1974 886Reduction of Some N-Alkyl- and N-Aryl-pyrrolidin-2-ones and -piperidin-2-ones by Lithium Aluminium HydrideBy George A. Swan and John D. Wiicock, Department of Organic Chemistry, University of Newcastleupon Tyne, Newcastle upon Tyne NE1 7RUReduction of 1 -methyl-pyrrolidin-2-one (1 ) and -piperidin-2-one (9) by lithium aluminium hydride (0-25 mol.equiv.) yielded 1 -methyl-3-(1 -methylpyrrolidin-Z-yl)-A2-pyrroline (7) and 1,2,3,4-tetrahydro-1 -methyl-5-(1-methyipiperidin-2-yl) pyridine (1 4). respectively. Similar reduction of 3.3-dimethyl-1 -p-tolylpyrrolidin-2-oneafforded 3.3-dimethyl-1 -p-tolylpyrrolidin-2-ol (1 5) ; but 1 -p-tolylpyrrolidin-2-one and 3-methyl-1 -p-tolyl-pyrrolidin-2-one gave 2,3,3a,3b,4,5,6,11 b-octahydro-1 O-methyl-1 -p-tolyl-1 H-dipyrrolo 1,2-a : 3',2'-cquinoline(1 9 ; R = Me) and 2,3,3a,3b,4,5,6,11 b-octahydro-3a.4.1 O-trimethyl-1 -p-tolyl-1 H-dipyrrolol,2-a: 3',2'-clquino-line (25).respectively, each being obtained in two stereoisomeric forms. In the case of 1 -p-tolylpiperidin-2-oneit was possible to isolate 1.2.3.4-tetrahydro-1 -p-tolylpyridine (30 ; R = Me).REDUCTION of 1-methylpyrrolidin-2-one (1) by lithiumaluminium hydride (0.25 mol. equiv.) was achieved byGalinovsky et aL,I who obtained the carbinolamine (2)existing in equilibrium with the amino-aldehyde ( 5 ) ,although neither of these products could be isolated.Loss of water from (2) should result in the formation ofthe enamine (3), which presumably dimerises rapidly.However, these workers were able to show the inter-mediate formation of a monomeric material by reactionwith acetonedicarboxylic acid, yielding hygrine andcuscohygrine.Leonard and Cook oxidised l-methyl-pyrrolidine with rnercury(I1) acetate in acetic acid, andI I I $1Q C h O IMe.b.+again failed to isolate the resulting l-methyl-A2-pyrroline(3). They obtained a dimer, which they formulated asthe pyrrolidinylpyrroline (7), and a trimer (6). Theyalso showed that in solution the dimer existed in equili-brium with the monomer, so that it underwent slowreaction with ethyl acetoacetate to yield hygrine.F. Galinovsky, A. Wagner, and R. Weiser, Sitzungsber.N. J. Leonard and A. G. Cook, J . Amer. Chem. SOC., 1959,81,E. Leete, J . Amer. Chem. SOC., 1967, 89, 7081.M.L. Rueppel and H. Rapoport, J . Amer. Chem. Soc., 1971,H. C. Brown and A. Tsukamoto, J . Amer. Chem. SOC., 1964,Akad. Wiss. Wien, 1951, Abt. 2B, 160, 551.5627.93, 7021.86, 1089.Leete,3 on the other hand, claimed that reduction ofl-methylpyrrolidin-2-one, followed by treatment of theproduct with acid, yielded the iminium salt (4), althoughhis evidence that this was monomeric was not strong.Rueppel and Rapoport isolated a correspondingiminium salt by acidification of the reduction product of1 ,3-dimethylpyrrolidin-2-one.We obtained (6) and (7) by reduction of l-methyl-pyrrolidin-2-one with either lithium aluminium hydrideor sodium bis-(2-methoxyethoxy)aluminium hydride ;and confirmed the identity of the i.r., n.m.r., and massspectra of the dimeric products with those of thatobtained by oxidation of l-methylpyrrolidine withmercury(I1) acetate.However, the reduction morereadily yielded a pure product than did the oxidationmethod. Oxidation of (7) with alkaline ferricyanideafforded 1-methyl-3-( l-methylpyrrolidin-2-y1)pyrrole (8).Likewise, by reduction of l-methylpiperidin-2-one (9)with lithium aluminium hydride (at room temperatureor at -70"), or with lithium diethoxyaluminium hydrideor triethoxyaluminium h ~ d r i d e , ~ we obtained 1,2,3,4-tetrahydro-l-methyl-5-( 1-methylpiperidin-2-y1)pyridine(la), which had previously been obtained by reductionof the same lactam with sodium and and alsoby oxidation of l-methylpiperidine with mercury(@acetate.8 The crude product of reduction of l-methyl-piperidin-2-one with lithium aluminium h ydride, whentreated with hydrochloric acid, afforded a syrup, then.m.r.spectra of which showed a broad singlet at T 1.12,characteristic of an iminium ion (=N=C-H); cf. refs. 9and 10. When this material was treated with benzoylchloride in the presence of sodium hydroxide, it yieldeda viscous syrup, which was evidently a benzoylateddimer, probably (12).We have earlier described the reduction of strychnineto 10,l l-didehydrostrychnidine ; l1 we have now in-vestigated the reduction of some l-aryl-pyrrolidin-2-onesand -piperidin-2-ones.R. Luke: and J. KovA?, Coll. Czech. Chem. Comm., 1954, 19,1215.C. Schopf and H. L. de Waal, Chem. Bey., 1966, 89, 909. * N. J. Leonard and F.P. Hauck, jun., J . Amer. Chem. SOC.,A. F. McDonagh and H. E. Smith, J . Org. Chem., 1968,33,8.10 H. Volz and H.-H. Kiltz, Annalen, 1971, 752, 86.l1 G. A. Swan and J. D. Wilcock, J.C.S. Perkin I, 1972, 1068.+1957, 79, 5279886 J.C.S. Perkin I(e.g. at 200"), although even at 120" the 392 peak wasstill small compared with the 205. The n.m.r. spectrumof the carbinolamine in carbon tetrachloride showed adoublet at 7 5-37 (J 6 Hz), attributed to the 2-proton ofthe pyrrolidine ring, coupled with the hydroxy-proton ;exchange with deuterium oxide appeared to convert thisinto a singlet, although it was not possible to be certain,because of the proximity of the HOD peak. Singletswere present at z 7.74 (ArMe) and at 8.9 and 9.11 (CMe,).Treatment of the above, crude material (mixture ofcarbinolamine and ether) with hydrochloric acid yieldeda salt, the n.m.r.spectrum of which suggested it to behomogeneous (1 7).Reduction of 1-fi-tolylpyrrolidin-one (18 ; R = Me)yielded a mixture of two isomeric bases, C,,H,N,, whichwere separated by chromatography on silica. Thefaster-running isomer (A) had m.p. 144-145" and theslower-running (B) m.p. 115-116". In most respectsthese two bases were alike. However, the U.V. spectrumof A showed maxima at 258 and 318 nm, unchanged(except for lowering in intensity) on acidification;whereas that of B showed maxima at 253 and 300,shifted to 262 and 335 nm on acidification. Initially itseemed possible that these isomers might have structuresThe 1-arylpyrrolidin-ones were conveniently ob-tained by condensation of appropriate primary aromaticamines with y-butyrolactones. 3-Methyl-1-fi-tolylpyr-rolidin-2-one was also obtained from l-fi-tolylpyrrolidin-2-one by treatment with diethyl carbonate in the KN/MeMeMe(131 ( 1 4 )presence of sodium hydride, followed by methylation,hydrolysis, and decarboxylation.We encountered difficulty in using Krimm's l2 methodfor the preparation of I-arylpiperidin-2-ones, as in ourhands the reaction of cyclopentanone with aniline re-sulted predominantly in a 2 : 1, rather than a 1 : 1 con-densation product.Therefore 1-aryl-2-pyridones, pre-pared essentially by Tschitschibabin and Jeletzky'smethod,13 were hydrogenated catalytically to the corre-sponding 1 -arylpiperidin-ones.Condensation of fi-toluidine with 8-valerolactone also afforded l-fi-tolyl-piperidin-2-one.Reduction of 3,3-dimethyl-l-~-tolylpyrrolidin-2-onewith lithium aluminium hydride afforded a crystallineproduct which appeared to consist mainly of thecarbinolamine (15), containing a little of the correspond-ing ether (16). Attempts to separate these by chromato-graphy on alumina or silica were only partly successful,interconversion evidently being easy. However, sublim-ation in vacuum yielded the ether, in the mass spectrumof which (at 120") the molecular ion peak (m/e 392) wascomparable in height to the m/e 205 peak. The n.m.r.spectrum of this showed singlets at z 5.05 (N-CH-0),7.78 (ArMe), and 8-78 and 9-12 (CMe,), but suggestedthat the compound was not pure, singlets (each re-presenting only a fraction of one proton) being presentat T 4.9 and 9.2.It was therefore assumed that thebulk of this material consisted of the meso-ether, contain-ing a small amount of the ()-ether.No mass spectrum of the carbinolamine was obtainedwhich failed to give a peak at m/e 392 due to the presenceof some of the ether (16), although in the best spectrathis peak was very small compared with the molecularion peak of the carbinolamine (i.e. m/e 205). Thebest spectra were obtained with rapid volatilisationla H. Krimm, Chem. Bey., 1958, 91, 1057.l3 A. E. Tschitschibabin and N. P. Jeletzky, Ber., 1924, 57,1168.Me Me(16)R(19)RI20 )nI21 1 ( 2 2 ) (23 1 ( 2 4 )(19; R = Me) and (20; R = Me), respectively. How-ever, the latter was eliminated for the following reasons:(a) absence of an absorption maximum at 1650 cm-l inthe i.r.spectrum of either A or B such an absorption isshown by (7); (b) failure to convert B into A by treat-ment with acid or by sublimation in vacuum; (c) th1974 887longest wavelength absorption band in the U.V. spectrumof (17) being at 282 nm; and (d) the great similaritybetween the mass spectra of A and B.Bases A and B are therefore regarded as being stereo-isomeric forms of the dipyrrolol,2-a:3’,2’-cquinoline(19 ; R = Me). Peaks at m/e 185 and 184 in the massspectra of both isomers were attributed to the ions (21;R = Me) and (22; R = Me), respectively.Howeverthe 185 peak is of much lower intensity than the 184peak in the case of B, whereas in A these peaks are ofcomparable intensity.The n.m.r. spectra of A and B showed doublets atT 5.07 and 5-66, respectively, representing the l l b -proton coupled to the 3a-proton (J 7 and 8 Hz, re-spectively). The spectra of the two isomers obtainedby reduction of 1-p-tolylpyrrolidin-2-one with lithiumaluminium deuteride lacked these doublets, and weresimpler in the T 6.25-7.1 region (loss of signal due to3b-proton).A phenhomazine structure (23) is also theoreticallypossible for either or both isomers; but is ruled out byreason of lack of symmetry: (a) A and B each show twoseparate methyl peaks in their n.m.r. spectra; and(b) the doublets at T 5.07 and 5.66 each represent onlyone proton.Reduction of 1 -phenylpyrrolidin-2-one likewise yieldedtwo isomers, formulated as (19; R = H), which corre-sponded in all the foregoing respects to compounds Aand B.The m.p. and U.V. spectrum of the faster-running isomer corresponded to those given by Wittigand Sommer14 for a compound which they obtainedsimilarly, but which they incorrectly formulated as1-phenyl-A2-pyrroline. However, we confirmed thatthe compound described by these authors as I-phenyl-A3-pyrroline was correctly formulated, although itsn.m.r. spectrum is deceptively simple (cf. ref. 15). Wefailed to isomerise this to 1-phenyl-A2-pyrroline byheating with Raney nickel.Very strong peaks at m/e 198 and 184 in the massspectra of both isomers of (19; R = Me), and (19;R = H), respectively are attributed to the ions (24;R = Me) and (24; R = H), respectively.This formu-lation received support from the appearance in thespectrum of (19; R = H) of a metastable transition:184’- 169+ +- 15 (Le. loss of Me). The metastabletransitions 290+ _t 171+ + 119 and 171+ _t 170+ +H also occur. The peaks at 171 and 170 represent(21; R = H) and (22; R = H), respectively.Reduction of 3-methyl-l-~-tolylpyrrolidin-2-one like-wise afforded two isomers, formulated as the dipyrrolo-1,2-a:3’,2’-cquinoline (25), the mass spectra of whichshowed intense peaks for fragment ions at m/e 213 and212, corresponding to (26) and (27), respectively.One possible explanation for the shift to longer wave-length in the U.V.spectra of the slower-running isomers(19; R = Me or H) and (25) on acidification might bethat C- rather than N-protonation occurs. The n.m.r.spectra of the pairs of isomers (19; R = Me and H) weretherefore compared by measurements (a) on the base indeuteriochloroform; (b) as (a) with the addition ofmethanesulphonic acid (1 equiv.); (c) as (a) with theaddition of methanesulphonic acid (2-3 equiv.); (d) inCF3*C02H ; and (e) in CF,*CO,D. Acidification causeda downfield shift of most peaks, but a peak appeared inthe T 4 region. This was a somewhat broad singlet inthe case oi (19; R = Me),fine structure) in the case ofMeMe( 2 5 )4 R( 2 9 )but a doublet(19; R = H);Me ge /Me(27)I 31)equally with CF,*C02H or CF3*C0,D.(with furtherand appearedr - l8 RThis signal wastherefore attributed to the aromatic 11-proton, which inthe salts must presumably be shielded by the arylresidue attached to N-1.Moreover, in the case of bothisomers, one of the methyl peaks moved upfield (to7 ca. 8.1) on the addition of 1 equiv. of acid, againpresumably owing to the same shielding effect. Thespectra of pairs of isomers under comparable conditionsof acidity were very similar; no evidence for C-proton-ation was deduced from these spectra.There was no evidence of rapid irreversible changeoccurring with dilute acid at room temperature, as thesame isomers (A and B) could be obtained either by anisolation procedure avoiding contact with acid, or byone in which the bases were taken into acid, and thenliberated by basification.However, long boiling of Bwith 2~-hydrochloric acid resulted in extensive decom-position, with the formation of a product which was notobtained pure, but which had the molecular formulaC2,H2,N,. Its spectra were consistent with the structure(28), which could have been formed through protonationat C-lla resulting in ring-opening to give (20; R = Me),followed by disproportionation.Four configurations (A-D) for (19) are possible(represented schematically in the Figure), of which Aand B appear to be consistent with the informationl5 L. F. Johnson, A. V. Robertson, W. J. R. Simpson, and B.G. Wittig and H. Sommer, Annalen, 1955, 594, 1.Witkop, Austral.J. Chem., 1966, 19, 116J.C.S. Perkin Iknown about the faster- and slower-running isomers,respectively. Of these configurations, Dreiding models(allowing reasonably full conjugation of each nitrogenatom with its appropriate aromatic ring) suggest that Ais unstrained, and B only slightly strained; C is alsoA B C D3 b - H . .3a-H . .116-H o oFIGUREsomewhat strained, and D is more strained than B.The torsion angle between the 3a- and llb-protons forisomers A, B, C, and D is seen to be ca. 30, 3540, 180,and NO", respectively. In A H-llb can lie approxi-mately in the plane of the aromatic ring attached toN-1; whereas in B this proton will be out of the plane.This could account for the lower field signal for H-llbin the n.m.r.spectrum of A than that of B. The shield-ing effects on H-11 and on the methyl group at position10 are explicable on the assumption that protonationoccurs first at N-1, the configuration of the latternitrogen atom thus changing from planar to tetrahedral.Under the acidic conditions used for the measurementof the n.m.r. spectra there seems little doubt thatN-protonation occurs in both isomers. It is, however,possible that under the acidic conditions used for themeasurement of U.V. spectra isomer B undergoes proton-ation at.C-lla to a sufficient extent to show weak longwavelength absorption.The formation of (19) could involve two steps, withthe intermediary formation of (20); on the other hand acycloaddition reaction might also be possible.Roy and Swan l6 showed that the cation (PhNMe:CH,)+condensed with ethyl vinyl ether to afford a quinolinederivative.We therefore carried out the reduction ofl-~-tolylpyrrolidin-2-one in the presence of ethyl vinylether, but failed to detect any product which might havebeen formed by the reaction of the latter with animinium ion corresponding to (4).From the products of reduction of l-$-tolylpiperid-2-one (29; R = Me), an oil was isolated by chromato-graphy, which although not pure, evidently consistedmainly of 1,2,3,4-tetrahydro-l-$-tolylpyridine (30; R =Me). The mass spectrum of this showed a molecularion peak at mle 173 (C,,H,,N); and its n.m.r. spectrumshowed a multiplet at T 5*05-5*4, attributed to the3-proton of the pyridine ring, and only one AxMe singletat T 7.75.This was unstable, and its isolation andpurification were carried out so far as possible in thedark, as otherwise a red product was formed; a similarbehaviour has been recorded for 1,4-dihydro-l-methyl-quinoline.17Attempts to distil the aryltetrahydropyridine (30;R = Me) yielded a product, the mass spectrum of whichshowed a molecular ion peak at m/e 346 (C,H,,N,),l6 R. B. Roy and G. A. Swan, J. Chem. SOC. (C), 1969, 1886.l7 J. W. Bunting and W. G. Meathrel, Tetrahedron Letters, 1971,133.together with an intense fragment ion peak at m/e 198probably (32; R = Me), suggesting that dimerisationhad occurred, giving the benzohpyrido2,1-fll,6naph-thyridine (31; R = Me); the n.m.r. spectrum of thecrude material was consistent with this.Similarreduction of 1-phenylpiperidin-one yielded 1,2,3,4-t e t rahydro- 1 -phenylp yridine.EXPERIMENTALN.m.r. spectra were measured in solutions in deuterio-chloroform (except where otherwise stated), using tetra-methylsilane as internal standard, with a Perkin-ElmerR10 spectrometer a t 60 MHz. U.V. spectra were measuredfor solutions in ethanol. Mass spectra were obtained byuse of an A.E.I. MS9 instrument, with direct insertion.1.r. spectra were measured for potassium bromide discs orliquid films. For column chromatography Hopkin andWilliams M.F.C. silica gel and aluminium oxide were used.Merck Kieselgel G and aluminium oxide G (type E) wereused for t.l.c., the spots being detected by exposure toiodine vapour.Light petroleum refers to the fraction ofb.p. 40-60°, except where otherwise stated. Lithiumaluminium hydride solution was prepared as describedearlier.llReduction of l-MethyZfirrolidin-2-one (1) .-To a stirredsolution of l-methylpyrrolidin-2-one (9 ml) in benzene(50 ml) cooled in ice, one of sodium bis-(2-methoxyethoxy)-aluminium hydride (70 in benzene; 24 ml) diluted withbenzene (30 ml) was added during 1 h. The mixture wasallowed to come to room temperature, boiled under refluxfor 1 h, then cooled in ice, decomposed with saturatedaqueous potassium sodium tartrate, and extracted withbenzene. The extract was then extracted with 2~-hydro-chloric acid, and the latter extract was basified (40sodium hydroxide), and extracted with ether.Distillationof the dried (K,CO,) ethereal extract afforded l-methyl-3-( l-methylpyrrolidin-2-y1)-A2-pyrroline (7), b.p. ca. 100" a t16 mmHg (1.64 g), and l-methyl-3-l-methyl-3-( l-methyl-pyrrolidin-2-yl) pyrrolidin-2-yl-A2-pyrroIine (6), b.p. ca.170" a t 16 mmHg (2 g). The former fraction when re-distilled had b.p. 100" a t 16 mmHg (Found: M+, 166.Calc. for C,,Hl,N2: M , 166), vm.. 1650 cm-1 (C=C str.);T (CCI,) 4-3br (s, H-2), 7-52 (s, pyrroline NMe), and 7-9 (s,pyrrolidine NMe). This redistilled material (0.97 g) wasadded to a solution of potassium ferricyanide (4.6 g) andsodium hydroxide (1.1 g ) in water (20 ml); the mixture waskept for 6.5 h in a refrigerator, and was then extracted withether. Distillation of the dried (K,CO,) extract gavel-unethyZ-3-( 1-methyZ~~rroZidin-2-yl)pyrroZe ( 8 ) , b.p.125' at16 mmHg (0.46 g) (Found: M', 164. C,,H,,N, requiresM , 164) ; 'I: (CC1,) 3-6 (2H, resembling d, with fine splitting,pyrrole H-2 and -5), 4-03 (lH, resembling t, with finesplitting, pyrrole H-4, J4,6 2 Hz), 6.4 (3H, s, pyrrole NMe),and 7-9 (3H, s, pyrrolidine NMe).l-M'ethylpiperidin-2-one (9) .-l-Methyl-2-pyridone l8 washydrogenated l9 to l-methylpiperidin-2-one, T 6.5-6.8 (2H,m, NCH,), 7.03 (3H, s, NMe), 7.4-7-8 (2H, m, CO-CH,),and 8.0-8.3 (4H, m, CO*CH2*CH2*CH2).Reduction of l-Methylpiperidin-2-one (9) .-An etherealsolution of Iithium aluminium hydride (0.25 mol. equiv.)l* P. Beak, J . Bonham, and J . T. Lee, jun., J. Amer. Chenz. Soc.,1968, 90, 1669.19 N.J . Leonard and E. Barthel, jun., J. Amer. Cham. SOC.,1949, 71, 30981974was added with stirring during 1 h to one of l-methyl-piperidin-2-one (2-15 g); the mixture was boiled underreflux for 1 h, then cooled, and decomposed with saturatedaqueous potassium sodium tartrate. The ether wasremoved from the dried (Na,SO,) organic layer; and theresidue was chromatographed on silica. Elution withchloroform afforded l-methylpiperidin-2-one (1.2 1 g), andelution with chloroform-ethanol (7 : 1) gave 1,2,3,4-tetra-hydro-l-methyl-( l-methylpiperidin-2-y1)pyridine (14)as an oil (0.47 g) (Found: M f , 194.1769. Calc. forC,,H,,N,: M , 194-1783), vmx. 1655 cm-l (C=C str.); 7 4.12(lH, s, H-3), 6.8-7.3 (5H, m, 2 x NCH, and NCH), 7.40(3H, s, pyridine NMe), 7.75 (3H, s, piperidine NMe), and7-8-8.6 (lOH, m, 5 x CH,); hmx, (neutral) 230 nm, (acidic) 262 nm.The dipicrolonate (from ethanol) hadm.p. 149-150" (lit.,l 155-156") (Found: C, 52.9; H, 5.0;N, 19.4. Calc. for Cl2H,,N2,2C1,H,N,O5: C, 53-1; H, 5.25;N, 19.4).A similar reduction was carried out on l-methylpiperidin-2-one (4.9 g ) , but the reaction mixture was decomposedwith sodium hydroxide solution (25; 20 ml), and theethereal layer was extracted with 2~-hydrochloric acid.Evaporation of the aqueous layer under reduced pressureafforded the salt (13), as a syrup, which was dried (P,05)in a vacuum. A portion of the residue was stirred for 3 hwith a mixture of benzoyl chloride (4 ml) and 2~-sodiumhydroxide (15 ml).Saturated aqueous sodium iodide wasthen added, and the resulting solid was collected, treatedwith 2~-hydrochloric acid, and then basified (NH,OH).The mixture was extracted with chloroform, and thesolvent was removed from the dried (N2t2S04) extract,yielding a syrup, m/e 298, 163, 136, 110, 105, and 77;T 24-24 (6H, m, Ar and olefinic H), 7.48 and 7.86 (each3H, s, 2 x NMe), and 6-3-9.0 (14H, m, remaining protons).3,3-Dinaethyl- l-p-tolylpywolidin- 2-one .-A mixture of 2,2-diniethyl-y-butyrolactone 2a (5.2 g) and P-toluidine (5.5 g)was heated in a sealed tube for 45 h at 290". A solution ofthe residue in a little ethanol was diluted with lightpetroleum, and the resulting solid was recrystallised fromlight petroleum (b.p. 60-80') (charcoal), yielding theproduct (7-35 g ) , m.p.78' (Found: C, 77.4; H, 8.7; N,6-85. C,,H,,NO requires C, 76.8; H, 8.4; N, 6.9);vmx. 1685 cm-l ( G O str.); T (CCl,) 242 and 2-8 (each 2H,J 8 Hz, Ar ABq), 6.3 and 8.08 (each 2H, t, J 8 Hz;K*CH2-CH,), 7.68 (3H, s, ArMe), 8-81 (6H, s, CMe,).Ethyl 2-0x0- l-p-tolyZpyrrolidine-3-carboxylate.-A soh tionof l-p-tolylpyrrolidin-one 21*2a (20 g) in dry benzene(80 ml) was added during 2 h to a mixture of sodiumhydride (10 g), diethyl carbonatz (46 g), and dry benzene(100 ml), boiling under reflux. The mixture was boiledfor a further 24 h and was then cooled, treated with water(60 ml), and acidified with acetic acid. The solvent wasremoved from the dried (Na,SO,) benzene layer, and theresidue was recrystallised twice from benzene-lightpetroleum, yielding the ester (16.3 g, 58), m.p.80-81"(Found: C, 67-9; H, 6.85; N, 5.7. Cl,H1,N03 requiresC, 68.0; €3, 6.9; N, 6.7); vmx. 1730 (ester C=O str.) and1685 cm-l (y-lactam C=O str.); T 2.49 and 2-79 (each 2H,J 9 Hz, Ar ABq), 5.72 (2H, q, J 8 Hz,CH,-CH,*O), 5.9-6.5(3H, m, NCH, and EtOCO-CHCO), 7.3-7.8 (2H, m,4-H,), 7-67 (3H, s, ArMe), and 8-7 (3H, t, J 8H2, CH3CH,-O).Ethyl 3-Methyl-2-oxo- 1 -p-toIylpyrro Iidine- 3-carboxylate.-A2o R. Luke5 and V. DGdek, Cull. CmA. Chem. Comm.. 1968,S.21 J. T. Braunholtz and F. G. Mann. J . Ckem. Soc., 1967, 4174.1981.solution of sodium (2 g) in ethanol (20 ml) was added to asolution of the above ester (16 g) in ethanol (200 ml),followed by methyl iodide (40 ml).The mixture wasboiled under reflux for 24 h, and then cooled. The solventwas removed under reduced pressure, the residue wastreated with water (30 ml), and the mixture was extractedwith chloroform. The chloroform was removed from thedried (Na,SO,) extract, yielding a red oil, which crystallisedwhen treated with hexane. Recrystallisation from hexaneafforded the product (14.2 g), m.p. 41-42' (Found: C, 68.9;H, 7.2; N, 5-4. Cl,H,,N03 requires C, 68.9; H, 7.25; N,5.4); vmx, 1742 (ester C=O str.) and 1690 cm-l (y-lactamC=O str.) ; T 2.52 and 2.87 (each 2H, J 9 Hz, Ar ABq),8.85 (2H, q, J 8 Hz, CH,*CH,*O), 6-05-6.5 (2H, m, NCH,),7.2-8.3 (2H, m, 4-H,), 7.70 (3H, s, ArMe), 8-60 (3H, s,3-Me), and 8.78 (3H, t, J 8 Hz, CH,*CH,*O).3-Methyl-2-oxo- l-p-tolylpyrrolidine-3-carboxylic A cid .-The above ester (14 g) was boiled under reflux for 20 h with2~-hydrochloric acid (100 ml) : the cooled mixture was thenbasified (NaHCO,), and extracted with chloroform.Theaqueous layer was acidified (HC1) and extracted withchloroform, affording the acid (12.1 g), m.p. 101-102"(Found: C, 67.0; H, 6.55; N, 6.15. Cl,Hl,N03 requiresC, 67.0; H, 6.45; N, 6.0); vmX. 1716 (acid G O str.) and1684 cm-l (y-lactam C=O str.) ; 7 - 1.0 (lH, s, CO,H), 2.54and 2.86 (each 2H, J 9 Hz, Ar ABq), 6-0--6.45 (2H, m,NCH,), 7.2-8.2 (2H, m, 4-H,), 7.71 (3H, s, ArMe), and8.52 (3H, s, 3-Me).3-Methyl-l-p-tolylpyrrolidin-2-one.-(a) The above acid(12 g) was heated for 5 h at 160". The residue, recrystal-lised from light petroleum, afforded the Product (10.1 g),m.p. 60-60.5" (Found: C, 76.15; H, 7.95; N, 7.55.C,,H,,NO requires C, 76.2; H, 7-95; N, 7.4); vmaL1680 cm-l ( G O str.); T 2.53 and 2.97 (each 2H, J 9 Hz,Ar ABq), 6.3-6-6 (2H, m, N*CH,), 74-8.6 (3H, m, 3-Hand 4-H,), 7-75 (3H, s, ArMe), and 8-85 (3H, d, J 7 Hz3-Me).(b) The same product (6 g) was obtained by heating2-methyl-y-butyrolactone (4 g) with p-toluidine (4-85 g)for 90 h at 270".Reaction between Aniline and Cyclopenianone.-A mixtureof aniline (31.5 g, 0.33 mol), cyclopentanone (25 g, 0.3 mol),and toluene (150 ml) was boiled for 15 h in an apparatuswith a water-separator.Distillation of the remainingsolution afforded an oil, b.p. 250" a t 20 mmHg, whichcrystallised; the product had m.p.53-54" (from ethanol)(Found: M+, 225.l-Phenylpiperidin-2-one (29; R = H) .-l-Phenyl-2-pyridone was prepared by Tschitschibabin and Jeletzky'smethod,13 except that the reaction mixture was heated for12 insteaiof 6 h. The product isolated had m.p. 123-125"(yield 15.8) ; further extraction of the reaction mixturewith chloroform, followed by chromatography on aluminaof the residue obtained by evaporation of the extract(elution with benzene), gave a further quantity (8.3).l-Phenyl-2-pyridone (17 g) was hydrogenated at 1 atmin acetic acid (60 ml) over Adams cataIyst (0.1 g), yieldingl-phenylpiperidin-2-one (16.9 g ) , m.p. 101-103" (frombenzene) (lit.,1z 99-100"); v- 1667 cm-1 (blactam C=Ol-p-Tolylpiperidin-2-one (29; R = Me) .-(a) This awl-piperidone, m.p.86-87" (lit.,23 87-88") (from hexane-22 E. Spath and J. Lintner, Bey., 1936, 69, 2727.23 0. V. Schickh, B.P. 919,404/1963 (Chem. Abs., 1963, 59,Calc. for C,,H,,N: M , 225).S t r . ) .1600) J.C.S. Perkin Ibenzene), was similarly obtained via 1-p-tolyl-2-pyridone ;m.p. 125-128', vmx 1670 cm-l ( G O str.), prepared asabove.(b) The same compound was obtained by heating to-gether equimolecular amounts of 8-valerolactone and p-toluidine for 30 h at 220".Reduction of 3,3-Dimethyl- 1-p-tolylpyrroZidin-2-one .-Ethereal lithium aluminium hydride solution (0.37 mol.equiv.) was added dropwise t o the lactam (2 g) in ether(25 ml) with stirring and cooling in ice; the mixture wasthen stirred for 2.5 h a t room temperature, cooled inice, and decomposed by dropwise addition of an excess of2~-hydrochloric acid.The aqueous layer was separated,extracted with ether, then basified (NaOH), and againextracted with ether. The latter extract was dried( K2C03) and evaporated. The residue, recrystallised fromlight petroleum, afforded a product consisting mainly of3,3-dimethyl-l-p-tolylpyrrolidin-2-ol (15) (0-69 g). A solu-tion of this in dilute hydrochloric acid was evaporated todryness; the residue consisted of (17) as the chloride,'t 1.95 and 2.65 (each 2H, J 8 Hz, Ar ABq), 2-58 (lH, s,==CH-), 5.05 (2H, t , J 7 Hz, =G-CH,), 7.6 (2H, t, J 7 Hz,4-H,), 7.6 (3H, s, ArMe), and 8.4 (6H, s, CMe,); A,,(neutral) 210, 248, and 303 nm; A,, (acidic) 210, 227sh,and 282 nm.Redwtion of 1-p-Tolylpyrrolidin-one (18; R = Me) .-Ethereal lithium aluminium hydride solution (0.25 mol.equiv.) was added dropwise t o a stirred solution of thelactam (4 g) in ether (50 ml) at room temperature.After3.5 h the mixture was decomposed by potassium sodiumtartrate solution: and the ether was removed from thedried (Na,SO,) organic layer. The residue (3-8 g) showed3 spots on t.1.c. (silica; benzene), Rp 0.95, 0.85, and 0.15;and was chromatographed on silica. Elution with benzene-light petroleum (1 : 1) yielded base A, 2,3,3a,3b,4,5,6,1 lb-octahydro-10-methyl-1-p-tolyl- lH-dipyrroZol , 2-a:3',2'-c-quinoline (19; R = Me), RF 0.95, m.p. 144-145' (fromethanol or benzene-light petroleum) (0.39 g) (Found: C,82.85; H, 8.1.C,,H,,N, requires C, 83.0; H, 8.2);T 2-6-34 (7H, m, Ar), 5-07 (lH, d, J 7 Hz, H-llb), 6.25-7.1 (5H, m, 2 x NCH, and NCH): 7.50-8.5 (7H, m,other CH, and CH groups), 7.76 and 7.93 (each 3H, s,2 x ArMe); m/e 318 (C22H26N2), 211, 210, 198, 185, 184,170, 159, 157, 120, 93, 91, and 57; lmx. 258 and 318 nm(E 24,800 and 6050). Elution with benzene gave thestereoisomeric base B, RF 0-85, m.p. 115-116' (from ethanolor benzene-light petroleum) (0.38 g) (Found: C, 82.75;H, 8.0); z 2.7-3.7 (7H, m, Ar), 5.66 (lH, d, J 8 Hz,H-llb), 6.25-7-6 (5H, m, 2 x NCH, and NCH), 7.6-8.5 (7H, m, other CH, and CH groups), 7.76 and 7-84 (each3H, s, 2 x ArMe); rn/e 318, 211, 210, 198, 185, 184, 170,159, 156, 120, 93,91, and 57; A,, (neutral) 253 and 300 nm(E 23,500 and 4140) ; Elutionwith chloroform yielded l-~-tolylpyrrolidin-2-one, Rp 0.15(2.85 g).Bases A and B were also isolated from a similar experi-ment in which the reaction products were taken into2~-hydrochloric acid, and liberated by basification (NaOH) .Reduction of 1-Phenylpyrrolidin-2-one (18 ; R = H) .-The lactam (3 g) was reduced as above.Elution withbenzene-light petroleum (2 : 1) afforded 2,3,3a,3b,4,5,6,1 lb-octahydro- 1-phenyl-lH-dipyrroZo 1,2-a:3', 2'-cquinoZine (1 9;R = H), RF 0.74, m.p. 165-166' (from methanol) (0.38 g)(Found: C, 82-6; H, 7-45; N, 9.75. C,,H,,N, requires(acidic) 262 and 335 nm.C, 8243; H, 7.6; N, 9.6); z 2-5-3.6 (9H, m, Ar), 4.85(lH, d, J 7 Hz, H-llb), 6.15-7.0 (5H, m, 2 x NCH, andN-CH), and 7-35-8.5 (7H, m, other CH, and CH) ; m/e 290(C20H2zN2), 196, 184, 171, 170, 156, 130, 115, 106, 91, and77; Lx 253 and 311 nm (E 22,200 and 5800).Elution withbenzene gave a stereoisomer of the preceding base, RF 0.52,m.p. 153-154' (from methanol) (0-36 g) (Found: C, 82.55;H, 7.55; N, 9.8); z 2-6-3.5 (9H, m, Ar), 5.55 (lH, d,J 8 Hz, H-llb), 6.2-7-2 (5H, m, 2 x NCH, and NCH),and 7-2-8.5 (7H, m, other CH, and CH); m/e 290, 196,184, 170, 156, 145, 130, 128, 115, 106, 91, and 77; A,,,.(neutral) 251 and 294 nm (E 19,700 and 3830); Am= (acidic)260 and 327 nm (c 13,000 and 2900). Elution with chloro-form yielded l-phenylpyrrolidin-2-one, RF 0.10 (2.1 g).l-Phenyl-A3-pyrroline.-This was prepared by Wittigand Sommer's method,14 and had m.p. 91-92' (lit.,1499-100'), v,, 1630 cm-l (C=C str.); 't 2.6-3.6 (5H, m,Ar), 4-12 (2H, s, 2 x olefinic H), and 5.95 (4H, s, 2 xNCH,).It was recovered unchanged after being boiledin xylene with Raney nickel for 15 h.Reduction of 3-Methyl- 1 -p-toZylpyrroZidin-2-one. -Thelactam (3 g) was reduced as above, and the crude productwas chromatographed on silica. Elution with benzene-light petroleum (1 : 8) yielded 3-methyl-1-p-tolylpyrrolidineas an oil (0.11 g), Rp 0.96; T 2.91 and 3-49 (each 2H, J 9 Hz,Ar ABq), 6-5-7-2 (4H, m, 2 x NCH,), 7.75 (3H, s, ArMe),8.88 (3H, d, J 8 Hz, 3-Me), and 7-8-8-6 (3H, m, CH,CH) ;(Found: M+, 175. Cl,H1,N requires M , 175). Elutionwith benzene-light petroleum (1 : 6) afforded 2,3,3a,3b,-4,5,6,1 lb-octahydro-3a,4,lO-trirnethyZ- 1-p-toZyl-lH-difiyrroZo-1,2-a:3',2'-cquinoZine (25) , Izp 0.88, m.p.180-181' (frombenzene) (0.26 g) (Found: C, 82.85; H, 8-45; N, 8.0;Mf, 346.2418. C,,H,,N, requires C, 83.3; H, 8.65; N,8.1; M , 346.2409); 7 2-7-3.7 (7H, m, Ar), 5.41 (lH, s,H-llb), 6.45-7.2 (5H, m, 2 x NCH, and NCH), 7.3-8.7(5H, m, other CH, and CH), 7.70 and 7.88 (each 3H, s,2 x ArMe), 8-82 (3H, d, J 6.5 Hz, 4-Me), and 9.02 (3H, s,3a-Me); m/e 346, 342, 332, 331, 328, 226, 222, 213, 212,198, 195, 184, 171, 144, 133, 120, 105, and 91; Lx. 258 and316 nm (c 26,400 and 5440). Elution with benzene-lightpetroleum (1 : 3) gave a stereoisomer of the preceding base,Rp 0.60, m.p. 118-120' (from benzene) (0-14 g) (Found:Mf, 346.2389); T 2.75-3-45 (7H, m, Ar), 6.10 (lH, s,H-llb), 6.2-7-5 (5H, m, 2 x NCH, and NCH), 7.6-8-6(5H, m, other CH, and CH), 7-70 and 7-80 (each 3H, s,2 X ArMe), 8.90 (6H, 1s and Id, 3a- and 4-Me); m/e 346,331, 267, 222, 213, 212, 196, 173, 120, 111, and 91; I,,(neutral) 264 and 300 nm (c 22,350 and 4230) ; = (acidic)260 and 333 nm (c 19,900 and 4200).Elution with benzene-chloroform ( 1 : 3) gave 3-methyl- l-p-tolylpyrrolidin-2-one,Action of Acid on the Base B (19; R = Me).-Base B wasrecovered unchanged after being boiled under reflux inchloroform with benzoic acid for 7 days.The base (0.26 g) was boiled under reflux for 10 h with2~-hydrochloric acid (10 ml). The solution was thencooled, basified (NH,OH) , and extracted with chloroform.Removal of the solvent from the extract yielded a darksyrup (0.22 g), RF 0.45, which was chromatographed onsilica.Elution with chloroform gave a syrup (0.14 g)(Found: Mf, 320. C,,H,8N2 requires M, 320) ; z 2-8-3.7(8H, m, Ar), 6.4-7.5 (7H, m, 3 x NCH, and NCH), 7.72(6H, s, 2 x ArMe), and 7.8-8.8 (7H, m, remaining CH,and CH).Reduction of 1-p-Tolylpiperidin-Zone (29; R = Me) .-Rp 0.15 (1.73 g)1974 891l-p-Tolylpiperidin-one (4 g) was reduced as above exceptthat the reaction and work-up were carried out in the darkso far as possible. The crude product was a yellow, viscousoil (3.6 g), consisting of a mixture of at least three com-ponents, RF 0.92, 0.57, and 0.10 (alumina; benzene) ;0.90, 0.54, and 0.10 (silica; benzene). The mass spectrumof this showed 3 molecular ion peaks, M+, 189, 175, and 173.Chromatography on alumina, and elution with benzene-light petroleum (1 : 2) gave l-$-tolylpiperidine, RF 0.92(0.43 g). (Found: Mf, 175. Calc. for C,,H,,N: M , 175).Elution with benzene afforded 1,2,3,4-tetrahydro-l-$-tolyl-pyridine, R p 0-57 (0.39 g) (Found: M+, 173. Calc. forC1,H,,N: M , 173), which turned red on exposure to light.The alumina column became bright red; and more polarsolvents failed to extract further material from it.In a similar experiment, the crude product was chromato-graphed on silica. Elution with benzene-light petroleum(1 : 2) gave 1-p-tolylpiperidine, Rzp 0.90 (0.37 g ) . Elutionwith benzene afforded 1,2,3,4-tetrahydro-l-$-tolylpyridine,RF 0.54 (0-26 g ) . Elution with chloroform yielded 1-p-tolylpiperidin-one (1.8 g) .Reduction of l-Phenylpiperidin-2-one (29; R = H).Similar reduction of l-phenylpiperidin-2-one (4 g) andchromatography on silica afforded 1-phenylpiperidine,R p 0.90 (0-38 g) eluted by benzene-light petroleum (1 : 2),1,2,3,4-tetrahydro-l-phenylpyridine, RF 0.60 (0-21 g )(Found: M+, 159. C,,H,,N requires M , 159) (eluted bybenzene), and l-phenylpiperidin-2-one, RF 0.20 (1-7 g)(eluted by chloroform).We thank Mr. P. Kelly for the mass spectrometry, andthe S.R.C. for a research studentship (to J. D. W.).3/2144 Received, 19th October, 1973
展开▼
