An efficient route to thermal rate constants in reduced dimensional quantum scattering simulations: applications to the abstraction of hydrogen from alkanes

We present an efficient approach to the determination of two-dimensional potential energy surfaces for use in quantum reactive scattering simulations. Our method involves first determining the minimum energy path (MEP) for the reaction by means of an ab initio intrinsic reaction coordinate calculati...

Full description

Saved in:
Bibliographic Details
Published inThe Journal of chemical physics Vol. 135; no. 9; p. 094311
Main Authors von Horsten, H F, Banks, S T, Clary, D C
Format Journal Article
LanguageEnglish
Published United States 07.09.2011
Online AccessGet more information

Cover

Loading…
More Information
Summary:We present an efficient approach to the determination of two-dimensional potential energy surfaces for use in quantum reactive scattering simulations. Our method involves first determining the minimum energy path (MEP) for the reaction by means of an ab initio intrinsic reaction coordinate calculation. This one-dimensional potential is then corrected to take into account the zero point energies of the spectator modes. These are determined from Hessians in curvilinear coordinates after projecting out the modes to be explicitly treated in quantum scattering calculations. The final (1+1)-dimensional potential is constructed by harmonic expansion about each point along the MEP before transforming the whole surface to hyperspherical coordinates for use in the two-dimensional scattering simulations. This new method is applied to H-atom abstraction from methane, ethane and propane. For the latter, both reactive channels (producing i-C(3)H(7) or n-C(3)H(7)) are investigated. For all reactions, electronic structure calculations are performed using an efficient, explicitly correlated, coupled cluster methodology (CCSD(T)-F12). Calculated thermal rate constants are compared to experimental and previous theoretical results.
ISSN:1089-7690
DOI:10.1063/1.3625960