(154f) Simulating Biomass Fast Pyrolysis In Fluidized-Bed Reactors for Bio-Oil Production

Xue, Q. - Presenter, Iowa State University
Fox, R. O. - Presenter, Iowa State University

A numerical study is conducted to investigate optimal operating conditions of biomass pyrolysis in a fluidized-bed reactor for condensable bio-oil (tar) yield. A MFIX CFD model is presented for biomass fast pyrolysis in a fluidized-bed reactor via coupling the multifluid model with a global biomass particle pyrolysis model. The multifluid model is derived from the kinetic theory of granular flows. The kinetic model is based on superimposed cellulose, hemicellulose, and lignin reactions for micro-particles (< 1000 μm). The pyrolysis products are categorized into three groups: tar vapor, producer gas, and char. The CFD model uses 8 solid species to differentiate general biomass feedstocks by species mass fractions with 2 additional gaseous species released into the gas phase. The interaction between the biomass pyrolysis and reactor environment is represented through mass, heat, and momentum exchange terms. The model is used to investigate the effect of various operating parameters on the tar yield in a fluidized-bed reactor. The model simulates sufficient physical time (200 secs) to reach a steady-state condition. Results indicate that at a fixed temperature and superficial gas velocity, the biomass particle size determines the particle residence time inside the bed, the heating and reacting processes, and ultimately influences the tar yield. At a fixed particle size, the temperature is the prominent parameter that influences tar yield. For the range of studied fluidized gas temperatures, the predicted optimum tar yield at steady state is achieved near 450oC. The product yields for different biomass feedstocks show a material with more cellulose mass fraction favors tar yield and more lignin mass fraction favors char production. It is also found that shallow fluidized beds and higher fluidized gas velocities have a positive effect on tar yield due to short tar vapor residence time in the reactor.