A novel atomic mobility model for alloys under pressure and its application in high pressure heat treatment Al-Si alloys by integrating CALPHAD and machine learning

High pressure solution treatment, followed by ambient pressure aging treatment, may serve as a powerful tool for enhancing the alloy properties by tailoring plenty of nanoscale precipitates. However, no theoretical descriptions of the microstructure evolution and prediction of mechanical properties during high pressure heat treatment (HPHT) exist. In this work, a novel atomic mobility model for binary system under pressure was first developed in the framework of CALculation of PHAse Diagram (CALPHAD) approach and applied to assess the pressure-dependent atomic mobilities of (Al) phase in the Al-Si system. Then, quantitative simulation of particle dissolution and precipitation growth for HPHT Al-Si alloys was achieved through the CALPHAD tools by coupling the present pressure-dependent atomic mobilities together with previously established thermodynamic descriptions. Finally, the relationship among composition, process, microstructure, and properties was constructed by combining the CALPHAD and machine learning methods to predict the hardness values for HPHT Al-Si alloys over a wide range of compositions and processes with limited experimental data. This work contributes to realizing the quantitative simulation of microstructure evolution and accurate prediction of mechanical properties in HPHT alloys and illustrates pathways to accelerate the discovery of advanced alloys.