Metal extraction by Arabidopsis halleri grown on an unpolluted soil amended with various metal-bearing solids: a pot experiment

Most studies dealing with phytoremediation have considered metal extraction efficiency in relation to metal concentration of bulk soil samples or metal concentration of the soil solution. However, little is known about the effect of various metal-bearing solids on plant growth and metal extraction of hyperaccumulators. In this study, we investigated the ability of Arabidopsis halleri to grow and extract metals from different substrates consisting in an unpolluted soil amended with various metal-bearing solids collected in soils around a Zn smelter complex. The metal-bearing solids used as amendments were: fresh and decomposing organic residues in the soil, a soil clay fraction and two waste slags. Pure mono-metallic salt (ZnSO4) was also used. Two series of substrates were produced, one moderately polluted, and the other highly polluted. An additional substrate was formed by the unamended soil, and used as an unpolluted control. Zn, Cd, Cu, and Pb were measured in the substrates, and in the roots and shoots of A. halleri. The dry matter yield of A. halleri was shown not to depend on the nature of the metal-bearing solid used, except when Cu-toxicity was suspected. On highly-polluted substrates, Zn extraction by A. halleri depended on the nature of metal-bearing solids used, showing the following trend: pure mono-metallic salt>waste slags and soil clay fraction>fresh and decomposing organic matter. We explained these differences by the high solubility of Zn in the mono-metallic salt, whereas in the mineral metal-bearing solids and in both fresh and decomposing organic matter, Zn release required mineral weathering or organic matter mineralization, respectively. This work clearly showed that phytoremediation studies have to consider the nature of metal-bearing solids in contaminated soils to better predict the efficiency of plant extraction. (C) 2001 Elsevier Science Ltd. All rights reserved.