Carbon 122 (2017) 281e286

Contents lists available at ScienceDirect

Carbon
journal homepage: www.elsevier.com/locate/carbon

Electronic properties of a graphene/periodic porous graphene

heterostructure
Junsu Lee a, Gunn Kim a, b, *
a
Department of Physics and Astronomy, Sejong University, Seoul 143-747, South Korea
b
Graphene Research Institute, Sejong University, Seoul 143-747, South Korea

a r t i c l e i n f o a b s t r a c t

Article history: Recently, two-dimensional van der Waals materials such as graphene and hexagonal boron nitride have
Received 6 February 2017 attracted interest as research topics. The two-dimensional material of polyphenylene superhoneycomb
Received in revised form network (PSN) is similarly interesting because it is a type of periodic porous graphene. In this paper, we
6 June 2017
report a first-principles study of the geometric and electronic properties of vertical heterostructures
Accepted 19 June 2017
Available online 22 June 2017
comprising graphene and PSN. AA, AA0 , and AB stacking configurations of a graphene sheet on a PSN
sheet produce band gaps of 63, 16, and 3 meV, respectively. We also determine the relationships between
the band gap and the interlayer distance between the graphene and PSN sheets. Finally, we present
Keywords:
Density functional theory
computationally simulated scanning tunneling microscopy images, which indicate the local electronic
Porous graphene structures of the surfaces of the graphene and PSN sides.
Vertical heterostructure © 2017 Elsevier Ltd. All rights reserved.

1. Introduction porous graphene. In addition, theoretical investigations have pre-

Graphene has attracted great interest because it possesses
unique characteristics such as high electrical conductivity and
outstanding mechanical properties [1e8]. The discovery of gra-
phene has significantly increased research on various two-
dimensional (2D) materials such as hexagonal boron nitride and
transition-metal dichalcogenides, and the field of 2D materials is
expanding to investigate heterostructures containing multiple 2D
materials. Recently, several van der Waals heterostructures were
prepared experimentally. These heterostructured materials were
found to have unique properties, and several new phenomena have
been observed within them [9e14]. Consequently, the need for
theoretical research to discover and understand the properties of
experimentally observed and theoretically simulated hetero-
structures has increased.
In 2009, Bieri et al. succeeded in synthesizing a polyphenylene
superhoneycomb network (PSN), a periodic porous graphene sheet
with a periodicity of 1 nm or less [15]. An attractive property of
porous graphene is its angstrom-scale periodicity. Regular, uniform,
and porous structures are necessary to increase the performance
and expand the applicability of porous carbon materials, including
* Corresponding author. Department of Physics and Astronomy, Sejong Univer-
sity, Seoul 143-747, South Korea.
E-mail address: gunnkim@sejong.ac.kr (G. Kim).
dicted that PSN can be used in hydrogen storage or refinement, and
that it has a semiconductor-like band gap [16,17].
In the present study, the structural and electronic characteristics
of vertical graphene/PSN (Gr/PSN) heterostructures, in which one
layer of PSN and one layer of graphene are held together by van der
Waals interaction, were investigated using ab initio calculations.
Like vertical graphene/hexagonal boron nitride heterostructures
[18e20], the vertical Gr/PSN heterostructures also exhibit asym-
metry in interlayer potential because PSN is a type of periodic
porous graphene sheet. Such interlayer potential asymmetry causes
band gap opening in graphene, which may improve the room-
temperature pinch-off characteristics of graphene-based field ef-
fect transistors [21]. In particular, we focus on the changes in the
electronic structure of graphene caused by interactions with PSN.
We consider three different stacking configurations for a hetero-
structure consisting of a graphene sheet and a PSN: AB, AA, and AA0 .
Regarding the band gap size, we find that a band gap in graphene is
induced by stacking with a PSN layer that is free of significant
atomic deformation or extrinsic defects. In addition, we obtain
computationally simulated scanning tunneling microscopy (STM)
images of the Gr/PSN heterostructures. The images show the
changes in the electronic structure of a PSN (graphene) layer caused
by that of a graphene (PSN) layer on each surface, as well as the
pore shapes exhibited on each side of the heterostructure for the
three stacking configurations.

http://dx.doi.org/10.1016/j.carbon.2017.06.049
0008-6223/© 2017 Elsevier Ltd. All rights reserved.
282 J. Lee, G. Kim / Carbon 122 (2017) 281e286

2. Computational details

Using density functional theory (DFT) [22], the geometric and

electronic characteristics of a PSN monolayer and Gr/PSN hetero-
structures were studied. We used plane-wave basis sets and the
projector augmented wave (PAW) pseudopotential implemented in
the Vienna Ab Initio Simulation Package (VASP) [23,24]. The cut-off
electron kinetic energy was set to 500 eV. The exchange-correlation
energy was modeled using the PerdeweBurkeeErnzerhof gener-
alized gradient approximation (GGA) functional [25]. Because GGA
functionals do not describe van der Waals interactions well, a
separate correction was necessary for systems where van der Waals
interactions were dominant. In this study, we used the Tkatch-
enkoeScheffler (TS) method [26] as a correction scheme. In order to
confirm that the TS method is sufficient to describe the van der
Waals heterostructures, we considered four additional van der
Waals correction methods: DFT-D2 [27], DFT-D3 [28], DFT-D3 with
the BeckeeJohnson damping function [DFT-D3 (BJ)] [29], and
optB86b-vdW [30,31] for comparison. In addition, we used
G-centered 10  10  1 and 30  30  1 k point grids for the
relaxation and electronic structure calculations, respectively, suit-
able for the hexagonal symmetry of the system. Lattice-constant
optimization was performed using the third-order
BircheMurnaghan equation of state [32,33]. Using the Ter-
soffeHamann method [34], STM images were obtained from the
integrated partial charge densities in a given energy window.

3. Results and discussion

We optimized the structure of one PSN layer and obtained the

minimum energy geometry and its corresponding electronic
structure. As shown in Fig. 1(a), the length of one side of a rhombus-
shaped primitive cell is 7.51 Å. This length, which is equal to the
distance between pores in the PSN, was determined experimentally
as approximately 7.4 Å [15]. In Fig. 1(a), Ca represents a carbon atom
bonded to one hydrogen and two carbon atoms, and Cb denotes a
carbon atom bonded to three neighboring carbon atoms. Our first-
principles calculations reveal that the CaeCb bond length is 1.40 Å,
and that the CaeH bond length is 1.09 Å. Meanwhile, the CbeCb
bond length is 1.49 Å. Fig. 1(b) shows the electronic structure of one
layer of the PSN structure. Interestingly, this structure has a band
gap of 2.40 eV at the K-point, which is in agreement with the
literature [16,17].
After the structural optimization of one PSN layer, the
minimum-force structure of the commensurate Gr/PSN hetero-
structure was obtained by considering van der Waals interactions.
As seen in Fig. 2(a), we considered three types of stacking struc-
tures: AB, AA, and AA0 . In AB stacking, the PSN layer is positioned on
top of the graphene layer but shifted, and a carbon atom in the
graphene layer sits at the center of each PSN hexagon. In AA
stacking, the C atoms in the graphene and PSN layers are perfectly
superimposed. In AA0 stacking, the AA-stacked PSN layer is shifted
by half a hexagon length relative to the graphene layer. The lattice
constant (a), interlayer distance (d), and interlayer binding energy
(Eint ) were determined preferentially. The interlayer binding energy
is defined by the following equation:
Eint ¼ E½Gr þ PSN  E½Gr  E½PSN: (1)
Here E½Gr þ PSN is the total energy of the Gr/PSN hetero-
structure. E½Gr and E½PSN are the total energies of the graphene
and PSN sheets, respectively. The results are shown in Fig. 2(b) and
Table 1. In calculating the interlayer binding energies, the TS and
optB86b-vdW methods tend to yield larger binding energies than
the other correction methods do. All methods retain the order of
Fig. 1. Atomic and electronic structure of PSN monolayer, calculated by TS method. (a)
Optimized atomic geometry by a 2  2 supercell to indicate the structure of the pore at
the center of the figure. The atom pairs Ca (black) and Cb (red) belong to different sub-
lattices inside each hexagonal carbon ring. The numbers show the distance between
atoms (Å). (b) Band structure (left) and pz orbital-projected local density of states
(right) for six Ca atoms and six Cb atoms. The Dirac points are shown at 2.45 and 2:13
eV at the K-point. (c) Simulated STM images in the constant-current mode at the bias
voltages of 3:5 (left) and 3.5 V (right). The Fermi energies corresponding to the zero-
energy level are all set to the center between the conduction-band minimum and the
valence-band maximum at the K-point in the band structure. (A colour version of this
figure can be viewed online.)
stabilities such that AB (AA) stacking is the most (least) stable.
In comparing the interlayer binding energies of the three types
of stacking structures, the AB structure is found to be the most
stable. The AA structure is 3.3 meV/atom higher in binding energy
per atom than the AB structure, and the AA0 structure is 0.6 meV/
atom higher in energy per atom than the AB. Therefore, AB, AA0 , and
AA are more stable in this order, which is also the order of stabilities
in two-layer graphene [35]. The bond lengths between the atoms in
PSN differ from those of pure PSN, since the periodicity of PSN is
matched to that of graphene in this study. The CaeCb bond length
inside the hexagonal carbon ring is 1.39 Å, the CbeCb bond length
connecting two hexagonal carbon rings is 1.47 Å, the CaeH bond
length is 1.08 Å, and the bond length between two carbons in the
graphene is 1.43 Å, throughout all stacking configurations. When
compared to the structure containing only PSN, the CaeCb and the
CaeH bond lengths are all decreased by 0.01 Å, while compared to
the structure of only graphene, the bond length between two car-
bons is increased by 0.01 Å. The distance between graphene and
PSN is 3.36 Å; this is closest to the value observed in AB. The
interlayer distances in AA and AA0 are 3.41 and 3.39 Å, respectively.
 J. Lee, G. Kim / Carbon 122 (2017) 281e286
Fig. 2. Atomic and electronic structure of Gr/PSN heterostructure. (a) Model structures
of three stacking configurations and transition state (Fig. 2(c)), calculated by TS
method. Calculations were performed using primitive cells. The 2  2 expanded figure
of the primitive cell is included to show the relationship between graphene and the
PSN pores. In addition, d is the distance between the graphene and PSN layers. (b)
Interlayer binding energy (Eint )-d curves, calculated by various van der Waals correc-
tion methods. The interlayer binding energy is the lowest in the following order for all
methods that we considered: AB, AA0 , and AA. (c) Total energies along the path of
sliding of the graphene on the PSN sheet, as calculated by the TS method. (A colour
version of this figure can be viewed online.)
283
Finally, we determined that the separation between the layers was
smaller when the interlayer binding energy was lower, as shown in
Fig. 2(b) and Table 1.
Using ab initio calculations, we also obtained the energy profile,
considering the lateral translation of the graphene sheet on the PSN
layer. As shown in Fig. 2(c), the translation path is from AB to AA via
AA0 stacking. We observed the transition state at the activation
energy barrier peak between the AB and AA0 stacking configura-
tions in Fig. 2(a). The activation energy is calculated as 29 meV/cell
(¼ 0.81 meV/atom). Compared to the translational activation en-
ergy of the graphene bilayer case (1.3 meV/atom) [36], the activa-
tion energy of the Gr/PSN heterostructure is lower.
Fig. 3(a) shows the band structures of the AB, AA, and AA0
stacking configurations in their equilibrium structures. Interest-
ingly, the AB structure, which is the most stable regarding total
energy, has the smallest band gap (3 meV) of the graphene layers.
AA0 and AA have the band gaps of 16 and 63 meV, respectively. The
side views of partial charge densities of the three stacking types of
Gr/PSN heterostructures are shown in Fig. 3(b). Their cross-
sectional views show that the p-states near the Fermi level in the
AA and AA0 stacking configurations are more changed than those in
AB stacking. Consequently, the AA and AA0 stacking configurations
have somewhat larger gaps than the AB stacking configuration in
the Gr/PSN heterostructure. As shown in the insets of Fig. 3, the AA
and AB stacking configurations maintain double degeneracies for
the valence and conduction bands. In contrast, AA0 stacking breaks
the degeneracy; the energy splitting size is 18e19 meV. This result
indicates that the pore in PSN is responsible for the formation of the
band gap in the graphene layer. We can discuss the origin of de-
generacy breaking using the tight-binding approximation. For the
AA stacking configuration, the interlayer hopping parameters be-
tween graphene and PSN are almost identical, except in the pore
region. On the other hand, the interlayer hopping parameters are
two for the AB stacking configuration, except in the pore region. In
contrast, several different carbon sites in graphene appear for the
interlayer hopping in AA0 stacking. This means that the Hamilto-
nian matrix for AA0 stacking is more complicated than those for the
AA and AB stacking configurations, causing the splitting of energy
bands near the Fermi level.
Next, we investigated the influence of an external electric field
on the vertical Gr/PSN heterostructures. For this calculation, an
electric field was applied in the direction of the normal vector of
graphene and PSN to investigate changes in the electronic struc-
ture. The energy bands of PSN, which has a large band gap, clearly
shift when an electric field is applied, whereas the energy band of
graphene does not show appreciable change. When the intensity of
the electric field is 0.1 V/Å, the band gap changes are  3 meV for
the three stacking configurations.
We also obtained the band gap Eg of the graphene, depending
on the interlayer distance d. As shown in Fig. 3(c), the band gap is
not sensitive to the interlayer distance for AB stacking. The band
gap increases exponentially as the interlayer distance decreases for
the AA and AA0 stacking configurations. AA stacking shows a
stronger influence on the band gap by the interlayer distance than
AA0 stacking. The curve fitting is achieved using an exponential
function form,
h i
Eg ¼ aexp  bðd þ gÞd : (2)
For AA stacking, a, b, g, and d are determined to be 5.34050 eV,
4.78277 Å1, 2:61618 Å, and 0.414853, respectively; for AA0
stacking, a, b, g, and d are 0.632699 eV, 4.21244 Å1, 2:74993 Å,
and 0.301891, respectively (see Table 2).
Finally, we provide computationally simulated STM images in
Fig. 4. On the graphene side, the STM images exhibit hexagonal and
284 J. Lee, G. Kim / Carbon 122 (2017) 281e286

Table 1
Optimized lattice constants (a), interlayer distances (d), and interlayer binding energies (Eint ), calculated by five van der Waals correction methods for AB-, AA-, and AA0 -
stacked Gr/PSN heterostructures.

a (Å) d (Å) Eint (meV/atom)

DFT-D2 AB 7.433 3.25 17:1

AA 7.433 3.40 13:8
AA0 7.433 3.27 16:6
DFT-D3 AB 7.433 3.44 16:4
AA 7.433 3.56 14:8
AA0 7.434 3.49 16:2
DFT-D3 (BJ) AB 7.430 3.38 17:9
AA 7.430 3.51 15:6
AA0 7.430 3.39 17:6
DFT-TS AB 7.424 3.36 26:4
AA 7.423 3.41 23:1
AA0 7.424 3.39 25:8
optB86b-vdW AB 7.433 3.36 25:2
AA 7.432 3.41 22:3
AA0 7.433 3.39 24:7

Fig. 3. (a) The band structures of AB-, AA-, and AA0 -stacked Gr/PSN heterostructures in their equilibrium states, respectively. The band gaps are 3 (AB), 63 (AA), and 16 (AA0 ) meV,
respectively. The Fermi energy (EF ) is set to zero. Numbers in the insets of (a) represent the degeneracies of the two bands. (b) Partial charge densities that are integrated in the
energy range between EF  0:5 eV and EF with the isosurface value of 5  106 e=a30 for the three stacking configurations. Blue color denotes cross sections of the charge densities at
the boundary of the unit cells. (c) Band gaps (Eg ) as a function of interlayer distance d for AB-, AA-, and AA0 -stacked Gr/PSN heterostructures, calculated by the TS method. (A colour
version of this figure can be viewed online.)

Table 2 extrinsic impurity atom as a defect [37]. The STM images exhibit
Fitting parameters a, b, g, and d in a function in Eq. (2). common trigonal patterns at bias voltages of 0:5 and 0.5 V for AB
a (eV) b (Å1) g (Å) d stacking, but bright-spot inversion occurs between the two
AA 5.34050 4.78277 2:61618 0.414853
different carbon sub-lattices in the graphene. In addition, the pore
AA0 0.632699 4.21244 2:74993 0.301891 sites are nearly indistinguishable in this case. The STM images for
AA and AA0 stacking configurations are very similar: flower-like and
half-turned hexagonal patterns appear at 0:5 and 0.5 V, respec-
tively. In contrast to AB stacking, the pores are easily distinguished
trigonal patterns at bias voltages of 2:0 and 2.0 V in all stacking by the darker pore shapes in these cases.
configurations, respectively. Differences in the stacking types are On the PSN side, meanwhile, the STM images exhibit adjoining
revealed at the pore sites, which have unique pore shapes. At 2:0 triangles and full hexagonal patterns at bias voltages of 2:0 and
V, the graphene features darker Y-, hexagonal, and H-shapes 2.0 V in all stacking configurations, respectively. In this case, dark
(marked with yellow-colored eye guides), and at 2.0 V, it also fea- empty and bright full spots are reflected at the pore for all three
tures brighter triangular, dot, and rhombic shapes (marked with stacking types. At 2:0 V, only Ca atoms (Fig. 1) have bright spots.
blue-colored eye guides) in the following order: AB, AA, and AA0 More interesting points are presented at bias voltages of 0:5 and
stacking. Therefore, the different pore shapes reflect different 0.5 V; we observe constructive and destructive interferences be-
defect structures, similar to the findings from an earlier study on tween atoms in the graphene and PSN layers. In the case of AB
the graphene/hexagonal boron nitride heterostructure with an stacking, at 0:5 V, only Ca atoms of PSN above the centers of the
 J. Lee, G. Kim / Carbon 122 (2017) 281e286 285

Fig. 4. Computationally simulated constant-height (3 Å) mode STM images for (a) AB-, (b) AA-, and (c) AA0 -stacked Gr/PSN heterostructures on graphene and PSN sides. The applied
bias voltages are ±2:0 (left panels) and ±0:5 V (right panels) for each stacking configuration. The solid red hexagons represent the pore positions, composed of six hydrogen atoms
each. (A colour version of this figure can be viewed online.)

graphene hexagons show bright spots, compared to the case of a
bias voltage of 2:0 V. As shown in Fig. 1(b), the Ca sites have larger
pz orbitals than the Cb sites near the valence and conduction band
edges. This trend is maintained even near the Fermi level. At a plane
3e4 Å from the graphene sheet, the electron density around the
hollow site is higher than that around the C atom site because of the
constructive sum of p-bonding states originating from the hy-
bridization of pz orbitals of six C atoms in graphene. In the AB-
stacked Gr/PSN heterostructure, graphene-derived states are
mainly projected to the only Ca atoms above the centers of the
graphene hexagons near the Fermi level. Therefore, the half of the
PSN appears masked in the STM image at 0:5 V. In AA0 stacking, at
0:5 V, only four Ca atoms among the six Ca atoms have bright
spots, and at 0.5 V, because of the two Ca atoms among the six Ca
atoms that have bright spots, the patterns look like vertical stripes.
When a Gr/PSN heterostructure is formed, graphene-derived states
occur in the PSN layer near the Fermi level, and they are dependent
on the stacking configuration. For the biases of ±0:5 V, the above-
mentioned graphene-derived PSN states are shown. In fact, their
intensities are much smaller than those of the intrinsic PSN states,
which can be detected at the biases of about ±2:0 V. Thus, the
corresponding STM images are more sensitive to the stacking
configuration at ±0:5 V than ±2:0 V.
4. Conclusion
Using density functional theory with van der Waals corrections,
we have investigated the geometric and electronic properties of Gr/
PSN heterostructures. Our calculated results show that the band
gap of a free-standing PSN sheet is 2.40 eV, which is in agreement
with previous reports. For the Gr/PSN heterostructure, interlayer
interaction induces a band gap of 63 meV (16 meV) in the AA (AA0 )
stacking configuration. We apply an external electric field of
0.1e0.5 V/Å to the Gr/PSN heterostructure in the direction normal
to the graphene; however, no noticeable changes occur in the
electronic structure characteristics, such as the band gap. Because
286 J. Lee, G. Kim / Carbon 122 (2017) 281e286
there are no electronic states in PSN near the Fermi level, very little
interaction occurs between the two layers near the Fermi level.
Both hexagonal boron nitride [21] and PSN should be good choices
for substrates that induce a band gap in graphene in our quest to
find efficient 2D materials.
Acknowledgments
JL and GK acknowledge financial support of the Priority
Research Center Program (Grant No. 2010-0020207) and the Mid-
career Researcher Program (Grant No. 2016R1A2B2016120)
through the National Research Foundation funded by the Korea
government.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.carbon.2017.06.049.
References
[1] K.S. Novoselov, A.K. Geim, S.V. Morozov, D. Jiang, Y. Zhang, S.V. Dubonos, et al.,
Electric field effect in atomically thin carbon films, Science 306 (5696) (2004)
666e669.
[2] K.S. Novoselov, A.K. Geim, S.V. Morozov, D. Jiang, M.I. Katsnelson,
I.V. Grigorieva, et al., Two-dimensional gas of massless Dirac fermions in
graphene, Nature 438 (7065) (2005) 197e200.
[3] Y. Zhang, Y.-W. Tan, H.L. Stormer, P. Kim, Experimental observation of the
quantum Hall effect and Berry's phase in graphene, Nature 438 (7065) (2005)
201e204.
[4] M.I. Katsnelson, K.S. Novoselov, A.K. Geim, Chiral tunnelling and the Klein
paradox in graphene, Nat. Phys. 2 (9) (2006) 620e625.
[5] A.F. Young, P. Kim, Quantum interference and Klein tunnelling in graphene
heterojunctions, Nat. Phys. 5 (3) (2009) 222e226.
[6] V.V. Cheianov, V. Fal’ko, B.L. Altshuler, The focusing of electron flow and a
Veselago lens in graphene p-n junctions, Science 315 (5816) (2007)
1252e1255.
[7] A. Rycerz, J. Tworzydło, C.W.J. Beenakker, Valley filter and valley valve in
graphene, Nat. Phys. 3 (3) (2007) 172e175.
[8] C. Park, H. Yang, A.J. Mayne, G. Dujardin, S. Seo, Y. Kuk, et al., Formation of
unconventional standing waves at graphene edges by valley mixing and
pseudospin rotation, Proc. Natl. Acad. Sci. U. S. A. 108 (46) (2011)
18622e18625.
[9] L.A. Ponomarenko, A.K. Geim, A.A. Zhukov, R. Jalil, S.V. Morozov,
K.S. Novoselov, et al., Tunable metal-insulator transition in double-layer gra-
phene heterostructures, Nat. Phys. 7 (12) (2011) 958e961.
[10] L. Britnell, R.V. Gorbachev, R. Jalil, B.D. Belle, F. Schedin, A. Mishchenko, et al.,
Field-effect tunneling transistor based on vertical graphene heterostructures,
Science 335 (6071) (2012) 947e950.
[11] S.J. Haigh, A. Gholinia, R. Jalil, S. Romani, L. Britnell, D.C. Elias, et al., Cross-
sectional imaging of individual layers and buried interfaces of graphene-based
heterostructures and superlattices, Nat. Mater. 11 (9) (2012) 764e767.
[12] C. Dean, A.F. Young, L. Wang, I. Meric, G.-H. Lee, K. Watanabe, et al., Graphene
based heterostructures, Solid State Commun. 152 (15) (2012) 1275e1282.
[13] R.V. Gorbachev, A.K. Geim, M.I. Katsnelson, K.S. Novoselov, T. Tudorovskiy,
I.V. Grigorieva, et al., Strong Coulomb drag and broken symmetry in double-
layer graphene, Nat. Phys. 8 (12) (2012) 896e901.
[14] T. Georgiou, R. Jalil, B.D. Belle, L. Britnell, R.V. Gorbachev, S.V. Morozov, et al.,
Vertical field-effect transistor based on graphene-WS2 heterostructures for
flexible and transparent electronics, Nat. Nanotechnol. 8 (2) (2013) 100e103.
[15] M. Bieri, M. Treier, J. Cai, K. Aït-Mansour, P. Ruffieux, O. Gro €ning, et al., Porous
graphenes: two-dimensional polymer synthesis with atomic precision, Chem.
Commun. 45 (2009) 6919e6921.
[16] Y. Li, Z. Zhou, P. Shen, Z. Chen, Two-dimensional polyphenylene: experi-
mentally available porous graphene as a hydrogen purification membrane,
Chem. Commun. 46 (21) (2010) 3672e3674.
[17] A. Du, Z. Zhu, S.C. Smith, Multifunctional porous graphene for nanoelectronics
and hydrogen storage: new properties revealed by first principle calculations,
J. Am. Chem. Soc. 132 (9) (2010) 2876e2877.
[18] S. Park, C. Park, G. Kim, Interlayer coupling enhancement in graphene/hex-
agonal boron nitride heterostructures by intercalated defects or vacancies,
J. Chem. Phys. 140 (13) (2014) 134706.
[19] B. Hwang, J. Hwang, J.K. Yoon, S. Lim, S. Kim, M. Lee, et al., Energy bandgap and
edge states in an epitaxially grown graphene/h-BN heterostructure, Sci. Rep. 6
(2016) 31160.
[20] B. Hunt, J.D. Sanchez-Yamagishi, A.F. Young, M. Yankowitz, B.J. LeRoy,
K. Watanabe, et al., Massive Dirac fermions and Hofstadter butterfly in a van
der Waals heterostructure, Science 340 (6139) (2013) 1427e1430.
[21] G. Giovannetti, P.A. Khomyakov, G. Brocks, P.J. Kelly, J. van den Brink, Sub-
strate-induced band gap in graphene on hexagonal boron nitride: ab initio
density functional calculations, Phys. Rev. B 76 (7) (2007) 073103.
[22] P. Hohenberg, W. Kohn, Inhomogeneous electron gas, Phys. Rev. 136 (3B)
(1964) B864eB871.
[23] G. Kresse, J. Furthmüller, Efficient iterative schemes for ab initio total-energy
calculations using a plane-wave basis set, Phys. Rev. B 54 (16) (1996)
11169e11186.
[24] G. Kresse, D. Joubert, From ultrasoft pseudopotentials to the projector
augmented-wave method, Phys. Rev. B 59 (3) (1999) 1758e1775.
[25] J.P. Perdew, K. Burke, M. Ernzerhof, Generalized gradient approximation made
simple, Phys. Rev. Lett. 77 (18) (1996) 3865e3868.
[26] A. Tkatchenko, M. Scheffler, Accurate molecular van der Waals interactions
from ground-state electron density and free-atom reference data, Phys. Rev.
Lett. 102 (7) (2009) 073005.
[27] S. Grimme, Semiempirical GGA-type density functional constructed with a
long-range dispersion correction, J. Comp. Chem. 27 (15) (2006) 1787e1799.
[28] S. Grimme, J. Antony, S. Ehrlich, H. Krieg, A consistent and accurate ab initio
parametrization of density functional dispersion correction (DFT-D) for the 94
elements H-Pu, J. Chem. Phys. 132 (15) (2010) 154104.
[29] S. Grimme, S. Ehrlich, L. Goerigk, Effect of the damping function in dispersion
corrected density functional theory, J. Comp. Chem. 32 (7) (2011) 1456e1465.
[30] J. Klimes, D.R. Bowler, A. Michaelides, Chemical accuracy for the van der Waals
density functional, J. Phys. Condens. Matter 22 (2) (2010) 022201.
[31] J. Klimes, D.R. Bowler, A. Michaelides, Van der Waals density functionals
applied to solids, Phys. Rev. B 83 (19) (2011) 195131.
[32] F. Birch, Finite elastic strain of cubic crystals, Phys. Rev. 71 (11) (1947)
809e824.
[33] F.D. Murnaghan, The compressibility of media under extreme pressures, Proc.
Natl. Acad. Sci. U.S.A. 30 (9) (1944) 244e247.
[34] J. Tersoff, D.R. Hamann, Theory of the scanning tunneling microscope, Phys.
Rev. B 31 (2) (1985) 805e813.
[35] Y.-W. Son, S.-M. Choi, Y. Hong, S. Woo, S.-H. Jhi, Electronic topological tran-
sition in sliding bilayer graphene, Phys. Rev. B 84 (15) (2011) 155410.
[36] C. Park, J. Ryou, S. Hong, B.G. Sumpter, G. Kim, M. Yoon, Electronic properties
of bilayer graphene strongly coupled to interlayer stacking and an external
electric field, Phys. Rev. Lett. 115 (1) (2015) 015502.
[37] J. Lee, M. Kim, J.R. Chelikowsky, G. Kim, Computational simulation of
subatomic-resolution AFM and STM images for graphene/hexagonal boron
nitride heterostructures with intercalated defects, Phys. Rev. B 94 (3) (2016)
035447.