Study of Graphene Layers Using Atomic Force Microscope and Raman Spectroscopy

COMPARATIVE STUDY OF GRAPHENE FLAKEON SiO2 SUBSTRATE USING ATOMIC FORCE MICROSCOPE AND RAMAN SPECTROSCOPY

Felicia Dhivya Thiagaraj

A comprehensive study of graphene flakes was conducted using atomic force microscope (AFM) and Raman spectroscopy. The sample preparation was done by micromechanical cleaving of bulk graphite on ~285 nm SiO2 layer. The graphene surface was not found to be flat but contained folds, wrinkles and impurities which contributed to the intrinsic stability of the graphene flake. The respective Raman spectra related to the electronic band structure of graphene were obtained at different regions of the flake with D band (~1360cm-1) G band (~1576 cm-1), 2D band (~2724 cm-1) and 2G band (~3250 cm-1). The asymmetric nature of the 2D band indicated that the selected graphene flake was a few-layered. Combining Atomic force microscope and Raman spectroscopy detailed information regarding the morphology, thickness and electronic structure of graphene is obtained.

1. Introduction

Graphene, the building block of graphite has been of great scientific interest in recent years due to its unique electronic and structural characteristics. In essence, graphene is a flat single layer of sp2 hybridized carbon atoms packed into a two dimensional honeycomb lattice.For a long time it was assumed that 2D crystals were thermodynamically unstable and could not exist under ambient conditions. However,in 2004 Novoselov et al. successfully demonstrated that it was possible to produce graphene and other 2D atomic crystals by extracting them from 3D materials [1]. These extracted 2D crystals were found to exhibit long range crystalline order. Also, the microscopic roughening on the surface of suspended graphene sheets contributed to the stability of the crystals [2].The fact that the roughness is reproducible for different positions on the flake, becomes notably smaller for bi-layer graphene and disappears for thicker flakes prove that these corrugations are intrinsic to graphene membranes. Further, theoretical investigations of 2D crystallite graphene had predicted their thermodynamic stability through static microscopic crumpling. This discovery has led to investigations into the electronic and mechanical properties of graphene. Presently, there are two methods of producing graphene. In the first method, graphene layers are mechanically exfoliated from bulk graphite [3]. This method obviously cannot be employed for large-scale applications since only a single sheet of graphene is removed at a time from highly ordered pyrolytic graphite. Hence,the second method which involves the full graphitization of silicon carbide under vacuum conditions is employed[4].

The Atomic Force Microscope is a widely used tool used to characterize morphology and to measure the thickness of the graphene flakes on the silicon oxide substrate but it has low throughput. Due to the chemical contrast-caused by variation in the tip-sample interaction between two samples-between the graphene and substrate, which results in an apparent chemical thickness of 0.5-1 nm, much greater than that expected from the interlayer graphite spacing, in practice it is only possible to distinguish between one or two layers by AFM when graphene films contain folds and wrinkles [5]. Though the measurements obtained from AFM operating in the contact mode are more reliable, this mode could nevertheless damage the sample surface. Hence, the AFM is commonly operated in the non-contact or tapping mode.

Raman Spectroscopy is a very reliable, non-destructive and high throughput tool used to identifysingle-layer and multi-layer graphene [6].The Raman spectra for single- and few-layer graphene reflect changes in the electronic structure and electron-phonon interactions. Variations have been observed in the spectral regions of the graphite G (~1580 cm-1) and 2D (~2700 cm-1) modes as a function of the number of graphene layers. In addition, the disorder induced D mode (~1350cm-1) is related to the high energy optical phonons in the vicinity of the reciprocal space vector κ point in the graphite. It essentially gives an indication of the defect, damage and impurity levelin the graphene. The G mode is the first-order optical phonon mode and is related to the Raman active tangential E2g phonon, where the two atoms in the graphene unit cell are vibrating tangentially one against the other.The 2D mode is the second-order optical phonon mode near the κ point in the graphene Brillouin zone and is due to the double inelastic phonon scattering.For single-layer graphene the 2D peak can be fitted to a single Lorentzian, whereas in few-layer graphene it requires fitting to two or more Lorentzians.The 2D band changes from a narrow and symmetric feature in a single-layer graphene to one that shows an asymmetric shape on the high-energy side in a few-layered graphene [6]. Also, the increase in the G-band intensity going from single- to few-layer graphene further supports the claim that the electronic band structures are unique to the number of graphene layers.

In this paper, the morphology of the graphene and its number of layers have been investigated and also a comparative study of the graphene flake using AFM and Raman Spectroscopy is performed.

1. Materials and Experimental method

Graphene films for this study were prepared by micromechanical cleavage of highly oriented pyrolytic graphite as described by Novoselov et.al in 2004 [2]. Using a scotch tape, flakes of graphite were repeatedly peeled and then deposited on the surface of a silicon wafer covered by a 285 nm thick silicon dioxide film. Graphene layers thinner than 50 nm are transparent to light. However, they become visible on the silicon dioxide surface as even a monolayer adds to the optical path of reflected light to change the interference color with respect to the substrate. Typically, the color of the silicon dioxide wafer is violet-blue but the deposition of graphene layers shifts the color to blue. Graphene flakes were then identified under an optical microscope.

Veeco Dimension 3100 Atomic Force Microscope was used in the tapping mode to characterize and identify the selected graphene flake. The images were collected and analyzed using “section or profile” analysis to measure the height of the graphene flake relative to silicon dioxide surface.

Raman spectra for the selected graphene flakewere recorded using Renishaw Raman Microscopy version 5 for which the excitation source was the 514 nm Ar ion laser with incident power in the mW range to avoid heating of the sample. The laser spot size is sub-micron. Also, the Raman measurements were performed with a microscope set-up in the backscattering geometry at room temperature. The spectra measurements were collectedin the range 1000-4000 cm-1.

1. Results and Discussion

An AFM image of the graphene flake on silicon dioxide wafer selected for this experiment is shown in Fig. 1. Various features of the flake were investigated under the AFM. It was confirmed that the surface of the graphene was not flat but rather consists of folds or wrinkles (Fig. 1) [3]. This means that the stability of the graphene flake is due to the presence of this3D roughening on its surface. Therefore, in order to determine the true height at which the graphene flake is relative to the substrate, the heights at which the different features are fromthe SiO2 surface should be taken into account. Impurities (Fig.1) on the surface of the graphene film were also observed. These impurities also cause roughening of the graphene surface and their presence can be attributed to the preparation of graphene on the SiO2 surface by repeated peeling using a scotch tape. It can be clearly seen that different features of the flake are at different heights from the SiO2 surface. This could be due to the pleating of graphene or anomalies induced by the preparation of graphene. Though like other scanning probe techniques the AFM is not free of measurement artifacts, it still gives reasonably accurate information regarding the structure and thickness of the graphene layers. (Insert Figure. 1)

By using Raman spectroscopy with 514 nm excitation laser the same graphene flake features were analyzed (Fig. 2). As per literature the G band occurs at approximately 1576 cm-1 and the 2D band occurs at approximately 2724 cm-1. The asymmetrical nature of the 2D band for features A,B and C within the same graphene flake (see Fig.3) tells us that these regions belong to a few layered graphene [6].Also, the 2D peak for a few-layered graphene is divided into 2D1 and 2D2[8,7]. This splitting of the 2D band can be explained in the following way: an electron is excited from point A in the valence л band to point B in the conduction л* band by absorbing a photon. The excited electron is inelastically scattered to a point C by emission of a phonon. Inelastic backscattering to the vicinity of point A by emission of another phonon and electron-hole recombination leads to emission of photon of energy less than that of the incident photon [9]. Also, the presence of the disorder induced D peak at about 1360cm-1indicates that small defects are present in the flake. Another peak can be seen at about 3250 cm-1 which corresponds to the double-resonant disorder induced G band or the 2G band. Its presence is again related to the defects in the graphene.The variation in the intensities and the FWHM of the G peaks and 2D peaks within the same graphene flake is because the spectra were obtained at different regions of the flake where folds, wrinkles or impurities would have been present (see Fig.2). Therefore,there is a variation in the electronic band structure within the same graphene flake while considering different regions. One set back of using the Raman Spectroscopy is that sample edges are always seen as defects giving rise to disorder induced peaks [10].(Insert Figure. 2, Figure. 3)

The darker flake shaped features found on the SiO2substrate were initially assumed to be thick graphene layers under the AFM. While, in reality these were damage/impurities on the surface of the SiO2substrate, as confirmed by Raman spectroscopy

Though the AFM provides information regarding the surface morphology and thickness of graphene layers, yet it cannot indicate whether the chosen graphene layer is single-layered or a few-layered. In contrast, excellent information can be obtained regarding electronic structure and number of layers by using the Raman Spectroscopy. However, this technique cannot provide information regarding the morphology of the graphene flakes. Also, it is better to use the Raman Spectroscopy first if studies on only the nature of single- or a few-layered graphene flake has to be performed. In conclusion, a comprehensive study of the nature of the graphene flake can be performed by using a combination of the AFM and Raman Spectroscopy.

4. Conclusion

The different features of the graphene flake were studied using both the AFM and Raman spectroscopy. The surface of the flake was found to contain folds and impurities which provide intrinsic stability. The intensities of the G bands and the asymmetry of the 2D bands of the Raman spectra indicate that the graphene flake was few-layered.Also, the 2D band was split into 2D1 and 2D2 in a few-layered graphene due to double resonant phonon scattering. By combining the AFM and the Raman Spectroscopy one can estimate the morphology, number of layers and electronic band structures of graphene.

Acknowledgments

I would like to thank Dr. Christina Giuzca and Chris Buxey at the Advanced Technology Institute,University of Surrey, UK for their technical support while using AFM and Raman Spectroscopy. Also, I would like to thank Dr. Vlad Stolojan for making this experiment a valuable learning experience.

Figure 1

Figure 2

Figure 3

CAPTIONS

Fig.1 shows the AFM image of a few layered graphene. In this figure the dark brown color corresponds to the SiO2 surface, the light brown is the graphene flake and the white spots and pleating on its surface corresponds to the impurities and folds. The height of the pleated layer is 8.25nm relative to the graphene surface andthat of the folded edge of the graphene flake relative to substrate is 11.68nm. This implies that the graphene surface is not flat but rather consists of folds and impurities which are intrinsic to its structure and provides stability.

Fig. 2 shows the Raman Spectra using 514nm laser excitation corresponding to regions A (dashed line), B (solid line) and C (dotted line)from Fig. 1. The D peak appears at ~1360 cm and gives an indication of the defects in the graphene layer. The G peak appears at ~ 1576.4cm-1, 2D peak appear at ~2724cm-1 and2G peak at ~3250cm-1.The variation in the intensities and FWHM of the G bands and 2D bandsis due to the presence of impurities and folds in those regions.

Fig. 3shows 2D bands of the Raman Spectra corresponding to regions A (dashed line), B (solid line) and C (dotted line) from Fig.1. The 2D band peaks are finger prints of the graphene layers and occur at ~2724 cm-1. Its asymmetrical nature indicates that the graphene flake is few-layered. The 2D band is split into 2D1 and 2D2due to double resonant phonon scattering.

REFERENCES

1. K.S. Novoselov, D. Jiang, F. Schedin, T.J. Booth, V.V Khotkevich,S.V Morozov, PNAS 102:10451–53(2005).
2. J.C. Meyer, A. K. Geim, M. I. Katsnelson, K. S. Novoselov, T. J. Booth, S. Roth,Nature446, 60-63(2007).
3. K.S. Novoselov, A.K. Geim, S.V. Morozov, D. Jiang, Y. Zhang, S.V. Dubonos, I.V. Grigorieva, & A.A. Firsov, Science 306, 666-669 (2004).
4. C. Berger, Z. Song, X. Li,X. Wu,N. Brown, C. Naud, D. Mayou, T. Li, M. Sprinkle, J. Hass, A. N. Marchenkov, E.H. Conrad,W.A. de Heer, W. A. Science312, 1191–1196 (2006).
5. A. C. Ferrari,J. C. Meyer,V. Scardaci,C. Casiraghi,M. Lazzeri,F. Mauri,S. Piscanec,D. Jiang, K. S. Novoselov,S. Roth,A. K. Geim, Phys Rev Letter 97, 187401 (2006).
6. S. Elena, T.R. Kwang, R. Sunmin, M. Janina, K. Philip, E.B. Louis, F.H. Tony, S.H. Mark,W.F. George, PNAS 104, 9209-9212 (2007).
7. M. S. Dresselhaus, G. Dresselhaus, R. Saito, A. Jorio, Phys. Rep. 409, 47 (2005).
8. R. J. Nemanich and S. A. Solin, Phys. Rev. B 20, 392 (1979).
9. D. Graf, F. Molitor, K. Ensslin, C. Stampfer, A. Jungen, C.Hierold, L.Wirtz, Nano Letters, 7, 238-242 (2007).
10. Y. Duhee, M. Hyerim, C.Hyeonsik, Journal of the Korean Physical Society, 55, 1299-1303 (2009).