1020 J.C.S. Perkin ISpectroscopic Properties of Axially and Equaforially Substitutedp-Trimethylstannyl Ketones and Compounds with Related ChromophoresBy John Hudec, Chemistry Department, The University, Southampton SO9 5NHThe synthesis of two optically active cyclohexanones, with a P-trimethylstannyl substituent, are described andtheir c.d. spectra are reported. Whereas the axial SnMes group confers a dissignate (anti-octant) contributionon the n+nt transition relative to the parent unsubstituted ketone, the equatorial SnMe, group contributesstrongly in a consignate (octant) sense. This contrasting behaviour is rationalized in terms of a conformationallydependent through-bond coupling between the lone pair on the carbonyl oxygen atom and the electron-donatingC-Sn bond which gives rise to a higher energy transition in the equatorially substituted compound.This bandis assigned to the second component of the split n 3 x " transition. The spectral properties of some Py-unsaturated ketones, in which the p orbital on the P-carbon atom is in the same orientation with respect to thecarbonyl group as the equatorial C-Sn bond, are identical. Thus the high energy band, formerly assigned to acharge transfer transition, should be reassigned as a second n+x+ transition.WE have related empirically the conformational require-ments of the lone pair on a nitrogen atom in a chiralaxial a-amino-ketone to the sign of the Cotton effect of( 2 )the red-shifted n+x* transiti0n.l The N atom induces are4tively strong consignate (octant) contribution whenthe lone pair is trans-antiperiplanar to the Ca-C(=O) bondsee (l), and a dissignate (anti-octant) contributionwhen it is gauche to the same C-C bond see (2)l.t Anew transition of opposite Cotton effect to the HZ* oneis observed at higher energy in the former case (1).The protonation of (1) causes several changes in thec.d.spectrum: (a) the consignate contribution of N tothe n-n* transition changes to a dissignate contribution ;(b) its red shift decreases; and (c) the high energy transi-t The contribution of the N atom is evaluated relative to theparent unsubstituted ketone or to a ketone in which the N atom isreplaced by a C atom, using 6Ae = Ar (amino-ketone) - AE(parent ketone). Throughout this paper a positive 6Ar impliesconsignate contribution and negative 6Ar a dissignate one, thesubstituent being placed in a positive rear octant.$ The coupling of the lone pair on oxygen can take place withany filled orbital provided it is correctly oriented and of compar-able energy to the no orbital.We distinguish its origin anddescribe it as a x-donor if it is derived from the n orbital of aheteroatom or a p orbital of a double bond, and a a-donor if it isderived from a C-X a-bond. A detailed discussion of x-/a-donors/acceptors can be found in L. Libit and R. Hoffmann, J .Amev. Chem. SOC., 1974, 96, 1370.tion disappears. We have thus concluded 1 that the redshift in the n+x* transition of (1) is due to a combinationof two effects: (i) the coupling of the X / X * orbitals of theC=O group with the a/a* orbitals of the C-N bond and(ii) the interaction of the N lone pairs with other orbitalsof the C=O group. Our recent calculations on variousconformers and rotamers of amino-acetone show that theN and 0 lone pairs couple strongly via the interveningCa-C(=O) bond in a conformation corresponding to (1)thus accounting for the second part of the red shift.Thehigh energy transition is now assigned to a second n+x*transition * (see Figure). As might be expected, the twon+n* transitions should have Cotton effects of oppositesign because the two n levels have opposite symmetries.The purpose of the present work was to isolate the roleof the N lone pair (a x-donor $) and test experimentallyits effect on the n+n* transition, without the complicat-ing effect of the axial C-N bond (cr-withdrawer). Inprinciple this can be achieved by replacing the N atom bya C atom to which is attached an atom X chosen so thatthe C-X bond acts as a a-donor,$ i.e.X must be lesselectronegative than C. We have chosen X = SnMe,for two reasons, the first being mainly synthetic and theil TrFIGUREsecond theoretical. The photoelectron spectra of tinalkyl compounds show that the first ionisation potentialis relatively low, comparable in energy to that of the Nlone pair.6J. Hudec, Chem. Comm., 1970,829.W. Klyne and D. N. Kirk, Tetrahedron Letters, 1973, 1483.3 N. L. Allinger, J. C. Tai, and M. A. Miller, J . Amer. Chem.C . C . Levin, R. Hoffmann, W.J. Hehre, and J. Hudec, J . C.S.5 A. Schweig, V. Weidner, and G. Mannel, J . OrganometallicSOC., 1966, 88. 4496.Perkin I I , 1973, 210.Chem., 1973, 64, 146.S. D. Worley, Chem. Rev., 1971,71, 2961975 1021The required compounds (3) and (4) were made by theconjugate addition of (Me,Sn),CuLi to the correspondingI c H3(LlHA( 9 )enones. In either case only one epimer, formed byaxial 1,4-addition, could be isolated by preparative t.1.c.This parallels the reported behaviour of Me,CuLi.' Thestructural assignments, and the conformation in case of(3), were confirmed by lH and lSC n.m.r. spectroscopy.The axial orientation of the trimethylstannyl group in (4)model compounds 3a- and 3~-trimethylstannylcholestanehave Sn(CH,), resonances 9.3 and 12.1 p.p.m.upfield,respectively from Me,Si.8 A similar difference in chemi-cal shift is observed for (4) (-7.5 p.p.m.) and (3) (-11.1p.p.m.), indicating an axial and an equatorial Me,Sngroup. The axial orientation of the C-methyl group in(3) follows from the upfield shift of the C-5 resonance(20.1 p.p.m.) in comparison with the C-3 resonance in3p-trimethylstannylcholestane (25.2 p.p.m.) , as well asfrom the upfield shift of the 3-CH3 resonance (20.7 p.p.m.)in (3) in comparison with that in 3-methylcyclohexanone(22-9 p.p.m.) where it is oriented equatorially. Theseupfield chemical shifts of the terminal C atoms of agauche-butane fragment relative to a tram-butane frag-ment are the result of the y-effe~t,~ a through-space inter?action.If the Me,Sn and Me groups in (3) were cis-diequatorial, the Me group should have the same chemicalshift as in 3-methylcyclohexanone because the amp;effect ofan equatorial Me,Sn group is amp;O-0 p.p.m.8p10DISCUSSIONThe c.d. and U.V. data (Table) show that when the lonepair on N is replaced by an electron-donating Q bond, theexpected spectral behaviour does occur, perhaps evenmore strongly than was anticipated. Thus (3) has twon+n* bands of opposite sign of comparable Ac, as has (5).However, a more detailed comparison between (3) and (53is difficult on account of the conformational inversion ofCompound(3S, 5R)-3-Methyl-6-trimethylstannylcyclohexanone (3)(3R)-3-Methylcyclohexanone(4R, 4aS, 8aR) -4a-Methyl-4-trimethylstannylperhydronaphthalen-2-one 294229(4aR, 8aR) -4a-Methylperhydronaphthalen-2-one(lR,6S)-Tropin-2-one (6)(lR,6R)-Bicyclo3.2.loctan-2-one Q (9aR)-Quinolizidin-l-one (6) b(4aR,8aS) -Perhydronaphthalen-l-one3a-Trimethylstannylcholestane3 P-Trime th yls tan n ylcholes taneC.d. U.V.Solvent * Amax./mn e-298 + 6-61 M220 - 9.06 216sh 1150290 + 0-69 M296.6 89204289319228297300203210* M = methanol; E = ethanol; D = dioxan; 10 = iso-octane; H = n-hexane.0 H.M. Walborsky, M. E. Baum, and A. A. Youssef, J . Amer. Chem. SOC., 1961, 83,J.C.S. Perkin I , 1974, 1076.- 1-50 286 33 + 0.66 M Continuously rising-2.06 to end absorption- 2.40- 1.24 M-/- 2.4 E 314 68-4.1 228 600- 0.84 D 287 18- 1.42 I0 300 37 + 1.00 M 280 43 + 1.4 E + 2.2 H tt Insoluble in both methanol and ethanol.988. b Ref.13. CD. N. Kirkand W. Klyne,follows from the vicinal coupling constants of the ABCsystem. The geminal AB protons adjacent to the C=Ogroup, JAB -14.9 Hz, 6 2.73, couple with the proton C,6 1.72, with J A c 8-0 and Jsa 1-9 Hz; thus proton C mustbe equatorial. If the Me,Sn group were equatorial, theaxial proton C should have vicinal coupling constants ofca. 13.5 and 6.0 Hz. Further confirmation of the axialorientation of SnMe, in (4) comes from 13C n.m.r. The$ Detailed analysis of the c.d. data will be published latertogether with the temperature-variable 13C n.m.r. spectra.N. L. Allinger and C. K. Riew, Tetrahedron Letters, 1966,1269; J . A. Marshall and N. H.Anderson, J . Org. Chem., 1967,81,667.J. Hudec, unpublished work.the N-methyl group in (5). We will therefore discuss ingreater detail the spectral properties of (5).f.Low temperature 1 n.m.r. measurements on (6) indi-cate that the Me group exists ca. 55--70 in the axialform at room temperature. By using 6Ac = -2.4 forthe N atom with an equatorial alkyl group in (6) (whichshould be a maximal value on account of the solventdifference), we can calculate a 8Ac range of +7.4 to +5.3* D. M. Grant and B. W. Cheney, J . Amer. Chem. SOC., 1967,89, 5316, 6319; H. J. Reich, M. Jautelat, M. T. Messe, F. J.Weigert, and J . D. Roberts, J . Amer. Cltem. SOC., 1969, 91, 7446.10 D. Dodrell, I. Burfitt, W. Kitching, M. Bullpitt, C. H. Lee,R. J . Minott, J .L. Considine, H. G. Kuivilla, and R. H. Sarma,J . Amer. Chem. SOC., 1974,96,16401022 J.C.S. Perkin Ifor (5) containing 55-70 of the conformer with theaxial N methyl group. In comparison, BAE for (3) is+6-5, if we assume that the axial methyl group contri-butes ~ A E ca. 0,l1 indicating comparable x-electron don-ation from the N lone pair and m donation from the C-Snbond. The red shift in the 290 nm band (c.d.) of (3) is0.1 1 eV, whereas the value is 0.39 eV for (5) ( Amax for theparent ketone was corrected for the solvent shift dioxan-,ethanol by 7 nm). A large part of this difference is dueto interaction in (5) between the G* orbital of the C-Nbond with the x* orbital of the carbonyl group. Suchinteraction cannot be expected in (3).The negative high energy Cotton effect of (5) isassigned to the n,+x* transition; by analogy the sameassignment applies to the high energy band of (3).Analternative assignment as a m+a* transition of Sn-Cis most unlikely as it occurs at even higher energy l2 (seealso c.d. of 3- and 3~-trimethylstannylcho~estane in theTable). The through-bond coupling of the nitrogen andoxygen lone pairs of (5) has a splitting of n, and n, levelsof 1.67 eV (Figure); the corresponding splitting for (3)is 1.47 eV, resulting from the coupling of the oxygen lonepair and the C-Sn bond (Figure; N replaced by aesn.The comparison of (4) and (6) l3 is difficult on accountof the lack of U.V. and c.d. data below 250 nm. Never-theless, both (4) and (6) show a dissignate behaviour ofthe w x * transition, ~ A E -0.26 and -2.4 respectively.In addition, (4) has two Cotton effects at 229 and 204 nm.It is quite probable that the 229 nm band is a weaksecond f l e x * transition as the sign is opposite to that ofthe 294 nm band.Indeed, our calculations4 predictsuch behaviour for a-amino-ketones such as (6). Theno and flN orbitals can still couple via the C,-C(=O) bondbut weakly, resulting in smaller splitting of the new n, and32, levels with the result that the two n+x* transitionsmove closer together. Thus it is possible that (6) has aweak positive band below 250 nm-indeed, the repro-duced spectra l3 indicate this. The nature of the 204 nmband of (5) is not certain; it may be a c+c* or an no+a*transition.The isotropic absorption of (3) and (5) parallels the c.d.both in transition probability and in Amax.for the n,+x*transition. The higher energy band is visible as ashoulder only for (3).With the analogy established between a a-donor C-Xbond and a x-donor lone pair on N with respect to theireffects on the spectral properties of ketones whenproperly oriented as in (3) and (6), it is relatively easy togeneralise the concept of orbital coupling to other sub-stituents. We predict similar spectral behaviour fromother m-donors such as SiR,, GeR,, HgR, and aqua- orl1 G. Snatzke, B. Ehrig, and J. Klein, Tetrahedron, 1969, 25,6601.la C. W. N. Cumper, A. Melinkoff, and A. I. Vogel, J . Chem.SOC. ( A ) , 1966,242; P. P. Shorygin, V.A. Petukov, 0. M. Nefedov,S. P. Kdesnikov, and V. I. Shiryaer, Teor. eksp. Khim., 1966, 2,190.l8 S. Yamada and T. Kunieda, Chem. and Pharm. Bull. (Japan),1967,16,490; S. F. Mason, K. Schofield, and R. J. Wells, J. Chem.SOC. (C), 1967, 626.l4 G. Snatzke and B. Wolfram, Tetrahedron, 1972, 28, 665.pyridine-bisdimethylglyoximatocobalt complexes. Incase of heteroatom x-donors other than N, complexspectra explainable in terms of through-bond couplinghave been observed only in the case of SJ8~14-16 one ex-ample being (7),14 although they can be also expectedwith Se, P, and probably I (axial a-iodo-ketones). Itappears that the main requirement, apart from thecorrect relative orientation of the no and nx orbitals, isthe nx (or C-X) energy level which should be very closeto no or lower.The recent work l6 on the c.d.of l-methyl- and l-ethyl-6-methylpiperidin-3-one ( 8 ; X = NMe or NEt) and 6-methyltetrahydrothiopyran-3-one (8 ; X = S) requiressome comment in the light of the above conclusions.The reported A,, (MeOH) 296 nm for (8; X = NEt)compares well with the A,, value of (6) and thus the lone-pair on N in (8; X = NEt) must be axial, resulting in adissignate contribution of the nitrogen atom. It isdifficult at present to rationalise the difference betweenthe magnitudes of the dissignate 8Ac contributions in (6)(8Ac 2.4) and (8; X = NEt) (0.8); it may be due to thedifference in pattern of substitution on the C atomsattached to the N atom8 (C-6 and C-2). Similar com-ment applies also to the large difference in the consignatecontributions of S in (7) l4 (8Ac +21.8) and (8; X = S)(+1.75) ; it is probably amplified further by a distortionof the tetrahydrothiopyranone ring in (8; X = S) whichcannot take place in (7).Compounds (7) and (8; X = S)have a second band of opposite chirality to the one at ca.315 nm, at 257 and 250 nm, respectively. This band hasbeen assigned in the latter case to the n+a* transitionof S.16 Our work on thioketones of the type (7) or ( 8 ;X = S) shows that this band should be assigned to thesecond n+x* transition, ryt,+x*.The last and probably most extensively studied sub-stituents that fall into this category are the double bondin a By-unsaturated ketone 17*18 and the phenyl group ofan a-axial phenyl ketone.l** l9 These chromophores ,when the relative orientation of the double bond and thecarbonyl group is as in (9), give rise to a red-shifted andgreatly enhanced n+x* transition with positive Cottoneffect as well as a high energy band of opposite sign.The latter transition was at first assigned to a chargetransfer transition Is and later to a n+x* transition.l'bComparison of (1) and (9) shows that the p orbital on CBhas the same orientation as the lone pair on N and is thusideally disposed for a through-bond coupling with no.It is thus possible to reassign the high-energy transitionof (9) to a second n,+x* transition.The amp;unsaturatedketone (9), a homoconjugated n-system, has beenclassified as an inherently disymmetric chromophore l816 C.H. Robinson, L. Milewich, G. Snatzke, W. Klyne, andS. R. Wallis, J . Chem. SOC. (C), 1968, 1245.16 M. M. Cook and C. Djerassi, J . Amer. Chem. SOC., 1973, 95,3678.17 (a) R. C. Cookson and N. S. Wariyar, J . Chem. SOL, 1966,2302; (b) D. E. Bays and R. C. Cookson, J . Chem. SOC. (B), 1967,226.18 K. Mislow, M. A. W. Glass, A. Moskowitz, and C. Djerassi,J . Amer. Chem. Soc., 1961, 88, 2771; 1962, 84, 1946.19 R. C. Cookson and J. Hudec, J . Chem. Soc., 1962,429; R. C.Cookson and S. Mackenzie, Proc. Chem. SOC., 1961, 4231975 1023on account of its very high SAE values relative to thesaturated ketone or to an axial a-halogeno-ketone, whichare classed as asymmetrically perturbed symmetricchromophores.l* Although this distinction was valuableoriginally, we consider that its retention is no longernecessary.2oEXPERIMENTAL1H N.m.r. spectra were measured for solutions in CDCl,with a Varian HR-100 and 13C n.m.r. spectra with a BrukerFT 20 instrument.lSC Assignments were made by com-parison of fully decoupled and off-resonance spectra. Abenzene lock was used for lH n.m.r. and Me,Si was em-ployed as internal reference for 13C n.m.r. The low-tempera-ture 13C runs were carried out in CD,Cl,-CS, (4 : 6) clown to153 K. Chemical shifts are quoted in p.p.m. downfield fromMe,Si. C.d. spectra were run on a JASCO UV-CD-5 instru-ment modified with an SS-20 attachment.The starting a@-unsaturated ketones were synthesized inthe following manner. Bromination of (3R)-3-methylcyclo-hexanone gave (2R,5R)-2-bromo-5-methylcyclohexanone,21m.p. 81-82', which was dehydrobrominated in refluxingcollidine to give (5R)-5-methylcyclohex-2-enone, b.p.164-170". 4a, 6,6,7,8,8a-Hexahydro-4a-methylnaphthalen-2( IH)-one was prepared as described 22 from the antipodalbicyclic intermediate 23 used for the total synthesis ofsteroids.General Method for Trimethylstannylation of up- Un-saturated Ketones.-Hexamethyldistannane (1.8 g, 5 mmol)was added to dry tetrahydrofuran (15 ml) in which weresuspended freshly cut lithium wire pieces (ca. 100 mg).The mixture was stirred under a stream of dry nitrogen for20 Cf. K. N. Houk, K. J. Northington, and R. E. Dude, J .Amer. Chem. SOC., 1972, 94, 6233.21 C. Djerassi, L.E. Geller, and E. J. Eisenbraun, J . Org. Chem.,1960, 25, 1.3-4 h a t room temperature. The black solution of tri-methylstannyl-lithium was transferred (syringe) into anotherflask containing copper(1) iodide (60 mg) under dry nitrogen.The mixture was stirred for 1 h, and the @-unsaturatedketone (7-8 mmol) was then added. Stirring a t roomtemperature was continued for another 2 h, then the tetra-hydrofuran was slowly distilled off (steam-bath) to onequarter of the original volume in 30-45 min. The mixturewas diluted with water and filtered through Celite. The resi-due was well washed with ether, which was used for extrac-tion of the filtrate. The extract was washed 5 times withwater, dried (Na,SO,), filtered, and evaporated. Theresidue was chromatographed on silica gel plates (0.8 mmthick) with 10 ether in n-pentane as the eluant.(3S, 5R) -3-Methyl-5-trimethylstannylcyclohexanone (3)had b.p. ca. 80' a t 0.1 mmHg, 8, -0-04 (MeSn, Jsn,= 51 and53 Hz) and 0.87 (Me, J 7.0 Hz); 8, -11.1 (Me,), 20.1(C-5), 20.7 (Me), 34.2 (C-3), 37.2 (C-a), 45.2 (C-6), 49.1 (C-2),and 211.7 (C-1); vmaL (neat) 1705 and 865 cm-l.(4R,4aS, 8aR)-4a-Methyl-4-trimethylstannylperhydro-naphthalen-2-one (4) had m.p. 58-60'; BE 0.17 (Me,,J 51 and 53 Hz), 1.18 (Me), 1.72 (Ha, J 8.0 and 1-9 Hz), 2-47(HB, J 14.9 and 1.9 Hz), and 3.00 (HA, J 14.9 and 8.0 Hz);So -7.5 (Me3Sn), 17.6 (Me), 22.1 (C-6), 26.0 (C-7), 29.6 (C-8),37.5 (C-4a), 41.7 (C-a), 42.4 (C-3), 44.1 (C-8a), 45.3 (C-1), and211.1 (C-2); u,,, (Nujol) 1705 and 875 cm-l.We thank Dr. W. S. Knowles, Monsanto Chemicals, for agenerous gift of the bicyclic intermediate, Professor W. A.Graham for help and advice, and the Chemistry Department,University of Alberta, Edmonton, for facilities.4/2243 Received, 30th October, 1974122 K. B. Woodward, F. Sondheimer, D. Taub. K. Heusler,23 A. J. Speciale, J. A. Stephens, and Q. E. Thompson, J . Amer.and W. M. McLamore, J . Amer. Chem. SOL, 1962,74,4223.Chem. Soc., 1964, 76, 6011
展开▼
