Conductive Tungsten Oxide Nanosheets for Highly ... - ACS Publications

Letter Cite This: Nano Lett. 2017, 17, 7968−7973

pubs.acs.org/NanoLett

Conductive Tungsten Oxide Nanosheets for Highly Efficient Hydrogen Evolution Tingting Zheng,† Wei Sang,† Zhihai He,† Qiushi Wei, Bowen Chen, Hongliang Li, Cong Cao, Ruijie Huang, Xupeng Yan, Bicai Pan, Shiming Zhou,* and Jie Zeng* Hefei National Laboratory for Physical Sciences at the Microscale, Key Laboratory of Strongly-Coupled Quantum Matter Physics of Chinese Academy of Sciences, Department of Chemical Physics, University of Science and Technology of China, Hefei, Anhui 230026, P. R. China S Supporting Information *

ABSTRACT: Exploring efficient and economical electrocatalysts for hydrogen evolution reaction is of great significance for water splitting on an industrial scale. Tungsten oxide, WO3, has been long expected to be a promising non-preciousmetal electrocatalyst for hydrogen production. However, the poor intrinsic activity of this material hampers its development. Herein, we design a highly efficient hydrogen evolution electrocatalyst via introducing oxygen vacancies into WO3 nanosheets. Our first-principles calculations demonstrate that the gap states introduced by O vacancies make WO3 act as a degenerate semiconductor with high conductivity and desirable hydrogen adsorption free energy. Experimentally, we prepared WO3 nanosheets rich in oxygen vacancies via a liquid exfoliation, which indeed exhibits the typical character of a degenerate semiconductor. When evaluated by hydrogen evolution, the nanosheets display superior performance with a small overpotential of 38 mV at 10 mA cm−2 and a low Tafel slope of 38 mV dec−1. This work opens an effective route to develop conductive tungsten oxide as a potential alternative to the state-of-the-art platinum for hydrogen evolution. KEYWORDS: Conductive nanosheet, O vacancies, gap states, HER

H

great promise as a potential alternative to Pt for HER electrocatalysis, the reported activities of WO3 are unsatisfactory, largely ascribed to high ΔGH, inferior electrical conductivity, and limited accessible active sites.32−34 Recently, several advances have been reported in promoting the HER activity of this metal oxide.35−37 For example, a modulation of local atomic structure of WO3 nanoparticles, which produces WO2.9 by a thermal treatment, was reported to activate the metal oxide for hydrogen evolution, where the W sites in the modified structure were proposed to show an enhanced H adsorption strength.35 Growth of oxygen-deficient tungsten oxide on conductive supports such as carbon nanofibers36 or mesoporous nanowires37 was also found to promote HER activity. In those studies, the presence of O vacancies was speculated to play a role in activating this oxide for hydrogen evolution. However, the intrinsic mechanism is yet to be completely elucidated, especially the role of O vacancies in tailoring electronic structure. Herein, we developed a highly efficient hydrogen evolution electrocatalyst via introducing oxygen vacancies into WO3 nanosheets. Based on density-functional-theory (DFT) calculations on the electronic structure of WO3 with O vacancies, O vacancies introduce gap states, which transform the metal oxide

ydrogen, a clean and sustainable energy vector, is a promising alternative to traditional fossil fuels, and its utilization has significant implications for addressing the energy crisis and environmental issues.1,2 Electrocatalytic hydrogen evolution reaction (HER) represents one of the most advanced technologies for hydrogen production.3−6 Although Pt-based materials are, at present, the most active catalysts for HER, their large-scale application is hindered by high cost and lack of abundance.7−10 This limitation has sparked tremendous interest in exploring earth-abundant HER catalysts that could potentially replace Pt.11−25 Generally, to achieve efficient HER performance, a suitable hydrogen adsorption free energy (ΔGH) for electrocatalyst serves as a prerequisite. According to the Sabatier principle,6 the optimal value of ΔGH should be neither too negative nor too positive, meaning that hydrogen is bonded neither too strongly nor too weakly. Meanwhile, a high intrinsic electrical conductivity for the catalyst is requested to ensure a fast electron-transfer process and to reduce the Schottky barrier at the charge transfer interface.26,27 In addition, abundant active sites are also necessary to be exposed at surface to accelerate the electrocatalytic process.28−30 Given this, the development of cost-effective HER electrocatalysts with these three factors simultaneously optimized is desirable but challenging. Owing to earth abundance, tunable composition, and high stability, tungsten oxide, WO3, is considered as one of the most attractive candidates for a wide variety of catalyses.31 Despite © 2017 American Chemical Society

Received: October 17, 2017 Revised: November 23, 2017 Published: November 27, 2017 7968

DOI: 10.1021/acs.nanolett.7b04430 Nano Lett. 2017, 17, 7968−7973

Letter

Nano Letters

Figure 1. (a, b, c) Crystal structure of WO3(010) (√2 × √2) R45° slab with a perfect surface (denoted as W24O72), a surface with one bridging O vacancy (denoted as W24O71), and a surface with all terminal O atoms and one bridging O atom removed (denoted as W24O67), respectively. (d) Corresponding calculated partial DOS of W 5d. (e) Free-energy diagram for hydrogen adsorption at the W site on the WO3(010) slab with different O vacancies.

degenerate semiconductor act more like metal than semiconductor. In our case, WO3 layer with O vacancies should be a degenerate n-type semiconductor with a relatively large number of electrons in the conduction band, giving rise to a potential of high electrical conductivity. Based on the electronic structures of the tungsten oxide with O vacancies, we further evaluated HER activities of the W sites on the (010) slab (see details in Figure 1, Figure S2, Table S1, and Note S1). The hydrogen adsorption free energy, ΔGH, scaling with activation energies, has been successfully used as a descriptor of HER activity.43−45 It has been demonstrated that the value of ΔGH strongly depends on the electronic structures of catalyst. As shown in Figure 1e, for the perfect WO3 slab, the calculated ΔGH at the W sites is about 2.3 eV uphill, indicating that the W sites in bulk WO3 are inert.43−45 With the introduction of O vacancies, the new gap states appearing at the CBM allow favorable hydrogen adsorption and reduce ΔGH dramatically. When five O atoms were removed, ΔGH at the adjacent W sites was reduced to a negative value, comparable to that of platinum. Therefore, our DFT calculations suggest that WO3 with O vacancies serves as a degenerate semiconductor, being a very promising earth-abundant HER catalyst. Enlightened by the theoretical calculations, we presumed that the introduction of O vacancies can activate tungsten oxides for hydrogen evolution. Meanwhile, we noticed that owing to large specific surface area and a great number of active sites, the twodimensional nanosheet is a promising structural motif for the improvement of HER performance. Therefore, we aimed to prepare WO3 nanosheets with ample O vacancies and thus endowed with high conductivity and abundant active sites. Experimentally, WO3 nanosheets rich in O vacancies (WO3-r NSs) were synthesized by liquid exfoliation of the as-prepared tungsten oxide precursors. In comparison, WO3 nanosheets with poor O defects (WO3-p NSs) were also obtained by annealing the as-exfoliated nanosheets in air atmosphere (Figure S4). A representative transmission electron microscopy (TEM) image (Figure 2a) reveals that the as-exfoliated

from a conventional semiconductor with low conductivity to a degenerate one with high conductivity. Moreover, this transformation is followed by a large reduction in ΔGH to close zero, which is optimal for HER. Then we adopted a liquid exfoliation strategy to experimentally prepare WO3 nanosheets with rich O vacancies. Electronic transport measurements confirm the nature of degenerate semiconducting behavior for the defect-rich tungsten oxide. Endowed with desirable ΔGH, high conductivity, and large active surface area provided by two-dimensional nanosheet morphology, the WO3 nanosheets with rich O vacancies exhibit excellent HER catalytic performance with a small overpotential of 38 mV at 10 mA cm−2 and a low Tafel slope of 38 mV dec−1. We started with the first-principles calculations to unveil the electronic structures of tungsten oxide tailored by O vacancies. Three structure models were constructed based on WO3(010) (√2 × √2) R45° slab with a perfect surface, a surface with one bridging O vacancy, and a surface with all terminal O atoms and one bridging O atom removed (Figure 1a−c, and Figure S1). For the perfect surface, that is, bulk WO3, the density of states (DOS) presents a typical semiconductor character, where the Fermi level is located within the middle of band gap without any electron distribution around the Fermi level (Figure 1d), consistent with the previous reported results.29 When O vacancies were introduced, new energy levels appear at the conduction band minimum (CBM) of W 5d. Moreover, the presence of more O vacancies endows the WO3 slab with obviously increased DOS at the CBM of W 5d. Notably, for the two models with O vacancies, though the narrowed band gaps still exist, the Fermi levels cross the CBM of W 5d, which is a typical feature of a degenerate semiconductor.37−41 The degenerate semiconductor is a highly doped semiconductor where the Fermi level is situated in the conduction or valence band, which has been primarily applied in electronic and optoelectronic devices.38−42 As shown in Figure S3, the presence of state density distribution around the Fermi level leads to quite a number of charge carriers, which makes the 7969

DOI: 10.1021/acs.nanolett.7b04430 Nano Lett. 2017, 17, 7968−7973

Letter

Nano Letters

Figure 2. (a) TEM and (b) HRTEM images of WO3-r NSs. (c) XRD patterns, (d) Raman spectra, and (e) XPS spectra for WO3-r NSs, WO3-p NSs, and WO3 bulk.

Figure 3. (a) Temperature dependence of electrical resistivity of WO3-r NSs, WO3-p NSs, and WO3 bulk. (b) Hall coefficient (RH) and carrier concentration (n) of WO3-r NSs, WO3-p NSs, and WO3 bulk at 300 K.

O−W−O. For the exfoliated nanosheets, all these modes are broader with a slight shift toward lower wavenumber as compared to bulk counterpart, probably ascribed to the presence of abundant O vacancies.46 To characterize the surface oxidation state of the products, Xray photoelectron spectroscopy (XPS) was carried out. Figure 2e displays the W 4f core level spectra of WO3 bulk and WO3-r and WO3-p NSs. For WO3 bulk, only two sharp peaks with binding energy at 35.8 and 37.9 eV can be observed, which are ascribed to the characteristic W 4f7/2 and 4f5/2 peaks of W6+ species.47 In the cases of WO3-r and WO3-p NSs, both the peaks are slightly broader with a shoulder at the lower binding energy region. The two extra peaks centered at 34.7 and 36.9 eV correspond to the typical binding energies of W5+ (ref 48), insinuating the coexistence of W5+ ions with the O vacancies for the two nanosheets. In addition, the obviously different peak areas of W5+ indicate that the WO3-r NSs possess much more O vacancies compared with the WO3-p NSs. Moreover, the O 1s XPS spectrum where the peak at 531.6 eV is associated with O vacancy further suggested that WO3-r NSs contained more oxygen vacancies than WO3-p NSs, while WO3 bulk showed no O vacancy (Figure S6).49,50 The calculated stoichiometries based on the relative atomic ratio of W5+/W6+ from the XPS

products take a nanosheet morphology with lateral size of hundreds of nanometers. Figure 2b shows a high-resolution transmission electron microscopy (HRTEM) image of WO3-r NSs. Two sets of lattice fringes are clearly observed with the lattice spacings of 0.384 and 0.365 nm, corresponding to the (002) and (200) planes of tungsten oxide, respectively. In addition, dislocations and distortions, which are probably caused by the formation of oxygen vacancies during the exfoliation process, can be found, suggesting a novel defect-rich structure.29 The atomic force microscopy (AFM) image and the corresponding height profile (Figure S5) further confirm the sheet morphology with a thickness of about 5 nm. Powder X-ray diffraction (XRD) and Raman spectroscopy were performed to reveal the crystalline structure of the obtained products, with the commercial WO3 bulk as reference. As shown in Figure 2c, the diffraction peaks of WO3 bulk, WO3-p NSs, and WO3-r NSs are all indexed to the standard pattern of monoclinic WO3 (JCPDS Card No. 43-1035), indicating the high purity of the products. Moreover, the exfoliated nanosheets exhibit a (020) preferred orientation, in agreement with the HRTEM results. Figure 2d presents the Raman spectroscopy of all the samples, where the peak at 273 and 327 cm−1 are attributed to the bending vibration of O−W−O while the peaks at 716 and 807 cm−1 are attributed to the stretching vibration of 7970

DOI: 10.1021/acs.nanolett.7b04430 Nano Lett. 2017, 17, 7968−7973

Letter

Nano Letters

Figure 4. (a) Polarization curves measured in H2 saturated 0.5-M H2SO4 solution for the WO3 bulk, WO3-p NSs, WO3-r NSs, and Pt/C. (b) Tafel plots. (c) Nyquist plots at η = −100 mV. (d) Polarization curves after 1000 CV cycles with WO3-r NSs.

electronic resistivity and large carrier concentration demonstrate that this oxide was highly conductive. A standard three-electrode setup in 0.5 M H2SO4 electrolyte was applied to conduct the electrocatalytic measurements. Figure 4a shows the representative polarization curves for the WO3 nanosheets with O vacancies, along with the commercial WO3 bulk and 20 wt % Pt/C benchmark for comparison. The WO3 bulk exhibits poor HER activity with a high onset potential surpassing 300 mV. By contrast, the WO3 nanosheets with O vacancies present appreciably lower onset potentials and higher current densities. Remarkably, the WO3-r NSs exhibit impressive HER activity with an onset potential near zero and a small overpotential of −38 mV at the current density of 10 mA cm−2, which is 100 and 207 mV less than those of WO3-p NSs, respectively. The corresponding Tafel plots (Figure 4b) display that the introduction of O vacancies significantly reduces the Tafel slope from 136 mV dec−1 for WO3 bulk to 38 mV dec−1 for WO3-r NSs, approaching that of Pt/C. As such, the ratelimiting step for WO3 bulk is the Volmer reaction, whereas that for WO3-r NSs is the Heyrovsky reaction.24,25 This highly efficient HER performance of the WO3-r NSs is superior to that of most previous tungsten-based HER catalysts and comparable to that of the best non-precious-metal based electrocatalysts (Table S2), which strongly supports our theoretical predictions. The accelerated HER reaction kinetics induced by O vacancies is further reflected by electrochemical impedance spectroscopy (EIS) measured under HER conditions. The Nyquist plots in Figure 4c reveal a much smaller charge transfer resistance (∼64 Ω) for WO3-r NSs than that for WO3-p NSs (∼142 Ω) and WO3 bulk (∼391 Ω). To further assess the durability of the WO3-r NSs, cyclic voltammetry (CV) between −0.3 and +0.1 V was swept 1000 times on the electrodes (Figure 4d). After the repeated potential sweeps, the WO3-r NSs only exhibit a faint loss of activity of 2 mV at the current density of 10 mA cm−2, implying a long-term stable performance. XPS spectroscopy revealed that this slight loss may be associated with a minor oxidation of

peak areas are WO2.82 and WO2.97 for WO3-r and WO3-p NSs, respectively, which approximate the theoretical models. Electrical transport measurements were performed to confirm the theoretical predictions on the electronic structures of tungsten oxides modified by O vacancies. Figure 3a shows the temperature-dependent electrical resistivity for WO3 bulk and WO3-r and WO3-p NSs. A decrease in the electrical resistivity with elevated temperature corresponding to a negative slope value reveals the semiconducting behavior for the WO3 bulk, while the slopes of WO3-r NSs and WO3-p NSs are both approximately equal to zero, revealing that their electrical resistivities are less dependent on the temperature. Previous studies have demonstrated that in highly degenerated compounds the carriers are mainly scattered by ionized impurities, giving rise to conductivity nearly independent of the temperature.41,42 Our result suggests that the nanosheets with O vacancies act as a degenerate semiconductor. Moreover, the electrical resistivities at 300 K are about 1.6 × 10−2 and 9.1 × 10−1 Ω cm for WO3-r and WO3-p NSs, respectively, much lower than that of about 4.0 × 101 Ω cm for the WO3 bulk. Introducing O vacancies significantly enhanced the conductivity of the tungsten oxide nanosheets. Specially, the value of resistivity of the WO3-r NSs approaches that of metals. The Hall coefficient (RH) at 300 K of the three samples was also measured, as shown in Figure S7 and Figure 3b. The negative RH indicates that their charge carriers are all electrons. When ample O vacancies were introduced, the electron concentration (n) increases from about 2.1 × 1019 for WO3 bulk to about 1.8 × 1021 cm−3 for WO3-r NSs. Furthermore, more O vacancies contribute to higher electron concentration when comparing the two nanosheets with O vacancies. These transport properties reveal that the electronic behavior of WO3 nanosheets with O vacancies is in many ways intermediate between semiconductor and metal, analogous to that of a degenerate semiconductor, which is in perfect accordance with our DFT calculations. Especially, for WO3-r NSs, the low 7971

DOI: 10.1021/acs.nanolett.7b04430 Nano Lett. 2017, 17, 7968−7973

Letter

Nano Letters *J.Z. E-mail: [email protected].

electrocatalyst induced by corrosion in the acid condition, where the ratio of W5+/W6+ ions underwent a slight decrease from about 1:1.8 to 1:3.6 after 1000 cycles (Figure S8). In addition, the reaction kinetics of WO3-r NSs after 1000 cycles was revealed by EIS measurement (Figure S9), in which the charge transfer resistance is increased a little, further coinciding with the oxidation of WO3-r NSs during the HER process. The long-term durability of WO3-r NSs was also evaluated by chronoamperometry at a constant potential of −100 mV in 0.5 M H2SO4 (Figure S10), exhibiting that the catalysts endured for a long period of 10 h with a nearly stable current density. The experimental observation of excellent HER performance for WO3-r NSs well agrees with the DFT results, which predicts that the O vacancies introduce gap states around the Fermi level in the tungsten oxide. The appearing states guarantee fast charge transport and allow favorable hydrogen adsorption simultaneously, both of which activate the metal oxide for hydrogen evolution. Meanwhile, for the tungsten oxide nanosheets, the 2D morphology with rich defects is also an important structural factor to achieve superior HER activity, since it provides larger surface areas and abundant active sites. To estimate the effective electrochemically active surface area (ECSA), the double-layer capacitance (Cdl) was derived from the CV measurements for the catalysts (Figure S11). The calculated Cdl values are 57.0, 53.6, and 3.9 mF cm−2 for WO3-r NSs, WO3-p NSs, and WO3 bulk, respectively, indicating that the nanosheets have a high exposure of active sites. In summary, we put forward an atom-level insight into the role of O vacancies in WO3 for hydrogen evolution. DFT calculations revealed that the introduction of O vacancies transforms conventional semiconducting WO3 into a degenerate semiconducting material, with enhanced conductivity and more suitable ΔGH. Experimentally, WO3 nanosheets with rich O vacancies were synthesized by liquid exfoliation. Electronic transport measurements unraveled the nature of the degenerate semiconductor for the O vacancy containing WO3 nanosheets. Benefiting from the O vacancies rendering high electrical conductivity and appropriate ΔGH, along with 2D morphology providing large active surface area, WO3 nanosheets rich in O vacancies exhibit excellent HER catalytic performance approaching that of Pt/C. This work paves the way to design semiconducting metal oxides as an efficient and cost-effective HER catalyst.

■

ORCID

Shiming Zhou: 0000-0002-1597-6060 Jie Zeng: 0000-0002-8812-0298 Author Contributions †

T.Z., W.S., and Z.H. contributed equally to this work.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by Collaborative Innovation Center of Suzhou Nano Science and Technology, MOST of China (2014CB932700), NSFC (21573206 and 51371164), Key Research Program of Frontier Sciences of the CAS (QYZDBSSW-SLH017), Anhui Provincial Key Scientific and Technological Project (1704a0902013), and Fundamental Research Funds for the Central Universities. The computational center of USTC is acknowledged for computational support.

■

(1) Turner, J. A. Science 2004, 305, 972−974. (2) Seh, Z. W.; Kibsgaard, J.; Dickens, C. F.; Chorkendorff, I.; Nørskov, J. K.; Jaramillo, T. F. Science 2017, 355, No. eaad4998. (3) Landman, A.; Dotan, H.; Shter, G. E.; Wullenkord, M.; Houaijia, A.; Maljusch, A.; Grader, G. S.; Rothschild, A. Nat. Mater. 2017, 16, 646−651. (4) Montoya, J. H.; Seitz, L. C.; Chakthranont, P.; Vojvodic, A.; Jaramillo, T. F.; Nørskov, J. K. Nat. Mater. 2017, 16, 70−81. (5) Nocera, D. G. Acc. Chem. Res. 2017, 50, 616−619. (6) Greeley, J.; Jaramillo, T. F.; Bonde, J.; Chorkendorff, I.; Nørskov, J. K. Nat. Mater. 2006, 5, 909−913. (7) Chen, Z.; Ye, S.; Wilson, A. R.; Ha, Y.-C.; Wiley, B. J. Energy Environ. Sci. 2014, 7, 1461−1467. (8) Sheng, W.; Zhuang, Z.; Gao, M.; Zheng, J.; Chen, J. G.; Yan, Y. Nat. Commun. 2015, 6, 5848. (9) Wang, P.; Jiang, K.; Wang, G.; Yao, J.; Huang, X. Angew. Chem. 2016, 128, 13051−13055. (10) Fan, Z.; Luo, Z.; Huang, X.; Li, B.; Chen, Y.; Wang, J.; Hu, Y.; Zhang, H. J. Am. Chem. Soc. 2016, 138, 1414−1419. (11) Roger, I.; Shipman, M. A.; Symes, M. D. Nat. Rev. Chem. 2017, 1, 0003. (12) Andreiadis, E. S.; Jacques, P. A.; Tran, P. D.; Leyris, A.; Chavarot-Kerlidou, M.; Jousselme, B.; Matheron, M.; Pecaut, J.; Palacin, S.; Fontecave, M.; Artero, V. Nat. Chem. 2013, 5, 48−53. (13) Konkena, B.; Junge Puring, K.; Sinev, I.; Piontek, S.; Khavryuchenko, O.; Dürholt, J.P.; Schmid, R.; Tüysüz, H.; Muhler, M.; Schuhmann, W.; Apfel, U.-P. Nat. Commun. 2016, 7, 12269. (14) Jin, Y.; Wang, H.; Li, J.; Yue, X.; Han, Y.; Shen, P. K.; Cui, Y. Adv. Mater. 2016, 28, 3785−3790. (15) Voiry, D.; Yamaguchi, H.; Li, J.; Silva, R.; Alves, D. C.; Fujita, T.; Chen, M.; Asefa, T.; Shenoy, V.; Eda, G.; Chhowalla, M. Nat. Mater. 2013, 12, 850−855. (16) Li, Y.; Wang, H.; Xie, L.; Liang, Y.; Hong, G.; Dai, H. J. Am. Chem. Soc. 2011, 133, 7296−7299. (17) Staszak-Jirkovský, J.; Malliakas, C. D.; Lopes, P. P.; Danilovic, N.; Kota, S. S.; Chang, K. C.; Genorio, B.; Strmcnik, D.; Stamenkovic, V. R.; Kanatzidis, M. G.; Markovic, N. M. Nat. Mater. 2015, 15, 197− 203. (18) Popczun, E. J.; Read, C. G.; Roske, C. W.; Lewis, N. S.; Schaak, R. E. Angew. Chem. 2014, 126, 5531−5534. (19) Li, H.; Tsai, C.; Koh, A. L.; Cai, L.; Contryman, A. W.; Fragapane, A. H.; Zhao, J.; Han, S. H.; Manoharan, H. C.; AbildPedersen, F.; Nørskov, J. K.; Zheng, X. Nat. Mater. 2016, 15, 48−53. (20) Cao, B.; Veith, G. M.; Neuefeind, J. C.; Adzic, R. R.; Khalifah, P. G. J. Am. Chem. Soc. 2013, 135, 19186−19192.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.nanolett.7b04430. Experimental procedures, atom models for WO3(010) slab, structure models for W24O72, W24O71, and W24O67, schematic adsorption sites for H on atomic models, adsorption energy, zero-point energy (ZPE) ,and free energy of adsorbed H on different adsorption sites, TEM image of WO3 nanosheet, AFM and XPS spectrum of WO3-r nanosheet, CVs of WO3 bulk, WO3-r NSs, and WO3-p NSs, and comparison of the reported HER electrocatalysts in acidic aqueous media (PDF)

■

REFERENCES

AUTHOR INFORMATION

Corresponding Authors

*S.Z. E-mail: [email protected]. 7972

DOI: 10.1021/acs.nanolett.7b04430 Nano Lett. 2017, 17, 7968−7973

Letter

Nano Letters (21) Zhang, J.; Qu, L.; Shi, G.; Liu, J.; Chen, J.; Dai, L. Angew. Chem. 2016, 128, 2270−2274. (22) Zhang, J.; Wang, T.; Liu, P.; Liao, Z.; Liu, S.; Zhuang, X.; Chen, M.; Zschech, E.; Feng, X. Nat. Commun. 2017, 8, 15437. (23) Morales-Guio, C. G.; Hu, X. Acc. Chem. Res. 2014, 47, 2671− 2681. (24) He, R.; Hua, J.; Zhang, A.; Wang, C.; Peng, J.; Chen, W.; Zeng, J. Nano Lett. 2017, 17, 4311−4316. (25) Yu, L.; Xia, B. Y.; Wang, X.; Lou, X. W. Adv. Mater. 2016, 28, 92−97. (26) Geng, X.; Sun, W.; Wu, W.; Chen, B.; Al-Hilo, A.; Benamara, M.; Zhu, H.; Watanabe, F.; Cui, J.; Chen, T. P. Nat. Commun. 2016, 7, 10672. (27) Zhao, X.; Zhang, H.; Yan, Y.; Cao, J.; Li, X.; Zhou, S.; Peng, Z.; Zeng, J. Angew. Chem., Int. Ed. 2017, 56, 328−332. (28) Deng, D.; Novoselov, K. S.; Fu, Q.; Zheng, N.; Tian, Z.; Bao, X. Nat. Nanotechnol. 2016, 11, 218−230. (29) Zhang, N.; Li, X.; Ye, H.; Chen, S.; Ju, H.; Liu, D.; Lin, Y.; Ye, W.; Wang, C.; Xu, Q.; Zhu, J.; Song, L.; Jiang, J.; Xiong, Y. J. Am. Chem. Soc. 2016, 138, 8928−8935. (30) Wu, Z.; Li, X.; Liu, W.; Zhong, Y.; Gan, Q.; Li, X.; Wang, H. ACS Catal. 2017, 7, 4026−4032. (31) Huang, Z. F.; Song, J.; Pan, L.; Zhang, X.; Wang, L.; Zou, J. J. Adv. Mater. 2015, 27, 5309−5327. (32) Ganesan, R.; Gedanken, A. Nanotechnology 2008, 19 (2), 025702. (33) Xie, X.; Mu, W.; Li, X.; Wei, H.; Jian, Y.; Yu, Q.; Zhang, R.; Lv, K.; Tang, H.; Luo, S. Electrochim. Acta 2014, 134, 201−208. (34) Zheng, H.; Mathe, M. Int. J. Hydrogen Energy 2011, 36, 1960− 1964. (35) Li, Y. H.; Liu, P. F.; Pan, L. F.; Wang, H. F.; Yang, Z. Z.; Zheng, L. R.; Hu, P.; Zhao, J. H.; Gu, L.; Yang, H. G. Nat. Commun. 2015, 6, 8064. (36) Chen, J.; Yu, D.; Liao, W.; Zheng, M.; Xiao, L.; Zhu, H.; Zhang, M.; Du, M.; Yao, J. ACS Appl. Mater. Interfaces 2016, 8, 18132−18139. (37) Wu, R.; Zhang, J.; Shi, Y.; Liu, D.; Zhang, B. J. Am. Chem. Soc. 2015, 137, 6983−6986. (38) Wolff, P. A. Phys. Rev. 1962, 126, 405−412. (39) Yun, J. N.; Zhang, Z. Y.; Yan, J. F.; Zhao, W. J. Appl. Phys. 2010, 107 (10), 103711. (40) Chandrasekaran, K.; Vijayaraghavan, R.; Varadaraju, U. V. Mater. Chem. Phys. 1998, 56, 63−69. (41) Regragui, M.; Jousseaume, V.; Addou, M.; Outzourhit, A.; Bernede, J. C.; El Idrissi, B. Thin Solid Films 2001, 397, 238−243. (42) Ho, C.-H.; Chan, C.-H.; Tien, L.-C.; Huang, Y.-S. J. Phys. Chem. C 2011, 115, 25088−25096. (43) Zheng, Y.; Jiao, Y.; Jaroniec, M.; Qiao, S. Z. Angew. Chem., Int. Ed. 2015, 54, 52−65. (44) Chou, S. S.; Sai, N.; Lu, P.; Coker, E. N.; Liu, S.; Artyushkova, K.; Luk, T. S.; Kaehr, B.; Brinker, C. J. Nat. Commun. 2015, 6, 8311. (45) Cabán-Acevedo, M.; Stone, M. L.; Schmidt, J. R.; Thomas, J. G.; Ding, Q.; Chang, H. C.; Tsai, M.-L.; Jin, S.; He, J.-H. Nat. Mater. 2015, 14, 1245−1251. (46) Yan, J.; Wang, T.; Wu, G.; Dai, W.; Guan, N.; Li, L.; Gong, J. Adv. Mater. 2015, 27, 1580−1586. (47) Ma, M.; Zhang, K.; Li, P.; Jung, M. S.; Jeong, M. J.; Park, J. H. Angew. Chem. 2016, 128, 11998−12002. (48) Xi, G.; Ouyang, S.; Li, P.; Ye, J.; Ma, Q.; Su, N.; Bai, H.; Wang, C. Angew. Chem., Int. Ed. 2012, 51, 2395−2399. (49) Cai, Z.; Bi, Y.; Hu, E.; Liu, W.; Dwarica, N.; Tian, Y.; Li, X.; Kuang, Y.; Li, Y.; Yang, X.-Q.; Wang, H.; Sun, X. Adv. Energy Mater. 2017, 1701694. (50) Bao, J.; Zhang, X.; Fan, B.; Zhang, J.; Zhou, M.; Yang, W.; Hu, X.; Wang, H.; Pan, B.; Xie, Y. Angew. Chem. 2015, 127, 7507−7512.

7973

DOI: 10.1021/acs.nanolett.7b04430 Nano Lett. 2017, 17, 7968−7973