English

• Sonication has been widely used for dispersing particles in liquids or cleaning glassware. Since Suslick proposed the concept of sonochemistry in the early 1990s [1], numerous studies have demonstrated the use of sonochemistry for materials synthesis [2, 3], catalysis [4-8], and biological material deactivation [9, 10]. The most important effects of ultrasound arise from acoustic cavitation: formation, growth, and implosive collapse of bubbles in liquid by passing ultrasonic waves through the medium [11]. It has been suggested that the implosive collapse of the bubbles generates the localized hot spots that greatly accelerated the H₂ evolution in H₂O sonolysis [4] and significantly enhanced the degradation rate of azo dyes [6]. Among these applications, sonophotocatalysis has attracted great attention due to its potential application in reactions involving environmental remediation and protection, such as environmental pollutant degradation [12-14] and H₂O reduction [15, 16]. The combination of ultrasonic and light irradiations was believed to significantly prompt the chemical reaction rate. For examples, Mrowetz et al. observed the synergistic effect of sonolysis and photocatalysis towards the degradation of acid orange 8 in the presence of TiO₂, and they believed that the sonication induced the desorption of organic substrates as well as the degradation of the intermediates from the photocatalyst surface [17]. Wang et al. found that the high intensity sonication treatment was a simple and effective way to improve the performance of ZnO nanoparticles in the photoelectrochemical water splitting under the tungsten halogen lamp excitation [16]. However, to date, the description of the function of light irradiation in coupling with the ultrasound to facilitate the sonophotocatalytic process still remains ambiguous. Previous studies have suggested that a synergistic effect of photocatalysis and sonolysis might reduce the band gap energy via the high temperature and high pressure produced by the bubble collapse under the ultrasonic irradiation [18]. Others have proposed that both the light irradiation and ultrasound irradiation could generate reactive free radicals to directly increase the reaction activities [19].

  Among these different understanding of the synergistic effect in sonophotocatalytic reaction, it has been generally accepted that the oxidizing species (^·OH) generated on photocatalyst surfaces is necessary [20], suggesting that the surface charges on the photocatalysts might play a crucial role in prompting the surface chemical reactions. It has been known that the UV irradiation can generate hot carriers (electrons and holes) in semiconductors (e.g., TiO₂) and alter their electronic structures [21]. Recent studies demonstrated that the visible light also can be used to produce hot carriers in metal/semiconductor heterostructures to initiate surface chemical reactions [22-24]. Upon irradiation, the surface plasmon resonance (SPR) of metal nanoparticles is excited and then mediates the transfer of hot electrons across an interfacial Schottky barrier (φ_B) at the metal-semiconductor heterojunction, resulting different surface charge distributions on the metal/ semiconductor heterostructures. Here, we use light irradiation of different wavelengths to manipulate the electronic structures of Au/TiO₂ investigate the roles of surface charges in enhancing the sonophotocatalytic water (H₂O) reduction reaction. Our study will improve the fundamental understanding of sonophotochemistry and promote its potential applications in real catalytic reactions.

  Au/TiO₂ was prepared using a deposition precipitation (DP) method [25, 26]. HAuCl₄·H₂O (≥99.9%, Aldrich) was dissolved in Nanopure H₂O to prepare a 0.0191mol/L aqueous solution. Then, the HAuCl₄ aqueous solutionwas mixed with TiO₂ powder (≥99.5% trace metals basis P25, Aldrich) and Nanopure H₂O in a flask and stirred at 323K for 30min. NH₄OH (28%–30% by weight, ACS reagent) was added to adjust the pH value of the system to ca. 9, after which it was stirred at 323K for 2h. The gel was centrifuged, washed with Nanopure H₂O threetimes, dried at 323K for 12h and calcined at 473K for 4h. The weight percentage (wt%) of Au onTiO₂ was estimated to be 1wt%.

  To test the sonophotochemical activity, 50mg of as-prepared Au/TiO₂ and 15mL Nanopure H₂O were added into a 50mL threeneck round-bottom flask, followed by sonication for 20min to achieve a better dispersion of the catalysts in the solution. Later, 5mL methanol was injected into the solution. Schlenk line was used to evacuate atmospheric oxygen(O₂) and H₂O, followed by a gas flow of inert Argon (Ar) into the reactor. The flask was then placed in the middle of a sonicator (Fisher Scientific, FS30) throughout the reaction. The reactor was irradiated by a 300W Xe lamp (Newport 6258) coupled with long pass filters (Newport) with specified wavelengths. The power intensitieswere measured to 1.0W/cm² by a power meter (Newport, 1916-C). For each cycle, the reactor was sonicated for 1 min and then cooled down for 5 min to ensure the stabilization of the system temperature. Finally, 200 μL gas products were withdrawn by a gas-tight syringe at given reaction time and analyzed by a Gas Chromatography (GC, Shimadzu, GC2014) equipped with a 60/80 Mol Sieve 5A column. During the measurement, the thermal conductivity detector was maintained at 308K with Ar as the carrier gas. Given that the intensities of both incident light and ultrasonic wave were greatly affected by the position of the reactor, the experiments under light irradiations of different wavelengths were carried out without changing the position of the reactor to ensure the validity of comparison.

  Photoluminescence (PL) spectra were measured on a Horiba Aramis Raman system, which is a software-selectable multiwavelength Raman/PL system. High-angle annular dark fieldscanning transmission electron microscope (HAADF-STEM) analysis was operated on a Hitachi HD 2700C. X-ray photoelectron spectroscopy(XPS)analysis was conducted on a Sisubstrate using a ULVAC-PHI 5000 VersaProbe Ⅲ System. The binding energies in the XPS spectra were referenced to the Ti 2p_3/2 binding energy in TiO₂ at 457.8eV.

  Fig. 1A presents a HAADF-STEM image of Au/TiO₂ catalyst. The average particle diameter of Au was measured to be 4.4 ±1.1nm over a hundred Au nanoparticles (Fig. 1C). The interface of the Au nanoparticle and TiO₂ support was also examined, in which the lattice fringes could be easily distinguished (Fig. 1B). The lattice fringes separated by 0.36 and 0.23nm were assigned to anatase TiO₂ (101) and Au (111), respectively. An intact interface (which was marked by a white frame) between TiO₂ (101) and Au (111) was observed.

  Figure 1

  

  Figure 1.  (A) HAADF-STEM images of Au/TiO₂, (B) high-resolution image that shows the interface of Au-TiO₂, (C) size distribution of Au nanoparticles in Au/TiO₂.

  DownLoad: Full-Size Img PowerPoint

  Fig. 2A shows the high-resolution XPS spectrum of Au 4f peaks acquired over the Au/TiO₂ heterostructures. The Au 4f_7/2 peak consists of a single component with a binding energy of 83.7eV, which is slightly red-shifted relative to Au⁽⁰⁾ at 84.0eV [27]. This indicates that a partially negative surface charge was imparted from the TiO₂ tothe Au, as the Fermi level of two components must align during the formation of the Au-TiO₂ interface (Fig. 2C). Such strong interaction between Au and TiO₂ was further confirmed by the photoluminescence (PL) studies. As shown in Fig. 2B, Au/TiO₂ exhibits much weaker PL intensity than pristine TiO₂ under a 325-nm laser excitation.

  Figure 2

  

  Figure 2.  (A) High-resolution XPS spectrum of Au 4f peaks acquired over the Au/TiO₂ heterostructures. (B) Photoluminescence spectra of Au/TiO₂ and pristine TiO₂. (C) Band diagrams show the formation of the Schottky barrier. (D) Band diagrams show the change in the electronic structures under different optical excitation conditions. Note: In both band diagrams, CB stands for conduction band and VB stands for valence band. E_F stands for Fermi level.

  DownLoad: Full-Size Img PowerPoint

  If surface charges account for the enhanced catalytic activity during sonication, then the rate of H₂O sonolysis would be different when the electronic structures of Au/TiO₂ are altered. In our approach, different optical excitation was used to manipulate the surface charges of Au/TiO₂. As shown in Fig. 2D, when TiO₂ is excited by UV irradiation (i.e., λ < 400nm), hotelectrons inject into Au due to a lower electronic potential, leading tothe formation of a negative charge on Au and a positive charge on TiO₂ (marked as Au^(-)/TiO₂⁽⁺⁾) [22, 23]. Alternatively, when using visit light (i.e., λ > 435nm), Au LSPR is excited and facilitates the transfer of hot electrons into the conduction band of TiO₂, leaving positively charged Au and negatively charged TiO₂ (marked as Au⁽⁺⁾/TiO₂^(-)) [22, 23].

  Following the methods outlined in previous literature [23, 28], the catalytic activities of the Au/TiO₂ photocatalysts were tested for H₂O reduction. Fig. 3A shows that Au/TiO₂ by sonication alone produced (32.7 ±1.7) μmol_H2/g_cat H₂ within 1min. Without sonication, the Au/TiO₂ photocatalysts still produced (11.3 ±0.8) μmol_H2/g_cat H₂ over Au^(-)/TiO₂⁽⁺⁾ under UV excitation within 1min. The above results suggest that both the sonication and UV excitation can assistant the H₂O reduction to produce H₂. However, under visible light, we did not detect H₂ production over Au⁽⁺⁾/TiO₂^(-) within 1min. Such observation is consistent with our previous finding that the hot electrons transferred from the small Au nanoparticles to TiO₂ do not have enough potential energy to overcome the proton reduction potential [23].

  Figure 3

  

  Figure 3.  Amount of evolved H₂ over Au/TiO₂ photocatalyst under different experimental conditions. (A) Sonication only or optical excitation only. (B) The accumulated amount of H₂ as a function of the sonication time under different conditions. (C) Sonication combined with optical excitation.

  DownLoad: Full-Size Img PowerPoint

  To investigate the possible synergistic effect of the light irradiation and ultrasound irradiation and identify the role of surface charges, the sonophotocatalytic H₂O reduction was performed in the following sequence: (1) in the dark, (2) under UV excitation, (3) under visible excitation, and (4) in the dark again. The accumulated H₂ production along time is shown in Fig. 3B and the amount of H₂ generated from the sonolysis of H₂O under each experimental condition was calculated and plotted in Fig. 3C. For each sonication cycle, (32.7 ± 1.7) μmol_H2/g_cat of H₂ production was produced in the dark. However, to our surprise, only (34.0 ± 1.9) μmol_H2/g_cat of H₂ were produced under the UV excitation and sonication whereas (35.7 ± 3.0) μmol_H2/g_cat of H₂ were produced under the visible light and sonication. These negligible changes in the H₂ production strongly suggest that the optically induced surface charge (electron and hole) distributions, regardless of their locations in Au/TiO₂, should not be responsible for the enchantment in the sonophotocatalytic reactions. More importantly, our results imply that the sonication has much higher efficiency to generate radical species than that of the photoexcitation by UV light. Moreover, under our experimental conditions, the ultrasonic irradiation should not reduce the band gap energy of TiO₂. Thus, our results confirm that the acoustic cavitation-facilitated removal of intermediates and recovery of active sites via the ultrasound irradiation should be the key step in enhancing the formation of H₂ in the photosonocatalytic H₂O reduction.

  In summary, the electronic structures of Au/TiO₂ photocatalysts were manipulated through both UV and visible light irradiations. However, no significant increase of the rate of H₂O sonophotocatalysis was observed to associate with the buildup of surface charges. Given that free radicals, such as ^·OH and ^·H are known as the important species for the photosonocatalytic water splitting, our findings indicate that neither electrons nor holes-enriched semiconductor surfaces are able to enhance the formation of free radicals under sonication. Thus, enhancement of the sonophotochemical process should primarily arise from the recovery of active sites through ultrasound irradiation. This study provides evidence to clarify the function of light irradiation in the sonophotochemical processes and thereby confirms the key role of acoustic cavitation-facilitated removal of intermediates in sonophotocatalysis.

  Acknowledgments

  The work is supported by the National Science Foundation (NSF, No. DMR-1352328). Materials characterization was primarily conducted at the Major Analytical Instrumentation Center (MAIC) and the Nanoscale Research Facility (NRF), College of Engineering Research Service Centers, University of Florida. XPS characterization was conducted using an instrument purchased with NSF grant No. MRI-DMR-1126115. Partial electron microscopy work was carried out at the Center for Functional Nanomaterials at Brookhaven National Laboratory (Upton, NY) through User Proposal Nos. BNLCFN-31913, BNL-CFN-33789 and BNL-CFN-35817, supported by the U.S. Department of Energy (DOE), Office of Basic Energy Sciences, under Contract No. DE-SC0012704.

• 1. [1]

     K.S. Suslick, Science 247(1990) 1439-1445. doi: 10.1126/science.247.4949.1439

  2. [2]

     P. Hasin, Y.Y. Wu, Chem. Commun. 48(2012) 1302-1304. doi: 10.1039/C2CC15741A

  3. [3]

     Z.Y. Shen, G. Chen, Q. Wang, et al., Nanoscale 4(2012) 2010-2017. doi: 10.1039/c2nr12045c

  4. [4]

     Y.F. Wang, D. Zhao, W.H. Ma, et al., Environ. Sci. Technol. 42(2008) 6173-6178. doi: 10.1021/es800168k

  5. [5]

     Y.F. Wang, D. Zhao, H.W. Ji, et al., J. Phys. Chem. C 114(2010) 17728-17733. doi: 10.1021/jp105691v

  6. [6]

     G. Cravotto, P. Cintas, Chem. Sci. 3(2012) 295-307. doi: 10.1039/C1SC00740H

  7. [7]

     C.G. Joseph, G.L. Puma, A. Bono, D. Krishnaiah, Ultrason. Sonochem. 16(2009) 583-589. doi: 10.1016/j.ultsonch.2009.02.002

  8. [8]

     Y.H. He, F. Grieser, M. Ashokkumar, Ultrason. Sonochem. 18(2011) 974-980. doi: 10.1016/j.ultsonch.2011.03.017

  9. [9]

     P. Grimaldi, L. Di Giambattista, S. Giordani, et al., Spectrochim. Acta A 84(2011) 74-85. doi: 10.1016/j.saa.2011.08.074

  10. [10]

      P. Qin, L. Xu, W.J. Zhong, A.C.H. Yu, Ultrasound Med. Biol. 38(2012) 1085-1096. doi: 10.1016/j.ultrasmedbio.2012.02.017

  11. [11]

      G.A.D. Briggs, Interdiscip. Sci. Rev. 15(1990) 190-191.

  12. [12]

      R. Daghrir, A. Dimboukou-Mpira, B. Seyhi, P. Drogui, Sci. Total Environ. 490(2014) 223-234. doi: 10.1016/j.scitotenv.2014.05.006

  13. [13]

      I.M. Khokhawala, P.R. Gogate, Ultrason. Sonochem. 17(2010) 833-838. doi: 10.1016/j.ultsonch.2010.02.012

  14. [14]

      C.D. Wu, X.H. Liu, D.B. Wei, et al., Water Res. 35(2001) 3927-3933. doi: 10.1016/S0043-1354(01)00133-6

  15. [15]

      H. Harada, Ultrason. Sonochem. 8(2001) 55-58. doi: 10.1016/S1350-4177(99)00050-4

  16. [16]

      H. Wang, L. Jia, P. Bogdanoff, et al., Energy Environ. Sci. 6(2013) 799-804. doi: 10.1039/c3ee24058d

  17. [17]

      M. Mrowetz, C. Pirola, E. Selli, Ultrason. Sonochem. 10(2003) 247-254. doi: 10.1016/S1350-4177(03)00090-7

  18. [18]

      C.Y. Ma, J.Y. Xu, X.J. Liu, Ultrasonics 44(2006) E375-E378. doi: 10.1016/j.ultras.2006.05.164

  19. [19]

      E. Selli, Phys. Chem. Chem. Phys. 4(2002) 6123-6128. doi: 10.1039/b205881b

  20. [20]

      C. Minero, F. Catozzo, E. Pelizzetti, Langmuir 8(1992) 481-486. doi: 10.1021/la00038a029

  21. [21]

      M. Ni, M.K.H. Leung, D.Y.C. Leung, K. Sumathy, Renew. Sustain. Energy Rev. 11(2007) 401-425. doi: 10.1016/j.rser.2005.01.009

  22. [22]

      J.S. DuChene, B.C. Sweeny, A.C. Johnston-Peck, et al., Angew. Chem. Int. Ed. 53(2014) 7887-7891. doi: 10.1002/anie.201404259

  23. [23]

      K. Qian, B.C. Sweeny, A.C. Johnston-Peck, et al., J. Am. Chem. Soc. 136(2014) 9842-9845. doi: 10.1021/ja504097v

  24. [24]

      Y.M. Zhai, J.S. DuChene, Y.C. Wang, et al., Nat. Mater. 15(2016) 889. doi: 10.1038/nmat4683

  25. [25]

      K. Qian, W.X. Huang, Z.Q. Jiang, H.X. Sun, J. Catal. 248(2007) 137-141. doi: 10.1016/j.jcat.2007.02.010

  26. [26]

      K. Qian, L.F. Luo, Z.Q. Jiang, W.X. Huang, Catal. Today 280(2017) 253-258. doi: 10.1016/j.cattod.2016.03.035

  27. [27]

      M.S. Chen, D.W. Goodman, Science 306(2004) 252-255. doi: 10.1126/science.1102420

  28. [28]

      A. Primo, T. Marino, A. Corma, et al., J. Am. Chem. Soc. 134(2012) 1892. doi: 10.1021/ja211206v

• 

  Figure 1  (A) HAADF-STEM images of Au/TiO₂, (B) high-resolution image that shows the interface of Au-TiO₂, (C) size distribution of Au nanoparticles in Au/TiO₂.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  (A) High-resolution XPS spectrum of Au 4f peaks acquired over the Au/TiO₂ heterostructures. (B) Photoluminescence spectra of Au/TiO₂ and pristine TiO₂. (C) Band diagrams show the formation of the Schottky barrier. (D) Band diagrams show the change in the electronic structures under different optical excitation conditions. Note: In both band diagrams, CB stands for conduction band and VB stands for valence band. E_F stands for Fermi level.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Amount of evolved H₂ over Au/TiO₂ photocatalyst under different experimental conditions. (A) Sonication only or optical excitation only. (B) The accumulated amount of H₂ as a function of the sonication time under different conditions. (C) Sonication combined with optical excitation.

  下载: 全尺寸图片 幻灯片
•