BiOBr/Bi₂MoO₆ composite in flower-like microspheres with enhanced photocatalytic activity under visible-light irradiation

Abstract

BiOBr/Bi₂MoO₆ composites in flower-like microspheres were synthesized by a facile solvothermal method. This photocatalyst could be activated by visible light owing to the narrow energy band gaps of both Bi₂MoO₆ and BiOBr. The BiOBr/Bi₂MoO₆ composites exhibited high activity in the photocatalytic degradation of rhodamine B and other organic pollutants under visible-light irradiation owing to enhanced light harvesting via multiple reflections inside the flower-like structure and due to the reduced photoelectron–hole recombination due to the formation of BiOBr–Bi₂MoO₆ heterojunctions. The dominant active species during the photocatalytic oxidation of RhB were determined as photogenerated holes and ·O₂⁻ was distinguished as the sub-main active site. The RhB mineralized into CO₂ through the de-ethylation route via forming five intermediates.

---

Introduction

Bismuth-containing semiconductors represent a new class of promising photocatalysts useful for environmental cleaning and H₂ production, owing to their narrow energy band gaps for absorbing visible light, and their hierarchical structure and morphology as well as excellent stability.^1,2 γ-Bi₂MoO₆ has received increasing interest due to its layered Aurivillius structure, in which perovskite slabs of corner-sharing distorted MoO₆ octahedra are sandwiched between (Bi₂O₂)₂⁺ layers,^3,4 leading to the fast separation of photo-induced electrons from holes and thus, a high quantum efficiency of photocatalysis.^5,6 To further improve its photocatalytic activity, both nanostructure engineering and composition doping have been explored.⁷ The combination of two semiconductors allows the transfer of photo-induced electrons from the valence band (VB) of one semiconductor to the conduction band (CB) of the other semiconductor, which could efficiently inhibit photoelectron–hole recombination. Up to now, Bi₂MoO₆-based composites with Bi₂O₃,⁸ ZnTiO₃,⁹ Bi₂WO₆,¹⁰ TiO₂,¹¹ Bi₂O₂CO₃,¹² graphene¹³ and C₆₀ (ref. 14) have been reported. However, their photocatalytic activities are still limited by the not well-matched band potentials to a certain extent.^5,6

BiOBr crystal with a tetragonal matlockite structure is composed of [Bi₂O₂]²⁺ slabs interleaved by double slabs of Br atoms. The layered BiOBr structure endows high carrier mobility and small probability of the recombination of photogenerated electrons and holes.^15,16 We found that well-matched band potentials in BiOBr/Bi₂MoO₆ heterojunctions could be obtained, since the wider E_g and low CB potential in BiOBr compared to those in Bi₂MoO₆ could facilitate the transfer of photogenerated electrons from Bi₂MoO₆ to the CB of BiOBr, leading to efficient charge separation and high photocatalytic activity.

Herein, we fabricated BiOBr/Bi₂MoO₆ flower-like microspheres with a uniform BiOBr/Bi₂MoO₆ combination by a facile solvothermal route. The microspheres exhibited high activity in the photocatalytic degradation of organic pollutants under visible-light irradiation owing to the narrow energy band gap and unique morphology and due to the BiOBr–Bi₂MoO₆ heterojunctions. Meanwhile, the BiOBr/Bi₂MoO₆ combination displayed an unusual photocatalysis process.

Results and discussion

Structure characteristics

EDX analysis (Fig. S1†) demonstrated that only Bi, O, Br and Mo existed in a 0.20-BiOBr/Bi₂MoO₆ sample. As shown in Table 1, the ICP analysis revealed that the BiOBr content increased gradually in X-BiOBr/Bi₂MoO₆ with the increase of X. From Fig. 1, the XRD patterns of X-BiOBr/Bi₂MoO₆ samples (X < 1.5) only show the diffraction peaks indicative of rhombohedral Bi₂MoO₆ (JSPDS 21-0102), possibly due to the low content and high dispersion of BiOBr. With the increase of BiOBr content, the diffraction peaks at (011), (012) and (212) characteristic of BiOBr (JCPDS 73-2061) appear in the X-BiOBr/Bi₂MoO₆ samples with X ≥ 1.5. No other diffraction peaks could be observed, implying the formation of a pure BiOBr/Bi₂MoO₆ composite.

Table 1 Structural parameters of different catalysts

Sample	ICP				XPS	SBET (m^2 g^−1)	RhB adsorption rate (%)
	Sample mass (g)	Bi concentration (mg L^−1)	Mo concentration (mg L^−1)	Bi : Br :  Mo molar ratio	Bi : Br :  Mo molar ratio
Bi2MoO6	—	—	—	—	—	19	8.7
0.10-BiOBr/Bi2MoO6	0.021	29	64	2.1 : 0.077 : 1.0	3.0 : 0.26 : 1.0	19	15
0.20-BiOBr/Bi2MoO6	0.026	35	76	2.1 : 0.14 : 1.0	3.5 : 0.45 : 1.0	19	17
0.33-BiOBr/Bi2MoO6	0.023	31	65	2.2 : 0.22 : 1.0	3.6 : 0.64 : 1.0	21	18
0.50-BiOBr/Bi2MoO6	0.024	33	64	2.4 : 0.35 : 1.0	4.1 : 0.75 : 1.0	29	21
BiOBr	—	—	—	—	—	35	23

---

Fig. 1 XRD patterns of X-BiOBr/Bi₂MoO₆, BiOBr and Bi₂MoO₆.

The FT-IR spectra in Fig. S2† revealed that all the samples displayed the vibration peaks of the surface-adsorbed water and hydroxyl groups at 3456 and 1625 cm⁻¹, respectively.^17,18 Corresponding to the above XRD results, 0.20-BiOBr/Bi₂MoO₆ displayed the same absorbance peaks as the pure Bi₂MoO₆ rather than the pure BiOBr due to the low content of BiOBr in the composite. The absorbance signals around 445 and 560 cm⁻¹ could be assigned to the vibrations from the bending Mo–O bond of MoO₆ and the bending Bi–O bond of (Bi₂O₂)²⁺ in Bi₂MoO₆,¹⁹ respectively. Meanwhile, the vibration peaks at 842, 797 and 733 cm⁻¹ could be attributed to the asymmetric stretching of MoO₆.¹⁹ In comparison, 1.5-BiOBr/Bi₂MoO₆ showed a well-resolved peak at 509 cm⁻¹, indicative of the vibration of Bi–O bond in BiOBr.²⁰

As shown in Fig. 2, the FESEM images demonstrated that 0.20-BiOBr/Bi₂MoO₆ was present in the uniform flower-like microspheres with the average diameter around 2.0 μm. The outer shell was covered by orderly assembled nanosheets with an average thickness of about 15 nm. The TEM image displayed a hollow chamber in the BiOBr/Bi₂MoO₆ microspheres, and the average shell thickness was estimated at 500 nm in Fig. 3. The HRTEM image clearly shows the lattice distances of 0.277 and 0.314 nm indicative of the (110) facet in the BiOBr crystal²¹ and the (131) facet in the Bi₂MoO₆ crystal,⁹ respectively, corresponding to the formation of BiOBr–Bi₂MoO₆ heterojunctions at the interface. Moreover, the chemical mapping (Fig. S3†) confirmed the uniform distribution of Bi, Mo, Br and O elements in the microspheres.

Fig. 2 FESEM (a, b), images of 0.20-BiOBr/Bi₂MoO₆.

Fig. 3 TEM (a) and HRTEM (b) images of 0.20-BiOBr/Bi₂MoO₆.

The XPS spectra in Fig. 4 confirmed that the pure Bi₂MoO₆ displayed a Bi binding energy around 158.7 and 164.0 eV in Bi 4f_5/2 and 4f_7/2 levels,^22,23 while the pure BiOBr showed higher Bi 4f binding energies of around 159.4 and 164.7 eV.^24,25 The Bi 4f binding energy of 0.20-BiOBr/Bi₂MoO₆ was between that of Bi₂MoO₆ and BiOBr, which implied an electronic interaction in the BiOBr/Bi₂MoO₆ composite. Additionally, 0.20-BiOBr/Bi₂MoO₆ displayed higher binding energies of both O 1s and Mo 3d compared to pure Bi₂MoO₆ as well as a lower binding energy of O 1s and Br 3d than that of pure BiOBr. This could be attributed to the electron transfer from Bi₂MoO₆ to BiOBr via the strong interaction in the BiOBr/Bi₂MoO₆ heterojunction, corresponding to the above HRTEM image. The higher molar ratio of both Bi/Mo and Br/Mo in the different X-BiOBr/Bi₂MoO₆ samples determined by XPS compared to those by ICP analysis (see Table 1) further suggested the surface enrichment of Bi and Br atoms, which was obviously due to the coverage of BiOBr on the surface of the Bi₂MoO₆ microspheres.

Fig. 4 XPS spectra of 0.20-BiOBr/Bi₂MoO₆, BiOBr and Bi₂MoO₆.

The TG analysis (Fig. S4†) revealed that pure Bi₂MoO₆ showed no significant weight loss, while the pure BiOBr displayed a significant weight loss around 510 °C due to the release of Br-contained gas via BiOBr oxidation.²⁵ The BiOBr/Bi₂MoO₆ showed a higher weight loss temperature than the pure BiOBr. This further confirmed the strong BiOBr–Bi₂MoO₆ interaction, in which the Bi₂MoO₆ could stabilize the BiOBr.

All the BiOBr, Bi₂MoO₆ and X-BiOBr/Bi₂MoO₆ samples displayed typical IV N₂ adsorption–desorption isotherms (Fig. S5†), indicating the presence of a mesoporous structure. As shown in Table 1, S_BET of the X-BiOBr/Bi₂MoO₆ samples gradually increased with the enhanced content of BiOBr in the BiOBr/Bi₂MoO₆ composite since the S_BET of pure BiOBr was higher than that of pure Bi₂MoO₆.

Optical properties

The UV-vis DRS spectra (Fig. S6†) revealed that pure Bi₂MoO₆ displayed stronger absorbance in the wider visible-light region compared to that of pure BiOBr due to its narrower energy band gap. The 0.20-BiOBr/Bi₂MoO₆ displayed slightly enhanced visible-light absorbance as the BiOBr nanosheets assembled on the outer shell of Bi₂MoO₆ microspheres could enhance light harvesting via the multi-reflections. Fig. S7a† shows that 0.20-BiOBr/Bi₂MoO₆ displays the highest photocurrent compared to pure BiOBr, pure Bi₂MoO₆ and a mechanical mixture of BiOBr and Bi₂MoO₆, which could be attributed to both the enhanced light harvesting and the diminished photoelectron–hole recombination rate. The PL spectra (Fig. S7b†) clearly confirmed a lower photoelectron–hole recombination rate in the X-BiOBr/Bi₂MoO₆ composites than in pure BiOBr, pure Bi₂MoO₆, and in a mechanical mixture of BiOBr and Bi₂MoO₆, corresponding to the weaker PL intensities.^21,26 These results demonstrate that the formation of BiOBr–Bi₂MoO₆ heterojunction plays a key role in inhibiting photoelectron–hole recombination via transferring photoelectrons from the VB of Bi₂MoO₆ to the CB of BiOBr.²⁷ The mechanical mixture of BiOBr and Bi₂MoO₆ exhibited a higher photoelectron–hole recombination rate than the BiOBr/Bi₂MoO₆ composites, obviously due to the poor interaction between BiOBr and Bi₂MoO₆, leading to fewer BiOBr–Bi₂MoO₆ heterojunctions. Besides, it was found that the photoelectron–hole recombination rate decreased with the increase of BiOBr from 0.10-BiOBr/Bi₂MoO₆ to 0.20-BiOBr/Bi₂MoO₆, corresponding to the increase in the number of BiOBr–Bi₂MoO₆ heterojunctions. However, a very high BiOBr content was harmful for inhibiting photoelectron–hole recombination (see 0.50-BiOBr/Bi₂MoO₆ and 0.33-BiOBr/Bi₂MoO₆), possibly due to the formation of BiOBr and Bi₂MoO₆ in separated phases.

Additionally, the flat band potentials (E_fb) of BiOBr and Bi₂MoO₆ were determined in EIS (see Fig. S8†). According to the intercept of the linear relationship between C_SC⁻² and E, the E_fb (C_SC⁻² = 0) of BiOBr and Bi₂MoO₆ were calculated. As shown in Table 2, the conduction band edge (E_CB) was 0.1 V more negative than the E_fb²⁸ in n-type BiOBr and Bi₂MoO₆ semiconductors owing to the positive line slope.²⁹ The edge position of the valence band could also be determined by considering the energy band gap calculated from ahv = A(hv − E_g)ⁿ, where a, v, E_g and A refer to the absorption coefficient, the light frequency, the band gap and a constant, respectively. The values of n for Bi₂MoO₆ and BiOBr were 1/2 and 2, respectively.^30,31 Therefore, the energy band structure in the BiOBr/Bi₂MoO₆ composite is proposed, see Scheme 1. Clearly, the band edges of BiOBr (E_VB = 2.57 eV, E_CB = −0.19 eV) and Bi₂MoO₆ (E_VB = 2.04 eV, E_CB = −0.55 eV) were well matched and thus, their heterojunctions were favourable for the rapid transfer of photoelectrons from the VB of Bi₂MoO₆ to the VB of BiOBr, which reduces the photoelectron–hole recombination rate.

Table 2 Band gap energy and position of BiOBr and Bi₂MoO₆

Sample	Efb (eV)	ECB (eV)	EVB (eV)	Eg (eV)
Bi2MoO6	−0.450	−0.550	2.04	2.59
BiOBr	−0.0900	−0.190	2.57	2.76

---

Scheme 1 Illustration of the energy band structure and photocatalytic reaction process of the BiOBr/Bi₂MoO₆ heterojunction.

Photocatalytic performance

As shown in Table 1, the BiOBr exhibited much higher adsorption capacity for RhB than the Bi₂MoO₆ owing to the larger S_BET. It was interesting to note that the BiOBr exhibited much higher adsorption capacity per unit S_BET than the Bi₂MoO₆, possibly due to the different adsorption models for RhB. Obviously, each BiOBr active site adsorbed more RhB molecules than each Bi₂MoO₆ active site, which could account for the higher adsorption capacity of the X-BiOBr/Bi₂MoO₆ than that of pure Bi₂MoO₆ with almost the same S_BET.

The visible-light-induced photocatalytic RhB degradation yields and TOC removal rates on different photocatalysts are summarized in Fig. 5. Although the BiOBr showed a much higher adsorption capacity than of the Bi₂MoO₆ (see Table 1), it still exhibited lower activity, which could be attributed to the broad energy band gap, corresponding to a poor utilization of visible light in comparison with the Bi₂MoO₆. The 0.20-BiOBr/Bi₂MoO₆ exhibited much higher activity than those of pure BiOBr, pure Bi₂MoO₆, the BiOBr + Bi₂MoO₆ mechanical mixture and 0.20-BiOBr/Bi₂MoO₆(G), in which 0.1 g of 0.20-BiOBr/Bi₂MoO₆ was ground for 120 min in the mortar, which damaged the flower-like structure of 0.20-BiOBr/Bi₂MoO₆ (see SEM images in Fig. S9†). Taking into account the similar adsorption capacities for RhB, the high activity of 0.20-BiOBr/Bi₂MoO₆ could be mainly attributed to both the enhanced light harvesting due to the multiple reflections in the unique flower-like structure and the diminished photoelectron–hole recombination rate owing to the formation of BiOBr–Bi₂MoO₆ heterojunctions resulting from their strong interaction.^32–34 The low activities of pure BiOBr and Bi₂MoO₆ were due to the easy recombination of photo-induced holes and electrons, and the slightly increased activity of the BiOBr + Bi₂MoO₆ sample may be due to the random and unstable contact between Bi₂MoO₆ and BiOBr in the physical mixture, which is favourable for the charge separation to some extent. However, it is impossible to form stable BiOBr–Bi₂MoO₆ heterojunctions to greatly improve the photocatalytic activity. The 0.20-BiOBr/Bi₂MoO₆(G) sample exhibited lower activity than 0.20-BiOBr/Bi₂MoO₆ due to the damage to the flower-like structure (see Fig. S9†). The TOC measurement showed the change in the total amount of organics, including the original RhB and the intermediates during the photocatalytic process. Both the original RhB and the intermediates were photodegraded simultaneously. Therefore, the decreased TOC value was obviously lower than that of the RhB degradation ratio together with a certain difference in activity (Fig. 5b), and RhB was photodegraded to CO₂. By plotting (C₀ − C)/C vs. reaction time, linear plots were obtained for all of the prepared photocatalysts (see Fig. S10†), which indicated that the RhB degradation reaction was a zero-order reaction with respect to RhB concentration. 0.22-BiOBr/Bi₂MoO₆ showed the very highest kinetic constants due to the highest photocatalytic efficiency. Similar results were also obtained during visible-light-induced photocatalytic degradation of methyl orange (MO) and phenol (Fig. S11†). It was also found that in the X-BiOBr/Bi₂MoO₆ series, their activities first increased with the increasing BiOBr content from pure Bi₂MoO₆ to 0.10-BiOBr/Bi₂MoO₆ to 0.20-BiOBr/Bi₂MoO₆ and then decreased with the further increase of BiOBr content from 0.20-BiOBr/Bi₂MoO₆ to 0.33-BiOBr/Bi₂MoO₆ to 0.50-BiOBr/Bi₂MoO₆. The 0.20-BiOBr/Bi₂MoO₆ sample displayed the maximum activity (Fig. S12†). Such results were exactly the same as the changing order of PL intensities (Fig. S7†), suggesting that the photoelectron–hole recombination rate plays a crucial role in determining photocatalytic activity. The insufficient BiOBr–Bi₂MoO₆ heterojunctions in the 0.10-BiOBr/Bi₂MoO₆ sample resulted in an inefficient inhibition of charge recombination, while more BiOBr catalyst in both the 0.33-BiOBr/Bi₂MoO₆ and 0.50-BiOBr/Bi₂MoO₆ samples induced easier charge recombination.

Fig. 5 RhB degradation (a) and TOC removal process (b) vs. reaction time during the visible-light-induced photocatalytic degradation of RhB on different photocatalysts. Reaction conditions: (a) 100 mL 20 mg L⁻¹ RhB aqueous solution, 0.10 g catalyst, 300 W Xe lamp (λ > 420 nm), T = 30 °C; (b) 100 mL 50 mg L⁻¹ RhB aqueous solution, 0.10 g catalyst, 300 W Xe lamp (λ > 420 nm), T = 30 °C.

In order to examine the formation of ˙OH radicals during the photodegradation reaction, a terephthalic acid photoluminescence (TA-PL) probing technique^35,36 with an excitation wavelength of 315 nm was employed. In Fig. S13,† PL signals were observed at 425 nm with the irradiation time (excitation at 315 nm). This demonstrated that ˙OH radicals were formed during the photodegradation process. However, as shown in Fig. S13,† 0.20-BiOBr/Bi₂MoO₆ exhibited similar TA-PL peak intensity at 425 nm compared to Bi₂MoO₆, but much lower than BiOBr, implying that ˙OH radicals were mainly generated on BiOBr in the BiOBr/Bi₂MoO₆ composite. It could be ascribed that only the photo-induced holes on the BiOBr photocatalyst rather than that on Bi₂MoO₆ were able to oxidize either H₂O or OH⁻ directly to ˙OH, since both redox potentials of ˙OH/H₂O (E^o = 2.38 V)³⁷ and ˙OH/OH⁻ (E^o = 2.27 V)^38,39 were more positive than the E_VB of Bi₂MoO₆, but were more negative than the E_VB of BiOBr. Therefore, it was difficult to oxidize H₂O or OH⁻ into ˙OH in the Bi₂MoO₆/BiOBr heterojunction due to the low content of BiOBr. Additionally, Fig. 6 reveals that the addition of isopropanol (IPA) allowed it to act as an ˙OH scavenger,^40,41 thus displaying no significant inhibiting effect on the photocatalytic RhB degradation. However, the addition of ammonium oxalate (AO) as h⁺ (photogenerated holes) scavengers⁴² could almost completely suppress the photocatalytic RhB degradation. The addition of benzoquinone (BQ) as an ˙O₂⁻ scavenger^40,43 and AgNO₃ as an e⁻ scavenger⁴⁴ also resulted in decreased activity, respectively. This demonstrated that h⁺ was the main active site and could oxidize the RhB directly rather than first following the process of forming ˙OH.

Fig. 6 Effects of various scavengers on the photocatalytic efficiencies of 0.20-BiOBr/Bi₂MoO₆ in RhB photocatalytic degradation. Reaction conditions are given in Fig. 5a.

According to the abovementioned results, the plausible mechanism of RhB photocatalytic degradation on the BiOBr/Bi₂MoO₆ composite photocatalyst is proposed in Scheme 1. Under visible-light irradiation, the electrons could be excited from the VB of Bi₂MoO₆ and BiOBr to their CB. The electrons in the CB of Bi₂MoO₆ were injected into the CB of BiOBr via their heterojunctions, followed by reacting with the adsorbed O₂ to produce ˙O₂⁻, owing to the higher redox potential of O₂/˙O₂⁻ (E^o = −0.046 eV)⁴⁵ than the E_CB of BiOBr. Finally, ˙O₂⁻ and the main active species of h⁺ led to the oxidation of RhB.

In the photodecomposition process in dye wastewater treatment, it is important to evaluate the mineralization ability of catalysts since the yielded intermediates may be more toxic than the dye molecule itself. Since the multiple yielded intermediates may also exhibit activity on the visible-light-driven RhB degradation process,³³ the photocatalytic RhB degradation yields and TOC removal rates under visible lights show certain difference of activity in Fig. 5a and b. Hence, the yielded intermediates of the visible-light-driven RhB degradation process on 0.20-BiOBr/Bi₂MoO₆ were further investigated by HPLC-MS (Fig. S14†). Moreover, the N,N,N′,N'-tetraethylrhodamine (RhB) molecule with an m/z of 443, five intermediates with m/z of 415, 387_a, 387_b, 359 and 331 were detected, corresponding to N,N,N′-triethylrhodamine (TER), N,N′-diethylrhodamine (DER_a), N,N-diethylrhodamine (DER_b), N-ethylrhodamine (MER) and rhodamine. The intermediates were not detected, as they were more toxic than the dye molecule itself. The RhB concentration decreased monologically with the increasing reaction time, while the concentrations of all the intermediates first increased and then decreased. According to the appearance order of multiple intermediates, we concluded that the RhB degradation followed the de-ethylation route and cleaved into TER, DER, MER and rhodamine in turn, and finally decomposed into CO₂.

Conclusions

This work developed a facile solvothermal approach to prepare a BiOBr/Bi₂MoO₆ composite photocatalyst in flower-like microspheres. It exhibited high activity in the photocatalytic degradation of organic pollutants under visible-light irradiation, mainly owing to enhanced light harvesting via multi-reflections and reduced photoelectron–hole recombination through the heterojunction. Kinetic studies demonstrated that the RhB degradation to CO₂ were mainly driven by h⁺ active sites and followed a de-ethylation route with the formation of five intermediates.

Experimental section

BiOBr/Bi₂MoO₆ preparation

3.4 g Bi(NO₃)₃·5H₂O and 0.84 g Na₂MoO₄·2H₂O were dissolved in 10 mL ethylene glycol (EG) and 10 mL glycerol under stirring for 120 min, respectively. After mixing the two solutions, 40 mL ethanol was slowly added, followed by stirring for 10 min. The clear solution was then transferred into a 100 mL Teflon-lined autoclave and heated at 160 °C for 24 h. The brown suspension with Bi₂MoO₆ precursor was obtained after the autoclave was cooled to room temperature naturally. Then, the desired amounts of Bi(NO₃)₃·5H₂O and cetyltrimethyl ammonium bromide (CTAB) with a 1.5 molar ratio to Bi(NO₃)₃·5H₂O was added with stirring for 120 min. The solution was transferred into a 100 mL Teflon-lined autoclave and heated at 160 °C for 10 h again. Finally, the obtained solid was filtered, washed with ethanol, dried at 80 °C in air and annealed at 400 °C for 4 h. The as-prepared sample was denoted as X-BiOBr/Bi₂MoO₆, where X refers to the Br/Mo molar ratio. For comparison, the pure Bi₂MoO₆ and BiOBr samples were prepared by the same procedures, respectively. The mechanical mixture of BiOBr and Bi₂MoO₆ was referred to BiOBr + Bi₂MoO₆, in which the Br/Mo molar ratio was the same as that in 0.20-BiOBr/Bi₂MoO₆. The 0.20-BiOBr/Bi₂MoO₆(G) sample was 0.20-BiOBr/Bi₂MoO₆ after being crushed.

Catalyst characterization

The structure and composition were determined by energy dispersive X-ray spectroscopy (EDX, HITACHI S-4800) and X-ray diffraction (XRD, Rigacu D/Max-2000). Surface morphologies were observed by scanning electron microscopy (FESEM, HITACHI S-4800) and by using a transmission electronic micrograph (TEM, JEM-2010). The thermal stability was investigated by thermogravimetric analysis (TGA) on a DTG-60H. The FTIR spectra were recorded by a Fourier transform infrared spectroscopy (NEXUS 470). X-ray photoelectron spectroscopy (XPS, Versa Probe PHI 5000) was employed to determine the surface electronic states. The shift of the binding energy due to the relative surface charging was corrected using the C_1S level at 284.6 eV as an internal standard. N₂ adsorption/desorption isotherms were measured on a TriStar II 3020. The Brunauer–Emmett–Teller (BET) method was used to calculate the specific surface area (S_BET) based on the adsorption branches. The optical properties were determined by UV-vis diffuse reflectance spectrum (UV-vis DRS, Varian Cary 500) and photoluminescence spectrum (PL, Varian Cary-Eclipse 500). The excitation wavelength was 325 nm provided by a UV xenon lamp with an incidence angle of 45°, and the spectra were recorded in the range from 400 to 600 nm at room temperature. To avoid reflecting the excitation light into the detector, all the samples were maintained at 60° to the detection plane, which was defined by the Xe lamp, the samples and the detector. Meanwhile, ˙OH-trapping photoluminescence spectra of terephthalic acid solution (PL-TA, λ_excition = 315 nm, λ_emission = 425 nm) were recorded on a photoluminescence spectrum (PL, Varian Cary-Eclipse 500). The metal ion concentrations were determined by using inductively coupled plasmatomic emission spectroscopy (ICP, VISTA-MPX). Photocurrent-time measurements and electrochemical impedance spectroscopy (EIS) were performed using an electrochemistry workstation (CHI660E). Photocurrent-time measurements were carried out on the film sample in 0.10 M Na₂SO₄ aqueous solution with an exposed surface area of 2.0 cm², which was irradiated by a 300 W Xe lamp (λ > 420 nm). EIS Experiments were carried out on the film sample in 0.10 M Na₂SO₄ solution at 1000 Hz within the potential region of −1.0 to 1.0 V vs. Ag/AgCl. A Mott–Schottky (M–S) plot showed the linear variation of C_SC²⁻ with respect to the applied potential (eqn (1)), where N_D, ε, E_fb, E and C_SC refer to the doping density, dielectric constant, flat band potential, external biasing voltage and capacitance, respectively.

C_SC⁻² = (1.41 × 1020/N_Dε)[E − E_fb − 0.026] (1)

Activity test

The liquid-phase photocatalytic degradation of dyes was carried out at 30 °C in a self-designed 250 mL reactor containing 0.10 g catalyst and 100 mL 20 mg L⁻¹ dye organics. A 300 W Xe lamp was used as a visible-light source, and a 420 nm filter was placed above the reactor to cut off the UV light. At given time intervals, 3.0 mL aliquots were collected from the suspension and immediately centrifuged. The concentration of dyes was determined using a UV-vis spectrophotometer (UV-7504 PC) at the characteristic peak. A liquid chromatography-mass spectrometry system (HPLC-MS, Agilent 1200) was used to detect the reaction products and intermediates. The measurement of total organic carbon (TOC) was carried out on a Vario TOC analyzer. The reproducibility of the results was checked by repeating the experiments at least three times and was found to be within acceptable limits (±6%).

Acknowledgements

This work is supported by National Natural Science Foundation of China (21237003 and 21577092), Program for Changjiang Scholars and Innovative Research Team in University (IRT1269), Shanghai Government (15520711300) International Joint Laboratory on Resource Chemistry (IJLRC), Science and technology project of Yunnan Province (C0120150543) and Research Fund of Yunnan Provincial Education Department (2015Z183).

Notes and references

1. W. Z. Yin, W. Z. Wang and S. M. Sun, Catal. Commun., 2010, 11, 647–650.
2. A. Martinez-de Cruz and S. Obregon Alfaro, J. Mol. Catal. A: Chem., 2010, 320, 85–91.
3. E. L. Cuéllar, A. M. Cruz, K. H. L. Rodríguez and U. O. Méndez, Catal. Today, 2011, 166, 140–145.
4. L. W. Zhang, T. G. Xu, X. Zhao and Y. F. Zhu, Appl. Catal., B, 2010, 98, 138–146.
5. X. Lin, X. Y. Guo, W. L. Shi, H. J. Zhai, Y. S. Yan and Q. W. Wang, J. Solid State Chem., 2015, 229, 68–77.
6. J. Di, J. X. Xia, M. X. Ji, H. P. Li, H. Xu, H. M. Li and R. Chen, Nanoscale, 2015, 7, 11433–11443.
7. J. Ren, W. Z. Wang, M. Shang, S. M. Sun and E. P. Gao, ACS Appl. Mater. Interfaces, 2011, 3, 2529–2533.
8. Y. S. Xu, Z. J. Zhang and W. D. Zhang, Mater. Res. Bull., 2013, 48, 1420–1427.
9. P. Zhang, C. L. Shao, M. Y. Zhang, Z. C. Guo, J. B. Mu, Z. Y. Zhang, X. Zhang and Y. C. Liu, J. Hazard. Mater., 2012, 217–218, 422–428.
10. F. J. Zhang, S. F. Zhu, F. Z. Xie, J. Zhang and Z. D. Meng, Sep. Purif. Technol., 2013, 113, 1–8.
11. N. Li, L. Zhu, W. D. Zhang, Y. X. Yu, W. H. Zhang and M. F. Hou, J. Alloys Compd., 2011, 509, 9770–9775.
12. Y. S. Xu and W. D. Zhang, Appl. Catal., B, 2013, 140–141, 306–316.
13. F. Zhou, R. Shi and Y. F. Zhu, J. Mol. Catal. A: Chem., 2011, 340, 77–82.
14. X. Zhao, H. J. Liu, Y. L. Shen and J. H. Qu, Appl. Catal., B, 2011, 1–2, 63–68.
15. H. Deng, J. W. Wang, Q. Peng, X. Wang and Y. D. Li, Chem.–Eur. J., 2005, 11, 6519–6524.
16. W. L. Huang and Q. S. Zhu, Comput. Mater. Sci., 2008, 43, 1101–1108.
17. J. C. Yu, L. Z. Zhang, Z. Zheng and J. Zhao, Chem. Mater., 2003, 15, 2280–2286.
18. T. Y. Peng, D. Zhao, H. B. Song and C. H. Yan, J. Mol. Catal. A: Chem., 2005, 238, 119–126.
19. M. Maczka, L. Macalik, K. Hermanowicz, L. Kepinski and J. Hanuza, J. Raman Spectrosc., 2010, 41, 1289–1296.
20. Z. S. Liu, Y. H. Bi, Y. L. Zhao, X. Huang and Y. B. Zhu, Mater. Res. Bull., 2014, 49, 167–171.
21. D. Q. Zhang, M. C. Wen, B. Jiang, G. S. Li and J. C. Yu, J. Hazard. Mater., 2012, 211–212, 104–111.
22. X. Zhao, J. H. Qu, H. J. Liu and C. Hu, Environ. Sci. Technol., 2007, 41, 6802–6807.
23. G. P. Dai, J. G. Yu and G. Liu, J. Phys. Chem. C, 2011, 115, 7339–7346.
24. X. M. Tu, S. L. Luo, G. X. Chen and J. H. Li, Chem.–Eur. J., 2012, 18, 14359–14366.
25. Y. C. Feng, L. Li, J. W. Li, J. F. Wang and L. Liu, J. Hazard. Mater., 2011, 92, 538–544.
26. J. C. Yu, J. G. Yu, W. K. Ho, Z. T. Jiang and L. Z. Zhang, Chem. Mater., 2002, 14, 3808–3816.
27. J. Zhang, J. X. Xia, S. Yin, H. M. Li, H. Xu, M. Q. He, L. Y. Huang and Q. Zhang, Colloids Surf., A, 2013, 420, 89–95.
28. S. R. Morrison, Electrochemistry at Semiconductor and Oxidized Metal Electrodes, Plenum Press, New York, 1980, ch. 4.
29. J. M. Herrmann, M. E. Jamal and M. Forissier, React. Kinet. Catal. Lett., 1988, 37, 255–260.
30. M. Y. Zhang, C. L. Shao, P. Zhang, C. Y. Su, X. Zhang and P. P. Liang, J. Hazard. Mater., 2012, 225–226, 155–163.
31. X. J. Shi, X. Chen, X. L. Chen, S. M. Zhou, S. Y. Lou, Y. Q. Wang and L. Yuan, Chem. Eng. J., 2013, 222, 120–127.
32. J. Cao, B. Y. Xu, H. L. Lin, B. D. Luo and S. F. Chen, Chem. Eng. J., 2012, 185–186, 91–99.
33. Y. C. Miao, G. F. Pan, Y. N. Huo and H. X. Li, Dyes Pigm., 2013, 99, 382–389.
34. H. X. Li, Z. F. Bian, J. Zhu, D. Q. Zhang, G. S. Li, Y. N. Huo, H. Li and Y. F. Lu, J. Am. Chem. Soc., 2007, 129, 8406–8407.
35. K. Ishibashi, A. Fujishima, T. Watanabe and K. Hashimoto, Electrochem. Commun., 2000, 2, 207–210.
36. Q. Xiao, Z. C. Si, J. Zhang, C. Xiao and X. K. Tan, J. Hazard. Mater., 2008, 150, 62–67.
37. A. J. Bard and L. R. Faulkner, Electrochemical Methods: Fundamentals and Applications, John Wiley & Sons, New York, 1980.
38. P. L. Allen and A. Hickling, Trans. Faraday Soc., 1967, 53, 1626–1635.
39. X. Li, J. Zhu and H. X. Li, Appl. Catal., B, 2012, 123–124, 174–181.
40. G. T. Li, K. H. Wong, X. W. Zhang, C. Hu, J. C. Yu, R. C. Y. Chan and P. K. Wong, Chemosphere, 2009, 76, 1185–1191.
41. L. S. Zhang, K. H. Wong, H. Y. Yip, C. Hu, J. C. Yu, C. Y. Chan and P. K. Wong, Environ. Sci. Technol., 2010, 44, 1392–1398.
42. N. Zhang, S. Q. Liu, X. Z. Fu and Y. J. Xu, J. Phys. Chem. C, 2011, 115, 9136–9145.
43. M. C. Yin, Z. S. Li, J. H. Kou and Z. G. Zou, Environ. Sci. Technol., 2009, 43, 8361–8366.
44. Y. N. Huo, J. Zhang, M. Miao and Y. Jin, Appl. Catal., B, 2012, 111–112, 334–341.
45. L. Kong, Z. Jiang, H. H. Lai, R. J. Nicholls, T. C. Xiao, M. O. Jones and P. P. Edwards, J. Catal., 2012, 293, 116–125.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra18987j

---