Jump to main content.


  Arsenic Removal from Drinking Water by Adsorptive Media U.S. EPA Demonstration Project at Covered Wells in Tohono O’odham Nation, AZ Final Performance Evaluation Report (EPA/600/R-11/027) March 2011

This report documents the activities performed and the results obtained from the arsenic removal treatment technology demonstration project at Covered Wells in Tohono O’odham Nation, AZ. The main objective of the project was to evaluate the effectiveness of AdEdge Technologies’ (AdEdge) AD-33 media in removing arsenic to meet the new arsenic maximum contaminant level (MCL) of 10 µg/L. Additionally, this project evaluated 1) the reliability of the treatment system (Arsenic Package Unit [APU]), 2) the required system operation and maintenance (O&M) and operator skills, and 3) the capital and O&M cost of the technology. The project also characterized the water in the distribution system and residuals produced by the treatment process. The types of data collected included system operation, water quality (both across the treatment train and in the distribution system), process residuals, and capital and O&M cost.

The treatment system consisted of two 36-in × 72-in composite vessels in parallel configuration, each containing approximately 19 ft3 of AD-33 pelletized media. AD-33 is an iron-based adsorptive media developed by Bayer AG and marketed under the name of AD-33 by AdEdge. The treatment system was designed for a flowrate of 63 gal/min (gpm) (31.5 gpm per vessel) and an empty bed contact time (EBCT) of 4.5 min. Over the performance evaluation period, the actual average flowrate was 29.5 gpm for Vessel A and 30.6 gpm for Vessel B, based on readings from the flow meter/totalizer installed on each adsorption vessel. The average EBCT was 4.8 min for Vessel A and 4.7 min for Vessel B.

Each of the two wells had its own chlorination system that consisted of a storage tank, a chemical feed pump, and an injector. The chemical feed pumps were tied to the well pumps, which operated on an alternating basis, and would start injecting sodium hypochlorite (NaOCl) solutions only when the well pumps turned on. The pre-existing chlorine addition systems were configured for prechlorination and could not be feasibly converted to post-chlorination as called for in the design since the predominate arsenic species was As(V) (>90%) and oxidation of As(III) (average concentration of 0.5 µg/L) was not needed. The prechlorination system was used to maintain a target free chlorine residual of 1.0 mg/L (as Cl2) or less in the distribution system for disinfection.

As part of the water treatment system, a Destin North Bay carbon dioxide (CO2) pH adjustment/control system was used to adjust pH values of raw water from as high as 8.4 to a target value of 7.0. Due to several operational issues, the Destin North Bay system was replaced with an AdEdge CO2 pH adjustment/control system approximately five months into the study. The AdEdge system consisted of a control panel/cabinet and a "Venturi style" injector. The control panel/cabinet contained components such as an "Inlet" solenoid valve, a non-venting single stage pressure regulator, a manual loop controlled by a needle valve, an automatic loop controlled by a Burkert pH controller and solid-state SensorX in-line pH probe, a rotameter, an "Outlet" solenoid valve, and a check valve. CO2 was injected downstream of the chlorination injection point.

The treatment system began regular operation on February 13, 2008. From February 13, 2008, through the end of the performance evaluation study on March 19, 2010, the treatment system operated for a total of 3,353 hr, treating approximately 11,686,000 gal (or 41,148 bed volume [BV]) of water. The average daily operation time was 4.38 hr/day and the average daily demand was 15,276 gal/day (gpd).

Total arsenic concentrations in raw water ranged from 29.0 to 38.6 µg/L and averaged 34.9 µg/L. Soluble As(V) was the predominating species, ranging from 33.3 to 36.4 µg/L and averaging 32.4 µg/L, based on the results from six speciation sampling events. At the end of the performance evaluation study on March 19, 2010, total arsenic levels in the treated water were 0.6 and 0.4 µg/L following Vessels A and B, respectively. (Note that treatment plant water sampling continued on June 15, September 29, and November 3, 2010, with arsenic concentrations increasing from 1.2 to 4.2 and then to 3.3 µg/L following Vessels A and from 0.7 to 3.2 and then 3.0 µg/L following Vessel B. By November 3, 2010, the treatment system had treated approximately 60,000 BV of water.) Concentrations of silica and phosphorus, which could interfere with arsenic adsorption by competing for adsorption sites, averaged 26.2 mg/L (as SiO2) and were less than the method detection limit (MDL) of 10.0 µg/L (as P), respectively, in raw water. Concentrations of iron, manganese, and other ions in raw water were not high enough to impact arsenic removal by the media.

Comparison of the distribution system sampling results before and after operation of the system showed a significant decrease in arsenic concentration (from an average of 36.5 to an average of 0.9 µg/L). Arsenic concentrations in the distribution system were somewhat higher than those in the system effluent, probably caused by redissolution and resuspension of arsenic particles. Lead and copper concentrations appeared to have elevated somewhat after operation of the system; however, the elevated levels were significantly under their respective action levels.

The capital investment cost of $115,306 included $86,018 for equipment, $12,897 for site engineering, and $16,391 for installation. Using the system’s rated capacity of 63 gpm (or 90,720 gpd), the capital cost was $1,832/gpm (or $1.27/gpd) of design capacity. The capital cost also was converted to an annualized cost of $10,884/yr using a capital recovery factor (CRF) of 0.09439 based on a 7% interest rate and a 20-year return period. At a 100% use rate, the unit capital cost would be $0.33/1,000 gal. At the actual use rate, the unit capital cost increased to $1.89/1,000 gal.

The O&M cost included only the cost associated with the media replacement and disposal, CO2 and chlorine usage, electricity consumption, and labor. Although media replacement did not occur during the performance evaluation study, the media replacement cost would have represented the majority of the O&M cost. Media replacement and O&M cost per 1,000 gal of water treated was estimated and plotted as a function of the projected media run length to the 10 µg/L arsenic breakthrough.


Thomas Sorg

You will need Adobe Reader to view some of the files on this page.
See EPA's PDF page to learn more.

Office of Research & Development | National Risk Management Research Laboratory

Local Navigation

Jump to main content.