Effect of heating rates on the synthesis of Al2O3-SiC composites by the self-propagating high-temperature synthesis (SHS) technique

Article abstract of DOI:10.1111/j.1151-2916.2001.tb00768.x

Various aspects of in situ formation of Al₂O₃-SiC composites by the self-propagating high-temperature synthesis (SHS) technique have been investigated using thermal analyses (TG/DTA) of a powder mixture (4Al, 3SiO₂, 3C) and pellets in an argon atmosphere at different heating rates. Both the reaction initiation and peak temperatures are found to increase with the heating rates. At lower heating rates, the powder samples do not reveal any exothermic peak possibly because of poor reactivity and sluggish exothermic reaction. The appearance of exothermic peaks in the DTA plots after melting of aluminum indicates reduction of silica by liquid aluminum. Conversion of aluminum is found to decrease marginally with an increase in heating rates. The apparent activation energy of the process compares well with the interdiffusion activation energy of silicon and oxygen, indicating that oxygen diffusion in Si formed at the reaction front may be the rate-controlling factor for this SHS process. From SEM studies it appears that the formation of SiC whiskers is through liquid-phase mass transfer.

Full text of DOI:10.1111/j.1151-2916.2001.tb00768.x

J. Am. Ceram. Soc., 84 [5] 915–20 (2001)
journal
Effect of Heating Rates on the Synthesis of Al O –SiC Composites by
2
3
the Self-Propagating High-Temperature Synthesis (SHS) Technique
Lokesh Chandra Pathak, Debajyoti Bandyopadhyay, Srinivasan Srikanth,
Swapan Kumar Das, and P. Ramachandrarao
National Metallurgical Laboratory, Jamshedpur 831 007, India
Various aspects of in situ formation of Al O –SiC composites
by the self-propagating high-temperature synthesis (SHS)
technique have been investigated using thermal analyses (TG/
The RH-SHS technique has shown the feasibility of fabrication
8,9
2
3
of Al O –SiC in situ composites, but several aspects, which
2 3
include heating rates, green density, composition, etc., have not yet
been investigated. In this paper, an attempt has been made to
investigate the influence of heating rates on the synthesis of
Al O –SiC composites.
DTA) of a powder mixture (4Al, 3SiO , 3C) and pellets in an
2
argon atmosphere at different heating rates. Both the reaction
initiation and peak temperatures are found to increase with
the heating rates. At lower heating rates, the powder samples
do not reveal any exothermic peak possibly because of poor
reactivity and sluggish exothermic reaction. The appearance of
exothermic peaks in the DTA plots after melting of aluminum
indicates reduction of silica by liquid aluminum. Conversion of
aluminum is found to decrease marginally with an increase in
heating rates. The apparent activation energy of the process
compares well with the interdiffusion activation energy of
silicon and oxygen, indicating that oxygen diffusion in Si
formed at the reaction front may be the rate-controlling factor
for this SHS process. From SEM studies it appears that the
formation of SiC whiskers is through liquid-phase mass
transfer.
2
3
II. Experimental Procedure
Stoichiometric amounts of silica (SiO ) (99% pure), aluminum
2
(
99% pure), and carbon (graphite, 98% pure) powders in 4:3:1
molar ratio were taken in a stainless steel container and milled for
6
2
1
h. Approximately 6 g of milled powder mixture was pressed to
.5 cm diameter pellets at a pressure of Ϸ100 MPa. Approximately
wt% of poly(vinyl alcohol) (PVA) was added as a binder during
pelletization. The green pellets were dried at 150°C for 3 h to
remove the moisture.
Simultaneous differential thermal analyses (DTA) and thermo-
gravimetric analyses (TGA) of the powder mixture and the pellets
were conducted in a Seiko TG/DTA apparatus (Model No. 320).
The average particle size of the powder sample was 30 ␮m and the
green densities of the pellets were found to be 60% of the
theoretical density. Before the thermal analysis experiments, the
I. Introduction
LUMINA-BASED ceramic-matrix composites have shown prom-
A
ise as advanced materials for high-temperature applications
sample chamber was flushed with pure argon gas (O ϳ 2 ppm)
2
because of their excellent refractoriness, low susceptibility to
oxidation, and good mechanical properties at elevated tempera-
and the investigations were conducted under continuous flow of
argon gas at a rate of 50 mL/min. The samples were heated from
the ambient temperature to 1200°C at several heating rates ranging
between 5 and 100°C/min. On attainment of the desired tempera-
tures the samples were cooled down to room temperature. The
experimental conditions used for this study are summarized in
1
–4
tures.
Besides the conventional powder metallurgical route,
synthesis of these composites by in situ fabrication methods is
gaining momentum due to certain inherent advantages. In situ
fabrication techniques eliminate the problems of using fine ceram-
ics, particularly whiskers, and the chances of segregation of the
constituents are relatively low compared with mechanical mixing
methods. There are also fewer processing steps in this route.
Several techniques for the synthesis of in situ composites have
Table I. The reaction initiation temperature (T ) was measured
i
from the DTA plots using the three-point method. The time
interval between T and the exothermic peak temperature, T , was
i
p
noted. In one of the experiments, the pellet was heated at a rate of
00°C/min from ambient temperature to 1450°C and soaked for 10
min in a graphite furnace. Then the pellet was furnace cooled and
characterized.
The samples were characterized by XRD, SEM, and energy-
dispersive X-ray spectroscopy (EDS). XRD analyses of the sam-
ples were conducted at different stages of processing. Fractured
surfaces of the samples were studied with a JEOL-840A SEM
system equipped with a Kevex EDS system.
5
–8
been reported in the literature.
4
A new combustion synthesis technique for the fabrication of
Al O –SiC in situ composite using a rapid-heating self-
2
3
w
propagating high-temperature synthesis (RH-SHS) has recently
8
,9
been reported by the authors. During synthesis of composites
through the SHS technique, heating rates may have a significant
influence on the reaction initiation temperature (T ), the mecha-
i
nism of the combustion reaction, as well as the quality of the
1
0–12
products.
The synthesis of Ti–Al intermetallic compounds by
the SHS route has shown that the mechanism of the reaction
changes from solid–liquid interaction to solid–solid interaction
1
2
III. Results and Discussion
with an increase in the heating rates. It has also been reported
that an increase in the heating rates improves the density of the
1
1
In the combustion synthesis technique, the self-sustainability of
the exothermic reaction and the propagation of a combustion wave
front mostly depend on the enthalpy change associated with the
SHS product.
1
3,14
reaction and the rate of energy dissipation from the system.
The adiabatic temperature, T , indicates that the maximum tem-
N. S. Jacobson—contributing editor
ad
perature attained by the products under adiabatic conditions is an
important SHS parameter. It has been reported that a value Tad
above 1800 K is required for the self-sustainability of the exother-
1
4,15
Manuscript No. 189888. Received September 8, 1998; approved January 18, 2000.
mic reaction.
It is also well known that most of the exothermic
9
15

9
16
Journal of the American Ceramic Society—Pathak et al.
Vol. 84, No. 5
Table I. Details of the Experimental Conditions
reaction. In the case of sample P (Fig. 1(c)), even though the
3
contact is poor, a higher rate of heat input appears to catalyze the
exothermic reactions at localized zones, resulting in exothermic
peaks in the thermograms.
(B) Pellets: A representative DTA plot for pellet sample C2
heated at a rate of 10°C/min is shown in Fig. 2. Similar to the
powder samples, two endothermic peaks are observed to initiate at
Sample mass
(mg)
Heating rate
(°C/min)
Final temperature
(°C)
Sample
P1
P2
P3
Form
15.941
14.578
16.892
25.982
47.893
48.451
38.734
32.091
37.955
34.351
5.0 g
Powder
Powder
Powder
Pellet
Pellet
Pellet
Pellet
Pellet
Pellet
Pellet
Pellet
10
40
1200
1200
1200
1200
1200
1200
1200
1200
1200
1200
1450
100
5
C1
C2
C3
C4
C5
C6
C7
5
72° and 660°C in the DTA thermograms. Besides these endo-
10
thermic peaks, exothermic peaks are also observed in the thermo-
grams for all of the heating rates. The appearance of prominent
exothermic peaks above 720°C in the pellets is due to better
contact between the reactant constituents. The shift of the exother-
mic peaks toward higher temperature sides for samples C , C , and
20
40
60
80
100
400
1
3
C8
C6 are shown in Fig. 3, which indicates that both the reaction
initiation temperature (T ) and the peak temperature (T ) increase
i
p
with an increase in the heating rates.
The exothermic peaks occurring just after the melting of
aluminum appear to be a superimposition of three exothermic
reactions. These overlapped peaks become more distinguishable at
higher heating rates. TG/DTA plots of pellet C7 with three
deconvoluted (using Gaussian fitting) DTA peaks are shown in
Fig. 4. The first exothermic peak is due to the aluminothermic
reduction of silica, i.e.,
reactions are initiated at a temperature (T ) above room tempera-
ture. Tad for the reaction
i
4
Al ϩ 3SiO ϩ 3C ϭ 2Al O ϩ 3SiC
(1)
2
2
3
was calculated to be 2375 K based on the thermodynamic data
1
6
from a recent compilation. The high value of the adiabatic
temperature of the reacting system under consideration suggests
the possibility of self-sustenance of the reaction propagation and
simultaneous sintering of the SHS products.
4Al ϩ 3SiO
ϭ 2Al
O
ϩ 3Si
(2)
2
2
3
and the second peak is due to the reaction of Si and C forming SiC
i.e.,
(1) Thermal Analysis
(
A) Powder System: DTA plots for the stoichiometric pow-
Si ϩ C ϭ SiC
(3)
der mixture of Al ϩ SiO ϩ C (in 4:3:1 molar ratio) heated at
2
various rates of 10, 40 and 100°C/min are shown in Fig. 1(a). A
small endothermic peak appears at 572°C followed by a major
endothermic peak at 660°C (Fig. 1(a)). The first endothermic peak
The third exothermic peak may be due to the oxidation of Al, Si,
or SiC by the traces of oxygen present in argon. The weight gain
of the sample as observed from the TG plot (Fig. 4) and the peak
of the differential thermogravimetry (DTG) plot (not shown in the
figure) strongly suggest an oxidation phenomenon.
is due to the transformation of SiO from ␣- to ␤-phase and the
2
second one represents the melting of aluminum. Except for these
two peaks, no other peak could be noticed in the thermogram of
The variations of T and T with heating rates for the alumino-
i
p
sample P (heating rate ϭ 10°C/min) in the temperature range of
1
thermic reduction of silica are shown in Fig. 5. The reaction
initiation temperature at a low heating rate, i.e., 5°C/min, is
observed to be 723°C, which matches well with the reported value
observation. However, for powder sample P heated at a rate of
3
1
00°C/min, two additional exothermic peaks at Ϸ890°C and at
Ϸ1100°C are observed. These exothermic peaks are likely to be
due to the formation of aluminum oxide and SiC, respectively (Fig.
7
of 715°–720°C in the literature. The variation of the second
exothermic peak temperature, Tp(SiC), for SiC formation as a
function of heating rate is also shown in Fig. 5. It can also be
observed that Tp(SiC) increases continuously with an increase of
heating rates and the lowest peak temperature for SiC formation is
1
(c)). The absence of an exothermic peak for sample P1 can
possibly be attributed to the poor contact between the reacting
components in the powdered samples. The segregation of liquid Al
during slow heating might have also hindered the combustion
noted to be 867°C for pellet sample C . The formation of SiC well
2
Fig. 1. DTA thermogram of the powder samples P (a), P (b), and P (c).
Fig. 2. Typical DTA plot of the compact C heated at 10°C/min. The two
1
2
3
2
Comparing the effect of heating rates on the various transformations during
the combustion synthesis of composites (the broken line implies the base
line, d).
endothermic peaks correspond to the ␣ 3 ␤ transition in quartz and the
exothermic peaks correspond to the formation of Al O and SiC, respec-
2
3
tively.

May 2001
Effect of Heating Rates on the Synthesis of Al O –SiC Composites
917
2
3
Fig. 4. Deconvoluted exothermic peak of sample C showing the occur-
7
rence of three exothermic peaks. The peaks correspond to (1) formation of
Al O , (2) formation of SiC, and (3) possible oxidation of the constituents.
2
3
Fig. 3. Plot showing the shift of the exothermic peak with the heating
rates for the pellets heated at (a) 10°C/min, (b) 40°C/min, and (c)
1
00°C/min.
1
7
below the reported value of the ignition temperature (1425°C) is
possibly due to the localized attenuation of high temperature.
As mentioned, the aluminothermic reduction of SiO2 is ob-
served to be initiated at temperatures Ն723°C, higher than the
melting point of aluminum (660.4°C). Poor wettability of liquid Al
on SiO may be one of the reasons for the higher reaction initiation
Fig. 5. Variation of reaction initiation temperature (T ) (a) and peak
2
i
1
2
temperature. Yi et al. have reported that the liquid wetting
mechanism operating in the SHS of the Ti–Al system is caused in
lowering the reaction initiation temperature with the increase of
the heating rate. However, the present investigation does not
follow the findings of Yi et al., possibly because of poor
wettability. During the cooling cycle of the pellets after attainment
of 1200°C, a minor exothermic peak could be observed at 545°C.
The exothermic peak is due to solidification of unreacted alumi-
num at undercooled temperature. The ratio of the area of the
endothermic peak in the heating cycle and the exothermic peak of
the cooling cycle can be used to derive information on the amount
of unreacted aluminum in the process.
Thermogravimetric analyses of the pellets show that there is
around 1.1% to 1.4% increase in weight of the samples during
heating. This increase can be attributed primarily to the oxidation
of either Al, Si, or SiC by the traces of oxygen (ϳ2 ppm) present
in Ar. It is also observed that practically no weight gain can be
noted up to 661°C, i.e., the melting point of aluminum. The weight
gain is also found to be independent of the heating rates.
temperature (T ) of aluminum oxidation (b) and formation of silicon
p
carbide (c), with the heating rates for the pellets.
Q ϭ m L
(4)
(5)
1
1
Q ϭ Ϫm L
2
2
where m1 and m2 are the weight of the initial and unreacted
aluminum, respectively, Q and Q are the total heat absorbed and
1
2
evolved for aluminum melting and solidification, respectively, and
L is the latent heat of fusion of Al per unit weight. From Eqs. (4)
and (5), it can be deduced that
Q
/Q
ϭ m
/m
ϭ A
/A
(6)
1
2
1
2
1
2
assuming the machine constant (proportionality constant) and the
mass of the system to be identical for the heating and cooling
cycles. Here, A and A are the areas of the peaks for melting and
1
2
(
2) Kinetic Analysis
Assuming that no appreciable change in the weight of the
solidification of aluminum, respectively. The fraction of aluminum
reacted f can be expressed as
sample occurs during cooling and that the melting and freezing
occur within a close temperature range, the heat equation can be
expressed as
m Ϫ m
A2
1
2
f ϭ
ϭ 1 Ϫ
(7)
m1
A1

9
18
Journal of the American Ceramic Society—Pathak et al.
Vol. 84, No. 5
Using Eq. (7), the fractions of aluminum reacted as a function
of heating rates were calculated and are presented in Table II. It is
apparent from the table that aluminum conversion to alumina has
a marginal negative dependence over the heating rates. This
observation can be explained in the light of the kinetics of reaction.
A higher heating rate, although it increases the rate of heat input,
decreases the time available for reaction. Transport of reactants,
wetting, and chemical reaction all being time-dependent phenom-
ena, scarcity of time is thus expected to lower the conversion of
aluminum.
Table II also incorporates the time lapse between the initiation
of the major exothermic reaction and attainment of peak temper-
ature. This time is considered to be the effective time for
exothermic reaction, assuming no significant conversion of reac-
tants beyond T . It has been noticed that the reaction time
p
decreases with an increase in heating rates. At high heating rates,
the number of reacting molecules acquiring energy to overcome
the activation energy barrier per unit time increases, resulting in a
lowering of the reaction time.
Fig. 6. Kissinger plots for the estimation of the activation energy of
aluminothermic reduction of silica (a) and the formation of SiC (b).
The most extensively used expression for determination of the
apparent activation energy of reaction in DTA is proposed by
1
8
Kissinger and can be expressed as
formation is controlled by the liberation of Si in the aluminother-
mic reduction process. However, the contribution of chemical
reaction between Si and C in controlling the overall reaction
cannot be ruled out.
2
ϪE/RTp
␤
/T ϭ ke
p
(8)
where ␤ is the heating rate, T is the peak temperature in kelvins,
E is the activation energy, k is the preexponential factor, and R is
the gas constant. A plot of ln (␤/T ) as a function of the inverse of
peak temperature (1/T ) for the aluminothermic reduction of silica
p
2
p
(3) X-ray and Microstructural Study
p
is presented in Fig. 6(a). The apparent activation energy for the
aluminothermic reduction process is calculated from the slope of
this linear plot and has been found to be 240 Ϯ 5 kJ/mol. The
estimated apparent activation energy is observed to match ex-
tremely well with the interdiffusion activation energy of Si and
oxygen (241.7 kJ/mol). This suggests that reduction of silica by
liquid aluminum in the present investigation is possibly controlled
by the diffusion of oxygen through Si formed at the reaction
interface.
The XRD pattern of powder sample P shows no evidence for
1
the formation of Al O , Si, or SiC in the product, indicating the
2
3
absence of any reaction during heat treatment. This is consistent
with the thermoanalytical observations mentioned earlier. How-
ever, formation of Al O , Si, and SiC was detected in powder
2
3
1
9
sample P , which was subjected to a higher heating rate (100°C/
3
min). XRD analyses of the pellets show that unreacted SiO , as
2
well as Al, is also present along with the reaction products (Al O
2
3
and SiC) after the SHS reaction, which supplements TG/DTA
observations. Further, the presence of metallic Si in the product
confirms that the reduction of SiO2 has taken place by the
aluminothermic route. From qualitative analyses using the peak
intensities of the X-ray diffractograms, it appears that the amount
of aluminum reacted during SHS decreases with an increase in the
heating rates, as has been suggested earlier from the thermoana-
lytical studies. The relative intensity ratios also show a higher
amount of SiC phase formation with an increase in the heating
rates (Table III).
A Kissinger plot for the estimation of activation energy for the
process of silicon carbide formation is shown in Fig. 6(b), and the
apparent activation energy has been calculated to be 255 Ϯ 7.5
kJ/mol. This activation energy is much higher than the reported
2
0
interdiffusion activation energy of Si and C, i.e., 12.98 kJ/mol.
The effect of heating rates on the SHS fabrication of TiC–Al O
2
3
from a TiO –Al–C mixture has shown that the SHS reaction is
2
2
1
controlled by the diffusion of carbon through solid TiC. Simi-
larly, the present SHS reaction involving the formation of SiC may
be controlled by carbon diffusion through SiC as well. The
formation of SiC has also been reported to be controlled by the
Figure 7 shows scanning electron microscopic images of sam-
ples C and C . From the SEM images (Figs. 7(a) and (b)), it may
7
8
2
2
formation of carbon-ion vacancies across the SiC layer. The
be noted that formation of SiC whiskers is restricted to a relatively
localized zone toward a particular surface instead of being distrib-
uted throughout the matrix. Similar observations have also been
made for all of the pellet samples (C through C ). To identify the
self-diffusion of carbon in polycrystalline ␤-SiC has been studied
2
3
by Hon et al. and they report values of 841 Ϯ 13.5 and 563 Ϯ
.7 kJ/mol for lattice and grain boundary diffusion, respectively.
The activation energy for the self-diffusion of Si in ␤-SiC is also
8
2
7
surface where whisker growth was favored, several TG/DTA
supplementary experiments on pellets with marked surfaces were
conducted at a heating rate of 20°C/min. It was identified that the
surface touching the alumina crucible promotes whisker growth,
whereas the other surfaces were free of any whiskers. The top and
side surfaces of the samples being exposed to the flowing gas are
2
4
estimated to be 912 Ϯ 4.8 kJ/mol. This indicates that the
diffusion of neither Si nor C through solid SiC is the rate-
controlling factor in the present system. The activation energy of
2
40 Ϯ 5 kJ/mol for the first exothermic peak is close to the second
activation energy, indicating that the process of silicon carbide
Table II. Details of the TG/DTA Analysis and Reaction Time during
Heating Cycle
Mass gain Ratio of areas
Al converted to
alumina (%)
Al actually reacted Time interval between
Sample
(%)
a and b
during SHS (%)
i p
T and T (min)
C2
C3
C4
C5
C6
C7
1.4
1.3
1.4
1.3
1.2
1.2
16.5
14.0
12.5
11.7
9.7
93.9
92.8
92.0
91.4
89.6
89.6
89.2
88.1
87.3
86.7
84.9
84.9
5.30
2.92
2.14
1.6
0.83
0.71
9.7

May 2001
Effect of Heating Rates on the Synthesis of Al O –SiC Composites
919
2
3
Table III. Details of the XRD Analyses of Samples C and C
2
7
Intensity of the peaks
Intensity ratio of the peaks
Sample
Si
SiO
2
SiC
Al
2
O
3
Al
SiO
2
/Si
SiC/Si
2 3
Al/Al O
C2
C7
3196
1516
375
356
364
393
1102
1143
290
394
0.117
0.234
0.11
0.259
0.263
0.343
cooled by convection and radiation, whereas the bottom surface
undergoes cooling only by conduction, and consequently, the heat
transfer is expected to be much slower. The slow rate of heat
transfer from the bottom surface of the pellets will enable the
surface to remain at a relatively higher temperature for a longer
period and that has possibly promoted the growth of SiC whiskers.
SEM images of sample C8 heated at 400°C/min from room
temperature to 1450°C and soaked for 10 min showed the
formation of SiC whiskers throughout the sample and the quantity
of whiskers was also found to be more (Fig. 7(c)). The formation
of whiskers is known to be by either a vapor-phase or liquid-phase
mass transfer process. The vapor-phase growth process is invari-
ably associated with loss of mass and should have been reflected
in the thermoanalytical studies. The absence of any significant
mass loss during processing emphasizes the formation of SiC
whiskers by the liquid-phase mass transfer process in the present
investigation. The growth behavior of SiC whiskers in sample C8
as observed in Fig. 7(c) suggests the above mechanism to be
operative. The growth of whiskers also depends on the time. The
absence of liquid Si for a sufficient length of time in the exposed
surfaces of the pellets studied in TG/DTA has led to the formation
of SiC particulates through the solid-state diffusion process,
whereas the retention of liquid Si for longer periods of time in the
bottom surfaces of pellets has helped in the formation of SiC
whiskers.
IV. Conclusions
A systematic study on the effect of heating rates on the SHS
formation of Al O –SiC composites has been conducted. The
2
3
powder mixture does not undergo any exothermic reaction at lower
heating rates possibly because of poor contact between the
reactants and segregation of liquid Al. The heating rates have
shown a pronounced effect on both the reaction initiation temper-
ature (T ) and the peak temperature (T ). Further, both T and T are
i
p
i
p
found to increase with an increase in the heating rates. The
reduction of silica is found to take place by the aluminothermic
route and the exothermic reduction reaction is initiated only after
melting of aluminum, suggesting reaction between solid silica and
liquid aluminum. Three exothermic peaks due to aluminothermic
reduction of silica, formation of SiC from the reaction of Si and C,
and oxidation of Al, Si, or SiC have been observed. The conver-
sion of aluminum marginally decreases with a gradual increase in
the heating rates. The apparent activation energies for the alumi-
nothermic reduction of silica and the formation of silicon carbide
are estimated to be 240 Ϯ 5 and 255 Ϯ 7.5 kJ/mol, respectively.
This indicates that possibly the diffusion of oxygen in the reaction
interface is the rate-controlling factor for this present SHS process.
The SEM results suggest that the formation of whiskers is possibly
due to the liquid-phase mass transfer process.
Acknowledgment
We are thankful to Mr. Ravi Kumar for his help in carrying out the XRD studies
during different stages of this investigation.
References
Fig. 7. SEM images of the two surfaces of pellet C7 showing the
formation of particulate in one surface (a) and formation of SiC whiskers
in the other surface (b). Extensive growth of SiC whiskers in pellet C8
1
C. Wei and P. F. Becher, “Development of SiC Whisker Reinforced Ceramics,”
Am. Ceram. Soc. Bull., 64 [2] 298–304 (1985).
2
2 3
T. N. Tiegs and P. F. Becher, “Sintered Al O –SiC Whisker Composites,” Am.
(heated at 400°C/min) is shown in (c).
Ceram. Soc. Bull., 66 [2] 339–42 (1987).

9
20
Journal of the American Ceramic Society—Pathak et al.
Vol. 84, No. 5
3
13
J. Homeny and W. L. Vaughn, “Silicon Carbide Whisker–Alumina Composites:
Z. A. Munir, “Synthesis of High Temperature Materials by Self-Propagating
Effect of Whiskers Surface Treatment on Fracture Toughness,” J. Am. Ceram. Soc.,
Combustion Methods,” Am. Ceram. Soc. Bull., 67 [2] 342–49 (1988).
1
4
7
3 [2] 394–402 (1990).
J. Zhao, L. C. Stearns, M. P. Harmer, H. M. Chan, G. A. Miller, and R. F. Cook,
N. P. Novikov, J. P. Borovinskaya, and A. G. Merzhanov; “Thermodynamic
4
Analysis of Self-Propagating High-Temperature Synthesis Reactions”; pp. 174–87 in
Combustion Processes in Chemical Technology and Metallurgy. Edited by A. G.
Merzhanov. Institute of Chemical Physics, Chernogolovka, Russia, 1975.
“
Mechanical Behavior of Alumina–Silicon Carbide ‘Nancomposites’,” J. Am. Ceram.
Soc., 76 [2] 503–10 (1993).
5
A. C. D. Chaklader, S. Das Gupta, E. C. Y. Lin, and B. Gutowski, “Al
2
O
3
–SiC
15
C. R. Bowen and B. Derby, “Self Propagating High Temperature Synthesis of
Composites from Aluminosilicate Precursors,” J. Am. Ceram. Soc., 73 [8] 2283–85
1992).
Ceramic Materials,” Br. Ceram. Trans., 96, 25–31 (1997).
(
16
I. Barin and G. Platzki, Thermochemical Data of Pure Substances, 3rd ed. VCH
6
2 3
H. M. Lee, H. L. Lee, and H. J. Lee, “Submicron Al O /SiC Composite Powder
Publishing, New York, 1995.
Preparation by SHS Technique,” J. Mater. Sci. Lett., 14, 1515–17 (1995).
17
A. Peng and Z. A. Munir, “The Effect of an Electric Field on Self-Sustaining
7
R. A. Cutler, K. M. Rigtrup, and A. V. Virkar, “Synthesis, Sintering, Microstruc-
Combustion Synthesis: Part II, Field Assisted Synthesis of ␤-SiC,” Met. Mater. Trans.
B, 26B, 587–93 (1995).
ture and Mechanical Properties of Ceramics Made by Exothermic Reaction,” J. Am.
Ceram. Soc., 75 [1] 36–43 (1992).
18
8
H. E. Kissinger, “Variation of Peak Temperature with Heating Rate in Differen-
D. Bandyopadhyay, L. C. Pathak, S. K. Das, I. Mukhergee, R. G. Ganguly, and P.
tial Thermal Analysis,” J. Res. Natl. Bur. Stand. (U.S.), 57 [4] 217–21 (1956).
Ramachandrarao, “Fabrication of Al O –SiC In-Situ Composites by a New Combus-
2
3
19
C. Hass, “The Diffusion of Oxygen in Silicon and Germanium,” J. Phys. Chem.
tion Technique,” Mater. Res. Bull., 32 [1] 75–82 (1997).
9
Solids, 15, 108–11 (1960).
D. Bandyopadhyay, L. C. Pathak, S. K. Das, R. Kumar, and P. Ramachandrarao,
2
0
R. Newmann and J. Wakefield, Semiconductive Materials; p. 201. Wiley-
Interscience, New York, 1962.
“
Synthesis of Ceramic Matrix Composites by RH-SHS”; pp. 599–604 in Proceedings
of the International Conference on Recent Advances in Metallurgical Processes.
Edited by D. H. Sastry, E. S. Dwarakadasa, G. N. K. Iyengar, and S. Subramanian.
New Edge Publishing, Delhi, India, 1997.
2
1
Y. Choi and S. W. Rhee, “Reaction of TiO
2
–Al–C in the Combustion Synthesis
of TiC–Al O Composites,” J. Am. Ceram. Soc., 78 [4] 986–92 (1995).
2
3
2
2
1
0
H. Zhou and R. N. Singh, “Kinetics Model for the Growth of Silicon Carbide by
L. L. Wang, Z. A. Munir, and J. B. Holt, “The Combustion Synthesis of Copper
Aluminides,” Metall. Trans., 21B, 567–77 (1990).
the Reaction of Liquid Silicon with Carbon,” J. Am. Ceram. Soc., 78 [9] 2456–62
1
1
W. C. Lee and S. L. Chang, “Ignition Phenomenon and Reaction Mechanisms of
(1995).
2
3
14
the Self-Propagating High Temperature Synthesis Reaction in the Titanium–Carbon–
M. H. Hon and R. F. Davis, “Self-Diffusion of C in Polycrystalline ␤-SiC,” J.
Mater. Sci., 14, 2411–21 (1979).
Aluminum System,” J. Am. Ceram. Soc., 80 [1] 53–61 (1997).
1
2
24
30
H. C. Yi, A. Petric, and J. J. Moore, “Effect of Heating Rate on the Combustion
Synthesis of Ti–Al Intermetallic Compounds,” J. Mater. Sci., 27, 6797–806 (1992).
M. H. Hon, R. F. Davis, and D. E. Newbury, “Self-Diffusion of Si in ␤-SiC,”
J. Mater. Sci., 15, 2073–80 (1980).
Ⅺ