摘要:

1974 853Kinetics and Mechanism of Cleavage of Sulphur-Silicon, -Tin, -German-ium, and -Lead Bonds in Aqueous Dioxan in Some Qrganometallic Com-pounds of Bivalent SulphurBy G. Pirazzini, R. Danieli, A. Ricci,' and C. A. Boicelli, Laboratorio dei composti del carbonio contenentieteroatomi e loro applicazioni del Consiglio Nazionale delle Ricerche, Via Tolara di Sotto 89, Ozzano Emilia,Bologna, ItalyRates of cleavage have been measured spectrophotometrically for some PhS-M R, compounds (M = Si, Ge, Sn, andPb) in neutral and acidic aqueous dioxan. The reactivity sequences are unrelated to the electronegativity of themetals and the ease of cleavage increases in the order Si < Ge -= Sn < Pb. The substituent and solvent isotopeeffects indicate some differences in the cleavage mechanism of these compounds.This is governed by sulphur-metal bond polarization enhanced in the heavier Sn and Pb derivatives by co-ordination between water and themetal. A dual mechanism which fits the experimental results is proposed.WE have previously reported1 a kinetic study of thehydrolysis in aqueous dioxan of some thiosilanes andthe cleavage mechanism of the sulphur-silicon bond inacidic and basic media. In order to extend this investi-gation on the properties and reactivity of sulphur-metal bonds to the other Group IVB elements, we havenow studied spectrophotometrically the hydrolysis of theorganogermanium, organotin, and organolead analogues.RESULTS AND DISCUSSIONIn aqueous dioxan trimethyl(phenylthi0)-silane and-germane are very sensitive to hydrolysis, the productsbeing benzenethiol and hexamethyl-disiloxane or -digerm-oxane.In contrast to the reactivity of these com-pounds the corresponding tin and lead derivatives arePseudo-first-order rate constants for neutral and acidhydrolysis with varying amounts of water are listed inTable 1. Substituent effects in the acid catalysedhydrolysis of the silicon and tin derivatives are reportedin Table 2. For both systems a plot of log k againsta,, (ref. 2) is linear but for phenylthiostannanes the slopep of 0.3 is significantly smaller than that (1-4) for phenyl-thiosilanes.Solvent isotope effects have also been measured andthe results are collected in Table 3. One interestingfeature of these results is that, for both the compoundsinvestigated, there is a positive solvent isotope effectwhich increases from 1.87 to 3.0 going from trimethyl-(pheny1thio)-silane to -tin.The effects of groups bonded to the metal on the rateTABLE 1First-order rate constants at 25' for the hydrolysis of Me,MSPh in aqueous dioxanNeutral medium Acid medium A/nmM H,O (yo) a = 33 0.26 0.40 0.80 1.33 2-601 0 3 ~ IS-'cI \ Si 44,670 0.26 C 0.40 1.00 280Ge 230 0.12 265Sn No reaction 20.9 1327 31,630 260Pb No reaction 79,430 272a rnl per 100 ml solution.8.0 x 10-5~-HC1. In the acid hydrolysis of the Si and Ge compounds a t 0.8% water, allowancemust be made for neutral hydrolysis which affects the rates by a factor not exceeding 6%.hydrolytically stable in a neutral medium even inthe presence of a large excess of water.In an acid medium however the hydrolysis of the S-Snand S-Pb bonds is much faster than those of the corre-sponding Si and Ge derivatives.The cleavage of thephenylthio-derivatives of silicon and germanium in thepresence of hydrochloric acid leads to the products ob-tained from neutral hydrolysis, whereas trimethyltinchloride and trimethyl-lead chloride are obtained besidebenzenethiol in the acid hydrolysis of the phenylthio-tinand -lead derivatives [reactions (la and b)].2R,MSPh + 2H20 +- HC1-BR,MOH + 2PhSH + HC1 +M = Si, Ge (la)(R3M),O + H2OR,MSPh + H20 + HC1-R3MC1 + PhSH + H,OM = Sn, Pb (lb)of hydrolysis of Si and Sn derivatives are shown inTable 4.TABLE 2First-order rate constants for the acid catalysed cleavageof XC,H,SMMe, in aqueous dioxan at 25 & 0.1'Xp-OMep-MeHm-OMem-C1p-Cla 1.33% v / V8 x 10-6~-HC1.M = Si 0 , ~ M = S n b1 03k/s-l k1s-l0.2 3 25-120.29 25.120.62 31-640.871-793-23 39-81H,O; 13 x ~O-~M-HC~.b 0 4 % v/v H,O;c Corrected for neutral hydrolysis.The results for cleavage in acid medium of PhSXMe,(Table 1) exhibit a sequence similar to that observed inthe acid catalysed hydrolysis of ArMR, bonds byR. Danieli and A. Ricci, J.C.S. Perkin 11, 1972, 1471.R. W. Taft, jun., J . Phys. Chem., 1960,64, 1806854 J.C.S. Perkin I1aqueous ethanolic perchloric acid.3 The electronegati-vity of the metals is unrelated to the reactivity sequenceand does not correlate with the large difference in ratesTABLE 3Isotope effect on the acid hydrolysis of Me,MSPhIO5[HC1]/ H 2 0 103Pa,o/ 103k~,o/ ~ H ~ o /M &I (%VlV> S-1 S-1 kDzOSi 133 1.33 6-92 3.70 1-87Sn 8.0 0.8 31,630 10,540 3.00@ Taken from ref.1.TABLE 4Relative rates (&l) for the cleavage of R,MSPh byaqueous dioxan in the presence of HCL at 25 & 0.1"M Me Et Prn PhSi a 85 1.2 1Sn b 14 8 1 0-04a 1.33 X IO-,M-HCI; 1.337; H20. ' 8.0 X 10-5M-HC1; 0.87;H20.between the S-Si and S-Sn compounds, whatever electro-negativity scale is used.* Trimethyl(pheny1thio)-tinand -lead derivatives are cleaved by acid lo5 timesfaster than the silicon and germanium analogues andthis requires considerable electron density to be trans-ferred to the reaction centre in the transition state forthe first two compounds.Therefore the main factorlikely to govern the relative rates of acid cleavage ofPhSMMe, is the ease of electron release from MMe,.OR the electronic effect of MR,groups (M = Si, Ge, Sn, and Pb) suggest that the trendin properties of these Group IVB elements can begenerally interpreted in terms of inductive electronsupply and of electron withdrawal by a pn-+- d,,mechanism. Recently good evidence has been foundthat in addition to the inductive release of electrons byan MR, group hyperconjugative contributions are alsolargely responsible for the relative reactivity of organo-metallic derivatives containing C-M bonds. In sub--strates containing sulphur-metal bonds no conclusiveevidence for either p,, --)- d, inductive or hypercon-jugative contributions has yet been produced, How-ever from the 13C n.m.r.data in Table 5 the negligibleshielding at the para-carbon atom and the trend ofchemical shifts for aliphatic and substituted carbonatoms, coupled with the S2p,,z ionization energies de-rived from e.s.c.a. measurements reported earlier 7suggest greater importance in the ground state of thesemolecules of the -M effect than of electron release.Moreover recent evidence 8 from mass spectrometricmeasurements exhibits an unusual trend of ionizationC. Eaborn and K. C. Pande, J . Chem. SOC., 1960, 1566.A. Allred and E. G. Rochow, J . Inorg. Nuclear Chein., 1958,5, 269.R. W. Bott, C. Eaborn, and D.R. M. Walton, J . Organo-metallic Chem., 1964, 2, 154; H. Soffer and T. De Vries, J , Amev.Chem. Soc., 1951, 73, 6817; J. D. Roberts, E. A. McEhill, and K.Armstrong, ibid., 1949, 71, 2923; H. Freiser, M. V. Eagle, and J .Speier, ibid., 1953, 75, 2821; W. Adcock, S. Q. A. Rizvj, W.Kitching, and A. J. Smith, ibid., 1972, 94, 369; E. W. Abel, D. A.Armitage, and -4. A. Williams, Trans. Faraday Soc., 1964, 60,Extensive studies1 0 E Vpotentials (IP) with the order of electron acceptance asSi > Ge = C > Sn > Pb for PhSMMe, and as Si > Ge >Sn = C for (Me,M),S.This trend reflects quite clearly a fi,, --t d, contribu-tion superimposed on the S-M bond strengths for the Siand Ge derivatives. The picture is less clear for the tinand lead analogues: the presence of a d,, (Sn) -t p,, (S)interaction supported by the electronic spectra ofMe,SnSC,H,X compounds is somewhat in contrast withTABLE 5Substituent 13C chemical shifts (p.p.m.) of PhSMMe, invarious solventsDioxan-CCI, a CDCI, Me2S0 CP6 *-* * M C-S CH, C-S CH, CH, CH, C-S CH,C 133.0 32.0 135-6 30.7Si 131.6 2.0 130.1 1.4 131.4 1.63Ge 133.4 1.5Sn 134.8 -3.0 133.6 -3.8 137-1 -2.63 136.4 -4.80Pb 138.0 10.3 133.5 8.56 Taken from ref.7.the IP sequence from which Me,Sn and Me,Pb behave asweak electron donors. From this spectroscopic evidencethe ground and excited state electronic properties ofthese compounds seem better interpreted in terms ofd,(M) _+ 9, (S) interactions whereas hyperconjugativecontributions represented by structures such as (I) areprobably less important.( 1 )To fit the spectroscopic properties to the kinetic resultsin acid medium (Table 1) which show for PhSMMe, anorder of electron release Me,Pb > Me,Sn 9 Me,GeMe,Si, it is also important to consider the effects of themedium.Comparison of the results in Table 5 showsthat whereas 13C chemical shifts for the silicon derivativesare insensitive to the changes in the solvent, trimethyl-(pheny1thio)tin exhibits a larger dependence of shift onsolvent owing to interactions between solvents and themetal. In the aqueous dioxan used for the kineticexperiments the extent of the interaction between waterand the metal is shown by lH n.m.r. spectroscopy sincewater molecules bound to a metal have different chemicalshifts from those of the bulk solvent.1° The lH chemicalshifts recorded in partially deuteriated dioxan arereported in Table 6: the downfield shift of the watersignal on changing the metal from silicon to lead,A.R. Bassindale, C. Eaborn, D. R. M. Walton, and D. J.Young, J . Organometallic Chem., 1969, 20, 49; A. R. Bassindale,C . Eaborn, and D. R. M. Walton, ibid., 1970, 21, 91; W. Han-stein, H. J. Berwin, and T. G. Traylor, J . Amer. Chem. Soc., 1970,92, 7476 and references therein; M. A. Cook, C. Eaborn, andD. R. M. Walton, J . Organometallic Chein., 1970, 24, 293. ' S. Pignataro, L. Lunazzi, C. A. Boicelli, R. Di Marino, A.Ricci, A. Mangini, and R. Danieli, Tetrahedroiz Letters, 1972, 5341.G. Di Stefano, A.Ricci, R. Danieli, A. Foffani, G. Innorta,and S. Torroni, J . Ovganometallic Chenz., 1974, 65, 205.T. A. George, J . Organometallic Chem., 1971, 31, 233.Chem. Comm., 1971, 1265 and references therein.lo J. C. Boubel, J. J. Delpuech, M. R. I<hadda.r, and A. Peguv1974 855supports stronger co-ordination of the water to the metalin descending Group IVB.TABLE 6Internal lH n.m.r. chemical shifts a t 100 MHz for thewater signal in the dioxan--water-PhSMMe, systemM H2O PhSMMe, A/Hz 6A/Hz cDioxan a- H,O-74.2 88.13C 74.2 1.4 87.75 0.38Si 74.2 1.4 87.38 0.75Ge 74.2 1.4 87.21 0.92Sn 74.2 1-4 85.50 2-65Pb 74.2 1.4 84.38 3.75(1 Partially (66.6%) deuteriated dioxan. 6 Shifts in Hz; theaccuracy is A-0-05 Hz averaged over four scans, two in theincreasing and two in the decreasing mode.6A Is thevariation of the internal shift of water due to products.A change of the solvent from CCL, or CHCl, to theaqueous dioxan used in the kinetic study is thus likelyto be important in that the strong interaction betweenwater and the metal can increase the extent of polariza-tion in the sulphur-metal bond leading to a large en-hancement of the electron-releasing over the mesomericeffect l1 for the trimethyl-tin and -lead groups in linewith the observed trend in reactivity (Table 1).The results in Tables 1 4 suggest some differences inthe cleavage mechanism of the sulphur-metal bond indescending Group IVB. Some features of these resultsand comments are as follows. (i) The increase of thesolvent isotope effect on going from the silicon (1.87) tothe tin (3.0) derivatives is large enough to be used as acriterion of mechanism l2 and suggests that in the acidcat alysed hydrolysis of trimethyl (phenylt hio) tin protontransfer from the oxonium ion to the organometalliccompound in the transition state is much more rapid.(ii) The lower sensitivity to the effect of substituents forthe tin derivatives compared with that for the silicon ana-logues provides some support to nucleophilic assistanceby water in the hydrolysis of trimethyl(pheny1thio) tin.In fact in a co-ordinative intermediate the electrons ofthe highly polarizable sulphur-metal bond are readilyavailable and there is a small demand on the electronsof the ring (p 0.3) in marked contrast to that previouslyfound for the acid catalysed hydrolysis of trimethyl-(pheny1thio)silane (p 1.4).(iii) A plot of log k againstlog [H,O] for the hydrolysis of trimethyl(pheny1thio)tinin aqueous dioxan (Table 1) shows the rate is given bythe expression h[PhSSnMe,] [H20]4*5. The increasedkinetic order in water with respect to that (1.5) of tri-methyl(pheny1thio)silane in the same medium (Table 1)supports greater solvation l3 of the transition state in thehydrolysis of the tin derivative. (iv) The results inTable 4 suggest less steric hindrance for PhSSnR, thanfor the silicon analogues involving a highly ordered andcrowded transition state. The kinetically rapid forma-* Electron release irom R,M (M = Sn and Pb) is also likely tobe supplemented t o some extent by hyperconj ugative electronrelease similar to that observed for C-M bonds.6 The importanceof this contribution t o the transition state in these cleavagereactions cannot be evaluated unambiguously and deserves moredetailed study.tion of a pentacovalent tin complex thus appears plaus-ible.The simplest way of fitting these experimental resultsis based on a dual mechanism; owing to the minor roleplayed by co-ordination in the organosilicon and organo-germanium derivatives the previously established mech-anism of fast protonation of the sulphur atom followedby rate-determining attack of the nucleophile on themetal, proposed for the acid catalysed hydrolysis ofphenylthiosilanes, explains in the best way the results inTable 1 for the silicon and germanium derivatives.Thepartial double bond character of the sulphur-metal bondrevealed in these compounds by spectroscopic measure-ments 879 can also partially account for their lower re-activity .In the phenylthio-derivatives of tin and lead inaqueous dioxan, on the other hand, features (i)-(iv)suggest that the reactivity of the sulphur-metal bond isbest interpreted by considering that the driving force isstretching of the S-M bond by nucleophilic assistance ofwater. Interaction between water and the metal centrewill in fact favour polarization of the S-M bond givingrise to a certain amount of negative charge on thesulphur atom and thus facilitating the subsequent cleav-age.*k1 6 - 8+H20 4- ArSMR3 ArS.a-MR3OHZArS - - *[ HAr SH i- C1 MR, t HZOSCHEMEThe neutral and acid hydrolysis of these compoundscan thus be written as in the Scheme. In a neutralmedium bimolecular attack of water on the complexintermediate leading to the reaction products (k, inScheme) is unimportant : the fast reverse reaction( k - , 9 k,) between thiol and trimethyltin hydroxide,which instantly gives the starting trimethyl(pheny1thio)-tin (see Experimental section) stabilizes the intermediatewith respect to the final products and accounts for thestability of the tin and lead derivatives in neutral aqueousdioxan.For the acid hydrolysis of these derivatives thesimplest mechanism which best accommodates theresults in Tables 1-6 involves a rate-determining attack(k,) of a proton on the sulphur atom.This step will belargely favoured by an increase of electron release from11 A. J . Smith, W. Adcock, and W. Kitching, J . Amtv. Chem.SOG., 1970, 92, 6140.12 C. A. Bunton and V. J . Shiner, jun., J . Amer. Chem. SGC.,1961, 83, 3207, 3214.13 R. M. Prince and R. E. Timms, Inorg. Chim. Acta, 1967,129J.C.S. Perkin I1tin and lead toward sulphur owing to the prior co-ordination of water to the metal.Nucleophilic attack of the halide ion on the metal togive Me,MCl will probably occur in a fast subsequentstep even if at the present time the possibility of asynchronous process cannot be ruled out.EXPERIMENTALReagent grade dioxan purified according to standardprocedures,14 and conductivity water were used for thekinetic experiments.Materials.-t-Butyl phenyl sulphide l6 and trimethyl-(phenylthio) silane l6 were prepared according to literaturemethods.Substituted trimethyl(pheny1thio)tin deriva-tives were synthesized by adding a suitable thiol (0.026 mol)to trimethyltin chloride (0-025 mol) dissolved in water(ca. 50 ml). After addition of 1N-sodium hydroxide (25ml) the mixture was stirred for 2 h. The resulting oil wasseparated from the aqueous layer, washed with water,dried (Na,SO,), and distilled under high vacuum. Thephysical properties and yields of the products are in Table 7.TABLE 7Analytical data and physical properties ofXC,H,SSnMe,Found (yo) Required (%)B.p./"C a Yield c-, - X (p/mmHg) (%) C H S C H Sp-OMe 94-95 55 40.0 5.6 10.0 39.65 5.3 10-6p-Me 83-84 60 42.0 6.0 10.9 41.85 5.6 11.16m-C1 91-92 53 35.5 4.4 9.9 35-15 4-25 10.4(0.1)(0.3)(0.1)Uncorrected.Trimethyl(pheny1thio) germane was synthesized by a modi-fication of the procedure of Abel et a1.l' To dry sodiumthiolate (1.5 g, 0.0125 mol) in dry ether (ca.50 ml) trimethyl-germyl bromide (2.5 g, 0.0125 mol) in dry ether (20 ml) wasadded dropwise with vigorous stirring. The mixture wasrefluxed for 3 h, the solid was filtered off, and the etherremoved. The residual oil was distilled under high vacuumto give the product (2.2 g, 57%) (b.p. 59-69' a t 0.3 mmHg).Trimethyl(phenylthi0) lead was prepared by treating sodiumthiolate (1.49 g, 0.01 mol) in methanol (ca.10 ml), withtrimethyl-lead chloride (2-88 g, 0.01 mol) la in methanol(10 ml). The mixture was stirred for 30 min, filtered, andthe methanol removed. Careful distillation of the residualoil under high vacuum gave trimethyZ(phenyZthio)lead (1.7 g,62y0), b.p. 105-107° a t 0.2 mmHg (Found: C, 29.5; H,3.75. CgH,,PbS requires C, 29.9; H, 3.9%). Triphenyl-,19triethyl-, 2o and tri-n-propyl-(phenylthio) tin were syn-thesized by literature methods.Cleavage Reactions.-(a) Trimethyl(pheny1thio)silane. Totrimethyl(pheny1thio)silane (0.6 g 0.0027 mol) in dioxan(2 ml) was added lwhydrochloric acid (2.7 ml, 0.0027 mol).After 10 min a t room temperature the reaction mixture wasdried (MgSO,) . Benzenethiol and hexamethyldisiloxanel* A. Weissberger, ' Technique of Organic Chemistry,' Inter-science, New York, 1967, vol.17.l6 V. N. Ipatieff, H. Pines, and B. S. Friedman, J . Awcr. Chew.SOG., 1938, 60, 2731.l6 K. A. Henry and A. Allred, Inorg. Chem., 1965, 4, 671.l7 E. W. Abel, D. A. Armitage, and D. B. Brady, J . Organo-metallic Chem., 1966, 5, 130.were characterized by lH n.m.r. spectrum. G.1.c. (5%Carbowax 20M on Chromosorb W; 5 f t x 0.12 in; 60')of the reaction mixture afforded three peaks identified ashexamethyldisiloxane, benzenethiol, and dioxan. No evid-ence for the presence of trimethylsilanol was obtained.(b) TrimethyZ(phenyZt~zio)gervnane. To trimethyl(pheny1-thio)germane (0.5 g, 0.0022 mol) in dioxan (2 ml.) LN-hydro-chloric acid (2.4 ml, 0.0024 mol) was added. After drying(MgSO,) the mixture was carefully distilled to give hexa-rnethyldigermoxane, 22 b.p.139".(c) m-Chlorophenylthiotrirnethyltin. 1N-Hydrochloricacid (3-2 ml, 0.0032 mol) was added to a solution of m-chlorophenylthiotrimethyltin ( 1 g, 0.0032 mol) in dioxan(2 ml). Distillation of the mixture gave initially dioxanand then trimethyltin chloride, b.p. 152", and m-chloro-thiophenol.(d) Trimet~yZ(phenyZthio)leaa. To trimethyl(phenylthi0)-lead (1 g, 0.0027 mol) in dioxan (2 ml) IN-hydrochloric acid(3 ml, 0.003 mol) was added. After drying (MgSO,), theorganic layer was diluted with light petroleum to afford,on cooling, trimethyl-lead chloride,ls m.p. 195'.Rates and U.V. Measurements.-For slower runs thekinetic experiments were carried out spectrophotometricallyas previously described 1 using a Perkin-Elmer 402 spectro-photometer. For very fast reactions, the rate constantswere measured using a Durrum-G type stopped flow appara-tus.After the rapid mixing of the dioxan-organometallicsolution (0.8-3.0 x ~O-*M) and aqueous dioxan, the re-action progress was followed by observing on the oscillo-scope trace, the change of transmittance a t a suitablewavelength (Tables 1 and 2). The quantitative nature ofthe reaction was indicated by the close superimpositionupon the t, spectrum by the U.V. spectra of the thiols.Pseudo-first-order rate constants, calculated as previouslydescribed,l were averages of three or more independent runs.The experimental error was ca. &3%.Reactions between products to give starting compoundswere checked by means of U.V. spectra; reverse reactionswere shown to be unimportant for the silicon and german-ium derivatives under the conditions employed ; howeverby addition of benzenethiol (1.5 x l O W 4 ~ ) in neutralaqueous dioxan to an equimolecular amount of trimethyltinhydroxide, the resulting solution gave immediately a U.V.spectrum similar t o that of the starting trimethyl(pheny1-thio)tin. No reverse reaction was found by mixing tri-methyltin chloride and benzenethiol.N.m.r. Meas~rernents.-~~C Spectra were recorded a t25-15 MHz using a JEOL PS-100 spectrometer. All 13Cspectra were obtained for 60% (w/w) solutions of theorganometallic compounds, the solvent being used asinternal standard. lH N.m.r. spectra in partially deuteri-ated dioxan (66.6% isotopic purity) in the presence of1.33% (v/v) water were obtained a t 100 MHz using a JEOLPS- 100 spectrometer. In these conditions hydrolysis wasslow and did not interfere with the spectroscopic measure-ments.[3/1597 Received, 30th July, 19731l8 G. Calingaert, F. J. Dykstra, and H. Shapiro, J . Amer. Chewl9 D. Blake, G. E. Coates, and J. M. Tate, J. Chem. Soc., 1961,2o R. Sasin and G. S. Sasin, J. Org. Chem., 1955, 20, 770.21 R. Sasin and G. S. Sasin, J. Oyg. Chem., 1955, 20, 387.22 J. E. Griffiths and M. Onyszchuk, Canad. J . Chem., 1961,SOG., 1945, 67, 190.618.39, 339

ISSN:1472-779X    

DOI:10.1039/P29740000853

出版商:RSC    

年代:1974

数据来源: RSC