Nanoscale COMMUNICATION
Cite this: Nanoscale, 2014, 6, 699
Published on 23 October 2013. Downloaded on 26/12/2013 15:34:14.
Received 24th August 2013 Accepted 20th October 2013
View Article Online View Journal | View Issue
Scalable fabrication of high-performance and flexible graphene strain sensors† He Tian,‡ab Yi Shu,‡ab Ya-Long Cui,‡ab Wen-Tian Mi,ab Yi Yang,ab Dan Xieab and Tian-Ling Ren*ab
DOI: 10.1039/c3nr04521h www.rsc.org/nanoscale
Graphene strain sensors have promising prospects of applications in detecting human motion. However, the shortage of graphene growth and patterning techniques has become a challenging issue hindering the application of graphene strain sensors. Therefore, we propose wafer-scale flexible strain sensors with high-performance, which can be fabricated in one-step laser scribing. The graphene films could be obtained by directly reducing graphene oxide film in a Light-Scribe DVD burner. The gauge factor (GF) of the graphene strain sensor (10 mm 10 mm square) is 0.11. In order to enhance the GF further, graphene micro-ribbons (20 mm width, 0.6 mm long) has been used as strain sensors, of which the GF is up to 9.49. The devices may conform to various application requirements, such as high GF for low-strain applications and low GF for high deformation applications. The work indicates that laser scribed flexible graphene strain sensors could be widely used in medical-sensing, bio-sensing, artificial skin and many other areas.
Introduction Graphene has attracted a huge amount of research due to its superior properties,1–3 such as ultra-high mobility,4 transparency,5 Young modulus6 etc. Graphene devices have been widely applied,7–13 especially in sensing elds.9,14,15 As the market of strain sensors in 2013 is expected to exceed 4.5 billion USD, graphene strain sensors have enormous potential application value in this eld. Attributed to its piezoresistive effect, graphene strain sensors with gauge factors (GF) of 0.55 and 6.1 have been demonstrated.16,17 Jing Zhao et al. showed nanographene strain sensors with GF over 300.18 Marek Hempel et al. a
Institute of Microelectronics, Tsinghua University, Beijing 100084, China. E-mail: [email protected]
b
Tsinghua National Laboratory for Information Science and Technology (TNList), Tsinghua University, Beijing 100084, China † Electronic supplementary information (ESI) available: Testing results and discussion of graphene strain sensors. See DOI: 10.1039/c3nr04521h ‡ These authors contributed equally to this work.
This journal is © The Royal Society of Chemistry 2014
used a sprayed solution to make high performance strain sensors.19 Further, Xiao Li et al. made graphene woven fabrics to enhance the stretch ability of the GF up to 103.20 Those approaches indicate better performance than conventional metal gauge sensors, of which the GF ranges at 2–5.21,22 However, the shortage of graphene growth and patterning techniques has become a challenging issue hindering the application of graphene strain sensors. Current fabrication processes for graphene strain sensors are mainly based on CVD graphene, in which hours of graphene growth, transfer and patterning are needed. This process has insuperable shortcomings, such as time consumption, photoresist contamination and graphene lms cracking. It is necessary to seek other methods to make large-scale high-performance graphene strain sensors. Moreover, the mechanism for graphene strain sensors is still not understood clearly. In this work, we fabricate scalable exible strain sensors with high-performance in one-step laser scribing. The graphene could be obtained by direct reduction of graphene oxide lm in a Light-Scribe DVD burner.23–25 The growth and patterning of graphene could be done in one step in 25 minutes. The resistance change of graphene has a linear relation with the applied strain and the device shows good multi-cycle operation ability. In order to enhance the GF further, graphene microribbons (20 mm width, 0.6 mm long) have been used as strain sensors, of which the gauge factor is up to 9.49. The cracking of graphene lms under the applied strain is observed in the experiments. A model of the graphene lm under strain is also established, and the theoretical results agree with the experimental ones well. The work indicates that laser scribed exible graphene strain sensors could be widely used in medicalsensing, bio-sensing, articial skin and many other areas. This work shows several achievements. First of all, a laser scribing technology is applied to fabricate large-scale graphene strain sensors. This technology has numerous advantages, such as forming graphene in a designed shape at precise locations, low cost with large scale fabrication ability and simple process. Secondly, robust and exible graphene strain sensors with low
Nanoscale, 2014, 6, 699–705 | 699
View Article Online
Nanoscale
Communication
GF (0.11) are realized for high deformation applications. In addition, graphene micro-ribbon strain sensors with high GF (9.49) for low-strain applications are achieved. Lastly, a model of the piezoresistance effect of graphene lms is established and the crack in the graphene lms could be the main mechanism to induce the resistance change.
Published on 23 October 2013. Downloaded on 26/12/2013 15:34:14.
Results Fig. 1 shows schematic diagrams of the fabrication process for exible graphene-on-PET strain sensors. The fabrication process could be described as follows. Firstly, a PET lm is coated on a DVD disc (Fig. 1a). Then, GO solution is drop-cast on the DVD disc (Fig. 1b). Aer the GO layer is dried, the disc is inserted into a Light-Scribe DVD drive and a computer-designed circuit is etched onto the lm. The laser inside the drive converts the golden-brown GO into black graphene at precise locations to produce graphene strain sensors. With the laser scribing technology, large areas of precise graphene patterns completed in 25 minutes are achieved (Fig. 1c). Finally, wafer-scale exible graphene-on-PET strain sensors could be obtained by peeling off the substrate from the disc (Fig. 1d). Fig. S1† shows scalable fabrication of laser scribed graphene on a wafer-scale. Fig. 2a is the SEM image of the surface prole of GO lm. Fig. 2b is the SEM image of the laser scribed graphene surface. Fig. 2c shows wafer-scale fabrication of laser scribed graphene-on-PET strain sensors. Different types of strain sensors could be fabricated in one-step laser scribing. Fig. 2d is a zoomed-in photograph of a graphene-on-PET strain sensor. Fig. 2e indicated the graphene-on-PET strain sensor with good exibility in hand. Fig. 2f presents the cross-sectional view of the graphene lm. A stack of graphene layers could be
Schematic diagrams of the fabrication process for flexible graphene-on-PET strain sensors. (a) A PET film is coated on a DVD disc. (b) GO solution is drop-cast on the DVD disc. (c) The disc is inserted into a Light-Scribe DVD drive and a computer-designed circuit is etched onto the film. The laser inside the drive converts the golden-brown GO into black graphene at precise locations to produce interdigitated graphene circuits. Large areas of precise graphene patterns could be obtained in 20–30 minutes by the laser scribing technology. (d) Peeling off the PET to obtain wafer-scale flexible graphene-on-PET strain sensors.
Fig. 1
700 | Nanoscale, 2014, 6, 699–705
The morphology and structure of the laser scribed grapheneon-PET strain sensors. (a) The surface profile of GO film under SEM. (b) The SEM image of the laser scribed graphene surface. (c) Wafer-scale fabrication of laser scribed graphene-on-PET strain sensors. Different types of strain sensors could be fabricated in one-step. (d) A zoomedin photograph of a graphene-on-PET strain sensor. (e) The grapheneon-PET strain sensor showing good flexibility. (f) Cross-sectional view of graphene film. A stack of graphene layers could be clearly identified. (g) An array of graphene-on-PET strain sensors. (h) A TEM image of single-layer laser scribed graphene deposited on carbon TEM grids. Electron diffraction on the graphene film is shown as an inset. Fig. 2
clearly identied. Fig. 2g shows an array of graphene-on-PET strain sensors. Fig. 2h is a TEM image of single-layer laser scribed graphene deposited on carbon TEM grids. Electron diffraction on the graphene lm is shown as an inset. The XPS, Raman and electrical results of GO and graphene are shown in Fig. S2–S4.† The schematic structures of the graphene-on-PET strain sensor in its original state and stretched state are shown in Fig. 3a and b. Fig. 3c shows the strain vs. relative resistance change, which has favorable linearity. The GF shown through the graph is 0.11, which is quite suitable for high deformation applications. Fig. 3d shows multi-cycle operation of the device. Fig. 3e presents the endurance testing up to 150 cycles under 7.5% strain, which demonstrated that the performance of the device is repeatable. The schematic structure of the graphene micro-ribbon strain sensor in its original state and stretched state is shown in Fig. 4a and b. Fig. 4c is the optical image of the graphene micro-
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 23 October 2013. Downloaded on 26/12/2013 15:34:14.
Communication
Nanoscale
Fig. 3 The schematic structures and electrical experimental results of the laser scribed graphene-on-PET strain sensor. (a) The schematic structure of the graphene-on-PET strain sensor in its original state. (b) The schematic structure of the graphene-on-PET strain sensor in a stretched state. (c) The strain vs. relative resistance change. The response line is quite linear. (d) Multi-cycle operation of the device under 2.3% strain. (e) Endurance testing up to 150 cycles under 7.5% strain. The inset shows the zoomed in performance over 10 cycles.
ribbon with 20 mm width. Fig. S5† shows the laser scribed graphene microribbon with a minimum patterning resolution of 20 mm. Fig. 4d presents two-probe measurement of the graphene micro-ribbon. Through the tting line, the sheet resistance of the graphene is 165 U mm1. There is a good linear relationship between the relative resistance change and strain, which is demonstrated by the response line shown in Fig. 4e. Fig. 4f shows the relative resistance change vs. time over four cycles of applied strain. Fig. S6† shows the graphene micro-ribbon strain sensors failed to reversibly operate aer 35 cycles.
Discussion Although graphene strain sensors have been investigated for several years, the mechanism of graphene strain sensors remains unclear. The main reason leading to the uncertainty of the mechanism is the difference of graphene fabrication process and lm quality. Three kinds of mechanism have been This journal is © The Royal Society of Chemistry 2014
proposed before.26 One assumes that the strain could induce the graphene band gap open, which can result in a resistance change.27 Another one is based on the model of over connected graphene sheets.19 The conductivity between neighboring akes is determined by their overlap area and the contact resistance. The last one is based on the tunneling effect between neighboring graphene sheets.18 It is predicted that the resistance changes exponentially with the distance. The tunneling effect would cause an exponential increase of graphene strain sensor's resistance, which is in contradiction with the experimental results. In the structure of stacked graphene, different graphene layers are mainly bonded by van der Waals forces, but graphene is a single planar sheet of sp2bonded carbon atoms, which interaction is far more robust compared with van der Waals forces. Sliding of different graphene sheets would take place when the graphene stain sensor was stretched. Hence, we propose to use change of graphene sheets’ overlap to explain our strain devices. The schematic layer structure of the graphene micro-ribbon strain sensor in its
Nanoscale, 2014, 6, 699–705 | 701
View Article Online
Published on 23 October 2013. Downloaded on 26/12/2013 15:34:14.
Nanoscale
Communication
Fig. 4 The schematic structures and electrical experimental results of the graphene micro-ribbon strain sensor. (a) The schematic structure of the graphene micro-ribbon strain sensor in its original state. (b) The schematic structure of the graphene micro-ribbon strain sensor in a stretched state. (c) The optical image showing the graphene micro-ribbon with 20 mm width. (d) Two-probe measurement of the graphene micro-ribbon. Through the fitting line, the sheet resistance of the graphene is 165 U mm1. (e) The relative resistance change vs. strain. The response line is quite linear. (f) Relative resistance change vs. time over four cycles of applied strain.
original state and stretched state are presented in Fig. 5a and b respectively. Film cracking started from the graphene upper layers, which will make the resistance become larger. In order to demonstrate this mechanism, the optical images combined with SEM images of the graphene micro-ribbon are given. The optical images of the 0.6 mm-long graphene micro-ribbon in its original state and the 0.72 mm-long graphene micro-ribbon in a strained state, which stretched 20% compared to the original state, are shown in Fig. 5c and d. The lm is uniform with minor pores, which is shown in the SEM image of the graphene micro-ribbon in its original state (Fig. 5e). In contrast, obvious cracks could be identied, which is shown in the SEM image of the graphene micro-ribbon in the stretched state (Fig. 5f). As mentioned, an important factor to evaluate performance of the strain sensor is the gauge factor (GF), which represents the relative change in resistance with unit strain. GF ¼ (DR/R)/3
(1)
In which, the DR/R is the relative change in resistance and 3 is the mechanical strain.
702 | Nanoscale, 2014, 6, 699–705
The GF of the laser scribed graphene is 0.11 on the macroscale (Fig. 3c). As the GF increases along with the resistance,19 the decrease of the width of graphene could lead to higher resistance, and then result in higher GF. The graphene microribbon (20 mm width, 5 mm long) is used as strain sensors, of which the gauge factor is up to 9.49 (Fig. 4d). The devices may conform to various application requirements, such as high GF for low-strain applications and low GF for high deformation applications. To analyze the mechanism concretely, we employ the nite element method and set a model of a network of randomly positioned graphene akes. The resistance is calculated by Ohm's law, R ¼ V/I under the xed current. In the graphene akes network, the voltage drop is not uniform at current direction, but depends on the electrical conductivity distribution. Ohm's Law could be rewritten as below in twodimensions. V(sV4) ¼ j
(2)
In the equation above, s is the electrical conductivity. j is the current density ux into the plane, which is 0 in the area of
This journal is © The Royal Society of Chemistry 2014
View Article Online
Communication
Nanoscale
diagram. The extracted gauge factors of the devices with varying thickness are proportional to their initial resistance as shown in the Fig. 6c. Relevance between gauge factor and device resistance is also considered, which is shown in Fig. 6d. It is indicated that the gauge factor increases with resistance, which is due to the decrease of the overlap area. The sub-circuit disconnection effect would be more remarkable since high resistance means lower graphene akes density. Fig. 6e indicates the resistance increasing quickly aer the limiting strain. As shown in Fig. 6f, the limiting strain is increased with the graphene thickness, which indicates that our proposed working mechanism is reasonable.
Published on 23 October 2013. Downloaded on 26/12/2013 15:34:14.
Conclusion
Fig. 5 The schematic mechanism, optical and SEM results of the graphene micro-ribbon strain sensor. (a) The schematic layered structure of the graphene micro-ribbon strain sensor in its original state. (b) The schematic layered structure of the graphene microribbon strain sensor in the stretched state. Cracks started from the upper layers. (c) The optical image showing the graphene microribbon (0.6 mm long) in its original state. (d) The optical image showing the graphene micro-ribbon (0.72 mm long) in the strained state. (e) The SEM image of the graphene micro-ribbon in its original state. (f) The SEM image of the graphene micro-ribbon in the stretched state. Obvious cracks could be identified.
strain sensor. 4 is the electronic potential. The potential distribution is computed by the nite element method under electrical conductivity distribution and xed current. The strain effect is analyzed by modeling the percolation of current through a network of randomly positioned circular akes with a uniform size distribution. As shown in Fig. 6a, the electrical conductivity distribution depends on the distribution of graphene akes and it is higher in the overlap area of multi-layer graphene akes. The decrease of the overlap of graphene akes changes the electrical conductivity of the overlap area, which means the whole resistance increases since the overlap area decreases when the device is stretched. A few sub-circuits are even disconnected as the graphene akes move to non-overlapping positions. All these effects increase the resistance of the device as the experimental result. As shown in Fig. 6b, the voltage drop is increased along with the increase of the strain under the xed current. It is consistent with the experiment results. The effect of the graphene lm on the sensitivity of the modeled strain sensor is demonstrated in Fig. 6c. It is shown that devices with a lower thickness exhibit a larger slope in the resistance–strain
This journal is © The Royal Society of Chemistry 2014
In summary, wafer-scale exible strain sensors with highperformance have been demonstrated. A Light-Scribe DVD burner could be used to obtain high quality graphene stack-layers. The graphene strain sensor has a good linear response to strain and good multi-cycle operation. Graphene micro-ribbons (20 mm width, 0.6 mm long) have been used as strain sensors to enhance the GF, of which the gauge factor is up to 9.49. Through the experimental and theoretical results, the lm cracking could be the main mechanism to explain the piezoresistance effect in our strain sensors. Our work shows that exible graphene strain sensors could be fabricated by laser scribing, which could open wide applications in medical-sensing, bio-sensing, articial skin and many other areas.
Method Material preparation GO dispersion with 2 mg mL1 concentration was provided by XFNANO Materials Tech CO., Ltd (Nanjing, China) which were synthesized from the graphite powders using a common Hummers method. 10 mL GO solution was dropped on the surface of the laser-scribed DVD disk coated with PET. Then the GO solutions were le overnight to dry on the DVD disk. The GO coated DVD disk could be patterned by the LightScribe DVD Drive (HP inc. 557S). Using the Nero StartSmart soware, the designed structure or photograph could be directly converted onto the GO lm by laser reduction of the GO into graphene. Characterization The surface morphology is observed by Quanta FEG 450 SEM (FEI Inc.) and JEM-2010 (JEOL Inc.). The Raman spectroscopy is obtained using a laser with wavelength of 514.5 nm (HORIBA Inc.). Testing for strain sensors PET substrate with the graphene device is mounted on a micrometer caliper, which can push the PET substrate to cause strain. Electrical properties are collected by a digital multimeter DM3068 (RIGOL).
Nanoscale, 2014, 6, 699–705 | 703
View Article Online
Published on 23 October 2013. Downloaded on 26/12/2013 15:34:14.
Nanoscale
Communication
Modeling of the graphene film under strain. (a) Diagram of electrical conductivity distribution of the graphene strain sensor. Different colors denote the value of electrical conductivity. (b) Representation of voltage distribution at a fixed current in a graphene film at different levels of strain. (c) Resistance vs. strain for different thicknesses of graphene film. (d) Gauge factor as a function of unstrained resistance. (e) Resistance vs. strain under large strain showing the resistance increasing fast after the limiting strain. (f) The limiting strain vs. graphene thickness.
Fig. 6
Acknowledgements This work was supported by the National Natural Science Foundation of China (61025021, 60936002, 51072089, and 61020106006), the National Key Project of Science and Technology (2011ZX02403-002) and the Special Fund for Agroscientic Research in the Public Interest (201303107). He Tian is additionally supported by the Ministry of Education Scholarship of China.
References 1 K. Novoselov, A. K. Geim, S. Morozov, D. Jiang, Y. Zhang, S. Dubonos, I. Grigorieva and A. Firsov, Science, 2004, 306, 666–669. 2 K. Novoselov, A. K. Geim, S. Morozov, D. Jiang, M. K. I. Grigorieva, S. Dubonos and A. Firsov, Nature, 2005, 438, 197–200. 3 A. K. Geim and K. S. Novoselov, Nat. Mater., 2007, 6, 183–191. 4 K. I. Bolotin, K. Sikes, Z. Jiang, M. Klima, G. Fudenberg, J. Hone, P. Kim and H. Stormer, Solid State Commun., 2008, 146, 351–355.
704 | Nanoscale, 2014, 6, 699–705
5 S. Bae, H. Kim, Y. Lee, X. Xu, J.-S. Park, Y. Zheng, J. Balakrishnan, T. Lei, H. R. Kim and Y. I. Song, Nat. Nanotechnol., 2010, 5, 574–578. 6 C. Lee, X. Wei, J. W. Kysar and J. Hone, Science, 2008, 321, 385–388. 7 F. Xia, T. Mueller, Y.-m. Lin, A. Valdes-Garcia and P. Avouris, Nat. Nanotechnol., 2009, 4, 839–843. 8 H. Tian, T.-L. Ren, D. Xie, Y.-F. Wang, C.-J. Zhou, T.-T. Feng, D. Fu, Y. Yang, P.-G. Peng and L.-G. Wang, ACS Nano, 2011, 5, 4878–4885. 9 H. Tian, Y. Yang, D. Xie, T.-L. Ren, Y. Shu, C.-J. Zhou, H. Sun, X. Liu and C.-H. Zhang, Nanoscale, 2013, 5, 890–894. 10 F. Schwierz, Nat. Nanotechnol., 2010, 5, 487–496. 11 H.-Y. Chen, H. Tian, B. Gao, S. Yu, J. Liang, J. Kang, Y. Zhang, T.-L. Ren and H.-S. P. Wong, Electrode/oxide interface engineering by inserting single-layer graphene: Application for HfO x-based resistive random access memory, 2012. 12 H. Tian, D. Xie, Y. Yang, T.-L. Ren, Y.-F. Wang, C.-J. Zhou, P.-G. Peng, L.-G. Wang and L.-T. Liu, Nanoscale, 2012, 4, 3345–3349. 13 H. Tian, H.-Y. Chen, B. Gao, S. Yu, J. Liang, Y. Yang, D. Xie, J. Kang, T.-L. Ren and Y. Zhang, Nano Lett., 2013, 13, 651–657.
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 23 October 2013. Downloaded on 26/12/2013 15:34:14.
Communication
14 Z. Cheng, Q. Li, Z. Li, Q. Zhou and Y. Fang, Nano Lett., 2010, 10, 1864–1868. 15 Y. Lu, B. R. Goldsmith, N. J. Kybert and A. C. Johnson, Appl. Phys. Lett., 2010, 97, 083107. 16 Y. Wang, R. Yang, Z. Shi, L. Zhang, D. Shi, E. Wang and G. Zhang, ACS Nano, 2011, 5, 3645–3650. 17 Y. Lee, S. Bae, H. Jang, S. Jang, S.-E. Zhu, S. H. Sim, Y. I. Song, B. H. Hong and J.-H. Ahn, Nano Lett., 2010, 10, 490–493. 18 J. Zhao, C. He, R. Yang, Z. Shi, M. Cheng, W. Yang, G. Xie, D. Wang, D. Shi and G. Zhang, Appl. Phys. Lett., 2012, 101, 063112. 19 M. Hempel, D. Nezich, J. Kong and M. Hofmann, Nano Lett., 2012, 12, 5714–5718. 20 X. Li, R. Zhang, W. Yu, K. Wang, J. Wei, D. Wu, A. Cao, Z. Li, Y. Cheng and Q. Zheng, Sci. Rep., 2012, 2, 870.
This journal is © The Royal Society of Chemistry 2014
Nanoscale
21 S. Beeby, MEMS mechanical sensors, Artech House on Demand, 2004. 22 M.-H. Bao, Micro mechanical transducers: pressure sensors, accelerometers and gyroscopes, Elsevier Science, 2000. 23 M. F. El-Kady and R. B. Kaner, Nat. Commun., 2013, 4, 1475. 24 V. Strong, S. Dubin, M. F. El-Kady, A. Lech, Y. Wang, B. H. Weiller and R. B. Kaner, ACS Nano, 2012, 6, 1395– 1403. 25 M. F. El-Kady, V. Strong, S. Dubin and R. B. Kaner, Science, 2012, 335, 1326–1330. 26 J. Zhao, G. Y. Zhang and D. X. Shi, Chin. Phys. B, 2013, 22, 057701. 27 K. S. Kim, Y. Zhao, H. Jang, S. Y. Lee, J. M. Kim, K. S. Kim, J.-H. Ahn, P. Kim, J.-Y. Choi and B. H. Hong, Nature, 2009, 457, 706–710.
Nanoscale, 2014, 6, 699–705 | 705
Cite this: Nanoscale, 2014, 6, 699
Published on 23 October 2013. Downloaded on 26/12/2013 15:34:14.
Received 24th August 2013 Accepted 20th October 2013
View Article Online View Journal | View Issue
Scalable fabrication of high-performance and flexible graphene strain sensors† He Tian,‡ab Yi Shu,‡ab Ya-Long Cui,‡ab Wen-Tian Mi,ab Yi Yang,ab Dan Xieab and Tian-Ling Ren*ab
DOI: 10.1039/c3nr04521h www.rsc.org/nanoscale
Graphene strain sensors have promising prospects of applications in detecting human motion. However, the shortage of graphene growth and patterning techniques has become a challenging issue hindering the application of graphene strain sensors. Therefore, we propose wafer-scale flexible strain sensors with high-performance, which can be fabricated in one-step laser scribing. The graphene films could be obtained by directly reducing graphene oxide film in a Light-Scribe DVD burner. The gauge factor (GF) of the graphene strain sensor (10 mm 10 mm square) is 0.11. In order to enhance the GF further, graphene micro-ribbons (20 mm width, 0.6 mm long) has been used as strain sensors, of which the GF is up to 9.49. The devices may conform to various application requirements, such as high GF for low-strain applications and low GF for high deformation applications. The work indicates that laser scribed flexible graphene strain sensors could be widely used in medical-sensing, bio-sensing, artificial skin and many other areas.
Introduction Graphene has attracted a huge amount of research due to its superior properties,1–3 such as ultra-high mobility,4 transparency,5 Young modulus6 etc. Graphene devices have been widely applied,7–13 especially in sensing elds.9,14,15 As the market of strain sensors in 2013 is expected to exceed 4.5 billion USD, graphene strain sensors have enormous potential application value in this eld. Attributed to its piezoresistive effect, graphene strain sensors with gauge factors (GF) of 0.55 and 6.1 have been demonstrated.16,17 Jing Zhao et al. showed nanographene strain sensors with GF over 300.18 Marek Hempel et al. a
Institute of Microelectronics, Tsinghua University, Beijing 100084, China. E-mail: [email protected]
b
Tsinghua National Laboratory for Information Science and Technology (TNList), Tsinghua University, Beijing 100084, China † Electronic supplementary information (ESI) available: Testing results and discussion of graphene strain sensors. See DOI: 10.1039/c3nr04521h ‡ These authors contributed equally to this work.
This journal is © The Royal Society of Chemistry 2014
used a sprayed solution to make high performance strain sensors.19 Further, Xiao Li et al. made graphene woven fabrics to enhance the stretch ability of the GF up to 103.20 Those approaches indicate better performance than conventional metal gauge sensors, of which the GF ranges at 2–5.21,22 However, the shortage of graphene growth and patterning techniques has become a challenging issue hindering the application of graphene strain sensors. Current fabrication processes for graphene strain sensors are mainly based on CVD graphene, in which hours of graphene growth, transfer and patterning are needed. This process has insuperable shortcomings, such as time consumption, photoresist contamination and graphene lms cracking. It is necessary to seek other methods to make large-scale high-performance graphene strain sensors. Moreover, the mechanism for graphene strain sensors is still not understood clearly. In this work, we fabricate scalable exible strain sensors with high-performance in one-step laser scribing. The graphene could be obtained by direct reduction of graphene oxide lm in a Light-Scribe DVD burner.23–25 The growth and patterning of graphene could be done in one step in 25 minutes. The resistance change of graphene has a linear relation with the applied strain and the device shows good multi-cycle operation ability. In order to enhance the GF further, graphene microribbons (20 mm width, 0.6 mm long) have been used as strain sensors, of which the gauge factor is up to 9.49. The cracking of graphene lms under the applied strain is observed in the experiments. A model of the graphene lm under strain is also established, and the theoretical results agree with the experimental ones well. The work indicates that laser scribed exible graphene strain sensors could be widely used in medicalsensing, bio-sensing, articial skin and many other areas. This work shows several achievements. First of all, a laser scribing technology is applied to fabricate large-scale graphene strain sensors. This technology has numerous advantages, such as forming graphene in a designed shape at precise locations, low cost with large scale fabrication ability and simple process. Secondly, robust and exible graphene strain sensors with low
Nanoscale, 2014, 6, 699–705 | 699
View Article Online
Nanoscale
Communication
GF (0.11) are realized for high deformation applications. In addition, graphene micro-ribbon strain sensors with high GF (9.49) for low-strain applications are achieved. Lastly, a model of the piezoresistance effect of graphene lms is established and the crack in the graphene lms could be the main mechanism to induce the resistance change.
Published on 23 October 2013. Downloaded on 26/12/2013 15:34:14.
Results Fig. 1 shows schematic diagrams of the fabrication process for exible graphene-on-PET strain sensors. The fabrication process could be described as follows. Firstly, a PET lm is coated on a DVD disc (Fig. 1a). Then, GO solution is drop-cast on the DVD disc (Fig. 1b). Aer the GO layer is dried, the disc is inserted into a Light-Scribe DVD drive and a computer-designed circuit is etched onto the lm. The laser inside the drive converts the golden-brown GO into black graphene at precise locations to produce graphene strain sensors. With the laser scribing technology, large areas of precise graphene patterns completed in 25 minutes are achieved (Fig. 1c). Finally, wafer-scale exible graphene-on-PET strain sensors could be obtained by peeling off the substrate from the disc (Fig. 1d). Fig. S1† shows scalable fabrication of laser scribed graphene on a wafer-scale. Fig. 2a is the SEM image of the surface prole of GO lm. Fig. 2b is the SEM image of the laser scribed graphene surface. Fig. 2c shows wafer-scale fabrication of laser scribed graphene-on-PET strain sensors. Different types of strain sensors could be fabricated in one-step laser scribing. Fig. 2d is a zoomed-in photograph of a graphene-on-PET strain sensor. Fig. 2e indicated the graphene-on-PET strain sensor with good exibility in hand. Fig. 2f presents the cross-sectional view of the graphene lm. A stack of graphene layers could be
Schematic diagrams of the fabrication process for flexible graphene-on-PET strain sensors. (a) A PET film is coated on a DVD disc. (b) GO solution is drop-cast on the DVD disc. (c) The disc is inserted into a Light-Scribe DVD drive and a computer-designed circuit is etched onto the film. The laser inside the drive converts the golden-brown GO into black graphene at precise locations to produce interdigitated graphene circuits. Large areas of precise graphene patterns could be obtained in 20–30 minutes by the laser scribing technology. (d) Peeling off the PET to obtain wafer-scale flexible graphene-on-PET strain sensors.
Fig. 1
700 | Nanoscale, 2014, 6, 699–705
The morphology and structure of the laser scribed grapheneon-PET strain sensors. (a) The surface profile of GO film under SEM. (b) The SEM image of the laser scribed graphene surface. (c) Wafer-scale fabrication of laser scribed graphene-on-PET strain sensors. Different types of strain sensors could be fabricated in one-step. (d) A zoomedin photograph of a graphene-on-PET strain sensor. (e) The grapheneon-PET strain sensor showing good flexibility. (f) Cross-sectional view of graphene film. A stack of graphene layers could be clearly identified. (g) An array of graphene-on-PET strain sensors. (h) A TEM image of single-layer laser scribed graphene deposited on carbon TEM grids. Electron diffraction on the graphene film is shown as an inset. Fig. 2
clearly identied. Fig. 2g shows an array of graphene-on-PET strain sensors. Fig. 2h is a TEM image of single-layer laser scribed graphene deposited on carbon TEM grids. Electron diffraction on the graphene lm is shown as an inset. The XPS, Raman and electrical results of GO and graphene are shown in Fig. S2–S4.† The schematic structures of the graphene-on-PET strain sensor in its original state and stretched state are shown in Fig. 3a and b. Fig. 3c shows the strain vs. relative resistance change, which has favorable linearity. The GF shown through the graph is 0.11, which is quite suitable for high deformation applications. Fig. 3d shows multi-cycle operation of the device. Fig. 3e presents the endurance testing up to 150 cycles under 7.5% strain, which demonstrated that the performance of the device is repeatable. The schematic structure of the graphene micro-ribbon strain sensor in its original state and stretched state is shown in Fig. 4a and b. Fig. 4c is the optical image of the graphene micro-
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 23 October 2013. Downloaded on 26/12/2013 15:34:14.
Communication
Nanoscale
Fig. 3 The schematic structures and electrical experimental results of the laser scribed graphene-on-PET strain sensor. (a) The schematic structure of the graphene-on-PET strain sensor in its original state. (b) The schematic structure of the graphene-on-PET strain sensor in a stretched state. (c) The strain vs. relative resistance change. The response line is quite linear. (d) Multi-cycle operation of the device under 2.3% strain. (e) Endurance testing up to 150 cycles under 7.5% strain. The inset shows the zoomed in performance over 10 cycles.
ribbon with 20 mm width. Fig. S5† shows the laser scribed graphene microribbon with a minimum patterning resolution of 20 mm. Fig. 4d presents two-probe measurement of the graphene micro-ribbon. Through the tting line, the sheet resistance of the graphene is 165 U mm1. There is a good linear relationship between the relative resistance change and strain, which is demonstrated by the response line shown in Fig. 4e. Fig. 4f shows the relative resistance change vs. time over four cycles of applied strain. Fig. S6† shows the graphene micro-ribbon strain sensors failed to reversibly operate aer 35 cycles.
Discussion Although graphene strain sensors have been investigated for several years, the mechanism of graphene strain sensors remains unclear. The main reason leading to the uncertainty of the mechanism is the difference of graphene fabrication process and lm quality. Three kinds of mechanism have been This journal is © The Royal Society of Chemistry 2014
proposed before.26 One assumes that the strain could induce the graphene band gap open, which can result in a resistance change.27 Another one is based on the model of over connected graphene sheets.19 The conductivity between neighboring akes is determined by their overlap area and the contact resistance. The last one is based on the tunneling effect between neighboring graphene sheets.18 It is predicted that the resistance changes exponentially with the distance. The tunneling effect would cause an exponential increase of graphene strain sensor's resistance, which is in contradiction with the experimental results. In the structure of stacked graphene, different graphene layers are mainly bonded by van der Waals forces, but graphene is a single planar sheet of sp2bonded carbon atoms, which interaction is far more robust compared with van der Waals forces. Sliding of different graphene sheets would take place when the graphene stain sensor was stretched. Hence, we propose to use change of graphene sheets’ overlap to explain our strain devices. The schematic layer structure of the graphene micro-ribbon strain sensor in its
Nanoscale, 2014, 6, 699–705 | 701
View Article Online
Published on 23 October 2013. Downloaded on 26/12/2013 15:34:14.
Nanoscale
Communication
Fig. 4 The schematic structures and electrical experimental results of the graphene micro-ribbon strain sensor. (a) The schematic structure of the graphene micro-ribbon strain sensor in its original state. (b) The schematic structure of the graphene micro-ribbon strain sensor in a stretched state. (c) The optical image showing the graphene micro-ribbon with 20 mm width. (d) Two-probe measurement of the graphene micro-ribbon. Through the fitting line, the sheet resistance of the graphene is 165 U mm1. (e) The relative resistance change vs. strain. The response line is quite linear. (f) Relative resistance change vs. time over four cycles of applied strain.
original state and stretched state are presented in Fig. 5a and b respectively. Film cracking started from the graphene upper layers, which will make the resistance become larger. In order to demonstrate this mechanism, the optical images combined with SEM images of the graphene micro-ribbon are given. The optical images of the 0.6 mm-long graphene micro-ribbon in its original state and the 0.72 mm-long graphene micro-ribbon in a strained state, which stretched 20% compared to the original state, are shown in Fig. 5c and d. The lm is uniform with minor pores, which is shown in the SEM image of the graphene micro-ribbon in its original state (Fig. 5e). In contrast, obvious cracks could be identied, which is shown in the SEM image of the graphene micro-ribbon in the stretched state (Fig. 5f). As mentioned, an important factor to evaluate performance of the strain sensor is the gauge factor (GF), which represents the relative change in resistance with unit strain. GF ¼ (DR/R)/3
(1)
In which, the DR/R is the relative change in resistance and 3 is the mechanical strain.
702 | Nanoscale, 2014, 6, 699–705
The GF of the laser scribed graphene is 0.11 on the macroscale (Fig. 3c). As the GF increases along with the resistance,19 the decrease of the width of graphene could lead to higher resistance, and then result in higher GF. The graphene microribbon (20 mm width, 5 mm long) is used as strain sensors, of which the gauge factor is up to 9.49 (Fig. 4d). The devices may conform to various application requirements, such as high GF for low-strain applications and low GF for high deformation applications. To analyze the mechanism concretely, we employ the nite element method and set a model of a network of randomly positioned graphene akes. The resistance is calculated by Ohm's law, R ¼ V/I under the xed current. In the graphene akes network, the voltage drop is not uniform at current direction, but depends on the electrical conductivity distribution. Ohm's Law could be rewritten as below in twodimensions. V(sV4) ¼ j
(2)
In the equation above, s is the electrical conductivity. j is the current density ux into the plane, which is 0 in the area of
This journal is © The Royal Society of Chemistry 2014
View Article Online
Communication
Nanoscale
diagram. The extracted gauge factors of the devices with varying thickness are proportional to their initial resistance as shown in the Fig. 6c. Relevance between gauge factor and device resistance is also considered, which is shown in Fig. 6d. It is indicated that the gauge factor increases with resistance, which is due to the decrease of the overlap area. The sub-circuit disconnection effect would be more remarkable since high resistance means lower graphene akes density. Fig. 6e indicates the resistance increasing quickly aer the limiting strain. As shown in Fig. 6f, the limiting strain is increased with the graphene thickness, which indicates that our proposed working mechanism is reasonable.
Published on 23 October 2013. Downloaded on 26/12/2013 15:34:14.
Conclusion
Fig. 5 The schematic mechanism, optical and SEM results of the graphene micro-ribbon strain sensor. (a) The schematic layered structure of the graphene micro-ribbon strain sensor in its original state. (b) The schematic layered structure of the graphene microribbon strain sensor in the stretched state. Cracks started from the upper layers. (c) The optical image showing the graphene microribbon (0.6 mm long) in its original state. (d) The optical image showing the graphene micro-ribbon (0.72 mm long) in the strained state. (e) The SEM image of the graphene micro-ribbon in its original state. (f) The SEM image of the graphene micro-ribbon in the stretched state. Obvious cracks could be identified.
strain sensor. 4 is the electronic potential. The potential distribution is computed by the nite element method under electrical conductivity distribution and xed current. The strain effect is analyzed by modeling the percolation of current through a network of randomly positioned circular akes with a uniform size distribution. As shown in Fig. 6a, the electrical conductivity distribution depends on the distribution of graphene akes and it is higher in the overlap area of multi-layer graphene akes. The decrease of the overlap of graphene akes changes the electrical conductivity of the overlap area, which means the whole resistance increases since the overlap area decreases when the device is stretched. A few sub-circuits are even disconnected as the graphene akes move to non-overlapping positions. All these effects increase the resistance of the device as the experimental result. As shown in Fig. 6b, the voltage drop is increased along with the increase of the strain under the xed current. It is consistent with the experiment results. The effect of the graphene lm on the sensitivity of the modeled strain sensor is demonstrated in Fig. 6c. It is shown that devices with a lower thickness exhibit a larger slope in the resistance–strain
This journal is © The Royal Society of Chemistry 2014
In summary, wafer-scale exible strain sensors with highperformance have been demonstrated. A Light-Scribe DVD burner could be used to obtain high quality graphene stack-layers. The graphene strain sensor has a good linear response to strain and good multi-cycle operation. Graphene micro-ribbons (20 mm width, 0.6 mm long) have been used as strain sensors to enhance the GF, of which the gauge factor is up to 9.49. Through the experimental and theoretical results, the lm cracking could be the main mechanism to explain the piezoresistance effect in our strain sensors. Our work shows that exible graphene strain sensors could be fabricated by laser scribing, which could open wide applications in medical-sensing, bio-sensing, articial skin and many other areas.
Method Material preparation GO dispersion with 2 mg mL1 concentration was provided by XFNANO Materials Tech CO., Ltd (Nanjing, China) which were synthesized from the graphite powders using a common Hummers method. 10 mL GO solution was dropped on the surface of the laser-scribed DVD disk coated with PET. Then the GO solutions were le overnight to dry on the DVD disk. The GO coated DVD disk could be patterned by the LightScribe DVD Drive (HP inc. 557S). Using the Nero StartSmart soware, the designed structure or photograph could be directly converted onto the GO lm by laser reduction of the GO into graphene. Characterization The surface morphology is observed by Quanta FEG 450 SEM (FEI Inc.) and JEM-2010 (JEOL Inc.). The Raman spectroscopy is obtained using a laser with wavelength of 514.5 nm (HORIBA Inc.). Testing for strain sensors PET substrate with the graphene device is mounted on a micrometer caliper, which can push the PET substrate to cause strain. Electrical properties are collected by a digital multimeter DM3068 (RIGOL).
Nanoscale, 2014, 6, 699–705 | 703
View Article Online
Published on 23 October 2013. Downloaded on 26/12/2013 15:34:14.
Nanoscale
Communication
Modeling of the graphene film under strain. (a) Diagram of electrical conductivity distribution of the graphene strain sensor. Different colors denote the value of electrical conductivity. (b) Representation of voltage distribution at a fixed current in a graphene film at different levels of strain. (c) Resistance vs. strain for different thicknesses of graphene film. (d) Gauge factor as a function of unstrained resistance. (e) Resistance vs. strain under large strain showing the resistance increasing fast after the limiting strain. (f) The limiting strain vs. graphene thickness.
Fig. 6
Acknowledgements This work was supported by the National Natural Science Foundation of China (61025021, 60936002, 51072089, and 61020106006), the National Key Project of Science and Technology (2011ZX02403-002) and the Special Fund for Agroscientic Research in the Public Interest (201303107). He Tian is additionally supported by the Ministry of Education Scholarship of China.
References 1 K. Novoselov, A. K. Geim, S. Morozov, D. Jiang, Y. Zhang, S. Dubonos, I. Grigorieva and A. Firsov, Science, 2004, 306, 666–669. 2 K. Novoselov, A. K. Geim, S. Morozov, D. Jiang, M. K. I. Grigorieva, S. Dubonos and A. Firsov, Nature, 2005, 438, 197–200. 3 A. K. Geim and K. S. Novoselov, Nat. Mater., 2007, 6, 183–191. 4 K. I. Bolotin, K. Sikes, Z. Jiang, M. Klima, G. Fudenberg, J. Hone, P. Kim and H. Stormer, Solid State Commun., 2008, 146, 351–355.
704 | Nanoscale, 2014, 6, 699–705
5 S. Bae, H. Kim, Y. Lee, X. Xu, J.-S. Park, Y. Zheng, J. Balakrishnan, T. Lei, H. R. Kim and Y. I. Song, Nat. Nanotechnol., 2010, 5, 574–578. 6 C. Lee, X. Wei, J. W. Kysar and J. Hone, Science, 2008, 321, 385–388. 7 F. Xia, T. Mueller, Y.-m. Lin, A. Valdes-Garcia and P. Avouris, Nat. Nanotechnol., 2009, 4, 839–843. 8 H. Tian, T.-L. Ren, D. Xie, Y.-F. Wang, C.-J. Zhou, T.-T. Feng, D. Fu, Y. Yang, P.-G. Peng and L.-G. Wang, ACS Nano, 2011, 5, 4878–4885. 9 H. Tian, Y. Yang, D. Xie, T.-L. Ren, Y. Shu, C.-J. Zhou, H. Sun, X. Liu and C.-H. Zhang, Nanoscale, 2013, 5, 890–894. 10 F. Schwierz, Nat. Nanotechnol., 2010, 5, 487–496. 11 H.-Y. Chen, H. Tian, B. Gao, S. Yu, J. Liang, J. Kang, Y. Zhang, T.-L. Ren and H.-S. P. Wong, Electrode/oxide interface engineering by inserting single-layer graphene: Application for HfO x-based resistive random access memory, 2012. 12 H. Tian, D. Xie, Y. Yang, T.-L. Ren, Y.-F. Wang, C.-J. Zhou, P.-G. Peng, L.-G. Wang and L.-T. Liu, Nanoscale, 2012, 4, 3345–3349. 13 H. Tian, H.-Y. Chen, B. Gao, S. Yu, J. Liang, Y. Yang, D. Xie, J. Kang, T.-L. Ren and Y. Zhang, Nano Lett., 2013, 13, 651–657.
This journal is © The Royal Society of Chemistry 2014
View Article Online
Published on 23 October 2013. Downloaded on 26/12/2013 15:34:14.
Communication
14 Z. Cheng, Q. Li, Z. Li, Q. Zhou and Y. Fang, Nano Lett., 2010, 10, 1864–1868. 15 Y. Lu, B. R. Goldsmith, N. J. Kybert and A. C. Johnson, Appl. Phys. Lett., 2010, 97, 083107. 16 Y. Wang, R. Yang, Z. Shi, L. Zhang, D. Shi, E. Wang and G. Zhang, ACS Nano, 2011, 5, 3645–3650. 17 Y. Lee, S. Bae, H. Jang, S. Jang, S.-E. Zhu, S. H. Sim, Y. I. Song, B. H. Hong and J.-H. Ahn, Nano Lett., 2010, 10, 490–493. 18 J. Zhao, C. He, R. Yang, Z. Shi, M. Cheng, W. Yang, G. Xie, D. Wang, D. Shi and G. Zhang, Appl. Phys. Lett., 2012, 101, 063112. 19 M. Hempel, D. Nezich, J. Kong and M. Hofmann, Nano Lett., 2012, 12, 5714–5718. 20 X. Li, R. Zhang, W. Yu, K. Wang, J. Wei, D. Wu, A. Cao, Z. Li, Y. Cheng and Q. Zheng, Sci. Rep., 2012, 2, 870.
This journal is © The Royal Society of Chemistry 2014
Nanoscale
21 S. Beeby, MEMS mechanical sensors, Artech House on Demand, 2004. 22 M.-H. Bao, Micro mechanical transducers: pressure sensors, accelerometers and gyroscopes, Elsevier Science, 2000. 23 M. F. El-Kady and R. B. Kaner, Nat. Commun., 2013, 4, 1475. 24 V. Strong, S. Dubin, M. F. El-Kady, A. Lech, Y. Wang, B. H. Weiller and R. B. Kaner, ACS Nano, 2012, 6, 1395– 1403. 25 M. F. El-Kady, V. Strong, S. Dubin and R. B. Kaner, Science, 2012, 335, 1326–1330. 26 J. Zhao, G. Y. Zhang and D. X. Shi, Chin. Phys. B, 2013, 22, 057701. 27 K. S. Kim, Y. Zhao, H. Jang, S. Y. Lee, J. M. Kim, K. S. Kim, J.-H. Ahn, P. Kim, J.-Y. Choi and B. H. Hong, Nature, 2009, 457, 706–710.
Nanoscale, 2014, 6, 699–705 | 705
