Modelling gas-liquid flow in trickle-bed reactors

The performance of a trickle-bed reactor is affected, not only by reaction kinetics, pressure, and temperature, but also by reactor hydrodynamics, which are commonly described by means of global parameters such as pressure drop, liquid holdup, dispersion of gas and liquid phases, catalyst wetting, and mass- and heat-transfer coefficients. Due to the complicated nature of trickle-bed reactor hydrodynamics, there is no straightforward method for the scale-up of these parameters from laboratory to industrial scale. Scale-up has therefore been based on a combination of different procedures: 1) constant dimensionless groups, 2) mathematical modeling, and 3) pilot planting. In general, not all dimensionless groups can be kept constant and pilot planting is both time-consuming and expensive. Thus mathematical modeling is an interesting option for the design of new trickle-bed reactors. In addition, mathematical models can be used in performance and sensitivity analysis. In this work mathematical models have been developed for pressure drop, liquid holdup, dispersion, and catalyst wetting efficiency and different modeling options are also discussed. Global pressure drop and liquid holdup can be estimated from the gas-liquid-solid phase interaction models by formulating a simplified mass and momentum equation for gas and liquid phase. New improved phase interaction models have been developed assuming uniformly distributed gas and liquid flows. Although it is assumed that the liquid is distributed uniformly, the perfect wetting of the catalyst is not assumed. A new model for liquid-solid wetting efficiency has been developed concurrently with the phase interaction models. The liquid-solid wetting efficiency is used to determine the significance of gas-liquid, liquid-solid and gas-solid phase interactions. Phase interaction and wetting efficiency model parameters have been optimized against a large experimental database for pressure drop, liquid holdup, and wetting efficiency. CFD is used to model phase dispersion in non-uniform gas and liquid flow conditions, where plug flow cannot be assumed. Spreading is attributed to three separate factors: overloading, mechanical dispersion, and capillary dispersion. The first of the spreading mechanisms derives from the phase interactions and does not require a separate model. New models have been developed for mechanical and capillary dispersion. The performance of the dispersion models is discussed on the basis of multiple case studies. The ability of the hydrodynamic model developed here is compared to liquid dispersion experiments from four different literature sources, including the author's own experiments. In addition to CFD modeling, an alternative, potential tool for liquid distribution studies is also presented - the cellular automata model. Thanks to single event modeling, a cellular automata model is simpler and thus faster and would be better suited for the modeling of larger reactors, when information on pressure drop and liquid holdup is not required