Uncertainty associated with the leaching of aerosol filters for the determination of metals in aerosol particulate matter using collision/reaction cell ICP-MS detection

High quality observational data with a firm uncertainty assessment are essential for the proper validation of biogeochemical models for trace metals such as iron. Typically, concentrations of these metals are very low in oceanic waters (nM and sub nM) and ICP-MS is therefore a favoured technique for quantitative analysis. Uncertainties in the measurement step are generally well constrained, even at sub-nM concentrations. However, the measurement step is only part of the overall procedure. For the determination of trace metal solubilities from aerosols in the surface ocean, aerosol collection on a filter paper followed by a leaching procedure is likely to make a significant contribution to the overall uncertainty. This paper quantifies the uncertainties for key trace metals (cobalt, iron, lead and vanadium), together with aluminium as a reference element, for a controlled, flow through laboratory leaching procedure using filters collected from three different sampling sites (Tudor Hill (Bermuda), Heraklion (Crete) and Tel-Shikmona (Israel)) and water, glucuronic acid and desferrioxamine B as leachants. Relative expanded uncertainties were in the range of 12-29% for cobalt, 12-62% for iron, 13-45% for lead and 5-11% for vanadium. Fractional solubilities for iron ranged from 0.2 +/- 0.1% to 16.9 +/- 3.5%.