Highly Accurate Local Pseudopotentials of Li, Na, and Mg for Orbital Free Density Functional Theory

Abstract

We present a method to make highly accurate pseudopotentials for use with orbital-free density functional theory (OF-DFT) with given exchange-correlation and kinetic energy functionals, which avoids the compounding of errors of Kohn-Sham DFT and OF-DFT. The pseudopotentials are fitted to reference (experimental or highly accurate quantum chemistry) values of interaction energies, geometries, and mechanical properties, using a genetic algorithm. This can enable routine large-scale ab initio simulations of many practically relevant materials. Pseudopotentials for Li, Na, and Mg resulting in accurate geometries and energies of different phases as well as of vacancy formation and bulk moduli are presented as examples.

15 pages, 1 figure, 3 tables

Cited by in corpus (7)

• Data-driven kinetic energy density fitting for orbital-free DFT: linear vs Gaussian process regression
• Non-parametric Local Pseudopotentials with Machine Learning: a Tin Pseudopotential Built Using Gaussian Process Regression
• Machine learning of kinetic energy densities with target and feature averaging: better results with fewer training data
• A machine-learned kinetic energy model for light weight metals and compounds of group III-V elements
• Truncated Non-Local Kinetic Energy Density Functionals for Simple Metals and Silicon
• Machine learning-guided construction of an analytic kinetic energy functional for orbital free density functional theory
• Alkali-metal vibrations in bcc, fcc, hcp, and 9R structures: Implications for the energetics of Li and Na martensitic phases