A systematic approach to determine the fouling index for a RO/NF membrane process

A membrane fouling index such as the silt density index (SDI) and modified fouling index (MFI) is an important parameter in design and operation of reverse osmosis (RO) and nanofiltration (NF) membrane processes for desalination. Nevertheless, SDI and MFI often fail to predict actual membrane fouling due to their limited capability to measure fouling potential by colloidal deposition and organic adsorption. In this study, a novel method to determine fouling index was attempted using a set of different membrane filters for better prediction of RO/NF fouling potential. Three MF/UF membranes were used for MFI tests in order to increase sensitivity to fouling potential of particles, hydrophobic organics, and sub-micron colloids. The fouling indices determined using different membrane filters were compared with RO filtration results. Experimental results indicated that fouling indices using different test membranes showed variable results for untreated and pretreated waters. Nevertheless, no single method was sufficient for accurate measurement of fouling potential. Accordingly, a new fouling index (CFI: combined fouling index) was developed by combining various fouling indices. The CFI exhibited good agreement with experimental data compared with other fouling indices such as SDI, MFI, and MFI–UF.

In this study, a novel method of fouling index using a set of different membrane filters was developed and applied for better prediction of RO/NF fouling potential. The following conclusions can be drawn from this work: C Pretreatments using MF or UF were effective to control NF flux decline due to its high removal of particles and colloids. On the other hand, ozonation and PAC adsorption was less effective as a pretreatment for NF because they were not suitable for removing these colloidal foulants. C The results of SDI and MFI tests were dependent on not only the characteristics of feed waters but also the pore size and hydrophilicity of test membranes. Due to the properties of raw water used in this study, MFI– HP was better than the other fouling indices to predict fouling potential than MFI–UF. The CFI using multiple test membranes was found to be a promising tool for measuring the potential of RO/NF fouling by various mechanisms. Moreover, it can be used to compare the efficiency of various pretreatment processes. Future work should focus on advancing the CFI test for RO/NF membrane filtration systems.