Abstract
We present a semiconductor gas sensor based on mesoporous In2O3 (m-In2O3). The m-In2O3 was successfully fabricated by a simple sol-gel process, using block copolymer PE6800 as a soft template. The results of gas sensing reveal that the m-In2O3 prepared at room temperature shows higher resistance, which plays the key role in its greater sensitivity. The pore structure of material has an influence on gas adsorption on the material surface, which further affects response-recovery time of gas sensor.
1. Introduction
Over the past few years, mesoporous materials have received considerable attention due to their many attractive characteristics, such as well-ordered pore systems, large specific surface areas, high thermal stability, and these noble characteristics lead to their various areas of application including molecular separation [1], heterogeneous catalysis [2], nanocomposite preparation [3, 4], and gas sensors [5, 6].
In recent years, many mesoporous oxide materials like Al2O3, WO3, ZrO2 have been successfully synthesized [7, 8]. However, the research on m-In2O3 is still quite sparse. To our best knowledge, In2O3 is an n-type semiconductor with a wide bandgap close to GaN (3.5–3.7 eV) that shows the unusual combination of good optical transparency in the visible region and high electrical conductivity. Besides, In2O3 is also a promising gas sensing material for various oxidizing gases, such as O3 [9, 10], O2 [11], or nitric oxides [12, 13], as well as for reducing gases like CO or H2 [14, 15].
As we all know, the working principle of gas-sensing materials is based on measurement of gas adsorption and the changes of the conductivity caused by the surface reaction process. Increased surface area can improve the response performance. Therefore, the application of mesoporous materials can enhance the gas sensing performance owing to their large specific surface area. Though some examples of m-In2O3 have already been reported [16–18], little attention has been paid to study the effect of material structure on the gas-sensing performance. Herein, we successfully synthesized m-In2O3 with different structures by controlling the temperature during preparing process and investigated the effect of pore structure on the gas sensing performance.
2. Experimental Section
All the analytical chemicals were purchased from Shanghai Chemical Industrial Co. Ltd. (Shanghai, China), except for amphiphilic triblock copolymer PE6800, poly(ethylene oxide)-poly(propylene oxide)-poly (ethylene oxide) (EO73-PO28-EO73, Mw=8050), obtained from BASF and used without further purification.
2.1. Synthesis
The m-In2O3 materials, utilized amphiphilic triblock copolymer PE6800 as a soft template to control their morphology, were synthesized according to a literature produced [18].
In a typical synthesis, 0.01 mol of InCl3·4H2O was introduced into the 100 mL aqueous solution which contains 1 wt% triblock copolymer PE6800. Afterwards, the ammonia was dropped into the solution at 25°C, and the pH value of the final solution was controlled at 6.5. The mixed solution was then gently shaken until intimate mixing, and then white latex-like solution was obtained. After that, the above solution was aged for 3 days; subsequently the precipitate was poured out after centrifugal separating and washed with deionized water until no chloride ions were detected. Next, the products were dried overnight. Ultimately, the dried sample was calcined at 500°C for 2 hours to remove the surfactant (namely product-M-1), and the other product was prepared in the identical process at 40°C (namely, product-M-2).
2.2. Gas-Sensing Properties Research
The gas sensor was prepared, and the gas response was measured in the static state. Paste of In2O3 powders was coated onto an alumina ceramic tube attached with a pair of Au electrodes. The sensors were calcined at 500°C for 2 h to provide good mechanical strength and to remove the organic additives. Finally, a small Ni-Cr alloy heating coil was inserted into the tube to form a heater. The sensor should be kept at its working temperature for aging several days to evaluate the long-term stability. The electrical resistance of a sensor was measured in air and in detected gases. The gas-sensing responses of the material to C2H5OH, CH3OH, NH3, and NO2 gases were studied and correlated with the structural properties of the active materials.
3. Results and Discussion
3.1. Characterization of m-In2O3 Powders
XRD analysis is the most useful technique for identification of crystalline structure of the sample. Figure 1 displays wide-angle XRD of m-In2O3 synthesized at different temperatures after calcinations at 500°C for 2 h. The intensities and locations of the peaks were identified as a cubic structure of In2O3 (JCPDS card 6-0416). No impure peaks were observed in the wide-angle XRD patterns, which indicated high purity of the sample. From the patterns, it is perceptible that as the reaction temperature increased, the intensities of peak (222) to (400), (222) to (440) were enhanced, accordingly, which were larger than those of standard ratio. The surfactant PE6800 was found to be as a structure directing agent. Furthermore, the role of structure-oriented was boost up as the reaction temperature increased, which was in agreement with the results suggested by Yu et al. [19] using P123-synthesized mesoporous molecular sieves.
Low-angle XRD patterns, shown in Figure 2, exhibit a single XRD peak at and 0.62°, respectively, which confirm that the materials synthesized at different temperatures both have been occupied by the mesopore cavums. However, the single XRD peak indicates slightly lower overall nanostructural order or smaller coherent scattering domains. This is also confirmed by N2 adsorption and desorption isotherm (Figure 1), and the structural parameters are included in Table 1.
(a)
(b)
A more detailed analysis of the fine structure of the m-In2O3 (not allowed by the XRD) was obtained by High-Resolution Transmission Electron Microscopy (HRTEM), Field Emission Scanning Electron Microscopy (FE-SEM).
HRTEM images as depicted in Figure 3 clearly indicate that samples scattered evenly with an average diameter of 30 nm and 50 nm synthesized at 25°C and 40°C, respectively, which is well consistent with the results of XRD analysis.
(a)
(b)
FE-SEM image (Figure 4(a)) clearly shows that extremely aggregated nano particles with 30 nm were produced. Moreover, it can be seen that the product has mesoporous structures. The diameter of nano particles is up to 50 nm, and the structures of the mesoporous In2O3 are preserved without appreciable change when the reaction temperature was increased to 40°C, as shown in Figure 4(b).
(a)
(b)
As can be seen from Table 1, the structure of m-In2O3 is affected slightly by the reaction temperature. Both products are in round particle size; however, the related parameter increased as the temperature rises which was shown in Table 1.
Table 1 shows that the specific surface area of M-1 (54.78 m2g−1) is smaller than that of M-2 (62.82 m2g−1). The average diameter of the crystals can be estimated as about 30 nm, 50 nm, respectively. Generally, mesoporous material with the smaller pore size enables larger specific surface area. However, we find an opposite result that M-1 has smaller pore diameter than M-2, and specific surface area of M-1 is still smaller. So, we thought the result was concerned with the structure of bimodal pore size distribution. In Figure 5, we see that the rate of small pores is lower in M-1 than in M-2, which is beneficial for enlarging specific surface area.
(a)
(b)
Then N2 adsorption and desorption isotherm for m-In2O3 of different reaction temperatures can be seen in Figure 5. The BJH pore size distribution curve showed that the pore size of M-1 was mainly in the range of 2–8 nm and 12–48 nm, while that of M-2 was mainly in the range of 4–10 nm and 20–60 nm, which matched well with the results of N2 adsorption and desorption isotherm. Simultaneously, the BET surface areas are 54.78 m2/g and 62.82 m2g−1, respectively.
The N2 adsorption-desorption isotherm of samples made by different reaction temperatures for m-In2O3 is markedly dissimilar, which indicated that the pore structures of these two samples are different. In this case, the process of gas adsorption on the material surface is varied, and it further affects response-recovery time of gas sensor.
3.2. Sensing Properties of Gas Sensor
We have fabricated chemical sensors successfully with as-prepared nano architectures. Firstly, the gas response was measured in the static state under different operating temperatures. The results were shown in Figure 6, and m-In2O3 showed excellent gas-sensing properties to these tested gases. As shown in Figure 6, the sensitivities of m-In2O3 to C2H5OH, CH3OH, and NO2 are higher than those of HCHO in the testing gases. In addition, M-1 synthesized at 25°C performed higher sensitivity compared to 40°C (except NO2).
Accordingly, we can obtain that the resistance of materials in the air plays a key role in their response. In this paper, the response (S) of the sensor is defined as (for reducing gas), where and are the resistances of the sensors in the testing gas/air mixture and in air, respectively, so the bigger resistance in air will induce the higher sensitivity. Generally, the response of surface adsorption-controlled materials depends on specific surface area of materials, that is, the larger the specific surface area is, the higher sensitivity it achieves, which is contrary to the experimental results. In order to explain the abnormal phenomenon, the components of resistance in the air were studied, which was shown in Figure 7.
In Figure 7 the resistance of the two materials presented consistent change as the heating temperature increased, which shows typical characteristic of surface adsorption-controlled materials. However, M-1 is of higher resistance clearly than that of M-2. It could be assessed that M-1 prepared at lower temperature has higher surface activity, because surface-bound oxygen atoms are able to abstract the electron easier from the material, thus the surface electron potential barrier increased, causing the resistance of materials in air to be increased, thereby the sensitivity of the material enhanced.
The response and recovery time of our sensors has also been investigated. Here the response time is defined as the time required for the sample variation resistance to reach 90% of the equilibrium value following a step increase in the test gas concentration, while the recovery time is defined as the time necessary for the sample to return to 10% above its original resistance in air following the zeroing of the test gas. Figure 8 shows the response-recovery curve of m-In2O3 prepared to 50 ppm of C2H5OH. It can be seen that the response times are both in 5 s, whereas recovery times are totally different. M-1 showed faster recovery time than M-2. Combined with BET tests (see Figure 5), we find that the M-1 mainly cause the occurrence of single-molecule adsorption and with easy desorption of target gases, which leads to very short recovery time. M-2 with multi-molecular layers adsorption shows a larger recovery value. In summary, the m-In2O3 synthesized at room temperature has better gas-sensing properties.
4. Conclusions
In summary, the pore structure of nanomaterial plays a vital role in the application of semiconducting metal oxides as gas sensors. The m-In2O3 prepared at room temperature performs greater sensitivity, which plays the key role in its better gas-sensing properties. The pore structure of material has an influence on gas adsorption on the material surface and further affects response-recovery time of gas sensor.
Acknowledgments
This work was supported by the National Nature Science Foundation of China (no. 61071040, 50872142), the Leading Academic Discipline Project of Shanghai Municipal Education Commission (no. J50102), and Innovation Fund of Shanghai University (2010). The authors also thank the Analysis and Research Center at Shanghai University for sample characterization.