Modeling hydrodynamics of gas–liquid airlift reactor

Authors

  • Samuel Talvy,

    1. Laboratoire d'Ingénierie des Systèmes Biologiques et des Procédés (UMR INSA-CNRS 5504, UMR INSA-INRA 792), 31077 Toulouse cedex, France
    Search for more papers by this author
  • Arnaud Cockx,

    1. Laboratoire d'Ingénierie des Systèmes Biologiques et des Procédés (UMR INSA-CNRS 5504, UMR INSA-INRA 792), 31077 Toulouse cedex, France
    Search for more papers by this author
  • Alain Liné

    Corresponding author
    1. Laboratoire d'Ingénierie des Systèmes Biologiques et des Procédés (UMR INSA-CNRS 5504, UMR INSA-INRA 792), 31077 Toulouse cedex, France
    • Laboratoire d'Ingénierie des Systèmes Biologiques et des Procédés (UMR INSA-CNRS 5504, UMR INSA-INRA 792), 31077 Toulouse cedex, France
    Search for more papers by this author

Abstract

This article deals with the physical modeling and numerical simulation of two-phase bubbly flow in an airlift internal loop reactor. The objective is to show the ability of computational fluid dynamics (CFD) to correctly simulate hydrodynamics and axial dispersion in such a bubbly reactor. The modeling of two-phase bubbly flow is based on the so-called two-fluid model derived from Reynolds-averaged Navier–Stokes equations in two-phase flow. From the local perspective, CFD leads to the distributions of phases, interfacial area, and velocity field in the whole volume of the airlift. Numerical simulations are discussed after comparison with experimental data. Sensitivity analysis is then presented to highlight the main parameters that must be taken into account in two-fluid modeling, especially in terms of interfacial transfer of momentum and turbulence. Once local hydrodynamics has been discussed and validated, the axial dispersion is then addressed. The axial dispersion coefficient is estimated from simulation of transport equation of salt concentration. Given the time evolution of the concentration, measured by a “numerical” probe located in computed airlift reactor, it is possible to numerically estimate the axial dispersion in the airlift, in the same way as in the experiments. The axial dispersion coefficient determined after simulation is compared with the experimental one and the ability of CFD to simulate mixing time and axial dispersion is shown. In addition, a physical analysis of axial dispersion is proposed. © 2007 American Institute of Chemical Engineers AIChE J, 2007

Ancillary