[Journal logo]

Volume 66 
Part 12 
Pages i99-i102  
December 2010  

Received 12 May 2010
Accepted 9 November 2010
Online 19 November 2010

The structural phase transition in SrV6O11

aDepartment of Applied Physics, National Defense Academy, 1-10-20 Hashirimizu, Yokosuka, Kanagawa 239-8686, Japan,bAdvanced Nano Materials Laboratory, National Institute for Materials Science, 1-1 Namiki, Tsukuba 305-0044, Japan, and cDepartment of Applied Physics, University of Tsukuba, Ibaraki 305-8537, Japan
Correspondence e-mail: hata@nda.ac.jp

Single-crystal X-ray diffraction and specific heat studies establish that strontium hexavanadium undecaoxide, SrV6O11, undergoes a P63/mmc to inversion twinned P63mc structural transition as the temperature is lowered through 322 K. The P63/mmc and P63mc structures have been determined at 353 K and at room temperature, respectively. For the room-temperature structure, seven of the ten unique atoms lie on special positions, and for the 353 K structure all of the seven unique atoms sit on special positions. The P63/mmc to P63mc structural phase transition, accompanied by a magnetic transition, is a common characteristic of AV6O11 compounds, independent of the identity of the A cations.

Comment

A series of AV6O11 compounds (A = Na, K, Sr, Ba, Pb; de Roy et al., 1987[Roy, M. E. de, Besse, J. P., Chevalier, R. & Gasperin, M. (1987). J. Solid State Chem. 67, 185-189.]; Kanke, 1999[Kanke, Y. (1999). Phys. Rev. B, 60, 3764-3776.]; Kanke et al., 1991[Kanke, Y., Izumi, F., Takayama-Muromachi, E., Kato, K., Kamiyama, T. & Asano, H. (1991). J. Solid State Chem. 92, 261-272.]; Friese et al., 2006[Friese, K. & Kanke, Y. (2006). J. Solid State Chem. 179, 3277-3285.]; Mentre et al., 1996[Mentre, O. & Abraham, F. (1996). J. Solid State Chem. 125, 91-101.]) have generated interest because of their structural phase transitions, magnetic and transport properties (Kanke et al., 1990[Kanke, Y., Takayama-Muromachi, E., Kato, K. & Matsui, Y. (1990). J. Solid State Chem. 89, 130-137.], 1994[Kanke, Y., Izumi, F., Morii, Y., Akiba, E., Funahashi, S., Kato, K., Isobe, M., Takayama-Muromachi, E. & Uchida, Y. (1994). J. Solid State Chem. 112, 429-437.]; Uchida et al., 1991[Uchida, Y., Kanke, Y., Takayama-Muromachi, E. & Kato, K. (1991). J. Phys. Soc. Jpn, 60, 2530-2533.], 2001[Uchida, Y., Onoda, Y. & Kanke, Y. (2001). J. Magn. Magn. Mater. 226, 446-448.]; Mentre et al., 2001[Mentre, O., Kanke, Y., Dhaussy, A.-C., Conflant, P., Hata, Y. & Kita, E. (2001). Phys. Rev. B, 64, 174404.]). The crystal structures consist of hexagonal close-packed layers of A and O atoms, and three types of V atoms (Fig. 1[link]). The V1O6 octahedra form a Kagomé lattice by edge sharing. The V2O6 octahedra form face-sharing dimers between the layers of the Kagomé lattice. The coordination of V3O5 is a trigonal bipyramid.

While the AV6O11 compounds show common characteristics in their paramagnetic states, they exhibit individual characteristics for their magnetically ordered states. In the paramagnetic states, each AV6O11 shows one phase transition at a characteristic temperature, Tt. Above Tt, the compounds crystallize in the centrosymmetric hexagonal space group P63/mmc, and show Curie-Weiss paramagnetism. Below Tt, the compounds lose the centre of symmetry and show second-order transitions to hexagonal P63mc (Kanke et al., 1994[Kanke, Y., Izumi, F., Morii, Y., Akiba, E., Funahashi, S., Kato, K., Isobe, M., Takayama-Muromachi, E. & Uchida, Y. (1994). J. Solid State Chem. 112, 429-437.]). The V1O6 octahedron forms a regular Kagomé lattice above Tt. It distorts to form a V1O6 trimer with a regular triangular shape below Tt. Two V2O6 octahedra forming a face-sharing dimer are crystallographically equivalent above Tt. Below Tt, they become inequivalent. V3 is no longer at the centre of the V3O5 polyhedron below Tt. In the P63mc form of AV6O11, V1 exhibits a spin gap behaviour with a spin-singlet ground state, while V2 and V3 possess magnetic moments (Uchida et al., 2001[Uchida, Y., Onoda, Y. & Kanke, Y. (2001). J. Magn. Magn. Mater. 226, 446-448.]). Tt values for KV6O11 (Kanke, 1999[Kanke, Y. (1999). Phys. Rev. B, 60, 3764-3776.]), BaV6O11 (Friese et al., 2006[Friese, K. & Kanke, Y. (2006). J. Solid State Chem. 179, 3277-3285.]) and PbV6O11 (Kato et al., 2001[Kato, H., Kato, M., Yoshimura, K. & Kosuge, K. (2001). J. Phys. Condens. Matter, 13, 9311-9333.]) are 190, 250 and 560 K, respectively.

As first reported, the room-temperature structure of SrV6O11 was assigned to P63/mmc with a relatively high R factor (Kanke et al., 1992[Kanke, Y., Kato, K., Takayama-Muromachi, E. & Isobe, M. (1992). Acta Cryst. C48, 1376-1380.]). The Tt = 320 K of SrV6O11 exceeds room temperature, which suggests an incorrect assignment of the space group. We have pointed out briefly the existence of the P63/mmc to P63mc transition at 320 K (Hata et al., 1999[Hata, Y., Kanke, Y., Kita, E., Suzuki, H. & Kido, G. (1999). J. Appl. Phys. 85, 4768-4770.]). Kato et al. studied the transition by X-ray powder diffraction. They refined the structures at 100 K in P63mc and at 623 K in P63/mmc (Kato et al., 2001[Kato, H., Kato, M., Yoshimura, K. & Kosuge, K. (2001). J. Phys. Condens. Matter, 13, 9311-9333.]). However, we considered a single-crystal diffraction study to be indispensable for an accurate structural characterization of the acentric phase. In the present study, the crystal structure of SrV6O11 was determined in detail both above Tt (353 K) and below Tt (room temperature, RT) by single-crystal X-ray diffraction. The transition temperature was determined precisely by a specific heat study.

At both room temperature and 353 K, diffraction data showed hexagonal symmetry and an extinction rule, l [not equal to] 2n absent for hhl, indicating possible space groups P63/mmc, P[\overline{6}]2c and P63mc. P63/mmc is centrosymmetric and gives a unique structural model. But the other two are acentric, and each gives a pair of single-domain models, (x, y, z) and (-x, -y, -z), and one inversion twin model, [(x, y, z) + (-x, -y, -z)].

To examine the possible models, reflections with h [greater-than or equal to] 0, k [greater-than or equal to] 0, l [greater-than or equal to] 0, |h| [less-than or equal to] |k|, 2[theta] < 90°, and those with h [less-than or equal to] 0, k [less-than or equal to] 0, l [less-than or equal to] -1, |h| [less-than or equal to] |k|, 2[theta] [less-than or equal to] 90° were collected at both temperatures using a four-circle diffractometer (Enraf-Nonius CAD-4) with Mo K[alpha] radiation. As the diffractometer is equipped with a scintillation counter, reflections that were too weak were regarded as unobserved. Thus, 185 of a total of 1284 reflections for room temperature, and 349 of 1290 reflections for 353 K were assigned as unobserved. The refinements did not use the unobserved data. The data, however, are of high resolution, 0.5 Å, and have what we consider sufficient completeness; for 2[theta] < 50°, the completeness is 97.4% for room temperature and 92.9% for 353 K, even without the unobserved data. As a result, the model examinations give clear results as follows.

The models for room temperature were examined using 1093 Friedel-unaveraged reflections with I > 1.5[sigma](I), applying the common weighting scheme of 1/[sigma](I). As shown in Table 1[link], the inversion twin P63mc model gave trouble-free convergence and low enough R factors (R1 = 0.0231, wR2 = 0.0515, 44 parameters). However, all of the remaining models resulted in nonpositive definite atomic displacement parameter(s) and significantly higher R factors. Consequently, the twinned P63mc model was selected. This clear differentiation of the results for the different models also indicates high enough quality of the specimen and high enough resolution of the diffraction data.

The models for 353 K were examined using 919 unaveraged reflections with I > 1.5[sigma](I), applying the common weighting scheme of 1/[sigma](I). All seven models gave trouble-free convergence and low enough R factors (R1 = 0.0236-0.0242, wR2 = 0.0511-0.0526, 26-44 parameters). Consequently, there was no reason to choose any of the acentric space groups, and the P63/mmc model, with the highest symmetry, was selected.

The room-temperature structure of SrV6O11 had earlier been described by Kanke et al. (1992[Kanke, Y., Kato, K., Takayama-Muromachi, E. & Isobe, M. (1992). Acta Cryst. C48, 1376-1380.]) using P63/mmc. They compared the P63/mmc model, two single-domain models of P[\overline{6}]2c and two single-domain models of P63mc, using 2031 unaveraged intensities. The P63/mmc model, the better P[\overline{6}]2c model and the better P63mc model gave R = 0.070 with 24 parameters, R = 0.069 with 24 parameters and R = 0.064 with 35 parameters, respectively. The difference in the R factors was concluded to be insignificant, considering the numbers of parameters. Consequently, they chose the P63/mmc model with the highest symmetry. However, the study did not examine the twin models for the two acentric space groups. None of the models was free from negative temperature factors, and anisotropic displacement parameters were not applied to O2 even in the final refinement. In the present study, in fact only the P63mc model with inversion twinning is free of nonpositive definite displacement parameters, and this model clearly gives the best result among the seven models tested. We thus correct the previous report with this determination that SrV6O11 crystallizes in P63mc with inversion twinning at room temperature.

Between the two temperatures, viz. room temperature and 353 K, the specific heat of SrV6O11 shows only one anomaly, at 322 K. Consequently, we conclude that the structural phase transition takes place at 322 K and coincides with the magnetic transition.

The structural refinements of the P63mc forms of NaV6O11 (Kanke et al., 1994[Kanke, Y., Izumi, F., Morii, Y., Akiba, E., Funahashi, S., Kato, K., Isobe, M., Takayama-Muromachi, E. & Uchida, Y. (1994). J. Solid State Chem. 112, 429-437.]) and PbV6O11 (Mentre et al., 1996[Mentre, O. & Abraham, F. (1996). J. Solid State Chem. 125, 91-101.]) converged promptly without applying twinning. Kanke (1999[Kanke, Y. (1999). Phys. Rev. B, 60, 3764-3776.]) suggested that this may be due to the small anomalous dispersion term for Na for Mo K[alpha] in NaV6O11 or may suggest that the volume fraction ratio (x, y, z) / (x, y, -z) is far from 1.0 in PbV6O11 and/or in NaV6O11.

The P63mc and P63/mmc forms of SrV6O11 are illustrated in Figs. 1[link] and 2[link], respectively. In both forms, the V1 ellipsoid is elongated towards the centre of the V1 trimer. V2 is nearly isotropic in the P63/mmc form. In the P63mc form, though, V21 is oblate, compressed into (001) and V22 is prolate along [001]. V3 shows extended displacement along [001] in the P63/mmc form, but is rather isotropic in the P63mc form.

For the P63mc form, seven of the ten unique atoms lie on special positions, and for the P63/mmc form, all of the seven unique atoms sit on special positions. In the centric structure, the V1O6 octahedra form a regular Kagomé lattice with a uniform V1...V1 distance of 2.8887 (1) Å (Table 2[link]). In the P63mc phase, though, the Kagomé lattice distorts, with the V1O6 octahedra forming a trimer with a regular triangular shape; and the V1...V1 distance separates into two types, viz. inter-trimer [2.9736 (6) Å] and intra-trimer [2.7966 (6) Å] (Table 3[link]). It is noteworthy that SrV6O11 shows a much smaller change in the V2...V2 distance with the phase transition, as compared to the other AV6O11 compounds. In both the P63/mmc and P63mc forms, analogous V1-O distances show similar values, independent of the nature of A. On the other hand, in both forms the V2-O and the V3-O2 distances tend to be longer for divalent A cations and shorter for monovalent A.

The P63/mmc to P63mc structural phase transition, accompanied by a magnetic transition, is a common characteristic of AV6O11 compounds, independent of the A cations. The acentric form of SrV6O11 shows features in common with the corresponding P63mc forms of AV6O11 (A = Na, K, Sr, Ba, Pb). Below Tt, the V1O6 octahedron no longer forms a regular Kagomé lattice, but distorts to form a V1O6 trimer with a regular triangular shape. A pair of the V2O6 octahedra forming a face-sharing dimer at higher temperatures become inequivalent. V3 moves away from the centre of the V3O5 polyhedron. The V1 trimer formation accompanying the structural transition is considered to be the factor that suppresses the paramagnetism below Tt.

[Figure 1]
Figure 1
The crystal structure of SrV6O11 at room temperature, shown with 99% probability displacement ellipsoids.
[Figure 2]
Figure 2
The crystal structure of SrV6O11 at 353 K, shown with 99% probability displacement ellipsoids.

Experimental

Sr2V2O7 and V2O3 were mixed in a 1:5 molar ratio. The mixture was sealed in a platinum capsule and heated at 1073 K for 1 d and at 1473 K for 2 weeks, successively. Crystals of SrV6O11 were hexagonal plates with principal faces {001}. Sizes were typically 0.2 mm across the plate and 0.1 mm in thickness.

Data collection at 353 K was achieved by blowing hot nitrogen gas onto the specimen. The temperature was calibrated by a chromel-alumel thermocouple with a water-ice standard. The specific heat of single-crystal SrV6O11 was measured using a Quantum Design Physical Property Measurement System (PPMS). The temperature range of the measurements was from 2.4 to 350 K.

SrV6O11 at 353 K

Crystal data
  • SrV6O11

  • Mr = 569.26

  • Hexagonal, P 63 /m m c

  • a = 5.7773 (1) Å

  • c = 13.0852 (3) Å

  • V = 378.234 (13) Å3

  • Z = 2

  • Mo K[alpha] radiation

  • [mu] = 14.15 mm-1

  • T = 353 K

  • 0.31 × 0.14 × 0.08 mm

Data collection
  • Enraf-Nonius CAD-4 diffractometer

  • Absorption correction: Gaussian (SDP; B. A. Frenz & Associates Inc., 1985[B. A. Frenz & Associates Inc. (1985). SDP. Structure Determination Package, 4th ed. Enraf-Nonius, Delft, The Netherlands.]) Tmin = 0.413, Tmax = 0.532

  • 1290 measured reflections

  • 645 independent reflections

  • 447 reflections with I > 1.5[sigma](I)

  • Rint = 0.016

  • 3 standard reflections every 240 min intensity decay: 0.7%

Refinement
  • R[F2 > 2[sigma](F2)] = 0.021

  • wR(F2) = 0.054

  • S = 1.37

  • 447 reflections

  • 26 parameters

  • [Delta][rho]max = 0.84 e Å-3

  • [Delta][rho]min = -0.81 e Å-3

SrV6O11 at room temperature

Crystal data
  • SrV6O11

  • Mr = 569.26

  • Hexagonal, P 63 m c

  • a = 5.7702 (1) Å

  • c = 13.0784 (3) Å

  • V = 377.109 (13) Å3

  • Z = 2

  • Mo K[alpha] radiation

  • [mu] = 14.19 mm-1

  • T = 295 K

  • 0.31 × 0.14 × 0.08 mm

Data collection
  • Enraf-Nonius CAD-4 diffractometer

  • Absorption correction: Gaussian (SDP; B. A. Frenz & Associates Inc., 1985[B. A. Frenz & Associates Inc. (1985). SDP. Structure Determination Package, 4th ed. Enraf-Nonius, Delft, The Netherlands.]) Tmin = 0.382, Tmax = 0.529

  • 1284 measured reflections

  • 1247 independent reflections

  • 1065 reflections with I > 1.5[sigma](I)

  • Rint = 0.016

  • 3 standard reflections every 240 min intensity decay: 0.4%

Refinement

Table 1
Results for refinement of different models for SrV6O11 at room temperature

Space group Model Nra NPb R1 wR2
P63/mmc Uniquec 1093 26 0.0574 0.1111
P63mc (x, y, z)d 1093 43 0.0459 0.1002
P63mc (-x, -y, -z)d 1093 43 0.0390 0.0832
P63mc (x, y, z)+(-x, -y, -z)e 1093 44 0.0231 0.0515
P[\overline{6}]2c (x, y, z)f 1093 32 0.0571 0.1110
P[\overline{6}]2c (-x, -y, -z)f 1093 32 0.0571 0.1110
P[\overline{6}]2c (x, y, z)+(-x, -y, -z)g,h 1093 33 0.0558 0.1100
Notes: (a) number of reflections; (b) number of parameters; (c) displacement parameters of V1 are negative; (d) displacement parameters of O2 are negative; (e) the volume fraction (x, y, z):(-x, -y, -z) = 0.434 (7):0.566 (7); (f) displacement parameters of V1 and O2 are negative; (g) displacement parameters of V1, O1 and O2 are negative; (h) the volume fraction (x, y, z):(-x, -y, -z) = 0.519:0.481 (72).

Table 2
Selected bond lengths (Å) for SrV6O11 at 353 K

Sr1-O1 2.7232 (13)
Sr1-O2 2.892 (2)
V1-O1 1.9522 (9)
V1-O3 2.0438 (12)
V2-O1 1.9312 (14)
V2-O2 2.0423 (16)
V3-O2i 1.810 (2)
V3-O3 2.090 (2)
V1-V1ii 2.8887 (1)
V2-V2iii 2.7166 (9)
Symmetry codes: (i) [x-y, x, z+{\script{1\over 2}}]; (ii) -x+y, -x+1, z; (iii) [x, y, -z+{\script{1\over 2}}].

Table 3
Selected bond lengths (Å) for SrV6O11 at room temperature

Sr1-O11 2.738 (2)
Sr1-O12 2.700 (2)
Sr1-O2 2.8887 (9)
V1-O11i 1.9422 (18)
V1-O12 1.9598 (16)
V1-O31 2.069 (2)
V1-O32 2.018 (2)
V21-O11 1.911 (2)
V21-O2 2.063 (3)
V22-O12 1.951 (2)
V22-O2 2.025 (3)
V3-O2ii 1.8057 (15)
V3-O31 2.038 (4)
V3-O32ii 2.136 (3)
V1-V1iii 2.7966 (6)
V1-V1iv 2.9736 (6)
V21-V22 2.7151 (8)
Symmetry codes: (i) [y, -x+y, z+{\script{1\over 2}}]; (ii) [x-y, x, z+{\script{1\over 2}}]; (iii) -y+1, x-y, z; (iv) -x+y, -x+1, z.

Structural parameters including one single-domain model or two inversion twin models, scale factors and one free parameter for extinction correction were refined with SHELXL97 (Sheldrick, 2008[Sheldrick, G. M. (2008). Acta Cryst. A64, 112-122.]).

For both temperatures, data collection: CAD-4 (Enraf-Nonius, 1981[Enraf-Nonius (1981). CAD-4 Operations Software. Enraf-Nonius, Delft, The Netherlands.]); cell refinement: CAD-4; data reduction: SDP (B. A. Frenz & Associates Inc., 1985[B. A. Frenz & Associates Inc. (1985). SDP. Structure Determination Package, 4th ed. Enraf-Nonius, Delft, The Netherlands.]); program(s) used to solve structure: SHELXS97 (Sheldrick, 2008[Sheldrick, G. M. (2008). Acta Cryst. A64, 112-122.]); program(s) used to refine structure: SHELXL97 (Sheldrick, 2008[Sheldrick, G. M. (2008). Acta Cryst. A64, 112-122.]); molecular graphics: VESTA (Momma & Izumi, 2008[Momma, K. & Izumi, F. (2008). J. Appl. Cryst. 41, 653-658.]); software used to prepare material for publication: SHELXL97.


Supplementary data for this paper are available from the IUCr electronic archives (Reference: FA3226 ). Services for accessing these data are described at the back of the journal.


References

B. A. Frenz & Associates Inc. (1985). SDP. Structure Determination Package, 4th ed. Enraf-Nonius, Delft, The Netherlands.
Enraf-Nonius (1981). CAD-4 Operations Software. Enraf-Nonius, Delft, The Netherlands.
Flack, H. D. (1983). Acta Cryst. A39, 876-881.  [CrossRef] [details]
Friese, K. & Kanke, Y. (2006). J. Solid State Chem. 179, 3277-3285.  [ISI] [CrossRef] [ChemPort]
Hata, Y., Kanke, Y., Kita, E., Suzuki, H. & Kido, G. (1999). J. Appl. Phys. 85, 4768-4770.  [ISI] [CrossRef] [ChemPort]
Kanke, Y. (1999). Phys. Rev. B, 60, 3764-3776.  [CrossRef] [ChemPort]
Kanke, Y., Izumi, F., Morii, Y., Akiba, E., Funahashi, S., Kato, K., Isobe, M., Takayama-Muromachi, E. & Uchida, Y. (1994). J. Solid State Chem. 112, 429-437.  [CrossRef] [ChemPort] [ISI]
Kanke, Y., Izumi, F., Takayama-Muromachi, E., Kato, K., Kamiyama, T. & Asano, H. (1991). J. Solid State Chem. 92, 261-272.  [CrossRef] [ChemPort] [ISI]
Kanke, Y., Kato, K., Takayama-Muromachi, E. & Isobe, M. (1992). Acta Cryst. C48, 1376-1380.  [CrossRef] [details]
Kanke, Y., Takayama-Muromachi, E., Kato, K. & Matsui, Y. (1990). J. Solid State Chem. 89, 130-137.  [CrossRef] [ChemPort] [ISI]
Kato, H., Kato, M., Yoshimura, K. & Kosuge, K. (2001). J. Phys. Condens. Matter, 13, 9311-9333.  [ISI] [CrossRef] [ChemPort]
Mentre, O. & Abraham, F. (1996). J. Solid State Chem. 125, 91-101.  [CrossRef] [ChemPort] [ISI]
Mentre, O., Kanke, Y., Dhaussy, A.-C., Conflant, P., Hata, Y. & Kita, E. (2001). Phys. Rev. B, 64, 174404.  [CrossRef]
Momma, K. & Izumi, F. (2008). J. Appl. Cryst. 41, 653-658.  [ISI] [CrossRef] [ChemPort] [details]
Roy, M. E. de, Besse, J. P., Chevalier, R. & Gasperin, M. (1987). J. Solid State Chem. 67, 185-189.
Sheldrick, G. M. (2008). Acta Cryst. A64, 112-122.  [CrossRef] [details]
Uchida, Y., Onoda, Y. & Kanke, Y. (2001). J. Magn. Magn. Mater. 226, 446-448.  [ISI] [CrossRef]
Uchida, Y., Kanke, Y., Takayama-Muromachi, E. & Kato, K. (1991). J. Phys. Soc. Jpn, 60, 2530-2533.  [CrossRef] [ChemPort] [ISI]


Acta Cryst (2010). C66, i99-i102   [ doi:10.1107/S0108270110046299 ]