Data-driven prediction of grain boundary segregation and disordering in high-entropy alloys in a 5D space

Grain boundaries (GBs) can critically influence the microstructural evolution and various material properties. However, a fundamental understanding of GBs in high-entropy alloys (HEAs) is lacking because of the complex couplings of the segregations of multiple elements and interfacial disordering, which can generate new phenomena and challenge the classical theories. Here, by combining large-scale atomistic simulations and machine learning models, we demonstrate the feasibility of predicting the GB properties as functions of four independent compositional degrees of freedom and temperature in a 5D space, thereby enabling the construction of GB diagrams for quinary HEAs. The artificial neural network (ANN), support vector machine (SVM), regression tree, and rational quadratic Gaussian models are trained and tested, and the ANN model yields the best machine learning based predictions. A data-driven discovery further reveals new coupled segregation and disordering effects in HEAs. For instance, interfacial disordering can enhance the co-segregation of Cr and Mn at CrMnFeCoNi GBs. A physics-informed data-driven model is constructed to provide more physical insights and better transferability. Density functional theory (DFT) calculations are used to validate the prediction generality and reveal the underlying segregation mechanisms. This study not only provides a new paradigm enabling the prediction of GB properties in a 5D space, but also uncovers new GB segregation phenomena in HEAs beyond the classical GB segregation models.