Determinação de cádmio, chumbo, tálio e zinco em águas e alimentos por espectrometria de absorção atômica com injeção de amostra em forno aquecido por chama (BIFF-AAS).

Abstract

A flame atomic absorption method based on the introduction of a liquid sample jet into a heated tube placed in the burner head of the spectrometer was evaluated for the determination of Cd, Pb, Tl and Zn in foods and waters. The beam injection flame furnace system consisted of a piston pump, a Rheodyne valve for sample injection, PEEK tube transmission lines (0.7 mm inner diameter), a nozzle made from PEEK with a channel (50 µm inner diameter; 200 µm long) and an integrated filter, for liquid jet generation. The flame furnace consisted of a Ni-base superalloy tube (1 cm inner diameter; 10 cm long) positioned in the flame, held by ceramic pins fixed at both ends of the burner head. The effects of sample carrier flow rate (1 to 2 ml min-1), sample injection volume, composition of fuel/oxidant mixture (C2H2 / air), total flow rate of gaseous mixture, diameter and number of holes in the tube atomizer on analytes atomization are presented. The importance of flame gases introduced through the holes into the tube atomizer was also investigated. For better understanding the atomization process, an optical pyrometer was used to measure external tube wall temperature, and the two lines method with Sn and Pb thermometric atomic lines (SnI 286.3 and 284.0 nm; PbI 368.3 and 280.2 nm) for measuring the gas phase temperature inside the tube. Entrance of flame gases into the tube atomizer was decisive for the atomization of Cd, Pb, Tl, and Zn. On the other hand, it was demonstrated that the gas phase temperature drops from ca. 1600 °C to 1200 °C when the water jet is introduced into the atomizer. This explains why this method is restricted to more volatile elements. The effect of concomitants usually present in food and water samples was negligible in spectrometers furnished with D2 background correction system. Without background correction, only Na in high concentrations (> 1000 mg l-1) showed significant spectral interference. Limits of detection were 0.65 µg l-1 Cd, 32.0 µg l-1 Pb, 8.0 µg l-1 Tl e 5.0 µg l-1 Zn. The relative standard deviation of measurements (n=3) varied from 3 to 9% and up to 60 samples can be analyzed per hour. In comparison with flame atomic absorption spectrometry with sample introduction by pneumatic nebulization, the limits of detection with BIFF-AAS improved 26 times for Cd, 20 times for Pb, 4 times for Zn and 12 times for Tl. Results obtained for determination of Cd, Pb, and Zn in water and food samples by BIFF-AAS showed good agreement with those obtained by graphite furnace atomic absorption spectrometry, and no differences were found between results at 95% confidence level by the t-test, when using certified reference materials. Recoveries of thallium spiked in food and water samples were in the 98-102% range. For food analysis the BIFF-AAS system showed compatibility with centrifuged slurry food extracts obtained by ultrasound assisted extraction, as well as with the acid digests. Finally, this simple method can be successfully used for the determination of Cd and Pb in foods (useful for checking WHO Codex Alimentarius Commission criteria). It can also be recommended for the determination of Cd and Pb (except classes 1 and 2) in waters, following the CONAMA (National Environmental Council of Brazil) legislation.