Feasibility of eliminating interferences in graphite furnace atomic absorption spectrometry using analyte transfer to the permanently modified graphite tube surface

A procedure is proposed to avoid spectral and/or non-spectral interferences in graphite furnace atomic absorption spectrometry (GF AAS) by transferring the analyte during the pyrolysis stage from a solid sampling platform to the graphite tube wall that has been coated with a permanent modifier, e.g. by electrodeposition of a platinum-group metal. The direct determination of mercury in solid coal samples was chosen as a model to investigate the feasibility of this idea. The graphite tube surface was coated with palladium and the analyte was transferred from the solid sampling platform to the tube wall at a temperature of 500±50 °C. A characteristic mass of m0=64 pg Hg was obtained for an atomization temperature of 1300 °C, proposing a quantitative transfer of the analyte to the tube wall. Calibration against aqueous mercury standards was not feasible as this element was lost in part already during the drying stage and could not be trapped quantitatively on the modified graphite tube surface. However, the results for all except one of the coal reference materials were within the 95% confidence interval of the certificate when the slope of a correlation curve between the integrated absorbance, normalized for 1 mg of sample, and the certified value for mercury was used for calibration. A detection limit of 0.025–0.05 μg g−1 Hg in coal, calculated from three times the standard deviation of the investigated coal samples, could be obtained with the proposed method. The spectral interference due to excessive background absorption in the direct determination of mercury in coal could be eliminated completely. It is expected that this analyte transfer can be used in a similar way to eliminate other spectral and/or non-spectral interferences in the GF AAS determination of other volatile analytes.