Perchlorate-contaminated drinking water is of concern within the Department of Defense (DOD), as ammonium perchlorate is a component of explosives and solid rocket propellants used at numerous DOD installations. Perchlorate (ClO4-), a negatively charged anion which does not readily degrade, has been identified in groundwater, surface water, soil and food. Ion exchange (IX) is currently the most frequently used method for the treatment of perchlorate-contaminated drinking water.
In addition to perchlorate contamination, contamination of drinking water by organic compounds also is quite common and of great concern. The Environmental Protection Agency (EPA) regulates many of these organic compounds, including chlorinated solvents like trichloroethylene, tetrachloroethylene, vinyl chloride and carbon tetrachloride. Within DOD, these solvents are widely used for cleaning and metal degreasing. DOD also is concerned with cleaning up water contaminated by nitroaromatic compounds like HMX (1,3,5,7-tetranitro-1,3,5,7-tetrazocane), RDX (hexahydro–1,3,5– trinitro–1,3,5–triazine) and TNT (2,4,6-trinitrotoluene), which are used as explosives. Granular activated carbon (GAC) is conventionally used to treat water contaminated with these non-polar organic contaminants.
Many contaminated groundwater sites have plumes with perchlorate comingled with chlorinated solvents or nitroaromatics. The Massachusetts Military Reservation, Stringfellow Superfund Site and the National Aeronautics and Space Administration Jet Propulsion Laboratory are a few of the larger sites having perchlorate plumes with organic co-contaminants. At these sites, the conventional approach is to use a train of treatment technologies, with IX to remove perchlorate and, typically, GAC to remove the organics. The use of two different systems is often expensive and requires that personnel have the training, supplies, and equipment to operate both systems.
Tailored Granular Activated Carbon
Tailored granular activated carbon (T-GAC) is an innovative technology developed at Pennsylvania State University (PSU) that is currently being evaluated to determine if it can perform better than conventional IX to treat perchlorate-contaminated water. T-GAC is GAC that has been amended with positively charged surfactants commonly found in personal care products such as mouthwash. Tailoring gives the GAC particles a positive charge, which allows them to attract and remove from the water the negatively-charged perchlorate anions.
To evaluate the efficacy of using T-GAC to treat perchlorate-contaminated water, tests were conducted at three scales. At the smallest scale, in the PSU environmental engineering laboratory, researchers conducted rapid small-scale column tests (RSSCT), which have been used for decades to help design GAC systems at water treatment plants.
At the two larger scales, testing was done on-site at the Fontana, Calif., water treatment facility. The groundwater treated by the Fontana Water Company (FWC) is contaminated with perchlorate. Currently, FWC uses conventional IX technology to remove perchlorate from the water. DOD’s Environmental Security Technology Certification Program (ESTCP) funded a project to evaluate whether T-GAC would be a more cost-effective technology for treating the perchlorate-contaminated water. As part of the project, 1.5-gallon per minute (gpm) and 38-gpm pilot-scale tests were conducted at FWC. The 1.5-gpm tests were run with six sets of T-GAC columns.
FWC water was spiked with various constituents to see how the T-GAC treatment technology operated under varying influent water quality conditions. For example, one of the six columns had influent water that was spiked with TCE to see how effective the column was in removing both perchlorate and TCE. At the largest, 38-gpm scale, raw untreated water from the FWC wells was run through two T-GAC contactors followed by a GAC contactor. ARCADIS conducted the pilot-scale studies at FWC, using T-GAC systems from Siemens. Details of the project are in ESTCP Final Report for Project ER-0546.
Cost and Performance Analysis
Based on the experimental results at all three scales, combined with theory that was developed to predict GAC performance at water treatment plants and EPA models of costs for GAC treatment of water, researchers at the Air Force Institute of Technology developed an Excel-based screening program that combines cost and performance sub-models. The program uses input from a system designer. The designer must input both design criteria and economic assumptions such as the flow rate of water to be treated; the influent concentrations of perchlorate and any organic co-contaminants that might also be present; the influent concentration of other anions, like nitrate, which compete with perchlorate for adsorption sites on the T-GAC; and the discount rate and amortization period.
Based on the input, the performance sub-model uses a Freundlich adsorption isotherm that accounts for competition among anions to predict the volume of perchlorate-contaminated water that can be treated before perchlorate begins to breakthrough in the effluent of the T-GAC column. The performance sub-model also predicts the volume of water that can be treated before organic co-contaminants break through a GAC column, which operates in series with the T-GAC column. These break-through volumes are used to calculate the carbon usage rate (mass of carbon used per volume of water treated), a key parameter that is automatically entered as input into the cost sub-model of the Excel screening program.
Applying the carbon usage rate calculated by the performance sub-model with other inputs from the designer, the cost sub-model provides an estimate of capital and annual operating costs. Then, using standard engineering economic analysis techniques, the cost sub-model estimates costs to treat an acre-foot of water, which may be readily compared with the published costs of alternative treatment technologies.
Craig (2008) used the screening program to investigate the sensitivity of T-GAC treatment costs to various engineering and influent water-quality parameters. His study showed, for example, that the presence of nitrate at high concentrations as a co-contaminant led to significantly increased costs due to the fact that nitrate competes with perchlorate for adsorption sites on the T-GAC, resulting in early perchlorate breakthrough. Using the program, the study also demonstrated that when comparing conventional IX with T-GAC to treat perchlorate-contaminated waters, T-GAC was not as cost-effective. This is largely due to the fact that the cost per acre-foot of IX treatment has dramatically decreased in the past 10 years.
Downen (2009) extended Craig’s study by examining how cost effective a T-GAC system would be in treating water with comingled perchlorate and organic contaminants. The study compared the costs of T-GAC and GAC systems predicted using the Excel screening program with cost data obtained from sites where comingled plumes are currently being treated using conventional treatment technologies, which typically involve a combination of IX and GAC in series. Results from Downen’s study indicated that a combined T-GAC/GAC system may be more cost effective for the treatment of these comingled plumes than conventional treatment. An additional benefit of the T-GAC/GAC system is that GAC is already a widely accepted and understood technology. Site personnel are familiar with operating and maintaining GAC systems. Utilizing T-GAC will require no further training and will allow for utilities that already use GAC to easily transition to T-GAC.
The screening program discussed in this article has the ability to help with technology transfer, as it is simple to use and allows a designer with known influent water chemistry and site characteristics to determine a treatment cost per acre-foot of perchlorate- and organic-contaminated water. Further research is needed to validate this model, but the technology evaluation has shown that a T-GAC/GAC system has the potential to cost-effectively treat waters at the many sites that have perchlorate contamination co-existing with organic contaminants.