Numerical multi-criteria optimization methods for alloy design. Development of new high strength nickel-based superalloys and experimental validation

A new approach for the design of optimum balanced metallic alloys is presented. It is based on a mathematical multi-criteria optimization method which uses different property models to predict the alloy behavior in dependency of composition. These property models are mostly based on computational thermodynamics (CALPHAD-method). The full composition range of the alloying elements can be considered using these models. In alloy design usually several contradicting goals have to be fulfilled. This is handled by the calculation of so-called Pareto-fronts. The aim of our approach is to guide the experimental research towards new alloy compositions that have a high probability of having very good properties. Consequently the number of required test alloys can be massively reduced. The approach will be demonstrated for the computer-aided design of a new Re-free superalloy with nearly identical creep strength as that of Re-containing superalloys. Our starting point for the design was to maintain the good properties of the gamma prime-phase in well-known alloys like CMSX-4 and to maximize the solid solution strengthening of W and Mo. The presented experimental measurements proof the excellent properties.