Criterion of minimum state density in the transition state theory of bimolecular reactions

We discuss two minimum-density-of-states criteria for the location of generalized transition states for chemical reactions. One is due to Bunker and Pattengill; the other is due to Wong and Marcus. We prove that both provide upper bounds on the exact classical equilibrium rate constant. In addition, we show that for several-dimensional systems both methods are exact at threshold, and in the limit of an infinite number of dimensions they agree with the variational theory of reactions of Wigner, Horiuti, and Keck. However, it is also shown that for a finite number of degrees of freedom both methods yield rate constants which are only as accurate as or less accurate than rate constants given by the variational theory of reactions. We note that, where tested by others for actual systems, the differences of the results obtained with the variational and Bunker-Pattengill criteria have been minor.