Inuence of alumina addition on grain growth and room

temperature mechanical properties of 8YSCZ/Al

2
O
3
composites
S. Tekeli
*
Materials Division, Metallurgy Education Department, Faculty of Technical Education, Gazi University, 06500, Besevler-Ankara, Turkey
Received 8 July 2004; received in revised form 5 October 2004; accepted 31 October 2004
Available online 16 December 2004
Abstract
The eect of the amount of alumina addition on grain growth and room temperature mechanical properties (hardness and frac-
ture toughness) of 8YSCZ/Al
2
O
3
composites were determined using 8 mol% yttria-stabilized cubic zirconia (8YSCZ) and various
amounts of Al
2
O
3
(010 wt%). The experimental results showed that the grain growth in Al
2
O
3
doped 8YSCZs occured slowly
and more sluggish than that in undoped 8YSCZ. Also, the grain growth rate decreased with increasing Al
2
O
3
content. The grain
growth exponent value and the activation energy for undoped 8YSCZ were found to be 2 and 289 kJ/mol, respectively. The addition
of Al
2
O
3
raised the grain growth exponent value to 3 and activation energy for the grain growth process was increased from 289 to
410 kJ/mol. Also, the introduction of Al
2
O
3
enhanced the hardness and fracture toughness of 8YSCZ/Al
2
O
3
composites. The hard-
ness and fracture toughness increased with increasing Al
2
O
3
content. The maximum hardness and fracture toughness values reached
1459 kgf/mm
2
and 2.41 MPa m
1/2
, respectively, with the addition of 10 wt% Al
2
O
3
while these values were 1314 kgf/mm
2
and
1.5 MPa m
1/2
for undoped 8YSCZ.
2004 Elsevier Ltd. All rights reserved.
Keywords: Cubic zirconia; Grain growth; Fracture toughness; Hardness and alumina
1. Introduction
Zirconia is a widely used ceramic that exists in three
dierent crystal structures, i.e., monoclinic, tetragonal
and cubic depending on temperature and chemistry.
Varying amounts of oxide additives, such as CaO,
MgO, Y
2
O
3
and most rare-earth oxides, can stabilize
ZrO
2
in the tetragonal and cubic phases [1]. Among
these phases, cubic zirconia with uorite structure is
the most attractive zirconia ceramic to be used as a solid
electrolyte in solid oxide fuel cells (SOFC) due to its high
oxygen ionic conductivity and chemical stability over
wide ranges of temperature and oxygen partial pressure
[2]. In SOFC applications, not only high ionic conduc-
tivity, but also better mechanical, chemical and electrical
stability are required at high working temperatures
(1000 C is usually needed for SOFC and such a high
working temperature leads to a lower working e-
ciency). Also low mechanical strength of cubic zirconia
at higher and longer working temperatures would limit
their use as electrolytes, because they may fracture due
to thermal stresses, mechanical stresses during opera-
tion, lack of transformation toughening and severe grain
growth [3]. It is known that cubic zirconia has large
grain sizes and high grain growth rates [4]. Therefore
the enhancement of mechanical properties of the solid
electrolyte is important problem to be solved. Many ap-
proaches have been made to enhance the mechanical
strength and hinder the severe grain growth in cubic zir-
conia [5]. One such approach is to use the composite
way by dispersing second phase particles. In the com-
posites, it has been known that the remarkable mis-
matching of lattice parameters and thermal expansion
0266-3538/$ - see front matter 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.compscitech.2004.10.024
*
Tel.: +90 312 4399760; fax: +90 312 2120059.
E-mail address: stekeli@gazi.edu.tr.
Composites Science and Technology 65 (2005) 967972
COMPOSITES
SCIENCE AND
TECHNOLOGY
www.elsevier.com/locate/compscitech
coecients between matrix and second phase particles
creates grain growth inhibition and thus the enhance-
ment of room temperature mechanical properties. The
second phase particles having high elastic modulus and
large aspect ratio also give a remarkable shielding eect
during crack propagation by crack-bridging and crack-
deection mechanisms. Therefore, the aim of the present
study was to determine the eect of the Al
2
O
3
dopant in
amounts up to 10 wt% on the grain growth, fracture
toughness and hardness behaviour of 8YSCZ/Al
2
O
3
composites.
2. Materials and methods
8YSCZ/Al
2
O
3
composites containing up to 10 wt%
Al
2
O
3
were produced from the commercial powders:
8 mol% yttria-stabilized cubic zirconia (8YSCZ) pow-
der, Tosoh, Japan and high purity (>99.999%) a-Al
2
O
3
powder, Sumitoma, Japan. The average particle sizes
were 0.3 lm for 8YSCZ and 0.4 lm for Al
2
O
3
. The
chemical composition of the powders is listed in Table 1.
A colloidal processing was used for the preparation
of specimens for fracture toughness, hardness, and grain
growth measurements. The details of specimen prepara-
tion have been given elsewhere [6]. The specimens were
presintered at 950 C to make them more handleable
and sintered at 1450 C in air at a constant heating rate
of 200 C/h for 1 h.
For fracture toughness, hardness and grain growth
measurements, 8YSCZ/Al
2
O
3
specimens containing
Al
2
O
3
dopant in amounts up to 10 wt% were used.
The density of sintered specimens was determined by
the Archimedes method. The theoretical densities of
the composites were estimated by the rule of mixtures,
using 3.99 and 5.95 g/cm
3
for the theoretical densities
of alumina and cubic-zirconia, respectively. The relative
density was estimated on the assumption that the sin-
tered body was of the cubic phase and based on the the-
oretical density of 5.92, 5.81 and 5.67 g/cm
3
for the
specimens doped with 1, 5 and 10 wt% Al
2
O
3
,
respectively.
The specimens for grain growth measurements were
rst sintered at 1400 C for 1 h before annealing. Grain
growth was examined by annealing the sintered speci-
mens for 10, 50 and 100 h at temperatures between
1400 and 1600 C. After annealing, the specimens were
sectioned, ground, polished to 1 lm surface nish and
nally thermally etched in air for 30 min at a tempera-
ture 50 C lower than annealing temperature. Scanning
electron microscopy (SEM) equipped with an energy
dispersive X-ray spectrometer (EDS) attachment was
used to characterize the microstructure of as-sintered
and annealed specimens. Grain sizes were measured by
the mean linear intercept method. An average grain size,
D, was obtained as
D 1:78L, 1
where L is the average grain boundary intercept length
of random lines on the photomicrographs. Phase con-
tent was determined using XRD with Cu Ka radiation.
A scan speed of 0.1 2h/min was employed in the 2060
2h range.
The hardness and fracture toughness of the specimens
were determined to attempt to quantify the eect of
Al
2
O
3
addition on room temperature mechanical prop-
erties. Both hardness and fracture toughness were mea-
sured using a Vickers indenter. Vickers hardness
measurements were carried out on 8YSCZ/Al
2
O
3
com-
posites containing up to 10 wt% Al
2
O
3
, sintered at
1450 C for 1 h. Indentations were made on polished
samples with a load of 20 kg held for 15 s. Ten tests were
conducted for each composite. The hardness was calcu-
lated from the diagonal length of the indentation opti-
cally determined for each indentation, using the
following equation
H
v
1:854P=d
2
; 2
where H
v
is Vickers hardness, P is applied load (kg) and
d is mean of the diagonal length (mm). The fracture
toughness was determined by the indentation technique.
Polished specimens were indented at 10 dierent loca-
tions. Before indentation, a drop of oil was deposited
onto the surface of the specimens to minimize mois-
ture-enhanced crack growth. The size of the cracks
emanating the indentation centre was measured. Mea-
surements were performed immediately after each
indentation in order to avoid any dierence in crack
length due to the environmental variables such as mois-
ture and time. The indentation fracture toughness was
calculated using the formula proposed by Anstis et al.
[7] (half-penny crack)
K
IC
0:0016E=H
v

1=2
P=C
3=2
, 3
Table 1
Chemical composition of the powders
Materials Composition in wt%
ZrO
2
(+HfO
2
) Y
2
O
3
Al
2
O
3
SiO
2
TiO
2
Fe
2
O
3
Na
2
O CaO
8YSCZ 85.9 13.6 0.25 0.1 0.10 0.003 0.01 0.02
a-Al
2
O
3
99.9 0.040.08 0.010.02 0.08
968 S. Tekeli / Composites Science and Technology 65 (2005) 967972
where K
IC
is fracture toughness, E is Youngs modulus,
H
v
is Vickers hardness, P is the load and C is the crack
length.
3. Results and discussion
X-ray diraction analysis showed that all the com-
posites were composed of only cubic phase and there
were no diraction peaks of second phases for all com-
positions. XRD results also showed that Al
2
O
3
had very
limited solubility of 0.3 wt% into 8YSCZ [6]. As the
solubility limit of Al
2
O
3
in 8YSCZ is 0.3 wt%, Al
2
O
3
can hardly form a solid solution with 8YSCZ. There-
fore, Al
2
O
3
mostly segregates around the 8YSCZ parti-
cles and at grain boundaries. Fig. 1 shows representative
microstructures of the series of 8YSCZ/Al
2
O
3
compos-
ites with Al
2
O
3
addition in amount up to 10 wt%, an-
nealed at 1500 C for 10 h. The comparison of the
grain size of the alloys at the same heat treatment indi-
cated that grains were larger in undoped 8YSCZ than in
Al
2
O
3
doped 8YSCZs. Also, the grain size decreased
with increasing Al
2
O
3
content. As shown in Fig. 1,
Al
2
O
3
particles can be seen for specimens with Al
2
O
3
content >5 wt%. This suggests that only small part of
Al
2
O
3
dissolved in 8YSCZ and the rest of Al
2
O
3
segre-
gates at grain boundaries of 8YSCZ.
Isothermal grain growth can be determined by the
following phenomenological grain growth equation:
D
n
D
n
0
k
0
t expQ=RT, 4
where D is the grain size at time t, D
0
is the initial grain
size at time t
0
, n is a constant for a given grain growth
mechanism, k is a temperature-dependent constant, k
0
is a temperature insensitive constant and Q, R, T are
activation energy, gas constant and absolute tempera-
ture, respectively. The grain growth exponent can have
a value between 1 and 4 depending on the rate limiting
step during grain boundary migration. The grain growth
exponent, n, is calculated from the slope of the logD
(grain size) vs. logt (time) plot which is equal to 1/n.
An example for such a plot is given in Fig. 2 for anneal-
Fig. 1. SEM micrographs of 8YSCZ/Al
2
O
3
composites annealed at 1500 C for 10 h, containing: (a) 0 wt% Al
2
O
3
; (b) 1 wt% Al
2
O
3
; (c) 5 wt% Al
2
O
3
;
(d) 10 wt% Al
2
O
3
.
S. Tekeli / Composites Science and Technology 65 (2005) 967972 969
ing at 1400 C. It is evident that grains became larger
during high temperature annealing and the extent of
grain growth increased with increase in annealing tem-
perature, time and with decrease in Al
2
O
3
content. Also
it can be seen from Fig. 2 that grain growth rates could
be controlled by the deliberate addition of grain bound-
ary phase of Al
2
O
3
. The grain growth exponent value of
2 for undoped 8YSCZ was obtained. However, with the
addition of Al
2
O
3
up to 10 wt%, the n value was in-
creased to 3. In the case of single phase growth, the
exponent n usually takes a value of 2 which shows nor-
mal grain growth. In a dual phase structure, the steady
state grain growth would occur such that the growth
of one phase is aected by other phase grains, i.e., the
grain growth is mutually restricted by Zeners pinning
because of minor phase grains, as in metallic materials
and that the coarsening of the minor phase is approxi-
mated by Ostwald ripening then n = 3 can be predicted
for lattice-diusion controlled grain growth [8].
The activation energy for grain growth is calculated
from the slope of logD
n
D
n
0
=t vs: 1=T as shown in
Fig. 3. It can be seen from this gure that, the activation
energy, Q, for Al
2
O
3
doped 8YSCZs is higher than that
for undoped 8YSCZ. The activation energy of undoped
8YSCZ was calculated as 289 kJ/mol. However the
addition of Al
2
O
3
to 8YSCZ up to 10 wt% resulted in
a gradual increase in activation energy. The activation
values obtained are 361 kJ/mol for 1 wt% Al
2
O
3
doped
8YSCZ, 391 kJ/mol for 5 wt% Al
2
O
3
doped 8YSCZ and
410 kJ/mol for 10 wt% Al
2
O
3
doped 8YSCZ. These re-
veal an increase in activation energy for the grain
growth of 8YSCZ/Al
2
O
3
with increasing Al
2
O
3
content.
This increase in activation energy is accompanied by a
decrease in grain size.
In order to reduce or suppress the grain growth the
grain boundary mobility and energy can be reduced
for example by impurities, pores, dopants or particles
of a second phase. In polycrystalline materials, it has
been shown that particles of a second phase with limited
solubility are especially eective in pinning grain bound-
aries and thus in minimizing static and dynamic grain
growth [9]. As stated by Sturm et al., in order to provide
ecient grain boundary pinning in a nanocrystalline
structure the particles of the second phase have to full
several of the following conditions: (a) small diusion
coecient of the solute cations into the matrix; (b) small
or no solubility of the solute cations in the matrix; (c) a
small interfacial energy between the two phases; (d) a
homogeneous distribution of the second phase and; (e)
stability against dissolution and coarsening [10]. The
grain boundary mobility in pure matter is written as [10]
M D
gb
=rRT, 5
where D
gb
is the grain boundary diusion coecient, r
is the grain boundary width, R is gas constant and T
is absolute temperature. As can be seen from this equa-
tion, the grain boundary mobility decreases with
increasing grain boundary width. Increasing the grain
boundary width does not only decrease grain boundary
mobility and inhibit grain growth, but also increases the
diusion length over which stress concentration can be
rapidly relaxed. Additives and dopants that increase
grain boundary width may be eective in minimizing
grain growth. However, the mobility is not the only sig-
nicant inuence on grain growth. Grain growth is also
dependent on the grain boundary energy. A rate equa-
tion for grain growth can be given as [11]
D
n
D
n
0
2McVt, 6
where D is the instantaneous grain size at time t, D
0
is
the initial grain size at time t = 0, n is the grain growth
exponent, M is the mobility, c is the grain boundary en-
ergy and V is the atomic volume. Additives that decrease
grain boundary energy have an additional eect on min-
imizing grain growth. A reduction in grain boundary en-
ergy occurs when intergranular phases pin the grain
boundaries. The eectiveness of using Al
2
O
3
to pin grain
boundaries and limit grain growth can be estimated by
Zener pinnig [2] which shows the dependence of the
10 100
1
10
l
o
g

(
g
r
a
i
n

s
i
z
e
,

m
)
log (time, h)
undoped 8YSCZ, n:2
1 wt% Al
2
O
3
doped 8YSCZ, n:3
5 wt% Al
2
O
3
doped 8YSCZ, n:3
10 wt% Al
2
O
3
doped 8YSCZ, n:3
Fig. 2. Log (grain size) vs. log (time) for annealing at 1400 C.
5.3 5.4 5.5 5.6 5.7 5.8 5.9 6.0
-11
-10
-9
-8
-7
-6
-5
-4
undoped 8Y SCZ , Q:289
1 w t% Al
2
O
3
doped 8Y SCZ , Q:361
5 w t% Al
2
O
3
doped 8Y SCZ , Q:391
10 w t% Al
2
O
3
doped 8Y SCZ , Q:410
l
o
g

(
(
D
n
-
D
o
n
)
/
t
)
10
4
/T (1/K)
Fig. 3. Arrhenius plots for 8YSCZ containing 010 wt% Al
2
O
3
.
970 S. Tekeli / Composites Science and Technology 65 (2005) 967972
grain size of matrix on the volume and grain size of sec-
ond phase
G kr=f , 7
where G is grain size of matrix, k is constant, r is grain
size of second phase and f is volume fraction of second
phase. This equation also indicates that grain size of ma-
trix decreases with increasing volume fraction of second
phase [2].
In the present study, the dierences in grain growth in
undoped 8YSCZ and Al
2
O
3
doped 8YSCZs can be re-
lated to dierences in segregation of the solute cations
at grain boundaries. In the light of the above results, it
is deduced that the grain size stability is associated with
segregation of solute cation to grain boundaries, which
lowers grain boundary mobility and grain boundary en-
ergy, thus increasing the cohesive strength of the grain
boundary and the diusion distance across the bound-
ary. Low solubility of Al
2
O
3
acts as a barrier against dif-
fusion. Due to the lower solubility of Al
2
O
3
in 8YSCZ
[6], grain growth is expected to be slower in Al
2
O
3
con-
taining specimens. Compared to Al
2
O
3
doped 8YSCZ,
undoped 8YSCZ suers fast grain growth and only lim-
ited cation segregation to grain boundaries; conse-
quently, undoped 8YSCZ has a lower grain boundary
cohesive strength and a higher grain boundary mobility
and energy.
Hardness of ceramics is usually aected by the intrin-
sic deformability of the ceramic and microstructural fea-
tures such as multiphases, grain size and orientation,
porosity and grain boundary constitution [12]. In the
present study, the room temperature mechanical proper-
ties evaluated were hardness and fracture toughness.
The hardness and fracture toughness (K
IC
) of the
8YSCZ/Al
2
O
3
composites sintered at 1450 C for 1 h
were determined to investigate the eect of the Al
2
O
3
addition and content on room temperature mechanical
properties. These were performed using Vickers hard-
ness testing machine with a load of 20 kg for dwelling
time of 15 s. The result of the Vickers hardness carried
out on the various amount of Al
2
O
3
doped 8YSCZ com-
posites is shown in Fig. 4. It was seen that the addition
of Al
2
O
3
resulted in an enhancement of Vickers hard-
ness and also, the hardness values increased with
increasing Al
2
O
3
content. The hardness of the compos-
ites containing up to 10 wt% Al
2
O
3
is larger than that
of monolith 8YSCZ. This increase in hardness can be
attributed to smaller grain size of composites. As it
was mentioned before, the grain size of composites de-
creased with increasing Al
2
O
3
content. Plastic deforma-
tion of 8YSCZ matrix with up to 10 wt% Al
2
O
3
can be
much harder than that of monolith 8YSCZ only, be-
cause grain boundaries work as an obstacle for plastic
deformation. The increase in hardness with increasing
Al
2
O
3
content can also be explained by increased
amount of hard Al
2
O
3
grains in the matrix.
Ceramics are brittle materials and many attempts
have been made to increase their toughness by manipu-
lating their microstructures. One technique is to add a
reinforcing phase or a matrix phase to the polycrystal-
line ceramic. A second technique is to add a second
phase which has a thermal mismatch with the matrix
such that internal stress elds can deect or pin cracks.
The use of Al
2
O
3
as a second phase addition can be in-
cluded within this group. Al
2
O
3
particles have been
added to many polycrystalline ceramics to increase
toughness [12]. In a single crystal and glass, the crack
can follow a planar, smooth path and meet no obstruc-
tion. As a result, such materials have very low fracture
toughness. Most ceramics in commercial use are poly-
crystalline. Each grain generally has a dierent crystal
orientation than adjacent grain. A crack passing
through a polycrystalline ceramic does not follow a
smooth planar path. It follows grain boundaries around
some grains and fractures other grains. This results in
greater fracture energy and greater fracture toughness
[13].
In the present study, the eect of various amount of
Al
2
O
3
addition on the fracture toughness of 8YSCZ/
Al
2
O
3
composites is studied. It is observed that fracture
toughness values increased as the Al
2
O
3
content in-
creased. For example, undoped 8YSCZ had a fracture
toughness of 1.5 MPa m
1/2
whereas 8YSCZ/Al
2
O
3
com-
posite containing 10 wt% Al
2
O
3
had a fracture tough-
ness of 2.41 MPa m
1/2
. The relation between fracture
toughness and Al
2
O
3
content is shown in Fig. 5. The in-
crease in fracture toughness in specimens with higher
amount of Al
2
O
3
particles could be attributed to incre-
ment of Al
2
O
3
and thus smaller grain size. The increase
in fracture toughness with increasing amount of Al
2
O
3
can also be explained in term of the crack deection at
crack tip, bridging at process zone wake by second
phase, the mode of fracture and the residual stress in
composites. SEM observations showed that the cracks
propagated straight through the 8YSCZ grains, i.e.,
0 2 4 6 8 10
1300
1350
1400
1450
H
a
r
d
n
e
s
s

(
K
g
f
/
m
m
2
)
Al
2
O
3
Amount (wt %)
Fig. 4. The eect of Al
2
O
3
content on hardness of 8YSCZ/Al
2
O
3
composites sintered at 1450 C for 1 h.
S. Tekeli / Composites Science and Technology 65 (2005) 967972 971
transgranular fracture mode for monolith, while it was
deected by Al
2
O
3
particles in 8YSCZ/Al
2
O
3
composite
in which reinforcing particles are dispersed nely and
uniformly to achieve an eective toughening of the
8YSCZ matrix. Therefore crack deection was the rea-
son for fracture toughness improvement in composites.
The deection was caused as a result of the residual
stresses associated with dierential Youngs modulus
and the coecients of the thermal expansion between
8YSCZ (10.5 10
6
K
1
) and Al
2
O
3
(8.8 10
6
K
1
)
grains [14]. In the 8YSCZ matrix, tensile and compres-
sive stresses are developed. On the other hand, the
Al
2
O
3
grains and 8YSCZ/Al
2
O
3
interface are in com-
pression. When a crack propagates through composite,
residual stress eld in the matrix attracts the crack to-
wards the Al
2
O
3
particles and the crack either pass
through the Al
2
O
3
particles or propagate along the par-
ticle/matrix interface. In either case, crack deection and
fracture energy of Al
2
O
3
particles and/or 8YSCZ/Al
2
O
3
interface leads to an improvement in fracture toughness
of the composite [14].
4. Conclusions
The grain growth and room temperature mechanical
properties of hardness and fracture toughness of
8YSCZ/Al
2
O
3
composites having Al
2
O
3
addition of
010 wt% were investigated. It was found that the grain
growth exponent value and the activation energy for un-
doped 8YSCZ were found to be 2 and 289 kJ/mol,
respectively. The addition of Al
2
O
3
raised the grain
growth exponent value to 3 and activation energy for
the grain growth process was increased from 289 to 410
kJ/mol for the addition of Al
2
O
3
from 0 to 10 wt%.
Grain growth in Al
2
O
3
doped 8YSCZ occurred slowly
and more sluggish than that in undoped 8YSCZ. This
is mainly due to the lower grain boundary mobility and
energy which results from solute segregation in the grain
boundary and its drag in Al
2
O
3
doped 8YSCZ but not in
undoped 8YSCZ. The drag eect arises from any pre-
ferred segregation of impurities either to or from grain
boundary area because of size and charge dierences.
Al
2
O
3
addition is expected to segregate to grain bound-
aries. This segregation layer is believed to hinder grain
growth by resulting in limiting matter transfer along
the grain boundary. The results of the Vickers hardness
and indentation fracture toughness of 8YSCZ/Al
2
O
3
composites showed that the hardness of 8YSCZ/Al
2
O
3
composites increased with increasing Al
2
O
3
content
and this increase was due to the decreased grain size.
Also fracture toughness increased with the increase of
Al
2
O
3
content. Generally, the increase of fracture tough-
ness can be attributed to smaller grain size, the crack
deection and change of fracture mode from transgran-
ular to intergranular mode.
Acknowledgements
This work has been supported by DPT (the State
Planing Organization of Turkey) under Project No.
2003K120470-18. The author is grateful to the DPT
for nancial support and Gazi University for the provi-
sion of laboratory facilities.
References
[1] Adams JW, Ruh R, Mazdiyasni KS. J Am Ceram Soc
1997;80(4):9038.
[2] Bamba N, Choa YH, Sekino T, Niihara K. J Eur Ceram Soc
1998;18:6939.
[3] Zhang T, Zeng Z, Huang H, Hing P, Kilner J. Mater Lett
2002;57:1249.
[4] Lee IG, Chen I-W. In: Somiya S, Yoshimura M, Watanabe R,
editors. Sintering 87, 1. London: Elsevier; 1988. p. 340.
[5] Kwon N-H, Kim G-H, Song HS, Lee H-L. Mater Sci Eng A
2001;99:18594.
[6] Tekeli S, Demir U. Ceram Int, in press.
[7] Anstis GR, Chantikul P, Lawn BR, Marshall DB. J Am Ceram
Soc 1981;64:53343.
[8] Zhao J, Ikuhara Y, Sakuma T. J Am Ceram Soc
1998;81(8):208792.
[9] Sturm A, Betz U, Scipione U, Hahn H. NanoStruct Mater
1999;11(5):65161.
[10] Turnbull D. Trans AIME 1951;191:78.
[11] Sharif AA, Imamura PH, Mecartney ML. Mater Sci Forum
1999;304306:44350.
[12] Lee WE, Mark Rainforth W. Ceramic microstructures: property
control by processing. UK: Chapman and Hall; 1994.
[13] Richerson DW. Modern ceramic engineering; properties,
processing and use in design. New York: Marcel Dekker,
Inc.; 1992.
[14] Kwon N-H, Kim G-H, Song HS, Lee H-L. Mater Sci Eng A
2001;299:18594.
0 2 4 6 8 10
1.4
1.6
1.8
2.0
2.2
2.4
F
r
a
c
t
u
r
e

T
o
u
g
h
n
e
s
s

(
M
P
a
m
1
/
2
)
Al
2
O
3
Amount (wt %)
Fig. 5. The eect of Al
2
O
3
content on fracture toughness of 8YSCZ/
Al
2
O
3
composites sintered at 1450 C for 1 h.
972 S. Tekeli / Composites Science and Technology 65 (2005) 967972