Pentaeuropium dicadmium pentaantimonide oxide, Eu5Cd2Sb5O

The title compound, Eu5Cd2Sb5O adopts the Ba5Cd2Sb5F-type structure (Pearson symbol oC52), which contains nine crystallographically unique sites in the asymmetric unit, all on special positions. One Eu, two Sb, and the Cd atom have site symmetry m..; two other Eu, the third Sb and the O atom have site symmetry m2m; the remaining Eu atom has 2/m.. symmetry. Eu atoms fill pentagonal channels built from corner-sharing CdSb4 tetrahedra. The isolated O atom, i.e., an oxide ion O2−, is located in a distorted tetrahedral cavity formed by four Eu cations.

The title compound, Eu 5 Cd 2 Sb 5 O adopts the Ba 5 Cd 2 Sb 5 Ftype structure (Pearson symbol oC52), which contains nine crystallographically unique sites in the asymmetric unit, all on special positions. One Eu, two Sb, and the Cd atom have site symmetry m..; two other Eu, the third Sb and the O atom have site symmetry m2m; the remaining Eu atom has 2/m.. symmetry. Eu atoms fill pentagonal channels built from corner-sharing CdSb 4 tetrahedra. The isolated O atom, i.e., an oxide ion O 2À , is located in a distorted tetrahedral cavity formed by four Eu cations.

Related literature
For related ternary pnictides, see: Xia & Bobev (2007a,b, 2008a; Saparov et al. (2008aSaparov et al. ( ,b, 2010Saparov et al. ( , 2011Park & Kim (2004). For related antimonide fluorides and oxides [A 5 Cd 2 Sb 5 F (A = Sr, Ba, Eu); Ba 5 Cd 2 Sb 5 O x ], see: Saparov & Bobev (2010). For another related bismuthide oxide (Ba 2 Cd 2.13 Bi 3 O), see: Xia & Bobev (2010). For ionic and covalent radii, see: Shannon (1976);Pauling (1960 Band structure calculations highlight the importance of ionic Ba-F interations near the Fermi level to optimize bonding in Ba 5 Cd 2 Sb 5 F, but exact electron balance is achieved in the corresponding oxide Ba 5 Cd 2 Sb 5 O x only when x = 0.5 (Saparov & Bobev, 2010). Whereas the fluoride is free of disorder, the oxide exhibits underoccupancy of the oxygen site, causing positional disorder of the next-nearest Ba atoms, as revealed by elongated atomic displacement parameters on the Ba2 site (modeled as a split position) in Ba 5 Cd 2 Sb 5 O 0.59 (3) . The Ba2 atoms move away from their equilibrium positions towards the empty space that results when the oxygen site is vacant. Thus, it is surprising that the present structure of Eu 5 Cd 2 Sb 5 O contains a fully occupied oxygen site, because the formula would show a one-electron deficiency, viz. (Eu 2+ ) 5 (Cd 2+ ) 2 (Sb 3-) 3 (Sb 2-) 2 (O 2-). A possible resolution is to propose the occurrence of Eu in both +2 and +3 oxidation states. Because phasepure samples were unavailable, magnetic measurements could not be performed to verify this proposal. Nevertheless, we note that the Eu1-O distance [2.528 (4) Å] is only 0.4% longer than the Eu1-F distance in Eu 5 Cd 2 Sb 5 F (Saparov & Bobev, 2010); this increase does not scale with the difference between the ionic radii of O 2and F -, the former being nearly 5% bigger than the latter (Shannon, 1976

Experimental
The reagents were handled in an argon-filled glove box or under vacuum. All metals were with a stated purity higher than 99.9% (metal basis). They were purchased from Alfa, kept in the glove box, and were used as received.
A mixture of elemental Eu, Cd, Sb, and Pb (flux) in a molar ratio Eu:Cd:Sb:Pb = 2:1:2:10 was loaded in an alumina crucible. To prevent oxidation, the elements were weighed inside the glove box, but the mixture was accidently left in contact with ambient air for ca. 20-30 minutes, prior to sealing it under vacuum inside a silica tube. After that, the reaction mixture was put in a box-furnace and heated to 1273 K at a rate of 200 K h -1 , homogenized at this temperature for 24 supplementary materials sup-2 h, and then slowly cooled to 823 K at a rate of 5 K h -1 . After an equilibration step at 823 K for 96 h, the crystals were separated from the Pb flux.
This reaction was aimed at obtaining large single-crystals of Eu 11 Cd 6 Sb 12 (Saparov et al., 2008b), which was indeed the major product. However, alongside the needle crystals of Eu 11 Cd 6 Sb 12 , a small block-shaped crystal was also found. After the X-ray data were collected and the structure was solved, it turned out to be that of Eu 5 Cd 2 Sb 5 O. Other reactions using the same starting materials produced only the intermetallic phase, suggesting that the likely source of oxygen in this particular experiment was an unexpected partial oxidation. Attempts to increase the yield by using Eu 2 O 3 as a deliberate source of oxygen were not successful and yielded multiple phases. Reactions aimed at obtaining the isostructural Eu 5 Cd 2 Sb 5 F (Saparov & Bobev, 2010) were successful -they were performed using CdF 2 (Alfa), Eu, Cd, and Sb.

Refinement
Because the determined unit-cell dimensions and space group suggested isomorphism with Ba 5 Cd 2 Sb 5 F (Saparov & Bobev, 2010), the diffraction data were readily refined using this model. The refinements smoothly converged to low conventional residuals and a flat difference Fourier map. The maximum peak and deepest hole were located 0.86 Å from Eu2 and 0.93 Å from O, respectively.
We note that when oxygen was excluded from the model, a residual peak of about 15 e -Å -3 , located ca. 2.6-2.7 Å from Eu, remained in the difference Fourier map. Unlike the case of Ba 2 Cd 3-δ Bi 3 O (Xia & Bobev, 2010), where the residual density lacked the typical oxoanion coordination and was modeled as a partially occupied Cd site, here the tetrahedral coordination by Eu matches very well the bonding requirements of O 2-. The distances are reasonable and the oxygen site was fully occupied, as verified by freeing the site occupation factor, which led to an occupancy factor of 1.05 (2). In the final refinement cycles, all atoms were refined as fully occupied. Fig. 1

Special details
Experimental. Selected in the glove box, crystals were put in a Paratone N oil and cut to the desired dimensions. Chosen crystal was mounted on a tip of a glass fiber and quickly onto the goniometer. The crystal was kept under a cold nitrogen stream to protect from ambient conditions.
Geometry. All e.s. Refinement. Refinement of F 2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F 2 , conventional R-factors R are based on F, with F set to zero for negative F 2 . The threshold expression of F 2 > σ(F 2 ) is used only for calculating Rfactors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F 2 are statistically about twice as large as those based on F, and R-factors based on ALL data will be even larger.
Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å 2 )