Highly compressible behavior of polymer mediated three-dimensional network of graphene foam

Abstract

The compressive behavior of graphene foam (GF) and its polymer (polydimethyl siloxane) (PDMS) infiltrated structure are presented. While GF showed an irreversible compressibility, the GF/PDMS structure revealed a highly reversible mechanical behavior up to many cycles of compression and also possesses a six times higher compressive strength. In addition, the strain rate demonstrated a negligible effect on both the maximum achieved stress and energy absorption in the GF/PDMS structure. The mechanical responses of both GF and GF/PDMS structure are compared with carbon nanotubes based cellular structure and its composite with PDMS, where GF/PDMS presented a dominant mechanical characteristic among other carbon based micro foam structures. Therefore, the improved mechanical properties of GF/PDMS suggest its potential for dampers, cushions, packaging, etc.

---

1. Introduction

Graphene^1,2 is a monolayer of carbon atoms arranged in a two-dimensional honeycomb like lattice that exhibits excellent optical, electrical, thermal and mechanical properties.^3–7 Therefore, it has numerous applications in fields such as in sensor technology, optoelectronics and energy applications.^8–13 Recently, a three-dimensional counterpart of graphene has been developed using exfoliation of graphite into graphene oxide (GO) as well as growing a graphene layer on a foam network of nickel metal.^14,15 Although exfoliated GO presents a foam like structure, it offers poor control over structural characteristics in a random network. However, graphene on a nickel network not only allows controlling the number of layers but also structural porosity. So far, a carbon nanotubes (CNT) based cellular structure has shown strong potential for a flexible and lightweight material for cushioning and impact protection due to its entangled microstructure that is referred to as foam, turf, cellular structure, etc. in several studies.^16–18 These studies have evaluated the mechanical behavior of CNT under both static and dynamic impacts for cushioning, actuators, shock absorbers, and dampers.^19–21 Although CNT has demonstrated excellent energy absorbing capability²² its entangled microstructure means there is little control over its structural characteristics, which are crucial parameters for compressive studies.²³ On the other hand, a three-dimensional network of graphene foam (GF) grown on a nickel network has emerged as a new material due to its variable porosity with a high surface area. In addition, a conductive network of GF provides continuous channels for electron transport even after impregnation with nonconductive polymers.²⁴ Hence, it provides a novel filler material to fabricate conductive polymeric foam applicable for sensors and actuators. Recently, a GO based polymeric network of foam was demonstrated to exhibit large compressive strain where the maximum achieved stress was in the range of ∼0.4 MPa for one compression cycle only.²⁵ Another study reported the compressive behavior of GO coated polyurethane foam that revealed even lower stress of a few kPa for a strain of 50%.²⁶ So far, none of the studies on three-dimensional networks of graphene oxide have presented a better mechanical response than CNT foam. Hence, materials derived from a nickel coated graphene network (GF) could provide improved mechanical properties because of the tailored porosity and continuous network as mentioned above. Though bending, stretching and twisting tests have been performed by Chen et al., to evaluate the mechanical behavior of a GF/polydimethyl siloxane (PDMS) composite,²⁴ no report has yet presented a systematic study on the compressive behavior. Since GF/PDMS is a promising composite material, a detailed study on its compressive behavior is a necessity. This report presents a detailed study on the mechanical behavior of both GF and GF/PDMS foam under compression. The resulting responses are also compared with CNT foam and its composite with PDMS to demonstrate the superior mechanical performance of the GF derivatives.

2. Experimental section

GF was synthesized by atmospheric pressure chemical vapor deposition (CVD) using nickel foam as a template for the growth of a three-dimensional network. In this process, the nickel foam template was kept inside a horizontal tube furnace. Initially, argon gas was purged to maintain an inert atmosphere inside the chamber and the temperature was ramped up to 1000 °C over 25 minutes duration. Substrate annealing was conducted in the presence of hydrogen and argon gases for 15 minutes to clean the nickel surface from oxide layer. GF growth was conducted in the presence of methane gas introduced for 2.5 minutes. After growth, the furnace was cooled down to room temperature in the presence of both hydrogen and argon gases. The cooling procedure causes the formation of wrinkles in the graphene layer deposited on the nickel truss because of the difference between the coefficients of thermal expansion of nickel and of graphene. Wrinkles are generated when a flake starts growing and overlaps with another flake generated from the next step edge. These wrinkles might also help in improving the mechanical performance of the GF.^24,27

The transfer process of GF is mediated through poly(bisphenol carbonate) (PC) by coating it on graphene deposited on nickel foam. Thereafter, nickel was etched by dipping in a 3 M HCl solution at 80 °C for 8 hours. In order to remove PC, the GF/PC stack was dipped in chloroform for 30 minutes. The steps involved in the growth of GF and preparation of GF/PDMS are depicted in Fig. 1. Flexible GF/PDMS foam (as shown in the optical photograph in Fig. 1) was prepared from GF by infiltrating graphene foam with PDMS.²⁴ The monomer to linker ratio was maintained at about 1 : 10 and the air bubbles were removed by applying a vacuum. Finally, the composite was cured at 80 °C for 3 hours in order to strengthen the bonds between the linkers. During the composite formation, polymer fills the cells of the GF structure as well as coating the wrinkles of the graphene (as will be shown later). The bulk densities of GF and GF/PDMS were calculated to be 0.54 g cm⁻³ and 1.06 g cm⁻³, respectively.

Fig. 1 Schematic diagram showing the experimental steps involved in the growth of graphene foam by CVD and synthesis of the GF/PDMS composite. Optical photographs show the GF and flexible GF/PDMS composite.

Vertically aligned CNT mats were also synthesized for a comparative mechanical study with the GF. The bulk structure of CNT was grown on a silicon dioxide substrate using a CVD process and the details of the process can be found elsewhere.¹⁷ The bulk density of the individual CNT specimen was ∼0.3 g cm⁻³. The CNT/PDMS composite was prepared by infiltrating CNT mats with PDMS,²⁸ in a process similar to that of the GF/PDMS preparation as described earlier. The bulk density of the resulting CNT/PDMS was measured to be 1.2 g cm⁻³.

The compressive behavior of GF, GF/PDMS, CNT and CNT/PDMS were studied at various strains of 20, 40, 60 and 70% for several cycles of compression. Uni-axial compression tests were performed for both loading and unloading the sample with a cross-head speed of 1 mm min⁻¹ under a displacement controlled mode using a universal testing machine (UTM). At each strain, the compression characteristics such as peak stress (maximum stress at maximum strain), energy absorption capability and elastic modulus were measured along with analyzing the effect of the strain rate varying from 50 to 1000 min⁻¹ on the mechanical behavior of the GF/PDMS foam. To reduce experimental error, each experiment was performed three times and the corresponding standard deviation was calculated. A comparative study on the performance of each above-mentioned material is presented.

3. Results and discussion

Fig. 2 illustrates scanning electron microscope (SEM) images to reveal the morphology of the three-dimensional GF microstructure. Fig. 2a depicts the microstructure of the as-grown GF, showing a micro-porous structure with a pore size of 100–200 μm. The right inset in Fig. 2a is a magnified image revealing the ripples/wrinkles at the grain boundaries of the graphene flakes.²⁹ A small trace of nickel was also observed in the GF even though prolonged etching was performed, and this is evident in the energy dispersive X-ray analysis (EDAX) in the left inset of Fig. 2a. The SEM image in Fig. 2b demonstrates GF/PDMS where GF is uniformly covered with PDMS after impregnation and the magnified image in the inset depicts the covering of the ripples/wrinkles (earlier shown in Fig. 2a) by PDMS.

Fig. 2 SEM images showing the microstructure: (a) three-dimensional GF networks; (b) GF/PDMS composite; the right insets show the magnified images, and the left inset to (a) shows the EDAX of GF; (c) Raman spectrum of GF showing a single layer; (d) Raman spectrum of GF showing a few layers of graphene.

The Raman spectra of the as-grown GF shown in Fig. 2c and d reveal two prominent peaks at 1560 and 2700 cm⁻¹ corresponding to G and 2D bands of graphene. The absence of a D peak at 1350 cm⁻¹ represents a high quality of as-grown graphene on nickel substrate. The intensity ratio of G to 2D is calculated to be 0.61, which indicates the presence of single layer graphene at some places,³⁰ while at other places this ratio is 2.68, which indicates the presence of the growth of a few layers of graphene. This difference may be attributed to the different crystal orientation of the polycrystalline nickel grains in the foam, resulting in different diffusion rates of carbon atoms to the surface.²⁹

The stress–strain behavior of GF and GF/PDMS is shown in Fig. 3a and b. The loading curve is divided into two regions; elastic and densification. The elastic modulus is calculated from the linear regime of the loading curve. Both GF and GF/PDMS demonstrated a large hysteresis between loading and unloading curves, which is similar to the behavior observed in open cell foams.^23,31 The peak stress of GF was measured as ∼6 MPa at a strain of 70%, which is larger than any other reported values on GF. Moreover, the peak stress of GF/PDMS increased by 6 times compared with GF alone. However, when unloading was conducted after the loading strain reached its maximum value, the GF did not recover to its initial position. This demonstrates that the cells in the GF break during the loading of the sample, revealing very weak and fragile mechanical characteristics.²⁴

Fig. 3 Stress–strain curves of (a) GF and (b) GF/PDMS composite; (c) peak stress vs. strain plot and (d) energy absorbed vs. strain plot for GF and GF/PDMS composite.

On the other hand, the GF/PDMS structure demonstrated a complete recovery after unloading at all strains, as can be clearly observed in Fig. 3b, which reveals a strengthened structure of GF after PDMS is filled into the GF pores. In order to systematically evaluate the compressibility of both the GF and the GF/PDMS composite, the peak stress values were calculated at various strains, as illustrated in Fig. 3c. It is clear that as strain increased, there was a nonlinear increase in the stress for both GF (bottom curve) and GF/PDMS. In GF, this increase in the stress can be attributed to the bending of cells in graphene foam at a smaller strain of 20% hence reveals an elastic regime and results in a smaller peak stress of 0.44 MPa. Upon further increasing the strain to 40%, the load reached a critical value when the cells in GF started buckling after bending. Hence, the stress does not change considerably with the strain. At a strain of 60%, the edges of adjacent cells started touching each other and formed a dense structure of graphene that resulted in a considerable increase in the stress of 3.7 MPa. Upon further increasing the strain to 70%, the stress significantly increased to 5.4 MPa, which was due to the collapse of cell walls that resulted in a denser foam structure. It is worth mentioning that the presence of nickel (as mentioned earlier) could also contribute to enhanced mechanical behavior of the GF. In addition, the presence of trace amount of nickel helps in better handling of GF, by providing a support, since GF with no nickel content is very fragile and brittle in nature.²⁴

Similarly, the mechanical behavior of the GF/PDMS composite was analyzed and the results are shown in Fig. 3c. At a strain of 20% the composite showed a peak stress of 1.8 MPa, which is four times higher than that of GF alone. For higher strain values of 40%, 60% and 70%, the peak stress values for the GF/PDMS were 6.5 MPa, 19.8 MPa and 30.3 MPa, respectively; these values are much higher than the corresponding peak stress values of GF alone as discussed above as well as in any other report. This increase in peak stress is attributed to the strengthening of the structure because of the PDMS filling inside the foam cells and it also covers the wrinkled graphene, thereby resulting in thickening of the cell walls. Hence, a higher bulk density of GF/PDMS assists in the increased peak stress.

Furthermore, the energy absorption capabilities of the GF and the GF/PDMS were evaluated from the area under the loading curve, as shown in Fig. 3d. At lower strains, both the specimens are shown to follow Hooke's law, as can be seen in Fig. 3a and b. The energy absorption in GF was measured to be 0.04 and 0.14 MJ mm⁻³ at 20% and 40% strains, respectively. However, at higher strains of 60% and 70%, the energy absorption capabilities enhanced to 0.63 and 0.98 MJ mm⁻³, respectively. Thus this maximum strain region reveals a densification regime, where the energy absorption increases considerably with the increase in strain.²³ However, GF/PDMS showed an energy absorption of 0.15 MJ mm⁻³, at a strain of 20%, as compared with 0.04 MJ mm⁻³ observed in GF alone. Similarly for other strains, the energy absorption values for GF/PDMS were 0.92, 3.3 and 6 MJ mm⁻³ respectively, which is again much higher than the corresponding energy absorption values of GF. At a maximum strain of 70%, the energy value for GF/PDMS was found to be more than six times the energy absorption in GF.

Cyclic compression tests were conducted on the GF/PDMS composite for 100 cycles to study the recoverability. Fig. 4a shows the stress–strain curves for compression of the composite at 30% strain. The area under the stress–strain curve remained constant even after compression for 100 cycles, showing a constant energy absorption, and hence a stable composite. Fig. 4b shows similar stress–strain curves for 50% strain, which demonstrates the stability and recoverability of the composite even at higher strains. Similar experiments were performed at a highest strain of 70% (Fig. 4c). At a higher strain of 70%, the energy absorption decreased in the initial cycles, and finally achieved stabilization after a gradual densification. The variation in the energy absorption with the number of compression cycles is shown in Fig. 4d.

Fig. 4 Cyclical compression results showing stress vs. strain plots for the cyclic compression of GF/PDMS composite for 100 cycles at (a) 30% and (b) 50% strain, and (c) 50 cycles at 70% strain. (d) Variation in the energy absorption with 50 cycles at 70% strain.

Moreover, the strain rate effect on the compressive behavior of the GF/PDMS composite was elucidated with varying strain rates of 50, 200, 500 and 1000 min⁻¹. Both the peak stress and energy absorption capability were monitored with the variation in the strain rate, as illustrated in Fig. 5a and b. At a strain rate of 50 min⁻¹, the peak stress and the energy absorption were 34 MPa and 5.47 MJ mm⁻³, respectively, upon further increasing the strain rate to 500 and 1000 min⁻¹, a nominal decrease in the peak stress was observed (32 and 30 MPa, respectively). A similar trend of decrease was also observed in energy absorption with the variation in the strain rate, as revealed in Fig. 5b. This nominal effect of strain rate on both the peak stress and energy absorption was observed due to the deformation of the sample by localized stresses that soften it.²³ Therefore, these results present a novel continuous composite with excellent mechanical strength as compared with other carbon based composites such as CNT having discrete structural characteristics.

Fig. 5 (a) Peak stress and (b) energy absorption plotted vs. the strain rate for the GF/PDMS composite.

Finally, the mechanical response of both GF and GF/PDMS were compared with the most studied mechanical responses of CNT and CNT/PDMS, as shown in Fig. 6. Fig. 6a shows SEM images of the CNT microstructure, before (left) and after (right) impregnation with PDMS. Fig. 6b and c illustrate the peak stress and the energy absorption of CNT and GF and also of their composites with PDMS.

Fig. 6 (a) SEM images of CNT (left) and CNT/PDMS (right). Comparisons of mechanical properties of GF, CNT, CNT/PDMS, and GF/PDMS foams: (b) peak stress; (c) energy absorption; (d) elastic modulus.

The CNT structure under uni-axial compression revealed peak stress and energy absorption values of about 9.1 MPa and 2.76 MJ mm⁻³, respectively,¹⁷ while much higher values of 23 MPa and 4.3 MJ mm⁻³ were measured for CNT foam infiltrated with PDMS. The stronger and recoverable composite of CNT/PDMS was due to the interfacial interactions resulting in a strong coupling between its constituents.^28,32 Both CNT and CNT/PDMS demonstrated higher mechanical responses than GF alone but GF/PDMS demonstrated superior mechanical behavior among all four derivatives of CNT and GF. Fig. 6d shows a comparison of the elastic modulus of the different foams. GF had an elastic modulus of 1.88 MPa, while the CNT mat had a higher value of 5.34 MPa, suggesting an enhanced strength. When infiltrated with PDMS, the elastic modulus of CNT increases to 7.63 MPa. However, the highest elastic modulus among all the foams and their composites was shown by the GF/PDMS composite (14.4 MPa), which is more than seven times higher than that of GF and almost double that of CNT foam. From the above results it is clear that the GF composite, due to its continuous structure, demonstrated an enhanced mechanical response upon compression as compared with the CNT. The weakening of the CNT structure was probably due to the sliding among each other during compression.

4. Conclusion

In this work, a study of the compressive behavior of GF and GF/PDMS composite is presented. The compression measurements were conducted at different strains of 20, 40, 60 and 70%. Interestingly, the mechanical properties such as peak stress, energy absorption and elastic modulus of GF were enhanced by more than six times upon infiltration with PDMS. Cyclic tests were performed on the GF/PDMS composite to reveal a high stability, indicating its potential application in shape memory, dampers, shock absorbers and cushioning, etc. In addition, the peak stress and energy absorption were shown to be independent of the strain rate. The GF composite was shown to exhibit much higher compressive strength than the CNT foam and its composite.

Acknowledgements

We would like to acknowledge Mr Srikanta, MNCF, CeNSE, IISc, for the help in conducting mechanical tests. We would also like to acknowledge Mr Varadaraja and Mr Prahlada, MNCF, IISc, for the SEM images and Raman analysis, respectively.

Notes and references

1. A. K. Geim and K. S. Novoselov, Nat. Mater., 2007, 6, 183.
2. A. K. Geim, Science, 2009, 324, 1530.
3. M. Liu, X. Yin, E. Ulin-Avila, B. Geng, T. Zentgraf, L. Ju, F. Wang and X. Zhang, Nature, 2011, 474, 64.
4. K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y. Zhang, S. V. Dubonos, I. V. Grigorieva and A. A. Firsov, Science, 2004, 306, 666.
5. A. A. Balandin, S. Ghosh, W. Bao, I. Calizo, D. Teweldebrhan, F. Miao and C. N. Lau, Nano Lett., 2008, 8, 902.
6. A. A. Balandin, Nat. Mater., 2011, 10, 569.
7. C. Lee, X. Wei, J. W. Kysar and J. Hone, Science, 2008, 321, 385.
8. F. Schedin, A. K. Geim, S. V. Morozov, E. W. Hill, P. Blake, M. I. Katsnelson and K. S. Novoselov, Nat. Mater., 2007, 6, 652.
9. J. D. Fowler, M. J. Allen, V. C. Tung, Y. Yang, R. B. Kaner and B. H. Weiller, ACS Nano, 2009, 3, 301.
10. P. Gowda, T. Sakorikar, S. K. Reddy, D. B. Ferry and A. Misra, ACS Appl. Mater. Interfaces, 2014, 6, 7485.
11. Z. Sun and H. Chang, ACS Nano, 2014, 8, 4133.
12. F. Bonaccorso, Z. Sun, T. Hasan and A. C. Ferrari, Nat. Photonics, 2010, 4, 611.
13. M. D. Stoller, S. Park, Y. Zhu, J. An and R. S. Ruoff, Nano Lett., 2008, 8, 3498.
14. Y. He, N. Zhang, F. Wu, F. Xu, Y. Liu and J. Gao, Mater. Res. Bull., 2013, 48, 3553.
15. F. Yavari, Z. Chen, A. V. Thomas, W. Ren, H.-M. Cheng and N. Koratkar, Sci. Rep., 2011, 1, 166.
16. P. D. Bradford, X. Wang, H. Zhao and Y. T. Zhu, Carbon, 2011, 49, 2834.
17. S. K. Reddy, A. Suri and A. Misra, Appl. Phys. Lett., 2013, 102, 241919.
18. P. Jagtap, P. Gowda, B. Das and P. Kumar, Carbon, 2013, 60, 169.
19. P. Gowda, P. Kumar, R. Tripathi and A. Misra, Carbon, 2014, 67, 546.
20. W. Guo and Y. Guo, Phys. Rev. Lett., 2003, 91, 115501.
21. A. Misra, J. R. Greer and C. Daraio, Adv. Mater., 2009, 21, 334.
22. A. Cao, P. L. Dickrell, W. G. Sawyer, M. N. Ghasemi-Nejhad and P. M. Ajayan, Science, 2005, 310, 1307.
23. L. J. Gibson and M. F. Ashby, Cellular solids: Structure and properties, Cambridge University Press, Cambridge, 2nd edn, 1997.
24. Z. Chen, W. Ren, L. Gao, B. Liu, S. Pei and H. M. Cheng, Nat. Mater., 2011, 10, 424.
25. Y. Zhao, J. Liu, Y. Hu, H. Cheng, C. Hu, C. Jiang, L. Jiang, A. Cao and L. Qu, Adv. Mater., 2013, 25, 591.
26. C. Wu, X. Huang, X. Wu, R. Qian and P. Jiang, Adv. Mater., 2013, 25, 5658.
27. S. J. Chae, F. Gunes, K. K. Kim, E. S. Kim, G. H. Han, S. M. Kim, H. J. Shin, S. M. Yoon, J. Y. Choi, M. H. Park, C. W. Yang, D. Pribat and Y. H. Lee, Adv. Mater., 2009, 21, 2328.
28. L. Ci, J. Suhr, V. Pushparaj, X. Zhang and P. M. Ajayan, Nano Lett., 2008, 8, 2762.
29. D. A. C. Brownson, L. C. S. Figueiredo-Filho, X. Ji, M. Gomez-Mingot, J. Iniesta, O. Fatibello-Filho, D. K. Kampouris and C. E. Banks, J. Mater. Chem. A, 2013, 1, 5692.
30. D. Graf, F. Molitor, K. Ensslin, C. Stampfer, A. Jungen, C. Hierold and L. Wirtz, Nano Lett., 2007, 7, 238.
31. L. Chen, D. Rende, L. S. Schadler and R. Ozisik, J. Mater. Chem. A, 2013, 1, 3837.
32. I. Szleifer and R. Yerushalmi-Rozen, Polymer, 2005, 46, 7803.

---