8 Photochemistry By H. A. J. CARLESS Department of Chemistry Birkbeck College Malet Street London WClE 7HX Interest in photochemistry continues unabated and the most recent volume of Specialist Periodical Report’ suggests that more than 1400 papers on organic photochemistry are appearing each year. The present review is of necessity highly selective but overall there is a clear trend towards the use of photochemistry as a routine synthetic method. Alkene photochemistry is an area of research where much effort continues to be expended both in intramolecular photochemical reactions of alkenes and in [2 + 21 photocycloaddition reactions in which an alkene plays a role. An interesting review has summarized the numerous intramolecular photochemical rearrangements and fragmentations undergone by alkenes and polyenes.* Srinivasan and his group have begun a series of investigations into the photo- chemistry of alkene~~.~ in pentane solution using high-energy and cyclopropane~~ light (185 nm 6.7 eV).Notable features include the rearrangement of cycloalkenes such as cyclohexene to (1) (@=0.05)and (2) (@=Oo.07).3Experiments using 3,3,6,6-[*H4]cyclohexenesuggest that both carbenes (3)and (4)are involved arising 6860 from C-C and C-H bond migration in the excited cyclohexene. These results amplify the previous observations by Kropp and co-workers6 that a carbene may be an intermediate in the rearrangement of alkenes caused by direct irradiation. Curiously these rearrangements are not reported in the irradiation of cyclo-octene at 185 nm where cis-trans photoisomerization OCCU~S.~The sum of the quantum yields of cis -+ trans and trans +cis isomerization is ca.2 which indicates that common intermediate is not reached. The excited species responsible for ‘Photochemistry’ a Specialist Periodical Report ed. D. Bryce-Smith. The Chemical Society London 1978,Vol. 9. G.Kaupp Angew. Chem. Internat. Edn. 1978,17 150. (a)R. Srinivasan and K. H. Brown J. Amer. Chem. SOC.,1978,100,4602;(6)Tetrahedron Letters 1978 3645. R. Srinivasan and K. H. Brown J. Amer. Chem. Soc. 1978,100 2589. R. Srinivasan and J. A. On J. Amer. Chem. SOC.,1978,100,7089. (a)T.R.Fields and P. J. Kropp J Amer. Chem. Soc. 1974,% 7559; (b)P. J. Kropp Mol. Photochem. 1978-9.9 39. 147 H.A. J. Carless isomerization cannot therefore be the TT*state but could be a Rydberg state of the cycloalkene. Another approach to the photochemistry of alkenes which does not involve vacuum U.V. excitation is that of irradiation of the copper(1) complexes of alkenes at 254 nm. Thus irradiation of the copper(1) trifluoromethanesulphonate (triflate) complexes of methylenecyclopropanes yields an array of rearrangement products which suggests a copper(1) carbenium ion as an intermediate.' Irradiation of copper(1)-cyclohexene or -cycloheptene complexes produces copper(1)-trans- cycloalkene isomers which may lead to dimerization may be trapped in a Diels- Alder reaction by dienes,8a or (at lower concentrations) rearrange to products analogous to (1)and (2).86 A reaction which might have synthetic potential is the use of copper(1) triflate to promote [2 +21 photocycloadditions such as those of norbor- nenes to ally1 alc~hol.~ Aromatic esters act as sensitizers for the cis-trans isomerization of cyclo-octene probably uia a singlet excited complex.The use of chiral aromatic esters can thus induce optical activity in the trans- cyclo-octene produced (enantiomeric excess up to 4%)." The rather surprising fact that photolysis of the tricyclic azo-compound (5) (5) gives cis,trans- cyclo-octa- 1,5 -diene as the major product becomes understandable when account is taken of the conformational preference of the 1,4-biradical inter- mediate.'' Excited states of phenylalkenes are known to have hydrogen-abstracting ability and Padwa et u1.12 have provided an interesting example of this process in the triplet-sensitized conversion of the cyclopropene (6)into (8) and (9).The products arise by hydrogen abstraction of the triplet cyclopropene from a cyclopropenyl y-hydrogen atom followed by coupling or disproportionation of the biradical(7) in a Ph Ph Ph Ph Ph Ph (6) (7) (8) (9) process which bears some resemblance to the Norrish Type 11reaction of carbonyl compounds. Electron transfer reactions are playing an increasingly important role in alkene photochemistry. Irradiation of cyanoaromatic sensitizers in the presence of phenyl- R. G. Salomon A. Sinha and M. F. Salomon J. Amer. Chem. SOC.,1978,100,520. (a) J. Th.M. Evers and A. Mackor Tetrahedron Letters 1978 2317; (b)ibid.,1978,2321.R. G. Salomon and A. Sinha Tetrahedron Letters 1978 1367. lo Y. Inoue Y. Kunitomi S. Takamuku and H. Sakurai J.C.S. Chem. Comm. 1978,1024. H.-D. Martin B. Heiser and M. Kunze Angew. Chem. Internat. Edn. 1978,17,696. l2 A. Padwa U. Chiacchio and N. Hatanaka J. Amer. Chem. Soc. 1978 100,3928. Photochemistry alkenes leads by electron transfer to alkene radical cations. For example irradiation of 1,l -diphenylethylene (10) and potassium cyanide in acetonitrile-2,2,2-trifluoro-ethanol solution in the presence of 1-cyanonaphthalene yields the anti-Markovnikov addition product (1l),via the alkene radical cation.13 Conversely the 1,l-diphenyl- ethylene radical anion can be produced by the use of electron-donating sensitizers as in the conversion of (10)under the conditions shown into the Markovnikov addition product (12)in a mild non-acidic route.14 Generation of an alkene radical cation in 1 -cyanonaphthalene 1 -methoxynaphthalene hv hw Ph2CHCH2CN-KCN Ph2C=CH2 -+ Ph2C-Me KCN I CHKN-CHKN-CN CF3CH20H CF3CH2OH the manner described can lead to interesting reactions such as photoaddition of phenylalkenes to furans,” and it also serves to explain the anti-Markovnikov addition of methanol to homoconjugated arylalkenes such as methyl-enebenzonorbornene.16 A related process of photocyanation occurs on irradiation of pyrroles in methanolic sodium cyanide solution using 1,4-dicyanobenzene as sensitizer.” A new type of photosensitized cyclobutane cleavage can be induced by an electron-transfer route.Thus selective excitation of phenanthrene in the presence of 1,4-dicyanobenzene and indene dimers results in cleavage of indene dimers to indene via a radical chain mechanism (ad 8.2).’* A novel route to medium-ring and macrocyclic compounds depends upon such electron transfer. Irradiation of N-alkenyl-phthalimides in alcohols results in cyclization with incorporation of ~olvent,~’ as in the conversion of (13)into (14).lga 0 0 0 (13) A satisfactory mechanism would involve electron transfer from the alkene as donor to the phthalimido group producing the alkene radical cation which is trapped by alcohol and then undergoes radical coupling to give the cyclized product (14). Similar l3 A. J.Maroulis Y. Shigemitsu and D. R. Arnold J. Amer. Chem. SOC.,1978,100 535. l4 D. R. Arnold and A. J. Maroulis J. Amer. Chem. SOC., 1977,99,7355. ’’ K. Mizuno R. Kaji H. Okada and Y. Otsuji J.C.S. Chem. Comm. 1978,594. l6 (a)D.D. Neidigk and H. Morrison J.C.S. Chem. Comm. 1978,600;(6)T.Nylund and H. Morrison J. Amer. Chem. SOC.,1978,100,7364. ” K. Yoshida J.C.S. Chem. Comm. 1978 1108. Is T.Majima C. Pac A. Nakasone and H. Sakurai J.C.S. Chem. Comm. 1978,490. l9 (a)K. Maruyama and Y. Kubo J. Amer. Chem. Soc.,1978,100,7772;(b)K. Maruyama Y. Kubo M. Machida K. Oda Y. Kanaoka and K. Fukuyama J. Org. Chem. 1978,43,2303. H. A. J. Carless electron transfer could be responsible for the unusual 6-hydrogen abstraction and cyclization found on photolysis of cycloalkanone 2-carbo~amides.~' Michl and co-workers2' have published a full account of the reactions in which irradiation at long wavelength of a cyclobutene [e.g.(15)I2l" converts it into a butadiene [e.g. pleiadene (16)] at 77 K by a mechanism which involves successive (15) (16) absorption of two photons with triplet cyclobutene as the intermediate. The implication of this work is that the possible involvement of higher excited states in solution photochemical processes should not be overlooked. [2 +21 Photocycloadditions are well known but there has been a resurgence of interest into possible synthetic conversions of the resulting cyclobutanes. For example the photoaddition of an enolized P-diketone to alkenes can be followed by a retro-aldol reaction to produce a 1,5-dicarbonyl compound.22 An intramolecular version of this transformation has allowed Oppolzer and G0de1~~ to provide an interesting alternative synthesis of longifolene (20) by [2 +21 photocycloaddition of the enol derivative (17),followed by retro-aldol reaction of (18)to yield (19) and subsequent conversion to (20).Pattenden and co-w~rkers~~ have also used this (19) X=H,Y=Z=O (20) X=Me Y =(Me)2 Z =CH2 approach in an ingenious route to the bicyclo[3,2,l]octane system of the gibberellins by irradiation of the enol acetates derived from 4-(prop-2-enyl)cyclopentane-1,3-diones followed by retro-aldol reaction. Cyclobutene-1-carboxylicacid2' and its methyl ester26 have been used in [2 +21 photocycloadditions to give functionalized bicyclo[2,2,0]hexanes.Treatment of the bicyclohexane with lithium in liquid ammonia produces cis-fused cyclohexanes,26 2o T. Hasegawa M. Inoue H. Aoyama and Y. Omote 1. Org. Chem. 1978,43 1005. 21 (a)A.Castellan J. Kolc,andJ.Michl J. Amer. Chem.Soc. 1978,100,6687; (b)M. A. Souto J. Kolc,and J. Michl ibid. 1978,100 6692. 22 P. de Mayo Acc. Chem. Res. 1971,4,41. 23 W. Oppolzer and T. Godel J. Amer. Chem. Soc.. 1978,100,2583. 24 M. Mellor D. A. Otieno and G. Pattenden J.C.S. Chem. Comm.. 1978 138. 25 G. L. Lange and F. C. McCarthy Tetrahedron Letters 1978,4749. 26 P. A. Wender and J. C. Lechleiter I. Amer. Chem. SOC.,1978,100 4321. Photoehemistry whilst heating transforms the bicyclohexanes into derivatives having the germacrane or cadine ring ~ystems.~’ The carbon skeleton of the ring system found in hirsutic acid and related sesquiterpenes can be produced by [2+21 photocycloaddition of dicyclopent-1-enylmethanessuch as (21) in methanol to give the cis,syn cis-tricyclic derivatives[e.g.(22)]which are presumably formed by in situ addition of methanol to (21) (22) the intermediate bicycl0[2,1,0]pentanes.~~Two new routes to propellanes which have appeared involve the intermolecular [2+21photocycloadditions of diphenyl- acetylene or trans-stilbene to bicyclic alkenes,28 and the intramolecular cyclization of an alkenylcyclohexenone (23) to give a tricyclic ketone (24) followed by its unusual acid-catalysed conversion into (25).29 (23) (24) (25) There have been several other notable advances in the field of intermolecular [2+21 photocycloadditions.Lewis and Johnson3* claim to have observed different addition reactions from excited singlet stilbene monomer and excimer with dimethyl fumarate; the monomer reaction leads to a cyclobutane whilst the excimer yields oxetans. Cantrel13’ has clearly shown the features which control the site of attack of alkenes on benzonitrile yielding cycloaddition either at the nitrile group or at the aromatic ring. Irradiation of thiocarbonyl compounds in the presence of alkynes has been shown by two groups32 to give the first stable examples of [2 +21 cycloaddition products thietes. The regioselectivity of rearrangement of a divinylmethane depends on the substi- tution att tern;^^,^^ for example the cyano or methoxy substituents in (26) ensure exclusive photorearrangement to quite different products (27)and (28) respec-ti~ely.~~ The stereochemistry of two of the three major types of di-wmethane rearrangement has already been established.Whereas the divinylmethane re- arrangement gives inversion of configuration at the methane carbon atom the 27 J. S. H. Kueh M. Mellor and G. Pattenden J.C.S. Chem. Comm. 1978 5. 28 G. Kaupp and M. Stark Angew.Chem. Internut. Edn. 1978,17,758. 29 R. L. Cargill J. R. Dalton S. O’Connor and D. G. Michels Tetrahedron Letters 1978,4465. F. D. Lewis and D. E. Johnson J. Amer. Chem. Soc. 1978,100,983. 31 T. S. Cantrell J. Org. Chem. 1977,42,4238. 32 (a)H. Gotthardt and 0.M. Huss Tetrahedron Lelters 1978,3617; (b)A.C. Brouwer A. V. E. George D. Seykens and H. J. T. Bos. ibid. 1978,4839. 33 D. W. Alexander A. C. Pratt D. H. Rowley and A. E. Tipping J.C.S. Chem. Comm. 1978,101. 34 H. E. Zimmerman and R. T. Klun Tetrahedron,1978,34,1775. H.A. J. Carless oxa-di-vmethane case leads to a loss of stereochemistry. Zimmerman and co- worker~~~ have now determined the stereochemistry of the third type involving an arylvinylmethane(29) in which it rearranges to (30)and (31)with complete inversion (29) (30) (31) at the methane carbon atom. However the fast rate of photoracemization of the cyclopropanes meant that rearrangement had to be extrapolated to zero conversion to establish this stereospecificity. Zimmerman's continued detailed study of the rates and regioselectivities of rearrangement of a variety of aryl~inylmethanes~~ and tetra-aryl substituted di~inylmethanes~~ has led him to propose a theory which is capable of explaining such reactions using SCF-CI calculation^.^^ Briefly AP for each pair of orbitals is the change in bond order of the excited state compared to the ground state and the constructionof a 'APmatrix' gives an idea of how excitation energy is partitioned in a molecule.The extension of the concept to cover the kind of molecular motions likely to convert electronic into vibrational energy can be applied to a wide variety of photochemical reactions.39 One of the most remarkable syntheses reported this year must be that of tetra-t-butyltetrahedrane (32). Whilst matrix irradiation of the trisubstituted cyclopentadienone (33) has been shown to produce the [2 +21 cycloadduct 'housenone' (35),,' the addition of an extra t-butyl group as in (34) causes exclusive [2 +21 criss-cross addition on matrix isolation to give tricyclopentanone (36).,l Bu' n R B;' 'H Bu' Bu' (33) R=H (35) (34) R=Bu' 35 H.E. Zimmerman,T. P. Gannett apd G. E. Keck J. Amer. Chem. Soc. 1978,100 323. 36 H. E. Zimmerman M. G. Steinmetz and C. L. Kreil J. Amer. Chem. Soc. 1978,100,4146. 37 (a)H. E. Zimmerman and W. T. Gruenbaum,J.Org. Chem. 1978,43,1997; (b)H. E. Zimmerman and T. R. Welter J. Amer. Chem. SOC.,1978,100,4131. 38 H. E. Zimmerman W. T. Gruenbaum R. T. Klun M. G. Steinrnetz and T. R. Welter J.C.S. Chem. Comm..1978,228. 39 H. E. Zimmerman and M. G. Steinmetz J.C.S. Chem. Comm. 1978,230. 40 G. Maier U. Schiifer W. Sauer H. Hartan R. Matusch and J. F. M. Oth Tetrahedron Letters 1978 1837. 41 G. Maier S. Pfriem U. Schafer and R. Matusch Angew. Chem. Internat. Edn. 1978,17,520. Photochemistry 153 Irradiation of this latter material in diethyl ether at -100 "Cgives decarbonylation to yield the tetrahedrane (32) which is sufficiently stable that it can be made to isomerize to tetra-t-butylcyclobutadieneonly by heating above 130 "C. The Norrish Type 11 photoelimination reaction of carbonyl compounds is well known to occur uia a l,4-biradical. A competing process which is available to the 1,4-biradical(38) from 6-substituted valerophenones (37) is the radical loss of X'to produce a yS-unsaturated ketone (39).Wagner's group4* have examined the competition between formation of Type I1 products and (39) for a variety of &substituted ketones which allows a calculation of the relative rates of radical p-cleavage in these biradicals. Amongst the halogens the radical leaving abilities are I > Br > C1 to the extent that (39) becomes the only product detected from irradiation of (37; X = I). (37) PhCOCHZCHzCH = CHz (39) Scaiano's group4345 have continued to find applications for their important technique in which 1,4-biradical lifetimes are measured by trapping with paraquat dication (l,l'-dimethyl-4,4'-bipyridylium)and following the production of the coloured paraquat radical cation Lifetimes of triplet biradicals from phenyl alkyl ketones (typically -100 ns) are surprisingly insensitive to changes in temperature substitution pattern and solvent suggesting a process like intersystem crossing (to singlet) as the main deactivation pathway.43 In contrast an aliphatic aldehyde such as valeraldehyde gives rise to a much longer lived 1,4-biradical (2 ps in water-a~etonitrile).~~ In the photoenolization of o-methylacetophenone Small and Scai- an045 have found evidence for only one biradical intermediate (lifetime 300 ns) which helps to clarify the mechanisms proposed for this complex and controversial photochemical reaction.One of the usual assumptions about photochemical processes is that they are independent of temperature wavelength and light intensity.Each of these assump- tions has been shown in certain circumstances to be invalid for carbonyl compounds especially when radical intermediates may be involved in such photochemical processes. Thus the ratio of products from the photoaddition of aldehydes to bicyclo[2,2,2]octan-2,3-dionedepends strongly on light intensity,46 and acenaph- thoquinone appears to undergo wavelength-dependent processes of photoaddition and photo-oxidation in ~olution.~' Citral (40) is predominantly converted into '' (a)P. J. Wagner J. H. Sedon and M. J. Lindstrom J. Amer. Chem. SOC.,1978 100 2579; (b)P.J. Wagner and J. H. Sedon Tetrahedron Letters 1978 1927. 43 R. D. Small and J. C. Scaiano J. Phys. Chem. 1977,81 2126. O4 M.V.Encinas and J. C. Scaiano J. Amer.Chem. Soc. 1978,100 7108. " R.D.Small and J. C. Scaiano J. Amer. Chem. SOC.,1977,99,7713. " M.B.Rubin and S. Inbar J. Amer. Chem. Soc. 1978,100,2266. 47 T.-S. Fang and L. A. Singer J. Amer. Chem. SOC.,1978,100,6276. H.A. J. Carless aldehydes (41) and (42) on irradiation at high temperature (165-190 "C),perhaps via a 1,2-formyl shift whereas these products are not seen at all on irradiation at room temperat~re.~' CHO hv 280"C e (40) (41) R' = Me R2=CHO (42) R' = CHO R2=Me The much-studied ring expansion of cyclobutanones to cyclic acetals which occurs on irradiation in alcohols has been exploited in a synthesis of the prostaglandin precursor (44) by irradiation of the bicycloheptanone (43).49 Yields of (44) are .. .I I.0 0 increased to 42% in the presence of sodium bicarbonate and 2,5-dimethylhexa-2,4- diene though the function of this latter triplet quencher in increasing the yield is not obvious. Photolytic rearrangement of 2,3-epoxycyclohexa- 1,4-diones such as (45) occurs on irradiation in acetone to yield the y-alkylidene butyrolactones (46),"" R-Go U= 1;:. 0 H CHO (45) (46) which may in turn be used to provide a route to the y-alkylidene-A"*'-butenolide structures found in a variety of natural Working to a plan conceived nearly 15 years ago Quinkert Barton and co- workersS' have now completed a three-step synthesis of dimethyl crocetin (49). The key step involves the photochemical ring opening of a bis-cyclohexadienone (47) to a bis-ketene (48).F. Barany S. Wolff and W. C. Agosta J. Amer. Chem. SOC.,1978,100 1946. O9 N. M. Crossland S. M. Roberts and R. F. Newton,J.C.S. Chem. Cornm. 1978,661. 50 (a)T. Kitamura T. Imagawa and M. Kawanisi Tetrahedron Letters 1978 3443; (6)T.Kitarnura Y. Kawakami T. Irnagawa and M. Kawanisi Tetrahedron Letters 1978,4297. 51 G. Quinkert K. R. Schmieder G.Durner K. Hache. A. Stegk and D. H. R. Barton Chem. Ber. 1977 110,3582. Photochemistry 155 (49) Most py-unsaturated ketones give two characteristic photoreactions triplet sensitized oxa-di-m-methane rearrangement [e.g.(50)+ (5l)] from the lowest (50) (51) triplet mr* state or a 1,3-acyl shift [e.g. (50)-+ (52)] on direct irradiation that has been variously ascribed to a singlet nw* or a second triplet nr* state.In an interesting attempt to show the state responsible for this latter reaction Schaffner et al.52have generated the triplet n7zX state of the enone (50) by thermolysis of the dioxetan (53) comparing the product distribution with that obtained from direct and O-O OMe 4 x’. 0 (52) (53) sensitized photolyses. There is certainly evidence that the 1’3-acyl shift on irradia-tion of (50)can occur from the triplet nn* state although the additional involvement of the singlet n7r* state cannot be excluded. CIDNP Studies during irradiation of y-methyl substituted acyclic py-unsaturated ketones also suggest that the 1,3-acyl shift occurs mainly from a triplet excited Photolysis in micelles can have a very important effect on the fate of radical intermediates.For example irradiation of p-tolyl benzyl ketone in a micelle (at low ketone to micelle ratios) yields only 1-p-tolyl-2-phenylethane (ArCH2CH2Ph) with none of the cross-coupled products (PhCH2CH2Ph or ArCH2CH2Ar) which are found on irradiation under normal solution condition^.^^ Under the intriguing title of ‘sunlight and soap for the efficient separation of I3Cand ’*Cisotopes’ Turro and ’’ M. J. Mirbach A. Henne and K. Schaffner,J. Amer. Chem. Soc. 1978,100,7127. s3 A.J. A. van der Weerdt H. Cerfontain J. P. M. van der Ploeg and J. A. den Hollander J.C.S. Perkin IZ 1978,155. N. J. Turro and W. R. Cherry J. Amer. Chem. Soc. 1978,100,7431. H.A. J. Carless KraeutlerS5 describe how the irradiation of dibenzyl ketone in micelles can lead to a process of high isotope enrichment at the carbonyl group of the unreacted ketone by means of a magnetic isotope effect.Another example of the importance of environment in photochemistry is provided by the irradiation of 16-0x0- 16-p- tolylhexadecanoic acid where Type I1 elimination is enhanced in micelles compared to solutions but effectively prevented in monolayer Crown ethers have inevitably found their ways into photochemistry with reports which range from dramatic changes in photophysical properties of a crown ether naphthalene derivative on complexation with a caesium cation,” to catalysis by potassium ions of the Type I1 elimination process on irradiation of the valerophenone derivative prepared from dibenzo-18-crown-6.’* Lithium ions can stabilize the 12-crown-4 product formed by intramolecular photochemical cycloaddition of two anthracene residues linked by a polyoxyethylene chain although they do not appear to increase the efficiency of the cy~lization.~’ Since last year’s Report,60 more evidence has accumulated for the possibility that attack of singlet oxygen (lo2) on electron-rich alkenes occurs by way of a zwitterionic peroxide intermediate,61 rather than in a concerted process.The reaction of ‘02with norbornenyl ethers in methanol leads to zwitterion trapping by the whilst reaction of lo2with the norbornadiene (54) gives intramolecular trapping of the zwitterion by the remaining double bond to yield peroxide (55).61b There is 0-0 increasing recognition that photosensitized oxidation does not need to involve lo2 but may instead proceed through formation of the superoxide radical anion (02‘-) in an electron-transfer reaction.62 Research in this area may be assisted by the publication of a photochemical method for the generation of superoxide ion in aqueous Schultz et al.have applied the photocyclization of aryl vinyl and sulphides6’ to the synthesis of dihydrofurans and dihydrothiophens respectively. These reactions can provide a useful arylation process exemplified in the total ” N. J. Turro and B. Kraeutler J. Amer. Chem. SOC.,1978,100,7432. 56 P.R. Worsham D. W. Eaker and D. G. Whitten J. Amer. Chem. SOC.,1978,100,7091. ” J. M.Larson and L. R. Sousa J. Amer. Chem. SOC.,1978,100,1943.’* R. R. Hautala and R. H. Hastings J. Amer. Chem. SOC., 1978,100,648. 59 J.-P. Desvergne and H. Bouas-Laurent J.C.S. Chem. Comm. 1978,403. 6o H. A.J. Carless Ann. Reports 1977,74 165. (a)C. W. Jefford and C. G. Rimbault J. Amer. Chem. Soc.,1978,100,6437;(b)ibid.,1978,100,6515. 62 (a)C. W. Jefford and A. F. Boschung Helv. Chim. Acta 1977,60,2673;(b)J.Eriksen C. S. Foote and T. L. Parker J. Amer. Chem. SOC.,1977 99,6455; (c)J. P.McCormick and T. Thomason. ibid.,1978 100 312; (d)V.S.Srinivasan D. Podolski N. J. Westrick and D. C. Neckers ibid. 1978,100 6513. 63 R. A. Holroyd and B. H. J. Bielski J. Amer. Chem. SOC.,1978 100 5796. A. G. Schultz R. D. Lucci W. Y. Fu M. H. Berger J. Erhardt and W. K. Hagmann J. Amer. Chem. SOC. 1978,100,2150. Photochemistry 157 synthesis of the alkaloid lycoramine by the photocyclization of (56) to the dihy- drofuran (57).Irradiation of phenethyl vinyl ether (PhCH2CH20CH=CH2) in solution gives the unusual intramolecular cycloadduct (58),resulting from attack of the vinyl group on the 2,5-positions of the aromatic ring.66 qyyj C0,Me R C0,Me (56) R = CH2CH2NHC02Me (57) (58) Photorearrangement of five-membered ring heterocycles continues to provide results of mechanistic interest. The phototransposition of 2-cyanopyrroles to 3-cyanopyrroles has been proposed to occur by 2,5-bonding and 'walk' of the aziridine nitrogen atom as shown in Scheme 1. Conclusive evidence for (59) as an inter- Scheme 1 mediate in 2-cyano-N-methylpyrrole rearrangement results from its trapping by irradiation in methanol or f~ran.~' The related 1,3-phototransposition of an indene provides no evidence for a bicyclopentene intermediate but seems more in accord with a concerted di-.rr-methane mechanism;68 inversion occurs cleanly at the migrat- ing centre as for the arylvinylmethane rearrangement.35 Another pathway for photochemical rearrangement in five-membered ring heterocycles is that of the dihydrofuran (60)to acylcyclopropane (61),being a reaction used as an entry to the chrysanthemum monocarboxylic acids.69 szMe po2Me COMe (60) (61) " A.G. Schultz W. Y. Fu R.D. Lucci,B. G. Kurr K. M. Lo and M. Boxer J. Amer. Chem. Soc. 1978,100 2140. '' A. Gilbert and G. Taylor J.C.S. Chem. Comm. 1978 129. '' J. A.Barltrop A. C. Day and R. W. Ward J.C.S. Chem. Comm. 1978 131. 68 D. Giacherio and H. Morrison J. Amer. Chem. Soc. 1978,100,7109. 69 K. Ohkata T. Isako and T. Hanafusa Chem. and Ind. 1978 274. H. A. J. Carless Ring expansion of aryl azides to 3H-azepines on irradiation in the presence of nucleophiles is a well-known reaction examples of which are found in the recently- observed reactions of 0-azidobenzoic acid derivatives70a and azidoquin~lines.~~~ The reaction has been assumed for the last 20 years to involve an azirine intermediate (62) formed from phenylnitrene (63). However Chapman and Le Roux71 have shown that irradiation of phenyl azide in a matrix at 8 K leads to a product (vmu 1895 cm-l) assigned the ketenimine structure (64). Reaction of the keteni- mine with nucleophile (e.g.R,NH) would lead to 1H-azepine (65) followed by isomerization to 3H-azepine (66) which is the normal end-product. Recent e.s.r. studies also suggest that 2-pyridylmethylene (67)may be formed by irradiation of the (62) (63) (64) (65) (66) (67) ketenimine (64).72 When all else fails another means of providing evidence for intermediates in such ring-expansion reactions may come from the observation of isotope effects such as those found in the conversion of 1-iminopyridinium ylides into 1,2-diazepine~.~~ One of the most useful methods for the preparation of cyclophanes involves the extrusion of sulphur or sulphur dioxide from bis-sulphides or bis-sulphones respec- tively. Givens and W~lie~~ report that the efficient extrusion of sulphur dioxide by photolysis of bis-sulphones suspended in benzene represents a reasonable alter- native to pyrolysis.The extrusion of sulphur is an important method for C-C coupling in macrocyclic systems often carried out by photolysis of sulphides in the presence of trialkyl phosphites. This latter reaction has allowed the preparation of triple-layered paracy~lophanes,’~ and of the extraordinarily insoluble hydrocarbon Kekulene with its cyclic fused system of 12 benzenoid rings.76 A new method for the cyclopropanation of alkenes involves photolysis of di-iodomethane in the presence of an alkene and it has the advantage over the Simmons-Smith reaction in being little affected by steric constraints.” The absence of C-H insertion products suggests that methylene itself is not involved.70 (a) R. Purvis R. K. Smalley W. A. Strachan and H. Suschitzky J.C.S. Perkin I 1978 191; (6)F. Hollywood E. F. V. Scriven H. Suschitzky D. R. Thomas and R. Hull J.C.S.Chem. Comm. 1978,806. 71 0.L. Chapman and J.-P. Le Roux 1Amer. Chem. SOC.,1978,100,282. 72 0.L.Chapman R. S. Sheridan and J.-P. Le Roux J. Amer. Chem. SOC.,1978 100,6245. 73 (a)H.Kwart D. A. Benko J. Streith D. J. Harris and J. L. Schuppiser J. Amer. Chem. SOC.,1978,100 6501;(6)H.Kwart D. A. Benko J. Streith and J. L. Schuppiser ibid. 1978 100 6502. 74 R. S. Givens and P. L. Wylie Tetrahedron Lztters 1978 865. ” T. Otsubo T. Kohda and S. Misumi Tetrahedron Letters 1978 2507. 76 F.Diederich and H. A. Staab Angew. Chem. Internat. Edn. 1978,17,372. 77 N.J. Pienta and P. J. Kropp J. Amer. Chem. SOC.,1978,100,655.