pitts, Jr . , "Photochemistry", John Wiley & Sons, Inc., ( New York, 1967 ), p. 48. 6. J.G. Calvert and J.N. Pitts, Jr . , i b id . , p. 258. 7. D. Phill ips, "Photochemistry", vol 1, The Chemical Society, ( London, 1970 ), pp. 4-22. 8. P.A. Leermakers, op. c i t . , p. 11. 9. A.A. Lamola, "Technique of Organic Chemistry", vol. XIV, ( Toronto, 1969 ), p. 34. 10. R.C. Cookson, E. Grundwell, R.R. H i l l , and J. Hudec, J. Chem. Soc., 3062 ( 1964 ). 11. J.R. Scheffer, J . Trotter, R.A. Wostradowski, C.S. Gibbons, and K.S. Bhandari, J. Amer. Chem. Soc , 93, 3813 ( 1971 ). - 115 - 12. J.R. Scheffer, K.S.Bhandari, R.E. Gayler, and R.H. Wiekenkamp, J. Amer. Chem. Soc., 94, 285 ( 1972 ). 13. R.B. Woodward and R. Hoffmann, "The Conservation of Orbital Symmetry", Academic Press, ( New York, 1970 )• 14. R.L. Cargil l , B.M. Gimarc, D.M. Pond, T.Y. King, A.B. Sears, and M.R. Willcott, J. Amer. Chem. Soc, 92, 3809 ( 1970 ). 15. (a) A.F. Thomas and B. Willhalm, Tet. Let., 1309 ( 1965 ), and (b) A.F. Thomas, R.A. Schneider, and J. Meinwald, J. Amer. Chem. Soc., 89, 68 ( 1971 ) . 16. N.H. Werstiuk and R. Taillefer, Can. J. Chem., 48, 3966 ( 1970 ). 17. J.R. Scheffer, J. Trotter, R.E. Gayler, and C A . Bear, Tet. Let., 2871 ( 1973 ) . 18. J.A. Barltrop and B. Hesp, J. Chem. Soc , 1625 ( 1967 ). 19. R.E. Gayler, PhD Thesis, University of B.C., 1973. 20. (a) W. Herz and M.G. Nair, J . Amer. Chem. Soc , 89, 5474 ( 1967 ), and (b) W. Herz and M.G. Nair, J. Org. Chem., 39, 117 ( 1974 ). 21. (a) S. Wolff, W.L. Schreiber, A.B. Smith, III, and W.C Agosta, J . Amer. Chem. Soc, 94, 7797 ( 1972 ), (b) A.B. Smith, III, and W.C. Agosta, J. Amer. Chem. Soc , 95, 1961 ( 1973 ), and (c) A.B. Smith, III, and W.C. Agosta, J. Amer. Chem. Soc, 96, 3289 ( 1974 ) . 22. T. Kobayashi, M. Kurono, H. Sato, and K. Nakanishi, J. Amer. Chem. Soc., 94, 2863 ( 1972 ) . 23. J.R. Scheffer, J. Trotter, A.A. Dzakpasu, and S. Phill ips, work to be published. - 116 - 24. H.M.R. Hoffmann, Angew. Chem. Int. Ed. Engl., 8̂ , 556 ( 1969 ). 25. R.A. Cormier and W.C. Agosta, J. Amer. Chem. Soc., 96, 618 ( 1974 ). 26. M.F. Ansell, B.W. Nash, and D.A. Wilson, J. Chem. Soc, 3012 ( 1963 ). 27. J.A. Barltrop and D. Giles, J. Chem. Soc, 105 ( 1969 ). 28. D. Craig, J. Amer. Chem. Soc, 65, 1006 ( 1943 ). 29. A.A. Lamola, op. c i t . , pp. 102 & 105. 30. G. Porter and F. Wilkinson, Proc. Roy. Soc. ;( London ), A 264, 1 ( 1961 ). 31. (a) D. Valentine, N.J. Turro, Jr . , and G.S. Hammond, J. Amer. Chem. Soc., 86, 5202 ( 1964 ), and (b) G.O. Schenck, S.P.Mannsfeld, G. Schomburg, and C.H. Kranck, Z. Naturforsch., B 19, 18 ( 1964 ). 32. P.J. Wagner, J Amer. Chem. Soc , 89, 5715 ( 1967 ). 33. C.J. Brooks, I.G. Betteley, and S.M. Loxton, "Mathematics and Statistics for Chemists", John Wiley & Sons, ( New York,,1966 ) p. 346 ff. 34. H.E. Zimmerman and K.G. Hancock, J. Amer. Chem. Soc, 90, 3749 ( 1968 ). 35. (a) G.S. Hammond and J . Saltiel , J. Amer. Chem. Soc, 84, 4983 ( 1962 ) and (b) G.S. Hammond, J. Sa^ ie l , A.A. Lamola, N.J. Turro, J.S. Bradshaw, D.O. Cowan, R.C. Counsell, V. Vogt, and C. Dalton, J . Amer. Chem. Soc, 86, 3197 ( 1964 ). 36. S.L. Murov, "Handbook of Photochemistry", Marcel Dekker Inc., ( New York, 1973 ), p. 89. 37. N.J. Turro, "Molecular Photochemistry", W.A. Benjamin, Inc. ( New York, 1967 ), p. 196. - 117 - 38. T. Forster and K. Kasper, Z. Electrochem., 59, 977 ( 1955 ). 39. The involvement of excimers in triplet-triplet annihilation was demonstrated first for the system pyrene in ethanol by C.A. Parker and C.G. Hatchard, Trans. Faraday Soc., 59, 284 ( 1963 ); bimolecular triplet annihilation of 2,3-pentanedione was proposed by N.J. Turro and T.J . Lee, J. Amer. Chem. Soc., 91, 5651 ( 1969 ), and ib id . , 92, 7468 ( 1970 ), and references therein. 40. B.M. Jennings and J.R. Scheffer, results to be published. 41. (a) R.L. Cargil l , W.A. Bundy, D.M. Pond, and A.B. Sears, Mol Photochem. , 3_, 123 ( 1971 ), and (b) R.L. Cargil l , A.B. Sears, J. Boehm, and M.R. Willcott, J. Amer. Chem. Soc , 95, 4346 ( 1973 ). 42. P.J.'Wagner and G.S. Hammond, Advan. Photochem., _5» 2 1 ( 1 9 6 8 ) - 43. D. Bellus, D.R. Kearns, and K. Schaffner, Helv. Chim. Acta, 52, 971, ( 1969 ) 44. J.R. Scheffer, unpublished results. 45. F. Dupuy, C. Leibovici, and M. Sebti ( Fac. Sc i . , Bordeaux, France ), C.R. Acad. Sci . , Paris, Ser. A,B, 263 B, 1321-3 ( 1966 ) (Fr.). 46. P.J. Wagner, Acc. Chem. Res. , k_> 168 ( 1971 ). 47. Phosphorescence measurements were attempted using a Hitachi- Perkin Elmer Spectrofluorimeter-Phosphorimeter by Dr. G. Kaupp, University of Freiburg, W. Germany. The author wishes to thank Dr. Kaupp for a l l his efforts. 48. A. Mandelbaum and M. Cais, J. Org. Chem., 27, 2243 ( 1962 ). 49. E.F. Ullman and W.A. Henderson, J. Amer. Chem. Soc. , 88, 4942 ( 1966 ). - 118 - 50. I. Fleming and D.H. Williams, "Spectroscopic Methods in Organic Chemistry", Mc Graw-Hill, ( London, 1966 ), p. 15. 51. J.R. Scheffer, K.S. Bhandari, R.E. Gayler, and R.A. Wostradowski, J. Amer. Chem. Soc. , 97, 2178 ( 1975 ). 52 E.C. Alexander and J.A. Uliana, J. Amer. Chem. Soc., 96, 5644 ( 1974 ). 53. W.C. Agosta and S. Wolff, J. Amer. Chem. Soc., 97, 457 ( 1975 ). 54. E.L. E l i e l , "Stereochemistry of Carbon Compounds", McGraw-Hill, ( Toronto, 1962 ), pp. 124-6. 55. (a) N.J. Turro, P. Lechtken, N.E. Schore, G. Schuster, H.C. Steinmetzer, and A. Yekta, Acc. Chem. Res. , 4_, 97 ( 1974 ), and (b) K.R. Kopecky, private communication. 56. N.J. Turro, "Molecular Photochemistry", W.A. Benjamin. Inc., ( New York, 1967 ), p. 50. 57. H.E. Zimmerman, Mol. Photochem., 3_» 2 8 1 ( 1971 ). 58. J.C. Gardner and F.A. Case, U.S. Department of Commerce, Bureau of Standards, Circular 389 ( 1931 ). 59. The cells can only be dipped into the epoxy paint. Brushing cuases scratches in the mirror. Two coats of paint were sufficient to give strength and protection to the mirror. 60. C.G. Hatchard and C.A. Parker, Proc. Roy. Soc. ( London ), Ser. A, 235, 518 (1956 ). 61. T.R. Evans, "Technique of Organic Chemistry", vol XIV, Interscience Publishers ( Toronto, 1969 ), p. 329. 62. (a) C.A. Parker and C.G. Hatchard, J. Phys. Chem., 63, 22 ( 1959 ), and (b) G.B. Porter, J.G.W. Doering, and S. Kararika, J. Amer. Chem. Soc., 84, 4027 ( 1962 ). - 119 - 63. K.C. Kurien, J . Chem. Soc.(B), 2081 ( 1971 ). 64. J . Lee and H.H. Seliger, J . Chem. Phys., 40, 519 ( 1964 ). 65. W.M. Moore, G.S. Hammond, and R.P. Foss, J. Amer. Chem. S o c , 82, 2789 ( 1961 ). 66. W.M. Moore and M. Ketchum, J . Amer. Chem. Soc. , 84, 1368 ( 1962 ). 67. R.L. Hoffmann and CD. Evans, J . Gas Chrom. , 4_, 318 ( 1966 ). 68. (a) L.I. Smith and F.J. Dobrovblny, J . Amer. Chem. S o c , 48, 1420 ( 1926 ), and (b) L.I. Smith, Organic Synthesis 4, 254 ( 1943 ). 69. P.A. Leermakers, Mol. Photochem., _3> 2 0 1 ( 1 9 7 1 ). 70. (a) Y.L. Bahurel, D.J. MacGregor, T.L. Penner, and G.S. Hammond, J. Amer. Chem. Soc., 94, 637 ( 1972 ), and (b) L.R. Chapman and D.F. Keummel, Anal. Chem., 37, 1598 ( 1965 )

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
China 7 0
United States 6 0
Germany 3 0
France 2 0
Russia 1 0
Ethiopia 1 0
City Views Downloads
Shenzhen 5 0
Unknown 4 0
Wurzburg 3 0
Beijing 2 0
Redmond 2 0
Syracuse 1 0
Sunnyvale 1 0
Dire Dawa 1 0
Ashburn 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}

Share

Share to:

Comment

Related Items