Processes impacting the dissolution of dense nonaqueous phase liquids (DNAPLs) in fractured bedrock are not well understood, thereby limiting efforts to predict the longevity and persistence of DNAPL sources, select appropriate remedial technologies, effectively design and implement in situ technologies, evaluate remedial performance, and estimate the time needed to remediate a DNAPL source zone.  The rate of DNAPL dissolution is typically controlled by interfacial mass transfer between the DNAPL and surrounding groundwater, especially during implementation of in situ remediation technologies such as chemical oxidation and bioaugmentation.  Thus, changes in the effective DNAPL-water interfacial area in bedrock fractures due to dissolution of the DNAPL source will have a substantial impact on the overall remedial process.  The presence of oxidation precipitates, biofilm growth, and dead-end fractures also can impact overall DNAPL dissolution rates.  However, comprehensive studies have never been performed to evaluate and quantify these phenomena in fractured bedrock systems.

This project focused on measuring and evaluating the architecture, dissolution rate, and impact on groundwater quality of residually trapped tetrachloroethene (PCE) DNAPL from discrete bedrock fractures and fracture networks.  Treatments using both chemical oxidation and bioaugmentation were evaluated in order to compare their relative effectiveness and identify potential limitations in fractured settings.

The specific objectives of this project were as follows.

• Determine residual DNAPL saturation, distribution, and interfacial area characteristics in various bedrock fractures and fracture networks.
• Measure and evaluate DNAPL-water interfacial area and dissolution rates as a function of DNAPL saturation within the fractures.
• Measure and evaluate DNAPL dissolution kinetics during chemical oxidation and bioaugmentation treatment in individual bedrock fractures and fracture networks.
• Identify potential limitations (e.g., fouling due to chemical precipitation or biological growth, DNAPL in dead-end fractures) of chemical oxidation and bioaugmentation in bedrock fractures.
• Evaluate the impact of partial DNAPL removal on groundwater quality.
• Develop and validate models to describe and predict DNAPL dissolution.

The overall approach for this project consisted of constructing laboratory fractures at two scales: a single fracture scale and a fracture network scale.  Single fracture systems were created by inducing a single fracture in sandstone blocks.  Fracture network systems, which contained multiple vertical and horizontal fractures, were constructed by assembling sandstone blocks; points of contact between the blocks defined the fracture network.  Once constructed, the experimental fracture systems were contaminated with residual PCE DNAPL.  Using conservative and interfacial tracers, the DNAPL architecture was assessed.  In addition, dissolution experiments were performed to evaluate DNAPL dissolution kinetics.  DNAPL dissolution also was evaluated during application of bioaugmentation, and during application of chemical oxidation using both permanganate and persulfate.

Results from both the single fracture and fracture network experiments demonstrated that, compared to unconsolidated systems, residual DNAPL is not well contacted by migrating water.  This resulted in reduced dissolution rates, and persistence of DNAPL sources within the bedrock fractures.  While DNAPL dissolution during application of bioaugmentation was hindered by these intrinsic mass transfer limitations, results showed that bioaugmentation was effective at substantially enhancing the rate of DNAPL removal, despite dissolved PCE concentrations that were at or near solubility.  Microbially-enhanced dechlorination rates during bioaugmentation were well described by a modified version of our previously developed bioaugmentation model.  Results also showed that removal of a relatively small fraction of the DNAPL mass resulted in a large decrease in the dissolved PCE concentration.

Chemical oxidation was shown to be ineffective for treating DNAPL sources in bedrock fractures in both the single fracture and fracture network experiments.  This ineffectiveness was due to decreases in the effective DNAPL-water interfacial area (as measured using interfacial tracers) that were likely caused by oxidation reaction byproducts.

This research demonstrated the difficulty in contacting DNAPL sources in fractured bedrock, and highlighted the importance of mass transfer processes even at the single fracture scale.  However, results show that only partial removal of the DNAPL mass may be needed to sufficiently improve groundwater quality.  In terms of evaluating treatment technologies, results of this study showed that bioaugmentation is a potentially effective remedial approach for DNAPL sources in fractured rock, and likely is a better long term option than either persulfate or permanganate oxidation.