cardiGAN: A Generative Adversarial Network Model for Design and Discovery of Multi Principal Element Alloys

TL;DR

This paper introduces cardiGAN, a GAN-based model that accelerates the discovery and design of multi-principal element alloys by generating new compositions and predicting their phases, thus reducing costly experimental efforts.

Contribution

It presents a novel GAN-based framework combining generative and discriminative neural networks for efficient MPEA exploration and phase prediction, advancing materials discovery methods.

Findings

Successfully generated novel MPEA compositions.

Predicted phases of generated alloys with high accuracy.

Validated model with experimental alloy design.

Abstract

Multi-principal element alloys (MPEAs), inclusive of high entropy alloys (HEAs), continue to attract significant research attention owing to their potentially desirable properties. Although MPEAs remain under extensive research, traditional (i.e. empirical) alloy production and testing is both costly and time-consuming, partly due to the inefficiency of the early discovery process which involves experiments on a large number of alloy compositions. It is intuitive to apply machine learning in the discovery of this novel class of materials, of which only a small number of potential alloys has been probed to date. In this work, a proof-of-concept is proposed, combining generative adversarial networks (GANs) with discriminative neural networks (NNs), to accelerate the exploration of novel MPEAs. By applying the GAN model herein, it was possible to directly generate novel compositions for MPEAs, and to predict their phases. To verify the predictability of the model, alloys designed by the model are presented and a candidate produced; as validation. This suggests that the model herein offers an approach that can significantly enhance the capacity and efficiency of development of novel MPEAs.