Journal of Physics D: Applied Physics
ACCEPTED MANUSCRIPT
Influence of moisture absorption on electrical properties and charge
dynamics of polyethylene silica-based nanocomposite
To cite this article before publication: Yan Wang et al 2018 J. Phys. D: Appl. Phys. in press https://doi.org/10.1088/1361-6463/aadb7b
Manuscript version: Accepted Manuscript
Accepted Manuscript is “the version of the article accepted for publication including all changes made as a result of the peer review process,
and which may also include the addition to the article by IOP Publishing of a header, an article ID, a cover sheet and/or an ‘Accepted
Manuscript’ watermark, but excluding any other editing, typesetting or other changes made by IOP Publishing and/or its licensors”
This Accepted Manuscript is © 2018 IOP Publishing Ltd.
During the embargo period (the 12 month period from the publication of the Version of Record of this article), the Accepted Manuscript is fully
protected by copyright and cannot be reused or reposted elsewhere.
As the Version of Record of this article is going to be / has been published on a subscription basis, this Accepted Manuscript is available for reuse
under a CC BY-NC-ND 3.0 licence after the 12 month embargo period.
After the embargo period, everyone is permitted to use copy and redistribute this article for non-commercial purposes only, provided that they
adhere to all the terms of the licence https://creativecommons.org/licences/by-nc-nd/3.0
Although reasonable endeavours have been taken to obtain all necessary permissions from third parties to include their copyrighted content
within this article, their full citation and copyright line may not be present in this Accepted Manuscript version. Before using any content from this
article, please refer to the Version of Record on IOPscience once published for full citation and copyright details, as permissions will likely be
required. All third party content is fully copyright protected, unless specifically stated otherwise in the figure caption in the Version of Record.
View the article online for updates and enhancements.
This content was downloaded from IP address 152.78.209.172 on 21/08/2018 at 17:24
IOPPublishing JournalTitle
JournalXX(XXXX)XXXXXX https://doi.org/XXXX/XXXX
xxxx‐xxxx/xx/xxxxxx 1 ©xxxxIOPPublishingLtd
Influence of Moisture Absorption on Electrical
Properties and Charge Dynamics of Polyethylene
Silica‐BasedNanocomposites
YanWang,DayuanQiang,FuadN.F.Alhabill,ZhiqiangXu,GeorgeChenandAlun
Vaughan
1
The Tony Davies High Voltage Laboratory, University of Southampton
E-mail: yw14g13@soton.ac.uk
Received xxxxxx
Accepted for publication xxxxxx
Published xxxxxx
Abstract
The use of nanocomposites as dielectric materials is expected to lead to improved electrical
performance. However, recent research has shown that moisture absorption can cause a
deterioration in the electrical performance of nanocomposites. Although it is generally
accepted that hydroxyl groups attached to nanoparticle surfaces are the main cause of
moisture absorption, the impact of this absorption on the electrical properties of
nanocomposites is still not fully understand. In this paper, a series of measurements,
including thermogravimetric analysis, DC breakdown, surface potential decay and space
charge, are conducted with the aim of determining the impact of moisture absorption on the
electrical properties of polyethylene/silica nanocomposites. The results show that the loading
ratio of nanosilica and the humidity of the conditioning environment determine the amount of
absorbed moisture. According to the Zhuravlev model, the main contribution to the
deterioration in electrical properties of nanocomposites comes from the large amount of
moisture absorbed in multilayer form. It is found that the loading ratio of nanosilica is the
most significant factor in reducing DC breakdown strength.
Keywords: nanocomposite; moisture absorption; polyethylene; nanosilica; DC breakdown; charge dynamics
1.Introduction
The concept of a nanodielectric was first proposed in 1994
[1], and such materials have attracted considerable attention
during the following two decades. A number of interesting
results have been obtained showing the desirable electrical
properties of nanodielectrics. For example, magnesium
oxide/low-density polyethylene (LDPE) nanocomposites
exhibit improved DC breakdown strength and effectively
suppressed space-charge formation compared with neat
LDPE [2]. However, it has been found that absorption of
moisture by the incorporated nanoparticles can lead to a
deterioration in the electrical properties of nanocomposites.
This moisture absorption is generally related to the aspect
ratio and surface chemical groups of the nanoparticles [3,4].
For instance, nanosilica-based cross-linked polyethylene
(XLPE) nanocomposites are able to absorb a greater amount
of moisture than the host XLPE owing to the hydroxyl
groups attached to the surface of the nanosilica particles,
leading to a decrease in AC breakdown strength, an increase
in space-charge formation, and a reduction in water tree
aging of wetted specimens [5]. The results for
polyethylene/nanosilica indicate that moisture absorption can
lead to increases in permittivity and loss tangent [4]. For
polyimide/aluminium oxide nanocomposites, a reduction in
breakdown strength and facilitation of surface potential
decay due to moisture absorption have been reported [6].
Page 1 of 10 AUTHOR SUBMITTED MANUSCRIPT - JPhysD-116264.R3
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript
JournalXX(XXXX)XXXXXX Authoretal
2
Ethylene-vinyl acetate incorporated with organically
modified nanofiller clay showed worsened electrical
properties owing to moisture absorption dependent on the
aspect ratio of the nanofiller, and it was found that a higher
aspect ratio led to greater moisture absorption [3]. Moreover,
the aspect ratio of the nanoparticles strongly affects the stress
distribution within the matrix reinforcement, and this stress
can further affect the diffusion rate of absorbed moisture [7-
9].
Present understanding of moisture absorption behaviour
in nanocomposites is incomplete because, although moisture
can be present in several forms on the nanoparticle surface,
this has not been considered carefully in previous
investigations. Moreover, although there have been many
experimental and theoretical investigations of the impact of
moisture absorption on the electrical properties of
nanocomposites, a reasonable description that links moisture
absorption with electrical properties and charge dynamics is
still lacking. For example, percolation theory has been used
to model and explain the function of the water shell in charge
dynamics [5,10], and the results obtained can be used to
estimate the water shell size in terms of nanoparticle loading
ratios, but this approach cannot deal with the impact of
different forms of moisture on charge transport in
nanocomposites from the perspective of the different
chemical groups attached to the nanoparticle surface.
In this paper, the influence of moisture absorption on the
electrical properties of blended polyethylene/silica
nanocomposites containing untreated or
trimethoxy(propyl)silane-treated nanosilica is studied. In this
material system, the nanosilica is more hygroscopic than the
polyethylene. Therefore, moisture absorption is primarily by
the nanoparticles rather than the host polymer. The moisture
absorbed is measured by weighing specimens and finding the
change in mass and by thermogravimetric analysis (TGA).
Experiments based on DC electric fields, investigating space
charge, surface potential decay and breakdown strength, are
carried out. The surface chemical groups of nanosilica and
their influence on absorbed moisture are discussed in terms
of the Zhuravlev model [11]. The purpose of this research is
to reveal the forms in which moisture is absorbed and study
their impact on the DC breakdown strength and charge
dynamics associated with hydroxyl groups attached to the
nanosilica surface.
2.SurfaceChemicalGroupsofNanosilica
Many species of nanoparticles have been reported to
cause moisture absorption, including Si O
, Si
N
, MgO,
Al
O
, BN and nanoclays [3,6,12-16]. The moisture uptake
for each specific nanoparticle may be different, but in all
cases the absorption process can occur either before or after
specimen manufacture when the nanoparticle or
nanocomposite is exposed to a humid environment.
Nanosilica has frequently been used in research on
nanodielectrics because it is widely available in a range of
sizes, from tens to hundreds of nanometres, and also because
of a number of favourable properties, such as a relatively
high breakdown strength and electrical resistivity [14]. In
addition, from an economic perspective, it can be produced
at low cost by flame hydrolysis or polymerisation. However,
these production methods also make nanosilica hydrophilic.
To obtain spherical nanosilica, both flame hydrolysis and
polymerisation processes will hydroxylate the surface of the
silica, generating Si–OH (silanol) groups on this surface
[11,14]. The hydroxyl (OH) group is covalently bonded with
Si atoms on the surface and inside the particles. It is known
that the presence of OH groups on the surface of nanosilica
particles can change the surface properties, depending on the
concentration of these groups [11].
Besides causing moisture absorption, the OH groups are
also responsible for the agglomeration and compatibility of
nanosilica particles in the host materials. Agglomeration is
dependent on the high surface energy of the nanoparticles,
and surface OH groups enhance this tendency through the
formation of a large number of hydrogen bonds among the
nanoparticles. Therefore, the removal of OH groups from the
surface of nanosilica is a widely accepted approach for
reducing agglomeration. Functionalisation is a common
method for forming stable covalent bonds by replacing the
OH groups using a coupling agent. However, it has been
found that changes in the surface chemistry of nanoparticles
will cause a chain reaction in the nanocomposite, which may
not be in line with expectations. For example, nanosilica
processed at a high temperature (1050 °C with dry nitrogen)
can effectively prevent moisture absorption, but it also forms
larger and denser agglomerates than unprocessed nanosilica
[14].
3.ExperimentalDetails
3.1PreparationandConditioningofSpecimens
The nanosilica powder was received from Sigma-Aldrich
with a size range of 10–20 nm. To reduce the amount of
surface OH groups, trimethoxy(propyl)silane (C3) treatment
was applied to the surface of the received nanosilica via an
anhydrous route, and the related Fourier transform infrared
results indicated that the propyl group was successfully
bonded [4]. The methoxy group of the silane coupling agent
reacts with the OH groups on the nanosilica, forming Si–O–
Si bonds attached to the propyl (–C
3
H
7
) functional group.
The nanocomposites incorporating untreated nanosilica will
be referred to as ASR and those with C3-treated nanosilica as
C3. The host polymer used was 20% high-density
polyethylene (HDPE; Rigidex HD5813A, BP Chemical)
blended with 80% LDPE (LD100BW, ExxonMobil
Chemicals). The control group without nanosilica will be
Page 2 of 10AUTHOR SUBMITTED MANUSCRIPT - JPhysD-116264.R3
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript
JournalXX(XXXX)XXXXXX Authoretal
3
referred to as BPE. Four nanosilica loading ratios were used:
0.5, 2 , 5 and 10 wt%. The nanocomposite materials were
manufactured using a solvent method at a temperature of up
to 160 °C [4,12]. Film specimens were melting-pressed with
a hydraulic press at 160 °C. Once removed from the press,
each specimen was placed directly into an oil bath at 115 °C
for 1 hour of isothermal crystallisation. To investigate the
effect of differences in conditioning humidity on moisture
absorption by the nanocomposites, all specimens were
grouped and conditioned at room temperature under
conditions of different relative humidity (RH), including a
vacuum desiccator (RH5), a climate room (RH60) and in
deionised water (immersed condition) for about 9 days.
3.2MeasurementofMoistureAbsorptionandTGA
Moisture absorption by the nanocomposites was measured
by the change in mass, taken as the average value of five
specimens. In addition, a Perkin Elmer Pryis 1 TGA system
was employed to perform TGA for untreated and C3-treated
nanosilica particles that had already been conditioned in a
humid environment (over RH90). The quantity of nanosilica
tested was about 5 mg. The heating rate of TGA was
10 °C/min in dry air over a test range from 50 to 900 °C.
3.3DCBreakdownStrength
DC breakdown measurements were conducted in silicone
oil using spherical electrodes that were changed every 10
tests. The voltage ramping rate was 100 V/s. The specimen
thickness of the nanocomposite films was 70 ± 5 μm. The
results obtained were analysed in terms of the Weibull
distribution.
3.3SurfacePotentialDecay
The schematic diagram of surface potential decay (SPD) was
the same as in the previous study [12]. After negative corona
charging, the first decay reading was taken after 5 s owing to
system delay. In this work, the initial potential was set at 4.8
kV. The corona charging lasted 180 s, and was followed by a
900 s decay period. The thickness of the specimen was 120 ±
5 μm.
3.4SpaceCharge
The pulsed electro-acoustic (PEA) technique was employed
for measuring space-charge behaviour. The equipment used
was the same as in the previous study [17]. The thickness of
the specimen was 120 ± 5 μm. The space charge was tested
for 2 hours.
4.Results
4.1MoistureAbsorptionandTGA
The percentage change in mass for each specimen was
calculated as
M%=
M
d
M
i
M
i
×100%, (1)
where M
d
is the mass measured after conditioning as a
function of the number of days and
is the initial mass of
the specimen. The results are shown in Fig. 1. The changes in
mass of specimens conditioned in RH5 did not changed,
indicating that the greater increase in mass of the specimens
conditioned at RH60 and Immersed was primarily due to
direct contact with moisture. Therefore the results for
specimens conditioned at RH5 are not shown here.
Fig. 1. Moisture absorption of nanocomposites conditioned (a) at RH60 or (b)
by immersion in water.
The changes in mass of the specimens reached a steady
state by 7 days. From Fig. 1 (a) and (b), it can be seen that
the humidity of the conditioning environment directly affects
the amount of moisture intake and the absorption rate, with a
clear distinction between the RH60 and immersed specimens,
especially for nanocomposites with high loading ratios (5 and
10 wt%). The increase in mass of the control, BPE, was
Page 3 of 10 AUTHOR SUBMITTED MANUSCRIPT - JPhysD-116264.R3
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript
JournalXX(XXXX)XXXXXX Authoretal
4
much lower than that of the nanocomposites in both humid
environments, which indicates that nanosilica is the main
component causing moisture absorption in the
nanocomposites. The amount and rate of moisture intake of
nanocomposites with 5 and 10 wt% loading ratios are both
greater than those of nanocomposites with 0.5 and 2 wt%
loading ratios, since a higher proportion of nanosilica leads to
a larger number of OH groups. Hence, apart from the
environmental humidity, the loading ratio of nanosilica also
has a significant effect on moisture absorption. Additionally,
agglomerates with a high aspect ratio have frequently been
reported in nanocomposites with loading ratios higher than 5
wt% [15,18]. These high aspect ratios can further contribute
to a significantly increased moisture intake. For the C3
nanocomposites, moisture absorption can still be seen,
although the amount and rate are smaller than for the ASR
nanocomposites. This indicates that the OH groups on the
surface of nanosilica can be reduced in number by the C3
treatment, but not eliminated completely, and the residual OH
groups are still able to absorb moisture via hydrogen bonding
until a steady state is reached. It should be noted that in the
case of the 5 and 10 wt% nanocomposites, the immersed
specimens were replaced in the vacuum desiccator at room
temperature after their conditioning in deionised water. As
can be seen from Fig. 1(b), their masses were reduced sharply
after this drying process.
Fig. 2. TGA and DTG of ASR and C3 nanosilica.
The results of TGA and derivative thermogravimetric
analysis (DTG) are shown for both ASR and C3 nanosilica
particles in Figs. 2 and 3. Distinct DTG peaks can be seen
from 50 to 200 °C, which are attributed to loss of absorbed
moisture from the surface of the nanosilica. In addition, there
is another shallow but broad DTG peak from 200 to 500 °C,
mainly due to the OH groups on the nanosilica. The absorbed
moisture contributes to the primary weight loss during the
heating process. The TGA behaviour of the ASR nanosilica
is different from that of the C3 nanosilica. For instance, from
50 to 200 °C, the ASR nanosilica exhibits a sharp peak,
whereas the C3 nanosilica exhibits two small peaks. As the
C3 nanosilica is exposed longer to the humid environment,
the first peak slowly comes to make up a large percentage of
the weight loss, as shown in Fig. 3, which reveals that the C3
treatment can reduce moisture absorption compared with the
ASR specimen. Furthermore, from 200 to 500 °C in Fig. 2,
the C3 nanosilica loses more weight and does so even faster
than the ASR nanosilica because it contains chemical groups
that possess lower bond strength but higher molecular weight
than OH, such as carbon chains (C–C bonds).
Fig. 3. TGA and DTG of C3 nanosilica with a longer conditioning time.
On heating from 50 to 900 °C, the weight loss from the
nanoparticles was about 11%. Multiplying the loading ratios
by this percentage gives values close to the changes in mass
of the nanocomposites conditioned by immersion with the
corresponding loading ratios. For example, the M% of 10 wt%
nanocomposites is about 1.25%, and that of 5 wt%
nanocomposites is 0.55%. In addition, according to the DTG
curve, the majority of the weight loss occurred from 50 to
200 °C. The absorbed moisture is mainly responsible for the
change in mass of the nanocomposites, which is in line with
the results of weighing.
4.2DCBreakdownStrength
The results for DC breakdown strength as a function of
nanosilica loading ratio and conditioning humidity are shown
in Fig. 4. Twenty tests were conducted for each type of
specimen and the results were processed according to a
cumulative failure probability of 63.2% based on the Weibull
distribution. The BPE control is represented by 0 wt%. For
the specimens conditioned at RH5, moisture absorption is not
significant. An increase in loading ratio for both ASR and C3
nanosilica leads to a poorer DC breakdown strength
regardless of how much moisture is absorbed by the
Page 4 of 10AUTHOR SUBMITTED MANUSCRIPT - JPhysD-116264.R3
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Accepted Manuscript