Direct determination of cobalt and zinc in samples of different volatility by means of solid sampling-graphite furnace atomic absorption spectrometry

In this work, a method has been developed for the direct determination of cobalt and zinc in two different real-life samples (a vitamin complex and an agricultural soil) by means of solid sampling-graphite furnace atomic absorption spectrometry. The selection of the elements and of the samples was based on their different volatility and complexity, the final aim being to check the maturity of solid sampling-graphite furnace atomic absorption spectrometry to deal with different situations, and to further contribute to the development of a general methodology that could be used for a large variety of cases. In order to achieve these goals a systematic study was carried out by selection of the most appropriate working conditions and optimization of the sample mass. It can be stressed that the maintenance of the Ar flow (250 ml min(-1)) during the atomization step was found to be especially useful to improve the signal to background ratio for those situations in which simultaneous release of the matrix and atomization of the analyte was unavoidable (Co determination in soil). Comparison of the results for both samples showed no significant differences (other than some disparity regarding the optimum mass range), proving that solid sampling-graphite furnace atomic absorption spectrometry can be as appropriate for soil analysis as it is for the analysis of organic samples, if appropriate working conditions are chosen. The method finally proposed shows interesting features for the determination of the analytes in the samples such as: the ability to use aqueous standard solutions for calibration, a high sample throughput (15-20 min analysis time per analyte in a sample), the fact that it is unnecessary to use hazardous reagents, a low sample consumption, adequate limits of detection (0.5 mg kg(-1) for Co and 4.1 mg kg(-1) for Zn in the soil sample), suitable precision values (R.S.D. similar to8%) and a reduced risk of analyte losses and contamination. (C) 2003 Elsevier Science B.V. All rights reserved.