Numerical investigation of surface wettability on gas - Liquid flow hydrodynamics in microchannel: Application to trickle bed reactors

Two-phase gas-liquid flows are very important in trickle bed catalytic reactors. To investigate the effect of catalyst wettability on the two-phase flow hydrodynamics in the bed, microchannels are considered in order to simulate the pore scale phenomena. Since the surface area to volume ratio is a key factor at the microscale, the role of surface interaction becomes dominant. In this work, the channel hydrophobicity is varied in order to study the effect of the wetting properties of a rectangular horizontal micro-channel on the gas-liquid two-phase flow behavior. Air and water are injected through a Y junction microchannel where water is injected in the middle and air from side inlets, to investigate the air/water two-phase flow behavior in the straight section. The surface wettability of two types of catalysts is experimentally measured and used as input to the numerical computational model. The microchannel is of 300 mu m, 100 mu m and 1 cm in width, height and length, respectively. The CFD simulation is done on COMSOL Multiphysics, using the Level Set method. The level-set technique is used to represent a discrete fluid interface between the two phases. Navier Stokes equations and conservation of mass, coupled to the level set method keep track of the interface between the air and water in the flow field, including the interfacial interactions with the microchannel walls. The developed numerical model is used to characterize the hydrodynamics of cocurrent gas-liquid flow through the microchannel. The model was validated against published experimental data and good agreements are obtained. The prediction results demonstrate that the wetting properties of the microchannel affect drastically the two-phase flow hydrodynamic behaviors. We identified eight different flow patterns in the hydrophobic microchannel, whereas, in the case of hydrophilic microchannel, only two flow patterns were observed. All of these results were in good agreement with published experimental ones. The developed numerical model was thus used to predict trickle bed flow hydrodynamics at the microscale.