Molecular Electronic Structure Theory: Methods and Applications

University Of Georgia Research Foundation Inc, Athens GA

Investigators

Abstract

Henry (Fritz) Schaefer of the University of Georgia is supported by an award from the Chemical Theory, Models and Computational Methods Program in the Chemistry Division and the Computational and Data-Enabled Science and Engineering Program (CDS&E) to develop new quantum chemistry methods to study the behavior of electrons in molecules. The electrons in atoms and molecules are responsible for making and breaking chemical bonds. Very small particles, such as electrons, follow the laws of quantum mechanics. Thus, quantum chemistry or electronic structure methods are essential for a fundamental understanding of chemical reactions as well as other important atomic and molecular processes. The basic equation of quantum mechanics, the Schroedinger equation, is solved on a computer. An exact quantum mechanical description of all but the smallest molecules would require vast amounts of computer time. A goal of quantum chemists is to develop more efficient and more accurate methods. Schaefer and his co-workers have developed important new methods called Density Cumulant Theory. (DCT) This research project has two, distinct but related aims. The first is the hypothesis, formulation, development, implementation, and testing of the DCT method. Second, these new techniques are used to solve important prototypical chemical problems. These methods are implemented in a publicly available, user-friendly quantum chemistry software package called PSI4. Many of the current developers of PSI4 are currently members of Schaefer's research group. In the Schaefer laboratory, an important new method is being developed, Density Cumulant Theory (DCT). More specifically being pursued are (1) N-representability conditions from analyses of unitary transformations; (2) N-representability from the idempotency of the two-particle density matrix; (3) alternative DCT hierarchies using the irreducible contracted Schroedinger equation; (4) three-electron correlation effects in DCT; and (5) efficient DCT implementations using pair natural orbitals for large scale chemical applications. Also being tackled are linear-scaling algorithms for very large molecular systems. The freely available, open-source PSI4 suite of next generation computer programs for electronic structure theory is being developed and distributed. Applications ranging from physical and materials to inorganic, organic and biological chemistry are successfully being completed with state-of-the-art theoretical methods.