12 Alicyclic Chemistry By A. COX Department of Chemistry and Molecular Sciences University of Warwick Coventry CV4 7AL 1 Introduction This year has seen the publication of a further volume of 'Organic Syntheses" and a survey of the 1975 literature on mechanisms has appeared.2 Reviews have been published on the chemistry of rings of various sizes,3 intramolecular [4+2] and [3 +21 cycl~additions,~ pericyclic ~ynthesis,~ the Robinson Annelation6 and strained organic rnolec~les.~ Finally further work is reported on the rules for ring closure' and a new mechanistic criterion has appeared for early and late transition ~tates.~ 2 Synthesis Three-and Four-membered Rings.-A high-yield route to cyclopropyl and cyclo- propenyl ethers has appeared" which involves trapping of an alkoxycarbenoid as the key step.For example treatment of a mixture of EtOCH2CI and cyclohexene with lithium 2,2,6$-tetramethylpiperidine (LiTMP) leads to 7-ethoxynorcarane (55%). In a related paper an improved synthesis of cyclopropanols is reported." Use is again made of LiTMP to generate simple alkoxycarbenoids which are then allowed to react with an alkene but in this case they are formed by a-elimination of HCI from P-chloroethyl ethers. The product ethers are subsequently cleaved using n-butyl-lithium; use of acyloxycarbenes generated similarly is also reported. l2 A method has been de~cribed'~ for the conversion of @-unsaturated aldehydes and esters into cyclopropanes. The transformation (Scheme 1)involves conversion of the carbonyl oxygen into a leaving group while the /3-carbon is rendered nucleo- philic by Birch reduction of the thiophenyl moiety.A communication describing an 'Organic Syntheses' ed. G. H. Biichi Wiley New York 1977 Vol. 56. 'Organic Reaction Mechanisms' ed. A. R. Butler and M. J. Perkins Interscience London 1977. 'Alicyclic Chemistry' ed. W. Parker (Specialist Periodical Reports) The Chemical Society London 1977 Vol. 5. W. Oppolzer Angew. Chem. Internat. Edn. 1977,16 10. A. G. Anastassiou Accounts Chem. Res. 1976 9,453. R.E. Gawley Synthesis 1976 777. ' J. F. Liebman and A. Greenberg Chem. Rev. 1976.74 311. J. E. Baldwin and L. I. Kruse J.C.S. Chem. Comm. 1977,233. R. Huisgen and R. Schug J. Amer. Chem. Soc. 1976,98,7819. lo R. A. Olofson K.D. Lotts and G. N. Barber Tetrahedron Letters 1976 3779. G. N. Barber and R. A Olofson Tetrahedron Letters 1976 3783. l2 R. A. Olofson K. D. Lotts and G. N. Barber Tetrahedron Letters 1976 3381. l3 Y.-H. Chang D. E. Campbell and H. W. Pinnick Tetrahedron Letters 1977 3337. 285 286 A. Cox Reagents i PhSH-Na; ii LiAIH4; iii MeSOzCI-py; iv Li-NH3 Scheme 1 efficient and operationally straightforward route to divinylcyclopropanes has appeared.14 Conversion of trans -1-bromo-2-vinylcyclopropane into trans-lithio- vinylcyclopropane is accomplished using t-butyl-lithium in ether-pentane at -78 "C and following addition of 3-ethoxy-2-methylcyclopent-2-enoneand subsequent hydrolysis (Scheme 2) the trans-divinylcylopropane product can be isolated in high yield.Scheme 2 The generation and trapping of the interesting hydrocarbon methylene- cyclopropane has been reported (Scheme 3).15 Reaction of (1) with Bu'OK in Reagents i Bu'OK-MeS-; ii H202-MeC02H; iii Bu'OK-Me2SO Scheme 3 Me2S0 in the presence of MeS- yields (2) together with the sulphide (3). Oxidation of (3) gives (4) which with Bu'OK leads to (2) as the only volatile product Since l4 P. A. Wender and M. P. Filosa J. Ore. Chem. 1976,41 3490. l5 W. E. Billups A. J. Blakeney and W. T.Chamberlain J. Org. Chem. 1976 41 3771. Alic yclic Chemistry sulphones are well known to give alkenes via p-elimination this provides compel- ling evidence for the intermediacy of methylenecyclopropane. Also reported16 is the synthesis of methylenecyclopropenes lacking a stabilizing polar substituent on the exocyclic double bond (Scheme 4).It has been found that treatment of (5) with (5) R' =Hor But (6) R2= But or H Scheme 4 Bu'OK in dry THF at low temperatures leads to (6) a structure observed to be extremely reactive towards water. A synthesis of four-membered rings has appeared" which allows cyclobutane cyclobutene and cyclobutanone to be generated from both saturated and @-unsaturated aldehydes and ketones triu the intermediacy of the adducts of 1-1ithiocyclopropyl phenyl sulphide to the carbonyl partner. In another communication further syntheses of cyclobutanones poten- tially useful in the terpene field are reported." The reaction involves acid rear- rangement of a variety of alcohols having the 1-oxycyclopropyl structure and prepared from CY -alkoxy-@-unsaturated ketones and esters by way of reduction followed by Simmons-Smith reaction of the resultant (Y -alkoxyallyl alcohols.Five- and Six-membered Rings.-A new synthesis of five-membered rings accord- ing to a (C3+Cz) pattern has been anno~nced'~ and provides a unique route to derivatives having the functionality shown in two readily executed steps (Scheme 5). Slow addition of solutions of the lithium enolates of the ketones (7) to a solution R2 K-R'. 'f R2 (9) Scheme 5 l6 W. E. Billups and A. J. Blakeney J. Amer. Chem. SOC.,1976,98,7817. l7 B. M.Trost D. E. Keeley H. C. Arndt and M. J. Bogdanowicz J. Amer. Chem. Soc. 1977,99,3088. E.Wenkert N.F. Golob R.P.Hatch D. Wenkert and R. Pellicciari HeZu. aim. Am 1977,60,1. l9 D.Seebach M.S.Hoekstra and G. Protschuk Angew. Chem. Internut. Edn. 1977,16,321. 288 A. Cox of P-nitropropionyl chloride in THF at -80 "C furnishes the diketone (8). Cycliza-tion can be effected by a buffer solution of pH 8.5 to give hydroxycyclopentenone which rearranged to (9) on prolonged treatment. In another method based on the (C,+C,) pattern and one of considerable generality the very highly reactive dimethoxytetrachlorocyclopentadiene is heated in the presence of olefins.20 This leads to the Diels-Alder adducts which on reaction with sodium in liquid ammonia- ethanol removes all the chlorine atoms. Oxidation of the norbornene-type double bond is accomplished using potassium permanganate in a pH 7 buffer system (Na2HP04,-KH2P04) with t-butyl alcohol as co-solvent.Yields of between 22 and 40% overall have been achieved for the four steps involved. The construction of five-membered rings of potential use in prostaglandin synthesis continues to attract interest and one such example is a newly reported21 ene route to 3-hydroxy- cyclopentanone derivatives. The vinylogous aldol (lo) protected as its t-butyl- dimethylsilyl derivative is submitted to the kinetic aldol reaction with propynal to give (1 1). Reaction of (11) with trimethylsilyl cyanide promotes cyanohydrin formation and simultaneously protects the two free hydroxyls as their trimethylsilyl ethers. By heating in toluene at 250°C thermal cyclization of (12) is readily effected giving (13) (Scheme 6).An interesting report has appeared22 of the H OH I OH application in 11-deoxyprostaglandin El synthesis of thiophenol behaving as a nucleophile in stereospecific ring-opening reactions. The key transformation involves treatment of methyl exo-6-(trans-hept-l-enyl)-2-oxobicyclo[3,1,0] hexane-1-carboxylate (14)with potassium thiophenolate in Bu'OH at room temperature to produce the 2,3-disubstituted cyclopentanone (Scheme 7) in 89% 2o M. E. Jung and J. P. Hudspeth J. Amer. Chem. SOC.,1977,99,5508. 21 G. Stork and G. Kraus J. Amer. Chem. SOC.,1976,98,6747. 22 K. Kondo T. Umemoto Y. Takahatake and D. Tunemoto Tetrahedron Letters 1977 113. Alicyclic Chemistry Scheme 7 yield. The desired 11-deoxyprostaglandin El was then reached in four steps by standard means.A synthesis of PGA2 using the same procedure has also been reported.23 The reaction between some cyclic olefins (C5-Cs inclusive) and diethyl(methy1)silane and CO in the presence of CO~(CO)~ has been In the particular case of cyclopentene a 48% yield of the enol silyl ether of cyclo-hexanone has been achieved and much better yields are reported for the higher homologues. The reaction affords direct access to enol silyl ethers from olefins. A report has appeared2’ of the preparation of a versatile new diene 2-methoxy-3- phenylthiobutadiene which can serve as an annelating agent to introduce a masked @-ketosulphide moiety as an integral part of an annelation; this masking allows structural modification elsewhere.A second point of importance is that the regio- chemistry complements that normally obtained with 2-oxygenated dienes and the ease with which sulphur can be removed from organic molecules suggests this as a possible general approach to reversing the normal orientation of Diels-Alder reactions studied. However although this diene reacts highly successfully with maleic anhydride at room temperature yields of adducts derived from the easily polymerizable methacrylyl dienophiles are low. An investigation of the iodocy- clization of homologous dienes reveals2’ that in the case of hepta-l,&diene introduction of dialkyl substitutents at the 4-position completely changes the reaction path from that of addition to cyclization (Scheme 8).The scope of the iodocyclization of several other dienes has also been investigated. Scheme 8 Seven- and Eight-membered Rings.-A description has been given2* for the con-struction of functionalized cycloheptadienes which is based on the use of 2-vinyl-cyclopropylcopper reagents. For example treatment of the homocuprate (1 5) with 23 D. F. Taber J. Amer. Chem. SOC.,1977,99 3513. 24 Y.Seki A. Hidaka S. Murai and N. Sonoda Angew. Chem. Internat. Edn. 1977.16 174. 25 B. M. Trost and A. J. Bridges J. Amer. Chem. SOC.,1976,98 5017. 2b S. Danishefsky C. F. Yan and P. M. McCurry jun. J. Org. Chem. 1977,42 1819. 27 H. J. Giinther V. Jager and P. S. Skell Tetrahedron Letters 1977 2539. J. P. Marino and L. J. Browne Tetrahedron Letters 1976 3245. 290 A.Cox methyl propiolate at -78 "C gives the trans-divinylcyclopropene(16) (80%) and the rearranged cycloheptadiene (17) (15%) as shown in Scheme 9. Subsequent rearrangement of (16) at 140"C yields (17) quantitatively. A new general route29 to 3-substituted cyclohepta-l,4-dienes involves [4 +21 cycloaddition of the appro-priate 7-substituted cycloheptatriene to 4-phenyl-172,4-triazoline-3,5-dione. CHC0,Me CuLi HCEECCO~M~ d -78 "C d+ (15) (16) ""'"-o // (16) Scheme 9 Following catalytic reduction basic hydrolysis of the heterocyclic ring of these products and Cu2' oxidation of the resulting hydrazo-compounds nitrogen is readily lost to give 3-substituted cyclohepta-1,4-dienes. An improved one-step method for preparing trans-cyclo-octene has been pub-lished3' involving direct photoisomerization of the cis-isomer in pentane.The practical procedure consists of direct irradiation at 185 nm of the 0.12or 0.6 molar solution of cis-cyclo-octene and subsequent extraction of the product with aqueous silver nitrate. The same authors also report3' a detailed investigation of the direct and sensitized cis-trans photoisomerizations of cyclo-octene in the liquid phase along with photosensitization in the vapour phase. A discussion is presented of the effects of spin multiplicity and vibrationa1 activation of the excited states on the photostationary state and potential curves accounting for the results are proposed. A convenient synthesis of some substituted cis,trans-cyclo-octadienes (18) has been which can be either isolated or trapped depending on their reactivity.R' R' R2 (18) R' = C02Me R2=H R' =CN R~= H R' =R2 = C02Me Annu1enes.-The first report has appeared of the 175-methano[lolannulene (bicy-clo[5,3,l]undeca-1,3,5,7,9-pentaene)system" and several of its derivative^.^^ The 29 I. Pikulik and R. F. Childs Canad. J. Chem. 1977 55 251. 30 Y. Inoue S. Takamuku and H. Sakurai Synthesis 1977 111. 31 Y. Inoue S. Takamuku and H. Sakurai J. Phys. Chem. 1977,81,7. 32 H. D. Martin M. Hekman G. Rist H. Sauter and D. BelluH Angew. Chem. Internat. Edn. 1977 16 406. 33 S. Masarnune and D. W. Brooks Tetrahedron Letters 1977 3239. 34 S. Masamune D. W. Brooks K. Morio and R. L. Sobczak J. Amer. Chem. SOC.,1976,98 8277. Alicyclic Chemistry 291 13 C n.m.r.spectrum of (19 R'=R2= H) exhibits only seven signals even at -80 "C and the molecule therefore possesses two-fold symmetry. Induction of a diamag- netic ring current in this monocyclic 10welectron system is suggested by the high-field H signals due to the C-11 protons and the appearance of signals in the aromatic region. In contrast to the 1,6-methano counterpart the U.V. spectrum of R' (19) (19) extends deeply into the visible region. Sondheimer has described3' a new and improved synthesis of derivatives of [14]annuleno[ 14]annulene [14lannu-leno[ 16]annulene and [14]annuleno[ 18lannulene. These vinylogues of naphthalene are the first examples of bicyclic compounds consisting of two fused macrocyclic conjugated v-systems and the effect of one ring upon another has been investigated by 'H n.m.r.spectroscopy. The synthesis and properties of a third example of a fused non-benzenoid aromatic hydrocarbon (20),incorporating two 14~-electron systems has been reported.36 Analysis of the 100MHz 'H n.m.r. spectrum reveals that (20) is strongly diatropic showing the inner proton signals at S -2.85 and those of the outer protons at S 10.16. The electronic spectrum of (20) shows a considerable hypsochromic shift as compared with that of 3,11,14,22-tetra-t-butyl-l,12-bisdehydro[22]annulene. Both of these observations suggest that the bicyclic annulene (20) is a higher analogue of napthalene and may be represented by the delocalized formula (21). The synthesis of the tosylhydrazone sodium salt (22) (20) (21) has been and pyrolysis of the salt in refluxing diglyme gives contrary to anticipation not the [4,4,2]propellane (24) via a carbene+arbene rearrangement but rather only products believed to originate from carbene (or allene) dimers.The probable intermediates (23) are thought to lose hydrogen spontaneously to give the unsaturated hydrocarbon (25) with the double bonds localized as shown in Scheme 10. Polycyclic Systems.-A new reversible formation of bicyclobutanes involving a 1,3-bismethylenecyclobutanehas been described.38 Specifically solutions of (26) in ethanolic lithium perchlorate react rapidly with iodine to give the bicyclobutanes 35 T. M. Cresp and F. Sondheimer J. Amer. Chem. SOC.,1977,99 194. 36 S.Akiyama M. Iyoda and M. Nakagawa J. Amer. Chem. SOC.,1976,98,6410. 37 U. H. Brinker R. W. King and W. M. Jones,J. Arner. Chem. SOC.,1977,99,3175. 38 M. Horner and S. Hiinig J. Amer. Chem. Soc.,1977,99. 6120. 292 A. Cox N-N-Ts (23) syn +anti (25) syn +anti 1 I Scheme 10 (27) in high yield (Scheme 11). It is also observed that the bridging bond in (27) is readily cleaved by reduction e.g. Na-Hg in acetonitrile to regenerate the original (26) R' = H R2 =Me R' = H R2= PhCH2 ~1~ R~= M~ R' =Me R2 = PhCH2 Scheme 11 1,3-bismethylenecyclobutanes.Evidence has been made available3' testifying to the fleeting existence of bicycle[ 1,l,O]but-1(3)-ene a compound suggested by calculations to exist on a local minimum on the C4H4potential energy hypersur- face.It has been observed that reaction of (28) with n-butyl-lithium leads to an 87% yield of l-n-butyltricyclo[4 1,0,02*']heptane in a very rapid reaction. Of the three mechanistic possibilities considered for this conversion elimination of hydrogen chloride from (28) by the organometallic reagent with the formation of a bicyclo[1,l,O]but-l(3)-ene derivative and addition of the organolithium compound to the strained double bond is considered the most likely. The first member has been reported4' of a class of hitherto unknown fused bicylic trans-cycloalkenes in which two rings share a common double bond. This class of chiral olefin is 39 G. Szeimies J. Harnisch and 0.Baumgartel J. Amer. Chern. SOC.,1977,99 5183.40 J. A.Marshall and M. Lewellyn J. Amer. Chem. SOC.,1977 99 3508. Alicyclic Chemistry designated [a,b]betweenanenes (29) and the one reported is [ 10,lolbetweenanene (30). No report is available of the chemical and physical properties which could be bcl L (28) (29) (30) unusual as a consequence of the buried nature of the double bond. Semibenzilic acid ring contraction of 6-halogenotricyclo[3,2,1 ,03*6]octan-7-one (3 1) has been observed to give4' tricyclo[3,1,1 ,03.6]heptane-6-carboxylicacid (32) a compound 0 R (31) X=C1 or Br (32) R=C02H containing a new ring system characterized by three cyclobutane rings fused to a common carbon atom. The structure of this acid was determined unequivocalIy by examination of the 'H and 13Cn.m.r.spectra of the corresponding methyl ester. Studies are reported 42 of the synthesis of the novel bridgehead olefin tri- cyc10[3,3,3,0~.~] undec-2(6)-ene (33). In Wiseman's analysis of bridgehead olefins this compound may be viewed as a bisethano-derivative of trans-cycloheptene and although the olefin carbons in such molecules may be somewhat pyramidalized overlap between the atomic orbitals forming the rr-part of the double bond is principally diminished by torsion. Although (33) has not been isolated decom- position of (34) in refluxing triethyl phosphite containing diphenylisobenzofuran as trapping agent gives the Diels-Alder adduct of (33) in 85% isolated yield Detailed 'S (33) (34) reactivity studies are reported43 of substrates which can serve as precursors to the 'anchored' trishomoaromatic ions (35)-(37).Particular reference is made to gain- ing access to (35) and a number of remarkable interconversions on the (CH), energy surface are described. For example reduction of tri-cyclo[5,4,1 ,04v'2]dodeca-2,5,8 10-tetraene (38) using potassium metal in ammonia gives a mixture of four hydrocarbons of which the major product (39) has been '' S. A. Monti and J. M. Harless J. Amer. Chem. SOC.,1977,99 2690. 42 R.Greenhouse T. Ravindranathan and W. T. Borden J. Amer. Chem Soc. 1976,98,6738. 43 L. A. Paquette M. J. Kukla S. V. Ley and S. G. Traynor J. Amer. Chem. Soc.,1977,99,4756. 294 A. Cox defined spectroscopically. On the basis of these measurements it is not possible to distinguish between direct two-electron reduction of (38) to dianion (35) or a stepwise reduction sequence mediated by a radical anion.In a related paper,44 consideration is given to (37). However all attempts to deprotonate nortriquinacene (40) failed and this lack of acidity is interpreted to mean that the (38) (39) (40) transition state for ex0 -hydrogen abstraction does not experience significant stabil- izing interaction with the proximate double bonds. Metal-reduction studies also failed to provide evidence for formation of a stabilized carbanion. Further studies are reported4’ on attempts to synthesize dodecahedrane (41) and of the successful synthesis of the pentasecododecahedrane derivative (42) the most highly conden- sed precursor of (41) known.Elaboration of dodecahedrane from (42) appears to require development of a dehydrative retro-Baeyer-Villiger sequence which will not trigger transannular reactions and ultimate closure by threefold C-C bond formation. The first synthesis of two vinylidenedicyclopropanesdoubly halogenated in their allylic positions has been together with their intramolecular coupling to give the dispiroheptanes (43) and (44). By cyclopropanation using the Simmons- Smith reagent (43) can be quantitatively transformed into [3]rotane and the higher rotanes (45) can all be prepared by way of (44). In an accompanying paper4’ (45) n =4,5 or 6 4.1 L. A. Paquette H. C. Berk C. R. Degenhardt and G. D. Ewing J. Amer. Chem SOC.,1977,99,4764. *’ L. A. Paquette M.J. Wyvratt 0.Schallner D. F. Schneider W. J. Begley and R. M. Blankenship J. Amer. Chem. SOC.,1976,98,6744. 46 L. Fitjer Angew. Chem. Internat. Edn. 1976 15 762. 47 L. Fitjer Angew. Chem. Internat. Edn. 1976.15 763. Alicyclic Chemistry 295 by the same author a method permitting synthesis of each of the higher rotanes (45) is reported which is based upon a closed homologation sequence and which can in principle be repeated without limit. 3 Stereochemistry The 'Hn.m.r. spectrum of a solution of phenylcyclopropane has been in a CS2 solution and the spin-spin coupling constant between the para-proton of the phenyl group and the a!-proton in the cyclopropyl moiety found to be -0.23 f 0.02Hz. The authors conclude that this indicates a two-fold barrier to internal rotation of 8.4 f1.3kJ mol-' and implies that the low-energy conformation is such that the C-H bond prefers the plane of the phenyl group.This result is consistent with other theoretical and experimental data. Silver-ion assisted methanolysis of the epimeric tricyclic derivatives (46) and (47) has been studied4' in an attempt to provide insight into the stereochemistry of the cyclopropyl cation predicted by calculation to be planar. The results suggest that a planar encumbered ion is not formed and also exclude extensive participation of unsaturation in the transition state during the solvolysis of (47). A study of the effect of increasing pressure on (46) (47) the ring inversion of cyclohexane reports" that pressure rises promote an increase in inversion rate.In qualitative terms this can be explained by assuming that the twist or boat conformations adopted by the molecule during ring inversion have slightly smaller molar volumes than the energetically more favourable chair forms. Results are reported51 of ab initio calculations involving cis- 1,4-di-t-butylcyclo- hexane made in an attempt to decide whether a chair or boat form of this molecule is the more stable. By employing standard single-determinant molecular orbital theory on the STO-3G minimal basis set level the total energies have been calculated and a boat emerges as the most stable conformer. Similar calculations show that in trans- 1,2-di-t-butylcyclohexane both the diaxial chair and diaxial boat are of importance in its ground state.SCF-INDO-FPT calculations have been performed for various geometries of the flat cyclohexa-1,4-diene molecule and the values of 'Jcis/'JR,,,s found'* to remain within the relatively narrow range 1.22- 1.29 Hz. This result is claimed to preclude the need to postulate a rapidly inverting cyclohexa-174-diene ring. Similar on [2,3,4,5,6-2H5,carboxy-'3C]-1,4-dihydrobenzoic acid (48) [2,3,4,5,6,7,8-2H7,carboxy-'3C]-1,4-dihydro-1-48 W. J. E. Parr and T. Schaefer J. Amer. Chem. SOC.,1977,99 1033. 49 D. B. Ledlie W. Barber and F. Switzer Tetrahedron Letters 1977 607. H.-D. Liidemann R. Rauchschwalbe and E. Lang Angew. Chem. Internat. Edn. 1977,16,331. M. Askari D. L. Merrifield and L. Schafer Tetrahedron Letters 1976 3497. 52 P.W. Rabideau J. W. Paschal and J. L. Marshall J.C.S. Perkin ZI 1977 842. 53 J. L. Marshall L. G. Faehl C. R. McDaniel jun. and N. D. Ledford I. Amer. Chem. SOC.,1977 99 321. 296 A. Cox naphthoic acid (49) and [carboxy-13C]-9 10-dihydro-9-anthroic acid (50) show that the ratio Jcq-eq/Jax-eq(for JHH)decreases and the ratio Jax,.ax/Jax-eq (for J,,) increases in the series (48)-(50) indicating that the puckering in (49) is inter-mediate between that of (48) and (50). This leads to the identification of three DmD D D D \ D \ / H' 'D D H' 'D HH (48) (49) (50) categories 'flat' 'flattened boat' and 'true boat'. The results of calculations based on a potential function derived using vibrational datas4 to determine conformations of cycloheptane oxepan and 1,3-dioxepan are compareds5 with n.m.r.results and with rotational Raman spectra. This leads to a consistent picture of the seven- membered rings with twist-chairs as the lowest-energy conformations. The con- formation of 1-dimethylphosphono-1 -hydroxycycloheptane has been determined56 in the solid state by X-ray analysis and it has been found that both the twist-chair and chair conformers occur in the crystal structure. The relative abundances as determined by a refinement of occupancy factor are 0.93 and 0.07. Values of energy barriers associated with conformational ring inversion processes occurring in certain trans -cycloalkenes have been by analysis of their tempera- ture-dependent 13Cn.m.r. spectra. In cis trans-cyclo-octa- 1,5-diene (5 1) and the derived compounds (52) these barriers are estimated to be 146.4 173.6 188.3 (51) (52) X =0,CH2,or NPh and 186.2 kJ mol-' respectively.An investigation of the 251 MHz'H and 63.1 MHzI3C n.m.r. of cis cis-cyclonona-l,5-diene extending over the tempera- ture range -5 to -170 "C has been carried outs8 and the data interpreted in terms of a pair of chiral chair-like conformations with C2symmetry. A I3C n.m.r. study has been reported5' in which the thermodynamic parameters for equilibration of the chair-chair conformational enantimorphs of cis-10-methyldecalin-9-01 and cis-10-methyldecal-1-one have been determined. N.m.r. spectroscopy has also been used to determine the conformational preferences of a number of mobile ring systems including a series of 9-R-9,lO-dihydrophenanthrenes(R =Me CN But COMe C02Me C02Et OH or SiMe3) as well as two related 5,6-dihydro- chrysenes with special regard to the location of substituents in pseudoaxial or " D.F. Bocian and H. L. Strauss J. Amer. Chem. SOC.,1977,99 2866. " D. F. Bocian and H. L. Strauss J. Amer. Chem. SOC.,1977 99 2876. s6 G. I. Birnbaum G. W. Buchanan and F. G. Morin J. Amer. Chem. SOC.,1977 99 6652. 57 S. G. Davies P. F. Newton and G. H. Whitham J.C.S. Perkin 11 1977 1371. 58 F. A. L. Anet and I. Yavari J. Arner. Chem. SOC.,1977,99,6496. 59 J. W. Blunt J. M. Coxon N. B. Lindley and G. A. Lane Austral. J. Chem. 1976 29,967 Alicyclic Chemistry 297 pseudoequatorial positions. With the exception of CN all the substituents are found6' to adopt the pseudoaxial conformation preferentially.Ab initio energies have been reported61 for several minimum-energy conformations using standard single-determinant MO theory with an STO-3G minimal basis set for a number of compounds including trans-decal-2-one lO-methyl-trans-decal-2-one la-methyl-trans-decal-2-one 3a -methyl-trans-decal-2-one and 1,l ,lo-trimethyl- trans -decal-2-one. The calculations predict a significant population of flexible states for the latter two compounds at moderate temperatures. The results of calculations carried out using a 1973 force field on the 25 isomers of dodeca-hydrophenanthrene are presented6* in a report which lists the values of AH; for each conformation together with the values of AH for each compound (con- formational mixture) and the relative entropies.4 Structural Properties and Orbital Interactions In order to account for the fact that substituents on cyclopropane are known to alter the bond lengths in the ring ab initio wavefunctions have been obtained63 for cyclopropane and a number of relevant related molecules. Comparison of orbital composition energies and overlap populations of the parent with substituted molecules has been utilized to ascertain the origin of these substituent and ring- closure effects. The analysis shows that the substituent induced electron with- drawal and charge redistribution in the cyclopropane orbitals so that the adjacent bonds were shortened while the opposite bond was lengthened. Although no cases are known in which simple cyclopropyl cations have been unambiguously demon- strated as intermediates in solvolysis reactions by use of ion cyclotron double- resonance techniques it has been that cyclopropene forms a C3H5+ ion by proton transfer from dimethyloxonium ion.Such experiments suggest that the C3H5+ ion produced has a heat of formation of 996 kJ mol-'. This value is not consistent with the heat of formation of the ally1 cation and suggests that a cyclopropyl cation may have been formed. The surfaces of the ground state and first excited singlet and triplet states for the conversion of cyclopropene into vinylcarbene have been using the MIND0/3 semi-empirical MO method. In general the results appear to be in agreement with the known thermal and photochemical reactivities of cyclopropenes and the results of some ab initio calculations on these species.Generalized valence-bond calculations on cyclo- propene and vinylmethylene have been carried out.66 These predict that (a) the allyl-type .rr-system has a methylene-like triplet but a 1,3-biradical-like singlet state (b) the lowest singlet state of vinylmethylene is of the form of a singlet methylene and (c) the ground state of vinylmethylene should exist in two geometrical forms. These results suggest that the ring opening of cyclopropene proceeds directly to a biradical planar intermediate. The reaction of tri-t-butyl-6o R. G. Harvey P. P. Fu and P. W. Rabideau J. Org. Chem. 1976,41 3722. 61 M. Askari N. S. Ostlund and L. Schafer J. Amer. Chem. SOC.,1977,99 5246.62 H. Honig and N. L. Allinger J. Org. Chem. 1977,42 2330. 63 C. A. Deakyne L. C. Allen and V. W. Laurie J. Amer. Chem. SOC.,1977,99,1343. 64 D. H. Aue W. R. Davidson and M. T. Bowers J. Amer. Chem. SOC.,1976,98,6700. 65 J. A. Pincock and R. J. Boyd Canad. J. Chem. 1977,55 2482. 66 J. H. Davis W. A. Goddard and R.G. Bergman J. Amer. Chem. SOC.,1977,99,2427. 298 A. Cox cyclopropene with photochemically generated t-butoxyl radicals has been moni- t~red~~ using e.s.r. It is found that the magnitude of the 13Ccoupling constants is inconsistent with (53) involving an aromatic .rr-radical and is in agreement with (54) involving equilibrium of energetically equivalent a-radicals. The l-isocyano- R R R R R RboR =RbR eRX R (54) 2,2-diphenylcyclopropyl anion has been found68 to be configurationally stable at -72 "C in contrast to the corresponding 1-cyano-derivative.This stability is not affected by changing the cation from Li to Na to K nor is it affected by change in solvent or by the addition of co-ordinating agents such as crown ethers triglyme or HMPA. A series of 1-methylcyclopropylcarbinyl cations has been prepared and the 'H and 13C n.m.r. spectra inve~tigated~~ under stable-ion conditions. The parent ion has been found to be best represented in terms of a dynamic equilibrium involving significant contributions from both the 1-methylcyclobutyl cation and the set of three equivalent 8-delocalized 1-methylcyclopropyl cations a conclusion supported by ab initio MO calculations at the split-valence-shell 4-3 1G level.Ab initio calculations of the equilibrium structure of cyclobutane have been reported7' in which use is made of a single-determinant restricted Hartree-Fock theory involving an extended basis set of Gaussian orbitals augmented by polariza- tion functions. The molecule is found to be non-planar with a degree of puckering of 0.23 8,and a barrier to planarity of 3.8 kJ mol-'; tilting of the methylene groups proves to be essential for relieving steric strain. Semi-empirical SCFMO cal-culations (CNDO/I CNDO/II and MIND0/3) have been used7' to estimate the contribution of 1,3-carbon-carbon interactions to the strain energy of cyclobutane and perturbation energies are found to lie between 85.4 and 140.2 kJ mol-' depending upon the particular calculation and molecular geometry.The analyses suggest that cognisance of 1,3-carbon*arbon interactions is obligatory in the cyclobutane series. Cyclobutadiene continues to attract interest. The energy rela- tionships between the singlet and triplet states of cyclobutadiene have been dis- cussed'* in terms of MIND0/3 calculation and ab initio 'SCF calculations using the 4-31G basis set; calculations have also been made of molecular vibration frequen- cies. The results support the suggestion that the (CH) species obtained by pho- 6' K. Schriener and A. Berndt Angew. Chem. Internat. Edn. 1976,15 698. "M.P. Periasamy and H. M. Walborsky J. Amer. Chem. Soc.,1977,99,2631. 69 G.A. Olah R. J. Spear P. C. Hiberty and W. J. Hehre J.Amer. Chem. SOC.,1976.98 7470. 'O D. Cremer J. Amer. Chem. SOC.,1977,99 1307. N. L. Bauld and J. Cessac J. Amer. Chem. SOC.,1977,99 942. 72 M. J. S. Dewar and A. Komornicki J. Amer. Chem. SOC.,1977,99 6174. Alicyclic Chemistry tolyses in matrices at low temperatures is a metastable triplet. The fluxional isomerism of cyclobutadiene has been analy~ed~~ at the rr-electronic level in terms of an orbital correspondence analysis in maximum symmetry which shows that the process is allowed under a skew-planar distortion. Analysis of the ground-state eigenvector indicates that the most significant structure is the singlet biradical which is consistent with cycloaddition reactions of 1,3-di-t-butylcyclobutadiene.A full spectral characterization of tetrakis(trifluoromethy1)cyclobutadiene has been The compound has a singlet ground state and if the assignment of the i.r.band at 1700 cm-’ to the C=C stretching vibration proves to be correct then its geometry is not square but rectangular. This study also shows that at low temperature direct irradiation does not lead to the formation of the tetrahedrane system at least as a major course of reaction. In another comm~nication,~~ tri-t-butylcyclobutadiene (59 a molecule for which a rectangular structure is assumed has been shown to undergo reactions typical of a triplet molecule. Thus if (55) prepared by matrix isolation of 2,3,4-tri-t-butylcyclopentadienone,is allowed to react after thawing with an excess of CC14 the product is found to contain no dimer of (55) but exclusively the adduct (56).A possible mechanism for the formation of (56)is shown (Scheme 12). A number of substituted cyclobutadiene dications have Scheme 12 been prepared under stable-ion conditions and st~died’~ by ‘H 19F,and 13Cn.m.r. spectroscopy. The 13C n.m.r. spectra of the phenylated cyclobutadiene dications indicate that the aromatic 277-electron cyclobutenediylium ring interacts mesomerically with phenyl rings and accepts a significant amount of rr-electron density from them. The question as to whether the n.m.r. spectroscopic obser- vations of cyclobutadiene dications correspond to several species in dynamic equil- ibrium or to static non-equilibrating dications has been systematically investigated and unequivocal evidence for the latter presented.The photoelectron spectrum of bicyclo[4,1,1] octa-2,4-diene has been Based upon assignment of the first bands in the spectrum it is concluded that the interaction between the Walsh orbitals of the cyclobutane ring and the olefinic moiety can be described by a resonance integral p = -183 kJ mol-’. The e.s.r. spectra of the radical anions of 1,4-and 1,5-dimethylcyclo-octa- tetraene have been meas~red’~ and it is found that in the former compound the rr-spin density is spread almost equally over the eight ring carbons whereas in the 73 E. A. Halevi F. A. Matsen and T. L. Welsher J. Amer. Chem. SOC.,1976.98,7088. 74 S.Masamune T. Machiguchi and T. Aratani J. Amer. Chem. Soc.,1977,99,3524. 75 G.Maier and W. Sauer Angew. Chem.Zntemat. Edn. 1977,16,51. 76 G. A. OIah and J. S.Staral. J. Amer. Chem. Soc. 1976 98 6290. 77 R.Gleiter P. Bischof W. E. Volz,and L. A. Paquette J. Amer. Gem. Soc. 1977,99,8. 78 J. H. Hammons C. T. Kresge and L. A. Paquette J. Amer. Chcm. Soc. 1976,98 8172. 300 A. Cox latter compound it is localized on the four odd-numbered ring carbons. It is argued that these results provide support for symmetry-orbital models of substituted COT radical anions. The structures of the 1,4-dimethyl- 1,3,5,7-tetramethyl- 1,3,5,7- tetraphenyl- and sym -dibenzo-cyclo-octatetraenedications prepared by oxidation of suitably substituted cyclo-octatetraenes using SbF in S02CIF solution have been in~estigated'~ by 'Hand I3Cn.m.r. spectroscopy. The observations reported indicate that these ions possess those characteristics expected of a C8 67r aromatic system.Calculations using MIND0/3 have been reported*' for a number of potential homoaromatic molecules including (57) and (58) and the homocon- jugate interactions in the cyclobutyl-fused compounds studied were shown to be negligible whereas (57) appears to be homoaromatic. Judging the compounds investigated as a whole the conclusion is drawn that the energetic impact of monohomoaromaticity is only profound for the cyclobutenyl and homotropylium cations. Metallation routes have been announced" to the trilithium salts of the cycloheptatrienyl trianion its n-butyl derivative and its linearly conjugated analogue the heptatrienyl trianion. Among the evidence given is the appearance of the e.s.r.signal expected for cycloheptatrienyl dianion radical dilithium (aH= 3.52 and uLi= 0.83 G) and a singlet in the 'H n.m.r. and also in the I3Cn.m.r. spectra. By use of the newly synthesized bicyclobutanes (27) in which the bridge- head carbons are connected to pyridinium rings and positions 2 and 4 to substi-tuents of different size quantitative data have been provided82 concerning the effect of Ir-substituents and geometry on the LUMO energies in bicyclobutanes. Electrochemical oxidation transforms the 1,3-bisrnethylenecyclobutane(26) into (27) which in turn is reduced back to (26) and the evidence gained from cyclic voltammograms indicates a large decrease in the energy of the LUMO in (27) into which the two electrons are added as compared with that required for reduction of normal bicyclobutanes.The U.V. spectrum of bicyclo[ l,l,O]butane has been examineds3 and the band occurring near 50 000 cm-I assigned to a transition between bonding and antibonding 'T-like' central C-C bond orbitals. The lowest electronic state (A)has been examineds4 with a resolution of 0.3 cm-' permitting the rotational band contours for the vibronic components to be observed. Analysis of the band contours has shown the first excited state to have A2 symmetry. A number of spiroconjugated hydrocarbons have been studieds5 theoretically and by use of a modification within the MIND0 framework an orbital splitting is obtained which is within 5% of experiment in the case of spiro[4,4] nonatetraene. The 79 G.A. Olah J. S. Starel G. Liang L. A. Paquette W. P. Melega and M. J. Carmody J. Amer. Chem. SOC.,1977,99 3349. W. L. Jorgensen J. Amer. Chem. SOC.,1976,98,6784. 81 J. J. Bahl R. B. Bates W. A. Beavers and C. R. Launer J. Amer. Chem. SOC.,1977,99,6126. M. Horner and S. Hiinig J. Amer. Chem. SOC.,1977,99 6122. 83 K. B. Wiberg G. B. Ellison and K. S. Peters J. Amer. Chem. SOC.,1977 99 3941. 84 K. B. Wiberg K. S. Peters G. B. Ellison and F. Alberti J. Amer. Chem. SOC.,1977,99 3946. M. D. Gordon T. Fukunaga and H. E. Simmons J. Amer. Chem. SOC.,1976,98,8401. Alicyclic Chemistry conclusion is also drawn that the MIND0 technique using properly corrected overlap can yield realistic predictions on the magnitude of ‘through space’ inter- actions.The orbital energy differences for a number of spiro-compounds of type (59) and for the analogues in which the spiro-centre is replaced by a four-membered ring (60) have been calculated.86 This shows that the interaction between two mutually perpendicular n-systems linked by a spiro-centre becomes considerably stronger on replacing the spiro-centre by a four-membered ring and this is considered to occur by the Walsh orbitals of the four-membered ring acting as a ‘relay’. 5 Reactions Metal-promoted Reactions.-Methylenecyclopropane has been catalytically tri- merized using Nio compounds modified by trialkylphosphines to give a mixture of three open-chained and three cyclic CI2Hl8isome~s.~’The composition of the mixture depends on the nature of the alkyl groups attached to the phosphorus and a suggested route to one of the products (61) is shown in Scheme 13.Exclusive e- Qlx Scheme 13 [2u+ 27r] cycloaddition of methylenecyclopropane with alkenes has been found” on Pdo catalysts where in contrast to the Nio-catalysed reactions the three- membered ring is opened between C-2 and C-3. The catalytic cycloadditions 86 P. Bischof R.Gleiter and R. Haider Angew. Chem. Zniemaf. Edn. 1977 16 110. 13’ P. Binger A. Brinkmann and J. McMeeking Annalen 1977 1065. P. Binger and U. Schuchardt Angew. Chem. Internal. Edn. 1977 16 249. 302 A. Cox described might proceed via palladacyclobutanes of type (62) and palladacyclo- hexanes of type (63). Vinylcyclopropanes have beeri foundR9 to undergo a novel R" R2 epimerization in the presence of dicarbonylrhodium(1) chloride dimer and this has been shown to be coupled with trans-cis isomerization about the C-C r-bond.In related work kinetic studies involving rearrangement of l,l-dimethyl-2-vinyl-cyclopropane and other compounds provide evidence in support of a mechanism for ring cleavage involving initial co-ordination of the vinyl group with Rh' and stereospecific cis-P-hydride elimination from a subsequent metallocyclic allyl- rhodium(II1) alkyl intermediate. Methylenecyclobutane and its 3-substituted derivatives are reported" to undergo a Pd"-catalysed ring expansion to the cor- responding cyclopentanones in good yield. Analogous but somewhat less efficient reactions are observed for camphene methylenecamphor and methyl-enecyclopentane.The transformation has been found to be solvent sensitive yields being highest in benzene but falling off with increasing solvent polarity. An examination has been made91 of the silver(1) perchlorate-catalysed rearrangement of [l-2H]benzvalene to [2Hl]benzene in which rupture of two diagonally opposed bonds of the bicyclobutane moiety occurs. By following the reaction with the 2H-('H} FTn.m.r. technique it is found that the benzvalene deuterium label is scrambled by a catalysed degenerate rearrangement in competition with the aromatization. Together with labelling studies the observations provide unequivocal evidence that the reaction of the Ag' ion with the cyclobutane moiety is a reversible process involving rupture and then reclosure of a C-C bond.A study has been made92 of the heats of metal-catalysed rearrangements of some small-ring hydrocarbons in an attempt to rectify the lack of experimental ther- mochemical data. In particular the isomerizations of quadricyclane to norbor- nadiene and of tricycle[$ 1,0,02v7]heptaneto 3-methylenecyclohexene have been studied. The synthesis of anti-tetramantane (64) a C22H28 hydrocarbon having the regular topology of the diamondoid lattice has been achieved,93 in which the key step is a ring expansion+yclization reaction of a polycyclic diene (65) in hydrogen in the gas phase on a platinum-silica catalyst. The mode of expansion of the four-membered ring in the [2+2] dimers of adamantane in the gas phase on platinum has also been in~estigated~~ and the results support the view that loss of a single bridgehead hydrogen atom from the alkane is sufficient to initiate rear- rangements.This is to say that essentially covalently bonded alkyl groups rear- 89 R. G. Salomon M. F. Salomon and J. L. C. Kachinski J. Amer. Chem. SOC.,1977,99 1043. 90 P. Boontanonda and R. Grigg J.C.S. Chem. Comm. 1977 583. 91 U. Burger and F. Mazenod Tetrahedron Letters 1977 1757. 92 K. B. Wiberg and H. A. Connon J. Amer. Chem. SOC.,1976,98,5411. 93 W. Burns T. R. B. Mitchell M. A. McKervey J. J. Rooney G. Ferguson and P. Roberts J.C.S. Chem. Comm. 1976,893. 94 W. Burns M. A. McKervey J. J. Rooney N. G. Samman J. Collins P. von R. Schleyer and E. Osawa J.C.S. Chem. Comm. 1977 95. Alicyclic Chemistry (64) (65) range on a platinum surface in a manner commonly observed with carbonium ions.This represents an alternative pathway to the formation and further reaction of aah -triadsorbed intermediate (66) and is presumably followed as the aah Me Me fi -Pt-Pt-(66) mechanism is ruled out on simple geometric grounds. It has been that (diphenylcarbene)pentacarbonyltungsten without added co-catalysts is an effective initiator of metatheses of numerous cis-cycloalkenes converting them into polyalkenamers 90% or more of whose double bonds are cis. Cyclobutene cyclopentene cyclohexene cycloheptene cyclo-octene and norbornene are all transferred in this way. In an accompanying paper,96 (phenylmethoxy- carbene)pentacarbonyltungstenis also reported as being able to bring about meta- theses of cyclobutene and norbornadiene to the corresponding polyalkenamers.The importance of this discovery is that although ways have yet to be found to replace the stabilizing substituents in the metal carbenes such as (phenyl-methoxycarbene)pentacarbonyltungstenin order to transform them into initiators for olefin metathesis it emerges that this carbene is itself an effective catalyst. The photochemical decomposition of a number of substituted metallocyclobutane derivatives of the general form [(75-C5H5)2WCH2CHR2CHR'] have been studied9' and the major olefin formed on decomposition is shown to have one carbon less than the appropriate carbon chain of the initial metallocyclobutane. It is suggested that the initial photochemical reaction causes a q5-C5Hs to-q3-C5Hs ring shift and that this 16-electron intermediate allows the rearrangement of the metallocyclo- butane ring to occur.Equilibria of the type shown in Scheme 14 have been postulated as the central step in the olefin dismutation reaction and these results provide strong support for this mechanism. It has been found98 that l-methyl- trans-cyclo-octene undergoes metathesis to yield a polymer that within the limits of the detection method is perfectly alternating. Thus to the extent that this reaction is a valid measure the selectivity for equation (1)is >50 times that for equation (2) (Scheme 15). The experiments also show the stereochemistry of trisubstituted olefin metathesis (E-olefins yield mainly E-products) and indicate 95 T.J. Katz S. J. Lee and N. Acton Tetrahedron Letters 1976 4247. 96 T. J. Katz and N. Acton Tetrahedron Letters 1976 4251. 97 M. Ephritikhine and M. L. H. Green J.C.S. Chem. Comm. 1976,926. 98 S. J. Lee J. McGinnis and T. J. Katz J. Amer. Chem. SOC.,1976 98 7818. 304 A. Cox 16-electron 18-electron 1 R’ = R2 =R3 = R4 = H; R’ = R3=H R2 = R4 = Me; or R’ =Me R2 = H R3 = Me or H R4 = H or Me Scheme 14 the presence of the metathesis initiator at the ends of the polymer chains. A model has been proposedg9 to account for both the observed metathesis and the cis-trans isomerization as measured by the relative amounts of the various olefins produced at low conversion from the reaction of cis-pent-2-ene on a tungsten catalyst.A stereochemical model is proposed based on the belief that an olefin co-ordinated to a metallo-carbene unit passes through a four-centre transition state to form a new olefin and a new metallo-carbene. 99 J. L. Bilou J. M. Basset R. Mutin and W. F. Graydon J.C.S. Chem. Comm. 1976 970. Alicyclic Chemistry Thermally Induced Reactions.-A laminar flow reactor has been described'" which is useful for gas-phase kinetic measurements on small amounts of compounds; the vinylcyclopropane rearrangement and the retro-Diels-Alder reaction of dicyclobutadiene were the reactions used in test runs to demonstrate the reliability of the results. Light has been cast on the stereochemistry of the cyclo- pentene-forming reaction by reports'01 of a study of the thermal isomerization of (+)-(1S,2S)-trans,trans-2-methyl-l-propenylcyclopropanefor which the figures shown in Scheme 16 have been obtained.It is suggested that there is competition among four concerted processes two allowed and two forbidden with the possible intervention of a planar .rr-vinylcyclopropane intermediate having an orbital struc- ture appropriate for disrotatory ring closure to racemic (67). A study has been made1O2of the pyrolysis of 6-exo-vinylbicyclo[3,1 ,O]hex-2-ene in order to deter- mine whether isomerization to the 6-endo-isomer occurs through a one-centre epimerization at C-1 and C-5. The results show that the one-centre epimerization pathway is followed and for the first time this is demonstrated in a cyclopropane for which the alternative two-centre option is geometrically possible.Investigation of the thermal decomposition of bicyclic divinyl- A'-pyrazolines and divinylcyclo- propanes has that identical product decomposition patterns arise from (68) and (69) (Scheme 17). Comparison of the kinetic parameters for decomp- osition of (68) with those for trans-3,5-dimethyl-A1-pyrazoline and 3-vinyl-A'- pyrazoline suggests a contribution towards a lowering of the activation enthalpy \ Scheme 17 loo L.-U. Meyer and A. de Meijere Chem. Ber. 1977,110 2545. lo' G.D.Andrews and J. E. Baldwin J. Amer. Chem. SOC.,1976,98,6705. J. E.Baldwin and K. E. Gilbert 1.Amer. Chem. SOC.,1976,98,8283. '03 M. P. Schneider and B. Csacsko J.C.S. Chem. Comm.1977,330. 306 A. Cox consistent with formation of a diallylic 1,3-biradical and this is taken as supporting evidence that the thermal rearrangement of trans -divinylcyclopropane proceeds through a biradical intermediate. In order to attempt to define the mechanism for thermal geometrical isomerizations of cyclopropanes the rates have been meas~red"~ for three distinct degenerate processes shown by the cyclopropyl moiety of (Z)-l-methyl-2-deuteriospiro[2,4]hepta-4,6-diene.The results show that even in this favourable case the first cyclopropane constituted to enhance prospects for one-centre epimerization through a planar intermediate epimeriza- tion at C-3 is definitely not prominent and may well be absent. Studies"* of the thermal rearrangement of methylenecyclobutane indicate that electronic factors are least likely to be the cause of the observed sterepspecificity in this reaction and they also reveal a similarity between the reaction of cyclopropane arid cyclobutane with that of methylenecyclobutane in that all pass through a biradical transition state (or intermediate) which is not bound by large rotational barriers.Experi- ments have been reported *06which are claimed to demonstrate unambiguously 'hot molecule' effects at an atmosphere of nitrogen and exclusive operation of the C-1 -C-4 bond-cleavage mechanism in the transformation of bicyclo[2,1,0]pent-2-ene to cydopentadiene. It is suggested that the assumption that collisional de- activation in solution always pre-empts thermal reactions of vibrationally excited molecules can no longer be credited.The first experimental evidence has been reportedlo7 which demonstrates that stabilization of norcaradiene relative to cyclo- heptatriene can be achieved by replacement of hydrogen at C-7 with a wdonor. The equilibria (70)*(71) (Scheme 18) have been established and the time- averaged values of &,.6 in (70a) r,(71b) (4.45) and (70b) r* (71b) (4.88) indicate that in contrast to (71c) (5.74) substantial quantities of the norcaradiene form are present in both mixtures. From n.m.r. data can be obtained KZ5.,[(70a) a(71b)] -0.7 [41%(71a)l and K25aC[(70b)* (7lb)]-0.3 124% (71b)l. This establishes that a piperidino-group stabilizes the norcaradiene form relative to the cycloheptatriene PF Ph Ph Ph (70) a; R = 1-piperidino b; R = cyclohexyl c; R = €I Scheme 18 form slightly more than does the sterically similar cyclohexyl group.It is suggested that HOMO-LUMO interactions between v-electron donors and a cyclopropyl ring whether in a perpendicular or a bisected conformation must cause stabiliza- lo4 K. E. Gilbert and J. E. Baldwin J. Amer. Chem. SOC.,1976,98 1593. lo' W. W. Schoeller J. Amer. Chem. SOC.,1977,99 5919. G. D. Andrews and J. E. Baldwin J. Amer. Chem. SOC.,1977,99,4853. lo' S. W. Staley M. A. Fox and A. Cairncross J. Amer. Chem. SOC.,1977 99 4524. Alicyclic Chemistry tion not destabilization. An investigation has been undertaken"' of the thermal rearrangements of spiro[4,4 Jnonatetraene (72) and the related less unsaturated analogues spiro[4,4]nona- 1,3,6-triene (73) spiro[4,4]nona- 1,3,7-triene (74) and spiro[4,4]nona- 1,3-diene (75).Low activation barriers are observed for (72) and (73) relative to (74) and (75) and a transition state is favoured which implicates the w-system at the migrating carbon for (72) and (73). The transition state (76) for a WQQm (72) (73) (74) (75) simple sigmatropic shift of a carbon unit attached to a cyclopentadiene ring is appropriate for reactions of (74) and (75) but may be modified in the cases of (72) and (73) owing to interactions of a porbital on the migrating carbon (LUMO of an olefin or 1,3-diene unit with the cyclopentadienyl r-system HOMO). This HOMO-LUMO interaction (77) would stabilize the transition states leading to the (76) Orbital interaction during 15 (77) Secondary n-orbital interaction alkyl shift during 1,5-vinyl shift low barriers to rearrangement for (72) and (73) as compared with (74) and (75).Thermal isomerization'Og of cyclopropylallene at 380 "Cusing a contact time of ca. 2 minutes leads to a mixture of five components of which two 1,4-dimethylcyclohexa- 1,3-diene and pxylene cannot be explained without invoking a facile 1,7-hydrogen migration The thermolysis of bicyclo[2,2,0]hex-2-ene has been approached'".by generating a complete list of mechanistic alternatives both plausible and otherwise and then by experiment rigorously excluding as many as possible. By this method there is found no pericyclic alternative to the symmetry-forbidden least-motion mechanism shown in Scheme 19.The kinetic distribution of isomeric xylenes formed on thermal aromatization of dl-and meso-1,l-dimeth~l-3,3'-bicyclopropenyls and Scheme 19 M. F. Semmelhack H. N. Weller and J. S. Foos J. Amer. Chem. SOC.,1977,W. 292. lo9 D. E. Minter and G. J. Fonken Tetrahedron Letters 1977 1717. M.J. Goldstein R. S. Leight and M. S.Lipton J. Amer. Chem. Soc. 1976,98. 5717. 308 A. Cox of l,l'-dimethyl-3,3'-bicyclopropenylhas been determined."' The data are most consistent with a mechanism involving initial cleavage of one of the cyclopropane rings followed by expansion of the other ring closure to a Dewar-benzene and finally opening of the Dewar intermediate to form aromatic products. A report has appeared112of a study of the kinetic and thermochemistry of the valence iso- merization of benzvalene to benzene.The transformation is found to be essentially non-chemiluminescent and the valence isomerization may be viewed as a symmetry- forbidden [,2+,2] process and as such it may be supposed that a biradical mechanism is involved. 'I1 J. H. Davis K. J. Shea and R. G. Bergman J. Amer. Chern. Soc. 1977,99 1499. N.J. Turro C. A. Renner and T.J. Katz TerrahedronLettks 1976,4133.