Fouling of polyelectrolyte multilayer based nanofiltration membranes during produced water treatment: The role of surfactant size and chemistry

Large volumes of water become contaminated with hydrocarbons, surfactants, salts and other chemical agents during Oil & Gas exploration activities, resulting in a complex wastewater stream known as produced water (PW). Nanofiltration (NF) membranes are a promising alternative for the treatment of PW to facilitate its re-use. Unfortunately, membrane fouling still represents a major obstacle. In the present work, we investigate the effect of surface chemistry on fouling of NF membranes based on polyelectrolyte multilayers (PEM), during the treatment of artificial produced water. To this end, oil-in-water (O/W) emulsions stabilized with four different surfactants (anionic, cationic, zwitterionic and non-ionic) were treated with PEM-based NF membranes having the same multilayer, but different top layer polymer chemistry: crosslinked poly(allylamine hydrochloride) (PAH, nearly uncharged), poly(sodium 4-styrene sulfonate) (PSS, strongly negative), poly(sulfobetaine methacrylate-co-acrylic acid) (PSBMA-co-AA, zwitterionic) and Nafion (negative and hydrophobic). First, we study the adsorption of the four surfactants for the four different surfaces on model interfaces. Second, we study fouling by artificial produced water stabilized by the same surfactants on PEM-based hollow fiber NF membranes characterized by the same multilayer of our model surfaces. Third, we study fouling of the same surfactants solution but without oil. Very high oil retention (>99%) was observed when filtering all the O/W emulsions, while the physicochemical interactions between the multilayer and the surfactants determined the extent of fouling as well as the surfactant retention. Unexpectedly, our results show that fouling of PEM-based NF membranes, during PW treatment, is mainly due to membrane active layer fouling caused by surfactant uptake inside of the PEM coating, rather than due to cake layer formation. Indeed, it is not the surface chemistry of the membrane that determines the extent of fouling, but the surfactant interaction with the bulk of the PEM. A denser multilayer, that would stop these molecules, would benefit PW treatment by decreasing fouling issues, as would the use of slightly more bulky surfactants that cannot penetrate the PEM.