J Sol-Gel Sci Technol (2014) 72:443–454 DOI 10.1007/s10971-014-3454-x

ORIGINAL PAPER

Synthesis and characterization of g-C3N4/BiVO4 composite photocatalysts with improved visible-light-driven photocatalytic performance Man Ou • Qin Zhong • Shule Zhang

Received: 7 May 2014 / Accepted: 24 July 2014 / Published online: 12 August 2014 Ó Springer Science+Business Media New York 2014

Abstract Novel visible-light-driven g-C3N4/BiVO4 composite photocatalysts were fabricated via sol–gel and simple mixing and heating methods. The photocatalysts were characterized by X-ray diffraction, thermogravimetric, Fourier transform infrared, transmission electron microscope, Brunauer-Emmett-Teller, X-ray photoelectron spectroscopy, diffuse reflectance spectroscopy, and photoluminescence spectra. The results indicated that BiVO4 was well dispersed on g-C3N4 sheet and an interaction between g-C3N4 and BiVO4 was confirmed, which were facile to the electron transfer from g-C3N4 to BiVO4 species. The mechanism was further induced to the heterojunction effect to improve the photocatalytic efficiency. The g-C3N4/BiVO4 heterojunction at a weight ratio of 80 % calcined at 500 °C exhibited the most excellent photocatalytic ability for RhB decolorization under visiblelight irradiation (k [ 420 nm) which was extraordinary more active than that of pure components. Keywords g-C3N4/BiVO4  Heterojunction  Photocatalytic decolorization  Visible light

Electronic supplementary material The online version of this article (doi:10.1007/s10971-014-3454-x) contains supplementary material, which is available to authorized users. M. Ou  Q. Zhong (&)  S. Zhang School of Chemical Engineering, Nanjing University of Science and Technology, Nanjing 210094, Jiangsu, People’s Republic of China e-mail: [email protected] M. Ou  Q. Zhong  S. Zhang Nanjing AIREP Environmental Protection Technology Co., Ltd, Nanjing 210091, Jiangsu, People’s Republic of China

1 Introduction The photocatalytic technology has been received keen interest in recent years, due to energy saving, absence of secondary pollution and efficient photocatalytic activity for environmental remediation with the utilization of solar energy. Since Fujishimal and Honda [1] found the photocatalytic hydrogen/oxygen generation from water spitting in the n-type semiconductor TiO2 electrode under light irradiation, the new field of semiconductor photocatalysis was opened up. TiO2 has been considered as the attractive photocatalyst owning to its superior redox ability, low cost and nontoxicity, physical and chemical inertness against photo and chemical corrosion. However, the wide-bandgap semiconductor photocatalyst with 3.2 eV only absorbing UV-light (no more than 4 % of the solar spectrum) is a significant drawback for practical application in environment and clean energy area [2, 3]. Therefore, many attempts have been made to explore the photocatalysts responsive to visible light. Visible-light-driven photocatalysts discovered so far mainly consist of transition metal ions with d0 electronic configuration and post-transition metal ions with d10 configuration, such as WO3, AgPO4, and ZnO [4–6], while the appearance of graphitic carbon nitride (g-C3N4) as the first metal-free visible light induced semiconductor greatly attracts public concern [7]. The carbon nitrides are composed of inexpensive earth-abundant elements and easily obtained by heating cyanamide [8]. Furthermore, the soft polymer easily coats on the surface of other compounds as well as exhibits high thermal and chemical stability [9, 10]. Nevertheless, the high recombination rate of photo-generated electron–hole pairs is a main issue limiting the photocatalytic efficiency of g-C3N4 [11]. Therefore, many studies have been carried out to improve the charge

123

444

separation efficiency and then enhance the visible light photocatalytic activity. For example, preparing nanosheets structures [12], doping controlled metal or nonmetal impurities [13, 14], coupling with graphene [15] and forming composite heterojunction, such as g-C3N4/ ZnWO4, g-C3N4/Bi2WO6 and g-C3N4/AgPO4 [16–18] have been reported. Unlike other approaches, the composite heterojunction of g-C3N4 with the material possessing appropriate band position can combine the merits of each component to exhibit synergistic effects. BiVO4, as an novel stable and non-toxic semiconductor material, has attracted considerable interest because of its high response to visible light [19]. However, the photocatalytic efficiency of pure BiVO4 is not high due to its poor adsorptive performance and the difficulty in migrating photogenerated charge carriers [20]. It has been reported that the values of the construction band and the valence band levels of BiVO4 are 0.32 and 2.75 eV, respectively, both of which are lower than that of g-C3N4 [21]. Hence, the BiVO4 hybridized with g-C3N4 may be more helpful for the charge carriers transfer and separation at the heterojunction surface, consequently improving photocatalytic performance. In the present study, we demonstrated a simple mixing and heating method for the synthesis of g-C3N4/BiVO4 composite catalysts. And the photocatalytic performance of g-C3N4/BiVO4 was tested on the photodecolorization of RhB under visible light irradiation. The hybrid photocatalysts showed considerable photocatalytic activity compared with pure g-C3N4 and BiVO4, and the photoactivity remained almost barely change after cycling photocatalytic experiments. Furthermore, the mechanism of the photodecolorization of RhB over g-C3N4/BiVO4 heterojunctions was also discussed.

2 Experimental 2.1 Synthesis of g-C3N4/BiVO4 photocatalysts All chemicals are reagent grade and used without further purification. The synthesis of graphitic carbon nitride (gC3N4) polymer was performed according to the methods reported by some researchers [7, 16]. g-C3N4 was obtained by directly heating melamine at 550 °C for 4 h at a heating rate of 3 °C/min in a semi-closed alumina crucible with a cover, and further deammoniation treatment was set at 550 °C for 4 h. After cooling to room temperature, the products were collected and ground into yellow powders. Pure phase monoclinic BiVO4 was synthesized by sol– gel method: 0.01 mol of BiNO35H2O was firstly dissolved in 40 ml of 2 mol/L HNO3, followed by the addition of 0.02 mol citric acid to form a homogeneous solution A.

123

J Sol-Gel Sci Technol (2014) 72:443–454

Secondly, NH4VO3 was dissolved in 80 °C distilled water with mechanical agitation until the dissolution was complete, and then followed by the addition of 0.02 mol citric acid to form a black solution B. After that, solution A was added drop-wisely into solution B under continuously magnetic stirring, in which process the color has changed several times and finally obtained blue suspension with a Bi/V molar ratio of 1:1. The pH of the mixture suspension was adjusted to 7 by aqueous NH3H2O and then obtained the dark-blue solution. The resultant solution was stirred for 1 h vigorously at 80 °C and subsequently heated in the water bath at 80 °C for 3 h to get BiVO4 sol–gel. Afterwards the sample was dried at 110 °C overnight, and finally the BiVO4 precursor was submitted to a further calcination at 500 °C for 5 h in a muffle furnace. The typical preparation process of g-C3N4/BiVO4 photocatalysts was as follows: 0.8 g g-C3N4 and 0.2 g BiVO4 were adequately mixed and ground in an agate mortar and then calcined at 500 °C for 2 h to obtain the 80 wt% g-C3N4/BiVO4 photocatalyst. Other g-C3N4/ BiVO4 catalysts with different g-C3N4 content or calcined temperature were synthesized by the similar method. Briefly, the catalysts were abbreviated by the way of x wt% CB-T, where C represented g-C3N4, B represented BiVO4, x denoted the g-C3N4/BiVO4 mass ratio and T presented the calcined temperature. As a reference, the mixture of g-C3N4 and BiVO4 (80 % g-C3N4 content) by the simple physical mixture without heat treatment was also prepared, which was noted as 80 wt% CB-25. 2.2 Characterization X-ray diffraction (XRD) patterns were recorded on a Beijing Purkinjie general instrument XD-3 X-ray diffraction (Cu Ka, voltage 36 kV, electrical current 20 mA, 2h from 5° to 80°). The Brunauer-Emmett-Teller (BET) surface area was determined by N2 adsorption–desorption measurements at 77 K (Gold App V-sorb 2008). The thermogravimetric (TG) analysis of samples were performed from 50 to 800 °C with a Netzsch STA 449C Jupiter instrument under the flow of air gas at a heating rate of 10 °C/min. Transmission electron microscope (TEM) observations were carried out using a Philips CM-10 at 80 kV and a CM-12 at 120 kV. Fourier transform infrared (FT-IR) spectroscopy was recorded on an IS10 FTIR spectrometer, (Nicolet, USA). X-ray photoelectron spectroscopy (XPS) was performed on a RBD upgraded PHI-5000C ESCA system (Perkin-Elmer) with Mg Ka radiation (hm = 1,253.6 eV). Diffuse reflectance spectroscopy (DRS) was performed on a Shimadzu UV-2550 UV–vis spectrophotometer using BaSO4 as the reference sample. Photoluminescence spectra (PL) were carried out on a Labram-HR800-type spectrophotometer (Jobin–Yvon Co, France) with a He-Cd laser (k = 325 nm) as the light source.

J Sol-Gel Sci Technol (2014) 72:443–454

445

Fig. 2 TG curves for g-C3N4 and 80 wt% g-C3N4/BiVO4 composites calcined at different temperature

wavelength of 554 nm during the photodecolorization process.

3 Results and discussion 3.1 XRD, TG analysis

Fig. 1 XRD patterns of a g-C3N4, BiVO4 and g-C3N4/BiVO4 photocatalysts, b 80 wt% g-C3N4/BiVO4 composites calcined at different temperature

2.3 Photocatalytic activity The photocatalytic activities of g-C3N4/BiVO4 catalysts for RhB decolorization were evaluated under visible-light irradiation using a 350 W Xe lamp with a 420 nm cut-off filter as the light source. Experiments were conducted at ambient temperature and procedures were as follows: an aqueous suspension of RhB (50 mL, 10 mg/L) was placed in a quartz tube, and then 50 mg photocatalysts were added. Before illumination, the suspensions were mechanically stirred in the dark for 60 min to ensure the adsorption–desorption equilibrium between the catalysts and the dye. Then 5 mL aliquots were withdraw and centrifuged to remove the photocatalyst powders for analysis every 30 min. The concentration of remnant dye was subsequently determined by UV–vis spectroscopy at the

Figure 1 shows the X-ray diffraction patterns of as-synthesized pure g-C3N4, BiVO4 and their composites with different mass ratios. It is observed that pure g-C3N4 has the strongest peak at 27.4°, which corresponds to the characteristic interlayer stacking peak of aromatic systems, as in graphite [7]. Furthermore, all the diffraction peaks of BiVO4 match well to the monoclinic phase, which are accordance with the XRD patterns reported in the literature [22]. No typical crystalline peaks of g-C3N4 appear in the g-C3N4/BiVO4 composites when the content of g-C3N4 is \80 %. It may be due to the relatively high content and good crystallinity of BiVO4 in the composites. Therefore, the g-C3N4 (\80 %) peaks are not observed in the XRD pattern. In addition, the half-width and the position of the XRD peaks of both g-C3N4 and BiVO4 remain unchanged, suggesting that the crystal phase of g-C3N4 and BiVO4 is kept the same, respectively, and no other phases are observed in the patterns of the g-C3N4/BiVO4 composites. Figure 1b exhibits the XRD patterns of 80 wt% g-C3N4/ BiVO4 composites calcined at different temperatures. The XRD peak of g-C3N4 at low temperatures (B450 °C) is higher than one at 500 °C. This result illustrates that the g-C3N4 is destructed at higher temperature, which can be proven by TG experiment (Fig. 2). The weight of g-C3N4 begins to loss at 550 °C, implying that the g-C3N4 can be used to construct a composite

123

446

J Sol-Gel Sci Technol (2014) 72:443–454

Table 1 Specific surface area and the real g-C3N4 content of g-C3N4/BiVO4 composites Catalysts

S (m2g-1)

g-C3N4 content ( wt%)

Catalysts

S (m2g-1)

g-C3N4 content (wt%)

g-C3N4

8.95

100

40 % CB-500

5.01

4.4

BiVO4

4.92

–

20 % CB-500

3.97

–

90 % CB-500

14.17

86.3

80 % CB-25

8.63

83.9

80 % CB-500

7.68

72.2

80 % CB-400

5.30

83.6

60 % CB-500

5.83

28.6

80 % CB-450

9.93

83.3

photocatalyst at proper temperature due to the well stability confirmed by Fig. 2. Li et al. [23] have reported that only part of g-C3N4 was used to construct the g-C3N4/SmVO4 heterojunction with SmVO4 during the heating and oxidizing process. Therefore, the real g-C3N4 content is necessary to detect. The mass loss of mechanical mixed 80 wt% g-C3N4/BiVO4 is almost the same as the theoretical value. Hence, the amount of g-C3N4 in the composites can be easily calculated from the corresponding weight reminder after heating the samples over 800 °C. The precursors of 80 wt% g-C3N4/BiVO4 under different calcination temperatures are also investigated. It can be seen that a rapid weight loss region occurring from 510 to 640 °C for the composites at different temperatures is observed, which is similar to that reported in WO3/g-C3N4 composites [24]. The weight loss is almost 80 % both at 400 and 450 °C, which is consistent with the theoretical value, while the g-C3N4 content is approximately calculated to be 70 % at 500 °C. Other composites with different mass ratios at 500 °C (Fig. S1) show that the concentration of g-C3N4 decreases with the increase of BiVO4, which may contribute to that the catalytic oxidation of g-C3N4 is facilitated with the increase of BiVO4 (Table 1). It is consistent with the results reported by Huang [25]. The interaction between g-C3N4 and BiVO4 during the heating and oxidizing process will be discussed in XPS. 3.2 FT-IR analysis Fig. S2 shows the FT-IR spectra of g-C3N4, BiVO4 and a series of g-C3N4/BiVO4 photocatalysts with different g-C3N4 content. The FT-IR spectrum of the g-C3N4 exhibits the similar features with those in the literatures [26–28]. The band at 1,640 cm-1 corresponds to the typical stretching vibration modes of C–N heterocycles. The peaks at 1,250, 1,330, 1,420 and 1,570 cm-1 are attributed to aromatic C–N stretching [26]. And the band at 810 cm-1 is ascribed to the out-of plane bending modes of C–N hetero-cycles [27]. Additionally, the band at around 3,180 cm-1 corresponds to the characteristic stretching modes of terminal NH2 or NH groups at the defect sites of the aromatic ring [26, 28]. In the case of BiVO4, the absorption bands at 734 cm-1 and 837 cm-1 are assigned

123

to the m1(VO4) and m3(VO4) in the structure [29]. All the bands in Fig. S2 corresponding to g-C3N4 and BiVO4 are fully presented in the g-C3N4/BiVO4 samples, which further confirms that the composite photocatalyst is composed of g-C3N4 and BiVO4. This result further testifies the XRD result. 3.3 TEM and BET analysis The morphologies of g-C3N4, BiVO4 and the 80 wt% g-C3N4/BiVO4 sample calcined at 500 °C were investigated by TEM and HRTEM. Figure 3a, b show the TEM images of g-C3N4 and BiVO4, where g-C3N4 has a crumpled layered structure and BiVO4 particles with a diameter of about 100–200 nm present a better dispersion, although some small aggregates are also observed. Figure 3c shows the TEM images of 80 wt% CB-500, where BiVO4 particles are lay on the surface of g-C3N4 sheet and well dispersed. The HRTEM image (Fig. 3d) of the composite demonstrates that the lattice spacing of 0.304 nm corresponds to the (121) lattice plane of BiVO4. The phase in the middle without fringes is assigned to g-C3N4. It can be seen that BiVO4 particles correlates well with the g-C3N4 sheet, which would favor the formation of a g-C3N4/BiVO4 heterojunction. The similar phenomena have been reported by He et al. [30]. The results of the BET-surface area of the composites with different g-C3N4 content were listed in Table 1. It is obvious that BiVO4 has a lower surface area than g-C3N4 and the BET value of the heterojunction catalysts calcined at 500 °C gradually decreases with the increase of BiVO4. The specific surface area of 80 wt% g-C3N4/BiVO4 at different calcination temperatures is also observed. The change of BET-surface area is not aligned with the photocatalytic activity. It indicates that the specific surface area is not the most significant factor for the photocatalytic ability. 3.4 XPS analysis The surface chemical composition of g-C3N4, BiVO4 and 80 wt% CB-500 and the interaction of g-C3N4 with BiVO4 were determined by XPS. The overall XPS spectra of the samples are shown in Fig. 4a. The main peaks on the

J Sol-Gel Sci Technol (2014) 72:443–454

447

Fig. 3 TEM images of g-C3N4 (a), BiVO4 (b), 80 wt% CB-500 photocatalyst (c), HRTEM images of 80 wt% CB-500 composite (d)

surface of g-C3N4 are C 1s and N 1s, and the key peaks on the surface of BiVO4 are Bi 4f, V 2p and O 1s. All of the above peaks exist on the surface of 80 wt% CB-500, which is consistent with the chemical composition of the photocatalysts. Figure 4b displays the high-resolution XPS spectra of C 1s for g-C3N4 and 80 wt% CB-500. With respect to g-C3N4, two peaks at 284.0 and 287.4 eV are assigned to the adventitious carbon species and the carbon bonded to three neighbouring nitrogen [31]. However, the two peaks with the binding energy of 284.3 and 287.8 eV are observed in the composite, which are higher than that of pure g-C3N4. Similarly, in Fig. 4c, the main N 1s peak at 397.9 eV and the second contribution at 399.6 eV for pure g-C3N4 are identified as sp2-hybridized nitrogen (C = N– C) and tertiary nitride (N-C3) [32]. Compared with g-C3N4, the peak of g-C3N4/BiVO4 at 399.6 eV does not change while the key peak of the composite shifts to higher banding energy at 398.1 eV. It may be due to the fact that BiVO4 interacting with g-C3N4 induces to the inner shift of C 1s and N 1s orbitals. Figure 4d shows the high-resolution XPS of Bi 4f, in which two bands at 158.5 and 163.9 eV, ascribed to Bi 4f5/2 and Bi 4f7/2 binding energies of pure BiVO4, respectively, are observed. Since the binding energy values of Bi 4f5/2 and Bi 4f7/2 in 80 wt% CB-500

remain the same as those of BiVO4 and a similar phenomenon is also found in the XPS spectra of O 1s (Fig. 4e), the possibility of the surface charging effect inducing a shift in these binding energies can be negligible, just as Sun reported [16]. From Fig. 4f, the binding energies of V 2p2/3 (516.0 eV) and V 2p1/2 (523.5 eV) of 80 wt% CB-500 are less than that of V 2p2/3 (516.2 eV) and V 2p1/2 (523.7 eV) of pure BiVO4. The negative shift is also due to the interaction between BiVO4 and g-C3N4. It is noted that the positive shift of C 1s and N 1s indicates the decrease in the electron density on C, N and the negative shift of V 2p exhibites the increase in the electron density on V. Such results exhibit an interaction between C, N and V atom in the composite. It can be concluded that there exists a chemical interaction between g-C3N4 and BiVO4 in the heterojunction rather than a simple physical mixture. 3.5 DRS analysis Figure 5a depicts the UV–vis diffuse reflectance spectra of g-C3N4, BiVO4 and their composites at 500 °C with different g-C3N4 content. The adsorption edge of g-C3N4 or BiVO4 powders is higher than 475 nm, which signifies their visible-light-induced photocatalytic activity.

123

448

J Sol-Gel Sci Technol (2014) 72:443–454

Fig. 4 XPS spectra of g-C3N4, BiVO4 and 80 wt% CB-500 photocatalyst: a the survey scan; b C 1s; c N 1s; d Bi 4f; e O 1s; f V 2p

123

J Sol-Gel Sci Technol (2014) 72:443–454

449

Fig. 5 a UV–vis diffuse reflectance spectra of g-C3N4, BiVO4 and g-C3N4/BiVO4 composites. b The plots of (ahm)2 versus energy (hm) for the band gap energy of the g-C3N4/BiVO4 photocatalysts with different g-C3N4 content

Compared with pure g-C3N4 and BiVO4, the absorption edges of g-C3N4/BiVO4 composites are extended apparently towards visible light range with longer wavelength, implying that the g-C3N4/BiVO4 heterojunctions display better optical absorption ability. The value of the band gap energies of g-C3N4 and BiVO4 can be evaluated by the equation: ahv ¼ Aðhv  EgÞn=2 where a, hv, Eg and A are absorption coefficient, discrete photo energy, band gap energy and a constant. In addition, n is the parameter depending on the characteristics of the optical transition in a conductor (n = 1 for a direct transition and n = 4 for an indirect transition). For g-C3N4 [7, 32] and BiVO4 [33], both of them pertain to direct transition. Consequently, the band-gap energy (Eg) of the resulting samples can be estimated from a plot of (ahm)2 versus energy (hm). The interception of the tangent to the x-axis would give a good approximation of the Eg of the samples. In inset of Fig. 5b, the band-gaps of pure g-C3N4 and BiVO4 are estimated about 2.65 and 2.44 eV, respectively, which agree with the values reported in previous literatures [18, 33]. However, the band gaps of the composites in Fig. 5b present lower than that of the pure phase alone. Furthermore, 80 wt% CB-500 possesses the lowest band gap with respect to other different mass ratios of g-C3N4 and BiVO4. The results indicate that the introduction of the suitable amount of BiVO4 generates impurity band and narrows the band gap, which is similar to previous reports [34]. The conduction band potentials of a semiconductor at the point of zero charge can be predicted by Mulliken electronegativity theory:

ECB ¼ v  Ec  0:5Eg Herein, ECB is the conduction band potentials, the v is the absolute electronegativity of the semiconductor, which is the geometric mean of the constituent atoms, Ec is the energy of the free electron in the hydrogen scale (about 4.5 eV), and Eg is the band-gap energy of the semiconductor. The valence band potential can be calculated by EVB = ECB ? Eg. The v value for BiVO4 is 6.04 eV [21]. Hence, the ECB value of BiVO4 is calculated to be 0.32 eV, and the EVB value is estimated to be 2.76 eV. Based on the band gap positions, the positions of the conduction and the valence bands of g-C3N4 are determined to be 1.57 and 1.13 eV, respectively [11]. 3.6 Photocatalytic activity of the composite photocatalysts The photoactivity of pure g-C3N4, BiVO4 and their composites was investigated through RhB decolorization under visible-light irradiation (k [ 420 nm). The results are shown in Fig. 6. It is clearly observed that the blank experiment without photocatalysts demonstrates that RhB is stable under visible light illumination. BiVO4 shows extraordinarily low photocatalytic activity about 20 %. P25 also shows a rather poor performance similar to that of BiVO4 (Fig. 6a). It indicates that the dye sensitization effect is limited because P25 can not absorb visible light, just as some researchers reported [18]. The photocatalytic performance of g-C3N4/BiVO4 samples increases and then decreases with the increase of g-C3N4. The highest photocatalytic activity 96.4 % is obtained over the 80 wt% g-C3N4/BiVO4 heterojunction after 120 min under visiblelight irradiation. To confirm the significance of the role of

123

450

J Sol-Gel Sci Technol (2014) 72:443–454

Fig. 6 a Photocatalytic activity of g-C3N4/BiVO4 photocatalysts with different g-C3N4 content on the decolorization behavior of RhB; b Photocatalytic activity of g-C3N4, BiVO4 and 80 wt% g-C3N4/ BiVO4 at different temperature; c The UV–vis spectra of RhB

solution by 80 wt% CB-500 in different performing time; d The photographs of the products before and after 120 min photocatalytic process by 80 wt% CB-500 heterojunction

heterojunction on the photocatalytic activity, the photocatalytic decolorization experiments over the mixed 80 wt% g-C3N4/BiVO4 and calcined at different temperatures are investigated in Fig. 6b. The photocatalytic activity of the mixed powders and the samples calcined at lower temperatures exhibits much lower than that of calcined at 500 °C. It can be obtained that the heating treatment significantly affects the heterojunction interface between g-C3N4 and BiVO4, which can restrain the recombination of photoinduced charges effectively [32], thus enhancing the photocatalytic performance. Figure 6c shows the UV–vis spectra taken over performing time in the process of RhB decolorization by 80 wt% CB-500 particle. After irradiation for 120 min, the maximum spectral peak of RhB shifts from 554 to 498 nm, indicating that the degradation of RhB in this work is ascribed to N-deethylation process [35]. From the photographs of the

degradation system in Fig. 6d, it can be seen that the color of the dye solution changes from bright-red to light-yellow after 120 min. The above results indicate that the 80 wt% CB-500 bulk heterojunction possesses superior photocatalytic activity under visible-light irradiation. In addition to the photocatalytic performance, the stability of photocatalysts is also important for practical application. The cycling runs for the decolorization behavior of RhB by using 80 wt% CB-500 are performed to evaluate the stability of the bulk heterojunction. As shown in Fig. 7a, the activity does not obviously decrease after four successive cycles experiment. Furthermore, the XRD patterns and FT-IR spectra of the sample before and after the reaction (Fig. 7b, c) show that the crystal structure does not change and no additional band appears during the experimental process. These suggest that the heterojunction is not photocorroded and presents relatively high efficiency and stability.

123

J Sol-Gel Sci Technol (2014) 72:443–454

451

Fig. 8 Photodecolorization behavior of RhB over 80 wt% CB-500 composite with different scavengers under visible-light irradiation

3.7 Discussion of photocatalytic mechanism 3.7.1 Roles of reactive species To elucidate the mechanism of the visible-light-driven composite photocatalyst g-C3N4/BiVO4, the trapping experiment of the main active species such as superoxide radical (O2-), holes (h?) and hydroxyl radical (OH) during the photocatalytic reaction is necessary to explore. Different scavengers of ammonium oxalate (AO, a quencher of h?), benzoquinone (BQ, a quencher of O2-) and 2-propanol (IPA, a quencher of OH) are introduced. The use of the scavengers refers to previous studies [36, 37]. From Fig. 8, the photocatalytic efficiency of RhB on 80 wt% CB-500 is about 96 % without scavengers, and the addition of 1 mM IPA induces a small change in the degradation rate. On the contrary, 1 mM AO and 1 mM BQ added into the RhB solution lead to a decrease of the photocatalytic activity to 65 and 30 %, respectively. These results indicate that holes and O2- are the predominant active species, while OH does not play a major role in the photodecolorization behavior in the presence of 80 wt% CB-500 heterojunction. 3.7.2 Enhanced photocatalytic activity mechanism for gC3N4/BiVO4 heterojunction

Fig. 7 a Cycling runs of 80 wt% CB-500 for photodecolorization RhB under visible light. b XRD patterns of 80 wt% CB-500 before and after the recycling photodecolorization of RhB; c FT-IR spectra of 80 wt% CB-500 before and after the recycling photodecolorization of RhB

It is widely accepted that high adsorption activity and efficient charge separation of photoinduced charge carriers play important roles in the enhancement of photocatalytic performance. The adsorption ability of g-C3N4/BiVO4 has been already investigated in dark for 60 min, which is

123

452

J Sol-Gel Sci Technol (2014) 72:443–454

surface O2 of composite semiconductor system to O2because the CB of g-C3N4 is more negative than E0 (O2/ O2- = -0.284 eV) [32]. The holes and O2- as the main reactive roles are significantly contributed to the decolorization of RhB. The OH may be generated according to the following routs: (1) H2O ? h? ? OH ? H? E0 = 2.72 eV [11] (2) O2 ? 2e- ? 2H? ? H2O2 E0 = 0.695 eV [11] H2O2 ? e- ? OH ? OH(3) H2O ? h? ? H2O2 ? 2H? E0 = 1.35 eV [38] Scheme 1 Schematic diagram of photogenerated electron–hole pairs separation and transport mechanism over g-C3N4/BiVO4 composite

shown in Fig. 6a and b. It should be noted that the inconsistency between the surface area and the adsorption activity is observed in this work. Our results exhibit that the adsorption of RhB on the surface of the composite photocatalysts originates from not only a simple physical adsorption, but also the p-p stacking between RhB and g-C3N4, which is similar to the conjugation between aromatic molecules and grapheme [16]. The improved adsorption of RhB in the composite system is a major reason for the increased photocatalytic activity. The photocatalytic results have shown that the 80 wt% CB-500 performs more excellent photocatalytic activity than pure g-C3N4 and BiVO4 under visible light irradiation, which is due to the synergistic effect between BiVO4 and g-C3N4 sheet. The high charge carriers separation efficiency contributed to the synergistic effect originates from suitable matching band potentials and high-quality heterojunction interface between g-C3N4 and BiVO4 in this sample. Combining the above effect, we construct the potential energy diagram for the composite sample in Scheme 1. When the composite system is irradiated with the visible light, electrons are excited from the VB of g-C3N4 and BiVO4 to the CB of theirs, respectively, leaving holes in the VB in both samples. Furthermore, the CB and VB edge potentials of g-C3N4 are more negative than that of BiVO4. Therefore, the suitable band makes the photoinduced electrons from the CB of g-C3N4 particle easily transfer to the CB of BiVO4 through the welldeveloped interfaces. Simultaneously, the remained holes in the VB of BiVO4 can freely migrate to the VB of g-C3N4 by the internal static electric fields. This charge transfer in the two semiconductors can hold back the electron–hole pairs recombination, and maximized charge carriers separation of photoinduced charge carriers is obtained. The electrons in the process of migrating from the CB of g-C3N4 to the CB of BiVO4 could reduce the adsorbed

123

H2O2 ? e- ? OH ? OH-

The low concentration of OH may be due to the decrease of reducibility of electrons and oxidation ability of holes in the process of synergy effect [38]. 3.8 PL analysis Photoluminescence technique is considered to be a useful method to reveal the migration, transference, and recombination processes of the photo-generated electron–hole pairs in the semiconductor particles. It is widely accepted that the effective separation and transfer of photo-induced carriers are the most crucial factor influencing the photocatalytic activity. The fluorescence intensity with lower values indicates lower recombination rate of photo-induced electron–hole pairs and higher photocatalytic activity. In order to prove the electron–hole separation efficiency of photocatalyst, PL spectra of g-C3N4 and g-C3N4/BiVO4 composite samples are necessary. Fig. 9 illustrates the PL emission spectra of g-C3N4, 80 wt% CB-25 and 80 wt% CB-500 heterojunction. The g-C3N4 shows a strong emission peak centered at approximately 468 nm, which can be attributed to the band–band PL phenomenon with the energy of light approximately equal to the band gap energy of pure g-C3N4 [17], as shown by UV–vis spectra we did. Compared with pure g-C3N4, the 80 wt% CB-25 and 80 wt% CB-500 heterojunctions shift towards longer wavelength by about 20 nm, and this red shift is associated with the decrease of band gap. It is also found that PL intensity of both 80 wt% CB-25 and 80 wt% CB-500 heterojunction is lower than that of g-C3N4. This demonstrates that it is beneficial for the separation of electron– hole pairs after the BiVO4 is introduced. And the lower PL peak of 80 wt% CB-500 heterojunction further illustrates the importance of the heterojunction to retard the recombination of photo-generated charge carriers. It will be more conducive for photo-induced charge carriers seperation on the heterojunction interfaces and then the photocatalytic performance is greatly improved.

J Sol-Gel Sci Technol (2014) 72:443–454

Fig. 9 Photoluminescence spectra of g-C3N4, 80 wt% CB-25 sample and 80 wt% CB-500 heterojunction

4 Conclusions We succesfully synthetized a series of g-C3N4/BiVO4 visible-light-induced composite photocatalysts by a mixing-calcination method. The g-C3N4/BiVO4 photocatalysts possess significantly improved visible light photocatalytic activity in RhB decolorization with respect to pure g-C3N4 and BiVO4. The 80 wt% g-C3N4/BiVO4 calcined at 500 °C with narrow band gap exhibits the highest photocatalytic performance. Such a remarkable enhancement of photocatalytic efficiency under visible light is mainly attributed to the match of conduction and valence band levels between the g-C3N4 and the BiVO4, which can induce the high separation of photo-generated electron–hole pairs in the heterojunction system. The study demonstrates that the combination of g-C3N4 and BiVO4 may be an ideal system for practical application in environmental purification. Acknowledgments This work was financially supported by the Assembly Foundation of The Industry and Information Ministry of the People’s Republic of China 2012 (543), the National Natural Science Foundation of China (U1162119), Scientific Research Project of Environmental Protection Department of Jiangsu Province (2013003) and (201112), Research Fund for the Doctoral Program of Higher Education of China (20113219110009), Industry-Academia Cooperation Innovation Fund Projects of Jiangsu Province (BY2012025) and the research fund of Key Laboratory for Advanced Technology in Environmental Protection of Jiangsu Province (AE201001).

References 1. Fujishima A, Honda K (1972) Electrochemical photolysis of water at a semiconductor electrode. Nature 238:37–38

453 2. Lv KL, Li XF, Deng KJ, Sun J, Li XH, Li M (2010) Effect of phase structures on the photocatalytic activity of surface fluorinated TiO2. Appl Catal B Environ 95:383–392 3. Hong SJ, Lee S, Jang JS, Lee JS (2011) Heterojunction BiVO4/ WO3 electrodes for enhanced photoactivity of water oxidation. Energy Environ Sci 4:1781–1787 4. Chatchai P, Nosaka AY, Nosaka Y (2013) Photoelectrocatalytic performance of WO3/BiVO4 toward the dye degradation. Electrochim Acta 94:314–319 5. Yan J, Wang C, Xu H, Xu YG, She XJ, Chen JJ, Song YH, Li HM, Zhang Q (2013) AgI/AgPO4 heterojunction composites with enhanced photocatalytic activity under visible light irradiation. Appl Surf Sci 287:178–186 6. Yu JG, Yu XX (2008) Hydrothermal synthesis and photocatalytic activity of zinc oxide hollow sphere. Environ Sci Technol 42:4902–4907 7. Wang XC, Maeda K, Thomas A, Takanabe K, Xin G, Carlsson J, Domen JK, Antonietti M (2009) A metal-free polymeric photocatalyst for hydrogen production from water under visible light. nat. Mater 8:76–80 8. Ge L, Han CC (2012) Synthesis of MWNTs/g-C3N4 composite photocatalysts with efficient visible light photocatalytic hydrogen evolution activity. Appl Catal B Environ 117–118:268–274 9. Xu L, Xia JX, Xu H, Yin S, Wang K, Wang LG, Huang LY, Li HM (2014) Reactable ionic liquid assisted solvothermal synthesis of graphite-like C3N4 hybridized a-Fe2O3 hollow microspheres with enhanced supercapacitive performance. J Power Sources 245:866–874 10. Huang ZJ, Li FB, Chen BF, Lu T, Yuan Y, Yuan GQ (2013) Well-dispersed g-C3N4 nanophases in mesoporous silica channels and their catalytic activity for carbon dioxide activation and conversion. Appl Catal B Environ 136–137:269–277 11. Wang SM, Li DL, Sun C, Yang SG, Guan Y, He H (2014) Synthesis and characterization of g-C3N4/Ag3VO4 composites with significantly enhanced visible-light photocatalytic activity for triphenylmethane dye degradation. Appl Catal B Environ l144:885–892 12. Niu P, Zhang LL, Liu G, Cheng HM (2012) Graphene-like carbon nitride nanosheets for improved photocatalytic activities. Adv Funct Mater 22:4763–4770 13. Han CC, Ge L, Chen CF, Li YJ, Xiao XL, Zhang YN, Guo LL (2014) Novel visible light induced Co3O4-g-C3N4 heterojunction photocatalysts for efficient degradation of methyl orange. Appl Catal B Environ 147:546–553 14. Yuan YP, Cao SW, Liao YS, Yin LS, Xue C (2013) Red phosphor/g-C3N4 heterojunction with enhanced photocatalytic activities for solar fuels production. Appl Catal B Environ 140–141:164–168 15. Li XH, Chen JS, Wang XC, Sun JH, Antonietti M (2011) Metalfree activation of dioxygen by graphene/g-C3N4 nanocomposites: functional dyads for selective oxidation of saturated hydrocarbons. J Am Chem Soc 133:8074–8077 16. Sun LM, Zhao X, Jia CJ, Zhou YX, Cheng XF, Li P, Fan WL (2012) Enhanced visible-light photocatalytic activity of g-C3N4-ZnWO4 by fabricating a heterojunction: investigation based on experimental and theoretical studies. J Mater Chem 22:23428–23438 17. Ge L, Han CC, Liu J (2011) Novel visible light-induced g-C3N4/ Bi2WO6 composite photocatalysts for efficient degradation of methyl orange. Appl Catal B Environ 108–109:100–107 18. Zhang FJ, Xie XZ, Zhu SF, Liu J, Zhang J, Mei SF, Zhao W (2013) A novel photofunctional g-C3N4/Ag3PO4 bulk heterojunction for decolorization of RhB. Chem Eng J 228:435–441 19. Zhang L, Chen DR, Jiao XL (2006) Monoclinic structured BiVO4 nanosheets: hydrothermal preparation, formation mechanism, and coloristic and photocatalytic properties. J Phy Chem B 110: 2668–2673

123

454 20. Fu YS, Sun XQ, Wang X (2011) BiVO4-graphene catalyst and its high photocatalytic performance under visible light irradiation. Mater Chem Phys 131:325–330 21. Guan ML, Hu SW, Chen YJ, Huang SM (2011) From hollow olive-shaped BiVO4 to n-p core-shell BiVO4@Bi2O3 microspheres controlled synthesis and enhanced visible-light-responsive photocatalytic properties. Inorg Chem 50:800–805 22. Wang M, Zheng HY, Liu Q, Niu C, Che YS, Dang MY (2013) High performance B doped BiVO4 photocatalyst with visible light response by citric acid complex method. Spectrochim. Acta. Part A Mol Bio 11:74–79 23. Li TT, Zhao LH, He YM, Cai J, Luo MF, Lin JJ (2013) Synthesis of g-C3N4/SmVO4 composite photocatalyst with improved visible light photocatalytic activities in RhB degradation. Appl Catal B Environ 129:255–263 24. Huang LY, Xu H, Li YP, Li HM, Cheng XN, Xia JX, Xu YG, Cai GB (2013) Visible-light-induced WO3/g-C3N4 composites with enhanced photocatalytic activity. Dalton Trans 42:8606–8616 25. Huang LY, Li YP, Xu H, Xu YG, Xia JX, Wang K, Li HM, Cheng XN (2013) Synthesis and characterization of CeO2/g-C3N4 composites with enhanced visible-light photocatalytic activity. RSC Adv 3:22269–22279 26. Yan HJ, Yang HX (2011) TiO2-g-C3N4 composite materials for photocatalytic H2 evolution under visible light irradiation. J Alloys Compd 509:L26–L29 27. Kang HW, Lim SN, Song DS, Park SB (2012) Organic-inorganic composite of g-C3N4-SrTiO3: Rh photocatalyst for improved H2 evolution under visible light irradiation. J Hydrogen Energy 37:11602–11610 28. Yang M, Huang Q, Jin XQ (2012) ZnGaNO solid solution-C3N4 composite for improved visible light photocatalytic performance. Mater Sci Eng B 177:600–605 29. Liu JB, Wang H, Wang S, Yan H (2003) Hydrothermal preparation of BiVO4 powders. Mater Sci Eng B 104:36–39 30. He YM, Cai J, Zhang LH, Wang XX, Lin HJ, Teng BT, Zhao LH, Weng WZ, Wan HL, Fan MH (2014) Comparing two new

123

J Sol-Gel Sci Technol (2014) 72:443–454

31.

32.

33.

34.

35.

36.

37.

38.

composite photocatalysts, t-LaVO4/g-C3N4 and m-LaVO4/gC3N4, for their structures and performances. Ind Eng Chem Res 53:5905–5915 Miranda C, Mansilla H, Ya´n˜ez J, Obrego´n S, Colo´n G (2013) Improved photocatalytic activity of g-C3N4/TiO2 composites prepared by a simple impregnation method. J Photochem Photobiol A Chem 253:16–21 Wang XJ, Wang Q, Li FT, Yang WY, Zhao Y, Hao YJ, Liu SJ (2013) Novel BiOCl-C3N4 heterojunction photocatalysts: in situ preparation via an ionic-liquid-assisted solvent-thermal route and their visible-light photocatalytic activities. Chem Eng J 234: 361–371 Wang WZ, Huang XW, Wu S, Zhou YX, Wang LJ, Shi HL, Liang YJ, Zou B (2013) Preparation of p-n junction Cu2O/BiVO4 heterogeneous nanostructures with enhanced visible-light photocatalytic activity. Appl Catal B Environ 134–135:293–301 Xu YG, Xu H, Yan J, Li HM, Huang LY, Xia JX, Yin S, Shu HM (2013) A plasmonic photocatalyst of Ag/AgBr nanoparticles coupled with g-C3N4 with enhanced visible-light photocatalytic activity. Colloid Surf A-Physicochem Eng Asp 436:474–483 Liu W, Cao LX, Su G, Liu HS, Wang XF, Zhang L (2010) Ultrasound assisted synthesis of monoclinic structured spindle BiVO4 particles with hollow structure and its photocatalytic property. Ultrason Sonochem 17:669–674 Liu W, Wang ML, Xu CX, Chen SF (2012) Facile synthesis of g-C3N4/ZnO composite with enhanced visible light photooxidation and photoreduction properties. Chem Eng J 209:386–393 Ye LQ, Liu JY, Jiang Z, Peng TY, Zan L (2013) Facets coupling of BiOBr-g-C3N4 composite photocatalyst for enhanced visiblelight-driven photocatalytic activity. Appl Catal B Environ 142–143:1–7 He YM, Cai J, Li TT, Wu Y, Lin HJ, Zhao LH, Luo MF (2013) Efficient degradation of RhB over GdVO4/g-C3N4 composites under visible-light irradiation. Chem Eng J 215–216:721–730