Journal Search Engine
Search Advanced Search Adode Reader(link)
Download PDF Export Citaion korean bibliography PMC previewer
ISSN : 1225-0562(Print)
ISSN : 2287-7258(Online)
Korean Journal of Materials Research Vol.26 No.12 pp.741-750
DOI : https://doi.org/10.3740/MRSK.2016.26.12.741

Excellent Carbon Monoxide Sensing Performance of Au-Decorated SnO2 Nanofibers

Jae-Hun Kim1, Yifang Zheng1, Mirzaei Ali2, Sang Sub Kim1
1Department of Materials Science and Engineering, Inha University, Incheon 22212, Republic of Korea
2The Research Institute of Industrial Science, Hanyang University, Seoul 04763, Republic of Korea
Corresponding author sangsub@inha.ac.kr (S. S. Kim, Inha Univ.)
October 1, 2016 November 11, 2016 November 14, 2016

Abstract

Nanofibers(NFs), because of their high surface area and nanosized grains, have appropriate morphologies for use in chemiresistive-type sensors for gas detection applications. In this study, a highly sensitive and selective CO gas sensing material based on Au-decorated SnO2 NFs was fabricated by electrospinning. SnO2 NFs were synthesized by electrospinning and subsequently decorated with various amounts of Au nanoparticles(NPs) by sputtering; this was followed by thermal annealing. Different characterizations showed the successful formation of Au-decorated SnO2 NFs. Gas sensing tests were performed on the fabricated sensors, which showed bell-shaped sensing behavior with respect to the amount of Au decoration. The best CO sensing performance, with a response of ~20 for 10 ppm CO, was obtained at an optimized amount of Au (2.6 at.%). The interplay between Au and SnO2 in terms of the electronic and chemical sensitization by Au NPs is responsible for the great improvement in the CO sensing capability of pure SnO2 NFs, suggesting that Au-decorated SnO2 NFs can be a promising material for fabricating highly sensitive and selective chemiresistive-type CO gas sensors.


초록


    National Research Foundation

    Ministry of Science, ICT and Future Planning
    No.2016M2B2A4911989)

    © Materials Research Society of Korea. All rights reserved.

    This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

    1.Introduction

    Carbon monoxide(CO), which is notorious as a silent killer, is a colorless, tasteless, and odorless gas.1) The toxic effect of CO due to its strong affinity for hemoglobin, which is nearly 245 times that of oxygen, is well known.2) CO is the leading cause of poisoning in the U.S., and it accounts for more than 50 % of the fatal poisonings reported in many industrialized countries.3) Furthermore, CO has been considered as a marker of oxidative stress and is thought to be involved in the development of various diseases.4) In addition, CO detection can be used in early-fire-monitoring systems because the CO gas is generated by the incomplete combustion of hydrocarbons, and the presence of CO is an indication of fire.5) In U.S., the threshold limit value(TLV) of CO is set at 50 ppm by the Occupational Safety and Health Administration(OSHA). However, when the concentration of CO is higher than 15 ppm it becomes dangerous for humans, and in Europe, the limit of CO(maximum 8 h daily) is set at 8.75 ppm.6-8) This has motivated researchers to develop simple and inexpensive sensors for the detection of low concentrations of CO gas; currently, considerable effort is being devoted for developing sensing materials with high sensitivity and selectivity toward CO gas.

    SnO2, which has the O=Sn=O structure, is an important n-type semiconductor with a wide band gap of 3.6 eV.9) It has been recognized as a promising semiconductor material for many applications such as lithium-ion batteries,10) electrocatalysts,11) supercapacitors,12) and solar cells.13) In particular, SnO2 is widely used for gas sensing applications because of its low preparation cost, high electron mobility, and thermal stability under sensor operating conditions.14) Until now, SnO2 has been extensively used for sensing toxic gases such as NO2,15) CO,16) H2,17) H2S,18) and CH419) as well as volatile organic compounds (VOCs)20) including formaldehyde,21) ethanol,22) methanol,23) and toluene.24) However, like most of the metal oxide gas sensors, there is a lack of gas selectivity in SnO2, so enhancing its gas performance toward a specific target gas still remains a challenge. To overcome this challenge, some strategies such as incorporation of dopants;25) sensitization using noble metals such as Au,26) Ag,27) and Pd;28) formation of composites with other metal oxides such as CuO29) as well as the use of ultraviolet irradiation30) have been proposed to improve the sensing performance of SnO2 sensors. Among these, decoration with noble metals is regarded as one of the best and most effective strategies for improving gas sensitivity and selectivity.31) In particular, Au has been reported to be the most useful catalyst for the detection of several reductive gases such as CO32) and VOCs.33)

    In recent years, significant effort has been devoted to enhance the gas sensing properties by utilizing onedimensional( 1D) metal oxide nanostructure sensors, such as nanowires,34) nanobelts,35) nanotubes,36) nanorods,37) and nanofibers(NFs).38) 1D nanostructures have advantages such as higher sensitivity, superior spatial resolution, and rapid response because of the high surface-to-volume ratio compared to thin film gas sensors.39) In particular, the extraordinary sensing properties of NFs, in addition to their large surface to volume ratio compared to other nanostructures, are because of the fact that they contain many small grains, and consequently, they possess large grain boundary areas. Therefore, large amounts of analytes can easily diffuse along the boundaries, resulting in enhanced sensing characteristics.40)

    Another advantage of NFs is their facile synthesis by the electrospinning method. Many synthesis procedures for NFs such as electron-beam or focused ion-beam writing, lithography, hydrothermal, chemical vapor deposition, and template-directed methods have been reported over the last few decades. However, most of these methods have limitations such as material restrictions, high cost, and high process complexity.41) Therefore, electrospun NFs are preferred in gas sensor applications because of the easy control over processing parameters and characteristics like morphology, diameter, and aspect ratio so that NFs with different architectures, for example, simple NFs, core-shell NFs,42) porous NFs,43) and hollow NFs,44) can be produced by the electrospinning method. In addition, the electrospinning method is incredibly effective for low-cost mass production with minimal usage of materials, which makes it the most suitable method for industrial applications on a commercial scale. Hence, electrospun NFs have been employed in a diverse range of sensing materials.40)

    Accordingly, the use of Au decoration on SnO2 NFs has a favorable synergistic effect on the detection of a specific gas. However, the sensing properties of the SnO2 NFs decorated with Au nanoparticles(NPs) are likely to be greatly dependent on their amount. Usually, an optimized amount of metal NPs provides the maximum sensing performance, showing a bell-shaped curve with respect to the amount of decoration.45) However, many researchers simply compare the gas sensing properties of pristine and metal NP-loaded sensing materials without optimization of the metal NPs.46) Hence, systematic studies on this will facilitate the use of metal NP-loaded NFs in sensing applications.

    In this study, SnO2 NFs synthesized by electrospinning were decorated with Au NPs by depositing Au layers onto the SnO2 NFs by sputtering and subsequent thermal annealing. The amount of Au NPs was controlled by altering the thicknesses of the deposited Au layers. Various characterization techniques were utilized to examine the morphology of the Au NP-loaded SnO2 NFs. Comparison of the gas sensing properties of pristine and Au-decorated SnO2 NFs to different concentrations of CO, at various Au concentrations, was also conducted. A high sensor response of ~8 for 1 ppm CO was obtained at an optimized amount of Au NPs(2.6 at.%) at 300 °C, which is a remarkable result in terms of CO detection. Regardless of the significant developments in CO sensors, there has been little work on exceptionally high sensitivity to CO gas at low concentrations of around 1 ppm. In this respect, Au-decorated SnO2 NFs investigated in this study have the potential for use in CO sensing platform materials, especially with respect to the morphology and incorporation of catalytic elements.

    2.Experimental Details

    An electrospinning method was used to synthesize the SnO2 NFs. Tin(II) chloride dihydrate (SnCl2·2H2O, Sigma- Aldrich) was used as the precursor material. First, a mixed solvent comprising 9.42 g of anhydrous N,Ndimethylformamide (DMF, Sigma-Aldrich) and 11.7 g of ethanol (J. T. Baker) was prepared. Then, 1.54 g of polyvinyl acetate(PVAc) with a molecular weight of 850,000 and 1 g of SnCl2 were dissolved in the mixed solvent by stirring for 4 h at room temperature. These prepared viscous solutions were loaded into a syringe with a 21- gauge stainless steel needle and an inner diameter of 0.51 mm. The distance between the tip of the needle and the collector was fixed at 20 cm. A positive voltage of 15 kV was applied to the needle and the collector was simultaneously grounded. The feeding rate of the solution was adjusted to 0.03 mL/h by using an accurate syringe pump. All the electrospinning experiments were performed at room temperature under air ambient. After 15 min of electrospinning, the as-spun fibers were distributed uniformly over the SiO2-grown Si wafers that were placed on the collector. The as-spun fibers were subsequently calcined at 650 °C for 2 h in air with a heating rate of 5 °C/min.

    Afterward, the substrate on which SnO2 NFs were distributed was transferred to a turbo sputter coater (Emitech K575X, Emitech Ltd., Ashford, Kent, UK). The target was a circular Au, the sputtering current was 65 mA, and the sputtering was performed in high-purity(99.999 %) Ar ambient at room temperature. Au thin films with different thicknesses(3, ~5, ~10, 12, 15, and 20 nm) were successfully deposited by changing the sputtering time from 0 to 2 min. Subsequently, the Au-deposited SnO2 NFs were annealed for 30 min at 500 °C in air ambient. This thermal treatment promotes the formation of Au NPs, improving homogeneity of their distribution by layer thickness, and stabilizing properties of gas sensing materials.47)

    The microstructure and morphology of the Au NPdecorated SnO2 NFs were characterized by field-emission scanning electron microscopy(FE-SEM) and transmission electron microscopy(TEM). The compositional analysis was performed using energy dispersive X-ray spectroscopy (EDS).

    For the sensing measurements, Ti(~50 nm thick) and Au (~200 nm thick) were sequentially deposited via sputtering on the Au NP-decorated SnO2 NFs using an interdigital electrode mask. Their sensing capability to CO was investigated at different temperatures using a gas dilution and sensing system. The CO gas concentration in the sensing measurement chamber was controlled by changing the mixing ratio of the target gas CO to dry air through accurate mass flow controllers(MFCs). CO diluted with dry air, supplied by the manufacturer (Daeduk Gas Co., Korea), was used as the CO source. With a 100 ppm CO source and using MFCs, the CO concentrations could be controlled to a very low amount. The total flow rate was set to 500 sccm to avoid any possible variation in the sensing properties. The water vapor content in the CO container was less than 1.5 ppm, indicating that the effect of humidity would be negligible. To evaluate the selective sensing properties, other typical reducing gases such as toluene(C7H8) and benzene(C6H6) were also tested. The response(R) was calculated as R = Ra/Rg, where Ra and Rg are the resistances in the absence and presence of the target gas, respectively.

    3.Results and Discussion

    3.1.Morphological studies

    The FE-SEM images of the SnO2 and Au NPdecorated SnO2 NFs with different Au concentrations are presented in Fig. 1(a-g). Fig. 1(h) is a low magnification FE-SEM image of SnO2 NFs with 2.6 at.% Au NPs, showing the overall features of the samples. It is obvious that the Au layers are broken into small clusters or islands after the thermal annealing process. It can also be seen that the surfaces of the SnO2 were significantly coarsened after loading of the Au NPs. Furthermore, the size of the Au NPs increases with increasing thickness of the sputtered Au layer. The average diameter of the SnO2 NFs can be estimated to be about 150-250 nm with varying sizes of Au NPs. Overall, the NPs were homogeneously distributed over the surfaces of the SnO2 NFs, as shown in Fig. 1(h) and Fig. 2.

    TEM analyses were performed to further examine the surface modification of the SnO2 NFs by addition of 2.6 at.% Au NPs. Fig. 3(a) shows the TEM image of a SnO2 NF, where the presence of Au NPs is clearly observed. The high-resolution TEM(HRTEM) image of Au NP-decorated SnO2 NFs is presented in Fig. 3(b). A closer examination of the HRTEM image revealed fringes in both the Au NPs and the SnO2 NFs. The measured dspacing between adjacent fringes(0.264 nm) corresponds to the d-spacing of the (101) plane of tetragonal SnO2 NFs, and the lattice fringes of Au with a d-spacing of 0.236 nm can be indexed to the cubic Au (101) lattice spacing. Fig. 3(c) shows the selected area electron diffraction pattern(SAED) of Au NP-decorated SnO2 NFs. The well-defined SAED pattern clearly shows diffraction spots corresponding to the (111), (200), (220), and (311) lattice planes of Au. EDS analysis was performed to confirm the presence of elemental Au, Sn, and O. Elemental analyses were performed using EDS, as shown in Figs. 3(d), 4(a), and 5. From a comparison of Fig. 1, Fig. 4(a), and Fig. 5, it was proven that the SnO2 NFs with different amounts of Au NPs were successfully synthesized. As shown in Fig. 4(b), the amount of Au clusters present after the heat treatment increases linearly with increasing thickness of the Au layers prepared by sputtering. This is because the thin layers tend to break easily and are then converted into small islands, whereas it is more difficult for the thicker layers to break under the same heat treatment conditions. Therefore, the thicker Au layers are likely to form larger islands or clusters. According to the FE-SEM images, the diameters of the Au NPs increase with increasing thickness of the Au layers. As reported previously,45) it is noteworthy that there is a reasonably linear relationship between the Au layer thickness and the Au particle size. Using such a relationship, the Au particle size can be easily controlled in the nanoscale by adjusting the thickness of the Au layers deposited initially. Therefore, for a specific particle size distribution, we can simply adjust the thickness of the Au layers deposited initially. For example, for very fine Au NPs, the initial thickness must be as small as possible, whereas for larger Au particles, the thickness should be increased accordingly.

    3.2.Gas sensing studies

    It is well-known that the operating temperature greatly influences the sensing characteristics of metal oxide sensors, because it greatly affects the electrical conductivity and electron mobility. Hence, the pristine sensors were exposed to 10 ppm CO gas at different operating temperatures. Fig. 6(a) shows the transient responses of Au-decorated SnO2 NFs to CO gas at 150 to 400 °C. Fig. 6(b) shows the corresponding plot, indicating variations of the response as a function of operating temperature. It can be seen that the response increases with temperature up to 300 °C, and then decreases with a further increase in the temperature. Upon exposure to oxygen at low temperatures, the inactive ionosorption of oxygen because of a few active adsorption sites at the SnO2 NFs surfaces leads to inactive oxidation of CO gas, resulting in poor response. Vacancies and kink sites at the surface of a material are preferential sites for the adsorption of impurity atoms or ions. At low temperatures, the surface contains lower concentrations of vacancies and kink sites, resulting in the inactive oxidation of CO gas and a poor response. As the operating temperature increases, the adsorption of oxygen on the sensor surface becomes active, and oxidation of CO becomes more active, resulting in an enhanced response to CO gas. In contrast, further increase in the temperature beyond the optimal temperature will result in the desorption of the species.1,48) Based on these results, all other sensing tests were performed at 300 °C.

    The Au loading level is one of the most important factors that greatly affect the sensing response. Therefore, the sensing performance and the effect of the amount of Au were evaluated at 300 °C using 1, 5, and 10 ppm concentrations of CO gas. Fig. 7(a) shows the typical resistance curves and responses of the SnO2 NFs to CO gas with varying amounts of Au NPs. It can be seen that for all of the sensors, the resistance decreases upon exposure to CO gas. It returns to its original value after stopping the CO supply. This is a typical n-type sensing behavior of SnO2-based sensors. Furthermore, all of the sensors show the reversible behavior and the resistances come back to their initial values after three successive cycles. The highest response was observed from the NFs with 2.6 at.% Au NPs, as shown in Fig. 7(b), where the responses are plotted as a function of CO concentration. Fig. 7(c) presents the responses of the Au NP-decorated SnO2 NFs to 10 ppm CO gas as a function of Au concentration. The response of the pristine SnO2 NFs has been included for comparison. It is evident that the decoration of Au NPs significantly enhanced the CO response of the SnO2 NFs. Moreover, Fig. 7(c) shows a bell-shaped curve as a function of the amount of Au NPs where there is an initial improvement in the response with smaller amounts of Au, and then there is a deterioration with larger amounts. For 1.2, 2.1, 2.6, 4.6, 5.0, and 19.0 at.% Au NP-decorated SnO2 NFs, the values of the responses are 5.25, 13.72, 18.98, 15.53, 11.60, and 3.18, respectively. Note that the response of pristine SnO2 NFs is as low as 1.52. This behavior can be explained as follows. When the amount of Au decoration is negligible, Au NPs cannot actively participate in the reactions and the response is low. However, with an increase in the amount of Au NPs, the response increases, which is because of the presence of sufficient Au NPs that contribute to the electronic and chemical sensitizations, which will be discussed in the sensing mechanism section. The drop in the sensor response with increasing Au loading level is because there are not enough exposed surfaces of SnO2 NFs to receive the dissociated oxygen, and less exposed surfaces of the SnO2 NFs also affected the functionalities of the sensors.38) Thus, it is important to control the amount of Au NPs on the surfaces of SnO2 NFs to an appropriate amount. In other words, at high Au loading, Au islands or NPs are partially or completely connected with each other; therefore, the sensor resistance is not governed by the gas sensing properties of SnO2 NFs, but determined by the insensitive conducting Au NPs. Thus, negligible gas responses in the 19.0 at.% Au NP-decorated SnO2 NFs sensor is attributed to the formation of conducting channels by the Au islands. These results are consistent with the literature data that show an enhancement and deterioration of the gas response induced by low and high loading concentrations of metal NPs, respectively.49-51)

    Selectivity is another important parameter of a gas sensor. A good gas sensor must selectively detect a specific gas when it is exposed to an environment that contains multiple gases. The selectivity tests of the 2.6 at.% Au-decorated SnO2 sensors were performed by comparing their response to CO against C6H6 and C6H7, the transient responses to 1, 5, and 10 ppm of the mentioned gases are shown in Fig. 8(a), and the responses are summarized in Fig. 8(b). It can be seen that the sensors show the highest response to CO gas. The responses to C6H6 and C6H7 were significantly lower than that for the CO gas. In Fig. 8(c), the responses of the Au-decorated SnO2 NFs are compared to that of pristine SnO2 NFs for a 10 ppm concentration of the reducing gases at 300 °C. The results clearly demonstrate the highly selective CO detection capability of Au NP-decorated SnO2 NFs. Similar selective and enhanced catalysis behaviors by Au NPs toward CO have been reported in other studies,52-54) which show that the Au NPs exhibit efficient catalytic activity for enhancing the diffusion and interaction with CO gas compared to other reducing gases.

    It is useful to understand why Au NPs preferentially sense CO, rather than C6H6 and C7H8. We know that a strong bonding between the surface and the gas molecules will decrease the catalytic efficiency of metal NPs, because the binding will be too strong to be broken and form the transition state. Weak bonding, on the other hand, also decreases the catalytic efficiency since the gas molecules may not stick to the surface for the subsequent reaction. With a moderate bonding strength, the molecules stick to the surface but are flexible enough to move around and form transition states. Therefore, in the first step, it is necessary to analyze the binding strength of CO, C6H6, and C6H7 on the Au NPs. According to our previous analysis,55) C6H6 interacts weakly with Au, with an adsorption energy of 0.64 eV. In addition, the Au catalytic activity for C7H8 was relatively low at high temperatures, because all the C6H6 and its dissociated biphenyl species desorb from the Au surfaces at > 127 °C. Furthermore, the reaction energy barrier for complete oxidation of CO in the presence of Au was 0.21 eV, which is a very low value. Good catalytic activity of Au for oxidation of CO was also reported by other researchers. 53) For example, Haruta et al. 56) have also reported that Au NPs are remarkably active to lowtemperature oxidation of CO when they are highly dispersed and deposited on reducible metal-oxide-semiconductors. Therefore, in this study, it is not unexpected to see that Au NPs show a higher response to CO compared to C6H6 and C7H8.

    The d-band theory55) is also suitable for explaining the bond formation and reactivity among various transition metals that exhibit interactions between the adsorbate valence states and the s- and d-states of a transition metal surface. A higher d-band center(the Fermi level as reference) corresponds to an increase in energy(relative to the Fermi level) and to less filling of the antibonding (d-σ)* state. This means that the metal-adsorbate system is more stable, and stronger binding between the metal and the adsorbate is expected. The increased filling of the unstable antibonding state will result in the destabilization of the metal-adsorbate interaction, and thus, a weaker bonding energy is expected. The d-band center of Au was relatively low (−4.32), and by plotting the response to CO, C6H6, and C7H8 gases as a function of the d-band center, it was demonstrated that the response to CO in the presence of Au is higher than responses to the other two gases.

    Until now, several studies have been reported on the sensing of CO gas, but there are few reports on the preparation of highly sensitive CO sensors operating at low concentrations below 10 ppm. Table 1 compares the CO sensing properties of some metal oxide gas sensors with our current sensor. It can be seen that the response of the optimized Au NP-decorated SnO2 NFs sensor is much higher than the responses of other gas sensors, suggesting the potential of SnO2 NFs decorated with optimized Au NPs in sensitive, selective, and reliable CO sensors.

    3.3.Sensing mechanism

    It is known that the adsorption of gas molecules on the metal oxide semiconductor surface can cause a considerable change in the electrical resistivity. Thus, a change in the surrounding gaseous environment easily leads to a change in the conduction(or the resistance) of the semiconductor. For pristine SnO2 NFs sensor, the gas sensing mechanism is similar to other semiconductor oxide sensors.57,58) It is generally accepted that in ambient air, oxygen molecules are adsorbed to form O2 below 100 °C, O− between 100 and 300 °C, or O2− above 300 °C by capturing the conductive electrons from the sensor surface.57) Indeed, oxygen molecules are adsorbed on the surface of the SnO2 NFs when the sensor is exposed to air and are ionized to O or O2− at high temperatures (> 100 °C) by capturing free electrons from the conduction band of SnO2. Thus, an electron depletion layer is formed by reducing the free charge carrier(electron) concentration, and the conduction will be limited to the conduction channel along the SnO2 NFs, as schematically shown in Fig. 9(a). At the same time, a potential barrier will be formed as a barrier to the flow of electrons through the SnO2 nanograins. Therefore, the sensing mechanism of the pristine SnO2 NFs is related to the modulation of resistance on the SnO2 NF surface, as well as at the SnO2 nanograin boundaries. When the CO molecules react with the adsorbed oxygen species, they release the trapped electrons back into the conduction band of the SnO2 NFs according to the following reaction (1):

    COads = 2OCO2(gas) + 2e
    (1)

    As a consequence, the concentration of free electrons increases, the potential barrier will be considerably decreased and the conduction channel will be increased, ultimately decreasing the resistance of the pristine SnO2 sensors. Accordingly, the sensor response to CO gas depends on the number of electrons captured by the oxygen molecules and the number of electrons released by the CO molecules reacting with the adsorbed oxygen molecules.

    The enhanced response of the Au-decorated SnO2 NFs sensor toward CO could be attributed to both electronic and chemical sensitization effects induced by the Au NPs. The loading of Au NPs on the surface of SnO2 NFs is expected to induce electronic sensitization, as shown in Fig. 10(a). The work functions of Au(5.1 eV)59) and SnO2 (4.9 eV)60) are different, and in order to equate to the Fermi levels, electrons will be transferred from SnO2 to Au. Thus, Au acts as an electron acceptor for SnO2. As a result of the charge transfer, a potential barrier will be generated at the heterointerfaces, along with the bending of bands and a Schottky barrier will formed at the interface of SnO2/Au(Fig. 10(b)), which increases the resistance of the Au-decorated SnO2 NF sensors. It is possible that the electrical current across the Au/SnO2 heterointerfaces modulates the resistance. As shown in Fig. 9(b), when the electrons are transferred to the Au NPs, the width of the conduction channel becomes narrower compared to the case of the pristine SnO2 NFs. Upon exposure to CO gas, the width of the depletion layer will be substantially suppressed(Figs. 10(b) and 10(c)) in such a manner that the change in the width of the conduction channel becomes significantly larger than the case of the pristine SnO2 NFs. Therefore, a higher modulation in resistance, and consequently, a higher response is observed.

    In the chemical sensitization effect due to the spillover effect from Au NPs on SnO2, more oxygen species will be generated to react with the gas molecules and the adsorbed oxygen can diffuse faster to the surface vacancies and capture electrons from the conduction band of SnO2. In fact, the amount of active oxygen ions(O or O2−) increases significantly with the presence of Au NPs because of its better oxygen dissociation properties than SnO2. Because of both the increase in the adsorbed oxygen and the rate of conversion of molecules to ions, the degree of electron depletion at the Au/SnO2 interface is larger than that at the pristine SnO2 surface, which eventually causes a higher response to CO gas.61)

    4.Conclusion

    In summary, pristine and Au NP-decorated SnO2 NFs with different amounts of Au NPs were synthesized by electrospinning and subsequent sputtering and thermal annealing of Au. Gas sensing tests were performed in the presence of CO, C6H6, and C6H7 and the effects of the Au concentration and operating temperature on the CO sensing properties were studied. A bell-shaped behavior in the sensing curve was observed as a function of the amount of loading of Au NPs, with a maximum response at 2.6 at.% Au. It was found that the Au decoration greatly enhances the CO response of the SnO2 NFs. This was attributed to the synergistic effect caused by the electronic sensitization originating from the difference in the work functions between Au and SnO2 and the chemical sensitization of the Au NPs. The results showed that the optimization of the amount of loading of metal NPs is one of the major factors determining the response of metal NP-decorated oxide NFs sensors.

    Acknowledgment

    This work was supported by the National Research Foundation of Korea(NRF) grant funded by the Korean government(MSIP: Ministry of Science, ICT and Future Planning)(No.2016M2B2A4911989).

    Figure

    MRSK-26-12-741_F1.gif

    SEM images of (a) pristine SnO2 NFs and SnO2 NFs decorated with Au NPs containing (b) 1.2 at.%, (c) 2.1 at.%, (d) 2.6 at.%, (e) 4.6 at.%, (f) 5.0 at.%, and (g) 19.0 at.% Au. (h) Low-magnification SEM image of SnO2 NFs functionalized with 2.6 at.% Au NPs.

    MRSK-26-12-741_F2.gif

    Low-magnification SEM images of (a) pristine SnO2 NFs, and SnO2 NFs decorated with Au NPs containing (b) 1.2 at.%, (c) 2.1 at.%, (d) 4.6 at.%, (e) 5.0 at.%, and (f) 19.0 at.% Au.

    MRSK-26-12-741_F3.gif

    (a) TEM image of 2.6 at.% Au-decorated SnO2 NFs. (b) Highresolution TEM image, (c) SAED pattern, and (d) EDS analysis.

    MRSK-26-12-741_F4.gif

    (a) Elemental analysis of 2.6 at.% Au-decorated SnO2 NFs using EDS. (b) Relationship between the amount of Au and the Au layer thickness.

    MRSK-26-12-741_F5.gif

    EDS elemental analysis of SnO2 NFs decorated with Au NPs containing (a) 1.2 at.%, (b) 2.1 at.%, (c) 4.6 at.%, (d) 5.0 at.%, and (e) 19.0 at.% Au.

    MRSK-26-12-741_F6.gif

    (a) Resistance curves of pristine SnO2 NFs exposed to 10 ppm CO gas at various temperatures. (b) Responses of pristine SnO2 as a function of operating temperature.

    MRSK-26-12-741_F7.gif

    (a) Resistance curves of Au-decorated SnO2 NFs exposed to CO gas at 300 °C. (b) Calibration curve for various CO concentrations. (c) Responses to 10 ppm CO gas.

    MRSK-26-12-741_F8.gif

    (a) Resistance curves of 2.6 at.% Au-decorated SnO2 NFs for various reducing gases at 300 °C. (b) Calibration curve and (c) comparison of the responses of 2.6 at.% Au-decorated SnO2 NFs with pristine SnO2 NFs for 10 ppm concentration of various reducing gases.

    MRSK-26-12-741_F9.gif

    Schematic representations of sensing mechanism of (a) pristine and (b) Au-decorated SnO2 NFs.

    MRSK-26-12-741_F10.gif

    Band structures of Au and SnO2 (a) before contact, (b) after contact in air, and (c) after contact in CO.

    Table

    Comparison of some CO gas sensors with the sensor developed in this study.

    Reference

    1. Mirzaei A , Park S , Sun GJ , Kheel H , Lee C (2016) J. Hazard. Mater, Vol.305 ; pp.130
    2. Durante W , Johnson FK , Johnson RA (2006) J. Cell. Mol. Med, Vol.10 ; pp.672
    3. Fine GF , Cavanagh LM , Afonja A , Binions R (2010) Sensors, Vol.10 ; pp.5469
    4. Montuschi P , Khartimonov SA , Barnes PJ (2001) Chest, Vol.120 ; pp.496
    5. Patnaik P (2007) A comprehensive guide to the hazardous properties of chemical substances, John wiley & sons. inc, ; pp.403
    6. Korotcenkov G , Brinzari V , Cho BK (2016) Microchim. Acta, Vol.183 ; pp.1033
    7. Yanagimotoa T , Yu YT , Kaneko K (2012) Sens. Actuators. B, Vol.31 ; pp.166-167
    8. Yu MR , Wu RJ , Chavali M (2011) Sens. Actuators. B, Vol.153 ; pp.321
    9. Das S , Jayaraman V (2014) Progr. Mater. Sci, Vol.66 ; pp.112
    10. Lou XW , Wang Y , Yuan CL , Lee JY , Archer LA (2006) Adv. Mater, Vol.18 ; pp.2325
    11. Jiang LH , Sun GQ , Zhou ZH , Sun SG , Wang Q , Yan SY , Li HQ , Tian J , Guo JS , Zhou B (2005) J. Phys. Chem. B, Vol.109 ; pp.8774
    12. Kuo SL , Wu NL (2003) Electrochem. Solid State Lett, Vol.6 ; pp.A85
    13. Snaith HJ , Ducati C (2010) Nano Lett, Vol.10 ; pp.1259
    14. Kim HJ , Lee JH (2014) Sens. Actuators. B, Vol.192 ; pp.607-627
    15. Kaur J , Kumar R , Bhatnagar MC (2007) Sens. Actuators. B, Vol.126 ; pp.478
    16. Wang Q , Wang C , Sun H , Sun P , Wang Y , Lin J , Lu G (2011) Sens. Actuators, Vol.222 ; pp.257
    17. Liewhiran C , Tamaekong N , Wisitsoraat A , Phanichphant S (2009) Sensors, Vol.9 ; pp.8996
    18. Kong X , Li Y (2005) Sens. Actuators. B, Vol.105 ; pp.449
    19. Pourfayaz F , Khodaddadi A , Mortazavi Y , Mohajerzadeh SS (2005) Sens. Actuators. B, Vol.108 ; pp.172
    20. Nayak AK , Ghosh R , Santra S , Guha PK , Pradhan D (2015) Nanoscale, Vol.7 ; pp.29
    21. Du H , Wang J , Su M , Yao P , Zheng Y , Yu N (2012) Sens. Actuators. B, Vol.746 ; pp.166-167
    22. Sun P , Wang C , Liu J , Zhou X , Li X , Hu X , Lu G (2015) ACS Appl. Mater. Interfaces, Vol.7 ; pp.19119
    23. Rahman MM , Khan SB , Jamal A , Faisal M , Asiri AM (2012) Talanta, Vol.95 ; pp.18
    24. Liu L , Zhang Y , Wang G , Li S , Wang L , Han Y , Jiang X , Wei A (2011) Sens. Actuators. B, Vol.160 ; pp.448
    25. Kim T , Guo B (2011) J. Ind. Eng. Chem, Vol.17 ; pp.158
    26. Zhang J , Liu X , Wu S , Cao B , Zheng S (2012) Sens. Actuators. B, Vol.169 ; pp.61
    27. Uddin ASMI , Lee KW , Chung GS (2015) Sens. Actuators. B, Vol.216 ; pp.33
    28. Kim JH , Kim HW , Kim SS (2017) Sens. Actuators. B, Vol.239 ; pp.578
    29. Yuanda W , Maosong T , Xiuli H , Yushu Z , Guorui D (2001) Sens. Actuators. B, Vol.79 ; pp.187
    30. ACSPark S , Sun GJ , Jin C , Kim HW , Lee S , Lee C (2016) Appl. Mater. Interfaces, Vol.8 ; pp.2805
    31. Kim S , Park S , Park S , Lee C (2015) Sens. Actuators. B, Vol.209 ; pp.180
    32. Joshi RK , Hu Q , Alvi F , Joshi N , Kumar A (2009) J. Phys. Chem. C, Vol.113 ; pp.16199
    33. Murdoch M , Waterhouse GIN , Nadeem MA , Metson JB , Keane MA , Howe RF , Llorca J , Idriss H (2011) Nat. Chem, Vol.3 ; pp.489
    34. Park JY , Choi SW , Kim SS (2011) J. Phys. Chem. C, Vol.115 ; pp.12774
    35. Bai S , Chen C , Zhang D , Luo R , Li D , Chen A , Liu C (2014) C. Sens. Actuators. B, Vol.204 ; pp.754
    36. Xu L , Song H , Dong B , Wang Y , Chen J , Bai X (2010) Inorg. Chem, Vol.49 ; pp.10590
    37. Jiaqiang X , Yuping C , Yadong L , Jianian S (2005) J. Mater. Sci, Vol.40 ; pp.2919
    38. Zheng W , Lu X , Wang W , Li Z , Zhang H , Wang Y , Wang Z , Wang C (2009) Sens. Actuators. B, Vol.142 ; pp.61
    39. Park S , Kim H , Jin C , Choi SW , Kim SS , Lee C (2012) Thermochim. Acta, Vol.542 ; pp.69
    40. Abideen ZU , Katoch A , Kim JH , Kwon YJ , Kim HW , Kim SS (2015) Sens. Actuators. B, Vol.221 ; pp.1499
    41. Greiner A , Wendorff JH (2007) Angew. Chem. Int. Ed, Vol.46 ; pp.5670
    42. Yarin AL (2011) Polym. Adv. Technol, Vol.22 ; pp.310
    43. Wei S , Zhao G , Du W , Tian Q (2016) Vacuum, Vol.124 ; pp.32
    44. Zhang Z , Li X , Wang C , Wei L , Liu Y , Shao C (2009) J. Phys. Chem. C, Vol.113 ; pp.19397
    45. Lee JS , Katoch A , Kim JH , Kim SS (2016) Sens. Actuators. B, Vol.222 ; pp.307
    46. Park S , Kim S , Park S , Lee WI , Lee C (2014) Sensors, Vol.14 ; pp.15849
    47. Korotcenkov G (2005) Sens. Actuators. B, Vol.107 ; pp.209
    48. Park S , Sun GJ , Kheel H , Hyun SK , Jin C , Lee C (2016) Met. Mater. Int, Vol.22 ; pp.156
    49. Hwang IS , Choi JK , Woo HS , Kim SJ , Jung SY , Seong TY , Kim ID , Lee JH (2011) ACS Appl. Mater. Interfaces, Vol.3 ; pp.3140
    50. Wongrat E , Hongsith N , Wongratanaphisan D , Gardchareon A , Choopun S (2012) Sens. Actuators. B, Vol.230 ; pp.171-172
    51. Li H , Xu J , Zhu Y , Chen X , Xiang Q (2010) Talanta, Vol.82 ; pp.458
    52. Matsumiya M , Qiu FB , Shin W , Izu N , matsubara I , Murayama N , Kanzaki S (2004) J. Electrochem. Soc, Vol.151 ; pp.H7
    53. Korotcenkov G , Cho BK , Gulinac L , Tolstoy V (2009) Sens. Actuators. B, Vol.141 ; pp.610
    54. Neri G , Bonavita A , Milone C , Galvagno S (2003) Sens. Actuators. B, Vol.93 ; pp.402
    55. Kim JH , Wu P , Kim HW , Kim SS (2016) ACS Appl. Mater. Interfaces, Vol.8 ; pp.7173
    56. Haruta M , Kobayashi T , Sano H , Yamada N (1987) Chem. Lett, Vol.2 ; pp.405
    57. Mirzaei A , Hashemi B , Janghorban K (2016) J. Mater. Sci. Mater. Electron, Vol.27 ; pp.3109
    58. Mirzaei A , Janghorban K , Hashemi B , Bonavita A , Bonyani MS , Leonardi G , Neri G (2015) Nanomaterials, Vol.5 ; pp.737
    59. Ramgir NS , Kaur M , Sharma PK , Datta N , Kailasaganapathi S , Bhattacharya S , Debnath AK , Aswal DK , Gupta SK (2013) Sens. Actuators. B, Vol.187 ; pp.313
    60. Kwon YJ , Na HG , Kang SY , Choi MS , Bang JH , Kim TW , Mirzaei A , Kim HW (2017) Sens. Actuators. B, Vol.239 ; pp.180
    61. Guo J , Zhang J , Gong H , Ju D , Cao B (2016) Sens. Actuators. B, Vol.226 ; pp.266
    62. Yamaura H , Iwasaki Y , Hirao S , Yahiro H (2011) Sens. Actuators. B, Vol.153 ; pp.465
    63. Neri G , Bonavita A , Micali G , Rizzo G , Callone E , Carturan G (2008) Sens. Actuators. B, Vol.132 ; pp.224
    64. Hjiri M , Mir LE , Leonardi SG , Pistone A , Mavilia L , Neri G (2014) Sens. Actuators. B, Vol.196 ; pp.413
    65. Buono-Core GE , Klahn AH , Cabello G , Lillo L (2013) Polyhedron, Vol.62 ; pp.1
    66. Yu JH , Choi GM (2001) Sens. Actuators. B, Vol.75 ; pp.56
    67. McCue JT , Ying JY (2007) Chem. Mater, Vol.19 ; pp.1009
    68. Yamaura H , Jinkawa T , Tamaki J , Moriya K , Miura N , Yamazoe N (1996) Sens. Actuators. B, Vol.36 ; pp.325
    69. Zeng W , Li Y , Miao B , Lin L , Wang Z (2014) Sens. Actuators. B, Vol.191 ; pp.1
    70. Wang CT , Chen MT (2010) Sens. Actuators. B, Vol.150 ; pp.360