Au@In₂O₃ core–shell composites: a metal–semiconductor heterostructure for gas sensing applications

Abstract

Hybrid Au@In₂O₃ microstructures with a distinctive core–shell configuration have been successfully synthesized by employing Au@carbon spheres as sacrificial templates. The In₂O₃ shell can be easily decorated on the Au core by a facile aging process at room temperature (25 °C) combined with a subsequent calcination. Field emission electron microscopy and transmission electron microscopy images revealed that the Au@In₂O₃ core–shell structures had an average diameter of about 150 nm and the thickness of the porous In₂O₃ shell was ca. 50 nm. When tested as a potential sensing material for gas sensing, the resulting hybrid Au@In₂O₃ core–shell structures exhibited a higher response to formaldehyde compared with the pure In₂O₃ spheres. The enhanced sensing properties of Au@In₂O₃ core–shell structures were attributed to their intense electron depletion that arose from the catalytic activity of Au nanoparticles and the formation of metal–semiconductor junction.

---

1. Introduction

Metal oxide semiconductors, as one of the predominant gas sensing materials, have attracted tremendous attention owing to their industrial and domestic applications in toxic- and explosive-gas detection.^1,2 Recently, great efforts have been centered on the design of the semiconductor oxides with excellent sensing performance and numerous results have demonstrated that the sensing performance of metal oxide semiconductors is not only highly dependent on their microstructures, including morphology, crystalline size, specific area, exposed surface, etc.,^3–7 but also related to their further functionalization with metal elements,^8–11 heterogeneous oxide,^12,13 graphene¹⁴ and so on. Owing to their exceptional catalytic activities, noble metals, such as Pt, Pd, Ag, and Au, are often introduced into metal oxide semiconductors as sensitizers or promoters to improve the sensor performance under certain conditions. To date, a series of obvious improvement of sensing properties have been observed in these noble metals-functionalized semiconductors.^15–18 However, the noble metal, especially Au nanoparticles, will suffer from undesired aggregation arising from its low melting point and increased mobility at high operating temperature, which results in a loss of catalytic activity.^19,20 Thus, to overcome such issues, the exploration of noble metals-functionalized semiconductors with novel configuration is highly required.

In recent years, metal core–semiconductor shell hybrid nanomaterials, as a new type of promising functional materials, have attracted tremendous attention because of their appealing heterostructures, controllable chemical composition and synergistic properties.^21–23 On account of such amazing characteristics, metal–semiconductor hybrid nanomaterials can be used in a variety of areas such as photovoltaic devices,^24,25 nanoreactor,²⁶ chemical sensor^27–29 and photocatalysis.^30,31 In particular, by encapsulating noble metal nanoparticles in a protective shell, the hybrid core–shell structure could effectively increase the stability of catalyst against undesirable aggregation during practical operation.³² Up to now, a variety of methods have been developed to synthesize the metal core–semiconductor shell composites. One common strategy is based on two-step route. The metallic nanoparticles are first prepared by the reduction of metal salt in solution. Subsequently, metal oxide semiconductor is deposited onto the surface of metal core with the assistance of linker reagent or continuous heating procedure. Despite its conceptual simplicity, this synthetic process will bring lots of difficulties in practice. For instance, to obtain well-defined metal core–semiconductor shell nanoparticles, it is needed to choose suitable reducing agent, proper surfactants and stabilizer, appropriate linker reagent as well as reasonable reaction conditions, which undoubtedly increases the complexity and cost of the experiment.³³ Therefore, it still remains a challenge to develop a facile and effective method to fabricate hybrid core–shell nanomaterials.

Indium oxide (In₂O₃), an important n-type semiconductor with a direct band gap of 3.55–3.75 eV, has been recognized as the most potential sensing material due to its high electric conductance.^34,35 As motivated by the driving force of developing the sensor with enhanced sensing performance, great efforts have been exhausted on the design of In₂O₃ with diverse nanostructures.^36–38 For this reason, we demonstrate the synthesis of Au@In₂O₃ core–shell structures for the first time using a facile, mild and green method in this work. During the preparation process, only glucose, HAuCl₄ and indium trichloride were used as raw materials. The superiority of such Au@In₂O₃ hybrid composites on gas sensing was investigated by comparing the sensing performance of Au@In₂O₃ with that of pure In₂O₃ spheres. In addition, the sensing mechanism of Au@In₂O₃ core–shell structures toward formaldehyde has been discussed in details.

2. Experimental section

2.1 Preparation of carbon template

Carbonaceous template spheres were first prepared according to the reported method with some modification.³⁹ In a typical experiment, 6 g of glucose was dissolved in 30 mL of deionized water under magnetic stirring, and then the resulting clear solution was transferred into a Teflon-lined stainless steel autoclave and continuously heated at 180 °C for 6 h. After the autoclave cooled down naturally, the black precipitate was collected and washed by centrifugation at 15 000 rpm several times before being vacuum-dried at 80 °C.

2.2 Preparation of Au@Carbon core–shell structures

The synthesis process of Au@carbon spheres is similar to that of preparing pure carbon spheres. Instead, 0.4 mL of HAuCl₄ solution (0.05 M) was added dropwise into 30 mL of aqueous glucose solution (0.6 M) under vigorous stirring. The above mixture solution was then treated with the same procedure as the synthesis of carbon spheres and finally the Au@carbon spheres were obtained.

2.3 Deposition of In₂O₃ nanoshells

For the synthesis of In₂O₃ nanoshells, 50 mg of as-prepared carbon or Au@carbon spheres were first added to 10 mL of aqueous indium trichloride solution (1.0 M) and well dispersed with the assistance of ultrasonication. The resulting homogeneous suspension was subsequently aged under ambient conditions for 24 hours. After that, the precipitate was separated by centrifugation, washed with deionized water, and then dried in vacuum at 80 °C. To obtain pure In₂O₃ or Au@In₂O₃, the products were finally calcined at 450 °C for 2 h with a heating rate of 4 °C min⁻¹.

2.4 Characterization

The crystalline phases, morphologies, crystal structure and chemical composition of the as-synthesized products were characterized by X-ray powder diffraction (XRD, Rigaku D/max-2550, operated at 40 kV per 200 mA), field emission scanning electron microscopy (FESEM, JEOL JSM-7500F microscope, operated at 15 kV) and transmission electron microscopy (TEM, JEOL JEM-2100F, with an energy dispersive X-ray spectrometer (EDX), operated at an accelerating voltage of 200 kV). Nitrogen adsorption–desorption isotherm was measured on a Gemini VII surface area and porosity system at 77 K. Pore size distribution was automatically calculated based on the desorption branch of the isotherm.

2.5 Fabrication and measurement of gas sensors

The as-obtained pure In₂O₃ and Au@In₂O₃ powder were first dispersed into deionized water to form homogeneous slurry. Subsequently, the resulting slurries were coated onto an alumina tube (4 mm in length, 1.2 mm in external diameter, and 0.8 mm in internal diameter), on which a pair of gold electrodes was pre-installed at each end, and each electrode was connected with two platinum wires. Then, the devices were aged at 400 °C for 2 h to improve the thermal stability of the sensor. After cooling down naturally, a Ni–Cr alloy coil was inserted through the alumina tube as a heater to control the operating temperature of the sensor. Gas sensing measurements were carried out by a static process under laboratory conditions (50 RH%, 25 °C): the sensor was alternately placed into test chambers with air or target gas. The response (S) of the sensor was defined as the ratio of the resistance measured in fresh air (R_a) to that tested in target gases (R_g). While the response and recovery time were defined as the time taken by the sensor to achieve 90% of the total resistance change in the case of adsorption and desorption, respectively.

3. Results and discussion

3.1 Structural and morphological characteristics

As sacrificial templates, carbonaceous spheres play an important role during the synthesis of pure In₂O₃ and Au@In₂O₃ core–shell structures, since the sacrificial templates directly determine the shape and configuration of the resultant microstructures. Fig. 1 presents the typical FESEM and TEM images of the as-synthesized carbonaceous templates. As can be seen, both of the templates (with or without Au core) were composed of numerous well-dispersed spherical particles with uniform size. The FESEM images shown in Fig. 1a and b reveals that the diameter of Au@carbon spheres was about 500 nm. The discrepance of contrast between the center and the fringe of these spherical particles (Fig. 1c) indicated that Au nanoparticles had been successfully encapsulated in the center of carbonaceous spheres. In contrast, homogeneous carbonaceous spheres without Au core were also obtained, their corresponding FESEM and TEM observations were displayed in Fig. 1d–f.

Fig. 1 (a and b) low- and high-magnification FESEM images of Au@carbon core–shell structures; (c) typical TEM image of Au@carbon spheres; (d and e) low- and high-magnification FESEM images of carbon spheres; (f) TEM image of carbon spheres.

Fig. 2 shows the typical X-ray diffraction (XRD) pattern of the Au@In₂O₃ composites prepared with the assistance of Au@carbon templates. As can be seen, the diffraction peaks were composed of two crystal phases, most of the diffraction peaks could be indexed to cubic structure of In₂O₃ (JCPDS file no. 6-416) with lattice parameters a = 10.118 Å, while the residual peaks matched well with those from the standard JCPDS card of Au (no. 4-784). No other impurity peaks appeared in the XRD pattern, which indicates that the final product only consisted of In₂O₃ and Au nanoparticles.

Fig. 2 X-ray diffraction patterns of the obtained Au@In₂O₃ core–shell structures.

The microstructures and morphologies of the resulting Au@In₂O₃ core–shell structures were illuminated by FESEM and TEM observations. As can be seen in Fig. 3a, the as-synthesized product was composed of numerous monodisperse spheres with quite uniform size. The enlarged FESEM image in Fig. 3b reveals that the Au@In₂O₃ composites were constructed with plenty of randomly packed nanoparticles, thus presenting a coarse and cracked surface. The average diameter of these spheres was about 150 nm. Fig. S1† gives the N₂ adsorption–desorption isotherms and corresponding pore size distribution curve. As can be seen, the pore size of the obtained sample was mainly distributed around about 30 nm. The corresponding surface area was measured to be 10.4 m² g⁻¹. To get more detail information about the interior configuration and crystalline structure, the TEM and HRTEM observations of Au@In₂O₃ composites were further characterized by transmission electron microscopy. As can be seen, the strong contrast between the dark center and the relatively light edge reveals the formation of typical Au core@In₂O₃ shell microstructure (Fig. 3c). Moreover, it should be noted that Au@In₂O₃ composites possesses a loose and porous In₂O₃ shell, which will facilitate the diffusion and permeation of test gases. The high-magnification TEM image and corresponding SAED (selected-area electron diffraction) pattern of outer shell are shown in Fig. 3d and its inset. The grain boundaries and diffraction fringes indicate that the In₂O₃ shell was polycrystalline in nature and each single grain has a crystal size about 10 nm. Fig. 3e shows an enlarged HRTEM image of In₂O₃ nanoparticle marked by the white rectangle in Fig. 3d. It is obvious that the In₂O₃ nanoparticle attached on the shell was highly crystallize with a lattice spacing of 0.292 nm, corresponding to the distance between (222) planes in the cubic In₂O₃. In addition, the TEM image and elemental mapping of a single Au@In₂O₃ sphere were displayed in Fig. 3f–i. It can be seen that Au (Fig. 3g) was mainly distributed in the core region, while the O (Fig. 3h) and In (Fig. 3i) distributed homogeneously within the entire structure.

Fig. 3 SEM images of the Au@In₂O₃ core–shell structures: (a) low-magnification and (b) high-magnification; TEM analysis of Au@In₂O₃ core–shell structures: (c) TEM image, (d) HRTEM image and SAED pattern (inset), (e) enlarged HRTEM image taken from the area marked by rectangle; (g–i) elemental mapping of individual Au@In₂O₃ composites shown in (f).

Using the obtained carbon spheres as alternative template, pure In₂O₃ sample was synthesized by repeating the procedure used to prepare the Au@In₂O₃ core–shell structures. Fig. 4a shows the typical XRD pattern of the as-prepared In₂O₃ powder. All of the diffraction peaks were coincident with those from the standard JCPDS card no. 6-416, that is, can be readily indexed as a pure cubic phase of In₂O₃. The SEM image (Fig. 4b) shows that the resulting In₂O₃ sample consists of a number of uniform spheres with a diameter around 130 nm. The enlarged FESEM image shown in Fig. 4c reveals that these In₂O₃ spheres were constructed with lots of nanoparticles with a size of about 15 nm. Further structural analysis of these spheres was recorded on TEM observations (Fig. 4d), which confirmed the formation of porous In₂O₃ spheres without Au. Moreover, for the pure In₂O₃ sample, the BET (Brunauer–Emmett–Teller) surface area was measured to be 9.8 m² g⁻¹ and the average pore size was mainly distributed at 30 nm (Fig. S2†), which is quite similar to the results of Au@In₂O₃ core–shell structures.

Fig. 4 (a) XRD pattern of as-prepared In₂O₃ spheres; (b and c) SEM images of In₂O₃ spheres with different magnifications; (d) TEM image of In₂O₃ spheres.

On the basis of our experimental design, the formation mechanism of In₂O₃ shell and Au@In₂O₃ composites was schematically illustrated in Fig. 5. It is well known that the surfaces of the carbonaceous templates are hydrophilic and decorated with –OH and –C=O groups.⁴⁰ When these carbonaceous spheres were dispersed in aqueous indium trichloride solutions, a large quantity of cations (In³⁺) would bind with the functional groups on the surface layer. Accompany with the following calcination process in air, the carbonaceous templates shrinked gradually while the surface layers containing In³⁺ were condensed to accumulate sufficient mechanical strength to keep the well-defined configuration of spherical nanoparticles. After the carbonaceous templates were removed completely, highly crystalline In₂O₃ spheres with closely packed nanoparticles were finally obtained.

Fig. 5 Schematic formation process of the Au@In₂O₃ core–shell structures.

3.2 Gas sensing properties

To demonstrate the potential applications of these novel In₂O₃ microstructures, two gas sensors based on the as-prepared Au@In₂O₃ and pure In₂O₃ spheres were fabricated and their gas sensing performances were investigated. Fig. 6a shows the response of the gas sensors to 100 ppm formaldehyde (HCHO) as a function of operating temperatures. It can be observed that the response of tested sensors varied with operating temperature and exhibited a temperature-dependent feature. For pure In₂O₃ spheres, the response first increased with the operating temperature, up to 225 °C, and then gradually decreased on further increasing the operating temperature. The maximum response to 100 ppm HCHO was 5.3. As can be seen, the sensor based on Au@In₂O₃ composites shows an enhanced response at each temperature in comparison to the pure In₂O₃. The maximum response of Au@In₂O₃ composites could reach to 17.0 at the optimal operating temperature of 200 °C, which is about three times higher than that of In₂O₃ spheres.

Fig. 6 (a) Responses of the sensors versus operating temperature to 100 ppm HCHO; (b) dynamic curves of the sensors when orderly exposed to low-concentration HCHO at their optimal operation temperatures.

Fig. 6b shows the dynamic response of the gas sensors when orderly exposed to low-concentration formaldehyde ranging from 5 to 50 ppm. It can be seen that the corresponding responses of the sensors were highly dependent on the concentration of formaldehyde. For Au@In₂O₃ composites, the gas responses to 5, 10, 20, 30 and 50 ppm HCHO were about 2.1, 2.6, 3.9, 5.7 and 7.8, respectively. Further analysis about the sensing properties of the sensors to high concentration gas was conducted and the results were presented in Fig. S3(a).† Fig. S3(b)† gives the response of the sensors with respect to the varying formaldehyde concentration. It can be seen that the response of the sensors showed an approximately linear increase at low concentration, and then the response increased slowly with the increase of gas concentration, which indicated that the sensor tended to saturation gradually.

The selectivities of the sensors towards HCHO were further investigated by comparing the response to some common interfering gases (especially ethanol), as shown in Fig. 7a and S4.† It is obvious that both of the sensors exhibited higher response to formaldehyde in comparison to other target gases, which indicated that the sensors based on pure In₂O₃ and Au@In₂O₃ composites possess good selectivities to formaldehyde when working at optimal operation temperatures. Besides, by contrast, the sensor using the Au@In₂O₃ core–shell microstructures showed better sensing performance for each gas than its pure In₂O₃ counterpart, which in turn directly verifies the promotion effect of Au nanoparticles.

Fig. 7 (a) Response of the sensors based on Au@In₂O₃ and In₂O₃ to 100 ppm various gases (b) response transient of Au@In₂O₃ composites to 100 ppm formaldehyde at 200 °C.

Another important criterion of gas sensors is the response and recovery properties. Fig. 7b displays the dynamic response–recovery curves of the Au@In₂O₃ composite to 100 ppm HCHO at 200 °C. It can be found that the resistance changed immediately when the sensor was exposed to HCHO and then reached a steady state slowly. The time taken in this process was only 7 s. Soon afterwards, the sensor was transferred into air to recover and the time consumed by recovery was about 135 s.

Moreover, it can be observed from the inset of Fig. 7b that the sensor maintained its initial response amplitude without a clear attenuation upon alternately exposed to air and HCHO gas, which indicates that the sensor exhibited an excellent stability and reproducibility.

3.3 Gas sensing mechanism

The most widely accepted theory for the gas sensing mechanism of metal oxides is based on the change in resistance, which proposed that the change in resistance of the sensor is caused by the chemical adsorption and the reaction of test gas molecules on the surface of the sensing materials.^41–43 In the air, oxygen molecules would adsorb on the surface of the In₂O₃ spheres, and then formed the chemisorbed oxygen species (O²⁻, O⁻ or O₂⁻) by capturing free electrons from the conduction band of In₂O₃. Therefore, the electron concentration of sensing materials decreased, electron depletion layers would emerge on the surface of sensing materials, which is shown schematically in Fig. 8a-1 and b-1. When the sensors were exposed to HCHO atmosphere, the gas molecules would react with adsorbed oxygen species on the surface of In₂O₃(HCHO + 2O⁻ ↔ CO₂ + H₂O + 2e⁻). As a result, the trapped electrons were released back to the conduction band of In₂O₃, which eventually led to a remarkable decrease of resistance.

Fig. 8 (a) Schematic presentation of electron transportation and gas sensing principle; (b) schematic energy band diagram of the sensing materials (In₂O₃ and Au@In₂O₃) upon exposure to air.

Since the test conditions are identical for In₂O₃ spheres and Au@In₂O₃ composites, the enhanced sensing performance of Au@In₂O₃ hybrid microstructures should be directly related to the catalytic activity of Au nanoparticles. It is well known that Au nanoparticle can serve as an effective adsorption sites to bind and activate oxygen molecules due to the catalytic function.^44,45 Thus, more absorbed oxygen species will diffuse to the surface of the sensing semiconductor, resulting in a larger degree of electron extractions from the conduction band of In₂O₃. The high coverage of chemisorbed oxygen species makes the Au@In₂O₃ composites more sensitive to target gas (HCHO), which directly results in a high sensitivity. More importantly, owing to the strong electronic interaction between Au core and In₂O₃ shell, a typical metal–semiconductor junction was formed near the interfaces, as shown in Fig. 8a-2 and b-2. Because the work function of Au (5.1 eV) is larger than that of In₂O₃ (4.8 eV), the electron will transfer from conduction band of In₂O₃ to Au core. Therefore, the energy band of In₂O₃ exhibits greater bending at the Au@In₂O₃ interface compared with pure In₂O₃ spheres, which leads to a broadening of electron depletion layer and an increase of resistance. During the following detection of formaldehyde gas, more trapped electrons are transferred backwards to the conduction band of In₂O₃, and the significant variation in electron concentration eventually causes an obvious change of resistance. Therefore, the sensor based on hybrid Au@In₂O₃ core–shell structures exhibits higher gas response compared to the pure In₂O₃ spheres.

4. Conclusions

In summary, we have demonstrated that Au@In₂O₃ core–shell structures can be directly prepared using the combination of hydrothermal synthesis of carbonaceous template, aging process at room temperature and annealing treatment of resulting precursor. This facile preparation route is believed to be applicable in synthesis of well-defined metal–semiconductor core–shell microstructures with varying components. The as-prepared Au@In₂O₃ core–shell microstructure was about 150 nm in diameter with ∼50 nm porous In₂O₃ shell. On the basis of the comparative gas sensing test to HCHO between such unique Au@In₂O₃ composites and pure In₂O₃ spheres, it was found that the Au@In₂O₃ core–shell microstructures can be a potential candidate for gas sensing due to its enhanced sensing performance. Moreover, the gas sensing mechanism was discussed and the improved sensing properties stated here were mainly attributed to the intense electron depletion at the surface of Au@In₂O₃ core–shell structures.

Acknowledgements

Thanks for the financial support from the National Nature Science Foundation of China (nos 61374218, 61134010 and 61327804), Program for Chang Jiang Scholars and Innovative Research Team in University (no. IRT13018) and National High-Tech Research and Development Program of China (863 Program, no. 2013AA030902 and 2014AA06A505).

Notes and references

1. Y. F. Sun, S. B. Liu, F. L. Meng, J. Y. Liu, Z. Jin, L. T. Kong and J. H. Liu, Sensors, 2012, 12, 2610–2631.
2. E. Comini, Anal. Chim. Acta, 2006, 568, 28–40.
3. X. Wang, M. Zhang, J. Liu, T. Luo and Y. Qian, Sens. Actuators, B, 2009, 137, 103–110.
4. C. Xu, J. Tamaki, N. Miura and N. Yamazoe, Sens. Actuators, B, 1991, 3, 147–155.
5. W. J. Tseng, T.-T. Tseng, H.-M. Wu, Y.-C. Her, T.-J. Yang and P. Gouma, J. Am. Ceram. Soc., 2013, 96, 719–725.
6. X. Li, X. Zhou, Y. Liu, P. Sun, K. Shimanoe, N. Yamazoe and G. Lu, RSC Adv., 2014, 4, 32538–32543.
7. X. W. Li, P. Sun, T. L. Yang, J. Zhao, Z. Y. Wang, W. N. Wang, Y. P. Liu, G. Y. Lu and Y. Du, CrystEngComm, 2013, 15, 2949–2955.
8. P. Sun, C. Wang, X. Zhou, P. Cheng, K. Shimanoe, G. Lu and N. Yamazoe, Sens. Actuators, B, 2014, 193, 616–622.
9. V. D. Kapse, S. A. Ghosh, F. C. Raghuwanshi and S. D. Kapse, Sens. Actuators, B, 2009, 137, 681–686.
10. P. Rai, R. Khan, S. Raj, S. M. Majhi, K.-K. Park, Y.-T. Yu, I.-H. Lee and P. K. Sekhar, Nanoscale, 2014, 6, 581–588.
11. C. Dong, X. Liu, X. Xiao, G. Chen, Y. Wang and I. Djerdj, J. Mater. Chem. A, 2014, 2, 20089–20095.
12. W. Zeng, T. Liu and Z. Wang, J. Mater. Chem., 2012, 22, 3544–3548.
13. I.-S. Hwang, J.-K. Choi, S.-J. Kim, K.-Y. Dong, J.-H. Kwon, B.-K. Ju and J.-H. Lee, Sens. Actuators, B, 2009, 142, 105–110.
14. Z. Wang, Y. Xiao, X. Cui, P. Cheng, B. Wang, Y. Gao, X. Li, T. Yang, T. Zhang and G. Lu, ACS Appl. Mater. Interfaces, 2014, 6, 3888–3895.
15. X.-j. Wang, W. Wang and Y.-L. Liu, Sens. Actuators, B, 2012, 168, 39–45.
16. Y. Wang, Y. Wang, J. Cao, F. Kong, H. Xia, J. Zhang, B. Zhu, S. Wang and S. Wu, Sens. Actuators, B, 2008, 131, 183–189.
17. A. Kolmakov, D. Klenov, Y. Lilach, S. Stemmer and M. Moskovits, Nano Lett., 2005, 5, 667–673.
18. Y. Zhang, J. Xu, P. Xu, Y. Zhu, X. Chen and W. Yu, Nanotechnology, 2010, 21, 285501.
19. P. M. Arnal, M. Comotti and F. Schüth, Angew. Chem., 2006, 118, 8404–8407.
20. C. Galeano, R. Guttel, M. Paul, P. Arnal, A. H. Lu and F. Schuth, Chem.–Eur. J., 2011, 17, 8434–8439.
21. N. Zhang, S. Liu and Y. J. Xu, Nanoscale, 2012, 4, 2227–2238.
22. R. Ghosh Chaudhuri and S. Paria, Chem. Rev., 2012, 112, 2373–2433.
23. L. Zhang, D. A. Blom and H. Wang, Chem. Mater., 2011, 23, 4587–4598.
24. L. Zhou, X. Yu and J. Zhu, Nano Lett., 2014, 14, 1093–1098.
25. T. Hirakawa and P. V. Kamat, Langmuir, 2004, 20, 5645–5647.
26. C. H. Lin, X. Y. Liu, S. H. Wu, K. H. Liu and C. Y. Mou, J. Phys. Chem. Lett., 2011, 2, 2984–2988.
27. Y.-T. Yu and P. Dutta, Sens. Actuators, B, 2011, 157, 444–449.
28. S. M. Majhi, P. Rai, S. Raj, B. S. Chon, K. K. Park and Y. T. Yu, ACS Appl. Mater. Interfaces, 2014, 6, 7491–7497.
29. X. Li, X. Zhou, H. Guo, C. Wang, J. Liu, P. Sun, F. Liu and G. Lu, ACS Appl. Mater. Interfaces, 2014, 6, 18661–18667.
30. N. Zhang, X. Fu and Y.-J. Xu, J. Mater. Chem., 2011, 21, 8152–8158.
31. N. Zhang, S. Liu, X. Fu and Y.-J. Xu, J. Phys. Chem. C, 2011, 115, 9136–9145.
32. I. Lee, J. B. Joo, Y. Yin and F. Zaera, Angew. Chem., 2011, 123, 10390–10393.
33. R. Ghosh Chaudhuri and S. Paria, Chem. Rev., 2012, 112, 2373–2433.
34. S. Noguchi and H. Sakata, J. Phys. D: Appl. Phys., 1980, 13, 1129–1133.
35. Y. Shigesato, S. Takaki and T. Haranoh, J. Appl. Phys., 1992, 71, 3356–3364.
36. S. J. Kim, I. S. Hwang, J. K. Choi, Y. C. Kang and J. H. Lee, Sens. Actuators, B, 2011, 155, 512–518.
37. J. Liu, T. Luo, F. Meng, K. Qian, Y. Wan and J. Liu, J. Phys. Chem. C, 2010, 114, 4887–4894.
38. H. Y. Lai, T. H. Chen and C. H. Chen, CrystEngComm, 2012, 14, 5589–5595.
39. Y. Mi, W. Hu, Y. Dan and Y. Liu, Mater. Lett., 2008, 62, 1194–1196.
40. X. Sun and Y. Li, Angew. Chem., Int. Ed. Engl., 2004, 43, 597–601.
41. N. Yamazoe, G. Sakai and K. Shimanoe, Catal. Surv. Asia, 2003, 7, 63–75.
42. N. Yamazoe and K. Shimanoe, Sens. Actuators, B, 2009, 138, 100–107.
43. M. Batzill and U. Diebold, Phys. Chem. Chem. Phys., 2007, 9, 2307–2318.
44. X. Li, W. Feng, Y. Xiao, P. Sun, X. Hu, K. Shimanoe, G. Lu and N. Yamazoe, RSC Adv., 2014, 4, 28005–28010.
45. B. K. Min and C. M. Friend, Chem. Rev., 2007, 107, 2709–2724.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra12467g

---