A comparison of sequential extraction techniques for determining arsenic fractionation in synthetic mineral mixtures.

The sequential extraction methods according to Tessier et al., Borovec et al., Zhang and Moore and Hall et al. have been tested for their suitability for arsenic fractionation in samples of artificially prepared mineral mixtures. Mixtures containing different amounts of As-containing phases were prepared so that their compositions corresponded to weathering products on As-bearing ore deposits. A comparison of different procedures on simple mineral mixtures containing calcium arsenate (CaHAsO(4).H(2)O), As-bearing goethite (FeOOH) and arsenopyrite (FeAsS) showed that only the results of the Hall method satisfactorily correspond to the expected arsenic distribution. A detailed verification of the Hall method was subsequently carried out on most complex synthetic mineral mixtures with varying amounts of As-containing kaolinite and carbonate, calcium arsenate, As-bearing goethite and arsenopyrite. The results confirm that the Hall method cannot be fully employed for an accurate As speciation but may be applied for a route identification of As distribution between "labile", "medium-labile" and "residual" forms in heavily polluted soils.