3 Reaction Mechanisms Part (ii) Orbital Symmetry Correlations and Pericyclic Reactions By R. GRIGG Department of Chemistry Queen's University Belfast B T9 5AG Northern lreland 1 General and Theoretical Aspects The now numerous theoretical treatments of pericyclic processes have provided a number of perspectives of concerted reactions. For the organic chemist some of these have been helpful and some incomprehensible but their conclusions have been overwhelmingly in accord with those of the original Woodward and Hoffmann treatment. In this context a recent series of papers by Epiotis' deserves mention for its attempt to encompass substituent effects the principle of least motion and non-stereoselectivity. In a series of papers which are a sequel to those mentioned last year,2 the concept of configuration interaction is applied to pericyclic reactions.This approach leads to the prediction that configuration interaction can reverse the stereoselectivity of a 4N pericyclic process by rendering the stereoselectivity of reactions towards the polar end of the spectrum opposite to the stereoselectivity of the non-polar reactions. In 4N pericyclic processes non-stereoselectivity is then the result of two competing concerted processes and should be found near the middle of the polarity spectrum. Polarity in this context refers to electron-donating and -attracting substituents on the substrate(s). In contrast 4N + 2 electron pericyclic processes are predicted to maintain their stereoselective preference throughout the entire range of substrate type.Looking at this from an aromaticity ~iewpoint,~ the conclusions can be restated in the form that an aromatic transition state is not influenced by substitution whereas an anti-aromatic transition state is responsive to substitution. The latter conclusion accords with the prediction4 that the orbital properties of cyclo- butadiene and its anti-aromaticity strongly depend on substitution and geometry. The original treatment of pericyclic processes by Woodward and Hoffmann leads to the prediction that the energy ordering of transition states for a reaction ' N. D. Epiotis J. Amer. Chem. Soc. 1973 95 1191 1200 1206 1214. R. Grigg Ann. Reporrs (B) 1972 69 120. M. J. S. Dewar Angew. Chem. Internat. Edn. 1971 10 761.R. Hoffmann. Chem. Comm. 1969 240. 75 76 R.Grigg should be ‘allowed’ < biradical or zwitterionic -= ‘forbidden’. Therefore the interaction between reactive sites in a biradical or zwitterionic case should be minimal since such interaction generates ‘forbidden’ character. This minimizing of interactions should in turn result in loss of stereospecificity. However a number of ‘forbidden’ processes occur with facility and stereospecificity and these are attracting more attention.2 Configuration interaction first applied to these problems by Schmidt,’ has been utilized in rationalizing a number of these cases6 Baldwin6 has criticized the use of ‘concerted’ and ‘allowed’ as inter- changeable synonyms and points out that orbital symmetry is conserved in some but not all concerted reactions and not conserved in some but not all non-concerted reactions.The sigmatropic sub-group has been discussed and the importance of subjacent orbital control as a significant electronic factor favouring stereospecific concerted ‘forbidden’ reactions has been proposed. Thus the transition state of a thermal suprafacial 1,3-sigmatropic process occurring with retention at the migrating carbon centre i.e. a forbidden .2 + 02 process is appreciably stabilized by interaction of the carbon p orbital of the migrating centre with a subjacent bonding ally1 orbital. This results in two of the four electrons involved being accommodated in a more stable orbital than would be the case in a non-interacting biradical transition state (Figure 1).Both 11/1 and t,h3 mix with p [see Figure l(a)] but the interactions approximately cancel each other. Subjacent orbital control of the transition state is not expected to become important until steric factors become unfavourable to the operation of Woodward-Hoffmann control. This revised ordering of transition state energies ‘allowed’ < ‘forbidden’ concerted < non-interacting biradical or zwitterion will itself be susceptible to variation and calculations on the degenerate methylenecyclo- propane rearrangement show the orthogonal biradical transition state is more stable than the ‘forbidden’ concerted ,2 + .2 pathway.8 The first paper applying trajectory techniques on quantum-mechanical potential surfaces to the dynamics of organic reactions’ is concerned with the insertion of singlet methylene into a hydrogen molecule.A number of reviews on pericyclic processes have appeared lo some of which are devoted exclusively or in part to applications in inorganic chemistry.” The proposed new class of orbital symmetry controlled reactions dyotropic rearrangements,12 which W. Schmidt Tetrahedron Letters 1972 58 1 ; Helv. Chim. Acra 197 1 54 862. J. E. Baldwin A. H. Andrist and R. K. Pinschmidt Accounts Chem. Res. 1972,5,402. J. A. Berson Accounts Chem. Res. 1972 5 406; J. A. Berson and L. Salem J. Amer. Chem. SOC. 1972,94 8917. W. T. Borden and L. Salem J. Amer. Chem. SOC. 1973,95,932. I. S. Y. Wang and M. Karplus J. Amer. Chem. SOC. 1973,95 8160. lo R. F. Hudson Angew.Chem. Internat. Edn. 1973,12 36; K.-W. Shen J. Chem. Educ. 1973 50 238; J. Mathieu Bull. SOC. chim. France 1973 807. ‘ ‘ K. Fukui and H. Fujimoto Kyoto Daigaku Nippon Kagakuseui Kenkyusho Koeushu 1972 29 27 (Chem. Abs. 1973 78 146 899); H. Fujimoto S. Kato S. Yamabe and K. Fukui Bull. Chem. SOC. Japan 1973 46 1071; R. G. Pearson Fortschr. Chem. Forsch. 1973 41 75. R. Grigg Ann. Reports (B) 1972 69 121. Reaction Mechanisms -Part (ii) Orbital Symmetry I E =4f10 E =2f10 I I I- I \ \ I +3-s" \ \ I $3-\ \ E \ I \ \ I 0 ++A $-f\:sfpI~2+ -4-+ -fP I I I I I I 3t "1#-2PO I I I I Figure 1 MO energy ordering for (a)a tforbidden' concerted *2 +,2 process and (b) a non-interacting biradical process involves the simultaneous migration of two a-bonds has been further studied.' MIND0/2 calculations and orbital symmetry considerations are utilized to analyse the stereochemical fate of migrating groups in thermal and photochemical dyotropic processes.An explanation of the a-effect the enhanced reactivity of nucleophiles possess- ing an unshared electron pair adjacent to a nucleophilic centre has been given in terms of Mobius-Hiickel theory.14 The a-effect has only been definitely observed for attack of nucleophiles on substrates with n-bonding. This is now rationalized in terms of a six-electron aromatic transition state to which the unshared electron pair adjacent to the nucleophilic centre contributes e.g. the addition of hydrazine to a carbonyl compound (1).An interesting application l3 M. T. Reetz Tetrahedron 1973 29 2189. l4 J. F. Liebman and R. M. Pollack J. Org. Chem. 1973 38 3445. 78 R. Grigg of frontier orbital theory to the direction of attack of nucleophiles and electro- philes on cyclohexanones has appeared. ’ One of the current problems which restricts the full practical exploitation of pericyclic processes is that it is not yet possible to predict when applicable which of several symmetry-allowed possibilities will be preferred. A contribution to this problem comes from the study of the thermal rearrangement of homo- bullvalenone (2) to (3).l6 Deuterium-labelling studies provide evidence for a vinylketone intermediate and are rationalized in terms of two ‘allowed’ eight- electron steps when a single six-electron process could have achieved the same result.2 Electrocyclic Reactions A book on valence isomerizations has appeared.” A bond order rule for electro- cyclic reactions has been developed.’8 Disrotatory closure is preferred if the bond order between the reacting centres is positive and conrotatory closure preferred if the bond order is negative. This rule applies to both symmetrical and non-symmetrical molecules and highlights a class of electrocyclic reactions where the bond order between reacting centres is zero or very small. This will result in non-stereospecific closure but the product ratio can be altered by introducing substituents. Ab initio calculations have been carried out on various cyclopropyl species.” MO calculations predict that C-C bond opening of thi-irans should be conrotatory but the corresponding sulphoxide should open in a disrotatory manner.” The anion (4) undergoes electrocyclic opening ca.lo4 times faster than anion (5).2’ Anion (4) can undergo conrotatory opening whereas (5) cannot and it is concluded that conrotatory opening is the favoured path by at least 23.8 kJ mol- The key step in the thermal rearrangements of aziridinyl ketones to pyrroles 15 J. Klein Tetrahedron Letters 1973 4307. 16 M. J. Goldstein and S.-H. Dai J. Amer. Chem. SOC.,1973 95 933. 17 G. Maier ‘Valence Isomerizations’ Chemical Pocket Books Verlag Chemie Wein- heim 1973 Vol. 17. 18 E. E. Weltin J. Amer. Chem. SOC.,1973 95 7650. 19 A.Liberles A. Greenberg and A. Lesk J. Amer. Chem. SOC.,1972,94,8685; L. Farnell and W. G. Richards J.C.S. Chem. Comm. 1973 334; L. Radom P. C. Hariharan J. A. Pople and P. v. R. Schleyer J. Amer. Chem. SOC.,1973 95 6531. 20 R. Hoffmann H. Fujimoto J. R. Swenson and C.-C. Wan J. Amer. Chem. SOC.,1973 95 7644. 21 M. Newcomb and W. T. Ford J. Amer. Chem. SOC.,1973,95 7186. Reaction Mechanisms -Part (ii) Orbital Symmetry [e.g.(6)+(7)]22and trans-divinyloxiran to a dihydrofuran [(8)-+(9)]23is the initial conrotatory opening of the heterocycle to a dipolar intermediate i.e. (10) and (1 l),respectively.A related ring-opening occurs in the thermal rearrangement of (u).~~Rearrangement is preceded by an endo-ex0 isomerization [(12)-B (1411 which it is suggested involves (13)formed by a disrotatory electrocyclic process.The intermediate can undergo a 1,4-hydrogen shift to give (15) while (12) and (14) rearrange to dienals by a concerted mechanism involving fission of an external cyclopropyl bond. Stereomutation of allyl cations usually proceeds via perpendicular (i.e. non- delocalized) allyl cations. However ub initio calculations show that variation of X in (16) should when X = Me or F lead to stereomutation [(16)+ (IS)] via cyclopropyl cation (17).25 When X = OH or NH the cyclopropyl cations (17) are more stable than the corresponding allyl cations. The two isomeric pentadienyl cations (19a) and (19b) have been observed in FS03H-S02C1F at -125 "C.They undergo thermal conrotatory cyclization to cyclopemtenyl cations.26 The bicyclo[5,1,O]octadienyl anion (20) undergoes disrotatory photo- equilibration with the monocyclic isomer (21).27 The diazo-group participates in a number of electrocyclic reactions which lead to diazepines e.g.(22)-+ (23).28 22 A.Padwa D. Dean A. Mazzu and E. Vega J. Amer. Chem. SOC.,1973,95 7168. 23 R. J. Crawford V. Vukov and M. Tokunaga Canad. J. Chem. 1973,51 3718. 24 J. Wolfhugel A. Maujean and J. Chuche Tetrahedron Letters 1973 1635. '' L. Radom J. A. Pople and P. v. R. Schleyer. J. Amer. Chem. SOC.,1973,95 8193. 26 N. W. K. Chin and T. S. Sorensen Canad. J. Chem. 1973,51 2776. 27 S. W. Staley and N. J. Pearl J. Amer. Chem. SOC.,1973,95 2731. A. A. Reid J. T. Sharp H. R. Soad and P. B. Thorogood J.C.S. Perkin I 1973,2543.80 R.Grigg Me Me (19) a; R' = RZ = Me R3 = H Me b; R' = R3 = Me R2 = H (20) (21) Formation of the iron tricarbonyl complex of cis4-cyclononatetraene (24) stabilizes the ligand and retards the electrocyclic closure to a dihydroindene (25)29 whereas attachment of acetylenic groups to cyclo-octatriene (26) facili-tates ring opening to (27).30 Kinetic studies on the thermal disrotatory cycliza- tion of a series of I-and 3-alkyl-l,3,5-hexatrienesyielded relative rates in the order 3Bu' > 3-Et 3-Me > 1-Et > 1-Me H.31 The activation enthalpies of the 3-alkyl series were in general 12.5 kJ mol-' less than either the 1-alkyl counterparts or the parent hydrocarbon. Six-electron anionic electrocyclic processes have been observed in dihydrobenzthiophenes [(28) -+ (29)13*and enolate anions of eucarvone [(30) -+(31)]; the related enolate (32) undergoes equilibration via an anionic Cope rearrangement at 150 "C[(32) -+ (33),4 5)].33 FCZCR -+ 29 E.J. Reardon and M.Brookhart J. Amer. Chem. SOC.,1973 95,431 1. 30 H. Straub J. M. Rao and E. Muller Annafen 1973 1339 1352. 31 C. W. Spangler T. P. Jondahl and B. Spangler J. Org. Chem. 1973 38 2478. 32 H. Kloosterziel and J. A. A. van Drunen Tetrahedron Letters 1973 1023. 33 A. J. Bellamy and W. Crilly Tetrahedron Letters 1973 1893. Reaction Mechanisms -Part (ii) Orbital Symmetry Me Me ca. I 2 B” 0- Me Me Me Me Irradiation of a-pyrones in a dilute argon matrix at low temperatures allows the i.r. spectra of the intermediate ketens to be observed (34).34 The complexity of the keten bands indicates several rotamers are formed and this is rationalized in terms of electrocyclic opening of electronically excited a-pyrone and formation of a mixture of rotamers in the process of demotion to the ground state.N-Benzoyl- enamines undergo photocyclization followed by 1,5-H shift to give trans-fused ring systems e.g. (35)+(36).35 I .5-H shift (35) 3 Cycloaddition Reactions Perturbation theory has been used to predict the preferred orientation (regio- selectivity) of concerted cycl~additions.~~ In many but not all cases 2 + 2 thermal cyclodimerizations are expected to occur in a head-to-tail manner and 2 + 2 photodimerizations in a head-to-head fashion if the reactions are concerted.34 0.L. Chapman C. L. Mclntosh and J. Pacansky J. Amer. Chem. SOC.,1973,95,244; R. G. S. Pong and J. S. Shirk ibid. p. 249. 35 I. Ninomiya T. Naito and T. Mori J.C.S. Perkin I 1973 505; 1. Ninomiya T. Naito T. Kiguchi and T. Mori ibid. p. 1696; I. Ninomiya T. Naito and T. Kiguchi ibid. pp. 2257 2261 ;G. R. Lenz Tetrahedron Letters 1973 1963. 36 N. D. Epiotis J. Amer. Chem. Soc. 1973 95 5624. 82 R.Grigg A similar treatment of 4 + 2 cycloadditions is also presented. However since the forces governing regioselection are weak the stabilization-energy difference between two distinct arrangements of the reactants in the transition state is comparatively small and steric and dipolar effects may be crucial. PMOtheory also predicts quite accurately the perispecificity of cycloadditions involving some cross-conjugated systems3' Polar cycloaddition~~~ have and cycloaddition reactions of cyclopr~penes~~ been reviewed and Kaupp has continued to demolish cherished examples of concerted photocycloadditi~ns.~~ New methods4' and techniques for generating singlet oxygen have appeared including polymer-based dye sen~itisers.~~ A detailed study of the reaction of singlet oxygen with 2,5-dimethylhexa-2,4dieneshows that both the rate of the reaction and the product distribution are solvent deper1dent.4~ Photo-oxidation of the diene in methanol at -78 "C allowed a 1,Zdioxetan (37) to be isolated.This and some solvent-incorporated products suggest that a perepoxide (38) is the precursor of (37).Related studies with adamant~lideneadarnantane~~ also favour a perepoxide intermediate. F~kui~~ has made the interesting suggestion that 2 + 2 cycloadditions of benzyne and of singlet oxygen to ethylene derivatives have a common feature. He proposes a novel approach geometry in which only one lobe of the addend ('02or benzyne) overlaps with the ethylene Ir-bond [e.g. (39) for benzyne]. This novel reaction path is clearly different from the 2 + 2 'allowed' mode and is supported by extended Huckel calculations. A detailed calculation of the keten 2 + 2 cycloaddition reaches similar con- clusion~.~~ This suggestion for benzyne additions should be contrasted with the configuration interaction treatment which suggests the 2 + 2 process may occur.' The thermal generation of electronically excited carbonyl compounds from dioxetans has been further studied and the results rationalized in terms of a concerted mechanism?' Studies on the thermal decomposition of tetramethyl- 1,Zdioxetan in degassed benzene show the reaction is autocatalytic and suggest a chain decomposition of the dioxetan molecule occurs as the direct result of interaction with its electronically excited cleavage product.48 The overwhelming majority of photochemical processes go from an initial reactant electronically excited state to a ground-state product.49 Conversion of an excited state of a 37 M.N. Paddon-Row P. L. Watson and R. N. Warrener Tetrahedron Letters 1973 1033. 38 R. R. Schmidt Angew. Chem. Internat.Edn. 1973 12 212. 39 M. L. Deem Synthesis 1972 675. O0 G. Kaupp Annalen 1973 844; Angew. Chem. Internat. Edn. 1973,12 765. K. Gollnick and G. Schade Tetrahedron Letters 1973,857. 42 D. C. Neckers A. L. Thayer and A. P. Schaap J. Amer. Chem. SOC.,1973,95 5820; J. R. Williams G. Orton and L. R. Unger Tetrahedron Letters 1973 4603. O3 N. M. Hasty and D. R. Kearns J. Amer. Chem. SOC.,1973.95 3380. O4 A. P. Schaap and G. R. Faler J. Amer. Chem. SOC.,1973,95 3381. O5 S. Inagaki and K. Fukui Bull. Chem. SOC.Japan 1973,46,2240. O6 R. Sustmann A. Ausmann and F. Vahrenholt J. Amer. Chem. SOC.,1972,944 8099. 41 N. J. Turro and P. Lechtken J. Amer. Chem. SOC.,1973,95 264. 48 P. Lechtken A. Yekta and N. J. Turro J. Amer. Chem. SOC.,1973,95 3027. 49 R.Grigg Ann. Reports (B) 1972,69 121. Reaction Mechanisms -Part (ii) Orbital Symmetry reactant into an electronically excited product molecule is exceedingly rare for molecules possessing a high number of vibrational modes. Two such processes the photochemical cleavage of tetramethyldioxetan to acetone (43 % triplet) and of naphthvalene (40)to naphthalene have now been uncovered.so Some triplet- state benzene is produced during the thermal rearrangement of certain Dewar benzene^.^ ’ Further studies have been reported on the thermal gas-phase rearrangement of bicyclo[2,l,O]pent-2-enes (41; R’ = H R2 = Me; R’ = Me RZ = H) into methylcyclopentadienes. None of the evidence requires or suppports a concerted #2 + #2 isomeri~rn.~~ In contrast the results of a kinetic study on the thermal rearrangement of hexamethylbicyclo[2,2,0]hexanes e.g.(42) -.) (43) + (a), are felt to support a concerted mechani~m.’~ The products and relative rates of cycloaddition of a series of olefins and acetylenes to dimethylketen have been reported. Steric and electronic factors accord with a concerted reaction in which the keten adds antarafacially with differing degrees of bond formation at the reaction termini in the transition state.54 The ketens (45; R = Me or Ph) undergo regiospecific intramolecular 50 N. Turro P. Lechtken A. Lyons R. R. Hantala E. Carnahan andT. J. Katz J. Amer. Chem. SOC.,1973,95,2035. 51 P. Lechtken R. Breslow A. H. Schmidt and N. J. Turro J. Amer. Chem. SOC.,1973 95 3025. 52 J.I. Brauman W. E. Farneth and M. B. D’Amore J. Amer. Chem. SOC.,1973,95,5043. 53 A. Sinnema F. van Rautwijk A. J. De Koning A. M. van Wijk and H. van Bakkum J.C.S. Chem. Comm. 1973 36.2.. 54 N. S. Isaacs and P. Stanbury J.C.S. Perkin 11 1973 166. 84 R.Grigg cycloaddition to (46) in high yield; none of the alternative product (47) could be detected.5s Studies of keten-allene cycl~additions~~ accord with a two-step non-concerted mechanism in contrast to those reported last year." Chloro-sulphonyl isocyanate adds to a-pinene to give an adduct (48)with an unrearranged carbon skeleton which supports a concerted me~hanism.'~ Photocycloaddition of powdered single mixed crystals containing 15% of (49) in (50) consistently gave a slightly optically active mixed dimer (51).5g This simple production of stable optically active samples from inactive starting materials is relevant to current hypotheses on the pre-biological origin of optical activity on the earth.A study of the photochemistry of the allenes (52; R' = H R2 = Ph; R' = Ph R2 = H) shows that the singlet state gives the .2 + .2 cycloadduct (53) as the major product and the di-lr-methane rearranged com- pounds (54; R' = H R2 = Ph; R' = Ph R2 = H) as minor products.60 S02CI I CI Ar= 0 Th Ar -c1 (49) 55 S. W. Baldwin and E. H. Page J.C.S. Chem. Comm. 1972 1337. " P. E. Brook,J. M. Harrison and K. Hunt J.C.S. Chem. Comm. 1973 733; W. G. Duncan W. Weyler and H. W. Moore Tetrahedron Letters 1973 4391. 57 R. Grigg Ann.Reports (B) 1972 69 128. 58 G.T. Furst M. A. Wachsman J. Pieroni J. G. White and E. J. Moriconi Tetrahedron 1973,29 1675. 59 A. Elgavi B. S. Green and G. M. J. Schmidt J. Amer. Chem. SOC.,1973,95 2058. 6o D. C. Lankin D. M. Chihal G. W. Griffin and N. S. Bhacca Tetrahedron Letters 1973 4009. Reaction Mechanisms -Part (ii) Orbital Symmetry 85 A general reaction of cyclopropyl-substituted allylic cations is rearrangement to cyclohexenyl and dienylic cations. Labelling of the a-cyclopropyl position with a methyl group shows a very specific skeletal change accompanies the rearrangement (55)-+(56). The reaction is best described as a concerted .2 + ,2 or .2 + ,2 cycloaddition.6' Stereospecific cycloaddition of the oxyallyl-Fe" complex (57) generated from ad-dibromoketones and Fe,(CO) to aryl-substituted olefins [(57) -+ (B)] has been observed but is thought to involve a two-step mechanism.62 (55) Ar D Fel'L H/-\H Me Me (57) Several reviews of 1,3-dipolar cycloadditions have appeared63 and further refinements of the molecular orbital model for 1,3-dipolar cycloadditions have been reported.64 Frontier orbitals are found to be of general utility in the ration- alization of relative rates and regioselectivity of 1,3-dipolar cycloadditions.Of the two regioisomeric adducts the favoured isomer will be that one in which the largest coefficients of the highest occupied and lowest unoccupied orbitals of the two addends are united. The primary I4C kinetic isotope effects in the 1,3-dipolar cycloaddition of N,a-diphenylnitrone and styrene to yield 2,3,5- triphenylisoxazolidine are consistent with a concerted mechanism65 and inconsistent with Firestone's biradical mechanism.66 The unidirectional addition of many 1,3-dipoles to both electron-rich and electron-deficient monosubstituted dipolarophiles is no longer observed when the dipolarophile is highly electron deficient in accord with prediction based on an MO treatment.67 Studies of 61 K.Rajeswari and T. S. Sorensen J. Amer. Chem. SOC. 1973 95 1239. 62 R. Noyori K. Yokoyama and Y. Hayakawa J. Amer. Chem. SOC. 1973,95 2722. 63 P. K. Kadaba Synthesis 1973 71; C. G. Stuckwisch ibid. p. 469; J. Bastide J. Hame-lin F. Texier and Y. Vo Quang Bull. SOC.chim. France 1973 2555 2871. 64 K.N. Houk J. Amer. Chem. SOC.,1972 94 8953; R. E. Duke R. W. Strozier and J. K. George ibid. 1973,95,7287; K. N. Houk,J. Sims C. R. Watts and L. J. Luskus ibid. p. 7301. 65 B. M. Benjamin and C. J. Collins J. Amer. Chem. SOC. 1973 95 6145. 66 R. A. Firestone J. Org. Chem. 1972 37 2181. 6' J. Sims and K. N. Houk J. Amer. Chem. SOC. 1973,95 5799. 86 R. Grigg the stereochemistry and regioselectivity of 1,3dipolar cycloadditions have provided information on steric and electronic factors in these processes.68 When treated with hydroxylamine or N-methylhydroxylamine in the presence of a protondonor catalyst fruns-A'("hecosteroids (59) are converted stereo- specifically and in good yield into the corresponding isoxazolidine derivatives (61) by a transannular 1,3-dipolar cycloaddition of the intermediate nitrone e.g.(60).69The naphthotriazine derivative (62) undergoes a remarkable 1,ll- dipolar cycloaddition to acetylenic esters to give acenaphthotriazines e.g. (63) after spontaneous dehydrogenation. 70 AcO N-0 Me Me I I N4Y.N-McO,C=CCO Me (I5 A+ /\ ,-MeOzC C0,Me Me Oxazaphospholes (64) react with isocyanides by a [3 + 11-cycloaddition via formation of a nitrile ylide (65) to give (66).71 Kinetic studies of the thermal R. Gree F. Tonnard and R. Carrie Tetrahedron Letters 1973 453; M. Joucla R. Gree and J. Hamelin Tetrahedron 1973 29 2315; G. Bianchi C. De Micheli R. Gandolfi P. Grunanger P. V. Finzi and 0.V. de Pava J.C.S. Perkin I 1973 1148. '' M. Lj. Mihailovic Lj.Lorenc Z. Maksimovic and J. Kalvoda Terrahedron 1973 29 2683. 70 C. W. Rees R. W. Stephenson and R. C. Storr J.C.S. Chem. Comm. 1972 1281. " K. Burger. J. Fehn and E. Muller Chem. Ber.. 1973,106 I ;K. Burger W. Thenn and E. Muller Angew. Chem. Internat. Edn. 1973 12 154. Reaction Mechanisms -Part (ii) Orbital Symmetry cycloaddition of N-phenylmaleamide to (67) suggest the dipolar intermediate (68) is involved.72 Oxycarbene intermediates (69) generated by thermolyses of lactone tosylhydrazone salts decompose by several pathways. One of these is a highly stereospecific fragmentation to olefin which is suggested to be a [4 + 23-cycloreversion [(69)+(70)].73 Conflicting views74 are still held about the perepoxide or concerted ene mechanism for formation of allylic hydroperoxides from olefins and singlet oxygen.75 The proposed ene mechanism for the selenium dioxide oxidation of olefins (71) has been supported by the trapping of the intermediate seleninic acid.76 Further studies have been reported of intramolecular ene reac-tion~’~,~* leading from acyclic precursors to 5-9 membered rings e.g.(72) -+(73; X = NMe). In cases where enolization is favourable e.g. (72; X = CH,) the ene reaction is thought to involve the enol (74) whereas when enolization is not facile a direct H-shift is more probable.77 72 S. R. Tanny and F. W. Fowler J. Amer. Chem. SOC.,1973,95 7320. 73 A. M. Foster and W. C. Agosta J. Amer. Chem. SOC.,1973,95 608. 74 A. Nickon J. B. DiGiorgio and P. J. L. Daniels J. Org.Chem. 1973 38 533; L. M. Stephenson D. E. McClure and P. K. Sysak J. Amer. Chem. SOC.,1973,95 7888. 75 R. Grigg Ann. Reporrs (B) 1972 69 133. ’‘ K. B. Sharpless and R. F. Lauer J. Amer. Chem. SOC.,1972,94 7154; D. Arigoni A. Vasella K. B. Sharpless and H. P. Jensen ibid. 1973 95 7917. 77 M. Bortolussi R. Block and J. M. Conia Tetrahedron Letters 1973 2499 4171. J. B. Lambert and J. J. Napoli J. Amer. Chem. SOC., 1973 95 294. 88 R.Grigg A number of papers dealing with competition between the ene reaction [2 + 21-cycloadditions and [4 + 21-cycloadditions have appeared." The reaction of vinyloxyboranes with carbonyl compounds appears to involve an ene mechanism [(75) -+(76)]?' The retro-ene reaction of organo-selenium compounds (77) proceeds under mild conditions and is finding numerous synthetic uses." The nitrones (78) undergo a retro-ene reactiong2 at 80-210 "C and the sulphones (79) fragment to sulphur dioxide and olefin (80) with the transfer of a carbon substit~ent.~~ RS (77) R' (79) The cycloaddition of ally1 cations to conjugated dienes has been reviewedg4 and the stereochemistry of the cycloaddition of oxyallyl to conjugated dienes has been further studied.85 Approximate frontier orbital energies and coefficients 79 P.Crews and J. Beard J. Org. Chem. 1973,38,522; V. Usieli and S. Sarel ibid. p. 1703; S. Sarel A. Felzenstein and J. Yovell J.C.S. Chem. Comm. 1973 859. T. Mukaiyama K. Inomata and M. Muraki J. Amer. Chem. SOC.,1973,95 967. " K. B. Sharpless and R.F. Lauer J. Amer. Chem. SOC.,1973 95 2697; H. J. Reich I. L. Reich and J. M. Renga ibid. p. 5813; K. B. Sharpless R. F. Lauer and A. Y. Teranishi ibid.,p. 61 37. 82 D. R. Boyd Tetrahedron Letters 1973 3467. 83 J. B. Hendrickson and R. Bergeron Tetrahedron Letters 1973 3609. 84 H. M. R. Hoffmann Angew. Chem. Internat. Edn. 1973 12 819. 85 S. Ito H. Ohtani and S. Amiya Tetrahedron Letters 1973 1737; R. Noyori Y. Baba S. Makino and H. Takaya ibid. p. 1741; C. E. Hudson and N. L. Bauld J. Amer. Chem. SOC.,1973,95 3822. Reaction Mechanisms -Part (ii) Orbital Symmetry for the HO (highest occupied) and LU (lowest unoccupied) orbitals of Diels- Alder participants allow prediction of the preferred Diels-Alder regioisomer. The larger terminal coefficient of each addend will become bonded preferentially in the transition state.For ‘normal’ Diels-Alder reactions involving n-rich diene and n-poor dienophile the ‘ortho’ regioisomer is favoured for l-substituted butadienes and the ‘para’ regioisomer with 2-substituted diene~.~~** A similar treatment of Lewis acid catalysed Diels-Alder reactions provides an MO explanation for the large rate acceleration and greatly increased regioselectivity.88 Studies of the endo-exo selectivity of various Diels-Alder reactions have appeared8’ and evidence obtained for attractive interactions between diene and halogen atoms in halogeno-dienophiles.” Cycloaddition to (81) occurs stereo- specifically in a manner such that repulsive interactions in the transition state are minimized.’* Thus the non-synchronous but concerted cycloaddition of cyclopentadiene could give rise to two transition states (82) and (83).The sole product is (84) arising from the more stable transition state (83). The Diels-Alder addition of singlet oxygen to acyclic 1,3-dienes is a major process even in the presence of allylic hydrogen^.'^ The dithenium fluoroborate (85) is the synthetic equivalent of carbon monoxide in that cycloaddition of dienes gives (86)’ which can be converted to (87) in three steps.93 A one-step 86 K. N. Houk J. Amer. Chem. SOC. 1973,95,4092. ’’ 0.Eisenstein and N. T. Anh Bull. SOC. chim. France 1973 2721 2723. 88 N. T. Anh and J. Seyden-Penne Tetrahedron 1973 29 3259; K. N. Houk and R. W. Strozier J. Amer. Chem. SOC.1973 95 4094. 89 D. W. Jones and G. Kneen J.C.S. Chem. Comm. 1973 420; D. W. Jones and R. L. Wife ibid. p. 421; D. Bellus K.v. Bredow H. Sauter and C. D. Weis Helv. Chim. Acta 1973 56 3004; G. Desimoni G. Colombo P. P. Righetti and G. Tacconi Tetrahedron 1973 29 2635; K. Imagawa K. Sisido and M. Kawanisi Bull. Chem. SOC.Japan 1973,46 2922. ’O K. Seguchi A. Sera and K. Maruyama Tetrahedron Letters 1973 1585; E. T. McBee M. J. Keogh R. P. Levek and E. P. Wesseler J. Org. Chem. 1973 38 632. ’’ C. Kt Bradsher F. H. Day A. T. McPhail and P.4. Wong J.C.S. Chem. Comm. 1973 156. 92 K. Kondo and M. Matsumoto J.C.S. Chem. Comm. 1972 1332. ’’ E. J. Corey and S. W. Walinsky J. Amer. Chem. SOC.,1972 94 8932. 90 R.Grigg condensation of thioamide aldehyde and olefin gives dihydr0-1,3-thiazines.~~ The reaction is thought to involve the thioamidoalkyl ion (88) and addition to the olefin is regiospecific and stereospecific [(88) +(89)].R2 SWS BF,-R' v R' (88) (89) The suggestion has been made9' that the photochemical formation of bicyclo-[3,1,O]hex-2-enes from hexatrienes [e.g. (90) -+(91)] formally a ,4 + .2 or .4 + .2 process proceeds from an excited state in which the centre double bond has twisted to give a bisallyl system. Closure to the cyclopropane ring then occurs to give (92) and is followed by closure to (91). The syn-benzene dioxide (93) undergoes valence tautomerism to (94) at 50 "C whereas anti-benzene dioxide is stable up to 150°C. The tris-homobenzene dioxides also exhibit considerably different thermal ~tabilities,~~ and activation parameters have been reported for systems of this type.97 Homofulvenes react with dieno- philes via a 6 + 2 cycloaddition [(95) 3(96)],'* whilst cycloheptatrienethione undergoes a rare .8 + ,2 cycloaddition with maleic anhydride.99 Paracyclo- phadiynes undergo a remarkable multicycloaddition reaction [(97) +(98)].loo 94 L.Abis and C. Giordano J.C.S. Perkin I 1973 771. " W. Dauben and M. S. Kellog J. Amer. Chem. SOC.,1972,94 8951. 96 E. Vogel H.-J. Altenbach and E. Schmidbauer Angew. Chem. Internar. Edn. 1973 12 838. " A. de Meijere D. Kaufmann and 0. Scheller Tetrahedron Letters 1973 553; H. Prinzbach and D. Stusche Helv. Chim. Acta 1971 54 755. " R. Askani and J. P. Chesick Chem.Ber. 1973 106 8. '9 T. Machiguchi M. Hoshino S. Ebine and Y. Kitahara J.C.S. Chem. Comm. 1973 196. loo T. Kaneda T. Ogawa and S. Misumi Tetrahedron Letters 1973 3373. Reaction Mechanisms -Part (ii) Orbital Symmetry NC. CN 4 Sigmatropic Reactions A study of [1,2]-shifts in carbonium ions in which either a methyl group-or the electronegative diphenylphosphinyl group (Ph,PO) can migrate shows that Ph,PO migration slightly predominates. It is suggested that it is the ability of the group that remains behind to support the positive charge rather than 'migra- tory aptitude' which is important in these reactions.'" The bicyclo[4,2,0]octenes (99; X = OAc or OSiMe,) were chosen as ideal substrates to investigate subjacent orbital control' in a thermal [1,3]- sigmatropic rearrangement [(99)+(loo)].The endo-methyl groups produced a sufficient steric blockade to the allowed .2 + .2 process (Si) to allow the 'forbidden' .2 + ,2 process (S,) to predominate. Rate ratios (S,/Si)of 12 (X= OAc) and 15 (X = OSiMe,) were observed and interpreted in terms of subjacent orbital control.' O2 Careful study of the thermal methylenecyclopropane rearrangement using the four racemic diastereoisomeric 2-cyano-3-methylethylidenecyclo-propanes allows both a Mobius transition state and subjacent orbital control to be ruled out in this case.' O3 However optically active (101 ;R = H) is thermally racemized more slowly than (101; R = D) equilibrates with (102). Using a trideuterio-derivative of (101 ; R = D) the 1'3-shift was shown to be 65 % antara-facial with respect to the allylic component.' O4 Photochemical and thermal 1'3-boron shifts have been observed (103; R = Me or Ph).'" Photochemical rearrangement of some diphenylbicyclo[3,1,O]hexenes [e.g.(104)-+ (1091 in the excited singlet state showed a preference for involvement of the internal cyclo- propyl bond.A concerted 1,3-suprafacial process was primarily utilized although lo' D. Howells and S. Warren J.C.S. Perkin 11 1973 1645. lo* J. A. Berson and R. W. Holder J. Amer. Chem. SOC.,1973,95 2037. lo' W. von E. Doering and L. Birladeanu Terrahedron 1973,29 499. lo* J. E. Baldwin and R. H. Fleming J. Amer. Chem. SOC.,1973,95 5256 5261. Io5 K. G. Hancock and J. D. Kramer J. Amer. Chem. SOC.,1973,95. 3425. 6463.92 R.Grigg a possible 1,3-antarafacial shift with inversion at the migrating centre was also observed. The cyclopropyl inversion process was designated a 1,l -sigma- tropic shift.'06 H R R\ B-NMe Me Ph Ph It is suggested that any electromeric substituent at C-3 in cis-penta-1,3-dienes should tend to favour the anti-aromatic antarafacial 1,5-H shift but the effect should be most pronounced with -E substituents particularly those with HOMOSof very high energy e.g.(106; X = NMe or 0-).Similar symmetrically placed substituents at C-2 and C-4 should have the same effect.lo7 The thermal scrambling of substituents [(107) +(108)] may occur via a 1,5-shift in the bicyclic valence tautomer. O8 X Me The thermal rearrangement of the ortho-oxyanilium ylide (109) gives (1 10) by a [1,4]-shift and (1 11) by successive [2,3]- to give (1 12) and [3,3]-~hifts."~ Reactions with deuteriated substrates allied with CIDNP studies show the rearrangement involves competing but distinct concerted and radical pair lo6 H.E. Zimmerman and G. A. Epling J. Amer. Chem. SOC.,1972 94 8749. lo' R. C. Bingham and M. J. S. Dewar J. Amer. Chem. SOC. 1972,94,9107. lo* L. A. Paquette R. H. Meisinger and R. E. Wingard J. Amer. Chem. SOC.,1972 94 9224. Io9 W. D. Ollis I. 0.Sutherland and Y.Thebtaranonth J.C.S. Chem. Comm. 1973 653 654; S. Mageswaran W. D. Ollis I. 0.Sutherland and Y. Thebtaranonth ibid. p. 651. Reaction Mechanisms -Part (ii) Orbital Symmetry processes with a contribution of 15-22% from the radical pair mechanism at 60°C.Thus the observation of a CNDIP effect does not rule out the existence of a concerted mechanism which might even be the predominant pathway. These results clearly reopen the question of the mechanistic detail of [1,2]-anionic rearrangements for which CNDIP evidence forms the basis of assignment of an exclusive radical pair mechanism. The vinylogous compounds (1 13) rearrange at 0 "C by competing [1,4]- and [4,5]-shifts.' lo Related [4,5]-shifts occur even in the more flexible (1 14) and allow some conclusions to be drawn on the geometry of the transition state.' l1 The relative rates of the [2,3]-sigma- tropic processes (1 15) +(1 16; X = NMe or S) compared with their acyclic counterparts are cited as positive evidence for bonding in the transition state consistent with a concerted mechanism.' R' R2 Comparison of substituent effects on the relative rates of Cope rearrangement of a number of 1,Sdienes suggests that hexa-1,5-diene is more or less poised between a biradical and concerted mechanism and that appropriate substitution 'lo W.D. Ollis R. Somanathan and I. 0.Sutherland J.C.S. Chem. Comm. 1973 661. ' I I T. Laird and W. D. Ollis J.C.S. Chem. Comm. 1973 658. 'I2 S. Mageswaran W. D. Ollis and I. 0. Sutherland J.C.S. Chem. Comm. 1973 656; W. D. Ollis I. 0.Sutherland and Y. Thebtaranonth ibid. p. 657. 94 R. Grigg at C-2 and C-5 by electromeric substituents should stabilize a cyclohex-1,4-ylene intermediate (1 17) and hence favour the biradical me~hanism."~ cis-Divinyl- cyclo-propane has been synthesized and shown to rearrange to cyclohepta-1,4- diene below room temperature (AG* = 84 kJmol-').' l4 Cope rearrangements in bicyclic systems have continued to attract attention especially bicyclo[6,1,0]- nonadiene systems (118; X = CH, CHMe 0 or N-C0,Et).''5 For the process (1 18) -+ (1 19) the rates decrease in the order N-C0,Et > CH > 0.0 Regiospecificity of Claisen rearrangements in polysubstituted aromatic com- pounds is affected by the presence of intramolecular hydrogen-bonding which results in energetic preference for one of two possible transition states.' l6 Thus in (120) rearrangement occurs exclusively to C-3 whereas the corresponding acetate gives products arising from rearrangement to both C-3 and C-5.The site specificity of the [3,3]-sigmatropic rearrangements in 3-allylindolenines has been studied and a prior imine-enamine tautomerism uncovered.' '' Ally1 ethers (121) thermally rearrange to (122; R2= C,H, R3= Ac) and (122; R2 = Ac R3= C,H,); the latter arises via initial Claisen rearrangement to the o-acetyl site followed by a [ 1,5]-acetyl shift in the intermediate dienone.' ' Reaction of a-bromoesters of allylic or acetylenic alcohols with zinc dust provides a useful synthesis of y,d-unsaturated acids via a [3,3]-sigmatropic rearrangement of the intermediate zinc enolate [(123) -+ (124)I.ll9 An extensive series of papers by Schmid and his co-workers'20 reports on charge-induced and charge-controlled [3,3]-sigmatropic processes.Lewis acids such as zinc chloride boron trifluoride silver salts and Brsnsted acids are shown to appreci- ably accelerate a wide range of [3,3]-sigmatropic processes. 'I' M. J. S. Dewar and L. E. Wade J. Amer. Chem. SOC. 1973,95 290. 'I4 J. M. Brown B. T.Goulding and J. J. Stofko J.C.S. Chem. Comm. 1973 319. 'I5 W. Grimme and K. Seel Angew. Chem. Internat. Edn. 1973 12 507; W. Grimme J. Amer. Chem. SOC. 1973 95 2381. 'I6 S. Marcinkiewicz Bull. Acad. polon. Sci. St+. Sci. chim. 1972 20 861; D. G. Clark L. Crombie and D. A. Whiting J.C.S. Chem. Comm. 1973 580; F. Vogtle and E. Goldschmitt Angew. Chem. Internat. Edn. 1973,12,767. 'I7 R. K. Bramley J. Caldwell and R. Grigg J.C.S. Perkin I 1973 1913; Tetrahedron Letters 1973 3207. ' C.P.Falshaw S. A. Lane and W. D. Ollis Chem. Comm. 1973,491. J. E. Baldwin and J. A. Walker J.C.S. Chem. Comm. 1973 117. lZo R. Borgulya R.Madeja P. Farhrni H.-J. Hansen H. Schmid and R. Barner Helu. Chim. Acta 1973,56 14; A. Wunderli J. Zsindely H.-J. Hansen and H. Schmid ibid. p. 989; U. Widmer J. Zsindely H.-J. Hansen and H. Schmid ibid. p. 75; M. Schmid H.-J. Hansen and H. Schmid ibid. p. 105; H.Schlossarczyk W. Sieber M. Hesse H.-J. Hansen and H. Schmid ibid. p. 875; U. Kock-Pomeranz H.-J. Hansen and H. Schmid ibid. p. 2981. Reaction Mechanisms -Part (ii) Orbital Symmetry 0 OZnBr A A. R~R~C 0 R~R~C'-o 11-Br CHR4 R3CH)$CHR4 R3CH=/ Deuterium labelling uncovered a degenerate butadienylcyclopropane re-arrangement 12' which occurs in (125; arrows) above 1 10 "C.The [ 1,3]-sulphur migration (126)* (127) occurs such that the new C-S bond is formed with retention of configuration by what appears to be a concerted mechanism.'22 The inspired coupling of a photochemical [1,16]-H shift with a 16-electron electrocyclic process used in the elegant construction of the corrin ring'23 has now been applied [(128)-+ (129)] to the synthesis of a l-hydro~ycorin.'~~ W. Grimme and W. von E. Doering Chem. Ber. 1973 106 1765. lz2 J. Kitchin and R. J. Stoodley J.C.S. Perkin I 1973 2460. lZ3 A. Eschenmoser Quart. Rev. 1970,24 366. lZ4 E. Gotschi and A. Eschenmoser Angew. Chem. Internat. Edn. 1973 12 912. 96 R.Crigg 5 Cheletropic Reactions Non-empirical MO studies of several carbonyl-nitrenes have been reported' 25 and by use of orbital correlation diagrams the direct insertion of the lowest singlet state nitrene into a-bonds is shown to be a 'forbidden' process.However an energetically favourable insertion pathway occurs when the nitrene approaches the o-bond in an orientation which aligns its lowest unoccupied orbital with one end of the bond thus avoiding the symmetry restrictions. A large amount of singlet biradical character can develop in the transition state. The photochemical cheletropic loss of sulphur dioxide from (1 30; X = SO, R' = R2 = alkyl R3 = C0,Me) is non-stereospecific and episulphone inter- mediates may be involved.'26 Contrary to a previous report,'27 the thermal loss of sulphur monoxide from (130; X = SO R' = R2 = alkyl R3 = C0,Me) is reasonably stereoselective and proceeds in high yield.However this discrepancy could well be due to the presence of methoxycarbonyl groups in one case (130; X = SO R3 = C0,Me)'26 but not in the other.'27 The less than 100% stereo-selectivity is circumstantial evidence for a biradical intermediate. '26*1,' Studies on the cheletropic loss of phosphorus derivatives e.g. [130; X = P(OEt),Me R' = Me R2 = R3 = provide support for the suggestion that the cheletropic process will involve axial-axial or equatorial-equatorial bond breaking at trigonal-bipyramidal phosphorus.' 29 Formation of high-energy silicon atoms by the nuclear recoil technique in the presence of butadiene gives rise to silylene (SIH,) consisting of 80% triplet and 20 % singlet both configura- tions add to butadiene to give (130; X = SiH, R' = R2 = R3 = H).130 The question of linear uersus non-linear cheletropic processes has been studied with (131tj133).'3' Bridge extrusion under the influence of basic hydrosulphite is rapid in (131) and (133) where a linear cheletropic process (4n + 2 case) is possible.In contrast (132) for which a non-linear cheletropic process (4n case) is predicted evolves very little gas (ca. 10%) and fails to generate a hydrocarbon fragment. Thermolysis of cyclopropyl azides generates olefins stereo-specifically.'32 NNO NNO NNO R' P. E. Alewood P. M. Kazmaier and A. Rank J. Amer. Chem. Soc. 1973 95 5466. 12' W. L. Prins and R. M. Kellog. Tetrahedron Lerrers 1973 2833. ''' D.M. Lemal and P. Chao J. Amer. Chem. SOC.,1973 95 920 922. C. D. Hall J. D. Bramblett and F. F. S. Liu J. Amer. Chem. SOC.,1972 94 9264. R. Hoffmann J. M. Howell and E. L. Mutterties J. Amer. Chem. SOC.,1972,94 3047. I3O G. P. Gennaro. Y.-Y. Su 0. F. Zeck S. H. Daniel and Y.-N. Tang J.C.S. Chem. Comm. 1973,637. I3l A. G. Anastassiou and H. Yamamoto J.C.S. Chem. Comm. 1973 840. 132 G. Szcimies and J. Harnisch J.C.S. Chem. Comm. 1973 739. Reaction Mechanisms -Part (ii) Orbital Symmetry 6 Metal Catalysis A further model for the catalysis of 'forbidden' processes has been suggested.' 33 Dramatically lower temperatures suffice for the Cope rearrangements of cis-divinylcyclobutanes to cyclo-octa-1,5-dienes and of cis,rrans-cyclodecadienes (1 34) -+ (135) in the presence of Pd".'34 Rh'-catalysed cheletropic processes have been observed [( 136)-+ (1 37)]' and Cu' catalysis of an electrocyclic process has been discovered [(138)-+ (139)].'36 (136) (1 37) Me Ph J==*=( CuCl Ph conrot' Me)=*-Ph 6Me Me Ph 13' F.D. Mango Tetrahedron Letters 1973 1509. '34 P. Heimbach and M. Molin J. Organometallic Chem. 1973 49 477 483. '35 H. C. Volger H. Hogeveen and C. F. Roobeek Rec. Trav. chim. 1973,92 1223. '36 K. Kleveland and L. Skattebol J.C.S. Chem. Comm. 1973 432.