Enhanced Photocatalytic Water Splitting in a C2NMonolayer by C-Site Isoelectronic SubstitutionM.R. Ashwin Kishore[a, b] and P. Ravindran*[a, b, c, d]

1. Introduction

Photocatalytic water splitting has been a research focus sincethe pioneering work of Fujishima and Honda on the evolutionof hydrogen using TiO2-supported Pt electrodes.[1] Hydrogen(H2) fuel production through sunlight-driven water splittinghas the potential to be a clean energy source and a sustainablesolution to the depletion of fossil fuels and our environmentalproblems. However, the currently available photocatalysts havevery low hydrogen production efficiency. Hence, there is anurgent demand to put in greater efforts to design and fabri-cate catalytic materials with excellent selectivity, long term sta-bility, and cost effectiveness. Exploiting earth-abundant andlow-cost photocatalysts is a promising way to reduce the costof hydrogen production when compared with the use of fossil-based fuels. Extensive research has been devoted to improving

photocatalytic H2 evolution activity and yet the relatively low-energy conversion efficiency limits its practical applications.[2–5]

The fundamental prerequisites for enhancing the photocatalyt-ic conversion efficiency are to enable the semiconductors toharvest the more abundant visible light by modifying theirband gap to around 1.5 to 2.0 eV, suppressing the recombina-tion rate of photogenerated charge carriers by efficient separa-tion of electrons and holes, and tuning the band edge posi-tions with respect to water redox potentials that are favorablefor the overall water splitting.

Notably, a promising recent advancement in the field ofphotocatalytic water splitting has been in utilizing nanoscalematerials because of their tunable shapes and dimensions,thus resulting in interesting properties and phenomena thatare distinct from their bulk counterparts. Among variousclasses of nanomaterials, two-dimensional (2D) materials havedrawn much attention because of their unique thickness-de-pendent electronic and optical properties. Many 2D materialshave been predicted and also synthesized after the experimen-tal realization of graphene because the band gap shortcom-ings of graphene limits its practical applications to high-speedswitching devices, photocatalysts, etc.[6] Other 2D materialssuch as transition-metal dichalcogenides (TMDs),[7–10] phosphor-ene,[11, 12] Group III–VI monochalcogenides,[13] MXene,[14–16] andg-C3N4

[17–19] provide unique platforms to enhance the photoca-talytic activity because of their large surface area, which pro-vides more active sites for catalytic reactions. Furthermore, thedistance that photogenerated electrons and holes have totravel to the reactive sites is shorter for 2D materials. There-fore, the electron–hole recombination is greatly suppressed,

Two-dimensional (2D) semiconductors have shown greatpromise as efficient photocatalysts for water splitting. Tailoringthe band gap and band edge positions are the most crucialsteps to further improve the photocatalytic activity of 2D ma-terials. Here, we report an improved photocatalytic water split-ting activity in a C2N monolayer by isoelectronic substitutionsat the C-site, based on density functional calculations. Our op-tical calculations show that the isoelectronic substitutions sig-nificantly reduce the band gap of the C2N monolayer and thusstrongly enhance the absorption of visible light, which is con-sistent with the observed redshift in the optical absorptionspectra. Based on the HSE06 functional, the calculated band

edge positions of C2�xSixN and C2�xGexN monolayers are evenmore favorable than the pristine C2N monolayer for the overallphotocatalytic activity. On the other hand, for the C2�xSnxNmonolayer, the conduction band minima is more positive thanthe oxygen reduction potential and, hence, Sn substitution inC2N is unfavorable for the water decomposition reaction. In ad-dition, the isoelectronic substitutions improve the separationof e�–h+ pairs, which, in turn, suppress the recombinationrate, thereby leading to enhanced photocatalytic activity inthis material. Our results imply that Si-, and Ge-substituted C2Nmonolayers will be a promising visible-light photocatalysts forwater splitting.

[a] M. A. Kishore, Prof. P. RavindranDepartment of Physics, Central University of Tamil NaduThiruvarur, Tamil Nadu, 610101 (India)E-mail : raviphy@cutn.ac.in

[b] M. A. Kishore, Prof. P. RavindranSimulation Center for Atomic and Nanoscale MATerials (SCANMAT)Central University of Tamil NaduThiruvarur, Tamil Nadu, 610101 (India)

[c] Prof. P. RavindranDepartment of Materials Science, Central University of Tamil NaduThiruvarur, Tamil Nadu, 610101 (India)

[d] Prof. P. RavindranCenter for Materials Science and Nanotechnology and Department ofChemistryUniversity of OsloBox 1033 Blindern, 0315 Oslo (Norway)

The ORCID identification number(s) for the author(s) of this article canbe found under https://doi.org/10.1002/cphc.201700165.

ChemPhysChem 2017, 18, 1 – 8 � 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim1 &

These are not the final page numbers! ��These are not the final page numbers! ��

ArticlesDOI: 10.1002/cphc.201700165

which is beneficial for enhancing the photocatalytic watersplitting.

Recently, a new class of 2D materials that display a widerange of electronic, optical, and mechanical properties hasbeen explored and many are potential candidates for photoca-talytic water splitting application.[20, 21] One such material is ni-trogenated holey graphene, which was synthesized by Mah-mood et al.[22] This novel layered 2D structure possesses evenlydistributed holes and nitrogen atoms with a C2N stoichiometryin its basal plane. It is a direct band gap semiconductor andthe optically measured band gap value is 1.96 eV. The fabricat-ed field-effect transistor based on C2N multilayers exhibitsa high on/off ratio of 107. Given the direct band gap andporous nature, it is expected to have significant impact on var-ious applications ranging from nanoscale electronics, optoelec-tronics, catalysis, energy conversion and storage, gas storage,biological application, etc.

Sahin[23] reported that the C2N monolayer has quite high-fre-quency phonon modes that are close to that of graphene, indi-cating that the C2N monolayer has high structural stability.Zhang et al.[24, 25] theoretically predicted that the band gap andband edge positions of C2N can be tuned by varying theirstacking order, layer number, and external electric field for thespontaneous water splitting under visible light. Heterostruc-tured g-C3N4/C2N[26] and C2N/graphene[27] photocatalystshowed an enhanced efficiency because of the efficient separa-tion of photogenerated electron–hole pairs and favorableband edge positions for water splitting. From computationalstudies, Guan et al.[28] reported the strain tuning of the elec-tronic band structure and optical properties of the C2N mono-layer. Theoretical reports have shown that the magnetism ofthe C2N monolayers can be tuned by transition-metal embed-ding into the cavity of the porous C2N monolayer[29] and carrierdoping.[30] Earlier reports have shown that the C2N monolayeris an excellent candidate for gas separation because of its uni-formly distributed subnanometer pores with designable pore

sizes.[31–33] Furthermore, C2N has been employed for wide-spread applications such as single-atom catalysts,[34, 35] hydro-gen evolution reactions (HER),[36] oxygen evolution reaction(OER) catalysts,[37] and water desalination.[38]

To design more efficient C2N photocatalysts, one should usea strategy that would lower the band gap and expand thelight absorption further into the visible range, while ensuringthat the band edges straddle the redox potentials of water.One promising way to make this possible is by chemical modi-fication. In general, isovalent impurities do not induce deepdefect energy levels in the band gap but they can modify elec-tronic properties, which is beneficial for the solar energy con-version efficiency. Herein, we show an enhanced photocatalyticactivity in C2N by isoelectronic substitutions at the carbon-sitebased on first-principles calculations. We find that both Si- andGe-substitutions significantly reduce the band gap of the C2Nmonolayer and thereby strongly enhance the absorption ofvisible light. Furthermore, the band edge positions are evenmore favorable than the pristine C2N for the overall photocata-lytic activity. The factors that predominantly determine thephotocatalytic efficiency is photoinduced e�–h+ separationsand migration. In our case, the isoelectronic substitutions im-prove the separation of e�–h+ pairs, which leads to an en-hanced photocatalytic activity in this material.

2. Results and Discussion

Figure 1 shows a schematic structure of an optimized C2Nmonolayer. It is completely different from graphene—the twobenzene rings are bridged by pyrazine rings, which consist ofa six-membered D2h ring with two nitrogen atoms facing eachother. Unlike graphene, the presence of uniform holes and ni-trogen atoms in the C2N layer is expected to open up the gap.The primitive unit cell consists of 12 carbon atoms and 6 nitro-gen atoms and uniform holes in the C2N layer. The optimizedequivalent lattice parameter of monolayer C2N is 8.324 �, with

Figure 1. Left panel : Schematic structure of the C2N monolayer in the top view. The unit cell is indicated by dashed blue lines. The atoms are denoted as C(black) and N (gray). Right panel : Band structure of the C2N monolayer calculated by using the GGA-PBE functional.

ChemPhysChem 2017, 18, 1 – 8 www.chemphyschem.org � 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim2&

�� These are not the final page numbers!�� These are not the final page numbers!

Articles

in-plane covalent bond lengths of 1.429/1.468 � for the C�Cbond and 1.336 � for the C�N bond, which are consistent withprevious experimental[22] and theoretical reports.[25, 28] The cal-culated band structure obtained by using the GGA-PBE func-tional is depicted in Figure 1, which shows that the C2N mono-layer is a direct band gap semiconductor with band gap valueof 1.66 eV at the G point. This band gap value is much smallerthan the value obtained from optical measurements(1.96 eV)[22] and the value calculated theoretically[25] by theHeyd–Scuseria–Ernzerhof (HSE06) functional because of theunderestimation of the Kohn–Sham treatment in DFT.[39] Thus,for the present study, we have employed the HSE06 functionalto calculate all electronic structures. To simulate C2�xMxN (M =

Si, Ge, Sn) alloy, we have replaced one C atom in the C2N unitcell with an isovalent atom such as Si, Ge, and Sn, which corre-sponds to a dopant concentration of x = 0.083. The band struc-tures of pristine and C2�xMxN (M = Si, Ge, Sn) monolayers com-puted with the HSE06 functional are shown in Figure 2. It is

apparent that the pristine and C2�xMxN monolayers belong todirect band gap semiconductors with their valence band maxi-mum (VBM) and conduction band minimum (CBM) located atthe G point. The calculated band gap values are 2.47, 1.52,1.53, and 1.43 eV for C2N, C2�xSixN, C2�xGexN, and C2�xSnxNmonolayers, respectively. This significant reduction in the bandgap due to substitution leading to a redshift in the optical ab-sorption will be discussed below. This observed redshift in theoptical absorption greatly favors the absorption of more abun-dant visible light and hence photoconversion efficiency.

On closer examination of the band structure, the pristineC2N monolayer has two bands with different dispersions butthey are degenerate at the G point in the VBM. It is wellknown that in zinc blende type semiconductors, two degener-ate bands in the valence band edge are indexed as heavy holeand light hole bands to distinguish them. Therefore, we followsuch a description to describe these bands. The band withlower energy along the G-M direction is well dispersed andhence we call it the light hole (LH) band and the top-most va-lence band has less dispersion, which will result in heavy holeeffective mass so we call it the heavy hole (HH) band. In CBM,three bands are degenerate at the G point and, among thesethree bands, the lowest energy band shows a flat band naturealong G-M. This originates from the p state, contributed by thelocalized p orbitals of the nitrogen atoms.[22] However, thestrain in the lattice induced by isoelectronic substitutions splitthe HH and LH bands in VBM and also the degenerate bandsin CBM. More interestingly, the flat band along G-M in theCBM becomes more dispersive when we introduce isoelectron-ic atoms because of the increase in overlap interaction by thelarger size of the substituents. This indicates that the effectivemass would be lower than pristine C2N monolayer and hencethe transfer of electrons to the reactive sites are more efficientin isoelectronic substituted C2N monolayers. Moreover, C-sitesubstitutions remove the degeneracy of CBM and bands at thehigh symmetry points are well separated. Apart from a fewchanges in the band dispersion by C-site substitution, thedirect band behavior is preserved. However, the band gap getssystematically reduced and hence optical absorption startsmore in the visible range of the solar spectrum; a point thatwill be discussed in detail later. Nevertheless, the calculatedband gap values for C-site substituted systems are alwayshigher than the free energy of water splitting of 1.23 eV, andthis ensures the suitability of these materials for photocatalyticwater splitting. The observed direct band gap nature in thesematerials tends to result in better optical absorptivity than theindirect ones, because the indirect transition of electrons mustbe mediated by phonons.[40] Moreover, the large band gap (ca.1 eV) reduction without changing the direct band gap behav-ior in the C-site substituted C2N monolayers may also pave theway for the use of these materials in photovoltaic applications.

To understand the character of electrons involved in theVBM and CBM and also the origin of band gap narrowing forthe isoelectronic substitution at the C-site in C2N monolayers,the total and partial density of states (PDOS) were computed.The calculated total and orbital projected density of statesgiven in Figure 3 show that the VBM of the pristine C2N mono-layer mainly originates from the hybridization of C-2p and N-2p states (46 %), whereas the CBM is dominated by N-2p states(27 %) with a noticeable contribution from C-2p states (13.4 %).In the case of C2�xSixN, C2�xGexN, and C2�xSnxN monolayers,VBM mainly originates from the hybridization of p states of C,N, and isoelectronic substituents, whereas the CBM is mainlycontributed to by the hybridization of N-2p and p states of iso-electronic substituents with noticeable contribution from the dstates of isoelectronic substituents. The energetically degener-ate nature of DOS distribution coming from C-2p and N-2p,

Figure 2. The calculated band structure of C2N, C2�xSixN, C2�xGexN, andC2�xSnxN monolayers obtained by using the HSE06 functional are shown in(a), (b), (c), and (d), respectively. The Fermi level is set to zero.

ChemPhysChem 2017, 18, 1 – 8 www.chemphyschem.org � 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim3 &

These are not the final page numbers! ��These are not the final page numbers! ��

Articles

along with its strong hybridization feature in the whole va-lence band, indicates a strong covalent character, which is typi-cal for layered 2D materials. Notably, broad band features arefound at the band edges because of this covalent bonding. Ashas been noted earlier, isoelectronic substitutions do notinduce any defect levels in the band gap. Consistent with theabove view, no localized states are present in the DOS of C-sitesubstitution in the C2N monolayer, which can act as recombi-nation centers for the photogenerated electron–hole pairs. In-stead, this C-site substitution modifies the electronic structureand especially broadens the band edges, which significantlynarrow the overall band gap suitably for photocatalytic watersplitting. This band gap reduction is attributed to the broaden-ing of the band edges because of the hybridization interac-tions mentioned above, along with the shift of the CBM toa lower energy.

To probe the photocatalytic activity of pristine and C-sitesubstituted C2N monolayers, the band edge positions of VBMand CBM with respect to the redox potentials of water werecalculated. Here, we adopted the commonly used value for ox-idation and reduction potential of water at pH = 0 and thoseare �5.67 eV and �4.44 eV, respectively, with respect to thevacuum level.[41] The energy values of CBM/VBM and the localvacuum level of C2�xMxN (M = Si, Ge, Sn) monolayers werebased on DFT calculations. The absolute energy positions ofCBM/VBM were then obtained by shifting the local vacuumlevel to zero energy. Figure 4 shows the band edge alignmentof C2N, C2�xSixN, C2�xGexN, and C2�xSnxN monolayers with re-spect to redox potentials of water. As can be seen, the bandedge positions of the pristine C2N monolayer ensures that theredox reactions are thermodynamically favorable for water

splitting, which is consistent with other works.[24] Notably, theCBM is located 0.28 eV higher than the proton reduction po-tential (H+/H2). However, the VBM is 0.96 eV lower than theoxygen reduction potential (O2/H2O), which could permit trans-fer of holes with a low driving force and this would result inan imbalance between the reduction and oxidation reactionsthat will further prevent the process of the overall redox reac-tion. In contrast, the CBM and VBM in a C2�xSixN monolayer arelocated 0.13 eV higher than the proton reduction potentialand 0.16 eV lower than the oxygen reduction potential, respec-tively, and for the C2�xGexN monolayer, the CBM and VBM arelocated 0.24 eV higher than the proton reduction potentialand 0.07 eV lower than the oxygen reduction potential, respec-tively. This clearly leads to high photocatalytic efficiency be-cause of the smooth transfer of charge carriers to the waterredox species. In the case of the C2�xSnxN monolayer, the VBM

Figure 3. The calculated density of states for C2N, C2�xSixN, C2�xGexN, and C2�xSnxN monolayers obtained by using the HSE06 functional are shown in (a), (b),(c), and (d), respectively. The Fermi level is set to zero.

Figure 4. Variation in the band edge positions by isoelectronic substitutionin the C-site of C2N obtained by using the HSE06 functional. The dashedlines indicate water redox potentials at pH 0.

ChemPhysChem 2017, 18, 1 – 8 www.chemphyschem.org � 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim4&

�� These are not the final page numbers!�� These are not the final page numbers!

Articles

is more positive (0.21 eV) than the oxygen reduction potential,implying that it is unsuitable for use in photocatalytic watersplitting. It is worthwhile to mention that the Si and Ge substi-tutions not only reduce the band gap suitable for photocata-lytic water splitting, but also align the band edge positions ap-propriately to make the photocatalytic water redox reactionmore efficient. Our band alignment calculations clearly showthat Si- and Ge-substituted C2N monolayers have an evenmore favorable band edge position for water splitting than thepristine C2N monolayer, which would subsequently cater to theenhancement of the photocatalytic activity of these materials.

The photoconversion efficiency of semiconductors for watersplitting depends not only on the band edge alignment withrespect to water redox potential but also with the optical ab-sorption.[42] Therefore, to understand the role of C-site substitu-tion in the optical absorption of C2N, we have calculated theoptical absorption spectra for C2N, C2�xSixN, C2�xGexN, andC2�xSnxN monolayers. Considering the band gap underestima-tion of the PBE functional, we compensate for it by a rigid shiftof the absorption curve by the band gap value obtained fromthe HSE06 functional.[44] As shown in Figure 5, the optical ab-sorption for the pristine case increases significantly withphoton energy over the range of visible light and reachesa maximum absorption at 3.2 eV. This finding is consistent withprevious results obtained from DFT calculations.[28] This absorp-tion peak at 3.2 eV arises from the optical interband transitioncoming from the doubly degenerate flat band present in thevalence band at �1 eV, to a flat band feature present in theCBM. However the C2�xSixN, C2�xGexN, and C2�xSnxN monolayerspossess absorption edges that extend more into the visibleregion, as evident from Figure 5. The observed redshift in theabsorption edge for isoelectronic C-site substituted monolay-ers is attributed to the band gap reduction. Hence, the isoelec-

tronic substituted materials could efficiently harvest a broadrange of frequencies of visible light in the solar spectrum. Al-though the C-site substitutions of isoelectronic elements sys-tematically reduce the band gap, which enhances the opticalabsorption in the visible region, when one considers the bandalignment with respect to the water redox potential for thesecompounds, it supports our contention that Si- and Ge-substi-tuted C2N monolayers are promising photocatalysts for watersplitting.

Finally, to explore the separation efficiency of photogenerat-ed electrons and holes, which is another crucial prerequisitefor an efficient photocatalyst, we have illustrated the isosurfa-ces of band decomposed charge density corresponding toCBM and VBM at the G point in Figure 6. Both the VBM andCBM of the C2N monolayer are mainly composed of the nitro-gen Pz states and the C�C antibonding p states. Our band de-

Figure 5. Optical absorption spectrum calculated by using the PBE function-al followed by a rigid energy shift to take into account the band gap under-estimation of the PBE functional.

Figure 6. The calculated charge density corresponding to CBM and VBM of C2N, C2�xSixN, C2�xGexN, and C2�xSnxN monolayers are shown in (a), (b), (c), and (d),respectively. The isovalue for visualization is 0.008 e ��3. Drawings were produced by using VESTA visualization software.[43]

ChemPhysChem 2017, 18, 1 – 8 www.chemphyschem.org � 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim5 &

These are not the final page numbers! ��These are not the final page numbers! ��

Articles

composed charge density analysis shows that VBM of C2�xSixN,C2�xGexN, and C2�xSnxN monolayers distinctly comprise the s/pstates of C whereas the CBM is mainly dominated by s/p statesof N atoms. Hence, the hole will reside at the C site and theelectron will reside at the N site, so both the electrons andholes are spatially well separated (1.34 �) and this extends thelifetime of the carriers. If the lifetime of the carriers are veryhigh, then the recombination rate will be low and this wouldfurther improve the photocatalytic water splitting efficiency.

3. Conclusions

The photocatalytic activity of pristine and isoelectronic atoms(Si, Ge, and Sn) substituted at the C-site in the C2N monolayerhave been investigated under the framework of density func-tional theory. The calculated HSE06 band structure of the isoe-lectronic C-site substituted C2N monolayers show a significantreduction in the band gap, which is suitable to harvest a rela-tively large amount of visible light. The calculated band gapvalues and band edge positions indicate that the C2�xSixN andC2�xGexN monolayers are promising candidates for water split-ting applications compared with the pristine C2N monolayer,whereas the C2�xSnxN monolayer is unsuitable for the use inwater splitting because its VBM is more positive than thewater oxidation potential. The observed redshift in the opticalabsorption spectra for the isoelectronic C-site substituted C2Nmonolayers enhances the utilization of more abundant visiblelight than the pristine C2N and this further improves the hydro-gen production efficiency. The band decomposed charge den-sity analysis shows that the isoelectronic C-site substitutionsimprove the spatial separation of e�–h+ pairs, which, in turn,suppress the recombination rate, thereby enhancing the pho-tocatalytic activity. Our results support the idea that both theband gap and band edge positions of the C2N monolayer canbe tuned by substituting isoelectronic atoms such as Si or Gefor spontaneous photocatalytic water splitting.

Computational Details

Calculations were performed by using the projector augmentedwave (PAW) method, which is implemented in the Vienna ab InitioSimulation Package (VASP).[45] For the structural relaxation, we haveemployed the generalized gradient approximation proposed byPerdew, Burke, and Ernzrhof (GGA-PBE)[46] with van der Waals(vdW) correction proposed by Grimme because of its good descrip-tion for long-range vdW interactions.[47] The Heyd–Scuseria–Ernzer-hof (HSE06) hybrid functional[48] was used to calculate all electronicstructures because of the band gap underestimation of semicon-ductors by the PBE functional. The Brillouin zone was sampled byusing the Monkhorst–Pack scheme[49] and employs a 5 � 5 � 1 k-mesh for geometry optimization and a 3 � 3 � 1 k-mesh for themore expensive HSE06 calculations. A plane-wave basis set withcutoff energy of 520 eV was used. A large vacuum of ca. 20 � wasused to avoid interaction between two adjacent periodic images.The convergence criterion for energy was taken to be 10�5 eV andthe Hellmann–Feynman force acting on each atom was less than0.01 eV ��1 upon ionic relaxation.

To calculate the optical absorption spectrum of pristine and isoe-lectronic substituted C2N, the frequency-dependent dielectric func-tion e(w) =e1(w) + ie2(w) was calculated. The interband contributionto the imaginary part of the dielectric functions e2(w) was calculat-ed by summation over empty states by using Equation (1):[50]

eab2 ðwÞ ¼

4p2e2

Wlimq!0

1q2

X

c;v;k

2wkdðeck � evk � wÞ � hmckþqeajmvkihmvkjmckþqeb

i*

ð1Þ

where the indices c and v refer to conduction and valence bandstates, respectively. mck represents the cell periodic part of thewavefunctions at the k-point. The real part of the components ofthe dielectric tensor e1(w) were then calculated by using theKramer–Kronig transformation. The absorption coefficient a(w) wasevaluated according to Equation (2):[51]

aðwÞ ¼ffiffiffi2p

wffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffie2

1ðwÞ þ e22ðwÞ

q� e1ðwÞ

� �12

ð2Þ

where e1(w) and e2(w) are the real and imaginary parts of the fre-quency dependent complex dielectric function e(w), respectively.To account for the tensor nature of the dielectric function, e1(w)and e2(w) were averaged over three polarization vectors (along x, y,and z directions).

Acknowledgements

The authors are grateful to the Research Council of Norway forcomputing time on the Norwegian supercomputer facilities. Thisresearch was supported by the Indo-Norwegian Cooperative Pro-gram (INCP) through UGC Grant No. F.No.58-12/2014(IC) and De-partment of Science and Technology, India through Grant No. SR/NM/NS-1123/2013. M.R.A.K thanks Xu Zhang and Zhen Zhou forcomputation-practical help.

Keywords: density functional calculations · electronicstructure · graphene derivatives · photophysics · watersplitting

[1] A. Fujishima, K. Honda, Nature 1972, 238, 37.[2] Y. Tachibana, L. Vayssieres, J. R. Durrant, Nat. Photonics 2012, 6, 511 –

518.[3] M. G. Walter, E. L. Warren, J. R. McKone, S. W. Boettcher, Q. Mi, E. A. San-

tori, N. S. Lewis, Chem. Rev. 2010, 110, 6446 – 6473.[4] R. E. Blankenship, D. M. Tiede, J. Barber, G. W. Brudvig, G. Fleming, M.

Ghirardi, M. R. Gunner, W. Junge, D. M. Kramer, A. Melis, T. A. Moore, C.Moser, D. G. Nocera, A. J. Nozik, D. R. Ort, W. W. Parson, R. C. Prince, R. T.Sayre, Science 2011, 332, 805 – 809.

[5] F. E. Osterloh, Chem. Soc. Rev. 2013, 42, 2294 – 2320.[6] X. Li, Y. Dai, M. Li, W. Wei, B. Huang, J. Mater. Chem. A 2015, 3, 24055 –

24063.[7] A. K. Geim, I. V. Grigorieva, Nature 2013, 499, 419 – 425.[8] Q. H. Wang, K. Kalantar-Zadeh, A. Kis, J. N. Coleman, M. S. Strano, Nat.

Nanotechnol. 2012, 7, 699 – 712.[9] Z. Zeng, Z. Yin, X. Huang, H. Li, Q. He, G. Lu, F. Boey, H. Zhang, Angew.

Chem. Int. Ed. 2011, 50, 11093 – 11097; Angew. Chem. 2011, 123, 11289 –11293.

[10] K.-G. Zhou, N.-N. Mao, H.-X. Wang, Y. Peng, H.-Li. Zhang, Angew. Chem.2011, 123, 11031 – 11034.

[11] E. S. Reich, Nature 2014, 506, 19.[12] H. O. Churchill, P. J. Herrero, Nat. Nanotechnol. 2014, 9, 330 – 331.

ChemPhysChem 2017, 18, 1 – 8 www.chemphyschem.org � 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim6&

�� These are not the final page numbers!�� These are not the final page numbers!

Articles

[13] H. L. Zhuang, R. G. Hennig, Chem. Mater. 2013, 25, 3232 – 3238.[14] M. Naguib, V. N. Mochalin, M. W. Barsoum, Y. Gogotsi, Adv. Mater. 2014,

26, 992 – 1005.[15] J.-C. Lei, X. Zhang, Z. Zhou, Frontiers of Physics 2015, 10, 276 – 286.[16] M. Naguib, M. Kurtoglu, V. Presser, J. Lu, J. Niu, M. Heon, L. Hultman, Y.

Gogotsi, M. W. Barsoum, Adv. Mater. 2011, 23, 4248 – 4253.[17] X. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J. M. Carlsson, K.

Domen, M. Antonietti, Nat. Mater. 2009, 8, 76 – 80.[18] Y. Zheng, L. Lin, B. Wang, X. Wang, Angew. Chem. Int. Ed. 2015, 54,

12868 – 12884; Angew. Chem. 2015, 127, 13060 – 13077.[19] S. Yin, J. Han, T. Zhou, R. Xu, Catal. Sci. Technol. 2015, 5, 5048 – 5061.[20] S. Z. Butler, S. M. Hollen, L. Cao, Y. Cui, J. A. Gupta, H. R. Guti�rrez, T. F.

Heinz, S. S. Hong, J. Huang, A. F. Ismach, E. Johnston-Halperin, M. Kuno,V. V. Plashnitsa, R. D. Robinson, R. S. Ruoff, S. Salahuddin, J. Shan, L. Shi,M. G. Spencer, M. Terrones, W. Windl, J. E. Goldberger, ACS Nano 2013,7, 2898 – 2926.

[21] G. R. Bhimanapati, Z. Lin, V. Meunier, Y. Jung, J. Cha, S. Das, D. Xiao, Y.Son, M. S. Strano, V. R. Cooper, L. Liang, S. G. Louie, E. Ringe, W. Zhou,S. S. Kim, R. R. Naik, B. G. Sumpter, H. Terrones, F. Xia, Y. Wang, J. Zhu, D.Akinwande, N. Alem, J. A. Schuller, R. E. Schaak, M. Terrones, J. A. Robin-son, ACS Nano 2015, 9, 11509 – 11539.

[22] J. Mahmood, E. K. Lee, M. Jung, D. Shin, I. Y. Jeon, S. M. Jung, H. J. Choi,J. M. Seo, S. Y. Bae, S. D. Sohn, N. Park, J. H. Oh, H. J. Shin, J. B. Baek, Nat.Commun. 2015, 6, 6486.

[23] H. Sahin, Phys. Rev. B 2015, 92, 085421.[24] R. Zhang, J. Yang, arXiv:1505.02768, 2015.[25] R. Zhang, B. Li, J. Yang, Nanoscale 2015, 7, 14062 – 14070.[26] H. Wang, X. Li, J. Yang, ChemPhysChem 2016, 17, 2100 – 2104.[27] D. Wang, D. Han, L. Liu, L. Niu, RSC Adv. 2016, 6, 28484 – 28488.[28] S. Guan, Y. Cheng, C. Liu, J. Han, Y. Lu, S. A. Yang, Y. Yao, Appl. Phys. Lett.

2015, 107, 231904.[29] J. Du, C. Xia, W. Xiong, X. Zhao, T. Wang, Y. Jia, Phys. Chem. Chem. Phys.

2016, 18, 22678 – 22686.[30] Z. Liang, B. Xu, H. Xiang, Y. Xia, J. Yin, Z. Liu, RSC Adv. 2016, 6, 54027 –

54031.

[31] L. Zhu, Q. Xue, X. Li, T. Wu, Y. Jin, W. Xing, J. Mater. Chem. A 2015, 3,21351 – 21356.

[32] B. Xu, H. Xiang, Q. Wei, J. Liu, Y. Xia, J. Yin, Z. Liu, Phys. Chem. Chem.Phys. 2015, 17, 15115 – 15118.

[33] Y. Qu, F. Li, H. Zhou, M. Zhao, Sci. Rep. 2016, 6, 19952.[34] D. W. Ma, Q. Wang, X. Yan, X. Zhang, C. He, D. Zhou, Y. Tang, Z. Lu, Z.

Yang, Carbon 2016, 105, 463 – 473.[35] B. L. He, J. S. Shen, Z. X. Tian, Phys. Chem. Chem. Phys. 2016, 18, 24261 –

24269.[36] J. Mahmood, S. M. Jung, S. J. Kim, J. Park, J. W. Yoo, J. B. Baek, Chem.

Mater. 2015, 27, 4860 – 4864.[37] X. Li, W. Zhong, P. Cui, J. Li, J. Jiang, J. Phys. Chem. Lett. 2016, 7, 1750 –

1755.[38] Y. Yang, W. Li, H. Zhou, X. Zhang, M. Zhao, Sci. Rep. 2016, 6, 29218.[39] J. P. Perdew, Int. J. Quantum Chem. 1985, 28, 497 – 523.[40] M. Dong, J. Zhang, J. Yu, APL Materials 2015, 3, 104404.[41] S. Trasatti, Pure Appl. Chem. 1986, 58, 955 – 966.[42] J. Wang, Q. Meng, J. Huang, Q. Li, J. Yang, J. Chem. Phys. 2014, 140,

174705.[43] K. Momma, F. Izumi, J. Appl. Crystallogr. 2011, 44, 1272.[44] R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki, Y. Taga, Science 2001, 293, 269.[45] G. Kresse, J. Furthm�ller, Comp. Mater. Sci. 1996, 6, 15.[46] J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1997, 78, 1396.[47] S. Grimme, J. Comput. Chem. 2006, 27, 1787.[48] J. Heyd, G. E. Scuseria, M. Ernzerhof, J. Chem. Phys. 2003, 118, 8207.[49] H. J. Monkhorst, J. D. Pack, Phys. Rev. B 1976, 13, 5188.[50] M. Gajdos, K. Hummer, G. Kresse, J. Furthm�ller, F. Bechstedt, Phys. Rev.

B 2006, 73, 045112.[51] M. Chhowalla, H. S. Shin, G. Eda, L. J. Li, K. P. Loh, H. Zhang, Nat. Chem.

2013, 5, 263.

Manuscript received: February 15, 2017Accepted Article published: April 5, 2017Final Article published: && &&, 0000

ChemPhysChem 2017, 18, 1 – 8 www.chemphyschem.org � 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim7 &

These are not the final page numbers! ��These are not the final page numbers! ��

Articles

ARTICLES

M. A. Kishore, P. Ravindran*

&& –&&

Enhanced Photocatalytic WaterSplitting in a C2N Monolayer by C-SiteIsoelectronic Substitution

Splitting cats: Density functional calcu-lations have shown that Si- and Ge-sub-stitution at the C-site in a C2N monolay-er enhances the photocatalytic watersplitting efficiency compared with pris-tine C2N by modifying the band gapand band edge to positions that aremore suitable for water splitting.

ChemPhysChem 2017, 18, 1 – 8 www.chemphyschem.org � 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim8&

�� These are not the final page numbers!�� These are not the final page numbers!