Nanotechnology

ACCEPTED MANUSCRIPT

Thermally and electrically conductive multifunctional sensor based on

epoxy/graphene composite
To cite this article before publication: Sensen Han et al 2019 Nanotechnology in press https://doi.org/10.1088/1361-6528/ab5042

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 © 2019 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 192.236.36.29 on 23/10/2019 at 15:21

Page 1 of 25 AUTHOR SUBMITTED MANUSCRIPT - NANO-122604.R2

1
2 Thermally and electrically conductive multifunctional
3
4 sensor based on epoxy/graphene composite
5
6 Sensen Han1,2, Aron Chand1, Sherif Araby1,3,4*, Rui Cai5, Shuo Chen6, Hailan Kang7,
7
8 Rongqiang Cheng1, Qingshi Meng1,2*

pt
9
10 1
College of Aerospace Engineering, Shenyang Aerospace University, Shenyang 110136, China
2
11 Shenyang Aircraft Design Institute, Shenyang 110136, China
3
12 Department of Mechanical Engineering, Benha Faculty of Engineering, Benha University, Benha, Egypt
4
13 School of Engineering, University of South Australia, SA, 5095, Australia

cri
5
14 School of Mechanical, Aerospace and Automotive Engineering, Coventry University, Coventry, UK
6
College of Medicine and Biological Information Engineering, Northeastern University, Shenyang 110819, Liaoning, China
15 7 School of Materials Science and Engineering, Shenyang University of Chemical Technology, Shenyang, China
16
17 *
Corresponding Authors： Sherif Araby: SherifAraby.Gouda@unisa.edu.au
18
19
Qingshi Meng; Email: mengqingshi@hotmail.com
20

us
21
22
23 Abstract
24
25 Flexible electronics is expected to be one of the most active research areas in the next decade. In this
26
27
28
an
study, a mechanically strong and flexible epoxy/GnP composite film was fabricated having

29 percolation threshold of electrical conductivity at 1.08 vol% GnPs and high thermal conductivity as
30
31 1.07 W m-1 K-1 at 10 vol% GnPs. The composite film shows high mechanical performance: Young’s
32
dM
33 modulus and tensile strength were improved by 1344% and 66.7%, respectively at 10 vol%. The film
34
35
has demonstrated high sensitivity to various mechanical loads: (i) it has gauge factors of 2 at strain
36
37
38 range 0–7% and 6 at range 7–10% (ii) it gives good electrical response with bending and twisting
39
40 angles up to 180o; and (iii) it displays a good compressive load response up to 2N where the absolute
41
42 value of electrical resistance change increased by 71%. Furthermore, the film has showed an excellent
43
pte

44 reliability up to 5.5×103 cycles with minor zero-point error. Above 20 ºC, the film solely acts as a
45
46 temperature sensor; upon cyclic temperature testing, the film demonstrated a stable resistive response
47
48 in the range of 30 −75 ºC with a temperature sensitivity coefficient of 0.0063℃-1. This flexible
49
50 composite film has remarkable properties that enable it to be used as a full-fledged sensor for
51
ce

52
universal applications in aerospace, automotive and civil engineering.
53
54
55 Key words: Graphene; Epoxy; Electrical and thermal conductivity; Strain; Sensor
56
Ac

57
58
59
60
1
 AUTHOR SUBMITTED MANUSCRIPT - NANO-122604.R2 Page 2 of 25

1
2 1. Introduction
3
4
5 Recently, high-performance flexible multifunctional polymer composite films have attracted
6
7
8
extensive attention in both academia and industries for potential applications in automobile, energy,

pt
9
10 bio-health monitoring and communication industries [1, 2]. When polymers are combined with
11
12 electrically and/or thermally conductive nanofillers, a polymeric nanocomposite films featuring thin,
13

cri
14
lightweight, flexible, wearable, and electric and thermal properties are developed. These developed
15
16
17 composites could have potential in sensing signals depending on the attained features.
18
19
20
Strain sensors refer to a class of electronic devices, which transduce mechanical deformations into

us
21
22 electrical signals. They are highly desirable in wearable devices to record human physical motions
23
24 for therapy and health condition monitoring such as heart rate monitoring. Soft stretchable strain
25
26
sensors are increasingly demanded to measure movement on stretchable and curved surfaces (wrist
27
28
29
an
patch on skin) and the developed hoop strain on curved pressure vessels.
30
31
32
Polymers are intrinsically poor in electrical and thermal conductivity. Adding a conductive phase
dM
33
34 (fillers) into the polymeric matrix is one way to develop conductive polymer composites. The host
35
36 polymer is transformed into conductive composite when the concentration of the conductive phase
37
38
exceeds the percolation threshold where the conductive fillers overlap and contact each other
39
40
41 throughout the matrix. Another mechanism to understand the conduction behavior in a polymeric
42
43 composite is through tunneling-transfer mechanism where the conductive nanofillers are very close
pte

44
45 but not touching each other; minimum distance between two conductive fillers to observe tunneling
46
47
48 effect is 1.8 nm [3].
49
50
Various conductive materials including metal nanostructures (gold nanoparticles, gold and silver
51
ce

52
53 nanowires) [4-6] and carbon-based nanomaterials ‒ graphene [7-9] and carbon nanotubes (CNTs) [10,
54
55 11]‒ and carbon black [12] ‒ were investigated to develop multifunctional polymer-based composite
56
Ac

57 films. Metal-based sensors normally have poor stretchability and undesirable heavy weight due to its
58
59
60 high density. Graphene is highly electrically and thermally conductive material with outstanding
2
Page 3 of 25 AUTHOR SUBMITTED MANUSCRIPT - NANO-122604.R2

1
2 mechanical properties: specific surface area 2630 m2 g−1, Young’s modulus ~1 TPa, tensile strength
3
4 ~130 GPa and electrical conductivity up to 6000 S/cm which are close to those of CNTs [13].
5
6
7 Graphene platelets (GnPs) are few graphene layers (~3) stacked together with an average thickness
8

pt
9 of 3.55±0.32nm [14, 15]. In our previous work, GnPs were fabricated by thermal shocking graphite
10
11 intercalated compound (GIC) at 700℃ following by ultrasonication [15, 16]. GnPs proved to
12
13
markedly enhance mechanical, electrical and thermal properties of various types of polymers when

cri
14
15
16 they added into them [8, 15, 16].
17
18
19 Polymeric composite films and sensors developed using carbon nanotubes and graphene, possess
20

us
21 light weight, flexibility, stretchability and promising sensitivity and durability compared to metallic
22
23 nanoparticles. Plethora studies have investigated the structure-property relations of polymer/graphene
24
25
26 and CNT nanocomposites and films [3, 17-23]. The minimum concentration of CNTs in a polymer
an
27
28 needed to reach percolation threshold (φth) was theoretically calculated suggesting that only 3.33%
29
30 of CNTs contribute to the conductive network at φth [18]. This is mainly because CNTs have high
31
32
attracting force (van der Waal force) along the tube surface leading to agglomeration and bundling
dM
33
34
35 inside the matrix. Also, it is important to mention that polymer properties, nanofiller conductivity and
36
37 its structure characteristics, dispersion quality, and processing conditions are key factors in
38
39 determining the composite-based sensor behavior [19, 20].
40
41
42 Lubineau G. et al [3] modified multi-walled CNT (MWCNT) by coating them with a conductive
43
pte

44
polymer, poly(3,4-ethylenedioxythiophene)poly(styrenesulfonate), and added to polycarbonate
45
46
47 matrix. The electrical resistivity was dropped by 11 orders when 1 wt% of modified MWCNT added
48
49 compared to 8 orders when pristine MWCNTs were added at the same fraction. This confirms the
50
51
ce

crucial role of high degree of filler dispersion that was attained by modification via coating with
52
53
54 conductive polymer. However, the piezoresistive response was deteriorated upon coating. The
55
56 piezoresistive phenomena is one of the outstanding property of MWCNTs which makes them ideal
Ac

57
58 candidate to develop polymer-based strain sensors [24]. Gau C. et al [22] succeeded in developing a
59
60
3
 AUTHOR SUBMITTED MANUSCRIPT - NANO-122604.R2 Page 4 of 25

1
2 promising pressure sensor composite containing polyimide and MWCNTs. The composite possessed
3
4 highly linear piezoresistive and rapid response, high sensitivity and thermal stability which make the
5
6
7 developed sensor competitive to a commercial pressure sensor. Hu N. et al. [19] developed
8

pt
9 epoxy/CNT based strain sensors with gauge factor (sensitivity) up to 25. They also emphasized that
10
11 higher tunneling resistance can raise the sensitivity of the composite sensor. In recent study [25],
12
13
epoxy/graphene platelet strain sensor was developed possessing a gauge factor of 22.54. Initially, the

cri
14
15
16 sensors were mainly applied in strain sensing [26, 27]. However, with advancement in technology
17
18 and fabrication methods, the sensors now have the ability to detect other parameters such as exposure
19
20
to gas [28], ability to detect pressure [29] and infrared detection [30].

us
21
22
23 Most of previous studies develop nanocomposite which focuses on sensing one parameter–
24
25
26 unifunctional sensor–especially strain sensing. Limited studies have been conducted on developing
an
27
28 epoxy/GnP-based sensor which has the potential to simultaneously detect multiple parameters
29
30 (pressure, temperature and strain). Hence, this research stands out from other previous work in
31
32
developing an efficient and multifunctional sensor based on epoxy/GnP nanocomposite.
dM
33
34
35 In this work, a mechanically strong, flexible and highly conductive (electrically and thermally)
36
37
38 epoxy/GnP composite film sensor is fabricated. It is not only limited to sensing strain, but also able
39
40 to respond to angular movement. The developed sensor has high reliability with zero-point error up
41
42 to 5.5×103 cycles. Additionally, our composite film can efficiently work as a temperature sensor at
43
pte

44
range 30‒70℃.
45
46
47
48
2. Experiment details
49
50
51
2.1 Materials
ce

52
53
The expandable graphite (graphite intercalation compound, GIC, Asbury 1721) was provided by
54
55
56 Asbury Carbons, Asbury, NJ, USA. Epoxy – (diglycidyl ether of bisphenol A; WSR618, 184–200 g
Ac

57
58 per equiv.) was purchased from Nantong Xingchen Synthetic Material. The hardener, Jeffamine D-
59
60 2000 (J2000) was purchased from Huntsman, China.
4
Page 5 of 25 AUTHOR SUBMITTED MANUSCRIPT - NANO-122604.R2

1
2 2.2 Fabrication of composite films
3
4
5 The preparation of GnPs is fully described elsewhere [31]. In brief, 1 g GIC was expanded by
6
7 transferring it into a preheated crucible at 700℃ in a common furnace; after 1 min, the crucible was
8

pt
9
10 taken out and cooled down. The expansion ratio of GIC was calculated ~200 times by volume [15,
11
12 32]. Using a magnetic stirrer, a determined weight of the expanded GIC was suspended in acetone at
13

cri
14 1 wt% in a metallic container for 10 min followed by 1 hour ultrasonication below 25℃; the low
15
16
17 temperature of ultrasonication enhances the delamination and suspension of the expanded GIC [33].
18
19 Upon completion of sonication, the end-product is graphene platelets (GnPs) [16, 31, 34]. The mixture
20

us
21 will be used for the following preparation of epoxy/GnP composite film.
22
23
24 Epoxy resin was added and mixed with GnPs/acetone suspension by magnetic stirrer for 10 min to
25
26 reach dissolving. The mixture was then sonicated for 30 min at room temperature and then sat on a
27
28
29
an
hot plate at 70℃ with magnetic stirring till approximately 70% (by volume) of acetone was
30
31 evaporated. The mixing and evaporation processes were performed in a Pyrex Griffin beaker, 500ml
32
dM
33 and 1000ml. When the mixture cooled down to room temperature, the hardener–J2000 was slowly
34
35
added to the epoxy/GnP solution and manually stirred using a glass bar. The entire mixture was then
36
37
38 dropped into a polypropylene mould and left for 12 hrs at room temperature producing semi-cured
39
40 film and meanwhile the remaining acetone is evaporated. The semi-cured film is transferred to an
41
42 oven at 120℃ for 8 hours for full curing and removing any traces of acetone. The fabrication process
43
pte

44
45 is summarized schematically in Figure 1. It is noteworthy to mention that all composite constituents
46
47 were weighed in order to achieve a certain volume percentage. Equation 1 was used to convert weight
48
49 percent to volume percent where density of epoxy and GnPs are defined as 1.1 and 2.26 g/cm3 ,
50
51
ce

52 respectively [32].
53
54 Vf =
⍴m W f Eq.1
55 ⍴f (1 − Wf ) + ⍴m Wf
56
Ac

57 where ⍴ and w are the density and weight fraction, respectively; subscript m refers to epoxy and
58
59 hardener together, and f refers to GnPs.
60
5
 AUTHOR SUBMITTED MANUSCRIPT - NANO-122604.R2 Page 6 of 25

1
2
3
4
5
6
7
8

pt
9
10
11
12
13

cri
14
15
16
17 Figure 1. The fabrication process of epoxy/GnP composite film.
18
19 2.3 Characterizations
20

us
21 Transmission electron microscopy (TEM) Philips CM200, was used to study the morphology of
22
23 graphene platelets (GnPs) at 200 kV accelerating voltage. The samples were prepared by dispersing
24
25
26
graphene platelets into acetone via sonication with a subsequent dilution until it reaches concentration
an
27
28 of 0.0004 wt%. Few drops of the 0.0004 wt% were dropped on 200-mesh copper grids then dried.
29
30 The microstructure of the composite films was examined using scanning electron microscopy (SEM)
31
32
(JEOL JSM-7800F) at 5 kV accelerating voltage. A thin layer of platinum was used to coat the surface
dM
33
34
35 of the sample. Raman spectra was collected using Renishaw inVia Raman spectrometer and 633nm
36
37 laser excitation.
38
39
40 Graphene platelet (GnP) powder was produced as follow. After the expanded GIC was ultrasonicated
41
42 for one hour in acetone, the acetone was separated from the sonicated mixture using vacuum filtration.
43
pte

44
The wet product left overnight in a ventilated oven at 70℃ to completely remove any traces of the
45
46
47 solvent. The produced powder was GnP powder which was pressed using bench-top hydraulic press
48
49 to form a thin sheet of overlapped GnPs with thickness of ~ 0.5mm. To measure the intrinsic electric
50
51
ce

conductivity of GnP-sheet, four-probe method was used to measure the in-plane electrical
52
53
54 conductivity by commercial DC four-probe resistance measuring apparatus (RTS-8, 4 PROBES
55
56 TECH Co., Ltd., China). The four probes were gently pressed onto the sheet to acquire good contact.
Ac

57
58 The probes were by default separated ~1mm distance. The measurement current was DC and set at 1
59
60
6
Page 7 of 25 AUTHOR SUBMITTED MANUSCRIPT - NANO-122604.R2

1
2 mA. The four-probe method determines the sheet resistance as Rs (Ω/□). The bulk conductivity is
3
4 calculated according to Eq.2 [35]
5
6
7
8
1 Eq.2
σ (S⁄cm) =

pt
9 R s (Ω⁄□) × t(cm)
10
11
12 where σ and t are respectively the bulk conductivity and sheet thickness. The measurement was
13

cri
14
15 repeated over three prepared sheets and bulk resistivity was averaged.
16
17
The through-plane electrical volume resistivity of the developed epoxy/GnP composite was measured
18
19
20 using Agilent 4339B high-resistivity meter connected with a 16008B resistivity cell. The volume

us
21
22 resistivity measurement was performed in accordance to ASTM D257-99. The Agilent resistivity
23
24 meter was set at 5V and the electrical current was DC type. Three measurements were conducted for
25
26
27
28
29
an
each fraction to obtain the average. Each sample was 5±1.1 mm in thickness and 26 ± 0.5 mm in

diameter. The thickness of the composite was measured by a digital micrometer.

30
31
32 Thermal conductivity of the composite was measured using TA Instruments DTC300 analyzer. The
dM
33
34 test samples were placed between the upper and lower heating plates where the temperature difference
35
36 is 30℃. The lower plate is part of the calibration heat flowmeter. An axial temperature gradient is
37
38
39 developed when the heat transferred from the upper plate through sample to lower plate. The test was
40
41 in accordance with the ASTM E1530 standard.
42
43
pte

44 The piezoresistive property is the change of the electrical resistance of a material upon applying
45
46 mechanical strain. The piezoresistive performance of the composite film was evaluated by conducting
47
48 different types of mechanical loads including tensile, compressive, torsion, bending and thermal
49
50
51 strains. The corresponding electrical resistance change in all mechanical loads was recorded using a
ce

52
53 FLUKE 2638A data acquisition system at current source and electrical potential of 100 μA and 12 V,
54
55 respectively. Details of the applied mechanical thermal loads are described as follows.
56
Ac

57
58 Mechanical properties and piezoresistive property of the composite film under tensile loading were
59
60 measured using a quasi-static uniaxial tensile test. It was carried out using a universal tensile testing
7
 AUTHOR SUBMITTED MANUSCRIPT - NANO-122604.R2 Page 8 of 25

1
2 machine (XIANGMIN) equipped with 2 kN load cell and ran at cross-head speed 0.5 mm/min. Stress-
3
4 strain curves were plotted using the collected data and thus Young’s modulus, tensile strength and
5
6
7 elongation at break were determined. The test was replicated five times for each fraction. The
8

pt
9 recorded mechanical properties were the average values. While the tensile testing was running, the
10
11 change in electrical resistance was determined. Two identical-copper wires were attached to the
12
13
composite film by silver past and the other ends of the wires were connected to FLUKE data-

cri
14
15
16 acquisition system to simultaneously record the resistance change upon tensile strain.
17
18 The piezoresistive response of the composite film upon compressive strain was performed using a
19
20
manual hydraulic press. Pressure was exerted upon the film by placing it between two flat surface in

us
21
22
23 the press, the upper surface of the press is manually moved till touching the film surface at which the
24
25 pressure is 0. Then the pressure was step-wisely increased to 2N with 0.5N increment. Similar to the
26
27
28
29
an
tensile test, the film was connected to a FLUKE data-acquisition system to simultaneously record the

30 resistance change with pressure increment at room temperature.

31
32 Thermal strain was applied over the epoxy/GnP composite film by placing the composite film inside
dM
33
34 a program-controlled temperature medium. The temperature range was programmed to work from -
35
36
37 20 to 110 °C at a rate of 5 °C/min. The resistance was monitored using the FLUKE data acquisition
38
39 unit. However, we noticed that resistance does not change much at temperature range -20‒20°C.
40
41 Therefore, the plotted data are from room temperature to 110°C.
42
43
pte

Durability and reliability of the epoxy/GnP film was determined by fatigue test. It was conducted
44
45
46 under tensile load mode using the universal tensile testing machine (XIANGMIN) equipped with 2
47
48 kN load cell. The fatigue test was performed within strain rage 0‒5% at frequency of 1 Hz. The cyclic
49
50 loading was carried out for 10,000 cycles. The resistance across the specimen during cyclic testing
51
ce

52
53 was monitored using FLUKE data acquisition unit. Figure 2. Summarizes fabrication process and
54
55 piezoresistive measurements carried out over the 10 vol% epoxy/GnP composite film.
56
Ac

57
58
59
60
8
Page 9 of 25 AUTHOR SUBMITTED MANUSCRIPT - NANO-122604.R2

1
2 (c)
3 Fabrication of Epoxy/GnP film
4 (a)
5
6
7
8

pt
9
10
11
12
13

cri
14
15
16 (b)
17
18 Mechanical test
19
20

us
21
22
23
24
25
26
27
28
29
an
30
31
32
dM
33 (d) (e)
34
35
36
37
38
39
40
41
42
43
pte

44
45
46 (f)
47 (g)
48
49
50
51
ce

Temperature cycle sensing

52
53
54
55
56
Ac

57
58
59
Figure 2. Complete digital-photo chart representing fabrication and measurement processes
60
9
 AUTHOR SUBMITTED MANUSCRIPT - NANO-122604.R2 Page 10 of 25

1
2 3. Result and discussion
3
4 3.1 Morphology of graphene platelets
5
6 Transmission electron microscopy
7
8 Figure 3a–c shows representative micrographs of GnPs. In image (a) giant piece of plate-like structure

pt
9
10
is found, they are layers of graphene stacked together. When part of the layer is magnified in image
11
12
13 (b), edges of individual layers are observed confirming that our GnP might compromise of ~ 5 layers.

cri
14
15 Some other areas are bright and featureless (red circle) which mean this area contains one or two
16
17 layers. Also, the blue arrows show the common wrinkled morphology of the thermally produced
18
19
20 graphene in image (b) [14, 15]. A randomly selected area at red circle was selected to study the

us
21
22 electron diffraction (ED) pattern. Image (c) contains a typical highly crystalline structure of graphene.
23
24
25 Raman spectroscopy
26
27
28
29
an
The graphitic structure and integrity of graphene platelets were studied and compared to graphite

30 intercalation compounds (GIC) by Raman spectroscopy as shown in Figure 3d. The absorptions of
31
32 D, G and 2D bands of GnPs were observed at 1360, 1580 and 2723cm-1, respectively confirming the
dM
33
34
35
graphitic characteristics of the synthesized GnPs. Similar absorptions were noticed for GIC with
36
37 marginal shifts. In graphene, the intensity of the D-band refers to in-plane sp3 hybridized carbon
38
39 structure associating with quantity of the disordered structure and defects such as vacancies due to
40
41 oxidation and reduction, and grain boundaries. On the other hand, G-band corresponds to the ordered
42
43
pte

44 structure of sp2-hybridized carbon atoms. Thus, the ID/IG ratio quantifies the degree of structural
45
46 integrity of GnPs which was calculated as 0.08 compared to ~1.0 in case of reduced graphene oxide
47
48 [31]. This indicates that our GnPs possess high structural integrity reflecting high electrical
49
50
51 conductivity and mechanical performance. The ID/IG ratio of GIC (0.25) depicts that our starting
ce

52
53 material has fair structural integrity with low oxidation degree yielding GnPs with high structural
54
55 integrity. Since the synthesized GnPs possess more than 4 layers, it is difficult to distinguish its 2D-
56
Ac

57
58
band from graphite [36, 37]. However, the GnPs’ thickness was confirmed by atomic force microcopy
59
60 in our previous studies [16, 32, 34, 38].
10
Page 11 of 25 AUTHOR SUBMITTED MANUSCRIPT - NANO-122604.R2

1
2
3 (a)
(a) (b)
4
5
6
7
8

pt
9
10
11
12
13

cri
14 (d)
15
16
17
18
19
20

us
21
22
23
24
25
26
27
28
29
an
30
31
32
dM
33 Figure 3. (a,b) TEM images of GnPs, (c) electron diffraction pattern selected from the red circle, and (d) Raman spectra
34 of graphene platelets (GnPs) and graphite intercalation compounds (GIC)
35
36 3.2 Electrical conductivity of GnPs
37
38
39
The intrinsic electrical conductivity of graphene platelet sheet measured as 1460 S/cm using the four-
40
41 probe method and Eq.2. For comparison, the electric conductivity of a monolayer graphene produced
42
43 by chemical vapor deposition is 2000 S/cm [39]. Due to the overall high contact resistance between
pte

44
45 the overlapped sheets, the GnP electrical conductivity is lower than that of monolayer graphene.
46
47
48 However, GnP electrical conductivity is far higher than that of graphene oxide (0.05 S/cm),
49
50 chemically reduced graphene oxide (5 S/cm) or thermally reduced graphene oxide (350 S/cm) [40].
51
ce

52 This conforms the results of Raman test. Taking into account the scalable production of GnPs and its
53
54
55 cost-effectiveness, little difference with conductivity of monolayer graphene and the absolute non-
56
conductive nature of most polymers, GnPs are promising fillers to develop electrically conductive
Ac

57
58
59 polymer composites at low cost.
60
11
 AUTHOR SUBMITTED MANUSCRIPT - NANO-122604.R2 Page 12 of 25

1
2 3.3 Composite films properties
3
4
5 Electrical conductivity of the composite films
6
7 In this study, GnPs have substantially enhanced the electrical conductivity of the epoxy composites.
8

pt
9 At low fractions of GnPs, the composites are expected to have high volume resistivity. Therefore, the
10
11 electrical volume conductivity of composites containing GnP in range 0 ‒3 vol% was measured by
12
13
Agilent 4339B high-resistivity meter where displayed in Figure 4a. Figure 4a displays that epoxy

cri
14
15
16 conductivity does not substantially change within the low fractions (vol%<0.75%). Once the GnP
17
18 content is beyond 1 vol%, the electric conductivity of epoxy/GnP composite exponentially increases
19
20

us
21
transforming the epoxy matrix from non-conductive material into a semi-conductive one whose
22
23 inherited from the intrinsic superb electrical conductivity of GnPs. For example, the electrical
24
25 conductivity of neat epoxy augmented from 10-14 to 10-7 S/cm at 3 vol% recording 7 orders increase.
26
27
28
29
an
Putting the measurements into the power low equation (Eq.3) gives more understanding when the

30 percolation threshold is formed [16, 41].

31
32
dM
33 𝜎𝑐 = 𝜎𝑓 (𝜑 − 𝜑𝑡ℎ )𝑡 Eq. 3
34
35
36
37
where σc, σf are respectively the composite and filler conductivities, and φ, φth are the filler content
38
39 and percolation threshold in vol%, respectively; t is the critical exponent. The calculated percolation
40
41 threshold value for the epoxy/GnP composite film was ~1.08 vol%. At this fraction, a connected
42
43
pte

global network of GnPs was formed inside the epoxy matrix facilitating the electron mobility across
44
45
46 the matrix and breaking its nonconductive behavior.
47
48
49
When the fraction of GnPs increases beyond 3 vol%, the composite has high electrical conductivity
50
51 and thus Agilent 4339B high-resistivity meter was not suitable to measure the composite
ce

52
53 conductivity. A four-probe method was employed to record the surface resistivity then using Eq.2 to
54
55 convert it into bulk conductivity as presented in Figure 4b. After surpassing percolation threshold,
56
Ac

57
58 the electrical conductivity linearly and steadily increased with GnP vol% due to the multitude of
59
60 conductive phase creating numerous conductive paths across the composite. For example, at 10% vol
12
Page 13 of 25 AUTHOR SUBMITTED MANUSCRIPT - NANO-122604.R2

1
2 of GnPs, the epoxy composite possesses electrical conductivity of ~0.01 S/cm, representing an
3
4 increase of 11 orders of magnitude compared to neat epoxy.
5
6
7 Thermal conductivity of the composite films
8

pt
9 Thermal conductivity of epoxy/GnP composite films was studied at different levels of GnP vol%.
10
11 Thermal conductivity measurements and enhancement percentage are shown in Figure 4c. The results
12
13
revealed that thermal conductivity rose steadily and linearly with the addition of GnPs. For example,

cri
14
15
16 the thermal conductivity of composite reached 1.07 W m-1 K-1 at 10 vol% GnPs recording 569%
17
18 improvement compared to neat epoxy.
19
20

us
21 Thermal conductivity mainly relies on phonon transfer via lattice vibration. Neat polymers usually
22
23 scatter the phonons when they transfer from one side to another due to nonconnected chain structure
24
25
of polymers. When GnPs are added, phonons have better paths to pass with lower scattering compared
26
27
28
29
an
to the neat polymer. However, the Kapitza resistance; thermal resistance at interface between filler

30 and matrix (GnP and epoxy); highly affects the overall thermal conductivity of the composite [42-
31
32 44]. This explains the linear augment in thermal conductivity compared to the percolation behavior
dM
33
34
35 in electrical conductivity.
36
37
Mechanical performance of the composite films
38
39 Young’s moduli and tensile strengths of neat epoxy and its composite films are depicted in Figure 4d.
40
41 Obviously, adding GnPs into epoxy has a significant influence on the mechanical performance of the
42
43
pte

44 films. With the increase of GnP content, the Young’s modulus and tensile strength of the composite
45
46 film exhibit a prominent enhancing trend. For example, at 10 vol% GnP, the Young’s modulus and
47
48 tensile strength reached 26±1.7 MPa and 1.0±0.08 MPa, demonstrating 1344% and 66.7%
49
50
51 improvement respectively, compared to the neat epoxy proving successful reinforcement of epoxy
ce

52
53 matrix by GnPs. However, it is noticed that tensile strength trend inflected at 5 vol% GnP. Initially,
54
55 the tensile strength increased to 1.4 ± 0.07 MPa at 5 vol% GnP and then decreased to 1 ± 0.08 MPa
56
Ac

57
at 10 vol%, demonstrating an increment of 133.3% at 5 vol% and then reduction by 28.6% at 10 vol%.
58
59
60 Similar trend was reported by Meng, Q et al. [45]. The increase in tensile strength at low fractions is
13
 AUTHOR SUBMITTED MANUSCRIPT - NANO-122604.R2 Page 14 of 25

1
2 due to: (i) uniform dispersion of nanofillers and (ii) strong mechanical interlocking between the filler
3
4 and matrix. The noticed drop in tensile strength after 5 vol% GnP is due to the toughening and
5
6
7 stiffening effect of the matrix upon introducing a stiff nanofiller (Young’s modulus of graphene is ~
8

pt
9 1TPa). Figure 4(e) shows the elongation at break decreases upon filler addition, emphasizing that the
10
11 matrix is undergoing stiffening effect. Similar results were reported by Ma J et al. [46]. Based on the
12
13
revealed mechanical properties, electrical and thermal conductivity, the composite with 10 vol%

cri
14
15
16 GnPs was chosen in order to study its piezoresistive performance.
17
18
19 (a) (b)

Electrical conductivity (S/cm)

0.01
Electrical conductivity (S/cm)

1E-7
20

us
21 1E-3
t=1.08 vol%
22 1E-9
23 1E-4
24 -7.0

1E-11
25 -7.2 1E-5
Log(c)

26
-7.4

27
28
29
1E-13
-7.6

-7.8

-0.4 -0.2
Slope, t= 1.21
R2 = 99%
0.0
Log(−t)
0.2
an 1E-6

1E-7
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3 4 5 6 7 8 9 10 11
30 GnP fraction (vol%) GnP fraction (vol%)
31
32 1.25 600
(d) Young's modulus
Thermal conductivity (W/m.k)

(c) Thermal conductivity

dM
30 Tensile strength 2.5
33 Enhancement ratio
Young's modulus (MPa)
Enhancement ratio (%)

Tensile strength (MPa)

1.00
34 450
2.0
35 20
0.75
36
300
37 10
1.5
0.50
38
39 0.25
150 1.0
0
40
41 0.00 0 0.5
42 0 2 4 6
GnP fraction (vol%)
8 10 0 2 4 6 8 10
43 GnP fraction (vol%)
pte

44
45 21
(e)
46
Elongation @ break (%)

47
48 18
49
50
15
51
ce

52
53 12
54
55
56 0 2 4 6 8 10
Ac

57 GnP fraction (vol%)

58 Figure 4. Functional and mechanical properties of epoxy/GnP composite film: (a,b) electrical percolation
59 threshold and conductivities at different filler fractions, (c) thermal conductivity, (d) Young’s
60 modulus and tensile strength, and (e) elongation at break
14
Page 15 of 25 AUTHOR SUBMITTED MANUSCRIPT - NANO-122604.R2

1
2 3.4 Morphology of epoxy/GnP composite film
3
4
5 Since microstructure of neat epoxy is featureless and possesses a smooth cross-section surface
6
7 structure [47], its SEM image is not presented here. The SEM micrographs for epoxy/GnP at 5 vol%
8

pt
9 (Figure 5a1‒a4) and 10 vol% (Figure 5b1‒b4) were presented in order to show changes in
10
11
microstructure with the progress of GnP content. At both contents, GnPs generally show relatively
12
13

cri
14 uniform dispersion with good mechanical interlocking with the matrix leading to high mechanical
15
16 stiffness.
17
18 At low fraction 5 vol%, Image (a1) shows sponge-like microstructure displaying micropores
19
20

us
21 throughout the cross-section with rough surface due to the breakage of GnP flakes. When the
22
23 micrograph is magnified to Image (a2 & a3), giant GnPs are overlapped and tightly connected to each
24
25 other leading to a conductive epoxy/GnP composite (~3×10-7 S/cm at 5 vol%). This supports the
26
27
28
29
an
observed percolation threshold at 1.08 vol%. Image (a4) presents sharp and rough edge of GnP which

30 enhance the mechanical interlocking with the matrix leading to high stiffness and maximum tensile
31
32 strength ‒Figure 4d.
dM
33
34
At high fraction, 10 vol%, microstructure images show dense and compact structure of GnPs over the
35
36
37 entire cross-section. However, few small micropores are presented in Images (b2‒b4) (blue circles).
38
39 The compact structure does not allow much epoxy to penetrate into GnPs networks, nonetheless the
40
41 5 vol% structure does. Also, at high volume fractions, small quantity of bonding phase (epoxy) is
42
43
pte

44 available leading to relatively weak interface between the matrix and GnPs compared to 5 vol%;
45
46 epoxy works as bonding material between GnPs and matrix as the same time. Consequently, the
47
48 composite is not able to resist the mechanical deformation once the applied stress is beyond the yield
49
50
51 point, demonstrating lower tensile strength compared to 5 vol%. However, Young’s modulus
ce

52
53 increases with the GnP fraction because it is calculated at the elastic region.
54
55 Apart from mechanical performance, the multitude of GnPs in the matrix at high fractions provides
56
Ac

57
several conductive paths throughout the matrix recording electrical conductivity of ~0.01 S/cm at 10
58
59
60 vol% GnPs.
15
 AUTHOR SUBMITTED MANUSCRIPT - NANO-122604.R2 Page 16 of 25

1
2
3
4
5
6
7
8

pt
9
10
11
12
13

cri
14
15
16
17
18
19
20

us
21
22
23
24 Figure 5. SEM images of epoxy/GnP composite films; (a1-a4) 5 vol% GnP content and (b1-b4) 10 vol% GnP content
25
26
3.5 Piezoresistive performance under different mechanical loads
27
28
29
an
In this section, 10 vol% GnP composite film was used to examine its piezoresistive performance
30
31
32 under various dynamic engineering applications. In the discussion below, the resistance change refers
dM
33
34 to ΔR=R-Ro and relative resistance change is ΔR/Ro where R is the measured film resistance
35
36 corresponding to the stimulating signal and Ro is the film resistance at zero signal (zero strain for
37
38
39 instance). Gauge factor is calculated as:
40
41 Δ (ΔR/R o )
GF = Eq. 4
42 Δε
43
pte

44 Figure 6a shows the relative resistance change as a function of strain. Since tensile strain is one of
45
46
47
the important mechanical deformation, it was thoroughly studied. It is observed that the relative
48
49 resistance change in Figure 6a can be divided into two linear segments; first one is from 0 to 7 %
50
51 strain and the second is within strain range 7‒10 %. At range 0‒7%, strain sensitivity (gauge factor)
ce

52
53 is calculated as 2 while at the second range the gauge factor is 6 which are the slopes of the fitting
54
55
56 lines presented in the insets. These results tell that the film is able to sense and distinguish between
Ac

57
58 low and high strain affecting a structure. The change in sensitivity of the composite film at low and
59
60 high strains is due to that (i) at low strain range, the overlap and connection between graphene
16
Page 17 of 25 AUTHOR SUBMITTED MANUSCRIPT - NANO-122604.R2

1
2 platelets is firm and thus resistance does not change much resulting in low gauge factor ,2, and (ii) at
3
4 high strain, some of the network branches are broken and disconnected, increasing the overall
5
6
7 resistance of the composite leading to high gauge factor, 6 .
8

pt
9
Figure 6b‒d shows that the epoxy/GnP film is able to detect not only tensile strain but various
10
11
12 mechanical deformations including compression, bending and torsion. Figure 6b shows relative
13

cri
14 resistance change of the composite film at various compressive loads. When the compressive load
15
16 augments to 2N, the value of ǀΔR/R0ǀ increased from 0 to 71%. This result demonstrates that the
17
18
19 composite film has high sensitivity to compressive loads and is able to sense loads as small as 0.01N.
20

us
21 Under compressive load, the distance between adjacent graphene sheets planes within the composite
22
23 film decreases resulting in better connection between GnPs in the 3D network paths. This would lead
24
25
26 to a decrease in the composite’s electric resistance and ultimately an increase in absolute value of
an
27
28 relative resistance change, ǀΔR/R0ǀ. This gives our sensor the ability to work as a load cell with high
29
30 resolution. Figure 6c shows electrical response of the composite film as a function of bending angle.
31
32
The composite film could be bent up to 180º, demonstrating a good flexibility; with the increase of
dM
33
34
35 bending angle to 180º, the ΔR/R0 elevated from 0 to 45%. This may be due to the fact that the distance
36
37 between GnPs at the bend increases as the bending angle increases, which makes the sensing ability
38
39 more pronounced.
40
41
42 Figure 6d shows the linear dependence of the ΔR/R0 with the angle of torsion. The composite film
43
pte

44 was twisted by 180º, which demonstrates an excellent flexibility within this range (0‒180º). With the
45
46 torsional angle increasing to 180º, the ΔR/R0 increased from 0 to 30%.
47
48
49
50
51
ce

52
53
54
55
56
Ac

57
58
59
60
17
 AUTHOR SUBMITTED MANUSCRIPT - NANO-122604.R2 Page 18 of 25

1
2 80
3
4
(b)
5 60
6

ôDR/R0ô (%)
7
8 40

pt
9
10
11 20
12
13

cri
14 0
15
16 0.0 0.5 1.0 1.5 2.0
17 Compressive load (N)
18
19
20 60
(c) (d)

us
40
21
22 45
23 30
DR/R0 (%)

DR/R0 (%)
24
25 30
20
26
27
28
29
15
an 10

30
0
31 0
32 0 45 90 135 180 0 45 90 135 180
dM
33 Bending angle () Torsion angle ()
34 Figure 6. The piezoresistive responses of epoxy/GnP composite films to different deformations: (a) tensile strain, (b)
35 compressive deformation (c) bending deformation, and (c) torsional deformation.
36
37
38
3.6 Temperature sensing
39
40 Of most signals, temperature monitoring is imperative in many physical, electronic, chemical,
41
42 mechanical and biological systems [48-50]. The signal in temperature sensors is provided by a sudden
43
pte

44
45
increase or decrease in resistivity or capacitance with the temperature [51]. Figure 7 presents the
46
47 overall relative resistance change in the composite film within the temperature range 20‒110℃. It is
48
49 noted that relative resistance change is linearly with temperature within the 20‒110℃ range. The
50
51
ce

adjusted regression coefficient is 99%, which justifies that the linear regression is well fit and accurate
52
53
54 with minimum offset points. The slope of the linear fit represents the temperature sensitivity (α) of
55
56 the film according to Eq.5 [52]
Ac

57
58
59 ΔR 1
α= × Eq.5
60 R o ΔT
18
Page 19 of 25 AUTHOR SUBMITTED MANUSCRIPT - NANO-122604.R2

1
2 In temperature range 20–110℃, the temperature sensitivity is calculated as 0.63% (0.0063℃-1) which
3
4 is much higher than the standard commercial platinum-based temperature sensor (0.0039℃-1) [52,
5
6
7 53] revealing that our film possesses outstanding temperature sensing performance.
8

pt
9 In order to test the stability and reliability of the developed film as a temperature sensor within range
10
11 20–110oC, cyclic thermal test was carried out for 40 cycles (Figure 8a and b). The film response in
12
13
the first 5 cycles and last 5 cycles are compared (Figure 8b inset). Initially, the relative resistance

cri
14
15
16 change is slowly decreasing until it stabilizes with the number of cycles. It is observed that the relative
17
18 resistance change of first 5 cycles is decreasing and not consistent with the steady state of the sensor
19
20
revealing that the film is initially unstable till it passes the settle-in phase of a new sensor upon being

us
21
22
23 newly manufactured. On the other hand, the relative resistance change in the last 5 cycles has reached
24
25 a steady state response indicating a stable behavior as temperature sensor after 40 cycles.
26
27
28
29
60
an
Experimental data
50 Fitting line
30
31 40
32
DR/R0 (%)

dM
33 30
34
Fitting details
35 20
R2=99%
36 Slope=0.63
37 10
38
39 0

40 10 20 30 40 50 60 70 80 90 100 110 120

41 o
Temperature ( C)
42
Figure 7. The electric resistance response of epoxy/GnP composite film (10 vol%) during heating and
43
pte

its temperature sensitivity coefficient.

44
45 150
46
90
(a) Temperature
50
(b)
47 DR/R0 120
Temperature (oC)

30

48 75 20
DR/R0 (%)

(R-R0)/R0 (%)

25 90
49
DR/R0 (%)

10
-15
0
50
(R-R0)/R0 (%)

-20

60 60 -10
0
51
ce

-25
-20

52 30 0 20 40 60
Time (min)
80 100 -30

53 -25 -35
45 750 780 810 840
Time (min)
54 0
55 -50
30 -30
56
Ac

57 0 200 400 600 800 0 200 400 600 800

58 Time (min) Time (min)
59 Figure 8. (a) Reliability of the composite film as thermal sensor within temperature range 30–75 ℃ for
60 40 cycles; and (c) comparison between first and last five thermal cycles
19
 AUTHOR SUBMITTED MANUSCRIPT - NANO-122604.R2 Page 20 of 25

1
2 3.7 Fatigue testing
3
4
5 Reliability represents how consistently a strain sensor can perform normal function without
6
7 degradation, upon repeating it number of times. The cyclic tensile testing was conducted to study the
8

pt
9
10 stability and durability of the composite film as a strain sensor. Since the developed epoxy/GnP
11
12 composite film is proposed to detect structure health monitoring, low strain and frequency were
13

cri
14 applied during the fatigue test [54-56]. The epoxy/GnP composite film (at 10 vol%) was tested for
15
16
17 10×103 cycles under 5% strain at 1Hz frequency, as shown in Figure 9a. Figures 9b‒d are
18
19 magnifications for composite film responses within the ranges of 0‒1000, 4500‒5500 and 9×103‒
20

us
21 10×103 cycles, respectively. The relative resistance change (ΔR/R0) at first and last 10 cycles were
22
23
compared (Figures 9b‒d insets) in each range. In Figure 9b, the epoxy/composite film demonstrates
24
25
26 high consistency at the first and last 10 cycles revealing the resilience and repeatability of the film as
27
28
29
a strain sensor.
an
30
31 On the other hand, the insets of Figure 9c shows a weak degeneration after 6000 cycles; upon
32
dM
33 unloading, the resistance change came back to 2.3% instead of 0 as in Figure 9b. The observed
34
35
degeneration is progressed in the last 1000 cycles (within range 9×103‒10×103 cycles) as seen in
36
37
38 Figure 9d; when the load was released, the relative resistance change maintained 4% instead of getting
39
40 back to zero. This zero-drift error is due to two main reasons: (i) the irreversible destruction of the
41
42 GnP conductive networks caused by the accumulation of microcracks and further yielding of the film
43
pte

44
45 due to fatigue damage, and (ii) the deterioration at the interface between the nanofiller and matrix
46
47 upon cyclic loading [57]. This concludes that our epoxy/GnP composite film has high repeatability
48
49 and resilience up to 5500 cycles.
50
51
ce

52
53
54
55
56
Ac

57
58
59
60
20
Page 21 of 25 AUTHOR SUBMITTED MANUSCRIPT - NANO-122604.R2

1
2
3
4
5
6
7
8

pt
9
10
11
12
13

cri
14
15
16
17
18
19
20

us
21
22
23
24
25
26
27
28
29
an
30
31
32
dM
33
34
35
36
37
38
39
40
41
42
43
pte

44
45
46
47
48
49
50
51
ce

52
53
54
55
56
Ac

57
Figure 9. The cyclic tensile test of epoxy/GnP composite films at 5% strain and 1Hz, (a) the overall cyclic test over
58 10000 cycles, (b‒d) magnifications of composite film responses within ranges of 0‒1000, 4500‒5500 and
59 9000‒10000 cycles, respectively.
60
21
 AUTHOR SUBMITTED MANUSCRIPT - NANO-122604.R2 Page 22 of 25

1
2 Conclusions
3
4 In this work, a flexible, mechanically strong and thermally stable piezoresistive graphene-based film
5
6
sensor was fabricated. Their microstructure and piezoresistive properties were investigated. At 10 vol%
7
8
GnPs, Young’s modulus and tensile strength of the composite film are 26 MPa and 1.0 MPa, recording

pt
9
10
11 1344% and 66.7% improvement respectively. The percolation threshold of epoxy/GnPs composite
12
13 film is observed at ~1.08 vol% of GnPs breaking the electrically nonconductive nature of the epoxy

cri
14
15
16 matrix. The thermal conductivity of composite film also showed prominent increase from 0.16 to
17
18 1.07 W m-1 K-1 reporting 569% improvement.
19
20 The epoxy/GnP composite film showed two different gauge factors upon tensile strain; 2 at strain

us
21
22
23
range 0–7% and 6 at 7–10% range. It also has good pressure response up to 2N, by demonstrating a
24
25 71% reduction in relative resistance change. Furthermore, the flexible composite films proved an
26
27
28
29
an
excellent reliability and durability up to 5500 cycles with minimum zero-point drift and also showed

responses to bending and twisting angles up to 180°. Within the temperature range 20‒110°C, the
30
31
32 film displayed an excellent sensitivity to temperature change recording temperature sensitivity
dM
33
34 coefficient of 0.0063℃-1 which is far higher than the standard commercial temperature sensor
35
36 (0.0039℃-1).
37
38
39 Overall, this multifunctional composite film possesses substantial improvements in mechanical,
40
41 electrical and thermal properties upon adding graphene platelets leading to outstanding piezoresistive
42
43
pte

performance under various mechanical and thermal loads. The composite film has the potential to
44
45
46
work as a fully-fledged sensor in various engineering applications where temperature and mechanical
47
48 loads require monitoring.
49
50
51
ce

52
53
54
55
56
Ac

57
58
59
60
22
Page 23 of 25 AUTHOR SUBMITTED MANUSCRIPT - NANO-122604.R2

1
2 Acknowledgements
3
4
5 Authors thank Asbury (Asbury, NJ, USA) for providing the graphite intercalation compounds. This
6
7 work was financially supported by the Natural Science Foundation of Liaoning Province-China
8

pt
9 (20170520142 and 2019-MS-256), Aeronautical Science Foundation of China (2018ZF54036),
10
11
12 China Postdoctoral Science Foundation (2019M651151) and National Natural Science Foundation
13

cri
14 (51973123).
15
16
17 Reference
18
19
20 1. Shi, G., et al., Highly Sensitive, Wearable, Durable Strain Sensors and Stretchable Conductors Using

us
21 Graphene/Silicon Rubber Composites. Advanced Functional Materials, 2016. 26(42): p. 7614-7625.
22 2. Zhang, B.-X., et al., Multi-dimensional flexible reduced graphene oxide/polymer sponges for multiple
23 forms of strain sensors. Carbon, 2017. 125: p. 199-206.
24 3. Aguilar Ventura, I., J. Zhou, and G. Lubineau, Drastic modification of the piezoresistive behavior of
25 polymer nanocomposites by using conductive polymer coatings. Composites Science and Technology,
26 2015. 117: p. 342-350.
27
28
29
4.
an
Zhang, R.-C., et al., Gold nanoparticle-polymer nanocomposites synthesized by room temperature
atmospheric pressure plasma and their potential for fuel cell electrocatalytic application. Scientific
Reports, 2017. 7: p. 46682.
30
5. Hicks, J.F., Y. Seok-Shon, and R.W. Murray, Layer-by-Layer Growth of Polymer/Nanoparticle Films
31
Containing Monolayer-Protected Gold Clusters. Langmuir, 2002. 18(6): p. 2288-2294.
32
dM
33
6. Zhang, W., et al., Conducting polymer/silver nanowires stacking composite films for high-performance
34 electrochromic devices. Solar Energy Materials and Solar Cells, 2019. 200: p. 109919.
35 7. Shi, G., et al., Facile Fabrication of Graphene Membranes with Readily Tunable Structures. ACS
36 Applied Materials & Interfaces, 2015. 7(25): p. 13745-13757.
37 8. Meng, Q., et al., Effect of interface modification on PMMA/graphene nanocomposites. Journal of
38 Materials Science, 2014. 49(17): p. 5838-5849.
39 9. Meng, Q., et al., Flexible strain sensors based on epoxy/graphene composite film with long molecular
40 weight curing agents. Journal of Applied Polymer Science, 2019. 136(35): p. 47906.
41 10. Rahman, R. and P. Servati, Effects of inter-tube distance and alignment on tunnelling resistance and
42 strain sensitivity of nanotube/polymer composite films. Nanotechnology, 2012. 23(5): p. 055703.
43 11. Mo, L., et al., Flexible transparent conductive films combining flexographic printed silver grids with
pte

44 CNT coating. Nanotechnology, 2016. 27(6): p. 065202.

45 12. Zhang, Q., et al., Conductive mechanism of carbon black/polyimide composite films, in Journal of
46 Polymer Engineering. 2018. p. 147.
47 13. Araby, S., et al., Elastomeric composites based on carbon nanomaterials. Nanotechnology, 2015.
48 26(11): p. 112001.
49 14. Araby, S., et al., Electrically and thermally conductive elastomer/graphene nanocomposites by
50
solution mixing. Polymer, 2014. 55(1): p. 201-210.
51
ce

15. Araby, S., et al., A novel approach to electrically and thermally conductive elastomers using graphene.
52
53
Polymer, 2013. 54(14): p. 3663-3670.
54 16. Zaman, I., et al., A Facile Approach to Chemically Modified Graphene and its Polymer
55 Nanocomposites. Advanced Functional Materials, 2012. 22(13): p. 2735-2743.
56 17. Amjadi, M., et al., Stretchable, Skin-Mountable, and Wearable Strain Sensors and Their Potential
Applications: A Review. 2016. 26(11): p. 1678-1698.
Ac

57
58 18. Mora, A., F. Han, and G. Lubineau, Estimating and understanding the efficiency of nanoparticles in
59 enhancing the conductivity of carbon nanotube/polymer composites. Results in Physics, 2018. 10: p.
60 81-90.
23
 AUTHOR SUBMITTED MANUSCRIPT - NANO-122604.R2 Page 24 of 25

1
2 19. Hu, N., et al., Investigation on sensitivity of a polymer/carbon nanotube composite strain sensor.
3 Carbon, 2010. 48(3): p. 680-687.
4 20. Wichmann, M.H.G., et al., Piezoresistive response of epoxy composites with carbon nanoparticles
5 under tensile load. Physical Review B, 2009. 80(24): p. 245437.
6 21. Bautista-Quijano, J.R., et al., Strain sensing capabilities of a piezoresistive MWCNT-polysulfone film.
7 Sensors and Actuators A: Physical, 2010. 159(2): p. 135-140.
8 22. Gau, C., H.S. Ko, and H.T. Chen, Piezoresistive characteristics of MWNT nanocomposites and

pt
9 fabrication as a polymer pressure sensor. Nanotechnology, 2009. 20(18): p. 185503.
10
23. Hamdi, K., et al., Improvement of the electrical conductivity of carbon fiber reinforced polymer by
11
incorporation of nanofillers and the resulting thermal and mechanical behavior. Journal of Composite
12
13
Materials, 2018. 52(11): p. 1495-1503.
24. Liu, W.O.T., A Review: Carbon Nanotube-Based Piezoresistive Strain Sensors. Journal of Sensors,

cri
14
15 2012. 2012: p. 15.
16 25. Lu, S., et al., Strain sensing behaviors of GnPs/epoxy sensor and health monitoring for composite
17 materials under monotonic tensile and cyclic deformation. Composites Science and Technology, 2018.
18 158: p. 94-100.
19 26. Wichmann, M.H., et al., Piezoresistive response of epoxy composites with carbon nanoparticles under
20 tensile load. 2009. 80(24): p. 245437.

us
21 27. Wang, X., et al., Highly Stretchable and Wearable Strain Sensor Based on Printable Carbon Nanotube
22 Layers/Polydimethylsiloxane Composites with Adjustable Sensitivity. ACS Applied Materials &
23 Interfaces, 2018. 10(8): p. 7371-7380.
24 28. Collins, P.G., et al., Extreme Oxygen Sensitivity of Electronic Properties of Carbon Nanotubes. Science,
25 2000. 287(5459): p. 1801-1804.
26 29. Nela, L., et al., Large-Area High-Performance Flexible Pressure Sensor with Carbon Nanotube Active
27
28
29
30.
an
Matrix for Electronic Skin. Nano Letters, 2018. 18(3): p. 2054-2059.
Liu, Y., et al., Room Temperature Broadband Infrared Carbon Nanotube Photodetector with High
Detectivity and Stability. 2016. 4(2): p. 238-245.
30
31. Zaman, I.K., H. C. Dai, J. Kawashima, N. Michelmore, A. Sovi, A. Dong, S. Luong, L. Ma, J., From
31
32
carbon nanotubes and silicate layers to graphene platelets for polymer nanocomposites. Nanoscale,
dM
33 2012. 4(15): p. 4578-86.
34 32. Araby, S., et al., Melt compounding with graphene to develop functional, high-performance elastomers.
35 Nanotechnology, 2013. 24(16): p. 165601.
36 33. Zaman, I., et al., From carbon nanotubes and silicate layers to graphene platelets for polymer
37 nanocomposites. Nanoscale, 2012. 4(15): p. 4578-4586.
38 34. Meng, Q., et al., Free-standing, flexible, electrically conductive epoxy/graphene composite films.
39 Composites Part A: Applied Science and Manufacturing, 2017. 92: p. 42-50.
40 35. Pei, S. and H.-M. Cheng, The reduction of graphene oxide. Carbon, 2012. 50(9): p. 3210-3228.
41 36. Ferrari, A.C. and D.M. Basko, Raman spectroscopy as a versatile tool for studying the properties of
42 graphene. Nature Nanotechnology, 2013. 8: p. 235.
43 37. Malard, L.M., et al., Raman spectroscopy in graphene. Physics Reports, 2009. 473(5): p. 51-87.
pte

44 38. Zaman, I., et al., From carbon nanotubes and silicate layers to graphene platelets for polymer
45 nanocomposites. Nanoscale, 2012. 4(15): p. 4578-4586.
46 39. Wu, Z.-S., et al., Synthesis of Graphene Sheets with High Electrical Conductivity and Good Thermal
47 Stability by Hydrogen Arc Discharge Exfoliation. ACS Nano, 2009. 3(2): p. 411-417.
48 40. Marcano, D.C., et al., Improved Synthesis of Graphene Oxide. ACS Nano, 2010. 4(8): p. 4806-4814.
49
41. Stankovich, S., et al., Graphene-based composite materials. Nature, 2006. 442(7100): p. 282-286.
50
42. Su, Y., J.J. Li, and G.J. Weng, Theory of thermal conductivity of graphene-polymer nanocomposites
51
ce

52
with interfacial Kapitza resistance and graphene-graphene contact resistance. Carbon, 2018. 137: p.
53 222-233.
54 43. Han, Z. and A. Fina, Thermal conductivity of carbon nanotubes and their polymer nanocomposites: A
55 review. Progress in Polymer Science, 2011. 36(7): p. 914-944.
56 44. Chen, Y., et al., Advances in graphene-based polymer composites with high thermal conductivity.
Ac

57 Veruscript Functional Nanomaterials, 2018. 2: p. oosb06.

58 45. Meng, Q., et al., Mechanically robust, electrically and thermally conductive graphene-based epoxy
59 adhesives. Journal of Adhesion Science and Technology, 2019. 33(12): p. 1337-1356.
60 46. Ma, J., et al., Development of polymer composites using modified, high-structural integrity graphene
24
Page 25 of 25 AUTHOR SUBMITTED MANUSCRIPT - NANO-122604.R2

1
2 platelets. Composites Science and Technology, 2014. 91: p. 82-90.
3 47. Kuan, H.-C., J.-B. Dai, and J. Ma, A reactive polymer for toughening epoxy resin. Journal of Applied
4 Polymer Science, 2010. 115(6): p. 3265-3272.
5 48. Kunzelman, J., et al., Shape memory polymers with built-in threshold temperature sensors. Journal of
6 Materials Chemistry, 2008. 18(10): p. 1082-1086.
7 49. Mohiuddin, M., et al., Flexible cellulose acetate/graphene blueprints for vibrotactile actuator. RSC
8 Advances, 2015. 5(43): p. 34432-34438.

pt
9 50. Robinson, J.T., et al., Reduced Graphene Oxide Molecular Sensors. Nano Letters, 2008. 8(10): p.
10
3137-3140.
11
51. Al-Mumen, H., et al., Thermo-flow and temperature sensing behaviour of graphene based on surface
12
13
heat convection. Iet Micro & Nano Letters, 2013. 8(10): p. 681-685.
52. Wang, Z., et al., 3D-Printed Graphene/Polydimethylsiloxane Composites for Stretchable and Strain-

cri
14
15 Insensitive Temperature Sensors. ACS Applied Materials & Interfaces, 2019. 11(1): p. 1344-1352.
16 53. Kuo, J.T.W., L. Yu, and E. Meng, Micromachined Thermal Flow Sensors—A Review. Micromachines,
17 2012. 3(3): p. 550-573.
18 54. Tung, T.T., et al., Engineering of graphene/epoxy nanocomposites with improved distribution of
19 graphene nanosheets for advanced piezo-resistive mechanical sensing. Journal of Materials Chemistry
20 C, 2016. 4(16): p. 3422-3430.

us
21 55. Moriche, R., et al., Strain monitoring mechanisms of sensors based on the addition of graphene
22 nanoplatelets into an epoxy matrix. Composites Science and Technology, 2016. 123: p. 65-70.
23 56. Eswaraiah, V., K. Balasubramaniam, and S. Ramaprabhu, Functionalized graphene reinforced
24 thermoplastic nanocomposites as strain sensors in structural health monitoring. Journal of Materials
25 Chemistry, 2011. 21(34): p. 12626-12628.
26 57. Qiu, A., et al., A Path Beyond Metal and Silicon:Polymer/Nanomaterial Composites for Stretchable
27
28
29
Strain Sensors. 2019. 29(17): p. 1806306.
an
30
31
32
dM
33
34
35
36
37
38
39
40
41
42
43
pte

44
45
46
47
48
49
50
51
ce

52
53
54
55
56
Ac

57
58
59
60
25