LUMINESCENT DETERMINATION OF TRACE AMOUNTS OF BORON IN HIGH-PURITY TIN

Yu. L. Lel’chuk, V. A. Ivashina

(Presented by the Scientific-Methodological Seminar of the Faculty of Chemical Technology)

There is no description in the literature of methods for determining trace amounts of boron in high-purity tin. For the luminescent determination of traces of boron in other objects [1—4], use is made of the property of an alcoholic solution of benzoin to form with boron, in a weakly alkaline medium, a complex compound that fluoresces with a bright blue light when the solution is exposed to ultraviolet or violet light.

This method was studied in detail by White and co-workers [2, 3] and was later improved by Shcherbov, Korzheva, and Ponomarenko [4], who applied it to the determination of boron in certain types of mineral raw materials.

Benzoin, as a reagent for the luminescent determination of trace amounts of boron, compares favorably with the alizarin originally proposed for this purpose [5, 6], and later with quinalizarin [7], 1-amino-4-oxyanthraquinone [8], and resacetophenone [9]. Unlike benzoin, which forms with boron a fluorescent complex in a weakly alkaline alcoholic solution, all the above-mentioned reagents form the corresponding fluorescent compounds with it in concentrated sulfuric-acid and phosphoric-acid solutions, which is less convenient for the practical use of these reactions.

The ratio of boron to benzoin in the fluorescent complex is 1:1. The fluorescence intensity of the complex depends on the pH of the medium, on the amount of benzoin added, and on the alcohol content in the solution. The greatest brightness of luminescence is observed in a 95% solution of ethyl alcohol at pH 12.8. To maintain the pH of the solution and increase the stability of the complex, a glycine buffer is used [10—12].

A number of factors associated with photochemical changes both in the complex and in the excess benzoin not bound in the complex affect the luminescence intensity of the boron–benzoin complex [2, 13]. In this connection, in the procedure for the luminescent determination of trace amounts of boron in high-purity tin developed by us, until the moment of measuring the intensity of the complex, the solution being analyzed is protected from exposure to ultraviolet rays; during the preparation of all solutions being compared, the constancy of all conditions is strictly observed; and the determination of boron is carried out by the method of additions.

The high degree of selectivity of the reaction of boron with benzoin was demonstrated by White, Shcherbov, and co-workers [2—4]. In addition to boron, fluoresci-

compounds with benzoin are formed only by beryllium and germanium, and in the presence of magnesium and silica gel—by antimony and zinc. However, some elements quench the fluorescence of the complex to one degree or another and therefore must be removed from the solution being analyzed. Such elements include aluminum, iron, manganese, vanadium, and chromium.

The boron–benzoin complex has the greatest stability at pH 12.8. Under these conditions tin also interferes with the determination of boron.

For separating boron from the elements that interfere with its determination, the method of distilling off boron in the form of methyl borate is widely used. However, this method is lengthy, requires special quartz apparatus, and consumes a comparatively large amount of methyl alcohol. In addition, under the conditions for separating microamounts of boron, the accuracy of the method is low.

Having tested various methods of masking and removing tin and microimpurities of heavy metals in it, we chose their removal in the form of sulfides by hydrogen sulfide. To purify the hydrogen sulfide from possible boron impurities, it was passed through a wash bottle containing a 0.01% mannitol solution.

In the course of the study, the effects of the solution pH, the concentrations of benzoin and alcohol, the holding time of the complex before measurement of the fluorescence intensity, and the method of completing the determination on the accuracy, reproducibility, and sensitivity of the determination of microamounts of boron by this method were studied.

The water used is twice distilled in a quartz apparatus; hydrochloric acid is especially pure; hydrogen peroxide, chemically pure, 30% solution; ethyl alcohol is purified by distillation; sodium hydroxide, chemically pure; mannitol, analytical grade; sodium chloride, chemically pure; glycine, analytical grade. Evaporation was carried out under a drying electric lamp in a cabinet made of organic glass.

A glycine buffer solution with pH 12.8 is obtained by mixing, in strictly defined volumes, a solution of glycine and sodium chloride with a decinormal solution of sodium hydroxide. The first solution is prepared by dissolving 0.75 g of glycine and 0.60 g of sodium chloride in 100 ml of bidistillate; the second solution, by dissolving 2 g of sodium hydroxide in 500 ml of bidistillate. To obtain the glycine buffer with pH 12.8, 25 ml of the first solution are measured with a pipette, transferred to a 250-ml volumetric flask, made up to the mark with the second solution, and thoroughly mixed.

The glycine buffer is stored in a polyethylene vessel. The solution remains without noticeable changes for no more than 14 days.

A standard boron solution containing 1000 μg of boron in 1 ml is prepared by dissolving 0.5715 g of chemically pure boric acid in 100 ml of bidistillate. The solution is stored in a polyethylene vessel. A working standard solution containing 10 μg of boron in 1 ml is prepared on the day of its use by appropriate dilution of the stock standard solution with bidistillate.

Course of Analysis

To a weighed portion of high-purity tin of up to 0.5 g in a quartz beaker of 100–120 ml capacity, add 5 ml of especially pure hydrochloric acid and, dropwise in the cold with stirring, a 30% solution of hydrogen peroxide. After complete dissolution of the weighed portion, add 30 ml of bidistillate and pass a moderate stream of hydrogen sulfide from a Kipp apparatus for 30 min. To purify the hydrogen sulfide from possible traces of boron, a wash bottle containing a 0.01% mannitol solution is placed after the Kipp apparatus. The precipitate is then allowed to settle for 10–15 min-

The solution is filtered through a blue-ribbon filter into a platinum dish; the precipitate is washed several times by decantation, transferred to the filter, and washed another 2–3 times with bidistillate.

The filtrate and wash waters are neutralized with \(10\,N\) NaOH to pH 5.6–6.0 and evaporated under a lamp in an organic-glass cabinet at 95–90° almost to dryness. The residue is dissolved in 8 ml of water, transferred to a 25-ml volumetric flask, made up to the mark with ethyl alcohol, and mixed thoroughly.

After the addition of alcohol, a white precipitate of salts separates from the solution. It is filtered off through a No. 4 glass filter.

The determination of boron is completed by the method of additions. For this purpose, 4 aliquots of 5 ml each are taken from the filtered solution and transferred into 25-ml volumetric flasks. To three of them, boron additions of different amounts are added (from 0.5 to 2 μg). Then to each flask are added 1 ml of glycine buffer, 10 ml of alcohol, and 3 ml of a 0.5% alcoholic solution of benzoin; the volume is brought to the mark with alcohol and mixed thoroughly.

At the same time a blank solution is prepared, to which all the listed reagents are added in the same quantities and in the same sequence as to the solution being analyzed.

The fluorescence intensities of all five solutions are measured after 20 min in 20-mm cuvettes at a wavelength of 420 mμ, with an entrance slit width of the monochromator of 1.15 mm, an exit slit of 0.8 mm, and a photomultiplier voltage of 1500 V. The stability of the luminescence apparatus is checked and adjusted each time with the aid of a standard quinine bisulfate solution.

The fluorescence intensity should be measured with minimal exposure of the solutions to ultraviolet light, since under the action of ultraviolet rays the fluorescence intensity of the complex gradually decreases. In this connection, until the time of measurement the cuvettes should be covered on top with plates made of tin or another material that does not transmit ultraviolet and violet rays.

To increase sensitivity, it is recommended that the fluorescence intensity be measured using a mirror to shield the luminescence from the inactive walls of the cuvette.

The boron content in the aliquot, in μg, is calculated by the formula:

\[ B_{\mathrm{b}}=\frac{D_{\mathrm{b}}(I_{\mathrm{a}}-I_{0})}{I_{\mathrm{a+d}}-I_{\mathrm{a}}} \]

where \(B_{\mathrm{b}}\) is the weight of boron in the aliquot (5 ml), in μg;
\(D_{\mathrm{b}}\) is the weight of added boron, in μg; \(I_{\mathrm{a}}\) is the microammeter reading when measuring the luminescence intensity of the aliquot of the analyzed sample; \(I_{\mathrm{a+d}}\) is the microammeter reading when measuring the luminescence intensity of the aliquot of the analyzed sample with the given addition of boron; \(I_{0}\) is the microammeter reading when measuring the luminescence intensity of the blank solution.

From the 3 values of \(B_{\mathrm{b}}\) obtained by this formula, taking into account each boron addition, the arithmetic mean is taken and the percentage of boron in the tin being analyzed is calculated by the formula:

\[ \%B=\frac{B_{\mathrm{b}}\cdot V_{\mathrm{n}}\cdot 10^{-4}}{V_{\mathrm{a}}\cdot H} \]

where \(B_{\mathrm{b}}\) is the weight of boron in μg in the aliquot portion of the solution;
\(V_{\mathrm{n}}\) is the volume of the solution in which the tin sample is dissolved, ml;
\(V_{\mathrm{a}}\) is the volume of the aliquot, ml; \(H\) is the tin sample weight, g.

Table 1

Verification of the accuracy of boron determination in standard solutions carried through all stages of the analysis

Table 2

Verification of the accuracy of boron determination by the method of additions

Table 3

Results of parallel determinations of boron in two samples of high-purity tin

For calculating the percentage content of boron in the tin being analyzed, instead of the two formulas given above one may use the following general formula, taking into account that the aliquot volume is 5 ml and that the volume of the solution in which the entire sample is dissolved is 25 ml:

\[ \%B=\frac{D_b\left(I_a-I_0\right)}{H\left(I_{a+d}-I_a\right)\cdot 2000}. \]

The accuracy of the developed procedure was checked on standard boron solutions passed through all stages of the analysis (Table 1) and by the method of additions (Table 2). The results of parallel determinations of boron in a sample of high-purity tin are given in Table 3.

The luminescence setup consists of a PRK-2 mercury–quartz lamp, a UFS-6 light filter, a UM-2 universal monochromator, an FEU-10M photomultiplier, a VS-22 high-voltage stabilizer, and an M198/2 millivolt microammeter.

Conclusions

A procedure has been developed for the luminescent determination of boron in high-purity tin using benzoin. The sensitivity of the method is \(5\cdot 10^{-5}\%\). The relative error of determination does not exceed 20%. The duration of the analysis is 5 hours. The tin sample weight is within 0.5 g.

References

1. I. A. Radley. Analyst 63, 266, 1938.
2. Ch. E. White, A. Weissler, Busher. Anal. Chem. 19, 802, 1947.
3. Ch. E. White, D. E. Hoffman. Anal. Chem. 29, 1105, 1957.
4. D. P. Shcherbov, R. N. Korzheva, A. I. Ponomarenko. Methods of Luminescence Analysis, Proceedings of the 8th Conference on Luminescence, Publishing House of the Academy of Sciences of the BSSR, Minsk, p. 37, 1960.
5. P. W. Dankwort, I. Eisenbrand. Lumineszenz-Analyse im filtrierten ultravioletten Licht, Leipzig, 1956.
6. L. Szebledy, St. Tanay. Z. Anal. Chem. 107, 26, 1956.
7. L. Rosenthaler. Mikochim. 23, 191, 1937.
8. I. A. Radley. Analyst 69, 47, 1944.
9. Brozozowski, Ianowski. Z. Anal. Chem. 171, 49, 1959.
10. Q. Elliot, I. A. Radley. Analyst 80, 1018, 62—69, 1961.
11. D. D. Shcherbov, R. N. Plotnikova. Methods of Analysis of Chemical Reagents and Preparations, issue 4, IREA Publishing House, Moscow, p. 65, 1962.
12. G. I. Kuchmistaya. Methods of Analysis of Chemical Reagents and Preparations, issue 4, IREA Publishing House, Moscow, p. 69, 1962.
13. E. A. Bozhevolnov. Proceedings of IREA, issue 22. Goskhimizdat, p. 60, 1958.