COMMUNICATIONS TO THE EDITOR
June 3, 1961
ethvlene. Evidence for it is based on studies of hydrogen isotope exchange between gaseous deuterium and gamma ray irradiated Marlex-50 polyethylene. Such exchange duri?tg the irradiation has been studied by Varshavskii, et al.? These authors postulated a chain reaction R. + D) +RD D (1)
+
(2)
D+RH+R.+HD
Akdditionalreactions can be suggested R. D. R,
+ D +K D + D . +D> + R. +RR
(3) (4)
(5,
However, in the Russian work reaction (1) mtght have involved electronically excited or ionized methylene groups because of the existence of these species rather than the free radical R. during the irradiation. I n order to clarify this point we have irradiated Marlex-30 a t liquid nitrogen temperature and in vacuo, warmed the sample to room temperature, pumped off the evolved hydrogen gas and then introduced deuterium gas subsequent to the irradiation. During this treatment probably two thirds of the free radicals decayed6; nevertheless copious exchange up to 25yG conversion of DL to HD involving the remaining free radicals occurred over a period of about 100 hours. During this period there was less then O.2Y0 decrease in the total gas pressure; thus reactions (3) and (4) may be considered neg1igible.j Letting y represent PHJPOD?where POD,is the initial deuterium pressure, it can be shown that for various mechanisms where 2bi.a is the D-H fractionation factor, and and /3 have the values:
01
2585
existed in the polyethylene subsequent to the irradiation. I t is further evident that the R. of Eq. (2) is not a t the same location in the solid polyethylene as the R. of Eq. (1). Thus we have a n additional mechanism for free radical migration in solid polyethylene. Increase of deuterium gas pressure should not increase dy,idt, for reasons to be discussed later, but should lower the extent of D-H exchange because of acceleration of R. decay. This prediction of the mechanism postulated here is borne out by the data. There is a possibility that some of the exchange may be the result of a reaction in which the D-atom replaces the H-atom on the alkyl free radical without free radical migration as suggested by Voevodskii, et ~ l . in, ~a study involving exchange between deuterium gas and ethyl radicals. While such exchanges may occur to a small extent, evidence will be given later that this mechanism cannot account for the major effect. By R.-decay? in this note we include also the possibility of conversion of the alkyl to the allyl free r a d i ~ a l . ' , ~Because ,~ of the greater stability of the allyl free radical, reaction (1) involving it would be less likely by a factor of about lo-". This research was supported by the U. S. Atomic Energy Commission. (7) V. V . Voevodskii, G. K. Lavrovskaya and R. E. Mardalelshvili, Doklady Acad. N a u k , S.S.S.R., 81, 215 (1951). (8) Fulbright travel grantee from t h e University of Louvain, Belgium. DEPARTMENT OF CHEMISTRY
NORTHWESTERN UNIVERSITY MALCOLM DOLE ILLINOIS FRAYCISC u c c o * EVANSTON, RECEIVED MARCH 30, 1961 THE ACID-CATALYZED ISOMERIC REARRANGEMENT OF cis AND ~ Y U ~ , ~ - ~ - M E T H Y L - ~ CY CLOHEXENOL-0 '8
Sir: Several years ago the acid-catalyzed (HClOI) decay at bt rearrangement of cis- (I) and trans-5-methyl-2First order R. cyclohexenol (11) in 35% aqueous acetone was in(a/&) (1 - g j (blkr) (1 - q ) decay q = e-kd vestigated in these laboratories and i t was found Second order R. that with both isomers the pseudo first-order rate decay (a,/Bj In (1 + Bt) (blB) In (1 + Bt) of loss of optical activity (ka) is several times The meaning of the kinetic constants a , b, B and ka larger than that of geometric isomerization ( k i ) . will be discussed in the complete paper; t is the This means that the acid-catalyzed allylic retime. Suffice i t to say that for t very small, for the arrangement is stereospecific in the sense that each three sets of a and /3 values Eq. (6) reduces to y isomer is converted to its enantiomer faster than to at. At long times, however, there is a marked its geometric isomer-in this symmetrical system, difference between the three possibilities. Our allylic rearrangement without geometric isomerizaevidence indicates that R.decays by a first order tion results in the interconversion of enantiomers. process at least a t long times (the mechanism of this I t was suggested that the racemization with preswill be considered later); the extensive and ex- ervation of geometric configuration may involve cellent work of Lawton, Balwit and Powell6 also either an SNi intramolecular rearrangement of demonstrates a first order decay of the alkyl free the conjuga.te acids of the alcohols or an interradical in irradiated Marlex-50. ' molecular stereospecific (cis) S N ~process. Referring to Eqs. (1) and ( 2 ) it is now evident that these reactions must occur provided that HX excited or ionized methylene groups no longer CY
KOfree radical
B
(4) Ya. hl. Varshavskii, G . Ya. Vasil'ev, V. I,. Karpov, Yu. S. 1,azurkin and I. Ya. Petrov, D o k i a d y A k a d . Ilauk. S.S.S.R., 118,315 (1958). ( 5 ) T h e possibility of a back thermal reaction between molecular hydrogen and irradiated polyethylene suggested by t h e work of hrvia and Dole. Proc. 2nd Conf. Peaceful Uses of Atomic Enerzy, United Nations, Geneva, 29, 171 (1058) is being further investigated. (6) E. J. Lawtun, J. S . Balwit and R. S. Powell, J . Chew%.Phys., 33, 395 (1960).
H36@J I
I1
I\.c have now reexamined this rearrangement and have confirmed the earlier report' that k, > ki (1) H. I,. Goering and E. F. Silversmith, J. A m . Ckent. S O L . ,79, 3-18(1057).
COMMUNICATIONS TO THE EDITOR
2586
for hoth isomers. In addition we have determined the pseudo first-order rate of oxygen exchange for both isomers to learn more about the excess racemization. The three measured reactions: (a) loss of optical activity (eq. 1)) (b) isomerization (eq. 2) and (c) ka active-ROH +inactive-ROH
(1)
ki c i r
cis-ROH
7 trans-ROH
(2)
ki tram kczc
RO1aH --3 ROH
(3)
oxygen exchange (eq. 3), are cleanly pseudo firstorder, i.e., these processes are first-order in acid (the concentration of which remains constant) and alcohol. The kinetic behavior is consistent with the interpretation that these processes involve reversible protonation of the alcohols followed by first-order transformations of the con jugate acids of the alcohols.' If the three rates are compared a t the same acid concentration the relative magnitudes of the pseudo first-order constants correspond to those of the first-order constants for transformations of the conjugate acids of the alcohols. The pseudo first-order constant for the stereospecific racemization (krac, eq. 4) is k, - Ki. krm rir
d-I
+dl-I
(44
kme tvona
d-I1
-+
dl-I1
(4b)
Pseudo first-order constants for reactions 1 4 a t 30" in 35% aqueous acetone containing 0.095 M Hc10.1are given in Table I. Rate constants for isomerization, k i clr and ki were determined from the pseudo first-order rate of equilibration ( k i cis 4- k i trans) and the equilibrium constant for equation 2 ; K,, = ki cis/ki tra,z.7 = 1.22. The same k i l r a n s ) were obtained values for K,, and (ki
