Labile rhizosphere soil solution fraction for prediction of bioavailability of heavy metals and rare earth elements to plants.

A labile rhizosphere soil solution fraction has been recommended to predict the bioavailability of heavy metals and rare earth elements to plants. This method used moist rhizosphere soil in combination with a mixture of 0.01 mol L(-1) of low-molecular-weight organic acids (LMWOAs) as extractant. The extracted soil solutions were fractionated into two colloidal fractions of <0.45 microm (F(3)) and <0.2 microm (F(2)), and one truly dissolved fraction including free metal ions and inorganic and organic complexes (fraction<molecular weight cutoff of 1 kD). Hereafter, the fraction of the truly dissolved fraction refers to labile rhizosphere soil solution fraction, F(lrss). For the soil solutions extracted with a mixture of LMWOAs the concentrations of heavy metals and rare earth elements in F(2) and F(3) were quite similar. However, the mean concentrations of Cr, Ni, Zn, Cu, Pb, Cd, La, Ce, Pr, and Nd in F(lrss) accounted for 79.9%, 91.3%, 90.8%, 60.1%, 77.5%, 75.3%, 81.2%, 77.2%, 80.3%, and 79.5%, respectively, of their concentrations in F(2). In contrast, there were no differences in the extractable metal concentrations between the three fractions while the first step of the method recommended by the European Community of Reference (BCR), where 0.1 mol L(-1) acetic acid was used as an extractant. The single correlation analysis was made between metal concentrations in the different fractions of soil solutions and their concentrations in wheat. If the first step of BCR method was used there was no good correlation between heavy metals in soil pools and that in wheat shoots and roots. When LMWAOs were used a good correlation was obtained between the concentrations of heavy metals in soil pools and that in wheat roots, which followed a general order of r(1 kD, LMWOAs) >r(0.2 microm, LMWOAs) approximately r(0.45 microm, LMWOAs). In the case of rare earth elements the good correlation was obtained for both the wheat roots and shoots. Generally, the correlation coefficients obtained by LMWAOs were better than that obtained by the first step of BCR method. Therefore, LMWAOs and F(lrss) were strongly recommended to predict the bioavailability of metals in soil pools to plants.