Author(s) |
Austin M. Hernandez, Sona Avetian, Sharmila Karumuri, Zachary McClure, Logan Ware, Alejandro Strachan, Ilias Bilionis, Kenneth Sandhage, Michael Titus |
Abstract Scope |
Refractory complex concentrated alloys (RCCAs) are promising materials for replacing nickel-based superalloys in turbine engines, as RCCAs exhibit an attractive combination of properties, including high temperature strength, ductility, and phase stability. However, RCCAs are often prone to catastrophic oxidation at intermediate and high temperatures. To design high strength and oxidation resistant alloys, we begin with a design space of 9 refractory elements (Ti, Zr, Hf, V, Nb, Ta, Cr, Mo, W), along with aluminum, encompassing millions of possible compositions. We first evaluate these compositions using CALPHAD methods and utilize an active learning loop, coupled with machine learning models, to systematically interrogate the oxidation and hardness of alloys generated from down-selected CALPHAD calculations along a Pareto front. To allow for high-throughput experimentation, diffusion couples of Al-containing alloys along the pareto front with Al-free compositions have been examined, and the hardness and oxidation behavior along the resulting concentration gradients will be discussed. |