- Email: [email protected]
Extraction and preconcentration of selenium from aqueous solutions and its determination in water and hair samples by atomic-absorption spectrophotometry
Talanta, Vol. 34, No. 3, pp. 337-340, 1987 Printed in Great Britain. All rights reserved
0039-9140/8783.00+ 0.00 Copyright 0 1987Pcqamon Journals Ltd
EXTRACTION AND PRECONCENTRATION OF SELENIUM FROM AQUEOUS SOLUTIONS AND ITS DETERMINATION IN WATER AND HAIR SAMPLES BY ATOMICABSORPTION SPECTROPHOTOMETRY M. EJAZ Department of Chemical Engineering, College of Engineering King Abdulaxiz University, P.O. Box 9021, Jeddah-21413, Saudi Arabia M. A. QURESHI* Pakistan Institute of Nuclear Science and Technology, Nilore, Rawalpindi, Pakistan (Received 16 October 1984. Revised 16 September
1986. Accepted 10 October 1986)
Summary-Several organic solvents, including benzene, xylene, toluene, nitrobenzene, chloroform, carbon tetrachloride, chlorobenxene and high molecular-weight pyridines such as 6(5-nonyl)pyridine, 2-hexylpyridine and benxylpyridine have been investigated as components of systems for the extraction and preconcentration of selenium from nitric acid solutions containing iodide. The results are discussed in terms of choice of reagents and the acid and iodide concentrations, and of several other parameters affecting the extraction. The utility of the method for separation of selenium from aqueous solution has been evaluated. The method has been used for preconcentration of trace levels of selenium from water and hair samples for determination by atomic-absorption spectrophotometry.
In our investigations on the use of high molecularweight pyridines in extraction of selenium, we have found that when benzene is used as diluent it enhances the extraction under certain experimental conditions. The present work was therefore devoted to study of the distribution of selenium between several common organic solvents (in addition to various new nitrogen-donor compounds) and nitric acid containing iodide. There seems to be no published information concerning solvent extraction of selenium(N) with the solvents and aqueous conditions reported here, although chloride complexes of selenium have been extracted’-’ with liquid-anion exchangers, solvating solvents, and chelating extractants. The method is rapid and selective, and the reaction conditions have wide tolerance; it has been used for determination of the element in water and hair samples by atomic-absorption spectrophotometry (AAS). Selenium can be determined without preconcentration, by using hydride generation and silica-tube AAS but this is prone to several interferences.~’ EXPERIMENTAL Reagents
Standard WOO-pg/ml stock solution of selenium(IV) was prepared by dissolving 0.5 g of pure selenium pellets in a mixture of 2 ml of concentrated sulphuric acid and 3 ml of concentrated nitric acid and making up to volume in a 500~ml standard flask with demineralized distilled water. All *Present address: Research Institute, University of Petroleum and Minerals, Dhahran, Saudi Arabia.
other reagents used were Merck analytical grade. 4(5-NonyBpyridine (Npy), 2-hexylpyridine- (Hpy); and benzvlovridine (BDv) were obtained from K & K Laboratories Inc., Plain&~, N.Y. Their characteristics have been reported elsewhere.89 A Zeiss F M D-3 atomic-absorption spectrophotometer equipped with a Perkin-Elmer graphite-furnace atomizer unit (HGA-74), a Servogor S RE/541 recorder and a deuterium lamp for background compensation, was used. The selenium hollow-cathode lamp was operated at 20 mA and the wavelength selector was set at 392.1 nm for monitoring the 196-nm line, in accordance with the instruction manual for the atomic-absorption spectrometer. Eppendorf pipettes were used for injecting the sample (10 ~1) into the graphite tube of the HGA unit. The heating programme was 60 set drying at 150”, 60 set ashing at 600”, 5 set atomization at 2650”, 10 set cleaning at 2650”, in the gas-stop mode, with background correction. Tracers
Selenium-75 (r,,2 = 120.4 days) tracer was obtained by neutron-activation of SeG, in the PARR-l research reactor of the Pakistan Institute of Nuclear Science and Technology. The concentration of selenium in the original aqueous solutions was <10-6M. All other tracers used were obtained by (n, y) reactions, or by separation of daughter nuclides from the parents without a carrier,iO or were obtained from the Radiochemical Centre, Amersham. The equipment used for the radioassay has been described in earlier papers.9,” Extraction procedure The distribution coefficients of selenium(IV) and the other elements were determined by procedures analogous to those used for zinci and copper. I3 One ml of 0.W benzene solution of ligand, or of the pure ligand, was shaken vigorously for 5 min with I ml of an aqueous solution of
337
338
M.
EJAZ
and M. A,
nitric acid containing potassium iodide and the radioisotope of the test element (- 10’ counts/100 set/ml). The phases were then separated by centrifugation. Aliquots (500 ~1) were removed from each phase and the concentration of the test element determined radiometrically. Preparation of calibration graph Standard aqueous solutions of selenium in the range
0.14.0 pg/ml were prepared by dilution of the stock solution and used for calibration by extraction of the selenium with toluene, and its determination by AAS. Analysis of hair samples A representative sample of hair was taken in an acidwashed polyethylene tube and washed with detergent, then rinsed three times with distilled, demineralized water, dried at _ 40” and cooled. Part of it was then immersed in 20 ml of I-pg/ml selenium standard solution for 90 hr. A 0.5-g portion of each hair sample (selenium-treated and untreated) was added to 4 ml of concentrated nitric acid and left to stand in it overnight to prevent foaming on subsequent heating. The samples were then heated at 60-70” until a clear solution was obtained. The reliability of the decomposition method was checked by adding radiotracer to samples and standards before the digestion. The clear solution was evaporated (at 60-70”) nearly to dryness, and the residue was taken up in 0.5M nitric acid, made 0.2-OSM in potassium iodide, and extracted. Standard selenium solutions and a reagent blank were prepared and treated in the same way. The selenium in the extract was determined by AAS. About 0.5 g of the hair sample which had been dipped in selenium solution was weighed into a narrow quartz tube and irradiated for 30 hr in the PINSTECH reactor at a neutron flux of 2 x 10” n.cm-2.sec’, with a selenium standard taped alongside. The irradiated samples were allowed to decay for 4 weeks. The 265-keV peak was then used for measurement with a Ge(Li) detector coupled with a charge-sensitive preamplifier (Canberra Model 970D) and a spe&oscopic amplifier (O&c Model 451). The pulses
from the amulifier were analvsed by a Nuclear Data NCD4410 compuierized multichannel analyser. The resolution of the detector was 2.9 keV at the 1332 keV gamma-ray of
QURFSHI
Analysis of water samples
Water samples were not given any prior treatment and the amount of selenium was determined by the standardaddition method. A lo-ml water sample, made 0.5M in potassium iodide and containing known added amounts of selenium, was equilibrated with 1 ml of toluene. The digestion procedure used for the hair samples had been found not to be necessary for water samples. RESULTS AND DISCUSSION
In preliminary experiments, the partition behaviour of selenium between nitric acid and O.lM 4-(S-nonyl)pyridine (Npy) in benzene was investigated but the distribution coefficients (D) were very low. In line with some of our previous investigations ‘Z-14these experiments were repeated with 0.02M potassium iodide in the aqueous phase, and it was found that selenium(IV) could be quantitatively extracted from 1M nitric acid/O.O2M potassium iodide by O.lM Npy in benzene. It was also found that even with only 10e4M Npy in benzene the value of D was 40, suggesting that benzene itself had some solvent action. This led to investigation of the extraction of selenium by benzene and several other common solvents, as shown in Fig. 1. Xylene and heptane were also investigated but gave relatively poor extraction. In general, with increasing acid concentration the D values passed through a maximum at 0%1M nitric acid concentration. The decrease at higher acidity is probably due to decrease in the iodide concentration by oxidation to molecular iodine, as shown by most of the organic phases becoming dark red at >4M nitric acid concentration. The efficiency of the extraction systems ranks in the order 0.1 M Npy/benzene > carbon tetrachloride > benzene 2 cyclohexane > chloroform 2 nitrobenzen
‘r
(cl
log [acid], M Fig. 1. Dependence of the distribution coefficient of selenium on the concentration of nitric acid solutions which were 0.02M with respect to iodide, with various solvents. A: (1) toluene; (2) carbon tetrachloride; (3) cyclohexane; (4) benzene. B: (1) O.OM Npy/benzene; (2) Hpy; (3) Bpy. C: (1) nitrobenzene; (2) chlorobenzene; (3) chloroform.
Extraction of selenium > chlorobenzene > toluene > benzylpyridine > 2hexylpyridine. The high efficiency of the Npy/benzene system may be due to synergism. Carbon tetrachloride gives high extraction efficiency over a wider acidity range than the other systems, and this may be associated with its low dielectric constant, presumably because extraction of ion-association complexes is less favourable with solvents of relatively high dielectric constant. The effect of the iodide concentration on the benzene, O.lM Npy/benzene and toluene systems was investigated as a function of nitric acid concentration (Fig. 2). There is generally a linear increase in log D with increasing iodide concentration up to 0.2M, and then a levelling off. In general the efficiency is highest with an acid concentration of -0SM. The effect of chloride, oxalate, acetate, citrate and ascorbate on extraction of selenium(IV) at 0.5M nitric acid and 0.5M iodide concentrations was found to be generally insignificant up to concentrations of about 0. lM, but in some cases (depending on the combination of anion and solvent system) the degree of extraction was decreased at higher concentrations, offering convenient stripping systems. The mechanism of the extraction of selenium by the pure solvents is not known. The extraction of nitric acid with aromatic hydrocarbons (Ar) has been attributed to formation of a species such Ar.&O.HNO,. I5 It is possible that a species of the type H2SeI, could be formed and extracted in this way, but it seems more likely that a species such as SeI, is extracted. The similarity of the extraction curves certainly suggests a common mechanism for all the systems. The increased extraction with pure
3
3- (Al
r
(Bl
339
Bpy or with benzene containing Npy could be due to the additional formation of an ion-association complex of the type SeI; . NpyH+. We have found that several solvents give quantitative extraction, including benzene, toluene, carbon tetrachloride, cyclohexane and Npy. Of these we examined Npy/ benzene and toluene more extensively, with O.OlM nitric acid-1M potassium iodide and OSM nitric acidaSM potassium iodide respectively. For the Npy/benzene system, the conditions chosen were those giving simultaneous extraction of selenium and common toxic metal ions.‘2,‘6 Table 1 shows the distribution coefficients obtained for both systems. The toluene system is much the more selective, but still allows co-extraction of mercury, the halide complexes of which are soluble in a variety of inert solvents.“~i8 The methods can thus be used for the preconccntration of certain toxic elements in solutions. At this stage in the work our stock of Npy became exhausted and the suppliers had stopped making it, so for investigation of applications we had to use the toluene system. Atomic-absorption
measurements
Selenium was extracted into toluene from aqueous solutions and determined at 392 nm by AAS. Extraction of selenium is highest in benzene, toluene, carbon tetrachloride and cyclohexane, but toluene was preferred because of its higher boiling point. The calibration graph was very convex (Fig. 3). When water samples spiked with known amounts of selenium were extracted with toluene as described in
3
i 2
(Cl
. * F 3
1
.3
*
l
4
1
0
-1: -2
-1
-1
0
,
I
-1
1
0
Fig. 2. Dependence of the distribution coeflicient of selenium on aqueous iodide concentration for extraction by different solvents from nitric acid media. A, Benzene: (1) O.lM HNO,; (2) OSM HNO,; (3) 1M HNO,. B, O.lM Npy in benzene: (1) 0.05M HNO,; (2) O.lM HNQ; (3) OSM HNO,; (4) lMHN0,. C, Toluene: (1) 0.02M HNOl; (2) O.lM HNO,; (3) 0.5M HNO,; (4) 1M HNO,.
M.
340
BAZ
and M. A. QURESHI
Table 1. Distribution coefhcients of various metal ions with respect to selenium
Table 2. AAS determination water
Distribution coefficients* Concentration, M
Species Z) Mo(VI) Cr(vI) MO(V) Hf(IV)
10-6 C.Ft 10-5 CF 10-S 10-7 lo-’ 10-n IO-’ CF IO-’ lo-’ 10-a
ZrfIW
sn(Ivj
cr’+
La’+ Nd’+ Ce’+ Fe’+
4181.57 1.26 1.09 1.21 0.0 0.0 43 0.0
lolo--* 1O-8 10-s 10-s 10-7 lo-’ 10-7 10-6 1o-5 1O-8 1o-6
d’ In’+ Pm’+ sr2+ car+ Hg2+ Zn2+
10-s
I@+ cu2+ CS+
&+
0.08 0.0 0.0
1.6 32 0.0 0.0 0.2 1400 910 640 33.2 0.0
0.1 0.0 0.0 0.0 907 0.02 0.03 1.0 0.0 -
10-7
0.0
the results shown in Table 2 were It was further found that the extraction is > 90% even with 100: 1 aqueous phase/toluene
volume ratio. Triplicate AAS analysis of the hair sample gave selenium contents of 0.04, 0.03, OMpg/g. For the hair sample impregnated by immersion in l-fig/g selenium solution triplicate AAS analysis gave 0.22, 0.19, 0.21 rg/g, and neutron-activation analysis give 0.21-0.25 pg/ml, in good agreement with the AAS result.
Acknowledgements-The authors are grateful to Dr. Shamim A. Chaudri and Miss W. Dil for their useful contribution to certain atomic-absorption measurements.
REFERENCES
1.
0.98 0.0 0.0
*A: O.lM HNO,/lM
KI (O.lM Npy/benzcne); HNO,/O.SM ICI (toluene). tC.F = Carrier-free.
B: 0.5M
2. 3. 4. 5. 6. 7.
800 700
8. 9.
F 600 t f
500
x 8 *
400
2
300
10. 11. 12. 13.
I
14. 15. .’ I
0
I
I 3
1 Concentrhm
@pm)
Fig. 3. Calibration curve.
0.09 f 0.00 0.18 f 0.02 0.30 f 0.02
obtained.
0.0
0.9 0.0 0.0
0.16 91 0.0
0.10
the procedure,
0.05
lo-’ 10-3 1o-5 10-6 1O-8
Se found in I-ml toluene extract, pg
0.20 0.30
303.2 0.21 0.0 0.0 0.15
-
1O-6
Se added to 10 ml of HTO, fig
B
A
of selenium in
A 4
16. 17. 18.
C.
Fisher, K. Gloe, P. Muhl and V. V. Bagreev, J. fnorg. Nucl. Chem., 1978, 40, 1793. Y. Marcus, Co-ord. Chem. Reo., 1967, 2, 195. N. Jordanov and L. Futekov, Talunfa, 1965, 12, 371. G. A. Cutter, Anal. Chim. Acta, 1978, 98, 59. F. C. Pierce and H. R. Brown, Anal. Chem., 1976, 48, 693. I&m, ibid., 1977, 49, 693. R. M. Brown, Jr., R. C. Fry, J. L. Moyers, S. J. Northway, M. B. Denton and G. S. Wilson, ibid., 1981, 53, 1560. M. Ejaz and M. Iqbal, ibid., 1975, 47, 936 M. Ejaz, M. Iqbal, S. A. Chaudri and R. Ahmed, Sepn. Sci., 1976, 11, 255. M. S. Fadeeva, 0. N. Panlov and V. V. Bakunina, Zh. Neorgan. Khim., 1958, 3, 165. M. Ejaz, Anal. Chem., 1976, 48, 1158. M. Ejaz and Shamus-ud-Zuha, ibid., 1978, 50, 740. M. Ejaz, Shamus-ud-Zuha, D. Wasim and S. A. Chaudri, Talanta, 1981, 28, 441. M. Ejaz, S. Ahrned, D. Wasim and S. A. Chaudri, ibid., 1979, 26, 503. A. P. Bobylev and L. N. Komissarova, Russ. J. Inorg. Chem., 1978, 23, 1186, and references therein. M. Ejaz, M. A. Qureshi and Shamus-ud-Zuha, Sepn. Sci Technol., 1981, 16, 291. I. Eliezer, J. Chem. Phys., 1965, 42, 3625. Idem, ibid., 1965, 41, 3276.
0039-9140/8783.00+ 0.00 Copyright 0 1987Pcqamon Journals Ltd
EXTRACTION AND PRECONCENTRATION OF SELENIUM FROM AQUEOUS SOLUTIONS AND ITS DETERMINATION IN WATER AND HAIR SAMPLES BY ATOMICABSORPTION SPECTROPHOTOMETRY M. EJAZ Department of Chemical Engineering, College of Engineering King Abdulaxiz University, P.O. Box 9021, Jeddah-21413, Saudi Arabia M. A. QURESHI* Pakistan Institute of Nuclear Science and Technology, Nilore, Rawalpindi, Pakistan (Received 16 October 1984. Revised 16 September
1986. Accepted 10 October 1986)
Summary-Several organic solvents, including benzene, xylene, toluene, nitrobenzene, chloroform, carbon tetrachloride, chlorobenxene and high molecular-weight pyridines such as 6(5-nonyl)pyridine, 2-hexylpyridine and benxylpyridine have been investigated as components of systems for the extraction and preconcentration of selenium from nitric acid solutions containing iodide. The results are discussed in terms of choice of reagents and the acid and iodide concentrations, and of several other parameters affecting the extraction. The utility of the method for separation of selenium from aqueous solution has been evaluated. The method has been used for preconcentration of trace levels of selenium from water and hair samples for determination by atomic-absorption spectrophotometry.
In our investigations on the use of high molecularweight pyridines in extraction of selenium, we have found that when benzene is used as diluent it enhances the extraction under certain experimental conditions. The present work was therefore devoted to study of the distribution of selenium between several common organic solvents (in addition to various new nitrogen-donor compounds) and nitric acid containing iodide. There seems to be no published information concerning solvent extraction of selenium(N) with the solvents and aqueous conditions reported here, although chloride complexes of selenium have been extracted’-’ with liquid-anion exchangers, solvating solvents, and chelating extractants. The method is rapid and selective, and the reaction conditions have wide tolerance; it has been used for determination of the element in water and hair samples by atomic-absorption spectrophotometry (AAS). Selenium can be determined without preconcentration, by using hydride generation and silica-tube AAS but this is prone to several interferences.~’ EXPERIMENTAL Reagents
Standard WOO-pg/ml stock solution of selenium(IV) was prepared by dissolving 0.5 g of pure selenium pellets in a mixture of 2 ml of concentrated sulphuric acid and 3 ml of concentrated nitric acid and making up to volume in a 500~ml standard flask with demineralized distilled water. All *Present address: Research Institute, University of Petroleum and Minerals, Dhahran, Saudi Arabia.
other reagents used were Merck analytical grade. 4(5-NonyBpyridine (Npy), 2-hexylpyridine- (Hpy); and benzvlovridine (BDv) were obtained from K & K Laboratories Inc., Plain&~, N.Y. Their characteristics have been reported elsewhere.89 A Zeiss F M D-3 atomic-absorption spectrophotometer equipped with a Perkin-Elmer graphite-furnace atomizer unit (HGA-74), a Servogor S RE/541 recorder and a deuterium lamp for background compensation, was used. The selenium hollow-cathode lamp was operated at 20 mA and the wavelength selector was set at 392.1 nm for monitoring the 196-nm line, in accordance with the instruction manual for the atomic-absorption spectrometer. Eppendorf pipettes were used for injecting the sample (10 ~1) into the graphite tube of the HGA unit. The heating programme was 60 set drying at 150”, 60 set ashing at 600”, 5 set atomization at 2650”, 10 set cleaning at 2650”, in the gas-stop mode, with background correction. Tracers
Selenium-75 (r,,2 = 120.4 days) tracer was obtained by neutron-activation of SeG, in the PARR-l research reactor of the Pakistan Institute of Nuclear Science and Technology. The concentration of selenium in the original aqueous solutions was <10-6M. All other tracers used were obtained by (n, y) reactions, or by separation of daughter nuclides from the parents without a carrier,iO or were obtained from the Radiochemical Centre, Amersham. The equipment used for the radioassay has been described in earlier papers.9,” Extraction procedure The distribution coefficients of selenium(IV) and the other elements were determined by procedures analogous to those used for zinci and copper. I3 One ml of 0.W benzene solution of ligand, or of the pure ligand, was shaken vigorously for 5 min with I ml of an aqueous solution of
337
338
M.
EJAZ
and M. A,
nitric acid containing potassium iodide and the radioisotope of the test element (- 10’ counts/100 set/ml). The phases were then separated by centrifugation. Aliquots (500 ~1) were removed from each phase and the concentration of the test element determined radiometrically. Preparation of calibration graph Standard aqueous solutions of selenium in the range
0.14.0 pg/ml were prepared by dilution of the stock solution and used for calibration by extraction of the selenium with toluene, and its determination by AAS. Analysis of hair samples A representative sample of hair was taken in an acidwashed polyethylene tube and washed with detergent, then rinsed three times with distilled, demineralized water, dried at _ 40” and cooled. Part of it was then immersed in 20 ml of I-pg/ml selenium standard solution for 90 hr. A 0.5-g portion of each hair sample (selenium-treated and untreated) was added to 4 ml of concentrated nitric acid and left to stand in it overnight to prevent foaming on subsequent heating. The samples were then heated at 60-70” until a clear solution was obtained. The reliability of the decomposition method was checked by adding radiotracer to samples and standards before the digestion. The clear solution was evaporated (at 60-70”) nearly to dryness, and the residue was taken up in 0.5M nitric acid, made 0.2-OSM in potassium iodide, and extracted. Standard selenium solutions and a reagent blank were prepared and treated in the same way. The selenium in the extract was determined by AAS. About 0.5 g of the hair sample which had been dipped in selenium solution was weighed into a narrow quartz tube and irradiated for 30 hr in the PINSTECH reactor at a neutron flux of 2 x 10” n.cm-2.sec’, with a selenium standard taped alongside. The irradiated samples were allowed to decay for 4 weeks. The 265-keV peak was then used for measurement with a Ge(Li) detector coupled with a charge-sensitive preamplifier (Canberra Model 970D) and a spe&oscopic amplifier (O&c Model 451). The pulses
from the amulifier were analvsed by a Nuclear Data NCD4410 compuierized multichannel analyser. The resolution of the detector was 2.9 keV at the 1332 keV gamma-ray of
QURFSHI
Analysis of water samples
Water samples were not given any prior treatment and the amount of selenium was determined by the standardaddition method. A lo-ml water sample, made 0.5M in potassium iodide and containing known added amounts of selenium, was equilibrated with 1 ml of toluene. The digestion procedure used for the hair samples had been found not to be necessary for water samples. RESULTS AND DISCUSSION
In preliminary experiments, the partition behaviour of selenium between nitric acid and O.lM 4-(S-nonyl)pyridine (Npy) in benzene was investigated but the distribution coefficients (D) were very low. In line with some of our previous investigations ‘Z-14these experiments were repeated with 0.02M potassium iodide in the aqueous phase, and it was found that selenium(IV) could be quantitatively extracted from 1M nitric acid/O.O2M potassium iodide by O.lM Npy in benzene. It was also found that even with only 10e4M Npy in benzene the value of D was 40, suggesting that benzene itself had some solvent action. This led to investigation of the extraction of selenium by benzene and several other common solvents, as shown in Fig. 1. Xylene and heptane were also investigated but gave relatively poor extraction. In general, with increasing acid concentration the D values passed through a maximum at 0%1M nitric acid concentration. The decrease at higher acidity is probably due to decrease in the iodide concentration by oxidation to molecular iodine, as shown by most of the organic phases becoming dark red at >4M nitric acid concentration. The efficiency of the extraction systems ranks in the order 0.1 M Npy/benzene > carbon tetrachloride > benzene 2 cyclohexane > chloroform 2 nitrobenzen
‘r
(cl
log [acid], M Fig. 1. Dependence of the distribution coefficient of selenium on the concentration of nitric acid solutions which were 0.02M with respect to iodide, with various solvents. A: (1) toluene; (2) carbon tetrachloride; (3) cyclohexane; (4) benzene. B: (1) O.OM Npy/benzene; (2) Hpy; (3) Bpy. C: (1) nitrobenzene; (2) chlorobenzene; (3) chloroform.
Extraction of selenium > chlorobenzene > toluene > benzylpyridine > 2hexylpyridine. The high efficiency of the Npy/benzene system may be due to synergism. Carbon tetrachloride gives high extraction efficiency over a wider acidity range than the other systems, and this may be associated with its low dielectric constant, presumably because extraction of ion-association complexes is less favourable with solvents of relatively high dielectric constant. The effect of the iodide concentration on the benzene, O.lM Npy/benzene and toluene systems was investigated as a function of nitric acid concentration (Fig. 2). There is generally a linear increase in log D with increasing iodide concentration up to 0.2M, and then a levelling off. In general the efficiency is highest with an acid concentration of -0SM. The effect of chloride, oxalate, acetate, citrate and ascorbate on extraction of selenium(IV) at 0.5M nitric acid and 0.5M iodide concentrations was found to be generally insignificant up to concentrations of about 0. lM, but in some cases (depending on the combination of anion and solvent system) the degree of extraction was decreased at higher concentrations, offering convenient stripping systems. The mechanism of the extraction of selenium by the pure solvents is not known. The extraction of nitric acid with aromatic hydrocarbons (Ar) has been attributed to formation of a species such Ar.&O.HNO,. I5 It is possible that a species of the type H2SeI, could be formed and extracted in this way, but it seems more likely that a species such as SeI, is extracted. The similarity of the extraction curves certainly suggests a common mechanism for all the systems. The increased extraction with pure
3
3- (Al
r
(Bl
339
Bpy or with benzene containing Npy could be due to the additional formation of an ion-association complex of the type SeI; . NpyH+. We have found that several solvents give quantitative extraction, including benzene, toluene, carbon tetrachloride, cyclohexane and Npy. Of these we examined Npy/ benzene and toluene more extensively, with O.OlM nitric acid-1M potassium iodide and OSM nitric acidaSM potassium iodide respectively. For the Npy/benzene system, the conditions chosen were those giving simultaneous extraction of selenium and common toxic metal ions.‘2,‘6 Table 1 shows the distribution coefficients obtained for both systems. The toluene system is much the more selective, but still allows co-extraction of mercury, the halide complexes of which are soluble in a variety of inert solvents.“~i8 The methods can thus be used for the preconccntration of certain toxic elements in solutions. At this stage in the work our stock of Npy became exhausted and the suppliers had stopped making it, so for investigation of applications we had to use the toluene system. Atomic-absorption
measurements
Selenium was extracted into toluene from aqueous solutions and determined at 392 nm by AAS. Extraction of selenium is highest in benzene, toluene, carbon tetrachloride and cyclohexane, but toluene was preferred because of its higher boiling point. The calibration graph was very convex (Fig. 3). When water samples spiked with known amounts of selenium were extracted with toluene as described in
3
i 2
(Cl
. * F 3
1
.3
*
l
4
1
0
-1: -2
-1
-1
0
,
I
-1
1
0
Fig. 2. Dependence of the distribution coeflicient of selenium on aqueous iodide concentration for extraction by different solvents from nitric acid media. A, Benzene: (1) O.lM HNO,; (2) OSM HNO,; (3) 1M HNO,. B, O.lM Npy in benzene: (1) 0.05M HNO,; (2) O.lM HNQ; (3) OSM HNO,; (4) lMHN0,. C, Toluene: (1) 0.02M HNOl; (2) O.lM HNO,; (3) 0.5M HNO,; (4) 1M HNO,.
M.
340
BAZ
and M. A. QURESHI
Table 1. Distribution coefhcients of various metal ions with respect to selenium
Table 2. AAS determination water
Distribution coefficients* Concentration, M
Species Z) Mo(VI) Cr(vI) MO(V) Hf(IV)
10-6 C.Ft 10-5 CF 10-S 10-7 lo-’ 10-n IO-’ CF IO-’ lo-’ 10-a
ZrfIW
sn(Ivj
cr’+
La’+ Nd’+ Ce’+ Fe’+
4181.57 1.26 1.09 1.21 0.0 0.0 43 0.0
lolo--* 1O-8 10-s 10-s 10-7 lo-’ 10-7 10-6 1o-5 1O-8 1o-6
d’ In’+ Pm’+ sr2+ car+ Hg2+ Zn2+
10-s
I@+ cu2+ CS+
&+
0.08 0.0 0.0
1.6 32 0.0 0.0 0.2 1400 910 640 33.2 0.0
0.1 0.0 0.0 0.0 907 0.02 0.03 1.0 0.0 -
10-7
0.0
the results shown in Table 2 were It was further found that the extraction is > 90% even with 100: 1 aqueous phase/toluene
volume ratio. Triplicate AAS analysis of the hair sample gave selenium contents of 0.04, 0.03, OMpg/g. For the hair sample impregnated by immersion in l-fig/g selenium solution triplicate AAS analysis gave 0.22, 0.19, 0.21 rg/g, and neutron-activation analysis give 0.21-0.25 pg/ml, in good agreement with the AAS result.
Acknowledgements-The authors are grateful to Dr. Shamim A. Chaudri and Miss W. Dil for their useful contribution to certain atomic-absorption measurements.
REFERENCES
1.
0.98 0.0 0.0
*A: O.lM HNO,/lM
KI (O.lM Npy/benzcne); HNO,/O.SM ICI (toluene). tC.F = Carrier-free.
B: 0.5M
2. 3. 4. 5. 6. 7.
800 700
8. 9.
F 600 t f
500
x 8 *
400
2
300
10. 11. 12. 13.
I
14. 15. .’ I
0
I
I 3
1 Concentrhm
@pm)
Fig. 3. Calibration curve.
0.09 f 0.00 0.18 f 0.02 0.30 f 0.02
obtained.
0.0
0.9 0.0 0.0
0.16 91 0.0
0.10
the procedure,
0.05
lo-’ 10-3 1o-5 10-6 1O-8
Se found in I-ml toluene extract, pg
0.20 0.30
303.2 0.21 0.0 0.0 0.15
-
1O-6
Se added to 10 ml of HTO, fig
B
A
of selenium in
A 4
16. 17. 18.
C.
Fisher, K. Gloe, P. Muhl and V. V. Bagreev, J. fnorg. Nucl. Chem., 1978, 40, 1793. Y. Marcus, Co-ord. Chem. Reo., 1967, 2, 195. N. Jordanov and L. Futekov, Talunfa, 1965, 12, 371. G. A. Cutter, Anal. Chim. Acta, 1978, 98, 59. F. C. Pierce and H. R. Brown, Anal. Chem., 1976, 48, 693. I&m, ibid., 1977, 49, 693. R. M. Brown, Jr., R. C. Fry, J. L. Moyers, S. J. Northway, M. B. Denton and G. S. Wilson, ibid., 1981, 53, 1560. M. Ejaz and M. Iqbal, ibid., 1975, 47, 936 M. Ejaz, M. Iqbal, S. A. Chaudri and R. Ahmed, Sepn. Sci., 1976, 11, 255. M. S. Fadeeva, 0. N. Panlov and V. V. Bakunina, Zh. Neorgan. Khim., 1958, 3, 165. M. Ejaz, Anal. Chem., 1976, 48, 1158. M. Ejaz and Shamus-ud-Zuha, ibid., 1978, 50, 740. M. Ejaz, Shamus-ud-Zuha, D. Wasim and S. A. Chaudri, Talanta, 1981, 28, 441. M. Ejaz, S. Ahrned, D. Wasim and S. A. Chaudri, ibid., 1979, 26, 503. A. P. Bobylev and L. N. Komissarova, Russ. J. Inorg. Chem., 1978, 23, 1186, and references therein. M. Ejaz, M. A. Qureshi and Shamus-ud-Zuha, Sepn. Sci Technol., 1981, 16, 291. I. Eliezer, J. Chem. Phys., 1965, 42, 3625. Idem, ibid., 1965, 41, 3276.