(400q) Numerical Evaluation of Solid-Liquid Drag Models for a Fluidized Bed Bioreactor

Bioreactors are used in many different processes. Among the various existing configurations, fluidized bed reactors offer advantages such as high heat and mass transfer. Fluidized beds in tapered bioreactors (TBR) offer as an additional advantage the decrease of the ascending velocity along the height, which naturally prevents the bioparticle from being carried out of the bioreactor. Several techniques can be used to evaluate these units. Among them, numerical techniques such as computational fluid dynamics (CFD) can provide detailed information for the existing fields in a non intrusive manner. In most of the literature available, there are many numerical models developed for gas-liquid or gas-solid systems. On the other hand, the dynamics in bioreactors is dominated by interactions between the liquid and solid phases (substrate and biomass, respectively). Unlike the other cases, in solid-liquid systems present in bioreactors, both phases have densities with close values. Thus, the objective of this study is to evaluate the numerical predictions of solid-liquid fluidized beds obtained in a tapered bioreactor, in which bioparticles (granulated activated carbon particles coated with biofilm) immersed in a solution of synthetic acetate wastewater were subjected to different ascending velocities. The present study also aims to contribute to the scarce literature regarding solid-liquid systems. Different drag models were evaluated in the simulation of these fluidized beds, since the drag is the main interfacial force acting in this system. The influence of the lift force on the phases dynamics was also evaluated, as well as the effect of a correction of this force for its action on solid particles. In order to conduct this work, the OpenFOAM code was used due to its low cost, and the possibility of easily changing the models available in the original code, which is required when performing adjustments in the existing models. For the conduction of the simulations, a two-phase, transient and turbulent approach was adopted. The models were solved in a symmetrical three-dimensional geometry for about 130 s, of which the last 80 s were used for the calculation of time-averaged values. A mesh sensitivity study was performed using the grid convergence index (GCI) method, which defined that a mesh containing 14.7 thousand control volumes is adequate for the prediction of reliable values. Results of this study indicate that the models developed for drag in gas-solid fluidized beds are unsuitable for predicting the height of the liquid-solid fluidized bed in the simulated conditions. It was also observed that the lift force contributes to the elevation of the bed height, and its correction to model the lift force on solid particles improves the predicted results. However, it is still not enough to accurately predict the height of the fluidized bed. Thus, although the inclusion of a corrected model for the prediction of the lift force provides little better results for the prediction of the TBR behavior, it can be concluded that there is still a need to adequate the currently available models or to obtain new correlations, in order to actually obtain accurate predictions for the bioparticles behavior in fluidized bed bioreactors.