Oxygen Intercalation of Graphene on Transition Metal Substrate: An

Sep 25, 2015 - Consequently, four edge C atoms (or two C—C dimers) in the supercell form two M—O—C units and two carbonyl groups in the sequence...
0 downloads 6 Views 7MB Size
Letter pubs.acs.org/JPCL

Oxygen Intercalation of Graphene on Transition Metal Substrate: An Edge-Limited Mechanism Liang Ma,†,‡ Xiao Cheng Zeng,*,‡ and Jinlan Wang*,† †

Department of Physics, Southeast University, Nanjing 211189, China Department of Chemistry, University of NebraskaLincoln, Lincoln, Nebraska 68588, United States



S Supporting Information *

ABSTRACT: Oxygen intercalation has been proven to be an efficient experimental approach to decouple chemical vapor deposition grown graphene from metal substrate with mild damage, thereby enabling graphene transfer. However, the mechanism of oxygen intercalation and associated rate-limiting step are still unclear on the molecular level. Here, by using density functional theory, we evaluate the thermodynamics stability of graphene edge on transition metal surface in the context of oxygen and explore various reaction pathways and energy barriers, from which we can identify the key steps as well as the roles of metal passivated graphene edges during the oxygen intercalation. Our calculations suggest that in well-controlled experimental conditions, oxygen atoms can be easily intercalated through either zigzag or armchair graphene edges on metal surface, whereas the unwanted graphene oxidation etching can be suppressed. Oxygen intercalation is, thus, an efficient and low-damage way to decouple graphene from a metal substrate while it allows reusing metal substrate for graphene growth.

C

intercalation can largely decouple the interior graphene layer from the TM substrate whereas the possible rate-limiting step is to unbind the graphene edge from metal substrate.14,15 However, the underlying mechanism of O intercalation is still unclear. In this letter, we systemically investigate thermodynamic behavior and kinetic process of O intercalation through the metal surface passivated graphene edge on the molecular level by using density functional theory (DFT) calculation, from which we identify the rate-limiting step and effect of graphene edge types on O intercalation. As a result, a simple and low-damage method as well as optimized experimental parameters for decoupling graphene from metal catalyst substrate is proposed. Spin-polarized DFT calculations were performed with generalized gradient approximation (GGA) of Perdew, Burke, and Ernzerhof (PBE) exchange-correlation functional30 implemented in the Vienna Ab initio Simulation Package (VASP).31 The interactions between valence electrons and ion core were described by the projected augmented wave (PAW) potentials.32,33 The minimum energy paths (MEP) and the transition states were explored by using the climbing-image nudged elastic band (CI-NEB) method.34 We chose Cu(111) and Ni(111) surfaces as two representative metal substrates to explore the mechanism of O intercalation under graphene on metal surfaces because the Cu8,9,12 and Ni11,27 are widely used as catalytic surfaces in CVD graphene growth. Moreover, Cu and Ni represent two limiting cases in that Cu interacts with the graphene weakly but with Ni relatively strongly.

hemical vapor deposition (CVD) growth of graphene on transition metal (TM) catalyst surface with hydrocarbon as precursor has been broadly recognized as the most promising scalable approach for synthesis of high quality graphene1,2 for future industrial applications.3,4 The CVD grown graphene inevitably interacts strongly with the TM substrate. Not only this interaction but also the lattice mismatch with the TM substrate can perturb the properties and morphology of the supported graphene.5−7 More importantly, transferring CVD grown graphene to desired substrates is a prerequisite for various applications. Thus, decoupling CVD grown graphene from the TM substrate is a key step either to restore its freestanding properties or to facilitate its transfer for practical applications. Several techniques have been developed to transfer graphene from metal surface to desired substrate,8−11 such as the so-called roll-up-roll12 or face-to-face13 technique, in which the decoupling of graphene from TM substrate is achieved by etching the metal substrate in solution. However, these techniques involve aggressive process that may cause additional damage to the graphene layer and the metal substrate is not renewable. Alternatively, decoupling graphene from TM substrate can be realized by atom or molecule intercalation between the graphene and the TM surface. Experimentally, intercalations underneath CVD grown graphene on various metal surfaces, such as iridium (Ir),14−17 ruthenium (Ru),18−22 copper (Cu),23−25 nickel (Ni)26−28 surfaces, have been intensively studied. Particularly, oxygen atom intercalation underneath graphene on Ir(111),14,15 Ru(0001),18,19,21,29 Cu(111) and (110)23,24 and polycrystalline Cu25 surface have been achieved to make the graphene float on the metal surface. These experimental results suggest that the O © XXXX American Chemical Society

Received: August 23, 2015 Accepted: September 25, 2015

4099

DOI: 10.1021/acs.jpclett.5b01841 J. Phys. Chem. Lett. 2015, 6, 4099−4105

Letter

The Journal of Physical Chemistry Letters The zigzag (ZZ)- and armchair (AC)-edged graphene nanoribbons (GNRs) were placed on a three layer metal slab with bottom layer fixed to mimic the ZZ/AC graphene edge on metal surface, respectively. A test calculation to compare the results based on the PBE and the more accurate PBE-D235 methods indicates that the van der Waals correction on the reaction barrier of the oxygen atom and the metal−graphene edge interface is very small (∼4%). Considering the much-increased computing time for using the PBE-D2 method for a large system considered here, the PBE functional is thus selected for all calculations. (See part I in Supporting Information (SI) for details.) Carbon atoms at the edges of CVD grown graphene typically form covalent bonds with the metal surface. Here, unbinding graphene edges from the metal substrate upon oxygen intercalation means that the metal surface passivated graphene edges (labeled as G−M) are repassivated by oxygen atoms on metal surface (labeled as G−O@M). If the oxygen-terminated graphene edge is energetically more favorable than the metal passivation, oxygen intercalation would be possible, and as a result, preempt the metal−graphene interface. Otherwise, the intercalation would be unfeasible. First, we assume that oxygen atoms at the graphene edge and on the metal surface are in thermal equilibrium with O2 gas. When the chemical potential of oxygen atom (half of O2 molecule) is considered, we can assess the competition between the metal passivated ZZ (AC) graphene edge and the oxygenterminated ZZ (AC) graphene edges and obtain a phase diagram as shown in Figure 1 (See part II of SI for details of obtaining

What’s more, the binding strength of the graphene edge and the metal substrate can be greatly weakened by the oxygen passivation. We defined the binding energy (Ebind) as follows: E bind = (E T − EM − EG(G−O))/Nedge‐C

(1)

where ET is the total energy of metal-passivated graphene or oxygen-terminated graphene on the metal surface; EM is the energy of metal slab; EG and EG−O are the energy of suspended graphene without or with the oxygen termination at the edge, respectively; Nedge‑C is the number of carbon atoms on the graphene edge. The computed binding energies (Ebind) of the metal surface and the graphene edge (See part III of SI for details) show that, upon oxygen passivation, such a binding strength is reduced by ∼3/4 for ZZ edge on Cu(111)/Ni(111) surface and by ∼2/3 for AC edge on Cu(111)/Ni(111) surface (See data in Table S1 of SI). These results clearly prove that the oxygen termination will greatly weakens the binding strength of the graphene edge to the metal surface, thereby greatly benefit the oxygen intercalation. We have now shown that from the thermodynamic point of view, the oxygen terminated graphene edge is superior to the metal−graphene interface in common experimental conditions (e.g., 10−10 ∼ 1 bar, 300−600 K) and that the binding of the graphene edge and the metal surface is greatly weakened upon the oxygen termination. However, the kinetics of oxygen intercalation under graphene on metal surface is still incompletely known. Here, possible kinetic steps for the oxygen intercalation under the supported graphene island exposed in O2 gas or air are suggested as the following (also see Scheme 1), Scheme 1. Scheme of O Intercalation Underneath CVD-Grown Graphene on Metal Catalyst Substrate: (i) O2 Molecules Dissociate on Metal Surface; (ii) O Atoms Diffuse on Metal Surface; (iii) O Atoms Attach to the Graphene Edge; (iv) O Atoms Penetrate the Interface and Diffuse to the Graphene-Covered Surface; (v) Graphene Floats above Metal Surface with the Fully Intercalated O Atoms

Figure 1. Phase diagram for metal surface passivated graphene edges (G−M) and oxygen terminated graphene edges on metal surface (G−O@M). The insets represent the structures of G−M and G−O@M, respectively. The green, purple, and pink ellipses denote the experimental conditions of oxygen intercalation under graphene on Ir(111),14,15 Ru(0001),18,19,21 and Cu(111)24 surfaces, respectively.

from which we will identify the key step that dominates the reaction pathway: (i) O2 molecules are adsorbed on the bare metal surface and subsequently dissociated in the form of O radical pairs. (ii) Dissociated O atoms diffuse on the bare metal surface and some of them can meet the metal−graphene interface. (iii) At the metal−graphene interface, the O atoms can attach to the graphene edge C atoms while breaking the MC bond; as a result, the graphene edge starts to decouple from the metal surface. Depending on the MC distance and CC distance, either the MOC structures (where the graphene edge C atoms and metal surface are still bridged by the O atoms but the binding strength is greatly weakened) or carbonyl groups (CO) are formed.

the phase diagram). Clearly, the metal passivated graphene edge is thermodynamically favorable at high temperature and extremely low O2 pressure, for example, T = 800 K and pO2 < 10−19 bar. However, the oxygen-terminated graphene edge on a metal surface is thermodynamically more favorable over a broad parameter range, consistent with the experimental evidence of the oxygen intercalation under graphene on Ir(111),14,15 Ru(0001),18,19,21 or Cu(111)24 surfaces. Note that even under high O2 pressure the temperature should not be higher than a critical value, for example, ∼ 670 K, above which the oxygen exposure would cause rapid graphene etching.19,36 4100

DOI: 10.1021/acs.jpclett.5b01841 J. Phys. Chem. Lett. 2015, 6, 4099−4105

Letter

The Journal of Physical Chemistry Letters

Figure 2. (a) Side and top view of ZZ graphene edge on Cu(111); (b) the reaction pathway of the first O atom attaching to the ZZ graphene edge on Cu(111) surface; (c) side and top view of O bridged ZZ graphene edge and Cu(111); (d) the reaction path of an O atom penetrating the MOC line at ZZ graphene edge on Cu(111) surface. The black, red, pink balls denote the C, O, and Cu atoms, respectively. The incoming O atom for intercalation in (d) is highlighted in green. The relative energies are in electronvolts.

hollow site near the MOC line, an O atom of the MOC interface can be lifted up by forming a carbonyl group (CO) with its neighboring C atom (substep 1 in Figure 2d) due to the strong repulsion between the incoming O atom and the interface O atom. The incoming O atom then diffuses to the adjacent hollow site closer to the MOC line and lifts another neighboring OC pair in the MOC interface by forming the second carbonyl group (substep 2 in Figure 2d). The relevant activation barrier and the energy rise are 0.45 and 0.35 eV for substep 1 and 0.63 and 0.46 eV for substep 2, respectively. The two lifted carbonyl groups can facilitate the incoming O atom to the interior region of graphene covered metal surface. The incoming O atom further crosses the interface line (substep 3 in Figure 2d) by overcoming a low reaction barrier of 0.21 eV with a small energy gain of 0.12 eV. Therefore, the overall O atom intercalation encounters a relatively high reaction barrier of 1.02 eV and an energy rise of 0.69 eV, making this reaction unfavorable. Nevertheless, the incoming O atom can continue to migrate toward the interior region and only needs to overcome a rather low reaction barrier of 0.21 eV with an energy drop of 0.69 eV (substep 4 in Figure 2d). Therefore, the systems before and after O atom intercalation are energetically comparable. The relatively low reaction barrier and significant energy drop in substep 4 suggest that once an incoming O atom penetrates the MOC interface successfully, it would migrate into the interior region with ease. This is very important that the intercalated O atom can further diffuse toward the interior region easily to make room for the ensuing intercalation with newly and continuously supplied O atoms. The activation barriers in steps iii and iv are nearly same, ∼ 1.0 eV, suggesting that both steps are rate-limiting for the O intercalation. O Intercalation through the AC Graphene Edge on Cu(111). The mechanism of O intercalation through the AC graphene edge on

(iv) Continuously supplied O atoms pass through the MOC/CO interface and diffuse into the interior region between the graphene and the metal surface. (v) Eventually, the O intercalation enables the graphene layer to fully decouple from the O terminated metal surface. Our calculations (see Figures S5 and S6 of SI) show that the O2 molecules can be easily adsorbed and dissociate into radical pairs on Cu(111) or Ni(111) surface near or below room temperature, which is in good agreement with previous experimental37,38 and theoretical39,40 studies. Moreover, the diffusion barrier of O atom on the metal surface is only in the range of 0.25−0.45 eV for Cu(111)39 and 0.5−0.6 eV for Ni(111),40 so that O atoms are able to diffuse very quickly on the metal surface around or above the room temperature. Thus, neither the step i nor ii is the ratelimiting step for the O atom intercalation. Next, we consider several possible reaction pathways and compute the corresponding activation barriers involved in steps iii and iv to seek the rate− limiting step for the O atom intercalation. O Intercalation through the ZZ Graphene Edge on Cu(111). In step iii, for an O atom attaching to a C atom of the ZZ graphene edge on Cu(111), as shown in Figure 2b, the computed activation barrier is 0.99 eV with a relatively large energy drop of 1.09 eV from the reactant to the product. This energy drop indicates that the O atom passivated graphene edge is energetically favorable, and the formation of O-passivated graphene is exothermic. Upon the attachment with the O atom, the edge C atom is lifted upward and forms the MOC structure. Due to the large energy gain along with a modest reaction barrier, continuous attachment of O atoms to the edge C atoms would eventually decouple the ZZ graphene edge from the Cu(111) substrate by forming a closed MOC line along the interface (see Figure 2c). In step iv, a new incoming O atom penetrates the MOC line at ZZ graphene edge. When the O atom diffuses to a surface 4101

DOI: 10.1021/acs.jpclett.5b01841 J. Phys. Chem. Lett. 2015, 6, 4099−4105

Letter

The Journal of Physical Chemistry Letters

Figure 3. (a) Side and top view of AC graphene edge on Cu(111); (b) the reaction pathway of the first O atom attach to the AC graphene edge on Cu(111) surface; (c) side and top view of O passivated AC graphene edge on Cu(111); (d) the reaction path of an O atom penetrating the MOC/ CO line at AC graphene edge on Cu(111) surface.

the lifted carbonyl group pairs open a gap, making the step iv of O intercalation proceed easily. Taking the activation barrier in the rate-limiting step as a reference, the O intercalation through ZZ graphene edge on Cu(111) surface is very close to that of the AC edge (∼1.0 eV). Thus, the intercalation of O atom through either the ZZ or AC graphene edge on Cu(111) surface can occur on the order of a second at a mild temperature ∼420 K (estimated from the transition state theory, 1/τ ≈ 1012exp(−Eb/kT) ∼ 1, where Eb = 1.0 eV, τ = 1 s and the pre-exponential factor 1012 s−1 can be viewed as the lower limit of the (kT)/h for T in the range of 300− 700 K). This estimated reaction temperature is close to that reported in recent experimental study of O intercalation underneath graphene on Ir(111) surface,14,15 which the surface also interacts with graphene weakly. On the other hand, at room temperature (300 K) the reaction time is estimated to be ∼17.5 h, qualitatively consistent with the experimental result24 (a few days). Previous experimental studies show that at a critical temperature, for example, ∼ 670 K, the reaction rate of oxidation etching is comparable to that of the oxygen intercalation for graphene on the metal surface.19,36 But the rate of oxidation etching decreases more rapidly as the temperature is lowered and can be about two or more orders of magnitude lower than that of the oxygen intercalation when the T is in the range of 300−500 K.19 Thus, it is expected that the O intercalation on Cu(111) supported graphene would occur at a temperature well below ∼670 K, whereas the unwanted oxidation etching of graphene would be largely suppressed. O Intercalation through the ZZ and AC Graphene Edge on Ni(111). Compared with the Cu(111) surface, the O intercalation through ZZ graphene edge on Ni(111) surface is predicted to be much more difficult due to the higher reaction barrier (1.34 eV) for the attachment an O atom to the ZZ graphene edge on Ni(111) surface (see Figure 4b). The physical reason is that the binding of ZZ graphene edge with Ni(111) surface is much stronger than

Cu(111) surface appears to be different. The surface metal atoms and carbon atoms on the AC edge are not 1:1 matched (or bonded) along metal−graphene interface. The dimerization of C atoms at AC graphene edge makes the distance of edge C atoms alternatively change between 1.42 and 2.85 Å (Figure 3a). Hence, the oxygen passivation of AC edge on Cu(111) is expected to proceed differently than that of the ZZ edge. The activation barrier of an O atom attaching to the C atom of AC graphene edge is about 1.02 eV with an energy drop of 0.81 eV by forming the MOC structural unit (see Figure 3b). This reaction requires overcoming a similar activation barrier as that of ZZ edge on Cu(111). When all the edge C atoms are attached by the O atoms, the O atoms and the adjacent edge C dimers can form alternately staggered MOC pairs and carbonyl group pairs. This is different from the case of ZZ edge on Cu(111) in that only the MOC structure is formed. This difference stems from the strong repulsion of oxygen atoms at the AC graphene edge caused by the short CC distances (1.42 Å) of C dimer at the AC graphene edge. Consequently, four edge C atoms (or two CC dimers) in the supercell form two MOC units and two carbonyl groups in the sequence of 2MOC··· 2CO (See Figure S7 in SI for top and front view), and two neighboring carbonyl groups are lifted up. The two lifted neighboring carbonyl groups upon O passivation of AC graphene edge can greatly facilitate the intercalation of an O atom. In this step, the new incoming O atom can easily migrate from the bare Cu(111) surface to the graphene covered region by overcoming two low energy barriers of 0.67 and 0.31 eV, respectively (Figure 3d). The overall reaction barrier (0.67 eV) is much lower than that of ZZ graphene edge (1.02 eV) on Cu(111) surface due to the wider gap opened by two lifted neighboring carbonyl groups. In summary, for O intercalation through AC graphene edge, only the attachment of O atom to the graphene edge C atom, namely, step iii, is the rate-limiting step. Once all AC graphene edge C atoms are attached by the O atoms, 4102

DOI: 10.1021/acs.jpclett.5b01841 J. Phys. Chem. Lett. 2015, 6, 4099−4105

Letter

The Journal of Physical Chemistry Letters

Figure 4. (a) Side and top view of ZZ graphene edge on Ni(111); (b) the reaction pathway of the first O atom attach to the ZZ graphene edge on Ni(111) surface; (c) side and top view of O bridged ZZ graphene edge and Ni(111); (d) the reaction path of an O atom penetrating the MOC/ CO interface at ZZ graphene edge on Ni(111) surface. The cyan balls denote the Ni atoms.

O atoms, two neighboring MOC and two neighboring carbonyl groups are formed at the O passivated AC graphene on Ni(111) surface. Similar to the AC graphene edge on Cu(111) surface, this can also be attributed to the strong repulsion of O atoms caused by the short CC distance (1.42 Å) of AC edge dimer (See Figure S8 in SI for top and front views). The incoming O atom migrates to the graphene covered surface by overcoming the reaction barriers of 0.70, 0.58, and 0.80 eV, respectively (see Figure 5d), giving an overall energy barrier of 1.24 eV. Thus, the rate-limiting step of O intercalation through the AC graphene edge on Ni(111) surface is the penetration of O atoms (step iv). On the basis of the transition state theory, the estimated time required for the O intercalation through AC graphene edge on Ni(111) surface is about one second at ∼520 K (Eb = 1.24 eV). For the ZZ graphene edge on Ni(111), however, the O intercalation on the same time scale would require a high temperature of ∼760 K (Eb = 1.81 eV), at which point graphene oxidation etching would be pronounced. In conclusion, we have theoretically investigated the atomic mechanism of oxygen intercalation through metal passivated graphene edges to decouple CVD grown graphene from metal substrate on thermodynamic and kinetics aspects, and successfully identified the rate-limiting steps. It is suggested that the O intercalation through the ZZ and AC graphene edge can be achieved on Cu(111) surface by overcoming an energy barrier of ∼1.0 eV. However, on Ni(111) surface, intercalating O atom through the ZZ graphene edge is much more difficult than the AC edge due to the reaction barriers being as high as 1.81 and 1.24 eV, respectively. The estimated temperature required for the O intercalation through the ZZ and AC edge on Cu(111) to be on the time scale of second is ∼420 K, and ∼520 K through the

that with Cu(111) surface (2.24 eV vs 1.64 eV per edge carbon atom). Although the step iii is still energetically favorable, the smaller energy drop (0.21 eV) implies that the capability of ZZ graphene edge to adsorb O atom on Ni(111) surface is much weaker than that on Cu(111) surface. Moreover, the step iv for the incoming O atom to penetrate the MOC interface of ZZ graphene edge on Ni(111) surface is rather complicated. It actually involves three substep reactions: an incoming O atom approaches the MOC interface; two neighboring O atoms are then lifted up; and two carbonyl groups are finally formed. The corresponding three reaction barriers are 0.83, 1.00, and 0.45 eV, respectively, accompanying with relatively high energy rises of 0.58 and 0.78 eV, but a rather small energy drop of 0.12 eV, respectively (see substep 1−3 in Figure 4d). Thus, the overall reaction barrier of step iv is as high as 1.81 eV and the total energy rise is as high as 1.24 eV, suggesting that the step iv of O intercalation through the ZZ graphene edge on Ni(111) surface is very difficult and it is the rate-limiting step. The high reaction barrier is due to the strong O−Ni interaction between the O atom and Ni(111) surface of MOC interface, whereas the high energy rise stems from the strong repulsion between the incoming O atom and MOC interface. Similar to that of ZZ edge on Cu(111) surface, the O atom can easily diffuse toward the interior region by overcoming two low reaction barriers of 0.25 and 0.32 eV with energy drops of 0.37 and 0.60 eV, respectively (see substep 4 and 5 in Figure 4d). In contrast, the O intercalation through the AC graphene edge on Ni(111) surface is much easier. The attachment of an O atom to the AC edge on Ni(111) only needs to overcome a reaction barrier of 1.0 eV (see Figure 5b). When the AC graphene edge (four C atoms in dimer form) on Ni(111) is fully attached by 4103

DOI: 10.1021/acs.jpclett.5b01841 J. Phys. Chem. Lett. 2015, 6, 4099−4105

Letter

The Journal of Physical Chemistry Letters

Figure 5. (a) Side and top view of AC graphene edge on Ni(111); (b) the reaction pathway of the first O atom attach to the AC graphene edge on Ni(111) surface; (c) side and top view of O passivated AC graphene edge on Ni(111); (d) the reaction path of an O atom penetrating the MOC/ CO interface at AC graphene edge on Ni(111) surface.

Notes

AC graphene edge on Ni(111) surface. Both are well lower than the critical temperature (e.g., ∼ 670 K) for the unwanted graphene oxidation etching to become considerable. Besides, the rough graphene edge, the wrinkle of graphene layer, and the step edge of metal surface in realistic experiments can further promote the oxygen intercalation.41,42 Thus, O intercalation is indeed a promising and fast way to decouple graphene from the metal catalyst surface, thereby producing free-standing graphene in large scale but with less damage if the experimental conditions are delicately tuned within the temperature range of 300−500 K and low O2 gas pressure range of 10−10−10−6 bar. Once the graphene is decoupled from the metal surface, the metal catalyst substrate can be reusable for catalytic graphene CVD growth.



The authors declare no competing financial interest.

ACKNOWLEDGMENTS



REFERENCES

J.W. is supported by the NBRP (2011CB302004), NSFC (21173040, 21373045), NSF of Jiangsu (BK20130016), SRFDP (20130092110029) of China. XCZ is supported by the National Science Foundation (NSF) through the Nebraska Materials Research Science and Engineering Center (MRSEC) (grant No. DMR-1420645). The authors thank the computational facilities of SEU, National Supercomputing Center (TH-1A) in Tianjin, and Holland Computing Center in University of Nebraska Lincoln.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpclett.5b01841. The model details and the top/side view of optimized supercell of graphene edges on metal surface, the comparison between the PBE and the PBE-D2 methods for reaction barrier calculation, the details of the thermodynamic phase diagram and binding energy calculations, the reaction pathways and barrier heights of O2 molecule dissociation and oxygen atom migration on metal surface, the top and front view of oxygen atom attached ZZ/AC graphene edge on metal surface. (PDF)





(1) Zhang, Y.; Zhang, L.; Zhou, C. Review of Chemical Vapor Deposition of Graphene and Related Applications. Acc. Chem. Res. 2013, 46, 2329−2339. (2) Zhang, X.; Li, H.; Ding, F. Self-Assembly of Carbon Atoms on Transition Metal SurfacesChemical Vapor Deposition Growth Mechanism of Graphene. Adv. Mater. 2014, 26, 5488−5495. (3) Tang, Q.; Zhou, Z.; Chen, Z. Graphene-Related Nanomaterials: Tuning Properties by Functionalization. Nanoscale 2013, 5, 4541−4583. (4) Ma, L.; Wang, J.; Ding, F. Recent Progress and Challenges in Graphene Nanoribbon Synthesis. ChemPhysChem 2013, 14, 47−54. (5) Pletikosić, I.; Kralj, M.; Pervan, P.; Brako, R.; Coraux, J.; N’Diaye, A. T.; Busse, C.; Michely, T. Dirac Cones and Minigaps for Graphene on Ir(111). Phys. Rev. Lett. 2009, 102, 056808. (6) Stradi, D.; Barja, S.; Dı ́az, C.; Garnica, M.; Borca, B.; Hinarejos, J. J.; Sánchez-Portal, D.; Alcamí, M.; Arnau, A.; Vázquez de Parga, A. L.; et al. Role of Dispersion Forces in the Structure of Graphene Monolayers on Ru Surfaces. Phys. Rev. Lett. 2011, 106, 186102.

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. 4104

DOI: 10.1021/acs.jpclett.5b01841 J. Phys. Chem. Lett. 2015, 6, 4099−4105

Letter

The Journal of Physical Chemistry Letters

Graphene Growth on Polycrystalline Copper. Nano Lett. 2013, 13, 4769−4778. (26) Dedkov, Y. S.; Fonin, M.; Rüdiger, U.; Laubschat, C. GrapheneProtected Iron Layer on Ni(111). Appl. Phys. Lett. 2008, 93, 022509. (27) Varykhalov, A.; Sánchez-Barriga, J.; Shikin, A. M.; Biswas, C.; Vescovo, E.; Rybkin, A.; Marchenko, D.; Rader, O. Electronic and Magnetic Properties of Quasifreestanding Graphene on Ni. Phys. Rev. Lett. 2008, 101, 157601. (28) Voloshina, E. N.; Generalov, A.; Weser, M.; Böttcher, S.; Horn, K.; Dedkov, Y. S. Structural and Electronic Properties of the Graphene/ Al/Ni(111) Intercalation System. New J. Phys. 2011, 13, 113028. (29) Lizzit, S.; Larciprete, R.; Lacovig, P.; Dalmiglio, M.; Orlando, F.; Baraldi, A.; Gammelgaard, L.; Barreto, L.; Bianchi, M.; Perkins, E.; et al. Transfer-Free Electrical Insulation of Epitaxial Graphene from Its Metal Substrate. Nano Lett. 2012, 12, 4503−4507. (30) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made Simple. Phys. Rev. Lett. 1996, 77, 3865−3868. (31) Kresse, G.; Furthmüller, J. Efficient Iterative Schemes for Ab Initio Total-Energy Calculations Using a Plane-Wave Basis Set. Phys. Rev. B: Condens. Matter Mater. Phys. 1996, 54, 11169−11186. (32) Blöchl, P. E. Projector Augmented-Wave Method. Phys. Rev. B: Condens. Matter Mater. Phys. 1994, 50, 17953−17979. (33) Kresse, G.; Joubert, D. From Ultrasoft Pseudopotentials to the Projector Augmented-Wave Method. Phys. Rev. B: Condens. Matter Mater. Phys. 1999, 59, 1758−1775. (34) Henkelman, G.; Uberuaga, B. P.; Jónsson, H. A Climbing Image Nudged Elastic Band Method for Finding Saddle Points and Minimum Energy Paths. J. Chem. Phys. 2000, 113, 9901−9904. (35) Grimme, S. Semiempirical GGA-Type Density Functional Constructed with a Long-Range Dispersion Correction. J. Comput. Chem. 2006, 27, 1787−1799. (36) Starodub, E.; Bartelt, N. C.; McCarty, K. F. Oxidation of Graphene on Metals. J. Phys. Chem. C 2010, 114, 5134−5140. (37) Sueyoshi, T.; Sasaki, T.; Iwasawa, Y. Molecular and Atomic Adsorption States of Oxygen on Cu(111) at 100−300 K. Surf. Sci. 1996, 365, 310−318. (38) Shayegan, M.; Cavallo, J. M.; Glover, R. E.; Park, R. L. Precursor Adsorption of Oxygen on Ni(111) and the Activation Energy for Chemisorption. Phys. Rev. Lett. 1984, 53, 1578−1581. (39) Xu, Y.; Mavrikakis, M. Adsorption and Dissociation of O2 on Cu(111): Thermochemistry, Reaction Barrier and the Effect of Strain. Surf. Sci. 2001, 494, 131−144. (40) Mittendorfer, F.; Eichler, A.; Hafner, J. Molecular Precursors in the Dissociative Adsorption of O2 on Ni(111). Surf. Sci. 1999, 433−435, 756−760. (41) Kimouche, A.; Renault, O.; Samaddar, S.; Winkelmann, C.; Courtois, H.; Fruchart, O.; Coraux, J. Modulating Charge Density and Inelastic Optical Response in Graphene by Atmospheric Pressure Localized Intercalation through Wrinkles. Carbon 2014, 68, 73−79. (42) Vlaic, S.; Kimouche, A.; Coraux, J.; Santos, B.; Locatelli, A.; Rougemaille, N. Cobalt Intercalation at the Graphene/Iridium(111) Interface: Influence of Rotational Domains, Wrinkles, and Atomic Steps. Appl. Phys. Lett. 2014, 104, 101602.

(7) Preobrajenski, A. B.; Ng, M. L.; Vinogradov, A. S.; Mårtensson, N. Controlling Graphene Corrugation on Lattice-Mismatched Substrates. Phys. Rev. B: Condens. Matter Mater. Phys. 2008, 78, 073401. (8) Li, X.; Cai, W.; An, J.; Kim, S.; Nah, J.; Yang, D.; Piner, R.; Velamakanni, A.; Jung, I.; Tutuc, E.; et al. Large-Area Synthesis of HighQuality and Uniform Graphene Films on Copper Foils. Science 2009, 324, 1312−1314. (9) Gao, L. B.; Ren, W. C.; Zhao, J. P.; Ma, L. P.; Chen, Z. P.; Cheng, H. M. Efficient Growth of High-Quality Graphene Films on Cu Foils by Ambient Pressure Chemical Vapor Deposition. Appl. Phys. Lett. 2010, 97, 183109. (10) Kang, J.; Hwang, S.; Kim, J. H.; Kim, M. H.; Ryu, J.; Seo, S. J.; Hong, B. H.; Kim, M. K.; Choi, J.-B. Efficient Transfer of Large-Area Graphene Films onto Rigid Substrates by Hot Pressing. ACS Nano 2012, 6, 5360−5365. (11) Kim, K. S.; Zhao, Y.; Jang, H.; Lee, S. Y.; Kim, J. M.; Kim, K. S.; Ahn, J.-H.; Kim, P.; Choi, J.-Y.; Hong, B. H. Large-Scale Pattern Growth of Graphene Films for Stretchable Transparent Electrodes. Nature 2009, 457, 706−710. (12) Bae, S.; Kim, H.; Lee, Y.; Xu, X. F.; Park, J. S.; Zheng, Y.; Balakrishnan, J.; Lei, T.; Kim, H. R.; Song, Y. I.; et al. Roll-to-Roll Production of 30-Inch Graphene Films for Transparent Electrodes. Nat. Nanotechnol. 2010, 5, 574−578. (13) Gao, L.; Ni, G.-X.; Liu, Y.; Liu, B.; Castro Neto, A. H.; Loh, K. P. Face-to-Face Transfer of Wafer-Scale Graphene Films. Nature 2014, 505, 190−194. (14) Grånäs, E.; Knudsen, J.; Schröder, U. A.; Gerber, T.; Busse, C.; Arman, M. A.; Schulte, K.; Andersen, J. N.; Michely, T. Oxygen Intercalation under Graphene on Ir(111): Energetics, Kinetics, and the Role of Graphene Edges. ACS Nano 2012, 6, 9951−9963. (15) Larciprete, R.; Ulstrup, S.; Lacovig, P.; Dalmiglio, M.; Bianchi, M.; Mazzola, F.; Hornekær, L.; Orlando, F.; Baraldi, A.; Hofmann, P.; et al. Oxygen Switching of the Epitaxial Graphene−Metal Interaction. ACS Nano 2012, 6, 9551−9558. (16) Meng, L.; Wu, R.; Zhou, H.; Li, G.; Zhang, Y.; Li, L.; Wang, Y.; Gao, H.-J. Silicon Intercalation at the Interface of Graphene and Ir(111). Appl. Phys. Lett. 2012, 100, 083101. (17) Herbig, C.; Kaiser, M.; Bendiab, N.; Schumacher, S.; Förster, D. F.; Coraux, J.; Meerholz, K.; Michely, T.; Busse, C. Mechanical Exfoliation of Epitaxial Graphene on Ir(111) Enabled by Br 2 Intercalation. J. Phys.: Condens. Matter 2012, 24, 314208. (18) Zhang, H.; Fu, Q.; Cui, Y.; Tan, D.; Bao, X. Growth Mechanism of Graphene on Ru(0001) and O2 Adsorption on the Graphene/Ru(0001) Surface. J. Phys. Chem. C 2009, 113, 8296−8301. (19) Sutter, P.; Sadowski, J. T.; Sutter, E. A. Chemistry under Cover: Tuning Metal−Graphene Interaction by Reactive Intercalation. J. Am. Chem. Soc. 2010, 132, 8175−8179. (20) Mao, J.; Huang, L.; Pan, Y.; Gao, M.; He, J.; Zhou, H.; Guo, H.; Tian, Y.; Zou, Q.; Zhang, L.; et al. Silicon Layer Intercalation of Centimeter-Scale, Epitaxially Grown Monolayer Graphene on Ru(0001). Appl. Phys. Lett. 2012, 100, 093101. (21) Sutter, P.; Albrecht, P.; Tong, X.; Sutter, E. Mechanical Decoupling of Graphene from Ru(0001) by Interfacial Reaction with Oxygen. J. Phys. Chem. C 2013, 117, 6320−6324. (22) Jin, L.; Fu, Q.; Yang, Y.; Bao, X. A Comparative Study of Intercalation Mechanism at Graphene/Ru(0001) Interface. Surf. Sci. 2013, 617, 81−86. (23) Walter, A. L.; Nie, S.; Bostwick, A.; Kim, K. S.; Moreschini, L.; Chang, Y. J.; Innocenti, D.; Horn, K.; McCarty, K. F.; Rotenberg, E. Electronic Structure of Graphene on Single-Crystal Copper Substrates. Phys. Rev. B: Condens. Matter Mater. Phys. 2011, 84, 195443. (24) Lu, A.-Y.; Wei, S.-Y.; Wu, C.-Y.; Hernandez, Y.; Chen, T.-Y.; Liu, T.-H.; Pao, C.-W.; Chen, F.-R.; Li, L.-J.; Juang, Z.-Y. Decoupling of CVD Graphene by Controlled Oxidation of Recrystallized Cu. RSC Adv. 2012, 2, 3008−3013. (25) Kidambi, P. R.; Bayer, B. C.; Blume, R.; Wang, Z.-J.; Baehtz, C.; Weatherup, R. S.; Willinger, M.-G.; Schloegl, R.; Hofmann, S. Observing Graphene Grow: Catalyst−Graphene Interactions During Scalable 4105

DOI: 10.1021/acs.jpclett.5b01841 J. Phys. Chem. Lett. 2015, 6, 4099−4105