NO2 Sensing Properties of Cr2WO6 Gas Sensor in Air and N2 Atmospheres

Introduction

National nitrogen oxide (NO_x) emissions were 20.6 million metric tons (Mt) in 2015, with annual growth rates of 5.9% since 1949 in China (Richter et al., 2005; Sun et al., 2018). Among them, owing to its toxic effects to animals and plants, NO₂ is irritant and corrosive even at ppm level with serious harm to the respiratory tract and causticity particularly in children and elderly (Ling and Leach, 2004; Kida et al., 2009). Thus, determination of NO_x emissions is vital to regional and global ozone air pollution, acid deposition, and climate change (Jaegle et al., 2005). There have been great demands for cheap, reliable, and effective methods of real-time monitoring of NO₂ level in the environment (Stǎnoiu et al., 2012). To date, gas sensors based on metal oxide semiconductors (MOS) have been extensively investigated and successfully commercialized for NO_x detection due to their superior properties, simple structure, and low cost (Afzal et al., 2012). Among them, WO₃ based sensors show an excellent property to NO₂ in respect of sensitivity (Choi et al., 2004; Hua et al., 2018d). It was reported that WO₃ sensors fabricated through a wet process were sensitive to ppb levels of NO₂ with a low temperature due to a strong adsorption ability of NO₂ onto tungsten atoms compared with the weak adsorption of oxygen (Choi et al., 2004). Recently, gas sensors based on P-type MOS materials have been reported to have a sensitive and selective response to NO₂ with a low cross-sensitivity to humidity (Nguyen and El-Safty, 2011; Stǎnoiu et al., 2012). It was reported that the NO₂ sensors based on Cr₂O₃ showed a selectivity relative to CO and low cross-sensitivity to humidity, which was due to the higher surface reactivity toward NO₂ than CO through the electrical resistance and work function changes both in dry and humid air (Stǎnoiu et al., 2012). Gas sensors based on NiO nanosheets were fabricated by a hydrothermal method, which showed high sensitivity and selectivity to NO₂ (Nguyen and El-Safty, 2011). That may be due to NO₂ having higher electron affinity than the pre-adsorbed oxygen (Hoa et al., 2009). In this paper, chromium tungstate (Cr₂WO₆) was prepared by a wet chemistry method followed by hydrothermal treatment. The sensing properties to NO₂ in oxygen and non-oxygen atmosphere were investigated under dry and humid conditions, respectively. It was found that sensors based on P-type Cr₂WO₆ were very sensitive to NO₂ even with a high humid condition. In addition, the sensing mechanism was verified by the oxygen adsorption behavior in different conditions. Eventually, the fundamental sensing mechanism to NO₂ was explained.

Experimental

Chromium(III) nitrate nonahydrate (99.95%) and sodium tungstate dehydrate (ACS 99.0–101.0%) were provided by Shanghai Aladdin Biochemical Technology Co., Ltd. Chromium tungstate (Cr₂WO₆) nanoparticles were synthesized by a hydrothermal assisted process as described in Supplementary Materials A (Zhou et al., 2015). Subsequently, the powders were annealed at 1,000°C in air for 2 h. The samples were characterized using X-ray diffraction (XRD; D8 FOCUS, Bruker, Germany) in Cukα radiation with corresponding wavelengths of 1.54 Å and a filament current and voltage of 15 mA and 40 kV and a field-emission scanning electron microscope (FE-SEM; Nova Nano SEM 450, FEI). The surface morphology of the material was analyzed by transmission electron microscopy (TEM; Tecnai-F20, FEI, USA) with an accelerating voltage of 200 kV. The sample powders were mixed with glycerin to form a homogeneous paste and then was screen-printed on the alumina substrate. In order to obtain good stability, all sensors were aged at 400°C for 24 h. The gas sensing performance was measured by DC resistance with a homemade apparatus equipped with a dynamic gas distribution system as shown schematically in Figure 1. Target gases were supplied by gas cylinders with appropriate concentrations balanced with the carrier gases (air or nitrogen). The humidity and oxygen concentrations were calibrated using a humidity sensor (SHT31-ARP, Sensirion, Switzerland) and oxygen analyzer (SST, England), respectively. Keithley multimeter (Keithley 2000, USA) was used to record all data in real time. The sensor response was defined as S = R_a/R_g, where R_a and R_g were the resistances in the presence of air/N₂ and oxidizing gases, respectively.

FIGURE 1

Figure 1. Experimental system for the sensing characteristics of sensor devices.

Results and Discussion

In order to obtain a high purity of Cr₂WO₆ phase, a high sintering temperature of 1,000°C was used and the crystal structure was characterized by XRD. Figure S3 shows the XRD patterns of prepared Cr₂WO₆ powders. Obviously, XRD patterns exhibit very narrow peaks indicating a good crystal quality and the patterns could be well-fit with the tetragonal phase (JCPDS 35-0791) indicating a high purity of the prepared Cr₂WO₆. The morphology of Cr₂WO₆ powders was characterized by SEM and TEM. Figures 2A,B show SEM images of Cr₂WO₆ powders, which consisted of huge number of particles in a grain shape with a large size. According to the insert SEM images the grain shape is estimated around 300 nm by counting. TEM images in Figure 2C show that the grain size is consistent with the SEM images. Additionally, HRTEM images in Figure 2D also suggest a good crystalline quality of Cr₂WO₆ nanoparticles and the lattice spacing of 0.323 nm and 0.248 nm is in good accordance with (110) and (103) planes of tetragonal Cr₂WO₆ (JCPDS 35-0791) and results of XRD (Zhou et al., 2015).

FIGURE 2

Figure 2. The morphology analysis of powders (A,B) SEM images, (C) TEM images, and (D) HRTEM micrograph with lattice diffraction pattern of nanoparticles.

The sensing properties were characterized with NO₂ ranging from 0.2 to 5 ppm balanced with syntheses with an operation temperature of 200–350°C. According to previous reports, chemiresistive type sensors with a low concentration of carriers, i.e., holes in the present case, have a strong transduction ability with high cost of resistance (Bârsan et al., 2010; Hua et al., 2018a). It is worth noting that two identical sensor devices fabricated from the same materials were measured; however, only one sensors' data was presented for simplicity. The other one was used as a reference and not shown. Figure 3A presented the time and temperature dependence of sensor resistance. It was found that sensor resistance significantly increased with reduction in operation temperatures. When the temperature decreased from 250 to 200°C, sensor resistance was almost increased by 3 times. However, when the temperature was lower than 200°C, sensors gave an extremely high resistance reaching to 10⁸ Ω and over the range of measurement (Keithley 2000), which is also very difficult for practical applications. Thus, sensing response was only characterized from 200 to 350°C. One can note that sensor resistance is significantly reduced when exposed to NO₂, suggesting a P-type response of Cr₂WO₆. Moreover, the sensor responses were found to be highly temperature-dependent and increased with reduction in temperatures (Bodneva et al., 2019). When the working temperature increased to 350°C, the responses of 5 ppm NO₂ were reduced by 4.5 times.

FIGURE 3

Figure 3. (A) The transient response of sensor devices and (B) the sensor responses as a function of NO₂ concentration and (C) temperatures.

Detection of NO₂ in the non-oxygen atmosphere was highly required and could be applied to some specific scenes, such as the fuel leaks in aerospace systems. The sensor responses to NO₂ balanced with N₂ were investigated from 200 to 350°C in the absence of oxygen. Owing to its p-type conduction, the baseline resistance in N₂ was largely enhanced caused by the increase in electrons released by adsorbed oxygens on the surface. Consequently, the sensor responses were greatly increased to NO₂ higher than 1 ppm and much higher than in air atmosphere as shown in Figure 3B. It was shown that the responses in N₂ conditions were 4 times as much as that in air conditions at 200°C as shown in Figure 3C. It was well-known that in the presence of oxygen there could be a competitive adsorption between oxygen and NO₂, and therefore, that may give a reason for the clear difference of sensing responses in the presence and absence of oxygen (Hua et al., 2018d). What's more, the competitive relation between NO₂ and oxygen will be analyzed by the oxygen adsorption behavior in the following. In addition, it was indicated that the sensing responses in N₂ atmosphere were highly dependent on the working temperatures similar to that in the air atmosphere. When the temperature decreased from 350 to 200°C, sensor resistance was almost promoted by 2 orders and sensor responses were enhanced by almost 4 times. As shown in Figure S4, it seems that the sensors gave a relatively poor response to NO₂ below 1 ppm, almost no response in a dry condition. In other words, the limit of detection (LOD) of NO₂ for Cr₂WO₆ nanoparticles is better than 1 ppm. This is very similar to a typical n-type MOS gas sensor when exposed to reducing gases such H₂ and CO (Hua et al., 2018c). A small concentration of gas could be shielded by adsorption of oxygen due to the release of free electrons by the reaction of adsorbed oxygens with reducing gases. Thus, small concentration of reducing gases could not be detected (Hua et al., 2018d). However, in the present case, it is believed that the shielding effect of NO₂ could be caused by a competitive adsorption of oxygen and NO₂ on the surface of Cr₂WO₆. When NO₂ adsorbed onto the surface and shared the same sites with oxygen adsorption leading to desorption of oxygen, there was no electron charge transfer. As NO₂ molecular trap the electrons, which released from desorption of O₂, the surface density will not generate a net increase. As a result, there was no resistive response, which was known as a chemical shielding effect for MOS gas sensors (Hua et al., 2018d). Finally, a comparison of MOS sensors to NO₂, including various n-type and p-type materials, was given in Table 1 (Ling and Leach, 2004; Hoa et al., 2009; Nguyen and El-Safty, 2011; Vyas et al., 2013; Marichy et al., 2015). In conclusion, the Cr₂WO₆ sensors have reat sensitivity and lower operation temperatures to NO₂ compared with the n-type materials including SnO₂/WO₃ and ZnO both in air and N₂ atmospheres. Furthermore, some typical p-type materials were listed in Table 1, and the Cr₂WO₆ sensors displayed high sensitivity, which had great potential to be applied to some specific scenes. The stability of sensors have been also investigated in Figure S7 of Supplementary Material E.

TABLE 1

Table 1. The response to NO_x compared with other MOS sensors.

Moreover, the sensing properties to NO₂ balanced with air in the presence of humidity were also investigated at 200°C and the transient responses were shown in Figure S5 of Supplementary Materials C. It was clear that all sensor resistances were a direct proportion to the relative humidity at 25°C (RH at 25°C) compared with that in dry conditions as revealed in Figure 4. The phenomenon can be explained by water vapor competing for adsorption sites of NO₂ on the surface of Cr₂WO₆ nanoparticles. When the humidity reached 4% RH at 25°C, sensor resistance was promoted by ~2 times and the sensor responses were enhanced in the humid conditions in Figure 4. The sensor response was increased by almost 2 times to 2 ppm NO₂ in the presence of 4% RH at 25°C. In the sensing processes, NO₂ acted as the acceptor, meanwhile the water vapor was the donor. Moreover, the water vapor gave an inhibition to the oxygen adsorption on the material surface. Interestingly, the sensor responses were reduced when the humidity increased to 14 and 20% RH at 25°C. That may owe to the chemical adsorption of water vapor, which occupies the adsorption sites of NO₂. However, the responses were still higher than that in dry conditions. Moreover, the sensing properties balanced with N₂ in humid atmosphere were also analyzed by Keithley multimeter; nevertheless, the water vapor gave a promotion to sensor resistance compared with that in dry conditions. Therefore, we cannot evaluate the performance of the sensor devices based on the existing equipment in the laboratory.

FIGURE 4

Figure 4. The sensor responses of Cr₂WO₆ as a function of humidity and NO₂ concentration balanced with air or N₂ at 200°C.

According to our previous reports, oxygen adsorption, and reaction play a basic role in the sensing process with the formation of ${\text{O}}_2^{-}$, O⁻, or O²⁻ on the surface of metal oxides (Hua et al., 2018a,b,c). The oxygen adsorption behavior could be analyzed by the relationship between sensor resistance (R_g) and partial pressure of oxygen (P_O2) in different atmospheres, i.e., power-law response. Figure 5 shows the linear plot of R_g on the P_O2 ranging from 0.1 to 0.7 atm (1 atm = 100% in volume) in a double logarithm-scale at an operation temperature of 200°C. It was obvious that the R_g decreased with increasing P_O2, indicating the presence of oxygen adsorption on the surface of Cr₂WO₆ nanoparticles. Adsorption of oxygen on the surface traps electrons and releases holes resulting in a reduction in sensor resistance. Moreover, Figure 5A presents the power-law response to oxygen, and the fitting slope (n) was −0.15 at 200°C. However, for p-type materials, the receptor and transducer functions have not been built yet. Thus, the fitting power-law exponent could not be well-clarified. However, it was quite clear that oxygen adsorption served as the receptor function for Cr₂WO₆ without difference with typical n-type WO₃ or p-type Cr₂O₃ (Hua et al., 2018a,b,c). As mentioned before, water vapor forms chemical adsorption on surface leading to a block on oxygen adsorption as shown in Figure 5A. With humidity increasing from 0 to 10% RH at 25°C, R_g was promoted at all P_O2 and the absolute value of n was decreased, suggesting a strong electronic interaction of oxygen with the Cr₂WO₆ surface. This may be caused by the competitive adsorption between oxygen and water molecules (Hua et al., 2018c). With the presence of NO₂, the power-law response to oxygen was also investigated to clarify the role of oxygen on the sensing process of NO₂ (Hua et al., 2018d). When exposed to NO₂ of 1 ppm, the sensor resistance greatly decreased. According to Figure 5B, the power-law response to oxygen is not linear, suggesting that the adsorption of oxygen is influenced by NO₂ adsorption. More specifically, the electronic interaction of oxygen with the Cr₂WO₆ surface was significantly inhibited by NO₂, i.e., shielded by NO₂. This could be due to a competitive adsorption of NO₂, which may share the same adsorption sites, in case of typical MOS, are metal atoms and oxygen vacancies. When exposed to humidity in the presence of NO₂ (1 ppm balanced with N₂), sensor resistance was almost independent of P_O2, and the power-law exponent was merely 0.06, which is much smaller than that in a dry condition. The corresponding transient responses of power law response were shown in Figure S6 of Supplementary Material D. In addition, it was noted that the absolute value of the power-law exponent, n became much smaller compared with that in the absence of NO₂. In other words, humidity not only promotes the sensing response but also changes the basic mechanism of p-type Cr₂WO₆ to NO₂ according to the power-law response. It is possible that in the presence of humidity oxygen adsorption may take place with a competitive adsorption with water, which forms a weak adsorption on the surface of Cr₂WO₆. This causes a weak adsorption.

FIGURE 5

Figure 5. (A) The power-law response to oxygen for pristine Cr₂WO₆ in the absence and presence of water vapor and (B) a transient response to O₂ in the presence of 1 ppm NO₂ at 200°C.

Conclusion

A new sensing material Cr₂WO₆ was prepared and its sensing properties to NO₂ were investigated in both dry and humid conditions. The basic sensing mechanism was also analyzed by surface oxygen adsorption behavior in different atmospheres. Here are the conclusions:

➢ Sensors based on Cr₂WO₆ nanoparticles showed great response to NO₂ based on air atmosphere down to 0.2 ppm. The sensor responses in N₂ conditions were higher than in air conditions over 1 ppm of NO₂.

➢ Sensor resistances were raised up with the relative humidity increasing. However, sensor responses to all gases were enhanced by water vapor over the range of 0–20% RH at 25°C.

➢ The oxygen adsorption played a vital basic role in the sensing process. However, the oxygen adsorption behavior was inhibited by the presence of water vapor and NO₂.

In conclusion, sensing material Cr₂WO₆ shows a great response to detect NO₂ in dry and humid conditions.

Data Availability Statement

All datasets generated for this study are included in the article/Supplementary Material.

Author Contributions

MY and CT performed the experiments and analyzed the data with the help from YW, YL, and ZH. YW and ZH conceived and modified the manuscript based on experimental data.

Funding

This study was supported by the National Natural Science Foundation of China (Grant Nos. 51977059 and 51777056). The characterization results were supported by Beijing Zhongkebaice Technology Service Co., Ltd, and we acknowledge critical and quantity of testing work supported by Beijing Zhongkebaice Technology Service Co., Ltd.

Conflict of Interest

YL was employed by the Beijing Institute of Mechanical Equipment that is a government-owned non-profit organization.

The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Supplementary Material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/article/10.3389/fchem.2019.00907/full#supplementary-material

References

Afzal, A., Cioffi, N., Sabbatini, L., and Torsi, L. (2012). NO_x sensors based on semiconducting metal oxide nanostructures: Progress and perspectives. Sens. Actuators B Chem. 171, 25–42. doi: 10.1016/j.snb.2012.05.026

CrossRef Full Text | Google Scholar

Bârsan, N., Simion, C., Heine, T., Pokhrel, S., and Weimar, U. (2010). Modeling of sensing and transduction for p-type semiconducting metal oxide based gas sensors. J. Electroceram. 25, 11–19. doi: 10.1007/s10832-009-9583-x

CrossRef Full Text | Google Scholar

Bodneva, V., Ilegbusi, O., Kozhushner, M., Kurmangaleev, K., Posvyanskii, V., and Trakhtenberg, L. (2019). Modeling of sensor properties for reducing gases and charge distribution in nanostructured oxides: a comparison of theory with experimental data. Sens. Actuators B Chem. 287, 218–224. doi: 10.1016/j.snb.2019.02.034

CrossRef Full Text | Google Scholar

Choi, Y., Sakai, G., Shimanoe, K., and Yamazoe, N. (2004). Wet process-based fabrication of WO₃ thin film for NO₂ detection. Sens. Actuators B Chem. 101, 107–111. doi: 10.1016/j.snb.2004.02.031

CrossRef Full Text | Google Scholar

Hoa, N., Quy, N., and Kim, D. (2009). Nanowire structured SnO_x-SWNT composites: high performance sensor for NO_x detection. Sens. Actuators B Chem. 142, 253–259. doi: 10.1016/j.snb.2009.07.053

CrossRef Full Text | Google Scholar

Hua, Z., Li, Y., Zeng, Y., and Wu, Y. (2018a). A theoretical investigation of the power-law response of metal oxide semiconductor gas sensors I: Schottky barrier control. Sens. Actuators B Chem. 255, 1911–1919. doi: 10.1016/j.snb.2017.08.206

CrossRef Full Text | Google Scholar

Hua, Z., Qiu, Z., Li, Y., Zeng, Y., Wu, Y., Tian, X., et al. (2018b). A theoretical investigation of the power-law response of metal oxide semiconductor gas sensors II: Size and shape effects. Sens. Actuators B Chem. 255, 3541–3549. doi: 10.1016/j.snb.2017.09.189

CrossRef Full Text | Google Scholar

Hua, Z., Tian, C., Huang, D., Yuan, W., Zhang, C., Tian, X., et al. (2018c). Power-law response of metal oxide semiconductor gas sensors to oxygen in presence of reducing gases. Sens. Actuators B Chem. 267, 510–518. doi: 10.1016/j.snb.2018.04.002

CrossRef Full Text | Google Scholar

Hua, Z., Tian, C., Qiu, Z., Li, Y., Tian, X., Wang, M., et al. (2018d). An investigation on NO₂ sensing mechanism and shielding behavior of WO₃ nanosheets. Sens. Actuators B Chem. 259, 250–257. doi: 10.1016/j.snb.2017.12.016

CrossRef Full Text | Google Scholar

Jaegle, L., Steinberger, L., Martin,. R., and Chance, K. (2005). Global partitioning of NO_x sources using satellite observations: relative roles of fossil fuel combustion, biomass burning and soil emissions. Faraday Discuss. 130, 407–423. doi: 10.1039/b502128f

PubMed Abstract | CrossRef Full Text | Google Scholar

Kida, T., Nishiyama, A., Yuasa, M., Shimanoe, K., and Yamazoe, N. (2009). Highly sensitive NO₂ sensors using lamellar-structured WO₃ particles prepared by an acidification method. Sens. Actuators B Chem. 135, 568–574. doi: 10.1016/j.snb.2008.09.056

CrossRef Full Text | Google Scholar

Ling, Z., and Leach, C. (2004). The effect of relative humidity on the NO₂ sensitivity of a SnO₂/WO₃ heterojunction gas sensor. Sens. Actuators B Chem. 102, 102–106. doi: 10.1016/j.snb.2004.02.017

CrossRef Full Text | Google Scholar

Marichy, C., Donato, N., Latino, M., Georg Willinger, M., Tessonnier, J.-P., Neri, G., et al. (2015). Gas sensing properties and p-type response of ALD TiO₂ coated carbon nanotubes. Nanotechnology 26:024004. doi: 10.1088/0957-4484/26/2/024004

PubMed Abstract | CrossRef Full Text | Google Scholar

Nguyen, D., and El-Safty, S. (2011). Synthesis of mesoporous NiO nanosheets for the detection of toxic NO₂ gas. Chem. Eur. J. 17, 12896–12901. doi: 10.1002/chem.201101122

PubMed Abstract | CrossRef Full Text | Google Scholar

Richter, A., Burrows, J., Nüß, H., Granier, C., and Niemeier, U. (2005). Increase in tropospheric nitrogen dioxide over China observed from space. Nature 437, 129–132. doi: 10.1038/nature04092

PubMed Abstract | CrossRef Full Text | Google Scholar

Stǎnoiu, A., Simion, C., Diamandescu, L., Tǎrǎbǎşanu-Mihǎilǎ, D., and Feder, M. (2012). NO₂ sensing properties of Cr₂O₃ highlighted by work function investigations. Thin Solid Films 522, 395–400. doi: 10.1016/j.tsf.2012.09.003

CrossRef Full Text | Google Scholar

Sun, W., Shao, M., Granier, C., Liu, Y., Ye, C., and Zheng, J. (2018). Long-term trends of anthropogenic SO₂, NO_x, CO, and NMVOCs emissions in China. Earths Future 6, 1112–1133. doi: 10.1029/2018EF000822

CrossRef Full Text | Google Scholar

Vyas, R., Sharma, S., Gupta, P., Prasad, A. K., Dhara, S., Tyagi, A. K., et al. (2013). Nitrogen dioxide induced conductivity switching in ZnO thin film. J. Alloys Comp. 571, 6–11. doi: 10.1016/j.jallcom.2013.03.217

CrossRef Full Text | Google Scholar

Zhou, W., Huang, J., Li, J., Xu, Z., Cao, L., Yao, C., et al. (2015). Cr₂WO₆ nanoparticles prepared by hydrothermal assisted method with selective adsorption properties for methylene blue in water. Mater. Sci. Semicond. Process. 34, 170–174. doi: 10.1016/j.mssp.2015.02.010

CrossRef Full Text | Google Scholar