Material informatics for uranium-bearing equiatomic disordered solid solution alloys. (December 2021)

• Record Type: Journal Article
• Authors: Huang, He; Wang, Xin; Shi, Jie; Huang, Huogen; Zhao, Yawen; Xu, Haiyan; Zhang, Pengguo; Long, Zhong; Bai, Bin; Fa, Tao; Ma, Ce; Li, Fangfang; Meng, Daqiao; Li, Xiaoqing; Schönecker, Stephan; Vitos, Levente
• Is Part Of: Materials today communications. Volume 29(2021)
• Publication Date: 2021-12
• Subjects: Uranium alloys -- Disordered solid solution phase -- Machine-learning model; Materials science -- Periodicals; 620.11
• DOI: 10.1016/j.mtcomm.2021.102960
• Languages: English
• ISSNs: 2352-4928

Abstract

Abstract: Near-equiatomic, multi-component alloys with disordered solid solution phase (DSSP) are associated with outstanding performance in phase stability, mechanical properties and irradiation resistance, and may provide a feasible solution for developing novel uranium-based alloys with better fuel capacity. In this work, we build a machine learning (ML) model of disordered solid solution alloys (DSSAs) based on about 6000 known multi-component alloys and several materials descriptors to efficiently predict the DSSAs formation ability. To fully optimize the ML model, we develop a multi-algorithm cross-verification approach in combination with the SHapley Additive exPlanations value (SHAP value). We find that the Δ S C, Λ, Φ s, γ and 1∕Ω, corresponding to the former two Hume − Rothery ( H − R ) rules, are the most important materials descriptors affecting DSSAs formation ability. When the ML model is applied to the 375 uranium-bearing DSSAs, 190 of them are predicted to be the DSSAs never known before. 20 of these alloys were randomly synthesized and characterized. Our predictions are in-line with experiments with 3 inconsistent cases, suggesting that our strategy offers a fast and accurate way to predict novel multi-component alloys with high DSSAs formation ability. These findings shed considerable light on the mapping between the material descriptors and DSSAs formation ability. Graphical Abstract: ga1 Highlights: A supervised machine learning (ML) model is built based on the phase information of 6000 known multi-component alloys (not containing uranium) and 22 initial materials descriptors. The ΔSC, Λ, Φs, γ and 1/Ω, corresponding to the former two H-R rules, are the most important materials descriptors affecting DSSP formation ability. We apply our ML models to predict DSSP formation probabilities for 375 alloys through the various ML algorithms. 20 uranium alloys out of these 375 alloys are selected for synthesis (arc-melting) and crystal structure characterization (X-ray diffraction). In 17 out of these 20 cases, our predictions of the phase content agree with the measurements conducted. The Λ and Φs can distinguish single phase and dual-phase U-bearing disordered solid solution.