CrossMat: Integrating Material Databases for Multi-Domain Cross-Application

Big data and artificial intelligence (AI) have emerged as a transformative force in materials science. However, the field still faces significant challenges, primarily the scarcity of comprehensive material datasets and the inefficiencies in utilizing existing data to its full potential. In this work, we propose the CrossMat platform, integrating material databases to accelerate material discovery across diverse applications. Currently encompassing fields such as electrocatalysis, thermocatalysis, photocatalysis, solid-state electrolyte materials, hydrogen storage materials, lithium battery electrolytes, thermoelectric materials, and superconducting materials, CrossMat is driven by large language models and machine learning algorithms to significantly broaden the scope of material prediction. By systematically identifying synthesizable, cost-effective, and environmentally stable materials, CrossMat facilitates their adaptation to previously unexplored domains. This approach not only addresses critical limitations in current methodologies but also opens up innovative avenues for the discovery and development of advanced materials.