C. GARDNER SWAIN, LACRA E. KAISERA N D TERENCE E. C. KNEE
4098
with ethyl ether, the ether evaporated, and 50 ml. of 50'% aqueous ethanol added to t h e residues in glass-stoppered erlenrneyer flasks. Addition of a n aqueous ethanolic solution of sodium 3,5-dinitrobenzoate to one flask yielded benzyl 3,5-dinitrobenzoate, m.p. 108-1 10.8" ( 111.1-1 12.O" after recrystallization from alcohol). Volhard titration of t h e other 21 days later showed 100.3y0 chloride ion. This is evidence for benzyl chloride. From a third aliquot the dibenzpl sulfide was precipitated out of the ether layer by extraction with a n alcoholic solution of mercuric chloride. The dibenzyl sulfide-mercuric chloride complex was filtered and dried in a desiccator, yield 88.4%, m.p. 134.6-13E1.5O.l~ Kinetic Measurements.-All rates were measured by the usual sealed ampoule technique with about 6 ml. of solution in soft glass tubes. Zero time was taken after the tubes had been in the thermostat for about five minutes, a t which time the initial concentration was determined by titration. For reaction of tribenzylsulfonium chloride, tubes were (17) Reported as 136' b y L. Bermejo and J. J. Herrera, Anales lis. qzrinz., 32, 682 (1934); C. A , , 28, 6709 (1934).
IOC. espnfi.
[CONTRIUUTIOX FROM TIiE
Vol. SO
quenched in ice-water and a n accurate 5-ml. aliquot of the contents added t o a separatory funnel containing 10 ml. of ethyl ether. About 15 ml. of distilled water was added and the sample shaken thoroughly to remove benzyl chloride. When the emulsifying properties of the sulfonium salts prevented a clear separation of layers, a 570 solution of sodium acetate was used instead of distilled water t o overcome this difficulty. The aqueous layer and two extracts of the ether layer were combined and titrated for chloride ion by the Volhard method (with nitrobenzene) with a blue (daylight) lamp bulb t o facilitate recognition of the end-point. Titrations of samples of the reaction mixture for acid at the start of the reaction and at the end were made in order t o check on any hydrolysis of benzyl chloride. S o hydrolysis wds detectable during the period of time of decomposition of the sulfonium salt. In the runs with perchlorate salts and no added chloride ion, titrations for acid were made to follow the reaction. Typical kinetic runs are given in Tables I1 and 111. CAMBRIDGE,
hIASSACHUSE1TS
DEPARTMENT O F CHEMISTRY, hTASSBCIIUSETI'S INSTITUTU
O F 'I'ECHXOLOGY]
Mechanisms of Decomposition of Neutral Sulfonium Salts in Solution. 11. t-Butyldimethylsulfonium BY C. GARDNER SWAIN, LAURA E. KAISERAND TERENCE E. C. K N E E RECEIVED MARCH3, 1958 'The hydrolysis of t-butyldi~netliylsulfoiiiuiiichloride is iiot significantly faster than t h a t of the perchlorate in 90% acetone-lO% water solution at 50". Therefore anions do not appear to be involved in the mechanism. Of 15 solvents and solvent-mixtures studied, 90% acetone-lO% water gave the largest first-order rate constant for solvolysis and 100% water the smallest, but the range was less than a factor of four. The rate constant for solvolysis in 100% acetic acid is 011ly slightll- (28y0) larger than that for hydrolysis in 100% water. This relatively lon rate in 100% acetic acid suggests that the rate-determining step is reaction of acetic acid with trimeth\.lcarbonium ion.
The decomposition of neutral t-butyldiniethylsulfoniuin salts in goyo acetone-10y0 water solution is a solvolysis not involving a nucleophilic displacement on carbon by the anion, in contrast to the decompositions of trimethyl- and tribenzyl-sulfonium salts described in the previous paper.2 Measurem e n t ~a ~t 50" and 0.01 JI ionic strength gave relative first-order rate constants of 1.00 for fbutyldimethylsulfonium perchlorate E ' S . 1.17 for the chloride, 1.19 for thc bromide and 1.13 for the iodide. This less than ?O(% variation is in contrast to the greater than thousand-fold difference between perchlorate and chloride salts of trimethyland tribenzyl-sulfonium ioris.?,5 In agreement (1) Supported b y t h e Oflice of Yaval Research. Reproduction permitted for a n y pupose of t h e United States Government. ( 2 ) C j . Paper I , C. C.Swain and L.E. Kaiser, THISJ O U R N A L , 80, 1089 (1958). (3) For further details. CJ. T . 13. C . Knee, 1'h.D T h c s i s in Organic Cheniistty, hI.I.T., September, 19Zlj. (41 L E, Kaiser, Ph.D. Thesis i n Organic Chemistry, 11, I. 'l'., l:cbniary, 18,jl. (.i) IVc :+re intlcl,tcd to Air, l ) : ~ v i < lA . K u h n lor t h e ohservalinii t h a t t h e \ arixtion is even l e i , w h e n a Ixrger excesb of inert salt is u.ied. 'I'hus t h e r.rte o f solvolysis of 0.010 .\I sulfonium perchlorate a t 50' 111 9(ii'( iicetilne kontaining 0.11 J f sodium perchlurate was t h e samc .,\ x i t h 0.10 .\I sodium perchlorate and 0.010 .21 sodium chloride .Use t h e rate of 0.00.5 . I I sulfonium perchlorate a t 50' !\ithiti I " ; :it 100r; acetic acid containing 0.012 .\I sodium acetate a n d 0.11 .V sodium perchlorate was t h e same as with 0.012 M sodium acetate, 0.10 M sodium perchlorate and 0.010 M sodium chloride within 1%. T h e r a t e in acetic acid was unaffected b y t h e concentration of sodium xcetatr ( 0 010 .If sodiiim perchlorate, 0.010-0.021 .\I sodium acetate, 1 h r r ~ d l t ) Increa.;ini: the ionic 5trength from 0.01 to 0.12 If oi111 v x l i i t i i i i ~ ? r r h l t a r a t rd ~ i ~ r v + vl ldi r rute i i i t h e w two w i v ? t t l . ~ ~ r \ iV Li \ ~ , t ~ , ~ , i i t i \ i t i i itrolt:ii>lv 111.tt t l t r 17": \ , i i i : i t i i i i l \ w t I i
