Coupling with a narrow-band-gap semiconductor for enhancement of visible-light photocatalytic activity: preparation of Bi₂S₃/g-C₃N₄ and application for degradation of RhB

Abstract

A coupled system for the photodegradation of Rhodamine B dye was realized using a Bi₂S₃/g-C₃N₄ composite as a photocatalyst under visible light irradiation. The Bi₂S₃/g-C₃N₄ composite was prepared by a hydrothermal method and characterized by Fourier transform-infrared spectroscopy (FT-IR), X-ray diffraction (XRD), UV-vis diffuse reflectance spectroscopy (DRS), scanning electron microscopy (SEM) and transmission electron microscopy (TEM). Compared with pure g-C₃N₄, the Bi₂S₃/g-C₃N₄ sample exhibits an enhanced photocatalytic activity and the best photocatalytic efficiency is 3.68 times more than that of pure g-C₃N₄. The obtained results indicate that a coupled system of Bi₂S₃ and g-C₃N₄ could overcome the drawback of low photocatalytic efficiency brought by electron–hole recombination and a narrow photoresponse range. On the basis of the corresponding energy band positions, the mechanism of photocatalytic activity enhancement was proposed.

---

1. Introduction

Recently, semiconductor-based photocatalysts have attracted much attention due to their potential employment in organic and inorganic pollutant remediation in wastewater, and water splitting for hydrogen production.^1,2 Among the numerous semiconductors reported, titanium dioxide (TiO₂) is by far the most popular photocatalyst for its higher photocatalytic activity, good photostability, non-toxicity, and low price.³ However, the most widely used TiO₂ photocatalyst is only active under UV irradiation (about 4% of solar photons can be used).^4,5 To effectively eliminate electron–hole recombination in the photocatalytic reaction, many attempts have been made to improve the photocatalytic efficiency. Moreover, upon the viewpoint of utilizing solar light, developing a visible light response photocatalyst is more significant since 45% of the sunlight spectrum is visible light.⁶

Compared with traditional catalyst TiO₂, graphitic-carbon nitride (g-C₃N₄) possesses a proper mid-wide band gap (2.7 eV) to absorb visible light efficiently.⁷ Recent years, g-C₃N₄ has drawn much attention for its high photocatalytic performance for degradation ability of organic pollutants under visible light irradiation.⁸ Unlike conventional organic semiconductor counterparts, g-C₃N₄ exhibits a unique stability, including the heat endurance and chemical resistance. From reports, the as-prepared g-C₃N₄ is non-volatile up to as high as 600 °C, and will be almost completely decomposed until the temperature rises to 700 °C (ref. 9–11) and chemical resistance emerged in that g-C₃N₄ are almost insoluble in water, ethanol, toluene, diethyl ether and tetra hydro furan.¹² The lone pair of nitrogen and electrons delocalization endows the tri-s-triazine derivatives, so g-C₃N₄ with unique electronic structure, which is controllable due to their tunable band gap, and the excellent thermal and chemical stability¹³ make it become one of the most promising semiconductor materials in the exciting research field.

According to the above, good performance seems to have endowed g-C₃N₄ a bright future in the application of catalyst. Unfortunately, two main drawbacks of low separation efficiency of photogenerated electron–hole pairs and narrow visible light response range limit its practical application as an efficient visible-light photocatalyst.^14,15 To improve photocatalytic efficiency, tremendous efforts have been made to improve the efficiency by doping with metal ions or non-metal ions, designing optimizing heterojunctions, and morphological modification. Research by Tian and coworkers showed that the g-C₃N₄/BiIO₄ composite photocatalyst displays a higher photocatalytic activity than the two individuals, which can be attributed to its heterojunction structure.¹⁶ Li and coworkers reported that heterojunctions of graphitic carbon nitride (g-C₃N₄) and Bi₂MoO₆ were synthesized by a hydrothermal method and the heterojunction composites exhibited higher photocatalytic activity than pure g-C₃N₄ or Bi₂MoO₆.¹⁷ Research by Theerthagiri showed that alpha-Fe₂O₃–g-C₃N₄ composite photocatalysts of various compositions by a wet impregnation method and exhibited remarkably improved visible-light induced photocatalytic activity.¹⁸

To overcome the drawbacks of low photocatalytic efficiency brought by electron–hole recombination and narrow photoresponse range, it was proposed that g-C₃N₄ could be coupled with narrow-band-gap semiconductor for enhancement of visible-light photocatalytic activity. In this study, bismuth sulfide (Bi₂S₃) with a narrow band gap of 1.3 eV was prepared and selected as modifier.¹⁹ The photocatalytic activity evaluation was tested by the degradation of Rhodamine B (RhB) under visible light. The influence of Bi₂S₃ on optical property and the photocatalytic activity of g-C₃N₄ were studied. The influence of time on photocatalytic activity and the reaction kinetics of the composites were also investigated. The obtained results may provide an important indication that how to improve the photocatalytic activity under visible light, not only for g-C₃N₄, but also for the other photocatalysts. As far as we know, a novel Bi₂S₃/g-C₃N₄ composite photocatalyst was designed and used for the degradation of organic pollutants for the first time.

2. Experimental section

2.1. Materials

Dicyandiamide (C₂H₄N₄), ethylene glycol (HOCH₂CH₂OH, EG), thiourea (NH₂CSNH₂), bismuth nitrate pentahydrate (Bi(NO₃)₃·5H₂O), nitric acid (HNO₃) and other reagents used in the experiments were obtained from Aladdin Chemical Reagent Co., Ltd. Deionized water was used throughout this study. All chemicals were of analytical grade and were used without further purification.

2.2. Preparation of g-C₃N₄

The graphitic-carbon nitride (g-C₃N₄) samples were prepared by directly heating dicyandiamide.⁴ Typically, 2 g of dicyandiamide powder was put into a quartz crucible with a cover, then heated at a rate of 10 °C min⁻¹ to reach a temperature of 600 °C; and then tempered at this temperature for 2 h in a flowing-nitrogen atmosphere. After the sample was cooled naturally to room temperature, the resulting powder was grinded in the agate mortar and used in subsequent studies.

2.3. Preparation of photocatalyst

In a typical procedure, 200 mg of g-C₃N₄ was dissolved in 30 mL ethylene glycol under vigorous stirring for 20 min to get the dispersed solution, named solution A. A certain amount of Bi(NO₃)₃·5H₂O according to the designed mass ratio of Bi(NO₃)₃·5H₂O to g-C₃N₄ was dissolved in diluted nitric acid to form a clear solution. Meanwhile, moderate thiourea was dissolved in 20 mL deionized water. After that, these two solutions were mixed together and the pH of this mixed solution was adjusted to 7 using certain amounts of NaOH solution, which was named solution B. Then, solution A and solution B were mixed together and was stirred for 2 h. The obtained suspension was transferred into a 100 mL Teflon-lined stainless steel autoclave up to 80% of the total volume, and then was heated to 160 °C and kept at this temperature for 24 h. After being cooled naturally to room temperature, the product was collected and washed using water and ethanol for 3 times, repeatedly, and dried at 60 °C for 12 h. The pure Bi₂S₃ photocatalyst was obtained by the same conditions without adding g-C₃N₄ powder. The process flow chart is illustrated in Scheme 1.

Scheme 1 Preparation process of Bi₂S₃/g-C₃N₄ composite by thermal process.

With the above method, a series of Bi₂S₃/g-C₃N₄ composites with various amounts of Bi₂S₃ were prepared according to different mass ratios of Bi(NO₃)₃·5H₂O to g-C₃N₄. The final products were named as Bi₂S₃/g-C₃N₄-1, Bi₂S₃/g-C₃N₄-2, Bi₂S₃/g-C₃N₄-3, Bi₂S₃/g-C₃N₄-4 and Bi₂S₃/g-C₃N₄-5 samples, respectively (the mass ratios of Bi(NO₃)₃·5H₂O to g-C₃N₄ were 10%, 20%, 30%, 40% and 50%, respectively). In addition, a catalyst with mass ratio of 40% (Bi(NO₃)₃·5H₂O to g-C₃N₄) was prepared by a simple mechanical mixing and named as Bi₂S₃ & g-C₃N₄-4.

2.4. Characterizations

X-ray diffraction (XRD) patterns of samples were scanned on the Shimadzu LabX-6000 X-ray Diffractometer (40 kV, 30 mA) with a Cu Kα radiation source at a scanning rate of 5° min⁻¹ within the range of 5–90°. Fourier transform-infrared spectra (FT-IR) of all the catalysts (KBr pellets) were recorded on the equipment (AVATAR 360, Madison, Nicolet). UV-vis diffuse reflectance spectroscopy (DRS) was carried out on a Hitachi UV-3010 UV-vis spectrophotometer. BaSO₄ was as reference sample. Thermogravimetric analysis (TGA) was performed on a Netzsch STA 449C instrument (NETZSCH Corporation, Germany). The programmed heating range was from room temperature to 800 °C at a heating rate of 10 °C min⁻¹ under air atmosphere. The measurement was taken with 6–10 mg samples. The size and morphology of the nanoparticles were viewed on a PHI-Tecnai 12 transmission electron microscope (TEM). High resolution transmission electron microscopy (HRTEM) was performed on the FEI-TecnaiG2F30S-Twin transmission electron with a Philips CM200 field-emission gun microscope operating at 197 kV. The photocurrents were measured with an electrochemical analyzer (CHI660B, CHI Shanghai, Inc.). Photoluminescence (PL) spectra of the catalyst were measured on the QuantaMaster™ 40 (Photon Technology International, Inc.).

2.5. Photocatalytic activity

The photocatalytic activity of the photocatalyst was confirmed by the degradation of Rhodamine B (RhB) in an apparatus with a tungsten lamp (500 W) as the irradiation source. The visible-light (λ ≥ 420 nm) used in the present study was obtained by the filter with cut-off wavelength of 420 nm. Photocatalyst (50 mg) was dispersed into 100 mL RhB solution with the concentration of 5 mg L⁻¹, and then the mixture was stirred for 30 min in the dark to ensure absorption–desorption equilibrium, after which the reaction suspension was irradiated for 130 min under visible light. The temperature of the suspension was kept at about 25 °C by an external cooling jacket with recycled water. The samples were analyzed every 10 min by UV-vis spectrophotometer at maximum absorption characteristic peak of 553 nm (as shown in Fig. S1, ESI†). A linear calibration curve and correlation coefficient R² = 0.999 are obtained over the range 0 to 5.0 mg L⁻¹ (as shown in Fig. S2, ESI†). Concentration at time t was generally labeled as C. The photocatalytic activity and degradation efficiency were calculated in the form of C/C₀ and (C₀ − C)/C₀, respectively. Where C₀ (mg L⁻¹) is the initial concentration of RhB. C (mg L⁻¹) is the RhB concentration at time t (min).

3. Results and discussion

3.1. Photocatalytic activity comparisons of photocatalysts prepared by mechanical mixing and hydrothermal method

The photocatalytic activity of RhB for blank experiment, Bi₂S₃ & g-C₃N₄-4 (prepared by mechanical mixing method) and Bi₂S₃/g-C₃N₄-4 (prepared by hydrothermal method) samples were investigated (as shown in Fig. S3, ESI†). From the obtained result, the photocatalytic activity of Bi₂S₃/g-C₃N₄-4 is bigger than that of Bi₂S₃ & g-C₃N₄-4. This is mainly because the composition of Bi₂S₃ & g-C₃N₄-4, prepared by the physically mixing, is easy to separate and could not form coupled system. So, series of Bi₂S₃/g-C₃N₄ photocatalysts were selected for the further study.

3.2. XRD analysis

X-ray diffraction (XRD) studies were applied to study the phase, crystallinity and structure of the synthesized materials. XRD patterns of g-C₃N₄ (a), Bi₂S₃ (b) and Bi₂S₃/g-C₃N₄-4 (c) samples are shown in Fig. 1. For g-C₃N₄, a strong peak at 2θ = 27.4° marked with square is observed, which was corresponded to the characteristic interplanar staking peak (002) of aromatic systems.²⁰ For Bi₂S₃/g-C₃N₄-4 sample, the intensity of the peak at 2θ = 27.4° appears an enhancement compared with pure Bi₂S₃, due to the addition of g-C₃N₄. All other prominent peaks at 2θ = 22.19°, 23.57°, 24.95°, 28.58°, 31.72°, 32.97°, 33.85°, 35.73°, 39.96°, 42.72°, 45.48°, 46.61°, 52.53°, 59.15°, 62.55°, 64.89°, 67.60° and 69.31° marked with circle, which were attributed to the (220), (101), (130), (211), (221), (301), (311), (240), (141), (421), (002), (431), (351), (242), (152), (721), (532) and (820) planes of Bi₂S₃, respectively (JCPDS 00-17-0320).²¹

Fig. 1 XRD patterns of g-C₃N₄ (a), Bi₂S₃ (b) and Bi₂S₃/g-C₃N₄-4 (c).

3.3. FT-IR analysis

The chemical structures of g-C₃N₄ and Bi₂S₃/g-C₃N₄ samples were analyzed by FT-IR and the results are shown in Fig. 2, from which it can be seen that the peaks at 1240, 1321, 1408, 1456, 1560 and 1641 cm⁻¹ are contributed to the typical stretching modes of CN heterocycles.²² The absorption peaks near at 1560 and 1641 cm⁻¹ are corresponded to C=N stretching, while the peaks observed at 1240, 1321 and 1408 cm⁻¹ are attributed to aromatic C–N stretching.²³ The band around 806 cm⁻¹ is corresponded to the stretching vibrations of triazine ring.²⁴ A broad band near 3200 cm⁻¹ corresponds to the stretching vibration of O–H of the absorbed water molecule and the stretching modes of terminal NH₂ groups at the defect sites of the aromatic ring.²⁵ It is worth noting that there were no differences of FT-IR peak between bulk g-C₃N₄ and Bi₂S₃/g-C₃N₄ samples, from which it can be seen that the main characteristic peaks of g-C₃N₄ appeared in all Bi₂S₃/g-C₃N₄ photocatalysts. It is clear that the modification with Bi₂S₃ does not alter the FT-IR absorption bands of the g-C₃N₄ obviously.

Fig. 2 FT-IR spectra of g-C₃N₄ (a), Bi₂S₃/g-C₃N₄-1 (b), Bi₂S₃/g-C₃N₄-2 (c), Bi₂S₃/g-C₃N₄-3 (d), Bi₂S₃/g-C₃N₄-4 (e) and Bi₂S₃/g-C₃N₄-5 (f).

3.4. SEM and TEM analysis

To study the morphology of the as-prepared Bi₂S₃/g-C₃N₄ sample, SEM micrographs of pure g-C₃N₄, Bi₂S₃ and Bi₂S₃/g-C₃N₄-4 were taken, as shown in Fig. 3(a)–(c), respectively. As can be seen from Fig. 3(a), pure g-C₃N₄ is composed of big irregular particles with a diameter of micron size.²⁰ The morphology of pure Bi₂S₃ is presented in Fig. 3(b), showing monodisperse nanorods²⁶ with a diameter about 70 nm (seen as the inset of Fig. 3(b)). Fig. 3(c) shows the SEM morphology of Bi₂S₃/g-C₃N₄-4 composite. Different region exhibits the characteristic morphology of g-C₃N₄, and the similar morphology of Bi₂S₃. The corresponding EDS spectrum and result of Bi₂S₃/g-C₃N₄-4 are displayed in Fig. 3(d). It can be seen that the Bi₂S₃/g-C₃N₄-4 consists of C, N, Bi and S.

Fig. 3 SEM images of pure g-C₃N₄ (a), Bi₂S₃ (b) and Bi₂S₃/g-C₃N₄-4 (c) and EDS results (d).

TEM and HRTEM were used to characterize morphology of the samples, and to identify the detailed crystallographic structure and orientation of the Bi₂S₃ nanorod. Fig. 4(a)–(c) showed typical TEM of pure g-C₃N₄, Bi₂S₃/g-C₃N₄-4 and amplification of Bi₂S₃/g-C₃N₄-4. Compared with Fig. 4(a), it can be seen clearly that the TEM of Bi₂S₃/g-C₃N₄-4 is different from that of pure g-C₃N₄. As seen in Fig. 4(b), the structure of Bi₂S₃ is exhibited as nanorods, which keep consistent with that shown in SEM. Fig. 4(d) and (e) presented the HRTEM of Bi₂S₃ nanorod, and clear fringe spacing with an interval of 0.36 nm could be indexed to (130) lattice plane of Bi₂S₃.^22,27 The crystallographic of Bi₂S₃ is consistent with the result of XRD measurement.

Fig. 4 TEM images of pure g-C₃N₄ (a), Bi₂S₃/g-C₃N₄-4 (b), Bi₂S₃/g-C₃N₄-4 (c) and HRTEM images (d and e) of Bi₂S₃ nanorod.

3.5. TG analysis

TGA experiments were carried out on NETZSCH STA449C with the heating rate 10 °C min⁻¹ under air atmosphere. TG tests of single Bi₂S₃, g-C₃N₄ and all Bi₂S₃/g-C₃N₄ samples were carried out; and TG curves are shown in Fig. 5. The first mass loss process is light from the room temperature to 200 °C, which is originating from the removal of adsorbed O₂ and H₂O. From the curve of pure g-C₃N₄, it can be seen that the g-C₃N₄ is fairly stable when the heat temperature is below 600 °C, and total weight loss is nearly 100% when the temperature is up to 750 °C, which implies that g-C₃N₄ can decompose completely. From the Bi₂S₃ curve, it becomes unstable when the temperature is above 280 °C. This may be due to that Bi₂S₃ can decompose and transform into metal Bi, when the sample was heated about 400 °C. For single Bi₂S₃, the Bi content is calculated with the value of 81.3%. Compared with pure g-C₃N₄, Bi₂S₃/g-C₃N₄ samples become unstable when the temperature is above 280 °C. This is mainly because Bi₂S₃ component was decomposed and transform into metal Bi in the former stage; and g-C₃N₄ component of Bi₂S₃/g-C₃N₄ can decompose completely in the later stage when the temperature is up to 750 °C. The residual mass ratio increases with the increasing addition amount of Bi(NO₃)₃·5H₂O. For Bi₂S₃/g-C₃N₄-5 sample preparation process, the mass ratio of Bi(NO₃)₃·5H₂O to g-C₃N₄ is 50% and the metal Bi content in Bi₂S₃/g-C₃N₄-5 sample could be easily calculated with the value of 16.99%. As can be seen from Fig. 5(a) and (b), the residual mass ratio is about 79.8 and 18.1% when the sample was heated over 750 °C. In consideration of small amount of impurity existing, the test result is according with the theoretical calculation value.

Fig. 5 TG curves of pure Bi₂S₃ (a), Bi₂S₃/g-C₃N₄-5 (b), Bi₂S₃/g-C₃N₄-4 (c), Bi₂S₃/g-C₃N₄-3 (d), Bi₂S₃/g-C₃N₄-2 (e), Bi₂S₃/g-C₃N₄-1 (f) and g-C₃N₄ (g).

3.6. Optical properties

The UV-vis diffuse reflection spectra (DRS, A and B) of the pure g-C₃N₄ (a), Bi₂S₃/g-C₃N₄-4 (b) and Bi₂S₃ (c) samples are shown in Fig. 6. From Fig. 6(A), it can be seen that pure g-C₃N₄ has photo-absorption not only in UV light range, but also in visible light range. But the absorption range is relatively narrow, which is less than 500 nm. It is worthy being noted that Bi₂S₃ has strong absorption in nearly the whole range of visible light. Compared with pure g-C₃N₄, the light absorption ability of the composite is significantly enhanced after Bi₂S₃ was introduced, which is attributed to the narrow band gap and large absorption coefficient of Bi₂S₃. As shown in Fig. 6(B), the proposed band gap of Bi₂S₃/g-C₃N₄-4 shifts to lower energy of 2.3 eV with the addition of Bi₂S₃. The band gap of pure g-C₃N₄ (a) is about 2.7 eV, and which of Bi₂S₃ (c) is about 1.3 eV. The results indicate that the introduction of Bi₂S₃ generated an impurity band and narrowed the band gap of g-C₃N₄.

Fig. 6 (A) UV-vis diffuse reflection spectra (DRS) and (B) plots of (Ahν)² versus energy (hν) of pure g-C₃N₄ (a), Bi₂S₃/g-C₃N₄-4 (b) and Bi₂S₃ (c).

Furthermore, it is well known that semiconductors with intrinsic band gap structures possess reliable photoelectric properties related to efficient photo-electric conversion. Transient photocurrent responses for the pure g-C₃N₄ (a) and Bi₂S₃/g-C₃N₄-4 (b) were measured to supply an evidence to support that coupling Bi₂S₃ played an important role in the photocatalytic reaction. From the results shown in Fig. 7, it can be seen that the transient photocurrent response of Bi₂S₃/g-C₃N₄-4 (b) is much higher than that of pure g-C₃N₄ (a), which is strong evidence that g-C₃N₄ coupled with Bi₂S₃, the higher efficient separation efficiency of electron–hole pairs would achieve.

Fig. 7 Transient photocurrent response for the pure g-C₃N₄ (a) and Bi₂S₃/g-C₃N₄-4 (b).

3.7. Photocatalytic activity measurement

Photocatalytic activity of pure g-C₃N₄ and Bi₂S₃/g-C₃N₄ samples with variable Bi₂S₃ amount were tested by degradation of RhB dye under visible light irradiation and the relative removal of RhB dye molecules in terms of irradiation time were shown in Fig. 8. From the results shown in Fig. 8, direct photolysis of RhB is almost negligible and bare g-C₃N₄ showed a relatively low photocatalytic activity mainly through a photosensitive pathway, while RhB dye degradation efficiency is just 26.61%, as shown in Table 1. Bi₂S₃ displayed a great impact on photocatalytic performance among all samples tested. With the increasing of Bi₂S₃ amount, Bi₂S₃/g-C₃N₄ samples display a dramatically enhancement upon identical conditions. Bi₂S₃/g-C₃N₄-4 has a best photocatalytic activity with the degradation efficiency of 98.10%. However, Bi₂S₃/g-C₃N₄-5 has a relative low photocatalytic activity, compared with Bi₂S₃/g-C₃N₄-4. It is probably because the larger Bi₂S₃ leads to the agglomeration among nanoparticles, and leads to its decreasing photocatalytic activity (as shown in Fig. S4, ESI†). The evolutions of absorption spectra of RhB in the presence of Bi₂S₃/g-C₃N₄-4 under visible irradiation with different time are shown in Fig. S5 (ESI†).It is clearly seen that the characteristic absorption band of RhB solution at 553 nm gradually diminishes in intensity with the increasing irradiation time. Moreover, no new absorption peaks appear, indicating the degradation of RhB and no other organic molecules generating. To clarify the final products after decomposition, the total organic carbon (TOC) values of residual RhB after the photodegradation of 5 mg L⁻¹ RhB solution over Bi₂S₃/g-C₃N₄-4 were measured (as shown in Fig. S6, ESI†) and are obviously reduced within 130 min. The result indicates that RhB molecules have been decomposed into CO₂ and H₂O.

Fig. 8 The photodegradation of RhB for blank experiment (a), g-C₃N₄ (b), Bi₂S₃/g-C₃N₄-1 (c), Bi₂S₃/g-C₃N₄-2 (d), Bi₂S₃/g-C₃N₄-3 (e), Bi₂S₃/g-C₃N₄-4 (f) and Bi₂S₃/g-C₃N₄-5 (g) samples.

Table 1 Degradation efficiency and kinetics parameters for RhB degradation with Bi₂S₃/g-C₃N₄

Sample	RhB degradation efficiency (%)	First-order kinetics
		Correlation coefficient (R^2)	Apparent kinetic constant (k, min^−1)
g-C3N4	26.61	0.992	0.00259
Bi2S3/g-C3N4-1	49.29	0.955	0.00571
Bi2S3/g-C3N4-2	61.45	0.948	0.00833
Bi2S3/g-C3N4-3	74.69	0.993	0.0111
Bi2S3/g-C3N4-4	98.10	0.985	0.0310
Bi2S3/g-C3N4-5	87.40	0.988	0.0148

---

The experimental data were fitted with a pseudo-first-order model to study reaction kinetics of RhB degradation. As shown in Fig. 9, the corresponding ln(C₀/C) plot has a good linearity against time, indicating that the visible-light-driven photocatalytic degradation of RhB solutions in the presence of Bi₂S₃/g-C₃N₄ follows the first-order kinetics. From Table 1, all tested samples display a good linearity and possess high correlation coefficient. Bi₂S₃/g-C₃N₄-4 sample shows the largest reaction rate among all Bi₂S₃/g-C₃N₄ samples that is nearly 12.0 times higher than that of pure g-C₃N₄.

Fig. 9 The first-order kinetics of RhB degradation in the presence of g-C₃N₄ (a), Bi₂S₃/g-C₃N₄-1 (b), Bi₂S₃/g-C₃N₄-2 (c), Bi₂S₃/g-C₃N₄-3 (d), Bi₂S₃/g-C₃N₄-4 (e) and Bi₂S₃/g-C₃N₄-5 (f) samples.

3.8. Photocatalytic mechanism of Bi₂S₃/g-C₃N₄ photocatalyst

To improve photogenerated carriers separation and enhance the efficiency of the interfacial charge transfer, one of the most valuable is to use two semiconductors in contact with different redox energy levels of conduction band (CB) and valence band (VB). In this study, g-C₃N₄ was coupled with narrow-band-gap semiconductor (Bi₂S₃) for enhancement of visible-light photocatalytic activity. It is generally accepted that the relative positions of energy bands of g-C₃N₄ (E_VB = 1.57 eV, E_CB = −1.13 eV)²⁸ and Bi₂S₃ (E_VB = 1.42 eV, E_CB = 0.12 eV)¹⁹ are shown in Scheme 2. When the Bi₂S₃/g-C₃N₄ composites are irradiated under visible light, both g-C₃N₄ and Bi₂S₃ can easily absorb visible light and be excited to produce photogenerated electron–hole pairs. The photogenerated holes (h⁺) in VB of g-C₃N₄ and Bi₂S₃ can directly oxidize RhB. The photogenerated electrons (e⁻) in CB of g-C₃N₄ and Bi₂S₃ can be captured by oxygen adsorbed on the surface of Bi₂S₃/g-C₃N₄ composites to generate ˙O²⁻, further to form ˙OH, which would be responsible for the degradation of RhB.

Scheme 2 Schematic diagram of the separation and transfer of photogenerated charges in the Bi₂S₃/g-C₃N₄ photocatalyst under visible light irradiation.

According to the relative positions of energy bands, the photogenerated holes (h⁺) in VB of g-C₃N₄ are easily be transferred to Bi₂S₃. However, the photogenerated electrons in CB of g-C₃N₄ are difficultly injected into Bi₂S₃ because of the potential difference between CB of g-C₃N₄ (−1.13 eV) and Bi₂S₃ (0.12 eV).^29,30 Due to g-C₃N₄ coupled with Bi₂S₃, h⁺ in VB of g-C₃N₄ are easily be transferred to Bi₂S₃, which is the main cause for the efficient separation of the photogenerated electron–hole pairs in g-C₃N₄. Hence, this coupled system effectively reduces the recombination of photoinduced electrons and holes, resulting in enhanced photodegradation efficiency of g-C₃N₄.

The photoluminescence (PL) spectrum is an effective approach to evaluate the separation capacity of the photogenerated charge carriers. In order to prove the effective separation of the photogenerated charges, PL spectra of g-C₃N₄ (a), Bi₂S₃ (b) and Bi₂S₃/g-C₃N₄-4 (c) were shown in Fig. 10. The higher PL intensity means more efficient carriers participate in the recombination. On the contrary, the lower PL intensity means more carriers participate in photocatalytic process. As can be seen from Fig. 10(a) and b), there is a strong emission peak for g-C₃N₄ and a relative low emission peak for Bi₂S₃ at around 482 nm, which could be related to the recombination of the photoexcited electron–hole. However, PL intensity of Bi₂S₃/g-C₃N₄-4 is lower than that of g-C₃N₄ and Bi₂S₃, which suggests the improved charge carrier separation inside the composite.

Fig. 10 PL spectra of pure g-C₃N₄ (a), Bi₂S₃ (b) and Bi₂S₃/g-C₃N₄-4 (c).

3.9. Reusability studies

The reusability is an important parameter of the photocatalytic process. In order to examine the photocatalytic stability of the photocatalyst, Bi₂S₃/g-C₃N₄-4 was selected for the recycling experiment and the result is shown in Fig. 11. As can be seen from Fig. 11(A), the degradation efficiency of RhB is about 96.15% in the fifth run. Therefore, it can be concluded that the photocatalyst has a good stability in the experimental conditions. The XRD patterns of Bi₂S₃/g-C₃N₄-4 were examined further to verify durability before and after the photocatalytic reaction. As shown in Fig. 11(B), the XRD patterns are almost identical, indicating that the sample is stable under light irradiation and has the ability to be reused.

Fig. 11 Reuse experiments of the Bi₂S₃/g-C₃N₄-4 in the photodegradation of RhB under visible light irradiation (A), and the XRD patterns of Bi₂S₃/g-C₃N₄-4 before and after photocatalytic reaction (B).

4. Conclusions

In summary, visible-light-driven Bi₂S₃/g-C₃N₄ composites were prepared by a hydrothermal method and studied as photocatalysts for degradation of RhB dye molecules. The photocatalytic activity of the composites increased then decreased with increasing Bi₂S₃ content and Bi₂S₃/g-C₃N₄-4 presents the best photocatalytic efficiency, which is more than 3.68 times that of pure g-C₃N₄. Moreover, from the study of reaction kinetics, it can be seen that Bi₂S₃/g-C₃N₄-4 sample showed the largest reaction rate among all Bi₂S₃/g-C₃N₄ samples which is nearly 12.0 times higher than that of pure g-C₃N₄. On the basis of mechanism discussion, the enhanced activity is attributed to the coupled system which offers a wide response wavelength range and the effective separation of electron–holes pairs. Hence, this study has a guiding significance for the design of coupled system and has potential in environmental remediation applications.

Acknowledgements

This project was supported by the Innovation Program for Graduate Education of Jiangsu Province (KYLX_1063), the Natural Science of Jiangsu Province (BK20141298) and the Society Development Fund of Zhenjiang (SH2013020).

References

1. D. J. Martin, P. J. T. Reardon, S. J. A. Moniz and J. W. Tang, J. Am. Chem. Soc., 2014, 136, 12568–12571.
2. V. M. D. Rocha, M. D. Pereira, L. R. Teles and M. O. D. Souza, Mater. Sci. Eng., B, 2014, 185, 13–20.
3. M. D. Hernandez-Alonso, F. Fresno, S. Suarez and J. M. Coronado, Energy Environ. Sci., 2009, 2, 1231–1257.
4. M. Karmaoui, D. M. Tobaldi, A. S. Skapin, R. C. Pullar, M. P. Seabra, J. A. Labrincha and V. S. Amaral, RSC Adv., 2014, 4, 46762–46770.
5. S. Z. Hu, L. Ma, J. G. You, F. Y. Li, Z. P. Fan, G. Lu, D. Liu and J. Z. Gui, Appl. Surf. Sci., 2014, 311, 164–171.
6. H. X. Zhao, H. T. Yu, X. Quan, S. Chen, Y. B. Zhang, H. M. Zhao and H. Wang, Appl. Catal., B, 2014, 152, 46–50.
7. M. Shalom, M. Guttentag, C. Fettkenhauer, S. Inal, D. Neher, A. Llobet and M. Antonietti, Chem. Mater., 2014, 26, 5812–5818.
8. Y. G. Xu, H. Xu, J. Yan, H. M. Li, L. Y. Huang, J. X. Xia, S. Yin and H. M. Shu, Colloids Surf., A, 2013, 436, 474–483.
9. J. H. Yang, X. T. Wu, X. F. Li, Y. Liu, M. Gao, X. Y. Liu, L. N. Kong and S. Y. Yang, Appl. Phys. A: Mater. Sci. Process., 2011, 105, 161–166.
10. Y. G. Li, J. A. Zhang, Q. S. Wang, Y. X. Jin, D. H. Huang, Q. L. Cui and G. T. Zou, J. Phys. Chem. B, 2010, 114, 9429–9434.
11. S. C. Yan, Z. S. Li and Z. G. Zou, Langmuir, 2009, 25, 10397–10401.
12. E. G. Gillan, Chem. Mater., 2000, 12, 3906–3912.
13. J. X. Xia, J. Di, S. Yin, H. M. Li, H. Xu, L. Xu, H. M. Shu and M. Q. He, Mater. Sci. Semicond. Process., 2014, 24, 96–103.
14. A. Thomas, A. Fischer, F. Goettmann, M. Antonietti, J. O. Muller, R. Schlogl and J. M. Carlsson, J. Mater. Chem., 2008, 18, 4893–4908.
15. J. Chen, S. H. Shen, P. H. Guo, M. Wang, P. Wu, X. X. Wang and L. J. Guo, Appl. Catal., B, 2014, 152, 335–341.
16. N. Tian, H. W. Huang, Y. He, Y. X. Guo and Y. H. Zhang, RSC Adv., 2014, 4, 42716–42722.
17. H. P. Li, J. Y. Liu, W. G. Hou, N. Du, R. J. Zhang and X. T. Tao, Appl. Catal., B, 2014, 160–161, 89–97.
18. J. Theerthagiri, R. A. Senthil, A. Priya, J. Madhavan, R. J. V. Michael and M. Ashokkumar, RSC Adv., 2014, 4, 38222–38229.
19. Z. J. Zhang, W. Z. Wang, L. Wang and S. M. Sun, ACS Appl. Mater. Interfaces, 2012, 4, 593–597.
20. Y. M. He, J. Cai, T. T. Li, Y. Wu, H. J. Lin, L. H. Zhao and M. F. Luo, Chem. Eng. J, 2013, 215, 721–730.
21. S. Vadivel and V. P. Kamalakannan, Ceram. Int., 2014, 40, 14051–14060.
22. F. Jiang, T. T. Yan, H. Chen, A. W. Sun, C. M. Xu and X. Wang, Appl. Surf. Sci., 2014, 295, 164–172.
23. K. Dai, L. H. Lu, C. H. Liang, Q. Liu and G. P. Zhu, Appl. Catal., B, 2014, 156, 331–340.
24. X. S. Rong, F. X. Qiu, J. Qin, J. Yan, H. Zhao and D. Y. Yang, J. Ind. Eng. Chem., 2014, 20, 3808–3814.
25. L. Y. Chen and W. D. Zhang, Appl. Surf. Sci., 2014, 301, 428–435.
26. Y. F. Luo, H. Chen, X. Li, Z. Q. Gong, X. J. Wang, X. F. Peng, M. D. He and Z. Z. Sheng, Mater. Lett., 2013, 105, 12–15.
27. Y. M. Cui, Q. F. Jia, H. Q. Li, J. Y. Han, L. J. Zhu, S. G. Li, Y. Zou and J. Yang, Appl. Surf. Sci., 2014, 290, 233–239.
28. L. Y. Huang, H. Xu, R. X. Zhang, X. N. Cheng, J. X. Xia, Y. G. Xu and H. M. Li, Appl. Surf. Sci., 2013, 283, 25–32.
29. H. Katsumata, Y. Tachi, T. Suzuki and S. Kaneco, RSC Adv., 2014, 4, 21405–21409.
30. D. S. Wang, H. T. Sun, Q. Z. Luo, X. L. Yang and R. Yin, Appl. Catal., B, 2014, 156, 323–330.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra15715j

---