4. PHOTOCHEMISTRY By A. C. Day (The Dyson Perrins Laboratory Oxford) RECENT general textbooks,’ volume 4 of Advances in Photochemistry and the first volume of a new review series (Organic Photochemistry) all attest to the continued rapid growth of this subject. Literature surveys have appeared for 1965 and 1966,2and during 1966-1967 there have been numerous reviews on specific topics. A wide-ranging yet particularly detailed review by Warrener and Bremner4 discusses systematically the photochemistry of unsaturated systems. It is impossible to overstress the importance of photophysical considerations in mechanistic organic photochemistry ;and relevant topics which have recently been reviewed include excited states’ and fluorescence lifetimes6 of aromatic N.J. Turro ‘Molecular Photochemistry,’ Benjamin New York 1965; J. G. Calvert and J. N. Pitts jun. ‘Photochemistry,’ Wiley New York 1966; R. 0. Kan ‘Organic Photochemistry,’ McGraw-Hill New York 1966; D. C. Neckers ‘Mechanistic Organic Photochemistry,’ Reinhold New York 1967. B. Capon M. J. Perkins and C. W. Rees ‘Organic Reaction Mechanisms 1965,”Interscience London 1966 p. 285; ibid. ‘Organic Reaction Mechanisms 1966,’ Interscience London 1967 p. 369. (a) Olefins G. J. Fonken Org. Photochem. 1967 1 197; (b) Conjugated dienes and trienes R. Srinivasan Ado. Photochelfi. 1966 4 113; (c) Polyenes M. Mousseron ibid. p. 195; (d) Intra- molecular cycloadditions of non-conjugated olefins W. L. Dilling Chem. Rev. 1966 66 373; (e) Cycloaddition reactions 0.L.Chapman and G. Lenz Org. Photochem. 1967,1,283 and R. Steinmetz Fortschr. Chem. Forsch. 1967 7 445; v) Carbonyl compounds R. B. Cundall and A. S. Davies Progr. Reaction Kinetics 1967 4 149 J. N. Pitts jun. and J. K. S. Wan in ‘The Chemistry of the Carbonyl Group,’ ed. S. Patai Wiley London 1965 p. 823 and K. Tokumaru A. Sugimori T. Akiyama and T. Nakata J. SOC. Org. Synthetic Chem. (Japan) 1966 24 1183; (9) Small-ring carbonyl compounds A. Padwa Org. Photochem. 1967,1 91 ; (h) Cyclohexenones and cyclohexa- dienones P. J. Kropp ibid. p. 1 and K. Schaffner Adu. Photochem. 1966 4 81; (i) Photo-Fries reaction :V.I. Stenberg Org. Photochem. 1967,1 127; (j)Quinones J. M. Bruce Quart. Rev. 1967 21 405; (k) Photocyclization of stilbenes F. R. Stermitz Org.Photochem 1967 1 247; (I) Photo-chromism and reversible photoisomerisation E. Fisher Fortschr. Chem. Forsch. 1967 7 605 ; (m)Troponoid compounds D. J. Pasto Org. Photochem. 1967,1,155 and K. F. Koch Adv. Alicyclic Chem. 1966,1 257; (n)Photoalkylation D. Elad Fortschr. Chem. Forsch. 1967 7 528; (0)Photo-oxbation :M. Pap ibid. p. 559; (p) Photo-oxidation :K. Gollnick and G. 0.Schenck in ‘1,CCyclo- addition Reactions,’ ed. J. Hamer Academic Press New York 1967 p. 255 and M. Niclause J. Lemaire and M. Letort Adv. Photochem. 1966,4 25; (4)Diazirines H. M. Frey ibid. p. 225; (r) Chemiluminescence F. McCapra Quart. Rev. 1966,20,485 ;R. F. Vassil’ev Progr. Reaction Kinetics 1967 4 305; (s) Nucleic acids K. L. Wierzchowski Postepy Biochem. 1967 13 127; L. Musajo Chimica e Industria 1967,49,131; (t)Flavins S.Paszyc Postepy Biochem. 1967,13,161. R. N. Warrener and J. B. Bremner Rev. Pure Appl. Chem. (Australia) 1966 16 117. W. A. Noyes jun. and I. Unger Adv. Photochem. 1966,4,49. J. B. Birks and I. H. Munro Progr. Reaction Kinetics 1967 4 239. 162 A. C. Day molecules energy transfer,' the triplet state,** and photochemistry and re- action kinetics." General.-Kearns and his co-workers have developed the phosphorescence excitation technique for the measurement of singlet -+ triplet (So + T) absorption spectra and recorded So + T spectra for aromatic hydrocarbons ketones and aldehydes. l1 The method is simple in practice highly sensitive and much less influenced by impurities than other methods of determining So-+ T spectra.Transitions involving n,n* states are less susceptible to heavy atom effects than are those involving x,n* states;12 and by use of heavy-atom solvents it is possible to distinguish between S+T,,,%* and S+K,%*transitions in the phosphorescence excitation method. Media containing heavy atoms (including xenon) also promote non-radiative S-+ T intersystem crossing,13 and Cowan and Drisko have provided a chemical demonstration of this effect. l4 In the photodimerisation of acenaph-thylene the cis-dimer is formed predominantly via an excited singlet state or excimer whereas the trans-dimer is formed uia the triplet state. Irradiation of Ph wh \ 0 h-0-co (3) (4) F. Wilkinson Adv. Photochem. 1964 3 241; Quart. Rev. 1966 20 403; R.G. Bennett and R. E. Kellogg Progr. Reaction Kinetics 1967 4 215. S. K. Lower and M. A. El-Sayed Chem. Rev. 1966,66,199. 'The Triplet State,' Proceedings of a Symposium held at the American University of Beirut 1967 ed. A. B. Zahlan G. M. Androes H. F. Hameka J. H. van der Waals F. W. Heineken C. A. Hutchinson jun. and G. W. Robinson Cambridge 1967. lo 'Photochemistry and Reaction Kinetics,' ed. P. G. Ashmore F. S. Dainton and T. M. Sugden Cambridge 1967. R. F. Borkman and D. R. Kearns Chem. Comm. 1966 446; D. R. Kearns and W. A. Case J.Amer. Chem. Soc. 1966,88,5087; A. P.Marchetti and D. R. Kearns ibid. 1967,89,768; W. Rothman and D. R. Kearns Photochem. and Photobiol. 1967,6,775. l2 M. A. El-Sayed J. Chem. Phys. 1964,41 2462. l3 A.R. Horrocks T. Medinger and F. Wilkinson Photochem. ad Photobiol. 1967,6 21. l4 D. 0.Cowan and R. L. Drisko Tetrahedron Letters 1967 1255; J. Amer. Chem. SOC. 1967 89 3068. Photochemistry R$3 R (9) a:R=H a:R=H b R = CO,H b R = CO,H acenaphthylene in solvents containing ethyl iodide or n-propyl bromide produced much greater yields of trans-dimer than were obtained in solvents containing only light atoms.14 In contrast no heavy-atom effect was detected in the dimerisation of coumarin” or the Type I1 fission of aliphatic ketones.16 These reactions both involve n,n* states and the absence of a heavy-atom effect is thereforeI2 understandable. The use of optical rotatory dispersion measurements coupled with flash photolysis has been suggested as a possible method of studying structural changes occurring on electronic excitation as well as the absolute configura- tion of excited states.Preliminary work showed that excited states of (+)-and (-)-benzoin give roughly mirror-image 0.r.d. curves.17 Full details of a colour test for compounds having short-lived triplet states have appeared.18 This is based on the photochemical valence tautomerisation of 2,3-epoxy-2,3-diphenyl-indanone (1) to the diphenylbenzopyrylium oxide (2). Excitation of the benzophenone chromophore in 3-(a-naphthyl)-5a-androstan- 17p-yl p-benzoylbenzoate (3) results in intramolecular transfer of triplet energy to the naphthalene residue which then phosphoresces. The analogous 9’-carbazolylacetate behaved similarly.1g Chemiluminescence in- ’ H.Morrison H. Curtis and T. McDowell J. Amer. Chem. SOC.,1966,88,5415; cf. H. Morrison and H. Curtis Abstr. 151st Meeting Amer. Chem. SOC. March 22 1966 K55. l6 P. J. Wagner J. Chem. Phys. 1966,452335. l7 P. A. Carapellucci H. H. Richtol and R. L. Strong J. Amer. Chem. SOC.,1967,89,1742. E. F. Ullman and W. A. Henderson jun. J. Amer. Chem. Soc. 1967,89,4390. l9 R. A. Keller and L. J. Dolby J. Amer. Chem. SOC.,1967,89,2768. * Application of the appropriate orbital symmetry rules6 to the concerted thermal reversion of an anti-4a.4b-dihydrophenanthreneto the corresponding stilbene leads to an impossible situation ! F* 164 A. C.Day volving intramolecular energy transfer has been observed by White and Roswell:20 e.g.oxidation of the luminol analogue (4) leads via the singlet excited state of the dicarboxylate di-anion to chemiluminescence at the fluorescence wavelengths of 9,lO-diphenylanthracene. Several observations by Hammond’s group demonstrate that efficient inter- molecular transfer of triplet energy requires close contact between donor and acceptor. The efficiency of benzophenones as sensitisers2 and metal complexes as quenchers22 is reduced by the. introduction of bulky alkyl groups. Asym- metric induction was found in the photosensitised cis-trans isomerisation of 1,2-diphenylcyclopropaneswith an optically active ~ensitiser.~~ Cautions have appeared concerning the use of diene~~~ and triphenylene2 in triplet-transfer studies. Care is also necessary in the general Stern-Volmer treatment of quenching processes.26 ‘Topochemical’ studies of the relationship between crystal structure and solid-state photochemistry have been extended to heterocyclic analogues of cinnamic acid2’ and to muconic acid derivatives.28 Irradiation of poly(viny1 cinnamate) produces spectral changes in the U.V.spectrum similar to those observed in the photosensitised cyclodimerisation of ethyl inna am ate.^' Two groups have used polymeric phenyl ketones as heterogeneous photosensiti~ers.~~ Comparative studies of the photochemical behaviour of molecules in solution and adsorbed on silica gel have been reported.31 Mercury-sensitised photolytic degradation coupled with gas chromato- graphy has been investigated as a tool for organic structure determinati~n.~~ Several photochemical techniques have been described.33 Olefms.”304 Two recent interesting developments relate to very simple olefins.’O E. H. White and D. F. Roswell J. Amer. Chem. SOC. 1967,89,3944. W. G. Herkstroeter L.B. Jones and G. S. Hammond J. Amer. Chem. SOC. 1966,884777. 22 A. J. Fry,R. S. H. Liu and G. S. Hammond J. Amer. Chem. SOC. 1966,88,4781. ” G. S. Hammond and R. S. Cole J. Amer. Chem. SOC. 1965,87,3256. 24 L. M. Stephenson D. G. Whitten G. F. Vesley and G. S. Hammond J. Amer. Chem SOC. 1966,88,3665,3893; S. D. Andrews and A. C. Day Chem. Comm. 1967,477. ’’ W. M. Hardham and G. S. Hammond J. Amer. Chem. SOC.,1967,89,3200 footnote 27. 26 R. J. Campbell E. W. Schlag and B. W. Ristow J. Amer. Chem. SOC.1967 89 5098; P. J. Wagner ibid. p. 5715. ” M. Lahav and G. M. J. Schmidt J. Chem. SOC. (B) 1967,239; cf. J. Rennert E. M. Ruggiero and J. Rapp Photochem. and Photobiol. 1967,6,29. ” M. Lahav and G. M. J. Schmidt J. Chem. SOC. (B) 1967,312. ’9 H. G. Curme C. C. Natale and D. J. Kelley J. Phys. Chem. 1967,71,767. 30 R. Searle J. L. R. Williams J. C. Doty D. E. deMeyer S. H. Merrill and T. M. Laakso Makromol. Chem. 1967,107,246; P. A. Leermakers and F. C. James J. Org. Chem 1967,32,2898. 31 J. L. Ruhlen and P. A. Leermakers J.Amer. Chem. SOC. 1967,89,4944; T. R. Evans A. F. Toth and P. A. Leermakers ibid. p. 5060; C. Balny and P. Douzou Compt. rend. 1967,264 C 417. 32 R. S.Juvet jun. R. L. Tanner and J. C. Y.Tsao J. Gas Chromatog. 1967,5 15. 33 S.D.Cohen M. V. Mijovic G. A. Newman and E. Pitts Chem. and Ind. 1967,1079; D. Bryce- Smith J. A. Frost and A. Gilbert Nature 1967,213,1121; D. A. Warwick and C. J. H. Wells J. Sci. Instr. 1967 44 483; J. H. Allen and J. F. McKellar Lab. Practice 1967 16 991. * The phenomenology of organic photochemistry presents some difficulty and it has been decided to systematise this year’s Report according to compound rather than reaction type. This naturally leads to some dispersal of closely related material e.g. on cycloadditions. Cross-referencing within the compound divisions will it is hoped mitigate this situation somewhat. The ‘orthogonal’ classi- fication into reaction types presents at least asmany problems. Photochemistry 165 1- Alkylcyclohexenes and 1-methylcycloheptene undergo a light-induced ionic reaction in alcohols containing aromatic sensitisers such as benzene toluene or ~ylene.~~’ 35 Products are the corresponding exocyclic olefin and tertiary ether formed by Markovnikov addition across the double bond.For example menthene (5)irradiated in 0-deuteriomethanol containing benzene gave the deuteriated olefin (6)and stereoisomeric ethers (7).34In two cases,349 36 molecular rearrangements characteristic of cationic species were observed and an ionic mechanism involving protonation has been proposed. 34 Reaction is very slow with 1-methylcyclopentene and is not observed at all with 1- methylcyclo-octene acyclic or highly strained 01efins.~~’ In methanol containing xylene norbornene gave only dimers and products characteristic of radical reactions e.g.(8).37 From the premise that an olefinic n,n* triplet prefers to exist in the orthogonal conformation Kr~pp~~. 37 has rationalised the photochemical behaviour of cyclic olefins. Triplets of larger-ring cyclic olefins and acyclic and exocyclic olefins can decay to cis-or trans-olefin with an efficiency which precludes alternative processes. In cyclohexenes and cyclo- heptenes the orthogonal triplet (or derived trans-olefin) relieves excessive strain mainly by protonation although hydrogen-atom abstraction becomes possible in an aprotic solvent.34 In strained olefins the orthogonal conformation is impossible and the triplet state may therefore be sufficiently long-lived or energetic to take part in intermolecular processes of a radical nature.Cyclo- pentenes are apparently a borderline case between the last two categories. Formally analogous reactions have been observed with phenols ;3 mechanistic-ally though these reactions probably involve the well known enhancement of phenol acidity in the singlet excited state. In favour of this explanation is the close similarity in isomer distribution between products from photocyclisa- tions and ‘dark’ acid-catalysed reactions of o-allyl- and o-b~t-3’-enylphenols.~~~ Irradiation of tetramethylethylene for long periods with unfiltered light from a medium pressure mercury lamp gives octamethylcyclobutane in reasonable yield.39 Reaction involves the n,n*singlet state for which dimerisa- tion appears to be fast enough to compete with other deactivation processes (e-g.,intersystem crossing).This is in marked contrast to the n,n*triplet situation in simple olefins where deactivation is so rapid that dimerisation cannot com- pete. In fact sensitised dimerisation is commonly only observed in cases of exceptional strain (cf. ref. 37) e.g. 1,2,3-triphenyl~yclopropene,~~ bridged and intramolecular ‘caging’ processes.41b The dimerisation of 34 P. J. Kropp J. Amer. Chem. SOC. 1966 88 4090; P. J. Kropp and H. J. Krauss ibid. 1967 89 5199. 35 J. A. Marshall and R. D. Carroll J. Amer. Chem. SOC.,1966,88,4092. 36 J. A. Marshall and A. R. Hochstetler Chem. Comm. 1967 732. ” P. J. Kropp J. Amer. Chem. SOC.,1967,89,3650. 38 (a) W. M. Horspool and P. L. Pauson Chem.Comm. 1967 195; (b)G. Frhter and H. Schmid Helv. Chim. Acta 1967 50,255. 39 D. R. Arnold and V.Y. Abraitys Chem. Comm. 1967,1053. 40 C. Deboer and R. Breslow Tetrahedron Letters 1967,1033 ;H. Diirr ibid. p. 1649. *’ Znter ah (a) H. D. Scharf Tetrahedron 1967 23 3057; (b) C. G. Chin H. W. Cuts and S. Masamune Chem. Comm. 1966 880; J. C. Barborak and R. Pettit J. Amer. Chem. SOC. 1967,89 3080. 166 A. C.Day tetramethylethylene is of some significance since examples involving simple non-conjugated olefins are rare.42 As a (2 + 2)7t type cycloaddition it is photo- chemically allowed on orbital symmetry grounds.43 The tetracyclononene (9a) undergoes ‘caging’ to the saturated compound (1Oa) on direct photolysis in ether.44 This internal cycloaddition of an ethylenic bond to a cyclopropane must involve singlet states since triplet sensitisation with acetone gave only dimers( 1 l) together with an acetone-addition pr~duct.~’ The carbonyl-containing analogue (9b) gives the ‘cage’ isomer ( Analogues of the photochemical valenq isomerisation of bicyclo[2,2,l]hepta- diene to q~adricyclene~’ have been described.48 The reaction of the (less strained) cyclohexa-1,4-diene (1 2) follows a different course giving the isomer (13) presumably by the sequence shown.49 This rearrangement which bears a formal resemblance to some light-induced reactions of cycl~hexadienones,~” is one example of a general photochemical reaction in which a divinylmethane is converted into a vinylcycl~propane.~~ Most of the examples listed by Zimmerman et a1.” are relatively complex but its seems possible that the mercury-photosensitised rearrangement of acyclic 1,4-dienes to vinylcyclo- propanes’ proceeds similarly.An example is the rearrangement of 3,3-dimethylpenta-l,6diene (14) to the cyclopropane (16) and isomeric diene (15); (1 5) is also convertible into (16) on sensitised irradiation. Meinwald and Smith have suggested a homolysis-recombination mechanism for reactions of this kind.’ la Other products of mercury-sensitised rearrangement of 1,4-dienes are bicyclo[2,1,0]cyclopentanes5 and bicycle[ l,l,l]pentanes.’ lb Srinivasan and Carlough’ lb studied the sensitised photolysis of several 1,4- 1,5- and 1,6-dienes and obtained in all cases except cyclo-octa-1,5-diene (vide inpa) a mixture of two internal adducts a bicyclo[n,2,0]alkane (‘parallel adduct’) and a bicyclo[n 1 llalkane (‘crossed adduct’).The crossed adduct :-parallel adduct ratio was found to be dependent on chain length values of 0.10 233 and 0-04 being found for penta-1,4-diene (17) hexa-1,5-diene (18) and hepta-1,6- diene (19) respectively. To explain the selectivity of these reactions Srinivasan and Carlough suggested that cycloaddition is a two-step process and that the 42 G. S. Hammond N. J. Turro and A. Fischer J. Amer. Chem. SOC.,1961,83,4674;J. R. Chesick ibid. 1963,85 3718. 43 R. Hoffmann and R. B. Woodward J. Amer. Chem. SOC. 1965,87,2046. 44 E. Wiskott and P. von R. Schleyer Angew. Chem. 1967,79,680 (Angew. Chem. Internat. Edn. 1967,6 694).4s H.-D. Scharf and G. Weisgerber Tetrahedron Letters 1967 1567. 46 C. F. Huebner E. Donoghue L. Dorfman E. Wenkert W. E. Streth and S. W. Donely Chem. Comm. 1966,419; P. K. Freeman and D. M. Balls J. Org. Chern. 1967,3& 2354; H. Prinzbach and D. Hunkler Angew. Chem. 1967,79,232 (Angew. Chem. Internat. Edn. 1967,6,247). 47 G. S. Hammond P. Wyatt C. D. DeBoer and N. J. Turro J. Amer. Chem. SOC.,1964,86,2532. 48 E. Payo L. Cortts J. Mantech C. Rivas and G. de Pinto Tetrahedron Letters 1967 2415; H. Prinzbach and J. Rivier ibid. p. 3713. 49 W. Reusch and D. W. Frey Tetrahedron Letters 1967 5193. H. E. Zimmerman R. W. Binkley R. S. Givens and M. A. Sherwin J. Amer. Chem. SOC. 1967,89 3932 and refs. therein cited. (a) J. Meinwald and G. W.Smith J. Amer. Chern. SOC.,1967,89 4923; (b) R. Srinivasan and K. H. Carlough ibid. p. 4932. 167 Photochemistry c -a:-a preferred initial step is formation of afiue-membered ring cf. (20) (21) and (22). The second step then gives the preferred adducts (23) (24) and (25). Formation of the alternative adducts cannot involve biradicals containing a five-membered ring. Cyclo-octa-l,5-diene (26) gives solely ( >97 %) the crossed adduct (27),”’ via a biradical analogous to (21).52 Internal cycloadditions of acyclic trienes have been discussed in similar terms by Hammond and L~u.~~ Photoisomerisa-tion of o-divinylbenzene (28) gives (29) not the expected bicyclo[2,1,1]- compound (30).Deuterium-labelling studies showed that the reaction involves a rearrangement of the carbon skeleton rather than hydrogen rnigrati~n.’~ A different type of behaviour is shown by the 1,5-diene (31).Photolysis at 254 52 I. Haller and R. Srinivasan J. Amer. Chem. SOC.,1966,88 5084;cf. J. E.Baldwin and R. H. Greeley ibid.,1965,87 4514. 53 R.S. H. Liu and G. S. Hammond J. Amer. Chem. SOC.,1967,89,4936. 54 M. Pomerantz J. Amer. Chem. SOC.,1967,89,694;J. Meinwald and P. H. Mazzocchi ibid. p. 696. 168 A. C. Day mp gives the cis-trans isomers (32) and (33). The position of the deuterium in the products rules out a Cope rearrangement and the reaction appears to involve a concerted 1,3-shift of the 3-deuterioallyl group without allylic in~ersion.~~ Numerous electro~yclic~~ reactions involving conjugated olefins have appeared.The highly strained cis-trans-diene (34) is an intermediate in the photosensitised conversion of cis-cis-cyclo-octa-l,3-diene(35)into the bicyclic compound (36) in boiling benzene. The cis-trans-diene (34) can be isolated in 85 % yield from photolyses conducted at a lower temperature and it is quanti- tatively converted into (36) at 80” by a thermally all~wed,’~ conrotatory (30) (28) H (35) (37) (38) ’’ R.F. C.Brown R C. Cookson and J. Hudec Chem. Comm. 1967,823. 56 R. B. Woodward and R. Hoffman J. Amer. Chem. SOL 1965,87,395; H.C.Longuet-Higgins and E.W. Abrahamson ibid. p. 2045. Photochemistry 169 process.57cis-cis-Octa- 1,3,5,7-tetraene has been detected in the flash photolysis of cyclo-octa-1,3,5-triene in solution.58" In previous (preparative) studies only the bicyclic and tricyclic isomers (37) and (38) had been obtained.58b Cyclohexa- 1,3-diene photoisomerises initially to hexa-1,3,5-triene ; prolonged photolysis leads to a 1 :1 mixture of bicyclo[3,1,0]hex-2-ene (39) and 3-vinyl~yclobutene.~~" The cyclohexadiene-+bicyclo[3,l,O]hexene conversion which seems to be rather common,4* 6o has been discussed in terms of several mechanisms desig- nated by Meinwald and MazzocchiSb paths a b and c(see formulae).Photolysis of the labelled triene (40) gave the bicyclohexene (41) labelled as shown. Path c can therefore be eliminated since it would have given uiu the symmetrical intermediate (42),both (41) and the isomer with deuterium and the indicated hydrogen inter~hanged.~ Two photoisomerisations of cycloheptatriene are 9b known valence tautomerisation to bicyclo[3,2,0]hepta-2,6-diene (43) and a 1,7-shift of hydrogen.Molecular orbital considerations show that the 1,7-shift is allowed in the first excited state.61* 62 A similar 1,7-shift of an alkyl group has been observed in the photolysis of the trimethylcycloheptatriene(44) which Q '' R. S. H. Liu J. Amer. Chem. SOC.,1967,89 112. (a) T. D. Goldfarb and L. Lindqvist J. Amer. Chem. SOC.,1967,89,4588 ;(b) 0.L. Chapman G. W. Borden R. W. King and B. Winkler J. Amer. Chem. SOC.,1964,86 2660; W. R. Roth and B. Peltzer Angew. Chem. Internat. Edn. 1964 3 440; 1.Zirner and S. Winstein Proc. Chem. SOC. 1964,235.59 J. Meinwald and P. H. Mazxocchi J. Amer. Chem. SOC.(a) 1966,88,2850; (b)1967,89 1755. 6o W. G. Dauben and J. H. Smith J. Urg. Chem. 1967,32,3244. 61 R. B. Woodward and R. Hoffmann J. Amer. Chem. SOC.,1965,87,2511. G. W. Borden 0.L. Chapman R. Swindell and T. Tezuka J. Amer. Chem. Soc. 1967,89,2979. 170 A. C. Day gives (45)and (46)as primary photo product^.^^ In 7-substituted cyclohepta- trienes the relative importance of the two processes depends on the nature of the substituent.62* 64 Substituent dependence is also shown in the photo- chemistry of 7,8-dimethylenecyclo-octa-1,3,5-trienes(47). The unsubstituted compound (47; R = H)gave (48),whilst the chloro-compound (47;R = C1) gave (49)and (50) i.e. chlorine deactivates the em-diene and the compound behaves like a conjugated cyclo-~ctatriene.~~ Intriguing thermal and photochemical inter-relationships exist between a series of compounds of general formula (CH),, e.g.bullvalene (51),the 9,lO-dihydronaphthalenes (52) and (53) and Nenitzescu's hydrocarbon (54).The formulae indicate those interconversions which seem at the present time to be definitely established.66 The tetracyclic compound (59,a conceivable inter- mediate in some of these transformations still resists detection. The identifica- tion of the (4n + 2)-hydrocarbon cyclodecapentaene (56),seems secure but its geometry is as yet unknown.67 The photochemistry of the (CH) series is not yet so rich in detail. Barrelene (57) undergoes photosensitised isomerisation to semibullvalene (58) and cyclo-octatetraene.68 Studies" with deuteriated barrelene show that the conversion of (57) into (58)involves the divinylmethane- vinylcyclopropane rearrangement referred to above.Photoadditions across the ethylenic bond have been reported with cyclic ethers,69 f~rmamide,~' ~hloramine,'~~ nitro soar nine^,^' N-chl~rourethane,~~" and haloacetic acids.73 Most of these seem to be straightforward radical reactions some of synthetic value ;and rather similar reactions with saturated compounds have been described74 (cf. refs. 3n and 30). 63 L. B. Jones and V. K. Jones J. Amer. Chem. SOC. 1967,89 1880. 64 A. P. ter Borg E. Razenberg and H. Kloosterziel Chem. Comm. 1967 1210. 65 J. A. Elix M. V. Sargent and F. Sondheimer 1. Amer. Chem. SOC.1967 89 180 5081; cf. ref. 58b. 66 M. Jones jun. J. Amer. Chem. SOC. 1967 89 4236 and refs. therein cited; G. Schroder and J. F. M. Oth Angew. Chem. Internat. Edn. 1967,6,414. 67 E. E. van Tamelen and T. L. Burkoth J. Amer. Chem. SOC. 1967,89 151. 68 H. E. Zimmerman and G. L. Grunwald,J. Amer. Chem. SOC. 1966,88 183; cf.,J. P. N. Brewer and H. Heaney Chem. Comm. 1967,811. 69 I. Rosenthal and D. Elad Tetrahedron 1967,23 3193. 70 J. Rokach and D. Elad J. Org. Chem. 1966,31 4210; cf. M. Pfau and R. Dulou Bull. SOC. chim. France 1967,3336. 71 Y. L. Chow C. Colon and S. C. Chen J. Org. Chem. 1967,32,2109. 72 (a) K. Schrage Tetrahedron Letters 1966 5795; Tetrahedron 1967 23 3033; (b) Y. Ogata Y. Izawa and H. Tomioka Tetrahedron 1967,23 1509. 73 N. Kharasch P.Lewis and R. K. Sharma Chem. Comm.,1967,435. 74 R. C. Cookson J. Hudec and N. A. Mirza Chem Comm.,1967,824; C. Pac and S. Tsutsumi Bull. Chem. SOC. Japan 1966,39 1926; Y. Shigemitsu T. Tominaga T. Shimodaira Y. Odaira and S. Tsutsumi ibid. p. 2463. Photochemistry Y = H; Z = C1 orY = C1:Z = H (52) z (51) A A '"\ A t (53) (55) (57) (58) Photocycloadditions involving olefins are discussed in the sections on car- bony1 compounds aromatic compounds and quinones. Carbonyl Compounds. 3f-h.-The photochemistry of very simple carbonyl compounds many in the vapour phase has received attention.75 It has been '' (a) Inter al. B. A. Degraff and J. G. Calvert J. Amer. Chem. SOC.,1967,89,2247; P. J. Wagner ibid. 1966,88 5672; J.C. W. Chien ibid. 1967,89 1275; E. K. C. Lee J. Phys. Chem. 1967,71,2804; E. K. C. Lee and N. W. Lee ibid. p. 1167; R. E. Rebbert and P. Ausloos J. Amer. Chem. SOC. 1967 89 1573; M. J. Yee Quee and J. C. J. Thynne Trans. Faraday SOC. 1967 63 1656; 1966,62 3154; (b)P. J. Wagner J. Amer. Chem. SOC. 1967,89 2503; Tetrahedron Letters 1967 1753. 172 A. C.Day suggested that the Type-I1 photofragmentation of ketones may be a concerted six-centre process in the singlet-excited state but a two-step process initiated by abstraction of y-hydrogen in the triplet state.75b* 76 The vapour-phase photolysis of a series of ketones n-C,H COR has been studied correlations between structure and efficiency of the Type I1 fission were found.76 Ketones are photoreduced by amine~~~ Contrary and trib~tylstannane.~’~ to e~pectation,~~ singlet-excited acetone is less reactive in hydrogen abstraction than the triplet by a factor of 1000.75b Benzophenones with an o-alkyl substituent give photoenols on irradiation.Further examples have been reported this year.’’ Enolic species of a different kind e.g. ‘isobenzpinacol’ (59) have been detected spectroscopically in the photolysis of benzophenone in isopropyl alcohol and other hydrogen-donating solvents.79 (60) %c 76 C. H. Nicol and J. G. Calvert J. Amer. Chem. SOC.,1%7,89,1790. 77 S. G. Cohen and J. I. Cohen J. Amer. Chem. SOC.,1967 89 164; S. G. Cohen and R. J. Baumgarten ibid. p. 3471. 78 P. J. Wagner and G. S. Hammond J. Amer. Chem. SOC.1966,88 1245. 79 G. 0. Schenck M. Cziesla K. Eppinger G. Matthias and M. Pap Tetrahedron Letters 1967,193. Photochemistry 173 The photoisomerisation of ketones containing a cyclopropane ring con- jugated with carbonyl to ap-unsaturated ketones involves specific breakage of that cyclopropane bond which overlaps best with the carbonyl x-bond. Thus Dauben and Shaffer have found that in solution n,x* excitation of a series of alkylated bicyclo[4,1,0]heptan-2-ones results in fission of the 1-7 bond (60)-*(61)-*(62). In no case was 1-6 cleavage observed." When as in (63) both ap-bonds in the cyclopropane ring overlap equally with carbonyl cleavage gives the more highly substituted biradical (64).809 Cyclopropane ring opening however only competes with cleavage of the 2-3 bond [cf.(60)] when C-3 is unsubstituted or C-7 is substituted.80 A similar but slightly less specific cyclopropane cleavage has been observed for bicyclo[3,1 ,O] hexan-2- one in the vapour phase.82 An elegant labelling study by Beugelmans defines the stereochemistry of rearrangement of a 3,5-cyclosteroidal ketone (65). Photolysis gave a mixture of the two unsaturated ketones (66) and (67) by respectively 4a+3 hydrogen and 4p+5 deuterium shifts.83 A discussion of the U.V. spectra of cyclopropyl ketones which is of relevance to the foregoing has been given by Dauben and Bere~in.'~ The photochemical rearrangement of ap-epoxy-ketones to P-diketonesy38 (68)-+(70) is characterised by an unusual order of migratory aptitudes of p groups (Ph,CH and PhCHz > H > RCH > Me % Ph).85 The initial cleavage of the C-0 bond to (69) resembles the cyclopropane cleavage already discussed.Recent studies'' seem to favour a concerted mechanism for the rearrangement of (69) to (70) rather than two-step fragmentation in- volving a caged-radical pair (71) although there is some evidence to suggest that (69) may be diverted to (71) at least in part when the group p can form a particularly stable radical.85" The fission of ap-bonds in cyclopropyl and epoxy- ketones has a formal analogy in the x*-assisted elimination of the substituent in a-substituted ketones as (72).86 For the photolysis of py-epoxy ketones see Padwa et 'O W. G. Dauben and G. W. Shaffer Tetrahedron Letters 1967,4415. R.E. K. Winter and R. F. Lindauer Tetrahedron Letters 1967,2345. '' L. D. Hess and J. N. Pitts jun. J. Amer. Chem. SOC.,1967,89,1973; L. D. Hess J. L. Jacobson K. Schaffner and J. N. Pitts jun. ibid. p. 3684. " R. Beugelmans Bull. SOC.chim. France 1967,244. 84 W. G. Dauben and G. H. Berezin J. Amer. Chem. SOC.,1967,89 3449. 85 (a) C. S. Markos and W. Reusch J. Amer. Chem. SOC. 1967 89 3363; (6) refs. cited in (a) especially H. Wehrli C. Lehmann P. Keller J. J. Bonet K. Schaffner and 0.Jeger Helu. Chim. Acta 1966,49 2218. 86 Literature cited in ref. 82 and C. L. McIntosh P. de Mayo and R. W. Yip Tetrahedron Letters 1967,37; Y. Saburi K. Minami and T. Yoshimoto J. Chem. SOC. Japan 1967,88,557. '' A. Padwa D. Crumiine,R. Hartman and R. Layton J. Amer.Chem SOC. 1967,89,4435. 174 A. C.Day Photolysis of cyclobutanone in the vapour phase involves Type I fission to the biradical CH CH CH CO which may fragment to keten and ethylene or lose carbon monoxide. Decarbonylation leads to cyclopropane in a vibra- tionally excited state which may be collisionally deactivated or rearrange to propene.” A third reaction which becomes competitive with these processes in solution is cyclisation without decarbonylation to give an oxacarbene. Thus 2,2-dimethylcyclobutanone(73) photolysed in methanol gives (75) via (74).” The process is structurally specific cleavage only occurring at the more highly substituted a-bond. The importance of the oxacarbene pathway relative to keten and cyclopropane formation increases with increasing substitution at the a-carbon atoms.” An attempt by Hostettler to trap with olefins an oxacarbene in the photolysis of 2,2,4,4-tetramethylcyclobutanonesfailed perhaps because of steric hindrance,” but an analogous oxacarbene (76) from benzocyclobutenedione did undergo carbenoid addition to olefins.” The reaction is not confined to cyclobutanones for the strained ( +)-cyclocampha-none (77) in cyclohexene gave the adduct (78).’ An oxacarbene may also be involved in the photolysis of ( f)-fenchone (79) in aqueous ethanol to give (80).93 Photolysis of 3-methylenecyclobutanone in a glass at -196” gave triplet *’ H.0.Denschlag and E. K. C. Lee J. Amer. Chern. SOC. 1967,89 4795; R. J. Campbell and E. W. Schlag ibid. p.5103. 89 N. J. Turro and R. M. Southam Tetrahedron Letters 1967,545. H. U. Hostettler,Helu. Chim. Actu 1966,49 2417. 91 H. A. Staab and J. Ipaktschi Tetrahedron Letters 1966,583. 92 P. Yates and L. Kilmurry J. Amer. Chem. SOC.,1966,88,1563. 93 P. Yates and A. G. Fallis Tetrahedron Letters 1967 4621 ;cf. G. E. Gream J. C. Paice and C. C. R. Ramsay Austral. J. Chem. 1967,20,1671. Photochemistry 60 (79) @PhPh Ph4 (86) Ph a:R=Me h:R = Ph R (88) (89) (a Ar = p-C6H4.CN) (b Ar = p-C6H4*OMe) trimethylenemethane identified by its e.s.r. spectrum.94 Stereospecific transfer of the em-hydrogen atom occurs in the photolysis of carvone camphor in methanol e.g. (81)+(82).95 A heavy-atom effect (see General Section) has been observed in a study of the photolysis of non-enolisable ~-diketonesg6-somewhat surprising in a reaction which probably involves n,n* states.Mesityl oxide is per se photostable but when irradiated with 1849 8 light in methanol or isopropyl alcohol it gives several products in a reaction which is probably initiated by photodecomposition of the Vinylic esters derived from dimedone undergo photohydrolysis in an aqueous medium probably by an initial hydration of the ethylenic bond.98 The stereochemistry of addition of alcohols to ap-unsaturated ketones to give P-alkoxy-ketones has been studied.99 94 P. Dowd and K. Sachdev J. Amer. Chem. SOC. 1961,89,715. 95 J. Meinwald R. A. Schneider and A. F. Thomas J. Amer. Chem. SOC.,1967,89,70. 96 H.Nozaki Z. Yamaguti T.Okada R. Noyori and M. Kawanisi Tetrahedron 1967,23,3993. ’’)N.C.Yang and Do-Minh Thap J. Org. Chem. 1967,32,2462. ’* P.de Mayo and J. S. Wasson Chem. Comni..1967,970. 99 B.J. Ramey and P. D. Gardner J. Amer. Chem. SOC.,1967,89,3949. 176 A. C. Day Bridged py-unsaturated ketones (83) are isomerised by light to the bicyclic cyclobutanones (84). loo Cyclo-oct-4-enone undergoes photoinduced cleavage of the 2-3-bond to give the biradical (85) which subsequently recyclises to form 3-vinylcyclohexanone as major product. l The cleavage is unprecedented in solution photochemistry-though reminiscent of the z*-assisted cleavages noted above86-and is probably a conformational effect. The interest in cyclo hexenones and cross-conj ugated cyclo hexadienone~~~ continues unabated.The 'type A' rearrangement of cyclohexenones exemplified by the conversion of 4,4-dimethylcyclohexenone(86a) into (87) is not followed by the 4,4-diphenyl compound (86b) which gives instead the stereoisomeric products (88) formed by phenyl migration."' 'Type A' rearrangement though predominates in the rearrangement of 3,5-diphenylcyclohexenone(89). The difference in behaviour between (86b) and (89) was ascribed to the ability of the phenyl groups to stabilise different intermediates in the two cases though steric factors may also be important."' Both p-methoxyphenyl and p-cyano- phenyl migrate in preference to phenyl in the photolysis of the 4,4-diaryl- cyclohexenones (90a) and (gob) which suggests that C-C=C-0 is more helpful than the dipolar form C-C=C-O-as a guide in predicting reactivity + in this series.lo4 The cyclohexadienone (91) is converted by light into the bicycle[3,l ,O] hexenone (92) without aryl migration.Irradiation of the bicyclo- hexenone (92) gives a mixture of the isomeric phenols (93a) and (93b) i.e. phenyl migrates in preference to p-cyanophenyl. O5 Preferential migration of phenyl with (92) provides a sharp contrast to the case of cyclohexenone (90a)lo4 above. Different chemical behaviour is of course to be expected of electronically excited species and the corresponding ground-state intermediates and Zimmerman OH Ph Ar 6 (93b) (91) loo W. F. Erman and H. C. Kretschmar J. Amer. Chem. SOC. 1967,89,3842. lo' K. J. Crandall J. P.Arrington and R. J. Watkins Chem. Comm. 1967 1052. lo2 H. E. Zimmerman and J. W. Wilson J. Amer. Chem. SOC.,1964,86,4036. H. E. Zimmerman and D. J. Sam J. Amer. Chem. SOC. 1966,88,4905. lo* H. E. Zimmerman R. D. Rieke and J. R. Scheffer,J. Amer. Chem. SOC. 1967,89,2033. H. E. Zimmerman and J. 0.Grunewald J. Amer. Chem. Soc. 1967,89,5163. Photochemistry has suggested that the excited state of (92) suffers electron demotion to aground- state zwitterion (94) before migration. In the rearrangement of the cyclo- hexenones (go) aryl migration might precede electron demotion and thus involve species of essentially biradical as opposed to ionic character.lo5 The zwitterionic intermediate (95) supposed to be implicated in the photochemical rearrangement of 4,4-diphenylcyclohexadienone to the corresponding bi-cyclo[3,l,0]hexenone has been generated non-photochemically.It rearranged spontaneously to the expected product.lo6 As hydrocarbon analogues of cyclohexenones and cyclohexadienones the exo-methylene compounds (96) and (97) have been investigated photochemically by Zimmerman and his co-workers. Like the ketone (86b) (96) rearranged with phenyl migration giving (98).'07" The triene (97) underwent a similar phenyl shift to give (stereo- selectively) the product (99) and not the product of 'type A' rearrangement which might have been expected by analogy with the behaviour of 4,4-diphenyl-cyclohe~adienone.'~~~ Of course the analogy is rather a superficial one for the ketones react through their n,n* triplets whilst n,n* singlets were shown to be involved in the photochemistry of the hydrocarbons (96) and (97).lo7 The photochemistry of cyclohexadienones containing t-butyl groups displays some new features attributable to steric effects.lo8 The representation of carbonyl n,n* states has been discussed by Taylor who has argued for a modified convention (100) and discussed its applica- bility to cyclohexadienones.logThe mnemonic value of representations such as (100) in stressing correspondences between mass spectra and photochemical processes'l0 is obvious; but it is not clear to the Reporter that they offer any &h R -+-R,c=o .- Ph hh (100) x+Y (98) B- xQY Hz (99) A- MeOH-"Gy' \ / z z (104) (101) (102) (103) lo6 H. E. Zimmerman D.Dopp and P. S. Huyffer J. Amer. Chem. SOC.,1966,%8,5352. lo' (a) H. E. Zimmerman and G. E. Samuelson J. Amer. Chem. SOC. 1967 89 5971; (b) H. E. Zimmerman P. Hackett D. F. Juers and B. Schroder ibid. p. 5973. lo* B. Miller and H. Margulies J. Amer. Chem. SOC. 1967 89 1678; B. Miller ibid. p. 1690; T. Matsuura and K. Ogura ibid. pp. 3846 3850. log G. A. Taylor Chem. Comm. 1967,896. 'lo Inter alia N. J. Turro,D. S. Weiss W. F. Haddon and F. W. McLafferty J. Amer. Chem. SOC. 1967 89 3370; A. L. Burlingame C. Fenselau W. J. Richter W. G. Dauben G. W. Shaffer and N. D. Vietmeyer ibid. p. 3346; C. Djerassi and B. Zeeh Chem. and Ind. 1967,358; M. M. Bursey L. R. Dusold and A. Padwa Tetrahedron Letters 1967,2649. 178 A. C.Day advantage in interpretative or predictive power over other valence-bond representations of photochemical mechanisms (e.g.cf. refs. 102-107). Two photochemical paths are available to conjugated cyclohexadienones (A) reversible ring-opening to a cis-keten which may then react with a protic solvent and (B) rearrangement to a bicyclo[3,1,0]hexenone.11' Collins and Hart studied a series of methyl-substituted cyclohexadienones (101 ; X Y Z = H or Me) and found the course of photolysis to be markedly dependent on the number and position of the methyl groups e.g. (101 ; X = Y = Z = Me) follows path B giving(l04; X = Y = 2 = Me),112 (101; X = Z = H Y = Me) follows path A and (101 ; X = Z = Me; Y = H) reacts by both pathways. It was further shown that the cis-ketens (102) react with methanol to give cis-fly :&-unsaturated esters (103) which may subsequently photoisomerise to the corresponding trans-isomers.There was no evidence for cis-trans isomerisation of the ketens (102) under the reaction conditions."' .".c;. * EtO CN (109) (110) The photodimerisations of cyclopentenone' and cyclohexenone1'4 occur by triplet mechanisms. The relative proportions of head-to-head and head-to- tail dimer in both cases are subject to 'polar solvent effe~ts'.''~~"~ The photocycloaddition of cyclopentenone to cyclohexene may involve a higher triplet state having ET-73 kcal./mole (ETfor the lowest triplet is ca. 61 kcal./mole).' ' The photochemistry of troponoid compounds is discussed in a separate section. Since late 1966 most of the aliphatic photocyc1oadditions3' reported between unlike molecules have been of a preparative nature and embody no new fundamental advance.Oxetan formation occurs in the light-induced reactions of olefins with the P. M. Collins and H. Hart J. Chem. SOC.(C),1967 1197 and refs. therein cited. 'I2 cf. H. Hart and R. K. Murray,jun. J. Org. Chem. 1967,32,2448. 'I3 P. E. Eaton and W. S. Hurt J. Amer. Chem. SOC.,1966,88 5038; J. L. Ruhlen and P. A. Leer-makers ibid. p. 5671. E. Y. Y. Lam D. Valentine and G. S. Hammond J. Amer. Chem. SOC. 1967,89 3482. 'I5 P. de Mayo J.-P. Pete and M. Tchir J. Amer. Chem. SOC.,1967,89 5712. Photochemistry 179 carbonyl groups of benzophenone,' diethyl mesoxalate,' ' ethyl cyano- formate [to give with for example 1,l-diphenylethylene the oxetan (lOS)],' '* acetyl cyanide,' l9 and ap-acetylenic ketones.' 2o The last case contrasts with that of ap-ethylenic ketones which usually react at the carbon-carbon double bond to give cyclobutanes.In all of these cases the orientation of addition to an unsymmetrical olefin is that expected for an initial attack by carbonyl oxygen (n~* triplet) in such a way as to give the more stable biradical. Nucleo- philic attack by carbonyl in the n,~*singlet state seems to be implicated in the oxetan formation between ketones and trans-l,2-dicyanoethyleneor maleic anhydride.' 21 Ketones give spiro-oxetans with allenes ; e.g. acetone gives two 2:1 adducts (106) and (107) with tetramethylallene.'22 Ketenimines give 1-imino-oxetans with aromatic ketones.123 Cyclobutane formation has been reported between various components maleic anhydride with olefins' 24 and dienes,' 25 acrylonitrile with indene,'26 and chromone with 01efins.'~~ Butadiene gives only cyclobutanes (108) with or-acetoxyacrylonitrile on unsensitised irradiation but additionally the 1,4-adduct (109) on photosensitisation.' 28 Light-induced cycloadditions to 3-acetoxycyclopentenones provided a convenient route to seven-membered rings in the synthesis of stipitatonic acid' 29a and ( f)-p-himachalene.' 29b Cycloadditions and cyclodimerisation of vinylene carbonate (1 10)offer routes to lY2-dihydroxy- and 1,2,3,4-tetrahydroxycyclobutanes,respectively.'30 The photochemistry of 0x0-sulphides,' 31 silyl ketones,'32 and imonium salts133 has received attention.Irradiation of degassed solutions of a variety of a-substituted cinnamic and crotonic acids gives the isomeric p-lactones ; analogous amides similarly give p-lactams. 34 Aromatic Compounds.-Last year Bryce-Smith and Longuet-Higgins' M. Ogata and H. Kanb Chem. and Znd. 1967,321. M. Hara Y. Odaira and S. Tsutsumi Tetrahedron Letters 1967,2981. Y. Odaira T. Shimodaira and S. Tsutsumi Chem. Comm. 1967 757. Y. Shigemitsu,Y.Odaira and S. Tsutsumi Tetrahedron Letters 1967 55. M. J. Jorgenson Tetrahedron Letters 1966 5811. N. J. Turro P. Wriede J. C. Dalton D. Arnold and A. Glick J.Amer. Chem. Soc. 1967,89,3950. H. Gotthardt R. Steinmetz and G. S. Hammond Chem. Comm. 1967,480; D. R. Arnold and A. H. Glick ibid.1966 813; H. Hogeveen and P. J. Smit Rec. Trao. chim. 1966,85 1188. IZ3 L. A. Singer and G. A. Davis J. Amer. Chem. SOC. 1967,89,941. R. L. Cargill and M. R. Willcott J. Org. Chem. 1966 31 3938; W. Metzner H. Partale and C. H. Krauch Chem. Ber. 1967,100,3156. H.-D. Scharf Tetrahedron Letters 1967 4231; H.-D. Scharf and F. Korte Chem. Ber. 1966 99 1299. 126 J. J. McCullough and C. W. Huang Chem. Comm. 1967 815. J. W. Hanifin and E. Cohen Tetrahedron Letters 1966 5421. W. L. Dilling and J. C. Little J. Amer. Chem. SOC.,1967,89,2741; W. L. Dilling ibid. p. 2742. (a)G. L. Lange and P. de Mayo Chem. Comm. 1967,704; (b)B. D. Challand G. Kornis G. L. Lange and P.de Mayo ibid. p. 704. 130 W. Hartmann and R. Steinmetz Chem. Ber. 1967,100,217. 131 P. Y.Johnson and G. A. Berchtold J. Amer. Chem. SOC. 1967,89,2761. 13' A. G. Brook and J. M. Duff,J.Amer. Chem. Soc. 1967,89,454; H. G. Kuivila and P. L. Maxfieid J. Organometallic Chem. 1967 10 41. 133 G. Adam C. Horstmann and K. Schrieber Chem. Ber. 1967,100,1753. 134 0.L. Chapman and W. R. Adams J. Amer. Chem. SOC. 1967,89,4243. 135 D. Bryce-Smith and H. C. Longuet-Higgins Chem. Comm. 1966,593. 180 A. C.Day 2531A Benzene L . (113) 1 (112) SCHEME rationalised the known photochemistry of benzene as illustrated in Scheme 1. Initial excitation gives the lowest excited singlet state (’ &) which may undergo intersystem crossing to the triplet state (3B1u). Orbital correlation diagrams showed that the excited singlet may pass adiabatically into the singlet form of (11l) the triplet into the triplet states of (1 12) and (1 13).The intermediate (1 11) readily accounts for the formation of fulvene benzvalene (114) and 1,3-adducts with olefins e.g. (1 15) from a 1,2-disubstituted olefin. Prismane Dewar benzene and 1,2-adducts with olefins are derivable from (1 12) and 1,4-adducts from (113). The transposition of ring atoms which can be observed with substituted benzenes could occur by 1,2-shifts in the reversible conversion of benzene into benzvalene and by both 1,2-and 1,3-hifts through prismane. Leading references to the experimental basis for these conclusions are readily a~ailable,’~~ and only the more recent work is discussed here. Photolysis of benzene vapour at 1470 A causes extensive fragmentati~n,’~’ but at longer wavelengths fulvene is formed.’38 Benzvalene (1 14) was not detected in vapour phase photolyses but irradiation of liquid benl e at 2537 8 gives both benzvalene and f~1vene.l~’ Photolysis of o-xylene vapoufi,in Zen ’R (114) (1 15) (1 16) (117) (118) (119) (120) 136 Ref.135 and I. 0.Sutherland,Ann. Reports 1966,63 398; cf. also ref. 2 (1966 p. 379). 13’ W. M.Jackson J. L. Faris and B. Donn J. Phys. Chem. 1967,71,3346. 13’ H.R. Ward J. S. Wishnok and P. D. Sherman jun. J. Amer. Chem. SOC. 1967 89,162; L. Kaplan and K. E. Wilzbach ibid. p. 1030. 13’ K. E. Wilzbach J. S. Ritscher and L. Kaplan J. Amer. Chem. SOC. 1967,89 1031. Photochemistry X (=9 a-X = OAc b:X =OH (124) (125) the vacuum ultraviolet or at 2537 A gives chiefly m-xylene possibly via di-methylbenzvalene along with smaller quantities of other hydrocarbons includ- ing p-xylene and benzocy~lobutene.'~~ A detailed study of the light-induced cycloaddition of maleic anhydride to benzene has a~peared,'~' and of the maleic anhydride-hexamethylbenzene system (which does not undergo 1,2-photocycloaddition).142 Photo-oxidation of liquid benzene gives trans trans-hexa-2,4-dienedial and a C1,-dialdehyde OCH -[CH=CH] CHO.Possible intermediates are the 1,2-adducts with oxygen (1 16) and (1 17).'43 Analogously Perrins and Simons have detected acyclic tetraenes spectroscopic- ally when benzene is irradiated in the presence of chloro-olefins or benzonitrile in the presence of trimeth~lethylene.'~~ The photochemical reaction between benzene and butadiene gives a complex mixture of products including the highly reactive trans-olefin (1 18) which may dimerise or react with a second molecule of diene to give the 2 1 adduct (119) in non-photochemical steps.In the presence of nitric oxide (1 18) isomerises to the more stable cis-isomer. 14' Formation of (1 18) is readily rationalised as involving the reaction of transoid butadiene with the biradical (1 13). Biradical(ll3) may also be involved in the photochemical 1,6addition of amines to benzene.'46 Pyrrole gives the 1 -cyclo- hexadienylpyrrole (1 2O),' 460 whilst simple primary and secondary amines react at nitrogen to give 3-alkylaminocyclohexa-1,4-dienes.'46b Since N-methylpyrrole failed to react it was proposed that compound (120) is formed by intramolecular hydrogen abstraction in the intermediate (121).The possibility that these reactions involve attack on ground-state benzene by R ficould be excluded since no traces of the corresponding aminobenzenes were found 146 140 H. R. Ward J. Amer. Chem. Soc. 1967,89,2367. 141 W. M. Hardham and G. S. Hammond J. Amer. Chem. Soc. 1967,89,3200; cf. D. Bryce-Smith B. Vickery and G. I. Fray J. Chem. Soc. (C),1967 390. 142 Z. Raciczewski J. Chem. Soc. (B),1966,1142 1147. 143 Kei Wei J.-C. Mani and J. N. Pitts jun. J. Amer. Chem. Soc. 1967,89 4225. 144 N. C. Perrins and J. P. Simons Chem. Comm. 1967,999. 14' K. Kraft and G. Koltzenburg Tetrahedron Letters 1967,4357; cf.ibid. p. 4723. M. Bellas D. Bryce-Smith and A. Gilbert Chem. Comm. 1967 (a)p. 263; (b)p. 862. 182 A. C. Day Benzene undergoes photochemical reactions with acetic acid and aqueous phosphoric acid to give the exo-adducts (122a) and (122b) re~pectively.'~'" To explain inter alia,147b the stereospecificity of the reaction the carbonium ion (123) has been proposed as an intermediate.148 This could be formed either by protonation of the species (1 11) discussed above or by direct photogxcitation of a-protonated benzene. The cation (123) should be attacked by nucleophiles preferentially in the exo sense and thus accounts for the stereochemistry of the products.'48 It is not yet clear whether any relationship exists between these reactions and the photosensitised protonation of 01efins~~-~~ noted earlier.Other discussions relevant to the photochemistry of benzene concern the geometry of the triplet and a hypothetical intermediate having the topology of a Mobius strip."' Hexamethylbicyclo[2,2,0]hexa-2,5-diene (hexamethyl Dewar benzene) is converted by light into hexamethylprismane (124)'' and small amounts of the bicyclic isomer (125). ''lb Further examples of light-induced valence isomerisation will be found in the sections on heterocyclic and troponoid compounds. Dianthracene the 9,lO-photodimer of anthracene dissociates when irradiated in the solid state to give a new sandwich crystal form of anthracene which has a yellow-green excimer fluorescence.' 52 Cross-adducts analogous to dianthra- cene have been prepared by irradiation of mixtures of substituted anthra- cenes.lS3Also analogous to dianthracene is the bridged system formed by reversible p hotoisomerisation of 1,2-di-(9'-an t hry1)ethane.'54 [2,2JParacyclo-naphthane (126) is converted by light into the heptacyclic hydrocarbon (127) ('dibenzoequinene'). lS5Transannular bonding of this kind was not observed in the photolysis of [2,2]paracyclophane; cleavage of one of the saturated bridging groups occurred instead. lS6 The photolysis of sodium cyclopentadienyl in tetrahydrofuran-t-butyl alcohol gives in low yield a 1 :1 mixture of the meso-and (+)-isomers (128). Deuterium-labelling implied that the excited anion abstracts a hydroxy hydrogen atom from the solvent. Subsequent protonation gives the cyclopent- 2-enyl radical which dimerises.'' 14' (a) E. Fahrenhorst and A. F. Bickel Tetrahedron Letters 1966 5911; (b) L. Kaplan J. S. Ritscher and K. E. Wilzbach J. Amer. Chem. SOC. 1966,88,2881. 148 D. Bryce-Smith A. Gilbert and H. C. Longuet-Higgins Chem. Comm. 1967,240. 14' G. C. Nieman and D. S. Tinti J. Chem. Phys. 1967,46,1432. 150 E. Fahrenhorst Tetrahedron Letters 1966,6465; 6.footnote in ref. 148. 15' (a)D. M. Lemal and J. P. Lokensgard J. Amer. Chem. SOC. 1966 88 5934; W. Schiifer R. Criegee R. Askani and H. Griiner Angew. Chem. Internat. Edn. 1967 6 78; W. Schafer and H. Hellmann ibid. p. 518; (b)H. Hogeveen and H. C. Volger Chem. Comm. 1967 1133. 152 E. A. Chandross and J. Ferguson,J. Chem. Phys. 1966,45 3564. 153 H.Bouas-Laurent and R. Lapouyade Compt. rend. 1967 264 C 1061. 154 R. Livingston and K. S. Wei J. Amer. Chem. SOC.,1967,89 3098. 155 H. H. Wasserman and P. M. Keehn J. Amer. Chem. SOC.,1967,89,2770; cf. ibid. 1966,88,4522. R. C. Helgeson and D. J. Cram J. Amer. Chem. SOC. 1966,88 509. E. E. van Tamelen J. I. Brauman and L. E. Ellis J. Amer. Chem. SOC.,1967,89 5073. Photochemistry The photochemical reduction of a phenol by sodium borohydride has been reported. Sodium bisulphite and dithionite were also effective. '58 Photosubstitution. A recent lecture described contributions made by Havinga's group to aromatic photochemical substitution. '59 Aromatic nitro-compounds undergo substitution at the ortho-and para- positions when irradiated in liquid ammonia with unfiltered light.Nitrobenzene yields p-nitroaniline and a smaller amount of the ortho-isomer. p-Chloro- nitrobenzene gives 5-chloro-2-nitroaniline and by displacement of chlorine p-nitroaniline. rn-Chloronitrobenzene is attacked at the 4-and 6-positions.' 6o Preferential substitution at the positions ortho and para to the nitro-group is consistent with earlier observations on the light-induced reaction of nitro- anisoles with amines.16 'In contrast displacement of the methoxy-group meta to the nitro-group occurred in the reactions between methylamine and 4-nitro- veratro1e,162a and between hydroxide ion and m-nitroanisole.' 62b Attack meta to the nitro-group has also been observed recently in the photochemical reactions of cyanide ion with p-and m-nitroanisoles in the latter case with displacement of meth0~y.I~~ Nucleophilic attack at the meta-position was discussed by Letsinger and McCain in terms of a bicyclic intermediate e.g.(129) for the reaction of m-nitroanisole with cyanide.'63 A formal analogy is discernible between protonated prefulvene (123) and representations such as (130) [which could give (129) with cyanide ion] and it therefore seems worth considering whether any useful correlation can be drawn between these 15* J. A. Waters and B. Witkop J. Amer. Chem. SOC. 1967,89 1022; cf. T. Tokuyama S. Senoh T. Sakan K. S. Brown,jun. and B. Witkop ibid. p. 1017. lS9 E. Havinga R. 0. de Jongh and M. E. Kronenberg Helo. Chim. Acta 1967 50 2550; E. Havinga 2nd IUPAC International Symposium on Photochemistry Enschede Holland 16-22 July 1967 Pure Appl.Chem. to be published. A. van Vliet M. E. Kronenberg and E. Havinga Tetrahedron Letters 1966,5957. M. E. Kronenberg A. van der Heyden and E. Havinga Rec. Trao. chim. 1966,85,56. (a) M. E. Kronenberg A. van der Heyden and E. Havinga Rec. Trao. chim. 1967 86 254; (b)R.0.de Jongh and E. Havinga ibid. 1966,85,275. 163 R. L. Letsinger and J. H. McCain J. Amer. Chem. Soc. 1966,88,2884. 184 A. C.Day NO reactions and the photoinduced addition of acetic acid etc. to benzene. However a satisfactory rationale of substituent effects will not be easy to reach without more experimental data in view of complexities such as the dependence of the reaction path on the nucleophile,'6'* 162a the formation of complexes between benzene and arnine~,'~~ and the light-induced 1,4-addition of amines to benzene noted above.In a study of photoinduced hydrolysis striking differences have been observed between substituted pyridines and the corresponding nitrobenzenes. 3-Bromo- pyridine undergoes photohydrolysis in dilute base easily whereas m-bromo- nitrobenzene is unreactive. Conversely 3-methoxypyridine hydrolyses photo- chemically much less readily than m-nitroanisole. ''' Havinga has stressed that care is necessary in interpreting differences of this kind.'59* '" The reason is obvious but merits repetition. The quantum yield of a photochemical reaction with respect to a given product is a measure of the efficiency of that reaction compared with all other pathways available to the excited species.Therefore comparisons based on the quantum yields for the formation of analogous products from different reactants are meaningless in absence of a detailed knowledge of the reaction mechanism and the lifetimes of the relevant excited states. Monosubstituted benzenes such as fluorobenzene chlorobenzene and anisole undergo photonucleophilic displacement of the substituent in the presence of alkoxide ion cyanide ion and piperidine. '"Displacement of the nitro-group occurs in the light-induced reaction between piperidine and 4-nitropyridine-1 -oxide.' 67 Replacement of the trimethylamino-group by hydrogen when substituted trimethylanilinium salts are irradiated in methanol involves free radicals.The effect of different anions suggests that the initial step is usually charge-transfer excitation of the associated ions. '68 Recent studies of aromatic photosubstitution also include the fol!owing photochlorination of nitro-compounds with hydrochloric acid,' 69 alkylation 164 H. Kehiaian Abstracts of Chemical Society Anniversary Meeting Exeter 3-6 April 1967 C 12 quoted in ref. 146b. 165 G. H.D. van der Stegen E. J. Poziomek M. E. Kronenberg and E. Havinga Tetrahedron Letters 1966,6371. 166 J. A. Barltrop N. J. Bunce and A. Thomson J. Chem. SOC.(C) 1967,1142. "'R. M. Johnson and C. W. Rees J. Chem. SOC.(B),1967 15. 16* T.D. Walsh and R. C. Long J. Amer. Chem. SOC.,1967,89,3943. 16' R.L. Letsinger and G. G. Wubbels J. Amer. Chem. SOC.1966,88 5041. Photochemistry of polycyclic aromatic compounds by lithium alkyls,' 70 photochemical de- alkylati~n,'~oxidative coupling of phenols,' 72 reactions of diary1 ketones ' with phenols,173 and photoarylation.'74 The 9-anthroate ion in aqueous solution is converted by light into the anion of 9-hydroxyanthracene and carbon monoxide. A mechanism analogous to the photochemical rearrangement of vinyl and aryl nitro-compounds to nitrites is proposed. 17' Photo-Fries rearrangement. Two mechanisms have been proposed for the photochemical rearrangement of aryl esters to 0-and p-hydroxyaryl ketones (i) a cyclic mechanism and (ii) homolysis of the acyl-0 bond to give two radicals which may recombine within the solvent cage to ketone e.g.(133) from a p-substituted ester (131) (path a in Scheme 2) or diffuse apart and give O*COR1 0 [0 -R'cO'] a OCOR' hv R2 R2 R2 (132) 033) 1 (fll/ h from cage diffusion 0 6-0 R2 R1 RZ R2 (135) SCHEME 2 phenols e.g. (135) as cleavage products (path l~).~' For p-tolyl acetate (131; R' = R2 = Me) Sandner and Trecker have now shown that the quantum yield of p-cresol (135; R2 = Me) is reduced by an increase in the viscosity of the medium whilst formation of (133; R' = R2 = Me) is insensitive to vis- cosity changes. An analogous effect was caused by oxygen. '76 This behaviour is consistent with mechanism (ii) provided that the radicals ArO and RCO recombine to give (133) only within the solvent cage (132). (The insensitivity to external effects commonly observed of caged radical pairs is well known).Sandner and Tre~ker'~~ give a somewhat less economical interpretation of their results. 17* H. J. S. Winkler R. Bollinger and H. Winkler J. Urg. Chem. 1967,32 1700. T. Matsuura and Y. Kitaura Tetrahedron Letters 1967 3311. J. M. Bobbitt J. T. Stock k Marchand and K. H. Weisgraber Chem. and Ind. 1966,2127. 173 H.-D. Becker,J. Org. Chem. 1967,32,2115,2124. P. W. Jeffs and J. F. Hansen J. Amer. Chem. Soc. 1967,89 2798; S. M. Kupchan and R. M. Kanojia Tetrahedron Letters 1966 5353. 17' A. W. Bradshaw and 0.L. Chapman J. Amer. Chem. SOC.,1967,89 2372. 17' M. R. Sandner and D. J. Trecker J. Amer. Chem. SOC. 1967,89 5725. 186 A. C.Day Side reactions which may accompany the photo-Fries rearrangement are decarboxylation (ArO*COR -+ArR) and decarbonylation (ArO- COR -+ ArOR).177a Photolysis of the optically active ester (136) gives optically inactive ether (137) and optically active hydrocarbon (138).' 77bFinnegan and Knutson suggested that the ether (137) is formed by decarbonylation of RCO within the solvent cage to give a new caged radical pair [cf.(134)] in which inversion of R is faster than radical recombination. The retention of optical activity in (138) was taken to imply a concerted mechanism for decarb~xylation.'~~ Steric and electronic effects in the photo-Fries rearrangement have been studied. '78a No isotope effect was observed in the rearrangement of p-methoxy- phenyl [1-'4C]acetate.1 78b /Me Me 0-CO-CH 'Et (139) (140) (141) The rearrangement of aryl cinnamates has been de~cribed.'~' Aryloxy- acetones180a and alkyl aryl ethers180b undergo reactions analogous to the photo-Fries rearrangement.Other reactions involving aromatic side-chains which have been reported include the photolysis of 2,4-dinitrophenylamino-a~ids'~ la and aryl phos- phates,l8lb and the use of photosensitive protecting groups.181c Stilbenes. Triplet stilbene has at last been detected. It was produced by flash photolysis of trans-stilbene in EPA glass at 77"~ (lifetime 2.2 x sec.).'82 The curious multiple maxima reported in the Saltiel plot for photosensitised 177 R. A. Finnegan and D. Knutson (a) Chem. Comm. 1966 172; (b)J Amer. Chem. SOC. 1967 89 1970.17* (a) G. M. Coppinger and E. R. Bell J. Phys. Chem. 1966 70 3479; (b) L. Schutte and E. Havinga Tetrahedron 1967,23,2281. 179 H. Obara and H. Takahashi Bull. Chem. SOC.Japan 1967,40 1012. 180 (a)J. Hill Chem. Comm. 1966 260; (b) D. P. Kelly J. T. Pinhey and R. D. G. Rigby Tetra-hedron Letters 1966 5953; cf. also enamides N. C. Yang and G. R. Lenz ibid. 1967,4897. (a)D. J. Neadle and R. J. Pollitt J. Chem. SOC.(C),1967,1764;(b)A. J. Kirby and A. G. Varvoglis Chem. Comm. 1967,405; (c)J. A. Barltrop P. J. Plant and P. Schofield ibid. 1966 822. 183 G. Heinrich H. Blume and D. Schulte-Frohlinde Tetrahedron Letters 1967,4693. Photochemistry 187 cis-trans isomerisation of stilbene' 3a do not have the significance originally attributed to them. The latest version is a more or less typical Saltiel plot having a single maximum in the region of ET= 50 k~al./mole.'~~~ Evidence for the intervention of triplet states in the unsensitised photoisomerisation of stilbene has been advanced.184 Pyrylium salts have been used as photosensitisers for the isomerisation of ~tilbene,'~ and SCF-MO calculations on stilbene have been reported.' The reversible photocyclisation of cis-stilbene to 4aY4b-dihydrophenan- threne exemplifies a general class of photochemical reactions3k applicable to a wide variety of compounds of the type Ar*X=Y*Ar'. Additionally it has interest as one of the simplest photochromic system^.^' Muszkat and Fi~cher'~~ have studied in great' detail the reversible photoisomerism of 1,2-diphenylcyclopentene (1 39) to the dihydrophenanthrene (140) and the oxidation of the latter by molecular oxygen to the phenanthrene (141).The photoisomerisation involves singlet states and on the basis of orbital symmetry arguments56 the anti-configuration as shown in (140) has been predicted for the dihydr~phenanthrene'~' (conrotatory process).* Synthetically stilbene photocyclisations are normally carried out under conditions which permit oxidation of the dihydro-compound to the phenan- threne. Iodine has often been used as oxidant ;but cupric halides in the presence of air are particularly effective for the cyclisation of m-disubstituted stil-benes,'88 and selenium radicals for the synthesis of sterically hindered phen- an throne^.'^^ In the presence of diphenyl diselenide 1,2-di-(a-naphthyl)- ethylene is converted by light quantatitively into picene.189 Examples reported recently are of considerable diversity and illustrate the great versatility of the reaction for the synthesis of polycyclic aromatic and heterocyclic compounds. P-Styrylnaphthalene cyclises solely at the a-position to give the benzophenanthrene (142) and a variety of P-naphthyl analogues of stilbene behave ~irnilarly.'~~ 9-Styrylphenanthrenes (143 ; R = H or Me) cyclise at the 10-position to give benzo[g]chrysenes (144).190a o-Terphenyl gives triphenylene in 88 % yield when irradiated in the presence of iodine.Ig1 The metacyclophane (145) undergoes reversible photoisomerisation to the coloured hexaene (146) which is rapidly converted by oxygen into the di- ls3 (a) G.S. Hammond J. Saltiel A. A. Lamola N. J. Turro J. S. Bradshaw D. 0. Cowan R. C. Counsell V. Vogt and C. Dalton J. Amer. Chem. Soc. 1964,86 3197; (b)W. G. Herkstroeter and G. S. Hammond ibid. 1%6,88,4769. lS4 K. A. Muszkat D. Gegiou and E. Fischer J. Amer. Chem. Soc. 1967,89,4814. R. Searle J. L. R. Williams D. E. DeMeyer and J. C. Doty Chem. Comm. 1967,1165. lS6 P. Borrell and H. H. Greenwood Proc. Roy. Soc. 1967 A 298 453. K. A. Muszkat and E. Fischer J. Chem. Soc. (B) 1967,662; cf. K. A. Muszkat D. Gegiou and E. Fischer Chem. Comm. 1965,447. 188 D. J. Collins and J. J. Hobbs Austral. J. Chem. 1967,20 1905. E. J. Levi and M. Orchin J. Org. Chem. 1966,31,4302. 190 (a) W. Carruthers J. Chem. Soc. (C) 1967 1525; (b)M.Scholz M. Miihlstadt and F. Dietz TetrahedronLetters 1967,665. 19' T. Sato. Y. Goto and K. Hata Bull. Chem. Soc. Japan 1967,40,1994. G 188 A. C.Day \/ (142) (143) (144) hv -@ -hv' / (145) (146) (147) Ph Ph Ph HO \/ dR2 (148) hydropyrene (147).192 Related systems which undergo light-induced cyclisa- tion are tetraphenylcy~lopentadienone,'~~ tetraphenylcyclopentenone,'94 and the hydroxylactone (148)8f cis-3-benzoyl-2,3-diphenylacrylic acid. lg2 H. Blaschke and V. Boekelheide J. Arner. Chern. SOC. 1967 89 2747; cf. H.-R. Blattmann D. Meuche E. Heilbronner R. J. Molyneux and V. Boekelheide ibid. 1965,87 130. 19' N. Toshima and I. Moritanj Bull. Chem.SOC.Japan 1967,40,1495; Tetrahedron Letters 1967 357; I.Moritani and N. Toshima ibid. p. 467. 194 I. Moritani N. Toshima S. Nakagawa and M. Takushiji Bull. Chem SOC.Japan 1967,40,2129. 195 G. Rio and J. C. Hardy Bull. SOC.chirn. France 1967,2642. Photochemistry 189 Applications in the heterocyclic field have included the oxidative photo- cyclisation of 4-styrylpyrimidine to benzo[f]quinazoline,' 96 and 4-styryl- pyridine (4-stilbazole) to benz[h]isoquinoline and the related reaction of 4-styrylquinoline. lg7 The last two reactions which were conducted in cyclo- hexane also give cyclohexyl-substituted products e.g. (149a) and (149b) from 4-styrylpyridine.' 97 Quantum yields have been measured for the photo- isomerisation of isomeric 1,2-dipyridylethylenes.'g* Examples containing furan and thiophen rings have also been reported.' 99 For example compound (150) cyclises to give (151) in good yield.1,2-Di-(2'-furyl)ethylene gives a cyclisation product analogous to (1 5 1) in very poor yield 199a unless cupric chloride is employed as oxidant.'99b Photocyclisation is also observed when the -CH=CH group of stilbene is replaced by the CH-N -or N=N groups. Benzylideneaniline is converted into 9-a~aphenanthrene;~OO and azobenzenes are converted into benzo[c]cin- nolines. Photocyclisations of azobenzenes require strongly acidic conditions and part of the azo-compound is reduced during the reaction to give benzidines as by-products. Reactions which bear some formal relationship to these cyclisations are the photochemical conversion of diphenylamine and diphenyl ether into carbazole and dibenzofuran respectively,202 and the photocyclisation of ben~anilides~~~ and anilides of substituted acrylic In the last case a deuterium- labelling study established that 1,3-shifts of hydrogen occur in the conversion of the presumed intermediate or excited state (1 52) to 3,4-dihydrocarbo~tyrils.~~~ 2,2'-Di(phenylethynyI)biphenyl is converted by light or thermally into the (152) 196 C.E. Loader and C. J. Timmons J. Chem. SOC.(C),1967,1343. 197 C. E. Loader and C. J. Timmons J. Chem. SOC.(C) 1967,1457. H.-H. Perkampus G. Kassebeer and P. Miiller Ber. Bunsengesellschafi Phys. Chem. 1967 71,40. 199 (a) C. E. Loader and C. J. Timmons J. Citem. SOC.(C),1967 1677; (b) R. M. Kellogg M.B. Groen and H. Wynberg J. Org. Chem. 1967,32,3093. 2oo C. E. Loader and C. J. Timmons J. Chem. SOC.(C) 1966 1078; cf. V. M. Clark and A. Cox Tetrahedron 1966,22,3421. '01 N. C. Jamieson and G. E. Lewis Austral. J. Chem.,1967,20,321; C. P. Joshua and G. E. Lewis ibid. p. 929; G. E. Lewis and J. A. Reiss ibid. p. 1451 and refs. therein cited. '02 H. Stegemeyer,Naturwiss. 1966,53 582; CJ ref. 3k. 203 B. S. Thyagarajan N. Kharasch H. B. Lewis and (in part) W. Wolf Cheni. Comm. 1967,614. 204 P. G. Cleveland and 0.L. Chapman Chem. Comm. 1967 1064. 190 A. C. Day polycyclic hydrocarbon (1S3).205 Examples where photocyclisation according to the stilbene pattern might have been expected but where alternative reac- tions occurred have been reported in connection with some heterocyclic syntheses.2o Quinone~~j-The structure of the photodimer of p-benzoquinone has been reinve~tigated.~” The photodimerisation and photoreduction of 1,2-naphtha-quinone in various media have been studied.208 The light-induced reaction between p-quinones and aldehydes to give acylquinols probably involves attack by acyl radicals on the q~inone.~” Several photoadducts between p-dioxen and 1,Znaphthaquinones have been described. They are believed to have the cis-configuration (1S4).21 Full details have appeared of Bryce-Smith’s work on the photocycloaddition of olefins to p-benzoquinone. Addition occurs at the carbonyl group to give spiro-oxetans e.g. cyclo-octene gives the adduct (155) when irradiated in benzene in the presence of p-benzoquinone.The adducts readily undergo dienone-phenol rearrangement to hydroxy-coumarans when treated with acid.2l1 The light- induced addition of diphenylacetylene to p-benzoquinone gives the adduct (156) presumably uia the intermediate spiro-oxeten (157).21 The photo- addition of diphenylacetylene and other alkynes to methoxy-p-benzoquinone provides an instructive contrast illustrating the marked influence of substi- tuents in the photochemistry of quinones. The products were cyclobutenes (158a<) formed by attack upon a carbon-carbon double The photocycloaddition of conjugated dienes to quinones has been studied by Barltrop and He~p.~~~ Chloranil l,Cnaphthaquinone and 2,S-dimethyl-p- benzoquinone reacted with simple dienes to give adducts containing a cyclo- butane ring e.g.(159; R = H or Me) from the reaction of chloranil with 205 E. H. White and A. A. F. Sieber Tetrahedron Letters 1967,2713. 206 N. C. Yang A. Shani and G. R. Lenz J. Amer. Chem. Soc. 1966,88 5369; G. R. Lenz and N. C. Yang Chem. Comm. 1967,1136; K. H. Grellmann and E. Tauer Tetrahedron Letters 1967,1909. ’*’ E. H. Gold and D. Ginsburg J. Chem. SOC.(C),1967,15. 208 J. Rennert S. Japar and M. Guttman,Photochem. and Photobiol. 1967,6,485. ’09 J. M. Bruce D. Creed and J. N. Ellis J. Chem. SOC.(C) 1967,1486. 210 W. M. Horspool and G. D. Khandelwal Chem. Comm. 1967,1203. ‘I1 D. Bryce-Smith A. Gilbert and M. G. Johnson J. Chem. SOC.(C) 1967 383. ’12 H. E. Zimmerman and L. Craft Tetrahedron Letters 1964,2131 ;D.Bryce-Smith G. I. Fray and A. Gilbert ibid. p. 2137. 213 S. P. Pappas and B. C. Pappas Tetrahedron Letters 1967 1597. ’14 J. A. Barltrop and B. Hesp J. Chem. SOC. (C) 1967 1625. 191 Photochemistry (157) (158) (159) a R' = RZ = Ph b:R' = R2 = Me c R' = Ph R2 = H or R' = H R2 = Ph I butadiene and 2,3-dimethylbutadiene7 respectively [cf. also (1 60)]. A number of other products were obtained in most cases notably the 2 1 adduct (161) and the spirodihydropyran (1 62) from 1,4-naphthaquinone and 2,3-dimet hyl-butadiene. The structures of the products with the exception of (162) can all be rationalised in terms of intermediate biradicals such as (163). The spiro- compound (162) probably arises from attack of carbonyl oxygen on the diene.21 These and earlier examples cited by Bruce3' and Barltrop and He~p,~'~ illustrate two competing tendencies in quinone photochemistry attack on carbonyl oxygen versus attack at a ring carbon atom.The opposing tendencies as typified by the difference in behaviour between chloranil and p-benzoquinone have received some rationalisation from molecular orbital calculations.214 Heterocyclic Compounds.-The phototransposition of ring atoms in benzenes (see section on aromatic compounds) has an analogy in the photo- isomerisation of pyrazine to pyrimidine.2 ' Methyl derivatives of pyrazine behave similarly when irradiated at 2537 A in the vapour For (a)F. Lahmani N. Ivanoff and M. Magat Compt. rend. 1966,263 C 1005;(b)F.Lahmani and N. Ivanoff Tetrahedron Letters 1967,3913.192 A. C. Day example 2,5-dimethylpyrazine gives 4,6- and 2,5-dimethylpyrimidine and 2-methylpyrazine gives 4- and 5-(and possibly also 2-) methylpyrimidine. Singlet n,n* states seem to be involved. The observed products are all derivable from the pyrazines by 1,2-shiftsY and Lahmani and Ivanoff have therefore suggested that phototransposition occurs through a benzvalene (114) analogue. The light-induced rearrangement of substituted thiophens involves similar transpositions within the thiophen ring as demonstrated in 14C labelling experiments. A series of paperszi6 describes the thiophen rearrangements in detail and the following generalisations emerge (a) 2-arylthiophens re-arrange irreversibly to 3-arylthiopnens ; (b) the major process occurring in 2-phenylthiophens containing phenyl methyl or deuterium substituents is interchange of C-2 and C-3 without concomitant interchange of C-4 and C-5; and (c) in 3-phenylthiophens containing phenyl methyl or deuterium sub- stituents C-2 and C-4 are interchangeable and also C-2 may rearrange to become C-5 and C-5 may rearrange to become C-4.In the case of 4-substituted 3-phenylthiophens a second possibility is rearrangement to a 3-substituted 2-phenylthiophen. Wynberg and his collaborators have rationalised these observations in terms of structures of the valene type e.g. (164) and related canonicals involving the d-orbitals of the sulphur for the C-2-C-3 trans-position in 2-phenylthiophen. Structures such as (1 64) may of course represent some (electronically or vibrationally) excited species.Alternative structures such as (165) are less attractive for various even though two analogies are known. Thus the azirine aldehyde (1 66) has been characterised as an intermediate and actually isolated in the photochemical rearrangement of 3,5-diphenylisoxazole to 2,5-diphenyloxa~ole.~'~ Also cyclopropene-2-carbaldehyde seems to be implicated in the vapour-phase mercury-photo- sensitised decarbonylation of furan to propyne and cyclopropene since one of the minor products of this reaction is the aldehyde (167) which could well have been formed by Diels-Alder addition of the cyclopropene aldehyde to furan.218 Furan decomposition is however a rather complex process since in addition vinylketen can be detected by trapping experiments with methanol." Benzophenone undergoes photocycloaddition to furan to give a 1 :1 adduct and two 2:l adducts the stereochemistry of which has been determined.219 Pyridine is reversibly photohydrolysed in aqueous base to 5-aminopenta-2,4-dienal.220A convenient new synthesis of cyclobutadieneirontricarbonyl(168) from a-pyrone has been described.22' Photolysis of this complex gave free 'I6 H.Wynberg R. M. Kellogg H. van Driel and G. E. Beekhuis J. Amer. Chem. SOC. 1967,89 3501 and preceding papers. 217 E. F. Ullman and B. Singh J. Amer. Chem. SOC.,1966,88 1844. R Srinivasan J. Amer. Chem. SOC.,1967,89 1758,4812. 219 M. Ogata H. Watanabe and H. Kanb Tetrahedron Letters 1967,533; J. Leitich ibid. p.1937; G. Evanega and E. B. Whipple ibid. p. 2163; S. Toki and H. Sakurai ibid. p 4119. 220 J. Joussot-Dubien and J. Houdard Tetrahedron Letters 1967,4389. 221 M. Rosenblum and C. Gatsonis J. Amer. Chem. SOC. 1967,89,5074. Photochemistry Ph 12 5aph -4 As+ v-(164) Ph 3q S' H COzMe I 9 0 1 C0,Me Fe (CO) (170) (168) (1 69) cyclobutadiene.222 The photochemical alkylation of nitrogen-heterocycles by acetic and acidic methanol and has received attention. The electronic states of azoalkanes have been discussed theoretically.224 The photolysis of 2,3-diazabicyclo[2,2,l]hept-2-ene in the vapour phase gives vibrationally excited bicyclo[2,1,0]pentane as initial Cycloaddition of diazomethane to cyclobutenes containing a conjugated carbonyl or cyano- group followed by photolysis of the adduct provides a convenient route to bicycl0[2,1,0]pentanes.~~~Starting with dimethyl acetylenedicarboxylate and 2-diazopropane a similar sequence gave the bicyclobutane (169).227 The photolysis of several pyrazolenines has been reported.228 The photolysis of a 2,3-dihydropyrazine with ring-contra~tion,~~~ and of a 3-benzoylazetidine with ring-expan~ion~~' have been described.The photosensitised extrusion of sulphur dioxide from the cis-and trans-dimethylsulpholenes (1 70) gives hexa-2,4-dienes by a conrotatory process as 222 W. J. Tyerman M. Kato P. Kebarle S. Masamune 0.P. Strausz and H. E. Gunning Chem. Comm. 1967,497. 223 (a) H. Nozaki M. KatB R. Noyori and M. Kawanishi Tetrahedron Letters 1967 4259; (b)M.Ochiai and K. Morita ibid. p. 2349. 224 M. B. Robin R. R. Hart and N. A. Kuebler J. Amer. Chem. SOC.,1967,89 1564. 225 T. F. Thomas C. I. Sutin and C. Steel J. Amer. Chem. SOC. 1967,89 5107. 226 T. H. Kinstle R. L. Welch and R.W. Exley J. Amer. Chem. SOC.,1967,89 3660. 227 M. Franck-Neumann Angew. Chem. Internat. Edn. 1967,6 79. G. L. Closs L.R. Kaplan and V.I. Bendall J. Amer. Chem. SOC.,1967,89 3376; A. C. Day and M. C. Whiting J. Chem. SOC.(B) 1967,991. 229 P. Beak and J. L. Miesel J. Amer. Chem. SOC.,1967,89,2375. 230 A. Padwa R. Gruber and L. Hamilton J. Amer. Chem. SOC.,1967,89 3077. 194 A. C. Day expected for concerted fragmentation of an electronically excited sulpho- lene.23la Disrotatory fragmentation occurs (as predicted) in the thermal decomposition of the sulpholenes (170).231bHuisgen and his co-workers have elegantly demonstrated that the thermal cleavage of aziridines to azomethine ylids is a conrotatory process whilst photochemical cleav.age is disrotatory.The ylids were generated in the presence of dimethyl acetylenedicarboxylate as 1,3-dipoIarophile and thereby stereospecifically trapped as 3-pyrrolines e.g. the cis-ester (172) was obtained in the thermally induced cleavage of the trans-aziridine (171).232 Aziridines thus behave in the way predicted for a system isoelectronic with the cyclopropyl anion.56 Carbenoid fragmentations of three-membered heterocyclic rings are deferred to the last section. Ar Ar (17 1) X =C0,Me;Ar =p-C,H,*OMe (172) 0 0 (174) hv /\ Q -(173) qCN + H UN -C(CN)z CN (177) (175) (176) Interest in the photochemical rearrangements of heterocyclic N-oxides continues.As with simple nitr~nes,~~~ the most common process is isomerisa- tion to an oxaziridine which may subsequently rearrange; but loss of the N-oxygen atom occurs in some cases. Such photodeoxygenations have been found with N-oxides in the ~yridine,~~~ and pyrida~ine~~~ pyra~ine,~~’ series. 231 (a)J. Saltiel and L. Metts J. Amer. Chem. Soc. 1967,89 2232; (b)W. L. Mock ibid. 1966,88 2857; S. C. McGregor and D. M. Lemal ibid. p. 2858. 232 R. Huisgen W. Scheer and H. Huber J. Amer. Chem Soc. 1967,89 1753. 233 For recent examples see L. S. Kaminsky and M. Lamchen J. Chem.Soc. (C) 1966 2295; J. B. Bapat and D. St. C. Black Chem. Comm. 1967 73. 234 J. Streith B. Danner and C. Sigwalt Chem. Comm. 1967 979. 235 N. Ikekawa and Y. Honma Tetrahedron Letters 1967 1197. 236 M. Ogata and K. Kanb Chem. Comm. 1967 1176. Photochemistry 195 (Azoxybenzene yields azobenzene as major product on photosen~itisation.~~~) Recent studies with N-oxides of quinolines isoquinolines quinoxalines quinazolines and phenanthridines provide examples of ring expansion to ring c~ntraction,~~’ oxazepines and o~adiazepines,~~~ and 1,2-shifts of oxygen.240 The quinoxaline di-N-oxide (1 73) photoisomerises with ring contraction to the benzimidazole derivative (174).241 The various types of rearrangement product can all be rationalised in terms of intermediate oxaziridines.The ylid (1 73 an N-oxide analogue rearranges photochemically to the aziridine (1 76) and 2-(2’,2’-dicyanoviny1)pyrrole( 177).242 Troponoid Compounds3’”-Further studies of the photodimerisation of tropone have appeared.243. 244 Spectroscopic evidence has been presented favouring the trans-configuration (178) for the (4 + 2)-dimer; this could be formed either concertedly from a ‘Mobius’ tropone or by stepwise trans- cy~loaddition.~~~~ The (6 + 4)-dimer (179) may also be formed by non-243n9 concerted cycloaddition. 2440 Sensitisation and quenching experiments have shown that tropone dimerisation occurs by a triplet mechanism.244a Bicyclo[5,1 ,O]octa-3,5-dien-2-one (2,3-homotropone) gives mainly isomeric products on irradiati~n.~~’ Two paths are available for the photoisomerisation of a-tropolones and their derivatives.As exemplified for a 5-substituted methyl ether (180) the possibilities are (i) 2,5-bonding to give (18 l),which may subsequently rearrange to (182) and (ii) 4,7-bonding to give (183). Path (i) is favoured for simple a-tropolones and their derivatives whilst path (ii) occurs in the colchicine series.’’” Mukai and his co-workers have described the first examples of path (ii) amongst simple monocyclic a-tropolones. 5-Phenyltropolone methyl ether (180; R = Ph) followed path (ii) exclusively on photolysis yielding compound (183 ; R = Ph). 5-Phenyl- and 5-chloro-tropolone and 5-chloro- tropolone methyl ether (180; R = C1) gave products by both pathways.246 The methyl ether of 6-phenyltropolone similarly gave analogues of (181) and (183); a further product in this case was the dimer (184).247 The light-induced 237 R.Tanikaga K. Maruyama R. Goto and A. Kaji Tetrahedron Letters 1966,5925. 0. Buchardt and J. Feeney Acta Chem. Scand. 1967 21 1399; 0.Buchardt Tetrahedron Letters 1966 6221 ; 0.Buchardt C. Lohse A. M. Duffield and C. Djerassi ibid. 1967 2741 ; C.Kaneko S. Yamada I. Yokoe and M. Ishikawa ibid. p. 1873;C. Kaneko and S. Yamada ibid. p. 5233. 239 0. Buchardt J. Becher and C. Lohse Acta Chem. Scand. 1966 20 2467; J. Streith H.K. Darrah and M. Weil Tetrahedron Letters 1966 5555. C. Kaneko I. Yokoe and M. Ishikawa Tetrahedron Letters 1967 5237; E.C.Taylor and G. G. Spence Chem. Comm. 1966,767.241 M. J. Haddadin and C. H. Issidorides Tetrahedron Letters 1967 753. 242 J. Streith and J.-M. Cassal Compt. rend. 1967,264,C 1307. 243 T. Tezuka Y. Akasaki and T. Mukai Tetrahedron Letters 1967,(a) p. 1397;(b) p. 5003; (c) cf. T. Mukai T. Tezuka and Y. Akasaki J. Amer. Chem. SOC. 1966,88 5025. 244 (a) A. S. Kende and J. E. Lancaster J. Amer. Chem. SOC. 1967,89 5283; (b) cf. A.S. Kende ibid. 1966,88 5026. 245 L. A. Paquette and 0.Cox J. Amer. Chem. SOC.,1967,89,1969. ’*’ T.Mukai and T. Miyashi Tetrahedron 1967 23 1613; T. Mukai and T. Shishido J. Org. Chem. 1967,32,2744. 247 T. Mukai T. Miyashi and M. C. Woods Tetrahedron Letters 1967,433. G* 196 A. C.Day (178) (179) a hv - R R (181) (182) rearrangement of tetra-0-methylpurpurogallin (185 ; R = OMe) to methyl 6,7,84rimethoxynaphthoate has received further attention.248 The desmethoxy- analogue (185; R = H) behaved differently giving the photoproduct (186).249 Miscellaneous.-Mixed solvents containing carbon disulphide have been recommended for dye-sensitised photo-oxidation~.~~~ Photo-oxidations of trop~lones~~ and several heterocyclic compounds252 have been reported.1,4-Dialkoxyanthracenes give 1,Cphoto-oxides (1 87) which on continued illumination isomerise to diepoxides (188).25 The 1,4-photo-oxide of 1,4-dimethoxy-5,8-diphenylnaphthaleneis the first such derivative to be isolated 248 0.L. Chapman and T. J. Murphy J. Amer. Chem. Soc. 1967,89 3476; cf. E. J. Forbes and R. A. Ripley J. Chem. SOC.,1959,2770.249 E. J. Forbes and J. Griffiths J. Chem. SOC.(C),1966,2072. E. J. Forbes and J. Grifiths Chern. Comm. 1967,427. 251 E. J. Forbes and J. Griffiths Chem. Comm. 1966,896;J. Chem. SOC.(C) 1967,601. 252 C. Dufraisse G. Rio and A. Ranjon Compt. rend. 1967 265 C 310; C. S. Foote M. T. Wuesthoff S. Wexler I. G. Burstain R. Denny G. 0.Schenck and K.-H. Schulte-Elte Tetrahedron 1967,23,2583. 253 J. Rigaudy N. C. Cohen and Nguyen Khim Cuong Compt. rend. 1967,264 C 1851. Photochemistry I97 in the naphthalene series. It reverts slowly to the parent hydrocarbon at room temperature.254 Cyclopropanes containing phenyl substituents undergo heterolytic photo- cleavage in protic solvents as well as the previously observed fragmentation to carbene~.~~~ A similar heterolytic reaction is the photochemical alcoholysis of oxiran~.~~~ New examples of the carbenoid photofragmentation of aryl- oxirans illustrate further the generality of the reaction.*''I Diphenylcarbene \/ (187) IOR ROT0 (188) NEt (189) (190) has been detected by e.s.r. and luminescence techniques in the photolysis of triphenyl- and tetraphenyl-oxiran at 77°K.258 Irradiation of the oxaziridine (189) with 2537 8 light in the presence of diethylamine gives the azepine (190) presumably by fragmentation of the oxaziridine ring to phenyl nitrene and cyclohexanone.259 Photolyses of 2-ben~oylaziridines~~' and a thi-iran-l-oxide26 have been described. Papers have appeared on the photochemistry of N-nitroso-compounds262 triphenylbor~n,~~~" triphenylmethyl hal- and sodium tetra~henylborate~~~~ 254 J.Rigaudy C. Delttang and J. J. Basselier Compt. rend. 1966,263 C 1435. 255 C. S. Irving R. C. Petterson I. Sarkar H. Kristinsson C. S. Aaron G. W. Griffin and G. J. Boudreaux J. Amer. Chem. SOC. 1966,88,5675. 256 K. Tokumaru Bull. Chem. SOC.Japan 1967,40,242. '"P. C. Petrellis H. Dietrich E. Meyer and G. W. Griffin J. Amer. Chem. SOC.,1967,89 1967; P. C. Petrellis and G. W. Griffin Chem. Comm. 1967,691. 258 A. M. Trozzolo W. A. Yager G. W. Griffin H.Kristinsson and I. Sarkar J. Amer. Chem. SOC. 1967,89 3357. '" E. Meyer and G. W. Griffin Angew. Chem. Internat. Edn. 1967,6,634. 260 A. Padwa and L. Hamilton J. Amer. Chem. SOC.,1967,89,102. 261 D. C. Dittmer G. C. Levy and G. E.Kuhlmann J. Amer. Chem. SOC.,1967,89,2793. 262 L. P. Kuhn G. G. Kleinspehn and A. C. Duckworth J. Amer. Chem. SOC. 1967 89 3858; 0.E. Edwards and R. S. Rosich Canad. J. Chem. 1967 45 1287; Y.L. Chow and A. C. H. Lee Chem. and Ind. 1967,827; T. Axenrod and G. W. A. Milne Tetrahedron Letters 1967,4443. 263 (a)J. L. R. Williams P. J. Grisdale and J. C. Doty J. Amer. Chem. SOC.,1967 89 4538; (b) J. L. R. Williams J. C. Doty P. J. Grisdale R. Searle T. H. Regan G. P.Happ and D. P. Maier ibid. p. 5153; J. L. R. Williams J. C. Doty P. J. Grisdale T. H. Regan and D. G. Borden Chem. Comm 1967. 109. I98 A. C. Day ides264 ~ulphones~~~ thiocyanates and isothiocyanates266 and a cyanatel"' i~ocyanate~~ b and nitrile Pyrimidine bases of biochemical interest have received considerable attention268 during 1967.Phosphorescence and e.s.r. studies show that the phosphorescence of native DNA originates in the thymine residues.z69 Thymine dimers and dihydrothymine have been isolated from DNA after photoly~is.~~~ vestigated.27 ' The photolysis of S-S bonds in proteins has been in-264 H. G. Lewis and E. D. Owen J. Chem. SOC.(B) 1967,422. 26s N. Kharasch and A. I. A. Khodair Chem. Comm. 1967,98. U. Mazzucato G. Beggiato and G. Favaro Tetrahedron Letters 1966 5455; G. Favaro and U. Mazzucato Photochrm. and Photobiol. 1967,6,589. 267 (a) M. Hara Y. Odaira and S. Tsutsumi Tetrahedron Letters 1967 1641; (b)J. S. Swenton ibid. p. 2855; (c)G. Just and W. Zehetner ibid. p. 3389. 268 N. Camerman S. C. Nyburg and D.Weinblum Tetrahedron Letters 1967,4127; E. Sztumpf- Kulikowska D. Shugar and J. W. Boag Photochem. and Photobiol. 1967,6 41 ;V. I. Danilov ibid. p. 233; D. Elad C. Kriiger and G. M. J. Schmidt ibid. p. 495; I. von Wilucki H. Matthaus and C. H. Krauch ibid. p. 497; M. Charlier and C. HBlbne ibid. p. 501; H. Becker J. C. LeBlanc and H. E. Johns ibid. p. 733; C. L. Greenstock I. H. Brown J. W. Hunt and H. E. Johnson Biochem. Biophys. Res. Comm. 1967,27,431; J. Eisinger and A. A. Lamola ibid. 1967,28 558; V. I. Danilov 0.V.Shramko and G. G. Dyadyusha Biokhimiya 1967,12,544. 269 R. 0.Rahn R. G. Shulman and J. W. Longworth J. Chem. Phys. 1966,452955. 270 T. Yamane B. J. Wyluda and R. G. Shulman Proc. Nat. Acad. Sci. U.S.A. 1967 58 439; A.A. Lamola and T. Yamane ibid. p. 443. 271 S. Risi K. Dose T. K. Rathinasamy and L. Augenstein Photochem. and Photobiol. 1967,6 423 ;K. Dose ibid. p. 437.