Reac-Discovery is a dataset for AI-driven flow reactor design and reaction performance optimization released by Jaume I University in 2025. The related paper results are "Reac-Discovery: an artificial intelligence–driven platform for continuous-flow catalytic reactor discovery and optimization".

This dataset is generated automatically during the experiment using the team's proprietary Reac-Discovery platform, without using any external public data sources. It covers three categories of data: geometry, printability, and reaction performance, corresponding to the platform's Reac-Gen, Reac-Fab, and Reac-Eval modules:

• Structural parameterization dataset (Reac-Gen): Generates periodic open cell structures (POCs) through mathematical parameterization models and records geometric descriptors such as size, hierarchy, surface area, free volume, and tortuosity;
• Printability Dataset (Reac-Fab): Based on 3D printing experiments, it establishes the correspondence between design parameters and printing accuracy and completeness;
• Reaction Performance Dataset (Reac-Eval): Experiments were performed in an automated flow reaction system using a self-propelled laboratory platform, with real-time recording of reaction parameters such as temperature, flow rate, concentration, and yield.

All data is standardized and structured and stored in XLSX and STL files.