Computational models for the prediction of yields in the autothermal pyrolysis of biomass

Abstract

Autothermal pyrolysis of biomass, conducted in fluidized bed reactors, addresses the heat transfer challenges of conventional fast pyrolysis by injecting a small amount of oxygen to allow for partial oxidation of pyrolysis products. This article presents two computational models for predicting yields in autothermal pyrolysis using a comprehensive chemical kinetic mechanism for devolatilization, char combustion, and secondary gas-phase reactions. Initially, the reaction mechanism is studied in stages using a homogeneous model in OpenFOAM®, excluding the fluidized bed hydrodynamics. The model performs well to estimate yields based on biomass feedstock and operating conditions. An Euler-Euler multiphase model is then used to describe the fluidized bed hydrodynamics coupled with the kinetic model. Early-stage simulations are compared to experimental data, showing that secondary gas-phase reactions can be omitted in lab-scale devices. However, homogenous models show that such reactions are useful when considering longer residence times typical of plant-scale devices.