Construction of a magnetic Z-scheme photocatalyst with enhanced oxidation/reduction abilities and recyclability for the degradation of tetracycline

Abstract

Here, we successfully designed and prepared the magnetic recyclable Z-scheme photocatalyst WO₃/Fe₃O₄/g-C₃N₄ for the first time. The structures, morphology and photocatalytic properties of as-prepared photocatalysts were characterized by a series of modern characterization methods. According to the photoluminescence, transient fluorescence, transient photocurrent and electrochemical impedance spectra, the introduction of Fe₃O₄, with its high conductivity, into WO₃/Fe₃O₄/g-C₃N₄ enhanced the lifetime of photogenerated holes and electrons based on the Z-scheme charge carrier transfer mechanism. More importantly, WO₃/Fe₃O₄/g-C₃N₄ showed good recyclability without loss of apparent photocatalytic activity even after five cycles. Therefore, this kind of material may be a promising candidate for applications in the fields of environmental protection and sewage treatment.

---

1 Introduction

Over the past few decades, increasing environmental problems have become a huge challenge to the development of modern human society,^1–3 with water pollution being an especially challenging issue. The antibiotic drug tetracycline shows strong sterilization abilities, displays antibacterial activity, and is widely used for the treatment of diseases in humans and animals. However, the large residual amounts of tetracycline in the aquatic environment have serious negative effects on the lives of people. For example, tetracycline has been shown to cause gastrointestinal maladies such as nausea, vomiting, abdominal distension, and diarrhea. Therefore, finding a reasonable and efficient method to treat this antibiotic pharmaceutical in wastewater is very important. Many approaches and technologies such as adsorption,⁴ microbial degradation,⁵ and electrolysis⁶ have been developed and widely used to address this problem, but they all appear to us to have their own drawbacks. However, photocatalysis has drawn significant attention due to the ability of this approach to remove organic pollutants from aqueous solutions by completely decomposing mineralized forms of these organic pollutants to carbon dioxide.

Recently, g-C₃N₄, an effective organic semiconductor photocatalyst, has attracted particular attention for treatment of organic pollutants, because of its low cost, thermal stability, and narrow band gap of 2.7 eV.⁷ However, the disadvantages of pure g-C₃N₄ limit its application. First, it undergoes a rapid recombination of photogenerated electron–hole pairs, and this process results in a low efficiency of its visible-light utilization. Thus, the visible-light activity of g-C₃N₄ is low in practical applications.^8–10 In order to enhance the photocatalytic activity of g-C₃N₄, many strategies such as coupling or doping g-C₃N₄ with semiconductor materials, nonmetals, and metals have been used to modify g-C₃N₄.

Tungsten oxide (WO₃), a typical n-type semiconductor, is recognized as a promising photocatalyst with an optical band gap of 2.4–2.8 eV, excellent chemical stability, a long photo-generated electron lifetime, and high electron mobility, and for these reasons as well as its low cost has been extensively studied for environmental remediation.^11–13 Unfortunately, WO₃ does suffer from certain drawbacks hindering the photocatalytic process, mainly (i) its conduction band (CB) level being more positive than that of other semiconductors, which results in the electrons generated in its CB having a limited reductive ability, and (ii) its relatively high electron–hole recombination rate during photocatalytic reactions.¹⁴ Therefore, the problems described above remain unsolved and require urgent attention.

Recently, the concept of all-solid-state Z-scheme photocatalysts with strong redox capacities and improved photocatalytic performance relative to the single-component photocatalysts was proposed. The sandwich-structured CdS/Au/TiO₂ photocatalyst¹⁵ was reported as such a Z-scheme photocatalyst, with the Au nanoparticles (NPs) anchored between CdS and TiO₂ and acting as electron-transfer mediators. It was found that the photogenerated electrons in the CB of TiO₂ could transfer through the Au layer and recombine with the photogenerated holes left in the valence band (VB) of CdS, and prolong the lifetime of the carriers, and hence further enhance the photocatalytic activity.

Inspired by this concept and example, the Z-scheme WO₃/Fe₃O₄/g-C₃N₄ photocatalyst was constructed to overcome the individual above-described drawbacks of WO₃ and g-C₃N₄. Moreover, the introduced Fe₃O₄ not only has preferable magnetic properties, and can be hence easily separated from the mixture with a magnet and be recycled conveniently, but it can also serve as a conductor between the WO₃ and g-C₃N₄ components and can participate in the transmit of photogenerated charge carriers. Combining magnetic Fe₃O₄ with a semiconductor for improving the photocatalytic activity has been reported, with such composites including TiO₂/Fe₃O₄, WO₃/Fe₃O₄, and graphene/Fe₃O₄.^16–18 Therefore, Fe₃O₄ was introduced between g-C₃N₄ and WO₃ and was able to significantly overcome the above drawbacks.

In the present study, magnetic Z-scheme-structured WO₃/Fe₃O₄/g-C₃N₄ photocatalysts with high activity was successfully prepared for the first time. Moreover, this as-prepared photocatalyst was extensively characterized using X-ray diffraction (XRD), Fourier transform infrared spectroscopy (FT-IR), transmission electron microscopy (TEM), vibrating sample magnetometry (VSM), UV-vis diffuse reflectance spectroscopy (UV-vis), photoluminescence spectroscopy (PL), photocurrent-time measurements, transient photocurrents and electrochemical impedance spectroscopy (EIS), Brunauer–Emmett–Teller (BET) measurements, and electron spin resonance (ESR). As expected, WO₃/Fe₃O₄/g-C₃N₄ displayed strong photocatalytic activity for removing tetracycline under visible light. Finally, the mechanism by which WO₃/Fe₃O₄/g-C₃N₄ catalyzed the degradation of tetracycline was also systematically investigated.

2 Experimental

2.1. Chemicals

Tetracycline was purchased from the National Institutes for Food and Drug Control. Melamine (AR), iron(III) chloride hexahydrate (FeNO₃·9H₂O, AR), C₂H₃O₂Na·3H₂O (AR), polyethylene glycol 1500 (PEG, 1500), ethylene glycol (98.0%), polyvinyl pyrrolidone (PVP) and sodium tungstate (Na₂WO₄·2H₂O, AR), were all supplied by Aladdin Chemistry Co., Ltd. NH₄OH in solution form (25.0%), HCl in solution form (37%), p-benzoquinone (BQ, AR), isopropanol (IPA, AR), triethanolamine (OA, AR), and ethanol (C₂H₅OH, AR) were all purchased from Sinopharm Chemical Reagent Co., Ltd. Deionized water was used throughout this work.

2.2. Structural analysis

The as-prepared samples were characterized by using an X-ray diffractometer (XRD, Rigaku, Japan) between 5° and 80° with a scanning step of 5° min⁻¹. Transmission electron microscope (TEM) images were acquired with a JEM-2100 transmission electron microscope (JEOL, Japan). Fourier transform infrared (FT-IR) spectra were recorded on a Nicolet Nexus 470 FT-IR (Thermo Nicolet Co., USA) at a resolution of 2.0 cm⁻¹ between 400 cm⁻¹ and 4000 cm⁻¹. Magnetic measurements were taken using a vibrating sample magnetometer (VSM) (7300, Lakeshore) under a magnetic field up to 10 kOe. UV-vis diffuse reflectance spectra (UV-vis DRS) of the samples were obtained using a Shimadzu UV-3600 spectrometer with BaSO₄ as a reference. Photoluminescence (PL) spectra were obtained from a F4500 PL detector (Hitachi, Japan). PL of C₃N₄ was investigated by using a spectrophotometer (Cary Eclipse spectrophotometer, VARIAN, USA). Transient photocurrents and electrochemical impedance spectroscopy (EIS) were investigated by using a ZENNIUM electrochemical workstation (Zahner Instruments, Germany).

2.3. Preparation of the magnetic WO₃/Fe₃O₄/g-C₃N₄ photocatalyst

First, g-C₃N₄ was prepared by using a previously published method with some modifications. Typically, a mass of 3.0 g of melamine in an open crucible was subjected to a constant temperature of 500 °C for 2 h in a muffle, then heated with a ramping rate of 2.3 °C min⁻¹ to 550 °C and held at this temperature for another 2 h, after which it was allowed to naturally cool to room temperature to obtain the g-C₃N₄ powders.

Second, a mass of 0.5 g Na₂WO₄·2H₂O was dissolved in 30 mL deionized water with vigorous stirring for 30 min, and 0.2 mol L⁻¹ HCl was introduced dropwise into the above aqueous solution to adjust the pH to 1. The resulting solution was then transferred into a 50 mL Teflon-lined stainless steel autoclave and solvothermally treated at 120 °C for 5 h. Afterwards, the product was then dried at 80 °C for 15 h and calcined at 500 °C for 1 h to yield the WO₃ powder.

Finally, certain amounts of the as-prepared g-C₃N₄ and WO₃ were dispersed in 30 mL ethanediol containing 0.3 g Fe(NO₃)₃·9H₂O, 0.1 g C₂H₃O₂Na·3H₂O, 0.03 g PEG, and 0.005 g PVP. After the mixed components fully dissolved, the solution was transferred into a 50 mL Teflon-lined stainless steel autoclave and kept at 190 °C for 15 h. After the product was cooled to ambient temperature, it was washed with deionized water and ethanol several times to obtain WO₃/Fe₃O₄/g-C₃N₄.

2.4. Photocatalytic and active species trapping experiments

The photocatalytic performances of the as-prepared samples were evaluated by monitoring the degradation of tetracycline under irradiation with visible light. A 300 W xenon lamp covered with a UV filter (λ > 400 nm) was used as a light source. In brief, a mass of 0.1 g of catalyst in 100 mL of an aqueous solution of tetracycline (20 mg L⁻¹) was magnetically stirred in the dark to reach an adsorption–desorption equilibrium. Every 20 min, a volume of 3 mL of the mixture was removed and separated by using a magnet, and the absorbance of each such sample was measured by using a UV-vis spectrophotometer at λ = 357 nm. The active species trapping experiments were carried out in a manner similar to the way the photocatalytic experiments were performed, except that only 1 mmol triethanolamine (OA, a quencher of h⁺), 1 mmol isopropanol (IPA, a quencher of ·OH), and 1 mmol benzoquinone (BQ, a quencher of ·O₂⁻) were, respectively, added into the aqueous tetracycline solution (20 mg L⁻¹, 100 mL) before the start of the reaction; the next steps were the same in the degradation experiments described above.

2.5. Photoelectrochemical (PEC) measurements

To prepare the electrodes for the photoelectrochemical (PEC) experiments, a mass of 0.1 g of the as-prepared photocatalyst was dispersed in a mixture of 1.0 mL ethanol and 1.0 mL ethanol glycol, then subjected to spin-coating on a fluorine-doped tin oxide glass electrode (1 × 1 cm²), and calcined in air at 80 °C. The PEC measurements were taken on a CHI760D electrochemical analyzer (ChenHua Instruments, Shanghai, China). The working electrodes were photocatalyst thin films. A platinum wire electrode and Ag/AgCl electrode were used as the counter electrode and reference electrode, respectively. The PEC measurements were taken with 0.5 M Na₂SO₄ electrolyte, and a 300 W Xe lamp served as the light source for these measurements.

3 Results and discussion

3.1. XRD and FT-IR analyses

Fig. 1 shows the XRD patterns of pure g-C₃N₄, Fe₃O₄/g-C₃N₄, and WO₃/Fe₃O₄/g-C₃N₄. Two distinct diffraction peaks, at 27.5° and 13.1°, were observed for the prepared g-C₃N₄ (Fig. 1a), which were indexed to the (002) and (100) peaks in JCPDS 87-1526. Fe₃O₄/g-C₃N₄ also yielded distinct diffraction peaks at 2θ = 30.2, 35.5, 43.2, 57.3 and 62.6° (Fig. 1b), consistent with diffraction peaks of Fe₃O₄.^19,20 More importantly, WO₃O/Fe₃O₄/g-C₃N₄ yielded additional strong diffraction peaks (Fig. 1c), and all of the peaks here matched well with the monoclinic phase WO₃ (JCPDS card no. 72-0677).^21,22 In addition, all the diffraction peaks at 27.4° and 35.5°, 43.2°, 57.3°, 62.6° are corresponding to lattice plane of (002) of g-C₃N₄ and (311), (400), (511), (440) of Fe₃O₄ could also be found in WO₃/Fe₃O₄/g-C₃N₄. These observations hence suggested that the phase of g-C₃N₄ did not change with the introduction of WO₃, and that WO₃/Fe₃O₄/g-C₃N₄ was successfully synthesized.^23,24

Fig. 1 XRD patterns of g-C₃N₄ (a), Fe₃O₄/g-C₃N₄ (b), and WO₃/Fe₃O₄/g-C₃N₄ (c).

FT-IR spectra of g-C₃N₄, Fe₃O₄/g-C₃N₄ and WO₃/Fe₃O₄/g-C₃N₄ were also acquired, and are shown in ESI (Fig. S1†). A series of peaks from 1650 cm⁻¹ to 1200 cm⁻¹ (1251, 1325, 1419, 1571, and 1639 cm⁻¹) and a peak near 808 cm⁻¹ appeared (Fig. S1a–c†), which corresponded to the typical stretching modes of CN heterocycles.²⁵ Clearly, the main peaks of g-C₃N₄ appeared in the spectrum of Fe₃O₄/g-C₃N₄ (Fig. S1b†) and were similar to those of pure g-C₃N₄, indicating that g-C₃N₄ in Fe₃O₄/g-C₃N₄ maintained the same chemical structure as did pure g-C₃N₄ after solvothermal treatment. However, Fe₃O₄/g-C₃N₄ also yielded an extra peak at 550–650 cm⁻¹ (Fig. S1b†) which might have been due to an Fe–O bond vibration.^26,27 The peaks mentioned above were all weakened in the WO₃/Fe₃O₄/g-C₃N₄ spectrum (Fig. S1c†), i.e., after deposition of WO₃ nanoparticles, because the surface functional groups of g-C₃N₄ may have been covered by the deposited WO₃, and as the FT-IR results showed the introduction of Fe₃O₄ and WO₃ to have no effect on the chemical structure of g-C₃N₄.

3.2. TEM analysis

The morphology of the as-prepared sample was further investigated by using TEM. The g-C₃N₄ nanosheets were clearly observed to be thin and to overlap each other (Fig. 2a). Meanwhile, the TEM image of Fe₃O₄/g-C₃N₄ (Fig. 2b) showed the Fe₃O₄ nanoparticles to have an average diameter of about 50–100 nm, and to be dispersed homogeneously on the g-C₃N₄ surface. And the TEM image of WO₃/Fe₃O₄/g-C₃N₄ (Fig. 2c) showed the presence of small particles on the Fe₃O₄/g-C₃N₄ surface; moreover, the Fe₃O₄ nanoparticles and g-C₃N₄ nanosheet were less transparent in this composite than in the other two forms examined. In addition, analysis of the EDS spectrum of WO₃/Fe₃O₄/g-C₃N₄ (Fig. 2d) clearly revealed the presence of the elements Fe, O, W, C and N. The EDS and XRD results taken together confirmed the presence of WO₃, Fe₃O₄, and g-C₃N₄ in the WO₃/Fe₃O₄/g-C₃N₄ Z-scheme system. In addition, the specific surface areas of g-C₃N₄, Fe₃O₄/g-C₃N₄ and WO₃/Fe₃O₄/g-C₃N₄ were 6.9, 50.3 and 87.6 m² g⁻¹, respectively. Fe₃O₄/g-C₃N₄ were lower than that of WO₃/Fe₃O₄/g-C₃N₄. The surface roughness levels of the as-prepared samples were also observed to differ. This difference might have contributed to the photocatalytic activity of WO₃/Fe₃O₄/g-C₃N₄ being greater than those of g-C₃N₄ and Fe₃O₄/g-C₃N₄: the more expansive and rougher BET surface of WO₃/Fe₃O₄/g-C₃N₄ can adsorb more tetracycline and further improve the photocatalytic degradation efficiency.

Fig. 2 TEM images of g-C₃N₄ (a), Fe₃O₄/g-C₃N₄ (b), WO₃/Fe₃O₄/g-C₃N₄ (c), and an EDS spectrum of WO₃/Fe₃O₄/g-C₃N₄ (d).

3.3. Photocatalytic activity

We first sought to determine the amount of Fe₃O₄ in Fe₃O₄/g-C₃N₄ that optimized its photocatalytic performance. From degradation dynamics curves of tetracycline over Fe₃O₄/g-C₃N₄ with different contents of Fe₃O₄ (Fig. 3a), Fe₃O₄/g-C₃N₄ was determined to exhibit the highest photocatalytic activity when the percentage of Fe₃O₄/g-C₃N₄ made up of Fe₃O₄ was 5.0 wt%. Thus, Fe₃O₄ not only has the magnetic separation performance, but also played an important role in enhancing the photocatalytic activity. We also determined the efficiency levels of the degradation of tetracycline by WO₃/Fe₃O₄/g-C₃N₄ with different amounts of WO₃ (Fig. 3b). The photocatalytic activity of WO₃/Fe₃O₄/g-C₃N₄ was optimal when 1.0 wt% of it was made up of WO₃ nanoparticles: this composition degraded 89% of tetracycline over the course of a photocatalytic reaction of 120 min. This result suggested that too much WO₃ covering the Fe₃O₄/g-C₃N₄ surface would diminish the ability of the material to harvest light. In contrast, too little WO₃ would weaken the formation of heterojunctions between WO₃ and g-C₃N₄. Therefore, the formation of heterojunctions between WO₃ and g-C₃N₄ appeared to be a major enhancer of photocatalytic activity owing to the ability of these heterojunctions to efficiently separate the charge carriers. The degradation dynamics curves of tetracycline in Fig. 3c showed the decomposition effect of g-C₃N₄ to have been significantly enhanced after it was modified with Fe₃O₄ and WO₃. We also acquired absorbance spectra of tetracycline over WO₃/Fe₃O₄/g-C₃N₄ for various photodegradation reaction times (Fig. 3d). The absorbance nearly disappeared in 120 min. This result suggested that the tetracycline molecules were completely decomposed into smaller molecules.²⁸

Fig. 3 (a and b) Degradation of tetracycline over different samples under visible light irradiation. (a) Fe₃O₄/g-C₃N₄ with different Fe₃O₄ contents: 40 wt%, 20 wt%, 10 wt%, 5.0 wt%, and 3.0 wt%. (b) WO₃/Fe₃O₄/g-C₃N₄ with different WO₃ contents: 10 wt%, 5.0 wt%, 2.0 wt%, 1.0 wt%, and 0.5 wt%. (c) The photocatalytic effect curves of photo-degradation solutions over pure g-C₃N₄, Fe₃O₄/g-C₃N₄, and WO₃/Fe₃O₄/g-C₃N₄. (d) Absorbance of 20 mg L⁻¹ tetracycline solutions after various reaction times with WO₃/Fe₃O₄/g-C₃N₄.

3.4. UV-vis analysis

UV-vis diffuse reflectance spectroscopy was used to determine the absorption levels of the as-prepared pure g-C₃N₄ and the Fe₃O₄/g-C₃N₄ and WO₃/Fe₃O₄/g-C₃N₄ composites (Fig. 4). The pure g-C₃N₄ displayed an absorption edge around 450 nm, which corresponded to a band gap of 2.76 eV. The absorption edge of Fe₃O₄/g-C₃N₄ in the visible-light range was significantly increased, with a red shift of more than 700 nm relative to that of the pure g-C₃N₄. More importantly, the diffuse reflectance absorption of WO₃/Fe₃O₄/g-C₃N₄ was stronger than that of Fe₃O₄/g-C₃N₄, and the enhanced light absorption was caused by the introduced WO₃ nanoparticles. The enhanced light absorption may have generated more photo-induced electron–hole pairs under visible-light irradiation, which subsequently enhanced the photocatalytic activity. Therefore, the heterojunction character of the WO₃/Fe₃O₄/g-C₃N₄ catalyst was able to effectively enhance its light absorption and photocatalytic activity. UV-vis diffuse reflectance spectra with different Fe₃O₄ contents are shown in Fig. S2.†

Fig. 4 UV-vis absorbance spectra of g-C₃N₄, Fe₃O₄/g-C₃N₄, and WO₃/Fe₃O₄/g-C₃N₄.

3.5. Photoluminescence emission spectra

PL emission spectra were acquired to investigate the recombination and separation of the photoexcited electron–hole pairs, and the intensity of the PL emission spectra indicated the recombination speeds of these pairs. Generally, a higher PL intensity means a higher recombination rate of electron–hole pairs.^29,30 As shown in Fig. 5, the PL intensity of g-C₃N₄ was the highest of the three as-prepared samples, and as expected, the PL response of WO₃/Fe₃O₄/g-C₃N₄ was the lowest. This difference indicated the importance of WO₃ and Fe₃O₄ in the Z-scheme system and indirectly showed WO₃/Fe₃O₄/g-C₃N₄ to have the lowest electron–hole pair recombination rate.

Fig. 5 Photocurrent response curves of g-C₃N₄, Fe₃O₄/g-C₃N₄, and WO₃/Fe₃O₄/g-C₃N₄ under irradiation with visible light.

3.6. Photocurrent tests

Photocurrent testing is considered to be an efficient method for demonstrating the generation and separation of electron–hole pairs in photocatalytic materials. Here, the higher the photocurrent, the higher the electron–hole pair separation efficiency, and the higher the photocatalytic activity.^31,32 Fig. 6 shows the photocurrent responses of pure g-C₃N₄, Fe₃O₄/g-C₃N₄ and WO₃/Fe₃O₄/g-C₃N₄. The photocurrent of pure g-C₃N₄ was the lowest of the three samples. And satisfyingly, the photocurrent of WO₃/Fe₃O₄/g-C₃N₄ was the highest, about three times that of pure g-C₃N₄, and remained at a nearly constant value after the light was turned on and after it was turned off. Therefore, the results of the photocurrent test indirectly showed the WO₃/Fe₃O₄/g-C₃N₄ to have the lowest electron–hole pair recombination rate and the highest photocatalytic activity. The photocurrent response curves of WO₃/Fe₃O₄/g-C₃N₄ with different contents of WO₃ were also acquired, and are displayed in Fig. S3.†

Fig. 6 Photocurrent response curves of g-C₃N₄, Fe₃O₄/g-C₃N₄, and WO₃/Fe₃O₄/g-C₃N₄ under visible light irradiation.

3.7. Electrochemical impedance spectroscopy

Electrochemical impedance spectroscopy (EIS) measurements were taken (Fig. 7) to investigate the charge transfer resistance and separation efficiency of the photoinduced charge carriers. Of the three samples tested, the diameter of the Nyquist semicircle was greatest and lowest for g-C₃N₄ and WO₃/Fe₃O₄/g-C₃N₄, respectively, indicating them to have the highest and lowest levels of resistance, respectively. These EIS results indicated that the introduction of Fe₃O₄ and WO₃ into g-C₃N₄ enhanced the separation and transfer efficiency of photoinduced electron–hole pairs, which is a favorable condition for improving the photocatalytic activity. These result reveals that the WO₃/Fe₃O₄/g-C₃N₄ has a preferable efficiency of charge transfer after modified by WO₃ nanoparticles. In addition, EIS spectra of WO₃/Fe₃O₄/g-C₃N₄ with different contents of WO₃ were also acquired and are shown in Fig. S4.†

Fig. 7 EIS spectra of g-C₃N₄, Fe₃O₄/g-C₃N₄, and WO₃/Fe₃O₄/g-C₃N₄.

3.8. Detection of reactive species

When photocatalysts are irradiated by visible light, active species such as holes (h⁺), hydroxyl radicals (·OH), and superoxide radicals (·O₂⁻) are generated during the photocatalytic process.^33,34 In the current work, we carried out radical trapping experiments (results shown in Fig. 8), using benzoquinone (BQ) to reduce the levels of ·O₂⁻, isopropanol (IPA) to impair the formation of ·OH, and triethanolamine (TEAO) to remove h⁺. Only 10% of the tetracycline was degraded over WO₃/Fe₃O₄/g-C₃N₄ in the presence of IPA, in contrast to the near-90% degradation without any trapping agents. This result revealed the very strong effect that ·OH had during the photo-degradation of tetracycline. Also, only about 25% of the tetracycline was degraded in the presence of TEAO, indicating that h⁺ also played crucial role in the degradation of tetracycline. In addition, when BQ was used as the superoxide radical scavenger agent, 30% of the tetracycline was degraded. These results taken together indicated the order of the influence of the activated species on the tetracycline photo-degradation process to be ·OH > h⁺ > ·O₂⁻.

Fig. 8 Effects of reactive species of different scavengers on the photocatalytic activity of WO₃/Fe₃O₄/g-C₃N₄.

3.9. Detection of reactive oxygen species using ESR

The ESR technique has been used to detect the presence of ·OH and ·O₂⁻ radicals in photocatalytic reaction systems. As shown in Fig. 9a, four characteristic peaks for DMPO–·OH adducts were observed (1 : 2 : 2 : 1 quartet pattern) with WO₃/Fe₃O₄/g-C₃N₄ subjected to irradiation with visible light. Similarly, four characteristic peaks (Fig. 9b) corresponding to the DMPO–·O₂⁻ adduct were observed, indicating that ·O₂⁻ radicals were also produced when WO₃/Fe₃O₄/g-C₃N₄ was irradiated with visible light. The ESR results confirmed that both ·O₂⁻ and ·OH were present, and that the active oxidation species played an important role in the photodegradation of tetracycline.

Fig. 9 ESR spectra of WO₃/Fe₃O₄/g-C₃N₄ (a) DMPO–·OH and (b) DMPO–·O₂⁻ resulting from irradiation with visible light.

3.10. VSW and reusability

A vibrating sample magnetometer (VSM) was employed to compare the magnetic properties of Fe₃O₄/g-C₃N₄ and WO₃/Fe₃O₄/g-C₃N₄. Their symmetric hysteresis loops are shown in Fig. S5.† The values of magnetization saturation (M_s) of Fe₃O₄/g-C₃N₄ and WO₃/Fe₃O₄/g-C₃N₄ were 30.06 Oe and 20.41 Oe, respectively. The relatively weakened M_s value of WO₃/Fe₃O₄/g-C₃N₄ could be explained by the introduced WO₃ nanoparticles. However, when an external magnetic field was applied, the WO₃/Fe₃O₄/g-C₃N₄ became attracted to the wall of the beaker and the dispersion became clear and transparent, indicating that the WO₃/Fe₃O₄/g-C₃N₄ still showed good magnetic properties. In addition, the photocatalytic stability of WO₃/Fe₃O₄/g-C₃N₄ was also tested, because such stability is crucially important for practical applications. The recycling capability of WO₃/Fe₃O₄/g-C₃N₄ was verified by carrying out a five-run test, which is shown in Fig. S6.† The results of this test revealed no obvious decrease in the degradation rate, which revealed the excellent photocatalytic stability of the Z-scheme WO₃/Fe₃O₄/g-C₃N₄ photocatalyst.

3.11. Mechanism of enhanced photocatalytic performance

A schematic drawing illustrating the proposed synergistic effect in the photocatalytic degradation of tetracycline over WO₃/Fe₃O₄/g-C₃N₄ is shown in Fig. 10. Under irradiation with visible light, the electrons and holes in this proposal are excited from WO₃ and g-C₃N₄ (eqn (1) and (2)).³⁵ And then, the weak reductive photoelectrons in the conduction band (CB) of g-C₃N₄ could transfer to the CB of WO₃ (eqn (3)). Simultaneously, the weak oxidative holes in the VB of WO₃ could transfer to the VB of g-C₃N₄ (eqn (4)). Then, the photogenerated electrons and holes could annihilate at Fe₃O₄ (eqn (5)).^36–38 And the lifetime of charge carriers photogenerated from g-C₃N₄ and WO₃ might be increased. As a result, the WO₃/Fe₃O₄/g-C₃N₄ photocatalyst would show higher photocatalytic activity. Moreover, according to this proposal, the photogenerated electrons from g-C₃N₄ react with O₂ and reduce it to ·O₂⁻ (eqn (6)). Subsequently, the ·O₂⁻ would react with H⁺ in the solution and generate ·OH (eqn (7)). As both the ·O₂⁻ and ·OH are the reactive species, they would be the agents causing tetracycline to decompose into smaller molecules (eqn (8) and (9)). Meanwhile, the holes generated from the valence band of g-C₃N₄ and WO₃ can also directly oxidize the tetracycline dye (eqn (10)). Finally, the tetracycline gradually becomes consumed as the reaction progresses. This photocatalytic mechanism can be summarized as follows:

WO₃ + hν → WO₃ (e⁻ + h⁺) (1)

g-C₃N₄ + hν → g-C₃N₄ (e⁻ + h⁺) (2)

g-C₃N₄ (e⁻) → WO₃ (CB) (3)

WO₃ (h⁺) → g-C₃N₄ (VB) (4)

WO₃ (e⁻) + g-C₃N₄ (h⁺) → Fe₃O₄ (annihilate) (5)

O₂ + g-C₃N₄ (e⁻) → ·O₂⁻ (6)

·O₂⁻ + 2H⁺ → 2·OH (7)

·O₂⁻ + tetracycline → small-molecule products (8)

·OH + tetracycline → small-molecule products (9)

h⁺ + tetracycline → small-molecule products (10)

Fig. 10 Schematic illustration of the generation, separation and transport of photogenerated charge carriers at the WO₃/Fe₃O₄/g-C₃N₄ interfaces.

4 Conclusions

In summary, a novel Z-scheme photocatalyst, WO₃/Fe₃O₄/C₃N₄, was successfully prepared and showed high photocatalytic activity (90%) towards the degradation of tetracycline according to the results of PL spectroscopy, transient photocurrent, EIS spectroscopy and scavenger experiments. The increased photocatalytic activity was attributed to the high conductivity of Fe₃O₄, which acted as an intermediate agent for making the holes (g-C₃N₄) and electrons (WO₃) annihilate effectively, and further enhanced the activity of the photocatalysts. On the basis of the results of this study, the present approach used to prepare the photocatalyst and the described mechanism can provide valuable information for the development of magnetic-based Z-scheme photocatalysts that remove pollutants from the environment.

Acknowledgements

This work was financially supported by the Young Teacher Funding Scheme from Pingdingshan University (No. 2014037) and the National Scientific Research Project Cultivating Foundation of Pingdingshan University (No. PXY-PYJJ2016005).

References

1. F. Sastre, A. V. Puga, L. Liu, A. Corma and H. García, Complete Photocatalytic Reduction of CO₂ to Methane by H₂ under Solar Light Irradiation, J. Am. Chem. Soc., 2014, 136, 6798–6801.
2. C. H. Lin, J. H. Chao, C. H. Liu, J. C. Chang and F. C. Wang, Effect of Calcination Temperature on the Structure of a Pt/TiO₂ Nanofiber and its Photocatalytic Activity in Generating H₂, Langmuir, 2008, 24, 9907–9915.
3. Y. H. Zhang, Z. R. Tang, X. Z. Fu and Y. J. Xu, TiO₂–Graphene Nanocomposites for Gas-Phase Photocatalytic Degradation of Volatile Aromatic Pollutant: Is TiO₂–Graphene Truly Different from Other TiO₂–Carbon Composite Materials, ACS Nano, 2010, 4, 7303–7314.
4. L. L. Ji, W. Chen, S. R. Zheng, Z. Y. Xu and D. Q. Zhu, Adsorption of sulfonamide antibiotics to multiwalled carbon nanotubes, Langmuir, 2009, 25, 11608–11613.
5. G. Buitron and I. Moreno-Andrade, Biodegradation kinetics of a mixture of phenols in a sequencing batch moving bed biofilm reactor under starvation and shock loads, J. Chem. Technol. Biotechnol., 2011, 186, 669–674.
6. L. Sun, C. Wang, M. Ji and X. Kong, Treatment of mixed chemical wastewater and the agglomeration mechanism via an internal electrolysis filter, Chem. Eng. J., 2013, 50–56, 215–216.
7. Z. Zhu, X. Tang, S. Kang, P. W. Huo, M. S. Song, W. D. Shi, Z. Y. Lu and Y. Y. Yan, Constructing of the magnetic photocatalytic nanoreactor MS@FCN for cascade catalytic degrading of tetracycline, J. Phys. Chem. C, 2016 DOI:10.1021/acs.jpcc.6b05642.
8. Y. S. Jun, J. Park, S. U. Lee, A. Thomas, W. H. Hong and G. D. Stucky, Three-Dimensional Macroscopic Assemblies of Low-Dimensional Carbon Nitrides for Enhanced Hydrogen Evolution, Angew. Chem., Int. Ed., 2013, 125, 11289–11293.
9. P. Niu, L. Zhang, G. Liu and H. Cheng, Graphene-Like Carbon Nitride Nanosheets for Improved Photocatalytic Activities, Adv. Funct. Mater., 2012, 22, 4763–4770.
10. Z. Lin and X. Wang, Nanostructure engineering and doping of conjugated carbon nitride semiconductors for hydrogen photosynthesis, Angew. Chem., 2013, 125, 1779–1782.
11. A. Srinivasan and M. Miyauchi, Chemically Stable WO₃ Based Thin-Film for Visible-Light Induced Oxidation and Superhydrophilicity, J. Phys. Chem. C, 2012, 116, 15421–15426.
12. T. Nakajima, T. Kitamura and T. Tsuchiya, Visible light photocatalytic activity enhancement for water purification in Cu(II)-grafted WO₃ thin films grown by photoreaction of nanoparticles, Appl. Catal., B, 2011, 47–53, 108–109.
13. F. G. Wang, C. D. Valentin and G. Pacchioni, Electronic and Structural Properties of WO₃: A Systematic Hybrid DFT Study, J. Phys. Chem. C, 2011, 115, 8345–8353.
14. S. Chen, Y. Hu, S. Meng and X. Fu, Study on the separation mechanisms of photogenerated electrons and holes for composite photocatalysts g-C₃N₄/WO₃, Appl. Catal., B, 2014, 150–151, 564–573.
15. H. Tada, T. Mitsui, T. Kiyonaga, T. Akita and K. Tanaka, All-solid-state Z-scheme in CdS–Au–TiO₂ three-component nanojunction system, Nat. Mater., 2006, 5, 782–786.
16. W. J. Chen, P. J. Tsai and Y. C. Chen, Functional Fe₃O₄/TiO₂ Core/Shell Magnetic Nanoparticles as Photo killing Agents for Pathogenic Bacteria, Small, 2008, 4, 485–491.
17. J. Xie, K. Chen, H. Y. Lee, C. H. Xu, A. R. Su, S. Peng, X. Chen and S. Sun, Ultra small c(RGDyK)-Coated Fe₃O₄ Nanoparticles and Their Specific Targeting to Integrin rv3-Rich Tumor Cells, J. Am. Chem. Soc., 2008, 130, 7542–7543.
18. L. Ren, S. Huang, W. Fan and T. Liu, One-Step Preparation of Hierarchical Superpara magnetic Iron Oxide/Graphene Composites Via Hydrothermal Method, Appl. Surf. Sci., 2011, 258, 1132–1138.
19. Z. Zhu, Z. Y. Lu and X. X. Zhao, Surface imprinting of a g-C₃N₄ photocatalyst for enhanced photocatalytic activity and selectivity towards photodegradation of 2-mercaptobenzothiazole, RSC Adv., 2015, 5, 40726–40736.
20. G. Quan, Y. Liao, L. Yin, J. Long, X. Wang and C. Xue, Template-free synthesis of porous graphitic carbon nitride microspheres for enhanced photocatalytic hydrogen generation with high stability, Appl. Catal., B, 2015, 165, 503–510.
21. T. Wang, W. Quan, D. Jiang, L. Chen, D. Li and S. Meng, Synthesis of redox-mediator-free direct Z-scheme AgI/WO₃ nanocomposite photocatalysts for the degradation of tetracycline with enhanced photocatalytic activity, Chem. Eng. J., 2016, 300, 280–290.
22. A. K. Chakraborty and M. A. Kebede, Preparation and characterization of WO₃/Bi₃O₄Cl nanocomposite and its photocatalytic behavior under visible light irradiation, React. Kinet., Mech. Catal., 2012, 106, 83–98.
23. Y. Q. Yang, Facile synthesis of ZnO/Ag nanocomposites with enhanced photocatalytic properties under visible light, Mater. Lett., 2016, 180, 97–100.
24. Y. Guo, S. Lin, X. Li and Y. Liu, Amino acids assisted hydrothermal synthesis of hierarchically structured ZnO with enhanced photocatalytic activities, Appl. Surf. Sci., 2016, 384, 83–91.
25. S. Cao, W. Yuan, Y. P. Fang, J. Shahjamali and M. M. Barber, Growth of CdS quantum dots on g-C₃N₄ nanosheets for highly efficient photocatalytic hydrogen generation under visible light irradiation, Int. J. Hydrogen Energy, 2013, 38, 1258–1266.
26. J. Xie, K. Chen, H. Y. Lee, C. H. Xu, A. R. Su, S. Peng, X. Chen and S. Sun, Ultrasmall c(RGDyK)-Coated Fe₃O₄ Nanoparticles and Their Specific Targeting to Integrin rv3-Rich Tumor Cells, J. Am. Chem. Soc., 2008, 130, 7542–7543.
27. J. Park, K. J. An and Y. S. Wang, Ultra-large-scale syntheses of monodisperse nanocrystals, Nat. Mater., 2004, 3, 891–895.
28. Z. Zhu, X. Tang, C. Ma, M. Song, N. Gao and Y. Wang, Fabrication of conductive and high-dispersed PPy@Ag/g-C₃N₄, composite photocatalysts for removing various pollutants in water, Appl. Surf. Sci., 2016, 387, 366–374.
29. J. Cao, B. D. Luo, H. L. Lin, B. Y. Xu and S. F. Chen, Thermodecomposition synthesis of WO₃/H₂WO₄ heterostructures with enhanced visible light photocatalytic properties, Appl. Catal., B, 2012, 111–112, 288–296.
30. C. Li, C. Gang, J. Sun, Y. Feng, J. Liu and H. Dong, Ultrathin nanoflakes constructed erythrocyte-like Bi₂WO₆ hierarchical architecture via, anionic self-regulation strategy for improving photocatalytic activity and gas-sensing property, Appl. Catal., B, 2015, 163, 415–423.
31. Z. Y. Lu, F. Chen, M. He, M. S. Song, Z. F. Ma, W. D. Shi, Y. S. Yan and J. Z. Lan, Microwave synthesis of a novel magnetic imprinted TiO₂ photocatalyst with excellent transparency for selective photodegradation of enrofloxacin hydrochloride residues solution, Chem. Eng. J., 2014, 249, 15–26.
32. F. Tian, Y. Zhang, J. Zhang and C. Pan, Raman Spectroscopy: A New Approach to Measure the Percentage of Anatase TiO₂ Exposed (001) Facets, J. Phys. Chem. C, 2012, 11, 7515–7519.
33. Z. A. Huang, Q. Sun, K. L. Lv, Z. H. Zhang, M. Li and B. Li, Effect of contact interface between TiO₂ and g-C₃N₄ on the photoreactivity of g-C₃N₄/TiO₂ photocatalyst (001) vs. (101) facets of TiO₂, Appl. Catal., B, 2015, 164, 420–427.
34. K. Dai, D. Li, L. Lu, Q. Liu, C. Liang and J. Lv, Plasmonic TiO₂/AgBr/Ag ternary composite nanosphere with heterojunction structure for advanced visible light photocatalyst, Appl. Surf. Sci., 2014, 314, 864–871.
35. S. Chen, Y. Hu, X. Jiang, S. Meng and X. Fu, Fabrication and characterization of novel Z-scheme photocatalyst WO₃/g-C₃N₄ with high efficient visible light photocatalytic activity, Mater. Chem. Phys., 2015, 149–150, 512–521.
36. Z. Zhu, Z. Y. Lu, D. D. Wang, X. Tang, Y. S. Yan, W. D. Shi and Y. S. Wang, Construction of high-dispersed Ag/Fe₃O₄/g-C₃N₄ photocatalyst by selective photo-deposition and improved photocatalytic activity, Appl. Catal., B, 2016, 182, 115–122.
37. W. Li, F. Chang and S. Dai, Fabrication of sulfur-doped g-C₃N₄/Au/CdS Z-scheme photocatalyst to improve the photocatalytic performance under visible light, Appl. Catal., B, 2015, 168–169, 465–471.
38. Z. Zhu, Z. Y. Lu, D. D. Wang, X. Tang, Y. S. Yan, W. D. Shi and Y. S. Wang, Construction of high-dispersed Ag/Fe₃O₄/g-C₃N₄ photocatalyst by selective photo-deposition and improved photocatalytic activity, Appl. Catal., B, 2016, 182, 115–122.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra24096h

---