Solid State Gas Sensors - Industrial Application

It is often thought that technology drives innovation – a process that creates great values from ideas. But historically, it is a great vision that drives innovation and seeks for technologies as enablers. Take computer as a technology. “There is no reason for any individual to have a computer in their home,” said Mr. Ken Nelson, the then president of Digital Equipment Corporation or DEC in 1977 (Nova Online. “Traveling Through Time – Part 2”. Public Broadcasting Service. http://​www.​pbs.​org/​wgbh/​nova/​time/​through2.​html). Although immensely successful in minicomputer business, DEC failed to imagine what Home PC can do, and hence, unfortunately, saw one of the most dramatic demise of a technology company. It was lack of vision and imagination and not technology that led to such demise. Currently, Apple’s success in mobile industry is driven by Steven Jobs’ vision of delivering consumer infotainment by bundling technologies with flair and sleek products.

Therefore, it is important to paint a vision of what gas sensing can do in the future (Hagleitner et al., Nature 414:293–296, 2001; Moos et al., Sensors 9:4323–4365, 2009). What are the possibilities? How it can touch and benefit environment and human lives? Below is an attempt to paint vision through common-life examples. The goal here is to inspire and motivate us to think or dream big as how sensing can change our daily life experience. The dream should not be bounded by what can be achieved or not. That shall ultimately be governed by the real-life constraints such as costs, available technology, etc. But such lofty sensing vision should set a roadmap by defining requirements and seeking solutions for future.

This chapter presents a vision of solutions, particularly with gas sensing, that are achievable today with various matured and emerging technologies. The landscape of sensing solutions is changing fast as more and more technologies are becoming commercially available at even faster rate. Hence, once again, it is our vision that should set the pace of what we can expect tomorrow. So, let us keep imagining.

The implementation of gas sensors in automotive environments aims to improve the air quality for vehicle occupants. These sensors provide output signals corresponding to the gas concentration of the prevalent pollutant and the degree of odor contamination. This output signal is primarily used for the automatic recirculation control of the vehicle’s Heating, Ventilation, and Air Conditioning (HVAC) system.

This paper outlines the requirements of solid state gas sensors for use in automotive air quality applications. Implementing these sensors in an automotive environment poses a number of challenges, due to the wide range of possible temperature, atmospheric pressure, humidity, and vibration profiles. Additionally, the sensors must fulfill strict cost requirements and meet high standards of reliability and quality.

Based on the example of a metal oxide semiconductor gas sensor, the technical specification, data interpretation, application criteria, and automotive suitability are demonstrated. Furthermore, a brief overview of the use of solid state gas sensors for detecting the indoor cabin air quality via odor pattern recognition is given.

Hydrogen concentration monitoring in ambient air and in exhaust gas of fuel cell vehicles is important for a safe vehicle operation. This can be done by using automotive hydrogen sensors. This chapter describes the current and future sensor requirements and their implication on manufacturing and calibration. Its objective is to provide a requirement-based guideline for future sensor development.

In the field of fire detection, for more than 30 years the opto-electric principle was used for detecting a fire. In recent years, increasingly fire detectors are based on gas sensors. This chapter, based on some theoretical aspects, describes the different kinds of sensors used for fire detection. In the end, the role of standardization on different levels will be highlighted.

The challenge of nanotechnology is to discover new effects on already known materials and to convert exciting new findings into advanced technologies that are useful for industrial applications.

In these years, researchers have achieved the ability to produce quasi-one-dimensional (Q1D) structures in a variety of morphologies such as nanowires, core shell nanowires, nanotubes, nanobelts, hierarchical structures, nanorods, nanorings. In particular, Q1D Metal OXides (MOX) are attracting an increasing interest in gas sensing application: nanosized dimension ensures high specific surface that leads to the enhancement of catalytic activity or surface adsorption. Moreover, single-crystalline structures with well-defined chemical composition and surface terminations are not prone to thermal instabilities suffered from MOX polycrystalline counterpart. All these peculiarities can help to fill the gap between research and industrial application needs, aiming at the development of a reliable, low cost gas sensor.

This chapter presents an up-to-date survey of the research on Q1D metal oxide materials for gas sensing application, addressing the preparation techniques of sensing nano-crystals in connection with their electrical and optical properties. The application as resistive, transistor-based or optical-based gas sensors will be treated.

Looking at the literature about chemical sensors, it is evident that numerous sensor effects are reported but only a very few sensor concepts remain, which are suitable for industrial applications. This is in contrast to the measurement of physical parameters such as pressure, temperature, acceleration etc. There are several reasons for this discrepancy. First of all, for chemical reactions usually there exist much more cross correlations, which have to be considered in order to determine a correct concentration of chemical species, and secondly, environmental influences cannot be neglected at all. Therefore, it becomes extremely difficult to fulfill all the requirements, which have been listed in the introduction of this paper.

One interesting and successful gas sensor concept is based on the measurement of surface work function changes due to chemical reactions. Besides a Kelvin probe with a vibrating capacitance, field effect transistors (FET) are well suited to act as transducers for the determination of the corresponding potential variations. This chapter reviews the detection mechanisms, which lead to work function changes and gives an overview of the various transducer concepts. Beginning with the so-called Lundström FET, the historical development all the way to the hybrid mounted “floating gate FET (FG-FET)” is presented. This latest concept is extremely flexible and depending on the chemical-sensitive layer, it can be used for the detection of a large variety of gases.

As an example the development of a hydrogen sensor for future automotive application is presented in more detail. Using platinum as chemical-sensitive layer it is shown that under harsh environmental conditions, it is not sufficient to consider exclusively the reaction of hydrogen with the platinum, but rather, it is necessary to also take into account oxygen, which is always present in air. As a result, a characteristic catalytic ignition point, i.e., the adsorbed hydrogen atoms are consumed by a water-forming reaction, occurs at around 60°C. Only if the complete reaction scheme is considered, it is possible to understand the complex and partially intriguing transducer signals. The development of two-layer systems for the chemical-sensitive layer solves these problems by using a phase transition that comes along with the catalytic ignition and allows a stable sensor operation in the required temperature regime between −40°C and +120°C. Based on the theoretical modeling the temperature-dependent phase transition has been evaluated in order to measure hydrogen concentration up to 4% with high accuracy and to fulfill the automotive requirements.

Finally, in the last section, a new GasFET concept is presented, which extends the operating regime to temperatures as high as 400°C. In the future, this allows the incorporation of a variety of new gas-sensing materials, which need temperatures above 200°C in order to show a decent adsorption–desorption equilibrium.

Altogether, it has been proven that GasFETs are a very promising candidate for future industrial applications.

Chromium titanium oxide (Cr_2–xTi_xO_3+z, CTO) is a solid solution with the corundum crystal structure of the pure chromium oxide if x is in the range of 0.01–0.45. When heated to temperatures above 300°C, CTO shows a very strong and fast resistivity response to the presence of ammonia in air. The conductivity of CTO is primarily determined by chromium imperfections. In gas measurements, CTO shows a p-type semiconductor behavior. At even higher temperatures (>400°C), CTO is an excellent material for ammonia (NH₃) detection with a reduced cross sensitivity to humidity. This has been the key to the successful development of ammonia sensors based on CTO.

We investigated CTO as a sensitive material for NH₃ sensors operating at room and slightly elevated temperatures. It is based on the change of work function of Cr_1.8Ti_0.2O₃ upon gas exposure. CTO exhibits fast response and relaxation, no baseline drift induced by exposure and little influence of changing ambient humidity. The cross sensitivity to other gases is low, in particular to NO₂.

Temperature and humidity are very important parameters by measuring gas concentration. Most of the measuring procedures are showing a high temperature dependency; hence, the exact knowledge of the medium temperature is necessary in order to carry out high accurate measurements. While temperature compensation can be found in most gas measuring systems, the humidity of the medium is neglected in many cases. For rough concentration measurement this limitation can be accepted, but for highly accurate measurements humidity acts like an interfering gas.

The measurement of both these parameters can happen via different procedures. For the temperature measurement, an often implemented system is the use of characteristics of the gas measuring procedures (e.g., diode structures at silicon-based semiconductor gas sensors) or present resources in the application (e.g., resistance temperature sensor in processing units). The measurement of the moisture is clearly more difficult to integrate into the systems. Most frequently, type of moisture detection is realized with capacitive humidity sensors. These sensors allow the measurement of the relative humidity by evaluation of a capacitive signal obtained by a humidity-sensitive polymer. The knowledge of the temperature in these sensors is also indispensable for the determination of the absolute humidity respectively of the dew point.

The relative humidity has a strong temperature dependency (especially in the high humidity range). Therefore, it is an advantage to measure the temperature close to the humidity sensor. Thereby measurement mistakes could be inhibited effectively.

In this chapter, the measurement principle of thin film temperature sensors and capacitive humidity sensors, and examples for integration of both these technologies are described.

The response of tin oxide and tungsten oxide sensor elements to the volatiles of heated polymer fibres of textile fabrics was investigated. As an example, polyethylene terephthalate (PET) fibres covered with finish were chosen. Both the contributions to the sensor signal from the fibre and the finish are discussed. All compounds originating from the fibres contain at least one aromatic ring, the dominating components are benzoic acid, phenol, and 1-phenyl-1,2-propandion. The finish contributes with several aliphatic aldehydes and ketones of chain lengths from 5 to 19 carbon atoms. Rings with five carbon atoms are also due to the finish. The investigated tin oxide sensor and one of the tungsten oxide sensors respond only to one decomposition compound of the fibres, namely to 1-phenyl-1,2-propandion. The second investigated tungsten oxide sensor does not respond to any PET decomposition product. However, all sensors react with high signals to decomposition compounds of the finish.

Steadily increasing emission standards for passenger cars and heavy duty vehicles combined with the need for fuel efficiency lead to novel powertrain concepts, for example to leanly operated gasoline direct injection engines, or to novel exhaust gas aftertreatment concepts such as Lean NO_x Traps (LNT), ammonia selective catalytic reduction catalysts for NO_x reduction (SCR), or even to a combination of both. Also, diesel particulate filters (DPF) are in series production.

To control these novel exhaust gas aftertreatment systems and to monitor on-board the proper operation of these systems (on-board diagnosis, OBD), novel exhaust gas sensors are required or are at least be very helpful. Since the development of exhaust gas sensors has always to be seen in interaction with the corresponding exhaust gas aftertreatment systems, novel types of exhaust gas sensors have gained in importance just recently, when the time was ripe for novel exhaust gas aftertreatment concepts. This article reports on several types of NO_x sensors and ammonia sensors.

Additionally, a very recent novel concept is presented. Here, the catalyst itself works as a sensing device that gives directly information on its own status. The readout can be wirebound (demonstrated for LNT and SCR) or even be wireless by applying radio frequency techniques. It will be shown that this allows to detect the oxygen loading degree of three-way catalysts very precisely. It can be also applied to determine the ammonia loading of SCR catalysts and the soot loading of DPF.

As a conclusion, these novel methods may provide a future alternative for low emission-aiming engine control as well as for OBD of low emission vehicles with novel exhaust gas aftertreatment systems. However, it is clear that all novel sensors or systems do not only have to meet the technical requirements but also have to be very inexpensive, reliable, and cost effective.

The development of SiC-FET gas sensors has proceeded for about fifteen years. The maturity of the SiC material and a deeper understanding of the transduction mechanisms and sensor surface processes behind the sensitivity to a number of target substances have recently allowed the development of market-ready sensors for certain applications. Some examples presented below are a sensor system for domestic boiler control, an ammonia sensor for control of the SCR (selective catalytic reduction) and SNCR (Selective Non-Catalytic Reduction) NO_x abatement processes as well as other more or less market-ready applications. In parallel, the basic research continues in order to reach more demanding markets/new applications and also to possibly lower the production costs of the sensors. Therefore, current research and future challenges are also treated, such as the development of new types of conducting ceramics for ohmic contacts to SiC in order to increase the operation temperature beyond the present state of the art.

In the past, we have deeply investigated the development of a β-alumina potentiometric sensor with two coplanar metallic electrodes. Recently, many papers have demonstrated an interest for similar non-Nernstian devices based on stabilised zirconia (YSZ) electrolyte. The common particularity of such sensors is related to the use a dissymmetry of electrodes on an electrolyte and to work in a single gas cell. Our main experimental results obtained with β-alumina sensors are reported with the special attention to develop gas sensors for exhaust applications. The sensors are evaluated under CO, HC and NO_x CO. The selectivity depends on the sensor temperatures. The performances appear sufficient to satisfy in the future automotive applications like the control of NOxTrap or DeNOx devices. The measure of NH₃ concentrations seems also possible. The comparison of performances of sensors replacing β-alumina electrolyte by YSZ is also presented. Similar behaviours are obtained for both types of sensors. These comparisons reinforce our proposal of a capacitive model mainly based on the difference of reactivity of the electrodes. It is pointed out that mixed potentials cannot explain the behaviour of our sensors under oxygen. Very good agreements are obtained between this capacitive model and the experimental results, not only versus oxygen, but also with mixture of oxygen and CO.

Atmospheric humidity is strongly related to precipitation and more general to weather and weather prediction and as the most important greenhouse gas water vapour is strongly related to climate. For weather forecasting purposes, boundary layer and lower troposphere humidity observations are of prime interest. With climate change, accurate humidity measurements in the upper troposphere and lower stratosphere became very important. Upper air observations started in the second part of the nineteenth century and regular radiosonde balloon soundings are made since nearly 70 years. Humidity sensors on radiosondes had large uncertainties for many years but experienced a rapid evolution and were strongly improved over the last three decades. This article presents an overview of operational and scientific sensors used in upper air humidity measurements and then focuses on thin-film capacitive gas sensors, which are the most used hygrometers in present-day operational radiosondes. Information on international developments and future needs for upper air humidity observations are given as a motivation for further improvements of such humidity gas sensors.