Permeability modeling of fibrous media with bimodal fiber size distribution

Fibrous media as aerosol filtration devices are widely used in many industrial applications and present a wide variety of features like the fiber composition, the solid volume fraction, the fiber diameter distribution and the fiber orientation. Among them are nonwovens materials, which are mostly composed of fibers blends. The two main characteristics of these filters are the filtration efficiency and the pressure drop. These properties depend on the filter thickness, the face velocity, the fluid viscosity, etc. Recently, predictive expressions for bimodal media have been developed based on the combination of unimodal equivalent radius models and dimensionless permeability. The main contribution of this work is to propose combinations for a wide range of realistic fibrous filters in addition to experimental validation. The building of a universal model to predict the permeability of bimodal fibrous media for a wide range of solid volume fraction, α and fiber diameter ratio, R is envisaged. Therefore our approach has been to test different analytical correlations for 0.05<α<0.2 and 2<R<5. Based on well-known permeability models, combinations between dimensionless permeability models given by Davies (1952) or Jackson and James (1986) and unimodal equivalent radius from Brown and Thorpe (2001), Clague and Phillips (1997) or Tafreshi et al. (2009) were compared with numerical results obtained with GeoDict code. Even if the building of a universal model is far from being completed due to the large number of variables and their dependence, this comparison categorizes previous models in a unique way, which made it easy for readers to use them in different applications.