Machine Learning Enabled Prediction of Stacking Fault Energies in Concentrated Alloys

Abstract

Recent works have revealed a unique combination of high strength and high ductility in certain compositions of high-entropy alloys (HEAs), which is attributed to the low stacking fault energy (SFE). While atomistic calculations have been successful in predicting the SFE of pure metals, large variations up to 200 mJ/m2 have been observed in HEAs. One of the leading causes of such variations is the limited number of atoms that can be modeled in atomistic calculations; as a result, due to random distribution of elements in HEAs, various nearest neighbor environments may not be adequately captured in small supercells resulting in different SFE values. Such variation further increases with the increase in the number of elements in a given composition. In this work, we use machine learning to overcome the limitation of smaller system sizes and provide a methodology to significantly reduce the variation and uncertainty in predicting SFEs. We show that the SFE can be accurately predicted across the composition ranges in binary alloys. This capability then enables us to predict the SFE of multi-elemental alloys by training the model using only binary alloys. Consequently, SFEs of complex alloys can be predicted using a binary alloys database, and the need to perform calculations for every new composition can be circumvented.

Accelerated prediction of stacking fault energy in FCC medium entropy alloys using multilayer perceptron neural networks: correlation and feature analysis

“…(The SFE estimation using the microstrain method is attached in supplementary information I, section S5). The table shows that the predicted SFE is similar to the reported SFE values calculated from experimentation and atomistic calculations [5,29,30,[55][56][57][58]. The error percentage is below 25% showing a good resemblance with the reported value and acceptable for the SFE prediction.…”

“…A few approaches have strived to overcome this limitation using data obtained from atomistic calculations. Arora and Aidhy [29] reportedly built a framework for the prediction of SFE for multi-elemental alloys, trusting on the inputs of binary alloys i.e. Ni-Fe, Fe-Cr, and Cr-Ni data obtained via atomistic calculations.…”

Machine Learning Based Methodology to Predict Point Defect Energies in Multi-Principal Element Alloys

“…Similarly, Sharma et al (2020) used ML based framework to predict substitutional defect formation energies in ABO 3 perovskites. Other materials properties such as vibrational entropy (Manzoor and Aidhy, 2020) and stacking fault energies (Arora and Aidhy, 2020) have also been recently predicted using a combination of ML and atomistic calculations. Thus, application of ML models in materials science is rapidly becoming mainstream that is being used not only to bypass the computational expense but also to predict new properties.…”

Correlation analysis of materials properties by machine learning: illustrated with stacking fault energy from first-principles calculations in dilute fcc-based alloys

“…Vilalta et al [39] examined the relationship between yield stress and the γ SFE landscape in high entropy alloys by means of a ML approach. Arora and Aidhy [40] indicated that γ SFE in concentrated multi-elemental alloys can be predicted using a ML based framework. However, the ML based study of γ SFE affected by dilute alloying elements is not available in the literature.…”