Data-driven design of Ni-based turbine disc superalloys to improve yield strength

Increasing the thrust-weight ratio of aeroengines requires development of high-strength and stable high -temperature materials. A data-driven design of Ni-based turbine disc superalloys is performed to improve the yield strength to reach the target. Through first-principles calculations determining the design super -alloy system, the theoretical models and Calculation of Phase Diagram (CALPHAD) screening composi-tions, and machine learning extrapolating prediction, 14 compositions are selected from 2,865,039 com-position combinations. Ni-17Cr-8Co-1Mo-1W-6Al-3Ti-1Nb-1Ta is selected to verify the design accuracy. Experimental tests prove that the designed alloy has trade-offs of microstructure with satisfying design targets, and then, the yield strength is higher in the designed alloy than in commercial superalloys, reach-ing 728 MPa at 850 degrees C. A scheme for increasing the performance of the designed alloy is proposed by discussing the strengthening mechanisms, machine learning process, and alloying chemistry effect. The cross-scale data-driven design is regarded as an accurate and efficient way to design novel high-strength Ni-based turbine disc superalloys, whose significance is the obvious reduction of trial-and-error tests. (c) 2023 Published by Elsevier Ltd on behalf of The editorial office of Journal of Materials Science & Technology.

Automated summary

The development of high-strength and stable high-temperature materials is essential to increase the thrust-weight ratio of aeroengines. A data-driven design approach was used to improve the yield strength of Ni-based turbine disc superalloys. Experimental tests showed that the designed alloy had satisfactory microstructure and higher yield strength compared to commercial superalloys.