No abstract

Quasiclassical trajectory studies of the H+H2 reaction on an accurate potentialenergy surface. III. Comparison of rate constants and cross sections with experiment Quasiclassical trajectory studies of the H+H2 reaction on an accurate potential energy surface. II. Effect of initial vibration and rotation on reactivity Quantum mechanical calculations are compared to quasiclassical trajectory forward (QCT) calculations for the collinear. endoergic reaction Hlnl)+I->H+HI for two different potential energy surfaces, a rotated-Morse-curve (RMC) surface and the semiempirical valence-bond surface of Raff et al. Vibrationally stateselected reaction probabilities and rate constants and Arrhenius parameters are presented. Thermally averaged rate constants and their Arrhenius parameters are also given. For one of the potential energy surfaces, quasiclassical trajectory reverse histogram (QCTRH) calculations were also performed. The results show that classical mechanics and quantum mechanics are in significant qualitative agreement for state-selected properties. Specifically, for the n I = 0 state of the Raff et al. surface the quantum mechanical reaction probabilities are very small (less than 0.005) and the QCT method predicts this state to be totally non-reactive. For all other states on both surfaces quantum mechanics and QCT and QCTRH results all agree that reaction probabilities attain much higher values (up to 0.85). For both surfaces quantum mechanical and QCT results predict that excited vibrational states make significant contributions to the thermal reaction rates, although the methods disagree as to which vibrational state is quantitatively most important. Quantitative agreement with quantum mechanical results is obtained only with the QCTRH method for thermally averaged rate constants (agreement within 2%) and with both QCT and QCTRH methods for the Arrhenius parameters (agreement within a few tenths kcal mol-1 for activation energy). However, to achieve such agreement the QCT method had to be suitably modified to correct unphysical discrepancies in the threshold energy region. We present tables of these and many other results as a function of temperature. These should be useful in assessing the validity of trajectory studies of various kinds of reaction attributes under conditions where they are used to interpret experiments.

No abstract