- Email: [email protected]
Determination of sub microgram amounts of selenium in rocks by atomic-absorption spectroscopy
547
!ZHORT COMMUNICATIONS
SPnmmry-A method is described for the amperometric titration of nickel and eudcessive amperometric determination of copper and nickel. Nickel (10-16.0 mg) and copper (10-l l-0 mg) could be determined with an average error of less than 1%. Cobalt interferes but chloride does not. Interference by aluminium, iron(II1) and chromium can be eliminated Zinc and manganese do not interfere if the correct applied voltage is chosen. The procedures can be utilized in the analysis of alloys such as nichrome, Raney nickel, constantan, german silver and manganin. It is best to use the standard addition method for less than 3 mg of nickel.
Talanta, Vol. 22, pp 547-549
Pergamon
Press
1975. Printed in Great Britain
DETERMINATION OF SUB MICROGRAM AMOUNTS OF SELENIUM ROCKS BY ATOMIC-ABSORPTION SPECTROSCOPY
IN
(Received 23 June 1974. Revised 6 December 1974. Accepted 29 December 1974)
Selenium and its compounds have been used commercially for many years in the production of rectifiers, photocells, pigments, etc. with no long-term effects on industrial workers. However, selenium has long been regarded as a toxic substance, because of the well-known seleniumpoisoning 1 of cattle (“alkali disease” and “blind-staggers”) caused by the consumption of selenium-bearing plants. Elemental selenium is relatively non-toxic toward humans, but some selenium compounds, especially hydrogen selenide, are toxic. Selenium is also nutritionally important and is valuable in the treatment of a number of deficiency diseases. Considerable attention has been devoted to trace analysis for selenium in environmental,’ geological,” and biological“ samples by various techniques. Gravimetric procedures have been employed in the determination of selenium in geological samples, but interfering ions must first be removed. Iodometry, calorimetry and polarography have also been proposed as analytical methods; however, they are dependent upon the chemical state and separation of selenium, and the reagent blanks may be significant. Neutron-activation analysis and spark-source mass spectrometry offer the required accnracy and sensitivity, but spark-source mass spectrometry is a specialized and expensive technique not commonly found in analytical laboratories, while neutron-activation analysis necessitates that an analyst have access to a nuclear reactor or a suitable neutron generator. Therefore, it was the purpose of this investigation to determine submicrogram amounts of selenium with accuracy, reproducibility and sensitivity, by using a conventional and relatively-inexpensive ato&&bsorp%on spectrophotometer with a detection limit of 01 ppm Se; with ihe advent of the tantalum sampling-boat sy&ms the limit may be extended down to 0.01 ppm. In this investigation, some United States Geological Survey standard rocks, GSP-1, W-l and BCR-1 were analyscd for their selenium content and the results compared with those obtained by the slower but generally contaminationfree method of neutron-activation analysis.’
EXPERIMENTAL
Apparatus
A Perkin-Elmer Model 403 atomic-absorption spectrophotometer equipped with a three-slot burner head and
a tantalum sampling-boat system was used. An Intensitron selenium hollow-cathode lamp was operated at a current of 16 mA and the reasonanoe line of 196.1 nm was used with a slit setting of 12 A, and an air-acetylene flame. The @put signal was monitored on a Perkin-Elmer Model 165 lO-mV strip-chart recorder with a chart speed of 66mm/min. The absorption peaks were recorded so that peak heights could be measured after all samples had been run. Procedure
To determine possible losses due to volatilization of selenium, three l-g samples of each rock were dissolved in 20 ml of a 1:l mixture of 40% hydrofluoric acid and 16 M nitric acid, then “Se tracer in the form of selenious acid (New England Nuclear Corp., Boston, Mass.) and of known activity was added to each sample, and the samples were digested and reduced to dryness on a steam-bath. Owing to the relatively long half-life of “Se (120.4 days) the amount lost through decay would be negligible. Then 10 ml of 16M nitric acid were added to the residue and a l-ml aliquot was counted, the observed activity being corrected to correspond to the original conditions. No loss of selenium was noted; this observation substantiates the work of Chau and Riley.6 For AAS analysis, 2-g samples of the three rocks were accurately weighed and digested in Teflon beakers with 40 ml of the 1: 1 acid mixture. The beakers were covered and heated on a steam-bath for 12hr. The covers were removed, the solutions evaporated to dryness, and another 40 ml of the 1: 1 mixture of acids were added to each and again evaporated to dryness. Then 10 ml of 16h4 nitric acid were added to each, and the solutions evaporated to dryness on a steam-bath. This was repeated twice more. Then 25 ml of 4 M hydrochloric acid were added to each residue and the resilting solutions boiled gently until the volume was reduced to 05 ml. Two uCi of “Se (as selenious acid) were added to each solution together with 1Oml of 9 M hydrobromic acid. Each beaker was placed on an NaI(T1) crystal and the total counts were determined by the method of Covel17 from the 401-keV peak. Afterwards each solution was poured into a 40-ml separatory funnel and 11 ml of 1% solution of phenol in benzene were added in accordance with the procedure reported by McGee et aZ.* Each funnel was shaken for 2.5 min, after which the whole of each solution was transferred to a 50-ml tube and allowed to separate for 2.5 min.
548
SHORT COMMUNICATIONS
The aqueous and organic phases were then divided and counted. The degree of extraction of selenium into the benzene could thus be determined. Each benzene phase was evaporated to dryness and the residues were dissolved in water (demineralized and quartz-distilled), and diluted to the mark in lO-ml volumetric flasks. These solutions were then analysed by AAS, an aliquot (Sml) being evaporated in the tantalum boat, which was then inserted into the flame. The primary standard solution for the AAS was prepared by dissolving 16.34g of selenious acid in 1 litre of demineralized quartz-distilled water and standardized by the method outlined by Hillebrand et al.’ The quartzdistilled water used was analysed by neutron-activation analysis and no evidence of selenium was found. A l-ml aliquot of the stock solution was diluted to 1 litre and the resulting solution was used for the preparation of other standards for AAS. RESULTS
AND
Table 2. Trace selenium concentrations USGS rock GSP-1 W-l BCR-1
0.058 f O-001 0.110 f cat5 0.100 f OaOl
0~057ao59 0103+116 0 099 + 0.100
Table 3. The selenium:sulphur rocks
0.059 0.110 a103
ratio of some U.S.G.S.
USGS. rock GSP-1 W-l BCR-1
0058 0110 0.100
005 0.014 0.05
1 x lo-’ 8.2 x lo-’ 2 x lo-’
10e4, and this serves as further evidence that the results obtained are valid. Severne and Brooks have recently employed AAS for the determination of selenium at levels exceeding 1 ppm in geological samples, i2, i3 In our investigation, we have extended the detection of selenium down to 0% ppm with simplicity, accuracy and precision. We suggest that the relatively inexpensive and rapid technique of AAS and an associated boat sampling system be utilized in conjunction with activation analysis to enhance accuracy in studies of this nature.
Acknowledgement~The authors are grateful to the Ansul Corporation, Marinette, Wisconsin, which provided a fellowship for T.J.G. and to the United States Geological Survey, Washington, D.C. for providing samples of standard rocks. Acknowledgement is also made to the donors of the Petroleum Research Fund, administered by the American Chemical Society, for partial support of this research. Presented in part at the Pittsburgh Conference on Analytical Chemistry and Applied Spectroscopy, Cleveland, Ohio, March 5, 1973.
Department of Chemistry Lowell Technological Institute Lowell, Massachusetts, U.S.A. New England Nuclear Corp. Billerica, Massachusetts
Activity in aqueous phase. cps
Actlvlty m orgatuc phase, cps
Extraction, %
A B
71.2 30.
I
202.2 1875
74.0 86.2
A B
150.9 134.5
1055 124.3
41.2 48.0
A B
159 65.6
230.9 216.9
936 76.8
T.
GOLEMBESKI
REFERENCES 1.
Table 1. Degree of extraction of selenium with use of “Se tracer
BCR-1 Sample Sample W-l Sample Sample GSP-1 Sample Sample
Range, ppnr
selelllum, ppm (neutron-activation analysa)’
DISCUSSION
The absorption signal obtained by the sampling-boat technique represents the total mass of selenium in the sample aliquot rather than its concentration in solution. This amount was established by interpolation in a plot of peakheight us. pg of selenium, a fixed volume (1 ml) of two standard selenium solutions, which closely bracketed the selenium content of the solutions obtained from the rocks, being evaporated in the tantalum boat, which was then placed in the flame. Standards were run before and after each rock solution in order to ensure accuracy and to minimize the effect of a slight decrease in the background absorbance of the boat. The chemical yields for the solvent-extraction process appear in Table 1. The variation in degree of extraction is large and is a result of the selenium concentration being low; McGee, Lynch and Boswells reported that the reproducibility of their procedure below a selenium concentration of 0.07 mg/ml(70 ppm) was very poor. As with other elements at tracer levels, selenium is also subject to losses by adsorption on the walls of the container and on colloidal particles. The AAS results were corrected by means of the chemical-yield factor. The results for the rock samples appear in Table 2 together with trace selenium concentrations for the same geological specimens reported by Brunfelt and Steinnes’. The agreement may be considered excellent in view of the fact that Brunfelt and Steinnes used a different procedure (distillation) to separate selenium in the rocks, before the neutron-activation analysis. Inspection of Table 2 also indicates that the precision as measured by standard deviation and range is quite good and this provides further evidence that the chemical-yield correction is adequate. The selenium:sulphur ratio has been determined for various geological materials and found to be about 1 x 10d4 except for sea-water (1 x 10e7) and evaporites (2 x 10-s).i”~ it Therefore, this ratio may be used as a check on the reliability of selenium determinations. Table 3 shows that our results approximate to the ratio of 1 x
U.S G.S. rock
S&mum, ppm (this method)
for U.S.G.S. rocks
E. A. Cerwenka, Jr. and W. C. Cooper, Arch. Enoiron.
Health, 1961, 3, 71. 2. Y. Hashimoto and J. W. Winchester, Enoiron. Sci. Technol., 1967, 1, 338. 3. V. Lavrakas, T. Golembeski, G. Pappas, J. E. Gregory and H. Wedlick, And. Chem. 1974, 46, 952. 4. National Academy of Sciences, Selenium in Nutrition,
Washington, DC., 1971. 5. A. 0. Brunfelt and E. Steinnes, Geochim. Cosmochim. Acta, 1967, 31, 283. 6. Y. K. Chau and J. P. Riley, Anal. Chim. Acta, 1965, 33, 36. 7. D. F. Covell, Anal. Chem., 1959, 31, 1785. 8. T. McGee, J. Lynch and G. G. J. Boswell, Talonta,
1968, 15, 1438.
SHORT
COMMUNiCATlONS
9. W. F. Hi&brand, G. E. Lundell, H. A. Bright and J. L. Hoffman, Applied Inorganic Analysis, 2nd Ed., Wiley, New York, 1953. 10. K. Rankama and G. Sahama, Geochemistry, University of Chicago Press, Chicago, 1950. 11. D. F. Schutz and K. K. Turekian, Geochim. Cosmochim. Acta, 1965, 29, 259.
549
12. B. C. Severne and R. R. Brooks, Anal. Chim. Acta, 1972, 58, 216. 13. Idem, Talanta, 1972, 19, 1467. 14. F. J. Flanagan, Geochim. Cosmochim. Acta, 1969, 33, 81. 15. M. Fleischer and R. E. Stevens, ibid., 1962, 26, 525.
Summary-Atomic-absorption spectroscopy was used to determine trace amounts of selenium accurately in U.S. Geological Survey standard rocks, GSP-1, W-l and BCR-1. The results obtained were compared with those obtained by neutron-activation analysis and excellent agreement was found; in addition, the selenium:sulphur ratio was calculated and agreed with results obtained by other workers.
!ZHORT COMMUNICATIONS
SPnmmry-A method is described for the amperometric titration of nickel and eudcessive amperometric determination of copper and nickel. Nickel (10-16.0 mg) and copper (10-l l-0 mg) could be determined with an average error of less than 1%. Cobalt interferes but chloride does not. Interference by aluminium, iron(II1) and chromium can be eliminated Zinc and manganese do not interfere if the correct applied voltage is chosen. The procedures can be utilized in the analysis of alloys such as nichrome, Raney nickel, constantan, german silver and manganin. It is best to use the standard addition method for less than 3 mg of nickel.
Talanta, Vol. 22, pp 547-549
Pergamon
Press
1975. Printed in Great Britain
DETERMINATION OF SUB MICROGRAM AMOUNTS OF SELENIUM ROCKS BY ATOMIC-ABSORPTION SPECTROSCOPY
IN
(Received 23 June 1974. Revised 6 December 1974. Accepted 29 December 1974)
Selenium and its compounds have been used commercially for many years in the production of rectifiers, photocells, pigments, etc. with no long-term effects on industrial workers. However, selenium has long been regarded as a toxic substance, because of the well-known seleniumpoisoning 1 of cattle (“alkali disease” and “blind-staggers”) caused by the consumption of selenium-bearing plants. Elemental selenium is relatively non-toxic toward humans, but some selenium compounds, especially hydrogen selenide, are toxic. Selenium is also nutritionally important and is valuable in the treatment of a number of deficiency diseases. Considerable attention has been devoted to trace analysis for selenium in environmental,’ geological,” and biological“ samples by various techniques. Gravimetric procedures have been employed in the determination of selenium in geological samples, but interfering ions must first be removed. Iodometry, calorimetry and polarography have also been proposed as analytical methods; however, they are dependent upon the chemical state and separation of selenium, and the reagent blanks may be significant. Neutron-activation analysis and spark-source mass spectrometry offer the required accnracy and sensitivity, but spark-source mass spectrometry is a specialized and expensive technique not commonly found in analytical laboratories, while neutron-activation analysis necessitates that an analyst have access to a nuclear reactor or a suitable neutron generator. Therefore, it was the purpose of this investigation to determine submicrogram amounts of selenium with accuracy, reproducibility and sensitivity, by using a conventional and relatively-inexpensive ato&&bsorp%on spectrophotometer with a detection limit of 01 ppm Se; with ihe advent of the tantalum sampling-boat sy&ms the limit may be extended down to 0.01 ppm. In this investigation, some United States Geological Survey standard rocks, GSP-1, W-l and BCR-1 were analyscd for their selenium content and the results compared with those obtained by the slower but generally contaminationfree method of neutron-activation analysis.’
EXPERIMENTAL
Apparatus
A Perkin-Elmer Model 403 atomic-absorption spectrophotometer equipped with a three-slot burner head and
a tantalum sampling-boat system was used. An Intensitron selenium hollow-cathode lamp was operated at a current of 16 mA and the reasonanoe line of 196.1 nm was used with a slit setting of 12 A, and an air-acetylene flame. The @put signal was monitored on a Perkin-Elmer Model 165 lO-mV strip-chart recorder with a chart speed of 66mm/min. The absorption peaks were recorded so that peak heights could be measured after all samples had been run. Procedure
To determine possible losses due to volatilization of selenium, three l-g samples of each rock were dissolved in 20 ml of a 1:l mixture of 40% hydrofluoric acid and 16 M nitric acid, then “Se tracer in the form of selenious acid (New England Nuclear Corp., Boston, Mass.) and of known activity was added to each sample, and the samples were digested and reduced to dryness on a steam-bath. Owing to the relatively long half-life of “Se (120.4 days) the amount lost through decay would be negligible. Then 10 ml of 16M nitric acid were added to the residue and a l-ml aliquot was counted, the observed activity being corrected to correspond to the original conditions. No loss of selenium was noted; this observation substantiates the work of Chau and Riley.6 For AAS analysis, 2-g samples of the three rocks were accurately weighed and digested in Teflon beakers with 40 ml of the 1: 1 acid mixture. The beakers were covered and heated on a steam-bath for 12hr. The covers were removed, the solutions evaporated to dryness, and another 40 ml of the 1: 1 mixture of acids were added to each and again evaporated to dryness. Then 10 ml of 16h4 nitric acid were added to each, and the solutions evaporated to dryness on a steam-bath. This was repeated twice more. Then 25 ml of 4 M hydrochloric acid were added to each residue and the resilting solutions boiled gently until the volume was reduced to 05 ml. Two uCi of “Se (as selenious acid) were added to each solution together with 1Oml of 9 M hydrobromic acid. Each beaker was placed on an NaI(T1) crystal and the total counts were determined by the method of Covel17 from the 401-keV peak. Afterwards each solution was poured into a 40-ml separatory funnel and 11 ml of 1% solution of phenol in benzene were added in accordance with the procedure reported by McGee et aZ.* Each funnel was shaken for 2.5 min, after which the whole of each solution was transferred to a 50-ml tube and allowed to separate for 2.5 min.
548
SHORT COMMUNICATIONS
The aqueous and organic phases were then divided and counted. The degree of extraction of selenium into the benzene could thus be determined. Each benzene phase was evaporated to dryness and the residues were dissolved in water (demineralized and quartz-distilled), and diluted to the mark in lO-ml volumetric flasks. These solutions were then analysed by AAS, an aliquot (Sml) being evaporated in the tantalum boat, which was then inserted into the flame. The primary standard solution for the AAS was prepared by dissolving 16.34g of selenious acid in 1 litre of demineralized quartz-distilled water and standardized by the method outlined by Hillebrand et al.’ The quartzdistilled water used was analysed by neutron-activation analysis and no evidence of selenium was found. A l-ml aliquot of the stock solution was diluted to 1 litre and the resulting solution was used for the preparation of other standards for AAS. RESULTS
AND
Table 2. Trace selenium concentrations USGS rock GSP-1 W-l BCR-1
0.058 f O-001 0.110 f cat5 0.100 f OaOl
0~057ao59 0103+116 0 099 + 0.100
Table 3. The selenium:sulphur rocks
0.059 0.110 a103
ratio of some U.S.G.S.
USGS. rock GSP-1 W-l BCR-1
0058 0110 0.100
005 0.014 0.05
1 x lo-’ 8.2 x lo-’ 2 x lo-’
10e4, and this serves as further evidence that the results obtained are valid. Severne and Brooks have recently employed AAS for the determination of selenium at levels exceeding 1 ppm in geological samples, i2, i3 In our investigation, we have extended the detection of selenium down to 0% ppm with simplicity, accuracy and precision. We suggest that the relatively inexpensive and rapid technique of AAS and an associated boat sampling system be utilized in conjunction with activation analysis to enhance accuracy in studies of this nature.
Acknowledgement~The authors are grateful to the Ansul Corporation, Marinette, Wisconsin, which provided a fellowship for T.J.G. and to the United States Geological Survey, Washington, D.C. for providing samples of standard rocks. Acknowledgement is also made to the donors of the Petroleum Research Fund, administered by the American Chemical Society, for partial support of this research. Presented in part at the Pittsburgh Conference on Analytical Chemistry and Applied Spectroscopy, Cleveland, Ohio, March 5, 1973.
Department of Chemistry Lowell Technological Institute Lowell, Massachusetts, U.S.A. New England Nuclear Corp. Billerica, Massachusetts
Activity in aqueous phase. cps
Actlvlty m orgatuc phase, cps
Extraction, %
A B
71.2 30.
I
202.2 1875
74.0 86.2
A B
150.9 134.5
1055 124.3
41.2 48.0
A B
159 65.6
230.9 216.9
936 76.8
T.
GOLEMBESKI
REFERENCES 1.
Table 1. Degree of extraction of selenium with use of “Se tracer
BCR-1 Sample Sample W-l Sample Sample GSP-1 Sample Sample
Range, ppnr
selelllum, ppm (neutron-activation analysa)’
DISCUSSION
The absorption signal obtained by the sampling-boat technique represents the total mass of selenium in the sample aliquot rather than its concentration in solution. This amount was established by interpolation in a plot of peakheight us. pg of selenium, a fixed volume (1 ml) of two standard selenium solutions, which closely bracketed the selenium content of the solutions obtained from the rocks, being evaporated in the tantalum boat, which was then placed in the flame. Standards were run before and after each rock solution in order to ensure accuracy and to minimize the effect of a slight decrease in the background absorbance of the boat. The chemical yields for the solvent-extraction process appear in Table 1. The variation in degree of extraction is large and is a result of the selenium concentration being low; McGee, Lynch and Boswells reported that the reproducibility of their procedure below a selenium concentration of 0.07 mg/ml(70 ppm) was very poor. As with other elements at tracer levels, selenium is also subject to losses by adsorption on the walls of the container and on colloidal particles. The AAS results were corrected by means of the chemical-yield factor. The results for the rock samples appear in Table 2 together with trace selenium concentrations for the same geological specimens reported by Brunfelt and Steinnes’. The agreement may be considered excellent in view of the fact that Brunfelt and Steinnes used a different procedure (distillation) to separate selenium in the rocks, before the neutron-activation analysis. Inspection of Table 2 also indicates that the precision as measured by standard deviation and range is quite good and this provides further evidence that the chemical-yield correction is adequate. The selenium:sulphur ratio has been determined for various geological materials and found to be about 1 x 10d4 except for sea-water (1 x 10e7) and evaporites (2 x 10-s).i”~ it Therefore, this ratio may be used as a check on the reliability of selenium determinations. Table 3 shows that our results approximate to the ratio of 1 x
U.S G.S. rock
S&mum, ppm (this method)
for U.S.G.S. rocks
E. A. Cerwenka, Jr. and W. C. Cooper, Arch. Enoiron.
Health, 1961, 3, 71. 2. Y. Hashimoto and J. W. Winchester, Enoiron. Sci. Technol., 1967, 1, 338. 3. V. Lavrakas, T. Golembeski, G. Pappas, J. E. Gregory and H. Wedlick, And. Chem. 1974, 46, 952. 4. National Academy of Sciences, Selenium in Nutrition,
Washington, DC., 1971. 5. A. 0. Brunfelt and E. Steinnes, Geochim. Cosmochim. Acta, 1967, 31, 283. 6. Y. K. Chau and J. P. Riley, Anal. Chim. Acta, 1965, 33, 36. 7. D. F. Covell, Anal. Chem., 1959, 31, 1785. 8. T. McGee, J. Lynch and G. G. J. Boswell, Talonta,
1968, 15, 1438.
SHORT
COMMUNiCATlONS
9. W. F. Hi&brand, G. E. Lundell, H. A. Bright and J. L. Hoffman, Applied Inorganic Analysis, 2nd Ed., Wiley, New York, 1953. 10. K. Rankama and G. Sahama, Geochemistry, University of Chicago Press, Chicago, 1950. 11. D. F. Schutz and K. K. Turekian, Geochim. Cosmochim. Acta, 1965, 29, 259.
549
12. B. C. Severne and R. R. Brooks, Anal. Chim. Acta, 1972, 58, 216. 13. Idem, Talanta, 1972, 19, 1467. 14. F. J. Flanagan, Geochim. Cosmochim. Acta, 1969, 33, 81. 15. M. Fleischer and R. E. Stevens, ibid., 1962, 26, 525.
Summary-Atomic-absorption spectroscopy was used to determine trace amounts of selenium accurately in U.S. Geological Survey standard rocks, GSP-1, W-l and BCR-1. The results obtained were compared with those obtained by neutron-activation analysis and excellent agreement was found; in addition, the selenium:sulphur ratio was calculated and agreed with results obtained by other workers.