Selenium determination in biological fluids using Zeeman background correction electrothermal atomic absorption spectrometry.

Procedures for the direct determination of total selenium in urine, serum, and blood using electrothermal atomic absorption spectrometry are presented. In the selected experimental conditions, Zeeman correction is mandatory to compensate for the high background signals. The sample diluted and containing 0.1% (w/v) Triton X-100 is introduced directly into the electrothermal atomizer. A solution containing 15% (w/v) hydrogen peroxide, 0.65% (w/v) nitric acid, and 0.5% (w/v) nickel is injected separately into the atomizer. Calibration is carried out using the standard additions method. The detection limit is 30 pg selenium. If palladium, instead of nickel, is used as the chemical modifier, calibration can be carried out against aqueous standards, and the detection limit is 45 pg. In this case, three separate injections are required to prevent precipitation problems in the automatic injector. The reliability of the procedures is checked by analyzing three certified reference materials and by recovery studies. Mean recoveries are 99.7% for serum, 99.4% for urine, and 100.8% for blood samples. Relative standard deviation values are +/-4.0% for serum, +/-3.9% for urine, and +/-4.5% for blood.