Copper fractionation and extractability in two contaminated variable charge soils

The impact of heavy metal contamination on crop growth and water quality has become a public concern in southern China where variable charge soils are widespread. A laboratory experiment was designed with contaminated levels of copper (Cu) to estimate the distribution of anthropogenic Cu in different constituents of two variable charge soils using a sequential extraction procedure (SEP) and to evaluate Cu extractability with a single extraction procedure. Soil Cu was chemically fractionated into water-soluble, exchangeable, weakly specifically adsorbed, Fe/Mn oxide-bound, organically bound, and residual fractions using the SEP. One mole NH4Ac l(-1) (ammonium acetate, pH 5.0), 0.1 mol HCl l(-1), and Mehlich 3 were employed as the single extractant to evaluate the mobility or availability of anthropogenic Cu added in the variable charge soils. The two soils were an inceptisol (clayey mixed siliceous thermic typic Dystrochrept) and an ultisol (clayey kaolinitic thermic Plinthudults), and the Cu levels were treated at 250, 500, 1000 or 1500 mg Cu kg(-1) soil as Cu(NO3)(2). During a 6-week incubation, most of the anthropogenic Cu was associated with the mobile fractions, i.e. water-soluble, exchangeable, and weakly specifically adsorbed fractions in the inceptisol, whereas in the ultisol the mobile fractions became dominant only at the Cu amendments of 1000 and 1500 mg Cu kg(-1) soil. The Fe/Mn oxide-bound fractions that are relatively stable prevailed in both soils except for the high Cu level treatments. Soil organic matter had a minimal effect on the anthropogenic Cu distribution in the soils because of its low concentration range. Transformations between the mobile fractions and Fe/Mn oxide-bound fraction occurred during the 6-week incubation in both soils. The 1 mol NH4Ac l(-1) (pH 5.0) and Mehlich 3 extractable Cu in both Cu-amended soils were statistically associated with the mobile fractions according to the stepwise multiple regressions. However, the 0.1 mol HCl l(-1) extractions resulted in an overestimation of Cu availability in these soils due to the dissolution of Fe/Mn oxides, which released part of the adsorbed Cu that may not be available to plants under normal soil conditions. (C) 2004 Elsevier B.V. All rights reserved.