Morphology-controlled synthesis of WO2.72 nanostructures and their photocatalytic properties

Xiaoxiao Guoab, Xiaoyun Qinbc, Zhenjie Xuebc, Changbo Zhangbc, Xiaohua Sunbcd, Jibo Hou*a and Tie Wang*bc
aSchool of Material Science and Engineering, North University of China, China. E-mail: 13934134279@139.com
bKey Laboratory of Analytical Chemistry for Living Biosystems, Institute of Chemistry, Chinese Academy of Sciences, China. E-mail: wangtie@iccas.ac.cn
cUniversity of Chinese Academy of Sciences, Beijing 100049, P. R. China
dState Key Laboratory of Heavy Oil Processing, Institute of New Energy, China University of Petroleum (Beijing), Beijing 102249, China

Received 3rd April 2016 , Accepted 9th May 2016

First published on 11th May 2016


Abstract

The morphologies of nanomaterials have great influence on their properties. In this work, we used the solvothermal method to prepare tungsten oxides with two different morphologies: WO2.72 nanowires and urchin-like WO2.72 nanostructures. The photocatalytic activities of these two bare WO2.72 nanostuctures were evaluated by their efficiency in the degradation of pollutants, during which the influence of the morphology was taken into account. In the experiments, material structures and oxygen vacancies were altered with the change of the morphology. One-dimensional WO2.72 nanowires with fewer oxygen vacancies showed higher photocatalytic activity than three-dimensional urchin-like WO2.72 nanostructures with more oxygen vacancies. Thus, we ascertained the prominent role of the structure, rather than the number of oxygen vacancies, in enhancing photocatalytic activity. Surface photocurrent (SPC) measurements further confirmed that WO2.72 nanowires were more conducive to photo electron transfer than urchin-like WO2.72 nanostructures, which corresponded with the results of photocatalysis. Compared with commercial nanostructured tungsten oxide, both the WO2.72 nanowires and urchin-like WO2.72 nanostructures exhibited enhanced photocatalytic activities for the degradation of pollutants under UV light irradiation.


1 Introduction

Tungsten oxides are the key ingredient for many advanced functional materials and smart devices. The interest in tungsten oxides dates back to the 17th century when the properties of LiWO3 and techniques for the synthesis of WO3 and NaWO3 were first studied.1 Recently, WOx (tungsten oxides) were investigated extensively in chromism,2 photocatalysis3 and sensing.4 In addition, tungsten oxides have been studied for solar-cells, catalysis5 and surface enhanced Raman scattering.6

In comparison to many other metal oxide nanostructures, such as TiO2, ZnO and NiO, the chromism studies of WOx are much more advanced, while metal oxides such as SnO2, TiO2 and ZnO, as well as their substoichiometric forms, are commonly used in photocatalytic systems and sensors. For solar-cell research, nanostructured WOx has received relatively little attention compared to materials such as TiO2, ZnO and the like. However, the application of WOx in solar cells and photocatalysis should be encouraged, since it offers similar functional properties to TiO2 and ZnO, which are now widely used in relevant industries.

Because of the bright prospects and the wide applications of nanostructured WOx, various morphologies of WOx were investigated rigorously ranging from one-dimensional quantum dots to three-dimensional hierarchical structures. For example, nanorods7–9, nanowire,10 nanoplate,11 nanobelt, mesoporous structure12 and some other nanostructures13,14 were synthesized, for applications in electrochromism, photocatalysis, photoluminescent and gas sensing. Among them, thin films made of nanostructured WOx have been increasingly investigated and applied for the electrochromic devices. At the same time, a mass of approaches for the synthesis of nanostructures WOx have been developed, including vapor, liquid and solid methods. Vapor method is a costly method limited to energetic sources and precise control on the parameters. Solid method is simple. However, it is usually accompanied by the releasing of poisonous gas. Compared with the methods mentioned above, hydrothermal and solvothermal treatment is a facile, cost-effective and well-studied liquid-phase technique, which has the capability to produce WOx with different nanomorphologies.15

WOx are good photocatalysts. Previous studies on the photocatalytic properties of WOx mainly focused on the visible irradiation induced pollutant degradation. However, these studies did not take account of the morphological effect on degradation process. Like other materials, the properties of WOx are strongly relied on the morphology, structure, composition, crystal phase and so on. As we know, the morphology of photocatalyst7 has great influence on its photocatalytic properties.

WOx are well known for their non-stoichiometric properties,15 as the lattice can withstand a considerable number of oxygen vacancies. The existence of a number of oxygen vacancies can narrow the bandgap and increase the conductivity by a large degree, which is beneficial to photocatalysis.

As mentioned above, morphology plays an important role in the performance of WOx. We adopted the solvothermal method to fabricate WO2.72 with different morphologies, to investigate the influence of morphology on the properties. Monoclinic WO2.72 is of great interest owing to its unusual defect structure and promising properties in the nanometer regime. In this report, WO2.72 of different morphologies was obtained when the precursor concentration were changed. The morphology is considered to be the only variable in determining the photocatalytic properties; therefore, the other factors including the precursor and surfactant could be ignored. The prepared WO2.72 nanowires, with a high aspect ratio, were less than 50 nm in diameter and tens of micrometers in length. The urchin-like WO2.72 nanostructures were about 1 μm in diameter. Different photocatalytic efficiencies were shown between nanowires and urchin-like nanostructures. SPC measurements further confirmed that WO2.72 nanowires were more conducive to photo electron transfer than urchin-like WO2.72 nanostructures, which were corresponding with the results of photocatalysis. Both nanowires and urchin-like nanostructures exhibited significantly enhanced photocatalytic activities than commercial nanostructured tungsten oxide.

2 Experimental sections

2.1 Synthesis of urchin-like WO2.72 nanostructures and WO2.72 nanowires

In a typical synthesis, 375 mg of WCl6 was added into 75 mL of absolute ethanol, and a clear yellow solution was formed immediately. The solution was then loaded into a 100 mL Teflon-lined autoclave. The autoclave was sealed, heated in an oven at 180 °C for 10 h, and then naturally cooled to ambient condition. The formed urchin-like WO2.72 nanostructures were collected and washed with ethanol and distilled water several times to remove ions and possible remnants.

Similarly, uniform WO2.72 nanowires with high aspect ratios were obtained as follows. 75 mg of WCl6 was added into 75 mL of ethanol. The mixture was introduced into a 100 mL Teflon-lined stainless steel autoclave, and the container was then closed and maintained at 180 °C for 10 h. After that, the autoclave was cooled to ambient condition. Subsequently, the products were centrifuged and washed with absolute alcohol and distilled water to remove the residual surfactants.

2.2 Photocatalytic properties

The photocatalytic activities of the samples were evaluated by the degradations of rhodamine B (RhB), methylene blue (MB) and methyl orange (MO) in aqueous solutions under UV light from a 175 W mercury lamp equipped with a cut-off filter L42 and a water filter. The photocatalyst (2 mg) was poured into 30 mL of RhB, MB or MO aqueous solution (5 mg L−1) in a Pyrex reactor at room temperature in air respectively. Before the light was turned on, the solution was continuously stirred for 30 min in dark to ensure the establishment of an adsorption–desorption equilibrium. The concentrations of RhB, MO and MB during the degradation were monitored by colorimetry using a UV/vis spectrometer (Shimadzu UV-1800).

Surface photocurrent (SPC) measurements were performed as described elsewhere.16 SPC measurements were carried out by using a self-assembly multi-function spectrometer, which consists of a lock-in amplifier (SR830-DSP) and a 500 W xenon lamp. The lamp was regulated to 23 Hz with a light chopper (SR540), as a light source. The sample was located between a transparent indium tin oxide (ITO) electrode and a Cu electrode, forming a sandwich-like structure (Fig. S1). The SPC signals were recorded at wavelength of 380 nm. The SPC value could reach steady-state during the intervals that the shutter was open or closed.

2.3 Characterizations

Crystallographic information of the samples was collected by using powder X-ray diffraction (XRD, D/max 2500) and high resolution transmission electron microscope (HRTEM, JEM-2100F). Transmission electron microscope (TEM, JEM-1011) images and selected area electron diffraction (SAED, JEM-1011) patterns were obtained to probe crystallinity and crystal structure of the samples. Surface morphologies and energy-dispersive X-ray spectroscopy (EDS) of the samples were obtained using scanning electron microscopy (SEM, S-4800). X-ray photoelectron spectroscopy (XPS) analyses were performed and calibrated by the binding energy of C1s 284.6 eV to check the valence of W atoms. The UV-vis-NIR absorption spectra of the as-synthesized WO2.72 nanostructures were recorded to observe the light absorption. The concentrations of RhB, MB and MO during the degradation were monitored by colorimetry using a UV-vis spectrometer. SPC measurements were performed to observe the photocurrent response of the samples.

3 Results and discussion

3.1 Characterizations

TEM images of the as-synthesized WO2.72 nanostructures revealed the different morphologies with hierarchical and one-dimensional linear structures (Fig. 1a and d). The nanowires were 5–50 nm in diameter and tens of micrometers in length. Urchin-like WO2.72 nanostructures were about 1 μm in diameter. The SEM image further showed that the hierarchical structure was composed of a large number of radial nanowires (Fig. S2). Selection area electron diffraction (SAED) patterns revealed the single-crystalline nature of the urchin-like WO2.72 nanostructures and WO2.72 nanowires (Fig. 1c and f). Energy-dispersive X-ray spectroscopy (EDS) (Fig. S3) confirmed that the samples contained only the elements W and O (W[thin space (1/6-em)]:[thin space (1/6-em)]O = 2.72).
image file: c6ra08551b-f1.tif
Fig. 1 TEM image (a), SEM image (b) and SAED pattern (c) of the urchin-like WO2.72 nanostructures. TEM image (d), SEM image (e) and SAED pattern (f) of the WO2.72 nanowires.

XRD patterns of the urchin-like nanostructures and nanowires displayed mainly two intense diffraction peaks (Fig. 2a), corresponding to the (010) and (020) crystal faces of the monoclinic WO2.72 structure (JCPDS no. 36-0101). All of the other diffraction peaks were much weaker. The narrow (010) and (020) diffraction peaks strongly suggested that the possible crystal growth direction of the samples is [010]. The close-packed planes of the monoclinic WO2.72 crystal are (010), which was in line with the results of the high resolution transmission electron microscopy (HRTEM) analysis. XRD pattern of the urchin-like WO2.72 nanostructures showed the same diffraction peaks with the nanowires. It demonstrated that the crystal phase was insensitive to the morphology.


image file: c6ra08551b-f2.tif
Fig. 2 XRD patterns of the urchin-like WO2.72 nanostructures (black) and WO2.72 nanowires (red) (a). HRTEM images of the urchin-like WO2.72 nanostructures (b) and WO2.72 nanowires (c).

The spacing between the adjacent lattice planes was 0.37 nm (Fig. 2b and c), corresponding to the (010) planes of the monoclinic WO2.72. This result indicates that the preferential growth direction is [010]. WO2.72 is a monoclinic-type structure (P2/m) with lattice parameters of a = 18.318 Å, b = 3.782 Å, and c = 14.028 Å.

The WO2.72 nanostructures showed unusual photophysical properties, as demonstrated by the UV/Vis absorption spectroscopy (Fig. 3). The absorption tails presented in the visible and near infrared (NIR) regions of the absorption spectra give clear evidence that the urchin-like WO2.72 nanostructures and WO2.72 nanowires consisted of a large number of oxygen vacancies,6,13,17 which is very different from those of WO3 nanostructures. For WO3 nanostructures, no absorption was observed in the visible and NIR regions and no absorption tail were present. The spectra of WO2.72 nanostructures were similar to what has been reported previously for W18O49 nanowires,10 and it exhibited a short-wavelength absorption edge at approximately 420 nm, which agreed well with the reported value for the bandgap (Eg = 2.9 eV).


image file: c6ra08551b-f3.tif
Fig. 3 UV/Vis absorption spectra of the WO2.72 nanowires (red) and urchin-like WO2.72 nanostructures (black).

Compared with WO2.72 nanowires, urchin-like WO2.72 nanostructures exhibited an enhanced light harvesting in IR region, which indicated that urchin-like WO2.72 nanostructures had more oxygen vacancies. This result was further confirmed by the XPS spectra. (Fig. 4).


image file: c6ra08551b-f4.tif
Fig. 4 XPS spectra of the urchin-like WO2.72 nanostructures and the WO2.72 nanowires.

The valence of W atoms in the surface of urchin-like WO2.72 nanostructures and WO2.72 nanowires were characterized by XPS (Fig. 4). W5+ species were confirmed by the peaks located at 34.9 eV and 37 eV. And the other two peaks at 36 eV and 38.1 eV were corresponding to W6+. The XPS spectra revealed that the urchin-like WO2.72 nanostructures had more W5+ species comparing to WO2.72 nanowires. The W5+/W6+ ratio of the urchin-like nanostructures is 1/7.83, whereas that of WO2.72 nanowires is 1/8.0. This suggested that the urchin-like WO2.72 nanostructures had more oxygen vacancies than WO2.72 nanowires, which were corresponding to the result of the UV-vis spectra.

In order to investigate how the oxygen vancancies were formed, a structure illustration of WO3 and WO2.72 was made, as in Fig. S4. WO3 crystals are generally formed by corner-sharing WO6 octahedra (Fig. S4, left). For WOx (x < 3), the crystals are formed by corner-sharing WO6, which alternate with octahedra that are partially established by edge-sharing1 (Fig. S4, right). The urchin-like WO2.72 nanostructures had more oxygen vacancies because they may contain more edge-sharing WO6 units than WO2.72 nanowires.

It is supposed that the nanowires were more likely to be oxidized than the urchin-like structures, because more oxygen vacancies were exposed in the environment.

We noted that although the color (Fig. S5), structure, XPS spectrum and UV/Vis absorption spectrum of the WO2.72 nanowires were different from those of the urchin-like WO2.72 nanostructures. However, the XRD patterns of WO2.72 nanowires and urchin-like WO2.72 nanostructures were very similar and were corresponding to the monoclinic WO2.72 structure (JCPDS no. 36-101). These results clearly demonstrated that the oxidations only took place on the surface of the urchin-like WO2.72 nanostructures and WO2.72 nanowires, and that their inside structure were still composed of WO2.72.10

3.2 Growth mechanism

WO2.72 nanowires were obtained when the precursor concentration was 0.5 g L−1 and the reaction time was 10 h. With the increase of precursor concentration, different WO2.72 nanostructure morphologies appeared (Fig. S6). When the precursor concentration increased from 0.5 g L−1 to 1 g L−1, WO2.72 nanowires developed into bundle-like WO2.72. The mixture of urchin-like WO2.72 nanostructures and WO2.72 nanowires were obtained when the precursor concentration increased to 3 g L−1. Once the precursor concentration was equal to 5 g L−1, urchin-like nanostructures composed of a number of nanowires with larger diameters were predominantly obtained. The urchin-like nanostructures turned into nanospheres with the increased precursor concentration.

Time-dependent experiments were carried out to monitor the formation process of the urchin-like WO2.72 nanostructures (Fig. S7). After solvothermal treatment for 3 h, WO2.72 nanowires tangled with each other to form bundle-like WO2.72. The bundle-like WO2.72 continued to pack together and assembled into urchin-like WO2.72 nanostructures at 6 h. Urchin-like WO2.72 nanostructures became compact when reacted for 10 h. Spherical structure was finally formed after 24 h. The XRD patterns of the samples treated in different times were similar (Fig. S8), which demonstrated that the crystal phase and closed-packed planes were remained with the grain growth.

In addition, the influence of temperature on the morphology was also studied (Fig. S9). Fine structure was only obtained at 180 °C. Although the exact formation mechanism of the urchin-like WO2.72 nanostructures was unknown at the moment, we believed the crystal-structure features of the WO2.72 are crucial in the formation of different morphologies.

In conclusion, we proposed a growing process of the nanostructured WO2.72. The alcoholysis of WCl6 was first used to synthesize the WO2.72 nanostructures.18 WO2.72 crystallized with the nucleation and grain growth. The WO2.72 nanocrystals would preferentially grow along the [010] direction. As the diameter of the nanowires is very small (their surface energy is very large), the nanowires tend to tangle with each other to reduce the surface energy. Thus, bundle-like WO2.72 nanostructures were formed and further developed into urchin-like WO2.72 nanostructures. Nevertheless, the exact formation mechanism should be further investigated.

3.3 Photocatalysis

WOx are good photocatalysts. However, previous studies on the photocatalytic properties of WOx mainly focused, for example, on the pollutant degradation under visible irradiation, without considering their morphological effects. As is known to all, the morphology of photocatalyst has great influence on their photocatalytic properties. Herein, we tried to verify the influence of morphology on the photocatalytic properties. The structure and oxygen vacancies were altered with the change of morphology. One-dimensional structure was more conducive to photo electron transfer,19 which would enhance the photocatalytic activity. Meanwhile, the existence of oxygen vacancies can narrow the bandgap and significantly increase the conductivity,1 which is also beneficial to the photocatalytic process. Three-dimensional urchin-like WO2.72 nanostructures had more oxygen vacancies, while one-dimensional WO2.72 nanowires had less oxygen vacancies. To investigate which parameter has more influence on the photocatalytic activities, the efficiency in the pollutant degradation of these two WO2.72 nanostructures were compared.

The photocatalytic degradation of toxic pollutants is of great significance in environmental pollutant treatments, and represents a commonly used approach to characterize the activities of photocatalysts. We focus our studies on the morphology-dependent photocatalytic properties by the degradation of rhodamine B (RhB), methylene blue (MB) and methyl orange (MO). Fig. S10 showed that the absorption of RhB, MB and MO gradually decreased during the photodegradation under UV light irradiation. The corresponding plots for the concentration changes of RhB, MB and MO determined from its characteristic absorption peak were shown in Fig. 5.


image file: c6ra08551b-f5.tif
Fig. 5 Photocatalytic activities of three catalysts for the degradation reactions of MB (a), MO (b), and RhB (c) respectively: commercial WO3 (yellow), urchin-like WO2.72 nanostructures (black), WO2.72 nanowires (red).

As a comparison, the degradation of dyes over commercial WO3 samples and photocatalysis on dyes (without photocatalyst) under UV light irradiation were also measured at the same time. It is obvious that under the same experimental conditions, the urchin-like WO2.72 nanostructures and WO2.72 nanowires exhibited superior photocatalytic activity over the commercial WO3. It might be contributed to an increase in BET surface area (Table S1), as well as the number of surface active sites for our nanostructures. In addition, the dyes cannot be degraded under dark conditions in the presence of the photocatalysts, confirming that the photocatalytic activity indeed originated from the nanostructured WO2.72. Moreover, the oxygen vacancies appearance of tungsten oxide nanostructures might be significant for their enhanced photocatalytic activities. Among the three nanostructured tungsten oxide, WO2.72 nanowires exhibited the highest photocatalytic activity for dye photodegradation. It is most likely that WO2.72 nanowires are of one-dimensional structure, which allowed electrons/holes to travel from the top to the bottom of the line without changing courses.19 This property may result in the decrease of the travel distance and may solve the short electrons/holes diffusion length problem that complicates the three-dimensional urchin-like WO2.72 nanostructures, as will be further demonstrated by the following SPC analysis (Fig. 6). Thus, we got that the structure plays the leading role in enhancing the photocatalytic activity of WO2.72. Furthermore, the oxygen vacancies were not stable during the photocatalytic process. The UV-vis absorption spectra (Fig. S11) after photocatalysis showed no absorption tail in the visible and near infrared (NIR) regions, indicating that the oxygen vacancies on the surface of the samples had disappeared. However, the structures maintained original as shown in the Fig. S12. This result further proves that it was the one-dimensional structure, rather than oxygen vacancies, which played the leading role in enhancing the photocatalytic activity of the photocatalyst.


image file: c6ra08551b-f6.tif
Fig. 6 Surface photocurrent spectra of urchin-like WO2.72 nanostructures (black) and WO2.72 nanowires (red).

The results of the SPC (surface photocurrent) measurements further confirmed that WO2.72 nanowires were more conducive to photo electron transfer than urchin-like WO2.72 nanostructures (Fig. 6). Both urchin-like WO2.72 nanostructures and WO2.72 nanowires revealed fast and uniform photocurrent responses under UV light irradiation. The photocurrent of the WO2.72 nanowires electrode was about double as high as the urchin-like WO2.72 nanostructures. This photocurrent enhancement of the WO2.72 nanowires indicated an enhanced photoinduced separation and transfer of the electrons and holes, which could be attributed to the one-dimensional linear nanostructures.

4 Conclusions

WO2.72 nanowires and urchin-like WO2.72 nanostructures were synthesized by the simple solvothermal method. The morphology of the WO2.72 nanostructures can be controlled by tuning the precursor concentration. WO2.72 nanowires were 10–50 nm in diameter and tens micrometers in length, while the urchin-like WO2.72 nanostructures were about 1 μm in diameter. Both the urchin-like WO2.72 nanostructures and WO2.72 nanowires showed light absorption from the visible to the NIR region. Morphology-dependent photocatalytic activity of bare WO2.72 nanostructures was evaluated by its efficiency in degradation of rhodamine B (RhB), methylene blue (MB) and methyl orange (MO). Compared with urchin-like WO2.72 nanostructures, one-dimensional WO2.72 nanowires with less oxygen vacancies exhibited enhanced photocatalytic activities, due to its linear nanostructures. This result demonstrated that the structure played the leading role in enhancing the photocatalytic activity of WO2.72 nanostructures. This work presents not only a possibility for using oxygen vacancy-rich nanostructured WO2.72 as a functional material in the RhB, methylene blue (MB) and methyl orange (MO) photodegradation, but also an important concept that morphology-control can be used as a strategy to design materials with high photochemical activities.

Acknowledgements

This research was financially supported by the 1000 Young Talents program, the National Natural Science Foundation of China (Grant No. 21422507, 21321003) and the Chinese Academy of Sciences.

References

  1. H. Zheng, J. Z. Ou, M. S. Strano, R. B. Kaner, A. Mitchell and K. Kalantar-zadeh, Adv. Funct. Mater., 2011, 21, 2175 CrossRef CAS.
  2. B. J. Liu, J. Zheng, J. L. Wang, J. Xu, H. H. Li and S. H. Yu, Nano Lett., 2013, 13, 3589 CrossRef CAS PubMed.
  3. D. Chen and J. Ye, Adv. Funct. Mater., 2008, 18, 1922 CrossRef CAS.
  4. T. Kida, A. Nishiyama, Z. Hua, K. Suematsu, M. Yuasa and K. Shimanoe, Langmuir, 2014, 30, 2571 CrossRef CAS PubMed.
  5. Y. Cao, J. Wang, M. Kang and Y. Zhu, J. Mol. Catal. A: Chem., 2014, 381, 46 CrossRef CAS.
  6. S. Cong, Y. Yuan, Z. Chen, J. Hou, M. Yang, Y. Su, Y. Zhang, L. Li, Q. Li, F. Geng and Z. Zhao, Nat. Commun., 2015, 6, 7800 CrossRef CAS PubMed.
  7. A. Yella, M. N. Tahir, S. Meuer, R. Zentel, R. Berger, M. Panthofer and W. Tremel, J. Am. Chem. Soc., 2009, 131, 17566 CrossRef CAS PubMed.
  8. K. Manthiram and A. P. Alivisatos, J. Am. Chem. Soc., 2012, 134, 3995 CrossRef CAS PubMed.
  9. J. Wang, E. Khoo, P. S. Lee and J. Ma, J. Phys. Chem. C, 2008, 112, 14306 CAS.
  10. G. Xi, S. Ouyang, P. Li, J. Ye, Q. Ma, N. Su, H. Bai and C. Wang, Angew. Chem., Int. Ed. Engl., 2012, 51, 2395 CrossRef CAS PubMed.
  11. J. Y. Zheng, G. Song, J. Hong, T. K. Van, A. U. Pawar, D. Y. Kim, C. W. Kim, Z. Haider and Y. S. Kang, Cryst. Growth Des., 2014, 14, 6057 CAS.
  12. J. K. Kim, J. H. Moon, T. W. Lee and J. H. Park, Chem. Commun., 2012, 48, 11939 RSC.
  13. Z. F. Huang, J. Song, L. Pan, F. Lv, Q. Wang, J. J. Zou, X. Zhang and L. Wang, Chem. Commun., 2014, 50, 10959 RSC.
  14. S. Jeon and K. Yong, Chem. Commun., 2009, 45, 7042 RSC.
  15. Z. F. Huang, J. Song, L. Pan, X. Zhang, L. Wang and J. J. Zou, Adv. Mater., 2015, 27, 5309 CrossRef CAS PubMed.
  16. L. Kronik and Y. Shapira, Surf. Sci. Rep., 1999, 37, 1 CrossRef CAS.
  17. T. Nütz and M. Haase, J. Phys. Chem. B, 2000, 104, 8430 CrossRef.
  18. G. Xi, J. Ye, Q. Ma, N. Su, H. Bai and C. Wang, J. Am. Chem. Soc., 2012, 134, 6508 CrossRef CAS PubMed.
  19. X. Shi, I. Y. Choi, K. Zhang, J. Kwon, D. Y. Kim, J. K. Lee, S. H. Oh, J. K. Kim and J. H. Park, Nat. Commun., 2014, 5, 4775 CrossRef CAS PubMed.

Footnote

Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra08551b

This journal is © The Royal Society of Chemistry 2016
Click here to see how this site uses Cookies. View our privacy policy here.