Growth of g-C3N4 Layer on Commercial TiO2 for Enhanced Visible Light Photocatalytic Activity
Novel visible light photocatalytic graphitic carbon nitride/TiO2 (g-C3N4/TiO2) composite samples were synthesized by heating mixtures of melamine and commercial TiO2(TO) at different weight ratios. The samples were characterized by X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), transmission electron microscopy (TEM), photoluminescence spectroscopy (PL), Fourier transform infrared spectroscopy (FTIR), and UV-visible diffused reflectance spectroscopy (UV-vis DRS). Characterization confirms formation of nanocomposites of g-C3N4/TiO2. At the optimized precursor weight ratio (), the samples exhibited highest adsorption capacity and visible light photocatalytic activity, measured by degradation of methylene blue (MB). Under visible light irradiation, the excited electrons on the surface of g-C3N4 transfer easily to the conduction band (CB) of TiO2 via the well-built heterojunction. The g-C3N4/TiO2 nanocomposites exhibit enhanced visible light catalytic activity due to increased visible light adsorption and effective separation of photogenerated electron-hole pairs. These g-C3N4/TiO2 nanocomposites could find broad applicability in environmental protection due to their excellent visible light photocatalytic property and facile, cost-effective preparation process.
Visible light photocatalysis has attracted much attention due to its extensive application in the fields of hydrogen generation [1, 2] and environmental protection [3, 4]. TiO2 has been proved to be a competent photocatalyst for environmental applications; however, due to the wide band gap of anatase TiO2 (3.2 eV), it cannot be excited under visible light irradiation ( nm) . Therefore, there is much interest in developing visible light-responsive TiO2 with catalytic activity high enough for practical applications.
Recently, graphite-like carbon nitride (g-C3N4) has drawn wide attention for its potential application in the fields of catalytic support, gas storage, drug delivery, and optical and electronic materials  and as a metal-free, -type semiconductor with tri-s-triazine units . The band gap of g-C3N4 is 2.73 eV, imbuing a response wavelength of up to 450 nm, creating enormous potential in the photocatalysis field. Research has focused on synthesizing the g-C3N4 with several precursors [8–11]. Ma et al. reported a strategy for enhancing the photoactivity of g-C3N4 via doping of nonmetal elements .
Recently, Yan and Yang reported a TiO2-g-C3N4 composite material for photocatalytic H2 evolution under visible light irradiation . The TiO2-g-C3N4 composites were prepared by thermal treatment of the mixture of TiO2 and g-C3N4. Yang et al. reported that N-doped TiO2/g-C3N4 composites with enhanced daylight photocatalytic properties were prepared by heating titanium tetrachloride ethanol solutions with C3N4 . However, commercial g-C3N4 is not readily available and easily obtainable; inexpensive precursors for g-C3N4 need to be used requiring tedious preparation which may prevent large-scale application. Fu et al. reported a solid-state approach to synthesizing g-C3N4 coated TiO2 nanocomposites from urea and commercial TiO2 precursors ; however, the preparation conditions and proposed mechanism for enhancing photocatalytic activity of g-C3N4/TiO2 require further research.
In the present work, all raw materials were purchased from commercial sources. The g-C3N4/TiO2 composite photocatalysts were prepared by a facile solid-phase method of directly heating melamine and commercial TiO2 mixtures. In the process of melamine pyrolysis, sublimation and thermal condensation of melamine occurred simultaneously in temperature range of 297–390°C  and triazine-ring-based radicals and NHx were formed as a result of the decomposition . Based on XPS, FTIR, and TEM results, it was confirmed that g-C3N4 existed in prepared composites. Photocatalytic efficiency of the as-prepared g-C3N4 coated TiO2 nanocomposites was determined by degradation of methylene blue (MB) dye under visible light irradiation.
2.1. Synthesis of Photocatalysts
g-C3N4-TiO2 photocatalysts with various mass ratios ( : = 1.5, 2.5, 3.5, and 4.5, denoted TO-M-1.5, 2.5, 3.5, and 4.5, resp.) were prepared as follows. 2 g of TiO2 (TO) and Xg of melamine were combined in an alumina crucible and stirred with deionized water at room temperature. The mixture was dried at 60°C for 1 h and then annealed to 520°C in a covered muffle furnace for 5 h. Finally, the sample was cooled to room temperature to yield the g-C3N4-TiO2 photocatalyst. The g-C3N4 photocatalyst was also synthesized through pyrolysis of melamine, via direct heating of melamine at 520°C for 4 h. For comparison, TO-520-5 and M-520-5 were prepared analogously in the absence of melamine and TiO2, respectively. Finally, precursors with the same mass ratios ( : = 2.5) were prepared at temperatures of 320°C, 420°C, 520°C, and 620°C to yield TO-M-320, TO-M-420, TO-M-520, and TO-M-620 for 5 h, respectively.
2.2. Materials Characterization
Comprehensive structural characterization of as-prepared samples was undertaken. The crystal structures of samples were determined by an X-ray diffractometer with Cu-Kα radiation (XRD: modelD/max RA, Rigaku Co., Japan). The surface area and porosity of the samples were estimated by measuring the nitrogen adsorption-desorption isotherms on a Quantachrome Autosorb-1 system. FTIR spectra were recorded on a Nicolet Nexus spectrometer on samples embedded in KBr pellets. XPS spectra of the products were collected on a photoelectron spectrometer (Kratos Co., UK). The morphology of samples was characterized by TEM (JEM-2010). UV-vis diffuse reflection spectra were taken on a scan UV-vis spectrophotometer (UV-vis DRS: UV-2550, Shimadzu, Japan). Photoluminescence spectra of samples were obtained by using a fluorescence spectrometer (HITACHI, F-7000) at room temperature.
2.3. Evaluation of Photocatalytic Activity
Methylene blue (MB) oxidation reactions were performed on a photochemical reactor (BL-GHX) using a 500 W Xe lamp fitted with a thimble containing NaNO2 solution (1 mol/L) to block UV light ( nm). Additionally, immediate heat removal from the lamp was achieved via water flow through the inner thimble of the annular quartz tube. 0.02 g of catalyst was dispersed in MB aqueous solution (50 mL, 5 mg/L) with stirring for 40 min without irradiation until adsorption-desorption equilibrium was reached. During the irradiation, 5 mL aliquots were taken from the reaction cell at hourly intervals for subsequent dye concentration analysis following centrifugation. The degradation rate of MB was evaluated via the intensity of UV-vis absorption spectra at 665 nm.
3. Results and Discussion
Figure 1 depicts the XRD patterns of the as-prepared, mass ratio differentiated photocatalysts alongside g-C3N4 and TiO2. The g-C3N4 has two main diffraction peaks at 13.1° and 27.5° as a result of a polymeric layered structure . The characteristic peaks at 27.5° and 13.1° correspond to the stacking of the conjugated aromatic system and the periodic arrangement of the condensed tri-s-triazine units of g-C3N4, respectively. The peak intensity at 27.5° corresponding to g-C3N4 increases with increasing weight ratio of melamine, which confirms that g-C3N4 has been produced in the as-prepared samples and the yield of it also gradually increased. For all g-C3N4/TiO2 composites and TiO2, characteristic anatase TiO2 peaks with similar position, intensities and widths were detected. This implies that the phase structure and the crystallite size of TiO2 particles in as-prepared samples are conserved.
3.2. Visible Light Catalytic Activity
The degradation of MB is carried out under visible light ( nm) illumination in aqueous sample suspensions. Figure 2(a) shows degradation of MB in the presence of samples fabricated from varying precursor ratios ( : = 1.5, 2.5, 3.5, and 4.5). Contrastingly, TO-520-5 and M-520-5 were prepared from single precursors by heating TiO2 and melamine, respectively, under analogous reaction conditions. The photocatalytic activity of g-C3N4/TiO2 nanocomposites first increases and then decreases with the increasing melamine additive. The excess g-C3N4 prevented pollutants from being in contact with TiO2. As shown in Figure 2(a), during adsorption and the 5 hours of visible light photocatalytic degradation, TO-M-2.5 exhibits the highest visible light catalytic activity of all compounds including TiO2 (TO) and g-C3N4. The degradation rate (%) can be calculated as where is the initial concentration of dye and is the time dependent concentration. Degradation rate of MB induced by TO-M-2.5 is about 91%. It is surmisable that when combined TiO2 and g-C3N4 produce a synergistic effect and effectively enhance the visible light photocatalytic efficiency.
The photocatalytic activity of the samples prepared at different temperatures is shown in Figure 2(b). The variation in degradation observed (Figure 2(b)) suggests that the calcination temperature is also a key factor with regard to photocatalytic activity.
3.3. Microstructure Analysis
The BET surface areas and porous structures of TiO2 and TO-M-2.5 were investigated by nitrogen adsorption/desorption. In Figure 3, the two samples show a type IV adsorption isotherm with a H2 hysteresis loop in the range (/) of 0.6–1.0. The surface area and pore volume (Table 1) of pure TiO2 are 81.63 m2/g and 0.221 cm2/g, higher than those of the TO-M-2.5 (50.57 m2/g, 0.154 cm2/g). Compared with TiO2, the specific surface area of TO-M-2.5 is reduced, potentially due to the presence of g-C3N4 layers on the TiO2 surface. However, the surface area of g-C3N4/TiO2 is not a positive factor. The enhanced visible light activity of g-C3N4/TiO2 should be ascribed to the enhanced visible light adsorption and dispersion of TiO2 resulting in increased adsorption capacity due to the presence of g-C3N4 (Figures 6(c) and 7).
The FTIR spectra (Figure 4) of TO-M-2.5, TO-M-3.5, and TO-M-4.5 were similar to that of g-C3N4, particularly in three regions: 805 cm−1, 1200~1650 cm−1, and 3100 cm−1~3400 cm−1. The peaks at 805 cm−1 and 1200~1650 cm−1, in the spectra of TO-M-2.5, TO-M-3.5, and TO-M-4.5, are assigned to the characteristic stretching modes of g-C3N4, indicating the existence of g-C3N4. The band at 805 cm−1 corresponds to the characteristic breathing mode of the triazine units [8, 13]. The absorption peak at ca. 1329 cm−1 and ca. 1635 cm−1 can be attributed to C–N and C=N stretching modes, respectively [15, 16]. Additionally, the wide band between 3100 cm−1 and 3400 cm−1 corresponds to the N–H and O–H stretching vibration [14, 17]. These characteristic peaks were largely absent in the FTIR patterns of TO-M-1.5, which probably results from the low content of g-C3N4 in the sample.
The elemental chemical states of TO-M-2.5 were detected by XPS as shown in Figure 5 and were consistent with the FTIR results. The C 1s peaks (Figure 5(a)) at 284.8 eV and 288.1 eV were assigned to the C–C bond [15, 18] and the sp3 C–N bond [15, 16, 18], respectively. The N 1s XPS spectrum of TO-M-2.5 can be deconvoluted into two peaks at 400.1 eV and 398.6 eV (Figure 5(b)), which were attributed to the binding energy of N–(C)3 and C–N–C, respectively. These C 1s and N 1s binding energies were consistent with previously reported XPS data [8, 9], which further confirms that our as-prepared samples contain the graphite-like carbon nitride. The O 1s XPS spectrum (Figure 5(c)) included two peaks at 529.8 eV and 531.6 eV, which were closely related with C–O functional groups and surface hydroxyl groups . The adsorbed water molecule was a major cause of surface hydroxyl groups benefitting capture of photogenerated holes and inhibition of electron-hole recombination. The peaks located at 458.7 and 464.4 eV in the Ti 2p spectra correspond to the Ti 2p3/2 and Ti 2p1/2 (Figure 5(d)). There were no Ti-C or Ti-N bond peaks, which implied that C and N elements do not enter the lattice of TiO2. By the analysis of XPS, it can be ensured that the g-C3N4 coated TiO2 composites were prepared by heating mixture of melamine and TiO2.
Figure 6(a) shows the TEM images of pure TiO2, consisting of nanoparticles with ca. 10 nm radius, consistent with the size of TiO2 from microstructure analysis (Table 1). Figure 6(d) shows clear lattice fringes for the identification of crystallographic spacing of ca. 0.35 nm, matching well with an anatase TiO2 (101) plane. The TEM and HRTEM images of TO-M-2.5 samples (Figures 6(b), 6(d), and 6(e)) also indicate that partial surface of TiO2 was covered with a secondary phase and in Figures 6(b) and 6(d), it is clearly shown that TiO2 is well distributed on a layered structure. The XPS and FTIR have confirmed the existence of g-C3N4. As the g-C3N4 is a semicrystallized material, it is hard to obtain the clear lattice fringe. However, by comparison of the morphology of g-C3N4 (Figure 6(c)) and TO-M-2.5, it can be concluded that the secondary phase in TO-M-2.5 is g-C3N4. In Figure 6(d), g-C3N4/TiO2 (TO-M-2.5) displayed a connection between TiO2 and g-C3N4, indicating that the formation of heterojunction was possible.
Figure 7(a) shows the UV-vis DRS spectra of pure TiO2, g-C3N4, and g-C3N4/TiO2 nanocomposites. Visible light absorption of composites is enhanced by the presence of g-C3N4; however, it cannot be simply concluded that visible light absorption is consistent with photoactivity as many factors influence catalytic activity under visible light . Enhanced activity is due to the presence of the TiO2, which can promote effective interfacial electron transfer  and conduce efficient space separation of photogenerated electron-hole pairs of g-C3N4. The band gaps of TiO2 and g-C3N4 are showed in Figure 7(b), which are 3.2 eV and 2.7 eV, respectively.
3.8. Fluorescence Analysis
Figure 8 shows the photoluminescence spectra of as-prepared samples. Excluding TiO2 and M-TO-1.5, the as-prepared composites and g-C3N4 exhibit similar profiles with a broad emission band from 350 to 600 nm under an excitation wavelength of 330 nm, with intensity correlated to g-C3N4 content. These spectra imply that PL intensities of as-prepared composites are strongly dependant on recombination of electron-hole pairs in g-C3N4. The PL quenching was observed in g-C3N4/TiO2 nanocomposites as the content of TiO2 increases, probably due to the charge transfer occurring from g-C3N4 to TiO2 [11, 15].
4. Proposed Photocatalytic Mechanism
The g-C3N4 is formed by sp2 hybridization between C and N atoms, which depicts the π conjugate structure. Presence of g-C3N4 benefits the dispersion of TiO2 resulting in increased adsorption capacity. When the as-prepared composite is irradiated with visible light, the electrons get promoted from the valence band (VB) to the conduction band (CB) in g-C3N4 (Figure 9), providing a route for electronic transition. The CB edge potential of g-C3N4 is more negative than that of TiO2 allowing the excited electron on the surface of g-C3N4 to transfer easily to TiO2 via the well-built heterojunction [22, 23] (Figure 6(d)). The recombination of photogenerated charge was inhibited and photocatalytic activities were enhanced effectively  for TiO2 which provides a site for electron translation. The previously mentioned characterization indicates that coupling TiO2 with g-C3N4 produces a well-contacted solid-solid heterojunction interface between g-C3N4 and TiO2 semiconductor particles, which promotes the effective interfacial electron transfer [21, 25] and separation of photogenerated charge carriers, significantly enhancing photocatalytic activity.
g-C3N4/TiO2 nanocomposite samples were simplistically synthesized by thermal treatment of mixtures of industrial TiO2 and melamine at different weight ratios. It was observed that TiO2 particles were distributed on the layered g-C3N4 formed in situ from melamine. TO-M-2.5, the g-C3N4/TiO2 composite fabricated with a precursor ratio ( : ) of 2.5, exhibited the highest visible light catalytic activity. The introduction of g-C3N4 increased visible light adsorption of the g-C3N4/TiO2 nanocomposite. The enhanced visible light catalytic activity arises from the heterojunction between g-C3N4 and TiO2 which effectively enhances separation of photogenerated electron-hole pairs. The novel g-C3N4/TiO2 nanocomposites prepared by this facile approach could have broad application in environmental protection.
Conflict of Interests
The authors declare that there is no conflict of interests regarding the publication of this paper.
This research is financially supported by the Science and Technology Project of CQ CSTC (stc2013yykfB50008, cstc2013jcyjA20018), the Science and Technology Project from Chongqing Education Commission (KJ120713, KJ130725), and the Innovative Research Team Development Program in University of Chongqing (KJTD201314).
X. Ma, Y. Lv, J. Xu et al., “A strategy of enhancing the photoactivity of g- via doping of nonmetal elements: a first-principles study,” The Journal of Physical Chemistry, vol. 116, no. 44, pp. 23485–23493, 2012.View at: Google Scholar
M. Fu, J. Pi, F. Dong, Q. Duan, and H. Guo, “A cost-effective solid-state approach to synthesize g-C3N4 coated TiO2 nanocomposites with enhanced visible light photocatalytic activity,” International Journal of Photoenergy, vol. 2013, Article ID 158496, 7 pages, 2013.View at: Publisher Site | Google Scholar
Y. Qiu and L. Gao, “Chemical synthesis of turbostratic carbon nitride, containing C-N crystallites, at atmospheric pressure,” Chemical Communications, vol. 9, no. 18, pp. 2378–2379, 2003.View at: Google Scholar
Q. Y. Li, J. W. Zhang, Z. S. Jin et al., “A novel with a large amount of bulk intrinsic defects-visible-light-responded photocatalytic activity,” Chinese Science Bulletin, vol. 11, no. 58, pp. 1007–1013, 2013.View at: Google Scholar