Download citation
Download citation
link to html
A novel compound, vanadium aliovalent substituted zirconium tungstate, ZrW1.8V0.2O7.9, was prepared with vanadium substituting tungsten rather than the common zirconium substitution. The structure of the high-temperature phase was refined from combined neutron and X-ray powder diffraction data gathered at 530 K. This phase is the disordered centric modification (space group Pa\overline{3}) and the average crystal structure is similar to that of β-ZrW2O8. The V atom occupies only a W2 site and charge compensation is achieved through oxygen vacancy, i.e. the oxygen vacancy occurs at only the O4 site. [Atom names follow the established scheme; Evans et al. (1996). Chem. Mater. 8, 2809–2823.]

Supporting information

cif

Crystallographic Information File (CIF) https://doi.org/10.1107/S0108270109028856/fa3187sup1.cif
Contains datablocks global, I, IXRD, INPD

rtv

Rietveld powder data file (CIF format) https://doi.org/10.1107/S0108270109028856/fa3187Isup2.rtv
Contains datablock I

Comment top

The remarkable observation of isotropic negative thermal expansion (NTE) for cubic ZrW2O8 was described by Mary et al. (1996), who found that an order–disorder phase transition occurs at approximately 440 K. Below this temperature, a secondary bond along the threefold axis of the cubic unit cell links a pair of WO4 tetrahedra to form W2O8 units, in which the W—Oterminal bond of each WO4 unit in a given chain points in the same direction. Above the transition temperature, the structure becomes disordered, with the WO4 tetrahedra oriented in two opposite directions with equal probability (Fig. 1). The space group symmetry consequently gains an inversion center on the threefold axis and changes from P213 to Pa3 (Evans et al., 1999).

To improve the material properties of NTE compounds with the cubic ZrW2O8 structure, solid solutions have been prepared by substituting Zr with lower-valent ions to obtain Zr1 − xMxW2O8 − x/2 (M = Sc3+, Y3+, Yb3+, Er3+, Eu3+ and In3+). Phase transition temperatures for these solid solutions are lower due to local structural features, including the introduction of local WO4 pair orientational disorder (Yamamura et al., 2004, 2007), and an ionic conductivity increase derived from oxygen vacancy defects (Li, Han et al., 2007; Li, Xia et al., 2007). However, the work reported so far has been limited to Zr-site substitution. In the present work, we report the successful preparation of ZrW1.8V0.2O7.9, an aliovalent tungsten-substituted solid solution. The crystal structure at 530 K was analyzed using Rietveld refinement of combined X-ray diffraction (XRD) and neutron power diffraction (NPD) data.

The power of X-ray and neutron powder diffraction methods in structural studies of materials is well known. Joint refinement from NPD and XRD data, in addition to simply adding extra data points to the observables, provides two special advantages. Firstly, the neutron scattering length (analogous to the X-ray atomic scattering factor) does not fall off with sinθ/λ, significantly enhancing Debye–Waller sensitivity. Secondly, the peak shapes of NPD data are well behaved compared with those obtained using X-rays, giving a smoother refinement. The complementary nature of the two methods has been successfully utilized in the simultaneous determination of structural parameters for both O and heavy atoms (Carrio et al., 2002) in structures in which a large contrast in neutron scattering lengths exists (Rodriguez et al., 2004). NPD was previously used to distinguish V and Cr by Douglas & Anthony (1996); as vanadium is almost transparent to thermal neutrons, the combined XRD and NPD data analysis supplied information on vanadium to obtain a complete structure characterization.

In the present work, we combine XRD and NPD data to overcome the transparency of vanadium to thermal neutrons and to stabilize the refinement of O-atom structural parameters, especially the O atomic displacement parameters (ADPs). The process began the refinement with preliminary XRD data, which provided nearly full parameter information for vanadium and the heavy atoms. However, the ADPs of the O atoms were not positive definite in this refinement, so the NPD data were included in order to stabilize the O ADPs. Although the NPD data have better resolution than the XRD data, in the actual refinement the structural parameters of the heavy atoms show large fluctuations and the R(F2) value is large (0.86) when only NPD data are used. Therefore, for this structure and these data, the combination of XRD and NPD data is necessary. Equal weighting was used for both XRD and NPD data so that one type would not dominate the other.

Comparing the structure of β-ZrW2O8 with that of the vanadate ZrV2O7, we find that both structures contain AO6 octahedra and MO4 tetrahedra linked by shared corners only (Evans et al., 1998). The configuration of [VO3O1/2]2 in ZrV2O7 (Fig. 2a) is analogous to that of [W2(O1)3O3]2, except that in the case of the vanadate, a single bridge O atom on the center of symmetry completes the tetrahedral coordination of each V atom, forming a pyrovanadate group, while in β-ZrW2O8, atoms O3 are terminal, completing the W2O4 tetrahedra (Fig. 1) (Evans et al., 1996). The distance of the O3 atomic site from the symmetry center is 0.011 Å in β-ZrW2O8, so in interpreting the present structure we assume that what was the O site on the symmetry center in ZrV2O7 is here superposed with the two congeners of O3 from the W-containing structure. (We maintain the name O3 for the combined site; our O3 site is half-occupied.) The structural unit [V(O1)3O3]2 is comparable in shape with [W2(O1)3O3]2 (Fig. 2b). The disordered average structure for ZrW1.8V0.2O7.9 is thus modeled as 0.45 units of W2O8 in either of the centrosymmetrically related positions (black or gray in Fig. 1) and 0.1 units of [V(O1)3O3]2 (Fig. 2), superimposed on each other to give a composite pattern that could be illustrated by the superposition of the atomic sites represented in Figs. 1 and 2(b). The M2 site is occupied by W and V atoms with a stoichiometric ratio because they have compatible coordination numbers and ionic radii (r[WIV(6+)] = 0.42 Å and r[VIV(5+)] = 0.355 Å; Shannon, 1976). The [V(O1)3O3]2 group substitutes [W2(O1)3O3]2 tetrahedra. The resulting occupancy fractions of the O1 and O3 sites are contributed from the atoms of the [W2(O1)3(O3)]2 and [V(O1)3O3]2 (in which Obridge is split into equal parts and equivalent to O3 in the parent ZrW2O8 structure) in addition to the [W1(O1)3]2 unit, leaving a partial (0.05) O vacancy at each of the O4 sites. The site O3 in the present structure is a superposition of O3 from [W2(O1)3O3]2 and O3 from [V(O1)3O3]2. The atomic coordinates of O3 represent a mean position. The results thus describe an average crystal structure of a vanadium aliovalent substituted solid solution ZrW1.8V0.2O7.9 with the structure of β-ZrW2O8, as shown in Fig. 1, where M2 = W/V.

The final refinement plots for the XRD and NPD data are shown in Fig. 3. Total Rietveld refinement residual factors for the combined NPD and XRD data are Rp = 0.0385 and wRp = 0.0577, which are comparable with those for ZrW2O8 at 483 K (Rp = 0.0431 and wRp = 0.0569; Evans et al., 1996). The cell parameter [9.11787 (6) Å] of ZrW1.8V0.2O7.9 at 530 K is slightly smaller than that of ZrW2O8 at 483 K [9.1371 (5) Å], which is attributed to two factors: the radius of V5+ is smaller than that of W6+, and V5+ substituting a W6+ atom introduces O vacancy defects.

The refinement yields structural parameters for ZrW1.8V0.2O7.9, including ADPs and atomic coordinates, which are similar to those of β-ZrW2O8 at 483 K (Mary et al., 1996), except that the ADP of atom O4 is slightly smaller than that in the matrix. The distortion indices D (bond length; Baur, 1974) of the W1O4 and (W2/V)O4 tetrahedra are 0.042 and 0.016, respectively, while they are 0.045 and 0.013, respectively, in β-ZrW2O8. Thus V substition at the W2 site causes little distortion in the crystal structure.

Experimental top

Aqueous solutions of analytical grade ZrOCl2·8H2O, (NH4)6W7O24·6H2O and NH4VO3·6H2O in a 1:1.8:0.2 molar ratio were used as starting materials. They were mixed under continuous stirring for 3 h and then concentrated at about 350 K for 3–5 h to yield a buff powder. After grinding with an agate mortar, the powder was placed in a Teflon container, sealed in a stainless steel autoclave and heated at 473 K for 20 h with 6 M hydrochloric acid (10 ml) as an acidic steam source to prepare the hydrated precursor; this process is called the acidic steam hydrothermal (ASH) route (Guo et al., 2007). The precursor was then heated for 2 h at 873 K and then quenched in air to produce a yellow crystalline powder of the title compound. No notable mass loss occurred for this final sample.

Refinement top

The coordinates and displacement parameters reported here are the final results of the combined refinement from the XRD and NPD data. For the refinement, the initial structural model was based on the assumptions described in the Comment, and the initial cell-dimension and structural parameters were taken from β-ZrW2O8 (Evans et al., 1996). During the refinement, the atomic coordinates and anisotropic displacement parameters (ADPs) of atoms V2 and W2 were constrained to be the same. The O4 site-occupancy fraction was fixed at 0.45 to maintain charge neutrality. The occupancy fraction of each O3 in [V(O1)3O3]2 is 0.05. The ADP of O3 was also fixed to the value found for β-ZrW2O8 (Evans et al., 1996) during the combined refinement of the XRD and NPD data. In all, 21 structural parameters were refined, including a lattice parameter, seven atomic coordinates and 13 ADPs. The conditions in the parallel refinement only using the NPD data (results available as Supplementary material) are the same as those for the combined refinement of the XRD and NPD data, except that the ADP of atom O3 was refined and the atomic coordinate of atom O3 was fixed to the value found for β-ZrW2O8 (Evans et al., 1996).

Computing details top

Figures top
[Figure 1] Fig. 1. The structure of ZrW1.8V0.2O7.9, which also serves as a schematic representation of the high-temperature disordered structure of β-ZrW2O8. Displacement ellipsoids are drawn at the 75% probability level. The basic asymmetric unit and its neighbor are dark-colored, while the second disorder group is light gray. Atoms on the threefold axis (W1, M2 and O3) have an occupancy fraction of 1/2, while that of O4 is 0.45. [Symmetry codes: (i) 1 − x, 1 − y, 1 − z; (ii) z, x, y; (iii) y, z, x; (iv) 1 − y, 1 − z, 1 − x; (v) 1 − z, 1 − x, 1 − y.]
[Figure 2] Fig. 2. (a) A generic representation of the V2O7 unit in ZrV2O7 structure and (b) its breakdown for structural model building. V2O7 is divided into two equal parts [V(O1)3O3]2 (denoted with different colors) from part (a) to part (b), in order to be consistent with the [W2(O1)3O3]2 unit in ZrW2O8. The occupancy fraction of each O3 congener in [V(O1)3O3]2 is half that of V. Displacement ellipsoids are drawn at the 75% probability level.
[Figure 3] Fig. 3. Fitted (a) X-ray and (b) neutron powder diffractograms for ZrW1.8V0.2O7.9.
zirconium vanadium tungstate top
Crystal data top
ZrW1.80V0.20O7.90Dx = 4.896 Mg m3
Mr = 558.73Cu Kα; Neutron radiation
Cubic, Pa3T = 530 K
Hall symbol: -P 2ac 2ab 3Particle morphology: powder
a = 9.11787 (6) Åyellow
V = 758.02 (2) Å3flat sheet; cylinder, 2; 50 × 10; 11 mm
Z = 4
Data collection top
Philips XPert MPD
diffractometer; GPPD, IPNS, Argonne National Laboratory (USA)
2θfixed = 145.00
Scan method: step; time of flight
Refinement top
Rp = 0.039Profile function: CW Profile function number 2 with 18 terms Profile coefficients for Simpson's rule integration of pseudovoigt function (Howard, 1982; Thompson et al., 1987) #1(GU) = 19.273 #2(GV) = -63.284 #3(GW) = -36.015 #4(LX) = 8.048 #5(LY) = 13.844 #6(trns) = -5.075 #7(asym) = 3.4886 #8(shft) = 0.0000 #9(GP) = 53.393 #10(stec)= 0.00 #11(ptec)= 0.00 #12(sfec)= 0.00 #13(L11) = 0.000 #14(L22) = 0.000 #15(L33) = 0.000 #16(L12) = 0.000 #17(L13) = 0.000 #18(L23) = 0.000 Peak tails are ignored where the intensity is below 0.0001 times the peak Aniso. broadening axis 0.0 0.0 1.0; TOF Profile function number 3 with 21 terms Profile coefficients for exponential pseudovoigt convolution (Von Dreele (1990)) #1 (alp ) = 3.7446 #2 (bet-0) = 0.037328 #3 (bet-1) = 0.005382 #4 (sig-0) = 69.6 #5 (sig-1) = 53.1 #6 (sig-2) = 16.0 #7 (gam-0) = 12.94 #8 (gam-1) = 15.82 #9 (gam-2) = 0.00 #10(gsf ) = 0.00 #11(g1ec ) = 0.00 #12(g2ec ) = 0.00 #13(rstr ) = -1.641 #14(rsta ) = -1.291 #15(rsca ) = 3.050 #16(L11) = 0.000 #17(L22) = 0.000 #18(L33) = 0.000 #19(L12) = 0.000 #20(L13) = 0.000 #21(L23) = 0.000 Peak tails are ignored where the intensity is below 0.0010 times the peak Aniso. broadening axis 0.0 0.0 1.0
Rwp = 0.05851 parameters
Rexp = 0.031; 0.0180 restraints
R(F2) = 0.09763; 0.10396(Δ/σ)max = 0.03
χ2 = 6.200Background function: GSAS Background function number 6 with 11 terms. Power series in Q**2n/n! and n!/Q**2n 1: -4034.91 2: 530.394 3: 22855.3 4: -61.9106 5: -24227.6 6: 5.54910 7: 8206.00 8: -0.328951 9: -970.646 10: 9.608920E-0311: 34.4864; GSAS Background function number 6 with 7 terms. Power series in Q**2n/n! and n!/Q**2n 1: -0.316794 2: 1.040930E-02 3: 10.9721 4: -1.689380E-04 5: -54.4476 6: 1.464300E-06 7: 60.7960
6470; 8823 data points
Crystal data top
ZrW1.80V0.20O7.90Z = 4
Mr = 558.73Cu Kα; Neutron radiation
Cubic, Pa3T = 530 K
a = 9.11787 (6) Åflat sheet; cylinder, 2; 50 × 10; 11 mm
V = 758.02 (2) Å3
Data collection top
Philips XPert MPD
diffractometer; GPPD, IPNS, Argonne National Laboratory (USA)
2θfixed = 145.00
Scan method: step; time of flight
Refinement top
Rp = 0.039χ2 = 6.200
Rwp = 0.0586470; 8823 data points
Rexp = 0.031; 0.01851 parameters
R(F2) = 0.09763; 0.103960 restraints
Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) top
xyzUiso*/UeqOcc. (<1)
Zr0.00.00.00.0116 (5)*
W10.33847 (17)0.33847 (17)0.33847 (17)0.017410.45
W20.39543 (14)0.39543 (14)0.39543 (14)0.009290.45
V20.39543 (14)0.39543 (14)0.39543 (14)0.009290.1
O10.20745 (14)0.43148 (18)0.4460 (2)0.03189
O30.4954 (8)0.4954 (8)0.4954 (8)0.0440.5
O40.2321 (3)0.2321 (3)0.2321 (3)0.05030.45
Atomic displacement parameters (Å2) top
U11U22U33U12U13U23
W10.0174 (8)0.0174 (8)0.0174 (8)0.0073 (9)0.0073 (9)0.0073 (9)
W20.0093 (8)0.0093 (8)0.0093 (8)0.0023 (8)0.0023 (8)0.0023 (8)
V20.0093 (8)0.0093 (8)0.0093 (8)0.0023 (8)0.0023 (8)0.0023 (8)
O10.0129 (7)0.0401 (11)0.0427 (12)0.0136 (7)0.0131 (7)0.0037 (7)
O30.0440.0440.0440.012790.012790.01279
O40.0503 (17)0.0503 (17)0.0503 (17)0.0216 (17)0.0216 (17)0.0216 (17)
Geometric parameters (Å, º) top
Zr—O1i2.0520 (12)W2—O11.8050 (13)
Zr—O1ii2.0520 (12)W2—O1iv1.8050 (13)
Zr—O1iii2.0520 (12)W2—O1v1.8050 (13)
Zr—O12.0520 (12)W2—O31.723 (13)
Zr—O12.0520 (12)V2—O11.8050 (13)
Zr—O12.0520 (12)V2—O1iv1.8050 (13)
W1—O11.7627 (13)V2—O1v1.8050 (13)
W1—O1iv1.7627 (13)V2—O31.723 (13)
W1—O1v1.7627 (13)W1—O32.623 (13)
W1—O41.681 (6)
O1i—Zr—O1ii90.79 (8)W1—W2—O1v72.80 (8)
O1i—Zr—O1iii90.79 (8)O1—W2—O1iv111.64 (7)
O1i—Zr—O189.21 (8)O1—W2—O1v111.64 (7)
O1i—Zr—O189.21 (8)O1—W2—O3107.20 (8)
O1ii—Zr—O1iii90.79 (8)O1—W2—O3107.20 (8)
O1ii—Zr—O189.21 (8)O1iv—W2—O1v111.64 (7)
O1ii—Zr—O189.21 (8)O1iv—W2—O3107.20 (8)
O1iii—Zr—O189.21 (8)O1iv—W2—O3107.20 (8)
O1iii—Zr—O189.21 (8)O1v—W2—O3107.20 (8)
O1—Zr—O190.79 (8)O1v—W2—O3107.20 (8)
O1—Zr—O190.79 (8)W1—V2—O172.80 (8)
O1—Zr—O190.79 (8)W1—V2—O1iv72.80 (8)
W2—W1—O178.02 (10)W1—V2—O1v72.80 (8)
W2—W1—O1iv78.02 (10)O1—V2—O1iv111.64 (7)
W2—W1—O1v78.02 (10)O1—V2—O1v111.64 (7)
V2—W1—O178.02 (10)O1—V2—O3107.20 (8)
V2—W1—O1iv78.02 (10)O1—V2—O3107.20 (8)
V2—W1—O1v78.02 (10)O1iv—V2—O1v111.64 (7)
O1—W1—O1iv115.81 (7)O1iv—V2—O3107.20 (8)
O1—W1—O1v115.81 (7)O1iv—V2—O3107.20 (8)
O1—W1—O4101.98 (10)O1v—V2—O3107.20 (8)
O1iv—W1—O1v115.81 (7)O1v—V2—O3107.20 (8)
O1iv—W1—O4101.98 (10)Zrvi—O1—W1154.78 (12)
O1v—W1—O4101.98 (10)Zrvi—O1—W2172.75 (11)
W1—W2—O172.80 (8)Zrvi—O1—V2172.75 (11)
W1—W2—O1iv72.80 (8)
Symmetry codes: (i) y1/2, z+1/2, x; (ii) x, y1/2, z+1/2; (iii) z+1/2, x, y1/2; (iv) z, x, y; (v) y, z, x; (vi) y, z+1/2, x+1/2.

Experimental details

Crystal data
Chemical formulaZrW1.80V0.20O7.90
Mr558.73
Crystal system, space groupCubic, Pa3
Temperature (K)530
a (Å)9.11787 (6)
V3)758.02 (2)
Z4
Radiation typeCu Kα; Neutron
Specimen shape, size (mm)Flat sheet; Cylinder, 2; 50 × 10; 11
Data collection
Data collection methodPhilips XPert MPD
diffractometer; GPPD, IPNS, Argonne National Laboratory (USA)
Specimen mounting
Data collection modeReflection
Scan methodStep; Time of flight
2θ values (°)2θfixed = 145.00
Refinement
R factors and goodness of fitRp = 0.039, Rwp = 0.058, Rexp = 0.031; 0.018, R(F2) = 0.09763; 0.10396, χ2 = 6.200
No. of data points6470; 8823
No. of parameters51

Computer programs: GSAS (Larson & Von Dreele, 2000), DIAMOND (Brandenburg, 2006).

 

Follow Acta Cryst. C
Sign up for e-alerts
Follow Acta Cryst. on Twitter
Follow us on facebook
Sign up for RSS feeds