AWWA ACE68565

The Western Corridor Recycled Water Project is the largest water recycling project in the Southern Hemisphere estimated to be worth $AUD2.4 billion. The completed project will deliver up to 232ML (61 MG) a day of purified recycled water to supplement South East Queensland's (Australia) water supplies by the end of 2008. The Project consists of three Advanced Water Treatment Plants (AWTPs), one of which is the Luggage Point plant. Stringent target water quality parameters have been set to enable indirect potable reuse (IPR). The Luggage Point plant consists of an equalization tank, flocculation, clarification, microfiltration (MF), reverse osmosis (RO), advanced oxidation, water stabilization and treated water storage tank and pump station. Monochloramine is used for membrane biofouling control. All solids are also treated on site using a gravity filter and centrifuges. When designing an AWTP, there are many competing factors to be optimized in order to meet the target water quality, in the product water and the concentrate, as well as to protect and operate the RO membrane efficiently. At the Luggage point AWTP, a pilot plant was constructed and operated for seven months in order to assist in this optimization process. This paper presents the results related to the optimization of the RO system. The main conclusion drawn from the pilot plant operations has been the establishment of important operational guidelines for ferric chloride dose, pH, monochloramine dose and ratio of ammonia to hypochlorite. For the Luggage Point AWTP design, these issues were resolved by having a pH of between 6.5 and 7.0, ferric chloride dose of between 60 and 100 mg/L, by having a flux at 18 LMH (15 GFD) or less, by having a monochloramine optimum concentration of 0.5 ppm to 1.0 ppm, and by having sufficient instrumentation to ensure that the RO membrane is protected from oxidative damage. The use of ferric chloride to reduce the phosphorus concentration was demonstrated to be effective as 85% recovery was achieved throughout the operational period. The antiscalant dose of 2.5 mg/L was also found to be satisfactory. The RO membrane was demonstrated to achieve a rejection rate of total nitrogen of 88% and that pH would not significantly affect this. Reducing flux and increasing nitrogen concentrations as well as increasing temperature may reduce this rejection rate. The monochloramine injection point was moved to just prior to the MF system to minimize the residence time for NDMA production. By optimizing the use of monochloramine, the following conclusions can be drawn by the pilot results: bio-fouling was controlled satisfactorily with monochloramine dose of 0.5 mg/L; the cleaning requirement was less frequent than the membrane manufacturer predicted (which was once every four months); lower rejection of monochloramine than expected in fact appears to be zero rejection; and, lower production of NDMA than expected and appears to be negligible, if any. This paper demonstrates the issues presented by the many process variables in designing and operating an RO plant for IPR, and how the issues were resolved for the Luggage Point AWTP. Includes 7 references, tables, figures.