Semi-mechanistic modelling of ionic liquid-based biomass fractionation

Fractionation of lignocellulosic biomass is a crucial step to provide cellulose, lignin, and hemicellulose for further processing. This work focuses on modelling the fractionation of woody biomass using the ionoSolv process, which employs low-cost ionic liquid water mixtures. We model a simple reaction network to describe the solvent-extraction of three main biopolymers from solid lignocellulosic biomass. We estimate the corresponding kinetic parameters and their credibility intervals using Bayesian parameter estimation and then exploit the calibrated model for a multi-criterion analysis employing three process metrics: glucan (cellulose) recovery, hemicellulose removal, and lignin removal (delignification). Specifically, we construct a probabilistic design space by propagating the model parameter uncertainty, with a view to predicting a feasible operating window for key process variables (pretreatment time, temperature and solids loading) to meet certain thresholds for each metric. Overall, the development of semi-mechanistic models provides a novel framework for the analysis and optimisation of ionic liquid-based biomass pretreatment.