Berlin 2018 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

MM: Fachverband Metall- und Materialphysik

MM 51: Topical Session (Symposium MM): Big Data in Materials Science - Managing and exploiting the raw material of the 21st century

MM 51.5: Vortrag

Donnerstag, 15. März 2018, 11:15–11:30, H 0107

Data-driven assessment of the transferability of effective interatomic potentials — •Y.V. Lysogorskiy, T. Hammerschmidt, and R. Drautz — Atomistic Modelling and Simulation, ICAMS, Ruhr-Universität Bochum, D-44801 Bochum, Germany

Many materials properties are directly accessible with high-precision by quantum-mechanical calculations using density-functional theory (DFT). However, the computational costs of the DFT calculations make many important materials properties difficult to calculate. To tackle this problem interatomic interaction potentials are widely used. A large number of interatomic potentials is available for a wide range of chemical elements and their mixtures. A central difficulty for applying potentials is the typically very limited information about their reliability for a particular simulation. We performed high-throughput calculations using DFT and interatomic potentials in order to investigate the transferability of interatomic potentials statistically. The results of our calculations are collected in a specially designed database for further analysis. The consequent statistical and correlation analysis of properties and their deviations with respect to reference values allow us to quantify the errors and transferability of interatomic potentials.