Regensburg 2025 – wissenschaftliches Programm
Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
MM: Fachverband Metall- und Materialphysik
MM 31: Data-driven Materials Science: Big Data and Worksflows
MM 31.1: Vortrag
Donnerstag, 20. März 2025, 15:00–15:15, H10
Thermodynamic stability of the materials in the Materials Cloud three-dimensional crystals database (MC3D) — •Timo Reents1,2, Marnik Bercx1, and Giovanni Pizzi1,2 — 1Laboratory for Materials Simulations (LMS) and National Centre for Computational Design and Discovery of Novel Materials (MARVEL), Paul Scherrer Institut (PSI), CH-5232 Villigen PSI, Switzerland — 2École Polytechnique Fédérale de Lausanne (EPFL), Lausanne, Switzerland
High-throughput studies based on ab initio methods such as Density Functional Theory (DFT) enable the analysis of physical properties across a broad chemical space. Here, we present the Materials Cloud three-dimensional crystals database (MC3D), a DFT optimized and curated structural database of experimentally known inorganic crystals. All calculations are managed and driven by the AiiDA [1, 2] workflow engine, allowing to browse the full provenance graph and to share the results in the Materials Cloud [3]. We introduce the protocols behind MC3D, the new frontend, and we then focus on the thermodynamic stability. To improve the agreement between the theoretical and experimental thermodynamic stability, we apply empirical [4] and machine-learning [5] based corrections, and improve upon them, discussing the agreement with experimental data on stability.
[1] Huber, S.P. et al., Sci Data, 2020, 7, 300.
[2] Uhrin, M. et al., Comp. Mat. Sci., 2021, 187, 110086.
[3] Talirz, L. et al., Sci Data 7, 299 (2020).
[4] Stevanović, V. et al., Phys. Rev. B, 2012, 85, 115104.
[5] Gong, S. et al., JACS Au, 2022, 2, 1964-1977.