JOURNAL OF THE CHEMICAL SOCIETY PERKIN TRANSACTIONS II Physical Organic Chemistry Stability of a-Sulphur- and a-Oxygen-substituted Carboniurn Ions By Giorgio Modena and Gianfranco Scorrano,* Centro CNR Meccanismi di Reazioni Organiche, lstituto di Chimica Organica, via Marzolo 1, 35100 Padova, Italy Paolo Venturello, lstituto di Chimica Organica, Universith, Torino, Italy The rates of hydrolysis of dimethoxymethane and methoxymethylthiomethane have been measured in aqueous sulphuric acid at 25 "C. The rate constant ratios for the oxygen and sulphur compounds (ko/$) vary between 0.12 and 0.08 depending on the acid concentration. The solvolysis of chloromethyl methyl ether and sulphide gives, in dioxan-water, a k,/k, ratio of ca. 115. Comparison with other published data on reactions leading to or-sulphur- or a-oxygen-substituted carbonium ions shows that ko/k, ratios may range between ca.0.1 and ca. 440. This is explained by implying different degrees of carbonium ion character in the transition states and the greater ability of sulphur to stabilize a fully developed carbonium ion as opposed to the greater overall x stabilization for the oxygen derivative in the ground state. CONTRASTINGreports are found in the literature on the relative ability of CH,O or CH,S groups to stabilize an adjacent positive charge. A frequently cited work on the hydrolysis of chloromethyl methyl ether and sulphide indicates greater x donating ability for oxygen than for sulphur. Examination of the U.V.spectra has led,, however, to the conclusion that sulphur is a better donor than oxygen, at least in the photoexcited state. Stabil-ization energies of substituted methyl cations generated by electron impact in the gas phase show that the methylthiomethyl cation is more stable than the meth- oxymethyl cation by ca. 5 kcal mol-l. In agreement with this finding, Field and Weeks4 have found that methylthiomethyl cation is formed by chemical ioniz- ation much more readily than methoxymethyl cation. Finally, in recent ab initio calculations on CH,OH+ and CH,SH+ cations, more positive charge was found on sulphur than on oxygen, as well as a greater x overlap population in the CH2SH+than in the CH,OH+ cation. This difference in behaviour between the gas phase and solution is not unique and it is obvious to attribute it to solvation effects.4 However, we have recently shown that the relative basicities (ability to share a lone pair with a proton) of oxygen and sulphur bases are strongly solvent dependent even in aqueous sulphuric acid.6 This indicates that, perhaps, a further variable should be taken into consideration when discussing the relative electron donor abilities of sulphur and oxygen, i.e. the composition of the solvent in which the experiments are run.To investigate this point we have studied the hydroly- sis of dimethoxymethane and methoxymethylthio-methane in aqueous sulphuric acid and the solvolysis of chloromethyl methyl sulphide in dioxan-water. The results are discussed in this paper together with other pertinent data taken from the literature.RESULTS Hydrolysis of Dimethoxymethane and Methoxymethylthio- methane.-The rates of hydrolysis of dimethoxymethane and methoxymethylthiomethane were measured by an n.m.r. technique in aqueous 0.5-4.5~-sulphuric acid. Both reactions follow pseudo-first-order kinetics at each acid concentration and the rates increase steadily with acidity. This has been already observed by other authors when studying the hydrolysis of dimethoxymethane.' The pertinent data are collected in Table 1. TABLE1 Rates of dimethoxymethane and methoxymethylthio- methane hydrolysis a Dimethoxymethane Methoxymethylthiome thane h r r %+/M 1OPk$/s; CH+h 104k$p-'A 0.85 0.336 0.50 0.534 1.20 0.649 0.85 1.38 1.61 1.32 1.20 2.64 2.49 4.09 1.69 6.88 2.87 6.95 2.49 21.4 4.45 34.2 2.87 44.2 a In aqueous sulphuric acid at 25 "C; monitored by n.m.r.The acid dependence of the pseudo-first-order rate co-efficient (KG) was analysed in terms of the Bunnett-Olsen 8 equation (1). As we discussed re~ently,~ equation (1) must be used when dealing with bases which are too weak to log '$h + Ho = +x(Ho + log CH+) f log(Ko/KSH+) (1) be substantially protonated in the range of acidity studied. The rapid hydrolysis of dimethoxymethane and methoxy- methylthiomethane prevents direct evaluation of their pKBH+ values. However, from published data on the protonation of sulphides lo and ethers 6y11 it may be safely assumed that : (i) both substrates are protonated on oxygen (ethers are stronger bases than sulphides in aqueous sulphuric acid solutions below 60-65% H2S04); 6, lo911 (ii) neither compound should be protonated to any signifi- cant extent even in the most concentrated acid soIution studied, 2.e.4.45~-H,sO,. In this solution the protonation ratio (cBH+/cu) for dimethyl ether, which should be more basic than either dimethoxymethane or methoxymethyl-thiomethane, may be estimated as 2.6 x The slope parameter 4~ obtained from equation (1) is related9 to the changes in solvation which the reacting system experiences in going from the initial to the transition state. We have discussed this point elsewhere by expressing the linear free energy relationship leading to equation (l),and the similar expressions for moderately and strongly basic substrates, in terms of activity coefficients [equation (2)]. logfH+ -log (fx/fS) = (l -4:)[l0gfH+ -log (fBH+/fB)l (2) It suffices to say here that a negative ~$zvalue indicates that the activity coefficient ratio f:/fs increases with acidity less than the ratio fn~+/fn,and therefore that the solvation requirements of the transition state are not very high, as happens in those reactions where the positive charge is fairly well delocalized in the transition state.This is expected for reactions leading to carbonium i0ns.991~ The $I values obtained for hydrolysis of dimethoxymethane and methoxymethylthiomethane are -0.13 and -0.59 respec- tively.R CGCH CH=CH2 CH2Cl CH,0C(0)CH3CH,0CH3 J.C.S. Perkin I1 metric technique. To evaluate the slower rates for chloro- methyl methyl sulphide we have used a potentiometric technique, checked by evaluating the rate of solvolysis for the chloro-ether in 95% dioxan-water. Our results are within the experimental error reported by Jones and Th0rnt0n.l~ The data for the solvolysis of the chloro-ether and sulphide in dioxan-water are collected in Table 2. TABLE 2 Rates of solvolysis of chloromethyl methyl ether and sulphide in dioxan-water (v/v)a Dioxan Ether Sulphide(%I Yb 103k++-1 103k4lsi 80 -0.83 300.6 2.63 85 -1.43 149.0 1.31 90 -2.03 35.1 0.36 95 -3.05 3.4 & 0.29 0.030 3.7 a At 25".From E. Grunwald and S. Winstein, J. Anzev. Chem. SOC.,1948, 70, 846. From ref. 13 unless otherwise indicated. This work. The most significant data from the literature relevant to the problems discussed in this paper are those for the hydrolysis of ethynyl methyl ethers l4 and sulphides,15 of methyl vinyl ethers l6 and sulphides,16 and of methoxy- methyl acetate *7 and methylthiomethyl acetate.ls They are summarized, together with our results, in Table 3 which also presents the reactivity ratios between the oxygen and the sulphur compounds (ko/ks). The experimental pseudo-first-order constants (Kd) for the acid-catalysed solvolysis of methoxymethyl acetate, diniethoxymethane, and their sulphur analogues, and hence the reactivity ratios reported in Table 3, depend [see equ- ation (l)]on the solvent composition (+I and H,) and the pKRH+ of the substrates.As mentioned above, it is not possible to evaluate the protonation equilibria of these compounds directly because of their fast hydrolysis. We are, however, only interested in the relative pKBH+ values and therefore on the relative effect on the basicity of the TABLE3 Hydrolysis of several CH,XR derivatives Solvent Acidic H20 Acidic H20 80% dioxan-20% water 95% dioxan-5% water Aq. O.28M-H2SO4 Aq. ~.~~M-I-H,SO, Aq. O.85M-HRSO4 Aq. 2.87M-HZSO4 x=o x=s k4ls-l k4ls-l ko/ks Reference 77 f0.8" 7.14 x 10-l 1.78 x lo-*" 2.10 x 10-2 " 437 34 14, 15 16 3.0 x 10-1 2.63 x 10-3 114 13, this work 3.4 x 10-3 3.0 x 10-4 113 7.43 x 10-4 2.04 x 3.36 x 10-5 4.50 x 10-4 1.14 x lop21.38 x 10-4 1.65 1.79 0.12 17, 18 This work 6.95 x 10-4 4.42 x 10-3 0.08 Catalytic rate constants for the hydronium ion, expressed in 1 mol-l s-l, obtained by dividing the experimental rate constant by the activity of the proton.The rate constants for the dimethoxymethane solvolysis have been divided by a factor of 2, to allow for statistical correction of ~KBH+ values (see text). As mentioned before, the hydrolysis of dimethoxy-methane has been studied by M.cIntyre and Long 7 who measured the rates in aqueous 1.2-1.9~-H,SO, solutions. Our rate data in this range are in fair agreement (ca. 10% slower) with the published value^.^ Solvolysis of Chlorovnethyl Methyl Ether and Sulphide.-The rates of solvolysis of chloromethyl methyl ether have been carefully determined by Jones and Thornton l3 for a series of solvents and solvent mixtures by using a conducti- groups CH,OCH, and CH,SCH,.Both the steric (E,) lg and the polar (q)2o parameters for the two groups are very similar: E, -1.43 and -1.58; 01 +0.07 and +0.04 for the oxygen- and sulphur-containing group, respectively. This, and the fact that only modest effects on pKB~+ upon alkyl group substitution have been reported for alkyl acetates 21 and for alkyl methyl ethers,6 leads us to believe that, as a first approximation, we may take as equal the overall sub- stituent effect of the two groups on the protonation equili- brium constant.No correction has been made for the sol- vent effect and the ratios Ko/hs are reported at two acid concentrations. Dimethoxymethane, but not methoxymethylthio-methane, has two equivalent sites of protonation and, therefore, the experimental rate constants have been divided by a statistical factor of 2 in evaluating the Ko/Ks ratios. DISCUSSION Two reaction pathways (a) and (b) are available for the acid-catalysed hydrolysis of methoxymethylthio-methane. Although paths (a) and (b) give the same Having thus established the site of fragmentation, we may now turn to considering in more detail the reaction mechanism for hydrolysis of the methoxymethylthio- methane and dimethoxymethane. Clear evidence has been collected 29 to suggest the assignment of dimethoxy- methane hydrolysis to the A-1 type of reaction.The analysis of our results on the acid dependence of the rates of hydrolysis of dimethoxymethane and methoxy- methylthiomethane confirms that both reactions occur via rate-determining unimolecular cleavage of the conjugate acid of the substrate as shown in the Scheme. In fact, the +$ values obtained from equation (1) CH3kH,0CH3 +CH3SH + CH30CHZ+IHZH CH3SCH20CH3 H+$ CH3SCH20CH3 I H final products and it is very difficult to distinguish between them on a kinetic basis, we believe that the reaction occurs exclusively via the sulphur stabilized cation [path (b)] for the following reasons. First, taking as a model the protonation of dimethyl ether (DME) and dimethyl sulphide (DMS),ll the fraction of 0-protonated substrate at the highest acidity studied (2.87~-H,so,), although small ([BH+]/[B,,] for DME ca.1.4 x is ca. 34000-fold higher than in the S-protonated case ([BH+]/[B,,] for DMS ca. 4 x Secondly, the leaving group ability from a cationic centre is much larger for CH,OH than for CH3SH. In fact, we have measured the rates of methyl t-butyl ether fragmentation in aqueous sulphuric acid to give methanol and t-butyl ~ation.~ The reaction is too fast to be followed with conventional techniques at 8M-H2S0,, where the protonation fraction [BH+] /[Bst] is ca. 0.Z.9 By contrast, the protonation of methyl t- butyl sulphide may be studied even in solutions as concentrated as 17.9~-H,sO,, where it is completely in the acid form.10 The greater stability to cleavage of protonated sulphides, compared with the corresponding protonated ethers, has been also observed by Olah et al.for FS0,H-SbF5-S0, solutions.22 They reported that protonated methyl t-butyl sulphide is stable at -60" and very slowly cleaves to CH,SH and t-butyl cation at -15O.23 The corresponding protonated ether is rapidly cleaved at -70°.24 Finally, the proposal by Fife et al. that 1,3-oxathiolans 25a and benzaldehyde methyl S-substituted-phenyl thioacetals 25b are cleaved in acid at the C-S bond has been convincingly challenged by Pihlaja 26 and by Lamaty et Of particularaZ.2792s relevance is the finding by Lamaty et aZ.27 that 2,2- dimethyl-l,3-oxathiolan is cleaved in FS0,H-SbF,-SO, to give exclusively the (CH3)26SCH,CH20H2+ dication, i.e.that derived from C-0 bond breaking. CH30H + CH3SCH2-' (-0.13 and -0.59, for the oxygen and sulphur deriv- ative, respectively) is a clear indication9 that the CH,XCH,OCH, + H+__ CH,XCH,ifCH,IH CH,XCH,i)CH, szCH,OH + CH,XdH, (1) (2)4 SCHEME transition states of the two reactions have low solvation and hence that they must be closer to a carbonium ion [such as (Z)] than to an oxonium ion [such as (1)or (1) plus a water molecule, if one consider an A-2 type re- action]. The 4: values, more negative for methoxymethylthio- methane than for dimethoxymethane, suggest more developed carbonium ion character in the transition state for the sulphur than for the oxygen derivative.How-ever, since a part of the positive charge should be carried by the heteroatom, the difference in $$values may also reflect intrinsically different solvation requirements for the two cations. Indeed, on the basis of activity co- efficient~,~~~~~one would expect that the larger sulphur cation would have a more dispersed charge than the oxygen analogue and, hence, a smaller solvation energy. The rates of hydrolysis of methoxymethylthiomethane increase with acid concentration more rapidly than those of dimethoxymethane, making, obviously, the reactivity ratios different in the various acid solutions. However, as the ratio KO/& is less than unity over the range of acid concentration, these reactions represent the first studied example in the condensed phase in which sulphur appears to behave as a better electron donor than oxygen.We will return to this point later. Chloromethyl methyl ether has been shown' to solvolyse by an SN1-like mechanism. In particular, the J.C.S. Perkin I1 m value obtained by the Winstein m-Y correlation 32 in a series of solvents and solvent mixtures was within experimental error of that defined for t-butyl chloride.’ From the solvolysis rate constants of chloromethyl methyl sulphide in dioxan-water we have evaluated m as 0.90 & 0.08, compared with the value of 0.90 0.07 found for the ether 7 in the same solvent mixture. This is a good indication of the similarity between the two reaction mechanisms.It must be pointed out that in this case the ratio ko/ks is ca. 110 over the solvent range. The reactivity ratio is much smaller than that derived from Bohme’s data.] Before considering in some detail the data reported in Table 3, we must briefly discuss the mechanism of the other reactions leading to oxygen- or sulphur-stabilized carbonium ions. Drenth and his co-workers 14315 have clearly proved that the hydrolyses of ethynyl methyl ether and sulphide are general acid catalysed processes, with protonation of the ethynyl group as the rate-determining step. McClelland l6 has shown that the hydrolysis of vinyl sulphides proceed via a mechanism analogous to that for vinyl ethers,33 with slow proton transfer to the carbon-carbon double bond.The acid- catalysed hydrolyses of methoxymethyl acetate and methylthiomethyl l8 acetate were shown to occur by A-1 type mechanisms. The +$ values which may be evaluated from published data 17918 are -0.20 and -0.15 for the oxygen and sulphur compound, respectively, in agreement with the hypothesis of a delocalized positive charge in the transition state. Relative Stabilities of Methoxymethyl and Methyl-thiomethyl Cations.-The relative st abilities of carbo-cations R+ have been frequently estimated from rate constants for solvolysis under identical conditions. To use the same approach in the case of CH,OCH,+ and CH3SCH,+ cations it is necessary (i) that all the reactions studied reach the transition state at the same point along the reaction co-ordinate and that the transition state is close in energy to the cation; (ii) that the solvation requirements in going from the reactants to the transition states are similar in all cases; and (iii) that the reactants have similar energies.Inspection of the data reported in Table 3 suggests that these three condi- tions are not very likely to be obeyed. As a matter of fact the reactivity ratio ko/ks ranges from ca. 400 to ca. 0.1. Clearly, caution must be used in interpreting these rate data in terms of sulphur or oxygen stabilizing ability for cations. It was suggested 3* that, qualitatively, the stabilizing effect on adjacent cations by heteroatoms should depend on the ionization potential (which predicts sulphur as a better donor than oxygen) and on the strength of the x bond formed in ‘ resonance structures ’ (which favour oxygen over sulphur for carbocations) .Recent cal-culations offer a more quantitative interpretation of these ideas5 The x stabilization energy (SE) due to the con- jugate interaction of a heteroatom X with an acceptor is given by equation (3) where HaB is the interaction matrix element between +A ($, lone pair of the hetero- atom) and a,bB (the vacant 2pn orbital of the acceptor fragment), and AE is the energy separation between (3) #A and #B. It has been found 5 that the matrix element (HA,) tends to favour lower group elements (0> S; F > C1) whereas the energy term (AE) always favours the higher group element (S > 0; C1 > F) as the ioniz- ation potential of the heavier element is always lower than that of the corresponding lighter one.It follows that when the LUMO of the acceptor is high in energy, as for example for the antibonding x orbital of the ethylene fragment in CH,=CHX, the matrix element is dominant and the overall x stabilization is greater for oxygen than for sulphur. On the other hand when the LUMO is of low energy, as for the empty p orbital of the carbonium ion in +CH2X, the energy term is dominant and the S-stabilized cation is more stable than the 0-stabilized one. It is therefore possible to rationalize the data in Table 3 by considering the differences in carbonium ion character of the transition states.Let us first consider reactions involving a proton, i.e. the hydration of ethynes and ethylenes and the hydrolysis of acetals and acylals. The hydration reactions occur through the rate-determining attack of the proton (see above). Hence, it is expected that the transition states resemble the reactant more than the intermediate carbonium ion or vinyl cation. In this case, as suggested by calculations and by n.m.r.S5*36 and dipole moment 36 data, oxygen must be a better donor than sulphur and hence the oxy- gen derivative must be more reactive than the sulphur one, as found. The acid catalysed hydrolysis of acylals (CH3XCH2- OCOCH,) and acetals (CH3XCH,0CH3) has been proved to occur via an A-1 type mechanism (see abo~e).~~~~~ On the basis of basicity measurements on methanol (PKBH+-2.05 +e 0.87) l1 and acetic acid (pKBH+ CII.-3.8, estimated from data on methyl acetate37), as well as from the acid catalysed fragmentation rates, faster for t-butyl acetate than for methyl t-butyl ether in aqueous sulphuric acid,g we expect methanol to be a worse leaving group than acetic acid. Hence, the transition state for the hydrolysis of acylals should be reached earlier along the reaction co-ordinate than for the hydrolysis of acetals, and it should have less car- bonium ion character. This is also reflected by the 4; values. As discussed elsewhere? a reaction going from an oxonium to a carbonium ion should have negative ($1 -+e) values,* because of the lower solvation requirements of the latter ion.The greater the car-bonium ion character of the transition state, the more * The (C$t -C$e) value is the slope of log -log (GBH+/C,t) uevsus H, + log CH+ plots.g This plot correlates the reaction rates with acidity for moderately basic substrates, i.e. for com-pounds which are substantially protonated in the range of acidity studied. The -C$e) is related to the difference between the activity coefficients of the protonated substrate and the transition state. The slope t$e is related to the solvent effect on the protonation equilibrium and may be evaluated as the slope of the plot of log (cBH+/cB) + H, against H, + log ~~+.~e@*~~9~~ 1979 negative are the ($1 -&) values found.g In the present case ($$-&) values may be estimated as ca.-0.8 for the acylals and -0.95 to -1.4 for the acetals on the basis of the experimental 41 values and assuming that the +e values are similar to those of unsubstituted methyl acetate and dimethyl ether (+0.6 and +0.82, respectively).11937 It follows from the larger carbonium ion character for acetal hydrolysis that the reactivity ratio ko/ks should be smaller than in the hydrolysis of acylals, as found (ca. 0.1 versus ca. 1.7). We must finally consider the solvolysis of chloro-methyl methyl ether and sulphide. These reactions are not strictly comparable with those discussed above since they have been studied in different solvents and do not require acid catalysis. Since the solvolysis of the chloro derivatives involves charge separation, we expect that the transition state is reached through a much steeper reaction path than in the solvolysis of the protonated acetals and acylals.Even if the chloro derivatives are more stable than the protonated oxy- compounds, they will reach the transition state earlier and this will be doser to the reactants than to the carbonium ion. This implies, according to the above discussion, that the oxygen derivative reacts faster than the sulphur and, in fact, the ratio ko/ks is 110. It is evident from the above data and discussion that is is not possible simply to take reactivity ratios as representing the relative ability of heteroatoms in stabilizing an adjacent positive charge.In the parti- cular case of sulphur- and oxygen-substituted carbonium ions it is concluded that sulphur has a greater ability to disperse the positive charge of the fully formed carbonium ion, although the oxygen derivative is often more reactive since the transition state is reactant-like, so that interactions are more important with the oxygen than with the sulphur heteroatom. EXPERIMENTAL N.ni.r. spectra were recorded on a Bruker HX-90 spectro- meter. B.p.s are uncorrected. Materials.-Reagents grade dioxan was purified according to the procedure given by Fieser.,* Sulphuric acid solu- tions were made by dilution of concentrated reagent grade acid with distilled water. The solutions were standardized by titration with 1M-NaOH.Commercial dimethoxy-methane, chloromethyl methyl ether, and chloromethyl methyl sulphide were purified by distillation. Methoxynzethylthiovnethune. Methanol (5 g) was added to a solution of chloromethyl methyl sulphide (5g) in pyridine (12.5 ml) and the mixture was stirred at 0 "C for 3 h. The title compound was recovered pure by dilution of the mixture with a large excess of saturated NaCl aqueous solution, b.p. 98-99 "C; F(CCl,), 2.05 (SCH,), 3.28 (OCH,), and 4.51 (CH,).39 Hydrolysis of Methoxynzethy1thiovnethane.-The title re-action was followed by monitoring the changes in the n.m.r. spectra of a sulphuric acid solution of methoxymethylthio- methane. The initial spectrum in 2.45~-H,sO,shows three signals at 466 (CH,), 350 (CH,O), and 239 Hz (CH,S) from tetramethylsilane as external standard.Three different signals appear with time: at 350 (attributed to CH,OH by comparison with an authentic sample) and at 470 and 243 Hz, attributed respectively to the methylene and methyl protons of the a-hydroxy-sulphide CH,SCH,OH, the first product of methoxymethylthiomethane hydrolysis. It is known that carbonyl compounds and thiols are in equili- brium with hydroxy sulphides in aqueous media.,* When the conversion to the a-hydroxy-sulphide is almost complete three signals appear at 482 [CH,(OH),], 380 [CH,(SCH,),], and 240 Hz [CH,(SCH,),], attributed by comparison with authentic samples. The overall reaction may be therefore written as (4). 2 CH,OCH,SCH, + 2 H,O -+ CH,SCH,SCH, + 2 CH,OH + CH,(OH), (4) Reaction of formaldehyde with methanethiol in 2.46~-H,S04 gives bismethylthiomethane (3),as expected. Hydrolysis of Diwethoxymethane.-The n.m.r.spectrum of dimethoxymethane in 2.46hf-H2S0,shows two signals at 463 (CH,) and 351 Hz (OCH,) from tetramethylsilane as external standard. With time another signal appears at 482 Hz, attributed to hydrated formaldehyde (see above). The signal corresponding to the other product, methanol, should appear at 350 Hz but it is not distinguishable from the OCH, signal of dimethoxymethane. Kinetic Procedure.-The hydrolysis reactions of dime-thoxymethane and methoxymethylthiomethane were monitored by following the disappearance of the methylene n.m.r. signal, using a slightly different procedure according to the acidity of the solution. In more dilute solutions (<2.5~-H,So,),the methylene proton signal appears near the water signal, and therefore an extraction technique was followed. Substrate (ca.30 p1) was added to thermostatted (25 "C) aqueous H,SO, (10 ml) of the appropriate concen- tration. Portions (1 ml) were withdrawn at intervals and extracted with 1 ml of a solution of CH,Cl, (25 pl) in CCl, (30 ml). The n.m.r. spectra of the carbon tetrachloride solutions were then taken and the ratio between the methy- lene signal of dimethoxymethane or methoxymethylthio- methane and that of the CH,Cl, used as standard (hcH,/h,t) was evaluated by comparing the peaks heights. In more concentrated acid solutions ( >2.~M-H,SO,), dimethoxymethane or methoxymethylthiomethane (ca.3 pl) were added to aqueous H,SO, (1 rnl) of the appropriate concentration containing dimethyl sulphoxide (ca. 1 pl) or dioxan (ca. 0.3 pl). DMSO and dioxan do not react in these acid solutions and give a sharp singlet in the n.m.r. spectrum. The spectra of the acid solutions were taken at intervals. The ratios between the methylene signal of dimethoxymethane or methoxymethylthiomethane and that of the internal standard (hcH,/h,,) were evaluated by comparing the peaks heights. The pseudo-first-order rate constants (kg) were evaluated from the slope of plots of log (hcH2/hsb)veysus time. The hydrolysis of chloromethyl methyl ether and sul- phide were followed by a potentiometric technique.In a three-necked thermostatted flask equipped with glass and calomel electrodes, substrate (50 pl) was added to dioxan- water (25 ml), made by mixing the appropriate amount of thermostatted solvents. The increase in acidity of the solu- tion was monitored by using a Radiometer 26 pH meter. Calibration curves, showing the linearity of the electrode response with the acid concentration, were made for each solvent mixture by using different amounts of HCl. The pseudo-first-order rate constants (K4) were evaluated from the slopes of plots of [(mV), -(rnv),]versus time. [7/1843 Received, 20th October, 19771 REFERENCES H. Bohme, Ber., 1941, 74, 248; H. Bohme, H. Fischer, and R. Fank, Annulen, 1949, 563, 54.A. Mangini, Gazzetta, 1958, 88, 1063; R. C. Passerini, in ‘ Organic Sulphur Compounds,’ ed. N. Kharasch, Pergamon,London 1961, voI. 1, p. 57. R. W. Taft, R. H. Martin, and F. W. Lampe, J. Amer. Chem. SOC.,1965, 87, 2490. F. H. Field and D. P. Weeks, J.Amer. Chem. Soc., 1970, 92, 6521; F. H. Field, in ‘ Ion-Molecule Reactions,’ ed. J. I,. Frank- lin, Butterworths, London, 1972. F. Bernardi, I. G. Csizmadia, H. B. Schlegel, and S. Wolfe, Canad. J, Chem., 1975, 53, 1144; F. Bernardi, A. Mangini, N. D. Epiotis, J. K.Larson, and S. Shaik, J.Amer. Chem. SOC.,1977,99, 7465. P. Bonvicini, A. Levi, V. Lucchini, G. Modena, and G. Scorrano, J. Amer. Chem. SOC.,1973, 95, 5960. D. McIntyre and F. A. Long, J. Amer. Chem. SOC.,1954, 76, 3240.J. F. Bunnett and F. P. Olsen, Canad. J. Chem., 1966, 44, 1899. n V. Lucchini, G. Modena, G. Scorrano, and U. Tonellato, J. Amer. Chem. Sac., 1977, 99, 3387. lo P. Bonvicini, A. Levi, V. Lucchini, and G. Scorrano, J.C.S. Perkin II,1972, 2267. l1 G. Perdoncin and G. Scorrano, J. Amer. Chem. SOC.,1977, 99, 6983. l2 G. Modena, F. Rivetti, G. Scorrano, and IT.Tonellato, J. Amer. Chem. Soc., 1977, 99, 3392. l3 T. C. Jones and E. R. Thornton, J. Amer. Chem. SOC.,1967,89, 4863. l4 E. J. Stamhuis and W. Drenth, Rec. Tvav. chim., 1961, 80, 797. l5 H. Hogcveen and W. Drenth, Rec. Trav.chim., 1963, 82, 375. l6 R. A. McClelland, Canad. J. Chem., 1977, 55, 548. l7 R. A. McClelland, Canad. J. Chem., 1975, 53, 2763. l8 R.A. McClelland, Canad. J. Chem., 1975, 53, 2772. S. H. Unger and C. Hansch, Progv. Phys. Org. Chern., 1976, 12, 91. J.C.S. Perkin I1 20 The GI values have been computed from data and equations reported in R. W. Taft, E. Price, I. R. Fox, I. C. Lewis, K. K. Andersen, and G. T. Davis, J.Amer. Chem. SOC.,1963, 85, 709. 21 D. G. Lee and M. H. Sadar, J. Amer. Chem. Suc., 1974, 96, 2862. 22 G. A. Olah, A. M. White, and D. H. O’Rrien, Chem. Rev., 1970, 70, 561. 23 G. A. Olah, D. H. O’Brien, and C. U. Pittman, jun., J.Amer. Chem. SOC.,1967, 89, 2996. 24 G. A. Olah and D. H. O’Brien, J.Amer. Ckem. SOC.,1967,89, 1725. 25 (a)T. H. Fife and L. K. Jao, J. Amer. Chem. SOC.,1969, 91, 4217; (b) T. H. Fife and E. Anderson, ibid., 1970, 92, 5464.26 K. Pihlaja, J. Amer. Chem. SOC.,1972, 94, 3330. 27 F. Guinot, G. Lamaty, and H. Miinsch, Bull. SOC.chim. France, 1971, 541. 28 F. Guinot and G. Lamaty, Tetrahedron Letters, 1972, 2569. 29 E. H. Cordes and H. G. Bull, Chem. Rev., 1974, 74, 581. 3O E. M. Arnett and G. Scorrano, Adv. Phys. Org. Chem., 1976, 13, 83. 31 K. Yates and R. A. McClelland, Prugr. Phys. Org. Chem., 1974, 11, 323. 32 J. E. Leffler and E. Grunwald, ‘ Rates and Equilibria of Organic Reactions,’ Wiley, New York, 1963. 33 A. J. Kresge, H. J. Chen, and Y. Chiang. J. Amer. Chem. SOC.,1977,99, 802 and previous papers of this series. a4 C. K. Ingold, ‘ Structure and Mechanism in Organic Chem- istry,’ Cornell University PresF, London, 1969. 36 W. Drenth and A. Loewenstein, Rec. Trav. chim., 1962, 81, 635; H. S. Gutowsky, M. Karplus, and D. M. Grant, J. Chem. Ph-ys.,1959,31, 1278; J. Feeney, A. Ledwith, andC. H. Sutcliffe, J. Chem. SOC., 1962, 2021; R. T. Hobgood, jun., G. S. Reddy,and J. H. Goldstein, J. Phys. Chem., 1963, 67, 110. 36 W. Drenth, G. L. Hekkert, and B. G. Zwanenburg, Rec. Trav. chim., 1960, 79, 1056. 37 G. Scorrano, unpublished results. 38 L. F. Fieser, ‘ Experiments in Organic Chemistry,’ D. C. Heath, Boston, 1957, p. 284. 3* K. Torssell, Acta Chem. Scand., 1967, 21, 1. *O I. B. Johns and R. M. Hixon, J. Amer. Chem. SOC., 1934,56, 1333; G. E. Lienhard and W. P. Jencks, ibid., 1966, 88, 3982; R. G. Kallen and W. P. Jencks, J. Biol. Chem., 1966, 241, 5851; I,. Field and B. J. Sweetman, J. Org. Chem., 1969, 34, 1799; L. Field, B. J. Sweetman, and M. Bellas, J. Medicin. Chern., 1969, 12, 624.