A Tailor Made Solution

Using tailored activated carbons to treat perchlorate-contaminated water

In 1997, researchers first used newly developed contaminant analytical capabilities to detect low concentrations of the perchlorate ions in groundwaters and surface waters.1 These tools helped substantiate the U.S. Environmental Protection Agency's (EPA) assertion that perchlorate is leaching into the drinking water supplies of approximately 15 million to 18 million Americans.

Concerns from this discovery of widespread contamination are based on the human health effects of consuming low levels of perchlorate. Perchlorate is known to inhibit the thyroid gland's ability to accumulate and retain iodine. The thyroid gland plays a major role in healthy child development, and also regulates metabolism in children and adults alike. Impaired thyroid function in expectant mothers may affect fetal development, resulting in behavioral changes, delayed development, decreased learning ability, and possibly even thyroid gland tumors.2

In 1999, EPA established a provisional acceptable range of perchlorate concentrations in drinking water of 4 micrograms per liter (m g/L) to 18 m g/L. EPA has suggested a drinking water equivalent level of 1 m g/L as a safe drinking water standard for pregnant women and children. On March 11, 2004, the California Department of Health Services (DHS) published a public health goal (PHG) of 6 m g/L for perchlorate in drinking water.3 The PHG is the first step to establishing a maximum contaminant level (MCL). California is expected to set a final MCL of 4 m g/L to 10 m g/L for perchlorate in another two to three years. Establishing MCLs this low may necessitate groundwater treatment for a large number of aquifers in Southern California and for some surface water sources such as the Colorado River where clean, uncontaminated groundwater is not readily available for blending to reduce perchlorate concentrations to below the proposed MCLs.

The Perchlorate Problem
The perchlorate ion is a high-energy molecule, comprised of one chlorine atom surrounded by four oxygen atoms. The chlorine atom is in its most oxidized state (+7) and the anion carries an overall negative charge of -1. Perchlorate salts are very soluble in water, with some salts showing solubilities that exceed 2 kilograms per liter (kg/L).

A poor complexing agent, the perchlorate ion is very mobile when released into the environment and tends to migrate within an aquifer at about the same rate as the groundwater moves. Although perchlorate is very reactive thermodynamically, it is kinetically very stable at low concentrations in groundwater and surface water under ambient conditions.4

Perchlorate salts are used in many military applications such as solid fuels for rockets and missiles, military and commercial munitions, explosives, and blasting agents. They are also used in many industrial applications such as in the manufacture of matches, additives in lubricating oils, leather finishing, electroplating, aluminum refining, rubber manufacturing, fertilizers, and in the production of paints and enamels. The only known natural source of perchlorate has been found in potassium nitrate fertilizers derived from bird guano in Chile.

EPA has identified perchlorate users in 44 states and significant releases to the environment in 23 of these states.5 The agency has conducted numerous treatability studies, in conjunction with universities and remediation service providers, to develop and demonstrate various technologies for removing perchlorate from water.

The cost-effective use of existing technologies to remove low concentrations of perchlorate from groundwaters and surface waters has proven to be somewhat challenging. Technologies being used and further developed include ion exchange resins, reverse osmosis (RO) membranes, electrodialysis, and tailored activated carbons. Of these technologies, tailored activated carbon is one of the most recently developed. To-date testing results indicate these carbons are extremely cost-effective in treating water that contains low concentrations of perchlorate.

Tailored Activated Carbons
Water utilities have used granular and powdered activated carbons (GAC and PAC, respectively) for more than a half-century to remove organic contaminants from potable water. Worldwide use of activated carbon has increased steadily over time, and today in the United States activated carbon use exceeds 210,000 metric tons annually.6

Traditionally, GAC is used to remove organic contaminants, especially those that are toxic when consumed (pesticides, herbicides, industrial solvents, disinfection byproducts, etc.) or that degrade the water's aesthetics and cause taste, odor, or color problems (geosmin, 2-methylisoborneol, humic acids, etc.). Activated carbons are generally very cost-effective when used in water treatment applications, especially when the contaminants to be removed are non-polar in nature and show low water solubility. However, activated carbons' ability to remove contaminants decreases rapidly as the solubility and polarity of the contaminant increases, as well as when inorganic salts needing removal are present.7

The graphitic plates that make up the structure and contribute to the high surface area of activated carbons are in general very clean and contain few functional groups. This lack of functional groups means activated carbons are very hydrophobic in nature. However, the few functional groups that do exist give the surface a slight charge and provide the carbon with a small amount of ion exchange capacity.

Although virgin activated carbons are generally considered ineffective for removing such inorganic ions as perchlorate from water, laboratory studies done by Dr. Fred Cannon and his associates at Penn State University have demonstrated these carbons have some capacity for adsorbing perchlorate.8 During the initial phases of an ongoing study, the Penn State researchers used a coal-based GAC to treat 80 parts per billion (ppb) of perchlorate-contaminated groundwater from the city of Redlands, Calif. They were able to treat approximately 1,100 bed volumes of water before breakthrough was detected.9 While Redlands was satisfied with the results for use in an emergency situation, few bed volumes were actually treated before the carbon was exhausted. At this rate, water treatment would cost more than $800 per acre-foot.

To circumvent the high costs typically associated with low perchlorate capacity, Penn State researchers began working on a number of different processes in 2000 that would significantly increase the GAC's perchlorate-adsorption capacity. The American Water Works Association Research Foundation and the U.S. Department of Energy Consortium for Premium Carbon Products from Coal funded much of the university's research on the use of activated carbons for perchlorate removal, with USFilter Westates providing in-kind support.

Penn State's research eventually led to a new approach to the capacity problem -- the development of a tailoring process for treating the activated carbons with a quaternary amine. This tailoring approach uses the activated carbons' high surface area as a low-cost backbone for creating ion exchange sites, instead of using the costly polymers associated with standard ion exchange resins. Tailored carbons made by this patent-pending process have been extensively tested by Penn State for their ability to remove perchlorate from the city of Redlands' groundwater. Case in point, the GAC's perchlorate capacity increased from about 1,100 bed volumes (using the virgin carbon) to about 34,000 bed volumes using the tailored material yet to be National Science Foundation (NSF) certified for use in drinking water. This large increase in perchlorate capacity reduced the projected cost of treating Redlands' water from more than $180 per acre-foot using ion exchange resins to between $100 and $125 an acre-foot (See Table 1).

Working with Penn State, USFilter Westates conducted a series of rapid small-scale column tests (RSSCT) on groundwater from several contaminated sites in Southern California.10 Both virgin and tailored carbons were used in each test. Results from these site tests, as well as from the Redlands and an East Coast test, are given in Table 1.

As a result of the RSSCTs, the research team concluded that tailored carbons using a non-NSF-certified tailoring agent with water similar to what is at the Redlands site are as, if not more, cost-effective than ion exchange in removing perchlorate from drinking water. Overall water chemistry affects which technology works best for a specific application. The presence of other ions such as nitrate, sulfate and carbonates, total dissolved solids (TDS), water pH, and the presence of organic and other contaminants significantly impact the cost of perchlorate removal Table 2.

Ion exchange resins hold the most promise for handling a wide variety of water chemistries. They can be custom manufactured to have a high selectivity for the perchlorate ion, and can therefore successfully treat groundwaters that contain high concentrations of other ions. Building this high selectivity into the resin increases its capacity as well as its cost of manufacture. These new high selectivity resins have not been tested under high flow conditions for the amount of time that would be required for them to reach their full loading capacity. In light of the high initial cost for these resins, shorter than expected service lives due to compaction or the resin's physical deterioration could significantly increase treatment costs.

The fluidized bed reactor (FBR) offers some of the lowest treatment costs. Relying on microorganisms to degrade the perchlorate, the FBR requires that nutrients be added. Depending on the system loading, the technology can be sensitive to sudden changes in contaminant concentrations, water temperature; and flow rates -- and generally has low operating costs.

Reverse osmosis (RO) membranes effectively remove ionic species from water. Even though significant progress has been made in developing more effective and lower cost membranes, water produced by this technology costs more than water produced by other methods. Nanofiltration membranes are also capable of removing perchlorate by establishing a charge differential on the membrane for perchlorate selectivity. The cost differential for treatment by membranes is largely due to the membranes' high initial capital and annual operating costs.

The use of tailored carbons for perchlorate remediation is fairly recent, and further studies need to be done to determine which applications are the most appropriate for this technology. The effects other present anions (nitrate, carbonate, and sulfate) and water chemistry have on the tailored carbons' performance are also not yet fully understood.

However, in applications where dissolved organic contaminants are present, tailored carbons can minimize both system costs and complexity. In the same adsorber, the tailored carbons can remove both the dissolved organics and perchlorate. The tailoring process does reduce the activated carbons' adsorption capacity of organic contaminants. For a particular site, the amount of tailoring can be adjusted so that breakthrough of perchlorate and dissolved organics occurs at the same time, maximizing the performance of the system and minimizing costs.

The Future of Tailored Carbons
Recent studies have demonstrated that tailored carbons may be used cost-effectively to remove perchlorate ions from groundwaters and surface waters. How widespread the carbons' use becomes remains to be seen.

Gaining acceptance from the state DHS and NSF for using tailored carbons for potable water treatment applications promises to be challenging. Before the DHS accepts this technology, the tailored carbons or tailored carbon systems (tailored carbon followed by a scavenger bed of virgin carbon) must be shown to comply with all the requirements of American National Standards Institute (ANSI)/NSF Standard 61. Standard 61 establishes minimum requirements for the control of potential adverse human health effects from products that come in contact with drinking water. While this standard does not address product performance requirements, it does set appropriate standards for metals and organic leachability as well as other requirements established by the Food Chemical Codex and ANSI/American Water Works Association (AWWA) Standard B-604. Municipalities may also be required to conform with ANSI/NSF Standard 53, as tailored carbon is a new material being introduced into water treatment systems.

Answers to many of the issues concerning the use of tailored carbons in potable water applications will soon be available. By the end of the year, USFilter Westates, Penn State, the city of Redlands, and their other research partners will complete full-scale testing of a closely monitored demonstration project, as part of a study at the Texas Street pumping station in Redlands, Calif.

1. Samples containing low concentrations of perchlorate can be easily analyzed using a Dionex DX-120 equipped with an AS-16 column, in accordance with EPA method 314.0. A 4,000-microliter sample loop is used for the semi-quantitative detection to a concentration of ~ 0.2 ppb and rigorously quantitative to 0.5-1.0 ppb. This adaptation of method 314.0 has been recommended and approved by EPA staff.

2. Shanahan, P.; Groff, K.; and Wilson, J. August 2003. Risk Analysis, Communication, Evaluation and Reduction at LANL -- Perchlorate in Groundwater. RAC Report No. 1-RACER LANL-2003-Final.

3. California Department of Health Services, 2004. Perchlorate Action Level, www.dhs.ca.gov/ps/ddwem/chemicals/perchl/actionlevel.htm.

4. Gullick, R. W.; M. W. LeChevallier; and T. S. Barhorst, 2001. "Occurrence of Perchlorate in Drinking Water Sources." Journal AWWA, 93:1: 66-77.

5. (U.S. EPA 2001)EPA, 1999. Method 314.0 Detection of Perchlorate in Drinking Water Using Ion Chromatography, www.epa.gov/ogwdw/methods/met314.html.

6. The Freedonia Group, Inc., November 2000. Industrial Study 1355 on Activated Carbon.

7. Patrick, J. W., 1995. Porosity in Carbons. New York, John Wiley & Sons, Inc.

8. Cannon, F. and N. Chongzheng, 2000. Perchlorate Removal Using Tailored Granular Activated Carbon. American Water Works Association Research Foundation. Denver.

9. A bed volume is equal to the volume in an adsorber vessel that is occupied by the GAC. For example an adsorber that holds 20,000 pounds (lbs.) or 668 cubic feet (ft3) of GAC would have a bed volume equivalent to 668 ft3 or 5,000 gallons of water.

10. Crittenden, J. C.; P. S. Reddy; D. W. Hand; & H. Arora, 1989. Prediction of GAC Performance Using Rapid Small-Scale Column Tests. American Water Works Association Research Foundation and American Water Works Association.

Table 1. Cost Comparison Table for Perchlorate Removal Technologies

Perchlorate Removal Technology

Estimated Cost per Acre-Foot*

Tailored Granular Activated Carbons

$100 - $175

Ion Exchange

$100 - $350 (non-regenerable)

$150 - $500 (regenerable)

Fluidized Bed Reactor

$100 - $175

Reverse Osmosis Membranes

Costs depend on TDS of water -- generally $300 - $800

* One acre-foot is 326,000 gallons of water, or enough to supply a family of four with enough water for two years.

Table 2. Prediction of GAC Performance Using Rapid Small-Scale Column Tests

Perchlorate Conc.

m g/L

Other Conditions

Virgin GAC

Bed Volumes

Tailored GAC

Bed Volumes

Estimated Cost Per Acre-Foot*

Using Tailored Carbon


70 - 85



$100 - $125


70 - 85



$500 - $800

Military Base

13 - 15

High Sulfate, High TDS



$250 - $400



High Sulfate and Carbonate



$600 - $900

East Coast Site

1 - 5

Soft Water,




$10 - $15

* One acre-foot is 326,000 gallons of water, or enough to supply a family of four with enough water for two years.
**Tailoring agent not NSF certified for use in drinking water
***Tailoring agent NSF certified for use in drinking water

This article originally appeared in the 03/01/2005 issue of Environmental Protection.

About the Author

James R. (Jim) Graham, PhD, is technical director for USFilter Westates and is based out of Santa Fe Springs, Calif. He has more than 30 years of industry experience in research, development, and project management. Graham has authored numerous publications and holds patents in the areas of GAC applications for air and water treatment, catalysis, adsorption, and coal conversion chemistry. He is past chairman of the American Society for Testing Materials (ASTM) D28 Committee on activated carbon, a position he held for six years. Graham holds a BS degree in chemistry and mathematics from the Indiana State University in Terre Haute and a PhD in inorganic chemistry from the Iowa State University in Ames.