Theoretical study of collinear Be+FH(v1) →BeF(v2) +H

Publisher Website
Google Scholar
Abstract
The potential energy surface for collinear Be+FH→BeF+H has been studied at various levels of ab initio approximation. A final surface was obtained from a first order configuration interaction wavefunction, using the iterative natural orbital method and a medium‐sized basis set of Slater atomic functions; this is expected to give a semiquantitative description of the reactive process. The exothermicity is computed to be 6 kcal/mole which can be compared with the best experimental value of 2±4 kcal/mole. The barrier height is predicted to be 28 kcal/mole at a geometry where both internuclear separations are extended by about 0.4 bohr from their asymptotic equilibrium values. This surface differs qualitatively from simple LEPS models. The curvature of the reaction path is much more abrupt, the atom effecting little distortion of the partner molecule until quite close approach in both entrance and exit channels. The surface was fit with bicubic splines and dynamics was studied by the quasiclassical trajectory method as a function of initial kinetic energy for the reactant initially in v1=0 and v1=1. The reaction probability and final energy distributions were found to depend sensitively and selectively on the initial kinetic and vibrational energy. Most of the available energy is channeled into product translation; for v1=0 at higher initital kinetic energies, less than 10% of the available energy becomes product vibration. Also, addition of reactant vibrational energy has a profound effect on reaction probability and final vibrational distributions. Examination of typical trajectories made it possible to identify the surface features responsible for the dynamical behavior. For comparison, calculations were also done on a LEPS surface constructed to have the same barrier position and height. Because the LEPS surface has a more gently curved reaction path, with better coupling of vibrational and translational energy, it results in less specific energy use and disposal. For example, 40%–50% of the available energy was channeled into product vibration on the LEPS surface, and addition of reactant vibrational energy effected only small changes in the dynamics. These results underline the dangers of using oversimplified potential surfaces in the study of reactive collision dynamics.