EUBCE 2026 - Conall MCNAMARA - Alkyl Levulinate Production from Lignocellulose with Mechanistic Learning by Hierarchical Surrogate Kinetic Modelling

Alkyl Levulinate Production from Lignocellulose with Mechanistic Learning by Hierarchical Surrogate Kinetic Modelling

Short Introductive summary

Advanced biofuels derived from lignocellulosic biomass are central to accelerating near term decarbonisation in transport, particularly where electrification faces technical barriers. Alcoholysis offers a promising one-pot route to convert heterogeneous biomass into levulinate-based fuel blends, yet progress is constrained by inconsistent yield reporting, limited mass balance closure, and an incomplete understanding of biomass decomposition under acidic conditions. This work addresses these challenges by establishing steady state conditions for alcoholysis of glucose, cellulose, xylan, and real biomass, revealing that both cellulose and hemicellulose contribute directly to alkyl levulinate formation. Comprehensive carbon accounting quantifies all major co-products, including dialkyl ethers, alkyl formates, and alkyl acetates, clarifying the true composition of the fuel blend. A hierarchical surrogate kinetic model, built on molecular group additivity principles and validated from monomers to whole biomass, links biochemical composition to predicted fuel yields, providing a predictive basis for developing efficient and scalable alcoholysis routes for sustainable biofuel production.

C.. Mcnamara, Trinity College Dublin, IRELAND
S. Dooley, Trinity College Dublin, IRELAND
A. O'Shea, Trinity College Dublin, IRELAND