Received 15 March 2010
aDepartment of Ecology and Protection of the Environment, Volyn National University, Voli Avenue 13, 43009 Lutsk, Ukraine,bW. Trzebiatowski Institute of Low Temperature and Structure Research, Polish Academy of Sciences, Okólna Street 2, PO Box 1410, 50-950 Wroclaw, Poland, and cDepartment of General and Inorganic Chemistry, Volyn National University, Voli Avenue 13, 43009 Lutsk, Ukraine
A previously unknown modification of dicopper(I) triselenostannate(IV), Cu2Se3Sn, has been obtained from the Cu2Se-SnSe2 quasi-binary system and investigated using X-ray single-crystal diffraction. The Se atoms are stacked in a closest-packed arrangement with the layers in the sequence ABC. The Cu atoms occupy one-third of the tetrahedral interstices, whereas the Sn atoms are located in one-sixth of the tetrahedral interstices. All the atoms occupy general positions. The structure possesses pseudo-inversion symmetry. The Cu2Se3Sn structure investigated in this paper (96 atoms per unit cell, ordered distribution of Cu and Sn over 12 cation positions) is a superstructure of the reported cubic (eight atoms per unit cell, random distribution of Cu and Sn over one cation position) and monoclinic (24 atoms per unit cell, ordered distribution of Cu and Sn over three cation positions) modifications.
As a continuation of our studies of ternary chalcogenides, we have examined the Cu2Se-SnSe2 system because of the reported formation of several phases of composition Cu2Se3Sn that belong to the family of low-melting-point compounds having a tetrahedral lattice, which are of interest for their semiconducting and optical properties (Sharma et al., 1977; Fernandez et al., 1996). Knowledge of the crystal structure of the Cu2Se3Sn compounds is important for understanding their properties. Sharma et al. (1977) indicated that Cu2Se3Sn crystallizes in the cubic sphalerite structure (space group F3m, a = 5.6877 Å). Recently, Delgado et al. (2003) described the structure of Cu2Se3Sn in a monoclinic unit cell [a = 6.9670 (3) Å, b = 12.0493 (7) Å and c = 6.9453 (3) Å, and = 109.19 (1)°] of the Cu2GeS3 structure type (space group Cc). Both studies were based on X-ray powder diffraction data. We present here the crystal structure of a previously unknown modification of Cu2Se3Sn based on X-ray single-crystal diffraction analysis.
The asymmetric unit of the title compound contains eight Cu atoms, four Sn atoms and 12 Se atoms (Fig. 1). Each of the formally CuI and SnIV ions is surrounded by four Se2- anions at distances that agree well with the sums of the respective ionic radii (Wiberg, 1995). The crystal lattice consists of corner-sharing [CuSe4] and [SnSe4] tetrahedra. Since the metal-centred tetrahedra are connected only by the corners, the Se-centred coordination environment is also tetrahedral. Similar values of the Cu-Se and Sn-Se interatomic distances for tetrahedral surroundings are observed in the structures of LnCuSe2 and Eu2SnSe5 (Daszkiewicz et al., 2008; Evenson & Dorhout, 2001). However, from the bond-valence point of view the eight symmetry-independent CuI ions are overbonded, because the bond-valence sums (BVS) for these ions [based on Cu-Se distances ranging from 2.380 (5) to 2.495 (5) Å] are greater than the formal oxidation state, 1.40 (Table 1) (Brown, 1996). Similarly, each of the Se2- anions is overbonded. In this case, the Sn4+ ions must be underbonded, as indicated by the calculated BVS [based on Sn-Se distances in the range 2.488 (3)-2.627 (3) Å] of 3.74, because the difference between the BVS for the cations and anions must be zero.
The Cu2Se3Sn structure described here, with 96 atoms in the unit cell, is a superstructure of the cubic (Sharma et al., 1977) (space group F3m, eight atoms per unit cell) and monoclinic (Delgado et al., 2003) (space group Cc, 24 atoms per unit cell) modifications reported earlier. The known modifications were investigated using X-ray powder diffraction, while the present monoclinic superstructure was investigated using X-ray single-crystal diffraction. The structures are similar (Fig. 2), in that the Se atoms in all modifications of Cu2Se3Sn are stacked in a closest-packed arrangement with the layers in the sequence ABC (cubic closest packing). In the cubic modification, a mixture of randomly distributed Cu+ and Sn4+ ions (Cu + Sn) occupy half of the tetrahedral interstices. In the structures of both monoclinic modifications, the Cu+ ions occupy one-third of the tetrahedral interstices, whereas the Sn4+ ions are located in one-sixth of the tetrahedral interstices. The distribution of the cation positions in both monoclinic modifications is ordered. In both modifications, the Sn-centred tetrahedra create zigzag chains along the c axis. However, the period of the chain is one-quarter as long in the previously reported structure than in the present superstructure, and this is reflected in the relation of the lattice parameters, c 4c'. Moreover, the amplitude of the chain is 1.5 times larger in the superstructure.
The presence of three structures for Cu2Se3Sn can be explained in two ways. The first is that the three modifications really exist. The cubic modification (random distribution of Cu+ and Sn4+ atoms over one position) is a high-temperature modification, while the monoclinic modifications with different ordered distributions of the positions of Cu and Sn are low-temperature modifications. The second is that only one or two modifications exist. The basic fragments of the structures for all three structures are similar. It is possible that the superstructure reflections measured in the X-ray single-crystal investigation were missed in the previous powder studies. Further work will be required on the Cu2Se-SnSe2 system to determine how many structural modifications actually exist.
| || Figure 1 |
The coordination environments for the four symmetry-independent Sn and eight Cu atoms in the asymmetric unit of Cu2Se3Sn. Displacement ellipsoids are drawn at the 50% probability level. [Symmetry codes: (i) x + , -y + , z + ; (ii) x + , y + , z; (iii) x - 1, y, z; (iv) x - , -y + , z + ; (v) x + , y - , z; (vi) x + 1, y, z; (vii) x - , y - , z; (viii) x - , y + , z.]
| || Figure 2 |
The packing of the Cu- and Sn-centred tetrahedra in (a) the cubic (Sharma et al., 1977), (b) the first monoclinic (Delgado et al., 2003) and (c) the second monoclinic (this work) modifications of Cu2Se3Sn.
A sample of composition Cu2Se3Sn was prepared by melting the high-purity (better than 99.9 wt%) elements in an evacuated silica tube. The ampoule was heated at a rate of 100 K h-1 in a tube furnace to a temperature of 770 K, then heated at a rate of 20 K h-1 to a maximum temperature of 1380 K and kept at this temperature for 2 h. The ampoule was then cooled slowly (at a rate of 10 K h-1) to 620 K and annealed at this temperature for 500 h. After annealing, the sample was quenched in cold water. A diffraction-quality single crystal was selected from the sample.
The systematic absences were found to be consistent with the space group Cc, which was assigned for the crystal structure determination. Eight positions for Cu, four for Sn and 12 for Se were determined. All positions are fully occupied. A mixed occupation of Cu and Sn on each cation site was checked, but in all cases the refinement was unstable. The structure was checked with PLATON (Spek, 2009), which detected a pseudo-inversion centre. A refinement in C2/c was unsuccessful. The structure was refined as a twinned model with a twin fraction of 0.058 (14), because the Flack parameter (Flack, 1983) initially refined to 0.078 (15).
Data collection: CrysAlis CCD (Oxford Diffraction, 2007); cell refinement: CrysAlis RED (Oxford Diffraction, 2007); data reduction: CrysAlis RED; program(s) used to solve structure: SHELXS97 (Sheldrick, 2008); program(s) used to refine structure: SHELXL97 (Sheldrick, 2008); molecular graphics: DIAMOND (Brandenburg, 2009); software used to prepare material for publication: publCIF (Westrip, 2010).
Supplementary data for this paper are available from the IUCr electronic archives (Reference: SQ3242 ). Services for accessing these data are described at the back of the journal.
Brandenburg, K. (2009). DIAMOND. Release 3.2c. Crystal Impact GbR, Bonn, Germany.
Brown, I. D. (1996). J. Appl. Cryst. 29, 479-480.
Daszkiewicz, M., Gulay, L. D., Shemet, V. Ya. & Pietraszko, A. (2008). Z. Anorg. Allg. Chem. 634, 1201-1204.
Delgado, G. E., Mora, A. J., Marcano, G. & Rincon, C. (2003). Mater. Res. Bull. 38, 1949-1955.
Evenson, C. R. & Dorhout, P. K. (2001). Z. Anorg. Allg. Chem. 627, 2178-2182.
Fernandez, B. J., Henao, J. A. & Pelgado, J. M. (1996). Cryst. Res. Technol. 31, 65-68.
Flack, H. D. (1983). Acta Cryst. A39, 876-881.
Oxford Diffraction (2007). CrysAlis CCD and CrysAlis RED. Versions 126.96.36.199. Oxford Diffraction Ltd, Abingdon, Oxfordshire, England.
Sharma, B. B., Ayyar, R. & Shing, H. (1977). Phys. Status Solidi A, 40, 691-697.
Sheldrick, G. M. (2008). Acta Cryst. A64, 112-122.
Spek, A. L. (2009). Acta Cryst. D65, 148-155.
Westrip, S. P. (2010). J. Appl. Cryst. 43. Submitted.
Wiberg, N. (1995). Lehrbuch der Anorganischen Chemie, pp. 1838-1841. Berlin: Walter de Gruyter.