Biomass pyrolysis is a promising renewable source instead of fossil fuels. In this study, a multiscale computational fluid dynamics (CFD) model for high-temperature biomass pyrolysis was developed in open-source MFiX with an intraparticle model and a detailed multistep kinetic scheme. The intraparticle transport phenomena were simulated with a one-dimensional particle model. The pyrolysis kinetics was built up with a detailed multistep kinetic scheme of 32 reactions and 58 species and a secondary cracking scheme consisting of 22 reactions. The effects of intraparticle transport phenomena and secondary tar cracking were investigated by the following four different models: isothermal model, isothermal model with tar cracking, nonisothermal model, and nonisothermal model with tar cracking. It was found that the nonisothermal model with tar cracking performs the best for both low- and high-temperature biomass pyrolysis by comparing the experimental data. The model was then applied to study the effects of particle size, inflow rate of external heat, and operation temperature on biomass pyrolysis. A machine learning-based surrogate model was built to predict the pyrolysis product distributions.
|Original language||American English|
|Journal||Industrial and Engineering Chemistry Research|
|State||Published - 4 Nov 2022|