Bioaccessibility of lead in phosphorus amended urban soils: impact of particle size and extraction method.

Phosphate amendments are one of the most studied management strategies to immobilize lead (thus reducing his bioaccessibility) in heavily contaminated soils (brownfields, mining areas), but their applicability on urban soils is poorly known. In these areas the most likely pathways of exposure are ingestion and inhalation; consequently, Pb-bound to particles <200 μm, than can be easily retained by hands, and <10 μm, removable from soil through wind or water erosion, are the most important fractions in relation to bioaccessibility. The objective of this study was to evaluate the effect of phosphate amendments on a historically roadside contaminated soil (RO) and an agricultural soil (SP) spiked with 500 mg kg-1 of Pb. The effectiveness of the treatments was evaluated through acetic acid (AA) and EDTA extractions, while bioaccessibility with four different methods (USEPA 1340 and modifications). Specifically, we assessed: i) the effect of the aging of the contamination on the efficacy of the amendment; ii) the effectiveness at lower-than-in-literature P:Pb rates (3:5, 6:5 and 12:5), applicable urban soils; iii) the effectiveness on different size fractions. The aging study lasted two years. At the end, the weakly bound Pb tended to decrease, while EDTA-extractable remained stable. For both extractions, contamination aging resulted in a lower efficacy of the amendment, as P additions led to lower percentages of immobilized (not extractable) metal. On SP soil, P additions managed to decrease AA-extractable Pb linearly with the added quantity, while EDTA extraction evidenced a high effect (75% reduction) only for the highest addition. Bioaccessibility results reported large differences between extractions; with the official method (at pH 1.5) no immobilization could be seen, while using modified extractions at pH 2.5 an effect on SP soil, although not significant, is visible. On RO soil, no effect was detectable with all extractions. Lead is enriched in the <10 μm fraction of both soils and, in general, the two lowest additions did not display an immobilization trend, while the highest has an effect only on coarsest particles of both soils. The finest particles displayed a significant immobilization only with the milder extraction (at pH 2.5). In conclusion, amendments were effective only at the highest application rate and on SP soil, while not on RO soil and they did not had any effect on fine particles. As the evaluation of the efficacy depends on the size fraction and the extraction method, without an accepted method for screening P-treated soils rigorous conclusions cannot be drawn, but our results indicate a limitation of the method for historically contaminated urban soils.