Fundamentals of Particle Fouling in Membrane Processes

The permeate flux decline due to membrane fouling can be addressed using a variety of theoretical stand-points. Judicious selection of an appropriate theory is a key toward successful prediction of the permeate flux. The essential criterion f3r such a decision appears to be a detailed characterization of the feed solution and membrane properties. Modem theories are capable of accurately predicting several properties of colloidal systems that are important in membrane separation processes from fundamental information pertaining to the particle size, charge, and solution ionic strength. Based on such information, it is relatively straight-forward to determine the properties of the concentrated colloidal dispersion in a polarized layer or the cake layer properties. Incorporation of such information in the framework of the standard theories of membrane filtration, namely, the convective diffusion equation coupled with an appropriate permeate transport model, can lead to reasonably accurate prediction of the permeate flux due to colloidal fouling. The schematic of the essential approach has been delineated in Figure 5. The modern approaches based on appropriate cell models appear to predict the permeate flux behavior in crossflow membrane filtration processes quite accurately without invoking novel theoretical descriptions of particle back transport mechanisms or depending on adjust-able parameters. Such agreements have been observed for a wide range of particle size ranging from small proteins like BSA (diameter ${\~}$6 nm) to latex suspensions (diameter ${\~}1\;{\mu}m$). There we, however, several areas that need further exploration. Some of these include: 1) A clear mechanistic description of the cake formation mechanisms that clearly identifies the disorder to order transition point in different colloidal systems. 2) Determining the structure of a cake layer based on the interparticle and hydrodynamic interactions instead of assuming a fixed geometrical structure on the basis of cell models. 3) Performing well controlled experiments where the cake deposition mechanism can be observed for small colloidal particles (< $1\;{\mu}m$). 4) A clear mechanistic description of the critical operating conditions (for instance, critical pressure) which can minimize the propensity of colloidal membrane fluting. 5) Developing theoretical approaches to account for polydisperse systems that can render the models capable of handing realistic feed solutions typically encountered in diverse applications of membrane filtration.