Speaker
Description
Pressure drop is a key performance indicator in any system involving flow through porous media (filters, catalytic beds, membranes, soil, packed columns, fuel cells, etc.). Many studies have been concerned with the understanding of the microscopic influence on the macroscopic transport properties. Hence, various porous media flow models and techniques have been developed over the years. Besides the advanced numerical modeling procedures, which rightfully owns its place in the literature, the focus will be on the analytical drag models, mainly because these models are aimed at providing physical meaning to the empirical coefficients in empirical curve fitting models. The drag resistance models are based on statistical averages, e.g. the unit cell models, and rectangular Representative Unit Cell (RUC) model. A well known empirical model is the Ergun equation, based on the capillary tube model. Many adaptations and improvements have been added to this equation by several authors in the literature, resulting in, for instance, the tube-sphere model. The drawback of the empirical models is that they are only applicable to the media from which the empirical coefficients have been obtained. The Ergun equation is nonetheless a successful model based on its extensive use, despite its empiricism. Following a fundamentally different modelling approach, although also involving the capillary tube theory, are the fractal models. Although these models account accurately for micro-structural complexity, such as pore irregularity and surface roughness, it is usually difficult to assign numerical values to the various fractal dimensions involved. In this study an overview will be provided of some existing models with the main focus on the predictive capabilities of the RUC model. The latter model has served well over the years and was initially introduced to predict the pressure drop and permeability for Newtonian flow through different types of porous media, i.e. granular, foamlike (i.e. metal foams) and fibrous media. An RUC is introduced for each of the three different porous medium geometries. In the modelling approach involved, macroscopic equations are derived from the spatial averaging of the microscopic equations over a representative elementary volume (REV), assuming that the pore structure within the REV can be statistically represented by the averaged kinematic and geometric properties. The adaptability of the model will be illustrated to result from sound physical reasoning and consequently extends its range of applicability to different applications in which porous media are used. This includes model adaptations to predict (i) non-Newtonian flow behaviour, (ii) the permeability of low porosity sandstone by taking pore blockage into account, (iii) the effective diffusivity in the case of diffusion and (iv) the formation factor for electrical conduction. An overview will, furthermore, be given of the use of the model in collaboration with IMT Atlantique in Nantes, France, over the years for predicting the permeability of fibrous filters used in air filtration as well as predicting the pressure drop over a biofilter by taking particle surface roughness into account. Finally, planned future adaptations will be mentioned and an invitation extended to join in future collaborative projects.
| Country | South Africa |
|---|---|
| Acceptance of the Terms & Conditions | Click here to agree |
