Journal ofMaterials Chemistry A

PAPER

A simple and effic

aDepartment of Chemistry, Beijing Key L

Instrumentation, Tsinghua University, Be

tsinghua.edu.cnbAcademy of State Administration of Grain

100037, ChinacEco-Materials and Renewable Energy R

Engineering and Applied Sciences, Nanji

Nanjing, Jiangsu 210093, China

† Electronic supplementary informationspectra, XRD patterns, and the elem10.1039/c4ta02781g

Cite this: J. Mater. Chem. A, 2014, 2,17521

Received 3rd June 2014Accepted 26th August 2014

DOI: 10.1039/c4ta02781g

www.rsc.org/MaterialsA

This journal is © The Royal Society of C

ient strategy for the synthesis of achemically tailored g-C3N4 material†

Xiaojuan Bai,ab Shicheng Yan,c Jiajia Wang,c Li Wang,a Wenjun Jiang,a Songling Wu,b

Changpo Sunb and Yongfa Zhu*a

Tailored nanostructures offer a new way to facilitate electron–hole separation and offer additional

opportunities to generate unique photocatalysts that demonstrate novel light absorption,

thermodynamic and kinetic properties. A simple and efficient approach to the synthesis of a large variety

of g-C3N4 tailored nanostructures is reported. Herein, NH3 and H2O2 were used as controllable chemical

scissors to tailor bulk g-C3N4 to a large variety of g-C3N4 nanostructures, which include exfoliated

porous, quantum dot, nanomites and nanospindles. The tailored g-C3N4 shows a photoreactivity of H2

evolution 3.0 (pure water) and 4.1 (saturated KCl solution) times higher than bulk g-C3N4 under l >

420 nm. We believe this strategy affords new opportunities for structural tuning of X-doped (X ¼ N, S, P,

and O) carbon materials, as well as their exploration in catalysis, organic synthesis, nanomedicine and

energy storage.

Introduction

Research on graphite carbon nitride (g-C3N4) and relatednanostructures has seen enormous increase in the level ofinterest in recent years.1 Nanostructured g-C3N4 is an appealingclass of nanomaterials to complement carbon in a variety ofapplications. g-C3N4 with different structures possesses not onlyintriguing properties including high hardness, low frictioncoefficient, and reliable chemical inertness, but also greatpotential for energy conversion and storage, and environmentalapplications.2,3 Although the electronic and optical properties ofbulk g-C3N4 are unique and potentially useful in future devices,there is currently no easily applied method for large-scale pro-cessing of tailored nanostructured g-C3N4 and related objectssuch as nanoporous sheets, quantum dots or other nano-g-C3N4

materials.1 According to the superiority of porous graphene andcarbon quantum dots when compared with conventional inor-ganic quantum dots and other carbon materials,4–10 as well asthe inherent unique properties of bulk g-C3N4 materials, thetailored g-C3N4 nanostructures are expected to have attractive

aboratory for Analytical Methods and

ijing, 100084, China. E-mail: zhuyf@

, No. 11 Baiwanzhuang Street, Beijing,

esearch Center (ERERC), College of

ng University, No. 22, Hankou Road,

(ESI) available: TEM images, FTIRental composition result. See DOI:

hemistry 2014

properties and so exhibit excellent performances in manyapplications.10,11 In this sense, an macroscopic and effectivemethod that can produce uniform tailored g-C3N4 with tunablesize on a large scale is in great need. However, papers on thepreparation of tailored g-C3N4 are limited, and further study onproperties and performances is more seldom reported. In mostof the previously reported g-C3N4 nanostructures, variousstructures and morphologies of g-C3N4 have been produced bypyrolysis of nitrogen-rich precursors to incorporate s-triazinerings or to generate them during synthesis, and a templatingmethod.12–16 Although good quality, nanoporous, nanosheets ornanospheres of g-C3N4 can be grown by this way, theseapproaches cannot easily be implemented for a series of varioussmall-sized products on a large-scale due to the intrinsic tech-nological difficulties. Recently, Sadhukhan et al. reportedmicrowave assisted bottom-up fabrication of g-C3N4 quantumdots from formamide (HCONH2), and the quantum dots wereproduced as an intermedium, which have a nonuniformthickness ranging from 0.5 to 2.5 nm and their width rangesfrom 5 nm to 20 nm.17 A convenient on-surface synthesis routewould thus be a powerful alternative for the bottom-up creationof g-C3N4 nanostructures.

A tailored method, which was found to induce the formationof the ordered morphologies,18,19 is a promising newmethod forsuch anisotropic chain- or stripe-like structures in the polymermatrix. Although a great number of investigations have beencarried out on the formation of shear-induced patterns, thistechnique was rarely used in the materials sciences becausethese shear-induced patterns were in a metastable state. Oncethe shear eld is changed or removed, so does the shear-induced patterns.20 Unlike the previous reports, in this work,

J. Mater. Chem. A, 2014, 2, 17521–17529 | 17521

Journal of Materials Chemistry A Paper

NH3 and H2O2 were used as controllable chemical scissors fortailored g-C3N4 which are easier and stable to process due to alower evaporation temperature and surface functional groupsresulting from a signicantly reduced p–p stacking, makingthem ideal candidates for the bottom-up production of tailoredg-C3N4 nanostructures. Being functional materials, the grainsize, surface area, surface chemistry, and crystallinity are ofimportance to performances, i.e. the adsorption and active sitesin the catalyst are generally related to dangling bonds, surfacegroups, and surface electron conguration, and all the abovefactors are related to the synthesis process.21,22 Using tunabletailored oxidation, graphene sheets can be easily stripped fromgraphite or nanotubes, and can be further tailored into poroussheets or quantum dots.20,23

Based on the above analysis, herein, NH3 and H2O2 wereused as controllable chemical scissors tunable for g-C3N4

materials which will be effective for further structural tailoring.The tailored method using NH3 and H2O2 as the chemicalscissors and the g-C3N4 nanostructures can be reasonablycontrolled by tuning the NH3/H2O2 stoichiometric proportion.The facile solution method capable of large scale production oftailored g-C3N4 is developed based on the selective cleavage of–NH– between C3N3 and C6N7 in g-C3N4 layers. This techniqueis readily scalable and conceptually different from thosereported for preparing g-C3N4 nanostructures. The tailoredg-C3N4 presents superior photocatalytic performances anduniformly distributed sizes and can be stably dispersed inaqueous media, which can be attributed to the high efficiencyof electron–hole separation and nanometer size effect oftailored g-C3N4.

Experimental sectionSynthesis of tailored g-C3N4 samples

Bulk g-C3N4. The bulk g-C3N4 photocatalysts were synthe-sized as described in a previous paper.2 Dicyandiamide (3 g)(Aldrich, 99%) in an open crucible was heated in static air with aramping rate of 2.3 �Cmin�1 to 550 �C where it was held for 4 h.The product was collected and ground to powder using an agatemortar for further characterization and performance measure-ments. It should be claimed that the widely used “g-C3N4” in theliterature is actually nonstoichiometric. Here we use “g-C3N4” todescribe the products just to be consistent with the generalusage.

Tailored g-C3N4. The tailored g-C3N4 was obtained bycontrollable tailoring of as-prepared bulk g-C3N4 in NH3 andH2O2 aqueous solutions. Typically, 50 mg of bulk g-C3N4

powder was dispersed in 100 mL H2O2 aqueous solution, andthen ultrasonically dispersed for about 3 hours and aerwardsthe resulted suspension was stirred for another 15 h. Thesuspension was then centrifuged at 1000 rmp to remove theresidual precipitation. Subsequently, excess NH3 was added inthe above suspension and stirred at 70 �C for a certain time. Atlast, the nal suspension was washed several times withdeionized water. In addition, the NH3/H2O2 ratio is one of thekeys for tuning tailored g-C3N4 nanostructure formation.

17522 | J. Mater. Chem. A, 2014, 2, 17521–17529

Characterization

Transmission electron microscopy (TEM) images were obtainedby using a JEOL JEM-2011F eld emission transmission elec-tron microscope with an accelerating voltage of 200 kV. Atomicforce microscopy (AFM) images were recorded by using a SPM-9700 scanning probe microscope (Shimadzu Corporation).X-ray diffraction (XRD) patterns of the powders were recorded atroom temperature by using a Bruker D8 Advance X-ray diffrac-tometer. The UV-vis absorption spectra of the suspension wererecorded in the range from 200 to 800 nm using a Hitachi U-3010 spectroscope. Fourier transform infrared (FTIR) spectrawere recorded using a Perkin-Elmer spectrometer in thefrequency range of 4000–450 cm�1 with a resolution of 4 cm�1.The Brunauer–Emmett–Teller (BET) surface area measurementswere performed by using a micromeritics (ASAP 2010 V5.02H)surface area analyzer. The nitrogen adsorption and desorptionisotherms were measured at 77 K aer degassing the sampleson a Sorptomatic 1900 Carlo Erba Instrument. Electrochemicaland photoelectrochemical measurements were performed in athree-electrode quartz cell with 0.1 M Na2SO4 electrolyte solu-tion. A platinum wire was used as a counter electrode and asaturated calomel electrode (SCE) was used as a referenceelectrode.

Photocatalytic experiments

Photocatalytic tests were carried out using a Pyrex top-irradia-tion reaction vessel connected to a closed glass gas system.Hydrogen production was performed by dispersing 50 mg ofcatalyst powders in an aqueous solution (100 mL) containingtriethanolamine (15 vol%) as a sacricial electron donor. Co-catalysts Pt nanoparticles were introduced by an in situ photo-deposition method, where 3 wt% (respect to Pt) H2PtCl6$6H2Owas added and well distributed in the reaction solution. Thereaction solution was evacuated several times to remove aircompletely prior to irradiation under a 300 W xeon-lampequipped with a 420 nm-cut-off lter. The temperature of thereaction solution was maintained at room temperature by a owof cool water during the reaction. The evolved gases wereanalyzed by using a gas chromatograph equipped with athermal conductive detector (TCD) and a 5 A molecular sievecolumn, using nitrogen as the carrier gas.

Theoretical simulation calculation

Our calculations are based on density functional theory (DFT),as implemented in the VASP32,33 with the projected augmentedwave (PAW34) method. Generalized gradient approximation(GGA35) in the scheme of Perdew–Bueke–Ernzerhof (PBE36) isused for the exchange correlation functional. For C and N, the2s22p2 and 2s22p3 orbitals, respectively, are included as valencestates. The cutoff energy is 500 eV. Geometry relaxations areperformed until the residual forces on each atom converged tobe smaller than 0.03 eV A�1.

This journal is © The Royal Society of Chemistry 2014

Fig. 1 The evolution process of the morphology for tailored g-C3N4. (a–l) TEM images of tailored g-C3N4 by treatment of different NH3 : H2O2

volume ratios: (a and d) NH3 : H2O2 ¼ 1 : 3; (b and e) NH3 : H2O2 ¼ 2 : 3; (c and f) NH3 : H2O2 ¼ 3 : 3; (g and j) NH3 : H2O2 ¼ 4 : 3; (h and k)NH3 : H2O2 ¼ 6 : 3; (i and l) NH3 : H2O2 ¼ 10 : 3.

Paper Journal of Materials Chemistry A

Results and discussionTailored g-C3N4 morphology

The variation of lateral size and morphology of g-C3N4 by NH3/H2O2 treatment was clearly observed by TEM measurement(Fig. 1).

The lateral size of bulk g-C3N4 prepared directly by thethermal polycondensation method was relatively large and

This journal is © The Royal Society of Chemistry 2014

thick. The size distribution ranges widely from ten to more thantwo hundred nanometers. Though the bulk g-C3N4 particles aremicrometers in size, they are in fact formed by closely com-pacted thin sheets (Fig. S1†). Aer adding the “chemicalscissor” NH3/H2O2 mixture, the lateral size of g-C3N4 wasdecreased and became uniform more and more with increasingconcentration of NH3/H2O2 (Fig. 1a and d). At NH3/H2O2¼ 2 : 3,the pore size distribution of g-C3N4 is around 15–25 nm (Fig. 1b

J. Mater. Chem. A, 2014, 2, 17521–17529 | 17523

Journal of Materials Chemistry A Paper

and e). Less than 3% g-C3N4 is over 30 nm. With increasingvolume ratio of NH3/H2O2 to 3 : 3, 4 : 3, 6 : 3 and 10 : 3, the sizedistribution varied to 10–15 nm, 100–200 nm, 50–100 nm, and400–500 nm, respectively (Fig. 1c, f, g, j, h, k, i and l). Aertailoring at NH3/H2O2 ¼ 2 : 3, pores start to appear on thesurface which are uniformly distributed in g-C3N4 nanosheets.By further increasing the tailor time, the small pores began tocleave to form the well-ordered quantum dots until they nallyform nanomites and then nanospindles.

As shown in Fig. 2, the lateral size of a uniform thickness wascalculated to be 2.5 nm and 0.7 nm according to the heightprole along AB and CD, corresponding to g-C3N4 nanoporoussheets for about 7 layers (Fig. 2a–c) and quantum dots for 2layers (Fig. 2e and g), respectively. The tailored g-C3N4

morphology strongly depends on the ratio of NH3/H2O2, whichdetermines the tailored degree. The g-C3N4 nanoporous sheetsare formed with micrometer size in large under mild oxidation,the average pore size is 20 nm and the pores are uniformlydistributed over g-C3N4 sheets. The narrow pore-size distribu-tion indicates that a large number of pores are around 4 and 15nm (Fig. 2d). Their Brunauer–Emmett–Teller (BET) specicsurface area (SBET) is calculated to be as high as 14.2 m2 g�1. It isabout 1.6 times as large as that of bulk g-C3N4 (8.5 m2 g�1) butless than the theoretical value. The tightly packing nature oftailored g-C3N4 may be responsible for the difference betweentheoretical and test values. The large total pore volume stronglysupports the fact that the as-prepared samples are highlynanoporous. The width of g-C3N4 quantum dots is 10 nm to15 nm under deep oxidation. Thus, the nearly transparentfeature of tailored g-C3N4 in aqueous media indicates its ultra-

Fig. 2 TEM and AFM images of tailored g-C3N4 nanoporous sheets (a–d)(CNPSs); (b) AFM image of CNPSs; (c) the corresponding height along ABimage of g-C3N4 quantum dots (CNQDs); (f) TEM image of tailored grap

17524 | J. Mater. Chem. A, 2014, 2, 17521–17529

small size (Fig. S2†). The tailored g-C3N4 was negatively charged,with zeta potentials of about �47.1 mV and �52.1 mV (Fig. S3and S4†) respectively. Beneting from the size and chargedsurface,24 the solutions are very stable, without detectableaggregation aer standing for even more than two weeks. Asshown in Fig. 2f, the tailored method is also effective in gra-phene pore formation.

Photocatalytic activity and tailored structure

The actual photocatalytic activity of the tailored g-C3N4

prepared at NH3/H2O2 ¼ 2 : 3 for hydrogen evolution undervisible light was measured in a water–triethanolamine solutionin the presence of 3 wt% Pt as a cocatalyst. As presented inFig. 3a, aer 8 hours, the average hydrogen evolution amount ofthe g-C3N4 nanosheets is 46 mmol, which is 3 times higher thanthat of bulk g-C3N4 (16 mmol). The hydrogen production rate canbe further increased to 66 mmol, a record rate for the carbonnitride materials, being 4.1 times higher than bulk g-C3N4 byusing a saturated salt solution of KCl to increase the active sitesfor hydrogen production. Electrochemical impedance spec-troscopy (Fig. 3b) studying the charge-transfer rate discloses theexpected semicircular Nyquist plots for both tailored g-C3N4

and bulk g-C3N4, but with a signicantly decreased diameter forthe former. This suggests that the former has a more effectiveseparation of photogenerated electron–hole pairs and fasterinterfacial charge transfer, thus comes with a signicantlyincreased photoreactivity.25 The crystal and chemical structuresof the g-C3N4 nanosheets were analyzed by diffuse reectanceabsorption spectra (DRS), X-ray photoelectron spectra (XPS) andX-ray diffraction (XRD) patterns. DRS of bulk g-C3N4 and

and quantum dots (e–g). (a) TEM image of g-C3N4 nanoporous sheetsof CNPSs; (d) pore size distribution of bulk g-C3N4 and CNPSs; (e) AFMhene oxide; (g) the corresponding height along CD of CNQDs.

This journal is © The Royal Society of Chemistry 2014

Fig. 3 (a) A typical time course of H2 production under visible light (l > 420 nm) over 3.0 wt% Pt-deposited tailored CNPS photocatalyst. (i) bulkg-C3N4 in water containing 15 vol% triethanolamine; (ii) CNPSs in water containing 15 vol% triethanolamine, (iii) CNPSs in saturated aqueoussolution of KCl containing 15 vol% triethanolamine. (b) EIS Nyquist plots of CNPSs (a and a0: light on; b and b0: dark) and bulk g-C3N4 (c and c0: lighton; d and d0: dark) with light on/off cycles under visible light ([Na2SO4 ¼ 0.1 M], l > 420 nm).

Paper Journal of Materials Chemistry A

tailored g-C3N4 were recorded and as shown in Fig. S5,† there isnot too obvious change in the absorption edge, compared tothat of bulk g-C3N4 powder. The absorption edge in the reec-tance spectra at approximately 459 nm suggests that the energyband gap of tailored g-C3N4 samples is similar to that of bulkg-C3N4. When the particle size reaches a certain degree, theband structure will be changed. In this work, the size of g-C3N4

is not enough to change the band structure. Therefore, thetailored g-C3N4 samples exhibit a similar light absorptionprole to the bulk g-C3N4.

Fig. 4 High-resolution XPS N1s spectra of bulk g-C3N4 and tailored g-C

This journal is © The Royal Society of Chemistry 2014

To further probe the chemical state of nitrogen in theresulting g-C3N4 nanosheets, we conducted the XPS measure-ments. As shown in Fig. 4, the high resolution N1s spectra canbe also deconvoluted into three different peaks at bindingenergies of z400.2 (N1), 398.5 (N2) and 404.2 eV (N3), respec-tively. The dominant N1 is commonly attributed to sp2 N atomsinvolved in triazine rings, while the medium N2 is assigned tobridging N atoms in N-(C)3 or N bonded with H atoms. The veryweak N3 can be assigned to the charging effects or positivecharge localization in heterocycles and the cyano-group.26–29 The

3N4.

J. Mater. Chem. A, 2014, 2, 17521–17529 | 17525

Journal of Materials Chemistry A Paper

decreased intensity of N2 shows that the less bridging N atomsexist in N-(C)3 or N bonded with H atoms aer tailored treat-ment. The disappeared N3 peak for tailored g-C3N4may indicatethat more negative charge is introduced by surface hydroxylgroups during the tailored process. The result indicated that the–NH–/–NH2 group decreases due to cleavage of chemical bond–NH– between C3N3 and C6N7.

The X-ray diffraction (XRD) pattern of the tailored g-C3N4

sample shows a decrease in intensity for line (002) (Fig. S6†),indicating that the tailored g-C3N4 tending to be assembled bythe xy-plane are in a good z-orientation compared with the bulkg-C3N4 powders.24 Remarkably enough, aer tailoring, theintensity of this (002) peak signicantly decreases, clearlydemonstrating that the layered g-C3N4 has been successfullytailored as we expected. This is consistent with the observationsfrom TEM and AFM images. The appearance of new peaks fortailored g-C3N4 compared to bulk g-C3N4 is strong evidence forthe creation of a new arrangement. Evidence for an in-planepattern is shown in the well-resolved peaks at 19.24 and 23.8�,which can be indexed as (110) and (200) of the in-planarpacking, respectively.30 The other tiny peaks could be attributedto the surface microstructure of tailored g-C3N4, and theassumption can be demonstrated by theoretical calculation.31–36

Carbon nitride (g-C3N4) network materials have been producedas disordered structures by precursor-based methods, whichmay contain two main construction isomers, triazine and hep-tazine units, not an ideal structure. And the proportion of themstrongly depends on the precursors and condensation processduring synthesis.31

The triazine and heptazine based on C3N4 are shown inFig. 5a and b, and denoted as tri-C3N4 and hep-C3N4, respec-tively. To simulate the combination of tri-C3N4 and hep-C3N4,we construct a 4 � 4 � 1 supercell of hep-C3N4 which contains192C and 256N atoms. Then, based on this supercell, the hep +trilow-C3N4 (Fig. 5c) and hep + trihigh-C3N4 (Fig. 5d) are con-structed, which correspond to the low and high concentrationsof tri-C3N4 in hep-C3N4, respectively. The theoretical X-raydiffraction (XRD) patterns are simulated using the PowderCell37

program package. Parameters used for simulation are Bragg–Branteno geometry and CuKa radiation. The XRD patterns ofhep-C3N4, tri-C3N4, hep + trilow-C3N4 and hep + trihigh-C3N4 areshown in Fig. 5e, respectively. It is seen that there were moresignicant peaks for the surface microstructure with more C3N3

units. As demonstrated by the XRD experimental result, thesurface microstructure is not observed in bulk g-C3N4, which isattributed that there may contains more C6N7 units. Aertailoring, the amount of C6N7 units becomes less, which meansthat the increment of C3N3 units is the major cause for theappearance of the surface microstructure.

To compare the stability among different g-C3N4 structures,the formation energy is calculated using the followingrelationship:

Ef ¼ Et � nCmC � nNmN

where Et is the total energy of different C3N4 structures, mi (i¼ Cor N) is the chemical potential of constituent i referenced to

17526 | J. Mater. Chem. A, 2014, 2, 17521–17529

elemental solid/gas, and ni is the number of ion i. In this study,mi (i ¼ C or N) simply use energy per atom in their corre-sponding elemental phases. According to this denition, thesmaller the formation energy is, the more stable the structurewill be. The formation energies of hep-C3N4, tri-C3N4, hep +trilow-C3N4 and hep + trihigh-C3N4 are listed in Table 1. It is seenthat (a) hep-C3N4 is the most stable structure, agreeing well withother work.38 (b) The formation energy of hep + trilow-C3N4 islarger than that of hep-C3N4 but smaller than that of tri-C3N4.This means that when the concentration of tri-C3N4 in hep-C3N4

is low, the combination of tri-C3N4 and hep-C3N4 is one properthermodynamically stable structure. (c) However, the formationenergy of hep + trihigh-C3N4 is larger than that of hep-C3N4 andtri-C3N4, suggesting that hep-C3N4 with high concentration oftri-C3N4 is thermodynamically unstable.

Proposed tailored mechanism

Based on the above analysis, there are two types of structuralisomers in g-C3N4 materials, triazines (C3N3) and heptazines(C6N7), and the proportion of them strongly depends on theprecursors and condensation process during synthesis.31 Theheptazines unit is energetically more stable than the triazinesunit, due to the different electronic environment of nitrogenatoms. Besides, it is reported that the vacancy-free directanalogue of graphite is energetically unstable.39 In this work,dicyandiamide (C2H4N4), a triazine-based compound, is used tosynthesize the bulk g-C3N4 rapidly. However, it was shown thatthe bulk reaction can hardly result in an ideal composition ofthe g-C3N4 material due to incomplete condensation or pyrol-ysis of C2H4N4. It is known that the electronic band structureand band gap, which are very important for the function of g-C3N4, depend on the condensation degree. It is underlined thatg-C3N4 is functionally not only a semiconductor but also a solidstate organic reagent with its surface-terminating amino groupsand lone-pair N heteroatoms of the poly(tri-s-triazine) frame-work.39 For example, bulk g-C3N4 solids perform rather poorly insome catalytic processes, while polymeric g-C3N4 with muchdisordered structure exhibits desirable activity. Structuraldefects or surface terminations seem to play a key role in thecatalytic activation. Therefore, it is valuable to cleave –NH–

between C3N3 and C6N7 to obtain more disordered andcomplete polymerized “g-C3N4 structure”. According to theelemental composition of g-C3N4 (Table S1, Fig. S7†), thesample has a H/C + N atomic ratio of 3/100. However, H/C + Natomic ratios of C3N3 and C6N7 are 0 and 1/15 respectively.Then, the NC3N3 : NC6H7 ratio is about 3 : 1, indicating that oneC6N7 unit is mostly surrounded by three C3N3 units. Then in asingle g-C3N4 layer, the C6N7 units inlay the C3N3 units.Combined with the value of bond length (C–N, C]N and N–H),the size of basic structural unit for tailored g-C3N4 is about 5nm, which means that the dimension of tailored g-C3N4

quantum dots is mainly distributed in 10–15 nm, correspond-ing to about 2–3 basic structural units. The calculation value isconsistent with the experimental result. Based on the aboveanalysis, the different nanostructures were formed by adifferent tailored degree of g-C3N4 which consists of C3N3 and

This journal is © The Royal Society of Chemistry 2014

Fig. 5 Calculated XRD pattern based on the supercell of tri-C3N4 and hep-C3N4: (a) supercell of tri-C3N4; (b) supercell of hep-C3N4; (c) lowconcentrations of tri-C3N4 in hep-C3N4; (d) high concentrations of tri-C3N4 in hep-C3N4; (f) calculated XRD pattern based on C3N4 of the abovestructural unit.

Table 1 Formation energies (Ef) of hep-C3N4, tri-C3N4, hep + trilow-C3N4 and hep + trihigh-C3N4

Structures Ef (eV)

Hep-C3N4 128.3Tri-C3N4 154.9Hep + trilow-C3N4 145.6Hep + trihigh-C3N4 156.7

Scheme 1 Schematic illustration of the tailored process from bulk g-treatment of different NH3/H2O2 volume ratios.

This journal is © The Royal Society of Chemistry 2014

Paper Journal of Materials Chemistry A

C6N7 by using NH3 and H2O2 as chemical scissors. According tothe tailored process,40,41 a possible formation mechanism fordifferent nanostructures is proposed in Scheme 1. H2O2 isinclined to heterolysis in an alkaline environment to form aperoxide hydroxyl anion, which would give C3N3 terminalgroups to imine and subsequent oxidation of imine terminalgroups, and then undergoes rearrangement to give C–OHterminal groups.42 Thus, the –NH– linking groups between C3N3

and C6N7 units are broken, and the morphology of the residue

C3N4 to two typical tailored morphologies (CNPSs and CNQDs) by

J. Mater. Chem. A, 2014, 2, 17521–17529 | 17527

Journal of Materials Chemistry A Paper

varies with the tailored degree of C3N3 and C6N7 units. Specif-ically, g-C3N4 nanoporous sheets are produced when mildoxidation was conducted while the quantum dots are obtainedwhen deep oxidation was proceeded. The tailored method hereis simpler, safer, and cheaper than previous approaches, whichgenerally involve various templates.

Conclusions

In conclusion, a chemically tailored method is developed whichis a cheap, convenient and effective approach for large synthesisof a large variety of g-C3N4 nanostructures by using NH3 andH2O2 as chemical scissors. It may be readily scalable to theindustrial level on both a theoretical and a practical level and isalso demonstrated to be effective in graphene pore formation.The tailored g-C3N4 nanostructures from the bulk g-C3N4

precursor can be well controlled by simply regulating the ratioof NH3/H2O2. Furthermore, the formation of new tailoredstructures improves the charge separation process, resulting inenhanced photocatalytic activity in hydrogen production. Inaddition to their promising application as a photocatalyst forwater splitting, we anticipate that tailored g-C3N4 becomesattractive for various potential applications, including catalysis,energy storage, nanomedicine, sensors and environmentalprotection.

Acknowledgements

This work was partly supported by the National Basic ResearchProgram of China, 973 Program (2013CB632403), National HighTechnology Research and Development Program of China(2012AA062701) and Chinese National Science Foundation(20925725 and 21373121).

Notes and references

1 Y. Zheng, J. Liu, J. Liang, M. Jaroniec and S. Z. Qiao, EnergyEnviron. Sci., 2012, 5, 6717–6731.

2 X. C. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin,J. M. Carlsson, K. Domen and M. Antonietti, Nat. Mater.,2009, 8, 76–80.

3 L. M. Zambov, C. Popov, N. Abedinov, M. F. Plass,W. Kulisch, T. Gotszalk, P. Grabiec, I. W. Rangelow andR. Kassing, Adv. Mater., 2000, 12, 656–660.

4 S. N. Baker and G. A. Baker, Angew. Chem., Int. Ed., 2010, 49,6726–6744.

5 L. Cao, X. Wang, M. J. Meziani, F. Lu, H. Wang, P. G. Luo,Y. Lin, B. A. Harruff, L. M. Veca, D. Murray, S. Y. Xie andY. P. Sun, J. Am. Chem. Soc., 2007, 129, 11318–11319.

6 H. X. Zhao, L. Q. Liu, Z. D. Liu, Y. Wang, X. J. Zhao andC. Z. Huang, Chem. Commun., 2011, 47, 2604–2606.

7 F. Wang, Y. Chen, C. Liu and D. Ma, Chem. Commun., 2011,47, 3502–3504.

8 H. Li, X. He, Z. Kang, H. Huang, Y. Liu, J. Liu, S. Lian,C. H. A. Tsang, X. Yang and S. T. Lee, Angew. Chem., Int.Ed., 2010, 49, 4430–4434.

17528 | J. Mater. Chem. A, 2014, 2, 17521–17529

9 P. Luo, C. Li and G. Q. Shi, Phys. Chem. Chem. Phys., 2012, 14,7360–7366.

10 S. Zhu, S. Tang, J. Zhang and B. Yang, Chem. Commun., 2012,48, 4527–4539.

11 L. A. Ponomarenko, F. Schedin, M. I. Katsnelson, R. Yang,E. W. Hill, K. S. Novoselov and A. K. Geim, Science, 2008,320, 356–358.

12 E. Kroke and M. Schwarz, Coord. Chem. Rev., 2004, 248, 493–532.

13 A. Thomas, A. Fischer, F. Goettmann, M. Antonietti,J. O. Muller, R. Schlogl and J. M. Carlsson, J. Mater. Chem.,2008, 18, 4893–4908.

14 Y. Wang, X. Wang and M. Antonietti, Angew. Chem., Int. Ed.,2012, 51, 68–89.

15 Y. Wang, X. Wang, M. Antonietti and Y. Zhang,ChemSusChem, 2010, 3, 435–439.

16 F. Goettmann, A. Fischer, M. Antonietti and A. Thomas,Angew. Chem., Int. Ed., 2006, 45, 4467–4471.

17 M. Sadhukhan and S. Barman, J. Mater. Chem. A, 2013, 1,2752–2756.

18 K. B. Migler, Phys. Rev. Lett., 2001, 86, 1023–1026.19 J. A. Pathak and K. B. Migler, Langmuir, 2003, 19, 8667–8674.20 C. Mao, J. R. Huang, Y. T. Zhu, W. Jiang, Q. X. Tang and

X. J. Ma, J. Phys. Chem. Lett., 2013, 4, 43–47.21 Y. C. Zhao, Z. Liu, W. G. Chu, L. Song, Z. X. Zhang, D. L. Yu,

Y. J. Tian, S. S. Xie and L. F. Sun, Adv. Mater., 2008, 20, 1777–1781.

22 Y. Wang, J. S. Zhang, X. C. Wang, M. Antonietti and H. R. Li,Angew. Chem., Int. Ed., 2010, 49, 3356–3359.

23 D. Wang, L. Wang, X. Y. Dong, Z. Shi and J. Jin, Carbon, 2012,50, 2147–2154.

24 X. Zhang, X. Xie, H. Wang, J. Zhang, B. Pan and Y. Xie, J. Am.Chem. Soc., 2013, 135, 18–21.

25 H. L. Guo, X. F. Wang, Q. Y. Qian, F. B. Wang and X. H. Xia,ACS Nano, 2009, 3, 2653–2659.

26 A. Vinu, Adv. Funct. Mater., 2008, 18, 816–827.27 Y. G. Li, J. A. Zhang, Q. S. Wang, Y. X. Jin, D. H. Huang,

Q. L. Cui and G. T. Zou, J. Phys. Chem. B, 2010, 114, 9429–9434.

28 R. C. Dante, P. Martin-Ramos, A. Correa-Guimaraes andJ. Martin-Gil, Mater. Chem. Phys., 2011, 130, 1094–1102.

29 Y. J. Cui, J. S. Zhang, G. G. Zhang, J. H. Huang, P. Liu,M. Antonietti and X. C. Wang, J. Mater. Chem., 2011, 21,13032–13039.

30 M. Shalom, S. Inal, C. Fettkenhauer, D. Neher andM. Antonietti, J. Am. Chem. Soc., 2013, 135, 7118–7121.

31 M. Deifallah, P. F. McMillan and F. Cora, J. Phys. Chem. C,2008, 112, 5447–5453.

32 G. Kresse and J. Furthmiiller, Comput. Mater. Sci., 1996, 6,15–50.

33 G. Kresse and J. Hafner, Phys. Rev. B: Condens. Matter Mater.Phys., 1993, 47, 558–561.

34 P. E. Blochl, Phys. Rev. B: Condens. Matter Mater. Phys., 1994,50, 953–979.

35 J. P. Perdew, J. A. Chevary, S. H. Vosko, K. A. Jackson,M. R. Pederson, D. J. Singh and C. Fiolhais, Phys. Rev. B:Condens. Matter Mater. Phys., 1992, 46, 6671–6687.

This journal is © The Royal Society of Chemistry 2014

Paper Journal of Materials Chemistry A

36 J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett.,1996, 77, 3865–3868.

37 W. Kraus and G. Nolze, Computer Programs, J. Appl.Crystallogr., 1996, 29, 301–303.

38 E. Kroke, M. Schwarz, E. Horath-Bordon, P. Kroll, B. Noll andA. D. Norman, New J. Chem., 2002, 26, 508–512.

39 F. Z. Su, S. C. Mathew, G. Lipner, X. Z. Fu, M. Antonietti,S. Blechert and X. C. Wang, J. Am. Chem. Soc., 2010, 132,16299–16301.

This journal is © The Royal Society of Chemistry 2014

40 K. Suzuki, T. Watanabe and S. I. Murahashi, J. Org. Chem.,2013, 78, 2301–2310.

41 M. T. Schumperli, C. Hammond and I. Hermans, ACS Catal.,2012, 2, 1108–1117.

42 M. T. Schumperili, C. Hammond and I. Hermans, Phys.Chem. Chem. Phys., 2012, 14, 11002–11007.

J. Mater. Chem. A, 2014, 2, 17521–17529 | 17529