- Email: [email protected]
Tensile strain sensing of buckypaper and buckypaper composites
Materials and Design 88 (2015) 414–419
Contents lists available at ScienceDirect
Materials and Design journal homepage: www.elsevier.com/locate/jmad
Tensile strain sensing of buckypaper and buckypaper composites Xiaoqiang Wang a, Shaowei Lu a,⁎, Keming Ma a, Xuhai Xiong a, Haijun Zhang a, Meijuan Xu b a b
Faculty of Aerospace Engineering, Shenyang Aerospace University, Shenyang 110136, China AVIC SAC Commercial Aircraft company Ltd., Shenyang 110136, China
a r t i c l e
i n f o
Article history: Received 23 January 2015 Received in revised form 7 September 2015 Accepted 8 September 2015 Available online 9 September 2015 Keywords: Composites Buckypaper Sensor Strain Piezoresistive
a b s t r a c t The effectiveness of buckypaper as a strain sensor is investigated. The key contribution of this paper is the study of piezoresistive response of both buckypaper and buckypaper composites. In addition, the manufacture of buckypaper and buckypaper composites is mentioned. The specimen is subjected to a tensile loading, and the resistance of the buckypaper is obtained using a four-point probe method and examined as a function of applied strain. Experimental results of buckypaper and buckypaper composites demonstrate that there are two different linear change sensing stages in the resistance of buckypaper with applied strain, and the linear relationship is recoverable and stable for the first sensing stage (0 – 30,000 με). From the results obtained, it is evident that buckypaper is very suitable for strain sensing. © 2015 Elsevier Ltd. All rights reserved.
1. Introduction Strain sensors are widely used in nuclear, space, aviation, shipbuilding and weaponry industries for monitoring the deformation and damage of structures. Conventional strain gauge is always arranged along a certain direction, and is used to measure the strain in designated directions. With the development of the engineering and science, a multidirectional sensor which can be used measure the strain is demanded. Since the discovery of carbon nanotubes (CNTs) by Iijima [1], they have attracted remarkable attention as raw materials for the development of nanomaterials owing to their remarkable thermal and electrical conductivities, superior mechanical properties and low density [2]. Fortunately, it has been confirmed that the conductance of CNTs can be dramatically changed by changing the chirality in a single-walled carbon nanotubes (SWNTs), and mechanical strain can efficient affect the chirality [3]. All these prove that CNT-based sensors can fulfill the above mentioned engineering requirement for their properties [4,5]. Hu et al. developed a CNT-based flexible sensor which can be used to detect normal and shear forces [6]. Yimazoglu et al. raised a simple technique for integrating flexible, vertically aligned, multi-walled CNTs (MWCNTs) arrays sandwiched between carbon layers and examined the electromechanical properties of the CNT arrays for sensing pressure, tactile and vibration [7]. Lai et al. proposed a novel sensing material which was prepared by dispersing CNTs and silver nanoparticles in a polydimethylsiloxane (PDMS) polymer by using the dielectrophoresis technique [8]. Lipomi
⁎ Corresponding author. E-mail address: [email protected] (S. Lu).
http://dx.doi.org/10.1016/j.matdes.2015.09.035 0264-1275/© 2015 Elsevier Ltd. All rights reserved.
et al. proposed a film pressure and strain sensor which was composed of SWCNTs [9]. Buckypaper (BP) is an outstanding material which contains entangled networks of CNTs formed by van der Waals interactions [10]. BP can be fabricated by using SWCNTs [11], double-walled CNTs [12], and MWCNTs [13]. A number of techniques for fabricating BP have been proposed. Hennrich et al. proposed a vacuum filtration method for fabricating large-area BP less than 200 nm thick [14]. Rigueur et al. fabricated BP by liberation of electrophoretically deposited carbon nanotubes [15]. Wang and Zhang et al. have fabricated highly oriented BP made of aligned carbon nanotubes [16–18]. Due to the component material (CNTs), microstructure and properties of BP, in the past decade, BP and BP composites have become a hot topic in the CNTs research. It is believed to be a good candidate for many engineering applications, such as electrodes, actuators, sensor, and heat conductors and as reinforcement for polymer composites [19–21]. Li et al. demonstrated the potential of carbon nanotube films in measuring strain at the macroscale [22]. Kang et al. developed a composite electrical resistance strain sensor based on SWNTs, and it was used to measure the strain of a structure at the macroscale [23]. Li et al. have studied the possibility of using multiwalled carbon nanotube (MWCNTs) films as strain sensors [24]. Gao et al. reported a simple approach to deposit multi-walled carbon nanotube (MWNTs) networks onto glass fiber surfaces achieving semiconductive MWNTs-glass fibers, along with application of fiber/polymer interphase s as in situ multifunctional sensors [25]. In this paper, we propose a simple vacuum filtration method for fabricating buckypaper and a vacuum bag molding technique for fabricating buckypaper composites. The buckypaper within the composites can be used as a sensing element for strain loading. Based on this sensing element, the tensile strain of buckypaper composites is investigated.
X. Wang et al. / Materials and Design 88 (2015) 414–419
415
2. Experiments 2.1. Preparation of buckypaper The buckypaper preparation strategy was based on CNT dispersion in a solvent followed by successive filtration steps of this dispersion to form a film. 700 mg MWCNTs (10–30 nm in diameter, up to 50 μmin length and purity = 98%, Chengdu Organic Chemistry, Chinese Academy of Sciences) were ultrasonically dispersed in 1000 ml deionized water (1% Triton X-100, Tianjin Shakespeare Company) using Misonix Sonicator Q700 (USA) at 100 W (20 KHz) for 1 h. Upon completion of the dispersion process, the solution was ejected onto a 0.45 μm porous diameter filtration membrane, and filtered by a vacuum filtration setup as shown in Fig. 1(a). After filtration, the buckypaper was thoroughly washed with lots of deionized water to remove the adsorbed surfactant and dried in air. After the drying process, the buckypaper was peeled off from the filter membrane carefully and resulting in ~ 20 μm thick film as shown in Fig. 1(b).
Fig. 2. Composites with MWCNTs buckypaper sensor.
3. Results and discussion 3.1. Microscopic morphology and pore size distribution of buckypaper
2.2. Preparation of buckypaper composites The buckypaper sensor was cut out from the buckypaper film obtained in Section 2.1 with scissors, and we ensure that the sensor has a rectangle shape with length 30 mm, width 10 mm. The sensor was located in the center of a 12 × 2.5 × 0.25 inch3 glass fiber reinforced epoxy resin composite's (0°unidirectional composite, Weihai Guangwei composites Co., Ltd.) tensile specimen using vacuum bag molding technique for 2 h under 120∘ Cand 0.5 MPa. The purpose of this process was to obtain a well resin infiltration into the buckypaper sensor and avoid a debonding between composites and buckypaper sensor. Subsequently, four copper wires (diameter: 0.1 mm) were fixed onto the buckypaper sensor surface using silver conductive adhesive and Fiber Bragg Grating strain sensor was used to measure the applied strain as shown in Fig. 2.
From Fig. 3(a), it can be seen that the SEM image of the buckypaper shows a densely packed mass of randomly oriented MWCNTs without any agglomeration, and this orientation gives rise to its isotropic
2.3. Characterization and sensing test of buckypaper Field Emission Scanning Electron Microscope (FE-SEM) was used to get the microstructure of buckypaper, and Barret–Joyner–Halenda (BJH) method [26] was used to evaluate the pore size distribution of buckypaper. With the aim of characterizing the electric resistance response of the buckypaper sensor, a four probe methodology was employed for collecting data continuously at a sampling rate of 0.5 Hz. For the tensile testing, a MTS landmark universal testing machine equipped with 100 kN load cell was employed. The loading protocol involved incremental loading steps at load control mode with constant loading rate of 0.5 mm/min.
Fig. 1. (a) Vacuum filtration setup, (b) fabricated buckypaper.
Fig. 3. (a) The FE-SEM image of a buckypaper, (b) the curve of average pore size distribution.
416
X. Wang et al. / Materials and Design 88 (2015) 414–419
properties. These properties will be benefit to the transfer of electrons in the buckypaper and it raises the strain sensing ability of buckypaper. Fig. 3(b) shows that the buckypaper is a typical mesoporous material with an average pore size 26.6 nm and most of the pore size are between 2 nm and 45 nm. 3.2. Piezoresistive tensile response of buckypaper The mechanical and piezoresistive tensile response of buckypaper is shown in Fig. 4, where the buckypaper is loaded in axial tension. In Fig. 4(a), the normalized changes in electrical resistance (ΔR/R0) are plotted against the applied strain from 0 to 43,000 με, where R0is the value of electrical resistance of buckypaper before loading and ΔR =R− R0 is the instantaneous change in R. From Fig. 4(b) to (d), the displayed curves are according to strain 0 μεto 12,000 με, 12,000 με to 30,000 με and 30,000 με to 43,000 με, respectively. The gauge factor measures the sensitivity of the strain sensor is expressed as the ratio of relative change in resistance to applied strain, for instance, K = ΔR/εR0. Based on Fig. 4, the gauge factors for different strain stages are displayed in Fig. 5. It shows K is increasing as the increase of strain which indicates the sensitivity of buckypaper depends on the applied strain to it, while the average value of K is 9 from 0 μεto 43,000 με. For strain stages 0–12,000 με and 12,000 με–30,000 με, the gauge factor is 8 and 9, respectively. They are close to the average value 9. We define these two strain stages as the first sensing stage. For strain stage 30,000 με–43,000 με, the gauge factor is 13 which is much bigger than the average value 9, and we define this stage as the second sensing stage. The difference between these two sensing stage is result of that with the increasing of applied strain, the buckypaper morphology has be changed, such as CNTs dispersion state and CNTs integrity which affect the linearity of the piezoresistive behavior [27,28]. Owing to the microscopic morphology of buckypaper, the buckypaper bulk resistance consists of CNTs contact resistance, tunneling resistance and CNTs intrinsic resistance. In the first sensing stage, the tensile loading is small, hence the gauge factor is mainly affected by CNTs intrinsic resistance.
Fig. 5. Gauge factor K for different strain stages.
Because the large stiffness of CNTs, the change of CNTs intrinsic resistance is small, and a smaller gauge factor is obtained. While in the second sensing stage, the gauge factor is mainly affected by CNTs contact resistance and tunneling resistance, and a relative bigger gauge factor is obtained [29]. 3.3. Tensile strain response of buckypaper composites The buckypaper composites is shown in Fig. 6. It can be found that the interface between the glass fiber reinforced composites and buckypaper is a well infiltration layer and the pores of buckypaper is filled with epoxy resin. All of these ensure load transfer continuity from fiber reinforced composites to buckypaper.
Fig. 4. Resistance change plotted against applied strain for buckypaper sensor: (a) the curve of response from 0 μεto 43,000 με, (b) the curve of response from 0 μεto 12,000 με, (c) the curve of response from 12,000 μεto 30,000 με, and (d) the curve of response from 30,000 μεto 43,000 με.
X. Wang et al. / Materials and Design 88 (2015) 414–419
417
Fig. 6. Well interface infiltration of buckypaper composite.
Based on the results obtained above, the applied tensile strain to the buckypaper composites is less than 12,000 με. This load level is contained in the first sensing stage. A four probe methodology was employed to collect the resistance value of the buckypaper sensor continuously at a sampling rate of 0.5 Hz. Fig. 7 gives the relationship curve between the strain of composite and the resistance change of the buckypaper sensor. It was found that the relationship can be well linear fitted and the curve indicates the gauge factorK = 9. Although K = 9is different from the value obtained in Fig. 4(b), it equals to the average value of K from 0 μεto 43,000 με. The difference is caused by the infiltration of the polymer of the composites into the buckypaper, but the infiltration does not affect the linear piezoresistivity of the buckypaper sensor. In order to evaluate the cyclicity of the sensing ability of the buckypaper sensor within the buckypaper composites, the resistance change-strain-time curves are given in Fig. 8. The applied maximum stress is 150 MPa. Fig. 8 shows the curves of the first six cycles. It can be found that the resistance of buckypaper sensor changes correspond better with the strain change in cycle 5 and 6 than in cycles 1 to 4. Fig. 9 shows the relationships between the change of resistance and the strain of sensor for cycle 1 to cycle 6 respectively. It is apparent that the synchronicity, reversibility and stability of the changes of
Fig. 7. Fitted line for resistance change versus strain of buckypaper composite.
both the resistance and strain are getting better from cycle 1 to cycle 6. It is because that the randomly distribution of the conductive carbon nanotubes in the buckypaper will be changed due to the external load situation. After four cycles, external load situation could not affect the spatial distribution of the carbon nanotubes, so the piezoresistive sensing characteristics will keep stable. Fig. 10 shows the gauge factors of both loading and unloading history for each cycle. It indicates that the buckypaper sensor has a stable gauge factorK = 9. 4. Conclusions A simple vacuum filtration method for fabricating buckypaper and a vacuum bag molding technique for fabricating buckypaper composites are proposed, respectively. A series of tensile experiments and loadingunloading experiments are performed to study the piezoresistive feature of buckypaper and the tensile strain response of buckypaper composites. The experimental results lead us to following conclusions: a) Based on the method proposed, an excellent buckypaper can be obtained, and this buckypaper has been verified to be a typical mesoporous material. In addition, it can be used to manufacture buckypaper composite materials easily.
Fig. 8. Curves of the change of resistance-strain-time for sample under maximum stress 150 MPa.
418
X. Wang et al. / Materials and Design 88 (2015) 414–419
Fig. 9. Relationships between the change of resistance and the strain for cycle 1 to cycle 6 respectively.
b) The obtained buckypaper has a superior piezoresistivity which can be affected by buckypaper morphology, and the morphology will change as a result of the applied strain. Though the gauge factor of buckypaper is different for different sensing stages, the relationship between resistance and strain is linear for each sensing stage, and it has an average gauge factor 9. The results demonstrate that buckypaper can be used as a strain sensor. c) Tensile experiment results shows that the relationship between the resistance change of buckypaper and the strain applied to the composites is linear. Tensile loading-unloading experiment results shows that an outstanding linear relationship between the resistance change of buckypaper sensor and the strain of the composites can be obtained at least four cycles later.
Fig. 10. Gauge factors of both loading and unloading history for each cycle.
Acknowledgments This work was financially supported by the Doctoral Start-up Foundation of Shenyang Aerospace University (13YB13), the Scientific Research Fund of Liaoning Provincial Education Department (L2013078, L2013074, and L2014072), the Defense Industrial Technology Development Program of China (A35201106) and the Aeronautical Science Foundation of China (2013ZA54012, 2013ZA54004). The financial contributions are gratefully acknowledged.
References [1] S. Iijima, Helical microtubules of graphitic carbon, Nature 354 (1991) 56–58. [2] N. Hu, Y. Karube, Y. Cheng, Z. Masuda, H. Fukunaga, Tunneling effect in a polymer/ carbon nanotube nanocomposite strain sensor, Acta Mater. 56 (2008) 2929–2936. [3] T.W. Tombler, X.C. Zhou, L. Alexseyev, J. Kong, H. Dai, L. Liu, Reversible electromechanical characteristics of carbon nanotubes under local-probe manipulation, Nature 405 (2000) 769–772. [4] R.H. Baughman, A.A. Zakhidov, W.A. de Heer, Carbon nanotubes—the route toward applications, Science 297 (2002) 787–792. [5] B. Mahar, C. Laslau, R. Yip, Y. Sun, Development of carbon nanotube-based sensors — a review, IEEE Sensors J. 7 (2007) 266–284. [6] C.-F. Hu, W.-S. Su, W.-L. Fang, Development of patterned carbon nanotubes on a 3D polymer substrate for the flexible tactile sensor application, J. Microelectromech. Syst. 21 (2011) 115012. [7] O. Yimazoglu, A. Popp, D. Pavildis, J. Schneider, D. Garth, F. Schuttler, G. Batten-berg, Vertically aligned multiwalled carbon nanotubes for pressure, tactile and vibration sensing, Nanotechnology 23 (2003) 085501. [8] Y.-T. Lai, Y.-M. Chen, Y.-J. Yang, A novel CNT–PDMS-based tactile sensing array with resistivity retaining and recovering by using dielectrophoresis effect, J. Microelectromech. Syst. 21 (2012) 217–223. [9] D.J. Lipomi, M. Vosgueritchian, B.C.K. Tee, S.L. Hellstrom, J.A. Lee, C.H. Fox, Z.N. Bao, Skin-like pressure and strain sensors based on transparent elastic films of carbon nanotubes, Nat. Nanotechnol. 6 (2011) 788–792. [10] R.H. Baughman, Carbon nanotube actuators, Science 284 (5418) (1999) 1340–1344. [11] L.C. Teague, S. Banerjee, S.S. Wong, C.A. Richter, B. Varughese, J.D. Batteas, Effects of ozonolysis and subsequent growth of quantum dots on the electrical properties of freestanding single-walled carbon nanotube films, Chem. Phys. Lett. 442 (2007) 354–359. [12] T. Gong, Y. Zhang, W.J. Liu, J.Q. Wei, C.G. Li, K.L. Wang, D.H. Wu, M.L. Zhong, Connection of macro-sized double-walled carbon nanotube strands by bandaging with double-walled carbon nanotube films, Carbon 45 (2007) 2235–2240.
X. Wang et al. / Materials and Design 88 (2015) 414–419 [13] G.H. Xu, Q. Zhang, W.P. Zhou, J.Q. Huang, F. Wei, The feasibility of producing MWCNT paper and strong MWCNT film from VACNT array, Appl. Phys. A Mater. Sci. Process. 92 (2008) 531–539. [14] F. Hennrich, S. Lebedkin, S. Malik, J. Tracy, M. Barczewski, H. Rosner, M. Kappes, Preparation, characterization and applications of free-standing single walled carbon nanotube thin films, Phys. Chem. Chem. Phys. 4 (2002) 2273–2277. [15] J. Rigueur, S. Hasan, S. Mahajan, J. Dickerson, Buckypaper fabrication by liberation of electrophoretically deposited carbon nanotubes, Carbon 48 (2010) 4090–4099. [16] D. Wang, P. Song, C. Liu, W. Wu, S. Fan, Highly oriented carbon nanotube papers made of aligned carbon nanotubes, Nanotechnology 19 (2008) 075609. [17] J.W. Zhang, D.Z. Jiang, H.X. Peng, Highly oriented carbon nanotube papers made of aligned carbon nanotubes, Microporous Mesoporous Mater. 184 (2014) 127–133. [18] Y. Inoue, Y. Suzuki, Y. Minami, J. Muramatsu, Y. Shimamura, K. Suzuki, A. Ghemes, M. Okada, S. Sakakibara, F. Mimura, Highly oriented carbon nanotube papers made of aligned carbon nanotubes, Carbon 49 (2011) 2437–2443. [19] J. Che, P. Chen, M.B. Chan-Park, High-strength carbon nanotube buckypaper composites as applied to free-standing electrodes for supercapacitors, J. Mater. Chem. A 1 (2013) 4057–4066. [20] Y.W. Chen, C.Y. Cheng, H.Y. Miao, M. Zhang, R. Liang, C. Zhang, Application of response surface methodology in the optimization of laser treatment in buckypaper lighting for field emission displays, Int. J. Adv. Manuf. Technol. 64 (2013) 515–536.
419
[21] J.H. Liu, H.Y. Miao, L. Saravanan, L.C. Wang, R.H. Tsai, Fabrication of metal alloydeposited flexible MWCNT buckypaper for thermoelectric applications, J. Nanomater. 6 (2013) 635647. [22] Z. Li, P. Dharap, S. Nagarajaiah, E.V. Barrera, J.D. Kim, Carbon nanotube film sensors, Adv. Mater. 16 (2004) 640–643. [23] I. Kang, M.J. Schulz, J.H. Kim, V. Shanov, D. Shi, A carbon nanotube strain sensor for structural health monitoring, Smart Mater. Struct. 15 (2006) 737–748. [24] X. Li, C. Levy, L. Elaadil, Multiwalled carbon nanotube film for strain sensing, Nanotechnology 19 (2008) 045501. [25] S.-l. Gao, R.-C. Zhuang, J. Zhang, J.-W. Liu, E. Mäder, Glass fibers with carbon nanotube networks as multifunctional sensors, Adv. Funct. Mater. 20 (2010) 1885–1893. [26] E.P. Barrett, L.G. Joyner, P.P. Halenda, The determination of pore volume and area distributions in porous substances. I. Computations from nitrogen isotherms, J. Am. Chem. Soc. 73 (1951) 373–380. [27] J. Alamusi, N. Hu, H. Fukunaga, S. Atobe, Y. Liu, J. Li, Piezoresistive strain sensors made from carbon nanotubes based polymer nanocomposites, Sensors 11 (2011) 10691–10723. [28] G. Yin, N. Hu, Y. Karube, Y. Liu, Y. Li, H. Fukunaga, A carbon nanotube/polymer strain sensor with linear and antisymmetric piezoresistivity, J. Compos. Mater. 45 (2011) 1315–1323. [29] M.K. Njuguna, C. Yan, N. Hu, Sandwiched carbon nanotube film as strain sensor, Compos. Part B 43 (2012) 2711–2717.
Contents lists available at ScienceDirect
Materials and Design journal homepage: www.elsevier.com/locate/jmad
Tensile strain sensing of buckypaper and buckypaper composites Xiaoqiang Wang a, Shaowei Lu a,⁎, Keming Ma a, Xuhai Xiong a, Haijun Zhang a, Meijuan Xu b a b
Faculty of Aerospace Engineering, Shenyang Aerospace University, Shenyang 110136, China AVIC SAC Commercial Aircraft company Ltd., Shenyang 110136, China
a r t i c l e
i n f o
Article history: Received 23 January 2015 Received in revised form 7 September 2015 Accepted 8 September 2015 Available online 9 September 2015 Keywords: Composites Buckypaper Sensor Strain Piezoresistive
a b s t r a c t The effectiveness of buckypaper as a strain sensor is investigated. The key contribution of this paper is the study of piezoresistive response of both buckypaper and buckypaper composites. In addition, the manufacture of buckypaper and buckypaper composites is mentioned. The specimen is subjected to a tensile loading, and the resistance of the buckypaper is obtained using a four-point probe method and examined as a function of applied strain. Experimental results of buckypaper and buckypaper composites demonstrate that there are two different linear change sensing stages in the resistance of buckypaper with applied strain, and the linear relationship is recoverable and stable for the first sensing stage (0 – 30,000 με). From the results obtained, it is evident that buckypaper is very suitable for strain sensing. © 2015 Elsevier Ltd. All rights reserved.
1. Introduction Strain sensors are widely used in nuclear, space, aviation, shipbuilding and weaponry industries for monitoring the deformation and damage of structures. Conventional strain gauge is always arranged along a certain direction, and is used to measure the strain in designated directions. With the development of the engineering and science, a multidirectional sensor which can be used measure the strain is demanded. Since the discovery of carbon nanotubes (CNTs) by Iijima [1], they have attracted remarkable attention as raw materials for the development of nanomaterials owing to their remarkable thermal and electrical conductivities, superior mechanical properties and low density [2]. Fortunately, it has been confirmed that the conductance of CNTs can be dramatically changed by changing the chirality in a single-walled carbon nanotubes (SWNTs), and mechanical strain can efficient affect the chirality [3]. All these prove that CNT-based sensors can fulfill the above mentioned engineering requirement for their properties [4,5]. Hu et al. developed a CNT-based flexible sensor which can be used to detect normal and shear forces [6]. Yimazoglu et al. raised a simple technique for integrating flexible, vertically aligned, multi-walled CNTs (MWCNTs) arrays sandwiched between carbon layers and examined the electromechanical properties of the CNT arrays for sensing pressure, tactile and vibration [7]. Lai et al. proposed a novel sensing material which was prepared by dispersing CNTs and silver nanoparticles in a polydimethylsiloxane (PDMS) polymer by using the dielectrophoresis technique [8]. Lipomi
⁎ Corresponding author. E-mail address: [email protected] (S. Lu).
http://dx.doi.org/10.1016/j.matdes.2015.09.035 0264-1275/© 2015 Elsevier Ltd. All rights reserved.
et al. proposed a film pressure and strain sensor which was composed of SWCNTs [9]. Buckypaper (BP) is an outstanding material which contains entangled networks of CNTs formed by van der Waals interactions [10]. BP can be fabricated by using SWCNTs [11], double-walled CNTs [12], and MWCNTs [13]. A number of techniques for fabricating BP have been proposed. Hennrich et al. proposed a vacuum filtration method for fabricating large-area BP less than 200 nm thick [14]. Rigueur et al. fabricated BP by liberation of electrophoretically deposited carbon nanotubes [15]. Wang and Zhang et al. have fabricated highly oriented BP made of aligned carbon nanotubes [16–18]. Due to the component material (CNTs), microstructure and properties of BP, in the past decade, BP and BP composites have become a hot topic in the CNTs research. It is believed to be a good candidate for many engineering applications, such as electrodes, actuators, sensor, and heat conductors and as reinforcement for polymer composites [19–21]. Li et al. demonstrated the potential of carbon nanotube films in measuring strain at the macroscale [22]. Kang et al. developed a composite electrical resistance strain sensor based on SWNTs, and it was used to measure the strain of a structure at the macroscale [23]. Li et al. have studied the possibility of using multiwalled carbon nanotube (MWCNTs) films as strain sensors [24]. Gao et al. reported a simple approach to deposit multi-walled carbon nanotube (MWNTs) networks onto glass fiber surfaces achieving semiconductive MWNTs-glass fibers, along with application of fiber/polymer interphase s as in situ multifunctional sensors [25]. In this paper, we propose a simple vacuum filtration method for fabricating buckypaper and a vacuum bag molding technique for fabricating buckypaper composites. The buckypaper within the composites can be used as a sensing element for strain loading. Based on this sensing element, the tensile strain of buckypaper composites is investigated.
X. Wang et al. / Materials and Design 88 (2015) 414–419
415
2. Experiments 2.1. Preparation of buckypaper The buckypaper preparation strategy was based on CNT dispersion in a solvent followed by successive filtration steps of this dispersion to form a film. 700 mg MWCNTs (10–30 nm in diameter, up to 50 μmin length and purity = 98%, Chengdu Organic Chemistry, Chinese Academy of Sciences) were ultrasonically dispersed in 1000 ml deionized water (1% Triton X-100, Tianjin Shakespeare Company) using Misonix Sonicator Q700 (USA) at 100 W (20 KHz) for 1 h. Upon completion of the dispersion process, the solution was ejected onto a 0.45 μm porous diameter filtration membrane, and filtered by a vacuum filtration setup as shown in Fig. 1(a). After filtration, the buckypaper was thoroughly washed with lots of deionized water to remove the adsorbed surfactant and dried in air. After the drying process, the buckypaper was peeled off from the filter membrane carefully and resulting in ~ 20 μm thick film as shown in Fig. 1(b).
Fig. 2. Composites with MWCNTs buckypaper sensor.
3. Results and discussion 3.1. Microscopic morphology and pore size distribution of buckypaper
2.2. Preparation of buckypaper composites The buckypaper sensor was cut out from the buckypaper film obtained in Section 2.1 with scissors, and we ensure that the sensor has a rectangle shape with length 30 mm, width 10 mm. The sensor was located in the center of a 12 × 2.5 × 0.25 inch3 glass fiber reinforced epoxy resin composite's (0°unidirectional composite, Weihai Guangwei composites Co., Ltd.) tensile specimen using vacuum bag molding technique for 2 h under 120∘ Cand 0.5 MPa. The purpose of this process was to obtain a well resin infiltration into the buckypaper sensor and avoid a debonding between composites and buckypaper sensor. Subsequently, four copper wires (diameter: 0.1 mm) were fixed onto the buckypaper sensor surface using silver conductive adhesive and Fiber Bragg Grating strain sensor was used to measure the applied strain as shown in Fig. 2.
From Fig. 3(a), it can be seen that the SEM image of the buckypaper shows a densely packed mass of randomly oriented MWCNTs without any agglomeration, and this orientation gives rise to its isotropic
2.3. Characterization and sensing test of buckypaper Field Emission Scanning Electron Microscope (FE-SEM) was used to get the microstructure of buckypaper, and Barret–Joyner–Halenda (BJH) method [26] was used to evaluate the pore size distribution of buckypaper. With the aim of characterizing the electric resistance response of the buckypaper sensor, a four probe methodology was employed for collecting data continuously at a sampling rate of 0.5 Hz. For the tensile testing, a MTS landmark universal testing machine equipped with 100 kN load cell was employed. The loading protocol involved incremental loading steps at load control mode with constant loading rate of 0.5 mm/min.
Fig. 1. (a) Vacuum filtration setup, (b) fabricated buckypaper.
Fig. 3. (a) The FE-SEM image of a buckypaper, (b) the curve of average pore size distribution.
416
X. Wang et al. / Materials and Design 88 (2015) 414–419
properties. These properties will be benefit to the transfer of electrons in the buckypaper and it raises the strain sensing ability of buckypaper. Fig. 3(b) shows that the buckypaper is a typical mesoporous material with an average pore size 26.6 nm and most of the pore size are between 2 nm and 45 nm. 3.2. Piezoresistive tensile response of buckypaper The mechanical and piezoresistive tensile response of buckypaper is shown in Fig. 4, where the buckypaper is loaded in axial tension. In Fig. 4(a), the normalized changes in electrical resistance (ΔR/R0) are plotted against the applied strain from 0 to 43,000 με, where R0is the value of electrical resistance of buckypaper before loading and ΔR =R− R0 is the instantaneous change in R. From Fig. 4(b) to (d), the displayed curves are according to strain 0 μεto 12,000 με, 12,000 με to 30,000 με and 30,000 με to 43,000 με, respectively. The gauge factor measures the sensitivity of the strain sensor is expressed as the ratio of relative change in resistance to applied strain, for instance, K = ΔR/εR0. Based on Fig. 4, the gauge factors for different strain stages are displayed in Fig. 5. It shows K is increasing as the increase of strain which indicates the sensitivity of buckypaper depends on the applied strain to it, while the average value of K is 9 from 0 μεto 43,000 με. For strain stages 0–12,000 με and 12,000 με–30,000 με, the gauge factor is 8 and 9, respectively. They are close to the average value 9. We define these two strain stages as the first sensing stage. For strain stage 30,000 με–43,000 με, the gauge factor is 13 which is much bigger than the average value 9, and we define this stage as the second sensing stage. The difference between these two sensing stage is result of that with the increasing of applied strain, the buckypaper morphology has be changed, such as CNTs dispersion state and CNTs integrity which affect the linearity of the piezoresistive behavior [27,28]. Owing to the microscopic morphology of buckypaper, the buckypaper bulk resistance consists of CNTs contact resistance, tunneling resistance and CNTs intrinsic resistance. In the first sensing stage, the tensile loading is small, hence the gauge factor is mainly affected by CNTs intrinsic resistance.
Fig. 5. Gauge factor K for different strain stages.
Because the large stiffness of CNTs, the change of CNTs intrinsic resistance is small, and a smaller gauge factor is obtained. While in the second sensing stage, the gauge factor is mainly affected by CNTs contact resistance and tunneling resistance, and a relative bigger gauge factor is obtained [29]. 3.3. Tensile strain response of buckypaper composites The buckypaper composites is shown in Fig. 6. It can be found that the interface between the glass fiber reinforced composites and buckypaper is a well infiltration layer and the pores of buckypaper is filled with epoxy resin. All of these ensure load transfer continuity from fiber reinforced composites to buckypaper.
Fig. 4. Resistance change plotted against applied strain for buckypaper sensor: (a) the curve of response from 0 μεto 43,000 με, (b) the curve of response from 0 μεto 12,000 με, (c) the curve of response from 12,000 μεto 30,000 με, and (d) the curve of response from 30,000 μεto 43,000 με.
X. Wang et al. / Materials and Design 88 (2015) 414–419
417
Fig. 6. Well interface infiltration of buckypaper composite.
Based on the results obtained above, the applied tensile strain to the buckypaper composites is less than 12,000 με. This load level is contained in the first sensing stage. A four probe methodology was employed to collect the resistance value of the buckypaper sensor continuously at a sampling rate of 0.5 Hz. Fig. 7 gives the relationship curve between the strain of composite and the resistance change of the buckypaper sensor. It was found that the relationship can be well linear fitted and the curve indicates the gauge factorK = 9. Although K = 9is different from the value obtained in Fig. 4(b), it equals to the average value of K from 0 μεto 43,000 με. The difference is caused by the infiltration of the polymer of the composites into the buckypaper, but the infiltration does not affect the linear piezoresistivity of the buckypaper sensor. In order to evaluate the cyclicity of the sensing ability of the buckypaper sensor within the buckypaper composites, the resistance change-strain-time curves are given in Fig. 8. The applied maximum stress is 150 MPa. Fig. 8 shows the curves of the first six cycles. It can be found that the resistance of buckypaper sensor changes correspond better with the strain change in cycle 5 and 6 than in cycles 1 to 4. Fig. 9 shows the relationships between the change of resistance and the strain of sensor for cycle 1 to cycle 6 respectively. It is apparent that the synchronicity, reversibility and stability of the changes of
Fig. 7. Fitted line for resistance change versus strain of buckypaper composite.
both the resistance and strain are getting better from cycle 1 to cycle 6. It is because that the randomly distribution of the conductive carbon nanotubes in the buckypaper will be changed due to the external load situation. After four cycles, external load situation could not affect the spatial distribution of the carbon nanotubes, so the piezoresistive sensing characteristics will keep stable. Fig. 10 shows the gauge factors of both loading and unloading history for each cycle. It indicates that the buckypaper sensor has a stable gauge factorK = 9. 4. Conclusions A simple vacuum filtration method for fabricating buckypaper and a vacuum bag molding technique for fabricating buckypaper composites are proposed, respectively. A series of tensile experiments and loadingunloading experiments are performed to study the piezoresistive feature of buckypaper and the tensile strain response of buckypaper composites. The experimental results lead us to following conclusions: a) Based on the method proposed, an excellent buckypaper can be obtained, and this buckypaper has been verified to be a typical mesoporous material. In addition, it can be used to manufacture buckypaper composite materials easily.
Fig. 8. Curves of the change of resistance-strain-time for sample under maximum stress 150 MPa.
418
X. Wang et al. / Materials and Design 88 (2015) 414–419
Fig. 9. Relationships between the change of resistance and the strain for cycle 1 to cycle 6 respectively.
b) The obtained buckypaper has a superior piezoresistivity which can be affected by buckypaper morphology, and the morphology will change as a result of the applied strain. Though the gauge factor of buckypaper is different for different sensing stages, the relationship between resistance and strain is linear for each sensing stage, and it has an average gauge factor 9. The results demonstrate that buckypaper can be used as a strain sensor. c) Tensile experiment results shows that the relationship between the resistance change of buckypaper and the strain applied to the composites is linear. Tensile loading-unloading experiment results shows that an outstanding linear relationship between the resistance change of buckypaper sensor and the strain of the composites can be obtained at least four cycles later.
Fig. 10. Gauge factors of both loading and unloading history for each cycle.
Acknowledgments This work was financially supported by the Doctoral Start-up Foundation of Shenyang Aerospace University (13YB13), the Scientific Research Fund of Liaoning Provincial Education Department (L2013078, L2013074, and L2014072), the Defense Industrial Technology Development Program of China (A35201106) and the Aeronautical Science Foundation of China (2013ZA54012, 2013ZA54004). The financial contributions are gratefully acknowledged.
References [1] S. Iijima, Helical microtubules of graphitic carbon, Nature 354 (1991) 56–58. [2] N. Hu, Y. Karube, Y. Cheng, Z. Masuda, H. Fukunaga, Tunneling effect in a polymer/ carbon nanotube nanocomposite strain sensor, Acta Mater. 56 (2008) 2929–2936. [3] T.W. Tombler, X.C. Zhou, L. Alexseyev, J. Kong, H. Dai, L. Liu, Reversible electromechanical characteristics of carbon nanotubes under local-probe manipulation, Nature 405 (2000) 769–772. [4] R.H. Baughman, A.A. Zakhidov, W.A. de Heer, Carbon nanotubes—the route toward applications, Science 297 (2002) 787–792. [5] B. Mahar, C. Laslau, R. Yip, Y. Sun, Development of carbon nanotube-based sensors — a review, IEEE Sensors J. 7 (2007) 266–284. [6] C.-F. Hu, W.-S. Su, W.-L. Fang, Development of patterned carbon nanotubes on a 3D polymer substrate for the flexible tactile sensor application, J. Microelectromech. Syst. 21 (2011) 115012. [7] O. Yimazoglu, A. Popp, D. Pavildis, J. Schneider, D. Garth, F. Schuttler, G. Batten-berg, Vertically aligned multiwalled carbon nanotubes for pressure, tactile and vibration sensing, Nanotechnology 23 (2003) 085501. [8] Y.-T. Lai, Y.-M. Chen, Y.-J. Yang, A novel CNT–PDMS-based tactile sensing array with resistivity retaining and recovering by using dielectrophoresis effect, J. Microelectromech. Syst. 21 (2012) 217–223. [9] D.J. Lipomi, M. Vosgueritchian, B.C.K. Tee, S.L. Hellstrom, J.A. Lee, C.H. Fox, Z.N. Bao, Skin-like pressure and strain sensors based on transparent elastic films of carbon nanotubes, Nat. Nanotechnol. 6 (2011) 788–792. [10] R.H. Baughman, Carbon nanotube actuators, Science 284 (5418) (1999) 1340–1344. [11] L.C. Teague, S. Banerjee, S.S. Wong, C.A. Richter, B. Varughese, J.D. Batteas, Effects of ozonolysis and subsequent growth of quantum dots on the electrical properties of freestanding single-walled carbon nanotube films, Chem. Phys. Lett. 442 (2007) 354–359. [12] T. Gong, Y. Zhang, W.J. Liu, J.Q. Wei, C.G. Li, K.L. Wang, D.H. Wu, M.L. Zhong, Connection of macro-sized double-walled carbon nanotube strands by bandaging with double-walled carbon nanotube films, Carbon 45 (2007) 2235–2240.
X. Wang et al. / Materials and Design 88 (2015) 414–419 [13] G.H. Xu, Q. Zhang, W.P. Zhou, J.Q. Huang, F. Wei, The feasibility of producing MWCNT paper and strong MWCNT film from VACNT array, Appl. Phys. A Mater. Sci. Process. 92 (2008) 531–539. [14] F. Hennrich, S. Lebedkin, S. Malik, J. Tracy, M. Barczewski, H. Rosner, M. Kappes, Preparation, characterization and applications of free-standing single walled carbon nanotube thin films, Phys. Chem. Chem. Phys. 4 (2002) 2273–2277. [15] J. Rigueur, S. Hasan, S. Mahajan, J. Dickerson, Buckypaper fabrication by liberation of electrophoretically deposited carbon nanotubes, Carbon 48 (2010) 4090–4099. [16] D. Wang, P. Song, C. Liu, W. Wu, S. Fan, Highly oriented carbon nanotube papers made of aligned carbon nanotubes, Nanotechnology 19 (2008) 075609. [17] J.W. Zhang, D.Z. Jiang, H.X. Peng, Highly oriented carbon nanotube papers made of aligned carbon nanotubes, Microporous Mesoporous Mater. 184 (2014) 127–133. [18] Y. Inoue, Y. Suzuki, Y. Minami, J. Muramatsu, Y. Shimamura, K. Suzuki, A. Ghemes, M. Okada, S. Sakakibara, F. Mimura, Highly oriented carbon nanotube papers made of aligned carbon nanotubes, Carbon 49 (2011) 2437–2443. [19] J. Che, P. Chen, M.B. Chan-Park, High-strength carbon nanotube buckypaper composites as applied to free-standing electrodes for supercapacitors, J. Mater. Chem. A 1 (2013) 4057–4066. [20] Y.W. Chen, C.Y. Cheng, H.Y. Miao, M. Zhang, R. Liang, C. Zhang, Application of response surface methodology in the optimization of laser treatment in buckypaper lighting for field emission displays, Int. J. Adv. Manuf. Technol. 64 (2013) 515–536.
419
[21] J.H. Liu, H.Y. Miao, L. Saravanan, L.C. Wang, R.H. Tsai, Fabrication of metal alloydeposited flexible MWCNT buckypaper for thermoelectric applications, J. Nanomater. 6 (2013) 635647. [22] Z. Li, P. Dharap, S. Nagarajaiah, E.V. Barrera, J.D. Kim, Carbon nanotube film sensors, Adv. Mater. 16 (2004) 640–643. [23] I. Kang, M.J. Schulz, J.H. Kim, V. Shanov, D. Shi, A carbon nanotube strain sensor for structural health monitoring, Smart Mater. Struct. 15 (2006) 737–748. [24] X. Li, C. Levy, L. Elaadil, Multiwalled carbon nanotube film for strain sensing, Nanotechnology 19 (2008) 045501. [25] S.-l. Gao, R.-C. Zhuang, J. Zhang, J.-W. Liu, E. Mäder, Glass fibers with carbon nanotube networks as multifunctional sensors, Adv. Funct. Mater. 20 (2010) 1885–1893. [26] E.P. Barrett, L.G. Joyner, P.P. Halenda, The determination of pore volume and area distributions in porous substances. I. Computations from nitrogen isotherms, J. Am. Chem. Soc. 73 (1951) 373–380. [27] J. Alamusi, N. Hu, H. Fukunaga, S. Atobe, Y. Liu, J. Li, Piezoresistive strain sensors made from carbon nanotubes based polymer nanocomposites, Sensors 11 (2011) 10691–10723. [28] G. Yin, N. Hu, Y. Karube, Y. Liu, Y. Li, H. Fukunaga, A carbon nanotube/polymer strain sensor with linear and antisymmetric piezoresistivity, J. Compos. Mater. 45 (2011) 1315–1323. [29] M.K. Njuguna, C. Yan, N. Hu, Sandwiched carbon nanotube film as strain sensor, Compos. Part B 43 (2012) 2711–2717.