6 Molecular Rearrangements By G. TENNANT Department of Chemistry University of Edinburgh West Mains Road. Edinburgh EH9 aJ 1 Introduction The continuing importance of molecular rearrangements in mechanistic and synthetic organic chemistry is demonstrated by the multitude of publications which appeared in the literature during 1973. Unhappily lack of space permits the discussion of only some 250 of the more than 1500 relevant papers and review articles noted. Heterocyclic rearrarlgements have not been allocated a separate section this year. The originator of a now well-known class of ylide rearrangements is co-author of an excellent new monograph on molecular rearrangements.' Two authoritative articles concerned with the application of CIDNP as a mechanistic probe for rearrangement processes have been published.2 Aliphatic Rearrangements Anidc Rearrangements.-A theoretical treatment of sigmatropic rearrange- ments involving polar transition states indicates that configuration interaction can reverse the stereoselectivity of [1,2] but not of [1,4] anionic shift^.^ Ultimate formation of the stable aromatic dianion (3) provides the driving force for the hitherto unknown [1,2] anionic shift of carbon involved in the transformation of the spiro[2,7]decatrienyl anion (1) into the bicyclo[6,2,0]decatrienyl anion (2).4 ' T. S. Stevens and W. E. Watts 'Selected Molecular Rearrangements' Van Nostrand Reinhold London 1973. * A. R. Lepley in 'Chemically Induced Magnetic Polarization' ed. A. R. Lepley and G.L. Closs Wiley-Interscience New York 1973 pp. 324-375; D. Bethel1 and M.R. Brinkman Adv. Phys. Org. Chem. 1973,10 115-120. N. D. Epiotis J. Amer. Chem. SOC.,1973 95 1206. S. W. Staley G. M. Cramer and W. G. Kingsley J. Amer. Chem. SOC.,1973,95 5052. 206 Molecular Rearrangements The facility of this otherwise forbidden carbon migration is explicable in terms of a concerted pathway involving a [1,8] carbon shift in an aromatic (ten electron) transition state. The rearrangements (4) +(5) and (6)-B (7) exemplify novel [1,2] and [1,4] C +C trimethylsilyl shifts to carbanion sites.’ Formation of (7) by a direct [1,3] anionic trimethylsilyl shift in (4) was excluded and competing phenyl shifts were not observed.’ Similar [1,2] N +C trimethylsilyl shifts to \ SiMe ‘Ph H Ph (4) BuLi 1 Li + CH,SiMe, I Ph Li+ Li’ SiMez \-c-c -/ / I -\ H ‘Ph H Li’ Ph (7) benzyl carbanion centres in acetamidine derivatives have also been reported.6 The identical product mixtures obtained from the reaction of lithium benzo- phenone ketyl with hex-5-enyl iodide and the Wittig rearrangement of benzhydryl hex-5-enyl ether show similar ratios for products derived by cyclization of hex-5-enyl radicals and subsequent coupling with ketyl but divergent ratios for products of direct hex-5-enyl-ketyl coupling.These results are interpreted in terms of competing intermolecular and intramolecular radical pathways for the Wittig process.’ Further evidence has accumulated in support of non-concerted pathways for Meisenheimer rearrangements.The lack of stereoselectivity observed in the [1,2] Meisenheimer rearrangements of (2)-and (E)-N-benzhydryl nitrones is consistent with the operation of a radical-pair process.8 Likewise the low cage effect (33%) observed in the Meisenheimer rearrangement of N-benzyl-N-methyl- aniline N-oxide is in accord with a predominant non-geminate radical-pair pathway to product.’ On the other hand the highly stereospecific character of the thermal reorganizations of N-ally1 amine N-oxides demands their formula- tion as concerted [2,3] sigmatropic processes.’o9 ’’ Thus the transformation of J. J. Eisch and M.R. Tsai J. Amer. Chem. SOC.,1973,95,4065. 0. J. Scherer and G. Schnabl J. Organometallic Chem.1973 52 C18. J. F. Garst and C. D. Smith J. Amer. Chem. Soc. 1973,95,6870. * T. S. Dobashi and E. J. Grubbs J. Amer. Chem. SOC. 1973,95 5070. J. P. Lorand R. W. Grant P. A. Samuel E. M. O’Connell J. Zaro J. Pilotte and R. W. Wallace J. Org. Chem. 1973 38 1813. Y. Yamamoto J. Oda and Y. Inouye J.C.S. Chem. Comm. 1973 848. lo V. Rautenstrauch Helv. Chim. Acta 1973 56 2492. 208 G. Tennant the chiral amine N-oxide (8) into the hydroxylamine (9) proceeds with essentially complete conservation of optical activity in keeping with rearrangement via a cyclic five-membered half-chair transition state. ' Aldoxime 0-ally1 ethers H I WIY1v ' 0N-NMe undergo the reverse thermal [2,3] shift to nitrones to the complete exclusion of alternative [3,3] sigmatropic pathways.l2 In contrast hydroxylamine 0-ally1 ethers rearrange by [1,3] shifts in presumed radical-pair processes." Meisen-heimer and Stevens rearrangements at the bridgehead in homoadamantane derivatives have been reported.' The competing [1,2] (Stevens) and [2,3] (Sommelet-Hauser) rearrangements of dibenzyl sulphide carbanion have been shown to be markedly solvent- and temperature-de~endent.'~ The concomitant formation of products derived by coupling (e.g. bibenzyl) and the predominance of Stevens rearrangement at high temperatures and in non-polar media support a radical-pair mechanism for the [1,2] shift. Conversely a concerted pathway for the [2,3] shift is indicated by its prevalence at low temperatures in polar media.14 Competition between Stevens and Sommelet-Hauser rearrangements is also observed in sulphur ylides generated in situ by the reaction of aryl carbenes with dialkyl sulphides.15 The formal nitrogen counterpart of this type of process is represented by the 'one pot' reaction of an aniline with t-butyl hypochlorite in the presence of a dialkyl sulphide.This elegant new synthetic method leads in high yield by exclusive Sommelet-Hauser rearrangement in an intermediate sulphilimine ylide to the corresponding specifically ortho-alkylated aniline.'6*' ' The considerable syn- thetic potential of this novel process is demonstrated by its application to the alkylation of aminopyridines'* and in a general synthesis of indole derivatives.16 Considerable experimental effort continues to be devoted to the study of rearrangements undergone by ammonium and sulphonium ylides.Particularly noteworthy are the high-order rearrangements of a variety of ammonium ylides reported by Ollis and his co-workers. Of these pride of place goes to the remark- able high-yield (ca.60%) rearrangements of ally1 (pentadienyl) ammonium ylides l2 S. Ranganathan D. Ranganathan R. S. Sidhu and A. K. Mehrotra Tetrahedron Letters 1973. 3577. " B. L. Adarns and P. Kovacic J. Amer. Chem. SOC.,1973 95 8206. l4 J. F. Bielmann and J. L. Schmitt Tetrahedron Letters 1973 4615. l5 W. Ando M. Yamada E. Matsuzaki andT. Migita J. Org. Chem. 1972 37 3791. l6 P. G. Gassman T. J. Van Bergen and G. Gruetzmacher J. Amer.Chem. SOC.,1973 95 6508 and other recent papers cited therein. P. G. Gassman and H. R. Drewes J.C.S. Chem. Comm. 1973,488. P. G. Gassman and C. T. Huang J. Amer. Chem. SOC.,1973,95,4453. Molecular Rearrangements (10) to octatrieneamines (l2).I9 Except in two instances [i.e. (10d) and (loe)] by-products derived by [2,3] sigmatropic and radical-pair pathways did not exceed 5%. The lack of crossover observed in these rearrangements excludes R' R2 Y R3 (10) R' R2 R3 a; H b; Me Me H Ph Ph c; Me d; H Me H Ph Ph e H Me H R3 NMe mechanisms based on sequential [2,3] and [3,3] migrations and supports their formulation as concerted [4,5] sigmatropic shifts involving aromatic (ten electron) bishomoazonine transition states.l9 Interestingly ethynyl (pentadienyl) ammo- nium ylides rearrange by competing [1,2] (Stevens) and [2,3] proces~es.'~ The [2,3] shift typical of carbonyl-stabilized N-allyiammonium ylides is largely suppressed in favour of other rearrangement modes when the ylide system is part of an aromatic framework. Rearrangements of this are exemplified by the thermal transformations of the 2-oxyanilinium ylides (14) into a mixture of the ethers (15) (79%) and the phenols (16) (11 %).20 Deuterium-labelling studies2' demonstrate that the ethers (15) are the result of a single concerted [1,4] sigmatropic shift whereas the phenolic products (16) stem from consecutive b; R=Ph (15) I9 T Laird and W. D. Ollis J.C.S. Chem. Comm. 1973 658. 2o S. Mageswaran W. D.Ollis I. 0. Sutherland and Y. Thebtaranonth J.C.S. Chem. Comm. 1973,651. " W. D. Ollis I. 0.Sutherland and Y. Thebtaranonth J.C.S. Chem. Comm. 1973 653. 22 W. D. Ollis 1. 0.Sutherland and Y. Thebtaranonth J.C.S. Chem. Comm. 1973 654. 210 G. Tennant [2,3] and [3,3] ally1 migrations. However the detection of CIDNP emission and deuterium scrambling between C-1 and C-3 of the ally1 sidechain in the rearrange- ment (14a) +(15a) + (16a) reveals a concomitant though minor radical-pair pathway to products.22 These observations emphasize the need for caution when using CIDNP as a mechanism probe. In contrast to their ortho counter-parts para-oxyanilinium ylides rearrange to products derived either by consecu- tive concerted [2,3] and [3,3] shifts or by radical fragmentation-recombination pathways the outcome depending on the allylic substitution pattern.20 ortho-and para-N-Pentadienylox yanilinium ylides also contrast in their rearrangement behaviour.Whereas the paru-compounds undergo orthodox stepwise [2,3] and [3,3] sigmatropic shifts ortho-N-pentadienyloxyanilinium ylides (17) afford mixtures of ethers (18) and phenols (19) derived respectively by novel concerted [1,4] and [4,5] shifts.2J The transition state for the latter process is constrained to a geometry (20) which differs from that propo~ed'~ for the acyclic [4,5] shift R' R2 R3 (17) a; H H Me b; Me H H (19) c; H Me H [cf. (1311. The lower rates of rearrangement of six-membered cyclic allylic ammonium and sulphonium ylides are most readily interpreted in terms of concerted processes involving strained 'bicyclic' transition The apparent [2,3J sigmatropic rearrangement of N-ethynylammonium ylides is not inhibited when the ylide system is part of a bicyclic structure.This result contrasts with the inhibition of rearrangement already noted for similar bicyclic N-allyl- ammonium ylides [cf Ann. Reports (B) 1971 68 2431 and suggests a non-23 W. D. Ollis R. Somanathan and I. 0.Sutherland J.C.S. Chem. Comm. 1973 661. '' S. Mageswaran W. D. Ollis and I. 0.Sutherland J.C.S. Chem. Comm. 1973 656. Molecular Rearrangements 21 1 concerted zwitterionic pathway for rearrangement of the acetylene derivative^.^' The rearrangement of cyclic allylammonium ylides derived from pyrrolidines provides the key step in a new general route to By-unsaturated aldehydes which is claimed to be superior to that based on the corresponding rearrangements of sulphonium ylides.26 Lithioderivatives [e.g.(2111 of 1-benzyl-4vinyl-azetidin-2-ones rearrange at -78 "C by allylic participation and concomitant ring expansion yielding azepinones [e.g. (22)]. The failure of N-allyl-N-benzyl-benzamide to rearrange similarly lends weight to the suggestion that this interest- ing /3-lactam ring expansion derives its driving force from relief of ring l train.^' The creation of chirality in an allylic sulphonium ylide and its transfer to quaternary carbon by sigmatropic rearrangement has been demonstrated.28 Thus treatment of S-methyl-S,S-bis-(y,ydimethylallyl)sulphoniumfluoroborate with (R)(-)-2,2,2-trifluorophenylethoxidein (R)(-)-2,2,2-trifluorophenylethanol at -10"Cgenerates the chiral ylide (23) which spontaneously rearranges to the thioether (24) with [cx]~'= -1.45 +_ 0.12'.Despite the low optical yield these results demonstrate the suitability of allylic sulphonium ylide rearrangement as a model for the process involved in the in vivo construction of head -P head-linked polyisoprenoids.28 The high optical induction (>94 %) observed in the [2,3] sigmatropic rearrangement of adamant-1-ylallylethylsulphonium ylide is attributed to concerted rearrangement via the most conformationally stable half- chair transition state.2g Similar stereoselectivity is exhibited by the [2,3] shifts undergone by allylic sulphonium ylides derived from conformationally fixed cyclohexylidene derivative^.^' Allylic sulphonium ylide rearrangement also " W.D. Ollis 1. 0.Sutherland and Y. Thebtaranonth J.C.S. Chem. Comm. 1973,657. 26 L. N. Mander and J. V. Turner J. Org. Chem. 1973 38 2915. '' T. Durst R. Van Den Elzen and M. J. Le Belle J. Amer. Chem. SOC.,1972 94 9261. 28 B. M. Trost and W. G. Biddlecom J. Org. Chem. 1973 38 3438. 29 B. M. Trost and R. F. Hammen J. Amer. Chem. SOC.,1973 95,962. 30 G. Andrews and D. A. Evans Tetrahedron Letters 1972 5 12 1. 212 G. Tennant provides one of the key steps in an elegant general synthesis of A3-cyclopent- en one^.^' The reaction of benzyne with 1-alkynyl or allenyl sulphide anions to give terminal acetylenes is rationalized by [2,3] sigmatropic rearrangement of intermediate allenyl sulphur ylide~.~ The negative entropies of activation observed for the allyl sulphoxide-ally1 sulphenate rearrangements of enantiomeric steroidal 6fl-sulphoxides demonstrate the concerted nature of such processes.The substantially lower rate of rearrangement observed for the (R) 6fl-sulphoxide is attributed to steric crowding in the transition state and illustrates the strictly suprafacial character of the allyl sulphoxide-ally1 sulphenate transf~nnation.~~ The selenium analogue [(25) +(2611 of the thiosulphinate-thiosulphoxylate equilibrium [cf:Ann. Reports (B) 1971 68 2441 has been demonstrated and as anticipated occurs at a faster rate than the sulphur system.34 The novel [2,3] sigmatropic rearrangement of allylseleninic acids to allyl selenium esters is implicated in a mechanistic rationale for the selenium dioxide oxidation of 01efins.~’ Novel Pummerer rearrangements of intermediate sulphinylsulphonium cations are invoked to account for the products of the thermal disproportionation reactions of alkyl thiol~ulphinates.~~ Attempts to trap the oxosulphonium cation intermediates proposed for Pummerer rearrangements have been un- successf~l.~ ’ Several new versions of the Favorskii rearrangement have been reported.The ring contraction of 2-acetylcyclopentanones to 2-(2’-chlorovinyl)cyclo-butene carboxylic acids which occurs on treatment with chloral in the presence of potassium carbonate is rationalized by a mechanism involving the enolate- induced rearrangement of trichloroethylidenecyclopentanone intermediate^.^^ The precise mechanism of these interesting (though low-yield) reactions which also occur in acyclic s~bstrates,~’ awaits the outcome of further experimentation.Deuterium-labelling studies support the involvement of cyclopropanone inter- mediates in the sodium methoxide-catalysed transformation of terminal alkyne 3’ E. J. Corey and S. W. Walinsky J. Amer. Chem. SOC.,1972 94 8932. 32 L. Brandsma S. Hoff and H.D. Verkruijsse Rec. Trav. chim. 1973 92 272. 33 D. N. Jones J. Blenkinsopp A. C. F. Edmonds E. Helmy and R. J. K. Taylor J.C.S. Perkin I 1973 2602. 34 K. B. Sharpless and R. F. Lauer J. Org. Chem. 1972 37 3973. 35 K. B. Sharpless H.P. Jensen D. Arigoni and A.Vasella J. Amer. Chem. SOC.,1973 95 791 7; K. B. Sharpless and R. F. Lauer ibid. 1972 94 7154. ’‘ E. Block and J. O’Connor J. Amer. Chem. Soc. 1973,95 5048. 37 T. Durst K. C. Tin and M.J. V. Marcil Cunad. J. Chem. 1973 51 1704. 38 A. Takeda S. Tsuboi F. Sakai and M. Tanabe Tetrahedron Letters 1973 4961. 39 A. Takeda and S. Tsuboi J. Org. Chem. 1973,38 1709. Molecular Rearrangements 213 diazotates to rearranged esters.40 Major pathways for these rearrangements involving planar oxyallyl cation intermediates are excluded by the demonstration that rearrangement of a chiral diazotate occurs with predominant (88%) inver-~ion.~~ The failure of sodium methoxide to catalyse the Favorskii rearrangements of a'-chloro-a-phenylketones which occur readily in the presence of secondary amines is adduced as evidence for the intermediacy of cyclopropanimonium cations (as opposed to cyclopropanones) in the amine-catalysed transforma- tion~.~' Ketens have been eliminated as viable intermediates in certain homo- Favorskii rearrangement^.^^ Relief of ring strain accounts for the base-catalysed conversion of cyclo- propylmethyl tosylate or bromide in moderate yield into cyclob~tene.~~ The lack of rearrangement in the absence of base and the exclusion (by deuterium- labelling studies) of rearrangement (cyclopropylcarbene ring expansion) induced by initial a-elimination is adduced as evidence for the bimolecular character of these new cyclopropylmethyl ring expansion^.^^ Ring expansion of lithio halohydrin intermediates rationalize^^^ the lithium halide-catalysed rearrange- ments of oxaspiropentanes to cyclob~tanones.~~~~~ Kinetic studies suggest that the rearrangements of 2-bromo- and 2-tosyloxycyclobutanone to cyclopropane- carboxylic acid in neutral media are best explained in terms of the formation and ring contraction of the corresponding cyclobutanone hydrates.46 The reaction of hydroxycyclobutenones with amines results in ring contraction to hydroxycyclopropane carbo~amides.~~ The pattern of deuterium uptake observed in the sodium methoxide-catalysed conversion of the diol (27) in [O-2H]methanol into the rearranged diketone (30) is rationalized by a novel 'double-barrelled' cyclobutane ring contraction-ring expansion (28) -+ (29) preceded and followed by hom~ketonization.~~ The intermediacy of geminal (rather than vicinal) dichlorosulphones and the derived thiiren S-dioxides in the reactions of bis(primary alkyl) sulphones with potassium hydroxide-carbon tetrachloride to give alkenesulphonic acid salts has been established conclu- sive]~.~~ Reactions of a-chloro- and a,a-dichlorodibenzyl sulphides with tri- phenylphosphine followed by potassium t-butoxide provide useful routes to stilbene and tolane derivatives respectively.These transformations exhibit features at varislnce with their formulation as Ramberg-Backlund-type rearrange- men ts. s' 40 W. Kirmse A. Engelmann and J. Hesse Chem. Ber. 1973,106 3073; W. Kirmse and A. Engelmann ibid. p. 3086. 4' F. G. Bordwell and J.Almy J. Org. Chem. 1973 38 571. 42 R. H. Bisceglia and C. J. Cheer J.C.S. Chem. Comm. 1973 165; S. Wolff and W. C. Agosta ibid. p. 771. " W. R. Dolbier and J. H. Alonso J.C.S. Chem. Comm. 1973 394. 44 D. H. Aue M. J. Meshishnek and D. F. Shellhamer Tetrahedron Letters 1973 4799. 45 B. M. Trost and M. J. Bogdanowicz J. Amer. Chem. Soc. 1973 95 5321. 4b J. Salaun B. Garnier and J. M. Conia Tetrahedron 1973 29 2895. 47 W. Reid A. H. Schmidt and H. Medem Annalen 1973 1530. 48 R. D. Miller and D. Dolce Tetrahedron Letrers 1973 1151. 49 C. Y. Meyers L. L. Ho G. J. McCollum and J. Branca Tetrahedron Letters 1973 1843. 50 R. H. Mitchell J.C.S. Chem. Comm. 1973 955; Tetrahedron Letters 1973 4395. 214 G. Tennant H / (29) (30) Cationic Rearrangements.-Photochemical valence isomerizations in carbo- cation^,^ ’ rearrangements accompanying Koch-Haaf carboxylati~n,~~ and the application of physical methods (i.r.Raman n.m.r. and ESCA spectroscopy) to the study of carbocation rearrangements5’ have been reviewed. The current position regarding the respective roles of edge- and corner- protonated cyclopropane intermediates in secondary and tertiary alkyl cation rearrangements is discussed in an authoritative article by the Yale group.54 It has been suggested on theoretical grounds that configuration interaction in the polar transition states for cationic rearrangements can reverse the stereo- selectivity of [1,4] but not of [1,2] cationic shifts3 Retention of configuration at the migrating centre for [1,2] cationic shifts in non-sterically constrained frame- works has been demonstrated for the first time.55*56 This stereoselectivity is in accord with the orbital-symmetry prediction that a thermal cationic [1,2] shift will be allowed in the [,O + a2J mode corresponding to a ‘slither’ type transition state (31) but disallowed in the LO + ,2,] mode corresponding to the ‘pivotal’ type transition state (32).55 The flaws inherent in the traditional methods for establishing migratory preferences in [1,2] cationic shifts particularly in relation to quantitative aspects have been pointed out.” Despite considerable ingenuity 51 P.W. Cabell-Whiting and H. Hogeveen Ado. Phys. Org. Chem. 1973 10 130-144. 52 H. Hogeveen Adu. Phys. Org.Chem. 1973 10.29-52. 53 G. A. Olah Angew. Chem. Internat. Edn. 1973 12 173. 34 M. Saunders P. Vogel E. L. Hagen and J. Rosenfeld Accounts Chem. Res. 1973,6,53. 53 T. Shono K. Fujita and S. Kumai Tetrahedron Letters 1973 3123. st W. Kirmse and W. Gruber Chem. Ber. 1973 106 1365; W. Kirmse W. Gruber and J. Knist ibid. p. 1376. 37 D. Howells and S. Warren J.C.S. Perkzn ]I 1973 1645. Molecular Rearrangements the construction of a suitable reference framework for the study of competing methyl and diphenylphosphinyl shifts was not entirely successful. However the demonstrably negligible difference in migratory aptitude for methyl uersus diphenylphosphinyl tends to support the contention that it is the ability of the non-migrating group not the migrating group to stabilize a positive charge which determines migratory preferen~e.~' A minimum free energy difference of 6.9 kcal mol-' has been estimated for competing em and endo [3,2] hydride shifts in methylated norbornyl cations.'* This result demonstrates the highly stereospecific nature of such shifts and implies an activation energy for [3,2] endo hydrogen and alkyl shifts ca.3 kcal mo1-l in excess of that required for alterna- tive sequential Wagner-Meerwein and [6,2] hydride shifts. However the question of [3,2] endo-methyl shifts in norbornyl cations is still to some extent contentious. Recent evidence59 purporting to support the operation of such a shift has been rejected.60 The possible involvement of a [3,2] endo-methyl shift in the cationic sultone rearrangement (33) +(34) has been shown by deuterium-labelling studies to be unlikely.61 The circumvention of the normal [3,2] exo-methyl shift in bicyclo[2,2,l]heptyl systems forms the basis of a new synthetic route to 8-substituted camphor derivatives.62 Rearrangements associated with vinyl cations have been re~iewed.~' Solvolysis of the trans-triflates (35) in 60% aqueous ethanol results in ca.50% phenyl shift [to rearranged ketone (40)] and occurs cu. 20-40 times faster than similar solvolysis of the cis-isomers (38) in which allene rather than ketone formation 58 L. Huang K. Ranganayakulu and T. S. Sorensen J. Amer. Chem. SOC.,1973,95 1936. 59 C. W. David B. W. Everling R. J. Kilian J. B. Stothers and W. R. Vaughan J. Amer. Chem.SOC.,1973 95 1265. 6o C. J. Collins and M. H. Lietzke J. Amer. Chem. SOC.,1973 95 6842. 6' D. R. Dimmel and W. Y. Fu J. Org. Chem. 1973,38 3778 3783. 62 C. R. Eck R. W. Mills and T. Money J.C.S. Chem. Comm. 1973,911. " P. J. Stang in 'Progress in Physical Organic Chemistry' ed. A. S. Streitwieser and R. W. Taft Wiley-Interscience New York 1973 Vol. 10 p. 205. 216 G.Tennani b; R' = CD, R2 = Me \ C=C=CH, c; R' = Me R2 = CD / Me (37) I Ph R2 \/ c=c R' /\ OTf (38) (39) + R2 \ CHCOR' / Ph (41) Scheme 1 predominate^.^^ These results are interpreted in terms of divergent solvolytic pathways for the trans-and cis-triflates (Scheme 1) involving respectively the bridged vinylidene phenonium ion (39) (the unsaturated analogue of the now well documented ethylene phenonium ion) and the open cation (36).The former cation leads directly to ketone whereas the latter mainly undergoes deprotonation to allene and to a minor extent sdvolysis either directly or via (39),to ketone.64 In contrast the kinetic nature and essentially non-stereospecific character of the anisyl shift which occurs in the course of the silver-promoted acetolysis of 2,2- dianisyl- 1-phenylvinyl bromide militate against the intermediacy of a bridged cation.65 The ratio of ring expansion (to cyclobutenes and alkylidenecyclo- butanes) to substitution in 1-cyclopropylvinyl cations increases with increasing electron donation at C-2. This effect is attributed to preferential charge stabiliza- tion in the transition state leading to rearrangement.66 The direct ring contraction of a cyclobutenyl cation (43) to a cyclopropylidene cation (44)is implicated in the solvolytic rearrangement of 1-bromo-2-phenylcyclobutene(42) to cyclopropyl phenyl ketone (45).An alternative mechanism involving protonation of (42) and subsequent cat ionic cyclobutykyclopropylmethyl ring contraction is excluded by the lack of deuterium uptake in deuteriated solvents.67 64 P. J. Stang and T. E. Dueber J. Amer. Chem. SOC.,1973,953 2683 2686. 6s Z. Rappoport A. Gal and Y. Houminer Tetrahedron Letters 1973 641. 66 D. R. Kelsey and R. G. Bergman J.C.S. Chem. Comm. 1973 589. 67 J. L. Derocque F. B. Sundermann N. Youssif and M. Hanack Annufen 1973. 419 Molecular Rearrangements 217 N.m.r.studies in super-acid media continue to reveal deep-seated yet otherwise undetected carbocation reorganizations. Cyclopropylallyl cations rearrange at -95 to -25 "C in FS0,H by smooth first-order processes to give hexadienyl and cyclohexenyl cations.68 Formation of the latter is envisaged as occurring via bicyclo[3,l,0]hex-2-enylcations (49) produced by an allowed but hitherto unprecedented [,2 + .2,] or [,2 + .2,] cycloaddition of a polarized a-bond to an allyl cation [(46) -* (47)-+ (4911 (Scheme 2). The corresponding stepwise 0 + ..... . .... . .... Scheme 2 process [(46) -+(48)-+ (4911 (Scheme 2) would require the energetically unfavour- able conversion of an allyl cation (46) into a primary cyclopropylmethyl cation (48).68 Low-temperature n.m.r.studies in FS0,H have also revealed a remark- able series of rearrangements interrelating methylated cyclohexenyl and norbornyl cation^.'^ A thorough kinetic study has completely unravelled the complex series of [1,2] hydrogen and methyl shifts and skeletal bond reorganiza- tions involved.58 Acid-catalysed rearrangements of readily accessible oxaspiro- pentanes provide general high-yield routes to cyclobutanone deriva-tives.44,45.69-7 1 These formal cyclopropylmethyl-cyclobutylcation ring expan- sions occur with preferential migration of the most electron-rich cyclopropyl K. Rajeswari and T. S. Sorensen J. Amer. Chem. SOC.,1973 95 1239. 69 B. M. Trost and M. J. Bogdanowicz J. Amer. Chem. SOC.,1973,95 531 1. 70 B.M. Trost and M. J. Bogdanowicz J. Amer. Chem. Soc. 1973,95,2038. " C. R. Johnson and E. R. Janiga J. Amer. Chem. SOC.,1973,95 7692. G.Tennant 218 bond and are highly stereospecific involving inversion at the migration termin~s.~'*~~* 70 Despite the apparent analogy between the purely thermal rearrangements of oxaspiropentanes to cyclobutanone~~~ and of a 2-azaspiro- [2,2]pentane to a cycl~butanimine,~' and the spiropentane methylenecyclobutane rearrangement the heterocyclic processes occur at much lower temperatures and consequently are suggested7' to be cationic in nature. Interestingly ring-unsub- stituted 2-azaspiro[2,2]pentanes are stable to heat and under a variety of acidic conditions give ring-opened rather than ring-expanded products.73 Cationic mechanisms are also proposed for the chlorinative ring expansions of l-vinyl- cycloalkanols to cycloalkanones.74 The closely related acid-catal ysed rearrange- ments of 1-vinylcyclopropanols and pinacolic rearrangements of a variety of 1 -substituted cyclopropanols occur readily and afford cyclobutanones in high In contrast more stringent conditions (5 % H2S04 100"C 30 min) are required for the corresponding acid-catalysed ring contractions of 2-alkyl-idenecyclobutanols to 1-alkylcyclopropyl carbonyl derivative^.^^ Skeletal reorganizations associated with adamantane and its derivatives continue to attract attention. Particular effort has been devoted to the elucida- tion of the complex structural changes involved in the Lewis-acid-catalysed hydrocarbon rearrangements leading to adamantane structures.The magnitude of the problem is indicated by the ob~ervation~~ that even if only [1,2] alkyl shifts are involved there are still no fewer than 2897 pathways available for the Lewis-acid-catalysed rearrangement of a tricyclodecane to adamantane ! The report77 that a new theoretical approach (based on molecular mechanics calcula- tions) correctly predicts the low-energy pathway for the aluminium bromide- catalysed rearrangement of endu-tetrahydrocyclopentadiene to adamantane represents an important break through in this area. Protodiamantane has been shown experimentally to be the most likely penultimate intermediate in the acid-catalysed rearrangement sequence leading from tetrahydro Binor-S to diamantane.78 Treatment of 1,l'- or 2,2'-biadamantane with aluminium bromide in cyclohexane at 60"C leads to an equilibrium mixture in which the 2,2'-isomer largely predominate^.^^ The implied greater stability of a 2-substituted adaman- tane in comparison with a- 1-substituted adamantane is unprecedented but is in accord with predictions based on molecular mechanics calculations. '' The aluminium bromide-catalysed rearrangement of 2,2'-binoradamantane affords [2]diadamantane the parent of a new class of adamantane derivatives." The 72 J. K. Crandall and W. W. Conover J.C.S. Chem. Comm. 1973 33. 7J D. H. Aue R. B. Lorens and G. S. Helwig Tetrahedron Letters 1973 4795. 74 C. R. Johnson and R. W. Herr J. Org. Chem. 1973,38 3153.75 J. P. Barnier B. Garner C. Girard J. M. Denis J. Salaun and J. M. Conia Tetra-hedron Letters 1973 1747; B. M. Trost D. Keeley and M. J. Bogdanowicz J. Amer. Chem. Soc. 1973,95 3068. 76 J. P. Barnier J. M. Denis J. R. Salaun and J. M. Conia J.C.S. Chem. Comm. 1973 103. 77 E. M. Engler M. Farcasiu A. Sevin J. M. Cense and P. von R. Schleyer J. Amer. Chem. Soc. 1973,95 5769. 78 T. M. Gund and P. von R. Schleyer Tetrahedron Letters 1973 1959. 79 J. Slutsky E. M. Engler and P. von R. Schleyer J.C.S. Chem. Comm. 1973 685. W. D. Graham P. von R. Schleyer E. W. Hagaman and E. Wenkert J. Amer. Chem. SOC.,1973 95 5785. Molecular Rearrangements protoadamantane-adtane rearrangement is exemplified by the reaction of 4methyleneprotoadamantane with formic acid to give 1-methyl-2-adamantyl formate.'' The acid-catalysed conversion of the diol(50) into the ketone (51) is interpreted in terms of consecutive [1,3] and [1,2] hydride shifts.The persistence of these novel rearrangements under conditions of high dilution demonstrates their intramolecular character.82 The preference for migration of axial hydrogen and the results of deuterium-labelling studies likewise conclusively demonstrate the intramolecularity of the novel [1,3] hydride shift implicated in the rearrange- ment (52) +(S1).82 The absence of label scrambling in the [42H]homotropilium ion at elevated temperatures (65-80 "C) in acidic media (FS03Hor H2S04)83places a minimum value on the energy barrier to circwnambulatory rearrangement in the cyclo- heptatrienylmethyl cation of 26-27 kcal mol-' which agrees well with the value (37 kcal mol- ') predicted on theoretical ground^.'^ This prohibitively high energy barrier to orbital symmetry-allowed rearrangement can be attributed to the loss of aromatic stabilization which would accrue in the course of the degenerate shift.The demonstration of two new degenerate rearrangements of bicyclo[3,2,1 Ioctadienyl cations is of considerable interest in this context." Changes in the 'H n.m.r. spectrum of the nonamethylbicyclo[3,2,1 Joctadien-Zyl cation (53) in FSOJHSOzCIF at -100 to -50 "C in the course of which only the basal methyl groups at C-1 and C-5 retain their integrity are ascribed to the novel circumambulation of the one carbon bridge and its attendant bridgehead carbons round the residual five carbon framework [cf.Scheme 3; (53) (54) (55)]." On the other hand the complete scrambling of the basal methyl groups observed in specifically labelled (53) at -78 "C demonstrates the concomitant 81 B. L. Adams and P. Kovacic J.C.S. Chem. Comm. 1972 1310. 82 E. Boelema J. H. Wieringa H. Wynberg and J. Strating Tetrahedron Letters 1973 2377. 83 J. A. Berson and J. A. Jenkins J. Amer. Chem. SOC.,1972,94. 8907. 84 W. J. Hehre J. Amer. Chem. SOC.,1972 94 8908. 85 M. Kuzuya and H. Hart J. Amer. Chem. SOC.,1973 95 4096; Tetrahedron Letters 1973 3887. 220 G. Tennant operation ofa slower degenerate process involving a [1,2] methano bridge shift [cf:Scheme 3; (53)$(56) *(57)].85 8v9 8v9 4J& 3 + 4 6&' 3 2 (54) (55) "v9 8 9 (53) = 6&2 5 5 Scheme 3 As anticipated a number of studies relating to polytopal cation rearrangements have followed hot on the heels of the theoretical predictions of Stohrer and Hoffmann regarding the (CH),' energy surface [cj Annual Reports (B) 1972 69 2451.The Stohrer-Hoffmann prediction that a cation of square-pyramidal C, geometry [cf Ann. Reports (B) 1972 69 2451 is the sole stable species on the (CH),' potential energy surface is only partly substantiated by recent theoretical studies.86-88 Although supporting the stability of the C4"configura-tion CND0,86 MIND0/3,87 and ab initioB8methods are in agreement that a non-planar version of the cyclopentadienyl cation also corresponds to an energy minimum.However the three methods are at variance in their respective estimates of the relative stabilities of the C,, and cyclopentadienyl species.86-88 Experimental support for the Stohrer-Hoffmann predictions is provided by the demonstration that esters of homotetrahedran-3-01s undergo smooth solvolytic rearrangement to cyclopentene derivative^.^^ On the other hand evidence for the intermediacy of the C4, cation in these rearrangements was inconclusive in so far as deuterium-labelling studies failed to demonstrate the associated polytopal rearrangement process [(58) (59) (60)JB9 More encouraging evidence for the incidence of polytopal cations of the type (58)comes from studies of the solvolytic rearrangements of labelled substrates at low temperatures in super-acid media using 13C n.m.r.as a structure probe. Thus the pattern of 86 H. Kollmar H. 0.Smith and P. von R. Schleyer J. Amer. Chem. SOC.,1973,95 5834. M. J. S. Dewar and R. C. Haddon J. Amer. Chem. SOC.,1973,95 5836. W. J. Hehre and P. von R. Schleyer J. Amer. Chem. SOC.,1973 95 5837. 89 S. Masamune M. Sakai and H. Ona J. Amer. Chem. SOC.,1972,94. 8955. Molecular Rearrangements 22I D p. p:::.:. 7 .. ..... D-C.’.’.............. .......!:.C-H ..... .... H (59) D (60) label scrambling and 13C n.m.r. absorption observed for solutions of homo-tetrahedran-3-01,’’ endo-tricyclo[3,2,0,02~7]heptan-4-0191 and related bridged and their derivatives in S0,ClF-FS0,H or S0,ClF-SbF at -1 15 to -50 “Careplausibly interpreted in terms ofcollapse to square-pyramidal cations [c$ (SS)] or their bishomo-anal~gues.’@-~~ Similar studies of the solvolytic rearrangements of bicyclo[2,l,l]hexane and tricyclo[3,1,0,02~6]hexane derivatives provide compelling e~idence’~ for the existence of the remarkable polytopal dication of C5”symmetry (61).‘H and 13C n.m.r. studies have also c. 2+ .’+. ........ . .of ’.., c, ........... .’ ... ... . : .’..... *. ......... demonstrated the operation in FS0,H at -150 to -20“C of the remarkably facile five-fold degenerate double circumambulatory rearrangement [(62) (63) etc.] which occurs by the formal corner alkylation of a cyclopropane by an ally1 ~ation.’~ The suggestiong4that the novel polytopal cation (64)(termed the [3,5,3]armilenium cation by the authors and corresponding to the co-ordina-tion of two mutually perpendicular allylic cation units above and below the plane of a cyclopentadienyl anion) is an intermediate in the process [(62)* (63) etc.] awaits further substantiation.Exclusive ortho-migration of the arene sulphonyloxy-group is observed in the spontaneous rearrangements of arenesulphonyloxy-N-phenylbenzohydrox-amic acids in aprotic media whereas in methanol the para-rearrangement 90 S. Masamune M. Sakai H. Ona and A. J. Jones J. Amer. Chem. SOC.,1972,94 8956. 91 S. Masamune M. Sakai A. V. Kemp-Jones H. Ona A. Vernot andT. Nakashima Angew. Chem. Internat. Edn. 1973 12 769. 92 H. Hart and M. Kuzuya J.Amer. Chem. SOC.,1972 94 8958; Tetrahedron Letters 1973,4123. 93 H. Hogeveen and P. W. Kwant Tetrahedron Letters 1973 1665; H. T. Jonkman and W. C. Nieuwpoort ibid. p. 1671; H. Hogeveen and P. W. Kwant ibid. p. 3747. 94 M. J. Goldstein and S. A. Kline J. Amer. Chem. Sor. 1973 95 935. 222 G.Tennant H H product predominates.” The lack of crossover and the results of ‘80-labelling experiments are interpreted in terms of a caged phenylnitrenium-sulphonate ion pair mechanism for the ortho-rearrangement.” Nitrenium cation inter- mediates are also postulated to account for the skeletal reorganizations of N-aroyloxy- and N-arenesulphonyloxy-azabornanederivatives?6 and in the ring expansion of a hydroxycyclopropylhydroxylaminetosylate to a #?-lactam.” The formally analogous oxidative ring expansions of hydroxycyclopropyl toluenesulphonylhydrazidesto #?-lactams have also been rep~rted.’~ a-Peroxyamine-based rearrangements have been re~iewed.~’ The extent of alkyl migration in the Baeyer-Villiger oxidations (by peroxytrifluoroacetic and peroxymaleic acids) of simple ketones increases with increasing bulk of the alkyl group to a value of 13for neopentyl versus methyl.’ O0 Baeyer-Villiger rearrange- ments of cyclobutanones to y-butyrolactones occur with oxidizing agents (basic hydrogen peroxide sodium hypobromite) which are normally incapable of effecting such transformations in larger cyclic ketone^.^^'^' Ring expansions of this type show the same stereospecificity (retention of configuration at the migrating centre) and migratory preferences (tertiary > secondary > primary alkyl) as acid-catalysed Baeyer-Villiger processes and owe their facility to relief of strain in the four-membered ring.44*45 The controversial Story mechanism 9s D.Gutschke and A. Heesing Chem. Ber. 1973 106 2379. q6 P. G. Gassman and G. D. Hartman J. Amer. Chem. SOC.,1973,95,449. ’’ H. H. Wasserman. E. A. Glazer and M. J. Hearn Tetrahedron Letters 1973,4855. 98 F. D. Greene R. L. Camp V. P. Abegg and G. 0.Pierson Tetrahedron Letters 1973 409 1. q9 E. G. E. Hawkins Angew. Chem. Internat. Edn. 1973 12 783. loo M. A. Winnick and V. Stoute Canad. J. Chem. 1973,51 2788. Molecular Rearrangements for ozonolysis [cf:Ann. Reports (B) 1971 68 257; 1972 69 2481 now appears to be untenable in the light of recent critical studies.The key role attributed to the Staudinger molozonide by Story is refuted by the demonstration"' that a primary (trioxolan) ozonide can readily effect the Baeyer-Villiger oxidation of propionaldehyde to propionic acid and by the lack of firm evidence for the formation of dioxetan by-products.' O'J O2 In addition products (e.g.6-hexano-lide) supposedly produced by Baeyer-Villiger oxidation of added carbonyl compounds by the molozonide are shown not to be primary but to arise by subsequent peroxide decomposition.' 02*'O3 Carbene Nitrene and Related Rearrangements.-Rearrangement processes associated with carbenes are included in a new textbook'04 and in a survey"' of the reactions of atomic carbon.The products of the thermal and photochemical decomposition of 1,2-diphenyl- 1-diazopropane stem from competing hydrogen and phenyl shifts originating in the singlet and triplet states respectively of a common carbene intermediate. '06 The formation of the same product mixture in the photolysis or thermolysis of 2,2-diphenyl-l-diazopropaneis explained by a [1,2] phenyl shift in (1-phenethy1)- phenylcarbene to a transient intermediate which is tentatively formulated as the 'phantom' singlet state of cr-methyl~tilbene.'~~ The demonstration'06 that c1 % methyl migration occurs in the thermal and direct photochemical decomposition of 1-([4'-2H]phenyl)-2-phenyldiazopropane is in accord with the aptitude H >Ph >Me for migration to a carbene centre.A carbene pathway for the [1,2] methyl shift observed in the photolysis of ethyl trimethylsilyldiazoacetate is supported by the demonstration that carbonium ion and silacyclopropane intermediates are not inv~lved.'~' Ring expansion of bridgehead carbenes generated in the gas phase provides a potentially valuable method for the synthesis of bridgehead olefins (notably Bredt violators). Thus the thermolysis of norbornane- and adamantane- 1-carboxaldehyde tosylhydrazone salts affords products which require the intermediacy of the Bredt violators bicyclo[2,2,2]- oct-1-ene and homoadamant-3-ene respecti~ely.'~~~'~~ Thermolysis of 4,4-dimethyl- and 4,4-diethyl-cyclohexadienylideneat 380 "Cin the gas phase gives moderate yields (35-40 %)of the alkyl-shift products p-xylene and 1,4-diethyl- benzene respectively.' The co-formation of monoalkylbenzenes and the P.S. Bailey T. P. Carter C. M. Fisher and J. A. Thompson Canad. J. Chem. 1973 51 1278. lo' K. R. Kopecky P. A. Lockwood J. E. Eilby and R. W. Reid Cunad. J. Chem. 1973 51 468. '03 D. R. Kerur and D. G. M. Diaper Canad. J. Chem. 1973,51 31 10. Io4 'Carbenes' Vol. 1 ed. M. Jones jun. and R. A. Moss Wiley-Interscience New York 1973. Io5 P. S. Skell J. J. Havel and M. J. McGlinchey Accounts Chem. Res. 1973 6 97. Io6 M. Pomerantz and T. W. Witherup J. Amer. Chem. SOC. 1973,95 5977. lo' W. Ando T. Hagiwara and T. Migita J. Amer. Chem. SOC. 1973 95 7518. lo' A. D. Wolf and M. Jones jun. J. Amer. Chem. SOC. 1973,95 8209. M. Farcasiu D.Farcasiu R. T. Conlin M. Jones jun. and P. von R. Schleyer J. Amer. Chem. SOC. 1973,95 8207. 'IoT. E. Berdick R. H. Levin A. D. Wolf and M. Jones jun. J. Amer. Chem. SOC. 1973 95. 5087. 224 G. Tennant demonstration of crossover in these rearrangements support their formulation as carbene-radical fragmentation-recombination processes. The extension of such rearrangements to spirocyclic cyclohexadienylidenes provides an elegant route to small-ring paracyclophanes (e.g. [7]paracyclophane).' ' New examples of carbene<arbene rearrangements have been reported. The formation of ring- expanded dimeric products in the solution-phase (135 "C,diglyme) thermolysis of a methanoannulene carboxaldehyde tosylhydrazone salt is rationalized by the novel arylcarbene-arylcarbene transformation [(65)-+ (66)] cited as the first example of such a process under non gas-phase The rearrange- ment of acenaphthylcarbene to phenalenylidene (the first example of a five- to a six-membered arylcarbene ring expansion) is implicated in the formation of peropyrene and phenalene by gas-phase thermolysis of the sodium salt of acenaphthylene-1-carboxaldehyde tosylhydrazone.Several studies provide unequivocal support for vinylcarbene intermediates in thermolytic and photolytic cyclopropene transformations. The activation parameters and product distribution noted for the thermolysis of 1-t-butyl-3,3- dimethylcyclopropene are rationalized in terms of ring-opening in both possible directions to vinylcarbenes which subsequently undergo competing insertion and hydrogen and methyl migration.'14 The faster rate of racemization compared with product formation in the thermolysis of 1,3-diethylcyclopropene is a measure of the lower rates of [1,2] and [1,4] hydrogen shifts compared with ring closure in vinylcarbenes.' l5 [1,4] Hydrogen shift in a vinylcarbene intermediate also accounts for the photolytic conversion of 1,2-diphenyl-3,3-dimethylcyclopropene into a mixture of cis-and trans-1,2-diphenyl-3-methylbuta-1,3-diene.Con-versely the products of the photolysis of 2,3,3-triphenylcyclopropene-l-carbox-aldehyde are most readily explained by Wolff rearrangement in an intermediate vinylketocarbene. l7 Rearrangement in vinylketocarbenes produced by nitrogen loss from intermediate vinyldiazoketones also accounts for the photodecomposi- A.D. Wolf V. V. Kane R. H. Levin and M. Jones jun. J. Amer. Chem. Soc. 1973,95 1680. I12 P. H. Gebert R. W. King R. A. LaBar and W. M. Jones J. Amer. Chem. Soc. 1973 95 2357. I13 T. T. Coburn and W. M. Jones Tetrahedron Letters 1973 3903. I14 R. D. Streeper and P. D. Gardner Tetrahedron Letters 1973 767. I I5 E. J. York W. Dittmar J. R. Stevenson and R. G. Bergman J. Amer. Chem. SOC. 1973,95 5680. I16 J. A. Pincock R. Morchat and D. R. Arnold J. Amer. Chem. SOC.,1973 95 7536. Ill L. Schrader and W. Hartmann Tetrahedron Lerrers 1973 3995. Molecular Rearrangements 225 tion of 5-acyl-3,3-dimethyl-3H-pyrazolesto vinyl-ketens. '**' However in the light of the work discussed before,' '1-acylcyclopropenes (significantly the major products of 4-acyl-3,3-dimethyl-3H-pyrazole photolysis' '8 should also be considered as vinylketocarbene precursors in pyrazole photolyses.Interestingly products derived by competing [1,4] hydrogen shift in the alkylated vinylketo- carbene intermediates were not isolated in these reactions though they may lurk in the polymeric' l9 by-products. Bicyclopropenyl to vinylcyclopropenyl- carbene ring-opening and subsequent ring expansion to a Dewar benzene provides the basis for a new mechanistic rationale for the bicyclopropenyl to benzene rearrangement. 2o The intermediacy of oxiren intermediates in photo-Wolff rearrangements and their absence (except possibly at elevated temperatures) in the corresponding thermal processes previously' established experimentally [cf Ann.Reports (B) 1972 69 248-2491 has now received theoretical support.'21 It is suggested that the energetically favourable but otherwise symmetry-forbidden singlet excited ketocarbene to oxiren transition is overcome by internal conversion to a ground-state singlet which still has sufficient vibrational energy to permit access to oxiren.' 2' However this interpretation is based on the calculated' 2' greater stability of the ketocarbene relative to the oxiren. This theoretical prediction is at variance with a MIND0/3 study'22 which suggests that keto- carbenes are substantially less stable than their oxiren counterparts and should rearrange to the latter without activation. On the other hand ab initio calculations indicate that the difference in stability between oxiren and formylcarbene is only CQ.0.49 kcal mol-' in favour of the former species.'23 Attempts to matrix- trap the oxirens produced by photolysis of diazoacetaldehyde and ethyl diazo- acetate have been unsuccessful. '24 In contrast ethoxycarbonylcarbene is trapped (as ethyl propionate) by hydrogen transfer in hydrocarbon matrices.' 24 These experimental findings appear to conflict with the predictions of the MIND0/3 study.' 22 * 3C-Labelling studies,' 25 which demonstrate label scrambling and its absence in ketens derived respectively by photo- and thermal Wolff rdrrange- ments supplement studies reported last year [cf Ann. Reports (B) 1972 69 2491. The differing product ratios observed for the identical reaction mixtures resulting from the peracid oxidative rearrangement of cycloalkynes and the thermal rearrangement of the corresponding diazocycloalkanones may be attributed to the involvement of oxiren intermediates in the former reactions and their absence in the latter.126 Migratory aptitudes have been determined A.C. Day A. N. McDonald B. F. Anderson T. J. Bartczak and 0.J. R. Hodder J.C.S. Chem. Comm. 1973,247. I I' M. Franck-Neumann and C. Buchecker Tetrahedron Letters 1973 2875. I2O R. Weiss and S. Andrae Angew. Chem. Internat. Edn. 1973,12 150 152. I I. G. Csizmadia H. E. Gunning R. K. Gosavi and 0.P. Strausz J. Amer. Chem. Soc. 1973 95 133. 122 M. J. S. Dewar and C. A. Ramsden J.C.S. Chem. Comm.1973 688. I 23 A. C. Hopkinson J.C.S. Perkin II 1973 794. 12d A. Krantz J.C.S. Chem. Comm. 1973 670. J. Fenwick G. Frater K. Ogi and 0.P. Strausz J. Amer. Chem. SOC.,1973 95 124. P. W. Concannon and J. Ciabattoni J. Amer. Chem. SOC.,1973 95 3284. 226 G. Tennant for therma1'2'~'28 and photochemical'28 [1,2] shifts in diketocarbenes. In accord with rearrangements occurring by anionotropic shift to electron-deficient termini the migratory aptitude in both processes increases with increasing nucleophilicityin the migrating group." However for the thermal rearrange- ment steric as well as electronic effects appear to be important and tend to dominate when the two effects are opposed.'28 The failure of the ketocarbene (67) to rearrange to the keten (68) is ~uggested'~' to reflect a ring strain barrier to successful photo-Wolff ring contraction of ca.50 kcal mol- '. The failure of methylmercuricdiazoacetoneto undergo photo-Wolff rearrangement is attributed to excited state to ground state relaxation induced by the proximity of the heavy metal atom to the carbene centre in methylmercuriketocarbene.'30 The products of the gas-phase (600"C)pyrolysis of 1-alkyl and l-aralkyl-l,2,3-triazolesare rationalized by competing [1,2] and [1,4] shifts in iminocarbene intermediates.13' A new version of the Schmidt rearrangement accounts for the reactions of malonic acid half esters with diphenylphosphoryl azide to give high yields of a-amino-acid derivatives.' 32 The failure of added alkenes to intercept nitrene intermediates is cited' 33 as evidence for the concerted nature of photochemically induced C-N methyl shifts in t-alkyl azides.The exclusive methyl shift (in competition with acetonyl shift) observed in the photolysis of 4-azido-4-methyl- pentan-2-one implies that electronic as well as steric factors are important in determining migratory preference in such rearrangements.' 33 New examples of ring expansion and ring contraction originating in thermal and photochemical azide decompositions have been reported. The activation parameters for the high-yield thermolytic rearrangements of cyclopropyl azides to azetines are inconsistent with ring bond shift being concerted with nitrogen loss. The facility of these novel ring expansions is explained in terms of stepwise mechanisms involving cyclopropyl-stabilized nitrene intermediates.'34 The interesting 12' K.P. Zeller H. Meier and E. Muller Tetrahedron 1972 28 5831. G. Heyes and G. Holt J.C.S. Perkin I 1973 1206. 12' B. M. Trost and P. L. Kinson Tetrahedron Letters 1973 2675. P. S. Skell and S. J. Valenty J. Amer. Chem. SOC.,1973,95 5042. 13' T. L. Gilchrist G. E. Gymer and C. W. Rees J.C.S. Chem. Comm. 1973 835. 13' S. I. Yamada K. Ninomiya and T. Shioiri Tetrahedron Letters 1973 2343. '33 S. Solar E. Koch J. Leitch P. Margaretha and 0.E. Polansky Monarsh. 1973 104 220. t34 G. Szeimies U. Siefken and R. Rinck Angew. Chem. Internut. Edn. 1973 12 161. Molecular Rearrangements 227 photochemical and thermal ring contractions of 1-azidocyclobutenes to cyano- cyclopropanes are similarly considered to involve ene-nitrenes which can be trapped by added dipolarophile~.'~~ However the low temperatures which suffice to promote the ring contractions of 2-azidopyridine N-oxides to 2-cyano N-hydroxypyrroles militate against the involvement of nitrene intermediates.' 36 Ring-opening in concert with nitrogen loss and electrocyclization of the nitroso- nitriles produced accounts satisfactorily for these synthetically valuable trans- formations.' 36 The nitrene-nitrene rearrangement of 2-quinolylnitrene to 1-isoquinolylnitrene is to be involved in oxadiazolo- and tetrazolo- quinoline thermolyses.Evidence has been obtained for the intermediacy of N-nitrenes in diazene-hydrazone rearrangements.* Thermal Photochemical and Metalcatalysed Rearrangements.-Valence iso-merizations of fluxional molecules' and photochemical rearrangements of cyclohe~adienones'~~ and cy~loheptadienones'~' have been reviewed. A new theoretical treatment of concerted sigmatropic rearrangements has been published'42 and the steric course of dyotropic rearrangements [cf Ann. Reports (B) 197269,2511 has been di~cussed.'~~ Di-n-methane and oxa-di-n-methane rearrangements have been reviewed.'44 Biradical intermediates are proposed to account for the non-stereospecificity of singlet-state photo-[ 1,2] methyl shifts in /3-t-butyl styrenes.'4s The inhibition of the [1,2] methyl shift by electron-donating groups in the aryl nucleus reveals the operation of polar effects in the n-n* excited states for such rearrangements akin to those governing [1,2] shifts in carbocation~.'~~ The successful photo- rearrangement of a rigid 1,Cdiene system constrained to react by the syn-disrotatory mode demonstrates that access to the normally preferred anti-disrotatory pathway is not a prerequisite of successful di-n-methane rearrange- ment.'46 Preferential ethynyl (sp)shift in competition with vinyl (sp2)migration is observed in the singlet-state photo-rearrangements of cis-and trans-l,5-diphenyl- 3-methyl-3-methoxypent-l-en-4-yne. The facility of these novel stereospecific [l,2] ethynyl shifts contrasts with the difficulty of the corresponding alkyl and hydrogen migrations and supports their formulation as di-n-methane-like [,2 + ,2 + ,2 J cycloadditions.'47 The success of these transformations 13' G.Buhr Chem. Ber. 1973 106 3544. 136 R. A. Abramovitch and B. W. Cue J. Org. Chem. 1973 38 173. 13' R. F. C. Brown F. Irvine and R. J. Smith Austral. J. Chem. 1973 26 2213. 13' B. V. Ioffe and L. A. Kartsova Tetrahedron Letters 1973 623. 139 B. Decock-Le-Reverend and P. Goudmand Bull. SOC. chim. France II 1973. 389. I4O G. Quinkert Angew. Chem. Internat. Edn. 1972 11 1072. 14' H. Hart Pure Appl. Chem. 1973,33 347. 142 .I.Mathieu Bull. Soc. chim. France 11 1973 807. 143 M. T. Reetz Tetrahedron,1973 29 2189. 144 S. S. Hixson P. S. Mariano and H. E. Zimmerman Chem. Rev. 1973,73 531. 14' S. S. Hixson and T.P. Cutler J. Amer. Chem. SOC. 1973 95 3031 3032. P. S. Mariano and R.B. Steitle J. Amer. Chem. SOC. 1973.95 61 15. 14' J. Perreten D. M. Chihal G. W. Griffin andN. S. Bhacca J. Amer. Chem. SOC.,1973 95 3427. 14' 228 G. Tennant contrasts with the lack of di-lr-methane rearrangement reported'48 for hexa- 1,2,5-trienes. The demonstration that the photoisomerization of a sterically unconstrained By-unsaturated ketone occurs with retention at C-2 requires that if concerted the oxa-di-7c-methane rearrangement must be formulated as a [,2 + .2J rather than a [,2 + .2,] cycloaddition.'49 The question of electron-repulsive destabilization of pericyclic transition states for sigmatropic rearrangements has been Considered.' Theoretical calculations of electron repulsion energies indicate that although electron-repulsive destabiliza- tion cannot reverse the order of transition state energies for the concerted- forbidden and non-concerted pathways open to [1,3] carbon shift this need not invariably be so.A case in point is the concerted-forbidden [,2 + .2,] thermal methylenecyclopropane rearrangement repulsive interaction in the transition state for which is predicted to be destabilizing compared with the biradical pathway.' 'O A detailed kinetic and stereochemical analysis of thermal [1,3] carbon shifts in em-and endo-bicyclo[4,3,0]oct-2-enesdemonstrates that steric blockade of the allowed suprafacial-inversion (si) mode permits the operation of the forbidden concerted suprafacial-retention (sr) pathway. ''' The gradual increase in the sr :si ratio for [1,3] carbon shift along the series bicyclo[2,1,1]- hexenyl < bicyclo[3,2,0]heptenyl < bicyclo[4,2,0]octenyl < vinylcyclobutyl is indicative of subjacent orbital control commensurate with a gradual increase in the efficiency of overlap between the front lobe on the migrating carbon atom and the C-2 suprafacial lobe of the allylic framework."' The racemization and lack of antarafacial participation associated with thermal vinylcyclobutane to cyclohexene rearrangements are best explained in terms of competing allowed (si) and forbidden (sr) pathways for these thermal [1,3] carbon shift~.''~ Kinetic parameters for the thermal equilibrations and degenerate automerizations of ethylidenecyclobutanes are also considered to be inconsistent with fully stepwise processes involving biradical intermediates.'' Conversely the demonstrably greater rate of racemization (k = 4.93 k1.13 x lO-'s-l) compared with deuterium scrambling (degenerate automerization) (kds= 8.15 & 0.6 x lo-' s-') in (2)-1-ethylidene-2-methylcyclobutane is construed as evidence for ca. 40% antarafacial allylic participation in the [1,3] carbon shift.lS3 A detailed kinetic and stereochemical analysis of the thermal rearrangement of l-(Z)-( 1-deuterio- ethylidene)-2-methyl-trans-3,4,4-trideuteriocyclobutane indicates a ratio of 77 :23 for the allowed and disallowed rearrangement pathways.' 53 In contrast a kinetic and stereochemical analysis of thermal methylenecyclopropane re- arrangements clearly excludes allowed and forbidden concerted paths for such [1,3] carbon migrations.' 54 Other studies provide compelling evidence for 14* D.C. Lankin D. M. Chihal G. W. Griffin and N. S. Bhacca Tetrahedron Letters 1973,4009. 149 J. I. Seeman and H. Ziffer Tetrahedron Letters 1973,4413. W. T. Borden and L. Salem J. Amer. Chem. SOC.,1973,95,932. Is' J. A. Berson and R. W. Holder J. Amer. Chem. Soc. 1973 95 2037. '52 J. A. Berson and P. B. Dervan J. Amer. Chem. SOC.,1973 95 267 269. IS3 J. E. Baldwin and R. H. Fleming J. Amer. Chem. Soc. 1973,95 5249 5256 5261. lS4 W. von E. Doering and L. Birladeanu Tetrahedron 1973 29 499. Molecular Rearrangements 229 tetramethylene methane biradical intermediates in dimethylenecyclobutane and methylenespiropentane rearrangements.' 55 Activation parameters for the thermal rearrangements of 6-methylenebicyclo[3,2,0]pent-2-ene'56 and aryldi- methylvinylidenecyclopropanes'5 'accord best with stepwise processes involving orthogonal biradical intermediates.Interestingly 1-dimethylvinylidene-2-iso-propylidene-3,3-dimethylcyclopropane which embodies structural features requisite both for vinylidene- and methylene-cyclopropane rearrangement under- goes biradical mediated shift not to the anticipated radialen but to the corres- ponding cross-conjugated dienyne.' 58 Although activation parameters fail to differentiate concerted ([ 1,3] or [3,3]) and biradical pathways for the vinyl- methylenecyclopropane to 3-methylenecyclopentene isomerization the pattern of label scrambling observed in this rearrangement is consistent with a non- concerted process.' 59 Thermal methyleneaziridine automerizations and methyl- eneaziridine-cyclopropanimine interconversions have been conclusively demon- strated for the first time.'60 It now seems probable that the predominant pathway for the thermal rearrangement of 2-methylbicyclo[2,1,0]pent-2-eneto 1-and 2-methylcyclopentadienesinvolves stepwise rupture of the C-1 -C-4 cross-link and subsequent [1,5] hydrogen shifts in the vibrationally activated 2-methylcyclo- pentadiene produced.' 61 However reports'61,'62 that rearrangement to crossed product still occurs to a significant extent in solution requires the vibrationally activated intermediate to exhibit longevity unprecedented for solution-phase isomerizations.Examples of new light-induced [1,3] shifts have been reported.The triplet- state rearrangement of 1,1,4-triphenyl-3,3-dimethylpenta-1 ,Cdiene occurs by [1,3] vinyl shift to the exclusion of the alternative di-n-methane rearrangement.' 63 Photolysis of 3,4benzotropilidene and its 7,7-dimethyl derivatives affords products derived by rare [1,3] hydrogen and methyl migration which thus compete to a significant extent (ca. 10%) with the more commonplace [1,7] shifts.'64 The first-order [1,3] benzyl shift involved in the thermal rearrangement of 1,4dibenzyl-1,4-dihydro-2,6-diphenylpyrazineto the 1,2-dibenzyl-1,2-di-hydro-isomer occurs intramolecularly and with 295 % inversion at the migrating centre. This interesting transformation is claimed as the first unequivocal example of suprafacial [1,3] migration in a nitrogen-containing ally1 frame- Concerted allylic [1,3] shifts of selenium34 and boron ligands'66 have W.R. Roth and G. Erker Angew. Chem. Internat. Edn. 1973,12,503,505; W. Grimme and H. J. Rother ibid. p. 505. 156 D. Hasselmann Tetrahedron Letters 1973 3739. Is' ' ' I. H. Sadler and J. A. G. Stewart J.C.S. Perkin II 1973 278. G. Kobrich and B. Posner Tetrahedron Letters 1973 203 1. J. C. Gilbert and D. P. Higley Tetrahedron Letters 1973 2075; W. E. Billups K. H. Leavell E. S. Lewis and S. Vanderpool J. Amer. Chem. SOC., 1973 95 8096. 160 H. Quast and W. Risler Angew. Chem. Internat. Edn. 1973 12 414. 16' J. I. Brauman W. E. Farneth and M. B. D'Amore J. Amer. Chem. SOC.,1973,95,5043. 16' G. D.Andrews M. Davalt and J. E. Baldwin J. Amer. Chem. SOC.,1973,95 5044. 163 H. E. Zimmerman D. W. Kurtz and L. M. Tolbert J. Amer. Chem. SOC.,1973 95 8210. L64 K. A. Burdett D. H. Yates and J. S. Swenton Tetrahedron Letters 1973 783. 165 J. W. Lownand M. H. Akhtar J.C.S. Chem. Comm. 1973 511. L66 K. G. Hancock and J. D. Kramer J. Amer. Chem. SOC.,1973,95 3425 6463. 230 G. Tennant also been reported. Conversely the demonstration' 67 of concomitant I3C CIDNP emission demonstrates a radical fragmentation-recombination pathway for the [1,3] O-+S shift involved in the oxime thionocarbonate rearrangement.I6' Activation parameters for thermal acyl shifts in 1-acyl-1-methylcyclohexa-2,4-dienes are consistent with their formulation as concerted [1,5] sigmatropic processes.Trapping experiments excluded alternative sequential [1,3] acyl shifts for these novel rearrangements which are shown to exhibit the migratory preference CHO > H x COMe > C02Me.16* Analogous [1,5] shifts likewise rationalize acetyl migrations accompanying the Claisen rearrangements of ortho-allyloxyacetophenones.'69 The interesting thermal rearrangement of bicyclo[3,2,0]hepta-1,3-diene (69; n = 2) to spiro[2,4]hepta-4,6-diene(70; n = 2) illustrates the greater rate of [1,5] carbon shift compared with the corresponding hydrogen shift in this cyclopentadiene system. The facility of the process (69) n = 2 or 3 (70) (69;n = 2)-+ (70; n = 2) may be a consequence of relief of ring strain and the development of a degree of aromatic character in (70; n = 2).I7O The formal reversal of this [l,5] carbon shift (70; n = 3)-(69; n = 3) occurs in the gas phase at 400 "C.'" The label equilibration undergone by octadeuteriobicyclo- [5,l,O]octa-2,4-diene (71) at 110 "C is explained in terms of novel butadienylcyclo- propane equilibria (71) $(72) and (73) (74) interconnected by a reversible [1,5] hydrogen shift (72) $(73) [cf:Scheme 41.' 72 The stereospecificity observed for the degenerate butadienylcyclopropane rearrangement in (7 1) is consistent with its formulation as an antarafacial [1,5] carbon shift involving inversion at the migrating centre.' 72 Interestingly the degenerate isomerization exhibited by the iron tricarbonyl derivative of (71) appears to involve a different transition- state stereochemistry.' 73 Several new examples of thermal and photochemical [1,7] carbon and hydrogen shifts have been reported.'74s'75 Enynols react at -10 "C with diethyl chlorophosphite to give intermediate enynol phosphites which rearrange spontaneously by [2,3] sigmatropic shift to afford ene-allenyl- phosphonates.'76 In the case of 3-methylhex-2-en-Qynol the intermediate phosphite (75) can be isolated but rearranges at 130 "C by a rare and presumably concerted [2,5] shift to give the phosphonate (76).' " Allylsulphones undergo 167 C.Brown R. F. Hudson and A. J. Lawson J. Amer. Chem. SOC.,1973,95 6500. '68 P. Schiess and P. Funfschilling Tetrahedron Letters 1972 5191 5195. C. P. Falshaw S. A. Lane and W. D. Ollis J.C.S. Chem. Comm. 1973,491.M. Oda and R. Breslow Tetrahedron Letters 1973 2537. 171 A. de Meijere and L. U. Meyer Angew. Chem. Internat. Edn. 1973 12 858. W. Grimme and W. von E. Doering Chem. Ber. 1973 106 1765. 173 R. Aumann Angew. Chem. Internat. Edn. 1973 12 574. 174 K. A. Burdett T. J. Ikeler and J. S. Swenton J. Amer. Chem. SOC.,1973,9!5,2702. 17) E. E. Waali and W. M. Jones J. Amer. Chem. SOC.,1973,95,8114; J. Org. Chem. 1973 38,2573. * 76 M. Huche and P. Cresson Tetrahedron Letters 1973 4291. Molecular Rearrangements 231 D D H D D (71) D D D D D .D (74) (73) Scheme 4 clean high-yield (80%) fragmentation-rearrangement at 300 "Cin the gas phase providing a new and probably stereospecific method for constructing carbon- carbon bonds.' Evidence has been obtained for biradical-mediated pathways in the Cope rearrangements of hexa-1,5-dienes.78 2,5-Diphenylhexa-l,5dene undergoes Cope rearrangement two thousand times faster than hexa-1,5-diene. This rate enhancement is inexplicable in terms of a [3,3] sigmatropic shift and is interpreted as evidence for a stepwise mechanism involving a biradical intermediate (1,4-diphenyl-1,kyclohexylene). '* The Cope rearrangement of a structure involving rare doubly bound silicon [(77)$(7831 has been ingeniously demonstrated by the ready stereomutation (E = 39.2 Ifr 0.4 kcal mol-') of cis-and trans-pro- penylallyldimethylsilane [(77) (78) (7911at 500 "C in the gas phase. The facility of this novel transformation (which requires only CQ. 3 kcal mol-I more activation than the allcarbon analogue) is attributed to the ability of silicon to stabilize the pericyclic transition state.' 79 The apparently low energy barrier '" J.B. Hendrickson and R. Bergeron. Tetrahedron Letters 1973 3609. l" M. J. S. Dewar and L. E. Wade. J. Amer. Chem. SOC.,1973 95 290. 179 J. Slutsky and H. Kwart J. Org. Chem. 1973 38,3658. 232 G. Tennant to diaza-Cope rearrangement in 1,2-diarylideneamino- 1,2-diarylethanes lends weight to theoretical predictions [cf Ann. Reports (B) 1972 69 2561relating to transition-state energies for Cope rearrangements.' Deuterium-labelling studies suggest that the formal diaza-Cope rearrangements of N-allylhydrazones are not entirely concerted but follow to the extent of ca. 18% a radical-pair pathway.181 Remarkably (and contrary to previous belief) rearrangement of cis-1,2-divinylcyclopropane to cycloheptadiene is slow enough between 5 and 20 "C to allow kinetic measurements which reveal an activation free energy of ca.20 kcal mol- Unprecedented Cope rearrangement involving a benzene nucleus is proposed to account for the thermal base-catalysed conversion of cis-Zphenylvinylcyclopropaneinto l-phe11ylpenta-l,3-diene.l~~ The sterically difficult thermal transformations of trans- 1,2-divinylcyclobutanes into cyclo-octa- 1,5-dienes have been shown to occur (at least in part) by prior stereomutation to the cis-isomers and subsequent Cope rearrangement. The reluctance of the trans-isomers to rearrange directly to product is a measure of the unfavourable strain barrier to cyclization in the presumed biradical interrnediate.la4 Divinyl- cyclopropane rearrangement provides a mechanistic rationale for the biogenesis of the naturally occurring cyclohepta-1,5-dienes dictyopterins C and D.185 Thermal rearrangements of sterically constrained diacetylene frameworks' 86 are exemplified by the interesting valence isomerizations (80) -+ (8l).' ' (80)X = 0 or S The thermal rearrangements of fused cis-divinylcyclopropanescontinue to stimulate much experimental effort.Preferential rearrangement from folded conformations leading to boat-like transition states accounts for the kinetic 180 F. Vogtle and E. Goldschmit Angew. Chem. Internat. Edn. 1973 12 767. I8 1 R. V. Stevens E. E. McEntire W.E. Barnett and E. Wenkert J.C.S. Chem. Comm. 1973,662. 182 J. M. Brown B. T. Golding and J. J. Stofko J.C.S. Chem. Comm. 1973 319. 183 E. N. Marvel1 and C. Lin Tetrahedron Letters 1973 2679. 184 J. A. Berson and P. B. Dervan J. Amer. Chem. SOC.,1972 94 8949. 185 W. Pickenhagen F. Naf G. Ohloff P. Muller and J. C. Perlberger Hefu. Chim. Acta 1973,56 1868. I86 R. G. Bergman Accounts Chem. Res. 1973 6 25. 187 K. P. C. Vollhardt and R. G. Bergman J. Amer. Chem. Soc. 1972,94,8950; ibid. 1973 95. 7538. Molecular Rearrangements 233 parameters and stereospecificity exhibited by the new heteroanalogous homo- tropilidine valence isomerization (82) S(83)’88 and by new bicyclo[6,1 ,O]nona- 2,6-diene to bicyclo[5,2,0]octa-2,5-diene inter conversion^'^^ and their hetero- analogues.’90 Stereochemical control is also manifest in the thermal rearrange- ments of bicyclo[6,1,O]nona-2,4,6-trienescontaining a bulky C-9 substituent.(82) (83) Thus the stereospecific high yield (9&94 %) thermal rearrangements of syn-and unti-9-t-butylbicyclo[6,1 ,O]nonatrienes to truns-and cis-8,9-dihydroindenes respectively are consistent with previous mechanistic proposals [cf Ann. Reports (B) 1971 68 2611 for these controversial rearrangements.”’ The reluctance of syn-9-cyanobicyclo[6,1,0]nona-2,4,6-trieneto undergo thermal rearrangement can be attributed to stabilization of the C-1-C-8 cross-link lending weight to the contention [cf:Ann. Reports (B),1971,68,261] that syn-9-substituted bicyclo- r6,l ,Olnonatrienes rearrange by direct symmetry-allowed C- 1-C-8 rupture in extended conformations.lg2 The stabilizing effect of a syn-9-cyano-group may also account for the apparent dichotomy in bicyclo[6,1,0]nonatriene isomeriza-tions noted last year [cf Ann. Reports (B),1972,69 2561. Theoretical studies [cf:Ann. Reports (B) 1972 69 2561 predict that electron- releasing substituents will destabilize the transition state for the Cope rearrange- ments of biallyl systems. In accord with this prediction is the demon~tration’~~ that complexation with pentacarbonyltungsten (equivalent to net electron release) raises the activation energy for the degenerate semibullvalene rearrange- ment. Despite other evidencelg4 to the contrary a reinterpretation’” of 13C n.m.r.data’ 96 indicates the absence of homoaromatic character in the hydro- carbon (84). It follows that the ‘Cope rearrangement which cannot be frozen out’ has yet to be demonstrated. (84) I88 H. Klein W. Kursawa and W. Grimme Angew. Chem. Internat. Edn. 1973 12 580. 189 W. Grimme J. Amer. Chem. SOC.,1973,952381. I90 W. Grimme and K. Seel Angew. Chem. Internat. Edn. 1973 12 507. I91 A. G. Anastassiou and R. C. Griffith J. Amer. Chem. SOC.,1973,95 2379. 192 A. G. Anastassiou and R. C. Griffith Tetrahedron Letters 1973 3067. I93 R. M. Moriarty C. L. Yeh E. L. Yeh and K. C. Ramey J. Amer. Chem. SOC.,1972 94 9229. 194 G. P. Ceasar J. Green L. A. Paquette and R. E. Wingard Tetrahedron Letters 1973 1721. 195 E. Vogel U. H. Brinker K. Nachtkamp J.Wassen and K. Mullen Angew. Chem. Internat. Edn. 1973 12 758. I96 E. Wenkert E. W. Hagaman L. A. Paquette R. E. Wingard and R. K. Russell J.C.S. Chem. Comm. 1973 135. 234 G. Tennant The use of Claisen-type rearrangements for the synthesis of yd-unsaturated carbonyl compounds and their thioanalogues continues to attract attention. An elegant variation on this general theme (the so-called Reformatskyxlaisen reaction) involves the in situ [3,3] sigmatropic rearrangement of zinc enolates derived from a-bromo-ally1 or -propargyl esters and provides a new route to yd-unsaturated acids avoiding both strongly acid and strongly basic conditions. 97 Similar rearrangements of enol phosphates derived from allyl trichloroacetates have also been described.lg7 In accord with fully concerted processes involving chair-like transition states the [3,3] sigmatropic rearrangements of allyl vinyl ethers to y6-unsaturated carboxylic acid derivatives proceed with >90% optical induction.The enhanced optical stereospecificity of these rearrangements recommends them as methods for carbon-carbon bond formation at asymmetric centres. 98 Thio-Claisen rearrangements of allyl vinyl sulphides and related substrates likewise provide methods for the generation of double bonds with a high degree of stereochemical The thermal rearrangement of allyl 1,1-dimethylbut-3-enyl ethers (85) to 7-methyloct-6-enol (86) competes to a 35% significant extent with alternative retro-ene cleavage or cyclization. The timing of the bond-making and -breaking processes in this novel formal [4,4] sigma- tropic shift has yet to be established."' Rearrangements involving metal-carbene complexes,2o' and those induced by thallium salts,202 have been reviewed.The known facilitation of the thermally difficult [3,3] sigmatropic propargyl ester-allenyl ester shift [(87) S (91)] by Ag' ions is of interest in relation to orbital symmetry considerations. This Ag'-catalysed process has now been subjected to detailed mechanistic scrutiny by Schmid and his co-workers with notable results.203 Rearrangement is shown to involve the prior formation of an Ag' ion n-complex (88 W),and occurs intra-molecularly and with high stereospecificity. These features prompt its formulation as a fully concerted process (cf Scheme 5) involving a cyclic metal-bonded transition state (89) unperturbed in an orbital symmetry sense by the presence of the metal cation which merely serves in a polar capacity to accelerate the 19' J.E. Baldwin and J. A. Walker J.C.S. Chem. Comm. 1973 117. R. K. Hill R. Soman and S. Sawada J. Org. Chem. 1972 37 3737. 199 H. Takahashi K. Oshima H. Yamamoto and H. Nozaki J. Amer. Chem. Soc. 1973 95 2693 5803; K. Oshima. H. Yamamoto and H. Nozaki ibid. p. 4446. E. N. Marvel1 and M. Fleming Tetrahedron Letters 1973 3789. zoi D. J. Cardin B. Cetinkaya M. J. Doyle and M. F. Lappert Chem. SOC. Rev. 1973,2 99. 202 A. McKillop and E. C. Taylor Chem. Brit. 1973 9 4. 203 H. Schlossarczyk W. Sieber M. Hesse H.-J. Hansen and H. Schmid Helv.Chim. Acta 1973 56 875. 19' Molecular Rearrangements Ar Ar I I / C\o -I / C\o 0 Ag+ . 0 I R'-C fast R' -C R2/\C R2/\c=c-...... 2; \R3 (87) jl slow Ar I C ON \o I slow C -R c(';...*g ......... LCN I I R3 Ar I L ON '0 I I R2 Scheme 5 otherwise sluggish thermal [3,3] shift (i.e.in Schmid's n~menclature~'~ a 'charge induced' [3,3] sigmatropic shift). The probability that such metal-catalysed transformations will not be restricted to the propargyl ester-allenyl ester '04 U. Widmer J. Zsindely H.-J. Hansen and H. Schmid Helu. Chim. Acta 1973,s. 75. 236 G. Tennant equilibrium is substantiated by the dernon~tration~~' of similar Ag'-ion catalysed [3,3] shifts in propargyl phenyl ethers.The nature of the intermediates and the precise mechanisms involved in transition-metal-catalysedstrained a-bond rearrangements are still controversial. It has been suggested206 that the hitherto unconsidered ylide canonical form (93) makes a significant contribution to the hybrid structures of the metal-complexed- carbene intermediates [cf.Ann. Reports (B),1971,68,262; 1972,69,258] proposed for such isomerizations. The course of the observed skeletal reorganizations can then be explained in terms of the relative contributions from (92t(94) which in turn depend specifically on the interplay between the n-donor and a-acceptor capacities of the particular metal. Consequently overriding n-donor capacity should favour the ylide form (93) and overriding a-acceptor capacity the carbonium ion form (94).It follows that metal-carbene intermediates in Rh'- (high n-donor capacity) catalysed rearrangements ought to exhibit ylide character (93). In support of these suggestions is the observation that metal- carbene intermediates in Rh'-catalysed bicycle[l,l,O]butane isomerizations can be trapped (as cyclopropane derivatives) by electron-poor but not by electron- rich 01efins.~" On the other hand there is now little doubt that products of Rh'- catalysed isomerizations in methanol ostensibly derived from metal-carbonium ion precursors [cf. Ann. Reports (B) 1971 68 262; 1972,69 2581 are in fact the result of pure acid-catalysed carbonium ion processes.208~209 Evidence for the nature of the acid-producing species in bicycle[l,l,O]butane-Rh'-methanol systems has been presented.208 Products of the Rh'-catalysed isomerization of naphtho[1,8]tricyclo{4,1,0,02~7]heptenein methanol attributed to the inter- mediacy of metal-carbonium ion species presumably have a similar origin.210 However the intramolecular nature of the Ag'- and Rh'-catalysed rearrangements of exo-l,2,3,4,5,6-hexamethyltricyclo[2,2,O,O2~6]heptane to cis-hexamethylcyclo- hexa- 1,3-diene is most readily explained by [1,2] hydride shifts in intermediate metal-complexed carbonium ions2' In contrast the 2-em hydrogen dependence 2os U.Koch-Pomeranz H.-J. Hansen and H. Schmid Helv. Chim. Acta 1973 56 2981. ''' R. Noyori Tetrahedron Letters 1973 1691. 207 P.G. Gassman and R. R. Reitz J. Organometallic Chem. 1973 52 C51. '08 W. G. Dauben A. J. Kielbania and K. N. Raymond J. Amer. Chem. SOC.,1973 95 7 166. 209 P. G. Gassman and R. R. Reitz J. Amer. Chem. Soc. 1973,95 3057; L. A. Paquette S. E. Wilson G. Zon and J. A. Schwartz ibid. 1972,94,9222; E. Muller Tetrahedron Letters 1973 1201 1203; G. F. Koser P. R. Pappas and S.-M. Yu ibid. p. 4943. 'lo '" I. Murata K. Nakasuji and H. Kume Tetrahedron Letters 1973 3401 3405. H. Hogeveen and J. Thio Tetrahedron Letters 1973 3463. Molecular Rearrangements and deuterium exchange found for Rh'-catalysed isomerizations of bicyclo[2,1,0]- pentanes to cyclopentenes accord best with hydrogen transfer in an allylrhodium hydride intermediate.' '' 3 Aromatic Rearrangements Dienone-phenol2 and arene oxide' rearrangements have been reviewed.The use of e.s.r. spectroscopy has permitted the first direct observation of the 2-methyl-2-phenylpropyl to 1-methyl-2-phenylethyl (neophyl) radical rearrange- ment. The activation energy (E = 43 & 9 kJ mol-') for this [1,2] radical shift is inconsistent with its formulation as a simple dissociation-recombination process.215 New examples of rare thermal [1,4] aryl shifts have been described. Diarylmethane formation in the thermal decomposition of benzyl triarylacetates is rationalized in terms of [ 1,4] aryl shifts which exhibit a ca. twofold preference for p-tolyl as opposed to phenyl migration.'16 Formation of a resonance-stabilized carbanion provides the driving force for the [1,2] phenyl shift which occurs in the base-catalysed rearrangement of 1,2,3,4,5-pentaphenylcyclopenta-2,4-dien- 1-01 to 2,3,4,5,5-pentaphenylcyclopent-2-en-1-one.' ' Novel ortho semibenzene-benzene rearrangements of 2-allenyl- and 2-pro- pargyl-1-methylenecyclohexa-3,5-dienesare implicated in the thermal rearrange- ments of dimethylenetricyclo[3,2,1,02~7]oct-3-enes to butynylbenzenes and in the Wit tig reactions of 2-propargy lcyclohexa-3,5-dienes with tripheny lp hos- phonium methylide to give 4-arylbuta-l,2-dienes.' ' The formulation of these rearrangements as concerted [3,3] sigmatropic processes is supported by the lack of by-products due to radical fragmentation-recombination and the side- chain inversion demonstrated by deuterium labelling.' ' In contrast radical- pair pathways are proposed for the [3,3] shifts undergone by 4-propargyl-l- methylenecyclohexa-2,5-dienes.2' The dimerization of a-fluorobenzyl radicals to give 1,2-difluorobibenzyl has been shown to occur by the intermediate forma- tion and subsequent semibenzene rearrangement of a fluoromethylenecyclohexa-2,4-diene.' ' Semibenzene rearrangements in intermediate cyclohexadienyl zwitterions account for the concomitant group shifts observed in the cyclo- addition reactions of l-methylenecyclohexa-2,5-dienes with tetracyanoethylene.The preference for [ 1,2] phenyl as opposed to methyl migration in these rearrange- ments is a measure of the greater migratory aptitude of phenyl compared with methyl in carbonium ion-induced shifts.220 Crossover experiments support an intermolecular radical fragmentation-recombination mechanism for the semi- phosphabenzene-phosphabenzene isomerization (95)-+ (96).2211 -Nitro-4-acet- 212 K.B. Wiberg and K. C. Bishop Tetrahedron Letters 1973 2727. 'I3 R. S. Ward Chem. Brit. 1973 9 447. 'I4 J. W. Daly D. M. Jerina and B. Witkop Experientia 1972 28,1129. 'I5 E.J. Hamilton and H. Fischer Helu. Chim. Acta 1973 56 795. 'I6 W. S. Trahanovsky D. E. Zabel and M. L. S. Louie J. Org. Chem. 1973 38,757. A. K.Youssef and M. A. Ogliaruso J. 3rg. Chem. 1973 38,2023. 2'8 P.Gilgen J. Zsindely and H. Schmid Helu. Chim. Acta 1973 56 681. 'I9 D. Bethell M. R. Brinkman and J. Hayes J.C.S. Chem. Comm. 1972 1324. ''O N. K. Hamer and M. E. Stubbs J.C.S.Perkin I 1972 2971. 22' G. Mark1 and D. E. Fischer Tetrahedron Letters 1973 223. 238 G. Tennunt PhvPh oxycyclohexa-2,5-dienes undergo novel thermal [1,3] nitro shifts the intra- molecular nature of which is demonstrated by the lack of nitration in added mesitylene.222 Formal [1,2] and [1,3] nitro shifts likewise account for the acid- catalysed conversion of the ‘ipso’ adduct (97) into the isomeric nitro-compounds (98) and (99). The concurrence of [1,2] carbon shifts in these rearrangements was excluded by deuterium labelling.223 A kinetic study indicates that the [1,2] shift rather than the protonation step is rate determining. in simple dienol- benzene rearrangement^.^^^ The acid-catalysed rearrangements of hydroxy-pentadienyl-1,2-dihydronaphthalenes occur almost entirely by [1,2] pentadienyl shift.The hoped for high-order [5,5] and [5,6] pentadienyl migrations were not observed.22 Trifluoroacetic acid-catalysed rearrangement of allylcyclohexadienones affords allylphenols derived exclusively by concerted [3,3] sigmatropic allyl shifts. Conversely the rearrangement of allylcyclohexadienones in the presence of trifluoroacetic anhydride or acetic anhydride-sulphuric acid yields allylphenols derived by competing [1,2] and [3,4] as well as [3,3] sigmatropic allyl shifts.204 The discovery of this dichotomy in the allylcyclohexadienone-allylphenol rearrangement has led to a new classification for charge-mediated sigmatropic processes.204 It is proposed that the [3,3] shifts (designated ‘charge induced processes’) originate in oxygen-protonated allylcyclohexadienone intermediates and involve charge-stabilized but otherwise thermally equivalent cyclic transition states.Conversely the [1,2] and [3,4] shifts (designated as ‘charge controlled processes’) are suggested to stem from acyloxybenzenium ion intermediates and consequently to involve highly polarized transition states having quite different orbital symmetry requirements compared with the purely thermal processes.204 222 A. Fischer and C. C. Greig J.C.S. Chem. Comm. 1973 396. 223 R. C. Hahn and M. 9. Groen J. Amer. Chem. SOC. 1973 95 6128. 224 V. P. Vitullo and M. J. Cashen Tetrahedron Letters 1973 4823. 22s H. Greuter H.-J. Hansen and H. Schmid Helv. Chim. Acta 1973 56 2479.Molecular Rearrangements 239 The dienone-phenol rearrangements of 2-propargylcyclohexa-3,5-dienones catalysed by trifluoroacetic anhydride-boron trifluoride or acetic anhydride- sulphuric acid occur by exclusive ‘charge-controlled’ [1.21 and [3,4] pathways affording 3-propargyl- and 3-allenyl-phenols respectively. 226 The novel Ag’- ion-catalysed dienone-phenol rearrangement of a 2-allenylcyclohexa-3,5-dienone to a 3-allenylphenol may likewise be classified as a ‘charge controlled’ [1,2] shift involving a silver benzonium ion complex.205 A kinetic study of the acid- catalysed rearrangement of 4-ethyl-4-methylcyclohexa-2,5-dienone to 3-ethyl-4- methylphenol reveals a twenty-five-fold greater preference for ethyl shift compared with methyl shift.22 Deuterium-labelling studies support a purely intramolecular (as opposed to addition-elimination) pathway for the dienone-phenol rearrange- ments of ortho-quinol acetates.228 A detailed kinetic study has revealed that the rearrangements involved in the conversion of indane 8,9-oxide into indan-4-01 and indan-5-01 occur mainly by an arene oxide-arene oxide (‘oxygen-walk’) pathway and only to a minor extent by alternative spirodienone-phenol rearrange- ment.229 The novel thermal arene oxide-arene oxide rearrangement (100) -+ -(101) occurs readily at 80-100 “C (E = 28.2 f0.8 kcal mol l) and shows a lack of rate dependence on solvent polarity which prompts its formulation as an allowed suprafacial [1,5] shift akin to the cycloheptatriene-norcaradiene valence i~omerization.~~~ The interesting photochemical rearrangement of 9,10-dihydro-9,10-epoxyphenanthrene(102) to 1,2 :3,4-benzoxepin (104) may also be explained by an ‘oxygen-walk’ mechanism involving a concerted suprafacial [1,5] shift [(102) +(10311 (allowed in the first excited state with inversion at oxygen) and subsequent thermal valence isomerization [( 103)* (104)].23’ [1,2] Shifts in iminonium cation intermediates are proposed to account for alkyl migrations which accompany the potassium ferricyanide oxidation of aniline derivatives to q~inone-anils.~~~ that peroxidase The dem~nstration’~~ induces similar shifts in a variety of alkyl groups (e.g.methyl t-butyl cyclohexyl) in uitro implies a possible relationship with similar enzymically promoted rearrangements.226 U. Widmer H.-J. Hansen and H. Schmid Helu. Chim. Acra 1973 56 1895. 227 J. W. Pilkington and A. J. Waring Tetrahedron Letters 1973 4345. 228 H. Budzikiewicz and J. Gunavan Monatsh. 1973 104 876. 229 G. J. Kasperek P. Y. Bruice T. C. Bruice H. Yagi and D. M. Jerina J. Amer. Chem. SOC.,1973 95 6041 1673. 230 F. G. Klarner and E. Vogel Angew. Chem. Internat. Edn. 1973 12 840. ”* N. E. Brightwell and G. W. Griffin J.C.S. Chem. Comm. 1973 37. 232 S. L. Goldstein and E. McNelis J. Org. Chem. 1973 38 183. 233 P. B. Baker V. R. Holland and B. C. Saunders Tetrahedron 1973 29 85. 240 G. Tennant &-[;:I,& --*& -( 102) ( 103) (104) The trifluoromethane-sulphonic acid-catalysed transformations of ortho-hydroxybenzophenones to phenyl benzoates exemplify ret ro-Fries rearrangements which are not therefore so rare as previously thought.The implication that acid- catalysed Fries rearrangements are subject to thermodynamic control requires the reappraisal of mechanistic proposals for such transformations the reversibility of which has not been considered hitherto.234 The observation of CIDNP effects attributable to the intermediacy of aryloxy and aroyl radicals in the photochemical rearrangement of para-cresyl 2~chlorobenzoate to 2-(2’-chloro- benzoy1)-p-cresol provides convincing evidence for the radical-pair mechanism of the photo-Fries rearrangement.235 The acylamino fragmentation which ac- companies photo-acyl shift in NN-diacylanilines and NN-diacylnaphthylamines likewise supports radical-pair pathways for these ph~to-rearrangements.~~~ On the other hand the absence of phenolic by-products and products derived by para-substitution in the photo-rearrangement of 2-phenoxybenzimidazole to 2-(2’-hydroxypheny1)-benzimidazoleis tentatively suggested to imply a con-certed mechanism.237 The latter reaction is claimed237 as the first example of a photo-Fries rearrangement to involve the preferential migration of a heterocyclic nucleus.Photo-Fries rearrangement of N-acyl- and N-alkoxy-indoles provides routes to relatively inaccessible 3,4-and 6-substituted indole derivatives.238 The site specificity of the high-yield triplet-state photo-rearrangement of ortho- phenoxybenzoic acid to phenyl salicylate excludes a radical fragmentation- recombination pathway.A mechanism involving intramolecular nucleophilic aromatic substitution in a polar excited-state complex is suggested to account for this novel photoisomerization.239 It is of interest that the reverse of this process appears to be involved in the photochemical transformation of benzyl salicylate into ortho-benzyloxybenzoic The inhibition of .rearrangement in the presence of a radical scavenger (nitric oxide)241 and the observation242 of CIDNP emission associated with phenoxyl and ally1 radicals clearly support a radical fragmentation-recombination mechanism for the photo-Claisen re-arrangement. ortho- And para-products of such rearrangements originate in 234 F. Effenberger H. Klenk and P. L. Reiter Angew.Chem. Internat. Edn. 1973 12 775. 235 W. Adam J. Arce de Sanabia and H. Fischer J. Org. Chem. 1973 38 2571. 236 Y. Katsuhara H. Maruyama Y. Shigemitsu and Y. Odaira Tetrahedron Letters 1973 1323. 237 P. D. Hobbs and P. D. Magnus J.C.S. Perkin I 1973,469. 238 M. Somei and M. Natsume Tetrahedron Letters 1973 2451. 239 N. C. Yang P. Kumler and S. S. Yang J. Org. Chem. 1972 37 4022. 240 M. Afzal Chem. and Ind. 1973 37. 241 F. A. Carroll and G. S. Hammond J. Amer. Chem. SOC.,1972 94 7151. 242 W. Adam H. Fischer H.-J. Hansen H. Heimgartner H. Schmid and H. R. Waespe Angew. Chem. Internat. Edn. 1973 12 662. Molecular Rearrangements 241 singlet radical-pair encounters whereas rneta-products are the result of encounters in triplet radical pairs.242 An elegant stereochemical study has shown that thermal and boron trifluoride-catalysed para- ortho Claisen rearrangements of steroidal allylcyclohexadienones exhibit identical optical stereospecificity consistent with their formulation as [3,3] concerted sigmatropic processes involving essentially identical and largely ( >95%) chair-like transition states.243 This significant study supports the ont tent ion^^^,^^^ that ‘charge-induced’ sigmatropic shifts in oxygen- co-ordinated allylcyclohexadienones involve essentially unperturbed ‘thermal’ transition states the presence of the charge merely acting as a stabilizing and therefore rate-enhancing influence.Boron trichloride-catalysed ortho-Claisen rearrangements of allyl aryl ethers show all of the mechanistic features associated with their purely thermal counterparts thereby permitting their formulation as true concerted [3,3] sigmatropic shifts.244 Consequently the 10’o-fold Jate en- hancement exhibited by such processes compared with their thermal analogues is a measure of the charge induction resulting from allyl shift in a boron trichloride- co-ordinated allylcyclohexadienone as opposed to an unco-ordinated species.244 Charge induction resulting from co-ordination at nitrogen likewise accounts for the facility of the zinc chloride-catalysed amino-Claisen rearrangements of N-allylanilines compared with the difficulty of the corresponding thermal processes.245 A kinetic has shown that the rates of ortho-Claisen rearrangements of allyloxynaphthoquinones are accelerated in protic media.These results which amplify observations reported last year [cf:Ann. Reports (B) 1972 69 2131 are attributed to the stabilization of the polar transition state for rearrangement by hydr~gen-bonding.~~~ Kinetic evidence indicates that the products of the thermal rearrangements of 3-substituted prop-Zynyl tropolone ethers result from novel [3,7] sigmatropic shifts.247 An example of a thermal [5,5] pentadienyl shift in an aryl pentadienyl ether has been reported.23 Lack of crossover demonstrates the intramolecular character of the thermal 0-arylhydrazonate-hydrazide rearrangement [cf Ann. Reports (B) 1971 68 267].248*249 The polar effects of substituents on the rate of these rearrangements differ from those of the formally analogous Chapman rearrangement of aryl imidates and prompt a radical-pair mechanism for the hydrazonate processes.248 Benzidine rearrangements of N-acetyl- and NN-dimethyl-hydrazobenzene have been rep~rted.~~~,~~’ Deuterium-labelling studies demonstrate the intra-molecularity of the rearrangement of N-acetylhydrazobenzene in strong acid to give by a ‘one proton’ mechanism N-acetylbenzidine.It is suggested that 243 A. Wunderli J. Zsindely H.-J. Hansen and H. Schmid Helv. Chim. Acta 1973 56 989. 244 J. Borgulya R. Madeja P. Fahrni H.-J. Hansen and H. Schmid Helv. Chim. Acta 1973 56 14. 245 M. Schmid H.-J. Hansen. and H. Schmid Helv. Chim. Acta 1973,56 105. 246 J. A. Miller and C. M. Scrimgeour J.C.S. Perkin II 1973 1137.247 R. M. Harrington and J. D. Hobson J.C.S. Perkin I 1973 1960. 248 A. F. Hegarty J. A. Kearney and F. L. Scott J.C.S. Perkin II 1973 1422. 249 A. S. Shawali and H. M. Hassaneen Tetrahedron 1972 28 5903. 250 J. R. Cox and M. F. Dunn J. Org. Chem. 1972 37,4415. 251 D. V. Banthorpe and M. O’Sullivan J.C.S. Perkin II 1973 551. 242 G.Tennant the N-acetyl group aids N-N bond heterolysis in the conjugate acid and consequently that rearrangement follows a n-complex path~ay.~” The lack of ring deuteriation observed in the course of these studies is cited2s0 as evidence against the CN-diprotonation hypothesis proposed [cJ:Ann. Reports (B) 1972 69,2631 to account for the courses of ‘two proton’ benzidine rearrangements. Despite the formation of products (semidines fission amine) attributable to radical fragmentation-recombination the absence of crossover and the ineffec- tiveness of radical scavengers support an intramolecular polar transition state mechanism for the acid-catalysed benzidine rearrangement of NN-dimethyl- hydra~obenzene.~” Deuterium- and ’N-labelling studies demonstrate that ortho-and para-semidine formation in ‘one proton’ benzidine rearrangements occurs intramolecularly.2s2 The acid-catalysed rearrangement of the cyclic hydrazobenzene (105) affords (106) the first isolable ortho-semidine inter- (105) ( 1 06) mediate.253 The preferential formation of this product is attributed to steric restraints in (105) which inhibit alternative benzidine andpara-semidine rearrange- ment~.~ The acid-catalysed arenesulphenanilide-aminodiarylsulphide re-s arrangement has been shown to be quite general and to exhibit features in common with the benzidine and nitramine rearrangements.The available evidence supports an intramolecular radical-pair mechanism for these rearrange- ment~.~~~ It has been shown that the photo-Wallach rearrangement is preceded by protonation in the first excited state.2ss The acid-catalysed rearrangement of 2,2’,4,4’,6,6’-hexamethylazoxybenzene to 4-hydroxymethyl-2,2‘,4,6,6’-penta-methylazobenzene illustrates a new variant of the Wallach rearrangement.2s6 The ring contraction (107) +( 108)76 is proffered as the year’s most unusual molecular rearrangement albeit in a kinetic rather than a structural sense ! 2s2 A.Heesing and U. Schinke Chem. Ber. 1972 105 3838. lS3 W. W. Paudler A. G. Zeiler and M. M. Goodman J. Heterocyclic Chem. 1973 10 423. 254 F. A. Davis E. R. Fretz and C. J. Horner J. Org. Chem. 1973 38 690; F. A. Davis C. J. Horner E. R. Fretz and J. F. Stackhouse ibid. p. 695. *” R. H. Squire and H. H. Jaffe J. Amer. Chem. Sac. 1973,95 8188. 256 R. A. Cox and E. Buncel Canad. J. Chem. 1973 51 3143.