G. St-Pierre Lemieux and P. Proulx
SuperMarine is a geometry generator for OpenFOAM that can be used for stirred tanks commonly used in chemical and biochemical engineering. It aims to give an easy solution for academic research and students that which to use OpenFOAM on common and simple geometry. The scripts create a blockMesh file ready to be used within a OpenFOAM case. SuperMarine was first developed to build a marine impeller, but can also be used to create a wide variety of impeller. The present version of the code is sufficient to create accurately a basic stirred tank design with its impeller, but can be tweaked in order to have more sophisticated shapes. Since it is a python script, the user can change the code has he wish and adapt it to his case.
E. Askari Mahvelati, G. St-Pierre Lemieux, P. Proulx
In the present paper, two gas‐liquid stirred tanks, one agitated by a radial impeller and another by an axial impeller, are modelled using the open‐source computational fluid dynamic (CFD) package OpenFOAM (open source field operation and manipulation). The combined effect of the bubble break‐ up and coalescence in the tank is considered by a population balance model (PBM) called extended quadrature method of moments (EQMOM). The three‐dimensional simulation is made using a multiple reference frame (MRF), a well‐established method for the modelling of mixers. Dispersed gas and bubble dynamics in the turbulent flow are modelled using the Eulerian‐Eulerian approach (E‐E) with mixture k‐epsilon turbulent model and the modified Tomiyama drag coefficient for the momentum exchange. The model is developed to predict the spatial distribution of gas phase fraction, Sauter mean bubble diameter (d32), number density function (NDF), dissolved oxygen (DO) evolution, and flow structure. The numerical results are compared with experimental data and a fair agreement is achieved. The results of the axial impeller are discussed based on four impeller rotational speeds with different volumetric mass transfer coefficients.
E. Askari Mahvelati, G. St-Pierre Lemieux, C. Braga Vieira, G. Litrico and P.Proulx
In the present paper, the oxygen dispersion in a laboratory scale (3 litres) bioreactor is modelled using open source Computational Fluid Dynamic (CFD) package OpenFOAM (Open Source Operation and Manipulation). The combined effect of the bubble breakup and coalescence in the tank is accounted by a novel method of Population Balance Model (PBM) called Extended Quadrature Method of Moments (EQMOM). The three dimensional simulation is made within a Multiple Reference Frame (MRF), which is a well established method for the modelling of mixers. Dispersed gas and bubbles dynamics in the turbulent flow are modelled using Eulerian-Eulerian (E-E) approach with mixture k-turbulent model. A modified Tomiyama drag coefficient was used for the momentum exchange, as well. Parallel computing is employed to make efficient use of computational power to predict the spatial distribution of gas phase fraction, Sauter mean bubble diameter (d32), Number Density Function (NDF), oxygen mass fraction in water and flow structure. The numerical results are compared with experimental data, and good agreement is achieved.
Camila Braga Vieira, Giuliana Litrico, Ehsan Askari, Gabriel Lemieux and Pierre Proulx
This paper presents an in-depth numerical analysis on the hydrodynamics of a bubble column. As in previous works on the subject, the focus here is on three important parameters characterizing the flow: interfacial forces, turbulence and inlet superficial Gas Velocity (UG). The bubble size distribution is taken into account by the use of the Quadrature Method of Moments (QMOM) model in a two-phase Euler-Euler approach using the open-source Computational Fluid Dynamics (CFD) code OpenFOAM (Open Field Operation and Manipulation). The interfacial forces accounted for in all the simulations presented here are drag, lift and virtual mass. For the turbulence analysis in the water phase, three versions of the Reynolds Averaged Navier-Stokes (RANS) k-ε turbulence model are examined: namely, the standard, modified and mixture variants. The lift force proves to be of major importance for a trustworthy prediction of the gas volume fraction profiles for all the (superficial) gas velocities tested. Concerning the turbulence, the mixture k-ε model is seen to provide higher values of the turbulent kinetic energy dissipation rate in comparison to the other models, and this clearly affects the prediction of the gas volume fraction in the bulk region, and the bubble-size distribution. In general, the modified k-ε model proves to be a good compromise between modeling simplicity and accuracy in the study of bubble columns of the kind undertaken here.
G. St-Pierre-Lemieux, E. Askari, D. Groleau, P. Proulx
15th International Conference on Flow Dynamics (ICFD2018) , Japan, November 2018
E. Askari, G. St-Pierre Lemieux, C. B. Vieira, G. Litrico, P. Proulx
15st International Conference of Numerical and Applied Mathematics (ICNAAM 2017) , Greece, September 2017