A new high-temperature cubic fluorite-type phase Mo0.16Bi0.84O1.74 with a Rare three-dimensional incommensurate modulation

Article abstract of DOI:10.1006/jssc.2000.8742

A new high-temperature phase Mo_0.16Bi_0.84O_1.74 was obtained by quenching from 800°C. Its average structure is of the cubic Fm3m flourite type with a = 5.6478(1) A and was refined from single-crystal Mo Kα X-ray diffraction (XRD) data to a weighted R-value of 3.7% for 16 unique reflections and isotropic thermal displacement parameters refined for the metal atoms, U(iso) = 0.0564(4) A², and the O atoms, U(iso) =0.186(6) A². The structure exhibits in addition a novel 3-D incommensurate modulation that was studied by electron diffraction and both powder and single crystal XRD. Extinction conditions indicate the symmetry P:Fm3m:Fm3m, which has previously only been reported once. The three q-vectors are along the unit cell axes and have lengths of 0.29256(6) unit cells in reciprocal space. The modulated structure is compared with the previously reported and similarly modulated structure of the solid solution Nb(x)Bi(1-x)O(1.5+x), with 0.12 ≤ x ≤ 0.46 and symmetry P:Fm3m: Fd3m. (C) 2000 Academic Press.

Full text of DOI:10.1006/jssc.2000.8742

Journal of Solid State Chemistry 152, 573}576 (2000)
doi:10.1006/jssc.2000.8742, available online at http://www.idealibrary.com on
RAPID COMMUNICATION
A New High-Temperature Cubic Fluorite-Type Phase Mo0.16Bi0.84O1.74
with a Rare Three-Dimensional Incommensurate Modulation
M. Valldor,ꢀ S. Esmaeilzadeh, C. Pay-Gomez, and J. Grins
Department of Inorganic Chemistry, Arrhenius Laboratory, University of Stockholm, SE-106 91, Sweden
Received February 22, 2000; accepted April 7, 2000
et al. (3). Its structure has so far not been determined, but it
A new high-temperature phase Mo0.16Bi0.84O1.74 was obtained
by quenching from 8003C. Its average structure is of the cubic
Fm3m 6ourite type with a ؍ 5.6478(1) A and was re5ned from
single-crystal Mo Ka X-ray di4raction (XRD) data to a weighted
R-value of 3.7% for 16 unique re6ections and isotropic thermal
displacement parameters re5ned for the metal atoms,
is #uorite-related with an orthorhombic unit cell with
a"16.8065(2), b"28.6004(4), and c"16.8898(2) A and the
probable space group symmetry Pccn (2,3). High-temper-
ature #uorite-related Bi}Mo oxide phases have not been
characterized in any detail. The synthesis and structural
characterisation of the 6-D incommensurate structure of the
U
iso ؍ 0.0564(4) A2, and the O atoms, Uiso ؍0.186(6) A2. The
phase Mo
Bi
O
is reported here. The observed
structure exhibits in addition a novel 3-D incommensurate
modulation that was studied by electron di4raction and both
powder and single crystal XRD. Extinction conditions indicate
the symmetry P:Fm3m:Fm3m, which has previously only been
reported once. The three q-vectors are along the unit cell axes
and have lengths of 0.29256(6) unit cells in reciprocal space. The
modulated structure is compared with the previously reported
and similarly modulated structure of the solid solution
NbxBi1؊xO1.5؉x, with 0.124x40.46 and symmetry P:Fm3m:
Fd3m. ( 2000 Academic Press
ꢇꢈꢀꢃ ꢇꢈꢅꢉ ꢀꢈꢆꢉ
symmetry is rare and the existence of the phase is therefore
of general interest for studies of aperiodic structures.
EXPERIMENTAL
The Mo
Bi
O
phase was synthesized by melting
ꢇꢈꢀꢃ ꢇꢈꢅꢉ ꢀꢈꢆꢉ
a stoichiometric mixture of analytical grade Bi O and
ꢁ ꢂ
MoO in a Pt crucible at 9503C for a few minutes, cooling
ꢂ
Key Words: Mo}Bi}O; 6uorite type structure; incommensur- the sample to 8003C at a rate of 7.53/min, and then removing
ate; 3D modulation.
it rapidly from the furnace. The obtained sample was homo-
geneous and contained large reddish-yellow and transpar-
ent crystals.
INTRODUCTION
The unit cell parameter and q-vector length were deter-
mined from Guinier}Hagg "lm data, using Cu Ka radi-
ꢀ
The high-temperature #uorite-type d form of Bi O is ation and Si as internal standard. The "lms were evaluated
ꢁ ꢂ
stable above 7303C up to its melting temperature 8243C and with an LS-KEJ20 scanner (4). Powder XRD data were also
is one of the best oxygen ion conductors known (1). The recorded with a STOE STADI!P di!ractometer, with Cu
pure Bi phase cannot be quenched to room temperature but Ka radiation and the sample in symmetric transmission
ꢀ
closely related modi"cations can be stabilized at lower tem- mode. Data were collected from 5}1303 in 2h in steps of 0.23
peratures by addition of numerous cations (1), either with a recording time of 350 seconds at each step. Single-
ꢂ>
ꢂ>
isovalent (Y , Ln ) or with smaller ionic radii and higher crystal X-ray di!raction (XRD) data were obtained using
ꢃ>
ꢄ>
ꢃ>
ꢄ>
valence states (e.g., W , Ta , Mo , Nb ). A number of a rotating anode Mo Ka X-ray source and a STOE image
Bi O -based systems of the latter type are described by Ling plate detector system. The data set consisted of 100 di!rac-
et al. (2). In the Bi O }MoO system there exists a low- tion images with the crystal rotated 2 degrees between every
temperature phase Bi Mo O having a metal composi- 5 min exposure. The intensities of the #uorite-type re#ec-
tion around 15% Mo, reported previously also by Buttery tions were integrated using the STOE software and the
X-RED (6)/X-SHAPE (7) programs used for an absorption
ꢁ ꢂ
ꢁ ꢂ
ꢂꢅ
ꢂ
ꢆ ꢆꢅ
correction. The JANA98 program (8) was used for the
structure re"nement. Electron di!raction (ED) patterns
*To whom correspondence should be addressed. Fax: 08-15 21 87. E-mail:
beb@inorg.su.se.
573
0022-4596/00 $35.00
Copyright ( 2000 by Academic Press
All rights of reproduction in any form reserved.

574
RAPID COMMUNICATION
were obtained with a transmission electron microscope
(TEM) JEOL 2000FX operating at a 200 kV accelerating
voltage. The samples investigated were crushed and disper-
sed in buthanol before they were placed on holey carbon-
coated copper grids. Energy dispersive X-ray spectroscopy
(EDS) analysis of the metal composition was made in a
JEOL SEM820 scanning electron microscope (SEM) with
a LINK AN10000 EDS system.
RESULTS
The synthesis yielded a homogeneous sample containing
large single crystals. For the EDS analyses in the SEM
a part of the material was molded in conducting plastic
which was surface polished. The metal composition was
determined from 20-point analyses to be Mo16(1):Bi84(1).
The determined composition agrees very well with the nom-
FIG. 1. ED pattern of Mo
image reveals both "rst- and second-order satellites along 11112* as well as
many cross terms re#ections, e.g., 111111.
Bi with zone axis 11102. The
O
ꢇꢈꢀꢃ ꢇꢈꢅꢉ ꢀꢈꢆꢉ
inal one and shows that no signi"cant Bi O loss by evapor-
ꢁ ꢂ
ation had occurred during the synthesis.
ED patterns, see Fig. 1, revealed satellite re#ections, of
several orders, to the basic #uorite re#ections. The satellite
A crystal with an average dimension of 40}50 lm was
re#ections could be indexed using six indices hklmnp and investigated by single-crystal XRD. A section of the recipro-
three q-vectors along the unit cell axes with equal lengths of cal space is given in Fig. 3 for comparison with the ED
*
ca. 0.29a in reciprocal space. The presence of cross term pattern in Fig. 1. The basic #uorite-type structure was
&
re#ections in all investigated crystallites excluded the possi- re"ned to a weighted R-value of 0.037 using 16 unique basic
bility of twinning. Since the satellites are spaced equally re#ections, an Mo/Bi ratio of 16:84 on the 4a sites (0, 0, 0)
along all space diagonals, it was concluded that the modula- and 7/8 (88%) of the 8c sites (1/4, 1/4, 1/4) occupied by
tion is three-dimensional and does not break the cubic O atoms, the latter occupancy calculated by charge balanc-
ꢃ>
ꢂ>
symmetry. The same systematic extinction conditions was ing, assuming Mo and Bi ions. The only parameters
observed for the satellites as for the main re#ections, with re"ned were two isotropic thermal displacement for respec-
ꢁ
hkl or mnp all odd or even. This, together with the fact that tively the metal atoms, ; "0.0564(4) A , and the
*ꢀꢁ
ꢁ
the point symmetry for the basic nonmodulated and O atoms, ; "0.186(6) A . The re"ned values show that in
modulated part of the structure must be equal, implies that the modulated structure the O atoms are signi"cantly dis-
*ꢀꢁ
the symmetry for both parts are Fm3m. No satellites could placed but that the metal atoms are close to the positions in
furthermore be observed that connect the real space three an ideal cubic #uorite-type structure.
dimensions with the three modulation dimensions, i.e., satel-
lites to systematically absent basic re#ections. This implies
that the 6-D structure is primitive and that the symmetry is
P:Fm3m:Fm3m. The strongest satellites are found in the
11112* directions (type hkl111) and the next-strongest ones
in the 12002* directions (type hkl200).
The Guinier}Hagg "lm data yielded a unit cell parameter
of a"5.6484(5) A, which equals that reported for pure
d-Bi O , a"5.6485(6) A (8). Using the approximate q-vec-
ꢁ ꢂ
*
tor length of e+0.29a , obtained from the ED patterns, it
&
was possible to index several satellite Bragg peaks in the
powder XRD pattern using six indices hklmnp, see Fig. 2,
and thereby derive a better estimate of e. Finally the unit cell
parameter and e were re"ned to a"5.6478(1) A and
*
\ꢀ
e"0.29256(6)a A
from 37 re#ections at 2h4843 by
&
nonlinear least squares. The indexed powder pattern for the
"rst 20 re#ections is given in Table 1. The strongest and
next-strongest satellite re#ections are of the kind hkl111 and
hkl200, in consistency with the ED data.
FIG. 2. Powder XRD pattern of Mo
lite re#ections in the inset.
Bi
O
with indexed satel-
ꢇꢈꢀꢃ ꢇꢈꢅꢉ ꢀꢈꢆꢉ

RAPID COMMUNICATION
575
TABLE 1
for both ED and XRD, increase with 2h angle in compari-
son with those of the main re#ections. This indicates that
the modulation has a substantial positional component that
probably involves both metal and O atoms. The substitu-
Powder X-ray Di4raction Pattern of Mo0.16Bi0.84O1.74
hklmnp
2h
ꢁ"ꢀ
*2h
d
(A)
I/I (&)
ꢁ"ꢀ
ꢇ
ꢂ>
tion of Bi ions by the much smaller and higher charged
000111
000200
000220
200400
111111
200220
111000
111220
111111
200000
111111
200111
220000
220111
222111
311000
222000
331111
400000
420111
7.903
9.097
!0.017
!0.051
!0.041
0.005
11.17
9.71
6.849
6.806
4.609
3.687
3.261
2.941
2.881
36
6
1
2
7
ꢃ>
Mo ions should e!ect local metal}O con"gurations that
di!er considerably from those in the average structure,
where each metal site is coordinated by eight O atoms at
a distance of 2.45 A. The local Mo}O distances in particular
should be much shorter. When discussing the O atom posi-
tions in the phases it should, however, be realized that the
12.915
12.997
19.241
24.121
27.319
30.538
31.005
31.653
35.562
37.041
45.380
46.139
47.580
53.786
56.385
64.145
66.128
67.839
0.004
0.017
1
!0.008
0.005
!0.004
!0.004
0.006
!0.009
!0.001
!0.007
0.005
!0.003
!0.003
!0.008
0.003
1000
1
12
411
4
6
394
5
O atom arrangement in d-Bi O itself has been shown not
to be like in an ideal cubic #uorite structure (11, 12). Since
ꢁ ꢂ
2.824
2.5223
2.4250
1.9969
1.9657
1.9095
1.7029
1.6304
1.4506
1.4118
1.3803
the scattering of X-rays and electrons are dominantly due
to the metal atoms in the structure, little can be inferred
from the present data about the behavior of the O atoms. It
has, however, been shown from neutron powder di!raction
data that for the Nb-Bi solid solution with composition
6
381
90
6
38
9
Nb
Bi
O
the O atoms exhibit in addition to an
ꢇꢈꢀꢅ ꢀꢈꢅꢁ ꢀꢈꢃꢅ
occupational modulation also a displacive modulation
along [111]* (10).
Attempts to determine the structure of Mo
!0.014
Bi
O
ꢇꢈꢀꢃ ꢇꢈꢅꢉ ꢀꢈꢆꢉ
Note. *2h"2h !2h
0.29256(6) A\ꢀ. Re#ection intensities originate from di!ractometer data.
. j"1.5406 A. a"5.6478(1) A and e"
from the collected single-crystal XRD data have so far not
been successful, mainly due to di$culties to accurately inte-
grate the intensities of the satellite re#ections. The high
symmetry of the structure makes, however, powder XRD
data nearly equally useful for a structure determination due
ꢁ"ꢀ
#!ꢂ#
DISCUSSION
The high-temperature phase Mo
Bi
O
has not
ꢇꢈꢀꢃ ꢇꢈꢅꢉ ꢀꢈꢆꢉ
been reported previously. A well-crystallized sample of the
orthorhombic low-temperature modi"cation was obtained
when remelting the quenched material at 10003C and cool-
ing the sample to 5003C at a rate of 53/h.
Few 6-D structures are known and only a small number
of them determined. The P:Fm3m:Fm3m symmetry derived
here for Mo
ꢇꢈꢀꢃ ꢇꢈꢅꢉ ꢀꢈꢆꢉ
observed previously only for Cu
for any oxide compound. A similar 6-D solid solution phase
is, however, found in the Nb}Bi}O system and a structure
model has been proposed for it (10). Its composition is
Bi
O
has to our knowledge been
Bi S (9) and not
ꢄꢈꢂꢆꢅ ꢇꢈꢄꢄꢁ ꢉ
Nb Bi
O
, 0.124x40.46, and the symmetry is
V
ꢀ\V ꢀꢈꢄ>V
P:Fm3m:Fd3m. The Nb and Mo phases have oxygen va-
cancy contents of 2}19% and 12.5%, respectively, and
stabilizing metal atom concentrations of 6}23% and 16%,
respectively. The q-vector for the Mo
Bi
O
phase
ꢇꢈꢀꢃ ꢇꢈꢅꢉ ꢀꢈꢆꢉ
spans in real space 3.418(1) unit cells or 19.307(4) A, while
the q-vector for the Nb}Bi solid solution is signi"cantly
shorter, 15.6}14.6 A. Further synthesis and structural work
is evidently needed to evaluate the dependence of the modu-
lations/di!use scattering on added atom type and concen-
tration. Possible displacive and occupational modulations
have been discussed for the Nb}Bi solid solution (10). A few
tentative inferences may however be made. In the case of
FIG. 3. A section of reciprocal space for single crystal XRD data in
a +110, plane, illustrating the intensity distribution for the basic #uorite
type and satellite re#ections. Up to second-order satellites and cross term
Mo
Bi
O
, the intensities of the satellite re#ections, satellite re#ections in 12002* directions are present.
ꢇꢈꢀꢃ ꢇꢈꢅꢉ ꢀꢈꢆꢉ

576
RAPID COMMUNICATION
to the fact that (i) peak overlap of re#ections with di!erent
hklmnp is small and (ii) that the multiplicity of satellite peaks
is high, which provides for a good determination of their
compounded intensities. A structure determination using
X-ray and neutron powder di!raction data is scheduled.
3. D. J. Buttery, D. A. Je!erson, and J. M. Thomas, Mater. Res. Bull. 21,
739 (1986).
4. K. E. Johansson, T. Palm, and P.-E. Werner, J. Phys. E: Sci. Instrum.
13, 1289 (1980).
5. STOE & Cie GmbH, Darmstadt, Germany, 1996, v. 1.07.
6. STOE & Cie GmbH, Darmstadt, Germany, 1996, v. 1.01.
7. V. Petrmcek and M. Dusek, Institute of Physics AVCR, Praha, Czech
Republic, 1997.
8. P. D. Battle, C. R. A. Catlow, J. W. Heap, and L. M. Moroney, J. Solid
State Chem. 63, 8 (1986).
ACKNOWLEDGMENTS
This study has been "nancially supported by the Swedish Natural
Science Research Council.
9. &&Aperiodic '94'' (G. Chapuis and W. Paciorek, Eds.), pp. 475}477,
World Scienti"c, Singapore, 1995.
10. R. L. Withers, C. D. Ling, and S. Schmid, Z. Kristallogr. 214, 296
(1999).
REFERENCES
11. P. D. Battle, C. R. A. Catlow, J. Drennan, and A. D. Murray, J. Phys.
C 16, L561 (1983).
12. L. E. Depero and L. Sangaletti, J. Solid State Chem. 122, 439
(1996).
1. J. C. Boivin and G. Mairesse, Chem. Mater. 10, 2870 (1998).
2. C. D. Ling, R. L. Withers, S. Schmid, and J. G. Thomson, J. Solid State
Chem. 137, 42 (1988).