Chemical Engineering Journal 175 (2011) 192–200 Contents lists available at SciVerse ScienceDirect Chemical Engineering Journal jo u r n al hom epage: www.elsevier.com/locate/cej Study of membrane fouling using synthetic model solutions in UF and NF processes A. Aouni a , C. Fersi a , B. Cuartas-Uribe b,∗ , A. Bes-Piá b , M.I. Alcaina-Miranda b , M. Dhahbi a,∗∗ a Water and Membrane Technologies Laboratory, BP 273, Soliman 8020 Tunisia b Department of Chemical and Nuclear Engineering, Universidad Politécnica de Valencia, Camino de Vera s/n, 46022 Valencia, Spain a r t i c l e i n f o Article history: Received 21 January 2011 Received in revised form 20 September 2011 Accepted 21 September 2011 Keywords: Ultrafiltration Nanofiltration Membrane fouling Filtration resistance Dyes solutions a b s t r a c t One of the main drawbacks for the application of membranes technologies to the treatment of textile effluents is the fouling. Fouling phenomenon causes a flux decline and reduces the quality of the treated water. Previous studies have pointed at dyes that remain in the water after rinse processes, as one of the fouling agents. In this study several UF and NF membranes were tested using three synthetic reactive dyes solutions as feed streams. The influence of applied transmembrane pressure (TMP) on normalised flux, color, COD and salt retentions was studied. The extent of fouling phenomenon was evaluated in terms of filtration resistance (R s /R t ratio) and normalized flux also with TMP. The obtained results showed that R s /R t for NF and 1 kDa membranes were not nearly affected by TMP, except for 10 kDa one. In these cases R s /R t ratio was lower than 0.5 showing a negligible fouling phe- nomenon. Similar tendency was observed with the normalized flux results. For the 10 kDa one, the normalized flux decreased progressively with TMP reaching a total normalized flux reduction of 60%. © 2011 Elsevier B.V. All rights reserved. 1. Introduction Textile industry is one of the oldest and primmest water pol- luters in the world. Due to their high volume and nature complexity, textile effluents have traditionally been object of big interest for the water treatment research. These industries mainly require a large consumption of fresh water and dyes which have acquired notoriety as hazardous sub- stances, since dyes consist of highly structured molecules with potential toxicity for organisms and high persistence in the envi- ronment [1,2], thus not allowing their easy natural biological degradation. In general, approximately 20–40% of the input dye remains in the dye-baths [3]. Their concentrations range varies from 10 to 1000 mg L -1 [4,5]. This fact makes the treatment of textile wastew- aters a difficult task requiring the examination of each factory operating conditions individually for the determination of the best treatment methodology. In addition, the fluctuating compounds concentrations and flow rates make the conventional treatments quite insufficient for the treatment of textile wastewaters, espe- cially for color removal [6]. ∗ Corresponding author. Tel.: +34 963877000x76381; fax: +34 963877639. ∗∗ Corresponding author. E-mail addresses: beacuau1@iqn.upv.es (B. Cuartas-Uribe), mahmoud.dhahbi@certe.rnrt.tn (M. Dhahbi). A special attention was given to the application of membrane processes, specifically ultrafiltration (UF) [7,8] and nanofiltration (NF) [9–14], to remove dyes. However, the main drawback of this technology is the fouling. Canic ¸ o-Gomes et al. [15] found that the main factor causing flux decline in textile solutions was the dye adsorption. This result has been reinforced by other authors [16,17]. Although there are many works about textile wastewaters, no many of them have been performed studying AZO dyes, its fouling phe- nomena and its effect on membrane behaviour. In this work, UF and NF membranes were tested to treat syn- thetic reactive dyes solutions (using black, blue and red Everzol AZO dyes). The filtration resistances were estimated in order to evaluate the membranes fouling. Besides, for each dye solution, normalized flux and salt, COD and color retentions were also determined to compare the membranes behaviour. 2. Materials and methods 2.1. Membranes tested In this work, two UF and two NF membranes were tested. The UF membranes were 10 kDa and 1 kDa from Rhodia Orelis and Synder Filtration, respectively. The NF membranes (NF-200 and NF-270) were provided by Dow Chemical. Table 1 collects main membranes characteristics according to manufacturers and liter- ature data. Also this table includes some parameters determined 1385-8947/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.cej.2011.09.093