1: \documentclass[12pt,showpacs,aps,prb,amsmath,amssymb,superscriptaddress]{revtex4}
2: \usepackage{graphicx}
3: \usepackage{dcolumn}
4: \usepackage{bm}
5:
6: \begin{document}
7:
8: \baselineskip=0.8cm
9: \renewcommand{\thefigure}{\arabic{figure}}
10: \def\be{\begin{equation}}
11: \def\ee{\end{equation}}
12: \def\bea{\begin{eqnarray}}
13: \def\eea{\end{eqnarray}}
14: \def\E{{\rm e}}
15: \def\bearst{\begin{eqnarray*}}
16: \def\eearst{\end{eqnarray*}}
17: \def\peleven{\parbox{11cm}}
18: \def\peffec{\peight{\bearst\eearst}\hfill\peleven}
19: \def\pspace{\peight{\bearst\eearst}\hfill}
20: \def\ptwelve{\parbox{12cm}}
21: \def\peight{\parbox{8mm}}
22: %\twocolumn[\hsize\textwidth\columnwidth\hsize\csname@twocolumnfalse\endcsname
23:
24: \title{ Roughness of undoped graphene and its short-range induced gauge
25: field }
26: \author{N. Abedpour}
27: \affiliation{Department of Physics, Sharif University of Technology,
28: P.O. Box 11365-9161, Tehran 11365, Iran}
29: \author{M. Neek-Amal}
30: \affiliation{Department of Nano-Science, IPM, Tehran 19395-5531,
31: Iran}
32: \author{Reza Asgari}
33: \affiliation{Institute for Studies in Theoretical Physics and
34: Mathematics, Tehran 19395-5531, Iran}
35: \author{F. Shahbazi}
36: \affiliation{Department of Physics, Isfahan University of
37: Technology, Isfahan 84156-83111, Iran}
38: \author{ N. Nafari}
39: \affiliation{Institute for Studies in Theoretical Physics and
40: Mathematics, Tehran 19395-5531, Iran}
41: \author{M. Reza Rahimi Tabar $^{1,}$ }
42: \affiliation{ CNRS UMR 6529, Observatoire de la C$\hat o$te d'Azur,
43: BP 4229, 06304 Nice Cedex 4, France \\$^5$ Carl von Ossietzky
44: University, Institute of Physics, D-26111 Oldenburg, Germany }
45:
46: \begin{abstract}
47: We present both numerical and analytical study of graphene roughness
48: with a crystal structure including $500 \times 500$ atoms. The
49: roughness can effectively result in a random gauge field and has
50: important consequences for its electronic structure. Our results
51: show that its height fluctuations in small scales have scaling
52: behavior with a temperature dependent roughness exponent in the
53: interval of $ 0.6 < \chi < 0.7 $. The correlation function of
54: height fluctuations depends upon temperature with characteristic
55: length scale of $ \approx 90 {\AA}$ (at room temperature). We show
56: that the correlation function of the induced gauge field has a
57: short-range nature with correlation length of about $\simeq 2-3
58: {\AA}$. We also treat the problem analytically by using the
59: Martin-Siggia-Rose method. The renormalization group flows did not
60: yield any delocalized-localized transition arising from the graphene
61: roughness. Our results are in good agreement with recent
62: experimental observations.
63: \end{abstract}
64: \pacs{68.35.Ct ,89.75.Da, 71.15.Pd, 81.05.Uw} \maketitle
65: \hspace{.3in}
66: \section{Introduction}
67:
68:
69: An isolated two dimensional (2D) sheet of carbon atoms having a
70: simple honeycomb structure is known as graphene~\cite{wallace}.
71: According to the recent detailed measurements, these 2D zero-gap
72: semiconductors reveal new features in their electronic properties.
73: In particular, the low-energy quasi-particles of the system can
74: formally be described by the massless Dirac-like
75: fermions~\cite{novoselov}. Doped graphene sheets are pseudo-chiral
76: 2D Fermi liquids with abnormal effective electron-electron
77: interaction physics~\cite{yafis}. From the experimental point of
78: view, the melting temperature of thin films decreases with
79: decreasing thickness and become unstable when their thickness
80: reduces to a few atomic layers~\cite{evens}. This observation
81: supports the theoretical prediction that states: No strictly 2D
82: crystal can be thermodynamically stable at finite temperatures
83: ~\cite{mermin}. Consequently, the expectation of not observing a
84: free 2D material in nature lived on till the Geim's group discovered
85: graphene~\cite{mayer}. Furthermore, because of the existing coupling
86: between bending and stretching energy modes in any 2D material, one
87: expects to observe a measurable undulations or at least very small
88: roughness on the graphene sheet and will reduce electronic transport
89: in graphene and affecting its thermal conductivity~\cite{novoselov,
90: mayer}.
91:
92:
93: In this paper, we are interested in determining the temperature
94: dependence of graphene {\it roughness}. Here roughness is defined as
95: the variance of the height fluctuations in graphene due to
96: temperature. Our numerical treatments suggest that its height
97: fluctuations have a scaling behavior in small scale with a
98: temperature dependent roughness exponent. We also determine the
99: temperature dependence of the amplitudes of the height structure
100: function. We show that the induced random gauge field has a short-
101: range correlation function with correlation length $\simeq 2-3
102: {\AA}$. Finally, we formulate a field theoretical method to
103: investigate the electron dynamics of the undoped graphene in such a
104: random gauge field and show that there is no disorder-induced
105: transition from delocalized to localized states$^7$. This important
106: result is in good agreement with the experimental observation that
107: due to the graphene roughness localization is
108: suppressed~\cite{morozov}.
109:
110: The rest of this paper is organized as follows. In Sec.\,II, we
111: introduce an effective interaction between carbon atoms that enters
112: the molecular dynamic simulation to model the roughness of graphene
113: . Section III contains our numerical calculations of graphene
114: simulation and then using the latest results calculating the $\beta$
115: function analytically by using the field theory of
116: Martin-Siggia-Rose method. We finally conclude in Sec.\,IV with a
117: brief summary.
118:
119:
120: \section{Roughness Exponent}
121:
122: To study and arrive at a quantitative information of a graphene
123: surface one may consider a surface with size $L$ and define the mean
124: height of the surface, $\overline{h}(L,\lambda)$, and its roughness,
125: $w(L,t,\lambda)$, by the expressions like
126: $\overline{h}(L,\lambda)=\frac{1}{L}\int_{-\frac{L}{2}}^{\frac{L}{2}}h(\bf
127: {x},\lambda)dx $ and
128: $w(L,t,\lambda)=(<(h-\overline{h})^2>)^{(\frac{1}{2})}$,
129: respectively. The symbol $<...>$ denotes an ensemble averaging.
130: Here, $\lambda$ is an external factor which could be temperature in
131: this problem ~\cite{sang} and $t$ is the time. In the limit of large
132: $t$, the roughness saturates and behaves as $w(L,\lambda) \sim
133: L^{\chi (\lambda)}$. The roughness exponent, $\chi$, characterizes
134: the self-affine geometry of the surface. The common procedure for
135: measuring the roughness exponents of a rough surface is to use a
136: surface structure function, $S(r)=<|h(x+r)-h(x)|^2>$ which depends
137: on the length scale $\Delta x = r$. The surface structure function
138: is equivalent to the statistics of height-height correlation
139: function, $C(r)$, and are related by $S(r)=2w^2(1-C(r))$ for
140: stationary surfaces. The second order structure function $S(r)$
141: scales with $r$ as $r^{\xi_2}$ where $\chi=\xi_2/2$
142: ~\cite{bara,jafari}.
143:
144:
145: The atomic structure of graphene will force us to define two scaling
146: exponents in $x-$ (zig-zag) and $y-$ (arm-chairs) directions. The
147: exponents can be found via the second order structure functions, $
148: S_x(r_x)=<|h(x+r_x,y)-h(x,y)|^2> $ and $
149: S_y(r_y)=<|h(x,y+r_y)-h(x,y)|^2>$. The different scaling exponents
150: in the zigzag and arm-chairs directions show the anisotropic nature
151: of the roughness in graphene. We have used the empirical
152: inter-atomic interaction potential, i.e., carbon-carbon interaction
153: in graphite~\cite{brenner}, which has in addition three-body
154: interaction for molecular dynamics simulation of graphene sheet to
155: investigate its morphology and its dependence of roughness exponents
156: on temperature. The two-body potential gives a description of the
157: formation of a chemical bond between two atoms. Moreover, the
158: three-body potential favors structures in which the angle between
159: two bonds is made by the same atom. Many-body effects of electron
160: system, in average, is considered in the Brenner potential, through
161: the bond-order and furthermore, the potential depends on the local
162: environment.
163:
164: It is well known that the harmonic approximation resulting in
165: bending instabilities due to soft wavelength phonons leads to
166: crumpling of a membrane. It is important to note that the Brenner
167: potential has anharmonic coupling between bending and stretching
168: modes which prevents crumpling. We have considered a graphene sheet
169: including a size of $500 \times 500$ atoms with periodic boundary
170: condition. Considering the canonical ensemble (NVT), we have
171: employed Nos\'{e}-Hoover thermostat to control temperature. Our
172: simulation time step is $1~fs$ in all cases and the thermostat's
173: parameter is $5~fs$. Therefore, we have found a stable 2D graphene
174: sheet in our simulation.
175:
176: \section{Numerical results}
177:
178:
179: In the top graph of Fig.~1, we have shown a snapshot of the graphene
180: fluctuations at temperature $300 K$. Moreover, the magnetic field
181: induced by the roughness is shown in the bottom graph to emphasize
182: the randomness of the surface structure. Order of height
183: fluctuations are about $\sim 5 {\AA}$ which is in good agreement
184: with experimental observation\cite{mayer}. In Fig.~2, we have
185: plotted (in log-log scales), the structure function in the
186: arm-chairs direction, namely $S_y(r_y)$, simulated at temperatures
187: $10,30,100$ and $300$ $K$. Fig.~2 shows that for some characteristic
188: length scales, the scaling behavior of the structure functions do
189: not exist. The typical characteristic length scales are
190: approximately $90{\AA}$ at room temperature which is in good
191: agreement with experimental findings~\cite{mayer} that is in the
192: range of $50-100{\AA}$.
193:
194: In Fig.~3, the temperature dependence of the scaling exponents,
195: $\chi$`s, in both zigzag and arm-chairs directions are given. As
196: shown at low temperatures, the exponents for zigzag- and armchairs-
197: directions are about $0.7$ (see also Ref.~\cite{L}). However, the
198: exponent for armchairs direction is greater than the one for zigzag
199: direction at large enough temperatures. We have also used the bond
200: order potential proposed by Ghiringhelli {\it et al.}~\cite{los},
201: and found good agreement between the results of the two potentials
202: ~\cite{Faso}. This is physically understandable, since our
203: simulation has been performed for $T < 700K$ and the two potentials
204: mainly differ at higher temperatures. Fig.~3 (inset) shows also the
205: amplitude of the second moments, $C_x$ and $C_y$, which are defined
206: as, $S_x(r_x) = C_x r_x ^{{\xi_2}_x}$ and $S_y(r_y) = C_y r_y
207: ^{{\xi_2} _y}$ in the scaling region, in terms of temperature.
208:
209: To determine the characteristic length scales, we define the
210: quantity $Q(r)$ as the difference between the joint probability
211: distribution function (PDF) of height fluctuations at two points.
212: For instance, given $y$ and $y+r$ points, the $Q(r)$ is calculated
213: by $P(h_1,y;h_2,y+r)$ and product of two PDFs, $P(h_1,y) $ and
214: $P(h_2,y+r)$~\cite{CMB}. Thus, \be Q(r) = \int dh_1 dh_2 |
215: P(h_1,y;h_2,y+r) - P(h_1,y) P(h_2,y+r)|~. \ee
216: In Fig.~4 we have shown $Q(r)$ as a function of
217: $r$. This figure clearly indicates that the height fluctuations
218: at scales of $90{\AA}$ and $125{\AA}$ for $T=300K$ and $T=30K$,
219: respectively, are almost independent. Obviously, the $Q(r)$ becomes
220: $r-$ independent after these values of $r$ (by considering its error
221: bars). Furthermore, we have found the same value for correlation
222: length for graphene with fixed boundary condition. As a consequence,
223: this length scale is not an artifact of boundary conditions.
224:
225: \section{MARTIN-SIGGIA-ROSE EFFECTIVE ACTION}
226: Roughness of graphene results in a random gauge field and affects
227: its electronic structure~\cite{iordanskii}. The dependence of the
228: hopping integral $\Gamma$ on the deformation tensor is expressed by
229: $ \Gamma = \Gamma_0 + \frac{\partial \Gamma}{\partial u_{ij}}
230: u_{ij}$, where $u_{ij}$~\cite{nelson} is given by $ u_{ij} =
231: \frac{1}{2} \{ \frac{\partial u_i}{\partial x_j} + \frac{\partial
232: u_j}{\partial x_i} + \frac{\partial u_k}{\partial x_i}
233: \frac{\partial u_k}{\partial x_j} + \frac{\partial h}{\partial x_i}
234: \frac{\partial h}{\partial x_j} \}$. Here, $x_i\equiv(x,y)$ are
235: coordinates in the plane and $u_i$ are the corresponding components
236: of the displacement vector. In the presence of roughness, an
237: effective Dirac Hamiltonian describes the electron states near the
238: K-point, $ H= v_F \sigma ( -i \hbar \nabla - \frac{e}{c} \bf A )$,
239: where $v_F = \sqrt{3} \Gamma_0 a / 2 \hbar $ and $\bf A$ is the
240: gauge field. The gauge field can be written in terms of the hopping
241: integral $\Gamma$ as: $A_x= \frac{c}{2 e v_F} ( \Gamma_2 + \Gamma_3
242: - 2 \Gamma_1$ ) and $A_y = \frac{\sqrt{3} c}{2 e v_F} ( \Gamma_3 -
243: \Gamma_2)$~\cite{katsnelson}. Labels $1,2$ and $3$ refer to the
244: nearest neighbors atoms with vectors $( -a/\sqrt{3}, 0), (
245: a/2\sqrt{3}, -a/2)$ and $ (a/2\sqrt{3}, a/2)$, respectively. In
246: Fig.~5, we have plotted the structure function of the gauge field
247: $\bf A$, namely $S_A (r) = < | (\bf A(\bf x + \bf r) - \bf A ( \bf
248: x)) \cdot \hat x |^2 > $ versus the scale in $y$-direction. The
249: vectors $\bf r$ and $\hat {\bf x}$ can be chosen in $x-$ and $y-$
250: directions. Therefore, we have four different types of structure
251: functions. As shown in Fig.~5, the gauge field and the related
252: magnetic field have small scale correlation with correlation length
253: $l_c \simeq 2-3{\AA}$. We have checked the other three structure
254: functions of the induced gauge field and their cross correlation
255: functions and find that they have small scale correlation and are
256: almost statistically independent.
257:
258: In what follows we have developed a renormalization group analysis
259: to investigate the Dirac equation in random gauge field and show
260: that no delocalization-localization transition occurs for electrons
261: in such a random gauge potential$^{20,21}$.
262:
263:
264: The Lagrangian of Dirac fermions in (2+1) dimensions and in the
265: presence of gauge potential $A_{\mu}$ is given by: $
266: {\mathcal{L}}=i\int dt \int d^{2}{\bf x}
267: \bar{\psi}\gamma^{\mu}(\partial_{\mu}-A_{\mu})\psi $, in which
268: $\gamma^{0}=\sigma_{z},\gamma^{1}=i\sigma_{y},
269: \gamma^{2}=-i\sigma_{x}$, $\sigma$'s are the Pauli matrices and
270: $\gamma$'s satisfy the Clifford algebra
271: $\{\gamma^{\mu},\gamma^{\nu}\}=2g^{\mu \nu}$. The wave functions
272: $\psi(\bf x,t)$ and $\bar{\psi}(\bf x,t)$ are 2D Dirac spinors and
273: $A_{\mu}$ is a static random gauge field with a Gaussian
274: distribution having zero mean value. The covariance is given by $
275: \langle(A_{i}({\bf x})A_{j}({\bf
276: x}')\rangle=2D_{0}\delta_{ij}\delta({\bf x}-{\bf x}')$ where $ i,j
277: \equiv 1,2$ and $D_{0}$ is the intensity of its spatial
278: fluctuations. This relation shows the spatially uncorrelated nature
279: of the gauge filed. We note that the Fermi velocity of electrons in
280: graphene is of the order of $10^6$ m/s and the typical velocity of
281: the height fluctuations is of the order of $30$ m/s. Therefore, the
282: random gauge potential will act as a quenched random field on
283: electrons. Focusing on a single mode with energy $\epsilon$ we get
284: the following expression for the Lagrangian:
285:
286: \be
287: {\label {lagrangian2}} {\mathcal{L}}=\int d^{2}{\bf x}
288: \bar{\psi}(i\gamma^{k}\partial_{k}-i\gamma^{k}A_{k}+\epsilon\gamma^{0})\psi~.
289: \ee
290:
291: The expectation value of any operator $\mathcal{O}$ can be
292: calculated as follows
293:
294: \be \langle {\mathcal{O}}\rangle=\frac{\int
295: {\mathcal{D}}\psi{\mathcal{D}}{\bar\psi} {\mathcal{O}}
296: \exp(-i{\mathcal{L}})}{Z}~, \ee in which $Z$ the partition function
297: is defined by
298: \bea
299: Z&=&\int {\mathcal{D}}\psi{\mathcal{D}}{\bar\psi}
300: {\mathcal{O}} \exp(\int d^{2}{\bf x}
301: \bar{\psi}(-\gamma^{k}\partial_{k}+\gamma^{k}A_{k}-i\epsilon\gamma^{0})\psi
302: ) \nonumber \\
303: &=&
304: \det(-\gamma^{k}\partial_{k}+\gamma^{k}A_{k}-i\epsilon\gamma^{0})~.
305: \eea
306:
307: Introducing the Dirac bosons $\chi$ and ${\bar\chi}$, one can
308: re-expressing the above determinant as follows
309:
310: \be
311: Z^{-1}=\int {\mathcal{D}}\chi{\mathcal{D}}{\bar\chi}
312: \exp\left(-\int d^{2}{\bf x}
313: \bar{\chi}(-\gamma^{k}\partial_{k}+\gamma^{k}A_{k}-i\epsilon\gamma^{0})\chi\right),
314: \ee
315:
316: and implicitly we have
317:
318: \be \langle {\mathcal{O}}\rangle=\int
319: {\mathcal{D}}\psi{\mathcal{D}}{\bar\psi}
320: {\mathcal{D}}\chi{\mathcal{D}}{\bar\chi}{\mathcal{O}} \exp\{\int
321: d^{2}{\bf x}
322: \bar{\psi}(-\gamma^{k}\partial_{k}+\gamma^{k}A_{k}-i\epsilon\gamma^{0})\psi
323: -\int d^{2}{\bf x}
324: \bar{\chi}(-i\gamma^{k}\partial_{k}+\gamma^{k}A_{k}-i\epsilon\gamma^{0})\chi\}.
325: \ee
326:
327: Now by integrating the above result over the Gaussian variable
328: $A_{\mu}$ whose probability density function is given by
329:
330: \be P\propto \exp(-\frac{1}{4D_{0}}\int d{\bf x} A_{k}^{2}), \ee one
331: reaches the following result for the averaging of expectation values
332: over quenched random gauge
333: \be \langle\langle {\mathcal{O}}\rangle\rangle=\int
334: {\mathcal{D}}\psi{\mathcal{D}}{\bar\psi}
335: {\mathcal{D}}\chi{\mathcal{D}}{\bar\chi}
336: {\mathcal{O}}\exp(-S_{0}-S_{int})~, \ee where the free part of the
337: effective action is
338:
339: \be S_{0}=\int d^{2}{\bf x}
340: \bar{\psi}(\gamma^{k}\partial_{k}+i\epsilon\gamma^{0})\psi +\int
341: d^{2}{\bf x}
342: \bar{\chi}(\gamma^{k}\partial_{k}+i\epsilon\gamma^{0})\chi~, \ee and
343: the interaction part
344:
345: \be S_{int}=-D_{0}\int
346: d^{2}{\bf
347: x}\left[\left({\bar\psi}\gamma^{k}\psi+{\bar\chi}\gamma^{k}\chi)\right)\cdot
348: \left({\bar\psi}\gamma^{k}\psi+{\bar\chi}\gamma^{k}\chi\right)\right]~,
349: \ee where $k=1,2$. The $\beta$ function of the coupling $D_0$ will
350: determine its behavior under changing the scale.
351: We found that in one-loop order the correction to the short-range roughness
352: intensity ($D_0$) is proportional to $k^2$ , therefore this correction vanishes
353: in long-wave length limit. One can see that the same will happen at higher order
354: of of perturbation, leading us to the conclusion that the vanishing of the beta function in all orders should be
355: the consequence of the Ward identity due the conservation of
356: Dirac current which leads to an incompressible flow of electrons in the low energy limit( $\epsilon \sim 0$).
357:
358: This result shows that the
359: resistance against the electron flow due the interaction of Dirac
360: fermions and roughness of the Graphene, remains unchanged under
361: renormalization group flow towards large scales, and this in
362: turn, excludes the possibility of the localization of low energy
363: states (For more details, see for instance
364: Ref[\onlinecite{morpurgo,morozov,demartino,aleiner, ziegler}]).
365:
366: \section{Summary}
367:
368: In Conclusion, we find the temperature dependence of the roughness
369: exponents in different directions of a graphene sheet by simulating
370: the surface within molecular dynamics approach. We have used the
371: Brenner empirical inter-atomic interactions for graphite which is a
372: semiconductor. The point
373: we are making in this publication is the further detailed explanations
374: raised in Ref.[\onlinecite{Faso}] which
375: roughness could affect. We answer the question whether the roughness
376: could lead to localized electrons in graphene or not?. The
377: correlation function of height fluctuations shows that depending on
378: the temperature, there are characteristic length scales in the order
379: of $ \approx 90 {\AA}$ at room temperature. We show that the induced
380: gauge field has a short- range nature with correlation lengths
381: approximately $\simeq 2-3 {\AA}$. More importantly, roughness
382: essentially can affect to the electronic properties like
383: conductivity and modulation of the hopping integrals~\cite{castro}.
384: We treat the problem analytically by using the Martin-Siggia-Rose
385: method. The renormalization group flows do not yield any
386: delocalized-localized transition due to roughness. In the present
387: work, the effect of Dirac-like electrons on roughness are not
388: considered. It would be of intrest to develop our work may use the
389: quantum molecular dynamics simulation or ab initio Car-Parinnello
390: molecular dynamics for Dirac-like electrons to investigate the
391: dependence of Dirac-like electron on the graphene roughness.
392: \begin{acknowledgments}
393: We are grateful to A. Geim for illuminating discussions
394: and comments. M.R.R.T. would like to express his deep
395: gratitude to the Alexander von Humboldt Foundation and
396: Universitat Oldenburg for their financial support and providing
397: an excellent environment for research.
398: \end{acknowledgments}
399:
400: \begin{thebibliography}{99}
401: \bibitem{wallace} P. R. Wallace, Phys. Rev. {\bf 71}, 622 (1947)~.
402: \bibitem{novoselov}
403: K. S. Novoselov, D. Jiang, F. Schedin,T. J. Booth, V. V. Khotkevich,
404: S. V. Morozov, A. K. Geim, Proc. Natl. Acad. Sci.U.S.A. {\bf 102},
405: 10451 (2005), K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang,
406: M. I. Katsnelson, I. V. Grigorieva, S. V. Dubonos, A. A. Firsov,
407: Nature {\bf 438}, 197 (2005)~; A. K. Geim and K. S. Novoselov,
408: Nature Materials~{\bf 9}, 183, (2007)~.
409:
410: \bibitem{yafis}
411: Y. Barlas, T. Pereg-Barnea, M. Polini, R. Asgari, and A. H.
412: MacDonald, Phys. Rev. Lett.{\bf 93}, 236601 (2007)~; M. Polini, R.
413: Asgari, Y. Barlas, T. Pereg-Barnea and A. H. MacDonald, Solid State
414: Commun. {\bf 143}, 58 (2007), M. Polini {\it et al.}
415: Cond-mat/0707.4230~.
416: \bibitem{evens}
417: J. W. Evens, P. A. Thiel and M. C. Bartelt, Sur. Sci. Rep. {\bf 61},
418: 1 (2006)~.
419: \bibitem{mermin}
420: N. D. Mermin, Phys. Rev. {\bf 176}, 250 (1968)~.
421: \bibitem{mayer}
422: J. C. Meyer, A. K. Geim, M. I. Katsnelson, K. S. Novoselov, T.
423: J.Booth and S. Roth, Nature {\bf 446}, 60 (2007)~.
424: \bibitem{MRS}
425: P. C. Martin, E. D. Siggia, and H. A. Rose, Phys. Rev. A {\bf 8},
426: 423 (1973)~.
427: \bibitem{morozov}
428: S. V. Morozov, K. S. Novoselov, M. I. Katsnelson, F. Schedin, L. A.
429: Ponomarenko, D. Jiang, and A. K. Geim Phys. Rev. Lett. {\bf 97},
430: 016801 (2006)~.
431: \bibitem{sang}
432: P. Sangpour, O. Akhavan, A. Z. Moshfegh, G. R. Jafari and M. Reza
433: Rahimi Tabar, Phys. Rev. B {\bf 71}, 155423 (2005)~.
434: \bibitem{bara}
435: A.L. Barabasi and H. E. Stanley {\it Fractal Concepts in Surface
436: Growth} (New York: Cambridge University Press) (1995)~.
437: \bibitem{jafari}
438: G. R. Jafari, S. M. Fazeli, F. Ghasemi, S. M. Vaez Allaei, M. Reza
439: Rahimi Tabar, A. Iraji zad and G. Kavei, Phys.Rev.Lett. {\bf 91},
440: 226101 (2003)~.
441: \bibitem{brenner}
442: D. W. Brenner, Phys. Rev. B {\bf 42}, 9458 (1990)~.
443: \bibitem{CMB}
444: F. Ghasemi, A. Bahraminasab, M. Sadegh Movahed, Sohrab Rahvar, K.
445: R. Sreenivasan, M. Reza Rahimi Tabar, J. Stat. Mech. P11008 (2006)~.
446: \bibitem{L} L. Radzihovsky and D. R. Nelson, Physical Review A, 44, 3525
447: (1991)~; D. R. Nelson and L. Radzihovsky, Europhysics Letters, 16,
448: 79 (1991)~.
449: \bibitem{los}
450: Jan H. Los, Luca M. Ghiringhelli, Evert Jan Meijer, and A. Fasolino,
451: Phys. Rev. B {\bf 72}, 214102 (2005)~.
452: \bibitem{Faso} A. Fasolino, J. H. Los, M. I. Katsnelson,
453: cond-mat/07041793.
454: \bibitem{iordanskii}
455: S. V. Iordanskii and A. E. Koshelev, JETP Lett. {\bf 41}, 574
456: (1985)~.
457: \bibitem{nelson}
458: D. R. Nelson, T. Piran, and S. Weinberg, {\it Statistical Mechanics of Membranes
459: and Surfaces}, World Scientific, Singapore, (2004)~.
460: \bibitem{katsnelson}
461: M. I. Katsnelson and K. S. Novoselov, Solid State
462: Commun. {\bf 143}, 3 (2007)~.
463: \bibitem{Shahbazi}
464: F. Shahbazi, A. Bahraminasab, S. M. Vaez Allaei, M. Sahimi, and M.R.
465: Rahimi Tabar, Phys. Rev. Lett. {\bf 95}, 165505 (2005).
466: \bibitem{Alireza}
467: A. Bahraminasab, S. Mehdi Vaez Allae, F. Shahbazi,
468: Muhammad Sahimi, M.D. Niry, M. Reza Rahimi Tabar, Phys. Rev. B{\bf 75}, 064301(2007)~.
469:
470: \bibitem{aleiner}
471: I. L. Aleiner, K. B. Efetov, Phys. Rev. Lett. {\bf 97}, 236801
472: (2006), S. Das Sarma, E.H. Hwang, and Wang-Kong Tse, Phys. Rev. B {\bf 75}, 121406(R) (2007)~.
473: \bibitem{ziegler}
474: K. Ziegler, Phys. Rev. Lett. {\bf 80}, 3113 (1998)~.
475: \bibitem{morpurgo}
476: A. F. Morpurgo and F. Guinea, Phys. Rev. Lett. {\bf 97}, 196804
477: (2006)~.
478: \bibitem{demartino}
479: A. De Martino, L. Dell'Anna and R. Egger, Phys. Rev. Lett. {\bf 98},
480: 066802 (2007)~.
481: \bibitem{castro}
482: A. H. Castro Neto and E. A. Kim, Cond-mat/0702562~.
483:
484: \end{thebibliography}
485:
486: \newpage
487:
488: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
489: \begin{figure}[t]
490: \begin{center}
491: \includegraphics[width=1.\linewidth]{1.eps}
492: \caption{(color online) Snapshot of the graphene surface at $300 K$
493: (upper graph). The sample contains a lattice size of $500 \times
494: 500$ atoms. In the lower graph, we have plotted the induced magnetic
495: field due to the roughness of graphene surface.}
496: \end{center}
497: \end{figure}
498: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
499:
500: \newpage
501:
502:
503:
504: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
505:
506: \begin{figure}[t]
507: \begin{center}
508: \includegraphics[width=0.8\linewidth]{2.eps}
509: \caption{(color online) Log-log plot of the second moment of height
510: difference as a function of $r$, in $y$ (arm-chairs) direction,
511: which shows that for samples with temperatures $10,30,100$ and $300$
512: $K$ the hight fluctuations have scaling behavior in small scales. It
513: indicates that the roughness exponent (the slope of the plots)
514: decreases with temperature, and means that the surface will be rough
515: at high temperatures. A similar figure can be found for the
516: height-height structure function in $x $ (zigzag) direction. }
517: \end{center}
518: \end{figure}
519: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
520:
521: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
522: \begin{figure}[t]
523: \begin{center}
524: \includegraphics[width=0.8\linewidth]{3.eps}
525: \caption{(color online) The temperature dependence of the roughness
526: exponents show that graphene is smoother in arm-chairs direction as
527: compared to zigzag direction. For very smooth surface the exponent
528: will be $\sim 1.0$. At high temperature limit the exponents approach
529: a random noise exponent (i.e. 0.5). The anisotropy of the graphene
530: is due to
531: the fact that the lattice spacing in arm-chairs and zigzag directions are
532: different. It shows that (inset) the amplitude of the second
533: moments, $C_x$ and $C_y$, are increasing function of temperature. }
534: \end{center}
535: \end{figure}
536:
537: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
538:
539: \newpage
540:
541: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
542: \begin{figure}[t]
543: \begin{center}
544: \includegraphics[width=0.8\linewidth]{4.eps}
545: \caption{(color online)Scale dependence of $Q(r)$, defined by Eq.(1)
546: as a function of $r$. It shows that the height fluctuations have
547: characteristic scales of $90{\AA}$ and $ 125{\AA}$ for $T=300K$ and
548: $T=30K$, respectively. }
549: \end{center}
550: \end{figure}
551: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
552:
553:
554: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
555:
556: \begin{figure}[t]
557: \begin{center}
558: \includegraphics[width=0.8\linewidth]{5.eps}
559: \caption{(color online) Log-log plot of the second moment of the
560: induced gauge field (inset structure function of magnetic field) in
561: $y$ direction, for samples with temperatures $30,100$ and $300$ $K$.
562: The correlation length is about $2-3 {\AA}$. Similar figures can
563: be found for the other three gauge field structure functions.}
564: \end{center}
565: \end{figure}
566: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
567:
568:
569:
570: \end{document}
571: