EPA LEAF Testing of Chlorobenzene-Impacted Sands and Soil–Cement Mix Designs

Abstract

This paper reports on laboratory-based treatability and field quality assurance (QA) testing completed on interbedded sands and silts at an industrial site heavily impacted by a chlorobenzene (CB)-rich nonaqueous phase liquid (NAPL) to depths of approximately 30 ft (9.1 m) below the ground surface (bgs). Total CB concentrations were measured up to 4,200 and 11,000 mg/kg in untreated sandy and silty soil samples, respectively. EPA 1316M testing indicated that the effective solubility of CB was on the order of 90–210 mg/L in two of the most impacted sand samples, the highest of which (from hotspot location 4370) was selected as the principal threat waste condition. A full depth [25 ft (7.6 m)] soil core from this location was homogenized to form a soil surrogate containing approximately 1,500 mg/kg CB. Three stabilization/solidification (S/S) mix designs using Type I/II portland cement (PC) doses ranging from 5 to 10 wt% were then prepared using the soil surrogate. The unconfined compressive strength (UCS) and hydraulic conductivity (K) values after 28 days of curing were, respectively, on the order of 190–265 lb/in.2 (1.31–1.83 MPa) and 1.0 × 10−9 cm/s, easily meeting the performance criteria. All mix designs showed essentially the same EPA 1315M leaching response, with percent leaching reductions (%LRs) greater than 99% for CB and benzene (B) compared with the untreated sand. This was well above the minimum of 90% LR established for successful mix designs. During remedial construction, bucket and auger S/S-mixing techniques using 7.5 wt% PC were used to treat approximately 9,335 CY of impacted soils to depths up to 30 ft (9.1 m) bgs (12 bucket-mixed cells and 527 auger columns) in approximately 6 months. The UCS and K criteria were both satisfied. The relative vertical positioning of the EPA 1315M cumulative mass release curves for CB from the field QA samples appeared to be more strongly influenced by the total CB content than by K, because all S/S-treated samples were characterized by K < 1 × 10−6 cm/s (i.e., were diffusion-controlled). EPA 1315M testing of the field-mixed sample from hotspot location 4370 again indicated %LRs greater than 99% for both CB and B. Samples from a second location in the western portion of the ISS treatment area demonstrated similar results.