ORNL is devoted to discovery and development of materials and processes needed to revolutionize energy storage and conversion for transportation, grid, and other applications. A significant part of this effort concerns the theory, modeling and simulation of materials. In order to enable the rapid discovery of new materials and processes, as well as to further optimize already identified materials, theory involving first principles many-body methods is required. This is because many materials can have exotic properties, or undergo novel behavior when confined or placed under particular conditions, which pose significant challenges to standard theoretical approaches. Indeed, there are still major difficulties in predicting and explaining many phenomena related to bonding, cohesion, optical properties, conductivity, and other quantum effects. The key to accurately predicting the structural, electronic, and optical properties of materials is an accurate and reliable solution of the Schrödinger equation of quantum mechanics for a large number of electrons. Quantum Monte Carlo (QMC) is a novel computational approach to the solution of the many-body Schrödinger equation. Owing to recent algorithmic improvements of energy minimization, self-healing diffusion Monte Carlo, and fast algorithms for evaluating wave functions, QMC methods are positioned to treat materials at unprecedented accuracy and scale on leadership computing facilities, enabling accurate/reliable calculations of challenging materials for the first time.
For more information, contact: