Super-elastic graphene/carbon nanotube aerogels and their application as a strain-gauge sensor

Abstract

A synergistic assembly strategy was developed to fabricate super-elastic all-carbon aerogels by integrating carbon nanotubes (CNTs) into three-dimensional graphene. The CNTs in the aerogels prevented the sliding between the graphene sheets and enhanced the stiffness of cell walls, which provided the aerogels with super-elasticity. Mechanical tests showed that the graphene/CNT aerogels could fully recover without fracture even after 90% compression. The potential application of the sensing compressive strain for the aerogels was also demonstrated. The electrical resistance response of the aerogels was highly constant to the multiple cycles of compression. The strain-gauge sensitivity of graphene/CNT aerogels could be tuned to considerably different values by controlling the aerogel density. Upon applying a strain of 30% and 60% to the graphene/CNT aerogel with a density of 37.8 mg cm⁻³, its strain-gauge sensitivities (gauge factor) reached 230% and 125%, respectively, which were superior to most of carbon-based compressible conductors.

---

1 Introduction

Graphene, a two-dimensional (2D) allotrope of carbon, has attracted tremendous attention owing to its extraordinary electrical and mechanical properties.^1,2 Effectively assembling individual graphene nano-sheets into macroscopic functional architectures is important for practical applications. There are several types of three-dimensional (3D) graphene macrostructures developed in various forms, such as graphene sponges, graphene aerogels and graphene foam (GF), which have been investigated intensively in the field of energy storage, catalysis, oil cleanup, and high performance electronics.^3–6 The porously interconnected network of 3D graphene integrates the dramatic properties of 2D graphene sheets into a single structure.⁷ Taking advantage of the combination of the electronic conductivity and the mechanical flexibility, 3D graphene can be considered as an alternative material for the fabrication of strain-gauge sensors.^8,9

Some GF/polymer composites with high elasticity, mainly due to the skeleton of elastomeric polymers, have been used for the strain-gauge sensors, such as GF/PDMS (PDMS = polydimethylsiloxane),¹⁰ GF/PU (PU = polyurethane),¹¹ and GF/PDMS/PET (PET = polyethyleneterephthalate).¹² Recently, super-elastic graphene monoliths were also prepared using small molecules or ions as binding agents to enhance the strengths of graphene frameworks and were directly tested for the pressure/strain sensitivity in compression. For example, Li et al.¹³ prepared a free-standing macro-porous graphene monolith and studied its potential application for sensing pressure deformations. Qiu et al.¹⁴ showed that the response of bulk electrical resistance of the super-elastic graphene monolith was constant over multiple cycles of compression. Sun et al.¹⁵ showed the brightness fluctuation of an LED lamp connected with a super-elastic graphene aerogel upon compressing and releasing it, indicating its promising applications as pressure-responsive sensors. It can be noted that the super-elastic graphene monoliths have higher affordable compressive strain (up to 80–90%)^11,16,17 than of the GF/polymer composites, which implied the advantages of a larger deformation testing range for strain-gauge sensors.

However, these super-elastic graphene monoliths have poor electrical conductivities (0.5 S m⁻¹ at density of 5 mg cm⁻³,¹⁶ 0.6 S m⁻¹ at a density of 1.4 mg cm⁻³,¹⁵ and 2.5 S m⁻¹ at a density of 3 mg cm⁻³ (ref. 17)), which hinder the fabrication of high-performance strain-gauge sensors. Although the electrical conductivity of 3D graphene can be improved by increasing its density, a relative low density is normally necessary for the super-elasticity because it is critical to the unique pore wall structure comprising face-to-face oriented graphene sheets.^14,17 The common methods such as the hydrothermal reduction method for preparing 3D graphene with high density easily result in high brittleness due to the randomly distributed pore wall structure. Therefore, it is still a challenge to prepare 3D graphene monoliths with both super-elasticity and high electrical conductivity for high performance strain-gauge sensors.

In the present study, we integrated carbon nanotubes (CNTs) into 3D graphene using a hydrothermal reduction method followed by a freeze-drying process. The entangled CNTs covered the graphene sheets in the as-prepared aerogels to “bundle” and “fix” them and prevented their sliding. The introduction of CNTs provided 3D graphene with an ability to maintain the structural integrity under compression and brought the graphene/CNT aerogels super-elasticity. In addition, the electrical conductivity of the aerogels could be further improved by increasing their density to tune the strain/sensing capability for the potential application in low- and high-pressure-sensing. In addition, the variation of the electrical resistance and gauge factor of the as-prepared graphene/CNT aerogels under mechanical deformation was tested in addition to the mechanical and micro-structural analysis.

2 Experimental sections

2.1 Preparation of graphene/CNT aerogels

Graphene oxide (GO) was prepared using the modified Hummers' method from natural graphite and exfoliated GO sheets were obtained in their aqueous dispersions according to the previous process.¹⁸ CNTs (diameter of 30–50 nm, length of 50–100 μm, purchased from Beijing DK nano-technology Co. LTD.) were functionalized by refluxing in a mixture of concentrated H₂SO₄ and HNO₃ (3 : 1 by volume) for approximately 80 min and then collected by repeated centrifuging and washing with deionized water. Typically, the fCNT aqueous dispersion was added to a GO aqueous dispersion with a mass ratio of GO : fCNT = 3 : 1, and the mixture was sonicated for 30 min. The graphene/CNT hydrogel assembled with GO sheets and fCNTs was prepared using a hydrothermal reaction maintained at 180 °C for 12 h in a Teflon-lined autoclave. After washing the as-obtained hydrogel with distilled water followed by freeze-drying, the graphene/CNT aerogels were produced. A graphene aerogel without CNTs was also prepared in the same way for comparison. In addition, to examine the effects of the electrical conductivity of the aerogels on their strain-gauge sensitivity, graphene/CNT aerogels with various densities tuned by GO/fCNT concentrations were prepared. The as-prepared aerogels were defined as graphene/CNT-n (n = 2, 4, 6, 9, 12) based on a GO/fCNT concentration of 2, 4, 6, 9 and 12 mg mL⁻¹, respectively.

2.2 Characterizations

Scanning electron microscopy (SEM) was conducted on a Hitachi S4800 field-emission SEM system with a 5 kV accelerating voltage. The compressive stress–strain measurements were performed using an Instron Model 5848 microtester with two test plates. The aerogels (5 mm × 4 mm × 3 mm) were set on the lower plate and compressed by the upper plate connected to a load cell. The strain ramp rate was controlled to be 1 mm min⁻¹ and the cyclic loading was carried out at a displacement of 30 mm min⁻¹. A 1000 N load cell was used to test the mechanical properties of the aerogels. The electrical resistance variation was measured by a two-probe method under mechanical deformation. During the measurement, two copper sheets served as electrodes to connect to aerogels and a Keithly 2410 Source Meter instrument (Keithly Inc. USA). Every electromechanical experiment was repeated 3 times and the electrical resistance was the average value.

3 Results and discussion

SEM (Fig. 1a and S1†) showed that graphene/CNT aerogels exhibited a porous structure with interconnected hollow cells ranging from 0.5 to 3 μm in diameter, sharing a similar structure to the graphene aerogel without CNTs (Fig. S2†). The cell walls were made up of assemblies of graphene sheets and CNTs produced during the hydrothermal reduction processes. With increasing the GO/fCNT concentration, the porous structure became denser and the average pore size became smaller due to more graphene sheets and CNTs packed in the 3D architecture (Fig. S1†). Herein, graphene/CNT-9 was taken as an example for studying the interaction between graphene sheets and CNTs in the aerogels. Different from the cell walls of graphene aerogels assembled with only overlapping and wrapping graphene sheets (Fig. S2†), the cell walls of the graphene/CNT aerogels consisted of graphene sheets and CNTs coated on the sheets (Fig. 1a–c). The structural changes in the GO sheets and fCNTs indicated that the hydrothermal reduction process removes the partial-oxygen containing functional groups of GO and fCNTs (Fig. S3†) and provides strong π–π interactions between the graphene sheets and CNTs. Therefore, the entangled CNTs could tightly cover the assembled graphene sheets of the cell walls then “bundled” and “tied up” them (Fig. 1b and c). This offered a mechanism to limit the sliding graphene sheets and enhance the stiffness of the cell walls (corresponding cartoons of the interconnections between the CNTs and graphene sheets are displayed in Fig. 1d–f), which provided aerogels with the ability to maintain the structural integrity upon cyclic compression.

Fig. 1 (a) SEM images of the porous structure of graphene/CNT-9; both of (b) over-lapping graphene sheets and (c) wrapping graphene sheets of the cell walls are “bundled” and “tied up” with entangled CNTs; (d–f) cartoon models corresponding to the micro-structures shown in (a–c), respectively.

The excellent resilience was of core interest for graphene/CNT aerogels in the application of strain-gauge sensors. Digital images of the graphene/CNT-9 aerogel recorded during the loading/unloading process are demonstrated in Fig. 2a, indicating a far better elasticity than graphene aerogel without CNTs (Fig. S4†). Once the external pressure was withdrawn, graphene/CNT-9 aerogel could recover almost to its original shape rapidly. The stress–strain curves of the graphene/CNT-9 aerogel are shown in Fig. 2b. Similar to other elastomeric foams,^19–21 three regimes of deformation could be observed in the loading stress–strain curve: strain < 10%, nearly linear elastic regime, corresponding to bending of cell walls; 10% < strain < 60%, relatively flat stress plateau, corresponding to elastic buckling of cell walls; and strain > 60%, abrupt stress increasing regime, corresponding to the densification of cells. The stress loops with 30%, 60% and 90% strain indicated that a large portion of energy (70–80%) was absorbed during compression. The energy dissipation was caused mostly by the friction between the cell blocks or movement of air through the pores. It can be noted that the recoverable strain of the graphene/CNT aerogels reached up to 90%, showing a much higher compressibility than graphene aerogels without CNTs (Fig. S5†). We even could not measure this during unloading of the graphene aerogel because of the permanent deformation during loading to strain of ∼40%. The fragility of the graphene aerogel was due to the lack of a restorative force to act on the graphene frameworks. Previous studies suggested that when 3D graphene monoliths made of a few layers of graphene was severely compressed, the intersheet van der Waals adhesion would overwhelm the stored elasticity preventing elastic recovery.¹⁴ In contrast, graphene/CNT aerogels presented super-elasticity, which was attributed to their unique hierarchical structure and the interactions between CNTs and graphene sheets. First, the entangled CNTs attached to the graphene sheets and tightly bonded them together, which ensured the deformation of cell walls rather than the sliding between them during compression. Second, the coating of CNTs reinforced the relatively flexible graphene substrate and endowed their intrinsic elasticity to the co-organized aerogel.¹⁵ When the aerogels were subjected to an external stress, the load can effectively transfer from the loosely connected graphene sheets to the entangled CNT nets, which could greatly enhance the strength and elastic stiffness of the cell walls, allowing the van der Waals adhesion to be overcome by the elastic energy.

Fig. 2 (a) Set of real-time images of the compressed graphene/CNT-9 showing the recovering process. (b) Stress–strain curves of graphene/CNT-9 at different maximum strain of 30%, 60% and 90%, respectively. (c) Stress–strain curves of 1^st and 20^th cycles on the graphene/CNT-6, -9 and -12, respectively, during repeated compression.

The stress–strain curves of graphene/CNT-6, -9 and -12 were similar in their shapes (Fig. 2c). The maximum stress at 60% strain was 21.2 kPa for graphene/CNT-6, whereas it was 62.5 kPa for graphene/CNT-12. The mechanical properties of the graphene/CNT aerogels depended strongly on their micro-morphology. Small pore size will provide the aerogels with high stiffness and compressive strengths. The thickness reduction (plastic deformation) of the graphene/CNT aerogels, which was derived from the intersection of the stress curve with the strain coordinate, was only about 10%, 8% and 5% for graphene/CNT-6, -9, and -12, respectively, at a strain of 60% after 20 cycles, indicating excellent mechanical robustness.

The electrical properties of the graphene/CNT aerogels were also studied. Their electrical conductivities were measured using a two-point probe method. For most of the 3D graphene, the electrical conductivity was influenced by the weight densities and the thicknesses of cell walls. As shown in Fig. 3a, the densities of the graphene/CNT aerogels could be tuned well by GO/fCNT concentrations. The electrical conductivity of the graphene/CNT aerogels increased by around 35 times (up to 28.1 S m⁻¹) when the density was increased to 7.6 times (from 2.8 to 21.3 mg cm⁻³). The electrical conductivity was much higher than that of other super-elastic graphene monoliths made through chemical modification routes without further treatments,^11,16,17 highlighting the advantage of graphene/CNT aerogels for electrical applications. Graphene/CNT aerogels also showed high electromechanical stability with little variation of conductivity at a fixed strain after 100 cycles (Fig. 3b). The electrical conductivity increased slowly with increasing strain from 30% to 60% and increased exponentially with strain of >60%, which is consistent with the mechanical behavior of the graphene/CNT aerogels. At large deformation stages (strain > 60%), the density of contact spots between the graphene/CNT skeletons increased rapidly, resulting in an exponential increase in electrical conductivity.

Fig. 3 (a) Density and electrical conductivity of graphene/CNT aerogels with various GO/fCNT concentrations. (b) Electrical conductivity change in the graphene/CNT-9 aerogels when compressed to a set strain and then being recovered for several cycles.

Combining good electrical conductivity and excellent elasticity, the graphene/CNT aerogels were suitable for sensing pressure deformations. To demonstrate this potential application, graphene/CNT-9 aerogel as a compressible conductor was connected with a lamp under a 3 V circuit. The lamp brightness fluctuated upon compression, releasing the aerogels (Fig. 4a). The electrical resistance was measured while deforming the graphene/CNT-9 aerogels under cyclic compressive loading to quantify the change in electrical resistance of the aerogels in response to an external pressure. The relative change in electrical resistance (ΔR/R₀) of the graphene/CNT-9 aerogels at a strain of 30%, 60% and 90% is shown in Fig. 4b. The response of the bulk electrical resistance of the aerogels behaved in the same way over multiple cycles of compression, further indicating their remarkable structural resilience. The change in the electrical resistance of graphene/CNT-9 aerogel increased from ∼45% to ∼100% of the original value corresponding to the strain from 30% to 90%. The mechanism for the strain-resistance relationship during compression could be explained by the change in contact area in the aerogels. When the compression strain was loaded on the aerogels, the volume and height shrank along the force. The micro-structure of graphene/CNT aerogels could effectively accommodate deformation without significant sliding and collapse under strain. The external strain induced an increase in contact area between the cell walls and decreased the resistance of the aerogels. Therefore, larger compressive strain would lead to a higher contact area in the aerogels and larger change in electrical resistance. The strain-resistance mechanism could be verified under SEM (Fig. 5). Once the strain was applied, bending or buckling of the cell walls decreased the interlayer distances in the graphene conductive network and increased the contact area, eventually leading to resistance variations.

Fig. 4 (a) Circuit constructed with a graphene/CNT aerogels as compressible conductors. (b) Electrical resistance variation of graphene/CNT-9 aerogel after a strain of 30%, 60% and 90% and then releasing for each cycle.

Fig. 5 SEM images of the loading and unloading states of graphene/CNT-9 aerogels under compression.

In addition, an irreversible change in the original resistance (when the strain is zero) of graphene/CNT-9 aerogels occurred in the first cycle (Fig. 4a), then stabilized afterwards. The irreversible part of the change in resistance in the first cycle became larger with higher levels of strain induced in the aerogels, which may be due to partial breakage or restructuring of a larger network structure than the lower strain values. Hence, prior to sensor assembly, several cycles of loading-releasing could be implemented as a preconditioning step to achieve reproducible results during the subsequent measurements.

A strain-gauge sensor should have a tuneable strain/pressure sensing capability to be implemented successfully in low- and high-pressure-sensing applications. Graphene/CNT aerogels with various densities were prepared to investigate the effect of the electrical conductivity on the sensitivity. Plots of ΔR/R₀, as a function of 60% strain, for graphene/CNT-6, -9 and -12 aerogels are depicted in Fig. 6a (the plots of graphene/CNT-2 and -4, as shown in Fig. S6†). A 60% compressive strain caused a ∼60% change in the electrical resistance for the graphene/CNT-9 aerogel. Under the same compressive strain, the graphene/CNT-6 and -12 aerogels result in a ΔR/R₀ value of ∼40% and ∼90%, respectively. In terms of the strain gauge sensitivity, the gauge factor was usually defined as the relative change in electrical resistance (ΔR/R₀) to the mechanical strain. The gauge factors of the graphene/CNT aerogels with various densities at different strains were calculated, as plotted in Fig. 6b. It was found that the graphene aerogels with a higher density possessed larger gauge factors at the same strain. Therefore, the strain-gauge sensitivity could be tuned to a desired value range by controlling the density of the aerogels, because it caused a change in the electron conductivity of the aerogels. This trend of the gauge factor as a function of the strain and density was consistent with previous reports.^23,24 In addition, under large compression strain, a few cycles were needed for the gauge factor of the graphene/CNTs aerogels to become stable (Fig. 6c). Table 1 showed a comparison of the gauge factors between graphene/CNT-12 aerogels and carbon-based compressible materials reported previously. At strain of 30% and 60%, the gauge factors of the aerogels in our studies reached 230% and 125%, respectively, which were superior to some CNT sponges and graphene monoliths and even comparable to some GF/polymer composites.

Fig. 6 (a) Electrical resistance change in graphene/CNT-6, -9, -12 aerogels when compressed repeatedly up to 60% strain for over 100 cycles. (b) Gauge factors (for the 2^nd cycle) of graphene/CNT aerogels at a strain of 30%, 60% and 90%, respectively. (c) Gauge factor of graphene/CNT aerogels at 60% strain for each cycle.

Table 1 Comparison of the gauge factor between our work and previous work, showing the excellent performance of our designed network in strain-gauge sensors. The electrical conductivity of conductors was measured without compression

Compressible conductors	Density (mg cm^−3)	Conductivity (S m^−1)	Stress (kPa)	Gauge factor	Ref.
CNT sponges	7.3	47	∼32 kPa at 60% strain	∼24% at 60% strain	22
CNT/Ag sponges	17.9	N.A.	3.5 kPa at 40% strain	∼45% at 40% strain	23
Graphene monoliths	8.6	0.75	∼7 kPa at 30% strain	∼58% at 30% strain	13
Graphene monoliths	5.1	12	∼20 kPa at 50% strain	∼120% at 50% strain	14
Graphene/PDMS	N.A.	N.A.	∼300 kPa at 30% strain	∼300% at 30% strain	24
Graphene/PU	1.54	0.22	∼15 kPa at 60% strain	∼140% at 60% strain	21
Graphene coated CNT aerogels	14	30	∼190 kPa at 60% strain	∼50% at 60% strain	25
Graphene/CNT-12 aerogels	21.3	28.1	43 kPa at 30% strain	230% at 30% strain	Our works
			53 kPa at 60% strain	125% at 60% strain

---

4 Conclusions

3D porous graphene/CNT aerogels constructed with cell walls of graphene sheets and CNT ribs were synthesized using a hydrothermal reduction method. The entangled CNTs covered the graphene sheets to prevent sliding between each other, which gave the aerogels super-elasticity. The electrical conductivity of the aerogels could also be improved by increasing their densities. Upon the application of strain-gauge sensors, graphene/CNT aerogels generated large variation of electrical resistance through the change in the contact area during mechanical deformation. In addition, the strain-gauge sensitivity of graphene/CNT aerogels could be tuned to the desired value by controlling the density of the aerogels. The gauge factor of graphene/CNT-12 aerogels reached 230% and 125% at a strain of 30% and 60%, respectively. This architectural design of the high performance aerogels would provide new insight into the development of strain-gauge sensors and elastic conductors.

Acknowledgements

This study was supported by the National Natural Science Foundations of China (No. 51503102 and 61274054), the Natural Science Foundation of Jiangsu Province, China (No. BK20140869), and the NUPTSF (No. NY214055 and NY213082).

References

1. A. K. Geim, Science, 2009, 324, 1530.
2. D. Li and R. B. Kaner, Science, 2008, 320, 1170.
3. Z. P. Chen, W. C. Ren, L. B. Gao, B. L. Liu, S. F. Pei and H. M. Chen, Nat. Mater., 2011, 10, 424.
4. H. C. Bi, X. Xie, K. B. Yin, Y. L. Zhou, S. Wan, L. B. He, F. Xu, F. Banhart, L. T. Sun and R. S. Ruoff, Adv. Funct. Mater., 2012, 22, 4421.
5. X. T. Zhang, Z. Y. Sui, B. Xu, S. F. Yue, Y. J. Luo, W. C. Zhan and B. Liu, J. Mater. Chem., 2011, 21, 6494.
6. Z. Xu, Y. Zhang, P. G. Li and C. Gao, ACS Nano, 2012, 6, 7103.
7. X. C. Dong, X. W. Wang, L. H. Wang, H. Song, H. Zhang, W. Huang and P. Chen, ACS Appl. Mater. Interfaces, 2012, 4, 3129.
8. Y. R. Jeong, H. Park, S. W. Jin, S. Y. Hong, S. S. Lee and J. S. Ha, Adv. Funct. Mater., 2015, 25, 4228.
9. J. Kuang, L. Q. Liu, Y. Gao, D. Zhou, Z. Chen, B. H. Han and Z. Zhang, Nanoscale, 2013, 5, 12171.
10. Y. A. Samad, Y. Q. Li, A. Schiffer, S. M. Alhassan and K. Liao, Small, 2015, 11, 2380.
11. H. B. Yao, J. Ge, C. F. Wang, X. Wang, W. Hu, Z. J. Zheng, Y. Ni and S. H. Yu, Adv. Mater., 2013, 25, 6692.
12. R. Q. Xu, Y. Q. Lu, C. H. Jiang, J. Chen, P. Mao, G. H. Gao, L. B. Zhang and S. Wu, ACS Appl. Mater. Interfaces, 2014, 6, 13455.
13. Y. R. Li, J. Chen, L. Huang, C. Li, J. D. Hong and G. Q. Shi, Adv. Mater., 2014, 26, 4789.
14. L. Qiu, J. Z. Liu, S. L. Y. Chang, Y. Z. Wu and D. Li, Nat. Commun., 2012, 3, 1241.
15. H. Y. Sun, Z. Xu and C. Gao, Adv. Mater., 2013, 25, 2554.
16. J. Zhong, J. Meng, X. C. Gui, T. Hu, N. Xie, X. Y. Lu, Z. Y. Yang and N. Koratkar, Carbon, 2014, 77, 637.
17. H. Hu, Z. B. Zhao, W. B. Wan, Y. Gogotsi and J. S. Qiu, Adv. Mater., 2013, 25, 2219.
18. L. W. Peng, Y. Y. Feng, P. Lv, D. Lei, Y. T. Shen, Y. Li and W. Feng, J. Phys. Chem. C, 2012, 116, 4670.
19. T. A. Schaedler, A. J. Jacobsen, A. Torrents, A. E. Sorensen, J. Lian, J. R. Greer, L. Valdevit and W. B. Carter, Science, 2011, 334, 962.
20. J. H. Zou, J. H. Liu, A. S. Karakoti, A. Kumar, D. Joung, Q. Li, S. I. Khondaker, S. Seal and L. Zhai, ACS Nano, 2010, 4, 7293.
21. C. Wu, X. Y. Huang, X. F. Wu, R. Qian and P. K. Jiang, Adv. Mater., 2013, 25, 5658.
22. X. C. Gui, A. Y. Cao, J. Q. Wei, H. B. Li, Y. Jia, Z. Li, L. L. Fan, K. L. Wang, H. W. Zhu and D. H. Wu, ACS Nano, 2010, 4, 2320.
23. S. F. Zhao, Y. J. Gao, G. P. Zhang, L. B. Deng, J. H. Li, R. Sun and C. P. Wong, Carbon, 2015, 86, 225.
24. Y. A. Samad, Y. Q. Li, S. M. Alhassan and K. Liao, ACS Appl. Mater. Interfaces, 2015, 7, 9195.
25. K. H. Kim, Y. Oh and M. F. Islam, Nat. Nanotechnol., 2012, 7, 562.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra20342b

---