Conformal graphene coating on high-aspect ratio Si nanorod arrays by a vapor assisted method for field emitter

Abstract

Carbon materials like nanotubes and graphene have been previously used for field emission application due to their high emission currents and low turn-on voltages. However, in most cases, these devices show low reliability and poor endurance after a few hours of testing. The poor performance is usually attributed to lack of alignment, poor structure quality, and/or non-conformal coating. In this paper, a hybrid structure of graphene–silicon nanorod arrays (NRAs) was demonstrated by direct growth of self-crystallized graphene with Ni vapor-assisted growth via a conventional chemical vapor deposition (CVD) system. By carefully adjusting parameters and reducing the deposition rate, thicknesses of graphene layers can be systematically coated in a controllable manner, even on high aspect ratio surfaces such as aligned silicon NRAs. Detailed surface morphologies and microstructures of the graphene–NRAs core–shell hybrid structures were investigated. Findings in field emission measurements indicate that the graphene coating exhibits a remarkable enhancement by lowering the turn-on field, increasing the current density over 4 orders of magnitude, and greatly improving the endurance. The endurance test shows a stable current density of 1000 μA cm⁻² after more than 15 hours of operation under a constant applied high bias stress.

---

Introduction

Graphene possess many illustrious properties^1–3 due to its particular structure such as excellent mobility,^4,5 electromechanical properties,⁶ outstanding thermal conductivity,^7–9 and high current capacity^10–12 (∼10⁹ A cm⁻²). Graphene grown by chemical vapor deposition (CVD) providing a well crystallized structure is believed to be a promising material for devices operating under high current densities as interconnects and field emission (FE) sources.^11–14

Excellent field emission caused by electrons tunneling out of the emitter to vacuum highly relies on the geometry of the tip region, electron affinity, and conductivity.^15–19 Therefore, the shape morphology at tip region, the low electron affinity, the good conductivity, and the relevant stacking density can usually result in an excellent field emission. Some studies made use of CVD graphene combined with nanowires and nanotubes arrays were demonstrated to enhance the field emission behavior due to its large geometry enhancement.^17,20,21 Growth of vertically aligned few layers graphene (graphene walls) by plasma enhanced chemical vapor deposition (PECVD) was demonstrated, further showing improvement by lowing the turn-on field while major concerns as low emission current densities and current degradation under a constant applied field after few hours of continuous operation remind.^16,18,19 Alternatively, nanowires with hetero-structure core–shell configuration, manipulate the gradient of the electric potential on the surface of nanowires for electrons tunneling to vacuum, revealing an enhanced field emission performance.^21–23 The concept of the core–shell configuration has been applied using a SiO₂ shell layer coated on a SnO₂ nanowire in order to reduce the electron affinity, enhancing field emission property with a lower turn-on field.²²

However, limited transfer techniques of graphene layers make a smooth coating of CVD graphene on the surface of nanostructures with high aspect-ratio difficult. As a result, it is unable to concentrate the electric field, resulting in high turn-on field and low emission current.^14,24 In this regard, we demonstrate a conformal coating of the graphene layer on complex structures such as Si nanorods arrays (NRAs) with a transfer-free process to form graphene–NRAs core–shell hybrid structures for FE applications via direct growth of self-crystallized graphene with Ni vapor-assisted growth in a conventional chemical vapor deposition (CVD) system. By greatly reducing the deposition rate and adjusting the growth time, we are able to precisely control the thickness of the graphene coating layer. Different from other works, our CVD graphene on Si nanostructures plays as the main layer to provide lower electron affinity and better conductivity, resulting in the enhancement of the performance under continuous operation at large current densities. The quality of the graphene layer on Si NRAs was studied by Raman spectroscopy. Detailed surface morphologies and microstructures of graphene–NRAs core–shell hybrid structures were investigated. Findings in field emission measurements indicate that the graphene–NRAs core–shell hybrid structures has better performance than other FE devices made by CVD graphene^19,25 and similar to those made of other carbon structures such as carbon nanotubes (CNTs)^26,27 and reduced graphene oxide (r-GO), respectively.^22,23

Experimental section

Synthesis of Si nanorod arrays (Si NRAs) by chemical etching process

Si NRAs were fabricated on n⁺-Si (100) wafer (resistivity ∼ 0.001 ohm cm) after a standard RCA clean. Later, O₂ plasma for 60 s at 30 watt was applied to change the surface of the samples from hydrophobic to hydrophilic states, allowing the polystyrene (PS) nanospheres surface adsorption. Subsequently, 40 μl solution of PS nanospheres with diameter smaller than 1 μm (Ps03N, Bangs Laboratories) were evenly separated in a petri dish and self-assembled into a monolayer after mixing with 90 μl methanol. Polyoxyethylene(10)nonyl-phenyl-ether (NP-9, Alfa Aesar) diluted 1 : 30 000 was added to increase the density of PS nanospheres, following by immersing Si substrates into the solution to attach the PS nanospheres. O₂ plasma was used at 80 watts for 100 s to shrink the size of the PS nanospheres from 0.96 to 0.67 μm and samples were annealed at 110 °C for 30 minutes to increase the adhesion between the PS balls and she substrate. A 30–40 nm thick Au film was deposited by e-beam evaporator. Ultrasonic vibration was used to remove the PS nanospheres, resulting in Au nano-network. Consequently, the sample was immersed into 5 M HF mixed with 0.49 M H₂O₂ solution for 5 min. The etching occurred at only the interface between Au and Si, allowing the formation of vertically extended Si nanorod arrays.²¹ Finally, the Au nano-network was removed by aqua regia so that the pure Si NRAs can be obtained.

Conformal coating of graphene layer on Si NRAs by vapor assisted method

The pure Si NRAs samples were placed at the center of a low pressure CVD system. At the same time, a Ni ingot was placed on the upstream of the tube. Then, both substrate and Ni ingots were heated up to 1100 °C with H₂ 20 sccm and Ar 100 sccm in 9 × 10⁻¹ 10⁻¹ Torr for 40 min. The amount of methane (CH₄) flow was optimized to 20 sccm at a stable pressure of 60 Torr. The growing time was set to 20 min achieving 3–5 graphene layer coating.

Material characterization

A high-resolution Micro Raman Spectrometer (LabRaman 800, Horiba Jobin Yvon) equipped a laser λ = 514 nm and a power close to 1 mW. A 100× objective lens was used to reduce the spot size down to 1 μm. TEM samples were prepared by ultrasonic vibration from the as-prepared sample in ethanol for 3 minutes and dripped directly on a 300 mesh Cu grid supported by silicon monoxide lacey layer. JEOL 2010F HR-TEM at a 200 kV was used to observe the lattice information of graphene and Si NRAs, respectively. For field emission measurements, the sample was stuck on a copper stage that works as anode electrode and a stainless steel probe with a diameter of 2 mm works as a cathode. The chamber was evacuated to a base pressure of 10⁻⁵ Pa during measurements. The inter-electrode distance between the probe and graphene–Si NRAs array was fixed at 100 μm. The electrical bias was applied and the current–voltage (I–V) characteristics were recorded by a Keithley 237.

Results and discussion

Si NRAs were synthesized by the metal-assisted chemical etching process as shown in Fig. 1(a)–(f).²¹ First, polystyrene (PS) nanospheres with diameters smaller than 1 μ were deposited on n⁺-Si (100) and later the volume was shrunk by O₂ plasma treatment (Fig. 1a–c). The PS nanospheres work as the shielding maker during metal deposition as shown in Fig. 1(d). The uniform deposition of the PS nanospheres is essential for the uniform distribution and the plasma treatment can further control the diameter of the nanosphere, defining the diameter of the Si NRAs. Finally, the Si NRAs were formed after etching the wafer in HF solution and the metal layer was removed as shown in Fig. 1(e) and (f), respectively. Additional details of the process can be found in the experimental section. Fig. S1(a)† displays a top view SEM image for the Si NRAs. The Si NRAs are clearly arranged with diameters ranging from 350 to 420 nm and the average distance between Si NRAs is approximately ∼600 nm. After the etching process, the Si NRAs did not form a consummately circular shape; instead, the nanowires form a star-like shape due to the different lateral etching rate and oxidation of Si lattice during the etching process into the corrosive aqueous solution at a relatively slow etching rate. Fig. S1(b)† shows an SEM image with a closer view of the star-like shape of the Si NRAs arrays after the etching process. Notice that this is an actual desirable effect to improve the thermo-electrical performance of the device as shown by Hochbaum et al.²⁸ Subsequently, graphene was deposited on the Si NRAs by the direct growth of the self-crystallized graphene with the Ni vapor-assisted growth via conventional chemical vapor deposition (CVD) system.²⁹ Fig. 1(g) and (h) illustrate schematics of the direct growth of the self-crystallized graphene on Si NRAs. Detailed growth of the self-crystallized graphene on Si NRAs was addressed in the experimental section. Briefly, the Si NRAs were placed in the center of the main reactor while a Ni ingot was located at the upstream of the tube. Note the amount of methane (CH₄) flow were found to significantly influence the thickness of the graphene layer. For example, the growth time was set to 20 min to achieve 3–5 graphene coating layers with the methane flux of 20 sccm while thicknesses of the graphene layers increase to ∼10 nm and ∼25 nm at only 5 min growth with methane fluxes of 100 sccm and 50 sccm (Fig. S(2a) and S(2b)),† respectively.

Fig. 1 (a) to (f) show the process flows for the fabrication of the Si NRAs by Au-assisted chemical etching. (g) and (h) show a schematic diagram of the CVD chamber used for graphene coating onto the Si NRAs by the Ni vapor-assisted graphene deposition, respectively. The OM image inserted in (g) shows the top view of the Si NRAs.

Raman spectrometer was used to confirm the quality of the conformal coating of the graphene layer on the surface of Si NRAs. The Raman spectra of the Si NRAs before and after the graphene deposition for 20 min at the methane flux of 20 sccm were shown in Fig. 2(a). Obviously, after the vapor-assisted process, the spectra show the typical graphene features, for which D, G, and 2D bands located at 1327.4, 1603.5, and 2656.4 cm⁻¹ were measured, respectively. In addition, the strongest peak located at 520.6 cm⁻¹ and 938.8 to 993.0 cm⁻¹ denote the first and secondary order Raman signals of silicon, respectively.³⁰ The position and the full width at half maximum (FWHM) of Si peaks (520.6 cm⁻¹ and 938.8 to 993.0 cm⁻¹) remain unchanged before and after the graphene deposition. The FWHM of the 2D band is about ∼65 cm⁻¹ supporting the fact that the thickness of the conformal coating of the graphene layer is few layers rather than monolayer.³¹ Note that the broaden G band and blue-shift of G and 2D from the deposition on flat substrate (1597 and 2645 cm⁻¹ for G and 2D respectively)²⁹ are most likely resulted from the strain induced by the rough surface of the Si NRAs acting as a substrate for graphene growth after the HF etching process.³² The intensity ratio of the I_D/I_G of ∼2 implies that the vapor-assisted deposited graphene is mainly composed by small domains.^33,34

Fig. 2 (a) Raman spectra of the pure Si NRAs before and after graphene deposition. (b and c) The corresponding cross-sectional SEM images of Si NRAs before and after the coating of graphene with same scale bar.

Based on the intensity ratio of the I_D/I_G, the effective domain size of the deposited film was further estimated to be around ∼12 nm by using L_a (nm) = (2.4 × 10⁻¹⁰)λ_l⁴(I_D/I_G)⁻¹ where λ_l is the wavelength of the laser (λ = 514 nm) used for Raman measurements and L_a is the lateral domain size of the graphene layer.³⁵ The small intensity of the 2D band is the consequence of the strong graphene/substrate interaction and small domain size.^29,36,37 During the annealing process, the vaporous Ni atoms tend to be aggregated with each other by reducing its surface energy, resulting in formation of Ni clusters. The Ni clusters further accelerate the decomposition of the methane (CH₄) into carbon atoms and trigger the graphitization process by assembling carbon atoms into the graphene surrounding to Ni clusters, namely self-crystallization process. In addition, the small domain size of the graphene layer is most likely attributed to the polycrystalline structure of Ni clusters and the uncontrolled graphene nucleation.³⁴ Fig. 2(b) and (c) show the SEM cross-section images of the Si NRAs array before and after the graphene deposition, in which the average length of 1.8–2.0 μm and the diameter of 400 nm with aspect ratios between 4.0–5.0 were confirmed. Distinctly, the morphology of Si NRAs remains after the conformal coating of the graphene layer. The results support the fact that the crystalline structure and morphology of Si NRAs do not suffer major changes during the annealing process required for conformal coating of the graphene layer.

High resolution (HR) TEM images for Si NRAs and graphene–Si NRAs core–shell hybrid structures with different conformal coating thicknesses of the graphene layers are shown in Fig. 3. Fig. 3(a1) shows a TEM image of a pure Si NRAs prepared by metal (Au)-assisted chemical etching. Note that different contrast of the lines parallel to the etched direction from top to bottom denotes the different lateral etching rates. A thin oxide layer (SiO_x) of ∼4–5 nm surrounding to the Si NRAs was found as shown in Fig. 3(a2), which is much thicker than a common native oxide (1–2 nm).³⁸ The thicker SiO_x layer may be resulted from the etching process under the mixture of HF/H₂O₂ solution required during the metal-assisted etching. Fig. 3(b1) and 3(b2) show HRTEM images of the graphene–Si NRAs core–shell hybrid structure for 20 min growth. As can be seen, the conformal coating of the graphene layer with the thickness of ∼1–2 nm, corresponding to 3–5 graphene layers along the whole NRAs, was confirmed (Fig. 3(b2)). In addition, the thickness of the coating graphene layer can be also increased to ∼30 layers, which almost exhibits a linear dependency with the deposition time (Fig. S3†). Interestingly, after extending the deposition time to 150 min with same growth condition, a cap-like structure of graphene forms on the top of Si NRAs as shown in Fig. 3(c1). The cap-like structure of the graphene extends the diameter of the NRAs near 20% and the deposition rate becomes non-uniform at the top and bottom-side of the NRAs. In a close observation of this cap-like structure, some of the graphene sheets were found to be warped together to form graphene walls on the surface of the Si NRAs as shown in Fig. 3(c2). The formation of the cap-like structure and graphene walls is more likely due to the rough surface caused by the lateral etching during the formation of the NRAs. Fig. S4(a)† shows an SEM image that confirms the larger lateral etching close to the top compared to the base of the NRAs. The longer arrows denote the areas with larger lateral etching in a single Si NRAs. Fig. S4(b)† shows schematics comparing the surface morphology at different areas of the Si NRAs. The increased defective surface at the tip of the nanorod can act as the nucleation site to trigger a faster coating rate of the graphene layer on the tip region than other regions along the Si NRAs, namely the faster lateral growth. In addition, the increased defect density and strain observed in the film by Raman can result in the formation of small graphene walls out-of-plane after releasing of internal force along in-plane direction, possibly resulting in the formation of a cap-like structure thereby degradation of the conformal coating with the increased numbers of layers.³⁹

Fig. 3 (a1) and (a2) TEM and HRTEM images of Si NRAs before the coating of the graphene. (b1 and b2) TEM and HRTEM images of the graphene–Si NRAs core–shell hybrid structure after 20 minutes deposition with CH₄ = 20 sccm at 1100 °C and 60 Torr. (c1) to (c3) TEM and HRTEM images of the graphene–Si NRAs core–shell hybrid structure after the deposited time was increased to 150 min. A cap-like structure can be observed.

Notably, conformal coating of the graphene layer on the graphene–Si NRAs core–shell hybrid structure provides the low electron affinity and the good conductivity while the morphology of the Si NRAs has an important role of concentrating the electrical field.

To shed light the enhanced performance of the graphene–Si NRAs core–shell hybrid structure as emitter; field emission properties of the graphene–Si NRAs core–shell hybrid structures were measured. Details of field emission measurements were described in the experimental section. Fig. 4(a) shows the typical field emissive current densities, J (current/total area of probe) as the function of applied electric field E (bias/inter-electrode distance) for the graphene–Si NRAs core–shell hybrid structure with only 3–5 graphene coating layers. For comparison, field emission properties of the pure Si NRAs and the graphene layer on SiO₂ substrate were measured as shown in Fig. 4(a), respectively. The turn-on field (E_on) is defined as the electric field required for generating a current density of 0.01 mA cm⁻² while 0.1 mA cm⁻² is sufficient for operation of display panel devices. Therefore, the turn-on electric field, E_on, for the graphene–Si NRAs core–shell hybrid structure was found to be ∼8 V μm⁻¹ (Fig. S5(a)†). In addition, no obvious field emission behavior was observed for the flat graphene layer on the SiO₂ substrate and the pure Si NRAs without a conformal coating of the graphene layer (Fig. S5(b)†). The results confirm that the electrical field can be effectively enhanced by the NRAs structure, lowing the energy barrier to allow tunneling of the electrons.⁴⁰ A linear relationship, so-called ln(J/E²) − (1/E) plot as shown in inset of Fig. 4(a), indicates a field-emission behavior, which follows Fowler–Nordheim (F–N) relationship, i.e. electrons tunneling through a potential barrier and can be expressed as follows:

where J is emission current density, E is applied field, ϕ is the work function of emitter material, β is enhancement factor, A (1.56 (10⁻¹⁰ A V⁻² eV) and B (6.8 × 10³ eV^−3/2 V μm⁻¹) are constant. The field enhancement factor, β, reflects the degree of the field emission enhancement of the tip shape on a planar surface, which also depends on the geometry of the nanowire, the crystal structure, and the density at the emitting point. Consequently, the β of the graphene–Si NRAs core–shell hybrid structure can be determined to be 423.6 by the slope of the ln(J/E²) − (1/E) plot with a work function of the graphene (4.48 eV).⁴¹ Notably, the thinner graphene coating outperforms the thicker one. For thicker graphene, the greatly reduced turn-on field at 1.1 V μm⁻¹ was found while the J–E curve becomes unstable and was burned out where a flash was observed soon after at a low current density as shown in Fig. S6.† Although some other peaks located at 4.4 and 9.1 V μm⁻¹ were observed by further increasing the electric field, the results were not repeatable. The low turn-on field can be attributed to the enhanced of the local electric field due to the graphene nanowalls form at the top of the Si NRAs (Fig. 3c2). However, these standing graphene flakes cannot afford the large current and burn out. These results highlight the importance of the homogeneous coating observed under few layers, which is loosened by further increasing the number of layers, resulting in amorphous shapes as the cap-like and leading to the formation of single graphene walls growing out of plane. It is worthy to mention that pure Si NRAs does not show and stable emissive current under 11 V μm⁻¹ caused by the poor electron transport between layers along the c-axis. The lifetime, long-term emission under large current densities, is still an issue for field emission devices. Therefore, for practical applications, a stable emissive current at large current densities along the time is imperative. The corresponding the endurance test of graphene–Si NRAs core–shell hybrid structure is shown in Fig. 4(b). Obviously, the applied electrical field was fixed at 10 V μm⁻¹ and the relative emissive current density remains almost unchanged at very high current, ∼2600 μA cm⁻² after continuous operation for 15 hours. The findings show a large enhancement of the performance under continuous emission operation compared to similar devices.

Fig. 4 (a) J–E measurements of Si NRAs, graphene, and graphene–Si NRAs core–shell hybrid structure. The insert shows an F–N plot of the graphene–Si NRAs core–shell hybrid structure. (b) Endurance test of the graphene–Si NRAs core–shell hybrid structure.

To shed light on this part, Fig. 5 compares the performance of the vapor-assisted graphene coating with other works prepared using CVD graphene^19,25 based on the time and current density used for continuous emission as emission layer and similar to those prepared with CNTs,^26,27 r-GO,^22,23 hybrid structures as CVD graphene/CNTs,²⁰ and other materials used for field emission as ZnO and Boron wires, respectively. The solid line observed in some cases represents the current decay during the emission test while single symbol with no line indicates stable emission without obvious current degradation. As can be seen, the stability of graphene–Si NRAs hybrid structure is comparable to emitter using r-GO deposited by electrophoretic methods from Wu et al.²³ and double wall CNTs with STO and PMMA from Pandey et al.²⁷ We also compared the results with other field emission devices prepared with ZnO and B.^42–44 However, the endurance tests were held for a short time and/or the current densities were quite low compared to our endurance test. As a result, our research demonstrated a potentially useful approach for other material/substrate architecture systems to significantly improve the field emission properties.

Fig. 5 Performance comparison with other carbon based FE devices. The x-axis represents the time used for the continuous operation test while y-axis shows the current density used for the test. Solid line indicates unstable endurance during continuous operation.

Conclusions

A novel approach to coat graphene on high aspect ratio surfaces as Si NRAs was demonstrated. By using Ni vapor-assisted deposition, graphene layer can be uniformly coated in structures down to nanometer scale with good coverage. Excellent field emission performance of this graphene–Si NRAs hybrid structure was demonstrated. For FE applications, the performance of the Si NRAs array was greatly improved after coating of graphene using the vapor-assisted method. The turn-on voltage of graphene–Si NRAs hybrid structure is ∼8 V μm⁻¹ and showing current densities up to ∼3600 μA cm⁻², which is much higher than that of intrinsic Si NRAs and flat graphene. In addition, the uniformity of the coating plays a key role in the large performance enhancement even after 15 hours of continuous operation. This method has a great potential and paves the way to a new series of applications for graphene that requires a conformal graphene coating in structures up to nanometer scale with ultimate good contact.

Acknowledgements

The research was supported by the Ministry of Science and Technology through Grant no. 101-2112-M-007-015-MY3, 101-2218-E-007-009-MY3, 102-2633-M-007-002, and National Tsing Hua University through Grant no. 102N2022E1. Y. L. Chueh greatly appreciates the use of facility at CNMM the National Tsing Hua University through Grant no. 102N2744E1.

References

1. A. K. Geim, Science, 2009, 324, 1530.
2. Y.-L. Zhang, Q.-D. Chen, Z. Jin, E. Kim and H.-B. Sun, Nanoscale, 2012, 4, 4858.
3. Y.-L. Zhang, L. Guo, H. Xia, Q.-D. Chen, J. Feng and H.-B. Sun, Adv. Opt. Mater., 2014, 2, 10.
4. K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, M. I. Katsnelson, I. V. Grigorieva, S. V. Dubonos and A. A. Firsov, Nature, 2005, 438, 197.
5. K. I. Bolotin, K. J. Sikes, Z. Jiang, M. Klima, G. Fudenberg, J. Hone, P. Kim and H. L. Stormer, Solid State Commun., 2008, 146, 351.
6. S.-H. Pan, H. Medina, L.-J. Chou, Z. Wang, K.-H. Chen, L.-C. Chen and Y.-L. Chueh. Nanoscale,  10.1039/c4nr00495g.
7. A. A. Balandin, S. Ghosh, W. Bao, I. Calizo, D. Teweldebrhan, F. Miao and C. N. Lau, Nano Lett., 2008, 8, 902.
8. D. L. Nika, S. Ghosh, E. P. Pokatilov and A. A. Balandin, Appl. Phys. Lett., 2009, 94, 203103.
9. M. Freitag, M. Steiner, Y. Martin, V. Perebeinos, Z. Chen, J. C. Tsang and P. Avouris, Nano Lett., 2009, 9, 1883.
10. A. D. Liao, J. Z. Wu, X. Wang, K. Tahy, D. Jena, H. Dai and E. Pop, Phys. Rev. Lett., 2011, 106, 256801.
11. Q. Shao, G. Liu, D. Teweldebrhan and A. A. Balandin, Appl. Phys. Lett., 2008, 92, 202108.
12. C.-H. Yeh, H. Medina, C.-C. Lu, K.-P. Huang, Z. Liu, K. Suenaga and P.-W. Chiu, ACS Nano, 2013, 8, 275.
13. J. Robertson, G. Zhong, S. Esconjauregui, C. Zhang and S. Hofmann, Microelectron. Eng., 2013, 107, 210.
14. S. Lee, S. Lee and E.-H. Yang, Nanoscale Res. Lett., 2009, 4, 1218.
15. J. O. Hwang, D. H. Lee, J. Y. Kim, T. H. Han, B. H. Kim, M. Park, K. No and S. O. Kim, J. Mater. Chem., 2011, 21, 3432.
16. A. Malesevic, R. Kemps, A. Vanhulsel, M. P. Chowdhury and A. Volodin, J. Appl. Phys., 2008, 104, 084301.
17. W. T. Zheng, Y. M. Ho, H. W. Tian, M. Wen, J. L. Qi and Y. A. Li, J. Phys. Chem. C, 2009, 113, 9164.
18. J. L. Qi, X. Wang, W. T. Zheng, H. W. Tian, C. Q. Hu and Y. S. Peng, J. Phys. D: Appl. Phys., 2010, 43, 055302.
19. N. Soin, S. Roy, K. S. Hazra, D. S. Misra and T. H. Lim, J. Phys. Chem. C, 2011, 115, 5366.
20. I. Lahiri, V. P. Verma and W. Choi, Carbon, 2011, 49, 1614.
21. Z. Huang, X. Zhang, M. Reiche, L. Liu, W. Lee, T. Shimizu, S. Senz and U. Gösele, Nano Lett., 2008, 8, 3046.
22. J. M. Wu and and C.-H. Kuo, J. Phys. D: Appl. Phys., 2009, 42, 125401.
23. Z.-S. Wu, S. Pei, W. Ren, D. Tang, L. Gao, B. Liu, F. Li, C. Liu and H.-M. Cheng, Adv. Mater., 2009, 21, 1756.
24. L. Gao, G.-X. Ni, Y. Liu, B. Liu, A. H. Castro Neto and K. P. Loh, Nature, 2014, 505, 190.
25. C.-K. Huang, Y. Ou, Y. Bie, Q. Zhao and D. Yu, Appl. Phys. Lett., 2011, 98, 263104.
26. S. I. Jung, S. H. Jo, H. S. Moon, J. M. Kim, D.-S. Zang and C. J. Lee, J. Phys. Chem. C, 2007, 111, 4175.
27. A. Pandey, A. Prasad, J. P. Moscatello, M. Engelhard, C. Wang and Y. K. Yap, ACS Nano, 2012, 7, 117.
28. A. I. Hochbaum, R. Chen, R. D. Delgado, W. Liang, E. C. Garnett, M. Najarian, A. Majumdar and P. Yang, Nature, 2008, 451, 163.
29. W.-C. Yen, Y.-Z. Chen, C.-H. Yeh, J.-H. He, P.-W. Chiu and Y.-L. Chueh, Sci. Rep., 2014, 4, 4739.
30. J. H. Parker, D. W. Feldman and M. Ashkin, Phy. Rev., 1967, 155, 712.
31. Y. Y. Wang, Z. H. Ni, T. Yu, Z. X. Shen, H. M. Wang, Y. H. Wu and W. Chen, J. Phys. Chem. C, 2008, 112, 10637.
32. T. M. G. Mohiuddin, A. Lombardo, R. R Nair, A. Bonetti, G. Savini, R. Jalil, N. Bonini, D. M. Basko, C. Galiotis, N. Marzari, K. S. Novoselov, A. K. Geim and A. C. Ferrari, Phys. Rev. B: Condens. Matter Mater. Phys., 2009, 79, 205433.
33. A. C. Ferrari, Solid State Commun., 2007, 143, 47.
34. A. C. Ferrari and D. M. Basko, Nat. Nanotechnol., 2013, 8, 235.
35. L. G. Cançado, K. Takai, T. Enoki, M. Endo, Y. A. Kim, H. Mizusaki, A. Jorio, L. N. Coelho, R. Magalhães-Paniago and M. A. Pimenta, Appl. Phys. Lett., 2006, 88, 163106.
36. P.-Y. Teng, C.-C. Lu, K. Akiyama-Hasegawa, Y.-C. Lin, C.-H. Yeh, K. Suenaga and P.-W. Chiu, Nano Lett., 2012, 12, 1379.
37. H. Medina, Y.-C. Lin, C. Jin, C.-C. Lu, C.-H. Yeh, K.-P. Huang, K. Suenaga, J. Robertson and P.-W. Chiu, Adv. Funct. Mater., 2012, 22, 2123.
38. M. Morita, T. Ohmi, E. Hasegawa, M. Kawakami and M. Ohwada, J. Appl. Phys., 1990, 68, 1272.
39. L. Jiang, T. Yang, F. Liu, J. Dong, Z. Yao, C. Shen, S. Deng, N. Xu, Y. Liu and H.-J. Gao, Adv. Mater., 2013, 25, 250.
40. X. Fang, Y. Bando, U. K. Gautam, C. Ye and D. Golberg, J. Mater. Chem., 2008, 18, 509.
41. G. Giovannetti, P. A. Khomyakov, G. Brocks, V. M. Karpan, J. van den Brink and P. J. Kelly, Phys. Rev. Lett., 2008, 101, 026803.
42. F. Liu, H. Gan, D.-M. Tang, Y. Cao, X. Mo, J. Chen, S. Deng, N. Xu, D. Golberg and Y. Bando, Small, 2014, 10, 685.
43. D. Pradhan, M. Kumar, Y. Ando and K. T. Leung, Nanotechnology, 2008, 19, 035603.
44. A. Wei, X. W. Sun, C. X. Xu, Z. L. Dong, M. B. Yu and W. Huang, Appl. Phys. Lett., 2006, 88, 213102.

---

Footnote

† Electronic supplementary information (ESI) available: Top-view SEM images of as fabricated Si NRAs; HRTEM images of graphene film grown at CH₄ = 100 and 50 sccm; number of graphene coated layers vs. growing time at the controlled deposition rate; a schematic diagram of the surface roughness of the Si NRAs after the metal-assisted etching process; J–E measurement thicker graphene coated on Si NRAs for 150 min. See DOI: 10.1039/c4ra03310h

---