Automated Reaction Mechanism Constructor

• Published in: Computer Aided Chemical Engineering Vol. 53; pp. 2803 - 2808
• Main Authors: de Carvalho Servia, Miguel Ángel, Kuok (Mimi) Hii, King, Hellgardt, Klaus, Zhang, Dongda, del Rio Chanona, Antonio
• Format: Book Chapter
• Language: English
• Published: 2024
• Subjects: automated reaction mechanism constructor; Chemical reaction engineering; Chemical reaction engineering; automated reaction mechanism constructor
• ISBN: 9780443288241; 0443288240
• ISSN: 1570-7946
• DOI: 10.1016/B978-0-443-28824-1.50468-3

In catalytic reaction engineering, the discovery of reaction mechanisms is paramount but challenging due to the intricate involvement of intermediates and limited prior knowledge. This study introduces a novel approach to generate the smallest feasible reaction mechanism (SFRM) that can accurately represent kinetic data sets. We propose an iterative algorithm based on a rule-based formulation, aiming to uncover the SFRM with minimal prior information about a system. Starting with the simplest conceivable mechanism, the algorithm advances by adding layers of complexity, estimating kinetic parameters, and evaluating mechanisms through the Akaike information criterion (AIC). Once a simpler mechanism demonstrates a smaller AIC value, it is selected as the optimal solution. Applied to the fructose to 5-hydroxymethylfurfural (HMF) mechanism, the methodology successfully uncovered the structure utilized for kinetic rate data generation. This novel framework offers experts a robust foundation for suggesting reaction mechanisms, addressing current limitations in mechanism discovery, emphasizing its utility not as a replacement but as a useful tool for experts.