Alloying effects and effective alloy design of high-Cr CoNi-based superalloys via a high-throughput experiments and machine learning framework

Abstract

The discovery of a stable γ/γ' microstructure in Co-Al-W-based alloys brought forth possibilities for Co-based superalloys as potential alternatives to Ni-based superalloys in high temperature applications. To date, only a few γ′-strengthened Co-based superalloys with good comprehensive performance have been reported due to the limited knowledge of alloying effects, mainly limited by the “trial and error” approach. In the present work, an accelerated framework integrating a multicomponent diffusion-multiple and machine learning was developed to efficiently and systematically study the effects of alloying elements, e.g. Ni, Al, W, Ti, Ta, Cr, Mo and Nb on the microstructures including phase equilibrium, γ′ volume fraction, γ′ size and γ′ morphology of multicomponent CoNi-based superalloys with high Cr content. The multicomponent diffusion-multiple enabled the collection of a large amount of experimental data relating composition and microstructural parameters, while machine learning enabled the exploration of alloying effects on the microstructural stability and parameters in the compositional space. To validate the efficacy of this approach, two multicomponent CoNi-based superalloys with high Cr addition were designed to possess a γ/γ' two-phase microstructure of relatively high γ' volume fraction, low γ' coarsening rate, with one having a spherical and the other cuboidal γ' morphology. This framework and the insight obtained in this study will be helpful to accelerate the development and application of multicomponent CoNi-based superalloys with high Cr content for engineering applications.