The Membrane Technologist today has a wide variety of ultrafiltration (UF) available as a repertoire for applications development.
The concept of molecular weight cut-off (MWCO) can be arbitrary. The consequences of non-standard characterization protocols, differences in membrane morphology, generic polymer composition and the chemical nature of the MWCO test solute all contribute to this uncertainty
This work is the first of a series of investigations designed to identify key test parameters which need to be controlled to allow objective performance comparisons between different UF membrane products. The current study shows the effect of operating pressure and crossflow velocity on the polydispersed dextran retention of cellulose acetate UF membranes.
The operating pressures between 70-280 kPa and crossflow velocities between 0.3-1.5 m/s were investigated in this work. Within the scope of this work (Reynolds numbers between 1500 and 5600), it was demonstrated that the dextran retention was strongly dependent on the test parameters. In general, an increase in transmembrane pressure and/or a decrease in the average crossflow velocity will both decrease the dextran retention. The effect of the pressure and crossflow velocity can be understood based on shear induced deformation of the solute molecule.
The retention properties of the membrane are strongly dependent on the solute shape, solute flexibility, solute/membrane interaction, test operating conditions and the test apparatus and/or configuration. This is a common conclusion and typical introductory sentence used in all legitimate solute retention studies of past work. This paper is not an exception.
The immediate purpose of this work is to elucidate the effects of test parameters under the conditions typically used to characterize membrane performance (i.e. molecular weight cutoff work) and the conditions approaching that of sepralator modules.
Dextran was chosen as the challenge solute because of its relatively non-adsorbing characteristic and its demonstrated utility in past retention studies. The dextran retention curves (% dextran retention verses dextran molecular weight; Figures 3-10) also lend themselves well to the presentation of results and analysis of membrane performance.
A parameter study can not possibly cover all the available parameter combinations and researchers are forced to impose limits and establish an experimental 'window' for the study. Hence, this is the beginning of several such studies planned and accumulation of several of these studies should provide a more complete understanding of membrane separation.