| Abstract Scope |
Grain boundary (GB) segregation governs intergranular cohesion, oxidation resistance, and embrittlement in high-entropy alloys, yet no analytic predictor exists for the multicomponent composition space these alloys occupy. We close this gap by extending the macroscopic atom model to GBs through a fractional surface-character description, generalizing the resulting expression to multicomponent matrices via surface-fraction averaging of contact enthalpies, and fusing it with a strained-coordination size term. Calibrated once against 240 binary molecular-dynamics pairs, the three-term model predicts segregation across any HEA composition from tabulated parameters. Gibbs adsorption emerges as the dominant driver, cross-validated classification reaches 75% on unseen matrices, and the predicted spectral width independently recovers the slope reported in literature. An application to nine CoCrFeMnNiAl subfamilies reveals Mg as the most dominant solute in GB segregation. |