Berlin 2024 – wissenschaftliches Programm

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

KFM: Fachverband Kristalline Festkörper und deren Mikrostruktur

KFM 31: Perovskite and Photovoltaics III (joint session HL/KFM)

KFM 31.9: Vortrag

Freitag, 22. März 2024, 11:45–12:00, EW 203

Assessing machine-learning force fields for defect calcula- tions in halide perovskites — •Frederico Delgado¹, Frederico Simões¹, Leeor Kronik², and David A. Egger¹ — ¹Physics Department, TUM School of Natural Sciences, Technical University of Munich, Germany — ²Department of Molecular Chemistry and Materials Science, Weizmann Institute of Science, Israel

The excellent optoelectronic properties exhibited by bulk halide perovskites (HaPs) are important for their photovoltaic performance and have been extensively investigated. Despite this, the ubiquity of both point defects and extended ones like surfaces requires further careful examination of their impact on such properties. Precise investigations of dynamic effects in this context via ab-initio molecular dynamics imply large computational costs, especially when the need for accurate exchange correlation functionals arises. Therefore, in order to sample sufficiently long time scales and sufficiently large supercells, usage of on-the-fly machine learning (ML) forces fields appears to be particularly appealing. In this study, we investigate such defects and their dynamical properties in CsPbBr3 using ML force fields. Our results can aid rationalizing the correlations between local structural dynamics and the observed optoelectronic behavior.

Keywords: Machine-learning force fields; Perovskites; Defects