Regensburg 2022 – wissenschaftliches Programm

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

MM: Fachverband Metall- und Materialphysik

MM 31: Computational Materials Modelling: Physics of Ensembles 2

MM 31.2: Vortrag

Donnerstag, 8. September 2022, 12:00–12:15, H44

Atomic cluster expansion parametrization of carbon for a fast, accurate and transferable interatomic potential — •Minaam Qamar, Matous Mrovec, Yury Lysogorskiy, Anton Bochkarev, and Ralf Drautz — Interdisciplinary Centre for Advanced Materials Simulation (ICAMS), Ruhr-University Bochum, Germany

We present a parametrization of the atomic cluster expansion (ACE) for carbon that can be employed in large-scale atomistic simulations of complex phenomena. The ACE model is parametrized over an exhaustive dataset of important carbon structures at extended volume and energy ranges, computed using highly accurate density functional theory (DFT). Dispersion corrections are explicitly added to properly account for long-ranged van der Waals interactions. A rigorous validation against DFT data reveals that ACE predicts accurately a broad range of properties of both crystalline and amorphous C phases while being significantly more computationally efficient than other popular machine learning models. We demonstrate the predictive power of ACE on two distinct applications: (1) brittle crack propagation in diamond at finite temperature, and (2) evolution of amorphous carbon structures at different densities and quench rates.