DALTON FULL PAPER J. Chem. Soc., Dalton Trans., 1998, Pages 1947–1949 1947 Antimony(III) tellurium(IV) chloride trioxide SbTeO3Cl: synthesis and ab initio structure determination from X-ray and neutron powder diVraction data José Antonio Alonso* Instituto de Ciencia de Materiales de Madrid, C.S.I.C., Cantoblanco, 28049 Madrid, Spain The mixed chloride oxide SbTeO3Cl has been prepared in polycrystalline form by solvolysis reaction from SbCl3 and TeCl4. The X-ray and neutron powder diVraction patterns were indexed in an orthorhombic unit cell; the structure was partially solved in the space group Pnma from the X-ray data, and completed and refined from neutron data.The crystal consists of strongly covalent [SbTeO3]1 layers, perpendicular to the a axis, corrugated as a consequence of the electrostatic repulsion of the lone pairs of SbIII and TeIV. Both semimetals are three-fold coordinated to oxygens in pyramidal configuration. The layers are held together by discrete Cl2 anions.This is one of the few examples of a positively charged two-dimensional network, very uncommon in the crystal chemistry of inorganic oxo compounds. Antimony and tellurium are typical examples of semimetallic elements for which the isoelectronic character in the oxidation states SbIII and TeIV, similarities of ionic radii and coordination features (determined by the presence of an electron lone pair) allow one to expect the formation of mixed oxo compounds of both elements, adopting closely related environments in the crystal structure.In a previous work1 the preparation and structure of the first example of a mixed chloride oxide of both semimetallic elements, Sb3TeO6Cl, obtained in single-crystal form, were described. In the layered network of composition (Sb3TeO6 1)n the positions of antimony(III) and tellurium( IV) cations were indistinguishable by X-ray diVraction, hence they were assumed to be distributed at random over the semimetal positions. Later, I reported on the existence of three new halide oxides of Sb and Te, of compositions SbTeO3Cl, Sb3Te2O7Cl3 and SbTeO3Br, for which the X-ray diVraction patterns, IR spectra and thermal behaviour were described.2 Attempts to grow single crystals failed, but, recently, I optimized the preparation procedure of SbTeO3Cl, obtaining a well crystallized polycrystalline material. The development of X-ray (XRD) and neutron powder diVraction (NPD) techniques for ab initio crystal structure determination encouraged me to investigate the structure of SbTeO3Cl from powder diVraction data.This paper reports on the results of this study. Results and Discussion The compound SbTeO3Cl was prepared as a white microcrystalline powder by solvolysis from SbCl3 and TeCl4 at 70 8C (see Experimental section), according to equation (1). After an age- SbCl3 1 TeCl4 1 3H2O 70 8C SbTeO3Cl 1 6HCl (1) ing period of 3 d, powder SbTeO3Cl samples exhibited an excellent crystallinity despite the relatively low reaction temperature, as seen from the X-ray powder diagrams (see Fig. 1). Since the X-ray diVraction (XRD) scattering factors of Sb and Te are considerably higher than that of chlorine and, particularly, oxygen, neutron diVraction techniques were necessary to complement the XRD data. The first 20 peaks of the XRD diagram of SbTeO3Cl were unambiguously indexed in an * E-Mail: jalonso@fresno.csic.es orthorhombic unit cell (de WolV figure of merit M20 = 20).Given the observed density, the unit cell contains four formula units. The observed reflection conditions suggested the space groups Pna21 (no. 33) or Pnma (no. 62). The latter, centric, was considered to solve the structure. For this crystal symmetry the matching of the observed and calculated XRD profiles without including a structural model led to excellent residuals. The unitcell parameters after the pattern matching of the XRD data were a = 11.1935(2), b = 5.4281(1), c = 7.2401(1) Å.The matching allowed the precise integration of the diVraction peaks. A set of structure factors in the range 2q 10 to 608 was used to solve the structure. A Patterson map followed by Fourier synthesis allowed the localization of Sb, Te, Cl and O(1) atoms from the XRD data. A first Rietveld refinement of the neutron data at this stage led to discrepancy factors of RI = 0.207. Atom O(2) was located from Fourier synthesis on the NPD data.After the refinement of the anisotropic thermal factors for Cl the RI factor dropped to 0.064. Given the relatively poor resolution of the neutron diVractometer at high angles, a further fully anisotropic refinement was not considered reliable. The agreement between the observed and calculated NPD profiles is shown in Fig. 2. In order to check the consistency of the neutron-derived structure with the X-ray data, a subsequent refinement of the XRD pattern was performed.A significant preferred orientation eVect was observed, which could be minimized in the Fig. 1 The XRD diagram for SbTeO3Cl, indexed according to an orthorhombic unit cell1948 J. Chem. Soc., Dalton Trans., 1998, Pages 1947–1949 refinement by considering platey-habit crystallites normal to the [0 1 0] direction, leading to acceptable discrepancy factors (Rp = 0.131, Rwp = 0.170, c2 = 5.21, RI = 0.0954). However, no improvement in the atomic coordinates (as far as standard deviations are concerned) was observed after the XRD refinement; therefore the final description of the crystal structure refers only to the neutron-derived coordinates.Table 1 includes the final atomic coordinates and thermal factors obtained from NPD data, and Table 2 lists the main interatomic distances and angles. Two views of the crystal structure are shown in Fig. 3. It is constituted by SbO3 and TeO3 trigonal pyramids (see Fig. 4), the apices of which are occupied by the semimetal atoms.Bonding distances are in the range 1.94–2.00 Å for Sb]O and 1.88–1.93 Å for Te]O. Both kinds of polyhedra are linked together via common oxygens, giving rise to corrugated layers of composition [SbTeO3]1, parallel to the bc plane. The strongly covalent two-dimensional networks are positively charged. These layers are held together by Cl2 anions, which are located between the layers at relatively long distances from metal atoms: 3.10 Å for Sb]Cl bonds, 3.16 Å for Te]Cl bonds.The very irregular oxygen environment around SbIII and TeIV shown in Fig. 4 is due to the electrical repulsion of the electron Fig. 2 Observed (crosses), calculated (full line) and diVerence (below) neutron powder diVraction profiles for SbTeO3Cl at 295 K. The short vertical lines indicate the allowed Bragg positions Table 1 Physical and crystallographic data and parameters for neutron powder data collection and refinement Formula M, Dc/g cm23 Space group, Z Crystal symmetry a/Å b/Å c/Å U/Å3 2q Range, step/8 Collection time/h, sample weight/g No.reflections Refined positional parameters Rwp, Rp, Rexp, RI ClO3SbTe 332.8, 5.02 Pnma (no. 62), 2 Orthorhombic 11.197(2) 5.427(1) 7.239(1) 439.93(1) 10.0–89.9, 0.01 4, 4 302 11 0.031, 0.024, 0.020, 0.064 Atom Sba Te a O(1) O(2) Cl Site 4c 4c 8d 4c 4c x 0.2485(6) 0.0718(7) 0.3729(4) 0.7187(7) 0.9226(5) y 0.25 0.25 0.9985(10) 0.25 0.25 z 0.5718(14) 0.1820(10) 0.5113(7) 0.1918(12) 0.5950(7) Beq/Å2 1.3(2) 0.8(2) 0.9(1) 0.9(2) 3.0(2) b a Better labelled as Sb-rich sites and Te-rich sites, given the possibility of partial mixed occupancy between antimony(III) and tellurium(IV) cations.b Anisotropic thermal factors for Cl, Uij (Å2, ×103): U11 = 44(3), U22 = 49(3), U33 = 22(3), U12 = U23 = 0, U13 = 13(1). lone pair, which is thought physically to occupy a volume similar to that of an oxygen anion.3 Considering the lone pair as a sterically significant sphere, the average volume per anion in SbTeO3ClE2 (E = electron lone pair of both SbIII and TeIV) is 18.3 Å3, which compares with the corresponding values of other oxides and chloride oxides of Sb and Te: 16.5 Å3 for Sb2Te2O9,4 22.8 Å3 for Sb3TeO6Cl,1 19.5 Å3 for Sb4O5Cl2 5 and 16.2 Å3 for Te6O11Cl2.6 Fig. 3 Two representations of the SbTeO3Cl structure: (a) along [0 1 0], showing the puckered [SbTeO3]1 layers perpendicular to the a axis; (b) view of one single layer, along the [1 0 0] direction.Key: Sb and Te, dark and light green spheres, respectively; O and Cl, red and orange balls, respectively. The c axis is oriented from left to right ( a) ( b) Fig. 4 Oxygen co-ordination polyhedra about Sb and Te: the electron lone pair physically occupies vacant sites in the structure, in the neighbourhood of each (Sb,Te) atom, as suggested Sb TeJ. Chem. Soc., Dalton Trans., 1998, Pages 1947–1949 1949 Distinction between the positions of SbIII and TeIV in the crystal is possible, in spite of the isoelectronic character of both elements and their similar fermi lengths (see Experimental section), given the significantly shorter observed Te]O than Sb]O bond lengths.In relation to this, the calculation of the bond valences 7 for the antimony and tellurium co-ordination polyhedra is enlightening. As shown in Table 3, bond valence sums for the co-ordination polyhedra of Sb and Te are close to the expected valences of 3 and 4, respectively.However, a partial mixed occupancy of the antimony and tellurium sites should not be discarded, given the size of the thermal motion. It is possible that each site is significantly contaminated with the other ion, leading to the observed valences, slightly higher than 31 for Sb and lower than 41 for Te. An average mixed occupancy of about 20% can be estimated from the valence deviations. The contribution of the bonds to chlorine to the total valence of the semimetals is very small, suggesting that the (Sb,Te)]Cl interactions are weak and predominantly ionic.Examples of positively charged networks are relatively rare in the crystal chemistry of inorganic oxo compounds, considering the large number of known examples of negatively charged networks. With the isolated exception of some crystal structures, like that of [Te2O4H]1[NO3]2,8 in most of the complex compounds of the heavier p elements of the Groups 5–7, which have been proposed to be formed by positively charged one- or two-dimensional networks, either 8 a discrete anion does not exist or the anion forms covalent bonds which complete the primary co-ordination polyhedra of the semimetal atoms in the network.In two-dimensional SbTeO3Cl, puckered layers of [SbTeO3]1 are stacked in a direction perpendicular to the a axis, with discrete Cl2 anions localized between the layers. Thus, SbTeO3Cl is one of the few oxo compounds exhibiting a positively charged two-dimensional network structure.Experimental The compound SbTeO3Cl was prepared from an equimolar mixture of SbCl3 and TeCl4 (total weight of 5 g) which was hydrolysed by addition of a 2 M HCl aqueous solution (100 cm3). The initially amorphous precipitate was digested in the reaction media at 70 8C, with stirring, for 3 d. This procedure led to a microcrystalline material which was thoroughly washed with water and dried at 120 8C in air. The chemical analysis of SbTeO3Cl was performed as fol- Table 2 Main interatomic distances (Å) and angles (8) Sb]O(1) Sb]O(2) Sb]Cl Te]O(1) Te]O(2) Te]Cl 1.998(7) ×2 1.937(13) 3.101(10) 1.932(8) ×2 1.881(11) 3.158(5) O(1)]Sb]O(1) O(1)]Sb]O(2) O(1)]Te]O(1) O(1)]Te]O(2) 86.1(4) 84.5(5) ×2 81.6(4) 91.8(6) ×2 Table 3 Bond valences * (si) for (Sb,Te)](O,Cl) bonds, multiplicity of the bonds [m] and valences (Ssi) for antimony and tellurium cations within the respective co-ordination polyhedra in the SbTeO3Cl structure si [ m] Atom Sb Te O(1) 0.93 [2] 1.13 [2] O(2) 1.10 [1] 1.30 [1] Cl 0.13 [1] 0.12 [1] Ssi 3.09 3.68 * Bond valences are calculated as si = exp[(r0 2 ri)/B]; B = 0.37; r0(SbIII]O) = 1.973, r0(SbIII]Cl) = 2.35, r0(TeIV]O) = 1.977, r0(TeIV]Cl) = 2.37 Å (from ref. 7). Individual distances (ri) are taken from Table 2. lows: 0.1 g was dissolved in concentrated HCl, then diluted to a HCl concentration of about 3 M. The antimony content (quantitatively present as SbIII) was determined by titration with KBrO3.In the resulting solution the tellurium content was determined gravimetrically, by reduction to Te metal with an excess of Na2SO3 and NH2NH2?2HCl. The chlorine content was obtained gravimetrically as AgCl: a new portion of SbTeO3Cl (0.1 g) was dissolved in an aqueous 3 M solution of KOH, then HNO3 was added until neutral pH, and the Cl2 anions from the chloride oxide were precipitated with AgNO3 [Found: Cl, 11.0; O (by diVerence), 13.7; Sb, 37.1; Te, 38.2. SbTeO3Cl requires Cl, 10.66; O, 14.42; Sb, 36.58; Te, 38.34%].The density was determined by immersion in CCl4, Dobs = 5.01(5) g cm23. The XRD patterns were collected with Cu-Ka radiation in a Siemens D-501 goniometer controlled by a DACO-MP computer, by step-scanning from 2q 10 to 1008, in increments of 0.058, and a counting time of 4 s each step. The NPD diagram of SbTeO3Cl was collected at room temperature in the multidetector DN5 diVractometer at the Siloé reactor of the Centre d’Etudes Nucléaires, Grenoble (l = 1.344 Å).The conditions of the data collection are summarized in Table 1. The XRD pattern was indexed with the TREOR 4 program.9 The structure was solved from a Patterson map from XRD data (SHELXS 86 program10) and subsequently completed and refined from NPD data. The profile refinements (pattern matching for XRD data and structural refinement for neutron data) were performed with the FULLPROF program.11 A pseudo- Voigt function was chosen to generate the line shape of the diVraction peaks.The coherent neutron scattering lengths for Sb, Te, O and Cl were, respectively, 5.57, 5.80, 5.803 and 9.577 fm. In the final run the following parameters were refined: background coeYcients, zeropoint, half-width, pseudo-Voigt and asymmetry parameters for the peak shape; scale factor, positional and thermal isotropic factors (anisotropic for Cl) and unit-cell parameters. The maximum shift for atomic coordinates in the final refinement cycle of the neutron data was lower than 1024.Acknowledgements This work was supported by the Spanish Dirección General de Investigación Cientifica y Técnica, under the project PB94-0046. The author thanks the MDN group at the CEN-Grenoble for their hospitality and the facilities at the Siloé reactor. References 1 J. A. Alonso, E. Gutiérrez-Puebla, A. Jerez, A. Monge and C. Ruiz- Valero, J. Chem. Soc., Dalton Trans., 1985, 1633. 2 J. A. Alonso, Ph.D. Thesis, University Complutense of Madrid, 1987. 3 J. Galy, G. Meunier, S. Anderson and A. Astrom, J. Solid State Chem., 1975, 13, 142. 4 J. A. Alonso, A. Castro, R. Enjalbert, J. Galy and I. Rasines, J. Chem. Soc., Dalton Trans., 1992, 2551. 5 Ch. Särnstrand, Acta Crystallogr., Sect. B, 1978, 34, 2402. 6 G. Giester, Acta Crystallogr., Sect. B, 1994, 50, 3. 7 N. E. Brese and M. O’Keefe, Acta Crystallogr., Sect. B, 1991, 47, 192. 8 J. B. Anderson, M. H. Paposch, C. P. Anderson and E. Kostiner, Monatsh. Chem., 1980, 11, 798. 9 P. E. Werner, L. Eriksson and M. Westdahl, J. Appl. Crystallogr., 1985, 16, 367. 10 G. M. Sheldrick, in Crystallographic Computing 3, eds. G. M. Sheldrick, C. Kruger and R. Goggard, Oxford University Press, Oxford, 1985, p. 175. 11 J. Rodríguez-Carvajal, FULLPROF, version 3.1, Institut Laue Langevin, 1995. Received 29th January 1998; Paper 8/00799C