Facile synthesis and enhanced visible-light photocatalytic activity of graphitic carbon nitride decorated with ultrafine Fe₂O₃ nanoparticles

Abstract

Hybrid nanocomposites based on graphitic carbon nitride (g-C₃N₄) nanosheet supported ultrafine Fe₂O₃ nanoparticles have been successfully prepared by a facile thermal polymerization and deposition-precipitation method. Characterization results demonstrate that Fe₂O₃/g-C₃N₄ nanocomposites exhibit a well-defined morphology, in which Fe₂O₃ nanocrystals of 3 nm size with a narrow particle distribution are uniformly dispersed on the layers of the g-C₃N₄ nanosheets. Photocatalytic reaction results indicate that the photocatalytic activity of g-C₃N₄ is significantly enhanced after introduction of a small amount of Fe₂O₃, and the optimum activity of the Fe₂O₃/g-C₃N₄ nanocomposites with a weight ratio of Fe₂O₃ at 0.1% is up to about 3 times and 62 times higher than those of pure g-C₃N₄ and pure Fe₂O₃, respectively, for the degradation of organic dye Rhodamine B (RhB) under visible light irradiation. The high performance of the Fe₂O₃/g-C₃N₄ photocatalysts is mainly attributed to the synergistic effect at the interface of the heterojunction between Fe₂O₃ nanoparticles and g-C₃N₄ nanosheets, including improved separation efficiency of the charge carriers, a suppressed recombination process and suitable band position of the composites. These results underline the potential for the development of effective, low-cost and earth-abundant photocatalysts for the promotion of water splitting and environmental remediation under natural sunlight by construction of sustainable g-C₃N₄ polymeric materials.

---

1. Introduction

In view of the current increasing crisis of energy shortage and environmental problems, semiconductor-based photocatalysis technology has attracted considerable attention due to its extensive applications in hydrogen generation from water splitting,^1,2 degradation of environmental pollutants,³ CO₂ reduction,⁴ selective organic synthesis⁵ and air purification.⁶ At present, developing efficient, inexpensive, non-toxic and stable materials as photocatalysts capable of harvesting solar light in the visible region still remains a great challenge in this field.^1,2,7,8 Recently, a novel polymeric semiconductor, graphitic carbon nitride (g-C₃N₄) was firstly reported by Wang et al.⁹ as a metal-free photocatalyst for visible-light photocatalytic water splitting. Moreover, because of its simple fabrication, low-cost, high chemical and thermal stability, suitable band gap and tunable electronic structure, g-C₃N₄ is considered to be a very promising photocatalyst for various potential applications.^10–13 However, the low separation efficiency and high recombination rate of photogenerated electrons and holes greatly restrict the photocatalytic activity of pure g-C₃N₄ for practical use. In addition, the band gap of g-C₃N₄ is still as large as 2.7 eV, corresponding to an optical wavelength just at 450 nm, which also seriously limits its visible-light response and photocatalytic performance.

To enhance the visible-light photocatalytic activity of g-C₃N₄, a variety of methods have been developed, including metal and nonmetal doping,^6,14–17 nanostructure designing,^18,19 sensitizing with organic dyes^20,21 and coupling with other semiconductors to form composite photocatalysts.^22–27 Among these strategies, semiconductor combination is one of the most effective and convenient methods for improving the photocatalytic performance with a formation of a heterojunction structure. Ge et al.²⁸ reported a CdS quantum dots coupled g-C₃N₄ photocatalyst with a hydrogen evolution rate 9 times that of pure g-C₃N₄ owing to the synergistic effect of the heterostructure. He et al.²³ prepared SnO_2−x/g-C₃N₄ composite photocatalyst via a calcination process and suggested that the enhanced RhB photodegradation activity benefited from a direct Z-scheme mechanism, which promoted light absorption and the separation of charge carriers. Huang et al.²⁹ developed WO₃/g-C₃N₄ heterojunction photocatalyst by a calcination method that showed much higher efficiency for the photodegradation of organic pollutants than that of pure samples.

As a promising candidate photocatalyst, Fe₂O₃ is a typical non-toxic semiconductor with a narrow band gap of about 2.2 eV (much smaller than that of g-C₃N₄), which can absorb light up to 600 nm, corresponding to approximately 40% of the solar spectrum energy. Besides, the superior properties of good chemical stability, high electrical conductivity, earth abundance and low cost make Fe₂O₃ advantageous over other semiconductor materials.³⁰ However, the photocatalytic activity of Fe₂O₃ is rather low due to high electron–hole recombination rate. It has been reported that fabrication of nanosized Fe₂O₃ particles could largely enhance the photocatalytic performance of this material.^31,32 Therefore, it is then expected that if g-C₃N₄ was combined with Fe₂O₃ nanoparticles of high dispersion to form heterojunction nanocomposites, some intrinsic disadvantages of g-C₃N₄ such as the low separation efficiency of charge carriers and the limited visible-light absorption ability can be well overcome and it can be highly possible to achieve excellent photocatalytic activity under visible light irradiation.

Up to now, for synthesizing composites of g-C₃N₄ coupled with another semiconductor, most of the preparation processes employed conventional calcination or hydrothermal methods by simply combining the mixture of two pure samples or precursors together, while the direct and controllable growth of highly dispersed semiconductor nanoparticles on the layers of g-C₃N₄ has been rarely exploited, and the relationship between well-defined structure and photocatalytic performance needs to be further clarified. Ye et al.²⁵ synthesized Fe₂O₃/g-C₃N₄ photocatalyst by calcination two precursors (FeCl₃·6H₂O and melamine) at 520 °C. Theerthagiri et al.²⁶ mixed the two pure composites (Fe₂O₃ and g-C₃N₄) and sintered them at 300 °C to synthesize Fe₂O₃/g-C₃N₄ hybrid. In this work, we report a well-defined Fe₂O₃/g-C₃N₄ nanocomposite for the photocatalytic degradation of organic pollutant Rhodamine B (RhB) under visible light illumination. A facile strategy via a deposition-precipitation method has been developed and Fe₂O₃ nanoparticles with a diameter of 3 nm are homogeneously deposited on g-C₃N₄ nanosheets. Our results reveal that the photocatalytic activity of g-C₃N₄ is significantly improved after introduction of Fe₂O₃ nanoparticles and the optimum photocatalytic activity of Fe₂O₃/g-C₃N₄ composites is achieved at a weight ratio of 0.1% Fe₂O₃. XRD, SEM, high-resolution TEM, XPS, DRS and PL spectra are employed to analyze the origin of the superior photocatalytic performance and a possible mechanism is proposed.

2. Experimental section

2.1 Preparation of photocatalysts

Urea and ferric nitrate nonahydrate (analytical grade purity) were purchased from Sinopharm Chemical Reagent Co., Ltd. and were used without further purification. All other reagents used in this work were of analytical grade and were used as received.

Metal-free g-C₃N₄ nanosheets were prepared by a thermal polymerization method. In a typical synthesis, urea was loaded into an alumina crucible with a lid and heated in a muffle furnace at a ramping rate of 5 °C min⁻¹ to 550 °C. This temperature was held for 4 h prior to cooling to room temperature naturally. The yellow product was collected and ground into a fine powder.

Fe₂O₃/g-C₃N₄ nanocomposites were synthesized using a facile deposition-precipitation method. As illustrated in Fig. 1, 200 mg g-C₃N₄ powder was first dispersed in 50 ml water by ultrasonication for 30 min. Then the mixture was transferred into a water-bath and continuously stirred at 70 °C, followed by adding of 10 ml Fe(NO₃)₃–ethanol solution (calculated amount of Fe(NO₃)₃·9H₂O dissolved in ethanol) dropwise. After that, the pH value of the mixture was adjusted to 8 by careful adding of 10% ammonium hydroxide. Subsequently, the obtained solid after evaporation of the solvent was gradually heated to 400 °C in Ar atmosphere and kept for 3 h to reach complete decomposition of Fe precursors. The final composite Fe₂O₃/g-C₃N₄ photocatalysts with various Fe₂O₃ content (0.05 wt%, 0.1 wt%, 0.3 wt%, 0.6 wt% and 1 wt%) were obtained. For comparison, pure Fe₂O₃ nanopowder was also prepared following the above-mentioned procedure in the absence of g-C₃N₄.

Fig. 1 Schematic diagram of the preparation process of Fe₂O₃/g-C₃N₄ nanocomposites.

2.2 Characterization of photocatalysts

X-ray diffraction (XRD) was performed on a Rigaku RINT-2200 diffractometer using Cu Kα radiation source at 40 kV and 20 mA. The morphologies and microstructures of the samples were investigated by scanning electron microscopy (SEM) on a Hitachi SU-8020 cold field emission instrument and by high resolution transmission electron microscopy (HRTEM) run on a JEOL JEM-2100 microscope at 200 kV. Energy-dispersive X-ray (EDX) spectroscopy attached to SEM was used to determine the chemical composition of the composite. X-ray photoelectron spectroscopy (XPS) was measured on a VG-ESCALAB 250Xi instrument with a monochromatic Al Kα source. The UV-vis diffuse reflectance spectra (DRS) were recorded on a JASCO V-750 instrument equipped with an integrating sphere in the range of 200–800 nm. The photoluminescence (PL) spectra were obtained at room temperature by an Edinburgh Instruments FLS980 Fluorescence Spectrometer with an excitation wavelength of 325 nm.

2.3 Photocatalytic reaction

The photocatalytic activities of the synthesized samples were evaluated by the degradation of Rhodamine B (RhB) under visible light irradiation. Typically, 0.03 g of the photocatalyst was dispersed into 200 ml of 20 mg L⁻¹ aqueous solution of RhB via ultrasonic treatment for 10 min. Prior to irradiation, the suspension was magnetically stirred in the dark for an hour to establish the adsorption–desorption equilibrium of RhB onto the catalyst surface at room temperature. A 300 W Xe lamp with a UV cutoff filter (λ > 400 nm) was used as the light source to provide the visible light. At regular intervals, 4 ml of the suspension was withdrawn and centrifuged to remove the catalyst particles for analysis. The degradation of RhB was monitored by a UV-vis spectrophotometer (Hitachi, U-3310) by measuring its absorbance at 553 nm.

3. Results and discussion

3.1 Catalyst characterization

The XRD patterns of Fe₂O₃/g-C₃N₄ nanocomposites are shown in Fig. 2a, together with that of pure g-C₃N₄ and Fe₂O₃ nanopowder for comparison. Fe₂O₃ nanopowder exhibits diffraction peaks at 2θ 24.1, 33.1, 35.6, 40.9, 49.5, 54.1, 62.4, 64.0°, which are indexed to the (012), (104), (110), (113), (024), (116), (214) and (300) planes of hematite α-Fe₂O₃ (JCPDS 33-0664), respectively. For pure g-C₃N₄ and Fe₂O₃/g-C₃N₄ composites, two broad peaks at 2θ 13.1 and 27.4° are observed, corresponding to the (100) and (002) crystal planes of g-C₃N₄, which are ascribed to the in-planar repeat period of the tri-s-triazine units and the interlayer spacing between graphitic layers, respectively.⁹ However, no significant peaks for Fe₂O₃ can be detected in the Fe₂O₃/g-C₃N₄ patterns, probably due to the relatively low content of Fe₂O₃ in the composite (≤1 wt%) and high dispersion of Fe₂O₃ nanoparticles on the g-C₃N₄. To get the XRD information of Fe species in the composite, we have added more Fe₂O₃ onto the g-C₃N₄ (20 wt%). It is found that obvious diffraction peaks of hematite α-Fe₂O₃ show up, which demonstrates the existence of Fe₂O₃ in the composite in the form of α-phase (ESI Fig. S1†). The energy-dispersive X-ray (EDX) spectrum was used to analyze the chemical composition of the Fe₂O₃/g-C₃N₄ nanocomposite. As shown in Fig. 2b, the composite was composed of Fe, O, C, N and Si, among which the Si signal was from silicon foil used as SEM support for the powder sample, indicating that Fe₂O₃ was indeed loaded onto g-C₃N₄ nanosheets.

Fig. 2 (a) XRD patterns of Fe₂O₃/g-C₃N₄ nanocomposites with different Fe₂O₃ weight percentage; (b) EDX analysis of 1 wt% Fe₂O₃/g-C₃N₄.

The SEM image of pure g-C₃N₄ is shown in Fig. 3a. It can be seen that pure g-C₃N₄ prepared from urea by high temperature calcination presents flowerlike structure, which is composed of thin graphitic nanosheets with imbedded open nanopores of tens to several hundred nanometers diameter. The pure g-C₃N₄ of high porosity contributes to a large surface area of 38.9 m² g⁻¹, as determined by BET nitrogen adsorption/desorption isotherm (ESI Fig. S2†). A high surface area means exposure of more active sites for g-C₃N₄ in the photocatalytic reaction. Fig. 3b–d exhibit typical TEM images of 1 wt% Fe₂O₃/g-C₃N₄ nanocomposite. As can be seen, ultrafine Fe₂O₃ nanoparticles are uniformly dispersed on the nanosheets of g-C₃N₄, with close contact between each other via forming some chemical bonds. This tight interaction is favorable for the formation of a heterojunction structure in the composites and promotes effective charge transfer and separation.²³ The d-spacing determined from HRTEM shown in Fig. 3e is 0.221 nm, corresponding to the (113) plane of hematite α-Fe₂O₃. To further confirm the microstructure of the composite, selected-area electron diffraction (SAED) was carried out on the area shown in Fig. 3d. The SAED exhibits concentric diffraction rings, indicating that the dispersed Fe₂O₃ nanoparticles are polycrystalline and well crystallized (Fig. 3f). The particle size distribution of Fe₂O₃ nanocrystals is presented in Fig. 3g. By measuring more than 300 particles from HRTEM images, it is found that Fe₂O₃ particles have a narrow size distribution, with most of them falling in the range of 2.5–4.0 nm with a mean size of approximately 3 nm. The above results demonstrate that Fe₂O₃/g-C₃N₄ hybrid composites with well-defined structure have been successfully prepared by the facile deposition-precipitation method used in our work, which may have promising application in the photocatalytic reaction.

Fig. 3 FESEM image of pure g-C₃N₄ (a); TEM images (b–d), HRTEM image (e), SAED pattern (f) and particle size distribution (g) of 1 wt% Fe₂O₃/g-C₃N₄ nanocomposite.

XPS measurements were carried out to investigate the chemical states of the surface composition in pure g-C₃N₄ and 0.1 wt% Fe₂O₃/g-C₃N₄ nanocomposite. Fig. 4a shows the survey scan XPS spectra of both samples. The strong peaks of C 1s, N 1s and O 1s are observed in Fig. 4a, indicating the presence of C, N and O elements in both samples. While for 0.1 wt% Fe₂O₃/g-C₃N₄ sample, the Fe 2p signal is negligible, owing to the low content of Fe₂O₃ in the composite. High resolution spectra of C 1s, N 1s and Fe 2p are shown in Fig. 4b–d. As illustrated in Fig. 4b, three binding energy values of C 1s are observed at 284.8, 286.2 and 288.2 eV for both samples, which are assigned to sp² C–C bonds from carbon-containing contaminations, sp³ hybridized carbon from the defects of g-C₃N₄ surface and sp²-bonded carbon of N–C=N in the graphitic structure, respectively, with the latter constituting the major carbon composition in the g-C₃N₄.^33–35 In the N 1s spectrum of g-C₃N₄ (Fig. 4c), three peaks at 398.7, 399.8 and 401.2 eV can be clearly resolved, corresponding to sp² hybridized nitrogen (C–N=C), sp³ hybridized nitrogen (H–N–(C)₃) and amino functional groups (C–NH_x), respectively.^17,34,36 However, two N 1s core levels of 0.1 wt% Fe₂O₃/g-C₃N₄ are found to be slightly shifted to lower binding energies (399.5 and 401.1 eV) compared with that of g-C₃N₄. This is probably due to the strong interaction between Fe₂O₃ and g-C₃N₄ in the heterostructure, which gives rise to charge transfer between each other.³⁷ The atomic ratio of C/N determined by the quantitative analysis results of XPS is 0.80 (the contribution of adventitious C 1s signal at 284.8 eV was deducted from the total C 1s peak area), which is close to the expected value of C/N in ideal g-C₃N₄ (0.75), reflecting the high graphitization degree of the g-C₃N₄ used in our experiments. The Fe 2p spectrum of 0.1 wt% Fe₂O₃/g-C₃N₄ (Fig. 4d) exhibits two characteristic peaks at 710.9 and 723.4 eV, attributed to typical Fe 2p_3/2 and Fe 2p_1/2 signals of Fe₂O₃ phase according to the literatures.^38,39 Therefore, the XPS results are in good agreement with the results of XRD and TEM that ultrafine Fe₂O₃ nanoparticles are evenly distributed on g-C₃N₄ nanosheets and thus heterostructures with close interconnection are formed.

Fig. 4 XPS of synthesized g-C₃N₄ and 0.1 wt% Fe₂O₃/g-C₃N₄ nanocomposite: (a) survey spectra, (b) C 1s spectra, (c) N 1s spectra and (d) Fe 2p spectrum.

To investigate the optical absorption properties of the as-prepared catalysts, we examined UV-vis diffuse reflectance spectra (DRS) of pure g-C₃N₄ and Fe₂O₃/g-C₃N₄ composites. As can be seen in Fig. 5, the g-C₃N₄ nanosheets show a strong absorption edge at about 450 nm, corresponding to the band gap of approximately 2.7 eV.^9,29,35 Compared to g-C₃N₄, all of the Fe₂O₃/g-C₃N₄ composites display similar visible light absorbance property but exhibit some red-shifts in absorption wavelength, and the absorption edge gradually moves towards longer wavelength as the Fe₂O₃ loading increases (the color of the composites evolves from yellow to light brick-red), implying the strong interaction between Fe₂O₃ and g-C₃N₄ in the heterojunction structure, which results in modifications of optical properties of Fe₂O₃/g-C₃N₄ composites.²³

Fig. 5 UV-vis diffuse reflectance spectra of Fe₂O₃/g-C₃N₄ nanocomposites and pure g-C₃N₄.

3.2 Photocatalytic performance

RhB photodegradation was used to evaluate the photocatalytic activity of Fe₂O₃/g-C₃N₄ nanocomposite under visible light irradiation. Fig. 6 represents a comparison of spectral changes during the photocatalytic degradation of RhB over g-C₃N₄ and 0.1 wt% Fe₂O₃/g-C₃N₄ composite. For pure g-C₃N₄ (Fig. 6a), the photocatalytic degradation rate of RhB is quite slow and about 50% of the starting RhB molecules still remains after 90 min of the reaction time. However, it is evident from Fig. 6b that the intensity of the absorbance peak of RhB at 553 nm decreases rapidly within 90 min, indicating the superior photocatalytic activity for 0.1 wt% Fe₂O₃/g-C₃N₄. Thus, 0.1 wt% Fe₂O₃/g-C₃N₄ composite exhibits a much better photocatalytic performance than that of pure g-C₃N₄ nanosheets under the same conditions.

Fig. 6 Changes in UV-vis absorption spectra of RhB by (a) pure g-C₃N₄ and (b) 0.1 wt% Fe₂O₃/g-C₃N₄ nanocomposite under visible light irradiation.

The detailed photocatalytic degradation activities for Fe₂O₃/g-C₃N₄ nanocomposites with various Fe₂O₃ content were further investigated and shown in Fig. 7a together with that of pure Fe₂O₃ nanopowder and g-C₃N₄ for comparison. A control experiment in the absence of a photocatalyst was also carried out, and it is observed that RhB self-degradation is almost negligible. No obvious degradation is observed with Fe₂O₃ nanopowder alone. For pure g-C₃N₄, only 50% of RhB has been degraded under visible light after irradiation for 90 min. With regard to the degradation of RhB over Fe₂O₃/g-C₃N₄ composites, the case is quite different. It is found that the Fe₂O₃ content has a great influence on the photocatalytic activities of g-C₃N₄, which is significantly enhanced after loading a very low amount of Fe₂O₃ nanoparticles (lower than 1 wt%). The photocatalytic activity first increases as the Fe₂O₃ content increases from 0.05 wt% to 0.1 wt%, and then decreases at a higher Fe₂O₃ content. When further increasing the loading of Fe₂O₃ higher than 1 wt% leads to activity lower than g-C₃N₄. These results suggest the existence of the synergistic effect of Fe₂O₃/g-C₃N₄ heterojunction, which can effectively improve the photocatalytic activity of Fe₂O₃/g-C₃N₄ composite via facilitated separation of charge carriers on the heterojunction interface. However, it should be noted that excess Fe₂O₃ species are detrimental to RhB degradation, which may be related to the recombination property of Fe₂O₃, more Fe₂O₃ greatly reducing the separation efficiency of photogenerated carriers and consequently lowering the photocatalytic activity of Fe₂O₃/g-C₃N₄. As a result, the optimum photocatalytic activity is achieved for the catalyst of 0.1 wt% Fe₂O₃/g-C₃N₄ composite (about 90% of RhB has been photodegraded within 90 min).

Fig. 7 (a) Photocatalytic activity, (b) first-order kinetics plot and (c) the kinetic constants for the degradation of RhB by g-C₃N₄, Fe₂O₃ nanopowder, and Fe₂O₃/g-C₃N₄ nanocomposite photocatalysts under visible light irradiation.

The experimental data in Fig. 7a can be well fit to a pseudo first-order kinetics equation

−ln(C/C₀) = kt (1)

where k is the apparent first-order rate constant (min⁻¹), C and C₀ are the RhB concentration at reaction time t and 0, respectively. Fig. 7b shows the kinetics plot of Fe₂O₃/g-C₃N₄ nanocomposites for RhB photodegradation reaction. Upon varying the Fe₂O₃ content, the plot of ln(C/C₀) against the irradiation time (t) is nearly a straight line, in which the slope of the fitting line is equal to the value of rate constant. The corresponding kinetic constants with different Fe₂O₃ content are shown in Fig. 7c. Even though we used less amount of catalyst and higher concentration of RhB in this reaction, the photocatalytic activities of Fe₂O₃/g-C₃N₄ composites have been already obviously better or comparable to that of some previously reported similar materials, and their preparation methods are much more complicated.^17,25,26 The highest apparent rate constant under our experimental conditions is determined to be 0.02097 min⁻¹ for 0.1 wt% Fe₂O₃/g-C₃N₄ composite, which is about 3 times and 62 times higher than that of pure g-C₃N₄ (0.00765 min⁻¹) and pure Fe₂O₃ (3.39 × 10⁻⁴ min⁻¹), respectively, revealing the superior photocatalytic activity for highly dispersed Fe₂O₃/g-C₃N₄ composite materials.

3.3 Proposed photocatalytic mechanism

The significantly enhanced photocatalytic activity of the hybrid composite catalysts motivates us to further study the photocatalytic reaction mechanism of the degradation process. Generally, photoinduced holes (h⁺), hydroxyl radicals (˙OH) and superoxide radicals (˙O₂⁻) are known to be the main reactive species involved in the photodegradation reaction.²⁰ In our radical trapping experiments, 1 mmol L⁻¹ of isopropanol (IPA), 1,4-benzoquinone (BQ) and triethanolamine (TEOA) were added into the solution and used as the hydroxyl radical, superoxide radical and hole scavengers, respectively, and their influences on the photocatalytic degradation efficiency for RhB are shown in Fig. 8. For both pure g-C₃N₄ and 0.1 wt% Fe₂O₃/g-C₃N₄, the photocatalytic degradation activities for RhB decrease remarkably by the adding of BQ and TEOA. However, when IPA is added, the activity is also declined, but not as significant as that with BQ and TEOA added. These results suggest that, under visible light irradiation, g-C₃N₄ and Fe₂O₃/g-C₃N₄ nanocomposite follow the similar photocatalytic mechanism for the degradation of RhB, in which the holes and superoxide radicals play a more important role than hydroxyl radicals. Thus, both holes and superoxide radicals are the primary active species in this reaction, which is consistent with previous reports.^20,40,41 Moreover, compared with g-C₃N₄, the photodegradation efficiency of Fe₂O₃/g-C₃N₄ sample declines more significantly with the same addition of BQ and TEOA, indicating that there may exist more superoxide radicals and holes in Fe₂O₃/g-C₃N₄ system than in g-C₃N₄ system due to the effective separation of photogenerated carriers.

Fig. 8 Influences of various hole and radical scavengers in the system of photocatalytic degradation of RhB by pure g-C₃N₄ and 0.1 wt% Fe₂O₃/g-C₃N₄ under visible light irradiation for 90 min.

The lifetime of charge carriers is of great importance to the performance of a photocatalytic reaction. The formation of heterojunction structures in the composite photocatalysts may lead to the fast separation of electrons and holes and thus result in the high photodegradation activity. As is well known that the photoluminescence (PL) emission spectra mainly stem from the recombination of free charge carriers in semiconductors,⁴² we carried out the room-temperature photoluminescence studies on the as-prepared samples to investigate the migration and separation efficiency of photogenerated electron–hole pairs. Fig. 9 represents the comparison of PL spectra for pure g-C₃N₄ and 0.1 wt% Fe₂O₃/g-C₃N₄ with an excitation wavelength of 325 nm. As for pure g-C₃N₄, a strong emission peak at about 445 nm and a shoulder peak around 475 nm are observed. After hybridization with Fe₂O₃ nanoparticles, the overall intensity of PL signals shows an obvious decrease, indicating that the recombination rate of photoinduced electron–hole pairs is greatly suppressed, which lengthens the lifetime of photogenerated carriers and subsequently favors the enhancement of photocatalytic performance.^23,43 Therefore, it can be deduced from Fig. 9 that the higher photocatalytic activity for Fe₂O₃/g-C₃N₄ compared with pure g-C₃N₄ may arise from the higher separation efficiency of electron–hole pairs.

Fig. 9 Photoluminescence emission spectra of as-prepared pure g-C₃N₄ and 0.1 wt% Fe₂O₃/g-C₃N₄ samples.

Based on above experimental results, a possible mechanism has been proposed for the enhanced visible-light activity of Fe₂O₃/g-C₃N₄ composite catalysts and a schematic diagram is illustrated in Fig. 10. The significant improvement of photocatalytic performance can be ascribed to the synergistic effect between Fe₂O₃ nanocrystals and g-C₃N₄ nanosheets. The band gaps of Fe₂O₃ and g-C₃N₄ are 2.2 eV (ref. 44) and 2.7 eV,⁹ respectively, so both pure samples can be excited by visible light and produce photoinduced electrons and holes. After Fe₂O₃ nanoparticles are dispersed on g-C₃N₄ nanosheets, the two types of semiconductors closely combine together and form heterojunction interface. Since both the bottom of the conduction band (CB) and the top of the valence band (VB) of g-C₃N₄ (−1.3 V and 1.4 V vs. NHE, pH = 7)^45,46 are more negative than that of Fe₂O₃ (0.28 V and 2.48 V vs. NHE, pH = 7),⁴⁴ the photogenerated electrons could transfer easily from g-C₃N₄ into the CB of Fe₂O₃, while the holes could migrate from Fe₂O₃ into the VB of g-C₃N₄ via the interface of the heterojunction. As a result, the redistribution of electrons and holes on each side of the heterojunction is achieved (as shown in Fig. 10), which greatly hinders the recombination process of electron–hole pairs and thus improves the efficiency of interfacial charge separation. In addition, the photogenerated electrons in the CB of Fe₂O₃ can reduce the dissolved O₂ to yield abundant active ˙O₂⁻, together with the holes in the VB of g-C₃N₄ employed for further effective oxidation of RhB. Therefore, owing to the high transfer and separation efficiency of charge carriers in this heterojunction system, Fe₂O₃/g-C₃N₄ composites demonstrate superior photocatalytic performance compared with that of pure g-C₃N₄ and Fe₂O₃.

Fig. 10 Proposed mechanism for the photocatalytic degradation of RhB over Fe₂O₃/g-C₃N₄ nanocomposites under visible light irradiation.

4. Conclusions

In summary, we have successfully fabricated an efficient visible-light photocatalyst for RhB degradation by loading a small amount of Fe₂O₃ onto the g-C₃N₄ nanosheets through a facile deposition-precipitation method. The crystal structure, morphology, element chemical states and optical absorption of the composites have been investigated, and the results confirm the strong interaction and the formation of heterojunction structure between highly dispersed Fe₂O₃ nanocrystals and g-C₃N₄ nanosheets. Photocatalytic RhB degradation results demonstrate that introducing of Fe₂O₃ nanoparticles brought about significant improvement of photocatalytic activity of g-C₃N₄ and that the optimum photocatalytic activity of Fe₂O₃/g-C₃N₄ composites with a weight ratio of 0.1% Fe₂O₃ is much higher than that of pure g-C₃N₄ and Fe₂O₃ under visible light irradiation. Radical trapping experiments indicate that both holes and superoxide radicals are the main oxidative species for RhB photodegradation. A possible photocatalytic mechanism is proposed based on the experimental results. The accelerated photoinduced electron–hole transfer and separation, decreased recombination rate and band energy matching lead to the enhancement of photocatalytic performance for Fe₂O₃/g-C₃N₄ heterojunction. This work provides a promising approach for developing an efficient composite heterojunction photocatalyst by tuning and control of dispersion and contact between different components for visible-light-driven photocatalytic reaction.

Acknowledgements

This work was financially supported by National Natural Science Foundation of China (No. 21503100) and Natural Science Foundation of Jiangxi Province of China (No. 20151BAB213010). Ailing Jin was supported by the Scientific Research Foundation of Graduate School of Jiangxi Province (YC2015-S123).

Notes and references

1. A. Kudo and Y. Miseki, Chem. Soc. Rev., 2009, 38, 253.
2. X. Chen, S. Shen, L. Guo and S. S. Mao, Chem. Rev., 2010, 110, 6503.
3. M. Pelaez, N. T. Nolan, S. C. Pillai, M. K. Seery, P. Falaras, A. G. Kontos, P. S. M. Dunlop, J. W. J. Hamilton, J. A. Byrne, K. O'Shea, M. H. Entezari and D. D. Dionysiou, Appl. Catal., B, 2012, 125, 331.
4. S. N. Habisreutinger, L. Schmidt-Mende and J. K. Stolarczyk, Angew. Chem., Int. Ed., 2013, 52, 7372.
5. X. Lang, X. Chen and J. Zhao, Chem. Soc. Rev., 2014, 43, 473.
6. Z. Zhao, Y. Sun, F. Dong, Y. Zhang and H. Zhao, RSC Adv., 2015, 5, 39549.
7. M. R. Gholipour, D. Cao-Thang, F. Beland and D. Trong-On, Nanoscale, 2015, 7, 8187.
8. Y. Qu and X. Duan, Chem. Soc. Rev., 2013, 42, 2568.
9. X. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J. M. Carlsson, K. Domen and M. Antonietti, Nat. Mater., 2009, 8, 76.
10. X. Wang, S. Blechert and M. Antonietti, ACS Catal., 2012, 2, 1596.
11. S. Cao, J. Low, J. Yu and M. Jaroniec, Adv. Mater., 2015, 27, 2150.
12. J. Zhang, B. Wang and X. Wang, Progr. Chem., 2014, 26, 19.
13. J. Zhu, P. Xiao, H. Li and S. A. C. Carabineiro, ACS Appl. Mater. Interfaces, 2014, 6, 16449.
14. S. Hu, L. Ma, J. You, F. Li, Z. Fan, G. Lu, D. Liu and J. Gui, Appl. Surf. Sci., 2014, 311, 164.
15. S. Zhang, J. Li, M. Zeng, J. Li, J. Xu and X. Wang, Chem.–Eur. J., 2014, 20, 9805.
16. K. Wang, Q. Li, B. Liu, B. Cheng, W. Ho and J. Yu, Appl. Catal., B, 2015, 176, 44.
17. S. Hu, R. Jin, G. Lu, D. Liu and J. Gui, RSC Adv., 2014, 4, 24863.
18. J. Zhang, F. Guo and X. Wang, Adv. Funct. Mater., 2013, 23, 3008.
19. J. Sun, J. Zhang, M. Zhang, M. Antonietti, X. Fu and X. Wang, Nat. Commun., 2012, 3, 1139.
20. D. Chen, K. Wang, W. Hong, R. Zong, W. Yao and Y. Zhu, Appl. Catal., B, 2015, 166, 366.
21. S. Min and G. Lu, J. Phys. Chem. C, 2012, 116, 19644.
22. Y. He, Y. Wang, L. Zhang, B. Teng and M. Fan, Appl. Catal., B, 2015, 168, 1.
23. Y. He, L. Zhang, M. Fan, X. Wang, M. L. Walbridge, Q. Nong, Y. Wu and L. Zhao, Sol. Energy Mater. Sol. Cells, 2015, 137, 175.
24. Z. Tong, D. Yang, T. Xiao, Y. Tian and Z. Jiang, Chem. Eng. J., 2015, 260, 117.
25. S. Ye, L.-G. Qiu, Y.-P. Yuan, Y.-J. Zhu, J. Xia and J.-F. Zhu, J. Mater. Chem. A, 2013, 1, 3008.
26. J. Theerthagiri, R. A. Senthil, A. Priya, J. Madhavan, R. J. V. Michael and M. Ashokkumar, RSC Adv., 2014, 4, 38222.
27. Y. Liu, Y.-X. Yu and W.-D. Zhang, Int. J. Hydrogen Energy, 2014, 39, 9105.
28. L. Ge, F. Zuo, J. Liu, Q. Ma, C. Wang, D. Sun, L. Bartels and P. Feng, J. Phys. Chem. C, 2012, 116, 13708.
29. L. Huang, H. Xu, Y. Li, H. Li, X. Cheng, J. Xia, Y. Xu and G. Cai, Dalton Trans., 2013, 8606.
30. M. Mishra and D.-M. Chun, Appl. Catal., A, 2015, 498, 126.
31. S. Yang, Y. Xu, Y. Sun, G. Zhang and D. Gao, CrystEngComm, 2012, 14, 7915.
32. S. K. Apte, S. D. Naik, R. S. Sonawane and B. B. Kale, J. Am. Ceram. Soc., 2007, 90, 412.
33. T. Li, L. Zhao, Y. He, J. Cai, M. Luo and J. Lin, Appl. Catal., B, 2013, 129, 255.
34. S.-W. Cao, X.-F. Liu, Y.-P. Yuan, Z.-Y. Zhang, Y.-S. Liao, J. Fang, S. C. J. Loo, T. C. Sum and C. Xue, Appl. Catal., B, 2014, 147, 940.
35. Z. Zhang, J. Huang, M. Zhang, L. Yuan and B. Dong, Appl. Catal., B, 2015, 163, 298.
36. D. J. Martin, K. Qiu, S. A. Shevlin, A. D. Handoko, X. Chen, Z. Guo and J. Tang, Angew. Chem., Int. Ed., 2014, 53, 9240.
37. S. Kumar, S. Tonda, A. Baruah, B. Kumar and V. Shanker, Dalton Trans., 2014, 16105.
38. K. Volgmann, F. Voigts and W. Maus-Friedrichs, Surf. Sci., 2010, 604, 906.
39. H. Y. Lin, Y. W. Chen and W. J. Wang, J. Nanopart. Res., 2005, 7, 249.
40. Y. Tian, B. Chang, Z. Yang, B. Zhou, F. Xi and X. Dong, RSC Adv., 2014, 4, 4187.
41. S. C. Yan, Z. S. Li and Z. G. Zou, Langmuir, 2010, 26, 3894.
42. Y.-S. Xu and W.-D. Zhang, ChemCatChem, 2013, 5, 2343.
43. L. Ge, C. Han and J. Liu, J. Mater. Chem., 2012, 22, 11843.
44. S. B. Rawal, S. Bera, D. Lee, D.-J. Jang and W. I. Lee, Catal. Sci. Technol., 2013, 3, 1822.
45. S. Chu, Y. Wang, Y. Guo, J. Feng, C. Wang, W. Luo, X. Fan and Z. Zou, ACS Catal., 2013, 3, 912.
46. Y. Cui, Z. Ding, P. Liu, M. Antonietti, X. Fu and X. Wang, Phys. Chem. Chem. Phys., 2012, 14, 1455.

---

Footnotes

† Electronic supplementary information (ESI) available: XRD pattern of 20 wt% Fe₂O₃/g-C₃N₄ nanocomposite and nitrogen adsorption/desorption isotherm of pure g-C₃N₄. See DOI: 10.1039/c5ra18466e

‡ The authors contribute equally to this work.

---