The control of solute segregation at grain boundaries is of significance in engineering alloy properties. However, there is currently a lack of a physics-informed predictive model for estimating solute segregation energies. Here we propose novel electronic descriptors for grain-boundary segregation based on the valence, electronegativity and size of solutes. By integrating the non-local coordination number of surfaces, we build a predictive analytic framework for evaluating the segregation energies across various solutes, grain-boundary structures, and segregation sites. This framework uncovers not only the coupling rule of solutes and matrices, but also the origin of solute-segregation determinants, which stems from the d- and sp-states hybridization in alloying. Our scheme establishes a novel picture for grain-boundary segregation and provides a useful tool for the design of advanced alloys.

中文翻译：

---

用于确定溶质在晶界偏析中的影响的通用描述符


控制晶界处的溶质偏析在工程合金性能中具有重要意义。然而，目前缺乏用于估计溶质偏析能的物理学预测模型。在这里，我们提出了基于溶质的价、电负性和大小的晶界偏析的新型电子描述符。通过整合表面的非局部配位数，我们建立了一个预测分析框架，用于评估各种溶质、晶界结构和偏析位点的偏析能。该框架不仅揭示了溶质和基质的耦合规则，还揭示了溶质偏析行列式的起源，它源于合金化中的 d 态和 sp 态杂化。我们的方案为晶界偏析建立了一种新的图景，并为先进合金的设计提供了有用的工具。