Effects of Preformed Composition and Pore Size on Microstructure and Properties of SiCf/SiC Composites via Reactive Melt Infiltration

Abstract

This study investigated the influence of preformed composition and pore size on the microstructure and properties of SiCf/SiC composites fabricated via reactive melt infiltration (RMI). The process began with the impregnation of SiC fiber cloth with phenolic resin, followed by lamination and pyrolysis. Subsequent steps included further impregnations with phenolic resin, SiC slurry, and carbon black slurry, each followed by additional pyrolysis. This process resulted in three types of preforms, designated as PP, PS, and PC. These preforms exhibited a multimodal distribution of pore size, with peak pore diameters around 5 μm for PP, ranging from 200 nm to 4 μm for PS, and approximately 150 nm for PC. The preforms were then subjected to molten silicon infiltration at 1600 °C under vacuum for 1 h to create SiCf/SiC composites. The PP preform contained only pyrolytic carbon, leading to a composite with high closed porosity and unreacted carbon, resulting in poor mechanical properties. The PS preform, which was impregnated with SiC particles, displayed an optimized pore size distribution but retained significant amounts of residual silicon and carbon in the final composite. In contrast, the PC preform featured both an ideal pore size distribution and an adequate amount of carbon, achieving high density and low porosity with reduced residual phases in the final composite. This optimization led to a flexural strength of 152.4 ± 15.4 MPa, an elastic modulus of about 181.1 ± 0.1 GPa, and a thermal conductivity of 27.7 W/mK in the SiCf/SiC composites product. These findings underscore the importance of preform optimization in enhancing the performance of SiCf/SiC composites, potentially paving the way for more reliable nuclear fuel cladding solutions.

Full text

Citation: Nie, H.; Si, P.; Ren, Q.; Yin,
Z.; Cao, T.; Huang, Z.; Huang, Q.; Li,
Y. Effects of Preformed Composition
and Pore Size on Microstructure and
Properties of SiC
f
/SiC Composites via
Reactive Melt Infiltration. Materials
2024,17, 5765. https://doi.org/
10.3390/ma17235765
Academic Editor: Alexey Smolin
Received: 31 October 2024
Revised: 20 November 2024
Accepted: 21 November 2024
Published: 25 November 2024
Copyright: © 2024 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Article
Effects of Preformed Composition and Pore Size on
Microstructure and Properties of SiCf/SiC Composites via
Reactive Melt Infiltration
Haifeng Nie 1,2,3, Pingzhan Si 1, Quanxing Ren 2,3, Ziqiang Yin 2,3, Tihao Cao 2,3 , Zhengren Huang 2,3,4,5,
Qing Huang 2,3,4,5 and Yinsheng Li 2,3,4,5,*
1College of Materials Science and Engineering, China Jiliang University, Hangzhou 310018, China;
niehaifeng@nimte.ac.cn (H.N.); pzsi@cjlu.edu.cn (P.S.)
2Zhejiang Key Laboratory of Data-Driven High-Safety Energy Materials and Applications, Ningbo Key
Laboratory of Special Energy Materials and Chemistry, Ningbo Institute of Materials Technology and
Engineering, Chinese Academy of Sciences, Ningbo 315201, China; renquanxing@nimte.ac.cn (Q.R.);
yinziqiang@nimte.ac.cn (Z.Y.); zhrhuang@nimte.ac.cn (Z.H.); huangqing@nimte.ac.cn (Q.H.)
3Qianwan Institute of CNITECH, Ningbo 315336, China
4Advanced Energy Science and Technology Guangdong Laboratory, Huizhou 516003, China
5University of Chinese Academy of Sciences, Beijing 100049, China
*Correspondence: liyinsheng@nimte.ac.cn
Abstract: This study investigated the influence of preformed composition and pore size on the
microstructure and properties of SiC
f
/SiC composites fabricated via reactive melt infiltration (RMI).
The process began with the impregnation of SiC fiber cloth with phenolic resin, followed by lamination
and pyrolysis. Subsequent steps included further impregnations with phenolic resin, SiC slurry,
and carbon black slurry, each followed by additional pyrolysis. This process resulted in three types
of preforms, designated as PP, PS, and PC. These preforms exhibited a multimodal distribution
of pore size, with peak pore diameters around 5
µ
m for PP, ranging from 200 nm to 4
µ
m for PS,
and approximately 150 nm for PC. The preforms were then subjected to molten silicon infiltration
at
1600 ◦C
under vacuum for 1 h to create SiC
f
/SiC composites. The PP preform contained only
pyrolytic carbon, leading to a composite with high closed porosity and unreacted carbon, resulting in
poor mechanical properties. The PS preform, which was impregnated with SiC particles, displayed
an optimized pore size distribution but retained significant amounts of residual silicon and carbon
in the final composite. In contrast, the PC preform featured both an ideal pore size distribution
and an adequate amount of carbon, achieving high density and low porosity with reduced residual
phases in the final composite. This optimization led to a flexural strength of 152.4
±
15.4 MPa, an
elastic modulus of about 181.1
±
0.1 GPa, and a thermal conductivity of 27.7 W/mK in the SiC
f
/SiC
composites product. These findings underscore the importance of preform optimization in enhancing
the performance of SiC
f
/SiC composites, potentially paving the way for more reliable nuclear fuel
cladding solutions.
Keywords: SiC
f
/SiC composites; reactive melt infiltration; preformed composition; pore size; properties
1. Introduction
SiC fiber-reinforced SiC matrix composites (SiC
f
/SiC) possess significant potential for
applications in aerospace, defense, and nuclear industries due to their exceptional proper-
ties, including their lightweight design, high strength, high toughness, corrosion resistance,
high-temperature resistance, radiation resistance, and oxidation resistance
[1–5]
. Further-
more, SiC
f
/SiC ceramic matrix composites typically do not experience catastrophic failure
due to their non-brittle fracture characteristics, which are less sensitive to cracks
[1,2,6]
. Ben-
efiting from these excellent properties, SiC
f
/SiC composites have become highly promising
Materials 2024,17, 5765. https://doi.org/10.3390/ma17235765 https://www.mdpi.com/journal/materials

Materials 2024,17, 5765 2 of 13
candidates for accident-tolerant nuclear fuel cladding. They could effectively prevent dis-
asters like the Fukushima nuclear accident, which was primarily caused by the zirconium–
water reaction of zirconium alloy cladding tubes at high temperatures, resulting in the
production of large amounts of hydrogen and triggering a hydrogen explosion [
7
]. Com-
pared to zirconium alloys, SiC
f
/SiC composites offer several advantages: (1) superior
temperature resistance, allowing for long-term use at 800
◦
C and short-term exposure
to temperatures exceeding 1200
◦
C; (2) a neutron absorption cross-section that is 15%
lower than that of zirconium alloys; and (3) enhanced mechanical properties, including
lightweight construction, high strength, high toughness, and high hardness [
6
,
8
]. These ad-
vantages have enabled SiC
f
/SiC composites to be an ideal structural material in advanced
nuclear energy systems.
SiC
f
/SiC composites are primarily fabricated through various techniques, such as the
chemical vapor infiltration (CVI) [
9
–
11
], polymer infiltration and pyrolysis (PIP) [
12
–
14
], re-
active melt infiltration (RMI) [
15
–
17
], and nano-infiltration transient eutectic (NITE) [
18
–
20
]
methods. Among these, the RMI approach yields SiC
f
/SiC composites with exceptionally
low porosity and high density, which facilitate net shaping during material preparation.
This capability allows for the fabrication of composite materials with intricate geometries.
The RMI process involves the creation of a porous SiC
f
/C preform, followed by the for-
mation of a SiC ceramic matrix to produce the SiC
f
/SiC composites. Typically, the porous
SiC
f
/C preform is generated by introducing carbon into the woven SiC fiber structure
using CVI or PIP processes, utilizing carbon sources like carbon particles or resin. The
SiC
f
/SiC composites are then formed through reactions between the carbon in the SiC
f
/C
preform and molten silicon via liquid-phase silicon infiltration (LSI) or vapor-phase silicon
infiltration (VSI). Although SiC
f
/SiC composites produced by the RMI method exhibit
extremely low porosity, high density, and facilitate net shaping during material prepara-
tion, controlling the silicon and carbon contents during RMI presents challenges. Indeed,
controlling the contents of residual silicon and carbon in reaction-bonded silicon carbide
ceramics (RBSC) has been effectively addressed in a previous study [
21
]. Several common
scientific methods and mechanisms can be beneficially referenced for SiC
f
/SiC compos-
ites fabricated via the RMI method. However, it is important to note that RMI-SiC
f
/SiC
composites present more complexities compared to RBSC, primarily due to two factors:
(1)
The green body of SiC ceramics can be obtained through simple isostatic pressing or
colloidal forming processes, whereas the fabrication of SiC
f
/SiC composite preforms
involves more intricate methods such as precursor infiltration pyrolysis, chemical
vapor infiltration, or slurry impregnation. Additionally, regulating the composi-
tion and pore structure of these preforms is more challenging than for SiC ceramic
green bodies.
(2) Due to the critical issue of SiC fiber damage at high temperatures, the RMI processing
temperature for SiC
f
/SiC composites needs to be kept relatively low (
≤
1600
◦
C),
unlike the higher temperatures (≥1650 ◦C) feasible for SiC ceramics.
The abovementioned difficulties could lead to the presence of unnecessary free sil-
icon within the SiC ceramic matrix. The melting of residual silicon and its subsequent
reactions with the interfacial layer and fibers at elevated temperatures during RMI could
adversely affect the performance of the final product. The introduction of pyrolytic carbon
(PyC) or BN/SiC interphase layers could partially mitigate the issues related to interfacial
reactions [22].
Minimizing the amount of residual silicon in the final product is crucial and requires
precise control of both the carbon content and the volume/size of the pores in the preforms.
Achieving this level of control in the traditional RMI process for preparing SiC
f
/C preforms
is challenging, particularly because it relies solely on carbon sources. In the RMI process, the
pore structure [
23
–
25
] and size [
16
,
26
] of the preform could strongly affect the wettability
of liquid Si and the C–Si reaction. The pore structure and size in preforms can be adjusted
by synthesizing mesoporous carbons [
27
] or by changing solid loads [
28
]. One of the main
challenges of the RMI process is designing preforms that yield the least amounts of residual

Materials 2024,17, 5765 3 of 13
silicon and carbon in the final SiC
f
/SiC composite product. In the past, several ideal
preforms have been developed to achieve effective RMI. Guo et al. proposed an alternative
method to enhance the RMI approach by synthesizing a special porous carbon (C
g
) and
incorporating it into porous 2D SiC
f
/SiC composites prepared by chemical vapor infiltration
(CVI) [
29
]. This resulted in a two-stage pore structure, leading to more complete RMI and
less content and better dispersion of the residual silicon. The SiC
f
/SiC composites achieved
a high flexural strength of 808.7
±
10.2 MPa. Chen et al. prepared C
f
/B
4
C-C preforms with
different pore structures using slurry impregnation and sol impregnation methods [
30
].
Preforms made by the sol impregnation method exhibited more uniform pore structures
and a reduced amount of residual silicon, consequently increasing the composite
′
s flexural
strength from 145 MPa to 192 MPa. Optimized pore structure prefabrication is key to
producing high-quality SiC matrix composites via the RMI method. However, these
methods involve complex and costly precast production processes, making them less
favorable for mass production.
The primary objective of this study was to develop a simple and cost-effective method
for the fabrication of high-performance SiC
f
/SiC composites based on the RMI approach.
The initial step involved creating a fiber preform by impregnating SiC fiber cloth with phe-
nolic resin, followed by laminating and pyrolyzing the assembly. Subsequently, the preform
underwent further impregnation with either phenolic resin, silicon carbide slurry, or carbon
black slurry. This process was repeated with additional pyrolysis cycles to achieve preforms
with tailored compositions and pore structures. These preforms were then subjected to
molten silicon infiltration, resulting in the formation of SiC
f
/SiC composites. A systematic
investigation was conducted to examine the microstructure and properties of the resultant
SiC
f
/SiC composites. Through comparative analysis, the optimal preformed composition
and pore structure for the preparation of SiC
f
/SiC composites were identified. Furthermore,
the underlying scientific mechanisms governing these observations were elucidated.
2. Experimental Procedure
2.1. Materials Preparation
The sample preparation procedures are schematically depicted in Figure 1. Initially, a
two-dimensional (2D) satin-woven SiC fiber cloth (Cansas 3303, Fujian Liya New Material
Co., Ltd., Quanzhou, China) was coated with PyC via chemical vapor deposition (CVD)
technology. Subsequently, a SiC coating was applied using polymer infiltration and py-
rolysis (PIP) techniques. The fiber cloth with PyC/SiC coating was then immersed in a
50 wt.%
ethanol solution of phenolic resin (BR2123, Henan Borun New Materials Co., Ltd.,
Zhengzhou, China), placed in an impregnation tank (DN300, Shenyang Weike Vacuum
Technology Co., Ltd., Shenyang, China) and subjected to a pressure of 3 kPa for 15 min,
followed by an increase to 20 bar for 2 h. Afterwards, 20 stacked sheets of the SiC fiber
cloth were compressed using a graphite fixture (Custom specifications, Yuyao Yujun Metal
Materials Business Department, Yuyao, China) and subsequently dried and cured at 120
◦
C
for 12 h.
These compacted bodies were then pyrolyzed at 900
◦
C for 1 h under an argon
atmosphere. Following pyrolysis, the compacted bodies were impregnated with phenolic
resin, SiC slurry, and carbon black slurry and then pyrolyzed again to obtain three types of
preforms: PP, PS, and PC. Specifically, the PP preform was obtained through additional
cycles of impregnation with a 50 wt.% phenolic resin solution followed by pyrolysis. The
PS preform was prepared using Slurry Infiltration (SI) technology with an aqueous slurry
containing 20 vol.%
α
-SiC powder (D
50
= 0.5
µ
m, Qinhuangdao Eno Material Co., Ltd.,
Qinhuangdao, China) and 2.5 wt.% tetramethylammonium hydroxide (TMAH, Shanghai
Aladdin Bio-Chem Technology Co., Ltd., Shanghai, China) as a dispersant based on the
mass of
α
-SiC powder. Similarly, the PC preform was produced using SI technology with
an aqueous slurry comprising 30 vol.% carbon black (particle size 20 nm, Qinhuangdao
Eno Material Co., Ltd., China) and 2.25 wt.% polyvinylpyrrolidone (PVP, Shanghai Aladdin
Bio-Chem Technology Co., Ltd., China) as a dispersant based on the mass of carbon black.

Materials 2024,17, 5765 4 of 13
For all samples, the second round of impregnation and pyrolysis conditions remained
consistent with the first round.
Materials 2024, 17, x FOR PEER REVIEW 4 of 14
Figure 1. Schematic diagram illustrating the fabrication process of SiCf/SiC composite samples.
These compacted bodies were then pyrolyzed at 900 °C for 1 hour under an argon
atmosphere. Following pyrolysis, the compacted bodies were impregnated with phenolic
resin, SiC slurry, and carbon black slurry and then pyrolyzed again to obtain three types
of preforms: PP, PS, and PC. Specically, the PP preform was obtained through additional
cycles of impregnation with a 50 wt.% phenolic resin solution followed by pyrolysis. The
PS preform was prepared using Slurry Inltration (SI) technology with an aqueous slurry
containing 20 vol.% α-SiC powder (D50 = 0.5 μm, Qinhuangdao Eno Material Co., Ltd.,
Qinhuangdao, China) and 2.5 wt.% tetramethylammonium hydroxide (TMAH, Shanghai
Aladdin Bio-Chem Technology Co., Ltd., China) as a dispersant based on the mass of α-
SiC powder. Similarly, the PC preform was produced using SI technology with an aque-
ous slurry comprising 30 vol.% carbon black (particle size 20 nm, Qinhuangdao Eno Ma-
terial Co., Ltd., China) and 2.25 wt.% polyvinylpyrrolidone (PVP, Shanghai Aladdin Bio-
Chem Technology Co., Ltd., China) as a dispersant based on the mass of carbon black. For
all samples, the second round of impregnation and pyrolysis conditions remained con-
sistent with the rst round.
During the subsequent reactive melt inltration process, the PP, PS, and PC preforms
were individually placed in graphite crucibles (Custom specications, Yuyao Yujun Metal
Materials Business Department, Yuyao, China), each topped with an adequate amount of
silicon powder (particle size 1–3 mm, purity 99.9999%, Qinhuangdao Eno Material Co.,
Ltd., Qinhuangdao, China). The samples were then heated to 1600 °C and held under vac-
uum conditions for 1 hour in a graphite furnace (ZGXYS-250, Shenyang Sante Vacuum
Technology Co., Ltd., Shenyang, China). This process ultimately yielded SiCf/SiC compo-
site samples. Prior to the analyses and characterizations, the composite samples were sec-
tioned, surface ground, and polished to a 1 μm nish.
2.2. Characterizations
The densities (ρ) and apparent porosities of the preforms and nal SiCf/SiC composite
products were determined using the Archimedes water displacement method. The pore
size distributions in the preforms (measuring 10 mm × 10 mm × 10 mm) were analyzed
through Mercury Porosimetry (MIP, AutoPore IV 9500, Micromeritics Instrument Co.,
Ltd., Norcross, GA, USA). The phase compositions of the samples were examined via X-
ray diractometry (XRD: D8 Advance, Bruker Co., Ltd., Karlsruhe, Germany). The micro-
structures and elemental distributions of the specimens were observed using a scanning
electron microscope equipped with an energy dispersive spectrometer (SEM-EDS, Quanta
Figure 1. Schematic diagram illustrating the fabrication process of SiCf/SiC composite samples.
During the subsequent reactive melt infiltration process, the PP, PS, and PC preforms
were individually placed in graphite crucibles (Custom specifications, Yuyao Yujun Metal
Materials Business Department, Yuyao, China), each topped with an adequate amount of
silicon powder (particle size 1–3 mm, purity 99.9999%, Qinhuangdao Eno Material Co., Ltd.,
Qinhuangdao, China). The samples were then heated to 1600
◦
C and held under vacuum
conditions for 1 h in a graphite furnace (ZGXYS-250, Shenyang Sante Vacuum Technology
Co., Ltd., Shenyang, China). This process ultimately yielded SiCf/SiC composite samples.
Prior to the analyses and characterizations, the composite samples were sectioned, surface
ground, and polished to a 1 µm finish.
2.2. Characterizations
The densities (
ρ
) and apparent porosities of the preforms and final SiC
f
/SiC composite
products were determined using the Archimedes water displacement method. The pore
size distributions in the preforms (measuring 10 mm
×
10 mm
×
10 mm) were analyzed
through Mercury Porosimetry (MIP, AutoPore IV 9500, Micromeritics Instrument Co., Ltd.,
Norcross, GA, USA). The phase compositions of the samples were examined via X-ray
diffractometry (XRD: D8 Advance, Bruker Co., Ltd., Karlsruhe, Germany). The microstruc-
tures and elemental distributions of the specimens were observed using a scanning electron
microscope equipped with an energy dispersive spectrometer (SEM-EDS, Quanta
250 FEG
,
FEI Co., Ltd., Hillsboro, OR, USA). The three-point bending strengths of bar-shaped sam-
ples (measuring 3 mm
×
4 mm
×
36 mm) were measured using a universal material testing
machine (UMT, Z030TE+TEE, ZwickRoell Co., Ltd., Ulm, Germany) with a 20 mm span and
a crosshead speed of 0.5 mm/min. The elastic moduli of plate-shaped samples (measuring
100 mm
×
30 mm
×
10 mm) were evaluated using a non-destructive ultrasonic tester
(GrindoSonic system, IET-01, Luoyang Zhuosheng Testing Instrument Co., Ltd., Luoyang,
China). The thermal diffusivities (
α
) and specific heats (C
p
) of square-shaped samples
(measuring 10 mm
×
10 mm
×
2 mm) were estimated using the laser flash method (LFA
467 Hyperflash, NETZSCH-Gerätebau GmbH, Selb, Germany). Thermal conductivity (
κ
)
was calculated based on the relationship κ=ρ·Cp·α.

Materials 2024,17, 5765 5 of 13
3. Results and Discussion
3.1. Preforms
The densities and open porosities of the three preform samples are presented in Table 1.
The PS preform exhibited the highest density at 1.651 g/cm
3
and also had the highest
apparent porosity of 42.33% among the preforms. The high density of the impregnated
SiC powder, at 3.21 g/cm
3
, contributed to the higher density of the PS preform. Unlike
nano-sized carbon black, it was more challenging to impregnate submicron-sized SiC
particles into the preforms. As a result, the PS preform also had the highest porosity.
Furthermore, although both the PP and PC preforms were composed of SiC fiber and C, the
PP preform had a lower density of 1.217 g/cm
3
compared to the PC preform’s 1.428 g/cm
3
.
Additionally, the open porosity of the PP preform was 35.17%, which was lower than the
PC preform’s 36.39%. This difference is primarily attributed to the fact that phenolic resin
tends to generate cracks and closed pores during pyrolysis [
30
]. As the number of PIP
cycles increased, the number of closed pores within the preforms also increased.
Table 1. The densities and apparent porosities of the green preforms.
Preform Density (g/cm3)Open Porosity (%)
PP 1.217 35.17
PS 1.651 42.33
PC 1.428 36.39
Figure 2illustrates the microstructures of the PP, PS and PC preforms. As shown in
Figure 2a,d, a significant number of pores were observed on the cross-section of the PP
preform, attributed to cracks or large pores formed due to the substantial shrinkage of
the matrix during the pyrolysis of the phenolic resin. These pores likely include some
closed pores that became exposed during the cutting process for sample preparation. The
carbon produced by the pyrolysis of phenolic resin was primarily distributed within the
SiC fiber bundles, with minimal carbon content present between the bundles, leading to
pore formation. The woven structure of the SiC fiber cloth facilitated the development
of closed pores and interlayer cracks between the fiber bundles. Additionally, shrinkage
cracks and structural collapse from the pyrolysis of phenolic resin further contributed to
the formation of closed pores. Conversely, the carbon matrix within the SiC fiber bundles
remained largely intact, featuring a relatively thicker carbon layer, which could hinder
subsequent molten silicon infiltration [31].
Materials 2024, 17, x FOR PEER REVIEW 6 of 14
Figure 2. The cross-sectional SEM micrographs of the preforms: low-magnification images of (a) PP,
(b) PS preforms, and (c) PC; high-magnification images of (d) PP, (e) PS, and (f) PC preforms.
Figure 2(b) and 2(e) depict that the the interstitial spaces between the fiber bundles
of PS preform are filled with a significant amount of SiC powder, effectively reducing
porosity. These chemically inert SiC powders did not participate in the reactive melt
infiltration process and may have decreased the residual silicon content post RMI. The
embedded SiC powders within the fiber bundles create ideal channels for molten silicon
infiltration. For the PS preform, the phenolic resin pyrolysis process was conducted only
once, resulting in a thinner carbon matrix within the fiber bundles compared to the PP
preform. However, large pores were still observed within the fiber bundle. The driving
force for molten silicon infiltration into the preform stems from capillary action within
these pores; smaller pore sizes indicate stronger capillary forces and enhanced infiltration
dynamics [16,23,29]. This mechanism can partially mitigate pore blockage caused by the
silicon/carbon reaction. Notably, capillary action within the SiC fiber bundles was more
pronounced than between the bundles, ensuring deep penetration of silicon into the
preform’s interior and achieving complete infiltration.
The microstructure of the PC preform is demonstrated in Figure 2(c) and (f ). The
carbon within the fiber bundles primarily originates from the pyrolysis of phenolic resin,
while the interstitial carbon between the bundles mainly comes from impregnated carbon
black. The pores within the fiber bundles are extremely small and not visible to the naked
eye. Besides, most adjacent SiC fibers are interconnected by a continuous carbon
framework formed during the pyrolysis of the phenolic resin. This thermally decomposed
carbon framework, along with the impregnated carbon black, would provide an adequate
supply of carbon necessary for the subsequent reaction to form SiC during the RMI
process.
The pore size distributions of the preforms are illustrated in Figure 3. The pores in
the preforms are primarily categorized into micrometer-scale and nanometer-scale pores.
Due to the fabrication process involving stacking two-dimensional satin SiC fiber cloth,
all preforms inherently contain micrometer-scale pores that are difficult to completely fill.
Additionally, all preforms exhibit a multimodal pore size distribution: the PP preform
shows peak pore diameters around 5 µm; the PS preform has pores ranging from 200 nm
to 4 µm; and the PC preform features pores approximately 150 nm in size.
Figure 2. The cross-sectional SEM micrographs of the preforms: low-magnification images of (a) PP,
(b) PS preforms, and (c) PC; high-magnification images of (d) PP, (e) PS, and (f) PC preforms.

Materials 2024,17, 5765 6 of 13
Figure 2b,e depict that the the interstitial spaces between the fiber bundles of PS
preform are filled with a significant amount of SiC powder, effectively reducing porosity.
These chemically inert SiC powders did not participate in the reactive melt infiltration
process and may have decreased the residual silicon content post RMI. The embedded SiC
powders within the fiber bundles create ideal channels for molten silicon infiltration. For
the PS preform, the phenolic resin pyrolysis process was conducted only once, resulting in
a thinner carbon matrix within the fiber bundles compared to the PP preform. However,
large pores were still observed within the fiber bundle. The driving force for molten silicon
infiltration into the preform stems from capillary action within these pores; smaller pore
sizes indicate stronger capillary forces and enhanced infiltration dynamics [
16
,
23
,
29
]. This
mechanism can partially mitigate pore blockage caused by the silicon/carbon reaction.
Notably, capillary action within the SiC fiber bundles was more pronounced than between
the bundles, ensuring deep penetration of silicon into the preform’s interior and achieving
complete infiltration.
The microstructure of the PC preform is demonstrated in Figure 2c,f. The carbon
within the fiber bundles primarily originates from the pyrolysis of phenolic resin, while the
interstitial carbon between the bundles mainly comes from impregnated carbon black. The
pores within the fiber bundles are extremely small and not visible to the naked eye. Besides,
most adjacent SiC fibers are interconnected by a continuous carbon framework formed
during the pyrolysis of the phenolic resin. This thermally decomposed carbon framework,
along with the impregnated carbon black, would provide an adequate supply of carbon
necessary for the subsequent reaction to form SiC during the RMI process.
The pore size distributions of the preforms are illustrated in Figure 3. The pores in
the preforms are primarily categorized into micrometer-scale and nanometer-scale pores.
Due to the fabrication process involving stacking two-dimensional satin SiC fiber cloth,
all preforms inherently contain micrometer-scale pores that are difficult to completely fill.
Additionally, all preforms exhibit a multimodal pore size distribution: the PP preform
shows peak pore diameters around 5
µ
m; the PS preform has pores ranging from 200 nm to
4µm; and the PC preform features pores approximately 150 nm in size.
Materials 2024, 17, x FOR PEER REVIEW 7 of 14
Figure 3. The pore size distributions of the green preforms: (a) PP, (b) PS, and (c) PC.
The relatively large pores in the PP preform result from the twice-repeated impreg-
nation and pyrolysis of the phenolic resin. For the PS preform, the pyrolysis of the phe-
nolic resin generated large pores, while some of the impregnated SiC powder inltrated
these pores, creating a gradient of pore sizes with smaller pores forming between the ber
bundles. In contrast, for the PC preform, although the pyrolysis of the phenolic resin pro-
duced large pores, subsequent inltration by carbon black eectively lled these pores,
resulting in dense packing and leaving only small pores [16].
3.2. Composite Materials
The XRD paerns of the SiCf/SiC composites are presented in Figure 4. For the PP
composite, the predominant phase was identied as β-SiC, with minor peaks correspond-
ing to silicon. Although diraction peaks of carbon were not identied, its existence could
not be excluded due to the amorphous state of carbon black. The formation of β-SiC dur-
ing the RMI process can be aributed to the following reaction:
Si(l) + C(s) → β-SiC(s)
(1)
Figure 4. XRD paerns of the nal SiCf/SiC composite samples.
The PS composite primarily consisted of β-SiC, accompanied by small amounts of
both silicon and α-SiC, which were introduced during the slurry inltration process. In
contrast, the PC composite exhibited only diraction peaks associated with β-SiC and a
tiny amount of residual silicon. Among all the samples, the PC composite demonstrated
Figure 3. The pore size distributions of the green preforms: (a) PP, (b) PS, and (c) PC.
The relatively large pores in the PP preform result from the twice-repeated impregna-
tion and pyrolysis of the phenolic resin. For the PS preform, the pyrolysis of the phenolic
resin generated large pores, while some of the impregnated SiC powder infiltrated these
pores, creating a gradient of pore sizes with smaller pores forming between the fiber bun-
dles. In contrast, for the PC preform, although the pyrolysis of the phenolic resin produced
large pores, subsequent infiltration by carbon black effectively filled these pores, resulting
in dense packing and leaving only small pores [16].
3.2. Composite Materials
The XRD patterns of the SiC
f
/SiC composites are presented in Figure 4. For the PP
composite, the predominant phase was identified as
β
-SiC, with minor peaks corresponding
to silicon. Although diffraction peaks of carbon were not identified, its existence could not

Materials 2024,17, 5765 7 of 13
be excluded due to the amorphous state of carbon black. The formation of
β
-SiC during
the RMI process can be attributed to the following reaction:
Si(l) + C(s)→β-SiC(s) (1)
Materials 2024, 17, x FOR PEER REVIEW 7 of 14
Figure 3. The pore size distributions of the green preforms: (a) PP, (b) PS, and (c) PC.
The relatively large pores in the PP preform result from the twice-repeated impreg-
nation and pyrolysis of the phenolic resin. For the PS preform, the pyrolysis of the phe-
nolic resin generated large pores, while some of the impregnated SiC powder inltrated
these pores, creating a gradient of pore sizes with smaller pores forming between the ber
bundles. In contrast, for the PC preform, although the pyrolysis of the phenolic resin pro-
duced large pores, subsequent inltration by carbon black eectively lled these pores,
resulting in dense packing and leaving only small pores [16].
3.2. Composite Materials
The XRD paerns of the SiCf/SiC composites are presented in Figure 4. For the PP
composite, the predominant phase was identied as β-SiC, with minor peaks correspond-
ing to silicon. Although diraction peaks of carbon were not identied, its existence could
not be excluded due to the amorphous state of carbon black. The formation of β-SiC dur-
ing the RMI process can be aributed to the following reaction:
Si(l) + C(s) → β-SiC(s)
(1)
Figure 4. XRD paerns of the nal SiCf/SiC composite samples.
The PS composite primarily consisted of β-SiC, accompanied by small amounts of
both silicon and α-SiC, which were introduced during the slurry inltration process. In
contrast, the PC composite exhibited only diraction peaks associated with β-SiC and a
tiny amount of residual silicon. Among all the samples, the PC composite demonstrated
Figure 4. XRD patterns of the final SiCf/SiC composite samples.
The PS composite primarily consisted of
β
-SiC, accompanied by small amounts of
both silicon and
α
-SiC, which were introduced during the slurry infiltration process. In
contrast, the PC composite exhibited only diffraction peaks associated with
β
-SiC and a
tiny amount of residual silicon. Among all the samples, the PC composite demonstrated the
highest proportion of
β
-SiC and the lowest amount of residual Si, highlighting its optimal
composition for RMI-SiCf/SiC composites [32].
The microstructures of the SiC
f
/SiC composites are illustrated in Figure 5. As shown
in Figure 5a, a distinct transition region was observed between the SiC fibers and the matrix
in the PP composite sample, primarily composed of
β
-SiC and Si, as confirmed by previous
XRD results. Figure 5a also indicates that in certain areas, the SiC fibers and the matrix were
indistinguishable, suggesting the formation of a high-density
β
-SiC matrix. Figure 5b,c
display the distribution of C and Si elements based on EDS analysis, indicating the presence
of residual silicon and carbon. Notably, a thin layer of SiC had formed at the C/Si interface,
likely inhibiting further reactions between Si and C through spatial hindrance effects. This
phenomenon can be attributed to the composition and pore structure of the PP preform.
The carbon from phenolic resin pyrolysis formed a continuous C matrix with large pores
up to 5
µ
m within the fiber bundles. The significant contact area for the Si/C reaction
led to the formation of a continuous SiC layer, which in turn hindered further interaction
between Si and C due to the low solubility of C in SiC and the slow reaction progression [
33
].
However, the PP preform’s pore structure facilitated the infiltration of molten silicon. The
silicon melted, filling the larger pores before entering smaller ones through capillary action.
The smaller pores generated higher capillary forces, aiding the entry of the melt into the
preform’s interior. This partly offset the adverse effects of pore blockage caused by the
Si/C reaction. Although the Si/C reaction occurred early at the interfaces during the RMI
process, leading to matrix volume expansion, the molten silicon was unlikely to backflow
and continued to infiltrate the preform’s interior [
29
]. Additionally, the larger pores allowed
for an increased amount of residual silicon after the RMI process.

Materials 2024,17, 5765 8 of 13
Materials 2024, 17, x FOR PEER REVIEW 8 of 14
the highest proportion of β-SiC and the lowest amount of residual Si, highlighting its
optimal composition for RMI-SiCf/SiC composites [32].
The microstructures of the SiCf/SiC composites are illustrated in Figure 5. As shown
in Figure 5(a), a distinct transition region was observed between the SiC fibers and the
matrix in the PP composite sample, primarily composed of β-SiC and Si, as confirmed by
previous XRD results. Figure 5(a) also indicates that in certain areas, the SiC fibers and the
matrix were indistinguishable, suggesting the formation of a high-density β-SiC matrix.
Figure 5(b) and 5(c) display the distribution of C and Si elements based on EDS analysis,
indicating the presence of residual silicon and carbon. Notably, a thin layer of SiC had
formed at the C/Si interface, likely inhibiting further reactions between Si and C through
spatial hindrance effects. This phenomenon can be aributed to the composition and pore
structure of the PP preform. The carbon from phenolic resin pyrolysis formed a
continuous C matrix with large pores up to 5 µm within the fiber bundles. The significant
contact area for the Si/C reaction led to the formation of a continuous SiC layer, which in
turn hindered further interaction between Si and C due to the low solubility of C in SiC
and the slow reaction progression [33]. However, the PP preform’s pore structure
facilitated the infiltration of molten silicon. The silicon melted, filling the larger pores
before entering smaller ones through capillary action. The smaller pores generated higher
capillary forces, aiding the entry of the melt into the preform’s interior. This partly offset
the adverse effects of pore blockage caused by the Si/C reaction. Although the Si/C
reaction occurred early at the interfaces during the RMI process, leading to matrix volume
expansion, the molten silicon was unlikely to backflow and continued to infiltrate the
preform’s interior [29]. Additionally, the larger pores allowed for an increased amount of
residual silicon after the RMI process.
Figure 5. Cross-sectional SEM images and the corresponding EDS mapping of SiCf/SiC composites:
(a-c) Sample PP, (d-f) Sample PS, (g-i) Sample PC.
In the case of the PS composite, as presented in Figure 5(d), a higher proportion of β-
SiC was formed, and there was no distinct boundary between Si and SiC in the matrix,
indicating a dispersed distribution. Figure 5(e) and 5(f) show the distribution of C and Si
elements. The smaller pore size of the PS preform compared to the PP preform facilitated
a more thorough RMI process. Furthermore, the inter-fiber bundles were uniformly filled
Figure 5. Cross-sectional SEM images and the corresponding EDS mapping of SiC
f
/SiC composites:
(a–c) Sample PP, (d–f) Sample PS, (g–i) Sample PC.
In the case of the PS composite, as presented in Figure 5d, a higher proportion of
β
-SiC was formed, and there was no distinct boundary between Si and SiC in the matrix,
indicating a dispersed distribution. Figure 5e,f show the distribution of C and Si elements.
The smaller pore size of the PS preform compared to the PP preform facilitated a more
thorough RMI process. Furthermore, the inter-fiber bundles were uniformly filled with
SiC powder in the preform, leading to a dispersed distribution of residual Si within the
SiC powder in the composite [
34
]. The presence of residual carbon in the PS composite
was primarily due to the similar morphology of carbon produced from the pyrolysis of the
phenolic resin, resembling that found in the PP preform. As previously discussed, the peak
pore size of the PS preform ranged from 200 nm to 4
µ
m. Some pores within the powder
were smaller than those within the SiC fiber bundles, creating regions where molten Si had
difficulty penetrating, resulting in incomplete Si/C reactions.
In the PC composite, as shown in Figure 5g, over 90% of the observed area displayed
an indistinguishable boundary between the fibers and the matrix, indicating a highly ideal
microstructure. Figure 5h,i illustrate the distribution of C and Si elements, respectively. The
dark regions in Figure 5h predominantly consist of residual carbon. The extensive presence
of Si elements in Figure 5i signifies that the Si/C reaction during the RMI process was
thorough. The PC preform exhibited the smallest internal pores among the three samples.
A smaller pore structure generates higher capillary forces, facilitating a faster and more
complete melt infiltration process. The variations in the red’s brightness, indicating differ-
ences in Si content, suggest uneven distribution of Si elements between SiC and residual Si
regions. Residual carbon may be present in areas encapsulated by SiC, preventing further
reaction with Si. The chemical stability of SiC hinders the dissolution of excess carbon
into the silicon melt. Additionally, gaps between SiC fiber bundles were filled with SiC,
allowing carbon to dissolve in molten silicon rather than forming an inert layer within the
fiber bundles. These conditions resulted in a thicker SiC layer compared to the other two
samples, contributing to a reduction in residual carbon content.
Table 2presents the physical properties of the SiC
f
/SiC composite materials. The
open porosities of the three samples ranged from 1% to 2%. Ideally, the RMI process
filled residual pores with molten silicon, resulting in zero open porosity. However, factors

Materials 2024,17, 5765 9 of 13
such as silicon shrinkage during solidification and measurement inaccuracies typically
led to an open porosity below 2%. The highest density measured was 2.80 g/cm
3
in the
PC composite sample, which was expected to exhibit better comprehensive properties
than the other samples. Under ideal circumstances, RMI-derived SiC
f
/SiC composites
consist of SiC fibers, an interfacial layer, and a SiC ceramic matrix, yielding a theoretical
density of approximately 3.21 g/cm
3
. Nevertheless, the practical density fell short of this
ideal value due to closed pores formed during the SiC
f
/C preform preparation and the
RMI process, as well as the presence of unreacted residual carbon and silicon [
35
]. The
PC composite achieved the highest density and lowest porosity, demonstrating that the
technical approach of first impregnating with phenolic resin followed by carbon black
slurry enables the production of an ideal preformed composite and pore size. These
characteristics are crucial prerequisites for subsequent RMI processes aimed at achieving
better performance of SiC
f
/SiC composites. The elastic modulus of the PC composite
reached a maximum value of 181.1
±
0.1 GPa, which was higher than the
176.3 ±0.1 GPa
of the PS composite and the 145.0
±
0.1 GPa of the PP composite. The low elastic modulus
of the PP composite was attributed to the abundant closed pores within the preform, which
inherently possessed zero elastic modulus. For the PS and PP preforms, during the RMI
process, the generated SiC was insufficient to fully fill these pores, leaving residual Si to
occupy the remaining spaces, along with abundant residual carbon. These pores, along with
the residual Si and C, detrimentally reduced the density, strength, and elastic modulus of
the composite. The moderate elastic modulus of the PS composite arose from the formation
of a relatively thick Si-SiC layer during the fabrication process, which combined with the
in situ generated SiC to form a dense SiC structure. The residual Si and C within the
PS composite also contributed to a decrease in its elastic modulus. The PC composite
preparation route was proved to be more effective in this study.
Table 2. The densification results and physical properties of the SiCf/SiC composites.
Composites
Bulk Density
(g/cm3)
Open
Porosity
(%)
Flexural
Strength
(MPa)
Elastic
Modulus
(GPa)
Thermal
Conductivity
(W/mK)
PP 2.73 1.64 135.7 ±15.5 145.0 ±0.1 21.3
PS 2.75 1.96 142.0 ±7.6 176.3 ±0.1 24.1
PC 2.80 1.50 152.4 ±15.4 181.1 ±0.1 27.7
Table 2presents the densification results and physical properties of the final SiC
f
/SiC
composites. The densities of the PP, PS, and PC samples were 2.73 g/cm
3
, 2.75 g/cm
3
, and
2.80 g/cm
3
, respectively, with open porosities ranging from 1% to 2% for all the samples,
indicating a high densification degree. Typically, the RMI process fills residual pores with
molten silicon, resulting in zero open porosity. However, factors such as silicon shrinkage
during solidification typically lead to an open porosity below 2%. The intrinsic density of
SiC is 3.21 g/cm
3
. Nevertheless, the actual densities of the composite samples were lower
than this intrinsic value, primarily due to the presence of internal pores, residual silicon,
and residual carbon [
35
]. The highest density and lowest porosity of the PC composite
demonstrate that the composition and pore size of the PC preform are most favorable for
preparing ideal RMI-SiCf/SiC composites.
Figure 6illustrates the bending stress-strain curves of three SiC
f
/SiC composite sam-
ples. The curves can be divided into three regions: an initial linear increase, followed by a
nonlinear rise, and finally a nonlinear decrease. These features indicate that all compos-
ite samples exhibit typical non-brittle failure behavior. However, there were significant
differences in the fracture behavior among the three samples. The highest points of the
stress-strain curves, representing the flexural strength, showed significant variation among
the composite samples. The flexural strengths, ranked from lowest to highest, were as
follows: PP, PS, and PC. This order corresponds to the flexural strength values listed in
Table 2: 135.7
±
15.5 MPa for PP, 142.0
±
7.6 MPa for PS, and 152.4
±
15.4 MPa for PC.

Materials 2024,17, 5765 10 of 13
Additionally, the area projected to the x-axis by the stress-strain curves represents the work
of fracture; a larger projected area indicates better toughness. The order of the projected
areas, from smallest to largest, was PP, PS, and PC.
Materials 2024, 17, x FOR PEER REVIEW 10 of 14
indicating a high densication degree. Typically, the RMI process lls residual pores with
molten silicon, resulting in zero open porosity. However, factors such as silicon shrinkage
during solidication typically lead to an open porosity below 2%. The intrinsic density of
SiC is 3.21 g/cm3. Nevertheless, the actual densities of the composite samples were lower
than this intrinsic value, primarily due to the presence of internal pores, residual silicon,
and residual carbon [35]. The highest density and lowest porosity of the PC composite
demonstrate that the composition and pore size of the PC preform are most favorable for
preparing ideal RMI-SiCf/SiC composites.
Table 2. The densication results and physical properties of the SiCf/SiC composites.
Compo-
sites
Bulk den-
sity
(g/cm3)
Open poros-
ity
(%)
Flexural
strength
(MPa)
Elastic modu-
lus
(GPa)
Thermal Conduc-
tivity
(W/mK)
PP
2.73
1.64
135.7 ± 15.5
145.0 ± 0.1
21.3
PS
2.75
1.96
142.0 ± 7.6
176.3 ± 0.1
24.1
PC
2.80
1.50
152.4 ± 15.4
181.1 ± 0.1
27.7
Figure 6 illustrates the bending stress-strain curves of three SiCf/SiC composite sam-
ples. The curves can be divided into three regions: an initial linear increase, followed by a
nonlinear rise, and nally a nonlinear decrease. These features indicate that all composite
samples exhibit typical non-brile failure behavior. However, there were signicant dif-
ferences in the fracture behavior among the three samples. The highest points of the stress-
strain curves, representing the exural strength, showed signicant variation among the
composite samples. The exural strengths, ranked from lowest to highest, were as follows:
PP, PS, and PC. This order corresponds to the exural strength values listed in Table 2:
135.7 ± 15.5 MPa for PP, 142.0 ± 7.6 MPa for PS, and 152.4 ± 15.4 MPa for PC. Additionally,
the area projected to the x-axis by the stress-strain curves represents the work of fracture;
a larger projected area indicates beer toughness. The order of the projected areas, from
smallest to largest, was PP, PS, and PC.
Figure 6. The exural stress-strain curves of the PP, PS, and PC composite samples.
To verify these results, typical fracture surfaces of SiCf/SiC composite samples were
observed and compared using SEM, as shown in Figure 7. The fracture surface of the PP
sample (Figure 7a) was very planar with only a few short pullout bers. The number of
pullout bers slightly increased in the PS sample (Figure 7b). In contrast, numerous long
pullout bers were observed in the PC sample (Figure 7c). Fiber pullout is a primary
mechanism for toughening, and a higher number of ber pullouts indicates a more pro-
nounced tough fracture behavior in the SiCf/SiC composites, consistent with the results
presented in Figure 6. Generally, a weaker ber/matrix interface promotes ber pullout
[36,37]. In this study, the bonding strength at the ber/matrix interface of the SiCf/SiC
composite samples likely varied due to residual silicon content. Specically, the PP sample
Figure 6. The flexural stress-strain curves of the PP, PS, and PC composite samples.
To verify these results, typical fracture surfaces of SiC
f
/SiC composite samples were
observed and compared using SEM, as shown in Figure 7. The fracture surface of the PP
sample (Figure 7a) was very planar with only a few short pullout fibers. The number of
pullout fibers slightly increased in the PS sample (Figure 7b). In contrast, numerous long
pullout fibers were observed in the PC sample (Figure 7c). Fiber pullout is a primary mech-
anism for toughening, and a higher number of fiber pullouts indicates a more pronounced
tough fracture behavior in the SiC
f
/SiC composites, consistent with the results presented in
Figure 6. Generally, a weaker fiber/matrix interface promotes fiber pullout [
36
,
37
]. In this
study, the bonding strength at the fiber/matrix interface of the SiC
f
/SiC composite samples
likely varied due to residual silicon content. Specifically, the PP sample had the highest
residual silicon content, possibly allowing molten silicon to penetrate the SiC interlayer
and erode the PyC interlayer during the reactive melt infiltration, thereby enhancing the
fiber/matrix interface bond and hindering fiber pullout. Conversely, the PC sample had the
lowest residual silicon content, preventing molten silicon from breaching the SiC interfacial
layer and preserving the PyC interlayer, ensuring a weaker fiber/matrix interface bond
and facilitating fiber pullout.
Materials 2024, 17, x FOR PEER REVIEW 11 of 14
had the highest residual silicon content, possibly allowing molten silicon to penetrate the
SiC interlayer and erode the PyC interlayer during the reactive melt inltration, thereby
enhancing the ber/matrix interface bond and hindering ber pullout. Conversely, the PC
sample had the lowest residual silicon content, preventing molten silicon from breaching
the SiC interfacial layer and preserving the PyC interlayer, ensuring a weaker ber/matrix
interface bond and facilitating ber pullout.
Figure 7. The SEM micrographs of typical fracture surfaces of SiCf/SiC composite samples: (a) PP,
(b) PS, and (c) PC.
The elastic moduli of the PP, PS, and PC samples were 145.0 ± 0.1 GPa, 176.3 ± 0.1
GPa, and 181.1 ± 0.1 GPa, respectively. Since the elastic moduli of silicon and carbon are
signicantly lower than that of SiC, a reduction in residual silicon and carbon content
eectively enhances the elastic modulus of the SiCf/SiC composites.
The thermal conductivities of the SiCf/SiC composites are also presented in Table 2.
The thermal conductivity values for the PP, PS, and PC samples were 21.3 W/mK, 24.1
W/mK, and 27.7 W/mK, respectively [38-40]. Due to the strong covalent bonding charac-
teristics of Si-C, the primary mechanism of thermal conduction in SiCf/SiC composites is
phonon transport. The dierences in thermal conductivity among the samples can be at-
tributed to two main factors:
(1) According to the formula for thermal conductivity, (κ = ρCpα), a higher density (ρ)
contributes to a higher thermal conductivity. The order of thermal conductivity for
the PP, PS, and PC samples aligned with their respective densities.
(2) There were variations in the matrix composition among the dierent composite sam-
ples, specically signicant dierences in the contents of residual silicon and carbon,
as indicated by XRD and SEM results. As reported in previous studies, the intrinsic
thermal conductivities of SiC single crystals (both β- and α-SiC), silicon single crys-
tals, and carbon black are 490 W/mK, 156 W/mK, and 2 W/mK, respectively. Thus,
composite samples with higher contents of residual silicon and carbon black exhib-
ited lower thermal conductivity.
Additionally, SiCf/SiC composite samples were primarily composed of SiC bers, -
ber/matrix interfaces, and ceramic matrix. The ceramic matrix included β-SiC (formed
through inltration reactions), α-SiC (present only in PS samples), residual silicon, resid-
ual carbon, and a small amount of porosity. Consequently, SiCf/SiC composite samples do
not exhibit an ideal single-crystal structure but contain numerous defects that cause pho-
non scaering. These defects include ber/matrix interfaces, as well as point defects, dis-
locations, stacking faults, grain boundaries, and pores within the matrix. They disrupt the
ecient transfer of phonons, leading to signicantly lower thermal conductivity than the
intrinsic value of SiC single crystals.
4. Conclusions
This study demonstrates that high-performance SiCf/SiC composites can be achieved
using a straightforward and cost-eective RMI method by carefully controlling the pre-
formed composition and pore size. The ndings highlight the critical role of preform de-
sign in achieving optimized mechanical and thermal properties.
Figure 7. The SEM micrographs of typical fracture surfaces of SiC
f
/SiC composite samples: (a) PP,
(b) PS, and (c) PC.
The elastic moduli of the PP, PS, and PC samples were 145.0
±
0.1 GPa,
176.3 ±0.1 GPa
,
and 181.1
±
0.1 GPa, respectively. Since the elastic moduli of silicon and carbon are
significantly lower than that of SiC, a reduction in residual silicon and carbon content
effectively enhances the elastic modulus of the SiCf/SiC composites.
The thermal conductivities of the SiC
f
/SiC composites are also presented in
Table 2
.
The thermal conductivity values for the PP, PS, and PC samples were 21.3 W/mK,
24.1 W/mK
,
and 27.7 W/mK, respectively [
38
–
40
]. Due to the strong covalent bonding characteristics
of Si-C, the primary mechanism of thermal conduction in SiC
f
/SiC composites is phonon

Materials 2024,17, 5765 11 of 13
transport. The differences in thermal conductivity among the samples can be attributed to
two main factors:
(1)
According to the formula for thermal conductivity, (
κ
=
ρ
C
pα
), a higher density (
ρ
)
contributes to a higher thermal conductivity. The order of thermal conductivity for
the PP, PS, and PC samples aligned with their respective densities.
(2)
There were variations in the matrix composition among the different composite
samples, specifically significant differences in the contents of residual silicon and
carbon, as indicated by XRD and SEM results. As reported in previous studies, the
intrinsic thermal conductivities of SiC single crystals (both
β
- and
α
-SiC), silicon single
crystals, and carbon black are 490 W/mK, 156 W/mK, and 2 W/mK, respectively.
Thus, composite samples with higher contents of residual silicon and carbon black
exhibited lower thermal conductivity.
Additionally, SiC
f
/SiC composite samples were primarily composed of SiC fibers,
fiber/matrix interfaces, and ceramic matrix. The ceramic matrix included
β
-SiC (formed
through infiltration reactions),
α
-SiC (present only in PS samples), residual silicon, resid-
ual carbon, and a small amount of porosity. Consequently, SiC
f
/SiC composite samples
do not exhibit an ideal single-crystal structure but contain numerous defects that cause
phonon scattering. These defects include fiber/matrix interfaces, as well as point defects,
dislocations, stacking faults, grain boundaries, and pores within the matrix. They disrupt
the efficient transfer of phonons, leading to significantly lower thermal conductivity than
the intrinsic value of SiC single crystals.
4. Conclusions
This study demonstrates that high-performance SiC
f
/SiC composites can be achieved
using a straightforward and cost-effective RMI method by carefully controlling the pre-
formed composition and pore size. The findings highlight the critical role of preform design
in achieving optimized mechanical and thermal properties.
(1)
Optimized Preform: The PC preform, prepared through carbon black slurry impreg-
nation, contained an adequate amount of carbon and exhibited an ideal pore size
distribution. These characteristics facilitated a more efficient RMI process, resulting in
a composite with high density, low porosity, and minimal residual silicon and carbon.
This significantly enhanced the mechanical properties of the composite.
(2)
Enhanced Properties: Among all the samples tested, the PC composite demonstrated
the highest flexural strength (152.4
±
15.4 MPa), elastic modulus (181.1
±
0.1 GPa),
and thermal conductivity (27.7 W/mK). These exceptional properties are crucial for
applications requiring high performance under extreme conditions.
(3) Cost-Effective Approach: The methodology developed in this study provides a simple
and cost-effective approach for preparing high-performance SiC
f
/SiC composites. By
optimizing the preformed composition and pore size, the RMI process was signifi-
cantly improved, leading to better performance in SiCf/SiC composites.
Author Contributions: Conceptualization, Y.L.; methodology, Q.R.; software, Z.Y.; validation, Q.H.
and Y.L.; formal analysis, H.N. and Q.R.; investigation, H.N.; resources, Q.H. and Y.L.; data curation,
H.N.; writing—original draft preparation, H.N.; writing—review and editing, P.S., Q.R., T.C. and Y.L.;
visualization, H.N.; supervision, P.S. and Y.L.; project administration, Y.L.; funding acquisition, Z.H.,
Q.H. and Y.L. All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by the National Natural Science Foundation of China (Grant no.
52302077), the Key Research and Development Program of Ningbo (Grant no. 2022Z084), the 3315
Innovation Team of Ningbo (Grant no. 2018A-03-A), the Strategic Priority Research Program of the
Chinese Academy of Sciences (Grant no. XDA041030301), the Top-talent Team Program of Ningbo,
Advanced Energy Science and Technology Guangdong Laboratory (Grant no. HND20TDTHGC00),
and the Hundred Talents Program of the Chinese Academy of Sciences.
Institutional Review Board Statement: Not applicable.

Materials 2024,17, 5765 12 of 13
Informed Consent Statement: Not applicable.
Data Availability Statement: The data presented in this study are available on request from the
corresponding author.
Conflicts of Interest: The authors declare no conflicts of interest.
References
1.
Wang, P.; Liu, F.; Wang, H.; Li, H.; Gou, Y. A review of third generation SiC fibers and SiC
f
/SiC composites. J. Mater. Sci. Technol.
2019,35, 2743–2750. [CrossRef]
2.
Katoh, Y.; Snead, L.; Henager, C.; Nozawa, T.; Hinoki, T.; Ivekovic, A.; Novak, S.; Vicente, S. Current status and recent research
achievements in SiC/SiC composites. J. Nucl. Mater. 2014,455, 387–397. [CrossRef]
3.
Masaki, K.T.; Akira, K.; Kiyohito, O.; Yutai, K. Consolidation of polymer-derived SiC matrix composites: Processing and
microstructure. Compos. Sci. Technol. 2002,62, 2179–2188. [CrossRef]
4.
Kiyoshi, S.; Atsushi, T.; Osamu, F.; Takeshi, I.; Yoshiharu, T.; Shinji, K.; Misao, I. Fabrication and pressure testing of a gas-turbine
component manufactured by a preceramic-polymer-impregnation method. Compos. Sci. Technol. 1999,59, 853–859. [CrossRef]
5. Brewer, D. HSR/EPM combustor materials development program. Mater. Sci. Eng. A 1999,26, 284–291. [CrossRef]
6.
Naslain, R. Design, preparation and properties of non-oxide CMCs for application in engines and nuclear reactors: An overview.
Compos. Sci. Technol. 2004,64, 155–170. [CrossRef]
7.
Kristine, B.; Shannon, B. Advanced LWR Nuclear Fuel Cladding System Development Trade-Off Study; Idaho National Laboratory:
Idaho Falls, ID, USA, 2012. [CrossRef]
8.
Zinkle, S.J.; Terrani, K.A.; Gehin, J.C.; Ott, L.J.; Snead, L.L. Accident tolerant fuels for LWRs: A perspective. J. Nucl. Mater. 2014,
448, 374–379. [CrossRef]
9.
Tao, P.; Wang, Y. Improved thermal conductivity of silicon carbide fibers-reinforced silicon carbide matrix composites by chemical
vapor infiltration method. Ceram. Int. 2019,45, 2207–2212. [CrossRef]
10.
Bhatt, R.T.; Kiser, J.D. Creep behavior and failure mechanisms of CVI and PIP SiC/SiC composites at temperatures to 1650
◦
C in
air. J. Eur. Ceram. Soc. 2021,41, 6196–6206. [CrossRef]
11.
Gao, Y.; Zhou, H.; Liu, M.; Ye, L.; Xia, H. Mechanical and thermal properties of chemical vapor infiltration engineered 2D-woven
and 3D-braided carbon silicate composites. Ceram. Int. 2015,41, 10949–10956. [CrossRef]
12.
Gu, J.; Lee, S.; Kim, D.; Lee, H.; Kim, J. Improved thermal stability of SiC
f
/SiC ceramic matrix composites fabricated by PIP
process. Process. Appl. Ceram. 2021,15, 164–169. [CrossRef]
13.
Yu, P.; Lin, Z.; Yu, J. Mechanical, thermal, and dielectric properties of SiC
f
/SiC composites reinforced with electrospun SiC fibers
by PIP. J. Eur. Ceram. Soc. 2021,41, 6859–6868. [CrossRef]
14.
Liu, R.; Wang, F.; Zhang, J.; Chen, J.; Wan, F.; Wang, Y. Effects of CVI SiC amount and deposition rates on properties of SiC
f
/SiC
composites fabricated by hybrid chemical vapor infiltration (CVI) and precursor infiltration and pyrolysis (PIP) routes. Ceram.
Int. 2021,47, 26971–26977. [CrossRef]
15.
Jiang, Y.; Hu, C.; Liang, B.; Pang, S.; Li, J.; Tang, S. Cyclic ablation resistance at 2300
◦
C of (Hf
0.4
Zr
0.4
Ta
0.2
)B
2
-SiC-Si coating for
C/SiC composites prepared by SiC-assisted reactive infiltration of silicon. Surf. Coat. Technol. 2022,451, 129072. [CrossRef]
16.
Dong, S.; Zhang, J.; Chen, X.; Liao, C.; Guo, F.; Yang, J.; Zhang, X. Optimizing Microstructure and Properties of SiC
f
/SiC
Composites Prepared by Reactive Melt Infiltration. J. Inorg. Mater. 2021,36, 1103–1110.
17.
Yan, C.; Li, Y.; Cheng, H.; Wu, Z.; Liu, F.; Liu, R. Mechanical properties and microstructure of C
f
/ZrC-SiC and C
f
/SiC composites
produced by precursor infiltration and pyrolysis combined with gaseous silicon infiltration. J. Alloys Compd. 2022,908, 164706.
[CrossRef]
18.
Fitriani, P.; Sharma, A.S.; Yoon, D.-H. Effects of sintering additives on the microstructural and mechanical properties of the
ion-irradiated SiCf/SiC. J. Nucl. Mater. 2018,503, 226–234. [CrossRef]
19.
Chen, X.; Cheng, G.; Zhang, J.; Guo, F.; Dong, S. Residual stress variation in SiC
f
/SiC composite during heat treatment and its
effects on mechanical behavior. J. Adv. Ceram. 2020,9, 567–575. [CrossRef]
20.
Shimoda, K.; Park, J.S.; Hinoki, T.; Kohyama, A. Microstructural optimization of high-temperature SiC/SiC composites by NITE
process. J. Nucl. Mater. 2009,386–388, 634–638. [CrossRef]
21.
Suyama, S.; Kameda, T.; Itoh, Y. Development of high-strength reaction-sintered silicon carbide. Diam. Relat. Mater. 2003,12,
1201–1204. [CrossRef]
22.
Lee, S.P.; Yoon, H.K.; Park, J.S.; Katoh, Y.; Kohyama, A.; Kim, D.H. Reaction sintering process of tyranno SA/SiC composites and
their characterization. Fusion Eng. Des. 2002,61–62, 717–722. [CrossRef]
23.
Guo, G.; Cheng, L.F.; Ye, F.; Song, C. Preparation of SiC/SiC composites with high toughness by reaction of the SiC
P
+C
double-cladding layer with liquid silicon. J. Eur. Ceram. Soc. 2023,43, 4277–4289. [CrossRef]
24.
Xiao, C.; Song, Q.; Shen, Q.W.T.X.W. Understanding on interlaminar nano-reinforcement induced mechanical performance
improvement of carbon/carbon composites after silicon infiltration. Compos. Part B Eng. 2022,239, 109946. [CrossRef]
25.
Magnant, J.; Maille, L.; Pailler, R.; Guette, A.; Rebillat, F.; Philippe, E. Carbon fiber/reaction-bonded carbide matrix for composite
materials—Manufacture and characterization. J. Eur. Ceram. Soc. 2012,32, 4497–4505. [CrossRef]

Materials 2024,17, 5765 13 of 13
26.
Zhong, Q.; Zhang, X.; Dong, S.; Yang, J.; Hu, J.; Gao, L.; He, P.; Zhou, H.; Wang, Z.; Ding, Y. Reactive melt infiltrated C
f
/SiC
composites with robust matrix derived from novel engineered pyrolytic carbon structure. Ceram. Int. 2017,43, 5832–5836.
[CrossRef]
27.
Enterría, M.; Figueiredo, J.L. Nanostructured mesoporous carbons: Tuning texture and surface chemistry. Carbon 2016,108,
79–102. [CrossRef]
28.
Zhao, Y.; Xia, H.; Tang, R.; Shi, Z.; Yang, J.; Wang, J. A low cost preparation of C/SiC composites by infiltrating molten Si into
gelcasted pure porous carbon preform. Ceram. Int. 2015,41, 6478–6487. [CrossRef]
29.
Guo, G.; Ye, F.; Cheng, L.; Li, Z.; Zhang, L. A novel porous carbon synthesized to serve in the preparation of highly dense and
high-strength SiC/SiC by reactive melt infiltration. Compos. Part A Appl. Sci. Manuf. 2024,176, 107839. [CrossRef]
30.
Chen, X.; Feng, Q.; Kan, Y.; Ni, D.; Zhou, H.; Gao, L.; Zhang, X.; Ding, Y.; Dong, S. Effects of preform pore structure on infiltration
kinetics and microstructure evolution of RMI-derived C
f
/ZrC-ZrB
2
-SiC composite. J. Eur. Ceram. Soc. 2020,40, 2683–2690.
[CrossRef]
31.
Süß, F.; Schneider, T.; Frieß, M.; Jemmali, R.; Vogel, F.; Klopsch, L.; Koch, D. Combination of PIP and LSI processes for SiC/SiC
ceramic matrix composites. Open Ceram. 2021,5, 100056. [CrossRef]
32.
Yang, L.W.; Liu, H.T.; Cheng, H.F. Processing-temperature dependent micro- and macro-mechanical properties of SiC fiber
reinforced SiC matrix composites. Compos. Part B Eng. 2017,129, 152–161. [CrossRef]
33.
Kong, J.H.; Lee, S.Y.; Son, Y.I.; Kim, D.K. Synergistic reinforcement effects of ZrB
2
on the ultra-high temperature stability of
Cf/SiC composite fabricated by liquid silicon infiltration. J. Eur. Ceram. Soc. 2023,43, 3062–3069. [CrossRef]
34.
Chen, X.; Yin, J.; Huang, L.; Lee, S.H.; Liu, X.; Huang, Z. Microstructural tailoring, mechanical and thermal properties of SiC
composites fabricated by selective laser sintering and reactive melt infiltration. J. Adv. Ceram. 2023,12, 830–847. [CrossRef]
35.
Jiang, J.; Wang, S.; Li, W.; Chen, Z.; Zhu, Y. Preparation of 3D C
f
/ZrC–SiC composites by joint processes of PIP and RMI. Mater.
Sci. Eng. A 2014,607, 334–340. [CrossRef]
36.
Zhang, Y.; Hu, J.; Zhou, L.; Zhang, X.; Yang, J.; Liao, C.; Kan, Y.; Dong, S.; Li, Y. Influence of fiber surface properties of SiC
f
/SiC
composites on the interfacial debonding behavior. J. Eur. Ceram. Soc. 2024,44, 795–801. [CrossRef]
37.
Cao, X.; Yin, X.; Fan, X.; Cheng, L.; Zhang, L. Effect of PyC interphase thickness on mechanical behaviors of SiBC matrix modified
C/SiC composites fabricated by reactive melt infiltration. Carbon 2014,77, 886–895. [CrossRef]
38.
Slack, G.A. Thermal conductivity of pure and impure silicon, silicon carbide, and diamond. J. Appl. Phys. 1964,35, 3460–3466.
[CrossRef]
39.
Glassbrenner, C.J.; Slack, G.A. Thermal conductivity of silicon and germanium from 3 K to the melting point. Phys. Rev. 1964,134,
634–636. [CrossRef]
40.
Sahoo, B.; Patel, P.; Nanda, B. Carbon black: A thermally conductive reinforcement for epoxy based composite. Bull. Mater. Sci.
2024,47, 260–271. [CrossRef]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.