This demonstration was designed to evaluate the technical effectiveness of in situ bioremediation as a treatment technology for explosives in groundwater at the Picatinny Arsenal in Dover, New Jersey. A recirculation cell design with semi-passive operation was employed to distribute and mix cosubstrate with contaminated groundwater in order to promote the biodegradation of nitramine and nitroaromatic explosives by indigenous bacteria. Cheese whey was utilized as a cosubstrate based on extensive treatability testing. The overall performance of this design for remediation was determined during the demonstration. The impacts of the technology on the geochemistry of treated groundwater also were evaluated. In addition to technical performance, the demonstration provided the capital and operation and maintenance (O&M) costs of this type of system at a scale that can be extrapolated to different full-scale designs.

This project builds on recent microbiological research suggesting that explosives-degrading bacteria are widespread but that they require one or more cosubstrates to completely degrade most nitramine and nitroaromatic explosives. During the demonstration, a groundwater extraction-reinjection (ER) system was installed to distribute and mix cheese whey as a cosubstrate with explosives-contaminated groundwater in the subsurface. The system, consisting of two extraction wells and a single injection well, was operated in a semi-passive mode, pumping for 3-5 days during injection of soluble cheese whey constituents (“active” phase) and then shut down for 6-12 weeks (“passive” phase) once adequate mixing and distribution of the whey was achieved. The cheese whey was added in four active cycles during the initial 6 months of operation. A total of 830 kg of cheese whey was added during these cycles (dissolved constituents only), and the system was operated at approximately 38 liters per minute flow. The final groundwater sampling event was conducted more than a year after the final active cycle. This approach facilitated modification of the aquifer geochemistry to enhance subsurface biodegradation of energetic compounds by indigenous bacteria while minimizing system O&M issues due to biofouling.

The primary performance objective of this demonstration was to reduce explosives in groundwater at Picatinny to concentrations below regulatory concern. For 2,4,6-trinitrotoluene (TNT) and hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX), the U.S. Environmental Protection Agency has issued Lifetime Health Advisory Limits (maximum contaminant goal level [MCGL] Values) of 2 micrograms per liter (μg/L), and the equivalent value for octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine (HMX) is 400 μg/L. The New Jersey Department of Environmental Protection also issued Interim Groundwater Quality Criteria for both RDX and TNT in 2008. The specific criteria are 0.3 μg/L and 1 μg/L for RDX and TNT, respectively. The key performance objective for this demonstration was achieved.

Concentrations of TNT in the treatment zone monitoring wells (TZMWs) declined rapidly after cheese whey injection. Initial concentrations ranged from 5 to 190 μg/L during the final baseline sampling event. The concentration of TNT was below the analytical detection limit (practical quantitation limit [PQL] = 0.25 μg/L) in all of the TZMWs by Day 62 of the study and remained at or below this concentration in all TZMWs except one throughout the remainder of the 565-day demonstration. RDX concentrations in the TZMWs ranged from 5 to 170 μg/L during the final baseline sampling event, with a mean value of 66 μg/L. RDX loss occurred somewhat more slowly than for TNT, but 148 days after the initial injection of cheese whey, RDX concentrations were less than 5 μg/L in all six of the TZMWs, and concentrations in five of these wells were less than 1.5 μg/L. From Day 222 to Day 565, the concentration of RDX in all of the downgradient TZMWs was less than 1 μg/L, and all were less than 0.2 μg/L on Day 565. Thus, more than 1 year after the final injection of cheese whey on Day 181, RDX was less than 1 μg/L throughout the downgradient region of the treatment plot. A significant decline in HMX was also observed in all wells, and by Day 274, each of the four downgradient TZMWs had HMX concentrations less than 0.4 μg/L (from a starting mean concentration of 50 μg/L). A slight rebound was observed in one downgradient TZMW on Day 565, but HMX remained less than 1 μg/L in each of the other wells throughout the remainder of the study. Thus, as with RDX and TNT, the data from the downgradient TZMWs indicate that the addition of cheese whey to the Picatinny aquifer effectively promoted HMX biodegradation to sub μg/L concentrations.

Overall, this in situ bioremediation approach proved to be highly effective for the treatment of nitramine and nitroaromatic explosives in groundwater. The applicable regulatory guidance and/or action levels were achieved for RDX and TNT, there was no significant accumulation of degradation intermediates, and the active-passive treatment approach resulted in no significant O&M issues. Moreover, after only four active injection cycles, concentrations of total organic carbon (TOC) from the cheese whey remained high enough in downgradient monitoring wells to promote degradation of explosives and intermediates for more than a year after the final injection. The data showed that, as long as TOC concentrations greater than approximately 5 milligrams per liter (mg/L) were maintained, rebound of explosives was negligible. Thus, this project clearly shows that in situ bioremediation of explosives in groundwater using active-passive cosubstrate addition can be a viable long-term treatment approach. This technology is expected to be widely applicable at military installations across the United States.