Tangential electrokinetic characterization of hollow fiber membranes: Effects of external solution on cell electric conductance and streaming current

The methodology of tangential electrokinetic measurements with hollow fiber (HF) membranes was investigated. Two cell configurations were used, one configuration in which HF were immersed in the measuring solution and another one in which HF were embedded in a gel. The experiments were conducted with hydrophilic polysullone HF of different lumen cross sections but of the same cross section of porous body in order to evaluate the contributions of porous fiber body and its lumen to both streaming current and cell electric conductance. The good linearity of streaming current/potential data vs. pressure difference gives evidence that expressions of the streaming current/potential derived in laminar flow are still valid for turbulent flux conditions (at least for experimental conditions used in the present study). Significant differences in both electric conductances and streaming current coefficients were obtained between he two cell configurations. The differences in the electric conductances show that he solution around fibers makes contribution to he cell electric conductance (as demonstrated previously). The observed differences in the streaming current coefficients could be clue to either the contribution of the external surface of fibers to the measured streaming current or a non-negligible Starling recirculation flow. For tests conducted with HF embedded in the gel, linear variations of streaming current coefficient and cell electric conductance on channel cross section were obtained, as originally demonstrated by Yaroshchuk and Luxbacher for channels with porous walls (A.E. Yaroshchuk, T. Luxbacher, Interpretation of electrokinetic measurements with porous (films: role of electric conductance and streaming current within porous structure, Langmuir, vol. 26, 2010, pp. 10882-10889). Surprisingly, it is found that the (negative) zeta potential of pore surface is higher (in absolute value) than that of the lumen surface. The contribution of the fiber body to both streaming current and cell electric conductance turns out to be considerable, It is shown that neglecting this additional path (i.e. the fiber body) for streaming current leads to an overestimation (in absolute value) of the zeta potential by a factor of 1.7-5.7 depending on the fiber. This is a clear indication that the contribution of the fiber body to the streaming current cannot be neglected in the case of standard HF ultra-filtration membranes otherwise the interpretation of their electrokinetic properties could be distorted. With such materials, it is therefore recommended not to convert the measurement of the streaming current into zeta potential of lumen surface. (C) 2015 Elsevier B.V. All rights reserved.