Determination of analytical limits in solid sampling ETAAS: a new approach towards the characterization of analytical...

In the present study the lower analytical limits of solid sampling electrothermal atomization atomic absorption spectrometry (SS-ETAAS) were characterized by means of blank measurements and--for the first time--by means of the calibration curve method, where a calibration near the range of these limits (limit of decision, detection and quantification) was performed. The limit of decision as derived from blank measurements was calculated according to the 3sigma-criterion to be 0.003 and 0.019 ng for Cd and Pb, respectively. For Pb and Cd a roughly three-fold increase of these limits was observed when the calibration method according to DIN 32 645 was applied. When solid reference material was used, only a slight increase could be observed. The analytical limits were 2 to 20 times lower than reported for sample decomposition methods. The blank measurement and conventional calibration curve method, however, do not account for factors relating to solid sampling such as sample mass and matrix. Therefore, the calibration curve model was applied to data derived from comparisons between direct solid sampling ETAAS and a compound reference method (ETAAS following sample homogenization and digestion). The observed analytical limits were not found to be substantially increased if enough samples with low element contents were available for calibration. Coupling of the calibration curve model with the comparison of methods included real test samples and thus the relevant maximum sample mass and analyte content in the range of the lower analytical limits. As validation procedures frequently include comparisons of methods, the present approach might prove to be of some general interest for the characterization of analytical quality in rapid methods.