Full text of DOI:10.1557/JMR.2000.0220

Structural characterization of ceria–zirconia powder catalysts
prepared by high-energy mechanical milling:
A neutron diffraction study
S. Enzo and F. Delogu
Istituto Nazionale di Fisica per la Materia e Dipartimento di Chimica dell’Universita` di Sassari,
via Vienna 2, 07100 Sassari, Italy
R. Frattini
Istituto Nazionale di Fisica della Materia e Dipartimento di Chimica Fisica dell’Universita` di
Venezia, Dorsoduro 2137, 30123 Venezia, Italy
A. Primavera and A. Trovarelli
Dipartimento di Scienze e Tecnologie Chimiche, Universita` di Udine, via Cotonifico 108,
33100 Udine, Italy
(Received 3 August 1999; accepted 10 April 2000)
Neutron diffraction measurements were carried out on samples of CeO2–ZrO2 powder
catalysts prepared by high-energy mechanical milling. The formation of solid solution
was evidenced across the entire composition range examined. Quantitative phase
evaluation by the Rietveld method indicated formation of tetragonal structure for low
CeO2 content, whereas cubic solid solutions were the stable form at high CeO2
loading. In addition, a pseudocubic or tetragonal tЉ cell with axial ratio of unity and
with internal deformation of the oxygen sublattice was observed at intermediate
composition (50 mol% CeO2). Thermal annealing up to 1000 °C showed expansion of
the unit cell parameters; an increase in the degree of tetragonality at the expense of
cubic and monoclinic phase was observed for composition CexZr1−xO2 (x < 0.5).
mining their oxygen storage/release behavior. In
I. INTRODUCTION
particular, the substitution of Zr for Ce favors the reduc-
tion behavior of solid solution with a cubic cell because
of the radius of the oxygen migration channel and the
creation of structural defects.4 However, solid solu-
tions with tetragonal symmetry seem to be less affected
by this process.
Materials containing ceria and zirconia have become
the focus of an intense study over the last few years, as
a result of several applications in the field of high-
temperature materials for the current technology of fuel
cells and oxygen gas sensors,1 and as components of
auto-exhaust catalysts for the removal of noxious com-
pounds.2 Although in some cases (such as auto-exhaust
catalysis) the application has already reached the com-
mercial production stage, there are several fundamental
questions and points that still need clarification.
It has been reported that one of the main features that
contributes to the success of ceria–zirconia solid solu-
tions in the formulation of auto-exhaust catalysts is the
enhancement of oxygen transport properties which fol-
lows the introduction of Zr into the cubic lattice of ceria.
This allows better oxygen exchange between the solid
and the gas phases, thus enabling greater efficiency in the
removal of CO, NO, and hydrocarbons under various
driving conditions.3 The structural features of ceria–
zirconia solid solutions play an important role in deter-
The formation of solid solutions with different struc-
tures in the CeO2–ZrO2 system is mainly regulated by the
methodology employed for preparation, and by compo-
sition (i.e., Ce/Zr ratio), although the exact nature of the
phase diagram and transformation dynamics is still a
matter of debate. Generally, for CeO2 contents lower
than 10 mol%, monoclinic symmetry is preferred but for
CeO2 contents higher than 80 mol%, the phase diagram
shows a monophase region of cubic symmetry.5 In the
intermediate region, the exact nature of the phases is still
unclear, because of the stable and metastable phases of
tetragonal symmetry present. Yashima et al. reported the
formation of three different tetragonal phases, named
t, tЈ, and tЉ:6 the t is a stable phase formed through high-
temperature diffusional phase decomposition, the tЈ form
1538
J. Mater. Res., Vol. 15, No. 7, Jul 2000
© 2000 Materials Research Society
http://journals.cambridge.org
Downloaded: 25 Mar 2015
IP address: 169.230.243.252

S. Enzo et al.: Structural characterization of ceria–zirconia powder catalysts prepared by high-energy mechanical milling
is obtained through a diffusionless transition and is meta-
stable with axial ratio c/af > 1, and the tЉ form is also
metastable but with c/af ס 1.7
transition occurs by oxygen ion displacement from
the ideal fluorite sites (8c sites in space group Fm3m),
giving a tetragonal structure whose axial ratio c/a is close
to √2 (tЉ phase) since af ס a√2 (see Fig. 1).7 Therefore,
we used neutron diffraction techniques12 to better distin-
guish between cubic and tetragonal structure during the
preparation of ceria–zirconia (i.e., evolution of the cell
parameters during milling) and to follow the structural
transformation of samples during thermal treatments.
This also permitted the determination of the atomic
coordinates of oxygen, which are important for a com-
plete understanding of the phase transformations13 and
chemical reactivity14 induced on introducing Zr into the
CeO2 lattice.
It has recently been reported that high-energy me-
chanical milling (MM) is an effective method for the
preparation of ceria–zirconia mixed oxides with a good
degree of chemical homogeneity over almost the entire
composition range.8,9 Preliminary x-ray-diffraction
(XRD) studies pointed out the formation of a cubic solid
solution for CeO2-rich samples, (CexZr1−xO2, with
x > 0.5) and an increasing degree of tetragonality for
higher ZrO2 loading.9 Because it has been reported that
solid solutions having a cubic cell behave better than
those with tetragonal symmetry,4 it is important to fully
characterize the structural features of these materials.
The various reports on the structural transformations
induced on oxides by MM are not conclusive and a com-
plete understanding of structural transformation occur-
ring under MM has yet to be acquired.10,11 One of the
reasons for these discrepancies is that the nature of end
products depends on the energetic conditions of the MM
apparatus employed, such as oscillation frequency, num-
ber and composition of colliding balls, and ball density.
In addition, XRD data for ceria–zirconia cannot be used
satisfactorily to distinguish between cubic and tetragonal
structures, particularly when the cubic-tetragonal phase
II. EXPERIMENTAL
CeO2 (99.999%, surface area 7 m2/g) was purchased
from Aldrich (Milwaukee, WI). ZrO2 was obtained by
calcination of hydrous zirconia (MEL Chemicals,
Manchester, U.K.) at 973 K (surface area 27 m2/g).
Preparation of mixed oxides was carried out with a high-
energy vibratory ball mill (Spex 8000, Spex Industries,
Inc., Edison, NJ) at an oscillation frequency of 20 Hz and
an amplitude of approximately 20 cm. Powders were
loaded into a 50-ml ZrO2 vial equipped with six Y-
doped, high-wear-resistant zirconia balls (⌽ ס 10 mm,
Tosoh Corporation, Tokyo, Japan). The ball-to-powder
ratio was 18/1 (i.e., 18 g of balls to 1 g of powder) and
milling time ranged from 1 to 9 h. Three different com-
position were examined: a mixture of 20 mol% CeO2 and
80 mol% ZrO2 (hereafter referred to as CZ20), a mixture
of 50 mol% CeO2 and 50 mol% ZrO2 (CZ50), and a
mixture of 80 mol% CeO2 and 20 mol% ZrO2 (CZ80).
Neutron diffraction measurements were made at ISIS
neutron source, Rutherford Appleton Laboratory (RAL),
Chilton, U.K., on liquid and amorphous diffractometer
(LAD). The specimens were placed in thin-walled vana-
dium tubes with 6-mm internal diameter. The annealing
treatment for the 9-h MM specimens was carried out at
different temperatures from 600 to 1000 °C in a furnace
under vacuum on the LAD instrument; each diffraction
pattern was collected for about 2 h. The data were ana-
lyzed by the Rietveld method15 with FULLPROF code,16
according to the methodology reported in our previous
publications.17 The neutron small-angle scattering (SAS)
experiments were made on the low Q (LOQ) instrument,
also at the ISIS neutron source. The samples were con-
tained in Helma quartz cells and the SAS data were ana-
lyzed using the Colette package on the LOQ instrument.
III. RESULTS AND DISCUSSION
Figure 2(a) shows the neutron diffraction patterns
(data points) and the Rietveld fit (solid lines) of CZ20
mechanically treated for 0 (parental mixture), 1, and 9 h
FIG. 1. Spatial relations between (a) cubic and (b) tetragonal tЉ cell of
(Ce,Zr)–O2. The open circles are the cations and the solid circles are
the anions.
J. Mater. Res., Vol. 15, No. 7, Jul 2000
Downloaded: 25 Mar 2015
1539
IP address: 169.230.243.252
http://journals.cambridge.org

S. Enzo et al.: Structural characterization of ceria–zirconia powder catalysts prepared by high-energy mechanical milling
respectively, in the Q range 1.8–10 Å−1. The bars at the
bottom mark the peak occurrences of the three phases
used to fit the parental powder pattern. As may be seen,
the ZrO2 component comprises tetragonal (space group
P42/nmc) and monoclinic (space group P21/a) poly-
morphs. In contrast, CeO2 is present in its stable cubic
form (space group Fm3m). At this stage, the lattice pa-
rameters of phases refined by fit are in agreement with
literature values (a ס 5.411 Å for CeO2 JCPDS 34-593;
af ס 3.6204 Å, c ס 5.25 Å for t-ZrO2, JCPDS 17-923;
and a ס 5.1477 Å, b ס 5.2030 Å, c ס 5.3156 Å, ␤ ס
99°23Ј for m-ZrO2, JCPDS 13-307).18
Quantitative phase evaluation obtained by the Rietveld
approach is reported in Fig. 2(b). Initially, ZrO2 appeared
mainly in the monoclinic polymorphic form and cubic
CeO2 was present in concentration close to the nominal
value. After 1-h milling, the presence of cubic CeO2
decreased markedly, while the increase in the tetragonal
ZrO2-containing phase occurred at the expense of the
monoclinic component. This trend was confirmed in the
specimen mechanically treated for 9 h, where the cubic
component totally disappeared from the pattern and
about 90 wt% of the sample was tetragonal, the rest being
monoclinic. In any case, the tetragonal form was cer-
tainly a solid solution, on the basis of the noticeable
lattice parameter changes that can be identified from
the evident shift of the most intense peaks in Fig. 2(a).
This point will be discussed later. Note also that in
Fig. 2(a) there is a considerable line broadening, already
established after 1-h ball milling; this is due to the si-
multaneous effect of fragmentation and mechanically in-
duced lattice disorder. The two contributions may be
separated using a procedure illustrated in Ref. 17. This
shows that initially the broadening due to lattice dis-
order prevails whereas, at a later stage of milling,
fragmentation phenomena are increasingly important
( D
200 Å).
The axial ratio c/a of tetragonal phase lattice param-
eters, obtained by fitting the diffraction pattern of pow-
ders after 9-h milling, is 1.013*√2. This tetragonal phase
is named t when formed by diffusion and tЈ when for-
mation is diffusionless. It has been suggested that the
phase transition from c to tЈ is accompanied by oxygen
displacement from the ideal fluorite site (see Fig. 1).
In our samples, the coordinate of oxygen zO was deter-
mined as 0.05.
The neutron small angle scattering data are reported in
Fig. 3 as a bilogarithmic plot. As may be observed, the
starting mixture powder shows a two-stage regime and
the outward part of the curve has a slope s ס −4, as
predicted by the classic Porod law. In contrast, mechani-
cally treated specimens show a single monotonous re-
gime with s ס −3.2, very distinct from Porod behavior
and indicative of established fractal-like, dissipative phe-
nomena. It is conceivable that these phenomena might
FIG. 2. (a) Neutron diffraction patterns (data points) and Rietveld fit
(solid lines) for CZ20 ball-milled specimens for time indicated. Bars at
bottom refer to peak phase position, calculated by lattice parameters
obtained for unmilled (0-h) specimen and very similar to literature
values. (b) Phase percentage evolution as a function of milling time;
triangles refer to cubic phase, circles to monoclinic phase, and squares
to tetragonal phase. Data interpolation was performed using exponen-
tial functions.
1540
J. Mater. Res., Vol. 15, No. 7, Jul 2000
Downloaded: 25 Mar 2015
http://journals.cambridge.org
IP address: 169.230.243.252

S. Enzo et al.: Structural characterization of ceria–zirconia powder catalysts prepared by high-energy mechanical milling
FIG. 3. Small-angle neutron scattering for CZ20 specimen in biloga-
rithmic scale. Data points refer to 0-h sample. Curve slope is −3.83;
crosses refer to specimen mechanically milled for 1 h, s ס −3.14;
triangles refer to 9-h MM specimen, s ס −3.22.
also be linked to interdiffusivity, on the basis of the pre-
viously mentioned lattice parameter changes and the for-
mation of solid solutions.
Figure 4(a) shows the neutron diffraction patterns
(data points) and the Rietveld fit (solid line) of CZ50
powders at different milling times. In this case, too, the
starting material can be satisfactorily described by a Riet-
veld fit (Rwp 10%) performed with cubic CeO2, te-
tragonal and monoclinic ZrO2 with a lattice parameter
close to literature-reported values. The phase percentage
obtained by Rietveld fit for this pattern, and its evolution
as a function of milling time, is reported in Fig. 4(b). An
increasing trend of tetragonal phase is evident, together
with a decrease in monoclinic form. In addition, the lat-
tice parameter of the tetragonal ZrO2-based phase in-
creases as a function of MM time. Conversely, for the
cubic phase, after a small increment in the lattice param-
eter from 5.413(1) Å (0 h) to 5.419(6) Å (1 h), there is a
decrease to a ס 5.290(8) Å after 9-h MM. It may there-
fore be surmised that both cubic and tetragonal mixed
FIG. 4. (a) Neutron diffraction patterns (data points) and Rietveld fit
(solid lines) for CZ50 specimens ball-milled for time indicated. (b)
Phase percentage evolution as a function of milling time. Notations as
in Fig. 2.
CexZr1−xO2 solid solutions are formed. By assuming a
Vegard-type law for the entire cubic phase composition
range, we estimated the concentration of Ce to be about
58% at the end of the MM process. This implies that,
assuming the remaining monoclinic zirconia to be pure,
the amount of Zr atoms in the tetragonal phase is about
67%. However, there may be deviation from an ideal
Vegard-type law in our system and this point will be
discussed in greater detail later.
The ratio c/a for the tetragonal phase is 1.417 (a value
close to √2); this tetragonal form is therefore convention-
ally denominated tЉ. It is interesting to note here that,
according to recent findings, the transition from cubic to
J. Mater. Res., Vol. 15, No. 7, Jul 2000
Downloaded: 25 Mar 2015
1541
IP address: 169.230.243.252
http://journals.cambridge.org

S. Enzo et al.: Structural characterization of ceria–zirconia powder catalysts prepared by high-energy mechanical milling
pseudocubic or tetragonal tЉ cell for ceria–zirconia is
expected to occur at compositions containing around
80% ceria.5 The stability of this phase spans a region of
CexZr1−xO2 composition from 0.65 < x < 0.8. Other au-
thors have reported that the formation of tЉ phase extends
beyond this region up to compositions with x ס 0.5.19 It
is clear that the relationship between composition and
phase formation in this region requires serious further
study. However, owing to the metastable nature of the
phases formed, transition from one structure to the other
may also be critically influenced by particle dimensions,
and therefore by the methodology used for preparation.20
The “state-of-the-art” phase diagram for this system is
the one reported by Yashima et al.5 These authors used
high-temperature firing to prepare solid solutions that
have a very small surface area. In our case, the low
temperature used in the synthesis, and the continuous
bond rupture/formation typical of MM, produced a ma-
terial having a higher surface area with a particle size in
the range of 100–170 Å.21 This could explain the stabi-
lization of the tЉ phase even at a lower Ce content. The
formation of this phase (which was previously identified
as cubic because of the cation lattice, which remains
essentially in a fluorite structure)9 is characterized by the
shift of oxygen atoms from their ideal position in an
ordered fashion, which induces a distorted oxygen sub-
lattice. The presence of these modifications can be high-
lighted by Raman spectroscopy,5 which cannot, however,
give a quantitative estimate of the topological lattice dis-
placement. In our case, this additional parameter was
calculated as zO ס 0.04 (corresponding to an oxygen
displacement of 0.21 c units along the z direction, if com-
pared to the position of oxygens in ceria where zO is
0.25). A comparison with the only quantitative data ex-
isting in the literature on such oxygen shift in ceria–
zirconia (for CZ50, zO ס about 0.08, corresponding to an
oxygen displacement of about 0.17)13 indicates a higher
degree of distortion in our samples, which may be in-
dicative of the type of preparation and/or the higher sur-
face area of our sample. This anion lattice distortion, in
conjunction with the axial ratio value c/af ס 1, indicates
that the phase is closely related to a cubic cell, with a
large shift of oxygen atoms from their ideal positions.
This may explain why the introduction of Zr into the
CeO2 lattice is beneficial to catalytic activity up to com-
positions close to CexZr1−xO2 with x ס 0.5, where opti-
mal activity is observed.3 The benefit could be the result
of the high oxygen transport properties typical of the
fluorite lattice, coupled with the presence of structural
defects (like oxygen distortions) which enhance diffusion
of oxygen. Transport properties are indeed believed to be
a key factor in the enhancement of the oxygen storage/
release capacity of these materials under catalytic condi-
tions.4,22,23
The broadening of peaks, attributed to an increment in
internal strain and to the decrease in crystallite size, is
also evident for this composition. However, in this case
the effects are observed after 1-h MM. No significant
differences in the peak width parameters U and W are
found between samples milled for 1 and 9 h. The fact that
fragmentation and disorder processes are almost com-
pleted after 1-h milling is confirmed by the small-angle
measurements (Fig. 5), where no appreciable differences
between the patterns of 1- and 9-h specimens are ob-
served (s ס −3.1).
Figure 6(a) shows the neutron diffraction patterns of
the CZ80 specimen. Because the ceria component is now
preponderant, the remaining ZrO2 phase can be ac-
counted for by the monoclinic form alone, which seems
to disappear almost entirely after 1-h mechanical
treatment. A satisfactory fit (Rwp ס 10%) for the sam-
ple pattern after 9-h MM is obtained using the cubic
Fm3m structure factor with a lattice parameter value of
5.371(8) Å plus a relatively small quantity of the tetrago-
nal phase (18%), whose presence can be inferred from
the shoulder to the right-hand side of the most intense
220 peak of the cubic phase at about 3.4 Å−1. In this case,
the true nature of the tetragonal phase (t, tЈ, tЉ) remains
uncertain.
The SAS neutron curves of Fig. 7 conform to the trend
previously quoted for the CZ20 and CZ50 specimens.
It is significant that the curves of ball-milled speci-
mens lie above the parental mixture, confirming further
fragmentation and disordering due to interpenetrating
structures.
FIG. 5. Small-angle neutron scattering for CZ50 specimen in log–log
scale. Data points refer to 0-h sample. Curve slope is −3.87. Patterns
of 1- (crosses) and 9-h (triangles) MM samples have slope − 3.13.
1542
J. Mater. Res., Vol. 15, No. 7, Jul 2000
Downloaded: 25 Mar 2015
http://journals.cambridge.org
IP address: 169.230.243.252

S. Enzo et al.: Structural characterization of ceria–zirconia powder catalysts prepared by high-energy mechanical milling
The formation of a solid solution in the composition
range we have examined can be evaluated by analysis of
variation in the unit cell parameter with composition. It is
known that the introduction of Zr4+ ion (ionic radius
0.84 Å according to Shannon and Prewitt)24 into the
face-centered-cubic (fcc) lattice of CeO2 (the ionic radius
of Ce4+ is 0.97 Å) will cause a shrinking of the unit cell
parameter, which is due to the different sizes of the two
ions. Therefore, on the analogy of the Vegard law, a
linear relationship is expected to hold for all samples
in the cubic phase stability region. If the tetragonal re-
gion is also considered this linearity is expected to hold
better for the cell volume normalized to the number of
stoichiometric units contained, rather than to unit cell
parameters, because of the variation in the c/a ratio.
Figure 8 shows that a linear relationship has not been
obtained over the entire composition range but a small
positive deviation is observed instead. This deviation
is beyond the experimental uncertainty, due to the cor-
rection included in the Rietveld methodology which,
conversely, was not adopted in a previous work by one
of us on materials produced similarly by MM.9 This
behavior could have at least two different physicochemi-
cal origins: (i) compositional inhomogeneity of samples
(ഛ10% of nominal composition) and (ii) oxireduction
processes causing the enlargement of cell parameter
owing to the presence of Ce3+. Discriminating between
the two effects is a challenging proposition because
of the low discrepancies observed. A comparison of
these results with data obtained from different laborato-
ries and from empirical relationships is also shown in
Fig. 8. The results are in fair agreement with those re-
ported in the literature,25,26 and again, the small devia-
FIG. 7. Small-angle neutron scattering for the CZ80 specimen in log-
log scale. Data points refer to 0-h sample. Curve slope is −4.15. Pat-
terns of 1- (crosses) and 9-h (triangles) MM samples have −3.17 and
−3.04 slopes, respectively.
FIG. 6. (a) Neutron diffraction patterns (data points) and Rietveld fit
(solid lines) for CZ80 specimens ball-milled for time indicated. (b)
Evolution of majority cubic phase as function of milling time.
J. Mater. Res., Vol. 15, No. 7, Jul 2000
Downloaded: 25 Mar 2015
1543
IP address: 169.230.243.252
http://journals.cambridge.org

S. Enzo et al.: Structural characterization of ceria–zirconia powder catalysts prepared by high-energy mechanical milling
tions observed can be explained by local sample
inhomogeneity originating from the different methods
of synthesis.
An analysis of the variation in the lattice parameter for
the cubic phase with milling time is reported in Fig. 9. As
already mentioned, the length of the unit cell parameter
in all samples examined, regardless of the initial compo-
sition, increases from an initial value of a ס 5.411 Å,
calculated for pure CeO2, up to a value of a ס
5.419(6) Å found for the CZ50 sample after 1-h milling
time. At higher milling times, owing to the formation of
solid solution, this value decreases in proportion to the
zirconium content. While the shrinking of unit cell di-
mension can be easily explained by the progressive in-
troduction of the smaller Zr4+ into the cubic lattice of
CeO2, lattice expansion is a more puzzling issue. One
reasonable explanation might take into account the for-
mation of some Ce3+ during milling. In this case, the
larger dimensions of the Ce3+ ion (1.14 Å versus 0.97 Å
for Ce4+) lead us to expect an increase in the cell param-
eter proportional to the amount of reduction. A compari-
son of the unit cell parameter with that found for reduced
ceria27 would indicate a degree of reduction of about 2%
with formation of CeO1.990. In our conditions, the me-
chanical stress produced under high-energy milling at a
temperature that can easily reach local values of 600–
700 K may facilitate reduction of ceria. The reduction of
ceria under mechanical stress has already been observed
in ceria–zirconia compounds.28
In situ annealing of end product specimens at high
temperatures affects structural features in a similar
manner. For the CZ20 specimen ball milled for 9 h, the
ZrO2 monoclinic component disappeared at 700 °C.
The broadened profiles became progressively sharper
but this effect can be removed only partially after treat-
ment at 1000 °C. For the CZ50 sample, we can envisage
a similar behavior pattern: The tetragonal phase will
increase as a function of annealing temperature at the
expense of the remaining cubic component. Finally, for
the CZ80 sample, the cubic Fm3m structure factor is
invariably present in the patterns. The molar volume
data calculated are plotted in Fig. 10. Molar volume in-
creases in near-linear fashion as a function of annealing
temperature, a behavior pattern similar to that re-
FIG. 8. Molar volume as function of ZrO2 content. This quantity was
obtained after normalizing the unit cell volumes to the number of
stoichiometric units contained (two in the cubic cell, four in the te-
tragonal) times their volume fraction. Data of the present investigation
are compared to other work in the literature.
FIG. 9. Cubic phase lattice parameter as function of milling time for
different compositions. Cross refers to CZ20 samples at 0- and 1-h
milling. Squares refer to CZ50 and circles to CZ80 specimens. Points
are connected by a straight line.
FIG. 10. Molar volume data versus temperature. Data are calculated
from Rietveld fit of specimen patterns after in situ thermal treatment at
different temperatures. The linear increase is evident.
1544
J. Mater. Res., Vol. 15, No. 7, Jul 2000
Downloaded: 25 Mar 2015
http://journals.cambridge.org
IP address: 169.230.243.252

S. Enzo et al.: Structural characterization of ceria–zirconia powder catalysts prepared by high-energy mechanical milling
ported for pure ceria.29 This may be due either to the
thermal expansion coefficient of the lattice and/or to
reduction of CexZr1−xO2 to CexZr1−xO2−␦ processes in
3. A. Trovarelli, C. de Leitenburg, and G. Dolcetti, CHEMTECH 27,
32 (1997).
4. P. Fornasiero, R. Di Monte, G. Ranga Rao, J. Kaspar, S. Meriani,
A. Trovarelli, and M. Graziani, J. Catal. 151, 168 (1995).
the solid solution. The latter effect is indirectly con-
firmed by the observed embrittlement of the vanadium
used as a sample holder after high-temperature anneal-
ing, which is caused by oxygen extraction from the re-
duced samples.
5. M. Yashima, H. Arashi, M. Kakihana, and M. Yoshimura, J. Am.
Ceram. Soc. 77, 1067 (1994).
6. M. Yashima, K. Morimoto, N. Ishizawa, and M. Yoshimura,
J. Am. Ceram. Soc. 76, 2865 (1993).
7. M. Yashima, S. Sasaki, M. Kakihana, Y. Yamaguchi, H. Arashi,
and M. Yoshimura, Acta Crystallogr. Sect. B 50, 663 (1994).
8. F. Zamar, A. Trovarelli, C. de Leitenburg, and G. Dolcetti, Stud.
Surf. Sci. Catal. 101, 1283 (1996).
IV. CONCLUSIONS
9. A. Trovarelli, F. Zamar, J. Llorca, C. de Leitenburg, G. Dolcetti,
and J. Kiss, J. Catal. 169, 490 (1997).
10. D. Michel, F. Faudot, E. Gagget, and L. Mazerolles, J. Amer.
Ceram. Soc. 76, 2884 (1993).
11. M. Qi and H.J. Fecht, Mater. Sci. Forum 187, 269 (1998).
12. T. Egami, W. Dmowski, and R. Brezny, SAE Technical Paper Se-
ries No. 970461, presented at 1997 Int. Cong. and Exposition,
Detroit, MI, February 24–27 (1997).
The main conclusions to be drawn from the present
work can be summarized as follows:
(1) It is confirmed that high-energy MM is an effec-
tive method to prepare ceria–zirconia solid solutions
with a good degree of homogeneity over a wide compo-
sition range.
(2) The method allows the stabilization of CZ50 with
a pseudofluorite cell. The comparison of our data to those
previously reported on ceramic samples also points to the
importance of the preparation method to stabilize tЉ
phase at this composition. The higher value of oxygen
displacement is also of interest if the preparation of these
materials is focused to catalytic applications.
(3) Thermal annealing up to 1000 °C does not cause
phase segregation in all samples but only slight varia-
tions in phase composition in favor of tetragonal
structure.
13. M. Yashima, S. Sasaki, Y. Yamaguchi, M. Kakihana, M. Yoshimura,
and T. Mori, Appl. Phys. Lett. 72, 182 (1998).
14. G. Vlaic, P. Fornasiero, S. Geremia, J. Kaspar, and M. Graziani,
J. Catal. 168, 386 (1997).
15. The Rietveld Method, edited by R.A. Young (International Union
of Crystallography, Oxford University Press, Oxford, United
Kingdom, 1993).
16. J. Rodriguez-Carvajal, Physica B 192, 55 (1993).
17. S. Enzo, R. Frattini, R. Gupta, P.P. Macr`ı, G. Principi, L. Schiffini,
and G. Scipione, Acta Mater. 44, 3105 (1996).
18. JCPDS data base, Swarthmore, PA (1994).
19. P. Fornasiero, G. Balducci, R. Di Monte, J. Kaspar, V. Sergo,
G. Gubitosa, A. Ferrero, and M. Graziani, J. Catal. 164, 173
(1996).
20. W. Stichert and F. Schuth, Chem. Mater. 10, 2020 (1998).
21. C. De Leitenburg, A. Trovarelli, F. Zamar, S. Maschio, G. Dol-
cetti, and J. Llorca, Chem. Commun. 2181 (1995).
22. M.Yu. Sinev, G.W. Graham, L.P. Haak, and M. Shelef, J. Mater.
Res. 11, 1960 (1996).
23. A. Trovarelli, Comm. Inorg. Chem. 20, 263 (1999).
24. R.D. Shannon and C.T. Prewitt, Acta Crystallogr., Sect. B 25, 925
(1969).
25. P. Duran, M. Gonzales, C. Moure, J.R. Jurado, and C. Pascual,
J. Mater. Sci. 25, 5001 (1990).
26. D.J. Kim, J. Am. Ceram. Soc. 72, 1415 (1989).
27. H-W. Chiang, R.N. Blumenthal, and R.A. Fournelle, Solid State
Ionics 66, 85 (1993).
ACKNOWLEDGMENTS
Neutron measurements were funded with the aid of
CLRC (RB 8706 and RB 8745 experiments) and by the
European Community (Training Mobility and Research
Programme for large scale facility). The authors thank
Dr. A. Hannon and Dr. R.K. Heenan, as well as the staff
of LAD and LOQ instruments, for their generous help.
We also acknowledge the Ministry for Science and Edu-
cation (MURST) of Italy (former funding 40%).
REFERENCES
28. V. Sergo, C. Schmid, S. Meriani, and A.G. Evans, J. Am. Ceram.
Soc. 77, 2971 (1994).
29. M. Ozawa and C-K. Loong, Catal. Today 50, 329 (1999).
1. H. Inaba and H. Tagawa, Solid State Ionics 83, 1 (1996).
2. A. Trovarelli, Catal. Rev. Sci. Eng. 38, 439 (1996).
J. Mater. Res., Vol. 15, No. 7, Jul 2000
Downloaded: 25 Mar 2015
1545
IP address: 169.230.243.252
http://journals.cambridge.org