Dec., 1950
COMMUNICATfONS TO
Upjohn Co. and the University of Wisconsin Research Committee for support. A. L. WILDS JACKW. RALLS WILLIAMC. WILDMAN KIRTLAND E. MCCALEB RECEIVED OCTOBER 3, 1950
DEPARTMENT OF CHEMISTRY UNIVERSITY OF WISCONSIN MADISON 6, WISCONSIN
DRIVING FORCE OF THE HOMOALLYLIC REARRANGEMENT IN ACETOLYSIS OF exo-DEHYDRONORBORNYL @-BROMOBENZENESULFONATE' Sir :
THE EDITOR
5795
neighboring group participation in the rate-determining ionization) have rates nearly identical with cy~lohexyl.~I11 has a rate reduced somewhat (by ca. one power of ten) by the polar effect due to an unsaturated group (e.g., K A for phenylacetic or vinylacetic acid). These facts make it clear that in acetolysis of p-bromobenzenesulfonates the driving force in I (1:III = 7000: 1) is a t least as large and probably somewhat larger than in cholesteryl. Also, as would be expected, it is larger than in IV, the latter being measured by the f of 360 for IV :V . The essential difference between our results on sulfonates and those of Roberts and co-workers2 on halides lies in the high reactivities they report for I11 and V.
Roberts, Bennett and Armstrong2 have very recently reported the relative solvolysis rates of exo-dehydronorbonyl (I), nortricyclyl (11), and endo-dehydronorbornyl (111) halides (X = C1 or Br) in 80% ethanol as ca. 5 : 1: 1. Solvolysis of (4) (a) Winstein and Trifan, i b i d . , 71, 2933 (1940); (b) A. 11. I and I11 gives mainly the homoallylic rearrange- Schlesinger, unpublished work. ment product with the structure 11; I is not much DEPARTMENT OF CHEMISTRY S. WISWBIN OF CALIFORNIA IT. M. WALHORSKY more reactive than 111, and I and I11 are slower, UNIVERSITY 24, CALIFORNIA KURTSCHREIBER if anything, than the saturated analogs IV and V. Los ANGELES RECEIVED OCTOBER 11, 1950 These workers conclude that the double bonds in I and I11 exert no very substantial driving force of the type postulated for cholesteryl compounds, THE EXCHANGE REACTION BETWEEN HYDROpresumably because the geometry is less favorable GEN AND LITHIUM HYDRIDE. THE PREPARAfor participation of the olefinic linkage in the TION OF LITHIUM HYDRIDE-t AND LITHIUM ALUMINUM HYDRIDE-t ionization process. On the other hand, on the prediction that conditions in I were very favorable Sir : The exchange of hydrogen between hydrogen for substantial participation of the olefinic linkage in the ionization process, we had been studying gas and lithium hydride (solid), traced with both the acetolysis of the corresponding p-bromoben- deuterium and tritium, occurs under unexpectedly zenesulfonates, I, m.p. 78.4-79.8' (still heavily mild conditions. At high temperatures lithium hvdride exhibits a meas&able degree of dissociation' and should therefore exhibit exchange H through a mechanism of x dissociation and recoin'H 'X IV V bination. However, the I I1 I11 following observations coiitaminated with isomeric material; prepared indicate that exchange involves a surface reaction from alcohol derived by stereoisomerization of and a slower diffusion process occurring a t rates endo-dehydronorborneol), 11, m.p. 80.2-81.8' such that exchange can be observed at room tem(pure material prepared from hydrolysis product perature and is substantially complete a t 200' from 111), and 111, m.p. 87.4-89.0' (pure material within twenty-four hours. prepared from endo alcohol), all three of which Rate measurements were made on a sample of yield acetate in ca. 80% yield, largely with the lithium hydride-t which had been prepared by structure I1 from infrared and hydrogenation heating 200-mesh lithium hydride (Maywood data. Chemical Co.) with hydrogen gas containing triThe first order rates of acetolysis obtained give tium in a Pyrex flask a t 350'. The hydride was the sequence 1 : I I : I I I of 7000:2000:1 a t 25'. then brought into contact with inactive hydrogen This indicates a very substantial driving force in gas and the uptake of tritium in the gas phase was isomer I which has the proper configuration for followed by means of ion-current measurements2 delocalization of the neighboring electron cloud in on samples of the gas. Between runs the hydride the rate-determining ionization, whereas in I11 was heated with hydrogen gas for sixteen hours a t this must occur essentially subsequent to ioniza- 230' to ensure uniform distribution of tritium tion. Also, it indicates considerable reactivity of throughout the solid phase. The measurements the structure 11. extend over the range from 25 to 200' and are V and cholestanyl benzenesulfonates (with no shown in the accompanying figure as a plot of log (1) Research supported by the Office of Naval Research. (1 - F ) with time, where F is the ratio of specific (2) Roberts, Bennett and Armstrong, THISJOURNAL, 7.2, 3329 L
(1950). (3) Winstein and Adams, ibid., 70,838 (1948).
(1) Hurd and Moore, THISJOURNAL, 67, 332 (1935). (2) Wilzbach and Van Dyken, to be published.
5796
C)OMMUNIC.ZTIONS TO THE
activity of the gas at time, t, to that a t equilibrium. The rate of the exchange reaction is proportiona13to a d log (1 - F)/dt, where Q represents the quantity of hydrogen gas.
EDITOR
Vol. 72
obtained in a radiochemical yield of 80% by redaction of ethyl acetate. .in attempt to prepare lithium aluminum hydride-t by the direct exchange of commercial lithium aluminum hydride with a hydrogen-tritium mixture a t 100" was unsuccessful. Acknowledgments.-We are indebted to Dr. j l . U'einstock for the mass-spectrometric deuterium analyses, and to Prof. W. G. Brown for niany helpful suggestions. KEUNETH
E. WILZBACH
LOUISKAPLAV RECEIVED NOVEMBER 10, 1950 --
s
-\-=.i
1
1
'I 1me, l1outs
of lithium hydride with hydrogen gas in a one-liter flask; temp., mg. atoms of €12; x , 25", 5.5; A, 170", 36.5; 8 , 170°, 6.7; V, 770°, 2.25; 0 , 200°, 382. F is the fraction of equilibrium concentration of tritium in the gas. Ijig. 1 .--FLite of exchange of 18% milliinoles
The non-linearity of these plots, judged by the of Zimens, indicates that the rate of exchange is controlled by a diffusion process within the solid. This conclusion is substantiated by our observation that the rate of formation of H D in a mixture of Hz and Dz in contact with lithium hydride a t 40' is tenfold faster than the rate of exchange with the solid. The increase in exchange rate with hydrogen pressure a t constant temperature rules out diffusion of hydride ions as the ratecontrolling process but would be consistent with a iiiechanisrii iiivolving diffusion of hydrogen mole. ' cules, possibly as H3 - ions. Equilibration experiments a t 200*C. gave a value 3.7 for the equilibrium constant, I\: = (HT)(LiH)/(Hz)(LiT), in agreement with a value :i.CiG calculated from the published partition funclions5for the various isotopic species. The availability of lithium hydride-t has made possible the preparation6 of Iithium aluminum hydride-t, from which a wide variety of tritium labeled organic compounds can be prepared in high yields.' As an example, ethanol-1-t has been riter ria^,^
13) Norris,
J Phys Cdloid Chcm., 54, 777 (1050). Ceol , ASO, No 18 (1111j1
14) Zimens, Arkiu Kemi, Mineral , ( 3 ) Urey. J Chem Soc , 662 (1947)
(0) Finholt
(1947). (71 W press
