Peter Taylor is jointly funded by the Theoretical and Computational Chemistry Program in the Chemistry Division and the the New Technologies Program in the Advanced Scientific Computing Division to continue his research in the development and application of methods for the calculation of accurate structures, energetics, and properties of molecules. Methods will be developed for the computation of accurate, ab initio, small molecule, electronic wavefunctions using contracted Gaussian geminals (predetermined linear combinations of explicitly-correlated, two-particle, Gaussian basis functions). Geminal parameters for molecules will be extracted from optimized atomic wavefunctions. Scalable algorithms will be developed for implementation on parallel computers. Taylor asserts that two-electron basis sets offer much more rapid convergence and thus provide a route to higher accuracy for small systems, and a way to extend existing levels of accuracy to larger molecules. Applications include the determination of anharmonic frequencies and vibrational intensities for small molecules, and studies involving the electronic spectroscopy of organic and transition-metal systems and small elemental clusters. Orbital-based concepts and computational methods play a central role in modern chemical theory, but there is a practical limit to the accuracy attainable in orbital-based computations using existing supercomputers. The goal of this project is to develop improved computational methods using geminals (explicit functions of the coordinates of a pair of electrons) rather than orbitals. If successful, this will significantly increase the accuracy of theoretical predictions of gas phase reaction rates such as those required for improved computer models of the earth's atmosphere.
