Abiotic Reduction of Organic and Inorganic Compounds by Fe(II)-Associated Reductants: Comprehensive Data Sets and Machine Learning Modeling

摘要

The research constructed a comprehensive data set and developedmachine learning models to predict abiotic reduction rate constantsin iron-associated reductive transformations. @@@ Iron-associated reductants play acrucial role in providing electronsfor various reductive transformations. However, developing reliablepredictive tools for estimating abiotic reduction rate constants (logk) in such systems has been impeded by the intricate natureof these systems. Our recent study developed a machine learning (ML)model based on 60 organic compounds toward one soluble Fe(II)-reductant.In this study, we built a comprehensive kinetic data set coveringthe reactivity of 117 organic and 10 inorganic compounds toward fourmajor types of Fe(II)-associated reductants. Separate ML models weredeveloped for organic and inorganic compounds, and the feature importanceanalysis demonstrated the significance of resonance structures, reduciblefunctional groups, reductant descriptors, and pH in logk prediction. Mechanistic interpretation validated that the modelsaccurately learned the impact of various factors such as aromaticsubstituents, complexation, bond dissociation energy, reduction potential,LUMO energy, and dominant reductant species. Finally, we found that38% of the 850,000 compounds in the Distributed Structure-SearchableToxicity (DSSTox) database contain at least one reducible functionalgroup, and the logk of 285,184 compounds could bereasonably predicted using our model. Overall, the study is a significantstep toward reliable predictive tools for anticipating abiotic reductionrate constants in iron-associated reductant systems.

关键词