Treatment of arsenic-contaminated soils. II: Treatability study and remediation

Treatment of sandy soils contaminated with arsenic was investigated at a bench scale and carried through to remediation in the field. The initial treatability study looked at many combinations of cement binders and reagents. Salts of iron, barium, manganese, and magnesium were generally effective at reducing arsenic leachability. The most consistently low potential for leaching [toxicity characteristic leaching procedure (TCLP) and a modified version of the American Nuclear Society's ANS16.1] was observed when the soils were treated with a mixture of Type I portland cement and ferrous sulfate. For instance, the average of arsenic concentrations in TCLP leachates in many treated soil samples from four sites was 0.26 mg/L. Better protection against leaching was observed when the soil was pretreated with FeSO4.7H(2)O, then with portland cement. In addition to chemical containment, the mixture should prevent ground-water leaching by physical entrapment, because the permeabilities of the treated soils were in the range of 10(-9)-10(-10) m/s. Scanning electron microscope micrographs showed a dense mass with minimal void space, and a combination of X-ray diffraction, thermal analysis, and solid-state nuclear magnetic resonance (NMR) spectroscopy indicated the formation of a normal hydrated cement matrix, with some excess ettringite being present due to the extra sulfate being added to the formulation. Results from the bench-scale treatability study were reproduced very faithfully in the field, with permeabilities and compressive strengths being similar to those observed in the laboratory and TCLP leachability being even lower than predicted by the laboratory study.