Isotropic Growth of Graphene toward Smoothing Stitching - ACS Nano

Jul 12, 2016 - Deterministic patterned growth of high-mobility large-crystal graphene: a path towards wafer scale integration. Vaidotas Miseikis , Fed...
0 downloads 3 Views 9MB Size
Isotropic Growth of Graphene toward Smoothing Stitching Mengqi Zeng,† Lifang Tan,† Lingxiang Wang,† Rafael G. Mendes,‡ Zhihui Qin,§ Yaxin Huang,† Tao Zhang,† Liwen Fang,† Yanfeng Zhang,∥,⊥ Shuanglin Yue,# Mark H. Rümmeli,‡ Lianmao Peng,# Zhongfan Liu,∥ Shengli Chen,† and Lei Fu*,† †

College of Chemistry and Molecular Science, Wuhan University, Wuhan 430072, China IFW Dresden, Dresden 01069, Germany § State Key Laboratory of Magnetic Resonance and Atomic and Molecular Physics, Wuhan Institute of Physics and Mathematics, Chinese Academy Sciences, Wuhan 430071, China ∥ Center for Nanochemistry, College of Chemistry and Molecular Engineering, ⊥Department of Materials Science and Engineering, College of Engineering, and #Department of Electronics, Peking University, Beijing 100871, China ‡

S Supporting Information *

ABSTRACT: The quality of graphene grown via chemical vapor deposition still has very great disparity with its theoretical property due to the inevitable formation of grain boundaries. The design of single-crystal substrate with an anisotropic twofold symmetry for the unidirectional alignment of graphene seeds would be a promising way for eliminating the grain boundaries at the wafer scale. However, such a delicate process will be easily terminated by the obstruction of defects or impurities. Here we investigated the isotropic growth behavior of graphene single crystals via melting the growth substrate to obtain an amorphous isotropic surface, which will not offer any specific grain orientation induction or preponderant growth rate toward a certain direction in the graphene growth process. The as-obtained graphene grains are isotropically round with mixed edges that exhibit high activity. The orientation of adjacent grains can be easily self-adjusted to smoothly match each other over a liquid catalyst with facile atom delocalization due to the low rotation steric hindrance of the isotropic grains, thus achieving the smoothing stitching of the adjacent graphene. Therefore, the adverse effects of grain boundaries will be eliminated and the excellent transport performance of graphene will be more guaranteed. What is more, such an isotropic growth mode can be extended to other types of layered nanomaterials such as hexagonal boron nitride and transition metal chalcogenides for obtaining large-size intrinsic film with low defect. KEYWORDS: graphene, isotropic growth, liquid metal, smooth stitching

G

from the unidirectional alignment of graphene grains to be merged to uniform single-crystal graphene film on single-crystal substrates with extremely high crystalline. Artificial hydrogenterminated germanium (110) surface11 and copper (111) surface12 were employed to achieve the unidirectional or commensurate stitching to coalesce graphene domains. Nevertheless, one should pay a high cost to obtain a perfect singlecrystal surface, and ultimately cannot avoid the defects or impurities which will terminate the unidirectional alignment. Generally speaking, incommensurate stitching is rooted in the uncontrollability of a grain’s orientation and shape, which directly affect the final assembly manner.13 It was reported that the shape, orientation, and edge geometry of CVD graphene

raphene is the two-dimensional (2D) crystalline form of carbon with one-atom thickness. To fully exploit its extraordinary potential, the scalable growth of singlecrystal graphene over wafer-scale areas has to be realized.1,2 Chemical vapor deposition (CVD) succeeded in synthesizing predominantly monolayer, highly crystallized graphene on polycrystalline copper (Cu) foils.3−5 However, the large-scale synthetic graphene produced thus far is typically polycrystalline via stitching of graphene domains separated by defective grain boundaries that significantly degrade their electrical and mechanical properties.6,7 Two strategies exist to reduce the forming of graphene grain boundaries. One approach is to decrease the nucleation density to obtain as large as possible of graphene grains, in which various substrate pretreatments are employed, such as annealing,8 polishing,9 and preoxidizing.10 However, the growth of large single graphene crystals is easily interrupted by the uncertain disturbances and normally requires a long-time growth. Another thoroughgoing approach benefits © 2016 American Chemical Society

Received: June 3, 2016 Accepted: July 12, 2016 Published: July 12, 2016 7189

DOI: 10.1021/acsnano.6b03668 ACS Nano 2016, 10, 7189−7196

Article

www.acsnano.org

Article

ACS Nano

Figure 1. Isotropic graphene grains grown on liquid Cu. (a) Scheme of the IGG via isotropic growth on the liquid Cu. (b) SEM image of an individual IGG with round shape. (c) Typical Raman spectrum of the IGG transferred onto SiO2/Si. (d−f) Raman mapping of ID, IG, and I2D, in which the spatial and spectral resolutions of the measurements are 1 μm and 1 cm−1, respectively.

domains could be regulated by the crystallographic orientations of the underlying substrates.14 In addition, the shape of graphene grains also varies with the growth conditions when employing solid Cu foil as the growth substrate. The multifarious shapes have been verified, including hexagon,15,16 triangle,17,18 rectangle,19,20 four-lobe,21,22 and snowflake,23,24 demonstrating the serious uncontrollability in the shape of graphene grains, not to mention its orientation control on solid metal substrates. Or perhaps even worse, the observed graphene grains usually exhibit poor symmetry, and even for intrinsic hexagonal crystals, which usually appear in elongated hexagons. The complex atom configurations at each edge toward a various direction make it hard for the stitching of two adjacent graphene grains to be commensurate. Thus, a seamless coalescence of graphene domains at large scale seems impossible unless on a perfect single crystal substrate with specifically designed surface. Notably, totally different from the solid crystalline catalysts, liquid catalysts, such as molten Cu, exhibit extraordinary controllability for growing perfectly symmetrical structures. Perfect six-symmetrical structure, for instance, hexagonal, various snowlike, and even twelve-pointed graphene single crystals, can be obtained by using liquid catalyst.25,26 Its amorphous surface without any specific crystal orientation and good fluidity appear to be propitious for the isotropic growth of graphene, which will not offer any specific grain orientation induction or preponderant growth rate toward a certain direction in the graphene growth process. As we know, compared with solid metal, the microstructure of amorphous liquid metal exhibits short-range order while showing long-range disorder. When a metal is in a liquid state, the interatom interaction does not orient toward a specific direction due to the thermal motion of metal atoms, and thus, the liquid metal owns fluidity and shows isotropy at the macro level.27 Here we attempt to overcome the negative influence of the uncontrollability of graphene grain’s orientation and shape in the stitching process by facilitating the isotropic growth of graphene through taking advantage of the liquid’s isotropy. Thereby, obtained graphene grains will be isotropically round

with random mixed smooth edge, in which the atoms arrange alternatively in the form of zigzag and armchair and the whole edge exhibits high activity. Atom configurations at the edge of the isotropically round graphene grains were further presented clearly by scanning tunneling microscopy (STM) in our work. Providing this, the orientation of adjacent grains can be slightly adjusted to match with each other due to the high-activity edge of as-obtained graphene grains with isotropic shape which owns much lower rotation steric hindrance compared with those with angular shape, and furthermore, the facile atom delocalization of the liquid metals offers the platform for the rotation.28 Thus, the stitching of the isotropic graphene single grains will become ultrasmooth. The deleterious effect of grain boundaries will be significantly weakened, and the transport performances will be guaranteed. What is more, we believe that the concept of isotropic growth mode derived smooth stitching behavior can be extended to other types of layered nanomaterials for obtaining intrinsic film with low defects.

RESULTS AND DISCUSSION Figure 1a shows the schematic of a graphene domain grown on Cu in a melted state. During the CVD growth, carbon source with extremely low concentration (0.5/10/800 sccm CH4/H2/ Ar) is introduced, in which the slow growth process will ensure the carbon atoms full diffusion and reach the energetically favorable sites. The amorphous surface of the liquid Cu will not offer a dominant direction of carbon atoms arrangement during the graphene formation. When the underlying Cu is in a solid state (1050 °C), the typical shape of observed graphene grains is hexagon with angular edges, which is derived from the kinetic Wulff construction (Supporting Information Figure S1).29 It is expected that the rate-limiting step for graphene domain growth in this situation is the adsorption of carbon species. Because the diffusion rate at a crystal surface is anisotropic, the graphene domain exhibits fast growth in the {211̅ 0̅ } direction and slow growth in the {101̅0} direction. As a result, the typical hexagonal domain will form on the solid Cu. When the temperature increased to 1080 °C reaching the melting point of 7190

DOI: 10.1021/acsnano.6b03668 ACS Nano 2016, 10, 7189−7196

Article

ACS Nano

Figure 2. TEM characterizations of the IGG transferred onto a quantifoil. (a) Low magnification image of a complete IGG outlined by the green dashed line. The green spots depict the regions where the SAED was collected. The diffractions corresponding to each of these areas are shown in the images labeled with 1−6. (b) Intensity profile over the {102̅0} (outer) and {101̅0} (inner) spots from the SAED pattern (in panel (a)-6). (c) Low resolution image of the edge of the IGG. (d) Backfolded edge of the IGG. (e) HRTEM atom images of the IGG. (f) Magnified region of graphene from (e) showing the perfect atom-scale crystal structure of the IGG.

Figure 3. Raman mappings of two smoothly stitched IGGs. (a) Optical image of the IGGs transferred onto a SiO2/Si substrate. (b−d) Intensity mappings of the D, G, and 2D Raman peaks, respectively. The spatial and spectral resolutions of the measurements are 0.5 μm and 1 cm−1, respectively.

intensity ratio of I2D and IG is about 2, and no D peak related to its defect was identified, indicating single-layer and high-quality features of as-grown IGG.4 To estimate whether the entire IGG was a single crystal, Raman mapping was also conducted on the IGG transferred onto SiO2/Si. The intensity mapping of the three characterized peaks of D, G, and 2D were clearly shown in Figure 1d−f. The three mapping images accurately outlined the round shape, which was in accordance with that in Figure 1b. The G and 2D peak mapping showed a high uniformity across the whole IGG and demonstrated that the as-obtained IGG owned extremely high uniformity and quality. In addition, it is worth noting that no pronounced ID was observed on the

Cu, its surface will exhibit an amorphous state. The dissociative carbon species on the liquid Cu surface will possess higher diffusion velocity.27 Thus, although the zigzag edge along {101̅0} direction has a lower edge free energy than that of armchair edge, the high-activity Cu surface makes the atom assemble isotropically instead of following an intrinsic direction. Finally, isotropic graphene grain (IGG) is successfully obtained on the liquid Cu, as shown in Figure 1b. Raman spectrum was employed to evaluate the quality and layer numbers of the IGG (Figure 1c). The IGG displays typical single-layer features with two prominent Raman peaks located at ∼1580 and ∼2680 cm−1, corresponding to the G and 2D peaks, respectively. The 7191

DOI: 10.1021/acsnano.6b03668 ACS Nano 2016, 10, 7189−7196

Article

ACS Nano

Figure 4. STM images of two adjacent IGGs on Cu. (a) STM image near the stitching site that exhibited the graphene on the step of the Cu surface. (b, f) Enlarged STM images of graphene grown on Cu in two adjacent different regions near the hypothetical grain boundary line. (c, g) Honeycomb graphene structures corresponding to the regions marked by blue or green dashed squares in (b) and (f). (d, h) FFT results derived from (c) and (g), respectively. (e) Enlarged STM images of the exposed Cu.

indicating no obvious defective grain boundaries were observed. This was entirely different from the Raman studies at the grain boundaries between two coalesced hexagonal graphene grains in the previous work and in our contrast experiment (Figure S2). The absence of ID at the grains’ joint indicates the potential existence of smooth stitching of adjacent IGGs. In addition, the intensity distribution of I2D and IG also demonstrated the high uniformity of IGGs even across the assumed splice line. More sets of Raman mappings of two adjacent IGGs are exhibited in Figure S3. In fact, the smoothing stitching of the two IGGs does not show a significant relationship with the grain size. As compared with the graphene grains with angular shape, besides the macroscopic isotropic shape, the deeper reason lies in the mixed edges with high activity. The isotropic shape and highly active edges are the driven forces of the rotation, resulting the energy minimization at the stitched area. What is more, the ratio of islands merged with smoothing stitching is also evaluated by randomly collecting 100 Raman spectra sampling at the hypothetic grain boundary. The intensity ratio of ID/I2D is analyzed, as typically shown in Figure S4. The statistics of I D /I 2D corresponding to more adjacent IGGs is shown in Figure S5, in which the value is 8.77 ± 1.60‰, indicating that the defect can be nearly ignored. What is more, Raman line scans collected at the hypothetic grain boundary of the regions merged with smoothing stitching are exhibited, as shown in Figure S6. The absence of D peak shows the smoothing stitching at the multiple hypothetic boundaries, thus confirming that the seamless stitching could be well extended from two merged islands to several ones, which lays the foundation of the seamless stitching of the IGGs over the whole liquid surface. Our confocal Raman spectroscopy provided evidence of no detectable grain boundaries between two adjacent IGGs at the macroscale. Here we attempt to reveal the stitching behavior at atomic scale by STM measurements. This was performed by focusing the corner of two stitched IGGs on Cu without transfer, and thus, the in situ characterizations can be realized. Figure 4a shows the STM image across the hypothetical

edge of the grains, which are totally different from the previous Raman studies of graphene edges.30−32 Generally speaking, the ID at the edge can be attributed to the defects resulted from that the bonding of the carbon atoms at the edge is different from those in the central region. Moreover, the intensity of the D peak is weak at the zigzag edge due to momentum conservation and strong at the armchair edge of graphene.33−36 The type and disorder degree of the edge can be reflected in the D peak. The abnormal D peak at the edge of IGG should be contributed to the high-activity mixed edge which is dominated by zigzag type. The structure and morphology of the as-obtained IGG were further characterized by transmission electron microscopy (TEM). The transferred IGG on quantifoil grid was shown in Figure 2a, exhibiting isotropically round shape. Then selectedarea electron diffraction (SAED) patterns show the same 6-fold symmetric diffraction points at different regions marked 1−6 over the entire IGG. To affirm that the observed reflexes are from a monolayer graphene and not from a Bernal-stacked graphene, we examined the relative intensity of the inner {101̅0} and outer {102̅0} spots. For monolayer, the inner spots are more intense than the outer spots, as we observed (Figure 2b), confirming the single layer characteristics. The region marked by a red square in dashed line in Figure 2a corresponds to the enlarged Figure 2c. By counting lines of contrast along a backfolded edge of the graphene film, we could tell that the IGG was single layer, as shown in Figure 2d. We further demonstrated the single-layer feature and high crystallinity of the graphene by the low voltage aberration-corrected, high resolution transmission electron microscopy (LVAC− HRTEM) (Figure 2e,f), which highlights the perfect atomscale crystal structure and six-fold symmetry single-crystal nature of the graphene. Identification of the boundaries between two adjacent IGGs is important to understand the stitching. Raman mapping was employed to detect the defective grain boundaries that may exist at the grains’ joint. Figure 3a shows a typical optical image of two adjacent IGGs. ID is negligibly small (Figure 3b) over the entire area within each IGG as well as at the coalesced site, 7192

DOI: 10.1021/acsnano.6b03668 ACS Nano 2016, 10, 7189−7196

Article

ACS Nano stitching boundary, which exhibited graphene on the step of the Cu surface, and the white dashed line was drawn to show the edge of these two adjacent IGGs. In addition, the lowmagnification image of these two adjacent IGGs was offered in Figure S7 to clearly show the merging. The red dashed line outlined the hypothetical grain boundary of these two stitched IGGs. Enlarged STM images located in two adjacent different regions (Figure 4b and f) near the hypothetical line showed the graphene grown on Cu at two sides, which were totally different from that of the exposed Cu (Figure 4e). The atom images corresponding with the regions marked by blue or green dashed squares are also offered in Figure 4c and g, revealing the clear honeycomb graphene structures. The corresponding FFT results (Figure 4d and h) identified the identical orientation across the hypothetical line. Similar phenomena in other adjacent IGGs at the corner were also observed, as shown in Figure S8. In brief, no clear evidence for the existence of a grain boundary between two adjacent IGGs was noticed. Furthermore, low energy electron diffraction (LEED) was conducted at a large area, up to ∼0.7 × 1 cm2, which is the maximum size for the sample holder in the STM, to confirm that IGGs can really adjust their rotations on liquid Cu to achieve the smoothing stitching at the macroscale, as shown in Figure S9. The adopted electron energy is ∼90 eV. Two regions with distance as large as ∼1 cm were selected and the size of the regions employed for LEED characterization is about 1 × 1 mm2, which can elucidate the smooth stitching at a macro level. The graphene formed on the liquid substrate showed only one set of six bright diffraction spots, which well confirmed the orientation consistency over a large area and thus proved the smooth stitching via the self-alignment of the IGGs on the liquid substrate with excellent fluidity. Furthermore, SAED characterizations over the seamlessly stitched graphene film by TEM within a large area (0.5 × 0.5 cm2, the size of the Cu grid for TEM characterization) were also conducted, as shown in Figure S10. Ten sampling sites were randomly selected and all the patterns show nearly the same six-fold symmetric diffraction, which confirms the single crystal property of our obtained graphene film stitched by IGGs with typical size of 400 μm2. To investigate how the IGGs achieve the smooth stitching, the edge structures of the IGG were studied in detail by STM. Figure 5a exhibited a typical image of the IGG’s edge, which looks like a nanozipper with periodic sawtooth-like variations rather than a smooth edge observed at micrometer scale. In a IGG with a typical size, there are several hundred “sawteeth” at its edge. Therefore, the IGGs exhibited extremely high isotropic property, as compared to the hexagonal graphene grains. Thus, the steric hindrance of the IGGs toward each direction is nearly the same and the whole IGGs exhibit much lower rotation steric hindrance compared with the graphene grains with angular shape. And with the help of the underlying liquid Cu, which possessed good rheological behaviors to suit for rotation of IGGs, only an extremely low energy will be needed to achieve the rotation with an extremely low angle of the IGGs in the smooth stitching process. To further clarify the rotation of the IGGs grown on the liquid Cu, the surface state of the liquid Cu for graphene growth should be focused. The electron backscattered diffraction (EBSD) and X-ray diffraction (XRD) characterizations of the Cu covered with IGGs were conducted, as shown in Figure S11, from which we can consider that the crystal form of liquid-derived Cu surface is ultrauniform and can be assigned to (220) (2θ = 74.130°) (JCPDS 04-0836). The high lattice mismatch (59%) between graphene and Cu

Figure 5. Edge structures of the IGG and stitching process of two adjacent IGGs. (a) Typical STM image of the edge of an individual IGG. (b) Atom images at the edge of the IGG and the simulative atom structures derived from the actual atom arrangements. (c) Stitching process of two adjacent IGGs.

(220) indicate the edges of IGGs are highly active. It is noted that recrystallized Cu usually has energetically favorable (111) surface;37,38 however, with IGGs on the surface, (220) surface was formed instead. Related density functional theory (DFT) calculations had been performed, as shown in Figure S12. More isotropic graphene grains tend to be bound more tightly to Cu (220) surface. It might mean that the above IGGs with high active edges would have certain influence on the solidification of the liquid Cu. It has been reported that the 2D materials could facilitate the metallic atoms to self-assemble into a specific ordered structure.39,40 The freshly formed crystal face of liquid Cu during the solidifying process should be matched with the above IGGs to achieve the energy minimization. As shown in Figure 5b, the atom configuration of the IGG at the edge was exhibited and the carbon atoms at the edges were marked by hollow black spots one by one. To show more clearly, simulative atom structures derived from the actual atom arrangements were offered. The edge exhibits random mixed structures, consisting of both zigzag configurations and armchair configurations. The Raman spectra at the edge indicate that the edge is dominant by zigzag configurations. Such a mixed edge will be highly active on a liquid Cu surface with fluidity, and thus rotation of IGGs is easily achievable. The isotropic edges make that only a rotation of an extremely low angle can achieve the angular adjustment during the stitching between two adjacent IGGs. What is more, the driving force of the rotation comes from the energy minimization at the stitched area. The alignment of the crystal orientations between two adjacent IGGs will lower the energy of their active mixed edges. As shown in Figure 5c, the IGGs can self-adjust the orientation toward smoothing stitching and the atom vacancies will be filled by the Cu-catalyzed decomposed carbon atoms. Thus, a perfect and smooth stitching was achieved between two IGGs due to their highly active edge and good fluidity of the underlying liquid Cu, which can easily achieve facile atom delocalization. As the growth continues, more individual IGGs stitch together. Figure S13 shows the transient state from the 7193

DOI: 10.1021/acsnano.6b03668 ACS Nano 2016, 10, 7189−7196

Article

ACS Nano

Figure 6. Carrier mobility across a hypothetic grain boundary to show the smooth stitching. (a) Optical microscopy image of a device with multiple electrodes (numbered 1−4) contacting two coalesced IGGs. The white dashed line outlines the original coalesced IGGs, and the hypothetic grain boundary is marked by a gray dashed line. (b) Intragrain and intergrain Ids−(Vgs − VDirac,gs) curves measured at room temperature for this device. The mobility derived from the region between electrode 1 and electrode 2 is 2928.6 cm2/(V s), and for the region between electrode 3 and electrode 4 the value is 3155.9 cm2/(V s). While for the intergrain region, the mobility derived from the region between electrode 2 and electrode 3 is 3109.4 cm2/(V s), which basically matched with the mobility derived from the intragrain region and confirmed the smooth stitching.

merger of several IGGs to an intact film, in which circular and subcircular domains can be found on the edges of the film. On the other hand, once there are larger (already merged with nonisotropic shape) domains, the seamless stitch cannot be easily achieved contributed to the self-rotation process due to that the as-formed larger and irregular shape will have relatively large steric hindrance. However, the uniformity and smooth surface of the liquid Cu substrate facilitate the seamless stitching of IGGs over the whole liquid surface. The nucleation on a liquid substrate will be quite uniform and as well as the crystallization orientation in order to achieve the energy minimization. It has been reported that liquid fluidity of the catalyst could benefit the dispersion of graphene single crystals to some extent. What is more, in the subsequent growth, the mass transfer over the liquid surface will be also uniform. Thus, the spatial, size, and crystalline orientation of the isotropic graphene grains will be approximately the same. In addition, during the progressive stitching process, with the help of the underlying liquid Cu, which possesses good rheological property to suit for rotation of IGGs, the precise seamless stitching can be further ensured. Electrical transport measurements were performed on multiterminal devices fabricated from two adjacent IGGs transferred onto 300 nm SiO2/Si wafers to show the smoothing stitching. Figure 6a shows the field effect transistor (FET) fabricated on two coalesced IGGs that meet at a hypothetic grain boundary. Multiple electrodes (Cr/Au (5/50 nm)) were patterned to contact each grain to allow simultaneous measurements of both intragrain (within the grain) and intergrain (across the grain boundary) transport. Employing different electrode pairs as the source and drain electrodes, Si as the back gate and 300 nm SiO2 as the gate dielectric layer, we can obtain the transport characteristics (Ids−Vgs) of the selected graphene regions. Figure 6b shows representative intragrain and intergrain transport characteristics (Ids−Vgs) of the FET measured under ambient conditions. The source-drain voltage was 0.05 V and the mobility calculated from two intra grain region and one inter grain region is 2928.6, 3155.9, and 3109.4 cm2/(V s), respectively. The mobility was not degraded via such a smooth stitching mode of two adjacent isotropic graphene grains. To exhibit the experimental reproducibility of the electrical measurement, more sets of data were offered, as

shown in Figure S14. For the smoothly stitched graphene grains, the error bar of the mobility value is 1.39−6.03%. What is more, in order to show the quality of the whole film, we measured the sheet resistance to show the electrical performance at a macro level and the value was 290.4 Ω/□ and could be much lower than the reported work (2100 Ω/□).41 In other words, the hypothetic boundary in smoothly stitched IGGs is electrically transparent.

CONCLUSION In summary, we achieved the smoothing stitching of adjacent graphene by facilitating the isotropic growth of graphene through employing the accessible amorphous liquid metal instead of the expensive single crystalline solid ones. We recognized the effect of the crystal shape and edge configuration on the stitching behavior of the two adjacent graphene single crystals based on such an liquid substrate. The obtained uniform and high-quality graphene domains are isotropically round with mixed smooth edge dominant in zigzag configuration that exhibits high activity, which was identified by STM atom images. The isotropic edges make that only a rotation of one IGG with an extremely low angle can achieve the angular adjustment during the stitching with another IGG due to the much lower rotation steric hindrance of isotropic grains compared with those of the angular grains, and the appropriate rheological behavior of underlying liquid Cu ensures the rotation. Raman mapping and STM characterizations were conducted to show the smooth stitching of two adjacent IGGs at different scales. Furthermore, transport properties were also measured within each graphene grain and across the grain boundary, and the hypothetic boundary in smoothly stitched IGGs was electrically transparent. What is more, isotropic growth via liquid catalytic system can be extended to other types of layered nanomaterials such as hexagonal boron nitride (h-BN) and transition metal chalcogenides (TMDs) and for obtaining intrinsic film with low defect. METHODS AND MATERIALS CVD Growth of Isotropic Graphene Grains on Liquid Cu. Cu foils (100 μm thick, 99.8% purity) and W foils (100 μm thick, 99.95% purity) were employed. One piece of Cu foil was directly placed on the 7194

DOI: 10.1021/acsnano.6b03668 ACS Nano 2016, 10, 7189−7196

Article

ACS Nano W foil. Before graphene growth, the 1 in. quartz tube was pumped to ∼10 Pa to clean the growth system. Then 200 sccm (standard cubic centimeters per minute) H2 gas flow was filled into the tube. After that, the quartz tube was heated using a furnace (Lindberg/Blue M, HTF55322C) to 1080 °C for 40 min. In the graphene growth process, the H2 flow rate was set to the required value, and Ar and CH4 was then introduced into the chamber. Finally, CH4 was turned off, and the system was naturally cooled down to room temperature. Transferring the Graphene to the Target Substrates. The process of transferring involved spin-coating a poly(methyl methacrylate) (PMMA) film on the graphene-grown substrates and releasing the PMMA/graphene film by etching out the Cu in an iron(III) chloride (FeCl3) aqueous solution (∼1 M) for 2−3 h. This was followed by a rinse in ultrapure water to remove the metal ions. The PMMA layer was dissolved in an acetone bath at 100 °C for 5 min after the PMMA/graphene film was transferred onto SiO2/Si substrates or TEM grids for further characterization. Characterization. Optical images were taken with an optical microscope (Olympus DX51), and Raman spectroscopy was performed with a laser micro-Raman spectrometer (Renishaw in Via, 532 nm excitation wavelength). Scanning tunneling microscopy (STM) characterization was performed inside an Omicron ultrahigh vacuum (UHV) STM at room temperature. Low energy electron diffraction (LEED) characterization was performed by a Unisoku USM1300S 3He STM system at room temperature. Scanning electron microscopy (SEM) images were obtained by Hitach-S4800 SEM and ZEISS Merlin Compact SEM. The EBL (JEOL 6510 with NPGS) was employed to define the structures of Hall bars derived from two coalesced IGGs and corresponding electrode patterns followed by metal evaporation and lift-off processes. Cr/Au (5/50 nm) was deposited as contact electrode to ensure high-quality ohmic-like contact in the devices. The current (I)−voltage (V) data were collected in a probe station under ambient conditions using an Agilent 4155C. The transmission electron microscopy (TEM) images were obtained by an aberration-corrected high-resolution TEM system (model AC-HRTEM, FEI Titan), in which the operating voltage was 80 kV and the graphene films were transferred onto a quantifoil copper TEM grid.

ACKNOWLEDGMENTS The research was supported by the Natural Science Foundation of China (Grant 51322209, 21473124) and the Sino-German Center for Research Promotion (Grant GZ 871). We thank Prof. Daiwen Pang and Prof. Lei Liao for their assistance with electrical transport measurements. REFERENCES (1) Geim, A. K. Graphene: Status and Prospects. Science 2009, 324, 1530−1534. (2) Tan, L. F.; Zeng, M. Q.; Zhang, T.; Fu, L. Design of Catalytic Substrates for Uniform Graphene Films: From Solid-Metal to LiquidMetal. Nanoscale 2015, 7, 9105−9121. (3) Novoselov, K. S.; Fal’ko, V. I.; Colombo, L.; Gellert, P. R.; Schwab, M. G.; Kim, K. A Roadmap for Graphene. Nature 2012, 490, 192−200. (4) Li, X. S.; Cai, W. W.; An, J. H.; Kim, S.; Nah, J.; Yang, D. X.; Piner, R.; Velamakanni, A.; Jung, I.; Tutuc, E.; Banerjee, S. K.; Colombo, L.; Ruoff, R. S. Large-Area Synthesis of High-Quality and Uniform Graphene Films on Copper Foils. Science 2009, 324, 1312− 1314. (5) Li, X. S.; Cai, W. W.; Colombo, L.; Ruoff, R. S. Evolution of Graphene Growth on Ni and Cu by Carbon Isotope Labeling. Nano Lett. 2009, 9, 4268−4272. (6) Liu, Y. Y.; Yakobson, B. I. Cones, Pringles, and Grain Boundary Landscapes in Graphene Topology. Nano Lett. 2010, 10, 2178−2183. (7) Yazyev, O. V.; Louie, S. G. Electronic Transport in Polycrystalline Graphene. Nat. Mater. 2010, 9, 806−809. (8) Yan, Z.; Lin, J.; Peng, Z. W.; Sun, Z. Z.; Zhu, Y.; Li, L.; Xiang, C. S.; Samuel, E. L.; Kittrell, C.; Tour, J. M. Toward the Synthesis of Wafer-Scale Single-Crystal Graphene on Copper Foils. ACS Nano 2012, 6, 9110−9117. (9) Han, G. H.; Gunes, F.; Bae, J. J.; Kim, E. S.; Chae, S. J.; Shin, H. J.; Choi, J. Y.; Pribat, D.; Lee, Y. H. Influence of Copper Morphology in Forming Nucleation Seeds for Graphene Growth. Nano Lett. 2011, 11, 4144−4148. (10) Hao, Y. F.; Bharathi, M. S.; Wang, L.; Liu, Y. Y.; Chen, H.; Nie, S.; Wang, X. H.; Chou, H.; Tan, C.; Fallahazad, B.; Ramanarayan, H.; Magnuson, C. W.; Tutuc, E.; Yakobson, B. I.; McCarty, K. F.; Zhang, Y. W.; Kim, P.; Hone, J.; Colombo, L.; Ruoff, R. S. The Role of Surface Oxygen in the Growth of Large Single-Crystal Graphene on Copper. Science 2013, 342, 720−723. (11) Lee, J. H.; Lee, E. K.; Joo, W. J.; Jang, Y.; Kim, B. S.; Lim, J. Y.; Choi, S. H.; Ahn, S. J.; Ahn, J. R.; Park, M. H.; Yang, C. W.; Choi, B. L.; Hwang, S. W.; Whang, D. Wafer-Scale Growth of Single-Crystal Monolayer Graphene on Reusable Hydrogen-Terminated Germanium. Science 2014, 344, 286−289. (12) Nguyen, V. L.; Shin, B. G.; Duong, D. L.; Kim, S. T.; Perello, D.; Lim, Y. J.; Yuan, Q. H.; Ding, F.; Jeong, H. Y.; Shin, H. S.; Lee, S. M.; Chae, S. H.; Vu, Q. A.; Lee, S. H.; Lee, Y. H. Seamless Stitching of Graphene Domains on Polished Copper (111) Foil. Adv. Mater. 2015, 27, 1376−1382. (13) Biro, L. P.; Lambin, P. Grain Boundaries in Graphene Grown by Chemical Vapor Deposition. New J. Phys. 2013, 15, 035024. (14) Murdock, A. T.; Koos, A.; Britton, T. B.; Houben, L.; Batten, T.; Zhang, T.; Wilkinson, A. J.; Dunin-Borkowski, R. E.; Lekka, C. E.; Grobert, N. Controlling the Orientation, Edge Geometry, and Thickness of Chemical Vapor Deposition Graphene. ACS Nano 2013, 7, 1351−1359. (15) Robertson, A. W.; Warner, J. H. Hexagonal Single Crystal Domains of Few-Layer Graphene on Copper Foils. Nano Lett. 2011, 11, 1182−1189. (16) Wu, B.; Geng, D. C.; Guo, Y. L.; Huang, L. P.; Xue, Y. Z.; Zheng, J.; Chen, J. Y.; Yu, G.; Liu, Y. Q.; Jiang, L.; Hu, W. P. Equiangular Hexagon-Shape-Controlled Synthesis of Graphene on Copper Surface. Adv. Mater. 2011, 23, 3522−3525. (17) Liu, J. W.; Wu, J.; Edwards, C. M.; Berrie, C. L.; Moore, D.; Chen, Z. J.; Maroni, V. A.; Paranthaman, M. P.; Goyal, A. Triangular

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsnano.6b03668. Growth of hexagonal graphene grains on solid copper; Raman mappings of two adjacent hexagonal graphene grains with obvious grain boundary and of two adjacent IGGs; statistics about the ratio of islands merged with smoothing stitching; Raman line scans collected at the hypothetic grain boundary of the regions merged with smoothing stitching; low-magnification image of two adjacent IGGs to show hypothetic grain boundary; STM images of two adjacent IGGs on Cu; LEED characterization for IGGs; SAED characterization over a large area on graphene film via seamless stitching transferred onto a Cu grid; EBSD and XRD characterizations of the Cu covered with IGGs after the solidification; DFT calculations of adsorption energy of graphene grains with different shapes on Cu (220); merger of IGGs to continuous film; carrier mobility across a hypothetic grain boundary to show smooth stitching (PDF)

AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest. 7195

DOI: 10.1021/acsnano.6b03668 ACS Nano 2016, 10, 7189−7196

Article

ACS Nano Graphene Grain Growth on Cube-Textured Cu Substrates. Adv. Funct. Mater. 2011, 21, 3868−3874. (18) Chen, X.; Liu, S. Y.; Liu, L. C.; Liu, X. Q.; Liu, X. M.; Wang, L. Growth of Triangle-Shape Graphene on Cu(111) Surface. Appl. Phys. Lett. 2012, 100, 163106. (19) Wu, Y. A.; Robertson, A. W.; Schaffel, F.; Speller, S. C.; Warner, J. H. Aligned Rectangular Few-Layer Graphene Domains on Copper Surfaces. Chem. Mater. 2011, 23, 4543−4547. (20) Wang, H.; Wang, G. Z.; Bao, P. F.; Yang, S. L.; Zhu, W.; Xie, X.; Zhang, W. J. Controllable Synthesis of Submillimeter Single-Crystal Monolayer Graphene Domains on Copper Foils by Suppressing Nucleation (vol 134, pg 3627, 2012). J. Am. Chem. Soc. 2012, 134, 18476−18476. (21) Li, X. S.; Magnuson, C. W.; Venugopal, A.; An, J. H.; Suk, J. W.; Han, B. Y.; Borysiak, M.; Cai, W. W.; Velamakanni, A.; Zhu, Y. W.; Fu, L. F.; Vogel, E. M.; Voelkl, E.; Colombo, L.; Ruoff, R. S. Graphene Films With Large Domain Size by a Two-Step Chemical Vapor Deposition Process. Nano Lett. 2010, 10, 4328−4334. (22) Wofford, J. M.; Nie, S.; McCarty, K. F.; Bartelt, N. C.; Dubon, O. D. Graphene Islands on Cu Foils: the Interplay Between Shape, Orientation, and Defects. Nano Lett. 2010, 10, 4890−4896. (23) Li, X. S.; Magnuson, C. W.; Venugopal, A.; Tromp, R. M.; Hannon, J. B.; Vogel, E. M.; Colombo, L.; Ruoff, R. S. Large-Area Graphene Single Crystals Grown by Low-Pressure Chemical Vapor Deposition of Methane on Copper. J. Am. Chem. Soc. 2011, 133, 2816−2819. (24) Zhang, Y.; Zhang, L. Y.; Kim, P.; Ge, M. Y.; Li, Z.; Zhou, C. W. Vapor Trapping Growth of Single-Crystalline Graphene Flowers: Synthesis, Morphology, and Electronic Properties. Nano Lett. 2012, 12, 2810−2816. (25) Geng, D. C.; Meng, L.; Chen, B. Y.; Gao, E. L.; Yan, W.; Yan, H.; Luo, B. R.; Xu, J.; Wang, H. P.; Mao, Z. P.; Xu, Z. P.; He, L.; Zhang, Z. Y.; Peng, L. M.; Yu, G. Controlled Growth of Single-Crystal Twelve-Pointed Graphene Grains on a Liquid Cu Surface. Adv. Mater. 2014, 26, 6423−6429. (26) Wu, B.; Geng, D. C.; Xu, Z. P.; Guo, Y. L.; Huang, L. P.; Xue, Y. Z.; Chen, J. Y.; Yu, G.; Liu, Y. Q. Self-Organized Graphene Crystal Patterns. NPG Asia Mater. 2013, 5, e36. (27) Zhang, J. S. Shaping Principles of Liquid Metal; Chemical Industry Press: Beijing, 2010; pp 4−12. (28) Zeng, M. Q.; Wang, L. X.; Liu, J. X.; Zhang, T.; Xue, H. F.; Xiao, Y.; Qin, Z. H.; Fu, L. Self-Assembly of Graphene Single Crystals with Uniform Size and Orientation: The First 2D Super-Ordered Structure. J. Am. Chem. Soc. 2016, 138, 7812−7815. (29) Witten, T. A.; Sander, L. M. Diffusion-Limited Aggregation, a Kinetic Critical Phenomenon. Phys. Rev. Lett. 1981, 47, 1400−1403. (30) Ferrari, A. C.; Meyer, J. C.; Scardaci, V.; Casiraghi, C.; Lazzeri, M.; Mauri, F.; Piscanec, S.; Jiang, D.; Novoselov, K. S.; Roth, S.; Geim, A. K. Raman Spectrum of Graphene and Graphene Layers. Phys. Rev. Lett. 2006, 97, 187401. (31) Ferrari, A. C. Raman Spectroscopy of Graphene and Graphite: Disorder, Electron-Phonon Coupling, Doping and Nonadiabatic Effects. Solid State Commun. 2007, 143, 47−57. (32) Malard, L. M.; Pimenta, M. A.; Dresselhaus, G.; Dresselhaus, M. S. Raman Spectroscopy in Graphene. Phys. Rep. 2009, 473, 51−87. (33) Cancado, L. G.; Pimenta, M. A.; Neves, B. R. A.; Dantas, M. S. S.; Jorio, A. Influence of the Atomic Structure on the Raman Spectra of Graphite Edges. Phys. Rev. Lett. 2004, 93, 247401. (34) Cancado, L. G.; Pimenta, M. A.; Neves, B. R. A.; MedeirosRibeiro, G.; Enoki, T.; Kobayashi, Y.; Takai, K.; Fukui, K.; Dresselhaus, M. S.; Saito, R.; Jorio, A. Anisotropy of the Raman Spectra of Nanographite Ribbons. Phys. Rev. Lett. 2004, 93, 047403. (35) Casiraghi, C.; Hartschuh, A.; Qian, H.; Piscanec, S.; Georgi, C.; Fasoli, A.; Novoselov, K. S.; Basko, D. M.; Ferrari, A. C. Raman Spectroscopy of Graphene Edges. Nano Lett. 2009, 9, 1433−1441. (36) You, Y. M.; Ni, Z. H.; Yu, T.; Shen, Z. X. Edge Chirality Determination of Graphene by Raman Spectroscopy. Appl. Phys. Lett. 2008, 93, 163112.

(37) Wood, J. D.; Schmucker, S. W.; Lyons, A. S.; Pop, E.; Lyding, J. W. Effects of Polycrystalline Cu Substrate on Graphene Growth by Chemical Vapor Deposition. Nano Lett. 2011, 11, 4547−4554. (38) Chatain, D.; Ghetta, V.; Wynblatt, P. Equilibrium Shape of Copper Crystals Grown on Sapphire. Interface Sci. 2004, 12, 7−18. (39) Huang, X.; Li, S.; Huang, Y.; Wu, S.; Zhou, X.; Li, S.; Gan, C. L.; Boey, F.; Mirkin, C. A.; Zhang, H. Synthesis of Hexagonal Closepacked Gold Nanostructures. Nat. Commun. 2011, 2, 292. (40) Fan, Z.; Huang, X.; Han, Y.; Bosman, M.; Wang, Q.; Zhu, Y.; Liu, Q.; Li, B.; Zeng, Z.; Wu, J.; Shi, W.; Li, S.; Gan, C. L.; Zhang, H. Surface Modification-induced Phase Transformation of Hexagonal Close-packed Gold Square Sheets. Nat. Commun. 2015, 6, 6571. (41) Li, X.; Zhu, Y.; Cai, W.; Borysiak, M.; Han, B.; Chen, D.; Piner, R. D.; Colombo, L.; Ruoff, R. S. Transfer of Large-Area Graphene Films for High-Performance Transparent Conductive Electrodes. Nano Lett. 2009, 9, 4359−4363.

7196

DOI: 10.1021/acsnano.6b03668 ACS Nano 2016, 10, 7189−7196