1974 1723Stereochemistry of a-Halogeno-sulphoxides. Part 1V.l Halogenation ofThian 1 -OxidesBy Mauro Cinquini, Stefano Colonna, and Fernando Montanari,rsquo; Centro C.N.R. e lstituto di Chimicalndustriale dellrsquo;UniversitA, Via C. Golgi, 19, 201 33 Milano, ItalyChlorination of trans- and cis-4-p-chlorophenylthian 1 -oxide (I) and (11) affords as main product 2ax-chloro-4eq-p-chlorophenylthian 1 eg-oxide ( I I I a ) , together with minor amounts of 2eq-chloro-4eq-p-chlorophenylthianlax-oxide (IVa). Bromination of ( I ) gives a similar mixture of the bromo-sulphoxides ( I I I b ) and (IVb).In the presence of silver nitrate the main products from the trans-substrate ( I ) are the lax-oxides ( I V ) and the onlyproducts from the cis-substrate (11) are the 1 eq-oxides (111).The reactions in the absence and in the presence ofsilver ions seem to be subject to predominant thermodynamic and kinetic control, respectively.IN the a-halogenation of sulphoxides 1-3 a strict relation-ship between the stereochemical processes a t sulphurand a t carbon is always encountered, the nature of whichdepends on the structural features of the substrate. Inparticular, in alkyl aryl sulphoxides retention andinversion of configuration at sulphur are accompaniedby retention and inversion a t carbon, respectively, butin benzyl methyl sulphoxide and in the correspondingdeuteriated species retention and inversion at sulphurare accompanied by inversion and retention at carbon,re~pective1y.l~~ Even small variations in structure andin reaction conditions can give rise to changes in thestereochemistry at sulphur and/or at carbon ; neverthe-less the reaction retains in most cases its high stereo-specificity.This has been explained on the basis of thelsquo; unique rsquo; characteristics of a-sulphinyl carbanions,whose stereochemical fate probably depends on thePart 111, M. Cinquini, S. Colonna, ar,d F. Montanari,preceding paper. * 31. Cinquini, S. Colonna, R. Fornasier, and F. Montanari,J.C.S. Perkin I , 1972, 1886.P. Calzavara, amp;I. Cinquini, S. Colonna, R. Fornasier, and F.Montanari, J . Amev. Chfm. SOC., 1973, 95, 7431, and referencestherein.S. Wolfe, A. Rauk, and I. G. Csizmadia, Canad. J . Chem.,1969, 47, 113.R. R. Fraser, F. J. Schuber, and Y. Y. Wigfield, J .A m v .Chrm. SOC., 1972, 94, 8796.K. Nishihata and RZ. Nishio, J.C.S. Perkin 11, 1972, 1730.orientation of the carbanion system with respect to thesulphinyl ~ T O U P . ~ - ~Although some hypotheses can be made about thefavoured conformations of sulphoxides in the groundstate, it is more difficult to anticipate the conformationof transition states leading to a-sulphinyl carbanions,particularly in view of the discrepancies betweentheoretical calculations on model systems and experi-mental results on base-catalysed H/D exchange re-a c t i o n ~ . ~ - ~ ~ For this reason, and in order to evaluatethe importance of conformational effects on the stereo-chemistry of a-halogenation of sulphoxides, we havestudied systems with a low degree of conformationalfreedom, namely the diastereoisomeric trans- and cis-4-$-chlorophenylthian 1-oxides, (I) and (11) .ll Resultsreported recently by Marquet and Tsuchihashi on theB.J. Hutchinson, K. K. Andersen, and A. R. Katritzky, J .Amer. Chem. SOC., 1969, 91, 3839.* (a) K. Lett and .4. Marquet, Tetrahedron Letters, 1971, 2851,2855; (b) R. Lett, S. Bory, B. Moreau, and A. Marquet; ibid.,p. 3266.@ J. E. Baldwin, R. E. Hackler, and R. -11. Scott, Chem. Comm.,1969, 1416.lo (a) T. Durst, R. Viau, and M. R. McClory, J . Amer. Clrem.SOL, 1971, 93, 3077; (b) T. Durst, K. R. Fraser, M. R. McClory,R. B. Swingle, R. Viau, and Y . Y. Wigfield, Canad. J. C h e w ,1970, 48, 2148.l1 Preliminary communication, 31. Cinquini, S. Colonna, U.Folli, and F.Montanari, Boll. xi. FRC. Chinz. ind. Bologna, 1969,27, 2031724 J.C.S. Perkin Ia-halogenation of similar thian 1-oxides 1 2 p B are com-parable to those obtained in the present work.RESULTSThe tram- and cis-4-p-chlorophenylthian 1-oxides, (I)and (11), were obtained by oxidation of the correspond-ing sulphide with (dichloroiodo) benzene in aqueouspyridine,14 and with t-butyl hypochlorite in methanol at-40deg;.1S The cis-isomer (11) was also obtained viaalkylation of the tram-sulphoxide (I) with triethyl-oxonium tetrafluoroborate followed by alkaline hydro-lysis.16 a-Halogenations were carried out, as describedelsewhere, with (dichloroiodo) benzene,l' N-chlorobenzo-triazole,ls and bromine l7 in pyridine, in the presenceand in the absence of silver nitrate.The results arereported in Schemes 1 and 2.CI HH.1. C' II0II0ID) (ma1 ( ma)(it) 63 o/oR=P-CIC H ( i I 7 3 "/o(7Oo/o) 6"/0 ( 2 * 5 deg; / o )6 LSCHEME 1 (i) PhICl,, C,H,N, 24 h, 26" (in parentheses: N-chlorobenzotriazole, C5H5N, 4 h, 25") : (ii) PhICl,, AgNO,,C6H5N, 1 h, -40"The only products of the reaction, obtained in variousratios, were the 4-~-chlorophenyl-2-halogenothian 1-from (IVa), thus indicating that (IIIa) and (Wa) haveopposite configurations both at sulphur and a t carbon.As previously found with other substrates,lp3 chlorin-ation of the cis-thian l-oxide (11) and thus probablyBr HBr0 R = p - C I C S y(It) ( I Z b )I'i) No reactionf n-Me OH Ciii) 27"/* t (IYb 1SCHEME 2 (i) Br,, C5H6N, 6 days, 25"; (ii) Br,, AgNO,, C,H5N,15 h, 25"; (iii) Br,, AgNO,, CsHsN, 5 days, 25"that of the tram-isomer (I) and bromination of both (I)and (II) does not involve breaking of the sulphur-oxygen bond.Indeed the conversion of an 180-enrichedR- I + -Ii 1 Et30BFkoxides (IIIa and b) and (IVa and b). The configurations R (iil Nal80Hof the sulphinyl amp;oups' in the bromo-derivatives (IIIb)and (IVb) were established by reduction with zinc andmethanol l-3 to the corresponding sulphoxides (I) and(11). The configurations a t the halogenated carbonatoms in (IIIa and b) and (IVa and b) and a t thesulphinyl groups in the chloro-derivatives (IIIa) and(IVa) were deduced from 100 MHz lH n.m.r. data, withthe aid of the shift reagent Eu(dpm), (see Experimentalsection).Further evidence of configuration was ob-tained by conversion of (IIIa), via reaction with tri-ethyloxonium tetrafluoroborate and subsequent hydro-lysis, into the epimeric chloro-sulphoxide (V) with theinverse configuration at sulphur. Compound (V) differsl2 S. Bory, R. Lett, B. Moreau, and A. Marquet, Compt. rend.,1973, 276C, 1323.13 (a) S. Iriuchijima, M. Ishibashi, and G. Tsuchihasbi, Bull.Chem. SOC. Japan, 1973, 46, 921; (b) S. Iriuchijima and G.Tsuchihashi, ibid., p. 929.14 G. Barbieri, M. Cinquini, S. Colonna, and F. Montanari, J .Chem. SOC. (C), 1968, 659.PhlClZ -1C1sample of (11) gave compound (IIIa) with the sameisotopic content.1965, 87, 1109.(b) C. R. Johnsonand D. McCants, jun., ibid., 1966,87,6404.15 C.R. Johnson and D. McCants, jun., J . Amer. Ckem. Soc.,l6 (a) C. R. Johnson, J . Amer. Chem. SOC., 1963, 85. 1020;l7 M. Cinquini and S. Colonna, J.C.S. Perkh I, 1972, 1883.18 M. Cinquini and S. Colonna, Synthesis, 1972,2691974 1726;Chlorination of the trans-thian l-oxide (I) with(dichloroiodo) benzene and with N-chlorobenzotriazole(Scheme 1) afforded mainly (IIIa), with retention ofconfiguration at sulphur and introduction of the chlorineinto the axial position (cis with respect to both thesulphinyl oxygen atom and the aryl group). The minorproduct (IVa) has an axial sulphinyl oxygen atom and anequatorial chloro-substituent, i.e. configuration is in-verted a t both centres with respect to (IIIa). Bromin-ation of (I) is much slower than chlorination and led toa 2 : 1 mixture of bromo-sulphoxides (IIIb) and (IVb),the configurations of which are identical with those of(IIIa) and (IVa), respectively (Scheme 2).Chlorination of the cis-isomer (11) with either of thetwo reagents gave a mixture similar to that obtainedfrom (I); with bromine no reaction occurred.In the presence of silver nitrate all the reactions werefaster: the trans-sulphoxide (I) still afforded a mixtureof stereoisomers (IIIa) and (IVa) or (IIIb) and (IVb)but in every case the ratio between the isomers wassubstantially different from that obtained in the absenceof silver nitrate.Indeed (IVa) and (IVb), of invertedconfiguration at sulphur and with an equatorial halogeno-substituent, were the main products uhder these con-ditions.The cis-isomer (11) afforded only one product,the chloro-derivative (IIIa) or (in low yield) the analogousbromo-derivative (IIIb).In contrast with the behaviour of conformationallyfree substrates, halogenation of the thians (I) and (11)in the absence of catalysis by silver ions thus seemslargely subject to thermodynamic control. Whateverthe configuration of the starting sulphoxide and thenature of the halogen, compounds (IIIa and b) are infact the main products. The isomers (IVa and b) seemon the contrary to be the result of prevailing kineticcontrol of the halogenation of the trans-sulphoxide (I).DISCUSSIONThe results of the uncatalysed reaction tally withthose obtained by Marquet and Tsuchihashi for theanalogous 4-phenyl-, 4-chloro-, and 4-t-butyl-thian 1-oxides.l29 l3 Since the diastereoisomeric a-halogeno-sulphoxides formed were characterized by a trans-diaxial or a tram-diequatorial arrangement of thehalogen with respect to the sulphinyl lone pair, Marquet l2and Tsuchihashi l3 proposed an elimination-additionmechanism, involving the ylide (V) (Scheme 3).Elimination in the halogeno-oxosulphonium salts (VII)could occur either through a boat conformation, whichwould allow a tram-diaxial arrangement of halogen atomand proton, or after isomerization to the epimeric salt(VI) .The elimination-addition mechanism is supportedby the occurrence of preferential halogenation of2-methylthian l-oxide (VIII) a t the more substituteda-carbon atom,12 but it cannot account for the stereo-specificity of halogenation met with in acylic systems,and in particular for the high optical purity of theproducts derived from optically active s~lphoxides.~-~On the other hand, it seems reasonable that a singlefundamental mechanism should be operating in thereactions of all the substrates.We previously pro-po~ed,l-~ in agreement with kinetic data,lg that form-ation of an intermediate halogeno-oxosulphonium saltYIIL)SCHEME 3(IX) is followed by a base-promoted proton abstractionand by a concerted migration of halogen to the a-carbonatom (Scheme 4).X0 0 0(Ec)SCHEME 4The stereochemistry of the second step depends onthe nature of the substrate and the reaction conditions.It is in fact determined by the conformational equili-brium of (IX) in the transition state, by whether thehalogen migrates as cation or anion, and by the abilityof an incipient carbanion to maintain or invert itsconfiguration.The experimental results imply that a loss of con-formational freedom makes the stereochemistry of thesereactions more complex.In conformationally freesubstrates, such as alkyl aryl sulphoxides, protonabstraction and halogen migration are synchronous orhighly concerted.l-U This does not necessarily applyto thian l-oxides. Small variations in bond angles inrigid systems have been found by Fraser5 to causerelevant differences in the rates of formation of a-sulphinyl carbanions from apparently similar pairs ofdiastereotopic protons.Furthermore, Wolfersquo;s * andNishiorsquo;s results, as well as our work on benzylicsulphoxides,l seem to indicate that the configurations ofa-sulphinyl carbanions and of the reaction products areinfluenced only by the preferred conformation of thesystem. Small structural variations, together with aM. Cinquini, S. Colonna, and D. Landini, J.C.S. Perkin 11,1972, 2961726 J.C.S. Perkinrestricted conformational freedom could therefore, inthian l-oxides, dramatically influence both the ab-straction of an axial or equatorial proton, and retentionor inversion of configuration. Conformationally fixedsulphoxides appear indeed to be less reactive than open-chain systems.From all this one may deduce the occurrence, incyclic systems, of a less concerted mechanism, possiblyinvolving the ylides (X) and (XI) (Scheme 5 ) .A fast!s:0 1x1Irm,SCHEME 5rearrangement , as experimentally found in the silvercatalysed reactions, should lead to the halogeno-sulph-oxides (IV) and (111) from the two ylides, respectively,if one assumes a minimum of molecular reorganization.The rearrangement may occur through intimate ionpairs, leading in the case of a non-catalysed reaction toan equilibration of the ylides (X) and (XI), and thus tothermodynamic control.* Such a mechanism, althoughnot substantially different from that proposed byMarquet and Tsuchihashi, can be seen as a borderlinecase of the more general mechanism of halogenation ofsulphoxides already proposed by us.tEXPERIMENTAL4-p-Chlorophenylthian 1-Oxides (I) and (11) .-These wereobtained by oxidation of the corresponding thian with(dich1oroiodo)benzene in aqueous pyridine l4 and witht-butyl hypochlorite in methanol at - 40",15 respectively.,9n alternative synthesis of the cis-isomer (11) (60 yield)involved inversion of the trans-isomer (I) via alkylationwith triethyloxonium tetrafluoroborate, followed by hydro-lysis with dilute aqueous sodium hydroxide.16cis-4-p-Chlorophenylthian 1-180Oxide.-Inversion of thetvans-isomer (I) was carried out in the presence of Na180H* Equilibration via halogen exchange in (VI) and (VII) is alsopossible.t Added in proof: a-Halogenation of thian l-oxides hasrecently been studied by Klein.20$We thank Miss A.Marquet (Paris) for the lH n.ni.r. dataobtained in the presence of Eu(dpm),.(10 180) to give the cis-sulphoside, n1.p. 167--168O, 10l 8 0 (mass spectrometry).Chlorination of 4-p-Chlorophenylthian l-Oxides (I) and (11).Ahlorination with an equimolecular amount of (dichloro-iodo)benzene in aqueous pyridine was carried out aspreviously described." Treatment of the cis-isomer for24 h gave 2ax-chloro-4eq-p-chlorophenylthian leq-oxide(IIIa) (73y0), m.p. 123-125" (decomp.) (Found: C, 50.2;H, 4.65. C,,H,,Cl,OS requires C, 50.2; H, 4-6), togetherwith 2eq-chloro-4eq-p-chZoropheny1tJ:ian lax-oxide (IVa)(6), m.p. 155-157' (decomp.) (Found: C, 50.25; H,4.7). The two compounds were separated by columnchromatography on silica with ethyl acetate as eluant.Under the same conditions the trans-isomer (11) afforded amixture of (IIIa) (83) and (IVa) (4).In the presence of silver nitrate, chlorination of (I)afforded, after 1 h at -40", compounds (IIIa) (16) and(IVa) (26); under the same conditions compound (11)only gave (IIIa), in 63 yield.With N-chlorobenzotriazole as halogenating agent and areaction time of 4 h a mixture of (IIIa) (65) and (IVa)(6) was obtained from the trans-sulphoxide (I).Fromthe cis-isomer (11), the yields of (IIIa) and (IVa) were 70and 2-5y0, respectively.lH N.m.r. Spectra.-Spectra were recorded on a VarianA 60 and/or a JEOL 100 MHz instrument with CDCl, assolvent, and Me,Si as internal standard. The equatorialconfiguration of the chlorine atom in (IVa) was shown bythe presence of an axial H, (Jm,m 11, Jnz,ep 4 Hz).In theisomer (IIIa) axial configuration of the chlorine atom wasshown by the absence of axial-axial coupling and theobservation of Jaz,ep (2.5) and Jcp,eq (2.3 Hz). The con-figurations of the sulphmyl groups in (IIIa) and (IVa) wereestablished by running the spectra in the presence ofEu(dpm),. In compound (IVa) the strong deshielding ofthe two p-axial protons indicated the SO group to be axial(signals for all the protons were identified by double-resonance experiments). In the isomer (IIIa) all theprotons are strongly shielded, in agreement with anequatorial configuration of the SO group.$In the case of the bromo-sulphoxide (IVb), the equatorialconfiguration of the halogen atom was established on thebasis of the values of Jaz,.az (12) and Jaaen (4 Hz) (for euro;la).The axial configuration of the bromine atom in compound(IIIb) is in agreement with the 17alues of Jaz,ep (3) andJep.eq (3 Hz)-Bromination of the Diastereoisoiizeric TJaian l-Oxides (I)and (II).-This was carried out with bromine in pyridine inthe presence of silver nitrate, as previously described.17After 15 h at room temperature, the trans-isomer (I) gave2eq-bromo-4eq-p-chlorophenylthian lax-oxide (IVb) (30y0),m.p.159-160" (decomp.) (Found: C, 42.9; H, 4.05.C,,H,,BrClOS requires C, 42-95; H, 3.95) , together with2ax-bromo-4eq-p-chlorophenyZthian leq-oxide (IIIb) (6),m.p. 139--140deg; (decomp.) (Found: C, 43.1; H, 4-05y0).The two compounds were separated by column chromato-graphy on silica with diethyl ether as eluant.Under the same conditions, after 5 days at room tem-perature, the cis-isomer (11) afforded (IIIb) (27) as theonly product.In the absence of silver nitrate a mixture of (IIIb) (42)and ( I n ) (2176) was obtained aeer 6 days at room tem-perature from the trans-isomer (I). The cis-isomer (11) did2o J. Klein and H. Stollar, J . .41izrr. CJrem. SOC., 1973, 95,74371974 1727not react, even bsol;bsol;-hen longer reaction times (12 days) wereemployed.Inversion of the Chloro-sulphoxide (IIIa) .--This wascarried out according to Johnson's method l6 to give a 50yield of 2ax-chloro-4eq-p-chZorophenylthian lax-oxide (V),m.p. 66-67" (decornp.) (from light petroleum-cyclohexane,2 : 1) (Found: C, 50.25; H, 4.55. C1,H1,Cl,OS requiresC, 50.2; H, 4.6).Redztctiozt of the Bvol.iso-sztlplLox.ides (IIIb) aiid (IVb) .-This was effected with zinc dust in methanol in the presenceof a few drops of concentrated sulphuric acid at roomtemperature for 30 min.afforded the sulphoxides (I) (40) and (11) (30), re-spectively, together with unchanged material which wasseparated by column chromatography on silica (eluantchloroform-methanol, 9 : 1).Work-up as previously described4/183 Received, 30th Janiravy, 1974
展开▼
