Modeling Fatigue and Wear of Strongly Correlated Electron Materials

Nick Mosey

Mechanical and Aerospace Engineering, Princeton University

Transition metal oxides (TMOs) are used in various applications to protect surfaces from high stresses, temperatures and/or chemical degradation. The development of improved TMO coatings would benefit from the ability to predict the behavior of these systems under various external conditions. In this presentation, I will discuss techniques for performing accurate first principles calculations of TMOs and incorporating the information obtained through these calculations into multiscale simulation models of wear. The first portion of the talk will focus on our recent advances in performing accurate electronic structure calculations of TMOs, systems that contain strongly-correlated, localized electrons. Such calculations are challenging because these electronic states are poorly described by density functional theory (DFT) and suitable ab initio methods are too costly to be of practical use. This issue is largely addressed with the DFT+U method in which the localized states are treated with a parameterized potential and DFT methods are applied to the remainder of the system. To render the DFT+U methodology fully-predictive, we have developed a new approach for generating the necessary parameters entirely from ab initio calculations. This new approach will be outlined and its abilities will be demonstrated through calculations on chromia, Cr2O3, a representative TMO. The second portion of the talk will focus on the evaluation of mechanical properties of materials, e.g. the tensile and shear strength. In particular, strategies will be outlined for extrapolating the mechanical properties obtained through atomic-level DFT+U calculations to mesoscopic lengths scales, which are relevant to the wear model.