Sensing behavior to ethanol of tin oxide nanoparticles prepared by microwave synthesis with different irradiation time

Sensors and Actuators B 194 (2014) 96–104

Contents lists available at ScienceDirect

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

Sensing behavior to ethanol of tin oxide nanoparticles prepared by microwave synthesis with different irradiation time N. Rajesh a , J.C. Kannan b , T. Krishnakumar c,∗ , S.G. Leonardi d , G. Neri d a

Department of Physics, KSR College of Engineering, Tiruchengode 637215, Tamilnadu, India Department of Physics, KSR Institute for Engineering and Technology, Tiruchengode 637215, Tamilnadu, India Department of Physics, Tagore Institute of Engineering and Technology, Attur, Salem 636112, Tamilnadu, India d Department of Electronic Engineering, Chemistry and Industrial Engineering, University of Messina, Messina 98166, Italy b c

a r t i c l e

i n f o

Article history: Received 15 July 2013 Received in revised form 12 December 2013 Accepted 16 December 2013 Available online 24 December 2013 Keywords: Tin oxide Nanostructures Microwave technique Ethanol sensor

a b s t r a c t Crystalline tin oxide nanoparticles were successfully synthesized by microwave-assisted technique without any post annealing process. The morphology, microstructure and phase composition of the products obtained applying microwave irradiation for different time intervals were examined by XRD, FT-IR, SEM-EDX, TEM and HRTEM. Characterization results indicated that microwave irradiated products are composed of crystalline SnO2 nanoparticles which exhibit the cassiterite-type tetragonal crystal structure. The sensing properties of as-prepared SnO2 nanoparticles towards ethanol at low operating temperature were investigated. Such sensor devices exhibited good response to low concentrations of ethanol at temperature below 100 ◦ C. An abnormal sensing behavior was registered, that is the sensor resistance increases in the presence of ethanol maintaining, at the same time, the usual n-type behavior with other reducing gases such as CO. In contrast, after annealing the SnO2 nanoparticles at 400 ◦ C, the sensors show the expected regular behavior in all range of operating temperature investigated. A plausible mechanism, linked to a specific interaction between the surface of SnO2 and ethanol molecule through its hydroxyl group, was suggested in order to describe the unusual sensing behavior observed. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Metal oxide nanocrystals provide ideal systems not only for understanding nanoscale effects in a nanostructured system, but can effectively be used in practical solid state devices. A specific parameter displayed by metal oxide nanocrystals is the high surface-to-volume ratio. The high percentage of surface atoms introduce many size-dependent phenomena which can, for example, modify the interaction of the metal oxide nanocrystals with the surrounding gaseous atmosphere, a mechanism well exploited in catalysis and gas sensing. So, nanostructured metal oxides have receiving considerable attention in the last years for sensing applications [1]. SnO2 , a wide band gap semiconductor (∼3.6 eV at room temperature) with relatively low electrical resistivity (∼10−4  cm) and good chemical stability [2,3] is utilized in nanostructured form for highly reflective coatings, UV and IR filters, heating layers for protective wide screens, transparent electrodes in solar cells, flat panel displays, and gas sensors [4,5]. Many methods had been

∗ Corresponding author. Tel.: +91 9003911518. E-mail address: tkrishnakumar [email protected] (T. Krishnakumar). 0925-4005/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.snb.2013.12.060

developed to synthesize SnO2 nanoparticles such as high energy ball milling method [6], homogeneous precipitation [7], sonochemical [8], hydrothermal [9], solvothermal [10], microemulsion [11], sol–gel route [12], spray pyrolysis [13], polymerized complex citrate route [14] and non-aqueous approaches [15]. However, generally a thermal treatment at high temperature is performed in order to obtain a crystalline material [2,16–19], while only few articles have described the preparation of crystalline tin oxide nanostructures without high temperature treatments [20–22]. In the present work, we report the preparation of tin oxide nanostructures by a simple and low cost wet chemical route assisted by microwave irradiation without necessity of any time-consuming post-synthesis annealing treatment. Indeed, as dipoles in the solution absorb microwaves, the irradiation is converted into heat with high energy efficiency. Enormous accelerations in reaction time can be achieved, so a reaction that takes several hours under conventional conditions can be completed over the course of minutes. Here, we reported that the SnO2 nanostructures were prepared within few (5–15) min. SnO2 -resistive type ethanol sensors are commonly applied in the biomedical and chemical industries to assess wine quality, food degradation, to identify drunk drivers, and to monitor fermentation and other processes in chemical industries, etc. [23–25]. Generally,

N. Rajesh et al. / Sensors and Actuators B 194 (2014) 96–104

these sensors operate at high temperatures (>150 ◦ C) in order to promote the sensing mechanism and obtain higher response and reduced response/recovery time. Low temperature operating sensors are however desirable to decrease power consumption [26]. SnO2 nanoparticles synthesized in the present study were therefore investigated in order to assess their sensing properties towards ethanol monitoring at low operating temperature (<100 ◦ C). Interestingly, an anomalous behavior in the sensing of ethanol has been observed. Indeed, although the n-type behavior typical of SnO2 has been ascertained with other reducing gases such as CO on these sensors, ethanol behaves differently showing an inversion of response. Such an unusual behavior has been attributed to a specific interaction between the surface of SnO2 and ethanol molecule through its hydroxyl group. 2. Experimental

a Pt heater located on the backside. The sensors were introduced in a stainless-steel test chamber for the sensing tests. The experimental bench for the electrical characterization of the sensors, allows to carry out measurements in controlled atmosphere. Preliminary tests were carried out in order to evaluate the electrical resistance of the sensor by increasing the temperature from 25 to 100 ◦ C, with step of 8 ◦ C/min, under 100 cm3 /min dry air flow. Electrical sensing tests were carried out in the temperature range from room temperature (25 ◦ C) to 100 ◦ C, with steps of 25 ◦ C, under a dry air total stream of 100 sccm, collecting the sensors resistance data in the four point mode. Gases coming from certified bottles can be further diluted in air at a given concentration by mass flow controllers. A multimeter data acquisition unit Agilent 34970A was used for this purpose, while a dual-channel power supplier instrument Agilent E3632A was employed to bias the built-in heater of the sensor to perform measurements at super-ambient temperatures. The gas response, S, is defined as:

2.1. Synthesis S= A tin hydroxide solution was first prepared by dissolving tin chloride with double distilled water at 0.1 M concentration. The ammonia solution was added to above precursor solution under constant stirring until the pH of the solution is 8. The resulting precipitate was washed with double distilled water until no chlorine ions were detected in the silver nitrate test. The obtained precipitate was divided into two parts. While a part of solution was left to evaporate slowly and then dried at 120 ◦ C in a conventional oven (SnO2 -0-120), the remaining solution was further divided into three parts and these solutions were separately placed in a microwave oven (2.45 GHz, 800 W) and irradiated for 5 min; the resulting precipitated were dried at 120 ◦ C in a conventional oven and named (SnO2 -5-120), 10 min (SnO2 -10-120) and 15 min (SnO2 -15-120) respectively. Further, a portion of sample SnO2 -15, named (SnO2 -15-400), and has been annealed at 400 ◦ C for 2 h in air. 2.2. Characterization The crystalline structure of the samples were analyzed by Xray diffraction (XRD) using a Bruker AXS D8 Advance instrument ˚ The average crysand using the Cu K␣1 wavelength of 1.5406 A. talline size of the crystallites was evaluated using the Scherrer’s formula d = k/ˇ cos  where d is the mean crystalline size, k is a grain shape dependent constant (0.9),  is the wavelength of the incident beam,  is a Bragg’s reflection angle, and ˇ is the full width half maximum. The surface morphology of the nanostructures were observed by scanning electron microscopy (SEM), using a JEOL 5600LV microscope at an accelerating voltage of 10 kV. Transmission electron microscopy (TEM) was recorded on a TechnaiG20-stwin using an accelerating voltage of 200 kV. The Fourier transform infrared spectra (FT-IR) of the samples were recorded using a 5DX FTIR spectrometer. The chemical compositions have been investigated by means of X-ray photoelectron spectroscopy (XPS) acquiring the spectra by means of a K-Alpha system from thermo scientific equipped with a monochromatic Al K␣ source (1486.6 eV) and operating with an analyzer in CAE mode with a pass energy of 200 and 50 eV for survey and high resolution spectra, respectively. A spot size diameter on the samples of about 400 ␮m has been adopted. Surface charging effects have been avoided using an electron flood gun. 2.3. Electrical and sensing test Sensors were made by depositing by drop coating films (1–10 ␮m thick) of the nano-powders dispersed in water on alumina substrates (6 × 3 mm2 ) with Pt interdigitated electrodes and

97

 R  Rair

× 100

(1)

where R = Rgas − Rair , with Rgas representing the electrical resistance of the sensor at different ethanol concentrations in dry air and Rair the baseline resistance in dry air. 3. Results and discussion 3.1. Materials characterization Microwave treatment is known to be a rapid approach to the synthesis of SnO2 nanostructures, with a good capability to control the particle shape and particle size [27–31]. In many cases, a surfactant/template agent is also added. For example, Xi et al., obtained SnO2 nanoparticles via a microwave method in conjunction with the presence of a surfactant and template [32]. Instead, in our synthesis approach, no surfactant or template agents have been utilized. Fig. 1a shows the XRD pattern of the dried precipitate and microwave irradiated samples for 5 and 15 min, respectively. The XRD pattern of SnO2 -0-120 indicates that, in the absence of microwave irradiation, a tin (oxy) hydroxide phase having the composition Sn6 O4 (OH)4 (JCPDS no. 84-2157) has been formed. Sekar and coworkers also reported the formation of the Sn6 O4 (OH)4 phase under similar synthesis conditions [33]. However, they were unable to obtain the crystalline SnO2 phase after microwave irradiation and an annealing at high temperature (500 ◦ C) of the intermediate SnO formed was necessary to complete the conversion to crystalline SnO2 . On the contrary, our microwave irradiated samples shows the formation of SnO2 nanostructures. Indeed, the XRD patterns of all irradiated samples matches well with JCPDS card no. 41-1445 of SnO2 in the cassiterite-type tetragonal crystal structure [16,34]. This confirms that the microwave radiation causes the conversion of tin hydroxyl group into SnO2 nanostructures without necessity of post-synthesis heating. Furthermore, the increase of peak intensity suggests also that the microwave treatment improved the crystalline structure. Fig. 1b shows an enlargement of the [2 1 1] diffraction peak. Increasing the irradiation time from 5 to 15 min only a slight decrease of the line broadening at half the maximum intensity (FWHM) has been observed. By using Scherrer’s formula the average crystalline size of the tin oxide crystals were calculated to be 21 and 24 nm for the samples SnO2 -5-120 and SnO2 -15-120, respectively. FT-IR spectra of irradiated samples are shown in Fig. 2. Data collected are in agreement with XRD results. The narrow peaks at 623 cm−1 confirm the formation of crystalline SnO2 phase [17,18,35]. The spectra exhibit also a common broad band

98

N. Rajesh et al. / Sensors and Actuators B 194 (2014) 96–104

a

(A)

400

200

50

60

70

80

2 (degree)

3403.48

200

(C)

100

622.97

1616.23

Intensity (cps)

b

622.44

40

624.25

30

1618.07

20

3403.30

0

(B)

1621.36

3398.22

600

Transmittance (%)

Intensity (cps)

800

4000 3500 3000 2500 2000 1500 1000

0 50

51

52

53

54

2 (degree)

500

-1

Wave numbers (cm ) Fig. 2. FTIR spectra of microwave irradiated samples.

Fig. 1. (a) XRD patterns of as-precipitate tin oxide (SnO2 -0) and microwave irradiated samples for 5 and 15 min; (b) enlargement of the [2 1 1] diffraction peak.

around 3403 cm−1 due to the OH-stretching vibrations of free and hydrogen-bonded hydroxyl groups [19]. The peak at 1616 cm−1 may be due to the adsorbed water and ammonia [36]. Importantly, XRD and FT-IR analyses suggests that no significant variation in the phase composition and average grain size of SnO2 nanostructures occurred for the various microwave irradiation time adopted. The morphology and chemical composition of the microwave synthesized tin oxide nanostructures were analyzed by SEM-EDX (Fig. 3). The micrograph of SnO2 -10-120 sample shows the typical morphology of the SnO2 powders synthesized, characterized by the presence of particles with an average size of about 1 ␮m or less. The EDX spectrum consists only of tin and oxygen peaks. The absence of other significant peaks indicates the purity of samples, and confirms that highly pure tin oxide nanostructures were synthesized without post synthesis annealing. Average crystallite size calculated by XRD is much smaller than deduced by SEM observations, and then it is likely that particles seen in the micrograph are composed of smaller particles. Therefore, an investigation by means of TEM was undertaken. A collection of TEM micrographs of irradiated samples is reported in Fig. 4. It can be observed the presence of large agglomerate composed of smaller round-shaped crystallites ranging from 3 to 30 nm. The particle size observed from TEM micrograph match well with the average crystalline size calculated from XRD pattern, suggesting that the individual grains imaged by TEM are single crystallites. The formation of such single crystalline SnO2 nanostructures was also revealed by HRTEM. Lattice fringes allowed to calculate the spacing between adjacent lattice planes (0.34 nm, corresponding

to (1 1 0) plane of tetragonal Cassiterite SnO2 crystals), confirming the formation of the SnO2 crystalline nanostructure. The composition, oxidation state and stoichiometry of the samples of the SnO2 -15 series vs. thermal treatment were investigated by XPS. The survey spectra of all samples are dominated by the lines of Sn and O. Fig. 5 shows the Sn 3d and O 1s photoelectron peaks for the SnO2 -15-120 and SnO2 -15-400 samples. Sn 3d spectrum (Fig. 5a) for the SnO2 -15-120 is characterized by the presence of a spin–orbit doublet, namely Sn 3d5/2 and 3d3/2 core levels, located at 487.5 and 495.9 eV, respectively. Binding energies registered are in according with the presence of Sn+4 , and also the spin–orbit splitting of 8.4 eV is consistent with the data reported in literature [37]. The annealed sample, SnO2 -15-400, shows the peak positions shifted towards lower binding energies at 487.2 and 495.6 eV. However, discrimination between Sn+2 and Sn+4 in XPS was reported to be very difficult due to the small shift in the Sn 3d binding energy for the two oxidation state of tin [38]. In Fig. 5b the photoelectron peaks of the O 1s core levels are shown. The spectra are characterized by a peak positioned at about 531.3 eV for the SnO2 -15-120 sample, caused by oxygen inside nonstoichiometric oxides within the surface region [39]. Moreover a clear asymmetry towards the high binding energy side is evident, due to oxygen of the O–H bonds coming from hydroxylated groups on the surface. The SnO2 -15-400 sample has instead a more narrow peak at about 531.1 eV and a decreased asymmetry, indicating the decrease of hydroxylated groups on the tin dioxide surface on the annealed sample. Values of O/Sn ratio calculated are about 1 and of 1.4 for the SnO2 -15-120 and SnO2 -15-400 samples, respectively, suggesting that surface O vacancies are filled up, i.e. surface oxygen ions, especially on the topmost O-terminated layers, are increased

N. Rajesh et al. / Sensors and Actuators B 194 (2014) 96–104

99

Fig. 3. SEM image (a) and corresponding EDX pattern (b) of microwave irradiated SnO2 -10 sample.

by annealing in air. However, the SnO2 crystallite appears to remain deficient in oxygen on the surface. 3.2. Sensing tests The potential gas sensing capability of the SnO2 samples for ethanol at low concentrations in air were investigated. The thick film deposited on the sensor substrate was pretreated at 100 ◦ C before electrical and sensing experiments. This low temperature treatment offers the advantage to perform a fast thermal cycling with low power consumption. Preliminary, it has been observed that the sample not treated in microwave oven, SnO2 -0-120, show a very high resistance out of the measurement range of our instrumentation. Instead, the microwave-treated SnO2 samples deposited on the alumina substrate show low resistance in

air (in the range of k), well measurable at room temperature with conventional instrumentation. On the basis of characterization studies (see XRD above) it can be suggested that the increase of conductivity of the latter sensors is due to the formation of the more conductive SnO2 phase and its higher crystallinity. Further, preliminary has been verified that the characteristics of sensitivity and signal stability improved strongly increasing the irradiation time. This can be attributed to the improvement of structural stability caused by the increase of particle size with the increase of irradiation time. Then, sensing tests were discussed only for the sensors of SnO2 -15 series. Sensing experiments were carried in the temperature range between 25 and 100 ◦ C. All fabricated SnO2 sensors resulted sensitive to ethanol at these low temperatures. Fig. 6 reports as an example the responses of the SnO2 -15-120 sensor at 250 ppm of ethanol at different temperatures, in the range

Fig. 4. TEM ((a) and (c)) and HRTEM ((b) and (d)) images of microwave irradiated samples.

100

N. Rajesh et al. / Sensors and Actuators B 194 (2014) 96–104

Fig. 5. XPS analysis of SnO2 -15-120 and SnO2 -15-400 samples. (a) Sn3d spectrum; (b) O 1s spectrum.

45–100 ◦ C. Increasing the operating temperature from 45 to 100 ◦ C a slight increase in the response, from 16 to 25, was noted, as well as a faster response/recovery time. The response of the SnO2 -15-120 sensor to different ethanol concentrations and the related calibration curve, at the operating temperature of 100 ◦ C, are shown in Fig. 7a and b, respectively. The response is stable also for repeated cycles (see Fig. 7c). Surprisingly, it has noted that the resistance of the SnO2 -15-120 sensor increases after exposure to ethanol. The same behavior has been observed with the other sensors investigated. Metal oxides are classified as n-type or p-type according to whether they show a decrease or an increase of resistance after exposure to reducing gases (such as ethanol or CO). Then, on the basis of this classification, SnO2 sensors investigated appears to exhibit a p-type behavior. This is not understandable because pure tin oxide is known to behave as a n-type semiconductor and a modification of the surface or doping is necessary to provide a p-type behavior. Matsubara et al. showed that the SnO2 surface modification by organic components with hydroxyl groups could lead the formation of hydrogen bonding between them, increasing the resistivity of the sensing layer when exposed to reducing gases such CO

Response ( R/R) 100

50

40

30

20

10

0 0

500

1000

1500

2000

2500

3000

Time (s) Fig. 6. Responses of the SnO2 -15 sensor at 250 ppm of ethanol at different operating temperatures.

[40]. Sb-SnO2 nanowires exhibited a typical p-type behavior and detected ethanol species at room temperature. The conductance of their sensor rapidly decreased as the device was exposed to ethanol gas [41]. Abnormal gas sensing characteristics were also observed at low temperature with Pt–SnO2 nanorod gas sensors and ascribed to Pt-catalyzed morphological changes of ionsorbed oxygen at low temperature [42]. The sensors operated at 200 ◦ C exhibited opposite variations of resistances, and the change of resistance decreased with increasing ethanol concentration. In contrast, the sensors operated at 300 ◦ C showed regular behavior. Such behavior was ascribed to Pt-catalyzed morphological changes of ionsorbed oxygen at low temperature. The behavior of pure SnO2 in respect to ethanol detection has been largely studied [43–45]. It is well-known that chemisorbed oxygen at SnO2 surfaces plays an important role. As a result of chemisorption there can appear O2 − , O2− , and O− ions depending on the temperature [46]. These adsorbed oxygen species create a space charge region near the film surface by extracting electrons from the material. Ethanol, being reducing in nature, removes adsorbed oxygen species from the surface and re-injects the electrons back to the material, thereby reducing the resistance. Much of these studies are however devoted to the understanding the mechanism at high temperature (150–400 ◦ C) where the O− species were reported to be highly active and dominant [47]. On the other hand, it is well known that the behavior of resistive sensors at room temperature differ quite significantly from that at high temperature. Kim et al. [48] reported such “abnormal” behavior for a tungsten oxide nanorod sensor when the operating temperature was below 70 ◦ C. While the sensing responses for reducing vapor (including NH3 and ethanol) showed an increase in electrical conductivity at temperatures above 150 ◦ C as expected for n-type metal oxide sensors, they exhibited the opposite behavior of unusual conductivity decrease below 100 ◦ C. The increase in resistance under ambient conditions has been attributed to the competitive adsorption between ambient molecular oxygen and analyte vapors on the surface of the active layer. In a very recent paper, it also reported that a sensor based on V2 O5 hierarchical structure networks behave to ethanol in a similar manner. However, no data are reported for CO sensing [49]. The sensor showed temperature-dependent p- to n-type response characteristic reversal, resulting in dual working temperature

N. Rajesh et al. / Sensors and Actuators B 194 (2014) 96–104

101

300

250 60

Resistance (k )

Resistance (k )

70

50

40

200

150

100 30 3000

4000

5000

50 2000

2500

3000

3500

4000

Time (s) Fig. 8. Response of the SnO2 -15 sensor exposed to 50 ppm CO at 100 ◦ C.

Response

between neighboring grains. CO is oxidized on the surface with consuming the oxygen adsorbates. During this reaction, the electrons back into the conduction band, resulting in a decrease in the depth of the space-charge region and a drop in resistance. Then, differently by the other cases above presented, the anomalous behavior we observed appears to be specific for ethanol. At present we do not know the precise origin of the increased resistance with ethanol. On the other hand, it is well known that the sensing properties span a very broad range, depending on the processing route, the thermal history, morphology of the material, and so on. In addition, the achievement of gas-sensing properties is not only a function of the nature and microstructure of metal oxide grains but also of suitable surface-reception properties [50]. In our case, it is obvious that a different detection mechanism from that described above for CO is needed to account for the anomaly observed with ethanol. Xu et al. suggested, as a possible explanation for the abnormal response of ethanol on ZnO sensor at room temperature that oxygen molecules are limited to bond with the localized electrons at the oxygen vacancy sites under low (room) temperature [51]. Thus, when an ethanol molecule adsorbs on the surface it interacts directly with localized electrons at the oxygen vacancy sites resulting in the deactivation of the donor site, based on the following equation:

0 0

100

200

300

Ethanol (ppm)

Resistance (k )

250

200

150

C2 H5 OH + e− → C2 H5 OH−

100 0

10

20

30

40

50

60

Time (min) Fig. 7. (a) Response of the SnO2 -15 sensor exposed to different concentrations of ethanol at 100 ◦ C; (b) calibration curve; (c) response reproducibility.

characteristic with the dual response extremes reached at room temperature (20 ◦ C) and 250 ◦ C respectively. In order to better understand the behavior of our SnO2 samples, we also tested CO, another reducing gas (Fig. 8). The response to CO in air at the temperature of 100 ◦ C was observed with reduction of electrical resistance. Unlike what was observed in the presence of ethanol, sensing tests carried out with CO follow the n-type semiconductor behavior typical for tin oxide. In this case, the sensing mechanism relies on oxygen adsorbed on the surface (in the range of temperature here considered, the predominant oxygen species on the sensor surface is O2 − ) which removes electrons from SnO2 conduction band at elevated temperatures, realizing an electron depleted surface region (space-charge region). The space-charge region works as a potential barrier

(2)

The mechanism described abstracts electrons from the n-type semiconductor, and hence increases the sensor resistance. However, they do not formulate any hypothesis on how this can occurs and neither if this is restricted to ethanol or to other reducing gases. Morante and coworkers reported that CO responded with an increase of the resistance at low temperature [52]. The rise of the resistance of the sensor upon exposure to CO observed at temperatures as low as 100 ◦ C might be attributed to the reaction of the gas molecules with the ionosorbed oxygen according to the following possible reaction: − COgas + O− 2 + e

→

CO− + O− 2ads ads

(3)

The same mechanisms apply for CH4 , another reducing gas, but no data were reported for ethanol. They related this behavior to the available oxygen concentration at the surface and also to their typology. On the basis of these indications and the support of our experimental data, hydroxyl group present in the ethanol molecule, which may behaves as hydrogen-bond donors or proton donors (Brønsted acid), represents the key factor. We suppose that, at low

102

N. Rajesh et al. / Sensors and Actuators B 194 (2014) 96–104

Scheme 1. Schematic view of the interaction of ethanol and CO with the surface of SnO2 sensing layer at low temperature in air. The dashed arrows indicated the direction of charge transfer.

10000 500 ppm ethanol

SnO2-15-400 in

Resistance ( )

temperature, the direct interaction may take place via hydrogen bonding between the ethanol hydroxyl group and the sensing layer surface, as illustrated in Scheme 1. As deduced by XPS data here reported, at low temperature, the surface of SnO2 is mainly covered by hydroxyl groups (here denoted (Sn␦+Sn −OH␦− ) as in Ref. [53]) and adsorbed O2 − oxygen. The paramagnetic superoxide ion is assumed to be oriented parallel to the surface (side-on) and stabilized in the crystal field of the tin cations [54]. In the presence of ethanol, hydrogen bonding above described favors the direct interaction with the free oxygen vacancy (mechanism 1). In this case, the charge transfer is directed away the bulk and the resistance increases. Mechanism 1 appears to be specific for ethanol, while with other reducing gases lacking of the hydroxyl group the typical mechanism leading to an opposite charge transfer directed towards the bulk, i.e. through the surface reaction of the target gas (here exemplified by CO) with adsorbed oxygen which diminished its concentration on the surface (mechanism 2), is dominant. However, there are still a number of specific questions to be investigated and clarified. For example, it cannot be excluded a direct participation of (Sn␦+Sn −OH␦− ) groups in the sensing mechanism. On the other hand, sensing materials for gas sensors are usually heat-treated at high temperatures for stabilizing them. It is well known that after annealing treatment atomic arrangements as well as oxygen vacancies are quite different due to formation of lattice imperfections of different kinds. Therefore, we have performed gas sensor measurements after an annealing treatment at 400 ◦ C (Fig. 9). After the thermal treatment, the abnormal behavior towards ethanol disappeared in all range of operating temperature investigated (25–400 ◦ C), indicating that the surface reconstruction occurring during the thermal treatment (causing a density variation of the oxygen vacancy and leading to a diminution of (Sn␦+Sn −OH␦− ) groups) is likely the cause of the behavior observed. At last, from a practical point of view, due to the easy scale up, this method of SnO2 synthesis could provide a practical route for the fabrication of low-cost high performance ethanol sensors. Furthermore, its peculiar behavior can exploit for example to discriminate between ethanol and other reducing gases such as CO, which has important implications especially in food and fuel industries [55]. Another important advantage of our sensor is the large electrical conductivity at low temperature eliminating the need of high thermal activation of SnO2 particles thus enabling low temperature sensing.

1000

100°C

out

100

in 400°C

10

out

1 0

200

400

600

800

1000

1200

Time (s) Fig. 9. Response of the SnO2 -15-400 sensor exposed to 500 ppm ethanol at the operating temperature of 100 and 400 ◦ C.

4. Conclusion In summary, the synthesis of tin oxide nanoparticles in just 5–15 min has been reported by microwave irradiation without any post-synthesis annealing procedure. The resistive sensors fabricated by using these SnO2 nanoparticles as sensing layer have shown a peculiar sensing behavior in the presence of ethanol, i.e. an increase of resistance with ethanol concentration. Such unusual behavior has not instead observed for CO as target gas. On the basis of XPS measurements, this can be attributed to the temperature-dependent density variation of the oxygen vacancy. So, a sensing mechanism involving a direct interaction between the surface of SnO2 and ethanol through the hydroxyl group has been proposed to explain these findings. Further studies are planned in order to investigate the influence of water and the sensing behavior with other alcohols (e.g. methanol).

Acknowledgment The authors wish to thank Dr. Francesco Barreca (Dept. of Earth Science and Physics, Univ. of Messina) for technical collaboration in performing the XPS measurements and helpful discussion.

N. Rajesh et al. / Sensors and Actuators B 194 (2014) 96–104

References [1] E. Comini, Metal oxide nano-crystals for gas sensing, Anal. Chim. Acta 568 (2006) 28–40. [2] S. Goldsmith, E. Cetinorgu, R.L. Boxman, Modeling the optical properties of tin oxide thin films, Thin Solid Films 517 (2009) 5146–5150. [3] R.K. Gupta, K. Ghosh, R. Patel, S.R. Mishra, P.K. Kahol, Highly conducting and transparent tin-doped CdO thin films for optoelectronic applications, Mater. Lett. 62 (2008) 4103–4105. [4] V. Senthilkumar, P. Vickraman, M. Jayachandran, M. Sanjeeviraja, Structural and optical properties of indium tin oxide (ITO) thin films with different compositions prepared by electron beam evaporation, Vacuum 84 (2010) 864–869. [5] E. Benamar, M. Rami, C. Messaoudi, D. Sayah, A. Ennaoui, Structural, optical and electrical properties of indium tin oxide thin films prepared by spray pyrolysis, Sol. Energy Mater. Sol. Cells 56 (1999) 125–139. [6] L.M. Cukrov, T. Tsuzuki, P.G. McCormickG., SnO2 nanoparticles prepared by mechanochemical processing, Scr. Mater. 44 (2001) 1787–1790. [7] S. Kato, H. Unuma, T. Ota, M. Takahashi, Homogeneous precipitation of hydrous tin oxide powders at room temperature using enzymatically induced gluconic acid as a precipitant, J. Am. Ceram. Soc. 83 (2000) 986–988. [8] L.H. Xian, Y.-J. Zhu, S.-W. Wang, Sonochemical and microwave-assisted synthesis of linked single-crystalline ZnO rods, Mater. Chem. Phys. 88 (2004) 421–426. [9] N.S. Baick, G. Sakai, N. Miura, N. Yamozoe, Hydrothermally treated sol solution of tin oxide for thin-film gas sensor, Sens. Actuators, B 63 (2000) 74–79. [10] H. Zhaohui, G. Neng, L. Fanqing, Z. Wanqun, Z. Huaquiao, Q. Yitai, Solvothermal preparation and morphological evolution of stannous oxide powders, Mater. Lett. 48 (2001) 99–103. [11] K.C. Song, J.H. Kim, Synthesis of high surface area tin oxide powders via waterin-oil microemulsions, Powder Technol. 107 (2000) 268–272. [12] C.H. Shek, L. Lai, G.M. Lin, Grain growth in nanocrystalline SnO2 prepared by sol–gel route, Nanostruct. Mater. 11 (1999) 887–893. [13] D. Briand, M. Labeau, J.F. Curie, G. Delabouglise, Pd-doped SnO2 thin films deposited by assisted ultrasonic spraying CVD for gas sensing: selectivity and effect of annealing, Sens. Actuators, B 48 (1988) 395–402. [14] J.Y. Kim, I.J. Chung, J.K. Kim, J.-W. Yu, Solid-state photovoltaic devices based on perylene acid-sensitized nanostructural SnO2 with P(VdF-co-HFP) gel electrolyte, Curr. Appl. Phys. 6 (2006) 969–973. [15] N. Pinna, G. Garnweitner, M. Antonietti, M. Niederberger, Non-aqueous synthesis of high-purity metal oxide nanopowders using an ether elimination process, Adv. Mater. 16 (2004) 2196–2200. [16] T. Krishnakumar, R. Jayaprakash, V.N. Singh, B.R. Mehta, A.R. Phani, Synthesis and characterization of tin oxide nanoparticle for humidity sensor applications, J. Nanopart. Res. 4 (2008) 91–101. [17] A.R. Babar, S.S. Shinde, A.V. Moholkar, K.Y. Rajpure, Electrical and dielectric properties of co-precipitated nanocrystalline tin oxide, J. Alloys Compd. 505 (2010) 743–749. [18] P.S. Patil, R.K. Kawar, T. Seth, D.P. Amalnerkar, P.S. Chigare, Effect of substrate temperature on structural, electrical and optical properties of sprayed tin oxide (SnO2 ) thin films, Ceram. Int. 29 (2003) 725–734. [19] T. Krishnakumar, R. Jayaprakash, M. Parthibavarman, A.R. Phani, V.N. Singh, B.R. Mehta, Microwave-assisted synthesis and investigation of SnO2 nanoparticles, Mater. Lett. 63 (2009) 896–898. [20] A.A. Firooz, A.R. Mahjoub, A.A. Khodadadi, Preparation of SnO2 nanoparticles and nanorods by using a hydrothermal method at low temperature, Mater. Lett. 62 (2008) 1789–1792. [21] Y. He, Y. Li, J. Yu, Y. Qian, Chemical control synthesis of nanocrystalline SnO2 by hydrothermal reaction, Mater. Lett. 40 (1999) 23–26. [22] J. Kaur, S.C. Roy, M.C. Bhatnagar, Highly sensitive SnO2 thin film NO2 gas sensor operating at low temperature, Sens. Actuators, B 123 (2007) 1090–1095. [23] G. Neri, A. Bonavita, G. Micali, F. Deorsola, P. Mossino, I. Amato, et al., Ethanol sensors based on Pt-doped tin oxide nanopowders synthesised by gel-combustion, Sens. Actuators, B 117 (2006) 196–204. [24] H.-C. Chiu, C.-S. Yeh, Hydrothermal synthesis of SnO2 nanoparticles and their gas-sensing of alcohol, J. Phys. Chem. C 111 (2007) 7256–7259. [25] X. Zhou, W. Fu, H. Yang, D. Ma, J. Cao, Y. Leng, et al., Synthesis and ethanolsensing properties of flowerlike SnO2 nanorods bundles by poly(ethylene glycol)-assisted hydrothermal process, Mater. Chem. Phys. 124 (2010) 614–618. [26] Y. Wang, X. Jiang, Y. Xia, A solution-phase, precursor route to polycrystalline SnO2 nanowires that can be used for gas sensing under ambient conditions, J. Am. Chem. Soc. 125 (2003) 16176–16177. [27] A.K. Singh, U.T. Nakate, Microwave synthesis, characterization and photocatalytic properties of SnO2 nanoparticles, Adv. Nanopart. 2 (2013) 66–70. [28] A. Azam, S.S. Habib, N.A. Salah, F. Ahmed, Microwave-assisted synthesis of SnO2 nanorods for oxygen gas sensing at room temperature, Int. J. Nanomed. 8 (2013) 3875–3882. [29] P.G. Mendes, M.L. Moreira, S.M. Tebcherani, M.O. Orlandi, J. Andre’s, M.S. Li, N. Diaz-Mora, J.A. Varela, E. Longo, SnO2 nanocrystals synthesized by microwaveassisted hydrothermal method: towards a relationship between structural and optical properties, J. Nanopart. Res. 14 (2012) 750. [30] J. Jouhannaud, J. Rossignol, D. Stuerga, Rapid synthesis of tin(IV) oxide nanoparticles by microwave induced thermohydrolysis, J. Solid State Chem. 181 (2008) 1439–1444. [31] M. Krishna, S. Komarneni, Conventional vs. microwave hydrothermal synthesis of tin oxide, SnO2 nanoparticles, Ceram. Int. 35 (2009) 3375–3379.

103

[32] G. Xi, Y. He, Q. Zhang, H. Xiao, X. Wang, C. Wang, Synthesis of crystalline microporous SnO2 via a surfactant-assisted microwave heating method: a general and rapid method for the synthesis of metal oxide nanostructures, J. Phys. Chem. C 112 (2008) 11645–11649. [33] M. Parthibavarman, V. Hariharan, C. Sekar, High-sensitivity humidity sensor based on SnO2 nanoparticles synthesized by microwave irradiation method, Mater. Sci. Eng., C 31 (2011) 840–844. [34] R. Karslioglu, M. Uysal, H. Akbulut, The effect of substrate temperature on the electrical and optic properties of nanocrystalline tin oxide coatings produced by APCVD, J. Cryst. Growth 327 (2011) 22–26. [35] N.G. Deshpande, Y.G. Gudage, Ramphal Sharma, J.C. Vyas, J.B. Kim, Y.P. Lee, Studies on tin oxide-intercalated polyaniline nanocomposite for ammonia gas sensing applications, Sens. Actuators, B 138 (2009) 76–84. [36] H.H. Son, W-.G. Lee, Annealing effects for calcination of tin oxide powder prepared via homogeneous precipitation, J. Ind. Eng. Chem. 18 (2012) 317–320. [37] H.J. Ahn, H.C. Choi, K.W. Park, S.B. Kim, Y.E. Sung, Investigation of the structural and electrochemical properties of size-controlled SnO2 nanoparticles, J. Phys. Chem. B 108 (2004) 9815–9820. [38] M. Batzill, U. Diebold, The surface and materials science of tin oxide, Prog. Surf. Sci. 79 (2005) 47–154. [39] T.J. Chuang, C.R. Brundle, D.W. Rice, Interpretation of X-ray photoemission spectra of cobalt oxides and cobalt oxide surfaces, Surf. Sci. 59 (1976) 413–429. [40] I. Matsubara, K. Hosono, N. Murayama, W. Shin, N. Izu, Organically hybridized SnO2 gas sensors, Sens. Actuators, B 108 (2005) 143–147. [41] J.M. Wu, A room temperature ethanol sensor made from p-type Sb-doped SnO2 nanowires, Nanotechnology 21 (2010) 235501. [42] X.-Y. Xue, Z.-H. Chen, Y.-J. Chen, C.-H. Ma, L.-L. Xing, Y.-G. Wang, T.-H. Wang, Abnormal gas sensing characteristics arising from catalyzed morphological changes of ionsorbed oxygen, Nanotechnology 21 (2010) 065501. [43] J. Zhang, S. Wang, Y. Wang, M. Xu, H. Xia, S. Zhang, et al., Facile synthesis of highly ethanol-sensitive SnO2 nanoparticles, Sens. Actuators, B 139 (2009) 369–374. [44] O.K. Varghese, L.K. Malhotra, G.L. Sharma, High ethanol sensitivity in sol–gel derived SnO2 thin films, Sens. Actuators, B 55 (1999) 161–165. [45] L.A. Patil, M.D. Shinde, A.R. Bari, V.V. Deo, Highly sensitive ethanol sensors based on nanocrystalline SnO2 thin films, Curr. Appl. Phys. 10 (2010) 1249–1254. [46] M.J. Madou, S. Roy Morrison, Chemical Sensing with Solid State Devices, Academic Press, New York, NY, 1989. [47] N. Yamazoe, J. Fuchigami, M. Kishikawa, T. Seiyama, Interactions of tin oxide surface with O2 , H2 O and H2 , Surf. Sci. 86 (1979) 335–344. [48] Y.S. Kim, S.-C. Ha, K. Kim, H. Yang, S.-Y. Choi, Y.T. Kim, Room-temperature semiconductor gas sensor based on nonstoichiometric tungsten oxide nanorod film, Appl. Phys. Lett. 86 (2005) 213105–213107. [49] Y. Qin, G. Fan, K. Liu, M. Hu, Vanadium pentoxide hierarchical structure networks for high performance ethanol gas sensor with dual working temperature characteristic, Sens. Actuators, B 190 (2014) 141–148. [50] M. Epifani, J.D. Prades, E. Comini, E. Pellicer, M. Avella, P. Siciliano, et al., The role of surface oxygen vacancies in the NO2 sensing properties of SnO2 nanocrystals, J. Phys. Chem. C 112 (2008) 19540–19546. [51] C.H. Xu, H.F. Lui, C. Surya, Optical and sensor properties of ZnO nanostructure grown by thermal oxidation in dry or wet nitrogen, J. Electroceram. 28 (2012) 27–33. [52] R. Ferro, J.A. Rodríguez, I. Jiménez, A. Cirera, J. Cerdà, J.R. Morante, Gas-sensing properties of sprayed films of (CdO)x(ZnO) mixed oxide, IEEE Sens. J. 5 (2005) 1. [53] N. Barsan, U. Weimar, Understanding the fundamental principles of metal oxide based gas sensors; the example of CO sensing with SnO2 sensors in the presence of humidity, J. Phys. Condens. Matter 15 (2003) R813–R839. [54] A. Gurlo, Interplay between O2 and SnO2 : oxygen ionosorption and spectroscopic evidence for adsorbed oxygen, Chem. Phys. Chem. 7 (2006) 2041–2052. [55] I. Aruna, F.E. Kruis, S. Kundu, M. Muhler, R. Theissmann, M. Spasova, CO ppb sensors based on monodispersed SnOx:Pd mixed nanoparticle layers: insight into dual conductance response, J. Appl. Phys. 105 (2009) 064312.

Biographies N. Rajesh received Master of Philosophy in physics from Bharathidasan University in 2007. He is Assistant Professor in Physics at KSR College of Engineering, Tiruchengode, Tamilnadu, India and he is currently a Ph.D. student of Physics at Anna University, Chennai, India. His research activity concerns with the preparation, characterization and development of metal oxide nanostructures for gas sensing applications. J.C. Kannan received Ph.D. degree in Physics from Bharathiar University in 2009. He has been Head of the Department of Physics in KSR Institute for Engineering and Technology, Tiruchengode, Tamilnadu, India. His research activity concerns with the preparation, characterization and development of metal oxide nanostructures for gas sensing applications. T. Krishnakumar received Ph.D. degree in Physics from Bharathiar University in 2009. From 2009 to 2011 he was Associate professor in Vinayaka Missions University. From 2012, He has been Head of the Department of Physics in Tagore Institute of Engineering and Technology, Attur, Tamilnadu, India. His research activity concerns with the preparation, characterization and development of metal oxide nanostructures for gas sensing applications.

104

N. Rajesh et al. / Sensors and Actuators B 194 (2014) 96–104

Gianluca Leonardi received the degree in Materials Engineering from the University of Messina in 2011. He is currently a Ph.D. student of Engineering and Chemistry of Materials at University of Messina. His research activity is focused on the development of chemical and biochemical sensors based on nanostructured materials. Giovanni Neri received the degree in Chemistry in 1980. From 1987 to 1998 he was a researcher at the University of Reggio Calabria. From 1991 to 1996 he spent

several periods of research at the University of Michigan (USA). In 1998, he moved to the University of Messina as an Associate Professor. From 2001 he is Full Professor of Chemistry. From 2004 to 2007 he was Director of the Department of Industrial Chemistry and Materials Engineering of the University of Messina. His research activity covers many aspects of the synthesis and characterization of materials and studies of their catalytic and sensing properties. In the latter research area his work has been focused on the application in gas sensors of nanostructured metal oxides and novel organic–inorganic hybrid nanocomposites.