Customized development of promising Cu-Cr-Ni-Co-Si alloys enabled by integrated machine learning and characterization
Shaobin Pan,Jinxin Yu,Jiajia Han,Yanqing Zhang,Qinghua Peng,Mujin Yang,Youheng Chen,Xiang Huang,Rongbei Shi,Cuiping Wang,Xingjun Liu
DOI: https://doi.org/10.1016/j.actamat.2022.118484
IF: 9.4
2022-11-14
Acta Materialia
Abstract:Two types of alloys, Cu-Ni-Co-Si and Cu-Cr-Zr, are considered candidate materials for next-generation integrated circuits due to their superior comprehensive performance. However, the rapid development of these two types of alloys remains difficult using conventional simulation techniques. Machine learning offers a new tool for accelerating the design and discovery of new materials with required property profiles. Herein, composition-process-property database of the six-element Cu-Cr-Ni-Co-Si-Zr alloys were established, and a novel strategy of customized performance design for different application environments was proposed. Then, four alloys with different performance characteristics were rapidly screened from 850,500 candidates using a multi-property segmented screening method, and the predicted results agreed well with the experimental results. Importantly, the developed Cu-1.0Cr-1.0Ni-2.5Co-0.8Si alloy was used as a bridge alloy to link the Cu-Ni-Co-Si and Cu-Cr-Zr alloys together, filling the gap in the mid-segment performance (220-240 HV, 45-65% IACS) of Cu-based alloys. Interestingly, the studied alloy was a dual-phase precipitation-strengthened alloy. It was found that the small spherical (Co, Ni) 2 Si phase was the main influence on the micro-hardness and strength, while the large rod-shaped Cr 3 Co 5 Si 2 phase was the main reinforcing phase that affected ductility and electrical conductivity. The design method proposed in this paper accelerates the development of the Cu-Cr-Ni-Co-Si alloy system, which has great potential for application in integrated circuits and heat sinks.
materials science, multidisciplinary,metallurgy & metallurgical engineering