CFD of Multiphase Flow in Trickle Bed Reactors: Porous Media Concept
Multi-scale and/or multi-disciplinary approach to process-product innovation
CFD & Chemical Engineering- I (T3-4a)
Keywords: Trickle bed reactor, CFD, porous media
Trickle bed reactors (TBR) are packed beds of catalyst with cocurrent flow of gas and liquid reactants. There are various applications of trickle bed reactors, particularly in the petroleum industry for hydroprocessing of oils (e.g. hydrotreating, hydrocracking). While scale-up of trickle bed reactors in petroleum processing for certain well characterized crude oil feedstock is probably well-established as a proprietary industrial art, a priori prediction of trickle bed performance or scale-up from laboratory reactors for new feedstock is still considered very risky and therefore not considered reliable for design. One reason for this is that even after decades of research efforts, transport processes in trickle-beds are not completely understood and are not readily quantified. With current interest in numerical prediction of hydrodynamics of trickle bed reactors, it can be observed that most of the literature available dealing with trickle bed flow simulation uses a three-phase Eulerian model in which the solids velocity is identically set to zero. The fluid phases (gas and liquid) are allowed to flow freely through this “bed” of solids as per the conservation equations and closure models. Such a calculation is computationally demanding and yet very difficult to implement in the case of predicting hydrodynamics of high pressure TBRs. In this research work, we propose to use a CFD model for the estimation of pressure drop and liquid hold-up in trickle-bed reactors by implementing the concept of porous media in order to achieve the same goal even at normal and higher pressures but with less computational effort and without compromising on accuracy.
A two-phase Eulerian model, based on porous media concept, describing the flow domain as porous region is presented to estimate the hydrodynamics of two-phase flow in trickle-bed reactors (TBRs). The drag forces between phases have been accounted by employing the relative permeability concept (Sàez and Carbonell, 1985). The model has been validated with the different sets experimental data obtained from different independent sources as well as with the prediction of computationally intensive three-phase Eulerian CFD simulations. All the comparisons lead to the fact that even without changing the Ergun’s constants arbitrarily, this model can be used to predict pressure drop and liquid saturation in low interaction regime with enough confidence at atmospheric pressure. The developed model is very much flexible unlike the traditional CFD approach, i.e. three-phase Eulerian simulations. It has been also tested for high pressure operation successfully. While simulating for high pressure condition, we have applied the recently developed correlations (Nemec and Levec, 2005) for relative permeabilities.
A less computationally intensive, yet first-principle based CFD model has been presented in this work using the porous media concept. This model has a remarkable feature of being flexible for different particle size effect incorporation without much complexity. It also seems to be a promising alternative to multi-fluid Euler-Euler drag models for trickle bed reactors. This CFD model can productively be implemented for high pressure operation (most of the commercial TBRs operate) which is cumbersome to account for three-phase Eulerian simulation.
References:
Nemec, D., and Levec, J., (2005), Flow through packed bed reactors: 2. Two-phase concurrent downflow, Chem. Eng. Sci., 60, 6958.
Sàez, A. E., and Carbonell, R. G., (1985), Hydrodynamic parameters for gas-liquid cocurrent flow in packed beds, AIChE J., 31, 52.
See the full pdf manuscript of the abstract.
