About this Abstract |
Meeting |
2023 TMS Annual Meeting & Exhibition
|
Symposium
|
Materials Genome, CALPHAD, and a Career over the Span of 20, 50, and 60 Years: An FMD/SMD Symposium in Honor of Zi-Kui Liu
|
Presentation Title |
Stability of Transition Metal High Entropy Alloys: From First-principles and Machine Learning |
Author(s) |
Ying Chen, Nguyen-Dung Tran, Chang Liu, Xinming Wang, Jun Ni |
On-Site Speaker (Planned) |
Ying Chen |
Abstract Scope |
Along the first-principles study on a specific quinary high entropy alloy FeNiCoCrPd which is synthesized by substituting Mn in Cantor alloy by Pd, to reveal the mechanism of inhomogeneous feature of Pd in enhancing the mechanical property, we have accumulated 1,000 DFT data of sub-systems of binary, ternary, quaternary for all equiatomic composition and typical non-equiatomic compositions using the special quasi-random structures (SQS), further extended to some senaries, for fcc. bcc and hcp structures. Based on this FeCoNiCrMnPd data set, systematic predictions are conducted using machine learning. The mesh searching for virtual systems of FeCoNiCrMnPd+x (x=all 3d-, 4d-elements, Mg, Al, Si, etc.) gave a general picture of solid solution stability of the transition metal ternaries, quaternaries. Furthermore, the elemental convolution graph neural networks (ECNet) combing transfer learning are attempted to explore the stability and properties of the higher compositional systems mainly based on the data of binaries and ternaries. |
Proceedings Inclusion? |
Planned: |
Keywords |
High-Entropy Alloys, Computational Materials Science & Engineering, Machine Learning |