12 Alicyclic Chemistry By A. COX Department of Molecular Sciences Unjvemjty of w8rwick Coventry CV4 7AL 1 Synthesis Three-and Four-membered Rings.-A review containing 142 references on the chemistry of cyclopropanols has appeared.’ The use of carbenes in the synthesis of cyclopropanes has been further developed. For example’ irradiation of fluorodi-iodomethane at 350nm in the presence of an olefin trapping agent provides a synthetically useful one-step procedure for the preparation of a variety of monofluorocyclopropanes; yields as high as 45% have been obtained. The monofluorocyclopropanes are formed stereospecifically and this is in agreement with the view that the reaction involves formation of the fluorocarbene followed by addition of the carbene to the olefin rather than addition of the fluoroiodomethyl radical followed by ring closure with loss of iodine atom.The zinc bromocarbenoid reagent prepared from diethylzinc and bromoform in the presence of oxygen also leads to conversion of olefins into the corresponding monobromocyclopropanes in good yield.3 Styrene greatly inhibits the formation of the zinc bromocarbenoid reagent and polymerizes under the reaction conditions suggesting a radical mechan- ism for the formation of the carbenoid. Carbenes have also been shown4 to be intermediates in the reduction of p-substituted aromatic aldehydes by zinc in the presence of boron trifluoride several of which cycloadd to C=C. A synthesis of fused methylenecyclopropanes employing a carbenoid ring-contraction of bicyclic cyclobutanonetoluenesulphonylhydrazoneshas been announced.’ Dehalogenation of the adduct (1)with Zn-AcOH gives the corresponding cyclobutanone.Pyrolysis of the dry Li salt of the derived toluenesulphonylhydrazonegives the methylenecyc- lopropanes (2) and (3). These are found to exhibit an extraordinarily clean thermal equilibrium. Vinylcyclopropanation of olefins by means of the copper-catalysed D. H. Gibson and C. H. DePuy Chem Reu. 1974,74,605. J. L.Hahnfeld and D. J. Burton,TetrahedronLeners 1975,1819. S. Miyano,Y.Matsumoto and H. Hashimoto J.C.S. Chem Comm 1975,364. I. Elphimoff-Felkin and P. Sarda Tetrahedron 1975,31 2785. A.S.Kende and E. E. Riecke. J.C.S. Chem. Comm. 1974,383. 285 286 A. Cox reaction with vinyldiazomethane is a short and convenient synthesis of vinylcyclo- propanes.6 With cis- and trQns-but-2-enes the vinylcyclopropanation is stereo- specific.In contrast to cyclopropanations of cyclohexane using either ethyl diazo- acetate or trimethylsilyldiazomethane,both of which give almost exclusively the exo-isomer the copper-catalysed method leads to the sterically more congested endo-isomer. Similarly the syn-isomer is preferentially formed in the reaction of vinyldiazomethane with vinyl acetate. The preparation of cis-divinylcyclopropane has been reported7 by photolysis of either cis- or trans-3,5-divinyl- 1-pyrazoline at -45°C. The product consists of a mixture of the cis-and the trans-divinylcyclopropane. The cis-isomer does not rearrange to any great extent below 0 "C but at +11"C it rearranges to cyclohepta-1,4-diene.Formation of cyclic ketens by treatment of an acid chloride with a tertiary amine fails in the case of cyclo- propane probably owing to an increase in I-strain in going from the cyclopropyl to the cyclopropylidene structure. However preparation of carbonylcyclopropane has now been reported* by pyrolysis of 2,2-dimethyl-l,3-dioxan-4,6-dione-5-spirocyclopropane (Scheme 1). At room temperature carbonylcyclopropane gives the dimer (4). Scheme 1 Although 'normal' alkenes such as cyclohexenes are known not to react with diazomethane-palladium(I1) acetate in ether strained alkenes such as bicyclo[2,2,l]heptene bicycIo[2,2,l]heptadiene and hexamethylDewarbenzene reactg with this reagent to give the corresponding cyclopropane in good yield.There is a characteristic difference" between an organocopper-phosphine complex and an organosodium intermediate in the stereochemistry of cyclopropane formation from an a-chloro-compound and an cyp-unsaturated cyano-compound. Using Buf;P-CuOBu' the cyclopropane derivative is readily formed (Scheme 2). In the HCCl(Me)CO,Me B~~u~~~Bu; Bu;P,CuCCl(Me)CO,Me JCH,=CM~CN MUM' + MuC02Me C0,Me CN I -CICu,PBu; I MeC-CH -CMe I I CN C0,Me CN Me c1 CuPBu; Scbeme 2 6 R. G. Salomon M. F. Salomon and T. R. Heyne J. Org. Chem 1975,40,756. M. P. Schneider and J. Rebell J.C.S. Chem. Comm. 1975,283. * G.J. Baxter R. F. C. Brown F. W. Westwood and K. J. Harrington Terruhedron Letters 1975,4283.J. Kottwitz and H. Vorbriiggen Synthesis 1975,636. T. Tsuda F. Ohoi S. Ito and T. Saegusa J.C.S. Chem. Comm. 1975,327. Alicyclic Chemistry organosodium system the cis :trans ratio is increased with decrease in the polarity of the solvent. Assuming that the carbanionic group of the organocopper-phosphine complex firmly co-ordinates to copper it has been suggested that intramolecular cyclization will occur without chelation involving copper producing the more favoured trans-isomer due to non-bonded interactions between substituents. Decar- boxylation of substituted lactones using NaCl in DMSO at 160 "Cyields cyclopropyl ketones (Scheme 3).11 Support for the proposed mechanism is provided by the Scheme 3 observation that the product is 4-benzoylbutyronitrile when NaCN is used.Fluoride- ion-promoted elimination of @-halogenosilanes is a powerful method for generating strained alkenes and it has now been shown'* that 1,l-dichloro-2-trimethylsilylcyclopropane reacts with caesium fluoride at 80 "C to give 1-chlorocyclopropene in 54% yield. 3-Alkoxycyclobutanones are readily accessible by [2 +21 cycloaddition of ketens to enol ethers. During column chromatography on neutral or basic alumina these adducts eliminate13 ethanol and afford pure cyclobutenones in yields as high as 80%. A potentially general synthesis (Scheme 4) of cyclobutenones e.g. (6),starts from the FS03CH3) 'l:::, )+!+0 flo +Go To MeS II c MeS Me$ /\ Scheme 4 readily available compound (5).14 Keteniminium salts derived from a-chloroenamines have been used for fabricating four-membered rings.'s For exam- ple [2 +21 cycloaddition of tetramethylketeniminium tetrafluoroborate or trichlorozincate to alkynes leads to cyclobutenylideneammoni~m salts from which in favourable cases the cyclobutenones can be obtained by hydrolysis using NaOH. A simple synthesis of dimethylenecyclobutene has appeared (Scheme 5).l6 Glow discharges normally convert acetylene completely into polymeric material but under suitable conditions17 the products include vinylacetylene diacetylene benzene phenylacetylene styrene indene and naphthalene which probably arise by the initial dimerization of acetylene to cyclobutadiene. l1 S. Takei and Y. Kawano Tetrahedron Letters 1975,4389.l2 T. H. Chan and D. Massuda Tetrahedron Letters 1975,3383. l3 H. Mayr and R. Huisgen Angew. Chem. Internat. Edn. 1975,14,499. l4 T. R. Kelly and R. W. McNutt Tetrahedron Letters,1975,285. Is C. Hoornaert A. M. Hesbain-Frisque and L. Ghosez Angew. Chem. Internat. Edn. 1975,14,569. l6 H.-D. Martin S. Kagabu and H. J. Schiwek Tetrahedron Letters 1975 3311. G. Rosskamp and H. Suhr Annalen 1975 1478. Q 0 LiAlH4 Scheme5 A reactive intermediate may be generated'' from an insoluble polymer-bound precursor and trapped by a second solid phase suspended in the same reaction mixture viz. preparation of a polymer-bound o-phenanthroline derivative of cyc- lobutadieneiron carbonyl and a polymer-bound maleimide derivative. Oxidation of the former in the presence of the latter is shown to proceed with 96% transfer of free cyclobutadiene between the two polymers.These results support the conclusion that oxidation of the iron carbonyl complexes generates free cyclobutadiene and indicate that ions derived from cerium are not required as vehicles for cyclobutadiene transport. Irradiation of tetrafluorocyclobutene-3,4-dicarboxylicanhydride (7) in the vapour phase at 253.7 nm and in the presence of 500 mmHg of N2leads via tetrafluorocyclobutadieneto octafluoro[4,2,0,02~5]cyclo-octa-3,7-diene (8) which thermally isomerizes to octafluorocyclo-octatetraene(9) (Scheme 6).l9 Irradiation F8 FF Scheme6 J. Rebek and F. Gavina J. Amer. Chem. Soc.,1975,97,3453. l9 M.J. Gerace D. M.Lemal and H. Ertl J. Amer. Chem. SOC.,1975,97,5584. Alicyclic Chemistry 289 of (7) in the presence of furan yields compound (lo) in addition to (9). The substituted cyclobutadienes (1 1)and (12) are stable at room temperature for 48 and Et2N#sph Ph NEt, Et2Nnph PhS NEt (1 1) (12) 1h respectively.20 Cyclopropa[4,5]benzocyclobutene (1 3)21,22 is the first compound in which a benzene ring is annelated by both three- and four-membered rings. Bicyclobutylidene has now been prepared (Scheme 7).23 Reagents i H,NNH,-H,S -20 "C;ii Pb(OAc),; iii Ph,P; iv Wittig reaction. Scheme7 Six-,Seven- and Eight-membered Rings.-Cyclohexadienes should in principle be available from starting materials supplying two and four carbons respectively. More specifically an enolate should add to the terminal double bond of a butadienylphos- phonium salt to produce stereoisomeric ylides.If these are in equilibrium the Z-isomer should undergo an intramolecular olefin synthesis to afford the cyclohexa- diene (Scheme 8). A number of bicyclic cyclohexadienes have been thus made.24 Cyclo-octa-2,5,7-triene- 1,4-dione has been synthesized (Scheme 9).25Photolysis of the sodium salt of 2,3-homotropone toluene-p-sulphonylhydrazoneleads to tricyclo[8,6,0,02~9]hexadec-2,5,7,1 From this result and the 1,13,16-he~aene.~~ structure of the Diels-Alder adduct with cyclic dienes it was concluded that the ** R. Gompper S.Mensch and G. SeyboId Angew. Chem. internat. Edn. 1975,14 704. z1 D. Davalian and P. J. Garrett J.Amer. Chem. Soc. 1975 97,6883. 22 C. J. Saward and K. P.C. Vollhardt Tetrahedron Letters 1975,4539. 23 L. K.Bee J. Beeby J. W. Everett and P. J. Garrett J. Org. Chem. 1975,40,2212. 54 G. Buchi and M.Pawlak J. Org. Chem. 1975,40,100. z5 M.Oda Y. Kayama H. Miyazaki and Y. Kitahara Angew. Chem. 1975,87,414. 26 M. Oda Y. Ito and Y. Kitahara Tetrahedron Letters 1975,2587. 290 A. Cox -PRe(),+Q PPh3 PPh PPh, , 0' R R Scheme8 Zn-AcOH @ -0 @ (l0 0 0 Scheme 9 highly reactive species participating in the reactions is a new strained allene cyclo- octa- 1,2,4,6-tetraene which must have been formed by rearrangement of bicyclo[5,1,0]octa-3,5-dien-2-ylidene initially generated in the photolysis. The homocyclo-octatetraene dianion has been prepared by dimetallation of (2,2,2)-cyclonona- 1,3,6-triene using n-butyl-lithium in TMEDA.*' The 'Hn.m.r.coupling constants provide evidence regarding the geometry of the anion ruling out both a norcaradiene-like structure and the angle-strainless boat conformation (14) and suggesting that the eight carbons are not coplanar. H (14) Large Rings.-The reaction of allylic halides with Ni(CO) permits the synthesis of large-ring methylene cycloalkanes (Scheme 10).**Cyclization normally proceeds in high yield but curiously fails completely for the formation of 12-membered rings. Br& +H2h -k-i"H2h Br Scheme 10 27 M. Barfield R. B. Bates W. A. Beavers I. R. Blacksberg S. Brenner B. I. Mayall and C. S. McCulloch J. Amer. Chem.Soc. 1975,97,900. E. J. Gxey and P. Helquist Tetrahedron Letters 1975,4091. Alicyclic Chemistry 291 Polycyclic Compounds.-The optically active Dewar-benzene (16) has been pre- pared29 from the menthyl bornyl or s-octyl phenylacetylenic esters and the cyc- lobutane (15) followed by hydrolysis. Evidence has been presented3' that butalene Me Me Me (15) (16) (17) can be generated and trapped below room temperature. The starting material is 3-chlorobicyclo[2,2,O]hexadiene which although quite stable to most strong bases can be dehydrohalogenated using LiNMe in HNMe,-THF. Judging from the typical free-radical reactions of (18) in the gas phase and the reactions observed for (17) the prediction that (17) and (18) have a separate existence and different Iz[] \ 0 (17) (18) chemistry is confirmed.The vigorous conditions required to generate butalene support the conclusion that such fused ring systems are not strongly stabilized by their overall content of (4n+2) w-electrons. A convenient route to truns-tricyclo[5 1,0,0274]oct-5-ene (trans-bishomobenzen) allowing access to multigram quantities has appeared.31 A route from dicyclopentadiene to trishomocubanone has been described.32 Photocyclization of dicyclopentadiene leads to 1,3-bishomocubane which with [Rh(CO),Cl] undergoes carbonylation to trishomocubanone via a rhodium-containing complex. By analogy with some other acylrhodium compounds this is assumed to have a C1-bridged polymeric structure. Despite earlier failures tetra-asterane pentacyclo[6,4,0,02~7,04~11,05~'']dodecane has been ~ynthesized.~~ Iceane (20) has been prepared34 from the olefinic ketone (19).This highly symmetrical hydrocarbon (&) possesses five six-membered rings (19) (20) of which because of the rigidity imposed by the carbon skeleton two exist in chair conformations and three in non-twist boat conformations. The 'H n.m.r. spectrum (90 MHz) shows an AM pattern centred at S 0.94 and 1.90 (J= 12 Hi),consistent *9 J. H. Dopper B. Greijdanus D. Oudman and H. Wynberg J.C.S. Chern. Grnrn.,1975,972. 30 R.Breslow J. Napierski and T. C. Clarke J. Arner. Chern. Soc. 1975,97 6275. 31 R.T.Taylor and L. A. Paquette Angew. Chern. Internat. Edn. 1975,14,496. 32 J. Blum C. Uotogorski and A. Zoran Tetrahedron Letters 1975 1117.33 H. M.Hutmacher H.-G. Fritz and H. Musso Angew. Gem. Internat. Edn. 1975,14,180. 34 D.P.G. Hamon and G. F. Taylor Tetrahedron Lerters 1974,155. 292 A. Cox with the resonances due to the axial and equatorial protons respectively of each methylene group. This difference in chemical shift is probably enhanced by van der Waals deshielding of the equatorial protons due to their position in the rings having a boat conformation. The highly strained cage hydrocarbon hexacyclo[5,4,0,02~6,04~11,05~9,08~10]~ndecane (23)has been prepared35 by reaction of the ditosylate (21) with NaI in HMPA. Control experiments indicated that (23) must be formed via (22). In the field of propellane chemistry the synthesis of 3,3- dibutynylcyclopentanone and its conversion into the [3,3,3]propellane system have been reported;36 this is a key compound from which some equally ring- functionalized propellanes can be derived.(21) (22) (23) A general method3' for the preparation of bicyclo[2,1 ,l]hexenes uses the Ramberg-Backlund rearrangement of 2-chloro-3 -thiabicyclo[ 3,l ,llheptane 3,3- oxides. The synthetic sequence starts from a mixture of cis-and trans-1,3- cyclobutane dicarboxylic acids and because of the number of 1,3-cycIobutane dicarboxylic acids known e.g. the truxinic acids this method is claimed to allow the preparation of a variety of substituted bicyclo[2,l,l]hexenes. A synthesis of the strained but stable bridgehead olefin bicyclo[4,2 llnon- l(8)-ene has been de~cribed,~' involving a Wittig reaction as the key step.Chemical and physical data suggest the compound to be less strained than the isomeric bicylo[3,3,l]non- 1-ene. The utility of the intramolecular Wittig reaction for the preparation of strained bridgehead double bonds has been further inve~tigated~~ in the attempted synthesis of bicyclo[2,2,l]hept-l-ene and bicyclo[3,2,l]oct-5-ene. When the salt (24) was stirred in a refluxing solution of potassium t-butoxide and diphenylisobenzofuran the two isomeric Diels-Alder adducts (25) were formed. This sequence constitutes the first efficient synthesis of bicyclo[3,2,l]oct-5-ene. A similar sequence starting from the cyclopentanone analogue of (24) failed to yield any detectable trace of a Diels-Alder adduct derived from bicyclo[2,2,l]hept-l-ene.0 Ph 35 A. P. Marchand and T.-C. Chou Tetrahedron Letters 1975,3359. 36 J. Drouin F. Leyendecker and J. M. Conia Tetrahedron Letters 1975,4053. 37 R. G. Carlson and K. D. May Tetrahedron Letters 1975,947. 38 K. B. Becker Tetrahedron Letters 1975,2207. 39 W. G. Dauben and J. D. Robbins Tetrahedron Letters 1975 151. Alicyclic Chemistry 293 Biogenetically patterned synthesis in the field of tetra- and penta-cyclic triter- penoids has been re~iewed.~' 2 Stereochemistry The conformational analysis and electronic structure of t-butyl cyclopropyl ketones have been investigated41 within the framework of the CND0/2 approximation. In every case the cis-isomer has been found to be preferred with an unusual conforma- tion of the substituent with respect to the cyclopropane ring.Interactions have also been observed between the oxygen of the HO or Me0 substituents and the carbonyl oxygen atom. An interesting paper the results of a crystallographic analysis of a derivative of cyclobutadiene chosen so that (i) the electronic state is minimally perturbed but yet its reactivity is suppressed by the steric blocking of the approach of another molecule by the substituents thus allowing its isolation and (ii) the substituents are such as to preserve the symmetry in an important way. The compound analysed was methyl tri-t-butyl[4 Jannulene carboxylate and the analysis shows that it is definitely not a square but a slightly distorted rectangle. Ab initio MO theory has been to study the potential surfaces associated with ring puckering in some saturated five-membered ring compounds.A geometrical model is used which has fixed bond lengths and fixed angular conformations for methylene groups but all other degrees of freedom are varied. It is found that theory predicts puckering amplitudes in fair agreement with experimental data and for cyclopentane all methods give nearly free rotation. Cyclopentanone is found to have its energy minimum in the twist form but in this case conformational interconversion occurs most easily along a path passing close to the planar skeleton form as a transition state. Although planar projection formulae have long been employed to represent ring compounds the justification for their use has seldom been satisfactory.Polya's method has been employed44 to show that their use in counting isomers is allowed because the permutation symmetry of such formulae is equivalent to that of a non-planar ring in which interconversion between conformers occurs. Ther-modynamic considerations indicate that the concentration of the twist-boat form becomes quite substantial at high temperatures. By means of a high-vacuum deposition technique in which a high-temperature equilibrium is trapped at 20 K on a CsI plate the i.r. spectrum of the mixture of chair and twist-boat conformations of cyclohexanes has been The twist-boat is present to ca. 25% in cyc- lohexane deposited from 800 "C. Thus the percentage of the twist-boat in the gas phase at 800 "C is at least 25% and possibly more depending on the efficiency of trapping of the twist-boat.The induced shift ratios for several substituted cyclohex- anones have been determined46 using tris(dipivaloy1methanato)ytterbiumand the individual shift data have been used to ascertain some rotamer populations. Exten- sive calculations based on the induced shift data have been used to obtain structures 40 E. E. van Tamelen Accounis Chem. Res. 1975,8 152. 41 F. Crasnier,J. F. Labarre H. Cousse L. D. D. Hinterland and G. Mouzin Tetrahedron,1975,31,825. 42 L. T. J. Delbaere M. N. G. James N. Nakamura and S. Masamune J. Amer. Chem. Soc. 1975,97 1973. 43 D. Cremer and J. A. Pople J. Amer. Chem. Soc. 1975,97 1358. 44 J. E. Leonard G. S. Hammond and H. E. Simmons J. Amer. Chem. SOC.,1975,97 5052.4s M. Squillacote R. S. Sheridan 0.L. Chapman and F. A. L. Anet J. Amer. Gem. SOC.,1975,97,3244. 46 K. L. Servis and D. J. Bowler J. Amer. Chem. Soc. 1975,97 80. 294 A. Cox for the ketone-lanthanide chelate complexes which give the best fit between calculated and observed induced shifts for all the ring and side-chain protons. The configurations of dimethyl cyclohexanone-2,6-dipropionate and diethyl cyclohexane-l,3-dipropionatehave been determined47 by a 13C n.m.r. study. It was concluded that the former compound has a cis configuration and that the latter consists of both cis-and trans-isomers with a molar ratio of cis :truns =3. Ab initio calculations on planar and boat conformations of cyclohexa-l,4-diene have been made.48 Total energies for (26) and (27) were calculated to be -231.5610 and (26) (27) a = 159.3" -231.5491 a.u.respectively. This corresponds to (26) being 30.96 kJ mol-' more stable than (27) and the highest occupied r-level formally the SS combination of the two olefinic r-orbitals has CQ. 91.4 kJ higher energy in both conformations than the orbital represented by the SA combination. A relatively complete calculation of the low-energy conformations of cyclohep- tane that the pseudorotation paths of cycloheptane can be visualized as occurring on a helical track wound around a torus. Comparison of the results obtained for the equilibrium conformations of cycloheptane with results of other calculations is made and qualitative agreement found. The pseudorotation of cycloheptane has been described" by means of the Fourier expansion of the endocyclic torsion angle.The topography of the pseudorotational itineraries for the chair/twist-chair and the boat/twist-boat families was also discussed. Cyclo-octa- 1,5-diene and the corresponding dibenzo derivatives have been studieds1 by means of force-field calculations. The twist-boat form is found to be the most stable conformation in both the diene and dibenzo molecules with a conforma- tional preference over the chair of 6.07 and 5.44 kJ mol-'. The twist-boat form is characterized by a w234S of 37.9" for the diene and 26.5" for the dibenzo derivative. Differences between the parent-diene and the dibenzo derivative are noted and are attributed largely to the difference in torsional barriers of a saturated carbon bond when attached to a double bond as opposed to being attached to an aromatic ring.13C n.m.r. data and iterative strain-energy calculations have been announced;'* they support the view that the presence of eclipsing strains in the C-6-7-8-1 moiety of the symmetrical boat conformation of cyclo-octa- 1,3,5-triene render this conformation unfavourable. The measurements show that a distortion to give a twist-boat relieves these strains. A paper has appeareds3 giving the results of force-field calculations of the strain energies of nine cyclodeca- 1,6-diene conformers. The difference in steric energy 47 N. Matsurnoto and J. Kumanotani Tetrahedron Letters 1975 3643. 48 G. Ahlgren B. Akermark and J.-E. Backvall Tetrahedron Lemrs 1975,3501.49 D. F. Bocian H. M. Pickett T. C. Rounds and H. L. Strauss J. Amer. Chem. Soc. 1975,97,687. 50 W. M. J. Flapper and C. Romers Tetrahedron 1975,31,1705. 51 N. L. Allinger and J. T. Sprague Tetrahedron 1975,31,21. s2 F. A. L. Anet and 1. Yavari Tetrahedron Letters 1975,4221. 53 D. N. J. White and M. J. Bovill Tetrahedron Letters 1975 2239. Alicyclic Chemistry between the preferred cis,cis-and trans,trans-isomers was calculated to be 12.34 W mol-' in excellent agreement with the observation that only the former is detectable when the two isomers are equilibrated; it was predicted that truns,trans- cyclodeca-1,6-diene should consist solely of conformer (28) in the gas phase. (28) The 'H and 13C n.m.r. spectra of cyclohexadecanes4 are temperature dependent below cu.-100°C. The 13C n.m.r. spectra of [1-13C]cyclohexadecane show the presence of three resonances in the ratio 1:2 :1at -152 "C. Iterative strain-energy calculations indicate that the [4444] conformation is 7.95 kJ mol-' more stable than the [3535] conformation and it is concluded that cyclohexadecane exists predomin- antly in the [4444] conformation with a barrier to pseudorotation of 28.0* 0.84kJmol-'. The 25.2MHz 13C n.m.r. (FFT) spectra of 1-cyanobicyclo[ l,l,O]butane have been reported," both with and without hydrogen decoupling. The chemical shifts are discussed in terms of dipolar resonance con- tributions to the ground state and the unsaturated nature of the bicyclobutane ring. 3 Structural Properties and Orbital Interactions The protonation of cyclobutane has been studieds6 theoretically using ub initio methods.The protonation energy is found to be 527 kJ mol-l but is significantly lower than that obtained for the protonation of cyclopropane by similar theoretical calculations. The protonation energies for edge and corner protonation of cyc- lobutane were found to be essentially identical. MIND0/3 calculations have been carried outs7 on cyclobutadiene by the 'half-electron' method. The heats of forma- tion of cyclobutadiene triplet and singlet (29) and (30) respectively are as follows singlet [(30) rectangular] AHf 394.9 kJ mol-l CC bond lengths 1.533 1.342 A; triplet [(29) square] AH 419.6 kJ mol-' CC bond lengths 1.433 A. A square (29) (30) singlet structure for cyclobutadiene is predicted to lie well above the square triplet (54.8 kJ mol-').The intersection of the singlet and triplet surfaces therefore lies above (29) in energy and it is concluded that conversion of (29) into (30) should require activation. Using an electron monochromator-mass spectrometer combina- tion the ionization potentials of cyclopentyl cyclopentenyl and cyclopentadienyl radicals produced in pyrolytic reactions have been rnea~ured.'~ The stabilities of s4 F. A. L. Anet and A. K. Cheng J. Amer. Chem.Soc. 1975,97,2420. s5 M. Pomerantz and D. F. Hillenbrand Tetrahedron,1975,31,217. s6 T. Pakkanen and J. L. Whitten J. Amer. Chem. Soc. 1975,97,6337. 57 M. J. S. Dewar and H. W. Kollmar J. Amer. Chem. Soc. 1975,97,2933.5.9 F. P. Lossing and J. C. Traeger J. Amer. Chem. Soc. 1975,97,1579. 296 A. ax the three cations were compared by reference to enthalpies of the general reaction RH + R++H+e and it was concluded that while the introduction of one double bond into the C ring brings about a stabilization of 100kJ the second double bond destabilizes the cyclopentenyl cation by 132 kJ. The cyclopentadienyl cation is consequently destabilized by 31.8 kJ mol-' with respect to the cyclopentyl cation a figure which closely corresponds to the solvolysis rate difference of lo5found by Breslow. An analysis of the structure of the homotropenylium cation has been undertaked9 by the MIND0/3 method. This has lead to a homoconjugate bond length which is better accommodated in terms of (31) as opposed to-(32) in agreement with the Winstein picture of an open cyclopropane unit.Consideration has been given6' to whether a cyclobutane ring (33)is able to replace the traditional cyclopropane ring as (33) a homoaromatic linkage. While not ruling out the possibility of participation of the cyclobutane ring in this way the results suggest that particularly severe conditions will have to be employed in order to force the cyclobutane ring to act as a homoconjugate linkage. This may indicate that the cyclopropane ring occupies an exclusive position in homoaromaticity and may be an essential ingredient for its occurrence. An exception to the generalization concerning the additive nature of the enhance- ment in the rate of solvolysis upon introduction of multiple cyclopropane rings at positions adjacent to an incipient carbocation centre is afforded by the introduction of a second cyclopropane system into the nortricyclyl skeleton to form the quadricyc- lyl system.This puzzle has been explained61 in terms of the combination of the HOMO of cyclopropane with the LUMO of the cyclopropylcarbinyl cation to form the quadricyclyl cation. A large bonding interaction between the carbinyl carbon and the adjacent cyclopropane carbon predominates; however this is substantially counteracted by the two antibonding interactions across the small base of the tetracyclic ring system. The net effect is a small stabilization upon introduction of this second cyclopropane. Two series of substituted ally1 cations have been studied62 59 R.C. Haddon Tetrahedron Letters 1975,863. 6o R. C. Haddon Tetrahedron Letters 1974,4303. 61 L. M. hew and C. F. Wilcox J. Amer. Chem. SOC.,1975,97 2296. 62 G. A. Olah and R. J. Spear J. Amer. Chem. SOC.,1975,97 1539. Alicyclic Chemistry by I3C n.m.r. and 'H n.m.r. spectroscopy under ion-stable conditions. The 1,3-substituted cations (34a-e) exhibit strong charge delocalization between C-1and C-3 whereas the charge in the 1,l-substituted alkenyl cations (35a-e) is substan- tially higher at C-1. 1'3-Overlap does not appear to contribute significantly to the total ion structure. H a; R' = R2 = Me R'yy b; R' = Me; R2 = cyclopropyl c; R' = R2 = cyclopropyl H H d; R' = cyclopropyl;R2 = Ph e; R' = R2 = Ph (34) (35) Acetolysis of the vinyl tosylate (37) gives products closely similar to those from (36) and (38).63 It was concluded that (37) is intimately involved in the solvolyses of (36)and (38)and that all three compounds react via the same intermediate manifold implicating homoallenic participation.products 4 Acetolysis of 3,3-diphenyl- and 3,3-dimethyl-cyclobutyl tosylate proceeds more slowly than that of the unsubstituted cyclobutyl tosylate thereby militating against concerted ring-opening to the allylcarbinyl ion accompanying ionization.64 These and other data are in line with the hypothesis that most 3-substituted cyclobutyl derivatives ionize initially to the cyclopropylcarbinyl ions which then undergo a rapid ring-opening to the corresponding allylcarbinyl ions.Synthesis has been undertaken6' of the 'parallel' and 'perpendicular' isomers of trans trans- bicyclo[6,1,0]nona-4-enes.The properties of (39) are reported to be vastly different from those of the perpendicular isomer (40) or other analogous trans-alkenes and appear to be attributable to the transannular repulsions between the .rr-bond and the cyclopropane ring which distort the .rr-bond and thus greatly alter its chemical reactivity. To probe the effect of the cyclo-octatetraene ring on the stability of an adjacent cationic centre the hydrolysis rate of cyclo-octatetraenylcarbinylchloride in 50% aqueous ethanol has been compared66 with that of A'-cyclo-octenylmethyl chloride. The relative rates of these first-order reactions are such that the cot derivative is five 63 T.Von Lehman and R. S. Macomber J. Amer. Chem. Soc. 1975,97 1531. 64 C. S. Michejda and R. W. Cornnick J. Org. Chem. 1975 40 1046. 65 J. A. Deyrup and M. F. Betkouski J. Org. Chem. 1975,40 284. 66 W. Etching K. A. Henzel and L. A. Paquette J. Amer. Chem. Soc. 1975,97 4643. 298 A. Cox times less reactive at 25 "C. The involvement of a homotropylium ion cannot be dismissed. The barrier of the degenerate valence isomerization in 3,4-homotropylidene has been determined67 by complete lineshape analysis of the 13Cdynamic n.m.r. spectra and the values of physical constants obtained are in good agreement with the results of the 'H dynamic n.m.r. lineshape analysis of the octadeuterio derivative.The influence of a methyl group in the 1-or 3-position is small. The first experimental verification has been offered68 of predictions based on Winstein's concept of homoaromaticity and the refining applications of MO theory of Goldstein and Hoffmann to longicyclic systems of potential bicycloaromaticity that electronic factors in (41) should be reversed in (42). The failure of the attempted reduction of bicyclo[4,2,l]nona-2,4,7-trien-9-oneto bicyclo[4,2,l]nona-2,4-dien-9-one by di- imide was attributed6' to homoaromatic interaction of the electrons on the isolated (41) (42) double bond at C-7-C-8 with the carbonyl at C-9 leading to sharply reduced electron density at C-7-C-8 compared with the diene moiety. The electron density in the diene may even be increased by bis(homocyclopentadieny1) interaction with the carbonyl.Optimized semiempirical MIND0/2 calculations on the transition states of (45)-(48) in the sigmatropic rearrangements of (43) and (44) have been rep~rted.~'Inversion over retention at the migrating carbon centre is only slightly favoured in (47) but retention over inversion is strongly favoured in (48). 5 67 R. Bicker H. Kessler and W. Ott Chern. Ber. 1975,108,3151. 6a J. T. Groves.and K. W. Ma J. Amer. Chem. Soc. 1975,97,4434. 69 D. I. Schuster and C. W. Kim J. Org. Chem. 1975,40,505. 'O W. W. Schoeller J. Amer. Gem. SOC. 1975,97 1978. Alicyclic Chemistry The pentacyclotetradecane rearrangement graph culminating in diamantane (49) has been analy~ed.~~ The most probable mechanistic pathways from structures (50)-(56) were deduced by graphical analysis guided by empirical force-field calculations./ \ (54) exo-trans-exo (53) (55) em-trans-endo (56) endo-trans-endo The synthesis of spiro[3,4]octa-l,5,7-triene has been carried out7’ in anticipation that the combination of spiroconjugative interaction in the 7r-system and direct a-7r overlap between the four-membered ring and the diene would produce interesting spectral and chemical properties. However in the event the U.V. spectrum showed a broad featureless maximum at h =262 nm almost identical with that observed for the analogous compound saturated in the four-membered ring. An investigation has been of the degree of interaction between Walsh orbitals of a cyclopropyl group and the 7r-orbitals of a neighbouring double bond in dispiro[2,2,2,2]deca-4,9-diene (57) as compared with that in spiro[2,4]hepta-4,6-diene (58) and dispiro[2,0,2,4]deca-7,9-diene (59).Their photoelectron spectra indicate that spirocyclopropyl groups as in (57),can transmit the electronic effect between the two n-bonds in a cyclohexa-1,4-diene far better than C-C a-bonds as in (58),and the C-H a-bonds as in cyclohexa-1,4-diene itself. 71 T. M. Gund P. von R. Schleyer,P. H. Gund and W. T. Wipke J. Amer. Chem. Soc.,1975,!W 743. 72 R.D.Miller and M. Schneider Tetrahedron Letters 1975 1557. 73 P.Asmus M. Klessinger L.-U.Meyer and A. de Meijere Tetrahedron Letters,1975,381. 300 A. Cbx 4 Reactions Metal-promoted Reactions.-Evidence has been for the mechanism of the potentially general method for preparing .rr-allylpalladium chloride complexes from the reaction of cyclopropenes with palladium(I1) chloride in which the results strongly implicate symmetrical w-complex formation as the rate-determining step (Scheme 11).Treatment of 1,l’-trimethylenebicyclopropenylwith AgCIO has been I Ph H + (PhCN),PdCl +[~~~~~~~~ 1 Ph Ph I Ph Scbeme 11 found75 to lead to 1,4-trirnethylene(Dewar benzene) as the only detectable Dewar- benzene (Scheme 12).1-Methoxy-5-vinylbicyclo[3,2,0]hept-6-ene the first exam- ple of a solvent-trapped product of a cation of type (60),is also produced and must be Ag (60) Weme12 74 M. A. Battiste L. E. Friedrich and R.A. Fiato Tetrahedron Letters 1975,45. 75 I. J. Landheer W. H. De Wolf and F. Bickelhaupt Tetrahedron Letters 1975,349 Alicyclic Chemistry considered to be direct evidence for this mechanism. The major product of the oxidative decomposition of (cyclo-octa- 1,5-diene)PdC12 using Pb( OAc) has been to be 6-and not 4-exo-chloro-2-endo-acetoxybicyclo[3,3,O~ctane (Scheme 13). This suggests that a a-bonded complex rather than a 7r-bonded complex is cis- H cis- transfex insertion OAc OAc H OAc OAc cl H scbeme13 formed as the major intermediate and that cis-intramolecular ligand transfer from metal to olefin carbon is considerably faster then the well-established addition of nucleophile to the olefin-Pd" complex under these conditions.The reactions of cyclodeca- 1,2,5,8-tetraene and cyclodeca- 1,2,5-triene with mercuric sulphate in acetic acid have been found to give only rearranged product^.'^ These are the first known examples of rearrangement during oxymercuration of an allene. Compara- tive of the nickel(0)-catalysed reactions of quadricyclane and norbor- nadiene with acrylates suggest that this and the nickel(0)-catalysed isomerization of quadricyclane to norbornadiene involve a common intermediate in the product- determining step (Scheme 14). Paquette has examined the Ag'-catalysed valence isomerization of substituted 1,8-bishomocubanes. It is established" that of the four different rearrangements leading to two isomeric snoutane diesters the distal isomer (61) is formed preferen- tially.The bond-switching process is considered" to be triggered by electrophilic 76 S.-K.Chung and A. I. Scott Tetrahedron Lerrers 1975,49. 77 R.W.Thies P.-Kong Hong R. Buswell and J. L. Boop J. Org. Chem. 1975,40,585. 78 R.Noyori I. Umeda H. Kawauchi and H. Takaya J. Amer. Chem. Soc. 1975,97,812. 79 L.A.Paquette and R. S. Beckley J. Amer. Chem. Soc. 1975,97,1084. L. A. Paquette R. S. Beckley and W. B. Farnham J. Amer. Chem. Soc. 1975,97,1089. 302 A. Cox Ni Ni Ni L/\L L" 1 Z Scheme 14 R (61) attack at the C-1 -C-5 bond with direct formation of a delocalized cyclopropylcar- binyl cation. Such edge argentation accounts for all available kinetic data and conforms to theoretical conclusions that attack by Ag+ at an edge bond is energeti- cally preferable to bidentate co-ordination with one of the cubyl surfaces.To probe the effect of such groups as phenyl vinyl cyclopropyl and ethoxy on the rate of Ag+-catalysed homocubyl rearrangements the isomerizations of 4-substituted homocubanes were compared8' with those of model compounds where direct resonance interaction in the transition state was not anticipated. The bond switch- ings appear to proceed by Ag+-induced electrophilic ring-opening of the less substituted edge bonds. The rearrangements of several 1,8-bishomocubane deriva- tives by various Rh' and Pd" catalysts are sensitive not only to the nature of the ligands attached to the metal but also to the degree of substitution at the reaction site.82 The effect of varied 4-substitution of the homocubane ring system on the course of Rh'-and Pd"-catalysed rearrangements has been probed83 by direct 81 L.A. Paquette,J. S. Ward R. A. Boggs,and W. B. Farnham J. Amer. Chem. Soc. 1975,M 1101. 82 L. A. Paquette,R. A. Boggs,W.B. Farnham and R. S. Beckley,J. Amer. Chem.Soc.,1975,97,1112. g3 L. A. Paquette R. A. Boggs and J. S. Ward J. Amer. am.Soc. 1975,97,1118. Alicyclic Chemistry examination of product ratios and kinetics. The evidence suggests that the ring- opening reaction proceeds through an intermediate (62) (Scheme 15) where the transition metal has entered into oxidative addition in the rate-determining step. (62) scheme 15 A complex series of light-induced valence isomerizations of bullvalene and related CloHlohydrocarbons with Fe(C0)' in moist benzene has been reported.84 ThermallyInduced Reactions.-It has been concluded8' from an investigation of the thermal stereomutation of optically active trans-[ 1,2-2H2]cyclopropane that a double methylene rotation mechanism operates (Scheme 16) i.e.the reaction occurs == pp D D A DD Scheme 16 via a trimethylene in which the three ring carbons and four of the hydrogens occupy a common plane. In a related papers6 it has been reported that a synchronous double rotation is also prominent in the stereomutation of l-phenyl[2-2H]cyclopropane. The ring-opening of 2,3-diphenylcyclopropyl radicals to 1,3-diphenyldlyl radicals has been inve~tigated.~~ Product analyses and the dependence of the yields of the products of cage recombination on solvent viscosity as well as kinetic measurements demonstrate that ring-opening and cage recombination are competing reactions.The results are interpreted in terms of a favoured disrotatory ring-opening. Thermal rearrangement of each of the four isomeric ethyl 2-methyl-3-(trans-propeny1)cyclopropanecarboxylates has been reported8* to result in cis-trans isomerization to the other isomers and the formation of a number of rearrangement products. The results are rationalized on the basis of a model involving C-C bond cleavage rotations around bond a and/or b in the biradical formed (Scheme 17),and closure to cyclopentenes with stereochemical discrimination occurring during the final two steps. Stereochemical dissection of the 4-cyano- 1-methylcyclopentene 84 R.Aumann Chem. Ber. 1975,108,1974. J. A. Berson and L. D. Pedersen J. Amer. Chem. SOC.,1975,97 238. 86 J. A. Berson L. D. Pedersen and B. K. Carpenter J. Amer. Chem. Soc. 1975,97,240. S. Sustmann and C. Riichardt Chem. Ber. 1975,108,3043. P. H.Mazzocchi and H. J. Tarnburin J. Amer. Chem. Soc. 1975,97,555. 304 A. cox HH rotation + Et0,C eH3H CH3CH3 -1". stereorandom biradical CH3 CH3 % Et0,C H& H rotation ' Et0,C CH Scheme 17 obtained from thermal rearrangement of cis-and trans-1-cyano-2-isopropenylcyclopropane shows8' partial retention of optical activity owing to a predominance of retention of configuration from cis and of inversion of configura- tion from trans.Pending quadrisection of the rearrangement a tentative description in terms of a not obviously concerted reaction involving rotational preference within the framework of the continuous biradical hypothesis is given. Cyclopropylallene has been foundg0 to rearrange above 300 "Cto 3-methylene-l-cyclopentene,and thermolysis of the deuterium-labelled compound (Scheme 18) shows that a 4-5 times faster geometric isomerization accompanies the rearrangement. Scheme 18 The thermal rearrangement of 1,1'-dimethylbicyclopropenyl to 2,2'-dimethylbicycloprop-2-enyl is claimedg1 to be the first experimental realization of a Cope rearrangement of bicyclopropenyls. At higher temperatures two isomers were formed. Presumably one is the dl-isomer originating from the more stable chair transition state and the other the rneso-isomer originating from a boat transition state.This may have implications for the discussion surrounding the concerted or radical nature of the Cope rearrangement. 89 W. von E. Doering and K. Sachdev J. Amer. Chem Soc. 1975,97,5512. 90 W.R. Roth T. Schmidt and H. Humbert Chem. Ber. 1975,108,2171. 91 W. H.de Wolf I. J. Landheer and F. Bickelhaupt TetrahedronLetters 1975 179. Alicyclic Chemistry The semiempirical MIND0/2 method has been used to explain9* the experimental stereochemistry in the thermal rearrangements of methylenecyclobutane. Retention of configuration is shown to be due to ‘subjacent orbital control’ and antarafacial participation of the ally1 unit due to Jahn-Teller instability in the biradical transition state.Preliminary results have been reported93 on the thermal reorganizations of 3,3-divinylmethylenecyclobutane and 3-vinylmethylenecyclobutane which undergo smooth thermal conversion into (63) and (64) respectively. Rate studies coupled with simple calculations on the ther- mochemical kinetics of the system combined with kinetic isotope results are (64) Scheme 19 claimed to provide convincing evidence for a two-step (biradical) mechanism as opposed to a concerted pathway. MIND0/3 calculations have-been reported94 for the thermal rearrangements of some bicyclo[2 1,O]pent-2-enes to cyclopentadienes. Of particular interest is the case of the unsubstituted molecules in which the transition state is predicted to have a symmetrical structure (65) implying that the transformation is a normal concerted 1.43 8 (65) electrocyclic process in spite of the fact that it violates the Woodward-Hoffmann rules.The rearrangement to the acetylcyclopentene (68) has been showng5 to occur 92 W. W. Schoeller Chem. Ber. 1975,108 1285. 93 W. R. Dolbier and G. J. Mancini Tetrahedron Letters 1975 2141. 94 M. J. S. Dewar and S. Kirschner J.C.S. Chem. Comm. 1975,461. 95 J. P. Grosclaude H. U. Gonzenbach J. C. Perlberger and K. Schaffner,J. Amer. Chem. SOC.,1975,97 4147. 306 A. Cox specifically from the endo-acetylbicyclopentane(66);this process and the endo-exo interconversion (66)&(67) do not involve a common intermediate. The 7-40 (67) (66) (68) Scheme 20 stereochemistry of the thermal reorganization of methyl-substituted 6- methylenebicyclo[3,2,0]hept-2-enesinto one another and the four methylated 5-methylenebicyclo[2,2,l]hept-2-enes has been The results are consis- tent with the assumption of a set of concurrent uncatalysed unimolecular rearrange- ments and the transformations are believed to occur by a course involving non- equilibrating biradicals.The thermal behaviour of 3-methylenebicyclo[3,2,l]oct-6-ene (69) and of 3-oxobicyclo[3,2,l]oct-6-ene (70) has been in~estigated.~~ Transformation of the former (69) into 7-methylenebicyclo[3,3,0]oct-2-ene pro-(69) X = CH (70) X = 0 ceeds cleanly via a Cope rearrangement. Thermolysis of (70) occurs only at much higher temperatures apparently indicating a non-concerted process.syn [9-2H]-Bicyclo[6,1 ,O]nonatriene has been synthesized and its rate of epimerization The epimerization proceeds via the intermediate (71) (Scheme 21). I1 D D Scheme 21 96 D. Hasselmann Angew. Gem. Zntemat. Edn. 1975,14 257. 97 J. Japenga M. Kool and G. W. Klumpp Tetrahedron Letfers 1975,1029. 98 C. P. Lewis and M. Brookhart J. Amer. Chem. Soc. 1975,97,651. G. Boche W. Weber and J. Benz Angew. Chem. Zntemat. Edn. 1974,13,207 Alicyclic Chemistry The thermal isomerization reactions of cis-9,lO-dihydronaphthaleneand deriva- tives have been studied.'00 A propensity for degenerate behaviour is established in the parent and the 1,9-products which arise on thermal reorganization of cis-9,lO- dihydronaphthalenes substituted in the 9 and 10 positions arise by [1,5]suprafacial migration of one of four identical trigonal a-carbons.Conversion into 1,5-isomers involves transient intervention of cis5-cyclodecapentaene intermediates. Pyrolysis of the appropriate tosylhydrazone sodium generates syn-and anti-tricyclo[4 1,0,02,4]heptan-5-ylidene (72)and (73) respectively which fragment'" to the highly strained cyclohepta-l,2,5-triene (74). Product ratios show a total lack of dependence upon the stereochemistry of (72) and (73) and this speaks against a synchronous pathway for the formation of (74). Scheme 22 Tricycle[7,3 ,0,04*'2]dodeca-2,5 ,7,10- te traene has been synthesized. lo2 Its H n.m.r. spectrum is unchanged up to 141 "C in agreement with a static structure.Thus Cope rearrangement of this compound is slow on the n.m.r. time-scale and as the geometry might suggest otherwise this slow rate is probably due to the absence of a small ring and its accompanying strain. Thermolysis of syn-and anti-tricycl0[4,2,0,0~~~]octane (75) and (76) leads103 to both cis,cis- and cis,trans-cyclo- octadiene (77) and (78) (Scheme 23) most likely via biradicals sufficiently long- lived to aIlow internal rotation (75) (77) 1 Scheme 23 Synthesis of 4-and S-methylspiro[2,4]hepta-4,6-dienehas been accomplished,'04 and vapour-phase pyrolysis of either provides an equilibrium mixture (60:40)of both. Kinetic analysis and other results appear most adequately accommodated by a stepwise perambulation of the GH4 fragment and a hydrogen atom about the loo L.A. Paquette and M. J. Carmody J. Amer. Chem. SOC.,1975,97,5841. Iol W. R. Dolbier 0.T. Garza and B. H. Al-Sader J. Amer. Chem. Soc. 1975,97,5038. Io2 D. G. Farnum and A. A. Hagedorn Tetrahedron Letters 1975,3987. lo3 H. D. Martin and E. Eisenmann Tetrahedron Letters 1975 661. lo4 R. A. Clark W. J. Hayles and D. S. Youngs J. Amer. Chem. Soc.,1975,97,1966. 308 A. Cox periphery of the cyclopentadiene ring. Although the stepwise process appears operative a competitive concerted [1,5]-sigmatropic methylene migration cannot be ruled out. Reaction of 1,2-bis(trifluoromethy1)-3,3-difluorocyclopropenewith cyclopen- tadiene at -78 "Chas been reported to give two adducts the major one of which has been assigned an ex0 stereochemistry and assumed to be the kinetic product since on gentle heating it progressively isomerues to the minor adduct of endo configuration.However it is claimedlo5 that direct preparation of these compounds by the carbene route using norbornadiene shows the above configurational assignments to be in error and requires them to be reversed. Reaction of cyclo-octatetraene with antimony pentafluoride in liquid SO leads directly to 9-thiatricyclo[3,3,1 ,027s]nona- 3,6-diene 9,9-dioxide by an unprecedented 1,5-~ycloaddition;'~~9-thiabicyclo[4,2,l]nona-2,4,7-triene9,9-dioxide is formed concomitantly and is the thermodynamically more stable C,H,02S isomer. 10.5 C. W. Jefford J. C. E. Gehret and J. Mareda Tetrahedron Letters 1975 823.106 L. A. Paquette U. Jacobson and M. Oku J.C.S. Chem. Comm. 1975,115.