Discovery of Chemically Modified Higher Tungsten Boride by Means of Hybrid GNN/DFT Approach
Аннотация
High-throughput search for new crystal structures is extensively assisted by data-driven solutions. Here we address their perspectives for narrower targeted applications in a data-efficient manner. For verification and experimental validation of the approach proposed, we consider the structure of higher tungsten borides, WB4.2, and eight metals as W substituents to set a search space comprising 375k+ inequivalent crystal structures for solid solutions. Their thermodynamic properties are predicted with errors of a few meV/atom using the graph neural networks fine-tuned on the DFT-derived properties of ca. 200 entries. Among the substituents considered, Ta provides the widest range of predicted stable concentration and lead to the most considerable changes in mechanical properties. Vacuumless arc plasma method is used to perform synthesis of higher tungsten boride with different concentrations of Ta. Vickers hardness samples of WB5-x with different Ta-content is measured showing increase in hardness.
Похожие публикации
сотрудничества и партнерства