Determination of selenium by flow injection hydride generation inductively coupled plasma optical emission spectrometry
Author(s) -
Princess C. Hernandez,
Julian F. Tyson,
Peter C. Uden,
Dennis A. Yates
Publication year - 2006
Publication title -
journal of analytical atomic spectrometry
Language(s) - English
Resource type - Journals
eISSN - 1364-5544
pISSN - 0267-9477
DOI - 10.1039/b612658h
Subject(s) - chemistry , selenium , analytical chemistry (journal) , inductively coupled plasma , chromatography , hydride , hydrochloric acid , detection limit , nitric acid , elution , separator (oil production) , plasma , inorganic chemistry , metal , physics , organic chemistry , quantum mechanics , thermodynamics
Two flow injection hydride generation procedures for the determination of selenium with an inductively coupled plasma emission spectrometer capable of monitoring transient signals were developed. In one, the acidified sample was injected into an acid carrier, which merged downstream with a stream of thiourea (for samples containing significant concentrations of copper) then with a stream of borohydride. In the second, the analyte was preconcentrated by passing the sample through an on-line anion-exchanger, followed by elution of the retained selenium eluted with acid into the HG manifold. A systematic study was made of the effect of radiofrequency power, plasma gas flow, stripping gas flow, sodium borohydride and hydrochloric acid concentrations and flow rates, sample acidity and the lengths of mixing and stripping coils. The hydride was stripped from solution and transported to the plasma by the gas normally delivered to the nebulizer, through a direct connection between the gas-liquid separator and the injector tube. The limits of detection for the first method, which ranged from 0.7 to 0.3 m gl 1 for 200-1000 ml sample volumes, were 10 times lower than those for conventional nebulization. The reproducibility (relative standard deviation of 5 successive measurements) was between 2-5% for 10 m gl 1 -100 m gl 1 Se in 10% HCl. The repeatability, based on five measurements per day performed on three consecutive days, was 4-6% for 10 m gl 1 -100 m gl 1 Se. The overall procedure was validated by the determination of selenium in digests of spinach leaves (SRM 1570a), bovine liver (SRM 1577b) and dogfish muscles (DORM-2). The interference from copper was removed by the addition of thiourea at a confluence point in the FI manifold. In the second method, the detection limit was improved further so that for preconcentration for 2, 4 and 8 min at 8 ml min 1 , the limits of detection were 6, 2 and 0.6 ng l 1 , respectively. Improved tolerance to interfering ions due to the separation of the matrix was found for Cl, HCO3 ,S O 4 2 , As(V), Cd(II), Co(II), Cu(II), Fe(III), Ni(II) and Sb(III), although it was found that many of the metallic interfering species were partially retained. The post-column addition of 0.10% (w/v) thiourea improved the tolerance towards Cu(II). Recoveries of selenium from a variety of water samples showed that the resulting procedure was free from interferences.
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom