Assessment of single extractions for the determination of mobile forms of metals in highly polluted soils and sediments--analytical and thermodynamic approaches.

Highly polluted forest/tilled soils and stream sediments from a mining and smelting area were subjected to single-extraction procedures to determine the extractable contents of Cd, Cu, Pb, and Zn. The results obtained from four widely used operationally defined single extraction tests were compared: deionised water, 0.01 mol L(-1) CaCl2, 1 mol L(-1) NH4NO3 and 0.005 mol L(-1) DTPA. The analytical data were coupled with measurement of the pH and Eh in extracts, mineralogical investigations and thermodynamic modelling using the PHREEQC-2 code. The changes in the pH of the equilibrated suspensions significantly influenced the metal extractabilities, with higher values in the lower pH regions. Although the DTPA procedure generally extracted the highest amounts of metals, it was found to be unsuitable for highly organic acidic forest soils, where anionic metal-DTPA complexes are assumed to be re-adsorbed on the positively charged surfaces of soil organic matter and oxides. The NH4NO3 extraction was also unsuitable due to the high ionic strength (1 mol L(-1)), limiting the use of the thermodynamic speciation model and the formation of the Cu(NH3)2+ complex, leading to acidification of the suspension. 0.01 mol L(-1) CaCl2 can be proposed as the most appropriate extraction medium, suitable for speciation modelling and analytical determinations using ICP-techniques and having an ionic strength similar to that of the soil solution. The metals are present in free ionic forms or chlorocomplexes in the CaCl2 extracts, preventing their re-adsorption on the positively charged surfaces of soil solids (organic matter, Fe- and Mn-oxides) in acidic and circum-neutral conditions.