|
1 Department of Earth Sciences & Geography (NERC ICP Facility), Kingston University, Penrhyn Road, Kingston upon Thames KT1 2EE, UK kym.jarvis{at}kingston.ac.uk
2 Department of Geology, Royal Holloway, University of London, Egham Hill, Egham TW20 0EX, UK
Inductively coupled plasma analytical techniques are widely used in forensic geochemistry because they can provide concentration data for a wide range of major and trace elements relatively rapidly and at reasonable cost. A pilot study was undertaken to identify the relative importance of uncertainty resulting from instrumental measurement sources and that due to the procedures used to prepare the samples initially. Three soils with a range of major and trace element concentrations were collected to permit an evaluation of uncertainty. A reference sample of demonstrated homogeneity was also prepared and analysed. Samples were prepared in replicate (five preparations) of each, and assessment made of uncertainty in the instrumental measurement alone and for replicated preparations of the same material. Small sample sizes (0.05g) were used to mimic the situation common in forensic investigation. Results show that, while instrumental variability may be an important factor during measurement, between-sample variation has a dominant effect on uncertainty in the final result. It is clear that, without replicated measurement and preparation, the uncertainty of the measured data is unknown. Thus, critical samples, on which a case might depend, must be analysed in a way that defines clearly that uncertainty.
