Controllable synthesis of Bi₂S₃/CuS heterostructures by an in situ ion-exchange solvothermal process and their enhanced photocatalytic performance

Abstract

A novel Bi₂S₃/CuS hybrid photocatalyst with a 3D hierarchical configuration has been synthesized through an in situ solution-based cation exchange reaction that creates a heterojunction between a single-crystalline Bi₂S₃ nanotube/nanorod core and CuS nanoplates shell. The crystal structure, optical absorption and morphology evolutions of the nanocomposites with different Bi₂S₃ to CuS ratios were investigated systematically, and a possible formation process of the heterostructure was proposed. The Bi₂S₃/CuS hierarchical nanocomposite photocatalysts exhibit a broad range of absorption wavelength and good visible-light-driven photocatalytic activity for the degradation of rhodamine B (RhB) aqueous solution, compared with bare orthorhombic Bi₂S₃ nanotubes/nanorods. The enhanced photocatalytic performance of the Bi₂S₃/CuS composites could be attributed to the effective electron–hole separation at the interface of these two semiconductors, as well as wider and more intensified light absorption of the solar spectrum. The present study provides helpful insight into rational design and fabrication of novel and efficient chalcogenide-based heterostructure photocatalysts.

---

1. Introduction

Semiconductor photocatalysis technology, which provides a relatively simple and environmentally friendly route for directly converting solar energy into chemical energy, has received tremendous attention to tackle energy shortage and environmental pollution issues.^1–3 Currently, the most studied photocatalysts are primarily oxide semiconductors, exemplified by TiO₂,⁴ ZnO,⁵ Cu₂O,^6,7 SnO₂,⁸ SrTiO₃,⁹ BiVO₄,^10,11 Bi₂WO₆,^12,13 BiFeO₃,^14,15 and CeO₂,¹⁶ etc. However, the choice of available metal oxide materials with desirable band gaps ranging between 1.5 and 2.2 eV is still limited, and thus new semiconductors with a narrow band gap for efficient photocatalysis are needed for real applications. To this end, a number of sulfides, nitrides, oxynitrides and transition-metal chalcogenides have received considerable investigations as alternative materials for visible-light-driven photocatalysis.^17–26 Among them, metal sulfides have proved to be a group of highly efficient catalysts for photochemical reactions, as well as serving as cocatalysts for active sites for reducing or oxidizing.^19,27,28 As well-known sulfide semiconductors, both copper(II) sulfide (CuS) and bismuth sulfide (Bi₂S₃) have attracted intensive investigations for diverse applications including electric and optoelectronic devices,^29,30 thermoelectric devices,^31,32 hydrogen storage materials,³³ and high-capacity cathode material in lithium secondary batteries.³⁴ More recently, many researches were focused on the photocatalytic properties of CuS and Bi₂S₃ nanostructures, exhibiting promising perspective in this field.^35–37

It is generally believed that the use of heterojunction photocatalysts combining multi-components is an effective strategy for improving their photocatalytic performance, resulting from dramatic improvement of the separation of photogenerated electrons and holes, as well as the ability of absorbing a wider range of sunlight compared with the single component.³⁸ In this regard, various CuS and Bi₂S₃ based semiconductor composites such as Bi₂S₃/Bi₂WO₆,³⁹ Bi₂S₃/CdS,^19,40 Bi₂S₃/ZnS,⁴¹ CuS/ZnS,⁴² etc. have been explored more recently, which can enhance the visible-light absorption and photocatalytic properties effectively. Many different combinations of configurations may form when two components with various morphologies are integrated into hybrid structures, which would influence the performance of the hybrid photocatalysts.⁴³ The criteria of a well-designed geometry architecture of heterojunction photocatalysts includes broad light absorption, efficient separation and transfer of photogenerated electrons and holes, and large surface area exposed to redox reactions. Although Bi₂S₃/CuS nanocomposites with sphere shape were reported,^44,45 novel designs and configurations that meet those requirements, as well as new synthetic strategy under mild reaction conditions that can obtain designed morphologies are still lacking for sulfide based semiconductor systems for use as high efficient visible-light-driven photocatalysts.

Herein, we have designed and synthesized a new Bi₂S₃/CuS heterojunction photocatalyst with a 3D hierarchical configuration via a facile in situ ion-exchange solvothermal method. Such a 3D hierarchical structure consists of a core of Bi₂S₃ nanotube/nanorod and shell of CuS nanoplates. Compared with bare Bi₂S₃ nanotubes/nanorods, those covered by CuS nanoplates exert more than 3 times higher photocatalytic performance for degradation of RhB solution. The mechanism of enhanced photocatalytic activity has also been discussed based on the band structure of the Bi₂S₃/CuS heterojunctions and experimental results.

2. Experimental methods

Preparation of Bi₂S₃ nanotubes/nanorods

Bi₂S₃ nanotubes/nanorods were synthesized by homogeneous precipitation through a solvothermal reaction, as reported in our previous work.³² Typically, 3.75 mmol Bi(NO₃)₃·5H₂O was firstly added into 25 mL ethanediol and stirred for 20 min to get solution A. Meanwhile, 5.625 mmol Na₂S was introduced to 30 mL de-ionized water and stirred for 10 min, getting solution B. Then solution B was dripped slowly into solution A. A large number of black suspended matters were formed in the mixed solution, in which 0.032 mol CO(NH₂)₂ and 20 mL de-ionized water were added subsequently. The mixed solution was then transferred into a Teflon-lined stainless steel autoclave (100 mL capacity), sealed and maintained at 120 °C for 12 h. The resultant black solid product was filtered, washed with de-ionized water and ethanol for three times respectively and dried in air at 60 °C.

Preparation of Bi₂S₃/CuS composites

As a typical synthesis of Bi₂S₃/CuS composite, 0.50 g of the as-obtained Bi₂S₃ powder was mixed with a solution of designed stoichiometric amount of Cu(NO₃)₂·3H₂O dissolved in 60 mL ethanediol. After stirring for 1 h, the mixture was transferred into a Teflon-lined stainless steel autoclave (100 mL capacity), sealed and maintained at 140 °C for 3 h. Finally, the precipitates of Bi₂S₃/CuS with different theoretical mass percentages (10%, 20% and 40%) of CuS to the total weight of CuS and Bi₂S₃ (denoted as 10-CuS/Bi₂S₃, 20-CuS/Bi₂S₃, 40-CuS/Bi₂S₃, respectively) were collected, washed with de-ionized water and ethanol for three times respectively and dried in air at 60 °C.

Preparation of CuS nanoplates

CuS nanoplates were synthesized by a solvothermal method for comparison. 3 mmol Cu(NO₃)₂·3H₂O, 12 mmol CS(NH₂)₂ and 0.4 g PVP-K30 were introduced to 50 mL ethanediol with vigorous stirring for 1 h. The mixed solution was then transferred into a Teflon-lined stainless steel autoclave (100 mL capacity), sealed and maintained at 150 °C for 2 h. The resultant black solid product was filtered, washed with de-ionized water and ethanol for three times respectively and dried in air at 60 °C.

Characterization techniques

The crystal structure was analyzed by X-ray diffraction (XRD: D/max-RB, Rigaku Inc., Japan) with Cu Kα radiation (λ = 0.15406 nm) filtered through a Ni foil. The morphology and structure of the samples were studied by field emission scanning electron microscopy (FESEM: SUPRA™ 55, Carl Zeiss, Nakano, Japan) and transmission electron microscopy (TEM: Philip Tecnai F20, Dutch) equipped with an energy dispersive X-ray spectrometer (EDX). Diffuse reflectance UV-vis spectra were recorded by means of UV-visible spectrophotometer (UV-2800, UNICO Instruments Co., Ltd., China) equipped with an integration sphere. The band gaps of the as-prepared samples were estimated using the equation of E_g = 1240/λ based on their absorption edges. The photoluminescence (PL) spectra of the as-obtained samples were measured using fluorescence spectrophotometer (F-4500) with 325 nm excitation at room temperature. The electron spin resonance (ESR) spectra were measured by a Bruker model ESR 300E spectrometer equipped with a quanta-Ray Nd:YAG laser system as the irradiation source (λ = 355 nm, 10 Hz) using 5,5′-dimethyl-1-pyrroline-N-oxide (DMPO) as spin-trap reagent. And the corresponding parameters, such as center field, sweep width, microwave frequency, modulation frequency and power, for the spectrometer were set as 3486.70 G, 100 G, 9.82 GHz, 100 kHz and 5.05 mW, respectively.

The photocatalytic activity was evaluated by the degradation of RhB under visible light irradiation using a 500 W Xe lamp (Beijing institute of electrical light sources, China) with an optical cut-off filter (λ > 400 nm). In a typical experiment, 200 mL RhB solution (10 mg L⁻¹) was mixed with 0.1 g as-obtained powders. The mixture was firstly placed in the dark to reach the adsorption/desorption equilibrium between the photocatalysts and RhB before illumination. The photocatalytic test system was maintained at room temperature by circulation of water through an external cooling coil. At given intervals of illumination, ca. 5 mL of the mixture solution were taken out and centrifuged. The degradation of RhB was monitored from the intensity of absorption peak of the supernatant RhB (560 nm) relative to its initial intensity using the UV-vis spectrophotometer under various illumination time on the basis of the following formula: η = (A₀ − A_t)/A₀ × 100%, where η, A₀ and A_t stand for the decolourization ratio of the reaction, the absorbance of RhB solution before reaction and the absorbance of RhB solution after reacting for t min, respectively.

3. Results and discussions

Fig. 1 shows the XRD patterns of bare Bi₂S₃ and Bi₂S₃/CuS heterostructures with different Cu(NO₃)₂ contents, which was serving as raw materials for synthesizing CuS nanoplate shell. All the diffraction peaks for bare Bi₂S₃ sample in Fig. 1A can be well indexed to orthorhombic Bi₂S₃ (PDF #84-0279). Diffraction patterns for the sample 10-CuS/Bi₂S₃ show only Bi₂S₃ phase without any other detectable peaks, thus it is hard to judge the existence of CuS phase. Besides orthorhombic Bi₂S₃ phase, hexagonal CuS (PDF #78-2121) was also observed in the samples of 20-CuS/Bi₂S₃ and 40-CuS/Bi₂S₃. No peaks from other impurity phase was detected for all the samples. Apart from the phase characteristics, the diffraction peak shifting was also noticed by highlighting the strongest peak of these two phases ((121) peak of Bi₂S₃ and (110) peak of CuS, abbreviated as 2θ₍₁₂₁₎ and 2θ₍₁₁₀₎, respectively), as shown in Fig. 1B and C. Different from the identical 2θ₍₁₂₁₎ between bare Bi₂S₃ and its standard card (PDF #84-0279) in Fig. 1B, the 2θ₍₁₂₁₎ for 10-CuS/Bi₂S₃ shifts to higher diffraction angles by ca. 0.13°. As for 20-CuS/Bi₂S₃ with detectable CuS phase, the 2θ₍₁₂₁₎ of Bi₂S₃ continues to shift to higher angles by ca. 0.18° in comparison to bare Bi₂S₃ (Fig. 1B), along with a similar shift of ca. 0.18° to higher diffraction angles than the 2θ₍₁₁₀₎ of CuS (PDF #78-2121). In contrast, both the 2θ₍₁₂₁₎ (Fig. 1B) and 2θ₍₁₁₀₎ (Fig. 1C) in 40-CuS/Bi₂S₃ exhibit an opposite shift towards lower diffraction angles. After further increasing CuS content, stable Bi₂S₃/CuS hybrid structure could not be obtained while other phases such as Cu_7.2S₄ and Cu₄Bi₇S₁₂ appeared. The shifts of XRD peaks and corresponding changes on the lattice parameters will be discussed in detail later.

Fig. 1 XRD patterns for bare Bi₂S₃ and Bi₂S₃/CuS heterostructures in the range of (A) 13–65°, (B) 28–29.3° and (C) 47–48.8°.

Fig. 2 shows the FESEM images of bare Bi₂S₃ and Bi₂S₃/CuS heterostructure composites. As shown in Fig. 2A and B, bare Bi₂S₃ powder is composed of nanorod- and nanotube-like structures whose surfaces are relatively smooth with diameter in the range of 50 to 150 nm, which was confirmed to be a single crystal grown along [001] direction in our previous work.³² In the case of 10-CuS/Bi₂S₃, the surface of the nanotubes/nanorods became coarse with a large amount of tiny plate-shaped structures growing vertically on it. The EDS line scan profile in Fig. 2D suggests that Bi, Cu and S elements are uniformly dispersed throughout the nanotube/nanorod. As for 20-CuS/Bi₂S₃ in Fig. 2E, the plate-shaped structures on the surface continued to grow on the nanotube/nanorod along with an increased density. The EDS linear scan from the central part to the edge of the nanotube/nanorod in Fig. 2F shows that Cu element distribute steadily, while the concentration of S and Bi elements decreases. This phenomenon can be precisely explained by the formation of CuS on the surface of Bi₂S₃ nanotubes. In the central part of the nanotube/nanorod, the core (Bi₂S₃) and shell (CuS) of the composites can be detected simultaneously by EDS linear scanning, leading to a uniform dispersion of the three elements. However, at the edge only CuS nanoplates (inset of Fig. 2F) can be exposed to the detections. Therefore, due to the disappearance of Bi₂S₃ phase, Bi and S elements showed a decreased tendency. As for 40-CuS/Bi₂S₃ (Fig. 2G), the surface of the Bi₂S₃ nanotube/nanorod experienced a smoothing tendency, and a number of free standing hexagonal plates appeared, demonstrating splitting of the CuS nanoplates from the Bi₂S₃ nanotubes. Furthermore, the EDS spot scan (Fig. 2H) indicates that the free standing CuS hexagonal plates are rich in Cu while lack in S element. The splitting of CuS and Bi₂S₃ phases leads to non-stoichiometric S/Cu and S/Bi ratios, compared to that of stoichiometric CuS and Bi₂S₃ phases.

Fig. 2 FE-SEM images and EDS line-scan analysis of as-prepared samples: (A and B) bare Bi₂S₃; (C and D) 10-CuS/Bi₂S₃; (E and F) 20-CuS/Bi₂S₃; (G and H) 40-CuS/Bi₂S₃.

Fig. 3 shows a typical TEM image of an individual hybrid Bi₂S₃/CuS nanostructure (20-CuS/Bi₂S₃) in which hexagon-shaped nanoplates cover the nanorod/nanotube, consistent with the FESEM analysis. The high-resolution TEM (HRTEM) images reveal the interplanar spacing of 0.329 nm for the nanoplates (Fig. 3B) and 0.372 nm for the nanorod (Fig. 3C), corresponding to the (100) plane of CuS and the (110) plane of Bi₂S₃, respectively. In addition, the STEM-EDX mapping of an individual Bi₂S₃/CuS heterostructure (Fig. 3D–G) demonstrates that Cu and S elements have a uniform distribution in both the nanorod/nanotube and nanoplates, while Bi element mainly concentrates in the nanorod/nanotube. These results confirm the formation of hierarchical Bi₂S₃/CuS heterostructures consisting of CuS nanoplates shell and Bi₂S₃ nanorod/nanotube core.

Fig. 3 (A) Low-magnification TEM image of an individual hybrid Bi₂S₃/CuS structure (inset shows the SAED patterns). HRTEM images of the nanoplate shell (B) and nanotube core (C). STEM-EDX maps of the Bi₂S₃/CuS heterostructure: (D) overlay, (E) Cu, (F) Bi and (G) S elements.

Based on the experimental results, a possible formation process of the 3D hierarchical Bi₂S₃/CuS heterostructure nanocomposite is proposed, as illustrated in Fig. 4. So as to realize the formation of CuS phase from Bi₂S₃, the substitutional reaction between Cu²⁺ and Bi³⁺ was adopted. CuS has a lower solubility than Bi₂S₃ in ethanediol, thus Cu²⁺ are more likely to replace Bi³⁺ in Bi₂S₃ to form CuS phase under a proper reaction condition (solvothermal reaction at 140 °C in the present work). When a small amount of Cu(NO₃)₂ (for instance 10-CuS/Bi₂S₃) was introduced to the Bi₂S₃ system, the diffraction peaks (θ₍₁₂₁₎) of Bi₂S₃ shift to higher diffraction angles (Fig. 1B), indicating the substitution of Cu²⁺ for Bi³⁺ in the Bi₂S₃ host phase before forming CuS. And the partial substitution of Bi³⁺ (r = 0.96 nm) by Cu²⁺ (r = 0.72 nm) with smaller ionic radius at the interface of the Bi₂S₃/CuS composites during the ion-exchange substitutional reaction process will result in a shrinked Bi₂S₃ lattice, as expressed by the following defect equation:

Fig. 4 Schematic formation process of the Bi₂S₃/CuS heterostructures.

As further increasing the Cu(NO₃)₂ content, Bi₂S₃ diffraction peaks continue shifting to higher diffraction angles (20% CuS doping), suggesting an ongoing substitution of Bi³⁺ by Cu²⁺. Meanwhile, CuS nanoplates appeared on the surface of the Bi₂S₃ nanotubes/nanorods, as shown in Fig. 2E and F for the 20-CuS/Bi₂S₃ sample, suggesting that the reaction has reached the solid solubility limit for Cu²⁺ into Bi₂S₃ lattice, and CuS phase started to form at the same time. However, the XRD peaks (θ₍₁₁₀₎) of CuS phase shift to the higher diffraction angles (Fig. 1C), indicating that a typically insufficient reaction process occurred in copper sulfide systems at this moment, ascribed to the lack of Cu²⁺ in the CuS phase:

With adding more Cu²⁺ (up to 40%), the rapid growth of CuS phase resulted in the splitting of CuS from the surface of Bi₂S₃, leaving a smoother surface of the nanotube (Fig. 2G and H) and some free standing CuS nanoplates. The enlarged lattices of both Bi₂S₃ and CuS phases in 40-CuS/Bi₂S₃ evidenced by the XRD results (Fig. 1B and C) could be attributed to the lack of S²⁻ owing to the excessive consumption during the reaction between the Bi₂S₃ and Cu(NO₃)₂ solution, providing that their initial stoichiometric ratio was constant:

The lack of S²⁻ in CuS phases was also confirmed by the EDS spot scan in Fig. 2H as discussed above. Therefore, we can conclude that the Bi₂S₃/CuS composites were formed through the ion-exchange process between Bi₂S₃ and Cu²⁺ ions. The formation process and mechanism are similar with reported works in other sulfate based heterostructure systems such as CuS/ZnS and Bi₂S₃/ZnS.^41,42

Fig. 5 presents the UV-vis diffuse reflectance spectra (DRS) of different Bi₂S₃/CuS heterostructures, as well as that of pure Bi₂S₃ and CuS for comparison. A broad absorption spectra located at ca. 650 nm was observed for bare Bi₂S₃, being light-absorbable in the visible range. Due to the intrinsic absorption of CuS, pure CuS sample exerts two absorption peaks around 300 nm and 650 nm. Therefore, after introducing CuS, another obvious absorption peak appeared around 300 nm in the samples of 20-CuS/Bi₂S₃ and 40-CuS/Bi₂S₃, which can be attributed to the intrinsic absorption transition of CuS. No CuS intrinsic absorption peaks appeared in the 10-CuS/Bi₂S₃ sample, suggesting the absence of CuS phase, which well corresponds to both of the XRD and FESEM/EDS results. It is also worth noting that the band transition edge of the heterostructure shows a red-shift trend compared to that of pure Bi₂S₃, covering the entire visible light range. This might be the opposite effects of Burstein–Moss Shift (BM-Shift) resulting from the doping by Cu²⁺ into the lattice of Bi₂S₃. Doping Cu²⁺ will produce holes in the Bi₂S₃ (n-type) lattice which can reduce carrier (electrons) density in the conduction band of Bi₂S₃ and further reduce the bandgap of Bi₂S₃.⁴⁶ Moreover, an enhancement in the intensity of absorption peak in the whole absorption range was observed, which arises from the cavity-mirror effect of Bi₂S₃/CuS that can allow multiple reflections of visible light.^35,47 Serving as the “mirrors”, CuS nanoplates on the surface can reflect the incident light many times and enhance the photoabsorption of the nanocomposite photocatalyst. From the absorption edge of pure Bi₂S₃ (ca. 838 nm) and the band transition edge originated from CuS in the sample 20-CuS/Bi₂S₃ (ca. 720 nm), the band gap of Bi₂S₃ and CuS nanoplates could be estimated to be 1.48 and 1.72 eV, respectively, using the equation of E_g = 1240/λ. The band gap value of CuS is comparable with our previous results in CuS nanodisks, which has very similar morphology with the nanoplates in the present study.⁴⁷ These results demonstrate that the Bi₂S₃/CuS composites could achieve efficient utilization of solar spectrum and therefore can be expected to show improved visible-light-driven photocatalytic performance.

Fig. 5 UV-vis diffuse reflectance spectra of bare Bi₂S₃ and different Bi₂S₃/CuS heterostructures.

The photocatalytic degradation of RhB is chosen as a model to investigate the photocatalytic activity of the Bi₂S₃/CuS nanocomposite under visible light irradiation. In the light of the Lambert–Beer law, the actual concentration changes of RhB are directly proportional to the normalized absorption value during the photodegradation at given time intervals within certain concentrations (the initial concentration C₀ = 10 mg L⁻¹ in the present study). As shown in Fig. 6, the self-degradation of RhB solution was negligible under visible light (line A). As for the pure Bi₂S₃ nanotubes and CuS nanoplates sample, only about 15.6% and 11.3% of RhB can be degraded (line B & line F) respectively after 5 h illumination. Compared with pure Bi₂S₃, the decomposition efficiency of sample 10-CuS/Bi₂S₃ experienced a slight downward tendency which only reached ca. 14.1% at the same condition (line C). In contrast, enhanced photocatalytic performance was recorded for 20-CuS/Bi₂S₃ and 40-CuS/Bi₂S₃ (line D & E) in comparison to bare Bi₂S₃ and CuS, especially for 40-CuS/Bi₂S₃, which can degrade more than 45% of RhB under the same condition, 3 times higher than that of pristine sample.

Fig. 6 Photodegradation efficiencies of RhB as a function of irradiation time under visible light irradiation for all samples: bare Bi₂S₃, 10-CuS/Bi₂S₃, 20-CuS/Bi₂S₃ and 40-CuS/Bi₂S₃, as well as CuS nanoplates for comparison.

Because photoluminescence (PL) emission mainly results from the recombination of excited carriers, PL spectra is a useful technique to investigate the separation efficiency of the photogenerated charge carriers in a semiconductor. It is generally believed that a lower PL emission intensity is an indication of a lower recombination rate of photogenerated electron and hole (e⁻–h⁺) pairs.⁴³ The PL spectra of all the samples under the excitation wavelength of 325 nm in Fig. 7A show two emission peaks centered at ca. 368 nm and 460 nm. For bare Bi₂S₃, the well-defined PL peaks suggest remarkable charge recombination. By contrast, the PL intensity weakened as increasing the CuS content from 0 to 40 wt%, demonstrating that the recombination rate of photogenerated e⁻–h⁺ pairs is impeded after introducing CuS on the surface of the Bi₂S₃ nanotubes, leading to the enhanced photocatalytic activity. It is worth mentioning that no PL peak from CuS was observed. On one hand, this is possibly due to the transfer of photogenerated electrons/holes from CuS with small loading amount to the Bi₂S₃ host, resulting in high separation rate of e⁻–h⁺ pairs in the CuS. Similar phenomenon has been reported in Bi₂S₃/Bi₂WO₆,⁴⁸ Bi₂S₃/BiVO₄ (ref. 49) and CuS/TiO₂ (ref. 50) systems, in which only peaks from host Bi₂WO₆, BiVO₄ and TiO₂ appeared in the PL spectrum. On the other hand, it has also been reported that CuS and Bi₂S₃ show the PL peaks with similar positions, which may be overlapped by each other in the present work.^51–54 For the 10% sample, despite of reduced photocatalytic performance, it also exerts a suppressed charge recombination process indicated by the weaker PL intensity. As an n-type semiconductor, the introduction of Cu²⁺ into Bi₂S₃ lattices will reduce the carrier (electron) concentration (eqn (1)), which could suppress the PL intensity to some extent.⁵⁵ To further confirm the photoexcitation process for 40% sample, electron spin resonance (ESR) technique was adopted to monitor the reactive oxygen species generated under visible light irradiation. As shown in Fig. 7B, after 30 min, signals of DMPO–˙OH could be clearly detected, which suggests that e⁻–h⁺ pairs could be produced efficiently by the heterostructures under visible light illumination.^56,57 At the same time, the charge separation process is long enough for the photoinduced electrons and holes, especially holes, to transfer to the surface and react with the adsorbed oxygen/H₂O into some active oxygen radicals.

Fig. 7 (A) Photoluminescence spectra of the Bi₂S₃/CuS heterostructures in the range of 330 to 800 nm; (B) ESR signals of DMPO on the Bi₂S₃/CuS heterostructures (40-CuS/Bi₂S₃) in the dark and under visible light irradiation for 30 min.

To better understand the charge transfer mechanism, we estimated the energy level of these two semiconductors. The valence band potentials of a semiconductor at the point of zero charge can be calculated by the following empirical formula:^58,59

E_VB = X − E^e + 0.5E_g (5)

where E_VB is the valence band (VB) edge potential, X is the electronegativity of the semiconductor, which is the geometric mean of the electronegativity of the constituent atoms (herein, the electronegativity of an atom is the arithmetic mean of the atomic electron affinity and the first ionization energy), E^e is the energy of free electrons on the hydrogen scale (about 4.5 eV), and E_g is the band gap energy of the semiconductor. Consequently, the conduction band level (E_CB) can be determined by E_CB = E_VB − E_g. The X values for CuS and Bi₂S₃ are ca. 5.279 and 5.276 eV, and the E_VB of CuS and Bi₂S₃ were calculated to be 1.78 and 1.43 eV, respectively. Thus, E_CB of CuS and Bi₂S₃ was estimated to be 0.06 eV and 0.12 eV, respectively. Using the same calculation, the E_CB and E_VB of 10% sample can be calculated to be ca. −0.37 eV and 1.93 eV.

Based on the calculations, a possible band configuration of different Bi₂S₃/CuS heterostructures is proposed, as illustrated schematically in Fig. 8. For the pure Bi₂S₃ (on the left), the recombination of photogenerated e⁻–h⁺ pairs are relatively high, as indicated with a much stronger PL intensity. For the 10-CuS/Bi₂S₃ sample (in the middle of Fig. 8), the defects introduced by doping Cu²⁺ into Bi₂S₃ lattice could trap the photogenerated carriers, which on the one hand could suppress the PL intensity to some extent while on the other hand may hinder the transferring of photogenerated e⁻–h⁺ pairs to the surface. However, when CuS phase appeared at the doping level above 20% (on the right of Fig. 8), a heterojunction will form at their interfaces. The VB edge potential of CuS is 1.78 eV, more positive than that (1.52 eV) of Bi₂S₃, and thus a difference of band potentials exists between these two materials, which induces a contact electric field at the interface of the Bi₂S₃/CuS composite. Under visible light illumination, the photogenerated holes on the VB of CuS will transfer readily to the VB of Bi₂S₃. Considering that the CB potentials for CuS is slightly lower than that of Bi₂S₃, the photogenerated electrons on the CB of CuS and Bi₂S₃ will mainly concentrate in CuS phase, which as a consequence can be transferred to the surface of CuS. Therefore, the Bi₂S₃/CuS heterostructure photocatalysts showed an increase in the separation and lifetime of the photogenerated electrons, leading to effective photodegradation of RhB under visible light irradiation. Especially for the 40% sample, it shows almost over 3 times higher performance compared with pure Bi₂S₃ and CuS samples. A similar semiconductor heterostructure configuration and charge separation mechanism was reported by Su et al.⁶⁰ However, further theoretical and experimental works are still needed to verify the proposed mechanisms.

Fig. 8 Schematic diagram of band structure and the possible charge transfer process for different samples: pure Bi₂S₃ (left), CuS doping of 10% (middle) and CuS doping ≥20% (right).

Upon the above experimental results and analysis, several possible reasons behind the enhanced photocatalytic activity of the presently designed hierarchical Bi₂S₃/CuS nanocomposites could be explained as follows. (i) The enhanced photoabsorption capacity of the heterostructure plays an important role. The broader and more intensified absorption peak indicates the increased number of photogenerated charges upon the irradiation by sunlight. (ii) Since the CuS nanoplates in situ grow tightly onto the Bi₂S₃ nanotube, the photogenerated carriers could promptly transfer at the interface resulted from the proper matching of their band levels. Thereby, the charge recombination of e⁻–h⁺ pairs could be suppressed and the photocatalytic activity is therefore prompted. Compared pure CuS, Bi₂S₃ and 10% CuS doped sample with the 20% and 40% doped ones, this effect could be significantly highlighted. The initial introduction of Cu²⁺ into Bi₂S₃ lattice has resulted into a reduced performance, which could be fully reversed by the arising of CuS phase in the system. And this enhanced performance cannot be obtained either in pure CuS or Bi₂S₃ samples, indicating a necessity in the combination of these two phases. The band calculation and PL measurement also suggest the existence of charge transfer process between the two phases which significantly reduced the e⁻–h⁺ recombination rate. (iii) The 2D CuS nanoplates on the surface as compared to 1D or 0D nanostructures may provide a better anchoring surface for adsorbing molecules, as well as a lager surface area exposed to redox reactions. Therefore, the synergistic effect of all the above factors contributes to the final enhanced photocatalytic performance of the nanocomposites.

4. Conclusion

In summary, 3D hierarchical Bi₂S₃/CuS heterostructure photocatalyst consisting of Bi₂S₃ nanotube core and CuS nanoplates shell was prepared for the first time via simple solution based ion-exchange solvothermal reaction. Advantages of this solution-based method include its simplicity and potential low cost. The Bi₂S₃/CuS composites not only exhibited wide absorption wavelength in both UV and visible region but also showed good photocatalytic performance compared to bare Bi₂S₃ nanotubes and CuS nanoplates for the degradation of RhB. The highest photocatalytic activity was observed for the sample with CuS theoretical mass ratio of 40 wt%. The enhanced catalytic activities of the Bi₂S₃/CuS heterostructures could be attributed to both the enhanced light absorption and the effective separation of photoinduced carriers at their interface. Our results would provide useful information and guidance for the rational design and fabrication of efficient photocatalysts consisting of two or more sulfides with obvious solubility differences. We also believe that the development of novel photocatalyst with 3D hierarchical heterostructure and a broad light absorption window constructed by combination of 1D and 2D nanostructures is promising towards the high-performance photocatalysis applications.

Acknowledgements

This work was supported by National Natural Science Foundation of China (Grant No. 51272023), National Basic Research Program of China (Grant No. 2013CB632500).

References

1. M. Anpo, S. Dohshi, M. Kitano, Y. Hu, M. Takeuchi and M. Matsuoka, Annu. Rev. Mater. Res., 2005, 35, 1.
2. H. Tong, S. Ouyang, Y. Bi, N. Umezawa, M. Oshikiri and J. Ye, Adv. Mater., 2012, 24, 229.
3. M. G. Walter, E. L. Warren, J. R. McKone, S. W. Boettcher, Q. Mi, E. A. Santori and N. S. Lewis, Chem. Rev., 2010, 110, 6446.
4. U. I. Gaya and A. H. Abdullah, J. Photochem. Photobiol., C, 2008, 9, 1.
5. K. Maeda, T. Takata, M. Hara, N. Saito, Y. Inoue, H. Kobayashi and K. Domen, J. Am. Chem. Soc., 2005, 127, 8286.
6. J. N. Kondo, Chem. Commun., 1998, 3, 357.
7. H. Xu, W. Wang and W. Zhu, J. Phys. Chem. B, 2006, 110, 13829.
8. K. Vinodgopal and P. V. Kamat, Environ. Sci. Technol., 1995, 29, 841.
9. R. Konta, T. Ishii, H. Kato, H. Kato and A. Kudo, J. Phys. Chem. B, 2004, 108, 8992.
10. A. Kudo, K. Omori and H. Kato, J. Am. Chem. Soc., 1999, 121, 11459.
11. S. Tokunaga, H. Kato and A. Kudo, Chem. Mater., 2001, 13, 4624.
12. J. Tang, Z. Zou and J. Ye, Catal. Lett., 2004, 92, 53.
13. C. Zhang and Y. Zhu, Chem. Mater., 2005, 17, 3537.
14. S. Li, Y. H. Lin, B. P. Zhang, Y. Wang and C. W. Nan, J. Phys. Chem. C, 2010, 114, 2903.
15. S. Li, Y. H. Lin, B. P. Zhang and C. W. Nan, Chin. J. Inorg. Chem., 2010, 26, 495.
16. Y. Huang, B. Long, M. Tang, Z. Rui, M. S. Balogun, Y. Tong and H. Ji, Appl. Catal., B, 2016, 181, 779.
17. W. Ho, J. C. Yu, J. Lin, J. Yu and P. Li, Langmuir, 2004, 20, 5865.
18. Z. Fang, Y. Liu, Y. Fan, Y. Ni, X. Wei, K. Tang, J. Shen and Y. Chen, J. Phys. Chem. C, 2011, 115, 13968.
19. P. S. Lunawat, S. Senapati, R. Kumar and N. M. Gupta, Int. J. Hydrogen Energy, 2007, 32, 2784.
20. Y. Liu, P. D. Kanhere, C. L. Wong, Y. Tian, Y. Feng, F. Boey, T. Wu, H. Chen, T. J. White, Z. Chen and Q. Zhang, J. Solid State Chem., 2010, 183, 2644.
21. B. B. Kale, J. O. Baeg, S. M. Lee, H. Chang, S. J. Moon and C. W. Lee, Adv. Funct. Mater., 2006, 16, 1349.
22. Y. Huang, H. Li, M. S. Balogun, W. Liu, Y. Tong, X. Lu and H. Ji, ACS Appl. Mater. Interfaces, 2014, 6, 22920.
23. Y. Huang, W. Fan, B. Long, H. Li, F. Zhao, Z. Liu, Y. Tong and H. Ji, Appl. Catal., B, 2016, 185, 68.
24. B. Long, Y. Huang, H. Li, F. Zhao, Z. Rui, Z. Liu, Y. Tong and H. Ji, Ind. Eng. Chem. Res., 2015, 54, 12788.
25. Y. Huang, B. Long, H. Li, M. S. Balogun, Z. Rui, Y. Tong and H. Ji, Adv. Mater. Interfaces, 2015, 2, 1500249.
26. W. Fan, H. Li, F. Zhao, X. Xiao, Y. Huang, H. Ji and Y. Tong, Chem. Commun., 2016, 52, 5316.
27. A. Kudo and Y. Miseki, Chem. Soc. Rev., 2009, 38, 253.
28. X. Zong, H. Yan, G. Wu, G. Ma, F. Wen, L. Wang and C. Li, J. Am. Chem. Soc., 2008, 130, 7176.
29. G. Konstantatos, L. Levina, J. Tang and E. H. Sargent, Nano Lett., 2008, 8, 4002.
30. I. Grozdanov and M. Najdoski, J. Solid State Chem., 1995, 114, 469.
31. A. Casaca, E. B. Lopes, A. P. Gonçalves and M. Almeida, J. Phys.: Condens. Matter, 2012, 24, 015701.
32. Z. H. Ge, B. P. Zhang, Z. X. Yu and B. B. Jiang, CrystEngComm, 2012, 14, 2283.
33. L. Li, N. Sun, Y. Huang, Y. Qin, N. Zhao, J. Gao and L. Qi, Adv. Funct. Mater., 2008, 18, 1194.
34. J. S. Chung and H. J. Sohn, J. Power Sources, 2002, 108, 226.
35. Y. Q. Zhang, B. P. Zhang, Z. H. Ge, L. F. Zhu and Y. Li, Eur. J. Inorg. Chem., 2014, 2014, 2368.
36. T. Wu, X. Zhou, H. Zhang and X. Zhong, Nano Res., 2010, 3, 379.
37. J. Chen, S. Qin, G. Song, T. Xiang, F. Xin and X. Yin, Dalton Trans., 2013, 42, 15133.
38. H. Li, Y. Zhou, W. Tu, J. Ye and Z. Zou, Adv. Funct. Mater., 2015, 25, 998.
39. Z. Zhang, W. Wang, L. Wang and S. Sun, ACS Appl. Mater. Interfaces, 2012, 4, 593.
40. I. M. Kobasa and G. P. Tarasenko, Theor. Exp. Chem., 2002, 38, 255.
41. Z. Wu, L. Chen, C. Xing, D. Jiang, J. Xie and M. Chen, Dalton Trans., 2013, 42, 12980.
42. J. Yu, J. Zhang and S. Liu, J. Phys. Chem. C, 2010, 114, 13642.
43. S. Bai, J. Jiang, Q. Zhang and Y. Xiong, Chem. Soc. Rev., 2015, 44, 2893.
44. Z. Q. Liu, W. Y. Huang, Y. M. Zhang and Y. X. Tong, CrystEngComm, 2012, 14, 8261.
45. L. Chen, J. He, Q. Yuan, Y. W. Zhang, F. Wang, C. T. Au and S. F. Yin, RSC Adv., 2015, 5, 33747.
46. M. Grundmann, The physics of semiconductors, Springer-Verlag, Heidelberg, 2010.
47. Y. Q. Zhang, B. P. Zhang and L. F. Zhu, RSC Adv., 2014, 4, 59185.
48. A. Rauf, M. S. A. Sher Shah, G. H. Choi, U. B. Humayoun, D. H. Yoon, J. W. Bae, J. Park, W. J. Kim and P. J. Yoo, ACS Sustainable Chem. Eng., 2015, 3, 28475.
49. D. K. Ma, M. L. Guan, S. S. Liu, Y. Q. Zhang, C. W. Zhang, Y. X. He and S. M. Huang, Dalton Trans., 2012, 41, 5581.
50. Y. Y. Lu, Y. Y. Zhang, J. Zhang, Y. Shi, Z. Li, Z. C. Feng and C. Li, Appl. Surf. Sci., 2016, 370, 312.
51. N. Mukherjee, A. Sinha, G. G. Khan, D. Chandra, A. Bhaumik and A. Mondal, Mater. Res. Bull., 2011, 46, 6.
52. S. H. Chaki, J. P. Tailor and M. P. Deshpande, Mater. Sci. Semicond. Process., 2014, 27, 577.
53. Y. Yan, Z. Zhou, X. Zhao and J. Zhou, J. Colloid Interface Sci., 2014, 435, 91.
54. A. Alemi and M. Dolatyari, Radiat. Eff. Defects Solids, 2008, 163, 123.
55. H. J. Ko, Y. F. Chen, S. K. Hong, H. Wenisch, T. Yao and D. C. Look, Appl. Phys. Lett., 2000, 77, 3761.
56. T. Xu, L. Zhang, H. Cheng and Y. Zhu, Appl. Catal., B, 2011, 101, 382.
57. Z. Wang, W. Ma, C. Chen, H. Ji and J. Zhao, Chem. Eng. J., 2011, 170, 353.
58. M. A. Butler and D. S. Ginley, J. Electrochem. Soc., 1978, 125, 228.
59. J. Tang and J. Ye, Chem. Phys. Lett., 2005, 410, 104.
60. J. Su, X. X. Zou, G. D. Li, X. Wei, C. Yan, Y. N. Wang, J. Zhao, L. J. Zhou and J. S. Chen, J. Phys. Chem. C, 2011, 115, 8064.

---

Footnote

† Y. Zhang and S. Li contributed equally to this work.

---