DOI :
10.2240/azojomo0291
Written by AZoMAug 19 2009
Tatsuya Ono, Koji Matsumaru, Isaias Juarez-Ramirez, Leticia M.
Torres-Martinez and Kozo Ishizaki
Copyright AD-TECH; licensee AZoM.com Pty Ltd.
This is an AZo
Open Access Rewards System (AZo-OARS) article distributed under the terms of the
AZo-OARS https://www.azom.com/oars.asp
which permits unrestricted use provided the original work is properly cited but
is limited to non-commercial distribution and reproduction.
AZojomo (ISSN 1833-122X) Volume 6 August
2009
Topics Covered
Abstract
Keywords
Introduction
Experimental
Results
Discussion
Conclusions
References
Contact
Details
Abstract
Machines for manufacturing large flat display glass have grown larger as the
size of display glasses has increased. Our research group has been developing
high Young's modulus, low thermal expansion porous materials consisting of a
positive thermal expansion material and a negative thermal expansion material
bonded by a glassy material (GM). SiC and LiAlSiO_{4} powders were selected
because of their positive and negative thermal expansion coefficients, respectively.
In order to obtain a desirable Young's modulus and a low thermal expansion coefficient,
porous material compositions were selected from the SiC - GM - LiAlSiO_{4}
system at 20 vol % porosity. The empirical values of Young's modulus are close
to the theoretical values obtained using the composition diagram for high GM
amounts. However the empirical values of Young's modulus are lower than the
theoretical values for low GM amounts. This work estimates a minimum amount
of GM from the thickness of GM which covered SiC and LiAlSiO_{4} powders
to obtain the theoretical Young's modulus. It is concluded that the GM thickness
should be more than 0.6µm to obtain the theoretical Young's modulus.
Keywords
LiAlSiO_{4}, SiC, Glassy material, Porous ceramics, Young's
modulus
Introduction
Machines for manufacturing large flat display glass have increased as the
size of display glasses has increased. Large size precision plates over 2 m x 2
m are necessary for the machines in order to fix the position of the glass. To
reduce positioning errors, a precision plate should have a low or zero thermal
expansion coefficient, a high Young's modulus and low specific weight. Table 1
shows the density, Young's modulus and thermal expansion coefficients of
granite, cast iron, alumina ceramics and metal matrix composites (MMC).
Table 1. Density, Young's modulus and thermal expansion
coefficient of raw materials for a precision plate
Property |
Cast
iron FC25[1] |
Granite
[1] |
Alumina
ceramics AC270[1] |
MMC[2] |
Density/g cm^{-3} |
7.8 |
3.0 |
3.4 |
3.0 |
Young's modulus/GPa |
108 |
29-88 |
230 |
265 |
TEC/10^{-6} K^{-1} |
11 |
4-8 |
4.5 |
6 |
Those materials are commonly used for the precision plates. Koga [1] and
Ishii [2] proposed the use of composite materials with high Young's modulus.
Although the Young's modulus of those materials is higher than that of granite,
their density and thermal expansion coefficients are similar to those of
granite.
Our research group has proposed a porous material bonded by a
glassy material (GM) for a large size precision plate. The properties of the
fabricated porous material can be controlled by the respective volume fractions
in the composites. Porous materials with zero thermal expansion coefficient are
synthesized from a positive thermal expansion material and a negative thermal
expansion material bonded by a GM [3 - 6]. The empirical values of thermal
expansion coefficient were similar to the theoretical values. However, in the
case of low GM content the empirical values of Young's modulus were lower than
the theoretical values [6]. This behaviour suggested that a minimum amount of GM
is required to obtain the theoretical Young's modulus. This is probably because
Young's modulus decreases with decreasing neck size of the GM bridges between
SiC and LiAlSiO_{4} powders, i. e. the thickness of GM which covered SiC
and LiAlSiO_{4} powders [6].
In this study, the reason why Young's modulus became low for low
GM contents was clarified for porous fabricated materials in the SiC - GM -
LiAlSiO_{4} system. Different sizes of SiC grains were used to estimate
the minimum amount of GM for obtaining the theoretical Young's modulus.
Experimental
Table 2 shows the properties of the raw powders. SiC and
LiAlSiO_{4} were selected as positive and negative thermal expansion
materials, respectively. Those powders were bonded by GM to synthesize porous
materials with high Young's modulus and low thermal expansion coefficient. SiC
powders with particle sizes of #90 (222µm), #120 (157µm) and #150(128µm), GM
powder of 10µm median diameter and LiAlSiO_{4} powder of 17µm median
diameter were used as raw materials. LiAlSiO_{4} was synthesized by
solid-state reaction [3]. The particle sizes of GM and LiAlSiO_{4} were
measured by the centrifugal method and those of SiC powders were measured using
a laser diffraction particle size analyzer.
Table 2. Density, Young's modulus and thermal expansion
coefficient of raw materials
Property |
SiC
[7] |
LiAlSiO_{4}
[8] |
GM
[3] |
Density/g cm^{-3} |
3.1 |
2.7 |
2.4 |
Young's modulus/GPa |
410 |
83 |
70 |
TEC/10^{-6} K^{-1} |
4.0 |
-6.2 |
4.6 |
Table 3 shows the volume fraction and sintering temperatures in the
experiments. Mixed samples (#90 + #150) were prepared by using 40 vol% of SiC
#90 and 60 vol% of #150. The volume fraction was fixed by the diagram for a high
Young's modulus and low thermal expansion coefficient material in the SiC - GM -
LiAlSiO_{4} system at 20 vol% porosity [6]. Mixed powders in 45g batches
were formed into rectangular bars by uniaxial pressing at 60MPa for 1 min and
cold isostatically pressing at 300MPa for 1 min. The green compacts were
sintered in a conventional sintering furnace at 850°C or 950°C at a heating rate
of 132Kh^{-1} and held at these temperatures for 1 h.
Table 3. The volume fraction and sintering
temperatures
Sample |
Amount
/ vol% |
Sint.
temp./°C |
SiC |
LiAlSiO_{4} |
GM |
#90 Vg15 |
65.0 |
20.0 |
15.0 |
950 |
#90 Vg 20 |
61.2 |
18.8 |
20.0 |
#90 Vg 25 |
57.4 |
17.6 |
25.0 |
#120 Vg 15 |
65.0 |
20.0 |
15.0 |
950 |
#120 Vg 20 |
61.2 |
18.8 |
20.0 |
#120 Vg 25 |
57.4 |
17.6 |
25.0 |
#120 Vg 30 |
53.5 |
16.5 |
30.0 |
850 |
#90 + #150 Vg 20 |
61.2 |
18.8 |
20.0 |
950 |
#90 + #150 Vg 25 |
57.4 |
17.6 |
25.0 |
#90 + #150 Vg 30 |
53.5 |
16.5 |
30.0 |
850 |
The porosity was measured by the Archimedes method. The samples were held in
vacuo for 30 min in ethanol to fill up the open pores with ethanol. Their
Young's modulus was measured by the resonant frequency method (JE-RT, Nihon
Techno-Plus Co., Ltd.); the sample size was 50 mm x 7 mm x 1.5 mm. The specific
surface area of the raw materials was measured by the gas absorption (BET)
method (BELSORP-max, BEL JAPAN, INC.). The microstructure of the samples was
analyzed by scanning electron microscopy (SEM, VE-7800, KEYENCE, Corp.).
Results
Fig. 1(a) shows the porosity as a function of the GM amount. The data points
represent the empirical porosity values, which were around 22 to 25 vol%. Fig.
1(b) shows the Young's modulus as a function of the GM amount. The lines
indicate the theoretical Young's modulus for different porosity values. The
theoretical values were calculated by linear summation for each composition. The
empirical Young's modulus of each sample with different SiC particle size shows
a maximum value under the experimental conditions used. The theoretical Young's
modulus decreases as the GM amount increases, while the empirical Young's
modulus increases as the GM amount increases, reaching a maximum value, then
decreases. The maximum empirical Young's modulus for SiC #90, #120 and #90 +
#150 are obtained at Vg20, Vg25 and Vg25, respectively. As the SiC grain size
decreases, a larger amount of GM is required to get the highest empirical
Young's modulus.
Figure 1. (a) Porosity and (b) Young's modulus as a function
of the GM amount.
Discussion
Fig. 2 shows the skeleton structure and SEM micrograph of sintered
porous material. The neck size increases as the GM amount increases. Green et
al. concluded that the Young's modulus is related to the neck size [9]. The
present work revealed that the Young's modulus is lower than the theoretical
values for low GM amounts because GM is consumed to cover SiC and
LiAlSiO_{4}, not used to form reasonably thick necks. This behaviour
suggests that a minimum GM amount is required to obtain the theoretical Young's
modulus. The minimum GM amount depends on the specific surface area of the SiC
and LiAlSiO_{4} particles, i. e. particle size.
Figure 2. The skeleton structure and SEM micrograph of
sintered porous material.
Fig. 3 shows the Young's modulus as a function of the GM thickness around the
SiC and LiAlSiO_{4} particles.
Figure 3. Young's modulus as a function of the GM thickness.
The GM thickness is calculated by Equation (1), where Tg is the GM thickness,
mg, ms and ml are the mass of GM, SiC and LiAlSiO_{4}, respectively, Dg
is the density of GM, Ass and Asl are the specific surface area of SiC and
LiAlSiO_{4}, respectively.
Table 4 shows the specific surface area of SiC#90, #120 and #150. The
specific surface area increases as SiC grain size decrease. The empirical
Young's Modulus increases and then decreases with increase of estimated GM
thickness. The Young's modulus for SiC grain sizes has a maximum value at a GM
thickness of around 0.6µm. Therefore, a GM thickness above 0.6 µm is required to
obtain the theoretical Young's modulus.
Table 4. The specific surface area of SiC and LiAlSiO_{4}
Sample |
Specific
Surface Area, A_{s} / m^{2}g^{-1} |
SiC #90 |
6.31x 10^{-2} |
SiC #120 |
9.70 x 10^{-2} |
SiC #150 |
1.52 x 10^{-1} |
LiAlSiO_{4} |
4.62 x 10^{-1} |
Conclusions
SiC and LiAlSiO_{4} were selected as positive and negative thermal
expansion materials, respectively. Those powders were bonded by Glassy Material
to synthesize porous materials with high Young's modulus and low thermal
expansion coefficient. Compositions of powders for porous materials were
determined by using the composition diagram of SiC - GM - LiAlSiO_{4}
system at 20 vol% of porosity to obtain a desirable values of Young's modulus
and thermal expansion coefficient. The empirical values of Young's modulus are
close to the theoretical values for high GM amounts. However the empirical
Young's modulus is lower than the theoretical values for low GM amounts.
Therefore, a minimum GM amount is required to obtain the theoretical value of
Young's modulus. If the GM amount is lower than 0.6µm thickness because of the
coverage of SiC and LiAlSiO_{4} particles, the necks between those
particles do not grow enough to obtain the theoretical Young's modulus.
References
- N. Koga, "Large Engineering Ceramics for LCD Production Systems",Ceramics
Japan, 43 (2008) 468-476 [in Japanese].
- M. Ishii, "MMC for LCD Production Systems", Ceramics Japan,
43 (2008) 568-569 [in Japanese].
- I. J. Ramirez, K. Matsumaru and K. Ishizaki, "Development
of a Near Zero Thermal Expansion Porous Material", J. Ceram. Soc. Jap., 114
(2006) 1111-1114.
- I. J. Ramirez, K. Matsumaru, K. Ishizaki and L. M. Torres-Martinez,
"Comparison of Porous Ceramic Materials with Low Thermal Expansion Coefficient
Prepared with SiC and Black-Al2O3", Materials Science Forum, 569 (2008) 321-324.
- I. J. Ramirez, K. Matsumaru, K. Ishizaki and L.M. Torres-Martinez,
"Particle size effect of LiAlSiO_{4} on the thermal expansion of SiC
porous materials", Journal of Ceramic Processing Research, 9 [5] (2008) 509-511.
- T. Ono, K. Matsumaru, I. J. Ramirez, L. M. Torres-Martínez
and K. Ishizaki, "Development of Porous Material with High Young's Modulus
and Low Thermal Expansion Coefficient in SiC-Vitrified Bonding Material-LiAlSiO_{4}
System", Mater. Sci. Forum, in print.
- S. Suyama, T. Kameda and Y. Itoh, "Development of High-strength
Reaction-sintered Silicon Carbide", Diamond and Related Materials, 12 (2003)
1201-1204.
- R. Roy, D.K. Agrawal, and H.A. McKinstry, "Very Low Thermal
Expansion Coefficient Materials", Ann. Rev. Mater. Sci., 19 (1989) 59-81.
- D. J. Green, C. Nader and R. Brezny, "The
Elastic Behavior of Partially-sintered Alumina", Sintering of Advanced Ceramics,
(1990) 345-356.
Contact Details
Tatsuya Ono, Koji Matsumaru and Kozo Ishizaki
Nagaoka University of Technology
Nagaoka, Niigata 940-2188
Japan
E-mail: [email protected]
Isaias Juarez-Ramirez and Leticia M. Torres-Martinez
Universidad Autonoma de Nuevo Leon
Av. Universidad s/n
San Nicolas de los Garza, NL, C.P. 66451
Mexico
This paper was also published in print form
in "Advances in Technology of Materials and Materials Processing", 11[1] (2009)
25-30.