Chemistry

Machine learning–driven predictive modeling and process optimization of one-pot biomass conversion to FDCA via heterogeneous catalysis

AI Insight

This study presents a machine learning framework designed to predict and optimize the one-pot catalytic conversion of biomass-derived compounds into 2,5-furandicarboxylic acid (FDCA), a key monomer for bio-based polymers. By integrating heterogeneous catalysis data with predictive models, the researchers identified optimal reaction conditions including temperature, catalyst composition, and oxidant parameters to maximize FDCA yield. The approach demonstrates that data-driven modeling can effectively navigate the complex, multivariable space of biomass conversion reactions.


FDCA is a renewable alternative to terephthalic acid used in plastic production, and improving its synthesis efficiency could accelerate the transition toward bio-based, potentially more sustainable polymer industries. This work provides a replicable computational methodology that could be applied to other catalytic biomass valorization processes.


Source: Machine learning–driven predictive modeling and process optimization of one-pot biomass conversion to FDCA via heterogeneous catalysis