Detection of bias errors in ETAAS determination of copper in beer and wine samples

A method that evidences changes in the shape of the absorbance profiles obtained by graphite furnace atomic absorption spectrometry (ETAAS) is proposed. The method is based upon the apparent content curves model previously described for molecular spectroscopy and it permits the detection of possible sources of bias errors. Moreover, a procedure that allows to detect the existence of constant and/or proportional errors is also described. Both models has been applied to the determination of copper in wine and beer samples with and without pre-treatment of the samples. Results obtained evidence the usefulness of the proposed models. (c) 2005 Elsevier B.V. All rights reserved.