AI-driven identification of a novel malate structure from recycled lithium-ion batteries

The integration of Artificial Intelligence (AI) into the discovery of new materials offers significant potential for advancing sustainable technologies. This paper presents a novel approach leveraging AI-driven methodologies to identify a new malate structure derived from the treatment of spent lithium-ion batteries. By analysing bibliographic data and incorporating domain-specific knowledge, AI facilitated the identification and structure refinement of a new malate complex containing different metals (Ni, Mn, Co, and Cu). The synthesized compound was investigated through chemical and physical analyses, confirming its unique structure and composition. The present work proposes a significant difference from the classical use of AI in materials science, typically rooted in data-driven approaches relying on extensive datasets. This hybrid approach, combining AI's computational power with human expertise, not only expedited the structure determination process but also ensured the reliability and accuracy of the results. Finally, AI-driven material discovery highlights that waste materials can be transformed into valuable chemical products, suggesting their possible reuse, with several expected benefits, emphasising the role of AI in fostering not only innovation but also sustainability in material science.