Are reaction rate coefficients additive? Revised transition state theory calculations for OH + alkane reactions

Recent shock tube measurements [1,2] of the reactions of OH radicals with various hydrocarbons provide the incentive for revising an earlier model [3] used to carry out thermochemical transition state theory calculations for the reaction rate coefficients of OH radicals with alkanes. In this article, details of the revised model are presented and calculations are described and compared with experiments. Of particular interest is the question of the reliability of an additive formulation for H atom abstraction rate coefficients; i.e., can we evaluate the total rate coefficient as the sum of site‐specific rate coefficients, k total = Σ i n   H   ik i , where we sum either over primary, secondary, and tertiary H atoms or over some larger number of classes of atoms? It is argued that there are two major reasons for expecting different primary, secondary, or tertiary H atoms to have different rate parameters: (1) activation energies may depend on next‐nearest neighbors and (2) activation entropies can be mass‐dependent. In principle the first factor invalidates the common procedure of treating the total rate coefficient for OH + RH H abstraction processes as the sum of invariant primary, secondary, and tertiary rates multiplied by the respective number of such H atoms in the molecule. The second factor places limits on the accuracy of any additive formulation. A separate question is whether there is sufficient experimental evidence to justify these distinctions, or whether, given experimental and theoretical uncertainties, it is adequate to treat them all as equivalent. It is concluded that (1) mass‐dependent variations are just barely discernible; and (2) there are measurable differences among various primary H atom abstractions, and possibly among secondary atoms, but the data base does not justify distinguishing among tertiary H atoms.