Applicability of ion exchange for NOM removal from a sulfate rich surface water incorporating full reuse of the brine

The raw water of water treatment works ‘the Blankaart’ (Belgium) is characterized by high levels of NOM and alkalinity. From 2006 until present, the possibility of applying fluidized ion exchange for NOM removal at the Blankaart plant has been investigated. Although the efficiency of ion exchange with respect to NOM removal has already been demonstrated by numerous studies on a wide range of raw waters, a feasibility study was conducted at the Blankaart plant for several reasons. Firstly, the raw water at the Blankaart is characterized by a high sulfate concentration (100 mg/L on average), and it was anticipated that a large part of the exchange capacity of the resin would go to sulfate removal instead of NOM removal. Secondly, application of ion exchange for NOM removal results in a waste stream highly loaded with salts and NOM, which has to be discharged. However, at the Blankaart, discharge of liquid waste streams is not possible and therefore, the possibility of completely reusing the waste brine was investigated. Thirdly, most of the results described in literature were obtained from lab scale experiments and pilot studies using the patented Miex ® DOC system, whereas at Blankaart, the application of a fluidized bed system is under consideration. As anticipated, operation of a 1.25 m3/h pilot plant applying Miex ® resin in fluidized bed under standard operating conditions (effective resin dose 1 mL/L, regeneration with a 10% NaCl solution) revealed that 68% of the total amount of chloride exchanged is attributable to sulfate removal, whereas only 11% can be attributed to DOC removal. Nevertheless, a DOC removal efficiency of 40 to 50% and a 60 to 70% removal of UV254 adsorption was achieved. To assess the feasibility of full brine recuperation, a fluidized bed column with a nominal treatment capacity of 150 L/h was operated from 2008 until 2010, with in situ regeneration of the resin every 1500 bed volumes. The waste brine from each operating cycle was flocculated with a ferric salt and filter pressed, and the filtrate was reused as brine in the following cycle. To maintain the brine pH and salt concentration at a constant level, NaOH and NaCl were added to the filtrate prior to its reuse. Recuperation of the brine in this manner was found to have no significant impact on the NOM removal efficiency. Moreover, as a result of the accumulation of sulfates in the brine, the unwanted removal of sulfate, nitrate and bicarbonate was significantly reduced. Column experiments conducted with Miex ® on the one hand and with conventional type 1 anion exchange resins with a polyacrylic matrix and a macroporous structure on the other hand gave similar removal efficiencies. However, whereas for the Miex ® resin, a contact time of 40 seconds between raw water and resin sufficed to achieve maximum NOM removal, this was not the case for the conventional resins. Due to their larger bead size, conventional resins require higher contact times for exchange equilibrium to be reached. Increasing the contact time from 40 to 120 seconds was shown to have a significant impact on NOM removal efficiency.