AWWA MTC57556

The City of Scottsdale (City) treats approximately 11.5 million gallons per day (mgd) of groundwater contaminated with volatile organics using packed tower air strippers. The total dissolved solids (TDS) and hardness concentrations of air stripped effluent are respectively 830 mg/L and 390 mg/L. The City wants to improve the quality of this treated water, which serves as the primary potable water supply to their southern neighborhoods, by treating a portion of the air stripped effluent using reverse osmosis (RO) and blending the RO product water with the remaining air stripped effluent to achieve a target TDS of approximately 550 mg/L. In addition to lowering the TDS and hardness, the City expects the RO treatment to keep the levels of arsenic and nitrate below their target goals of 6 ug/L and 8 mg/L (as nitrogen), respectively. Pilot tests were conducted on air stripped effluent to confirm the feasibility of RO treatment and to develop criteria for full-scale plant design and operation. The pH of the air stripped effluent was between 8.2 and 8.4. To control potential scaling from calcium carbonate, barium sulfate and silica, the RO feed water pH was lowered to 7.4-7.6 (during initial weeks of testing) and to 7.0-7.2 (in the later weeks of testing). The feed water was also conditioned by adding scale (threshold) inhibitor (SI) at a dose that was estimated based on maintaining a Langelier Saturation Index (LSI) of +2.0 in the RO concentrate at 80% recovery. Initially, projections were performed on a wide range of membranes that included nanofiltration softening, low energy (LE) RO, low fouling (LF) RO and high rejection (HR) RO membranes. LE membranes provide the lowest energy consumption while meeting the blended water quality goals and were selected for pilot testing. Pilot tests were conducted using a two-stage system with 18, FilmTec BW30-LE, 4040 elements. Parallel tests were conducted on alternative RO membranes (BW30-FR and AK-F) and chloraminated feed water using two single-stage 3- element pilot units to evaluate benefit of HF and disinfected feedwater to control biofouling, should it become problematic with the two-stage system. The air stripping treatment increased the pH/LSI of the groundwater and resulted in super-saturation of calcium carbonate in the RO feed water. When the large RO unit was operated at feed water pH of 7.4-7.6, significant loss in normalized product flow was observed in the lag element. The autopsy results of the lag element (after citric acid cleaning) indicated presence of silica with minor amounts of iron and calcium. Silica scale was probably caused by the high feed water pH (7.4-7.6) and prior precipitation of calcium carbonate. Stable performance was restored by replacing the lag element with a new element and by lowering the feed water pH to 7.0-7.2. Combination of silica-effective SI and acid addition is necessary to control scale formation in the air stripped groundwater. Addition of chloramines to the RO feed resulted in membrane degradation as measured by the significant (~50%) increase in salt passage and normalized product flow. Provision of space in the fullscale design for including a biological control agent (chloramines or other agents) is recommended. If chloramines are used, then utmost care must be taken to avoid the presence of any free chlorine in the RO feed. This paper discusses the results of the pilot testing and also presents some key features of the preliminary design. Includes 2 references, tables, figures.