A method has been developed for the determination of trace selenium by generation of its gaseous hydride using either metallic zinc or sodium borohydride as a reductant and followed by the introduction of the hydride into a premixed argon (entrained air)-hydrogen flame where all atomic fluorescence lines of selenium are simultaneously detected by use of a nondispersive system. The comparison of the zinc and the sodium borohydride reduction methods is discussed in terms of detection limit, precision, and interference. The best attainable detection limits for selenium are 0.3 ng (15 pg/ml) and 0.4 ng (20 pg/ml) with the zinc and sodium borohydride methods, respectively. Analytical calibration curves obtained by measuring peak heights and integrated peak areas of the fluorescence signals are linear over the range of approximately 4 orders of magnitude from the detection limits. Most mineral acids examined in the range up to 2.0 M do not interfere; however, the nitric acid is an exception which gives a depressing interference. The interference was found to be much more severe in the zinc method than in the sodium borohydride technique. The presence of several elements including other hydride-forming elements in 1000-fold ratio to selenium causes a depressing interference, whereas an enhancing interference from tellurium is observed only in the zinc reduction method. The present method has been successfully applied to the determination of selenium in wastewaters and foods.
Taketoshi Nakahara, Syoji Kobayashi, Tatsushi Wakisaka, and Sôichirô Musha, "The Determination of Trace Amounts of Selenium by Hydride Generation-Nondispersive Flame Atomic Fluorescence Spectrometry," Appl. Spectrosc. 34, 194-200 (1980)