Meta-Database of MOF Synthesis–Structure–Property Correlations for Data-Driven Anomaly Mining

Gubsch et al. (2022) and Tan et al. (2023) highlight the need for better datasets, but the field still lacks a truly integrative, open-access resource that collates synthesis recipes, structural characterizations (including defects, disorder, dynamics), and measured properties (adsorption, conductivity, etc.) across thousands of MOFs. This idea proposes not just collecting such data (including from text mining of the literature) but using advanced anomaly detection algorithms to flag cases where properties deviate sharply from expectations. By systematically investigating these "data outliers," researchers can identify overlooked mechanisms (e.g., defect-induced conductivity, unexpected flexibility, synergistic effects in core-shell structures) and iteratively refine both experimental and theoretical models. This meta-database would be an engine for hypothesis generation and could catalyze a new wave of serendipitous discoveries in MOFs and reticular chemistry.

References:

1. One-Pot Synthesis of MOF@MOF: Structural Incompatibility Leads to Core-Shell Structure and Adaptability Control Makes the Sequence.. Hao Tan, Xiang Zhao, Liting Du, Bufeng Wang, Yongliang Huang, Yupeng Gu, Zhiyong Lu (2023). Small.
2. DigiMOF: A Database of MOF Synthesis Information Generated via Text Mining. Kristian Gubsch, Rosalee Bence, Lawson T. Glasby, Peyman Z. Moghadam (2022).