Models to Predict Organic Contaminant Removal by RO and NF Membranes

Abstract

This article summarizes work recently completed and funded by the WateReuse Research Foundation—Predictive Models to Aid in the Design of Membrane Systems for Organic Micropollutant Removal. The project's goal was to develop a model to predict organic contaminant removal during water reuse applications employing high-pressure membranes. Various modeling approaches were evaluated, including a phenomenological model, hydrodynamic model, Donnan steric-pore model, solution-diffusion model, surface force pore model, and quantitative structure property relationship models. In bench-scale experiments, approximately 110 organic compounds and two membranes (one reverse osmosis membrane and one nanofiltration membrane) were used to develop a robust dataset for model development and validation. Modeling approaches developed using bench-scale experimental data were incorporated into mass balance models to describe rejection performance for pilot- and full-scale membrane systems. Developed modeling approaches were subsequently validated using rejection data generated through pilot-scale testing and full-scale sampling campaigns performed at water reuse facilities.