Preparation and properties of Y2O3-doped ZrO2 thin films by the sol-gel process

Article abstract of DOI:10.1023/A:1018650424335

Y₂O₃-doped ZrO2 (YZ) thin films were prepared on alumina substrates by the dip-coating process. The dip-coating solution consisted of a homogeneous sol, and was prepared by using the respective chlorides as raw materials, with ethylene glycol, 2-butanol and distilled water as solvents. The thin films containing 0-20 mol % Y₂O₃ were successfully produced by thermal treatment above 600°C. The characterization for the film preparation was performed by means of thermogravimetric-differential thermal analysis for the thermal analysis, and scanning electron microscopy for the morphological analysis and thickness measurements. The properties of the films were characterized in terms of a study of the crystalline phase, the crystallite size, the microstructure and the electrical conductivity by using X-ray diffraction, scanning electron microscopy and the complex impedance techniques. In all YZ thin films, the tetragonal phase was stable at low temperatures as a result of the crystallite size effect. However, at higher temperatures, the tetragonal phase was transformed into either the monoclinic phase or the cubic phase, depending on the doping concentration. The YZ thin film of 8 mol % Y₂O₃ content was stabilized to almost cubic phase at 1000°C. Resonable conductivity behaviour at YZ was observed for the YZ thin films. The electrical conductivity of YZ thin films was similar to the values of the sintered body.

Full text of DOI:10.1023/A:1018650424335

JOURNAL OF MATERIALS SCIENCE 32 (1997) 5249 – 5256
Preparation and properties of Y O -doped ZrO thin
2
3
2
films by the sol–gel process
JOO-SIN LEE
Department of Materials Science and Engineering, Kyung-Sung University, Nam-gu, Pusan,
6
08-736, Korea
T. MATSUBARA, T. SEI, T. TSUCHIYA
Department of Materials Science and Engineering, Science University of Tokyo, Noda-shi,
Chiba, 278, Japan
Y O -doped ZrO (YZ) thin films were prepared on alumina substrates by the dip-coating
2
3
2
process. The dip-coating solution consisted of a homogeneous sol, and was prepared
by using the respective chlorides as raw materials, with ethylene glycol, 2-butanol and
distilled water as solvents. The thin films containing 0—20 mol % Y O were successfully
2
3
produced by thermal treatment above 600°C. The characterization for the film preparation
was performed by means of thermogravimetric—differential thermal analysis for the thermal
analysis, and scanning electron microscopy for the morphological analysis and thickness
measurements. The properties of the films were characterized in terms of a study of the
crystalline phase, the crystallite size, the microstructure and the electrical conductivity by
using X-ray diffraction, scanning electron microscopy and the complex impedance
techniques. In all YZ thin films, the tetragonal phase was stable at low temperatures as
a result of the crystallite size effect. However, at higher temperatures, the tetragonal phase
was transformed into either the monoclinic phase or the cubic phase, depending on the
doping concentration. The YZ thin film of 8 mol % Y O content was stabilized to almost
2
3
cubic phase at 1000°C. Resonable conductivity behaviour at YZ was observed for the YZ thin
films. The electrical conductivity of YZ thin films was similar to the values of the sintered
body.
1
. Introduction
Negishi [7] prepared zirconia and stabilized-zirco-
nia thin films by the thermal decomposition process
of painted films of zirconium 2-ethylhexonate and
yttrium 2-ethylhexonate. He showed that the process
is applicable to fuel cells and sensors by the thin solid
electrolyte film technique.
Y O -doped ZrO is widely used for structural mater-
ꢀ
ꢁ
ꢀ
ials because of a number of superior properties, such
as heat resistance, high hardness, and chemical dura-
bility, and for electronic materials using the oxygen
ionic conductivity. Using these properties, the dense
thin films are applied in fuel cells, oxygen sensors, and
films for corrosion resistance and heat resistance
against oxidation.
In all these applications, the polymorphic stability
of zirconia over a broad temperature range is very
important. It is well known that zirconia shows three
phases with increasing temperatures, e.g. monoclinic
up to &1170 °C, tetragonal up to &2370 °C and
cubic up to the melting temperature, at &2706 °C.
The stabilization of the high-temperature phases
at moderate temperatures ensures the excellent mech-
anical, thermal, and electrical properties of zirconia
ceramics.
Peshev et al. [10] also reported the preparation of
thin yttria-stabilized zirconia films by using an iso-
propanol sol of zirconium propoxide and yttrium iso-
propoxide. Crack-free yttria-stabilized zirconia films,
approximately 0.2—0.5 lm thick, were deposited on
continuous quartz substrates by Kueper et al. [11].
Sakurai et al. [12] also reported the coatings of
8.8 mol % yttria-doped zirconia, fabricated using hy-
drolysis of zirconium n-butoxide and yttrium nitrate.
The above reports were attempts to prepare stabil-
ized zirconia thin film for application to the solid
electrolyte [7, 10—12]. There are other reports in the
literature of the application of the chemical protection
of metal substrates from attack by acid and oxidation
[3, 4, 6, 8, 9, 13].
The coatings of pure ZrO [1—6] and Y O -doped
ꢀ
ꢀ ꢁ
ZrO [7—13] produced by the sol—gel process have
been researched for several years, because of several
advantages, such as low processing temperatures,
homogeneity, the possibility of coating on substrates
with large areas, and low cost.
ꢀ
Studies on pure ZrO thin-film preparation, using
ꢀ
several organozirconium compounds by Izumi et al.
[3], and zirconium isopropoxide by Neto et al. [4]
and Atik et al. [6], reported that the heat resistance
0
022—2461 ( 1997 Chapman & Hall
5249

against oxidation [3] and chemical corrosion resist-
ance [4, 6] were improved by the ZrO coating on
ꢀ
stainless steel sheets.
Shane et al. [8] reported that 8 wt % Y O —ZrO
was prepared by spin-coating a solution containing
zirconium alkoxides and yttrium acetate on stainless
ꢀ
ꢁ
ꢀ
steel substrates. In addition, stabilized ZrO thin films
ꢀ
were prepared by Maggio et al. [9] using zirconium
tetra-n-tuboxide and cerium (III)-2-4-petanedionate
for CeO —ZrO , and by Miyazawa et al. [13] using
ꢀ
ꢀ
zirconium tetra-n-butoxide and yttrium acetate tetra-
hydrate for Y O —ZrO .
ꢀ
ꢁ
ꢀ
Until now, the starting materials for the sol—gel thin
films have usually used the metal alkoxides [1—13].
However, these materials are expensive and they ex-
hibit differences in the hydrolysis rate of each com-
ponent. There has been very limited attention directed
at the preparation of stabilized zirconia thin films
by using the non-alkoxides as starting materials and
the systematic study of the variation of the stabilizer
content.
In this work, the inorganic salts which have not
usually been used, were used as starting materials to
fabricate the Y O -doped ZrO (YZ) thin films. The
characterization for the thin-film fabrication, such as
the thermal analysis, morphological analysis and
thickness measurements, was performed. The proper-
ties of thin films were also evaluated through phase
analysis, microstructure analysis, and electrical con-
ductivity measurements.
ꢀ
ꢁ
ꢀ
Figure 1 Preparation procedure for YZ thin film.
performed on dried powders of the coating solution at
a heating rate of 5 °C min\ꢂ by using TG and DTA
(Rigaku type 8078). Film thickness was determined by
SEM (Jeol JSM-T100) observation of the fractured
surface of specimens.
2
. Experimental procedure
The as-coated films were characterized to study the
crystalline phase, the crystallite size, the microstruc-
ture and the electrical conductivity. The phase
transitions during heat treatments were studied by
The thin films prepared in this study have the com-
position of (100!x)ZrO · xY O , where x"0, 2, 5,
ꢀ
ꢀ ꢁ
8
, 11 and 20, respectively. Fig. 1 shows the proced-
ures for YZ thin-film processing. Zirconium chloride
octahydrate (ZrOCl · 8H O) powder (Wako Pure
X-ray diffraction (XRD) using CuK radiation. The
a
X-ray diffractometer (Rigaku CN4148) was equipped
with a thin-film attachment and a carbon mono-
chromator. The X-ray diffraction patterns were re-
corded in the range 20°(2h(80°. Lattice constants
were calculated from the d-spacing of the (111) diffrac-
tion peak measured by XRD. a-Al O was used as
ꢀ
ꢀ
Chemical Industries, Ltd, Japan, purity 99%)
and yttrium chloride hexahydrate (YCl · 6H O) pow-
ꢁ
ꢀ
der (Wako Pure Chemical Industries, Ltd, Japan,
purity 99.9%) were used as raw materials, which were
dissolved in the mixed solvents of ethylene glycol, 2-
butanol and distilled water. They were stirred at room
temperature and dissolved to form a coating solution.
A homogeneous and transparent sol was obtained.
A commercial alumina plate was used as a substrate.
The substrates were degreased ultrasonically in
acetone before dipping. The substrate was dipped into
the solution and withdrawn at a rate of about
ꢀ
ꢁ
a reference.
For the estimation of crystallite sizes as a function
of heat-treatment temperature, X-ray diffraction-
line broadening (XRD-LB) measurements were
performed. The Scherrer equation was used for the
calculation:
1
cm min\ꢂ.
kk
D"
(1)
The coated substrate was calcined at 600 °C for 2 h
B cos h
and then heat treated at each given temperature for
h at a heating rate of 5 °C min\ꢂ. Calcining and heat
2
where D is the crystallite size (nm), k is the wavelength
treatment were performed in air. The coating and
calcining processes were repeated five times to obtain
a sufficient thickness for measurements of properties.
The thickness of the films dipped five times was about
of CuK radiation (nm), h is the Bragg angle (deg), k is
a
a constant, 0.89, and B is the calibrated width of
a diffraction peak at half-maximum intensity (rad).
Assessments of crystallite size of the thin film were also
obtained from the SEM observations.
0
.6 lm.
Characterization for the film preparation was per-
The electrical conductivity was measured by the
complex impedance method at 100 Hz to 15 MHz
using an impedance analyser (Hewlett Packard impe-
dance/gain-phase analyser 4194A). A comb pattern of
formed by means of thermogravimetric analysis (TG),
differential thermal analysis (DTA), and scanning
electron microscopy (SEM). Thermal analysis was
5
250

electrodes was screen-printed on the coating surface
using platinum paste (Tokuriki Chemical Research
and Development). The platinum paste was baked at
1
000 °C for 10 min.
The electrical conductivity r (S cm\ꢂ) was cal-
culated from Equation 2
l
r"
(2)
RtN¸
where R ()) is the resistance obtained from Nyquist
diagrams in which the imaginary impedance is plotted
against the real impedance, t (cm) is the film thickness,
N is the number of electrode pairs in the comb pattern,
¸
(cm) is the length of electrodes, and l (cm) is the
separation distance of electrodes.
3
. Results and discussion
Fig. 2 shows TG and DTA curves for the dried pow-
ders of 8 mol % YZ coating solution. The coating
solution was dried at 200 °C for 24 h. Endothermic
and exothermic peaks were found with weight loss
until 500 °C, and no other thermal effect was observed
above this temperature. Their origins may be at-
tributed to the evaporation of solvents or burning of
organic compounds.
Figure 3 Scanning electron micrographs of 8 mol % YZ film heat
treated at 1000 °C for 2 h: (a) cross-section, (b) surface.
An exothermic peak at around 350 °C can be at-
tributed to the crystallization of zirconia. This was
confirmed by the XRD data. All dried samples con-
taining 0—20 mol % Y O exhibit similar DTA curves.
ꢀ
ꢁ
From the results of TG and DTA data, the calcination
temperature was determined as 600 °C, with no other
thermal effects being observed above this temperature.
Fig. 3 shows scanning electron micrographs of
8
mol % YZ film heat treated at 1000 °C for 2 h. The
film was coated five times. A dense film uniformly
coated on an alumina substrate is shown. The thick-
ness of the film coated five times is about 0.6 lm. As
shown in Fig. 3b, the film is crack-free and consists of
Figure 4 Film thickness as a function of the number of dippings.
densely packed crystallites with a diameter of about
5
0 nm. Good uniformity in crystallite size is also
shown.
The film thickness increases linearly as the number
of dippings increases, as shown in Fig. 4. This con-
firms that thickness can be easily controlled by the
number of dippings. A film thickness of about 0.12 lm
was obtained by each dip-coating. We know that
a film thickness of submicrometre order can be con-
trolled by the sol—gel method, although the electro-
chemical vapour deposition (EVD) process mainly
studied for solid-oxide fuel-cell applications [14, 15]
can produce a thickness of micrometre order.
Fig. 5 shows X-ray diffraction patterns of pure
ZrO thin film obtained from different thermal treat-
ments. The open circles in this figure show the peaks
Figure 2 TG and DTA curves for the dried powders of 8 mol % YZ
coating solution.
ꢀ
5251

Figure 5 XRD patterns of ZrO thin film heat treated at various
ꢀ
temperatures for 2 h.
Figure 6 XRD patterns of YZ thin film heat treated at 600 °C for
h.
2
from the alumina substrate. In the figure, t and m de-
note the tetragonal phase and the monoclinic phase,
respectively. At 600 °C, the main peak (111 plane) of
tetragonal phase is shown with a very weak peak of
monoclinic phase. With increasing temperature, the
intensity of the tetragonal peaks decreases, and the
monoclinic phase is largely formed. The orderly pro-
gression of the phase transition is observed with in-
creasing temperature.
It is known that for the zirconia having small crys-
tallite size, the metastable tetragonal phase exists at
low temperatures as a result of the crystallite size effect
[
16]. The crystallite sizes of thin films obtained in this
study are smaller than 30 nm, the critical size of the
crystallite size effect (see Fig. 3b and Fig. 11 below).
For thin films having crystallite sizes below 30 nm, the
results of Fig. 5 are reasonable.
Figs 6—8 show the film composition dependence of
the X-ray diffraction patterns in the YZ specimens,
heat treated at 600, 1000 and 1300 °C, respectively.
All the YZ thin films heat treated at 600 °C have
the tetragonal phase, as shown in Fig. 6. This co-
incides with the stabilization of the tetragonal phase
at low temperatures. Fig. 6 is characteristic of the
trend of peak broadening with increasing yttria con-
tent. The similar tendency is observed up to the XRD
data of YZ heat treated at 800 °C; however, no such
tendency is observed above this temperature (see
Figs 7 and 8).
Figure 7 XRD patterns of YZ thin film heat treated at 1000 °C for
2 h.
These phenomena can be related to the crystalliza-
tion behaviour, which is strongly influenced by the
yttria content. In a detailed study of the dehydration
and crystallization behaviour of zirconia—yttria gels
obtained by coprecipitation, Ramanathan et al. [17]
reported that a sharp exothermic peak in the DSC
profiles due to the crystallization of zirconia was dras-
tically reduced by increasing the yttria content. They
also showed that the temperature of the crystalliza-
tion peak was shifted to higher temperatures with an
5
252

Figure 8 XRD patterns of YZ thin film heat treated at 1300 °C for
h.
2
increase in the yttria content, while the activation
energy and kinetic constant were increased.
Fig. 7 shows XRD patterns of YZ thin film heat-
treated at 1000 °C (c denotes the cubic phase). The
angle dependence of all XRD patterns is very similar
at 1000 °C, showing the tetragonal phase and/or the
cubic phase. From the high-angle pattern of XRD, we
can easily discriminate between tetragonal and cubic
phases [18] as shown in Fig. 9, even though the angles
of the diffraction peaks due to cubic phase overlap
with the peaks from the tetragonal phase in the low
angles below 2h"72°. The cubic phase was seen in
the lower-angle region of the high-angle XRD patterns
Figure 9 High-angle XRD patterns of YZ thin film heat treated at
000°C for 2 h.
1
TABLE I Phase transitions of YZ thin films heat treated at vari-
ous temperatures
of the range 2h"68°—78°. In the figure, (400) and
#
(
400) denote the cubic phase and the tetragonal
ꢀ
Y O content
Heat-treatment temperature
phase, respectively.
ꢀ
ꢁ
(mol%)
From these results, it can be seen that, at the lower
doping concentrations, only the tetragonal phase
exists, while the cubic plus tetragonal phases exist at
the higher doping concentrations. In particular, there
is a feature of the dominant cubic phase in the YZ of
6
00°C
800°C
1000°C
1300°C
0
2
5
8
t#[m]!
t#(m)"
m#(t)"
t#[m]!
t
c#(t)"
c#t
c#t
m#[t]!
m#(t)"
t#(m)"
t#c
t
t
t
t
t
t
t
t#c
t#(c)"
t
8
mol % Y O content. It is well known that the
ꢀ
ꢁ
11
c#t
amount of dopant required to stabilize fully the cubic
2
0
c#t
structure is about 8—9 mol % for Y O , and the con-
ꢀ
ꢁ
ductivity maximum in the electrical conductivity—
dopant concentration curves also corresponds to this
amount [19].
! Very small
Small
"
As the temperature of heat treatment is increased,
YZ of 2 mol % Y O content shows the monoclinic
behaviour, as shown in Fig. 8. This is qualitatively
phases is grown, and the cubic phase becomes
dominant.
ꢀ
ꢁ
similar to the pure ZrO data of Fig. 5, even though
the temperature of the phase transition is dependent
on the doping concentration. With increasing yttria
content the main peak of the tetragonal and/or cubic
The phase transitions of YZ thin films are sum-
marized in Table I. Notations t, m, and c denote the
tetragonal phase, the monoclinic phase and the cubic
phase, respectively. For all the YZ thin films, the
ꢀ
5253

tetragonal phase is stable at low temperatures as a re-
sult of the crystallite size effect. At the higher temper-
atures, the tetragonal phase is transformed into either
the monoclinic phase or the cubic phase, depending
on the doping concentration.
At the lower doping concentrations, the tetrag-
onal phase is transformed into the monoclinic phase.
With increasing doping concentration, the transition
temperature of monoclinic phase becomes higher.
This is the result of the stabilization of tetragonal
phases, due to the addition of a stabilizer [20]. A YZ
thin film of 5 mol % Y O content shows the tetrag-
ꢀ
ꢁ
onal phase over all the heat-treatment temperatures in
this study.
However, at higher doping concentrations, the tet-
ragonal phase of low temperatures is transformed into
the cubic phase. YZ thin film of 8 mol % Y O con-
ꢀ
ꢁ
tent is stabilized to almost cubic phase at 1000 °C.
Also, an increase in the content of tetragonal phase is
observed at 1300 °C.
Sakurai et al. [12] reported similar results in
8
.8 mol % yttria-doped zirconia coatings prepared by
Figure 11 Heat-treatment temperature dependence of the crystal-
lite size of YZ thin film heated for 2 h. The crystallite size was
calculated from the (111) diffraction peak. (᭝) 2 mol %, (ᮀ)
the hydrolysis of zirconium alkoxide with hydrogen
peroxide and nitric acid. They obtained the cubic
phase after heating at 1000 and 1200 °C, and the cubic
and tetragonal phases at 1350 °C. They explained the
results from the fact that the partial transformation to
the tetragonal phase is probably due to the diffusion of
yttrium into the grain boundary and/or into the
alumina substrate.
5
mol %, (᭜) 8 mol %, (᭺) 11 mol %, (᭹) 20 mol %.
Fig. 11 shows the relationship between heat-treat-
ment temperature and crystallite size of YZ thin
film heated for 2 h. The crystallite size was calculated
from the Scherrer equation by using the (111) dif-
fraction peak obtained from XRD-LB measurements.
The crystallite size of all the thin films used in this
study is smaller than the crystal size of 30 nm, above
which the metastable tetragonal phase cannot exist at
lower temperatures as a result of the crystallite size
effect.
The crystallite sizes increase with increasing heat-
treatment temperature, and become saturated at
1300 °C. This result indicates that the crystals of the
films may be grown up to 1300 °C. It is also shown
that the crystallite size at lower temperatures becomes
larger with decreasing doping concentration, which is
in contrast to the heat-treatment temperature of
Fig. 10 shows the compositional dependence of lat-
tice constants in the system of YZ heat treated at
1
000 °C. The closed and open circles represent the
lattice constants of the a-axis and c-axis, respectively.
The lattice constants of a-axis and c-axis overlap with
increasing yttria content, which reveals the phase
transitions from the tetragonal phase into the cubic
phase. This result is in good agreement with the third
column of Table I. Another feature is the increase in
lattice constants with the yttria content. This suggests
that yttrium ions are distributed uniformly in the solid
solution of cubic zirconia. The radius of the yttrium
ion is bigger than that of the zirconium ion
(Yꢁ>0.096 nm, Zrꢃ> 0.082 nm).
1
300 °C. This may be responsible for an interference of
zirconia crystallization with the existence of yttria,
showing that the nuclei formation during crystalliza-
tion is suppressed by yttria.
The variation of crystallite sizes with the doping
concentration at constant temperature reflects the
variation of XRD peak intensities, where the peak
becomes broad with increasing yttria content, as
shown in Fig. 6.
The electrical properties of YZ thin film are dis-
cussed. In order to measure the electrical conductiv-
ity of materials, the contribution of the electrode
resistance at the electrode/material interface must be
eliminated. This was effectively achieved with the
complex impedance techniques as introduced in this
study.
Fig. 12 shows the compositional dependence of
electrical conductivity in the system of YZ heat treated
at 1300 °C. The conductivity was measured at 600 °C
Figure 10 Lattice constants of YZ thin film heat treated at 1000 °C
for 2 h: (᭺) c-axis, (᭹) a-axis.
5
254

models. However, for concentrated solutions in the
low-temperature range, vacancies form associates by
the coulombic interactions. Thus, the carrier concen-
tration becomes diminished, and then the conductivity
is decreased with increasing doping concentration. We
cannot discriminate the correct mechanism for this
study, but the data of the conductivity behaviour are
reasonable.
4
. Conclusions
Y O -doped and ZrO thin films were prepared by
ꢀ
ꢁ
ꢀ
the sol—gel process using the dip-coating method. The
following conclusions are drawn.
1
. A uniform thin film was prepared by using zirco-
nium chloride octahydrate and yttrium chloride
hexahydrate as raw materials, with ethylene glycol,
2
-butanol and distilled water as solvents.
2. YZ crystalline films were produced by the ther-
Figure 12 Electrical conductivity of YZ thin film measured at
00°C in air. The specimens heat treated at 1300 °C for 2 h were
used.
6
mal treatment above 600 °C, at which temperature
raw materials were completely decomposed.
3
. In all YZ thin films, the tetragonal phase was
stable at low temperatures as a result of the crystallite
size effect; however, at the higher temperatures the
tetragonal phase was transformed into either the mono-
clinic phase or the cubic phase, depending on the
doping concentration. YZ thin film of 8 mol % Y O
in air. The line is drawn as a guide to the eye.
Logr (S cm\ꢂ) has a maximum value at around
1
0 mol %, and then decreases with increasing yttria
content. It is known that the location of the conductiv-
ity maximum corresponds to the minimum amount of
the dopant required to stabilize fully the fluorite
phase.
The conductivity of YZ thin films prepared in this
study was similar to the values of the sintered body,
the single crystal and the plasma-sprayed films re-
ported by Chiodelli et al. [21].
ꢀ
ꢁ
content was stabilized to almost cubic phase at
000 °C.
. Reasonable conductivity behaviour at YZ was
1
4
observed for the YZ thin films. The electrical conduct-
ivity of YZ thin films was similar to the values of the
sintered body.
According to the general expression r"r exp
ꢄ
(
!*H/R¹), the variations in log r and the activa-
ꢄ
tion energy, *H, with the dopant content were ob-
References
tained from the Arrhenius plots between the conduct-
ivity and the measured temperature. The result is
1. Y. TAKAHASHI, K. NIWA, K. KOBAYASHI and M. MAT-
SUKI, ½ogyo-Kyokai-Shi 95 (1987) 942.
L. YANG and J. CHENG, J. Non-Crystl. Solids 112 (1989)
442.
2
.
characteristic of the tendency for log r to increase
ꢄ
with increasing yttria content. The pre-exponential
3.
K. IZUMI, M. MURAKAMI, T. DEGUCHI and A. MORITA,
J. Amer. Ceram. Soc. 72 (1989) 1465.
factor, r , is known to be proportional to the carrier
ꢄ
concentration and a square of the jumping distance. In
4. P. DE LIMA NETO, M. ATIK, L. A. AVACA and M. A.
AEGERTER, J. Sol-Gel Sci. ¹echnol. 2 (1994) 529.
this case, the increase in log r can be approximated
ꢄ
5
6
.
.
R. CARUSO, N. PELLEGRI, O. DE SANCTIS, M. C.
CARACOCHE and P. C. RIVAS, ibid. 3 (1994) 241.
M. ATIK, C. R. KHA, P. DE LIMA NETO, L. A. AVACA,
M. A. AEGERTER and J. ZARZYCKI, J. Mater. Sci. ¸ett. 14
(1995) 178.
to the increase of the carrier concentration, because
the variation in the jumping distance with the yttria
dopant is much smaller than that in the carrier con-
centration.
Therefore, the increase in the conductivity up to the
maximum in the conductivity—dopant concentration
curves of Fig. 12 can be explained by the increase in
the carrier concentration. Many yttrium cations sub-
stitute for Zrꢃ> in the lattice sites and as a result
of their lower valence state, they also create more
vacancies in the oxygen sublattice.
The decrease in conductivity from a maximum
value with increasing yttria content, however, can be
explained in terms of a blocking effect of the larger
cations. The migration of vacancies is more effectively
blocked by larger dopant cations.
The conductivity behaviour in dilute solid solutions
and in the high-temperature range, when vacancies are
free and mobile, is more easily explained by such
7
8
.
.
A. NEGISHI, ½ogyo-Kyokai-Shi 93 (1985) 566.
M. SHANE and M. L. MECARTNEY, J. Mater. Sci. 25 (1990)
1
537.
9
.
R. D. MAGGIO, P. SCARDI and A. TOMASI, Mater. Res.
Soc. Symp. Proc. 180 (1990) 481.
10. P. PESHEV and V. SLAVOVA, Mater. Res. Bull. 27 (1992)
269.
1
1
1. T. W. KUEPER, S. J. VISCO and L. C. DE JONGHE, Solid
State Ionics 52 (1992) 251.
1
2. C. SAKURAI, T. FUKUI and M. OKUYAMA, J. Amer.
Ceram. Soc. 76 (1993) 1061.
13. K. MIYAZAWA, K. SUZUKI and M. Y. WEY, ibid. 78 (1995)
47.
4. U. B. PAL and S. C. SINGHAL, J. Electrochem. Soc. 137
1990) 2937.
5. M. C. CAROLAN and J. N. MICHAELS, Solid State Ionics 37
1990) 189.
16. R. C. GARVIE, J. Phys. Chem. 69 (1965) 1238.
3
1
1
(
(
5255

1
7. S. RAMANATHAN, R. V. MURALEEDHARAN, S. K.
(The American Ceramic Society, Westerville, OH, 1988)
p. 71.
ROY and P. K. K. NAYAR, J. Amer. Ceram. Soc. 78 (1995)
4
29.
8. J. S. LEE, J. I. PARK and T. W. CHOI, J. Mater. Sci., 31
1996) 2833.
9. T. H. ETSELL and S. N. FLENGAS, Chem. Rev. 70 (1970)
39.
0. V. S. STUBICAN, in ‘‘Advances in Ceramics’’, Vol. 24A,
edited by S. Somiya, N. Yamamoto and H. Yanagida
21. G. CHIODELLI, A. MAGISTRIS, M. SCAGLIOTTI and
F. PARMIGIANI, J. Mater. Sci. 23 (1988) 1159.
1
1
2
(
3
Received 4 December 1995
and accepted 10 March 1997
ꢀ
.
5
256