Portal de Periódicos da CAPES

Artigo Produção Nacional Revisado por pares

BIBTEX RIS

Determination of selenium in nutritionally relevant foods by graphite furnace atomic absorption spectrometry using arsenic as internal standard

2005; Elsevier BV; Volume: 93; Issue: 2 Linguagem: Inglês

10.1016/j.foodchem.2004.11.024

1873-7072

Adriana Paiva de Oliveira, José Anchieta Gomes Neto, Joaquim A. Nóbrega, Paulo Rogério Miranda Correia, Pedro V. Oliveira,

Heavy Metal Exposure and Toxicity

A method has been developed for the direct determination of Se in nutritionally relevant foods by graphite furnace atomic absorption spectrometry. Tungsten/rhodium carbide coating on the integrated platform of a transversely heated graphite atomizer or W coating with co-injection of Pd(NO3)2 were used as a permanent modifiers. Samples and reference solutions were spiked with 500 μg L−1 As and absorbance variations due to changes in experimental conditions were minimized. For 20 μL aqueous analytical solutions delivered into the graphite tube, analytical curves in the 5.0–40 μg L−1 with good linear correlation were established. Pyrolysis and atomization temperatures were evaluated using pyrolysis and atomization curves, respectively. The optimized heating program (temperature, ramp time, hold time) of the graphite tube of the Perkin–Elmer SIMAA 6000 atomic absorption spectrometer was: dry steps (110 °C, 5 s, 10 s; 130 °C, 15 s, 15 s); air-assisted pyrolysis step (600 °C, 20 s, 40 s; 20 °C, 1 s, 40 s); pyrolysis step (1300 °C, 10 s, 20 s); atomization step (2100 °C, 0 s, 4 s); clean step (2550 °C, 1 s, 5 s). The method was applied for Se determination in coconut water, coconut milk, soybean milk, cow milk, tomato juice, mango juice, grape juice and drinking water samples and four standard reference materials and results were in agreement at 95% confidence level. The lifetime of the tube was 500 firings and the relative standard deviations of measurements of typical samples containing 25 μgL−1 Se were 3.0% and 6.0% (n = 12) with and without internal standardization, respectively. The limits of detection were in the 0.35 μg L−1–0.7 μg Se L−1 range. The accuracy of the proposed method was evaluated by an addition-recovery experiment and all recovered values were in the 98–109% range.