Abstract
We introduce an effective tightbinding model to discuss pentagraphene and present an analytical solution. This model only involves the πorbitals of the sp^{2}hybridized carbon atoms and reproduces the two highest valence bands. By introducing energydependent hopping elements, originating from the elimination of the sp^{3}hybridized carbon atoms, also the two lowest conduction bands can be well approximated  but only after the inclusion of a Hubbard onsite interaction as well as of assisted hopping terms. The eigenfunctions can be approximated analytically for the effective model without energydependent hopping elements and the optical absorption is discussed. We find large isotropic absorption ranging from 7.5% up to 24% for transitions at the Γpoint.
Introduction
Arguably, carbon is the most versatile element being capable to form various stable structures with graphene^{1,2} being its most prominent twodimensional allotrope out of which most carbon structures can be built such as fullerene^{3}, carbonnanotubes^{4} or multilayer graphene^{5} and graphite.
Recently, a new allotrope was proposed which cannot be composed from a graphene sheet: pentagraphene^{6}. It entirely consists of carbon atoms forming pentagons within the Cairo tiling and it remains almost flat by adding to its sp^{2}hybridised carbon atoms also sp^{3}hybridised carbon atoms, arranging them in three parallel horizontal planes separated by approximately half an Angstrom. Pentagraphene, therefore, only experiences a buckling on the atomic scale of a unit cell and would represent another example of a twodimensional semiconductor. Also other pentagonal monolayer crystals of boron nitride and silver azide have recently been investigated^{7}.
Even though pentagraphene has not been synthesised experimentally, the theoretical analysis point out many intriguing properties which are worth discussing in more detail. Most strikingly in terms of applications is probably the large predicted bandgap of 3.25 eV which makes it a potential candidate for blue absorption/emission. Nevertheless, the optical properties such as absorption due to a linearly polarised light field have not been addressed, yet.
The objective of this work is twofold. First, we will discuss the possibility of a simple tightbinding description to model the valence and conduction bands closest to the neutrality point. Our model will only include the πorbitals of the sp^{2}hybridised carbon atoms for which an analytical solution is possible. This is reminiscent to widelyused tightbinding models for graphene^{8,9,10} and carbonnanotubes^{11,12}. Second, using the analytical approximation, we will also determine the absorption of linearly polarized light via Fermi’s Golden Rule.
The paper is organised as follows. In Sec. II, we discuss the general structure of pentagaphene. In Sec. III, we introduce the effective 4band model and present its analytical approximation. In Sec. IV, we discuss the absorption at the highsymmetry points of the Brillouin zone and close with a summary and conclusions. In the supplementary information, we point out the importance of correlation effects in order to justify the parameters of the extended tightbinding model.
Pentagraphene
Lattice structure
The regular Cairo pentagonal patterning is characterized by four bonds forming the pentagon, three of which have the distance a and one has the distance . The unit cell consists of six carbon atoms and is defined by the two lattice vectors and . The length of the quadratic unit cell is obtained from firstprinciple studies to be 3.64 Å^{6}, which translates into a = 1.49 Å and b = 1.09 Å.
The real distances as obtained from firstprinciple calculations turn out to be slightly different, i.e., C1 − C2 = 1.55 Å and b = C2 − C2 = 1.34 Å, where there are two C1atoms and four C2atoms denoting the carbon atoms with sp^{3} and sp^{2}hybridization, respectively. Furthermore, the atoms are arranged in three different horizontal planes, of which the C1atoms form the central plane and two of the four C2atoms the upper and the lower plane, respectively. The total distance between the C1 and C2atomic horizontal planes is h = 0.6 Å which yields the projected 2Ddistance a = C1 − C2(projected) = 1.43 Å and b = C2 − C2 = 1.34 Å.
The distorted Cairo pentagonal patterning is shown on the left hand side of Fig. 1 where the C1 and C2atoms are represented by red and black dots, respectively. The black horizontal and vertical bonds of length b connect the C2atoms whereas the red bonds of the projected length a connect the C1 with the C2atoms. There is a slight difference of 1% between the two bonds connecting the C1 − C2 atoms that form a pentagon, but this will be neglected. We also assume that the projected distance C1 − C2(projected) is the distance a of the regular Cairo patterning. The unit cell is denoted by the shaded square and consists of two C1atoms and four C2atoms.
Let us finally comment on the symmetry group. The threedimensional lattice possesses the S_{4} point group and D_{2d} full space group. The latter includes the following symmetry elements: one C2 axis along the direction perpendicular to the a_{1} − a_{2} plane, two C2′ axes perpendicular to C2, two dihedral planes σ_{d} bisecting the angles formed by pairs of the C2′ axes and two improper S_{4} axes. For the strictly twodimensional lattice, the symmetry elements is doubled going from D_{2d} to D_{4h} full space group.
Density Functional Theory calculations
We calculate the band structure of pentagraphene using the VASP code^{13,14}, based on density functional theory (DFT). All calculations were done using the Projector Augmented Wave potentials^{15} and employing the PerdewBurkeErnzerhof flavour of the generalized gradient approximation for the exchangecorrelation functional^{16}. Plane waves with a energy cutoff of 500 eV were employed to describe the valence electrons (2s^{2} 2p^{2}) of the C atoms. The employed Brillouin zone for the relaxation calculations is 9 × 9 × 1 in the MonkhorstPack scheme^{17}, while high accurate electronic calculations to estimate the relative C1/C2 weight were performed using 11 × 11 × 1. The band structure of free standing pentagraphene layer was calculated for a fixed lattice constants of a = b = 3.64 Å^{6}, and including a vacuum distance of around 20 Å. In all calculations the atomic positions are relaxed till forces are less than 0.015 e/Å.
We find slightly larger atomic distances than the one cited in the previous subsection which were taken from ref. 6. The obtained band structure shows the expected narrowing of the band gap, common to all local and semilocal exchangecorrelation approximations, which is corrected via a rigid shift of the conduction band to mimic the obtained direct ΓΓ band gap using hybrid functionals^{6}. After the rigid shift, the band dispersion shown in Fig. 2 as red stars agrees remarkably well to the more costly hybrid functionals.
In order to obtain more insight of how the atomic and orbital nature influences the electronic structure we plot on the right hand side of Fig. 2 the projected density of states (PDOS) of the valence bands (VBs) and conduction bands (CBs). We observe that the band gap after correction is of the order of 3.5 eV. For both the VBs and CBs there is a noticeable difference between the total and the sorbital, evidencing that the highest contribution comes from the p orbital, whose contribution is above 94 and 90% respectively for the VB and CB. Similarly we can discuss the atomic contribution using the C1/C2 ratio which is less than 9% and 15% respectively for the VBs and CBs. We thus conclude that a four band model is an adequate approximation.
Despite the useful comparison obtained using the PDOS, the projected values should be considered qualitatively since the associated numerical errors are noticeable. These errors can be estimated as the difference between the real band charge, which is exactly 2 electrons, and the projected charge of the VBs and CBs, which is 1.5 and 1.35 electrons respectively.
Tightbinding and analytical approach
Our goal is to introduce a tightbinding model that only considers the four C2atoms where the atoms A and B form the vertical dimer and the atoms C and D the horizontal dimers, see right hand side of Fig. 1. Both dimers are coupled by the hopping matrix element t_{0} which connects the two πorbitals. We will set this to the typical value t_{0} = 2.7 eV^{5}. The other hopping matrix elements involve hopping processes between the two dimers t and nextnearest hopping processes between the same dimers t′. To simplify our model, we will set t = t′ and determine its value from a fit to the DFTband structure. As we will show, this model can also be well described analytically. But first, we will discuss the full tightbinding model including the six atoms of the unit cell and 4 orbitals within the SlaterKoster formalism^{18}.
SlaterKoster approach
We start our tightbinding description with the full model that will include all six atoms of the unit cell and all four orbitals of the second atomic energy level, i.e., 2s, 2p_{x}, 2p_{y}, 2p_{z}. The hopping parameters are chosen to be the one of graphene scaled by the corresponding distance and taken from ref. 19: V_{ssσ} = −5.34 eV, V_{spσ} = 6.40 eV, V_{ppσ} = 7.65 eV, and V_{ppπ} = −2.80 eV. The atomic energy levels of the sp^{2}hybridized C2atoms are ε_{s} = −2.85 eV, , ; the ones of the sp^{3}hybridized C1atoms are chosen as ε_{s} = −4.05 eV, ε_{p} = 2 eV. The 24 × 24 Hamiltonian is obtained following the SlaterKoster parametrization including only nearestneighbor hopping^{18}.
In Fig. 2, the lowest energy bands obtained from the full tightbinding model are shown (black lines) and compared to the DFTresults (red stars). Qualitatively, the tightbinding can reproduce the lowest conduction bands and the highest valence bands. Still, quantitatively there are several disagreements: i) the energy gap is not accurately reproduced; ii) the splitting of the two conduction as well as of the two valence bands is not small; iii) bands further away from the Fermi level become worse and they can at most only be described qualitatively.
The above analysis does not contain any fitting parameter. Nevertheless, the quantitative disagreement suggests other terms to become important. In fact, in the case of the effective fourband model, we will be only able to adequately describe the bands by including an onsite interaction as well as an assisted hopping term. This shall be discussed in the following. Also the fitting parameters should be improved and extracted from firstprinciple calculations for pentagraphene, thus avoiding the use of parameters obtained for graphene.
Effective fourband model
As outlined above, we will build up an effective tightbinding model by only considering lattice sites with an unbounded πelectron. There are thus four atoms in the unit cell and direct hopping (t_{0}) within the unit cell is only between the dimers A, B and C, D, respectively. All other hopping processes involve intermediate lattice sites with sp^{3}hybridisation and we assume that this hopping is the same (t = t′) and considerably less than the direct π − π hopping. The Hamiltonian is thus given by with:
where with σ_{x}, σ_{y} the Paulimatrices and c_{1/2} = cos(k · a_{1/2}), s_{1/2} = sin(k · a_{1/2}). We further set E_{0} = 0. The coupling between the two dimers is given by
In Fig. 3, we compare the resulting band structure (black lines) to the full model obtained from DFTcalculations (red stars). Whereas the two valence bands (VBs) approximately agree with the exact firstprinciple band structure, the two conduction bands (CBs) do not show the large splitting in the Γ − X and Γ − M direction and seem to be inverted. Still, we want to emphasize that by setting t_{0} = 2.7 eV (which is the typical tightbinding hopping parameter of πbands^{5}) only one fitting parameter is involved by choosing t/t_{0} = 0.056. Further, we applied a rigid energy shift of E_{0} = 2.3 eV. Also, the basic features of the band structure are reproduced, i.e., an (no) energy separation of the valence (conduction) bands at the Γpoint and a twofold degeneracy along the X − M direction.
The conduction bands can be considerably improved within our fourband model by introducing an energydependent hopping parameter. This term is formally obtained by including an effective p_{z}orbital of the sp^{3}hybridized C1atoms which are then projected out, see appendix. The projected 4 × 4 Hamiltonian is again given by Eq. (1) with where the effective energy difference of the p_{z}atomic orbitals and . With , E_{X} = −3.75 eV and an additional rigid energy shift , the eigenenergies of the conduction band are then obtained selfconsistently to yield the upper blue lines of Fig. 3.
In order to justify the above parameters, it is necessary to include an onsite interaction at the C2atoms and assisted hopping terms, see SI. Neglecting these correlations, the onsite energy shift yields E_{X} = 2 eV and the two valence bands of the original fourband model with constant hopping amplitude t = 0.056 t_{0} are almost perfectly reproduced by the fourband model with energydependent hopping parameter of . Additionally, we introduced an energy shift of . This can be seen in Fig. 3 where the lower blue lines are almost on top of the black lines.
Our analysis indicates that correlation effects are important in order to reproduce the bandstructure of the conduction band obtained from DFTcalculations and that electronelectron interactions are effectively screened in the valence band. The change in and the different constant energy shifts and further suggest an assisted hopping amplitude which depends on the occupation number of the C2atoms^{20,21}.
The inclusion of both correlation terms yields a consistent picture when choosing the Hubbardinteraction U ≈ 10 eV and the assisted hopping term W ≈ 2 eV. The assisted hopping term may further lead to a superconducting condensate when pumping electrons into the conduction bands^{20,21,22,23}.
Analytical solution
The eigenenergies of the bands around the neutrality point can be well approximated by the (extended) 4 × 4 model, reproducing correctly the degeneracy of the VBs and CBs along the X − Mdirection and the band splitting along the other directions with a (zero) gap in the VBs (CBs). Nevertheless, to also obtain orthonormal eigenvectors, we will continue the discussion by approximating the energydependent hopping parameter by the constant hopping parameter t also for the conduction bands. In the following, we set t_{0} = 1 and assume . A first approximation is thus given by neglecting the interdimer coupling h_{12}. Comparing this approximation with the exact numerical solution, we see that the conduction bands are well described by neglecting h_{12}. For the eigenvalues, we then obtain with
However, the valence band experiences a splitting that cannot be account for by simply setting h_{12} = 0. Let thus U denote the unitary transformation which diagonalizes the Hamiltonian with h_{12} = 0 and lets only consider the upper left and lower right 2 × 2matrix of U^{†}HU that connects the two valence and conduction band states, respectively. Keeping now only terms of these matrices that are linear in t yields the linear approximation for and the following eigenvalues for the valence band:
The corresponding eigenvectors need to be transformed by U to obtain the approximate eigenstates of the original Hamiltonian. We thus end up with the following set of 4dimensional eigenvectors:
with and .
The above approximation matches excellently the numerically exact band structure for , see Fig. 4.
Optical properties
Pentagraphene displays a band gap of 3.25 eV which makes it a potential candidate for blue absorption/emission. In order to discuss the optical response of the system, we will obtain the coupling Hamiltonian via the Peierls substitution in kspace. Nevertheless, for systems with various atoms in the unit cell, it is crucial to represent the Hamiltonian within a basis that distinguishes the relative phase of the atoms within the same unit cell^{24}. The above representation has thus to be modified by using the kstates of the following basis:
Placing the origin of the unit cell in the middle of the ABdimer and choosing it in the direction of the yaxis, we have δ_{A} = (0, −b/2), δ_{B} = (0, b/2), δ_{C} = (b_{+}, 0), and δ_{D} = (b_{+} + b, 0) where . For the regular Cairo patterning, we have ; the unitcell vectors are and with .
With respect to the new basis, the 2 × 2 Hamiltonians of Eq. (1) now read
The eigenvectors read
Fermi’s “golden rule”
We are interested in the system response at small fields and therefore expand the Hamiltonian up to terms linear in k. With (minimal coupling), we obtain the following coupling Hamiltonian:
with , and
where the velocities ħv = bt_{0} and .
We can now apply Fermi’s “golden rule” to calculate the absorption due to an incoming electric field E(t) = −∂_{t}A. The incoming energy flux of a propagating sinusoidal linearly polarized electromagnetic plane wave of a fixed frequency is given by and the absorbed energy flux W_{a} = ηħω with η the transition rate. Since the momentum is conserved in the absorption process, only transitions from (k, v) to (k, c) are allowed. The total transition rate is then obtained by summing over all initial states k which yields:
where the sum goes over the first Brillouin zone. For a system with N_{c} = N^{2} unit cells, the possible kvalues thus read:
with and .
Only considering the leading term in t_{0}, we can neglect v_{12} and obtain for the matrixelements:
By interchanging k_{x} ↔ k_{y}, we have φ_{+} ↔ φ_{−}, indicating that the absorption displays the underlying fourfold symmetry within our approximation (degenerate conduction band). But we will neglect φ_{±} in the following since it would yield another correction of order t. We can thus write the transition rate as
Optical absorption
One measure for absorption is the joint density of states (DOS):
where the 2 comes from the twofold degeneracy of the conduction band.
The DOS of the (degenerate) conduction band ρ^{c} is also given by Eq. (19) by setting and neglecting the summation over m. With , we can obtain an analytical expression:
where K(x) is the complete elliptic function. At the bandedges, we thus have a constant density of state of . At the bandcenter, there is a vanHove singularity at the Xpoint with a logarithmically diverging density of states .
The DOS of the valence band (setting ε^{c} = 0) can also be expressed in closed form, but already this and especially the joint density of states are quite complicated analytical function and we will not pursue this any further. Instead, we will expand the dispersion around the highsymmetry points and calculate the optical absorption which should be particularly high at vanHove singularities.
The optical absorption is defined by the ratio of the absorbed and the incoming energy flux, . The expansion around the Γpoint (Γ = 0) yields an isotropic absorption for both transitions from the two (m = ±)valence bands. These are related to the transition energies ħω_{+} = 2t_{0} and ħω_{−} = 2t_{0} + 8t, respectively. To leading order in t_{0}/t, we get:
The numerical values are obtained from (b/a)^{2} ≈ 0.88 and t_{0}/t ≈ 18 in units of the universal absorption of suspended graphene with the finestructure constant^{25,26}. This universal absorption is also present in InAsquantum wells^{27} and other systems^{28}. But here, the expression depends on material constants and is thus nonuniversal. It is also substantially higher with . This is remarkable having in mind that the maximal absorption of suspended twodimensional materials is 50% and to our knowledge the highest value of a suspended twodimensional system^{29}. However, at such high absorption the simple “golden rule”approach cannot be trusted anymore and more accurate calculations based on first principles are needed. In the Supplementary Information, we report on DFTresults for the optical conductivity which reaches absorption values up to .
Summary
In this work, we have investigated the possibility of an analytical description of pentagraphene. We were able to adequately approximate the two highest valence and lowest conduction bands within a simple four band model, but to match the conduction bands an energy dependent hopping element was necessary. To justify the parameters of the effective model, the inclusion of an onsite energy at the sp^{2}hybridized C2atoms was necessary. This interaction should be screened for the valence band and we also expect an assisted hopping term in the effective model which might lead to a superconducting condensate at low temperatures when pumping electrons into the conduction bands^{20,21,22,23}. A possible extension of our approach would include multiorbitals in the spirit of the WeaireThorpe (WT) model^{30,31}.
Using a constant hopping element between dimers, we were able to present an analytical solution for the eigenvalues as well as for the eigenvectors and calculated the optical absorption within this approximation. For small field strengths and for transitions around the Γpoint, this yielded an isotropic absorption of up to 24% which is remarkably large compared to usual twodimensional materials such as graphene^{28}. This suggests the interpretation that the quasiisolated sp^{2}dimers are capable to act as antennas which efficiently couple to the incoming radiation.
It would be interesting to further investigate the influence of correlation effects on the bandstructure as well as on the optical properties. Also the inclusion of electronhole interaction and excitonic resonances are likely to change the absorption close to the band edges. Moreover, ferromagnetism is expected due to quasiflat bands^{32}. These issues shall be discussed in the future.
Additional Information
How to cite this article: Stauber, T. et al. Tightbinding approach to pentagraphene. Sci. Rep. 6, 22672; doi: 10.1038/srep22672 (2016).
References
 1.
Novoselov, K. S. et al. Electric field effect in atomically thin carbon films. Science 306, 666–669 (2004).
 2.
Novoselov, K. S. et al. Twodimensional atomic crystals. Proc. Natl. Acad. Sci. USA 102, 10451–10453 (2005).
 3.
Kroto, H. W., Heath, J. R., O’Brien, S. C., Curl, R. F. & Smalley, R. E. C60: Buckminsterfullerene. Nature 318, 162–163 (1985).
 4.
Iijima, S. Helical microtubules of graphitic carbon. Nature 354, 56–58 (1991).
 5.
Castro Neto, A. H., Guinea, F., Peres, N. M. R., Novoselov, K. S. & Geim, A. K. The electronic properties of graphene. Rev. Mod. Phys. 81, 109–162 (2009).
 6.
Zhang, S. et al. Pentagraphene: A new carbon allotrope. Proc. Natl. Acad. Sci. USA 112, 2372–2377 (2015).
 7.
Yagmurcukardes, M. et al. Pentagonal monolayer crystals of carbon, boron nitride, and silver azide. J. App. Phys. 118 (2015).
 8.
Wallace, P. R. The band theory of graphite. Phys. Rev. 71, 622–634 (1947).
 9.
McClure, J. W. Diamagnetism of graphite. Phys. Rev. 104, 666–671 (1956).
 10.
Slonczewski, J. C. & Weiss, P. R. Band structure of graphite. Phys. Rev. 109, 272–279 (1958).
 11.
Jorio, A. et al. Structural (n, m) determination of isolated singlewall carbon nanotubes by resonant raman scattering. Phys. Rev. Lett. 86, 1118–1121 (2001).
 12.
Reich, S., Maultzsch, J., Thomsen, C. & Ordejón, P. Tightbinding description of graphene. Phys. Rev. B 66, 035412 (2002).
 13.
Kresse, G. & Hafner, J. Ab initio molecular dynamics for liquid metals. Phys. Rev. B 47, 558–561 (1993).
 14.
Kresse, G. & Hafner, J. Ab initio moleculardynamics simulation of the liquidmetal amorphoussemiconductor transition in germanium. Phys. Rev. B 49, 14251–14269 (1994).
 15.
Kresse, G. & Joubert, D. From ultrasoft pseudopotentials to the projector augmentedwave method. Phys. Rev. B 59, 1758–1775 (1999).
 16.
Perdew, J. P., Burke, K. & Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 77, 3865–3868 (1996).
 17.
Monkhorst, H. J. & Pack, J. D. Special points for brillouinzone integrations. Phys. Rev. B 13, 5188–5192 (1976).
 18.
Slater, J. C. & Koster, G. F. Simplified lcao method for the periodic potential problem. Phys. Rev. 94, 1498–1524 (1954).
 19.
Yuan, S., Rösner, M., Schulz, A., Wehling, T. O. & Katsnelson, M. I. Electronic structures and optical properties of partially and fully fluorinated graphene. Phys. Rev. Lett. 114, 047403 (2015).
 20.
Hirsch, J. E. & Marsiglio, F. Hole superconductivity in oxides: A twoband model. Phys. Rev. B 43, 424–434 (1991).
 21.
Hirsch, J. E. Electron and holehopping amplitudes in a diatomic molecule. Phys. Rev. B 48, 3327–3339 (1993).
 22.
Guinea, F. Effect of assisted hopping on the formation of local moments in magnetic impurities and quantum dots. Phys. Rev. B 67, 195104 (2003).
 23.
Stauber, T. & Guinea, F. Assisted hopping in the anderson impurity model: A flow equation study. Phys. Rev. B 69, 035301 (2004).
 24.
Paul, I. & Kotliar, G. Thermal transport for manybody tightbinding models. Phys. Rev. B 67, 115131 (2003).
 25.
Nair, R. R. et al. Fine structure constant defines visual transparency of graphene. Science 320, 1308–1308 (2008).
 26.
Mak, K. F. et al. Measurement of the optical conductivity of graphene. Phys. Rev. Lett. 101, 196405 (2008).
 27.
Fang, H. et al. Quantum of optical absorption in twodimensional semiconductors 110, 11688–11691 (2013).
 28.
Stauber, T., NoriegaPérez, D. & Schliemann, J. Universal absorption of twodimensional systems. Phys. Rev. B 91, 115407 (2015).
 29.
García de Abajo, F. J. Colloquium: Light scattering by particle and hole arrays. Rev. Mod. Phys. 79, 1267–1290 (2007).
 30.
Weaire, D. & Thorpe, M. F. Electronic properties of an amorphous solid. i. a simple tightbinding theory. Phys. Rev. B 4, 2508–2520 (1971).
 31.
Hatsugai, Y., Shiraishi, K. & Aoki, H. Flat bands in the weairethorpe model and silicene. New J. Phys. 17, 025009 (2015).
 32.
Mielke, A. & Tasaki, H. Ferromagnetism in the hubbard model. examples from models with degenerate singleelectron ground states. Comm. Math. Phys. 158, 341–371 (1993).
Acknowledgements
We thank Félix Yndurain and Paul Wenk for useful discussions. This work has been supported by Spain’s MINECO under grants FIS201348048P and FIS201457432P, by the Comunidad de Madrid under grant S2013/MIT3007 MAD2DCM, and by Deutsche Forschungsgemeinschaft via GRK 1570. JIB thanks the ERC starting Investigator Award, grant #239739 STEMOX.
Author information
Affiliations
Instituto de Ciencia de Materiales de Madrid, CSIC, 28049 Madrid, Spain
 T. Stauber
 & J. I. Beltrán
GFMC and Instituto Pluridisciplinar, Departamento de Física Aplicada III, Universidad Complutense de Madrid, 28040 Madrid, Spain
 J. I. Beltrán
IMDEA Materials Institute, C/Eric Kandel 2, 28906 Getafe, Madrid, Spain
 J. I. Beltrán
Institute for Theoretical Physics, University of Regensburg, D93040 Regensburg, Germany
 J. Schliemann
Authors
Search for T. Stauber in:
Search for J. I. Beltrán in:
Search for J. Schliemann in:
Contributions
T.S., J.I.B. and J.S. were involved in the tightbinding calculations, J.I.B. performed the DFT calculations, and T.S. wrote the manuscript. All authors contributed to the scientific discussion and revised the manuscript.
Competing interests
The authors declare no competing financial interests.
Corresponding author
Correspondence to T. Stauber.
Supplementary information
PDF files
Rights and permissions
This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/
About this article
Further reading

1.
The art of designing carbon allotropes
Frontiers of Physics (2019)

2.
Tightbinding model for optoelectronic properties of pentagraphene nanostructures
Scientific Reports (2018)

3.
Artificial boundary conditions for outofplane motion in pentagraphene
Acta Mechanica Sinica (2017)
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.