Surface imprinting of a g-C₃N₄ photocatalyst for enhanced photocatalytic activity and selectivity towards photodegradation of 2-mercaptobenzothiazole

Abstract

In the present work, based on Fe₃O₄/g-C₃N₄ as the support, 2-mercaptobenzothiazole (MBT) as the template molecule, and pyrrole as the functional monomer, we synthesized magnetic conductive imprinted photocatalysts (MCIPs) that were significantly efficient and stable through a suspension polymerization method. Moreover, the MCIPs not only exhibited a higher photodegradation capacity for MBT under visible light irradiation, but also possessed good selectivity. The specific surface areas of the resultant MCIPs were approximately four times that of pure g-C₃N₄. Furthermore, the results indicated that h⁺ was the major reactive species in the photocatalytic reaction system. The photodegradation mechanism was also discussed by analyzing the mass spectrum (MS), and the results demonstrated that MBT was degraded to H₂O, CO₂ and other small molecules, step by step. In addition, after five cycles, the MCIPs still showed high photocatalytic efficiency, indicating that the as-prepared MCIPs had excellent photochemical stability and good prospects for application in water treatment.

---

1. Introduction

In recent years, much attention has been paid to environmental problems, including a number of organic pollutants in contaminated water, such as those that are highly toxic, persistent and malodorous.^1,2 In particular, mercaptan contaminants, which are common in waste water, emit a strong foul odor and are threatening public health.³ Inhalation of low concentrations of MBT (a kind of mercaptan) vapor can cause headache and nausea. High concentrations can cause fatal respiratory paralysis, and this serious problem indicates that selective removal^4–6 of MBT from waste water is necessary and important. Therefore, many traditional methods to remove pollutants from the aquatic environment have been used, such as filtering methods, adsorption and microbial degradation. Unfortunately, none of these methods can remove the pollutants completely. Nowadays, photocatalytic technology has received significant attention due to its attractive applications in environmental cleaning and energy conversion. Nevertheless, consistent properties of photocatalytic technology and photocatalyst materials cannot be identified and selective degradation of specific molecules is challenging.

In order to solve the problems above, many photocatalytic materials have attracted particular attention for the removal of organic pollutants from waste water.⁷ Currently, g-C₃N₄, as one of the most promising photocatalysts, has been the concern of many researchers,^8,9 due to its high stability in aqueous solution and easy preparation and due to it being a visible light responsive semiconductor with a moderate band gap (E_g = 2.7 eV).^10–14 Unfortunately, using pure g-C₃N₄ as a visible-light photocatalyst has three main drawbacks. (1) The difficult collection of these photocatalysts.¹⁵ (2) The poor selectivity of g-C₃N₄ for the removal of the target molecule from complicated pollutants. (3) Low visible-light photocatalytic efficiency. Thus far, a simple way to simultaneously address these issues has yet to be achieved.

In order to overcome the above basic disadvantages of g-C₃N₄, the following can be considered: (1) loading of Fe₃O₄ nanoparticles as a magnetic material on g-C₃N₄ nanosheets, due to their good magnetic properties for improving the recycling efficiency¹⁶ and their excellent conductivity, as high as 1.9 × 10⁶ S m⁻¹,^17,18 which leads to rapid transfer of electron–hole pairs. (2) For enhanced selectivity of g-C₃N₄, surface imprinting technology is used,¹⁹ which is the most attractive and widely used method for preparing imprinted polymers with the ability to recognize target molecules. Therefore, surface imprinting technology can modify the surface of g-C₃N₄ and improve its ability to selectively recognize MBT, then achieve the goal of photodegradation of MBT. (3) While using the imprinting technique and modifying g-C₃N₄ with special polymers²⁰ will improve the selectivity and also reduce the catalytic activity, the active site will be covered; therefore, there are special requirements for the formed imprinting layer. The imprinting layer should synergistically enhance the catalytic activity and selectivity. For this reason, the polymer is important for the synthesis imprinted layer. In particular, PPy has shown great optical properties, unique electrical properties, excellent stability, and photocatalysis properties.^21–23 Furthermore, PPy is also an efficient hole transporter and good electron donor with visible light excitation, which suggests that polypyrrole has the ability to oxidize pollutants. Some work has been carried out to investigate the photocatalytic activity of PPy-modified TiO₂ and a PPy/Bi₂WO₆ composite and found that the modification of PPy can effectively enhance the photoactivity.^24–26

To our knowledge, no method has been proposed to enhance the selectivity and catalytic activity of g-C₃N₄ towards the degradation of MBT until now. Therefore, using the surface imprinting technology, which is a high-concentration polymer preparation method with specific recognition of target molecules, our group applied an easy method to synthesize MCIPs for selective degradation of MBT. In the present work, the MCIPs were successfully prepared based on Fe₃O₄/g-C₃N₄ as the support, MBT as the template molecule, and pyrrole as the functional monomer. They were characterized using X-ray diffraction (XRD), Fourier-transform infrared (FT-IR) spectroscopy, transmission electron microscopy (TEM), high-resolution TEM (HRTEM), scanning electron microscopy (SEM), the Brunauer–Emmett–Teller (BET) method, thermogravimetric analysis (TGA), elemental analysis, a vibrating sample magnetometer (VSM), and photoluminescence (PL). In addition, a series of influencing factors were optimized (such as the amount of functional monomer and the polymerization time). Afterwards, the adsorption, photocatalytic degradation and selectivity were investigated. Finally, the mechanism for degradation of MBT over the MCIPs was also discussed.

2. Experimental section

2.1. Materials

All chemicals were used without further purification and distilled water was used in all experiments. Melamine (AR), iron(III) chloride hexahydrate (FeCl₃·6H₂O, AR), iron(II) chloride tetrahydrate (FeCl₂·4H₂O, AR), 2-mercaptobenzothiazole (MBT), and trimethylolpropanetrimethacrylate (TRIM, AR) were all supplied by Aladdin Chemistry Co., Ltd. Polyethylene glycol 4000 (PEG 4000, CP), trichloromethane (AR), NH₄OH solution (25.0%), pyrrole (CP), p-benzoquinone (BQ, AR), isopropanol (IPA, AR), triethanolamine (OA, AR), and ethanol (C₂H₅OH, AR) were all purchased from Sinopharm Chemical Reagent Co., Ltd. Azodiisobutyronitrile (AIBN, CP) was obtained from Shanghai No.4 Reagent & H.V. Chemical Co., Ltd.

2.2. Preparation of g-C₃N₄

The pure g-C₃N₄ sample was synthesized by heating melamine. Briefly, 10 g melamine was heated from room temperature to 500 °C with a heating rate of 4 °C min⁻¹ and kept at 500 °C for 2 h. Afterwards, the product was further heated to 520 °C with the heating rate of 4 °C min⁻¹ and kept at 520 °C for another 2 h. After cooling to room temperature, the g-C₃N₄ powder was obtained and ground for further use.

2.3. Preparation of Fe₃O₄/g-C₃N₄

The Fe₃O₄/g-C₃N₄nanocomposites were prepared using Santosh Kumar’s method, with minor changes.¹⁹ Firstly, g-C₃N₄ (2.2 g) was dispersed in 600 mL of ethanol–water (1 : 3) and subjected to ultrasonication (PCI Analytics, 12 mm probe, 33 Hz, 150 W) for 5 h at ambient temperature. Secondly, FeCl₃·6H₂O (2 mmol) and FeCl₂·4H₂O (1 mmol) were dissolved separately in 20 mL of double-distilled water and added to the suspension of g-C₃N₄, stirring at 80 °C for 30 min. Subsequently, 10 mL NH₄OH solution was quickly injected into the reaction mixture, and stirred for another 30 min. Finally, the Fe₃O₄/g-C₃N₄ was washed several times with water and ethanol, then dried under vacuum and collected with a magnet.

2.4. Preparation of MCIPs

1 g Fe₃O₄/g-C₃N₄ and 1 g PEG were added into a three-neck flask containing 100 mL deionized water, with stirring 30 min. Meanwhile, 0.3 mmol MBT (template molecule), 0.1 mL TRIM (cross-linker) and 0.3 mmol AIBN (initiator) were dissolved in 15 mL trichloromethane, which contained 6 mmol pyrrole, then sonicated for 10 min to dissolve completely. Afterward, the mixed solution was added to the above solution. The reaction was stirred for 24 h under nitrogen at 75 °C. Finally, after the solid product was washed by soxhlet extraction with methanol and glacial acetic acid and dried using a vacuum drier, the MCIPs were obtained. Non-imprinted catalysts (MCNIPs) were prepared using the same procedure as for the MCIPs, but in the absence of MBT. In order to clarify the preparative procedure of the MCIPs composites, the synthesis route and preparation process are described in Scheme 1.

Scheme 1 Schematic representation of MCIPs.

2.5. Adsorption experiments

In order to compare the adsorption capability of MCIPs and g-C₃N₄ for MBT, the adsorption capacity was determined as follows. 0.1 g photocatalyst was added to the catalytic reactor containing 100 mL 10 mg L⁻¹ aqueous MBT solution. 8 mL aliquots of solution were sampled using an injector at 10 min intervals, after stirring for 60 min in the dark at 30 °C. The samples were filtered and the concentration of MBT was measured using a UV-vis spectrophotometer.

2.6. Photocatalytic activity and selectivity experiments

In order to study the photocatalytic activity of g-C₃N₄, MCIPs and MCNIPs towards the degradation of MBT, the reaction should reach the adsorption–desorption equilibrium. The adsorption procedure was consistent with Section 2.5. After the reaction reached the adsorption–desorption equilibrium, a xenon lamp (300 W) and aeration were both turned on, and the reaction maintained for 1 h. Next, the absorbance of the sample was measured after magnetic separation using an ultraviolet spectrophotometer. In order to investigate the selectivity, another contaminant (danofloxacin mesylate) was chosen, and the experimental steps were the same as above.

2.7. Circulation experiment

The reusability of the MCIPs is very important from a practical application point of view. Therefore, five groups of successive MBT degradation experiments were carried out. In the first step, the photodegradation experiment was carried out according to Section 2.6. Subsequently, the sample was collected using a magnet and washed with ethanol. After drying, the sample was used in the second cycling experiment. Subsequently, the remaining four cycling steps were the same as the first step.

2.8. Characterization

In order to analyze the performance of the catalyst, lots of characterization has been carried out as follows: X-ray diffraction (XRD) was used to characterize the crystal structure, and the patterns of the photocatalyst were obtained with a D/max-RA X-ray diffractometer (Rigaku, Japan) with Ni-filtered Cu Kα radiation (40 kV, 200 mA) at 5–80°, with a scanning step of 0.02°/0.2 s. Both the scanning electron microscopy (SEM) and energy dispersive X-ray spectroscopy (EDS) were performed using a JSM-7001F scanning electron microscope (JEOL Ltd., Japan). The transmission electron microscopy (TEM) images were obtained using a JEM-2100 transmission electron microscope (JEOL, Japan). Fourier transform infrared (FT-IR) spectra were recorded on a Nicolet Nexus 470 FT-IR (America Thermo-Electricity Company) with a 2.0 cm⁻¹ resolution in the range of 400–4000 cm⁻¹, using KBr pellets. The magnetic characterisation was carried out using a vibrating sample magnetometer (VSM) (HH-15, Jiangsu University). Specific surface areas were obtained using the Brunauer–Emmett–Teller (BET) method and examined with a NOVA 2000e analytical system (Quantachrome Co., USA). Thermogravimetric analysis (TGA) was carried out using a thermal analyzer (NETZSCHGeratebau GmbH, Germany) in air, from room temperature up to 800 °C with a heating rate of 10 °C min⁻¹. The elemental content of each sample was analyzed using an elemental analyzer (EA-1112A, Thermos, Italy). The photoluminescence (PL) emission spectra of the samples were obtained using a fluorescence spectrophotometer (Cary Eclipse Spectrophotometer, VARIAN, USA) using a xenon lamp as the excitation source at room temperature. The degradation mechanism of an aqueous tetracycline solution was determined using mass spectrometry (MS) with a Thermo LXQ spectrometer.

3. Results and discussion

3.1. Characterization of the as-prepared samples

3.1.1. XRD analysis. Fig. 1 shows the XRD patterns of pure g-C₃N₄, Fe₃O₄/g-C₃N₄ and MCIPs. The characteristic peaks of g-C₃N₄ were seen at 2θ = 27.4° and 13.1°, corresponding to the PDF #50-1250 data file,²⁷ which also appeared in Fe₃O₄/g-C₃N₄ and the MCIPs. This result indicated that the structure of g-C₃N₄ did not change by adding Fe₃O₄ and PPy, or other materials. Moreover, six distinct typical peaks (2θ = 30.2°, 35.5°, 43.2°, 53.4, 57.3° and 62.6°) were observed in Fig. 1b and c, which were attributed to the magnetite of the Fe₃O₄ nanoparticles.¹⁶ However, the diffraction peaks of PPy were not detected in Fig. 1c. Therefore, it was difficult to judge from XRD pattern whether the MCIPs had been successfully prepared, but this was proven by the FT-IR, SEM, TEM and elemental analysis.

Fig. 1 XRD patterns of the samples: g-C₃N₄ (a), Fe₃O₄/g-C₃N₄ (b) and MCIPs (c).

3.1.2. FT-IR. The FT-IR spectra of pure g-C₃N₄, Fe₃O₄/g-C₃N₄ and MCIPs are shown in Fig. 2. A series of peaks from 1650–1200 cm⁻¹ (1251, 1325, 1419, 1571, and 1639 cm⁻¹) corresponded to the typical stretching modes of CN heterocycles,^28,29 such as C–N and C=N stretching.^30,31 The peak near 808 cm⁻¹ was attributed to the typical bending vibration of s-triazine units.³² The broad peaks from 3400 cm⁻¹ to 2800 cm⁻¹ in all three materials were the stretching vibrational modes of primary (=NH) and secondary (–NH) amines.³³ In the spectra of Fe₃O₄/g-C₃N₄ and the MCIPs, the peak at 550–650 cm⁻¹ was attributed to the Fe–O.^34,35 The results showed that Fe₃O₄ had been successfully loaded on the surface of g-C₃N₄. With regard to the MCIPs, as reported in the literature,³⁶ PPy exhibits characteristic absorption bands at 1575, 1468, 1298, 784 and 678 cm⁻¹. They were assigned to pyrrole ring stretching, C–H in-plane deformation and C–H outer-bending vibrations. Some characteristic absorption bands also appeared in the FT-IR spectrum of the MCIPs; however, these characteristic bands became stronger (1575 cm⁻¹) or unclear (784 cm⁻¹, 678 cm⁻¹) in the spectrum, which might be due to the overlap of the FT-IR spectrum of g-C₃N₄, the low concentration of PPy in the composites and the possible interplay of PPy with g-C₃N₄. In brief, the FT-IR spectra fully proved the existence of g-C₃N₄, Fe₃O₄ and PPy.

Fig. 2 The FT-IR spectra of g-C₃N₄ (a), Fe₃O₄/g-C₃N₄ (b) and MCIPs (c).

3.1.3. SEM analysis. The SEM images of g-C₃N₄, Fe₃O₄/g-C₃N₄ and the MCIPs are shown in Fig. 3. It can be seen that the surface of g-C₃N₄ was relatively smooth, but after coating with Fe₃O₄, the surface of g-C₃N₄ became uneven and bulged. Moreover, the material on the surface of g-C₃N₄ can be seen clearly. The above results indicated that Fe₃O₄ was successfully loaded onto the surface of g-C₃N₄. In addition, the image of the MCIPs was different to that of Fe₃O₄/g-C₃N₄. A significant change to the surface of the MCIPs could be clearly seen by comparison with Fe₃O₄/g-C₃N₄. This must caused by the imprinted layer covering. The EDS spectra of the different samples are also presented in Fig. 3. Compared with Fe₃O₄/g-C₃N₄, the C content of the MCIPs increased and the Fe content decreased, which was due to the introduction of the surface imprinted layer. In addition, compared with the BET surface areas of the pure g-C₃N₄ and MCIPs shown in Fig. S1,† the BET surface area of the pristine g-C₃N₄ was about 5.8 m² g⁻¹. After coating with PPy, the BET surface area of the MCIPs increased to 20.2 m² g⁻¹. The change in the BET surface area might be related to a large number of Fe₃O₄ nanoparticles and imprinted cavities. Moreover, the increased BET surface area was consistent with results of SEM, and further proved the perfection of the imprinting.

Fig. 3 SEM and EDS images of g-C₃N₄ (a and a₁), Fe₃O₄/g-C₃N₄ (b and b₁) and MCIPs (c and c₁).

3.1.4. TEM analysis. Fig. 4 shows the typical TEM images of pure g-C₃N₄, Fe₃O₄/g-C₃N₄ and MCIPs, and high-resolution TEM (HRTEM) of Fe₃O₄/g-C₃N₄. It can be seen that the big sheet structure of g-C₃N₄ was covered with small particles of Fe₃O₄, and the Fe₃O₄ nanoparticles have sizes ranging from 5 nm to 10 nm and were not agglomerated (Fig. 4b). Fig. 4d shows the HRTEM image of Fe₃O₄/g-C₃N₄. It is clear that the lattice plane has spacing of 0.25 nm, which confirmed that the g-C₃N₄ sheet could serve as a support to bind with Fe₃O₄ nanoparticles. In contrast, in the image of the MCIPs (Fig. 4c), it seemed that a thin layer aggregated on the surface of Fe₃O₄/g-C₃N₄ and the spherical Fe₃O₄ could not be seen clearly. The change between these two images must caused by the PPy covering. This result was in accordance with the SEM image. Therefore, the TEM image provided further evidence that PPy was successfully assembled onto the surface of Fe₃O₄/g-C₃N₄.

Fig. 4 TEM images of g-C₃N₄ (a), Fe₃O₄/g-C₃N₄ (b) and MCIPs (c); HRTEM of Fe₃O₄/g-C₃N₄ (d).

3.1.5. Elemental analysis. The elemental content of the different samples is shown in Table 1. The C/N molar ratio of g-C₃N₄ was 0.734, which was close to the theoretical value of g-C₃N₄. This result proved that g-C₃N₄ was successfully prepared by calcination of melamine.^37–39 In addition, comparing the pure g-C₃N₄ and the MCIPs, the contents of N, C, and H decreased from 59.82546%, 37.13942% and 2.03588% to 53.97383%, 35.88362% and 1.31395%, respectively. Through calculation, the content of Fe₃O₄ was nearly 10%, which was consistent with the initial dosage, meaning that Fe₃O₄ was also introduced onto the surface of g-C₃N₄. Furthermore, after Fe₃O₄/g-C₃N₄ was modified with the imprinting layer in the MCIPs, the contents of N, C and H were significantly increased. This result indicated that the surface of Fe₃O₄/g-C₃N₄ was successfully imprinted.

Table 1 Results of the elemental composition of different samples from elemental analysis

Samples	N (%)	C (%)	H (%)
g-C3N4	59.82546	37.13942	2.03588
Fe3O4/g-C3N4	53.97383	35.88362	1.31395
MCIPs	55.09838	37.14768	1.92486

---

3.1.6. TG analysis. The TG curves of g-C₃N₄ and the MCIPs are shown in Fig. 5. The g-C₃N₄ was stable from the room temperature up to 505 °C; however, the weight decreased rapidly in the temperature range from 505 °C to 720 °C, indicating that the g-C₃N₄ underwent combustion in the air and almost all of the g-C₃N₄ generated gas when the temperature reached 720 °C.⁴⁰ According to the curve of Fig. 5b, a slight weight loss of the MCIPs (5.4%) could be seen from room temperature up to 200 °C, which might due to the evaporation of water. Furthermore, with the temperature increasing from 200 °C to 505 °C, the weight loss of the MCIPs (5.9%) must be attributed to the loss of organic residues from the imprinted layer on the surface of Fe₃O₄/g-C₃N₄.³ The weight loss between 505 °C and 720 °C was assigned to the decomposition of the g-C₃N₄. The remaining mass (9.7%) was attributed to Fe₃O₄, because the melting point of Fe₃O₄ was much higher than the experimental temperature, and it also contained remnants of small amounts of ash. The analysis completely proved that both the MCIPs and g-C₃N₄ have good thermal stability.

Fig. 5 The thermogravimetric analysis of g-C₃N₄ (a) and MCIPs (b).

3.1.7. VSM analysis. The magnetic properties were measured using a vibrating sample magnetometer (VSM) system at room temperature. The magnetic hysteresis loops of the sample Fe₃O₄/g-C₃N₄ and the MCIPs are shown in Fig. 6. The magnetization saturation (Ms) values of Fe₃O₄/g-C₃N₄ and the MCIPs were 13.87 emu g⁻¹ and 3.64 emu g⁻¹, respectively. This weakened magnetism of the MCIPs might be caused by the effect of the incremental imprinted layer, as the sample content of iron oxide in unit weight could be reduced. However, from the photograph (inset) it could be clearly observed that the MCIPs were easily separated using a magnet, indicating that the MCIPs still possessed excellent magnetic separation performance.

Fig. 6 Hysteresis loops of Fe₃O₄/g-C₃N₄ (a) and MCIPs (b). Inset images are of MCIPs suspended in water (left) and MCIPs attracted by a magnetic field (right).

3.1.8. PL analysis. It is well known that the PL spectrum originates from the radiative recombination of the free electron–hole pairs and helps to clarify the migration and recombination processes of electron–hole pairs in a semiconductor photocatalyst. It can be clearly seen in Fig. 7 that the PL intensity of the MCIPs decreased significantly compared to pure g-C₃N₄. The weaker intensity of the peak represented the lower recombination rate of photoinduced electron–hole pairs in the MCIPs. This could be attributed to the easy transfer of carriers between PPy, Fe₃O₄ and g-C₃N₄. Due to the high conductivity of PPy and Fe₃O₄, as well as the synergistic effect of PPy, Fe₃O₄ and g-C₃N₄, the recombination rate of electron–hole pairs is reduced. Although the PL intensity of the MCIPs was slightly higher than the MCNIPs, the MCIP has higher activity for MBT. This also indicates that the imprinted cavities and recognition capability played an essential role in enhancing the photocatalytic performance and selective photodegradation.

Fig. 7 PL spectra of pure g-C₃N₄ (a), MCIPs (b) and MCNIPs (c).

3.2. Visible-light photocatalytic activity of MCIPs

3.2.1. Adsorption experiments. The adsorption experiments were carried out in accordance with Section 2.5. The adsorption curves of the MCIPs and g-C₃N₄ are shown in Fig. 8. After about 40 min, the solution almost reached the adsorption–desorption equilibrium, with the amounts of adsorption by the MCIPs being much greater than with g-C₃N₄. The different adsorption abilities could be attributed to the large numbers of imprinting cavities in the imprinted polymes surface, such that the increased BET surface area enhanced the adsorption ability. Another important reason is that the cavities are consistent with the target structure, so that the MCIP exhibits good recognition and adsorbs more target pollutant.

Fig. 8 Adsorption results of MCIPs (a) and g-C₃N₄ (b) with MBT.

3.2.2. The investigation of different polymerization times. Fig. 9 presents the time course of MBT concentrations during the process of photocatalysis. As polymerization time was a very important consideration, a series of polymerization times were used to synthesise different imprinted photocatalysts. The different polymerization times revealed the effect on catalysis. With an appropriate polymerization time, PPy was embedded well in the surface imprinted layer. The effect of polymerization time was rather complicated, as if the time was shorter, pyrrole could not polymerize well. In contrast, a long polymerization time might increase the degree of bonding of the polymer, and the strong bonding provides an obstacle to the removal of the template molecules. Therefore, in the subsequent experiments, 24 h was chosen as the optimum polymerization time to prepare the MCIPs. More importantly, the different polymerization times will influence the degree of polymerization of PPy and the different degrees of polymerization will affect the HOMO and LUMO levels of PPy. Therefore, the appropriate HOMO and LUMO levels of PPy must be found to match the band gap of g-C₃N₄ and achieve a synergistic catalytic effect, so we proved by experiments that 24 h was the most suitable time for polymerization.⁴¹

Fig. 9 Effect of polymerization time on degradation of MBT: (a) 6 h, (b) 12 h, (c) 24 h, (d) 36 h, (e) 48 h.

3.2.3. The investigation of different amounts of pyrrole. The dosage of pyrrole was another important factor in the synthesis of the imprinted photocatalysts. As displayed in Fig. 10, the photocatalyst exhibited the highest photocatalytic activity when the dosage of pyrrole was 6 mmol. On the one hand, with a smaller dosage of pyrrole, PPy could not polymerize on the surface of Fe₃O₄/g-C₃N₄ completely and as a result the transmitting ability of the photogenerated electrons was weakened. Meanwhile, the imprinted cavities were also decreased and reduced the recognition ability for MBT, further resulting in lower photocatalytic activity. On the other hand, as the amount of PPy increased, the imprinted layer became thicker, since superfluous PPy was generated. The thicker imprinted layer will affect the absorption of light, limiting charge transfer and parts of the g-C₃N₄ semiconductor surface would not be excited under visible light irradiation, thereby leading to lower photocatalytic efficiency. Consequently, in the subsequent experiments, the dose of pyrrole was 6 mmol.

Fig. 10 Effect of the dose of pyrrole on degradation of MBT: (a) 0.075 mmol, (b) 3 mmol, (c) 6 mmol, (d) 7.5 mmol, (e) 9 mmol.

3.2.4. Selectivity experiment. The purpose of the present work was the selective recognition of a template molecule. In order to test the advantage of the MCIPs, MCNIPs and pure g-C₃N₄ for degradation experiments, the samples were prepared under the same conditions. Fig. 11(A) shows the photocatalytic effect of the MCIPs, MCNIPs and g-C₃N₄ on the degradation of MBT for 60 min under visible light. Furthermore, we discuss why the degradation effect of the MCIPs was the highest and g-C₃N₄ was the lowest. For g-C₃N₄, the low degradation must be caused by the lack of a conductive imprinting layer, resulting in no surface imprinted cavities and loss of the recognition capability for MBT, so that the degradation of MBT was correspondingly decreased. However, despite the fact that the MCNIPs had a conductive imprinting layer, there were no imprinted cavities, resulting in the degradation effect being lower than for MCIPs. The MCIPs exhibited excellent activity toward the template pollutants in this work, which was attributed to the conductive imprinting layer and surface imprinting cavities formed by the template molecules. This fully demonstrated the superiority of the MCIPs.

Fig. 11 The degradation of MBT (A) and danofloxacin mesylate (B) with MCIPs (a and d), MCNIPs (b and e), and g-C₃N₄ (c and f).

In order to further highlight the advantage of the selectivity, another experiment on the degradation of danofloxacin mesylate was carried out, and the concentration curves are shown in Fig. 11 (B). The catalysts g-C₃N₄, MCIPs and MNCIPs were the same as above. The photodegradation effect of the MCIPs on the degradation of danofloxacin mesylate decreased rapidly. This was due to the different structure of danofloxacin mesylate, which is larger than the imprinted molecule, and the absence of binding sites, leading to lower selectivity. It was obvious that the photodegradation effect of the MCIPs and MCNIPs was similar and higher than pure g-C₃N₄ for the degradation of danofloxacin mesylate. This also showed that the conductive imprinted layer plays a critical role, which was to also promote the transmission and separation of electrons and holes. Thus further demonstrated the importance of the imprinting cavities on degradation of target pollutants.

3.2.5. Photocatalytic stability. For the MCIPs, besides the demand for photocatalytic activity and selectivity, the photocatalytic stability is crucially important for practical applications. The recycling capability of the MCIPs was verified by carrying out a five-run test of photocatalytic degradation of MBT. Fig. 12 reveals that the degradation rate showed no obvious decrease after the five-run test, which indicated that the MCIPs had excellent photocatalytic stability.

Fig. 12 Recyclability of MCIPs for degradation of MBT under visible-light irradiation.

3.3. Photocatalytic mechanism

3.3.1. Active species trapping and ˙O₂⁻, h⁺ and ˙OH quantification experiments. It is generally known that ˙O₂⁻, h⁺ and ˙OH are the major reactive species for photodegradation. To detect the active species of the MCIPs during photocatalytic reactivity, 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⁻)^42–46 were added to the reaction solution during the degradation experiment. The detection process was as the same as the experimental photodegradation process. The results are shown in Fig. 13. The degradation efficiency of MBT was 85% when no scavenger was added. However, when adding OA to the solution, the photodegradation rate of MBT decreased to 5%, revealing that the photogenerated holes played a central role in degradation. When adding IPA to the reactions, the degradation rate of MBT was hardly decreased, indicating that ˙OH was not the main active species for degradation. Also, the effect of ˙O₂⁻ was non-negligible, as could be seen from the graph which shows that adding BQ also inhibited nearly half of the degradation rate. Therefore, the degree of influence was h⁺ > ˙O₂⁻ > ˙OH for the MCIPs.

Fig. 13 Effects of a series of scavengers on the photocatalytic efficiency of MCIPs.

The photodegradation rate (η) was calculated by using the following formula, where C₀ was the initial concentration and C the concentration of the reaction solution:

3.3.2. Photocatalytic mechanisms of the MCIPs. Fig. 14 shows the photocatalytic degradation process of the degradation of MBT with the MCIPs. When the as-prepared composites were irradiated under visible-light, both g-C₃N₄ and PPy could easily absorb visible light. As PPy is a well known electronic conductor,⁴⁷ the polymer could absorb photons and promote an electron from the ground state into an excited state under visible-light irradiation. The lowest unoccupied molecular orbital (LUMO) levels and the highest occupied molecular orbital (HOMO) levels of PPy generate electrons and holes, respectively.^26,48 Meanwhile, electrons could be excited to the conduction band and leave holes in the valence band of g-C₃N₄. Subsequently, the photogenerated electrons might be injected into the conduction band of the inner g-C₃N₄. Furthermore, the remaining holes of g-C₃N₄ were transferred to the surface of the thick PPy layer. Therefore, the recombination rate of electron–hole pairs in the composite system was reduced significantly. As there were a large number of O₂ and H₂O molecules adsorbed on the PPy surface, these could react with e⁻, then generate ˙O₂⁻ and H₂O₂, and react with MBT, the same as the h⁺, then produce CO₂, H₂O and other small molecules. The possible photodegradation mechanism and charge transfer processes are displayed as follows:

g-C₃N₄ + hν → g-C₃N₄(h⁺ + e⁻) (2)

PPy + hν → PPy(h⁺ + e⁻) (3)

PPy(e⁻) → g-C₃N₄(h⁺) (4)

g-C₃N₄(h⁺) → PPy(e⁻) (5)

H₂O → H⁺ + OH⁻ (6)

2g-C₃N₄(e⁻) + 2H⁺ + O₂ → H₂O₂ (7)

H₂O₂ + e⁻ →˙OH + OH⁻ (8)

g-C₃N₄(e⁻) + O₂ → ˙O⁻₂ (9)

˙O₂⁻ + 2H⁻ + 2e⁻ → 2˙OH (10)

PPy(h⁺) + MBT → other small molecules (majority) (11)

˙O₂⁻ + MBT → other small molecules (12)

H₂O₂ + MBT → other small molecules (13)

˙OH + MBT → other small molecules (14)

Fig. 14 Schematic illustration of the degradation of MBT with MCIPs under visible light irradiation.

3.3.3. MS analysis. Mass spectrometry (MS) is widely used to discuss the possible mechanism of photodegradation. As shown in Fig. S2 (in the ESI†), it could be found that as time went on, the characteristic peak of MBT (m/z = 168) became smaller and smaller. This illustrated that MBT had decomposed into other intermediate degradation products, and then these substances were gradually decomposed to CO₂, H₂O and other small molecules. According to MS, the mechanism was speculated to be as follows:

In the MS, the peaks of m/z = 168, m/z = 139, m/z = 108 and m/z = 125 were analyzed in the processes of photodegradation. Firstly, the target was attracted by the active substance and lost a –SH group, generating (A) (Fig. 15). The formation of (B) might due to the fact that (A) quickly bound with H in the solution, which was why the peak of m/z = 168 became smaller. As N=C was not stable, the N–C double bond of (B) was broken. Thereby, (C) was generated easily under the irradiation of light. Subsequently, the –S-CH₃ group in (C) might fracture and form ˙CH₃ and ˙SH. Then the two groups of ˙CH₃ and ˙SH undergo rearrangement, along with the reaction, and a small amount of (D) and (E) might be generated. Finally, (C), (D) and (E) were further degraded to CO₂, H₂O and other small molecules.

Fig. 15 The photodegradation mechanism of MBT over MCIPs.

4. Conclusions

In summary, a novel magnetic conductive imprinted g-C₃N₄ has been fabricated successfully by suspension polymerization with an easy preparation method, coupled with low cost and stability. The imprinted catalyst, through a series of investigations on influencing factors, showed optimized photocatalytic activity and high adsorption capacity for the target pollutants as well as outstanding recognition behavior toward MBT when the polymerization time was 24 h and the dose of pyrrole was 6 mmol. The photodegradation rate for MBT with the MCIPs reached nearly 85% in 60 min under visible light irradiation, and the degradation rate of the MCIPs improved nearly 40% compared to pure g-C₃N₄. More importantly, the intermediate products were analyzed by MS while proving that there was no harmful substance generated in the degradative process. Moreover, the PL improved the separation of photogenerated electron–hole pairs and gave a higher efficiency of cavity recognition.

Conflict of interest

The authors declare no competing financial interest.

Acknowledgements

We gratefully acknowledge the Natural Science Foundation of China (no. 21306068 and 21407059), the Natural Science Foundation of Jiangsu province (nos BK20130477, BK20130480, BK20130487 and BK20140532), the financial support of the China Postdoctoral Science Foundation (no. 2014T70486 and 2011M500863) and the Innovation Programs Foundation of Jiangsu province (nos CXZZ13_0693).

References

1. S. C. Yan, S. B. Lv, Z. S. Li and Z. G. Zou, Organic–inorganic composite photocatalyst of g-C₃N₄ and TaON with improved visible light photocatalytic activities, Dalton Trans., 2010, 39, 1488–1491.
2. M. N. Chong and B. Jin, Photocatalytic treatment of high concentration carbamazepine in synthetic hospital wastewater, J. Hazard. Mater., 2012, 199, 135–142.
3. Y. Y. Luo, Z. Y. Lu, Y. H. Jiang, D. D. Wang, L. L. Yang, P. W. Huo, Z. L. Da, X. L. Bai, X. L. Xie and P. Y. Yang, Selective photodegradation of 1-methylimidazole-2-thiol by the magnetic and dual conductive imprinted photocatalysts based on TiO₂/Fe₃O₄/MWCNTs, Chem. Eng. J., 2014, 240, 244–252.
4. X. Shen, L. H. Zhu, N. Wang, T. Zhang and H. Tang, Selective photocatalytic degradation of nitrobenzene facilitated by molecular imprinting with a transition state analog, Catal. Today, 2014, 225, 164–170.
5. Z. Y. Lu, P. W. Huo, Y. Y. Luo, X. L. Liu, D. Wu, X. Gao, C. X. Li and Y. S. Yan, Performance of molecularly imprinted photocatalysts based on fly-ash cenospheres for selective photodegradation of single and ternary antibiotics solution, J. Mol. Catal. A: Chem., 2013, 378, 91–98.
6. X. L. Liu, P. Lv, G. X. Yao, C. C. Ma, P. W. Huo and Y. S. Yan, Microwave-assisted synthesis of selective degradation photocatalyst by surface molecular imprinting method for the degradation of tetracycline onto Cl–TiO₂, Chem. Eng. J., 2013, 217, 398–406.
7. Z. Y. Lu, W. C. Zhou, P. W. Huo, Y. Y. Luo, M. He, J. M. Pan, C. X. Li and Y. S. Yan, Performance of a novel TiO₂ photocatalyst based on the magnetic floating fly-ash cenospheres for the purpose of treating waste by waste, Chem. Eng. J., 2013, 225, 34–42.
8. L. M. Sun, X. Zhao, C.-J. Jia, Y. X. Zhou, X. F. Cheng, P. Li, L. Liu and W. L. Fan, Enhanced visible-light photocatalytic activity of g-C₃N₄–ZnWO₄ by fabricating a heterojunction: investigation based on experimental and theoretical studies, J. Mater. Chem., 2012, 22, 23428–23438.
9. S. F. Chen, Y. F. Hu, S. G. Meng and X. L. Fu, Study on the separation mechanisms of photogenerated electrons and holes for composite photocatalysts g-C₃N₄–WO₃, Appl. Catal., B, 2014, 564–573.
10. Q. J. Xiang, J. G. Yu and M. Jaroniec, Preparation and enhanced visible-light photocatalytic H₂-production activity of graphene/C₃N₄ composites, J. Phys. Chem. C, 2011, 115, 7355–7363.
11. G. H. Dong, K. Zhao and L. Z. Zhang, Carbon self-doping induced high electronic conductivity and photoreactivity of g-C₃N₄, Chem. Commun., 2012, 48, 6178–6180.
12. Y. J. Wang, X. J. Bai, C. S. Pan, J. He and Y. F. Zhu, Enhancement of photocatalytic activity of Bi₂WO₆ hybridized with graphite-like C₃N₄, J. Mater. Chem., 2012, 22, 11568–11573.
13. G. Z. Liao, S. Chen, X. Quan, H. T. Yu and H. M. Zhao, Graphene oxide modified g-C₃N₄ hybrid with enhanced photocatalytic capability under visible light irradiation, J. Mater. Chem., 2012, 22, 2721–2726.
14. J. S. Zhang, J. H. Sun, K. Maeda, K. Domen, P. Liu, M. Antonietti, X. Z. Fu and X. C. Wang, Sulfur-mediated synthesis of carbon nitride: band-gap engineering and improved functions for photocatalysis, Energy Environ. Sci., 2011, 4, 675–678.
15. S. Ye, L.-G. Qiu, Y.-P. Yuan, Y.-J. Zhu, J. Xia and J.-F. Zhu, Facile fabrication of magnetically separablegraphitic carbon nitride photocatalysts with enhanced photocatalytic activity under visible light, J. Mater. Chem. A, 2013, 1, 3008–3015.
16. Z. Y. Lu, F. Chen, M. He, M. S. Song, Z. F. Ma, W. D. Shi, Y. S. Yan, J. Z. Lan, F. Li and P. Xiao, 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.
17. G. C. Xi, B. Yue, J. Cao and J. H. Ye, Fe₃O₄/WO₃ Hierarchical Core–Shell Structure: High-Performance and Recyclable Visible-Light Photocatalysis, Chem.–Eur. J., 2011, 17, 5145–5154.
18. S. Kumar, B. Kumar, A. Baruah and V. Shanker, Synthesis of Magnetically Separable and Recyclable g-C₃N₄–Fe₃O₄ Hybrid Nanocomposites with Enhanced Photocatalytic Performance under Visible-Light Irradiation, J. Phys. Chem. C, 2013, 117, 26135–26143.
19. Z. Y. Lu, Y. Luo, M. He, P. W. Huo, T. T. Chen, W. D. Shi, Y. S. Yan, J. M. Pan, Z. F. Ma and S. Y. Yang, Preparation and performance of a novel magnetic conductive imprinted photocatalyst for selective photodegradation of antibiotic solution, RSC. Adv., 2013, 3, 18373–18382.
20. L. Ge, C. C. Han and J. Liu, In situ synthesis and enhanced visible light photocatalytic activities of novel PANI–g-C₃N₄ composite photocatalysts, J. Mater. Chem., 2012, 22, 11843–11850.
21. J. Wang and X. Ni, Photoresponsive polypyrrole–TiO₂ nanoparticles films fabricated by a novel surface initiated polymerization, Solid State Commun., 2008, 146, 239–244.
22. P. Wang, T. Xie, L. Peng, H. Li and T. Wu, Water-assisted synthesis of anatase TiO₂ nanocrystals: mechanism and sensing properties to oxygen at room temperature, J. Phys. Chem. C, 2008, 112, 6648–6652.
23. H. C. Liang and X. Z. Li, Visible-induced photocatalytic reactivity of polymer-sensitized titania nanotube films, Appl. Catal., B, 2009, 86, 8–17.
24. D. Chowdhury, A. Paul and A. Chattopadhyay, Photocatalytic polypyrrole–TiO₂-nanoparticles composite thin film generated at the air–water interface, Langmuir, 2005, 21, 4123–4128.
25. C. R. Zhang, Q. L. Li and J. Q. Li, Synthesis and characterization of polypyrrole/TiO₂ composite by in situ polymerization method, Synth. Met., 2008, 160, 1699–1703.
26. Z. J. Zhang, W. Z. Wang and E. Gao, Polypyrrole/Bi₂WO₆ composite with high charge separation efficiency and enhanced photocatalytic activity, J. Mater. Sci., 2014, 49, 7325–7332.
27. X. J. Bai, R. L. Zong, C. X. Li, D. Liu, Y. F. Liu and Y. F. Zhu, Enhancement of visible photocatalytic activity via Ag@C3N4 core–shell plasmonic composite, Appl. Catal., B, 2014, 147, 82–91.
28. X. F. Li, J. Zhang, L. H. Shen, Y. M. Ma, W. W. Lei, Q. L. Cui and G. G. Zou, Preparation and characterization of graphitic carbon nitride through pyrolysis of melamine, Appl. Phys. A: Mater. Sci. Process., 2009, 94, 387–392.
29. Y. C. Zhao, Z. Liu, W. G. Chu, L. Song, Z. X. Zhang, D. L. Yu, Y. J. Tian, S. Xie and L. F. Sun, Large-Scale Synthesis of Nitrogen-Rich Carbon Nitride Microfibers by Using Graphitic Carbon Nitride as Precursor, Adv. Mater., 2008, 20, 1777–1778.
30. Y. Bu, Z. Chen, J. Yu and W. Li, A novel application of g-C₃N₄ thin film in photoelectrochemical anticorrosion, Electrochim. Acta, 2013, 88, 294–300.
31. Y. J. Wang, X. J. Bai, C. S. Pan, J. He and Y. F. Zhu, Enhancement of photocatalytic activity of Bi₂WO₆ hybridized with graphite-like C₃ N₄, J. Mater. Chem., 2012, 22, 11568–11573.
32. Y. S. Jun, E. Z. Lee, X. C. Wang, W. H. Hong, G. D. Stucky and A. Thomas, From Melamine-Cyanuric Acid Supramolecular Aggregates to Carbon Nitride Hollow Spheres, Adv. Funct. Mater., 2013, 23, 3661–3667.
33. X. H. Li, X. C. Wang and M. Antonietti, Solvent-Free and Metal-Free Oxidation of Toluene Using O₂ and g-C₃N₄ with Nanopores: Nanostructure Boosts the Catalytic Selectivity, ACS Catal., 2012, 2, 2082–2086.
34. J. Xie, K. Chen, H. Y. Lee, C. Xu, A. R. Hsu, S. Peng, X. Chen and S. Sun, Ultrasmallc (RGDyK)-Coated Fe₃O₄ Nanoparticles and Their Specific Targeting to Integrin rv3-Rich Tumor Cells, J. Am. Chem. Soc., 2008, 130, 7542–7543.
35. J. Park, K. J. An and Y. S. Hwang, Ultra-large-scale syntheses of monodisperse nanocrystals, Nat. Mater., 2004, 3, 891–895.
36. Q. L. Cheng, V. Pavlinek, A. Lengalova, C. Z. Li, Y. He and P. Saha, Conducting polypyrrole confined in ordered mesoporous silica SBA-15 channels: preparation and its electrorheology, Microporous Mesoporous Mater., 2006, 93, 263–269.
37. J. S. Zhang, M. W. Zhang, C. Yang and X. C. Wang, Nanospherical Carbon Nitride Frameworks with Sharp Edges Accelerating Charge Collection and Separation at a Soft Photocatalytic Interface, Adv. Mater., 2014, 26, 4121–4126.
38. X. C. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J. M. Carlsson, K. Domen and M. Antonietti, A metal-free polymeric photocatalyst for hydrogen production from water under visible light, Nat. Mater., 2008, 8, 76–80.
39. J. S. Zhang, F. S. Guo and X. C. Wang, An optimized and general synthetic strategy for fabrication of polymeric carbon nitride nanoarchitectures, Adv. Funct. Mater., 2013, 23, 3008–3014.
40. J. X. Sun, Y. P. Yuan, L. G. Qiu, X. Jiang, A. J. Xie, Y. H. Shen and J. F. Zhu, Fabrication of composite photocatalyst g-C₃N₄–ZnO and enhancement of photocatalytic activity under visible light, Dalton Trans., 2012, 41, 6756–6763.
41. S. Ghosh, N. A. Kouamé, L. Ramos and S. Remita, Conducting polymer nanostructures for photocatalysis under visible light, Nat. Mater., 2015, 14, 505–511.
42. 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, 288–296.
43. W. J. Li, D. Z. Li, Y. M. Lin, P. X. Wang, W. Chen, X. Z. Fu and Y. Shao, Evidence for the active species involved in the photodegradation process of methyl orange on TiO₂, J. Phys. Chem. C, 2012, 116, 3552–3560.
44. Y. M. Lin, D. Z. Li, J. H. Hu, G. C. Xiao, J. X. Wang, W. J. Li and X. Z. Fu, Highly efficient photocatalytic degradation of organic pollutants by PANI-modified TiO₂ composite, J. Phys. Chem. C, 2012, 116, 5764–5772.
45. L. Q. Ye, J. Y. Liu, C. Q. Gong, L. H. Tian, T. Y. Peng and L. Zan, Two different roles of metallic Ag on Ag/AgX/BiOX (X = Cl, Br) visible light photocatalysts: surface plasmon resonance and Z-scheme bridge, ACS Catal., 2012, 2, 1677–1683.
46. L. Q. Ye, C. Q. Gong, J. Y. Liu, L. H. Tian, T. Y. Peng, K. J. Deng and L. Zan, Bin (Tu)xCl3n: a novel sensitizer and its enhancement of BiOCl nanosheets’ photocatalytic activity, J. Mater. Chem., 2012, 22, 8354–8360.
47. X. C. Li, G. L. Jiang, G. H. He, W. J. Zheng, Y. Yi Tan and W. Wu Xiao, Preparation of porous PPy-TiO₂ composites: improved visible light photoactivity and the mechanism, Chem. Eng. J., 2014, 236, 480–489.
48. H. Ullah, A.-u.-H. A. Shah, S. Bilal and K. Ayub, Doping and Dedoping Processes of Polypyrrole: DFT Study with Hybrid Functionals, J. Phys. Chem. C, 2014, 118, 17819–17830.

---

Footnote

† Electronic supplementary information (ESI) available: Nitrogen sorption isotherms of g-C₃N₄ (a) and MCIPs (b) calculated from desorption branch by BJH method, mass spectra of 2-mercaptobenzothiazole in the photodegradation process with MCIPs. See DOI: 10.1039/c5ra06209h

---