Spectral model for grain boundary segregation in systems with strong solute–solute interactions

Solute segregation at grain boundaries is emerging as a promising tool for the stabilization of polycrystals against grain growth. Modern approaches to segregation modeling are based on employing atomistic grain boundary segregation spectra. Currently, there is no spectral model that accounts for solute–solute interactions beyond the random mixing approximation. The paper shows that such interactions are crucial for the Ag(Ni) system, which is the focus of this study. We derive the Gibbs free energy for a spectral model of grain boundary segregation that properly captures solute–solute interactions strong enough to violate the random mixing assumption. Strong attractive interactions have been shown to be independent of solute concentration and can be directly incorporated into the segregation spectrum. Moreover, at zero temperature, all types of solute–solute interactions can be incorporated into the segregation spectrum, allowing the calculation of zero-temperature stability scores with regard to solute–solute interactions. This may enhance the reliability of screening for stable nanocrystalline alloys. Finally, the model incorporating solute–solute interactions into the segregation spectrum has been shown to provide a reliable approximation for systems exhibiting strong solute–solute interactions, even at moderate temperatures.