Since Sato et. al. (Science, 2006) reported the finding of coherent, ordered L12 (γ’) phase in the Co-Al-W system in 2006, efforts have been found everywhere in searching for novel γ’-strengthened Co-based superalloys with properties superior to those of Ni-based superalloys. To reach the goal, multiple alloying elements need to be used and variety of properties must be considered simultaneously, which is challenging. The traditional trial-and-error method is obviously inefficient. In a recent publication, (npj Computational Materials, 2020), Liu and colleagues used machine learning approach to assist the design of Co-based superalloys with multi-performance optimization. In their work, thermodynamic calculated properties and experimental data were used to develop machine learning models and optimization algorithm to search for the alloy compositions with desired multi-property performance. By using this strategy, Liu and colleagues were able to screen out a series of candidate alloys from >210,000 compositions for further experimental investigations. The best performer is identified as Co-36Ni-12Al-2Ti-4Ta-1W-2Cr which possesses the highest γ’-solvus temperature of 1266.5oC, without the precipitation of any deleterious phases, a γ’ volume fraction of 74.5% after aging for 1000 h at 1000 °C, a density of 8.68 g/cm3 and good high-temperature oxidation resistance at 1000 °C due to the formation of a protective alumina layer. Thermodynamic properties, such as γ’-solvus temperature, γ’ volume fraction, liquidus temperature, and solidus temperature were calculated by using Pandat software and PanCo database.
The Figure shows the calculated equilibrium phase fractions vs. temperature for the Co-36Ni-12Al-2Ti-4Ta-1W-2Cr (at%) composition using the most recent version of Pandat software and PanCo database, which confirms the findings in the paper.
P. Liu, H. Huang, S. Antonov, C. Wen, D. Xue, H. Chen, L. Li, Q. Feng, T. Omori, Y. Su, Machine learning assisted design of γ′-strengthened Co-base superalloys with multi-performance optimization, Npj Comput. Mater., 6 (2020) 62. doi:10.1038/s41524-020-0334-5
J. Sato, T. Omori, K. Oikawa, I. Ohnuma, R. Kainuma, K. Ishida, Cobalt-Base High-Temperature Alloys, Science, 312 (2006) 90-91. doi:10.1126/science.1121738.