The sustainable management of dredged waterway sediments requires on-site determination of the main pollutants to facilitate their safe reuse or treatment. Portable X-ray fluorescence (pXRF) is commonly used for similar applications with contaminated soil, but the high water content of dredged sediments precludes any application of standard methods. Measurements for Pb, Zn, Cu and As were performed on-site on raw wet sediments with 50 to 70 % water contents during dredging or mapping operations. These results, although two or three times lower than laboratory analyses on the same samples, were found to be related to absolute concentrations closely enough to rank samples. In order to investigate further the feasibility of field analyses on wet sediments, partial dehydration methods were tested. The most efficient technique is based on a hand press. It is simple and quick enough to be used on dredging boats during operations and produces sample pellets with 30 to 50 % water contents. The relationship between pXRF measurements on these pellets and laboratory analyses was found to be sufficiently linear to calculate estimated concentrations. Potential differences were found to be less than 20 % for Pb and Zn. Higher differences for Cu were due to very low concentrations, within twice the limit of detection (LOD). Some limitations were observed. The water content in pellets is variable depending on the sediment type or matrix. The correction factors vary between the measured elements and they may also vary with matrix chemistry. However, Pb-Zn-Cu-As concentrations were ranked and evaluated accurately and the geochemical signatures of the samples were preserved. We demonstrated that, with a simple partial dehydration procedure, pXRF measurements can be reliably related closely enough to absolute concentrations to make field decisions for sediment management. Since the approximately linear relationships between measurements on semi-wet samples and laboratory analyses are matrix- and site-dependent, they must be recognised before using pXRF on wet samples for decision-making.
Read full abstract