INVESTIGATION OF THE AUTOMATED-DETERMINATION OF AS, SB AND PI BY FLOW-INJECTION HYDRIDE GENERATION USING IN-SITU TRAPPING ON STABLE COATINGS IN GRAPHITE-FURNACE ATOMIC-ABSORPTION SPECTROMETRY

Authors
Citation
Ho. Haug et Yp. Liao, INVESTIGATION OF THE AUTOMATED-DETERMINATION OF AS, SB AND PI BY FLOW-INJECTION HYDRIDE GENERATION USING IN-SITU TRAPPING ON STABLE COATINGS IN GRAPHITE-FURNACE ATOMIC-ABSORPTION SPECTROMETRY, Fresenius' journal of analytical chemistry, 356(7), 1996, pp. 435-444
Citations number
31
Categorie Soggetti
Chemistry Analytical
ISSN journal
09370633
Volume
356
Issue
7
Year of publication
1996
Pages
435 - 444
Database
ISI
SICI code
0937-0633(1996)356:7<435:IOTAOA>2.0.ZU;2-B
Abstract
Flow-injection hydride generation and in situ concentration of As, Sb and Bi hydrides in graphite furnace atomic absorption spectrometry can be automated by means of a long-term stable trapping reagent replacin g the Pd modifier. In a systematic study, carbide-forming elements (Zr , Nb, Ta, W) and noble metals (Ir, Ir/Mg, Pd/Ir) were investigated as stable adsorbers which require only a single application. Trapping tem perature curves indicate high signals for trapping of As at 750-800 de grees C, Sb at 450-8000 degrees C and Bi at 100-500 degrees C on Zr-co ated tubes. Ir- and Ir/Mg-coated tubes showed a high response for Sb a nd Bi at lower temperatures, but based on signal stability and reprodu cibility (over 400 trapping and atomization cycles tested) the better performance was found with the Zr-coated tubes. The radiotracers Sb-12 5 and Bi-207 were used to measure the hydride generation (> 95% for bo th elements) and trapping efficiency (91% for Sb and 56% for Bi) on th e Zr-coated tube. An adsorptive ''carry-over effect'' was observed wit h Sb and Bi but not with As, and trapping temperatures above 450 degre es C with Sb and 350 degrees C with Bi (the ''critical temperatures'') can lead to errors in absorbance values. On a Zr-coated tube the char acteristic mass was about 16 pg for As, 15 pg for Sb and 9 pg for Bi ( peak height) and the detection limits (3 sigma) were about 0.015, 0.01 0 and 0.027 ng, respectively, with a 1 ml sample loop. The method was tested by the determination of the elements in NIST low-alloy steel ce rtified reference materials.