UDC 549.732

Crystallography

D. P. SHASHKIN, M. A. SIMONOV, Academician N. V. BELOV

CRYSTAL STRUCTURE OF URALBORITE Ca$_2$[B$_4$O$_4$(OH)$_8$]

The mineral uralborite*—a hydrous calcium metaborate, discovered and described by S. V. Malinko in 1961 ($^1$)—was kindly supplied by him to the crystal-chemistry laboratory of the All-Union Institute of Mineral Raw Materials for detailed structural investigation. Careful X-ray study of the specimens supplied established the presence of yet another calcium metaborate—vimsite ($^2$).

The occurrence of vimsite together with uralborite makes it necessary to treat critically the chemical analysis of the latter given in ($^1$). A repeated microchemical analysis of uralborite (T. I. Stolyarova, chemical laboratory of the All-Union Institute of Mineral Raw Materials) made it possible to conclude that the formula of the mineral CaB$_2$O$_4 \cdot 2$H$_2$O, previously derived from a complete chemical analysis (wt. %): SiO$_2$ 4.00; Al$_2$O$_3$ 1.12; Fe$_2$O$_3$ 2.13; CaO 35.27; MgO 0.67; B$_2$O$_3$ 38.06; H$_2$O$^+$ 19.08; $\Sigma = 100.33\%$, with the ratio CaO : B$_2$O$_3$ : H$_2$O = 1 : 1 : 2, is retained, and it was adopted as the basis for a detailed X-ray structural investigation.

The parameters of the monoclinic cell, refined by the powder method (RKU-114), are: $a = 6.92 \pm 0.01$ Å; $b = 12.35 \pm 0.02$ Å; $c = 9.80 \pm 0.02$ Å; $\beta = 83^\circ$. With specific gravity $d = 2.60$ g/cm$^3$, the cell contains $z = 8$ units of CaB$_2$O$_4 \cdot 2$H$_2$O.

The principal experimental material for the structure determination of uralborite was provided by layer-line scans (Weissenberg photographs, Mo radiation) about the axes $a$, $b$, and $c$: $0kl$—$5kl$, $hk0$, $hk1$, $h0l$, $h1l$ ($\max \sin \theta/\lambda = 0.95$ Å$^{-1}$). Reflection intensities were estimated on the standard $\gamma/2$ scale of blackening standards. Systematic extinctions on the scans unambiguously indicated the space group $C_{2h}^{5} = P2_1/n$. The solution of the structure of uralborite, as also of vimsite ($^3$), in accordance with a previously calculated weight criterion ($r' = 1, 2$), was carried out by the heavy-atom method ($^4$).

Table 1

Coordinates of the basis atoms of uralborite

A detailed analysis of the Patterson maps $p(uw)$ and $p(vw)$ and the first syntheses of electron density made it possible to localize, in the projections $xz$ and $yz$, the Ca atoms in general positions. The light O and B atoms were localized from three-dimensional complete and difference syntheses of electron density, constructed from the found coordinates of the Ca atoms.

The resulting structural model was refined by the method of least squares—

* Named after the locality of its discovery in the Urals.

Table 2

Interatomic distances in the structure of uralborite (Å)

(continued)

* Atoms that are related to the corresponding basis atoms by symmetry elements of the group.
** Common edges are shortened in accordance with Pauling’s rule.

tions on the M-20 machine at the Computing Center of Moscow University, using the programs of B. L. Tarnopol’skii and V. I. Andrianov (⁵). The final coordinates of the basis atoms of uralborite (54 parameters) are given in Table 1. The discrepancy factor corresponding to these coordinates for 1600 independent and nonzero reflections is \(R_{hkl}=15.1\%\) (\(B=0.4\ \text{Å}^2\)).

Interatomic distances are given in Table 2. From them, taking into account the valence balance and the gross chemical formula, it follows that in uralborite, as in vimsite, there are no “whole” \(\mathrm{H_2O}\) molecules, and in the anionic part \(\mathrm{O^{2-}}\) and \(\mathrm{OH^-}\) groups are specifically distinguished.

In the structure of uralborite all boron atoms are in tetrahedral coordination and form a compact island group of four crystallographically independent B-tetrahedra connected by common vertices. As usual, the O atoms common to two adjacent B atoms, while the eight anions participating in the coordination of only one B (and Ca), are represented by \(\mathrm{OH^-}\) groups, i.e., the formula of the island boron–oxygen radical in uralborite is \([\mathrm{B_4O_4(OH)_8}]^{4-}\). As is seen from Fig. 1, in the boron–oxygen radical three B-tetrahedra form a triple ring, similar to the silico–oxygen ring \([\mathrm{Si_3O_9}]\), and the fourth is attached to it through a common vertex. This radical may be regarded as the interesting feature of the structure of uralborite that has been revealed, primarily as a previously unknown new radical of tetrahedra; the discovery of this radical changes the position

Fig. 1. Uralborite. Projection \(yz\). Oxygen tetrahedra are emphasized; Ca atoms belonging to the cell are shown by spheres.

of uralborite in the classification of borates: contrary to the earlier classification (⁶, ⁷), which assigned it to the chain group, it falls into the category of island borates. The addition (external) of a fourth tetrahedron to the triple ring did not change the metacharacter of the radical: \([\mathrm{B_{3+1}O_{12}}]=[\mathrm{BO_3}]_4\).

Two independent Ca atoms are located in 8-vertex polyhedra in the form of 12-face-

approximately equal in volume. These delta-dodecahedra (all faces are triangles) join along a common edge into pairs which, obeying the glide plane \(n\) at two levels, are extended along the long diagonal \((a+c)\) of the (010) projection into an infinite ribbon; neighboring pairs in the ribbon are also connected along common edges. This ribbon is linked by vertices on two sides in the direction of the diagonal \((a-c)\) with two translationally identical ribbons, forming an infinite layer of Ca polyhedra (Fig. 2). For the period \(b=12.34\ \text{Å}\) there are two such layers at the levels \(y=\frac14\) and \(y=\frac34\), constituting the “engineering” basis of the structure. The layers of Ca polyhedra in uralborite resemble layers of Sc octahedra with a corundum motif—six-membered loops in the structure of tortveitite \((^{8,9})\). The boron–oxygen radicals \([\mathrm{B}_4\mathrm{O}_4(\mathrm{OH})_8]^{4-}\) fit into the six-membered rings of Ca polyhedra, cementing them and joining with them by common edges (in the two stories of Ca polyhedra, by pairs of very shortened edges; Table 2) and common vertices. In the section of the structure perpendicular to the \(a\) axis (Fig. 3), the cementation of two layers by B tetrahedra into a three-dimensional framework is clearly visible. The connection of the layers with one another by boron–oxygen groups is so strong that, despite the layered arrangement of the Ca polyhedra, cleavage is absent in the mineral. In accordance with the structure determined, the formula of the mineral is
\(\mathrm{Ca}_2[\mathrm{B}_4\mathrm{O}_4(\mathrm{OH})_8]\).

Fig. 2. Uralborite. Layer of Ca polyhedra

Fig. 3. Uralborite. Section \(x=\frac14\). Geometry of the connection of Ca polyhedra by boron–oxygen radicals in the direction of the \(b\) axis

All-Union Scientific-Research
Institute of Mineral Raw Materials
Moscow

Received
14 VII 1969

CITED LITERATURE

1. S. V. Malinko, Zap. Vsesoyuzn. mineralog. obshch., part 90, issue 6 (1961).
2. D. P. Shashkin, M. A. Simonov et al., DAN, 182, No. 6 (1968).
3. D. P. Shashkin, M. A. Simonov, N. V. Belov, DAN, 182, No. 4 (1968).
4. M. A. Simonov, Candidate dissertation, Crystalline structures of sodium–cadmium silicates, 1969.
5. B. L. Tarnopol’skii, V. I. Andrianov, ZhSKh, 4, No. 3, 434 (1963).
6. G. B. Bokii, V. B. Kravchenko, ZhSKh, 7, No. 6 (1966).
7. A. S. Povarenykh, Crystallochemical classification of mineral species, Kiev, 1966.
8. N. V. Belov, Crystallochemistry of silicates with large cations, Publishing House of the Academy of Sciences of the USSR, 1961.
9. W. H. Zachariasen, Zs. Kristallogr., 73, 1 (1930).