Influence of the hydrodynamics and the reaction-rate formulation in modeling infinitely fast irreversible reactions in a turbulent un-baffled chemical reactor

Abstract

Numerical results are presented from large eddy simulations (LES) of a turbulent flow in an un-baffled stirred vortex reactor. The open source code TrioCFD is employed with a discontinuous front-tracking algorithm to detect the free surface separating the two fluids in the reactor. The influence of increasing the rotation Reynolds number on the flow hydrodynamics is investigated; a very good agreement is reported between the LES and both theoretical/experimental data. Infinitely fast and irreversible chemical reactions taking place in the liquid phase are modeled and different reaction models are examined. It is figured out that the chemical reactions are highly influenced by both hydrodynamics and the reaction model. Two macro-mixing zones are identified in the reactor where the first is in the center due to the downward-oriented forced vortex, while the second is in the free helicoidal vortex that forms about 96% of the liquid-phase volume. Numerical results show that only reaction models based on an Eddy-Dissipation Concept (EDC) are capable to reproduce the effect of turbulence in enhancing mixing, unlike classical simple-rate law reaction models that hold no information concerning the micro-mixing. Such observations demonstrate the necessity of using the EDC-based models as far as turbulence and mixing enhancement are two important phenomena encountered in the precipitation process of un-baffled mixing reactors. This study is a first mandatory step preceding the description of crystals formation due to precipitation which takes place in a wide range of energy related applications or in waste management.

Keywords:

• Reactor with vortex effect (RVE)
• large eddy simulation (LES)
• chemical reactors
• micro-mixing
• Eddy Dissipation Concept (EDC)
• infinitely fast irreversible reactions

Acknowledgements

This work was granted access to the HPC resources of CINES under the allocations A0072A07571 and A0062B07712 made by GENCI (Grand Equipement National de Calcul Intensif). The authors gratefully thank Christophe Bourcier for his help with the postprocessing and the SALOME platform.

Additional information

Funding

The authors acknowledge the financial support of the Cross Disciplinary Program on Numerical Simulation of CEA, the French Alternative Energies and Atomic Energy Commission (PTC 18SN29 : TrioCFD-GC).