Functionalization of AgNWs with amino groups and their application in an epoxy matrix for antistatic and thermally conductive nanocomposites

Abstract

Highly dispersive silver nanowires (AgNWs) in an epoxy matrix were obtained by introducing amino groups onto the surface of AgNWs. Firstly, modification of thioglycolic acid onto the surface of the AgNWs was conducted via coordination bonds between the AgNWs and thiol groups (AgNWs@COOH). Then amido-functionalized AgNWs (AgNWs@NH₂) were prepared through an amidation reaction between triethylenetetramine (TETA) and the carboxyl groups. The chemical structure and properties of the AgNW–epoxy nanocomposites were investigated using TEM, EDS, FT-IR, UV, TAG, DSC and SEM. The results revealed that because the amino groups participated in the epoxy curing reaction, chemical bonds were formed at the interface to improve the dispersion of the AgNWs@NH₂. Thus, the antistatic and thermal conductivity performances were enhanced by a well-constructed electrostatic discharge channel in the epoxy matrix. The surface resistivity of the AgNWs@NH₂–epoxy nanocomposite was decreased to 1.24 × 10⁵ Ω at 0.5 Vt%, and the thermal conductivity coefficient was upgraded to 0.67 W m⁻¹ K⁻¹ at 10 Vt%.

---

1. Introduction

On entering the 21st century, precious metal nanomaterials presented diverse potential applications, especially in the microelectronics industry, optical materials and biological sensors.^1,2 Particularly, Ag@polymer nancomposites were of great interest because Ag nanoparticles were good candidates for optics, electronics, and catalysis.^3–5 With their high electrical and thermal conductivity, silver nanowires (AgNWs) were good options for the preparation of conductive materials.⁶ However, AgNWs agglomerated easily because of the high surface energy, making their application in polymer matrix nancomposites greatly confined. Therefore, surface modification of AgNWs seemed necessary.⁷ Reduction of the system surface energy by surface modification could improve the dispersion and compatibility of nanowires with other materials and create conditions for macromolecular grafting. At present, the main methods of surface modification for precious metals and their oxides have been micellar, ligand exchange and cyclodextrin methods.^8–10 The most efficient method was chemical functionalization,^11–15 which could provide functional groups on the AgNWs and then form covalent bonds with the matrix, making AgNWs a high performance filler for polymer matrix composites.

Previously, antistatic polymer composites were not available without blending a large number of electrically conductive materials into the polymer matrix. The polymer itself did not have antistatic properties, and its application field would expand tremendously once able to impart antistatic properties. Among the polymer composites, high strength epoxy systems are very important materials. It made sense to prepare high performance epoxy composites with high strength, light mass and multi-functional features. However, because of its unique insulation, epoxy resin brought negative effects to the materials, such as electrostatic problems. In modern industrial production and daily life, the accumulation of static charge tends to cause casualties and damage to equipment. Therefore, much research has been done to explore how to build up a discharge channel for avoiding accumulation of static electricity in insulated polymers.^16,17 The most common and effective method was to use an antistatic filler to reduce the surface resistivity of the polymer. However, the traditional antistatic effect couldn’t be maintained for a long time as the organic antistatic agents migrated and were lost.^18,19 Thus new antistatic fillers have been reported gradually, such as metallic stuffing and carbon materials. Kozlowski prepared electrically conductive polymer composites by filling carbon black into HDPE, PS, PP, HDPE/PS and HDPE/PP. The obtained composites achieved the lowest percolation threshold at 2 wt% which resulted in an electrical resistance of 10⁷ Ω m.²⁰ Zeng prepared antistatic multi-layer graphene filled PVC composite films, and investigated the electrical conductivity, tensile behavior and thermal properties of the composite films. The surface electrical conductivity of the composites was less than 3 × 10⁸ ohms per square meter when the graphene was about 3.5 wt%, meeting the antistatic requirement for commercial antistatic PVC films.²¹ Furthermore, Ag had the best electrical and thermal conductivity among all the metal materials.

When using a filled epoxy resin, the antistatic and thermal conductivity of the epoxy composites mainly depended on the metal fillers because there were a large number of free electrons in the metal fillers and their electrical/thermal conductivity were much better than that of the resin matrix.^22,23 A one-dimensional AgNW fibrous structure could form a conductive network more easily in the epoxy matrix, which could greatly reduce the percolation threshold of the filler.²⁴ The slenderness ratio was one of the main factors which could affect conductivity. In a certain range, the greater the L/D ratio was, the better the electrical conductivity, because of more chances to develop a continuous conductive network. In this paper, high aspect ratio AgNWs are prepared using a liquid phase polyol method.^25–28 The conductivity mechanism of the AgNWs in the epoxy matrix could be expressed in the following diagram (Fig. 1). Parts of the conductive AgNWs are completely in continuous contact with each other to form an unbroken current path, equivalent to the current flow through a resistor R_L. Another situation is incomplete contact of the conductive AgNWs leading to a tunnel effect, so that they could not form a complete current path, corresponding to the case of a resistor R_g and a capacitor C in parallel and then in series with another resistor R_L.

Fig. 1 Conductive mechanism of the AgNWs and AgNWs@NH₂ in the epoxy matrix.

AgNWs tend to aggregate in a polymer matrix due to the specific surface area and high surface activity. To solve this problem, surface modification of the AgNWs was necessary. In this paper, amino groups are equipped onto thioglycolic acid functionalized AgNWs (AgNWs@NH₂)²⁹via amide bonds using triethylenetetramine (TETA) as a source of amino groups (Fig. 2). Meanwhile, TETA was the epoxy curing agent, so the preparation technology was simplified greatly. This method could improve the interface affinity between the AgNWs and substrate, enhance the interfacial bond strength,³⁰ and improve the antistatic and thermal conductivity properties^31–33 of the composites as well. Uniform dispersion of the AgNWs in the matrix could help establish a continuous conductive network and make the AgNW–epoxy nanocomposites reach the standard of antistatic materials. Compared with pure AgNWs, for AgNWs@NH₂ it was easier to form a continuous conductive network in the matrix. This was because for the pure AgNWs a spacer layer would appear between the conductive nanowires, which could affect the antistatic properties of the nanocomposites.

Fig. 2 Schematic representation of the surface-modified AgNWs.

2. Experimental procedures

2.1. Materials

Silver nitrate (AgNO₃, 99%) was provided by Shanghai Prolong Biochemical Co. Ltd. Chloroplatinic acid (H₂PtCl₆) and thioglycolic acid (90%) were supplied by Sinophar Chemical Reagent Co. Ltd. Acetone (>99.5%), ethylene glycol (EG, 99%) and ethanol (>99.7) were purchased from Vas Chemical of China. Polyvinylpyrrolidone (PVP, M_w = 58 000) and triethylenetetramine (TETA, 70%) were furnished by Aladdin Chemistry Co. Ltd. Epoxy resin (E-51) was provided by Changshu Jaffa Chemical Co. Ltd.

2.2. Preparation of the surface-modified silver nanowires

2.2.1 Preparation of the silver nanowires (AgNWs). Silver nitrate was reduced by glycol to give silver atoms. Then, silver particles began to grow on platinum points that had already been generated in the presence of PVP in EG. 10 mL of EG was heated to 160 °C in a three-necked flask with stirring and congealing for 1 h. Then 0.5 mL of H₂PtCl₆ (3.0 × 10⁻⁴ M, in EG) was added into the mixture at 160 °C and reacted for 10 min. After that, 5 mL of PVP solution (0.72 M, in EG) and 2.5 mL of AgNO₃ solution (0.40 M, in EG) were added dropwise over 5 min into the mixture synchronously. The reaction ended after 1 h, and the reaction flask was placed in the air and cooled to room temperature. Then, the product was centrifuged at 6500 rpm for 30 min using in order acetone, deionized water and ethanol as the diluting solvent, and this cycle was repeated at least 3 times. Finally, the product was kept in ethanol for future use.

2.2.2 Preparation of the thioglycolic acid functionalized silver nanowires (AgNWs@COOH). The pure AgNWs (80 mg) were dispersed into ethanol (10 mL), then a certain amount of thioglycolic acid was mixed into the solution with stirring at room temperature for 24 h under a nitrogen atmosphere. Since thioglycolic acid could easily decompose in light, the reaction required the condition of avoiding light. In the process, thiol groups were adsorbed on the surface of the silver nanowires through effective coordination bonds, which contributed to the successful introduction of carboxyl groups. At last, the sample was washed in a high-speed centrifuge with deionized water and ethyl alcohol alternately, and collected in ethyl alcohol solution for the next modification.

2.2.3 Preparation of the TETA grafted AgNWs@COOH (AgNWs@NH₂). Owing to the highly active amino groups, an amidation reaction can easily take place between carboxyl and amino groups. The AgNWs@COOH solution (20 mL) prepared in the previous step was disposed uniformly using ultrasonic dispersion. In the next step, TETA (3.5 mL) was added into the above solution and reacted at room temperature for 24 h under a nitrogen atmosphere, and the primary amine group ligand was obtained on the surface of AgNWs@COOH. The sample, termed AgNWs@NH₂, was purified using centrifugation.

2.3. Preparation of the AgNW–epoxy nanocomposite

Surface-modified AgNWs (AgNWs@NH₂) (80 mg) and epoxy resin (E-51) (1 g) were mixed in a mixed solvent of ethanol and acetone (ethanol/acetone 1 : 1 in volume), then pre-reacted at room temperature for 20 min with stirring. Next, the mixture was heated to 50 °C, and the solvent rapidly removed using a vacuum system. The above steps had to be operated 3–4 times to ensure that all solvents had been removed completely. Then, half the amount of curing agent triethylenetetramine (0.5 g) was added with stirring. The mixture was out-gassed in a vacuum at 50 °C for 5 min, then cast into a PET mould and cured for 6 h at room temperature.

2.4. Measurements

A Fourier transform infrared spectrometer (FT-IR, Nicolet-Nexus 670) was used to characterize the surface-modified AgNWs, and the samples were measured as pellets with KBr. Thermo-gravimetric analysis (TGA, Netzsch-TG 209c) was performed at a heating rate of 10 °C min⁻¹ from 25 to 700 °C under nitrogen. High-resolution transmission electron microscopy (HR-TEM, JEOL JEM-3010 electron microscope) and scanning electron microscopy (SEM, Hitachi-s4700) were employed to observe the morphology of various AgNWs. Morphology images of the AgNWs in the epoxy matrix were obtained using scanning electron microscopy (SEM, Hitachi-s4700) imaging. The UV-visible extinction spectra were obtained at room temperature using a U-3010 spectrophotometer. Energy dispersive spectrometry (EDS) was undertaken using a Hitachi-S4700 scanning electron microscope. Differential scanning calorimetry (DSC) was performed using a Perkin-Elmer Pyris I and conducted at a heating rate of 10 °C min⁻¹ ranging from 30 to 180 °C. Electrical conductivity was measured with a two-probe method using a Megger instrument (ZC-90, Yuanzhong Electron) in Megger Test mode. The measured samples were clipped to 10 × 30 × 1 mm, and silver conductive adhesive was coated onto two ends of the sample in order to eliminate surface contact potential. The thermal conductivity measuring instrument (TC 3000E, Xi’an Xiatech Eletron) was equipped to measure the thermal conductivity.

3. Results and discussion

3.1. The characterization of the surface-modified silver nanowires

The morphology of the series of silver nanowires was observed using SEM and HR-TEM, including the pure AgNWs (Fig. 3a and d), thioglycolic acid treated AgNWs (AgNWs@COOH) (Fig. 3b and e) and amino group functionalized AgNWs (AgNWs@NH₂) (Fig. 3c and f). In the reduction of silver nitrate to silver wire using Pt nanoparticles as seeds, PVP as a kind of passivation agent was employed to inhibit the vertical growth of the lattice plane through Ag–O chemical bonds. Finally, marvelous AgNWs were obtained with the properties of a high aspect ratio and uniform size. The one-dimensional structure of the AgNWs could be found clearly in the SEM images. Moreover, the UV-visible extinction spectrum of the pure AgNWs is shown in Fig. 4a. The existence of two SPR peaks at 350 and 385 nm illustrates that AgNWs were generated. Using HR-TEM (Fig. 3e), it could be clearly observed that the surface of the AgNWs were effectively coated with a thin coating layer. On comparison with the pure AgNWs (Fig. 3d), a satisfactory conclusion was reached that thioglycolic acid had been successfully modified onto the surface of the AgNWs. The same fact could be proven by the UV-visible extinction spectrum (Fig. 4b) showing that the resonance absorption peaks red shifted to 385 and 425 nm.^34,35 When charge transfer occurs between the silver metal and adsorbed thioglycolic acid molecules, the metal electron cloud density would change which would lead to the red shift of the resonance absorption peaks.^36,37 The direction of the mobile charge was from Ag to the thiol group. Furthermore, there was also a small red shift of the absorption peaks for AgNWs@NH₂ (Fig. 4c). This was due to an increase of the size of the AgNWs after being coated with TETA, and it has long been known that absorbance shifts to a longer wavelength are related to a large particle size and/or changes in the particle size morphology.^38,39 The average diameter of the AgNWs had increased by 30 nm. In the two images Fig. 3c and f, the HR-TEM image of AgNWs@NH₂ shows an obvious thicker coating layer wrapped over the surface of the AgNWs with its diameter significantly increased, and this phenomenon was also shown in the SEM image. The results support that the amino groups have been introduced onto AgNWs@COOH successfully, as expected.

Fig. 3 SEM images of (a) the pure AgNWs, (b) AgNWs@COOH, and (c) AgNWs@NH₂; HR-TEM images of (d) the pure AgNWs, (e) AgNWs@COOH and (f) AgNWs@NH₂.

Fig. 4 UV-visible extinction spectra of (a) the pure AgNWs, (b) AgNWs@COOH and (c) AgNWs@NH₂.

The chemical structures of the pure AgNWs, the thioglycolic acid functionalized AgNWs@COOH and the amino group grafted AgNWs (AgNWs@NH₂) were characterized using FT-IR (Fig. 5), EDS (Fig. 6) and TGA (Fig. 7). In the FT-IR spectrum, the pure AgNWs (Fig. 5a) didn’t provide any obvious absorption peaks. The EDS spectrum (Fig. 6a) also shows a clean surface for the AgNWs, and this could be confirmed by the 100% silver weight fraction that is listed in the additional table. Although the absorption peaks of the thioglycolic acid which acted as a ligand were weak, there is a faint peak for the –SH stretch vibration at 2551 cm⁻¹ and a C=O stretching vibration of the carboxylic groups at 1718 cm⁻¹ in the spectrum of the thioglycolic acid functionalized AgNWs^40–42 (Fig. 5b). In addition, the EDS results could indicate successful functionalization because of the signal of the sulfur (S) element (Fig. 6b). All the evidence described above indicates that thioglycolic acid had been successfully modified onto the surface of the AgNWs through the coordinate bond Ag–S. AgNWs@NH₂ (Fig. 5c) provided characteristic stretch vibration peaks of C–N and C=O in an amide group at 1660 and 1446 cm⁻¹. A characteristic peak of the N element simultaneously appeared in the EDS results (Fig. 6c). Due to the low content of N element and muted sensitivity of the machine used for the EDS spectrum, the peak of the N element appeared to be very small. The above-mentioned data shows that a certain amount of amino groups were functionalized onto the surface of the thioglycolic acid functionalized AgNWs. In Fig. 5d, besides the vibration peaks of the amide in the spectrum, there were some characteristic peaks of the epoxy resin (E-51), which reacted with the TETA on the surface of the AgNWs. The characteristic vibration peaks of benzene were at 1607, 1509 and 1296 cm⁻¹, and the stretch vibration of the epoxy group was at 917 cm⁻¹. A conclusion could be drawn that the TETA had successfully reacted with the epoxy resin.

Fig. 5 FT-IR spectra of (a) the pure AgNWs, (b) AgNWs@COOH, (c) AgNWs@NH₂, and (d) AgNWs@NH₂–epoxy.

Fig. 6 EDS spectra from top to bottom are pure AgNWs, AgNWs@COOH and AgNWs@NH₂, respectively.

TGA curves were used to estimate the weight ratio of the grafted organic materials on the AgNWs. The pure AgNWs (Fig. 7a) could maintain a stable weight until 700 °C. After treatment with thioglycolic acid, the weight loss was increased by 5.43%, in curve b, which could be assigned to the decomposition of grafted thioglycolic acid. The weight of AgNWs@NH₂ (Fig. 7c) that remained was 90.72% (m₂) at 700 °C in Ar. There was another 3.85% weight loss, which was attributed to the decomposition of the organic material of triethylenetetramine.

Fig. 7 TGA curves of (a) the pure AgNWs, (b) AgNWs@COOH and (c) AgNWs@NH₂.

3.2. The characterization of the AgNW–epoxy nanocomposites

Fig. 8 presents the DSC results of the pure AgNW–epoxy (Fig. 8a) and surface-modified AgNWs@NH₂–epoxy (1 : 100 v/v) (Fig. 8b) nanocomposites. There is an exothermic peak at 81 °C shown in curve (b), which was attributed to the curing reaction of the epoxy and TETA grafted on the surface of the AgNWs. Because there was no curing agent added into the mixture, the exothermic peak must come from the reaction between the amino groups and epoxy. On the contrary, there was no exothermic peak in the curve (a) of the pure AgNW–epoxy (1 : 100 v/v) nanocomposite. This fact proves that no reaction happened between the pure AgNWs and epoxy without adding a curing agent.

Fig. 8 DSC curves of the AgNW–epoxy (a) and surface-modified AgNWs@NH₂–epoxy (b) nanocomposites.

The SEM micrographs of the epoxy nanocomposites with pure AgNWs and AgNWs@NH₂ are shown in Fig. 9a and b (content 1 Vt%). It could be observed that the pure AgNWs had an “integrated” structure, which is one of the crucial factors of AgNWs to reinforce nanocomposites. The image exhibits that the AgNWs were enveloped by lots of semitransparent substances. However, due to the silver nanowires being prone to self-aggregation, the high density metal nanowires could not form a uniform distribution in the epoxy matrix. It turned out that many AgNWs were deposited at the bottom of the substrate during the experiment. For the surface modified AgNWs@NH₂ (Fig. 9b), it is shown that the AgNWs were more evenly dispersed in the epoxy matrix, because of the crosslinked chain from the reaction between TETA (acted as curing agent) grafted on the AgNWs and the epoxy. TETA played the role of a bridge between the AgNWs and epoxy resin matrix. Moreover, the image (b) shows the toughening texture that might make a contribution to the mechanical properties. The same conclusion could be made from the powerful SEM image inset. It could be observed from the inset images that the fracture surface of the pure AgNWs (Fig. 9a) showed a flat section, and neat and tidy pattern, while the AgNWs@NH₂–epoxy nanocomposite’s image (Fig. 9b) showed more roughness than that of the pure AgNWs–epoxy nanocomposite and the matrix was found to be broken. Not only did the epoxy resin matrix appear as irregular concave, there was a certain amount of epoxy resin stuck on the surface of the AgNWs. It could be reasonably inferred that the matrix backbone could tightly hold the surface-modified AgNWs by interfacial bonding, and testify to the existence of strong interfacial bonding forces between the AgNWs and epoxy matrix.

Fig. 9 SEM images of (a) the pure AgNW–epoxy and (b) AgNWs@NH₂–epoxy nanocomposites.

3.3. Properties of the highly antistatic and thermally conductive nanocomposites

The surface conductivity increased steadily on increasing the loading of the pure AgNWs and surface-modified AgNWs@NH₂, respectively. At low concentrations of AgNWs, the fillers were dispersed in the polymer with rare opportunity to come into contact with each other. Therefore, the conductivity was very low. But as the conductive filler concentration reached a critical value, the AgNWs in the system connected with one another to form an infinite chain network, and the network chain of metal throughout the polymer built a conductive path, so the electrical conductivity rapidly rose. The non-linear seepage phenomenon could be described using the GEM equation:^43,44
where ϕ is the filler volume fraction and ϕ_c is the critical volume fraction (percolation threshold). σ, σ_l and σ_h represent the composite, low conductivity component and high resistivity component, respectively. The index “t” is a parameter. When ϕ > ϕ_c, the filler can form a conductive network to a considerable extent in the composite material and the surface resistivity would sharply decline. By comparing the curves a and b in Fig. 10, ϕ_c of the AgNWs@NH₂ nanocomposite is 0.25 Vt% while for the pure AgNW nanocomposite it is 1 Vt%, which follows the conclusion that the AgNWs@NH₂ nanocomposite exhibited a more pronounced enhancement in conductivity than its counterpart containing pure AgNWs. In addition, the percolation threshold of the AgNWs@NH₂ nanocomposite (Fig. 10) was 4.16 × 10⁷ Ω at 0.25 Vt%, which occurred earlier than for the pure AgNWs. The surface resistivity of the AgNWs@NH₂ nanocomposite was one to two orders of magnitude lower than for the pure AgNWs. This is owed to the functional amino groups that could encourage the dispersion of the AgNWs in the matrix. Due to the mismatch in polarity between the AgNWs and epoxy, the complex was not easy to closely and uniformly disperse. So surface treatment of the AgNWs was necessary. The electrostatic discharge channel of the AgNWs@NH₂ could be better constructed in the matrix, and the contact area between each AgNWs@NH₂ was effectively reduced by surface modification.

Fig. 10 Surface resistivity of the pure AgNW–epoxy and AgNWs@NH₂–epoxy nanocomposites with different filler content.

The thermal conductivities of the nanocomposites with a different volume content of pure AgNWs and surface-modified AgNWs@NH₂ are exhibited in Fig. 11. Lots of unfettered free electrons in the silver metal made the main contribution to its high thermal conductivity.⁴⁵ Because of the free electrons providing an “electron gas” movement, it was possible to achieve rapid heat transfer. Therefore, the thermal conductivity of the AgNW–epoxy nanocomposites was obviously increased. Compared with the epoxy matrix, there was an expected ascent in the thermal conductivity curves with the introduction of the pure AgNWs and AgNWs@NH₂. Comparing the curves of the pure AgNWs and AgNWs@NH₂ nanocomposites, the surface-modified AgNWs@NH₂ nanocomposite had an apparently better thermal performance that could increase to 0.67 W m⁻¹ K⁻¹. Because of the functionalization of the AgNWs, the aggregation of the silver nanowires could be reduced and the wettability of the AgNWs and epoxy could be enhanced. In this way, a good thermal conductivity of the nanocomposites could be created.

Fig. 11 Thermal conductivity of the pure AgNW–epoxy and AgNWs@NH₂–epoxy nanocomposites with different filler content.

4. Conclusions

In summary, amino group functionalized AgNWs (AgNWs@NH₂) were obtained through amidation with thioglycolic acid which was modified onto the surface of pure AgNWs. First and foremost, the surface resistivity of the AgNWs@NH₂–epoxy nanocomposite was significantly lower than that of the nanocomposite containing pure AgNWs, which confirmed the prominent advantage of AgNWs@NH₂ as an effective antistatic agent. The results showed that AgNWs@NH₂ could be homogenously and stably dispersed in the epoxy matrix, and behave as highly antistatic at a lower content (1 Vt%) via forming a large number of conductive pathways in the epoxy matrix with the help of the surface functionalization, which could help the nanocomposites become high performance antistatic materials. The antistatic behavior of the prepared AgNW–epoxy and AgNWs@NH₂–epoxy nanocomposites was investigated. Moreover, the thermal conductivity of the nanocomposites could also be increased. It is expected that our present work will have great value in the development of high performance antistatic and thermally conductive engineered polymers.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (No. 51203007).

References

1. X. Lu, M. Rycenga and S. E. Skrabalak, Annu. Rev. Phys. Chem., 1999, 37, 855–858.
2. Y. N. Xia, Y. J. Xiong and B. Lim, Angew. Chem., Int. Ed., 2009, 48, 60–103.
3. L. Wang, H. Li, J. Tian and X. Sun, ACS Appl. Mater Interfaces, 2010, 11, 2987–2991.
4. G. Chang, Y. Lou, W. Lu, X. Qin, A. Asiri, A. O. Al-Youbi and X. Sun, Catal. Sci. Technol., 2012, 2, 800–806.
5. J. Tian, Y. Lou, H. Li, W. Lu, W. Lu, G. Chang, X. Qin, X. Qin and X. Sun, Catal. Sci. Technol., 2011, 1, 1393–1398.
6. H. N. Li, C. B. Liu, B. Dai, X. H. Tang, Z. J. Zhang, Z. Q. Xiong and X. M. Liu, J. Appl. Polym. Sci., 2015, 132, 42498.
7. M. Gratzel, Prog. Photovolt: Res. Appl., 2000, 8, 171–185.
8. E. Dinda, M. Biswas and T. K. Mandal, J. Phys. Chem. C, 2011, 115, 18518–18530.
9. Y. Wang, J. F. Wong and X. W. Teng, Nano Lett., 2003, 3, 1555–1559.
10. T. R. Zhang, J. P. Ge, Y. X. Hu and Y. D. Yin, Nano Lett., 2007, 7, 3203–3207.
11. J. F. Shen, W. S. Huang, L. P. Wu, Y. Z. Hu and M. X. Ye, Compos. Sci. Technol., 2007, 6730, 41–50.
12. J. Liu and C. Grunlan, Adv. Funct. Mater., 2007, 17, 2343–2348.
13. J. G. Wang, Z. P. Fang and A. J. Gu, Polym. Eng. Sci., 2006, 635–642.
14. S. Barrau, P. Demont and C. Lacabanne, Macromolecules, 2003, 36, 9678–9680.
15. A. Yu, P. Ramesh, M. E. Itkis, E. Bekyarova and R. C. Haddon, J. Phys. Chem. C, 2007, 111, 7565.
16. T. T. Mai, J. W. Schulze and G. Staikow, Electrochim. Acta, 2003, 48, 3021–3027.
17. S. Koul, R. Chandra and S. K. Dhawan, Polymer, 2004, 41, 9305–9310.
18. A. Zheng, X. Xu, H. Xiao, N. Li, Y. Guan and S. Li, Appl. Surf. Sci., 2012, 258, 8861–8866.
19. R. Parra and A. R. Black, Mater. Sci. Forum, 1997, 240, 189–192.
20. M. Kozlowski and S. Frackowia, Sens. Actuators, B, 2005, 109, 141–145.
21. H. Wang, G. Xie, M. Fang, Z. Ying, Y. Tong and Y. Zeng, Composites, Part B, 2015, 79, 444–450.
22. W. J. Kim, M. Taya and M. N. Nguyen, Mech. Mater., 2009, 41, 1116–1124.
23. I. Gou, M. Chipara and J. M. Zaleski, Chem. Mater., 2007, 19, 1755–1760.
24. R. Kilik, R. Davise and S. M. Darwish, Int. J. Adhes. Adhes., 1989, 9, 219–223.
25. H. Y. Liang, W. Z. Wang and Y. Z. Huang, J. Phys. Chem. C, 2011, 114, 7427–7431.
26. Y. G. Sun and Y. N. Xia, Science, 2002, 298, 2176–2179.
27. Q. Zhang, C. Cobley and L. Au, ACS Appl. Mater. Interfaces, 2009, 1, 2044.
28. Y. Sun, Y. Yin, B. T. Mayers, Y. Herricks and J. Xia, Chem. Mater., 2002, 14, 4736–4745.
29. V. J. Gonzalez, E. Gracia-Espion, A. Morelos-Gomez, F. Lopez-Urias and H. M. Terrones, Carbon, 2015, 88, 51–59.
30. A. A. Bol, H. C. Gerritsen and A. Meijerlink, ChemPhysChem., 2002, 3, 871–879.
31. H. J. Jiang, K. S. Mong, J. X. Lu and C. P. Wong, J. Electron. Mater., 2005, 11, 1432–1439.
32. N. Ucar, N. Demirsoy, A. Onen, I. Karacan, N. Kizildag, O. Eren, O. F. Vurur, E. Sezer and B. Ustamehmetoglu, J. Mater. Sci., 2015, 4, 1855–1864.
33. S. M. Gawish, H. Avci, A. M. Ramadam, S. Mosleh, R. Monticello, F. Breidit and R. Kotek, J. Biomater. Sci., Polym. Ed., 2012, 1–4, 43–61.
34. M. Z. Si, R. G. Wu and P. X. Zhang, Spectrosc. Spectral Anal., 2011, 21, 343.
35. C. Petit, P. Lixon and M. P. Pileni, J. Phys. Chem., 1993, 49, 12974–12983.
36. C. S. Onnichsm, T. Franzl, T. G. Wilk and P. J. von, New J. Phys., 2002, 4, 931–938.
37. C. G. Tian, E. B. Wang and Z. H. Kang, J. Solid State Chem., 2006, 179, 3270–3276.
38. K. G. Sujit and P. S. Tara, Chem. Rev., 2007, 107, 4797–4862.
39. W. G. Kwak, M. H. Oh and M. S. Gong, Carbohydr. Polym., 2015, 115, 317–324.
40. H. Zhou, C. Ahang, H. Li and Z. Du, Carbon, 2011, 49, 126–132.
41. W. Zou, Z. J. Du, Y. X. Liu, X. Yang, H. Q. Li and C. Zhang, Compos. Sci. Technol., 2008, 68, 3259–3264.
42. H. Deng, T. Skipa, R. Zhang, D. Lellinger, E. Bilotti, I. Alig and T. Peijs, Polymer, 2009, 50, 3747–3754.
43. H. Lei, W. G. Pitt, L. K. McGrath and C. K. Ho, Sens. Actuators, B, 2004, 101, 122–1132.
44. V. Y. Petrusky and Z. S. Rak, J. Eur. Ceram. Soc., 2001, 21, 237–244.
45. X. Zhang and M. Fujii, Polym. Eng. Sci., 2003, 11, 1755–1764.

---