Ni-doped ZnO nanorods gas sensor: Enhanced gas-sensing properties, AC and DC electrical behaviors

Sensors and Actuators B 199 (2014) 403–409

Contents lists available at ScienceDirect

Sensors and Actuators B: Chemical journal homepage: www.elsevier.com/locate/snb

Ni-doped ZnO nanorods gas sensor: Enhanced gas-sensing properties, AC and DC electrical behaviors Mengmeng Xu, Qiang Li ∗ , Yuan Ma, Huiqing Fan State Key Laboratory of Solidification Processing, School of Materials Science and Engineering, Northwestern Polytechnical University, Xi’an 710072, China

a r t i c l e

i n f o

Article history: Received 26 December 2013 Received in revised form 4 March 2014 Accepted 28 March 2014 Available online 6 April 2014 Keywords: ZnO Doping Gas-sensing Current–voltage curve AC impedance spectroscopy

a b s t r a c t Ni-doped zinc oxide (ZnO) nanorods had been successfully fabricated via a fast microwave-assisted hydrothermal synthesis at 150 ◦ C. The morphology and composition were carefully characterized by X-ray diffraction, field emission scanning electronic microscopy, and transmission electron microscopy. Gas-sensing testing results demonstrated that Ni-doped ZnO nanorods had enhanced gas-sensing performance. Furthermore, AC impedance spectroscopy and DC current–voltage curves were observed to investigate the gas-sensing mechanism. Current–voltage curves are approximately close to a linear function, indicating the potential barriers formed at the electron-depleted surface layer occupy a dominant when carriers transport in the gas sensor, and AC impedance spectra indicates the potential barriers height of the electron-depleted surface layer. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Recently, metal oxide semiconductor gas sensors are among the most widely researched and used gas sensors due to their advantageous features, such as high sensitivity under ambient conditions, low power consumption, low price, prompt response and simple structure. A large number of semiconductor oxides such as TiO2 [1,2], SnO2 [3,4], ZnO [5–7], NiO [8,9], and Fe2 O3 [10,11] have been proven to be effective gas sensor materials. However, metal oxide semiconductor gas sensors have some deficiencies, such as: poor thermal stability, reliability, selectivity, anti-interference and high working temperatures, which limit the further development and widely application of metal oxide gas sensors. At present, numerous efforts have been made to improve the sensing properties of metal oxide semiconductor gas sensors, among which, the strategy of doping modification with various metallic elements, for example, noble metal [12], rare-earth metal [13,14], transition metal [4,15–17], and metal oxide [18], had been proven effective. Meanwhile, metal oxide semiconductors are promising candidates for gas sensor development and have received a majority of attention in recent decades. However, the fundamental understanding of sensing mechanism remains poor since an empirical optimization of gas-sensing performance has always been the investigation focus. To meet the demands for gas sensors which

are capable of detecting environmentally important gases within sub-ppb levels, a more fundamental basis understanding the gassensing mechanism is necessary. The working principle of a typical resistive gas-sensor material is based on a shift of the state of equilibrium of the surface oxygen reaction due to the presence of the target gas. The resulting evolution in chemisorbed oxygen is recorded as a change in resistance of the sensor material [19–22]. However, the change of resistance cannot reflect the real working process of the gas sensor. It is well known that AC impedance spectroscopy, DC resistance and current–voltage (I–V) characteristics have usually been used to analyze the electrical process [23,24], for example, the charge carriers translation, potential barriers [25]. Therefore, may be can they be applied to understand the gassensing mechanism. In this paper, we report an economical method for preparing Ni-doped ZnO nanorods. The morphology, structure, and gas sensing performance were carefully investigated. AC impedance spectroscopy and DC resistance, I–V characteristics were observed to analysis the gas-sensing mechanism. Furthermore, we expected such a gas senor based on Ni-doped ZnO nanorods could be reliably used for detection of inflammable gases. 2. Experimental 2.1. Synthesis

∗ Corresponding author. Tel.: +86 29 88494463; fax: +86 29 88492642. E-mail addresses: [email protected], [email protected] (M. Xu), [email protected] (Q. Li). http://dx.doi.org/10.1016/j.snb.2014.03.108 0925-4005/© 2014 Elsevier B.V. All rights reserved.

All the chemicals are of analytical reagent (AR) grade used without further purification, and purchased from Shanghai Chemical Industrial Co. Ltd. (Shanghai, China). Zinc nitrate hexahydrate

404

M. Xu et al. / Sensors and Actuators B 199 (2014) 403–409

(Zn(NO3 )2 ·6H2 O), nickel acetate tetrahydrate (Ni(Ac)2 ·4H2 O) d-(+)-glucose (C6 H12 O6 ·H2 O), hexamethylene tetramine (HMTA), and urea. In a typical procedure, 2.5 mmol of Zn(NO3 )2 ·6H2 O, 2 mmol of HMTA, 3.6 mmol of C6 H12 O6 ·H2 O and 1.25 mmol of urea were successively introduced into 40 mL deionized water while stirring for 10 min at room temperature. Then, 0, 0.125, 0.167, 0.25, 0.5, 2.5 mmol of Ni(Ac)2 ·4H2 O were added to solution, respectively. After being vigorously stirred for another 30 min at room temperature, the as-obtained mixture was transferred to a Teflon vessel of the MDS-6 (Microwave Digestion/Extraction System, Shanghai Sineo Microwave Chemical Technology Co. Ltd.). With a programmable temperature control, the desired reaction temperature was established to 150 ◦ C. The as-prepared powders were collected by centrifugation, rinsed several times with deionized water and pure ethanol, and then, vacuum-dried at 80 ◦ C for 8 h. Finally, the samples were obtained by calcining the precursor at 500 ◦ C for 2 h in air. 2.2. Characterization The phase identification of the as-obtained samples was performed by X-ray diffraction (XRD; X’pert, Philips, Holland) with ˚ at 40 kV, and 30 mA over the 2 of Cu K␣1 radiation ( = 1.5406 A) range 15–75◦ . The morphology and microstructure were carried out by using field emission scanning electronic microscopy (FE-SEM; JSM-6701F, JEOL, Japan). Transmission electron microscopy (TEM;

JEM-3010, Questar, New Hope, USA), high-resolution transmission electron microscopy (HRTEM) images combined with select area electron diffraction (SAED) and energy dispersive spectroscopy (EDS; FeatureMax, Oxford Instruments, Abingdon Oxfordshire, UK) were recorded on a FEI TecnaiF30G2 field emission microscope, operating at an acceleration voltage of 300 kV. The DC electric conductivity was measured by using a high resistance meter (4339B, Agilent, Santa Clara, CA, USA). AC impedance spectroscopy was performed using an Impedance Analyzer (4294A, Agilent, CA, USA) in the 100 Hz–1 MHz frequency range. I–V characteristics were measured using a Keithley instruments (Model 2410, Sourcemeter, Cleveland, USA) in the voltage range from −20 V to 20 V with a heating temperature varying from room temperature to 400 ◦ C. 2.3. Sensor fabrication and measurements The structure of the gas sensor belongs to side-heated type, and the basic fabricated process is as follows. The as-obtained Ni-doped ZnO nanorods were mixed and grinded with some terpineol adhesive in an agate mortar to form gas-sensing paste. The paste used as sensitive body was coated on a ceramic tube with Au electrodes and platinum wires, and then sintered at 500 ◦ C for 2 h to remove the binder. A Ni–Cr alloy crossing the ceramic tube was used as a heating resistor which is controlled by heating voltage Vh to ensure both substrate heating and temperature controlling. In order to improve their stability and repeatability, the gas sensors were aged at 300 ◦ C for 10 days in air. The gas-sensing properties were tested using a gas

Fig. 1. Typical SEM images of S1, S2, S3, S4, S5, S6.

M. Xu et al. / Sensors and Actuators B 199 (2014) 403–409

405

response instrument (HW-30A, Hanwei Ltd., Zhengzhou, China). The tested gas was introduced into a glass chamber with a volume of 15 L [26,27]. In the measuring electric circuit of gas sensor, a load resistor is connected in series with a gas sensor. The circuit voltage Vc is 5 V, and the output voltage (Vout ) is the terminal voltage of the load resistor RL . Gas response (S) is defined as the ratio of Rair /Rgas , where Rair and Rgas are the resistance values measured in air and tested gas, respectively. 3. Results and discussion 3.1. Morphology and crystalline structure Fig. 1(a–f) shows the FE-SEM images of samples with 0, 0.125, 0.167, 0.25, 0.5, 2.5 mmol of Ni(Ac)2 ·4H2 O added relating to S1, S2, S3, S4, S5, S6, respectively. As shown in Fig. 1(a), the pencil-like ZnO rods consist of a hexagonal trunk and two hexagonal tips on one side. The trunk has a diameter of about 2 ␮m in the middle and a length of about 10 ␮m. Then, the uniform, large-scale ZnO nanorods with regular hexagonal shape and flat end can be clearly observed in Fig. 1(b)–(d) when we increase the added amount of Ni element. The diameter and length of these nanorods are in the ranges of 100 ∼ 200 nm and 2 ∼ 3 ␮m, respectively. Further increase the amount of Ni(Ac)2 ·4H2 O to 0.5 mmol, ZnO nanorods covered with nanosheets are obtained. Under microwave hydrothermal conditions, the high temperature and pressure promote Ni atoms to nucleate anisotropic nanosheet structures rather than nanorods. As a results, when introduction of over amount of Ni element, more Ni atoms in nanosheets incorporate with nanorods. As shown in Fig. 1(f), when 2.5 mmol Ni(Ac)2 ·4H2 O was added, flower-like products composed of a large amount of nanosheets were fabricated. Fig. S1 displays the powder XRD pattern of nanosheets. Two sets of diffraction peaks can be observed from the spectrum of samples, which can be indexed to hexagonal wurtzite ZnO (JCPDS File No. 36-1451) and rhombohedral NiO (JCPDS File No. 44-1159). Moreover, the TEM and HRTEM images show two kinds of lattice spacing corresponding to the distance between the (1 0 0) planes

Fig. 2. Rietveld refinement of room-temperature XRD data of S4 using the FullProf program. Table 1 The cell parameters of all the samples (S1 ∼ S6). Samples

Crystal parameters (Å)

Unit cell volume (Å3 )

S1 S2 S3 S4 S5

a = 3.25(2), c = 5.21(4) a = 3.25(0), c = 5.21(1) a = 3.24(9), c = 5.20(6) a = 3.24(6), c = 5.20(4) a = 3.24(2), c = 5.20(5)

55.14(0) 55.04(1) 54.95(4) 54.83(2) 54.70(7)

in the ZnO crystal lattice and (1 0 1) planes in the NiO crystal lattice (Fig. S2). In addition, XRD Rietveld refinements for samples were carried out with P63 mc space group at room temperature by using the FullProf program. The XRD data were collected in a 2 range of 15–75◦ with a step size and time of 0.02◦ , 10 s, respectively. As shown in Fig. 2, the final factors, weighted profile factor (Rwp ), profile factor (Rp ) for S4 are 9.00%, 2.16%, respectively. The hexagonal cell parameters a and c of all samples extracted by Rietveld analysis are shown in Table 1. With the increase of Ni2+ concentration,

Fig. 3. (a) Selected image of S4 for the EDS mapping analysis. The corresponding elemental mapping images for Zn, O, Ni are presented in (b), (c), and (d), respectively.

406

M. Xu et al. / Sensors and Actuators B 199 (2014) 403–409

Fig. 4. (a) TEM of S4; (b, c) the corresponding SAED pattern and HRTEM images.

the volume of the unit cell decreases for the lower Ni2+ ions radii ˚ ˚ substitution for Zn2+ ions (0.74 A). (0.69 A) 3.2. Elements and composition analysis Fig. 3 shows the selected area for the elemental mapping of S4 by EDS and Fig. 3(b–d) presents the Zn, O, Ni elemental mappings, respectively. As expected, Ni is uniform distribution in the elemental mapping. In order to have a further investigation for S4, TEM, HRTEM, SAED are recorded and analyzed. Fig. 4(a) displays the TEM images. The HRTEM image in the Fig. 4(c) shows that ZnO rods have a lattice spacing of about 0.26(1) nm, corresponding to the distance between the (0 0 2) planes in the ZnO crystal lattice. Meanwhile, the SAED pattern confirms that ZnO hexagonal prism along the [0 0 0 2] direction. 3.3. Gas-sensing performance The working principle of metal oxide semiconductor gas sensors for detecting gas depends on the conductance variation of the sensing element, which considerably relies on both the test gas and the operating temperature of the sensors. Therefore, gas response and recovery characteristics of the sensors based on Ni-doped ZnO nanorods were investigated at different operating temperature to optimize the working temperature. Fig. 5(a) shows the typical response and recovery curves of sensors exposed to 50 ppm ethanol gas at various working temperature. One can observe that the highest response to ethanol gas appeared at 370 ◦ C. Moreover, the response and recovery curves in Fig. 5(b) indicated the sensors based on S4 exhibited the finest gas-sensing performance to 100 ppm ethanol. Therefore, the optimum operating temperature was determined to be 370 ◦ C for the subsequent detections of the sensors. It is well known that the sensitivity of the semiconductor oxide gas sensor can be empirically represented as S = 1 + Ag (Pg )ˇ

Fig. 5. (a) Typical response and recovery curve of sensors based on S4 exposed to 50 ppm ethanol vapor at different working temperatures; (b) typical response and recovery curve of sensors based on S1, S2, S3, S4, S5, S6 exposed to a 100 ppm ethanol vapor.

proportional to the gas concentration. Fig. 6 displays the typical response recovery characteristics of the sensors based on S4 exposed to ethanol vapor with the concentrations of 5, 10, 30, 50, 100, 200, 300, and 500 ppm at 370 ◦ C. It indicates that the gas response increases with ethanol vapor concentration within the

(1)

where, Pg is the target gas partial pressure, which is directly proportional to the gas concentration, Ag is a prefactor, and ˇ is the exponent on Pg . As a result, the sensitivity of gas sensor is usually

Fig. 6. Response and recovery curves of sensors based on S4 exposed to ethanol vapor ranging from 5 to 500 ppm (the inset is the plot of sensitivity corresponding to the different ethanol concentration).

M. Xu et al. / Sensors and Actuators B 199 (2014) 403–409

407

range of 5–500 ppm at 370 ◦ C, and the sensitivity of the sensors exposed to 500 ppm ethanol vapor reaches to 313. Furthermore, the sensors show a quick response and short recovery time, as shown in Fig. S3a in Supplementary data, the response and recovery times of the sensors based on S4 exposed to 100 ppm ethanol vapor are approximate 7 s and 10 s, respectively. In addition, selectivity of the gas sensor is another very important factor in the gas sensing application. The results of selectivity testing show that the sensors based on S4 have excellent selectivity for ethanol gas (Fig. S3b in Supplementary data). 3.4. Gas sensing mechanism It is well known that the working principle of a typical metal oxide semiconductor gas sensor is based on a shift in the state of equilibrium of the surface oxygen reaction due to the presence of the target gases. For n-type metal oxide semiconductors, the negatively charged oxygen species absorbed on the surface of materials establish an electron depletion layer on the surface of grains. The resistance of n-type metal oxide semiconductors gas sensor in air is therefore high, due to the development of potential barrier in electron depletion layer. When the sensor is exposed to atmosphere containing reducing gases at increased temperatures, the oxygen adsorbates are removed by reduction reaction, so that the steadystate surface coverage of the adsorbates is lowered leading to a decrease in resistance. In order to have a meaningful understanding for the gas-sensing mechanism, AC impedance spectroscopy, DC I–V resistance and I–V characteristics were carefully observed. Twoprobe I–V curves were measured between −20 and +20 V using a Keithley 2410 source meter. The characteristics of sensors based on S4 in the temperature range from 353 K to 673 K were measured, and the results are presented in Fig. 7. The current vs. voltage relationship is a nonlinear curve. Although it is approximately close to a linear function, one can see more clearly from the nonlinear behavior in the R/V curves as shown in Fig. 8. The entire R/V curve is nonlinear with the most pronounced region occurring in low voltages (<5 V). Fig. 7(a) shows I/V characteristics of sensor depending on the temperature in air. The inset (Fig. 7(a)) on the top left corner displays the plot of logarithm of conductivity vs. 1/T. The curves indicate three different regions which are correspond to the change of oxygen adsorbates on the surface of ZnO grains. The plot on the bottom right corner shows the slopes of the nearly linear I/V characteristics increase slowly with the increase of temperature. As a contrast, the I/V characteristics of sensor depending on the temperature in 50 ppm ethanol gas were also measured and shown in Fig. 7(b). It can be seen that the conductivity and I/V characteristics have made great change in the temperature range from 573 K to 673 K. The slopes increase rapidly with the increase of temperature. In order to investigate the effects of the tested gases on the overall resistance of the sensors, AC impedance spectra in air and 5 ppm ethanol vapor are observed, respectively. The complex impedance spectra are displayed in Fig. S4 in Supplementary data. Fig. 9 shows the variation of the real part (Z ) and imaginary part (Z ) of the impedance with frequency at various temperatures in air and 5 ppm ethanol vapor, respectively. As shown in Fig. 9, the Z increases with frequency at various temperatures and reach a max imum peak Zmax . The change in Z is an indication of the accumulation of space charges in the sensor. Meanwhile, a rapid decrease in the value of Z occurs at about 10 KHz, and finally, the values merge at high frequencies at each temperature. It may be due to the release   of space charges. In addition, both the Zmax and Zmax have a tremendous decrease at 573 K and 673 K in 5 ppm ethanol atmosphere   compared with the values in air. Moreover, the Zmax and Zmax have the minimum value at 673 K in 5 ppm ethanol atmosphere. The change of conductivity with the temperature can be explained by bulk conductivity changes in semiconducting oxides.

Fig. 7. I/V characteristics of sensors based on S4 depending on the temperature at air (a) and 50 ppm ethanol vapor (b) atmosphere. The inset on the top left corner displays the plot of logarithm of conductivity vs. 1/T and the inset on the bottom right corner shows the slopes increase slowly of the nearly linear I/V characteristics with the increase of temperature.

The change of the defect in semiconducting oxides at elevated temperatures is well recognized. This change affects the electrical conductivity of the materials. However, the conductivity shows a substantial change when a small conductivity of a combustible gas is present, and the mechanism of bulk conductivity change cannot explain the phenomenon. The assumption, therefore, is that surface processes control the conductivity. It is well known, at 100 ∼ 500 ◦ C, oxygen molecules adsorb onto the surfaces of oxide semiconductors and ionize into kinds of species such as O2 − , O− , and O2− by taking the electrons near the surfaces of the semiconductors [28].

Fig. 8. The R/V curves of sensors based on S4 at air and 50 ppm ethanol vapor atmosphere.

408

M. Xu et al. / Sensors and Actuators B 199 (2014) 403–409

revealed the enhanced gas-sensing performance of Ni-doped ZnO nanorods, the sensitivity of the sensors based on Ni-doped ZnO nanorods at 500 ppm ethanol gas vapor atmosphere is up to 313. Such a gas senor based on Ni-doped ZnO nanorods could be reliably used for detection of inflammable gases. The investigation of AC impedance spectroscopy and DC I–V resistance, I/V characteristics depending on the temperature at different gases promote to a meaningful understanding of the gas-sensing mechanism, such as: the change of oxygen species with temperature on the surface of grains, the building of the space charge region on the surface of grains, the changes of conductivity of ZnO gas sensor. Acknowledgments This work was supported by the National Natural Science Foundation of China (51172187), the SPDRF (20116102130002, 20116102120016) and 111 Program (B08040) of MOE, and Xi’an Science and Technology Foundation (CX1261-2, CX1261-3, CX12174, XBCL-1-08), and Shaanxi Province Science Foundation (2013KW12-02), and the NPU Fundamental Research Foundation (NPU-FRF-JC201232) of China, Shaanxi Province Foundation for Returned Scholars, Aeronautical Science Foundation (2013ZF53072), and Scientific Research Foundation for the Returned Overseas Chinese Scholars, State Education Ministry. Fig. 9. Variation of the imaginary part (Z ) and the real part (Z ) (inset) of impedance of the sensors based on S4 with frequency at various temperatures: (a) in air (b) in 50 ppm ethanol vapor.

Among these, O− is the most reactive with CH3 CH2 OH molecule in the temperature range of 300 ∼ 500 ◦ C. The following is the reaction when the sensor is exposed to ethanol atmosphere [29]. O2 (g) + 2e → 2O− (s)

(2)

CH3 CH2 OH(s) + 6O− (s) → 2CO2 (g) + 3H2 O(g) + 6e

(3)

In the case of n-type semiconducting oxides, the formation of oxygen adsorbates builds space charge region on the surfaces of the oxide grains resulting in an electron-depleted surface layer. Therefore, the changes of conductivity of ZnO gas sensor at different temperature and gas atmosphere origin from both the bulk conductivity and surface conductivity. Generally speaking, electron transport in ZnO gas sensor may overcome the potential barriers formed at the electron-depleted surface layer, grain boundaries potential barriers and Schottky barrier between the metal electrode semiconductor interfaces. The DC I/V characteristics and AC impedance spectra just provide the direct evidence for the building of the space charge region and the individual contributions to the sensor response from the grains, grain boundaries and oxide/electrode interfaces [30]. Each set of I/V curves are approximately close to a linear function. Then, we can conclude that the potential barriers formed at the electron-depleted surface layer occupy a dominant when carriers transport in the ZnO gas sensor. When the sensors react with tested gas at the optimum operating temperature, the surface potential barrier height and resistance have a maximal drop. Therefore, it shows the maximal conductivity as shown in I/V characteristics. Furthermore, one can also observe the accumulation and release process of the space charge with the frequency via the AC impedance spectra (Fig. 9) [31], which further indicate the lowest potential barrier height of the electron-depleted surface layer when the sensors work at the optimum operating temperature. 4. Conclusions We successfully prepared Ni-doped ZnO nanorods via fast microwave hydrothermal synthesis. The gas-sensing experiment

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.snb.2014.03.108. References [1] B. Lyson-Sypien, A. Czapla, M. Lubecka, P. Gwizdz, K. Schneider, K. Zakrzewska, K. Michalow, T. Graule, A. Reszka, M. Rekas, A. Lacz, M. Radecka, Nanopowders of chromium doped TiO2 for gas sensors, Sens. Actuators B: Chem. 0 (2012) 163–172. [2] S. Lin, D. Li, J. Wu, X. Li, S.A. Akbar, A selective room temperature formaldehyde gas sensor using TiO2 nanotube arrays, Sens. Actuators B: Chem. 2 (2011) 505–509. [3] B. Wang, L.F. Zhu, Y.H. Yang, N.S. Xu, G.W. Yang, Fabrication of a SnO2 nanowire gas sensor and sensor performance for hydrogen, J. Phys. Chem. C 17 (2008) 6643–6647. [4] G. Korotcenkov, I. Boris, V. Brinzari, S.H. Han, B.K. Cho, The role of doping effect on the response of SnO2 -based thin film gas sensors: analysis based on the results obtained for Co-doped SnO2 films deposited by spray pyrolysis, Sens. Actuators B: Chem. 0 (2013) 112–124. [5] K.V. Gurav, M.G. Gang, S.W. Shin, U.M. Patil, P.R. Deshmukh, G.L. Agawane, M.P. Suryawanshi, S.M. Pawar, P.S. Patil, C.D. Lokhande, J.H. Kim, Gas sensing properties of hydrothermally grown ZnO nanorods with different aspect ratios, Sens. Actuators B: Chem. 0 (2014) 439–445. [6] V. Talwar, O. Singh, R.C. Singh, ZnO assisted polyaniline nanofibers and its application as ammonia gas sensor, Sens. Actuators B: Chem. 0 (2014) 276–282. [7] F. Tong, K. Kim, Y. Wang, R. Thapa, Y. Sharma, A. Modic, A. Claude Ahyi, T. Issacs-Smith, J. Williams, H. Ahn, H. Park, D. Kim, S. Lee, E. Lim, K.K. Lee, M. Park, Growth of ZnO nanorod arrays on flexible substrates: effect of precursor solution concentration, ISRN Nanomater. 2012 (2012) 7. [8] H. Steinebach, S. Kannan, L. Rieth, F. Solzbacher, H2 gas sensor performance of NiO at high temperatures in gas mixtures, Sens. Actuators B: Chem. 1 (2010) 162–168. [9] L.T. Hoa, H.N. Tien, V.H. Luan, J.S. Chung, S.H. Hur, Fabrication of a novel 2Dgraphene/2D-NiO nanosheet-based hybrid nanostructure and its use in highly sensitive NO2 sensors, Sens. Actuators B: Chem. 0 (2013) 701–705. [10] L. Wang, T. Fei, Z. Lou, T. Zhang, Three-dimensional hierarchical flowerlike ␣-Fe2 O3 nanostructures: synthesis and ethanol-sensing properties, ACS Appl. Mater. Interfaces 12 (2011) 4689–4694. [11] X. Wang, W. Liu, J. Liu, F. Wang, J. Kong, S. Qiu, C. He, L. Luan, Synthesis of nest like ZnO hierarchically porous structures and analysis of their gas sensing properties, ACS Appl. Mater. Interfaces 2 (2012) 817–825. [12] M. Schweizer-Berberich, J.G. Zheng, U. Weimar, W. Göpel, N. Bârsan, E. Pentia, A. Tomescu, The effect of Pt and Pd surface doping on the response of nanocrystalline tin dioxide gas sensors to CO, Sens. Actuators B: Chem. 1–2 (1996) 71–75. [13] T. Zhang, F. Gu, D. Han, Z. Wang, G. Guo, Synthesis, characterization and alcoholsensing properties of rare earth doped In2 O3 hollow spheres, Sens. Actuators B: Chem. 0 (2013) 1180–1188.

M. Xu et al. / Sensors and Actuators B 199 (2014) 403–409 [14] P. Mohanapriya, H. Segawa, K. Watanabe, K. Watanabe, S. Samitsu, T.S. Natarajan, N.V. Jaya, N. Ohashi, Enhanced ethanol-gas sensing performance of Ce-doped SnO2 hollow nanofibers prepared by electrospinning, Sens. Actuators B: Chem. 0 (2013) 872–878. [15] A. Galdikas, V. Jasutis, S. Kaˇciulis, G. Mattogno, A. Mironas, V. Olevano, D. ˇ ˙ A. Setkus, Peculiarities of surface doping with Cu in SnO2 thin film Senuliene, gas sensors, Sens. Actuators B: Chem. 1–3 (1997) 140–146. [16] Y. Cai, H. Fan, M. Xu, Q. Li, C. Long, Fast economical synthesis of Fe-doped ZnO hierarchical nanostructures and their high gas-sensing performance, CrystEngComm 36 (2013) 7339–7345. [17] H. Ahn, Y. Wang, S.H. Jee, M. Park, Y.S. Yoon, D. Kim, Enhanced UV activation of electrochemically doped Ni in ZnO nanorods for room temperature acetone sensing, Chem. Phys. Lett. 4–6 (2011) 331–335. [18] H. Nanto, T. Morita, H. Habara, K. Kondo, Y. Douguchi, T. Minami, Doping effect of SnO2 on gas sensing characteristics of sputtered ZnO thin film chemical sensor, Sens. Actuators B: Chem. 1–3 (1996) 384–387. [19] A. Rothschild, Y. Komem, The effect of grain size on the sensitivity of nanocrystalline metal-oxide gas sensors, J. Appl. Phys. 11 (2004) 6374–6380. [20] S.R. Morrison, Mechanism of semiconductor gas sensor operation, Sens. Actuators 3 (1987) 283–287. [21] M.E. Franke, T.J. Koplin, U. Simon, Metal and metal oxide nanoparticles in chemiresistors: does the nanoscale matter? Small 1 (2006) 36–50. [22] N. Yamazoe, New approaches for improving semiconductor gas sensors, Sens. Actuators B: Chem. 1–4 (1991) 7–19. [23] T. Kuwabara, Y. Kawahara, T. Yamaguchi, K. Takahashi, Characterization of inverted-type organic solar cells with a ZnO layer as the electron collection electrode by ac impedance spectroscopy, ACS Appl. Mater. Interfaces 10 (2009) 2107–2110. [24] J.T.S. Irvine, D.C. Sinclair, A.R. West, Electroceramics characterization by impedance spectroscopy, Adv. Mater. 3 (1990) 132–138. [25] P.R. Bueno, E.R. Leite, M.M. Oliveira, M.O. Orlandi, E. Longo, Role of oxygen at the grain boundary of metal oxide varistors: a potential barrier formation mechanism, Appl. Phys. Lett. 1 (2001) 48–50. [26] Y. Cai, H. Fan, One-step self-assembly economical synthesis of hierarchical ZnO nanocrystals and their gas-sensing properties, CrystEngComm 44 (2013) 9148–9153.

409

[27] W. Yan, H. Fan, Y. Zhai, C. Yang, P. Ren, L. Huang, Low temperature solution-based synthesis of porous flower-like ␣-Fe2 O3 superstructures and their excellent gas-sensing properties, Sens. Actuators B: Chem. 1 (2011) 1372–1379. [28] N. Yamazoe, J. Fuchigami, M. Kishikawa, T. Seiyama, Interactions of tin oxide surface with O2 , H2 O and H2 , Surf. Sci. 0 (1979) 335–344. [29] Y. Liu, J. Dong, P.J. Hesketh, M. Liu, Synthesis and gas sensing properties of ZnO single crystal flakes, J. Mater. Chem. 23 (2005) 2316–2320. [30] J. Fleig, The influence of non-ideal microstructures on the analysis of grain boundary impedances, Solid State Ionics 1–2 (2000) 117–127. [31] S.K. Rout, A. Hussian, J.S. Lee, I.W. Kim, S.I. Woo, Impedance spectroscopy and morphology of SrBi4 Ti4 O15 ceramics prepared by soft chemical method, J. Alloy Compd. 1–2 (2009) 706–711.

Biographies Mengmeng Xu is now a master student at the School of Materials Science and Engineering, Northwestern Polytechnical University, under the supervision of Qiang Li. His research interests include the synthesis of function materials and their application in gas sensor. Qiang Li has been an associate Professor in Northwestern Polytechnical University. His research interests include crystal growth, nanocrystalline materials for photocatalysis as well as chemical sensors and thin-layer devices. Huiqing Fan obtained his BSc in Physics, MSc in Electronic Engineering, and Ph.D. in Electronic Materials Science from Xi’an Jiaotong University. He has been a Professor in Northwestern Polytechnical University. He has published more than 200 pear-reviewed papers. His research interests include ferroelectric, piezoelectric, pyroelectric and photo-electronic ceramics in polycrystalline and single crystal form, including nanocrystalline materials for photo-catalysis as well as chemical sensors and thin-layer devices.