CFD for the Modelling of Entrainment in Fluidised Bed Fast Pyrolysis of Biomass

Abstract

Bio-oil generated by fluidised bed fast pyrolysis of biomass holds considerable promise towards reducing dependence on fossil fuels. One problem stitl holding back progress is the efficient removal of char by entrainment from the reactor, which is impossible to model with traditional empirical correlations. A new model combining computational fluid dynamics (CFD), implemented in the commercial software package CFX, and particle entrainment theory has therefore been developed and found to give good qualitative agreement with experimental entrainment data. The CFD model uses a multiphase Eulerian-Eulerian technique for the bulk of the fluidised bed and individual particle tracking in the freeboard. Furthermore, the predictions of the CFD calculations for a number of important fluidised bed characteristics were checked, namely bubble diameter and velocity, terminal particle velocity and bed expansion. These were found to correspond closely to observations. Finally, several design modifications were tested on a laboratory cold flow model, which showed a significant improvement in char entrainment, and successfully modelled by computational simulation.