Development of a methodology for the description of dispersion interactions at the DFT level.

This project focuses on the inclusion of dispersion interactions into geometry optimizations carried out at the relatively low-cost DFT level, thus allowing to obtain accurate geometries for large systems, such as biomolecules. These systems cannot be treated by expensive high-level methods that provide good description of van der Waals forces, such as CCSD(T), but do require a good quantitative description of dispersion interactions for reliable results. Since even the smallest biologically active peptides contain hundreds of atoms, their treatment at DFT level still requires highly parallelized codes and access to an extendedinfrastructure. The addition of dispersion to the DFT treatment will enhance the quality of results significantly without increasing the computation cost substantially. In this project we propose to add a post-SCF correction to the energy of a system, based on pair-wise dispersion interactions between the atoms of the system. By applying analytic gradients to the energy correction, a fully automated optimization procedure can be constructed. In the preliminary stages of research the method displayed major advantages over other methods currently being developed for this type of interactions. In order to accomplish this project an extension of the available computer infrastructure is required.

Keywords:DFT METHOD, DISPERSION

Disciplines:Theoretical and computational chemistry, Other chemical sciences