Development and integration of calphad databases and software tools for alloy design

Efforts at LLNL regarding CALPHAD database development leveraging high-throughput DFT calculations, semi-empirical thermodynamic models and dedicated experiments, and benefiting from the latest developments in the pycalphad (https://pycalphad.org) and ESPEI (https://espei.org) open-source software packages will be presented and discussed for different classes of alloys including rare-earth elements containing and refractory alloys. Then, the LLNL’s Materials Acceleration Platform (MAP) running on high-performance computers and handling CALPHAD-based, analytical and machine-learning models will be introduced to tackle the challenging task of screening and optimizing new alloys for multiple properties in a vast composition space. Among others, we will highly the design of new refractory alloys targeting high yield strength and ductility at elevated temperatures with constraints on phase stability in a 10-element composition space. Finally, alloy design case studies using the Alloy Optimization Software (TAOS, https://softwarelicensing.llnl.gov/product/taos) – which is running on stand-alone computers and solving unconstrained and constrained optimization problems of extremely complex and non-smooth functions – will be presented with specific examples for structural and functional materials design.

This work was performed under the auspices of the U.S. DOE by LLNL under contract DE-AC52-07NA27344 and supported by LLNL-LDRD program under Project No. 22-SI-007 and the Critical Materials Innovation Hub, an Energy Innovation Hub funded by the U.S. Department of Energy, Office of Energy Efficiency and Renewable Energy, Advanced Manufacturing and Materials Technologies Office (AMMTO).