Customized development of promising Cu-Cr-Ni-Co-Si alloys enabled by integrated machine learning and characterization. (15th January 2023)
- Record Type:
- Journal Article
- Title:
- Customized development of promising Cu-Cr-Ni-Co-Si alloys enabled by integrated machine learning and characterization. (15th January 2023)
- Main Title:
- Customized development of promising Cu-Cr-Ni-Co-Si alloys enabled by integrated machine learning and characterization
- Authors:
- Pan, Shaobin
Yu, Jinxin
Han, Jiajia
Zhang, Yanqing
Peng, Qinghua
Yang, Mujin
Chen, Youheng
Huang, Xiang
Shi, Rongpei
Wang, Cuiping
Liu, Xingjun - Abstract:
- 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 Cr3 Co5 Si2 phase was the main reinforcing phase that affected ductility and electrical conductivity. The design method proposed in this paper accelerates theAbstract: 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 Cr3 Co5 Si2 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. Graphical abstract: Image, graphical abstract … (more)
- Is Part Of:
- Acta materialia. Volume 243(2023)
- Journal:
- Acta materialia
- Issue:
- Volume 243(2023)
- Issue Display:
- Volume 243, Issue 2023 (2023)
- Year:
- 2023
- Volume:
- 243
- Issue:
- 2023
- Issue Sort Value:
- 2023-0243-2023-0000
- Page Start:
- Page End:
- Publication Date:
- 2023-01-15
- Subjects:
- Cu-based alloy -- Alloy design -- Microstructure -- Precipitate -- Property
Materials -- Periodicals
Materials science -- Periodicals
Materials -- Mechanical properties -- Periodicals
Metallurgy -- Periodicals
Chemistry, Inorganic -- Periodicals
620.112 - Journal URLs:
- http://www.sciencedirect.com/science/journal/13596454 ↗
http://www.elsevier.com/journals ↗ - DOI:
- 10.1016/j.actamat.2022.118484 ↗
- Languages:
- English
- ISSNs:
- 1359-6454
- Deposit Type:
- Legaldeposit
- View Content:
- Available online (eLD content is only available in our Reading Rooms) ↗
- Physical Locations:
- British Library DSC - 0629.920000
British Library DSC - BLDSS-3PM
British Library HMNTS - ELD Digital store - Ingest File:
- 24649.xml