Scalable ZnO nanotube arrays grown on CVD ...

Scalable ZnO nanotube arrays grown on CVD-graphene films J. B. Park, H. Oh, J. Park, N.-J. Kim, H. Yoon, and G.-C. Yi Citation: APL Mater. 4, 106104 (2016); doi: 10.1063/1.4964490 View online: http://dx.doi.org/10.1063/1.4964490 View Table of Contents: http://scitation.aip.org/content/aip/journal/aplmater/4/10?ver=pdfcov Published by the AIP Publishing Articles you may be interested in Study on the structural and physical properties of ZnO nanowire arrays grown via electrochemical and hydrothermal depositions J. Appl. Phys. 110, 094310 (2011); 10.1063/1.3657843 Fe solubility, growth mechanism, and luminescence of Fe doped ZnO nanowires and nanorods grown by evaporation-deposition J. Appl. Phys. 110, 014317 (2011); 10.1063/1.3609073 Effect of annealing on the structural and luminescent properties of ZnO nanorod arrays grown at low temperature J. Appl. Phys. 109, 103508 (2011); 10.1063/1.3586243 Indirect optical transition due to surface band bending in ZnO nanotubes J. Appl. Phys. 108, 103513 (2010); 10.1063/1.3511345 Trimming of aqueous chemically grown ZnO nanorods into ZnO nanotubes and their comparative optical properties Appl. Phys. Lett. 95, 073114 (2009); 10.1063/1.3211124

Reuse of AIP Publishing content is subject to the terms at: https://publishing.aip.org/authors/rights-and-permissions. Download to IP: 192.30.84.210 On: Fri, 07 Oct 2016 17:40:05

APL MATERIALS 4, 106104 (2016)

Scalable ZnO nanotube arrays grown on CVD-graphene films J. B. Park,1 H. Oh,1 J. Park,1 N.-J. Kim,2 H. Yoon,1 and G.-C. Yi1,a 1

Department of Physics and Astronomy, Institute of Applied Physics, and Research Institute of Advanced Materials (RIAM), Seoul National University, Seoul 151-747, South Korea 2 Department of Physics and Chemistry, Korea Military Academy, Seoul 139-799, South Korea

(Received 12 May 2016; accepted 26 September 2016; published online 6 October 2016) We report the growth of wafer-scale arrays of individually position-controlled and vertically aligned ZnO nanotube arrays on graphene deposited by chemical vapor deposition (CVD-graphene). Introducing two-dimensional layered materials such as graphene as a growth buffer has recently been suggested for growing nanomaterials on traditionally incompatible substrates. However, their growth has been restricted to small areas or had limited controllability. Here, we study the distinct growth behavior of ZnO on CVD-graphene that makes the selective area growth of individual nanostructures on its surface difficult, and propose a set of methods to overcome this. The resulting nanotube arrays, as examined by scanning electron microscopy and transmission electron microscopy, exhibited uniform morphologies and high structural quality over a large area and could be prepared on a broad variety of substrates, including amorphous, metallic, or flexible substrates. C 2016 Author(s). All article content, except where otherwise noted, is licensed under a Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). [http://dx.doi.org/10.1063/1.4964490]

Bottom-up growth of one-dimensional (1D) semiconductor nanostructures has been extensively studied for fabricating high-density and high-aspect-ratio nanodevices including sensors, lightemitting diodes, transistors, and photovoltaic cells.1–3 In particular, individually position-controlled and vertically aligned 1D nanostructures are desirable for the practical manipulation of these nanodevices. While this can be achieved with a number of specific materials on single-crystalline substrates such as sapphire and silicon wafers,3–5 expanding this to more general material/substrate combinations remains challenging due to limitations in growth compatibility. For example, a direct growth on amorphous oxides or metals can often benefit nanoelectronic or optoelectronic applications,6–8 but is difficult to achieve in a well-controlled manner.9–11 Recently, the growth of 1D nanostructures on two-dimensional (2D) nanomaterials such as graphene has been studied as a method to allow the preparation of vertically aligned 1D nanostructures on traditionally incompatible substrates.12–14 There, the 2D nanomaterial acts as a growth buffer layer that can be mechanically attached to, and also detached from, arbitrary substrates. However, while position-controlled and vertically aligned growth has been performed on mechanically exfoliated graphene, this approach is not scalable to a wafer-sized area.15 On the other hand, previous reports of growth on graphene deposited by chemical vapor deposition (CVD-graphene) have shown unsatisfactory growth controllability, with a lack of individual position control (i.e., a bundled growth), low growth selectivity, or poor vertical alignment.16,17 Herein, we report a method to significantly enhance the growth controllability on CVD-graphene, enabling a wafer-scale preparation of individually position-controlled and vertically aligned 1D ZnO nanostructure arrays on arbitrary substrates including amorphous, metallic, or flexible substrates.

a Author to whom correspondence should be addressed. Electronic mail: [email protected]

2166-532X/2016/4(10)/106104/7

4, 106104-1

© Author(s) 2016.


106104-2

Park et al.

APL Mater. 4, 106104 (2016)

The growth behaviors of semiconductor nanostructures on exfoliated and CVD-graphene are drastically different, which has led to different strategies for achieving position-controlled growth. In previous work reporting ZnO nanotube arrays grown using metal-organic vapor-phase epitaxy (MOVPE) on exfoliated graphene,15 the mechanically cleaved surface of graphene was shown to exhibit a small density of nucleation sites, with slight growth occurring presumably along the step edges of the graphene films (Fig. 1(a)). Hence, nucleation sites were artificially introduced by selective oxygen (O2) plasma treatment to grow the nanomaterial with a desired density.18 On the other hand, the same growth conditions applied to CVD-graphene led to a significantly different growth behavior (Fig. 1(b)). Comparing the SEM images of ZnO nanostructures grown at 600 ◦C on the two graphene films, the density and morphology of ZnO nanostructures were both affected, with a row of ZnO nanoneedles on the exfoliated graphene film (Fig. 1(a)) and ZnO nanowall structures on the entire surface of the CVD-graphene film (Fig. 1(b)). This difference in growth behavior likely resulted from a much higher density of step edges and grain boundaries in CVD-graphene compared with exfoliated graphene, as evidenced by their differences in surface roughness (AFM data in Fig. S1 of the supplementary material) and structural defects (Raman spectra in Fig. S2 of the supplementary material). The high density of nucleation on CVD-graphene implies that selective suppression, rather than selective creation, of nucleation sites is needed to achieve position-controlled growth, as opposed to the case of exfoliated graphene. The surface morphology of ZnO nanostructures grown on CVD-graphene films depends significantly on the substrate temperature during the growth process. Figures 1(c)–1(f) show SEM images of ZnO nanostructures grown at 500, 600, 700, and 800 ◦C for 30 min on CVD-graphene films transferred onto SiO2/Si substrates. At a low growth temperature of 500 ◦C, the resulting nanostructures were a mixture of 20-nm-high nanowall structures and nanodots with a diameter of 20 ± 3 nm. The morphology of the nanostructures significantly changed with increasing growth temperature. At 600 ◦C, the ZnO products consisted of 60-nm-high nanowall structures, and no nanodots. Nanostructures grown at 700 ◦C comprised 120-nm-high nanowall structures mixed with 120 ± 30-nm-high nanoneedles. When the growth temperature was further increased to 800 ◦C, nanoneedles with a wide range of diameters of 20–100 nm were obtained. These results indicate that lower growth temperatures yielded nanowalls, and the general shape of the nanostructures

FIG. 1. 30◦-tiled scanning electron microscopy (SEM) images showing different nucleation densities of ZnO nanostructures on (a) mechanically exfoliated-graphene and (b) chemical vapor deposition (CVD)-graphene films at 600 ◦C of growth temperature. ((c)–(f)) 30◦-tiled SEM images of ZnO nanostructures grown on CVD-graphene films at different growth temperatures of (c) 500, (d) 600, (e) 700, and (f) 800 ◦C.


106104-3

Park et al.

APL Mater. 4, 106104 (2016)

slowly changed into nanoneedles with increasing growth temperature. This was presumably due to an inhomogeneous nucleation rate at higher temperatures caused by the rough surface of the CVD-graphene films. In subsequent experiments, growth parameters that maximized the coverage of nanowall structures (600 ◦C) were used for the selective growth of ZnO nanotube arrays, in a similar manner to previous reports.19 Position- and dimension-controlled growth of ZnO nanotubes was performed by selectively suppressing the nucleation of ZnO using a growth mask (Fig. 2(a)). First, a few-layered CVDgraphene film was transferred onto a supporting substrate; we show the results on SiO2/Si substrates in the figures, but arbitrary substrates with sufficient thermal tolerance, such as metallic or ceramic substrates, were also used and provided nearly identical results (Fig. S3 of the supplementary material). Next, a thin SiO2 film was deposited onto the graphene films using plasma-enhanced CVD (PECVD). The thickness was typically 10–200 nm; a 50-nm-thick film was used in the following experiments. Importantly, the oxide layer was annealed at 600 ◦C in oxygen (O2) before the patterning process to reduce the defects in the as-deposited SiO2 film that caused undesired growth and reduced the growth selectivity (Fig. S4 of the supplementary material). We note that this step was not required when the same SiO2 film was used for selective growth on GaN films, implying that the defects may have originated from the deposition of the SiO2 film on a defect-rich CVD-graphene film. Next, hole patterns on the growth mask were defined by either nanoimprint

FIG. 2. (a) Schematic illustrations of the process used to grow ZnO nanostructures on substrates coated with CVD-graphenefilms. (b) A photograph of ZnO nanotube arrays grown on a 2-in. SiO2 wafer. Inset: SEM images of ZnO nanotubes of different magnitudes.


106104-4

Park et al.

APL Mater. 4, 106104 (2016)

lithography (NIL) or e-beam lithography (EBL), where NIL allows a high-throughput wafer-scale process (Fig. S5 of the supplementary material) and EBL enables fine-diameter nanostructure growth. The typical diameters and pitches of the holes were 100–500 nm and 1–8 µm, respectively. After lithography, the SiO2 film was first etched by reactive-ion etching (RIE) using CF4 plasma until a very thin oxide film was left to prevent damage to the graphene by the plasma. The residual oxide layer was then completely removed by a brief wet etch using buffered oxide etchant (BOE). This mixed etching process provided better reproducibility for the preparation of a patterned oxide growth mask on the CVD-graphene films, as opposed to using a dry- or wet-etch alone. Then, ZnO nanotube arrays were selectively grown on SiO2-mask-patterned CVD-graphene films by catalyst-free MOVPE. The resulting product was a wafer-scale array of position-controlled and vertically aligned ZnO nanotubes. Figure 2(b) shows a photograph and corresponding SEM images of ZnO nanotube arrays prepared on the entire area of a 2-in.-diameter SiO2/Si wafer. Position-controlled nanostructures with designed dimensions were grown by varying the lithographic pattern and the growth duration. In this way, nanotubes of various dimensions, line arrays of nanowalls or complex-shaped letters formed by nanowalls (Fig. S6 of the supplementary material) were prepared. Combined with the diverse choice of substrates (Fig. S3 of the supplementary material), CVD-graphene films may provide a universal seed layer for large-area growth of individually position-controlled and vertically aligned ZnO nanotube arrays for various applications. The microstructural characteristics of ZnO nanotubes grown on CVD-graphene films and the interface between them were examined by cross-sectional transmission electron microscopy (TEM). For this analysis, a TEM specimen was fabricated from a ZnO nanotube/graphene/SiO2/Si sample using a focused-ion beam (FIB). Figure 3(a) shows a bright field (BF) cross-sectional image of a ZnO nanotube grown on a CVD-graphene film with an SiO2 mask, revealing the selective growth mechanism. The patterned SiO2 mask film was not only used to define the positions of the ZnO nanostructure array but also to effectively control undesired lateral growth at the bottom edge of the nanotubes. There was an undercut in the SiO2 mask, presumably formed by wet etching, and ZnO also grew in this region. However, further growth of this ZnO product was blocked by the upper SiO2 layer as shown in the TEM image, resulting in sharply defined nanostructures without undesired basal structures. This is in contrast to nanotubes grown on exfoliated graphene following selective O2 plasma treatment,15 where lateral tailing of ZnO reaching micrometer scales was observed at the base. The crystallographic orientation of the ZnO nanostructures was further investigated using high-resolution (HR) TEM imaging and fast Fourier transform (FFT) patterns. Figure 3(b) shows an HR-TEM image of the interfacial region indicated in Fig. 3(a); an ordered atomic structure was observed for ZnO in the initial stage of growth on CVD-graphene. Its corresponding FFT pattern (Fig. 3(c)) reveals a (0002) crystal orientation of ZnO parallel to the (0002)

FIG. 3. Microstructural characteristics of ZnO nanotubes. (a) Cross-sectional transmission electron microscopy (TEM) bright-field (BF) image of ZnO nanotubes on CVD-graphene films. The white and yellow dashed lines indicate SiO2 mask and CVD-graphene regions, respectively. (b) high-resolution (HR) image of the ZnO nanotube taken from the region indicated by the red box in (a). ((c) and (d)) Fast Fourier transform (FFT) patterns obtained from the HR image in the regions indicated by boxes in (b). Inset of (b): FFT pattern obtained from entire HR image of (b).


106104-5

Park et al.

APL Mater. 4, 106104 (2016)

plane of graphene (Fig. 3(d)). The FFT pattern at the interface shown in the inset of Fig. 3(b) also supports the c-aligned vertical growth. The graphene growth buffer layer also served as a buffer for mechanical lift-off20 to allow the transfer of 1D nanostructures obtained by MOVPE to thermally unstable substrates, for instance, to create flexible nanodevices. This is illustrated in Fig. 4(a), in which a thick polymeric polyimide (PI) film, spin-coated as a supporting layer, is detached from the surface together with ZnO nanotubes and graphene using punched a Kapton tape. The ZnO nanotubes on graphene can then be attached to arbitrary substrates including plastic films, after which the nanostructures are exposed by O2 plasma etching of the supporting PI layer to a depth of a few micrometers. Figure 4(b) shows a photograph of ZnO nanotubes on graphene after the mechanical lift-off from the original substrate. The punched tape was used for detaching and attaching the nanomaterials and to provide a mechanical handle for the freestanding 1D/2D hybrid system; the freestanding nanomaterial was supported by a 3-µm-thick polymeric film and was handled using the tape (ca. 50-µm-thick). The thin supporting layer allowed extremely bent ZnO nanotube arrays at a radius of curvature of 0.5 mm as shown in Figs. 4(c) and 4(d). The images suggest that no mechanical damage or fracture occurred during the bending test. This exfoliation method to fabricate a flexible nanosystem may provide a template for individually position-controlled and vertically aligned nanodevice arrays, which can be directly applied to flexible LEDs or touch sensors.21,22 Large-scale, precisely-controlled 1D ZnO nanotube arrays grown directly on CVD-graphene films offer a variety of flexible device applications, which have been difficult to achieved with randomly grown nanorods or controlled nanostructure arrays grown on a small-area substrate.15,21 For example, the position-controlled nanotube arrays can serve as a template for heteroepitaxial coating for flexible, large-scale GaN/ZnO nanoarchitecture LEDs.23 With precisely controlled shape and spacing between each nanostructure, we could control the chemical composition and thickness of quantum layers, which determines the emission color emitted from the LED nanorods.24 Accordingly, these position- and morphology-controlled nanostructures on large-area graphene films can potentially be used for flexible full-color displays. Compared with organic LEDs, the novel inorganic devices could be used for head-up automobile displays for use in harsh environments. In addition, 1D ZnO nanostructures could be used as piezoelectric mediators for pressure sensors.22,25,26 Our wafer-scale ordered nanotube array offers higher-resolution piezoelectric sensors.

FIG. 4. (a) Schematic illustration of the lift-off procedure of nanotube arrays through a simple exfoliation using the conventional polymer tape. (b) A photograph of bent nanotube arrays. ((c) and (d)) SEM images of transferred ZnO nanotubes at a bending radius of 0.5 mm (c) over a wide view and (d) at high magnification at the indicated area.


106104-6

Park et al.

APL Mater. 4, 106104 (2016)

Moreover, by using CVD-graphene as a growth buffer layer, a precisely controlled hybrid nanosystem can be prepared on an arbitrary substrate including conventional Si-electronics. The atomically thin graphene layers offer an epitaxial relation to the semiconductor crystal. Consequently, by inserting an intermediate graphene layer, high-quality crystal can be grown on the amorphous oxide or polycrystalline-metal at top surfaces of Si-electronics, which hinder elaborate growth of semiconductor nanostructures.7 In summary, we have demonstrated the growth of individually position-controlled and vertically aligned ZnO nanotubes on arbitrary wafer-scale substrates using catalyst-free MOVPE with a CVDgraphene film as an intermediate layer. The ZnO nanostructures had a precisely controlled shapes and highly crystalline structures as examined by SEM and TEM. Additionally, the nanostructures were easily detached with negligible damage from the host substrate using a simple mechanical lift-off to form flexible nanomaterial systems. Our method presents a versatile means for the preparation of 1D nanostructures as functional components in various applications by allowing vertically aligned growth of 1D nanostructures even on amorphous, metallic, or flexible substrate materials, with individually controlled positions and designed dimensions over a wafer-scale area. We believe that this simple method may be expanded to create more sophisticated nanoarchitectures required for various nanodevice applications such as nanosensors, photovoltaic cells, and memory. CVD growth of the graphene films. Large-area, multi-layered graphene films were synthesized on Cu foil using the CVD method. First, the Cu foil surface was cleaned using acetone and IPA and inserted into a tubular quartz tube and heated to 1030 ◦C with an H2 flow at 100 standard cubic centimeters per minute (SCCM) at 200 Torr. After reaching 1030 ◦C, the Cu foil was annealed for 15 min to coarsen the grain, while maintaining the flow rate and reactor pressure. Then, graphene films were grown on the Cu foil for 130 min under a mixture of CH4 and H2 at flow rates of 10 and 100 SCCM, respectively. During growth, the reactor pressure was maintained at 220 Torr. Finally, the sample was cooled to room temperature under flowing H2 at a pressure of 200 Torr. Forming the growth mask layer. A thin SiO2 layer was deposited onto the as-transferred CVDgraphene layer using a commercial system PECVD installed at the Inter-university Semiconductor Research Center (ISRC) at Seoul National University. The thickness of the oxide film was typically 10–200 nm, and the experiments reported herein were performed with a 50-nm-thick film. The oxide layer was annealed at 600 ◦C in N2 before the patterning process to reduce the number of defects in the as-deposited SiO2 film that would cause undesired growth and reduce growth selectivity. Next, hole-patterns at the growth-mask were defined by either NIL or EBL, where NIL readily allows a wafer-scale process and EBL enables growth of fine-diameter nanostructures. After lithography, the SiO2 film was etched using dry-etching, followed by wet etching using CF4 plasma and BOE. The residual oxide layer on the graphene after dry etching was completely removed using BOE. MOVPE growth of the ZnO nanotube array. ZnO nanotubes were selectively grown on a graphene layer using catalyst-free MOVPE. Diethylzinc (DEZn) and high-purity O2 (>99.9999%) were used as reactants, and high-purity Ar (>99.9999%) as the carrier gas. The flow rates of DEZn and O2 were 15–30 and 70–90 SCCM, respectively. During growth, Ar flowed into the quartz reactor through the bubbler with a DEZn bubbler temperature of −15 ◦C. To prevent premature reaction, the O2 gas line was separated from the main gas manifold line. The reactor pressure was kept at 0.3 Torr during growth, and the temperature ranged from 580 to 700 ◦C. Surface morphology and microstructure characterization. The morphology and crystal structural characteristics of the ZnO nanotubes were investigated using field-emission SEM (FE-SEM; TESCAN) and TEM (Tecnai F20). TEM specimen was prepared by using FIB (Helios 650) installed at the National Center for Inter-university Research Facilities (NCIRF) at Seoul National University. See supplementary material for data related to this article including AFM and Raman shift characteristics of graphene films and detailed results of ZnO nanostructure growth. This work was financially supported by the Samsung Research Funding Center of Samsung Electronics (No. SRFC-TA1503-05). 1 2

X. D. Wang, C. J. Summers, and Z. L. Wang, Nano Lett. 4, 423 (2004). W. Lu and C. M. Lieber, Nat. Mater. 6, 841 (2007).


106104-7

Park et al.

APL Mater. 4, 106104 (2016)

3

S. Roddaro, K. Nilsson, G. Astromskas, L. Samuelson, L.-E. Wernersson, O. Karlstrom, and A. Wacker, Appl. Phys. Lett. 92, 3509 (2008). 4 K. Tomioka, M. Yoshimura, and T. Fukui, Nature 488, 189 (2012). 5 A. I. Hochbaum, R. Fan, R. R. He, and P. D. Yang, Nano Lett. 5, 457 (2005). 6 R. Chen, T. T. D. Tran, K. W. Ng, W. S. Ko, L. C. Chuang, F. G. Sedgwick, and C. Chang-Hasnain, Nat. Photonics 5, 170 (2011). 7 F. L. Lu, T. T. D. Tran, W. S. Ko, K. W. Ng, R. Chen, and C. Chang-Hasnain, Opt. Express 20, 12171 (2012). 8 Q. Liao, Y. Huang, Z. Qin, Z. Zhang, J. Qi, Y. Zhang, and L. Xia, J. Nanosci. Nanotechnol. 12, 775 (2012). 9 X. J. Feng, K. Shankar, O. K. Varghese, M. Paulose, T. J. Latempa, and C. A. Grimes, Nano Lett. 8, 3781 (2008). 10 J. A. Jiang, Y. Y. Li, J. P. Liu, and X. T. Huang, Nanoscale 3, 45 (2011). 11 M. W. Ahn, K. S. Park, J. H. Heo, J. G. Park, D. W. Kim, K. J. Choi, J. H. Lee, and S. H. Hong, Appl. Phys. Lett. 93, 263103 (2008). 12 P. K. Mohseni, A. Behnam, J. D. Wood, C. D. English, J. W. Lyding, E. Pop, and X. L. Li, Nano Lett. 13, 1153 (2013). 13 Y. J. Hong, J. W. Yang, W. H. Lee, R. S. Ruoff, K. S. Kim, and T. Fukui, Adv. Mater. 25, 6847 (2013). 14 H. Park, S. Chang, J. Jean, J. J. Cheng, P. T. Araujo, M. S. Wang, M. G. Bawendi, M. S. Dresselhaus, V. Bulovic, J. Kong, and S. Gradecak, Nano Lett. 13, 233 (2013). 15 Y.-J. Kim, H. Yoo, C.-H. Lee, J. B. Park, H. Baek, M. Kim, and G.-C. Yi, Adv. Mater. 24, 5565 (2012). 16 A. M. Munshi, D. L. Dheeraj, V. T. Fauske, D.-C. Kim, A. T. J. van Helvoort, B.-O. Fimland, and H. Weman, Nano Lett. 12, 4570 (2012). 17 W. M. Choi, K.-S. Shin, H. S. Lee, D. Choi, K. Kim, H.-J. Shin, S.-M. Yoon, J.-Y. Choi, and S.-W. Kim, Nano Res. 4, 440 (2011). 18 Y.-J. Kim, J.-H. Lee, and G.-C. Yi, Appl. Phys. Lett. 95, 213101 (2009). 19 Y. J. Hong, H. S. Jung, J. Yoo, Y. J. Kim, C. H. Lee, M. Kim, and G. C. Yi, Adv. Mater. 21, 222 (2009). 20 S. R. Na, J. W. Suk, L. Tao, D. Akinwande, R. S. Ruoff, R. Huang, and K. M. Liechti, ACS Nano 9, 1325 (2015). 21 C. H. Lee, Y. J. Kim, Y. J. Hong, S. R. Jeon, S. Bae, B. H. Hong, and G. C. Yi, Adv. Mater. 23, 4614 (2011). 22 W. Z. Wu, X. N. Wen, and Z. L. Wang, Science 340, 952 (2013). 23 C.-H. Lee, J. Yoo, Y. J. Hong, J. Cho, Y.-J. Kim, S.-R. Jeon, J. H. Baek, and G.-C. Yi, Appl. Phys. Lett. 94, 213101 (2009). 24 Y. J. Hong, C. H. Lee, A. Yoon, M. Kim, H. K. Seong, H. J. Chung, C. Sone, Y. J. Park, and G. C. Yi, Adv. Mater. 23, 3284 (2011). 25 V. Q. Dang, D. I. Kim, L. T. Duy, B. Y. Kim, B. U. Hwang, M. Jang, K. S. Shin, S. W. Kim, and N. E. Lee, Nanoscale 6, 15144 (2014). 26 F. Xue, L. B. Chen, L. F. Wang, Y. K. Pang, J. Chen, C. Zhang, and Z. L. Wang, Adv. Funct. Mater. 26, 2104 (2016).
