Professor Jeremy K. Burdett at the Department of Chemistry, the University of Chicago, is supported by a grant from the Theoretical and Computational Chemistry Program at NSF to study the solid state chemistry of a variety of compounds. In particular, the supported research addresses problems in four important areas of structural chemistry where, although there is a great deal of experimental data, the chemical understanding of this data is very limited. The four areas being studied concern: (1) nonstoichiometric materials; (2) stabilities of differently coordinated crystal structures; (3) oxidation states of transition metal ions; and, (4) Zintl phases. The approach being used involves accurate LMTO and LAPW band structure calculations followed by more approximate tight binding, TB, calculations. The accurate LMTO and LAPW results are being fit to a set of TB parameters so as to be able to make maximal connections between chemists ideas of orbitals and solid state physicists band structure results. An important and, in some measure, unique feature of the work supported under this grant is that the accurate band structure methods are used in combination with the more approximate but conceptually valuable tight binding methods. This combination has the capability to provide real chemical insights which can guide understanding and which provide predictive power. The great advances in computer hardware and programs have made it increasingly easy to generate accurate numbers for the problems being studied under this grant. The challenge being addressed by the supported work lies in understanding the chemical and physical meaning of the accurate numbers.
