Global machine learning potentials for molecular crystals

Žugec, Ivan; Geilhufe, R. Matthias; Lončarić, Ivor (2024) Global machine learning potentials for molecular crystals. The Journal of Chemical Physics, 160 (15). ISSN 0021-9606

PDF - Accepted Version - article Download (665kB)

Abstract

Molecular crystals are difficult to model with accurate first-principles methods due to large unit cells. On the other hand, accurate modeling is required as polymorphs often differ by only 1 kJ/mol. Machine learning interatomic potentials promise to provide accuracy of the baseline first-principles methods with a cost lower by orders of magnitude. Using the existing databases of the density functional theory calculations for molecular crystals and molecules, we train global machine learning interatomic potentials, usable for any molecular crystal. We test the performance of the potentials on experimental benchmarks and show that they perform better than classical force fields and, in some cases, are comparable to the density functional theory calculations.

Item Type:	Article
Uncontrolled Keywords:	machine learning interatomic potentials
Subjects:	NATURAL SCIENCES > Physics > Condensed Matter Physics
Divisions:	Theoretical Physics Division
Projects:	Project title Project leader Project code Project type Povećanje prostorne i vremenske skale modeliranja materijala iz prvih principa pomoću strojnog učenja-ExtMatModelML Ivor Lončarić UIP-2020-02-5675 HRZZ
Depositing User:	Ivor Lončarić
Date Deposited:	04 Feb 2025 11:28
URI:	http://fulir.irb.hr/id/eprint/9529
DOI:	10.1063/5.0196232

Actions (login required)

View Item