Nanofiltration of secondary effluent for wastewater reuse in the textile industry

Textile industries represent an important environmental problem due to their high water consumption. In many Spanish regions with water scarcity, this fact can be an argument to make wastewater reuse necessary. In this work, a biologically treated wastewater from a cotton thread factory was subjected to nanofiltration (NF) in two ways, direct NF treatment and NF after a pre-treatment stage by ultrafiltration (UF). Nowadays, the factory effluent is treated by an activated sludge process. This effluent has high values of COD (200 mgO2 L−1) and TDS (5000 mg L−1) which makes its use in the textile processes impossible. In such situations, reverse osmosis (RO) has been typically used with the purpose of reuse. However, NF can achieve enough permeate quality for certain processes at a lower operating pressure than RO. To verify the possibility of reusing textile wastewater with NF, three NF membranes of different pore size (NF90, NF200 and NF270 from Dow-Filmtec) were studied. In order to select the most effective one, experiments were carried out at four different pressures in a pilot plant equipped with a flat-sheet membrane module. The studied responses were permeate flux, salt retention and COD removal. The NF90 yielded a COD reduction of 99% and the highest salt retention (75–95%). As the permeate quality obtained with this membrane was satisfactory, long duration experiments were performed with a spiral-wound module in order to scale-up the system and to study the effect of fouling. These experiments showed that the levels of COD removal and salt retention were not significantly affected by fouling and that a high flux percentage could be retrieved after cleaning. Finally, UF and NF experiences were coupled in order to study the effect of UF as pre-treatment in a NF system. In this case, permeate flux of NF increased a lot (about 50%) and COD concentration in NF feed was reduced (about 40%).

Comparing Fig. 7 and Fig. 10, the previous use of UF largely improved the flux of the NF membrane. The UF used removed between 30 and 40% of the COD. However, the COD fraction retained was responsible for most of the fouling. The UF module presented a severe fouling and had to be subjected to continuous cleaning operations. After the UF treatment, the NF flux nearly doubled and the initial flux after flushing with water was almost completely retrieved. As it can be seen in Fig. 11, the salt rejection was slightly smaller than in the case of direct UF.