Jump to main content.


Arsenic Removal from Drinking Water by Adsorptive Media
U.S. EPA Demonstration Project at Valley Vista, AZ
Final Performance Evaluation Report
(93 pp, 3673 Kb) November 2007

You will need Adobe Acrobat Reader to view some of the files on this page. See EPA's PDF page to learn more about PDF, and for a link to the free Acrobat Reader.


This report documents the activities performed during and the results obtained from the arsenic removal treatment technology demonstration project at an Arizona Water Company (AWC) facility in Sedona, AZ, commonly referred to as Valley Vista. The objectives of the project were to evaluate 1) the effectiveness of Kinetico's FA-236-AS treatment system in removing arsenic to meet the maximum contaminant level (MCL) of 10 g/L, 2) the reliability of the treatment system for use at small water facilities, 3) the required system operation and maintenance (O&M) and operator skill levels, and 4) the capital and O&M cost of the technology. The project also characterized water in the distribution system and residuals generated by the treatment process. The types of data collected included system operation, water quality, process residuals, and capital and O&M cost.

After engineering plan review and approval by the state and county drinking water officials, the treatment system was installed in May 2004 and became operational on June 24, 2004. The system consisted of two 36-in-diameter, 72-in-tall fiberglass tanks in series (lead/lag), each containing 16.7 to 22 ft3 of adsorptive media. The media types evaluated included AAFS50 (an iron-modified activated alumina medium manufactured by Alcan) for Media Runs 1, 2, 2a, and 3 and ARM 200 (an iron oxide/hydroxide medium manufactured by Engelhard/BASF) for Media Run 4. The system was designed to treat 37 gal/min (gpm) of flow using 22 ft3 of media per tank, which corresponded to an empty bed contact time (EBCT) of 4.5 min/tank and 9.0 min for both tanks. Due, in part, to the use of an incorrect AAFS50 media density and, thus, shipment weight, 16.7 ft3 of AAFS50 media was inadvertently loaded into each tank for Media Runs 1 and 2a, resulting in a shorter EBCT of 3.5 min/tank.

Source water contained 23.5 to 49.8 g/L of total arsenic, with As(V) being the predominating species, averaging 39.7 g/L. Prechlorination, although not required for oxidation, was initiated one month after system startup to inhibit biological growth in the adsorption tanks and to provide residual chlorine in the distribution system. The treatment system operated for 24 hr/day during Media Runs 1, 2, 2a, and 4, and 16 hr/day during Media Run 3 with less than 1% downtime for repairs and media replacement. Concentrations of iron, manganese, silica, orthophosphate, and other ions in source water did not appear to impact arsenic removal by the media.

After treating 8,240 bed volumes (BV) of water during Media Run 1 based on 33.4 ft3 of media in the lead and lag tanks, the system effluent exceeded the 10-g/L arsenic MCL. Source water pH, with values ranging from 7.5 to 8.4 and averaging 7.7, was then adjusted to approximately 6.9 using 37 to 50% sulfuric acid (H2SO4) at the end of Media Run 1 and throughout Media Runs 2 and 2a. Lowering the pH values, beginning on September 17, 2004, reduced the arsenic concentrations after both tanks, but not to the desired level of 10 g/L.

After media changeout of both tanks on October 25, 2004, Media Run 2 began with virgin AAFS50 media. pH adjustment increased the AAFS50 media run length to 23,030 BV at 10-g/L arsenic breakthrough in the system effluent based on 44 ft3 of media in the lead and lag tanks. Due to the increased media capacity, it was economical to rebed only the lead tank at this time and continue utilizing the remaining capacity of the lag tank after it was switched to the lead position. Thus, Media Run 2a began on April 29, 2005. Operational problems associated with system programming resulted in the tanks returning to their default positions following power outages. The system programming was later corrected by the vendor to allow the tanks to remain in their current positions following any power interruptions.

Media Run 3, which commenced on October 12, 2005, evaluated the use of AAFS50 media again under the unaltered pH condition, but with an intermittent run time of 16 hr/day and longer EBCT than Media Run 1 (i.e., 4.6 instead of 3.5 min/tank). Under these conditions, 10-g/L arsenic breakthrough in the system effluent occurred at approximately 10,360 BV based on 44 ft3 of media in the lead and lag tanks.

Media Run 4 began on March 7, 2006, with ARM 200 media, unaltered pH, and 24 hr/day operation after media changeout of both tanks. The system effluent reached 10 g/L of arsenic at 25,720 BV based on 44 ft3 of media in the lead and lag tanks. The treatment system was shut down on September 18, 2006, due to the conclusion of the demonstration study and well maintenance by AWC.

Comparison of the distribution system sampling results before and after the commencement of system operation showed a decrease in the average arsenic concentration at three locations (i.e., from 39.2 to 44.5 g/L to 8.7 to 27.4 g/L). Arsenic levels were reduced most prominently at the location closest to the treatment system and that received water most representative of the system effluent. Arsenic concentrations at the other two locations were much higher than those of the treatment effluent, presumably due to blending with other untreated wells supplying the distribution system. Similarly, alkalinity and pH values were reduced at the nearby location during pH adjustment, but they fluctuated widely at the other two locations. The lead, copper, manganese, iron, and aluminum concentrations at the three sampling locations did not appear to be significantly impacted by the arsenic treatment system.

Treatment system residuals included spent media and backwash water. All spent media including 9,100 lb of AAFS50 media and 2,200 lb of ARM 200 media passed EPA Toxicity Characteristic Leaching Procedure (TCLP) tests and could be disposed of as non-hazardous wastes at solid waste landfills. Backwash of the filter media was manually initiated monthly using treated water for 20 min/tank at 27 to 36 gpm (or 4 to 5 gpm/ft2) for AAFS50 media and for 15 min/tank at 34 to 42 gpm (or 5 to 6 gpm/ft2) for ARM 200 media. No significant pressure buildup was observed during the service runs. Backwash water from the lead tank generally contained higher concentrations of all analytes than the lag tank most likely because it removed the majority of the particulates from source water. A piping loop and a recycling tank enabled the system to recycle nearly 100% of the wastewater produced during normal system operation at a maximum flowrate of 3.6 gpm.

The capital investment cost of the system was $228,309 consisting of $122,544 for equipment, $50,659 for site engineering, and $55,106 for installation. Using the system's rated capacity of 37 gpm (or 53,280 gal/day [gpd]), the capital cost was $6,171/gpm (or $4.29/gpd). The capital cost also was converted to an annualized cost of $21,550/yr based on a 7% interest rate and a 20-yr return period. During the first year, the system produced 18,750,000 gal of water, so the unit capital cost was $1.15/1,000 gal. The capital cost does not include the cost of the enclosure to house the treatment system.

The O&M cost for the treatment system included cost for media replacement and disposal, chemical supply, incremental electricity consumption, and labor. Representing the majority of the O&M cost, the media replacement and disposal cost depended on the operating conditions affecting the media run length, the number of tanks to be changed out when the system effluent reached 10 g/L of arsenic, and labor and material cost. Due to the short duration of using AAFS50 without pH adjustment, it might be more cost-effective to replace the media in both lead and lag tanks when the system effluent reached 10 g/L of arsenic. System operations using AAFS50 with pH adjustment and ARM 200 without pH adjustment were able to last about three times longer, so it was sensible to replace the media of only the lead tank when the system effluent reached 10 g/L of arsenic. The combined chemical supply, electricity, and labor cost was $0.19/1,000 gal without pH adjustment and $0.91/1,000 gal with pH adjustment. The total O&M cost for AAFS50 media without pH adjustment and rebedding both tanks at the same time was $2.74/1,000 gal. The total O&M cost for rebedding one tank at a time was $1.49 or $1.79/1,000 gal using AAFS50 with pH adjustment or ARM 200 without pH adjustment, respectively.

For more information on this {and similar} research, please visit our research web site.


Thomas Sorg

Office of Research & Development | National Risk Management Research Laboratory

Local Navigation

Jump to main content.