Bad-Metal-Layered Sulfide Oxide CsV2S2O

Article abstract of DOI:10.1002/ejic.201501154

Through a solid-state reaction between stoichiometric amounts of a mixed cesium oxide Cs₂O_1.3, VS, S, and V₂O₅, a polycrystalline powder of CsV₂S₂O was obtained. Small single crystals could be grown in a CsCl melt by allowing Cs₂SO₄, V metal and S powders to react. The crystals have a plate-like morphology, consistent with the tetragonal crystal-structure symmetry [P4/mmm, a = 3.9455(1), c = 7.4785(1) ?]. Magnetic measurements suggest that CsV₂S₂O is a temperature-independent paramagnet, and resistivity data concur with a bad metal. The mixed oxidation state of V on one crystallographic site offers a tentative explanation of the electronic properties of the title compound. CsV₂S₂O has a tetragonal crystal structure and contains Cs-separated V-S-O layers with relatively short V-V distances of 2.790 ?. V has formally a charge of +2.5, resulting in temperature-independent paramagnetism and bad-metal-like electric conductivity.

Full text of DOI:10.1002/ejic.201501154

DOI: 10.1002/ejic.201501154
Communication
Mixed-Valent Layers
Bad-Metal-Layered Sulfide Oxide CsV2S2O
Martin Valldor,*[a] Patrick Merz,[a] Yurii Prots,[a] and Walter Schnelle[a]
Abstract: Through a solid-state reaction between stoichiomet- symmetry [P4/mmm, a = 3.9455(1), c = 7.4785(1) Å]. Magnetic
ric amounts of a mixed cesium oxide Cs2O1.3, VS, S, and V2O5, measurements suggest that CsV2S2O is a temperature-inde-
a polycrystalline powder of CsV2S2O was obtained. Small single pendent paramagnet, and resistivity data concur with a bad
crystals could be grown in a CsCl melt by allowing Cs2SO4, V metal. The mixed oxidation state of V on one crystallographic
metal and S powders to react. The crystals have a plate-like site offers a tentative explanation of the electronic properties
morphology, consistent with the tetragonal crystal-structure of the title compound.
Introduction
O0.92(7), with Cs scaled to 1. The discrepancy on the V content
might be related to the strong overlapping of V-L with O-K and
V-K with Cs-L X-ray lines in the spectrum. All data from the
single-crystal X-ray diffraction experiment and the results from
the refined atomic parameters can be found in Tables 1 and 2,
respectively. As the elemental analyses indicate slight devia-
tions from ideal stoichiometry, the Cs occupancy was set as a
free parameter during a trial crystal structure refinement; the
resulting value [1.001(3)] was close to ideal and, thus, fixed to
1 in the final crystal structure model.
Layered transition-metal (TM) systems are in focus of solid-state
research since the discovery of the high-temperature supercon-
ductivity in the copper oxides.[1] Partially introducing a heavier
chalcogenide in the oxygen anionic lattice while simultaneously
keeping the quasi-octahedral TM coordination led to the dis-
covery of several new layered structures: La2TM2Se2O3 (TM =
Fe,[2] Co[3]) with layers of 2 [TMSe4/4O2/4], CeCrS2O with a 2D net
∞
of ∞2 [CrS4/2S2/3,CrS4/3O2],[4] and Sr2TMCu2S2O2 (TM = Co, Ni, Cu)[5]
with layers of 2 [TMS2/1O4/2]. Further, Sm2Ti2S2O5 contains dou-
∞
ble layers of 2 [TiO5/2],[6] whereas layers of 2 [VS6/3] are found in
∞
∞
Sr6V9S22O2.[7,8] In fact, most TMs observed in layered chalcogen-
ide oxides possess a single oxidation state and insulating na-
ture. Thus, chalcogenide oxides have not gained the same at-
tention compared to the corresponding oxides. However, due
to lower electronegativity with the heavier chalcogens, there
should be more covalence in the bonding to metal atoms, re-
sulting in weaker electronic correlations. Here, a new layered
vanadium sulfide oxide is presented with an intermediate oxi-
dation state of vanadium that can be described as a bad metal.
Figure 1. Scanning electron microscopy image in backscattering mode of
typical CsV2S2O crystallites, as grown in the CsCl flux.
Results and Discussion
The polycrystalline powder sample, as obtained from the
solid-state reaction, was black. The graphical presentation of
the Rietveld refinement based on the structural model obtained
from the single-crystal data are represented in Figure 2. Three
unidentified peaks are observed, as indicated with arrows in the
inset of Figure 2. However, these intensities correspond to
about 1 % of the maximum intensity of the title phase, suggest-
ing that the sample is satisfactorily pure. The unit-cell parame-
ters obtained from the powder experiment replace those from
the single-crystal experiment, due to the higher data quality
on the peak position from the focusing camera. The largest
differences between observations and calculated pattern in the
powder pattern (Figure 2) are situated at the Bragg positions of
the main phase. The peak profiles are difficult to simulate,
which is most likely due to crystal-lattice strains; lower crystal
From a crystal-growth experiment, thin, black platelets with
nearly quadratic surfaces were obtained (Figure 1). X-ray pow-
der data of the resulting flux with crystals revealed that the title
compound was dominating in the sample, but CsCl and small
amounts of V2O3 were also present (data not shown). The crys-
tals of the title compound could be mechanically extracted
from the solidified CsCl salt melt. Elemental analyses confirm
the expected stoichiometry; energy-dispersive X-ray spectro-
scopy (EDXS) analyses of the eight crystalline individuals re-
sulted in the average composition of Cs1.00(5)V2.25(6)S1.83(3)
-
[a] Max Planck Institute for Chemical Physics of Solids
Nöthnitzer Straße 40, 01187 Dresden, Germany
E-mail: martin.valldor@cpfs.mpg.de
http://www.cpfs.mpg.de/
Eur. J. Inorg. Chem. 2016, 23–27
23
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
Table 1. Details on the single-crystal refinement of CsV2S2O.
Empirical formula
Formula mass [g mol–1
EDXS composition
Crystal system
Space group
a[a] [Å]
CsV2S2O
314.9
Cs1.00(5)V2.25(6)S1.83(3)O0.92(7)
tetragonal
P4/mmm (no. 123)
3.9455(1)
7.4785(1)
116.418(4)
]
c[a] [Å]
V[a] [Å3]
Z
1
4.49
ρ [g cm–3
]
Crystal size [mm]
Temperature [K]
Crystal color
0.005 × 0.060 × 0.060
295
black
Diffractometer
Detector radiation
Wavelength [Å]
θ range [°]
Rigaku AFC7
α
0.71073
2.72–35.54
μ [mm–1
]
12.41
Absorption correction
multi-scan
1159
199
0.035
172
I > 3σ(I)
Figure 2. X-ray diffraction powder data of the CsV2S2O solid-state reaction
sample, where the model from single-crystal structure, run through a Rietveld
refinement, has been superimposed (data = red circles, model = black line).
The Bragg positions are indicated with vertical lines. The bottom full line
displays the difference between calculated pattern and observation. The inset
is a blow-up of a selected angle range, where the unknown reflections are
marked with arrows.
No. of measured reflections
No. of independent reflections
R(int)
No. of observed reflections
Criterion for observation
h, k, l ranges
–6 ≤ h ≤ 6, –3 ≤ k ≤ 6, –9 ≤ l ≤ 11
Refinement on
F2
R(obs), Rw(obs)
R(all), Rw(all)
0.025, 0.055
0.031, 0.058
1.47, 1.50
10
GoF(obs), GoF(all)
No. of refined parameters
Weighting scheme
the title compound. In La2Fe2S2O3, a similar octahedra face-
sharing is found, and the corresponding Fe–Fe distance is
2.857 Å.[2] However, a similar V–V distance is found in the room-
temperature modification of V2O3 (2.697 Å)[10] across the face-
sharing two oxygen octahedra. This distance in the sesquioxide
is related to high-resistance metallic behavior above 160–
170 K,[11] below which an insulating state is found due to a
structural transition involving changes in the V–V distances.[12]
Hence, it can be expected that direct coupling between the V
atoms in the title compound delocalizes electrons. Cs is separat-
1/[σ2(I) + 0.004I2]
1.83, –1.15
430299
Residual peaks [e Å–3
Databank CSD no.[b]
]
[a] From Rietveld refinement. All STDs include the Berar–Lelann factor. [b]
Further details of the crystal structure information may be obtained from the
Fachinformationzentrum Karlsruhe, 76344 Eggenstein-Leopoldshafen, Ger-
many; E-mail: crysdata@fiz-karlsruhe.de.
Table 2. Refined atomic parameters from the single-crystal data.
[a]
ing the 2 [VS4/4O2/4] layers and coordinates to eight S species in
Atom Wyckoff, x, y, z, Uiso
U11, U22, U33
∞
a CsCl-like manner. Here, the Cs–S distance (3.53 Å) is similar
to that of the eight-coordinate Cs in Cs2LiVS4 (3.53–3.83 Å).[13]
CsV2S2O represents a new layered TM-based oxochalcogenide
with a sole ion separating the layers containing octahedrally
coordinated TM. There are only few such compounds, e.g.
CeCrS2O.[4] However, the title compound is different because of
the formally mixed valence of TM. The fact, that a simple ion
separates the TM layers, makes the distances between them
(CsV2S2O: 7.48 Å; CeCrS2O: 5.76 Å) relatively short in comparison
Cs
V
S
1b, 0, 0, 1/2, 0.0171(1)
2f, 1/2, 0, 0, 0.089(3)
0.0153(2), 0.0153(2), 0.0205(4)
0.0058(4), 0.080(4), 0.0131(5)
0.098(4), 0.098(4), 0.0122(8)
–
2h, 1/2, 1/2, 0.2117(2), 0.0106(3)
1a, 0, 0, 0, 0.014(1)
O
[a] U12 = U13 = U23 = 0.
symmetries were not tried, as the effect is most apparent at
low diffraction angles and neither peak splitting nor peak
shoulders are observed at high diffraction angles.
The crystal structure of the title compound is clearly layered to, e.g., 8.95 Å in La2Fe2Se2O3,[2] 8.83 Å in Sr2TMCu2S2O2,[5] and
(Figure 3) with entities that can be described as 2 [VS4/4O2/4]. 11.7 Å in Sr6V9S22O2.[7] This might influence the dimensionality
∞
Similar coordinations of Co and Fe are found in La2TM2Se2O3.[2,3] of the physical properties, where a short inter-layer distance
Accordingly, vanadium is trans-octahedrally coordinated. The V– could increase the 3D character of conductivity and magnetism.
S distance of 2.529 Å is slightly longer than the corresponding The character of the ion separating the TM-S-O entities seems
distance of 2.478–2.480 Å in the VS6 octahedron observed for to play a crucial role for the crystal structure. For the chemically
V3+ in Ba3V2S4O3.[9] As a compensation, the V–O distance in the similar composition BaFe2S2O,[14] a completely different crystal
title compound (1.973 Å) is shorter than the average corre- structure is formed consisting of tetrahedrally coordinated Fe2+
sponding distance of 2.010 Å in V2O3.[10] The octahedral face- by 3 S and 1 O in an extended lattice with quasi-1D ladders.
sharing within the layer results in a relatively short V–V distance The difference in size and/or charge might matter, since Cs+ is
of 2.790 Å, which is markedly shorter than that in the columns significantly larger than Ba2+. Further compositions need to be
[9]
in Ba3V2S4O3 (2.966 Å). This is due to the presence of the synthesized and characterized to draw more general conclu-
relatively smaller O at the faces connecting the octahedra in sions.
Eur. J. Inorg. Chem. 2016, 23–27
www.eurjic.org
24
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
due to a structural change, because a difference scanning calor-
imetry measurement across 280 K (inset Figure 5) reveals only
smooth temperature dependence without anomalies.
Figure 3. Crystal structure of CsV2S2O. The VS4O2 octahedra are highlighted,
and selected interatomic distances [Å] are marked.
The temperature-dependent magnetic susceptibility, χ(T), of
CsV2S2O is shown in Figure 4. Although several different fields
were used, the curves are quite similar in the temperature range
2–300 K; the almost temperature-independent and small sus-
ceptibility of CsV2S2O can be interpreted as van Vleck-like para-
magnetism or as Pauli paramagnetism from conducting elec-
trons. The diamagnetism from closed electron shells is over-
compensated by these paramagnetic contributions; van Vleck
magnetism is due a crystal-field scheme with a non-magnetic
singlet ground state and higher lying states, which can be ther-
mally populated to some extent. However, vanadium in a quasi-
octahedral symmetry is expected to have a magnetic ground
state where a singlet state is highly unlikely. Paramagnetism of
this size coming only from fermions would indicate a high den-
sity of states at the Fermi level and/or strong electronic correla-
tions.
Figure 5. Electric resistivity (ρ) as function of the temperature for a sintered
pellet of polycrystalline CsV2S2O. The inset displays temperature-dependent
DSC data, where the measuring directions are indicated with arrows.
At low fields (e.g. 2 mT) there is a typical diamagnetic signal
from a superconducting phase (inset Figure 4). By using the
density of CsV2S2O, the amount of superconducting volume in
the sample is estimated to be 0.5 %. This indicates that the
signal is due to a superconducting impurity that is not ob-
served in conventional powder X-ray data (Figure 2). Consider-
ing the elements involved, there are only few possible com-
pounds that can cause this effect: CsV6S8 (Tc = 0.32 K), V6S8
(Tc = 0.76 K),[15] or possibly VN (Tc = 1.3–3 K),[16] if nitrogen
entered the synthesis as impurity in one of the constituents. A
tentative explanation would involve a deficient Cs1–xV6S8, simi-
lar to what is observed for BaxV6S8 (x = 0.41–0.48), where Tc =
2.5 K,[17] perhaps increasing Tc to the here observed value by
deficiency on the Cs site. A minor Cs1–xV6S8 impurity could also
be responsible for the magnetic anomaly at 280 K, resembling
a CDW formation, as observed for isostructural Tl1–xV6S8.[18]
Simply from the composition of the title compound and by
assuming an ionic bonding character, vanadium has to possess
an intermediate oxidation state, +2.5. Moreover, there is only
one crystallographic site for V, suggesting an electron delocali-
zation, i.e. conductivity, which would fit the temperature-inde-
pendent paramagnetism (Figure 4). The electric resistivity and
the positive temperature coefficient of ρ(T) above 177 K (Fig-
ure 5) confirm the metal-like behavior of CsV2S2O. Below this
temperature, which is not coinciding with the anomaly in χ(T),
there is a gradual changeover to semiconductivity. Therefore,
we may exclude a classical CDW scenario for the main phase.
Moreover, it cannot be a classic charge-ordering phenomenon
in the sense metal-insulator transition. Instead, the increase of
ρ(T) with decreasing temperature might be some kind of elec-
tronic localization, e.g. of Anderson type.
Figure 4. Magnetic susceptibilities (χ) of CsV2S2O powder as function of tem-
perature at different applied static fields. The arrows indicate the measure-
ment directions. The inset displays the low-temperature data at low field,
including field-cooled (FC) and zero-field-cooled curves (ZFC).
A minor anomaly in χ(T) is discernible at ca. 280 K, which
However, a possible classification of CsV2S2O may be that it
is weaker with lower field; this is the reverse effect of typical is a material at the border between Mott-insulating and metallic
ferromagnetic impurities and cannot be due to a charge-den- (Fermi-liquid) states. The title compound actually exhibits some
sity wave (CDW) for the main phase, as χ does not decrease features that belong to a group of materials, which are crudely
significantly below the anomaly. Further, the anomaly is not called “bad-metals”, where quantum criticality may be involved.
Eur. J. Inorg. Chem. 2016, 23–27
www.eurjic.org
25
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
Their predicted temperature-dependent resistivity[19] resembles
that of CsV2S2O.
Single crystals of the title compound were grown from a metal
halide melt, where Cs2SO4 acted as cesium source. Inside the glove
box, Cs2SO4 (Alfa-Aesar 99.9 %), V metal, elemental S, and CsCl (Bio-
zol 99.9 %) in the molar ratio of 1:8:3:5 were mixed in an agate
mortar, and placed into a silica tube. The tube was shut by using a
rubber-hose-connected home-made glass-plastic vent before trans-
ferring the sample out of the box and seal-melting it after lowering
the inner pressure to less than 10–4 mbar. The sample was placed
in a semi-upright position in a muffle furnace and heated up to
750 °C at a rate of 100 °C/h. This temperature was kept for 100 h
before a slow cooling to 500 °C (300 h) began. The subsequent
cooling progressed at an ambient rate. Because of the toxicity of
H2S that could have formed due to small inclusions of water, the
tube was opened after the reaction inside a separate glove box
(MBraun GB2202-P-PAC) and the box exchange gas was pumped
through an adsorption filter (Miniabsorber, CS Clean Systems AG).
In comparison to the title compound, the quasi-2D vana-
dium-based sulfide Sr3V5S11 has relatively strong electronic cor-
relations, exhibiting paramagnetic, semiconducting proper-
ties.[20] Closer resemblance with CsV2S2O has the layered
Sr6V9S22O2 that is insulating, although its magnetic behavior is
similar to that of the title compound.[7,8] Hence, the title com-
pound seems to be even closer to an electronic instability, lack-
ing localized magnetic moments and having metallic conduc-
tivity at high temperatures. In a broader perspective, the elec-
tronic behavior of CsV2S2O in combination with its layered crys-
tal structure mimic the underdoped, non-superconducting
La2–xSrxCuO4 (x ≈ 0.05).[21] Hence, it will be interesting to dope
CsV2S2O with holes/electrons.
Crystals and powder of CsV2S2O deteriorate slowly in air but fast in
deionized water, forming an opaque bright purple solution.
Single-crystal data were obtained with a Rigaku AFC7 with a CCD
camera as detector (Saturn 724+) and an Mo-Kα X-ray source (λ =
0.71073 Å). The empirical absorption correction proved difficult due
to the flake-like crystal morphology. Hence, harmonic anisotropic
displacement parameters were only used for Cs, V, and S but iso-
tropic for O during the data treatment with JANA2006.[23] All posi-
tions were set to be fully occupied.
Conclusions
The tetragonal crystal structure of CsV2S2O contains layers of
∞2 [VS4/4O2/4], where V is trans-octahedrally coordinated; V occu-
pies only one crystallographic site and has the formal charge
+2.5. The title compound exhibits temperature-independent
paramagnetic behavior, and its conductivity behavior resembles
that of a poor- or bad-metal with a resistivity minimum close
to 177 K.
X-ray powder data were obtained with a Huber camera by using
Cu-Kα1 (λ = 1.54056 Å) radiation, an image plate detector, and a
double-foil flat sample holder in Guinier transmission mode. The
data were treated with Fullprof2k.[24]
The magnetization on a sintered powder sample was measured in
a SQUID magnetometer (MPMS-XL7, Quantum Design) in the tem-
perature range 2–300 K and at fields between 0.002 and 7 T.
Experimental Section
The cesium oxide precursor was prepared from elemental Cs, which
was purified by distillation prior to use. A calculated amount of dry
O2 was slowly added to the liquid Cs inside an evacuated quartz
tube. As no Cs2O is required for the further reaction, a small excess
of oxygen was used to avoid the formation of suboxides, CsxO
(x > 2), which would require to remove the excess metal by distilla-
tion.[22] Once the metal had absorbed all the oxygen, the sample
was ground under argon and heated at 200 °C for 1–2 d. In order
to obtain a homogeneous product, this grinding and heating proce-
dure was carried out 7 times. The orange, air-sensitive product was
sealed under argon in glass ampoules. The relative composition of
Cs2O and Cs2O2 was determined by powder X-ray diffraction to give
an average stoichiometry of Cs2O1.3. All further sample preparations
were done inside a glove-box with controlled Ar atmosphere (Lab-
master, O2 and H2O < 1 ppm). It proved to be dangerous to mix
the obtained Cs2O1.3 with vanadium metal as an obvious reaction
occurred inside the glove box on mixing. Hence, vanadium metal
(Alfa-Aesar 99.5 %) was pre-treated with elemental S (Alfa-Aesar
99.5 %) to form VS in a closed, evacuated quartz tube at 900 °C for
2 d with heating and cooling rates of about 100 °C/h. The obtained,
The electrical resistivity was determined on powder pressed in situ
in a sapphire die cell with four Pt contacts in a van der Pauw ar-
rangement. Direct current of periodically changing polarity was
used.
Differential scanning calorimetry (DSC) investigations were per-
formed in a PerkinElmer DSC-8500 by using an aluminum crucible
with lid and Ar gas flow. Heating and cooling ramps were made
with 10 °C/min.
A scanning electron microscope SEM XL30 with attached energy-
dispersive X-ray spectrometer (EDXS) from Philips working at 15 kV
was used for basic elemental analysis.
Acknowledgments
We like to thank Petra Scheppan for the microscope images and
the respective elemental analyses. Many thanks go to Marcus
Schmidt for the thermal analyses (DSC) and to Ralf Koban for
the assistance with obtaining the magnetic and conductivity
data.
optically homogeneous powder was used as a constituent with the
initial composition for the main reaction:
1
/ Cs2O1.3 + 1.86 VS +
2
0.07 V2O5 + 0.14 S, where V2O5 (Alfa-Aesar 99.5 %) was dried in air
before use. Subsequent to mixing the constituents in an agate mor-
tar, the powder was pressed into pellets and reacted in a corundum
crucible in a vertical, water-cooled furnace inside the glove-box. The
heating rate of the main reaction was 100 °C/h up to the optimized
temperature of 700 °C that was kept for 40 h, followed by cooling to
room temperature at an ambient rate. To obtain the here presented
powder sample, the sample was reground, repressed, and reheated
four times by using the same temperature program.
Keywords: Vanadium · Sulfur · X-ray diffraction · Magnetic
properties · Conducting materials
[1] J. G. Bednorz, K. A. Müller, Z. Phys. B 1986, 64, 189–193.
[2] J. M. Mayer, L. F. Schneemeyer, T. Siegrist, J. V. Waszczak, R. B. van Dover,
Angew. Chem. Int. Ed. Engl. 1992, 31, 1645–1647; Angew. Chem. 1992,
104, 1677–1678.
Eur. J. Inorg. Chem. 2016, 23–27
www.eurjic.org
26
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communication
[3] C. Wang, M.-Q. Tan, C.-M. Feng, Z.-F. Ma, S. Jiang, Z.-A. Xu, G.-H. Cao, K.
Matsubayashi, Y. Uwatoko, J. Am. Chem. Soc. 2010, 132, 7069–7073.
[4] P. J. Dugué, T. Vovan, J. Villers, Acta Crystallogr., Sect. B 1980, 36, 1294–
1297.
[5] C. F. Smura, D. R. Parker, M. Zbiri, M. R. Johnson, Z. A. Gal, S. J. Clarke, J.
Am. Chem. Soc. 2011, 133, 2691–2705.
[6] M. Goga, R. Seshadri, V. Ksenofontov, P. Gütlich, W. Tremel, Chem. Com-
mun. 1999, 979–980.
[7] J. B. Litteer, B.-H. Chen, J. C. Fettinger, B. W. Eichhorn, H. L. Ju, R. L.
Greene, Inorg. Chem. 2000, 39, 458–462.
[8] O. Gourdon, M. Evain, S. Jobic, R. Brec, H.-J. Koo, M.-H. Whangbo, B. Cor-
raze, O. Chauvet, Inorg. Chem. 2001, 40, 2898–2904.
[9] E. J. Hopkins, Yu. Prots, U. Burkhardt, Y. Watier, Z. Hu, C.-Y. Kuo, J.-C.
Chiang, T.-W. Pi, A. Tanaka, L. H. Tjeng, M. Valldor, Chem. Eur. J. 2015, 21,
7938–7943.
[15]
T. Ohtani, Y. Miyoshi, Y. Fujii, T. Koyakumaru, T. Kusano, K. Minami, Solid
State Commun. 2001, 120, 95–99.
W. Meissner, H. Franz, Z. Phys. 1930, 65, 30–54.
O. Fuentes, H.-H. Wang, B. H. Ward, J. Zhang, D. M. Proserpio, F. Calvagna,
C. E. Check, K. C. Lobring, C. Zheng, Chem. Mater. 2001, 13, 3051–3056.
C. F. Miclea, H. D. Hochheimer, B. Martin, C. Miclea, T. Ohtani, New J. Phys.
2013, 15, 083008.
J. Vučičević, D. Tanasković, M. J. Rozenberg, V. Dobrosavljević, Phys. Rev.
Lett. 2015, 114, 246402.
[16]
[17]
[18]
[19]
[20]
Y. Klein, H. Moutaabbid, N. Boubiche, A. Soyer, G. Rousse, A. Gauzzi, J.
Phys. Condens. Matter 2013, 25, 435604.
[21] H. Takagi, B. Batlogg, H. L. Kao, J. Kwo, R. J. Cava, J. J. Krajewski, W. F.
Peck Jr., Phys. Rev. Lett. 1992, 69, 2975–2978.
[22] A. Helms, W. Klemm, Z. Anorg. Allg. Chem. 1939, 242, 33–40.
[23] V. Petříček, M. Dušek, L. Palatinus, JANA2006, Institute of Physics, Acad-
emy of Sciences of the Czech Republic, Praha, Czech Republic, 2013.
[24] J. Rodríguez-Carvajal, Phys. Bl. 1993, 192, 55–69.
[10] M. G. Vincent, K. Yvon, J. Ashkenasi, Acta Crystallogr., Sect. A 1980, 36,
808–813.
[11] J. M. Honig, J. Solid State Chem. 1982, 45, 1–13.
[12] P. D. Dernier, M. Marezio, Phys. Rev. B 1970, 2, 3771–3776.
[13] F.-Q. Huang, B. Deng, J. A. Ibers, J. Solid State Chem. 2005, 178, 194–199.
[14] M. Valldor, P. Adler, Yu. Prots, U. Burkhardt, L. H. Tjeng, Eur. J. Inorg. Chem.
2014, 36, 6150–6155.
Received: October 12, 2015
Published Online: December 9, 2015
Eur. J. Inorg. Chem. 2016, 23–27
www.eurjic.org
27
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim