A detailed method validation of graphite-furnace atomic absorption spectrometry (GFAAS) with Zeeman background correction was performed. The aim is to perform a detailed investigation of short-term precision as opposed to long-term precision. It was suggested that release of graphite flakes into the light path during measurement significantly influenced the performance of the method. It was found that significant deviations with respect to the certified values were frequent and an estimate of reliable uncertainties was obtained only after a high number of repetitions. Uncertainty of Interlaboratory testing was evaluated as a method to estimate uncertainties that are comparable to uncertainties that were obtained by Interlaboratory testing and to uncertainties predicted by the Horwitz curve. To a large extent, the uncertainty in measurement that was predicted by pooled calibrations corresponded to the uncertainties that were obtained from multiple determinations of unknowns. It was thus proposed that a large proportion of the difference in uncertainty in measurement between laboratories could be explained by properties of the different detectors. In order to support accuracy, it is suggested that a higher level of uncertainty should be accepted in analytical investigations.
Financed by the National Centre for Research and Development under grant No. SP/I/1/77065/10 by the strategic scientific research and experimental development program:
SYNAT - “Interdisciplinary System for Interactive Scientific and Scientific-Technical Information”.