Elsevier

Sensors and Actuators B: Chemical

Volume 239, February 2017, Pages 841-847
Sensors and Actuators B: Chemical

Applicability of superposition for responses of resistive sensors in a diluted mixed gas environment

https://doi.org/10.1016/j.snb.2016.07.095Get rights and content

Highlights

  • The potential of using superposition to predict the sensitivity of resistive type gas sensors.

  • The measured sensitivity was compared with the calculated sensitivity based on our mathematical formula.

  • The actual concentrations of the input gases were compared to the gas concentrations calculated from the observed response of the sensors.

Abstract

The potential for using superposition to predict the response of resistive-type gas sensors when they are exposed to more than a single gas was tested using two SnO2-based sensors. It is proposed that the selectivity of resistive-type gas sensors could be improved, or individual gas concentrations extracted, based on the observed response values of the sensors and a mathematical formulation. In this study, the behavior of several resistive gas sensors was tested in a mixed-gas environment of toluene and acetone, using a GASENTEST, which is an automatic gas sensor testing machine developed for this study. The measured responses were compared with calculated responses based on our mathematical formula. Additionally, the actual concentrations of the input gases were compared to the gas concentrations calculated from the observed response of the sensors.

Introduction

With the ongoing concern for a clean environment, interest in applications that exploit the Internet of Things (IOT) continues to grow, and in particular, sensor technologies are increasingly being recognized for their potential value as signal generators [1]. One example of such a device are the gas sensors that are currently used in healthcare and environmental monitoring [2], [3]. Resistive type sensors in particular have attracted significant attention because they have the advantages of easy production, low cost and good sustainability. However, such devices have low selectivity when detecting typical gases in a mixed state because of the sensing mechanism of resistive gas sensors [4]. For most applications, in a mixed-gas environment gas concentrations can be obtained only for a specific gas. To overcome this weakness, many studies have focused on forming sensor arrays and enhancing the pattern response of multiple sensors [5], [6]. In this study, the behaviors of several resistive gas sensors were tested in a mixed-gas environment. The measured responses were compared with calculated values that were obtained using our mathematical formula to determine the potential of using superposition to improve the performance of resistive-type gas sensors.

E. Bornald [7] found that on exposure to a certain reducing gas atmosphere the conductance of a resistive-type sensing material such as SnO2 could be expressed asG=G0+kcimwhere G0, ci, k and m are the conductance of the sensor in air, the gas concentration of the ith species, the conductance dependence and the exponent constant, respectively. Since the conductance is inversely related to the resistance, Eq. (1) can be re-written asG=1R=1R0+kcimwhere R and R0 are the resistance of the sensor on exposure to a gas i at a concentration ci, and the resistance of the sensor in air, respectively. Thus, Eq. (2) can be further reduced to Eq. (3) using the concept of response S, defined as R0/RS=1+kcimwhere the response dependence, k′, corresponds to R0k. However, in reality it has been found that m can be approximated to 1 for a sensor exposed to low concentrations (a few hundred ppm) of a reducing gas [8]. On the other hand, according to the conductance model for a ceramic pellet sensor [9], the conductance can be written asG=G0exp(qVskT)where q, k, T and Vs are the electronic charge, the Planck constant, the temperature and the surface barrier height, respectively. According to the space charge model [10], Vs depends on the square of the amount of ionosorbed species on the surface, expressed as Eq. (5)Vs=qNs22εε0Niwhere ε, ε0, Ni, and Ns are the dielectric constant of the sensor, the permittivity of free space, the net density of ions in the space charge region and the density of the charged species on the surface, respectively. Substitution of Eq. (5) into Eq. (4) givesG=G0exp(q2Ns2kT2εε0Ni)

When ceramic sensors are exposed to a reducing-gas-containing environment, changes occur in Ns from Ns0 (Ns in air) to Ns0+ ΔNs (ΔNs < 0) through the reaction of the reducing gas with ionosorbed oxygen ions on the surface. Therefore, the conductance of the sensor measured from the reducing gas environment can be written asG=G0exp[q2(Ns0+ΔNs)2kTεε0Ni]=G0exp[q2(Ns02+2Ns0ΔNs+ΔNs2)kT2εε0Ni]=G0exp[q2Ns02kT2εε0Ni]exp[q2(2Ns0ΔNs+ΔNs2)kT2εε0Ni]=G0exp[2Ns0ΔNs+ΔNs2kT2εε0Ni]Because Ns0 is a small value of only a few percent according to the Weisz limitation [11], ΔNs  0, and a Taylor series approximation can be applied to Eq. (7) to giveG=G0'(12Ns0ΔNs+ΔNs2kT2εε0Ni)

Assuming ΔNs2  0, Eq. (8) can be rearranged asG=G0'(12Ns0ΔNskT2εε0Ni)=G0'+kΔNs

Comparing Eq. (1) with Eq. (9), ΔNs must be proportional to the concentration of the reducing gas, c. If that is true for various reducing gases, we can approximate the total change in the concentration of surface ionosorbed oxygen ions asΔNs,mix=k1'c1++kn'cnwhen the sensor is placed in a mixed-gas environment containing n different reducing gases. Substitution of Eq. (10) into Eq. (9) gives the total conductivity of the sensor under mixed-gas conditions asGmix=G0'+k(k1'c1++kn'cn)

or, in terms of responseSmix=1+k1''c1++kn''cn

Therefore, when the sensor is exposed to a single gas, the response returns to Eq. (3). If we can obtain responses for various gases independently (e.g., S1, S2, S3…), the total response of the sensor in a mixed-gas environment can be expressed asSmix=1+(S11)+(S21)+(S31)+=i=1nSi(n1)

Eq. (12) states that we can superimpose the responses of the sensor to individual gases to obtain a total response for the sensor under mixed-gas conditions if k″ and the concentration for each gas are available. This means that solutions can be found for Eq. (12), as we have the same number of equations with concentrations, c. We therefore need to have the same number of different sensors as the number of gases whose concentrations, c, we want to measure. This can be simulated as an array of sensors. For example, assuming that two sensors, sensor 1 and sensor 2, are being used to measure the gas components x and y in a mixed state, their responses S1 and S2 under mixed-gas conditions can be expressed asS1,mix=1+k1x[x]+k1y[y]S2,mix=1+k2x[x]+k2y[y]and the solutions for the gas concentrations of x and y can be extracted as[x]=S2,mix1k2yS1,mix1k1yk2xk2yk1xk1y[y]=S2,mix1k2xS1,mix1k1xk2yk2xk1yk1x

Section snippets

Experimental

To minimize heat-dissipation through the electrode lines, a three-electrode platform fabricated on a MEMS structure was adopted. Here, the center Au electrode was surrounded by an Au-coated Pt-heater line, as shown in Fig. 1(a). The sensing material was deposited by screen printing on this line-mesh, as shown in Fig. 1(b). The chip size was 1700 × 2000 um and the mesh was 175 × 195 um. This type of sensor usually consumes about 25 mW when 0.6 V is applied to the heater. Packaging is required to mount

Results and discussion

First, the response to a single gas species (toluene or acetone) was measured for each sensor, to obtain k in Eq. (3). The responses of sensor 1 for various respective concentrations of toluene and acetone gases are shown in Fig. 3(a) and (b), while those of sensor 2 for toluene and acetone gases are shown in Fig. 3(c) and (d), respectively. All measurements showed a fast response of less than a few seconds and acceptable recovery, even in highly dilute gas environments.

From Fig. 3(a–d), k″ was

Conclusion

The potential for using superposition to predict the response of resistive-type gas sensors when they are exposed to more than a single gas was tested using two SnO2-based sensors. It is proposed that the selectivity of resistive-type gas sensors could be improved, or individual gas concentrations extracted, based on the observed response values of the sensors and a mathematical formulation. The behavior of two SnO2-based resistive gas sensors was tested in a mixed-gas environment of toluene

Acknowledgement

This study was supported by the KAIST Institute for Information Technology Convergence as the “Mobile Sensor and IT Convergence (MOSAIC) Project”.

Kwang-Min Park received his B.S. and M.S. degrees in Advanced Materials Engineering from Sejong University in 2008 and 2010. Now, he is a Ph.D. candidate in Materials Science and engineering, at Korea Advanced Institute of Science and Technology (KAIST). His current research interests are focused on solid state electrochemical devices and mobile gas sensors.

References (11)

There are more references available in the full text version of this article.

Cited by (9)

  • Ultrasensitive flexible NO<inf>2</inf> gas sensors via multilayer porous polymer film

    2022, Sensors and Actuators B: Chemical
    Citation Excerpt :

    Gas sensor that targeting hazardous gas perception with high responsivity and low limit of detection are of critical demand in many existing and emerging applications, such as environmental quality monitoring, personal healthcare, disease diagnosis and artificial olfactory systems [1–3]. It encourages researches to continuously explore various fundamental gas sensing mechanisms, particularly these based on resistive [4], electrochemical [5], optical [6] and filed-effect transistor (FET) devices [7]. Especially, in-situ signal amplification capability and strong interaction of gas analytes with semiconductors facilitate FET compelling for the construction of gas sensors with ultra-high sensitivities and ease of integration [8].

  • Picosecond laser-direct fabrication of graphene-based electrodes for a gas sensor module with wireless circuits

    2019, Microelectronic Engineering
    Citation Excerpt :

    The advances in nanotechnology-based sensors have enabled the development of simple, portable, disposable, low power-consuming, and inexpensive gas sensor devices; these devices are small and are more accurate in gas detection [1]. Recently, gas sensors have rapidly advanced and have been extensively used in medicine, explosive detection, toxic substance detection, and environmental studies [2]. Generally, most toxic gases encountered in daily life, such as carbon dioxide (CO2), carbon monoxide (CO), and nitrogen dioxide (NO2), are odorless and colorless [3–5].

  • Fully integrated and portable semiconductor-type multi-gas sensing module for IoT applications

    2018, Sensors and Actuators, B: Chemical
    Citation Excerpt :

    In order to resolve this problem, we have previously suggested a highly integrated platform that enables multi-sensing in a micro-scale, consuming fairly low power around 6 mW [1]. Also, selectivity-enhancing techniques using multiple resistive sensors are being developed [2]. For circuit issues to interface the sensors, it needs to readout the resistance over a wide range (i.e. variation in several decades) [3], and heater drivers should provide constant power for sensors to maintain a controlled and optimized temperature.

View all citing articles on Scopus

Kwang-Min Park received his B.S. and M.S. degrees in Advanced Materials Engineering from Sejong University in 2008 and 2010. Now, he is a Ph.D. candidate in Materials Science and engineering, at Korea Advanced Institute of Science and Technology (KAIST). His current research interests are focused on solid state electrochemical devices and mobile gas sensors.

Tae-Wan Kim received his B.S. degree from KAIST in 2013. He joined the Department of Materials Science and Engineering at Korea Advanced Institute of Science and Technology (KAIST) as an Integrated Master’s & Ph.D. Program student in 2014. His current research interests include chemical sensors using semiconductor and solid electrolytes.

Jeong-Ho Park received the B.S. degree in electrical engineering from the Korea Aerospace University, Goyang, Korea, in 2010, and the M.S. degree in electrical engineering from Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea, in 2012. He is currently working towards Ph. D. degree in electrical engineering from KAIST. His research interests include sensor systems, RF systems for mobile communications, and sensor communications.

Chong-Ook Park received his B.S. degree from Seoul National University in 1979 and M.S. and Ph.D. degrees from Ohio State University in 1985. He joined the Department of Materials Science and Engineering at KAIST in 1988, where he currently works as a professor. He is an organizer of IMCS2016 and a chairman of the Korean Sensor Society for 2016. His current research interests include gas sensors of semiconductor and solid electrochemical types.

View full text