University of Michigan

Michigan Engineering

Skip to content. Skip to navigation

Materials Science and Engineering, University of Michigan

  MSE / Research / Projects / First-principles prediction of phase stability in multi-component solids

First-principles prediction of phase stability in multi-component solids

Materials: Metals Semiconductors Ceramics Organic
Application: Structural Electronic Energy
Technique: Computation

First-principles electronic structure methods are now capable of predicting basic properties of solids, including lattice parameters, relative stabilities and migration enthalpies, with an accuracy that often complements experiment. This is making it possible to design materials with specific properties on the computer. However, before a computer-designed material can ever be synthesized, it must be shown to be thermodynamically stable, or at least metastable. A major component of our research focuses on the first-principles prediction of finite temperature phase stability in multi-component solids. We do this by combining ab initio electronic structure methods (DFT and its extensions) and statistical mechanics techniques (cluster expansions, lattice dynamics, Monte Carlo). We develop statistical mechanics software to calculate temperature–composition phase diagrams of multi-component solids from first-principles. We have applied these methods to successfully investigate and predict phase stability in technologically important materials for lithium battery (LixCoO2, LixNiO2, LixMn2O4), fuel cell and catalysis (oxygen adsorption on Pt and Pt-Ru surfaces) applications.


Personal tools