0707.4575/AA.tex
1: \documentclass[floatfix,preprintnumbers,superscriptaddress,showpacs,prl,
2:                preprint]{revtex4}
3: %\documentclass[floatfix,preprintnumbers,superscriptaddress,showpacs,prl,
4: %               twocolumn]{revtex4}
5: 
6: \usepackage{graphicx}
7: \usepackage{amsmath,amsfonts,amssymb,bm}
8: 
9: \begin{document}
10: 
11: \bibliographystyle{apsrev}
12: 
13: \title{ Strong covalent bonding between two graphene layers }
14: 
15: \author{P.L. de Andres}
16: \affiliation{
17: Instituto de Ciencia de Materiales de Madrid (CSIC), Cantoblanco,
18: 28049 Madrid, Spain} 
19: 
20: \author{R. Ram\'{\i}rez}
21: \affiliation{
22: Instituto de Ciencia de Materiales de Madrid (CSIC), Cantoblanco,
23: 28049 Madrid, Spain} 
24: 
25: \author{J.A. Verg\'es}
26: \affiliation{
27: Instituto de Ciencia de Materiales de Madrid (CSIC), Cantoblanco,
28: 28049 Madrid, Spain} 
29: 
30: \date{\today}
31: 
32: \begin{abstract}
33: We show that two graphene
34: layers stacked directly on top of each other
35: (AA stacking) form strong chemical bonds when the distance between planes is
36: 0.156 nm. Simultaneously, C-C in-plane bonds are considerably
37: weakened from partial double-bond (0.141 nm) to single bond (0.154 nm).
38: This polymorphic form of graphene bilayer is
39: meta-stable w.r.t. the one bound by van der Waals forces at a
40: larger separation (0.335 nm) with an activation energy of 0.16 eV/cell.
41: Similarly to the structure found in hexaprismane,
42: C forms four single bonds in a geometry
43: mixing $90^{0}$ and $120^{0}$ angles. 
44: Intermediate separations between layers can be stabilized under external
45: anisotropic stresses showing a rich electronic structure changing
46: from semimetal at van der Waals distance, to metal when compressed, to
47: wide gap semiconductor at the meta-stable minimum.
48: \end{abstract}
49: 
50: \pacs{81.05.Uw,73.22-f,61.50.Ah,73.61.Cw}
51: 
52: \maketitle
53: 
54: Carbon shows one of the richest chemistry in the periodic table and
55: it is often found in allotropic forms. In molecules it is the basis 
56: for organic compounds, being central to different fields from biology to 
57: electronics in new materials. In solid state it shows very different
58: properties drifting from a soft metal (graphite, the most stable
59: phase at P=0 GPa, T= 0 K) to a hard wide gap 
60: semiconductor (diamond). New forms like fullerenes and nano-tubes have 
61: raised even more the interest in carbon for their potential applications. 
62: Recently, the realization of two-dimensional periodic systems made by the 
63: stacking of few graphene layers (FGL), going down to the single layer, 
64: has attracted much interest as the basis for new electronic 
65: devices\cite{novoselov04}. 
66: The peculiar linear dispersion found in the electronic band structure near 
67: the charge neutrality point (Dirac Point), where the carriers behave like 
68: mass-less chiral relativistic particles, translates in all sort of new 
69: phenomena related to transport properties on these 
70: systems\cite{heersche07}. 
71: Moreover, a variety of preparation techniques have been used
72: giving rise to samples showing
73: important differences\cite{rokuta99,horiuchi03,novoselov04,berger04,meyer07};
74: most notably: charge accumulation regions associated with physical
75: corrugation found in free standing graphene\cite{meyer07},
76: new properties induced in the graphene layers by
77: the epitaxial growth on a SiC substrate\cite{berger04}, 
78: or a modification of the stacking, from Bernal AB phase
79: to AA, found in carbon nanofilms grown from graphite
80: oxide\cite{horiuchi03}.
81: Accurate and detailed information on these samples 
82: obtained from structural techniques is sometimes 
83: difficult to interpret;
84: state-of-the-art theoretical total-energy methods are necessary
85: to understand the precise atomic and electronic structure of these 
86: films. In this paper, we report on the
87: formation of strong covalent bonds between graphene layers
88: stacked directly on top of each other (AA) at a distance
89: that is much smaller than $\sim 0.335$ nm, that is, the typical
90: distance for an alternating (AB) stacking based on weak
91: van der Waals forces (Fig. 1).
92: On this meta-stable polymorphic form of a graphene bi-layer
93: each carbon is bonded to the four 
94: nearest neighbours, at 0.154 and 0.156 nm for in-plane and 
95: out-of-plane bonds respectively. 
96: Under these conditions, the bi-layer is a wide gap
97: semiconductor (indirect gap of 0.91 eV). 
98: As a function of the separation between layers, transport properties of the 
99: un-doped AA stacking are rich: 
100: at large distances between planes (e.g., as found
101: in graphite) the system is very close to a semi-metal,
102: mostly dominated by the single graphene layer
103: properties. As the distance between layers decreases it is possible
104: to find interlayer distances and/or different 2D unit cell sizes where 
105: the bi-layer becomes metallic.
106: 
107: \begin{figure}
108: \includegraphics[clip,width=0.9\columnwidth]{fig1.eps}
109: \caption{(Color online) 
110: Meta-stable extended 2D carbon allotrope formed by two graphene layers 
111: at covalent C-C bond distance and direct on-top stacking (AA). 
112: The 2D unit cell is shown ($a=b=0.267$ nm, $\gamma=120^{o}$). 
113: }
114: \label{fig1}
115: \end{figure}
116: 
117: Our ab-initio calculations are based in
118: Density Functional Theory
119: (DFT)\cite{hohenberg65} 
120: and a local approximation
121: to describe exchange and correlation (LDA)\cite{kohn65,gga}. 
122: LDA calculations performed with CASTEP\cite{payne02,accelrys,precision} 
123: reproduce very well distances and angles for the strong
124: sp$^{2}$ bonds inside the graphene layer, 
125: predicting as the most stable configuration a honeycomb lattice 
126: with a C-C distance of 0.141 nm, 
127: and a bond population of 1.53.
128: Experimental value is 0.142 nm 
129: (fractional error less than 1{\%}), 
130: in between a carbon double bond 
131: (typical length 0.133 nm) and a single one (0.154 nm).  
132: A negligible charge transfer (0.3{\%}) takes place from 2s to 2p orbitals. 
133: These geometrical results, together with those obtained for electronic 
134: and vibrational properties, demonstrate the ability of DFT to describe 
135: the C-C bond at typical distances allowing
136: the formation of covalent bonds. 
137: A rough electron-counting picture for 
138: the graphene layer would be 
139: each C atom sharing one electron with each of the in-plane 
140: three nearest C neighbours, while the 
141: fourth electron is delocalized among them, 
142: making three stronger C-C bonds 
143: with a character somewhere in between a single and a double bond. 
144: In bulk graphite this fourth electron would be responsible for the 
145: appearance of pockets near the Fermi energy and the in-plane 
146: conductivity. 
147: This scenario makes plausible to use this extra electron to establish 
148: single bonds between carbons across the layers. 
149: While van der Waals interaction is weak and
150: not accurately described by a local DFT,
151: the formation of the new allotropic form of graphene 
152: bi-layer rather involves interactions between carbons
153: at shorter covalent bonding distances
154: where the DFT formalism is accurate and realistic.
155: This is independent of the basis choosen to solve
156: the equations;
157: we have checked that quantum chemistry calculations made
158: with localized basis sets and mixed functional methods
159: in small clusters concur with the
160: ones derived from plane-waves basis for extended 2D systems.
161: 
162: \begin{figure}
163: \includegraphics[clip,width=0.85\columnwidth]{fig2a.eps}
164: \includegraphics[clip,width=0.85\columnwidth]{fig2b.eps}
165: \caption{(Color online)
166: First-principles total energy landscape
167: for the graphene bi-layer as a function of the 
168: distance between layers ($d$) and the 2D lattice parameter ($a$). 
169: Label A shows the local minimum reported
170: in this work at $d=0.156$ nm and $a=0.267$ nm 
171: corresponding to the formation of covalent bonds across layers.  
172: Contour lines start at -618 eV and go up in steps of 0.05 eV 
173: (the minimum in A is at -617.208 eV).
174: }
175: \label{fig2}
176: \end{figure}
177: 
178: Recent papers have investigated the electronic structure of
179: the standard alternating AB stacking since it is energetically
180: favoured over the AA stacking\cite{latil06}.
181: However, the expected energy difference is necessarily small due to the 
182: weak interaction between layers, about 0.02 eV/cell in our calculations.
183: The barrier to transform one stacking into the
184: other might be higher, but of the same order.
185: Therefore, we have investigated the AA bi-layer, searching for 
186: new structural configurations; we find a meta-stable energy minimum 
187: at about half the usual distance between layers in graphite.
188: This new structure implies an important lateral relaxation of the 
189: 2D unit cell, and displays electronic 
190: properties quite different from the global van der Waals-like minimum. 
191: Fig. 2 shows a 2D total energy map for the system near the new minimum:
192: the meta-stable configuration appears around an interlayer distance 
193: $d=0.156$ nm and lattice parameter $a=0.267$ nm (label A in Fig. 2). 
194: The alternating Bernal stacking (AB) does not show a similar 
195: meta-stable local minimum in our calculations.
196: The reason for the different behaviour lies on the different
197: coordination of the C atoms in the bi-layer stacking AA and AB. 
198: While {\em all} C atoms can form an interlayer covalent
199: bond in the AA stacking, only half of the atoms have this
200: possibility for the AB case. Consequently, when a small separation is
201: forced in the AB bi-layer, buckling of both planes
202: can release stress efficiently, and
203: sp$^{3}$ coordination with nearly tetrahedral angles appears
204: (the resulting structure is a 2D diamond precursor).
205: In the AA case, the formation of a meta-stable configuration
206: is favoured because symmetry does not allow the relaxation of
207: structural strain by buckling. 
208: A similar idea has been put forward
209: to explain the meta-stability of n-prismanes\cite{jenkins00}.
210: Stacking of carbon layers with covalent
211: bonds accross layers seems unnoticed;
212: because this configuration is meta-stable it should require
213: contributing some external energy to the system. 
214: A natural way of doing this is to grow the layers
215: epitaxially on a substrate imposing a stretched length for the
216: 2D unit cell. 
217: %Until now, most of the structural information about FGL comes from 
218: %diffraction data, in particular Transmission Electron Microscopy
219: %(TEM). At normal incidence, TEM usually results in a 6-fold pattern 
220: %that cannot distinguish between the AA and domains of the AB stacking;
221: %therefore off-normal measurements and a theoretical analysis
222: %are needed to distinguish both cases\cite{meyer07}.
223: However, we should mention that the AA stacking has been reported
224: in the literature for some related system\cite{horiuchi03}. 
225: 
226: \begin{figure}
227: \includegraphics[clip,width=0.8\columnwidth]{fig3.eps}
228: \caption{Phonon spectrum
229: calculated at the local minimum A in 
230: Fig. 2. 
231: The x-axis samples the boundary of the irreducible 2D Brillouin zone. 
232: Lines between points are only meant to guide the eye.
233: }
234: \label{fig3}
235: \end{figure}
236: 
237: We have obtained the barrier to escape from minimum A to the global 
238: one (G) by applying first a Linear Synchronous Transit (LST) 
239: transition state search, followed by a Quadratic Synchronous Transit 
240: (QST) method. We find a barrier of $0.16 \pm 0.04$ eV/cell, 
241: allowing us to predict that the structure A is stable at room 
242: temperature. 
243: The energy barrier for the formation of the meta-stable state 
244: (from G to A) amounts to 4.80 eV/cell, the A configuration being 
245: 4.64 eV/cell higher in energy than the G one. 
246: The path from A to the transition state involves a simultaneous 
247: modification of parameters, $d$ and $a$ (Fig. 2) 
248: due to the correlation between bonds 
249: formed in and out the planes. 
250: Boundary conditions keeping the parameter $a$ fixed to a given value 
251: make a different scenario with interesting consequences. 
252: If $a$ is kept at a constant value of $0.279$ nm, 
253: the local minimum A is established at $d=0.155$ nm 
254: and the barrier grows to 0.8 eV/cell. 
255: For a constant value of $a= 0.291$ nm, 
256: A becomes the global minimum, and the barrier from A to G 
257: goes to 1.7 eV/cell, 
258: A being lower in energy than G by 1.03 eV/cell. 
259: 
260: \begin{figure}
261: \includegraphics[clip,width=1.0\columnwidth]{fig4.eps}
262: \caption{
263: Evolution of the electronic band structure for the bi-layer 
264: as separation changes from van der Waals-like distance (a)
265: to the small separation allowing the chemical
266: bonding of graphene sheets (d). 
267: (a) $d=0.358$ nm, $a=0.243$ nm (global minimum G, semi-metallic);
268: (b) $d=0.300$ nm, $a=0.250$ nm (2D metal); 
269: (c) $d= 0.1625$ nm, $a= 0.2645$ nm (near the transition state, 2D metal); 
270: and 
271: (d) $d= 0.156$ nm, $a=0.267$ nm (local minimum A, insulator).  
272: Fermi energy is used as the origin for energies.
273: }
274: \label{fig4}
275: \end{figure}
276: 
277: Let us further characterize the new bonding configuration
278: after the formation of chemical bonds between C atoms
279: located in different layers.
280: The building of these bonds produces a weakening of the
281: sp$^{2}$-like in-plane bonds, that elongate from
282: 0.141 nm to 0.154 nm.
283: In A, we observe a 0.1 electron charge transfer from the 
284: 2p to the 2s orbital, and the formation of a single bond between 
285: carbons across the two graphene layers at 0.156 nm with a calculated 
286: bond order of 0.92. 
287: This distance is typical of single C-C bonding for 
288: substances like diamond, propane, etc.\cite{pauling}, 
289: supporting the formation of a chemical bond in place of the 
290: previous weak van der Waals interaction. 
291: We notice that similar strained carbon structures have been observed in 
292: molecular systems known as n-prismanes\cite{allinger83}. 
293: Quantum chemical calculations performed with the program
294: GAMESS\cite{gamess} confirm
295: the building of single C-C bonds across parallel carbon 
296: hexagonal rings saturated with H to form
297: the hexaprismane. C-C bond distances and angles
298: are similar to those found in the
299: graphene bi-layer (A).
300: In agreement with our periodic solid-state calculations, 
301: this is a meta-stable molecular configuration w.r.t.
302: van der Waals-like separation between two benzene molecules. 
303: Our calculations give a barrier between the meta-stable structure 
304: and the global minimum (C$_6$H$_6$-C$_6$H$_6$, one hexagonal ring)
305: of about 0.83 eV per C-C bond. 
306: This value decreases consistently as more rings are added; 
307: already for three hexagonal rings (C$_{13}$H$_{8}$-C$_{13}$H$_{8}$) 
308: it goes down to several tenths of eV per C-C bond. 
309: As the number of hexagonal rings increases this barrier 
310: converges to our result for the graphene bi-layer.
311: 
312: Fig. 3 gives the phonon spectra at the minimum A calculated 
313: with a linear response formalism\cite{refson06}. 
314: The phonon spectra has no dispersion in the direction perpendicular 
315: to C layers and shows that the new minimum is stable with 
316: respect to small displacements that preserve the unit cell 
317: area\cite{nota}. 
318: The optical branches around 1600 cm$^{-1}$ at $\Gamma$ 
319: can be compared with those measured for 
320: graphite\cite{maultzsch04}, 
321: although bonding in the layer is now weaker than for graphite. 
322: Near 1100 and 1250  cm$^{-1}$ we observe a couple of optical 
323: modes related to vibrations perpendicular to the layers that are 
324: similar in energy to that found for two
325: C$_{6}$H$_{6}$ rings (hexaprismane) vibrating against each other 
326: at C-C covalent distances. These may be used to experimentally
327: identify the bilayer.
328: 
329: Transport properties on FGL-based devices are determined by the 
330: band-structure of the material. Therefore, we study 
331: the electronic structure of the bi-layer for different structural parameters 
332: (size of the 2D unit cell, $a$, and separation between layers, $d$). 
333: A single graphene layer displays a semi-metallic character
334: with valence and conduction bands touching in 
335: the corners of the Brillouin zone,
336: \{{\bf K}\},
337: and the dispersion relation being linear.
338: At the van der Waals-like separation between layers (0.358 nm), 
339: the interaction is weak, but already a marginal 2D metal starts to 
340: form. The 2D Fermi circle is centred at the corners of the Brillouin 
341: zone, {\bf K}, with a very small radius and the density of states 
342: at the Fermi energy is nearly zero (Fig. 4a). 
343: We notice that in the AA stacking
344: the bands near {\bf K} are still linear, unlike
345: the AB stacking where the bands approach {\bf K} quadratically\cite{latil06}.
346: A new situation emerges if the two layers are forced to
347: get closer to each other.
348: Fig. 4b shows the band structure for such a
349: non-equilibrium configuration ($a=0.250$ nm, $d=0.300$ nm).
350: For this geometry, repulsive forces on atoms on each layer are 0.024 eV/nm. 
351: %Internal stresses for this out-of-equilibrium configuration amount
352: %to $\sigma_{xx}$=$\sigma_{yy}$=10 GPa, and $\sigma_{zz}$=-2 GPa.
353: A comparison between panels a and b in Fig. 4 shows how the radius 
354: of the Fermi circle increases, yielding a distinctively non-zero 
355: density of states and making the bi-layer a 2D metal. 
356: This picture is still valid near the transition state, 
357: where the Fermi line is approaching the symmetry point $\Gamma$ 
358: in the Brillouin zone (Fig. 4c). 
359: Further down the distance between the two layers, the system develops 
360: strong single covalent bonds, and the bi-layer becomes a wide gap 
361: semiconductor.
362: % (dicho mas arriba)
363: % (an indirect gap of 0.91 eV is clearly visible in Fig. 4d).
364: 
365: Finally, we have explored the role of external stresses on the 
366: bi-layer by applying in-plane tensile stresses of 
367: $\sigma_{xx}$=$\sigma_{yy}$= 3, 6 and 9 GPa.
368: As expected, by forcing the 2D unit cell to extend, 
369: the minimum at A is stabilized and the barrier grows to 0.43, 
370: 0.88 and 1.4 eV/cell respectively. 
371: The local minimum A changes so the 2D unit cell size grows 
372: from 0.267 nm to 0.273, 0.280 and 0.289 nm respectively, 
373: while the two layers come closer together by a small distance 
374: (0.0008 nm for 6 GPa). 
375: We notice that around G the strain is approximately half the value 
376: around A (from 0.243 to 0.249 nm for the 6 GPa stress), 
377: a consequence of the existence of stronger sp$^{2}$ bonds. 
378: 
379: In conclusion, we have found a new polymorphic form for two extended 
380: flat 2D graphene layers stacked with AA sequence where carbon atoms 
381: located in atop positions establish new 
382: covalent bonds. This meta-stable configuration is not 
383: subject to thermodynamic instability
384: and shows a barrier 
385: large enough to make it feasible at room temperature.
386: As a function of the separation between the two layers, 
387: their electronic properties range from a 
388: semi-metal (layers far away apart) to a weak 2D metal 
389: (van der Waals distances, low density of states at the Fermi energy) 
390: to a stronger 2D metal (intermediate distances, higher density of 
391: states at the Fermi energy), and finally to a wide gap semiconductor 
392: (covalent bonding distance). External stresses can help to further 
393: stabilize these configurations, as well as to control the separation 
394: between layers. 
395: The new predicted semiconductor should allow
396: traditional doping with impurities (B, N) opening a well-defined
397: way towards strict 2D electronics.
398: 
399: This work has been financed by the CICYT (Spain) under contracts
400: MAT-2005-3866, MAT-2006-03741, FIS-2006-12117-C04-03, 
401: and NAN-2004-09183-C10-08. 
402: We acknowledge the use of the Spanish Supercomputing Network
403: and the CTI (CSIC).
404: 
405: %\bibliography{AA}
406: 
407: \begin{thebibliography}{20}
408: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi
409: \expandafter\ifx\csname bibnamefont\endcsname\relax
410:   \def\bibnamefont#1{#1}\fi
411: \expandafter\ifx\csname bibfnamefont\endcsname\relax
412:   \def\bibfnamefont#1{#1}\fi
413: \expandafter\ifx\csname citenamefont\endcsname\relax
414:   \def\citenamefont#1{#1}\fi
415: \expandafter\ifx\csname url\endcsname\relax
416:   \def\url#1{\texttt{#1}}\fi
417: \expandafter\ifx\csname urlprefix\endcsname\relax\def\urlprefix{URL }\fi
418: \providecommand{\bibinfo}[2]{#2}
419: \providecommand{\eprint}[2][]{\url{#2}}
420: 
421: \bibitem[{\citenamefont{Novoselov}(2004)}]{novoselov04}
422: \bibinfo{author}{\bibfnamefont{K.~S.} \bibnamefont{Novoselov}},
423:   \bibinfo{journal}{Science} \textbf{\bibinfo{volume}{306}},
424:   \bibinfo{pages}{666} (\bibinfo{year}{2004}).
425: 
426: \bibitem[{\citenamefont{Heersche et~al.}(2007)\citenamefont{Heersche,
427:   Jarillo-Herrero, Oostinga, Vandersypen, and Morpurgo}}]{heersche07}
428: \bibinfo{author}{\bibfnamefont{H.~B.} \bibnamefont{Heersche}},
429:   \bibinfo{author}{\bibfnamefont{P.}~\bibnamefont{Jarillo-Herrero}},
430:   \bibinfo{author}{\bibfnamefont{J.~B.} \bibnamefont{Oostinga}},
431:   \bibinfo{author}{\bibfnamefont{L.~M.~K.} \bibnamefont{Vandersypen}},
432:   \bibnamefont{and} \bibinfo{author}{\bibfnamefont{A.~F.}
433:   \bibnamefont{Morpurgo}}, \bibinfo{journal}{Nature}
434:   \textbf{\bibinfo{volume}{446}}, \bibinfo{pages}{56} (\bibinfo{year}{2007}).
435: 
436: \bibitem[{\citenamefont{Rokuta et~al.}(1999)\citenamefont{Rokuta, Hasegawa,
437:   Itoh, Yamashita, Tanaka, Otani, and Oshima}}]{rokuta99}
438: \bibinfo{author}{\bibfnamefont{E.}~\bibnamefont{Rokuta}},
439:   \bibinfo{author}{\bibfnamefont{Y.}~\bibnamefont{Hasegawa}},
440:   \bibinfo{author}{\bibfnamefont{A.}~\bibnamefont{Itoh}},
441:   \bibinfo{author}{\bibfnamefont{K.}~\bibnamefont{Yamashita}},
442:   \bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{Tanaka}},
443:   \bibinfo{author}{\bibfnamefont{S.}~\bibnamefont{Otani}}, \bibnamefont{and}
444:   \bibinfo{author}{\bibfnamefont{C.}~\bibnamefont{Oshima}},
445:   \bibinfo{journal}{Surf. Sci.} \textbf{\bibinfo{volume}{427}},
446:   \bibinfo{pages}{97} (\bibinfo{year}{1999}).
447: 
448: \bibitem[{\citenamefont{Horiuchi et~al.}(2003)\citenamefont{Horiuchi, Gotou,
449:   Fujiwara, Sotoaka, Hirata, Kimoto, Asaka, Yokosava, Matsui, Watanabe
450:   et~al.}}]{horiuchi03}
451: \bibinfo{author}{\bibfnamefont{S.}~\bibnamefont{Horiuchi}},
452:   \bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{Gotou}},
453:   \bibinfo{author}{\bibfnamefont{M.}~\bibnamefont{Fujiwara}},
454:   \bibinfo{author}{\bibfnamefont{R.}~\bibnamefont{Sotoaka}},
455:   \bibinfo{author}{\bibfnamefont{M.}~\bibnamefont{Hirata}},
456:   \bibinfo{author}{\bibfnamefont{K.}~\bibnamefont{Kimoto}},
457:   \bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{Asaka}},
458:   \bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{Yokosava}},
459:   \bibinfo{author}{\bibfnamefont{Y.}~\bibnamefont{Matsui}},
460:   \bibinfo{author}{\bibfnamefont{K.}~\bibnamefont{Watanabe}},
461:   \bibnamefont{et~al.}, \bibinfo{journal}{Jpn. J. Appl. Phys.}
462:   \textbf{\bibinfo{volume}{42}}, \bibinfo{pages}{L1073} (\bibinfo{year}{2003}).
463: 
464: \bibitem[{\citenamefont{Berger et~al.}(2004)\citenamefont{Berger, Song, Li, Li,
465:   Ogbazghi, Feng, Dai, Marchenkov, Conrad, First et~al.}}]{berger04}
466: \bibinfo{author}{\bibfnamefont{C.}~\bibnamefont{Berger}},
467:   \bibinfo{author}{\bibfnamefont{Z.}~\bibnamefont{Song}},
468:   \bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{Li}},
469:   \bibinfo{author}{\bibfnamefont{X.}~\bibnamefont{Li}},
470:   \bibinfo{author}{\bibfnamefont{A.~Y.} \bibnamefont{Ogbazghi}},
471:   \bibinfo{author}{\bibfnamefont{R.}~\bibnamefont{Feng}},
472:   \bibinfo{author}{\bibfnamefont{Z.}~\bibnamefont{Dai}},
473:   \bibinfo{author}{\bibfnamefont{A.~N.} \bibnamefont{Marchenkov}},
474:   \bibinfo{author}{\bibfnamefont{E.~H.} \bibnamefont{Conrad}},
475:   \bibinfo{author}{\bibfnamefont{P.~N.} \bibnamefont{First}},
476:   \bibnamefont{et~al.}, \bibinfo{journal}{J. Phys. Chem. B}
477:   \textbf{\bibinfo{volume}{108}}, \bibinfo{pages}{19912}
478:   (\bibinfo{year}{2004}).
479: 
480: \bibitem[{\citenamefont{Meyer et~al.}(2007)\citenamefont{Meyer, Geim,
481:   Katsnelson, Novoselov, Booth, and Roth}}]{meyer07}
482: \bibinfo{author}{\bibfnamefont{J.~C.} \bibnamefont{Meyer}},
483:   \bibinfo{author}{\bibfnamefont{A.~K.} \bibnamefont{Geim}},
484:   \bibinfo{author}{\bibfnamefont{M.~I.} \bibnamefont{Katsnelson}},
485:   \bibinfo{author}{\bibfnamefont{K.~S.} \bibnamefont{Novoselov}},
486:   \bibinfo{author}{\bibfnamefont{T.~J.} \bibnamefont{Booth}}, \bibnamefont{and}
487:   \bibinfo{author}{\bibfnamefont{S.}~\bibnamefont{Roth}},
488:   \bibinfo{journal}{Nature} \textbf{\bibinfo{volume}{446}}, \bibinfo{pages}{60}
489:   (\bibinfo{year}{2007}).
490: 
491: \bibitem[{\citenamefont{Hohenberg and Kohn}(1964)}]{hohenberg65}
492: \bibinfo{author}{\bibfnamefont{P.}~\bibnamefont{Hohenberg}} \bibnamefont{and}
493:   \bibinfo{author}{\bibfnamefont{W.}~\bibnamefont{Kohn}},
494:   \bibinfo{journal}{Phys. Rev.} \textbf{\bibinfo{volume}{136}},
495:   \bibinfo{pages}{B864} (\bibinfo{year}{1964}).
496: 
497: \bibitem[{\citenamefont{Kohn and Sham}(1964)}]{kohn65}
498: \bibinfo{author}{\bibfnamefont{W.}~\bibnamefont{Kohn}} \bibnamefont{and}
499:   \bibinfo{author}{\bibfnamefont{L.~J.} \bibnamefont{Sham}},
500:   \bibinfo{journal}{Phys. Rev.} \textbf{\bibinfo{volume}{140}},
501:   \bibinfo{pages}{A1133} (\bibinfo{year}{1964}).
502: 
503: \bibitem[{gga()}]{gga}
504: \bibinfo{note}{Including gradient corrections to improve the description of the
505:   exchange and correlation functional (GGA) does not qualitatively change our
506:   conclusions; it merely modifies the fine details (GGA overestimates bonding
507:   distances by a similar amount as LDA underestimates them, while the barrier
508:   to the global minimum slightly increases to 0.21 eV). We have decided to use
509:   LDA because using a simpler formalism yields a simpler physical
510:   interpretation.}
511: 
512: \bibitem[{\citenamefont{Segall et~al.}(2002)\citenamefont{Segall, Lindan,
513:   Robert, Pickard, Hasnip, Clark, and Payne}}]{payne02}
514: \bibinfo{author}{\bibfnamefont{M.~D.} \bibnamefont{Segall}},
515:   \bibinfo{author}{\bibfnamefont{P.~L.~D.} \bibnamefont{Lindan}},
516:   \bibinfo{author}{\bibfnamefont{M.~J.} \bibnamefont{Robert}},
517:   \bibinfo{author}{\bibfnamefont{C.~J.} \bibnamefont{Pickard}},
518:   \bibinfo{author}{\bibfnamefont{P.~J.} \bibnamefont{Hasnip}},
519:   \bibinfo{author}{\bibfnamefont{S.~J.} \bibnamefont{Clark}}, \bibnamefont{and}
520:   \bibinfo{author}{\bibfnamefont{M.~C.} \bibnamefont{Payne}},
521:   \bibinfo{journal}{J. Phys.: Cond. Matt.} \textbf{\bibinfo{volume}{14}},
522:   \bibinfo{pages}{2717} (\bibinfo{year}{2002}).
523: 
524: \bibitem[{acc()}]{accelrys}
525: \bibinfo{note}{Materials Studio 4.1; http://www.accelrys.com}.
526: 
527: \bibitem[{pre()}]{precision}
528: \bibinfo{note}{Accuracy for total energies have been converged below 0.01 eV
529:   and for barriers below 0.001 eV by using energy cutoffs around 750 eV
530:   (norm-conserving pseudo-potentials) and a set of about 100 k-points sampling
531:   the Brillouin zone.}
532: 
533: \bibitem[{\citenamefont{Latil and Henrard}(2006)}]{latil06}
534: \bibinfo{author}{\bibfnamefont{S.}~\bibnamefont{Latil}} \bibnamefont{and}
535:   \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{Henrard}},
536:   \bibinfo{journal}{Phys. Rev. Lett.} \textbf{\bibinfo{volume}{97}},
537:   \bibinfo{pages}{036803} (\bibinfo{year}{2006}).
538: 
539: \bibitem[{\citenamefont{Jenkins and King}(2000)}]{jenkins00}
540: \bibinfo{author}{\bibfnamefont{S.~J.} \bibnamefont{Jenkins}} \bibnamefont{and}
541:   \bibinfo{author}{\bibfnamefont{D.~A.} \bibnamefont{King}},
542:   \bibinfo{journal}{Chem. Phys. Lett.} \textbf{\bibinfo{volume}{317}},
543:   \bibinfo{pages}{381} (\bibinfo{year}{2000}).
544: 
545: \bibitem[{\citenamefont{Pauling}(1974)}]{pauling}
546: \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{Pauling}},
547:   \emph{\bibinfo{title}{The nature of the chemical bond}}
548:   (\bibinfo{publisher}{Cornell Univ. Press}, \bibinfo{address}{Ithaca (NY)},
549:   \bibinfo{year}{1974}).
550: 
551: \bibitem[{\citenamefont{Allinger and Eaton}(1983)}]{allinger83}
552: \bibinfo{author}{\bibfnamefont{N.~L.} \bibnamefont{Allinger}} \bibnamefont{and}
553:   \bibinfo{author}{\bibfnamefont{P.~E.} \bibnamefont{Eaton}},
554:   \bibinfo{journal}{Tetrahedron Lett.} \textbf{\bibinfo{volume}{24}},
555:   \bibinfo{pages}{3697} (\bibinfo{year}{1983}).
556: 
557: \bibitem[{\citenamefont{Schmidt et~al.}(1993)\citenamefont{Schmidt, Bladrige,
558:   Boatz, Boatz, Elbert, Gordon, Jensen, Koseki, Matsunaga, Nguyen
559:   et~al.}}]{gamess}
560: \bibinfo{author}{\bibfnamefont{M.~W.} \bibnamefont{Schmidt}},
561:   \bibinfo{author}{\bibfnamefont{K.~K.} \bibnamefont{Bladrige}},
562:   \bibinfo{author}{\bibfnamefont{J.~A.} \bibnamefont{Boatz}},
563:   \bibinfo{author}{\bibfnamefont{J.~A.} \bibnamefont{Boatz}},
564:   \bibinfo{author}{\bibfnamefont{S.~T.} \bibnamefont{Elbert}},
565:   \bibinfo{author}{\bibfnamefont{M.~S.} \bibnamefont{Gordon}},
566:   \bibinfo{author}{\bibfnamefont{J.~H.} \bibnamefont{Jensen}},
567:   \bibinfo{author}{\bibfnamefont{S.}~\bibnamefont{Koseki}},
568:   \bibinfo{author}{\bibfnamefont{N.}~\bibnamefont{Matsunaga}},
569:   \bibinfo{author}{\bibfnamefont{K.~A.} \bibnamefont{Nguyen}},
570:   \bibnamefont{et~al.}, \bibinfo{journal}{Comput. Chem.}
571:   \textbf{\bibinfo{volume}{14}}, \bibinfo{pages}{1347} (\bibinfo{year}{1993}).
572: 
573: \bibitem[{\citenamefont{Refson et~al.}(2006)\citenamefont{Refson, Tulip, and
574:   Clark}}]{refson06}
575: \bibinfo{author}{\bibfnamefont{K.}~\bibnamefont{Refson}},
576:   \bibinfo{author}{\bibfnamefont{P.~R.} \bibnamefont{Tulip}}, \bibnamefont{and}
577:   \bibinfo{author}{\bibfnamefont{S.~J.} \bibnamefont{Clark}}, 
578:   \bibinfo{journal}{Phys. Rev. B} \textbf{\bibinfo{volume}{73}},
579:   \bibinfo{pages}{155114} (\bibinfo{year}{2006}).
580: 
581: \bibitem[{not()}]{nota}
582: \bibinfo{note}{Geometrical optimization of large fragments of the bi-layer (up
583:   to 320 C atoms) saturated by H atoms to preserve the C coordination at the
584:   molecule boundaries shows the stability of the new bonds against simultaneous
585:   atomic reconstructions and atomic density changes.}
586: 
587: \bibitem[{\citenamefont{Maultzsch et~al.}(2004)\citenamefont{Maultzsch, Reich,
588:   Thomsen, Requardt, and Ordejon}}]{maultzsch04}
589: \bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{Maultzsch}},
590:   \bibinfo{author}{\bibfnamefont{S.}~\bibnamefont{Reich}},
591:   \bibinfo{author}{\bibfnamefont{C.}~\bibnamefont{Thomsen}},
592:   \bibinfo{author}{\bibfnamefont{H.}~\bibnamefont{Requardt}}, \bibnamefont{and}
593:   \bibinfo{author}{\bibfnamefont{P.}~\bibnamefont{Ordejon}},
594:   \bibinfo{journal}{Phys. Rev. Lett.} \textbf{\bibinfo{volume}{92}},
595:   \bibinfo{pages}{075501} (\bibinfo{year}{2004}).
596: 
597: \end{thebibliography}
598: 
599: \end{document}
600: