Abstract

Two sample pretreatment methods based on the use of ultrasound [i.e., automated ultrasonic slurry sampling (USS method) and ultrasound-assisted extraction (UAE method)] were developed for determination of total As in eleven fresh and frozen seafood samples by transverse heated electrothermal atomic absorption spectrometry with longitudinal Zeeman background correction and compared with the conventional microwave-assisted digestion method using the H₂O₂–HNO₃ mixture (MAD method). Non-parametric tests (i.e., Kruskal–Wallis test and Wilcoxon rank sum test) were used in order to compare analytical results obtained, thus avoiding the influence of outliers. In general, no significant differences in accuracy were observed when comparing the UAE and MAD methods. Significantly lower arsenic concentrations were observed for several seafood samples with the USS method even using large amounts of matrix modifier [i.e., 22.5 µg Pd + 13.5 µg Mg(NO₃)₂] which were attributed to non-spectral interferences caused by the sample matrix. When the solid particles were allowed to settle in the USS method, an improved arsenic recovery was obtained; the concentration found being similar to that obtained by the UAE and MAD methods. Plackett–Burman designs were used to assess the robustness of the USS and UAE methods. The slurry concentration was seen to cause a significant negative effect on As recovery in both the slurry and solid–liquid extraction techniques, whereas nitric acid concentration caused a significant positive effect only in the solid–liquid extraction techniques. The LOD of arsenic with the USS and UAE methods was 0.5 µg g⁻¹ and the between-batch precision expressed as relative standard deviation (RSD) was in the range of 1–6%.