678 J.C.S. Perkin IReduction of 5-Hydroxymethylene-cis-caran-4-one with ComplexH yd ridesBy Francesco Bondavalli and Mario Longobardi, lstituto di Chimica Farmaceutica e Tossicologica dellrsquo;-Pietro Schenone,rsquo; Cattedra di Chimica Organica, FacoltB di Farmacia, Viale Benedetto XV-3,16132 Genova,UniversitB, GenovaItalyReduction of 5-hydroxymethylene-cis-caran-4-one (I) with lithium aluminium hydride gave 5-methylene-cis-caran-cis-and -frans-4-ols (111) and (IV) as main products, and small quantities of cis-5-methyl-cis-caran-4-one (11) and 5-hydroxymethyl-cis-car-4-ene (V), whereas reduction with sodium borohydride gave mainly5-hydroxymethylcaran-4-01 (VI) with a lesser quantity of compounds (11)-(V). The mechanisms of thesereactions are discussed.REDUCTION of enolisable 1,3-dicarbonyl derivatives ofbicyclic terpenes, such as 3-hydroxymethylenebornan-Z-one and lJ8-epoxy-3-hydroxymet hylene-p-menthan-2-one,2 by complex metal hydrides, has been investigatedas a source of allylic alcohols and carbonyl derivatives.We report here the results of this reduction applied to{ -)-5-hydroxymethylene-cis-caran-4-one (I) .3Compound (I) was reduced with an excess of lithiumaluminium hydride and with sodium borohydride; theproducts were separated by column and gas chromato-graphy into four components (II)-(V) in the formercase, and five (II)-(VI) in the latter, and were identi-fied from i.r.and n.m.r. spectral data.cis-5-Methyl-cis-caran-4-one (11) shows an i.r. spec-1.-C. Richer and R. Clarke.Tetrahedron Letters. 1964. 935.G. Minardi, F. Bondavalli; and P. Schenone, A m . Chim.(Italy), 1970, 60, 71.trum similar to that of (-)-cis-caran-4-one, and then.m.r. spectrum shows the C-5 methyl signal at 6 1.04as a doublet; this deshielding by the carbonyl groupindicates a near equatorial or cis-configuration.*The i.r. spectra of the epimeric 5-methylene-cis-caran-4-01s (111) and (IV) are similar, but the n.m.r.spectra show significant differences. The 5-methylenesignal (essentially a singlet at 6 5.10) in the case of (111)is only consistent with an axial position of the C-4hydroxy-group, which thus does not influence the methyl-ene protons. Moreover, the equatorial C-4 methynesignal (broad singlet) shows deshielding (6 3.93). Incontrast, the equatorial position of the C-4 hydroxy-8 F.Bondavalli, P. Schenone, and M. Longobardi, I2 Farnzaco,Ed. sci., 1974, 29, 48.4 W. Cocker, P. V. R. Shannon, and P. -4. Staniland, J. Chein.SOC. ( C ) , 1967, 4851976 679group in (IV) causes nonequivalence of the 5-methyleneprotons (multiplets at 6 4.98 and 5.25; c j . Z-methylene-cyclohexano15), and the C-4 methyne signal (6 3.45)shows a broad multiplet pattern, typical of an axialproton with several couplings.1.r. and n.m.r. spectra of 5-hydroxymethyl-cis-car-4-eiie (V) (see Experimental section) are in agreement withthe proposed structure. For the diol (VI) we wereunable to assign the stereochemistry at C-4 and C-5 fromspectral data.Al possible pathway for these reductions is outlined inthe Scheme.The 1,3-diol (VI) is the major product ofby acid-catalysed rearrangement of (111) and/or (IV) , aswas supposed in the case of 3-methylenebornan-2-01,scan be excluded because of the basicity of the reactionmedium.In both reductions, the ratio (ca. 3 : 1) of (111) to (IV)can be explained in terms of easier attack by the reducinganion from the less hindered side.EXPERIMENTAL1.r. spectra were measured with a Perkin-Elmer 267spectrometer, and n.m.r. spectra with a Perkin-Elmer R12(II) ( 5 )SCHEMEthe reduction with sodium borohydride. Under thereaction conditions (with methanol as solvent), proton-ation of intermediate enolate is possible and the lowbasicity of the borohydride anion will reduce the tendencyfor enolate formation.6The reduction of compound (I) with lit hiurn aluminiumhydride, as in the cases of 3-hydroxymethylenebornan-2-one 1 and 1,8-epoxy-3-hydroxymethylene-fi-rnenthan-2-0ne,~ differs from that of Z-hydroxpethylenecyclo-hexanone,' especially in the formation of an a-methylketone.The lack of an a-methyl ketone in the case of2-hydroxymet hylenecyclohexanone may be ascribed tothe occurrence of 1,2- addition only,6 whereas in the othercases 1,4-addition also takes place. This 1,4-additionmay be a result of steric hindrance at the ketone carbonylgroup, very strong in the case of the bornan-2-one (60-6574, yield of 3-methylbornan-one l) and weaker in thecase of the epoxymenthanone and in the presentinstance.The possibility that the a-methyl ketone (11) is formedV.S. Joshi, N. P. Damodaran, and S. Dev, Tetrahedron, 1968,H. 0. House, ' Modern Synthetic Reactions,' 13enjamin.24, 6817.IIcnlo Park, California, 1972, p. 82.instrument (60 MHz ; tetramethylsilane as internal stan-dard). T.1.c. was carried out on silica gel plates, with cyclo-hexane-benzene-ethyl acetate (4 : 1 : 1) as developer.G.1.c. was performed on a Fractovap GI instrument (C.Erba) (2800 x 4 mm column packed with 2.5 SE 30 onsilanized Chromosorb W; temperatures : evaporator 170 "C,column 90 "C; carrier gas helium). M.p.s were determinedwith a Mettler FPl apparatus.Reduction of B-Hyd~oxymethylene-cis-caran-4-orze.--(i)With lithium aluminium hydride. To a solution of lithiumaluminium hydride (2.28 g, 60 mmol) in anhydrous ether(100 ml), a solution of compound (I) (5.41 g, 30 mmol) inthe same solvent (25 ml) was added dropwise with stirring.After refluxing for 1 h, the mixture was cooled with iceand decomposed with water (2.5 ml), 15 sodium hydroxidesolution (2.5 ml), and finally water again ( 5 ml).Theinorganic precipitate was washed thoroughly with ether andthe combined solutions were dried (MgSO,) and evaporated.The liquid residue (6.30 g) exhibited four components ont.1.c. Repeated column chromatography on neutral alu-mina (grade I) gave the pure components as colourless oils.A. S. Dreiding and J. A. Hartman, J . Amev. Chem. SOC.,1953, 75, 939; W. J. Bailey and J. C. Goossens, ibid., 1956, '78,2804.J.-C. Richer and C.Lamarre, Canad. J. Chem., 1967, 45,1581J.C.S. Perkin IThe relative quantities of the products were ascertained bycis-5-MethyZ-cis-caran-4-one (11), eluted by light petroleum(b.p. 40-70"), had b.p. 68-70" a t 5 mmHg (Found: C,79.2; H, 10.7. CllH180 requires C, 79.5; H, 10.9);vmx. (neat) 1 710 cm-l; 6 (CCl,) 1.04 (3 H, d, J 8 Hz, 5-CH3),1.02 (3 H, s, CH,), 1.00 (3 H, d, J 6 Hz, 3-CH3), and 0.82(3 H, s, CH,). 5-MethyZene-cis-caran-cis-4-oZ (111), elutedby light petroleum-benzene (4: l), had b.p. 60-63' a t0.5 inmHg (Found: C, 79.2; H, 11.1. CllH180 requires C,79.5; H, 10.9); v,,,. (neat) 3 470, 3 065, 1 625, and 890crn-'; 6 (CCl,) 5.10 (2 H, approx. s, C=CH,) and 3.93br( 1 H, s, CH-OH) . 5-Methylem-cis-caran-trans-4-oZ (IV),eluted by benzene, had b.p.70-73" a t 0.5 mmHg (Found:C, 59.4; H, 11.2. Cl1Hl8O requires C, 79.5; H, 10.9);vmax (neat) 3 320, 3 080, 1 630, and 895 cm-l; 6 (CCl,) 5.25and 4.98 (2 H, m, C=CH,) and 3.45br (1 H, m, CHmOH).5-Hydroxy~ethyZ-cis-car-4-ene (V) , eluted by ether, had b.p.83-85' at 0.5 mmHg (Found: C, 79.6; H, 10.8. Cl1Hl8Orequires C, 79.5; H, 10.9); vmaX. (neat) 3 300, 1 660, and825 cm-l; 6 (CCI,) 5.50 (1 H, m, C=CH), 3.93 (2 H, m,g.1.c.CH,*OH), and 2.13 (1 H, m, OH; disappears with deuteriumoxide).To a solution of compound(I) (5.41 g, 30 mmol) in methanol (50 ml), a solution ofsodium borohydride (4.56 g, 0.12 mol) in N-sodium hydroxide(12 ml) was added dropwise with stirring. The mixture wasrefluxed for 2 h, the solvent removed under reduced pressure,and the residue taken up in water and extracted with ether.The combined extracts were dried (MgSO,) and the solventwas distilled off. The semi-solid residue (5.30 g) exhibitedfive components (II)-(VI) on t.1.c. Washing the residuewith light petroleum gave 5-hydroxymethylcarun-4-01 (VI) aswhite needles, m.p. 116" (hexane) (Found: C, 71.8; H, 10.9.C11H,,02 requires C, 71.7; H, 10.9); vmax (KBr) 3 420,1060, and 1020 cm-l; 6 (CDCl,) 3.88 (2 H, m, CH,*OH),3.75 (1 H, approx. s, CH*OH), and 2.80-2.20br (2 H, 111,2 OH; disappears with deuterium oxide).We thank Dr. M. Canepa for the microanalyses and Mr. A.5/1769 Received, 15th Septembsr, 19751(ii) With sodium borohydride.Panaro and Dr. S. Morass0 for 1i.m.r. and i.r. spectra
展开▼
