CalPhaD Coffee Lecture – High-Throughput CALPHAD for Sustainable Materials Design in the Age of AI

The transition toward sustainable materials demands faster and more reliable design strategies capable of balancing performance, resource efficiency, and environmental impact. High-throughput CALPHAD calculations, when integrated with AI-based methods, offer a promising pathway to meet these challenges.

This lecture presents an integrated workflow that combines HT-CALPHAD data generation with automated data processing, machine learning property models, and multi-objective optimization. By embedding thermodynamic consistency into data-driven pipelines, HT-CALPHAD enables efficient pre-screening of candidate materials, supports informed design decisions, and enhances the reliability of AI-driven predictions.

Case studies demonstrate how this combined approach accelerates sustainable materials design by reducing experimental and computational effort while maintaining physical interpretability. The talk concludes with perspectives on how HT-CALPHAD and AI together can shape future materials discovery frameworks focused on sustainability.