Chemical fixation of arsenic in contaminated soils

Arsenic-contaminated soils have been successfully treated using fixation methods whereby chemicals are added to prevent As mobilization. However, the chemistry of the fixation process used in the field is poorly understood. We have examined one process which succeeded in immobilizing 0.1 to 0.2 weight % As in soil at a 69 a old dump site through the addition of ferrous sulfate and water followed by Ca(OH)(2), Portland cement and water. The intent of the first step is to form insoluble Fe arsenate, the second to bind the soil. The particle-size distribution of the untreated soil grains containing As is bimodal, with 70% of the As occurring in 2 size fractions: 38% in the 0.25-0.5 mm fraction mostly as the mineral hoernesite (Mg-3(AsO4)(2) . 8H(2)O), and 32% in the size fraction less than 5 mu m. From sequential extractions, As in the finest fraction is inferred to be present in an oxyanionic form adsorbed on mineral surfaces. Arsenic may also be associated with Fe or Mn oxide coatings, and may be present as fine grains of hoernesite. Hoernesite, which we infer formed because of the brackish nature of the ground water at this site, represents an example of natural fixation of As. Arsenic is more evenly distributed among grain size fractions in the field-fixed soils. Using X-ray diffraction, electron microscopy. and chemical analysis, we find no direct evidence for the formation of Fe arsenate phases in the fixed soils. Sequential extractions demonstrate that the fixation method reduces exchangeable As in the < 5 mu m fraction even when only Fe sulfate is added. While fixation occurs by reaction with FeSO4, it is unlikely that ferric arsenate forms as an insoluble phase under these conditions, as is generally assumed. Arsenic fixation occurs through precipitation of an unknown As-Fe phase or by incorporation during Fe oxide precipitation aided by immobilization by a cement coating. Copyright (C) 1996 Elsevier Science Ltd