Inter-method comparison for the determination of antimony and arsenic in peat samples

Four analytical approaches, based on different physical principles, for the determination of antimony (Sb) and arsenic (As) in ancient peat samples were critically evaluated: (a) open vessel digestion/hydride generation-atomic absorption spectrometry (HG-AAS), (b) closed-pressurized digestion in a microwave oven followed by sector field-inductively coupled plasma-mass spectrometry (SF-ICP-MS), (c) digestion in a microwave autoclave and subsequent quadrupole-inductively coupled plasma-mass spectrometry (Q-ICP-MS) measurements and (d) instrumental neutron activation analysis (E, AA). The quality control scheme applied, always included the use of adequate plant reference materials to ensure the accuracy and precision of the analytical procedures. Additionally, two internal peat reference materials were analyzed using all four analytical approaches, generally showing good agreement for both elements. Method detection limits for As and Sb provided by all procedures were approximately 5 and 2 ng g(-1) which is sufficiently low for the reliable quantification of both elements in ancient, pre-anthropoaenic peat samples. A comparison of As and Sb concentrations in a set of peat samples determined by INAA, HG-AAS and SF-ICP-MS revealed that INAA underestimated the values in a systematic manner, whereas HG-AAS and SF-ICP-MS data agreed very well. Best precision of the results was obtained by analytical procedures involving HG-AAS or Q-ICP-MS and varied from 3.6 to 4.3% and 7.1 to 7.5% for As (at about 0.5 mug g(-1)) and Sb (at about 0.1 mug g(-1)), respectively. The highest sample throughput (40 samples per run accomplished in 2h) combined with low risk of sample contamination could be realized in the high-pressure microwave autoclave. The amount of sample required by all approaches was 200 mg, except for INAA which needed at least 25 times more sample mass to achieve comparable detection limits. For the quantification of As and Sb, inductively coupled plasma-mass spectrometry (ICP-MS) was preferred over LNAA and HG-AAS, mainly because (a) less sample is needed and (b) As and Sb can be determined simultaneously. In addition, ICP-MS offers the possibility to measure concurrently a wide range of other elements which also are of environmental interest. (C) 2002 Elsevier Science B.V. All rights reserved.