1974 927Stereospecific Ring Contractions of 7-Chlorobicyclo3.2,Ohept-2-en-6-ones and Equilibration Studies of 7-Alkylbicyclo3.2.0hept-2-en-6-onesBy Peter R. Brook,* A. Jonathan Duke, John M. Harrison, and Keith Hunt, Department of Organic Chemistry,7-AI kyl-7-endu-chlorobicyclo 3.2.0 hept-2-en-6-ones (1 b-d) undergo ring contraction readily and stereospecific-ally with aqueous base. Because of conformational factors, the 7-exu-chloro-epimers (2b-d) show an increasingreluctance to undergo ring contraction as the size of the alkyl group increases, and in the case of (2c and d)2-alkyltropones are also formed. Equilibration studies show that for 7-alkylbicycIo3.2.0 hept-Z-en-6-ones, thelarge alkyl group prefers the endu-configuration, whereas in the 2.3-dihydro-derivatives it prefers the exo-con-figuration ; however in both equilibria, energy differences between endu- and exu-epimers are small.The University, Leeds LS2 9JTTHE idea of crosswise approach of the two components,proposed by Woodward and Hoffmann for + 712,cycloadditions,2 when applied to keten-cyclopentadienereactions leads to the prediction that the larger group ofan unsymmetrical keten will be found preferentially inthe endo-position of the product, a 7,7-disubstitutedbicyclo3.2.0 hept-2-en-6-one. Experimentally, this hasbeen confirmed by ourselves1 and other groups ofworkers notably those of Brady and Dreiding.4 Stereo-selectivity has also been found in cycloadditions ofketens to other ole fin^.^Our own investigation consisted of the addition of aseries of alkylchloroketens to cyclopentadiene ; the size1 Preliminary communication, P.R. Brook, J. M. Harrison,and A. J. Duke, Chem. Comm., 1970, 689.R. B. Woodward and R. Hoffmann, Angem. Chem. Intevnat.Edn., 1969, 8, 847.of alkyl group was systematically varied (series a-e).The reactive chloroketens were chosen so that additionoccurred under mild conditions, and they were generatedby dehydrochlorination of the appropriate a-chloro-acidchlorides. In the route to the ethyl- and isopropyl-chloroketens, chlorination of butyryl chloride and 3-methylbutyryl chloride with sulphuryl chloride led toconsiderable a-dichlorination in the former case (20)and some a-dichlorination and @-chlorination (tertiaryH) in the second case.Intermediates of satisfactorypurity were obtained by chlorination (SO,Cl,), then3 W. T. Brady and R. Roe, J . Amer. Chem. SOC., 1970. 92,4618; W. T. Brady and E. F. Hoff, J . Org. Chem., 1970,35, 3733,and references cited in these papers.M. Rey, S. Roberts, A. Dieffenbacher, and A. S. Dreiding,Helv. Chim. Acta, 1970, 53, 417.T. DoMinh and 0. P. Strausz, J . Amer. Chem. SOC., 1970,92,1766928decarboxylation of appropriate monoalkylmalonic acids.The cycloadditions have been fully reported in otherpapers, and we merely summarise our results andanalyses in Table 2. The present paper describes someof the chemistry of the chloroketen adducts.0 0(1 1 I21a ; R = H d ; R = P r ib; R=Me e ; R = P hc; R=Et f : R=ClAssignment of stereochemistry in the chloro-adducts(1) and (2) has normally been made by n.m.r.spectro-scopy; in particular the H-5 signal occurs at lower fieldwhen a 7-exo-chloro-group is present .3 Nuclear Over-hauser effects have also been employed.* After separ-ation of 7-epimeric adducts by chromatography, we*-+H H RJ.C.S. Perkin Iring contraction in excellent yields with 2N-sodiumhydroxide to give 6-exo-alkylbicyclo3. I .O) hex-2-ene-6-embcarboxylic acids (3b-d) equation (i). The stereo-chemistry a t C-6 was readily established by use of aniodolactonisation reaction which demonstrated the cis-relationship of acid group and double bond with respectto the cyclopropane.The epimeric 7-exo-chloro-7-alkyl ketones (2) showedan increasing reluctance to undergo ring contraction asthe size of the alkyl group increased.Whereas (2b)reacted readily to give (4b), the yield was lower for theethyl ketone (2c) and 2-ethyltropone was also formed.8The isopropyl ketone (2d) gave no bicyclohexane-carboxylic acid (4d) although 2-isopropyltropone wasformed; the reaction took a different and more com-plicated c0urse.B This systematic change in ease ofreaction is attributed to increasing steric interactionbetween the alkyl group and the 4-endo-H, in thatconformation in which the chloro-group becomesequatorial equation (ii) .Two further anomalies were noted in the ring con-tractions. The phenylchloro-ketone (2e) did not under-go ring contraction (so giving indirect support to theassignment of the endo-configuration to the large phenylgroup) but instead appeared to give lactonic products(VCO 1763 cm-l) by a ring-opening reaction equation(31 (iii). The products were only characterised spectro-scopically, but more vigorous treatment with base gavethe hydroxy-acids (5), for which satisfactory analyticalfigures were obtained.The monochloroketen adduct (la) did not undergoring contraction stereospecifically, but instead gave anepimeric mixture of bicyclohexane-6-carboxylic acids(3a) and (4a), separable by the iodolactonisation pro-(4) cedure.As the chloro-ketone (la) contained an ' enolis-able ' proton (H-7), it seemed likely that the lack ofstereospecificity was due to epimerisation prior to ringcontraction.Evidence that the C-7 enolate was beingformed in the reaction was obtained by use of NaOD-D,O : the resulting endo- and exo-bicyclohexanecarb-oxylic acids were 86 monodeuteriated and ca. 14dideuteriated. We consider the dideuteriation to ariseby replacement of H-5 as well as H-7 in the ketone (la)before ring contraction. This C-5 enolate anion appearsrather strained, but it has been invoked lo in explanationQ q 2 HH H R( i i IH:2H(iii )obtained chemical evidence of stereochemistry bymaking use of the stereospecific base-catalysed ringcontractions of a-chlorocyclobutanones first observed insimple cases by Conia and his co-workers.6 The 7-endo-chloro-7-alkyl ketones (1 b-d) underwent stereospecificJ . M. Conia and J .Salsun, Accounts Chem. Res., 1972, 5, 37.J. Meinwald, S . S. Labone, and M. S. Chadha, J . Amer.8 W. T. Bradyand J. P. Hieble, J . Amer. Chem. SOC., 1972,94,Chem. SOC., 1963, 85, 582.4278.of cirte-substitution of the 7-halogeno-group withmethoxide ion and other nucleophiles.In view of the foregoing results, a route to the 7-exu-chloro-ketone (2a) by base-catalysed epimerisation of thecorresponding endo-chloro-compound appeared feasible.In O.~N-N~,CO,-D,O, 87 of H-7 had exchanged after1 h at 20" (ring contraction was observed at loo"), butwithout detectable epimerisation. After longer reactiontimes only endo-chloro-derivative was found, but ketol9 P. R. Brook and J . M. Harrison, J.C.S. Chew. Comm., 1972,997.10 (a) D. L.Garin and K. L. Canmalk, J.C.S. Chem. Comm.,1972, 333; (b) P. D. Bartlett and T. Ando, J . Amer. Chem. SOC.,1970, 02, 7518; (c) W. T. Bradyand J. P. Hieble, J . Org. Chem.,1971, 36, 20331974dimer (6) (two stereoisomers by t.1.c.) was also observed.This lack of epimerisation shows that the C-7 enol has astrong preference for protonation from the exo-face andwith fairly weak base (Na,CO,) little or no equilibrationwas being observed. exo-Protonation of the 7-en01 wasconfirmed when partial dechlorination of the 7,7-dichloro-ketone (If) with zinc or triphenylphosphine l2gave only 7-endo-chloro-ketone (la).With triethyl phosphite the dichloro-ketone (If) gavethe enol phosphate (7) by a Perkow reaction.13 At-tempted ester interchange of this with methanolicmethoxide led to a low yield of ring-contracted methylesters related to (3a) and (4a), rather than monochloro-ketones (la and 2a).Milder hydrolysis conditions gaveback starting material.Our failure to epimerise the chloro-ketone led us tosuspect initially that it was the exo-derivative (2a) and(51 (610 PO(OEt12 0( 9 ) (10)the more thermodynamically stable epimer. Howeverthe enndo-configuration was firmly established by X-raycrystallography and n.m.r. studies.l* Ghosez andRoussel have successfully epimerised the 7-endo-chloro-ketone (la) by treatment with triethylamine in benzene,16but the equilibrium mixture only contained 22 ofexo-chloro-epimer. This result is surprising; mostworkers have assumed that the larger group wouldprefer the exo-configuration and that only kinetic controlin the cycloadditions leads to a greater proportion ofepimer with the larger group endo.16We have now extended this observation that a large7-substituent has a slight preference for the endo-l1 Zinc-acetic acid has also been used : for a brief report see M.Rey, U.A. Iliiber and A. S . Dreiding, Tetrahedron Letters, 1968,3683.12 For other examples of dehalogenation with this reagent seeI. T. Borowitz, K. C. Kirby, and R. Wirkhaus, J . Org. Chem.,1966, 31, 4031.l3 This reaction has been reported earlier: S.A. P. 6.706,947/1968 (Chew. Abs., 1968, 70, 106,068).configuration to the alkyl-substituted compounds (8)and (9). Dechlorination of the epimeric 7-chloro-7-alkylbicyclo3.2.0heptenones (1) and (2) with zinc led toa product consisting mainly of one epimer, as in the caseof the dichloro-ketone (If).In the methyl case, n.m.r.confirmed that the major product was the 7-endo-methylisomer (8b) the methyl doublet appearing at higher field(6 0.98) than in the case of the minor product (1.25)owing to shielding by the five-membered ring. Themajor product of dechlorination in the ethyl and iso-propyl series was assigned the elzdo-alkyl structure onlyby analogy with the methyl case, as here the n.m.r.spectra were more complicated. Base-catalysed epi-merisation of these 7-alkylbicycloheptenones (8) and(9) proceeded readily to equilibrium as reported inTable 1. Because of the unexpected preference for anendo-alkyl group the equilibrium in the methyl serieswas approached from both sides to confirm that truethermodynamic equilibrium had been established.Theresults are in line with that reported for the chloro-ketone and so suggest that torsional strain and stericeffects are major factors in all cases; any dipole-dipoleinteraction in the chloro-ketone case appears to operatein the same direction.The dihydro-ketones were readily obtained by catalyticreduction of the double bonds in the epimeric mixtures of(8) and (9), and again these compounds were readilyequilibrated with base. In the dihydro-compounds theexo-alkyl configuration was preferred but not to a greatextent (Table 1). Clearly, the change of C-2 from anTABLE 1Base-catalysed equilibria of 7-alkylbicyclo3.2.0 hept-2-en-6-ones and their 2,3-dihydro-derivativesBicyclohep tanonesdihydro-( 9) - R exo endo exo endoMe 34 66 68 42Et 37 63 73 27Pr' 46 54 86 14Bicycloheptenones dihydro- (S) and(8) and (9) -sp2- to an sp3-hybridised centre leads to increased un-favourable interaction of it with a 7-endo-alkyl group.From Dreiding models of 7-alkylbicycloheptenones, itappears that a planar cyclobutanone ring requires acyclopentene ring fused to it also to be planar.In thissituation unfavourable torsional strains arise, particu-larly for atoms adjacent to the 1,5- and 5,kbonds;again, with planar rings the terminal hydrogen atoms ofthe 7-endo-ethyl or -isopropyl group approach within 1of C-2, directly above the double bond. If the cyclo-butanone ring is bent by moving C-6 in the endo-direction, much of the unfavourable interaction isremoved: the cyclopentene ring adjusts to an envelopeconformation with C-5 out of the plane of the double14 P.R. Brook, A. J. Duke, and J. R. C. Duke, Chem. Comm.,1970, 574.l5 Albert Roussel, D.Sc. Thesis, Universitb Catholique deLouvain, 1970. We thank Professor Ghosez for a copy of thisthesis.l 6 See for example refs. 1, 3, and 4; and also for comments on a' masochistic steric effect,' B. M. Jacobson and P. D. Bartlett,J . Org. Chem., 1973, 38, 1036930 J.C.S. Perkin Ibond containing atoms 1, 2, 3, and 4, and the endo-alkylgroup swings away from the cyclopentene ring, becomingequatorial with respect to the four-membered ring as inconformation (10).Bending of the cyclobutanone ringby movement of C-6 in the exo-direction again brings thecyclopentene ring into an envelope conformation butthe 7-endo-alkyl group, now axial, is brought across thefive-membered ring close to the C-4 methylene group,also axial with respect to the cyclobutanone ring. This1,3-diaxial interaction produces a less favoured situation.For the 7-exo-alkyl series the planar cyclobutanonearrangement is still disfavoured, but, for the two bentcyclobutanones (C-6 moved endo or exo, respectively)the best conformation is not so obvious. The formerreduces torsional strains between H-1 and H-2, H-4(exo) and H-5, and H-4 (endo) and C-6, but at theexpense of increasing the 1,3-diaxial interaction of the7-exo-alkyl group with H-5.The latter conformation(C-6 exo) increases the torsional strains while reducingthe 1,3-diaxial interaction. On balance we suggest thatconformation (10) for the bicyclic system is againpreferred, with the exo-alkyl group axial.EXPERIMENTALDetails of general experimental techniques and theinstruments used have been given in earlier papers.17ChZoro(ethyl)malonic Acid.l8-Sulphuryl chloride (28.35 g,0.21 mol) was added dropwise to a solution of ethylmalonicYields, ratios, and analysesaction (about twice the volume of cyclopentadiene) . Themixture was poured into water, the organic phase waswashed with 2~-sodium carbonate, 2~-hydrochloric acid,and water; removal of the solvent left the crude adductswhich were analysed by n.m.r.and g.1.c. Chromatographyon Kieselgel G (Merck) with benzene-petroleum (b.p. 40-60") as eluant gave good separation of the stereoisomers inproportions agreeing with the results of n.m.r. analysis.The yields and ratios of adducts are reported in Table 2.All adducts had spectroscopic properties in agreement withthose reported. Chromatography was not used where theproduct was homogeneous : the monochloroketen adduct(la) was merely purified by distillation, and the phenyl-chloroketen adduct (2c) after removal of solvent appearedpure; attempts to distil or chromatograph the adductresulted in loss of product.Base-catalysed Ring Contractions of 7-A lkyZ-7-chlorobicyclo-3.2.0 hePt-2-en-6-ones .-General method.The pure stereo-isomer in 2~-sodium hydroxide (about 10-fold excess) wasshaken until the solution was homogeneous, or until nofurther reaction was occurring (t.1.c.). After extractionwith dichloromethane to remove any neutral product,acidification gave the pure acid.(a) The 7-endo-methylchloro-ketone (2b) (1.0 g) after10 min gave 6-endo-methyZbicyc203.1 .O Jzex-2-ene-6-carboxyZicacid (4b) (0.92 g, 97y0), m.p. (from aqueous acetone) 75.5-76" (Found: C, 68.9; H, 7.1. C,H,,O, requires C, 69.6;H, 7-25), vco 1695 cm-l, 6 11.2 (lH, s, OH), 5-85-5-55(2H, m, olefinic), 2-97-1.87 (4H, m, bridgehead and allylicprotons), and 1.01 (3H, s, Me).TABLE 2of chloroketen-cyclopentadiene adductsFound (yo) Required (yo) .- h----, AC58-8561.261.963.563.565-1565.2571-7acid (26.4 g, 0.2 mol) in anhydrous ether (120 ml).Afterthe initial vigorous reaction the mixture was heated underreflux for 3 h and the ether was removed. The chloro-acid,crystallised from AnalaR benzene, had m.p. 109-110"(yield 21 g, 63). When this acid (10 g) had been heatedto 120-125' for 0.5 h, distillation then gave 2-chloro-butyric acid (5.9 g, 81), b.p. 85-86" at 15 mmHg.Chloro (isofirofiyl) malonic A cid.18-Similarly, isopropyl-malonic acid (21.9 g, 0-15 mol) gave the chloro-acid (17.0 g,64), m.p. 104-106" from benzene-petroleum (b.p. 60-80") 3. On decarboxylation at 120-130" 2-chloro-3-methyl-butyric acid was obtained (7.1 g, 93y0), b.p. 98-100" a t15 mmHg, m.p. 25-28".Chloroketen-Cyclopentadiene Cycloadditions .-GeneraZ pro-cedure. Triethylamine (5-10y0 excess over acid chloride)was added to a refluxing rapidly stirred mixture of thea-chloroacyl chloride with a four-fold excess of freshlydistilled cyclopentadiene in sufficient pentane to maintaina mobile suspension of base hydrochloride during the re-* The i.r.spectrum of the crude product indicated a mixtureof 6- and y-lactones, the latter predominating. The structureof the purified product was not established.H c1 ' . C H c1'4.9 24.95 58.9 4.95 24.95.8 22.7 5.85 22.9 61.4 5-75 22.7 ::i: f::E5 63.4 6-45 20.87.3 65.00 7.05 19.3 6-95 19-06-35 71.6 6-4(b) The 7-exo-methylchloro-ketone (lb) (0.5 g) after 10min gave 6-exo-met?zylbicyclo3.1 .Ohex-2-ene-6-carboxyZicacid (3b), m.p.95-96.5" (0.46 g, 96) (Found: C, 69.4;H, 7.15), m/e 138 (M+) and 93 (100; Mf - CO,H),vco 1700 cm-l, 6 11.1 (lH, s, OH), 5-80-5-45 (2H, m,olefin), 2.75-1.53 (4H, extended m, bridgehead and allylicprotons), and 1.33 (3H, s, Me).Treatment of this endo-acid (3b) (0.25 g) with sodiumhydrogen carbonate-potassium tri-iodide yielded a 6-exo-methyliodo-lactone * (0.41 g), m.p. 84-84.5O (from methyl-cyclohexane) (Found: C , 68-65; H, 8.3; I, 48.4. C,H,IO,requires C, 68.6; H, 8.6; I, 48.2y0). The epimeric exo-acid on similar treatment only gave back starting acid.Catalytic hydrogenation (Pt in MeOAc) of the 6-endo-methylbicyclohexene-6-carboxylic acid (4b) gave the corre-sponding 6-endo-unethylbicyclohexane acid dihydro-(4b),m.p.74-76O (from water-5yo acetone) (Found: C, 68.35;H, 8.4. C,H,,O, requires C, 68-6; H, 8.6), Mf 140(looyo); 6 1-15 (3H, s, Me). In a similar way the 6-exo-l7 P. R. Brook and A. J . Duke, J.C.S. Pevkin I, 1973, 1013.The method of R. H. Horn, R. B. Miller, and S. N. Slater, J.1950, 29001974 931methyl epimer gave the 6-exo-methyZbicycZohexane-6-carb-oxyZic acid dihydro-(3b), m.p. 75-76.5" (Found: C,68.65; H, 8-3), M+ 140 (loo), 6 1.26 (3H, s, Me).The 2,3-double bond in the methylchloroketen adductsplayed no part in the stereochemical course of the ringcontraction. Hydrogenation of the 7-endo-methylchloro-ketone (2b) afforded the 2,3-dihydro-compound, not fullycharacterised, but having the expected spectral properties,and this when treated with base gave the ring-contracted6-endo-methylbicyclohexane acid dihydro-(4b).In thesame way, the 2,3-dihydro-7-exo-methyl adduct dihydro-( 1 b) gave the 6-exo-methylbicyclohexane acid dihydro-(3b)l. Stereochemical purity of the acids was >98(g.1.c. analysis of the methyl esters).(c) The 7-endo-ethyl-7-chloro-ketone (2c) (1.0 g) with2~-sodium hydroxide after 45 min gave 6-endo-ethyl-bicycZo3.1 .Ohex-2-ene-6-carboxyZic acid (4c) (0.59 g, 65),m.p. 68-69" (from methylcyclohexane, after sublimation)(Found: C, 70.8; H, 7.85. C,H1202 requires C, 71-0; H,7.85), 6 11-1 (lH, s, OH), 5-75 (2H, m, olefinic), 3.0-2.5(2H, m, allylic CH,), 2.45-1.9 (2H, m, H-1 and H-5), and1.9-0-7 p.p.m. (5H, m, Et). A small neutral fraction gave2-ethyltropone, vmx.1630 cm-l (GO), 6 6.95 (5H, m,aromatic H), 2.55 (2H, distorted q, CH,*CH,), and 1.13(3H, distorted t, CH,*CH,), Lx. (MeOH) 232 (E 21,000) and311 nm (7000).(d) The 7-exo-ethyl-7-chloro-ketone (lc) (0-52 g) after30 min gave in the same way 6-exo-ethylbicycZo3.1.0hex-2-ene-6-carboxylic acid (3c), purified as before, m.p. 53.5-54" (0.44 g, 94) (Found: C, 70.9; H, 7.9), 6 11.07 (lH, s,OH), 5.83-5.44 (2H, m olefin), 3.0-2-4 (ZH, m, allylicCH,), and 2.30-0.80 (7H, remaining H). No neutraltropone was formed in this case. This endo-acid affordeda 6-exo-ethyZiodo-Zactone,* m.p. 53-54" (from aqueousmethanol) (Found: C, 38-95; H, 3-95; I, 45.25. CgH1110,requires C, 38.85; H, 4.0; I, 45.7).(e) The 7-exo-isopropyl-7-chloro-ketone (Id) (0.2 g) after24 h gave 6-exo-iso~ro~yZbicycZo3.1.0hex-2-ene-6-carboxyZicacid (3d) (0.165 g, 93), m.p.68-71.5" (Found: C, 71.9;H, 8.3. C10H1402 requires C, 72.3; H, 8*5y0), 6 10.47 (lH,s, OH), 5.76 and 5.51 (2H, 2 x m, olefin), 2.92-2-4 (2H,In, allylic CH,), 2.10 (lH, m, H-l), 1-67 (lH, m, H-5), and1-08 (7H, m, including apparent t, non-equivalent methylsand CH of Pri).The endo-acid gave a 6-exo-isopro~yZiodo-Zactone, * m.p.56-57' (from methylcyclohexane) (Found: C, 41.5; H, 4-6;I, 43.2. Cl,Hl,I02 requires C, 41-2; H, 4-45; I, 43.4).(f) The 7-endo-isopropyl-7-chloro-ketone (2d) (2-0 g)after 36 h yielded 2-isopropyltropone (0.3 g, 18yo), but theacidic fraction contained no 6-endo-isopropylbicyclohexane-6-carboxylic acid.9 The 2-isopropyltropone was character-ised as 2-amino-7-isopropyltropone, obtained by treatmentwith hydrazine hydrate in ethanol; m.p.88-89' (lit., 88").(g) The 7-endo-chloro-ketone (la) was treated with baseunder various conditions. Thus the ketone (la) (5.83 g)was added dropwise to a stirred solution of potassiumhydroxide (16.0 g) in 70 aqueous dioxan (250 ml).Acidification after 30 min at room temperature, followed byextraction with dichloromethane yielded the bicyclohexene-6-endo- and 6-exo-carboxylic acids (3a) and (4a) (ratiocu. 40 : 60 by g.1.c. analysis of their methyl esters) (4.35 g,86). Iodolactonisation of this mixture in the normalway gave the known 6- and y-iodolactones7 (4.92 g) and the6-exo-bicyclohexenecarboxylic acid (4a) (1-60 g), m.p.* Same footnote as on page 930.77-79' from methylcyclohexane after sublimation (97"and 0.1 mmHg).Treatment of the iodo-lactones with zinc-methanolicammonium chloride and normal work-up gave the 6-endo-carboxylic acid (3a), m.p.92O.'Formation of a similar mixture of ring-contracted acids(3a) and (4a) was observed when the monochloro-ketone wasboiled in 2N-sodium carbonate.Ring Contraction with Sodium Deuteroxide-DeuteriumOxide.-When the monochloro-ketone (la) was added toa rapidly stirred solution of sodium deuteroxide sodium(0.9 g) in deuterium oxide (20 ml) an 88 yield of endo-and exo-acids (3a) and (4a), ratio 81 : 19 by g.1.c. analysis oftheir methyl esters, was obtained. After separation of thetwo epimers by the iodolactonisation procedure, the puredeuteriated endo- and exo-acids were analysed by massspectrometry and compared with undeuteriated material.Although the comparison was hindered by the presence ofan M - 1 peak it was estimated that 88y0 of endo-acidcontained one (non-exchangeable) deuterium atom and13 had two such atoms.For the exo-acid the figures were87 and 16. The error in these values is difficult tojudge but appears to be at least f4.(h) The 7-endo-phenyl-7-chloro-ketone (2e) on treatmentwith 2~-sodium hydroxide gave non-crystalline lactonicmaterial, vmx 1763 cm-1 (C=O), not fully characterised.Vigorous treatment of the phenyl ketone (Ze) (3.00 g)with potassium hydroxide (20 g) in water (30 ml) at the boilgave a dark oil which after bulb-to-bulb distillation IS200" (air-bath) and 0.1 mmHg gave Z-(a-hydroxybenzyZ)-eyelopent-3-ene- 1-carboxylic acid (5) as a gum (diastereo-isomers) (1.85 g, 65) (Found: C, 71-7; H, 6.35.C1,H1,Oarequires C, 71.6; H, 6.4y0), v-. 3700-2400 (OH andCO,H), 1730 (CO), and 1618 and 1610 cm-l (double bondand phenyl group).Dechlorination of the ChZoroketen Adduck.-(a) 7,7-Di-chZorobicycZo3.2.0hept-2-en-6-one (If). The dichloro-ketone(3-0 g), stirred and heated in methanol (15 ml) with zincdust (4 g) and ammonium chloride (1 g), gave after 30 hbicyclo3.2.0hept-2-en-6-one (92y0), b.p. 70-75" at 18mmHg. The semicarbazone had m.p. 226" (from ethanol)Slow passage of the dichloro-ketone (If) (2 g) in methanol(30 ml) and ammonium chloride (1 g) down a column ofzinc wool (1 x 10 in) previously activated with methanolicammonium chloride and set up over refluxing methanolgave 7-endo-chlorobicyclo3.2.0 hept-2-en-6-one ( la) (0.85 g,53), characterised spectroscopically (> 99 pure by g.1.c.).The dichloro-ketone (If) (0.8 g) in 80 aqueous dioxan(10 ml) was heated with triphenylphosphine (1.2 g) on awater-bath for 48 h.After aqueous work-up g.1.c. analysisindicated that the major product was the 7-endo-mono-chloro-ketone (la), produced together with a little startingketone and some fully dechlorinated ketone.The dichloro-ketone (10 g ) was heated under reflux inbenzene (250 ml) with triethyl phosphite (25 g) for 10 daysunder nitrogen.Distillation yielded 7-chZorobicycZo3.2.0-laepta-2,6-dien-6-yZ diethyl phosphate (7), b.p. 120-125" at0.3 mmHg (15.7 g, 99) (Found: C, 47.3; H, 5.95; C1,12.8. C1,Hl6ClO4P requires C, 47.4; H, 5.75; C1, 12.75y0),M f 278 and 280 (4: l), v,, 1680 (enol ester double bond),1612 (olefin), 1295 ( P O ) , and 1040 cm-l (P-0), 6 6-01-5.61(2H, m, olefinic H), 4.46-3.90 (4H, dq, J 8.6 and 7.4 Hz re-19 B. T. Brooks and G. Wilbert, J . Amer. Chem. Soc., 1941, 68,870.(iit.,19 221-2220)992 J.C.S. Perkin Ispective'ly, CH,*O), 3.84-3-39 (2H, m, H-1 and H-5),2.91-1.92 (2H, m, allylic CH,), and 1-56-1-23 (6H, dt,J 0.7 and 7.4 Hz, CH,*CW,).(b) 7-endo- and exo-Me2hyZbicycZo3.2.0hept-2-en-6-ones(8b) and (9b). Thc 7-chloro-7-methyl ketones (lb) and(2b) in the ratio 27 : 73 (10 g) were dechlorinated by zinc inmethanolic ammonium chloride as in (a) for 3 h, to yield amixture of dechlorinated ketones (7.1 g, 91), b.p.95-100" at 15 mmHg. G.1.c. analysis indicated two com-ponents, ratio 7 : 93, the minor product being the firsteluted. The mixture showed vmX. 1778 cm-' (GO), 6 1-25(d, with some inner lines due to virtual coupling, J 7.5 Hz,exo-Me) and 0.98 (d, J 6-9 Hz, with virtual coupling, endo-Me) in the ratio 7 : 93. Samples of the dechlorinatedketones (8b) and (9b) were obtained pure by preparativeg.1.c. The endo-methyl ketone (8b) yielded a semicarbazone,m.p. 202-204" (Found: C, 60.15; H, 7.35; N, 23.55.CgH1,N,O requires C, 60.3; H, 7.3; N, 23.6). Anattempt to form the corresponding derivative for the exo-methyl epimer (9b) failed.(c) 7-endo- and 7-exo-EthyZbicycZo3.2.Ohept-2-en-6-ones(8c) and (9c).7-exo-Chloro-7-endo-ethylbicyclo3.2.0hep-tenone (2c) (172.5 mg) was dechlorinated as with zinc inmethanolic ammonia chloride for 1 h to yield 7-exo- and7-endo-ethyl ketones (Sc) and (Sc) (114 mg, 84), b.p.120" at 30 mmHg in the ratio 11 : 89 (g.1.c.) (Found: C ,79-25; HI 8.85. Calc. for C,H,,O: C, 79.4; HI 8.9y0),v- 1770 cm-I (C=O), 6 5.81 (2H, m, H-2 and H-3), 3.60(3H, m, H-1, H-5, and H-7), 2.52 (2H, m, allylic CH,), 1.48(2H, m, CH,*CH,), and 0.95 (3H, m, CH,*CH,).(d) 7-endo- and 7-exo-IsopropylbicycZo 3.2.0 hept-2-en-6-ones (8d) and (9d). 7-exo-Chloro-7-isopropylbicyclohepten-one (2d) (2 g) in the same way during 1 h gave a mixtureof 7-endo- and 7-exo-isopropyl ketones (8d) and (Qd) ; ratio88 : 12, b.p.115-120" at 3 mmHg (Found: M+, 150.104093.Calc. for C,,H,,O: M , 150-104459), vmax 1770 cm-l ( G O ) ,6 5-87 (2H, narrow m, olefinik H), 3.70 (3H, m, H-1, H-5,and H-7), 2-25 (2H, m, allylic CH,), 2-13 (lH, m, CHMe,),and 8.95 and 9.10 (6H, 2 x d, CHMe,).Hydrogenation of Dechlom'nated Adducts.-The epimeric7-alkylbicycloheptenones were hydrogenated in ethylacetate over Adams catalyst (PtO,). Uptake was 1 mol.equiv. in all cases, and g.1.c. analysis indicated no changein the ratio of epimers during the hydrogenation.(a) Epimeric 7-methylbicycZo3.2.Oheptaut-B-ones di-hydro-(8b)f and dihydro-(gb). The epimeric 7-methyl-bicycloheptenones yielded 95 of dihydro-compounds, b.p.70" at 20 mmHg; ratio 93 f 1 : 7 f 1 (g.l.c.), M+ 124(C8H,,0), v- 1777 cm-1 ( G O ) , 8 3.77-2-70 (3H, m,cyclobutane CH), 2.15-1.2 (6H, m, CH,), and 1.18 and0.97 (3H, 2 x d, ratio 8 : 92, exo- and endo-Me).(b) 7-endo- and 7-exo-EthyZbicyclo3.2.0 heptan- ones di-hydro-( sc) and dihydro-(gc).The epimeric 7-ethylbi-cycloheptenones (8c) and (Sc), ratio 89 : 11, gave 94 ofdihydro-epimers (ratio 91 : 9 by g.l.c.), b.p. 120" at 30mmHg (Found: C, 77.9; H, 10.2. Calc. for C,H,,O: C,78-2; H, 10-15y0), vmak 1770 cm-l (GO), 6 3.53 (lH, m,H-5?), 3-08 (2H, m, H-1 and H-7?), and 2.00-0-90 (ZlH, m,including distorted t at 0.94, remainder).(c) 7-endo- and 7-exo-Iso~ropyZbicycZo3.2.0heptan-6-otzesdihydro-(8d) and dihydro-(9b).The epimeric 7-endo-and 7-exo-isopropylbicycloheptenones (8d) and (9d) gave93 of dihydro-epimers (ratio 87: 13 by g.l.c.), b.p. 120"at 30 mmHg (Found: M+, 152.120424. Calc. for C,,H,,O:152.120109), v-. 1768 cm-l (GO).Equilibation Studies of 7-A lkylbicyclohefitenones and theirDihydro-derivatives. (a) 7-Methyl series. Pure 7-exo-methylbicyclohepten-6-one (9b) (40 mg) in N-sodiummethoxide in dry methanol (0.4 ml) during 1 h at 25' gavea mixture with an exo-methyl : endo-methyl ratio of 35 : 65,as judged by n.m.r. analysis. In the same way pure7-endo-methyl epimer (8b) gave an exo : endo ratio of33 : 67 in 1 h. The equilibrium mixture is regarded ascomprising 34 f 1 exo-epimer and 66 f 1 endo-epimer.An epimeric mixture of 7-exo-methylbicycloheptan-6-onedihydro-(9b) and the 7-endo-methyl isomer dihydro-(8b)(ratio 7 : 93, exo : endo) was isomerised as above: theexo : endo ratios observed by n.m.r.analysis time (min) inparentheses were: 56 : 44 (30); 68 : 42 (60); and 58 : 42(150). The final figures were taken as the equilibrium ratio.(b) Ethyl series. As n.m.r. spectra were complicated,epimensation was followed by g.1.c. analysis. The originalepimeric mixture of 7-ethylbicyclo3.2.0hept-2-en-6-ones(8c) and (9c) from dechlorination (11 exo-epimer) gavethe following exo : endo ratios time (min) in parentheses:11 : 89 (0); 34: 66 (5); 35 : 65 (10); 35 : 65 (15); 36: 64(30); and 37 : 63 (60), after treatment with N-sodiummethoxide in methanol at 20".The dihydro-epimersdihydro-(8c) and dihydro-( Sc) similarly gave the follow-ing figures: 9 : 91 (0); 69 : 31 (5); 71 : 29 (15); 73 : 27 (30);and 73 : 27 (60).In the same way rY.-sodium nieth-oxide with epimeric 7-isopropylbicyclohept-2-en-6-ones (8d)and (9d) gave the following exo : endo ratios: 12 : 87 (0);29 : 71 (5); 42 : 58 (15); 44: 56 (30); and 46 : 54 (1320).The dihydro-epimers dihydro-(Sd) and dihydro-(Sd),equilibrated in the same way, gave the following figures:13: 87 (0); 81 : 19(5); 86: 14 (30); and 86: 14 (1020).Themonochloro-ketone (la) (0.5 g) was vigorously shaken witha solution of sodium carbonate (0.1 g) in deuterium oxide(99; 2.0 ml). A parallel run with water was also carriedout. After 1 h a sample was withdrawn and the chloro-ketone was analysed by n.m.r. The integrated intensity ofthe CHCl signal at 8 5.14 had been reduced by (87 f 2)as compared with the olefinic proton signals in the deuteri-ation experiment. The n.m.r. spectrum of the chloro-ketone in water-sodium carbonate had not changedperceptibly after 1 h and i t was concluded that no epimeris-ation was occurring.After 18 h 4 of CHCl remained in the deuteriumexchange experiment. The n.m.r. spectrum of the chloro-ketone recovered from aqueous base indicated the forma-tion of ketols, confirmed by t.1.c.The chloro-ketone ( l a ) (2 g), after 72 h shaking withwater (12 m1)-sodium carbonate (0.6 g) yielded 21 ofunchanged ketone and 75 of the 7-chloro-7-( 7-chloro-6-hydroxybicyclo 3.2.01 hex-2-en-6-yl) bicyclo3.2.0 hex-2-m-6-ones (6), b.p. 125-130" at 0.15 mmHg, as a thick oil(Found: C, 58.95; H, 4.9; C1, 24.9. Calc. for C,,H,,Cl,O,:C, 59.3; H, 5.15; C1, 24.8), vmar 3590 (OH), 1785 (GO),and 1610 cm-l (olefin), 8 6.18-5.6 (4H, m, olefinic), 5.16-4.90 ( lH, m, CHCl), 4.12-3.13 (4H, m, bridgehead protons),3-05-2.11 (4H, m, methylenes), and 2.9-2.78 (lH, 2 X s,2 x OH of isomers). T.1.c. (benzene) showed two almostcoincident spots of approximately equal intensity (RF 0-35),suggesting two isomers.(c) Isopropyl series.(d) 7-endo-ChZovobicycZo 3.2.0 hept-2-en- one.3/2240 Received, 30th October, 1973
展开▼
