The determination of Au, Pd and Pt in ashed vegetation by ICP-mass spectrometry and graphite furnace atomic absorption spectrometry

Abstract

Current methods to determine Au, Pd and Pt as a “package” in samples of ashed vegetation require a minimum of 10 g to achieve detection limits below 10 ppb. This precludes practical application of biogeochemistry as a prospecting method for these elements. This study is concerned with the development of a method whereby Au, Pd and Pt can be determined in 1–2-g samples of ash to detection levels of 0.5–2 ppb for these elements. The proposed procedure involves ashing at 870°C, decomposition in HF-aqua regia, separation of the analytes by coprecipitation on Te, and analysis by graphite furnace atomic absorption spectrometry (GFAAS) or inductively coupled plasma mass spectrometry (ICP-MS). The high-temperature ashing is employed to obtain full recovery for Pd. Alternatively, additional digestion and evaporation with HClO 4 can be used but with the attendant risk of loss of Au through volatilization of the chloride. Sample introduction in ICP-MS by electrothermal vaporization allows for detailed studies down to 0.2–0.3 ppb for these elements but at the sacrifice of the high productivity enjoyed in analysis by ICP-MS compared to GFAAs. If samples contain high amounts of Cu (e.g., 1000 ppm), an element which is carried through the separation procedure to varying degrees, certain precautions are necessary. Copper causes an isobaric interference (possibly CuAr +) on 105Pd in analysis by ICP-MS and therefore measurement should be made at 106Pd or 108Pd. Copper also suppresses the absorbance for Au in analysis by GFAAS; matrix-matching or the method of standard additions overcomes this interference. Good accuracy is indicated by comparison of results, at higher concentrations of these elements, with independent methods such as NiS fire-assay/neutron activation analysis and Pb fire-assay/ICP-MS. The homogeneity of the sample in Au, Pd and Pt distribution is the dominant factor in the precision obtained. Control samples of ashed spruce twigs and needles averaged relative standard deviations ( n = 7) of 15% for Au (at 40 and 230 ppb), 6% for Pd (at 230 and 1430 ppb) and 18% for Pt (at 90 and 800 ppb).