How to use a multi-ionic transport model to fully predict rejection of mineral salts by nanofiltration membranes

Prediction of ions rejection by nanofiltration membranes is of prime importance for many industrial or environmental applications. For this purpose, many knowledge models were developed in the last decades to understand which phenomena govern the selectivity between various ions. In the present work, the possibility of predicting salts rejection by nanofiltration membranes with a classical transport model has been investigated. The Pore Transport Model (PPTM) used in this study is based on the coupling between the equilibrium partitioning at the membrane-solution interfaces and the extended Nernst–Planck equation to describe transport within the membrane pores. The main difficulty of such a model is the parameters estimation and especially the volume membrane charge (Xd) and the dielectric constant inside pores (ɛp). The membrane charge is usually characterized by streaming potential whereas the difficulty of measuring dielectric constant leads to a numerical adjustment from experimental rejections. In this work, the values of ɛp and Xd are numerically identified from ternary mixtures and the possibility of fully predicting single salts and mixtures rejection is investigated. The identification method is implemented for two membranes and various divalent ions at various concentrations. The agreement between experimental and predicted curves, presented in this paper for many conditions, allows the feeling that the full prediction of ionic separations is not completely unrealistic.

Figure optionsDownload as PowerPoint slideHighlights
► Modeling the multi-ionic transport through nanofiltration membranes.
► Numerically identifying the dielectric constant within the pores and the membrane charge for mixtures separations.
► Fully predicting the rejection curves of various ionic solutions by extrapolating the trends obtained with the mixtures.