Insights from symmetry: Improving machine-learned models for grain boundary segregation - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Computational Materials Science Année : 2024

Insights from symmetry: Improving machine-learned models for grain boundary segregation

Y. Borges
L. Huber
G. Demange

Résumé

Grain boundary (GB) segregation substantially alters structural and functional properties of metallic alloys, including strength, fracture mode, and corrosion resistance. A critical factor causing the variability in the segregation of solute atoms at GBs is the atomic structure of the boundary. GB segregation and its link with GB structure should thus be included in modern computational strategies for polycrystalline materials design. Here, we first propose an efficient and user-friendly Machine-Learning (ML) framework capable of predicting the segregation of different solute atoms at grain boundaries in the form of a segregation energy density, from the corresponding undecorated GB atomic structure. We then show that ML provides a fresh and promising perspective to address the long-standing issue of GB structure-segregation property relationships, in the form of two correlated atomic parameters quantifying the degree of structural and segregation symmetry respectively.
Fichier non déposé

Dates et versions

hal-04302466 , version 1 (23-11-2023)

Identifiants

Citer

Y. Borges, L. Huber, H. Zapolsky, R. Patte, G. Demange. Insights from symmetry: Improving machine-learned models for grain boundary segregation. Computational Materials Science, 2024, 232, pp.112663. ⟨10.1016/j.commatsci.2023.112663⟩. ⟨hal-04302466⟩
7 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More