Jump to main content.


  Arsenic and Uranium Removal from Drinking Water by Adsorptive Media, U.S. EPA Demonstration Project at Upper Bodfish in Lake Isabella, CA, Interim Evaluation Report (EPA/600/R-08/026) April 2008

This report documents the activities performed during and the results obtained from the first 10 months of system operation of an arsenic and uranium removal technology being demonstrated at Upper Bodfish in Lake Isabella, California. The objectives of the project are to evaluate: (1) the effectiveness of a hybrid ion exchange (HIX) technology in removing arsenic and uranium to meet the respective maximum contaminant levels (MCLs) of 10 and 30 µg/L, (2) the reliability of the treatment system, (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 characterizes water in the distribution system and process residuals produced by the treatment system.

The HIX system designed by VEETech for the Upper Bodfish site consisted of two trailer-mounted, single-stage fiberglass reinforced plastic vessels, each capable of treating up to 50 gallons per minute (gpm) of flow. The vessels were 42 inches in diameter and 60 inches in height, each containing 27 cubic feet (ft3) of ArsenXnp, a hybrid anion exchange resin impregnated with hydrous iron oxide nano-particles manufactured by Purolite. During normal operation, one vessel was put into service while the other was on standby.

During the study period from October 13, 2005, through August 3, 2006, the HIX system operated for a total of 4,631 hours, treating approximately 6,693,700 gallons of water from the Upper Bodfish Well CH2-A. The average daily run time was 15.4 hours per day and the average daily production was 22,300 gallons per day (gpd). The system flow rates ranged from 21 to 29 gpm and averaged 24 gpm, which was 48 percent of the system design flow rate. The lower flow rates resulted in longer empty bed contact times (EBCT), i.e., 9.6 to 7.0 minutes, and lower hydraulic loading rates, i.e., 2.2 to 3.0 gallons per minute per square foot (gpm/ft2).

Source water from Well CH2-A had near-neutral pH values of 6.8 to 7.2, 88 to 145 milligrams per liter (mg/L) of alkalinity (as calcium carbonate), 36 to 41 mg/L of sulfate, and 40 to 48 mg/L of silica. In addition, the well water contained 36.5 to 47.3 µg/L of total arsenic with Arsenic (V) being the predominating species at an average concentration of 40.9 micrograms per liter (µg/L). The source water also contained 26.6 to 38.9 µg/L of total uranium, with concentrations exceeding the 30-µg/L MCL most of the time.

During the first 10 months of system operation, total arsenic concentrations in the treated water were reduced to <0.1 µg/L initially and gradually increased to 10.5 µg/L after 33,100 bed volumes (BV) of throughput. This run length was 65 percent higher than the vendor-provided estimate of 20,000 BV. Meanwhile, uranium was completely removed to below the detection limit of 0.1 µg/L throughout the 10-month study period. A laboratory rapid small-scale column test on the Upper Bodfish water using the ArsenXnp media achieved a similar run length of 28,000 BV for arsenic and over 50,000 BV for uranium. The better-than-expected performance of the full-scale system might have resulted from the lower flow rates and longer EBCTs experienced by the HIX system. The HIX system did not require backwashing due to an insignificant headloss buildup across the adsorption vessel.

Comparison of the distribution system water sampling results before and after system startup showed significant decreases in arsenic concentrations at three residences. The arsenic concentrations measured at the taps of these residences typically were higher than those of the plant effluent and mirrored the breakthrough behavior of arsenic in the plant effluent. Uranium was not present in the distribution system during the baseline sampling when Well CH2-A was not in service, and is not expected to be present after system startup due to the absence of uranium in the treatment effluent. The HIX system did not appear to have any effects on other water quality parameters in the distribution system.

At 33,100 BV, the uranium loading on the ArsenXnp media was estimated to be 0.13 percent (by wet weight). According to EPA’s A Regulators’ Guide to the Management of Radioactive Residuals from Drinking Water Treatment Technologies, uranium is considered “source material” and may be subject to the Nuclear Regulatory Commission’s (NRC’s) licensing requirements if a water system generates uranium-containing residuals. However, uranium is exempt from NRC regulations if it makes up less than 0.05 percent (by weight), or an “unimportant quantity,” of the residuals (10 CFR 40.13). Although it is not clear how this 0.05 percent is defined and how the “residuals” are quantified, there is a possibility that the spent media may be classified as nonexempt material, and thus can be subject to relevant regulations on storage, transportation, and disposal. If so, the spent media may not be regenerated at Mobile Processing Technology’s facility in Memphis, Tennessee, as planned because it is not licensed to process nonexempt material. Therefore, three options were proposed by the vendor and are being evaluated for spent media disposition, including 1) partial onsite regeneration to reduce the uranium loading to below the 0.05 percent “unimportant quantity,” followed by offsite regeneration to further remove arsenic and uranium, 2) complete onsite regeneration to remove both arsenic and uranium from the media, and 3) replacement and disposal of the spent media at a permitted facility. The approach for actual spent media disposition will be described in the Final Performance Evaluation Report.

The capital investment cost was $114,070, which included $82,470 for equipment, $12,800 for engineering, and $18,800 for installation. Using the system’s rated capacity of 50 gpm, the capital cost was $2,281/gpm (or $1.58/gpd).

The O&M cost for the HIX system included only incremental cost associated with the system operation, such as media regeneration or replacement and disposal as well as labor for routine operation. The vendor estimated $12,700 for partial onsite regeneration (not including any additional cost for the subsequent offsite regeneration), $15,900 for complete onsite regeneration, and $21,950 for media replacement and disposal. By averaging the media regeneration or replacement cost over the useful life of the media (i.e., 33,100 BV or 6,685,000 gallons), the cost per 1,000 gallons of water treated for these three options would be $1.90, $2.38, and $3.28/1,000 gallons, respectively. The HIX system did not require electricity to operate. Routine activities to operate and maintain the system consumed only 50 minutes per week and the estimated labor cost was $0.13/1,000 gallons of water treated.


Tom 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.