English

• Energy is the basic element for the sustainability of human society and it is crucial for social and economic development. Nowadays, owing to the severe environmental pollution, global warming and energy shortage, it is urgent to optimize our existing energy structures, integrate available and potential energy resources to find a way for the future of mankind. Solar energy is a clean and well-stocked energy source, which is the ultimate energy in the existing energy resources that can satisfy our demand of energy [1].

  Splitting water to produce hydrogen and oxygen is one convenient way for energy conversion [2]. The water splitting comprises of two half reactions, that is, one is the evolution of O₂ (2H₂O → O₂ + 4H⁺ + 4e^-) and the other is evolution of H₂ (4H⁺ + 4e^- → 2H₂) [3]. The oxygen evolution reaction involves a process of four-electron transfer, which is more challenging than hydrogen evolution reaction [4].

  Since Fujishima et al. [5] discovered that TiO₂ can split water under ultraviolet irradiation, many semiconductor materials have been researched for photoelectrochemical water splitting, such as BiVO₄ [6-8], Fe₂O₃ [9-11] and WO₃ [12, 13]. WO₃ has a bandgap of 2.6–2.7 eV for visible light absorbability, and its valence band position is about 3 V (vs. normal hydrogen electrode (NHE)) to efficiently oxidize water [14]. Therefore, WO₃ semiconductor photoanode is chosen by us for researching the water oxidation reaction. However, water oxidation reaction kinetics is sluggish when only WO₃ is used as the photoanode. In order to resolve this defect, using cocatalyst is a common strategy to promote the kinetics [15].

  Herein, we loaded MnO_x on the WO₃ photoanode through photodeposition and hydrothermal methods to improve the photoelectrochemical water oxidation performance. Manganese oxide existed on prepared composite photoanode WO₃ nanoflakes was confirmed by a series of characterizations. Photoelectrochemical tests illustrated that photodeposited MnO_x as the cocatalyst increases the photocurrent density of the photoanode.

  All chemicals were analytical grade and were used as purchased without further purification. Double-distilled water (18.2 MΩ cm) for the experiments was attained from a Molecular Lab water purifier. The fluorine-doped tin oxide conductive glass (FTO, Zhuhai Kaivo Optoelectronic Technology Co., Ltd., < 15Ω/sq.) was ultrasonically cleaned with acetone, ethanol and double-distilled water for 30 min each in sequence prior to use.

  WO₃ film was prepared on FTO conductive glass using a previously reported method [16]. A WO₃ seed layer was coated on a FTO conductive glass by spin coating a seed solution. The seed solution was made by dissolving 1.25 g of H₂WO₄ and 0.5 g of PVA (poly vinyl alcohol, the average polymerization degree was 1750) in 10 mL of 50 wt% H₂O₂. Then the coated FTO conductive glass was annealed at 500 ℃ for 2 h in air. A H₂WO₄ solution was prepared for solvothermal preparation, by dissolving 1.25 g of H₂WO₄ in 30 mL of double-distilled water and adding 10 mL of 50 wt% H₂O₂ with heating at 95 ℃. The resulting solution was diluted to 0.05 mol/L with double-distilled water. The annealed FTO conduction glass was placed in a 25 mL Teflon-lined stainless steel autoclave with 3 mL of H₂WO₄ (0.05 mol/L) solution, 0.02 g of oxalic acid, 0.02 g of urea, 0.5 mL of 6 mol/L HCl, and 12.5 mL of acetonitrile. Then the reaction was kept at 180 ℃ for 2 h. After reaction, the FTO conduction glass was washed with double-distilled water and dried at 60 ℃. The resulting WO₃ film was annealed at 500 ℃ for 1 h in air.

  Hydrothermal of MnO_x on WO₃ film was carried out hydrothermally starting with solution of MnSO₄·H₂O and KMnO₄ following the procedure reported method [17]. A mixed solution of MnSO₄·H₂O and KMnO₄ was transferred to a Teflon-lined stainless steel autoclave with the WO₃-coated FTO conduction glass and kept at 140 ℃. The time for hydrothermal reaction was 2 h, 4 h, 6 h, 8 h, and 10 h. Then the resulting electrode was washed with double-distilled water and followed by air-dried at room temperature. This sample is labeled as MnO_x/WO₃-H.

  Photodeposition of MnO_x on WO₃ film was performed from an aqueous solution (initial pH 5.1) containing MnSO₄ and NaIO₃ as precursors under irradiation [18]. In a usual preparation, a typical three electrode cell included a WO₃ film as work electrode, a Ag/AgCl reference, and a Pt counter electrode. A LED light source was used to irradiate the WO₃ film work electrode. During illumination, WO₃ semiconductor produced holes and electrons. Then, Mn²⁺ ions were oxidized to high valence species, which precipitated on the surface of the WO₃ film electrode. This sample is labeled as MnO_x/WO₃-P.

  X-ray diffraction (XRD) measurements were performed on PANalytical X'Pert Pro Diffractometer using Cu-K radiation as the X-ray source. The surface morphologies of the above-mentioned electrode were investigated by scanning electron microscopy (SEM, SU8020) and transmission electron microscopy (TEM, Tecnai G² TF20). X-ray photoelectron spectroscopy (XPS) spectra were measured by ESCALAB250xi with X-ray monochromatization and the binding energy was calibrated with respect to C 1s level 284.8 eV of contaminated carbon.

  Photoelectrochemical measurements were conducted using a three-electrode set up with a Ag/AgCl (3.5 mol/L KCl) as reference, a Pt electrode as a counter electrode. All photoelectrochemical measurements were implemented on a CHI 660D workstation (CH Instruments Co.) under simulated solar light irradiation with an AM 1.5G filter (100 mW/cm², Perfect Light). Throughout all the experiment, 0.1 mol/L potassium phosphate buffer (pH 7) was used as an electrolyte. All measured potentials were converted to V vs. RHE according to E_RHE = E_Ag/AgCl + 0.208 + 0.059 pH. The photocurrent-potential curves were measured by linear sweep voltammetry with a scan rate of 10 mV/s.

  The morphologies of pristine WO₃ and MnO_x/WO₃ were investigated by SEM and TEM. Fig. 1 presents typical SEM images of an illustrative as-prepared WO₃ nanoflake, which was grown on a FTO conduction glass substrate. Figs. 1a and d show that WO₃ nanoflakes pile up with rough surface, which could increases the contact area between electrode and electrolyte solution. After photodepositing MnO_x on WO₃ film, the surface of catalyst composite becomes smooth (Figs. 1b and e). It is obvious that nanoflake construct remains after photodeposition process. Unfortunately, hydrothermal deposition of MnO_x on WO₃ changes WO₃ nanoflake structure (Figs. 1c and f) and a great deal of MnO_x is found on the surface of the irregular WO₃ nanorods.

  Figure 1

  

  Figure 1.  SEM image of the WO₃ film (a and d), photodeposition of MnO_x/WO₃-P film (b and e), hydrothermal of MnO_x/WO₃-H film (c and f).

  DownLoad: Full-Size Img PowerPoint

  TEM images were carried out to better illustrate the configurations of the MnO_x catalyst on WO₃ film. In order to investigate the microstructure of WO₃ and MnO_x/WO₃-P film, TEM measurements were performed on a pristine sample and photodeposited MnO_x sample. TEM images show the nanoflake structure of WO₃ (Figs. 2a and d) remains unchanged with photodepositing MnO_x. The lattice spacing of 0.35 nm is clearly observed, corresponding to the (012) plane of monoclinic WO₃ (Fig. 2b). The lattice spacing of 0.32 nm is corresponding to the (201) plane of MnO₂ (Fig. 2e), which suggests the formation of MnO₂ on the surface of WO₃. Furthermore, it can be seen that the electron diffraction spectrum of WO₃ is a diffraction spot with certain symmetry (Fig. 2c). While a concentric diffraction ring is observed for the MnO_x/WO₃-P film (Fig. 2f), indicating that the measured material contains microcrystalline, this can prove that MnO_x loaded on the WO₃. Energy dispersive X-ray spectroscopy (EDX) characterizations further confirm that the Mn element is present on the WO₃ electrode after loading the MnO_x catalyst (Fig. S1 in Supporting information).

  Figure 2

  

  Figure 2.  TEM and SAED images of WO₃ film (a–c) and MnO_x/WO₃-P film (d–f).

  DownLoad: Full-Size Img PowerPoint

  Fig. 3 shows the XRD patterns of the composite photoanodes. The black line is the XRD pattern of pristine WO₃ film on FTO conduction glass. The green one is the XRD pattern of MnO_x/WO₃-H, which was obtained by hydrothermal synthesis method. The pink one is the XRD pattern of MnO_x/WO₃-P, which was made by photodeposition method. Curve a in Fig. 3 gives diffraction peaks of 23.1°, 23.6°, 24.4°, 34.0°, 41.9°, 49.9° and 55.9°, which are indexed to the (002), (020), (200), (202), (222), (140), and (420) lattice fringe diffractions of the monoclinic WO₃ (JCPDS: 43-1035) [19]. Compared with pure WO₃, the XRD patterns of the other two composites have changed after loading MnO_x catalyst (Fig. 3, curves b and c). The characteristic peak located at 63.1° is indexed to the (511) lattice diffraction of MnO₂ (JCPSD: 39-0375). MnO_x has been coated on the WO₃ photoanode through two different preparation methods, which is consistent with the SEM images. Indeed, MnO₂ is detected by XRD and TEM measurements in the composite photoanode. However, due to the very positive potentials of the top of the valence band of WO₃, it is possible for the composite photonode to generate some other Mn species with other valence states during the process of preparing MnO_x/WO₃ and photoelectrochemical test. In addition, we cannot rule out the other manganese oxide species since comparatively limited amount of manganese species loaded on WO₃ photoanode is detected hardly. Therefore, MnO_x is used in the manuscript instead of MnO₂.

  Figure 3

  

  Figure 3.  XRD of WO₃ film (black, a), hydrothermal of MnO_x/WO₃-H (green, b), and photodeposition of MnO_x/WO₃-P (pink, c).

  DownLoad: Full-Size Img PowerPoint

  From the survey spectrum of WO₃ (Fig. S2 in Supporting information), we can see the distinct peaks of W and O elements. The fine W4f spectrum is shown in Fig. S2b. The peaks located at 35.5 eV and 37.6 eV correspond with W 4f_7/2 and W 4f_5/2 of WO₃ [20], respectively. Fig. S2c is the Mn 3s fine XPS spectrum of MnO_x/WO₃-P, suggesting the successful loading of MnO_x on WO₃ [21].

  In order to evaluate the photoelectrochemical activity of MnO_x modified WO₃ photoanodes, a three-electrode electrochemical cell using a Pt plate as a counter electrode and a Ag/AgCl (3.5 mol/L KCl) as a reference electrode was set up for implementing linear sweep voltammograms (LSVs). Fig. 4 demonstrates the LSVs of two different ways to load MnO_x on WO₃ photoanode. The photocurrent density decreases after loading MnO_x by hydrothermal synthesis method compared with that of pristine WO₃ photoanode (Fig. 4a). After loading MnO_x by hydrothermal method, the photocurrent density of the electrode decreases with the time of hydrothermal synthesis. When the hydrothermal time is 10 h, the photocurrent density approaches approximately zero. The onset potentials of composite photoanodes change more positive than that of pristine WO₃ photoanode. The main reason for the negative effect is probably that the ordered structure of nanoflake WO₃ has changed during the hydrothermal procedure. As displayed in Figs. 1c and f, unordered structure is not conducive to the separation of photogenerated electron-hole pairs. The more time hydrothermal process continues, the more MnO_x is loaded on WO₃. Over much MnO_x can block the incident illumination and decrease the light-absorbed efficiency [10]. In addtiotion, holes are hard to arrive at the interface for oxygen evolution reaction. In all, the activity of composite photoanode MnO_x/WO₃-H declines due to WO₃ structure variation and much loaded MnO_x.

  Figure 4

  

  Figure 4.  LSV images of photoanodes: (a) MnO_x/WO₃-H by hydrothermal synthesis method, (b) MnO_x/WO₃-P by photodeposition.

  DownLoad: Full-Size Img PowerPoint

  Fig. 4b displays the LSV of the MnO_x/WO₃-P photoanode produced by photodeposition method. The dependence of photocurrent on deposition time is also studied, and the optimum time of the composite photoanode is 3 min. When the photodeposition time increases, the photocurrent density decreases in the whole test range of potential. In the process of photoelectrochemical water oxidation, holes must be transferred to the interface for oxidizing water and avoid the recombination with electrons [22]. The poor performance of the samples at 5 min and 8 min could be attributed to the accumulation of MnO_x among the nanoflakes of pristine WO₃, which perhaps aggravates the recombination of electrons and holes. Only suitable amount of MnO_x loaded on WO₃ can extract holes from WO₃ and boost surface reaction kinetics.

  In summary, two different methods were used to load MnO_x on WO₃ photoanode and the photoelectrochemical activity and some characterizations of MnO_x/WO₃ composite electrodes were carried out. It is found that the activity of WO₃ photoanode could be improved efficiently after loading MnO_x by photodeposition. The maximum photocurrent density of composite photoanode is achieved with a deposition time of 3 min, which is higher than that of pristine WO₃ photoanode around 40%. However, loading MnO_x by hydrothermal method will change the original structure of WO₃ photoanode and the photoelectrochemical water oxidation performance of WO₃ electrode declines after hydrothermal procedure. In a word, photodeposition of cocatalysts MnO_x on the surface of WO₃ semiconductor can effectively promote photoelectrochemical performance. Our research provides insightful guidelines for designing and synthesizing new efficient photoelectrochemical water oxidation catalysts in the future.

  Acknowledgements

  This work was financially supported by the National Natural Science Foundation of China (Nos. 21173105, 21773096), Fundamental Research Funds for the Central Universities (No. lzujbky-2016-k08), Open fund by Jiangsu Key Laboratory of Atmospheric Environment Monitoring and Pollution Control (No. KHK1701) and the Natural Science Foundation of Gansu (No. 17JR5RA186).

  Appendix A. Supplementary data

  Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.cclet.2017.12.010.

• 1. [1]

     I. Roger, M.A. Shipman, M.D. Symes, Nat. Rev. Chem. 1(2017) 0003. doi: 10.1038/s41570-016-0003

  2. [2]

     Y. Feng, J. Wei, Y. Ding, J. Catal. 339(2016) 186-194. doi: 10.1016/j.jcat.2016.04.009

  3. [3]

     M. Zheng, Y. Ding, L. Yu, X. Du, Y. Zhao, Adv. Funct. Mater. (2017) 1605846. doi: 10.1002/adfm.201605846

  4. [4]

     X. Du, J. Huang, Y. Feng, Y. Ding, Chin J. Catal. 37(2016) 123-134. doi: 10.1016/S1872-2067(15)61012-9

  5. [5]

     A. Fujishima, K. Honda, Nature 238(1972) 37-38. doi: 10.1038/238037a0

  6. [6]

     Y.H. Ng, A. Iwase, A. Kudo, R. Amal, J. Phys. Chem. Lett. 1(2010) 2607-2612. doi: 10.1021/jz100978u

  7. [7]

     Y. Feng, H. Cheng, J. Han, et al., Chin. Chem. Lett.28(2017) 2254-2258. doi: 10.1016/j.cclet.2017.10.025

  8. [8]

     R. Li, F. Zhang, D. Wang, et al., Nat. Commun. 4(2013) 1432. doi: 10.1038/ncomms2401

  9. [9]

     L. He, L. Jing, Y. Luan, L. Wang, H. Fu, ACS Catal. 4(2014) 990-998. doi: 10.1021/cs401122e

  10. [10]

      J. Huang, G. Hu, Y. Ding, M. Pang, B. Ma, J. Catal. 340(2016) 261-269. doi: 10.1016/j.jcat.2016.05.007

  11. [11]

      M. Wang, M. Wang, Y. Fu, S. Shen, Chin. Chem. Lett. 28(2017) 2207-2211. doi: 10.1016/j.cclet.2017.11.037

  12. [12]

      C.A. Bignozzi, S. Caramori, V. Cristino, et al., Chem. Soc. Rev. 42(2013) 2228-2246. doi: 10.1039/C2CS35373C

  13. [13]

      J. Huang, Y. Ding, X. Luo, Y. Feng, J. Catal. 333(2016) 200-206. doi: 10.1016/j.jcat.2015.11.003

  14. [14]

      Q. Mi, A. Zhanaidarova, B.S. Brunschwig, H.B. Gray, N.S. Lewis, Energy Environ. Sci. 5(2012) 5694-5700. doi: 10.1039/c2ee02929d

  15. [15]

      J. Yang, D. Wang, H. Han, C. Li, Acc. Chem. Res. 46(2012) 1900-1909. http://www.ncbi.nlm.nih.gov/pubmed/23530781/

  16. [16]

      J. Su, X. Feng, J.D. Sloppy, L. Guo, C.A. Grimes, Nano Lett. 11(2011) 203-208. doi: 10.1021/nl1034573

  17. [17]

      V. Subramanian, H. Zhu, R. Vajtai, P.M. Ajayan, B. Wei, J. Phys. Chem. B 109(2005) 20207-20214. doi: 10.1021/jp0543330

  18. [18]

      Y. Matsumoto, S. Ida, T. Inoue, J. Phys. Chem. C 112(2008) 11614-11616. doi: 10.1021/jp804625r

  19. [19]

      J. Huang, Y. Zhang, Y. Ding, ACS Catal. (2017) 1841-1845.

  20. [20]

      J.Y. Zheng, G. Song, J. Hong, et al., Cryst Growth Des. 14(2014) 6057-6066. doi: 10.1021/cg5012154

  21. [21]

      Y. Gorlin, T.F. Jaramillo, J. Am. Chem. Soc. 132(2010) 13612-13614. doi: 10.1021/ja104587v

  22. [22]

      T. Yao, R. Chen, J. Li, et al., J. Am. Chem. Soc. 138(2016) 13664-13672. doi: 10.1021/jacs.6b07188

• 

  Figure 1  SEM image of the WO₃ film (a and d), photodeposition of MnO_x/WO₃-P film (b and e), hydrothermal of MnO_x/WO₃-H film (c and f).

  下载: 全尺寸图片 幻灯片
  

  Figure 2  TEM and SAED images of WO₃ film (a–c) and MnO_x/WO₃-P film (d–f).

  下载: 全尺寸图片 幻灯片
  

  Figure 3  XRD of WO₃ film (black, a), hydrothermal of MnO_x/WO₃-H (green, b), and photodeposition of MnO_x/WO₃-P (pink, c).

  下载: 全尺寸图片 幻灯片
  

  Figure 4  LSV images of photoanodes: (a) MnO_x/WO₃-H by hydrothermal synthesis method, (b) MnO_x/WO₃-P by photodeposition.

  下载: 全尺寸图片 幻灯片
•