(619e) A Lattice-Boltzmann Approach to Two-Phase Flow with High Density Ratios in Inclined Channels of Structured Packings

In a Fischer-Tropsch reactor (part of a Gas-to-Liquid, GtL process), carbon monoxide and hydrogen are converted into liquid hydrocarbons with the help of a catalyst. One of the options for carrying out this highly exothermic chemical process is by depositing a very thin catalyst layer on a structured packing the gaseous and liquid components are forced to flow through. A particular type of packing consists of inclined channels which direct the flow towards cooling walls of the reactor resulting in better heat transfer towards an external coolant. This study aims at gaining a better insight in the exact way gas and liquid interact mutually in the presence of the solid walls of the inclined channels in a structured packing.

In this study, the Shan-Chen (S-C) pseudo-potential Lattice Boltzmann (LB) method was applied and adapted for simulating two-phase flow with high density ratios through a structured packing. In the S-C approach, the LB formulation is extended with an intermolecular force expressed as the gradient of a potential which is a function of the local fluid density. The form of the potential function dictates the resulting (non-ideal) equation of state (EOS). When operated in the subcritical temperature range, the non-ideal fluid comprises two coexisting states (phases) each having a different density. In addition to this density bifurcation, the interaction potential also provides a means of controlling the interfacial tension between the different phases.

In the original S-C LB method, the ratio of the densities of the two phases should be kept low as to avoid numerical instabilities. We succeeded in extending the S-C method towards higher density ratios, representative of a realistic gas-liquid system, by incorporating a different, more representative EOS. In the code written in house on the basis of the original S-C model, various EOS were implemented which describe the non-ideal fluids in the reactive system of interest.

The code has been validated for a variety of canonical cases such as Ostwald ripening (in the absence of flow), capillary rise, wall wetting with varying contact angle, and the rise of a single bubble through a stagnant liquid.

Flows with different capillary numbers, subject to wall-fluid adhesive forces, with several gas/liquid density ratios and in various channel geometries were studied. This study is expected to improve our understanding of flow behaviour in structured packings which may lead to increasing their performance in industrial reactors.