- Email: [email protected]
Piezoresistive response of carbon nanotube composite film under laterally compressive strain
Accepted Manuscript Title: Piezoresistive response of carbon nanotube composite film under laterally compressive strain Authors: Yin Wang, Shaokai Wang, Min Li, Yizhuo Gu, Zuoguang Zhang PII: DOI: Reference:
S0924-4247(17)30460-0 https://doi.org/10.1016/j.sna.2018.02.032 SNA 10654
To appear in:
Sensors and Actuators A
Received date: Revised date: Accepted date:
19-3-2017 18-2-2018 19-2-2018
Please cite this article as: Wang Y, Wang S, Li M, Gu Y, Zhang Z, Piezoresistive response of carbon nanotube composite film under laterally compressive strain, Sensors and Actuators: A Physical (2010), https://doi.org/10.1016/j.sna.2018.02.032 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Piezoresistive response of carbon nanotube composite film under laterally compressive strain Yin Wang, Shaokai Wang*, Min Li, Yizhuo Gu and Zuoguang Zhang Key Laboratory of Aerospace Advanced Materials and Performance (Ministry of Education), School of
100191, China
Highlights
This study explored the piezoresistive behavior of high CNT-loading carbon nanotube composite film
U
to advance the self-sensing ability of structural composite.
The relationship between microstructure and piezoresistive response for typical carbonaceous film
N
SC R
*Corresponding author: Tel & Fax: +86-10-82339575, E-mail: [email protected]
IP T
Materials Science and Engineering, Beihang University, No. 37 Xueyuan Road, Haidian District, Beijing
The piezoresistive behavior of CNT composite film was effectively tailored by polymeric matrix and
M
A
material was revealed by comparing CNT and graphite composite films.
ED
CNT functionalization.
Abstract：
PT
Carbon nanotubes (CNTs) have attracted great attention for strain sensor application due to their excellent electromechanical property. This paper focused on the piezoresistive response of high CNT loading composite
CC E
film. The piezoresistive behavior of floating catalyst chemical vapor deposition (FCCVD)-grown CNT composite film under tension and lateral compression was investigated and compared with a graphite composite film. The effects of matrix type, sidewall epoxidation and acid treatment on piezoresistive response of CNT
A
composite film were discussed. The results showed that CNT composite film exhibited positive piezoresistive response under tension and negative piezoresistive effect under lateral compression. After eliminating the effect of geometrical deformation on gauge factor, the change in resistivity was more sensitive under lateral compression than under tension. Different from CNT composite film, graphite composite film showed less obvious piezoresistive behavior, and showed positive piezoresistive response regardless of tensile and laterally compressive loads. Furthermore, the piezoresistive response of CNT composite film was found to closely
correlate with its modulus. Stronger interfacial bonding was proved to effectively enhance the piezoresistive response and increase the corresponding gauge factor. CNT/epoxy composite film with 1h treatment in potassium permanganate solution reached a gauge factor of -5.9. These results revealed the key factors tailoring piezoresistive response of CNT composite film, and presented one potential approach to monitor biaxial strain. The present work should be useful in the in-plane strain monitoring by incorporating the CNT film into
A
CC E
PT
ED
M
A
N
U
SC R
Key words: Carbon nanotube film; piezoresistive response; gauge factor; compression.
IP T
traditional composite.
1. Introduction Carbon nanotubes possess excellent electromechanical property and show great potential as piezoresistive strain sensor due to the alteration of band structure accompanied with its mechanical deformation [1, 2]. How to take CNT's advantages in macroscopic CNT-based piezoresistive sensor has attracted lots of research attention. As one of the most widely studied CNT-based sensor, polymer film mixed with CNTs has fully
IP T
demonstrated its piezoresistive behavior due to the change of tube-tube contact in CNT network. This kind of CNT-based sensors have been well developed, such as single-walled carbon nanotube (SWNT)-polyelectrolyte (PE) film [3], CNT-poly (methyl methacrylate) (PMMA) film [4], multi-wall carbon nanotubes (MWCNT)-
SC R
polysulfone film [5], MWCNT and natural rubber nanocomposite [6]. The MWCNTs/epoxy nanocomposites have been reported to display a gauge factor as high as 78, and also exhibited excellent stability and durability [7]. The microstructure of inside CNT network was recognized to play an important role on piezoresistive
U
response, including network density, degree of CNT alignment [8], aspect ratio of CNTs [9], CNT concentration
N
[10], and so on. Wang et al. [11] conducted numerical simulation, and found that the junction resistance between
A
CNTs decided the piezoresistive behavior of these polymer film sensor. The variation of average junction gap,
M
describing the conductance change of CNT network, was influenced by the orientation and diameter of CNTs, the poisson's ratio of polymer matrix, and CNT concentration. Oliva-Avile´s [12] found the positive impact of
ED
CNT alignment on the piezoresistive sensitivity of these CNT-based sensors. High CNT-loading assemblies, e.g. CNT fiber [13] and CNT film, have been well developed in recent years.
PT
These CNT assemblies were widely considered to be the next-generation high-performance structural reinforcements for aerospace-grade composite. Lekawa-Raus reported the piezoresistive response of CNT fiber
CC E
accompanied with the opening of bandgaps and reduction of the contact area between CNT bundles [14]. As a freestanding mat of densely-packed and inter-tangled carbon nanotubes, CNT film also displayed resistance change under applied force. Miao Y. et al [15] prepared pure CNT film and found its gauge factor was about
A
2.4. Although the gauge factors of these high loading CNT assembly were much smaller than the above CNT nanocomposites, their piezoresistive behavior may greatly advance the structure-function for these CNT materials. The loading transfer inside CNT composite film was greatly influenced by the interactions between CNTs and polymeric matrix. Functionalization and acid treatment on CNTs to improve its surface features have been widely investigated for higher mechanical and electrical properties [16, 17]. Still, how to tailor the
electromechanical performance of high-loading CNT materials remains challenge. Several researches have been carried out to investigate the piezoresistive response of various CNT-based sensors under different load types. Tensile force was the most common load type to measure sensor's piezoresistive behavior [18]. Besides, Kang et al adhered SWCNT buckypaper or CNT composite on a cantilever beam, and measured their resistance change with respect to the displacement due to bending. These
IP T
sensors showed linear resistance variation with strain, and had good sensitivity when these sensors were tensioned or compressed [4]. Dharap found the similar piezoresistive behavior of SWCNT buckypaper when it was subjected to unidirectional tensile and compressive stresses [19]. For CNT/polymer film, different matrices
SC R
may result in quite different piezoresistive behaviors, and it was found that the obstacle effect of polymer chains on the movement of CNTs played an important role on piezoresistive effect under compressive strain [20]. Freestanding MWCNT/ graphene nanoplatelet hybrid film also showed electrical resistance increase with
U
increasing strain undergoing flexural force, and the resistance change was more obvious in the tensile direction
N
[21]. These results showed the capability of CNT-based sensor for monitoring tensile and compressive strain.
A
This makes it possible to detect multi-axial strain with the same kind of CNT materials, and the multifunctional
M
composite made of CNT assembly may self-sense its mechanical behavior, e.g. tensile and compressive strain, and even Poisson effect under complex loading conditions. The structure–property–processing relationship has
ED
been widely studied under tension [22], but the influence factors on piezoresistive response under compression were rarely reported.
PT
This paper explored the potential of floating catalyst chemical vapor deposition (FCCVD)-grown CNT composite film as a strain sensor to monitor biaxial strain under tensile test. The piezoresistive behavior of CNT
CC E
composite film under tension and lateral compression was investigated. Focusing on the piezoresistive response under lateral compression, the effects of matrix type, functionalization and acid treatment were further studied. 2. Experimental
A
2.1. Materials
The experimentally used CNT film was supplied by the Suzhou Institute of Nano-tech and Nano-bionics.
The film was synthesized using floating catalyst CVD-growth method. In the process carbon nanotubes formed a continuous aerogel which was wound by a roller. The aerogel was further densified by spraying ethanol solution to form multilayered seamless CNT film. The prepared carbon nanotubes had 3-8 walls and the film
had a thickness of 8-12 μm. CNTs are randomly oriented in the as-received film. This kind of CNT film can be directly used as reinforcement in composite material [23] or hybridized with other fiber [24], fully taking advantages of its excellent mechanical and physical properties [25]. CNT composite film was further prepared by impregnating polymer materials. The polymeric matrix includes epoxy (EP), bismaleimide (BMI) and phenylacetylene (PAA). The epoxy E51 resin used in this
IP T
experiment was purchased from Jiangsu Wuxi Resin Works. BMI resin was provided by Institute of Chemistry, Chinese Academy of Sciences. PAA resin was provided by East China University of Science and Technology. To prepare CNT composite film, acetone was used to dissolve epoxy and PAA resin, and a homogeneous
SC R
solution was obtained with a resin concentration of 5 wt%. To achieve CNT nanocomposite film with different modulus, BMI resin was dissolved in N, N-dimethylformamide (DMF) with a concentration of 1 wt%. CNT films were soaked in resin solution for 1 h. Subsequently, the resin-impregnated CNT films were dried in a
U
vacuum oven at 30 oC for 2 h to remove the solvent. The pre-impregnated CNT films were free-standingly cured
N
in an oven to get the final composites. The epoxy resin was cured at 130 oC/2 h+150 oC /1 h+180 oC/2 h. BMI
A
resin was cured at 80 °C/1 h+150 °C/2 h+190 °C/1 h+220 °C/2 h. The curing schedule of PAA resin was
M
115 °C/10 h +120 °C/4 h +140 °C/2 h +160 °C/2 h +180 °C/2 h+250 °C/4 h. For comparison, another
Materials Technology Co., Ltd.
ED
carbonaceous film (graphite film) was used in this experiment, which was purchased from the Sixth Elements
In order to investigate the effect of CNT film's surface state on piezoresistive behavior, functionalization
PT
and acid treatment on CNT film were carried out, respectively. Potassium permanganate was used to functionalize CNT film and introduce carboxyl group. 5ml concentrated sulfuric acid was evenly dispersed in
CC E
100 mL KMnO4 solution with a concentration of 0.3 wt%. CNT film was soaked in KMnO4/H2SO4 solution for 1h or 2 h to achieve different degree of functionalization. Afterwards, the CNT film was washed five times with diluted hydrochloric acid and five times with deionized water to a pH value of 7. The treated film was then
A
dried in a vacuum oven at 70 oC for 2 h. Sulfric acid and nitric acid were used in acid treatment. CNT films were soaked in the mixture of sulfuric acid and nitric acid (3:1) for 15 minutes. Higher conductivity was achieved because the mixture solution can introduce p-type doping into the CNTs network. Then CNT film was washed with deionized water and dried in a vacuum oven.
2.2. Characterization The morphology of CNT film material was examined on a JEOL JSM-7500F scanning electronmicroscope (SEM). The electrical conductivity was measured using a dielectric impedance analyzer (Guangzhou Four Probes Tech Co., Ltd. RTS-9). The tensile property of CNT film material was measured on a Instron 3344 mechanical testing machine. The rectangular specimens with a gauge length of 10 mm were measured at a speed
IP T
of 0.5 mm/min. Five specimens were tested for each sample. To investigate the piezoresistive behavior of CNT composite film under in-plane tensile and lateral compressive loads, two pieces of composite film with dimensions of 5×10 mm were adhered to both surfaces
SC R
of tensile specimen as shown in Fig. 1. Polycarbonate specimens for tensile test were prepared according to ASTM D 638 standard. The specimen has a gauge length of 60 mm. CNT films were adhered to the specimen by using epoxy adhesive. After the adhesive was cured, electrodes were bonded with the film materials by using
U
silver paste. The specimen was then loaded onto Instron 5565 mechanical testing machine. A displacement
N
speed of 1.5 mm/min was applied on the sample. An auto DC digital multimeter was used to monitor the
A
resistance variation. The resistance changes along and perpendicular to the tensile direction were monitored,
M
respectively. The piezoresistive sensitivity was quantified by gage factor (GF) K as follow [26, 27]:
Δ𝑅 𝑅Δ𝜀
(1)
ED
K=
where R represents the initial resistance of film material, and ΔR is the resistance change in response to applied
PT
strain. Δε is the strain of CNT composite film. Δε corresponds to tensile strain or laterally compressive strain respectively when CNT composite film was placed along or perpendicular to tensile direction. Here, tensile and
CC E
laterally compressive strains are regarded as positive and negative strain, respectively. With the increase of strain, positive ΔR results in positive gauge factor indicating a positive piezoresistive effect, whereas a negative
A
piezoresistive effect is defined.
3. Results and Discussion 3.1 Comparison of piezoresistive behavior between different carbonaceous film materials In the tensile measurement, the specimen was stretched along tensile direction and shrank in the width direction. Correspondingly, the attached CNT composite films were tensioned and laterally compressed along and perpendicular to the tensile direction, respectively. Fig. 2 shows the piezoresistive behavior of CNT/epoxy
film under tension (along longitudinal direction) and lateral compression (along specimen's width direction), respectively. The resistance change showed good linear relationship with applied strain under both tensile and laterally compressive loads. By linearly fitting resistance change and strain plots, the coefficients of determination reached 0.99 and 0.98 in Fig. 2a and b, respectively. The resistance showed cyclical variation when exposed to cyclic loading as shown in Fig. 2. The linearity and stability of resistance-strain relationship
IP T
show the great potential for the application as strain sensor. Under tensile force the resistance linearly increased with increasing strain as shown in Fig. 2a. The piezoresistive response was generally considered to be caused by the change of CNTs' intrinsic resistance
SC R
subjected to applied strain and intertube resistance resulted from the change of CNT network structure. Along tensile direction, the intertube resistance increased due to the increase of tube-tube contact. Due to the Poisson effect, the deformation of polycarbonate also caused the shrinkage of adhered CNT composite film along the
U
width direction. The polycarbonate specimen had a Poisson's ratio of 0.39. When 1% strain was applied along
N
tensile direction, the transverse strain was about -0.39%. The changing trend in Fig. 2a indicated that the
A
resistance change was mainly controlled by the piezoresistive response along tensile direction. On the other
M
surface of polycarbonate sample, the contraction transverse to tensile direction resulted in negative strain of CNT composite film. Surprisingly, its resistance showed an increasing tendency with the increase of the absolute
ED
value of transverse strain under lateral compression, as shown in Fig. 2b. This was totally different from the reported results under unidirectional compression [19], where the resistance decreased with the increase in the
PT
absolute value of compressive strain. This phenomenon indicated that besides the lateral compressive strain, the deformation of CNT composite film along tensile direction also affected the transverse resistance. From the
CC E
SEM morphology of CNT film in Fig. 4a, numerous carbon nanotubes entangled in the film. The geometrical deformation caused the increase of contact resistance, and consequently the overall resistance increased with the increase in the absolute value of compressive strain. Moreover, after five loading-unloading cycles, ΔR/R
A
became negative, indicating the decrease of resistance. This should be related to the uncompacted microstructure and bundling effect of CNTs. Based on equation (1), CNT composite film had a gauge factor of 4.5 under tensile strain, which was bigger than reported CNT film without surfactant (GF=2.4) [15] but lower than CNT fiber produced with a similar process (GF=15) [14]. Under lateral compression a smaller gauge factor of -2.6 was calculated, which was the
result of a combination of geometrical deformation and resistivity change. The prominent piezoresistive response along both tensile and transverse direction suggested the possibility of CNT composite film as strain sensor monitoring biaxial strain. The gauge factor K is related to the geometrical deformation and the change in electrical resistivity, which is defined as equation (2) [28]. d𝜌 𝜌0 d𝜀
(2)
IP T
K=1 + 2ν +
where ρ and ρ0 are resistivities at strain ε and zero, respectively. ν is the Poisson’s ratio of the film. (1+2ν)
SC R
represents the effects of geometrical deformation, and dρ/ρ0dε is the contribution of the change in electrical resistivity. Since the CNT composite film was bonded on a substrate, its in-plane geometrical deformation was the same as the tensile specimen. The CNT composite film was free-standingly cured, which kept the
U
morphology of CNT network. Our previous study showed that the film’s thickness increased under uniaxial
N
tension [25]. Introducing polymer matrix may inhibit the expansion of CNT network. Herein, the geometrical deformation in thickness was ignored in this study. Under tension the effect of geometrical deformation was
A
simplified to 1+νpc, where νpc is the Poisson’s ratio of tensile specimen. According to the gauge factor, under
M
tensile strain dρ/ρ0dε is equal to 3.1. Under lateral compression, the effect of geometrical deformation is equal
ED
to 1+1/νpc, which causes the decrease in resistance. As a result, dρ/ρ0dε is calculated to be -6.2. This suggests
PT
that the resistivity change of CNT composite film is more sensitive under lateral compression than under tension.
In order to elucidate the relationship between structure and electromechanical property of carbonaceous
CC E
material, we also measured a graphite film. Fig. 3 shows the resistance change of graphite composite film under tension and lateral compression. The resistance change displayed an increase tendency with increasing tensile strain, which was the same as CNT composite film. However, ΔR/R and strain did not show good linear
A
relationship, and the fitting line had a coefficient of determination 0.92. The graphite film possessed a layered structure consisting of graphite grains as shown in Fig. 4. Under tension the contact resistance between graphite grains played an important role on resistance change. Graphite grains were connected by Van der Waal’s force, and the slippage between graphite grains might randomly occur, leading to the deviation from linear relation. Different from CNT composite film, the resistance change of graphite composite film under lateral compression showed a contrary tendency. The resistance of graphite composite film decreased when the
absolute value of compressive strain became bigger, indicating the occurrence of positive piezoresistive effect. The graphite grains were stiffer and more difficult to buckle than carbon nanotubes under the same strain. Strong van der Waals' force between graphite grains was beneficial to its structural stability. Furthermore, the compression stress may reduce the contact gap between graphite grains, resulting in the resistance decrease. After five loading cycles the resistance was observed to reach a bigger value than initial resistance, indicating
IP T
the instability of layered structure of graphite film. It should be pointed out that the resistance change was not consistent with strain for graphite film. When the biggest compressive strain was applied or the tensile strain was totally unloaded, the resistance did not reach its minimum value. This might be related to the local buckling
SC R
in the film under compression.
According to equation (1), the graphite film had a gauge factor of approximate 0.75 under tensile strain, and meanwhile, a gauge factor of 1.3 was calculated under lateral compression. Both gauge factors of graphite
U
composite film were much lower than the GF’s absolute values of CNT composite film. This proved the
N
superiority of network structure of CNT composite film as strain sensor. CNT film and graphite composite films
A
both showed positive piezoresistive effect under tension. Under lateral compression CNT composite film
M
showed a negative piezoresistive effect while positive piezoresistive effect of graphite composite film was clearly observed. Although graphite composite film showed poor piezoresistive behavior, it gave us a hint that
ED
the mechanical behavior of film component is one of the most important factors responsible for piezoresistive
PT
response, especially under lateral compression.
3.2 The effect of polymer matrix on piezoresistive behavior under lateral compression
CC E
The compression of composite is a very complex issue, and compressive properties are dependent upon
matrix modulus and strength, interfacial bonding and the orientation of reinforcement [29]. It can be seen from Fig. 4a that CNT film had a porous microstructure. Carbon nanotubes can be further coated and pores were
A
filled by polymeric matrix to obtain nanocomposite. Interested in the influence of resin matrix on the piezoresistive behavior under lateral compression, we chose the same CNT film and three different kinds of resin to prepare different nanocomposites, including epoxy (EP), bismaleimide (BMI) and polyarylacetylene (PAA) resin. Fig.5 shows the resistance change with strain under lateral compression for these CNT composite films.
All these nanocomposites exhibited good linear relationship between ΔR/R and compressive strain. Regardless of matrix type, the resistance nearly recovered to its original value after several loading cycles. Among these composite film, CNT/PAA showed the biggest resistance change with compressive strain, while the resistance change of CNT/BMI film was the least significant. The resistance change is generally considered to be caused by the change in the intrinsic resistance of CNTs and the intertube resistance. The various matrix may cause
IP T
different load transferring between CNTs and matrix, resulting in the change in intrinsic resistance. Based on equation (1), the calculated gauge factors of CNT/BMI, CNT/EP and CNT/PAA composite films showed big difference, which were -1.9, -2.6 and -5.5, respectively. All these composite films behaved negative
SC R
piezoresistive response. Resin type not only affected the piezoresistive response but also their mechanical properties. Fig. 6 shows the typical tensile stress-strain curves for these three composite films. It was surprisingly found that the effect of matrix type on gauge factor was in accordance with the modulus of
U
composite film. The stress-strain curve of CNT/PAA composite was approximately linear, and a sudden stress
N
drop was observed at the maximum stress. Both CNT/EP and CNT/BMI showed obvious deviation from linear
A
stress-strain relationship, and the samples yielded accompanied with plastic deformation. Although these
M
composite films have similar strength, their modulus decreased in the order of CNT/PAA, CNT/EP and CNT/BMI, which were 14.3, 9.5, and 7.2 GPa in the strain range of 0-1%, respectively. For the composite film
ED
with larger modulus, bigger stress was transferred to individual CNTs at the same stain. As a result, CNTs' intrinsic resistance played more important role on their piezoresistive response. Even though the resistance
PT
change perpendicular to tensile direction was recorded, the deformation of CNT-based sensor was mainly decided by the polycarbonate materials, which deformed in both tensile and compressive direction during the
CC E
measurement. Therefore, the effect of polymeric matrix was equally implemented on the resistance change along and perpendicular to tensile direction. Consequently, the composite film with higher modulus still showed negative piezoresistive response under lateral compressive, and the absolute value of gauge factor became much
A
higher.
3.3 Effect of functionalization and acid treatment on piezoresistive behavior of CNT film The interaction between CNTs and matrix plays important role on load transfer. The strong interfacial bonding may significantly improve the mechanical property of final nanocomposite. In order to improve the
interfacial bonding between CNTs and epoxy resin, pristine CNT film was functionalized with potassium permanganate solution to create carboxyl groups on CNT surface. The resultant covalent bonding between CNTs and epoxy resin may result in a stronger interfacial cohesion. CNT film was treated for 1 h and 2 h to achieve different degree of functionalization. The resultant composites had electrical resistivity of 1.4×10-5 Ωm and 1.7×10-5 Ωm with 1 h and 2 h treatment, respectively. Fig.7 shows the resistance change of composite films
IP T
under lateral compression. All these nanocomposite showed linear relationship between resistance change and strain, and negative piezoresistive response was observed. Particularly after 1h functionalization, CNT composite film showed more obvious resistance increase when the absolute value of laterally compressive strain
SC R
became bigger. This was closely related to the strong interfacial bonding which effectively prohibited CNT slippage. Thus, the variation of CNT's intrinsic resistance became more obvious subjected to the same applied strain, leading to stronger piezoresistive response. However, the CNT composite film with 2h functionalization
U
showed slighter piezoresistive response than pristine composite film. This was due to the structure damage after
N
long-period functionalization. Fig. 8 shows the typical tensile stress-strain curves of functionalized composite
A
films. One-hour treatment greatly increased tensile strength and modulus, and less plastic deformation was
M
observed. Two-hour treatment reduced the tensile properties. This also evidenced the enhancement of interfacial bonding after 1h treatment and structural damage after 2h treatment. Due to the improvement of CNT network
ED
stability, ΔR/R turns back to zero after five cycles. According to equation (1), 1h functionalization increased the GF’s absolute value of composite film under lateral
PT
compression by 127% which was -5.9. This value of gauge factor was even larger than PAA composite. The
CC E
composite film with 2h treatment had a gauge factor of -1.6 under lateral compression.
In addition, the mixture of sulfuric acid and nitric acid was used to treat CNT film. The electrical resistivity
was decreased to 9.1×10-6 Ωm due to p-type doping. Fig. 9 shows the resistance change of acid treated CNT
A
composite film under lateral compression. The piezoresistive response of final composite became less sensitive than pristine composite film. The resultant gauge factor was only -1.0. The acid treatment only changed electrical conductivity, but had no influence on load transferring between CNTs and matrix. By considering all these influence factors on piezoresistive response, it implied that the tailoring of mechanical property was one of the most effective method to improve the electromechanical characteristic of CNT composite film.
4. Conclusions This paper investigated the potential of floating catalyst chemical vapor deposition (FCCVD)-grown CNT composite as a sensor to monitor biaxial strain. The piezoresistive response of CNT composite film under tension and lateral compression was investigated, and it was compared with a graphite composite film to reveal
IP T
the relation between microstructure and electromechanical property. Focusing on the piezoresistive response under lateral compression, the effects of matrix type, sidewall epoxidation and acid treatment were studied. It was found that the resistance change varied linearly with increasing applied strain, and CNT composite film
SC R
exhibited positive piezoresistive response under tension and negative piezoresistive effect under laterally compressive load. After eliminating the effect of geometrical deformation on gauge factor, the change in resistivity is more sensitive under lateral compression than under tension, and the resultant dρ/ρ0dε of CNT
U
composite film is -6.2 under lateral compression, which is more sensitive than that under tensile strain. Different
N
from CNT composite film, graphite composite film showed less obvious piezoresistive behavior, and exhibited
A
positive piezoresistive response regardless of tensile and laterally compressive loads. The CNT composite film
M
with network structure consisting of numerous contact junctions was more sensitive than graphite composite film with layered structure. Moreover, piezoresistive response became stronger in the order of CNT/BMI,
ED
CNT/EP, and CNT/PAA. It was found that the modulus of composite films played an important role on its piezoresistive response. Stronger interfacial bonding was also proved to further enhance the piezoresistive
PT
response and increase the corresponding gauge factor. CNT/EP with 1h KMnO4 treatment reached a gauge factor of -5.9. These results revealed the key factors tailoring piezoresistive response of CNT composite film,
CC E
and suggested for one possible approach to monitor Possion effect. Acknowledgements
This work was supported by the National Natural Science Foundation of China (Grant No. 51403009 and
A
51503225). The authors thank Suzhou Institute of Nano-tech and Nano-bionics, Chinese Academy of Sciences for supplying carbon nanotube film. References [1] M.A. Cullinan, M.L. Culpepper, Effects of chirality and impurities on the performance of carbon nanotubebased piezoresistive sensors, Carbon 51 (2013) 59-63.
[2] C. Stampfer, A. Jungen, R. Linderman, D. Obergfell, S. Roth, C. Hierold, Nano-electromechanical displacement sensing based on single-walled carbon nanotubes, Nano Letters 6 (2006) 1449-53. [3] K.J. Loh, J.P. Lynch, B.S. Shim, N.A. Kotov, Tailoring piezoresistive sensitivity of multilayer carbon nanotube composite strain sensors, J. Intell. Mater. Syst. Struct. 19 (2008) 747-64. [4] I. Kang, M.J. Schulz, J.H. Kim, V. Shanov, D. Shi, A carbon nanotube strain sensor for structural health
IP T
monitoring, Smart Mater. Struct. 15 (2006) 737-48. [5] J.R. Bautista-Quijano, F. Avilés, J.O. Aguilar, A. Tapia, Strain sensing capabilities of a piezoresistive
SC R
MWCNT-polysulfone film, Sensors and Actuators A: Physical 159 (2010) 135-40.
[6] N. Tamil Selvan, S.B. Eshwaran, A. Das, K.W. Stöckelhuber, S. Wießner, et al., Piezoresistive natural rubbermultiwall carbon nanotube nanocomposite for sensor applications, Sensors and Actuators A: Physical 239
U
(2016) 102-13.
N
[7] A. Sanli, A. Benchirouf, C. Müller, O. Kanoun, Piezoresistive performance characterization of strain
A
sensitive multi-walled carbon nanotube-epoxy nanocomposites, Sensors and Actuators A: Physical 254
M
(2017) 61-68.
[8] A. Li, A.E. Bogdanovich, P.D. Bradford, Aligned carbon nanotube sheet piezoresistive strain sensors, Smart
ED
Mater. Struct. 24 (2015) 095004.
[9] J.W. Zha, K. Shehzad, W.K. Li, Z.M. Dang, The effect of aspect ratio on the piezoresistive behavior of the
PT
multiwalled carbon nanotubes/thermoplastic elastomer nanocomposites, J. Appl. Phys. 113 (2013) 014102. [10] Y. Wang, A.X. Wang, Y. Wang, M.K Chyu, Q.M. Wang, Fabrication and characterization of carbon
CC E
nanotube-polyimide composite based high temperature flexible thin film piezoresistive strain sensor, Sensors and Actuators A -Physical, 199 (2013) 265-71. [11] Z. Wang, X. Ye, A numerical investigation on piezoresistive behaviour of carbon nanotube/polymer
A
composites: mechanism and optimizing principle, Nanotechnology 24 (2013) 265704.
[12] A.I. Oliva-Avile´s, F. Avile´s, V. Sosa, Electrical and piezoresistive properties of multi-walled carbon nanotube/polymer composite films aligned by an electric field, Carbon 49 (2011) 2989-97. [13] J.L. Abot, C.Y. Kiyono, G.P. Thomas, E.C.N. Silva, Strain gauge sensors comprised of carbon nanotube yarn: parametric numerical analysis of their piezoresistive response, Smart Mater. Struct. 24 (2015) 075018.
[14] A. Lekawa-Raus, K.K. Koziol, A.H. Windle, Piezoresistive effect in carbon nanotube fibers, ACS Nano 8 (2014) 11214-24. [15] Y. Miao, L. Chen, Y. Lin, R. Sammynaiken, W.J. Zhang, On finding of high piezoresistive response of carbon nanotube films without surfactants for in-plane strain detection, J. Intell. Mater. Syst. Struct. 22 (2011) 2155-9.
IP T
[16] J. Pan, M. Li, S. Wang, Y. Gu, Q. Li, Z. Zhang, Hybrid effect of carbon nanotube film and ultrathin carbon fiber prepreg composites, J Reinf. Plast. Comp. 36 (2017) 452-63.
[17] A. Eitan, K. Jiang, D. Dukes, R. Andrews, L.S. Schadler. Surface modification of multiwalled carbon
SC R
nanotubes: toward the tailoring of the interface in polymer composites. Chem. Mater. 15 (2003) 3198-201. [18] F. Avilés, A. May-Pat, G. Canché-Escamilla, O. Rodríguez-Uicab, J.J.Ku-Herrera, S. Duarte-Aranda,
N
composites, J. Intell. Mater. Syst. Struct. 27 (2016) 92-103.
U
Influence of carbon nanotube on the piezoresistive behavior of multiwall carbon nanotube/polymer
strain sensing, Nanotechnology 15 (2004) 379-82.
A
[19] P. Dharap, Z. Li, S. Nagarajaiah, E.V. Barrera, Nanotube film based on single-wall carbon nanotubes for
M
[20] Z. Wang, X. Ye. An investigation on piezoresistive behavior of carbon nanotube/polymer composites: II.
ED
Positive piezoresistive effect, Nanotechnology 25 (2014) 285502. [21] S.H. Hwang, H.W. Park, Y.B. Park, Piezoresistive behavior and multi-directional strain sensing ability of
PT
carbon nanotube–graphene nanoplatelet hybrid sheets, Smart Mater. Struct. 22 (2013) 015013. [22] S. Luo, T. Liu, Structure-property-processing relationships of single-wall carbon nanotube thin film
CC E
piezoresistive sensors, Carbon 59 (2013) 315-24. [23] Q. Liu, M. Li, Y. Gu, Y. Zhang, S. Wang, et al., Highly aligned dense carbon nanotube sheets induced by multiple stretching and pressing, Nanoscale 6 (2014) 4338-44.
A
[24] S. Wang, R. Downes, C. Young, D. Haldane, A. Hao, R. Liang, Carbon fiber/carbon nanotube buckypaper interply hybrid composites: manufacturing process and tensile properties, Advanced engineering materials 17 (2015) 1442-53. [25] Q Liu, M Li, Y Gu, S Wang, Y Zhang, Q Li, et al., Interlocked CNT networks with high damping and storage modulus, Carbon 86 (2015) 46-53.
[26] A. Benchirouf, C. Müller, O. Kanoun, Electromechanical behavior of chemically reduced graphene oxide and multi-walled carbon nanotube hybrid material, Nanoscale Research Letters 11 (2016) 1-7. [27] L. Vertuccio, L. Guadagno, G. Spinelli, P. Lamberti, V. Tucci, S. Russo, Piezoresistive properties of resin reinforced with carbon nanotubes for health-monitoring of aircraft primary structures, Composites Part B: Engineering 107 (2016) 192-202.
multi-walled carbon nanotube films, Diam. Relat. Mater. 16 (2007) 388–92.
IP T
[28] C.L. Cao, C.G. Hu, Y.F. Xiong, X.Y. Han, Y. Xi, J. Miao, Temperature dependent piezoresistive effect of
[29] G. Zhang, R.A. Latorur, FRP composite compressive strength and its dependence upon interfacial bond
SC R
strength, fiber misalignment, and matrix nonlinearity, Journal of Thermoplastic Composite Materials 6 (1993)
A
CC E
PT
ED
M
A
N
U
298-311.
Author Biographies Yin Wang received the B.S. degree from Beihang University in 2014. In the same year he began to pursue his master degree in Materials Science and Engineering at Beihang University. His current research focuses on the electromechanical performance of carbon nanotube composite film.
IP T
Shaokai Wang received the Ph. D degree in Materials Science from Beihang University in 2011. In 2011 he joined High-Performance Materials Institute at Florida State University, where he was a research associate. Since 2014 he has been working at the School of Materials Science and Engineering, Beihang Univeristy. His
SC R
research interests include the mechanical property, electromechanical performance, and thermal conductivity of
U
nanocomposite.
Min Li received the Ph. D degree in Materials Science from Beihang University in 2005. In the same year, she
N
joined the School of Materials Science and Engineering at Beihang University, where she is currently a Professor.
M
functionalization of carbon nanotube film.
A
Her research interests include the processing, mechanical and physical property, microstructure control and
ED
Yizhuo Gu received the Ph. D degree in Materials Science from Beihang University in 2007. In the same year, he joined the School of Materials Science and Engineering at Beihang University, where he is currently an
PT
Associate Professor. His research interests include the processing and quality control of polymer composite.
CC E
Zuoguang Zhang received the B.S. degree from Beihang University, China, in 1978 and the M.S and Ph. D degrees in Materials Science from Beihang University in 1986 and 1997, respectively. In 1986 he joined the School of Materials Science and Engineering at Beihang University as a Professor. His current research interests
A
include the manufacturing technique and theory of advanced polymer composite.
Tensile force Tension specimen
Lateral compression at backside
SC R
Digital Multimeter
IP T
CNT composite film
Fig.1 Schematic of the setup used for measuring piezoresistive behavior under tension and lateral
A
CC E
PT
ED
M
A
N
U
compression.
1.2 6
1.0
0.6 2
0.4 0.2
0
0.0 0
50
100 150 Time (s)
N M
-1.0 -0.8
-0.4 -0.2
4
2
ED
-0.6
Compressive strain ΔR/R
0
PT
Strain (%)
6
A
-1.2
-2
U
(a)
200
0.0
A
CC E
0.2
0
50
100
150
200
Time (s) (b)
Fig. 2 Piezoresistive behavior of CNT composite film under tension and lateral compression.
-2
ΔR/R (%)
-0.2
SC R
ΔR/R
Tensile strain
ΔR/R (%)
4
IP T
Strain (%)
0.8
1.2
3
1.0
1
0.4 0.2
0
ΔR/R
Tensile strain 0
50
100 150 Time (s)
200
-1
N
U
(a)
SC R
-0.2
IP T 3
A
-1.2 -1.0
M
Compressive strain 2
ΔR/R
-0.8
ED
-0.6 -0.4
0
PT
-0.2
1
0.0
A
CC E
0.2
0
50
100
150
200
Time (s) (b)
Fig. 3 Piezoresistive behavior of graphite composite film under tension and lateral compression.
-1
ΔR/R (%)
Strain (%)
0.6
0.0
Strain (%)
ΔR/R (%)
2
0.8
IP T
(b)
PAA composite Lateral compressive strain
U N A
-0.2
50
PT
0
ED
0.0
100 150 Time (s)
1 0
200
CC E
Fig. 5 The resistance change of CNT/BMI, CNT/EP and CNT/PAA nanocomposite film.
A
3 2
M
Strain (%)
-0.4
4
-1
ΔR/R (%)
Epoxy composite BMI composite
SC R
Fig.4 SEM morphology of (a) CNT film and (b) graphite film.
200
BMI composite PAA composite Epoxy composite
50
0
IP T
100
0
1
2
3
Strain (%)
SC R
Stress (MPa)
150
4
5
M
2
ED
-0.2
4 3
1
PT
Strain (%)
-0.4
1h treated CNT composite film Lateral compressive strain
0
CC E
0.0
0
100 Time (s)
200
-1
A
Fig. 7 Resistance change of CNT nanocomposite film with different degrees of functionalization.
ΔR/R (%)
A
Pristine CNT composite film 2h treated CNT composite film
N
U
Fig. 6 The tensile stress-strain curves of CNT/PAA, CNT/EP and CNT/BMI composite films.
300
100
0
0
1
2 3 Strain (%)
SC R
1h treatment 2h treatment Pristine CNT composite film
IP T
Stress (MPa)
200
4
5
U
Fig.8 Stress-strain curve of CNT/EP nanocomposite film with and without functionalization.
N
4
Lateral compressive strain Acid treatment Pristine composite film
2 1
PT CC E
0
0
100 Time (s)
Fig. 9 Resistance change of CNT nanocomposite film after acid treatment.
200
-1
ΔR/R (%)
A
M -0.2
0.0
A
3
ED
Strain (%)
-0.4
S0924-4247(17)30460-0 https://doi.org/10.1016/j.sna.2018.02.032 SNA 10654
To appear in:
Sensors and Actuators A
Received date: Revised date: Accepted date:
19-3-2017 18-2-2018 19-2-2018
Please cite this article as: Wang Y, Wang S, Li M, Gu Y, Zhang Z, Piezoresistive response of carbon nanotube composite film under laterally compressive strain, Sensors and Actuators: A Physical (2010), https://doi.org/10.1016/j.sna.2018.02.032 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Piezoresistive response of carbon nanotube composite film under laterally compressive strain Yin Wang, Shaokai Wang*, Min Li, Yizhuo Gu and Zuoguang Zhang Key Laboratory of Aerospace Advanced Materials and Performance (Ministry of Education), School of
100191, China
Highlights
This study explored the piezoresistive behavior of high CNT-loading carbon nanotube composite film
U
to advance the self-sensing ability of structural composite.
The relationship between microstructure and piezoresistive response for typical carbonaceous film
N
SC R
*Corresponding author: Tel & Fax: +86-10-82339575, E-mail: [email protected]
IP T
Materials Science and Engineering, Beihang University, No. 37 Xueyuan Road, Haidian District, Beijing
The piezoresistive behavior of CNT composite film was effectively tailored by polymeric matrix and
M
A
material was revealed by comparing CNT and graphite composite films.
ED
CNT functionalization.
Abstract：
PT
Carbon nanotubes (CNTs) have attracted great attention for strain sensor application due to their excellent electromechanical property. This paper focused on the piezoresistive response of high CNT loading composite
CC E
film. The piezoresistive behavior of floating catalyst chemical vapor deposition (FCCVD)-grown CNT composite film under tension and lateral compression was investigated and compared with a graphite composite film. The effects of matrix type, sidewall epoxidation and acid treatment on piezoresistive response of CNT
A
composite film were discussed. The results showed that CNT composite film exhibited positive piezoresistive response under tension and negative piezoresistive effect under lateral compression. After eliminating the effect of geometrical deformation on gauge factor, the change in resistivity was more sensitive under lateral compression than under tension. Different from CNT composite film, graphite composite film showed less obvious piezoresistive behavior, and showed positive piezoresistive response regardless of tensile and laterally compressive loads. Furthermore, the piezoresistive response of CNT composite film was found to closely
correlate with its modulus. Stronger interfacial bonding was proved to effectively enhance the piezoresistive response and increase the corresponding gauge factor. CNT/epoxy composite film with 1h treatment in potassium permanganate solution reached a gauge factor of -5.9. These results revealed the key factors tailoring piezoresistive response of CNT composite film, and presented one potential approach to monitor biaxial strain. The present work should be useful in the in-plane strain monitoring by incorporating the CNT film into
A
CC E
PT
ED
M
A
N
U
SC R
Key words: Carbon nanotube film; piezoresistive response; gauge factor; compression.
IP T
traditional composite.
1. Introduction Carbon nanotubes possess excellent electromechanical property and show great potential as piezoresistive strain sensor due to the alteration of band structure accompanied with its mechanical deformation [1, 2]. How to take CNT's advantages in macroscopic CNT-based piezoresistive sensor has attracted lots of research attention. As one of the most widely studied CNT-based sensor, polymer film mixed with CNTs has fully
IP T
demonstrated its piezoresistive behavior due to the change of tube-tube contact in CNT network. This kind of CNT-based sensors have been well developed, such as single-walled carbon nanotube (SWNT)-polyelectrolyte (PE) film [3], CNT-poly (methyl methacrylate) (PMMA) film [4], multi-wall carbon nanotubes (MWCNT)-
SC R
polysulfone film [5], MWCNT and natural rubber nanocomposite [6]. The MWCNTs/epoxy nanocomposites have been reported to display a gauge factor as high as 78, and also exhibited excellent stability and durability [7]. The microstructure of inside CNT network was recognized to play an important role on piezoresistive
U
response, including network density, degree of CNT alignment [8], aspect ratio of CNTs [9], CNT concentration
N
[10], and so on. Wang et al. [11] conducted numerical simulation, and found that the junction resistance between
A
CNTs decided the piezoresistive behavior of these polymer film sensor. The variation of average junction gap,
M
describing the conductance change of CNT network, was influenced by the orientation and diameter of CNTs, the poisson's ratio of polymer matrix, and CNT concentration. Oliva-Avile´s [12] found the positive impact of
ED
CNT alignment on the piezoresistive sensitivity of these CNT-based sensors. High CNT-loading assemblies, e.g. CNT fiber [13] and CNT film, have been well developed in recent years.
PT
These CNT assemblies were widely considered to be the next-generation high-performance structural reinforcements for aerospace-grade composite. Lekawa-Raus reported the piezoresistive response of CNT fiber
CC E
accompanied with the opening of bandgaps and reduction of the contact area between CNT bundles [14]. As a freestanding mat of densely-packed and inter-tangled carbon nanotubes, CNT film also displayed resistance change under applied force. Miao Y. et al [15] prepared pure CNT film and found its gauge factor was about
A
2.4. Although the gauge factors of these high loading CNT assembly were much smaller than the above CNT nanocomposites, their piezoresistive behavior may greatly advance the structure-function for these CNT materials. The loading transfer inside CNT composite film was greatly influenced by the interactions between CNTs and polymeric matrix. Functionalization and acid treatment on CNTs to improve its surface features have been widely investigated for higher mechanical and electrical properties [16, 17]. Still, how to tailor the
electromechanical performance of high-loading CNT materials remains challenge. Several researches have been carried out to investigate the piezoresistive response of various CNT-based sensors under different load types. Tensile force was the most common load type to measure sensor's piezoresistive behavior [18]. Besides, Kang et al adhered SWCNT buckypaper or CNT composite on a cantilever beam, and measured their resistance change with respect to the displacement due to bending. These
IP T
sensors showed linear resistance variation with strain, and had good sensitivity when these sensors were tensioned or compressed [4]. Dharap found the similar piezoresistive behavior of SWCNT buckypaper when it was subjected to unidirectional tensile and compressive stresses [19]. For CNT/polymer film, different matrices
SC R
may result in quite different piezoresistive behaviors, and it was found that the obstacle effect of polymer chains on the movement of CNTs played an important role on piezoresistive effect under compressive strain [20]. Freestanding MWCNT/ graphene nanoplatelet hybrid film also showed electrical resistance increase with
U
increasing strain undergoing flexural force, and the resistance change was more obvious in the tensile direction
N
[21]. These results showed the capability of CNT-based sensor for monitoring tensile and compressive strain.
A
This makes it possible to detect multi-axial strain with the same kind of CNT materials, and the multifunctional
M
composite made of CNT assembly may self-sense its mechanical behavior, e.g. tensile and compressive strain, and even Poisson effect under complex loading conditions. The structure–property–processing relationship has
ED
been widely studied under tension [22], but the influence factors on piezoresistive response under compression were rarely reported.
PT
This paper explored the potential of floating catalyst chemical vapor deposition (FCCVD)-grown CNT composite film as a strain sensor to monitor biaxial strain under tensile test. The piezoresistive behavior of CNT
CC E
composite film under tension and lateral compression was investigated. Focusing on the piezoresistive response under lateral compression, the effects of matrix type, functionalization and acid treatment were further studied. 2. Experimental
A
2.1. Materials
The experimentally used CNT film was supplied by the Suzhou Institute of Nano-tech and Nano-bionics.
The film was synthesized using floating catalyst CVD-growth method. In the process carbon nanotubes formed a continuous aerogel which was wound by a roller. The aerogel was further densified by spraying ethanol solution to form multilayered seamless CNT film. The prepared carbon nanotubes had 3-8 walls and the film
had a thickness of 8-12 μm. CNTs are randomly oriented in the as-received film. This kind of CNT film can be directly used as reinforcement in composite material [23] or hybridized with other fiber [24], fully taking advantages of its excellent mechanical and physical properties [25]. CNT composite film was further prepared by impregnating polymer materials. The polymeric matrix includes epoxy (EP), bismaleimide (BMI) and phenylacetylene (PAA). The epoxy E51 resin used in this
IP T
experiment was purchased from Jiangsu Wuxi Resin Works. BMI resin was provided by Institute of Chemistry, Chinese Academy of Sciences. PAA resin was provided by East China University of Science and Technology. To prepare CNT composite film, acetone was used to dissolve epoxy and PAA resin, and a homogeneous
SC R
solution was obtained with a resin concentration of 5 wt%. To achieve CNT nanocomposite film with different modulus, BMI resin was dissolved in N, N-dimethylformamide (DMF) with a concentration of 1 wt%. CNT films were soaked in resin solution for 1 h. Subsequently, the resin-impregnated CNT films were dried in a
U
vacuum oven at 30 oC for 2 h to remove the solvent. The pre-impregnated CNT films were free-standingly cured
N
in an oven to get the final composites. The epoxy resin was cured at 130 oC/2 h+150 oC /1 h+180 oC/2 h. BMI
A
resin was cured at 80 °C/1 h+150 °C/2 h+190 °C/1 h+220 °C/2 h. The curing schedule of PAA resin was
M
115 °C/10 h +120 °C/4 h +140 °C/2 h +160 °C/2 h +180 °C/2 h+250 °C/4 h. For comparison, another
Materials Technology Co., Ltd.
ED
carbonaceous film (graphite film) was used in this experiment, which was purchased from the Sixth Elements
In order to investigate the effect of CNT film's surface state on piezoresistive behavior, functionalization
PT
and acid treatment on CNT film were carried out, respectively. Potassium permanganate was used to functionalize CNT film and introduce carboxyl group. 5ml concentrated sulfuric acid was evenly dispersed in
CC E
100 mL KMnO4 solution with a concentration of 0.3 wt%. CNT film was soaked in KMnO4/H2SO4 solution for 1h or 2 h to achieve different degree of functionalization. Afterwards, the CNT film was washed five times with diluted hydrochloric acid and five times with deionized water to a pH value of 7. The treated film was then
A
dried in a vacuum oven at 70 oC for 2 h. Sulfric acid and nitric acid were used in acid treatment. CNT films were soaked in the mixture of sulfuric acid and nitric acid (3:1) for 15 minutes. Higher conductivity was achieved because the mixture solution can introduce p-type doping into the CNTs network. Then CNT film was washed with deionized water and dried in a vacuum oven.
2.2. Characterization The morphology of CNT film material was examined on a JEOL JSM-7500F scanning electronmicroscope (SEM). The electrical conductivity was measured using a dielectric impedance analyzer (Guangzhou Four Probes Tech Co., Ltd. RTS-9). The tensile property of CNT film material was measured on a Instron 3344 mechanical testing machine. The rectangular specimens with a gauge length of 10 mm were measured at a speed
IP T
of 0.5 mm/min. Five specimens were tested for each sample. To investigate the piezoresistive behavior of CNT composite film under in-plane tensile and lateral compressive loads, two pieces of composite film with dimensions of 5×10 mm were adhered to both surfaces
SC R
of tensile specimen as shown in Fig. 1. Polycarbonate specimens for tensile test were prepared according to ASTM D 638 standard. The specimen has a gauge length of 60 mm. CNT films were adhered to the specimen by using epoxy adhesive. After the adhesive was cured, electrodes were bonded with the film materials by using
U
silver paste. The specimen was then loaded onto Instron 5565 mechanical testing machine. A displacement
N
speed of 1.5 mm/min was applied on the sample. An auto DC digital multimeter was used to monitor the
A
resistance variation. The resistance changes along and perpendicular to the tensile direction were monitored,
M
respectively. The piezoresistive sensitivity was quantified by gage factor (GF) K as follow [26, 27]:
Δ𝑅 𝑅Δ𝜀
(1)
ED
K=
where R represents the initial resistance of film material, and ΔR is the resistance change in response to applied
PT
strain. Δε is the strain of CNT composite film. Δε corresponds to tensile strain or laterally compressive strain respectively when CNT composite film was placed along or perpendicular to tensile direction. Here, tensile and
CC E
laterally compressive strains are regarded as positive and negative strain, respectively. With the increase of strain, positive ΔR results in positive gauge factor indicating a positive piezoresistive effect, whereas a negative
A
piezoresistive effect is defined.
3. Results and Discussion 3.1 Comparison of piezoresistive behavior between different carbonaceous film materials In the tensile measurement, the specimen was stretched along tensile direction and shrank in the width direction. Correspondingly, the attached CNT composite films were tensioned and laterally compressed along and perpendicular to the tensile direction, respectively. Fig. 2 shows the piezoresistive behavior of CNT/epoxy
film under tension (along longitudinal direction) and lateral compression (along specimen's width direction), respectively. The resistance change showed good linear relationship with applied strain under both tensile and laterally compressive loads. By linearly fitting resistance change and strain plots, the coefficients of determination reached 0.99 and 0.98 in Fig. 2a and b, respectively. The resistance showed cyclical variation when exposed to cyclic loading as shown in Fig. 2. The linearity and stability of resistance-strain relationship
IP T
show the great potential for the application as strain sensor. Under tensile force the resistance linearly increased with increasing strain as shown in Fig. 2a. The piezoresistive response was generally considered to be caused by the change of CNTs' intrinsic resistance
SC R
subjected to applied strain and intertube resistance resulted from the change of CNT network structure. Along tensile direction, the intertube resistance increased due to the increase of tube-tube contact. Due to the Poisson effect, the deformation of polycarbonate also caused the shrinkage of adhered CNT composite film along the
U
width direction. The polycarbonate specimen had a Poisson's ratio of 0.39. When 1% strain was applied along
N
tensile direction, the transverse strain was about -0.39%. The changing trend in Fig. 2a indicated that the
A
resistance change was mainly controlled by the piezoresistive response along tensile direction. On the other
M
surface of polycarbonate sample, the contraction transverse to tensile direction resulted in negative strain of CNT composite film. Surprisingly, its resistance showed an increasing tendency with the increase of the absolute
ED
value of transverse strain under lateral compression, as shown in Fig. 2b. This was totally different from the reported results under unidirectional compression [19], where the resistance decreased with the increase in the
PT
absolute value of compressive strain. This phenomenon indicated that besides the lateral compressive strain, the deformation of CNT composite film along tensile direction also affected the transverse resistance. From the
CC E
SEM morphology of CNT film in Fig. 4a, numerous carbon nanotubes entangled in the film. The geometrical deformation caused the increase of contact resistance, and consequently the overall resistance increased with the increase in the absolute value of compressive strain. Moreover, after five loading-unloading cycles, ΔR/R
A
became negative, indicating the decrease of resistance. This should be related to the uncompacted microstructure and bundling effect of CNTs. Based on equation (1), CNT composite film had a gauge factor of 4.5 under tensile strain, which was bigger than reported CNT film without surfactant (GF=2.4) [15] but lower than CNT fiber produced with a similar process (GF=15) [14]. Under lateral compression a smaller gauge factor of -2.6 was calculated, which was the
result of a combination of geometrical deformation and resistivity change. The prominent piezoresistive response along both tensile and transverse direction suggested the possibility of CNT composite film as strain sensor monitoring biaxial strain. The gauge factor K is related to the geometrical deformation and the change in electrical resistivity, which is defined as equation (2) [28]. d𝜌 𝜌0 d𝜀
(2)
IP T
K=1 + 2ν +
where ρ and ρ0 are resistivities at strain ε and zero, respectively. ν is the Poisson’s ratio of the film. (1+2ν)
SC R
represents the effects of geometrical deformation, and dρ/ρ0dε is the contribution of the change in electrical resistivity. Since the CNT composite film was bonded on a substrate, its in-plane geometrical deformation was the same as the tensile specimen. The CNT composite film was free-standingly cured, which kept the
U
morphology of CNT network. Our previous study showed that the film’s thickness increased under uniaxial
N
tension [25]. Introducing polymer matrix may inhibit the expansion of CNT network. Herein, the geometrical deformation in thickness was ignored in this study. Under tension the effect of geometrical deformation was
A
simplified to 1+νpc, where νpc is the Poisson’s ratio of tensile specimen. According to the gauge factor, under
M
tensile strain dρ/ρ0dε is equal to 3.1. Under lateral compression, the effect of geometrical deformation is equal
ED
to 1+1/νpc, which causes the decrease in resistance. As a result, dρ/ρ0dε is calculated to be -6.2. This suggests
PT
that the resistivity change of CNT composite film is more sensitive under lateral compression than under tension.
In order to elucidate the relationship between structure and electromechanical property of carbonaceous
CC E
material, we also measured a graphite film. Fig. 3 shows the resistance change of graphite composite film under tension and lateral compression. The resistance change displayed an increase tendency with increasing tensile strain, which was the same as CNT composite film. However, ΔR/R and strain did not show good linear
A
relationship, and the fitting line had a coefficient of determination 0.92. The graphite film possessed a layered structure consisting of graphite grains as shown in Fig. 4. Under tension the contact resistance between graphite grains played an important role on resistance change. Graphite grains were connected by Van der Waal’s force, and the slippage between graphite grains might randomly occur, leading to the deviation from linear relation. Different from CNT composite film, the resistance change of graphite composite film under lateral compression showed a contrary tendency. The resistance of graphite composite film decreased when the
absolute value of compressive strain became bigger, indicating the occurrence of positive piezoresistive effect. The graphite grains were stiffer and more difficult to buckle than carbon nanotubes under the same strain. Strong van der Waals' force between graphite grains was beneficial to its structural stability. Furthermore, the compression stress may reduce the contact gap between graphite grains, resulting in the resistance decrease. After five loading cycles the resistance was observed to reach a bigger value than initial resistance, indicating
IP T
the instability of layered structure of graphite film. It should be pointed out that the resistance change was not consistent with strain for graphite film. When the biggest compressive strain was applied or the tensile strain was totally unloaded, the resistance did not reach its minimum value. This might be related to the local buckling
SC R
in the film under compression.
According to equation (1), the graphite film had a gauge factor of approximate 0.75 under tensile strain, and meanwhile, a gauge factor of 1.3 was calculated under lateral compression. Both gauge factors of graphite
U
composite film were much lower than the GF’s absolute values of CNT composite film. This proved the
N
superiority of network structure of CNT composite film as strain sensor. CNT film and graphite composite films
A
both showed positive piezoresistive effect under tension. Under lateral compression CNT composite film
M
showed a negative piezoresistive effect while positive piezoresistive effect of graphite composite film was clearly observed. Although graphite composite film showed poor piezoresistive behavior, it gave us a hint that
ED
the mechanical behavior of film component is one of the most important factors responsible for piezoresistive
PT
response, especially under lateral compression.
3.2 The effect of polymer matrix on piezoresistive behavior under lateral compression
CC E
The compression of composite is a very complex issue, and compressive properties are dependent upon
matrix modulus and strength, interfacial bonding and the orientation of reinforcement [29]. It can be seen from Fig. 4a that CNT film had a porous microstructure. Carbon nanotubes can be further coated and pores were
A
filled by polymeric matrix to obtain nanocomposite. Interested in the influence of resin matrix on the piezoresistive behavior under lateral compression, we chose the same CNT film and three different kinds of resin to prepare different nanocomposites, including epoxy (EP), bismaleimide (BMI) and polyarylacetylene (PAA) resin. Fig.5 shows the resistance change with strain under lateral compression for these CNT composite films.
All these nanocomposites exhibited good linear relationship between ΔR/R and compressive strain. Regardless of matrix type, the resistance nearly recovered to its original value after several loading cycles. Among these composite film, CNT/PAA showed the biggest resistance change with compressive strain, while the resistance change of CNT/BMI film was the least significant. The resistance change is generally considered to be caused by the change in the intrinsic resistance of CNTs and the intertube resistance. The various matrix may cause
IP T
different load transferring between CNTs and matrix, resulting in the change in intrinsic resistance. Based on equation (1), the calculated gauge factors of CNT/BMI, CNT/EP and CNT/PAA composite films showed big difference, which were -1.9, -2.6 and -5.5, respectively. All these composite films behaved negative
SC R
piezoresistive response. Resin type not only affected the piezoresistive response but also their mechanical properties. Fig. 6 shows the typical tensile stress-strain curves for these three composite films. It was surprisingly found that the effect of matrix type on gauge factor was in accordance with the modulus of
U
composite film. The stress-strain curve of CNT/PAA composite was approximately linear, and a sudden stress
N
drop was observed at the maximum stress. Both CNT/EP and CNT/BMI showed obvious deviation from linear
A
stress-strain relationship, and the samples yielded accompanied with plastic deformation. Although these
M
composite films have similar strength, their modulus decreased in the order of CNT/PAA, CNT/EP and CNT/BMI, which were 14.3, 9.5, and 7.2 GPa in the strain range of 0-1%, respectively. For the composite film
ED
with larger modulus, bigger stress was transferred to individual CNTs at the same stain. As a result, CNTs' intrinsic resistance played more important role on their piezoresistive response. Even though the resistance
PT
change perpendicular to tensile direction was recorded, the deformation of CNT-based sensor was mainly decided by the polycarbonate materials, which deformed in both tensile and compressive direction during the
CC E
measurement. Therefore, the effect of polymeric matrix was equally implemented on the resistance change along and perpendicular to tensile direction. Consequently, the composite film with higher modulus still showed negative piezoresistive response under lateral compressive, and the absolute value of gauge factor became much
A
higher.
3.3 Effect of functionalization and acid treatment on piezoresistive behavior of CNT film The interaction between CNTs and matrix plays important role on load transfer. The strong interfacial bonding may significantly improve the mechanical property of final nanocomposite. In order to improve the
interfacial bonding between CNTs and epoxy resin, pristine CNT film was functionalized with potassium permanganate solution to create carboxyl groups on CNT surface. The resultant covalent bonding between CNTs and epoxy resin may result in a stronger interfacial cohesion. CNT film was treated for 1 h and 2 h to achieve different degree of functionalization. The resultant composites had electrical resistivity of 1.4×10-5 Ωm and 1.7×10-5 Ωm with 1 h and 2 h treatment, respectively. Fig.7 shows the resistance change of composite films
IP T
under lateral compression. All these nanocomposite showed linear relationship between resistance change and strain, and negative piezoresistive response was observed. Particularly after 1h functionalization, CNT composite film showed more obvious resistance increase when the absolute value of laterally compressive strain
SC R
became bigger. This was closely related to the strong interfacial bonding which effectively prohibited CNT slippage. Thus, the variation of CNT's intrinsic resistance became more obvious subjected to the same applied strain, leading to stronger piezoresistive response. However, the CNT composite film with 2h functionalization
U
showed slighter piezoresistive response than pristine composite film. This was due to the structure damage after
N
long-period functionalization. Fig. 8 shows the typical tensile stress-strain curves of functionalized composite
A
films. One-hour treatment greatly increased tensile strength and modulus, and less plastic deformation was
M
observed. Two-hour treatment reduced the tensile properties. This also evidenced the enhancement of interfacial bonding after 1h treatment and structural damage after 2h treatment. Due to the improvement of CNT network
ED
stability, ΔR/R turns back to zero after five cycles. According to equation (1), 1h functionalization increased the GF’s absolute value of composite film under lateral
PT
compression by 127% which was -5.9. This value of gauge factor was even larger than PAA composite. The
CC E
composite film with 2h treatment had a gauge factor of -1.6 under lateral compression.
In addition, the mixture of sulfuric acid and nitric acid was used to treat CNT film. The electrical resistivity
was decreased to 9.1×10-6 Ωm due to p-type doping. Fig. 9 shows the resistance change of acid treated CNT
A
composite film under lateral compression. The piezoresistive response of final composite became less sensitive than pristine composite film. The resultant gauge factor was only -1.0. The acid treatment only changed electrical conductivity, but had no influence on load transferring between CNTs and matrix. By considering all these influence factors on piezoresistive response, it implied that the tailoring of mechanical property was one of the most effective method to improve the electromechanical characteristic of CNT composite film.
4. Conclusions This paper investigated the potential of floating catalyst chemical vapor deposition (FCCVD)-grown CNT composite as a sensor to monitor biaxial strain. The piezoresistive response of CNT composite film under tension and lateral compression was investigated, and it was compared with a graphite composite film to reveal
IP T
the relation between microstructure and electromechanical property. Focusing on the piezoresistive response under lateral compression, the effects of matrix type, sidewall epoxidation and acid treatment were studied. It was found that the resistance change varied linearly with increasing applied strain, and CNT composite film
SC R
exhibited positive piezoresistive response under tension and negative piezoresistive effect under laterally compressive load. After eliminating the effect of geometrical deformation on gauge factor, the change in resistivity is more sensitive under lateral compression than under tension, and the resultant dρ/ρ0dε of CNT
U
composite film is -6.2 under lateral compression, which is more sensitive than that under tensile strain. Different
N
from CNT composite film, graphite composite film showed less obvious piezoresistive behavior, and exhibited
A
positive piezoresistive response regardless of tensile and laterally compressive loads. The CNT composite film
M
with network structure consisting of numerous contact junctions was more sensitive than graphite composite film with layered structure. Moreover, piezoresistive response became stronger in the order of CNT/BMI,
ED
CNT/EP, and CNT/PAA. It was found that the modulus of composite films played an important role on its piezoresistive response. Stronger interfacial bonding was also proved to further enhance the piezoresistive
PT
response and increase the corresponding gauge factor. CNT/EP with 1h KMnO4 treatment reached a gauge factor of -5.9. These results revealed the key factors tailoring piezoresistive response of CNT composite film,
CC E
and suggested for one possible approach to monitor Possion effect. Acknowledgements
This work was supported by the National Natural Science Foundation of China (Grant No. 51403009 and
A
51503225). The authors thank Suzhou Institute of Nano-tech and Nano-bionics, Chinese Academy of Sciences for supplying carbon nanotube film. References [1] M.A. Cullinan, M.L. Culpepper, Effects of chirality and impurities on the performance of carbon nanotubebased piezoresistive sensors, Carbon 51 (2013) 59-63.
[2] C. Stampfer, A. Jungen, R. Linderman, D. Obergfell, S. Roth, C. Hierold, Nano-electromechanical displacement sensing based on single-walled carbon nanotubes, Nano Letters 6 (2006) 1449-53. [3] K.J. Loh, J.P. Lynch, B.S. Shim, N.A. Kotov, Tailoring piezoresistive sensitivity of multilayer carbon nanotube composite strain sensors, J. Intell. Mater. Syst. Struct. 19 (2008) 747-64. [4] I. Kang, M.J. Schulz, J.H. Kim, V. Shanov, D. Shi, A carbon nanotube strain sensor for structural health
IP T
monitoring, Smart Mater. Struct. 15 (2006) 737-48. [5] J.R. Bautista-Quijano, F. Avilés, J.O. Aguilar, A. Tapia, Strain sensing capabilities of a piezoresistive
SC R
MWCNT-polysulfone film, Sensors and Actuators A: Physical 159 (2010) 135-40.
[6] N. Tamil Selvan, S.B. Eshwaran, A. Das, K.W. Stöckelhuber, S. Wießner, et al., Piezoresistive natural rubbermultiwall carbon nanotube nanocomposite for sensor applications, Sensors and Actuators A: Physical 239
U
(2016) 102-13.
N
[7] A. Sanli, A. Benchirouf, C. Müller, O. Kanoun, Piezoresistive performance characterization of strain
A
sensitive multi-walled carbon nanotube-epoxy nanocomposites, Sensors and Actuators A: Physical 254
M
(2017) 61-68.
[8] A. Li, A.E. Bogdanovich, P.D. Bradford, Aligned carbon nanotube sheet piezoresistive strain sensors, Smart
ED
Mater. Struct. 24 (2015) 095004.
[9] J.W. Zha, K. Shehzad, W.K. Li, Z.M. Dang, The effect of aspect ratio on the piezoresistive behavior of the
PT
multiwalled carbon nanotubes/thermoplastic elastomer nanocomposites, J. Appl. Phys. 113 (2013) 014102. [10] Y. Wang, A.X. Wang, Y. Wang, M.K Chyu, Q.M. Wang, Fabrication and characterization of carbon
CC E
nanotube-polyimide composite based high temperature flexible thin film piezoresistive strain sensor, Sensors and Actuators A -Physical, 199 (2013) 265-71. [11] Z. Wang, X. Ye, A numerical investigation on piezoresistive behaviour of carbon nanotube/polymer
A
composites: mechanism and optimizing principle, Nanotechnology 24 (2013) 265704.
[12] A.I. Oliva-Avile´s, F. Avile´s, V. Sosa, Electrical and piezoresistive properties of multi-walled carbon nanotube/polymer composite films aligned by an electric field, Carbon 49 (2011) 2989-97. [13] J.L. Abot, C.Y. Kiyono, G.P. Thomas, E.C.N. Silva, Strain gauge sensors comprised of carbon nanotube yarn: parametric numerical analysis of their piezoresistive response, Smart Mater. Struct. 24 (2015) 075018.
[14] A. Lekawa-Raus, K.K. Koziol, A.H. Windle, Piezoresistive effect in carbon nanotube fibers, ACS Nano 8 (2014) 11214-24. [15] Y. Miao, L. Chen, Y. Lin, R. Sammynaiken, W.J. Zhang, On finding of high piezoresistive response of carbon nanotube films without surfactants for in-plane strain detection, J. Intell. Mater. Syst. Struct. 22 (2011) 2155-9.
IP T
[16] J. Pan, M. Li, S. Wang, Y. Gu, Q. Li, Z. Zhang, Hybrid effect of carbon nanotube film and ultrathin carbon fiber prepreg composites, J Reinf. Plast. Comp. 36 (2017) 452-63.
[17] A. Eitan, K. Jiang, D. Dukes, R. Andrews, L.S. Schadler. Surface modification of multiwalled carbon
SC R
nanotubes: toward the tailoring of the interface in polymer composites. Chem. Mater. 15 (2003) 3198-201. [18] F. Avilés, A. May-Pat, G. Canché-Escamilla, O. Rodríguez-Uicab, J.J.Ku-Herrera, S. Duarte-Aranda,
N
composites, J. Intell. Mater. Syst. Struct. 27 (2016) 92-103.
U
Influence of carbon nanotube on the piezoresistive behavior of multiwall carbon nanotube/polymer
strain sensing, Nanotechnology 15 (2004) 379-82.
A
[19] P. Dharap, Z. Li, S. Nagarajaiah, E.V. Barrera, Nanotube film based on single-wall carbon nanotubes for
M
[20] Z. Wang, X. Ye. An investigation on piezoresistive behavior of carbon nanotube/polymer composites: II.
ED
Positive piezoresistive effect, Nanotechnology 25 (2014) 285502. [21] S.H. Hwang, H.W. Park, Y.B. Park, Piezoresistive behavior and multi-directional strain sensing ability of
PT
carbon nanotube–graphene nanoplatelet hybrid sheets, Smart Mater. Struct. 22 (2013) 015013. [22] S. Luo, T. Liu, Structure-property-processing relationships of single-wall carbon nanotube thin film
CC E
piezoresistive sensors, Carbon 59 (2013) 315-24. [23] Q. Liu, M. Li, Y. Gu, Y. Zhang, S. Wang, et al., Highly aligned dense carbon nanotube sheets induced by multiple stretching and pressing, Nanoscale 6 (2014) 4338-44.
A
[24] S. Wang, R. Downes, C. Young, D. Haldane, A. Hao, R. Liang, Carbon fiber/carbon nanotube buckypaper interply hybrid composites: manufacturing process and tensile properties, Advanced engineering materials 17 (2015) 1442-53. [25] Q Liu, M Li, Y Gu, S Wang, Y Zhang, Q Li, et al., Interlocked CNT networks with high damping and storage modulus, Carbon 86 (2015) 46-53.
[26] A. Benchirouf, C. Müller, O. Kanoun, Electromechanical behavior of chemically reduced graphene oxide and multi-walled carbon nanotube hybrid material, Nanoscale Research Letters 11 (2016) 1-7. [27] L. Vertuccio, L. Guadagno, G. Spinelli, P. Lamberti, V. Tucci, S. Russo, Piezoresistive properties of resin reinforced with carbon nanotubes for health-monitoring of aircraft primary structures, Composites Part B: Engineering 107 (2016) 192-202.
multi-walled carbon nanotube films, Diam. Relat. Mater. 16 (2007) 388–92.
IP T
[28] C.L. Cao, C.G. Hu, Y.F. Xiong, X.Y. Han, Y. Xi, J. Miao, Temperature dependent piezoresistive effect of
[29] G. Zhang, R.A. Latorur, FRP composite compressive strength and its dependence upon interfacial bond
SC R
strength, fiber misalignment, and matrix nonlinearity, Journal of Thermoplastic Composite Materials 6 (1993)
A
CC E
PT
ED
M
A
N
U
298-311.
Author Biographies Yin Wang received the B.S. degree from Beihang University in 2014. In the same year he began to pursue his master degree in Materials Science and Engineering at Beihang University. His current research focuses on the electromechanical performance of carbon nanotube composite film.
IP T
Shaokai Wang received the Ph. D degree in Materials Science from Beihang University in 2011. In 2011 he joined High-Performance Materials Institute at Florida State University, where he was a research associate. Since 2014 he has been working at the School of Materials Science and Engineering, Beihang Univeristy. His
SC R
research interests include the mechanical property, electromechanical performance, and thermal conductivity of
U
nanocomposite.
Min Li received the Ph. D degree in Materials Science from Beihang University in 2005. In the same year, she
N
joined the School of Materials Science and Engineering at Beihang University, where she is currently a Professor.
M
functionalization of carbon nanotube film.
A
Her research interests include the processing, mechanical and physical property, microstructure control and
ED
Yizhuo Gu received the Ph. D degree in Materials Science from Beihang University in 2007. In the same year, he joined the School of Materials Science and Engineering at Beihang University, where he is currently an
PT
Associate Professor. His research interests include the processing and quality control of polymer composite.
CC E
Zuoguang Zhang received the B.S. degree from Beihang University, China, in 1978 and the M.S and Ph. D degrees in Materials Science from Beihang University in 1986 and 1997, respectively. In 1986 he joined the School of Materials Science and Engineering at Beihang University as a Professor. His current research interests
A
include the manufacturing technique and theory of advanced polymer composite.
Tensile force Tension specimen
Lateral compression at backside
SC R
Digital Multimeter
IP T
CNT composite film
Fig.1 Schematic of the setup used for measuring piezoresistive behavior under tension and lateral
A
CC E
PT
ED
M
A
N
U
compression.
1.2 6
1.0
0.6 2
0.4 0.2
0
0.0 0
50
100 150 Time (s)
N M
-1.0 -0.8
-0.4 -0.2
4
2
ED
-0.6
Compressive strain ΔR/R
0
PT
Strain (%)
6
A
-1.2
-2
U
(a)
200
0.0
A
CC E
0.2
0
50
100
150
200
Time (s) (b)
Fig. 2 Piezoresistive behavior of CNT composite film under tension and lateral compression.
-2
ΔR/R (%)
-0.2
SC R
ΔR/R
Tensile strain
ΔR/R (%)
4
IP T
Strain (%)
0.8
1.2
3
1.0
1
0.4 0.2
0
ΔR/R
Tensile strain 0
50
100 150 Time (s)
200
-1
N
U
(a)
SC R
-0.2
IP T 3
A
-1.2 -1.0
M
Compressive strain 2
ΔR/R
-0.8
ED
-0.6 -0.4
0
PT
-0.2
1
0.0
A
CC E
0.2
0
50
100
150
200
Time (s) (b)
Fig. 3 Piezoresistive behavior of graphite composite film under tension and lateral compression.
-1
ΔR/R (%)
Strain (%)
0.6
0.0
Strain (%)
ΔR/R (%)
2
0.8
IP T
(b)
PAA composite Lateral compressive strain
U N A
-0.2
50
PT
0
ED
0.0
100 150 Time (s)
1 0
200
CC E
Fig. 5 The resistance change of CNT/BMI, CNT/EP and CNT/PAA nanocomposite film.
A
3 2
M
Strain (%)
-0.4
4
-1
ΔR/R (%)
Epoxy composite BMI composite
SC R
Fig.4 SEM morphology of (a) CNT film and (b) graphite film.
200
BMI composite PAA composite Epoxy composite
50
0
IP T
100
0
1
2
3
Strain (%)
SC R
Stress (MPa)
150
4
5
M
2
ED
-0.2
4 3
1
PT
Strain (%)
-0.4
1h treated CNT composite film Lateral compressive strain
0
CC E
0.0
0
100 Time (s)
200
-1
A
Fig. 7 Resistance change of CNT nanocomposite film with different degrees of functionalization.
ΔR/R (%)
A
Pristine CNT composite film 2h treated CNT composite film
N
U
Fig. 6 The tensile stress-strain curves of CNT/PAA, CNT/EP and CNT/BMI composite films.
300
100
0
0
1
2 3 Strain (%)
SC R
1h treatment 2h treatment Pristine CNT composite film
IP T
Stress (MPa)
200
4
5
U
Fig.8 Stress-strain curve of CNT/EP nanocomposite film with and without functionalization.
N
4
Lateral compressive strain Acid treatment Pristine composite film
2 1
PT CC E
0
0
100 Time (s)
Fig. 9 Resistance change of CNT nanocomposite film after acid treatment.
200
-1
ΔR/R (%)
A
M -0.2
0.0
A
3
ED
Strain (%)
-0.4