Influence of computational parcel resolution on hydrodynamics and performance of reacting fluidized bed reactors

Abstract

The Eulerian-Lagrangian computational models are critical for understanding gas-phase processes and addressing engineering challenges, yet their numerical accuracy in reactive gas-solid fluidized beds remains a concern. This work utilizes the multiphase particle-in-cell (MP-PIC) scheme to assess the impact of parcel number on the hydrodynamics and reaction predictions in the fluid catalytic cracking (FCC) catalyst regeneration process. By employing the energy-minimization multi-scale model for bubbling and turbulent flow regimes, the MP-PIC simulations closely match experimental solids concentration profiles. We developed and analyzed different parcel resolutions: 9.5×10³, 2.0×10⁴, 1.0×10⁵, 2.2×10⁵, 1.1×10⁶ in bubbling fluidized beds (BFB), and 6.5×10³, 1.3×10⁴, 7.1×10⁴, 1.4×10⁵, 6.8×10⁵ in turbulent fluidized beds (TFB). The findings reveal a significant sensitivity of coke combustion efficiency, flue gas evolution, and temperature to parcel resolution in BFB, attributable to mesoscale activities, with less impact observed in TFB. The study highlights the essential balance between the accuracy of Lagrangian particle configuration and computational costs.