The role of oxygen adsorption and gas sensing mechanism for cerium vanadate (CeVO₄) nanorods

Abstract

CeVO₄ nanorods (NRs) were successfully synthesized via a one-step hydrothermal method using disodium edentate (EDTA) as a chelating agent. The CeVO₄ NRs are assigned to the zircon-type tetragonal structure and exhibited pure single-crystals as determined by XRD analysis. FE-SEM images indicate that the as-prepared samples are present as square-section nanorods, and the length and sectional size of the CeVO₄ NRs are found to be ∼1.5 μm and ∼100 nm, respectively. Moreover, the HRTEM images and SAED diffraction patterns confirm that the main exposed surfaces of the CeVO₄ NRs were the (010) and (004) lattice planes with a high exposed percentage (ca. 96.77%) around the NRs and the growth direction was along the (200) lattice plane. The CeVO₄ NRs presents a pure phase, with no other impurity phases identified from the FTIR and Raman spectra. XPS results indicate that the vanadium atoms on the surface exhibit a mixture of valence states, i.e., pentavalent state (V⁵⁺) and trivalent state (V³⁺), as dangling bonds around the oxygen vacancies were induced by EDTA desorption during hydrothermal process. An acetone gas sensor based on the CeVO₄ NRs was fabricated, which exhibits a significant response (0.5 s) and recovery (80 s) with high selectivity at the optimum working temperature (108 °C). This is mainly due to the presence of the trivalent states (V³⁺), which serve as the active sites and provide a large number of oxygen vacancies (V_o) as identified by XPS and infrabar experiments at 300 ppm O₂ (0.003 atm). Moreover, it has been demonstrated that the response to acetone for the gas sensor was crucially dependent on the adsorbed oxygen (O_ads) on the (010) or (004) facets of the CeVO₄ NRs, where the redox reaction with acetone occurred reversibly.

---

Introduction

Many rare earth vanadates have been studied because of their outstanding optical, electrical and magnetic properties, and have been widely used in sensors, tribology and heat-resistant materials.¹ Rare earth orthovanadates (ReVO₄) such as cerium vanadate (CeVO₄) are important semiconducting derivatives of the vanadium oxides due to their useful electronic and catalytic properties for electrochromic materials and gas sensors.² In particular, CeVO₄ has shown excellent redox nature and optical properties because of the 4f electronic structure and diverse electronic transition mode of the rare earth elements.^3,4 As a consequence, the design and synthesis of CeVO₄ nanostructures with well-defined size and morphology have attracted a lot of attention.^5,6 Luo et al. prepared nanorods-assembled CeVO₄ hollow spheres by a simple hydrothermal synthesis method and used them as an active catalyst for the oxidative dehydrogenation of propane.⁷ In the gas sensing field, earlier, hydrothermally synthesized CeVO₄ and CeO₂–CeVO₄ nanopowders were investigated as ethanol sensors, and the gas sensor prepared using CeVO₄ nanopowders performed poorly even at a high temperature (400 °C).⁸ In addition, a type of CeVO₄ nanorods (NRs) with highly selective ethanol sensing properties was synthesized by a one-step hydrothermal method; however, the synthesis process was much more complicated and a V₂O₅ phase was still present.⁹ To date, vanadates have rarely been studied for the detection of acetone molecules. It should be noted that the redox reactions occur mainly at the active sites for gas sensing materials, and a larger specific surface area can provide more active sites. Thus, reasonable control over the hierarchical morphology of CeVO₄ nanomaterials with a high specific surface area will effectively improve the gas-sensing properties. With regards to the gas sensing mechanism, as reported, the gas response was inferred according to the oxygen vacancies and adsorbed oxygen on the sensing layer. However, this behaviour has still not been identified via experiments.¹⁰

Acetone has been widely applied in medical, pesticide and paint industries, and is used as a solvent, reagent, and extractant.¹¹ When emitted, it is harmful to the environment and humans. Slight irritation of the nose, throat, lungs and eyes in the presence of 300–500 ppm acetone has been reported.^12,13 Once exposed to a 2000 ppm acetone environment, humans suffer nausea and vomiting symptoms.¹⁴ Acetone is also a flammable gas with a lower explosive limit (LEL) and upper explosive limit (UEL) of 2.6 and 12.8%, respectively.¹⁵ Thus, analyzing the concentration of acetone in the environment is very important to for the health and industrial safety.¹⁴ Zhou et al. prepared a highly sensitive acetone gas sensor based on porous ZnFe₂O₄ nanospheres by annealing the precursor, which was synthesized via a simple template-free solvothermal route with a binary solvent comprising ethanol and ethylene glycol (EG).¹⁶ Do et al. prepared a conductometric acetone gas sensor using polypyrrole and polyaniline conducting polymers with chemical oxidation-casting (COC), chemical vapour deposition (CVD) and impregnated oxidation (IO) techniques.¹⁵

In the present study, we have successfully synthesized CeVO₄ NRs via a one-step hydrothermal method using disodium edetate (EDTA) as a surfactant to control the shape of the NRs and for maintaining the pH at ∼9.5 using sodium hydroxide solution. Various characterizations were carried out to obtain the crystal structure and morphological information of the as-prepared samples. Furthermore, to demonstrate their potential applications, the resulting CeVO₄ NRs were used to fabricate a gas sensor, which was then tested for response to a variety of gases. The response and selectivity towards acetone for the sensor were investigated at the optimal operating temperature. Before being used to detect a specific gas, the adsorption of oxygen molecules from air on the surface of the CeVO₄ NRs was studied using an infrabar experiment. A surface depletion layer model was also established to explain the gas sensing mechanism according to the results obtained from the XPS analysis and infrabar experiment.

Experimental

Synthesis reagents

The chemical reagents used in this study were cerium acetate [Ce(CH₃CO₂)₃] (Aladdin Co., Ltd.), sodium vanadate (Na₃VO₄) (Aladdin Co., Ltd.), EDTA (Sinopharm Chemical Reagent Co., Ltd.), alcohol (Sinopharm Chemical Reagent Co., Ltd.), acetone (CH₃COCH₃) (Sinopharm Chemical Reagent Co., Ltd.), hydrogen (H₂) (Fuzhou SIA industrial gases Co., Ltd.), and methane (CH₄) (Fuzhou SIA industrial gases Co., Ltd.). All the chemicals were of analytical grade and used as received without further purification.

Synthesis process

The approach for fabricating the CeVO₄ samples was as follows: an appropriate amount of Ce(CH₃CO₂)₃ (0.8 mmol) and EDTA (1 mmol) as template were added to distilled water in a 50 ml flask, forming a chelated cerium, and stirred for several minutes. A stoichiometric amount of Na₃VO₄ solution was added to the complex solution. A transparent yellow solution (64 ml) was obtained. The pH was adjusted to 9.5 using 4 M sodium hydroxide solution and the mixture was transferred into a Teflon-lined stainless-steel autoclave (80 ml), which was maintained at 180 °C for 24 h. The autoclave was allowed to cool to room temperature and the precipitated powders were separated by centrifugation and washed several times with deionized water and ethanol. Finally, the CeVO₄ nanosized products were dried at 60 °C for further characterization.

Characterization

X-ray diffraction analysis (XRD) was performed using a Rigaku-miniflex II (Rigaku Company, Japan) diffractometer operating with monochromatic Cu Kα₁ (λ = 0.154056 nm) radiation at 30 kV and 15 mA. Data were collected over a 2θ range of 10–80° at a scanning speed of 0.02° s⁻¹. The surface morphology of the samples across the entire substrate was characterized by field-emission scanning electron microscopy (FE-SEM, Hitachi S4800, Hitachi Company, Japan). To improve the resolution of the micrographs, the samples were sprayed with Au before conducting the SEM measurements. Transmission electron microscopy (TEM) and high-resolution TEM (HRTEM) images were obtained using a Tecnai G2 F20 S-TWIN, 200 kV (FEI Company, USA). The samples for HRTEM were obtained by dissolving a small amount of the product in ethanol, transferring a few drops of the solution onto the carbon supported on the copper network, and allowing the sample to dry. Raman spectra were recorded using a confocal Raman microspectrometer (Renishaw, UK) under the excitation of a 785 nm diode laser and was collected in the range of 400–1800 cm⁻¹. The FTIR spectra of the CeVO₄ NRs mixed with KBr were recorded on a Spectrum-2000 FTIR spectrophotometer (PerkinElmer Corp., USA) in the range of 400–4000 cm⁻¹. The surface properties of the samples were characterized using X-ray photoelectron spectroscopy (XPS) on a PHI Quantum 2000 Scanning ESCA Microprobe (PHI. Corp., USA) system with a monochromatic Al Kα_1,2 source (1486.60 eV). All binding energies were referenced to the C 1s peak at 284.8 eV corresponding to adventitious carbon found on the surface.

Gas sensing measurements

The fabrication process used to prepare the gas sensor is described as follows: first, CeVO₄ NRs were mixed with an appropriate amount of alcohol to obtain a paste, and then the paste was uniformly coated on the surface of a ceramic tube (with a thickness of about 0.2 mm) under dry conditions in the shade. Finally, a Ni–Cr alloy coil was inserted inside the alumina tube as a heater.¹⁷ After aging for 10 days at a heating voltage of 4 V, the gas sensing performance of the gas sensor was tested using a JF02E gas sensor test system (Jinfeng Tech. Co. Ltd., Kunming, China). The tested gases included H₂, CH₄ and CH₃COCH₃. The sensitivity (S) of the gas sensor towards the testing gases was determined using the relative ratio of the resistance variation, S = (R₀ − R_g)/R₀ (%), where R₀ is the resistance in air and R_g is the resistance in a testing gas. The response time is defined as the time required for the sensor resistance to reach 90% of the equilibrium value after acetone molecules were injected and the recovery time is taken as the time necessary for the gas sensor resistance to reach 90% of the baseline value in air. To identify the performance related with V³⁺, oxygen vacancies and adsorbed oxygen molecules, a low oxygen pressure experiment for the CeVO₄ NRs gas sensor was carried out based on the corresponding resistance in air and infrabar at 300 ppm O₂ (0.003 atm) with a flowing O₂–N₂ mixture.

Results and discussion

Characterization of the samples

As shown in Fig. 1, all the recorded diffraction peaks were well indexed to the pure tetragonal phase of CeVO₄ with lattice constants of a = b = 7.399 Å and c = 6.496 Å, which are in good agreement with those from the standard JCPDS card no. 12-0757.¹⁸ In contrast to the diffraction peaks, the growth rate of the (200) lattice plane at 2θ = 24.032°, is faster than the other lattice planes. No other impurity phases could be observed.

Fig. 1 XRD pattern of CeVO₄ NRs.

The morphologies and microstructures of the as-obtained CeVO₄ samples were investigated using FE-SEM (Fig. 2). Fig. 2a shows the low-magnification FE-SEM image of the products as uniform and dispersed square-section nanorods. The high-magnification FE-SEM images of the CeVO₄ NRs are presented in Fig. 2b and c, which revealed detailed information on the CeVO₄ NRs; the CeVO₄ NRs display a sectional width of ∼100 nm and length of ∼1.5 μm (Fig. 2d), corresponding to the tetragonal structure shown in the XRD analysis (Fig. 1). To the best of our knowledge, there is no related report on the CeVO₄ NRs synthesized.

Fig. 2 FE-SEM images (a–c) and particle size distribution (d) of CeVO₄ NRs.

Transmission electron microscopy (TEM) observations and corresponding selected area electron diffraction (SAED) were conducted to reveal additional information concerning the structural and surface properties of the as-prepared samples. From Fig. 3a and c, the TEM images of the CeVO₄ NRs indicate that the size and shape of the products were in good accordance with the FE-SEM observations throughout their entire lengths and rough surfaces. In fact, the clear lattice fringes in the HRTEM image (Fig. 3b) and SAED pattern (the inset of Fig. 3c) demonstrate that the CeVO₄ NRs are uniform single crystals. The 4.89 Å spacing of the crystallographic planes is assigned to the (101) lattice plane of the CeVO₄ NRs. It also demonstrates the monocrystalline features of CeVO₄ materials based on the (101), (004) and (200) lattice planes in the SAED pattern, corresponding to the HRTEM images. The above mentioned results indicate that the main exposed lateral surfaces of the CeVO₄ NRs are the (010) and (004) lattice planes and the growth direction of the CeVO₄ NRs is mainly along the (200) lattice plane, coinciding with the XRD analysis. Based on the morphology of the CeVO₄ NRs (Fig. 2), the exposed percentage for the (010) and (004) planes is ca. 96.77%. However, the growth rate of the (010) and (004) lattice planes was inhibited because of the chelation of disodium edentate. Therefore, the CeVO₄ NRs have a complete lattice shape and exhibit a nanorod structure with square section after EDTA decomposition during the hydrothermal process and after being washed with deionized water and ethanol.

Fig. 3 TEM images (a) and HRTEM images (b) and (c) of CeVO₄ NRs (the inset is the SAED pattern).

To monitor the presence of other organic phases and the functional groups of the CeVO₄ NRs, the Raman and FTIR spectra were obtained and are shown in Fig. S1.† In Fig. S1a,† two main Raman modes are observed at 801 and 863 cm⁻¹, which are assigned to A_1g vanadate symmetric stretching (ν₁) and Eg antisymmetric stretching of vanadates (ν₃) of VO₄³⁻, respectively. Moreover, in the FTIR spectrum (Fig. S1b†), two distinct peaks at 767 and 437 cm⁻¹ correspond to the stretching vibration of the V–O bond¹⁹ and Ce–O bond,²⁰ respectively. A smooth linear peak is observed from 4000–1000 cm⁻¹, which indicates the absence of both impurity phases in the CeVO₄ NRs and hydroxide radicals.

For investigating the surface nature, XPS was carried out to examine the surface components of the as-prepared samples. Fig. 4a shows the XPS survey spectra, indicating the presence of Ce 4p, V 2p, O 1s, V 2s,²¹ Ce MNN²² and Ce 3d. The C 1s peak appears due to the testing electrodes, confirming the high chemical purity of the CeVO₄ NRs. In the Ce 3d spectrum, peaks for two characteristic states, Ce 3d_5/2 and Ce 3d_3/2, were observed and assigned to the Ce³⁺ valance state (Fig. 4b).^9,18,22,23 The peaks related to the valencies of vanadium in Fig. 4c and Table 1 correspond to the two characteristic states, V 2p_3/2 and V 2p_1/2, and are assigned to the dominant pentavalent state V⁵⁺ and minor trivalent state V³⁺, respectively; the latter corresponds to the dangling bonds of the V⁵⁺ ions on the surface.^9,24,25 Accordingly, we speculated that the V³⁺ on the (010) and (004) lattice planes could serve as active sites, benefiting the redox reaction from V³⁺ to V⁵⁺ once exposed to oxygen in the air. The O 1s components also show their particular binding energies in the spectra (Fig. 4d and Table 1). In particular, the state of O 1s indicates that there are two types of oxygen species on the surface, i.e., the lattice oxygen (O_lattice) and the adsorbed oxygen (O_ads). The former could not be interacted with the reducing gas and is unable to affect the formation of main charge-carrier (electrons) in the n-type semiconductor. However, the latter, O_ads, readily reacts with the gas and is responsive as a medium.²⁶ Moreover, the integral area of the peaks in XPS can be used to quantify the chemical composition because the number of photoelectrons of an element depends on the atomic concentration of these elements in the samples. Thus, by calculating the area of the two components, the quantity ratio of V³⁺/V⁵⁺ (0.1981), O_ads/O_lattice (0.2581), and O_ads/V³⁺ (1.87) were obtained. This reveals that there is a O_ads–V³⁺–O_ads type structure on the surface of the CeVO₄ NRs, and the gas sensing properties may be enhanced because the O_ads and V³⁺ play an important role in the gas sensing process as described below.

Fig. 4 XPS spectra: (a) survey spectra, (b) cerium (Ce 3d), (c) vanadium (V 2p) and (d) oxygen (O 1s) of CeVO₄ NRs.

Table 1 XPS data for Ce 3d, V 2p and O 1s binding energies (eV) of CeVO₄ NRs

Phase	Ce 3d		V 2p		O 1s
	3d5/2	3d3/2	2p3/2	2p1/2
CeVO4	880.86 and 885.1	899.5 and 903.3	516.6	524.35	529.4
			515.4	522.65	531.1

---

Growth mechanism of the CeVO₄ NRs

Many studies have shown that the addition of disodium edetate under alkaline conditions have a significant impact on the growth of nanocrystals in the reaction system. In our study, Ce(CH₃CO₂)₃, Na₃VO₄ and disodium edetate were inferred to the chelating effect of EDTA on Ce³⁺ ions under alkaline conditions.²⁷ These initial CeVO₄ clusters, mediated by the adsorbed ligands on the (010) and (004) crystal planes, could serve as the seeds for the growth of highly anisotropic nanostructures in the solution–solid process (Fig. S2†). As the reaction proceeds, the number of generated CeVO₄ clusters gradually increases and begins to aggregate under the influence of static electricity or the high surface energy, making the CeVO₄ crystal nuclei grow along a certain direction. Herein, using the analysis of the XRD and HRTEM, we know that the growth rate of the (101) and (004) lattice planes was inhibited due to the chelation of EDTA, while the (200) lattice plane grew rapidly. Therefore, it is speculated that the growth direction of the CeVO₄ clusters was mainly along the (200) lattice plane. Finally, due to the tetragonal zirconia structure of CeVO₄, the growth rate of the (200) lattice plane is faster than the other lattice planes, and the square section for CeVO₄ NRs was reasonably obtained when the hydrothermal reaction was complete. A schematic representation of the growth mechanism of CeVO₄ NRs is illustrated in Fig. S2.†

Gas sensing properties and the gas-sensing mechanism

In recent years, environmental pollution and public safety have caused increasing concerns and as a consequence, alternative gas sensors with excellent performance have become a hot topic because they play an important role in the monitoring of poisonous gases. To demonstrate their potential application in gas sensing, a gas sensor (inset of Fig. S3†) was fabricated using the as-prepared CeVO₄ NRs. The gas sensor was placed inside a sensing chamber filled with normal atmospheric air, and the resistance of sensor was measured. Subsequently, the test gases were injected into the sensing chamber through a common syringe. First, the selectivity of the as-fabricated CeVO₄ NRs sensor was evaluated by exposing the sensor to different types of gases with a concentration of 1000 ppm at an operating voltage of 5 V. Fig. S3† presents a bar graph of the response of the sensor to a variety of gases such as acetone, hydrogen and methane. The response towards acetone was remarkably higher than that to the other gases. Therefore, it can be concluded that the as-fabricated gas sensor shows an excellent selectivity towards acetone. Subsequently, it is well-known that the gas response of a gas sensor is highly affected by the various operating voltages of the heating coil inside the ceramic tube; thus, the relationships between the operating voltage and gas response of the gas sensor based on the CeVO₄ NRs to 1000 ppm acetone were tested. Fig. S4† and 5a show the response–recovery curves and sensitivity of the gas sensor at different operating voltages, respectively. It can be seen that the responses to acetone varied with the voltage and both plots exhibited an ‘increase–maximum–decrease’ tendency. At a low operating voltage, acetone molecules cannot effectively react with the surface absorbed oxygen species, which led to a low response. While, with an increase in voltage, both the higher reaction activity and the conversion of surface absorbed oxygen species (O_2gas → O_2ads → O₂²⁻_ads → 2O⁻_ads) contributed to the higher response. As the optimum operating voltage was further increased, the response was reduced because of the low adsorption ability of the acetone molecules, which caused the low response of the sensing materials. Then, the operating voltage of 5 V (Fig. 5a), corresponding to a temperature on the surface of the sensor of 108 °C (monitored by a FLIR IR (Infrared Radiation) camera player, FLIR company, Germany) was chosen as the optimum operating voltage. Fig. 5b presents the response–recovery curves of the sensor to 1000 ppm acetone at 108 °C. According to the definition of response time and recovery time, the response time to 1000 ppm acetone at 108 °C was about 0.5 s, which presents a rapid response. In the same way, the recovery time to 1000 ppm acetone was about 80 s. Thus, the gas sensor illustrates a relatively rapid response and recovery to acetone. Finally, the response behaviour was further investigated with the exposure of CeVO₄ NRs to different concentrations of acetone at 108 °C and the results are shown in Fig. 6. Apparently, when each 100 ppm of acetone was injected into the sensing chamber, the resistance of the sensing material rapidly decreased. As the corresponding linear diagram, the inset of Fig. 6 also shows a good linear relationship between the resistance of the gas sensor and the acetone concentration (100–1000 ppm).

Fig. 5 Sensitivity (a) of CeVO₄ NRs gas sensor to 1000 ppm acetone gas at different voltages and response curve (b) of CeVO₄ NRs gas sensor at 5 V (108 °C).

Fig. 6 Response versus acetone concentration of CeVO₄ NRs (inset is the corresponding sensitivity curve).

Moreover, the gas sensing mechanism of the CeVO₄ NRs gas sensor (Fig. 7 and S5†) is explained according to a surface depletion layer model.^28,29 At first, when exposed to an air atmosphere, oxygen molecules are absorbed into the oxygen vacancies (V_o) on the exposed (010) or (004) lattice planes of the CeVO₄ NRs, forming O₂²⁻_ads; O⁻_ads captures free electrons from the conduction band of the sensing material [eqn (1)] and causes a decrease in the concentration of carriers and increases the resistance of the gas sensor.³⁰ Moreover, the trivalent state (V³⁺) will be oxidized to the pentavalent state (V⁵⁺) due to the loss of electrons [eqn (2)]. Subsequently, when the gas sensor is exposed to acetone gas, O⁻_ads is in favor of adsorbing more acetone molecules, thus facilitating the subsequent redox reaction and reaction with acetone molecules to generate CO₂ and H₂O [eqn (3)]. This induces those O⁻_ads capturing the free electrons to return to the sensing materials, CeVO₄ NRs, and thus the resistance reduces reversibly.

Fig. 7 Response in air and infrabar of CeVO₄ NRs gas sensor at 108 °C.

To further prove the existence of the oxygen vacancies of the CeVO₄ NRs before exposure to O₂ in air, we measured the resistance of the gas sensor in air and infrabar at 300 ppm O₂ (0.003 atm) (Fig. 7). This clearly explains that the corresponding resistance becomes smaller under infrabar (300 ppm O₂) due to the desorption of adsorbed oxygen (O_ads) at the active sites, i.e., oxygen vacancies (V_o), releasing the electrons to the CeVO₄ NRs. Moreover, the resistance of the CeVO₄ NRs decreases. Once in air, O₂ molecules in the air are adsorbed on the oxygen vacancies and V³⁺ dangling bonds present on the surface. This process is reversible as shown in Fig. 7 and eqn (1) and (2).

O_2gas + 2V_o ↔ O₂²⁻_ads + 2V_o˙ ↔ 2O⁻_ads + 2V_o˙ (1)

V³⁺ ↔ V⁵⁺ + 2e⁻ (2)

CH₃COCH₃ (g) + 8O⁻_ads → 3CO₂ (g) + 3H₂O (g) + 8e⁻ (3)

Conclusions

In summary, using a one step hydrothermal method, we have fabricated straight and uniform CeVO₄ NRs with a sectional width of ∼100 nm and length of ∼1.5 μm. During the hydrothermal reaction, the chelation of EDTA on the CeVO₄ NRs and control of the alkaline conditions played a key role on the growth process of the CeVO₄ NRs. FE-SEM and TEM observations demonstrated that the CeVO₄ NRs exhibited a square section, which was mainly due to the tetragonal zircon structure and rapid growth of the (200) lattice plane. Due to the presence of the V³⁺ active sites (dangling bonds) and oxygen vacancies on the exposed (010) or (004) lattice planes, the gas sensor based on the hydrothermally synthesized n-type CeVO₄ NRs exhibited highly efficient acetone sensing behaviour at a low operating voltage (108 °C), which assumes its novelty for use in medical and other industrial applications. Thus, the acetone gas sensor prepared using CeVO₄ NRs can be applied to monitor acetone concentrations between 100 and 1000 ppm with high selectivity. Further efforts will be devoted to the doping effect of CeVO₄ NRs in gas sensing behaviour and their dependence on the processing conditions.

Acknowledgements

The authors gratefully acknowledge the financial support from the National Science Fund of China (NSFC) (21201035, 61201397, J1103303 (J2013-004)), the development project of Fujian Provincial Economic and Information Technology Commission (2015–2017), the Youth Scientific Research Program of Fujian Provincial Health and the Family Planning Commission (2014-1-39), Nursery Scientific Research Foundation of Fujian Medical University (2014MP008), and Fujian Natural Science Foundation (2015J05020).

Notes and references

1. L. Adijanto, V. Balaji Padmanabhan, K. J. Holmes, R. J. Gorte and J. M. Vohs, J. Solid State Chem., 2012, 190, 12.
2. R. Jindal, M. M. Sinha and H. C. Gupta, Spectrochim. Acta, Part A, 2013, 113, 286.
3. G. Lucovsky and J. Phillips, Thin Solid Films, 2005, 486, 200.
4. G. Pagano, F. Aliberti, M. Guida, R. Oral, A. Siciliano, M. Trifuoggi and F. Tommasi, Environ. Res., 2015, 142, 215.
5. G. Li, K. Chao, H. Peng and K. Chen, Mater. Res. Bull., 2008, 112, 6228.
6. M. Oshikiri, J. Ye and M. Boero, J. Phys. Chem. C, 2014, 118, 8331.
7. F. Luo, C. Jia, R. Liu, L. Sun and C. Yan, Mater. Res. Bull., 2013, 48, 1122.
8. L. Chen, Mater. Lett., 2006, 60, 1859.
9. S. A. Hakim, Y. Liu, W. Jin, W. Chen, I. Mehmood and V. T. Nguyen, Mater. Lett., 2015, 154, 144.
10. T. Rakshit, S. Santra, I. Manna and S. K. Ray, RSC Adv., 2014, 15, 947.
11. F. Liu, Y. Guan, R. Sun, X. Liang, P. Sun, F. Liu and G. Lu, Sens. Actuators, B, 2015, 221, 673.
12. X. Xu, Y. Chen, S. Ma, W. Li and Y. Mao, Sens. Actuators, B, 2015, 213, 222.
13. F. Meng, N. Hou, Z. Jin, B. Sun, W. Li, X. Xiao, C. Wang, M. Li and J. Liu, Sens. Actuators, B, 2015, 219, 209.
14. S. Liu, F. Zhang, H. Li, T. Chen and Y. Wang, Sens. Actuators, B, 2012, 162, 259.
15. J. Do and S. Wang, Sens. Actuators, B, 2013, 185, 39.
16. X. Zhou, J. Liu, C. Wang, P. Sun, X. Hu, X. Li, K. Shimanoe, N. Yamazoe and G. Lu, Sens. Actuators, B, 2015, 206, 577.
17. Z. Han, J. Wang, L. Liao, H. Pan, S. Shen and J. Chen, Appl. Surf. Sci., 2013, 273, 349.
18. Y. Shen, Y. Huang, S. Zheng, X. Guo, Z. Chen, L. Peng and W. Ding, Inorg. Chem., 2011, 50, 6189.
19. S. Yuvaraj, R. Kalai Selvan, V. B. Kumar, I. Perelshtein, A. Gedanken, S. Isakkimuthu and S. Arumugam, Ultrason. Sonochem., 2014, 21, 599.
20. L. Tan, J. Xu, X. Zhang, Z. Hang, Y. Jia and S. Wang, Appl. Surf. Sci., 2015, 356, 447.
21. D. Guo, C. Hu and Y. Xi, J. Magnesium Alloys, 2013, 550, 389.
22. L. Armelao and D. Barreca, Surf. Sci. Spectra, 2001, 8, 247.
23. D. Barreca, A. Gasparotto, E. Tondello, C. Sada, S. Polizzi and A. Benedetti, Chem. Vap. Deposition, 2003, 9, 199.
24. D. Barreca, L. Depero, V. Noto, G. Rizzi, L. Sangaletti and E. Tondello, Chem. Mater., 1999, 11, 255.
25. V. Raju, J. Rains, C. Gates, W. Luo, X. Wang, W. F. Stickle, G. D. Stucky and X. Ji, Nano Lett., 2014, 14, 4119.
26. H. Zhang, J. Shu, X. Wei, K. Wang and J. Chen, RSC Adv., 2013, 3, 7403.
27. J. Liu, L. Wang, X. Sun and X. Zhu, Angew. Chem., Int. Ed., 2010, 49, 3492.
28. G. Yamamoto, T. Yamashita, K. Matsuo, T. Hyodo and Y. Shimizu, Sens. Actuators, B, 2013, 183, 253.
29. J. Khanderi, R. C. Hoffmann, A. Gurlo and J. J. Schneider, J. Mater. Chem., 2009, 19, 5039.
30. X. Han, M. Jin, S. Xie, Q. Kuang, Z. Jiang, Y. Jiang, Z. Xie and L. Zheng, Angew. Chem., Int. Ed., 2009, 48, 9180.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra20049k

---