Spray coated ultrathin films from aqueous tungsten molybdenum oxide nanoparticle ink for high contrast electrochromic applications

Abstract

Ultrathin tungsten molybdenum oxide nanoparticle films were fabricated from aqueous ink by a spray coating technique. With the in situ heating of the hot plate during the spray coating process, the detrimental effects of oxygen vacancies on electrochromic (EC) materials could be eliminated. The spray coated ultrathin films exhibit higher contrast than the drop casted films, which would provide a versatile and promising platform for energy-saving smart (ESS) windows, batteries, and other applications.

---

Assembly technologies would substantially affect the physicochemical properties and, ultimately, the performance of thin films.¹ Thus, the development of compliant fabrication of films is critical to the emerging electronics technology. Recently, EC devices have attracted increased attention due to their capability of tuning down energy consumption with their application in ESS windows. Therefore, a lot of research has been actively explored to obtain high performance EC films through various methods, including the Langmuir–Blodgett technique,² filtration,³ spin coating,^4,5 layer-by-layer deposition,^6,7 electrodeposition,⁸ the anodic growth method,^9–11 hydrothermal techniques,^12–14 and drop casting.^15,16 However, each of these methods mentioned above has one or more characteristic drawbacks, including complicated processes and high energy consumption, leading to the inconvenience of large area fabrication. The spray process has been proven to be a versatile and facile method to fabricate electrodes, and is considered to be one of the most promising film fabrication techniques. This is widely applied because the spray process can be done at high production speed and is compatible with large-scale electrode fabrication.^17,18

Recently, molybdenum doped tungsten oxide has attracted much attention because it could exhibit a broadened optical absorption band.^19,20 Besides, higher electrochromic efficiencies are expected as a result of enhanced electron intervalence transfer between Mo⁵⁺ and W⁶⁺ states, in addition to Mo⁵⁺, Mo⁶⁺ and W⁵⁺, W⁶⁺ transitions.^21,22 Thus, a spray coating technique was utilized to prepare molybdenum doped tungsten oxide films due to their enhanced EC performance.^23,24 However, the few reports regarding spray coated molybdenum doped tungsten oxide films had paid too much attention to the molecular precursors and finally the films were of bulk structures, which would greatly increase the diffusion paths of lithium ions in the active materials. To solve these issues, it is attractive to spray pre-formed tungsten molybdenum oxide nanoparticle ink, which could produce nanoparticle films and thus greatly shorten the diffusion paths of lithium ions in the EC materials and facilitate intimate contacts between EC materials and the electrolyte/current collector. The investigation into spray coating pre-formed tungsten molybdenum oxide provides potential insights into the ion transport mechanism and needs to be actively explored.

In this study, we firstly designed a facile chemical method to synthesize doped tungsten oxide nanoparticles with ultrasmall sizes by adding MoO₃ in a typical synthesis procedure of pure tungsten oxide. In the typical synthesis of pure tungsten oxide, hydrated tungsten oxide was prepared via a hydrothermal method (see the experimental section). Fig. 1(a) shows the X-ray powder diffraction (XRD) pattern of the pure hydrated tungsten oxide powder. All peaks of the as-prepared powder can be well indexed to the orthorhombic phase of WO₃·0.33H₂O (JCPDS Card No. 54-1012). The FE-SEM picture (Fig. 1(b)) shows that the sample has a cluster structure composed of nanoplates with a thickness of about 30 nm. The formation of the nanoplates could be attributed to the natural layered structure of the hydrated tungsten oxide²⁵ and the presence of ethylene glycol (EG) as explained in our previous report.¹² The average size of the nanoplate clusters is estimated to be 1 μm, which would increase the diffusion paths of intercalation ions in the solid phase and thus lead to a slow response. Moreover, the big sizes of the nanoplate clusters make it difficult to get uniform ink and films. As mentioned above, the doping with molybdenum will not only improve the EC performance, but will also increase the crystalline disorder and affect the grain growth.^19–24 In this case, molybdenum oxide was utilized to obtain doped tungsten oxide with smaller dimensions under the same hydrothermal conditions. Thus, the desired nanoparticulate tungsten molybdenum oxide can be synthesized via this novel and facile wet chemical doping method. Fig. 1(c) shows the XRD pattern of the as-synthesized powder. The optical photo of the as-prepared powder shows a dark blue color, which suggests the nonstoichiometric nature of the material and the existence of oxygen vacancies.^4,26–28 From the XRD pattern, it is confirmed that the as-synthesized powder is tungsten molybdenum oxide (W_0.71Mo_0.29O_3−x) with a monoclinic phase. The obvious broadened reflection peaks imply smaller crystalline sizes of the particles. The FE-SEM image (Fig. 1(d)) shows that the nanoplate clusters could be changed to nanoparticles upon the addition of MoO₃ with a drastic reduction in cluster size, in agreement with the XRD results. The TEM image (Fig. 1(e)) of the tungsten molybdenum oxide reveals that the nanoparticles with nearly spherical shapes possess an average diameter of 20 nm. The distances between the adjacent lattice fringes were measured to be 0.31 and 0.253 nm, corresponding to the lattice spacing of the (−112) and (−212) planes. As shown in the inset of Fig. 1(e), the uniform ink exhibits a pale-blue color, which is typically related to the presence of oxygen vacancies.^4,26–28

Fig. 1 (a) XRD pattern of as-prepared pure hydrated tungsten oxide. (b) FE-SEM images of the as-prepared hydrated tungsten oxide powder. (c) XRD pattern of the as-prepared tungsten molybdenum oxide. (d) FE-SEM image of the as-prepared tungsten molybdenum oxide powder. (e) Low-magnification and (f) high-resolution TEM images of the as-prepared tungsten oxide materials confirming the successful achievement of W_0.71Mo_0.29O_3−x nanoparticles with average sizes of 20 nm. Inset in (e), an optical photo of the dispersion in DI water showing pale-blue color, which suggests the existence of oxygen vacancies.

In addition to XRD and TEM analyses, X-ray photoelectron spectroscopy (XPS) is used to identify the stoichiometry of the doped nanoparticles as shown in Fig. 2. The XPS survey spectra shown in Fig. 2(a) indicate that the elements W, Mo, O exist in tungsten molybdenum oxide nanoparticles without other impurities. Corresponding high-resolution Mo 3d and W 4f XPS spectra are used to identify the stoichiometry of the doped nanoparticles and are shown in Fig. 2(b and c). From the Mo 3d XPS spectrum (Fig. 2b), it is found that Mo⁶⁺ and Mo⁵⁺ ions coexist in the nanoparticles. The Mo⁵⁺ peaks (centered at 234.8 and 231.6 eV) are as obvious as the Mo⁶⁺ peaks (centered at 236.1 and 232.9 eV), suggesting a significant sub-stoichiometry for the doped nanoparticles. The W 4f XPS core-level spectrum of the nanoparticles exhibits three peaks corresponding to W 4f_7/2, W 4f_5/2 and W 5p_3/2, which are located at 35.5 eV, 37.7 eV and 41.3 eV, indicating that W in the doped nanoparticles is at its highest oxidation state (W⁶⁺). These results reveal that Mo was introduced into the WO₃ lattice accompanied by the emergence of oxygen vacancies. The introduced Mo in the WO₃ lattice will highly increase the structural disorder, which could prevent the preferred orientation growth and thus nanoparticles with small sizes could be prepared.

Fig. 2 (a) XPS survey spectrum of tungsten molybdenum oxide nanoparticles. (b) Mo 3d XPS spectrum. (c) W 4f XPS spectrum.

To investigate the electrochromic performance of the as-prepared W_0.71Mo_0.29O_3−x nanoparticles, the pale-blue ink was drop casted on FTO glass to form an EC film with a thickness of around 1 μm (see Fig. 3(a)). Due to the detrimental effects of oxygen vacancies in EC materials,²⁹ the drop casted film shows a low modulation range (23.9%) even though there is a lot of active material on the film. To eliminate the detrimental effects of oxygen vacancies, a facile in situ oxidization with heating of the hot plate during the spray coating process was utilized as the scheme illustrated in Fig. 4 (a)and (b). An ink containing W_0.71Mo_0.29O_3−x nanoparticles is deposited as a continuous film on FTO glass using the spray coating technique (Fig. 4(a)). Subsequent removal of DI water and in situ oxidization with heating of the hot plate could induce the self-assembly of the W_0.71Mo_0.29O₃ film (Fig. 4(b)). The pale yellow film, the Mo 3d XPS spectrum of the spray coated film and the drop casted film reveal that the oxygen vacancies have been eliminated (see Fig. S1 and S2 in the ESI†). Fig. 4(c) and (d) exhibit the surface morphology and a cross-sectional view of the as-prepared spray coated W_0.71Mo_0.29O₃ film. W_0.71Mo_0.29O₃ nanoparticles are clearly seen with porous structures. The cross-section image (inset in Fig. 4(c)) shows that the thickness of the as-prepared spray coated W_0.71Mo_0.29O₃ film is around 200 nm, which is obviously thinner than that of the drop casted film (which is around 1 μm) and previous reports.^11,12 The surface roughness of the film is also shown in Fig. 4, with amplitude of 60 nm for the region investigated. EDS (Energy Dispersive Spectroscopy) mapping of molybdenum and tungsten atoms was conducted and is shown in Fig. S3 (ESI†). The uniform signal coverage of the elements across the selected area confirms that the spray coated film consists of tungsten molybdenum oxide.

Fig. 3 (a) FE-SEM images of the drop casted tungsten molybdenum oxide film. The inset shows the cross-sectional view of the film. (b) UV-vis transmittance spectra of the drop casted film measured at −1.0 and +1.0 V, respectively.

Fig. 4 (a) Pale blue solution is sprayed onto FTO glass. (b) Coated substrate is dried and oxidized subsequently as the droplet is sprayed onto the hot FTO glass. FE-SEM images of the spray coated W_0.71Mo_0.29O₃ film: (c) low magnification, (d) higher magnification. The inset shows the cross-sectional view of W_0.71Mo_0.29O₃ film. (e–g) AFM images and rugosity measurements of the as-prepared spray coated W_0.71Mo_0.29O₃ film.

The EC phenomenon of the spray coated W_0.71Mo_0.29O₃ film was investigated via an optical modulation test and characterization results are shown in Fig. 5(a). The digital photographs shown in Fig. 5(a) reveal that the logo placed below the film can be seen clearly, indicating fairly good transmittance of the film. The UV-vis transmittance spectra of the spray coated W_0.71Mo_0.29O₃ film in the colored and bleached states were measured at −1.0 V and +1.0 V, respectively (Fig. 5(a)). The transmittance of the spray coated W_0.71Mo_0.29O₃ film at 632.8 nm is measured to be about 80% when bleached, which is much higher than that of the drop casted film (Fig. 3(b)) and our previous reports.^12,13,30,31 In addition to the high transmittance of the bleached state, the spray coated W_0.71Mo_0.29O₃ film also has a high contrast when colored at a low voltage. The contrast of the spray coated W_0.71Mo_0.29O₃ film at 632.8 nm after applying voltage of −1.0 V is calculated to be 42.9%, which is much higher than that of the drop casted film (Fig. 3(b)) and previous reports.^31–33 To quantify the EC performance, the contrast density is defined as the contrast provided per unit film thickness. It has been calculated for the drop casted film and the spray coated film using the following equation:

Contrast density = (T_bleach − T_color)/D, (1)

where D is the thickness of the film. As shown in Fig. 5(b), the contrast density of the spray coated W_0.71Mo_0.29O₃ film is about 10 times of that of the drop casted W_0.71Mo_0.29O_3−x film. Besides, the contrast density of the spray coated film is also fairly good compared to that calculated from our previous report (see Fig. S4 in the ESI†).¹² This high contrast density can be attributed to the elimination of oxygen vacancies and the increased electron transitions between two types of metallic sites in the molybdenum doped tungsten oxide.

Fig. 5 (a) UV-vis transmittance spectra of the spray coated W_0.71Mo_0.29O₃ film in its colored and bleached state. (b) Contrast density of the spray coated film and drop casted film.

The assessment of the coloration–bleaching kinetics are vital for the evaluation of optical and electronic properties of the EC films.³⁴ Therefore, in situ normalized transmittance changes measured at an optical wavelength of 632.8 nm were measured using the chronoamperometric (CA) technique (Fig. 6). The coloration and bleaching times are always defined as the time taken to reach 90% of the optical modulation at a 632.8 nm wavelength.^35,36Fig. 6 shows the dynamic coloration/bleaching switching time for the drop casted W_0.71Mo_0.29O_3−x film and the spray coated W_0.71Mo_0.29O₃ film. For the drop casted film, the coloration time t_c and bleaching time t_b are found to be 5.5 and 7 s, respectively. Meanwhile, for the spray coated film, the coloration time t_c and bleaching time t_b are found to be 10 and 7.5 s, respectively. The response times of both films are all much faster than those of previous reports on pure tungsten oxide nanostructures.^{12,13,37–39} Electrochemical impedance spectroscopy (EIS) measurements were conducted to investigate the intrinsic reason why the spray coated film exhibited longer responsive time than the drop casted film. The Nyquist plots of both spray coated and drop casted films are shown in Fig. 6(b). The high-frequency arcs correspond to the charge-transfer impedance on the electrode/electrolyte surface. The semicircle of the spray coated film is similar to that of the drop casted film, indicating similar charge-transfer resistance. The inclined line at low frequency is related to the ion diffusion process. The line of the spray coated film exhibits a larger slope, suggesting its faster ion-diffusion process. These results demonstrate that the spray coated film possesses fast ion diffusion, and the longer responsive time of the spray coated film than that of the drop casted film can be attributed to the larger transmittance variation.

Fig. 6 (a) In situ normalized transmittance for the spray coated film and the drop casted film at ±1 V applied potentials for an optical wavelength of 632.8 nm. (b) Nyquist plots of the spray coated film and drop casted film.

Coloration efficiency (CE) is defined by the following formulas:

CE = ΔOD/(Q/A) (2)

ΔOD = log(T_b/T_c) (3)

where T_b and T_c refer to the transmittances of the film in its bleached and colored states, respectively. The calculated CE of the spray coated W_0.71Mo_0.29O₃ film is 36.3 cm² C⁻¹ at the wavelength of 632.8 nm (see Fig. S5 in the ESI†), which is comparable to other crystalline tungsten oxides.^{8,10,32,38–40}

Electrochemical stability is another essential parameter for EC materials. It is well known that in an electrochromic film, the degradation of optical modulation and reversibility is induced by Li⁺ ion trapping.⁴¹ Thus, the electrochemical stabilities of as-prepared films were characterized by the CA technique using square potentials (between −1.0 and 1.0 V). As shown in Fig. 7, the peak current densities of the spray coated film have almost a constant value during 2000 cycles, while those of the drop casted film decreased sharply during 500 cycles, indicating the high stability of the spray coated film. In addition, the spray coated film can stably keep the nanoparticle structure after 2000 cycles, while the nanoparticles turned into a dense film after 2000 cycles for the drop casted film (see Fig. S6 in the ESI†). These results indicate that the spray coated film possesses good cycling durability.

Fig. 7 Peak current evolution of the as-prepared films during the step chronoamperometric cycles.

Conclusions

In summary, a novel and facile method was used to synthesize W_0.71Mo_0.29O_3−x nanoparticles, and an in situ oxidization spray coating technique was utilized to assemble W_0.71Mo_0.29O₃ films. The spray coated W_0.71Mo_0.29O₃ film shows 42.9% optical contrast at a low voltage, with 10 s coloring and 7.5 s bleaching (to 90%), and a coloration efficiency of 36.3 cm² C⁻¹. Such a spray coated W_0.71Mo_0.29O₃ film demonstrates improved EC behaviors, which possess promising properties for potential applications in energy-saving smart windows and displays. Besides, due to the small size, excellent electrochemical accessibility and the facile film fabrication technique, the thin films of W_0.71Mo_0.29O₃ nanoparticles obtained here hold tremendous promise for nano-electronics, water photo-splitting cells and batteries.

Experimental section

Synthesis of precursor

All solvents and chemicals were of analytical grade and were used without further purification. The precursor solution was prepared through a modified approach based on our previously reported method.^12,13,30,31 In a typical synthesis route, H₂WO₄ (5 g) and MoO₃ (1.44 g) were added to 30 wt% H₂O₂ (60 mL) and heated at 95 °C with stirring. The resulting clear solution was diluted to 200 mL with equal volumes of DI water and EG, giving a concentration of 0.1 M. For comparison, a precursor solution without molybdenum doping was prepared without adding MoO₃via the same method.

Synthesis of WO₃·0.33H₂O and W_0.71Mo_0.29O_3−x

The precursor solution (12 mL) and DI water (12 mL) were added into a 40 mL Teflon-lined stainless steel autoclave and maintained at 120 °C for 5 h. After performing the reaction, the products were collected by centrifugation, and thoroughly washed with ethanol and DI water. Then the product was redispersed in DI water to form a pale blue ink for assembly of EC films.

Preparation of EC films

The obtained ink was diluted to 0.5 mg mL⁻¹ and then 30 drops were drop-casted onto FTO glass covering 8 × 35 mm² area to get a drop casted EC film.

The spray coated film was prepared based on the methods in our previous report.⁴² 2.5 mL diluted 0.5 mg mL⁻¹ ink was deposited onto pre-cleaned FTO glass substrates (2.5 × 3.5 cm²) by a spray gun in a fume hood. The ink could be dried immediately after being deposited onto the substrates by heating of a hot plate. Meanwhile, the oxygen vacancies could be eliminated due to the in situ heat treatment and a pale yellow EC film could be prepared.

Characterization

X-ray diffraction (XRD, Shimadzu discover diffractometer with Cu Kα-radiation (λ = 1.5406 Å)), field-emission scanning electron microscopy (FE-SEM, JSM 7600F) and transmission electron microscopy (TEM, JEOL2010) were used to identify the morphology, structure and composition of the products, respectively. The sample roughness was measured using Atomic Force Microscopy (AFM) with a Cyphers S (Asylum Research). XPS measurements were performed on a Thermo Scientific K-Alpha spectrometer with a monochromatic Al Kα X-ray source of 1486.6 eV. The ultraviolet-visible (UV-vis) spectra were recorded on a UV-vis spectrophotometer (Shimadzu UV3600). The in situ spectro-electrochemical properties of individual films were carried out in a quartz cell at room temperature in the UV-vis spectrophotometer, and the voltage supply was from Solartron 1470E. The three-electrode cell consisted of an Ag wire as the reference electrode, a Pt wire as the counter electrode and the as-synthesized EC film as the working electrode. EIS measurement was conducted by applying an AC voltage of 5 mV over a frequency range of 0.1 Hz to 100 kHz using a Zahner electrochemical workstation (Zennium CIMPS-1). The electrolyte used in this paper was 1 M LiClO₄ in PC.

Acknowledgements

We gratefully acknowledge the financial support by the Natural Science Foundation of China (No. 51172042), the Specialized Research Fund for the Doctoral Program of Higher Education (20110075130001), the Science and Technology Commission of Shanghai Municipality (13JC1400200), the Shanghai Natural Science Foundation (15ZR1401200), the Program for Professor of Special Appointment (Eastern Scholar) at Shanghai Institutions of Higher Learning, the Innovative Research Team in University (IRT1221), the Program of Introducing Talents of Discipline to Universities (No. 111-2-04) and the Fundamental Research Funds for the Central Universities (2232014A3-06). The work was also supported by the Fundamental Research Funds for the Central Universities (15D310604). We are grateful to Nanyang Technological University for supporting the Special Excellent PhD International Visit Program. This work was also partially supported by Donghua University.

Notes and references

1. J. J. Richardson, M. Björnmalm and F. Caruso, Science, 2015, 348, aaa2491.
2. J. Liu, J. Zheng, J. Wang, J. Xu, H. Li and S. Yu, Nano Lett., 2013, 13, 3589.
3. L. Liang, J. Zhang, Y. Zhou, J. Xie, X. Zhang, M. Guan, B. Pan and Y. Xie, Sci. Rep., 2013, 3, 1936.
4. S. Cong, Y. Tian, Q. Li, Z. Zhao and F. Geng, Adv. Mater., 2014, 26, 4260.
5. S. Y. Park, J. M. Lee, C. Noh and S. U. Son, J. Mater. Chem., 2009, 19, 7959.
6. M. Cui, W. S. Ng, X. Wang, P. Darmawan and P. S. Lee, Adv. Funct. Mater., 2015, 25, 401.
7. H. Ling, L. Liu, P. S. Lee, D. Mandler and X. Lu, Electrochim. Acta, 2015, 174, 57.
8. S. H. Lee, R. Deshpande, P. A. Parilla, K. M. Jones, B. To, A. H. Mahan and A. C. Dillon, Adv. Mater., 2006, 18, 763.
9. A. Ghicov, S. P. Albu, J. M. Macak and P. Schmuki, Small, 2008, 4, 1063.
10. J. Z. Ou, S. Balendhran, M. R. Field, D. G. McCulloch, A. S. Zoolfakar, R. A. Rani, S. Zhuiykov, A. P. O'Mullane and K. Kalantar-zadeh, Nanoscale, 2012, 4, 5980.
11. J. Zhang, X. L. Wang, X. H. Xia, C. D. Gu, Z. J. Zhao and J. P. Tu, Electrochim. Acta, 2010, 55, 6953.
12. H. Li, G. Shi, H. Wang, Q. Zhang and Y. Li, J. Mater. Chem. A, 2014, 2, 11305.
13. H. Li, J. Wang, G. Shi, H. Wang, Q. Zhang and Y. Li, RSC Adv., 2015, 5, 196.
14. D. Ma, G. Shi, H. Wang, Q. Zhang and Y. Li, J. Mater. Chem. A, 2014, 2, 13541.
15. W. Kang, C. Yan, X. Wang, C. Y. Foo, A. W. M. Tan, K. J. Z. Chee and P. S. Lee, J. Mater. Chem. C, 2014, 2, 4727.
16. Y. Lu, L. Liu, W. Foo, S. Magdassi, D. Mandlerb and P. S. Lee, J. Mater. Chem. C, 2013, 1, 3651.
17. M. Wong, R. Ishige, K. L. White, P. Li, D. Kim, R. Krishnamoorti, R. Gunther, T. Higuchi, H. Jinnai, A. Takahara, R. Nishimura and H. J. Sue, Nat. Commun., 2014, 5, 3589.
18. J. H. Lee, S. Yoshikawa and T. Sagawa, Sol. Energy Mater. Sol. Cells, 2014, 127, 111.
19. Y. Lin, T. Tsai, S. Hung and S. Tien, J. Solid State Electrochem., 2013, 17, 1077.
20. K. A. Gesheva and T. Ivanova, Chem. Vap. Deposition, 2006, 12, 231.
21. L. Kondrachova, B. P. Hahn, G. Vijayaraghavan, R. D. Williams and K. J. Stevenson, Langmuir, 2006, 22, 10490.
22. K. A. Gesheva, A. Cziraki, T. Ivanova and A. Szekeres, Thin Solid Films, 2007, 515, 4609.
23. J. M. O. R. de León, D. R. Acosta, U. Pal and L. Castaneda, Electrochim. Acta, 2011, 56, 2599.
24. L. M. Manceriu, A. Rougier and A. Duta, J. Alloys Compd., 2015, 630, 133.
25. H. Zheng, J. Z. Ou, M. S. Strano, R. B. Kaner, A. Mitchell and K. Kalantar-zadeh, Adv. Funct. Mater., 2011, 21, 2175.
26. H. Cheng, T. Kamegawa, K. Mori and H. Yamashita, Angew. Chem., Int. Ed., 2014, 53, 2910.
27. K. Manthiram and A. P. Alivisatos, J. Am. Chem. Soc., 2012, 134, 3995.
28. G. Song, J. Shen, F. Jiang, R. Hu, W. Li, L. An, R. Zou, Z. Chen, Z. Qin and J. Hu, ACS Appl. Mater. Interfaces, 2014, 6, 3915.
29. B. Dasgupta, Y. Ren, L. M. Wong, L. Kong, E. S. Tok, W. K. Chim and S. Y. Chiam, J. Phys. Chem. C, 2015, 119, 10592.
30. D. Ma, G. Shi, H. Wang, Q. Zhang and Y. Li, J. Mater. Chem. A, 2013, 1, 684.
31. D. Ma, H. Wang, Q. Zhang and Y. Li, J. Mater. Chem., 2012, 22, 16633.
32. S. Vankova, S. Zanarini, J. Amici, F. Cámara, R. Arletti, S. Bodoardo and N. Penazzia, Nanoscale, 2015, 7, 7174.
33. F. Zheng, H. Lu, M. Guo and M. Zhang, CrystEngComm, 2013, 15, 5828.
34. D. D. Yao, R. A. Rani, A. P. O'Mullane, K. Kalantar-zadeh and J. Z. Ou, J. Phys. Chem. C, 2014, 118, 476.
35. B. N. Reddy, P. N. Kumar and M. Deepa, ChemPhysChem, 2015, 16, 377.
36. C. Zhao, F. Du and J. Wang, RSC Adv., 2015, 5, 38706.
37. W. Xiao, W. Liu, X. Mao, H. Zhu and D. Wang, J. Mater. Chem. A, 2013, 1, 1261.
38. J. Wang, E. Khoo, P. S. Lee and J. Ma, J. Phys. Chem. C, 2009, 113, 9655.
39. J. M. Wang, E. Khoo, P. S. Lee and J. Ma, J. Phys. Chem. C, 2008, 112, 14306.
40. N. M. Vuong, D. Kim and H. Kim, J. Mater. Chem. C, 2013, 1, 3399.
41. R. T. Wen, C. G. Granqvist and G. A. Niklasson, Nat. Mater., 2015, 14, 996.
42. J. Wang, C. Yan, W. Kang and P. S. Lee, Nanoscale, 2014, 6, 10734.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5tc02802g

---