- Email: [email protected]
Elution and spectrophotometric determination of gold after its separation from non-volatile platinum metals by column extraction chromatography
Talanta, Vol. 29, PP. 946 to 948, 1982 Printed in Great Britain. All rights reserved
0039-9140/82/110946-03$03.00/O
Copyright 0 1982 Pergamon Press Ltd
ELUTION AND SPECTROPHOTOMETRIC DETERMINATION OF GOLD AFTER ITS SEPARATION FROM NON-VOLATILE PLATINUM METALS BY COLUMN EXTRACTION CHROMATOGRAPHY A. FLIEGER and S. PRZESZLAKOWSKI Department
of Inorganic
and Analytical
Chemistry,
Medical
School,
2&081
Lublin,
Poland
(Received 8 January 1982. Accepted 10 June 1982) Summary-Mixtures of gold(II1) and iridium(IV) were separated by column extraction chromatography on silica treated with a tri-n-octylamine (TOA) salt. A mixture 2.25M in hydrochloric acid and 5M in nitric acid was used for elution of iridium. Gold was eluted together with the TOA salt by acetone, and after evaporation of the acetone, the TOA chloroaurate was dissolved in chloroform, converted into TOA bromoaurate and determined spectrophotometrically at 395 nm (6 = 3.4 x 1031.mole~1.cm~‘). Beer’s law was obeyed over the concentration range 5-67 ppm of gold. The method was found suitable for determination of gold after its separation from other metals by extraction chromatography on supports treated with liquid anion-exchangers.
The determination of gold after its separation from other metals by chromatography on solid anionexchange resins requires combustion of the resin, owing to irreversible retention of this metal.1~2 A simple and rapid method of analysis for non-volatile platinum metals, based on their column chromatographic separation on silica treated with a tri-noctylamine (TOA) salt and elution with mixtures of hydrochloric and nitric acid, has been described,3 but it was found that gold(II1) was very strongly retained and could not be eluted even by a mixture of concentrated hydrochloric and nitric acids. Preliminary experiments in our laboratory indicated that although the gold could be eluted from the column together with the TOA salt and remaining aqueous solution with acetone, and subsequently determined spectrophotometrically with Rodamine B,4 this procedure was inconvenient since the presence of TOA interfered, and mineralization of the amine salt with a mixture of concentrated nitric acid, sulphuric acid and hydrogen peroxide required refluxing at a temperature of 22CL230” for 4 hr.’ Many spectrophotometric methods for gold are based on ion-pair formation between halide complexes of gold and basic dyes.6,7 A selective and precise extraction-spectrophotometric method utilizing the bromoaurate complex with tri-n-octylphosphine oxide was proposed by Holbrook and Rein.’ Since gold had been reported to form coloured spots on paper treated with tri-n-octylamine hydrobromide,9~‘0 it was thought that formation of an ionpair composed of bromoaurate and tri-n-octylammonium ions could be utilized for extraction-spectrophotometric determination of gold after elution of chloroaurate from TOA-coated silica with acetone.
EXPERIMENTAL Reagents Pure tri-n-octylamine (Fluka) was further purified by vacuum distillation, the fraction boiling at 19&200”/5-10 mmHg being collected. Reagent grade &loroform was also distilled before use. Standard gold solution was prepared by dissolving 46.5 mg of HAuC14.4Hz0 (pure, POCh, Poland) in 1M hydrochloric acid. Iridium(IV) solution was prepared in the manner described previously.3 All other reagents were of analytical grade. Procedure Glass tubes (10 x 300 mm) were used. packed with purified silica gel coated with TOA salt. The sorbent (10 g) was slurried with 50ml of the first eluent (2.25M HCI + 5M HNO,). poured into the column and covered with a cellulose filter paper. The column length was 190 mm, the volume of organic phase 1.4 ml and the hold-up volume 8 ml. The volume of organic phase was determined by eluting the TOA salt together with the remaining aqueous phase from the column with 35 ml of acetone; the solvent was then evaporated and the volume of TOA was measured in a narrow calibrated test-tube. After the column had been washed with 8 ml of the first eluent, 2 or 3 ml of a synthetic mixture containing 176 pg of iridium and 890 or 445 pg of gold was introduced into it. Elution under hydrostatic pressure (head 250 mm) was used for the separation. Iridium was eluted with 40 ml of the first eluent (and was found in a 15-ml fraction of the eluate). The column was then washed with 12 ml of 3M hydrochloric acid and gold was eluted together with the organic stationary phase and aqueous solution with 35 ml of acetone. The acetone was then evaporated on a waterbath and the bright yellow residue (TOA chloroaurate and excess of TOA salt) with remaining aqueous solution was treated with 40 ml of chloroform and transferred into a separatory funnel, which was then thoroughly shaken. When the phases had separated. the organic (lower) phase was run into a 50-ml standard flask and the aqueous phase 946
SHORT
947
COMMUNICATIONS
was washed with 5 ml of chloroform, the washings also being added to the flask. The combined extracts were diluted to the mark with chloroform and mixed. Then 5 ml of the chloroform phase were pipetted into a small separatory funnel and 5 ml of 1.8M aqueous hydrobromic acid (containing 4 x 10-40~ bromine) and 10 drops of concentrated phosphoric acid were added. The mixture was then shaken for 1 min. and after separation of the phases, the lower (organic) phase was dried by addition of a small amount of anhydrous sodium sulphate. The absorbance of the organic phase was then measured in a l-cm cell at 395 nm against a reagent blank similarly prepared (the concentration of TOA salt in the reagent blank was 0.06M). A calibration graph was prepared with standard TOA bromoaurate solutions made as follows. A 0.12M solution of TOA in chloroform was shaken with an equal volume of the first eluent (2.25M HCI + 5M HNO,), and after separation of the phases, the lower phase was filtered through a cellulose filter. More dilute solutions of the TOA salt were obtained by dilution of this 0.12M TOA salt solution with chloroform. A 5-ml volume of this organic solution was then introduced into a small separatory funnel and 5 ml of 1.8M hydrobromic acid (containing 4 x 10m4% bromine) 10 drops of concentrated phosphoric acid and 0.1-l ml of standard gold solution were added. The mixture was shaken for 1 min and then treated as described for samples. The procedure for column preparation and chromatographic separation has already been described.3 RESULTS
AND DISCUSSION
Since gold(M) can be very easily separated from other noble metals by column extraction chromatography, owing to the very strong retention of gold in the system TOA salt-(HCl + HNO,), a simple method for elution of gold and its determination was sought. Attention was paid to the spectrophotometric method based on the bromoaurate complex with trin-octylphosphine oxide, which was found to be very selective for gold, only iridium(W) and tin(I1) interfering (at low concentrations).’ Since it was found that gold and the tri-n-octylamine salt can be easily eluted from the column with acetone, and bromoaurate should form a coloured ion-pair with the tri-n-octylammonium cation, we decided to modify the method proposed by Holbrook and Rein* for the determination of gold in the eluate from the column. It was also supposed that chloroaurate tri-n-octylammonium ion-association complex should be convertible into the bromoaurate complex by treatment with a sufficient excess of hydrobromic acid owing to the
II
I 360
I
I 400
I
I 440
Wavelength,
nm
Fig. 1. Spectrum of bromoaurateeTOA form, measured us. reagent blank. 0.06M; gold concentration
I
I
I 520
complex in chloroTOA concentration 22.5 ppm.
higher stability of the bromide complexes of gold” and the higher affinity of bromide for alkylammonium cations. “3 3 Preliminary experiments indicated that gold(II1) is not extracted by chloroform from hydrobromic acid media, but the organic phase becomes yellow after addition of a small amount of tri-n-octylamine. In further experiments, a small amount of bromine was added to the hydrobromic acid solutions to prevent the reduction of gold, and phosphoric acid was added to the aqueous solution (before the extraction) as masking agent for iron(II1) if traces of this metal were present in the aqueous solution. The absorption spectrum of the TOA-bromoaurate complex in chloroform (Fig. 1) is analogous to the spectrum of the TOPO-bromoaurate complex reported by Holbrook and Rein,’ but the molar absorptivity is a bit lower (3.4 x lo3 l.mole- ’ .cm- i at 395 nm). The absorbance of the organic phase is stable for 20 min and then slowly decreases. The absorbance depends somewhat on the TOA salt concentration in the range 0.03-0.12M (see Table 1); since 40 ml of chloroform were used for the dissolution of the TOA-chloroaurate complex and the excess of TOA salt after their elution from the column, and the concentration of TOA salt was then 0.06&f, a tri-n-octylamine hydrobromide solution of this concentration in chloroform was used as the blank for gold determination. It should be remarked that the absorbance of the organic phase was the
Table 1. Influence of tri-n-octylamine salt concentration in chloroform on absorbance of the bromide complex of gold (18 ppm); the organic phase was shaken with 1.8M hydrobromic acid containing 4 x 10e4% bromine Absorbance
TOA blank us. CHCl, Sample us. CHCI, Sample us. TOA blank
I 480
at 395 nm
0.03M TOA
0.06M TOA
0.09M TOA
0.12M TOA
0.100
0.108
0.120
0.136
0.390
0.390
0.400
0.420
0.285
0.280
0.284
0.287
948
SHOR’I
COMMUNICATIONS
Table 2. The determination of gold after the separation of synthetic mixtures of iridium (176 extraction column gold by and pg) chromatography Au taken, /@3
Au found and standard deviation, p9
Error, %
890 445
880 + 12 432 k 5
- 1.1 -2.9
* Mean value from 6 successive determinations. same whether the gold was extracted from hydrobromic acid medium or the TOA-chloroaurate complex in chloroform was shaken with 1.8M hydrobromic acid. Beer’s law is obeyed over the gold concentration range 5-67 ppm. The results of gold determination after its separation from iridium(W) by column extraction chromatography on silica treated with TOA salt and eluted with mixture of hydrochloric acid and nitric acid (2.294 HCI + 5M HN03) are given in Table 2. In our opinion, the proposed method can be utilized not only for the determination of non-volatile noble metals after their separation by column extraction chromatography, but also for determination of gold in other materials containing larger amounts of this metal (owing to the relatively low sensitivity of the method) in the absence of metals which form coloured anionic bromide complexes extractable by TOA. Furthermore, extraction chromatography with the system TOA HCl-HN03 is more convenient than use of anion-exchange resins for the separation and determination of non-volatile noble metals, owing to the easier elution of platinum(W) and iridium(IV) and the simple and rapid method for determination of gold (which is very strongly retained in both anionexchange chromatographic systems). It should be noted that gold(II1) and the TOA salt can probably be eluted from the column with methyl isobutyl ketone, like gold bound with tributyl phos-
phate,14 and subsequently determined in the organic solution by AAS, in a modification of the method described by Groenewald.’ 5 The sensitivity of the method described here could be increased by using a smaller volume of chloroform (e.g., 10 ml) for the extraction, but a blank made in the same way would have to be used as reference solution because of the effect of the higher TOA hydrobromide concentration. Acknowledgements-This work was supported by the Institute of Chemistry of the Maria Curie-Sklodowska University in Lublin (Grant No. MR.L14.40-81). Thanks are due to Professor Edward Soczewidski for his interest in this work. REFERENCES F. E. Beamish, The Analytical Chemistry of the Noble Metals, Pergamon Press, Oxford, 1966. 2. F. E. Beamish and J. C. Van Loon, Recent Advances in the Analytical Chemistry of the Noble Metals, Perga1.
mon Press, Oxford, 1972. 3. S. Przeszlakowski and A. Flieger, Talanta, 1981, 28, 557.
4. B. J. MacNulty and L. D. Woodward,
Anal. Chim.
Acta, 1955, 13, 154.
5. A. Flieger, unpublished results, 1980. 6. Z. Marczenko, Spectrophotometric Determination of Elements, Horwood, Chichester, 1976. 7. H. Sikorska-Tomicka, Chem. Analit. (Warsaw), 198 I, 26, 257.
8. W. B. Holbrook and J. E. Rein, Anal. Chem., 1964, 36, 2451.
9. S. Przeszlakowski,
Chem. Analit. (Warsaw), 1966, 11,
895.
10. Idem, ibid., 1967, 12, 1071. 11. L. G. Sill& and A. E. Martell. Stability Constants of Metal-Ion Complexes, Chem. Sot., London, 1964. 12. A. S. Kertes, Y. Marcus and E. Yanir, Equilibrium Constants of Liquid-Liquid Distribution Reactions, Part 2, Alkylammonium Salt Extractants, Butterworths, Lon-
don, 1974. 13. S. Przeszlakowski and R. Kocjan, Chromatographia, 1977, 10, 358. 14. C. Pohlandt and T. W. Steele, Talanta, 1972, 19, 839. 15. T. Groenewald, Anal. Chem., 1969, 41, 1012.
0039-9140/82/110946-03$03.00/O
Copyright 0 1982 Pergamon Press Ltd
ELUTION AND SPECTROPHOTOMETRIC DETERMINATION OF GOLD AFTER ITS SEPARATION FROM NON-VOLATILE PLATINUM METALS BY COLUMN EXTRACTION CHROMATOGRAPHY A. FLIEGER and S. PRZESZLAKOWSKI Department
of Inorganic
and Analytical
Chemistry,
Medical
School,
2&081
Lublin,
Poland
(Received 8 January 1982. Accepted 10 June 1982) Summary-Mixtures of gold(II1) and iridium(IV) were separated by column extraction chromatography on silica treated with a tri-n-octylamine (TOA) salt. A mixture 2.25M in hydrochloric acid and 5M in nitric acid was used for elution of iridium. Gold was eluted together with the TOA salt by acetone, and after evaporation of the acetone, the TOA chloroaurate was dissolved in chloroform, converted into TOA bromoaurate and determined spectrophotometrically at 395 nm (6 = 3.4 x 1031.mole~1.cm~‘). Beer’s law was obeyed over the concentration range 5-67 ppm of gold. The method was found suitable for determination of gold after its separation from other metals by extraction chromatography on supports treated with liquid anion-exchangers.
The determination of gold after its separation from other metals by chromatography on solid anionexchange resins requires combustion of the resin, owing to irreversible retention of this metal.1~2 A simple and rapid method of analysis for non-volatile platinum metals, based on their column chromatographic separation on silica treated with a tri-noctylamine (TOA) salt and elution with mixtures of hydrochloric and nitric acid, has been described,3 but it was found that gold(II1) was very strongly retained and could not be eluted even by a mixture of concentrated hydrochloric and nitric acids. Preliminary experiments in our laboratory indicated that although the gold could be eluted from the column together with the TOA salt and remaining aqueous solution with acetone, and subsequently determined spectrophotometrically with Rodamine B,4 this procedure was inconvenient since the presence of TOA interfered, and mineralization of the amine salt with a mixture of concentrated nitric acid, sulphuric acid and hydrogen peroxide required refluxing at a temperature of 22CL230” for 4 hr.’ Many spectrophotometric methods for gold are based on ion-pair formation between halide complexes of gold and basic dyes.6,7 A selective and precise extraction-spectrophotometric method utilizing the bromoaurate complex with tri-n-octylphosphine oxide was proposed by Holbrook and Rein.’ Since gold had been reported to form coloured spots on paper treated with tri-n-octylamine hydrobromide,9~‘0 it was thought that formation of an ionpair composed of bromoaurate and tri-n-octylammonium ions could be utilized for extraction-spectrophotometric determination of gold after elution of chloroaurate from TOA-coated silica with acetone.
EXPERIMENTAL Reagents Pure tri-n-octylamine (Fluka) was further purified by vacuum distillation, the fraction boiling at 19&200”/5-10 mmHg being collected. Reagent grade &loroform was also distilled before use. Standard gold solution was prepared by dissolving 46.5 mg of HAuC14.4Hz0 (pure, POCh, Poland) in 1M hydrochloric acid. Iridium(IV) solution was prepared in the manner described previously.3 All other reagents were of analytical grade. Procedure Glass tubes (10 x 300 mm) were used. packed with purified silica gel coated with TOA salt. The sorbent (10 g) was slurried with 50ml of the first eluent (2.25M HCI + 5M HNO,). poured into the column and covered with a cellulose filter paper. The column length was 190 mm, the volume of organic phase 1.4 ml and the hold-up volume 8 ml. The volume of organic phase was determined by eluting the TOA salt together with the remaining aqueous phase from the column with 35 ml of acetone; the solvent was then evaporated and the volume of TOA was measured in a narrow calibrated test-tube. After the column had been washed with 8 ml of the first eluent, 2 or 3 ml of a synthetic mixture containing 176 pg of iridium and 890 or 445 pg of gold was introduced into it. Elution under hydrostatic pressure (head 250 mm) was used for the separation. Iridium was eluted with 40 ml of the first eluent (and was found in a 15-ml fraction of the eluate). The column was then washed with 12 ml of 3M hydrochloric acid and gold was eluted together with the organic stationary phase and aqueous solution with 35 ml of acetone. The acetone was then evaporated on a waterbath and the bright yellow residue (TOA chloroaurate and excess of TOA salt) with remaining aqueous solution was treated with 40 ml of chloroform and transferred into a separatory funnel, which was then thoroughly shaken. When the phases had separated. the organic (lower) phase was run into a 50-ml standard flask and the aqueous phase 946
SHORT
947
COMMUNICATIONS
was washed with 5 ml of chloroform, the washings also being added to the flask. The combined extracts were diluted to the mark with chloroform and mixed. Then 5 ml of the chloroform phase were pipetted into a small separatory funnel and 5 ml of 1.8M aqueous hydrobromic acid (containing 4 x 10-40~ bromine) and 10 drops of concentrated phosphoric acid were added. The mixture was then shaken for 1 min. and after separation of the phases, the lower (organic) phase was dried by addition of a small amount of anhydrous sodium sulphate. The absorbance of the organic phase was then measured in a l-cm cell at 395 nm against a reagent blank similarly prepared (the concentration of TOA salt in the reagent blank was 0.06M). A calibration graph was prepared with standard TOA bromoaurate solutions made as follows. A 0.12M solution of TOA in chloroform was shaken with an equal volume of the first eluent (2.25M HCI + 5M HNO,), and after separation of the phases, the lower phase was filtered through a cellulose filter. More dilute solutions of the TOA salt were obtained by dilution of this 0.12M TOA salt solution with chloroform. A 5-ml volume of this organic solution was then introduced into a small separatory funnel and 5 ml of 1.8M hydrobromic acid (containing 4 x 10m4% bromine) 10 drops of concentrated phosphoric acid and 0.1-l ml of standard gold solution were added. The mixture was shaken for 1 min and then treated as described for samples. The procedure for column preparation and chromatographic separation has already been described.3 RESULTS
AND DISCUSSION
Since gold(M) can be very easily separated from other noble metals by column extraction chromatography, owing to the very strong retention of gold in the system TOA salt-(HCl + HNO,), a simple method for elution of gold and its determination was sought. Attention was paid to the spectrophotometric method based on the bromoaurate complex with trin-octylphosphine oxide, which was found to be very selective for gold, only iridium(W) and tin(I1) interfering (at low concentrations).’ Since it was found that gold and the tri-n-octylamine salt can be easily eluted from the column with acetone, and bromoaurate should form a coloured ion-pair with the tri-n-octylammonium cation, we decided to modify the method proposed by Holbrook and Rein* for the determination of gold in the eluate from the column. It was also supposed that chloroaurate tri-n-octylammonium ion-association complex should be convertible into the bromoaurate complex by treatment with a sufficient excess of hydrobromic acid owing to the
II
I 360
I
I 400
I
I 440
Wavelength,
nm
Fig. 1. Spectrum of bromoaurateeTOA form, measured us. reagent blank. 0.06M; gold concentration
I
I
I 520
complex in chloroTOA concentration 22.5 ppm.
higher stability of the bromide complexes of gold” and the higher affinity of bromide for alkylammonium cations. “3 3 Preliminary experiments indicated that gold(II1) is not extracted by chloroform from hydrobromic acid media, but the organic phase becomes yellow after addition of a small amount of tri-n-octylamine. In further experiments, a small amount of bromine was added to the hydrobromic acid solutions to prevent the reduction of gold, and phosphoric acid was added to the aqueous solution (before the extraction) as masking agent for iron(II1) if traces of this metal were present in the aqueous solution. The absorption spectrum of the TOA-bromoaurate complex in chloroform (Fig. 1) is analogous to the spectrum of the TOPO-bromoaurate complex reported by Holbrook and Rein,’ but the molar absorptivity is a bit lower (3.4 x lo3 l.mole- ’ .cm- i at 395 nm). The absorbance of the organic phase is stable for 20 min and then slowly decreases. The absorbance depends somewhat on the TOA salt concentration in the range 0.03-0.12M (see Table 1); since 40 ml of chloroform were used for the dissolution of the TOA-chloroaurate complex and the excess of TOA salt after their elution from the column, and the concentration of TOA salt was then 0.06&f, a tri-n-octylamine hydrobromide solution of this concentration in chloroform was used as the blank for gold determination. It should be remarked that the absorbance of the organic phase was the
Table 1. Influence of tri-n-octylamine salt concentration in chloroform on absorbance of the bromide complex of gold (18 ppm); the organic phase was shaken with 1.8M hydrobromic acid containing 4 x 10e4% bromine Absorbance
TOA blank us. CHCl, Sample us. CHCI, Sample us. TOA blank
I 480
at 395 nm
0.03M TOA
0.06M TOA
0.09M TOA
0.12M TOA
0.100
0.108
0.120
0.136
0.390
0.390
0.400
0.420
0.285
0.280
0.284
0.287
948
SHOR’I
COMMUNICATIONS
Table 2. The determination of gold after the separation of synthetic mixtures of iridium (176 extraction column gold by and pg) chromatography Au taken, /@3
Au found and standard deviation, p9
Error, %
890 445
880 + 12 432 k 5
- 1.1 -2.9
* Mean value from 6 successive determinations. same whether the gold was extracted from hydrobromic acid medium or the TOA-chloroaurate complex in chloroform was shaken with 1.8M hydrobromic acid. Beer’s law is obeyed over the gold concentration range 5-67 ppm. The results of gold determination after its separation from iridium(W) by column extraction chromatography on silica treated with TOA salt and eluted with mixture of hydrochloric acid and nitric acid (2.294 HCI + 5M HN03) are given in Table 2. In our opinion, the proposed method can be utilized not only for the determination of non-volatile noble metals after their separation by column extraction chromatography, but also for determination of gold in other materials containing larger amounts of this metal (owing to the relatively low sensitivity of the method) in the absence of metals which form coloured anionic bromide complexes extractable by TOA. Furthermore, extraction chromatography with the system TOA HCl-HN03 is more convenient than use of anion-exchange resins for the separation and determination of non-volatile noble metals, owing to the easier elution of platinum(W) and iridium(IV) and the simple and rapid method for determination of gold (which is very strongly retained in both anionexchange chromatographic systems). It should be noted that gold(II1) and the TOA salt can probably be eluted from the column with methyl isobutyl ketone, like gold bound with tributyl phos-
phate,14 and subsequently determined in the organic solution by AAS, in a modification of the method described by Groenewald.’ 5 The sensitivity of the method described here could be increased by using a smaller volume of chloroform (e.g., 10 ml) for the extraction, but a blank made in the same way would have to be used as reference solution because of the effect of the higher TOA hydrobromide concentration. Acknowledgements-This work was supported by the Institute of Chemistry of the Maria Curie-Sklodowska University in Lublin (Grant No. MR.L14.40-81). Thanks are due to Professor Edward Soczewidski for his interest in this work. REFERENCES F. E. Beamish, The Analytical Chemistry of the Noble Metals, Pergamon Press, Oxford, 1966. 2. F. E. Beamish and J. C. Van Loon, Recent Advances in the Analytical Chemistry of the Noble Metals, Perga1.
mon Press, Oxford, 1972. 3. S. Przeszlakowski and A. Flieger, Talanta, 1981, 28, 557.
4. B. J. MacNulty and L. D. Woodward,
Anal. Chim.
Acta, 1955, 13, 154.
5. A. Flieger, unpublished results, 1980. 6. Z. Marczenko, Spectrophotometric Determination of Elements, Horwood, Chichester, 1976. 7. H. Sikorska-Tomicka, Chem. Analit. (Warsaw), 198 I, 26, 257.
8. W. B. Holbrook and J. E. Rein, Anal. Chem., 1964, 36, 2451.
9. S. Przeszlakowski,
Chem. Analit. (Warsaw), 1966, 11,
895.
10. Idem, ibid., 1967, 12, 1071. 11. L. G. Sill& and A. E. Martell. Stability Constants of Metal-Ion Complexes, Chem. Sot., London, 1964. 12. A. S. Kertes, Y. Marcus and E. Yanir, Equilibrium Constants of Liquid-Liquid Distribution Reactions, Part 2, Alkylammonium Salt Extractants, Butterworths, Lon-
don, 1974. 13. S. Przeszlakowski and R. Kocjan, Chromatographia, 1977, 10, 358. 14. C. Pohlandt and T. W. Steele, Talanta, 1972, 19, 839. 15. T. Groenewald, Anal. Chem., 1969, 41, 1012.