D-Epox-Cat-ML: Digital Catalyst Dataset for Cyclohexene Epoxidation – Enabling AI through Literature-Based Reaction Descriptors and Yields

This dataset, named D-Epox-Cat-ML: Digital Catalyst Dataset for Cyclohexene Epoxidation – Enabling AI through Literature-Based Reaction Descriptors and Yields, compiles experimental data on cyclohexene epoxidation reactions catalyzed by transition metal complexes. The dataset was extracted from peer-reviewed scientific publications and contains curated and normalized descriptors, including metal identity, oxidation state, catalyst type, ligand type, oxidant, solvent, temperature, time, and quantitative reaction outcomes (yield, conversion, selectivity, TON, and TOF). This structured dataset enables the training and evaluation of supervised machine learning models for predicting catalytic performance. The compilation aims to support AI-driven catalyst design and facilitate reproducibility in digital catalysis research.