Sequential hydride generation/pneumatic nebulisation inductively coupled plasma mass spectrometry for the fractionation of arsenic and selenium species

Abstract

The toxicity of certain elements is known to be related to their organic substituents and/or oxidation states. As such, total elemental determinations do not always yield sufficient information for accurate risk assessments and therefore speciation or fractionation data are required. In order to obtain fractionation data for trace levels of arsenic and selenium, a novel sequential pneumatic nebulisation (PN)/hydride generation (HG) inductively coupled plasma mass spectrometry (ICP-MS) method was developed. The method offers the advantage of sample introduction via either PN or HG by simply rotating a 4-way switching valve while the system is in operation. In PN mode, the liquid sample is aspirated into ICP, allowing for the determination of the total amount of each element, whilst in HG mode only the arsenic and selenium species that form volatile hydrides are determined. Conveniently, in the case of arsenic, this allows for differentiation between the four most toxic arsenic species (arsenate, arsenite, monomethylarsonic acid and dimethylarsinic acid), which form volatile hydrides, and the virtually non-toxic forms (arsenobetaine, arsenosugars, etc.), which do not. This allows for the rapid estimation of the amounts of toxic and non-toxic arsenic species present in a sample. For arsenic, the technique gave detection limits of 36 ng l −1 in PN mode and 1 ng l −1 in HG mode. For selenium, detection limits of 150 ng l −1 were achieved in PN mode and 220 ng l −1 in HG mode. The technique also gave good long- and short-term stabilities of under 6% RSD for both elements. A variety of samples, including water and urine standard reference materials, were analysed in both modes, and the precision and accuracy of the results for total arsenic and selenium levels were assessed. Using the technique in both modes also allowed for the fractionation of As and Se species into their volatile hydride-forming and non-hydride-forming species. This was particularly informative, with respect to As species present, in the case of a kelp powder extract. Digested tobacco samples were only analysed for their total As levels, in which case results obtained via both sample introduction modes showed good agreement.