Synthesis, Mo-valence state, thermal stability and thermoelectric properties of SrMoO3-xNx (x>1) oxynitride perovskites

Article abstract of DOI:10.1016/j.jssc.2007.06.035

Oxynitrides of the general composition SrMoO_3-xN_x (x>1) were synthesized by thermal ammonolysis of crystalline SrMoO₄. According to neutron and X-ray diffraction experiments, the materials crystallize in the cubic perovski

Full text of DOI:10.1016/j.jssc.2007.06.035

ARTICLE IN PRESS
Journal of Solid State Chemistry 180 (2007) 2649–2654
www.elsevier.com/locate/jssc
Synthesis, Mo-valence state, thermal stability and thermoelectric
properties of SrMoO N (x41) oxynitride perovskites
3ꢀx x
a
D. Logvinovich , R. Aguiar , R. Robert , M. Trottmann , S.G. Ebbinghaus ,
b
a
a
b
b
A. Reller , A. Weidenkaff
a,
Ã
a
Empa, Swiss Federal Laboratories for Materials Testing and Research, Solid State Chemistry and Catalysis, Ueberlandstr. 129,
CH-8600 Duebendorf, Switzerland
b
Universit a¨ t Augsburg, Festk o¨ rperchemie, Institut f u¨ r Physik, Universit a¨ tsstrasse 1, D-86159 Augsburg, Germany
Received 15 March 2007; received in revised form 5 June 2007; accepted 20 June 2007
Available online 20 July 2007
Abstract
Oxynitrides of the general composition SrMoO3 Nx (x41) were synthesized by thermal ammonolysis of crystalline SrMoO .
ꢀx
According to neutron and X-ray diffraction experiments, the materials crystallize in the cubic perovskite structure (space group Pm3m).
4
¯
X-ray absorption spectroscopy (XAS) shows evidence of local distortions of the Mo(O,N) octahedra. The oxidation states of Mo
6
determined by X-ray absorption near edge structure (XANES) spectroscopy are slightly lower than that calculated from the oxygen/
nitrogen(O/N) content. The disagreement arises from the higher covalence of the Mo–N bonding when compared to the Mo–O bonding
(‘‘chemical shift’’). The electrical transport properties of the samples are strongly different from SrMoO . It was found that the
3
conductivity of the samples decreases with increasing nitrogen content. The Seebeck coefficient values are up to three times higher than
those of SrMoO3.
r 2007 Elsevier Inc. All rights reserved.
Keywords: Oxynitride; Molybdate; XANES; Electronic properties
1
. Introduction
from semiconducting to metallic when increasing the
3
+
substitution level of La [3,4], all the oxynitride samples
remain semiconducting. Moreover, a decrease of the
electrical conductivity with increasing nitrogen content
was measured for the oxynitride samples with x40.53.
Thus, changing the nitrogen content can lead to a dramatic
change of the physical properties of oxynitrides.
Partial oxygen substitution by nitrogen in perovskite-
type oxides of the general formula ABO is an interesting
3
3
ꢀ
method to change their physical properties [1,2]. Since N
2
ꢀ
has a higher negative charge than O , nitrogen insertion
into the oxygen sublattice formally leads to an increase of
the B-site cation oxidation state, and consequently changes
the charge carriers concentration. This method has been
considered as an alternative to the cationic substitution to
change transport properties of the p-type conducting
Another example in which oxygen substitution for
nitrogen leads to materials with modified physical proper-
ties is the SrMoO3 N system. Up to now, only studies on
ꢀx
x
samples with xp0.5 (SrMoO2.6N0.4 and SrMoO2.5N0.5
)
2
+
2+
LaVO3 [1]. Similar to the Ba
LaVO , the synthesized oxynitrides of general formula
and Sr
substituted
have been described in the literature [5,6]. Both neutron
and X-ray diffraction (XRD) confirm no deviation from
3
¯
the cubic symmetry (space group Pm3m) for these
LaVO3 N (xo0.53) show an increased electrical con-
ꢀx
ductivity compared to that of LaVO . While the con-
x
compositions. Their electronic conductivity was several
orders of magnitude lower than that of SrMoO [7].
3
2
+
2+
ductivity of Sr
and Ba
containing samples changes
3
Herein we report the synthesis procedure for higher level
substituted oxynitrides of the composition SrMoO N ,
in particular of those with x41. Furthermore, their
Ã
Corresponding author.
E-mail address: anke.weidenkaff@empa.ch (A. Weidenkaff).
3ꢀx
x
0022-4596/$ - see front matter r 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.jssc.2007.06.035

ARTICLE IN PRESS
D. Logvinovich et al. / Journal of Solid State Chemistry 180 (2007) 2649–2654
2650
crystallographic structure and physical properties are
investigated in detail. The results obtained by neutron
diffraction (ND), XRD, X-ray absorption near edge
spectroscopy, Seebeck coefficient and electrical conductiv-
ity measurements are presented.
T ¼ 1073 K under ammonia (PanGas, 499.98%) flow of
150 mL/min. The synthesis temperature was reached with a
heating rate of 10 K/min. After the reaction, the furnace
was cooled down to room temperature with a cooling rate
of 10 K/min under flowing ammonia. Three samples with
different oxygen/nitrogen (O/N) content were obtained
after 11, 48 and 72 h of the ammonolysis.
2
. Experimental
The cationic composition was studied by energy
dispersive X-ray spectrometry (EDXS-Link Pentafet
5947, from Oxford Microanalysis group). The O/N content
of the different samples was measured by the hotgas-
extraction method using a LECO TC500 analyzer. Silicon
nitride and silicon oxide were used as calibration standards
for nitrogen and oxygen, respectively.
A total of 0.04 mol MoO3 (JMC, Specpure) was
dissolved in minimal amount of NH3 (aq, 25%) and
precipitated with 100 mL of an aqueous solution of
Sr(NO3)2 (Merck, 499%) with the concentration of
0.4 mol/L. The precipitate was washed with distilled water,
dried and annealed at 1073 K for 4 h to form phase pure,
well crystalline SrMoO as confirmed by XRD (Fig. 1A).
4
Structure and phase purity of the obtained oxynitrides
were investigated by X-ray diffraction using a Phillips
X’Pert PRO MPD Y–Y system equipped with a linear
detector X’Celerator. Lattice parameters were obtained
from Rietveld refinements of the XRD data.
ND data were recorded at the high resolution powder
diffractometer for thermal neutrons (HRPT) [9] located at
the Swiss Spallation Neutron Source (SINQ) of the Paul
Scherrer Institute in Switzerland as well as at the SPODI
diffractometer of the FRM2 in Garching, Germany.
Samples were placed in cylindrical vanadium cans with
8 mm diameter. The measurements were performed with a
SrMoO3 was prepared by reduction of 1 g freshly
synthesized SrMoO at T ¼ 1373 K with forming gas (5%
4
H /95% N , 99.999% purity, Pangas). A gas flow of
2
2
3
5
00 mL/min was applied. The reduction was completed in
h. The prepared material was of a purple-red color. Its
phase purity was confirmed by XRD (Fig. 1B). The lattice
˚
parameter refined from the XRD data, a ¼ 3.9752(1) A, is
in perfect agreement with the previously reported value of
˚
.9751(3) A for the sample with anionic stoichiometry
3
equal to 3 [7]. When stored at ambient conditions SrMoO3
transforms slowly to SrMoO . The presence of moisture
4
˚
promotes a faster SrMoO4 formation. Therefore, in
between handlings the material was kept in a desiccator.
The thermal ammonolysis of SrMoO was performed in
neutron wavelength l ¼ 1.54842 A in the angular range of
2–1521 with a step size of 0.051. The background was fitted
with a 6-coefficients polynomial function. The peak shape
was modeled using a Tompson–Cox–Hastings pseudo
Voigt function. Structure refinements were done using
Fullprof [10].
4
a rotating cavity reactor described elsewhere [8]. A total of
.25 g of the SrMoO4 powder was ammonolysed at
2
X-ray absorption near edge structure (XANES) mea-
surements at the MoK-edge were performed at the beam-
line CEMO at the Hamburger Synchrotron Radiation
Laboratory, HASYLAB. Data were collected in transmis-
sion mode. The energy scale was calibrated from the first
inflection point in the K-edge of Mo metal (19.999 keV),
which was measured simultaneously as reference. The
beamline was equipped with a double-crystal fixed exit
monochromator with Si(311) crystal pairs. Signal inten-
sities where detuned to 40% of their maximum levels to
minimize contributions of higher harmonics. Spectra were
acquired in 2 eV increments for the pre-edge region
(19.80–19.98 keV) and 0.20 eV near the edge (19.98–
20.03 keV). SrMoO and SrMoO were measured as
standards for Mo
3
+
4
4
6+
and Mo
respectively. For data
analysis the software package WinXAS v3.1 was used [11].
The electrical transport measurements were performed
on bars with dimensions of 1.65 mm ꢁ 5–10 mm ꢁ 1 mm.
The bars were obtained by applying an uniaxial pressure of
10 bar followed by a cold isostatic pressing at 2000 bar. The
bars were sintered at T ¼ 1073 K for half an hour under
flowing ammonia. This short sintering time has no
significant influence on the O/N content and purity of the
samples as verified by hotgas extraction and X-ray powder
diffraction. The measured density of the obtained ceramics
Fig. 1. XRD patterns confirming the phase purity of: (A) SrMoO
annealing at 1073 K for 4 h. Space group: I41/a. The reflections are
(Pm3
4
after
assigned according to the pdf entry 01-085-0809. (B) SrMoO
3
¯
m,
˚
a ¼ 3.9752(1) A) obtained by the reduction of SrMoO
4
.

ARTICLE IN PRESS
D. Logvinovich et al. / Journal of Solid State Chemistry 180 (2007) 2649–2654
2651
varied from 74% to 89% of the theoretical density.
Electrical conductivity (DC four-probe method) and
Seebeck coefficient were measured simultaneously as a
function of temperature in the temperature range of
respectively. Hence, the nitrogen content of our samples is
more than twice higher than reported before [5,6]. The
measured O/N content corresponds to a mixed oxidation
state between +5/+6 for Mo. EDX reveal an average
Sr:Mo ratio of all oxide and oxynitride samples equal to
0.98(1):1.02(2). Thus, within this method mistake (about
3%) the obtained ratio is identical with the ideal ratio 1:1.
The dependence of the measured lattice constant from
the nitrogen content of the samples is shown in Fig. 3. The
lattice parameter of the sample treated for 11 h is
3
40 KoTo950 K using the RZ2001i measurement system
from Ozawa science, Japan. For the measurements Pt-
contacts were used.
3
. Results and discussion
˚
According to the XRD results (Fig. 2), a perovskite-type
single-phase oxynitride is already obtained after 11 h of
3.9782(1) A, which is higher of that reported in the
˚
literature for SrMoO2.5
SrMoO
N
0.5 (3.9773(1) A) [6] and of
. Larger lattice constants were found for the
ammonolysis at T ¼ 1073 K, whereas the formation of
3
Mo N impurity phase was observed at higher tempera-
2
samples with higher nitrogen content. Nitrogen insertion is
expected to lead to an increase of the cell parameter since
˚
the effective ionic radius of N (1.32 A) is larger than that
2
of O (1.26 A). On the other hand, oxidation of Mo
tures.
3ꢀ
The samples are of a dark-blue color. Similar to SrMoO3
they are sensitive to the storing conditions. In particular, in
contact with moisture they evolve ammonia and transform
ꢀ
4+
˚
5
+
6+
˚
(0.79 A) to Mo
˚
(0.75 A) and Mo
˚
(0.73 A) [12,13]
into SrMoO within a period of a few months. Since the
4
should lead to a decrease of the lattice constant. Although
these two factors are expected to partly compensate each
other, an increase of the lattice constant with the nitrogen
content is measured, indicating that the influence of the
larger anionic radius is the dominating effect.
samples may also react with oxygen they were stored in dry
N after the synthesis.
2
The O/N content measured by hotgas extraction
corresponds to the compositions SrMoO1.95(5)N1.05(5)
,
SrMoO1.81(5)N1.19(5) and SrMoO1.73(5)N1.27(5) for the com-
pounds obtained after 11, 48 and 72 h of ammonolysis,
Perovskite-type compounds often possess pseudosym-
metry which originates from a tilting of the BO -octahedra,
6
while the cations occupy the same positions as in the cubic
aristotype structure. When the tilting is rather small it
cannot be probed by XRD due to a low X-ray scattering
power of anions. In contrast, the neutron scattering
lengths of oxygen and nitrogen are large and therefore
superstructure reflections, which occur due to the pseudo-
symmetry are often intense. One example for such
pseudosymmetry is SrNbO N, which, according to XRD,
2
crystallizes as a simple cubic perovskite while ND revealed
a tetragonal supercell [14]. Additionally, since oxygen and
nitrogen are distinguishable with neutrons, ND allows to
verify the O/N content measured by hotgas extraction.
Fig. 3. The influence of the nitrogen content (measured by hotgas
extraction) on the lattice constants (refined from XRD data) of the
samples.
4 3
Fig. 2. XRD patterns of SrMoO samples reacted with NH : (A) 11 h
reacted sample, (B) 48 h reacted sample, and (C) 72 h reacted sample.

ARTICLE IN PRESS
D. Logvinovich et al. / Journal of Solid State Chemistry 180 (2007) 2649–2654
2652
Rietveld refinement of the 11 h ammonolysed sample
¯
was carried out in the space group Pm3m. In the starting
structural model the Sr:Mo ratio was set to 1:1 (based on
the cationic compositions study), whereas the O:N ratio
was set to 2:1. Thermal displacement factors were refined
isotropically for all atoms. The occupancy factors for
oxygen and nitrogen were initially refined with the anionic
site constrained to be fully occupied. In succeeding runs the
occupancy factors of the anions were refined indepen-
dently. Finally, the lattice and the profile parameters, 2Y0,
the background coefficients, the thermal displacement
factors and the anionic occupancies were refined together.
Within the doubled standard deviation (2s) no difference
between the parameters obtained with the constrained and
the unconstrained model were found. For this reason in the
final refinement a complete occupation of the anionic sites
was assumed and only the O/N ratio was refined.
Refinements of cationic occupancies gave no hints for the
presence of the cationic vacancies. A summary of the
structural parameters refined from the neutron data can be
found in Table 1. The statistics of the refinement, the visual
inspection of the fit (Fig. 4) and the refined values of the
thermal displacement factors indicate that the chosen
model is correct. The refined O/N content corresponds to
the composition SrMoO1.89(2)N1.11(2), which is in a reason-
Fig. 5. Normalized MoK-edge XANES spectra of the oxynitrides in
comparison to the SrMoO and SrMoO standards.
3
4
able agreement with the result obtained by hotgas
extraction.
The MoK-edge XANES spectra of SrMoO , SrMoO ,
3
4
SrMoO1.95(5)N1.05(5) and SrMoO1.81(5)N1.19(5), are shown in
Fig. 5. The edge shifts are obtained by using a well resolved
feature above the absorption edge as shown in the inset of
Fig. 5, and are reported relative to the first inflection point
in the Mo metal K-edge at 19.999 keV [15]. The shift of the
MoK absorption edge to higher phonon energies with
respect to the metal standard gives information about the
average valence of the Mo while the pre-edge features
correspond to its coordination geometry [16].
Table 1
Structural parameters of the SrMoO1.95
data
N1.05 refined from neutron powder
2
˚
Name
x
y
z
B
iso (A )
Occupancy factor
Site
The XANES spectrum of SrMoO4 shows a strong
pre-edge peak which arises from an allowed 1s–4d
electronic transition for tetrahedral symmetry [17].
This characteristic can also be observed as a shoulder for
the oxynitrides with compositions SrMoO1.95(5)N1.05(5) and
SrMoO1.81(5)N1.19(5). Since no SrMoO4 impurity was
detected with XRD and ND, the feature most probably
corresponds to a distortion in the octahedral oxygen
Sr
Mo
1/2
0
1/2
0
1/2
0
0.92(4)
0.72(4)
0.64(2)
1
1b
1a
3d
1
0.64(1)/0.36(1)
O/N(1)
1/2
0
0
2
Space group: Pm 3¯ m (a ¼ 3.9835(1) A, wR
˚
¼ 6.78, R
¼ 5.20, w ¼ 1.36).
p
p
environment of Mo-ions reported before for MoO [16].
3
In the XANES spectrum of SrMoO , where Mo is in a
3
regular octahedron, this feature is not observed. Since for
the oxynitrides no deviations from the cubic symmetry and
no anomalous displacement parameters of their constituent
ions have been detected by using ND (see Table 1), the
local distortions in the surrounding of the Mo ions must
have a too small coherence length to be detected with
diffraction techniques. Most probably, the symmetry
2
ꢀ
reduction arises from the different polarizabilities of O
ꢀ
3
and N -ions, and their disordered arrangement around
the Mo-ions. This interpretation is in accordance with
recent EXAFS studies on BaTaO N, which reveal distorted
octahedral arrangement of the Ta -ions [18]. It should be
noted that like in our case neither XRD nor ND had given
any hints on the possible deviation of the local symmetry
2
5
+
Fig. 4. Rietveld refinement plot of the ND data for the 11 h ammonolyzed
sample. Space group: Pm3
¯
m, a ¼ 3.9835(1) A
˚
, wR ¼ 6.78, R ¼ 5.20,
from O . While diffraction methods give information on a
h
long range ordering averaged over the whole sample, XAS
p
p
2
w ¼ 1.36.

ARTICLE IN PRESS
D. Logvinovich et al. / Journal of Solid State Chemistry 180 (2007) 2649–2654
2653
Fig. 6. Linear relationship of average Mo valence and MoK-edge position
relative to Mo metal) calculated from K-edge position of the standards
SrMoO and SrMoO
Fig. 7. Temperature dependence of Seebeck coefficient of the oxynitrides.
The inset shows an irreversible decrease of the measured values when
samples were heated above their decomposition temperature.
(
3
4
.
is sensitive to the short-range surrounding of the respective
element and therefore this method is well suited to detect
local deviations from the average symmetry [19].
From the energy shift of the X-ray absorption edge, the
oxidation state of transition metals can be derived [20].
Fig. 6 depicts the calculated average Mo valence state for
SrMoO1.95(5)N1.05(5) (+4.62) and SrMoO1.81(5)N1.19(5)
(
+4.52). On the other hand, the valence state obtained
from the O/N content of these samples are +5.03 for
SrMoO1.95(5)N1.05(5) and +5.19 for SrMoO1.81(5)N1.19(5)
.
This seeming discrepancy can be explained by the fact that
the absorption edge shift originates from both the ‘‘valence
shift’’ (i.e. shift due to changing the oxidation state) and
the ‘‘chemical shift’’ (shift due to different electronegativ-
2
ꢀ
3ꢀ
ities of O and N -ligands). Since the oxide standards
used are not able to account for differences in electro-
negativity of the anions, a big discrepancy between the
calculated and the expected Mo valences arises.
Fig. 8. Temperature dependence of electrical conductivity of the
oxynitrides.
The measured Seebeck coefficient values for the synthe-
sized samples are up to three times higher than those
of Mo is the formation of the very stable N molecule
2
reported for SrMoO (S ¼ 4–9 mV/K) [7] but still close to
(941 kJ/mol).
3
those of metals. The Seebeck coefficient increases with
increasing nitrogen content of the samples. The evolution
of S with temperature and N content is shown in Fig. 7.
Apparently, the values for all samples increase with
temperature passing through a maximum at T ¼ 650 K
The electrical conductivity of the samples decreases with
increasing nitrogen content as shown in Fig. 8. Contrary to
the metallic SrMoO3 the electrical conductivity of the
oxynitrides increases with temperature, i.e. a semiconduct-
ing behavior is observed. The measured conductivity values
for SrMoO
T ¼ 550 K for SrMoO
N
, T ¼ 600 K for SrMoO
N
and
are 3 orders of magnitude lower than those of SrMoO
ꢀ1
1
.95 1.05
1.81 1.19
3
3
N . Further heating of the
.73 1.27
(12.8 ꢁ 10 S cm ) [7] and are of the same order of
1
ꢀ1
samples is leading to the formation of metallic Mo
impurity phase as revealed by XRD and an irreversible
decrease of the Seebeck coefficient. Thus, the temperature
at which the maximum of the Seebeck is reached can be
considered as decomposition temperature. Apparently,
the formation of metallic Mo is coupled with a nitrogen
release from the samples. We have already reported
magnitude as those of SrMoO2.6N0.4 (66.7 S cm ) [5],
but one order of magnitude higher than reported for
ꢀ
1
SrMoO2.5N0.5 (2 S cm ) [6]. It should also be noticed
that the temperature dependence of the SrMoO2.6N0.4
conductivity is semiconductor-like [5] similar to our
samples, whereas that of SrMoO2.5N0.5 was described as
metallic [6].
similar behavior for Ca-substituted LaTiO N heated under
2
The power factor of our samples increases with
temperature (Fig. 9). The measured values are lower than
N (1 atm) [8]. The main driving force for the reduction
2

ARTICLE IN PRESS
D. Logvinovich et al. / Journal of Solid State Chemistry 180 (2007) 2649–2654
2654
perovskite structure for the synthesized compounds,
whereas local distortions of the Mo(O,N) octahedra are
detected with XANES. These distortions may arise due to
6
2
ꢀ
3ꢀ
the different polarizabilities of O and N and anionic
disorder. Seebeck values of the oxynitrides are similar to
those of metals. Smaller values of electrical conductivity of
our samples compared to those of SrMoO are measured.
3
Electrical transport property measurements on single
crystalline thin films with the same compositions are
planned for the future to study the possible influences of
grain boundaries on the physical properties.
Acknowledgments
The authors acknowledge the German Science Founda-
tion (DFG-SPP 1136), Swiss Science Foundation (SNF-
MaNEP) and Empa for the financial support as well as
Dr. Denis Sheptyakov (SINQ) for the technical assistance.
This work is partly based on the experiments performed at
the Swiss Spallation Neutron Source SINQ, Paul Scherrer
Institute, Villigen, Switzerland.
Fig. 9. Variation of power factor of the oxynitrides samples with
temperature.
those reported for SrMoO [21] due to the lower electrical
3
conductivity of the oxynitrides.
The conductivity values and behavior of our samples are
typical for semiconductors or poor metals. The density
achieved for the ceramics (up to 89% of the theoretical
References
density) resembles the one reported for SrMoO (90.8%)
3
[
[
[
1] P. Antoine, R. Assabaa, P. L’Haridon, R. Marchand, Y. Laurent,
C. Michel, B. Raveau, Mater. Sci. Eng. B 5 (1989) 43–46.
2] P. Antoine, R. Marchand, Y. Laurent, C. Michel, B. Raveau, Mater.
Res. Bull. 23 (1988) 953–957.
[
7]. The observed high deviation between the conductivity
values measured on our samples and reported for SrMoO3
therefore cannot be attributed to the difference between the
samples density. Another factor that can affect the
conductivity and its temperature dependence is the
composition of grain boundaries. Indeed, taking into
account that the samples are air sensitive, it may happen
that the grain surface partly reoxidizes during the handling
and the measured slight temperature dependence of the
conductivity originates from the grain boundaries influ-
ence. The measured Seebeck values suggest that the
synthesized samples are metallic-like. To clarify, whether
the measured semiconductive behavior of the conductivity
is an intrinsic property of our samples further investiga-
tions are in progress. In particular, thin films resistivity
measurements free from grain boundaries effects can help
to verify the conductivity values obtained in present study.
3] P. Dougier, A. Casalot, J. Solid State Chem. 2 (1970) 396–403.
[4] P. Dougier, P. Hagenmuller, J. Solid State Chem. 15 (1975) 158–166.
[
[
5] G. Liu, X. Zhao, H.A. Eick, J. Alloys Compd. 187 (1992) 145–156.
6] I.D. Fawcett, K.V. Ramanujachary, M. Greenblatt, Mater. Res. Bull.
3
2 (1997) 1565–1570.
7] L.H. Brixner, J. Inorg. Nucl. Chem. 14 (1960) 225–230.
8] D. Logvinovich, A. Borger, M. Dobeli, S.G. Ebbinghaus, A. Reller,
[
[
¨
¨
A. Weidenkaff, Prog. Solid State Chem. 35 (2007) 281–290.
9] P. Fischer, G. Frey, M. Koch, M. Konnecke, V. Pomjakushin,
J. Schefer, R. Thut, N. Schlumpf, R. Burge, U. Greuter, Physica B
[
2
76–278 (2000) 146–147.
[
[
10] J. Rodriguez-Carvajal, Physica B 192 (1993) 55–69.
11] T. Ressler, J. Synchrotron Rad. 5 (1998) 118–122.
[12] R.D. Shannon, C.T. Prewitt, Acta Crystallogr. B: Struct. Sci. 25
1969) 925–946.
13] R. Shannon, Acta Crystallogr. A: Found. Crystallogr. 32 (1976)
51–767.
(
[
[
7
14] S.G. Ebbinghaus, A. Weidenkaff, A. Rachel, A. Reller, Acta
Crystallogr. C: Cryst. Struct. Commun. 60 (2004) i91–i93.
15] J.A. Bearden, A.F. Burr, Rev. Mod. Phys. 39 (1967) 125–142.
16] T. Ressler, O. Timpe, T. Neisius, J. Find, G. Mestl, M. Dieterle,
R. Schlogl, J. Catal. 191 (2000) 75–85.
4
. Conclusions
[
[
Solid solutions of the general composition SrMoO3 Nx
ꢀx
with x41 were successfully synthesized by thermal
ammonolysis of SrMoO . The nitrogen content of our
[
[
[
[
[
17] F.W. Kutzler, C.R. Natoli, D.K. Misemer, S. Doniach,
K.O. Hodgson, J. Chem. Phys. 73 (1980) 3274–3288.
4
samples is more than twice higher than previously reported
and is confirmed by both ND and hotgas extraction.
XANES studies result in a lower oxidation state of Mo
with respect to the values calculated from the nitrogen
content, resulting from a higher covalency of the Mo–N
bond compared to the Mo–O bond (‘‘chemical shift’’).
Both XRD and ND reveal no deviations from the cubic
18] B. Ravel, Y.-I. Kim, P.M. Woodward, C.M. Fang, Phys. Rev. B:
Condens. Matter Mater. Phys. 73 (2006) 184121–184127.
19] S.G. Ebbinghaus, A. Weidenkaff, R.J. Cava, J. Solid State Chem. 167
(
2002) 126–136.
20] S. Ebbinghaus, Z. Hu, A. Reller, J. Solid State Chem. 156 (2001)
94–202.
1
21] T. Maekawa, K. Kurosaki, H. Muta, M. Uno, S. Yamanaka,
J. Alloys Compd. 390 (2005) 314–317.