March 16, 2009
Journal Article

Ti 3p electrons: core or valence?

Abstract

The debate over where Ti 3p electrons should be placed (in the core or valence) has been addressed using pseudopotential-based density functional theory. This work has focused on the TiO2 rutile phase using two different Ti pseudopotentials with 4 or 10 valence electrons, depending on how the Ti 3p electrons are treated. The electronic structure of bulk TiO2 are very similar for the two potentials, and both potentials show similar bulk properties (lattice parameters, bulk modulus). Adsorption of several different types of molecules on a (110) surface gives similar results for organic and inorganic molecules, while adsorption of metal atoms leads to noticeable differences in adsorption energies (26% difference between the two pseudopotentials). This discrepancy is attributed to a larger change in the electronic state of Ti upon metal adsorption, compared to organic or inorganic molecule adsorption. The results also show that the two pseudopotential methods describe oxygen vacancies slightly differently, with a difference in vacancy formation energy being 0.32 eV. Ti in other oxidation states was modeled and both pseudopotentials work reasonably well. This study shows that the 3p electrons can often be treated as core states, but care must be taken to verify that this does not affect the simulation accuracy. Funding was provided by the Department of Energy, Office of Basic Energy Sciences. Computational resources were provided by the Molecular Science Computing Facility located at the Environmental Molecular Science Laboratory in Richland, WA and the National Energy Research Scientific Computing Center at Lawrence Berkeley National Laboratory. All work was performed at Pacific Northwest National Laboratory (PNNL). Battelle operates PNNL for the U.S. Department of Energy.

Revised: April 24, 2009 | Published: March 16, 2009

Citation

Deskins N.A. 2009. Ti 3p electrons: core or valence?. Chemical Physics Letters 471, no. 1-3:75-79. PNNL-SA-59111.