화학공학소재연구정보센터
Journal of Membrane Science, Vol.467, 292-302, 2014
Characterising nanostructure functionality of a cellulose triacetate forward osmosis membrane using electrical impedance spectroscopy
Electrical Impedance spectra generated in situ, in real time for cellulose triacetate forward osmosis (CTAES) FO membranes was resolved with the Maxwell-Wagner theory to reveal distinct structures including the active separation layer and porous support. Two distinct structural elements with capacitance of 7.7 x 10(-6) (F/m(2)) and 7.8 x 10(-4) (F/m(2)) and a corresponding thickness of 43(+/- 13) mu m and 80(+/- 11) nm representing the porous support and active separation layer respectively were determined from spectra acquired on membranes operating in Active Layer Draw Side (ALDS) mode on 0.5 M potassium chloride draw solutions. Overall membrane thickness of 33-61 Lim determined using Electrical Impedance Spectroscopy (EIS) compared favourably to a thickness range of 50-90 mu m measured by Scanning Electron Microscopy. A stationary ion layer with a capacitance of 7.7 x 10(-6) (F/m(2)) was visible on the porous support at draw solutions of 0.5 M KCI in the ALDS mode. However, as the concentration of the draw solution increased, thereby increasing the conductivity of the region, EIS was unable to interpret the interactions between stationary ion layer and the porous support. Reversing the membrane orientation to Active Layer Feed Side (ALFS) increased the amount of internal concentration polarisation (ICP) in the porous support compared with ALDS mode resulting in a decrease in FO flux from 3.9(+/-0.2) L/m(2) h to 2.5(+/- 0.2) L/m(2) h. The presence of ICP and its subsequent impact on flux decline may be revealed from EIS spectra by observing an overall increase in conductance. While EIS remains a viable technique to characterise membrane structure and thickness, identification of coupled effects of internal and external concentration polarisation in situ remains elusive and requires further improvement of signal to noise ratio at higher concentrations and improvement in Maxwell-Wagner fitting algorithms. (C) 2014 Elsevier B.V. All rights reserved.