Jump to main content.


  Arsenic Removal from Drinking Water by Adsorptive Media, U.S. EPA Demonstration Project at Dummerston, VT, Final Performance Evaluation Report
July 2008

This report documents the activities performed and the results obtained for the arsenic removal treatment technology demonstration project at Charette Mobile Home Park (CMHP) in Dummerston, Vermont. The objectives of the project were to evaluate: (1) the effectiveness of an Aquatic Treatment Systems (ATS) arsenic removal system in removing arsenic to meet the new arsenic maximum contaminant level of 10 micrograms per liter (µg/L), (2) the reliability of the treatment system, (3) the required system operation and maintenance (O&M) and operator skills, and (4) the capital and O&M cost of the technology. The project also characterized water in the distribution system and residuals produced by the treatment process.

The ATS system consisted of two parallel treatment trains, each having three 10-inch diameter, 54-inch tall sealed polyglass columns connected in series to treat up to 11 gallons per minute (gpm) of water. Water supplied from three source water wells was chlorinated to provide chlorine residuals and then passed through a 25-micrometer sediment filter and the three adsorption columns in each train. Each adsorption column was loaded with 1.5 cubic feet of A/I Complex 2000 adsorptive media, which consisted of an activated alumina substrate and a proprietary iron complex. Based on the design flow rate of 11 gpm through each train, the empty bed contact time (EBCT) in each column was one minute and the hydraulic loading rate to each column was 20.4 gpm per square foot. The actual flow rate was much lower, averaging only 2.8 and 3.3 gpm for trains A and B, respectively, throughout the evaluation period. A 50 percent reduction in flow was observed after the 23rd week of operation. The flow rate increased again after the 39th week but fluctuated greatly after this point. As a result, each adsorption column had a much longer EBCT, ranging from 1.6 to 56.1 minutes throughout the entire study period. The highly variable and low flow rates from the wells might be attributed, in part, to slow recovery rates of the aquifer resulting from a dry summer.

Between June 24, 2005, and October 10, 2006, the system operated at an average of 7.6 hours per day for a total of 3,636 hours, treating approximately 745,000 gallons of water that contained 20.8 to 101 µg/L of arsenic, existing predominately as soluble arsenic (V). During the first 34-week-long test run, arsenic concentrations following the lead columns reached 10 µg/L after treating 5,700 and 5,400 bed volumes (BV) of water through trains A and B, respectively. (BV was calculated based on 1.5 cubic feet [or 11.2 gallons] of media in an individual column.) Arsenic concentrations reached 10 µg/L in the system effluent (following the final columns) after treating approximately 17,400 and 17,600 BV through trains A and B, respectively (or 5,800 and 5,900 BV, respectively, if considering the three columns in each train as one large column).

Media were replaced after approximately 8 months of operation, and arsenic concentrations reached 10 µg/L in the system effluent (after the second lag column) after approximately 15,000 BV and 17,000 BV for trains A and B, respectively (or 5,000 and 5670 BV, respectively, if considering the three columns in each train as one large column). Arsenic concentrations in the effluent of the new lead columns were around 10 µg/L at the time of the media changeout.

Arsenic breakthrough occurred sooner than projected (at 40,000 BV in the lead column) by the vendor. It is presumed that relatively high pH values of source water (averaging 7.6), competing anions, such as silica, and higher influent arsenic concentrations (i.e., 41.3 µg/L, on average, compared to 30 µg/L observed during the initial site visit) might have contributed, in part, to early arsenic breakthrough from the adsorption columns. The arsenic mass removed by the adsorption media during the two runs ranged between 0.30 and 0.49 micrograms of arsenic per milligram of dry media per column.

Aluminum concentrations in the treated water following adsorption columns (existing primarily in the soluble form) were approximately 10 to 30 µg/L higher than those in raw water, indicating leaching of aluminum from the adsorptive media. Leaching of aluminum continued throughout the study period; however, there was a decreasing trend in aluminum concentration in the treated water during each test run.

Comparison of distribution system sampling results before and after operation of the system showed a significant decrease in arsenic concentrations at two of the three residences. One residence had elevated arsenic concentrations ranging from 16.3 to 26.0 µg/L through the first three months. Starting from the fourth month, all three residences had arsenic concentrations below 3.1 µg/L. After the sixth month, arsenic concentrations began to increase and media were changed out after 34 weeks of operation. Arsenic concentrations decreased again after the changeout. The wells were not able to generate enough water to meet the demand of CMHP, so water was hauled in and stored in the 5,500-gallon atmospheric storage tank (where water treated from the ATS system was stored). Therefore, distribution sampling was discontinued after April 2006 because the results were not representative of the treated water from the ATS system. Lead and copper levels did not appear to have been impacted by the treatment system.

The capital investment cost of $14,000 included $8,990 for equipment, $2,400 for site engineering, and $2,610 for installation. Using the system’s rated capacity of 22 gpm (or 31,680 gallons per day [gpd]), the capital cost was $636 per gpm (or $0.44 per gpd). Annualized capital cost was $1,321 per year, based upon a 7-percent interest rate and a 20-year lifespan. The unit capital cost was $0.11 per 1,000 gallons, assuming the system operated continuously 24 hours per day, 7 days a week at 22 gpm. At the current use rate of 1,565 gallons per day, the unit capital cost increased to $2.31 per 1,000 gallons.

Operation and maintenance (O&M) costs included only incremental costs associated with the adsorption system, such as media replacement and disposal, electricity consumption, and labor. The incremental cost for electricity was negligible. Media replacement of the lead and first lag columns in each train occurred on February 14, 2006, after 34 weeks of system operation. The cost to replace the four columns was $3,910 for media, labor, and travel. This cost was used to estimate the media replacement cost per 1,000 gallons of water treated as a function of the media run length to the 10-µg/L arsenic breakthrough from the third column in series.


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.