The form, distribution and mobility of arsenic in soils contaminated by arsenic trioxide, at sites in southeast USA

Soils from many industrial sites in southeastern USA are contaminated with As because of the application of herbicide containing As2O3. Among those contaminated sites, two industrial sites, FW and BH, which are currently active and of most serious environmental concerns, were selected to characterize the occurrence of As in the contaminated soils and to evaluate its environmental leachability. The soils are both sandy loams with varying mineralogical and organic matter contents. Microwave-assisted acid digestion (EPA method 3051) of the contaminated soils indicated As levels of up to 325 mg/kg and 900 mg/kg (dry weight basis) for FW and BH soils, respectively. However, bulk X-ray powder diffraction (XRD) analysis failed to find any detectable As-bearing phases in either of the studied soil samples. Most of the soil As was observed by scanning electron microscopy, coupled with energy dispersive X-ray spectroscopy (SEM/EDX), to be disseminated on the surfaces of fine-grained soil particles in close association with Al and Fe. A few As-bearing particles were detected in BH soil using electron microprobe analysis (EMPA). Synchrotron micro-XRD and X-ray absorption near-edge structure (XANES) analyses indicated that these As-rich particles were possibly phaunouxite, a mineral similar to calcium arsenate, which could have been formed by natural weathering after the application of As2O3. However, the scarcity of those particles eliminated them from playing any important role in As sequestration. Synthetic acid rain sequential batch leaching experiments showed distinct As leaching behaviors of the two studied soil samples: BH soil, which has the higher As content, showed a slow, steady release of As, while FW soil, with a lower As content, showed a much quicker release and lower overall retention of As upon leaching. Sequential chemical extraction experiments were carried out using a simplified 4-step sequential chemical extraction procedure (SCEP) previously developed to characterize the fractionation of As and better understand the different leaching behaviors of the two studied soils. It was shown that only about 50% of the total extractable As was removed by the first two extraction steps, which represented the most weakly bonded and readily available As for environmental leaching. Compared with the sequential leaching experiments, it was further indicated that only half of the As associated with phases extracted by the second SCEP step was mobilized by SPLP leaching. Although microwave-assisted acid digestion results showed similar Al and Fe contents in both soils, the sequential chemical extraction experiments indicated that BH soil has a much higher content of amorphous Al and Fe phases and that a comparably higher portion of soil As was associated with those materials. The experimental results suggest that remediation efforts for the contaminated sites can be directed towards enhancing the formation of more stable As-bearing compounds in the soils to reduce the environmental leachability of As.