1: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

2: %Version from May 2003

3: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

4: %

5: %\documentstyle[aps,epsfig,twocolumn]{revtex}

6: %\documentclass[twocolumn,aps,prb,showpacs,preprintnumbers,amsmath,amssymb,psfrac]{revtex4}

7: \documentclass[twocolumn,aps,floats,superscriptaddress,showpacs]{revtex4}

8: \usepackage{amsmath}

9: \renewcommand{\topfraction}{0.85}

10: \renewcommand{\textfraction}{0.1}

11: \renewcommand{\floatpagefraction}{0.75}

12: \usepackage{epsfig}

13: \usepackage{graphicx}% Include figure files

14: \usepackage{dcolumn}% Align table columns on decimal point

15: \usepackage{bm}% bold math

16: 

17: %http://www.ias.ac.in/currsci/oct252001/1011.pdf

18: 

19: \renewcommand{\dag}{^{\dagger}}

20: \newcommand{\dl}{\partial_\ell}

21: \def\gapp{\lower.35em\hbox{$\stackrel{\textstyle>}{\sim}$}}

22: \def\lapp{\lower.35em\hbox{$\stackrel{\textstyle<}{\sim}$}}

23: 

24: \begin{document}

25: %\bibliographystyle{apsrev}

26: %

27: 

28: 

29: \title{Electronic properties of curved graphene sheets}

30: %

31: \author{Alberto Cortijo}

32: \affiliation{Unidad Asociada CSIC-UC3M,

33: Instituto de Ciencia de Materiales de Madrid,\\

34: CSIC, Cantoblanco, E-28049 Madrid, Spain.}

35: 

36: \author{ Mar\'{\i}a A. H. Vozmediano}

37: \affiliation{Unidad Asociada CSIC-UC3M,

38: Universidad Carlos III de Madrid, E-28911

39: Legan\'es, Madrid, Spain. }

40: 

41: 

42: 

43: \date{\today}

44: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

45: \begin{abstract}

46: A model is proposed to study the electronic structure of slightly curved graphene

47: sheets with an arbitrary number of pentagon-heptagon

48: pairs and Stone-Wales defects based on

49: a cosmological analogy. The disorder induced by curvature produces

50: characteristic patterns in

51: the local density of states that can be observed in scanning tunnel and

52: transmission electron microscopy.

53: 

54: \end{abstract}

55: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

56: %

57: \pacs{75.10.Jm, 75.10.Lp, 75.30.Ds}

58: %

59: 

60: \maketitle

61: %\section{Introduction.}

62: The recent synthesis of single or few layers of graphite\cite{Netal05,Zetal05}

63: allows to test the singular transport properties predicted in early

64: theoretical studies\cite{S84,Hal88,GGV96,Khv01} and experiments\cite{Ketal03}.

65: The discovery of a

66: substantial field effect \cite{Netal04} and of ferromagnetic

67: behavior\cite{Eetal03} allows to envisage graphene as a reasonable

68: replacement of nanotubes in electronic applications.

69: Disorder plays a very important role in the electronic

70: properties of low dimensional materials.

71: 

72: Substitution of an hexagon by an n-sided ring in the hexagonal lattice

73: without affecting the threefold coordination of the carbon atoms leads to the warping

74: of the graphene sheet. Rings with $n>6$ $(n<6)$, induce locally positive (negative)

75: curvature.

76: Inclusion of an equal number of pentagons and heptagonal rings

77: would keep the flatness of the sheet at large scales and produce

78: a flat structure with curved portions that would be structurally stable

79: and have distinct electronic properties.

80: This defects give rise to long range modifications

81: in the electronic wave function  that affect the electronics

82: in a way different

83: from that produced by  vacancies or other impurities modelled by local potentials.

84: Pentagon-heptagon pairs and

85: Stone-Wales defects made of two adjacent heptagons and two pentagons

86: form naturally in experiments of ion bombarded nanotubes as a mechanism

87: to reduce the dangling bonds in large vacancies\cite{Aetal98} and

88: have been observed to form in situ

89: in single graphene layers by high-resolution transmission electron

90: microscopy (TEM)\cite{Hetal04}.

91: 

92: We propose a method, based on a cosmological analogy, to compute the

93: electronic structure and transport properties

94: of curved graphene sheets consisting of

95: an arbitrary number of topological defects

96: located at fixed positions of the lattice. We see that, unlike

97: vacancies and voids, the combination of positive and negative

98: curvature gives rise to characteristic

99: inhomogeneous patterns in the density of states that

100: affect the transport properties of the layers and can be observed

101: in scanning tunnel (STM) and

102: electron transmission spectroscopy (ETS). The present analysis can help

103: in the experimental  characterization of graphene samples

104: and in the correct analysis of STM images.

105: The results obtained can be related to

106: recent Electrostatic Force Microscopy (EFM) measurements

107: that indicate large potential differences

108: between micrometer large regions on the surface of highly oriented graphite\cite{GE06}

109: and to the suppression of magnetorresistance in

110: single layered graphene reported in \cite{Metal06}.

111: We apply the formalism to present the corrections to the local

112: density of states induced by pentagon-heptagon pairs and Stone-Wales

113: defects in the average planar graphene sheet.

114: 

115: {\it Description of the model.}

116: The conduction band of graphene is well described by a tight

117: binding model which includes the $\pi$ orbitals

118: which are perpendicular to the plane at each

119: C atom\cite{W47,SW58}. This model describes a semimetal, with zero

120: density of states at the Fermi energy, and where the

121: Fermi surface is reduced to two inequivalent K-points

122: located at the corners of the hexagonal

123: Brillouin Zone.

124: The low-energy excitations with momenta

125: in the vicinity of any of the Fermi points $K_{\pm}$

126: have a linear dispersion

127: and can be described by a continuous model which reduces to the

128: Dirac equation in two dimensions\cite{GGV92,GGV93,GGV94}.

129: In the absence of interactions or disorder mixing  the two

130: Fermi points the electronic properties of the system are well described

131: by the effective low-energy Hamiltonian:

132: \begin{align}

133: {\cal H}_{0\pm}= i\hbar v_{\rm F}

134: ( \sigma_x \partial_x \pm  \sigma_y \partial_y)

135: \;,\label{hamil}

136: \end{align}

137: where $\sigma_{x,y}$ are the Pauli matrices,

138: $v_{\rm F} = (3 t a )/2 $, and $a=  1.4 \AA$ is the

139: distance between nearest carbon atoms. The

140: components of the two-dimensional spinor:

141: $\bar{\Psi_i}( {\bf r} )= (

142: \varphi_A ( {\bf r} ),\varphi_B ( {\bf r }))_i$

143: correspond to the amplitude of the wave function in each of the two

144: sublattices (A and B) which build up the honeycomb structure.

145: Pentagons and heptagons can be viewed as disclinations

146: in the graphene lattice, and, when circling one such defect, the

147: two sublattices in the honeycomb structure as well as the

148: two Fermi points are exchanged.

149: The scheme to incorporate this change in a continuous

150: description was discussed in refs. \cite{GGV92,GGV93} and \cite{GGV01}.

151: The process can be described by means

152: of a non Abelian gauge field, which rotates the spinors in the

153: flavor space of the Fermi points. The two spinors associated to

154: each Fermi point can be  combined

155: into  a four component Dirac spinor $\bar{\Psi}( {\bf r} )= (

156: \Psi_+ ( {\bf r} ),\Psi_- ( {\bf r }))$ which in

157: %

158: %Conical defects with an arbitrary opening angle

159: %made by accumulation of pentagons at the tip

160: %have been observed experimentally in\cite{JRDG03}.

161: %When we try to describe an electron in the planar space with

162: %a conical defect

163: %with a deficit angle $2\pi b$ its wave function must

164: %obey the boundary condition

165: %\begin{equation}

166: %\psi({\bf r},\varphi=2\pi)=e^{i2\pi(1-b)

167: %\frac{\sigma_3}{2}}\psi({\bf r},\varphi=0).

168: %\end{equation}

169: %We can  convert the phase $b$ in a continuous

170: %variable and assume that a lattice distortion which rotates the lattice axis

171: %can be parametrized by the angle of rotation, $\theta ({\bf \vec{r}} )$,

172: %of the local axes with respect to a fixed reference frame. The spinor

173: %can then be written as

174: %\begin{equation}

175: %\psi({\bf r},\varphi)=e^{i(\int_x{\bf A}({\bf y}){\bf dy})\frac{\sigma_3}{2}}

176: %\psi_0({\bf r},\varphi),

177: %\label{spinor2}

178: %\end{equation}

179: %where the potential is that of a vortex

180: %\begin{equation}

181: % ${\bf A}( {\bf r} )  \sim\nabla \theta ( {\bf r} ).$

182: %\nonumber

183: %\end{equation}

184: %This is the basis of the gauge

185: %description of the defects that can be modelled by

186: %fictitious magnetic fields attached to the points of the

187: %graphene plane where the defects where located. This

188: %description supplemented with the effect of curvature,

189: % was successfully applied to obtain the

190: %electronic spectra of fullerenes in\cite{GGV93}. It has

191: %also been used to classify graphene cones in \cite{LC04}.

192: %We must notice that the boundary conditions alone do not

193: %account for the curvature that has to be added to the model

194: %as will be described later.

195: the presence of a single disclination is an eigenstate  of the

196: Hamiltonian\cite{CV06}

197: \begin{eqnarray}

198: H=-i v_{_{F}}{\vec \gamma}.{\vec\partial}+g\gamma^{q}

199: {\vec \gamma}.{\vec A}({\bf r}). \label{hamgauge}

200: \end{eqnarray}

201: $v_{_{F}}$ is the Fermi velocity,

202: $ \gamma^{i}$ are

203: $4\times4$ matrices constructed from the Pauli matrices,

204: $\gamma^q=\frac{\sigma_3}{2}\otimes I$,

205: the latin indices run over the two spatial dimensions and

206: g is a coupling parameter related to the deficit angle of the  defect.

207: 

208: We  use time-independent

209: perturbation theory to calculate corrections to the self-energy

210: %$\Sigma(\textbf{k},\omega)$

211: in the weak coupling regime of the parameter $\widehat{g}\equiv

212: \frac{\Phi}{2\pi L^2},\label{singlecoupling}$ where the constant

213: $\Phi$  is the strength of the vortex: $\Phi=\oint\vec{A}d\vec{r}

214: $  related to the opening angle of the defect by $\Phi=(1-b)$. L

215: is the dimension of the sample.

216: 

217: In a disordered system, in general, because of the presence of an

218: external potential or impurities, the space is inhomogeneous. The

219: Green's function doesn't depend on the difference ($\bf{r}-r'$)

220: and $\bf{k}$ is no longer a good quantum number. If the external

221: potential or the effect of the impurities are time-independent we

222: have elastic scattering and the states \textbf{k} and \textbf{k}'

223: have the same energy. We want to calculate the total density of

224: states $\rho(\omega)$ of the system perturbed by the defect via

225: the vector potential given in eq. (2). This density is the

226: imaginary part of the Green's function integrated over all

227: positions, in the limit $\textbf{r}'\rightarrow \textbf{r}$:

228: %

229: \begin{equation}

230: \rho(\omega)=\int Im

231: G(\omega,\textbf{r},\textbf{r})d\textbf{r}.\nonumber

232: \end{equation}

233: %

234: In terms of the Green's function in momentum representation,

235: $\rho(\omega)$ can be written as:

236: %

237: \begin{equation}

238: \rho(\omega)=Im\int\int \frac{d\textbf{k}}{(2\pi)^{2}} \int

239: \frac{d\textbf{k}'}{(2\pi)^{2}}

240: e^{i\textbf{kr}}e^{-i\textbf{k}'\textbf{r}}G(\omega,\textbf{k},\textbf{k}')d\textbf{r}.

241: \nonumber

242: \end{equation}

243: %

244: The integration over $\textbf{r}$  gives delta function

245: $4\pi^{2}\delta(\textbf{k}-\textbf{k}')$, and $\rho(\omega)$ then

246: reads:

247: \begin{equation}

248: \rho(\omega)=\int \frac{d\textbf{k}}{(2\pi)^{2}}Im

249: G(\omega,\textbf{k},\textbf{k}).\nonumber

250: \end{equation}

251: %The correction to the density of states

252: %induced by the defect is obtained from the Green{s function by

253: %\begin{eqnarray}

254: %\rho(\omega)=\frac{1}{\pi}Im\int Tr G(\omega, {\bf k}) %\nonumber

255: %\end{eqnarray}

256: %

257: 

258: 

259: {\it Generalization to negative curvature and an arbitrary number

260: of defects.} An alternative approach to the gauge theory of

261: defects  is to include the local curvature induced by an

262: n-membered ring by coupling the Dirac equation to a curved space.

263: This approach was introduced in \cite{GGV92,GGV93} to study the

264: electronic spectrum of closed fullerenes and has recently been

265: used for fullerenes  of  different shapes in \cite{PPO06}. The

266: holonomy discused in the gauge approach is included in the

267: formalism by means of the spin connection\cite{FMC06}. In most of

268: the previous works in this context, the graphene sheet was wrapped

269: on a geometrical surface that can be easily parametrized. The

270: situation is more complicated if we try to describe the samples

271: that are being obtained in the laboratory that are flat sheets

272: with local corrugation\cite{Metal06}.

273: 

274: 

275: In this context one can see that the substitution of an hexagon by

276: a polygon of $n<6$ sides gives rise to a conical singularity with

277: deficit angle $(2\pi/6)(6-n)$. This kind of singularities have

278: been studied in cosmology as they are produced by  cosmic strings,

279: a type of topological defect that arises when a U(1) gauge

280: symmetry is spontaneously broken\cite{vilenkin}. We can obtain the

281: correction to the density of states induced by a set of defects

282: with arbitrary opening angle by coupling the Dirac equation to a

283: curved space with an appropriate metric as described in ref.

284: \cite{AHO97}. The metric of a two dimensional space in presence of

285: a single cosmic string in polar coordinates is:

286: %

287: \begin{equation}

288: ds^{2}=-dt^{2}+dr^{2}+c^{2}r^{2}d\theta^{2},\label{metric}

289: \end{equation}

290: %

291: where the parameter $c$ is a constant related to the deficit angle

292: by $c=1-b$.

293: 

294: The dynamics of a massless Dirac spinor in a curved spacetime is

295: governed by the Dirac equation:

296: \begin{equation}

297: i\gamma^{\mu}\nabla_{\mu}\psi=0 \label{dircurv}

298: \end{equation}

299: The difference with the flat space  lies in the definition

300: of the $\gamma$ matrices that satisfy generalized anticommutation

301: relations

302: \begin{equation}

303: \{\gamma^{\mu},\gamma^{\nu}\}=2g^{\mu\nu},%\nonumber

304: \end{equation}

305: %

306: where $g_{\mu \nu}$ is given by (\ref{metric}),

307: and in the covariant derivative operator, defined as

308: \begin{equation}

309: \nabla_{\mu}=\partial_{\mu}-\Gamma_{\mu}

310: \end{equation}

311: where $\Gamma_{\mu}$ is the spin connection of the

312: spinor field that can be

313: calculated using the tetrad formalism\cite{birrell}.

314: \begin{figure}

315:   \begin{center}

316:     \epsfig{file=pentagonlfig2.eps,height=5cm}

317:     \caption{Electronic density  around a conical defect.

318:     }

319:     \label{1string}

320: \end{center}

321: \end{figure}

322: Fig. \ref{1string} shows the solution of the Dirac

323: equation (\ref{dircurv}) in the presence

324: of a single defect with a positive deficit angle

325: (positive curvature).

326: The electronic density is strongly peaked at the

327: position of the defect suggesting a bound state but

328: the behavior at large distances is a power law with an angle-dependent

329: exponent less than two

330: which corresponds to a non normalizable wave function.

331: This  behavior  is similar to the

332: one found in the case of a single vacancy\cite{PGLPC06}

333: and suggests that a

334: system with a number of this defects with overlapping

335: wave functions will be metallic.

336: 

337: The case of   a single cosmic string which

338: represents a deficit angle in the space can be generalized

339: to describe seven membered rings representing an angle

340: surplus by considering a value for the deficit angle c

341: greater than $1$.  This situation is non-physical from a general

342: relativity viewpoint as it would correspond to a string with negative mass

343: density but it makes perfect sense in our case. The scenario can also

344: be generalized to describe an arbitrary number of pentagons and

345: heptagons by using  the following metric:

346: %

347: \begin{equation}

348: ds^{2}=-dt^{2}+e^{-2\Lambda(x,y)}(dx^{2}+dy^{2}),\label{genmetric}

349: \end{equation}

350: %

351: where $\Lambda(\textbf{r})=\sum^{N}_{i=1}4\mu_{i}\log(r_{i})$ and

352: $r_{i}=[(x-a_{i})^{2}+(y-b_{i})^{2}]^{1/2}.$ This metric describes

353: the space-time around N parallel cosmic strings, located at the

354: points $(a_{i},b_{i})$. The parameters $\mu_{i}$ are related to the

355: angle defect or surplus by the relationship $c_{i}=1-4\mu_{i}$ in

356: such manner that if $c_{i}<1 (>1)$ then $\mu_{i}>0 (<0)$.

357: 

358: From equation (\ref{dircurv}) we can

359: write down

360: the Dirac equation for the electron propagator, $S_{F}(x,x')$:

361: %

362: \begin{equation}

363: i\gamma^{\mu}({\bf r})(\partial_{\mu}-\Gamma_{\mu})S_{F}(x,x')=

364: \frac{1}{\sqrt{-g}}\;\delta^{3}(x-x'),

365: \label{propcurv}

366: \end{equation}

367: %

368: where $x=(t, {\bf r})$.

369: The local density of states $N(\omega,\textbf{r})$ is obtained from (\ref{propcurv})

370: by Fourier transforming the time component and taking the limit

371: ${\bf r}\to {\bf r'}$:

372: \begin{equation}

373: N(\omega,\textbf{r})=Im

374: Tr S_{F}(\omega,\textbf{r},\textbf{r}).\label{LDOS}

375: \end{equation}

376: We solve eq. (\ref{LDOS}) considering the  curvature induced

377: by the defects as a perturbation of the flat graphene layer.

378: The details of the calculation will be given elsewere\cite{CV06}.

379: Here we will show the results obtained.

380: 

381: {\it Results.}

382: Fig. \ref{pair2} shows the correction to the local density of states

383: at energy $w=2.1 eV$ and for a large region of the graphene plane with a

384: pentagon (p)-heptagon (h) pair located out of plane

385: at $p=[-0.3,-5], h=[0.3,-5]$. The LDoS is normalized

386: with the clean DoS of graphene at the given energy. The color code is indicated in

387: the figure: green stands for the DOS of perfect graphene at the given energy

388: and red (blue) indicates an accumulation (depletion) of the density in the

389: area. The correction obtained is of the order of a few percent.

390: We can see that pentagonal (heptagonal) rings  enhance

391: (deppress) the electron density. A similar result has been obtained in

392: \cite{TT94} with numerical simulations. It is to note that a somehow

393: contradictory result was obtained in \cite{AFM98} where they studied

394: the electrostatics of a graphene plane with defects. They found that

395: disclinations corresponding to rings with more (less) than six carbon atoms function

396: as attractors (repellent) to point charges. It is obvious that this

397: issue needs further investigation.  The dipolar character of the defect is

398: clear from the figure and makes this type of defects unique. The effects

399: produced will extend spatially much beyond the scale of the defect what

400: can be related with

401: recent Electrostatic Force Microscopy (EFM) measurements

402: that indicate large potential differences

403: between micrometer large regions on the surface of highly oriented graphite

404: that do not show inhomogeneities in STM\cite{GE06}.

405: The interference patterns  in the DOS depend on the position and on the relative

406: orientation of the dipoles. Fig. \ref{stone-wales2}

407: shows the patterns produced by a Stone-Wales defect (two pairs

408: of pentagons ($p_i$) heptagons $(h_i)$)

409: located out of plane at positions $p_1=[5,5.6], p_2=[6.5,6],

410: h_1=[6,5.5], h_2=[6,6.9]$  for a frequency $w=2.1 eV$.

411: %

412: \begin{figure}

413:   \begin{center}

414:     \epsfig{file=newfig3.eps,height=5cm}

415:     \caption{(Color online) Image of the correction to the local density of states in a large portion

416:     of the plane with a heptagon-pentagon  pair located out of plane at position

417:     $[0.3, -5], [-0.3, -5]$ .

418:     }

419:     \label{pair2}

420: \end{center}

421: \end{figure}

422: %

423: \begin{figure}

424:   \begin{center}

425:     \epsfig{file=newfig4.eps,height=5cm}

426:     \caption{(Color online) Correction to the local density of states around a Stone-Wales defect

427:     located out of the image in the upper right corner.}

428:     \label{stone-wales2}

429: \end{center}

430: \end{figure}

431: %

432: 

433: {\it A random distribution of defects.}

434: A random distribution of

435: topological defects can be described in the continuous model

436: by a (non abelian) random gauge field.

437: The disorder is defined by a single dimensionless quantity,

438: $\Delta$, which is proportional to the average fluctuations

439: of the field:

440: \begin{equation}

441: \langle {\bf \vec{A}} ( {\bf \vec{r}} ) {\bf \vec{A}} (

442: {\bf \vec{r}'} ) \rangle = \Delta \delta^2 ( {\bf \vec{r}} -

443: {\bf \vec{r}'} )

444: \label{delta}

445: \end{equation}

446: The statistical properties of the gauge field induced by

447: topological defects were analyzed in ref. \cite{GGV01} following a

448: renormalization group scheme and a complete phase diagram as a

449: function of disorder and Coulomb interaction was obtained in

450: \cite{SGV05}.

451: 

452: In ref. \cite{GGV01} it was shown that

453: when  pentagons and heptagons are bound in dislocations

454: with average distance $b$,  the vector field of a vortex-antivortex

455: dipole decays as $r^{-2}$ and the behavior of $\Delta$ was found to be:

456: \begin{equation}

457: \Delta \propto \Phi_0^2 n b^2

458: \label{disloc}

459: \end{equation}

460: where $ n$ is the density of dislocations.

461: 

462: Assuming that random fields induced by topological defects have

463: the same statistical properties to those with gaussian disorder with

464: the same value of $\Delta$,

465: a renormalization group (RG) analysis using the replica trick

466: gives rise to an effective interaction between

467: fermion fields in different replicas. The resulting self-energy

468: is logarithmically divergent what  can be interpreted

469: as a renormalization of the density of states\cite{GGV01}.

470: The results obtained in this paper show

471: a qualitative picture of the corrected DoS. The influence of

472: these defects on the localization in graphene has been

473: analyzed recently in \cite{MG06}.

474: 

475: The type of disorder analyzed in  this article belongs to the

476: random gauge field in the classification of \cite{SGV05}. There it

477: was shown that the combination of this type of disorder with the

478: unscreened Coulomb interaction give rise to a line of infrared

479: stable fixed points. Topological defects give rise also to long

480: range correlated disorder  whose properties will be analyzed

481: elsewhere\cite{CV06}.

482: 

483: {\it Conclusions and open problems.} We have presented in this

484: article results on the electronic structure of a curved graphene

485: sheet with pentagon-heptagon pairs that we expect to be present in

486: the samples that are subject of present investigation. The

487: formalism can also apply to clean samples of graphite showing

488: strong anisotropy and to curved samples of graphene with an

489: arbitrary number of defects with any sign of the curvature. The

490: model can also be  used to study transport properties of the

491: curved graphene sheets and of nanographite and results will be

492: presented in \cite{CV06}. The main achievement  of this paper is

493: the proposal of a formalism that can describe the corrugations

494: observed in the experiments and predicts charge inhomogeneities

495: that can be used to characterize the samples. We have computed the

496: local density of states with the geometrical formalism  found that

497: it is enhanced around defects which induce positive curvature in

498: the lattice while the charge is "repelled" from regions with

499: negative curvature. Heptagon-pentagon pairs that keep the graphene

500: sheet flat in the long range behave as dipoles and give rise to

501: characteristic modulations of the DOS that can be observed by STM

502: and can explain recent experiments. The magnitude of the

503: oscillations increases with the frequency and the characteristic

504: patterns could be used to characterize the graphene samples. The

505: features predicted in this work should also be observable in other

506: layered materials with similar structure as boron

507: nitride\cite{TTM01}. The present analysis can help to clarify the

508: issue of the analysis and interpretation of STM images.

509: 

510: %\section{Acknowledgments.}

511: We thank B. Valenzuela, P. Guinea and P. Esquinazi for many illuminating discussions.

512: A. C. thanks N. Ozdemir for kindly  providing details of the calculations

513: in ref. \cite{AHO97}.

514: Funding from MCyT (Spain) through grant MAT2002-0495-C02-01

515: and from the European Union Contract No. 12881 (NEST) is

516: acknowledged.

517: 

518: \bibliography{EPL}

519: \end{document}

520: