(410b) Electrostatic Impacts On Fouling of Microfiltration Membrane

Over the last two decades, PNNL has been working on separation processes for the Waste Treatment Plant and in particular cross flow filtration. Because of the high cost of working with highly radioactive materials, it is desirable to develop methods to estimate filtration performance for Hanford HLW. However, development of predictive correlations from existing WTP test data is difficult due to the complexity and variability in actual waste and waste simulant properties such as solid composition, supernate composition, particle size, and rheology of each test sample. Despite these difficulties, a model of filtration transience behavior has been recently developed from simulant and actual waste testing. This model can capture the dynamics of particle transport to the filter surface and the interaction of that particle with the filter surface. With regard to particle-filter interactions, the model approximates: 1) irreversible fouling caused by penetration of the porous filter element by fine particulates, 2) reversible adherence of particles to the surface of the element, and 3) back transport of particles away from the surface as a result of fluid cross-flow. While the current model can adequately capture filter flux dynamics, it is semi-empirical and contains parameters that must be determined by fitting actual filtration data. As such, the model is not predictive with respect to fundamental material properties (such as size and charge). Fundamental particle properties play an important role in filter/membrane depth fouling and the formation of solids cakes on the filter surface. With respect to depth fouling, penetration of the porous network is governed by particle size and particle filter interactions are governed by standard DLVO (Derjaguin, Landau, Verwey, and Overbeek.) interactions. The surface charge of the filter and particle play a significant role in how strongly the filter fouls (and if that fouling is reversible). When the filter surface and particles have the same charge, fouling is expected to be minimal. When the filter and particles have opposite charges, fouling is expected to be strong. Because the filter medium and the particles comprising the dispersion are typically different materials, charge disparity between the two is likely. However, surface charge is also affected by pH and ionic strength of the slurry supernatant. To advance a fundamental understanding of filter fouling dynamics, the impacts of particle interactions (as governed by zeta potential) were tested and incorporated into existing models for filtration.