8 Photochemistry By W. M. HORSPOOL Department of Chemistry The University Dundee DO7 4HN Benzene photochemistry is an area which continues to provide interesting results. Barltrop and Day' have suggested that permutational analysis may be useful in recognizing which path has been followed in the transposition of groups in the rearrangement of aromatic systems e.g. the conversion of o-xylene into rn-xylene. They point out that there are twelve permutation patterns in the six-membered ring systems and the two which would account for the xylene transformation are shown in (1)and (2). In no case has it been established which of the 12 patterns actually occurs R3 R'OR' R'\+ 0 R5 in the transpositions since most of the experiments so far described in the literature have had insufficient labelling of atoms.They' also suggest that there are no cases where a connection has been established between benzvalene or Dewar and pris- mane isomers and the occurrence of photo-transposition. This statement has been challenged by Chambers et uZ.,~ who cite the photochemical rearrangements of perfluoroalkyl-pyridines. Permutation patterns for the transposition of ring-carbons have been used to analyse the photo-reactions of alkyl hydroxypyrilium cations (3)3 and of 2qanopyrr0les.~ J. A. Barltrop and A. C. Day J.C.S. Chem. Cbmm. 1975,177. * R. D. Chambers R. Middleton and R. P. Corbally J.C.S. Chem. Cbmm. 1975,731. J. A. Barltrop R. Carder A. C. Day J. R. Harding and C.J. Samuel,J.C.S. Chem. Comm. 1975,729. J.A. Barltrop A. C. Day P. D. Moxon and R. R. Ward J.C.S. Chem. Cornrn. 1975,786. 167 168 W.M. Horspool The interest in the photoaddition reactions of benzene and substituted derivatives has been maintained. Typical of this is the report of the 1,3-addition of diphenylacetylene to trimesic trimethyl ester to afford the adduct (4).' 1,2-Dimethylenecyclohexane has also been added to benzene photochemically to yield the first example (5) of 1,4-1',3'-additi0n.~ Compound (6) a 1,4-lf,4'-adduct was also isolated. Bryce-Smith et a1.' have published work which shows that the bicyclo-octadiene (7) is the immediate precursor of the 2 :1adduct in the benzene- maleic anhydride system. The diene (7) is the precursor in both the sensitized and direct reactions [in direct irradiation the zwitterion (8) is precursor to the diene (7)].0 0 0 0-(7) (8) The influence of diluents upon the reaction path has demonstrated that di- bromoethane can act as an aid to intersystem crossing (S +T,).A further report* on the photochemical addition of furan to benzene has explained that the differences between the original reports were due to different experimental The formation of (9) and (10) as the main products is dependent upon the use of a virtually monochromatic light source of low intensity. The formation of (1 1) requires a high-intensity source. (9) (10) (11) The influence of xenon on the fluorescence efficiency of benzene in oxygen-free cyclohexane has been studied. l1 Under these conditions the fluorescence intensity is decreased; however the photochemical yield of benzvalene is not affected.The results support the assumption that benzvalene formation arises from a non-relaxed state before the fluorescent level is reached. The photochemistry of benzvalene (12a) is inefficient with a quantum yield of 0.1. This inefficiency is a result of the degenerate photovalence isomerization (12b) +(13).12 This isomerization can be brought about by direct irradiation at 254nm or else by triplet sensitization providing that the sensitizer energy is <272 kJ mol-' but >222 kJ mol-'. When the triplet energy is >272 kJ rno1-l the photolysis affords benzene. The authors12 5 T. Teitei and D. Wells Tetrahedron Letters 1975 2299. 6 J. C. Berridge D.Bryce-Smith and A. Gilbert Tetrahedron Letters 1975 2325. 7 D. Bryce-Smith R. R. Deshpande and A. Gilbert Tetrahedron Letters 1975 1627. 8 J. C. Berridge D. Bryce-Smith A. Gilbert and T. S. Cantrell J.C.S. Chem. Cornm.,1975 6 11. 9 J. C. Berridge D. Bryce-Smith and A. Gilbert J.C.S. Chem. Cornm. 1974 965. 10 T. S. Cantrell Tetrahedron Letters 1974 3959. 11 Y. Ilan and G. Stein Chem. Phys. Letters 1975 31 441. 12 C. A. Renner T. J. Katz J. Pouliquen N. J. Turro and W. H. Waddell J. Arner. Chern. SOC.,1975,97 2568. Photochemisfry suggest that the higher energy sensitizers populate the benzvalene T2state which ring-opens to benzene triplets whose energy is sufficient to decompose a second benzvalene molecule i.e. a quantum chain process is in operation.The dependence on sensitizer concentration is in agreement with this proposal. D (12) a; R = H b;R=D Exciplex formation in arene-diene systems continues to be an area of active investigation. In this area Lewis and Hoyle13 have examined temperature depen- dence and have established that reversible exciplex formation is important. Thus it is possible to explain the decrease in the fluorescence quenching rate constants with diene ionization potential in terms of increased reversibility of exciplex formation rather than simply a decrease in the rate of exciplex formation. Similar observations have been made in a kinetic analysis of the a-cyanonaphthalene-olefin systems.14 Exciplexes are also involved in the reaction of 9-cyanoanthracene with furans,” and of anthracene with dienes.“J’In another report the piperylene-sensitized dimeriza- tion of 9-phenylanthracene has been shown to be a quite general process and several dienes show the ability to sensitize the dimeri~ation.~’,~~ The reaction mechanism apparently involves deactivation of the anthracene excimer by the diene. A similar effect has been observed in the dimerization of the parent anthracene. However in this case the authors2* conclude that an exciplex between the diene and the excited-state anthracene is involved. The generation of radical cations of 1-phenylcyclopentene 1-phenylcyclohexene and 2-phenylnorbornene has been reported in the photolysis of these olefins in the presence of 1-cyanonaphthalene as an electron acceptor.21 Photo-dehydrocyclizations of cis-stilbene analogues have long been an area of considerable activity.In the past year activity has been maintained and the R3 (14) a; R = a-C,,H (15) a; R2-R3 = (CH=CH),; R’ = H b; R = B-CloH b; R1-R2 = (CH=CH) ;R3= H l3 F. D. Lewis and C. E. Hoyle J. Amer. Chem. SOC.,1975,97,5950. l4 W. R. Ware D. Watt and 3.D. Holmes J. Amer. Chem. Soc. 1974,96,7853. l5 K. Mizuno C. Pac and H. Sakurai J.C.S. Perkin I 1974,2360. l6 N. C. Yang D. M. Shold and J. K. McVey J. Amer. Chem. SOC.,1975,97,5004. N. C. Yang K. Srinivasachar,B. Kim and J. Libman J. Amer. Chem. Soc. 1975,97,5006. R. 0.Campbell and R. S. H. Liu Mol. Photochem. 1974,6 207. l9 R. 0.Campbell and R. S. H. Liu Chem. Comm. 1970,1911. 2o J. Saltiel and D.E. Townsend J. Amer. Chem. SOC.,1973,95 6140. 21 Y.Shigemitsu and D. R. Arnold J.C.S. Chem. Comm. 1975,407. 170 W. M. Horspool photochemical cyclization of the benzothienylnaphthalenes (14) has been reported.22 The a-naphthyl isomer (14a) cyclizes normally to afford (15a) while the p-isomer (14b) yields (15b). This result is in contrast to normal cyclization paths encountered in other P-styrylnaphthalene cyclizations. Another interesting applica- tion of the photo-cyclization is in the formation of the bridged [18]annulene (16) which was obtained from the low-temperature (-80 "C)irradiation (254 nm) of a THF solution of (17).23 RR RR (16)a; R = H b;R-R = CH2CH2 Attention is still being focussed on photochemical reactions of dienes (either conjugated or not).The photochemical excitation either direct or sensitized of allenes (18) in acetic acid affords the enol acetates (19). The photo-addition to these substrates takes place in the reverse sense from the ground-state process.24 The penta-173-dienes have been the subject of attention over the past few years in reports which showed that direct irradiation (253.7 nm) of cis-and trans-penta-1,3-dienes R' R3 H R2 OAC R2Fc=( RkcR3 (18) a; R' = Ph; R2 = Me; R3= H b; R' = Ph; R2 = R3 = Me c; R' = rn-MeOC6H4; R2 = Me; R3 = H d; R' = n-C6H13; R2 = R3 = H in solution led to 3-methylcyclobutene 1,3-dimethylcyclopropene,as well as the geometric isomer of the diene.25 A recent study26 has now shown the presence of a wavelength effect in this system which was made evident by the lack of cyclobutene and cyclopropene products when the penta-1,3-dienes were irradiated at 228.8 nm.Interest has still been maintained in the photochemistry of non-conjugated dienes particularly the di-w-methane systems. In this respect the photochemistry of (20) and (21) which yield (22) and (23) respectively has shown that there is no demand on the di-w-methane reaction for either a cis-or a trans-rearrangement in the intermediate biradical [e.g. (24) which is the biradical proposed for the rearrange- ment of 1,1,5,5-tetraphenyl-3,3-dirnethylpenta-l74-diene into the vinylcyclo-propane].*' The replacement of one of the double bonds in the di-7r-methane 22 A. Croisy P. Jacquignon and F. Perin J.C.S. Chem.Comm. 1975 106. 23 R. B. Vernet T. Otsubo J. A. Lawson and V. Boekelheide J. Amer. Che.m. SOC.,1975,97 1629. 24 K. Fujita K. Matsui and T. Shono J. Amer. Chem. SOC.,1975,97 6256. 25 S. Boue and R. Srinivasan J. Amer. Gem. SOC.,1970 92 3226. z6 P. Vanderlinden and S. Boue J.C.S. Chem. Comm. 1975,932. z7 H. E. Zimmerman and L. M. Tolbert J. Amer. Chem. Soc. 1975,97,5497. Photochemistry 171 system by a cyclopropane ring as in (25) does not curtail photochemical reactivity and several products are formed.28 The path by which the bicyclopropenyl (26a) photochemically rearranges to a benzene derivative was originally thought to involve a prismane ir~termediate.~' However a reinvestigation of this process using suitably labelled bicyclopropenyls (26b c) has shown that a prismane intermediate does not account for the conversion R4 R' RZ R3 R Ph Ph Ph (26) a; R' = R2 = H; R3= R4 = Ph b; R1= H; R2 = Me; R3= R4 Ph c; R' = R2 = Me; R3 = R4 = Ph d; R' = R2 = Ph; R3 = H; R4 = Me e; R' = R2 = Ph; R3= R4 = Me R 2 v Ph (27) P h into the toluenes (2,3,4,5-tetraphenyltolueneand 2,3,4,6-tetraphenyltoluene)or for the isomerization of bicyclopropenyl (26b) into (26d) and of (26c) into (26e).The suggest that a biradical(27) is involved which can either aromatize into the toluenes or else revert to bicyclopropenyls. Hixson aEd Borovsky3' have established in the photoisomerization of the cyclo- propane (28a) that rupture of both bond 'a' yielding (28b) and bond 'b' yielding (29) cccurs. The analysis shows that upon direct irradiation 81% of the isomeriza- tion arises by bond 'a' fission while in sensitized photolysis 98% of the isomerization arises by bond 'a' fission.This indicates that the triplet state shows a greater preference for outside bond fission. Direct irradiation of either of the two alcohols (28) a; R' = H; R2 = CH,OH (29) (30) a; R' = OH; R2= H b; R' = CH20H; R2 = H b; R' = H; R2 = OH 28 H. E. Zimmerman and C. J. Samuel J. Amer. Chem. SOC.,1975,97,448,4025. 29 R. Breslow P. Gal H. W. Chang and L. J. Altman J. Amer. Chem. Soc. 1965,87,5139. 30 R. Weiss and H. Kolbl J. Amer. Gem. Soc. 1975 97 3222 3224. 31 S. S. Hixson and J. Borovsky J. Amer. Chem. Soc. 1975,97 2930. 172 W. M.Horspool (28a) or (28b) also affords ring-opened products (30)32 by a mechanism involving carbocation intermediates.Ring-opening in another system (3 1) has permitted the recognition of the fact that 1,5-hydrogen migration is an important feature in the photolysis of benzon~rcaradienes.~~ (31) In the review of the photochemistry of last year (1974) the report of the photoisomerization of alkenes into carbenes was mentioned.34 This past year (1975) has seen the publication of other reports of this conversion. Thus the direct irradiation of cycloheptene at moderate pressures in the gas phase affords methylenecyclohexane and bicyc1o[4,l70]heptane (32a) which are formed via the carbene (33a).35 The irradiation of 3-phenylcycloheptene in benzene (triplet ?) afforded products (32b c) and 2-phenylmethylenecyclohexane in a total yield of R' ,R2 b (32) a; R' = R2 = H (33) a; R = H (34) (35) b; R' = Ph; R2 = H b; R = Ph C; R' = H; R2 = Ph 20% as well as dimeric or polymeric material.A route to products involving the Hixs0x-1~~ carbene (33b) was post~lated.~~ has shown that irradiation of 1,l-diphenyl-3,3-dimethylbut-l-enein solution phase probably involves a carbene (34) intermediate which subsequently yields 1,2-diphenyl-3,3-dimethylbut-l-ene(not isolated) and the cyclopropane (35). The feasibility of the carbene scheme was demonstrated by deuterium-labelling studies. A re-interpretation of the photochemical conversion of triptycene (36a) into (37) by Iwamura et al.38,39 has implicated a carbene intermediate (38). Originally this reaction was thought4' to be another example of the di-n-methane reaction.Wheeler et d41 have joined in the fray and have examined the photochemistry of the dimethoxytriptycene (36b) which is converted into the aceanthrylene (39). No evidence for the intermediacy of carbenes was obtained. Iwamura and Tukada4* 32 S. S. Hixson and J. Borovsky J.C.S. Chem. Comm. 1975 607. 33 J. S. Swenton K. A. Burdett D. M. Madigan T.Johnson and P. D. Rosso J. Amer. Chem. Soc. 1975,97 3428. 34 T. R. Fields and P. J. Kropp J. Amer. Chem. SOC.,1974,96 7559. 35 Y. Inoue S. Takamuku and H. Sakurai J.C.S. Chem. Comm. 1975,577. 36 S. J. Cristol and C. S. Ilenda J. Amer. Chem. SOC., 1975,97 5862. 37 S. S. Hixson J. Amer. Chem. SOC., 1975,97 1981. 38 H. Iwamura and K. Yoshimura J.Amer. Chem. SOC.,1974,96,2652. 39 H. Iwamura Chem. Letters 1974 5. 40 T. D. Walsh J. Amer. Chem. SOC.,1969,91,515;N. J. Turro M. Tobin L. Friedman and J. B. Hamilton ibid.,p. 516. 41 R.0.Day V. W. Day S. J. Fuerniss and D. M. S. Wheeler J.C.S. Chem. Comm. 1975,296. 42 H. Iwamura and H. Tukada J.C.S. Chem. Comm. 1975,969. Photochemistry 173 .H (36) a;R = H (37) b; R = OMe C; R = OH have suggested that car "enes could be involved in the plotoreaction of (36b) and they have studied the closely related triptycene (36c) which is converted into (40)by a route thought to involve carbenes. The introduction of polar substituents into a 0" (39) (40) barrelene (41) has an effect on the sensitized photochemistry which yields 1,2- dicyanocyclo-octatetraene (20%) (the sole product from the direct irradiation) and two dicyanosemibullvalenes (42a 56%)and (42b 4%).This latter minor product is thought to involve a carbene (43) intermediate.43 (41) (42) a; R' = CN; R2 = H (43) b;R' = H;R2 = CN The intramolecular hydrogen-abstraction reaction encountered44 in the photo- chemistry of the N-(diphenylmethy1ene)acetamides (44a) is reminiscent of Norrish Type I1 hydrogen abstraction in o-methylbenzophenone~~~ or in the reactions of 1-o-tolyl- l-~henylethylene.~~ Hydrogen abstraction by the imino-group in (44a) leads /' ,'NHAc \\ (44)a;R=H (45) b;R=D 43 K. Saito and T. Mukai Bull. Chem. SOC.Japan 1975,48,2334. 44 M. Saeki N. Toshima and H. Hirai Bull. Chem.SOC.Japan 1975,48,476.45 e.g. H. Lutz E. Breheret and L. Lindqvist J.C.S. Faraday I 1973,69,2096. 46 A. C. Pratt J.C.S. Chem. Cornm. 1974 183. 174 W. M.Horspool to an o-quinomethide (45) which in MeOD will deuteriate by exchange on the N-H. Subsequent tautomerization affords the deuterium-incorporation product (44b). Another report on the photochemistry of imines has suggested that the low reactivity of the imine double bond towards hydrogen-abstraction reactions is due to rapid radiationless decay and twisting about the C=N double bond.47 Interest in the photochemistry of sulphur analogues of ketones has been con- tinued. One synthetic application in this area is the synthesis of (*)cuparene (1,2,2- trimethyl-1-p-tolylcyclopentane)by a route involving the desulphurization of the thiol(46) obtained from the photochemical cyclization of the thione (47) in benzene Remote oxidations in steroidal systems continue to be studied.In this p-MeC,H p-MeC,H respect the novel remote oxidation of the C-4 P-methyl group of the steroidal derivative (48) has been A further use of aryl iodide dichlorides for remote oxidation has been published for the photolysis of the cholestanyl aryl iodide (49) in the presence of PhIC12. After work-up unsaturation in the steroidal skeleton was found at the C-16/C-17 site." Irradiation of the esters (50) using the lSO-labelled compounds has demon- strated a previously undetected scrambling of the oxygen This indicates that radical recombination occurs in competition with decarboxylation in ester photolysis.The irradiation of the optically active ester (50b) ([a]i3:-121.8") showed that although the scrambling reaction was taking place the group migration occurred with considerable retention of the stereochemical integrity ([ax; -118.0O). A previous study on the photochemical reactions of benzofuran with sensitizers suggested that dimers were obtained when high-energy sensitizers were used while 47 J. M. Hornback G. S. Proehl and I. J. Starner J. Org. Chem. 1975,40 1077. 48 P. de Mayo and R. Suau J.C.S. Perkin I 1974,2559. 49 J. A. Nelson S. Chou and T. A. Spencer J. Amer. Chem. SOC.,1975,97 648. 5O B. B. Snider R. J. Corcoran and R. Breslow J. Amer. Chem. SOC.,1975,97,6580. 51 R. S. Givens and B. Matuszewski J.Amer. Chem. SOC.,1975,97 5617. Photochemistry 175 (50) a; R = H (51) a; R = Me b;R=Me b; R = Et oxetans were formed from lower-energy sen~itizers.~~ The reinvestigation of the reaction has shown that the high-energy sensitizers (propiophenone and acetophenone) also yield oxetans (51a b) the ratio of products (oxetans dimers) being sensitive to the ratio of the reactants a dependence which has been rationalized in terms of a reversible energy-transfer step in competition with oxetan formation. The results obtained from the study suggest that the energies of pro-piophenone and acetophenone are closer than previously established by low- temperature phosphorescence measurements. In fact the authorss3 suggest that in benzene solution at room temperature the triplets of acetophenone and pro- piophenone are iso-energetic.A study of the photochemistry of the keto-olefins (52),(53) has shown that the inefficiency in product (54) formation is a result of inefficient formation of the biradical (55).54 This biradical is the intermediate which either yields photoproduct (54) or brings about isomerization by a Cope process of (52)to (53) or vice versa. The inefficiency of the biradical formation is due to exciplex formation between the singlet n7r* state and the olefin. (52) (53) (54) (55) The cycloaddition reactions of enones and related systems still constitute a popular area of research. A reinve~tigation~~ of the photochemical reaction of coumarin with tetramethylethylene has confirmed that the photoproduct is the (2 +2) adduct (56),56 formed from both the triplet and the singlet state of the coumarin.Surprisingly the tetramethylethylene completely suppresses the dimerization of the singlet coumarin. (56) (57) a; X = 0 (58) a; X = 0 b; X = CH b; X = CH 52 C. H. Krauch W. Metzner and G. 0.Schenck Chem. Ber. 1966,99,1723. 53 S. Farid S. E. Hartman and C. D. DeBoer J. Amer. Chem. SOC.,1975,97,808. 54 J. C. Dalton and S. J. Tremont J. Amer. Chem. Soc. 1975,97,6916. s5 J. W. Hanifin and E. Cohen TetrahedronLetters 1966 1419. 56 P. P. Wells and H. Morrison J. Amer. Chem. Soc. 1975,97 154. 176 W. M. Horspool Previous work has shown that the dimerization of coumarin from the singlet arises from an exciplex. To account for the suppression of the dimerization the suggest that the mechanism involves the interception of the singlet exciplex by the olefin.The problem associated with charge distribution of cycloadditions to cyc- lohexenones has been examined further using the intramolecular additions encoun- tered in the enones (57) and (58).57 Padwa and DehmS* have reported phenyl migration arising on photochemical excitation of the furanones (59a b) yielding (60) in benzene solution. An odd- electron process (checked by migratory aptitude experiments) is indicated permit- ting phenyl migration to the enone terminus. ."a0 R (59) a; R = H (60) a;R = H b; R = Ph b; R = Ph P,y-Enones have had a special place in organic photochemistry for some time. The interest in these systems was heightened by the publication of theoretical predictions concerning their excited-state reactivity.P,y-Enones in the triplet excited state mainly undergo 1,2-acyl-migrations (an 0x0-di-v-methane reaction) affording cyclopropanes [e.g. (61a) and (61b) are converted into (62a) and (62b) respectively upon acetone sensiti~ation].~~,~~ The singlet reactivity is such that 1,3-acyl shifts result [e.g. the conversion of (61b) into (63)J6' Other work has shown that in some instances 1,3-acyl-migration can occur upon sensitization as well as upon direct irradiation. Thus enone (64) is converted into (65) upon direct as well as (61) a;n = 1 (62) a;n = 1 (63) b;n=2 b;n=2 acetone-sensitized irradiation. In the triplet reaction this product is accompanied by the usual 1,2-acyl-migration prod~ct.~' A further example of sensitized 1,3-acyl- migration has been detected in a reinve~tigation~~ of the photochemistry of the norbornenone (66) which is converted into (67).63This product is also photolabile and is converted by a 1,2-acyl shift into product (68).The reaction of the enone (66) can be sensitized by acetone acetophenone or benzophenone. However the efficiencyof the reaction falls off dramatically in the case of the last two sensitizers. It is particularly surprising that acetophenone should be so inefficient since it has been 57 D. Becker Z. Harel and D. Birnbaum J.C.S. Chem. Comm. 1975,377. 58 A. Padwa and D. Dehm J. Amer. Chem. SOC.,1975,97,4779. 59 R. K. Murray jun. T. K. Morgan jun.and K. A. Babiak J. Org. Chem. 1975,40 1079. 6o R. K. Murray jun. D. L. Goff and R. E. Ratych Tetruhedron Letters 1975,763. 61 P. S. Engel and M. A. Schexnayder J. Amer. Chem. SOC.,1975,97 145. 62 J. Ipaktschi Tetrahedron Letters 1969 2153; Chem. Ber. 1972,105 1840. 63 M. A. Schexnayder and P. S. Engel Tetrahedron Letters 1975 1153. Photochemistry (66) (67) (68) that the triplet energy of (66) is 291.2 kJ mol-’. This value was based on phosphorescence studies but the present paper points out that the enone (66) has no phosphorescence when it is pure. Thus the previous value is in some doubt. Kinetic studies have shown that triplet sensitizers (benzophenone and acetophenone) can form an excited-state complex with the enone and this competes favourably with energy transfer.The problem of interaction other than energy transfer between acetophenone and the enone (69) has also been 0m (69) In an earlier study of the photobehaviour of the cyclohexadienone (70)Schuster et .~~ ~1 interpreted the results from quenching studies using cyclohexa-1,3-diene in terms of two excited states. The nm* state was thought to be responsible for hydrogen abstraction from solvent yielding p-cresol while the mm* state formed the lumiketone. This interpretation was contrary to earlier proposals concerning the excited state responsible for the rearrangement. However in a re-interpretation of the photochemistry of the dienone (70)the recognition of a free-radical process casts doubt on the original p~stulate.~’ Closely associated with the problem of which excited state of a dienone is responsible for the rearrangement is the report of the thermal decomposition of the dioxetans (71).68 This decomposition yields 4,4-diphenylcyclohexa-2,5-dienone 6,6-diphenylbicyclo[3,l,O]hex-3-en-2-one (the usual photoproduct from the photolysis of the dienone) and the corresponding (j CCI Ph ‘Ph (71) a; R = Ph b; R = m-MeOC,H C; R = P-CloH arylmethyl ketone (acetophenone m-methoxyacetophenone or p- acetonaphthone).The authod8 reason that the decomposition of the dioxetan must yield the nn* triplet state of the dienone and it is this which is responsible for the formation of the bicyclic photoproduct in ca. 17% yield. 64 K. G. Hancock and R. 0.Grider J.C.S.Chem. Comrn. 1972,580. 65 P. S. Engel M. A. Schexnayder W. V. Phillips H. Ziffer and J. I. Seeman Tetrahedron Letters 1975 1157. 66 D. I. Schuster and K. V. Prabhu J. Amer. Chem. Soc. 1974,96 3511. 67 D. I. Schuster G. C. Barile and K. Liu J. Amer. Chem. SOC. 1975,97,4441. 68 H. E. Zimmerman and G. E. Keck J. Amer. Chern. Soc. 1975,97,3527. 178 W.M.Horspool An 0x0-di-n-methane rearrangement yielding (72) has been reported from the triple t-sensi tized pho to1 ysis of 2,2,7,7-te trame t h ylcyclo hep ta-3,5 -dien- 1-one .69 Direct irradiation of this dienone yields 2,7-dimethylocta-2,4,6-triene the product of decarbonylation. Population of the S (nn*)state by irradiation of the dienone at 254 nm also yields (72) (from the TI,nn* state) and the triene (from S1 nn*state) but two additional products (73)and (74) are also formed from the upper (S,) (72) (73) (74) Decarbonylation is also found on photolysis of the squaric acid derivative (75).The decarbonylation yields (76a) which is readily desilylated to the deltic acid (76b).70 Other 1,2-dicarbonyl compounds have also been studied particularly where they are known to undergo hydrogen-abstraction reactions. Ogata and Me,SiO OSiMe (75) (76) a; R = SiMe b;R=H Takagi7' recently published results which suggested that a photo-enol was involved in the photochemical conversion of the diones (77) into the hydroxyindanones (78). Hamer7* has suggested however that a triplet biradical (79) is involved in this conversion and by the use of SO as a radical trap (SO2 is a poor dienophile and would not capture the photo-enol) has isolated the adducts of this biradical from the photolysis of the dione in the presence of SO,.R R (77) R = H or Me (78) (79) Research has continued on the photochemical generation of nitrile ylides from azirines. Padwa and Carl~en~~ have examined the photochemistry of the substituted azirines (80) and have observed that the products (81) formed are the result of 1,l-addition. The suggest that molecular constraints prevent the normal addition of the ylide intermediate to the olefin. However rehybridization as 69 J. Eriksen K. Krogh-Jespersen M. A. Ratner and D. 1. Schuster J. Amer. Chem. Sac. 1975,97,5596. 70 D. Eggerding and R. West J. Amer. Chem. SOC.,1975,97 207.71 Y. Ogata and K. Takagi Bull. Chem. SOC.Japan 1974,41,2255;J. Org. Chem. 1974,39 1385. 72 N. K. Hamer J.C.S. Chem. Comm. 1975,557. 73 A. Padwa and P. H. J. Carlsen J. Amer. Chem. SOC.,1975 97 3862. Photochemistry suggested by Salem,74 would permit addition of a bent ylide to the isolated double bond. Interestingly this carbene-like addition takes place with the inversion of the Ph fl H' P(0Et) phYN ;y R' R2 NXo RT+ (82) a; R' = RZ= Me R 1-R2 (80)a; R' = R2 = H R' R2 R2 b; R' = H; R2 = Me b; R' = H; R2 = Ph c; R' = Me;R2 = H (81) c; R' = H; R2 = Me (83) geometry of the rr-system. Schmid and his co-worker~'~ have extended further the scope of the addition of nitrile ylides from the azirines (82) and have studied the addition to diethylbenzoylphosphonate,yielding (83).The addition is regiospecific but not stereospecific. The photoreaction (at 350 nm) of the betaine (84)in ethyl acetate yields a valence isomer (85) and a dimer (86) as the primary photo product^.^^ 00-P T0 03. ) F \+ N N Ph Ph ; NPh H H (84) (85) (86) 74 L. Salem J. Amer. Chem. Soc. 1974,% 3486. ?5 N. Gakis H.Heimgartner and H. Schmid Helv. Chim. Acta 1975 58 748. 76 A. R. Katritzky and H. Wilde J.C.S. Chem. Comm 1975 770.