Improved in Situ Synthesis of Heterostructured 2D/2D BiOCl/g-C3N4 with Enhanced Dye Photodegradation under Visible-Light Illumination.

A simple, in situ, and one-pot hydrothermal strategy was applied for the successful manufacturing of heterostructured 2D/2D BiOCl/g-C3N4 photocatalysts, and outstanding photodegradation of Rhodamine B in the condition of visible-light irradiation over the composites emerged. The investigation of various BiOCl/g-C3N4 ratios influencing the activity implied that the optimized B2C1 (mole ratio of BiOCl/g-C3N4 with 2:1) exhibited the higher degradation efficiency than that of the rest of the composites, even higher than that of pure BiOCl and pure g-C3N4, which yielded over 90% in the initial 30 min and reached almost 100% during the whole 70 min irradiation process. Kinds of characterizations demonstrated that the enhancement of photodegradation performance was caused by the intimate contact between BiOCl and g-C3N4 to form the heterostructure, which could benefit the generation of abundant visible-light photoinduced carriers and help enhance their separation and then promote their transportation to the surface.

Introduction

Environmental pollution and energy crises are two main global issues that have been the hot topics in the past decades.[1] For the former, water pollution influences the drinking water safety of humans directly, and among the abundant pollutants, dyes such as Rhodamine B (RhB) deserve to be focused on due to its massive emission from industries and daily use, difficult degradation, and toxicity.[2] Due to these harmful properties, kinds of methods were applied for the innocent treatment of RhB, including chemical degradation, physical and chemical adsorption, photocatalysis, or the combinations of these methods.[3−6] Among these treatments, photocatalysis is a rising star owing to the nature of the infinite driving force from solar energy and the green treating processes.

Until now, plentiful photocatalysts have been explored and synthesized containing TiO2, Bi-based composites, Cd-based materials, MOFs, TaON, etc.[7−15] Unfortunately, problems such as low solar light usage and fast carrier-binding need to be settled all the time. Recently, 2D materials like g-C3N4, graphene, C3N, MoS2, and BiOCl were developed due to their unique properties, such as large surface areas and adjustable layers.[16−23] Two-dimensional materials with few layers benefit the separation and transport of photoinduced carriers, thus enhancing photocatalytic performance.[24] Among these 2D materials, the BiOCl nanosheet attracts more attention owing to its stability, indirect band gap, and nontoxicity.[25] Numerous research studies confirmed the photocatalytic efficiency of BiOCl counting on its morphology and size. Hence, many strategies are dedicated to regulating the thickness of 2D materials. The pH value in the precursor was reported to influence the thickness of 2D BiOCl, and the pH value displayed a positive influence on the thickness in which BiOCl prepared in the condition of pH = 5 exhibited the highest activity for RhB photodegradation.[24] The addition of metal ions such as Fe3+, Co2+, and Na+ was also confirmed to affect the thickness of 2D BiOCl.[26] However, the wide band gap of BiOCl limits its utilization of sunlight. These methods are worth developing to overcome this disadvantage.

Constructing the heterostructure between BiOCl and other photocatalysts is a proven effective strategy, which can not only enlarge the use of sunlight but also promote the segregation of photoinduced charges further.[27,28] Especially, building 2D/2D heterostructures is a promising way to generate large contact areas and accelerate charge transfer. The construction of 2D BiOCl/Bi12O17Cl2 nanojunctions was reported, and the results confirmed that the composites displayed higher NO removal efficiency than that of pure materials.[29] The heterostructured 2D/2D BiOCl/K+Ca2Nb3O10– nanosheets were constructed, and they improved the photodegradation of tetracycline hydrochloride.[30] Among the vast 2D materials, g-C3N4 is focused on due to its dramatic properties such as easy preparation, metal-free composition, satisfied stability, and visible light response. g-C3N4 has been utilized for kinds of visible-light photocatalytic tests, such as photodegradation, water splitting, CO2 reduction, etc.[1,31,32] Heterostructured BiOCl/g-C3N4 composites have been widely reported for photodegradation application. Liu et al. prepared 2D BiOCl/g-C3N4 layered composites for photodegradation of methyl orange, and the degradation efficiency reached 84.28% after 180 min of visible-light illumination.[33] The preparation method could not make full use of BiOCl or g-C3N4; thus, the photoactivity could be enhanced further by improving the synthesis routes. Song et al. synthesized BiOCl/g-C3N4 composites for RhB photodegradation via two different preparation methods, including NH4Cl-templated hydrothermal and sonication-assisted deposition–precipitation routes, and they all exhibited satisfied photodegradation efficiency.[34,35] The microwave-assisted method was also applied for the preparation of BiOCl/g-C3N4 composites, and the excellent nizatidine photodegradation performance was achieved.[36] However, the contact area between BiOCl and g-C3N4 could be enlarged by adjusting the shape of BiOCl to be uniform to promote the transportation of carriers, thus further enhancing the photocatalytic performance. Therefore, it is necessary to explore a new strategy for the construction of 2D BiOCl/g-C3N4 composites with large contact areas.

Herein, in this work, a new simple one-pot hydrothermal method was developed for the preparation of 2D/2D heterostructured BiOCl/g-C3N4 with large contact areas successfully, and they were investigated by photodegradation of RhB in the condition of visible light irradiation. The influence of various ratios of BiOCl to g-C3N4 on the photocatalytic efficiency was investigated, and various characterizations such as XRD, SEM, TEM, XPS, UV–vis, PL, etc. were adopted to explain the relationship between the structure and activity. The mechanism of photodegradation of RhB over 2D/2D heterostructured BiOCl/g-C3N4 was proposed. This work aims to provide a strategy for the construction of two separate 2D materials to form a heterostructure with a large contact area in a simple and economic way, which can still exhibit the enhanced photocatalytic performance.

Results and Discussion

Physicochemical Structure and Morphology Observation

XRD patterns were performed to confirm the crystal forms of the sole BiOCl, sole g-C3N4, and BiOCl/g-C3N4 composites, displayed in Figure a. The Bi sample displayed the typical tetragonal BiOCl crystal structure (JCPDS no. 06-0249). However, the intensities of the peaks at 32.5° and 33.4° were attributed to the standard card, which corresponded to the (110) and (102) crystal faces, respectively. The results indicated that the main exposed face was (110), which was confirmed to be the active face for photocatalysis.[37] The same phenomenon was observed for BiOCl-containing samples. Sole g-C3N4 displayed two obvious diffraction peaks at 12.8° and 27.1°, ascribed to the (100) and (002) crystal faces of typical g-C3N4. The former peak was caused by the in-plane repeated tri-s-triazine moieties, while the latter originated from the stacked, conjugated aromatic system.[38] The distinct peaks that appeared in pure BiOCl and g-C3N4 confirmed their high crystallinity. For all BiOCl/g-C3N4 composites, two series peaks assigned to two pure materials were both observed, indicating the presence of BiOCl and g-C3N4 in the composites and both structures being retained. Moreover, the peak intensity corresponding to g-C3N4 intensified gradually with the increase of the g-C3N4/BiOCl mole ratio, showing that g-C3N4 was mostly adopted in the composites. No additional peaks were observed in the composites, displaying the pure formation of the composites.

Figure 1

(a) XRD patterns and (b) FT-IR of sole BiOCl, sole g-C3N4, and BiOCl/g-C3N4 composites.

The chemical structures on the surface of the samples were also explored via FT-IR techniques, as shown in Figure b. Typically, the wide peaks from 3000 to 3400 cm–1 belonged to the N–H band. Peaks at 1315 and 1240 cm–1 belonged to the typical extending modes of CN heterocycles.[39] Moreover, the peak ascribed to the typical vibration of triazine rings also appeared at 810 cm–1.[40] Additionally, the peak that appeared at 1010 cm–1 was caused by the Bi–O vibration. In the overview of all samples, it could be seen that the intensities of the peaks ascribed to BiOCl or g-C3N4 increased along with the ratio increase of the corresponding component in the composites. It was further explained that the BiOCl/g-C3N4 composites were constructed successfully.

The N2 adsorption–desorption isotherms over Bi, CN, and B2C1 samples were applied to investigate the BET surface area, as shown in Figure S1. Three samples all exhibited N2 isotherms of type IV with hysteresis loops of type H3,[41] confirming the formation of a mesoporous structure. The BET surface areas of sole BiOCl, sole g-C3N4, and the heterostructured BiOCl/g-C3N4 were 14.6, 112.3, and 47.1 m2/g, respectively. Compared to two pure samples, BiOCl/g-C3N4 displayed a decreased surface area value, which was ascribed to the introduction of BiOCl with a relatively low BET surface area. This phenomenon could also confirm the generation of the heterostructure between BiOCl and g-C3N4.

SEM images were conducted to observe all morphologies, displayed in Figure . The sole g-C3N4 displayed the aggregated rolled flake morphology, while pure BiOCl showed the typical irregular lamellar morphology. The surface of pure BiOCl was extremely smooth, confirming its purity after preparation. The size of pure BiOCl ranged from approximately 200 to 1100 nm in width and approximately 70 nm in thickness. For all BiOCl/g-C3N4 composites, two typical morphologies corresponding to two pure materials were observed, and the number of particles assigned to g-C3N4 increased with the increasing of g-C3N4/BiOCl mole ratio. Additionally, the EDS result of the selected B2C1 sample confirmed the existence of Bi, O, Cl, C, and N elements in the composites, displayed in Figure S2, indicating the successful introduction of Bi, O, Cl, C, and N elements, which was in accordance with XRD analysis.

Figure 2

SEM images of sole BiOCl, sole g-C3N4, and BiOCl/g-C3N4 composites.

HR-TEM characterization was adopted to observe more detailed morphologies of sole BiOCl, sole g-C3N4, and BiOCl/g-C3N4 composites, displayed in Figure . The irregular size and smooth surface of the pure BiOCl was further confirmed from TEM images. The distinct periodic lattice spacing was observed from HR-TEM, which exhibited the interplanar distance of 0.275 nm, corresponding to the (110) crystal plane. From the selected-area electron diffraction (SAED) images, the single-crystal diffraction spot appeared, which showed the two crystal planes of (110) and (200). For the pure g-C3N4 sample, the unfolded flake morphology was observed. The obvious surface of g-C3N4 could be seen from HR-TEM, and one polycrystal diffraction ring appeared in SAED images, which was ascribed to the (002) plane of g-C3N4. For the B2C1 composite, the BiOCl particles emerged mainly in TEM images, and the flake morphology of g-C3N4 could also be observed. The similar shape in the composites to the separate pure materials indicated that the combination between g-C3N4 and BiOCl could not alter their original morphologies. In HR-TEM images, g-C3N4 was capped over BiOCl particles, indicating the impinged contact between BiOCl and g-C3N4. This phenomenon confirmed the generation of the face-to-face heterostructure of BiOCl and g-C3N4, which could improve the transfer and separation of the photoinduced carriers, thus enhancing the photocatalytic performance. Moreover, the single-crystal diffraction spot and one polycrystal diffraction ring both appeared in the mixture, explaining the existence of BiOCl and g-C3N4.

Figure 3

TEM, HR-TEM, and SAED images of sole BiOCl, sole g-C3N4, and B2C1 composites.

SEM mapping of the B2C1 composite was applied to investigate the dispersion of BiOCl and g-C3N4 in the mixture, displayed in Figure S3. The phenomenon displayed the homogeneous distribution of Bi, O, Cl, C, and N elements in the B2C1 sample, indicating a high dispersion of g-C3N4 and BiOCl in the composite.

Band Structure Analysis

Since the photocatalytic performance was determined by the band structure of the photocatalyst, it is necessary to explore the band structure information of the component in the BiOCl/g-C3N4 mixtures. The valence band X-ray photoelectron spectroscopy (VB-XPS) was applied to explore the valence band positions (EVB) of pure BiOCl and pure g-C3N4, displayed in Figure a. The obtained values of BiOCl and g-C3N4 were 1.58 and 1.93 eV, respectively. The band gap (Eg) was obtained from the conversion of UV–vis DRS spectra. The conversion is based on the following equation.where h, v, and A is the Planck’s constant, frequency of photons, and absorbance, respectively. The n value in the condition of n = 1 is the direct transition, while n = 4 is the indirect transition. For both BiOCl and g-C3N4, the values of n are both selected as 4.

Figure 4

(a) VB-XPS data. (b) Plot of (Ahv)1/2 vs photon energy (hv), (c) Mott–Schottky profiles, and (d) illustrated band structures of sole BiOCl and sole g-C3N4.

Hence, the Eg values of BiOCl and g-C3N4 were 2.82 and 2.33 eV, respectively, displayed in Figure b. Mott–Schottky measurements were adopted to find out more information of the band structure, as shown in Figure c. The positive slope of the tangent of profiles over BiOCl and g-C3N4 indicated that they were both n-type semiconductors. The flat band positions (Vfb where the unit is V vs Ag/AgCl) could be obtained from the intersection between the tangent and y = 0 plot, which were −0.20 and −0.19 corresponding to BiOCl and g-C3N4. Based on the following formula, the unit of Vfb was converted from V vs Ag/AgCl to the normal hydrogen electrode (NHE) potential.[42]

After conversion, the Vfb values of BiOCl and g-C3N4 were −0.003 and 0.007 eV, respectively. Generally, the position of Vfb was close to the value of the Fermi level.[43] Based on the known value of EVB and Eg, the position of the conduction band (ECB) could be computed based on the following formula.

The calculated ECB values of BiOCl and g-C3N4 were −1.24 and −0.40 eV, respectively. From the above results, the band structures of sole BiOCl and sole g-C3N4 were obtained, illustrated in Figure d. Moreover, the Vfb values of BiOCl and g-C3N4 were extremely close to each other, which benefited the construction of the heterostructure between sole BiOCl and sole g-C3N4, confirming the analysis of TEM.

RhB Photodegradation Performance

Photodegradation of RhB tests under visible-light illumination for all samples was shown in Figure a. Among all photocatalysts, pure g-C3N4 displayed the lowest degradation efficiency, which reached only 65% after 70 min of visible-light irradiation. Sole BiOCl exhibited higher performance than that of sole g-C3N4 during the photodegradation process, and it reached 89% degradation efficiency after 70 min of irradiation. It could be seen that among all samples, the efficiency increased initially and then declined along with the increase of BiOCl to g-C3N4 ratios, and the B2C1 sample displayed the highest photodegradation activity, which reached almost 100% degradation efficiency after 70 min. Moreover, the efficiency of the B2C1 sample yielded over 90% after just 30 min of visible-light irradiation. Additionally, the blank test was conducted, and the results displayed that there were almost no photodegradation activity in the absence of photocatalysts, indicating that the self-degradation of RhB for the photocatalytic activity could be excluded.

Figure 5

(a) Photocatalytic degradation of RhB and (b) linear fit for the first-order kinetics model over sole BiOCl, sole g-C3N4, and BiOCl/g-C3N4 composites.

The degradation kinetics for all samples was conducted, displayed in Figure . The fitting results displayed that all samples meet well the first-order kinetics model with the following equation.where C, C0, k, and t is the rest concentration of RhB in the reaction system, fresh concentration of RhB, apparent rate constant, and irradiation time, respectively. The variation trend of the obtained k values from the fitting profiles was in accordance with the tendency of photodegradation activity. Moreover, the k value of the optimum B2C1 sample was 2.4 and 4.9 times larger than that of sole BiOCl and sole g-C3N4.

The catalytic reaction stability of the catalyst is crucial for practical application. Therefore, the photocatalytic stability over the B2C1 composite was performed via four time-cycle degradation experiments, as shown in Figure S4. After four runs of the photocatalysis reaction, the RhB degradation efficiency over the B2C1 sample after 30 min of visible-light illumination decreased slightly from 90.36% in the first run to 90.04% in the fourth run, indicating its satisfying photocatalytic stability.

Enhanced Photodegradation Investigation

XPS characterization was adopted to investigate the difference of surface valence states of different elements between pure materials and the composites, aiming to confirm the interaction between BiOCl and g-C3N4 and explore the enhancement on photodegradation, as shown in Figure . Typically, the optimum B2C1 sample was selected as the composite for comparison. The survey spectra confirmed the existence of Bi, C, O, Cl, and N elements on the surface, in accordance with EDS analysis. Displayed in Figure b, two main peaks were observed on the Bi-containing samples in which the peaks at low binding energy (BE) were ascribed to Bi 4f7/2, while the high BE peaks were assigned to Bi 4f5/2.[44] Compared to pure BiOCl, the peaks in B2C1 shifted to the negative BE value, explaining the interaction between two sole materials. Nevertheless, the generation of the heterostructure could not change the valence state of Bi, and the distance between two splitting peaks remained approximately 5.3 eV indicating the existence of Bi3+.[45] In the O 1s spectra, two peaks appeared on both samples. The peaks at a low BE value were attributed to the lattice oxygen, and the peaks at a high BE value were assigned to the chemisorbed oxygen, such as O2–.[46] The Cl 2p spectra displayed two obvious peaks, which belonged to Cl 2p3/2 and Cl 2p1/2, corresponding to the low and high BE peaks, respectively.[47] The same as Bi 4f spectra, the observed peak shift on the composite was due to the formation of the heterostructure. For the C 1s spectra, two maxima at approximately 283 and 287 eV appeared. The former was caused by the surface-adsorbed external carbon or defect-containing sp2 C elements from graphitic carbon, while the latter was ascribed to the N—C≡N coordinate in the framework of C3N4.[48] The N 1s spectra could be fit into three maxima, which are located at approximately 397, 399, and 403 eV. The lowest BE peaks were attributed to the sp2 N involved in triazine rings, and the middle BE peaks were assigned to the bridged N atoms in N–(C)3[49,50] or N bonded to H atoms.[51] The highest weak BE peaks were ascribed to π excitations.[52] It is worth mentioning that the highest BE peak in the B2C1 composite disappeared, indicating that BiOCl interacted with g-C3N4 through the π-electron cloud of CN heterocycles.[52] Additionally, the formation of the heterostructure was further confirmed by the fact that the shifts were also observed in the C 1s and N 1s spectra. Combined with photodegradation performance, it could be concluded that the formation of the heterostructure could benefit the segregation and transportation of photoinduced charges, consequently enhancing the degradation efficiency.

Figure 6

XPS spectra of sole BiOCl, sole g-C3N4, and BiOCl/g-C3N4 composites. (a) Survey, (b) Bi 4f, (c) O 1s, (d) Cl 2p, (e) C 1s, and (f) N 1s.

The technique of UV–vis DRS was adopted to explore the photoresponse at different wavelengths over these samples, and PL characterization was applied to investigate the segregation and transfer of charges, displayed in Figure a,b. Compared to pure BiOCl, the adsorption edges of all composites had undergone an obvious redshift, and when compared to sole g-C3N4, light absorptivity of the composites was higher. The results indicated that the formation of the heterostructure between two sole materials could promote the light adsorption efficiency, thus causing more excitation of valence electrons and then generating more active sites. The intensity of PL spectra generally reflects the amounts of the recombined electrons and holes, and a low intensity displays a suppressed recombination rate.[53] In the PL spectra, sole BiOCl and sole g-C3N4 exhibited stronger PL emission than that of the composites, indicating the fast binding of charges. The facts also explained that the construction of BiOCl and g-C3N4 could suppress the recombination and promote the segregation of charges. Among all composites, the B2C1 sample displayed the weakest intensity, confirming its most efficient segregation of charges, thus exhibiting the highest photodegradation performance.

Figure 7

(a) UV–vis DRS plots, (b) PL profiles, (c) EIS spectra, and (d) PC potential plots of sole BiOCl, sole g-C3N4, and BiOCl/g-C3N4 composites.

The electrochemical impedance spectroscopy (EIS) and photocurrent (PC) tests were applied to further study the segregation efficiency of charges. In general, the large radius of EIS curves explains the big carrier transportation resistance, and the large value of the photocurrent under irradiation implies the vast amount of photoinduced carriers.[54,55] As shown in Figure c, the small radii were appeared in sole BiOCl and sole g-C3N4, and after the combination, the radii of the composites were increased. Among all composites, the B2C1 sample displayed the largest value of radius. The above results confirmed that the construction of the heterostructure between BiOCl and g-C3N4 could weaken the charge transfer resistance, thus benefitting the transfer of the generated carriers. From Figure d, the similar phenomenon was observed in which sole BiOCl and sole g-C3N4 displayed a weak visible light response and the composites showed relatively strong intensity. Moreover, the B2C1 sample exhibited the most intense response under visible-light illumination. The results confirmed that the photoinduced carriers were generated in abundance, and the separation capacity was enlarged through the formation of the heterostructure, which was in accordance with PL results. Combined with photocatalytic performance, it could be concluded that the formation between BiOCl and g-C3N4 could generate the abundant visible-light photoinduced charges, enhance the segregation of these charges, and then promote their transportation to the surface, thus leading to outstanding photodegradation activity of RhB under visible-light illumination.

Photocatalysis Mechanism

Photodegradation of the RhB mechanism in the condition of visible light illumination over the BiOCl/g-C3N4 mixture was proposed, as displayed in Figure . During the first adsorption equilibrium process, the dye of RhB was initially adsorbed on the surface of BiOCl; then, it could act as a photosensitizer for photodegradation. Due to the small distance between HOMO and LUMO orbitals of RhB, they could be excited to form RhB* under visible-light illumination. Next, the photoexcited electrons in RhB were transferred to the conduction band (CB) of BiOCl, which possessed more positive potential.[56] However, due to the appropriate Eg of BiOCl and g-C3N4, they could be both excited under visible-light illumination. Hence, the jammed electrons on the CB of BiOCl were transferred to the more positive CB of g-C3N4; then, the formation of the heterostructure between BiOCl and g-C3N4 could promote this transportation. On account of the same excitation of g-C3N4 in the condition of visible-light illumination, the electrons on the CB crowded to a higher degree. Therefore, the electrons on the CB of g-C3N4 were transferred to the surface of the composite, and they were able to reduce the adsorbed oxygen to form O2–, caused by the CB potential of g-C3N4 being more negative than that of O2/O2–. In addition, the photoexcited holes in the valence band (VB) of g-C3N4 were transferred to the VB of BiOCl since the VB potential of g-C3N4 was more positive than that of BiOCl. However, the holes in the VB of BiOCl could not oxidize OH– to ·OH due to the more negative VB in BiOCl than the value of ·OH/OH–. From the above analysis, it could be found that the generated O2– played a crucial role in photodegradation of RhB.

Figure 8

Schematic figure of the proposed photodegradation of the RhB mechanism under visible light over the heterostructured BiOCl/g-C3N4 photocatalyst.

Conclusions

In summary, the heterostructured 2D/2D BiOCl/g-C3N4 photosamples were successfully synthesized through a facile in situ one-pot hydrothermal method, and the influence of different BiOCl/g-C3N4 ratios on visible light photodegradation of RhB was investigated. Compared to sole BiOCl and sole g-C3N4, the combination of BiOCl and g-C3N4 promoted the photocatalytic performance. Among all composites, the B2C1 sample displayed the highest degradation efficiency, which yielded over 90% in only 30 min and reached almost 100% within the whole 70 min reaction time. The enhancement on photodegradation was ascribed to the impinged contact between the two sole materials to form the heterostructure, which could benefit the generation of abundant visible-light photoinduced carriers and help enhance the segregation of these charges and then promote their transportation to the surface. This work provides a strategy for the preparation of 2D/2D hetero-structured photo-samples with high visible light photocatalytic efficiency in a simple and economic way.

Experimental Section

Chemical and Materials

Melamine, ethanol, bismuth nitrate (Bi(NO3)3·5H2O), cetyltrimethyl ammonium bromide (CTAB), carbamide, and glycerinum were all from Sinopharm Chemical Reagent Co., Ltd. (China). The reagents were all of analytical grade and utilized without further handling. Deionized water (H2O) after being Millipore system-purified was used throughout all experiments.

Synthesis of Pure g-C3N4

g-C3N4 was obtained from the simple calcination of melamine. Typically, 5 g of melamine was put in a 50 mL capped ceramic crucible and heated to 550 °C with the rate of 2 °C/min in the air atmosphere for 4 h. The final yellow powder was obtained, named as CN.

Preparation of BiOCl/g-C3N4

The 2D/2D heterostructured BiOCl/g-C3N4 was constructed via a simple in situ hydrothermal strategy. First, g-C3N4 was dispersed in 50 mL of ethanol. The suspension solution was stirred with a magneton for 1 h and further dispersed with ultrasound for another 1 h. Then, Bi(NO3)3·5H2O, CTAB, and urea were added into the above solution according to the mole ratio of 3:1:1; the additional 5 mL of glycerinum was also adopted. Another 1 h of stirring was conducted. Then, the suspension system was poured into a 100 mL Teflon container, sealed in an autoclave, and hydrothermally reacted at 180 °C for 24 h. When the container was cooled to room temperature naturally, the product was washed and centrifuged with H2O and ethanol several times. The product was obtained after vacuum drying at 60 °C for 6 h. The synthesized catalysts were noted as B3C1, B2C1, B1C1, B1C2, and B1C3 separately, which corresponded to the mole ratio of Bi(NO3)3·5H2O to g-C3N4 with 3:1, 2:1, 1:1, 1:2, and 1:3, respectively. The pure BiOCl was also prepared for comparison. The preparation process was like the construction of the BiOCl/g-C3N4 composite, except that g-C3N4 was absent in the solution. The pure BiOCl was named as Bi.

Characterizations

Powder X-ray diffraction (XRD) patterns of the photocatalysts were conducted on an XD-3 diffractometer (Beijing Purkinje General Instrument Co., Ltd., China). Fourier transform infrared (FT-IR) spectra were analyzed on an IS10 FT-IR spectrometer (Nicolet, U.S.A.) by mixing the samples with KBr. The morphologies of the photocatalysts were observed through a Quanta 250F field-emission scanning electron microscope (FE-SEM) (FEI, U.S.A.) and JEM-2100 high-resolution transmission electron microscope (HR-TEM) (JEOL, Japan). The energy-dispersive X-ray spectra (EDS) and elemental mappings were also collected on the SEM instrument. X-ray photoelectron spectroscopy (XPS) was applied to collect the electron states on the surface of the catalysts, which were obtained on an ESCALAB 250 spectrometer (Thermo, U.S.A.). Diffuse reflectance spectroscopy (DRS) on a Shimadzu UV-2550 UV–vis spectrophotometer (Shimadzu, Japan) was carried out for the characterization of UV–vis absorption spectra of catalysts. The photoluminescence spectra (PL) were studied by ELabram-HR800.

Photoelectrochemical Analysis

The photoelectrochemical properties of the catalysts were obtained on a CHI 760B electrochemical workstation (CH Instruments Inc., China) with a conventional three-electrode system in which a Pt wire was applied as the counter electrode and an Ag/AgCl electrode (3 M KCl) served as the reference. The electrolyte was obtained by the dissolution of Na2SO4 to form a 0.5 M aqueous solution. The working photoelectrodes were synthesized as follows. Ten milligrams of photocatalyst powder was dispersed in the mixed solution containing 1 mL of ethanol and 10 μL of naphthol. After sufficient dispersion, the suspension solution was dropped on the fluorine-doped tin oxide (FTO) conductive glass. The coated area was fixed at 1 cm2, and the amount of the drop was the same for all samples. After drying at 180 °C for 12 h in a vacuum oven, the final working electrodes were obtained.

Photocatalytic Section

The photodegradation tests of RhB were conducted for all samples at room temperature and atmospheric pressure. Typically, 15 mg of the photocatalyst was dispersed in a 100 mL RhB solution to form the suspension, and the concentration of RhB was set as 25 mg/L. The suspension was blended using a magneton for 0.5 h in the dark to evade the effect of adsorption. Next, the adsorption-balanced suspension was irradiated with 300 W of xenon (Xe) light (CEL-HXF300, Beijing CEAulight Co., Ltd., China) with a light filter of 420 nm, and 6 mL of the suspension was fetched out every 10 min. The suspension was centrifuged to remove the catalyst, and then 4 mL of the supernatant was taken for the following analysis. It is worth mentioning that circulating cooling water was placed around the reactor to avoid the influence of irradiation-generated heat. The RhB concentration in the fetched-out solution was measured via the peak absorption intensity at 553 nm. The blank test was conducted in the absence of a catalyst for comparison.