High-throughput computational screening of Aluminum alloys from scraps

Abstract

To meet societal goals like climate neutrality by 2050 and a circular economy, material discovery must be faster and more flexible, especially as raw material costs and supply chain issues increase. Advanced technologies like computational screening, 3D/4D characterization, and machine learning enable rapid responses to these challenges. A new Al-based alloy suitable for additive manufacturing (AM) derived from scrap metal mixtures was developed using high-throughput alloy screening paired with experimental validation. The new alloy is intended for aerospace applications and has to meet requirements such as low sensitivity to hot-cracking, appropriate strength and elongation and corrosion resistance. In high-throughput computational screening, up to 10k alloy compositions were generated by mixing the scrap alloys at different ratios. The CALPHAD method was used for equilibrium and non-equilibrium simulations to assess key parameters, including phases, solidification intervals, and mechanical properties like yield strength and hot-cracking sensitivity. A random forest model predicted properties from simulations and literature, while a multi-objective evolutionary algorithm filtered alloys with target characteristics. Uncertainty calculations addressed compositional variations in scrap, ensuring robust designs. Experimental methods, including SEM/EDX, laser flash analysis, DSC, synchrotron techniques, and laser track experiments, validated the simulations. These tests helped narrow down 1-2 optimal scrap mixtures for powder production and laser powder bed fusion (LPBF) processing.