CHEMISTRY

L. M. KOVBA, V. K. TRUNOV

ON THE INVESTIGATION OF DOUBLE OXIDES CONTAINING TUNGSTEN, TANTALUM, OR NIOBIUM

(Presented by Academician V. I. Spitsyn on 17 VII 1962)

There is a large amount of data in the literature on compounds of the oxides of tungsten, tantalum, and niobium with basic oxides, i.e., on tungstates, niobates, and tantalates. Double oxides have been less studied (somewhat better than others—the compounds with niobium pentoxide \((^1)\)). The present investigation to some extent fills this gap.

We investigated the reactions of thorium, cerium, and uranium dioxides with tantalum and niobium pentoxides, and of tantalum pentoxide with tungsten trioxide; new data were also obtained on oxygen compounds of uranium and tungsten. Mixtures of oxides were sintered in air, in nitrogen, or in evacuated quartz ampoules. Phase analysis of the sintering products was carried out in a Guinier camera. The results are given in Table 1.

Table 1*

Results of X-ray phase analysis of calcination products

* The \(\mathrm{WO_3—Ta_2O_5}\) system was investigated completely (at intervals of 10 at. %). Only the compounds indicated in the table were found.

Uranium, thorium, and cerium dioxides form with niobium and tantalum pentoxides a group of compounds of the same type, \(\mathrm{Me(ЭO_3)_4}\). The structures of these compounds are very close to perovskite, but on the X-ray patterns a considerable number of superstructure lines are observed, and the true cells are tetragonal. It is interesting to note that niobates and tantalates have different superstructures: in the case of tantalates all lines of the X-ray patterns are indexed with a period \(a\) twice as large as that of the perovskite cell, while the period \(c\) does not change; in niobates the period \(c\) is doubled, and the period \(a\) does not increase. Table 2 gives the results of indexing the X-ray patterns of thorium niobate and tantalate.

Cerium niobate crystallizes in a rhombic cell, but the deviation from tetragonal symmetry is small. The intensity of the lines indicates displacement of tantalum and niobium atoms from the corners of the perovskite cell. The superstructure in the tantalates corresponds to a completely ordered arrangement of the atoms of the tetravalent elements over \(1/4\) of the cuboctahedra.

In niobates, layers of empty cuboctahedra alternate along the \(c\) axis with filled layers. In the filled layers, statistically, half of the cuboctahedra are occupied.

In this connection it was of interest to check whether the phase \((\mathrm{U}, \mathrm{W})\mathrm{O}_3\), which we had found earlier \((^2)\), is not the lower limit of the phases \(\mathrm{U}_x\mathrm{WO}_3\), and to determine whether the phase \(\mathrm{U}(\mathrm{WO}_3)_4\) has a superstructure. The specimen \(\mathrm{U}(\mathrm{WO}_3)_4\) proved to be two-phase (a mixture of the perovskite phase and \(\mathrm{WO}_2\)). On the X-ray patterns of \(\mathrm{U}_{1/8}\mathrm{WO}_3\) and \(\mathrm{U}_{1/12}\mathrm{WO}_3\) only lines of the perovskite phase were present, without superstructure lines. The periods of the cubic lattices are \(3.813\ \mathrm{kX}\) for \(\mathrm{U}_{1/8}\mathrm{WO}_3\) and \(3.804 \pm 0.001\ \mathrm{kX}\) for \(\mathrm{U}_{1/12}\mathrm{WO}_3\). The lattice parameter of the phase \((\mathrm{U}, \mathrm{W})\mathrm{O}_3\) corresponds to the composition \(\mathrm{U}_{1/12}\mathrm{WO}_3\).

Table 2

Results of indexing the X-ray patterns of \(\mathrm{Th}(\mathrm{NbO}_3)_4\) and \(\mathrm{Th}(\mathrm{TaO}_3)_4\)

Thus, the tantalates and niobates of uranium, cerium, and thorium are, structurally, rather close to the meta-tantalates and niobates of the alkali and alkaline-earth metals. Apparently, this is explained by the sufficiently large radii of the thorium, cerium, and uranium ions. It is characteristic that their dioxides are isostructural (fluorite type) and form continuous series of solid solutions.

Table 3

Lattice parameters (Å) of niobates and tantalates of U, Th, and Cl

It would be of undoubted interest to obtain analogous compounds with dioxides of the rutile type, if such compounds are in fact formed (the available data are contradictory).

In the \(\mathrm{Ta}_2\mathrm{O}_5 — \mathrm{WO}_3\) system, two new phases were found: \(3\mathrm{WO}_3 \cdot \mathrm{Ta}_2\mathrm{O}_5\) and \(3\mathrm{Ta}_2\mathrm{O}_5 \cdot 4\mathrm{WO}_3\). The compound \(3\mathrm{WO}_3 \cdot \mathrm{Nb}_2\mathrm{O}_5\) had been described earlier \((^1)\). The X-ray patterns

Table 4

Results of indexing the X-ray diffraction patterns of \(3\mathrm{WO}_3\cdot\mathrm{Ta}_2\mathrm{O}_5\) and \(3\mathrm{WO}_3\cdot\mathrm{Nb}_2\mathrm{O}_5\)

The phases \(3\mathrm{WO}_3\cdot\mathrm{Nb}_2\mathrm{O}_5\) and \(3\mathrm{WO}_3\cdot\mathrm{Ta}_2\mathrm{O}_5\) are very similar and are indexed in tetragonal cells (Table 4) with the parameters: \(a = 12.166 \pm 0.003\ \mathrm{kX}\), \(c = 3.9265 \pm 0.0004\ \mathrm{kX}\) for \(3\mathrm{WO}_3\cdot\mathrm{Nb}_2\mathrm{O}_5\), and \(a = 12.25 \pm 0.01\ \mathrm{kX}\), \(c = 3.873 \pm 0.002\ \mathrm{kX}\) for \(3\mathrm{WO}_3\cdot\mathrm{Ta}_2\mathrm{O}_5\). There are two formula units \(3\mathrm{WO}_3\cdot\mathrm{Me}_2\mathrm{O}_5\) per cell. The cells of these compounds may be regarded as superstructures with respect to \(\mathrm{ReO}_3\) (the period \(a\) of the tetragonal cell corresponds to the 310 direction of the \(\mathrm{ReO}_3\) cell). An analogous superstructure was found by Magnéli\(^3\) in tetragonal tungsten bronzes, but the analogy is apparently formal, since the causes of the occurrence of the superstructure are obviously different.

Moscow State University
named after M. V. Lomonosov

Received
14 VII 1962

REFERENCES CITED

\(^1\) H. J. Goldschmidt, Metallurgia, 62, 373 (1960).
\(^2\) V. K. Trunov, L. M. Kovba, V. I. Spitsyn, DAN, 141, 114 (1961).
\(^3\) A. Magneli, Arkiv kemi, 1, 213 (1949).