On the Possibility of Determining Aluminum with Hematoxylin

V. A. Malevannyi, Yu. L. Lel’chuk

(Presented by the Department of Analytical Chemistry)

There is no consensus in the literature concerning the possibility of the colorimetric determination of aluminum with hematoxylin. Some authors [1–6] consider this reagent one of the principal reagents for determining aluminum; others [7–8] demonstrate that hematoxylin is unsuitable for determining aluminum in ferrous metals and their alloys. In the latest edition of Sandell’s monograph [9] it is stated that hematoxylin apparently yields to all other reagents for aluminum. The literature also gives contradictory data on the conditions for carrying out the reaction and on the analytical characteristics of the reagent. A summary of some literature data is given in Table 1.

Table 1

Summary of literature data on the conditions for determining aluminum with hematoxylin

To resolve the question of the suitability of hematoxylin for the photometric determination of small amounts of aluminum, we carried out work to study the reaction conditions and the principal analytical characteristics of the reagent and of its complex with aluminum.

Experimental Part

For the work, solutions of hematoxylin were used, prepared from a reagent-grade preparation produced by the English firm Lawson Chemical Ltd.

Working aluminum solutions with concentrations of \(5 \cdot 10^{-4}\) and \(5 \cdot 10^{-5}\) M were prepared on the day of use by diluting an initial standard solution with a concentration of \(5 \cdot 10^{-3}\) M. The initial standard solution was prepared from chemically pure aluminum sulfate. The titer of the solution was checked by the gravimetric ammonium method. To establish the required pH value, Sørensen buffer mixtures [11] were used. A control check of the pH of the solutions being photometered was carried out on a “Multoscope-5” pH meter (CSSR) with glass electrodes. The optical density of the solutions was measured on an SF-10 spectrophotometer.

Discussion of Results

The influence of the pH of the medium on the optical density of the aluminohematoxylin complex was studied in the range 1.4–14. Measurements were made at 620 mµ in a cuvette with a layer thickness of 50 mm in the presence of a 20-fold excess of reagent, after heating the solutions to be photometered on a boiling water bath for 10 minutes. As the comparison solution, a blank solution was used, to which all reagents were added in the same amount and in the same sequence, but without aluminum. The data obtained by us are shown in Fig. 1.

From the graph presented, it is evident that the aluminum complex with hematoxylin exists in the pH range 4–14 and has maximum optical density in the interval 7–11. It should be noted that, according both to our data and to the data of Molot [7], the aluminum complex with hematoxylin at pH 4.5 is destroyed to a considerable extent. This contradicts the recommendations available in the literature [3, 4] to stabilize the complex by lowering the pH of the solution to 4.5.

Fig. 1. Dependence of the optical density of the aluminum complex with hematoxylin on the pH of the solution.

\(C_{\mathrm{Al}} = 5 \cdot 10^{-5}\) M, \(C_R = 10^{-3}\) M,
\(V = 25\) ml, \(l = 50\) mm, \(\lambda = 620\) mµ,
comparison solution—reagent.

In the pH region 7–11, the optical density of the complex is practically independent of fluctuations in the pH of the solution; however, at pH 8 and above, the irreversible oxidation of hematoxylin to hematein [10] is sharply accelerated. Therefore, it is advisable to carry out the determination at pH 7–7.5.

In Fig. 2 are presented the spectrophotometric characteristics obtained by us for hematoxylin and its complex with aluminum at pH 7. As can be seen from the graph, the reagent solution has two absorption maxima—

tion. The first maximum—at 550 mµ—corresponds to the absorption maximum of hematein (the product of oxidation of the reagent by atmospheric oxygen [10]). The second maximum—at 400–450 mµ—corresponds to the absorption maximum of the unoxidized reagent. At pH 7 the absorption maximum of the aluminum complex with hematoxylin is located at 620 mµ. When the optical density is measured under these conditions, the maximum sensitivity of the reagent is achieved. The contrast of the oxidized reagent and its complex is equal to 70 mµ. The data obtained show that Sendel’s [3] recommendation to carry out photometry of the complex at 540 mµ is erroneous.

Fig. 2. Spectrophotometric curves of hematoxylin and its complex with aluminum. 1—oxidized reagent, 2—complex of aluminum and excess reagent, 3—aluminum complex. \(C_{\mathrm{Al}} = 5 \cdot 10^{-3}\ \mathrm{M}\), \(C_R = 10^{-3}\ \mathrm{M}\), \(V = 50\ \mathrm{ml}\), \(l = 50\ \mathrm{mm}\), pH = 7, reference solution—water.

According to Molot [7], the composition of the aluminum complex with hematoxylin, established by the saturation method, corresponds to 1 : 3. The data we obtained by the Ostromyslensky–Job method are shown in Fig. 3. They confirm Molot’s conclusion regarding the composition of the complex formed.

The molar extinction coefficient of the aluminum complex with hematoxylin, calculated from the data of the isomolar series, proved to be \(7.53 \cdot 10^4\ \mathrm{l\,mol^{-1}\,cm^2}\). Calculation of the molar extinction coefficient by the saturation method gave a similar result. On this basis one should consider hematoxylin a sensitive reagent for aluminum. It is known from the literature [7] that

Fig. 3. Determination of the composition of the complex by the Ostromyslensky–Job method. \(C_{\mathrm{Al}} = 5 \cdot 10^{-5}\ \mathrm{M}\), \(C_R = 5 \cdot 10^{-5}\ \mathrm{M}\), \(V = 50\ \mathrm{ml}\), \(l = 50\ \mathrm{mm}\), pH = 7, reference solution—water.

the molar extinction coefficients of aluminum complexes with aluminon, alizarin, and stilbazo are of the same order and are, respectively, \(1.24 \cdot 10^4\), \(1.01 \cdot 10^4\), and \(1.5 \cdot 10^4\ \mathrm{l\,mol^{-1}\,cm^2}\). However, in terms of photometric sensitivity (the relative difference between the molar extinction coefficients of the complex and the reagent), hematoxylin does not possess

of particular value. This is associated with the high value of the molar absorption coefficient of the reagent itself at a wavelength of 620 mµ, equal to \(2.6 \cdot 10^4\ \mathrm{mmole}^{-1}\ \mathrm{cm}^2\). Under the optimum conditions for determining aluminum, the photometric sensitivity is only 2.83. According to our data, the spectrophotometric sensitivity of the reagent toward aluminum is \(0.014\ \mu\mathrm{g}/\mathrm{cm}^2\).

Table 2

Effect of dilution on the optical density of a solution of the aluminum complex with hematoxylin

The aluminum complex with hematoxylin has low stability. Upon dilution of the solution, as is evident from the data in Table 2, its optical density decreases markedly. For the dilution, a solution of the complex with an Al concentration of \(10^{-5}\ \mathrm{M}\) was taken in the presence of a 20-fold excess of reagent. Calculation of the instability constant of the complex by the Babko method [12], from the data of Table 2, gave a value equal to \(2.4 \cdot 10^{-7}\).

From literature data it is known that the optical density of the aluminum complex with hematoxylin depends strongly on the standing time of the solution. We carried out experiments to study this dependence under the optimum determination conditions. The data obtained are given in Table 3. The determination was carried out with a 20-fold excess of reagent against the background of the reagent solution. To study the dependence of the optical density of the reagent on standing at \(18^\circ\mathrm{C}\), a solution with an aluminum concentration of \(5 \cdot 10^{-4}\ \mathrm{M}\) was taken; after 10-minute heating, a solution with an aluminum concentration of \(5 \cdot 10^{-5}\ \mathrm{M}\) was used. From the data presented it is evident that the optical density increases with time. However, upon storage for more than 6 hours, the color of highly diluted solutions of the complex disappears.

To determine the specificity of hematoxylin under the optimum conditions for its use, solutions of salts of a number of metals with a cation concentration equal to \(10^{-2}\ \mathrm{M}\) were used, with a 5-fold excess of reagent. The experiments showed that at pH 7.5 the cations \(\mathrm{Ti}^{4+}\) \((\lambda_{\max}=670\ \mathrm{m}\mu)\), \(\mathrm{Cr}^{3+}\) \((650\ \mathrm{m}\mu)\), \(\mathrm{Mn}^{2+}\) \((560\ \mathrm{m}\mu)\), \(\mathrm{Ca}^{2+}\) \((560\ \mathrm{m}\mu)\), \(\mathrm{Fe}^{2+}\) \((650\ \mathrm{m}\mu)\), \(\mathrm{Fe}^{3+}\) \((660\ \mathrm{m}\mu)\), and \(\mathrm{Cu}^{2+}\) \((750\ \mathrm{m}\mu)\) also react with hematoxylin. The complexes of \(\mathrm{V}^{5+}\), \(\mathrm{Mg}^{2+}\), \(\mathrm{Pb}^{2+}\), \(\mathrm{Zn}^{2+}\), and \(\mathrm{Ni}^{2+}\) have an absorption maximum in the absorption region of the reagent itself and do not noticeably affect the results of the determination of aluminum. Under the optimum conditions for determining aluminum, \(\mathrm{Fe}^{3+}\), \(\mathrm{Ti}^{4+}\), \(\mathrm{Cu}^{2+}\), and \(\mathrm{Cr}^{3+}\) interfere especially strongly.

The results obtained show that hematoxylin is of little use for determining aluminum in complex samples. In its specificity it differs little from such reagents as aluminon, alizarin, stilbazo, xylenol orange, and methylthymol blue. A very substantial drawback, however, is that it gives colored complexes with cations of both trivalent and divalent iron. As is known, in technical products iron

always accompanies aluminum and, consequently, will interfere with its determination with this reagent.

Despite the comparatively high value of the molar extinction coefficient of the aluminum complex, the photometric sensitivity of the reagent is very low. In terms of spectrophotometric sensitivity, hematoxylin is also considerably inferior to such reagents as alizarin, aluminon, eriochrome cyanine, and 8-hydroxyquinoline [9].

Table 3

Dependence of the optical density of the aluminum complex with hematoxylin on the standing time of the solution at 18°C and after 10-minute heating in a boiling water bath

A major drawback of hematoxylin is also the instability over time of the coloration of its complex with aluminum. This circumstance prevents the obtaining of accurate, well-reproducible results. Ten-minute heating of the solutions in a boiling water bath stabilizes the coloration to some extent, but even in this case it is not possible to achieve sufficient reproducibility and accuracy of the determination. With more prolonged heating, the alumino-hematoxylin lake precipitates, and the excess hematoxylin is partially oxidized to hematein, which leads to a shift of the absorption maximum into the long-wavelength region of the spectrum. In this connection, it is also necessary to take into account the possibility of oxidation of the lake itself.

Conclusions

1. The conditions for the use of hematoxylin for the photometric determination of aluminum have been studied. The principal analytical characteristics of the reagent and of its complex with aluminum have been determined: molar extinction coefficient, photometric and spectrophotometric sensitivity, stability of the complex over time, and instability constant of the complex. The range of interfering ions has been established.

2. The inadvisability of using hematoxylin for the photometric determination of aluminum in technical objects has been demonstrated.

References

1. J. H. Yoe. Photometric Chemical Analysis. Vol. 1, Moscow, ONTI, pp. 142–143, 1935.

2. F. D. Snell, C. F. Snell. Colorimetric Methods of Analysis, vol. 1, N. Y., 1945.

3. E. B. Sendel. Colorimetric Determination of Traces of Metals. Goskhimizdat, p. 138, 1949.

4. A. K. Babko, A. T. Pilipenko. Colorimetric Analysis. Moscow, Moskhimizdat, pp. 301–302, 1951.

5. F. Welcher. Organic Analytical Reagents. N. Y., p. 366, 1948.

6. J. H. Joe, L. A. Sarver. Organic Analytical Reagents. N. Y., p. 187, 1945.

7. L. A. Molot, I. S. Mustafin, N. S. Frulina. Proceedings of the Commission on Analytical Chemistry. Vol. 11, Moscow, Academy of Sciences of the USSR, pp. 231–242, 1960.

8. G. Ginsberg. Angew. Chem., 51, 663, 1938.

9. E. B. Sendel. Colorimetric Determination of Traces of Metals. Moscow, Mir Publishing House, pp. 211–212, 1964.

10. A. K. Babko, T. N. Nazarchuk. Journal of Analytical Chemistry, XIV, 2, 174–180, 1959.

11. E. Mislowitzer. Determination of the Concentration of Hydrogen Ions in Liquids. Leningrad, ONTI, pp. 349–352, 1930.

12. A. K. Babko. Physicochemical Analysis of Complex Compounds in Solutions. Kiev, Academy of Sciences of the Ukrainian SSR, 1955.