Construction of Formation Energy Dataset and Cluster Expansion for Au-Cu Alloys

This dataset mainly includes the formation energy dataset for real-space Cluster Expansion (CE), the training dataset for constituent strain energy, and the complete Effective Cluster Interaction (ECI) coefficients of the constituent strain energy corrected CE for the Au-Cu alloys. In this study, to construct the formation energy training dataset, the PBE+UDSCC method is employed to correct the energy levels of the Cu-3d and Au-5d orbitals, where the U and J parameters are self-consistently evaluated using the doubly screened Coulomb correction (DSCC) method. To construct the constituent strain energy training dataset, we use the PBE+UDSCC method to calculate the epitaxial strain energies of pure Au and Cu along different crystallographic orientations, and fit the constituent strain energy of Au-Cu alloys.

Based on these two datasets, a CE model corrected by constituent strain energy is established. Compared with the pure real-space CE model, both the fitting accuracy and the prediction accuracy of the convex hull are improved. The dataset mainly contains the following three files. File 1 contains the structural information and formation energy data of 241 conventional cells for CE (“Conventional Cell and Formation Energy of Au-Cu Alloys.csv”), corresponding to the PBE+UDSCC data in Figures 2, 8, and 12 of this article. File 2 contains the epitaxial strain energy data of pure elements for calculating constituent strain energy (“Epitaxial Strain Energy Dataset of Au.csv”, “Epitaxial Strain Energy Dataset of Cu.csv”) and the equilibrium constituent strain energy data (“Equilibrium Constituent Strain Energy Dataset of Au-Cu Alloys.csv”), corresponding to the data of the fitted curves in Figure 5 of this article. File 3 contains the complete ECIs of the strain-corrected CE (“All Effective Cluster Interaction (ECI) Coefficients of Strain-corrected CE.csv”), corresponding to the data in Figure 11(b) of this article.