An engineering model for solute transport in semi-aromatic polymeric nanofiltration membranes: Extension of Solution-Electro-Diffusion model to complex mixtures

Pressure-driven membrane processes, such as reverse osmosis (RO) and nanofiltration (NF), are preferred solutions for salinity reduction in ground, surface and sea waters as well as in mining and urban wastewaters. Interest in NF is increasing because it enables a more cost-efficient operation than RO. However, predictive projection tools for NF applications are still lacking. Whereas in RO the solute rejections are usually above 95% and approximately independent of feed composition, the performance of NF membranes is highly influenced by the feed-solution composition. Moreover, in both processes, concentration polarisation (CP) affects the transport of species and membrane scaling. This study presents an extended Solution-Electro-Diffusion model for the description of the transport of solutes through NF membranes in complex multi electrolytes (strong and weak) taking into account CP, namely Solution-Electro-Diffusion-Film coupled with Reactive Transport (SEDFRT) model. The model is first validated by using a semi-aromatic polymeric membrane (NF270) and electrolyte mixtures containing two dominant (MgSO4 and MgCl2) and two minor salts (NH4Cl and NaBr). The membrane permeances to all the species are determined as key parameters to be used in design projections. Finally, the model is applied for the description of the performance of NF270 membrane with model brackish surface water, influenced by urban, industrial and agriculture discharges.

Automated summary

Pressure-driven membrane processes like reverse osmosis (RO) and nanofiltration (NF) are efficient solutions for salinity reduction, but NF applications still lack predictive projection tools. NF membranes are heavily influenced by feed-solution composition, with concentration polarisation affecting species transport and membrane scaling.