Synthesis of 3-dimensional porous graphene nanosheets using electron cyclotron resonance plasma enhanced chemical vapour deposition

Abstract

Microwave plasma driven chemical vapour deposition was used to synthesize graphene nanosheets from a mixture of acetylene and hydrogen gas molecules. In this plasma, acetylene decomposes to carbon atoms that form nanostructures in the outlet plasma stream and get deposited on the substrate. The GNS consists of a few layers of graphene aligned vertically to the substrate. Graphene layers have been confirmed by high-resolution transmission electron microscopy, and Raman spectral studies were conducted to observe the defective nature of the sample. The growth of nanosheets in a vertical direction is assumed to be due to the effect of electric field and from the difference in the deposition rate in the axial and parallel directions. These vertical graphene sheets are attractive for various applications in energy storage and sensors.

---

1. Introduction

Carbon materials have received considerable attention in recent years due to their unique properties in various forms such as diamond, graphite, graphene, amorphous carbon, fullerenes and nanotubes. These structures have shown remarkable differences in properties and hence in potential applications, such as photovoltaics,^1–4 field emission,⁵ energy storage,⁶ transistors,⁷ biology,⁸ and hydrogen storage,⁹ due to their peculiar chemical, electrical, optical and mechanical properties.¹⁰

There are several methods for synthesizing graphene, which can be broadly classified as top down and bottom up approaches. In bottom up approach, graphene is synthesized from atomic or molecular species, whose precursor particles are allowed to gradually grow in size. In top down synthesis, graphite, carbon fibers and carbon nanotubes are the starting materials and individual graphene layers are extracted or peeled off either by physical, electrochemical or chemical methods.^11–14 The road block in making graphene devices is the absence of a viable method for mass production. Chemical vapour deposition (CVD),¹⁵ mechanical exfoliation,¹⁶ chemical exfoliation,¹⁷ liquid phase exfoliation¹⁸ and epitaxial growth¹⁹ generally produce single, bi-layer or tri-layer graphene sheets, but have low yield and thus are unsuitable for mass production. Recently, graphene has been synthesized from chemical routes,²⁰ in which graphite oxide readily exfoliates in water and the suspension can be chemically reduced to yield graphene. This highly reproducible chemical route is among the first techniques to promise the mass production of graphene, but the synthesis of graphene through chemical routes, which involves many toxic chemicals, is not compatible with the present microelectronic device technology.

Vertically oriented graphene (VG) nanosheets are a special type of two dimensional nanostructures that are oriented vertically on a substrate. These VG nanosheets are highly porous and consist of a few layers of graphene with a layer number of 1–15 and the pore diameter is in the range of 50–200 nm. These carbon nanostructures have unique features such as sharp edges, vertical morphology, high surface to volume ratio and can be found in a great number of applications, including energy storage (Li ion batteries^21,22 and supercapacitors), gas sensors, field emitters, biosensors and fuel cells.^1–10

Microwave plasma enhanced chemical vapour deposition (MW-PECVD) has been established as a key method for producing graphene nanosheets. PECVD technique has many advantages over thermal CVD, including low substrate temperature, higher growth selectivity, better control in nanostructure ordering,²³ and low operating pressure (10⁻³ to 10⁻⁴ mbar), leading to high purity films, as well as high ionization coefficient, that enabling a higher growth rate.^24–27

2. Experimental

Graphene nanosheets were synthesized in a home-made ECR-PECVD set-up (Fig. 1). The ECR system consisted of a plasma source chamber mounted on a stainless steel chamber. Beam diameter at the exit of the source was about 100 mm. A flexible substrate holder allowed the free positioning of the substrate along the axis of the source chamber. The substrate was kept at a distance of 10 cm from the exit of the plasma source and the ion density at the surface of the substrate was measured to be approximately 9.5 × 10¹⁰ cm⁻³. The details of the experimental system and working principle have been described in our group's previous study.²⁸

Fig. 1 Schematic of ECR plasma CVD setup.

Annealed polycrystalline copper (Cu) foil with a thickness of 0.25 mm (Alfa Aesar) and an area of 1 cm² was used as the substrate. The substrate was cleaned with dilute nitric acid and rinsed with de-ionized (DI) water. Prior to the deposition, the substrate was heated at 500 °C for 30 min. Subsequently, the substrate was treated in H₂ plasma for 10 min at 500 °C. The carbon deposition was carried out at 700 °C for 30 min. A mixture of hydrogen (H₂) and acetylene (C₂H₂) was used at a ratio of 2 : 1. Plasma power was set to 500 W and the working pressure was set at 7 × 10⁻⁴ mbar. Furthermore, GNS was deposited over different substrates such as platinum or nickel coated copper, SiC, SiO₂ and quartz.

3. Results and discussion

3.1 Growth of graphene nanosheets

Several reports have been published for the synthesis of graphene nanosheets. Wu et al.²⁵ and Zheng et al.²³ described the synthesis of graphitic carbon nanostructures composed of a few layers of graphene vertically standing on a substrate via microwave plasma enhanced chemical vapor deposition (MW-PECVD). Chen et al. reported the GNS synthesis using direct current (dc) PECVD at atmospheric pressure on various substrates.^29–31 The exact mechanism of the GNS growth is not known completely.

The vertical nature of GNS may be due to the difference of growth in horizontal to vertical directions. The exact mechanism of GNS growth is not well understood, and it is assumed that electric field plays an important role.^32,33 It has also been reported that electric fields play an important role in the alignment of CNTs in microwave plasmas.^34,35 The electric field tends to align vertically to the plate, when metallic objects are directly placed in the plasma field even without any biasing. In our experiment, we placed the substrate over a metallic plate from where the electric field can influence the GNS growth. In addition to the electric field, the vertical characteristic was also affected by H₂ dilution, which also contributed to the shaping of the nanowall morphology by etching the loosely bonded carbon atoms and amorphous carbon. Thus the graphitization of graphene nanowall could be enhanced.

From the SEM observations, it was found that an amorphous carbon film was deposited on the Cu substrates from the acetylene plasma. At a certain temperature, in the presence of plasma, the amorphous carbon film was bombarded to form random small size nanoflakes. These nanoflakes act as the nucleation sites. Subsequently, the GNS grew on the nucleation positions and aligned vertically on the substrate. The vertical nature of GNS is due to the difference in growth rates between the parallel and perpendicular directions to the graphene nanosheet.³⁶ The local electric field from the plasma also has an effect on the vertical orientation of GNS. The electric-field dependent orientation has been observed for carbon nanotubes and nanowalls.^37,38 Herein, we study the growth of GNS films by various growth parameters and discuss how the parameters, such as deposition time, flow rate of acetylene, temperature and microwave power, influence the growth. Hydrogen dilution with acetylene plasma also has an influence on the growth of GNS.

Various parameters influence the growth of graphene sheets for the plasma CVD deposition. The surface morphology of GNS over Cu substrate at different deposition time is given in Fig. 2. All the depositions were carried out at a microwave power of 500 W and at a substrate temperature of 700 °C. The flow rate of acetylene was set to 12.4 sccm and the ratio of C₂H₂ : H₂ was 1 : 2. After one min of deposition, no GNS nucleation was observed (Fig. 2a). An amorphous kind of carbon was observed after 3 min of deposition (Fig. 2b). Then, uniform nucleation of GNS started after 5 min of deposition (Fig. 2c). From SEM observations, it was found that the nucleation started after a certain period of deposition time elapsed. The thickness of the GNS film was measured about 200 nm (Fig. 2d).

Fig. 2 SEM images of graphene nanosheets grown on Cu substrate by ECR-PECVD at (a) 1 min of deposition, (b) 3 min of deposition and (c) 5 min of deposition. (d) Thickness of GNS. Deposition was carried out at 500 W microwave power, 700 °C and C₂H₂ : H₂ = 1 : 2.

Fig. 3 shows the SEM images of GNS grown by varying the flow rate of C₂H₂. All the depositions were carried out at a microwave power of 500 W, substrate temperature of 700 °C and deposition time of 5 min. At the flow rate of 5 sccm of C₂H₂, an amorphous carbon was deposited on the substrate and the GNS formation was not visible (Fig. 3a). When the flow rate was increased to 12 sccm, nucleation of GNS started (Fig. 3b) and at 25 sccm, there was good GNS formation on the substrate (Fig. 3c). The GNS was highly porous and uniformly distributed. The mixture of C₂H₂ and H₂ has been taken as 1 : 2. There should be a minimum concentration of C₂H₂ required to obtain uniform GNS over the entire substrate.

Fig. 3 SEM images of GNS on Cu substrate at various C₂H₂ flow rates (a) 5 sccm, (b) 12 sccm and (c) 25 sccm, at various microwave powers of (d) 100 W, (e) 200 W and (f) 500 W, at (g) room temperature, (h) 400 °C and (i) 700 °C.

The SEM images (Fig. 3d–f) show the effect of microwave power on the growth of GNS. The microwave power was varied from 100 W to 500 W. For the deposition of GNS, the flow rate of C₂H₂ was kept constant at 12 sccm. All the deposition was carried out at a substrate temperature of 700 °C and 5 min of deposition time. Furthermore, at the microwave power of 100 W (Fig. 3d), the GNS film contains mainly amorphous carbon. When the microwave power was increased to 300 W, the nucleation of GNS started (Fig. 3e) and at 500 W, well defined GNS growth was observed (Fig. 3f). The growth of GNS at different temperatures was studied at a constant C₂H₂ flow rate of 12 sccm, microwave power of 500 W and deposition time of 5 min. When deposition was carried out at room temperature, the film contained irregular and amorphous carbon, as shown in Fig. 3g. When the temperature was increased to 400 °C and 700 °C, uniform distribution and initiation of GNS nucleation could be observed (Fig. 3h and i).

Nonomura et al. studied the effect of H₂ dilution on GNS growth by hot wire CVD. It was found that GNS growth varies with different hydrogen dilution with methane (CH₄). The surface morphology of GNS changes at different hydrogen concentrations. One of the effects of H₂ dilution would be the etching activity of hydrogen radical. Hydrogen radical etches loosely bonded carbon and enhance the graphitization of diamond-like carbon (DLC) film.^39,40 Hydrogen radicals etch away disordered carbon atoms such as amorphous carbon rather than graphitic carbon. The bond strength of disordered carbon is weaker than that of ordered carbon.

Fig. 4a shows the SEM image of GNS film without hydrogen. A lot of amorphous carbon was deposited on Cu substrate and the thickness of the graphene sheet increased. Amorphous carbon was present in the film even after equal rates of (1 : 1) H₂ and C₂H₂ was used for deposition (Fig. 4b). Highly porous and uniformly distributed GNS were observed for the films that used 2 : 1 ratios of hydrogen and acetylene (Fig. 4c). The amorphous carbon was etched out due to the presence of more hydrogen radicals in the plasma. If the ratio of hydrogen is increased, more carbon will be etched out and the porous cavity will be increased (Fig. 4d).

Fig. 4 SEM images of GNS at various hydrogen dilutions with C₂H₂: (a) without hydrogen, (b) with H₂ : C₂H₂ = 1 : 1, (c) with H₂ : C₂H₂ = 2 : 1 and (d) with H₂ : C₂H₂ = 3 : 1 (all depositions were carried out at 12 sccm of C₂H₂, 500 W, 20 min of deposition time and 700 °C).

3.2 Morphology of GNS over different substrates

GNS films were deposited over different substrates using plasma enhanced CVD technique. The substrates used for the deposition were SiO₂, quartz, amorphous SiC (a-SiC), Cu, Pt and Ni. Pt and Ni were sputtered over the Cu substrate. The average thickness of the coating was ∼40 nm. An amorphous SiC coating was synthesised with reactive sputtering of silicon with acetylene (C₂H₄) over Si substrate and was then used as the substrate. All the substrates were cleaned with deionized (DI) water and ultrasonicated for 10 min. Furthermore, the substrates were dried with dry air and all the GNS deposition was carried out at a microwave power of 500 W, C₂H₂ flow rate of 12 sccm, substrate temperature of 700 °C and deposition time of 25 min. The ratio of H₂ and C₂H₂ mixture was about 2 : 1.

The morphology of the as-deposited GNS over metallic substrates, such as Cu, Pt and Ni, was observed by scanning electron microscopy (Fig. 5). The SEM images show a highly porous and vertically aligned thin wall type of morphology for the GNS electrode, which has a large surface to volume ratio. The morphology of NiGNS and PtGNS clearly reveals that the spherical flower-like graphene petals with a diameter of 1–2 μm were uniformly distributed over the surface of the substrate. The graphene walls formed over Ni and Pt coated Cu substrates are highly porous and branched (Fig. 5c–f), whereas GNSs over bare Cu substrate (Fig. 5a and b) are straight with a pore size of 50–150 nm in diameter.

Fig. 5 Scanning electron microscopy (SEM) images of graphene nanosheets on (a and b) copper, (c and d) nickel and (e and f) platinum substrates.

Fig. 6 shows the SEM image of GNS over insulating substrates such as SiO₂ (Fig. 6a and b), quartz (Fig. 6c and d) and a-SiC (Fig. 6e and f). The density of GNS over these substrates varies. The edges of the vertically grown GNS are highly curled and spiralled. From the SEM images, it can be observed that the nature of the substrate has an effect on GNS growth. High density GNS films were grown over the metallic substrates such as Cu, Pt and Ni. On the other hand, the growth of GNS on insulating substrates was less dense. The depositions of GNS over these substrates were performed after etching the surface in H₂ plasma for 10 min. While etching, the surface of Pt and Ni coated Cu substrates became very rough. The high density of GNS over Pt and Ni coated substrates may be due to the effect of roughness.

Fig. 6 SEM images of GNS on (a and b) SiO₂, (c and d) quartz and (e and f) a-SiC substrates.

The cross section views of CuGNS and PtGNS are shown in Fig. 7a and b. The highly branched morphology of NiGNS and PtGNS offers high surface area and high ratio of porosity. The HRTEM image further shows that very few number of graphene layers are present in the GNS (Fig. 8b).

Fig. 7 Cross section view of graphene nanosheets: (a) PtGNS, which are highly branched and vertically grown and (b) CuGNS electrode, aligned vertically and straight forward.

Fig. 8 Various characterizations for GNS. (a) XRD spectrum, (b) HR-TEM image, (c) Raman spectra of GNS on Cu substrate (inset: spectra of GNS on different substrates), (d) 3D AFM image, (e) line spectra from the AFM image and (f) contact angle measurement on GNS surface.

Various characterisations were carried out on the GNS samples. The XRD spectrum of the GNS sample is given in Fig. 8a. The peak observed at 2θ of ∼21° is assigned as the graphitic peak. The HR-TEM images of graphene flakes are given in Fig. 8b. Samples for the TEM study were prepared by scratching the GNS film deposited over Cu foil and dispersing it in ethanol with ultrasonication for 30 minutes. Thick graphite films and thin graphene flakes were dispersed well, wherein a few-layered light and airy graphene rise to the top. These few-layered GNS were collected, dispersed in ethanol and ultrasonicated for more than 40 min to produce even less few-layered graphene sheets. More than 15 layers of graphene were observed in the TEM image.

Raman spectra of the GNS film on Cu substrate at the initial stage and after the completion of growth are shown in Fig. 8c. The spectra clearly show the graphitization of graphene from the amorphous carbon. From the spectra it is observed that amorphous carbon is present at the initial time of synthesis and later it becomes graphitized. Raman spectra show that the G mode (due to bond stretching between the pairs of sp² carbon atoms) and D mode (due to breathing modes of sp² carbon atoms in rings) lie around 1580 and 1360 cm⁻¹, respectively. The D band appears when a finite wave vector mode becomes Raman active in the presence of in plane disorder such as vacancies, stacking faults or finite basal plane dimensions that break translational invariance.^41,42 The important feature appearing at approximately 2700 cm⁻¹ is usually called the 2D band in the studies related to graphite and is found in all the materials containing sp² carbon. The 2D peak of graphene sheet emerges in the GNS spectra when compared to the amorphous carbon. The 2D peak is used to confirm the presence of graphene and this second order Raman spectrum is due to a double resonance process involving two phonons of opposite wave vector.⁴³ Second order peaks include both overtones and combinations of first order peaks⁴⁴ and are seen at 2700 (2D) and 2930 cm⁻¹ (D + G). There is a blue shift of 13 cm⁻¹ in the G band position when compared to bulk graphite (1581 cm⁻¹) and this shift is attributed to the formation of bulk graphite crystal to graphene sheet.⁴⁵ The inset (Fig. 8c) shows the Raman spectra from different GNS samples grown on different substrates, such as Pt, Ni, Si, SiO₂ and a-SiC, after the completion of growth. All the spectra show clear 2D peaks, which confirm the growth of graphene on all the substrates.

The high resolution 3D AFM image of GNS is shown in Fig. 8d. The morphology of the GNS surface from the AFM image appears to be same as shown in the SEM images of GNS. The line scan of the AFM image for the GNS surface gives the details regarding the thickness of the vertically oriented graphene walls, as shown in Fig. 8e. From the AFM line scan, the thickness of each nanowall tip was observed to be about 15 nm, consisting of more than 20 layers of graphene. Carbon-based nanomaterials, such as carbon nanotubes and graphene, are excellent candidates for superhydrophobic surfaces because of their intrinsically high surface area and nonpolar carbon structure.⁴⁶ These nanostructured surfaces possess ultrahigh adhesion such as the rose petal effect. Many mechanisms involving the rose petal effect have been revealed such as morphological factors, composite micro-nanostructure, and chemical defects of hydrophobic surfaces.^47–49 To observe the hydrophobic properties of the graphene nanosheet matrix, water contact angle (CA) measurements were performed by placing droplets of distilled water on the surface of GNS samples. It was observed that the GNS surface showed strong hydrophobic properties. As shown in Fig. 8f, the optical image of a 2 μl water droplet exhibited a CA of 136°. This strong hydrophobicity was attributed to the surface roughness induced by the morphology of GNS.

4. Conclusion

Highly porous, vertically oriented graphene nanosheets were synthesised using microwave plasma CVD. These graphene nanostructures were shown to have strong hydrophobicity. Nanosheets formed on the surface of the substrate appeared to be parallel to the substrate. Each vertically grown nanowall exhibited more than ∼15 layers of graphene stacked together. The AFM measurements of the surface of GNS showed that the wall thickness was approximately 15 nm. A considerably faster growth rate of GNSs was observed for ECR plasma technique. SEM and HR-TEM images reveal the formation of carbon nanosheet and Raman spectra show the defective nature of the film. This plasma CVD synthesis technique for GNS is very much suitable for energy storage, in particular for the anodes for Li thin film batteries.

References

1. E. Stratakis, S. Savva, D. Konios, C. Petridi and E. Kymakis, Nanoscale, 2014, 6, 6925–6931.
2. G. Kakavelakis, E. Strataki, D. Konios and E. Kymakis, Chem. Mater., 2014, 26, 5988–5993.
3. D. Konios, C. Petridis, G. Kakavelakis, M. Sygletou, K. Savva, E. Stratakis and E. Kymakis, Adv. Funct. Mater., 2015, 25, 2213–2221.
4. N. Balis, D. Konios, E. Stratakis and E. Kymakis, ChemNanoMat, 2015, 1, 346–352.
5. G. Viskadouros, D. Konios, E. Kymakis and E. Stratakis, Appl. Phys. Lett., 2014, 105, 203104.
6. J. Hassoun, F. Bonaccorso and B. Scrosati, et al., Nano Lett., 2014, 14, 4901–4906.
7. F. Schwierz, Nanotechnology, 2010, 5, 487–496.
8. C. Chung, Y. K. Kim, D. Shin and D. H. Min, et al., Acc. Chem. Res., 2013, 46, 2211–2224.
9. V. Tozzini and V. Pellegrini, Phys. Chem. Chem. Phys., 2013, 15, 80–89.
10. D. S. Mao, X. Wang, X. H. Liu, Q. Li and J. F. Xu, Diamond Relat. Mater., 2000, 9, 1876–1880.
11. C. Berger, Z. Song, X. Li, X. Wu, N. Brown and C. Naud, et al., Science, 2006, 312, 1191–1196.
12. L. Zhi and K. Mullen, Mater. Chem., 2008, 18, 1472–1484.
13. A. Reina, X. Jia, J. Ho, D. Nezich, H. Son and V. Bulovic, et al., Nano Lett., 2009, 9, 30–35.
14. S. J. Park and R. S. Ruoff, Nat. Nanotechnol., 2009, 4, 217–224.
15. S. Bae, H. Kim and S. Iijima, et al., Nat. Nanotechnol., 2010, 5, 574–578.
16. M. Yi and Z. Shen, J. Mater. Chem. A, 2015, 3, 11700–11715.
17. W. S. Hummers and R. E. Fffeman, J. Am. Chem. Soc., 1958, 80, 1339.
18. Y. Hernandez, V. Nicolosi and J. N. Coleman, et al., Nat. Nanotechnol., 2008, 3, 563–568.
19. K. V. Emtse, A. Bostwick and T. Seyller, et al., Nat. Mater., 2009, 8, 203–207.
20. S. Gilje, S. Han, K. L. Wang and R. B. Kaner, Nano Lett., 2007, 7, 3394–3398.
21. R. Thomas and G. M. Rao, Electrochim. Acta, 2014, 125, 380–385.
22. R. Thomas, K. Y. Rao and G. M. Rao, Electrochim. Acta, 2013, 108, 458–464.
23. B. Zheng, Y. Yang, J. Chen, K. Yu, J. Yan and K. Cen, Nanoscale, 2013, 5, 5180–5204.
24. Y. Y. Wang, S. Gupta and R. J. Nemanich, Appl. Phys. Lett., 2004, 85, 2601–2603.
25. Y. Wu, P. Qiao, T. Chong and Z. Shen, Adv. Mater., 2002, 14, 64–67.
26. S. Virendra, J. Daeha, Z. Lei, D. Soumen, I. K. Saifal and S. Sudipta, Prog. Mater. Sci., 2011, 56, 1178–1271.
27. R. Vitchev, A. Malesevic, R. H. Petrov, R. Kemps, M. Mertens and A. vanhulsel, et al., Nanotechnology, 2010, 21, 095602.
28. K. V. Deenamma and G. M. Rao, Rev. Sci. Instrum., 2000, 71, 467–471.
29. K. Yu, Z. Bo, G. Lu, S. Mao, S. Cui, Y. Zhu, X. Chen, R. S. Ruoff and J. Chen, Nanoscale Res. Lett., 2011, 6, 202.
30. K. Yu, P. Wang, G. Lu, K. H. Chen, Z. Bo and J. Chen, J. Phys. Chem. Lett., 2011, 2, 537–542.
31. Z. Bo, K. Yu, G. Lu, S. Cui, S. Mao and J. Chen, Energy Environ. Sci., 2011, 49, 1849–1858.
32. Y. Wu, P. Qiao, T. Chong and Z. Shen, Adv. Mater., 2002, 14, 64–67.
33. Y. Wu, Nano Lett., 2002, 2, 355–359.
34. S. H. Tsai, C. W. Chao, C. L. Lee and H. C. Shih, Appl. Phys. Lett., 1999, 74, 3462.
35. C. Bower, W. Zhu, S. Jin and O. Zhou, Appl. Phys. Lett., 2000, 77, 830.
36. Z. Wang, M. Shoji and H. Ogata, Appl. Surf. Sci., 2011, 257, 9082–9085.
37. A. Ural, Y. M. Li and H. J. Dai, Appl. Phys. Lett., 2002, 81, 3464–3466.
38. Y. Wu and B. Yang, Nano Lett., 2002, 2, 355–359.
39. S. Shimabukuro, Y. Hatakeyama, M. Takeuchi, T. Itoh and S. Nonomura, Thin Solid Films, 2008, 516, 710–713.
40. C. L. Cheng, C. T. Chia, C. C. Chiu and I. N. Lin, Diamond Relat. Mater., 2002, 11, 262–267.
41. Z. H. Ni, H. M. Fan, Y. P. Feng, Z. X. Shen, B. J. Yang and Y. H. Wu, J. Chem. Phys., 2006, 124, 204703.
42. R. Saito, A. Jorio, A. G. S. Filho, G. Dresselhaus, M. S. Dresselhaus and M. A. Pimenta, Phys. Rev. Lett., 2001, 88, 027401.
43. M. S. Dresselhaus, G. Dresselhaus and M. Hofmann, Philos. Trans. R. Soc., A, 2008, 366, 231–236.
44. R. J. Nemanich and S. A. Solin, Phys. Rev. B: Condens. Matter Mater. Phys., 1979, 20, 392–401.
45. M. A. Pimenta, G. Dresselhaus, M. S. Dresselhaus, L. A. Cancado, A. Jorio and R. Sato, Phys. Chem. Chem. Phys., 2007, 9, 1276–1291.
46. Y. Lin, G. J. Ehlert, C. Bukowsky and H. A. Sodano, ACS Appl. Mater. Interfaces, 2011, 3, 2200–2203.
47. B. Bhushan and E. K. Her, Langmuir, 2010, 26, 8207–8217.
48. M. J. Liu and L. Jiang, Adv. Funct. Mater., 2010, 20, 3753–3764.
49. F. M. Chang, S. J. Hong, Y. J. Sheng and H. K. Tsao, Appl. Phys. Lett., 2009, 95, 064102.

---