DE102021210799A1 - Method for determining a gas concentration of a gas component in a gas using a sensor element - Google Patents

Abstract

A method for determining a gas concentration of a gas component in a gas using a sensor element (114) is proposed. The sensor element (114) has at least one pump cell (116) with at least two pump electrodes (134, 136) connected to one another by at least one solid electrolyte (118), wherein a first pump electrode (134) can be exposed to the gas and a second pump electrode (136) through at least one porous diffusion barrier (128) separated from the gas and disposed in a cavity (130) in the solid electrolyte (118). The method includes applying a predetermined pump voltage (UP) to the pump cell (116), the predetermined pump voltage (UP) being less than a pump voltage for a limit current case, applying a measuring AC voltage (Uz) to the pump cell (116) during the application the predetermined pump voltage (UP), measuring an impedance (Z) of the pump cell (116) and determining the gas concentration of the gas component in the gas based on the measured impedance (Z) of the pump cell (116).

Description

State of the art

Numerous exhaust gas sensors are known from the prior art.

The invention is described below, without restricting further possible configurations, essentially with reference to methods and devices which are used for the quantitative and/or qualitative detection of at least a portion of a gas in a measurement gas space. For example, the gas can be an exhaust gas from an internal combustion engine, particularly in the motor vehicle sector. The measurement gas space can be an exhaust tract, for example. The sensor element can be, for example, a lambda probe, in particular a broadband lambda probe. Lambda sensors are described, for example, in Robert Bosch GmbH: Sensors in motor vehicles, 1st edition 2010, pages 160-165. The proportion of the gas can be, for example, a target gas component, for example oxygen and/or nitrogen and/or nitrogen oxides and/or hydrocarbons and/or other types of gas components. In principle, the sensor element can also be another sensor, for example an NOx sensor. Sensor elements of the type mentioned can in particular be based on the use of one or more solid electrolytes, ie on the use of solid bodies, in particular ceramic solid bodies, which have ion-conducting, in particular oxygen-ion-conducting, properties. Examples of such solid electrolytes are zirconia-based solid electrolytes, such as yttrium-stabilized zirconia (YSZ) and/or scandium-doped zirconia (ScSZ). In the case of lambda probes, in particular in the case of broadband lambda probes, a quantity of oxygen (O ₂ ) and/or fat gas diffusing into a measurement cavity can be determined, for example, using a limit current, in particular in the case of unicellular organisms, and/or using a quantity necessary for regulating a cavity concentration to λ=1 Pump current, especially in double cells, are measured. For example, a flowing measurement current can be proportional to an O ₂ content and/or to a rich gas content in an exhaust gas. The cavity concentration can be measured by determining a Nernst voltage between a Nernst electrode in the cavity and an oxygen-flushed and/or air-flushed reference electrode in a reference space. A measurement of the oxygen partial pressure in the exhaust gas can be carried out from a linear relationship between the limiting current and an oxygen partial pressure.

In the exhaust gas or in other gases to be detected, however, there are not only oxygen components, but also other gas components whose detection is of interest. Such gas components are in particular hydrogen, carbon dioxide and water.

It is known, for example, that a carbon dioxide or water concentration measurement can be carried out by light absorption in the infrared range. With this principle, however, contamination of the optical components (e.g. the light window) is problematic, which leads to a decrease in sensitivity or signal drift over the operating time, which is why such sensors cannot be used in the area of an internal combustion engine.

Furthermore, the determination of water in gaseous media by a capacitive measurement is known. Typically, a polymer membrane, whose reversible water uptake correlates with the water content of the surrounding atmosphere, is placed in the electric field between two electrodes. By absorbing water, the dielectric strengthens and thus increases the capacitance of the capacitor. However, such sensors cannot be used in the field of internal combustion engines because of the very limited operating temperature of the polymer material.

In addition, carbon dioxide sensors are known which use high-temperature ion conductors such as complex phosphates (NASICON) or Na-β''-aluminate for carbon dioxide determination. The basic principle of such sensors is based on a powerless voltage measurement on a galvanic cell with alkali ion-conducting solid electrolytes. However, carbon dioxide sensors of this type can be irreversibly damaged both during operation and in the non-operated state by the action of liquid water and the leaching of alkali metal or alkaline earth metal ions.

The WO 2009/030547 A1 describes a method for determining the water or carbon dioxide concentration in a gas with a sensor element comprising a solid electrolyte body, a first electrochemical pump cell comprising a first inner pumping electrode arranged in a first inner gas space of the solid electrolyte body and a first inner pumping electrode on the side of the solid electrolyte facing the measurement gas first outer pumping electrode arranged on the rolyte body, the first inner pumping electrode and the first outer pumping electrode being connected to one another via a power source, and a second electrochemical pumping cell comprising a second inner pumping electrode arranged in a second inner gas space of the solid electrolyte body and one on the side of the measuring gas facing the Solid electrolyte body arranged second outer pump electrode, wherein the second inner pump electrode and the second outer pump electrode are connected to each other via a power source, in which the first pump cell is operated with a pump voltage that only leads to the reduction of molecular oxygen to O ^2- ions, and the second pump cell is operated with a higher pump voltage, which leads both to the reduction of molecular oxygen to O ^2- ions and to the splitting of water to O ^2- ions, and from the difference in the pump currents of the two pump cells and/or from from the pump currents res resulting voltage difference, the water concentration in the sample gas is determined; or the first pump cell is operated with a pump voltage that leads both to the reduction of molecular oxygen to O ^2- ions and to the splitting of water to O ^2- ions, and the second pump cell is operated with a higher pump voltage that both leads to the reduction of molecular oxygen to O ^2- ions and to the splitting of water to O ^2- ions and carbon dioxide to O ^2- ions, and from the difference in the pump currents of the two pump cells and/or from the pump currents The resulting voltage difference is used to determine the carbon dioxide concentration in the sample gas.

Despite the advantages of the methods known from the prior art for detecting gas components in a gas, these still contain potential for improvement. Until now, the detection of different gas components has been complex or only possible with several different sensors.

Disclosure of Invention

There is therefore a method for determining a gas concentration of a gas component in a gas using a sensor element, which largely avoids the disadvantages of known methods and devices for determining a gas concentration of a gas component in a gas. In particular, the advantages of a sensor element of a broadband lambda probe with regard to its possible uses, even in critical or difficult areas of use, should be combined with a reduction in the number of measuring units or measuring cells required for detecting different gas components.

• - Anlegen einer vorbestimmten Pumpspannung an die Pumpzelle, wobei die vorbestimmte Pumpspannung kleiner als eine Pumpspannung für einen Grenzstromfall ist,
• - Anlegen einer Mess-Wechselspannung an die Pumpzelle während des Anlegens der vorbestimmten Pumpspannung,
• - Messen einer Impedanz der Pumpzelle und
• - Bestimmen der Gas-Konzentration der Gaskomponente in dem Gas basierend auf der gemessenen Impedanz der Pumpzelle.

A method according to the invention for determining a gas concentration of a gas component in a gas by means of a sensor element, wherein the sensor element has at least one pump cell with at least two pump electrodes connected to one another by at least one solid electrolyte, wherein a first pump electrode can be exposed to the gas and a second pump electrode can be exposed to at least one porous diffusion barrier is separated from the gas and arranged in a cavity in the solid electrolyte, which comprises the following steps, preferably in the order given:

• - applying a predetermined pump voltage to the pump cell, the predetermined pump voltage being smaller than a pump voltage for a limit current case,
• - applying a measurement AC voltage to the pump cell while applying the predetermined pump voltage,
• - measuring an impedance of the pump cell and
• - determining the gas concentration of the gas component in the gas based on the measured impedance of the pump cell.

In this case, the measuring AC voltage can in particular have an effective value (rms—root mean square) of 30 mV. In principle, the effective value of the AC voltage can preferably be between 10 mV and 100 mV, particularly preferably 30 mV.

The method according to the invention has the advantage of using a sensor element with only a single pump cell to determine a concentration of gas components that differ from oxygen. The measuring principle is based on the fact that, in addition to the applied pump voltage, which is a DC voltage, as is known to those skilled in the art, an AC voltage is applied to the pump cell and an impedance of the pump cell is measured. The operating point is set in such a way that both oxygen and the gas component that is of interest are dissociated at the inner pumping electrode. This prevents damage to the pumping electrode, for example due to premature aging. The measured impedance, which corresponds to diffusion-driven processes, depends on the concentration of the respective gas component. By comparing a measurement of a pump current in the limit current operation, which gives the oxygen content, the concentration of the respective gas component can be deduced.

In this case, the concentration can be described as a qualitative and/or quantitative detection of a proportion of a gas component of the measurement gas, in particular with reference to a detection of an oxygen proportion in the measurement gas. The oxygen content can be recorded, for example, in the form of a partial pressure and/or in the form of a percentage.

The sensor element can be a sensor element for a lambda probe, in particular a broadband lambda probe, as is known, for example, from Konrad Reif (ed.): Sensors in motor vehicles, 1st edition 2010, pp. 160-165. Such a sensor element has a solid electrolyte and two electrodes separated from one another by the solid electrolyte. The solid electrolyte can be formed from one or more solid electrolyte layers. The sensor element also has at least one pump cell.

A broadband lambda sensor is usually made up of several layers using planar technology and has an integrated heater. Three parts are decisive for the measuring principle: the pump cell between the exhaust gas and the measuring gap/measuring gas, the diffusion channel leading through the pump cell between the exhaust gas and the measuring gas and the Nernst cell between the measuring gas and the reference gas. The oxygen content of the measuring gas in the measuring gap is determined on the one hand by the exhaust gas, which acts through a diffusion channel, and on the other hand is influenced by the current flow of the pump cell. Depending on the polarity of the applied pump voltage, the pump current pumps oxygen from the exhaust gas side of the zirconium membrane into or out of the measuring gap. The pump current is regulated by an external controller in such a way that the lambda value in the sample gas exactly balances the oxygen flow through the diffusion channel and keeps the sample gas in the measuring gap constant at λ = 1. A lambda value of 1 is always given when the voltage at the Nernst cell is 0.45 V. In the case of a rich mixture, the pump current pumps oxygen ions into the measuring gas in the measuring gap and out in the case of a lean mixture. The exhaust gas lambda can be determined via the sign and magnitude of this current. The current is regulated by a separate control chip in the engine control unit.

The application of the AC measurement voltage induces an AC measurement current through the pump cell. The impedance can be measured as a resistance to the induced measuring alternating current. This increases the measurement accuracy.

The method may further include varying a level of the predetermined pumping voltage depending on the type of gas component to be determined. This allows different gas components to be detected.

The magnitude of the pumping voltage can be set in such a way that it is greater than a pumping voltage required for the start of reduction or start of splitting of the gas component to be determined. This prevents overlapping by dissociation or splitting processes of other gas components.

The method to determine the concentration of several different gas components may include applying several different pumping voltages and measuring the impedance of the pumping cell at the different applied pumping voltages. As a result, the concentrations of different gas components in gas mixtures can also be determined using a single sensor element.

The magnitude of the applied pump voltages can increase. Accordingly, the gas components are detected in the order of their dissociation or decomposition reaction, so that the determination of the gas components is prevented from being overlapped.

The predetermined pump voltage can vary depending on the lambda and can be less than 1.2 V in air, for example, so that the onset of ZrO ₂ decomposition is not reached. This reliably prevents damage to the pumping electrode.

The sensor element can also have a Nernst cell, and the method can also include adjusting the Nernst cell to a lambda greater than 1.0. This increases the measurement accuracy, since the measurement is carried out in an oxygen-rich atmosphere.

A frequency of the measurement AC voltage can be from 0.05 Hz to 10 Hz, more preferably from 0.07 Hz to 5.0 Hz, and still more preferably from 0.08 Hz to 3.0 Hz. Since the impedance measurement is more sensitive in the lower frequency range, the measurement accuracy can be increased.

The gas component can be at least one gas component selected from the group consisting of water, hydrogen, carbon dioxide, carbon monoxide and nitrogen. This allows the most important gas components occurring during combustion processes to be recorded.

In the context of the present invention, a solid electrolyte is to be understood as meaning a body or object with electrolytic properties, ie with ion-conducting properties. In particular, it can be a ceramic solid electrolyte. This also includes the raw material of a solid electrolyte and therefore the formation of a so-called green body or brown body, which only becomes a solid electrolyte after sintering. According to this definition, a solid electrolyte layer is a body or object with electrolytic properties that is formed as a layer.

In the context of the present invention, an electrode is generally to be understood as an element which is able to contact the solid electrolyte in such a way that a current can be maintained through the solid electrolyte and the electrode. Accordingly, the electrode can comprise an element on which the ions can be incorporated into the solid electrolyte and/or removed from the solid electrolyte. Typically, the electrodes include a noble metal electrode, which can be applied, for example, as a metal-ceramic electrode or on the solid electrolyte or can be connected to the solid electrolyte in some other way. Typical electrode materials are platinum cermet electrodes. However, other noble metals, such as gold or palladium, can also be used in principle.

The pump voltage for a limiting current drop can be understood as the pump voltage applied to the pump cell from which no or only an insignificant increase in the pump current can be detected even with a further increase in the pump voltage, since no further oxygen molecules are reduced at the inner pump electrode. The pumping current that corresponds to this plateau of the characteristic is called the limiting current and is proportional to the concentration of oxygen or oxygen-containing gas species in a gas mixture. In normal operation of a broadband lambda probe, the actual measurement of the oxygen content takes place in the limiting current case.

All concentration data in the context of the present disclosure are given in percent by volume, unless something else is explicitly described.

The diffusion barrier can be understood, for example, as a layer made of a material which suppresses a flow of the gas and/or a fluid and/or the gas mixture and/or the gas component, while the layer prevents a diffusion of the gas and/or the fluid and/or of the gas mixture and/or the gas component and/or ions.

The cavity can be understood to mean a space within the sensor element which is structurally separate from the measurement gas space, but which can nevertheless be acted upon by the gas component and/or the gas mixture and/or the gas from the measurement gas space, for example via at least one gas access path and /or across the diffusion barrier. The cavity can be a chamber, for example. The device can include at least one reference gas space and/or at least one reference gas channel. The solid electrolyte can preferably be an ion-conducting solid electrolyte. A gas exchange, in particular of the gas and/or at least part of the gas, can preferably be possible via the diffusion barrier, in particular towards the cavity, preferably by diffusion.

The designations “first” and “second” serve as pure designations and in particular do not provide any information about a sequence or whether, for example, further pumping electrodes are included in the pumping electrodes, for example at least a third pumping electrode.

In a further aspect of the invention, a device for determining a gas component of a gas is proposed. The device comprises at least one sensor element. In principle, the determination of the concentration of the gas component can involve a qualitative and/or quantitative determination of the proportion of the gas component.

The device includes at least one control. The control is set up to carry out the method according to the invention for determining a gas concentration of a gas component in a gas using the sensor element, as described above or below. The control and/or the device can have at least one data processing device. For example, the data processing device can be integrated in the control. For example, the data processing device can also be arranged at least partially separately from the control. The control and/or the data processing device can be and/or can be connected to the sensor element, for example.

Control can be understood to mean a device that is set up to support and/or control at least one function of the device, in particular of the sensor element. “Connectable” can be understood, for example, as a property in which an electrical connection can be made or already exists. The control can be completely or partially separate from the sensor element, but can also be fully or partially integrated into the sensor element, for example in at least one plug of the sensor element and/or the device. The control can include at least one voltage measuring device and/or at least one current measuring device for detecting at least one pump current and/or at least one pump voltage and/or for controlling the pump voltage and/or for controlling the pump current. Alternatively or additionally, the control and/or the sensor element and/or the device can have at least one application device. The application device can in particular comprise at least one voltage source and/or at least one current source. For example, the application device can be set up to apply the pump current and/or the pump voltage to the sensor element.

character list

Further optional details and features of the invention result from the following description of preferred exemplary embodiments, which are shown schematically in the figures.

• 1 ein Ausführungsbeispiel einer erfindungsgemäßen Vorrichtung;
• 2 ein Schaubild eines erfindungsgemäßen Verfahrens;
• 3 einen Ausschnitt des Sensorelements im Bereich der Pumpzelle;
• 4 Spannungs-Strom-Kennlinien eines Sensorelements einer Breitband-Lambdasonde für drei verschiedene Gase;
• 5A und 5B Impedanzspektren, die bei den drei Gasen bei zwei unterschiedlichen Pumpspannungen gemessen wurden;
• 6 Impedanzspektren, die bei einer angelegten vorbestimmten Pumpspannung U_P von 0,9 V gemessen wurden, für Gase, die Wasserstoff, Sauerstoff und Wasser enthalten;
• 7 Impedanzspektren, die bei einer angelegten vorbestimmten Pumpspannung U_P von 1,05 V gemessen wurden, für die zweiten und dritten Gase sowie für ein sechstes und siebtes Gas;
• 8A bis 8E Kennlinien für die Impedanzabhängigkeit von der Konzentration;
• 9A bis 9E Kennlinien für die Abhängigkeit des Phasenwinkels von der Konzentration; und
• 10 Impedanzspektren, die bei einer angelegten vorbestimmten Pumpspannung U_P von 1,05 V gemessen wurden, für das oben genannte zweite Gas und ein achtes Gas.

Show it:

• 1 an embodiment of a device according to the invention;
• 2 a diagram of a method according to the invention;
• 3 a section of the sensor element in the area of the pump cell;
• 4 Voltage-current characteristics of a sensor element of a broadband lambda probe for three different gases;
• 5A and 5B Impedance spectra measured for the three gases at two different pump voltages;
• 6 Impedance spectra measured at an applied predetermined pumping voltage U _P of 0.9 V for gases containing hydrogen, oxygen and water;
• 7 Impedance spectra, measured with an applied predetermined pump voltage U _P of 1.05 V, for the second and third gases and for a sixth and seventh gas;
• 8A until 8E Characteristic curves for the dependence of the impedance on the concentration;
• 9A until 9E Characteristic curves for the dependence of the phase angle on the concentration; and
• 10 Impedance spectra measured with an applied predetermined pumping voltage U _P of 1.05 V for the above second gas and an eighth gas.

Embodiments of the invention

In 1 an exemplary embodiment of a device 110 according to the invention is shown. Device 110 according to the invention for detecting at least a portion of a gas 112 in a measurement gas chamber comprises at least one sensor element 114. Sensor element 114 has at least one pump cell 116 with at least two pump electrodes 120 connected to one another by at least one solid electrolyte 118. The sensor element 114 can include at least one heating element 122 . Device 110 includes at least one control 124. Control 124 is set up to carry out the method according to the invention for adjusting sensor element 114. The device 110 and/or the Driver 124 may include at least one data processing device 126 . Sensor element 114 also includes at least one porous diffusion barrier 128 and at least one cavity 130. Sensor element 114 can be connected to control unit 124 and/or to data processing device 126 via at least one interface 132. For example, a first pumping electrode 134 can be exposed to the gas 112 and a second pumping electrode 136 is separated from the gas 112 by at least one porous diffusion barrier 128 and is arranged in the cavity 130 . In this case, the first pumping electrode 134 is covered by a protective layer 138 . As shown, cavity 130 can be a space within sensor element 114 which is structurally separate from the measurement gas space, but which can nevertheless be charged with the gas component and/or the gas mixture and/or the gas from the measurement gas space, for example via at least one gas access path or a gas access hole 140 and/or via the diffusion barrier 128. Due to the described spatial arrangement of the first pumping electrode 134 and the second pumping electrode 136, the first pumping electrode 134 can also be referred to as the outer pumping electrode or outer pumping electrode and the second pumping electrode 136 as the inner pumping electrode or inner pumping electrode .

Optionally, the sensor element 114 also includes a reference electrode 142. The reference electrode 142 can be arranged at least partially in at least one reference gas channel 144, for example. The reference gas channel 144 can be at least partially filled with at least one porous medium. Furthermore, the sensor element 114 can comprise at least one further electrode, for example further electrodes, which in principle can be designed like a pump electrode, whereby the further electrodes can also be, for example, at least one Nernst electrode 146, i.e. an electrode which can detect a Nernst potential , for example in combination with the reference electrode 142. The Nernst electrode 146, the reference electrode 142 and the solid electrolyte 118 can form a Nernst cell 147 of the sensor element 114.

Furthermore, sensor element 114 is at least partially surrounded by a thermal shock protection layer 148 . The thermal shock protection layer 148 also covers the protective layer 138.

A method according to the invention is based on the 2 described. In step S10, the method includes applying a predetermined pump voltage to the pump cell, the predetermined pump voltage being less than a pump voltage for a limit current drop. In step S12, the method includes applying a measurement AC voltage to the pump cell while applying the predetermined pump voltage. In step S14, the method includes measuring an impedance of the pump cell. In step S16, the method includes determining the gas concentration of the gas component in the gas based on the measured impedance of the pump cell.

The basic principle of the method according to the invention is explained in more detail below 3 described. 3 shows a section of sensor element 114 in the region of pump cell 116. The method for determining the concentration is described using a water molecule (H ₂ O) as an example. As described above, the predetermined pump voltage U _P is applied to the pump cell 116 . The pump voltage U _P can be supplied by a current and/or voltage source 149 . The current and/or voltage source 149 can be part of the control 124 or can be connected to it. Analogously to oxygen, the water molecule reaches cavity 130 of sensor element 114, where it is adsorbed on the surface of negatively polarized second pump electrode 136 due to the predetermined pump voltage applied. The predetermined pump voltage U _P is slightly higher than the pump voltage at which the reduction of water begins, so that the water molecule is reduced or dissociated into a hydrogen molecule (H ₂ ) and an oxide ion (O ^2- ). The hydrogen molecule diffuses out of the cavity 130 again and the oxide ion is pumped to the first pumping electrode 134 by the pumping current I _P . In addition to the predetermined pump voltage U _{P ,} the measuring AC voltage U _z is applied to the pump cell 116 by means of the current and/or voltage source 149 . The application of the AC measurement voltage U _z induces an AC measurement current through the pump cell _116. The impedance Z of the pump cell 116 is measured as resistance to the AC measurement current I z induced by the AC measurement voltage U _z applied in addition to the pump voltage U _P .

The pump current I _P corresponding to the flow of oxide ions of the measured gas is very small and its sensitivity to the measured gas is weak, since the pump voltage U _P at the beginning of the reduction threshold is far from the pump voltage required to reach the limit current to the Pumping the gas is necessary is set. On the other hand, the impedance is much more sensitive at this potential level and the concentration dependence is measurable in this way. There are two advantages to using the impedance measurement with the previously described pump voltage settings. If Gases such as water are pumped at high enough potential to reach the limiting current, additional electrochemical processes with slow kinetics take place, which could lead to accelerated electrode aging. Setting the predetermined pumping voltage below the pumping voltage for a current limit case thus protects the pump cell 116. Additionally, gases such as water and carbon dioxide cannot be measured separately using the current limit because their reduction thresholds overlap. However, the onset potential for the reduction of water is about 100 mV lower than that for carbon dioxide. Therefore, by setting the operating point to a value between these two initial voltages, the impedance signal from water can be distinguished from the combined impedance signal from the two gases.

4 shows voltage-current characteristics of a sensor element 114 of a broadband lambda probe for three different gases measured at 700°C. The pump voltage U _P in V applied to the pump cell 116 is plotted on the X-axis 150 . The associated pump current I _P in A is plotted on the Y axis 152 . The first gas contains 4.5% oxygen, 2% water and 93.5% nitrogen. The second gas contains 4.5% oxygen, 3% carbon dioxide and 92.5% nitrogen. The third gas contains 4.5% oxygen and 95.5% nitrogen. With 154 is in 4 the voltage-current characteristic of the first gas, 156 the voltage-current characteristic of the second gas and 158 the voltage-current characteristic of the third gas. The characteristic curves 154, 156, 158 show one or two reduction stages. A first stage reduction occurs for all gas mixtures between about 0.0 to 0.9 V and results from the reduction of molecular oxygen and the pumping of that oxygen by pump cell 116. The pumping current at the plateau of the first stage reduction is proportional to the concentration of oxygen . The second reduction stage, beginning at 0.86 V for the first gas, as indicated by arrow 160, and at 0.93 V for the second gas, as indicated by arrow 162, is the reduction of water in the case of of the first gas and carbon dioxide in the case of the second gas and pumping the oxide ions removed from the molecules of these gases. The current measured at the plateau of the second reduction stage is proportional to the total concentration of oxygen and water in the case of the first gas and oxygen and carbon dioxide in the case of the second gas. The increase in current above the level corresponding to the pumping voltage of about 1.2 V in the case of the third gas, together with the hysteresis present in all three curves, means that electrochemical processes other than the reduction of gases at the electrode interface occur at higher voltages take place. These processes presumably involve the electrode or the electrolyte itself (e.g. ZrO ₂ decomposition) and can lead to accelerated electrode aging. The method according to the invention prevents the risk of irreversible electrode aging caused by repeated application of high pump voltages to the pump cell 116 by the impedance spectroscopy measurement in connection with a pump DC voltage applied to the pump cell 116. As a result, the sensor element 114 is in operation, ie oxide ion pumping, while the impedance measurement is being carried out.

As previously described, the predetermined pumping voltage applied to the pumping cell 116 is slightly higher than the initial voltage of the second reduction stage. This allows the total voltage applied to the pump cell 116 to be kept below the 1.2 V limit and the difference between the signals from water and carbon dioxide can be seen, which would be difficult if voltages close to the upper plateau of the current Voltage characteristics would correspond due to the overlapping of the reduction thresholds of water and carbon dioxide. While the pumping current measured for the first gas, which is slightly above the reduction of water (or carbon dioxide), is very close to the pumping current for oxygen reduction, the impedance measurement is much more sensitive in this voltage range.

The 5A and 5B show impedance spectra measured for the three gases at two different pump voltages. The real impedance ReZ in Ω is plotted on the x-axis 150 in each case. The imaginary impedance ImZ in Ω is plotted on the Y-axis 152 in each case. In each case, 164 indicates the impedance characteristic of the first gas, 166 indicates the impedance characteristic of the second gas and 168 indicates the impedance characteristic of the third gas. For 5A the impedance was measured when a predetermined pump voltage U _P of 0.9 V was applied. For 5B the impedance was measured when a predetermined pump voltage U _P of 1.0 V was applied. At both predetermined pumping voltages U _P , all of the oxygen that has entered the cavity 130 is pumped out immediately, causing a steep increase in the impedance measured in a low frequency range of approximately 3 Hz. With an applied predetermined pump voltage U _P of 0.9 V in 5A water is reduced at the second pumping electrode 136 in addition to oxygen and this new source of oxide ions reduces the impedance or more precisely the diffusion resistance part of the impedance in the low frequency range. On the other hand, carbon dioxide cannot with an applied predetermined pump voltage U _P of 0.9 V be reduced. Therefore, the impedance spectrum 164 of the first gas differs from the spectra 166, 168 of the second and third gases. When a predetermined pump voltage U _P of 1.0 V is applied, both water and carbon dioxide are reduced at the second pump electrode 136 and the spectra 164, 166 of the first and second gas deviate from the reference spectrum 168 of the third gas.

In this way, a concentration of water and carbon dioxide can be measured when they are mixed with oxygen. In a mixture containing all three gas components, the signal from water can be separated from the cumulative signal from water and carbon dioxide by first using the lower predetermined pumping voltage U _P , which is slightly above the start of water reduction, and then the higher predetermined pumping voltage U _P , which is slightly above the start of carbon dioxide reduction.

With the presently disclosed method it is also possible to measure the concentration of reducing gases present as minor components in mixtures with oxygen. 6 12 shows impedance spectra measured with an applied predetermined pump voltage U _P of 0.9 V for gases containing hydrogen, oxygen and water. The real impedance ReZ in Ω is plotted on the X-axis 150 . The imaginary impedance ImZ is plotted in Ω on the Y-axis 152 . 164 indicates the impedance characteristic of the first gas mentioned above, 168 the impedance characteristic of the third gas mentioned above, 170 the impedance characteristic of a fourth gas and 172 the impedance characteristic of a fifth gas. The fourth gas was prepared from the third gas by adding 0.5% hydrogen to the mixture, followed by adding an appropriate amount of oxygen to keep lambda constant at 1.3, which is the lambda value of the first through third gases. The fifth gas was prepared from the first gas by adding 0.5% hydrogen to the mixture, followed by an appropriate amount of oxygen to maintain a constant lambda of 1.3. In this case, hydrogen can be clearly detected by impedance measurement both in the fourth and fifth gas, which means that the signals or characteristic curves 168, 170 are additive. The signal corresponding to hydrogen is not overwritten by the signal for water.

Analogous results are found for carbon monoxide and carbon dioxide 7 shown. 7 shows impedance spectra, which were measured with an applied predetermined pump voltage U _P of 1.05 V, for the above-mentioned second and third gases and for a sixth and seventh gas. The real impedance ReZ in Ω is plotted on the X-axis 150 . The imaginary impedance ImZ is plotted in Ω on the Y-axis 152 . 166 indicates the impedance characteristic of the above-mentioned second gas, 168 the impedance characteristic of the above-mentioned third gas, 174 the impedance characteristic of the sixth gas and 176 the impedance characteristic of the seventh gas. The sixth gas was prepared by adding 0.5% carbon monoxide to the third gas, followed by adding an appropriate amount of oxygen to maintain the constant lambda of 1.3. The seventh gas was prepared by adding 0.5% carbon monoxide to the second gas, again followed by the appropriate amount of oxygen to maintain the constant lambda of 1.3. The signals are also additive in the case of carbon monoxide and carbon dioxide. However, a comparison of 6 and 7 that the measurement sensitivity for hydrogen is higher than for carbon monoxide.

The reason it is possible to detect both hydrogen and carbon monoxide, even though these cannot be reduced to oxide ions by the applied pumping voltage, is because these gases react with oxygen at the platinum-containing pumping electrodes 134, 136 to achieve equilibrium to reach: CO + ½ _{O2 ↔CO2} (1) _H2 + ½ _O2 ↔ _H2O (2)

Water and carbon dioxide generated by this reaction are then reduced at the second pumping electrode 136 by the applied DC pumping voltage, resulting in the in FIGS 6 and 7 shown signals or characteristics 174, 176 leads.

It should be noted that the exemplary gas mixtures described above were measured when the Nernst cell 147 was regulated with a lambda greater than 1. More precisely, the gas mixtures were measured with a constant lambda of 1.3, which means that addition of a reducing gas was always accompanied by addition of a corresponding amount of oxygen according to equations (1) and (2) above. The percentage of unreacted oxygen present in all specified gas mixtures, ie the compensated percentage as measured by an oxygen sensor was is constant. This is the correct approach, reflecting operating conditions in combustion processes where constant lambda control is the essential control measure.

In principle, the method according to the invention can be used to determine any gas component which is reduced to oxide ions in the safe potential range, ie at a voltage below the upper end of the plateau for oxygen reduction and below the start of the reduction stage of slow kinetic ZrO ₂ decomposition or which is oxidized to molecules which in turn are reducible under these conditions.

It is clear from the above explanations that the method according to the invention can be used to detect a wide range of gases using a conventional broadband lambda sensor. In order to evaluate the suitability of the method for also quantitatively measuring a concentration of the gas component, the concentration dependence of certain components on the complex impedance at a single frequency must be analyzed. Since the greatest effect of different gases on the impedance spectra can be seen in the low frequency range, where the diffusion-driven processes are manifested, the frequency of 0.1 Hz was chosen as the operating point here. Except for the real impedance part ReZ and the imaginary impedance part ImZ, which are included in the 5A until 7 plotted above, the complex impedance in the polar form can be represented by the magnitude |Z| (distance of the point from the origin in the complex plane) and the phase angle Φ (tangent of the angle is equal to the imaginary part of the impedance divided by the real part).

Table 1 summarizes concentration dependencies of all four terms measured at a frequency of 0.1 Hz for hydrogen, carbon monoxide, carbon dioxide, hydrogen over a constant background of water at 2%, and carbon monoxide over a constant background of carbon dioxide at 3%. Table 1 gas Up/V concentration / % ReZ / Ω ImZ/Ω |Z|/Ω Φ/Ω a) _H2 0.9 0 1246 -3413 3633 -69.9 0.02 1227 -3324 3543 -69.7 0.05 1355 -3169 3447 -66.9 0.1 1693 -2693 3181 -57.8 0.5 1021 -296 1064 -16.1 1 575 -85 582 -8.4 b) CO 1:15 0 2411 -4675 5260 -62.7 0.04 2464 -4852 5442 -63.1 0.06 2472 -4923 5509 -63.3 0.1 2484 -4981 5565 -63.5 0.2 2611 -5039 5675 -62.6 0.3 2693 -5128 5793 -62.3 0.4 2778 -5096 5804 -61.4 0.5 2939 -5040 5834 -59.7 c) _CO2 1:15 2 2606 -948 2773 -20.0 2.3 1998 -501 2060 -14.1 2.5 1754 -364 1791 -11.7 2.8 1486 -246 1506 -9.4 3 1420 -219 1437 -8.8 3.3 1286 -170 1297 -7.5 3.5 1211 -146 1219 -6.9 4 1073 -109 1079 -5.8 d) H ₂ with constant background of 2% H ₂ O 0.9 0 1668 -928 1909 -29.1 0.02 1592 -823 1792 -27.3 0.05 1473 -673 1619 -24.5 0.1 1303 -502 1396 -21.1 0.5 668 -114 677 -9.7 1 462 -51 465 -6.3 e) CO with a constant background of 3% CO ₂ 1:15 0 1460 -233 1478 -9.1 0.04 1412 -213 1428 -8.6 0.06 1395 -207 1410 -8.4 0.1 1359 -195 1373 -8.1 0.2 1321 -181 1334 -7.8 0.3 1262 -162 1272 -7.3 0.4 1217 -148 1226 -6.9 0.5 1176 -137 1184 -6.7

Table 1: Dependencies of the four complex impedance terms on the concentration of a) hydrogen, b) carbon monoxide, c) carbon dioxide, d) hydrogen over constant background of water (2%) and e) carbon monoxide over constant background of carbon dioxide (3%)

The dependencies of the magnitude and the phase angle are also shown graphically in FIGS 8A until 8th and 9A until 9E shown. Analogously to the first to seventh gases, all gas mixtures were obtained with a constant lambda value with 4.5% oxygen content measured by the reference lambda probe. The impedance measurements were carried out with an applied predetermined pump voltage _UP of 0.9 V for hydrogen and water and with an applied predetermined pump voltage _UP of 1.15 V for carbon monoxide and carbon dioxide.

The 8A until 8E show characteristic curves for the dependence of the impedance on the concentration. On the X-axis 150 is magnitude |Z| of the impedance in Ω. The concentration in % is plotted on the Y-axis 152 . With 178 is in 8A the impedance magnitude characteristic of hydrogen is shown. With 180 is in 8B the impedance magnitude characteristic of carbon monoxide is shown. With 182 is in 8C the impedance magnitude characteristic of carbon dioxide is shown. With 184 is in 8D the impedance magnitude characteristic of hydrogen is shown against the water background. With 186 is in 8E the impedance magnitude characteristic of carbon monoxide is shown against the carbon dioxide background.

The 9A until 9E show characteristic curves for the dependence of the phase angle on the concentration. The phase angle Φ of the impedance in ° (degrees) is plotted on the X-axis 150 . The concentration in % is plotted on the Y-axis 152 . With 188 is in 9A the phase angle characteristic of hydrogen is shown. With 190 is in 9B the phase angle characteristic of carbon monoxide is shown. With 192 is in 9C the phase angle characteristic of carbon dioxide is shown. With 194 is in 9D the phase angle characteristic of hydrogen is shown against the water background. With 196 is in 9E the phase angle characteristic of carbon monoxide is shown against the carbon dioxide background.

As in the 8A until 9E and Table 1, the complex impedance terms are non-linearly dependent on the concentration of all gases under study. It is interesting that the concentration dependencies of the amount ( 8A until 8E ) have convex character in the case of water, hydrogen and carbon dioxide, while they have concave character in the case of carbon monoxide in the measured concentration range. In fact, the curve 178, 188 measured for hydrogen appears to change curvature from concave at very low concentrations to convex at higher concentrations. The shape of the curve is likely affected by a combination of diffusion rate and concentration of the gas in question, with increases in these amounts resulting in a more convex shape of the curve. The concentration dependence of carbon monoxide measured over a Hin The background of carbon dioxide supports this conclusion since the relatively high constant level of carbon dioxide reduces the impedance magnitude while changing the shape of the curve measured for carbon monoxide. The 9A until 9E show that the concentration dependencies of phase angle follow trends in their concave shape in the opposite direction to the concentration dependencies of magnitude, and that increasing gas diffusion rate and concentration leads to a more concave shape of the curves.

It could thus be shown above that the concentration dependencies for different reducible and oxidizable gases can be measured when the gas component of interest is the only other component in the gas mixture besides oxygen and nitrogen or when a known amount of another measurable gas is present present. The situation becomes more complicated when the mixture consists of a combination of gases in unknown concentrations.

The difference between the sensitivity of the measurement to the concentration of carbon monoxide and carbon dioxide is in 10 shown. 10 Figure 12 shows impedance spectra measured with an applied predetermined pumping voltage U _P of 1.05 V for the above second gas and an eighth gas. The eighth gas contained 2.5% carbon dioxide and 0.5% carbon monoxide. The total carbon content remained the same as in the second gas. The oxygen content in the eighth gas was adjusted to maintain the constant lambda of 1.3 as with the other gases.

The real impedance ReZ in Ω is plotted on the X-axis 150 . The imaginary impedance ImZ is plotted in Ω on the Y-axis 152 . 166 in each case means the impedance characteristic of the second gas mentioned above, and 198 means the impedance characteristic of the eighth gas mentioned above. The total carbon concentration contained in carbon dioxide and carbon monoxide is the same for both gas mixtures, which means that if the sensitivity of the method for both carbon monoxide and carbon dioxide were identical, there would be no difference in the signal 166 of the second gas and in the signal 198 of the eighth gas should give. 10 shows that the sensitivity of the method is actually different for carbon monoxide and carbon dioxide. Since the difference is visible in the low frequency range of the spectra, it is mainly attributed to the difference between the diffusion rates of carbon dioxide and carbon monoxide. Since the total carbon content was the same for both gases, the CO ₂ concentration in the eighth gas is the same as the CO ₂ concentration in the second gas, even if the CO to CO ₂ in the eighth gas is completely converted. A different CO ₂ concentration is therefore ruled out as an explanation for the different impedance spectra.

However, the difference in the diffusion rates of carbon monoxide and carbon dioxide and in the respective sensitivities is not large enough to allow a reliable measurement with a conventional oxygen sensor when both gases are present in unknown concentrations. This means that the measurement is limited to cases where only one of these gases is present in an unknown concentration.

A larger difference in sensitivities is likely to be seen in the case of an analogous pair of gas mixtures containing water and hydrogen, since the difference between the diffusion rates of water and hydrogen is much larger than that of carbon monoxide and carbon dioxide. In this case, the simultaneous measurement of these two gases is likely to be more reliable. In addition, the water/hydrogen pair can be distinguished from the carbon monoxide/carbon dioxide pair since the former pair is measured using a lower DC pumping voltage at which the latter pair is not detected at all, as in FIGS 4 and 5 is shown.

QUOTES INCLUDED IN DESCRIPTION

This list of documents cited by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent Literature Cited

• WO 2009030547 A1 [0007]

Claims (10)

Method for determining a gas concentration of a gas component in a gas by means of a sensor element (114), the sensor element (114) having at least one pump cell (116) with at least two pump electrodes (134, 136) connected to one another by at least one solid electrolyte (118). , wherein a first pumping electrode (134) is exposable to the gas and a second pumping electrode (136) is separated from the gas by at least one porous diffusion barrier (128) and is arranged in a cavity (130) in the solid electrolyte (118), the method comprising : - application of a predetermined pump voltage ( _UP ) to the pump cell (116), the predetermined pump voltage ( _UP ) being smaller than a pump voltage for a limit current case, - application of a measuring AC voltage (U _z ) to the pump cell (116) during the application of the predetermined pump voltage ( _UP ), - measuring an impedance (Z) of the pump cell (116) and - determining the gas concentration of the gas component in the gas end on the measured impedance (Z) of the pump cell (116). Method according to the preceding claim, wherein the application of the measuring AC voltage (U _z ) induces a measuring AC current through the pump cell (116), the impedance (Z) being measured as resistance to the induced measuring AC current. Method according to one of the preceding claims, further comprising varying a level of the predetermined pump voltage ( _UP ) depending on the type of gas component to be determined. Method according to the preceding claim, wherein the magnitude of the pump voltage ( _UP ) is set such that it is greater than a pump voltage ( _UP ) required for the start of reduction or start of splitting of the gas component to be determined. Method according to one of the two preceding claims, wherein the method for determining the concentration of several different gas components comprises applying several different pump voltages ( _UP ) and measuring the impedance (Z) of the pump cell (116) at the different applied pump voltages ( _UP ). Method according to one of the two preceding claims, in which the level of the pump voltages ( _UP ) applied increases. Method according to one of the preceding claims, wherein the predetermined pump voltage (U _P ) is less than 1.2 V. Method according to one of the preceding claims, wherein the sensor element also has a Nernst cell, wherein the method further comprises adjusting the Nernst cell to a lambda greater than 1.0. Method according to one of the preceding claims, wherein a frequency of the measuring AC voltage (U _z ) is from 0.05 Hz to 10 Hz, more preferably from 0.07 Hz to 5.0 Hz and even more preferably from 0.08 Hz to 3.0 Hz is. A method according to any one of the preceding claims, wherein the gas component is at least one gas component selected from the group consisting of water, hydrogen, carbon dioxide, carbon monoxide and nitrogen.

Images