enhanced dielectric and thermal performance by fabricating
TRANSCRIPT
Enhanced dielectric and thermal performance by fabricatingcoalesced network of alumina trihydrate/boron nitride in siliconerubber for electrical insulation
M TARIQ NAZIR1,* , B TOAN PHUNG2, GUAN H YEOH1, GHULAM YASIN3,SHAKEEL AKRAM4, M SHOAIB BHUTTA5, M ALI MEHMOOD5, SHAHID HUSSAIN6,SHIHU YU7 and IMRANA KABIR1
1 School of Mechanical and Manufacturing Engineering, University of New South Wales, Sydney 2052, Australia2 School of Electrical Engineering and Telecommunications, University of New South Wales, Sydney 2052,
Australia3 State Key Laboratory of Chemical Resource Engineering, College of Energy, Beijing University of Chemical
Technology, Beijing 100029, People’s Republic of China4 Institut d’Electronique et des Systemes (IES), UMR 5214, CNRS, Universite de Montpellier, Montpellier 34095,
France5 The State Key Laboratory of Power Transmission Equipment and System Security and New Technology, Chongqing
University, Chongqing 400044, People’s Republic of China6 School of Materials Science and Engineering, Jiangsu University, Zhenjiang 212013, People’s Republic of China7 Electric Power Research Institute of Guangdong Power Grid Co. Ltd., Guangzhou 510080, People’s Republic of China
*Author for correspondence ([email protected])
MS received 20 January 2020; accepted 18 February 2020
Abstract. Silicone rubber filled with micron-alumina trihydrate (ATH) is a substantially used composite material for
high voltage outdoor insulators. This article investigates the effect of nano-boron nitride (BN) addition on the dielectric,
thermal stability and thermal conductivity of solely micron-ATH-filled silicone rubber by fabricating coalesced network
of particles. Micron-ATH/nano-BN-filled hybrid silicone rubber composites are fabricated with a ratio of 30/0 wt%
(ATH30), 29/1 wt% (ATBN1), 27/3 wt% (ATBN3), 25/5 wt% (ATBN5) and 23/7 wt% (ATBN7) using mechanical
mixing and water bath sonication techniques. Results suggest that the hybrid batch of silicone rubber composites (ATBN)
possess lower permittivity, dielectric loss, enhanced thermal stability and thermal conductivity relative to ATH30.
ATBN1 offers low permittivity and dielectric loss values of 3.64 and 0.0086 at 1000 Hz relative to 3.87 and 0.0224 of
ATH30, respectively. As far as thermal properties are concerned, ATBN5 emerges as the most promising candidate for
electrical insulation with 31 and 200�C higher temperatures for 10 and 15% mass loss, whilst it has shown 20% higher
thermal conductivity than ATH30.
Keywords. Silicone rubber; hybrid composites; dielectric properties; thermal stability; thermal conductivity.
1. Introduction
Over the past three decades, substantial attention has
been grabbed by silicone rubber, ethylene propylene
diene monomer (EPDM), ethylene-vinyl acetate and
cycloaliphatic epoxy for the fabrication of weather sheds
of high voltage insulators. The major attributes of
polymeric insulating technology are hydrophobic nature,
UV resistance, excellent dielectric, thermal properties
and excellent flashover performance [1]. The raw form
of polymers could not provide the required thermal,
dielectric and mechanical properties and therefore,
reinforcing fillers are introduced to enhance the insula-
tion performance [2].
Polymeric materials filled with alumina trihydrate (ATH)
are widely used for the manufacturing of insulators recently.
ATH is added by 30–50 wt% into pristine polymeric
materials to enhance the resistance against dry band arcing-
induced thermal deterioration of polymer in the form of
electrical carbon tracking and erosion [3]. Recent devel-
opments in nanodielectrics have opened new directions for
the insulation design. Relatively small contents of nano-
sized particles impart outstanding dielectric, thermal and
electrical properties to resultant composites [4]. One school
of thought suggests that micron-sized ATH and silica-filled
polymeric insulators are still a preferred choice for the
power utilities and manufacturing because such units are
more economical due to significantly lower consumption of
Bull Mater Sci (2020) 43:220 � Indian Academy of Scienceshttps://doi.org/10.1007/s12034-020-02201-8Sadhana(0123456789().,-volV)FT3](0123456789().,-volV)
polymer into the composites relative to nanoparticle-filled
counterparts. Hence, it motivates the authors to enhance the
dielectric and thermal properties of solely ATH-filled sili-
cone rubber by synthesizing the hybrid structure of micron
and nanoparticles.
Recently, several studies have been reported on the
synergistic impact of hybrid set of fillers on dielectric and
thermal properties of insulating materials. Zha et al [5]
explored the coalesced role of micro-Si3N4/nano-Al2O3 on
the thermal, dielectric and mechanical properties of silicone
rubber. Hybrid network with Si3N4:Al2O3 = 26:4 showed
the excellent improvement in the properties. Qi et al [6]
studied the polyvinylidene fluoride co-filled composites
with three types of boron nitride (BN) and carbon nanotubes
(CNTs) and it was found that the moderate size of BN along
with CNTs showed substantial improvement in the thermal
conductivity of the composites. The authors’ previous work
reported improvement in thermal conduction, stability and
dielectric performance of EPDM co-filled composites [7]. A
most recent study by Cai reported that small contents of
graphene nanosheets into BN-filled polytetrafluoroethylene
impart outstanding improvement in the thermal conductiv-
ity of composites [8].
To date, the literature on enhancing the dielectric and
thermal properties of solely ATH-filled silicone rubber is
sparse. To address this issue, four sets of samples with a
hybrid network of micron-ATH/nano-BN were fabricated
with a ratio of 29/1, 27/3, 25/5 and 23/7 wt%. Moreover, a
solely ATH-filled silicone rubber sample with 30 wt% was
also fabricated for the comparison.
2. Experimental
2.1 Materials
Room temperature-vulcanized (RTV) silicone rubber (trade
name of RTV 615) was procured from DC Products Pty Ltd,
Australia. RTV 615 (density: 1.02 g cm-3) is a high
transparent and easily flowable two-part platinum catalyst
that cures polydimethylsiloxane (PDMS) matrix with a
viscosity of 4300 cps [9]. Moreover, ATH with a size of
5 lm and a density of 2.50 g cm-3 was procured in
amorphous form from Southstar China Co., Ltd, China. BN
particles with a size of 100 nm were procured from HW
Nanomaterials, China.
2.2 Sample fabrication
Firstly, part A of RTV 615 was degaussed using the vacuum
pump until no air bubbles were seen on the top of the
matrix. Moreover, all the particles were dried overnight in a
laboratory oven at 100�C. The required amount of ATH and
BN particles was mixed in part A of PDMS using a hand
stirrer and then using a sharp blade mechanical mixer for
5 min, subsequently. After that, the matrix beaker was
placed in a water tank of the ultrasonic cleaner. Sonication
was continued for an hour at the frequency of 40 kHz.
Finally, the part B of RTV was added to the mixed com-
pound by a ratio of 10:1. Subsequently, the composites were
poured into preheated moulds and kept in a vacuum again to
remove the trapped air bubbles. Later, the moulds were kept
in the oven at 150�C for 30 min. All the samples were
cleansed with ethanol and dried in oven for 15 min at 80�C.
A detailed fabrication method is reported earlier in our
previous article [10,11]. The composition of samples used
in this study is given in table 1.
2.3 Characterization
In order to investigate the microscopic structure of the
composites, scanning electron microscopy (Hitachi S3400)
was used for the morphological study. All the composites
were kept in a drying chamber for 24 h. Subsequently,
samples were coated with a layer of gold in an argon
atmosphere. Figure 1 shows the micrograph captured on the
top surface of ATBN5. It is observed that several ATH and
BN particles are dispersed in the silicone rubber that
demonstrates a good quality of distribution and compati-
bility between particles and matrix. The coalesced set of
particles forms a compact structure that may help in thermal
transport. The dielectric properties of the samples were
examined by the Novocontrol dielectric spectrometer in the
frequency range of 10 to 106 Hz. The dielectric spec-
troscopy was performed at room temperature (30�C) and
relative humidity of RH = 50%. The test samples were
casted with a 4 cm diameter, whilst a gold electrode with a
diameter of 3 cm was sputter-coated on the top of the
samples. In dielectric response, the real and imaginary
permittivity of the tested sample is expressed as follows:
e ¼ e0 � je00: ð1Þ
Here e0 and e00 represent the real and imaginary parts of
complex permittivity, respectively, whilst the dielectric loss
(tan d) could be calculated by the following equation:
tan d ¼ e00
e0: ð2Þ
Table 1. Composition of silicone rubber samples.
Composite PDMS (wt%) ATH (wt%) BN (wt%)
ATH30 70 30 0
ATBN1 70 29 1
ATBN3 70 27 3
ATBN5 70 25 5
ATBN7 70 23 7
220 Page 2 of 5 Bull Mater Sci (2020) 43:220
Thermo-gravimetric analysis (TGA) experiments were
conducted using Mettler Toledo TGA-851. The sample
mass was controlled between 10 and 15 mg in a ceramic
crucible of the instrument. The heating was started from
70�C until 800�C with a rate of 10�C min-1. Moreover,
thermal conductivity experiments were performed using a
NETZSCH LFA 447 Nanoflash apparatus. The samples
were prepared by a size of 1 9 1 9 0.1 cm3 and the thermal
conductivity of the samples was calculated by the following
mathematical expression:
K ¼ a� q� CP: ð3Þ
Here a, q and CP represent the thermal diffusivity, den-
sity and specific heat capacity of the target sample,
respectively.
3. Results and discussion
3.1 Dielectric properties
Figure 2a shows the frequency dependency of real permit-
tivity of silicone rubber ATH and BN-filled composites. It
is found that the permittivity is substantially influenced by
the introduction of BN contents in the hybrid samples.
Interestingly, co-filled composites show significantly lower
values of permittivity relative to ATH30. The permittivity of
ATH30 is nearly 3.87 which is higher than 3.64 (ATBN1),
3.70 (ATBN3), 3.83 (ATBN5) and 3.82 (ATBN7) at 1000
Hz. Among ATBN batch, ATBN1 particularly emerges as a
potential candidate with a lower value of permittivity.
Figure 2b shows the frequency dependence of dielectric loss
tangent of silicone rubber composites. It is found that the
samples of ATBN batch have lower dielectric loss than its
counterpart (ATH30). Specifically, the loss tangent of
ATH30 is approximately 0.0224 at 1000 Hz, while the loss
tangent of ATBN1, ATBN3, ATBN5 and ATBN7 is
measured around 0.0086, 0.0140, 0.0116 and 0.0111,
respectively. ATBN1 shows substantial lower dielectric loss
tangent, which seems consistent with the permittivity results
as well. From the results, it is concluded that the addition of
small contents of BN-nanoparticles can enhance the dielec-
tric performance of solely ATH-filled silicone rubber.
3.2 Thermal stability performance
Figure 3a and b shows the TGA and differential thermal
gravimetry (DTG) profiles of silicone rubber composites.
The results exhibit no major degradation of samples until
230�C. Significant degradation is seen initially between 250
and 350�C. This is clearly demonstrated by a sharp dip in
DTG profiles and it is due to the release of ATH’s inherent
moisture contents by its degradation on this temperature. The
thermal degradation data on temperatures of different mass
loss are presented in table 2. The results suggest that no
Figure 1. SEM micrograph of ATBN5.
(a)
(b)
Figure 2. Frequency dependent (a) real permittivity and (b) di-
electric loss tangent of silicone rubber composites.
Bull Mater Sci (2020) 43:220 Page 3 of 5 220
notable variation is found in the temperatures of 5% mass
loss of composites, but pronounced variations are seen in the
temperatures of 10, 15 and 30% mass loss of samples. It is
clearly noted that all the hybrid-filled samples demonstrate
their outstanding thermal stability performance relative to
ATH30. Among the ATBN batch, ATBN5 shows impressive
resistance to degradation. The temperatures of 10 and 15%
mass loss of ATBN5 are measured at 496 and 740�C,
respectively, and they are significantly higher by 31 and
200�C relative to ATH30. Tmax represents the thermal
depolymerization of main siloxane chain [12] and it is
measured higher in ATBN samples. Moreover, char residue
is another indicator that gives information on the thermal
stability of composites [13]. Char residue percentage is seen
consistent with the mass loss temperatures of the composites.
3.3 Thermal conductivity
Figure 4a illustrates the effect of different nano-BN doping
on the variations in the thermal conductivity of co-filled
composites. It is a well-known fact that the addition of
particles enhances the thermal conduction capability of the
polymeric composites. The results exhibit that thermal
conductivity is linearly increased with the addition of a
hybrid set of particles to some extent. The thermal con-
ductivity of ATH30 is measured at 0.287 W m-1 K-1.
Interestingly, all hybrid samples demonstrate excellent
thermal conduction performance relative to ATH30. Among
ATBN samples, ATBN5 emerges as a potential candidate
with higher thermal conductivity of 0.346 W m-1 K-1
which is 20% higher than ATH30.
3.4 Discussion
Experimental findings suggest that dielectric and thermal
properties of solely ATH-filled silicone rubber can be
enhanced by fabricating coalesced network of micron-
ATH/nano-BN in silicone rubber. It is well known that
dielectric permittivity of the composites is enhanced with
the addition of particles due to interfacial polarization
between the additives and polymer matrix and dipole
polarization of additives [5]. Moreover, additives enhance
the electrical conductivity of composites and introduce the
electrical charge carriers and increase the dielectric loss
[14]. ATBN1 has emerged as a candidate with excellent
permittivity and dielectric loss in this study. It is highly
likely that nano-BN additives may fill the gaps between
micron-ATH particles in the hybrid samples, but only 1 wt%
of BN in ATBN1 shows a great deal of phase interaction
which could be the reason for excellent permittivity results.
Subsequently, this ideal phase interphase may strict the
(b)
(a)
Figure 3. Thermal stability of silicone rubber composites by
(a) TGA and (b) DTG curves.
Table 2. Information on thermal degradation of silicone rubber composites.
Sample T5% (�C) T10% (�C) T15% (�C) T30% (�C) Tmax (�C) Char residue (%)
ATH30 327 465 540 672 500 67
ATBN1 330 474 604 762 501 69
ATBN3 331 493 710 * 510 72
ATBN5 330 496 740 * 509 74
ATBN7 329 483 681 * 511 71
*Degradation is not achieved.
220 Page 4 of 5 Bull Mater Sci (2020) 43:220
chain mobility in the composites and it will reduce the
electrical conductivity of ATBN1. Hence, this mechanism
will restrict the transport of mobile charge carriers in the
silicone rubber composites.
Furthermore, thermal conductivity is enhanced by the
development of heat flow pathways in the composites. In
hybrid composites, it is believed that nano-BN particles
may cover the gaps between ATH particles by forming a
bridging network and provide more heat flow paths, as
shown in figure 4b. It is likely that 5 wt% of BN in ATBN5
may show a great deal of denser packing geometry and
enhance phonon transport [15]. Moreover, the thermal
conductivity of ATBN7 is marginally reduced relative to
ATBN5. It is likely that additional BN contents may form
more agglomerates in the hybrid structure and reduce the
thermal conductivity of ATBN7.
4. Conclusions
Dielectric, thermal stability and thermal conductivity of
silicone rubber is studied in this work by fabricating the
coalesced network of micron-ATH/nano-BN hybrid set of
particles. Results suggest that the addition of BN in the
ATH-filled silicone rubber significantly alters the dielectric
and thermal properties and the improvement is greater than
that achieved through solely 30 wt% ATH-filled composite
(ATH30). ATBN1 proclaims improved dielectric perfor-
mance relative to ATH30 with lower permittivity and
dielectric loss of 3.64 and 0.0086 at 1000 Hz, respectively.
Moreover, it is concluded that all the hybrid-filled samples
show their outstanding thermal stability and thermal con-
duction performance relative to ATH30. ATBN5 confirms
higher temperatures of 10 and 15% mass loss those are
measured higher by 31 and 200�C relative to ATH30,
respectively. ATBN5 shows higher conductivity of
0.346 W m-1 K-1 which is seen 20% higher than ATH30.
References
[1] Nazir M T, Jiang X and Akram S 2012 Int. J. Electr.Comput. Sci. 12 1
[2] Huang X, Chunyi Z, Pingkai J, Dmitri G, Yoshio B and
Toshikatsu T 2013 Adv. Funct. Mater. 23 1824
[3] Ali M and Hackam R 2008 IEEE Trans. Dielectr. Electr.Insul. 15 1368
[4] Yu S, Li S, Wang S, Huang Y, Nazir M T and Phung B T
2018 IEEE Trans. Dielectr. Electr. Insul. 25 1567
[5] Zha J W, Dang Z M, Li W K, Zhu Y H and Chen G 2014
IEEE Trans. Dielectr. Electr. Insul. 21 1989
[6] Qi X D, Wang W Y, Xiao Y J, Huang T, Zhang N, Yang J H
et al 2019 Compos. Commun. 13 30
[7] Nazir M T, Phung B T, Zhang Y and Li S 2018 Micro NanoLett. 14 150
[8] Cai X, Dong X, Lv W, Ji C, Jiang Z, Zhang X et al 2020
Compos. Commun. 17 163
[9] http://www.dcproducts.com.au/RTV_Silicone_Solutions/Tech_
Data_Sheets/RTV615-tds.pdf
[10] Nazir M T, Phung B T, Yu S, Zhang Y and Li S 2018 HighVolt. 3 289
[11] Nazir M T, Phung B T and Li S 2017 In 2017 IEEE Int’lSym. on Electr. Insul. Mater. (ISEIM) 1 370
[12] Chen X, Song W, Liu J, Chuanmei J and Yi Q 2015 J.Therm. Anal. Calorim. 120 1819
[13] Nazir M T, Phung B T, Yu S and Li S 2018 IEEE Trans.Dielectr. Electr. Insul. 25 2076
[14] Qi W and Chen G 2014 IEEE Trans. Dielectr. Electr. Insul.21 1809
[15] Zhou W, Wang C, An Q and Ou H 2008 J. Compos. Mater.42 173
(a)
(b)
Figure 4. (a) Thermal conductivity of silicone rubber compos-
ites and (b) schematic of thermal conduction in a coalesced
network.
Bull Mater Sci (2020) 43:220 Page 5 of 5 220