Bioremediation/natural attenuation continues after ISCO treatment
Scientists and engineers from the private sector and EPA are evaluating the potential for in situ bioremediation and/or natural attenuation to successfully treat residual soil and ground-water contamination after source-area in situ chemical oxidation (ISCO) treatment. The studies were prompted by concerns that the two treatment technologies may be incompatible due to detrimental impacts on microbial communities as well as decreases in aquifer permeability resulting from ISCO. Recent field evaluation at a former manufacturing site in Framingham, MA, and other investigations indicated that adverse impacts from ISCO were not permanent and did not affect subsequent enhanced in situ bioremediation or natural attenuation.
An approximate 2,000-ft2 TCE plume existed at the Framingham site as a result of past activities involved with electronics manufacturing. The plume contained TCE in high concentrations of approximately 22,000 ppb and 1,000 ppb of cis-DCE and vinyl chloride, respectively, and was situated 20-35 feet bgs within layered fine sands and silts. The water table ranges from 8 to 12 feet bgs. The presence of daughter products indicated that reductive dechlorination was active.
In December 2001, injection of 450 gallons of a 20% solution of sodium permanganate (NaMnO4) was performed through a single injection well screened at 24-34 feet bgs within the suspected source area. A total of 750 pounds (dry weight) of NaMnO4 was injected over two days. Permanganate was detected at least 30 feet downgradient of the injection point.
Analyses of ground water at downgradient monitoring wells showed that permanganate persisted in the aquifer for approximately six months following injection. The wells also showed an immediate reduction in TCE followed by temporary rebound attributed to slow mass transfer and mass transport mechanisms. Over the following 1.5 years, TCE concentrations steadily declined to 5,000 ppb, and a small increase in daughter product cis-DCE was observed, showing that reductive dechlorination activity had rebounded.
Bioremediation efforts were initiated in 2003, 30 months after the ISCO treatment, to accelerate degradation of the contaminant plume through bacterial processes. A total of 1,026 pounds of sodium lactate was administered over 1.5 years during five injection events in two of the monitoring wells located in the ISCO treatment area and downgradient of the ISCO injection well. Laboratory analysis of ground-water samples collected from treatment wells two years after the start of sodium lactate additions indicated TCE concentrations had decreased to approximately 2,500 ppb. Cis-DCE and vinyl chloride levels also increased following lactate injection.
NaMnO4 and other ISCO oxidants such as hydrogen peroxide, permanganate, and ozone exhibit antiseptic properties with potential for inhibiting or killing microbial organisms involved in subsurface biotic processes, particularly under anaerobic conditions. Oxidant injection results in a significant increase in redox potentia that can interfere with reducing conditions needed under anaerobic conditions, and consequently inhibit microbial activity and contaminant transformations. To evaluate these impacts more closely, microbial communities were assessed at another site, where approximately 60,000 pounds of NaMnO4 and potassium permanganate (KMnO4) were injected into a two-acre area over one year. Bio-Trap® devices were deployed in monitoring wells for passive, in-situ collection of microbes in ground water over an extended time.
Microbial sampling and PLFA analysis of three wells (a few months after and four years after ISCO) indicated an increase in post-oxidation biomass levels in monitoring wells impacted by permanganate, when compared to upgradient ground water representative of background conditions (Figure 3). PLFA data from permanganate-impacted wells also indicated a complex consortium of microbes including aerobes, anaerobes, and metal- and sulfate-reducing bacteria. Collectively, this consortium is capable of degrading hydrocarbons and enhancing reductive dechlorination of CVOCs.
No loss in permeability was observed at the Massachusetts facility, in contrast to common assertions that reduction of permanganate ion consistently leads to manganese oxide (MnO2) accumulation in porous media and associated reduction in permeability and treatment efficacy. A critical analysis of MnO2(s) accumulation in porous media indicated that MnO2(s) precipitation filled approximately 8% of the void volume in a porous medium with an oxidant demand of 50-60 g/kg. Other mechanisms such as MnO2(s) particle transport and filtration, mechanical straining, electrostatic interactions, chemical bridging, or specific adsorption may cause immobilization of MnO2(s) particles within porous media and lead to changes in permeability.
Study of ISCO applications at multiple sites showed that KMnO4 injections typically involve concentrations of 2-3 g/L, which are below the solubility (6.5 g/L at 20°C), but are sensitive to temperatures. For example, differences in temperature between the KMnO4 solution in a mixing tank and the (cooler) aquifer can result in precipitation of KMnO4 in the aquifer, causing rapid but temporary permeability reduction. In addition, insufficient pre-injection mixing duration or methods can produce an injection solution containing a significant quantity of KMnO4 particulates. Study findings indicated that although accumulation of KMnO4 particulates in the aquifer can decrease permeability, the effect typically is temporary due to dissolving of KMnO4 over time. Similarly, generation of carbon dioxide gas caused by microbial activity or oxidation of organics can temporarily decrease permeability before the gas dissolves in water.
Results at the Massachusetts site and two other sites gave no indication that ISCO resulted in sterilization of aquifer material or permanent inhibition of microbial activity. In subsurface systems, contact between oxidant and microbial populations can be limited due to preferential pathways and microniches, allowing microbiota to survive rigorous applications of oxidant. Spatial separation between oxidant injection into source areas and downgradient microbially active areas also diminishes the impact of the oxidant. Potential for successful bioremediation and/or natural attenuation following ISCO applications may vary with oxidant, physical and chemical characteristics of subsurface materials, or varying contaminant concentrations and characteristics.