Combining transition state theory with quasiclassical trajectory calculations. Part 1.—Collinear collisions

Abstract

Quasiclassical trajectory calculations of two kinds have been performed on collinear reactive collisions involving three atoms: i.e. A + BC → AB + C. In the first set of calculations, trajectories are initiated with A and BC (v= 0) well-separated. In the second, a normal mode analysis is carried out at the saddlepoint on the potential energy surface and the line (S*) orthogonal to the reaction coordinate is defined. Trajectories are then run “forward” and “backward” from points on S*, with the system initially being assigned zero-point energy associated with the motion along this line and corresponding to the “real” frequency in standard transition state theory. Results are compared for all 8 combinations of L(1 a.m.u.) and H(35 a.m.u.) atoms on three different potential energy surfaces, as well as for F + H₂ and H + F₂. On surfaces with early barriers, motion between the separated reagents and the transition state exhibits a high degree of vibrational adiabaticity and the reaction probabilities and patterns of energy disposal from the two sets of calculation are very similar. The implications of these results for standard transition state theory and for increasing the effectiveness of full-scale quasiclassical trajectory calculations are discussed.