I got started fitting tight-binding models in 1995, doing a postdoc at Los Alamos. I fitted tight-binding parameters for silicon with 36 adjustable parameters to a DFT database using force-matching techniques. That code has been extended over the years to fit spline-based pair potentials, EAM, MEAM, SW (Stillinger-Weber), MEAM+SW, and EAM+SW models. The MEAM+SW model consists of a MEAM term added to a SW term. More recently we have experimented with MEAM(n) and MEAM(n)+SW models which contain multiple MEAM terms. My fitting code contains sophisticated local and global optimizers, and can fit models with several dozen splines or several hundred parameters to large databases, using an ordinary workstation-class computer for fitting. We use statistical techniques and uncertainty quantification to validate model performance.
Over the last seven years, I have been self-employed and have had contracts with Los Alamos National Laboratory and Lawrence Livermore National Laboratory. This has resulted in a variety of unpublished work. Please contact me if you are interested in any other contract work, consulting, or commercial ventures.
Silicon (tight-binding and MEAM)
BCC, HCP, and FCC metals (classical potentials)
Titanium -- MEAM to model phase transformations between BCC, HCP, and omega phases
Most of my work in fitting potentials has been unpublished, however see publication list on Google Scholar:
Contributor of the following OpenKIM content: