Measurement of arsenic fractions in freshwater with SCF and DGT technology

The total quantity of a specific trace metal/metalloid in water is distributed into different fractions. Different fractions will dominate depending on witch metal/metalloid under study in addition to the physical and chemical conditions of the specific water system on time of investigation. The Size charge fractionation (SCF) technique is able to separate the filtered fraction from the colloidal and particulate fractions. In addition, the filtered portion is fractionated based on charge so that free ions (labile) are separated from anionic species. The trace metal/metalloid levels are determined by HR-ICP-MS. This presentation will show an example of how SCF together with the passive sampler DGT (Diffusive gradient in thin films) can be used as a technique in a monitoring process and indicate the distribution of arsenic in the different fractions. While DGT estimates a time-integrated concentration of labile species, SCF intercept the distribution between fractions at the sampling moment. In this case, arsenic levels were measured through a whole year in a pond in the vicinity of a phased out wood preservation plant in Norway. The results show a correlation between total arsenic and the anionic fraction (R2 = 0,939). This indicates that the total arsenic is dominated by anionic species. This fractionation technique can be used for different kind of metals/metalloids, and together with DGT be a valuable tool in environmental risk assessment. Not only because it gives an estimation of the amount of labile and bioavailable metals/metalloids, but also their reservoir in the water body. Anionic arsenic content (measured with SCF) ranged from about 0,15 µg/L to 1,4 µg/L inside a relatively small area. Sampling is a major contributor to uncertainties in the final conclusion in an environmental exploration. Selection of sampling points and sampling method in an environmental exploration should therefore have high priority also in use of SCF. Homogeneity and stability can be a problem even in aquatic systems (e.g. in mixing zones). How many different sampling points are necessary to cover the whole area of current interest? How many parallel samples are requisite from each sampling point? Depending on acceptable limits of data quality, decisions on this certain questions determine the sampling procedure.