Photocatalytic degradation and pathway of oxytetracycline in aqueous solution by Fe₂O₃–TiO₂ nanopowder

Abstract

The degradation behavior of oxytetracycline (OTC) in aqueous solution using Fe₂O₃–TiO₂ nanopowders (Fe₂O₃–TiO₂ NPs) as photocatalysts under UV/visible light was investigated in this paper. Fe₂O₃–TiO₂ NPs photocatalysts synthesized using co-precipitation were characterized by XRD, TEM, FT-IR and UV-vis DRS. Fe₂O₃–TiO₂ NPs with 45 wt% Fe₂O₃ exhibited higher photocatalytic activity than others. The photocatalytic degradation of OTC shows a maximum efficiency at pH = 5.5 under both UV/visible light due to the surface adsorption. The intermediate products during the photocatalytic degradation of OTC were detected by LC-MS/TOF. Seven main intermediates were formed, and their evolution was discussed. On the basis of the evidence oxidative intermediate formation and a detailed degradation path way of OTC by 45 Fe₂O₃–TiO₂ NPs are proposed.

---

1. Introduction

Antibiotic tetracyclines (TCs) are widely used in disease treatment for humans as well as disease prevention and as growth promoters in livestock. Among this family, oxytetracycline (OTC) is one of the most frequently detected tetracyclines in water bodies and sediments in many countries. Moreover, due to the antibiotic nature, hydrophilic property and stable naphthacene ring structure (seen in Fig. 1), OTC is hardly removable by conventional water treatment processes, including biological and chlorination processes. Therefore, the spread of OTC in surface water may directly threaten the safety of drinking water for human consumption.

Fig. 1 Structure of OTC.

Recently, photocatalytic degradation is considered as an affordable, effective, environment friendly technology to remove the antibiotic in aquatic system. Zhao et al.^1,2 has also studied the photocatalytic degradation of OTC using TiO₂-supported zeolite. As it is well know that TiO₂ alone can only response to ultraviolet light, thus limited its application. In order to improve the efficiency of solar energy utilization, visible light responsive (VLR) TiO₂ were developed. Among these materials, Fe₂O₃–TiO₂ composites exhibited excellent photocatalytic activity under visible light and were applied as photocatalyst in remove of various organic pollutants, such as auramine, acridine dye and aryl halides in solution.^3–5

It is clear that the destruction of OTC is complicated; a series of intermediate products are formed during the photocatalytic process, and some may be more toxic than OTC. Thus, it is necessary to identify the intermediate products during the photocatalytic degradation, and further develop an understanding of the degradation mechanism. Liu et al.⁶ suggested a possible degradation process of tetracycline (TC) using Cl–TiO₂ as photocatalysts and considered that the resulting ˙OH radical can oxidize most of TC antibiotics to mineral end-products. However, the photocatalytic degradation pathway and mechanism of OTC has been few reported.

The present study attempts to assess the heterogeneous photocatalytic degradation of OTC by Fe₂O₃–TiO₂ NPs under UV and visible light, and further investigated the degradation behavior of OTC. Fe₂O₃–TiO₂ NPs were synthesized by co-precipitation, and characterized by XRD, TEM, FT-IR and UV-vis DRS. The effects of Fe₂O₃ content, initial pH of solution on degradation efficiency were investigated under UV/visible light, and compared to pure TiO₂. The main intermediate products formed during photocatalytic degradation process were analyzed by LC-MS/TOF. And based on these results, the detailed degradation pathway of OTC using 45 Fe₂O₃–TiO₂ NPs is proposed.

2. Experimental section

2.1 Preparation of Fe₂O₃–TiO₂ NPs

Fe₂O₃–TiO₂ nanopowders were synthesized using co-precipitation method.⁷ Certain amounts of predetermined Fe(NO₃)₃·9H₂O and Ti (SO₄)₂ were dissolved in 10 mL dilute sulfuric acid, respectively, then both were poured into 400 mL ethanol to form a mixed solution. Aqueous ammonia (NH₃:25–28 wt%) was added drop by drop in this mixed solution with continuous stirring. The pH of solution was controlled at 9 during the dropping process. The precipitate was firstly aged for 2 h, then filtered and washed till that no SO₄²⁻ was detected with Ba²⁺. The precipitate was dried at 100 °C for 24 h, and calcined at 550 °C for 6 h to obtain Fe₂O₃–TiO₂ NPs.

2.2 Photocatalytic degredation of OTC

Photocatalytic degradation of oxytetracycline (OTC, 95% purity, Aladdin Chemistry Co., Ltd, Shanghai, China) was carried out quartz glass tubes (Φ = 30 mm, H = 260 mm), which was opened to the ambient air in order to reach the air-equilibrated condition. The reaction of unit used in this work was arranged as reported in literature.³ A 300 W high-pressure mercury lamp was used as a UV light source, with main wavelength of 250 nm–400 nm. A 300 W iodine tungsten lamp was used as visible light source. The top of the quartz glass tube was 10 cm away from the UV and visible source.

Fe₂O₃–TiO₂ NPs photocatalysts at the dose of 1 g L⁻¹ was dispersed into OTC aqueous solution (60 mg L⁻¹) and shaked vigorous by using magnetic stirrer. The pH value of the solution was adjusted with hydrochloric acid or sodium hydroxide buffer solution to the desired value (pH = 2.0, 5.5, 8.5 and 11.0). A pH meter (PHS-25, Shanghai, China) was used to measure the pH value. Before switching on irradiation, OTC solution was continuously stirred (stirring speed: 500 rpm min⁻¹) in the dark for 1 h to ensure adsorption–desorption equilibrium. Then the system was exposed under UV or visible light for different time. Each photocatalytic degradation experiment was conducted in triplicate. The reusability experimental of Fe₂O₃–TiO₂ NPs was similar to A. K. Patra et al.⁸ reported.

After irradiation, 5 mL of the suspension was extracted and centrifuged, and the residual concentration of OTC was measured using a UV-vis spectroscopy (Techcomp Ltd. U3010, Japan) at its maximum absorption of 350 nm. The photocatalytic degradation efficiency (DE) was calculated by the following formula (eqn (1)):

DE = (A₀ − A_t)/A₀ × 100% (1)

where A₀ is the initial absorbance of OTC solution, A_t is the absorbance of OTC solution at the irradiation time (t).

2.3 Characterization

The phases of obtained Fe₂O₃–TiO₂ NPs and commercial Fe₂O₃, TiO₂ reagents were analyzed by X-ray diffraction (Rigaku D/Max-2200, Japan). The average crystalline size of Fe₂O₃–TiO₂ NPs nanopowder was calculated by Scherrer equation (eqn (2)):where D is the average crystalline size, K refers to a constant of 0.89, β is the integration of half-height width, θ is diffraction angle, and λ is X-ray wavelengths (equals to 0.15418 nm).

The micrography of 45 Fe₂O₃–TiO₂ NPs was observed via a transmission electron microscopy (Jeol 200CX, Japan). Infra-red (IR) spectra of 45 Fe₂O₃–TiO₂ NPs was recorded with a Nicolet Avatar 370 FT-IR (Thermo Nicolet, Waltham, USA). The UV-vis absorption spectra of photocatalysts were recorded by a UV-vis spectrophotometer (Techcomp Ltd. U3010, Japan). Zeta potential of Fe₂O₃–TiO₂ NPs was measured with a nanoparticles and potential analyzer (Zetasizer 3000HS, United Kingdom).

LC-MS/TOF systems consisted of an Agilent 1260 infinity HPLC coupled with PDA detector and an Agilent 6230 Accurate Mass MS/TOF mass spectrometer (Agilent technologies Singapore(sale)Pte, America) were used for the structure information of unknown intermediary products formed during the photocatalytic degradation process. Each mass spectrometer was equipped with electrospray ionization (ESI†) interface and operated in positive ionization mode. The mobile phase was acetonitrile and a representative separation of OTC was achieved within 13 min.

The MS full scan spectra with accurate mass determination were obtained by Agilent 6230 TOF mass spectrometer. Ionization condition conditions: drying gas (N₂) flow rate, 10.0 L min⁻¹; dry temperature, 330 °C; nebulizer, 30 psi; fragmentor, 130 V; capillary voltage, 4000 V; skimmer potential, 50 V; hexapole RF, 250 V. The TOF spectra were set to scan a mass range of 100–3200 m/z. All the operations, acquisition and analysis were controlled by Mass Hunter Workstation.

3. Results and discussion

3.1 Characterization of Fe₂O₃–TiO₂ NPs

Fig. 2(a) presents XRD pattern of anatase TiO₂ (anatase TiO₂ content is 99.8%, particle size of 60 nm. Aladdin Chemistry Co., Ltd, Shanghai, China), hematite Fe₂O₃ (hematite Fe₂O₃ content is 99.5%, particle size of 30 nm. Aladdin Chemistry Co., Ltd, Shanghai, China) and 45 Fe₂O₃–TiO₂ NPs. The main phases in 45 Fe₂O₃–TiO₂ NPs are anatase TiO₂ and hematite Fe₂O_3. The crystalline size of Fe₂O₃ and TiO₂ are 18 nm and 14 nm calculated from Scherrer equation, respectively. In addition, the micrography of 45 Fe₂O₃–TiO₂ NPs obtained is shown as Fig. 2(b). The average particle size is around 20 nm, and well dispersed.

Fig. 2 (a) XRD patterns of anatase TiO₂, hematite Fe₂O₃ and 45 Fe₂O₃–TiO₂ NPs, (b) TEM photo of 45 Fe₂O₃–TiO₂ NPs.

FT-IR spectra of 45 Fe₂O₃–TiO₂ NPs is shown in Fig. 3(a). Bands at 3420.78 cm⁻¹ and 1650.15 cm⁻¹ are ascribed to hydroxyl.⁹ While the strong absorption at 471.05 cm⁻¹ is attributed to the stretching vibrations of the Ti–O in anatase TiO₂.¹⁰

Fig. 3 (a) FTIR spectra of 45 Fe₂O₃–TiO₂ NPs, (b) the UV-vis absorption spectra of 45 Fe₂O₃–TiO₂ NPs, anatase TiO₂ and hematite Fe₂O₃.

Fig. 3(b) shows the UV-vis absorption spectra of 45 Fe₂O₃–TiO₂ NPs, pure anatase TiO₂ and hematite Fe₂O₃. Anatase TiO₂ only absorbed ultraviolet radiation of less than 400 nm, while hematite Fe₂O₃ exhibits a high energy absorbance in the visible light region of 400–600 nm. The 45 Fe₂O₃–TiO₂ NPs shows not only stronger absorption in the ultraviolet region of less than 400 nm but also adequate and strong absorption in the visible light region of 400–600 nm. It indicates that 45 Fe₂O₃–TiO₂ NPs can also be used under visible light.

3.2 Effect of Fe₂O₃ content on photocatalytic activity

The representative UV-visible (vis) spectra changes of the OTC solution as a function of reaction time under UV condition were shown in Fig. 4(a). The result shows that the adsorption peaks significant reduced with the increase of the irradiation time at 350 nm in 5 h.

Fig. 4 (a) UV-vis spectral changes of OTC solution with reaction time; (b) DE of OTC using Fe₂O₃–TiO₂ NPs, TiO₂ and Fe₂O₃ under UV light (OTC: 60 mg L⁻¹, pH = 2. Holding time: 5 h, the photocatalyst concentration was 1 g L⁻¹).

Fig. 4(b) illustrated the dependence of different Fe₂O₃ content on photocatalytic degradation efficience (DE) of OTC under UV light. All Fe₂O₃–TiO₂ NPs exhibited lower photocatalytic activity than pure TiO₂, but much higher than pure Fe₂O₃. The photocatalytic activity of Fe₂O₃–TiO₂ NPs was found firstly increased with the increase of Fe₂O₃ content from 40 to 45 wt%. The 45 Fe₂O₃–TiO₂ NPs exhibited the highest photocatalytic DE of OTC. With the further increase of Fe₂O₃ content to 55%, the DE of OTC decreased slightly. Liu et al.³ considered the recombination of electron–hole pairs was reduced due to combining Fe₂O₃ with TiO₂, and resulted in an increase of photocatalytic activity. However, when the content of Fe₂O₃ is higher than 45%, the surface of TiO₂ could have been covered with part of Fe₂O₃ of high concentration, and led to a decrease of light absorption activated site at the surface of TiO₂.

3.3 Effect of pH on OTC degradation

It has been reported that OTC has four species at different pH ranges,¹¹ and each species has unique electric charge state. At pH = 2.0, OTC is protonated as H₃OTC⁺. When pH is increased to 5.5, the majority of OTC changed to H₂OTC^±. Besides, OTC at pH 8.5 is primarily represented for the fraction of HOTC⁻ while OTC mainly exists as OTC²⁻ at pH = 11.0. Thus the initial pH of solution may affects photocatalytic degradation of OTC under both UV and visible. The effect of pH on photocatalytic degradation of OTC using 45 Fe₂O₃–TiO₂ NPs under both UV and visible light are presented as a function of irradiation time in Fig. 5(a) and (b), respectively. The results showed that DE of OTC was firstly improved with increasing pH, then significantly lowered at pH = 8.5 and 11.0 under both UV and visible light.

Fig. 5 Effect of pH on DE of OTC (60 mg L⁻¹) under (a) UV and (b) visible light by 45 Fe₂O₃–TiO₂ NPs ([45 Fe₂O₃–TiO₂ NPs] = 1 g L⁻¹).

Fig. 6(a) illustrates the effects of pH on the adsorption for OTC by 45 Fe₂O₃–TiO₂ NPs. The experiment was conducted in darkness, the initial concentration of OTC was 60 mg L⁻¹ and the reaction time was 5 h. As seen in Fig. 6(b), the percent adsorption of OTC by 45 Fe₂O₃–TiO₂ NPs increased from 25% to 33% with increase in pH from 2 to 5.5, and decrease from 33% to 4% with increasing pH. This tendency was in good agreement with photocatalytic degradation behavior, indicating that the photocatalytic process was limited by the adsorption of OTC at the surface of catalyst in this experiment.

Fig. 6 Effect of pH on (a) adsorption of OTC (60 mg L⁻¹, 5 h) by 45 Fe₂O₃–TiO₂ NPs, (b) zeta potential of 45 Fe₂O₃–TiO₂ NPs ([45 Fe₂O₃–TiO₂ NPs] = 1 g L⁻¹) under dark.

Fig. 6(b) gives the pH-dependent zeta potential of 45 Fe₂O₃–TiO₂ NPs, and the zero point is around 8. When pH is higher than zero, 45 Fe₂O₃–TiO₂ NPs is negatively charged. As mention above, OTC mainly exists as HOTC⁻ at pH = 8.5, OTC²⁻ at pH = 11, the electric repulsion makes the adsorption difficult, and leads to the decrease in OTC degradation. When pH is less than zero point, the surface of 45 Fe₂O₃–TiO₂ NPs, presents positive charge. At pH = 5.5, OTC acts as neutral-zwitterionic (H₂OTC^±) may result in relatively stronger adsorption of OTC on the surface of 45 Fe₂O₃–TiO₂ NPs, and sharply increase the OTC degradation. At pH = 2, the H₃OTC⁺ is difficult to be absorbed on the positive surface. However, more OH⁻ absorbed on the surface of photocatalysts, generated more ˙OH and enhanced the photo-catalytic degradation.

The photocatalytic degradation efficiency of OTC using pure TiO₂ and Fe₂O₃ at pH = 5.5 are also given in Fig. 5. Under UV light irradiation, photocatalytic activity of pure TiO₂ was superior to the 45 Fe₂O₃–TiO₂ NPs, but under visible light, the photocatalytic activity of 45 Fe₂O₃–TiO₂ NPs was significantly higher than pure TiO₂. Since visible light takes 50% of solar light, 45 Fe₂O₃–TiO₂ NPs are good candidates for further application in photocatalytic degradation OTC.

3.4 Reusability of 45 Fe₂O₃–TiO₂ NPs

In order to evaluate the potential applications, the reusability 45 Fe₂O₃–TiO₂ NPs has been studied under same conditions for 3 cycles, and the results are shown in Fig. 7. Under both UV and visible light, the photocatalytic degradation of OTC decreased slightly during 3 cycles, showing that it could be used as an effective material for many cycles.

Fig. 7 Photocatalytic activities 45 Fe₂O₃–TiO₂ NPs ([45 Fe₂O₃–TiO₂ NPs] = 1 g L⁻¹) for the degradation of OTC (60 mg L⁻¹) at pH = 5.5 under UV-vis light for 3 cycles.

3.5 Analysis of degradation intermediates

The total ion chromatogram (TIC) of mass spectra for the samples of OTC solution before and after photocatalytic degradation is given in Fig. 8. On the basis of this experiment result, the intermediate products during the photocatalysis process are identified by MS/TOF, and iron spectra at different retention time (RT) of OTC photocatalysis sample before and after 5 h are presented in ESI Fig. S1.†

Fig. 8 The total ion chromatogram (TIC) of mass spectra for OTC before and after photo-catalytic degradation ([45 Fe₂O₃–TiO₂ NPs] = 1 g L⁻¹, OTC: 60 mg L⁻¹, pH = 5.5, 5 h UV-vis irradiation).

Five irons spectra at RT = 0.345, 0.354, 0.38, 4.882 and 6.656 min were detected in original OTC solution. The average mass spectra corresponding to these five irons are shown in Fig. 9, five main product ions at m/z 477.1501 ([M + H]⁺), m/z 461.1549 ([M + H]⁺), m/z 447.1386 ([M + H]⁺), m/z 446.1445 ([M + H]⁺), and m/z 465.1294 ([M + Na]⁺) might be identified as C₂₂H₂₄N₂O₁₀ (P₆), C₂₂H₂₄N₂O₉ (OTC), C₂₁H₂₂N₂O₉ (P₁), C₂₂H₂₃NO₉ (P₂) and C₂₂H₂₂N₂O₈ (P₃).

Fig. 9 Product ions spectra of OTC photocatalysis sample before and after photocatalytic degradation ([45 Fe₂O₃–TiO₂ NPs] = 1 g L⁻¹, OTC: 60 mg L⁻¹, pH = 5.5, 5 h UV-vis irradiation). (Superscript B: product ions detected in original OTC solution; superscript A: product ions detected in OTC solution after photocatalytic degradation).

In the OTC solution after photocatalytic degradation under UV-vis irradiation, P₁, P₆ and OTC are not detected, while P₂ at m/z 446.1443 and P₃ at m/z 465.1294 are still remained. Besides, three evident product ions at m/z 415.149 ([M + H]⁺), m/z 402.1188 ([M + H]⁺) and m/z 179.0349 ([M + H]⁺) are observed in the average mass spectra at 0.354 min, 5.706 min and 10.423 min, which are different from other main product ions in this experiment. Based on the molecular weight of the compound we therefore suggested that these peaks represent C₂₁H₂₂N₂O₇ (P₄), C₂₀H₁₉NO₈ (P₅) and C₉H₆O₄ (P₇). It was suggested that these three products were formed during the photocatalytic degradation process.

3.6 Mechanism of photodegradation

The mechanism of OTC degradation can be described as eqn (3)–(7). Firstly, the OTC molecular was adsorbed onto the surface of Fe₂O₃–TiO₂ nanopowders. It was well known that conduction band electron (e⁻) and valence band holes (h⁺) are generated when aqueous TiO₂ suspension is irradiated with UV light energy.¹² However, due to the large distance between VB and CB (3.31 eV), absorption of visible-light is negligible for pure TiO₂. In Fe₂O₃–TiO₂, as the band edge of Fe³⁺ is lower than TiO₂, which means that Fe³⁺ is situated in between conduction band (CB) and valence band (VB). When visible-light was illuminate on Fe₂O₃–TiO₂, the electrons excited from the valence band to forbidden band leaving holes in the valence band TiO₂. The photogenerated e⁻–h⁺ pairs can recombine in the bulk catalyst or diffuse to the catalyst particle surface and react with H₂O and O₂ molecules adsorbed there leading to the formation of reactive oxygen species (ROS) under visible light.¹³ As a very strong oxidizing agent, the ˙OH radical and ˙O₂⁻ can oxidize most of TC antibiotics to the mineral end-products.^14–16

Fe₂O₃ − TiO₂ + hν → Fe₂O₃ − TiO₂(e⁻ + h⁺) (4)

e⁻ + O₂ → O₂⁻ (5)

h⁺ + H₂O → ˙OH + H⁺ (6)

OTC + h⁺/˙OH/O₂⁻ → intermediate products → CO₂↑ + H₂O (7)

At same time, Fe³⁺ species in 45 Fe₂O₃–TiO₂ NPs photocatalysts acting as h⁺/e⁻ traps, that prevent photo-generated electron–hole recombination. h⁺ and ˙OH generated in the valance band of Fe₂O₃/TiO₂ under visible light irradiation (eqn (8)–(11)) and improve photo activities for the decomposition of pollutants under visible light (eqn (7)).^17–19

Fe³⁺ + e⁻ → Fe²⁺ (8)

Fe²⁺ + O₂ → Fe³⁺ + O₂⁻ (9)

Fe³⁺ + h⁺ → Fe⁴⁺ (10)

Fe⁴⁺ + OH⁻ → Fe³⁺ + ˙OH (11)

On the basis of LC-MS/TOF results, the degradation pathway of OTC using 45 Fe₂O₃–TiO₂ NPs is proposed in Scheme 1. In the OTC solution after photocatalytic degradation, P₂, P_3, P₄, P₅ and P₇ were founded by LC-MS/TOF. It is proposed that the photocatalytic degradation of P₁ includes de-methylation and de-hydroxylation, which may produce P₄.²⁰ Further degradation of P₂ leads to the generation of P₅ via further loss of two N-methyl at aromatic ring 4 by attacking of ˙OH radicals groups, because of the low bond energy of N–C.²¹ Product P₆ resulted from the hydroxyl groups connected to the degraded aromatic ring 1.²² OTC as a kind of antibiotics with stable naphthol ring as main structure and toxicity in wastewater. And some degradation byproducts of OTC have been proved to be more toxic than themselves.^21,23 The naphthol ring in P₆ have been proved easy to open in the present of ˙OH radicals groups compared to other intermediate products.²⁴ Thus, P₇ with a partial ring-ruptured structure may stems from P₆ by further attacked by ˙OH radicals, which is not reported in other literatures.

Scheme 1 Proposed degradation pathways of OTC ([45 Fe₂O₃–TiO₂ NPs] = 1 g L⁻¹, OTC: 60 mg L⁻¹, pH = 5.5, UV-vis experiment).

In summary, product P₅ could be formed in the reaction route of OTC (C₂₂H₂₄N₂O₉) → P₂ (C₂₂H₂₃NO₉) → P₅ (C₂₀H₁₉NO₈), product P₄ and P₆ could be produced as the pathways of OTC → P₁ (C₂₁H₂₂N₂O₉) → P₄ (C₂₁H₂₂N₂O₇) and OTC → P₆ (C₂₂H₂₄N₂O₁₀) → P₇ (C₉H₆O₄). While product P₃ could be came from OTC → P₃ (C₂₂H₂₂N₂O₈). It is suggested that products P₃–P₆ may be further scission to small amounts such as P₇ by ˙OH radicals through naphthol ring-rupturing reactions, even these intermediate products were not detected by LC-MS/TOF. At the end of photocatalytic degradation, H₂O, CO₂ gas and other products could be obtained as final products with properly conditions (such as enough reaction time, light irradiation and photocatalyst).^25,26 However, the detail mechanism of OTC degradation will be further studied in the future.

4. Conclusions

Fe₂O₃–TiO₂ NPs photocatalysts were synthesized using co-precipitation, and characterized by XRD, TEM, FT-IR and UV-vis DRS. These results shows that Fe₂O₃–TiO₂ NPs consisted of hematite Fe₂O₃ and anatase TiO₂ with particle size of 20 nm, exhibited strong absorption in visible region. Fe₂O₃–TiO₂ NPs with 45 wt% Fe₂O₃ exhibited higher photocatalytic activity than others, and all samples have more activity than pure TiO₂ under visible light. The optimized pH for remove OTC was 5.5 under both UV-vis light due to the surface adsorption. The reusability of 45 Fe₂O₃–TiO₂ NPs has also studied, and after 3 cycles the degradation efficiency still remained at same level. The intermediate products during the photocatalytic degradation of OTC were detected by LC-MS/TOF. Five main intermediates as P₂ (C₂₂H₂₃NO₉), P₃ (C₂₂H₂₂N₂O₈), P₄ (C₂₁H₂₂N₂O₇) and P₇ (C₉H₆O₄) were formed after photocatalytic degradation, among these products, P₇ has a partial ring-ruptured structure, indicating that the complete degradation of OTC to mineral end-products. However, the more details need to be investigated in the future work to understand the mechanism.

Acknowledgements

The authors would like to thank Instrumental Analysis and Research Center of Shanghai University for the experimental assistance on the TEM and XRD, Ms. Rong Jin of Institude of Nanochemistry and Nanobiology, Shanghai University for the experimental assistance on the UV-visible diffuse reflectance spectra measurement. We also greatly appreciate Ms. Qun Qian of Department of Chemistry, Shanghai University for the experimental assistance on LC-MS/TOF. This work was supported by the National Natural Science Foundation of China (nos 51272154, 51472156), the Innovation Foundation of Shanghai University (nos sdcx2012033, sdcx2012062), the Special Research Foundation for Training and Selecting Outstanding Young Teachers of Universities in Shanghai.

References

1. C. Zhao, H. P. Deng, Y. Li and Z. H. Liu, J. Hazard. Mater., 2010, 176, 884–892.
2. C. Zhao, M. Pelaez, X. Duan, H. P. Deng and K. O'Shea, Appl Catal B-Environ., 2013, l 134–135, 83–92.
3. H. Liu, H. K. Shon, X. Sun, S. Vigneswaran and H. Nan, Appl. Surf. Sci., 2011, 257, 5813–5819.
4. Y. Gao, B. H. Chen, H. L. Li and Y. X. Ma, Mater. Chem. Phys., 2003, 80, 348–355.
5. K. P. Astam and A. B. Arghyautta, ACS Appl. Mater. Interfaces., 2012, 4, 5022–5028.
6. X. L. Liu, P. Lv, G. X. Yao, C. C. Ma, P. W. Huo and Y. S. Yan, Chem. Eng. J., 2013, 217, 398–406.
7. B. M. Reddy, K. J. Ratnam and P. Saikia, J. Mol. Catal. A: Chem., 2006, 252, 238–244.
8. A. K. Patra, A. Dutta and A. Bhaumik, ACS Appl. Mater. Interfaces, 2012, 4, 5022–5028.
9. J. G. Yu, X. J. Zhao, J. J. Yu, G. R. Zhong, J. J. Han and Q. G. Zhao, J. Mater. Sci. Lett., 2001, 20, 1745–1748.
10. P. Bonamali, S. Maheshwar Sharon and N. Gyoichi, Mater. Chem. Phys., 1999, 59, 254–261.
11. S. A. Sassman and L. S. Lee, Environ. Sci. Technol., 2005, 39(19), 7452–7459.
12. H. H. Chen, E. N. Charith and H. G. Vicki, Chem. Rev., 2012, 112, 5919–5948.
13. B. Palanisamy, C. M. Babu, B. Sundaravel, S. Anandan and V. Murugesan, J. Hazard. Mater., 2013, 252–253, 233–242.
14. X. L. Liu, P. Lv, G. X. Yao, C. C. Ma, P. W. Huo and Y. S. Yan, Chem. Eng. J., 2013, 217, 398–406.
15. M. H. Khan, H. Bae and Y. J. Jin, J. Hazard. Mater., 2010, 181, 659–665.
16. Y. Wang, H. Zhang, J. H. Zhang, C. Liu, Q. Q. Huang, J. Wu and F. Liu, J. Hazard. Mater., 2011, 192, 35–43.
17. M. A. Ahmed, E. E. Emad and H. G. Zarha, J. Alloys Compd., 2013, 553, 19–29.
18. F. M. Dukes, E. Iuppa, B. Meyer and M. J. Shultz, Langmuir, 2012, 28, 16933–16940.
19. Q. L. Jin, F. Musashi and T. Hiroaki, J. Phys. Chem. C, 2011, 115, 6478–6483.
20. J. H. O. S. Pereira, V. J. P. Vilar, M. T. Borges, O. Gonźalez, S. Esplugas and R. A. R. Boaventura, Sol. Energ., 2011, 85, 2732–2740.
21. S. J. Jiao, S. R. Zheng, D. Q. Yin, L. H. Wang and L. Y. Chen, J. Environ. Sci., 2008, 20, 806–813.
22. V. M. Mboula, V. Héquet, Y. Gru, R. Colin and Y. Andrès, J. Hazard. Mater., 2012, 209–210, 355–364.
23. S. J. Jiao, S. R. Zheng, D. Q. Yin, L. H. Wang and L. Y. Chen, Chemosphere, 2008, 73(3), 377–382.
24. J. F. Niu, S. Y. Ding, L. W. Zhang, J. B. Zhao and C. H. Feng, Chemosphere, 2013, 93, 1–8.
25. X. D. Zhu, Y. J. Wang, R. J. Sun and D. M. Zhou, Chemosphere, 2013, 92, 925–932.
26. Z. Y. Liu, P. W. Huo, Y. Y. Luo, X. L. Liu, D. Wu, X. Gao, C. X. Li and Y. S. Yan, J. Mol. Catal. A: Chem., 2013, 378, 91–98.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra04540a

---