Computational Design of Thermally Stable Nanoprecipitates in Al‐Zn‐Mg Alloys: Insights From High‐Throughput DFT and Machine Learning

ABSTRACT The exceptional strength of 7000 series (Al‐Zn‐Mg) aluminum alloys originates from nanoscale, metastable η′ precipitates. However, the transformation of these precipitates into the stable η phase at elevated temperatures significantly degrades the alloys’ mechanical properties. To address this, we combine high‐throughput density functional theory (DFT) calculations with machine learning (ML) to screen 21 transition metals for their potential to enhance η′ stability via interfacial segregation. Our calculations identify Mn, Fe, Co, Ni, Cu, Ag, Pt, and Au as feasible segregants. The ML analysis reveals that segregation is primarily driven by minimizing elastic strain energy at the precipitate‐matrix interface. Smaller atoms—such as Cu, Mn, Ni, and Co—are attracted to the interface to relieve compressive strain, thus enhancing stability. Secondary chemical effects, governed by electronegativity and electron affinity, explain why larger atoms like Ag, Pt, and Au are also viable segregants. Considering both thermodynamic tendency and solute availability, we identify Ni as a uniquely potent, yet previously overlooked stabilizer. Experimental validation by Differential Scanning Calorimetry (DSC) confirmed that a Ni‐added alloy increases the transition temperature by ∼30°C over a conventional Cu‐added (AA7075) alloy, providing a robust framework for designing next‐generation high‐temperature aluminum alloys.