On the Analysis of Tantaloniobates

Separation of Tin, Zirconium, and Tungsten from Tantalum and Niobium

M. E. Efendi, I. P. Onufrienok

(Presented by Prof. B. V. Tronov, Doctor of Chemical Sciences)

In our works¹ we gave a new, simpler method for separating tantalum from niobium and titanium from tantalum and niobium. On the basis of these works it is possible to simplify considerably the scheme for the analysis of certain tantaloniobates and industrial products containing tantalum and niobium. For a radical restructuring of the schemes for the analysis of tantaloniobates, data were lacking on the behavior of the genetic companions of tantalum and niobium—tin, zirconium, and tungsten. It was unclear how the tannin compounds of tin, zirconium, and tungsten would behave under the conditions of the methods we had developed. Therefore, in the present work the chief attention was devoted to studying the properties of the tannin compounds of these elements and to using these properties for the separation of tin, zirconium, and tungsten from tantalum and niobium.

Tin

In analyses for tantalum and niobium, in practice one has to deal with tin ranging from traces to considerable amounts.

The existing methods for separating tin from tantalum and niobium amount to removing the tin with hydrogen sulfide immediately after conversion into a soluble state [1]. By these methods tin is separated well from tantalum and niobium, but they are rather laborious. We set ourselves the goal of finding a less laborious method of separation, which we succeeded in doing after studying the properties of the tannin compounds of tin.

To study the properties of the tannin compounds of tin, a stock solution with a titer of 1 mg/ml calculated as $\mathrm{SnO_2}$ was prepared; for this purpose a weighed portion of metallic tin was decomposed with sulfuric acid, the excess sulfuric acid was evaporated, the residue was fused with $\mathrm{KHSO_4}$, the melt was leached with a cold saturated solution of ammonium oxalate, and the solution was diluted with the latter in a volumetric flask. The tannin compounds from the stock solution were obtained under the same conditions in which

tannin compounds of tantalum and niobium are obtained; at the same time it was found that tin, like tantalum and niobium, is precipitated quantitatively by tannin when heated to boiling and ammonia is added to an alkaline reaction with Congo red, but acidic to litmus. The tin tannin compounds thus obtained were washed free of oxalate ions with a hot 2% solution of NH₄Cl and were studied for solubility in various solvents.

It was established that the tannin compound of tin is capable of being peptized by water that does not contain strong electrolytes, but when washed with a 2% ammonium chloride solution it is not prone to pass into a colloidal solution. It is readily soluble in dilute oxalic and hydrochloric acids and in caustic alkalis. It is somewhat less soluble in dilute sulfuric and gallic acids. It is also soluble in ammonia and in ammonium carbonate, but with the addition of a small amount of hydrogen peroxide.

Especially valuable, as repeated quantitative experiments showed, was the fact that the tannin compound of tin dissolves completely in a hot 1 N solution of hydrochloric acid. The latter made it possible to combine the operation for separating tin from tantalum and niobium with the separation from them of the tannin compound of titanium.

Following the method we developed for separating tannin compounds of tantalum and niobium from titanium, tin, and zirconium, based on dissolution in 1 N hydrochloric acid (the method is given below in the section “Zirconium”), we carried out the separation of tin and titanium from tantalum and niobium in two mixtures, in which the ratios of Ta₂O₅, Nb₂O₅, SnO₂, and TiO₂ were (1:1:1:1) for the first mixture and (1:1:2:2) for the second mixture. For both mixtures, excellent results were obtained for the separation of tin and titanium from tantalum and niobium. In the second mixture, only traces of titanium were found in the Ta₂O₅ + Nb₂O₅ residue. It was also noted that titanium, in the presence of tin, is separated more completely from tantalum and niobium.

Analogous experiments, in which 1 N hydrochloric acid was replaced by 1% gallic acid, did not give quantitative separation of tin and titanium.

Zirconium

Zirconium, if its minerals occur in an easily decomposable form, may likewise be present in significant amounts in solutions of pyrosulfate melts in ammonium oxalate, because ammonium oxalate does not precipitate zirconium sulfate. Therefore, in analyses for tantalum and niobium, one always has to deal with the separation of zirconium; in doing so, various methods are used [2], in which either the earth acids are brought into solution, leaving zirconium in the precipitate, or the reverse [1]. We were primarily interested in the question of the behavior of the tannin compound of zirconium in a 1 N solution of hydrochloric acid.

To study the solubility of the tannin compound of zirconium, we prepared a stock zirconium solution by treating a weighed portion of ZrOCl₂·8H₂O with sulfuric acid, followed by fusion with KHSO₄ and leaching of the melt in a cold saturated solution of ammonium oxalate. The stock solution had a ZrO₂ titer equal to 0.88 mg/ml.

The tannin compound of zirconium was obtained in the same way as for tin or titanium, but with the difference that neutralization with ammonia after introduction of tannin was carried out to an alkaline reaction to litmus. If, however, one adheres to the conditions for precipitating the tannin compounds of titanium and

earth acids, i.e., to complete the neutralization with ammonia in an alkaline solution according to Congo red and still acid according to litmus, then significant amounts of zirconium remain in solution unprecipitated. After thorough washing of the precipitate from oxalate ion with a 2-percent solution of NH₄Cl, the precipitate was tested for solubility in various solvents.

It was established qualitatively that the tannin compound of zirconium is soluble in oxalic acid, in dilute solutions of hydrochloric and sulfuric acids, in gallic acid after gentle heating; in ammonia, ammonium carbonate, and a 0.1 N borax solution, dissolution occurs only after the addition of a small amount of hydrogen peroxide.

Quantitative experiments showed complete solubility of the tannin compound of zirconium in 1 N hydrochloric acid; the latter made it possible to assign this compound to the group of tannin compounds separated from earth acids by treatment with a 1 N solution of hydrochloric acid.

Therefore, in subsequent experiments mixtures of tannin compounds of tantalum, niobium, titanium, tin, and zirconium in a ratio of (1:1:1:1:1) were subjected to testing. The experiments showed that with a single treatment with 1 N hydrochloric acid, only TiO₂ remains in the residue of the earth acids, in amounts not exceeding 0.2 mg. After double treatment, titanium remains only in traces.

The preparation of the tannin compounds and their separation were carried out by us according to the following procedure.

A pyrosulfate fusion of the oxides of tantalum, niobium, titanium, tin, and zirconium is leached with a cold-saturated solution of ammonium oxalate. To the resulting solution is added an equal volume of cold-saturated ammonium chloride and an excess of a 2-percent solution of tannin. The solution is heated to boiling and neutralized with ammonia to an alkaline reaction according to Congo red, but still acid according to litmus. The solution with the precipitate is boiled for 2–3 minutes and kept on a boiling water bath for 30 minutes. The precipitate is filtered off through a medium-density filter, washed 2–3 times with a hot 2-percent solution of ammonium chloride, after which, with the same wash liquid, it is washed back into the beaker, mixed well, and again filtered through the same filter; it is washed on the filter 7–8 times. Then the precipitate is washed as completely as possible from the filter back into the beaker with 70–80 ml of cold 1 N hydrochloric acid. The washed-down precipitate is heated with the acid in the beaker, while stirring with a glass rod, to boiling, then again filtered through the same filter. The precipitate on the filter is washed 7–8 times with hot 1 N hydrochloric acid, after which the precipitate is washed as completely as possible from the filter back into the beaker with a 0.1 N solution of oxalic acid. After adding to the washed-down precipitate in the beaker 1–2 ml of 1 N oxalic acid, the precipitate dissolves; on heating and stirring, the entire solution obtained is filtered through the same filter into another clean beaker. The filter is washed well with a hot 0.1 N solution of oxalic acid and discarded. In the collected oxalic-acid solution the tannin compounds are again precipitated; for this, to the solution is added an equal volume of cold-saturated ammonium chloride solution and 2 ml of a 2-percent solution of tannin. The solution is heated to boiling and neutralized with ammonia, as described above; once again, as before, the washed precipitate is treated with 1 N hydrochloric acid, after which the precipitate of tannin compounds of tantalum and niobium, washed with hot 1 N hydrochloric acid, is ignited in a weighed ...

in a quartz or platinum crucible and ignited to constant weight. It is weighed as $\mathrm{Ta_2O_5}$ and $\mathrm{Nb_2O_5}$.

Tungsten

In analyses for tantalum and niobium, tungsten may also be present in oxalic-acid solutions, from traces up to considerable amounts. From the analyst’s point of view, special attention must be paid to tungsten because tungsten is the most difficult of all to separate from tantalum and niobium, and also because the color of its tannin compound is the same as that of the corresponding tantalum compound. A distinguishing feature of the tannin compound of tungsten, according to our observations, is that it changes its color from yellow to reddish brown when treated with hot 1 N hydrochloric acid, whereas the same change in color in the tantalum compound occurs only when treated with hot 9 N hydrochloric acid. As regards the separation of tungsten from tantalum and niobium, two methods are known, both based on the ability of the earth acids to give, in an alkaline medium, insoluble sodium salts $4\mathrm{Na_2O}\cdot 3\mathrm{Ta_2O_5}$ and $7\mathrm{Na_2O}\cdot 6\mathrm{Nb_2O_5}$, and on the solubility under these conditions of the sodium salt of tungstic acid [2].

Our experiments on separating the tannin compound of tungsten from the corresponding compounds of tantalum and niobium were not successful. Preparation of the tannin compound of tungsten under the same conditions as for the analogous compounds of tantalum and niobium showed that tungsten is not completely precipitated by tannin; the amount of unprecipitated tungsten in this case reaches 15%. The tannin compound of tungsten is soluble in oxalic acid, in 1% gallic acid with slight heating, completely soluble in alkalis, and in ammonium carbonate only with the addition of a small amount of hydrogen peroxide; it is poorly soluble or completely insoluble in mineral acids. When boiled, 100 ml of 1 N hydrochloric acid dissolve 2.5 mg of the tannin compound of tungsten out of 8.4 mg taken for the experiment, calculated as $\mathrm{WO_3}$.

Mixtures of acids (100 ml of 1 N HCl and 1 ml of $\mathrm{H_3PO_4}$) completely dissolve the tannin compound of tungsten if these compounds are taken separately for the experiment; but if such a mixture is used to treat the tannin compounds of tungsten together with the corresponding compounds of tantalum and niobium, then the tannin compounds of the earth acids also pass partly into solution.

Conclusions

1. A new method has been developed for separating tin and zirconium from tantalum and niobium, based, as in the case of titanium, on the solubility of the tannin compounds of these elements in 1 N hydrochloric acid. The method makes it possible to separate tin, zirconium, and titanium simultaneously, which considerably accelerates and reduces the cost of determining tantalum and niobium.

2. The work has shown a certain possibility of separating tungsten from tantalum and niobium by means of a hydrochloric–phosphoric acid mixture, but the optimal conditions under which this separation is possible have not been found.

References

1. Collection Analysis of Mineral Raw Materials, edited by B. G. Karpov, Yu. M. Knipovich, and Yu. V. Morachevsky. ONTI Publishing House, Leningrad, pp. 173 and 171, 1936.

2. Hillebrand, W. F.; Lundell, G. E. Practical Handbook of Inorganic Analysis. ONTI Publishing House, Moscow, 1935.

1. See the article in this volume, “On the Chemistry of Tantalum and Niobium.” ↩