Nov. 5 , 1961
COMMUNICATIONS TO THE EDITOR
gated, that i t is probably the first system, investigated kinetically in which non-specifically-solvated primary carbonium ions have been formed in a solvolysis reaction, and that relative to the ground states the central carbon atoms in the transition states bear very large positive charges. Whereas the maximum differences in free energies of activation of solvolytic reactions in acetic acid previously reported for t-butyl and methyl compounds is approximately 8 kcal./mole, the difference for the solvolysis of t-butyl and methylmercuric salts is approximately 17.5 kcal./mole. The effect of solvent on the reaction is complex. This is to be expected since some of the more ionizing solvents coordinate with mercury t o a large extent. It has been found that with solvents of about the same ionizing power, the reaction occurs more slowly in the more nucleophilic solvent. The results given in Table I probably indicate that in these solvents the interaction between solvent and mercury is the most important effect, and this effect slows the reaction.
4479
atoms are highly charged and that this charge is not specifically solvated. It is to be expected that solvolysis of alkylmercuric salts will be very sensitive to structural changes and that the carbonium ion intermediates will not be highly selective in their reactions. The relative rate sequence per alkyl group on carbon is striking: kt-Bu/ki-Pr = 1 0 5 a 2 ; k i - P r / k E t = 104.52;k E t / k M e = 103.07. For solvolysis of arylsulfonates in acetic acid (75') the available relative rates3 are: k i - P r / k E t = lo'.' and k E t / k M e = 10-0.36. By projecting relative rates,4i t has been suggested that for limiting solvolysis in formic acid, kl-Bu/ lo6. This value is not achieved in acetic ki.pr acid in the present system. The change per alkyl group on carbon might be comparable. For alkylmercuric salts the ratio in formic acid is ki-Pr/kEt = 105J. By comparison, for solvolysis of arylsulfonates3 in formic acid the ratio is kj-Pr/kEt = 102.3. Thus, solvent-on-carbon participation is exceedingly low for the production of primary (except for perhaps methyl) and secondary carbonium ions from organomercuric ions.
TABLE I (3) S. Winstein and H. Marshall, J. A m . Chcm. Soc., 7 4 , 1120 PARAMETERS FOR THE REACTION OF CYCLOHEXYLMERCURIC (1952). PERCHLORATE IN VARIOUS SOLVENTS AT 25" (4) A. Streitwieser, Chcm. Revs., 66, 644 (1956). AF*, DEPARTMENT O F CHEMISTRY FREDERICK R. JENSEN AH*, kcal./ GNIVERSITY OF CALIFORNIA Solvent kcal./mole A S * , e.u. mole krel BERKELEY 4, CALIFORNIA ROBERTJ. OUELLETTE 22.1 1 HOAc 25.3 i 0 . 4 10.9 i z 0 . 9 RECEIVED SEPTEMBER 20, 1961 EtOH 3 0 . 3 zk . 2 24.5 0.017 19.6 zk . 6 MeOH 24.1 .030 29.6 f .4 18.4 f 1 . 2 n-BuOH 30.3 .2 1 9 . 5 zkO.6 24.5 .0165 AN ADDITIVITY RELATION FOR C'a-PROTON Dioxane 25.3 i . 3 9 . 9 i~ . 9 22.4 .60 COUPLING IN NUCLEAR MAGNETIC RESONANCE Acetone 29.1 .2 20.5 i .7 23.0 ,204 Sir : 23.4 ,098 28.5 zk .2 17.1 f .6 Hi0
N.m.r. C13-proton spin-spin coupling constants have previously been correlated with percentage s character of the carbon atomic orbital participating in the C-H bondlm2and the C-H bond length.2 For compounds of the type CH3X, an empirical equation expressing JCH in terms of the effective electionegativity of the substituent and TABLE I1 the C-X bond length has also been proposed.2 EFFECTO F ALKYL GROUPON THE SOLVOLYSIS O F ALKYL In predicting JCHvalues of compounds involving MERCURIC PERCHLORATES IN WATERA N D ACETICACIDAS various tetrahedral-type carbon atoms of the form SOLVENTS CHXYZ, the above-mentioned schemes are not AF*nio very practicable because, first, one seldom has A H * H ~ o- AF*HoA. AH*HoA~, kcal./ RHgClOa krel, ( H O A ~ detailed information concerning small deviations of R is kcal./mole mole, 25' 250 the percentage s character of the carbon atoms and, Me 0.4 0.7 1" second, bond lengths are extremely difficult to obEt 1.6 1.19 x 108 3.0 tain. The purpose of this note is to draw immedin-Butyl 3.4 1.8 1 . 7 x 103 ate attention to a simple, but precise, additivity Isopropyl ... 2.0 4 . 1 x 107 relation for predicting C13-proton couplings. The sec-Butyl 3.3 1.4 8.1 X l o 7 additivity function described here is strongly Cyclohexyl 3.2 1.35 1 . 3 X lo8 reminiscent of the well-known additivity relations t-Butyl ... ... 6.6 X 10'2 of molar refractivity and parachor. k 3.1 X sec.Extrapolated from data at It has been found in this laboratory that C13elevated temperature. proton coupling constants, for compounds of the The observations that the differences in activation type CHXYZ, can be separated into three comenergies for each compound are approximately the ponents according to the equation: JCH = px same in solvents of widely differing ionizing power by bz. In this equation J C His the C13-proton and nucleophilicity independent of the number of spin-spin coupling constant; components alkyl groups attached to the central carbon atom, and lz are contributions which are associated with and that the differences in relative rates are very substituents X, Y and Z, respectively. large and approximately proportional to the numThe coupling component for the hydrogen atom ber of alkyl groups attached t o carbon indicate (1) J. N.Shoolery, J Chcm. Phrs., Si, 1427 (1959). that in the transition states the central carbon (2) N Muller and D. E Pritchard, ibid., Si, 768, 1471 (1959). If it is assumed that the degree of coordination of a solvent with mercury is approximately independent of the alkyl group attached to mercury, information regarding the nature of the transition state can be deduced from the data in Table 11.
=i
+
+
rx,
COMMUNICATIONS TO THE EDITOR
4480
Vol. 83
TABLE I I. The accuracy of this simple additive scheme in C l 3 - P ~ 0 ~COUPLING o~ CONSTANTS OF METHYLPROTONS predicting J C H values for methylene and methine INCHSX AND CALCULATED COUPLING COMPONENTS protons is shown in Table 11. The differences Substituent
JCH
11 (CPS.) (av. value)
(CPS.)
-H 1252 41.7 -F 149: 65.6 -c1 1502 68.6" -Br 1522 68.6 -1 151*-' 67.6 -OW& 143' 59.6 -OH 1411, 144P 59.6 -S(O)CHs 1382 54.6 -NHz 1339 49.6 -"CHI 1321 48.6 --N(CH:)n 1312 47.6 -CN 1363.8 52.8 -CCL 1342 50.6 -CH21 1326 48.6 --CH 1322 48.6 -CHCla 1315 47.6 -COOH 103a, 131' 47.1 -CHpBr 1286 44.6 -CH&l 1285 44.6 -CHO 1272 43.0 -CHI 1262 42.6 -CC" 1262 42.6 -CHCHpBr 1266 42.6 -C( 0)CH, 1262, 122' 40.6 -C(CH:)r 1242, 120' 38.6 a Based on 152 cps. instead of 150 cps. for better correspondence with data of Table 11.
between calculated and experimental values are within the range of experimental error, that is, approximately f 2 cps. Upon further examination of Table I we notice that the zeta values seem to be primarily dependent upon the first atom of the substituent (the atom by which the substituent is bonded to the rest of the molecule); halogens > 0 > N > C. This variation parallels the order which these elements occupy in the Periodic Table. Besides its utility in spectral analysis, the present scheme should have a significant bearing on general theories of nuclear spin-spin coupling. The additivity and variations of zeta values probably are the result of independent, anisotropic, electron-orbital currents localized in the substituents. Consequently, we should expect the zeta values to be angular dependent. Further work along these lines is being conducted in this laboratory. DEPARTMENT O F CHEMISTRY AND CHEMICAL ENGINEERIXC STEVENS INSTITUTE OF TECHNOLOGY EDMUND R. MALIXOWSKI HOBOKEN, N. J. RECEIVED SEPTEMBER 13, 1961
SOME TRIMETHYLSILOXYTRIALKYLTIN COMPOUNDS1
Sir:
It is worth while to add some further results of our own studies to a recent note2 in the field of COMPARISON OF PREDICTED AND OBSERVED -PROTON trimethylsiloxytrialkyltin compounds. COUPLING CONSTANTS Trimethylsiloxytrialkyltin compounds, MeSSiOCompound JCH (calcd.) JCH (exp Difi. SnR3, (R = n-Pr, and n-Bu) are prepared easily G H ~ C H ~ C ~ H ~ 127 1276 0 by cohydrolysis, as previously reported,8 of a mix(cH,),CCH,OH 139 132' +7 ture of trimethylchlorosilane (3-5 mole) and CH,CH~I 152 1496 $3 trialkyltin chloride (1 mole) in benzene solution CH,CHtBr 153 1516 +2 with a slight excess of aqueous ammonia over that CaHsCH2CI 153 1526 +I necessary to neutralize the mixture. The properCaHsCH2Br 153 1536 0 ties of the products, after rigorous fractionation, (CH&CHBr 154 1516 +3 are as shown in the table. TABLE I1
BrCH2CH2Br ClCHzCHzCl CH2ClCN CH& CHlBr2 CHzCln CleCHCHCl2 CHC12CN CHCli
155 155 163 177 179 179 185 190 206
1576*6
-2
m5$6
+1
1616 1732 1 1782, 182' 1896 209'7'
$2 +4 +1 +1 $3
+I -3
TABLE I PROPERTIES OF TRIMETHYLSILOXYTRIALKYLTIN COMPOUNDS: MeiSiOSnRa R is
B.p., "C. a t mm.
n %o d Sn, %, calcd., found
n-Pr
126 20 1.4575 1.093 35.21 35.29
n-Bu
142 2 1,4582 1.059 31.30 31.28
ean be evaluated from the data of methane2 (JcII %, found" 42.75 42.70 47.51 47.51 8.97 8.89 9.57 9.81 = 125 cps.). According to the equation JcH(CH~) found" Mol. wt., calcd., found* 337.2 336 379.2 382 = 3 l ~ thus ; we find {H = 41.7 cps. a Carbon and hydrogen analyses were performed by the Zeta values for various substituents now can be Schwarzkopf Microanalytical Laboratory, Woodside, N. Y . calculated from Cla-proton coupling constants of b Cryoscopic measurement in benzene. methyl protons in C H 3 X type compounds, since These two transparent Oils have an irritating (= = JcH(CH&) - 2lH, Spectral data and the results of such calculations are presented in Table smell and are fairly stable in a laboratory atmosH9% 'I
(3) P.C. Lauterbur, ibid., 46, 217 (19571. (4) G.V. D.Tiers, J . Phys. Chcm., 64,373 (1960). (5) N. Sheppard and J. J. Turner, Proc. Roy. SOC.(London) 8 8 6 8 , 506 (1959). (6) Observad in this laboratory by Richard Magee, National Science Foundation Undergraduate Research Participant, 1961, (7) G. J. Karabatsos, J. A m . Chrm. Soc., 83, 1230 (1961).
(1) This work was presented a t The Chemical Society Symposium on Inorganic Polymers a t Nottinghsm University, England, July, 1961. (2)
H.Schmidbaur and M. Schmidt, J. Am. Chcm. Soc., 83, 2963
(1961).
(3) R.Okawara, D. G. White, K. Fujitani and H. Sato, ibid., 83. 1342 (1961).
