A multiphase multicomponent flow and transport model for liquid aerosol filtration in coalescing fibrous filters

Pore-scale CFD simulations of liquid aerosol filtration in coalescing fibrous filters are currently limited by severe computational constraints. In this paper, we present a new multiphase multicomponent framework for modeling such dynamic filtration process at the Darcy-scale. Moving from the pore- to the Darcy-scale allows for significant reduction in computational efforts with limited repercussion on the representation of the filter geometry. In our model, two phases coexist within the filter, namely the injected oil-mist and the coalesced oil phase. The oil-mist phase consists of a mixture of a gas phase and oil droplets of different sizes. The governing equations are the standard ones of Darcy flow and transport in porous media, where the filtration processes, such as droplet capture and drainage, are accounted for through ad hoc mass exchange terms, which are coherent with the Jump & Channel model by Kampa et al. (2014). The equations are strongly non-linear and are solved through the IMplicit Pressure Explicit Saturation (IMPES) algorithm. Cell-centered finite volumes are used for discretization in space. The model is validated by replicating experiments of oil-mist filtration available in the literature for both wettable and non-wettable media, as well as for combinations of media with different wettability. Simulations are then performed for a variety of operational conditions, demonstrating both the accuracy and robustness of our implementation.