Natalie Holzwarth and I conduct first principles computer modeling of electronic and structural properties of materials. We work closely with experimental groups on a variety of projects including the study of new materials, surfaces, and defects in crystals. The past five years have seen an impressive development of computational tools for understanding solids, both in terms of the introduction of clever new formalisms and in terms of the availability of more powerful computer hardware and software. This has facilitated the modeling of more complex materials in their ground state and the development of models for excited state and finite temperature treatments of materials. It is an exciting time to participate in these developments. Specifically, density functional theory provides the framework for treating the electronic structure of the materials, and classical and semiclassical molecular dynamics theory provides the framework for treating the nuclear motions.
The principle modeling approach used by the group is the Projector Augmented Wave method developed by Blöchl. This method promises improved accuracy over pseudopotential methods for some problems, while retaining the efficiency of such methods.
Rodney Dunning, a former graduate student and now at Longwood University, uses PAW to explore the properties of point defects in insulators, and has studied the structure of the F-center in calcium fluoride and lithium fluoride. He is also studying the migration kinetics of such centers. Since PAW yields the full nodal wavefunction, it can determine important experimental measurements such as ENDOR spectra and moments of defect absorption spectra that are inaccessible to pseudpotentials.
Alan Tackett, a recent research associate with this group and now with Vanderbilt University, led the development of the first real-space implementation of PAW. He has recently used a interpolating wavelets to produce a PAW implementation using a non-uniform real-space grid. A non-uniform grid enables accurate treatment of larger problems, by placing grid points where they are most needed to properly represent potentials and wave functions.
Rick Matthews home page
Department of Physics
Wake Forest University
[an error occurred while processing this directive]