Reac-Discovery: an artificial intelligence–driven platform for continuous-flow catalytic reactor discovery and optimization
Cristopher Tinajero, Marcileia Zanatta, Julián E. Sánchez‐Velandia, Eduardo García‐Verdugo, Víctor Sans
Abstract
Digital technologies, including artificial intelligence and additive manufacturing, have revolutionized chemistry and chemical engineering. In reactor engineering, performance improvements have been enabled by novel geometries, yet design approaches have traditionally relied on human input. This study introduces Reac-Discovery, a digital platform that integrates catalytic reactor design, fabrication, and optimization based on periodic open-cell structures (POCs). It combines the parametric design and analysis of advanced structures from mathematic models (Reac-Gen), high-resolution 3D printing and functionalization of catalytic reactors (Reac-Fab) with an algorithm validating the printability of reactor designs and a self-driving laboratory (Reac-Eval), capable of parallel multi-reactor evaluations featuring real-time nuclear magnetic resonance (NMR) monitoring and machine learning (ML) optimization of process parameters and topological descriptors. Two multiphase catalytic reactions-the hydrogenation of acetophenone and the CO₂ cycloaddition-were selected as case studies, where Reac-Discovery achieved the highest reported space-time yield (STY) for a triphasic CO₂ cycloaddition using immobilized catalysts.