ANALYTICAL SCIENCES AUGUST 1988, VOL. 4 417
Preconcentration of Gallium(III) and Indium(III) with Hafnium Hydroxide for Electrothermal
Atomic Absorption Spectrometry
by Coprecipitation
Joichi USDA and Chika Mizui
Faculty of Education, Kanazawa University, Marunouchi, Kanazawa 920
The trace amounts of gallium and indium in water were determined by graphite furnace atomic absorption spectrometry after being concentrated by the coprecipitation technique with hafnium hydroxide. Hafnium
hydroxide coprecipitates quantitatively 0.2-3 µg of gallium from 50-400 ml of sample solution at pH 6.0-10.5 and 0.2-4 µg of indium from 50-400 ml at pH 6.0-11.0. The atomic absorbances of gallium and indium are increased about 3 times by the presence of hafnium; the reproducibilities of the measurements are also improved. The
calibration curve is linear from 8 to 120 ng ml-' of gallium or from 8 to 160 ng ml-' of indium. Twenty seven diverse ions tested did not interfere seriously with either element. These methods are applicable to the determination of gallium and indium in river or seawater which contains down to 0.5 ng ml-' of these elements.
Keywords Gallium determination, indium coprecipitation, hafnium hydroxide
determination, electrothermal atomic absorption spectrometry,
Electrothermal atomic absorption spectrometry us- ing a graphite tube is highly sensitive and has been frequently utilized for trace metal analysis, though it can be prone to matrix or interference effects. For the determination of gallium or indium, this technique has been applied; the combined use of matrix modi- fiers'-'"s, a platform3,4,7-9, and a graphite tube impreg- nated with metal10" were recommended to remove those interference effects and to improve the sensitivity.
The coprecipitation method is one of the most useful ways for the concentration of trace elements or for the separation of analytes from sample matrix, and the variety of coprecipitants have been studied.12 We have been examining hafnium hydroxide as a coprecipitant and have suggested that it is useful for the preconcen- tration prior to the determination of trace cadmium13
and copper14 in water samples by the electrothermal atomic absorption spectrometry. This is because hafnium hydroxide is an excellent collector for these ions and coprecipitates few or no matrix ions such as alkali and alkaline earth elements. We tried to apply the hafnium hydroxide coprecipitation technique to the concentration of trace gallium and indium in water samples for subsequent electrothermal atomic absorp- tion spectrometric determination. It was found that hafnium hydroxide collects down to 0.5 ng ml-' of gallium or indium quantitatively, and that the presence of hafnium enhances the atomic absorbance of gallium or indium about three times and improves the reproducibilities of the measurements. The methods
proposed here are simple and reproducible. For the collection of gallium and indium, several kinds of coprecipitants15-22 have already been proposed. Lan- thanum hydroxide's has been used prior to determina- tion of indium by electrothermal atomic absorption spectrometry. However, the use of lanthanum hydroxide is disadvantageous because it does not give a linear calibration curve for indium.
This paper describes the fundamental conditions for the coprecipitation of gallium or indium with hafnium hydroxide and for the graphite furnace atomic absorption spectrometric determination of these ele- ments.
Experimental Apparatus
A Hitachi 170-70 Zeeman effect atomic absorption spectrometer with Hitachi hollow cathode lamp of gallium or indium was used for the atomic absorption measurements, and a Hitachi-Horiba model M-5 glass electrode pH meter for the pH measurements.
Reagents
All reagents used were of guaranteed reagent grade.
Gallium(III) and indium(III) standard solutions. A solution containing about 1 mg ml-' of gallium or indium was prepared by dissolving metal nitrate in a small amount of nitric acid and diluting with distilled
water. The solutions were standardized by the com- plexometric back titration with a standard thorium solution using Xylenol Orange as an indicator and diluted as required.
Hafnium solution. A solution containing about 5 mg ml-' of hafnium was prepared by dissolving hafnium chloride (Nakarai Chemicals) in distilled water and standardized complexometric back titration with a standard thorium solution using Xylenol Orange as an indicator.
Recommended procedure
To a sample solution (50 - 400 ml) containing 0.2 - 3 µg of gallium or 0.2 - 4 tg of indium, 20 mg of hafnium is added and the pH of the solution is adjusted to about 9.5 with aqueous ammonia (1+2). To settle hafnium hydroxide produced, the solution is allowed to stand for more than 10 min, then, the precipitate is collected on a 3G4 sintered-glass filter and washed with a small amount of aqueous ammonia (pH about 9.5).
The precipitate is dissolved with 1 ml of concentrated nitric acid, and the solution is diluted to 25 ml with distilled water. The atomic absorbance of gallium or indium is then measured under the operating condi- tions shown in Table 1. A blank is also run, using distilled water as a sample solution.
Results and Discussion
Optimum conditions for coprecipitation
According to the recommended procedure, the amount of hafnium which is necessary for the
coprecipitation of gallium and indium was studied with sample solutions (50 - 400 ml) containing 3 µg of gallium or 4 µg of indium. The required amount of hafnium for the quantitative collection of gallium or indium increased with increasing sample volume and more than 15 mg of hafnium was needed for 400 ml of sample solution. On the other hand, the presence of hafnium improved the precision of the atomic absorption measurements of gallium or indium, and the variations of the reproducibility of the peak heights observed from 200 repeated measurements became smaller compared with those in the absence of hafnium (Figs. 1 and 2). Furthermore, the sensitivities of the determinations were increased by the presence of hafnium, and the peak heights of gallium and indium became higher by 2.8 and 2.9 times, respectively, than those in the absence of hafnium (Figs. 3 and 4). These enhancing effects of hafnium were slightly affected by the change of the added amounts of hafnium, and almost constant peak heights were obtained over the concentration range from 0.2 to at least 2.0 mg ml-' of hafnium in cases of both gallium and indium. So, in these experiments, 20 mg of hafnium was used for the coprecipitation of gallium and indium.
To find the optimum pH range for the coprecipita- tion, the recovery of gallium or indium was examined with solutions (about 100 ml) containing 3 µg of gallium or 4 µg of indium. The maximum and almost constant recovery was obtained in the pH range 6.0 -
10.5 for gallium and in the pH range 6.0 -11.0 for indium. The recoveries of both metal ions were little influenced by the standing time of the precipitates.
Thus, almost 100% recoveries were obtained within a
Fig. 1 Reproducibility of gallium atomic absorbances exam- ined with a 25 ml solution containing 3 µg of gallium, l ml of concentrated nitric acid and 20 mg of hafnium (1) or 0 mg of hafnium (2). The relative standard deviations are plotted in such a manner that the relative standard deviation obtained from the first to tenth measurements is shown at the place of the tenth measurement.
Fig. 2 Reproducibility of indium atomic absorbances exam- ined with a 25 ml solution containing 4 µg of indium, l ml of concentrated nitric acid and 20 mg of hafnium (1) or 0 mg of hafnium (2). The relative standard deviations are plotted in such a manner that the relative standard deviation obtained from the first to tenth measurements is shown at the place of the tenth measurement.
ANALYTICAL SCIENCES AUGUST 1988, VOL. 4 419
few minutes after the formation of hafnium hydroxide and the recovered amounts remained unchanged for at least 10 h of standing in both cases. For the dissolution of hafnium hydroxide, nitric acid was prefered to hydrochloric acid, because hydrochloric acid seriously suppressed the gallium or indium absorbance (Fig. 5).
Optimization of operating conditions
Using the solution which contains 3 µg of gallium or 4 µg of indium, 20 mg of hafnium and 1 ml of concentrated nitric acid in 25 ml, the optimum conditions for the measurements of the atomic
absorbance of gallium or indium were studied.
At the drying stage, almost constant peak heights were obtained within 23 - 26 A of heating current and 30 - 60 s of heating time for gallium and within 22 - 26 A and 45 - 60 s for indium. With an increase of ashing current, the peak heights became higher and reached maximum values from 95 to 120 A for gallium and from 95 to 105 A for indium (Fig. 3). These absorbances remained almost constant from 20 to 50 s and from 30 to 60 s of ashing times, respectively. At the atomization stage, the maximum peak heights were obtained at 310 A (Fig. 4) and from 3 to 10 s for both ions. In Figs. 3 and 4, the results about the absence of hafnium are also shown. The absorbances of gallium and indium were not affected by the change of the argon sheath gas flow rate from 1 to 51 min-'.
Fig. 3 Effect of ashing current on gallium or indium peak height examined with a 25 ml solution containing 1 ml of concentrated nitric acid, 3 µg of gallium (1 and 3) or 4 µg of indium (2 and 4) and 20 mg of hafnium (1 and 2) or 0 mg of hafnium (3 and 4). Atomizing conditions: 310 A 5 s.
Fig. 4 Effect of atomizing current on gallium or indium peak height examined with a 25 ml solution containing 1 ml of concentrated nitric acid, 3 µg of gallium (1 and 3) or 4 µg of indium (2 and 4) and 20 mg of hafnium (1 and 2) or 0 mg of hafnium (3 and 4). Ashing conditions: (1 and 2) 100 A 30 s, (3)
118 A 305, (4)40A 30s.
Fig. 5 Effect of acids on gallium or indium peak height exam- ined with a 25 ml solution containing 20 mg of hafnium, 3 µg of gallium (1 and 3) or 4 µg of indium (2 and 4) and concen-
trated nitric acid (1 and 2) or concentrated hydrochloric acid (3 and 4).
Table 1 Operating enertrnmPtrv
conditions for the atomic absorption
However, the increase in the carrier gas flow rate up to 30 ml min' decreased the absorbances gradually. The optimum instrumental parameters obtained from these results are summarized in Table 1.
Calibration curves
The linear relationships through the point of origin between the peak height and the concentration were obtained over the range from 8 to 120 ng ml-' of gallium and from 8 to 160 ng ml-' of indium. The reproducibilities of these methods (relative standard deviations) for peak heights obtained from five repeated determinations were 2.3% for 3.tg of gallium and 2.9% for 4 tg of indium in about 100 ml of water.
The detection limits (signal/ noise=2) were 0.1 ng ml-' and 0.2 ng ml'' in 400 ml of the initial sample solution, respectively.
Interferences
The influences of each of 27 diverse ions on the
determination of 3.tg of gallium or 4 tg of indium were examined by coprecipitating gallium or indium from about 70 ml of sample solution. As shown in Table 2, large amounts of sodium, potassium, magnesium, and calcium did not affect the determinations. The other ions tested also did not cause serious interferences in either determination.
Recoveries of gallium and indium from spiked water samples
To evaluate the usefulness of these methods, the recoveries of gallium and indium from river and sea- water samples spiked with these metal ions were examined. The samples were filtered through a Toyo Roshi TM-2p membrane filter (pore size 0.45 µm) as soon as possible after sampling, and acidified with nitric acid to about pH 2 for storage. The results obtained are shown in Table 3. From the non-spiked water samples, gallium and indium were not detected by proposed methods, because of the very low levels usually found in these natural waters as shown in Table 3. From the spiked water samples, however, satis- factory results were obtained, indicating that the proposed methods are applicable for the analysis of river or seawater which contains down to 0.5 ng ml-' of gallium and indium.
References
1. D. C. Barron and B. W. Haynes, Analyst [London], 111, 19 (1986).
Table 2 Effect of diverse gallium and indium
ions on the determination of
Three micrograms of gallium or 4 µg of indium was copre- cipitated from about 70 ml of sample solution at pH 9.5.
a. Average of duplicate determinations.
Table 3 samples
Recovery of gallium and indium from spiked water
The recoveries obtained are the average of three replicate determinations. The contents of gallium are 9X 10'2 µg 1'' in river water and 3X 10'2 µg 1'' in seawater, and that of indium is 1 X 10'° µg 1'' in seawater. 23
a. Relative standard deviation.
n.d., not detected.
ANALYTICAL SCIENCES AUGUST 1988, VOL. 4
2. X. Shan, Z. Yuan and Z. Ni, Anal. Chem., 57, 857 (1985).
3. S. Hasegawa, T. Kobayashi, F. Hirose and H. Okochi, Bunseki Kagaku, 36, 371 (1987).
4. P. V. Botha and J. Fazakas, Anal. Chim. Acta,162, 413 (1984).
5. J. R. Clark, J. Anal. At. Spectrom., 1, 301(1986); Anal.
Abstr., 49, 4B28 (1987).
6. X. Shan, Z. Ni and Z. Yuan, Anal. Chim. Acta,171, 269 (1985).
7. P. V. Botha and J. Fazakas, Spectrochim. Acta, 39B, 379 (1984).
8. H. Han and Z. Ni, Fenxi Huaxue, 11, 571 (1983); Anal.
Abstr., 46, 7H4 (1984).
9. L. N. Sukhoveeva, G. G. Butrimenko and B. Ya.
Spivakov, Zh. Anal. Khim., 35, 649 (1980).
10. K. Kuga, Bunseki Kagaku, 30, 529 (1981).
11. L. Zhou, T. T. Chao and A. L. Meier, Anal. Chim. Acta, 161, 369 (1984).
12. A. Mizuike, in "Enrichment Techniques for Inorganic Trace Analysis", pp. 61- 66, Springer-Verlag, Berlin,
Heidelberg (1983).
13. J. Ueda and N. Yamazaki, Bull. Chem. Soc. Jpn., 59,1845 (1986).
421
14 J. Ueda and N. Yamazaki, Analyst [London], 112, 283 (1987).
15. K. Itsuki, H. Yagasaki and H. Fujinuma, Bunseki Kagaku, 34, T109 (1985).
16. A. A. Loginov, A. M. Tuzova and N. A. Rudnev, Zh.
Anal. Khim., 34, 1510 (1979).
17. E. Jackwerth and P. G. Willmer, Spectrochim. Acta, 33B, 343 (1978).
18. A. I. Novikov and V. I. Ruzankin, Soviet Radiochem.,15, 647 (1973); Anal. Abstr., 28, 4B71(1975).
19. S. Bruggerhoff, E. Jackwerth and S. Salewski, Spectrochim. Acta, 39B, 1 1 8 1 (1984).
20. E. Jackwerth and S. Salewski, Fresenius' Z. Anal. Chem., 310,108 (1982).
21. H. Berndt and J. Messerschmidt, Fresenius' Z. Anal.
Chem., 299, 28 (1979).
22. E. Jackwerth and P. G. Willmer, Fresenius' Z. Anal.
Chem., 279, 23 (1976).
23. S. Tunogai and S. Noriki, in "Kaiyoukagaku", p. 55, ed.
M. Nishimura, Sangyoutosho, Tokyo (1983).
(Received March 14, 1988) (Accepted June 17, 1988)
