Fe₃O₄ nanopearl decorated carbon nanotubes stemming from carbon onions with self-cleaning and microwave absorption properties

Abstract

Electromagnetic wave absorbing materials with a microwave absorption capacity over a wide frequency range and a superhydrophobic surface are of great importance for their applications in stealth technology, particularly, in high humidity environments. In this paper, Fe₃O₄ nanopearl decorated carbon nanotubes stemming from carbons onions (CNOs/CNTs@Fe₃O₄) have been successfully fabricated using a facile flame strategy, with brass foil as a substrate and catalyst. Fe₃O₄ nanopearls from the thermal decomposition of iron(III) acetylacetonate were in situ decorated on carbon nanotubes stemming from the ripening of the self-assembly of carbon onion aggregates from the incomplete combustion of ethanol. Notably, this nanocomposite film exhibited a good microwave absorption performance with a wide absorption frequency over a range of 6–18 GHz based on the cooperation of the dielectric loss of the carbon nanotubes and the magnetic loss of the Fe₃O₄ nanoparticles. Furthermore, the nanocomposite film displays superhydrophobic properties and a low adhesive force that makes it a good candidate for water shedding stealth materials.

---

1 Introduction

Microwave absorbing materials that can absorb electromagnetic (EM) waves effectively and dissipate microwaves by interference, are in high demand at present because of the development of military stealth technology and the increased demand of electronic devices.¹ Over the past decade, various EM materials have been developed for these purposes, and they are generally classified as either dielectric or magnetic loss materials, according to their loss characteristics.² In general, typical dielectric loss materials include conducting polymers,^3,4 reduced graphene oxide,^5–8 carbon nanotubes,^9–12 ZnO,^13,14 TiO₂,¹⁵ and BaTiO₃,¹⁶ while magnetic loss materials include Fe₃O₄,¹⁷ Co,¹⁸ Ni,¹⁹ and Fe₃Al,²⁰etc. Different from these, carbon onions and carbon nanotube (CNTs) are representative novel nanomaterials, possessing many fascinating properties like high specific surface area, thermal conductivity, mechanical properties and electron mobility. These properties make them good candidates for EM microwave materials. Unfortunately, there are still some problems unsolved as follows: (1) they have a poor magnetic loss performance, restricting the improvement of their absorption properties. (2) Carbon nanotubes in the powder form usually tend to aggregate and re-stack, leading to an obvious reduction of available surface area. Therefore, it is highly desirable to construct three-dimensional CNTs as microwave absorbers due to their ultra-high surface-to-volume ratio. Furthermore, as for high-efficiency microwave absorbers, it is also required to combine dielectric properties with magnetic materials to achieve an impedance match between complex permittivity and permeability, thus improving the absorption properties. This strategy is needed to construct novel hybrid nanostructures based on carbon nanomaterials and magnetic nanoparticles which can exhibit excellent dielectric polarization, magnetic loss performance and interface action compared with reported EM materials such as Fe₃O₄@TiO₂,²¹ Fe₃O₄/polyaniline,²² Fe₃O₄/CNT,²³ and Fe₃O₄/graphene.²⁴ However, multi-step procedures involved in the fabrication of these hybrid nanomaterials may limit their practical applications. Therefore, it is urgent to develop a facile and large-scale strategy for fabricating nanocomposites for high-efficiency microwave absorption.

In practical applications, it is inevitable for microwave absorbers to encounter high humidity environments that degrade the conductivity of the materials, thus leading to a decrease in the microwave absorption performance. There were some stealth fighter crashes due to their poor water shedding capacity. Some types of stealth fighters even had to be retired, because they lost the capacity of microwave absorption on rainy days.

Superhydrophobic surfaces have extreme water repellency that quickly removes water from the surface, and thus may protect their conductivity. For example, Zhu et al. reported a superhydrophobic, conductive, magnetic carbon nanofiber coating via an electrospinning method.²⁵ Megaridis et al. presented a solvent-based, mild method to fabricate superhydrophobic, carbon nanofiber/PTFE-filled polymer composite coatings for EMI shielding.²⁶ Although some progress has been made in the fabrication of electromagnetic materials with superhydrophobicity, it is of great urgency to develop a facile strategy for preparing high performance microwave absorbing materials with excellent water-shedding properties.

Herein, a facile one-step flame strategy has been developed to fabricate a superhydrophobic CNOs/CNTs@Fe₃O₄ nanocomposite film on a large scale using brass foil as a substrate. The microwave absorbing nanocomposite film has well-defined three-dimensional structures containing aligned carbon nanotubes decorated by magnetite nanopearls. The formation of the CNTs was from the ripening of carbon onion aggregate assemblies. This microwave absorbing nanocomposite exhibits a high performance microwave absorption capacity over a wide frequency range. Moreover, the as-prepared nanocomposite film shows superhydrophobic properties, with a contact angle (CA) of 155 ± 1.6° and low-adhesion, which makes it a promising candidate for water-shedding stealth materials.

2 Experimental section

Materials

Iron(III) acetylacetonate (Fe(acac)₃) was purchased from Aldrich and used after a 2-fold recrystallization. Brass foil was obtained from Goodtime Industry Limited (Beijing). Copper foil (thickness 0.25 mm, 99.95+%) was purchased from Sigma-Aldrich. Other chemical reagents with an analytical grade were used as received.

Preparation of aligned carbon nanotubes decorated by Fe₃O₄ nanopearls

In a typical fabrication, Fe(acac)₃ was dissolved in ethanol with a concentration of 0.4 mol L⁻¹. Then the Fe(acac)₃ solution was pumped to a fixed flamethrower at a certain speed from a raw material tank. After ignition of this solution, a brass foil was employed as a substrate to collect the nanocomposites from the flame of the flamethrower. The reaction temperature was about 488.6 ± 3.5 °C. In order to obtain a large area film, brass foil could be fixed on a computer-controlled rolling device. The thickness of the nanocomposite film could be controlled by deposition time and the moving speed of the brass foil. The brass foil is hydrophilic with a CA of 46.5 ± 1.2°. However, the brass foil with the nanocomposite film exhibits superhydrophobicity with a CA = 155 ± 1.6°. The water droplets could easily slide off this surface. The fabricated process and wettability of the CNTs@Fe₃O₄ aligned film is shown in Scheme 1.

Scheme 1 Synthetic procedure of aligned carbon nanotubes decorated with Fe₃O₄ nanocomposite film via a flame strategy.

Characterization

Scanning electron microscopy (SEM) and energy dispersive spectrometry (EDAX) studies were done using a Hitachi S-4800 microscope and FEI Nova Nanolab SEM, respectively. Transmission electron microscopy (TEM) images of the samples were taken using a JEM-100CXII electron microscope at an acceleration voltage of 100 kV. High resolution TEM (HRTEM) was performed using a JEOL 3011 High Resolution Electron Microscope with an acceleration voltage of 300 kV. Powder X-ray diffraction (PXRD) patterns of the particle samples were recorded on a Rigaku D/Max-2500 diffractometer using CuKα radiation (λ = 1.54056 Å). Raman spectra were recorded on a HORIBA Jobin Yvon (Laboratory RAM HR800) confocal micro-Raman spectrometer with a backscattered geometry through a 10× (NA = 0.25) microscope objective. Ar⁺ laser emission at a wavelength of 514.5 nm was used as a source of excitation. CAs were measured using an OCA20 machine (DataPhysics, Germany) contact angle system at ambient temperature. The magnetic properties of the resultant samples were characterized by using a vibrating sample magnetometer (VSM JDM-13, China). High speed movies were taken with a high speed camera (Phantom v9.1 vision research, America). X-Ray photoelectron spectroscopy data were obtained with an ESCALab220i-XL electron spectrometer from VG Scientific using 300 W Mg KR radiation. The base pressure was about 3 × 10⁻⁹ mbar. A powder sample of the composite material was first mixed with an equal weight of olefin. Then, the resultant mixture was used to form an O-ring shaped sample (i.d. 3 mm, o.d. 7 mm, and thickness 2 mm) under the melting state of paraffin. The relative complex permittivity and the relative complex permeability were determined using a vector network analyzer (Agilent N5230A) in the range of 2–18 GHz.

3 Results and discussion

Scheme 1 shows the synthetic procedure of the nanocomposite film via a simple flame strategy using brass foil as a substrate. The method was based on the combustion of ethanol, in situ thermal decomposition of Fe(acac)₃ and using the brass foil substrate to collect the products from the out-flame of the flamethrower. While the black film was forming on the brass foil, there was a sharp wetting transition from the hydrophilic state with a contract angle (CA) of 46.5 ± 1.2° to a superhydrophobic state with a CA of 155 ± 1.6°.

The morphology and detailed structures of the as-fabricated film and nanoparticles stripped from this film were investigated and characterized using SEM, TEM and HRTEM. From Fig. 1a, it can be clearly observed that the film has a three-dimensional (3D) structure. The high resolution SEM images indicate that the film is composed of curved nanofibers decorated by many nanoparticles with an average size of 48.0 ± 14.6 nm, as given in Fig. 1b and c. The TEM image further indicates that the nanofibers are hollow tubes that have an average length and diameter of 4.3 ± 1.1 μm and 19.4 ± 2.8 nm (Fig. 1c). The HRTEM images show a well-defined crystalline lattice spacing of 3.4 Å attributable to the (002) crystal plane of the carbon onions,²⁷ while the crystal lattice spacing of 3.8 Å is assigned as being characteristic of multi-walled carbon nanotubes (Fig. 1d and f). Furthermore, the HRTEM image clearly demonstrates that the lattice fringe pitch of 4.9 Å corresponds well with the d-spacing of the (111) reflections for single-crystalline Fe₃O₄ (Fig. 1e). The XRD pattern (Fig. 2) indicates that the broad diffraction peak located at 28° corresponds to the (002) plane of graphite which has an inter-planar spacing of 3.4 Å (Fig. 1e).²⁸ The other six peaks indexed (220), (311), (400), (422), (511) and (440) are characteristic of and can be attributed to Fe₃O₄ nanoparticles (JCPDF no. 88-0866), which is in good agreement with the HRTEM results.

Fig. 1 (a and b) SEM images of the nanocomposite film with different magnifications, (c) TEM image of a sample stripped from the as-prepared film, and (d, e and f) HRTEM images of the carbon onions, Fe₃O₄ nanoparticles, and carbon nanotubes in sequence.

Fig. 2 XRD patterns of the CNTs@Fe₃O₄ nanocomposites obtained using a flame strategy.

Fig. 3 shows the Raman spectrum of the nanocomposite film. It is observed that the G band centred at 1598 cm⁻¹ indicates that the nanocomposite film has a graphitic sp² phase corresponding to the E_2g phonon mode at the Brillouin zone atom.² The in-plane bond stretching of the sp²-bonded carbon atoms in the as-prepared nanocomposite film is also found, suggestive of some disordered graphitic structures. The D band at 1350 cm⁻¹ indicates the breathing mode of the six-fold aromatic rings in the carbon network and is assigned to photons of A_1g symmetry at the K point of the Brillouin zone. This mode indicates that the as-prepared material presents a structural disorder like the armchair edges of the carbon network. Generally, the growth in the D-band intensity is widely used to qualify the disorder. The ratio of the D-band to G-band intensity (I_D/I_G) is 0.69 which indicates that some disordered carbon exists in the as-prepared nanocomposite materials. The dimensionless A band at about 1528 cm⁻¹ is indicative of interstitial carbon with sp³ linking outside or inside the planes of the aromatic rings, which is assigned to amorphous carbon.²⁹ The ratio of the intensity of the A band to the intensity of the G band (I_A/I_G) is 0.19 which supports that carbon with an amorphous structure is present in the obtained products.

Fig. 3 Raman spectrum of nanocomposites stripped from the as-prepared superhydrophobic film.

The magnetic properties of the obtained composite film were also investigated (Fig. 4). It can be seen that the composite materials stripped from the brass substrate in the glass vessel can be easily attracted by a magnet, suggesting the good magnetic properties of the composite film (Fig. 4a–c). Furthermore, vibrating sample magnetometry (VSM) data indicate that the composite film is paramagnetic with a saturation magnetization (M_s) of 12.5 emu g⁻¹, coercivity (H_c) of 100.9 Oe and remnant magnetization (M_r) of 2.4 emu g⁻¹. The magnetic properties could be attributed to the Fe₃O₄ nanoparticles in the composite film, which provide the magnetic loss for microwave absorbing materials.

Fig. 4 (a) Photographs of the nanocomposite film on the brass foil, (b and c) nanocomposite sample stripped from this film captured in the absence or presence of a permanent magnet, and (d) room-temperature magnetization curves of the nanocomposites prepared using a flame synthesis. The inset shows a detailed view of the room temperature magnetization of this sample.

Fig. 5 Survey X-ray photoelectron spectra of (a) Cu 2p_3/2 and (b) Zn 2p_3/2 of the brass foil before (black lines) and after (red lines) being used as a substrate in the flame synthesis. The bold lines shown are the best fits for Cu 2p_3/2 and Zn 2p_3/2, respectively.

It is generally accepted that the Fe₃O₄ nanoparticles in this film were formed from the thermal decomposition of Fe(acac)₃via a flame strategy. However, the formation mechanism of CNOs and CNTs is still unclear. It is noticed that the composite film of CNOs/CNTs@Fe₃O₄ can form on the surface of the brass foil composed of Cu–Zn alloy with a molar ratio of 2.1 to 1 (Fig. S2†). When pure copper is used as a substrate, only CNOs are found in the products (Fig. S1†), no CNTs are observed in the products. It is easily concluded that the presence of zinc in the brass foil plays an important role in the synthesis of the CNOs/CNTs@Fe₃O₄ nanocomposite film, especially for the CNTs. To further understand the formation mechanism, fresh and used brass foils were examined using X-ray photoelectron spectroscopy (XPS) (Fig. 5). It is observed that the binding energies of Zn 2p_3/2 and Cu 2p_2/3 are 1021.9 eV and 933.5 eV, respectively, indicating the presence of Zn⁰ and Cu²⁺ in the brass foil. However, after brass foil was employed as a substrate, the binding energy of Zn 2p_3/2 shifted to a higher value of 1022.4 eV, suggestive of the formation of Zn²⁺, but the binding energy of Cu 2p_2/3 had almost no change. This result indicates that the presence of Zn⁰ in the brass foil can contribute to the formation of CNTs in the composite film.

The multi-walled nature of the CNTs is also investigated via varying the reaction times of 30 min, 60 min, and 180 min (Fig. 6a–c). It can be seen that at 30 min, the obtained products are spherical CNO aggregates with an average size of 31.1 ± 6.2 nm. With a prolonged reaction time, the CNO aggregates assemble into nanofibers with a relatively loose structure at 60 min. When the reaction time is prolonged to 180 min, multi-walled CNTs were observed. Therefore, it is reasonable to deduce that the CNTs were produced by a ripening of the assemblies of the nanotube-like structures of the CNO aggregates.

Fig. 6 SEM images of the products collected at (a) 30 min, (b) 60 min, and (c) 180 min.

Due to the presence of CNOs, MWCNTs and Fe₃O₄ nanoparticles, these nanocomposites can possess both magnetic and dielectric properties, thus making a film with special electromagnetic absorption properties. The electromagnetic parameters (relative complex permittivity ε_r = ε′ − jε′′ and relative complex permeability, μ_r = μ′ − jμ′′) of wax composites containing 5 wt% of the as-prepared nanocomposites were measured at room temperature.²⁴ The permittivity of the real part (ε′), permittivity of the imaginary part (ε′′), permeability of the real part (μ′), permeability of the imaginary part (μ′′), dielectric loss tangent (tan δε) and magnetic loss tangent (tan δμ) of the nanocomposite materials are presented in Fig. S4.† For contrast, the CNOs/CNTs nanocomposite materials were obtained by removing the Fe₃O₄ nanoparticles from the as-prepared film materials via HCl (10 M). It is found that the ε′ values of CNOs/CNTs@Fe₃O₄ and CNOs/CNTs decrease from 4.98 to 4.57 and from 5.74 to 4.72, respectively, with some fluctuations over the 2–18 GHz frequency range (Fig. S4a†), while the ε′′ value changes from 0.62 to 0.44 and from 1.10 to 0.66, respectively, in the frequency range of 2–18 GHz (Fig. S4b†). The μ′ and μ′′ values of CNOs/CNTs@Fe₃O₄ and CNOs/CNTs exhibit complex fluctuations in the frequency of 2–18 GHz (Fig. S4c and d†). It is observed that when magnetic Fe₃O₄ nanoparticles are removed, the μ′ and μ′′ values decrease sharply due to low magnetic loss properties. The dielectric tangent loss (tan δε = ε′′/ε′) and the magnetic tangent loss (tan δμ = μ′′/μ′) are shown in Fig. S4e and f,† respectively. It suggests that sample A has a strong dielectric loss against electromagnetic waves. By removing the magnetic Fe₃O₄ nanoparticles, the CNOs/CNTs nanocomposite has a greater ratio of conducting material than the CNOs/CNTs@Fe₃O₄ nanocomposite material. The strong magnetic tangent loss (tan δμ = μ′′/μ′) of sample A indicates that CNOs/CNTs@Fe₃O₄ has more magnetic loss than CNOs/CNTs between 5 to 12.3, and 15 to 18 GHz. These results indicated that the electromagnetic absorption properties of CNOs/CNTs@Fe₃O₄ originate from the coupling of the dielectric loss based on CNOs and CNTs and the magnetic loss from the Fe₃O₄ nanoparticles.

The reflection loss (R_L) values of the nanocomposites were calculated according to the following equations.³⁰

where Z_in is the input impedance of the absorber, f is the frequency of the electromagnetic waves, d is the thickness of the absorber, and c is the velocity of the electromagnetic waves in free space. The impedance matching condition is determined by the combination of the six parameters: ε′, ε′′, μ′, μ′′, f and d.³¹ In order to study the influence of CNOs, CNTs and magnetite nanoparticles on the microwave absorption, the calculated theoretical reflection loss values of the CNOs/CNTs@Fe₃O₄ and CNOs/CNTs with a filler loading of 5 wt% at different thicknesses in the range of 2–18 GHz were calculated according to eqn (1) and (2).³²Fig. 7a and b show the RL of the CNOs/CNTs@Fe₃O₄ and CNOs/CNTs. It can be obviously observed that CNOs/CNTs@Fe₃O₄ has three different microwave absorption areas including area I (thickness 2.5–4.5 mm, frequency 12–18 GHz), area II (thickness 4.5–5.5 mm, frequency 16–18 GHz) and area III (thickness 4.0–5.5 mm, frequency 6–12 GHz). By contrast, CNOs/CNTs has only two smaller absorption areas including area I (thickness 2.5–3.5 mm, 14–18 GHz) and area II (thickness 4.5–5.5 mm, frequency 16–18 GHz). According to the chemical difference between CNOs/CNTs@Fe₃O₄ and CNOs/CNTs, the magnetite nanoparticles in sample A contribute to the wider frequency range for microwave absorption at a lower thickness, but the film has a relatively good microwave absorption performance in the low frequency range. Notably, the as-prepared CNOs/CNTs@Fe₃O₄ nanoarray film exhibited good microwave absorption properties both at low and high frequencies.

Fig. 7 Microwave absorption properties of the as-prepared nanocomposite films calculated by using the measured relative complex permittivity and permeability values according to transmission line theory. Three-dimensional microwave RL curves of the (a) CNOs/CNTs@Fe₃O₄ and (b) CNOs/CNTs nanocomposites.

A superhydrophobic surface for microwave absorbing nanocomposite films is also very important for their practical civil and military applications. The as-prepared nanocomposite film exhibits superhydrophobic properties at a wCA of 155° in air. It is necessary to study the interaction between the water droplet and this surface. Fig. 8 shows the time-resolved images of a water droplet bouncing on the as-prepared CNOs/CNTs@Fe₃O₄ nanocomposite film, well studied using a high speed camera. It was obviously observed that there was no residual water during the bouncing process. The as-prepared nanocomposite film exhibited an isotropic structure, which was proved by the same value of the sliding angle (∼3°) for the different directions of this film. These results proved that this nanocomposite film with a 3D structure exhibited superhydrophobic properties. Considering the microwave absorption properties and superhydrophobic wettability with low adhesive properties, this film may have a future application in stealth technology as stealth materials with the capacity to absorb microwave signals with both high and low frequencies and excellent water shedding properties.

Fig. 8 Time-resolved images of the bouncing of a 5 μL water drop on a superhydrophobic as-prepared microwave absorber film.

4 Conclusions

In summary, a simple flame strategy is developed to construct a GNOs/CNTs@Fe₃O₄ nanoarray film on a brass foil substrate. The brass foil that is composed of Zn⁰ serves not only as the substrate, but also as the catalyst. CNOs were formed by incomplete combustion of ethanol, and then the tube-like structures were obtained by the self-assembly of CNO aggregates. With the reaction proceeding, CNT arrays were prepared by the ripening of the tube-like intermediates. At the same time, Fe₃O₄ nanoparticles were produced from the thermal decomposition of Fe(acac)₃, which decorated the CNTs in situ. Furthermore, the as-prepared nanocomposite film exhibits an excellent microwave absorption capacity over a wide range of frequencies and superhydrophobic wettability with low adhesive properties. It is believed that the CNO/CNTs@Fe₃O₄ nanoarray film can be effective for applications in water-shedding stealth materials.

Acknowledgements

The authors thank the financial support of the Beijing Natural Science Foundation (2132030), the National Natural Science Foundation of China (21103006), China Scholarship Council (201303070263), 863 Program (2012AA030305), the Fundamental Research Funds for the Central Universities (YWF-10-01-B16, YWF-11-03-Q-214, YWF-13-DX-XYJL-004) and the 111 Project (no. B14009). The authors thank Prof. Dr Nicholas A. Kotov of the University of Michigan, Ann Arbor for his kind help with the HRTEM measurements.

Notes and references

1. H. He and C. Gao, ACS Appl. Mater. Interfaces, 2010, 2, 3201–3210.
2. L. Wang, X. Jia, Y. Li, F. Yang, L. Zhang, L. Liu, X. Ren and H. Yang, J. Mater. Chem. A, 2014, 2, 14940–14946.
3. H. Ghasemi and U. Sundararaj, Synth. Met., 2012, 162, 1177–1183.
4. A. Ameli, M. Nofar, S. Wang and C. B. Park, ACS Appl. Mater. Interfaces, 2014, 6, 11091–11100.
5. M. Verma, A. P. Singh, P. Sambyal, B. P. Singh, S. K. Dhawan and V. Choudhary, Phys. Chem. Chem. Phys., 2015, 17, 1610–1618.
6. A. Joshi, A. Bajaj, R. Singh, P. S. Alegaonkar, K. Balasubramanian and S. Datar, Nanotechnology, 2013, 24, 455705–455708.
7. D.-X. Yan, P.-G. Ren, H. Pang, Q. Fu, M.-B. Yang and Z.-M. Li, J. Mater. Chem., 2012, 22, 18772–18774.
8. S. Kim, J.-S. Oh, M.-G. Kim, W. Jang, M. Wang, Y. Kim, H. W. Seo, Y. C. Kim, J.-H. Lee, Y. Lee and J.-D. Nam, ACS Appl. Mater. Interfaces, 2014, 6, 17647–17653.
9. S. Maiti, N. K. Shrivastava, S. Suin and B. B. Khatua, ACS Appl. Mater. Interfaces, 2013, 5, 4712–4724.
10. M. H. Al-Saleh and U. Sundararaj, Carbon, 2009, 47, 1738–1746.
11. A. P. Singh, B. K. Gupta, M. Mishra, G. A. Chandra, R. B. Mathur and S. K. Dhawan, Carbon, 2013, 56, 86–96.
12. J.-M. Thomassin, X. Lou, C. Pagnoulle, A. Saib, L. Bednarz, I. Huynen, R. Jerome and C. Detrembleur, J. Phys. Chem. C, 2007, 111, 11186–11192.
13. R. F. Zhuo, H. T. Feng, J. T. Chen, D. Yan, J. J. Feng, H. J. Li, B. S. Geng, S. Cheng, X. Y. Xu and P. X. Yan, J. Phys. Chem. C, 2008, 112, 11767–11775.
14. H. Li, Y. Huang, G. Sun, X. Yan, Y. Yang, J. Wang and Y. Zhang, J. Phys. Chem. C, 2010, 114, 10088–10091.
15. T. Xia, C. Zhang, N. A. Oyler and X. Chen, Adv. Mater., 2013, 25, 6905–6910.
16. Y.-F. Zhu, L. Zhang, T. Natsuki, Y.-Q. Fu and Q.-Q. Ni, ACS Appl. Mater. Interfaces, 2012, 4, 2101–2106.
17. G. Sun, B. Dong, M. Cao, B. Wei and C. Hu, Chem. Mater., 2011, 23, 1587–1593.
18. D. Zhang, F. Xu, J. Lin, Z. Yang and M. Zhang, Carbon, 2014, 80, 103–111.
19. C. Wang, X. Han, P. Xu, J. Wang, Y. Du, X. Wang, W. Qin and T. Zhang, J. Phys. Chem. C, 2010, 114, 3196–3203.
20. J. Wei, T. Wang and F. Li, J. Magn. Magn. Mater., 2011, 323, 2608–2612.
21. J. Liu, R. Che, H. Chen, F. Zhang, F. Xia, Q. Wu and M. Wang, Small, 2012, 8, 1214–1221.
22. C. Cui, Y. Du, T. Li, X. Zheng, X. Wang, X. Han and P. Xu, J. Phys. Chem. B, 2012, 116, 9523–9531.
23. Y. Liu, Z. Zhang, S. Xiao, C. Qiang, L. Tian and J. Xu, Appl. Surf. Sci., 2011, 257, 7678–7683.
24. M. Zong, Y. Huang, Y. Zhao, X. Sun, C. Qu, D. Luo and J. Zheng, RSC Adv., 2013, 3, 23638–23648.
25. Y. Zhu, J. C. Zhang, J. Zhai, Y. M. Zheng, L. Feng and L. Jiang, ChemPhysChem, 2006, 7, 336–341.
26. A. Das, H. T. Hayvaci, M. K. Tiwari, I. S. Bayer, D. Erricolo and C. M. Megaridis, J. Colloid Interface Sci., 2010, 353, 311–315.
27. J. Xiao, G. Ouyang, P. Liu, C. X. Wang and G. W. Yang, Nano Lett., 2014, 14, 3645–3652.
28. M. H. Rümmeli, P. Ayala and T. Pichler, in Carbon Nanotubes and Related Structures, Wiley-VCH Verlag GmbH & Co. KGaA, 2010, ch. 1, pp. 1–21.
29. K. R. Paton and A. H. Windle, Carbon, 2008, 46, 1935–1941.
30. R. C. Che, L. M. Peng, X. F. Duan, Q. Chen and X. L. Liang, Adv. Mater., 2004, 16, 401–405.
31. G. Wang, Z. Gao, S. Tang, C. Chen, F. Duan, S. Zhao, S. Lin, Y. Feng, L. Zhou and Y. Qin, ACS Nano, 2012, 6, 11009–11017.
32. Y. Naito and K. Suetake, IEEE Trans. Microwave Theory Tech., 1971, 19, 65–72.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra06042g

‡ These authors contributed equally to this work.

---