Estimation of uncertainties in ICP-MS analysis: a practical methodology

The measurement uncertainty associated with the determination of Ni-60 in a queous samples by ICP-MS has been calculated using a cause-and-effect appro ach. A cause-and-effect diagram was constructed to aid in the identificatio n of the sources of uncertainty associated with the method. The uncertainty estimate was calculated from a combination of existing quality control dat a and specially planned experiments. The uncertainty budget was based initi ally on precision data, followed by separate evaluation of the method bias and the effect of parameters not sufficiently covered by these estimates. T he construction of the cause-and-effect diagram, its reconciliation with ex isting data and the estimation of the individual components of the uncertai nty budget are described in detail. The expanded uncertainties for three di fferent nickel concentrations (3,10,35 ng g(-1)) were calculated as 1.1, 1. 5 and 5.3 ng g(-1), respectively. These were calculated using a coverage fa ctor of two approximating to a 95% level of confidence. The dominant contri butions to the uncertainty budget were method precision, instrument drift a nd bias measured as method recovery. The uncertainties associated with the concentration of the working standard and sample dilution were found to be insignificant. (C) 1999 LGC (Teddington Ltd). Published by Elsevier Science B.V. All rights reserved.