Liquid crystal functionalization of graphene nanoplatelets for improved thermal and mechanical properties of silicone resin composites

Abstract

A liquid crystalline molecule, polyurethane-imide (PUI), was used to functionalize graphene nanoplatelets (GNS) via covalent bond and π–π interactions. The PUI functionalized graphene nanoplatelets (PUI–GNS) were characterized by fluorescence spectroscopy, thermal gravimetric analysis (TGA), Fourier transform infrared spectroscopy (FTIR), X-ray diffraction (XRD) and Raman spectroscopy, and then mixed with silicone resin as fillers to fabricate silicon resin nanocomposites. The drastic quenching of the PUI fluorescence elucidated that the biphenyl anchoring unit of liquid crystalline PUI was strongly interacted with the surface of graphene sheets via π–π interactions. FTIR and Raman spectroscopy proved the existence of a covalent interaction between the PUI and GNS. The mechanical properties testing indicated that the tensile strength of silicon resin nanocomposites increased by 521% over that of a neat silicon resin when the mass fraction of PUI–GNS was 1.0%, and the elastic modulus of the silicon resin nanocomposite increased by 902% over that of the neat silicon resin if it came up to 2.0%. The thermal conductivity of the resin filled with the PUI–GNS was improved to be 1.3822 W (m K)⁻¹ at a mass fraction of 10.0%, which was enhanced more than 16.5 times over that of the neat silicon resin. The resulting thermally conductive and mechanically applicable silicon resin nanocomposites could be significant in a wide variety of electronic packaging applications.

---

1 Introduction

Graphene nanoplatelets (GNS) are a new generation of carbon-based nanoparticles that feature remarkable mechanical, electrical and thermal properties,^1–4 along with lower production costs than carbon nanotubes (CNTs).⁵ These characteristics have prompted numerous researchers to focus their efforts on investigating the enhancement of several physical properties of polymers reinforced with GNS.^6–10

Silicon resin has been extensively employed because of its potential application in electronic packaging.^11–13 However, its poor mechanical strength and thermal conductivity limit the range of applications. Nanoscale reinforcement is considered to improve the mechanical and thermal performance of silicon resin effectively.^14–16 Similar to other types of carbon nanoparticles, the main challenges in the fabrication of GNS-based nanocomposites are achieving a homogeneous dispersion of GNS in silicon resin without damaging its structure and improving the quality of the interface between the GNS and matrix. To address the former issue, the authors previously proposed an optimum and cost-effective dispersion method for GNS/polymer nanocomposites by taking into account the structural differences between GNS and CNTs.^17–20

In addition to the dispersion method, the functionalization of carbon nanoparticles has been established as an efficient means to achieve superior dispersion of nanoparticles in polymers.^21–26 For instance, Reed²⁷ recently reviewed the interface properties of functionalized CNTs and their matrix, showing that the main reason for selecting a functional group is to match the surface energy of the filler materials to the polymer matrix. Santosh Kumar Yadav et al.²⁸ researched a simple and convenient route to efficiently functionalize nanoscale nanostructure material on the surface of CNTs by entailing click coupling between azide moiety-functionalized polyhedral oligomeric silsesquioxane (POSS) and alkyne-functionalized multi-walled CNTs. Youdi Kuang et al.²⁹ investigated the effects of covalent functionalization on the thermal transport in carbon nanotube (CNT)/polymer composites, indicated that the cross-links between the polymer and CNTs can further remarkably suppress thermal conductivities of embedded functionalized CNTs by up to 50% compared with free-standing functionalized CNTs. While several attempts have been made toward the functionalization of CNTs, the literature suffers from a lack of a suitable functionalization technique for GNS. Muchun Liu et al.³⁰ reported a way to functionalize graphene nanosheets with aniline groups on their surfaces, and they attained functionalized graphene nanosheets by diazonium treatment. Ahmadi et al.¹⁷ reviewed a new strategy for functionalizing graphene nanoplatelets (GNS) by bonding a silane agent to its structure, and the mechanical properties such as elastic modulus and fracture toughness of epoxy resin nanocomposites fabricated functionalized GNS were enhanced greatly. To the best of the authors' knowledge, functionalization of graphene sheets through π–π interactions or/and covalent bond by liquid crystalline molecule have rarely been explored.

In this research, a liquid crystal molecule, polyurethane-imide (PUI), was synthesized according to our previous work³⁰ and used for the functionalization of GNS. Silicon resin specimens reinforced with PUI–GNS were prepared at different weight contents of nanoparticles along with graphene oxide (GO), and unfunctionalized GNS. PUI–GNS was characterized by fluorescence spectroscopy, thermal gravimetric analysis (TGA), Fourier transform infrared spectroscopy (FTIR), X-ray diffraction (XRD) and Raman spectroscopy. The thermal and mechanical performances of the nanocomposites were investigated using thermal conductivity meter and universal testing machine.

2 Materials and methods

2.1 Materials

All chemical reagents used in this study were of analytical laboratory grade. Graphite powder (325 mesh, carbon content 99.95 wt%), 2,4-tolylene diisocyanate, polyethylene glycol-600, 1-methyl-2-pyrrolidinone, pyromellitic dianhydride were obtained from Aladdin Chemical Reagent Co. Ltd. (Shanghai, China). Sulphuric acid (H₂SO₄), phosphoric acid (H₃PO₄), potassium permanganate (KMnO₄), hydrogen peroxide (H₂O₂) and oleic acid were procured from Tianjin Da Chemical Reagent Co. Ltd. (Tianjin, China) and used as received without further purification. Hydrogen containing silicone fluid and vinyl silicone resin prepolymer were self-prepared. All other chemicals were used as received commercially.

2.2 Synthesis of GO and GNS

GO was prepared by modifying the method proposed by Hummers and Offeman. 10 g of expanded graphite was added into 230 mL of cooled sulfuric acid. 30 g of KMnO₄ and 5 g of NaNO₃ were added gradually with stirring and cooling so that the temperature of the mixture was maintained at 10–15 °C. The mixture was then heated and stirred at 35 °C for 30 min. The temperature was raised to 90 °C after 250 mL of deionized water was slowly added. The mixture was stirred at 90 °C for 30 min. To stop the oxidation reaction, an additional 500 mL of deionized water and 50 mL of 30% hydrogen peroxide solution (H₂O₂), were added sequentially to reduce the excess KMnO₄. The obtained sample was filtered and washed with 100 mL of deionized water. The filtrate was resuspended in deionized water again, and followed by ultrasonic treatment for 5 min. The solid product was separated by centrifugation and then dried in a vacuum oven at 60 °C for 24 h and the product is GO.

Typically 10 mL of a 0.1 wt% GO aqueous solution was placed in a boiling flask of 50 mL. 10 μL of a 35 wt% aqueous solution of N₂H₄ (hydrazine) and 70 mL of a 28 wt% of an aqueous solution of NH₄OH (ammonia) were added to the mixture. The solution was homogenized and heated to 90 °C under stirring and refluxing during 1 h. The pH measured after the reaction was 11. The product was obtained after centrifugation and washing with distilled water and then dried in a vacuum oven at 60 °C for 24 h. The final product is GNS.

2.3 Liquid crystal functionalization of GNS

0.03 g of PUI was dissolved in 10 mL ethyl alcohol in a 50 mL flask, followed by the addition of 0.3 g graphene and stirred evenly. The resulting mixture was sonicated in ultrasonic bath for 30 min, and then filtered and washed with distilled water. The final products were dried in vacuum at 80 °C for 24 h.

2.4 Preparation of silicon resin nanocomposites

10 g of vinyl silicone resin prepolymer and a certain quality of PUI–GNS were mixed and sonicated for 30 min. Then, 4.5 g of hydrogen containing silicone as cross-linking agent, 0.5 g of platinum catalyst and 1-vinyl cyclohexanol as inhibitor were added, followed by vigorous stirring and then let stand for 30 min. The mixture was poured into the Teflon mold. The mold was kept in a vacuum oven and the mixture was vacuum degassed at 60 °C for 2 h. Then it was cured at 80 °C for 1 h, 100 °C for 1 h, and 150 °C for 2 h. The end product was PUI functionalized graphene/silicon resin nanocomposite (silicon resin/PUI–GNS). The silicon resin/GO nanocomposite and silicon resin/GNS nanocomposite were also prepared using the same procedure.

2.5 Characterization

Fourier transform infrared spectroscopy (FTIR) (8300PCS, Shimadzu, Japan), thermogravimetric analyzer (TGA) (STA409PC, NETZSCH, Germany) and fluorescence spectrometer (F7000, Hitachi, Japan), Raman spectroscopy (Rainie Salt Public Co. Ltd., Britain) and X-ray diffraction (XRD) were employed to characterize covalent and noncovalent interaction between PUI and GNS. The infrared spectra were evaluated in the range of 400–4000 cm⁻¹. The weight loss of solid samples in TGA analyzer was measured at heating rate of 10 °C min⁻¹ under nitrogen atmosphere. Samples characterized using X-ray diffraction with Cu Kα radiation (λ = 1.5418 Å) at a step size of 2° per second. The tensile properties of silicon resin nanocomposites were measured on an mechanical tester (CMT4304, SANS, China) under a 500 N load cell with a crosshead speed of 2 mm min⁻¹. The thermal conductivities of silicon resin nanocomposites were examined by using thermal conductivity meter (TC3000, Xiatech, China) at 25 °C.

3 Results and discussion

3.1 Fluorescence analysis

When light of sufficient energy is incident on a material, photons are absorbed and photoexcited electrons and holes are generated. Eventually, these excitations relax and the electrons return to the ground state. If radiative relaxation occurs, the emitted light is called photoluminescence (PL) emission.^31–34 Fig. 1 shows the PL emission spectra of PUI and PUI–GNS. The spectra of PUI exhibited a broad emission band in the range of 475–600 nm, which is ascribed to luminescence from localized surface states due to recombination of photogenerated electron–hole pairs. However, this emission was extensively quenched in PUI functionalized graphene sheets complexes, the quenching arises from efficient energy transfer between the two materials due to π–π interactions, rather than the disruption of π-conjugation caused by a conformational change. The drastic quenching of the PUI fluorescence elucidated that the biphenyl anchoring unit of liquid crystalline PUI was strongly interacted with the surface of graphene sheets via π–π interactions.

Fig. 1 Room-temperature PL emission spectra of PUI and PUI–GNS with excitation wavelength of 325 nm.

3.2 FTIR analysis

Results of FTIR experiments are shown in Fig. 2. The spectrum of GO (Fig. 2c) confirms the presence of C–O (ν_C–O at 1040 cm⁻¹), O–C–O (ν_O–C–O at 1260 cm⁻¹ and 804 cm⁻¹), carboxyl C=O (ν_C=O at 1730 cm⁻¹) and ketene C=O (ν_C=O at 1620 cm⁻¹), the peak at 3430 cm⁻¹ is corresponding to the O–H stretching vibrations of the C–OH groups and the water molecules intercalated into GO.³⁵ After reduction, the peak at 3440 cm⁻¹ is severely attenuated in the GNS spectrum, and strong absorptions at 1730 cm⁻¹, 1260 cm⁻¹, 1040 cm⁻¹ and 804 cm⁻¹ disappear, which indicate that most carboxyl and ketene groups were removed. The peaks emerged at 1630 cm⁻¹ is assigned to the skeleton vibration and plane bending vibration of graphitic domains. Although it is difficult to distinguish the characteristic peak of O–H (ν_O–H at about 3400 cm⁻¹) from the peak of water molecule (ν_O–H at 3430 cm⁻¹), the IR spectrum of PUI–GNS was still distinguished from that of GNS as evidenced by the stretching vibrations of acrylamide (1730 cm⁻¹ and 1535 cm⁻¹) and of PUI, and the presence of a series of peaks at 2900 cm⁻¹ and 2870 cm⁻¹, which attributed to the stretching vibrations of C–H of methyl group and methylene of PUI. The appearance of the peak at 1750 cm⁻¹ is ascribe to ester carbonyl of PUI–GNS, which proves the existence of chemical bond between the PUI and residual oxygen-containing groups of GNS.

Fig. 2 FT-IR spectra of PUI–GNS, GNS and GO.

3.3 Raman analysis

Raman spectroscopy is a non-destructive, powerful tool to distinguish between ordered and disordered crystal structures of carbon. As shown in Fig. 3, the Raman spectra show two prominent peaks in all samples, corresponding to the well-documented D (∼1340 cm⁻¹) and G (∼1590 cm⁻¹) bands. The G band is arising from the in phase vibration of the graphite lattice, and the D band is arising from the weak disorder band caused by the graphite edges.³⁶ The calculated I_D/I_G intensity ratios of GO, GNS and PUI–GNS are 0.86, 1.05, 1.12, respectively. The I_D/I_G of GNS (Fig. 3b) is higher than that of GO (Fig. 3c), which can be explained by the phenomenon assuming that the reduced state increases the number of aromatic domains of smaller size of GNS after the removal of most oxygen groups.³⁷ After PUI functionalization as described here, the increase in the I_D/I_G of PUI–GNS can be partly attributed to the formation of covalent bonds between hydroxy and surfaces of graphene, which further increases the number of aromatic domains of smaller size of GNS.

Fig. 3 Raman spectra of PUI–GNS, GNS and GO.

3.4 Thermo-gravimetric analysis (TGA)

The thermal behaviors of samples were recorded by thermal gravimetric analysis (TGA) in the nitrogen atmosphere (Fig. 4). In Fig. 4, the overall weight loss of GO, GNS, PUI, and PUI–GNS are 93 wt%, 29 wt%, 99 wt% and 59 wt%, respectively. GNS was found to have a major weight loss of about 10 wt% below 100 °C, which was attributed to the loss of absorbed water. And with weight loss of 19 wt% in the temperature ranges from 100 to 1000 °C due to the decomposition of some residual oxygen-containing groups, which is much smaller than that of GO.^38,39 However, a steady mass loss is also retained, suggesting that oxygen functionalities cannot be removed totally in the presence of strong reductants (such as hydrazine monohydrate). For the liquid crystalline PUI, the major weight loss started at about 220 °C, and ended at 400 °C with weight loss of 99 wt%. TGA data for the PUI functionalized graphene sheets indicated a three-step decomposition process. The first and the third step are consistent with the thermal evaporation of GNS, corresponding to the degradation of absorbed water and decomposition of residual oxygen-containing groups. The second decomposition step corresponds to the decomposition of attached PUI. In addition, the weight loss and residual weight obtained from PUI–GNS were more than that obtained from the unmodified GNS. These results also suggest that graphene sheets noncovalent functionalized with liquid crystalline PUI were successfully achieved.

Fig. 4 TGA curves of GNS, PUI–GNS, PUI and GO.

3.5 XRD analysis

Fig. 5 shows XRD patterns of GO, GNS and PUI–GNS. The XRD peak (001) of GO was observed at 2 = 11.15°, indicating an interlayer distance of 0.79 nm. This result indicates that the interlayer spacing of GO layers became much larger than that of pristine graphite because of many oxygen-containing functional groups on the graphite sheets. Following the conversion of GO to graphene, the peak (002) shows a much broader and weaker characteristic diffraction feature, owing to a loss of graphitic crystallinity. The XRD of GNS data clearly confirms the transition from millions of graphitic layers of natural graphite to a few graphitic layers graphene and the reduction of the GO sheets, which is in accordance with the above Raman data. The peak (001) of PUI–GNS shift from 11.15° to 10.51° and the corresponding interlayer distance of GNS change to 0.86 nm, indicating that the ordered layer structure of GNS remains after reduction and functionalization.

Fig. 5 XRD patterns of GO, GNS, and PUI–GNS.

3.6 Mechanical properties

Following the successful functionalization of GNS, the effect of PUI–GNS along with GO and GNS on the mechanical properties of the silicon resin were experimentally investigated. The elastic modulus and the tensile strength of the nanocomposites are depicted in Fig. 6. The results reveal that, in general, the elastic modulus of the silicon resin have been enhanced by all the above samples (Fig. 6a). In general, the nanocomposites reinforced with each content of PUI–GNS provided a higher elastic modulus than that obtained using GNS or GO. Moreover, the highest value for the elastic modulus of nanocomposites was obtained at 2.0 wt% for the PUI–GNS nanocomposite. These results clearly demonstrate that the functionalization of GNS impacts the stiffness of nanocomposites significantly. However, the elastic modulus of nanocomposites at the lowest GNS content (i.e., 0.5 wt%) did not vary noticeably compared to that of the neat silicon resin. By increasing the nanoparticle contents from 0.5 wt% to 1.0 wt%, a growth of 171% in the elastic modulus was observed for nanocomposites hosting the surface functionalized GNS. Furthermore, if continuing to increase the content of PUI–GNS to 2.0 wt%, the elastic modulus of silicon resin nanocomposite increased by 902% over neat silicon resin and only 322% increase obtained in nanocomposite hosting the unfunctionalized GNS.

Fig. 6 Mechanical properties of neat silicon resin and various types of nanocomposites (a) elastic modulus (b) tensile strength.

The variation tendency of tensile strength is showed in Fig. 6b. It could be seen that the nanocomposites reinforced with each content of PUI–GNS provided a higher tensile strength than that obtained using GNS or GO. The tensile strength of nanocomposites filled with 1.0 wt% GO, GNS and PUI–GNS nanoparticles were increased by 235%, 228% and 471% over that of neat silicone resin respectively. However, if the nanoparticle contents were increased to 2.0%, the tensile strength of silicon resin nanocomposite declined. It indicated that excessive contents of nanoparticle have influence on the curing process of silicon resin for their aggregation.

3.7 Thermal conductivity analysis

Fig. 7 shows the thermal conductivity of the samples with three different nanoparticles (GO, GNS and PUI–GNS) as a function of the mass fraction. It can be seen obviously that the thermal conductivity of silicon resin filled with these three nanoparticles increase with the content of the nanoparticles. The value of thermal conductivity of neat silicon resin is measured to be 0.0782 W (m K)⁻¹ at room temperature. In general, the functionalized nanocomposites reinforced with each mass fraction of PUI–GNS provided a higher thermal conductivity than that obtained using GNS or GO. The thermal conductivity of the resin filled with the PUI–GNS is improved to be 1.3822 W (m K)⁻¹ at the mass fraction of 10.0%, which is enhanced more than 16.5 times than that of the pure silicon resin. If continuing to increase the mass fraction of nanoparticles, the excessive contents of nanoparticles will have great influence on the curing process of silicon resin and can not be moulded.

Fig. 7 Thermal conductivities of silicon resin nanocomposites filled with PUI–GNS, GNS and GO with different mass fraction.

These results clearly demonstrate that the functionalization of GNS significantly improves the thermal conductivity of nanocomposites. Generally, such increase of thermal conductivity can be ascribed to phononic heat transport of nanoparticles. Owing to structural damage during acid oxidation, the ability of phononic heat transport of GO decreases tremendously, so the thermal conductivity of GO/silicon resin do not vary noticeably compared to that of the neat silicon resin. GNS is prone to agglomerate in polymer matrix and this defect hinder the formation of thermal conductive network, and the increase of thermal conductivity is not enough. After functionalization with PUI, the dispersion degree of GNS in silicon resin and the interfacial interactions between the GNS surfaces and the silicon resin are improved significantly, so the phonon scattering induced boundaries of silicon resin and GNS decrease dramatically.

4 Conclusions

A novel liquid-crystalline molecule (PUI) was synthesized, characterized and used to functionalize GNS, after which the mechanical and thermal properties of silicon resin reinforced with GO, GNS and functionalized GNS were experimentally assessed. The results manifested that graphene sheets functionalized with liquid crystalline PUI noncovalently (π–π interactions) and covalently were successfully achieved.

The results of mechanical and thermal measurements indicated that silicon resin filled with PUI–GNS gained great increases in both thermal conductivity and mechanical performance compared with GO/silicon resin, GNS/silicon resin and neat silicon resin. When the mass fraction of PUI–GNS was 1.0%, the tensile strength of silicon resin nanocomposites increased by 471% over neat silicon resin, and the elastic modulus of silicon resin nanocomposite increased by 902% over that of neat silicon resin if it came up to 2.0%. The thermal conductivity of the resin filled with the PUI–GNS was improved to be 1.3822 W (m K)⁻¹ at the mass fraction of 10.0%, which was enhanced more than 16.5 times over that of neat silicon resin.

Acknowledgements

Support from the National Basic Research Program of China (Program 973) (No. 2011CB605603), the Basic Research Project of Shenzhen (No. JCYJ20140418091413509) is greatly acknowledged.

References

1. K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y. Zhang and S. V. Dubonos, Science, 2004, 306, 2345.
2. E. Murray, S. Sayyar, B. C. Thompson, R. Gorkin III, D. L. Officer and G. G. Wallace, RSC Adv., 2015, 5, 45284–45290.
3. G. X. Zhao, T. Wen, C. L. Chen and X. K. Wang, RSC Adv., 2012, 2, 9286–9303.
4. H. Porwal, P. Tatarko, S. Grasso, J. Khaliq, I. Dlouhy and M. Reece, Carbon, 2013, 64, 359.
5. C. D. Scott, S. Arepalli, P. Nikolaev and R. E. Smalley, Appl. Phys. A, 2001, 72, 573–580.
6. A. Pandikumar, G. T. S. How, T. P. See, F. S. Omar, S. Jayabal, K. Z. Kamali, N. Yusoff, A. Jamil, R. Ramaraj, S. A. John, H. N. Limbe and N. M. Huang, RSC Adv., 2014, 4, 63296–63323.
7. C. Lee, X. Wei, W. J. Kysar and J. Hone, Science, 2008, 321, 385–388.
8. W. K. Chee, H. N. Lim, N. M. Huang and I. Harrison, RSC Adv., 2015, 5, 68014–68051.
9. P. F. Wei and S. B. Bai, RSC Adv., 2015, 5, 93697–93705.
10. H. Fan, L. Wang, K. Zhao, N. Li, Z. Shi, Z. Ge and Z. Jin, Biomacromolecules, 2010, 11, 2345.
11. C. P. Wong, Silicone resin electronic device encapsulant[P], US Pat., US5051275 A, 1991–09–24.
12. N. Gao, W. Liu, Z. Yan and Z. Wang, Opt. Mater., 2013, 35, 567.
13. A. W. Norris, M. Bahadur and M. Yoshitake, Proc. SPIE, 2005, 5941, 594115.
14. Y. Liu, W. Chi, H. Duan, H. Zou, D. Yue and L. Zhang, J. Alloys Compd., 2015, 657, 472–477.
15. D. Chen, F. Chen, X. Hu, H. Zhang, X. Yin and Y. Zhou, Compos. Sci. Technol., 2015, 117, 307.
16. Y. Song, J. Yu, L. Yu, F. E. Alama, W. Dai and C. Li, Mater. Des., 2015, 88, 950.
17. B. M. Ahmadi and F. Taheri, J. Mater. Sci., 2014, 49, 6180–6190.
18. H. Suhermana, A. B. Sulong and J. Sahari, Ceram. Int., 2013, 39, 1277–1284.
19. T. Ramanathan, A. A. Abdala, S. Stankovich, D. A. Dikin, M. A. Herrera and R. D. Piner, Nat. Nanotechnol., 2008, 3, 327–331.
20. L. Meng, C. Fu and Q. Lu, Prog. Nat. Sci., 2009, 19, 801–810.
21. M. Słoma, M. Jakubowska, A. Kolek, K. Mleczkoet, P. Ptakal and A. W. Stadler, J. Mater. Sci.: Mater. Electron., 2011, 22, 1321.
22. J. Chen, Q. Chen and Q. Ma, J. Colloid Interface Sci., 2012, 370, 32–38.
23. L. Kumari, T. Zhang, G. H. Du, W. Z. Li, Q. W. Wang, A. Datye and K. H. Wu, Ceram. Int., 2009, 35, 1775–1781.
24. M. Deborah, A. Jawahar, T. Mathavan, M. K. Dhas and A. M. Benial, Spectrochim. Acta, Part A, 2014, 139, 138–144.
25. X. Shi, B. Jiang, J. Wang and Y. Yang, Carbon, 2012, 50, 1005–1013.
26. H. Suhermana, A. B. Sulong and J. Sahari, Ceram. Int., 2013, 39, 1277–1284.
27. C. W. Reed, Dielectric polymer nanocomposites, Springer, US, 2010, vol. 4, pp. 95–131.
28. S. K. Yadav, S. S. Mahapatra, J. W. Cho and J. Y. Lee, J. Phys. Chem. C, 2010, 114, 11395–11400.
29. Y. Kuang and B. Huang, Polymer, 2015, 56, 563.
30. D. Y. Gao, X. Gao, J. F. Hao and J. H. Liu, Polym. Eng. Sci., 2014, 54, 1704–1711.
31. M. Ahmad, E. Ahmed, Z. L. Hong, W. Ahmed, A. Elhissi and N. R. Khalid, Ultrason. Sonochem., 2014, 21, 761–773.
32. C. Li and G. Shi, J. Photochem. Photobiol., C, 2014, 19, 20.
33. F. M. Winnik, Chem. Rev., 1993, 93, 587–614.
34. J. Chen, H. Y. Liu, W. A. Weimer, M. D. Halls, D. H. Waldeck and G. C. Walker, J. Am. Chem. Soc., 2002, 124, 9034–9035.
35. W. Gao, L. B. Alemany, L. J. Ci and P. M. Ajayan, Nat. Chem., 2009, 1, 403–408.
36. K. N. Kudin, B. Ozbas, H. C. Schniepp, R. K. Prud'homme, I. A. Aksay and R. Car, Nano Lett., 2008, 8, 36–41.
37. S. Stankovich, D. A. Dikin, R. D. Piner, K. A. Kohlhaas, A. Kleinhammes and Y. Y. Jia, Carbon, 2007, 45, 1558–1565.
38. H. Chen, M. B. Muller, K. J. Gilmore, G. G. Wallace and D. Li, Adv. Mater., 2008, 20, 3557–3561.
39. L. J. Ma, X. M. Yang, C. X. Guo, Y. W. Li, Y. F. Tu and X. L. Zhu, Carbon, 2013, 53, 269–276.

---