Materials Science and Engineering A 373 (2004) 180–186
Spark plasma sintering (SPS) consolidated ceramic composites from plasma-sprayed metastable Al2 TiO5 powder and nano-Al2 O3, TiO2 , and MgO powders Ren-Guan Duan a , Guo-Dong Zhan a , Joshua D. Kuntz a , Bernard H. Kear b , Amiya K. Mukherjee a,∗ a
b
Department of Chemical Engineering and Materials Science, University of California, Davis, CA 95616, USA Department of Ceramics and Materials Engineering, Rutgers-The State University of New Jersey, Piscataway, NJ 08854, USA Received 24 November 2003; received in revised form 10 January 2004
Abstract Spark plasma sintering (SPS) was applied to consolidate the nano-Al2 O3 , TiO2 , MgO powders and plasma-sprayed metastable Al2 TiO5 powder. The fully dense ceramics had been prepared successfully. It was found that the starting nano-Al2 O3 reacted with nano-TiO2 powder to form the Al2 TiO5 phase after SPS consolidation at 1150 ◦ C under a pressure of 63 MPa. The starting plasma-sprayed Al2 TiO5 was stable above 1150 ◦ C after SPS under 63 MPa, though in oxidizing atmospheres under the ambient air pressure, the stable temperature range of Al2 TiO5 was above 1280 ◦ C. There was a fast decomposition rate for plasma-sprayed Al2 TiO5 at 1000 ◦ C during SPS, whereas in oxidizing atmospheres under the ambient air pressure at 1000 ◦ C its decomposition is very slow. The microstructure of powder, applied pressure, and/or SPS characteristics, etc. obviously had an effect on the stable temperature and/or decomposition rate of Al2 TiO5 . The addition of MgO advanced the formation of Mg0.3 Al1.4 Ti1.3 O5 along with MgAl2 O4 in the ceramics prepared both from plasma-sprayed powder and from nano-powder. In comparison with the composition of the ceramics without MgO additive, ␣-Al2 O3 , TiO2 , and Al2 TiO5 , the ceramics with MgO additive consisted of only two phases, Mg0.3 Al1.4 Ti1.3 O5 and MgAl2 O4 spinel. The small equiaxed MgAl2 O4 spinel grains were distributed in a matrix of Mg0.3 Al1.4 Ti1.3 O5 with a large grain size. © 2004 Elsevier B.V. All rights reserved. Keywords: Spark plasma sintering (SPS); Nano-powder; Plasma-sprayed powder; Metastable; Decomposition
1. Introduction Alumina-based ceramics are utilized in many areas of modern industry because of their unique mechanical, electrical, and optical properties. It is anticipated that alumina-based nano-sized ceramic composites will demonstrate novel and favorable properties in comparison with their micro-sized crystalline counterparts [1,2]. Nanocrystalline materials have been drawing attention due to the expectations of high mechanical and functional characteristics. For example, the low temperature superplasticity in nanocrystalline ceramic materials is anticipated [3].
∗ Corresponding author. Tel.: +1-530-752-1776; fax: +1-530-752-9554. E-mail address:
[email protected] (A.K. Mukherjee).
0921-5093/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.msea.2004.01.006
In the preparation of nanocrystalline ceramics, when the densities are larger than 90% of theoretical, grain coarsening becomes particularly severe [4,5]. This means that it is difficult to get a fully dense ceramic while maintaining a nanocrystalline grain size. Therefore, it is necessary to explore new consolidation techniques that can accelerate sintering without increasing grain growth. The ideal choice for preparing nanoceramics is an optimization of a higher sintering pressure, a shorter sintering duration, and a use of the suitable additives. Spark plasma sintering (SPS) is a novel method that enables shorter sintering durations and relatively high sintering pressures, to consolidate ceramics. Nano-scale powders are used to conventionally prepare nanocrystalline ceramics. Another kind of powder, that is metastable and micro-sized, is also used to prepare nanocrystalline ceramics through transformation assisted consolidation [6]. This technique includes first a
R.-G. Duan et al. / Materials Science and Engineering A 373 (2004) 180–186
formation of metastable micro-scale ceramic powder by plasma-spraying and then high-pressure sintering (HPS) of the metastable powder to form nanocrystalline ceramics. This method possesses an advantage in that a nano-sized starting powder is not needed to prepare nanocrystalline bulk ceramics, and has been successfully applied to fabricate nano-Al2 O3 and nano-TiO2 single phase materials with a density higher than 99% of theoretical and with a grain size as small as 18 nm while consolidated above a pressure 5 GPa [6]. Plasma-sprayed aluminum titanate (Al2 TiO5 , tialite) is such a powder, which is metastable and micro-sized. It is being used in this investigation. In another regard, sintered Al2 TiO5 is a ceramic that has many potential applications such as components of internal combustion engines, thermal barriers, etc. due to its low thermal expansion coefficient (0.2 × 10−6 to 1 × 10−6 K−1 ), low thermal conductivity (0.9–1.5 W m−1 K−1 ), and excellent thermal shock resistance (about 500 W m−1 ) [7]. It is also a suitable additive as a second phase particle that can improve the thermal properties of ceramic matrix composites [8]. In this investigation, SPS was used to consolidate both the plasma-sprayed metastable Al2 TiO5 powder and a mixture of nano-Al2 O3 and TiO2 powders with or without nano-MgO powder additive. The decomposition rate and temperature stability of the Al2 TiO5 phase was also explored due to its importance in science and technology, and the influences of nano-MgO powder additive were analyzed.
181
and mixed by high energy ball-milling (HEBM, Spex 8000 Mixer/Mill, Spex Certiprep Industries Inc., Metuchen, NJ) for 2 h in alcohol with a tungsten carbide ball in a tungsten carbide vessel. Four powder mixtures were named Nano-AT, Plasma-AT, Nano-ATM, and Plasma-ATM, as listed in Table 1. Here AT refers to Al2 O3 /TiO2 powder mixtures and ATM refers to Al2 O3 /TiO2 /MgO powder mixtures. Spark plasma sintering (SPS) is a novel technique for consolidation of ceramics. The important steps during SPS process include the application of uniaxial pressure, the application of pulsed voltage to activate plasma, and the resistance heating of graphite die and powders. It is theorized that the pulsing electric field may produce plasma activation of powder surfaces, and along with the applied pressure and rapid heating is likely to consolidate the powders to close to the theoretical density. In this investigation, the apparatus, Dr. Sinter SPS-1050 from the Sumitomo Corporation of Japan, was used to sinter the sample. The powder was loaded into a graphite die with an internal diameter of 19 mm and then was heated by an electric current passing through the assembly in a vacuum. The sintering temperatures were selected at 1000, 1150, and 1250 ◦ C, with a holding time of 3 min. The heating rate was 300 ◦ C/min. During sintering, the applied uniaxial pressure on the sample is 63 MPa. The microstructural observations were carried out using a FEI XL30-SFEG high-resolution scanning electron microscope (SEM). The phase identification was conducted by X-ray diffraction (XRD), a Scintag XDS-2000 diffractometer, using Cu K␣ radiation.
2. Experimental procedures The metastable Al2 TiO5 powder was prepared by the following method: on the basis of the composition of Al2 TiO5 (60 wt.% Al2 O3 and 40 wt.% TiO2 ), commercially available Al2 O3 (grain size about 45 m) and TiO2 (about 5 m) powders were weighed and mixed uniformly. The powder mixture was melted in a plasma torch and then rapidly solidified by quenching in water [9]. The resultant powder possessed a micro-sized particle distribution. The above plasma-sprayed metastable Al2 TiO5 powder, nano-Al2 O3 powder (␥-Al2 O3 , average grain size ∼15 nm, from Nanotechnologies, Austin, TX), nano-TiO2 powder (average particle size ∼32 nm, from Nanophase Technologies Corporation, Burr Ridge, IL), and nano-MgO powder (average particle size ∼35 nm, NanomyteTM , from Nanopowder Enterprises Inc., Piscataway, NJ) were weighed on the basis of the compositions listed in Table 1, Table 1 The compositions of the starting powder mixtures (in wt.%)
Nano-AT Plasma-AT Nano-ATM Plasma-ATM
Nano-Al2 O3
Nano-TiO2
Nano-MgO
Al2 TiO5
60 – 54 –
40 – 36 –
– – 10 10
– 100 – 90
3. Results After the plasma-sprayed metastable Al2 TiO5 powder was consolidated by SPS at 1000 ◦ C (Fig. 1b), a great amount of powder is decomposed to ␣-Al2 O3 and TiO2 in comparison with the composition of the original Al2 TiO5 powder (Fig. 1a). The decomposition rate of the powder Al2 TiO5 is fast. At 1000 ◦ C during SPS under 63 MPa the plasma-sprayed Al2 TiO5 is not stable. When the Al2 TiO5 powder is consolidated under the same pressure at 1150 ◦ C (Fig. 1c) and 1250 ◦ C (Fig. 1d) for the same duration, it can be noted that the Al2 TiO5 is stable or its decomposition rate is very slow. When the nano-Al2 O3 and TiO2 powders are consolidated by SPS at 1150 ◦ C, a small amount of Al2 TiO5 phase was formed (Fig. 1e). This reveals that the plasma-sprayed Al2 TiO5 is stable above 1150 ◦ C during SPS under 63 MPa. The ceramic prepared by SPS from the plasma-sprayed metastable Al2 TiO5 powder at 1000 ◦ C is not fully dense (Fig. 2a). The size and shape of the particles in this ceramic are close to those in the original Al2 TiO5 powder. When this powder is consolidated at 1150 ◦ C, the ceramic is dense (Fig. 2b) and possesses a grain size distribution from 100 nm to 1 m (Fig. 3a). Most grains are in the range of 200–600 nm (Fig. 3a). When the powder is sintered at
182
R.-G. Duan et al. / Materials Science and Engineering A 373 (2004) 180–186
When the nano-Al2 O3 and TiO2 powders are consolidated at 1150 ◦ C, the resultant ceramic is fully dense (Fig. 2d). Most grains are located in the range of 300–600 nm (Fig. 3c). No grain is larger than 1.4 m. Though the chemical compositions are the same, 60 wt.% Al2 O3 and 40 wt.% TiO2 , for the Plasma-AT and Nano-AT samples (Table 1), the microstructures of the ceramics are very different (Fig. 2b and d). When nano-MgO powder is doped into the starting powder mixtures, either plasma-sprayed or nano-sized, the ceramics consolidated at 1150 ◦ C consist of only two phases, Mg0.3 Al1.4 Ti1.3 O5 (one composition of the Al2 TiO5 –MgTi2 O5 solid solution) and MgAl2 O4 spinel (Fig. 1f and g). In comparison with the compositions of the ceramics prepared without MgO additive (Fig. 1c and e), it is displayed that nano-MgO additive leads to the formation of Mg0.3 Al1.4 Ti1.3 O5 along with the formation of MgAl2 O4 . SEM observations show that the microstructures of these two kinds of ceramics are most similar (Fig. 2e and f), characterized by the extended grain size distributions (Fig. 3d and e). Most grains are located in the 300–900 nm range in the ceramic prepared from a mixture of plasma-sprayed powder and nano-powder (Fig. 3d), and most grains are distributed in the range from 200 to 700 nm in the ceramic prepared from nano-powders (Fig. 3e). However, some very large grains are also formed in these two ceramics, up to 2.3 m size in the ceramic prepared from plasma-sprayed powder Figs 2e and 3d) and up to 2.6 m size in the ceramic prepared from nano-powder (Figs. 2f and 3e). Combining XRD analysis (Fig. 1f and g) with SEM observation (Fig. 2e and f), it can be realized that the small equiaxed MgAl2 O4 spinel grains are distributed in the Mg0.3 Al1.4 Ti1.3 O5 matrix. In this investigation, the addition of MgO does not lead to a fine grain size.
4. Discussions Fig. 1. X-ray diffraction patterns of the original Al2 TiO5 powder and the ceramics prepared by spark plasma sintering (SPS) under 63 MPa pressure for 3 min: (a) original Al2 TiO5 powder; (b) at 1000 ◦ C from Al2 TiO5 powder; (c) at 1150 ◦ C from Al2 TiO5 powder; (d) at 1250 ◦ C from Al2 TiO5 powder; (e) at 1150 ◦ C from nano-Al2 O3 and TiO2 powders; (f) at 1150 ◦ C from plasma-sprayed Al2 TiO5 and nano-MgO powder; (g) at 1150 ◦ C from nano-Al2 O3 , TiO2 , and MgO powders.
1250 ◦ C, some grains exhibit tremendous growth (Fig. 2c), while other grains have very low growth or stay at the same sizes as those sintered at 1150 ◦ C (Fig. 2b and c). The grain size has an extended distribution, from 200 nm to 2.8 m (Fig. 3b). The smallest grains are contained within the large grains (Fig. 2c). Along with the XRD analysis (Fig. 1c and d), it can be suggested that the large grains are Al2 TiO5 , and the small grains are Al2 O3 and TiO2 . The Al2 TiO5 powders consolidated at different temperatures possess very different microstructures (Fig. 2b and c) and grain size distributions (Fig. 3a and b).
4.1. The stability of Al2 TiO5 during SPS Al2 TiO5 phase is metastable and has a tendency to decompose into the parent oxides below 1280 ◦ C in oxidizing atmospheres under the usual air pressure according to the following reaction [10]: Al2 TiO5 → ␣-Al2 O3 + TiO2 -rutile
(1)
the maximum decomposition rate of Al2 TiO5 is in the temperature range of 1100–1150 ◦ C, and above 1280 ◦ C no decomposition takes place [7]. Below 800–900 ◦ C the decomposition rate of Al2 TiO5 is usually very low, and decomposition cannot be detected even after long anneals [7]. So it can be considered stable from the kinetic view point. In this investigation, during SPS consolidation, the plasma-sprayed Al2 TiO5 has a fast decomposition rate at 1000 ◦ C (Fig. 1b) and is stable above 1150 ◦ C (Fig. 1c–e).
R.-G. Duan et al. / Materials Science and Engineering A 373 (2004) 180–186
183
Fig. 2. Scanning electron micrographs of the ceramic fracture surfaces prepared by spark plasma sintering (SPS) under 63 MPa pressure for 3 min: (a) from plasma-sprayed Al2 TiO5 powder at 1000 ◦ C; (b) from plasma-sprayed Al2 TiO5 powder at 1150 ◦ C; (c) from plasma-sprayed Al2 TiO5 powder at 1250 ◦ C; (d) from nano-Al2 O3 and TiO2 powders at 1150 ◦ C; (e) from plasma-sprayed Al2 TiO5 and nano-MgO powders at 1150 ◦ C; (f) from nano-Al2 O3 , TiO2 , and MgO powders at 1150 ◦ C.
The decomposition rate and stable temperature of Al2 TiO5 are affected by many factors, such as microstructure of the powder, applied pressure, atmosphere, etc. [7]. The Al2 TiO5 powder used in this investigation was made by plasma-spray: the powder mixture was melted by plasma and then rapidly solidified by quenching into water [9]. The microstructure of the resultant Al2 TiO5
powder can vary from those prepared by other methods, such as the coprecipitation method [11], reacting sintering [12,13], etc. This could be a factor affecting the decomposition rate and stable temperature of Al2 TiO5 powder. The decomposition of Al2 TiO5 powder is very sensitive to the composition of the atmosphere [7]. In an atmosphere
184
R.-G. Duan et al. / Materials Science and Engineering A 373 (2004) 180–186 35
35 30 25
25
Frequency (%)
Frequency (%)
30
20 15
10
5
5 0 100
200
300
(a)
400
500
600
700
800
900
1000
800 1000 1200 1400 1600 1800 2000 2200 2400 2600 2800
Grain Size (nm)
25
25
20
20
15
10
15
10
5
5
0
200 400 600
(b)
Grain Size (nm)
Frequency (%)
Frequency (%)
15
10
0
(c)
20
0 300
400
500
600
700
800
900
1000 1100 1200 1300 1400
200
400
600
800
1000 1200
(d)
Grain Size (nm)
1400 1600
1800 2000
2200 2400
Grain Size (nm)
25
Frequency (%)
20
15
10
5
0
200 400
600 800 1000 1200 1400 1600 1800 2000 2200 2400 2600 2800
(e)
Grain Size (nm)
Fig. 3. Grain size frequency distributions in the ceramics consolidated by spark plasma-sprayed (SPS) under 63 MPa pressure for 3 min, calculated according to the scanning electron micrographs: (a) from plasma-sprayed Al2 TiO5 at 1150 ◦ C; (b) from plasma-sprayed Al2 TiO5 at 1250 ◦ C; (c) from nano-Al2 O3 and TiO2 at 1150 ◦ C; (d) from Al2 TiO5 and MgO powders at 1150 ◦ C; (e) from nano-Al2 O3 , TiO2 , and MgO at 1150 ◦ C.
where the oxygen partial pressure is decreased, Ti4+ is progressively reduced to Ti3+ . According to Sperisen and Mocellin [14], in a reducing atmosphere, the decomposition of Al2 TiO5 powder should lead to the formation of Al2 Ti4+ O5 –Ti3+ 2 Ti4+ O5 (tialite–anosovite) solid solution. This implies that a reducing atmosphere can increase the decomposition rate of Al2 TiO5 . The densification of powder in this investigation was performed in vacuum. This can play a role in increasing the decomposition rate at 1000 ◦ C. On the other hand, in this investigation, the stable temperature
of Al2 TiO5 during SPS decreases from 1280 to 1150 ◦ C. It seems that the vacuum does not have an effect on stable temperature. The pressure dependence of the Gibbs free energy in a closed system is given as dG = −S dT + V dp
(2)
In the case that the temperature is a constant, the following equation can be obtained through the integration of Eq. (2)
R.-G. Duan et al. / Materials Science and Engineering A 373 (2004) 180–186
from pressure p0 to p: p G(p) = G(p0 ) + V dp p0
(3)
In the case that the pressure range is not too large, the volume of solid in Eq. (3) can be considered as a constant, so the following equation can be obtained: G(p) = G(p0 ) + V(p − p0 )
(4)
In this investigation, the applied pressure is not too large, so Eq. (4) can be applied in the analysis of decomposition process of Al2 TiO5 (Eq. (1)):
GRecT (p) = GRecT (p0 ) + V(p − p0 )
(5)
where GRecT (p) = GAl2 O3 (p) + GTiO2 (p) −
GAl2 TiO5 (p) and GRecT (p0 ) = GAl2 O3 (p0 ) +
GTiO2 (p0 ) − GAl2 TiO5 (p0 ) are the free energies of decomposition of Al2 TiO5 at temperature T under pressures p and p0 , respectively, and V = VAl2 O3 + VTiO2 − VAl2 TiO5 is the molar volume change during the decomposition at temperature T. When GRecT (p) < 0, thermodynamics indicate that the decomposition of Al2 TiO5 can progress. In this investigation, p0 represents the usual air pressure (1 atm). At room temperature, the densities of ␣-Al2 O3 , TiO2 , and Al2 TiO5 are 3.99, 4.25, and 3.70 g/cm3 , respectively, so the molar volumes are 25.56, 18.82, and 49.20 cm3 /mol, respectively. The V during the decomposition of Al2 TiO5 is negative (−4.82 cm3 /mol) at room temperature. The thermal expansion of Al2 TiO5 is very small [7] in comparison with those of ␣-Al2 O3 and TiO2 . With the increase of temperature, V changes gradually from the negative to a less negative or positive value. On the basis of Eq. (5), the pressure p can change the GRecT (p) from the negative to a less negative or the positive value. In this investigation, at 1150 ◦ C the nano-Al2 O3 reacted with TiO2 to form Al2 TiO5 (Fig. 1e). It means that at this temperature the decomposition free energy, GRecT (p), for Al2 TiO5 is greater than or equal to 0. The pressure should be one reason for decreasing the stable temperature of Al2 TiO5 from 1280 ◦ C down to at least 1150 ◦ C. The pressure should also have an effect on the decomposition rate of Al2 TiO5 . The decomposition of Al2 TiO5 is a nucleation-growth process [7]. A stress develops during the formation of new inclusion (precipitate) due to the difference of theoretical densities between inclusion and matrix, which can be increased or decreased by the applied pressure. In the present study, the decomposition of Al2 TiO5 leads to a volume contraction due to the difference in theoretical density between Al2 TiO5 and ␣-Al2 O3 as well as TiO2 , so the tensile stresses, developed during the nucleation process, can be decreased by the applied pressure. Pressure reduces the thermodynamic energy barrier for nucleation and advances the Al2 TiO5 decomposition. Another factor affecting the decomposition of Al2 TiO5 is the kinetic energy barrier for nucleation that relates with the ionic diffusion. Under 10 MPa mechanical pressure, at least
185
between 1475 ◦ C and 1630 ◦ C for the pure Al2 O3 , the lattice l diffusion coefficient of DAl,10 MPa of aluminum ion can be shown below [15]: 578 000 l 2 −1 7 (cm s ) = 1.87 × 10 DAl,10 exp − (6) MPa RT And under 5 MPa mechanical pressure, at least between 1350 and 1550 ◦ C for the pure Al2 O3 , the aluminum ion l lattice diffusion coefficient DAl,5 MPa can be given as [16] 578 000 l 2 −1 5 DAl,5 (cm s ) = 1.36 × 10 exp − (7) MPa RT where R is the gas constant and T is the absolute temperature. So pressure increases the lattice diffusion coefficient of the aluminum ions in pure Al2 O3 . Many other studies on the ionic diffusion in the solid have also shown that the pressure has an influence on the diffusion. In the amorphous Co92 Zr8 at 425 ◦ C, pressure decreases the self-diffusion coefficient of the zirconium ions below 800 MPa [17]. Pressure increases the diffusion coefficient of As in Ge matrix at 575 ◦ C [18], increases the diffusion coefficient of Ge passing through the interface between amorphous and crystalline Ge from 300 to 365 ◦ C [19], and increases the diffusion coefficient of Si passing through the interface between amorphous and crystalline silicon from 520 to 550 ◦ C [19]. In this investigation, the diffusion of ions passing through the inclusion/matrix interface during the nucleation of Al2 TiO5 possibly follows a diffusion mechanism where the activation volume is negative. During SPS, the applied pressure may increase the diffusion coefficient and decrease the kinetic energy barrier for nucleation, thus, increases Al2 TiO5 decomposition. SPS is a novel, pressure-assisted, fast sintering method that takes advantage of a high dc pulsed current (up to 5000 A) at a relatively high frequency (∼300 kHz) to consolidate the powders. The dc pulsing can generate several effects: spark plasma, spark impact pressure, Joule heating and an electrical field diffusion effect. These SPS characteristics could be another reason that the decomposition rate and stable temperature of Al2 TiO5 in this investigation are altered. 4.2. Influence of MgO on the structure and composition of ceramics During the transformation of ␥-Al2 O3 to ␣-Al2 O3 in the sintering process, a low intrinsic nucleation density leads to a large spacing between nucleation events and a formation of micrometer scale ␣-Al2 O3 grains with dendritic protrusions surrounded by continuous pore channels [20]. The resultant vermicular microstructure hinders the densification of ceramics during sintering and requires a sintering temperature above 1600 ◦ C to reach high densities [21]. In this investigation, SEM observations (Fig. 2a and c) of ceramics with or without MgO additive show that no vermicular microstructure is formed. This is possibly due to the addition of second or third phase.
186
R.-G. Duan et al. / Materials Science and Engineering A 373 (2004) 180–186
Due to large growth of the Al2 O3 /TiO2 grains during sintering, the ceramic prepared from nano-Al2 O3 and TiO2 powders in this study is not a nanocrystalline composite (average grain size < 100 nm) (Fig. 2d). Using high-pressure sintering (HPS) under 1 GPa at 850 ◦ C [22], the addition of MgO decreases the growth of grains and results in a nanocrystalline structure. However, in this investigation, the addition of MgO makes some grains grow faster (Fig. 2e and f). This is possibly due to the formation of the Mg0.3 Al1.4 Ti1.3 O5 phase that has a faster grain growth rate. The Mg0.3 Al1.4 Ti1.3 O5 phase is in effect a Al2 TiO5 – MgTi2 O5 solid solution with composition like Mgx Al2(1−x) Ti(1+x) O5 , that likely decomposes according to the following reaction:
out MgO additive contain three phases, ␣-Al2 O3 , TiO2 rutile, and Al2 TiO5 , whereas, the ceramics prepared from plasma-sprayed powder and from nano-powder with a MgO additive consist of only two phases, Mg0.3 Al1.4 Ti1.3 O5 and MgAl2 O4 . The small equiaxed MgAl2 O4 grains were distributed in the large grain Mg0.3 Al1.4 Ti1.3 O5 matrix. During SPS the nano-MgO additive does not decrease the grain size in the resultant ceramics. Acknowledgements This work is sponsored by UC Davis’ Subcontract #1289 with Rutgers University on their ONR Grant #N00014-01-C-0370.
Mgx Al2(1−x) Ti(1+x) O5 → xMgAl2 O4 + (1 − 2x)Al2 O3 + (1 + x)TiO2 (0.0 ≤ x ≤ 0.5)
References (8)
However, in this study, no Al2 O3 and TiO2 were formed when nano-MgO powder was added to either nano-Al2 O3 and TiO2 powder or plasma-sprayed Al2 TiO5 powder (Fig. 1f and g). This suggests that the formed solid solution Mg0.3 Al1.4 Ti1.3 O5 is stable at 1150 ◦ C during SPS under 63 MPa. In comparison with pure Al2 TiO5 , in oxidizing atmospheres and under the usual air pressure the solid solution, Mg0.3 Al1.4 Ti1.3 O5 , is more thermally stable [23,24]. This means that in this investigation, the stable temperature of Mg0.3 Al1.4 Ti1.3 O5 should be lower than 1150 ◦ C.
5. Conclusions Fully dense ceramics have been successfully prepared by spark plasma sintering (SPS), both from plasma-sprayed powder and from nano-powder. During spark plasma sintering (SPS), above 1150 ◦ C under 63 MPa pressure, the Al2 TiO5 is stable. Whereas, in the air the stable temperature range for Al2 TiO5 is above 1280 ◦ C. There is a fast decomposition rate for Al2 TiO5 at 1000 ◦ C during SPS, though in the air at the same temperature the decomposition rate for Al2 TiO5 is very slow. Applied pressure, microstructure of powder, and/or SPS characteristics, etc. have obviously had an effect on the decomposition rate and stable temperature of Al2 TiO5 . The addition of nano-MgO powder leads to the formation of Mg0.3 Al1.4 Ti1.3 O5 (one kind of Al2 TiO5 –MgTi2 O5 solid solutions) along with the formation of MgAl2 O4 spinel. The ceramics prepared both from plasma-sprayed powder and from nano-powder with-
[1] G.D. Zhan, J.D. Kuntz, J. Wan, J. Garay, A.K. Mukherjee, Scripta Mater. 47 (2002) 737. [2] G.D. Zhan, J.D. Kuntz, J.L. Wan, A.K. Mukherjee, Nat. Mater. 2 (2003) 38. [3] R.S. Mishra, J.A. Schneider, J.F. Shackelford, A.K. Mukherjee, Nanostruct. Mater. 5 (1995) 525. [4] R.S. Averback, H.J. Hofler, H. Harn, J.C. Logas, Nanostruct. Mater. 1 (1992) 173. [5] M.J. Mayo, D.C. Hague, Nonostruct. Mater. 3 (1993) 43. [6] B.H. Kear, J. Colaizzi, W.E. Mayo, S.C. Liao, Scripta Mater. 44 (2001) 2065. [7] V. Buscaglia, P. Nanni, J. Am. Ceram. Soc. 81 (1998) 2645. [8] S. Ananthakumar, K.G.K. Warrier, J. Eur. Ceram. Soc. 21 (2001) 71. [9] B.H. Kear, Z. Kalman, R.K. Sadangi, G. Skandan, J. Colaizzi, W.E. Mayo, J. Therm. Spray Technol. 9 (2000) 483. [10] E. Kato, K. Daimon, J. Takahashi, J. Am. Ceram. Soc. 63 (1980) 355. [11] M. Nagano, S. Nagashima, H. Maeda, A. Kato, Ceram. Int. 25 (1999) 681. [12] R. Uribe, C. Baudin, L. Mazerolles, D. Michel, J. Mater. Sci. 36 (2001) 5105. [13] J.J. Melendez-Martinez, M. Jimenez-Melendo, A. DominguezRodriguez, G. Wotting, J. Eur. Ceram. Soc. 21 (2001) 63. [14] T. Sperisen, A. Mocellin, J. Mater. Sci. Lett. 10 (1991) 831. [15] M.P. Harmer, R.J. Brook, J. Mater. Sci. 15 (1980) 3017. [16] P.A. Lessing, R.S. Gordon, J. Mater. Sci. 12 (1977) 2291. [17] P. Klugkist, K. Ratzke, F. Faupel, Phys. Rev. Lett. 81 (1998) 614. [18] S. Mitha, M.J. Aziz, D. Schiferl, D.B. Poker, Appl. Phys. Lett. 69 (1996) 922. [19] G.Q. Lu, E. Nygren, M.J. Aziz, J. Appl. Phys. 70 (1991) 5323. [20] C.S. Nordahl, G.L. Messing, J. Eur. Ceram. Soc. 22 (2002) 415. [21] F.W. Dynys, J.W. Halloran, J. Am. Ceram. Soc. 65 (1982) 442. [22] R.G. Duan, G.D. Zhan, J.D. Kuntz, B.H. Kear, A.K. Mukherjee, Scripta Mater., submitted for publication. [23] M. Ishitsuka, T. Sato, T. Endo, M. Shimada, J. Am. Ceram. Soc. 70 (1987) 69. [24] G. Tilloca, J. Mater. Sci. 26 (1991) 2809.
