Modeling micropollutant removal by nanofiltration and reverse osmosis membranes: considerations and challenges

Organic micropollutants (OMPs) in drinking water constitute a potential risk to human health; therefore, effective removal of these pollutants is required. Nanofiltration (NF) and reverse osmosis (RO) are promising membrane-based technologies to remove OMPs. In NF and RO, the rejection of OMPs depends on the properties and characteristics of the membrane, the solute, and the solution. In this review, we discuss how these properties can be included in models to study and predict the rejection of OMPs. Initially, an OMP classification is proposed to capture the relevant properties of 58 OMPs. Following the methodology described in this study, more and new OMPs can be easily included in this classification. The classification aims to increase the comprehension and mechanistic understanding of OMP removal. Based on the physicochemical principles used to classify the 58 OMPs, it is expected that other OMPs in the same groups will be similarly rejected. From this classification, we present an overview of the rejection mechanisms involved in the removal of specific OMP groups. For instance, we discuss the removal of OMPs classified as perfluoroalkyl substances (e.g., perfluorooctanoic acid, PFOA). These substances are highly relevant due to their human toxicity at extremely low concentration as well as their persistence and omnipresence in the environment. Finally, we discuss how the rejection of OMPs can be predicted by describing both the membrane-solution interface and calculating the transport of solutes inside the membrane. We illustrate the importance and impact of different rejection mechanisms and interfacial phenomena on OMP removal and propose an extended Nernst-Plank equation to calculate the transport of solutes across the membrane due to convection, diffusion, and electromigration. Finally, we show how the theory discussed in this review leads to improved predictions of OMP rejection by the membranes.