The high k~/ k~ ratio foundexperimentally for methyl radicalabstraction reactions is explained on the basis of classical rate theory. I t is shown that if the bending frequencies in the activated complex approach zero, good agreement is found between experimental and calculated kn/ko values. For abstraction of H and D from hydrocarbons, both the tunneling and classical rate theories predict acceptable k~/ k o values. Abstraction of H and D from Ha and Ds, however, can best be explained on the basis of a classical process.
ISTRODUCTIONThe ratio of rate constants, kH/kD, for the abstraction of H and D from hydrocarbons by methyl radicals has been found to be higher than that calculated from classical rate theory when the bending frequencies of the RC-H bond is assumed to remain coilstant from the initial to final states (1-6). T o explain these high values for kHlkD, several workers (1, 2) have proposed that the rates are largely determined by appreciable participation of a quantum mechanical tunneling process. The basis for the tunneling mechanism is the theory of Johnston and Rapp (3), who give the relation for methyl radical abstractions based on proton tunneling (2). This theory is attractive since it does predict the experimental values of kBlkD with a good degree of accuracy, but the theory does not explain the lower values of kH/kD found for methyl radical abstraction from H1 and Dq. In Table I , the kH/kD values for various reactions are given together with the ratio of frequency factors, AH/AD, and the difference in activation energies, A E , found experimentally. I t will now be shown that these observed values can be correlated with classical rate theory and that quantum mechanical tunneling need not be invoked.
CALCULATIOKSFrom absolute reaction rate theory (9), the ratio of rate constants is given by k~ 3 8 -7 3n-6