Fast and real-time acetone gas sensor using hybrid ZnFe₂O₄/ZnO hollow spheres

Abstract

Special hollow ZnFe₂O₄/ZnO hybrid spheres were successfully designed and synthesized via a facile hydrothermal procedure. Carbon spheres were used as the core template, which were removed during the calcination process in an air-rich atmosphere, thus leading to interior voids and an outer shell structure due to the contraction and combustion of the carbon spheres. The hybrid ZnFe₂O₄/ZnO hollow spheres not only possess a hollow structure, but also have a porous shell with a thickness of approximately 30 nm. Inspired by this special structure and design composition, a gas sensor based on the hybrid ZnFe₂O₄/ZnO hollow spheres showed a high response and fast response/recovery speed compared with nanoparticles. A short response time usually enables a gas sensor to accurately detect and real-time monitor a certain target gas. This study may give insight for the rational fabrication of high performance sensing materials.

---

Introduction

Currently, real-time sensing materials are in high demand for the detection of volatile organic compounds (VOCs) and other toxic gases in view of human health and environmental protection.^1,2 To fabricate real-time detection gas sensors, research on sensing materials is significant on account of their key role in gas sensors. In particular, considerable efforts have been concentrated on improving sensor response and recovery rate.^3–5 Some effective strategies have been reported, including loading various dopants or tuning morphologies.^6,7 On the one hand, the incorporation of secondary phases could include noble metals, carbon nanotubes, hydroxyapatite and metal oxides.⁸ Materials with hybrid components are widely used in the sensing, electrical, biological fields because the synergic and hybridization effect of pure organic or inorganic materials could improve gas sensing properties. For example, NiO/SnO₂ composite nanofibers exhibited ultrafast response and recovery properties (about 3 s) towards H₂.⁹ Our group reported that ring-like PdO–NiO showed a rapid response/recovery speed (2–3 s) towards CO.¹⁰ On the other hand, special structures, such as core–shell spheres, hollow spheres, hollow cubes, and hollow fibers, could offer thin and semi-permeable shells for the rapid and effective diffusion of target gases to the entire sensing surface so as to enhance gas sensitivity and response rate, simultaneously.¹¹ Wang et al. observed that hollow α-Fe₂O₃ spheres showed a fast ethanol response (2 s) in comparison with the solid spheres (6 s).¹² Bing and co-workers reported that the response times of a sensor based on double-shelled/single-shelled SnO₂ cages were about 2.3 s/2.0 s when exposed to toluene, which were fast enough to real-time detect target gases.¹³ In addition, hollow structures with a porous surface are also advantageous for a high gas molecule capacity. Thus, hollow structured materials with interior cavities and surface porosity are believed to favour gas sensing.^14,15

MOSs (metal oxide semiconductors) are widely used in gas sensors to detect various gas vapours such as ethanol, carbon monoxide, ammonia, and trimethylamine (TMA).^16–19 ZnO has attracted considerable attention of researchers because of its fascinating features, such as low cost, large electron mobility and nontoxicity, and has been used in many applications such as optoelectronic devices, lithium-ion batteries and sensors.^20–23 Zinc ferrites, as a typical spinel ferrite of transition metals (e.g. Co, Cd, Mg, Ni, Zn, and Mn) of spinel compounds (with the general formula AB₂O₄), are a promising candidate for gas sensors because of their low resistance, low preparation cost, high catalytic efficiency and fascinating electrochemical characters, including active interaction of oxygen–metal ions.^24–26

In this study, the synthesis of hybrid ZnFe₂O₄/ZnO hollow spheres is presented. Integration with two hollow structured materials not only favours gas diffusion, but also enhances the reaction speed on the inner and outer surface of the material. Compared with the nanoparticles-based sensor, the gas sensor based on the hybrid hollow spheres exhibits a fast response/recovery speed (7.1 s/10.1 s) towards acetone.

Experiment section

Chemical reagents

D-Glucose (C₆H₁₂O₆·6H₂O), zinc nitrate hexahydrate (Zn(NO)₃·6H₂O), iron(III) nitrate nonahydrate (Fe(NO)₃·9H₂O), N,N-dimethylformamide (DMF), ammonia solution (NH₃·H₂O), absolute ethyl alcohol (C₂H₅OH) and deionized water were the starting chemical reagents, which were analytical grade and used as received without further purification.

Synthetic process

Synthesis of hybrid ZnFe₂O₄/ZnO hollow spheres. A two-step synthetic method from the reported literature was used to obtain hybrid ZnFe₂O₄/ZnO hollow spheres.^27,28 The detailed procedure is described as follows: first, carbon spheres were obtained via a simple hydrothermal step. In brief, 0.5 M homogeneous solution of glucose was placed into a 50 mL Teflon-lined stainless steel autoclave, which then was sealed and heated at 180 °C for 5 h. After the Teflon-lined stainless steel autoclave cooled down to room temperature, the collected product was washed and then dried at 60 °C overnight. Second, hybrid ZnFe₂O₄/ZnO hollow spheres were obtained using the carbon spheres, which were synthesized in the first hydrothermal step, as templates. As the source of Zn and Fe, zinc nitrate and iron nitrate were dispersed in 200 mL DMF with a specific ratio under continuous stirring using a magnetic stirring bar. Then, 100 mg as-obtained carbon spheres and ammonia were added to the mixed solution, which was ultrasonicated for 30 min. Then, the mixture was stirred continuously for 12 h without heating. After the reaction, the collected product was washed by centrifugation. Finally, to remove the carbon spheres and obtain a hollow structure, the as-prepared samples were annealed at 450 °C for 6 h in air (5 °C min⁻¹). For comparison, different molar ratios of zinc nitrate and iron nitrate were mixed in the second step. The detailed content, including the specific molar ratio, the related XRD and gas sensing properties are listed in Table S1 and Fig. S(3–6).†

Synthesis of hybrid ZnFe₂O₄/ZnO nanoparticles. The synthesis of hybrid ZnFe₂O₄/ZnO nanoparticles was similar to the hollow spheres except for the absence of carbon spheres as templates, i.e., nanoparticles were obtained without the first hydrothermal step.

Characterization techniques

XRD (X-ray diffraction) patterns give information on the phase and crystallinity of the as-prepared materials, which were collected on a Rigaku D/Max-2550 diffractometer equipped with Cu-Kα radiation (λ = 0.15418 nm) at a scanning range of 20–80° and scanning speed of 5° min⁻¹. FESEM (field emission scanning electron microscope) images and EDS (energy dispersive spectroscopy) spectra were obtained using a JEOL 7500F microscope. TEM, together with SAED were carried on a Tecnai G220S-Twin transmission electron microscope at an accelerating voltage of 120 kV, and HRTEM images were observed at 200 kV instead of 120 kV. A JW-BK132F analyzer was used to obtain the nitrogen isotherm, which was the basis for the calculation of specific surface area.

Measurement of the gas sensing performance of the ZnFe₂O₄/ZnO-based sensor

The measurement process is as follows: after the as-fabricated sensor was connected to an electrical circuit, it was used to detect various gases on a Chemical Gas Sensor-8 intelligent gas sensing analysis system (Alite Tech, Beijing, China). A schematic of the structure of the ceramic tube and gas sensor is shown in Fig. S1.† Similar to typical metal oxide semiconductors, varying values of resistance appeared when the sensor was exposed to different concentrations of tested gas due to the process of adsorption/desorption and reaction with gas molecules. The resistance values towards C₃H₆O, NH₃, C₇H₈ and C₃H₉N were measured at different temperatures. The resistance increased when the sensor was exposed to these four reducing gases and decreased upon exposure to air, which indicated that the sensor based on hybrid ZnFe₂O₄/ZnO hollow spheres shows typical p-type sensing behavior. Thus, R_a and R_g were defined as the values of resistance of the sensors when exposed to air and tested gas, respectively. Gas sensing response was defined as the value of the ratio of R_g to R_a (S = R_g/R_a), and response/recovery time was the time variation from R_a to R_a + 90% × (R_g − R_a) or R_g to R_g − 90% × (R_g − R_a), which was abbreviated T_r1 and T_r2, respectively.

Results and discussion

Structural and morphological characteristics

In the following, the hybrid ZnFe₂O₄/ZnO hollow spheres (S-3) are characterized in detail as a representative sample. Scheme 1 sheds light on the procedure for the synthesis of the hybrid ZnFe₂O₄/ZnO hollow spheres, which was confirmed in Fig. S2 and (1–3). For the hybrid ZnFe₂O₄/ZnO hollow spheres samples, the proposed procedure involves: (I) carbon spheres synthesis; (II) Zn²⁺/Fe³⁺ adsorption; and (III) removal of the templates. First, the polymerization/carbonization step: carbonization and dehydration reactions were induced by intermolecular dehydration of oligosaccharides, which were formed from polymerization. The final nuclei grew isotropically by diffusion of the solutes up to the final size with the relatively uniform diameter of approximately 1.8 μm (Fig. S2†) along with functionalization by carboxyl (–COOH) and hydroxyl (–OH) groups.²⁷ Second, the adsorption step: after continuous stirring, Fe³⁺ and Zn²⁺ were immobilized on the surface of the carbon spheres homogeneously by electrostatic adsorption to form structures on the atomic scale and then hydrolyzed to Zn(OH)₂ and Fe(OH)₃. As a result, the C@Zn/Fe-precursor core–shell structure formed. Third, the calcination step: the carbon spheres shrunk and then progressively burned off as CO₂ by oxidation in air,²⁹ thus reducing the size of the final obtained hollow spheres (Fig. 1(d)). Release of CO₂ led to a porous surface and even open structure of the shell, which was in accordance with the FESEM and TEM images (Fig. 1 and 2). Moreover, ZnO and ZnFe₂O₄ nanocrystals were produced from hydroxide during the calcination. In addition, without the carbon spheres as templates, the ZnO and ZnFe₂O₄ nanoparticles aggregated resulting in solid hybrid nanoparticles, which is confirmed in Fig. 3.

Scheme 1 Schematic illustration of the synthetic process for hybrid ZnFe₂O₄/ZnO hollow spheres. (I) Hydrothermal fabrication of carbon spheres; (II) fabrication of C@Zn/Fe-precursor by stirring at room temperature; and (III) formation of hybrid ZnFe₂O₄/ZnO hollow spheres by the removal of the carbon spheres via calcination in air.

Fig. 1 (a–e) Low and high magnification FESEM and (f and g) TEM images of the hybrid ZnFe₂O₄/ZnO hollow spheres.

Fig. 2 (a and b) TEM images, (c) magnified TEM image and (d) HRTEM image of the hybrid ZnFe₂O₄/ZnO hollow spheres. The inset of (d) is the SAED pattern. (e–h) STEM image and corresponding elemental mapping showing the dispersion of Zn, Fe and O elements in the hybrid ZnFe₂O₄/ZnO hollow spheres.

Fig. 3 (a and b) FESEM images, (c) TEM image and (d) HRTEM image of the hybrid ZnFe₂O₄/ZnO nanoparticles.

A large amount of well-shaped, monodisperse and relatively uniform ZnFe₂O₄/ZnO spheres, which were conducive to gas diffusion, were obtained and confirmed by the typical FESEM images shown in Fig. 1(a and b). Moreover, we observed a hollow structure from broken ZnFe₂O₄/ZnO spheres (Fig. 1(c–e)), which was further confirmed through the corresponding low/high magnification TEM images (Fig. 1(f and g)). Obviously, the as-obtained spheres were hollow with no core in their shell (Fig. 1(c)), except for some debris appearing in the hollow spheres (Fig. 1(d)). The as-synthesized ZnFe₂O₄/ZnO composite spheres had a diameter of approximately 500 nm (Fig. 1(c)) and the shell layer had an average thickness of 30 nm (Fig. 1(e)).

Fig. 2(a and b) show representative TEM images of the hybrid ZnFe₂O₄/ZnO hollow spheres and the hollow structure was confirmed by the dark contour and bright interior. The edge and the center present strong contrast, which further proves the formation of well-defined hollow ZnFe₂O₄/ZnO spheres. Moreover, the littered bright areas identify the pores on the surface of hollow spheres. A local magnified image (Fig. 2(c)) was selected from part of the yellow imaginary box in Fig. 2(b) and the relevant HRTEM image is shown in Fig. 2(d). The recorded spacing values were 0.487 nm and 0.260 nm, which correspond to the (111) lattice plane of ZnFe₂O₄ (JCPDS: 22-1012) and (002) lattice plane of ZnO (JCPDS: 36-1451). The inset in Fig. 2(d) is the corresponding SAED pattern and it confirms that the hollow ZnFe₂O₄/ZnO spheres are polycrystalline structures. The hybrid ZnFe₂O₄/ZnO hollow spheres are composed of Zn, Fe and O according to the EDS elemental mapping (Fig. 2(e–h)) and spot scanning (Fig. S7†), which highlight the distribution of elements and confirm the hollow structure. Moreover, due to the supporting FESEM pedestal used during the measurement, silicon signals appeared in the spot scanning pattern.

Fig. 3(a and b) shows the surface morphology of the hybrid ZnFe₂O₄/ZnO nanoparticles. The hybrid ZnFe₂O₄/ZnO sample shows rough surface spherical-shape nanoparticles with a uniform size (diameter of approximately 100 nm), which comprise small nanoparticles. Detail information about the interior structure of the hybrid ZnFe₂O₄/ZnO samples was further determined via TEM.

Fig. 3(c) shows the hybrid ZnFe₂O₄/ZnO nanoparticles with a solid interior structure and their size is in agreement with the FESEM image (Fig. 3(a)). In addition, the obvious lattice spacing found in the HRTEM image further confirms that the ZnFe₂O₄/ZnO sample comprises two phase nanoparticles with the lattice spacing of 0.487 nm and 0.260 nm, which match well with the (111) lattice plane of ZnFe₂O₄ and (002) lattice plane of ZnO.

Fig. 4(a) exhibits the X-ray diffraction patterns of the as-synthesized ZnFe₂O₄/ZnO samples. The diffraction peaks of the two as-prepared materials could be well indexed with hexagonal ZnO (JCPDS: 36-1451) and cubic ZnFe₂O₄ (JCPDS: 22-1012). In addition, no drifting peaks and no peaks from other impurities were observed in the patterns. This can be explained by the fact that the materials are a mixture of ZnO and ZnFe₂O₄. In addition, the XRD patterns of S-1, S-2, S-4, S-5 are shown in Fig. S3.† The intensity of the diffraction peaks of ZnFe₂O₄ decreased, whereas the intensity of the diffraction peaks of ZnO increased with increase in the ZnO content. Nitrogen adsorption-desorption measurement was employed to further confirm the porous structure of the as-synthesized hybrid ZnFe₂O₄/ZnO samples (Fig. 4(b)). Due to their large internal space, the BET surface area of the hybrid ZnFe₂O₄/ZnO hollow spheres was calculated to be 30.8 m² g⁻¹. By contrast, the BET surface area of the hybrid ZnFe₂O₄/ZnO nanoparticles was 9.1 m² g⁻¹. The related pore-size distribution of the hybrid ZnFe₂O₄/ZnO hollow spheres was calculated from the N₂ adsorption–desorption isotherm (inset of Fig. 4(b)), which indicates that most of the pore sizes were 9.12 nm. The large exposed surface of the hybrid ZnFe₂O₄/ZnO hollow spheres provides sufficient space to facilitate adsorption–desorption and the interfacial charge transfer process, which is expected to provide excellent gas sensing properties.

Fig. 4 (a) X-ray diffraction patterns (pink line: standard ZnFe₂O₄ (JCPDS: 22-1012) and blue line: standard ZnO (JCPDS: 36-1451)) and (b) N₂ adsorption–desorption isotherm of the as-synthesized hybrid ZnFe₂O₄/ZnO hollow spheres and nanoparticles. Inset shows the BJH pore-size distribution plot of the as-synthesized hybrid ZnFe₂O₄/ZnO hollow spheres.

Gas sensing properties

For contrast, the gas sensing properties of all the samples are given in Fig. S(4–6).† S-3 displays a lower operating temperature than pure ZnO and higher response than pure ZnFe₂O₄. Moreover, compared with S-2 and S-4, S-3 has a higher response to acetone. Thus, S-3 was chosen as the optimal sample. Furthermore, there is a conclusion that the introduction of ZnO could enhance the acetone response of the sensor; however, this increases the working temperature. As displayed in the response plots in Fig. 5(a), the two sensors show the maximum response to acetone at 280 °C. The maximum response value of the hollow spheres-based sensor at 280 °C to 50 ppm acetone was 5.2, which is 1.7 times higher than that of the nanoparticles (3.1). Importantly, the hollow spheres-based sensor not only showed a high response, but also possessed a fast response speed at a working temperature of 280 °C (Fig. 5(b)). To study the selectivity of the sensor towards different gases, Fig. 5(c) shows the response of the sensor based on the hybrid ZnFe₂O₄/ZnO hollow spheres vs. temperature towards 50 ppm acetone, ammonia, methylbenzene and trimethylamine. It could be clearly observed that the maximum response towards methylbenzene (4.3) and acetone (5.2) was higher than that of ammonia (1.7) and trimethylamine (2.8). The response of the sensor based on the hybrid ZnFe₂O₄/ZnO hollow spheres reached its maximum value at 300 °C towards 50 ppm methylbenzene, which indicated that the sensor exhibited a higher response and lower optimum working temperature when the target gas was acetone. It seems that the working temperature is a vital parameter for a semiconductor to detect gases. Thus, acetone was selected to be the main tested gas at the working temperature of 280 °C due to the lower cost corresponding to the practical requirement. The single response transient curves of the sensor based on the hybrid ZnFe₂O₄/ZnO hollow spheres exposed to four types of reducing gases at 280 °C are shown in Fig. 5(d). The dynamic variation of the resistance response reveals that the resistance of the sensor increases rapidly when exposed to different reducing gases after a period of steady value in the air, which exhibits the properties of a p-type semiconductor.

Fig. 5 (a) Response of two gas sensors to 50 ppm C₃H₆O under different working temperatures. (b) Transient response of two gas sensors to 50 ppm C₃H₆O at 280 °C. Relationship of (c) responses to the working temperature and (d) resistance transients of the hybrid ZnFe₂O₄/ZnO hollow spheres-based sensor at the working temperature of 280 °C towards 50 ppm C₃H₆O, C₃H₉N, NH₃ and C₇H₈.

The typical dynamic response–recovery curves of the sensor based on the hybrid ZnFe₂O₄/ZnO hollow spheres to different concentrations of acetone at 280 °C are shown in Fig. 6(a), which indicate reproducible and reversible sensing properties. The response of the sensor increased abruptly when exposed to the target gas and recovered to its initial value after the tested gases were released. The responses towards 10, 20, 50, 100 and 200 ppm acetone were 3.2, 4.2, 5.2, 5.7 and 6.2, respectively. The plots of the response of the sensor based on the hybrid ZnFe₂O₄/ZnO hollow spheres to 10–2000 ppm acetone are shown in Fig. 6(b). The response increased quickly with an increase in acetone concentration (10–100 ppm), which is shown in the inset of Fig. 6(b). When the concentration was higher than 100 ppm, the response of the sensor began to be saturated, which was proven by the gradual response plateau.

Fig. 6 (a) Response of the hybrid ZnFe₂O₄/ZnO hollow spheres-based sensor towards 10–200 ppm acetone at 280 °C and (b) relationship of the sensor response to the concentration of acetone. (c) Resistance transient of the sensor based on the hybrid ZnFe₂O₄/ZnO hollow spheres at 280 °C towards 50 ppm acetone. (d) Five cycles response of the sensor towards 50 ppm acetone.

The response/recovery time is one type of important parameter of gas sensing properties. The magnifying response and recovery curve of the sensor based on the hybrid ZnFe₂O₄/ZnO hollow spheres towards 50 ppm acetone at 280 °C are shown in Fig. 6(c). The response time and recovery time were approximately 7.1 s and 10.8 s, respectively. The response/recovery time towards 10 ppm, 20 ppm, 50 ppm, 100 ppm and 200 ppm acetone were 9.5 s/20 s, 8.4 s/15 s, 7.1 s/10.8 s, 6.6 s/7.5 s and 5.8 s/6.7 s, respectively, which are demonstrated in Fig. S8.† Furthermore, they both decreased with an increase in temperature. In addition, the response and recovery time towards 50 ppm C₇H₈, C₃H₉N and NH₃ are shown in Fig. S9,† the values of which were in range of 5.6–7.2 s and 9–24 s, respectively. The experimental data led to the conclusion that the sensor based on the hybrid ZnFe₂O₄/ZnO hollow spheres shows not only a rapid response, but also fast recovery to C₃H₆O, along with C₇H₈, C₃H₉N and NH₃, which is believed to be beneficial for real-time gas detection. Fig. 6(d) displays the response reproducibility of the sensor based on the hybrid ZnFe₂O₄/ZnO hollow spheres within five cycles towards 50 ppm acetone at 280 °C to observe response fluctuations. The small deviation and no obvious change in the similar response shapes of the amplitude changes confirm that the sensor maintains a fast response/recovery state and good reproducibility.

The acetone responses of various materials reported in the literature are summarized in Table 1.^30–33 The fast response to acetone was probably due to two beneficial factors: on the one hand, the hollow structure could provide rapid diffusion of target molecules easily and enable gas molecules to adsorb onto the inner and outer surfaces of the materials. On the other hand, as a p-type semiconductor, ZnFe₂O₄ showed good catalytic reactivity (Fig. S10†).²⁶ Moreover, a fascinating electrochemical character ensured the electronic transmission through ZnFe₂O₄.²⁴

Table 1 Comparison of response times of various sensing materials-based sensors to acetone^a

Material	Gas	Tem./°C	Conc./ppm	Tr1/s	Ref.
a Tem.: temperature; Conc.: concentration; Tr1: response time Ref.: reference.
ZnFe2O4/ZnO	Acetone	280	50	7.1	This work
ZnFe2O4/ZnO	Acetone	320	50	13	30
Co3O4	Acetone	150	100	80	31
ZnO	Acetone	330	100	11	32
Ag–TiO2/SnO2	Acetone	275	50	28	33

---

Gas sensing mechanism

The excellent sensing properties can be explained as follows: the hollow structure could provide more space and active sites for target gases and adsorbed oxygen ions to react. In addition, the hollow structure could solve the problem of aggregation effectively, which is a bothersome obstacle for nanoparticles.¹³ Moreover, the surface possesses a porous structure, which can offer a sufficiently active surface and be beneficial to gas adsorption and gas diffusion because of good permeability (Fig. 7(a)).²⁰ In addition to the structural factor, the special hybrid component also plays an important role in sensing properties. Good conductivity ensures fast electronic transmission through p-type ZnFe₂O₄.^25,34 Through the combination of ZnFe₂O₄ and ZnO, oxygen molecules adsorb on the surface of hollow spheres more easily because of their catalytic activity to promote oxygen dissociation. Therefore, oxygen can diffuse faster to the surface vacancies; moreover, not only the quantity of adsorbed oxygen, but also the molecule-ion conversion rate increase. As a consequence, a fast response is obtained. It is believed that the resistance variation is due to the process of adsorption and desorption. When exposed to air, oxygen molecules could be adsorbed on the surface of the hollow spheres, and depletion layers form at the surface of the as-synthesized material, forming oxygen ions (O²⁻ and O⁻) from surface-adsorbed oxygen (O₂ + e⁻ → O²⁻ and O²⁻ + e⁻ → 2O⁻).³⁵ In addition, electrons would transfer from ZnO to ZnFe₂O₄, while holes would move in the opposite direction at the interface, resulting in the addition of a depletion layer (Fig. 7(b and c)).³⁶ When the sensor is exposed to reducing acetone gas, the adsorbed oxygen ions react with acetone as follows:³⁵

C₃H₆O + 8O⁻(ads) → 3CO₂↑ + 3H₂O + 8e⁻

Fig. 7 Schematic of (a and c) possible gas sensing mechanisms and (b) energy band structures of the hybrid ZnFe₂O₄/ZnO hollow spheres-based sensor.

The resistance increases because the electrons are sent back to the sensing material, which exhibits the properties of a p-type semiconductor and holes are the carrier.

Conclusions

In conclusion, hybrid ZnFe₂O₄/ZnO hollow spheres were synthesized using carbon spheres as sacrificial templates. The obtained hybrid ZnFe₂O₄/ZnO hollow spheres had a diameter of approximately 500 nm and a thickness of approximately 30 nm. The as-fabricated hollow spheres-based sensor showed excellent gas sensing properties to acetone compared with nanoparticles, especially a fast response/recovery speed. The response of the sensor was 5.2 towards 50 ppm acetone and the response/recovery time (T_r1/T_r2) was 7.1 s/10.8 s at 280 °C. All the properties mentioned indicate that the combination of ZnFe₂O₄ and ZnO p–n materials provides a possible approach for real-time acetone detection devices.

Conflict of interest

The authors declare no competing financial interest.

Acknowledgements

This study was supported by the Postdoctoral Science Foundation of China (No. 2015M571361), the Natural Science Foundation Committee (NSFC, Grant No. 51502110 and 61504136), the High Tech Project of Jilin Province (No. 20150204029GX), the Program for Chang Jiang Scholars and Innovative Research Team in University (No. IRT3018) and the Open Project from State Key Laboratory of Transducer Technology (Grant No. SKT1402).

References

1. J. Zhang, X. H. Liu, G. Neri and N. Pinna, Adv. Mater., 2016, 5, 795–831.
2. D. J. Liu, T. M. Liu, H. J. Zhang, C. L. Lv, W. Zeng and J. Y. Zhang, Mater. Sci. Semicond. Process., 2015, 221, 56–62.
3. Y. X. Liu, J. Parisi, X. C. Sun and Y. Lei, J. Mater. Chem. A, 2014, 2, 9919–9943.
4. A. Rehman and X. Q. Zeng, RSC Adv., 2015, 5, 58371–58392.
5. D. J. Wales, J. Grand, V. P. Ting, R. D. Burke, K. J. Edler, C. R. Bowen, S. Mintova and A. D. Burrows, Chem. Soc. Rev., 2015, 44, 4290–4321.
6. R. Zhang, L. L. Wang, J. N. Deng, T. T. Zhou, Z. Lou and T. Zhang, Sens. Actuators, B, 2015, 220, 1224–1231.
7. A. Kumar, S. Samanta, A. Singh, M. Roy, S. Singh, S. Basu, M. M. Chehimi, K. Roy, N. Ramgir, M. Navaneethan, Y. Hayakawa, A. K. Debnath, D. K. Aswal and S. K. Gupta, ACS Appl. Mater. Interfaces, 2015, 7, 17713–17724.
8. R. Sahay, P. S. Kumar, R. Sridhar, J. Sundaramurthy, J. Venugopal, S. G. Mhaisalkarc and S. Ramakrishna, J. Mater. Chem., 2012, 22, 12953–12971.
9. Z. Wang, Z. Li, J. Sun, H. Zhang, W. Wang, W. Zheng and C. Wang, J. Phys. Chem. C, 2010, 114, 6100–6105.
10. L. L. Wang, Z. Lou, R. Wang, T. Fei and T. Zhang, J. Mater. Chem., 2012, 22, 12453–12456.
11. H. J. Kim, K. Choi, A. Q. Pan, D. Kim, H. R. Kim, K. M. Kim, C. W. Na, G. Z. Cao and J. H. Lee, J. Mater. Chem. A, 2011, 21, 6549–6555.
12. L. L. Wang, Z. Lou, J. N. Deng, R. Zhang and T. Zhang, ACS Appl. Mater. Interfaces, 2015, 7, 13098–13104.
13. Y. F. Bing, Y. Zeng, C. Liu, L. Qiao and W. T. Zheng, Nanoscale, 2015, 7, 3276–3284.
14. L. L. Wang, J. N. Deng, T. Fei and T. Zhang, Sens. Actuators, B, 2012, 164, 90–95.
15. S. J. Kin, C. W. Na, I. S. Hwang and J. H. Lee, Sens. Actuators, B, 2012, 168, 83–89.
16. L. L. Wang, H. M. Dou, Z. Lou and T. Zhang, Nanoscale, 2013, 5, 2686–2691.
17. L. L. Wang, J. N. Deng, Z. Lou and T. Zhang, J. Mater. Chem. A, 2014, 2, 10022–10028.
18. S. W. Choi, A. Katoch, J. Zhang and S. S. Kim, Sens. Actuators, B, 2013, 176, 585–591.
19. X. G. Li, X. X. Li, J. Wang and S. W. Lin, Sens. Actuators, B, 2015, 219, 158–163.
20. L. L. Wang, Z. Lou, T. Fei and T. Zhang, J. Mater. Chem., 2012, 22, 4767–4771.
21. C. Yang, X. D. Cao, S. J. Wang, L. Zhang, F. Xiao, X. T. Su and J. D. Wang, Ceram. Int., 2015, 41, 1749–1756.
22. L. L. Gu, K. B. Zheng, Y. Zhou, J. Li, X. L. Mo, G. R. Patzke and G. R. Chen, Sens. Actuators, B, 2011, 159, 1–7.
23. N. Li, S. Jin, Q. Y. Liao and C. X. Wang, ACS Appl. Mater. Interfaces, 2014, 6, 20590–20596.
24. J. Y. Patil, D. Y. Nadargi, J. L. Gurav, I. S. Mulla and S. S. Suryavanshi, Ceram. Int., 2014, 40, 10607–10613.
25. B. I. J. William, P. B. John and U. M. K. Shahed, J. Am. Chem. Soc., 2004, 126, 10238–10239.
26. J. Wolska, K. Przepiera, H. Grabowska, A. Przepiera, M. Jablonski and R. Klimkiewicz, Res. Chem. Intermed., 2008, 34, 43–51.
27. X. M. Sun and Y. D. Li, Angew. Chem., Int. Ed., 2004, 43, 597–601.
28. H. Song, L. P. Zhu, Y. G. Li, Z. R. Lou, M. Xiao and Z. Z. Ye, J. Mater. Chem. A, 2015, 3, 8353–8360.
29. L. L. Wang, Z. Lou, T. Fei and T. Zhang, J. Mater. Chem., 2011, 21, 19331–19336.
30. S. R. Wang, X. L. Gao, J. D. Yang, Z. Y. Zhu, H. X. Zhang and Y. S. Wang, RSC Adv., 2014, 4, 57967–57974.
31. Z. Y. Zhang, Z. Wen, Z. Z. Ye and L. P. Zhu, RSC Adv., 2015, 5, 59976–59982.
32. M. Y. Ge, T. M. Xuan, G. L. Yin, J. Lu and D. N. He, Sens. Actuators, B, 2015, 220, 356–361.
33. V. K. Tomer and S. Duhan, J. Mater. Chem. A, 2016, 4, 1033–1043.
34. T. Nunome, H. Irie, N. Sakamoto, O. Sakurai, K. Shinozaki, H. Suzuki and N. Wakiya, J. Ceram. Soc. Jpn., 2013, 121, 26–30.
35. J. Zhang, J. M. Song, H. L. Niu, C. J. Mao, S. Y. Zhang and Y. H. Shen, Sens. Actuators, B, 2015, 221, 56–62.
36. Y. Y. Bu, Z. Y. Chen and W. B. Li, Dalton Trans., 2013, 42, 16272–16275.

---

Footnote

† Electronic supplementary information (ESI) available: Further experimental details on fabrication of gas sensors, the FESEM images, the EDX spot scanning images and detailed sensing performance of sensing materials are listed. See DOI: 10.1039/c6ra12201a

---