|Arsenic Removal From Drinking Water by Adsorptive Media, U.S. EPA Demonstration Project at Chateau Estates Mobile Home Park in Springfield, OH, Final Performance Evaluation Report (84 pp, 2.21 MB) (EPA/600/R-07/072)
This report documents the activities and results of the arsenic removal treatment technology demonstration project at the Chateau Estates Mobile Home Park in Springfield, Ohio. The objectives of the project were to evaluate the:
The project also characterized the water in the distribution system and process residuals produced by the treatment system. 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 costs.
The 250 gallon-per-minute (gpm) Arsenic Package Unit (APU-250) treatment system consisted of two integrated units referred to as AD-26 oxidation/filtration and AD-33 adsorption systems. The AD-26 pretreatment system was for iron and manganese removal, followed in series by the AD-33 adsorption system for arsenic removal. Both the AD-26 oxidation/filtration and AD-33 adsorption systems were skid-mounted and were composed of three carbon steel pressure vessels of similar construction and configuration but of different sizes.
AD-26 media was a manganese dioxide mineral commonly used for oxidation and filtration of iron and manganese. Because chlorine was added prior to using the AD-26 system, it helped precipitate soluble iron, oxidize arsenic (III) to arsenic (V), and form arsenic-laden solids, which were then filtered by the AD-26 media. The pretreated water was subsequently polished by the AD-33 media, an iron-based adsorptive media developed by Bayer AG for arsenic removal.
The APU-250 system began regular operation on September 21, 2005. Through September 24, 2006, the system treated approximately 16,873,000 gallons (about 19,726 bed volumes) of water, with a daily run time ranging from 3.7 to 17.3 hours per day and averaging 9.5 hours per day. The AD-26 system operated at the well pump flow rates with water supplied by two alternating wells at approximately 130 and 90 gallons per minute (gpm). The AD-33 system operated on demand from the distribution system, ranging from 9 to 71 gpm and averaging 37 gpm. Because of the low flow rates, long empty bed contact times, averaged at 23 minutes, were experienced by the AD-33 system.
The treatment system reduced the arsenic levels from between 9.5 and 35.4 µg/L (averaged 22.7 µg/L) in raw water to less than 10 µg/L in the treated water. Arsenic (III) was the predominating arsenic species in raw water, ranging from 5.6 to 25.8 µg/L and averaging 16.9 µg/L in both wells. Upon chlorination, arsenic (III) was oxidized to arsenic (V) that, in turn, was attached to the iron solids also formed during chlorination. The majority of arsenic was removed in the particulate form by the AD-26 media, leaving only 0.5 to 2.1 µg/L in solution, existing mainly as arsenic (V), to be further polished by the AD-33 media. The system also reduced total iron concentrations from an average of 1,000 µg/L to less than the method detection limit of 25 µg/L, while the total manganese concentrations decreased from an average of 35.6 to 0.1 µg/L.
The AD-26 system was backwashed initially every two days for 15 minutes with a 2-minute service-to-waste rinse, producing approximately 5,640 gallons of wastewater per backwash event. During a power outage, the backwash settings were reset to default values, prompting the system to produce almost twice as much wastewater per backwash event. This problem was resolved by manually adjusting the backwash settings, which, after a short time, were further reduced to every three days for 9 minutes with a 90-second rinse. Assuming that 83 milligrams per liter (mg/L) of total suspended solid were produced in 6,000 gallons of backwash wastewater, approximately 4 pounds (lbs) of solids (including 0.02, 1.51, and 0.03 lbs of arsenic, iron, and manganese, respectively) would be discharged during each backwash event. The AD-33 system was backwashed four times during the one-year demonstration period.
Comparison of the distribution system sampling results before and after system startup showed a significant decrease in arsenic concentration (from an average of 23.7 to 1.6 µg/L). The arsenic concentrations in the distribution system were similar to those in the system effluent. Iron and manganese were also significantly reduced in the distribution system. Neither lead nor copper concentrations appeared to have been affected by the operation of the system.
The most significant operational issue observed was related to the chlorine injection system. In spite of repeated efforts to fine-tune the chlorine injection system and reconfigure the system piping to allow the injection to be controlled by well pump flow rates instead of on-demand flow rates, as much as 4 and 3.8 mg/L (as chlorine) total and free chlorine, respectively, were measured in the treated water. This was significantly higher than the 1.5 and 1 mg/L (as chlorine) of total and free residuals targeted for the treatment. The problem seemed to have been resolved by the addition of an in-line filter placed just before the chlorine monitor to reduce clogging and coating of the chlorine probe due to iron particulates.
The capital investment cost of $292,252 included $212,826 for equipment, $27,527 for site engineering, and $51,899 for installation. The capital investment also included the cost (paid for by the park owner) to:
Using the system’s rated capacity of 250 gpm (360,000 gallons per day [gpd]), the capital cost was $1,170 per gpm of design capacity ($0.81/gpd) and equipment-only cost was $851 per gpm of design capacity ($0.59/gpd).
The O&M cost of $0.33/1,000 gallons included the incremental cost associated with the oxidation/filtration and adsorption system, such as media replacement and disposal, chemical supply, electricity, and labor. Although media replacement did not occur during the demonstration period, the adsorptive media replacement cost would represent the majority of the O&M cost and was estimated to be $34,230.
You will need Adobe Reader to view some of the files on this page.