Monitoring metal pollution in soils using portable-XRF and conventional laboratory-based techniques: Evaluation of the performance and limitations according to metal properties and sources.

Large variability in the spatial distribution and content of metals is generally recognised in anthropogenicallypolluted soils, hence, a detailed site investigation implying the collection and analysis of a large number of soil samples is often necessary. To this regard, the selection of a rapid, cost-effective and accurate analytical technique to assess the concentration of metals in soil is of paramount importance. The overall objective of thisworkwas to evaluate the possibility of assessing the aqua regia-extractable (AR) content of metals in soil from the multielement profile of the soil obtained by a portable X-ray fluorescence analyser (pXRF). To this objective, we attempted: (i) to establish, by simple linear regressions, the relations occurring between the metal contentsmeasured by pXRF and AR in laboratory setting on air-dried and 2mm-sieved soil samples fromtwo case studies (Aagricultural and B-industrial sites); (ii) to define metal-based linear models predicting metal AR contents from pXRF measurements; (iii) to assess the influence of metal properties and sources on relations found between the two analytical methods. Very satisfying correlations (R2 N 0.90) were observed between the AR and pXRF contents of Ca, Cu, Cr, Ni, Pb and Zn in the site A, and of Cd, Cu, Pb and Zn in the site B. For themajority ofmetals, lower AR than pXRF contents were measured, as result of the AR incomplete dissolution of metal-bearing silicates. Thiswas not observedwhen metals - of anthropogenic origin - occurred in soil in very high concentrations (i.e., Cr for A and Pb for B). In both sites, the comparison among different regression parameters revealed a strong metal-dependence. Moreover, for most of the metals, the parameters of each metal-regression line significantly differed between the two case studies, indicating site-dependence of regression fits.