Grand Challenges in Computational Materials Science: Towards Reliable Predictions of Quantum Materials

by Paul Kent (Oak Ridge National Laboratory)

Monday Dec 14, 2020, 2:00 PM → 3:00 PM America/Chicago

BIO:

Paul Kent is currently a distinguished R&D staff at ORNL. He completed his PhD in Theoretical Physics at the University of Cambridge in 1999. He joined ORNL after moving to the USA for postdoctoral positions at the National Renewable Energy Laboratory and the University of Tennessee.  Currently he directs the Center for Predictive Simulation of Functional Materials and leads the open source QMCPACK applications development as part of DOE’s Exascale Computing Project. He is a Fellow of the American Physical Society and was most recently awarded the 2020 ORNL Director’s Award for Outstanding Individual Accomplishment in Science and Technology.

Abstract:

One of the longstanding grand challenges in computational materials science and condensed matter theory is to be able to reliably and accurately solve for the quantum mechanical properties of general materials. This capability is necessary in many areas ranging from materials design through to prediction and interpretation of novel quantum phenomena. Today while there are many successful approximate methods, we lack general and widely applicable methods able to provide a ground truth. This is a general problem, but particularly problematic for systems such as quantum materials where the results are particularly delicate to approximations made. In this talk I will describe recent advances in Quantum Monte Carlo methods, which combined with the increased capabilities promised by Exascale computing promise to bring us significantly closer to the goal of reliable materials predictions.

 

Kent Poster