Validating Spin-Orbit QMC for Equation of State Calculations of Heavy Elements

Heavy element systems are difficult to model for ab initio electronic structure methods. The small energy scales involved, presence of non-negligible relativistic effects, and the presence of non-trivial electron correlations mean that traditional methods can sometimes fail to even qualitatively model these systems correctly. Moreover, for equation of state calculations, there are few methods available for modelling these systems—mainly consisting of different flavors of density functional theory and dynamical mean field theory. A recently proposed generalization of real-space projector Monte Carlo, “Spin-Orbit Diffusion Monte Carlo” (SODMC), has been deployed for molecular systems and a few solid-state systems, but its accuracy and limitations for equation of state calculations are currently unknown. The goal of this work is to benchmark and validate the use of SODMC for equation of state calculations of f-electron systems. By considering the cold curves of Ir, W, and Au, we can carefully quantify the impact of spin-orbit effects on the equation of state. We will do this by comparing SODMC against pseudopotential and all-electron density functional theory, as well as against experimental data. It is hoped that this work will establish confidence to model heavier f-electron systems within a real space QMC framework.

SNL is managed and operated by NTESS under DOE NNSA contract DE-NA0003525