Raman spectroscopic characterization of stacking configuration and interlayer coupling of twisted multilayer graphene grown by chemical vapor deposition
Jiang-Bin Wu, Huan Wang, Xiao-Li Li, Hailin Peng, Ping-Heng Tan
aa r X i v : . [ c ond - m a t . m t r l - s c i ] S e p Raman spectroscopic characterization of stacking configuration and interlayercoupling of twisted multilayer graphene grown by chemical vapor deposition
Jiang-Bin Wu a , Huan Wang b , Xiao-Li Li a , Hailin Peng b , Ping-Heng Tan a, ∗ a State Key Laboratory of Superlattices and Microstructures, Institute of Semiconductors, Chinese Academy of Sciences, Beijing 100083, China b Center for Nanochemistry, Beijing National Laboratory for Molecular Sciences, Key Laboratory for the Physics and Chemistry of Nanodevices,College of Chemistry and Molecular Engineering, Peking University, Beijing 100871, China
Abstract
Multilayer graphene (MLG) grown by chemical vapor deposition (CVD) is a promising material for electronic andoptoelectronic devices. Understanding the stacking configuration and interlayer coupling of MLGs is technologicallyrelevant and of importance for the device applications. Here, we reported a kind of twisted MLGs (tMLGs), which areformed by stacking one graphene monolayer on the top of AB-stacked MLG by rotating a certain angle between them.The twist angle of tMLGs are identified by the twist-related modes, R and R’. With increasing the total layer numberof N, the observed interlayer shear modes in the tMLG flake always follow those of AB-stacked (N-1)LG, while theobserved interlayer breathing modes always follow those of AB-stacked NLG, independent of its twist angle. The layerbreathing coupling of the tMLGs is almost identical to that of mechanically-exfoliated tMLGs, which demonstrates thehigh quality of MLGs grown by CVD. This study provides an applicable approach to probe the stacking configurationand interlayer coupling of MLGs grown by CVD or related methods. This work also demonstrates the possibility togrow MLG flakes with a fixed stacking configuration, e . g . , t(1 + n )LG, by the CVD method.
1. Introduction
Large-area multilayer graphene (MLG) grown bychemical vapor deposition (CVD), with transparency andhigh electrical conductivity and flexibility, is consideredas a candidate for transparent and conducting electrodes,which can be used in touch screen panels, organic light-emitting diodes and solar cells[1–8]. The plane-to-plane(vertical) conductivity, determined by the interlayer cou-pling of MLG, is the bottleneck of improving the over-all conductivity, due to the series resistance e ff ect[9, 10].Therefore, it’s important to produce and characterizethe MLG with an interlayer coupling close to that ofgraphite. The CVD method tends to produce twisted bi-layer graphene (2LG)[11–14], ∗ Corresponding author.
Email address: [email protected] (Ping-Heng Tan)
Twisted 2LG (t2LG) is non-AB stacked bilayergraphene in which one graphene monolayer sheet ro-tates by a certain angle ( θ t ) relative to the other[15, 16].t2LG brings about a series of novel physical properties, e . g . , lower Fermi velocity than single layer graphene(SLG)[16, 17] and θ t -dependent optical absorption[18,19], and can be employed as photoelectric detector[20]and pressure sensors[21, 22]. Similarly, by assemblingm-layer (mLG, m >
1) and n-layer (nLG, n ≥
1) flakes,a (m + n)-system can be formed, which is denoted ast(m + n)LG, a kind of twisted MLG (tMLG). In general,for a given total layer number N (with N = n + m + ...),t( n + m + ...)LG is assumed to denote the N layer graphene,which is stacked by each AB-stacked n , m ...LG with twistangles between them. The stacking configuration infor-mation in a t(n + m + ...)LG includes the layer number ofeach constituent and the stacking way between two adja-cent constituents.[23] tMLG (layer number >
2) also ex-
Preprint submitted to Carbon September 6, 2016 ibits a series of novel physical properties[23], similarto the case of t2LG. One can easily grow t2LG or AB-stacked 2LG (AB-2LG) by CVD to control its flake sizeand stacking configuration. However, it will become moreand more important to grow graphene flakes by CVDto create tMLG on demand with properties determinedby the interlayer interaction and stacking configuration.When more graphene layers are grown by CVD, it’s stilla challenge to characterize the stacking configuration andinterlayer coupling of CVD-grown MLG (CVD-MLG) di-rectly and nondestructively.Raman spectroscopy is one of the most useful char-acterization techniques in graphene[24, 25].The phononvibration modes of AB-stacked MLG (AB-MLG) canbe divided into the in-plane vibration modes (like Gand 2D modes) and out-plane vibration modes (like theshear (C) mode). Besides G and 2D modes, two ad-ditional R and R ′ modes, which are from the TO andLO phonon branches respectively, selected by a twistvector, are detected in t(n + m)LG[23, 26–28], and canbe a signature to distinguish the stacking configurationbetween twist and AB stacking. The layer number ofAB-MLG can be determined by the 2D[29] and C[30]modes. The 2D peak profile of the AB-MLG is layer-number dependent[29], because the electronic structuresof the AB-MLG are distinct[31] and the 2D peak pro-file is related to its electronic structures by the doubleresonant process[32]. However, the twist angle depen-dent electronic structures[17] in the tMLGs lead to muchmore complicated profile of 2D peaks[33, 34], whicharen’t suitable to identify the layer number of tMLGs.Moreover, twisting would block the interlayer shear cou-pling, resulting in the localization of the C modes inthe AB-stacked constituent[23]. Fortunately, the inter-layer breathing coupling remains almost constant at thetwisted interface, and the layer breathing (LB) modes(LBMs), which can not be measured in MLG, can be de-tected in tMLG under the resonant condition.[35] There-fore, by monitoring the C and LB modes, the stack-ing configuration between adjacent constituents in tMLGcan be determined. Moreover, the C and LB modescorrespond to interlayer vibrations, whose frequenciesare directly related to the strength of interlayer cou-pling strengh[30, 36]. Therefore, it is necessary to ex-tend the method of identifying stacking configuration andinterlayer coupling strength in mechanically-exfoliated tMLGs (ME-tMLGs) by ultralow-frequency (ULF) Ra-man spectroscopy to MLG flakes grown by CVD.Here, we employ CVD method to prepare MLGflakes containing di ff erent layer numbers ranging from1 to 7, whose layer number is determined by Rayleighimaging[37]. The CVD-MLGs are identified as twisted(1 + n )LG by the R and R ′ modes. In the ULF region,with increasing the total layer number of N, the observedC modes in the tMLG with a definite θ t always followthose of AB-(N-1)LG, while its observed LBMs alwaysfollow those of AB-NLG. The tMLG with a definite totallayer number of N exhibits similar spectral features ( e . g . ,the number and frequency of the C and LB modes), in-dependent of θ t . Based on the frequencies of C and LBmodes, we revealed that the interlayer C and LB cou-plings in tMLGs prepared by both CVD method and self-folding during the mechanical exfoliation process are al-most identical to each other, respectively, which indicatesthe high crystal quality of tMLGs grown by the CVDmethod.
2. Experimental
MLG flacks were grown on annealed copper substrate(Alfa Aesar) in a homemade low-pressure chemical vapordeposition (LPCVD) system. Before growth, the copperfoil was under pre-annealed treatment at 1020 ◦ C under100 sccm H with a pressure of about 200 Pa for 30minto reduce oxide at the surface of copper. H flow waschanged into 600 sccm and 1 sccm methane was intro-duced while keeping the temperature constant of 1020 ◦ C. After 40 min growth, the furnace was cooled downto room temperature. MLG flacks grown on copper weretransferred onto Si substrate with (90 nm) SiO thicknesswith the aid of Poly (methylmethacrylate). Raman spectra are measured in back-scattering at roomtemperature with a Jobin-Yvon HR800 Raman system,equipped with a liquid-nitrogen-cooled charge-coupleddevice (CCD), a 100 × objective lens (NA = + laser, and 488, 466 nm from an Ar + laser.2he resolution of the Raman system at 633 nm is 0.35cm − per CCD pixel. Plasma lines are removed from thelaser signals, using a BragGrate Bandpass filters. Mea-surements down to 5 cm − for each excitation are enabledby three BragGrate notch filters with optical density 3 andwith FWHM = − . [30] Both BragGrate bandpassand notch filters are produced by OptiGrate Corp. Thetypical laser power of ∼
3. Results and Discussions
The optical image of a typical CVD-grown polycrys-talline MLG flake is shown in Fig. 1(a). The layernumber of the MLG flake is identified by Rayleighscattering[37]. Fig. 1(b) shows contrast mapping ofRayleigh scattering of a MLG flake obtained by 532 nmexcitation. Rayleigh scattering signals are processed bythe mean intensity of the substrate: I ( contrast ) = ( I ( sub . )- I ( nLG )) / I ( sub . ), where I ( sub . ) is the Rayleigh intensityfrom bara substrate and I ( nLG ) is the Rayleigh inten-sity from the MLG flakes laid on the substrate. The re-gion of each graphene flake with a definite layer num-ber can be distinguished by I(contrast) or the color inthe contrast mapping of Rayleigh scattering, as markedin the image. The CVD-2LG tends to be t(1 + + θ t . The relationship be-tween VHS and θ t can be estimated by this formula[33]of E VHS ≈ πθ t ~ v f / a , where a is the lattice constant ofgraphene (2.46 Å), ~ is the reduced Planck’s constant, and v f is the Fermi velocity of SLG (10 m / s ). The mappingof the G mode intensity (I(G)) of the CVD-MLG flakeunder the excitation of 633 nm (1.96 eV) is shown in Fig.1(c). The CVD-MLG flake can be distinguished with dif-ferent zones by di ff erent enhancement levels of I(G). Eachzone is indicated by the white dash lines, and marked asZ1, Z2, Z3 and Z4, respectively. We also marked eachzone in Figs. 1(a) and 1(b). The Raman spectrum of theCVD-2LG flake from each zone is shown in Fig. 1(d).The so-called R and R ′ modes[23, 26] are observed inall the spectra, which suggests that the CVD-2LG flakesare t(1 + θ t from AB stacking. θ t ofgraphene flakes in the each zone can be indicated by theposition of R and R ′ modes,[23, 26, 28, 38] and are la-beled in the mapping image in Fig. 1(c). The t(1 + θ t of 11.5 ◦ corresponds to an optimalexcitation energy of 2.04 eV, which is close to the excita-tion energy of 1.96 eV, leading to a strong enhancementof I(G) in the Z1 zone. θ t is more close to 10.8 ◦ , the I(G)enhancement would be more significant under the 1.96 eV(633 nm) excitation.[23] We notice that the G mode inten-sity of the CVD-MLG flake, whose layer number is > + + + + + + + + + + + + + + ∼
720 nm in the optical contrast, dueto the presence of VHS in the JDOS.[18, 19], as shownin Fig. 2(a). Compared with AB-3LG, the adsorptionpeak at 720 nm still exists in the optical contrast of ME-t(1 + + ∼
620 nm. The same adsorption peak at ∼
720 nm arefound in ME-t(1 + + + + + + + − , whichcorresponds to the twisted interface with a θ t of 9.2 ◦ . Twoadditional peaks are measured at 1500 cm − and 1626cm − in the ME-t(1 + + − is the R ′ modecorresponding to the twisted interface between the bottomtwo graphene layers, which shares the same twisted vec-3 (cid:3)(cid:4)(cid:3)(cid:5)(cid:1)(cid:6)(cid:7)(cid:8)(cid:9)(cid:10)(cid:1)(cid:11)(cid:12)(cid:4) (cid:13)(cid:14) (cid:15)(cid:14)(cid:16)(cid:17)(cid:17) (cid:14)(cid:16)(cid:16)(cid:17) (cid:14)(cid:18)(cid:17)(cid:17) (cid:14)(cid:18)(cid:16)(cid:17) (cid:19) (cid:5) (cid:10) (cid:20)(cid:5) (cid:6) (cid:8) (cid:10) (cid:21) (cid:1) (cid:11) (cid:3) (cid:22) (cid:23) (cid:22) (cid:15) (cid:24)(cid:25)(cid:24)(cid:26)(cid:24)(cid:14)(cid:24)(cid:14) (cid:11)(cid:3)(cid:15) (cid:11)(cid:12)(cid:15) (cid:27)(cid:14) (cid:27)(cid:28)(cid:27)(cid:26)(cid:27)(cid:29)(cid:27)(cid:14) (cid:27)(cid:28)(cid:27)(cid:26)(cid:27)(cid:29) (cid:27)(cid:29)(cid:27)(cid:28)(cid:27)(cid:26)(cid:27)(cid:14) (cid:2) (cid:2)(cid:30) (cid:14)(cid:14)(cid:22)(cid:16) (cid:31) (cid:22)(cid:28) (cid:31) (cid:14)(cid:26)(cid:22)(cid:29) (cid:31) !(cid:22)(cid:18) (cid:31) (cid:16)(cid:1) (cid:2) (cid:4) (cid:11)"(cid:15) (cid:2)(cid:3)(cid:4)(cid:5)(cid:3)(cid:4)(cid:6)(cid:3)(cid:4)(cid:7)(cid:3)(cid:4) (cid:8)(cid:3)(cid:4)(cid:9)(cid:3)(cid:4)(cid:10)(cid:3)(cid:4) (cid:14) (cid:11)%(cid:15) (cid:14)(cid:14)(cid:22)(cid:16) (cid:31) (cid:22)(cid:28) (cid:31) (cid:14)(cid:26)(cid:22)(cid:29) (cid:31) !(cid:22)(cid:18) (cid:31) (cid:18)(cid:26)(cid:26)(cid:1)(cid:5)(cid:4)(cid:18)(cid:26)(cid:26)(cid:1)(cid:5)(cid:4) $(cid:19)(cid:11)$(cid:15) &’((cid:13)(cid:28) (cid:2) (cid:2) Figure 1:
Characterization of CVD-grown MLG. ( a ) Optical image of a CVD-MLG flake. ( b ) Contrast mapping of Rayleigh scattering of theCVD-MLG flake shown in (a) under the 532 nm excitation, from which the layer numbers of the MLG flakes in di ff erent regions are determinedand marked. ( c ) Raman image of the G mode intensity under the 633 nm excitation for the corresponding MLG flake. The zones with di ff erent Gmode intensities are indicated by white dash lines and marked as Z1, Z2, Z3 and Z4, respectively. The images in (a) (b) (c) are with the same scale.( d ) Raman spectra of CVD-2LG flakes in di ff erent zones in (c). The scale factor of each spectrum is marked. tor with the R peak at 1531 cm − of the ME-t(1 + − are denotedas R and R ′ respectively. The peak at 1500 cm − indi-cating a 12.4 ◦ twist angle is denoted as R , correspondingto the other twisted interface between the top two layersof ME-t(1 + + ff erence between the tMLG flake and the cor-responding AB-MLGs, and also from the sets of the R andR ′ modes in the corresponding Raman spectra.Next, we checked a CVD-MLG flake containing t N LG(2 ≤ N ≤ ∼
640 nmin the t N LG if we compare them with those of the cor-responding AB- N LG. This suggests that there only existsone twisted interface in CVD-grown t N LG for N >
2, andthe twisted interface should be the same as that of CVD-grown t2LG (i.e., t(1 + N LG can be identified as t(1 + n )LG( n = N -1). To further confirm the above conclusion, theRaman spectra of the CVD-grown t N LG are measured,as shown in Fig. 2(d). A couple of R and R ′ peaksare observed in 1509 and 1621 cm − in the CVD-2LG,which confirms that this CVD-2LG is a t(1 + ◦ . For the CVD-3LG, only the same couple of R and R ′ peaks are observed, indicating no more twisted inter-face in this flake, which means that this CVD-3LG is at(1 + + + + / R ′ -related Raman spec-tra of other CVD-MLG flakes also reveals their stackingway of t(1 + n )LG under the present growth condition.The stacking configuration of ME-tMLGs had been in-vestigated by the C and LB modes,[23, 35] which can onlybe observed within the resonant excitation window. The488 nm laser is used to excite the Raman spectra of theCVD-grown t(1 + n )LG with a θ t of 16.9 ◦ , as shown inFig. 3. For comparison, the Raman spectrum of ME-t(1 + + N LG, where N is the layer number, there are N -1 C andLB modes, which are denoted as C NN − i and LB NN − i ( i = , , ..., N − C N and LB N (i.e., i = N −
1) are the C and LB modes with the highest frequen-cies, respectively. All the frequencies of the C and LBmodes with di ff erent layer numbers can be calculated bya linear chain model (LCM), where each graphene layeris considered as one ball to calculate the frequency ofthe C and LB modes and only nearest-neighbor interlayerinteractions are taken into account.[23, 30, 35, 36] Fort(1 + ∼
28 cm − , iden-tified as C , because it’s close to the frequency of C modein AB-2LG[30], 31 cm − . According to the peak position,4 (cid:3)(cid:4)(cid:3)(cid:5)(cid:1)(cid:6)(cid:7)(cid:8)(cid:9)(cid:10)(cid:1)(cid:11)(cid:12)(cid:4) (cid:13)(cid:14) (cid:15)(cid:14)(cid:16)(cid:17)(cid:18) (cid:14)(cid:17)(cid:18)(cid:18) (cid:14)(cid:17)(cid:17)(cid:18) (cid:14)(cid:19)(cid:18)(cid:18) (cid:20) (cid:5) (cid:10) (cid:21)(cid:5) (cid:6) (cid:8) (cid:10) (cid:22) (cid:1) (cid:11) (cid:23) (cid:24) (cid:25) (cid:26)(cid:1) (cid:27) (cid:5) (cid:8) (cid:10) (cid:6) (cid:15) (cid:28)(cid:3)(cid:29)(cid:21)(cid:1)(cid:30)(cid:21)(cid:5)(cid:31)(cid:10)(cid:7)(cid:1)(cid:11)(cid:5)(cid:4)(cid:15)(cid:16)(cid:18)(cid:18) (cid:17)(cid:18)(cid:18) (cid:19)(cid:18)(cid:18) (cid:18)(cid:18) !(cid:18)(cid:18) " (cid:10) (cid:8) (cid:12) (cid:3) (cid:30) (cid:1) $ %(cid:5) (cid:10) (cid:24) (cid:3) (cid:6) (cid:10)(cid:1) (cid:11) & (cid:15) (cid:18)(cid:14)(cid:18)’(cid:18)((cid:18)(cid:16)(cid:18)(cid:17)(cid:18) (cid:28)(cid:3)(cid:29)(cid:21)(cid:1)(cid:30)(cid:21)(cid:5)(cid:31)(cid:10)(cid:7)(cid:1)(cid:11)(cid:5)(cid:4)(cid:15)(cid:16)(cid:18)(cid:18) (cid:17)(cid:18)(cid:18) (cid:19)(cid:18)(cid:18) (cid:18)(cid:18) !(cid:18)(cid:18) " (cid:10) (cid:8) (cid:12) (cid:3) (cid:30) (cid:1) $ %(cid:5) (cid:10) (cid:24) (cid:3) (cid:6) (cid:10)(cid:1) (cid:11) & (cid:15) (cid:18)(cid:14)(cid:18)’(cid:18)((cid:18)(cid:16)(cid:18) (cid:14))*+,(cid:13)(cid:10)(cid:11)(cid:14)-(cid:14)(cid:15))*+,(cid:13)(cid:10)(cid:11)(cid:14)-(cid:14)-(cid:14)(cid:15))*./(cid:25)(cid:6)(cid:26) (cid:11)(cid:3)(cid:15) (cid:2)(cid:3)(cid:4)(cid:3)(cid:5)(cid:1)(cid:6)(cid:7)(cid:8)(cid:9)(cid:10)(cid:1)(cid:11)(cid:12)(cid:4) (cid:13)(cid:14) (cid:15)(cid:14)(cid:16)(cid:18)(cid:18) (cid:14)(cid:16)(cid:17)(cid:18) (cid:14)(cid:17)(cid:18)(cid:18) (cid:14)(cid:17)(cid:17)(cid:18) (cid:14)(cid:19)(cid:18)(cid:18) (cid:20) (cid:5) (cid:10) (cid:21)(cid:5) (cid:6) (cid:8) (cid:10) (cid:22) (cid:1) (cid:11) (cid:23) (cid:24) (cid:25) (cid:26)(cid:1) (cid:27) (cid:5) (cid:8) (cid:10) (cid:6) (cid:15) (cid:2) +,(cid:13)(cid:10)(cid:11)(cid:14)-(cid:14)-(cid:14)(cid:15))*(cid:10)’)*(cid:10)()*(cid:10)(cid:16))*(cid:10)(cid:17))*(cid:10)(cid:19))* * (cid:2) (cid:14) +,(cid:13)(cid:10)(cid:11)(cid:14)-(cid:14)(cid:15))* (cid:2)0 (cid:11)(cid:25)(cid:15) $12(cid:13)(cid:31)(cid:24)%3(cid:5) % (cid:14)’(cid:26)(cid:16) % (cid:14)(cid:14)(cid:26)(cid:17) % (cid:2) (cid:14) (cid:2) (cid:14) (cid:2) ’ (cid:23)5(cid:13)(cid:16))*(cid:23)5(cid:13)()*(cid:23)5(cid:13)’)*(cid:10)’)*(cid:10)()*(cid:10)(cid:16))* % * (cid:23)5(cid:13)+)*(cid:10)’)*(cid:10)()*(cid:10)(cid:16))* $12(cid:13)(cid:31)(cid:24)%3(cid:5) (cid:23)5(cid:13)()*+,(cid:13)(cid:10)(cid:11)(cid:14)-(cid:14)-(cid:14)(cid:15))*(cid:23)5(cid:13)’)*+,(cid:13)(cid:10)(cid:11)(cid:14)-(cid:14)(cid:15))* (cid:11)(cid:12)(cid:15) (cid:11)6(cid:15) (cid:17)(cid:1) (cid:2) (cid:4) Figure 2:
Characterization of CVD-grown t N LG. ( a ) Optical contrast of ME-t(1 + + + + + + + + + + + + b ) Raman spectra of ME-t(1 + + + + ′ modes areobserved in ME-t(1 + + d ) Raman spectra of CVD-grown t N LG( N = ′ mode are observed in all the CVD-grown t N LG. the peak in the LBM region is identified as LB . Al-though the total layer number of ME-t(1 + + + is the observed at the low frequency zone in t(1 + + + + ff ect the interlayer breath- ing coupling, wherefore the LBMs are contributed fromall the graphene layers.[35] Therefore, it is reasonable thatonly the LB can be observed in the ULF region of ME-t(1 + + andC are observed in the t(1 + and LB , can be detected at the same time. Over-all, in the t(1 + n )LG, the C modes of n LG and the LBMsof (1 + n )LG are observed, confirming the identification byoptical contrast and the R and R ′ modes. Therefore, com-bining with high-frequency R / R ′ modes and ULF C / LB5 (cid:3)(cid:4)(cid:5) (cid:2)(cid:6)(cid:5) (cid:7) (cid:8) (cid:9) (cid:10)(cid:8) (cid:11) (cid:12) (cid:9) (cid:13) (cid:1) (cid:14) (cid:15) (cid:16) (cid:17) (cid:18)(cid:1) (cid:19) (cid:8) (cid:12) (cid:9) (cid:11) (cid:20) (cid:2)(cid:21)(cid:5) (cid:2)(cid:4)(cid:5) (cid:5) (cid:4)(cid:5) (cid:21)(cid:5) (cid:6)(cid:5) (cid:3)(cid:4)(cid:5) (cid:3)(cid:21)(cid:21)(cid:5) (cid:3)(cid:21)(cid:22)(cid:5) (cid:23)(cid:24)(cid:25)(cid:24)(cid:8)(cid:1)(cid:11)(cid:26)(cid:12)(cid:27)(cid:9)(cid:1)(cid:14)(cid:28)(cid:25)(cid:2)(cid:3)(cid:20) (cid:29)(cid:3)(cid:29)(cid:21)(cid:29)(cid:4)(cid:29)(cid:4)(cid:29)(cid:3)(cid:30)(cid:29)(cid:3)(cid:30)(cid:29)(cid:3)(cid:5)(cid:29)(cid:3)(cid:5) (cid:31)(cid:31) ! ! (cid:9)(cid:14)(cid:3)"(cid:4)(cid:20) (cid:29)(cid:3)(cid:29)(cid:3)(cid:29)(cid:3)(cid:29)(cid:3)(cid:29)(cid:3)(cid:30)(cid:29)(cid:3)(cid:30)(cid:29)(cid:3)(cid:5)(cid:29)(cid:3)(cid:5) (cid:23) (cid:29)%(cid:29)%(cid:29)%(cid:29)%(cid:29)(cid:30) (cid:29)(cid:30) (cid:29)(cid:3) (cid:29)(cid:3)(cid:29)(cid:4) (cid:2)(cid:3)(cid:4)(cid:5)(cid:6)(cid:7)(cid:8)(cid:7)(cid:8)(cid:7)(cid:9)(cid:10)(cid:11) (cid:23) (cid:3) (cid:23) (cid:4) (cid:23) (cid:3) &(cid:30)$$(cid:8)(cid:25) (cid:21)(cid:6)(cid:6)(cid:8)(cid:25)(cid:21)(cid:6)(cid:6)(cid:8)(cid:25)(cid:21)(cid:6)(cid:6)(cid:8)(cid:25)(cid:21)(cid:6)(cid:6)(cid:8)(cid:25)(cid:14)(cid:3)(cid:30)(cid:18)(cid:21) ’ (cid:20) ! $(cid:3) ! $(cid:3) ! (cid:21)(cid:3) ! (cid:21)(cid:4) ! %(cid:3) ! %(cid:4) ! (cid:30)(cid:3) ! (cid:30)(cid:4) ! (cid:30)$ (cid:31) %(cid:3) (cid:31) %(cid:4) (cid:31) %$ (cid:31) %(cid:21) (cid:31) (cid:21)(cid:3) (cid:31) (cid:21)(cid:4) (cid:31) (cid:21)$ (cid:31) $(cid:3) (cid:31) $(cid:4) (cid:31) (cid:4)(cid:3) Figure 3:
Raman spectroscopy of CVD-grown t(1 + n )LG. Stokes / anti-Stokes Raman spectra in the C and LB mode region and Stokes spectra inthe G spectral region for ME-t(1 + + + n )LG. The top one is the Raman spectra of t(1 + + + n )LG excited by 488 nm. The spectra are scaled and o ff set for clarity. The scaling factors of the individual spectraare shown. There are not any C modes observed in t(1 + + + n . The C modesof each n LG are observed, while the LBMs of (1 + n )LG are observed. Vertical dashed lines are a guide to the eye. (cid:2)(cid:3)(cid:4)(cid:5) (cid:2)(cid:6)(cid:5) (cid:7) (cid:8) (cid:9) (cid:10)(cid:8) (cid:11) (cid:12) (cid:9) (cid:13) (cid:1) (cid:14) (cid:15) (cid:16) (cid:17) (cid:18)(cid:1) (cid:19) (cid:8) (cid:12) (cid:9) (cid:11) (cid:20) (cid:2)(cid:21)(cid:5) (cid:2)(cid:4)(cid:5) (cid:5) (cid:4)(cid:5) (cid:21)(cid:5) (cid:6)(cid:5) (cid:3)(cid:4)(cid:5) (cid:22)(cid:23)(cid:24)(cid:23)(cid:8)(cid:1)(cid:11)(cid:25)(cid:12)(cid:26)(cid:9)(cid:1)(cid:14)(cid:27)(cid:24)(cid:2)(cid:3)(cid:20) (cid:28)(cid:4)(cid:28)(cid:3)(cid:28)(cid:3)(cid:28)(cid:6)(cid:28)(cid:3)(cid:5)(cid:28)(cid:4)(cid:28)(cid:4)(cid:28)(cid:4)(cid:5) (cid:29)(cid:29)(cid:30)(cid:31) (cid:30)(cid:31) (cid:3)(cid:21) (cid:5) (cid:3)!(cid:21)(cid:5) (cid:3)"(cid:3)(cid:5) (cid:28)(cid:3)(cid:28)(cid:3)(cid:28)(cid:3)(cid:28)(cid:3)(cid:28)(cid:3)(cid:5)(cid:28)(cid:4)(cid:28)(cid:4)(cid:28)(cid:4)(cid:5) (cid:22) (cid:29) $(cid:4) (cid:29) $(cid:3) (cid:30)(cid:31) (cid:21)(cid:4) (cid:30)(cid:31) (cid:21)(cid:3) (cid:28)(cid:4)(cid:5) (cid:28)(cid:4)(cid:5)(cid:28)! (cid:28)(cid:3)(cid:21)""(cid:8)(cid:24)(cid:21)(cid:6)(cid:6)(cid:8)(cid:24)!"(cid:6)(cid:8)(cid:24)"$$(cid:8)(cid:24) (cid:6)!(cid:8)(cid:24) (cid:6)(cid:18)(cid:3)%(cid:3)(cid:3)(cid:18)!%(cid:3)(cid:4)(cid:18)(cid:21)%(cid:3)"(cid:18)(cid:21)%(cid:3) (cid:18)(cid:4)%(cid:28)$(cid:28)(cid:6) (cid:28)(cid:3)(cid:28)(cid:6)!$(cid:5)(cid:8)(cid:24) (cid:3)(cid:21)(cid:18)(cid:21)% (cid:9)(cid:14)(cid:3)&$(cid:20)(cid:30) (cid:9)(cid:14)(cid:3)&(cid:8)(cid:20)(cid:30) (cid:14)(cid:23)(cid:20) (cid:2)(cid:2)(cid:2) *+(cid:17)(cid:11)(cid:9)(cid:16)(cid:23)(cid:9)(cid:10) (cid:14)(cid:17)(cid:20) Figure 4:
Schematic diagram of CVD-grown t(1 + n )LG and Raman spectra of CVD-grown t(1 + ff erent twist angles. ( a )Schematic diagram of t(1 + n )LG. ( b ) Stokes / anti-Stokes Raman spectra in the C and LB peak region and Stokes spectra in the R and G peakregions for CVD-grown t(1 + θ t of each t(1 + λ ) corresponding to each θ t is marked. The spectra are scaled and o ff set for clarity. Vertical dashed lines are a guide to the eye. ff use from the nuclear of the first layer to the sub-strate to form the second layer, and so on to form a MLG.The twist stacking is more likely to form between the toptwo layers, due to the influence of the copper foil. How-ever, the relative orientation between the layers below thetop one is more likely to be AB-stacked. The schematicdiagram of CVD-grown t(1 + n )LG is demonstrated in Fig.4(a). It is noteworthy that the LBMs are more di ffi cult tobe observed than the C modes in the AB-MLG[30, 39],and only the C N were observed in the Raman measure-ment at room temperature, due to the symmetry and weakelectron-phonon coupling.[30, 39] Here, LBM and morethan one C mode are detected in CVD-grown t(1 + n )LGflake with high signal to noise ratio, because the reso-nance from the VHS in JDOS enhances the Raman in-tensity and the twisting between two constituents leadsto lower symmetry, which makes more C and LB modesRaman active.[23, 35] In the graphite and ME-MLG, theAB-stacked MLG is the most stable stacking order and theABC stacked ones account for only about 15%.[15, 40]The C and LB modes of CVD-growth t(1 + n)LG are alsoin analog to those in ME-t(1 + n)LG, where nLG is withAB stacking. Furthermore, the transition temperatureof ABC-AB modification in bulk graphite is higher than1000 o C, however, the ABC-AB transition in ME-MLGhas been observed within two months even at room tem-perature, and thus it is expected that the transition temper-ature of ABC-stacked MLG will be much lower than thatin bulk case. Additionally, the growth temperature of theCVD-MLG is as high as 1020 o C. Therefore, it’s reason-able to conclude that the constituent nLG in the observedt(1 + n)LGs is with AB stacking.To comprehensively figure out the stacking configu-ration of t(1 + n )LG grown by the present CVD method,more tMLG flakes with di ff erent θ t are detected with thecorresponding resonant excitation energies, as shown inFig. 4(b). The t(1 + θ t of these six t(1 + ◦ to 17.2 ◦ , which can be determinedby the position of the R peak marked by the gray dashline. For each t(1 + and C modes are ob- served at ∼
36 cm − and ∼
21 cm − , respectively, which areclose to the corresponding peak position of ME-t(1 + ∼
37 cm − and ∼
22 cm − , respectively)[23]. Moreover,for all t(1 + and LB are measured in ∼
116 cm − and ∼
93 cm − , which are almost the same asthose of ME-t(1 + + n )LG is the universal stacking configurationof our CVD-MLGs under the present growth condition.The C and LB modes in all the t(1 + + / SiO substrate.The C and LB modes are the direct signature of theinterlayer coupling. The strength of the interlayer cou-pling can be determined from the position of the C andLB modes. To understand the interlayer coupling quan-titatively, the frequency of C and LB modes in CVD-MLG flakes is compared with that of the correspondingME-MLG. As shown in Figs. 5(a)(b), the optical im-ages of two CVD-MLG flakes are shown. The monocrys-talline zone of each flake is marked as ZA and ZB, re-spectively. The Raman spectra of ME-t(1 + + of those t(1 + ∼
109 cm − . However, C mode of theME-t(1 + ∼ − , which is higher than thatof the two CVD-t(1 + ∼ − and 28 cm − from ZA and ZB, respectively.For the 2D layered materials, according to the LCM,the frequencies of interlayer vibration modes are propor-tional to the square root of the interlayer force constant,which is usually assumed as α and β . In the ME-tMLGsystem, a softened factor with respect to the bulk caseis introduced at the twisted interface for shear coupling, α k t /α k ∼ α k t /α k ∼ α ⊥ ) at the twisted interface is as strong as that at theAB-stacked interface. However, to well predict the ex-7 (cid:3)(cid:4)(cid:5) (cid:2)(cid:6)(cid:5) (cid:7) (cid:8) (cid:9) (cid:10)(cid:8) (cid:11) (cid:12) (cid:9) (cid:13) (cid:1) (cid:14) (cid:15) (cid:16) (cid:17) (cid:18)(cid:1) (cid:19) (cid:8) (cid:12) (cid:9) (cid:11) (cid:20) (cid:2)(cid:21)(cid:5) (cid:2)(cid:4)(cid:5) (cid:5) (cid:4)(cid:5) (cid:21)(cid:5) (cid:6)(cid:5) (cid:3)(cid:4)(cid:5)(cid:22)(cid:23)(cid:24)(cid:23)(cid:8)(cid:1)(cid:11)(cid:25)(cid:12)(cid:26)(cid:9)(cid:1)(cid:14)(cid:27)(cid:24) (cid:2)(cid:3) (cid:20)(cid:28)(cid:3)(cid:28)(cid:3)(cid:29) (cid:30)(cid:30)(cid:31) (cid:31) (cid:30)!"(cid:2)(cid:9)(cid:14)(cid:3) + (cid:16) (cid:10),((cid:10)(cid:8) (cid:27)(cid:13) (cid:1) (cid:14) (cid:27) (cid:24) (cid:2) (cid:3) (cid:20) -(cid:5)(cid:3)(cid:5)(cid:5)(cid:3)(cid:3)(cid:5)(cid:3)(cid:4)(cid:5)(cid:31)(cid:23)(cid:13)(cid:10)(cid:16)(cid:1)(cid:8)((cid:24)(cid:17)(cid:10)(cid:16)(cid:1)(cid:14))(cid:20)*(cid:18)(cid:5) *(cid:18)% (cid:21)(cid:18)(cid:5) (cid:21)(cid:18)% %(cid:18)(cid:5) %(cid:18)% (cid:29)(cid:18)(cid:5) + (cid:16) (cid:10),((cid:10)(cid:8) (cid:27)(cid:13) (cid:1) (cid:14) (cid:27) (cid:24) (cid:2) (cid:3) (cid:20) (cid:3)(cid:5)(cid:4)(cid:5)*(cid:5)(cid:21)(cid:5) (cid:30)!"(cid:2)(cid:9)&(cid:31)$&’(cid:2)(cid:9)&(cid:31)$ (cid:14)(cid:23)(cid:20) (cid:14)(cid:17)(cid:20) (cid:14)(cid:27)(cid:20)(cid:14).(cid:20) (cid:14)(cid:10)(cid:20) (cid:14)(cid:26)(cid:20) (cid:14)/(cid:20) (cid:30)(cid:9)(cid:31)(cid:30)& (cid:31) (cid:4)(cid:31)(cid:30)&0(cid:15) 0(cid:15)0 (cid:30)!"(cid:2)(cid:9)&(cid:31)$&’(cid:2)(cid:9)&(cid:31)$(cid:30)(cid:9)(cid:31)(cid:30)& (cid:31) (cid:4)(cid:31)(cid:30)& (cid:2)(cid:3) (cid:9) (cid:2)(cid:3) (cid:5)(cid:9) (cid:2)(cid:3) (cid:5) (cid:2)(cid:3) (cid:5) (cid:2)(cid:3) (cid:5) (cid:2)(cid:3) (cid:5) (cid:2)(cid:4) (cid:5) (cid:2)(cid:4) (cid:5) (cid:30)!"(cid:2)$(cid:16)12(cid:8) (cid:30)!"(cid:2)$(cid:16)12(cid:8) (cid:30) (cid:4)(cid:3) (cid:31) *(cid:3) Figure 5:
Interlayer coupling of ME-t(1 + n )LG and CVD-grown t(1 + n )LG. ( a ) Optical image of one CVD-MLG flake. One of the monocrys-talline zones is marked as ZA. ( b ) Optical image of another CVD-MLG flake. One of the monocrystalline zones is marked as ZB. ( c ) Stokes / anti-Stokes Raman spectra in the C and LB peak region and Stokes spectra in the G peak region for ME-t(1 + + d ) The schematic diagram of tLCM for calculating the C frequencies intMLGs. ( e ) The schematic diagram of 2LCM for calculating the LBM frequencies in tMLGs. ( f ) Summary of experimental frequencies of the Cmodes (green triangle for ZA and blue circle for ZB). The calculated frequencies of the C modes are plotted by lines too (dash line for mechanicalexfoliation, dark red solid line for ZA and pink solid line for ZB). ( g ) The experimental and calculated LB peak frequencies for each sample. perimental frequencies, the second-nearest layer breath-ing coupling ( β ⊥ ) must be considered in the LCM, given2LCM.[35] tLCM and 2LCM, as shown in Figs. 5(d)(e),predict the frequencies of C and LB modes very well inthe ME-tMLG respectively, whose results are plotted inFigs. 5(f)(g) by dash black lines. tLCM and 2LCM arealso adopted to describe the interlayer vibration modesof CVD-grown samples. The C and LB modes frequen-cies of each CVD-tMLG are summarized in Figs. 5(f)(g).Compared with the frequencies in ME-tMLG, there arethe softening of the experimental frequency for all thewhole-family modes in CVD-tMLG. Therefore, a unifiedsoftened factor γ for each interlayer coupling in tLCMand 2LCM is assumed, as shown in Figs. 4(d)(e). For theC modes, γ is assumed as 95% and 90% for ZA and ZB,respectively, to fit the experimental results, which meansthat the interlayer shear coupling is softened slightly,which demonstrates the high quality of these CVD-grown tMLGs. However,For the LBMs, γ is 99.5% and 99% forZA and ZB, respectively, which indicates that the inter-layer breathing coupling almost keep constant as the bulkcase comparing with the ME-tMLGs, as shown in Fig.5(g).In general, there are two main reasons for the softeningof intrinsic interlayer coupling in CVD-grown tMLGs:the change of interlayer space changing resulting frommolecules intercalation and wrinkles, and the change ofstacking way changing resulting from twisting. The largerinterlayer space would significantly soften the frequencyof both C and LB modes. In the present case, the LBMfrequency almost keep constant as the bulk case, inde-pendent on the twist angle, which means the interlayerspace almost keep constant as the bulk case, independenton the twist angle. Moreover, twisting between two ad-jacent layers would just a ff ect the shear coupling at thetwisted interface. However, the softening happens for8ll C modes, which means the twisting isn’t the reasonof lower C modes in CVD-grown tMLGs. The CVD-grown tMLGs are with a bit of defects, like atom miss-ing or dislocation.[2, 41] These defects would a ff ect thealignment of carbon atoms of neighboring layer at thedefect region. The shear coupling are much more sensi-tive to the alignment of carbon atoms than layer breathingcoupling.[35, 42] Therefore, the C and LB modes also canbe used to evaluate the quality of CVD-grown tMLGs.
4. Conclusion
In conclusion, we prepared the MLGs by CVD method.The contrast mapping of Rayleigh scattering is employedto identify the layer number. The CVD-MLG flakes canbe distinguished with di ff erent zones by the Raman map-ping of the enhanced intensity of the G mode excited bya specific laser energy. Only one R mode is observed ineach zone containing di ff erent MLGs, which suggests thatthe MLG flakes are twisted ones as t(1 + n )LGs. For theCVD-grown MLGs with a definite layer number, they ex-hibit similar spectral features, including the number andmode frequency of the C and LB modes, independent ofthe twist angle, confirming the universality of the stackingconfiguration in the CVD-grown tMLGs under the presentgrowth condition. The frequencies of the C and LB modesin the CVD-grown MLGs are almost identical to those inthe corresponding ME-MLGs with the same total layernumber and sub-constituents, indicating the high crystalquality of the CVD-grown MLGs.
5. Acknowledgments
We acknowledge support from the National Basic Re-search Program of China Grant Nos. 2016YFA0301200and 2014CB932500, and the National Natural Sci-ence Foundation of China, grants 11225421, 11434010,11474277, 11504077 and 21525310.
References [1] Kim KS, Zhao Y, Jang H, Lee SY, Kim JM, KimKS, et al. Large-scale pattern growth of graphenefilms for stretchable transparent electrodes. Nature2009;457(7230):706–10. [2] Li X, Cai W, An J, Kim S, Nah J, Yang D,et al. Large-area synthesis of high-quality anduniform graphene films on copper foils. Science2009;324(5932):1312–4.[3] Gomez De Arco L, Zhang Y, Schlenker CW, Ryu K,Thompson ME, Zhou C. Continuous, highly flexi-ble, and transparent graphene films by chemical va-por deposition for organic photovoltaics. ACS Nano2010;4(5):2865–73.[4] Bae S, Kim H, Lee Y, Xu X, Park JS, Zheng Y,et al. Roll-to-roll production of 30-inch graphenefilms for transparent electrodes. Nat Nanotech2010;5(8):574–8.[5] Xue J, Sanchez-Yamagishi J, Bulmash D, JacquodP, Deshpande A, Watanabe K, et al. Scanning tun-nelling microscopy and spectroscopy of ultra-flatgraphene on hexagonal boron nitride. Nat Mater2011;10(4):282–5.[6] Wang Y, Tong SW, Xu XF, ¨Ozyilmaz B, LohKP. Interface engineering of layer-by-layer stackedgraphene anodes for high-performance organic solarcells. Adv Mater 2011;23(13):1514–8.[7] Wang G, Kim Y, Choe M, Kim TW, Lee T. Anew approach for molecular electronic junctionswith a multilayer graphene electrode. Adv, Mater,2011;23(6):755–60.[8] Han TH, Lee Y, Choi MR, Woo SH, Bae SH, HongBH, et al. Extremely e ffi cient flexible organic light-emitting diodes with modified graphene anode. NatPhoton 2012;6(2):105–10.[9] Liu Y, Yuan L, Yang M, Zheng Y, Li L, Gao L,et al. Giant enhancement in vertical conductivityof stacked cvd graphene sheets by self-assembledmolecular layers. Nat Commun 2014;5:5461.[10] Lee H, Qi Y, Kwon S, Salmeron M, Young PJ. Largechanges of graphene conductance as a function oflattice orientation between stacked layers. Nan-otechnology 2014;26(1):015702.[11] Lee S, Lee K, Zhong Z. Wafer scale homogeneousbilayer graphene films by chemical vapor deposi-tion. Nano Lett 2010;10(11):4702–7.912] Yan Z, Peng Z, Sun Z, Yao J, Zhu Y, Liu Z, et al.Growth of bilayer graphene on insulating substrates.ACS Nano 2011;5(10):8187–92.[13] Nie S, Walter AL, Bartelt NC, Starodub E, Bost-wick A, Rotenberg E, et al. Growth from be-low: Graphene bilayers on ir (111). ACS Nano2011;5(3):2298–306.[14] Yan K, Peng H, Zhou Y, Li H, Liu Z. Forma-tion of bilayer bernal graphene: Layer-by-layer epi-taxy via chemical vapor deposition. Nano Lett2011;11(3):1106–10.[15] Lui CH, Li Z, Chen Z, Klimov PV, Brus LE, HeinzTF. Imaging stacking order in few-layer graphene.Nano Lett 2010;11(1):164–9.[16] Dos Santos JL, Peres N, Neto AC. Graphene bilayerwith a twist: Electronic structure. Phys Rev Lett2007;99(25):256802.[17] Trambly de Laissardi`ere G, Mayou D, Magaud L.Localization of Dirac electrons in rotated graphenebilayers. Nano Lett 2010;10(3):804–8.[18] Havener RW, Zhuang H, Brown L, Hennig RG, ParkJ. Angle-resolved Raman imaging of interlayer ro-tations and interactions in twisted bilayer graphene.Nano Lett 2012;12(6):3162–7.[19] Moon P, Koshino M. Optical absorption in twistedbilayer graphene. Phys Rev B 2013;87:205404.[20] Yeh CH, Lin YC, Chen YC, Lu CC, Liu Z, SuenagaK, et al. Gating electron-hole asymmetry in twistedbilayer graphene. ACS Nano 2014;8(7):6962–9.[21] Nguyen VH, Dollfus P. Strain-induced modulationof dirac cones and van hove singularities in a twistedgraphene bilayer. 2D Materials 2015;2(3):035005.[22] Nguyen VH, Nguyen HV, Saint-Martin J, DollfusP. Strain-induced conduction gap in vertical devicesmade of misoriented graphene layers. Nanotechnol-ogy 2015;26(11):115201.[23] Wu JB, Zhang X, Ij¨as M, Han WP, Qiao XF, Li XL,et al. Resonant raman spectroscopy of twisted mul-tilayer graphene. Nat Commun 2014;5:5309. [24] Ferrari AC, Basko DM. Raman spectroscopyas a versatile tool for studying the properties ofgraphene. Nat Nanotech 2013;8(4):235–46.[25] Ferrari AC. Raman spectroscopy of graphene andgraphite: disorder, electron-phonon coupling, dop-ing and nonadiabatic e ff ects. Solid State Commun2007;143(1):47–57.[26] Carozo V, Almeida CM, Ferreira EH, Canc¸ado LG,Achete CA, Jorio A. Raman signature of graphenesuperlattices. Nano Lett 2011;11(11):4527–34.[27] Campos-Delgado J, Canc¸ado LG, Achete CA, JorioA, Raskin JP. Raman scattering study of the phonondispersion in twisted bilayer graphene. Nano Re-search 2013;6(4):269–74.[28] Jorio A, Canc¸ado LG. Raman spectroscopy oftwisted bilayer graphene. Solid State Communica-tions 2013;175:3–12.[29] Ferrari AC, Meyer J, Scardaci V, Casiraghi C,Lazzeri M, Mauri F, et al. Raman spectrum ofgraphene and graphene layers. Phys Rev Lett2006;97(18):187401.[30] Tan PH, Han W, Zhao W, Wu Z, Chang K, Wang H,et al. The shear mode of multilayer graphene. NatMater 2012;11(4):294–300.[31] Latil S, Henrard L. Charge carriers in few-layergraphene films. Phys Rev Lett 2006;97(3):036803.[32] Thomsen C, Reich S. Double resonant raman scat-tering in graphite. Phys Rev Lett 2000;85(24):5214.[33] Kim K, Coh S, Tan LZ, Regan W, Yuk JM, Chat-terjee E, et al. Raman spectroscopy study of ro-tated double-layer graphene: misorientation-angledependence of electronic structure. Phys Rev Lett2012;108(24):246103.[34] Cong C, Yu T. Evolution of raman G and G ′ (2 D ) modes in folded graphene layers. Phys RevB 2014;89:235430.[35] Wu JB, Hu ZX, Zhang X, Han WP, Lu Y, ShiW, et al. Interface coupling in twisted multilayergraphene by resonant raman spectroscopy of layerbreathing modes. ACS Nano 2015;9(7):7440–9.1036] Zhang X, Han WP, Wu JB, Milana S, Lu Y, LiQQ, et al. Raman spectroscopy of shear and layerbreathing modes in multilayer MoS . Phys Rev B2013;87:115413.[37] Casiraghi C, Hartschuh A, Lidorikis E, Qian H,Harutyunyan H, Gokus T, et al. Rayleigh imag-ing of graphene and graphene layers. Nano Lett2007;7(9):2711–7.[38] Carozo V, Almeida CM, Fragneaud B, Bedˆe PM,Moutinho MVO, Ribeiro-Soares J, et al. Resonancee ffff