7876
correct in theory, there exists, to our knowledge, no Among them are p-methoxybenzyl chloride (NaN, in example of a negative salt effect of this magnitude on 7 0 z acetone), 4 , 1 2 benzoyl chloride4," (0-nitroaniline in solvolysis of any substrate in any solvent by any salt; 50 wt aqueous acetone), two of the substrates conand in fact there exists but one authenticated example sidered in this manuscript (a-phenylethyl bromide and (NaCl in 50 wt % aqueous dioxane with neophyl tosya-p-tolylethyl chloride with Na- and SCN- in ethanol), late) of a relevant negative salt effect of any magnitude and a,y-dimethylallyl chloride (N3- and SCN- in (b = -0.22 to -0.37).45 We conclude that the availethanol and 90% ethanol).I3 The evidence in each able evidence is overwhelmingly opposed to an S N ~ case against mechanistic alternatives has been preinterpretation of the 2-octyl mesylate data. sented and argued. While the case for the operation Finally, it should be recognized that the 2-octyl of the ion-pair mechanism with any one of these subsystem, our first published example of the operation of strates singly might not be completely convincing as an the ion-pair mechanism, is now but one of several isolated example to all observers it is difficult for us to "borderline" systems which have been shown to consee how the cumulative weight of all of these studies can form to the predictions of the ion-pair mechanism. be denied.
Solvolytic Chemistry of 1,4-Dirnethylbenzene Oxide. A New and Novel Mechanism for the NIH Shift George J . Kasperek,l Thomas C. Bruice,*' Haruhiko Yagi,2 Nicholas Kaubisch,? and Donald M. Jerina2 Contributions from the Department of Chemistry, University of California, Santa Barbara, California 93106, and the National Institute of Arthritis and Metabolic Diseases, National Institutes of Health, Bethesda, Maryland 20014. Received March 31, 1972 The kinetics of the aromatization of 1,4-dimethylbenzeneoxide (I) into 2,5-dimethylphenol (XI) and 2,4-dimethylphenol(III), the latter arising cia the NIH shift of a methyl group, were measured over the pH range of 1-12, in 50% aqueous dioxane, I.( = 0.1 (KCl), and follow the equation -d[I]idr = [ 4 [ k 0 ( k x l k H 2 ) a H ] (where sec-'; k H 1 = 7.3 X l o 2 M-' sec-'; and k R 2 = 5.3 X I O 2 M-' sec-l). In this equation, k o is a k o = 4.8 X spontaneous rate giving rise to a 13:87 ratio of 1I:III and k R 1and kH2are acid-catalyzed terms giving rise to a 54:46 ratio of 1I:III. The reaction proceeding through the k H 1path does not involve detectable intermediates, (IV). whereas that through the kH2path proceeds cia the intermediate 1,4-dimethyl-2,5-cyclohexadiene-2,4-diol The k H 2path represents a new mechanism for the NIH shift. The mechanistic details of these reactions are Abstract:
+
+
discussed.
W
ith the recent demonstration that arene oxides function as the initial intermediates in the enzymatic formation of phenols and other metabolites of the aromatic nucleus, considerable interest has developed in the chemical and biological fate of these labile compounds. In particular, arene oxides have been implicated as agents in the toxicity,4 mutagenicity,; and carcinogenicity3.6 of mono- and polycyclic aromatics. These adverse reactions have been speculated to be the consequence of covalent binding of arene oxides to cellular nucleophiles. Thus, an understanding of the mechanisms by which arene oxides (1) Department of Chemistry, University of California, Santa Barbara, Calif. 93106. (2) National Institute of Arthritis and Metabolic Diseases, National Institutes of Health, Bethesda, Md. 20014. (3) D. M. Jerina, J. W. Daly, B. Witkop, P. Zaltzman-Nirenberg, and S . Udenfriend, J . Amer. Chem. Soc., 90, 6525 (1968); Biochemistry, 9 , 147 (1970). (4) B. R. Brodie, W. D. Reid, A. IC Cho, G. Sipes, G . Krishna, and J . R. Gillette, Proc. N u t . Acad. Sci. U . S., 68, 160 (1971). (5) See ref 6-9 in P. L. Grover, A. Hewer, and P. Sims, FEBS Lett., 18, 76 (1971). (6) P. L. Grover, P. Sims, E. Huberman, H. Marquardt, T. Kuroki, and C. Heidelberger, Proc. N u t . Acad. Sci. U . S.,68, 1098 (1971).
Journal of the American Chemical Society 1 94:22
isomerize to phenols and react with nucleophiles acquires considerable significance. A characteristic of the mixed function oxidasecatalyzed formation of phenols is the migration and subsequent retention of ring substituents originally present at the position of hydroxylation, the NIH shift.' Evidence that arene oxides are intermediates compatible with the NIH shift is found in the isomerizations of 1- 2H-4-methylbenzene oxideS and 1- 'H-naphthalene oxideg to the corresponding phenols. The magnitude of deuterium retention was found to be a function of pH in each case. Kinetics of the isomerization of benzene and naphthalene 1,2-oxide1" have shown that different mechanisms are operative at high and low pH. In the neutral to basic region, an un(7) D. M. Jerina, J. W. Daly, and B. Witkyp, "Biogenic Amines and Physiological Membranes in Drug Therapy, Part B, J. H. Biel and L. G. Abood, Ed., Marcel Dekker, New York, N. Y., 1971, p 413. (8) D. M. Jerina, J. W. Daly, and B. Witkop, J . Amer. Chem. Soc., 90, 6523 (1968). (9) D. R. Boyd, J. W. Daly, and D. M. Jerina, Biochemisrrj,, 11, 1961 ( 1972). (10) G. J. Kasperek and T. C. Bruice, J. Amer. Chem. SOC.,94, 198 ( 1972).
Nooember 1, 1972
7877
catalyzed, spontaneous rearrangement occurs while at low pH (loand is depicted in Scheme IV. The Scheme IV
CHJ XVI
I11 1-H'
VH+ I1 1-H'
protonated arene oxide XI11 can either open (path a) to form carbonium ions XIV and XV which rearrange to intermediate protonated ketones XVI and XVII,
Bruice, et al.
i Solvolytic Chemistry of 1,4- Dimethylbenzene Oxide
7882
or XI11 can proceed directly to XVI and XVII via a concerted mechanism (path b); it is also possible to envision direct formation of I1 from XV by loss of a proton. Whether the reaction proceeds through path a or b or both cannot be determined from the data; however, the involvement of both pathways must be considered. The rearrangement of l-*H- and 2-2Hnaphthalene 1,2-oxides is known to provide 1-naphthol containing different amounts of deuterium which indicate mechanistic pathways through both XVIII (retains deuterium) and XIX (loses deuterium).
Scheme V
XI I1
IV
1
I1
+
I11
+
H,O
?fH CK
x IX
XL‘III
The portion of the acid-catalyzed reaction going through the diol path (kH2),Scheme 111, may involve either reaction of the protonated arene oxide XI11 or carbonium ion XIV with water to produce diol IV. The fact that carbonium ion XIV is unusually stabilized (positive charge stabilized by two allylic and one methyl group XIVa) ac-
Scheme VI
IV
CH, XIVa
counts for its exchanging OH groups with solvent H,O faster than it aromatizes. There are two possible mechanisms for the formation of I1 and I11 from IV. In the first (Scheme V), I1 and I11 are produced via a steady-state concentration of protonated arene oxide XI11 in a mechanism identical with that in the direct route. The second mechanistic possibility (Scheme VI) is more complicated and involves a 1,2-diol V as an intermediate in the pathway to I1 while I11 is produced via the same route as in the direct mechanism. Although an unequivocal choice cannot be made between these two mechanisms, the one depicted in Scheme V is “preferred” because it predicts the II/III ratio to be the same in the direct path and the diol path because
Journal of the American Chemical Society
both pathways involve the same product-determining intermediate XIII. Examination of this ratio as calculated by the more accurate product ratio method yields 1.55 for the direct path and 1.32 for the diol path in good agreement with the mechanism of Scheme V.I6 Acknowledgment. This work was supported in part by a grant to T. C. Bruice from the National Institutes of Health. (16) Since submitting this study, conclusive proof that aromatization of arene oxides occurs cia rate-determining carbonium ion formation has appeared [see G. J. Kasperek, T. C. Bruice, H. Yagi, and D. M. Jerina, J . Chem. Soc., Chem. Commun., 784 (1972)l.
1 94:22 1 November I , 1972
