Fracture Toughness of Ceramics Fired at Different Temperatures

Abstract

The fracture toughness test was performed at room temperature on sets of 5 ceramic samples made from material for high voltage insulators (kaolin 36 wt. %, Al 2 O 3 30 wt. %, clay 12 wt. % and feldspar 22 wt. %) fired at temperatures 400, °C at heating and cooling rate of 5 °C/min. The precrack was made to each sample by indentation under the loads 10 N – 200 N, the dwell time was 45 s and the loading rate was 10 N/s. Results of the fracture toughness tests were in accordance with changes of structure of the samples after the partial firings. Fracture toughness from 20 °C to 500 °C is almost constant and it varies between 0.1 MPa·m 0.5 and 0.2 MPa·m 0.5 . Dehydroxylation (420 °C – 600 °C) does not influence the value of fracture toughness. At temperature interval where we assume sintering (700 °C – 1250 °C) we observe exponential dependence of fracture toughness up to 1.5 MPa·m 0.5 . From comparison of the fracture toughness, Young's modulus and flexural strength follows a correlation and proporcionality of these mechanical properties.

Full text

ISSN 1392–1320 MATERIALS SCIENCE (MEDŽIAGOTYRA). Vol. XX, No. XX. 2011
Fracture Toughness of Ceramics Fired at Different Temperatures
Peter SIN 1, Renno VEINTHAL 2, Fjodor SERGEJEV 2, Maksim ANTONOV 2, Igor STUBNA 1 ∗
1 Department of Physics, Faculty of Natural Sciences, Constantine the Philosopher University in Nitra, Slovakia
2 Department of Materials Engineering, Faculty of Mechanical Engineering, Tallinn University of Technology, Estonia
Received 18 May 2011; accepted 03 June 2011
The fracture toughness test was performed at room temperature on sets of 5 ceramic samples made from
material for high voltage insulators (kaolin 36 wt. %, Al2O3 30 wt. %, clay 12 wt. % and feldspar 22 wt.
%) fired at temperatures 400, 500, 600, 700, 800, 900, 1000, 1100, 1200, 1250, 1300, 1400, 1500 °C at
heating and cooling rate of 5 °C/min. The precrack was made to each sample by indentation under the
loads 10-200 N, the dwell time was 45 s and the loading rate was 10 N/s. Results of fracture toughness
tests were in accordance with changes of structure of the samples after the partial firings. Fracture
toughness from 20 to 500 °C is almost constant and it varies between 0.1 and 0.2 MPa·m0.5.
Dehydroxylation (420 – 600 °C) does not influence the value of fracture toughness. At temperature
interval where we assume sintering (700 – 1250 °C) we observe exponential dependence of fracture
toughness up to 1.5 MPa·m0.5. From comparison of the fracture toughness, Young modulus and flexural
strength follows a correlation and proporcionality of these mechanical properties.
Keywords: ceramic, fracture toughness, mechanical strength, solid-state sintering, liquid-state sintering,
dehydroxylation.
INTRODUCTION∗
Mechanical parameters are important characteristics
of the ceramic materials. Each ceramic product is
mechanically stressed in technological processes during
drying and firing as well as in actual service. For example
it is known that increasing of the temperature during
firing is possible only until limit, when thermomechanical
stress is lower than mechanical strength [1, 2].
The dependence of the mechanical strength during
firing was investigated in [3, 4]. Classical ceramics based
on kaolinite and illite is presented in many fields of
industry. In the last decades, a significant progress in
increasing of the mechanical strength was made for
alumina porcelain.
Mechanical strength after firing at different
temperatures was measured in [5]. However, in literature
we did not find results for the fracture toughness (KIc) for
porcelain, apparently because this value is low, so no
attention was given to this problem.
Ceramic materials are brittle. A lot of research is
conducted to decrease the brittleness of advanced ceramics
[6, 7]. This brittleness may be judged indirectly by KIc
which denotes material's resistance to brittle fracture when
a crack of critical size is present. The subscript Ic denotes
the mode I crack opening under the normal tensile stress
perpendicular to the crack path. Brittle fracture is very
characteristic for materials with low fracture toughness
[8]. A related concept is the work of fracture (γwof), which
is directly proportional to
EK
Ic
/
2
, where E is the Young's
modulus of material [9].
The aim of this article is to find out how KIc of the
ceramic material depends on the firing temperature.
 Corresponding author. Tel.: +421-37-6408-620.
E-mail address: istubna@ukf.sk (I. Stubna)
THEORETICAL PART
According to standard ASTM C1421-10 [10], the
dimensions of the sample subjected to fracture toughness
test, as well as the direction and position of force
application and precrack orientation, indentation location
and orientation, and indentation and (Palmqist) crack
dimensions are depicted at Fig. 1 where dW and dB means
the maximum deviation of sample dimensions acording to
standard [10]
Fig. 1. Dimension of the sample, indentation and location
orientation and determinimg the crack dimensions
For the sc (surface crack) method and the three-point-
bending in flexure we calculate the fracture toughness KIsc
from the following equation:
6
2
0
10
2
3
−








=
a
BW
SP
YK
max
Isc
,
(1)
where KIsc = fracture toughness [MPa·m0.5], Y = stress
intensity factor (Eq.2 or Eq.6, whichever is greater), Pmax =
maximum force (the breaking force) [N], So = outer span
(0.0145 m in this case), B = side to side dimension of the
test specimen perpendicular to the crack length (depth)
[m], W = top to bottom dimension of the tested specimen
217

parallel to crack length (depth) [m], a = crack depth [m],
c = crack half width [m].
Calculate the stress intensity shape factor coefficients
for both the deepest point of precrack periphery Yd and for
the point at the surface Ys which will give a maximum
error of 3 % for an ideal precrack and an estimated
maximum error of 5 % for a realistic precrack.
For the deepest point of precrack the stress intensity
factor is
Q
MH
Y
d
2
π
=
,
(2)
where:
65.1
)/(464.11)/( cacaQQ
+==
,
(3)
+−==
))/(09.013.1()/,/( caWacaMM
2
)/(
)/(2.0
89.0
54.0 Wa
ca







+
+−
424
)/()/1(14
)/(65.0
1
5.0 Waca
ca







−+
+
−+
,
(4)
)/)(/12.022.1(1)/,/(
22
WacaWacaHH
+−==
25.175.0
)/)()/(47.0)/(05.155.0( Wacaca
+−+
,
(5)
For the point at the surface the stress intensity factor is
Q
SMH
Y
s
1
π
=
,
(6)
where:
),/)(/11.034.0(1)/,/(
11
WacaWacaHH
+−==
(7)
caWaWacaSS /))/(35.01.1()/,/(
2
+==
.
(8)
Note 1: The stress intensity factor coefficients are
valid only for
1/ ≤ca
. They can be used for a/c ratios
slightly greater than 1 with slight loss of accuracy.
Note 2: The term in the brackets of Eq. (1) is the
flexural strength [MPa] of the beam with the surface
crack. It is often useful to compare this value with a range
of values of the flexural strength of test specimens without
a precrack, in which fracture occurs from the natural
fracture sources in the material.
Note 3: Precrack is Vickers indentation at the load
(10–200) N, dwell time 45 s.
MATERIALS AND EXPERIMENTAL PART
Ceramic samples were made from the wet ceramic
plastic mass for manufacturing the high voltage insulators
from alumina porcelain. Mineralogical composition of the
mass was 36 wt. % of kaolin, 30 wt. % of Al2O3, 12 wt. %
of clay and 22 wt. % of feldspar.
Green blank with diameter of 300 mm was prepared
from the wet mass, described above, by the vacuum
extruder in ceramic plant PPC Čab, Slovakia. Prismatic
samples were cut from the wet blank then dried in the
open air and then grinded and polished. The final
dimensions of samples were 5 mm × 5 mm × 20 mm. The
sets of 5 samples were fired at different temperatures 400,
500, 600, 700, 800, 900, 1000, 1100, 1200, 1250, 1300,
1400, 1500 °C. The heating and cooling rate was 5
°C/min without soaking at the highest temperature.
After firing, the four flat sides of samples were
polished with sandpaper rotating machine with an angular
frequency 600 rpm for 5 minutes. The sandpaper number
was 800. The final dimensions of the samples were close
to 3 mm × 4 mm × 20 mm. The structure of material is
highly porous with a net of open pores. The porosity
varies from 32 % for green ceramics to 37 % at 500 °C
and decreases to 5 % within interval 1000 - 1200 °C [11].
To make a precrack, an indentation was performed by
hardness tester Zwick-Indtentec 5030 SKV [12] in the
middle of the sample side under the load 10-200 N with
the dwell time of 45 s, see Fig 2. The diagonals of
indentation were parallel with sample sides. Indentations
of the samples were then colored with red dye penetrant
ARDROX F6R for 10-60 minutes and then cleaned with
towel to remove excessive penetrant.
Fig. 2. Picture of the indent at 1250 °C (left picture) and at 400
°C (right picture)
A fracture toughness test was performed on the
Instron dynamic testing system 8516 [12] under the
loading rate of 10 N/s according to standard [10].
y = 0.0024e
0.005 x
R
2
= 0.9265
0.0
0.4
0.8
1.2
1.6
2.0
0 200 400 600 800 1000 1200 1400 1600
firing temperature / °C
fracture toughness / MPa.m
0.5
Fig. 3. The dependence of fracture toughness on firing
temperature.
218

0
10
20
30
40
50
60
300 500 700 900 1100 1300
firing temperature / °C
Young m. / GPa, mech. str. / MPa
firing temperature / °C
Young m. / GPa, mech. str. / MPa
Fig. 4. The dependence of flexural strength (o) and Young
modulus (•) on firing temperature
RESULTS AND DISCUSSION
The dependence of the fracture toughness on firing
temperature is shown in Fig. 3. The standard deviation of
KIc varies from 2 to 18 %.
Similar graphs, as pictured in Fig. 3, were also
obtained for flexural strength and Young's modulus on
samples made of quartz porcelain mixture (50 wt. % of
kaolin, 25 wt. % of feldspar, 25 wt. % of quartz) measured
at the same conditions [3, 13], see Fig. 4. Although the
material composition of quartz samples is slightly
different, the processes and their consequences during
heating are the same in the alumina and quartz samples
[1, 2]. Therefore, we can use the results presented in Fig.
4 for qualitative explanation of the fracture toughness
results pictured in Fig. 3. The fracture toughness, Young's
modulus and flexural strength have similar temperature
dependences, as follows from Fig. 3 and Fig 4, so they
correlate together.
For the samples fired at 20-500 °C no changes in
structure occurred at these low temperatures [1, 2],
therefore the mechanical strength and Young's modulus,
Fig. 4, are nearly constant in this range. The same is valid
for fracture toughness, as can be seen in Fig. 3.
For samples fired at 500-700 °C, dehydroxylation,
which weakens the kaolinite crystals, takes place.
Metakaolinite created during dehydroxylation is very
porous material with internal vacancies [14] and lowers
the mechanical properties of the samples. In spite of that,
we do not register a significant decrease of the mechanical
properties within the temperature region of 500 – 700 °C,
see Fig. 3, Fig. 4. We can assume that the solid state
sintering prevails over the weakness of the metakaolinite
crystals, so the crystal interfaces remain stronger after
cooling up to room temperature.
For samples fired at 700 - 1250 °C, there is
exponential increase in the fracture toughness and also the
mechanical strength and Young's modulus increases
dramatically. The solid-state sintering continues up to
1150 °C where feldspar begins to melt and a liquid-phase
sintering starts. The samples contain glassy phase, their
density is higher, and this corresponds to higher values of
the flexural strength measured at room temperature.
In the samples fired above 1300 °C closed pores are
present [1, 2] which lowers the mechanical strength and
fracture toughness.
CONCLUSIONS
1. Fracture toughness from 20 to 500 °C is almost
constant and it varies between 0.1 and 0.2 MPa·m0.5.
2. Dehydroxylation (420 – 600 °C) does not influence the
value of fracture toughness.
3. At temperature interval where we assume sintering
(700 – 1250 °C), we observe exponential dependance of
the fracture toughness.
4. From comparison of the fracture toughness, Young’s
modulus and flexural strength follows a correlation and
proporcionality of mechanical properties which agrees
with theoretical assumptions.
REFERENCES
1. Norton, F. H. Fine ceramics – Technology and
Application. McGraw-Hill Book Co., New York, 1970.
2. Hankýř, V., Kutzendorfer, J. Technologie keramiky.
Silis Praha a Vega Hradec Králové, Praha, 2000.
3. Štubňa, I., Trník, A., Chmelík, F., Vozár, L.
Mechanical Properties of Kaolin-base Ceramics During
Firing Materiali in Tehnologie. InTech Open Access
Publisher, Rijeka, 2011: 229-244.
4. Štubňa, I., Kozík, T., Hanic, F. Young’s Modulus
and Mechanical Strength of Porcelain at the Firing Cooling
Stage Ceramics International 18 1992: pp. 353-354.
5. Štubňa, I., Trník, A., Šín, P., Sokolář, R.
Relationship between Mechanical Strength and Young’s
Modulus in Traditional Ceramics Materiali in Tehnologie
45 (4) 2011: pp. 375-378.
6. Okabe, N. Application Technique and Evaluation for
Functional and Structural Design of Ceramics J. Ceram.
Soc. Japan 109 (12) 2001: pp. S135-S143.
7. Guo, J. K. Stress Design of the Ceramic Grain
Boundary Materials Chemistry and Physics 43 (1996) pp.
99-102.
8. Hertzberg, R. W. Deformation and Fracture
Mechanics of Engineering Materials (4th Edition). Wiley,
1995-12.
9. Dos Santos, S. F., De Anchieta, R. J. Correlation
between Fracture Toughness, Work of Fracture, and Fractal
Dimensions of Alumina-Mullite-Zirconia Composites
Materials Research 6 (2003) pp. 219-226.
10. ASTM C1421-10: Standard Test Methods for
Determination of Fracture Toughness of Advanced
Ceramics at Ambient Temperature.
11. Štubňa, I., Kozík, T. Permeability of the
Electroceramics to gas and Its Dependence on the Firing
Temperature Ceramics International 23 (1997) pp. 247-
249.
12. http://www.ttu.ee/department-of-materials-
engineering/rd/ equipment
13. Štubňa, I., Šín, P., Trník, A., Vozár, L. Relationship
between Mechanical Strength and Young’s Modulus in
Traditional Ceramics Material-Acoustics-Place STU
Zvolen, 2010, pp. 121-126.
14. Freund, F. Kaolinite-metakaolinite, a Model of a
Solid with Extremely High Lattice Defect Concentration
Berichte Deutsche Keramische Geselschaft 44 (1967) pp.
5-13.
Presented at the 20th International Conference "Materials
Engineering’2011" (Kaunas, Lithuania, October 27-28, 2011)
219