Article

Cite This: ACS Nano 2019, 13, 1547−1554 www.acsnano.org

A Paper-Like Inorganic Thermal Interface

Material Composed of Hierarchically
Structured Graphene/Silicon Carbide
Nanorods
Wen Dai,†,∥,¶ Le Lv,†,¶ Jibao Lu,*,‡ Hao Hou,† Qingwei Yan,† Fakhr E. Alam,†,∥ Yifan Li,§
Xiaoliang Zeng,‡ Jinhong Yu,†,∥ Qiuping Wei,⊥ Xiangfan Xu,# Jianbo Wu,∇ Nan Jiang,*,†,∥
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

Shiyu Du,§ Rong Sun,‡ Jianbin Xu,○ Ching-Ping Wong,◆ and Cheng-Te Lin*,†,∥
†
Key Laboratory of Marine Materials and Related Technologies, Zhejiang Key Laboratory of Marine Materials and Protective Technologies, Ningbo Institute of
Materials Technology and Engineering (NIMTE), ‡Shenzhen Institutes of Advanced Technology, and §Ningbo Institute of Materials Technology and
Downloaded via MONASH UNIV on January 6, 2020 at 10:48:20 (UTC).

Engineering (NIMTE), Chinese Academy of Sciences, Ningbo 315201, PR China

∥
Center of Materials Science and Optoelectronics Engineering, University of Chinese Academy of Sciences, Beijing 100049, PR China
⊥
State Key Laboratory of Powder Metallurgy, School of Materials Science and Engineering, Central South University, Changsha 410083, PR China
#
Center for Phononics and Thermal Energy Science School of Physics Science and Engineering, Tongji University, Shanghai 200092, China
∇
State Key Laboratory of Metal Matrix Composites, School of Materials Science and Engineering, Shanghai Jiao Tong University, Shanghai, 200240, PR China
○
Department of Electronics Engineering, The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong 999077, China
◆
School of Materials Science and Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332, United States

*
S Supporting Information

ABSTRACT: With the increasing integration of devices in electronics

fabrication, there are growing demands for thermal interface materials
(TIMs) with high through-plane thermal conductivity for efficiently
solving thermal management issues. Graphene-based papers consisting of
a layer-by-layer stacked architecture have been commercially used as
lateral heat spreaders; however, they lack in-depth studies on their TIM
applications due to the low through-plane thermal conductivity (<6 W
m−1 K−1). In this study, a graphene hybrid paper (GHP) was fabricated by
the intercalation of silicon source and the in situ growth of SiC nanorods
between graphene sheets based on the carbothermal reduction reaction.
Due to the formation of covalent C−Si bonding at the graphene−SiC
interface, the GHP possesses a superior through-plane thermal conductivity of 10.9 W m−1 K−1 and can be up to 17.6 W m−1 K−1 under
packaging conditions at 75 psi. Compared with the current graphene-based papers, our GHP has the highest through-plane thermal
conductivity value. In the TIM performance test, the cooling efficiency of the GHP achieves significant improvement compared to that of
state-of-the-art thermal pads. Our GHP with characteristic structure is of great promise as an inorganic TIM for the highly efficient
removal of heat from electronic devices.
KEYWORDS: graphene hybrid paper, silicon carbide nanorods, hierarchical structure, through-plane thermal conductivity,
thermal interface materials

fficient heat dissipation is of crucial importance for high-

E power and high-frequency electronic devices to avoid

working under overheating conditions, which are
harmful to the device service life and reliability.1 In actuality,
microgaps between the mating surfaces of heat-generating
electronic components and the heat sink are full of air, which is a
very poor heat conductor with ultralow thermal conductivity (κ:
≈ 0.032 W m−1 K−1),2 leading to a high thermal interface
resistance for the suppression of thermal percolation.3 To solve
this problem, thermal interface materials (TIMs) are commonly
applied to fill these microgaps and bridge between the heat
source and the heat sink for removing excess heat from the
devices.4 Conventional TIMs are made of polymer-based
composites filled with thermally conductive materials, such as
Received: September 26, 2018
Accepted: February 6, 2019
Published: February 6, 2019

© 2019 American Chemical Society 1547 DOI: 10.1021/acsnano.8b07337

ACS Nano 2019, 13, 1547−1554
ACS Nano
Figure 1. Surface morphologies of (a) pristine and (b) SiO2 NPs-decorated GO in TEM. The corresponding (c) SEM image and (d) EDS
spectrum of panel b. (e) Schematic of the fabrication process and the photograph of GHP.
BN, AlN, Al2O3, etc., to achieve thermal conductivity values of
1−5 W m−1 K−1 (50−70 wt % filler content is required).5−7
However, with the rapid increase of integration density and
power consumption in electronics, the accompanying demand
for highly efficient heat dissipation is beyond the ability of
conventional TIMs with which to deal. Therefore, the
development of TIMs with improved through-plane thermal
conductivity has become a subject of considerable scientific and
technological interests.
Graphene is a single layer of densely packed sp2-hybridized
carbon atoms that exhibits a ultrahigh intrinsic thermal
conductivity of 3500−5300 W m−1 K−1 and large specific
surface area of ∼2630 m2 g−1.8−10 Due to the above
characteristics, graphene/polymer composites with enhanced
thermal conductivity,11−13 graphene coatings for efficient
radiative cooling,14 and graphene papers as heat spreaders15−17
have been extensively studied in recent years for development of
the next generation of thermal management applications.
Among them, graphene papers composed of layer-by-layer
stacked graphene sheets have been commercially used as high-
performance heat spreaders in portable devices because of its
excellent in-plane thermal conductivity(κ: 1000−2000 W m−1
K−1)18,19 and ease of mass production.20 Lian et al. fabricated a
graphene paper with the size of 18 in. × 4.5 in. and in-plane
thermal conductivity of 1434 W m−1 K−1 by integration of
electro-spray deposition and continuous roll-to-roll process.16
However, in contrast to the superior in-plane thermal
conductivity, conventional graphene paper exhibits a much-
lower through-plane conductivity (κ: 0.1−3.4 W m−1 K−1)20−23
due to the high barrier for phonon transmission across a van der
Waals interface between graphene layers.20 Recently, some
efforts have been done to solve this issue on improvement of the
through-plane thermal conductivity of graphene papers. Jiang et
al. constructed a graphene hybrid paper by growth of carbon
nanoring between graphene sheets through annealing with the
assistance of metal catalysts.20 Although the through-plane
thermal conductivity of the paper was improved by a factor of 3
after hybridization, the obtained value of 5.8 W m−1 K−1 is just
close to that of commercial TIMs.4 Moreover, in the practical
use, a proper packing pressure of 50−100 psi needs to be given
to guarantee a good contact between TIMs and two mating
surfaces.4 However, for the case of conventional graphene paper,
the horizontally aligned graphene sheets would become more
ordered under normal compression, resulting in a further
1548
Article
decrease of through-plane heat-transfer capability. Drzal et al.
applied a cold compaction process with 100 psi to graphene
paper and found that the through-plane thermal diffusivity
decreased by 60%.22 Therefore, so far, it is still lack of in-depth
studies on potential TIM applications based on paper-like
graphene structure. As a result, to meet the requirement for use
as TIM, it is a major challenge to create a paper-like graphene
structure with not only high through-plane thermal conductivity
but also appropriate mechanical properties.
In this work, a graphene/SiC nanorod hybrid paper with
hierarchical structure was fabricated by vacuum filtration of a
mixture of SiO2 nanoparticles (NPs)-decorated graphene sheets,
followed by high-frequency heat treatment to grow SiC
nanorods between graphene layers. The through-plane thermal
diffusivity of as-prepared graphene hybrid paper (GHP) can be
improved by a factor of ∼2.2 compared to pristine paper.
Interestingly, under compression of 75 psi, the through-plane
thermal conductivity of GHP can be further increased up to 17.6
W m−1 K−1, which significantly outperforms conventional
TIMs.4 The comparative tests between GHP and state-of-the-
art commercial TIMs also demonstrate the superior perform-
ance of GHP for cooling electronic devices.
RESULTS AND DISCUSSION
The hierarchically structured graphene/SiC nanorods was
developed by the interaction of SiO2 NPs between graphene
layers, followed by the growth of SiC nanorods by rapid heat
treatment at atmosphere. To act as the nucleation sites and
silicon sources, pristine graphene oxide (GO) was decorated
with SiO2 NPs through the hydrolysis of TEOS in an alkaline
environment. The morphology of GO before and after SiO2 NPs
decoration is present in panels a−c of Figure 1, respectively, in
which a uniform distribution of SiO2 NPs can be seen on the GO
surface after hydrolysis. The composition of SiO2 NPs-
decorated GO was identified using energy-dispersive X-ray
spectroscopy (EDS) analysis in transmission electron micros-
copy (TEM), by which a strong Si peak can be found in Figure
1d, and others peaks are assigned to the GO and copper grid,
respectively.25 Figure 1e shows the schematic diagram of the
fabrication process of GHP and also illustrates our concept of
this study to create a hierarchical microstructure composed of
graphene and SiC nanorods by a simple filtration and post-
annealing process. As displayed in Figure 1e, a paper was first
made by vacuum filtration of a mixture of the as prepared SiO2
DOI: 10.1021/acsnano.8b07337
ACS Nano 2019, 13, 1547−1554
ACS Nano
Figure 2. Top-view and cross-sectional SEM images of (a, c) GHP and (b, d) GP, respectively. The comparison of (e) XRD patterns, (f) Raman
spectra, (g) high-resolution XPS C 1s spectra, and (h) TGA curves between GHP and GP.
NPs-decorated GO and graphene sheets with the mass ratio of
1:9 to obtain SiO2 NPs-intercalated paper-like structure. After
filtration, a rapid annealing process was performed to grow SiC
nanorods between graphene layers at 1400 °C in air. The
formation of SiC nanorods is based on carbothermal reduction
reaction according to reactions 1−3,26 and the comparative
XRD patterns and Raman spectra of the papers before and after
annealing can be found in Figure S1:
SiO2(s) + 3C(s) = SiC(s) + 2CO(g)
(1)
SiO2(s) + CO(g) = SiO(g) + CO2(g)
(2)
3SiO(g) + CO(g) = SiC(s) + 2SiO2(l)
(3)
According to the previous report, a relatively strong covalent
bonding at the interface of graphene−SiC nanorods can be
achieved.27 As a result, a GHP with the diameter of 40 mm and
the thickness of 500 μm could be obtained.
The comparison of the microscopic morphologies of our
GHP and graphene paper (GP) prepared with the same process
are presented in Figure 2a−d. In Figures 2a and S2a−c, with
SiO2 NPs intercalation and post-annealing, the top surface of
GHP is fully covered by SiC nanorods with diameter of 20−100
nm and the length of 5−15 μm, in contrast to the smooth top
surface of GP (Figure 2b). TEM observations confirmed that the
crystalline 3C-SiC nanorods were synthesized with a [111]
preferred growth direction, and a thin amorphous SiO2 layer
(∼1 nm) can be seen on the outer surface of the nanorods (see
Figures S3 and S4). Moreover, SiC nanorods are not only grown
on the surface but also act as the bridge between graphene layers
within GHP to form a hierarchical structure, as shown in the
cross-sections in Figures 2c and S2d−f. The cross-sectional
morphology of GHP is fundamentally different from that of GP
(Figure 2d), which consisted of layer-by-layer stacking of
graphene sheets and was in agreement with the microstructure
of conventional filtrated paper.18
The crystal information and chemical composition of
graphene papers with and without the hybridization of SiC
nanorods are identified, as presented in Figure 2e−g. In Figure
2e, apart from the signal at 26.2° originated from (002) plane of
1549
Article
graphene sheets, other diffraction peaks at 35.8°, 60.2°, and
71.9° can be assigned to the (111), (220), and (311) crystal
planes of 3C-SiC (JCPDS card no. 29-1129), respectively. A pair
of tiny peaks can be seen at 28.6° and 47.3°, which may derive
from the fully reduced silicon. Raman spectroscopy (Figure 2f)
indicates that the characteristic peaks of graphene can be found
at 1337 (D-band), 1556 (G-band), and 2685 cm−1 (2D-band),
respectively.28 The evaluated I2D-to-IG ratio of 0.57 suggests the
multilayer nature of graphene sheets. A pair of peaks at 772 and
906 cm−1 are in correspondence with the vibrational frequencies
of transverse and longitudinal optical modes of 3C-SiC at the Γ
point, respectively.29 The chemical composition of GHP and GP
was also analyzed by X-ray photoelectron spectroscopy (XPS),
as exhibited in Figures 2g and S5. In Figure 2g, the C1s peak of
GHP can be deconvoluted into four major components: C=C
(44%, at 284.6 eV), C−O (13%, at 285.5 eV), C=O (6%, at
286.3 eV), and C−Si (37%, at 283.3 eV), respectively.25,30 The
deconvolution of high-resolution Si2p signal in Figure S5b
reveals two peaks at 101.4 (Si−C, 95.7%) and 103.8 eV (Si−O,
4.3%), respectively,25 indicating that the mass ratio of SiC
component in nanorods is ∼93.7 wt %. To investigate the mass
ratio of SiC nanorods in GHP, both GP and GHP were analyzed
by TGA in air, as in the curves shown in Figure 2h. Compared to
the complete decomposition of GP, the thermogravimetric
analysis (TGA) curve of GHP exhibits a residual weight of
∼6.94 wt % at 900 °C. In addition, the pure SiC nanorods were
individually analyzed in the same conditions, showing a 1.3 wt %
weight increment at 900 °C. Accordingly, the mass ratio of SiC
nanorods in GHP can be obtained to be ∼6.85 wt %, which is
composed of mainly crystalline SiC (∼6.42 wt %) and a small
amount of amorphous SiO2 (∼0.43 wt %), according to the
fitting results of Si2p XPS spectrum.
To investigate the effect of the formation of hierarchical
structure within GHP on the thermal properties, thermal
diffusivity (α) of GHP and GP was measured by the laser flash
method. The corresponding thermal conductivity (κ) can be
calculated by the equation:31 κ = α × ρ × Cp, where ρ is the
sample density and Cp is the specific heat capacity (0.75 J g−1
K−1; see Figure S6). As shown in Figure 3a, with the
development of graphene/SiC nanorod hybrid, the in-plane
DOI: 10.1021/acsnano.8b07337
ACS Nano 2019, 13, 1547−1554
ACS Nano Article

Figure 3. (a) In-plane and through-plane thermal diffusivities of GHP and GP. (b) The calculated thermal conductance at the graphene−SiC
nanorod interface bonded by covalent and van der Waals (vdW) interactions. (c) The test system configuration for determining the through-
plane heat-transfer capacity. (d) Surface temperature evolution and the (e) corresponding IR images of GHP and GP as a function of heating
time.

Figure 4. (a) Through-plane thermal properties of compressed GHP and compressed GP under 75 psi packing pressure. (b) Schematic
illustrating the difference of heat-flow across compressed GHP and compressed GP vertically. (c) Comparison of the through-plane thermal
conductivity of our GHP with the reported graphene-based papers. (d) Through-plane thermal conductivity as a function of environmental
temperature.

thermal diffusivity of GHP is ∼6% lower than that of GP. The
decrease is attributed to the formation of graphene/SiC
junctions on the graphene surface as scattering centers, which
hinder the phonon transport along the basal plane of graphene.20
In contrast, the through-plane thermal diffusivity of GHP (18.4
mm2 s−1) is ∼2.2 times higher than that of GP (8.2 mm2 s−1).
Due to the fact that the spacing between graphene sheets was
expanded by the intercalation of SiC nanorods, the density of
GHP (0.8 g cm−3) is lower than that of GP (1.1 g cm−3). As a
result, the through-plane thermal conductivity of GHP (10.9 W
m−1 K−1) shows a 60% enhancement compared to 6.8 W m−1
K−1 of GP (see Table S1). The more efficient heat-transfer
capacity of GHP along the through-plane direction is
contributed by both the intrinsic high thermal conductivity of
3C-SiC nanorods (bulk: 490 W m−1 K−1, nanorod: 76−88 W
m−1 K−1)32,33 and the creation of covalent C−Si bonding at the
interface of graphene/SiC nanorods due to the in situ growth. As
Drzal et al. reported, the insertion of Au NPs into graphene
layers would lead to a reduction of through-plane thermal
conductivity of graphene/Au NPs paper by 23%, owing to the
weak interfacial heat conduction between them based on the
weak van der Waals (VdW) interaction.22 In our case, based on
1550
the result of non-equilibrium molecular dynamics simulation
(Figure 3b), the interfacial thermal conductance between
graphene/SiC nanorods connected by covalent C−Si bonding
(1.43 GW m−2 K−1) is 1 order of magnitude higher than vdW
interactions (0.14 GW m−2 K−1) (details can be seen in Figure
S7). The high interfacial thermal conductance at the covalently
bonded interface using our proposed method is of significance to
phonon transport across graphene and SiC nanorods.
To compare the through-plane heat-transfer capacity between
GHP and GP, a pair of papers cut into the size of 10 mm × 10
mm × 0.5 mm were placed on a ceramic heater (60 W) and kept
at room temperature initially, followed by heating at the same
time (Figure 3c). The time dependence of their surface
temperature evolution on the heating time was recorded using
a thermocouple and calibrated infrared (IR) camera, respec-
tively, as shown in panels d and e of Figure 3. We found that the
surface temperature of GHP increases faster than that of GP, and
always shows a higher value (e.g., ΔT = 27 °C at 300 s),
confirming the superior thermal conductivity of GHP along the
through-plane direction.
The GHP composed of hierarchically structured graphene/
SiC nanorods has been demonstrated to exhibit a high through-
DOI: 10.1021/acsnano.8b07337
ACS Nano 2019, 13, 1547−1554
ACS Nano
Figure 5. (a) Schematic configuration of TIM performance evaluation system. Temperature evolution of ceramic heater as a function of (b)
heating time at the heater power of 30 W and (c) various applied powers after heating for 200 s. (d) The calculated heater temperature as a
function of effective thermal conductivity. (e) The comparison of heat dissipation capability based on simulated profiles.
plane thermal conductivity (up to 10.9 W m−1 K−1), by which
the GHP presents a great potential for TIM applications.
However, as the TIMs used in electronic packaging, generally, a
packing pressure between 50−100 psi is required to guarantee
good contact between the mating surfaces and TIMs to
eliminate the microgaps at their interface.4 Therefore, the
change of through-plane thermal conductivity of GHP and GP
under vertical compression (75 psi) was studied. As revealed in
Figure 4a, both of the samples show a lower thermal diffusivity of
14.7 mm2 s−1 for compressed GHP and 4.8 mm2 s−1 for
compressed GP, respectively, compared to that of pristine ones.
This is because of the rearrangement of graphene sheets toward
the horizontal direction during vertical compression (Figure
S8), leading to higher in-plane and lower through-plane thermal
diffusivities of the resulting paper with an increase of thermal
anisotropy.16 Interestingly, we noticed that the reduction in
thermal diffusivity of compressed GHP (19.7%) is lower than
that of compressed GP (42.2%). This can be ascribed to the fact
that the SiC nanorods covalently bonded between graphene
sheets lower the phonon diffusion barrier across the basal planes
of graphene (see Figure 4b), thus compensating for the decrease
of through-plane heat conduction capability caused by the
rearrangement of graphene sheets. As a result, after applying the
same vertical compression at 75 psi, the through-plane thermal
conductivity of compressed GHP (17.6 W m−1 K−1) is more
than 3 times higher than that of compressed GP (5.8 W m−1
K−1) under packaging conditions.
To the best of our knowledge, the GHP possesses the highest
through-plane thermal conductivity compared with the other
graphene-based papers in the previous literature (see Figure 4c
and Table S2)20−23,34 and also much better than that of
commercial TIMs, such as thermal gels, pads, and greases (up to
4−5 W m−1 K−1).4 Besides, the environmental temperature
1551
Article
dependence of through-plane thermal conductivity of GHP was
investigated, and it shows a linear drop by 30% from 25 to 145
°C in Figure 4d. Such behavior can be attributed the increased
relaxation time of phonons as the temperature increases and is in
reasonable agreement with Umklapp scattering mechanism of
crystalline materials.35−37
Based on its high through-plane thermal conductivity, GHP
may have the potential to act as a high-performance TIM to
efficiently transfer heat across the heater−heat sink interface.
Therefore, a verification system for simulating heat dissipation
process in electronic components was developed, as schemati-
cally shown in Figure 5a. A state-of-the-art thermal pad (∼5 W
m−1 K−1, 5000S35, Bergquist) was employed for comparison.
The GHP and 5000S35 with the size of 2 cm × 2 cm were
placed, respectively, between the ceramic heater and tge heat
sink with a same bond line thickness (BLT) of 250 μm and
vertical pressure of 75 psi. A water-cooling system was attached
to keep the heat sink at room temperature, and the evolution of
the heater temperature was recorded. In Figure 5b, after the
heater (30 W) turned on at 35 s, the heater temperature rise
rapidly and then reached the equilibrium state. Obviously,
compared to the heater temperature without TIM, the cooling
performance of GHP as TIM by a decrease of 18.3 °C at 200 s is
much better than that of 5000S35 thermal pad (9.8 °C). The
saturated temperatures of the heater versus applied power with
and without TIMs are plotted in Figure 5c, in which the slope
values are 1.32 (without TIM), 0.99 (5000S35), and 0.72 °C
W−1 (GHP), respectively, indicating that GHP achieves 45.5%
and 27.3% improvements of cooling efficiency compared to the
system without TIM and with 5000S35 thermal pad.
A commercial computational fluid dynamics software
(ANSYS Icepak) was then utilized for thermal analysis of our
test configuration at 30 W applied power (Figure S9), and the
DOI: 10.1021/acsnano.8b07337
ACS Nano 2019, 13, 1547−1554
ACS Nano
Figure 6. (a) Experimental setup and the schematic configuration for comparing the cooling efficiency between GHP and state-of-the-art
thermal pad based on desktop computer systems. (b) CPU core temperature evolution as a function of running time. (c) Comparative IR images
of the motherboard when 5000S535 and GHP were used as TIMs, respectively.
effective thermal conductivity (κeff) of two TIMs at the same
measurement conditions can be estimated (see Figure 5d).
Based on the values (κeff) for GHP (∼4.1 W m−1 K−1) and
5000S35 (∼2.1 W m−1 K−1), not only superior through-plane
thermal conductivity, the calculated thermal contact resistance
(two sides) of our GHP (47 K mm2 W−1) is also significantly
lower than that of 5000S35 thermal pad (69 K mm2 W−1). The
detailed calculation process can be found in Figure S10. The
lower thermal contact resistance can be attributed the
characteristic hierarchical structure of graphene/SiC nanorods,
which gives GHP not only a high through-plane thermal
conductivity but also lower compressive modulus compared to
that of 5000S35 (Figure S11). At 75 psi, the compressive
modulus of GHP (∼0.64 MPa) is only about one-third of that of
5000S35 (∼1.95 MPa). As a result, GHP with a soft
compressibility is easier to fill up the microgaps between the
mating surfaces under packaging, leading to a maximum of
contact area at the microscale and lower thermal contact
resistance.38 Combining the high through-plane thermal
conductivity and low thermal contact resistance, the simulated
temperature profiles in Figure 5e confirm the outstanding heat-
transfer properties of our sample for TIM use.
A real TIM application for cooling computer CPU was carried
out to demonstrate superior heat dissipation performance of
GHP compared to 5000S35 thermal pad. In Figure 6a, GHP and
5000S35 with the same area and BLT were directly physically
fastened between the CPU (Intel Core i5-7500) and the
aluminum heat sink by vertically applying a packing pressure,
which was carried out using spring screws. The CPU is fixed on
the motherboard, and the heat sink is connected to a cooling fan
through the screws to transfer heat to the air flowing. A
professional system diagnostics software (AIDA64)39 was
employed to drive the CPU running at maximum heat-
generating conditions (42 W) and record the CPU core
temperature (TCPU core). As the results presented in Figure 6b, it
can be seen that the TCPU core of the case using GHP as TIM
increases more slowly than that of 5000S35 and always shows a
lower value (e.g.: ΔT = 6.3 °C at 780 s). And the comparative
temperature profiles of the computer motherboard in Figure 6c
also indicate a lower temperature when GHP was used as TIM.
As a result, our GHP is a very promising candidate to replace the
state-of-the-art thermal pad for real electronic cooling
applications.
1552
Article
CONCLUSIONS
In summary, a graphene/SiC nanorod hybrid paper was
developed by a facile and easy to scale-up filtration method,
followed by rapid heat treatment to grow in situ SiC nanorods
between graphene sheets based on carbothermal reduction
reaction. The GHP shows a characteristic structure composed of
hierarchical graphene/SiC nanorod architecture, leading to an
enhanced through-plane thermal conductivity (10.9 W m−1
K−1) by 60% compared to that of GP. In addition, conventional
graphene-based papers lose their through-plane thermal
conductivity as a compressive force is vertically applied to
them, which is not capable of developing practical TIM
applications. In contrast, due to the formation of relatively
strong covalent C−Si bonding, the through-plane thermal
conductivity of compressed GHP can be even higher (up to 17.6
W m−1 K−1) with a compression force of 75 psi. Accordingly, we
demonstrated that the heater temperature of compressed GHP
by a decrease of 18.3 °C in TIM performance test is superior
than 9.8 °C of commercial thermal pads (κ: 5 W m−1 K−1) with
the cooling efficiency improving by 27.3%, which is also verified
under actual operating conditions. By getting rid of the aging
issue of conventional polymer-based TIMs, our GHP with
characteristically inorganic structure has great potential to be
used as high-performance TIMs with good thermal and
chemical stability.
METHODS
Materials. Graphene sheets prepared by the intercalation and
exfoliation of graphite were supplied by Ningbo Morsh Technology
Co., Ltd., and the sample characterizations have been done in our
previous work.24 The lateral size and thickness of graphene sheets are
5.4 ± 0.3 μm and 10.6 ± 0.3 nm, respectively. Graphene oxide (type
HG03-1) with the typical lateral size and thickness of 3.0 μm and 2.5
nm was supplied by Qingdao Huagaomoxi Technology Co., Ltd.
Tetraethoxysilane (TEOS), ammonia, and ethanol were purchased
from Sinopharm Chemical Reagent Co., Ltd. Pure SiC nanorods were
obtained from Changsha Sinet Advanced Materials Co., Ltd. All
chemicals were of analytical reagent grade and used without further
purification.
Preparation of SiO2 Nanoparticle-Decorated GO. A hydrolysis
method was employed to prepare SiO2 NPs-decorated GO. GO was
dispersed in a mixture of ethanol (240 mL) and deionized water (24
mL) with ultrasonication for 120 min. Then, 4 mL of ammonia was
added to this dispersion for adjusting the pH value, following by the
addition of 0.5 mL of TEOS. After hydrolysis with bath ultrasonication
for 180 min, the product was collected by vacuum filtration and washed
DOI: 10.1021/acsnano.8b07337
ACS Nano 2019, 13, 1547−1554
ACS Nano
with deionized water several times until a neutral pH was reached. SiO2
NPs-decorated GO was finally obtained by drying the product at 60 °C.
Fabrication of Graphene Hybrid Paper. Graphene sheets (450
mg) and as prepared SiO2 NPs-decorated GO (50 mg) were uniformly
dispersed in ethanol via ultrasonic treatment. The dispersion was then
filtered through a Teflon filter membrane (pore size: 0.22 μm) to obtain
a thin paper. The paper with the intercalation of SiO2 NPs was treated
in high-frequency furnace at 1400 °C for 4 min. A graphene paper as a
control sample was also made of 500 mg graphene sheets without SiO2
NPs decoration using the same experimental process.
Characterizations. Raman spectra were recorded using a Reflex
Raman System (Renishaw PLC) employing a laser wavelength of 532
nm. TEM images were taken by JEM-2100 (JEOL) with an acceleration
voltage of 200 kV. Electron energy loss spectroscopy was performed
with a Gatan parallel spectrometer, and the corresponding analysis was
conducted using the scanning transmission electron microscopy
(STEM) mode of TEM. The surface morphology of the samples was
examined with field emission scanning electron microscopy (Quanta
FEG250, FEI). X-ray diffraction patterns were recorded by D8 Discover
with GADDS (Bruker) with CuKa radiation (λ = 1.54 Å). X-ray
photoelectron spectroscopy was carried out with an AXIS Ultra DLD
spectrometer (Kratos Analytical). The compression tests were carried
out on an electron omnipotence tester of universal testing machine
(UTM, 5567A, Instron). All samples were fabricated into a standard
size. The loading rate was controlled as 0.1 mm min−1. Thermal
conductivity of the samples was measured with an LFA 467 NanoFlash
apparatus (Netzsch). The IR photos were captured by using an infrared
camera (Fluke, Ti400).
ASSOCIATED CONTENT
*
S Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acsnano.8b07337.
XRD patterns and Raman spectra of GHP before and after
annealing; SEM images of GHP; TEM images of SiC
nanorods grown on the surface of graphene; STEM image
and EELS spectra of an individual SiC nanorod; XPS
spectra of GP and GHP; DSC curves of GP and GHP; a
schematic and parameter setting of the NEMD
implementation; cross-sectional SEM images of GP
before and after compression; a comparison of through-
plane thermal conductivity of graphene-based papers; the
schematic, parameter setting, and results of Icepak
implementation; a schematic illustrating the calculative
process of the thermal contact resistance; and stress−
strain curves of GHP and 5000S35; additional exper-
imental details (PDF)
AUTHOR INFORMATION
Corresponding Authors
*E-mail: Jibao.lu@siat.ac.cn..
*E-mail: jiangnan@nimte.ac.cn.
*E-mail: linzhengde@nimte.ac.cn.
ORCID
Xiaoliang Zeng: 0000-0002-1389-8484
Xiangfan Xu: 0000-0001-7163-4957
Jianbo Wu: 0000-0002-3574-5585
Shiyu Du: 0000-0001-6707-3915
Rong Sun: 0000-0001-9719-3563
Jianbin Xu: 0000-0003-0509-9508
Ching-Ping Wong: 0000-0003-3556-8053
Cheng-Te Lin: 0000-0002-7090-9610
1553
Article
Author Contributions
¶
W.D. and L.L. contributed equally in this work for conceiving
and designing the preliminary experiments. W.D. and C.T.L.
wrote the paper and revised the manuscript at all stages. H.H.,
Q.Y., and X.T. performed partial materials characterization
including Raman, XRD and XPS. J.L., Y.L., S.D., and X.X.
provided the NEMD and heat flux simulation and data analysis.
N.J. measured the EELS sprectra of the sample and contributed
reagents, materials, and analysis tools. X.Z., J.Y., Q.W., and J.W.
helped the mechanism explanation and assisted with the
revision. R.S., J.X., N.J., and C.P.W. provided some constructive
suggestions and supervised this study.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
The authors are grateful for the financial support by the National
Key R&D Program of China (grant no. 2017YFB0406000),
Scientific Instrument Developing Project of the Chinese
Academy of Sciences (grant no. YZ201640), the Project of the
Chinese Academy of Sciences (grant no. KFZD-SW-409),
Science and Technology Major Project of Ningbo (grant nos.
2016S1002 and 2016B10038), and the International S&T
Cooperation Program of Ningbo (grant no. 2017D10016) for
financial support. We also thank the Chinese Academy of
Sciences for Hundred Talents Program, Chinese Central
Government for Thousand Young Talents Program, 3315
Program of Ningbo, and the Key Technology of Nuclear Energy
(CAS Interdisciplinary Innovation Team, 2014).
REFERENCES
(1) Liang, Q.; Yao, X.; Wang, W.; Liu, Y.; Wong, C. P. A Three-
Dimensional Vertically Aligned Functionalized Multilayer Graphene
Architecture: An Approach for Graphene-Based Thermal Interfacial
Materials. ACS Nano 2011, 5, 2392−2401.
(2) Lemmon, E. W.; Jacobsen, R. T. Viscosity and Thermal
Conductivity Equations for Nitrogen, Oxygen, Argon, and Air. Int. J.
Thermophys. 2004, 25, 21−69.
(3) Lv, P.; Tan, X. W.; Yu, K. H.; Zheng, R. L.; Zheng, J. J.; Wei, W.
Super-Elastic Graphene/Carbon Nanotube Aerogel: A Novel Thermal
Interface Material with Highly Thermal Transport Properties. Carbon
2016, 99, 222−228.
(4) Hansson, J.; Nilsson, T. r. M. J.; Ye, L.; Liu, J. Novel
Nanostructured Thermal Interface Materials: A Review. Int. Mater.
Rev. 2018, 63, 22−45.
(5) Gu, J.; Zhang, Q.; Dang, J.; Xie, C. Thermal Conductivity Epoxy
Resin Composites Filled with Boron Nitride. Polym. Adv. Technol. 2012,
23, 1025−1028.
(6) Prasher, R. Thermal Interface Materials: Historical Perspective,
Status, and Future Directions. Proc. IEEE 2006, 94, 1571−1586.
(7) Shahil, K. M. F.; Balandin, A. A. Graphene-Multilayer Graphene
Nanocomposites as Highly Efficient Thermal Interface Materials. Nano
Lett. 2012, 12, 861−867.
(8) Balandin, A. A.; Ghosh, S.; Bao, W.; Calizo, I.; Teweldebrhan, D.;
Miao, F.; Lau, C. N. Superior Thermal Conductivity of Single-Layer
Graphene. Nano Lett. 2008, 8, 902−907.
(9) Park, S. D.; Won Lee, S.; Kang, S.; Bang, I. C.; Kim, J. H.; Shin, H.
S.; Lee, D. W.; Won Lee, D. Effects of Nanofluids Containing
Graphene/Graphene-Oxide Nanosheets on Critical Heat Flux. Appl.
Phys. Lett. 2010, 97, 023103.
(10) Zheng, C.; Zhou, X.; Cao, H.; Wang, G.; Liu, Z. Synthesis of
Porous Graphene/Activated Carbon Composite with High Packing
Density and Large Specific Surface Area for Supercapacitor Electrode
Material. J. Power Sources 2014, 258, 290−296.
(11) Shen, X.; Wang, Z.; Wu, Y.; Liu, X.; He, Y. B.; Zheng, Q.; Yang, Q.
H.; Kang, F.; Kim, J. K. A Three-Dimensional Multilayer Graphene
DOI: 10.1021/acsnano.8b07337
ACS Nano 2019, 13, 1547−1554
ACS Nano Article

Web for Polymer Nanocomposites with Exceptional Transport for the Preparation of a Nanocomposite with High Thermal
Properties and Fracture Resistance. Mater. Horiz. 2018, 5, 275−284. Conductivity. ACS Nano 2017, 11, 5167−5178.
(12) Shen, X.; Wang, Z.; Wu, Y.; Liu, X.; He, Y. B.; Kim, J. K. (32) Slack, G. A. Thermal Conductivity of Pure and Impure Silicon,
Multilayer Graphene Enables Higher Efficiency in Improving Thermal Silicon Carbide, and Diamond. J. Appl. Phys. 1964, 35, 3460−3466.
Conductivities of Graphene/Epoxy Composites. Nano Lett. 2016, 16, (33) Lee, K.; Choi, T.; Lee, S.; Poulikakos, D. Focused Ion Beam-
3585−3593. Assisted Manipulation of Single and Double β-SiC Nanowires and
(13) Lian, G.; Tuan, C. C.; Li, L.; Jiao, S.; Wang, Q.; Moon, K. S.; Cui, Their Thermal Conductivity Measurements by the Four-Point-Probe
D.; Wong, C. P. Vertically Aligned and Interconnected Graphene 3ω-Method. Nanotechnology 2010, 21, 125301.
Networks for High Thermal Conductivity of Epoxy Composites with (34) Xiang, J.; Drzal, L. T. Thermal Conductivity of Exfoliated
Ultralow Loading. Chem. Mater. 2016, 28, 6096−6104. Graphite Nanoplatelet Paper. Carbon 2011, 49, 773−778.
(14) Hsiao, T. J.; Eyassu, T.; Henderson, K.; Kim, T.; Lin, C. T. (35) Kholmanov, I.; Kim, J.; Ou, E.; Ruoff, R. S.; Shi, L. Continuous
Monolayer Graphene Dispersion and Radiative Cooling for High Carbon Nanotube-Ultrathin Graphite Hybrid Foams For Increased
Power LED. Nanotechnology 2013, 24, 395401. Thermal Conductivity and Suppressed Subcooling in Composite Phase
(15) Peng, L.; Xu, Z.; Liu, Z.; Guo, Y.; Li, P.; Gao, C. Ultrahigh Change Materials. ACS Nano 2015, 9, 11699−11707.
Thermal Conductive yet Superflexible Graphene Films. Adv. Mater. (36) Cao, J. X.; Yan, X. H.; Xiao, Y.; Ding, J. W. Thermal Conductivity
2017, 29, 1700589. of Zigzag Single-Walled Carbon Nanotubes: Role of the Umklapp
(16) Xin, G.; Sun, H.; Hu, T.; Fard, H. R.; Sun, X.; Koratkar, N.; Borca Process. Phys. Rev. B: Condens. Matter Mater. Phys. 2004, 69, 073407.
Tasciuc, T.; Lian, J. Large Area Freestanding Graphene Paper for (37) Pop, E.; Mann, D.; Wang, Q.; Goodson, K.; Dai, H. Thermal
Superior Thermal Management. Adv. Mater. 2014, 26, 4521−4526. Conductance of an Individual Single-Wall Carbon Nanotube above
(17) Renteria, J. D.; Ramirez, S.; Malekpour, H.; Alonso, B.; Centeno, Room Temperature. Nano Lett. 2006, 6, 96−100.
A.; Zurutuza, A.; Cocemasov, A. I.; Nika, D. L.; Balandin, A. A. Strongly (38) Bhanushali, S.; Ghosh, P. C.; Simon, G. P.; Cheng, W. Copper
Anisotropic Thermal Conductivity of Free Standing Reduced Nanowire Filled Soft Elastomer Composites for Applications as
Graphene Oxide Films Annealed at High Temperature. Adv. Funct. Thermal Interface Materials. Adv. Mater. Interfaces 2017, 4, 1700387.
Mater. 2015, 25, 4664−4672. (39) Shi, Y.; Veidenbaum, A. V.; Nicolau, A.; Xu, X. Large-Scale
(18) Teng, C.; Xie, D.; Wang, J.; Yang, Z.; Ren, G.; Zhu, Y. Ultrahigh Neural Circuit Mapping Data Analysis Accelerated with the Graphical
Conductive Graphene Paper Based on Ball Milling Exfoliated Processing Unit (GPU). J. Neurosci. Methods 2015, 239, 1−10.
Graphene. Adv. Funct. Mater. 2017, 27, 1700240.
(19) Shen, B.; Zhai, W.; Zheng, W. Ultrathin Flexible Graphene Film:
An Excellent Thermal Conducting Material with Efficient EMI
Shielding. Adv. Funct. Mater. 2014, 24, 4542−4548.
(20) Zhang, J.; Shi, G.; Jiang, C.; Ju, S.; Jiang, D. 3D Bridged Carbon
Nanoring/Graphene Hybrid Paper as a High Performance Lateral Heat
Spreader. Small 2015, 11, 6197−6204.
(21) Kong, Q. Q.; Liu, Z.; Gao, J. G.; Chen, C. M.; Zhang, Q.; Zhou,
G.; Tao, Z. C.; Zhang, X. H.; Wang, M. Z.; Li, F.; et al. Hierarchical
Graphene-Carbon Fiber Composite Paper as a Flexible Lateral Heat
Spreader. Adv. Funct. Mater. 2014, 24, 4222−4228.
(22) Xiang, J.; Drzal, L. T. Electron and Phonon Transport in Au
Nanoparticle Decorated Graphene Nanoplatelet Nanostructured
Paper. ACS Appl. Mater. Interfaces 2011, 3, 1325−1332.
(23) Wu, H.; Drzal, L. T. Graphene Nanoplatelet Paper as a Light-
Weight Composite with Excellent Electrical and Thermal Conductivity
and Good Gas Barrier Properties. Carbon 2012, 50, 1135−1145.
(24) Alam, F. E.; Dai, W.; Yang, M.; Du, S.; Li, X.; Yu, J.; Jiang, N.; Lin,
C. T. In-situ Formation of a Cellular Graphene Framework in
Thermoplastic Composites Leading to Superior Thermal Conductivity.
J. Mater. Chem. A 2017, 5, 6164−6169.
(25) Dai, W.; Yu, J.; Wang, Y.; Song, Y.; Alam, F. E.; Nishimura, K.;
Lin, C. T.; Jiang, N. Enhanced Thermal Conductivity for Polyimide
Composites With a Three-Dimensional Silicon Carbide Nanowire@
Graphene Sheets Filler. J. Mater. Chem. A 2015, 3, 4884−4891.
(26) Li, X.; Chen, X.; Song, H. Preparation of Silicon Carbide
Nanowires via a Rapid Heating Process. Mater. Sci. Eng., B 2011, 176,
87−91.
(27) Meng, G. W.; Cui, Z.; Zhang, L. D.; Phillipp, F. Growth and
Characterization of Nanostructured β-SiC via Carbothermal Reduction
of SiO2 Xerogels Containing Carbon Nanoparticles. J. Cryst. Growth
2000, 209, 801−806.
(28) Malard, L. M.; Pimenta, M. A.; Dresselhaus, G.; Dresselhaus, M.
S. Raman Spectroscopy In Graphene. Phys. Rep. 2009, 473, 51−87.
(29) Bechelany, M.; Brioude, A.; Cornu, D.; Ferro, G.; Miele, P. A
Raman Spectroscopy Study of Individual SiC Nanowires. Adv. Funct.
Mater. 2007, 17, 939−943.
(30) Ma, Q. B.; Ziegler, J.; Kaiser, B.; Fertig, D.; Calvet, W.;
Murugasen, E.; Jaegermann, W. Solar Water Splitting With p-SiC Film
on p-Si: Photoelectrochemical Behavior and XPS Characterization. Int.
J. Hydrogen Energy 2014, 39, 1623−1629.
(31) Zeng, X.; Sun, J.; Yao, Y.; Sun, R.; Xu, J. B.; Wong, C. P. A
Combination of Boron Nitride Nanotubes and Cellulose Nanofibers

1554 DOI: 10.1021/acsnano.8b07337

ACS Nano 2019, 13, 1547−1554