Regensburg 2022 – wissenschaftliches Programm
Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
O: Fachverband Oberflächenphysik
O 43: Frontiers of Electronic Structure Theory: Focus on Artificial Intelligence Applied to Real Materials 1
O 43.2: Vortrag
Mittwoch, 7. September 2022, 10:45–11:00, S054
Realistic Structural Properties of Amorphous SiNx from Machine-Learning-Driven Molecular Dynamics — •Ganesh Kumar Nayak1, Prashanth Srinivasan2, Juraj Todt3, Rostislav Daniel1, and David Holec1 — 1Department of Materials Science, Montanuniversität Leoben, Leoben, Austria — 2Franz-Josef-Strasse 18 — 3Erich Schmid Institute of Materials Science of the Austrian Academy of Sciences, Jahnstrasse 12, Leoben, Austria
Machine-learning(ML)-based interatomic potentials can enable simulations of extended systems with an accuracy that is largely comparable to DFT, but with a computational cost, that is orders of magnitude lower. Molecular dynamics simulations further exhibit favorable linear (order N ) scaling behavior.
Amorphous silicon nitride (a-SiNx) is a widely studied noncrystalline material, and yet the subtle details of its atomistic structure and mechanical properties are still unclear. Due to the small sizes of representative models, DFT cannot reliably predict its structural properties and hence left an anisotropic order parameter. Here, we show that accurate structural models of a-SiNx can be obtained using an ML-based inter-atomic potential. Our predictions of structural properties are validated by experimental values of mass density by X-ray reflectivity measurements and by radial distribution function measured by synchrotron X- ray diffraction.
Our study demonstrates the broader impact of ML potentials for elucidating structures and properties of technologically important amorphous materials.