Facile Construction of All-Solid-State Z-Scheme g-C3N4/TiO2 Thin Film for the Efficient Visible-Light Degradation of Organic Pollutant.

The increasing discharge of dyes and antibiotic pollutants in water has brought serious environmental problems. However, it is difficult to remove such pollutants effectively by traditional sewage treatment technologies. Semiconductor photocatalysis is a new environment-friendly technique and is widely used in aqueous pollution control. TiO2 is one of the most investigated photocatalysts; however, it still faces the main drawbacks of a poor visible-light response and a low charge-separation efficiency. Moreover, powder photocatalyst is difficult to be recovered, which is another obstacle limiting the practical application. In this article, g-C3N4/TiO2 heterojunction is simply immobilized on a glass substrate to form an all-solid-state Z-scheme heterojunction. The obtained thin-film photocatalyst was characterized and applied in the visible-light photodegradation of colored rhodamine B and tetracycline hydrochloride. The photocatalytic performance is related to the deposited layers, and the sample with five layers shows the best photocatalytic efficiency. The thin-film photocatalyst is easy to be recovered with stability. The active component responsible for the photodegradation is identified and a Z-scheme mechanism is proposed.


Introduction
Water pollution has become particularly serious in recent decades due to the rapid development of industrialization [1]. The photocatalytic technique provides an alternative strategy to solve the above problem since it converts solar energy directly into useful energy for degrading organic pollutants into harmless components [2]. In 1977, Frank reported [3] the successful degradation of pollutants in water by using TiO 2 as the photocatalyst, which aroused extensive applications of TiO 2 in water treatment. In 1987, Mattews et al. [4] found that 34 kinds of organic pollutants in water could be degraded by TiO 2 under the irradiation of ultraviolet light (UV-light). Until now, TiO 2 is one of the most popular photocatalysts for environmental protection [5,6].
However, pristine TiO 2 photocatalyst has two intrinsic defects of a wide bandgap (3.2 eV) and the fast recombination of photogenerated carriers, which limits its widespread development. It is necessary to modify pristine TiO 2 to simultaneously satisfy the requirements for practical applications Preparation of g-C 3 N 4 /TiO 2 thin film. Typically, (1) the cleaning of the glass substrate would be carried out as follows: the diced glass substrate (20 × 15 × 1 mm 3 ) was washed under ultrasonic in isopropanol, deionized water, and anhydrous ethanol for 15 min in sequence. Then, the cleaned glass substrate was dried at 80 • C and put into a sealed container for future use. (2) In total, 24.0, 48.0, 145.0, 241.0 and 334.7 mg of g-C 3 N 4 powder was added into 20 mL of TiO 2 sol that had aged for 3 days, respectively. Then, they were constantly kept magnetically stirring for 20 min to obtain g-C 3 N 4 /TiO 2 powder-sol with its mass ratio of 0.05, 0.1, 0.3, 0.5 and 0.7. (3) The cleaned glass substrate was stuck to the homogenizer, 200 µL g-C 3 N 4 /TiO 2 pre-formed sol was dipped on the substrate. The sol was spin-coated on the substrate by a low speed of 500 rpm/min for 10 s and a successive high speed of 4000 rpm/min for 30 s. The coated sample was dried at 80 • C on a heating platform for 5 min, then one layer of film was obtained and the sample was denoted as a 1-layer g-C 3 N 4 /TiO 2 thin film. Then, 3, 5, and 7 layers of g-C 3 N 4 /TiO 2 thin film with the mass ratio of 0.5 and 5 layers of pristine TiO 2 thin film were prepared by a similar spin-coating process and the obtained samples were denoted as 3-layer g-C 3 N 4 (0.5)/TiO 2 , 5-layer g-C 3 N 4 (0.5)/TiO 2 , 7-layer g-C 3 N 4 (0.5)/TiO 2 , 5-layer TiO 2 , respectively. In addition, 5 layers of spin-coated g-C 3 N 4 /TiO 2 thin films with different g-C 3 N 4 /TiO 2 mass ratios of 0.05, 0.1, 0.3, 0.7 were prepared in the same way, and the obtained products were named as 5-layer g-C 3 N 4 (0.05)/TiO 2 , 5-layer g-C 3 N 4 (0.1)/TiO 2 , 5-layer g-C 3 N 4 (0.3)/TiO 2 , and 5-layer g-C 3 N 4 (0.7)/TiO 2 , respectively. In addition, g-C 3 N 4 (0.5)/TiO 2 powder photocatalysts were also prepared from the similar preparation processes of g-C 3 N 4 /TiO 2 thin-film photocatalyst. All the prepared samples were heated at 450 • C for 2 h with a ramp rate of 2.5 • C/min and automatically cooled down to room temperature.

Photocatalytic Activity Measurement
The photocatalytic activities of prepared g-C 3 N 4 /TiO 2 thin films were evaluated by the degradation of colored dye (rhodamine B, Rh B) and colorless antibiotic (tetracycline hydrochloride, TC-HCl) in water. A 300-W Xe lamp with a 420-nm filter was used as the visible light source (≥420 nm). The prepared thin-film photocatalysts were immersed into 50 mL of 5 mg·L −1 Rh B solution and 50 mL of 5mg·L −1 TC-HCl, and the distance between the light source and the glass substrate was set as 13 cm. The photocatalytic analyses were operated at a constant temperature by a circulating cooling system. Before the light irradiation, the solution with the photocatalyst was magnetically stirred in darkness for 0.5 h to establish adsorption equilibrium. During the light irradiation, 3 mL of the reaction liquid was taken out every 30 min and the solution was analyzed by the UV1600 spectrophotometer to determine the residual concentrate. Then, the measured solution was poured back into the reaction solution to continue the photodegradation.
Antibiotic TC-HCl was employed for stability measurement. Thin-film photocatalyst was immersed into 50 mL TC-HCl (5 mg·L −1 ) and the photodegradation was carried out in the same way as that of Rh B. After a light irradiation of 180 min, the thin-film photocatalyst was taken out and washed with de-ionized water. Then, it was immersed into TC-HCl solution for another cycle of photodegradation, and 5 cycles were run in a similar fashion. As a comparison, a blank experiment without the photocatalyst was done in a similar fashion to study the photo-resistance of Rh B and TC-HCl. Nanomaterials 2020, 10, 600 4 of 12

Trapping Experiment
The active species responsible for the photodegradation were identified by the trapping experiments. Diverse trapping agents such as TBA, BQ, EDTA were separately added into the reaction solution (Rh B, 5 mg·L −1 ) and the same operation as the above photodegradation was done.

Results and Discussions
3.1. Structure, Composition, and Morphology Figure 1A shows XRD patterns of the synthesized g-C 3 N 4 and two distinct diffraction peaks at 12.9 • and 27.6 • , corresponding to the (100) and (002) crystal planes of g-C 3 N 4 (PDF#87-1526) [22]. The same diffraction bands were also observed on other g-C 3 N 4 -containing samples. As for the TiO 2 -containing photocatalysts ( Figure 2B), the characteristic signals of anatase phase TiO 2 (PDF#21-1272) are presented, and no characteristic peaks associated with the rutile phase are observed in any of the samples. Compared with the 5-layer g-C 3 N 4 (0.5)/TiO 2 , the peak of powder g-C 3 N 4 (0.5)/TiO 2 has much sharper (101), (004), (200), (105), (211), crystal planes that all corresponds to clear peaks, which indicates the latter has higher crystallinity [23]. However, it is difficult to observe the clear diffraction bands from g-C 3 N 4 , which should be resulted from its low quantity and high dispersion in the thin film.
The FT-IR spectra of the prepared photocatalysts are shown in Figure 2. Characteristic peaks at 1640, 1409, 1321, and 1242 cm −1 can be clearly identified on g-C 3 N 4 , which are assigned to the stretching vibration of the aromatic CN heterocycle [24]. Additional peaks at 809 and 3100-3500 cm −1 can be attributed to the typical triazine ring vibration and the stretching vibration of N 2 H-and -OH [25]. As for 5-layer TiO 2 , the peaks at about 3400 and 1630 cm −1 are assigned to the bending and stretching vibration of OH from the adsorbed water, respectively, which is similar with the previous report [26]. Additionally, the characteristic absorption peak of the TiO 2 thin film detected in the wavelength range of 450-700 cm −1 may be attributed to the stretching vibration of the Ti-O-Ti bond [27]. Only weak characteristic peaks of the g-C 3 N 4 in 5-layer g-C 3 N 4 (0.5)/TiO 2 are detected in the FT-IR spectrum due to its low quantity. Nanomaterials 2020, 10, 600 4 of 12

Trapping Experiment
The active species responsible for the photodegradation were identified by the trapping experiments. Diverse trapping agents such as TBA, BQ, EDTA were separately added into the reaction solution (Rh B, 5 mg·L −1 ) and the same operation as the above photodegradation was done. Figure 1A shows XRD patterns of the synthesized g-C3N4 and two distinct diffraction peaks at 12.9° and 27.6°, corresponding to the (100) and (002) crystal planes of g-C3N4 (PDF#87-1526) [22]. The same diffraction bands were also observed on other g-C3N4-containing samples. As for the TiO2-containing photocatalysts ( Figure 2B), the characteristic signals of anatase phase TiO2 (PDF#21-1272) are presented, and no characteristic peaks associated with the rutile phase are observed in any of the samples. Compared with the 5-layer g-C3N4(0.5)/TiO2, the peak of powder g-C3N4(0.5)/TiO2 has much sharper (101), (004), (200), (105), (211), crystal planes that all corresponds to clear peaks, which indicates the latter has higher crystallinity [23]. However, it is difficult to observe the clear diffraction bands from g-C3N4, which should be resulted from its low quantity and high dispersion in the thin film. The FT-IR spectra of the prepared photocatalysts are shown in Figure 2. Characteristic peaks at 1640, 1409, 1321, and 1242 cm −1 can be clearly identified on g-C3N4, which are assigned to the stretching vibration of the aromatic CN heterocycle [24]. Additional peaks at 809 and 3100-3500 cm −1 can be attributed to the typical triazine ring vibration and the stretching vibration of N2H-and -OH [25]. As for 5-layer TiO2, the peaks at about 3400 and 1630 cm −1 are assigned to the bending and stretching vibration of OH from the adsorbed water, respectively, which is similar with the previous report [26]. Additionally, the characteristic absorption peak of the TiO2 thin film detected in the wavelength range of 450-700 cm −1 may be attributed to the stretching vibration of the Ti-O-Ti bond [27]. Only weak characteristic peaks of the g-C3N4 in 5-layer g-C3N4(0.5)/TiO2 are detected The surface morphology of the prepared g-C3N4/TiO2 thin-film photocatalyst was studied by SEM, as shown in Figure 3. A flat film can be clearly seen on the glass substrate, as shown in Figure  3A-C. All the films are composed of tightly contacted irregular large and small particles, and pores are also observed at the particle interstice. Along with the increase in film thickness, the surface roughness decreases first and increases afterward. Five-layer g-C3N4(0.5)/TiO2 has the most pores and the smallest particle size with a uniform size distribution, which indicates the presence of the photocatalytic property. The surface morphology of the prepared g-C 3 N 4 /TiO 2 thin-film photocatalyst was studied by SEM, as shown in Figure 3. A flat film can be clearly seen on the glass substrate, as shown in Figure 3A-C. All the films are composed of tightly contacted irregular large and small particles, and pores are also observed at the particle interstice. Along with the increase in film thickness, the surface roughness decreases first and increases afterward. Five-layer g-C 3 N 4 (0.5)/TiO 2 has the most pores and the smallest particle size with a uniform size distribution, which indicates the presence of the photocatalytic property. The surface morphology of the prepared g-C3N4/TiO2 thin-film photocatalyst was studied by SEM, as shown in Figure 3. A flat film can be clearly seen on the glass substrate, as shown in Figure  3A-C. All the films are composed of tightly contacted irregular large and small particles, and pores are also observed at the particle interstice. Along with the increase in film thickness, the surface roughness decreases first and increases afterward. Five-layer g-C3N4(0.5)/TiO2 has the most pores and the smallest particle size with a uniform size distribution, which indicates the presence of the photocatalytic property. Nanomaterials 2020, 10, 600 6 of 12 Figure 4 shows the UV-vis diffuse spectrum of the prepared samples. As seen in Figure 4A, all the prepared photocatalysts have strong absorption in the ultraviolet region. The absorption band edge moves towards the lower direction with the increase in coating layers, which should come from the increased g-C 3 N 4 . The bandgap is calculated through the Tauc formula and the results are given in Figure 4B [28]. The band gaps of the g-C 3 N 4 -containing photocatalysts are obviously narrower than that of pristine TiO 2 and become narrower with the increase in coating layers, and the specific band gap data are shown in Table 1. These results indicate that the addition of g-C 3 N 4 may improve light utilization. Nanomaterials 2020, 10, 600 6 of 12 Figure 4 shows the UV-vis diffuse spectrum of the prepared samples. As seen in Figure 4A, all the prepared photocatalysts have strong absorption in the ultraviolet region. The absorption band edge moves towards the lower direction with the increase in coating layers, which should come from the increased g-C3N4. The bandgap is calculated through the Tauc formula and the results are given in Figure 4B [28]. The band gaps of the g-C3N4-containing photocatalysts are obviously narrower than that of pristine TiO2 and become narrower with the increase in coating layers, and the specific band gap data are shown in Table 1. These results indicate that the addition of g-C3N4 may improve light utilization.

Effects of Film Thickness on the Photocatalytic Performance
The photocatalytic properties of the prepared photocatalysts were estimated by the visible-light degradation of Rh B. As shown in Figure 5A, the direct degradation of Rh B without a photocatalyst under visible light irradiation is extremely weak, which implies that the self-degradation of Rh B can be ignored. After the addition of photocatalyst, the removal rates were 5.1%, 17.9%, 31.2%, and 22.6% of Rh B under 180 min irradiation on 5-layer TiO2, 3-layer g-C3N4(0.5)/TiO2, 5-layer g-C3N4(0.5)/TiO2, 7-layer g-C3N4(0.5)/TiO2, respectively. These g-C3N4 containing-photocatalysts have much higher photocatalytic degradation efficiencies than that of pure TiO2 thin film, which may have resulted from the synergistic effect between g-C3N4 and TiO2. However, along with the increase in film layers, 7-layer g-C3N4 (0.5)/TiO2 has a decreased photodegradation efficiency. The kinetic rates for the photodegradation of Rh B are fitted according

Effects of Film Thickness on the Photocatalytic Performance
The photocatalytic properties of the prepared photocatalysts were estimated by the visible-light degradation of Rh B. As shown in Figure 5A, the direct degradation of Rh B without a photocatalyst under visible light irradiation is extremely weak, which implies that the self-degradation of Rh B can be ignored. After the addition of photocatalyst, the removal rates were 5.1%, 17.9%, 31.2%, and 22.6% of Rh B under 180 min irradiation on 5-layer TiO 2 , 3-layer g-C 3 N 4 (0.5)/TiO 2 , 5-layer g-C 3 N 4 (0.5)/TiO 2 , 7-layer g-C 3 N 4 (0.5)/TiO 2 , respectively. These g-C 3 N 4 containing-photocatalysts have much higher photocatalytic degradation efficiencies than that of pure TiO 2 thin film, which may have resulted from the synergistic effect between g-C 3 N 4 and TiO 2 . However, along with the increase in film layers, 7-layer g-C 3 N 4 (0.5)/TiO 2 has a decreased photodegradation efficiency. The kinetic rates for the photodegradation of Rh B are fitted according to the first-order reaction equation: ln(C 0 /C), where k, C 0 , and C is the rate constant, initial and real-time concentration of Rh B [24]. As shown in Figure 5B, the fitting curve of the photocatalytic degradation reaction is straight and is shown below. The k values of 5-layer TiO 2 , 3-layer g-C 3 N 4 (0.5)/TiO 2 , 5-layer g-C 3 N 4 (0.5)/TiO 2 , and 7-layer g-C 3 N 4 (0.5)/TiO 2 are 2.90 × 10 −4 , 11.2 × 10 −4 , 21.8 × 10 −4 , and 13.3 × 10 −4 min −1 , respectively. Among them, 5-layer g-C 3 N 4 (0.5)/TiO 2 shows the best photocatalytic performance and its rate constant is nearly four times higher than that of 5-layer TiO 2 . However, the photocatalytic activity decreases with the further increase in the coating layer, and 7-layer g-C 3 N 4 (0.5)/TiO 2 demonstrates a decreased efficiency. The difference in the photocatalytic activity should come from the different characteristics of the prepared g-C 3 N 4 /TiO 2 thin-film photocatalysts such as BET specific surface and particle size. According to the results of SEM analyses, the increase in the coating layers has an influence on the particle size and porous feature, and 5-layer g-C 3 N 4 (0.5)/TiO 2 has the smallest particle size and most of the pores. This may help the access between the pollutant and the active sites, which finally improves the photocatalytic performance. Nanomaterials 2020, 10, 600 7 of 12 to the first-order reaction equation: ln(C0/C), where k, C0, and C is the rate constant, initial and real-time concentration of Rh B [24]. As shown in Figure 5B, the fitting curve of the photocatalytic degradation reaction is straight and is shown below. The k values of 5-layer TiO2, 3-layer g-C3N4(0.5)/TiO2, 5-layer g-C3N4(0.5)/TiO2, and 7-layer g-C3N4(0.5)/TiO2 are 2.90 × 10 −4 , 11.2 × 10 −4 , 21.8 × 10 −4 , and 13.3 × 10 −4 min −1 , respectively. Among them, 5-layer g-C3N4(0.5)/TiO2 shows the best photocatalytic performance and its rate constant is nearly four times higher than that of 5-layer TiO2. However, the photocatalytic activity decreases with the further increase in the coating layer, and 7-layer g-C3N4(0.5)/TiO2 demonstrates a decreased efficiency. The difference in the photocatalytic activity should come from the different characteristics of the prepared g-C3N4/TiO2 thin-film photocatalysts such as BET specific surface and particle size. According to the results of SEM analyses, the increase in the coating layers has an influence on the particle size and porous feature, and 5-layer g-C3N4(0.5)/TiO2 has the smallest particle size and most of the pores. This may help the access between the pollutant and the active sites, which finally improves the photocatalytic performance.
The photodegradation rates on all the prepared samples are summarized in Figure 6D. As shown, both the thickness and composition of the prepared thin film have an impact on the photocatalytic

Photocatalytic Stability
Except for the easy operation, cyclic stability is another important factor for the industrial application of thin-film photocatalysts. The antibiotic residue is another refractory pollutant in water, and the photodegradation of tetracycline hydrochloride (TC-HCl) was evaluated as the model reaction to study the photocatalytic activity as well as the stability. Five-layer g-C3N4(0.5)/TiO2 is selected for this evaluation and the result is shown in Figure 7. A total of 25.8% TC-HCl is degraded after 180 min of visible-light irradiation, and no reduced photocatalytic performance is observed at the 2nd cycle. However, a slight decrease in the photocatalytic activity is present after the 2nd cycle, which may come from the loss of weakly attached photocatalysts. Then, no further activity loss in the other cycles is observed, indicating that the prepared thin-film photocatalyst has stability.

Photocatalytic Stability
Except for the easy operation, cyclic stability is another important factor for the industrial application of thin-film photocatalysts. The antibiotic residue is another refractory pollutant in water, and the photodegradation of tetracycline hydrochloride (TC-HCl) was evaluated as the model reaction to study the photocatalytic activity as well as the stability. Five-layer g-C 3 N 4 (0.5)/TiO 2 is selected for this evaluation and the result is shown in Figure 7. A total of 25.8% TC-HCl is degraded after 180 min of visible-light irradiation, and no reduced photocatalytic performance is observed at the 2nd cycle. However, a slight decrease in the photocatalytic activity is present after the 2nd cycle, which may come from the loss of weakly attached photocatalysts. Then, no further activity loss in the other cycles is observed, indicating that the prepared thin-film photocatalyst has stability.

Proposed Mechanism
The capture experiments were carried out to identify the main active substances in the photocatalytic process for understanding the corresponding reaction mechanism. Benzoquinone (BQ), tert butyl alcohol (TBA), and ethylene diamine tetraacetic acid (EDTA) were added into the Rh B solution as the trapping agent for superoxide radical (O2 − ), hydroxyl radical (OH), and hole (h + ), respectively [29]. As shown in Figure 8, the photocatalytic effect is significantly improved after the introduction of EDTA, which suggests that h + is not responsible for the photodegradation. However, the photocatalytic activity is prominently inhibited after the adding of BQ and TBA. This indicates the existence of OH and O2 − and they both play a vital role in the above photocatalytic degradation. The reaction mechanism is discussed according to the above trapping experiment. If a traditional heterojunction is formed at the interface of 5-layer g-C3N4(0.5)/TiO2, photo-generated electrons may be concentrated in the conduction band of TiO2, and photo-generated holes are transferred into the valence band of g-C3N4. However, the valence band energy of g-C3N4 is less than the energy required for the formation of OH, which cannot oxidize H2O/OH -into·OH to participate in the photocatalytic reaction. Therefore, the traditional heterojunction is not responsible

Proposed Mechanism
The capture experiments were carried out to identify the main active substances in the photocatalytic process for understanding the corresponding reaction mechanism. Benzoquinone (BQ), tert butyl alcohol (TBA), and ethylene diamine tetraacetic acid (EDTA) were added into the Rh B solution as the trapping agent for superoxide radical (O 2 • − ), hydroxyl radical (•OH), and hole (h + ), respectively [29]. As shown in Figure 8, the photocatalytic effect is significantly improved after the introduction of EDTA, which suggests that h + is not responsible for the photodegradation. However, the photocatalytic activity is prominently inhibited after the adding of BQ and TBA. This indicates the existence of •OH and O 2 • − and they both play a vital role in the above photocatalytic degradation.

Proposed Mechanism
The capture experiments were carried out to identify the main active substances in the photocatalytic process for understanding the corresponding reaction mechanism. Benzoquinone (BQ), tert butyl alcohol (TBA), and ethylene diamine tetraacetic acid (EDTA) were added into the Rh B solution as the trapping agent for superoxide radical (O2 − ), hydroxyl radical (OH), and hole (h + ), respectively [29]. As shown in Figure 8, the photocatalytic effect is significantly improved after the introduction of EDTA, which suggests that h + is not responsible for the photodegradation. However, the photocatalytic activity is prominently inhibited after the adding of BQ and TBA. This indicates the existence of OH and O2 − and they both play a vital role in the above photocatalytic degradation. The reaction mechanism is discussed according to the above trapping experiment. If a traditional heterojunction is formed at the interface of 5-layer g-C3N4(0.5)/TiO2, photo-generated electrons may be concentrated in the conduction band of TiO2, and photo-generated holes are transferred into the valence band of g-C3N4. However, the valence band energy of g-C3N4 is less than the energy required for the formation of OH, which cannot oxidize H2O/OH -into·OH to participate in the photocatalytic reaction. Therefore, the traditional heterojunction is not responsible The reaction mechanism is discussed according to the above trapping experiment. If a traditional heterojunction is formed at the interface of 5-layer g-C 3 N 4 (0.5)/TiO 2 , photo-generated electrons may be concentrated in the conduction band of TiO 2 , and photo-generated holes are transferred into the valence band of g-C 3 N 4 . However, the valence band energy of g-C 3 N 4 is less than the energy required for the formation of •OH, which cannot oxidize H 2 O/OHinto •OH to participate in the photocatalytic reaction. Therefore, the traditional heterojunction is not responsible for the photodegradation of Rh B, and the transport of photo-generated electrons and holes in photocatalytic reactions must follow the all-solid-state Z-scheme process, which is in correspondence with previous articles [12,18]. The transport process of all-solid-state Z-scheme heterojunction carriers is shown in Figure 9. The excited electrons in TiO 2 transfers from its conduction band to the valence band of g-C 3 N 4 . Then, the electrons are excited to the conduction band of g-C 3 N 4 and react with the substrate to form active species. It might be pointed out that •OH is generally produced from the direct oxidation on the holes. However, it is difficult to completely exclude the formation of •OH from the consecutive reactions [30,31]. Therefore, another possible formation of •OH from the consecutive conversion of O 2 • − is also given in Figure 9. for the photodegradation of Rh B, and the transport of photo-generated electrons and holes in photocatalytic reactions must follow the all-solid-state Z-scheme process, which is in correspondence with previous articles [12,18]. The transport process of all-solid-state Z-scheme heterojunction carriers is shown in Figure 9. The excited electrons in TiO2 transfers from its conduction band to the valence band of g-C3N4. Then, the electrons are excited to the conduction band of g-C3N4 and react with the substrate to form active species. It might be pointed out that •OH is generally produced from the direct oxidation on the holes. However, it is difficult to completely exclude the formation of •OH from the consecutive reactions [30,31]. Therefore, another possible formation of •OH from the consecutive conversion of O2 − is also given in Figure 9.

Conclusion
In summary, all-solid-state Z-scheme g-C3N4/TiO2 thin-film photocatalyst was successfully prepared by the facile sol-gel and spin-coating methods which showed good photocatalytic performance for the removal of colored Rh B and colorless TC-HCl with easy operation and good stability. It was found that the mass ratio of g-C3N4 to TiO2 and coating layers had a strong influence on the photocatalytic activity. Five-layer g-C3N4(0.5)/TiO2 showed the highest efficiency for the photodegradation and a Z-scheme mechanism was proposed. The construction of a thin-film photocatalyst may not only benefit the recovery from the reaction but also show good cyclic stability, which may be of great significance for practical application in the future.
Author Contributions: For research articles with several authors, a short paragraph specifying their individual contributions must be provided. The following statements should be used "Conceptualization, Z.C. and X.L.; methodology, Z.C.; software, Q.Y.; validation, X.Y., C.L. and X.Q.; formal analysis, W.Z.; investigation, W.Z.; resources, Y.W.; data curation, T.S.; writing-original draft preparation, W.Z.; writing-review and editing, Z.X.; visualization, W.C.; supervision, X.L..; project administration, Z.C.; funding acquisition, Z.C. All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by the Zhejiang Provincial Natural Science Foundation of China, grant number LY17E020008.

Conflicts of Interest:
The authors declare no conflict of interest.

Conclusions
In summary, all-solid-state Z-scheme g-C 3 N 4 /TiO 2 thin-film photocatalyst was successfully prepared by the facile sol-gel and spin-coating methods which showed good photocatalytic performance for the removal of colored Rh B and colorless TC-HCl with easy operation and good stability. It was found that the mass ratio of g-C 3 N 4 to TiO 2 and coating layers had a strong influence on the photocatalytic activity. Five-layer g-C 3 N 4 (0.5)/TiO 2 showed the highest efficiency for the photodegradation and a Z-scheme mechanism was proposed. The construction of a thin-film photocatalyst may not only benefit the recovery from the reaction but also show good cyclic stability, which may be of great significance for practical application in the future.