2019 AIChE Annual Meeting

(242c) Particle Back Transport Mechanisms in Crossflow Filtration with Nuclear Waste Simulants

Authors

Philip P. Schonewill - Presenter, Pacific Northwest National Laboratory
Richard C. Daniel, Pacific Northwest National Laboratory
Carolyn A. Burns, Pacific Northwest National Laboratory
Rick W. Shimskey, Pacific Northwest National Laboratory
Reid A. Peterson, Pacific Northwest National Laboratory
Crossflow filtration is a planned unit operation in the Hanford Waste and Immobilization Treatment Plant Pretreatment Facility and has also been considered for use in other pretreatment facilities on the Hanford site. Crossflow filtration may be used across a significant range of solids concentration, from a few hundred parts per million to 20 wt% undissolved solids. The wide range of possible concentrations, combined with the chemical complexity of the Hanford waste, introduces challenges in performing a priori estimation of filter fouling and production rates, i.e., filter flux.

More specifically, a persistent disconnect has been observed between crossflow filtration models in the literature and Hanford waste or waste simulant filtration data; this complicates the prediction of long time performance using macroscale physical and chemical parameters. Most models forecast that a cake quickly forms on the filter surface and reaches a steady-state thickness within minutes or a few hours. In general, Hanford filtration does not follow that trend; either a steady-state is not quickly achieved or the models over-estimate the steady-state flux (which is often only a pseudo steady-state) by factors of two to three. To probe this difference, historical Hanford waste and waste simulant crossflow filtration data were examined against a series of proposed back transport mechanisms. Across the examined data sets, shear-induced diffusion was found to best describe crossflow filter performance.