Installation Date:
03/01/1998

Contaminants:
Trichloroethene

Reactive Media:
Fe⁰

Cost:
$1,120,000

Construction:
Hydraulic Fracturing, Permeation Infilling

Point of Contact:
John Vidumsky
DuPont Specialty Chemicals
Tel: 302-892-1738
Fax: 302-892-7641
Email: john.e.vidumsky@
usa.dupont.com
Barley Mill Plaza, 27/2226
Lancaster Pike and Route 1
Wilmington , DE 19805

A full-scale permeable reactive barrier (PRB) system was installed at Operating Unit (OU) 2 of the Caldwell Trucking Superfund Site in northern New Jersey in 1998. The PRB is located approximately 3,000 ft downgradient of the source area, and immediately upgradient of a “seep” where groundwater discharges to surface water. The barrier is designed to reduce initial trichloroethylene (TCE) concentrations of 6,000-8,000 µg/L in the ground water to below 50 µg/L prior to discharge to surface water.

The Caldwell Trucking site encompasses 11 acres near the Passaic River that were used for disposal of septic wastes in unlined ponds from the 1950s to 1984 and industrial waste containing lead and TCE. The site contains areas of glacial deposition overlying basalt flows with an average conductivity of approximately 0.1 in/sec. Ground-water flow occurs in a 25-ft deep sand and gravel aquifer confined below an impermeable clay layer at an average elevation of 160 ft above mean sea level. The water table is located approximately 5-15 ft below ground surface. A fractured basalt zone is located below the sand/gravel aquifer at 100-125 ft above mean sea level. Approximately half of the groundwater flow is in the sand and gravel unit, and half in the upper fractured basalt. Water from both units discharges through the “seep”. Studies indicated that the rate of natural attenuation occurring at this site is 3,000 kg/yr.

The PRB system was installed in unconsolidated sands and a fractured basalt zone using a combination of hydraulic fracturing (in the unconsolidated formation) and permeation infilling (in the fractured basalt). The barrier system is 50 ft deep, beginning about 15 ft below ground. The system consists of two 3-in walls, 150 ft and 90 ft in length and uses 250 tons of zero-valent iron (Fe⁰) as the reactive material. Construction of the PRB system involved hydraulic fracturing of the upper sand/gravel zone, using 15 hydrofrac/infilling wells at 15-ft intervals, and permeation infilling of the lower sedimentary zone (pumping a gel containing the Fe⁰ down a well into the fractured bedrock through an open borehole). Total installation cost of the PRB system (both walls) at this site is estimated at $1,120,000, $670,000 for the 90-ft (hydrofracing) wall and $450,000 for the 150-ft (permeation infilling) wall. This includes the cost of design, construction, materials, and the reactive material.

Monitoring wells and surface waters have been sampled at least monthly for volatiles and metals, and other parameters have been measured. To date, the barrier has achieved only 50% degradation of TCE in the ground water, from an upgradient concentration of 7,000 µg/L to a downgradient concentration of less than 3,500 µg/L. The PRB is clearly not achieving the performance for which it was designed. Other remedial measures are currently being pursued.

Lessons Learned

The poor performance of this PRB is believed to be due to changes in the groundwater flow regime resulting from the injection of granular iron into the fractured bedrock. Our current theory is that the upper bedrock contains relatively large open fractures that acted as major conduits to groundwater flow. The hydraulic conductivity of these fractures was reduced by the infilling with granular iron. This reduction in hydraulic conductivity, coupled with the shallow gradients in the vicinity of the PRB, resulted in diversion of groundwater flow upward into the sand and gravel unit, as well as sideways around the infilled zones. These effects must be carefully considered in any application where granular iron is introduced into bedrock fractures.

Another unexpected condition was the slow breaking of the guar gum gel used for the iron installation. The low temperature and high pH at which the guar gum gel was formulated slowed its enzymatic degradation after it was in place. As a solution, a pH buffer and additional enzyme were injected. Guar breakdown then occurred and TCE reductions were observed. Otherwise, the gel has not interfered with the barrier's permeability nor impacted the iron's reactivity.