EPA 1315 One-Dimensional Leach Testing of Silts Impacted by Chlorobenzene-Rich NAPL

Abstract

An extensive treatability study was completed on a chlorobenzene-rich nonaqueous phase liquid (CB-NAPL) impacted source zone at an active industrial site. A baseline leaching condition for the low permeability silty site soils was established via one-dimensional (1D) semidynamic leach testing on undisturbed and remolded core samples using EPA 1315 modified (M) [Method 1315 – mass transfer rates of constituents in monolithic and compacted granular material using a semidynamic tank leaching procedure, revision 1. Washington, DC: EPA. Office of Resource Conservation and Recovery] for use with hydrocarbons under zero headspace conditions in a new, specially designed apparatus. Comparison of the most heavily impacted samples (average of 2,932 to 9,422 mg/kg CB) indicated that the EPA 1315M 1D leaching response was typically within a factor of two when comparing undisturbed versus remolded (homogenized) samples. Even though soil homogenization may have changed the initial soil fabric and contaminant distribution, the comparable outcomes suggested that it may be sufficiently accurate to work with remolded samples given the many challenges and uncertainties associated with undisturbed cores. In addition, CB leaching rates were observed to be more closely associated with total CB content than hydraulic conductivity (K) for samples characterized by K < 1 × 10−6 cm/s. This is perhaps not surprising since the concentration gradient is the key driver for mass transfer in diffusion-controlled media. Comparison of the EPA 1315M leaching results from an untreated, undisturbed silty soil layer from the hotspot location to the leaching results of a stabilization/solidification (S/S)-treated soil homogenized over the full treatment depth showed CB percent leaching reductions (%LRs) of approximately 85.5% for the contract-specified laboratory-based mix design and 74.5% for a field quality assurance (QA) sample taken during remedial construction. These %LR values for the S/S treatment were considerably lower than the calculated %LR of >99% using EPA 1316M on the most heavily impacted sand sample from the hotspot location. This discrepancy in perceived performance prompted a discussion on the appropriate use and limitations of the %LR framework for assessing S/S performance and stressed the importance of properly designating the principal threat waste condition as that hotspot location/interval characterized by the greatest contaminant mobility.