Journal of Molecular Modeling (2017) 23:355

https://doi.org/10.1007/s00894-017-3521-7

ORIGINAL PAPER

Effects of shape, size, and pyrene doping on electronic properties

of graphene nanoflakes
Thanawit Kuamit 1 & Manussada Ratanasak 1 & Chompoonut Rungnim 2 & Vudhichai Parasuk 1

Received: 16 May 2017 / Accepted: 6 November 2017

# Springer-Verlag GmbH Germany, part of Springer Nature 2017

Abstract
Effects of size, shape, and pyrene doping on electronic properties of graphene nanoflakes (GNFs) were theoretically investigated
using density functional theory method with PBE, B3PW91, and M06-2X functionals and cc-pVDZ basis set. Two shapes of
zigzag GNFs, hexagonal (HGN) and rhomboidal (RGN), were considered. The energy band gap of GNF depends on shape and
decreases with size. The HGN has larger band gap energy (1.23–3.96 eV) than the RGN (0.13–2.12 eV). The doping of pyrene
and pyrene derivatives on both HGN and RGN was also studied. The adsorption energy of pyrene and pyrene derivatives on GNF
does not depend on the shape of GNFs with energies between 21 and 27 kcal mol−1. The substituent on pyrene enhances the
binding to GNF but the strength does not depend on electron withdrawing or donating capability. The doping by pyrene and
pyrene derivatives also shifts the HOMO and LUMO energies of GNFs. Both positive (destabilizing) and negative (stabilizing)
shifts on HOMO and LUMO of GNFs were seen. The direction and magnitude of the shift do not follow the electron withdrawing
and donating capability of pyrene substituents. However, only a slight shift was observed for doped RGN. A shift of 0.19 eV was
noticed for HOMO of HGN doped with 1-aminopyrene (pyNH2) and of 0.04 eV for LUMO of HGN doped with 1-
pyrenecarboxylic acid (pyCOOH).

Keywords Pristine graphene . Pyrene doping . Diode . Graphene nanoflake . DFT calculations

Introduction [8–11]. Graphene can be potentially used for logic devices

Graphene is a crystalline allotrope of carbon with two-
dimensional properties that can be described as a one-atom
thick layer of graphite [1]. Graphene has good conductivities,
good thermal properties, extremely high surface area, light
weight, etc. [2–7]. Moreover, graphene is also known as a
zero-band gap semiconductor. Thus, it has been interesting
to large numbers of researchers for its wide applicability
Electronic supplementary material The online version of this article
(https://doi.org/10.1007/s00894-017-3521-7) contains supplementary
material, which is available to authorized users.
* Vudhichai Parasuk
vudhichai.p@chula.ac.th
1
Center of Excellence in Computational Chemistry, Department of
Chemistry, Faculty of Science, Chulalongkorn University,
Bangkok 10330, Thailand
2
National Nanotechnology Center National Science and Technology
Development Agency, 130 Thailand Science Park, Paholyothin Rd,
Klong Luang, Pathumthani 12120, Thailand
by inducing a band gap between its π and π* states. There
are several methods for introducing band gap to graphene.
Traditionally, this could be achieved by doping graphene with
boron or nitrogen atoms. The atomic doping technique can
induce the band gap and make p- or n-type graphene
[12–16]. However, the atomic doping is difficult to control
and often destroys the band structure of graphene. Another
technique for band gap modification is by using chemisorbed
adsorbates, the chemical doping method [17–20]. The chem-
ical doping changes the hybridization of carbon on graphene
from sp2 to sp3 and hence alters the band structure. However,
this technique causes an undesired adjustment or disrupting of
the basic electronic properties of graphene. Another method
for the electronic-property modification of graphene is by
using physisorbed adsorbates, the physisorption technique.
The advantage of the physisorption technique is that the
unique electronic property of graphene can be protected, since
the physisorption just slightly changes the band structure
[21–29]. Lee et al. [30] modified electronic properties of re-
duced graphene oxide (rGO) by using noncovalent
355 Page 2 of 9
interactions with aromatic molecules through π-π interaction.
Because pyrene could stably bind to the graphene through
strong π-π interactions [31–34] and electron withdrawing or
donating substituents can be attached to pyrene with simple
reactions and facile purification [35, 36]. They then doped
rGO with pyrene and pyrene derivatives and successfully de-
vised a bottom-gated field-effect transistor. Several theoretical
studies have been carried out regarding the properties of
physisorbed graphene. Deng et al. [37] using density function-
al theory (DFT) suggested that the computed band gap energy
of fully H-terminated armchair- and zigzag-edge graphene
decreases with increasing size of nanosheet. Zhou and
Zhang [38] theoretically investigated the adsorption behavior
of a series of benzene derivatives physisorbed on graphene.
They suggested that adsorption energies were insensitive to
the adsorption sites (hollow, bridging, on-top), since energy
differences between these sites are within 1 kcal mol−1.
Moreover, the adsorption energies for all the derivatives are
larger than that of benzene. This can be ascribed to the increas-
ing orbital overlap of substituents with graphene.
A graphene nanoflakes (GNF) is a finite sized
graphene. It can be prepared by both bottom up [39] and
top down [40–48] approaches. Whilst it has properties
similar to graphene, it can be made into many sizes and
shapes. Nagai et al. [49] have studied electrical properties
of functionalized GNFs with various shapes and sizes and
found that electrical properties of GNFs can be tuned by
making them the proper size and shape. Shakourian-Fard
et al. [50] used M06-2X/cc-pVDZ level of theory to study
the effect of defect types on electronic and optical prop-
erties of GNFs physisorbed by ionic liquids (ILs). They
reported that the increasing size of the defective GNF
affects the geometry of the surface and enhances the bind-
ing affinity of ILs by about 10%. Chutia et al. [51] inves-
tigated the properties of GNFs by modifying their surface
chemistry and stacking conformation. They have shown
that there is a remarkable relationship between the chem-
ical hardness with electrical conductivity of GNFs. They
also revealed that HOMO-LUMO energy gap can directly
be related with the chemical hardness and hence the elec-
trical conductivity. Like graphene, electronic properties of
GNF can also be modified by doping with π-conjugated
compound.
In this work, geometries and electronic properties of
two shapes of graphene nanoflakes, rhomboidal (RGN)
and hexagonal (HGN), at various sizes were investigated
using DFT calculations. Moreover, the physisorption on
HGN and RGN by pyrene and its derivatives with various
electron withdrawing/donating substituents and their ef-
fects on electronic properties of GNF were discussed.
This research aimed to understand the fundamental that
governs electronic properties of GNF and graphene re-
garding shape and size and especially pyrene doping.
J Mol Model (2017) 23:355
The knowledge could provide a new insight into the de-
sign of a complete circuit of donor- and acceptor-type
graphene diodes.
Computational models and methods
Two different shapes of zigzag-edge GNFs, HGN and
RGN, were selected for the study. The hexagonal carbon
ring is the core structure for the construction of HGN,
which expands along the C2 symmetry axis as C6 exten-
sion within the D6h point group. Thus, it expands along
the C2’ axis within D6h symmetry as zigzag circles around
the nanoflake. The total number of C atoms in HGN is
6n2, where n is the peripheral zigzag circle of HGN. Our
HGN models were generated for n = 2–6, which contains
C24H12, C54H18, C96H24, C150H30, and C216H36, respec-
tively. For RGN, the m x m model was used to generate
GNFs, where m is the number of benzene rings. Our RGN
models were produced for m = 3–8, which are comprised
of C 3 0 H 1 4 C 4 8 H 1 8 , C 7 0 H 2 2 , C 9 6 H 2 6 , C 1 2 6 H 3 0 , and
C160H34, respectively. All molecular structures were fully
optimized by DFT method [52] using Perdew Burke
Ernzerhof (PBE) functional [53] and cc-pVDZ basis set
[54]. Geometry optimizations were performed in gas
phase using the Gaussian 09 software [55]. Illustrations
of the HGNs and RGNs being investigated in this study
are given in Fig. 1.
To address the dispersion-type interaction for the study
of physisorption, we chose the Minnesota M06-2X func-
tional [56] with cc-pVDZ basis set in cooperation with
BSSE correction since from our previous study we found
that the method could yield a reasonable description for
the dispersion interaction (π-π stacking) [57]. Two differ-
ent adsorption sites, on-top and hollow, were investigated
for both doped HGN and RGN (Fig. 2). We chose the C96
model to study the physisorption because in this model
both HGN and RGN have the same number of carbon
atoms. Doped GNF structures at the two positions were
obtained by fully optimizing the structures in Fig. 2 at
BSSE corrected M06-2X/cm3-pVDZ level of theory.
Energies of HOMO and LUMO and band gap energies (Eg)
of HGNs, RGNs, and their physisorbed complexes were de-
termined using the DFT method with B3PW91 hybrid func-
tional [58, 59] and cc-pVDZ basis set. The B3PW91 hybrid
functional has been shown to provide excellent prediction for
band gap energies of 70 insulating compounds [60]. Band gap
energies of both HGNs (C24H12, C54H18, C96H24, C150H30,
C216H36) and RGNs (C30H14, C48H18, C70H22, C96H26,
C126H30, C160H34) were estimated by:
Eg ¼ ELUMO  EHOMO ð2Þ
where E LUMO is the energy of the lowest unoccupied
J Mol Model (2017) 23:355
Fig. 1 Hexagonal graphene nanoflakes (a) and rhomboidal graphene nanoflakes (b)
molecular orbital and EHOMO is the energy of the highest
occupied molecular orbital.
Results and discussion
Size and shape dependent
HOMO and LUMO energies of five HGNs, C24H12, C54H18,
C96H 24, C150H30, and C216H 36 and six RGNs, C30H 14,
C48H18, C70H22, C96H26, C126H30, and C160H34 are displayed
in Fig. 3. The relation between band gap energies and numbers
of carbon atoms are given in Fig. S1. Corresponding band gap
energies for HGNs are 3.96, 2.76, 2.06, 1.58, and 1.23 eV,
respectively and for RGNs are 2.12, 1.16, 0.62, 0.34, 0.20,
and 0.13 eV, respectively. It can be seen that HOMO and
LUMO energies and hence band gap energies depend on
shape and size of GNFs. The decreasing of band gap energy
with the increasing of size was observed for both HGN and
RGN consistent with the result of Deng et al. [37]. Thus, the
observation could be explained by the increment of free π
electrons in occupied orbitals perpendicular to the GNF plane
(pz orbital). RGN generally has a band gap smaller than HGN.
Fig. 2 Pyrene doped HGN (a) and RGN (b) at on-top and hollow adsorbed positions
Page 3 of 9 355
All HGNs being studied have large band gap. They could be
classified as semiconductors. From the relation between the
number of carbon atoms on HGN and its band gap energy,
HGN will have zero band gap when the number of carbon
atoms is beyond 480 (see Fig. S1). Thus, very large HGN will
have electronic properties similar to graphene. However, mod-
erately large RGNs (C > 200) have negligible band gap energy
and they show the properties of metals. Though, for small
RGNs there remains a sizable band gap and they can be made
into semiconductors.
To understand why HOMO and LUMO energies of GNFs
depend on shape, we compared HOMO and LUMO orbitals
of C96H24 (HGN) and C96H26 (RGN), Fig. S2 and S3. HOMO
and LUMO orbitals of HGN look very distinct; however,
HOMO and LUMO orbitals of RGN appear to be similar.
Thus, while HGN has large band gap, HOMO and LUMO
of RGN have similar energy and hence negligible band gap.
Physisorption
Adsorption energies (Ead1) and BSSE corrected adsorption
energies (Ead2) of pyrene and seven pyrene derivatives
adsorbed on HGN and RGN were given in Table 1. The
355 Page 4 of 9 J Mol Model (2017) 23:355

mol−1 were observed. So, in general there is no distinct pref-

erential adsorption site.
From Table 1, strong bindings between HGN/RGN and
adsorbates were observed. Although, the BSSE is reasonably
large (4–7 kcal mol−1), it is still an order of magnitude smaller
than the binding affinity. The BSSE corrected adsorption en-
ergies for all adsorbates are between 21 and 27 kcal mol−1.
Adsorption energies of pyrene/pyrene derivatives on HGN
and RGN are similar. The binding of GNF with pyrene/
pyrene derivatives is 2–3 times stronger than that between
nucleobase and graphene (9–14 kcal mol−1) [57]. This is prob-
ably because pyrene has more benzene rings than the
nucleobase. This binding affinity between π-conjugated com-
pound and graphene is, however, one order of magnitude
smaller than that between ion and graphene [61]. This sug-
Fig. 3 HOMO, LUMO energies (eV) of HGNs and RGNs with various gests the different nature of the interaction with graphene for
sizes
the two systems. Intriguingly, all pyrene derivatives have
stronger binding with GNF than the unsubstituted one in
preferred adsorption site for each physisorbed complex was
agreement with Zhou and Zhang [38]. Thus, it could be ratio-
denoted in parenthesis, where t refers to on-top site and h to
nalized that the substituent on pyrene allows greater orbital
hollow site. Optimized structures of adsorption complexes of
overlap between the GNF and the adsorbate. Computed ad-
HGN and RGN with pyrene/seven pyrene derivatives
sorption energies do not seem to depend on the electron
were displayed in Figs. 4 and 5, respectively. For most adsorp-
donating/withdrawing capability of the substituents.
tion complexes, the energy difference between the complex at
However, they are more or less contingent to the size of sub-
the hollow and the on-top sites is < 1 kcal mol−1 (see Table S1
stituent or rather to the surface area, Fig. S4. The py-
and S2) in agreement with Zhao and Zhang [38]. However, for
NHCOCH3 has the largest surface area and thus the strongest
RGN-pyNH2, HGN-pyNH2, HGN-pyCH3, and HGN-pyNO2
binding to GNF. This again agrees with Zhou and Zhang’s
the respective variances of 5.65, 1.88, 1.88, and 1.26 kcal
observation [38], while disagreeing with the Hunter–Sanders
rules (electron donating substituents weaken binding whereas
electron withdrawing substituents enhance binding) [62, 63].
Table 1 Adsorption energies (Ead1) and BSSE corrected adsorption
energies (Ead2) of pyrene and seven derivatives adsorbed on HGN and We cannot find any relation between the adsorbate-GNF dis-
RGN in kcal mol−1 together with the distance between pyrene/pyrene tance (rPG) and the adsorption energy. The rPG between 3.0
derivatives and GNF (rPG) in Å and remarks of adsorption sites in paren- and 3.2 Å was reported for HGN and RGN, where those for
thesis (h=hollow, t=on-top) RGN are slightly shorter.
GNF+adsorbates Ead1 (kcal mol−1) Ead2 (kcal mol−1) rPG (Å)
Effect of pyrenes doping
HGN-PyH(h) 25.64 21.74 3.115
HGN-PyNO2(h) 30.70 23.33 3.029 HOMO and LUMO energies of HGN (C96H24) and RGN
HGN-PySO3H(h) 31.49 23.90 3.144 (C96H26) doped with pyrene and seven pyrene derivatives
HGN-PyCOOH(h) 29.76 23.61 3.156 are presented in Figs. 6 and 7, respectively. The variation of
HGN-PyNHCOCH3(h) 31.45 26.75 3.064 HOMO and LUMO energies were observed for both HGN
HGN-PyCH3(h) 28.40 22.98 3.185 and RGN.
HGN-PyOH(h) 28.32 23.36 3.098 The change of HOMO and LUMO energies could be better
HGN-PyNH2(h) 30.74 23.95 3.156 described by the shift of HOMO/LUMO energy of doped
RGN-PyH(h) 25.21 21.64 3.043 GNF as compared to the undoped one as listed in Tables 2
RGN-PyNO2(h) 30.53 24.48 3.020 and 3 for HGN and RGN, respectively. Positive (energy
RGN-PySO3H(h) 29.67 23.65 3.052 higher than undoped GNF) and negative (energy lower than
RGN-PyCOOH(h) 29.94 24.04 3.041 undoped GNF) shifts were observed for HOMO of HGN dop-
RGN-PyNHCOCH3(t) 32.79 26.95 2.999 ing with pyrene/pyrene derivatives, while only positive shifts
RGN-PyCH3(h) 27.17 23.03 3.083 were seen for LUMO of HGN doping with pyrene/pyrene
RGN-PyOH(h) 28.31 23.10 3.012 derivatives and HOMO/LUMO of pyrene/pyrene derivatives
RGN-PyNH2(h) 27.18 22.62 3.032 doped RGN. However, we could not find any relation between
the electron withdrawing/donating capability of the
J Mol Model (2017) 23:355 Page 5 of 9 355

Fig. 4 Optimized structures for adsorption complexes of pyrene/seven pyrene derivatives and HGN (top view)

substituents on pyrene and the HOMO/LUMO energy shift.
For RGNs, only a small shift around 0.01–0.08 eV were
noted.
Thus, it could be concluded that the doping has a minimal
effect on the electronic structure of RGN. However, it should
be noticed that the doping with pyNH2 (the strongest electron
donating substituents) provides the largest positive shift. For
HGN, the more pronounced positive shift was observed on
HOMO energy when doping with pyNH2 (0.19 eV), while
HGN with pyCOOH doping provides the smallest postive
shift (0.04 eV) on LUMO energy. Likewise, the positive and
the negative shifts on HOMO/LUMO energy of HGN as well
as their magnitude do not have any relation with the electron
withdrawing/donating capability of the substituent. Thus, the
electronic effect on pyrene ring does not have any direct in-
fluence on the electronic properties of GNF. This conclusion
surprised us. To see whether the electronic effect of the sub-
stituent on pyrene affects its electronic structure, we carried
out B3PW91/cc-pVDZ calculations for pyrene and seven
pyrene dervatives. More pronouced shifts on HOMO (0.01–
0.55 eV) and LUMO (0.03–1.03 eV) as compared to those of
physisorbed complexes were found, see Table S3. The nega-
tive shift was noticed to be associated with pyrene with elec-
tron withdrawing substituents, while the positive shift was

Fig. 5 Optimized structures for adsorption complexes of pyrene/seven derivatives and RGN (top view)
355 Page 6 of 9 J Mol Model (2017) 23:355

Table 2 Energy shifts of HOMO and LUMO (eV) upon doping as

compared to undoped HGN

Substituent LUMO (eV) HOMO (eV)

HGN-pyH 0.10 −0.01

HGN-pyNO2 0.09 0.00
HGN-pySO3H 0.12 0.04
HGN-pyCOOH 0.04 −0.05
HGN-pyNHCOCH3 0.08 −0.01
HGN-pyCH3 0.10 −0.01
HGN-pyOH 0.09 −0.01
HGN-pyNH2 0.07 0.19

softness of those complexes have the same trend as their

Fig. 6 HOMO and LUMO energies (eV) of HGN (C96H24) and HGN
doped with pyrene (−H) and pyrene derivatives (electron donating -NH2,
−OH, −CH 3, −NHCOCH 3, electron withdrawing -NO2 , −SO3H,-
COOH)
found for those with electron donating substituents. In addi-
tion, the magnitude of the shift is also in accordance with the
Hammett sigma constant, see Fig. S5. Hence, the electronic
effect of substituents correlates with HOMO/LUMO energy
shifts on pyrene. Unlike the physisorbed complex, direct sub-
stitution has a stronger effect on the electronic structure of the
aromatic compound since it interacts with π electrons directly.
Therefore, we could not easily control the shift of HOMO and
LUMO energies of GNF by just adding substituents on the
adsorbate. Chutia et al. [51] suggested that the electrical con-
ductivity is related to the chemical hardness and softness. The
chemical hardness and softness of complexes between pyrene/
pyrene derivatives and HGN/RGN were calculated and listed
in Table S6 and S7. Indeed, the chemical hardness and
HOMO-LUMO gap. However, they could not explain the
shift of HOMO/LUMO energies in physisorbed complexes
of pyrene/pyrene derivatives and GNFs because no relation-
ship between values of chemical hardness and softness and
HOMO/LUMO energy shifts was observed. To understand
HOMO/LUMO energy shifts of the physisorbed complexes,
we performed population analysis for pyrenes in the complex
using the natural population analysis (NPA) method. Table 4
represents the sum of the atomic net charge of the adsorbate
(pyrene/pyrene derivatives) in HGN and RGN complexes.
From Table 4, for both HGN and RGN complexes pyrene
and all pyrene dervatives have negative net charge suggesting
that the adsorbate pulls electrons from GNF. Thus, all adsor-
bates act as the elecron withdrawing group such that regard-
less of the substituent they all provide the same electronic
effect to GNF and changing electron withdrawing/donating
capabilites of the substituent on pyrene will not have any
substantial effect on the electronic property of GNF.
One application for nanoelectronic devices of graphene is
diode. The diode is a fundamental electronic device, which
consists of donor and acceptor. Thus, it allows electricity to
flow in one direction when voltage is applied.
Typically, the donor is the n-type semiconductor and the
acceptor is the p-type semiconductor. For organic

Table 3 Enegy shifts of HOMO and LUMO (eV) upon doping as

compared to undoped RGN

Substituent LUMO (eV) HOMO (eV)

RGN-pyH 0.05 0.01

RGN-pyNO2 0.06 0.03
RGN-pySO3H 0.05 0.02
RGN-pyCOOH 0.05 0.02
RGN-pyNHCOCH3 0.06 0.02
RGN-pyCH3 0.05 0.01
Fig. 7 HOMO and LUMO energies (eV) of RGN (C96H26) and RGN
doped with pyrene (−H) and pyrene derivatives (electron donating -NH2, RGN-pyOH 0.04 0.00
−OH, −CH 3 , −NHCOCH 3 , electron withdrawing -NO 2 , −SO 3 H, RGN-pyNH2 0.08 0.06
−COOH)
J Mol Model (2017) 23:355 Page 7 of 9 355

Table 4 Sum of atomic net charge of pyrene/pyrene derivatives in HGN 3.96 eV). Thus, large RGNs have electronic properties closer
and RGN complexes
to graphene. Very large HGNs with twice the number of car-
Adsorbates Net charge bon atoms as large RGNs will also have a close-to-zero band
gap. Pyrene and all pyrene derivatives have large binding
HGN complex RGN complex affinity with GNFs (21–27 kcal mol−1, BSSE corrected).
PyH −0.0191 −0.01548
Adsorption energies with pyrene and pyrene derivatives are
similar for RGNs and HGNs. The strenght of adsorption does
PyNO2 −0.01755 −0.01643
not depend on the type of substituent (whether it is electron
PySO3H −0.01253 −0.01423
withdrawing or donating). All pyrene derivatives adsorbed to
PyCOOH −0.01559 −0.01348
GNF stronger than pyrene due to the orbital overlap of the
PyNHCOCH3 −0.01459 −0.01030
substituent. Doping can shift HOMO and LUMO of GNF
PyCH3 −0.01643 −0.01355
but the magnitude and the direction of the shift does not follow
PyOH −0.01045 −0.00800
the electron withdrawing/donating capability of the substitu-
PyNH2 −0.00699 −0.01085
ent. Thus, the shift of HOMO and LUMO in GNFs cannot be
controlled by the physisorption since all pyrene dervatives act
as electron withdrawing groups. Therefore, the substituent on
pyrene does not provide a substantially different electronic
effect on GNF. A slight shift in HOMO and LUMO was ob-
served for RGN, while a more significant shift is noticed for
HGN doped with pyNH 2 (0.19 eV). HGN doped with
pyCOOH provides the least positive shift on LUMO
(0.04 eV). Therefore, HGN-pyNH2 and HGN-pyCOOH can
be made into a diode.

Acknowledgments The authors would like to acknowledge the 90th an-

Fig. 8 Schemetic drawing for organic diode with p- and n-type junction
semiconductor [64, 65], the electron will move from HOMO
of the donor to LUMO of the acceptor as illustrated in Fig. 8.
The HGN doped with pyNH2 has the highest positive shift of
HOMO (0.19 eV) and slight positive shift of LUMO
(0.07 eV), whereas the HGN doped with pyCOOH has the
smallest positive shift of LUMO (0.04 eV) and slight negative
shift of HOMO (−004 eV). Therefore, doped GNFs can be
used as diode, for example using HGN-pyCOOH (lower
LUMO) for p-type and HGN-pyNH2 (higher HOMO) for n-
type. Interestingly, Lee et al. [30] reported that rGO doped
with pyNH2 can be prepared for n-type semiconductor, while
rGO doped with pyCOOH for p-type semiconductor. Their
result somehow agrees with our finding.
Conclusions
The energy band gap of GNF decreases with the size increase
in agreement with the observation by Chutia et al. [51]. RGN
(0.13–2.12 eV) has a smaller band gap than HGN (1.23–
niversary of Chulalongkorn University for Mr.Thanawit Kuamit’s schol-
arship and Ratchadapisek Somphot Endownment Fund of Chulalongkorn
University RES560530184-AM and CU-57-038-AM (Advanced
Material Cluster) and the world class university grant of the comission
for higher education WCU-031-AM-57 for funding this work.
Additionally, the Center of Excellence in Computational Chemistry
(CECC), Department of Chemistry, Chulalonkorn University and the
National Electronics and Computer Technology Center (NECTEC) under
the National e-Science Consortium are gratefully acknowledged for their
support in computing facilities.
This work is devoted to the remembrance of His Majesty King
Bhumibol Adulyadej (1927-2016), for his life-time dedication to
Thailand.
References
1. Novoselov K, Geim A, Morozov S, Jiang D, Zhang Y, Dubonos S,
Grigorieva I, Firsov A (2004) Electric field effect in atomically thin
carbon films. Science 306:666–669
2. Berger C, Song ZM, Li TB, Li XB, Ogbazghi AY, Feng R, Dai ZT,
Marchenkov AN, Conrad EH, First PN, de Heer WA (2004)
Ultrathin epitaxial graphite: 2D electron gas properties and a route
toward graphene-based Nanoelectronics. J Phys Chem B 108:
19912–19916
3. Lin YM, Jenkins KA, Valdes-Garcia A, Small JP, Farmer DB,
Avouris P (2009) Operation of graphene transistors at gigahertz
frequencies. Nano Lett 9:422–426
4. Wang Y, Shi Z, Huang Y, Ma Y, Wang C, Chen M, Chen Y (2009)
Supercapacitor devices based on graphene materials. J Phys Chem
C 113:13103–13107
5. Zhang L, Zhou R, Zhao X (2010) Graphene-based materials as
supercapacitor electrodes. J Mater Chem 20:5983–5992
355 Page 8 of 9
6. Ghosh S, An X, Shah R, Rawat D, Dave B, Kar S, Talapatra S
(2012) Effect of 1-pyrene carboxylic-acid functionalization of
graphene on its capacitive energy storage. J Phys Chem C 116:
20688–20693
7. Mohammed M, Li Z, Cui J, Chen TP (2012) Junction investigation
of graphene/silicon Schottky diodes, nanoscale. Res Lett 7:302–
307
8. Gowtham S, Scheicher RH, Ahuja R, Pandey R, Karna SP (2007)
Physisorption of nucleobases on graphene: density-functional cal-
culations. Phys Rev B 76:033401–033404
9. Lee J-H, Choi Y-K, Kim H-J, Scheicher RH, Cho JH (2013)
Physisorption of DNA nucleobases on H-BN and graphene:
Vdw-corrected DFT calculations. J Phys Chem C 117:13435–
13441
10. Cortés-Arriagada D, Sanhueza L, Santander-Nelli M (2013)
Modeling the physisorption of bisphenol a on graphene and
graphene oxide. J Mol Model 19:3569–3580
11. Duy L, Abdelkader K, Elsebeth S, Per H, Talat SR (2012)
Physisorption of nucleobases on graphene: a comparative van der
Waals study. J Phys Condens Matter 24:424210
12. Martins TB, Miwa RH, da Silva AJR, Fazzio A (2007) Electronic
and transport properties of boron-doped graphene nanoribbons.
Phys Rev Lett 98:196803–196806
13. Wang XR, Li XL, Zhang L, Yoon Y, Weber PK, Wang HL, Guo J,
Dai HJ (2009) N-doping of graphene through electrothermal reac-
tions with ammonia. Science 324:768–771
14. Wei DC, Liu YQ, Wang Y, Zhang HL, Huang LP, Yu G (2009)
Synthesis of N-doped graphene by chemical vapor deposition and
its electrical properties. Nano Lett 9:1752–1758
15. Panchokarla LS, Subrahmanyam KS, Saha SK, Govindaraj A,
Krishnamurthy HR, Waghmare UV, Rao CNR (2009) Synthesis,
structure, and properties of boron- and nitrogen-doped graphene.
Adv Mater 21:4726–4730
16. Li X, Wang H, Robinson JT, Sanchez H, Diankov G, Dai H (2009)
Simultaneous nitrogen doping and reduction of graphene oxide. J
Am Chem Soc 131:15939–15944
17. Choi J, Kim KJ, Kim B, Lee H, Kim S (2009) Covalent
functionalization of epitaxial graphene by azidotrimethylsilane. J
Phys Chem C 113:9433–9435
18. Elias DC, Nair RR, Mohiuddin TMG, Morozov SV, Blake P,
Halsall MP, Ferrari AC, Boukhvalov DW, Katsnelson MI, Geim
AK, Novoselov KS (2009) Control of graphene’s properties by
reversible hydrogenation: evidence for Graphane. Science 323:
610–613
19. Balog R, Jorgensen B, Nilsson L, Andersen M, Rienks E, Bianchi
M, Fanetti M, Laegsgaard E, Baraldi A, Lizzit S, Sljivancanin Z,
Besenbacher F, Hammer B, Pedersen TG, Hofmann P, Hornekaer L
(2010) Bandgap opening in graphene induced by patterned hydro-
gen adsorption. Nat Mater 9:315–319
20. Choi J, Lee H, Kim KJ, Kim B, Kim S (2010) Chemical doping of
epitaxial graphene by organic free radicals. J Phys Chem Lett 1:
505–509
21. Chen W, Chen S, Qi DC, Gao XY, Wee ATS (2007) Surface transfer
p-type doping of epitaxial graphene. J Am Chem Soc 129:10418–
10422
22. Voggu R, Das B, Rout CS, Rao CNR (2008) Effects of charge
transfer interaction of graphene with electron donor and acceptor
molecules examined using Raman spectroscopy and cognate tech-
niques. J Phys Condens Matter 20:472204–472208
23. Grimme S (2008) Do special noncovalent pi-pi stacking interac-
tions really exist? Angew Chem 47:3430–3434
24. Dong XC, Fu DL, Fang WJ, Shi YM, Chen P, Li LJ (2009) Doping
single-layer graphene with aromatic molecules. Small 5:1422–1426
25. Zhang YH, Zhou KG, Xie KF, Zeng J, Zhang HL, Peng Y (2010)
Tuning the electronic structure and transport properties of graphene
J Mol Model (2017) 23:355
by noncovalent functionalization: effects of organic donor, acceptor
and metal atoms. Nanotechnology 21:065201–065207
26. Zhang ZX, Huang HL, Yang XM, Zang L (2011) Tailoring elec-
tronic properties of graphene by pi-pi stacking with aromatic mol-
ecules. J Phys Chem Lett 2:2897–2905
27. Wang WZ, Sun T, Zhang Y, Wang YB (2014) Substituent effects in
the pi...pi interaction between graphene and benzene: an indication
for the noncovalent functionalization of graphene. Comput Theor
Chem 1046:64–69
28. Wang WZ, Sun T, Zhang Y, Wang YB (2015) Benchmark calcula-
tions of the adsorption of aromatic molecules on graphene. J
Comput Chem 36:1763–1771
29. Yin YF, Cervenka J, Medhekar NV (2015) Tunable hybridization
between electronic states of graphene and physisorbed hexacene. J
Phys Chem C 119:19526–19534
30. Lee J, Hwang E, Lee E, Seo S, Lee H (2012) Tuning of n- and p-
type reduced graphene oxide transistors with the same molecular
backbone. Chem Eur J 18:5155–5159
31. Chen RJ, Zhang YG, Wang DW, Dai HJ (2001) Noncovalent side-
wall functionalization of single-walled carbon nanotubes for protein
immobilization. J Am Chem Soc 123:3838–3839
32. Besteman K, Lee JO, Wiertz FGM, Heering HA, Dekker C (2003)
Enzyme-coated carbon nanotubes as single-molecule biosensors.
Nano Lett 3:727–730
33. Xu YX, Bai H, Lu GW, Li C, Shi GQ (2008) Flexible graphene
films via the filtration of water-soluble noncovalent functionalized
graphene sheets. J Am Chem Soc 130:5856–5857
34. An XH, Simmons TJ, Shah R, Wolfe C, Lewis KM, Washington M,
Nayak SK, Talapatra S, Kar S (2010) Stable aqueous dispersions of
noncovalently functionalized graphene from graphite and their mul-
tifunctional high-performance applications. Nano Lett 10:4295–
4301
35. Babu P, Sangeetha NM, Vijaykumar P, Maitra U, Rissanen K, Raju
AR (2003) Pyrene-derived novel one- and two-component
organogelators. Chem Eur J 9:1922–1932
36. Soustek P, Michl M, Almonasy N, Machahcky O, Dvorak M,
Lycka A (2008) The synthesis and fluorescence of N-substituted
1- and 2-aminopyrenes. Dyes Pigments 78:139–147
37. Deng JP, Chen WH, Chiu SP, Lin CH, Wang BC (2014) Edge-
termination and core-modification effects of hexagonal Nanosheet
graphene. Molecules 19:2361–2373
38. Zhou PP, Zhang RQ (2015) Physisorption of benzene derivatives
on graphene: critical roles of steric and stereoelectronic effects of
the substituent. Phys Chem Chem Phys 17:12185–12193
39. Li H, Zou L, Pan L, Sun Z (2010) Using graphene nano-flakes as
electrodes to remove ferric ions by capacitive deionization. Sep
Purif Technol 75:8–14
40. Berger C, Song Z, Li X, Wu X, Brown N, Naud C, Mayou D, Li T,
Hass J, Marchenkov AN, Conrad EH, First PN, de Heer WA (2006)
Electronic confinement and coherence in patterned epitaxial
graphene. Science 312:1191–1196
41. Schedin F, Geim AK, Morozov SV, Hill EW, Blake P, Katsnelson
MI, Novoselov KS (2007) Detection of individual gas molecules
adsorbed on graphene. Nat Mater 6:652–655
42. Molitor F, Güttinger J, Stampfer C, Dröscher S, Jacobsen A, Ihn T,
Ensslin K (2011) Electronic properties of graphene nanostructures.
J Phys Condens Matter 23:243201–243215
43. Tapas S, Das I, Banerjee S (2009) Contribution of energy-gap in the
ferromagnetic spin–wave spectrum on magnetocaloric parameters
of Ceru 2 Ge 2. J Phys Condens Matter 21:026010
44. Neubeck S, Ponomarenko LA, Freitag F, Giesbers AJM, Zeitler U,
Morozov SV, Blake P, Geim AK, Novoselov KS (2010) From one
electron to one hole: quasiparticle counting in graphene quantum
dots determined by electrochemical and plasma etching. Small 6:
1469–1473
J Mol Model (2017) 23:355
45. Datta SS, Strachan DR, Khamis SM, Johnson ATC (2008)
Crystallographic etching of few-layer graphene. Nano Lett 8:
1912–1915
46. Ci L, Xu Z, Wang L, Gao W, Ding F, Kelly KF, Yakobson BI,
Ajayan PM (2008) Controlled nanocutting of graphene. Nano
Res. 1:116–122
47. Ci L, Song L, Jariwala D, Elías AL, Gao W, Terrones M, Ajayan
PM (2009) Graphene shape control by multistage cutting and trans-
fer. Adv Mater 21:4487–4491
48. Campos LC, Manfrinato VR, Sanchez-Yamagishi JD, Kong J,
Jarillo-Herrero P (2009) Anisotropic etching and nanoribbon for-
mation in single-layer graphene. Nano Lett 9:2600–2604
49. Nagai H, Nakano M, Yoneda K, Fukui H, Minami T, Bonness S,
Kishi R, Takahashi H, Kubo T, Kamada K, Ohta K, Champagne B,
Botek E (2009) Theoretical study on third-order nonlinear optical
properties in hexagonal graphene nanoflakes: edge shape effect.
Chem Phys Lett 477:355–359
50. Shakourian-Fard M, Kamath G (2017) The effect of defect types on
the electronic and optical properties of graphene nanoflakes
physisorbed by ionic liquids. Phys Chem Chem Phys 19:4383–
4395
51. Chutia A, Sahnoun R, Deka RC, Zhu Z, Tsuboi H, Takaba H,
Miyamoto A (2011) Local electronic and electrical properties of
functionalized graphene nanoflakes. PhyB 406:1665–1672
52. Parr RG (1982) Density functional theory. In: Electron distributions
and the chemical bond. Springer, Boston, pp 95–100
53. Paier J, Hirschl R, Marsman M, Kresse G (2005) The Perdew–
Burke–Ernzerhof exchange-correlation functional applied to the
G2-1 test set using a plane-wave basis set. J Chem Phys 122:
234102–234114
54. Woon DE, Dunning Jr TH (1993) Gaussian basis sets for use in
correlated molecular calculations. III. The atoms aluminum through
argon. J Chem Phys 98:1358–1371
Page 9 of 9 355
55. Frisch M, Trucks G, Schlegel HB, Scuseria G, Robb M, Cheeseman
J, Scalmani G, Barone V, Mennucci B, Petersson G et al. (2009)
Gaussian 09, revision a.02. Gaussian Inc., Wallingford, p 200
56. Zhao Y, Truhlar DG (2008) The M06 suite of density Functionals for
main group thermochemistry, thermochemical kinetics, noncovalent
interactions, excited states, and transition elements: two new functionals
and systematic testing of four M06-class functionals and 12 other func-
tionals. Theor Chem Accounts 120:215–241
57. Rungnim C, Chanajaree R, Rungrotmongkol T, Hannongbua S,
Kungwan N, Wolschann P, Karpfen A, Parasuk V (2016) How
strong is the edge effect in the adsorption of anticancer drugs on a
graphene cluster? J Mol Model 22:85–93
58. Becke AD (1993) Becke’s three parameter hybrid method using the
LYP correlation functional. J Chem Phys 98:5648–5652
59. Perdew JP (1991) In: Ziesche P, Eschrig H (eds) Electronic structure
of solids. Akademie, Berlin
60. Crowley JM, Tahir-Kheli J, Goddard WA (2016) Resolution of the
band gap prediction problem for materials design. J Phys Chem Lett
7:1198–1203
61. Colherinhas G, Fileti EE, Chaban VV (2015) Can inorganic salts
tune electronic properties of graphene quantum dots? Phys Chem
Chem Phys 17:17413–17420
62. Hunter CA (1993) Arene—arene interactions: electrostatic or
charge transfer? Angew Chem 32:1584–1586
63. Hunter CA, Sanders JKM (1990) The nature of pi-pi interactions. J
Am Chem Soc 112:5525–5534
64. Lüssem B, Riede M, Leo K (2013) Doping of organic semiconduc-
tors. Phys Status Solidi A 210:9–43
65. Salzmann I, Heimel G (2015) Toward a comprehensive understand-
ing of molecular doping organic semiconductors (review). J
Electron Spectrosc 204:208–222