Dissolution and transport of 2,4-DNT and 2,6-DNT from M1 propellant in soil.

Live-fire training exercises can result in particulate propellant contamination on military training ranges and can potentially contaminate ground water. This study was conducted to evaluate dissolution of the 2,4-dinitrotoluene (2,4-DNT) and 2,6-dinitrotoluene (2,6-DNT) from the propellant formulation, M1 (87.6% nitrocellulose, 7.3% 2,4-DNT, 0.57% 2,6-DNT, 1.06% diphenylamine, 3.48% dibutyl phthalate) and their subsequent transport in soil. Batch dissolution studies were followed by saturated column transport experiments. Neat, dissolved 2,4-DNT, and M1 in solid and dissolved forms were used as influent to columns filled with Plymouth loamy sand (mesic, coated Typic Quartzipsamments) from Camp Edwards, MA. Dissolution rates and other fate and transport parameters were determined using the HYDRUS-1D code. M1 dissolution was limited by DNT diffusion from the interior of the pellet, resulting in an exponential decrease in dissolution rate with time. The HYDRUS-1D model accurately described release and transport of 2,4- and 2,6-DNT from M1 propellant. Dissolution rates for M1 in the stirred reactor and column studies were similar, indicating that batch dissolution rates are potentially useful to represent field conditions.

Dissolution and transport of TNT, RDX, and composition B in saturated soil columns.

Low-order detonations and blow-in-place procedures on military training ranges can result in residual solid explosive formulations to serve as distributed point sources for ground water contamination. This study was conducted to determine if distribution coefficients from batch studies and transport parameters of pure compounds in solution adequately describe explosive transport where compounds are present as solid particles in formulations. Saturated column transport experiments were conducted with 2,4,6-trinitrotoluene (TNT), hexahydro-1,3,5-trinitro-1,3,5-triazine (RDX), and the explosive formulation, Composition B (Comp B) (59.5 +/- 2.0% RDX, 39.5 +/- 2.3% TNT, and 1% wax) in solid and dissolved forms. The two soils used were Plymouth loamy sand (mesic, coated Typic Quartzipsamments) from Camp Edwards, MA and Adler silt loam (coarse-silty, mixed, superactive, thermic Fluvaquentic Eutrudepts) from Vicksburg, MS. Interrupted flow experiments were used to determine if explosives were at equilibrium distribution between soil and solution phases. The HYDRUS-1D code was used to determine fate and transport parameters. Results indicated that sorption of high explosives was rate limited. The behavior of dissolved Comp B was similar to the behavior of pure TNT and RDX. Behavior of solid Comp B was controlled by dissolution that depended on physical properties of the Comp B sample. Adsorption coefficients determined by HYDRUS-1D were different from those determined in batch tests for the same soils. Use of parameters specific to formulations will improve fate and transport predictions.

Reduction of suspended solids following hydroclassification of metal-contaminated soils.

Remediation of metals-contaminated soil typically uses solidification/stabilization and "dig and haul". Soil washing and physical separation have been applied to a much lesser extent to reduce soil volumes requiring aggressive treatment and to improve performance of follow-up treatments. In earlier work [J. Hazard. Mater. 66 (1999) 15], we used a simple, vertical-column hydroclassifier, to separate four soils contaminated with heavy metals, defining a "best case" performance for larger-scale (minerals processing) equipment. Such processes, using water-based slurries, generate substantial volumes of water with suspended solids. These typically contain disproportionately high concentrations of heavy metals. Here, we performed an initial screening of settling, coagulation, and centrifugation for reducing suspended solids, and thus suspended metals from soil slurries following processing. The four soils, previously hydroclassified, were sieved to <600 microm, slurried with a 4:1 weight ratio of water, and allowed to settle. Slurry samples were collected at settling times of 0, 0.0833, 1, 5, and 22-24h. Coagulant (alum) addition and centrifugation were investigated. The slurries were filtered, digested, and analyzed by atomic absorption for lead and chromium content. Two soil slurries clarified in <5 min. In all four cases, 90% of solids and metals settled within 5h. However, completion may require up to 24h, or other intervention, i.e. coagulants. The metal concentration in the residual suspended solids increased with settling time, implying an enrichment of metals in finer, suspended particles. Metals dissolved in the slurry water ranged from 3 to 5mg/l for chromium and lead. This screening study provides guidance for water treatment requirements and treatability studies for the integration of hydroclassification and solids removal.