Extending the reach and accuracy of Quantum Monte Carlo in materials

Paul Kent
Oak Ridge National Laboratory

Advances in quantum mechanics-based approaches to predict the properties of materials with high accuracy and reliability are required for many materials of current interest. This includes quantum materials where delicate couplings of spin, charge, spin-orbit, and geometry result in novel phenomena, and in general for materials where benchmark reference data is desired for validation or for use in a multiscale approach. In this presentation I will outline ongoing developments in ab initio Quantum Monte Carlo methods to meet this challenge including the development of spin-orbit, geometry relaxation, new trial wavefunctions, and computational developments to bring these methods to Exascale supercomputers. I will describe recent applications of these methods to quantum materials, highlighting currently open questions on how the methodology can be further improved.

Back to Workshop I: Increasing the Length, Time, and Accuracy of Materials Modeling Using Exascale Computing