6 Molecular Rearrangements By G. TENNANT Department of Chemistry University of Edinburgh The third volume of a series on molecular rearrangements has been published.’ 1 Aliphatic Rearrangements Anionic Rearrangements.-The Sommelet-Hauser2 v3 and Stevens2 rearrange- ments have been reviewed. Further mechanistic studies on the competing Sommelet and Stevens rearrangements of quaternary ammonium salts have been p~blished.~ In general 1,2-shifts are promoted by non-polar solvents and elevated temperatures whereas increase in base concentration favours the ortho-Sommelet rearrangement at the expense of the para-Sommelet and Stevens The formation of cleavage products in the latter reactions supports a radical-pair me~hanism.~~ A study of the effect of pressure on such rearrangements indicates that ylide formation is the rate-determining step in the ortho-Sommelet rearrange- ment whereas in the Stevens rearrangement ylide cleavage is rate-limiting.4c Pressure effects may provide an alternative means to c.i.d.n.p.for the diagnosis of the mechanisms of such intramolecular rearrangement^.^^ It has been pointed out’ that c.i.d.n.p. has been successfully detected only in those intramolecular rearrangements for which a concerted pathway is forbidden on orbital symmetry grounds. The demonstration of c.i.d.n.p. in the addition-rearrangement reactions of benzyne with tertiary amines,6 and in the rearrangements of the cyclic amini- mide (1)7 and the sulphonium ylide (2)’ provides support for radical dissociation- recombination mechanisms.Additional support for a non-concerted mechanism is adduced from the formation of both stereoisomers (4) and (5) in the Stevens rearrangement of the bis-ylide (3).* The first examples of the Stevens rearrange- ‘Mechanisms of Molecular Migrations,’ ed. B. s.Thyagarajan Interscience New York 1971 Vol. 3. A. R. Lepley and A. G. Giumanini ref. 1 p. 297. ’ G. Wittig Bull. Soc. chim. France 1971 1920. (u) S. H. Pine E. M. Munemo T. R. Phillips G. Bartolini W. D. Cotton and G. C. Andrews J. Org. Chem. 1971 36 984; (b) A. R. Lepley and A. G. Giumanini ibid. p. 1217; (c) D. A. Archer J. Chem. Soc. (0, 1971 1329. ’ H. Iwamura M. Iwamura T. Nishida M. Yoshida and J. Nakayama Tetrahedron Letters 197 1 63. ‘A. R. Lepley R. H. Becker and A.G. Giumanini J. Org. Chem. 1971 36 1222. H. P. Benecke and J. H. Wikel Tetrahedron Lerters 1971 3479. * R. H. Mitchell and V. Boekelheide Chem. Comm. 1970 1555. 241 242 G. Tennunt 0 Ph \+ ~ /S-cH.PhPhCH Me Me (3) (4) ment in an acyclic aminimide' and in an electrolytic process" have been reported. Details of the scope and mechanism of the thiomethoxymethylation of phenols by dimethylsulphoxide alone' la or in the presence of the pyridine-sulphur trioxide complex or dicyclohexylcarbodi-imide-phosphoricacid have been published. Predominant formation of ortho-thiomet hoxyme thylated products is explained in terms of the [2,3] sigmatropic rearrangement of an intermediate ylide [cf (6)-+(7)+(S)]. Ortho-blocked phenols give the corresponding para-thiomethoxymethylated phenol in a fragmentation-recombination process involving the methylmethylenesulphonium cation which can be trapped by an external nucleophile.N-Aryl-SS-dimethylsulphimidesare similarly re-arranged to ortho-thiomethoxymethylanilines in the presence of trieth~larnine.'~ The stereospecificity of a Wittig-type rearrangement has been elegantly demonstrated. In accord with a [2,3] sigmatropic process involving a doubly suprafacial transition state the chiral ether (9) undergoes 100% stereospecific rearrangement to a cis-trans mixture of the alcohols (10) and (11) on treatment E. A. Sedor Tetrahedron Letters 1971 323. lo P. E. Iversen Tetrahedron Letters 1971 55. (a) P. Claus N. Vavra and P. Schilling Monatsh.1971 102 1072; (6) P. Claus ibid. p. 913. l2 (a) J. P. Marino K. E. Pfitzner and R. A. Olofson Tetrahedron 1971 27 4181; (6) R. A. Olofson and J. P. Marino ibid. p. 4195. " P. Claus W. Vycudilik and W. Rieder Monarsh. 1971 102 1571. Molecular Rearrangements 243 Ph L Ph OH :nH BuLi ' HL 4-Me Me Me Me with butyl-1ithi~m.l~ The rare migration of a phenyl group in a Wittig rearrange- ment has been reported. ' The base-catalysed conversion of benzyloxytriethyl- silane into the corresponding a-silyl carbinol is an example of the hitherto unknown Wittig rearrangement of alkoxysilanes.' The driving force for this reversal of the well-known 'anti' Wittig rearrangement of silyl carbinols to silyl ethers is apparently the stability of the oxyanion which overrides the normal preference shown for formation of an Si-0 bond.The silyl carbinol to silyl ether transformation involves inversion at carbon and not retention as previously believed.' The [2,3] sigmatropic rearrangements of nitrogen and sulphur ylides continue to stimulate much experimental effort. A study of the effects of torsional strain on the course of the [2,3] sigmatropic rearrangements of ammonium ylides has produced the first example of a carbonyl-stabilized allylic ammonium ylide (12). Sigmatropic rearrangement of this compound is precluded by the rigid geometry of the azabicyclo[2,2,2]octane ring system which prevents efficient orbital overlap in the transition state. The greater efficiency of orbital overlap in the more flexible azabicyclo[3,3,l]nonane ring system is demonstrated by the smooth thermal rearrangement of the ylide (13) to the compound (14).18 On the basis of these findings it is suggested' * that sigmatropic rearrangements which involve considerable twisting of the n-system in the transition state (e.g.the [3,3]antara-antara type) probably follow non-concerted pathways. An ammonium ylide rearrangement which occurs with retention of configuration '' J. E. Baldwin and J. W. Patrick J. Amer. Chem. Soc. 1971 93 3556. Is D. R. Dimmel and S. B. Gharpure J. Amer. Chem. SOC., 1971 93 3991. R. West R. Lowe H. F. Stewart and A. Wright J. Amer. Chem. SOC.,1971 93 282. " M. S. Biernbaum and H. S. Mosher J. Amer. Chem. SOC.,1971,93,6221;A. G. Brook and J.D. Pascoe ibid.,p. 6224. Is S. Mageswaran W. D. Ollis I. 0.Sutherland and Y. Thebtaranonth Chem. Cnmm. 1971 1494. 244 G. Tennant at the carbanionic centre has been reported. The [2,3] sigmatropic rearrange- ment of dipolar diazene intermediates (16)is proposed to account for the oxidative formation of azo-compounds (17) from allylic hydrazines (1 5).20 Me Me Me -I Ph/N N Further examples of the [2,3] sigmatropic rearrangements of sulphonium ylides have been described.21 The isomer (19) is favoured in the novel thio- sulphinate-thiosulphoxylate equilibrium (18) (1 9)].22 The activation para- meters of the first-order rearrangements [(20)-(22)]of allylic disulphides are in accord with sequential [2,3] sigmatropic shifts in thiosulphoxide intermediates (21) which can be intercepted by triphenylph~sphine.~~ Delocalization of the l9 G.V. Kaiser C. W. Ashbrook and J. E. Baldwin f.Amer. Chem. SOC., 1971,93 2342. J. E. Baldwin J. E. Brown and G. Hofle J. Amer. Chem. SOC., 1971 93 788. 21 J. F. Biellmann and J. B. Ducep Tetrahedron Letters 1971 33; V. Rautenstrausch Hefu. Chim. Acta 1971 54 739; R. W. La Rochelle B. M. Trost and L. Krepski f. Org. Chem. 1971 36 1126. 22 J. E. Baldwin G. Hofle and S. C. Choi f.Amer. Chem. SOC.,1971 93 2810. 23 G. Hofle and J. E. Baldwin J. Amer. Chem. SOC.,1971 93 6307. 245 Molecular Rearrangements Me R R Me e' I Me (20)(R = H or Me) R Me negative charge in sulphonium ylides has been shown not to be a barrier to rapid sigmatropic rearrangement.24 The a-chlorination of dialkyl sulphoxides by sulphuryl chloride involves chlorination at sulphur rather than at carbon followed by a Pummerer-type rearrangement.2 The Pummerer rearrangements of sulphonium ylides [(23-P (24)]'1n312,13 and thioanhydrohexitol sulphoxides26 have also been reported.R' R2 /Ar.X.S -+ Ar-X-dHSR' \ CH.R~ (23)(X = 0 or NH) (24) The demonstration of e.s.r. and c.i.d.n.p. effects in the Martynoff rearrangement of nitrones implies a caged radical-pair mechanism.27 Thermodynamic parameters have been reported for thenilic acid rearrange- ments. The Hammett relationship is followed (p = +2.62-2.67).28 The quasi- Favorskii rearrangement of a-bromoketones involves inversion at the brominated carbon atom thereby providing firm support for a semibenzilic me~hanisrn.~~ Cyclobutanedione undergoes spontaneous rearrangement in aqueous solution affording 1-hydroxycyclopropane carboxylic acid.30 The first example of a base- catalysed Tiffeneau rearrangement has been rep~rted.~' Interest in the fast 24 J.E. Baldwin and W. F. Erickson Chem. Comm. 1971 359. 25 T. Durst and K. C. Tin Canad. J. Chem. 1971 49 2374. 26 J. Kuszmann P. Sohar and G. Horvath Tetrahedron 1971 27 5055. 2' D. G. Morris Chem. Comm. 1971 221 ; I. W. Jones D. A. Kerr and D. A. Wilson J. Chem. SOC.(0,1971 2595. '' G. P. Nilles and R. D. Schuetz J. Org. Chem. 1971 36 2489. 29 D. Baudry J. P. Begue and M. Charpentier-Morize Buff. Soc. chim. France 1971,1416.30 J. M. Coniaand J. M. Denis Tetrahedron Letters 1971,2845; H. G.Heine Chem. Ber. 197 1,104,2869. 31 B. J. Herold and J. J. R. P. Queiroga Angew. Chem. Internat. Edn. 1971 10 834. 246 G. Tennant reversible acyloin rearrangements of bridgehead ketols continues. In the equili- brium mixture of 3,3-dimethyl- 1-hydroxynorbornan-2-one (I-hydroxycampheni-lone) (25) and 1-hydroxyapocamphor (26) the latter isomer is favoured by a factor of two at 31 0C.32A variable-temperature n.m.r. study of the equilibrium of the anions derived from (25)and (26) gives a value of 1.1-1.3 for the equilibrium constant with AGO = -180ca1 ASo = -1.2 eu and E = 24.1 1.0 k~a1.~~ The rate of rearrangement of a-halogenoketones increases significantly with increase in the number of a-aryl groups.Rate enhancement is attributed to stabilization of a zwitterion intermediate produced by rate-determining halide release.34 The rare homo-Favorskii rearrangement is shown to occur readily in suitably constituted fi-halogenoketones. Base-catalysed rearrangement of the Me Mc M,e ,C02H dichloromethylcyclohexenone (27) yields the homo-Favorskii products (28) and (29) together with the bicyclo[3,1,0] hexenyl acids (30) and (31).35 The products (30) and (31) are rationalized in terms of the formation and subsequent semibenzilic rearrangement of a bicyclo[3,2,0]heptenone intermediate.3s The first examples of the Favorskii-type rearrangements of a-bromoketimines have been reported. The cyclopropanimine intermediates can be isolated and they 32 A.Nickon T. Nishida J. Frank and R. Muneyuki J. Org. Chrm. 1971 36 1075. 33 J. V. Paukstelis and D. N. Stevens Tetrahedron Letters 1971 3549. 34 F. G. Bordwell and R. G. Scamehorn J. Amer. Chem. Sue. 1971 93 3410. 35 E. Wenkert P. Bakuzis R. J. Baumgarten C. L. Leicht and H. P. Schenk J. Amer. Chem. Soc. I97 1.93 3208. Molecular Rearrangements 247 show the same regiospecificity of ring-opening as cyclopropanones. 36 Base-catalysed rearrangement of bicyclic sulphones (32) and subsequent extrusion of sulphur dioxide from the products (33) provides a valuable synthetic route to cyclo-octatetraenes. 37 The novel transformations (32) -+ (33) can be formulated either as bishomoconjugative Ramberg-Backlund rearrangements or as normal 1,3-eliminations followed by the bond reorganizations shown (Scheme l).37 (33) Scheme 1 The synthetic versatility of the Ramberg-Backlund rearrangement is exploited in a general synthesis of unsaturated propel lane^.^^ An unsuccessful attempt to synthesize bicyclopropylidene by Ramberg-Backlund rearrangement of a di- cyclopropyl sulphone is attributed to the instability inherent in cyclopropyl car bani on^.^^ Thiirene S-dioxides have been isolated from the Ramberg-Backlund reactions of zx-dihalogenodibenzyl s~lphones.~' Cationic Rearrangements.-Ester group migrations have been re~iewed.~ Equilibrium deuterium isotope effects provide detailed information on the mechanisms of fast degenerate Wagner-Meerwein rearrangements in acyclic carbonium ions.42 This powerful new method has been used to study rapid methyl and hydride shifts in the 2,3-dimethylb~tyl~~"~' 2,2,3-trimethylb~tyl,~~' 36 H.Quast E. Schmitt and R. Frank Angew. Chem. Internat. Edn. 1971 10 651. 37 L. A. Paquette R. E. Wingard and R. H. Meisinger J. Amer. Chem. Sac. 1971 93 1047. 38 L. A. Paquette J. C. Philips and R. E. Wingard J. Amer. Chrm. Sor. 1971 93 4516 4522. 39 L. A. Paquette and R.W. Houser J. Org. Chrm. 1971 36 1015. 4o L. A. Carpino L. V. McAdams R.H. Rynbrandt and J. W. Spiewak J. Amer. Chem. Sot... 1971 93 476; J. C. Philips J. V. Swisher D. Haidukewych and 0. Morales Chem. Comm. 197I 22. 4' R. M. Acheson Accounts Chem. Res. 1971,4 177. 42 (0)M. Saunders M. H. Jaffe and P. Vogel J. Amer. Chem. Soc.1971 93 2558; (h) M. Saunders and P. Vogel ihid. p. 2559; (r) M. Saunders and P. Vogel ihid. p. 2561. 248 G. Tennant and ~yclopentyl~~~ cations. A new degenerate rearrangement involving conver- sion into and return from a singly branched species is demonstrated by label scrambling in 2-trideuteriomethyl-2,3-dimethylbutyl It has been demonstrated that methyl participation is not important in the rearrangement of neopenty1-2,4-dinitrobenzenes~lphonate.~~ Deaminative rearrangements of primary aliphatic amines have been reviewed,44 and a series of papers con- cerning the steric course of such processes has been published.45 Rearrangements involving cyclopropylvinyl cations are attracting increasing attention. Treatment of 1 -iodovinylcyclopropane (34a) with silver acetate results mainly in solvent capture ring expansion to cyclobutene derivatives occurring only to a minor extent.46 In contrast predominant formation of hydroxycyclobutenes is observed in the solvolysis (AgN0,-CaC0,-H20 100 "C 12 h) of trans-(l-chlorovinyl)-l,2-dimethylcyclopropane.Product ratios and stereochemistry are explained in terms of vinylcyclopropyl and methylene- cyclobutyl cation intermediates4' Ring expansion of a vinylcyclopropyl cation to a cyclobutenyl cation is implicated in the conversion of l-bromovinylcyclo-propane (34b) into cyclo butanone by a smooth first-order process in aqueous trie thylamine at elevated temperature.48 The cy clopropylcar binylkyclo bu ty1 rearrangement provides a simple rationale for the key role played by presqualene alcohol pyrophosphate in squalene bio~ynthesis.~~ A similar homoallyl-cyclo- propylcarbinyl-cyclobutyl carbonium ion sequence has been demonstrated in the terpene Methanolysis of the labelled exo- and endo-bicyclo[3,1,0]- hexenyl trifluoroacetates (35) and (36)results in the sole formation of the exo-ether (37) which is exlusively deuteriated at positions 2 and 4.The absence of label scrambling to the 1 3 and 5 positions indicates that in these systems the rate of solvent capture is at least 1.6 x lo3faster than the rate of sigmatropic rearrange- ment.50 Photolysis of benzene in deuteriophosphoric acid results in specific incorporation of deuterium into the 6-endo position of the bicyclo[3,l,0]hexeny1 products.This specificity is incompatible with a mechanism involving the direct X (34) a; X = I b;X = Br 43 W. M. Schubert and W. L. Henson J. Amer. Chem. Soc. 1971 93 6299. 44 C. J. Collins Accounts Chem. Res. 1971 4 315. 45 W. Kirmse H. Arold and B. Kornrumpf Chem. Ber. 1971 104 1783; W. Kirmse and W. Gruber ibid. p. 1789 1795; W. Kirmse and H. Arold ibid. p. 1800. 46 S. A. Sherrod and R. G. Bergman J. Amer. Chem. SOC., 1971 93 1925; D. R. Kelsey and R. G. Bergman ibid. p. 1941. 47 M. Santelli and M. Bertrand Tetrahedron Letters 1971 3767. 48 T. Bassler and M. Hanack Tetrahedron Letters 1971 2171. 49 (a) E. E. van Tamelen and M. A. Schwartz J. Amer. Chem. Sac. 1971 93 1780; (b) L. J. Altman R. C. Kowerskii and H. C. Rilling ibid. p. 1782; H. C.Rilling C. D. Poulter W. W. Epstein and B. Larsen ibid. p. 1783. J. A. Berson and N. M. Hasty J. Amer. Chem. SOC.,1971,93 1549. Molecular Rearrangements OMe (37) rearrangement of benzeneonium ion to the bicyclohexenyl cation and supports the intermediacy of benzvalene in such proce~ses.~' Mechanisms proposed5 for the benzeneonium ion -+ bicyclohexenyl cation photosiomerization merit reconsideration in view of these results. Sequential sigmatropic shifts have been successfully demonstrated in the parent bicyclo[3,l,0]hexeny1 cation.52 However the activation energy required (AF* = 15 f1 kcal mol- at -90 "C) is sub- stantially higher than that found for the corresponding heptamethyl derivative (AF* = 9 kcal mol- ' at -89 0C).52A new and intriguing degenerate rearrange- ment in a carbonium ion has been reported.Treatment of the benzobicyclo- [3,2,1 Joctenone (38) with deuteriotrifluoroacetic acid results in deuterium exchange ofonly five of the six methyl groups that at the 8-anti position (labelled with an asterisk in Scheme 2) being ~naffected.~~ These results are reconciled by a stepwise (or possibly concerted) migration of the three-carbon bridge round the periphery of the five-membered ring in the protonated ketone accom- panied by stereospecific methyl shift in the same direction (Scheme 2). Deu- terium labelling demonstrates the operation of similar sequential migrations in the bicyclo[3,2,l]octenone (39).53 The 13C and 'H n.m.r. and Raman spectra of the 1,2-dimethylnorbornyl cation are in accord with a partially o-delocalized structure which is undergoing a rapid degenerate hydride shift [cf:(40)S(41)S(42)].54 Examples of hitherto unknown 2,3-endo endo methyl and hydride shifts have been observed in the deaminative rearrangements of norbornyl derivatives.These rearrangements " R. F. Childs and B. Parrington Chem. Comm. 1971 1540. " P. Vogel M. Saunders N. M. Hasty and J. A. Berson J. Amer. Chem. SOC., 1971,93 1551. 53 H. Hart and G. M. Love J. Amer. Chem. SOC.,1971,93 6264. 54 G. A. Olah J. R. DeMember C. Y. Lui and R. D. Porter J. Amer. Chem. SOC.,1971 93 1442. 55 S. Rengaraju and K. D. Berlin Tetrahedron 1971 27 2399; P. Wilder and W. C. Hsieh J. Org. Chern. 1971 36 2552. 250 G.Tennant -Me ___) *?jT \ Scheme 2 Molecular Rearrangements 251 must either involve a concerted process which bypasses the normal bicyclic carbonium ion or must occur by a fast 1,2-shift to the ‘hot’ carbonium ion produced on initial loss of nitrogen. Carbonium ion rearrangements of janusene (43)and its derivatives have been described [e.g.(43)-+ Rearrangement to bicyclo[4,3,1]-decatrienes [e.g.(47)]via the corresponding bishomotropylium cation [e.g. (46)]occurs on electrophilic addition to bicyclo[4,2,2]decatetraene (4515’The transformation “48)+(5O)Jrepresents a novel example of a Wagner- Meerwein shift in an organometallic complex. Deuterium labeiling showed that though the 1,2-shift in the cation (49)occurs predominantly by the direct pathway (a) rearrangement by pathway (b)is also in~olved.’~ Treatment of (43) (44) H Br H Br [2-14C]adamantane with aluminium bromide at 110“C results in extensive (78.4%) label scrambling.This unprecedented automerization (degenerate isomerization) of adamantane is viewed in terms of bond reorganization within the framework of the adamantyl cation.59 1,2-Hydride shifts on the adamantane skeleton are subject to an energy barrier of at least 2G30.5 kcal mol- and 5h S. J. Cristol and M. A. Imhoff J. Org. Chem. 1971,36 1849; 1854. ’’ G. Schroder U. Prange B. Putze J. Thio and J. F. M. Oth Chem. Ber. 1971 104 3406. 58 A. Eisenstadt and S. Winstein Tetrahedron Letters 1970 4603. 59 Z. Majerski S. H. Liggero P. von R. Schleyer and A.P. Wolf Chem. Comm. 1970 1596. 252 G. Tennanl consequently are strongly inhibited.60 Hydride shifts occurring in the course of the Koch-Haaf carboxylation of adamantanes have been shown to be inter- molecular processes.61 The versatility of the protoadamantane to adamantane rearrangement as a synthetic entry to otherwise inaccessible 1,2- and 2,4-di- substituted adamantanes has been demonstrated by two research groups.62 Reaction of 1-bromoadamantane with prop-2-ynyl alcohol results in ring ex- pansion to homoadamantyl methyl ketone.63 Ring expansion of the adamantane skeleton is also involved in the acid-catalysed conversion of the novel spiro- adamantane-homoadamantane (51) into a mixture of the fused bishomoadaman- tane (52) and adamantylidene adamantane (53).64 The Rupe and Meyer-Schuster rearrangements have been reviewed.65 The critique by Bordwel166 published last year has polarized the controversy sur- rounding the S,2‘ mechanism of allylic rearrangement and the criticisms have been ref~ted.~’ A major factor contributing to this problem is the lack of suitable analytical methods which are capable of distinguishing unequivocally between the various mechanistic possibilities.An analytical procedure based on isotopic labelling which is claimed to circumvent these difficulties has been described.68 Entry of hydride ion and loss of chloride ion are shown to occur synfacially in the hydride-ion-induced rearrangement of an allylic chloride. It is concluded that such processes are best designated as SNi’and that ‘the SN2’myth should no longer be perpet~ated’.~~ Ring-opening in the cyclo- propyl-ally1 chloride rearrangement is at least 95 % stereospecific recapture of 6o P.Vogel M. Saunders W. Thielecke and P. von R. Schleyer Tetrahedron Letters 1971 1429. 6* D. J. Raber R. C. Fort E. Wiskott C. W. Woodworth P. von R. Schleyer J. Weber and H. Stetter Tetrahedron 197 1 27 3. 62 D. Lenoir R. Glaser P. Mison and P. von R. Schleyer J. Org. Chem. 1971,36 1821 ; D. Lenoir P. von R. Schleyer C. A. Cupas and W. E. Heyd Chem. Comm. 1971,26; B. D. Cuddy D. Grant and M. A. McKervey ibid. p. 27. 63 J. K. Chakrabarti and A. Todd Chem. Comm. 1971 556. 64 E. Boelema H. Wynberg and J. Strating Tetrahedron Letters 1971 4029. 65 S. Swaminathan and K.V. Narayanan Chem. Ret;. 1971,71,429. 66 F. G. Bordwell Accounts Chem. Res. 1970 3 281. 67 P. B. D. de la Mare and C. A. Vernon J. Chem. SOC.(B) 1971 1699. 68 M. M. Shemyakin L. A. Neiman S. V. Zhukova Y. S. Nekrasov T. J. Pehk and E. T. Lippmaa Tetrahedron 1971 27 281 1. 69 C. W. Jefford A. Sweeney D. T. Hall and F. Delay Helv. Chim. Acta 1971,54 1691. Molecular Rearrangements the chloride ion following rapidly on its release.70 Valency isomerism of acyloxo- nium cations is described in a series of paper^.^' The acyloxonium cation(s) (54) derived from pentahydroxycyclopentanes are shown by n.m.r. to exhibit remarkable valency isomerism whereby the acyloxonium unit undulates round the five-membered ring in a ten-step cycle [i.e.(54) S(55)S(56)G (54)].'Ib Me OAc Me Me (54) (55) The first example of a 1,3-halogen shift in a chlorocarbonium ion has been re- ported.72 Rearrangements of Electron-deficient Intermediates.-The operation of the hitherto unknown thermal Wolff rearrangement of alkoxycarbonylcarbenes is implied by the formation of a product mixture containing methyl vinyl ether methyl acetate and methyl pyruvate in the gas-phase pyrolysis (>280 "C)of dimethyl diaz~malonate.~~ Photochemical Wolff rearrangement is involved in the formation74 of propargylene (58) from the diazoketone (57) and has also been employed in the first syntheses of [3,3,2]- and [4,2,2]-propellane~.~~ The ring expansion [(59)-+ (60)] represents the first example of dichlorocarbene c1 CI CHCI 'O I.Fleming and E. J. Thomas Tetrahedron Letters 1971 2485. 71 (a)H. Paulsen and H. Behne Chem. Bet-. 1971 104 1281 1311; (b) H. Paulsen and H. Behne ibid. p. 1299. 72 P. E. Petersen and W. F. Boron J. Amer. Chem. SOC.,1971,93 4076. 73 D. C. Richardson M. E. Hendrick and M. Jones J. Amer. Chem. SOC.,1971,93,3790; S. Julia H. Ledon and G. Linstrumelle Compt. rend. 1971 272 C 1898. 74 R. Selvarajan and J. H. Boyer J. Org. Chem. 1971 36 1679. 75 P. E. Eaton and K. Nyi J. Amer. Chem. SOC.,1971,93 2786. 254 G. Tennant insertion into a silicon-carbon bond.76 Flash vacuum pyrolysis (700 "C)of o- rn- and p-tolylcarbenes yields benzocyclobutene and styrene in 50 % overall yield. 3C-labelling supports a mechanism for these novel rearrangements involving interconversion of rn-and p-tolylcarbene with o-tolylcarbene by sequential bicyclo[4,l,0]heptatriene-cycloheptatrienylidene valence isomerism and subsequent cycli~ation.~ The quest continues for processes involving non-classical carbene intermediates ('foiled carbene' additions).The formation of tetracyclo[3,3,0,0,2.80,4~6]octane (64) in (63) and tricycl0[3,3,0,O,~~~]oct-2-ene the base-catalysed decomposition of endo-8-tricyclo[3,2,1,02~4]octanonetosyl-hydrazone (61) is adduced as evidence for trishomocyclopropenyl participation and hence non-classical character in endo-8-carbenatricyclo-octane(62).78 N-NHTs Migratory aptitudes in the Beckmann rearrangements of cyclic ap-unsaturated ketoximes have been and the thermal and solvolytic rearrangements of oxime-sulphur trioxide complexes have been described.8o A free-radical rather than an ionic process is favoured for the thermal rearrangement of benzo-hydroxamic chloride to phenylisocyanate.' I The mechanism of photochemical Beckmann rearrangement appears to be solvent dependent.In propan-2-01 formation of open-chain amides by cleavage of the most highly substituted bond implies a free-radical process. Conversely rearrangement in methanol is stereospecific and proceeds with retention of configuration at the migrating carbon atom supporting the concerted breakdown of an oxaziridine inter- mediate.82 Anilide oximes are rearranged in moderate yield to carbodi-imides by treatment with phosphorus oxychloride in pyridine.The migration aptitude (p-Me0 > p-Me > H > p-C1 > p-Ph > p-NO,) observed in these reactions and the absence of insertion products (benzimidazoles) indicate that aryl migra- tion is concerted with departure of the leaving Tetrazole formation '' D. Seyferth R. Damrauer S. B. Andrews and S. S. Washburne J. Amer. Chem. Soc. 197 1,93 3709. '' E. Hedaya and M. E. Kent J. Amer. Chem. Soc. 1971,93 3283. '' P. K. Freeman R. S. Raghavan and D. G. Kuper J. Amer. Chem. SOC., 1971 93 5288. 79 T. Sato H. Wakatsuka and K. Amano Tetrahedron 1971 27 5381; Y. Tamura Y. Kita and M. Terashima Chem. and Pharm. Bull. (Japan),1971 19 529. K. K. Kelly and J. S. Matthews J. Org. Chem. 1971 36 2159. " Y. H. Chiang J. Org. Chem. 1971 36 2155. 82 M. Cunningham L.S. Ng Lim and G. Just Canad. J. Chem. 1971,49 2891 ;see also H. Suginome and H. Takahashi Tetrahedron Letters 1970 51 19. 83 J. Garapon B. Sillion and J. M. Bonnier Tetrahedron Letters 1970 4905. Molecular Rearrangements in the Schmidt rearrangement of adamantanone is indicative of a stepwise process involving iminium cation intermediate^.^^ The rearrangement [(65)-+(66)]represents one of the first examples of a Schmidt reaction involving migration to oxygen rather than nitrogen.85 The nitrone to amide transforma- tion [(67) +(69)] is thoughtg6 to proceed via an intermediate hydroxylamine tosylate (68) and provides a valuable complement to Beckmann rearrangement for ring expansion of this type. N-Tosyloxy-2-azanorbornenesand norbornanes undergo novel concerted Lossen-type rearrangements to afford the corres- ponding 1-azanornornene and norbornane derivatives [e.g.(70)-+ (71)Jg7 HN &I \ / NC C0,Et v \ OTs (70) The Hofmann-Loeffler and Stieglitz rearrangements and rearrangements involving nitrenium ions have been reviewed.gg Kinetic parameters determined for the Hofmann rearrangements of N-chloro-and N-bromo-benzamides and 84 T. Sasaki S. Eguchi and K. Toru J. Org. Chem. 1971 36 2454. 85 D. H. R. Barton P. G. Sammes and G. G. Weingarten J. Chem. SOC.(0,1971 729. 8h D. H. R. Barton M. J. Day R. H. Hesse and M. M. Pechet Chem. Comm. 1971,945. " J. M. Biehler and J. P. Fleury Tetrahedron 1971 27 3171; J. Heterocyclic Chem. 1971 8 431. 88 P. Kovacic M.K. Lowery and K. W. Field Chem. Rev. 1970,70 739. 256 G. Tennant their derivatives support fully concerted mechanisms for these reaction^.^' Rearrangements involving nitrenium ions continue to attract attention. Genera- tion of singlet nitrenium ion exocyclic to a four-membered ring results in direct ring expansion to a pyrrolidine cation which can be trapped after reductive work-up [cf. (72) +(73) +(74)]. Hydrogen abstraction after spin inversion to triplet nitrenium ion competes with rearrangementg0 Conversely generation of the nitrenium ion on a four-membered ring is rapidly followed by ring con- traction to an aziridinium cation [ct (75) +(76) +(77)] whose formation is inferred from the nature of the products obtained after hydrolytic work-up in the presence of benzoyl ~hloride.~' The reverse of this ring contraction is exemplified by the silver nitrate-catalysed ring expansion of N-chloroamino- cyclopropanols to azetidin~nes.~' Full details of the aluminium chloride (73) Me catalysed nitrenium ion rearrangements of NN-dichloro-1 -aminoadamantane have been Skeletal rearrangements of nitrenium ions generated by nitrous acid deamination of N-aminoazanorbornanes have also been re- ported.93 89 T.Imamoto Y. Tsuto and Y. Yukawa Bull. Chem. SOC. (Japan) 1971 44 1632 1639 1644; (Chem. Abs. 1971 75 62 899 62 901 62 908). 90 P. G. Gassman and A. Carrasquillo Tetrahedron Letters 1971 109. 91 H. H. Wasserman H. W. Adickes and 0. Espejo de Ochoa J. Amer. Chem. SOC. 1971,93 5586. 92 P.Kovacic J. H. Liu E. M. Levi and P. D. Roskos J. Amer. Chem. SOC.,1971 93 5801. 93 P. G. Gassman and K. Shudo J. Amer. Chem. SOC.,1971,93 5899. Molecular Rearrangements 257 There is now a considerable body of evidence which supports the operation of fully concerted processes in a number of photolytic and pyrolytic azide rearrangements. Further support comes from the demonstration of non-statistical migration of alkyl and aryl groups in the photolysis of azides the least bulky group invariably undergoing preferential migration.94 This specificity is interpreted in terms of fully concerted non-nitrene processes in which the migration aptitude is determined by conformational preferences in the ground state. This interpretation is supported by the observation that rearrangement product ratios tend to the statistical value with increase in temperature a feature which militates against the involvement of a symmetrical nitrene intermediate.94 In contrast to photolysis which produces rearrangement thermolysis of the geminal diazide of dimethyl malonate results in insertion into the C=O bond.95 This clear-cut example of dichotomy in the behaviour of azides is rationalized by the proposal that the photochemical and thermal processes are respectively concerted and non-~oncerted.~' In the photochemical rearrangement of 1-azidonorbornane (78) the one- and two-carbon bridges migrate with equal facility to yield (79) and (80).The same product mixture is obtained in lower yield by thermolysis of (78) at 140-170 0C.96Nitrene and nitrenium ion intermediates can be excluded and a mechanism involving bridge migration concerted with loss of nitrogen from the solvated azide is suggested.96 The thermal decom- position of dichlorocyclopropyl azides is highly regiospecific and provides an excellent method for the synthesis of azetine~.~~ Ring expansion concerted with nitrogen loss is indicated by the negative entropy of activation (AS* = -18 eu) observed for these rearrangement^.^^ The thermal ring expansion of 2-azidoindane-l,3-dionesto azanaphthoquinones is an example of a rare acyl migration to nitr~gen.~' 0-N aryl migration has been reported in oxyni- trenes.Inhibition of ozonide formation by added aldehyde or ketone is attributed to formal reduction of the Staudinger molozonide in a Baeyer-Villiger type process [cJ(Sl)].The recognition of this phenomenon has led to a new rationale for the mechanism of ozonolysis."' An example of the rare Baeyer-Villiger oxidation of an afi-unsaturated carbonyl compound has been described. lo' Migration aptitudes have been reported for the Baeyer-Villiger rearrangements of a1dehydes,lo2 cycloalkyl ket~nes,''~ and a cage ketone.lo4 In the latter case exclusive migration of the cage structure is observed. The incidence of acid- y4 R. A. Abramovitch and E. P. Kyba J. Amer. Chem. Sac. 1971 93 1537. 95 R. M. Moriarty and P. Serridge J. Amer. Chem. Soc. 1971 93 1534. y6 J. 0. Reed and W. Lwowski J. Org. Chem. 1971,36,2864. Y'A. B. Levy and A. Hassner J. Amer.Chem. Soc. 1971 93 2051. 98 H. W. Moore and D. S. Pearce Tetrahedron Letters 1971 1621. 99 F. A. Carey and L. J. Hayes J. Amer. Chem. Soc. 1970 92 7613. loo P. R. Story J. A. Alford W. C. Ray and J. R. Burgess J. Amer. Chem. Sac. 1971 93 3044. lo' D. L. Coffen and D. G. Korzan J. Org. Chem. 1971 36 390. lo' J. Royer and M. Beugelmans-Verrier Compt. rend. 1971 272 C 1818. lo' S. A. Monti and C. K. Ward Tetrahedron Letters 1971 697. Io4 B. Zwanenburg and A. J. H. Klunder Tetrahedron Letters 1971 1717. 258 G. Tennant catalysed 1,2-shifts in alkyl hydroperoxides is governed by the strength of the acid medium. Strongly acid media promote rearrangement in accordance with preferential protonation of the hydroxyl-oxygen. 'O5 Acid-catalysed re-arrangement of tertiary cycloalk yl hydro peroxides occurs with predominant ring-bond migration; in the examples studied alkyl migration was not observed.Thermal Photochemical and Metal-catalysed Rearrangements.-The 'abnormal' Claisen rearrangement has been reviewed. lo' Aspects of concerted sigmatropic rearrangements are discussed in review articles by Dewar'08 and Fukui."' The effect of steric constraint on the transition state for Claisen rearrangement is illustrated by the thermal behaviour of the isomeric vinylfurans (82) and (83). Concerted thermal rearrangement of (82) is disallowed on steric grounds but occurs by initial isomerization to the exo-methylene compound (83) in which the requisite transition-state geometry is more readily attained.' ' The novel thermal transformation [(84)-+(86)] is formulated' '' as a bicyclic Claisen rearrangement [(84)-+(85)+(86)].The formation of a single phenolic product Me Me Me lo5 J. 0.Turner Tetrahedron Letters 1971 887. Io6 R. D. Bushick Tetrahedron Letters 1971 579. lo' H. J. Hansen ref. I p. 177. lo' M. J. S. Dewar Angew. Chem. Internat. Edn. 1971 10 761. lo9 K. Fukui Accounts Chem. Res. 1971 4 57. lo S. J. Rhoads and C. F. Brandenburg J. Amer. Chem. SOC.,1971,93 5805; S. J. Rhoads and J. M. Watson ibid. p. 5813. 'I' J. W. Hanifin and E. Cohen J. Org. Chem. 1971 36 910. Molecular Rearrangements 2 59 excludes a biradical pathway."' The rearrangements [(87)+(88)+(89)l show first-order kinetics and are relatively insensitive to radical inhibitors or to changes in solvent polarity.Cyclic concerted mechanisms are therefore proposed.' l2 The Claisen-type rearrangement of an allyl system containing three hetero-atoms has been reported.' l3 The Cope rearrangement is a classic example of a narcissistic reaction. ' l4 Full details of a kinetic study of the parent Cope rearrangement of 1,l-dideuterio- hexa-1,5-diene have been published.' ' The finely balanced interplay between steric and electronic factors in determining the preferred transition-state geo- metry for the Cope rearrangement is demonstrated by the thermal transformation of meso-3,4-diphenylhexa-1,5-dieneinto a mixture of cis,trans- and trans trans-1,6-diphenylhexa- 1,5-dienes in a combination of four-centre and six-centre pathways."6 Doubt has been cast on the antara-antara Cope mechanism pro- posed' '' for the thermal isomerization of bicyclo[3,2,0]hepta-2,6-dienes.' ' Me OCOAr Me I 'Me ee-3 0 O.COAr 0 -OH (87) (88) D JZD D An alternative mechanism' l8 involving conrotatory ring opening to a cis,-trarqcis-triene [cf (91)] and subsequent reclosure via the cis double bond to product is supported by the key role played by the cyclobutene ring in the closely related thermal rearrangement [(90) +(91)-+(92)]. The elusive sulpho-Cope rearrangement has been successfully demonstrated for allyl vinyl sulphone. ' ' Two remarkable new degenerate rearrangements of C,,H hydrocarbons have been described. The extent of label scrambling in hypostrophene (93) at 'I2 D.H. R. Barton P. D. Magnus and M. J. Pearson J. Chem. Soc. (0,1971 2231. 'I3 N. D. Heindel and M. C. Chun Tetrahedron Letters 1971 1439. L. Salem Accounts Chem. Res. 1971 4 322. 'I5 W. von E. Doering V. G. Toscano and G. H. Beasley Tetrahedron 1971 27 5299. 'I6 R. P. Lutz S. Bernal R. J. Boggio R. 0. Harris and M. W. McNicholas J. Amer. Chem. Soc. 1971 93 3985. 117 T. M'iyashi M. Nitta and T. Mukai J. Amcr. Chem. Soc. 1971. 93. 3441. J. E. Baldwin and M. S. Kaplan J. Amer. Clrrm. Soc. 1971 93 3969. 'I9 J. F. King and D. R. K. Harding Chem. Comm. 1971. 959. 260 G. Tennant 35 "C indicates fluxional behaviour which rivals that of bullvalene.'20 The cage hydrocarbon snoutene (94) undergoes deuterium scrambling at 500 "C by a formal [,2 + ,2 + .2,] process [(94)S(95)].12 A theoretical treatment (MIND0/2) correctly predicts rates and activation energies for the degenerate Cope rearrangements of bullvalene barbaralene and semibullvalene.122 The report'23 that the tautomer (96) is favoured in the equilibrium (96),(97) appears to vindicate the prediction that structures having electron-withdrawing substituents at the bridgehead are preferred in such equilibria. However it is interesting to note that in the barbaralone equilibrium a methyl group shows the same preference.' 24 Retention of configuration at the migrating centre and at both allylic centres has been established for the photochemically induced 1,3-allylic shift of a benzyl group. 12' Unexpectedly the corresponding thermal 1,3-allylic shift [(98) -+ (99)] also occurs with > 90% retention of configuration at the migrating centre.This apparent violation of orbital symmetry control is attributed to orbital perturba- tion in the highly polarized dicyanoallyl system.' 26 Two examples of thermal [ 1,5] sigmatropic acyl shifts have been reported. '27 The thermal rearrangement of 2-bromocyclobutanone ketals to cyclopropyl carboxylates occurs with inver- sion of configuration at the migrating centre demonstrating the absence of orbital symmetry control.12* The ob~ervation'~~ a linear correlation between of A L 7 hv Ph Me Ph Me NC. CN (99) (97) (98) lZo J. S.McKennis L. Brener J. S. Ward and R. Pettit J. Amer. Chem. SOC.,1971 93 4957. lZ1 L. A. Paquette and J.C. Stowell J. Amer. Chem. SOC.,1971 93 2459. lZ2 M. J. S. Dewar and W. W. Schoeller J. Amer. Chem. SOC.,1971 93 1481. 12' G.R.Crow and K. C. Ramey Tetrahedron Letters 1971 3141. J. C. Barborak S. Chari and P. von R. Schleyer J. Amer. Chem. SOC.,1971,93 5275. R. C. Cookson J. Hudec and M. Sharma Chem. Comm. 1971 107 108. lZ6 R. C. Cookson and J. E. Kemp Chem. Comm. 1971 385. I" J. A. Berson and R. G. Salomon J. Amer. Chem. SOC.,1971,93,4620; R. A. Baylouny ibid. p. 4621; D. W. Jones and G. Kneen Chem. Comm. 1971 1356. 12* J. Salaun and J. M. Conia Tetrahedron Letters 1971 4023. lZ9 J. E.Baldwin and A. H. Andrist J. Amer. Chem. SOC.,1971 93 3289. Molecular Rearrangements 261 U.V. transition energies and activation energies for the thermal bicyclo[2,1,0]- pentene4yclopentadiene rearrangement' 30 may indicate that low-lying excited states play an important role in such [,2 + ,2,] interconversions.The demonstration' ' that cis,2trans,cis-cyclonona-tetraene(CNT) (102) is probably an intermediate in the thermal reorganizations of cis-bicyclo[6,1 ,O]-nona-2,4,6-triene (100) to 8g-dihydroindenes has shed new light on the vexatious question of the mechanisms of these transformations. It is suggested that the preferred pathway for rearrangement involves initial Cope rearrangement from a folded conformation of (100) to afford bicyclo[5,2,0]nona-1,4,7-triene(101) followed by symmetry-allowed conrotatory ring-opening to the tetraene (102) and subsequent closure to trans-8,9-dihydroindene (103).Alternatively (101) undergoes symmetry-forbidden rupture of the cross-link to all-cis-(CNT) (1 04) the precursor of cis-8,9-dihydroindene (105). Where a folded conformation in (100)is not accessible [as in 9,9-disubstituted derivatives of (loo)] rearrangement takes place from an extended conformation by direct symmetry-allowed rupture of the cross-link in (100) to give ~is,~trmzs-(CNT), a known precursor of trans-8,9-dihydroindene (103).' 31 These proposals are in accord with the results of other ~tudies,'~' but it is of interest that the reversal of a Cope rearrangement of the type [(loo)+ (101)] has been The remarkable alteration in stereospecificity represented' 33b by the Rh'-catalysed rearrangement of the 9,9- dimethyl derivative of (100) to the corresponding cis-8,9-dihydroindene also remains to be clarified.The results of deuterium-labelling studies support either a concerted [,2 + ,2,] or a biradical mediated pathway for the penultimate I3O J. E. Baldwin and A. H. Andrist Chem. Comm. 1970 1561. ''I A. G. Anastassiou and R. C. Griffith J. Amer. Chem. SOC.,1971 93 3083; A. G. Anastassiou and R. C. Griffith Chem. Comm. 1971 1301. 132 J. C. Barborak T. M. Su P. von R. Schleyer G. Boche and G. Schneider J. Amer. Chem. SOC.,197 1,93 279. (a)L. A. Paquette and M. J. Epstein J. Amer. Chem. SOC., 1971,93 5936; (b)R. Grigg R. Hayes and A. Sweeney Chem. Comm. 1971 1248. 262 G. Tennant step [( 101)-+(104)] leading from (100) uia all-cis-(CNT)(104)to cis-8,9-dihydro- indene (105).'34 The Agl-catalysed rearrangements of strained o-bonded hydrocarbons have been reviewed.' The controversy surrounding the mechanism(s) of these and related transition-metal-catalysed skeletal reorganizations has stimulated much experimental effort during the past year. The recognition that these rearrange- ments formally represent metal-catalysed symmetry-disallowed thermal pro- cesses is particularly intriguing because of the implied capacity of the metal to modify the orbital symmetry requirements of the transition state. The question of the timing of these remarkable metal-catalysed transformations appears definitely to have been resolved in favour of a stepwise process (cf:refs. 136a-c and other recent work cited therein) in which the metal acting as a weak Lewis acid interacts with the o-bonded structure to give directly (or after formation of an initial complex) a discrete intermediate variously formulated as a metal- bonded carbonium ion '353'36b or as a metal-complexed-carbene-metal-bonded carbonium ion hybrid'36" [(106)++ (107)l.Rearrangement by a formal carbe- noid or carbonium ion process within such an intermediate to give the observed products is supported by results obtained in recent studies.' 36a*c However the precise role played by the metal catalyst'36a and by the attached ligands'37 in determining the electronic character of the intermediate (and hence the mode of rearrangement) remains to be clarified. Pentafluorophenylcopper 38 is an effective catalyst for strained a-bond rearrangements ; new examples of these processes have been reported for the bicyclo[2,1,0]pentane ring system.39 M-yq-yyf (106) ( 107) 2 Aromatic Rearrangements Spirodiene rearrangements were the theme of last year's Pedler Lecture. 140 The dienone-phen01,'~' ph~to-Fries,'~~ and nitramine 144 H~fmann-Martius,'~~ rearrangements have been reviewed. Other review articles deal with aspects of dienone-phenol '45 and benzidine'08 rearrangements. 134 J. E. Baldwin and A. H. Andrist J. Amer. Chem. SOC. 1971 93 4055. 135 L. A. Paquette Accounts Chem. Res. 1971 280. 13' (a) P. G. Gassman and T. J. Atkins J. Amer. Chem. Soc. 1971 93 4597; (6) L. A. Paquette R. P. Henzel and S. E. Wilson ibid. p. 2335; L. A. Paquette and S. E. Wilson ibid. p. 5934; (c) M.Sakai and S. Masamune ibid. p. 4610; M. Sakai H. H. Westberg H. Yamaguchi and S. Masamune ibid. p. 461 1. 13' P. G. Gassman G. R. Meyer and F. J. Williams Chem. Comm. 1971 842. 38 P. G. Gassman and F. J. Williams Tetrahedron Letters I97 1 1409. 13' P. G. Gassman T. J. Atkins and J. T. Lumb Tetrahedron Letters 1971 1643. 140 D. H. Hey Quart. Rev. 1971 25 483. 14' H. Hart Accounts Chem. Res. 1971 4 337. 14* D. Bellus Adv. Photochem. 1971 8 109. 143 G. F. Grillot ref. 1 p. 237. 14' W. N. White ref. I p. 109. R. M. Acheson Accounts Chem. Res. 1971 4 179. Molecular Rearrangements Methyl shifts in intermediate benzeneonium ions [e.g.(108)+(log)] account for the fragmentation of t-butyl groups observed in the course of nitration of 14' 2,4,6-tri-t-butylnitrobenzene.Deuterium and ''N-labelling studies reveal that the methyl group originates in the t-butyl group displaced. '46 Preferential migration of methyl groups is observed in the acid-catalysed rearrangements of the very crowded allyl benzene derivative (l10).'47 N.m.r. studies exclude a concerted mechanism for hydrogen migration 14' in benzeneonium ions and Me +I Me demon~trate'~~ the operation of sequential 1,2-methyl shifts in heptamethyl- benzeneonium ion rather than direct methyl transfer to non-adjacent positions. Jacobsen rearrangements induced in polymethyl-benzenes by trichloromethyl cations have been shown to occur by initial attack at a substituted carbon atom followed by a 1,2-methyl shift in the cation produced.' 50 The acid-catalysed [2,2]paracyclophane to [2,2]metaparacyclophane rearrangement [(1 1 1) -+(1 14)] involves a formal 1,2-alkyl shift in the benzeneonium ion intermediate (1 12) formed by preferential protonation at the bridgehead.' ' The demonstration' 52 that the conversion of (11 1) into (1 14) proceeds with retention of configuration is rationalized in terms of a bridged intermediate (or transition state) (113).The racemization observed in the photochemical rearrangement of the [2,2]- metaparacyclophane (1 14) to the [2,2]metacyclophane (115) is attributed to free rotation in a zwitterionic intermediate produced by cleavage of one of the benzyl-benzyl bonds.' 52 The extent of Ar '-5 rearrangement in cationic' and free-radical '54 cyclization reactions of benzene derivatives has been elegantly demonstrated by deu terium-la belling techniques.The available evidence suggests that the semibenzene rearrangement [( 116a)-+ (1 17a)l occurs by radical dissociation-recombination rather than sequential [3,3] sigmatropic allyl shifts. '' A thermally allowed [5,5] sigmatropic benzyl shift in the semibenzene (1 16b) is excluded by the formation of (1 17b) as the sole product. However rearrangement is not completely inhibited by radical traps 146 P. C. Myhre M. Beug K. S. Brown and B. Ostman J. Arner. Chern. SOC.,1971 93 3452. 14' K. H. Lai and B. Miller Tetrahedron Letters 1971 3575. A. J. Kresge Y.Chiang and S. A. Shapiro Canad. J. Chem. 1971 49 2777. 149 B. G. Derendyaev V. I. Mamatyuk and V.A. Koptyug tzvest. Akad. Nauk. S.S.S.R. Ser. khim. 1971 5 972 (Chern. Abs. 1971 75 62 855). H. Hart and J. F. Janssen J. Org. Chem. 1970 35 3637. D. T. Heffelfinger and D. J. Cram J. Amer. Chem. SOC.,1971 93 4754. lS2 M. H. Delton R. E. Gilman and D. J. Cram J. Amer. Chem. SOC.,1971 93 2329. V. R. Haddon and L. M. Jackman J. Arner. Chern. SOC.,1971 93 3832. J. C. Chottard and M. Julia Tetrahedron Letters 1971 2561. Is' B. Miller and K. H. Lai Tetrahedron Letters 1971 1617. 264 G. Tennant / and may involve at least in part an intramolecular pathway.'55 The isolation of oxepin derivatives in the dienol-benzene rearrangements of highly hindered cyclohexadienediols is explained by the valence isomerism of benzene oxide inter- mediates.' A 1,2-cyclopropyl shift in a dienone-phenol rearrangement has been reported.57 Exclusive aralkyl migration in competition with aryloxy migration is exemplified by the transformation [(118)-P (119)].158 A similar migration preference is shown in the rearrangements of the spirolactones (120) to coumaranones.l5 Tosylhydrazones of allylcyclohexadienones [e.g. (1 2 l)] -Me&: Me@ Me\ Me Me Me Me Me (1 16) (117) a; R = CH-CH b; R = Ph 156 S. Berger G. Henes and A. Rieker Tetrahedron Letters 1971 1257; A. Rieker Angew. Chem. Internat. Edn. 1971 10 425. 15' R. C. Hahn and G. W. Jones J. Amer. Chem. SOC., 1971 93 4232. 15* A. M. Choundhury K. Schofield and R. S. Ward J. Chem. SOC.(C),1970 2543. lS9 J. L. Chitwood P. G. Gott J. J. Krutak and J.C. Martin J. Org. Chem. 1971,36,2216. Molecular Rearrangements 0 0 (120) a; R = Me b;R=Ph undergo mild acid-catalysed [3,3] sigmatropic rearrangement to allylphenyl- hydrazines [e.g.(122)l. The product of a 1,2-shift was not detected.16' Fries rearrangement of 2-bromophenyl acetate affords products derived by intermolecular bromine and acyl migration. 16' This behaviour contrasts with the 'clean' rearrangement undergone by 2-chlorophenyl acetate. The formation of biradical intermediates in the photo-Fries rearrangements of phenolic car- bonates has been demonstrated by flash photolysis.'62 In support of a solvent cage radical-pair mechanism photo-Fries rearrangements of aryloxy-s-triazines are unaffected by changes in substrate concentration or by the presence of oxygen or triplet quenchers.63 Photo-Fries rearrangements have also been reported for ethyl phenyl carbonate 164aryloxya~etonitriles,~~~ and N-acyl-'66 and N-alkyl-arylamines. 165,167 The small positive p value ( + 0.426) observed for aryl migration accompanying peroxide-induced decarbonylation of fluorene derivatives [(123) -+(124)] is indicative of a high degree of free-radical character in the transition state for rearrangement.' 68 Alkali-metal-catalysed phenyl shifts in indene derivatives are claimed as the first examples of sigmatropic phenyl I6O M. Schmid H. J. Hansen and H. Schmid Helo. Chim. Acta 1971 54 937. J. A. Donnelley and J. J. Murphy J. Chem. SOC. (0,1970,2596. 16' J. S. Humphrey and R. S.Roller Mol. Photochem. 1971 3 35 (Chem. Abs. 1971 75 87 822). 163 H. Shizuka T. Kanai T. Morita Y. Ohoto and K. Matsui Tetrahedron 1971 27 4021. E. A. Caress and I. E. Rosenberg J. Org. Chem. 1971 36 769. 16' K. J. S. Arora M. K. M. Dirania and J. Hill J. Chem. SOC.(C) 1971 2865. Ih6 D. H. Hey G. H. Jones and M. J. Perkins J. Chem. SOC.(C) 1971,116;L. Munchausen 1. Ookuni and T. D. Roberts Tetrahedron Letters 1971 1917. 67 S. Naruto and 0.Yonemitsu Tetrahedron Letters 1971 2297. 16' P. N. Cote and B. M. Vittimberga J. Amer. Chem. SOC.,1971 93 276. 266 G. Tennant O R PhO R /I / h \/ Ar-C-C-Ph A \ /c=c\ R Ar R (125) (126) migrations in radical anions. '69 The corresponding thermally induced re-arrangements which show first-order rates and involve specific phenyl migration to the 2-position are thought to involve a transition state having a high degree of biradical character.' 70 The novel photochemical 1,3-aryl shift [(125) +(126)] has been reported.17' Deuterium labelling demonstrates that the aryl borate rearrangement [(127) +(128)] involves preferential phenyl migration in a radical fragmentation-recombination process.'' Further results supporting the polar transition state mechanism for the benzidine rearrangement have been published. 73 A benzidine-like rearrange- ment [(129) +(130)] of an intermediate ON-diarylhydroxylamine is proposed to account for the formation of biphenyl derivatives in the reactions of N- hydroxycarbamates with o-fluoronitro benzene. 174 Analogy with the benzidine rearrangement is further substantiated by the isolation of semidine- and di- phenyline-like by-products.'74 A comprehensive investigation' 75 of the thermal (129) H. 'H (130) L. L. Miller and R. F. Boyer J. Amer. Chem. SOC.,1971 93 646. L. L. Miller and R. F. Boyer J. Amer. Chem. Soc. 1971 93 650. 17' H. G. Heine Tetrahedron Letters 1971 1473. 17' P. J. Grisdale J. L. R. Williams M. E. Glogowski and B. E. Babb J. Org. Chem. 197 1 36 544. D. V. Banthorpe A. Copper and M. O'Sullivan J. Chem. SOC.(B) 1971 2054. T. Sheradsky and G. Salemnick Tetrahedron Letters 1971 645. 175 P. Wetzel Chem. Ber. 1971 104 808. Molecular Rearrangements rearrangements of N-nitroso-and N-nitro-diphenylamines demonstrates a close relationship to the thermal transformations of tetra-arylhydrazines '76 and provides compelling evidence for a radical fragmentation-recombination mechanism for these rearrangements.Conversely failure to detect c.i.d.n.p. in the nitramine rearrangement of N-nitroaniline is cited' as evidence against a radical mechanism. The intramolecular character of the thermal nitramine rearrangements of 1-nitropyrazoles to 3(5)-nitropyrazoles is strongly supported by the inability of such systems to initiate nitration of substrates such as anisole and by the failure of 2,3-dinitroindazole to undergo rearrangement. '77 In contrast the capacity of l-nitropyrazoles to effect the nitration of anisole under the conditions of their acid-catalysed rearrangement to 4-nitropyrazoles clearly reflects the intermolecular character of these processes.77 The photochemical rearrange- ment of "'-dimethylhydrazobenzenes leads predominantly to the formation of o-semidine derivatives. 78 The absence of o-benzidine products indicates a mechanism differing from that of the acid-catalysed benzidine rearrangement. The operation of a radical fragmentation-recombination process implied by the concomitant formation of scission products (N-methylarylamines) of the hydrazobenzene requires further substantiation. l7 Labelling studies demon- strate that the acid-catalysed rearrangement of sodium 1-naphthylsulphamate to l-aminonaphthalene-4-sulphonicacid is partly intramolecular. Bimolecular sulphonation by a multiply sulphonated intermediate is suggested as an alterna- tive to a n-complex mechanism which is considered unlikely.' 79 However the precise mechanism of this intriguing sulphamic acid rearrangement awaits the outcome of further experimentation.Rearrangements of the sulphamic acid type have also been reported for aryl-N-sulphohydroxylamines. '8o Ar0 0 \ I1 C=N.NHPh A Ph.C.NH.N/Ar / \ Ph Ph Aryl hydrazonates (13 1) undergo smooth thermal rearrangement at 100 "C to afford the hydrazides (132). It is tentatively suggested that this new type of O-+ N aryl migration involves an intramolecular process akin to that of the Chapman rearrangement. ' ' Smiles rearrangements involving nova1 intra- "' F. A. Neugebauer and H. Fischer Chem. Ber. 1971 104 886. '" J. W. A. M. Janssen and C.L. Habraken J. Org. Chem. 1971 36 3081; P. Cohen-Fernandes and C. L. Habraken J. Org. Chem. 1971 36 3084. "* €3. J. Shine and J. D. Cheng J. Org. Chem. 1971 36 2787. W. J. Spillane F. L. Scott and C. B. Goggin J. Chem. SOC.(B) 1971 2409. I8O D. Manson J. Chem. SOC.(0,1971 1508. A. F. Hegarty J. A. Kearney M. P. Cashman and F. L. Scott Chem. Comm. 1971 689. 268 G. Tennant (133) (134) molecular 0-0 [e.g. (133)+ (134)]'82 and N+N [e.g. (13.5)-(136)]'83 nucleophilic displacements have been described. Full details of the remarkable potassium anilide-catalysed isomerizations of trihalogenobenzenes (the 'base-catalysed halogen dance' !) have been pub-lished.'84 Compelling evidence is presented for a mechanism in which the key step is positive halogen transfer from a tetrahalogenobenzene intermediate (the '7-halogen' mechanism).An analogous process in a trihalogenobenzene intermediate (the '6-halogen' mechanism) may also intervene. 84 3 Heterocyclic Rearrangements Aziridine rearrangements ' heterocyclic valence isomerizations 86 and photo- chemical rearrangements '87 have been reviewed. The thermally disallowed aziridine rearrangement (137) +(138) has been reported.'88 The stimulus for this forbidden reaction is provided by relief of strain in (137) and the resonance stabilization attained in (138). Orbital symmetry Ph N.C,H, 18* M. Harfenist and E. Thom J. Org. Chem. 1971 36 1171; D. Horton and A. E. Luetzow Chem. Comm. 1971 79. lE3 N. W. Gilman P. Levitan and L. H.Sternbach Tetrahedron Letters 1970 4121; H. H. Otto ibid. p. 5189. J. F. Bunnett and C. E. Moyer J. Amer. Chem. SOC. 1971,93 1183; J. F. Bunnett and G. Scorrano ibid. p. 1190; D. J. McLennan and J. F. Bunnett ibid. p. 1198; J. F. Bunnett and I. N. Feit ibid. p. 1201. H. G. Heine ref. 1 p. 145. '" L. A. Paquette Angew. Chem. Internat. Edn. 1971 10 11. '*' A. Lablanche-Combier and M. A. Remy Bull. SOC. chim. France 1971 679; H. Wynberg Accounts Chem. Res. 1971 4 71. 18' J. W. Lown and K. Matsumoto J. Org. Chem. 1971 36 1405. Molecular Rearrangements 0-35°C tx 4 I OH (139) control has been demonstrated for the thermal ring-opening reactions of oxi- rans' 89 and oxaziridines. 190 Aziridine N-oxides [e.g. (139 ; X = NBu' R = H)]I9' and episulphoxides [e.g.(139; X = S; R = Me)]22 rearrange spontane- ously at &35 "C to afford allylic hydroxylamines [e.g. (140;X = NBu' R = H)] and unstable ally1 sulphenic acids [e.g. (140; X = S R = Me)]. In highly substi- tuted derivatives (139) fragmentation to the alkene and t-nitrosobutane or sulphur monoxide competes with rearrangement. The rate dependence on solvent polarity observed in the thermal interconversions of 3-chloro-1-azirines is interpreted in terms of an azacyclopropenyl cation or bridged chloronium ion intermediate. 192 A kinetic study of the thermal rearrangement of 2,3,4,4-tetra- methyloxeten to 1,2-dimethylpent-2-en-4-one reveals a marked rate acceleration in comparison with the thermal ring-opening of a structurally analogous cyclo- butene.l9 Similar electrocyclic ring-opening is illustrated by the thermal rearrangement of fused dioxetans to glycol diformates. 94 A number of novel heterocyclic rearrangements which derive their stimulus from the inherent reactivity of nitrene intermediates have been reported. The varied rearrangements observed in the nitrene-induced cyclizations of ortho-nitro and ortho-azido diaryl sulphides 195 and diaryl ethers,' 96 can all be attri- buted to the propensity of a zwitterionic spiran intermediate (141) to regain aromatic stability. In simple cases (141 ;R = H) this is achieved by the exclusive 1,2-shift of sulphur [cf (141; X = S R = or by competing oxygen and nitrogen 1,2-shifts [cf (141; X = 0,R = Nor is rearrangement deterred by the presence of substituents in both ortho-positions of the cyclohexadiene ring.Where the substituent is methyl proton transfer (from the methyl group) precedes the hetero-atom shift and the net result is ring expansion to a dibenzo- heterazepine. 195,196 On the other hand when the o-substituent is methoxy ultimate stabilization is achieved via an unprecedented transannular 1,4-migra- tion of a methoxy-group. 195,196 Nitrene-induced rearrangements are also observed in the reductive cyclizations of 2-nitrodiphenylamines to phenazine 189 A. Dahmen H. Hamburger R. Huisgen and V. Markowski Chem. Comm. 1971 1192. 190 J. S. Splitter T. M. Su H. Ono and M. Calvin J. Amer. Chem. Soc. 1971 93 4075. 191 J. E. Baldwin A. K. Bhatnagar S. C. Choi and T. J.Shortridge J. Amer. Chem. SOC. 1971 93,4082. 192 J. Ciabattoni and M. Cabell J. Amer. Chem. SOC.,1971 93 1482. 193 L. E. Friedrich and G. B. Schuster J. Amer. Chem. Soc. 1971,93,4602. 194 A. P. Schaap Tetrahedron Letters 1971 1757. 195 J. I. G. Cadogan and S. Kulik J. Chem. SOC.(0, 1971 2621 and references cited therein. 196 J. I. G. Cadogan and P. K. K. Lim Chem. Comm. 1971 1431. 270 G. Tennant derivatives,' 97a and in the pyrolytic cyclization of o-azidodiphenylmethanes to acridans and acridines. 197b N-Nitrenes are implicated in the novel oxidative ring-expansions of 1-and 2-aminoindazoles to benzo- 1,2,3-triazines 19' and in the oxidative rearrangements of N-aminoquinoxalones to benzo- 1,2,4-tri- azines.199 The latter reactions are particularly noteworthy in providing an elegant synthetic entry to the relatively inaccessible benzo-1,2,4-triazine ring system.The novel degenerate thermal valence isomerism exhibited by 7-acetyl- 3-methylanthranil (142)- (144) is claimed2'' as the first example of a con- certed [1,9] sigmatropic shift. R A mechanism involving intramolecular nucleophilic aromatic substitution is proposed to account for the novel base-catalysed rearrangements of N-aryloxypyridinium salts to 2-(2'-hydro~yaryl)pyridines.~~ Thermal N-oxide rearrangements [(145) -+(146)] involving the concerted suprafacial [1,4] sigma- tropic shift of alkyl and phenyl groups have been reported.202 However a 19' (a)Y. Maki T. Hosokami and M. Suzuki Tetrahedron Letters 1971 3509; (6) G.R. Cliff and G. Jones J. Chem. SOC.(0,1971 3418. 19' D. J. C. Adams S. Bradbury D. C. Horwell M. Keating C. W. Rees and R. C. Storr Chem. Comm. 1971 828. 19' B. Adger C. W. Rees A. A. Sale and R. C. Storr Chem. Comm. 1971 695. 2oo K. P. Parry and C. W. Rees Chem. Comm. 1971 833. 201 R. A. Abramovitch S. Kato and G. M. Singer J. Amer. Chem. SOC.,1971 93 3074. 202 U. Schollkopff and I. Hoppe Tetrahedron Letters 1970 4527. Molecular Rearrangements 27 1 radical-pair mechanism for the particular case of a 1,4-benzhydryl shift is sup- ported by the zero entropy of activation and c.i.d.n.p. observed. The demonstra- tion of c.i.d.n.p. in the acetic anhydride-catalysed rearrangement of 4-picoline N-oxide provides support for the oft-disputed radical-pair mechanism without excluding the simultaneous operation of an ion-pair process.203 The photo- chemical rearrangement of pyridine N-oxide to 2-formylpyrrole is shown to originate from a singlet excited state.204 Deuterium-labelling studies fully substantiate the mechanisms proposed to account for the products of the photochemical rearrangement of quinoline N-~xide.~'~ The first example of the photorearrangement of a 4H-pyran-4-one to the corresponding 2H-pyran- 2-one has been described.'06 Bridged structures [e.g.(147)] formed by the (4 + 2) cycloaddition of dieno-philes (e.g. keten acetals) to polycyclic azonia hydrocarbons followed by acid treatment are smoothly converted by warming with sodium acetate in acetic anhydride into products of the type (148).Despite the obvious driving force for formation of an aromatic system these rearrangements are remarkable because of the highly crowded nature of the products [cf (148)].207 Photochemical AcO Ac,O A R rearrangement of pentakis(pentafluoroethy1)pyridine affords the aza Dewar ben- zene (149) and the azaprismane (150) the first relatively stable examples of the valence-bond isomers of a six-membered heterocycle.208 Photorearrangement of a perfluoropyridazine is also reported to yield the diaza Dewar benzene (151).209 This result considered in conjunction with other studies210 of the perhalogenopyridazine to perhalogenopyrazine photorearrangement implies that a diaza Dewar benzene rather than a diazaprismane is the key intermediate'" in such photoisomerizations.Courses involving retro-Diels-Alder reaction followed by Cope rearrangement and subsequent extrusion of hydrogen cyanide are proposed to account for the deep-seated rearrangements of diazabasketene '03 H. Iwamura M. Iwamura T. Nishida and S. Sato J. Amer. Chem. SOC.,1970,92,7474. 'O' F. Bellamy L. G. R. Barragan and J. Streith Chem. Comm. 1971 456. 205 0. Buchardt K. B. Tomer and V. Madsen Tetrahedron Letters 1971 131 1. 206 N. Ishibe M. Odani and M. Sunami Chem. Comm. 1971 1034. '07 D. L. Fields and T. H. Regan J. Org. Chem. 1971 36 2986 2991; D. L. Fields T. H. Regan and R. E. Graves ihid. p. 2995; D. L. Fields ibid. p. 3002. 208 M. G. Barlow J. G. Dingwall and R. N. Haszeldine Chem. Comm. 1970 1580. '09 R.D. Chambers W. K. R. Musgrave and K. C. Srivastava Chem. Comm. 1971,264. 'lo D. W. McNeil M. E. Kent E. Hedaya P. F. D'Angelo and P. 0. Schissel J. Amer. Chem. SOC.,197I 93 38 17. 272 G. Tennant to azocine2" and of diazabasketene N-oxide to benzaldoxime.21 A number of new valence isomerizations involving sulphur heterocycles have been des- cri bed. * In conclusion attention is drawn to what is perhaps the year's most individual molecular rearrangement namely the remarkable transformation (152)-P (153). Readers are referred to the original paper213 for the mechanism proposed. J. P. Snyder L. Lee and D. G. Farnum J. Amer. Chem. SOC.,1971 93 3816. 'I2 M. S. Ao and E. M. Burgess J. Amer. Chem. SOC. 1971 93 5298; D. L. Coffen Y. C. Poon and M.L. Lee ibid.,p. 4627; R. M. Kellogg ibid. p. 2344; L. A. Paquette and S. Maiorana Chem. Comm. 1971 313. '13 G. Just and W. Zehetner Chem. Comm. 1971 81.