X-ray, Raman and FTIRS studies of the microstructural evolution of zirconia particles caused by the thermal treatment

Article abstract of DOI:10.1016/j.jssc.2006.04.057

Genesis of the structure of zirconia particles prepared by precipitation of amorphous hydrated zirconia by ammonia from the ZrO(NO₃)₂ solution followed by a mild hydrothermal treatment (HTT) of precipitate, washing and calcination under air up to 1000 °C has been studied by X-ray diffraction (XRD), Raman and FTIRS. As revealed by FTIRS of lattice modes, the local structure of amorphous zirconia subjected to HTT is close to that in m-ZrO₂. This helps to obtain nearly single-phase monoclinic nanozirconia (particle size 5-15 nm) already after a mild calcination at 500 °C. Stability of this phase with nanoparticles sizes below the critical value determined by thermodynamic constraints is due to its excessive hydroxylation demonstrated by FTIRS. Dehydroxilation and sintering of these nanoparticles at higher (600-650 °C) temperatures of calcination leads to reappearance of the (111) cubic reflection in XRD patterns. Modeling of XRD patterns revealed that this phenomenon could be explained by polysynthetic (001) twinning earlier observed by HRTEM.

Full text of DOI:10.1016/j.jssc.2006.04.057

ARTICLE IN PRESS
Journal of Solid State Chemistry 179 (2006) 2965–2971
www.elsevier.com/locate/jssc
X-ray, Raman and FTIRS studies of the microstructural evolution of
zirconia particles caused by the thermal treatment
a,
Dmitry A. Zyuzin , Svetlana V. Cherepanova , Ella M. Moroz , Elena B. Burgina ,
Ã
a
a
a
a
Vladislav A. Sadykov , Victor G. Kostrovskii , Valerii A. Matyshak
b
c
a
Boreskov Institute of Catalysis SB RAS, pr. Akademika Lavrentieva 5, 630090 Novosibirsk, Russia
b
Institute of Solid State Chemistry SB RAS, 630090 Novosibirsk, Russia
c
Semenov Institite of Chemical Physics RAS, Moscow, Russia
Received 26 October 2005; received in revised form 26 February 2006; accepted 8 April 2006
Available online 22 May 2006
Abstract
Genesis of the structure of zirconia particles prepared by precipitation of amorphous hydrated zirconia by ammonia from the
ZrO(NO solution followed by a mild hydrothermal treatment (HTT) of precipitate, washing and calcination under air up to 1000 1C
has been studied by X-ray diffraction (XRD), Raman and FTIRS. As revealed by FTIRS of lattice modes, the local structure of
amorphous zirconia subjected to HTT is close to that in m-ZrO . This helps to obtain nearly single-phase monoclinic nanozirconia
particle size 5–15 nm) already after a mild calcination at 500 1C. Stability of this phase with nanoparticles sizes below the critical value
3 2
)
2
(
determined by thermodynamic constraints is due to its excessive hydroxylation demonstrated by FTIRS. Dehydroxilation and sintering
of these nanoparticles at higher (600–650 1C) temperatures of calcination leads to reappearance of the (111) ‘‘cubic’’ reflection in XRD
patterns. Modeling of XRD patterns revealed that this phenomenon could be explained by polysynthetic (001) twinning earlier observed
by HRTEM.
r 2006 Elsevier Inc. All rights reserved.
Keywords: Monoclinic zirconia; Hydrothermal treatment; Polysynthetic twinning; Whole XRD pattern modeling
1
. Introduction
chlorides) can be considered [9–11]. Any future practical
application of such a technique requires elucidation of the
features of zirconia phase composition and real/defect
structure evolution in the course of subsequent progressive
annealing under air at increasing temperatures. For the
catalytic application, the phase composition and the real/
defect structure of zirconia determine the surface density
and strength of Lewis and Broensted acid sites [4–8,12]
responsible for the reagents activation and/or anchoring of
supported active components. However, the details of the
real structure of these nanosized zirconia samples remain to
be poorly characterized, and factors controlling them are
not properly elucidated.
In a previous paper [13], evolution of the defect structure
of nanosized zirconia particles prepared by HTT of
precipitated amorphous zirconium hydroxide and calcined
at temperatures in the range of 300–1000 1C was followed
by using HRTEM. However, TEM is rather poorly
sensitive to the local coordination environment of Zr
Zirconia is one of the most refractory, shock-resistant
and corrosion-resistant oxides, which determines its broad
industrial application including construction ceramics,
corrosion-resistant and thermal-barrier coatings, gas sen-
sors, solid electrolytes, etc. [1,2]. Pure zirconia is also used
as a dispersed phase in both oxide and non-oxide ceramics
to increase fracture toughness, strength and hardness [3].
Dispersed zirconia attracts great attention as a catalyst or
as a support for metal/oxide catalysts of red–ox/acid–base
reactions [4–8].
As an attractive option for synthesis of dispersed
zirconia, the hydrothermal treatment (HTT) at moderate
(100–200 1C) temperatures of a hydrous zirconia precipi-
tated by ammonia from inorganic salts (nitrates or
Ã
Corresponding author. Fax: +7 383 3308056.
E-mail address: zyuzin@catalysis.nsk.su (D.A. Zyuzin).
0022-4596/$ - see front matter r 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.jssc.2006.04.057

ARTICLE IN PRESS
D.A. Zyuzin et al. / Journal of Solid State Chemistry 179 (2006) 2965–2971
2966
cations, which can be changed due to generation of defects.
In addition, TEM data by itself are not easily referred to
the average structural features of dispersed samples, so
there is a danger of overestimation of the occurrence of
some structural features revealed by this method. Hence,
the relative importance of different defects revealed by
TEM for the average structural characteristics of samples is
to be estimated by using X-ray (powder) diffraction
monoclinic phase was determined by using the external
¯
standard method and the integral intensities of (111) and
(111) diffraction peaks. Sample of ZrO calcined at 1000 1C
2
was chosen as such a standard. The X-ray particle sizes in
the [111] and [202] directions for monoclinic phase and in
the [111] direction for the cubic phase were estimated from
the FWHM of corresponding diffraction peaks.
Modeling of diffraction patterns for nanoparticles of the
monoclinic phase with a different real structure has been
carried out by using approaches developed elsewhere [15].
(
XRD). Similarly, spectroscopic methods (Raman, IR
spectroscopy) are well known to be much more sensitive
to the local distortion of coordination polyhedra in oxides
as compared with XRD. Hence, the aim of the second
paper of the series is to compare HRTEM data with the
results of the same system studies by XRD, Raman and
FTIRS of lattice modes.
ꢁ
1
Infra-red spectra of samples in the 200–4000 cm range
were registered using a BOMEM MB-102 spectrometer
ꢁ
1
with resolution 4 cm . Samples for measurements were
prepared by a standard procedure of pressing wafers
comprised of 3 mg of zirconia powder with 600 mg of CsI.
The Raman scattering measurements were carried out
using a 100/S-Bruker Raman Fourier Spectrometer. The
2
. Experimental
1.06 mm line of an Nd–YAG laser was used for excitation,
an output laser power was 100 mW.
Samples of zirconia were prepared by a method similar
to that described in [11]. As a starting compound,
commercial product ZrO(NO ) of a high purity grade
3
2
3. Results and discussion
was used. Freshly prepared 1.5 M solution was aged for
5 h at room temperature. Hydrolysis of this solution was
1
3.1. XRD
carried out at nearly constant pH value ꢀ9.5 by rapidly
adding it to the aqueous ammonia solution under intensive
stirring. Thus obtained sol was subjected to HTT in the
mother liquor in an autoclave first for 24 h at 90 1C and
then at 110 1C for 8 h.
Fig. 1 shows fragments of the diffraction patterns of
studied samples. Data on the qualitative phase analysis and
X-ray particle sizes are presented in Table 1. Due to a high
dispersion and superposition of peaks corresponding to
different zirconia phases, it was impossible to clarify the
space group of a highly symmetric phase, which will be
tentatively termed henceforth as a ‘‘cubic’’ one.
As follows from the data presented in Fig. 1 and Table 1,
there is a non-monotonous variation of the phase
composition with the temperature of calcination. On one
hand, as follows from the presence of the diffraction peak
at 2Y ¼ 30.251 and its intensity variation, the cubic phase
first practically disappears at 500 1C reappearing again
after annealing at 600 1C in an amount ꢀ20%. On the
After cooling suspension and filtering precipitate, the
latter was thoroughly washed by 0.1 M solution of NH OH
4
and dried at 110 1C. Dried sample was separated into parts
then calcined at 300, 400, 500, 560, 600, 650, 750 and
1
000 1C under air in a muffle furnace. The temperature
ramp includes heating from ꢀ25 1C to required tempera-
ture, keeping at this temperature for 4 h followed by a slow
cooling down in a muffle to room temperature after
turning-off the power.
XRD patterns were obtained by using HZG-4C dif-
fractometer by scanning in the 2Y range 10–701 with a step
of 0.051 and acquisition time in each point 10 s. CuKa
radiation monochromatized by passing through the gra-
phite monochromator situated in a diffraction beam was
used. Qualitative phase analysis has been carried out by
using JCPDS–ICDD files and ICSD/retrieve database.
Diffraction pattern treatment for the quantitative phase
analysis has been carried out by using the PowderCell for
Windows software and approximation of diffraction peaks
s
by Pseudo-Voight function in the Origin
software.
Estimation of the amount of monoclinic phase has also
been carried out by using the expression:
mon
11
mon
1
I
þ I
1
11
/mon
¼
,
cub
mon
1
mon
I
þ I111 þ I111
11
i
where I
is the integral intensity of the (111) peak of a
1
11
given phase estimated by using area of the corresponding
diffraction peak [14]. In addition, an amount of the
Fig. 1. Experimental diffraction patterns of zirconia samples calcined at
different temperatures.

ARTICLE IN PRESS
D.A. Zyuzin et al. / Journal of Solid State Chemistry 179 (2006) 2965–2971
2967
Table 1
X-ray data for samples of ZrO
2
calcined at different temperatures
Calcination temperature (1C)
Monoclinic ZrO
2
Cubic ZrO
2
˚
˚
˚
Content (%)
D
111 (A)
D
202 (A)
Content (%)
D
111 (A)
a
4
5
6
6
7
00
00
00
50
50
44 (39)
75
—
56
4
75
96 (92)
82 (97)
79 (99)
96 (97)
96 (100)
165
190
205
260
400
165
240
235
260
400
110
150
190
—
18
21
4
1
000
4
—
a
2
The amount of m-ZrO determined by the external standard method is given in parenthesis.
other hand, as judged by the monoclinic phase peaks
intensity, its amount is nearly constant under studied
annealing conditions being close to 100% as estimated by
using the external standard. This result agrees with the
TEM data [13] revealing presence of only m-ZrO particles
2
in a sample annealed at 500 1C and appearance of
fragments with a t-ZrO structure as thin slabs between
the twin boundaries in samples annealed at 650 1C.
2
For samples annealed at 400 and 500 1C, the X-ray
particle sizes of both m- and c-phases are practically
identical coinciding with those estimated by TEM [13].
This agrees with conclusions about the absence of extended
defects in zirconia nanoparticle even after c–m martensitic
transformation. For samples calcined at 600 and 650 1C, X-
ray particle sizes of the monoclinic phase estimated for
various directions were found to differ noticeably. Thus,
Fig. 2. The IR spectra of the lattice modes of zirconia samples calcined at
different temperatures.
¯
the (111) and (111) diffraction peaks situated in the 2Y
range 28–321 are broader than the (202) peak situated at a
higher diffraction angle (2Y ¼ 54.11). As a result, the
particle sizes calculated for the [111] direction are smaller
than those for the [202] direction. This suggests existence of
additional defects due to a specific real/defect structure of
those samples containing monoclinic phase, which can be
polysynthetic twins revealed by HRTEM [13].
to facilitate polymerization of tetrameric species through
their progressive dehydroxylation (oxolation) into less
symmetric 3D structures with a local coordination of Zr
resembling that in m-ZrO . Certainly, the presence of‘
2
more symmetric species in amorphous hydroxide is not
excluded as well, since broad IR bands do not allow to
make more detailed analysis. At least, a high amount of
water and hydroxyls in the amorphous hydroxide reflected
2
. FTIRS and Raman spectroscopy
In the IR spectrum of the amorphous hydroxide dried at
ꢁ
1
ꢀ
100 1C, practically the same spectral features as for
samples calcined at higher temperatures and containing
in the high intensity of IR bands at ꢀ3400 and ꢀ1600 cm
(Fig. 3) corresponding to stretching and bending vibration
modes of the water molecules [19] suggests that more
symmetric structures with a local coordination similar to
that in the tetrameric units are present as well.
exclusively m-ZrO are observed (Fig. 2). For the tetra-
2
gonal zirconia phase, where Zr atom is surrounded by eight
O atoms forming a distorted cube [17], only two bands at
ꢁ
1
ꢀ
460 and ꢀ670 cm corresponding to IR-active E and
After calcinations at 400 1C, the intensity of the water
bands strongly declines (Fig. 3). In the lattice mode range
(Fig. 2), IR bands at 232, 264, 354, 420–450, 501, 576 and
u
A
modes should be observed [16]. Hence, at least for a
u
2
part of Zr cations in the amorphous hydroxide subjected to
HTT and dried, their local coordination is apparently
strongly distorted resembling that in m-ZrO . In olygo-
merized band-like or sheet-like zirconium, polynuclear
hydroxospecies based on the tetramer [Zr (m-OH) (OH)
ꢁ
1
741 cm are clearly resolved, which can be assigned to the
monoclinic phase [20,21]. The Raman spectrum for this
sample (Fig. 4) is of a low intensity. However, lines at 180,
188, 223, 234, 305, 333, 345, 378, 475, 505, 534, 569, 618
2
4
8
8
ꢁ
1
(H O) ] ꢂ H O [18], Zr cations are also coordinated by
eight hydroxyls and water molecules. Hence, HTT appears
and 634 cm typical for the m-ZrO2 [20–25] can be
identified. In addition, less intense lines at 140–150 and
2
8
2

ARTICLE IN PRESS
D.A. Zyuzin et al. / Journal of Solid State Chemistry 179 (2006) 2965–2971
2968
and cation and anion vacancies) still remaining in the
lattice of both phases formed after rather soft dehydrox-
ylation. Hence, the nearest coordination sphere of Zr
cations in both phases present in this sample appears to be
rather distorted, as was earlier proposed for so-called t-
ZrO2 phase by Caracoche et al [27] on the bases of
perturbed angular correlation spectroscopy results. Indeed,
XAS spectroscopy data [17] revealed for the cubic and
monoclinic zirconia phase, a similar Zr–O polyhedron with
a sevenfold coordination and a comparable bond length,
along with a large distortion or splitting of the next-
nearest-neighbor Zr–Zr shell.
For sample calcined at 500 1C, the splitting of IR bands
ꢁ
1
at 355–374, 413–448 and 570–617 cm becomes appar-
ently (Fig. 2) driven by the progressive dehydroxylation
2
Fig. 3. The IR spectra in the H O range of zirconia samples calcined at
(Fig. 3) and nearly complete transformation of the cubic
different temperatures (1C).
phase into the monoclinic one (Table 1). On the whole, by
bands position and their relative intensity, this spectrum
corresponds to the spectrum of m-ZrO2 [16]. Again, a
rather small difference of this spectrum with the spectrum
of the sample calcined at 400 1C, despite nearly complete
disappearance of the cubic phase, agrees with the similarity
of the Zr–O coordination polyhedra in both phases. On the
contrary, the intensity of the Raman bands for this sample
is strongly enhanced (Fig. 4), their position mainly
corresponding to m-ZrO [16]. Weak bands at 146 and
2
ꢁ
1
56 cm typical for the t-ZrO are still observed, thus
2
2
suggesting existence of a small admixture of this phase.
For samples calcined at 600 and 650 1C, the IR spectra
differ only slightly from that of samples calcined at 500 1C
(Fig. 2), only narrowing of bands is observed, thus
suggesting the increase of the lattice ordering due to
removal of residual hydroxyls and annealing of point
defects. Some redistribution of the intensities of bands at
ꢁ
1
Fig. 4. The Raman spectra of zirconia samples calcined at different
temperatures.
413 and 448 cm is registered.
In contrast, strong variation is observed in the Raman
spectra of these samples (Fig. 4). While for the sample
annealed at 600 1C the intensity of lines corresponding to
the m-zirconia declines, the spectrum completely disap-
pears for the sample calcined at 650 1C. For the sample
annealed at 600 1C, weak lines at ꢀ136, ꢀ270 and
ꢁ
1
270 cm characteristic for the t-ZrO phase [16,20–26]
ꢀ
2
are observed. The most intense Raman line of the cubic
ꢁ
1
phase at ꢀ630 cm , even if present, could not be
ꢁ
1
apparently observed being superimposed with the more
ꢁ
ꢀ640 cm appear which are close by position to those
1
intense lines of the m-zirconia at ꢀ618 and 634 cm
.
typical for the t-ZrO phase. This agrees with the increase
2
Hence, this sample appears to be a mixture of the
tetragonal and/or cubic and monoclinic phases. Though
this conclusion broadly agrees with the results of the X-ray
phase analysis (Table 1) and TEM data [13], some further
specificity can be inferred from more detailed data
comparison. Thus, broad IR bands and weak Raman
spectrum for this sample suggest its strong disordering, first
of all, at the level of the nearest coordination sphere and
the unit cell characterized by these spectral methods. At the
same time, TEM data revealed a rather good crystallinity
of nanoparticles of both phases, and an amorphous phase
was not detected by XRD as a typical halo in the
diffraction patterns. This suggests that disordering of both
phases is due to point defects (namely, residual hydroxyls
of the content of t-ZrO phase for these samples by XRD
2
(Table 1) and TEM [13] data. Complete disappearance of
the Raman spectrum for the sample with a high density of
polysynthetic twins is quite surprising and has no analogs
in the accessible literature on zirconia. However, similar
phenomena were observed when studying the Raman
spectra of ceria doped by rare-earth cations [28]. In these
systems, the increase of the dopant content leading to
progressive ordering of the dopants and oxygen vacancy
distribution in the lattice was also accompanied by the
disappearance of the Raman spectra while all XRD data
¯
were well fitted by the cubic Fm3m space group. Hence,
some ordering in the defects distribution in the lattice is
indeed able to cause disappearance of the Raman spectra.

ARTICLE IN PRESS
D.A. Zyuzin et al. / Journal of Solid State Chemistry 179 (2006) 2965–2971
2969
This suggests that in a similar way ordered defects—
polysynthetic twins observed by TEM in zirconia samples
calcined at 600–650 1C are able to suppress the Raman
spectra.
In agreement with this hypothesis, disappearance of
polysynthetic twins in samples calcined at 750 and 1000 1C
recovers the Raman spectra, while the IR bands position
and splitting remain practically unchanged (Figs. 2 and 4).
units. Coordinates x , y n z were derived by corresponding
i
i
i
calculation from the atom coordinates in the monoclinic
zirconia unit cell. In each modeling series, two types of
layers were used, including one described above (layer 1)
and another related by the mirror reflection in the (001)
plane (layer 2). The layer thickness was constant, and, as
variables, the relative number of layers and their stacking
order were used.
Fig. 6 shows model diffraction patterns with different
alternation probabilities of layers 1 and 2 at the each layer
thickness equal to the c parameter of the monoclinic phase
unit cell (C ¼ c) and an equal number of both types of
layers (W ¼ W ¼ 0.5). In the case of the strict alternation
3
. Modeling of XRD patterns
For dispersed materials, appearance of new peaks is
known to be often generated by their disordering [29]. The
1
2
˚
˚
order (after layer 1 only layer 2 appears, and vice versa,
monoclinic structure of ZrO (a ¼ 5.315 A, b ¼ 5.208 A,
2
˚
P
11 ¼ 0), the diffraction pattern contains a peak at position
corresponding to the (111) peak of the cubic phase (Fig.
.1). In this model diffraction pattern, the peaks corre-
c ¼ 5.148 A; b ¼ 99.2o, S.G. P2 /c) [30] can be considered
1
%
as comprised of alternating layers of oxygen and zirconium
ions situated along the (001) plane. For samples calcined at
6
¯
sponding to (111) and (111) reflections of the monoclinic
phase are absent. As P11 increases, the diffraction peak
broadens (Fig. 6.2 and 6.3), and when only identical layers
alternate (the twinned particle contains only one twinning
plane), the diffraction pattern corresponds to the mono-
clinic phase (Fig. 6.5). Hence, at a given layer thickness, it
is impossible to obtain the diffraction pattern containing
reflections of both cubic and monoclinic phases.
6
00–650 1C, HRTEM data revealed [13] polysynthetic
twins, so that crystallites are comprised of the alternating
slabs with a mirror symmetric structure. Here, the
thickness of slabs, their stacking and alteration can vary.
By using the program for simulation of the diffraction
patterns of layered structures [15], diffraction patterns
from polysynthetically twinned particles of the monoclinic
˚
ZrO with sizes up to 300 A have been computed.
2
For the triplicate layer thickness (C ¼ 3c), keeping
number of different layers equal (W ¼ W ¼ 0.5), diffrac-
The starting data for calculation include distribution of
atoms in two-dimensionally ordered layers; number of
types of different layers (W ); their stacking order
determined by the probability of appearance of a K-type
layer after a M-type layer (PKM).
1
2
tion peaks corresponding by position to both monoclinic
and cubic phases are observed (Fig. 7.1). With the increase
of P , the peak corresponding to the cubic phase is
N
11
smeared, though its integral intensity is preserved until its
complete disappearance (Fig. 7.2–4). Only when the
relative number of different layers in the particle is changed
In the two-dimensional ordered layer, the distribution of
atoms was determined by parameters of the two-dimen-
sional unit cell (A, B, angle b), the layer thickness C,
coordinates x , y , z in the relative units with respect to A,
(
W ¼ 0.3; W ¼ 0.7) while the layer thickness is preserved,
1
2
i
i
i
the ratio of the integral intensities of ‘‘cubic’’ and
‘monoclinic’’ diffraction peaks decreases (Fig. 8). The
B, C parameters.
‘
In our case, for the direction of the uniaxial disordering,
the [001] direction was chosen as a direction of the oxygen
layers stacking (Fig. 5). The unit cell parameters of the
two-dimensional layer—A, B, and b were equal to
parameters a, b, and b of the monoclinic ZrO structure.
2
In different series of modeled diffraction patterns, the layer
thickness C was equal to 1, 2 or 3 monoclinic structure
Fig. 6. Simulated diffraction patterns for model particles of zirconium
dioxide: A ¼ a, W ¼ 0.5: 1—P11 ¼ 0, 2—P11 ¼ 0.5, 3—P11 ¼ 0.6, 4—
¼ 0.8, 5—P ¼ 1.0.
1
Fig. 5. The layered structure of the oxygen sublattice of monoclinic ZrO
2
.
P
11
11

ARTICLE IN PRESS
D.A. Zyuzin et al. / Journal of Solid State Chemistry 179 (2006) 2965–2971
2970
Fig. 9. Fragment of the experimental diffraction pattern of the sample
calcined at 600 1C (1) in comparison with the theoretical curve for the
model twinned particles (2).
Fig. 7. Simulated diffraction patterns for model particles of zirconium
dioxide: C ¼ 3c, W ¼ 0.5: 1—P11 ¼ 0, 2—P11 ¼ 0.3, 3—P11 ¼ 0.5, 4—
11 ¼ 0.8.
1
P
explained as emerging due to multiply twinned particles of
the monoclinic phase.
4. Conclusions
A mild HTT of a precipitated hydrous zirconia makes its
local structure similar to that in m-ZrO , apparently by
2
facilitating polymerization of polynuclear Zr hydroxocom-
plexes based on a tetramer [Zr (m-OH) (OH) ꢃ (H O) ] ꢂ
4
8
8
2
8
H O via olation and oxolation. This allows to obtain, at
2
low (ꢀ400–500 1C) calcination temperatures, nanosized m-
ZrO phase with the particle sizes below the critical value
2
determined by the simple end-point thermodynamic
analysis of Garvie [31]. The relative stability of these
particles to sintering and transformation to c/t-ZrO is
2
Fig. 8. Simulated diffraction patterns for model particles of zirconium
dioxide: C ¼ 3c, W ¼ 0.3: 1—P11 ¼ 0, 2—P11 ¼ 0.05, 3—P11 ¼ 0.1, 4—
11 ¼ 0.2, 5—P11 ¼ 0.3.
ensured by a high content of residual water/hydroxyls in
the structure. At higher calcination temperatures, a
1
P
‘
‘pseudocubic’’ (111) XRD peak reappears, though the
amount of the monoclinic phase determined by using the
external standard remains nearly constant. Modeling of the
XRD diffraction patterns by using approaches earlier
developed for layered structures allowed to explain this
phenomenon by the polysynthetic mirror twinning along
increase of P11 from 0 to 0.3 also causes broadening and
disappearance of the ‘‘cubic’’ diffraction peak.
Diffraction patterns the most close to the experimental
one are ensured by models with the next parameters:
A ¼ 3a, W ¼ 0.3, W ¼ 0.7, P ¼ 0–0.1. However, even
the (001) oxygen planes revealed by HRTEM [13]. These
m
periodic planar defects and associated strains affect the
mode of ZrO coordination polyhedra ordering in slabs
1
2
11
in this case, complete coincidence of the theoretical and
experimental diffraction patterns is not achieved (Figs. 8
and 9), since modeling is based on invariable monoclinic
unit cell parameters. According to TEM data [13], the
monoclinic structure is distorted due to twinning (the
monoclinic angle and the unit cell parameters are changed
from layer to layer). A due regard for these variations is
expected to improve agreement between the simulated and
experimental diffraction patterns.
7
between the adjacent twin planes, which is reflected in the
disappearance of the Raman spectra. At higher (4700 1C)
calcination temperatures polysynthetic twins as well as
associated features in the XRD diffraction patterns
disappear while the Raman spectra are restored. Hence,
in contrary to single crystals and thin films where those
defects are known to be stable, in pure nanozirconia,
polysynthetic twins are observed only in the process of the
real structure evolution caused by progressive annealing/
sintering, which suggests their relative high excess energy
caused by their structural specificity.
Hence, experimental diffraction patterns for samples
calcined at 600 and 650 1C (Fig. 1) formally corresponding
to a mixture of the cubic and monoclinic phases can also be

ARTICLE IN PRESS
D.A. Zyuzin et al. / Journal of Solid State Chemistry 179 (2006) 2965–2971
2971
References
[17] P. Li, I.-W. Chen, J.E. Penner-Hahn, Phys. Rev. B. 48 (1993) 10063.
[
[
18] L.M. Zaitsev, Russ. J. Inorg. Chem. (Engl. Transl.) 11 (1966) 900.
19] P.N. Kuznetsov, L.I. Kuznetsova, A.M. Zhizhaev, S.M. Kolesniko-
va, G.L. Pashkov, V.V. Boldyrev, Zh. Neorg. Khim. 47 (2002)
[
[
1] J.B. Goodenough, Nature 404 (2000) 821.
2] T. Ishikawa, H. Yamaoka, Y. Harada, T. Fujii, T. Nagasawa, Nature
4
50.
4
16 (2002) 64.
3] R.H.J. Hannink, P.M. Kelly, B.C.J. Muddle, J. Am. Ceram. Soc. 83
2000) 461.
[
[
[
[
[
[
20] C.M. Phillippi, K.S. Mazdiyasni, J. Am. Ceram. Soc. 54 (1971) 254.
21] A. Feinberg, C.H. Perry, J. Phys. Chem. Solids 42 (1981) 513.
22] C.H. Perry, D.-W. Liu, R.P. Ingel, J. Am. Ceram. Soc. 68 (1985) 184.
23] B.-K. Kim, J.-W. Hahn, K.R. Han, J. Mater. Sci. Lett. 16 (1997) 669.
24] M. Ishigame, T. Sakurai, J. Am. Ceram. Soc. 60 (1977) 367.
25] Yu.K. Voron’ko, A.A. Sobol’, S.N. Ushakov, L.I. Tsymbal, Neorg.
Mater. 3 (1994) 803.
[
(
[
[
[
[
[
4] K. Tanabe, Mater. Chem. Phys. 13 (1985) 347.
5] Y. Nakano, T. Lizuka, H. Hattori, K.J. Tanabe, J. Catal. 57 (1979) 1.
6] M.M.V.M. Souza, M. Schmal, Appl. Catal. A: General 255 (2003) 83.
7] A.T. Bell, Stud. Surf. Sci. Catal. 136 (2001) 13.
8] O.V. Metelkina, V.V. Lunin, V.A. Sadykov, G.M. Alikina, R.V.
Bunina, E.A. Paukshtis, V.B. Fenelonov, A.Yu. Derevyankin, V.I.
Zaikovskii, U. Schubert, J.R.W. Ross, Catal. Lett. 78 (2002) 111.
9] G.K. Chuah, S. Jaenicke, B.K.J. Pong, J. Catal. 175 (1998) 80.
[
[
[
26] M. Li, Zh. Feng, G. Xiong, P. Ying, Q. Xin, C. Li, J. Phys. Chem. B.
1
05 (2001) 8107.
27] M.C. Caracoche, P.C. Rivas, M.M. Cervera, R. Caruso, E.
Benavides, O. de Sanctis, S.R. Mintzer, J. Mater. Res. 18 (2003) 208.
28] V.A. Sadykov, T.G. Kuznetsova, G.M. Alikina, Y.V. Frolova, A.I.
Lukashevich, Y.V. Potapova, V.S. Muzykantov, V.A. Rogov, V.V.
Kriventsov, D.I. Kochubei, E.M. Moroz, D.A. Zyuzin, V.I.
Zaikovskii, V.N. Kolomiichuk, E.A. Paukshtis, E.B. Burgina, V.V.
Zyryanov, N.F. Uvarov, S. Neophytides, E. Kemnitz, Catal. Today
[
[
10] E. Tani, M. Yoshimura, Sh.S o¯ miya,, J. Am. Ceram. Soc. 66 (1983)
1.
11] G.K. Chuah, Catal. Today 49 (1999) 131.
12] S. Benref, E.J. Knozinger, Mater. Chem. 9 (1999) 1203.
1
[
[
[
¨
13] V.A. Sadykov, V.I. Zaikovskii, D.A. Zyuzin, E.M. Moroz, E.B.
Burgina, A.V. Ishchenko, V.G. Kostrovskii, V.A. Matyshak, Mater.
Res. Soc. Symp. Proc. 878E (2005) Y4.8.1-6.
9
3–95 (2004) 45.
[
[
[
29] V.A. Drits, C. Tchoubar, X-ray Diffraction by Disordered Lamellar
Structures, Springer, Berlin, 1990.
[
14] R.C. Garvie, P.S. OLE_LINK1Nicholson, J. Am. Ceram. Soc. 55
(1972) 303.
30] B. Bondars, G. Heidemane, J. Grabis, K. Laschke, H. Boysen, J.
Schneider, F. Frey, J. Mater. Sci. 30 (1995) 1621.
31] R.C. Garvie, J. Phys. Chem. 82 (1978) 218.
[
[
15] S.V. Cherepanova, S.V. Tsybulya, J. Mol. Catal. 158 (2000) 263.
16] P. Bouvier, H.C. Gupta, G.J. Lucazeau, J. Phys. Chem. Solids 62
(2001) 873.