Unraveling dynamic analysis for ozonolysis pre-treatment of oil palm empty fruit bunch using sparse nonlinear optimizer and transport of diluted species interface

Ozonolysis pre-treatment offers an environmentally friendly alternative to improve cellulose accessibility for maximum utilization of lignocellulosic biomass. A diffusion-reaction model was developed for a well-mixed OzBiONY® reactor using COMSOL Multiphysics® software. The experimental work on ozonolysis pre-treatment under optimized operating conditions was conducted by measuring the changes in concentration of acid insoluble lignin (iLg), acid soluble lignin (sLg), and ozone versus time, to serve as input for kinetics analysis. Chemical Reaction Engineering Module was employed to create the numerical model for the ozonolysis reaction. The unknown kinetics parameters for the ozone assisted delignification reaction were evaluated using Parameter Estimation tools through Sparse Nonlinear OPTimizer (SNOPT) solver as it allows the incorporation of various objective functions and constraints. The effects of initial guess, lower boundary, and upper boundary values for unknown parameters on the simulation model were studied. Based on the simulation results, the estimated values for unknown kinetics parameters were successfully obtained with R² value greater than 0.75. Plausible pathway for delignification reactions during ozonolysis pre-treatment were also elucidated by the current diffusion-reaction model. The diffusion-reaction model was later coupled with Transport of Diluted Species (tds) interface to investigate the dynamic behavior of the species involved within an EFB biomass particle. This work establishes a framework for determination of kinetic parameters of ozonolysis pre-treatment and its potential applications in industrial scale biorefinery.