Grain boundary segregation prediction with a dual-solute model

Solute segregation along grain boundaries (GBs) profoundly affects their thermodynamic and kinetic behavior in polycrystalline materials. Recently, it has become a promising strategy for alloy design, mitigating grain growth by reducing excess GB energy and strengthening the GB network in nanocrystalline metals. In this context, the spectrum approach has emerged as a powerful tool to predict GB segregation. However, previous GB segregation predictions using this method relied heavily on single-solute segregation spectra, neglecting the crucial role of solute-solute interactions, which are often incorporated through a fitting parameter. In this work, we developed a dual-solute model whose segregation energy spectrum intrinsically considers the solute-solute interactions. Further improvement was made by describing the volume fraction of GBs as a varying parameter that scales with the total solute concentration and temperature. The refined dual-solute model was attempted to predict the GB segregation at finite temperatures in several binary systems. It shows significant improvement over the single-solute model and can accurately predict the hybrid Molecular Dynamics/Monte Carlo data within a broad temperature range with varying solute concentrations before forming secondary phases. This dual-solute model provides an effective way to statistically predict GB segregation with considerable accuracy in nanocrystalline metals.