Principal Investigator Michael A. Collins Project s01

Research School of Chemistry Machine VP

Co-Investigators: Meredith J. T. Jordan and Keiran C. Thompson

Research School of Chemistry

Molecular Potential Energy Surfaces by Interpolation

Chemical reactions are simply the combination or fragmentation of groups of atoms whose movement is governed by the molecular potential energy surface. We have developed a new method for constructing such a surface from ab initiio quantum chemistry calculations, so that the dynamics of chemical reactions can be studied and understood.

What are the basic questions addressed?

How efficiently can we construct a molecular potential energy surface (PES) by interpolation of local Taylor expansions of the surface? How can the process be optimised?

What are the results to date and future of the work?

Truncating the local Taylor expansions at second order is computationally most efficient, while we have also shown that if the ab initio calculation of third order derivatives of the electronic energy can be improved in efficiency, then third order expansions should prove to be superior in accuracy and computational efficiency. The ab initio calculation of a surface (at the Hartree-Fock level of theory) for the OH + H2 Æ H2O + H reaction has been accomplished. Further development of the methodology is required for applications to larger systems.

What computational techniques are used and why is a supercomputer required?

Two aspects of this project benefit from a supercomputing environment: ab initio quantum chemistry calculations and the use of classical trajectories in building the surface and evaluating the dynamics.


Convergence of molecular potential energy surfaces by interpolation: application to the OH + H2 Æ H2O + H reaction. M. J. T. Jordan, K. C. Thompson and M. A. Collins, Journal of Chemical Physics, 102, 5647-5657 (1995).

The utility of higher order derivatives in constructing molecular potential energy surfaces by interpolation, M. J. T. Jordan,. K. C. Thompson and M. A. Collins, Journal of Chemical Physics, 103, 9669 (1995).