Selenium in foods: evaluation of atomic absorption spectrometric techniques involving hydrogen selenide generation and carbon furnace atomization.

The performance of hydrogen selenide generation- and carbon furnace atomization-atomic absorption spectrometry, with a semiautomated trace metal accessory and a carbon rod atomizer, respectively, was evaluated for the determination of selenium in foods. Samples were digested with nitric, perchloric, and sulfuric acids for both procedures. Hydrogen selenide was generated with stannous chloride/potassium iodide/zinc from 20 ml aliquots in a hydrochloric-sulfuric acid medium and directed into an argon-hydrogen-entrained air flame. Sample matrix interferences in the carbon furnace technique were minimized by isolating selenium by precipitation with ascorbic acid and redissolution in nitric-perchloric acid prior to taking 5 mul aliquots for estimation; precision was significantly improved by incorporating 5000 mug nickel/ml into analytical solutions. For the hydrogen selenide and carbon furnace techniques, respectively, the following data were obtained for untreated standard solutions: sensitivity, 0.13 and 10.0 ng/ml; absolute sensitivity, 2.6 and 0.05 ng; detection limit, 0.4 and 90 ng/ml; absolute detection limit, 7 and 0.45 ng. The detection limit and absolute detection limit for samples were 2.5 ng/ml and 50 ng, respectively, for the hydrogen selenide method, and 25 ng/ml and 0.13 ng, respectively, for the carbon furnace method. Taking into consideration these figures and the practical aspects of both methods, the overall performance of the hydrogen selenide method was superior.