Tetragonal nanophase stabilization in nondoped sol-gel zirconia prepared with different hydrolysis catalysts

Article abstract of DOI:10.1006/jssc.1997.7586

Sol-gel zirconia was prepared with zirconium n-butoxide and HCl, H₂SO₄, C₂H₄O₂, and NH₄OH as hydrolysis catalysts. Samples were characterized with DTA and TG analysis, X-ray powder diffraction, and FTIR spectroscopy. The structure of the crystalline phases was refined by the Rietveld method. When samples were annealed below 300°C, they lost weight and had an amorphous structure that, by annealing at higher temperatures, crystallized into nanostructures. For H₂SO₄ as hydrolysis catalyst, the amorphous structure was stable even at higher temperatures, which was probably caused by the presence of SO_x ions in the structure. The local order in the amorphous phase was similar to the local order in the tetragonal zirconia. Crystallization of the amorphous phase produced tetragonal and monoclinic nanophases, with the tetragonal as the main phase. Both phases had a similar average crystallite size. By annealing, the tetragonal nanophase, which was more stable when C₂H₄O₂ was the hydrolysis catalyst, was transformed into the monolinic nanophase. Since not only OH^- ions in the structure were detected with FTIR spectroscopy but also Zr vacancies were measured with X-ray powder diffraction in the zirconia crystalline structure, we propose that these defects stabilized the tetragonal phase. Both defects disappeared when samples were annealed at high temperatures, which brought about the irreversible transformation of the tetragonal into the monoclinic structure.

Full text of DOI:10.1006/jssc.1997.7586

JOURNAL OF SOLID STATE CHEMISTRY 135, 28—35 (1998)
ARTICLE NO. SC977586
Tetragonal Nanophase Stabilization in Nondoped Sol–Gel Zirconia
Prepared with Different Hydrolysis Catalysts
X. Bokhimi,*ꢀꢁ A. Morales,* O. Novaro,*ꢀꢂ M. Portilla,† T. L o´ pez,‡ F. Tzompantzi,‡ and R. G o´ mez‡
*
Institute of Physics, The National University of Mexico (UNAM), A.P. 20-364, 01000 M e& xico, D.F., Mexico; †Faculty of Chemistry,
The National University of Mexico (UNAM), A.P. 70-197, 01000 M e& xico, D.F., Mexico; ‡Department of Chemistry,
Universidad Aut o& noma Metropolitana, A.P. 55-534, 09349 M e& xico, Mexico
Received March 6, 1997; accepted June 16, 1997
divalent or trivalent cations, its structure is deficient in
Sol–gel zirconia was prepared with zirconium n-butoxide and oxygen, which favors oxygen transportation and produces
HCl, H SO , C , and NH OH as hydrolysis catalysts. ionic conduction. Therefore, zirconia can be used as an
Samples were characterized with DTA and TG analysis, X-ray oxygen sensor (3) and in the fabrication of fuel cells (4).
2
4
2
H
O
4 2
4
powder diffraction, and FTIR spectroscopy. The structure of the
crystalline phases was refined by the Rietveld method. When
samples were annealed below 300°C, they lost weight and had an
amorphous structure that, by annealing at higher temperatures,
These mechanical and electronic properties were discovered
and studied mainly in microcrystalline zirconia (5), but re-
cently they have also been studied in nanocrystalline zirconia.
During the past 10 years, nanocrystalline zirconia has
been produced by using different methods (6—10). The small
size of these crystals favors zirconia sintering at low temper-
atures (7), the preparation of zirconia thin films (8), and their
use as a support in catalysis (11).
crystallized into nanostructures. For H
lyst, the amorphous structure was stable even at higher temper-
atures, which was probably caused by the presence of SO ions in
2
SO
4
as hydrolysis cata-
x
the structure. The local order in the amorphous phase was similar
to the local order in the tetragonal zirconia. Crystallization of the
amorphous phase produced tetragonal and monoclinic nano-
In contrast to microcrystalline pure zirconia, where the
phases, with the tetragonal as the main phase. Both phases had phase at room temperature is monoclinic, in nanocystalline
a similar average crystallite size. By annealing, the tetragonal pure zirconia, the phase at this temperature can be tetra-
nanophase, which was more stable when C
2
H
4
O was the hy-
2
gonal or monoclinic, depending on the preparation condi-
tions (12—15). In microcrystalline zirconia, the stabilization
of the tetragonal zirconia phase at room temperature is
done by doping it with a large ion such as Y, Ce, Mg, or Ca
drolysis catalyst, was transformed into the monolinic nano-
؊
phase. Since not only OH ions in the structure were detected
with FTIR spectroscopy but also Zr vacancies were measured
with X-ray powder diffraction in the zirconia crystalline struc-
ture, we propose that these defects stabilized the tetragonal
phase. Both defects disappeared when samples were annealed at
high temperatures, which brought about the irreversible trans-
formation of the tetragonal into the monoclinic structure. ( 1998
Academic Press
(5). By annealing, this stabilized tetragonal phase is revers-
ibly transformed into monoclinic zirconia through a mar-
tensitic transformation (5).
The initial crystalline phase obtained in nanocrystalline
nondoped zirconia depends on the preparation conditions
(6, 9, 16). It can be tetragonal or monoclinic. When nano-
crystalline nondoped zirconia is annealed, the tetragonal
phase is transformed into the monoclinic phase (12). This
transformation, however, is neither martensitic nor revers-
1
. INTRODUCTION
The interesting structural and electronic properties of ible (17). In this case, some research groups suggest that the
zirconia (ZrO ) have been put to use in many applications. transformation is associated with the growth of the tetra-
ꢂ
Since it has a high melting point and low thermal conductiv- gonal crystallite size (12), like the dependence of the tetra-
ity at high temperature, zirconia is used as a thermal shield gonal to monoclinic transformation observed in poly-
(
1). In zirconia composites, the martensitic transformation crystalline zirconia (18).
from tetragonal to monoclinic zirconia produces materials
One method for making nanocrystalline zirconia is the
with high toughness (2). When zirconia is doped with sol—gel technique, where the final product depends on many
parameters (6). Among them, the hydrolysis catalyst and
temperature play important roles, especially when multiple
nanocrystalline phases coexist in the system (19).
ꢁ
To whom correspondence should be addressed.
ꢂ Member of El Colegio Nacional, Mexico.
2
8
0
022-4596/98 $25.00
Copyright ( 1998 by Academic Press
All rights of reproduction in any form reserved.

STABILIZATION OF SOL—GEL ZIRCONIA
29
In the particular case of nanocrystalline zirconia, the given in parentheses. When the numbers corresponded to
formation of the tetragonal or monoclinic nanophase in the parameters obtained from the Rietveld refinement, the esti-
initial step of crystallization depends on pH, which in the mated standard deviations are not estimates of the analysis
sol—gel technique can be controlled with the hydrolysis as a whole but only of the minimum possible probable
catalyst.
The concentration of the nanophases in the samples and
errors based on their normal distribution (23).
their crystallography can be obtained from X-ray powder
diffraction and refinement of their crystalline structure (19).
Therefore, in the present work, we have used these tech-
niques for studying the nanophases in the sol—gel zirconia
prepared with different hydrolysis catalysts and annealed in
air at different temperatures.
FTIR Spectroscopy
Transparent disks, obtained by pressing dried sample
powder, were characterized with a 170-SX Nicolet FTIR
spectrometer.
3
. RESULTS AND DISCUSSION
2
. EXPERIMENTAL
Sample Preparation
Sol—gel zirconia samples lost weight when they were an-
nealed from room temperature to 300°C (Fig. 1). This
weight loss was produced by evaporation of the residual
volatile components involved in sample preparation. Above
this temperature, the sample weight loss depended on the
hydrolysis catalyst.
For all hydrolysis catalysts, the structure of the samples
annealed below 300°C was amorphous (Fig. 2A). The center
of the main broad peak of the amorphous phase, however,
corresponded to the position of the main peak associated
with the (011) reflection of the tetragonal zirconia (Fig. 2B).
This broad peak was produced by the most frequent ion—ion
distances in the amorphous structure, which, in the present
case, corresponded to Zr—O distances. This means that the
local order in the amorphous phase was similar to the local
order in the tetragonal zirconia phase. This explains why the
most abundant crystalline phase formed during the crystal-
lization of the amorphous phase was the tetragonal one.
With continuous stirring, 10 ml of zirconium n-butoxide
was dissolved in 150 ml of tert-butyl alcohol, and the mix-
ture was refluxed at 70°C. The corresponding hydrolysis
catalyst, HCl or H SO for pH 3, C H O for pH 5, and
ꢂ
ꢃ
ꢂ ꢃ ꢂ
NH OH for pH 9, was added until the desired pH was
ꢃ
obtained. Then the solution was stirred and refluxed for
1
0 min and cooled to 50°C. At this temperature, 25.66 ml of
water was added dropwise to the solution. Then its temper-
ature was again raised to 70°C, and the solution was stirred
and refluxed until forming a gel, which was dried in air at
1
00°C for 24 h (fresh sample).
Before the analysis, fresh samples were annealed in air at
00, 400, 600, and 800°C for 12 h at each temperature.
2
DTA and TG Analysis
These analyses were done in a Dupont Model 950 ther-
moanalyzer with DTA and TGA stages. Samples were an-
nealed in air from 25 to 1200°C at the annealing rate of
2
0°C/min.
X-Ray Diffraction
The crystalline structure of the phases in the samples was
characterized by X-ray powder diffraction and by refining it
by the Rietveld method. X-ray diffraction patterns were
measured at room temperature with a Siemens D-5000
diffractometer having CuKa radiation and a secondary-
beam graphite monochromator. Specimens were prepared
by packing sample powder into a glass holder. Intensity was
measured by step scanning in the 2h range between 20° and
1
10°, with a step of 0.02° and a measuring time of 2 s per
point. Crystalline structures were refined by using the
DBWS-9411 (20) and WYRET (21) programs. The peak
profiles, modeled with a pseudo-Voigt function, had average
crystallite size and microstrain as profile-breadth fitting
parameters (22). The standard deviations, showing the vari-
FIG. 1. TG curves as a function of temperature of the sol—gel zirconia
ation of the last figures of the corresponding number, are prepared with different hydrolysis catalysts.

3
0
BOKHIMI ET AL.
FIG. 2. X-ray diffraction patterns of the sol—gel zirconia prepared with
acetic acid as hydrolysis catalyst and annealed in air: (A) at 200°C; (B) at acetic acid as hydrolysis catalyst. Upper tick marks represent the tetra-
00°C. The tick marks and the (011) reflection represent the tetragonal gonal zirconia, while lower tick marks represent the monoclinic zirconia.
zirconia.
FIG. 4. X-ray diffraction patterns of the sol—gel zirconia prepared with
4
weight, and annealing at temperatures above 600°C caused
When HCl, C H O , and NH OH were the hydrolysis additional weight loss (Fig. 1), totaling 37% at 800°C. This
ꢂ ꢃ ꢂ
ꢃ
catalysts, the amorphous structure crystallized when sam- weight loss was similar to the total weight loss found in the
ples were annealed in air at 400°C (Figs. 3—5). During the samples prepared with the other hydrolysis catalysts. This
annealing from room temperature to 400°C, samples lost means that the interaction of SO ions with the amorphous
V
between 30 and 34% of their initial weight (Fig. 1). Their phase stabilized its amorphous structure and shifted its
differential thermal analysis showed that crystallization of crystallization to higher temperatures. This was confirmed
the amorphous phase for these hydrolysis catalysts occurred by the corresponding differential thermal analysis and the
between 300 and 350°C; the precise crystallization temper- weight loss observed above 600°C (Fig. 1). This result will be
ature depended on the specific hydrolysis catalyst.
of interest to those research groups studying zirconia super-
When H SO was the hydrolysis catalyst, samples cry- acids (24—27).
ꢂ
ꢃ
stallized at temperatures higher than 400°C (Fig. 6). Here,
the sample annealed until 400°C lost only 26% of its initial nanocrystalline phases, whose structures were refined by the
For all hydrolysis catalysts, crystallized samples had two
FIG. 3. X-ray diffraction patterns of the sol—gel zirconia prepared with
FIG. 5. X-ray diffraction patterns of the sol—gel zirconia prepared with
hydrochloric acid as hydrolysis catalyst. Upper tick marks represent the ammonium hydroxide as hydrolysis catalyst. Upper tick marks represent
tetragonal zirconia, while lower tick marks represent the monoclinic the tetragonal zirconia, while lower tick marks represent the monoclinic
zirconia.
zirconia.

STABILIZATION OF SOL—GEL ZIRCONIA
31
FIG. 8. Rietveld refinement plot of the sol—gel zirconia prepared with
acetic acid and annealed in air at 400°C (R "0.08). Experimental data
ꢀ
ꢁ
are indicated by dots, while the calculated curve obtained after the refine-
FIG. 6. X-ray diffraction patterns of the sol—gel zirconia prepared with ment is indicated with a continuous line. Upper and lower tick marks
sulfuric acid as hydrolysis catalyst. Upper tick marks represent the tetra- represent the tetragonal zirconia. R was 0.01 for each phase.
$
gonal zirconia, while lower tick marks represent the monoclinic zirconia.
phase was also modeled with the local order of the mono-
clinic phase, but the fit was not as good as with the tetra-
gonal local order.
Rietveld method (Figs. 7—10). One crystalline nanophase
was tetragonal, described by space group P4 /nmc and the
ꢂ
Weight phase concentrations (Table 4) were obtained
from the structure refinement. In the crystallization of the
amorphous phase, both tetragonal and monoclinic zirconia
nanophases coexisted (Tables 5 and 6), the tetragonal phase
being the main phase. Here, the monoclinic nanophase had
a smaller crystallite size than the tetragonal one. For both
hydrochloric acid and acetic acid, this difference in crystal-
lite size was present even after the samples were annealed at
atom positions given in Table 1. The other nanophase was
monoclinic, described by space group P2 /c and the atom
ꢁ
positions given in Tables 2 and 3. These nanostructured
phases had crystallite sizes that varied between 4 and 34 nm
(Table 4). When acetic acid was the hydrolysis catalyst,
the sample annealed at 400°C was composed of 62(3) wt%
of the crystalline tetragonal zirconia phase and 38(1) wt%
of an amorphous phase with the same local order as the
crystalline tetragonal phase (Fig. 8). This amorphous
6
00°C (Table 4). The observed monoclinic nanophase in the
initial crystallization was not associated with any trans-
formation of the tetragonal phase produced by the growing
of the tetragonal crystallite size (12).
FIG. 7. Rietveld refinement plot of the sol—gel zirconia prepared with
hydrochloric acid and annealed in air at 600°C. Experimental data are
indicated by dots, while the calculated curve obtained after the refinement
is indicated with a continuous line. The residue R generated when all
FIG. 9. Rietveld refinement plot of the sol—gel zirconia prepared with
ꢀ
ꢁ
contributing parts were taken into account was 0.08. The residue R
associated with only one phase was 0.009 for the tetragonal zirconia and mental data are indicated by dots, while the calculated curve obtained after
0
ammonium hydroxide and annealed in air at 800°C (R "0.08). Experi-
$
ꢀꢁ
.01 for the monoclinic zirconia. The continuous curve under the tick the refinement is indicated with a continuous line. Upper tick marks
marks in the difference between the experimental data and the calculated represent the monoclinic phase (R "0.01), and lower tick marks represent
$
curve.
the tetragonal zirconia (R "0.01).
$

3
2
BOKHIMI ET AL.
TABLE 2
Monoclinic Zirconia (Space Group P2 /c): Atomic Fractional
1
Coordinates
Atom
Site
x
y
z
Zr
O(1)
O(2)
4e
4e
4e
u
v
w
u
v
w
u
v
w
V
V
W
W
X
X
V
W
X
Note. These atomic fractional coordinates ranged within the limits
specified in Table 3.
vacancies had to be close to OH\ ions occupying oxygen
FIG. 10. Rietveld refinement plot of the sol—gel zirconia prepared with
sulfuric acid and annealed in air at 600°C (R "0.08). Experimental data sites in the structure.
ꢀ
ꢁ
are indicated by dots, while the calculated curve obtained after the refine-
ment is indicated with a continuous line. Upper tick marks represent the
To detect and study OH\ ions in the zirconia structure,
samples were analyzed by using FTIR spectroscopy.
FTIR spectra of fresh samples had absorption bands
associated with residual water and solvent present after
sample preparation. In all samples, these residua disap-
peared after sample annealing at 300°C.
tetragonal phase (R "0.008), and lower tick marks represent the mono-
$
clinic zirconia (R "0.01).
$
To stabilize the tetragonal phase in microcrystalline zir-
conia at low temperatures, zirconia is doped with large ions
such as Y, Ca, or Mg (5). In the sol—gel zirconia reported in
the present work, however, the solutions and precursors
used in the sample peparation did not include any of these
or similar ions.
When samples were annealed at higher temperatures,
FTIR absorption bands associated with stable—even at
6
00°C—OH groups in the phase structures were observed.
For pH 3 and HCl as hydrolysis catalyst (Fig. 11a), the
band produced by OH stretching vibrations was observed
at 3318 cm\ꢁ, while the band produced by its bending
vibrations was observed at 1615 cm\ꢁ. In the FTIR spec-
trum, bands produced by Zr—O binding were also detected.
They were, however, not well defined, except the broad band
at 381 cm\ꢁ.
When the hydrolysis catalyst was sulfuric acid (pH 3), the
bands produced by the stretching and bending vibrations of
the stable OH groups shifted to 3417 and 1636 cm\ꢁ, re-
spectively (Fig. 11b), and had larger intensities than for the
HCl hydrolysis catalyst. In the low-energy region of the
spectrum, Zr—O binding produced well-defined bands at
During sample preparation, H O, OH\ ions, H> ions,
ꢂ
H
ions, and vacancies occur. When samples were an-
!
$ꢂ
nealed to 300°C, the H O present left them. Therefore,
ꢂ
water did not stabilize the tetragonal structure. It is known
that zirconia prepared from strongly acidic solutions
(pH(1) gives rise to the monoclinic phase (9, 28, 29), which
also eliminated H> ions from stabilizing the tetragonal
structure. Since the tetragonal structure that we observed
was a bulk material, the adsorbed hydrogen was also dis-
carded as the stabilizer. These results suggest that a mixture
of OH\ ions and vacancies could cause the observed tetra-
gonal structure stabilization.
6
06 and 459 cm\ꢁ.
From the Rietveld refinement of the tetragonal nano-
phase, we found that zirconium occupancy was less than 1.0.
This corresponded to cation vacancies in the crystalline
structure. For compensating electrical charge, these
TABLE 3
Limits of the Fractional Coordinates of Monoclinic Zirconia
Fractional coordinate
Lower limit
Upper limit
u
V
0.2742(3)
0.0390(2)
0.2090(2)
0.2761(2)
0.0445(2)
0.2112(2)
TABLE 1
u
W
Tetragonal Zirconia (Space Group P4 /nmc): Atomic
2
u
X
Fractional Coordinates
v
V
v
W
0.064(1)
0.329(1)
0.328(1)
0.102(2)
0.339(1)
0.347(1)
Atom
Site
x
y
z
v
X
Zr
O
2a
4d
0.75
0.25
0.25
0.25
0.75
u
w
V
0.425(1)
0.756(1)
0.430(2)
0.451(1)
0.770(2)
0.484(4)
w
W
w
X
Note. u varied between 0.0467(9) and 0.050(2).

STABILIZATION OF SOL—GEL ZIRCONIA
33
TABLE 4
TABLE 6
Zirconia Phase Composition and Average Crystallite Size
as a Function of Hydrolysis Catalyst and Temperature
Zirconia Monoclinic Nanophase: Lattice Parameters as a
Function of Hydrolysis Crystal and Temperature
Hydrolysis
catalyst
¹
(°C)
Tetragonal Monoclinic Tetragonal Monoclinic
Hydrolysis
catalyst
¹
(°C)
a
(nm)
b
(nm)
c
b
(deg)
(wt%)
(wt%)
d(nm)
d(nm)
(nm)
H SO
600
85(3)
9(2)
15(3)
91(4)
12.4(4)
21.9(6)
12(2)
27.8(8)
H SO
600
800
0.5135(5)
0.5198(5)
0.5321(4)
98.88(4)
99.106(4)
ꢂ
ꢃ
ꢂ
ꢃ
8
00
0.51417(4) 0.52030(4) 0.53127(4)
HCl
400
78(5)
35(1)
13(2)
22(4)
65(2)
87(3)
17(1)
24(1)
31(3)
4.1(2)
21(1)
31.1(8)
HCl
400
600
800
0.5096(8) 0.5189(9) 0.5334(9)
0.51463(4) 0.51946(5) 0.53175(5)
0.51463(3) 0.51938(3) 0.53208(3)
98.1(1)
99.07(3)
99.163(3)
6
8
00
00
C H O
400 62(3), 38(1)
12.5(6), 1.3(1)
22.0(5)
C H O
400
600
800
0.494(2)
0.5146(5)
0.533(3)
0.5197(5)
0.556(2)
0.5313(5)
102.66(6)
98.97(6)
99.064(4)
ꢂ
ꢃ
ꢂ
ꢂ ꢃ ꢂ
6
8
00
00
91(1)
17(2)
8.6(2)
82(3)
12.8(1)
29(1)
22(3)
0.51463(4) 0.51973(4) 0.53165(4)
NH OH
400
70(5)
45(2)
11.7(4)
30(6)
55(1)
88.3(7)
8.4(3)
20.7(8)
32(3)
5.8(5)
22(1)
34(1)
NH OH
400
600
800
0.5158(5) 0.5192(5) 0.5305(5)
0.51436(5) 0.51986(5) 0.53147(5)
0.51433(2) 0.51981(3) 0.53170(3)
98.37(6)
99.169(6)
99.153(3)
ꢃ
ꢃ
6
8
00
00
For acetic acid as hydrolysis catalyst (pH 5), the band
produced by OH stretching vibrations was shifted to
FTIR and X-ray diffraction results showed that the tem-
perature behavior of the bands associated with the OH
groups and the amount of tetragonal phase were correlated.
The foregoing analysis confirmed the presence of OH\
ions in sol—gel nondoped zirconia and reinforced the as-
sumption that OH\ ions stabilized the tetragonal structure.
The assumption that OH\ ions and cation vacancies
stabilized the tetragonal structure was according to the
reported fact (30) that monoclinic zirconia is obtained when
3
381 cm\ꢁ and increased in intensity (Fig. 11c). The band
associated with the corresponding bending vibrations, how-
ever, was not observed.
The shift of the band produced by the stretching vibration
of the OH groups was stronger for ammonium hydroxide as
hydrolysis catalyst (pH 9). This band was at 3417 cm\ꢁ in
the FTIR spectrum (Fig. 11d), which suggested a stronger
interaction between OH\ ions and the zirconia structure.
After the samples were annealed at 800°C, the oberved
FTIR bands associated with the OH groups had a very
weak intensity.
Zr(NO ) is used as a precursor in its preparation, but
ꢄ ꢃ
tetragonal zirconia is obtained when Zr(OH) is the precur-
ꢃ
sor.
The annealing of the samples at even higher temperatures
increased the crystallite size of the nanophases and caused
the transformation of the tetragonal nanophase into the
monoclinic one (Figs. 3—6 and Table 4). According to our
model of the stabilization of the tetragonal structure, when
samples were annealed, the hydroxyl ions left them, de-
stabilizing the tetragonal structure and transforming it into
the monoclinic one. This transformation, however, was not
reversible. Since the Oꢂ\ ions associated to the OH\ ions
left the sample, the Ti : O molar ratio and the corresponding
Ti—O bindings increased.
TABLE 5
Zirconia Tetragonal Nanophase: Lattice Parameters as a
Function of Hydrolysis Catalyst and Temperature
Hydrolysis
catalyst
¹ (°C)
a
(nm)
c
(nm)
H SO
600
0.35951(1)
0.35945(5)
0.51800(7)
0.5182(2)
ꢂ
ꢃ
8
00
HCl
400
0.35975(5)
0.35945(2)
0.35945(2)
0.51749(4)
0.51816(4)
0.51839(7)
4
. CONCLUSIONS
6
8
00
00
The structure of the zirconia produced by the sol—gel
technique was amorphous when the samples were heated
below 300°C. When sulfuric acid was the hydrolysis cata-
lyst, however, the amorphous phase was stable at even
higher temperatures. The local order in the amorphous
structure was similar to the local order in the tetragonal
phase.
C H O
400
0.35964(3)
0.35941(1)
0.35954(3)
0.51685(8)
0.51804(3)
0.51882(8)
ꢂ
ꢃ ꢂ
6
8
00
00
NH OH
400
0.35989(5)
0.35939(2)
0.35953(3)
0.5165(2)
0.51809(3)
0.51877(8)
ꢃ
6
8
00
00

3
4
BOKHIMI ET AL.
FIG. 11. FTIR transmittance curves of sol—gel zirconia annealed at 300°C and prepared with different hydrolysis catalysts: (a) hydrochloric acid, (b)
sulfuric acid, (c) acetic acid, and (d) ammonium hydroxide.
For all hydrolysis catalysts, by annealing, the amorphous
We proposed a model for explaining the stabilization of
phase crystallized into the tetragonal and monoclinic zirco- the tetragonal phase. In this model, OH\ ions and Zr
nia nanophases, which coexisted since the beginning of crys- vacancies in the crystalline structure were the stabilizers.
tallization. Here, the tetragonal phase was the main phase.
The model also explained the irreversible structure trans-
The average crystallite size of the monoclinic and tetra- formation of the tetragonal into the monoclinic phase,
gonal phases was similar. This result showed that the mono- which was observed when samples were annealed.
clinic nanophase can exist without needing to be generated
by the transformation of the tetragonal phase when its
crystal grows.
REFERENCES
By annealing, the tetragonal phase was irreversibly trans-
formed into the monoclinic phase. When acetic acid was the
hydrolysis catalyst, the tetragonal zirconia nanophase was
more stable.
1
2
. P. Reyer-Didier and H. Werke, Ind. Ceram. 839, 449 (1989).
. Q.-M. Yuan, J.-Q. Tan, and Z.-G. Jin, J. Am. Ceram. Soc. 69, 265
(1986).

STABILIZATION OF SOL—GEL ZIRCONIA
35
3
4
5
. J. Fouletier, Ceramurgia 25, 209 (1995).
. Y. Ohno, Y. Kaga, and S. Nagata, Electr. Eng. Jpn. 107, 59 (1987).
16. A. Clearfied, J. Mater. Res. 5, 161 (1990).
17. Zh. Lin and L. Zhang, Rare Met. (Beijing) 9, 296 (1990).
. E. C. Subbarao, H. S. Maiti, and K. K. Srivastava, Phys. Status Solidi 18. P. F. Becher, J. Am. Ceram. Soc. 75, 493 (1992).
1, 9 (1974). 19. Bokhimi, A. Morales, O. Novaro, T. Lopez, E. Sa
J. Mater. Res. 10, 2788 (1995).
2
´
´
nchez, and R. Go
´
mez,
6
7
8
. B. E. Yoldas, J. Mater. Sci. 21, 1080 (1986).
. G. Skandan, Nanostruct. Mater. 5, 111 (1995).
20. R. A. Young, A. Sakthivel, T. S. Moss, and C. O. Paiva-Santos, J. Appl.
Crystallogr. 28, 366 (1995).
. K. T. Miller, C. J. Chan, M. G. Cain, and F. F. Lang, J. Mater. Res. 8,
1
69 (1993).
. R. P. Denkewicz, Jr., K. S. TenHuisen, and J. H. Adair, J. Mater. Res. 5,
698 (1990).
21. Margarita Schneider EDV-Vertrieb, Starnbergweg 18, D-8134,
9
P o¨ cking, Germany, 1992, Tel. 0049-8157-8727, Fax 0049-8157-4527.
22. P. Thompson, D. E. Cox, and J. B. Hastings, J. Appl. Crystallogr. 20, 79
(1987).
2
1
1
1
1
1
1
0. D. Michel, L. Mazerolles, P. Berthet, and E. Gaffet, J. Solid State Inorg.
Chem. 32, 673 (1995).
23. E. Prince, J. Appl. Crystallogr. 14, 157 (1981).
1. R. Go
ett. 1941 (1992).
2. A. Chatterjee, S. K. Pradhan, A. Datta, M. De, and D. Chakravorty, 25. C. Monterra, G. Cerrato, C. Emanuel, and V. Bolis, J. Catal. 142, 349
J. Mater. Res. 9, 253 (1994). (1993).
3. Y. Wang, K. Lu, D. Wang, Zh. Wu, and Zh. Fang, J. Phys.: Condens. 26. F. C. Wu and S. C. Yu, J. Mater. Sci. 25, 970 (1190).
´
mez, T. Lo
´
pez, G. Ferrat, J. M. Domı
´
nguez, and I. Schifter, Chem. 24. A. Corma, V. Fornes, M. I. Juan-Rajadell, J. M. Lo
´ pez-Nieto, Appl.
¸
Catal. A 116, 151 (1994).
Matter 6, 633 (1994).
4. M. L. Rojas-Cervantes, R. M. Martı
and J. de D. Lopez-Gonzalez, J. Mater. Sci. 29, 3743 (1994).
27. A. Clearfield, G. P. D. Serrette, and A. H. Khazi-Syed, Catal. ¹oday 20,
295 (1994).
28. H. Saricimen, Powder ¹echnol. 27, 23 (1980).
´
n Aranda, A. J. Lo´ pez-Peinado,
´
´
5. R. Caruso, N. Pellegri, O. De Sanctis, M. C. Caracoche, and P. C. 29. P. E. D. Morgan, J. Am. Ceram. Soc. 67, C-204 (1984).
Rivas, Mater. Res. Soc. Symp. 355, 557 (1995).
30. R. Srinivasan and B. H. Davis, Catal. ¸ett. 14, 165 (1992).