Coarse-grained CFD-DEM simulation on the fast pyrolysis of large biomass particles: From intra-particle to reactor-scale behavior

Journal article, 2026

Constrained by the prevalent use of thermally-thin particle assumptions and the high computational overhead of the traditional discrete element model (DEM), existing single-scale modeling methods are inadequate for simulating biomass fast pyrolysis. In this work, a one-dimensional particle-scale model for directly resolving intra-particle heat transfer is first developed and subsequently coupled with a coarse-grained (CG) CFD-DEM. Within this multi-scale framework, the sand phase is modeled using a CG method, while biomass particles are tracked via DEM, and the fast pyrolysis of large biomass particles in a bubbling fluidized bed is comprehensively investigated. The results indicate that the predicted results of gas–solid dynamics and thermal behavior are satisfactory compared to experimental data, both at the single particle-scale and reactor-scale (cold and hot) validation. The product yields accounting for the intra-particle temperature gradient exhibit closer agreement with experimental data. Compared to neglecting the temperature gradients, the predicted tar yield has increased by approximately 5.71%. Furthermore, heat transfer at the particle-scale and the macroscopic gas–solid dynamics characteristics are accurately captured in pyrolysis. These evolution details are statistically and visually analyzed using probability density distribution and particle trajectory tracing, highlighting the accuracy and efficiency of the constructed multi-scale method. Future work will focus on predicting the pyrolysis behavior of large biomass particles with irregular shapes, thereby further broadening the applicability of this multi-scale method.