Background

Ammonium perchlorate (NH4ClO4) has been used for several decades in the United States as an oxidizer in solid propellants and explosives. This compound and other perchlorate salts also are present in various commercial products, such as fireworks, safety flares, and matches. Discharges during the manufacture of perchlorate salts and from the periodic replacement of outdated solid fuels in military missiles and rockets have resulted in substantial perchlorate contamination in soils and groundwater in several states, including California, Texas, Utah, Maryland, and Nevada. Although as of 2010 there is no federal drinking water standard (maximum contaminant level; MCL) for perchlorate, the U.S. Environmental Protection Agency has issued a reference dose (RfD) of 0.7 μg perchlorate/kg body wt/day, which corresponds to a drinking water equivalent level (DWEL) of approximately 24.5 μg/L. In addition, California has established a state MCL of 6 μg/L and Massachusetts has set a drinking water MCL of 2 μg/L. Several other states, including Nevada, Maryland, New York, and Texas, also have instituted advisory levels for the oxidant.

While there has been a significant effort to develop practical remediation technologies for perchlorate in groundwater, there has been less consideration of perchlorate treatment in unsaturated soils, particularly in deep vadose zone soils that overlie many groundwater perchlorate plumes. Residual perchlorate contamination within vadose zone soils in source areas such as hog out operations, open-burn open-detonation areas, live fire ranges, and ammonium perchlorate production and fine grinding facilities continues to pose an ongoing threat to groundwater. Previous laboratory and field studies have demonstrated successful biological treatment of perchlorate-laden surface soils by addition of manure composting or injection of gaseous electron donors. Phytoremediation also has been tested for soil treatment. However, these approaches may not be cost-effective or feasible for sites with deep groundwater and high levels of residual perchlorate contamination within vadose zone soils. Due to the high solubility of perchlorate, these impacted soils will continue to act as a source of contamination to underlying groundwater aquifers. This ongoing source will increase the operating time frame and associated costs for hydraulic containment (pump and treatment) and in situ groundwater treatment systems.

Objective

The primary objective of this project was to demonstrate and validate the treatment of perchlorate within vadose zone soils through bioremediation and flushing via two electron donor delivery methods: Treatment #1, the infiltration of liquid electron donor using an engineered infiltration gallery; and Treatment #2, the addition of an electron donor source to the upper soil column and periodic watering to promote vertical distribution within the vadose zone. For Treatment #1, an engineered infiltration gallery was to be designed to effectively deliver and distribute the electron donor to the perchlorate-impacted vadose zone soils. For Treatment #2, a complex electron donor source such as emulsified vegetable oil or cow manure was to be mixed into the upper meter of the soil and an automated sprinkling system designed to supply water over the test plot area and promote vertical penetration of the electron donor agent along with the infiltrating waters.

Technology Description

Due to underground injection control permitting requirements, it was expected that Lake Mead water (unchlorinated) would be used during this project to provide water to both treatment plots. Lake Mead water was to be collected and amended with electron donor (for Treatment #1 only) and, if necessary, other amendments (e.g., pH buffering agents or nutrients). The most effective electron donor for each treatment method was determined in microcosm studies. Quantities of the lake water were to be directed to the infiltration gallery and the surface treatment areas, which would release the water into the vadose zone soils causing a vertical spreading of the electron donor and amendments throughout the vadose zone. Soil moisture probes, suction lysimeters, and a network of groundwater monitoring wells were proposed to measure the variability and flux of perchlorate within the vadose zone and groundwater (wells) throughout the demonstration. Two suction lysimeter nests were to be installed in each treatment plot (total of 4 nests) with screens set in the upper, middle, and lower thirds of the vadose zone for monitoring purposes. Additionally, soil moisture probes were proposed to be installed at approximately 10 and 20 ft below ground surface (bgs) in order to observe the propagation of the wetting front. The rate of fluid infiltration was to be varied within each treatment area during the study and the corresponding effect on contaminant mobilization and degradation rates monitored.

Demonstration Results

A site selection process was implemented, and from the results it was determined that Tronox LLC, Henderson, Nevada (Tronox) was the most appropriate area to perform the vadose zone field demonstration. A laboratory microcosm study was conducted to determine the most effective electron donor(s) for perchlorate treatment in the vadose zone of the Tronox demonstration site. A representative subset of the vadose zone soil samples collected during the investigation work completed at Tronox in January 2008 were used for this study. After field collection, the split spoon soil samples from each core were shipped to the Shaw treatability laboratory in Lawrenceville, New Jersey, for analysis of perchlorate concentrations.

The results of this microcosm study showed that three liquid amendments (EOS, ethanol, and citrate) were effective for promoting biological degradation of nitrate and perchlorate. Among these amendments, EOS resulted in the fastest and most consistent biodegradation of the target anions. At the conclusion of the study, perchlorate concentrations in the EOS-treated samples were consistently less than 0.3 mg/kg. Previous studies have shown that EOS can be an effective substrate for promoting perchlorate biodegradation in groundwater aquifers. However, this study apparently was the first to show that EOS can be effective in unsaturated vadose zone soils and particularly in soils with perchlorate concentrations exceeding 1,000 mg/kg. Based on the laboratory results, EOS was recommended for use in the engineered infiltration gallery design (Treatment # 1).

Several solid (or solid/liquid combination) amendments also were effective for stimulating perchlorate biodegradation in the Tronox vadose zone soils, including soybean oil with peat moss, bioreactor sludge with acetate, and cheese whey. Among these substrates, the former two mixtures resulted in the most rapid and consistent perchlorate biodegradation. Based on the laboratory results, a mixture of soybean oil and peat moss was recommended for use in the planned surface amendment field test (Treatment # 2).

Unfortunately, the project could not proceed to the field phase.

Implementation Issues

The field component of this project at Tronox was cancelled due to scheduling and other issues, but the laboratory studies completed to support this effort certainly suggest that perchlorate treatment in deep vadose zone soils is feasible if good amendment distribution can be achieved.