Construction of an artificial inorganic leaf CdS–BiVO4 Z-scheme and its enhancement activities for pollutant degradation and hydrogen evolution

Ruijie Yang abc, Rongshu Zhu *abc, Yingying Fan bc, Longjun Hu bc and Baiyang Chen *bc
aState Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology, Harbin 150090, P. R. China
bShenzhen Key Laboratory of Organic Pollution Prevention and Control, Environmental Science and Engineering Research Center, Harbin Institute of Technology (Shenzhen), Shenzhen 518055, P. R. China
cInternational Joint Research Center for Persistent Toxic Substances, Harbin Institute of Technology (Shenzhen), Shenzhen 518055, P. R. China. E-mail: rszhu@hit.edu.cn; poplar_chen@hotmail.com

Received 10th March 2019 , Accepted 8th April 2019

First published on 10th April 2019


The natural photosynthesis of plants is closely related to leaf structures, therefore, the preparation of photocatalysts with leaf-like structures is of great significance to improve the activity of artificial photosynthetic systems. Using the leaves of a Chongyang wood seedling as the biological template (BT), artificial inorganic leaf CdS–BiVO4 was constructed by the “biological template”–“dipping-calcination”–“successive ionic layer adsorption and reaction” (BT–DC–SILAR) method. The fine hierarchical leaf-like structures of the artificial inorganic leaf were confirmed by FE-SEM and TEM observations. In the process of synthesizing the artificial inorganic leaf, the doping of C, N, and Si elements (coming from a natural leaf) was determined by XPS and FTIR analyses. The enhanced optical properties of the artificial inorganic leaf were proved by UV-vis DRS and PL analyses. It was found that the artificial inorganic leaf demonstrated superior photocatalytic activity in both photocatalytic pollutant degradation and H2 evolution. After 2 h of visible light irradiation, the photocatalytic decomposition efficiency of RhB for the artificial inorganic leaf (92%) is 2.1 times higher than that of no template BiVO4 (45%), and the photocatalytic H2 evolution for the artificial inorganic leaf (9250 μmol g−1CdS h−1) is 13 times higher than that of pure CdS (706 μmol g−1CdS h−1). Driven by solar light, the artificial inorganic leaf also possesses a strong artificial photosynthetic activity. The enhanced photocatalytic performance is ascribed to the combined action of the unique structure, the C, N, and Si elemental doping and the formation of the Z-scheme. Based on this research, a new method, BT–DC–SILAR, for the construction of a micro-nano-Z-scheme photocatalytic system was proposed, which enables the simultaneous control of structure, elemental doping and Z-scheme photocatalytic systems.


1. Introduction

In attempts to simulate natural photosynthesis, Z-scheme photocatalysis has received increasing attention,1–7 especially the all-solid-state Z-scheme photocatalytic systems between BiVO4 and CdS.8–10 Wu et al. reported that the BiVO4–CDs–CdS Z-scheme photocatalyst working under visible light shows enhanced photocatalytic activity, with a H2 evolution rate of 1.24 μmol h−1 and an O2 evolution rate of approximately 0.61 μmol h−1 for complete splitting of water.8 Zhang et al. reported that depositing Au nanoparticles onto the (010) facet of BiVO4 single crystals and then depositing CdS onto the Au nanoparticles to constitute a Z-scheme photocatalyst can effectively enhance photocatalytic degradation of tetracycline and rhodamine B.9 Zhou et al. also reported BiVO4 nanowires decorated with CdS nanoparticles as a Z-scheme photocatalyst with enhanced H2 generation at 23[thin space (1/6-em)]060 μmol g−1 h−1.10 These research studies indicate that the Z-scheme photocatalytic system between BiVO4 and CdS has achieved good results in simulating natural photosynthesis.

Natural photosynthesis of plants is generally considered to be closely related to their leaf structures.11–13 A natural leaf has unique micro-nano-multilevel structures: lens-like epidermal cells, columnar cells in palisade parenchyma, irregularly arranged spongy mesophyll cells, the porous architecture of veins, and three-dimensional constructions of interconnected nanolayered thylakoid cylindrical stacks (grana) in chloroplasts.11,14 These structures are strongly favoured for transmitting matter and harvesting light.5,11,14,15 Therefore, the construction of artificial leaves with leaf-like structures is of great significance to improve the photocatalytic activity of photocatalytic systems. In recent years, researchers have carried out research in this area. For example, Fan et al. fabricated the artificial inorganic leaf Pt/N-doped TiO2, which enhanced the harvesting of light energy and photocatalytic activity for hydrogen production.11 Zhou et al. fabricated g-C3N4 nanosheets, which mimick the nanolayered thylakoid stacks and show an enhancement on water splitting and CO2 reduction.16 Currently, the materials that constitute artificial inorganic leaves are mostly single semiconductors, which only simulated the structure of leaves in plant photosynthesis. Meanwhile, Z-scheme photocatalytic systems simulate the efficient electron transport pathway of natural photosynthesis, and only simulate the function of leaves in plant photosynthesis. However, little research has been done on the integrated system that simulates both the structure of leaves and the function of photosynthesis.5

In this work, an artificial inorganic leaf Z-scheme photocatalytic system was constructed. This strategy may provide a potential idea for a more comprehensive design of artificial photosynthetic systems and broaden the horizon for the development of solar energy. Firstly, C–N–Si self-doped BiVO4 (CNSiBiVO4) was synthesized by the dipping-calcination (DC) method with the leaves of a Chongyang wood seedling (from Shenzhen University Town, Nanshan District, Shenzhen, China) as the biological template (BT). On this basis, an artificial inorganic leaf CdS–CNSiBiVO4 micro–nano Z-scheme photocatalytic system was synthesized via the successive ionic layer adsorption and reaction (SILAR) method. Then, the artificial inorganic leaf was subsequently characterized by various analytical techniques and its photocatalytic activities in photocatalytic pollutant degradation and H2 evolution were tested. Furthermore, a new method, BT–DC–SILAR, was proposed for the construction of a micro-nano-Z-scheme photocatalytic system.

2. Experimental section

2.1 Materials

All chemicals were of analytical grade and were used as received without further purification. Glutaraldehyde (25%), 37% hydrochloric acid, Bi(NO3)3·5H2O, glycerol, ethanol, NH4VO3, 65% nitric acid, Cd(NO3)2·4H2O, Na2S·9H2O, methanol, tert-butyl alcohol (TBA), EDTA-2Na and p-benzoquinone (BQ) were purchased from Sinopharm Chemical Reagent Co., Ltd., China. Tetramethylammonium hydroxide (TMAH) aqueous solution and 3-mercaptopropionic acid (MPA) were purchased from Shanghai Jingchun Industrial Co., Ltd., China, and H2PtCl6·6H2O was purchased from Shanghai July Chemical Co., Ltd., China. The leaves of the Chongyang wood seedling were picked from Shenzhen University Town, Nanshan District, Shenzhen, China.

2.2 Synthesis of the different photocatalysts

2.2.1 C–N–Si self-doped BiVO4 (CNSiBiVO4). CNSiBiVO4 was synthesized by the dipping-calcination (DC) method (inspired by Zhou's work11) with the leaves of the Chongyang wood seedling as the biological template (BT). In a typical procedure, the leaves were first pre-treated. The leaves were soaked in cell fixation liquid (2% glutaraldehyde phosphate buffer) and stored at 4 °C for 8 h. The leaves were then rinsed with deionized water and subsequently immersed in 5% hydrochloric acid solution for 3 h. After the pretreatment, the leaves were rinsed with deionized water and then immersed in BiVO4 impregnation solution for 4 d. The BiVO4 impregnation solution was prepared as follows:17 (1) Bi(NO3)3·5H2O (19.4 g) was dissolved in a mixture of 80 mL ethanol and 60 mL glycerol at 70 °C with magnetic stirring; (2) NH4VO3 (4.68 g) was dissolved in 20 mL TMAH with gentle shaking; (3) these two solutions were mixed together at 70 °C. After this operation, large amounts of a yellow precipitate appeared. Concentrated nitric acid (65%) (about 15 mL) was added in a timely manner until the solution became transparent. The leaves were washed with ethanol solution (50%) and then dried in a temperature gradient of 40–60–80 °C for 24 h. Finally, the leaves were placed in a programmable oven and calcined for 6 h at 600 °C in an air atmosphere. CNSiBiVO4 was thus obtained.
2.2.2 Artificial inorganic leaf CdS–CNSiBiVO4 (CdS–CNSiBiVO4). CdS–CNSiBiVO4 was synthesized by the successive ionic layer adsorption and reaction (SILAR) method.18 CNSiBiVO4 was first immersed in 0.3 M MPA ethanol solution for 12 h at 50 °C. CNSiBiVO4 was then successively immersed in 150 mL 0.1 mol L−1 Cd(NO3)2·4H2O ethanol solution, pure ethanol, 150 mL 0.2 mol L−1 Na2S·9H2O methanol solution with 1.5 mL of 3.75 mg mL−1 H2PtCl6, pure methanol, and pure ethanol for 30 s each. This operation cycle was repeated 5 times. After that, the samples were dried under vacuum at 60 °C for 6 h, and then CdS–CNSiBiVO4 was obtained. In terms of the significance and the function of Pt, it is generally considered that it can be used as the hydrogen evolution cocatalyst in the photocatalytic hydrogen production reaction.19–22 For comparison, CdS–CNSiBiVO4–(no-Pt) was also synthesized using the same procedure but without the addition of H2PtCl6.
2.2.3 No-template BiVO4 and CdS. For comparison, no-template BiVO4 and CdS were also prepared. The BiVO4 impregnation solution (as described in section 2.2.1) was dried at 80 °C for 48 h in an oven until the water was completely evaporated. BiVO4 was then successively collected, ground and calcined (6 h at 600 °C). The no-template BiVO4 was obtained. In addition, CdS was prepared by the following steps. First, 1.5 mL 3.75 mg mL−1 H2PtCl6 and 150 mL 0.1 mol L−1 Na2S·9H2O solution were mixed and stirred for 10 min. Then, 150 mL 0.1 mol L−1 Cd(NO3)2·4H2O solution was added dropwise to the aforementioned solution. The mixed solution was magnetically stirred for 2 h, precipitated in the dark for 24 h, then successively washed with water, filtered, dried at 60 °C and ground, and then CdS was obtained.

2.3 Characterization

Thermogravimetry-differential scanning calorimetry (TG-DSC) curves of the as-prepared sample were recorded with a Netzsch STA 449F3 synchronous thermal analyser. The morphologies of the prepared photocatalysts were observed using a HITACHI SU1080 field-emission scanning electron microscope (FE-SEM) and a Tecnai G2 F30 high-resolution transmission electron microscope (TEM). The Brunauer–Emmett–Teller (BET) surface area was evaluated by N2 adsorption with an ASAP2020M+C constant-volume adsorption apparatus. The X-ray diffraction (XRD) surveys were performed on a Rigaku D/max 2550 VB/PC X-ray diffractometer equipped with a Cu Kα radiation source. The X-ray photoelectron spectroscopy (XPS) spectral measurements were carried out on a PHI 5000 VersaProbe II X-ray photoelectron spectroscope equipped with an Al Kα radiation source operated at 250 W. The Fourier transform infrared spectroscopy (FTIR) spectra were recorded on a Thermo Scientific Nicolet IS50 spectrometer using a 360 nm LED laser as the light source; the detection wavelength was 540 nm. The UV-vis diffuse-reflectance spectra (DRS) were collected with a SHIMADZU UV-2450 spectrometer equipped with an integrating sphere assembly; BaSO4 was used as the reference material. The photoluminescence (PL) spectra were recorded at room temperature on a SHIMADZU RF5301PC using the 380 nm line of a Xe lamp as the excitation source. The content of CdS in CdS–CNSiBiVO4 was measured by inductively coupled plasma (ICP) on a Thermo Scientific™, iCAP™ RQ ICP-MS.

2.4 Photocatalytic activity measurements

2.4.1 Photocatalytic degradation of RhB. The photocatalytic activity of the samples was evaluated by the degradation of RhB under visible-light irradiation. Briefly, 0.1 g photocatalyst was dispersed in the RhB solution (150 mL, 10 mg L−1). Before light irradiation, the suspension was stirred for 60 min in the dark to attain adsorption–desorption equilibrium of RhB on the surface of the photocatalyst. Then, the suspension was irradiated by a 350 W Xe lamp (Shenzhen Stone-Lighting Opto Device Co., Ltd., China) under magnetic stirring. During the reaction, 2 mL suspension was withdrawn from the reaction vessel every 20 min. The photocatalyst was then removed from the suspension by centrifugation. Finally, the concentration of RhB in the solution was measured using a UV-vis spectrophotometer (SHIMADZU UV-2450). For the recycling experiments, the used photocatalyst was collected by suction filtration, washed several times with deionized water, and then retested in a new RhB solution under the same experimental conditions.
2.4.2 Photocatalytic hydrogen evolution. The photocatalytic hydrogen evolution experiments were conducted on an off-line photocatalytic hydrogen evolution system similar to the RhB system. Photocatalysts (0.1 g) were dispersed in aqueous solution (150 mL) containing 10% lactic acid as sacrificial reagents. Prior to the photocatalytic reaction, the suspension was stirred in the dark and purged with argon gas for 30 min to remove oxygen from the reactor. Hydrogen evolution was observed only under irradiation and was analysed with a gas chromatograph (Shanghai Precision & Scientific Instrument Co., Ltd., GC-112A). For the recycling experiments, the used photocatalyst was collected by suction filtration, washed several times with deionized water, and then retested in a fresh lactic acid solution under the same experimental conditions previously mentioned.
2.4.3 Solar-light-driven experiments. The solar-light-driven experiments were carried out from 6:00 to 20:00 on August 16, 2017, at the E building roof of the Harbin Institute of Technology (Shenzhen), which is located in Shenzhen, China (N 22.59°, E 113.97°). The only difference between the solar-light-driven experiments and the visible-light-driven experiments was the light sources, which were sunlight and a 350 W Xe lamp, respectively. Light intensity was measured using a radiometer (Beijing Normal University Photoelectric Instrument Factory, FZ-A) during the experiment.

3. Results and discussion

3.1 Synthetic route of artificial inorganic leaf CdS–CNSiBiVO4

Fig. 1 shows the synthetic route for the artificial inorganic leaf CdS–CNSiBiVO4 (routes A–F) and the appearance of the leaf in different periods (a–g). First, the natural leaf (Fig. 1a) was immobilized (route A) to prevent autolysis and to maintain the original structure of the cells. Second, the immobilized leaf (Fig. 1b) was soaked in hydrochloric acid (route B) to change the permeability of the membrane to transfer the BiVO4 particles into the tissues and cells of the leaf. Third, the acidified leaf (Fig. 1c) was dipped into BiVO4 impregnation solution for 120 h (route C) to make the BiVO4 particles fill the tissues and cells (Fig. 1d). To avoid the structural damage, which is caused by rapid evaporation of the water in the leaf during calcination at high temperatures, the leaf was dried (route D) before calcining. The dried leaf (Fig. 1e) was then placed into a calciner. In the calcination process (route E), the organic ingredients of the leaf changed to CO2 and were volatilized from the leaf.23 At the same time, BiVO4 particles crystallized and then became BiVO4 crystals. After this operation, CNSiBiVO4 (Fig. 1f) was obtained. On this basis, CdS was deposited onto CNSiBiVO4via the SILAR method (route F), and then artificial inorganic leaf CdS–CNSiBiVO4 (Fig. 1g) was obtained.
image file: c9cy00475k-f1.tif
Fig. 1 The synthesis of artificial inorganic leaf CdS–CNSiBiVO4. Routes A–F are the synthetic routes of artificial inorganic leaf CdS–CNSiBiVO4; a–g are the appearances of the leaf in different periods; (a) original leaf; (b) leaf after being fixed with cells; (c) leaf after being dipped in hydrochloric acid for 3 h; (d) leaf after being dipped in BiVO4 impregnation solution for 120 h; (e) leaf after being dried; (f) CNSiBiVO4; (g) CdS–CNSiBiVO4.

To explore the weight changes of the leaf during the calcination process, TG-DSC was performed. As shown in Fig. S1, the first weight loss of the leaf was from room temperature to 120 °C, with a weight loss of approximately 36.1% attributed to the evaporation of water in the leaf during the heating process. The second and third weight-loss processes occurred at approximately 120 °C to 400 °C and 400 °C to 480 °C, with weight losses of approximately 43.7% and 12.3%, corresponding to the first and second obvious exothermic peaks, respectively; these events are attributed to the burning of the leaf organic ingredients and the formation of crystalline BiVO4, respectively. Based on TG-DTA, the possible formation temperature of BiVO4 is between approximately 400 °C and 480 °C. Therefore, to obtain the desired catalyst, the calcination temperature should not be less than 400 °C.

3.2 Characterization of CdS–CNSiBiVO4

3.2.1 Morphology and microstructure analysis. The morphologies of the as-prepared samples were observed by FE-SEM (Fig. 2 and S2) and TEM (Fig. 3). In Fig. 2 and S2, the tubular structure (Fig. 2a and a′ and S2a and b), surface porous structure (Fig. 2b and b′ and S2c–f), side section structure (Fig. S2g) and inside section structure (Fig. S2h and i), consisting of numerous crystallites, are clearly observed. Comparing these graphs with the microscopy images of the natural leaf (Fig. S3), we can see that CNSiBiVO4 has copied the multilevel structure of the natural leaf successfully. For comparison, the morphology of the no-template BiVO4 was also observed (Fig. S4), which was composed of irregular large-sized particles and no leaf-like structure was observed. Moreover, the BET surface area of CNSiBiVO4 (4.56 m2 g−1) was almost ten times larger than that of the no-template BiVO4 (0.41 m2 g−1), which is consistent with the morphology analysis. As shown in Fig. 2c and c′, after CdS was deposited onto the surface of CNSiBiVO4 crystal particles, the new sample still maintained its multilevel leaf-like structures.
image file: c9cy00475k-f2.tif
Fig. 2 FE-SEM images of the as-prepared CNSiBiVO4 and CdS–CNSiBiVO4. (a and a′) The vein section and (b and b′) the surface section of the as-prepared CNSiBiVO4; (c and c′) the surface section of the as-prepared CdS–CNSiBiVO4.

image file: c9cy00475k-f3.tif
Fig. 3 Images of the as-prepared CNSiBiVO4 and CdS–CNSiBiVO4. (a) TEM image of CNSiBiVO4; (b–c) TEM images of CdS–CNSiBiVO4; (d) HR-TEM image of CdS–CNSiBiVO4.

Fig. 3 shows the TEM images of the as-prepared CNSiBiVO4 and CdS–CNSiBiVO4. As shown in Fig. 3a and b (observed in the same observation mode), compared with the surface morphology of CNSiBiVO4 (Fig. 3a), CdS–CNSiBiVO4 (Fig. 3b) shows numerous bumps, which may be CdS quantum dots. To observe the surface morphology of CdS–CNSiBiVO4 more clearly, the observation mode was adjusted to a higher resolution and the image is shown in Fig. 3c. In this image, many microspheres on the surface of CNSiBiVO4 with a size of approximately 10 nm are clearly observed. The lattice fringe pattern of the quantum dots was measured to be 0.32 nm (Fig. 3d), consistent with the interplanar spacing of the CdS (111) plane.24,25 This result confirms that the microspheres on the surface of CNSiBiVO4 are CdS quantum dots. A similar morphology was also observed on the surface of CdS–CNSiBiVO4–(no-Pt) (shown in Fig. S5), further confirming that the microspheres on the surface of CNSiBiVO4 are CdS quantum dots, and not Pt nanoparticles. Unfortunately, we failed to observe the lattice fringes of BiVO4 crystals because they are too thick to transmit an electron beam.

3.2.2 Crystal structure analysis. To determine the crystallographic structures of different samples, XRD analysis was performed. Fig. 4 shows the XRD patterns of BiVO4, CdS, CNSiBiVO4 and CdS–CNSiBiVO4. In Fig. 4, the patterns of BiVO4, CNSiBiVO4 and CdS–CNSiBiVO4 show the characteristic peaks of monoclinic scheelite phase BiVO4 (JCPDS file: 14-0688).26–29 This result indicates that the crystallographic structures of CNSiBiVO4 were not changed after the leaf was used as the template. In the XRD pattern of CdS, three characteristic peaks at approximately 27°, 46° and 53° are assigned to the (111), (220) and (311) planes of CdS, respectively, which matches well with the cubic phase of CdS (JCPDS no. 89-0440).30 Both CdS and CdS–CNSiBiVO4 present a CdS (111) diffraction peak, demonstrating that CdS was successfully anchored onto the CNSiBiVO4 surface.
image file: c9cy00475k-f4.tif
Fig. 4 XRD patterns of the as-prepared BiVO4, CNSiBiVO4, CdS and CdS–CNSiBiVO4.
3.2.3 Chemical composition analysis. The XPS spectra of CNSiBiVO4 were analysed to identify the elemental composition and chemical state. For comparison, the XPS spectra of BiVO4 were also analysed. Fig. 5 shows the XPS spectra of CNSiBiVO4 and BiVO4. As shown in Fig. 5a, the XPS survey spectrum of CNSiBiVO4 confirms the presence of Bi, V, O, Si, N and C. In Fig. 5b and c, the peaks at binding energies of 103.5 eV and 405.5 eV in the XPS spectrum of CNSiBiVO4 are attributed to Si 2p (103.4 eV) and N 1s (405.5 eV);31 these peaks do not appear in the spectrum of BiVO4. These results indicate that parts of the original Si and N of the leaf were doped into CNSiBiVO4 during the calcination process and that the Si element existed as Si4+.31 As displayed in Fig. 5d, the peaks with binding energies of 284.8 eV and 286.0 eV are attributed to the C–C bond and the C–V bond, whereas the peak of the C–V bond does not appear in the spectrum of BiVO4. This result confirms that the C element was also doped into CNSiBiVO4, like the Si and N elements. Fig. 5e shows the XPS spectra of Bi in BiVO4 and CNSiBiVO4. Each spectrum has two major peaks at binding energies of approximately 159.4 eV or 159.3 eV and 164.6 eV or 164.5 eV, which correspond to Bi 4f7/2 and Bi 4f5/2, respectively, demonstrating that the valence state of Bi in BiVO4 and CNSiBiVO4 is +3.32–34 The two peaks in the spectrum of CNSiBiVO4 are slightly shifted to lower energies, compared with BiVO4, possibly because of the elemental doping. A phenomenon similar to that was reported by Zhang et al.35 and Xu et al.,36 and they attributed the peak shift to elemental C being inserted into the BiVO4 lattice.
image file: c9cy00475k-f5.tif
Fig. 5 XPS spectra of the as-prepared BiVO4 and CNSiBiVO4. (a) Survey spectrum of CNSiBiVO4; (b) Si 2p; (c) N 1s; (d) C 1s; (e) Bi 4f; (f) V 2p; and (g) O 2p.

Fig. 5f shows the XPS spectra of V in BiVO4 and CNSiBiVO4. The peak at a binding energy of 524.4 eV or 524.3 eV is attributed to V 2p1/2, and the peak at approximately 517.0 eV corresponds to V 2p3/2.24,27 Furthermore, the two peaks at approximately 516.7 eV or 516.6 eV and 517.4 eV or 517.2 eV are ascribed to V4+ and V5+, respectively.37,38 For BiVO4, the ratio of V4+ to V5+ is 1.08. By contrast, for CNSiBiVO4, the ratio is 1.63. The decrease in the proportion of V5+ is likely attributable to V5+ being replaced by Si4+ in CNSiBiVO4 because the ionic radius of Si4+ (0.41 nm) is closer to that of V5+ (0.59 nm) than V4+ (0.64 nm).39Fig. 5g shows the O 1s spectra of BiVO4 and CNSiBiVO4, which can be divided into two peaks located at 529.9 eV or 529.6 and 531.4 or 531.3 eV, corresponding to lattice oxygen (Olatt) and surface-adsorbed oxygen (Oabs), respectively.40,41 For BiVO4, the ratio of Oabs to Olatt is 0.53. By contrast, for CNSiBiVO4, the ratio is 1.31. In general, surface adsorption oxygen is closely related to the production of oxygen vacancies. These results indicate that CNSiBiVO4 has more oxygen vacancies than BiVO4, which is likely related to the decrease of V5+ in CNSiBiVO4, because a decrease in the concentration of V5+ likely leads to an increase in the number of oxygen vacancies.42

The FTIR spectra of BiVO4, CNSiBiVO4, CdS, CdS–CNSiBiVO4, and the leaf after calcination are shown in Fig. 6. The band located at approximately 3445 cm−1 in the spectra of the leaf, CdS and CdS–CNSiBiVO4 is attributed to stretching vibrations of the adsorbed H2O.8 As shown in the FTIR spectrum of the leaf, the absorption peaks at approximately 700, 1100 and 1500 cm−1 can be assigned to the vibrations of phosphate, silicate, and carbonate groups, respectively. Moreover, the spectrum does not exhibit characteristic peaks of organic matter, demonstrating that the organic matter of the leaf was consumed completely during the calcining process at 600 °C. In the FTIR spectrum of BiVO4 and CNSiBiVO4, the strong absorption peaks at 475, 736–744 and 840–845 cm−1 correspond to the typical stretching vibration of Bi–O, the deformation bending vibration of V–O and the symmetrical stretching vibration of V–O, respectively.43 The peak at 1100 cm−1 in the spectrum of CNSiBiVO4 is ascribed to the typical antisymmetric stretching vibration of Si–O; this peak does not appear in the FTIR spectra of BiVO4. These results confirm that Si element exists in CNSiBiVO4, which is consistent with the XPS characterization results. For CdS, the absorption peak at 1620 cm−1 is attributed to the surface-adsorbed H2O.44 The typical characteristic absorption bands at 1380 and 1114 cm−1 are due to the vibrations of the Cd–S bonds.45 The FTIR spectrum of CdS–CNSiBiVO4 represents the superimposed spectra of CNSiBiVO4 and CdS. These results further confirm that the CdS–CNSiBiVO4 composites were successfully synthesized.


image file: c9cy00475k-f6.tif
Fig. 6 FTIR spectra of the leaf after calcination, BiVO4, CNSiBiVO4, CdS and CdS–CNSiBiVO4.
3.2.4 Optical properties analysis. The UV-vis diffuse-reflectance spectra of BiVO4, CNSiBiVO4, CdS and CdS–CNSiBiVO4 are presented in Fig. 7a. In comparison with BiVO4, CNSiBiVO4 not only exhibits a wider wavelength absorbance performance but also exhibits enhancement of the light absorbance intensity, which can be ascribed to the unique structure and the C, N, and Si elemental doping. CdS–CNSiBiVO4 exhibits an obvious redshift of the absorption edge (compared with CNSiBiVO4) and exhibits the strongest visible-light absorption intensity (in the range of 530–700 nm). These results indicated that artificial inorganic leaf CdS–CNSiBiVO4 possesses the best optical performance.
image file: c9cy00475k-f7.tif
Fig. 7 The optical properties of the catalysts. (a) UV-vis DRS spectra of BiVO4, CNSiBiVO4, CdS and CdSCNSiBiVO4; (b) PL spectra of BiVO4, CNSiBiVO4 and CdS–CNSiBiVO4.

PL spectroscopy is considered to be a very useful technique for analysing the efficiency of photogenerated electron–hole pair separation in photocatalysts.24,46Fig. 7b shows the PL spectra of BiVO4, CNSiBiVO4, and CdS–CNSiBiVO4. As shown in Fig. 7b, CNSiBiVO4 exhibits a weaker PL intensity than BiVO4, which may be attributed to the C elemental doping favouring the rapid electron transfer. A similar phenomenon was observed in our previous studies;47–49 in these studies, the weaker PL intensity appears in the photocatalyst, which contains a greater carbon content. Wang et al.50 investigated the temporal evolution of fluorescence from carbon nanotubes and pointed out that a low fluorescence quantum yield originated from the rapidly quenching fluorescence rather than from the inherent weakness of the radiative transitions. Because of its greater electron mobility, carbon affords quick electron transfer, resulting in a decrease of the emission intensity. As shown in Fig. 7b, apparently, CdS–CNSiBiVO4 displays the lowest PL intensity, implying that artificial inorganic leaf CdS–CNSiBiVO4 possesses the highest transfer efficiency of the photogenerated charge carriers.

3.3 Photocatalytic activity of CdS–CNSiBiVO4

3.3.1 Photocatalytic RhB degradation. The visible-light-driven photocatalytic oxidation activities of the catalysts were evaluated on the basis of the RhB degradation ratio, as shown in Fig. 8a. For comparison, a blank experiment (no catalyst) was also performed to exclude the effects of self-degradation of RhB under visible-light irradiation. The degradation ratio of RhB over BiVO4 was 46% after visible-light irradiation for 120 min. Compared with the degradation ratio of BiVO4, that of CNSiBiVO4 was substantially increased to 82%, implying that CNSiBiVO4 exhibits an evident enhancement of photocatalytic oxidation activity. The enhancement of CNSiBiVO4 is attributed to its better optical absorption properties and greater photogenerated electron–hole pair separation ability. Artificial inorganic leaf CdS–CNSiBiVO4 shows much higher photocatalytic activity towards the degradation of RhB (the degradation ratio was 92%) than CNSiBiVO4 because of its further improvement of optical absorption properties and photogenerated electron–hole pair separation ability. Furthermore, the mineralization ratio of RhB over CdS–CNSiBiVO4 (shown in Fig. 8a) is almost coincident with its degradation ratio, illustrating that CdS–CNSiBiVO4 also possesses strong photocatalytic mineralization activity. In Fig. 8a, although CdS exhibits a higher photocatalytic degradation activity than CdS–CNSiBiVO4, the mineralization ratio of RhB over CdS was apparently lower than the mineralization ratio of CdS–CNSiBiVO4. A similar phenomenon has been reported by Zhang et al.19 (they reported that MnOx@CdS shows a lower photocatalytic RhB degradation activity than the pure CdS, and they attributed the result to the oxidation-active species that mainly exist at the inner surface of MnOx@CdS. However, the fact is that the holes exposed at the outer surface of CdS shells much more easily react with RhB molecules).
image file: c9cy00475k-f8.tif
Fig. 8 The visible-light-driven photocatalytic activities of the catalysts. (a) Photocatalytic degradation of RhB (the solid line) over BiVO4, CNSiBiVO4, CdS and CdS–CNSiBiVO4 and photocatalytic mineralization of RhB (the dotted line) over CdS and CdS–CNSiBiVO4; (b) stability of the photocatalysts for RhB degradation over CdS and CdS–CNSiBiVO4; (c) hydrogen evolution over CdS and CdS–CNSiBiVO4; (d) stability of the photocatalysts for hydrogen evolution over CdS and CdS–CNSiBiVO4.

As photostability is a very important characteristic for the practical application of the photocatalyst, cyclic photocatalytic degradation experiments were performed here to measure the photostability of CdS and CdS–CNSiBiVO4. As displayed in Fig. 8b, in the case of CdS–CNSiBiVO4, the degradation ratio of RhB shows only a small decrease (7.2%) after four cycles. For CdS, after four cycles, the degradation ratio of RhB shows a rapid decrease (67%). These results indicate that, excluding the loss of the photocatalyst in the cycling tests, artificial inorganic leaf CdS–CNSiBiVO4 can be considered as a stable photocatalyst for the degradation of organic pollutants.

3.3.2 Photocatalytic hydrogen production. Except for the application to the photooxidation of RhB, the artificial inorganic leaf CdS–CNSiBiVO4 can also be applied to a photocatalytic water reduction reaction. Fig. 8c shows the hydrogen evolution curve of CdS and CdS–CNSiBiVO4 (for easy comparison, the photocatalytic activity is expressed by the hydrogen production per unit mass of CdS. The content of CdS in CdS–CNSiBiVO4 is only 2.4 wt%, as measured by inductively coupled plasma). For comparison, the activities of BiVO4 and CNSiBiVO4 were also tested, and both of them displayed insufficient activity to detect hydrogen (not shown here). As shown in Fig. 8c, the H2 evolution of CdS–CNSiBiVO4 is 9250 μmol g−1CdS h−1, which is higher than the H2 evolution of CdS (706 μmol g−1CdS h−1). Furthermore, the H2 evolution stability of CdS and CdS–CNSiBiVO4 was also detected. The results are displayed in Fig. 8d. As expected, CdS–CNSiBiVO4 demonstrates undiminished activity even after four cycles. However, the photocatalytic activity of CdS shows a decrease after four cycles (the rate of hydrogen production decreased to 16.6%). These results indicate that artificial inorganic leaf CdS–CNSiBiVO4 can also be considered as a stable photocatalyst for hydrogen production.
3.3.3 Solar-light-driven photocatalytic activities. Fig. 9 shows the solar-light-driven photocatalytic activity of artificial inorganic leaf CdS–CNSiBiVO4. When the sunlight intensity became strong and stable, the photocatalytic activity of CdS–CNSiBiVO4 reaches a certain level, which is close to the level of photocatalytic activity under irradiation by a 350 W xenon lamp (126 W cm−2). This result implies that CdS–CNSiBiVO4 also exhibits a good photocatalytic activity under sunlight irradiation. As shown in Fig. 9, the sunlight intensity was relatively weaker from 6:00 to 8:00; however, CdS–CNSiBiVO4 also maintained a high photocatalytic activity towards the degradation of RhB, indicating that RhB can also be degraded by CdS–CNSiBiVO4 under relatively weaker sunlight intensity.
image file: c9cy00475k-f9.tif
Fig. 9 The solar-light-driven photocatalytic activities of CdS–CNSiBiVO4. (a) Degradation of RhB; (b) changes of solar-light intensity with time in a day; (c) hydrogen evolution.
3.3.4 Photocatalytic mechanism. Artificial inorganic leaf CdS–CNSiBiVO4 exhibits better activity for oxidizing organic matter than CNSiBiVO4 and a stronger hydrogen production capacity than CdS (as shown in Fig. 8). To prove whether CdS–CNSiBiVO4 is a system photocatalyst and not a simple mixture of individual components, the mixture experiment was carried out. In the mixture experiment, 0.1 g photocatalysts (the mixture of 2.4 wt% pure CdS and 97.6 wt% pure CNSiBiVO4, according to the mass ratio of CdS to CNSiBiVO4 in the CdS–CNSiBiVO4 system) were dispersed in the RhB solution. After 120 min of visible-light irradiation, the degradation ratio of RhB was 84%, which is lower than the degradation ratio when the CdS–CNSiBiVO4 system (92%) was used. These results illustrate that the CdS–CNSiBiVO4 system is not a simple mixture of CdS and CNSiBiVO4.

To clarify the photocatalytic mechanism of artificial inorganic leaf CdS–CNSiBiVO4, the participation of reactive species such as hydroxyl radicals (˙OH), superoxide radicals (˙O2) and holes (h+) in the photocatalytic degradation of organic pollutants must be considered. To verify the main reactive species for the degradation of RhB over CdS–CNSiBiVO4 and CNSiBiVO4, we conducted a series of contrast experiments using different quenchers. Herein, EDTA-2Na was used to quench h+, whereas TBA and BQ were used to quench ˙OH and ˙O2, respectively. In Fig. 10a, the photocatalytic activity of CNSiBiVO4 shows an obvious decrease in the presence of TBA and EDTA-2Na, whereas it shows almost no change after BQ was added into the reaction solution. This result indicates that, for CNSiBiVO4, ˙OH and h+ play the main roles in RhB degradation and that ˙O2 basically does not work because of the weak reduction ability of e. In terms of CdS–CNSiBiVO4 (shown in Fig. 10b), after the addition of TBA, BQ and EDTA-2Na, all of the photocatalytic activity distinctly decreases. This result indicates that, for CdS–CNSiBiVO4, ˙OH, h+ and ˙O2 are all main reactive species in the RhB degradation reaction, revealing that the electrons (e) can reduce O2 to generate ˙O2. These results indicate that artificial inorganic leaf CdS–CNSiBiVO4 is the Z-scheme system photocatalyst.


image file: c9cy00475k-f10.tif
Fig. 10 Photocatalytic degradation curves of RhB over CNSiBiVO4 (a) and CdS–CNSiBiVO4 (b). (c) Proposed photocatalytic mechanisms of CdS–CNSiBiVO4 and CNSiBiVO4.

On the basis of the aforementioned results, the degradation mechanism of RhB over artificial inorganic leaf CdS–CNSiBiVO4 was proposed; it is shown in Fig. 10c. Both CNSiBiVO4 and CdS can absorb the visible light to generate (e) in conduction bands (CB) and h+ in valence bands (VB). The h+ on the VB of CNSiBiVO4 can oxidize hydroxyl to produce ˙OH and the e in the VB of CdS easily combines with the absorbed O2 to generate ˙O2. All of the h+ and the produced ˙OH and ˙O2 participate in the degradation reaction of organic pollutants in the CdS–CNSiBiVO4 Z-scheme photocatalytic system. In the absence of oxygen, the e in the VB of CdS reduces water to H2.

3.4 A new method for the construction of micro-nano-composite photocatalytic systems – BT–DC–SILAR method

In this work, taking the leaf as the biological template (BT), the BiVO4 substrate (CNSiBiVO4) was synthesized by the dipping calcination (DC) method. Then, CdS quantum dots were deposited onto the BiVO4 substrate via the successive ionic layer adsorption and reaction (SILAR) method, and then the micro-nano CdS–CNSiBiVO4 Z-scheme photocatalytic system was constructed successfully. Fig. S6 shows the structure and composition schematic diagram of the natural leaf, CNSiBiVO4 and CdS–CNSiBiVO4.

Based on the construction process of artificial inorganic leaf CdS–CNSiBiVO4, the BT–DC–SILAR method for the construction of a micro-nano-Z-scheme photocatalytic system was proposed. It is summarized as follows: (1) selection and pre-treatment of the appropriate BT. Some researchers have prepared the corresponding catalysts with butterfly wings17 and rice husks51 as the BT. (2) Preparation of the precursor solution, such as the BiVO4 precursor solution in this study or the TiCl4 precursor solution used in another study.11 (3) Fabrication of the substrate using the DC method, such as the BiVO4 substrate in this work. (4) Preparation of the ion precipitation solution, such as S2− solution and Cd2+ solution in this work. (5) Preparation of the quantum dots deposited onto the substrate via the SILAR method, then the micro-nano-composite photocatalytic system is constructed successfully.

The BT–DC–SILAR method provides a new path for the construction of micro-nano-Z-scheme photocatalytic systems, which can simultaneously achieve the successful preparation of biomimetic structures and the efficient combination of catalysts. This method possesses three outstanding characteristics: (1) in the structure, micro-nano-Z-scheme photocatalytic systems synthesized by the BT–DC–SILAR method replicate the micro-nano-multi-level structure of the biological template, which is favourable for efficient matter transmission and sunlight capture. (2) With respect to composition, the original elements in the natural BT such as C, N, and Si are doped into the photocatalyst lattice, which will improve the photocatalytic activity of the catalytic system. (3) The substrate synthesized in the process of the BT–DC–SILAR method possesses a certain macroscopic morphology (for instance, the BiVO4 substrate possesses a leaf shape in this research), which enables it to be placed into the grid, ultimately allowing the SILAR method to deposit the quantum dots evenly on it.

4. Conclusions

In this work, an artificial inorganic leaf CdS–CNSiBiVO4 Z-scheme photocatalytic system was synthesized by the BT–DC–SILAR method. The measured photocatalytic activities indicate that artificial inorganic leaf CdS–CNSiBiVO4 demonstrated superior photocatalytic activity in RhB degradation and H2 evolution. After 2 h of visible light irradiation, the photocatalytic decomposition efficiency of RhB for artificial inorganic leaf CdS–CNSiBiVO4 (92%) is 2.1 times higher than that of no template BiVO4 (45%), and the photocatalytic H2 evolution for artificial inorganic leaf CdS–CNSiBiVO4 (9250 μmol g−1CdS h−1) is 13 times higher than that of pure CdS (706 μmol g−1CdS h−1). The enhanced photocatalytic performance is ascribed to the combined action of the unique structure, the C, N, and Si elemental doping and the formation of the Z-scheme. The unique leaf structures of CdS–CNSiBiVO4 improve the visible light absorption; the C, N, and Si elemental doping contributes to the enhancement of the efficient separation and transfer of the photo-generated electron–hole pairs and the visible light absorption; and the formation of the Z-scheme further advances the efficient separation of electron–hole pairs. In addition, based on this research, a new method, BT–DC–SILAR, for the construction of a micro-nano-Z-scheme photocatalytic system was proposed: (1) selection and pre-treatment of the appropriate BT; (2) preparation of the precursor solution; (3) fabrication of the substrate using the DC method; (4) preparation of the ion precipitation solution; (5) preparation of the quantum dots deposited onto the substrate via the SILAR method, and then the micro-nano-composite photocatalytic system is constructed successfully. This method enables the simultaneous control of structure, elemental doping and Z-scheme photocatalytic systems.

Conflicts of interest

There are no conflicts of interest to declare.

Acknowledgements

All the authors gratefully acknowledge support from the Special Fund for the Development of Strategic and New Industry in Shenzhen (No. JCYJ20150731104949789) and the Fund for Knowledge Innovation in Shenzhen (No. JCYJ20180507183621817).

Notes and references

  1. M. M. May, H. J. Lewerenz, D. Lackner, F. Dimroth and T. Hannappel, Nat. Commun., 2015, 6, 8286 CrossRef CAS PubMed.
  2. H. Tada, T. Mitsui, T. Kiyonaga, T. Akita and K. Tanaka, Nat. Mater., 2006, 5, 782–786 CrossRef CAS PubMed.
  3. P. Zhou, J. Yu and M. Jaroniec, Adv. Mater., 2014, 26, 4920–4935 CrossRef CAS PubMed.
  4. S. Hara, M. Yoshimizu, S. Tanigawa, L. Ni, B. Ohtani and H. Irie, J. Phys. Chem. C, 2012, 116, 17458–17463 CrossRef CAS.
  5. H. Zhou, L. Ding, T. Fan, J. Ding, D. Zhang and Q. Guo, Appl. Catal., B, 2014, 147, 221–228 CrossRef CAS.
  6. H. Li, W. Tu, Y. Zhou and Z. Zou, Adv. Sci., 2016, 3, 1500389 CrossRef PubMed.
  7. D. Zheng, C. Pang and X. Wang, Chem. Commun., 2015, 51, 17467–17470 RSC.
  8. X. Wu, J. Zhao, L. Wang, M. Han, M. Zhang, H. Wang, H. Huang, Y. Liu and Z. Kang, Appl. Catal., B, 2017, 206, 501–509 CrossRef CAS.
  9. S. Bao, Q. Wu, S. Chang, B. Tian and J. Zhang, Catal. Sci. Technol., 2017, 7, 124–132 RSC.
  10. F. Q. Zhou, J. C. Fan, Q. J. Xu and Y. L. Min, Appl. Catal., B, 2017, 201, 77–83 CrossRef CAS.
  11. H. Zhou, X. Li, T. Fan, F. E. Osterloh, J. Ding, E. M. Sabio, D. Zhang and Q. Guo, Adv. Mater., 2010, 22, 951–956 CrossRef CAS PubMed.
  12. H. Zhou, R. Yan, D. Zhang and T. Fan, Chemistry, 2016, 22, 9870–9885 CrossRef CAS PubMed.
  13. H. Zhou, J. Guo, P. Li, T. Fan, D. Zhang and J. Ye, Sci. Rep., 2013, 3, 1667 CrossRef PubMed.
  14. E. Shimoni, O. Rav-Hon, I. Ohad, V. Brumfeld and Z. Reich, Plant Cell, 2005, 17, 2580–2586 CrossRef CAS PubMed.
  15. X. Li, T. Fan, H. Zhou, S.-K. Chow, W. Zhang, D. Zhang, Q. Guo and H. Ogawa, Adv. Funct. Mater., 2009, 19, 45–56 CrossRef CAS.
  16. H. Zhou, P. Li, J. Liu, Z. Chen, L. Liu, D. Dontsova, R. Yan, T. Fan, D. Zhang and J. Ye, Nano Energy, 2016, 25, 128–135 CrossRef CAS.
  17. R. Yan, M. Chen, H. Zhou, T. Liu, X. Tang, K. Zhang, H. Zhu, J. Ye, D. Zhang and T. Fan, Sci. Rep., 2016, 6, 20001 CrossRef CAS.
  18. S. Zhou and L. Yin, J. Alloys Compd., 2017, 691, 1040–1048 CrossRef CAS.
  19. M. Xing, B. Qiu, M. Du, Q. Zhu, L. Wang and J. Zhang, Adv. Funct. Mater., 2017, 27, 1702624 CrossRef.
  20. Y. Wang, J. Hong, W. Zhang and R. Xu, Catal. Sci. Technol., 2013, 3, 1703 RSC.
  21. P. Montes-Navajas, M. Serra and H. Garcia, Catal. Sci. Technol., 2013, 3, 2252 RSC.
  22. M. Zhu, Y. Du, P. Yang and X. Wang, Catal. Sci. Technol., 2013, 3, 2295 RSC.
  23. N. E. Ermolin and V. M. Fomin, Combust., Explos. Shock Waves, 2016, 52, 566–586 CrossRef.
  24. B. Weng, S. Liu, N. Zhang, Z.-R. Tang and Y.-J. Xu, J. Catal., 2014, 309, 146–155 CrossRef CAS.
  25. Z. Sun, H. Zheng, J. Li and P. Du, Energy Environ. Sci., 2015, 8, 2668–2676 RSC.
  26. D. Fang, X. Li, H. Liu, W. Xu, M. Jiang, W. Li and X. Fan, Sci. Rep., 2017, 7, 7979 CrossRef.
  27. M. Zalfani, B. van der Schueren, Z.-Y. Hu, J. C. Rooke, R. Bourguiga, M. Wu, Y. Li, G. Van Tendeloo and B.-L. Su, J. Mater. Chem. A, 2015, 3, 21244–21256 RSC.
  28. C. Lv, G. Chen, J. Sun, C. Yan, H. Dong and C. Li, RSC Adv., 2015, 5, 3767–3773 RSC.
  29. S. J. Hong, S. Lee, J. S. Jang and J. S. Lee, Energy Environ. Sci., 2011, 4, 1781 RSC.
  30. D. Wang, H. Shen, L. Guo, F. Fu and Y. Liang, New J. Chem., 2016, 40, 8614–8624 RSC.
  31. Min Wang, Chao Niu, Jiali Huang and Y. Che, Zhongguo Youse Jinshu Xuebao, 2015, 25, 440–448 CAS.
  32. S. Balachandran, N. Prakash, K. Thirumalai, M. Muruganandham, M. Sillanpää and M. Swaminathan, Ind. Eng. Chem. Res., 2014, 53, 8346–8356 CrossRef CAS.
  33. Y. Deng, L. Tang, G. Zeng, C. Feng, H. Dong, J. Wang, H. Feng, Y. Liu, Y. Zhou and Y. Pang, Environ. Sci.: Nano, 2017, 4, 1494–1511 RSC.
  34. J. Choi, P. Sudhagar, J. H. Kim, J. Kwon, J. Kim, C. Terashima, A. Fujishima, T. Song and U. Paik, Phys. Chem. Chem. Phys., 2017, 19, 4648–4655 RSC.
  35. C. Yin, S. Zhu, Z. Chen, W. Zhang, J. Gu and D. Zhang, J. Mater. Chem. A, 2013, 1, 8367 RSC.
  36. D. Zhao, W. Zong, Z. Fan, S. Xiong, M. Du, T. Wu, Y.-W. Fang, F. Ji and X. Xu, CrystEngComm, 2016, 18, 9007–9015 RSC.
  37. J. Sun, X. Li, Q. Zhao, J. Ke and D. Zhang, J. Phys. Chem. C, 2014, 118, 10113–10121 CrossRef CAS.
  38. Z. Wang, J. Xuan, B. Liu and J. He, J. Ind. Text., 2013, 44, 868–883 CrossRef.
  39. X. Zhang, X. Quan, S. Chen and Y. Zhang, J. Hazard. Mater., 2010, 177, 914–917 CrossRef CAS PubMed.
  40. X. Wu, J. Zhao, S. Guo, L. Wang, W. Shi, H. Huang, Y. Liu and Z. Kang, Nanoscale, 2016, 8, 17314–17321 RSC.
  41. S. H. Yun, P. G. Ingole, W. K. Choi, J. H. Kim and H. K. Lee, J. Mater. Chem. A, 2015, 3, 7888–7899 RSC.
  42. M. Wang, Q. Liu, Y. Che, L. Zhang and D. Zhang, J. Alloys Compd., 2013, 548, 70–76 CrossRef CAS.
  43. J. Liu, H. Wang, S. Wang and H. Yan, Mater. Sci. Eng., B, 2003, 104, 36–39 CrossRef.
  44. Y. Liu, K. Yan and J. Zhang, ACS Appl. Mater. Interfaces, 2016, 8, 28255–28264 CrossRef CAS PubMed.
  45. F. Jiang, T. Yan, H. Chen, A. Sun, C. Xu and X. Wang, Appl. Surf. Sci., 2014, 295, 164–172 CrossRef CAS.
  46. Y. He, L. Zhang, B. Teng and M. Fan, Environ. Sci. Technol., 2015, 49, 649–656 CrossRef CAS PubMed.
  47. F. Tian, R. Zhu, J. Zhong, P. Wang, F. Ouyang and G. Cao, Int. J. Hydrogen Energy, 2016, 41, 20156–20171 CrossRef CAS.
  48. M. Niu, R. Zhu, F. Tian, K. Song, G. Cao and F. Ouyang, Catal. Today, 2015, 258, 585–594 CrossRef CAS.
  49. R. Zhu, F. Tian, S. Che, G. Cao and F. Ouyang, Renewable Energy, 2017, 113, 1503–1514 CrossRef CAS.
  50. F. Wang, G. Dukovic, L. E. Brus and T. F. Heinz, Phys. Rev. Lett., 2004, 92, 177401 CrossRef PubMed.
  51. D. Yang, T. Fan, H. Zhou, J. Ding and D. Zhang, PLoS One, 2011, 6, 247–288 Search PubMed.

Footnote

Electronic supplementary information (ESI) available. See DOI: 10.1039/c9cy00475k

This journal is © The Royal Society of Chemistry 2019