RARE METALS Vol. 30, No. 5, Oct 2011, p. 477 DOI: 10.1007/s12598-011-0416-0 Corresponding author: Ramazan Gürkan E-mail: rgurkan@cumhuriyet.edu.tr A novel indicator system for catalytic spectrophotometric determination and speciation of inorganic selenium species (Se(IV), Se(VI)) at trace levels in natural lake and river water samples Ramazan Gürkan, Halil İbrahim Ulusoy, Mehmet Akçay, and Pınar Bulut Department of Chemistry, Faculty of Science, University of Cumhuriyet, TR-58140, Sivas, Turkey Received 10 August 2010; received in revised form 1 November 2010; accepted 10 November 2010 © The Nonferrous Metals Society of China and Springer-Verlag Berlin Heidelberg 2011 Abstract A novel catalytic kinetic method is proposed for the determination of Se(IV), Se(VI), and total inorganic selenium in water based on the catalytic effect of Se(IV) on the reduction of Celestine blue by sodium sulfide at pH 7.0 phosphate buffer. The fixed-time method was adopted for the determination and speciation of inorganic selenium. Under the optimum conditions, the two calibration graphs are linear with a good correlation coefficient in the range 2-20 and 20-200 μg⋅L −1 of Se(IV) for the fixed-time method at 30°C. The experimental and theo- retical detection limits of the developed kinetic method were found to be 0.21 and 2.50 μg⋅L −1 for the fixed-time method (3 min). All of the variables that affect the sensitivity at 645 nm were investigated, and the optimum conditions were established. The interference effect of various cations and anions on the Se(IV) determination was also studied. The selectivity of the selenium determination was greatly improved with the use of the strongly cation exchange resin such as Amberlite IR120 plus as long as chelating agents of thiourea and thiosulphate. The proposed kinetic method was validated statistically and through recovery studies in natural water samples. The relative standard deviations (RSDs) for ten replicate measurements of 2, 10, and 20 μg⋅L −1 of Se(IV) change between 0.35% and 5.58%, while the RSDs for ten replicate measurements of 3, 6, and 12 μg⋅ L −1 of Se(VI) change between 0.49% and 1.61%. Analyses of a certified standard reference material (NIST SRM 1643e) for selenium using the fixed-time method showed that the proposed kinetic method has good accuracy. The Se(IV), Se(VI), and total inor- ganic selenium in lake and river water samples have been successfully determined by this method after selective reduction of Se(VI) to Se(IV). Keywords: selenium speciation; Celestine blue; catalytic effect; kinetic spectrophotometry; natural waters 1. Introduction It is well known that Se is essential to normal functioning of organisms in a very narrow concentration range. Below this range, Se deficiency occurs, and, above it, Se becomes toxic [1]. Moreover, it has been recognized that the bioavailability and/or toxicity of Se are dependent on the species, e.g., selenite and selenate have different absorbing and metabolic pathways in rats [2]. The concentration of se- lenium generally encountered in natural waters is in the ng⋅L −1 or sub-mg⋅L −1 range. Thus, it is important to have selective and sensitive analytical methods to identify each species and determine their levels. For these reasons, Se de- termination and speciation in real samples have become a very interesting research subject in recent years [3-5]. The coupling of a powerful separation tool such as high-perfor- mance liquid chromatography (HPLC) with an element- specific detector has become very popular for this purpose. One of the popular detectors for Se determination is induc- tively coupled plasma mass spectrometry (ICP-MS) [6-7]. However, the detection limit of Se by conventional ar- gon-based ICP-MS is not as low as desired. This is not only because of its relatively high first ionization potential lead- ing to lower ionization efficiency in the plasma, but also be- cause of the severe interferences of 40 Ar 36 Ar, 40 Ar 38 Ar, and 40 Ar 40 Ar to the major isotopes of 76 Se (natural abundance 9.37%), 78 Se (23.77%), and 80 Se (49.61%), respectively. 40 Ar 37 Cl and 81 Br 1 H polyatomic spectral interferences to 77 Se (7.63%) and 82 Se (8.73%) also make the determination of Se very difficult when a sample contains even traces of chloride or bromide. Atomic fluorescence spectrometry (AFS) is a well-established and sensitive alternative detection tech- nique for Se speciation [8-13]. Not only the sensitivity for Se can be greatly improved by hydride generation (HG) sample introduction [14-15], but also the possible interfer- ences from the matrix are eliminated simultaneously. Prior