A number of sites around the United States have used activated carbon (AC) amendments to remedy contaminated sediments. Variation in site-specific characteristics likely influences the long-term fate and efficacy of AC treatment. The long-term effectiveness of an AC amendment to sediment is largely unknown, as the field performance has not been monitored for more than three years. As a consequence, the focus of this research effort was to evaluate AC’s long-term (6–10 yr) performance and efficacy at two sites where sediments have been treated with AC, representing a range of physical environments. Specifically, the individual objectives comprised the following:

• Determine the mass of AC present and its vertical distribution within the sediment following the initial introduction.
• Determine if the AC still retained its functional efficacy (PCB sorption, reduced bioavailability).
• Apply this information to develop lines of evidence for making conclusions on the long-term efficacy of in situ remediation using AC.
• Disseminate the study results to inform sediment remediation guidance.

The data collected enabled comparison of AC distribution, polychlorinated biphenyl (PCB) concentrations, and bioaccumulation measured over the short and long term (months to years).

Assessments were performed at two pilot-scale demonstration sites, Grasse River, Massena, New York and Canal Creek, Aberdeen Proving Ground (APG), Aberdeen, Maryland, representing two distinct physical environments. Sediment core samples were collected after 6 and 10 years of remedy implementation at APG and Grasse River, respectively.

At the Canal Creek pilot study site, two commercially available technologies designed to deliver AC in a pelletized form into sediments were evaluated: SediMite, produced and marketed by Sediment Solutions, and AquaGate, manufactured and marketed by AquaBlok.

In the Grasse River pilot study, mixed and layered carbon treatments were evaluated. In the mixed treatment, a tiller-like device mechanically mixed bituminous coal–based AC (Carbsorb, Calgon Carbon) into surface sediments, while in the layered treatment a coconut shell–based AC (055C-CNS-V000, Calgon Carbon) was layered by broadcasting of the material over the water surface and allowing it to settle onto the sediment surface and mix into the sediments via natural processes (for example, bioturbation).

Data collected at each of the demonstration sites enabled a comparison of AC distribution, PCB concentrations, and bioaccumulation measured over the short and long term under differing hydrologic regimes, AC amendment types, and application techniques.

Overall, the long-term monitoring at the two pilot-scale studies revealed that AC is stable in the sediment environment at two very distinct sites (that is, vegetated estuarine wetlands [Canal Creek, Maryland] and a riverine system [Grasse River, New York]) with different AC amendments deployed using different application technologies. Results from the present study provide the following key insights into the long-term performance of AC amendment to river sediments:

• AC applied directly into sediments is stable in a slow-flowing river system and could be found at or above target levels after 10 years in the field.
• AC applied on surface sediments using three different approaches was found to be stable in a tidal wetland environment after 6 years in the field.
• Deposition of new, cleaner sediments over time buried the AC-treated zone of sediments, which now functions as a barrier layer between the shallower, cleaner sediments and deeper, more contaminated sediments.
• The depth of penetration of the AC layer is a function of site characteristics. A higher rate of burial and mixing is observed in the river site compare to the tidal wetland.
• AC amended to sediments continues to reduce pore water PCB concentrations within the zone of application.
• Natural attenuation due to clean sediment deposition is high at Grasse River, which has resulted in a nearly one order of magnitude reduction in surface sediment PCB concentration over 10 years.
• Pore water PCB concentrations continue to be reduced in the zone where AC was found in each of the treatments.
• The presence of AC in sediments continues to reduce PCB bioavailability to benthic organisms.
• Initial mechanical mixing in the mixed treatment plot resulted in a more diffuse layer of AC over time, as additional natural mixing spread out the AC through the depth. This finding suggests that the initial mechanical mixing may not be necessary for sites where natural mixing is anticipated.

The data collected enabled comparison of AC distribution, PCB concentrations, and bioaccumulation measured over the short- and long-term (months to years). The study found an important issue with performing ex situ measurements for sediments from a tidally influenced marsh site. Intact cores in the laboratory do not appear to accurately reflect the field exposure conditions after treatment for passive sampling or organism exposure when there is very active groundwater movement due to diurnal tidal pumping. This can be especially true where a major source of the pollutants lies deeper in the sediments and there is an active tidal pumping process as seen at the Canal Creek site.