Leveraging AI/ML-driven Tools for Material Discovery and Critical Mineral Separations †

Read full text Summarise with VIRAT
Authors: Kunal Mondal, Christine E. Duval, Rigoberto C. Advíncula
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Critical materials and rare earth elements (CM/REEs) are essential components for modern energy technologies, communication systems, security, and defense applications. However, CM/REE recovery from complex mining wastes remains a challenge. Currently used processes, dominated by solvent extraction and precipitation, are not well suited to complex environmental feedstocks such as coal fly ash, phosphogypsum, and metal mine tailings. Although there is intense interest in extracting CM/REEs via adsorptive processes due to their attractive selectivity and low solvent usage, the discovery of novel ligands with well-defined structure-property correlations and fast cycling rates remains challenging, particularly through scientific experimental design. In this Perspective, we discuss the significant potential of artificial intelligence/machine learning (AI/ML) and self-driving laboratories (SDLs) for the design of selective and mass-transferefficient adsorbents for the sustainable separation of CM/REEs from complex mining wastes. 2 Some important examples of the application of AI/ML include accelerating calculations in quantum chemistry, molecular dynamics, and discovery science, for which SDLs can be harnessed to facilitate the rapid validation, optimization, and scale-up of material systems. In principle, the combination of AI-based material discovery, continuous-flow processing, analytics, and reinforcement learning is expected to yield novel material-processing methods for CM/REE separation.

Fields of Study & Subjects

Machine Learning in Materials Science Extraction and Separation Processes Radioactive element chemistry and processing

Publication details

Journal / source
ChemRxiv
Publication date
24 March 2026
Language
EN
Document type
Preprint