Thermogravimetry, differential thermal analysis, and mass spectrometry study of the silicon nitride-boron carbide-carbon reaction system for the synthesis of silicon carbide-boron nitride composites

Article abstract of DOI:10.1111/j.1151-2916.2002.tb00444.x

Thermogravimetry, differential thermal analysis, mass spectrometry, and X-ray diffractometry were used to study the reaction process of the in situ reaction between Si₃N₄, B₄C, and carbon for the synthesis of silicon carbide-boron nitride composites. Atmospheres with a low partial pressure of nitrogen (for example argon + 5%-10% nitrogen) seemed to inhibit denitrification and also maintain a high reaction rate. However, the reaction rate decreased significantly in a pure nitrogen atmosphere. The experimental mass spectrometry results also revealed that B₄C in the Si₃N₄-B₄C-C system inhibited the reaction between Si₃N₄ and carbon and, even, the decomposition of Si₃N₄. The present results indicate that boron could be a composition stabilizer for ceramic materials in the Si-N-C system used at high temperature.

Full text of DOI:10.1111/j.1151-2916.2002.tb00444.x

J. Am. Ceram. Soc., 85 [9] 2256–60 (2002)
journal
Thermogravimetry, Differential Thermal Analysis, and
Mass Spectrometry Study of the Silicon Nitride–Boron Carbide–Carbon
Reaction System for the Synthesis of Silicon Carbide–Boron Nitride Composites
,
†,‡
,§,¶
,†
Guo-Jun Zhang,* Jian-Feng Yang,* and Tatsuki Ohji*
Synergy Materials Research Center, National Institute of Advanced Industrial Science and Technology (AIST),
Nagoya, Aichi 463-8687, Japan
Japan Science and Technology Corporation (JST) at AIST Chubu, Nagoya, Aichi 463-8687, Japan
Thermogravimetry, differential thermal analysis, mass spec-
trometry, and X-ray diffractometry were used to study the
reaction process of the in situ reaction between Si N , B C, and
homogeneous microstructures with fine and isotropically distrib-
uted BN flakes located at the grain boundaries of the SiC.
1
2
3
4
4
According to thermodynamic calculations, B C inhibits the
4
carbon for the synthesis of silicon carbide–boron nitride
composites. Atmospheres with a low partial pressure of nitro-
gen (for example argon ؉ 5%–10% nitrogen) seemed to
inhibit denitrification and also maintain a high reaction rate.
However, the reaction rate decreased significantly in a pure
nitrogen atmosphere. The experimental mass spectrometry
results also revealed that B C in the Si N –B C–C system
reaction between Si N and carbon to degas nitrogen, and this
3
4
reaction has been verified by chemical analysis of the elements in
14,15
composites obtained by in situ hot pressing.
The present
investigation focuses on analyzing the reaction process of Si N ,
3
4
B C, and carbon by thermogravimetry (TG), differential thermal
4
analysis (DTA), mass spectrometry (MS), and X-ray diffractom-
etry (XRD). The effect of various atmospheres on the reaction are
discussed in this paper.
4
3
4
4
inhibited the reaction between Si N and carbon and, even, the
3
4
decomposition of Si N . The present results indicate that boron
3
4
could be a composition stabilizer for ceramic materials in the
Si-N-C system used at high temperature.
II. Experimental Procedure
I. Introduction
The raw powders used were Si N (E-10 grade, mean particle
3
4
size 0.5 ␮m, oxygen content Ͻ2%, UBE Industries, Ltd., Yamagu-
RAPHITIC, hexagonal BN-based composites are an important
chi, Japan), B C (F1 grade, particle size 1 ␮m, Denki Kagaku
G
composite family that shows excellent thermal shock resis-
4
tance, good thermal conductivity, a low thermal expansion coef-
ficient, and high corrosion resistance under many corrosive cir-
cumstances, such as metal melts. In addition, BN composites
exhibit self-lubrication, because of the graphitic hexagonal struc-
ture of BN, and excellent machinability when the BN fraction
reaches ϳ20 vol%. Accordingly, BN composites have the poten-
Kogyo Co., Ltd., Tokyo, Japan), carbon (No. 2600 grade, particle
size 13 nm, Mitsubishi Chemical Corp., Tokyo, Japan), Al O
2 3
(particle size 0.2 ␮m, Daimei Chemical Co., Nagano, Japan), and
Y O (particle size 1.06 ␮m, Shin-Etsu Chemical Co., Ltd.,
2
3
1
2,15
Tokyo, Japan). In previous work,
10 wt% Al O –Y O (YA)
2 3 2 3
additives (7:3 weight ratio of Al O :Y O ) related to the SiC
2
3
2 3
1
–6
tial for use in various industrial fields.
BN composites usually are fabricated by hot pressing mechan-
ically mixed component powders
can form the component phases from in situ reactions.
previous works,
contents in the composites had been added to promote densifica-
tion and reaction and to improve crystallization of the BN phase.
To create a similar reaction condition in the chemical composition,
2
–6
or by mixing reactants that
7
–11
In our
1
0 wt% YA additives related to the Si N contents in the mixed
1
2–15
3 4
the in situ reaction
powders were added in the present work. Three mixed-powder
specimens were prepared by ball milling for 24 h in ethanol, using
ZrO (Y O ) balls, and subsequently dried. The sample designa-
Si N ϩ B C ϩ 2C ϭ 3SiC ϩ 4BN
(1)
3
4
4
2
2 3
was used to prepare SiC–BN composites with high strength, low
elasticity, and good strain tolerance. Compared with the BN
composites produced by hot pressing mechanically mixed compo-
tions are given in Table I.
TG-DTA (Model 2020, MAC Science Co., Ltd., Yokohama,
Japan) for the silicon nitride–boron carbide–carbon (SN–BC–C)
specimen was conducted from room temperature to 1700°C, at a
heating rate of 10°C/min, in flowing, high-purity argon, 95%
argon ϩ 5% nitrogen, 90% argon ϩ 10% nitrogen, and nitrogen
gas of 150 mL/min, to investigate the influence of various
atmospheres on the reaction process. Measurement was performed
twice for each condition, and very good repeatability was obtained.
MS (TG-MS Analyzer, Shimadzu Corp., Kyoto, Japan) was
conducted from room temperature to 1500°C, at a heating rate of
1
2
nent powders, the obtained in situ SiC–BN composites showed
J. J. Petrovic—contributing editor
Manuscript No. 187626. Received June 25, 2001; approved April 29, 2002.
This work has been supported by AIST, METI, Japan, as part of the Synergy
Ceramics Project. The authors are members of the Joint Research Consortium of
Synergy Ceramics.
1
0°C/min, in flowing helium gas of 30 mL/min. The maximum
temperature was 1700°C for the TG-DTA and 1500°C for the
TG-MS experiments. The phase compositions of the SN–BC–C
specimen (mixed powder compact formed by uniaxial pressing,
under 60 MPa) pressureless heat-treated at various temperatures
from 1200° to 1700°C, for 30 min, under 1 atm of nitrogen or
argon, were determined by XRD (Model 2500V, Rigaku Denki
Co., Tokyo, Japan), using CuK␣ radiation.
*
Member, American Ceramic Society.
†
Synergy Materials Research Center, National Institute of Advanced Industrial
Science and Technology.
‡
Now with Synergy Ceramics Laboratory, Fine Ceramics Research Association
(
FCRA), Nagoya, Aichi 463-8687, Japan.
§
Japan Science and Technology Corporation.
Now with Synergy Materials Research Center, AIST, Nagoya, Aichi 463-8686,
¶
Japan.
2
256

September 2002
TG-DTA and TG-MS Study of the Si N –B C–C Reaction System
2257
3
4
4
Table I. Sample Designation and Chemical Composition of Mixed Powders
Chemical composition (wt%)
Specimen
Actual chemical composition
Si
3
N
4
B
4
C
C
Al
2
O
3
Y O
2 3
(
(
(
1) SN
90 wt% Si N ϩ 10 wt% YA
90
78.66
60.05
0
0
0
7
5.51
4.20
3
2.36
1.80
3
4
2) SN–C
3) SN–BC–C
1 mol Si N ϩ 2 mol C ϩ 10 wt% YA related to Si N
13.47
10.28
3
4
3 4
1 mol Si N ϩ 1 mol B C ϩ 2 mol C ϩ 10 wt% YA
23.66
3
4
4
related to Si N
3
4
III. Results
The TG-DTA results obtained from the SN–BC–C specimen in
argon, with different partial pressures of nitrogen, are shown in
Fig. 1. In all atmospheres, weight loss occurred after ϳ1150°C.
However, the values of weight loss up to 1500°C decreased in the
nitrogen-containing atmosphere, because of inhibited decomposi-
atmosphere but became stronger in the argon atmosphere, and the
peak for Y Al O disappeared in the argon atmosphere but
3
5 12
became stronger in the nitrogen atmosphere. Moreover, a peak
possibly corresponding to Y Si C (ICDD Card No. 38-0794)
5
3
appeared at Ͼ1450°C in the argon atmosphere. According to the
peak intensities of the products BN and ␤-SiC and the reactants of
Si N and B C, we conclude that the reaction rate was much faster
3
4
4
tion of Si N . On the other hand, Fig. 1 shows obvious differences
3
4
in the argon atmosphere than in the nitrogen atmosphere.
in the DTA curves for the different atmospheres. In the atmo-
spheres with a low partial pressure of nitrogen, a sharp exothermic
peak occurred at temperatures Ͼ1700°C. The exothermic reaction
was remarkably inhibited in the pure nitrogen atmosphere, but a
small exothermic peak appeared before the large exothermic peak
at 1590°C. This small exothermic peak moved to 1548° and
MS was performed in a helium atmosphere on the three
specimens listed in Table I, and the results are illustrated in Fig. 3
and Table II. As shown in Fig. 3(a), denitrification of all specimens
began after ϳ1000°C and became substantial after ϳ1300°C. With
the addition of carbon to the Si N , denitrification became more
3
4
intense after ϳ1400°C. This finding suggests that the reaction
1
536°C in atmospheres with 10% and 5% nitrogen, respectively.
In the pure nitrogen atmosphere, only a small exothermic peak
appeared. The results demonstrated that the rate of the exothermic
reaction was very slow in a pure nitrogen atmosphere.
Si ϩ 3C ϭ 3SiC ϩ 2N (2)
3
N
4
2
occurred intensely above this temperature, although the beginning
temperature of the reaction calculated by thermodynamics was
The XRD patterns of the SN–BC–C specimens heat-treated at
1
4
1
488°C. Denitrification in the SN–BC–C specimen decreased
1
200°–1700°C in nitrogen or argon, and that of the mixed raw
significantly (see also Table II) over that in either the SN or the
SN–C specimen. This result cannot be explained by the decreased
Si N content caused in the SN–BC–C specimen by the addition of
powder, are shown in Fig. 2. At 1200°C, although the main
reaction apparently had not occurred, all of the peaks for ␣-Si N
3
4
shifted slightly to larger diffraction angles in both atmospheres.
3 4
B C, compared with that in the SN or SN–C specimen (see Table
Thus, some compositional change occurred with ␣-Si N , even at
4
3
4
I and Table II), but it is coincident with the thermodynamic
analysis of the reaction system and the results of chemical
composition analysis of the final composites prepared by hot
such low temperature. In addition, because of the addition of YA
¶
sintering aids, peaks corresponding to YO0.04C0.5 (ICDD Powder
Diffraction File Card No. 38-1113), Al OC (ICDD Card No.
2
1
4,15
pressing.
3
1
6-0148), and Y Al O (ICDD Card No. 09-0310) appeared at
200°C. As the temperature increased, the peak for YO0.04C0.5
3 5 12
The degasification behavior of CO and CO for the SN–C and
2
SN–BC–C specimens was also different, as shown in Figs. 3(b)
and (c). These figures show that CO was the main gas species
evolved. For all specimens, the degasification of physically ab-
disappeared in both the nitrogen and the argon atmospheres. On
the other hand, the peak for Al OC disappeared in the nitrogen
2
sorbed CO occurred in the temperature range from room temper-
2
ature to ϳ600°C, and the peak temperature was ϳ300°C (Fig.
3(c)). Meanwhile, as shown in Table II, the degasification of CO
¶
International Centre for Diffraction Data, Newtown Square, PA.
Fig. 1. TG-DTA spectra of SN–BC–C specimen in various atmospheres.

2
258
Journal of the American Ceramic Society—Zhang et al.
Vol. 85, No. 9
Fig. 2. XRD patterns of SN–BC–C specimen pressureless heat-treated at 1200°–1700°C, for 30 min, in argon or nitrogen.
and CO was less intense in the SN–BC–C specimen than in the
“absorbed” the [N] or nitrogen gas, according to reaction (5), and
decreased the denitrification (reaction (3)). This inference was
confirmed by MS, as shown in Fig. 3(a).
2
SN–C specimen. Almost no CO or CO could be detected in the
2
SN specimen, except for physically absorbed CO2.
Although Si(g), Si N(g), Si (g), SiN(g), and SiO(g) also were
As shown in Fig. 3(a) and Table II, although denitrification
occurred in all three of the specimens, denitrification in the
SN–BC–C specimen was the lowest in the experimental temper-
2
2
1
6
observed by Rocabois et al. in the Si N system under MS, and
3
4
the existence of Al O(g) and Y O(g) was reasonably possible, the
2
2
pressure of these substances would have been very low at the
detector of the mass spectrometric apparatus at the temperature
ature range. This result indicates that the reactant B C not only
4
“instantly absorbed” the [N] or nitrogen gas decomposed by Si N
3
4
(
ϳ250°C) used in the present work; consequently, no attention
but also inhibited the reaction between Si N and carbon (reaction
3 4
12
was paid to these substances in the current study.
(2)), as expected from thermodynamic analysis. This result may
indicate that boron is a good composition stabilizer for ceramics in
the Si-N-C system used at high temperature. However, nitrogen
will still escape, because of reaction dynamics, including the
diffusion and reaction rates of the atoms. This component of
nitrogen escape could be decreased in the sintering process for
making such in situ composites by either hot pressing or increasing
the partial pressure of nitrogen in the sintering atmosphere.
A useful inference obtained from Fig. 1 is that using atmo-
spheres with a controlled low nitrogen partial pressure—for
example, argon ϩ 5%–10% nitrogen—while making the compos-
ites should be beneficial for inhibiting the denitrification process
while maintaining the high reaction rate (high exothermic).
The main reason for the degasification behaviors of CO and
CO was considered to be carbothermal reduction of the Al O and
IV. Discussion
Si N does not melt but, rather, sublimes at temperatures
3
4
1
6,17
Ͻ1900°C under 1 atm of nitrogen gas:
Si N ͑s͒ 3 3Si͑s͒ ϩ 2N ͑g͒
(3)
3
4
2
High nitrogen pressure decreases the decomposition process.
Si N has been used to produce SiC whiskers, by reduction with
carbon.
is lower, and thus, the reduction temperature is higher than in an
argon atmosphere. Moreover, the morphology of the in situ formed
3
4
1
8–20
In a nitrogen atmosphere, the reported reduction rate
1
9
SiC is dependent on the atmosphere.
2
2 3
In the present investigation, similar results were obtained; that
is, reaction (1) occurred rapidly in the argon atmosphere (Figs. 1
and 2). The reaction process was closely related to the decompo-
sition process of the reactant Si N . According to the calculated
Y O additives, as well as the SiO2 surface oxidation layer of
2 3
Si N . Possible reactions related to the formation of CO and CO
3
4
2
are as follows:
Al O ϩ C 3 Al O͑g͒ ϩ CO and/or CO
2
3
4
2
1
(7)
(8)
results of Greskovich and Procchazka, the equilibrium partial
pressures of nitrogen gas resulting from the decomposition of
Si N at different temperatures are ϳ0.001 atm at 1400°C, 0.005
2
3
2
Al O ϩ C 3 Al OC ϩ CO and/or CO
2
3
2
2
2
3
4
atm at 1500°C, 0.02 atm at 1600°C, and 0.1 atm at 1700°C. The
current DTA results showed that reaction process (1) markedly
decreased in the atmosphere with a nitrogen partial pressure of
Ͼ0.1 atm (Fig. 1). Consequently, the following reaction sequence
is suggested:
Y O ϩ C 3 Y O͑g͒ ϩ CO and/or CO
(9)
2
3
2
Y O ϩ C 3 YO0.04C0.5 ϩ CO and/or CO2
(10)
(11)
(12)
(13)
2
3
YO0.04C0.5 ϩ Si 3 Y Si C ϩ CO and/or CO
2
5
3
Si N ͑s͒ 3 3Si͑s͒ ϩ 4͓N͔ or 2N
2
(4)
(5)
(6)
3
4
SiO ϩ C 3 SiC͑s͒ ϩ CO and/or CO
2
2
B C ϩ 4͓N͔ or 2N 3 4BN ϩ C
4
2
SiO ϩ C 3 SiO͑g͒ ϩ CO and/or CO
2
2
Si͑s͒ ϩ C 3 SiC
Calculations made from the amounts of CO and CO , summarized
2
where [N] represents a highly reactive nitrogen atom. The differ-
in Table II, and comparison with the starting compositions of the
mixed powders, shown in Table I, indicate that not all of the YA
ence for specimen SN–BC–C was that B C in the reactants
4

September 2002
TG-DTA and TG-MS Study of the Si N –B C–C Reaction System
2259
3
4
4
Table II. Accumulated Weight Loss of
Various Gases by Degasification during MS
Measurements from Room Temperature to
1
500°C in Helium
Weight loss (%)
CO
Specimen
N
2
CO
2
SN
0.630
0.783
0.344
0.087
4.166
2.467
0.104
0.325
0.191
SN–C
SN–BC–C
In regard to the different degasification behaviors of CO and
CO for the specimens, one of the main reasons for the lower
2
degasification of CO and CO2 in the SN–BC–C specimen is
considered to be the kinetic competition of reaction (1) with
reactions (7)–(13) (contending for carbon). On the other hand, in
the case of the SN–BC–C specimen, another explanation may exist
for part of the difference in the degasification behaviors of CO and
CO , based on the thermodynamic calculation. For example,
2
another reaction might occur in the presence of B C:
4
3
SiO ϩ B C ϩ 2C 3 2B O ͑l,g͒ ϩ 3SiC͑s͒
(14)
2
4
2
3
The formed B O would either produce a glass phase with the
2
3
sintering additives Al O and Y O or partially evaporate as B O
2 3
2
3
2
3
gas. In either case, the formation of CO or CO2 would be
decreased by the formation of B O .
2
3
Meanwhile, the results demonstrated in Fig. 2 and Table II
indicate that the reaction process under the current condition of
pressureless heat treatment is much different from that under the
1
2,14
hot-pressing condition reported in our previous work.
Previ-
ously, we reported no obvious compositional changes when hot
pressing was used to prepare SiC–BN composites in an argon
atmosphere. This finding points out that the above carbothermal
reduction reactions can be inhibited by hot pressing, through
preventing the emittance of the gas products CO and/or CO . In
2
addition, under the present pressureless heat treatment, the rate for
reaction (1) (Fig. 2) was much slower than that under hot pressing,
where the reaction was almost complete at 1700°C in an argon
1
4
atmosphere. Thus, applied stress must be beneficial to reaction
1), by increasing the density of the specimen and the contact
(
surfaces of the reactant particles. This finding implies that the
reaction process of reaction (1) is diffusion controlled.
V. Conclusions
The reaction process of the in situ reaction between Si N , B C,
3
4
4
and carbon was studied by TG, DTA, MS, and XRD. The reaction
process proved to be closely related to the decomposition process
of Si N . Atmospheres with a low partial pressure of nitrogen
3
4
gas—for example, argon ϩ 5%–10% nitrogen—could inhibit
denitrification and, at the same time, maintain a high reaction rate.
The reaction rate decreased significantly in a pure nitrogen
atmosphere, according to the weakened exothermic peak on the
DTA curve. The obtained experimental results also revealed that
B C in the reaction system Si N –B C–C inhibited reaction
4
3
4
4
between Si N and carbon and, even, the decomposition of Si N ,
3
4
3 4
itself. In other words, boron should be a good composition
stabilizer for ceramic materials in the Si-N-C system used at high
temperature.
Fig. 3. MS results for (a) nitrogen, (b) CO, and (c) CO2.
additives were carbothermally reduced according to reactions
7)–(11). This finding implies that some reasons must exist for the
Acknowledgments
(
different degasification behaviors in these specimens. In addi-
tion, because of reaction (11), the XRD peaks for silicon,
formed by reaction (4), were not detected (Fig. 2). Meanwhile,
our results suggest that reactions (8), (10), and (11) are faster in
the argon atmosphere than in the nitrogen atmosphere, so that
almost no Y Al O forms at temperatures Ͼ1450°C in the
The authors are grateful to Ms. Miyuki Mori, Mr. Yoshihiro Takai, and Mr.
Toshimi Hosoi, Toray Research Center, Japan, for TG-DTA and TG-MS analyses.
References
1
3
5
12
A. Lipp, K. A. Schwetz, and K. Hunold, “Hexagonal Boron Nitride: Fabrication,
argon atmosphere.
Properties and Applications,” J. Eur. Ceram. Soc., 5, 3–9 (1989).

2
260
Journal of the American Ceramic Society—Zhang et al.
Vol. 85, No. 9
2
12
K. S. Mazdiyasni and R. Ruh, “High/Low Modulus Si
3
N
4
–BN Composite for
G. J. Zhang and T. Ohji, “In Situ Reaction Synthesis of SiC–BN Composites,”
J. Am. Ceram. Soc., 84 [7] 1475–79 (2001).
Improved Electrical and Thermal Shock Behavior,” J. Am. Ceram. Soc., 64 [7]
1
3
4
15–19 (1981).
G. J. Zhang and T. Ohji, “Effect of BN Content on Elastic Modulus and Bending
Strength of SiC–BN in Situ Composites,” J. Mater. Res., 15 [9] 1876–80 (2000).
3
E. H. Lutz and M. V. Swain, “Fracture Toughness and Thermal Shock Behavior
1
4
of Silicon Nitride–Boron Nitride Ceramics,” J. Am. Ceram. Soc., 75 [1] 67–70 (1992).
G. J. Zhang, Y. Beppu, T. Ohji, and S. Kanzaki, “Reaction Mechanism and
4
Microstructure Development of Strain Tolerant in Situ SiC–BN Composites,” Acta
Mater., 49 [1] 77–82 (20001).
P. G. Valentine, A. N. Palazotto, R. Ruh, and D. C. Larsen, “Thermal Shock
Resistance of SiC–BN Composites,” Adv. Ceram. Mater., 1 [1] 81–87 (1986).
1
5
5
2 3 2 3
G. J. Zhang, J. F. Yang, Z. Y. Deng, and T. Ohji, “Effect of Y O –Al O
G. J. Zhang and Z. Q. Bao, “BN–TiB
2
Electrical Conductive and Corrosive
Additive on the Phase Formation and Densification Process of in Situ SiC–BN
Composite,” J. Ceram. Soc. Jpn., 109 [1] 45–48 (2001).
Resistant Composite Ceramics for Aluminum Evaporation Source,” Bull. Chinese
Ceram. Soc., 9 [1] 28–33 (1990).
1
6
6
P. Rocabois, C. Chatillon, and C. Bernard, “Thermodynamics of the Si-O-N
T. Kanai, K. Tanemoto, and H. Kubo, “Hot-Pressed BN–AlN Ceramic Compos-
System: I, High-Temperature Study of the Vaporization Behavior of Silicon Nitride
ites of High Thermal Conductivity,” Jpn. J. Appl. Phys., 29 [4] 683–87 (1990).
7
by Mass Spectrometry,” J. Am. Ceram. Soc., 79 [5] 1351–60 (1996).
W. S. Coblenz and D. Lewis, “In Situ Reaction of B
2
O
3
with AlN and/or Si
3
N
4
to
17
D. W. Richerson, Modern Ceramic Engineering—Properties, Processing, and
Form BN-Toughened Composites,” J. Am. Ceram. Soc., 71 [12] 1080–85 (1988).
Use in Design; p. 37. Marcel Dekker, New York, 1982.
8
T. Kusunose, Y. H. Chao, T. Sekino, and K. Niihara, “Mechanical Properties of
18
J. B. Li, G. Peng, S. R. Chen, Z. G. Chen, and J. G. Wu, “Formation and
Si N /BN Composites by Chemical Processing,” Key Eng. Mater., 161–63, 475–80
3
4
Morphology of 2H–SiC Whiskers by the Decomposition of Silicon Nitride,” J. Am.
Ceram. Soc., 73 [4] 919–22 (1990).
(
1999).
9
T. Kanai, K. Tanemoto, and H. Kubo, “Reaction-Bonded Boron Nitride–Alumi-
num Nitride,” Jpn. J. Appl. Phys., 30 [6] 1235–38 (1991).
19
H. Wang and G. S. Fischman, “In Situ Synthesis of Silicon Carbide Whiskers
from Silicon Nitride Powders,” J. Am. Ceram. Soc., 74 [7] 1519–22 (1991).
1
0
20
G. J. Zhang, J. F. Yang, and T. Ohji, “In Situ Si
Preparation, Microstructures and Properties,” Mater. Sci. Eng. A, A328, 201–205
2002).
3
N
4
–SiC–BN Composites:
J. Tian, J. Li, and L. Dong, “Synthesis of Silicon Nitride/Silicon Carbide
Nanocomposite Powders through Partial Reduction of Silicon Nitride by Pyrolyzed
(
Carbon,” J. Am. Ceram. Soc., 82 [9] 2548–50 (1999).
1
1
21
G. J. Zhang, J. F. Yang, T. Ohji, and S. Kanzaki, “In Situ Reaction Synthesis of
Non-Oxide Boron Nitride Composites,” Adv. Eng. Mater., 4, 15–17 (2002).
C. Greskovich and S. Procchazka, “Stability of Si N and Liquid Phase(s) during
3
4
Sintering,” J. Am. Ceram. Soc., 64 [7] C-96–C-97 (1981).
Ⅺ