ANALYST. JANUARY 1992, VOL. 117 43 Determination of Selenium(iv) by Anodic Stripping Voltammetry Using Extraction With Pentyl Alcohol Hasan Aydin Department of Chemistry, Gazi Universitesi, Fen-Edebiyat Fakultesi, 06500 Ankara, Turkey Abdul Hamid Yahaya Department of Chemistry, University of Malaya, 59100 Kuala Lumpur, Malaysia A method is described for the specific and accurate determination of Se in samples by anodic stripping voltammetry following separation of the Se with pentyl alcohol. The separation procedure removes Cull ions with a single extraction. After the extraction, the acidic solution can be used for the determination of Cull because most of these ions remain in the solution. The height of the anodic stripping peak and the standard additions method were used as a quantitative measurement.Following the extraction, 1 x 10-9 mot dm-3 Se can be determined; the relative standard deviation is 1 % ( n = 4). Keywords : Selenium determination; pent yl alcoh o I; anodic stripping voltam m etr y Selenium can be determined by atomic absorption spec- trometry,' polarography,' cathodic stripping voltammetry,3 anodic stripping voltammetry,J-h titrimetry7 and spec- trometry.8 Although there are many methods for the determi- nation of Se, its determination by anodic stripping voltam- metry is still of interest because of the sensitivity and simplicity of the procedures. Vadja" stripped Se from a mercury electrode in acidic solution and was able to determine down to 8 ppb of Se'". Andrews and Johnson l o significantly extended the sensitivity of the method by electrodeposition of Se'" and stripping from a rotating gold disc electrode.In both of these methods, metal ions such as Cu" and Cd" can interfere. Hamilton et af.h reported a procedure for the determination of Se and Te in electrolytic Cu using anodic stripping voltammetry at a gold film electrode. In this method, Se and Te were separated from Cu by passing an ammoniacal solution of the sample through Chelex-100 resin. Adeloju et al." developed a method for the stripping voltammetric determination of Se in biological materials by direct calibration. They used the stripping voltammetric technique following ion-exchange separation of Se from the bulk of the acid digested matrix. Aydin and SomerS studied the anodic stripping voltam- metric determination of Se in the presence of Cu" ions and demonstrated that 2 x 10-8 mol dm--i Se could be determined in the presence of Cull at concentrations of Cu" up to 10% of that of Se with a relative standard deviation of 3.7% ( n = 6).I t was shown that when the amount of Se lay in the range 1.8 < Y < 11, where Y is the concentration ratio of Cu to Se, Se could be determined without elimination of Cu from the solution. When the concentration ratio was less than 1.8, the peak of the intermetallic Cu-Se compound did not increase with the addition of Se because of the absence of free Cu. At a ratio of Cu to Se greater than 11, the Cu peak overlapped the peak of the intermetallic compound. This paper describes a simple voltammetric method for the determination of Se in the presence of Cu" ions based on the use of a combination of voltammetric techniques and extrac- tion with pentyl alcohol.The proposed method permits the determination of Se in the presence of Cull ions at low and trace levels. Experimental Apparatus The apparatus and plating of the graphite wax test electrode with mercury have been described previously. l 2 Reagents All reagents were of analytical-reagent grade. Triply distilled water was used for preparation of all solutions and at all other stages of analysis. Stock solutions of 0.16 mol dm-3 HgCI?, 1 X lo-? mol dm-3 Se03'- and 1 x 10-1 mol dm-3 Cull were prepared by dissolving the appropriate amounts of the salts in water. These solutions were further diluted to the appropriate concentrations.Procedure The sample solution was prepared by mixing Sel" and Cull solutions containing various concentrations of HN03. It was then mixed with pentyl alcohol and the mixture allowed to stand so that the two phases could separate. Solutions of NaOH of various concentrations were added to the organic phase. After shaking, the mixture was set aside for 1 d to allow the two phases to separate. In order to determine the amount of Se and Cu, 70 PI of concentrated HC104 were added to 100 PI of the aqueous phase. The solution was heated to boiling for about 1-2 min and 5 ml of triply distilled water were added. Nitrogen was passed through the solution to effect de-oxygen- ation prior to plating of Sel" and Cu" ions at potentials between -0.10 and -0.30 V. After a short rest period (30-40 s) a potential sweep of 60 mV s-I was applied in the positive direction and the current was recorded.The standard addi- tions were made by using 1 x 10-5 or 1 x 10-6 mol dm-3 Se03'- or Cu" solutions. The same procedure was used for preparation of the blank solutions, without Se and Cu. Results and Discussion Optimization of Experimental Conditions Several experiments were performed in order to determine the optimum conditions following extraction with pentyl alcohol. The best results were obtained at pH 1-2. Therefore, the pH of the plating solution was adjusted to the appropriate value by using HC104. The sensitivity of the method was affected by the deposition potential, Ed, the deposition time, the scan rate and the thickness of the mercury film.The peak height of Se increased with an increase in the negative deposition potential up to a limiting value of -0.350 V. Hence, a potential between -0.10 and -0.30 V was used and the electrode was again plated to the appropriate thickness as described under Procedure.44 ANALYST, JANUARY 1992, VOL. 117 Interference of Cu The specific and accurate determination of Se in various samples by stripping voltammetry requires the separation of this element from interferents.11.13-16 Copper is the ion that interferes most seriously among the metal ions because its deposition potential is close to that of Se'". The procedure developed in this work is designed to prevent the interference of Cu. The procedure can also be used to eliminate other interferents, for example Pb and Cd, and also those having deposition potentials close to that of Se or fairly small solubility products with Se'" ions.As shown in Fig. 1, a broad peak was observed at about 0 V. After the application of the procedure, a peak was only observed at about 0.06 V. The height of this peak increased on addition of S e 0 3 2 - or Cu" ions. After the addition of a solution of Cu", the peak potential was shifted towards a more n Ix -0.2 0.0 0.2 EdN versus SCE Fig. 1 Interference of Cu on the determination of Se. A. Sample solution (100 pl) (23 O00 ppb of Cu and 707 ppb of Se) + HCIOl (70 pl) + triply distilled water (5 ml); B. sample solution (100 pl), taken after extraction with pentyl aleohol, + HCIOl (70 pl) + triply distilled water (5 ml); and C, as for B + 707 ppb of H2Se03 (100 PI).Measurement of the peak height is indicated by x Ix -0.2 0.0 0.2 EdN versus SCE Fig. 2 Effect of Cu as an interferent on the determination of Se. A, Sample solution (100 pl) (2300 pb of Cu and 707 ppb of Se) + HCI04 (70 pl) + triply distilled water 8 ml); and B, sam le solution (100 pl), taken after extraction with pentyl alcohol, + H&O4 (70 pl) + triply distilled water ( 5 ml). Measurement of the peak height is indicated by x positive value. This showed that the ratio of the concentration of Cu to Se in the medium was less than 1.8. When synthetic sample solutions (3-6 in Table 4) were used directly, either two peaks or a single peak were observed for Cu and Cu-Se, respectively. However, after the extraction of Sel", the peaks for Cu and Cu-Se were not observed (Fig.2). Extraction of Se The results in Tables 1-3 show the appropriate concentrations of HN03 and NaOH that are necessary for the successful determination of Se in samples containing Cull ions. When the NaOH concentration is less than 1 rnol dm-3, the amount of Se extracted decreases and the Cu peak is not observed. It was Table 1 Determination of Se in NaOH solution (90% confidence interval was used) Se found in NaOH Synthetic HN03/ NaOH/ Se added to solution* 1 1 3 707 7 538k 10 1 1 3 707 6 543+2 2 1 1 707 7 706+ 1 7 706+ 1 2 1 1 707 3 0.1 1 707 5 698k5 3 0.1 1 707 7 707k 1 4 0.1 0.1 707 6 21 k 6 4 0.1 0.1 707 6 21 + 2 sample mol dm-3 mol dm-3 acid (ppb) n (PPb) * k standard deviation. Table 2 Determination of Cu in NaOH solution (9070 confidence interval was used) Cu found Cu added in NaOH solution* sample mol dm-3 mol dm-3 (ppm) n (10-3ppm) 1 1 3 23 5 115f20 2 1 1 23 6 13 _+ 3 3 0.1 1 23 7 1 3 f 1 4 0.1 0.1 23 5 Nopeak Synthetic HN03/ NaOW to acid * k standard deviation.Table 3 Determination of Se in acidic solution after extraction with pentyl alcohol (90% confidence interval was used) Synthetic HN03/ NaOH/ Se added to Se found in sample mol dm-3 mol dm-3 acid (ppb) n acid* (ppb) 1 1 3 707 8 114 k 5 1 1 3 707 8 1 1 2 f 3 2 1 1 707 6 38 k 4 2 1 1 707 7 3 3 5 8 3 0.1 1 707 5 66 k 7 3 0.1 1 707 5 66 k 2 4 0.1 0.1 707 6 92 k 12 4 0.1 0.1 707 5 87 f 6 * k standard deviation. Table 4 Application of the method to synthetic samples containing Se'" and Cu" ions (90% Confidence interval was used) Concentration of ions in 0.1 mol dm-3 HN03 Synthetic sample Cu (ppb) Se (ppb) 1 23 000 707 2 23 000 707 3 2 300 707 4 2 300 707 S 690 707 6 690 707 * f standard deviation.Se extracted" n (PPb) 6 699 k 7 6 693 k 4 6 693 k 5 6 699 k 3 6 693 k 3 6 701 kSANALYST, JANUARY 1992, VOL. 117 4s found that the NaOH concentration had to be equal to or greater than 1 mol dm-3 for the best results. Pentyl alcohol was selected for the extraction because it extracted Se selectively under the given conditions and was immiscible with water. The results in Table 3 show that the extraction procedure permits almost complete recovery of Se in the presence of Cu. The results in Tables 1 and 2 show that HN03 and NaOH concentrations of 0.1 and 1 mol dm-3, respectively, are necessary for the success of the method.Above 5.5 mol dm-3 HN03, pentyl alcohol did not separate from water. In order to determine the sensitivity of the method, 23 ppm of Cu were used and the results are shown in Table 2. The results in Tables 2 and 4 demonstrate that the extraction procedure was successful in preventing the interfer- ence of Cu. As shown in Tables 1 and 3, Se could be extracted from acidic solution. The reaction occurring in acidic media is: SeO? H3SeO3 or + 2ROH $ (RO)?SeO + H20 (R = alkyl) I t was found that this reaction was reversible and depended on the pH of the medium. For this reason, the concentration of NaOH must be greater than 0.1 mol dm-3 for the quantitative extraction of Se. Conclusion Anodic stripping voltammetry has been shown to be a sensitive and accurate method for the determination of Se.The minimum recovery of Se is 76%. The method is reproducible with a relative standard deviation of 1% for the determination of Se. 1 2 3 4 5 6 7 8 9 10 1 1 12 13 14 1s 16 References Sturgeon. R. E., Willie. S. N . , and Berman. S. S.. Anal. Chcm.. 1985.57.6. Hasdcmir. E., and Somcr. G . , Analyst, 1990. 115. 297. Dennis, B . L . , Moyers. J . L . , and Wilson. G . S . . Anal. Cliem.. 1976.48, 161 1. Poscy, R . S . , and Andrews, R. W., Anal. Chim. Actu, 1981, 124. 107. Aydin, H., and Somcr, G.. Anul. Sci., 1989, 5 , 89. Hamilton, T. W.. Ellis. J . , and Florence, T. M.. Anul. Clrim. Acru, 1979. 110. 87. Aydin. H.. and Somcr, G.. Tulunra, 1989, 36, 723. Bodini. M. E . , Pardo, J., and Arancibia. V . . Tuluntu, 1990,37, 439. Vadja, F., Actu Chim. Acud. Sci. Hung., 1970, 63, 257. Andrcws, R. W.. and Johnson. D. C.. And. Cliem.. 1975, 47. 294. Adeloju, S. B., Bond, A. M., Briggs, M. H.. and Hughes, H. C., Anul. Cliem.. 1983, 55. 2076. Somer, G . , and Aydin, H., Analyst. 1985. 110. 631. Howard, A. G., Gray, M. R., Waters, A. J . . and Orcmichic. A. R.. Anal. Chim. Actu. 1980, 118, 87. Forbes, S . , Bound, G. P., and West. T., Tuluntu, 1979,26.473. Andrews, R . W., and Johnson, D. C., Anal. Chem., 1976,48, 1056. Rasmussenh. L., And. Clzirn. Acru, 1981, 125, 117. Paper 1l01519B Received April 2, 1991 Accepted August 12, 1991