Gas Sensors

Abstract

This gas sensor 100 is provided with a sensor element 101 and a control device. The control device performs a heater control process of energizing a heater 72 and causing the heater 72 to generate heat. The control device performs, after the heater control process has been started, a normality control process including a main pump control process of controlling a main pump cell 21, an auxiliary pump control process of controlling an auxiliary pump cell 50 so that voltage V1 reaches a target value, and a normality measurement pump control process of controlling a measurement pump cell 41 so that voltage V2 reaches a normality target value and pumping out oxygen in a third inner void 61. The control device detects the concentration of a specific gas in a gas subject to measurement, on the basis of a pump current Ip2 flowing through the measurement pump cell 41 due to the normality measurement pump control process. During an initial phase of the normality control process, the control device performs a corrective process of correcting the target value for the voltage V1 to a value higher than that in a period after the initial phase.

Description

[Technical field]

[0001] The present invention relates to a gas sensor. [Background technology]

[0002] Conventionally, gas sensors that detect the concentration of a specific gas such as NOx in a measurement gas such as an automobile exhaust gas are known. For example, Patent Document 1 describes a gas sensor that includes a sensor element having a plurality of oxygen ion conductive solid electrolyte layers, an outer pump electrode arranged on the outside of the sensor element, an inner main pump electrode, an inner auxiliary pump electrode, and a measurement electrode arranged in a measurement gas flow section inside the sensor element, and a reference electrode and a heater arranged inside the sensor element. In this gas sensor, the specific gas concentration is detected as follows in a state in which the heater adjusts the entire sensor element to a temperature at which the solid electrolyte is activated. First, the oxygen concentration in the measurement gas around the inner main pump electrode is adjusted by a control voltage applied between the outer pump electrode and the inner main pump electrode. Next, the oxygen concentration in the measurement gas around the inner auxiliary pump electrode is adjusted by a control voltage applied between the outer pump electrode and the inner auxiliary pump electrode. As a result, the measurement gas after the oxygen concentration is adjusted reaches the vicinity of the measurement electrode. The specific gas in the measurement gas after the oxygen concentration is adjusted is reduced around the measurement electrode, and the specific gas concentration in the measurement gas is detected based on the pump current that flows when the oxygen generated by the reduction around the measurement electrode is pumped out. The control voltage applied between the outer pump electrode and the inner auxiliary pump electrode is controlled so that the electromotive force (voltage V1) between the inner auxiliary pump electrode and the reference electrode becomes a target value. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2016-166871 A Summary of the Invention [Problem to be solved by the invention]

[0004] In such a gas sensor, the temperature difference between the inner auxiliary pump electrode and the reference electrode may temporarily increase due to the difference in the temperature rise rate between the two electrodes from when the heater starts to heat the sensor element until the temperature of each electrode becomes stable. When the temperature difference between the two electrodes is temporarily large, the thermoelectromotive force contained in the voltage between the two electrodes (called the auxiliary pump voltage) also temporarily increases, so that the difference between the auxiliary pump voltage and the electromotive force based on the oxygen concentration around the inner auxiliary pump electrode temporarily increases. Therefore, even if the auxiliary pump voltage is controlled to be the target value, the oxygen concentration around the inner auxiliary pump electrode may not be accurately controlled. This may result in a decrease in the detection accuracy of the specific gas concentration.

[0005] The present invention has been made to solve such problems, and has as its main object to accurately control the oxygen concentration around the inner auxiliary pump electrode when the thermoelectromotive force contained in the auxiliary pump voltage is temporarily large. [Means for solving the problem]

[0006] In order to achieve the above-mentioned main object, the present invention adopts the following means.

[0007] The gas sensor of the present invention comprises: an element body including an oxygen ion conductive solid electrolyte layer and having a measurement gas flow section therein through which a measurement gas is introduced and flows; a measurement pump cell including an outer measurement electrode provided on the outer side of the element body so as to come into contact with the measurement gas, and an inner measurement electrode disposed in a measurement chamber of the measurement gas flow portion, the measurement pump cell pumping oxygen from a periphery of the inner measurement electrode to a periphery of the outer measurement electrode; a reference electrode disposed inside the element body so as to come into contact with a reference gas serving as a reference for detecting a specific gas concentration in the measurement gas; a measurement voltage detection sensor cell for detecting a measurement voltage between the reference electrode and the inner measurement electrode; a main pump cell for adjusting an oxygen concentration in a first internal space provided upstream of the measurement chamber in the measurement gas flow portion; an auxiliary pump cell including an inner auxiliary pump electrode disposed in a second internal space provided in the measurement gas flow portion downstream of the first internal space and upstream of the measurement chamber, the auxiliary pump cell adjusting the oxygen concentration in the second internal space; an auxiliary pump voltage detection sensor cell for detecting an auxiliary pump voltage between the reference electrode and the inner auxiliary pump electrode; A heater for heating the element body; A sensor element having a heater control unit that performs a heater control process to energize the heater to generate heat; a pump cell control unit that performs normal control processes after the heater control process is started, the normal control processes including a main pump control process that controls the main pump cell, an auxiliary pump control process that controls the auxiliary pump cell so that the auxiliary pump voltage becomes a target value, and a normal measurement pump control process that controls the measurement pump cell so that the measurement voltage becomes a normal target value to pump out oxygen from the measurement chamber; a specific gas concentration detection unit that detects a specific gas concentration in the measurement target gas based on a measurement pump current flowing through the measurement pump cell by the normal measurement pump control process; Equipped with the pump cell control unit performs a correction process to correct the target value of the auxiliary pump voltage to a higher value in an initial period of the normal control process as compared with a period after the initial period. It is something.

[0008] In this gas sensor, in the normal control process performed after the heater control process is started, a main pump control process for controlling the main pump cell, an auxiliary pump control process for controlling the auxiliary pump cell so that the auxiliary pump voltage becomes a target value, and a normal measurement pump control process for controlling the measurement pump cell so that the measurement voltage becomes the normal target value to pump out oxygen from the measurement chamber are performed. Then, the specific gas concentration in the measurement gas is detected based on the measurement pump current flowing through the measurement pump cell by the normal measurement pump control process. In addition, a correction process is performed in the early stage of the normal control process to correct the target value of the auxiliary pump voltage in the auxiliary pump control process to a higher value compared to a period after the early stage. In this way, by increasing the target value of the auxiliary pump voltage in the early stage of the normal control process, it is possible to reduce the influence of the thermoelectromotive force between the inner auxiliary pump electrode and the reference electrode (i.e., the thermoelectromotive force included in the auxiliary pump voltage) being temporarily large. As a result, compared to the case where the correction process is not performed, the oxygen concentration around the inner auxiliary pump electrode can be accurately controlled when the thermoelectromotive force included in the auxiliary pump voltage is temporarily large.

[0009] In the gas sensor of the present invention, the pump cell control unit may correct the target value of the auxiliary pump voltage in the correction process based on a predetermined correction pattern such that the correction amount tends to decrease as time passes. When the sensor element is heated by a heater, the temperature difference between the inner auxiliary pump electrode and the reference electrode tends to decrease as time passes. Therefore, the thermoelectromotive force included in the auxiliary pump voltage also tends to decrease as time passes. By correcting the target value of the auxiliary pump voltage based on a predetermined correction pattern such that the correction amount tends to decrease as time passes, it is possible to reduce the influence of a temporarily large thermoelectromotive force included in the auxiliary pump voltage. Therefore, the oxygen concentration around the inner auxiliary pump electrode can be controlled with higher accuracy.

[0010] Here, "the tendency for the correction amount to decrease as time passes" includes, for example, cases where the correction amount decreases linearly, cases where the correction amount decreases curvilinearly, and cases where the correction amount decreases step-function-like.

[0011] The gas sensor of the present invention includes a reference voltage detection sensor cell for detecting a reference voltage between a measurement gas side electrode provided outside the element body so as to be in contact with the measurement gas and the reference electrode, the main pump control process is a process for controlling the main pump cell so that an auxiliary pump current flowing through the auxiliary pump cell by the auxiliary pump control process becomes a target current, and the pump cell control unit may correct a target value of the auxiliary pump voltage in the correction process in such a manner that the correction amount increases as the deviation between a calculated reference voltage derived based on the main pump current and corresponding to the oxygen concentration in the measurement gas around the sensor element and an actual reference voltage detected by the reference voltage detection sensor cell increases when the main pump current flowing through the main pump cell during the main pump control process is deemed stable. Here, the main pump current flowing when adjusting the oxygen concentration in the first internal space by the main pump control process is correlated with the oxygen concentration in the measurement gas around the sensor element. In addition, since the reference voltage is the voltage between the measured gas side electrode and the reference electrode on the outside of the element body, it is correlated with the oxygen concentration in the measured gas around the sensor element. Therefore, when the main pump current can be considered stable, it is possible to derive the calculated reference voltage, which is the reference voltage corresponding to the oxygen concentration in the measured gas around the sensor element, based on the main pump current. On the other hand, since the measured reference voltage, which is the reference voltage actually detected by the reference voltage detection sensor cell, also includes the thermoelectromotive force between the measured gas side electrode and the reference electrode, when the temperature difference between the two electrodes is temporarily large, the thermoelectromotive force included in the measured reference voltage also temporarily becomes large. Therefore, the larger the thermoelectromotive force included in the measured reference voltage, the larger the deviation between the calculated reference voltage derived based on the main pump current and the measured reference voltage actually detected by the reference voltage detection sensor cell. In addition, when the thermoelectromotive force between the measured gas side electrode and the reference electrode is large, the thermoelectromotive force between the inner auxiliary pump electrode and the reference electrode is often also large. Therefore, by correcting the target value of the auxiliary pump voltage in the correction process so that the correction amount tends to increase as the deviation between the calculated reference voltage and the measured reference voltage increases, the effect of the thermoelectromotive force contained in the auxiliary pump voltage being temporarily large can be reduced.

[0012] In this case, the pump cell control unit may be provided with a memory unit that stores a correspondence relationship between the main pump current and a calculated reference voltage corresponding to the oxygen concentration in the measured gas surrounding the sensor element, and the calculated reference voltage may be derived based on this correspondence relationship and the main pump current.

[0013] In the gas sensor of the present invention, the element body may have an elongated shape having a front end and a rear end which are both ends along the longitudinal direction, the inner auxiliary pump electrode and the reference electrode may be disposed on the front end side of the element body, and the reference electrode may be disposed at a position farther from the front end than the inner auxiliary pump electrode. When the inner auxiliary pump electrode and the reference electrode are disposed in such a positional relationship, a difference in the rate of temperature rise between the inner auxiliary pump electrode and the reference electrode occurs from the time when heating of the sensor element by the heater begins until the temperatures of both electrodes become stable, and the temperature difference between the two electrodes is likely to become temporarily large. For this reason, the correction process of the present invention is highly significant.

[0014] In the gas sensor of the above aspect in which correction is performed based on the deviation between the calculated reference voltage and the measured reference voltage, the measured gas side electrode may be disposed on the front end side of the element body, and the reference electrode may be disposed at a position farther from the front end than the measured gas side electrode. When the measured gas side electrode and the reference electrode are disposed in such a positional relationship, when the temperature difference between the inner auxiliary pump electrode and the reference electrode is temporarily large, the temperature difference between the measured gas side electrode and the reference electrode is also likely to be temporarily large. Therefore, in the case of such an electrode positional relationship, the larger the deviation between the calculated reference voltage and the measured reference voltage, the larger the thermoelectromotive force contained in the auxiliary pump voltage is, so that the above-mentioned correction based on the deviation is suitable.

[0015] Incidentally, a gas sensor requires a time from when the heater starts to be energized until it can correctly detect the concentration of a specific gas, and this time is called the light-off time. The light-off time tends to be longer as the time required to pump out the oxygen (oxygen not derived from the specific gas) that exists in the measurement chamber in which the measurement electrode is placed before the use of the sensor element to a level that does not affect the measurement accuracy is longer. In recent years, the need to shorten the light-off time in such gas sensors has increased due to the strengthening of exhaust gas regulations.

[0016] In the gas sensor of the present invention, the pump cell control unit may perform the normal control process during normal operation of the sensor element, and may perform a startup measurement pump control process during startup of the sensor element before the normal operation, in which the measurement pump cell is controlled so that the measurement voltage becomes a startup target value higher than the normal target value to pump oxygen from the measurement chamber. When such startup measurement pump control process is performed during startup of the sensor element before the normal operation, the target value of the oxygen concentration in the measurement chamber is set to a value lower than that during normal operation, and oxygen is pumped from the measurement chamber. This makes it possible to quickly remove oxygen that exists in the measurement chamber before the startup of the sensor element from the measurement chamber, compared to when the normal measurement pump control process is performed from the startup of the sensor element. Therefore, the light-off time of the sensor element is shortened. Here, examples of oxygen that exists in the measurement chamber before the startup of the sensor element include oxygen molecules (O2) that exist in the space in the measurement chamber, oxygen molecules (O2) attached to the surface of the inner measurement electrode, and oxygen that is bonded to the constituent material of the inner measurement electrode (oxidizing the constituent material).

[0017] In the gas sensor of the present invention, the heater control process may be a process of energizing the heater to generate heat so that a sensor element temperature, which is the temperature of the heater or the element body, reaches a predetermined target temperature, and the pump cell control unit may start the pump control process for startup measurement when the heater control process is started and the sensor element temperature reaches a predetermined threshold value that is equal to or lower than the target temperature. In this way, the start timing of the pump control process for startup measurement can be appropriately determined based on the sensor element temperature. The predetermined threshold value may be a value lower than the target temperature. Here, "energizing the heater to generate heat so that the sensor element temperature reaches a predetermined target temperature" includes a case where the heater is controlled based on the sensor element temperature itself and a case where the heater is controlled based on a value that can be converted to the sensor element temperature (e.g., the resistance value of the heater, the resistance value of an electric circuit including a solid electrolyte, etc.). Examples of the resistance value of the electric circuit including a solid electrolyte include the resistance value of the measurement pump cell and the resistance value of the measurement voltage detection sensor cell.

[0018] In the gas sensor of the present invention, the pump cell control unit performs the auxiliary pump control process and the main pump control process at the start of the sensor element, and the main pump control process controls the main pump cell so that the auxiliary pump current flowing through the auxiliary pump cell by the auxiliary pump control process becomes a target current. When the pump cell control unit determines that the auxiliary pump current is stable near the target current, the pump cell control unit may switch from the start-up measurement pump control process to the normal measurement pump control process to start the normal control process. Here, before the start of the sensor element, not only the measurement chamber but also the first and second internal spaces upstream of the measurement chamber are in a state (e.g., atmospheric air) in which a large amount of oxygen (oxygen not derived from a specific gas) is present. By performing the main pump control process and the auxiliary pump control process by the pump cell control unit at the start of the sensor element, the oxygen present in the first and second internal spaces before the start of the sensor element can also be removed. In addition, since the first and second internal spaces are connected to the measurement chamber, the oxygen in the measurement chamber can also be removed by the operation of the main pump cell and the auxiliary pump cell. Furthermore, since the first and second internal spaces communicate with the measurement chamber as described above, when the oxygen concentrations in the first and second internal spaces are stable, the oxygen in the measurement chamber is also sufficiently pumped out. When the auxiliary pump current is stabilized near the target current, the oxygen concentrations in the first and second internal spaces are likely to be stable. Therefore, by switching from the startup measurement pump control process to the normal measurement pump control process when it is determined that the auxiliary pump current is stabilized near the target current, it is possible to appropriately switch from the startup measurement pump control process to the normal measurement pump control process. Here, "when it is determined that the auxiliary pump current is stabilized near the target current, switching from the startup measurement pump control process to the normal measurement pump control process" includes a case where the switching is performed immediately when it is determined that the auxiliary pump current is stabilized near the target current, and a case where the switching is performed after the timing when it is determined that the auxiliary pump current is stabilized near the target current.An example of the latter is a case where switching is performed a predetermined time after it is determined that the auxiliary pump current has stabilized near the target current, or when other conditions are met in addition to determining that the auxiliary pump current has stabilized near the target current.

[0019] In the gas sensor of the present invention, the pump cell control unit may switch from the startup measurement pump control process to the normal measurement pump control process and start the normal control process when an open measurement voltage, which is the measurement voltage in a state where no control is performed to cause a current to flow between the inner measurement electrode and the reference electrode, reaches a predetermined threshold value or more. Since the open measurement voltage is a value corresponding to the oxygen concentration in the measurement chamber, it is possible to appropriately determine whether or not the oxygen in the measurement chamber has been sufficiently pumped out by comparing the open measurement voltage with the threshold value. Therefore, the startup measurement pump control process can be appropriately switched to the normal measurement pump control process.

[0020] In the gas sensor of the present invention, the inner measurement electrode may contain at least one of Pt and Rh. Precious metals such as Pt and Rh may combine with oxygen in the measurement chamber and be oxidized, and this oxygen reduces the detection accuracy of the specific gas concentration. Therefore, when the sensor element is started up, it is necessary to reduce the oxide of the precious metal contained in the inner measurement electrode and then pump this oxygen out of the measurement chamber. Since it takes more time to reduce the oxide of the precious metal and pump out oxygen than to pump out oxygen molecules in the measurement chamber, the light-off time is likely to be long if oxidized precious metal is present in the measurement chamber. In the gas sensor of the present invention, when the sensor element is started up, the oxide of the oxidized precious metal can be reduced early and the light-off time can be shortened by performing the start-up measurement pump control process instead of the normal measurement pump control process. Therefore, when the inner measurement electrode contains at least one of Pt and Rh, it is highly meaningful to perform the start-up measurement pump control process.

[0021] In the gas sensor of the present invention, the inner measurement electrode has a volume C of 8×10 -3 mm 332×10 or more -3 mm 3 The difference ΔV between the startup target value and the normal target value may be 120 mV or more and 200 mV or less. -3 mm 3 Above this level, the measurement pump cell has a sufficiently high ability to pump out oxygen around the measurement electrode. Also, when the difference ΔV is 200 mV or less, the voltage applied to the inner measurement electrode during the start-up measurement pump control process does not become too high. Also, the larger the volume C, the larger the difference ΔV required to sufficiently increase the effect of shortening the light-off time tends to be. However, when the volume C is 8×10 -3 mm 3 32×10 or more -3 mm 3 If the difference ΔV is 120 mV or more and 200 mV or less, the effect of shortening the light-off time can be sufficiently increased while satisfying the above-mentioned lower limit of the volume C and the upper limit of the difference ΔV.

[0022] In the gas sensor of the present invention, when the pump cell control unit determines that the oxygen concentration in the measurement chamber is stable based on the operation of the measurement pump cell, the pump cell control unit may switch from the startup measurement pump control process to the normal measurement pump control process to start the normal control process. When the oxygen concentration in the measurement chamber is stable, it is highly likely that the oxygen in the measurement chamber has been sufficiently pumped out by the startup measurement pump control process. Therefore, by switching from the startup measurement pump control process to the normal measurement pump control process in this case, the switching from the startup measurement pump control process to the normal measurement pump control process can be appropriately performed.

[0023] In the gas sensor of the present invention, the pump cell control unit may determine that the oxygen concentration in the measurement chamber is stable when the measurement pump current flowing through the measurement pump cell is stabilized by the startup measurement pump control process. In this way, it is possible to determine whether the oxygen concentration in the measurement chamber is stable based on the measurement pump current during the startup measurement pump control process.

[0024] In the gas sensor of the present invention, the pump cell control unit may determine whether the measurement pump current is stable based on a first derivative of the measurement pump current flowing through the measurement pump cell by the startup measurement pump control process. Alternatively, the pump cell control unit may determine whether the measurement pump current is stable based on a second derivative of the measurement pump current flowing through the measurement pump cell by the startup measurement pump control process. In either of these cases, the pump cell control unit may determine whether the measurement pump current is stable based on the measurement pump current flowing through the measurement pump cell after the measurement pump current flowing through the measurement pump cell by the startup measurement pump control process has increased once and then started to decrease. [Brief description of the drawings]

[0025] [Figure 1] FIG. 1 is a schematic cross-sectional view of a gas sensor 100. [Diagram 2] FIG. 4 is a block diagram showing the electrical connection relationship between the control device 90 and each cell and heater 72. [Diagram 3] 11 is a flowchart showing an example of a start-up control process. [Figure 4] 6 is a graph showing an example of a startup control process and a normal control process; [Diagram 5] 10 is a flowchart showing an example of a startup control process according to a modified example. [Figure 6] 10 is a graph showing an example of a startup control process and a normal control process according to a modified example. [Figure 7] Phase diagram of Rh2O3. [Figure 8] FIG. 11 is a schematic cross-sectional view of a sensor element 201 according to a modified example. [Figure 9] 11 is a graph showing the relationship between the difference ΔV between the volume C of the measuring electrode 44 and a target value, and the light-off time. [Figure 10] 13 is a graph showing an example of an undershoot of a pump current Ip2. [Figure 11] 4 is an explanatory diagram of the stable region of the pump current Ip2, and the first and second derivatives of the pump current Ip2. [Figure 12]10 is a flowchart showing an example of a startup control process according to a modified example. [Figure 13] 4 is a graph showing an example of a change over time in voltage V1 caused by a thermoelectromotive force. [Figure 14] 4 is a graph showing an example of a change over time in a pump current Ip2 caused by a thermoelectromotive force contained in a voltage V1. [Figure 15] 11 is a flowchart showing an example of a correction process. [Figure 16] 10 is a flowchart showing an example of another correction process. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0026] Next, an embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a schematic cross-sectional view showing an example of the configuration of a gas sensor 100 according to an embodiment of the present invention. FIG. 2 is a block diagram showing the electrical connection between a control device 90 and each cell and a heater 72. The gas sensor 100 is attached to a pipe such as an exhaust gas pipe of an internal combustion engine such as a diesel engine. The gas sensor 100 detects the concentration of a specific gas such as NOx in the measured gas, which is exhaust gas from the internal combustion engine. The gas sensor 100 includes a sensor element 101 having a long rectangular parallelepiped shape, each cell 15, 21, 41, 50, 80 to 83 including a part of the sensor element 101, a heater unit 70 provided inside the sensor element 101, and a control device 90 for controlling the entire gas sensor 100.

[0027] The sensor element 101 is an element having a laminate in which six layers, namely a first substrate layer 1, a second substrate layer 2, a third substrate layer 3, a first solid electrolyte layer 4, a spacer layer 5, and a second solid electrolyte layer 6, each of which is made of an oxygen ion conductive solid electrolyte layer such as zirconia (ZrO2), are laminated in this order from the bottom as viewed in the drawing. The solid electrolyte forming these six layers is dense and airtight. The sensor element 101 is manufactured, for example, by performing a predetermined processing and printing a circuit pattern on ceramic green sheets corresponding to each layer, laminating them, and further firing them to integrate them.

[0028] On the tip side (the left end side in FIG. 1) of the sensor element 101, between the lower surface of the second solid electrolyte layer 6 and the upper surface of the first solid electrolyte layer 4, a gas inlet 10, a first diffusion rate-controlling section 11, a buffer space 12, a second diffusion rate-controlling section 13, a first internal void 20, a third diffusion rate-controlling section 30, a second internal void 40, a fourth diffusion rate-controlling section 60, and a third internal void 61 are formed adjacent to each other and communicated with each other in this order.

[0029] The gas inlet 10, the buffer space 12, the first internal void 20, the second internal void 40, and the third internal void 61 are spaces inside the sensor element 101, which are defined by an upper portion cut out of the spacer layer 5, the lower portion cut out by the upper surface of the first solid electrolyte layer 4, and the sides cut out by the side surfaces of the spacer layer 5.

[0030] The first diffusion rate-controlling section 11, the second diffusion rate-controlling section 13, and the third diffusion rate-controlling section 30 are each provided as two horizontally elongated slits (with the opening extending in the direction perpendicular to the drawing). The fourth diffusion rate-controlling section 60 is provided as one horizontally elongated slit (with the opening extending in the direction perpendicular to the drawing) formed as a gap with the lower surface of the second solid electrolyte layer 6. The portion from the gas inlet 10 to the third internal space 61 is also referred to as a measurement gas flow section.

[0031] Further, at a position farther from the tip side than the measurement gas flow portion, a reference gas introduction space 43 is provided between the upper surface of the third substrate layer 3 and the lower surface of the spacer layer 5, and at a position defined at the side by the side surface of the first solid electrolyte layer 4. For example, air is introduced into the reference gas introduction space 43 as a reference gas when measuring the NOx concentration.

[0032] The air introduction layer 48 is a layer made of porous ceramics, and a reference gas is introduced into the air introduction layer 48 through the reference gas introduction space 43. The air introduction layer 48 is formed so as to cover the reference electrode 42.

[0033] The reference electrode 42 is an electrode formed in a manner to be sandwiched between the upper surface of the third substrate layer 3 and the first solid electrolyte layer 4, and as described above, is surrounded by an air introduction layer 48 connected to a reference gas introduction space 43. As will be described later, the reference electrode 42 can be used to measure the oxygen concentration (oxygen partial pressure) in the first internal space 20, the second internal space 40, and the third internal space 61. The reference electrode 42 is formed as a porous cermet electrode (for example, a cermet electrode of Pt and ZrO2).

[0034] In the measurement gas flow section, the gas inlet 10 is a section that opens to the external space, and the measurement gas is introduced into the sensor element 101 from the external space through the gas inlet 10. The first diffusion rate-controlling section 11 is a section that imparts a predetermined diffusion resistance to the measurement gas introduced from the gas inlet 10. The buffer space 12 is a space provided for guiding the measurement gas introduced from the first diffusion rate-controlling section 11 to the second diffusion rate-controlling section 13. The second diffusion rate-controlling section 13 is a section that imparts a predetermined diffusion resistance to the measurement gas introduced from the buffer space 12 into the first internal space 20. When the measurement gas is introduced from the outside of the sensor element 101 to the first internal space 20, the measurement gas is suddenly taken into the sensor element 101 from the gas inlet 10 due to pressure fluctuations of the measurement gas in the external space (exhaust pressure pulsations if the measurement gas is automobile exhaust gas), but is not introduced directly into the first internal space 20, but is introduced into the first internal space 20 after the pressure fluctuations of the measurement gas are canceled through the first diffusion rate-controlling section 11, the buffer space 12, and the second diffusion rate-controlling section 13. As a result, the pressure fluctuations of the measurement gas introduced into the first internal space 20 are almost negligible. The first internal space 20 is provided as a space for adjusting the oxygen partial pressure in the measurement gas introduced through the second diffusion rate-controlling section 13. The oxygen partial pressure is adjusted by the operation of the main pump cell 21.

[0035] The main pump cell 21 is an electrochemical pump cell including an inner pump electrode 22 having a ceiling electrode portion 22a provided on almost the entire lower surface of the second solid electrolyte layer 6 facing the first internal space 20, an outer pump electrode 23 provided in a region corresponding to the ceiling electrode portion 22a on the upper surface of the second solid electrolyte layer 6 so as to be exposed to the external space, and the second solid electrolyte layer 6 sandwiched between these electrodes.

[0036] The inner pump electrode 22 is formed across the upper and lower solid electrolyte layers (the second solid electrolyte layer 6 and the first solid electrolyte layer 4) that define the first internal space 20, and the spacer layer 5 that provides the side walls. Specifically, a ceiling electrode portion 22a is formed on the lower surface of the second solid electrolyte layer 6 that provides the ceiling surface of the first internal space 20, and a bottom electrode portion 22b is formed on the upper surface of the first solid electrolyte layer 4 that provides the bottom surface. Side electrode portions (not shown) are formed on the side wall surfaces (inner surfaces) of the spacer layer 5 that constitute both side wall portions of the first internal space 20 so as to connect the ceiling electrode portion 22a and the bottom electrode portion 22b, and are arranged in a tunnel-shaped structure at the positions where the side electrode portions are arranged.

[0037] The inner pump electrode 22 and the outer pump electrode 23 are formed as porous cermet electrodes (for example, a cermet electrode of Pt containing 1% Au and ZrO2). The inner pump electrode 22, which comes into contact with the measurement gas, is formed using a material with a weakened ability to reduce the NOx component in the measurement gas.

[0038] In the main pump cell 21, by applying a desired pump voltage Vp0 between the inner pump electrode 22 and the outer pump electrode 23 and flowing a pump current Ip0 in a positive or negative direction between the inner pump electrode 22 and the outer pump electrode 23, it is possible to pump oxygen in the first internal space 20 out to the external space, or to pump oxygen in the external space into the first internal space 20.

[0039] In addition, in order to detect the oxygen concentration (oxygen partial pressure) in the atmosphere in the first internal space 20, an electrochemical sensor cell, i.e., an oxygen partial pressure detection sensor cell 80 for controlling the main pump, is configured by the inner pump electrode 22, the second solid electrolyte layer 6, the spacer layer 5, the first solid electrolyte layer 4, the third substrate layer 3, and the reference electrode 42.

[0040] The oxygen concentration (oxygen partial pressure) in the first internal space 20 can be determined by measuring the electromotive force (voltage V0) in the main pump control oxygen partial pressure detection sensor cell 80. Furthermore, the pump current Ip0 is controlled by feedback controlling the pump voltage Vp0 of the variable power supply 24 so that the voltage V0 becomes a target value. This allows the oxygen concentration in the first internal space 20 to be maintained at a predetermined constant value.

[0041] The third diffusion rate-controlling section 30 is a section that applies a predetermined diffusion resistance to the measurement gas whose oxygen concentration (oxygen partial pressure) has been controlled by the operation of the main pump cell 21 in the first internal space 20, and guides the measurement gas to the second internal space 40.

[0042] The second internal space 40 is provided as a space for further adjusting the oxygen partial pressure by the auxiliary pump cell 50 for the measurement gas introduced through the third diffusion rate-controlling section 30 after the oxygen concentration (oxygen partial pressure) has been adjusted in advance in the first internal space 20. This makes it possible to keep the oxygen concentration in the second internal space 40 constant with high accuracy, thereby enabling the gas sensor 100 to measure the NOx concentration with high accuracy.

[0043] The auxiliary pump cell 50 is an auxiliary electrochemical pump cell composed of an auxiliary pump electrode 51 having a ceiling electrode portion 51a provided on almost the entire lower surface of the second solid electrolyte layer 6 facing the second internal space 40, an outer pump electrode 23 (not limited to the outer pump electrode 23, but any appropriate electrode on the outside of the sensor element 101 will suffice), and the second solid electrolyte layer 6.

[0044] The auxiliary pump electrode 51 is disposed in the second internal space 40 in a tunnel-shaped structure similar to that of the inner pump electrode 22 provided in the first internal space 20. That is, a ceiling electrode portion 51a is formed on the second solid electrolyte layer 6 providing the ceiling surface of the second internal space 40, a bottom electrode portion 51b is formed on the first solid electrolyte layer 4 providing the bottom surface of the second internal space 40, and side electrode portions (not shown) connecting the ceiling electrode portion 51a and the bottom electrode portion 51b are formed on both wall surfaces of the spacer layer 5 providing the side walls of the second internal space 40, respectively, forming a tunnel-shaped structure. Note that, like the inner pump electrode 22, the auxiliary pump electrode 51 is also formed using a material with a weakened ability to reduce NOx components in the measurement gas.

[0045] In the auxiliary pump cell 50, by applying a desired voltage Vp1 between the auxiliary pump electrode 51 and the outer pump electrode 23, it is possible to pump oxygen in the atmosphere within the second internal space 40 out to the external space or pump oxygen from the external space into the second internal space 40.

[0046] In addition, in order to control the oxygen partial pressure in the atmosphere within the second internal space 40, an electrochemical sensor cell, i.e., an oxygen partial pressure detection sensor cell 81 for controlling the auxiliary pump, is constituted by the auxiliary pump electrode 51, the reference electrode 42, the second solid electrolyte layer 6, the spacer layer 5, the first solid electrolyte layer 4, and the third substrate layer 3.

[0047] The auxiliary pump cell 50 performs pumping using a variable power supply 52 whose voltage is controlled based on the electromotive force (voltage V1) detected by the auxiliary pump control oxygen partial pressure detection sensor cell 81. This allows the oxygen partial pressure in the atmosphere in the second internal space 40 to be controlled to a low partial pressure that does not substantially affect the measurement of NOx.

[0048] In addition, the pump current Ip1 is used to control the electromotive force of the main pump control oxygen partial pressure detection sensor cell 80. Specifically, the pump current Ip1 is input as a control signal to the main pump control oxygen partial pressure detection sensor cell 80, and the above-mentioned target value of the voltage V0 is controlled to always keep a constant gradient of the oxygen partial pressure in the measurement gas introduced from the third diffusion rate-controlling part 30 into the second internal space 40. When used as a NOx sensor, the oxygen concentration in the second internal space 40 is kept constant at about 0.001 ppm by the action of the main pump cell 21 and the auxiliary pump cell 50.

[0049] The fourth diffusion rate-controlling part 60 is a part that applies a predetermined diffusion resistance to the measurement gas whose oxygen concentration (oxygen partial pressure) has been controlled by the operation of the auxiliary pump cell 50 in the second internal space 40, and guides the measurement gas to the third internal space 61. The fourth diffusion rate-controlling part 60 serves to limit the amount of NOx flowing into the third internal space 61.

[0050] The third internal space 61 is provided as a space for carrying out a process for measuring the nitrogen oxide (NOx) concentration in the measurement gas, which has been introduced through the fourth diffusion-controlling part 60 after the oxygen concentration (oxygen partial pressure) has been adjusted in advance in the second internal space 40. The measurement of the NOx concentration is mainly carried out in the third internal space 61 by the operation of the measurement pump cell 41.

[0051] The measurement pump cell 41 measures the NOx concentration in the measurement gas in the third internal space 61. The measurement pump cell 41 is an electrochemical pump cell including a measurement electrode 44 provided on the upper surface of the first solid electrolyte layer 4 facing the third internal space 61, the outer pump electrode 23, the second solid electrolyte layer 6, the spacer layer 5, and the first solid electrolyte layer 4. The measurement electrode 44 is a porous cermet electrode made of a material having a higher reduction ability for the NOx component in the measurement gas than the inner pump electrode 22. The measurement electrode 44 also functions as a NOx reduction catalyst that reduces NOx present in the atmosphere in the third internal space 61.

[0052] Specifically, the measurement electrode 44 is an electrode containing at least one of Pt and Rh, which are precious metals having catalytic activity. The measurement electrode 44 is preferably an electrode made of a cermet containing at least one of Pt and Rh and an oxide (here, ZrO2) having oxygen ion conductivity. In addition, the measurement electrode 44 is preferably a porous body. In this embodiment, the measurement electrode 44 is a porous cermet electrode of Pt, Rh, and ZrO2.

[0053] In the measurement pump cell 41, oxygen produced by decomposition of nitrogen oxides in the atmosphere surrounding the measurement electrode 44 is pumped out, and the amount of oxygen produced can be detected as a pump current Ip2.

[0054] In order to detect the oxygen partial pressure around the measurement electrode 44, an electrochemical sensor cell, i.e., an oxygen partial pressure detection sensor cell 82 for controlling the measurement pump, is configured by the first solid electrolyte layer 4, the third substrate layer 3, the measurement electrode 44, and the reference electrode 42. The variable power supply 46 is controlled based on the electromotive force (voltage V2) detected by the oxygen partial pressure detection sensor cell 82 for controlling the measurement pump.

[0055] The measurement gas introduced into the second internal space 40 reaches the measurement electrode 44 in the third internal space 61 through the fourth diffusion rate-controlling part 60 under the condition that the oxygen partial pressure is controlled. Nitrogen oxides in the measurement gas around the measurement electrode 44 are reduced (2NO→N2+O2) to generate oxygen. The generated oxygen is then pumped by the measurement pump cell 41, and the voltage Vp2 of the variable power supply 46 is controlled so that the voltage V2 detected by the measurement pump control oxygen partial pressure detection sensor cell 82 becomes constant (target value). The amount of oxygen generated around the measurement electrode 44 is proportional to the concentration of nitrogen oxides in the measurement gas, so that the nitrogen oxide concentration in the measurement gas is calculated using the pump current Ip2 in the measurement pump cell 41.

[0056] In addition, the second solid electrolyte layer 6, the spacer layer 5, the first solid electrolyte layer 4, the third substrate layer 3, the outer pump electrode 23, and the reference electrode 42 constitute an electrochemical sensor cell 83, and the electromotive force (voltage Vref) obtained by this sensor cell 83 makes it possible to detect the oxygen partial pressure in the measured gas outside the sensor.

[0057] In the gas sensor 100 having such a configuration, the main pump cell 21 and the auxiliary pump cell 50 are operated to supply a measurement gas in which the oxygen partial pressure is always kept at a constant low value (a value that has substantially no effect on the measurement of NOx) to the measurement pump cell 41. Therefore, the NOx concentration in the measurement gas can be known based on the pump current Ip2 that flows when oxygen generated by the reduction of NOx is pumped out of the measurement pump cell 41, which is approximately proportional to the NOx concentration in the measurement gas.

[0058] Furthermore, in order to increase the oxygen ion conductivity of the solid electrolyte, the sensor element 101 includes a heater section 70 that serves to adjust the temperature by heating and keeping the sensor element 101 warm. The heater section 70 includes a heater connector electrode 71, a heater 72, a through hole 73, a heater insulating layer 74, and a pressure relief hole 75.

[0059] The heater connector electrode 71 is an electrode formed in a manner to contact the lower surface of the first substrate layer 1. By connecting the heater connector electrode 71 to an external power source, it is possible to supply power to the heater section 70 from the outside.

[0060] The heater 72 is an electrical resistor sandwiched between the second substrate layer 2 and the third substrate layer 3. The heater 72 is connected to a heater connector electrode 71 via a through hole 73, and generates heat when power is supplied from a heater power source 76 (see FIG. 2) through the heater connector electrode 71, thereby heating and keeping warm the solid electrolyte that forms the sensor element 101.

[0061] The heater 72 is embedded throughout the entire area from the first internal space 20 to the third internal space 61, and is capable of adjusting the temperature of the entire sensor element 101 to a temperature at which the solid electrolyte is activated.

[0062] The heater insulating layer 74 is an insulating layer made of an insulator such as alumina and formed on the upper and lower surfaces of the heater 72. The heater insulating layer 74 is formed for the purpose of obtaining electrical insulation between the second substrate layer 2 and the heater 72, and between the third substrate layer 3 and the heater 72.

[0063] The pressure relief hole 75 is a portion that penetrates the third substrate layer 3 and the atmosphere introduction layer 48 and is provided so as to communicate with the reference gas introduction space 43, and is formed for the purpose of mitigating the increase in internal pressure that accompanies an increase in temperature within the heater insulation layer 74.

[0064] 2, the control device 90 includes the variable power supplies 24, 46, 52, the heater power supply 76, and a control unit 91. The control unit 91 is a microprocessor including a CPU 92 and a memory unit 94. The memory unit 94 is a device for storing, for example, various programs and various data. The control unit 91 receives the voltage V0 detected by the oxygen partial pressure detection sensor cell 80 for controlling the main pump, the voltage V1 detected by the oxygen partial pressure detection sensor cell 81 for controlling the auxiliary pump, the voltage V2 detected by the oxygen partial pressure detection sensor cell 82 for controlling the measurement pump, the voltage Vref detected by the sensor cell 83, the pump current Ip0 detected by the main pump cell 21, the pump current Ip1 detected by the auxiliary pump cell 50, and the pump current Ip2 detected by the measurement pump cell 41. The control unit 91 also outputs control signals to the variable power sources 24, 46, 52 to control the voltages Vp0, Vp1, Vp2 output by the variable power sources 24, 46, 52, thereby controlling the main pump cell 21, the measurement pump cell 41, and the auxiliary pump cell 50. The control unit 91 outputs control signals to the heater power source 76 to control the power supplied from the heater power source 76 to the heater 72. The memory unit 94 also stores target values ​​V0*, V1*, V2a*, V2b*, etc., which will be described later. The CPU 92 of the control unit 91 controls each of the cells 21, 41, 50 by referring to these target values ​​V0*, V1*, V2a*, V2b*.

[0065] The control unit 91 performs an auxiliary pump control process to control the auxiliary pump cell 50 so that the oxygen concentration in the second internal space 40 becomes a target concentration. Specifically, the control unit 91 controls the auxiliary pump cell 50 by feedback-controlling the voltage Vp1 of the variable power source 52 so that the voltage V1 becomes a constant value (referred to as a target value V1*). The target value V1* is determined as a value that makes the oxygen concentration in the second internal space 40 a predetermined low concentration that does not substantially affect the measurement of NOx.

[0066] The control unit 91 performs a main pump control process to control the main pump cell 21 so that the pump current Ip1 flowing when the auxiliary pump cell 50 adjusts the oxygen concentration in the second internal space 40 by the auxiliary pump control process becomes a target current (referred to as the target current Ip1*). Specifically, the control unit 91 sets (feedback controls) a target value (referred to as the target value V0*) of the voltage V0 based on the pump current Ip1 so that the pump current Ip1 flowing by the voltage Vp1 becomes a constant target current Ip1*. Then, the control unit 91 feedback controls the pump voltage Vp0 of the variable power supply 24 so that the voltage V0 becomes the target value V0* (i.e., so that the oxygen concentration in the first internal space 20 becomes the target concentration). This main pump control process keeps the gradient of the oxygen partial pressure in the measurement gas introduced from the third diffusion rate-controlling unit 30 into the second internal space 40 constant. The target value V0* is set to a value such that the oxygen concentration in the first internal space 20 is higher than 0% and is a low oxygen concentration. The pump current Ip0 flowing during the main pump control process changes according to the oxygen concentration of the measurement gas (i.e., the measurement gas around the sensor element 101) flowing into the measurement gas flow section from the gas inlet 10. Therefore, the control section 91 can also detect the oxygen concentration in the measurement gas based on the pump current Ip0.

[0067] The above-mentioned main pump control process and auxiliary pump control process are also collectively referred to as an adjustment pump control process. The first internal space 20 and the second internal space 40 are also collectively referred to as an oxygen concentration adjustment chamber. The main pump cell 21 and the auxiliary pump cell 50 are also collectively referred to as an adjustment pump cell. The control unit 91 performs the adjustment pump control process, causing the adjustment pump cell to adjust the oxygen concentration in the oxygen concentration adjustment chamber.

[0068] Furthermore, the control unit 91 performs a measurement pump control process to control the measurement pump cell 41 so that the voltage V2 becomes a constant value (target value) (i.e., the oxygen concentration in the third internal space 61 becomes a predetermined low concentration). Specifically, the control unit 91 controls the measurement pump cell 41 by feedback-controlling the voltage Vp2 of the variable power supply 46 so that the voltage V2 becomes the target value. This measurement pump control process pumps out oxygen from the third internal space 61. In addition, the control unit 91 performs, as the measurement pump control process, a normal measurement pump control process performed during normal driving of the sensor element 101 and a startup measurement pump control process performed when the sensor element 101 is started up before the normal driving. The normal measurement pump control process and the startup measurement pump control process have different target values ​​for the voltage V2. The target value of the voltage V2 in the startup measurement pump control process is referred to as a startup target value V2a*. The target value of the voltage V2 in the pump control process for normal measurement is referred to as the normal target value V2b*. The start-up target value V2a* is set to a value higher than the normal target value V2b*. That is, V2a* and V2b* are set so that V2a*>V2b*. Here, the voltage V2 is a value related to the oxygen concentration difference between the periphery of the reference electrode 42 and the third internal space 61, and the lower the oxygen concentration in the third internal space 61, the larger the oxygen concentration difference becomes, and the larger the voltage V2 becomes. Therefore, the fact that the start-up target value V2a* is a value higher than the normal target value V2b* means that the target value of the oxygen concentration in the third internal space 61 is set to a lower value when the pump control process for normal measurement is executed, compared to when the pump control process for normal measurement is executed. In this embodiment, the normal target value V2b* is set to 400mV, and the start-up target value V2a* is set to 600mV.

[0069] By performing the normal measurement pump control process, oxygen is pumped out from the third internal space 61 so that the amount of oxygen generated by reduction of NOx in the measurement gas in the third internal space 61 becomes substantially zero. Then, the control unit 91 obtains the pump current Ip2 as a detection value corresponding to the oxygen generated in the third internal space 61 derived from the specific gas (here, NOx), and calculates the NOx concentration in the measurement gas based on this pump current Ip2.

[0070] As the correspondence relationship between the pump current Ip2 and the NOx concentration, a relational expression (for example, a linear function expression), a map, etc. are stored in the storage unit 94. Such a relational expression or map can be obtained in advance by experiment.

[0071] An example of a startup control process, which is a process performed by the control unit 91 of the gas sensor 100 configured as described above when starting up the sensor element 101, will be described below. Fig. 3 is a flowchart showing an example of the startup control process. Fig. 4 is a graph showing an example of the startup control process and the normal control process.

[0072] When the start-up control process is started, the CPU 92 of the control unit 91 first starts a heater control process in which the heater 72 is energized to generate heat so that the heater temperature Th of the heater 72 becomes the target temperature Th* (step S100). The heater temperature Th can be expressed by a linear function of the resistance value of the heater 72. In the heater control process of this embodiment, the CPU 92 calculates the resistance value of the heater 72 and feedback controls the heater power supply 76 so that the calculated resistance value becomes the target resistance value (resistance value corresponding to the target temperature Th*). The CPU 92 can, for example, obtain the voltage of the heater 72 and the current flowing through the heater 72, and calculate the resistance value of the heater 72 based on the obtained voltage and current. The CPU 92 may calculate the resistance value of the heater 72 by, for example, a three-terminal method or a four-terminal method. The CPU 92 outputs a control signal to the heater power supply 76 so that the calculated resistance value of the heater 72 becomes the target resistance value, and feedback controls the power supplied by the heater power supply 76. The heater power supply 76 adjusts the power supplied to the heater 72, for example, by changing the value of the voltage applied to the heater 72. As shown in Fig. 4, when the heater control process is started at time t0, for example, the heater temperature Th rises to the target temperature Th* over time, and after time t2 when the heater temperature Th reaches the target temperature Th*, the heater temperature Th is maintained near the target temperature Th*. The target temperature Th* is predetermined as a temperature (e.g., 800°C) that can sufficiently activate the solid electrolyte of the sensor element 101. Note that, in Fig. 4, the change in the heater temperature Th over time is shown as a straight line for ease of explanation, but in reality, the heater temperature Th may increase in a curved manner or an overshoot may occur in the heater temperature Th before it stabilizes near the target temperature Th*.

[0073] After starting the heater control process in step S100, the CPU 92 judges whether or not the start condition of the start-up measurement pump control process is satisfied. In this embodiment, the CPU 92 judges whether or not the heater temperature Th has reached a predetermined threshold value Thref or more (step S110), and when the judgment result is a positive judgment, it is deemed that the start condition of the start-up measurement pump control process is satisfied. The threshold value Thref is a value equal to or less than the target temperature Th*, and may be a value less than the target temperature Th*. The threshold value Thref is previously determined as a lower limit value of the heater temperature Th required to activate (to develop ion conductivity of the solid electrolyte) the solid electrolyte (in this embodiment, the second solid electrolyte layer 6, the spacer layer 5, and the first solid electrolyte layer 4) contained in the measurement pump cell 41 controlled by the start-up measurement pump control process. For example, the threshold value Thref may be previously determined by experiment as a lower limit value of the heater temperature Th required to heat the second solid electrolyte layer 6, the spacer layer 5, and the first solid electrolyte layer 4 to 600° C. or more. The threshold value Thref may be determined as a value obtained by multiplying the target temperature Th* by a predetermined ratio (a value less than 1). In this embodiment, the threshold value Thref=0.8×Th*. That is, when the heater temperature Th reaches 80% or more of the target temperature Th*, the CPU 92 determines that the start condition for the startup measurement pump control process is satisfied.

[0074] If the determination in step S110 is negative, the CPU 92 repeatedly executes step S110 and waits until the determination is positive. If the determination in step S110 is positive, the CPU 92 starts the above-mentioned startup measurement pump control process (step S120). In the example of FIG. 4, the heater temperature Th reaches the threshold value Thref at time t1 before time t2, so the CPU 92 starts the startup measurement pump control process at time t1. When the startup measurement pump control process is started, the CPU 92 feedback controls the voltage Vp2 of the variable power supply 46 so that the voltage V2 becomes the startup target value V2a*, thereby controlling the measurement pump cell 41. The measurement pump cell 41 pumps oxygen present in the third internal space 61 to the periphery of the outer pump electrode 23 by applying the voltage Vp2. At this time, a pump current Ip2 corresponding to the pumped amount of oxygen flows through the measurement pump cell 41. Here, before the start of the sensor element 101, the measurement gas flow section including the third internal space 61 is in a state where there is a lot of oxygen (oxygen not derived from a specific gas) (for example, air atmosphere, etc.). By performing the start-up measurement pump control process, such oxygen is pumped out to the periphery of the outer pump electrode 23. Therefore, as shown in the graph (solid line) of the pump current Ip2 of the embodiment shown in FIG. 4, the pump current Ip2 rises sharply from time t1, and a relatively large pump current Ip2 flows. Specifically, the oxygen present in the third internal space 61 before the start of the sensor element 101 includes oxygen molecules (O2) present in the space within the third internal space 61, oxygen molecules (O2) attached to the surface of the measurement electrode 44, and oxygen bonded to the constituent material of the measurement electrode 44. In this embodiment, the measurement electrode 44 contains Rh and Pt, so that at least one of Rh2O3 and PtO2 may be present in the measurement electrode 44. The oxygen (O) in this Rh2O3 and PtO2 is oxygen that is bonded to the constituent material of the measurement electrode 44. By the start-up measurement pump control process, not only oxygen molecules but also oxygen that is bonded to the constituent material of the measurement electrode 44 can be pumped out, and as a result, Rh2O3 and PtO2 are reduced.In this embodiment, since the measurement electrode 44 is a porous body, oxygen molecules (O2) may exist in the open pores and closed pores of the measurement electrode 44, but these oxygen molecules can also be pumped out by the start-up measurement pump control process. In this embodiment, the CPU 92 starts the start-up measurement pump control process and also starts the above-mentioned adjustment pump control process (main pump control process and auxiliary pump control process). As a result, oxygen that exists in the oxygen concentration adjustment chamber (first internal space 20 and second internal space 40) before the start-up of the sensor element 101 is also pumped out to the periphery of the outer pump electrode 23. Therefore, the pump current Ip0 and the pump current Ip1 rise sharply from time t1, and relatively large pump currents Ip0 and Ip1 flow. Note that the pump current Ip0 is omitted in FIG. 4. For convenience of illustration, the pump current Ip1 is shown to be smaller than the pump current Ip2 in FIG. 4, but the magnitude relationship of the actual pump currents is basically Ip0>Ip1>Ip2.

[0075] After starting the startup measurement pump control process in step S120, the CPU 92 determines whether or not the end condition of the startup measurement pump control process is satisfied. In this embodiment, the CPU 92 determines whether or not the oxygen concentration in the oxygen concentration adjustment chamber is stable based on the operation of the adjustment pump cell (step S130), and if the determination result is a positive determination, it considers that the end condition of the startup measurement pump control process is satisfied. More specifically, the CPU 92 determines that the oxygen concentration in the oxygen concentration adjustment chamber is stable when it determines that the pump current Ip1 is stable near the target current Ip1*.

[0076] For example, as shown in Fig. 4, when the start-up measurement pump control process, the main pump control process, and the auxiliary pump control process are started at time t1, the pump currents Ip1, Ip2 (and pump current Ip0) rise sharply to a relatively large value as described above. As a result, as oxygen in the measurement gas flow section is pumped out, the voltage V2 approaches the start-up target value V2a*, the voltage V1 approaches the target value V1*, and the pump current Ip1 approaches the target current Ip1*, so that the pump currents Ip1, Ip2 (and pump current Ip0) rise once and then gradually decrease. Then, when the oxygen present in the measurement gas flow section before the start-up of the sensor element 101 is sufficiently pumped out (time t3), the voltage V1 thereafter stabilizes at a value substantially equal to the target value V1*, and the pump current Ip1 stabilizes at a value substantially equal to the target current Ip1* (the voltage V1 is not shown in Fig. 4). Here, since the second internal space 40 and the third internal space 61 are in communication with each other, the operation of the auxiliary pump cell 50 can remove oxygen not only from the second internal space 40 but also from the third internal space 61. For example, when the oxygen concentration in the second internal space 40 drops to near the target concentration (oxygen concentration corresponding to the target value V1* of the voltage V1) and the oxygen concentration in the third internal space 61 is higher than the target concentration (oxygen concentration corresponding to the target value V2a* at the start of the voltage V2), the oxygen in the third internal space 61 may reversely diffuse (flow back) into the second internal space 40, and the oxygen can be pumped out by the auxiliary pump cell 50. Therefore, when the oxygen concentration in the oxygen concentration adjusting chamber (the first internal space 20 and the second internal space 40) is stabilized, that is, when the pump current Ip1 is stabilized near the target current Ip1*, the oxygen present in the third internal space 61 before the start of the sensor element 101 is also sufficiently pumped out. Therefore, based on whether the pump current Ip1 has stabilized near the target current Ip1*, it is possible to determine whether the oxygen present in the third internal space 61 has been sufficiently pumped out since before the activation of the sensor element 101. In the determination of step S130, the CPU 92 may determine that the pump current Ip1 has stabilized near the target current Ip1* when the pump current Ip1 increases once, then decreases and reaches a predetermined threshold or lower.In this case, the predetermined threshold value is a value at which the pump current Ip1 can be regarded as having dropped to the vicinity of the target current Ip1*, and can be previously determined as a value equal to or greater than the target current Ip1*, such as the same value as the target current Ip1* or a value 5% greater than the target current Ip1*. Alternatively, the CPU 92 may determine that the pump current Ip1 has stabilized near the target current Ip1* when the state in which the pump current Ip1 is within a predetermined range continues for a predetermined period of time or more. In this case, the predetermined range is a range at which the pump current Ip1 can be regarded as having a value near the target current Ip1*, and can be previously determined as a range within ±5% of the target current Ip1*. In this embodiment, the CPU 92 determines that the pump current Ip1 has stabilized near the target current Ip1* when the pump current Ip1 rises once, then falls and reaches a value equal to or less than the target current Ip1* (time t3 in FIG. 4).

[0077] If the determination in step S130 is negative, the CPU 92 repeatedly executes step S130 and waits until the determination is positive. If the determination in step S130 is positive, the CPU 92 ends the above-mentioned startup measurement pump control process, starts the normal measurement pump control process (step S140), and ends the startup control process. As a result, the process performed by the CPU 92 shifts from the startup control process to the normal control process. In the normal control process, the CPU 92 performs the normal measurement pump control process and continues the above-mentioned adjustment pump control process (main pump control process and auxiliary pump control process) from the startup control process. Then, the CPU 92 calculates the NOx concentration in the measurement gas based on the pump current Ip2 flowing due to the normal measurement pump control process. In the example of FIG. 4, the CPU 92 makes a negative determination in step S130 because the pump current Ip1 is not stable until time t3, and makes a positive determination in step S130 when the pump current Ip1 is stable at time t3. Therefore, the CPU 92 starts the normal control process from time t3.

[0078] As shown in FIG. 4, while the pump control process for start-up measurement is being performed (time t1 to t3), oxygen that exists in the third internal space 61 before the start-up of the sensor element 101 is pumped out, so the pump current Ip2 becomes high, and the value of the pump current Ip2 does not correspond to the NOx concentration in the measured gas. After time t3 when the oxygen is sufficiently pumped out by the pump control process for start-up measurement, almost all of the oxygen existing in the third internal space 61 becomes oxygen derived from the NOx in the measured gas, so the value of the pump current Ip2 corresponds to the NOx concentration in the measured gas. Therefore, after time t3, it becomes possible to detect the NOx concentration in the measured gas based on the pump current Ip2. Thus, the time (t0 to t3) from the start-up of the sensor element 101 until the value of the pump current Ip2 corresponds to the NOx concentration in the measured gas is the light-off time. In order to make the behavior of the pump current Ip2 easier to understand, FIG. 4 shows a state in which the NOx concentration in the measurement gas is constant, and the pump current Ip2 is constant after time t3. When the sensor element 101 is actually used, the NOx concentration in the measurement gas varies from moment to moment, so the value of the pump current Ip2 after time t3 varies according to the NOx concentration. Therefore, it is difficult to determine whether the sensor element 101 has lit off based on the value of the pump current Ip2. Therefore, in this embodiment, as described above, the CPU 92 determines whether the end condition of the startup measurement pump control process is satisfied based on the value of the pump current Ip1, not the pump current Ip2.

[0079] Here, in this embodiment, at the time of starting up the sensor element 101, as described above, the measurement pump cell 41 is controlled so that the voltage V2 becomes the startup target value V2a*, which is higher than the normal target value V2b*, to pump out oxygen from the third internal space 61. That is, in the startup measurement pump control process, the target value of the oxygen concentration in the third internal space 61 is set to a value lower than that during normal driving, and oxygen is pumped out from the third internal space 61. As a result, compared to the case where the normal measurement pump control process is performed from the start of the sensor element 101, oxygen that exists in the third internal space 61 before the start can be quickly removed from the third internal space 61. For example, the graph (broken line) of the pump current Ip2 of the comparative example shown in FIG. 4 shows the time change of the pump current Ip2 when the adjustment pump control process (main pump control process and auxiliary pump control process) is performed from time t1 and the normal measurement pump control process is performed instead of the startup measurement pump control process from time t1. In this case, the measurement pump cell 41 is controlled so that the voltage V2 becomes the normal target value V2b*, which is lower than the startup target value V2a*. Therefore, the pump current Ip2 does not become large compared to the graph (solid line) of the pump current Ip2 of the embodiment, and the amount of oxygen pumped out from the third internal space 61 becomes small. Therefore, the time when oxygen is sufficiently pumped out from the third internal space 61 is time t4, which is later than time t3. That is, the light-off time is longer in the comparative example than in the embodiment. Thus, in this embodiment, the light-off time of the sensor element 101 can be shortened by performing the startup measurement pump control process instead of the normal measurement pump control process at the time of startup of the sensor element 101. Although not shown in the figure, when the adjustment pump control process (main pump control process and auxiliary pump control process) is performed together with the normal measurement pump control process from time t1, the time until the pump current Ip1 stabilizes near the target current Ip1* is also time t4, which is later than time t3.

[0080] In FIG. 4, the time when the pump current Ip2 becomes constant (the time when the sensor element 101 lights off) and the time when the pump current Ip1 stabilizes near the target current Ip1* are the same as time t3, but in reality, the time when the pump current Ip1 stabilizes near the target current Ip1* may be slightly earlier than time t3. As described above, the auxiliary pump cell 50 can also pump out oxygen that has flowed back from the third internal space 61 to the second internal space 40, but oxygen that is bonded to the constituent material of the measurement electrode 44 and oxygen molecules present in the closed pores of the measurement electrode 44 do not flow back to the second internal space 40, so the auxiliary pump cell 50 cannot pump them out. Therefore, after the pump current Ip1 stabilizes near the target current Ip1*, the sensor element 101 may not light off until the measurement pump cell 41 pumps out the oxygen that the auxiliary pump cell 50 cannot pump out. In this case, it is preferable that the CPU 92 continues the pump control process for startup measurement without performing step S140 from when the pump current Ip1 becomes stable near the target current Ip1* until the sensor element 101 lights off. For example, the CPU 92 may perform step S140 after a predetermined time has elapsed since the positive determination was made in step S130. Alternatively, if the CPU 92 makes a positive determination in step S130 when the state in which the pump current Ip1 is within a predetermined range has continued for a predetermined time or more, the predetermined time may be set to be approximately the same as the time from when the pump current Ip1 becomes stable near the target current Ip1* until the sensor element 101 lights off.

[0081] Here, the correspondence between the components of this embodiment and the components of the present invention will be clarified. A laminate in which six layers, the first substrate layer 1, the second substrate layer 2, the third substrate layer 3, the first solid electrolyte layer 4, the spacer layer 5, and the second solid electrolyte layer 6, are laminated in this order in this embodiment corresponds to the element body of the present invention, the outer pump electrode 23 corresponds to the outer measurement electrode, the third internal space 61 corresponds to the measurement chamber, the measurement electrode 44 corresponds to the inner measurement electrode, the measurement pump cell 41 corresponds to the measurement pump cell, the reference electrode 42 corresponds to the reference electrode, the measurement pump control oxygen partial pressure detection sensor cell 82 corresponds to the measurement voltage detection sensor cell, the sensor element 101 corresponds to the sensor element, the pump current Ip2 corresponds to the measurement pump current, and the control device 90 corresponds to the pump cell control unit and the specific gas concentration detection unit. In addition, the control device 90 corresponds to a heater control section, the first internal space 20 and the second internal space 40 correspond to oxygen concentration adjusting chambers, the inner pump electrode 22 and the auxiliary pump electrode 51 correspond to the inner adjusting pump electrode, the outer pump electrode 23 corresponds to the outer adjusting pump electrode, the main pump cell 21 and the auxiliary pump cell 50 correspond to the adjusting pump cells, the main pump control process and the auxiliary pump control process correspond to the adjusting pump control process. In addition, the inner pump electrode 22 corresponds to the inner main pump electrode, the outer pump electrode 23 corresponds to the outer main pump electrode, the auxiliary pump electrode 51 corresponds to the inner auxiliary pump electrode, the outer pump electrode 23 corresponds to the outer auxiliary pump electrode, and the pump current Ip1 corresponds to the auxiliary pump current.

[0082] According to the gas sensor 100 of this embodiment described above in detail, at the start-up of the sensor element 101, a start-up measurement pump control process is performed to pump out oxygen from the third internal space 61 by controlling the measurement pump cell 41 so that the voltage V2 becomes the start-up target value V2a*, which is higher than the normal target value V2b*, so that oxygen present in the third internal space 61 before the start-up can be removed more quickly than when the normal measurement pump control process is performed from the start-up of the sensor element 101. Therefore, the light-off time of the sensor element 101 is shortened.

[0083] Furthermore, the CPU 92 starts the pump control process for startup measurement when the heater temperature Th reaches or exceeds a predetermined threshold value Thref that is equal to or lower than the target temperature Th* after the heater control process is started. Therefore, the CPU 92 can appropriately determine the start timing of the pump control process for startup measurement based on the heater temperature Th.

[0084] Furthermore, the sensor element 101 has an adjustment pump cell that adjusts the oxygen concentration in the oxygen concentration adjustment chamber provided upstream of the third internal space 61 in the measurement gas flow section. When the sensor element 101 is started up, the CPU 92 performs an adjustment pump control process to operate the adjustment pump cell, and when it is determined that the oxygen concentration in the oxygen concentration adjustment chamber is stable based on the operation of the adjustment pump cell, it switches from the startup measurement pump control process to the normal measurement pump control process. Here, before the sensor element 101 is started up, not only the third internal space 61 but also the oxygen concentration adjustment chamber upstream of it is in a state where there is a large amount of oxygen (oxygen not derived from a specific gas) (for example, air atmosphere, etc.). By performing the adjustment pump cell control process by the CPU 92 when the sensor element 101 is started up, the oxygen that exists in the oxygen concentration adjustment chamber before the sensor element 101 is started up can also be removed. In addition, since the oxygen concentration adjustment chamber and the third internal space 61 are in communication with each other, the oxygen in the third internal space 61 can also be removed by the operation of the adjustment pump cell. Furthermore, since the oxygen concentration adjustment chamber and the third internal space 61 are in communication with each other as described above, when the oxygen concentration in the oxygen concentration adjustment chamber is stabilized, the oxygen in the third internal space 61 is also sufficiently pumped out. Therefore, by switching from the startup-time measurement pump control process to the normal-time measurement pump control process when it is determined that the oxygen concentration in the oxygen concentration adjustment chamber is stabilized based on the operation of the adjustment pump cell, it is possible to appropriately switch from the startup-time measurement pump control process to the normal-time measurement pump control process.

[0085] The oxygen concentration adjusting chamber has a first internal space 20 and a second internal space 40 provided downstream of the first internal space 20 and upstream of the third internal space 61, and the adjusting pump cell has a main pump cell 21 that adjusts the oxygen concentration in the first internal space 20 and an auxiliary pump cell 50 that adjusts the oxygen concentration in the second internal space 40. The adjusting pump control process includes an auxiliary pump control process that controls the auxiliary pump cell 50 so that the oxygen concentration in the second internal space 40 becomes a target concentration, and a main pump control process that controls the main pump cell 21 so that the pump current Ip1 flowing through the auxiliary pump cell 50 by the auxiliary pump control process becomes the target current Ip1*. When the CPU 92 determines that the pump current Ip1 has stabilized near the target current Ip1*, it switches from the start-up measurement pump control process to the normal measurement pump control process. This allows the switching from the start-up measurement pump control process to the normal measurement pump control process to be appropriately performed based on the pump current Ip1.

[0086] The measurement electrode 44 contains at least one of Pt and Rh. Precious metals such as Pt and Rh may combine with oxygen in the third internal space 61 and oxidize, and this oxygen reduces the detection accuracy of the specific gas concentration. Therefore, when the sensor element 101 is started, it is necessary to reduce the oxide of the precious metal contained in the measurement electrode 44 and then pump this oxygen from the third internal space 61. Since it takes more time to reduce the oxide of the precious metal and pump oxygen than to pump oxygen molecules in the third internal space 61, the light-off time is likely to be long if oxidized precious metal is present in the third internal space 61. In the gas sensor 100 of this embodiment, when the sensor element 101 is started, the oxidized precious metal can be reduced early and the light-off time can be shortened by performing the start-up measurement pump control process instead of the normal measurement pump control process. Therefore, in this embodiment in which the measurement electrode 44 contains Pt and Rh, it is very meaningful to perform the start-up measurement pump control process.

[0087] It goes without saying that the present invention is not limited to the above-described embodiment, and can be embodied in various forms within the technical scope of the present invention.

[0088] For example, in the above embodiment, when the CPU 92 determines that the pump current Ip1 has stabilized near the target current Ip1*, it considers that the end condition of the startup measurement pump control process has been established, and switches from the startup measurement pump control process to the normal measurement pump control process, but this is not limited to the above. The end condition of the startup measurement pump control process may be determined so that the startup measurement pump control process can be ended when the sensor element 101 is lighted off, in other words, when the oxygen in the third internal space 61 has been sufficiently pumped out. For example, the startup measurement pump control process may be ended when a predetermined time has elapsed since the heater 72 was started to be energized, or when a predetermined time has elapsed since the startup measurement pump control process was started.

[0089] The CPU 92 may determine whether or not the end condition of the pump control process for measurement at startup is satisfied based on the open measurement voltage V2open, which is the voltage V2 in a state where control of current flow to the measurement electrode 44 and the reference electrode 42 is not performed. FIG. 5 is a flowchart showing an example of the control process for measurement at startup according to a modified example. FIG. 6 is a graph showing an example of the control process for measurement at startup and the control process for normal operation according to a modified example. In FIG. 5, the same steps as those in FIG. 3 are given the same step numbers as those in FIG. 3, and their description is omitted. FIG. 6 is a diagram in which the time change of the measurement voltage V2open in the open state is added to FIG. 4. In the control process for measurement at startup according to the modified example shown in FIG. 5, after starting the pump control process for measurement at startup in step S120, the CPU 92 determines whether or not the end condition of the pump control process for measurement at startup is satisfied by determining whether the measurement voltage V2open in the open state, which is the voltage V2 in a state where control of current flow to the measurement electrode 44 and the reference electrode 42 is not performed, has reached a predetermined threshold value V2opref or more (step S230). For example, in the above-described embodiment, there is no control to pass a current through the reference electrode 42, and after the start-up measurement pump control process is started in step S120, the pump current Ip2 passes through the measurement electrode 44 by the start-up measurement pump control process. Therefore, in step S230, the CPU 92 measures the voltage V2 in a state in which the start-up measurement pump control process is temporarily stopped, specifically, in a state in which the application of the voltage Vp2 from the variable power supply 46 is stopped, and sets the measured value as the open-state measurement voltage V2open. The open-state measurement voltage V2open is a value corresponding to the oxygen concentration in the third internal space 61, and the lower the oxygen concentration, the larger the value becomes. In addition, since the open-state measurement voltage V2open is measured in a state in which no control is performed to pass a current through the measurement electrode 44 and the reference electrode 42, it is not easily affected by a voltage drop due to the current. Therefore, the open-state measurement voltage V2open corresponds more accurately to the oxygen concentration in the third internal space 61 than the voltage V2 measured in a state where control is performed to pass a current through the measurement electrode 44 (the voltage V2 during execution of the startup measurement pump control process). The threshold value V2opref is determined in advance as the value of the open-state measurement voltage V2open in a state where oxygen has been sufficiently pumped out of the third internal space 61.For example, as shown in FIG. 6, when the start-up measurement pump control process is started, the value of the open measurement voltage V2open rises, and becomes a substantially constant value (steady state value) after time t3 when the pumping out of the oxygen existing in the third internal space 61 before the start-up is completed, that is, after the sensor element 101 is turned off. In this embodiment, this steady state value measured in advance by an experiment is set as the threshold value V2opref. In this embodiment, the threshold value V2opref is set to 200 mV. The threshold value V2opref may be a value slightly lower than the steady state value of the open measurement voltage V2open. If the determination in step S230 is negative, the CPU 92 repeatedly executes step S230 and waits until the determination is positive. If the determination in step S230 is positive, the CPU 92 executes the above-mentioned step S140 and ends the start-up control process. As described above, since the open measurement voltage V2open is a value corresponding to the oxygen concentration in the third internal space 61, it is possible to appropriately determine whether or not the oxygen in the third internal space 61 has been sufficiently pumped out by comparing the open measurement voltage V2open with the threshold value V2opref. Therefore, even when the CPU 92 performs step S230 in FIG. 5 instead of step S130 in FIG. 3, it is possible to appropriately switch from the start-up measurement pump control process to the normal measurement pump control process. As described above, the time when the pump current Ip1 stabilizes near the target current Ip1* may be slightly earlier than the time t3 when the sensor element 101 lights off. In contrast, since the open measurement voltage V2open is a value corresponding to the oxygen itself present in the third internal space 61, unlike the pump current Ip1, there is almost no time difference between the time when the open measurement voltage V2open becomes a steady value and the time when the sensor element 101 lights off.

[0090] In the above embodiment, the oxygen in the third internal space 61 is pumped around the outer pump electrode 23 in the start-up measurement pump control process, but the oxygen may be pumped around an electrode arranged outside the gas flow section, not limited to the outer pump electrode 23. For example, in the start-up measurement pump control process, a voltage may be applied between the measurement electrode 44 and the reference electrode 42 to flow a pump current, thereby pumping the oxygen in the third internal space 61 around the reference electrode 42. That is, in the start-up measurement pump control process of the above embodiment, the oxygen in the third internal space 61 is pumped outside the element body, which is around the outer pump electrode 23, but it may be pumped inside the element body, such as around the reference electrode 42. When pumping oxygen into the element body, it is preferable to pump it around an electrode arranged outside the measurement gas flow section, such as the reference electrode 42. In other words, when pumping oxygen into the element body, it is preferable to pump oxygen into a region of the element body that is not connected to the third internal space 61.

[0091] Although not specifically described in the above embodiment, the voltage Vp2 applied to the measurement pump cell 41 is preferably less than 1500 mV. When the voltage Vp2 is 1500 mV or more, oxygen ions in the solid electrolyte are deficient, electronic conduction of the solid electrolyte is generated, and the sensor element 101 may become black and become unusable. However, by setting the voltage Vp2 to less than 1500 mV, such a defect can be suppressed. The higher the start-up target value V2a* is, the higher the voltage Vp2 during the start-up measurement pump control process becomes, so it is preferable to set the start-up target value V2a* as a value that does not make the voltage Vp2 1500 mV or more. In addition, if the start-up measurement pump control process is performed in a state in which the solid electrolyte is not sufficiently activated, oxygen in the third internal space 61 is not pumped out and the voltage V2 does not approach the start-up target value V2a*, so that the voltage Vp2 is likely to become a high value due to feedback control. Therefore, if the start-up measurement pump control process is performed when the solid electrolyte is not sufficiently activated, the voltage Vp2 is likely to exceed 1500 mV during the start-up measurement pump control process. In consideration of this, it is preferable to appropriately determine the start condition of the start-up measurement pump control process (the value of the threshold value Thref in the above embodiment) so that the voltage Vp2 does not exceed 1500 mV.

[0092] In the heater control process of the above embodiment, the CPU 92 controls the heater 72 so that the heater temperature Th becomes the target temperature Th*, but the heater 72 may be controlled so that the sensor element temperature, which is the temperature of the heater 72 or the element body, becomes the target temperature, without being limited to the heater temperature Th. For example, the resistance value of the circuit of the measurement pump cell 41 or the resistance value of the circuit of the measurement pump control oxygen partial pressure detection sensor cell 82 may be measured as a value representing the temperature of the element body (a value that can be converted to temperature), and the heater 72 may be controlled so that this resistance value becomes the target resistance value. Similarly, in the above embodiment, the startup measurement pump control process is started when the heater temperature Th reaches a predetermined threshold value Thref or more, but this is not limited to this. The startup measurement pump control process may be started when the sensor element temperature reaches a predetermined threshold value or more (for example, when the resistance value of the circuit of the measurement pump cell 41 reaches a predetermined threshold value or less), without being limited to the heater temperature Th. Alternatively, the gas sensor 100 may be provided with a temperature detection unit such as a thermocouple, and the sensor element temperature (the temperature of the heater 72 or the temperature of the element body) itself may be directly measured by this temperature sensor. In this case, the CPU 92 may control the heater 72 based on the measured sensor element temperature so that the sensor element temperature becomes a target temperature.

[0093] In the above-described embodiment, the CPU 92 determines that the start condition for the startup measurement pump control process is satisfied when the heater temperature Th reaches or exceeds a predetermined threshold value Thref, but this is not limited to the above. The start condition for the startup measurement pump control process may be determined in advance so that the startup measurement pump control process can be started when the solid electrolyte contained in the measurement pump cell 41 controlled by the startup measurement pump control process is activated. For example, the startup measurement pump control process may be started when a predetermined time has elapsed since the heater 72 was started to be energized.

[0094] Alternatively, the start condition of the start-up measurement pump control process may be determined so that the start-up measurement pump control process can be started when the state becomes suitable for pumping oxygen bonded to the constituent material of the measurement electrode 44 according to the constituent material of the measurement electrode 44. For example, the start condition of the start-up measurement pump control process may be determined based on a phase diagram of the constituent material of the measurement electrode 44. An example of the start condition when the measurement electrode 44 contains Rh as a constituent material will be described. FIG. 7 is a phase diagram of Rh2O3 (Source: VK Tagirov, DM Chizhikov, EK Kazenas, and LK Shubochkin, Zh. Neorg. Khim., Title: Study of thermal dissociation of ruthenium dioxide and rhodium sesquioxide, Journal: Zhurnal Neorganicheskoj Khimii; v. 20(8);p. 2035-2037 (1975)). The vertical axis of FIG. 7 is the oxygen partial pressure, and the horizontal axis is the temperature. The thick line in Fig. 7 is a line (equilibrium line) showing the relationship between oxygen partial pressure and temperature at which Rh2O3 and Rh are in equilibrium. Based on this equilibrium line, a start condition for the startup measurement pump control process may be determined so that the startup measurement pump control process is performed in a state suitable for the reduction of the oxide (here, Rh2O3) of the constituent material (here, Rh) of the measurement electrode 44 (i.e., a state in Fig. 7 where the corresponding point between temperature and oxygen partial pressure is located in the region to the lower left of the equilibrium line). For example, the logarithm (logP O2) is -2 (i.e., when the oxygen partial pressure is 0.01 Torr), the temperature of the measurement electrode 44 on the equilibrium line corresponding to this oxygen partial pressure is Ta°C (about 630°C), as shown by the dashed line added to the state diagram of FIG. 7. Therefore, if the voltage V2 is near the startup target value V2a* by the startup measurement pump control process and the temperature of the measurement electrode 44 is Ta°C or higher, Rh2O3 is easily reduced to Rh. Therefore, a start condition for the startup measurement pump control process may be determined so that the startup measurement pump control process can be started when the temperature of the measurement electrode 44 becomes Ta°C or higher. For example, a threshold value Thref may be determined as a lower limit value of the heater temperature Th required to make the temperature of the measurement electrode 44 Ta°C or higher. Alternatively, the startup measurement pump control process may be started when a predetermined time has elapsed since the start of energization of the heater 72, and the predetermined time may be determined based on the time required for the temperature of the measurement electrode 44 to become Ta°C or higher from the start of the heater control process. As can be seen from FIG. 7, the higher the temperature of the measurement electrode 44 and the lower the oxygen partial pressure in the third internal space 61, the easier it is to reduce Rh2O3. In other words, the lower the oxygen partial pressure in the third internal space 61, the lower the temperature of the measurement electrode 44 suitable for starting the start-up measurement pump control process. Therefore, if the start-up target value V2a* is set to a large value, the start-up measurement pump control process can be started in a shorter time from the start of the heater control process. However, as described above, it is preferable to set the start-up target value V2a* as a value that does not cause the voltage Vp2 to be 1500 mV or more, so it is preferable to set the value of the start-up target value V2a* so that the start-up target value V2a* does not become too large, taking this point into consideration. Also, as described above, it is preferable to perform the start-up measurement pump control process after the solid electrolyte is activated, so it is preferable to set the start-up measurement pump control process based on the higher temperature between the temperature required for activating the solid electrolyte and the temperature suitable for reducing the oxide of the constituent material of the measurement electrode 44. When the measuring electrode 44 contains Pt as a constituent material, the start condition of the pump control process for startup measurement suitable for reducing Pt oxide may be determined based on the Pt phase diagram in the same manner as in the above example.Furthermore, when the measurement electrode 44 contains both Pt and Rh as constituent materials, it is preferable to determine the start condition of the startup measurement pump control process based on the higher of the temperature suitable for the reduction of Pt oxide and the temperature suitable for the reduction of Rh oxide.

[0095] In the above embodiment, the CPU 92 starts the adjustment pump control process (main pump control process and auxiliary pump control process) at the same time as starting the startup measurement pump control process, but this is not limited to this. There may be a time difference between the start timing of the adjustment pump control process and the start timing of the startup measurement pump control process. Since the adjustment pump cells (main pump cell 21 and auxiliary pump cell 50) have a higher oxygen pumping capacity than the measurement pump cell 41, it is preferable to start the adjustment pump control process (main pump control process and auxiliary pump control process) at the same time as or before the start of the startup measurement pump control process. In addition, it is preferable to also perform the adjustment pump control process (main pump control process and auxiliary pump control process) during the startup measurement pump control process.

[0096] In the above-described embodiment, the oxygen concentration adjusting chamber has the first internal space 20 and the second internal space 40, but is not limited thereto. For example, the oxygen concentration adjusting chamber may have another internal space, or one of the first internal space 20 and the second internal space 40 may be omitted. Similarly, in the above-described embodiment, the adjusting pump cell has the main pump cell 21 and the auxiliary pump cell 50, but is not limited thereto. For example, the adjusting pump cell may have another pump cell, or one of the main pump cell 21 and the auxiliary pump cell 50 may be omitted. For example, if the oxygen concentration of the measurement gas can be sufficiently reduced by the main pump cell 21 alone, the auxiliary pump cell 50 may be omitted. In this case, the CPU 92 may perform, as the adjusting pump control process, a process of controlling the main pump cell 21 so that the oxygen concentration in the oxygen concentration adjusting chamber (first internal space 20) becomes a target concentration. More specifically, the target value V0* may be determined in advance, and the CPU 92 may control the main pump cell 21 by feedback-controlling the pump voltage Vp0 of the variable power source 24 so that the voltage V0 becomes the target value V0* (i.e., so that the oxygen concentration in the first internal space 20 becomes the target concentration). In this case, when the CPU 92 determines that the oxygen concentration in the first internal space 20 is stable based on the operation of the main pump cell 21, the CPU 92 may switch from the pump control process for start-up measurement to the pump control process for normal measurement. For example, the CPU 92 may determine whether the oxygen concentration in the first internal space 20 is stable based on the value of the voltage V0 or the value of the pump current Ip0. More specifically, the CPU 92 may determine that the oxygen concentration in the first internal space 20 is stable when the voltage V0 reaches a predetermined threshold value or more, or may determine that the oxygen concentration in the first internal space 20 is stable when the pump current Ip0 rises once, then drops, and reaches a predetermined threshold value or less.

[0097] In the above-described embodiment, the outer pump electrode 23 serves as an outer main pump electrode which is a part of the main pump cell 21 and is disposed in a portion of the sensor element 101 exposed to the measured gas, an outer auxiliary pump electrode which is a part of the auxiliary pump cell 50 and is disposed in a portion of the sensor element 101 exposed to the measured gas, and an outer measurement electrode which is a part of the measurement pump cell 41 and is disposed in a portion of the sensor element 101 exposed to the measured gas, but is not limited to this. Any one or more of the outer main pump electrode, the outer auxiliary pump electrode, and the outer measurement electrode may be provided outside the sensor element 101 separately from the outer pump electrode 23.

[0098] In the above-described embodiment, the outer pump electrode 23 is exposed to the outside of the sensor element 101, but the present invention is not limited thereto and the outer pump electrode 23 may be provided outside the element body (layers 1 to 6) so as to be in contact with the measured gas. For example, the sensor element 101 may include a porous protective layer that covers the element body (layers 1 to 6), and the outer pump electrode 23 may also be covered by the porous protective layer.

[0099] In the above-mentioned embodiment, the sensor element 101 of the gas sensor 100 includes the first internal space 20, the second internal space 40, and the third internal space 61, but is not limited thereto. For example, the sensor element 201 of FIG. 8 may not include the third internal space 61. In the modified sensor element 201 shown in FIG. 8, the gas inlet 10, the first diffusion rate-controlling section 11, the buffer space 12, the second diffusion rate-controlling section 13, the first internal space 20, the third diffusion rate-controlling section 30, and the second internal space 40 are adjacently formed in this order between the lower surface of the second solid electrolyte layer 6 and the upper surface of the first solid electrolyte layer 4 in a manner that they communicate with each other. The measurement electrode 44 is disposed on the upper surface of the first solid electrolyte layer 4 in the second internal space 40. The measurement electrode 44 is covered with the fourth diffusion rate-controlling section 45. The fourth diffusion rate-controlling portion 45 is a film made of a ceramic porous body such as alumina (Al2O3). The fourth diffusion rate-controlling portion 45, like the fourth diffusion rate-controlling portion 60 of the above-mentioned embodiment, serves to limit the amount of NOx flowing into the measurement electrode 44. The fourth diffusion rate-controlling portion 45 also functions as a protective film for the measurement electrode 44. The ceiling electrode portion 51a of the auxiliary pump electrode 51 is formed up to just above the measurement electrode 44. Even with the sensor element 201 having such a configuration, the NOx concentration can be detected based on, for example, the pump current Ip2, like the above-mentioned embodiment. In this case, the periphery of the measurement electrode 44 functions as a measurement chamber.

[0100] In the above-mentioned embodiment, the element body of the sensor element 101 is a laminate having a plurality of solid electrolyte layers (layers 1 to 6), but is not limited thereto. The element body of the sensor element 101 may include at least one solid electrolyte layer having oxygen ion conductivity, and may have a measurement gas flow section provided therein. For example, in FIG. 1, layers 1 to 5 other than the second solid electrolyte layer 6 may be structure layers made of a material other than a solid electrolyte (for example, a layer made of alumina). In this case, each electrode of the sensor element 101 may be disposed on the second solid electrolyte layer 6. For example, the measurement electrode 44 in FIG. 1 may be disposed on the lower surface of the second solid electrolyte layer 6. Also, the reference gas introduction space 43 may be disposed on the spacer layer 5 instead of the first solid electrolyte layer 4, the air introduction layer 48 may be disposed between the second solid electrolyte layer 6 and the spacer layer 5 instead of between the first solid electrolyte layer 4 and the third substrate layer 3, and the reference electrode 42 may be disposed behind the third internal space 61 and on the lower surface of the second solid electrolyte layer 6.

[0101] In the above embodiment, the control device 90 sets (feedback controls) the target value V0* of the voltage V0 based on the pump current Ip1 so that the pump current Ip1 becomes the target current Ip1* in the main pump control process, and feedback controls the pump voltage Vp0 so that the voltage V0 becomes the target value V0*. However, other controls may be performed. For example, the control device 90 may feedback control the pump voltage Vp0 based on the pump current Ip1 in the main pump control process so that the pump current Ip1 becomes the target current Ip1*. That is, the control device 90 may omit obtaining the voltage V0 from the oxygen partial pressure detection sensor cell 80 for controlling the main pump and setting the target value V0*, and directly control the pump voltage Vp0 (and thus control the pump current Ip0) based on the pump current Ip1. In this case, the CPU 92 can also determine whether the end condition of the start-up measurement pump control process is satisfied by using, for example, the process of step S130 in FIG. 3 or step S230 in FIG. 5 described above.

[0102] In the above-described embodiment, the gas sensor 100 detects the NOx concentration as the specific gas concentration, but the specific gas concentration is not limited to this, and other oxide concentrations may be used as the specific gas concentration. When the specific gas is an oxide, oxygen is generated when the specific gas itself is reduced in the third internal space 61 as in the above-described embodiment, and the CPU 92 can detect the specific gas concentration based on a detection value corresponding to this oxygen. The specific gas may also be a non-oxide such as ammonia. When the specific gas is a non-oxide, the specific gas is converted, for example, to an oxide in the first internal space 20 (for example, ammonia is oxidized and converted to NO), and oxygen is generated when the converted oxide is reduced in the third internal space 61, and the CPU 92 can obtain a detection value corresponding to this oxygen to detect the specific gas concentration. In this way, whether the specific gas is an oxide or a non-oxide, the gas sensor 100 can detect the specific gas concentration based on the oxygen generated in the third internal space 61 derived from the specific gas.

[0103] In the above-described embodiment, the value of the startup target value V2a* may be set according to the volume of the measurement electrode 44. The inventors have found that even if the startup target value V2a* is the same, the light-off time may differ depending on the volume of the measurement electrode 44, that if the startup target value V2a* is too high or too low, the effect of shortening the light-off time may be small, that there is an optimal value for the startup target value V2a*, and that if the volume of the measurement electrode 44 changes, the optimal value for the startup target value V2a* also changes. This will be described in detail below.

[0104] The inventors have determined the volume C [mm 3 The relationship between the difference ΔV (=V2a*-V2b*) [mV] between the startup target value V2a* and the normal target value V2b*, and the light-off time [sec] was examined as follows. First, the sensor element 101 and the gas sensor 100 of the embodiment described above with reference to Figs. 1 and 2 were used as Experimental Example 1. The volume C of the measurement electrode 44 in Experimental Example 1 was 4 x 10 -3 mm 3The measurement electrode 44 was a porous cermet electrode of Pt, Rh and ZrO2 as described above. The gas sensor 100 was the same as that of the experiment example 1 except that the volume C was changed as shown in Table 1 by changing the thickness and the length in the front-rear direction of the measurement electrode 44. Specifically, the thickness of the measurement electrode 44 in the experiment example 2 was twice as large as that of the experiment example 1. The thickness of the measurement electrode 44 in the experiment example 3 was twice as large as that of the experiment example 2. The thickness of the measurement electrode 44 in the experiment example 4 was 1.5 times as large as that of the experiment example 2 and the length in the front-rear direction was twice as large as that of the experiment example 2. The thickness of the measurement electrode 44 in the experiment example 5 was twice as large as that of the experiment example 2. The light-off time was examined for the gas sensor 100 in the experiment example 1 when the startup control process shown in FIG. 3 was performed with the startup target value V2a* set to 500 mV and the normal target value V2b* set to 400 mV (therefore, the difference ΔV was set to 100 mV). Similarly, in the gas sensor 100 of Experimental Example 1, the difference ΔV was changed by 10 mV increments between 110 mV and 200 mV as shown in Table 1 by changing the startup target value V2a*, and the light-off time in each case was examined. Similarly, in the gas sensors 100 of Experimental Examples 2 to 5, the difference ΔV was changed between 100 mV and 200 mV, and the light-off time in each case was examined. Table 1 summarizes the volume C of the measurement electrode 44 and the light-off time corresponding to the difference ΔV (=V2a*-V2b*) between the startup target value V2a* and the normal target value V2b* in each of Experimental Examples 1 to 5. FIG. 9 is a graph of Table 1, showing the relationship between the volume C of the measurement electrode 44, the difference ΔV, and the light-off time. In addition, in all of the experimental examples 1 to 5, when the difference ΔV was within the range of 100 mV to 200 mV, the light-off time was shorter than when the pump control process for startup measurement was not performed (in other words, when the difference ΔV was 0 mV). Therefore, all of the experimental examples 1 to 5 correspond to the embodiments of the present invention.

[0105] [Table 1]

[0106] As can be seen from Table 1 and FIG. 9, it was confirmed that the light-off time is not necessarily shorter as the difference ΔV is larger, in other words, the light-off time is not necessarily shorter as the start-up target value V2a* is higher. For example, in Experimental Example 3, the light-off time was shortest when the difference ΔV was 160 mV, and it was confirmed that the light-off time tends to be longer as the difference ΔV becomes lower or higher than 160 mV. A similar tendency was confirmed in Experimental Examples 2 and 4. In Experimental Example 1, it was confirmed that the light-off time tends to be longer as the difference ΔV becomes larger in the range of the difference ΔV of 100 mV to 200 mV. Therefore, in Experimental Example 1, it is presumed that the value of the difference ΔV that can shorten the light-off time is less than 100 mV. In Experimental Example 5, it was confirmed that the light-off time tends to be shorter as the difference ΔV becomes larger in the range of the difference ΔV of 100 mV to 200 mV. Therefore, it is presumed that the value of the difference ΔV that can shorten the light-off time is greater than 200 mV in Experimental Example 5.

[0107] The reason why the relationship between the volume C of the measurement electrode 44, the difference ΔV, and the light-off time has the above-mentioned tendency will be described. First, the inventors have found that when switching from the pump control process for measurement at startup to the pump control process for measurement at normal time, an undershoot of the pump current Ip2 may occur. FIG. 10 is a graph showing an example of an undershoot of the pump current Ip2. Times t0, t1, and t3 in FIG. 10 are the same as the times shown in FIG. 4. As shown in FIG. 10, when switching from the pump control process for measurement at startup to the pump control process for measurement at normal time is performed at time t3, the pump current Ip2 does not immediately become a value corresponding to the NOx concentration in the measurement gas, but an undershoot of the pump current Ip2 occurs, and the value of the pump current Ip2 may become a value corresponding to the NOx concentration in the measurement gas after time t3'. In this case, the light-off time is the time from time t0 to t3', so the light-off time is extended by the period during which the undershoot occurs (time t3 to t3').

[0108] The reason for such an undershoot is considered as follows. When switching from the pump control process for startup measurement to the pump control process for normal measurement, the target value of the voltage V2 changes from the startup target value V2a* to the normal target value V2b* (<V2a*). Therefore, the oxygen concentration in the third internal cavity 61 immediately after the switch may be lower than the oxygen concentration corresponding to the normal target value V2b*. For this reason, immediately after switching to the pump control process for normal measurement, the control unit 91 may control the measurement pump cell 41 to draw oxygen into the third internal cavity 61 instead of pumping oxygen out of the third internal cavity 61 in order to increase the oxygen concentration in the third internal cavity 61 to the oxygen concentration corresponding to the normal target value V2b*. In this case, the pump current Ip2 becomes a negative value (the positive and negative are reversed compared to the pump current Ip2 during the pump control process for startup measurement). Also, during the pump control process for startup measurement, water in the gas to be measured may be decomposed to generate hydrogen due to the voltage Vp2 applied to the measurement electrode 44. After switching to the pump control process for normal measurement, the hydrogen reacts with oxygen, resulting in a decrease in the oxygen concentration in the third internal cavity 61. Thus, this may also cause the pump current Ip2 to decrease or become a negative value. Also, the larger the volume C of the measurement electrode 44, the larger the surface area of the measurement electrode 44 (including not only the outer surface of the measurement electrode 44 but also the surface area of the pores inside the measurement electrode 44). Therefore, the larger the volume C, the greater the tendency for the amount of hydrogen generated during the pump control process for startup measurement to increase. From the above, the larger the difference ΔV (=V2a* - V2b*) and the larger the volume C, the longer the period during which the undershoot occurs when switching from the pump control process for startup measurement to the pump control process for normal measurement, and the longer the light-off time tends to be.

[0109] On the other hand, the larger the difference ΔV (= V2a* - V2b*), the more quickly the oxygen present in the third internal space 61 before the start of the sensor element 101 can be removed in the startup measurement pump control process. Therefore, the larger the difference ΔV (= V2a* - V2b*), the shorter the time during which the startup measurement pump control process is performed (e.g., times t1 to t3 in FIG. 10) and the shorter the light-off time tends to be. And the shorter the total time of the time during which the startup measurement pump control process is performed (e.g., times t1 to t3 in FIG. 10) and the time during which undershoot occurs (e.g., times t3 to t3' in FIG. 10), the shorter the light-off time (e.g., times t0 to t3' in FIG. 10) can be.

[0110] From the above, it is considered that if the value of the startup target value V2a*, more precisely, the difference ΔV, is too small, the time for which the startup measurement pump control process is performed becomes longer (the time required to pump out the oxygen present in the third internal space 61 before startup becomes longer), thereby reducing the effect of shortening the light-off time. Also, if the difference ΔV is too large, the time during which the undershoot occurs becomes longer, thereby reducing the effect of shortening the light-off time. Therefore, it is considered that there is an optimal difference ΔV for a certain sensor element 101 at which the effect of shortening the light-off time is the highest. Also, as described above, the time during which the undershoot occurs changes depending on the volume C of the measurement electrode 44, so when the volume C changes, the optimal difference ΔV at which the effect of shortening the light-off time is the highest also changes, and specifically, it is considered that the optimal difference ΔV tends to increase as the volume C increases. For the above reasons, it is considered that the relationship between the volume C of the measurement electrode 44, the difference ΔV, and the light-off time is as shown in Table 1 and FIG. 9.

[0111] Here, the volume C of the measurement electrode 44 is 8×10 -3 mm 3 More than 8×10 is preferable. -3 mm 3 In the above condition, the pumping capacity of the measurement pump cell 41 for pumping out oxygen around the measurement electrode 44 is sufficiently high. Therefore, compared to the experiment example 1, the volume C is 8×10 -3 mm 3Experimental Examples 2 to 5, which are equal to or greater than 100 mV, are preferred. If the difference ΔV is too large, the voltage Vp2 during the start-up measurement pump control process will be high, and as described above, the sensor element 101 may turn black and become unusable. If the voltage Vp2 during the start-up measurement pump control process is too high, excessive voltage Vp2 is applied to the measurement electrode 44, which may accelerate deterioration of the measurement electrode 44, increasing the resistance of the measurement electrode 44 and decreasing the sensitivity of the NOx concentration detection. For this reason, the difference ΔV is preferably 200 mV or less. From the results of FIG. 9, it is presumed that if the volume C is larger than that of Experimental Example 5, the effect of shortening the light-off time will be small in the range of the difference ΔV being 200 mV or less. Therefore, the volume C should be equal to or less than the value of Experimental Example 5, that is, 32×10 -3 mm 3 In addition, in Experimental Examples 2 to 5, the light-off time can be set to the minimum value or its vicinity, and the difference ΔV is preferably in the range of 120 mV to 200 mV, so that the difference ΔV does not exceed 200 mV. From the above, the volume C is 8×10 -3 mm 3 32×10 or more -3 mm 3 It is preferable that the difference ΔV is 120 mV or more and 200 mV or less. The larger the volume C, the larger the difference ΔV required to sufficiently increase the effect of shortening the light-off time tends to be. -3 mm 3 32×10 or more -3 mm 3 If the difference ΔV is 120 mV or more and 200 mV or less, the effect of shortening the light-off time can be sufficiently increased while satisfying the above-mentioned lower limit of the volume C and the upper limit of the difference ΔV.

[0112] In addition, when adjusting the volume C of the measurement electrode 44, the thickness of the measurement electrode 44 may be, for example, 10 μm or more and 40 μm or less. The area of ​​the upper surface of the measurement electrode 44, in other words, the product of the left-right width and the front-back length of the measurement electrode 44, is 0.2 mm 2 More than 1.2mm 2The left-right width of the measurement electrode 44 may be 0.5 mm or more and 2.5 mm or less. In adjusting the difference ΔV, the startup target value V2a* is preferably 600 mV or less.

[0113] In the above-described embodiment, the CPU 92 may determine that the end condition of the start-up measurement pump control process is satisfied when it is determined that the oxygen concentration in the third internal space 61 is stable based on the operation of the measurement pump cell 41. For example, the CPU 92 may determine that the oxygen concentration in the third internal space 61 is stable and determine that the end condition of the start-up measurement pump control process is satisfied when the pump current Ip2 flowing through the measurement pump cell 41 is stabilized by the start-up measurement pump control process. The present inventors have found that a stable region in which the pump current Ip2 is stable appears as the oxygen concentration in the third internal space 61 becomes stable during the start-up measurement pump control process. FIG. 11 is an explanatory diagram of the stable region of the pump current Ip2 and the first and second derivatives of the pump current Ip2. Times t0, t1, and t3 in FIG. 11 are the same as the times shown in FIG. 4, and the start-up measurement pump control process is started at time t1, and the normal control process is started at time t3. Also, time t3' in FIG. 11 is the same as time t3' shown in FIG. 10. As shown in FIG. 11A, a stable region where the pump current Ip2 is stable may appear between the start of the pump control process for measurement at startup at time t1 and time t3 (the pump current Ip1 is omitted in FIG. 11) when it is determined that the pump current Ip1 is stabilized near the target current Ip1*. In the pump control process for measurement at startup, the CPU 92 feedback-controls the voltage Vp2 of the variable power supply 46 so that the voltage V2 becomes the target value V2a* at startup. Therefore, the fact that the pump current Ip2 is stable at this time is considered to mean that the oxygen concentration in the third internal space 61 is stable, and that the oxygen present in the third internal space 61 before startup has been sufficiently pumped out by the pump control process for measurement at startup. Therefore, by switching from the pump control process for measurement at startup to the pump control process for measurement at normal time when the pump current Ip2 is stable, it is possible to appropriately switch from the pump control process for measurement at startup to the pump control process for measurement at normal time. As shown in FIG. 11, the stable region of the pump current Ip2 may appear at a timing earlier than time t3.For example, when the oxygen concentration in the measurement gas is low, the pump current Ip2 may become stable earlier than the pump current Ip1 becomes stable near the target current Ip1*, and the stable region of the pump current Ip2 may appear earlier than time t3 as shown in FIG. 11. This is considered to be due to the fact that the volume of the space in the third internal space 61 is smaller than that in the second internal space 40, and the target concentration of the oxygen concentration in the third internal space 61 is smaller than that in the second internal space 40. In this case, by switching from the pump control process for start-up measurement to the pump control process for normal measurement when the pump current Ip2 becomes stable, the switching can be performed earlier than when the pump current Ip1 becomes stable near the target current Ip1* as in the above-mentioned embodiment. This can also shorten the light-off time of the sensor element 101. Conversely, the pump current Ip1 may become stable near the target current Ip1* earlier than the time when the pump current Ip2 becomes stable. This is because, as described above, the auxiliary pump cell 50 cannot pump out oxygen that does not flow back from the third internal space 61 to the second internal space 40 (oxygen bonded to the constituent material of the measurement electrode 44 or oxygen molecules present in the closed pores of the measurement electrode 44). In this case, if the pump control process for startup measurement is switched to the pump control process for normal measurement when the pump current Ip1 is stabilized near the target current Ip1*, the oxygen in the third internal space 61 may not be sufficiently pumped out and the switching may be too early. In contrast, by switching from the pump control process for startup measurement to the pump control process for normal measurement when the pump current Ip2 is stabilized, it is possible to prevent the switching from being too early.

[0114] The determination of whether the pump current Ip2 is stable may be performed based on the first derivative of the pump current Ip2 flowing through the measurement pump cell 41 by the start-up measurement pump control process. For example, as shown in FIG. 11A and FIG. 11B, when the pump current Ip2 reaches a stable region, the first derivative (dIp2 / dt) of the pump current Ip2 becomes 0 or close to 0. Therefore, when this first derivative can be regarded as 0 (i.e., when the slope of the pump current Ip2 can be regarded as 0), the pump current Ip2 may be determined to be stable. For example, in the graph of the time change of the first derivative of the pump current Ip2 shown in FIG. 11B, the pump current Ip2 can be determined to be stable at the time tb when the first derivative becomes 0. The CPU 92 may determine that the pump current Ip2 is stable when the first derivative is included in a predetermined region that can be regarded as 0. This predetermined region may be, for example, a region close to 0, or a region exceeding a predetermined threshold value slightly smaller than 0. For example, the predetermined threshold value may be set to -0.3 [μA / s], and the pump current Ip2 may be determined to be stable when (dIp2 / dt)>-0.3 [μA / s] is satisfied. The first derivative (dIp2 / dt) of the pump current Ip2 may be calculated by repeatedly measuring the pump current Ip2 at a predetermined time interval, and dividing the difference between the current measured value and the previous measured value by the predetermined time interval. As can be seen from FIG. 11B, even after the pump control process for startup measurement is started at time t1 and before the pump current Ip2 enters the stable region, the first derivative of the pump current Ip2 may be 0 or close to 0. Therefore, it is preferable that the CPU 92 determines whether the pump current Ip2 is stable or not based on the first derivative of the pump current Ip2 after the time when the pump control process for startup measurement is started and the pump current Ip2 increases once and then starts to decrease (the time when the pump current Ip2 exceeds the peak, for example, time ta in FIG. 11B). For example, the CPU 92 may determine that the pump current Ip2 has increased once and then started to decrease when the pump current Ip2 has exceeded a predetermined threshold and then fallen below the predetermined threshold. Alternatively, the CPU 92 may determine that the pump current Ip2 has increased once and then started to decrease when the first derivative of the pump current Ip2 falls below a predetermined threshold (negative value).

[0115] The determination as to whether the pump current Ip2 has stabilized may be made based on a second-order differential value of the pump current Ip2 flowing through the measurement pump cell 41 by the startup measurement pump control process. For example, as shown in FIGS. 11A and 11C, when the pump current Ip2 reaches a stable region, the second-order differential value (d 2 Ip2 / dt 2) becomes a value of 0 or near a value of 0. Therefore, when the second derivative of can be regarded as a value of 0 (i.e., when it can be regarded as an inflection point of the pump current Ip2), it may be determined that the pump current Ip2 is stable. For example, in the graph of the time change of the second derivative of the pump current Ip2 shown in the lower part of FIG. 12, it can be determined that the pump current Ip2 is stable at the time tc when the second derivative becomes a value of 0. The CPU 92 may determine that the pump current Ip2 is stable when the second derivative is included in a predetermined region where the value is regarded as a value of 0. This predetermined region may be, for example, a region near the value 0, or a region slightly larger than the value 0 and less than a predetermined threshold value. The second derivative of the pump current Ip2 (dIp2 / dt) may be calculated by, for example, repeatedly measuring the pump current Ip2 at a predetermined time interval to repeatedly derive the first derivative, and dividing the difference between the first derivative derived this time and the first derivative derived last time by the predetermined time interval. In the case where the determination is made based on the second derivative, similarly to the case where the determination is made based on the first derivative, it is preferable that the CPU 92 determines whether the pump current Ip2 is stable or not based on the second derivative of the pump current Ip2 after the start of the pump control process for startup measurement and the pump current Ip2 increases once and starts to decrease. In addition, the second derivative of the pump current Ip2 may become close to 0 after the time when the pump current Ip2 increases once and starts to decrease and before the pump current Ip2 is stabilized (for example, between time ta and time tb in FIG. 11C). In order to distinguish this case from the case where the pump current Ip2 is stabilized, the CPU 92 may determine that the pump current Ip2 is stable when the second derivative of the pump current Ip2 is included in the area that can be regarded as the value 0 from a state higher than a predetermined area that can be regarded as the value 0 (when the second derivative decreases and enters the predetermined area that can be regarded as the value 0) after the time when the pump current Ip2 increases once and starts to decrease.

[0116] As described in the above embodiment, if the NOx concentration in the measurement gas varies from moment to moment, the value of the pump current Ip2 varies according to the NOx concentration. Therefore, if the NOx concentration in the measurement gas varies during the start-up measurement pump control process, the stable region of the pump current Ip2 may not appear. Therefore, when the NOx concentration in the measurement gas can be considered stable, the CPU 92 preferably determines whether the end condition of the start-up measurement pump control process is satisfied based on whether the pump current Ip2 is stable. When the NOx concentration in the measurement gas can be considered stable, the measurement gas may be considered to have no NOx. Specific examples of the cases when the NOx concentration in the measurement gas can be considered to be stable include low load operation of the internal combustion engine, idling, and deceleration of the vehicle to which the gas sensor 100 is attached. When the vehicle to which the gas sensor 100 is attached is a hybrid vehicle, the hybrid vehicle's electric operation (engine stopped) can also be cited as a specific example of the cases when the NOx concentration in the measurement gas can be considered to be stable. The CPU 92 may determine whether the NOx concentration in the measured gas can be considered stable (for example, whether it falls under any of the specific examples above) based on information obtained from an engine ECU (not shown) of the internal combustion engine, and if the determination is positive, may determine whether the pump current Ip2 has stabilized.

[0117] Alternatively, the CPU 92 may perform a combination of a process for determining whether the end condition of the start-up measurement pump control process is satisfied based on whether the oxygen concentration in the third internal space 61 is stable, and a process for determining whether the end condition of the start-up measurement pump control process is satisfied in any of the above-mentioned modes. That is, whether the end condition of the start-up measurement pump control process is satisfied may be determined based on whether any of a plurality of conditions is satisfied. In this way, even if it is not possible to determine whether the end condition of the start-up measurement pump control process is satisfied based on whether the oxygen concentration in the third internal space 61 is stable (for example, when the NOx concentration in the measurement gas is fluctuating), it is possible to switch from the start-up measurement pump control process to the normal measurement pump control process by determining whether another condition is satisfied (for example, the determination in step S130 in FIG. 3 described above). In this case, the determination of whether the NOx concentration in the measurement gas can be regarded as stable can be omitted.

[0118] An example of a case where a combination of a plurality of determinations is used to determine whether or not the end condition of the start-up measurement pump control process is satisfied will be described. FIG. 12 is a flowchart showing an example of the start-up measurement pump control process according to a modified example. In FIG. 12, the same steps as those in FIG. 3 are given the same step numbers as those in FIG. 3, and the description thereof will be omitted. In the start-up measurement pump control process in FIG. 12, after starting the start-up measurement pump control process in step S120, the CPU 92 determines whether or not the oxygen concentration in the measurement chamber (the third internal space 61 in this case) is stable (step S125). For example, when the pump current Ip2 increases once and then starts to decrease, and the CPU 92 determines that the pump current Ip2 is stable based on at least one of the first-order differential value and the second-order differential value of the pump current Ip2, the CPU 92 determines that the oxygen concentration in the third internal space 61 is stable. If the pump current Ip2 is rising, if it has risen but has not yet started to decrease, or if the pump current Ip2 is not stable, it is determined in step S125 that the oxygen concentration in the third internal space 61 is not stable, and the process of step S130 is performed. If the determination in step S130 is negative, the process of step S125 is performed. If the determination in either step S125 or step S130 is positive, the CPU 92 executes step S140 and transitions to normal control processing. In the startup control processing of FIG. 12, if a stable region of the pump current Ip2 appears, the process of step S125 can transition to the normal control processing, so that the startup measurement pump control processing can be appropriately switched to the normal measurement pump control processing. Even if the stable region of the pump current Ip2 does not appear, the process of step S130 can transition to the normal control processing at time t3, so that the transition to the normal control processing can be reliably performed. In Fig. 12, step S125 and step S130 are combined to determine whether the end condition of the startup measurement pump control process is satisfied, but step S125 may be combined with step S230 in Fig. 5, or steps S125, S130, and S230 may be combined. In the startup control process in Fig. 12, a determination may be added as to whether the NOx concentration in the measurement gas is considered to be stable.For example, after step S120 in Fig. 12, the CPU 92 may determine whether the NOx concentration in the measured gas can be regarded as stable, and if the NOx concentration can be regarded as stable, perform the determination of step S125, and if the NOx concentration cannot be regarded as stable, perform the determination of step S130. If the CPU 92 makes a negative determination in step S125 or step S130, it may again determine whether the NOx concentration in the measured gas can be regarded as stable, and if the CPU 92 makes a positive determination in step S125 or step S130, it may perform step S140. In other words, if the NOx concentration in the measured gas can be regarded as stable, the determination of step S130 may not be performed. By doing this, even if the pump current Ip1 stabilizes near the target current Ip1* too early and using the judgment of step S130 would result in switching from the startup measurement pump control process to the normal measurement pump control process too early, the judgment of step S125 is made if the NOx concentration is stable, so that switching can be done at an appropriate timing that is not too early.

[0119] The voltage V1 (an example of the auxiliary pump voltage) detected by the auxiliary pump control oxygen partial pressure detection sensor cell 81 (an example of the auxiliary pump voltage detection sensor cell) of the above-mentioned embodiment includes not only the electromotive force based on the oxygen concentration difference between the periphery of the auxiliary pump electrode 51 and the periphery of the reference electrode 42, but also the thermoelectromotive force based on the temperature difference between the auxiliary pump electrode 51 and the reference electrode 42. Therefore, in the above-mentioned normal control process, a correction may be performed taking into account the thermoelectromotive force contained in the voltage V1. Specifically, in the initial period of the normal control process, a correction process may be performed to correct the target value V1* of the voltage V1 to a higher value compared to a period after the initial period. Here, the thermoelectromotive force contained in the voltage V1 will be described first. FIG. 13 is a graph showing an example of the time change of the voltage V1 caused by the thermoelectromotive force. The time t0 in FIG. 13 is the same as the time t0 shown in FIG. 4. In FIG. 13, in order to make it easier to understand the time change of the voltage V1 caused by the thermoelectromotive force, the state of the voltage V1 when the oxygen concentration difference between the periphery of the auxiliary pump electrode 51 and the periphery of the reference electrode 42 is constant is shown. When the heater control process is started at time t0, both the auxiliary pump electrode 51 and the reference electrode 42 are heated and the temperature rises, and after a sufficient time has passed, the temperatures of the auxiliary pump electrode 51 and the reference electrode 42 become stable, and the thermoelectromotive force between the two electrodes also becomes stable at a substantially constant value. However, before the temperatures of the two electrodes become stable, the temperature difference between the two electrodes may temporarily become large due to the difference in the temperature rise rate between the two electrodes. When the temperature difference between the two electrodes is temporarily large, the thermoelectromotive force contained in the voltage V1 between the two electrodes also becomes temporarily large. Therefore, even if the electromotive force based on the oxygen concentration difference between the periphery of the auxiliary pump electrode 51 and the periphery of the reference electrode 42 is constant, the voltage V1 becomes temporarily high after the heater control process is started at time t0, as shown by the solid line in FIG. 13. Then, as the temperature difference between the two electrodes becomes smaller with the passage of time, the voltage V1 decreases, and at time te when the thermoelectromotive force becomes stable, the voltage V1 becomes the stable value A. Value A includes an electromotive force based on the oxygen concentration difference between the periphery of the auxiliary pump electrode 51 and the periphery of the reference electrode 42, and a thermoelectromotive force between the two electrodes in a stable state in which the temperature difference between the two electrodes is small.Therefore, for example, at time td before time te, voltage V1 contains value B based on a temporarily large thermoelectromotive force between the auxiliary pump electrode 51 and the reference electrode 42 in addition to value A, so that voltage V1 becomes large by the amount of value B. Therefore, at time td, the actual value of voltage V1 (=A+B) and the value of the electromotive force based on the oxygen concentration difference between the periphery of the auxiliary pump electrode 51 and the periphery of the reference electrode 42, which is the electromotive force that is actually desired to be detected (almost equal to value A), deviate by the amount of value B. In this manner, in a state in which the deviation between the actual voltage V1 and the electromotive force based on the oxygen concentration difference between the periphery of the auxiliary pump electrode 51 and the periphery of the reference electrode 42 is temporarily large, the oxygen concentration around the auxiliary pump electrode 51 may not be accurately controlled even if the above-mentioned auxiliary pump control process is performed. Specifically, even if the voltage Vp1 of the variable power supply 52 is feedback-controlled so that the voltage V1 becomes the target value V1* in the auxiliary pump control process, the voltage V1 temporarily contains a large thermoelectromotive force, and therefore the oxygen concentration difference between the periphery of the auxiliary pump electrode 51 and the periphery of the reference electrode 42 is controlled to a value smaller than the oxygen concentration difference that is originally intended to be adjusted. As a result, the oxygen concentration around the auxiliary pump electrode 51 is controlled to a value higher than the target concentration that is originally intended to be adjusted (a value closer to the oxygen concentration around the reference electrode 42 than the target concentration). When the oxygen concentration around the auxiliary pump electrode 51 is controlled to a value higher than the target concentration, the amount of NOx reduced in the second internal space 40 changes, and the amount of NOx reaching the third internal space 61 also changes. Specifically, when the oxygen concentration around the auxiliary pump electrode 51 is controlled to a target concentration, the NOx in the measurement gas is slightly reduced in the second internal space 40. However, when the oxygen concentration around the auxiliary pump electrode 51 is controlled to a value higher than the target concentration, the amount of reduced NOx decreases, and as a result, the amount of NOx reaching the third internal space 61 may increase. This causes the pump current Ip2 flowing in the normal measurement pump control process to increase, causing an error in the detection value of the NOx concentration. FIG. 14 is a graph showing an example of the time change of the pump current Ip2 caused by the thermoelectromotive force contained in the voltage V1. Times t0, td, and te in FIG. 14 are the same as the times shown in FIG. 13.In addition, in order to make it easier to understand the time change of the pump current Ip2 caused by the thermoelectromotive force, Fig. 14 shows the state of the pump current Ip2 when the NOx concentration in the measured gas is constant. When the value of the voltage V1 temporarily increases due to the thermoelectromotive force as shown by the solid line in Fig. 13, the pump current Ip2 also temporarily increases for the above reason as shown by the solid line in Fig. 14, and an error occurs from the NOx concentration in the measured gas that is originally to be measured. Therefore, even if the NOx concentration in the measured gas is constant, the pump current Ip2 temporarily becomes high after the heater control process is started at time t0 as shown in Fig. 14. Then, as the thermoelectromotive force decreases over time, the pump current Ip2 decreases, and at time te when the thermoelectromotive force stabilizes, the pump current Ip2 becomes a stable value D (the value of the pump current Ip2 corresponding to the NOx concentration in the measured gas that is originally to be measured). Therefore, for example, at time td prior to time te, as shown by the solid line in Figure 13, voltage V1 includes value B resulting from thermoelectromotive force, causing pump current Ip2 to become greater than value D, resulting in an error in the detection value of the NOx concentration and reducing the detection accuracy of the NOx concentration.

[0120] Therefore, as described above, a correction process is performed to correct the target value V1* of the voltage V1 to a higher value in the early stage of the normal control process compared to the period after the early stage. By increasing the target value V1* of the voltage V1 in this way in the early stage of the normal control process, the influence of the thermoelectromotive force contained in the voltage V1 being temporarily large can be reduced. As a result, compared to the case where the correction process is not performed, the oxygen concentration around the auxiliary pump electrode 51 can be accurately controlled when the thermoelectromotive force contained in the voltage V1 is temporarily large. For example, if the target value V1* is corrected to a value higher than the original value by a value B at the time td in FIG. 13, the influence of the voltage V1 being temporarily larger by a value B due to the thermoelectromotive force can be offset, and the oxygen concentration in the second internal space 40 can be controlled at the target concentration. As a result, the temporary increase in the pump current Ip2 shown in FIG. 14 can be suppressed, and the deterioration of the detection accuracy of the NOx concentration can be suppressed.

[0121] The correction of the target value V1* may be performed based on a correction pattern that is determined in advance so that the correction amount tends to become smaller as time passes. As described above, the temperature difference between the auxiliary pump electrode 51 and the reference electrode 42 tends to become smaller as time passes. Therefore, the thermoelectromotive force included in the voltage V1 also tends to become smaller as time passes. Therefore, by correcting the target value V1* of the voltage V1 based on a correction pattern that is determined in advance so that the correction amount tends to become smaller as time passes, it is possible to further reduce the influence of the thermoelectromotive force included in the voltage V1 being temporarily large. For example, a correction pattern that follows the curve of the time change of the voltage V1 shown by the solid line in FIG. 13 may be determined in advance by experiment, and a correspondence relationship between the time (e.g., the time elapsed from the time t0) and the correction amount (or the value of the corrected target value V1*) may be stored in the storage unit 94. Then, the CPU 92 may derive the correction amount corresponding to the current time based on this correspondence relationship to derive the corrected target value V1*. In this way, the temporary increase (deviation from value A) of the voltage V1 shown by the solid line in FIG. 13 is offset, and the deviation of the pump current Ip2 from the value D can be reduced as shown by the dashed line in FIG. 14 in the state with correction, compared to the pump current Ip2 without correction shown by the solid line. Although the pump current Ip2 with correction shown by the dashed line in FIG. 14 deviates slightly from the value D, ideally, the deviation from the value D can be reduced to zero by correction. The pattern of the time change of the voltage V1 shown by the solid line in FIG. 13, that is, the pattern of the time change of the thermoelectromotive force between the auxiliary pump electrode 51 and the reference electrode 42, is basically the same pattern of time change if the temperature rise pattern of the heater 72 in the heater control process is the same. Therefore, the correction pattern can be determined in advance by experiment. Furthermore, by using the correction pattern prepared in advance in this way, the target value V1* can be appropriately corrected without measuring the thermoelectromotive force between the auxiliary pump electrode 51 and the reference electrode 42.

[0122] The "tendency for the correction amount to become smaller as time passes" includes, for example, a case where the correction amount becomes smaller linearly, a case where the correction amount becomes smaller curvedly, and a case where the correction amount becomes smaller step-function-like. For example, instead of using the correction pattern along the curve of the time change of the voltage V1 shown by the solid line in FIG. 13 as described above, a correction pattern that becomes smaller step-function-like as shown by the dashed line in FIG. 13 may be used. In this case, it is preferable to determine the correction pattern so that the difference between the correction pattern shown by the dashed line and the time change of the actual voltage V1 shown by the solid line (for example, the value C at time td) is not too large (for example, within 10 mV). Alternatively, instead of using the correction pattern along the curve of the time change of the voltage V1 shown by the solid line in FIG. 13, a correction pattern along an approximation straight line based on this curve may be used.

[0123] An example of the case where the target value V1* is corrected based on the predetermined correction pattern will be described. FIG. 15 is a flow chart showing an example of the correction process. This correction process is executed when the normal control process of step S140 shown in FIGS. 3, 5, and 12 is started. When the correction process is started, the CPU 92 derives a correction amount ΔV1* of the target value V1* based on the correction pattern (step S310). For example, when a correction pattern along the curve shown by the solid line in FIG. 13 is stored in the storage unit 94, the CPU 92 derives a correction amount ΔV1* corresponding to the current time (for example, the elapsed time from the time t0 when the heater control process is started) from this correction pattern. For example, if the current time is time td in FIG. 13, the CPU 92 derives a value B as the correction amount ΔV1*. Next, the CPU 92 corrects the target value V1* based on the derived correction amount ΔV1* (step S320). For example, the CPU 92 derives the corrected target value V1* by adding the correction amount ΔV1* to the original target value V1*. Next, the CPU 92 determines whether or not the correction period of the target value V1* has ended (step S330). For example, the CPU 92 determines that the correction period has ended when the current time has passed the last time of the correction pattern stored in the storage unit 94 or when the current time has passed a predetermined time. For example, the CPU 92 determines that the correction period has ended when the current time has passed the time te in FIG. 13 or when the current time has passed the time (a predetermined time slightly before the time te) at which the difference between the solid line and the value A in FIG. 13 becomes negligible. If the CPU 92 determines in step S330 that the correction period has not ended, it executes the process of step S310. As a result, the corrected target value V1* is set so that the correction amount tends to become smaller as time passes based on the correction pattern until the correction period ends. Then, when the CPU 92 determines in step S330 that the correction period has ended, it ends the correction process. In this way, the auxiliary pump control process is performed based on the corrected target value V1* during the initial period of the normal control process, that is, from the start of the normal control process to the end of the correction period, so that the oxygen concentration around the auxiliary pump electrode 51 can be accurately controlled even if the thermoelectromotive force contained in the voltage V1 is temporarily large as described above.As a result, a temporary increase in the pump current Ip2 can be suppressed, and a decrease in the detection accuracy of the NOx concentration can be suppressed.

[0124] Instead of correcting the target value V1* based on a predetermined correction pattern, the CPU 92 may correct the target value V1* in the correction process in such a manner that the correction amount increases as the difference between the calculated reference voltage Vrefs, which is the voltage Vref derived based on the pump current Ip0 and corresponding to the oxygen concentration in the measurement gas around the sensor element 101, and the measured reference voltage Vrefa, which is the voltage Vref detected by the sensor cell 83 (an example of a reference voltage detection sensor cell), increases, when the pump current Ip0 (an example of a main pump current) flowing through the main pump cell 21 during the main pump control process is considered to be stable. Here, the main pump control process is a process for controlling the main pump cell 21 so that the pump current Ip1 becomes the target current Ip1* as described above, and the pump current Ip0 flowing when adjusting the oxygen concentration in the first internal space 20 by the main pump control process is correlated with the oxygen concentration in the measurement gas around the sensor element 101. In addition, the voltage Vref (an example of a reference voltage) is a voltage between the outer pump electrode 23 (an example of a measured gas side electrode) on the outside of the sensor element 101 and the reference electrode 42, and is therefore correlated with the oxygen concentration in the measured gas around the sensor element 101. That is, the pump current Ip0 is correlated with the oxygen concentration in the measured gas around the sensor element 101, and the oxygen concentration in the measured gas around the sensor element 101 is correlated with the voltage Vref. Therefore, when the pump current Ip0 is considered to be stable, it is possible to derive a calculated reference voltage Vrefs, which is the voltage Vref corresponding to the oxygen concentration in the measured gas around the sensor element 101, based on the pump current Ip0. The calculated reference voltage Vrefs is a value equivalent to an electromotive force based on the oxygen concentration difference between the measured gas around the sensor element 101 and the reference gas around the reference electrode 42. On the other hand, since the measured reference voltage Vrefa, which is the voltage Vref actually detected by the sensor cell 83, also includes the thermoelectromotive force between the outer pump electrode 23 and the reference electrode 42, if the temperature difference between the two electrodes is temporarily large, the thermoelectromotive force included in the measured reference voltage Vrefa also becomes temporarily large.Therefore, the larger the thermoelectromotive force included in the measured reference voltage Vrefa, the larger the deviation between the calculated reference voltage Vrefs derived based on the main pump current Ip0 and the measured reference voltage Vrefa actually detected by the sensor cell 83. In addition, when the thermoelectromotive force between the outer pump electrode 23 and the reference electrode 42 is large, the thermoelectromotive force between the auxiliary pump electrode 51 and the reference electrode 42 is often large. Therefore, by correcting the target value V1* in the correction process with a tendency that the correction amount increases as the deviation between the calculated reference voltage Vrefs and the measured reference voltage Vrefa increases, the influence of the thermoelectromotive force included in the voltage V1 being temporarily large can be reduced. As a result, similar to the case where the above-mentioned correction pattern is used, the oxygen concentration around the auxiliary pump electrode 51 can be accurately controlled when the thermoelectromotive force included in the voltage V1 is temporarily large, compared to the case where the correction process is not performed. As a result, the temporary increase in the pump current Ip2 can be suppressed, and the deterioration of the detection accuracy of the NOx concentration can be suppressed.

[0125] For example, the correspondence between the pump current Ip0 and the calculated reference voltage Vrefs may be stored in the storage unit 94 in advance, and the CPU 92 may derive the calculated reference voltage Vrefs based on this correspondence and the value of the pump current Ip0 flowing through the main pump cell 21. The calculated reference voltage Vrefs is a value corresponding to the actual measurement value of the voltage Vref after a sufficient amount of time has passed since the start of the heater control process, the temperature difference between the outer pump electrode 23 and the reference electrode 42 has become small, and the thermoelectromotive force between the two electrodes has stabilized. Therefore, such a correspondence can be determined in advance by measuring combinations of the actual measurement value of the voltage Vref after the thermoelectromotive force has stabilized and the value of the pump current Ip0 during the main pump control at that time while changing the oxygen concentration in the measurement gas around the sensor element 101. This correspondence is such that the larger the value of the pump current Ip0 (i.e., the higher the oxygen concentration around the sensor element 101), the smaller the calculated reference voltage Vrefs (i.e., the smaller the oxygen concentration difference between the measurement gas around the sensor element 101 and the reference gas).

[0126] An example of the case where the target value V1* is corrected based on the calculated reference voltage Vrefs and the measured reference voltage Vrefa in this way will be described. FIG. 16 is a flow chart showing an example of a correction process different from that shown in FIG. 15. In FIG. 16, the same processes as those in FIG. 15 are given the same step numbers as those in FIG. 15, and detailed descriptions thereof will be omitted. The correction process in FIG. 16 is executed when the normal control process of step S140 shown in FIGS. 3, 5, and 12 is started. When the correction process is started, the CPU 92 determines whether the pump current Ip0 flowing in the main pump control process is stable or not, in other words, whether the oxygen concentration of the measurement gas around the sensor element 101 is stable or not (step S300). For example, the CPU 92 repeatedly measures the pump current Ip0 at predetermined time intervals, and determines that the pump current Ip0 is stable when the difference between the current measurement value of the pump current Ip0 and the previous measurement value is within a predetermined range. Alternatively, the CPU 92 may determine whether the pump current Ip0 is stable based on a first or second derivative of the pump current Ip0, similar to the determination of whether the pump current Ip2 is stable described above. If it is determined in step S300 that the pump current Ip0 is not stable, the CPU 92 performs the process of step S310 described in FIG. 15 to derive a correction amount ΔV1* based on a correction pattern previously stored in the storage unit 94. On the other hand, if it is determined in step S300 that the pump current Ip0 is stable, the CPU 92 derives a calculated reference voltage Vrefs based on the pump current Ip0 at that time and the above-mentioned correspondence relationship (step S312). Next, the CPU 92 derives a correction amount ΔV1* of the target value V1* based on the derived calculated reference voltage Vrefs and the measured reference voltage Vrefa, which is the current voltage Vref detected by the sensor cell 83 (step S314). For example, the CPU 92 may derive the correction amount ΔV1* based on the difference between the measured reference voltage Vrefa and the calculated reference voltage Vrefs. The difference between the measured reference voltage Vrefa and the calculated reference voltage Vrefs corresponds to a temporary increase in the thermoelectromotive force between the outer pump electrode 23 and the reference electrode 42. Therefore, this difference may be regarded as being the same as the temporary increase in the thermoelectromotive force between the auxiliary pump electrode 51 and the reference electrode 42, and the value of this difference itself may be used as the correction amount ΔV1*.Alternatively, the correction amount ΔV1* may be derived as a value obtained by multiplying the difference between the measured reference voltage Vrefa and the calculated reference voltage Vrefs by a predetermined coefficient. When the correction amount ΔV1* is derived in step S310 or step S314, the CPU 92 performs step S320 to correct the target value V1* based on the correction amount ΔV1*. Then, when the CPU 92 determines in step S330 that the correction period has not ended, it executes the process of step S300, and when it determines that the correction period has ended, it ends the correction process. In step S330, it may be determined that the correction period has ended when the correction amount ΔV1* derived in step S314 is a small value that can be regarded as zero (for example, when it is a value less than a predetermined threshold value). By performing the correction process of FIG. 16, when the pump current Ip0 is stable, the target value V1* can be corrected based on the calculated reference voltage Vrefs and the measured reference voltage refa. On the other hand, when the pump current Ip0 is not stable, the target value V1* can be corrected based on a predetermined correction pattern. Regardless of whether the correction amount ΔV1* is derived in the process of step S310 or S314, the correction can be performed taking into account the thermoelectromotive force contained in the voltage V1.

[0127] In addition, when the temperatures of the auxiliary pump electrode 51 and the reference electrode 42 are not yet stable at the beginning of the normal control process, the voltage V1 contains a temporarily increased thermoelectromotive force as shown in FIG. 13, and the pump current Ip2 temporarily increases as shown in FIG. 14. Therefore, the sensor element 101 lights off when the thermoelectromotive force stabilizes (for example, at time te in FIG. 14). For this reason, when the normal control process is started from time t3 in FIG. 4, if time t3 is time td in FIG. 14, there is an error in the pump current Ip2 from time t3 (time td) to time te in a state where the correction process is not performed, and the light-off does not occur, so the actual light-off time may be long. In contrast, by performing the correction process to suppress the temporary increase in the pump current Ip2, the sensor element 101 can be lighted off from time td or immediately after time td, for example, and the light-off time can be shortened.

[0128] As described above, the element body of the sensor element 101 has an elongated shape having a front end and a rear end, which are both ends along the longitudinal direction. As shown in FIG. 1, the auxiliary pump electrode 51 and the reference electrode 42 are disposed on the front end side of the element body (specifically, closer to the front end than the center in the front-rear direction), and the reference electrode 42 is disposed at a position farther from the front end in the front-rear direction than the auxiliary pump electrode 51 (i.e., rearward). When the auxiliary pump electrode 51 and the reference electrode 42 are disposed in such a positional relationship, a difference in the temperature rise rate between the auxiliary pump electrode 51 and the reference electrode 42 occurs from the start of heating the sensor element 101 by the heater 72 until the temperatures of both electrodes become stable, and the temperature difference between the two electrodes is likely to temporarily increase. For this reason, it is highly significant to perform the above-mentioned correction process. Generally, the inner pump electrode 22 of the electrodes of the sensor element 101 is located at the most distal end, and the main pump cell 21 pumps out a larger amount of oxygen than the auxiliary pump cell 50 and the measurement pump cell 41, so that the heater 72 is often arranged so that the inner pump electrode 22 has the highest temperature among the electrodes. Therefore, the fact that the auxiliary pump electrode 51 is arranged closer to the inner pump electrode 22 than the reference electrode 42 in the front-to-rear direction is also cited as a reason why the temperature difference between the auxiliary pump electrode 51 and the reference electrode 42 is likely to become temporarily large.

[0129] In addition, the outer pump electrode 23 is disposed on the front end side of the element body of the sensor element 101 (specifically, closer to the front end than the center in the front-rear direction), and the reference electrode 42 is disposed at a position farther from the front end in the front-rear direction than the outer pump electrode 23 (i.e., rearward). When the outer pump electrode 23 and the reference electrode 42 are disposed in such a positional relationship, when the temperature difference between the auxiliary pump electrode 51 and the reference electrode 42 is temporarily large, the temperature difference between the outer pump electrode 23 and the reference electrode 42 is also likely to be temporarily large. Therefore, in the case of such an electrode positional relationship, the larger the deviation between the calculated reference voltage Vrefs and the measured reference voltage Vrefa, the larger the thermoelectromotive force contained in the voltage V1 is, so that this is suitable for correction based on the deviation. Note that, as shown in FIG. 1, the outer pump electrode 23 is disposed at a position closer to the inner pump electrode 22 than the reference electrode 42 in the front-rear direction, which is also cited as a reason why the temperature difference between the outer pump electrode 23 and the reference electrode 42 is likely to be temporarily large.

[0130] In the above-described embodiment, the outer pump electrode 23 is a part of the sensor cell 83 and serves as both the measured gas side electrode provided outside the element body of the sensor element 101 so as to be in contact with the measured gas and another electrode (for example, the outer main pump electrode of the main pump cell 21). However, the present invention is not limited to this. The measured gas side electrode may be provided outside the element body of the sensor element 101 separately from the outer pump electrode 23.

[0131] In the above-described embodiment, the CPU 92 performs the startup measurement pump control process before proceeding to the normal control process, but the startup measurement pump control process may not be performed. In this case, the normal control process may be started after starting the heater control process. For example, the CPU 92 may start the normal control process when a positive determination is made in step S110 in FIG. 3, FIG. 5, and FIG. 12. In this case, too, by performing the above-described correction process at the beginning of the normal control process, the influence of the thermoelectromotive force contained in the voltage V1 being temporarily large can be reduced, and thus the deterioration of the detection accuracy of the specific gas concentration can be suppressed.

[0132] The CPU 92 may also perform the above-described correction process during the auxiliary pump control process during the startup control process before the start of the normal control process. However, since the specific gas concentration is not detected based on the pump current Ip2 during the startup control process, there is little need to perform the correction process.

[0133] The gas sensor of the present invention may be configured as follows.

[0134] The gas sensor of the present invention comprises: an element body including an oxygen ion conductive solid electrolyte layer and having a measurement gas flow section therein through which a measurement gas is introduced and flows; a measurement pump cell including an outer measurement electrode provided on the outer side of the element body so as to come into contact with the measurement gas, and an inner measurement electrode disposed in a measurement chamber of the measurement gas flow portion, the measurement pump cell pumping oxygen from a periphery of the inner measurement electrode to a periphery of the outer measurement electrode; a reference electrode disposed inside the element body so as to come into contact with a reference gas serving as a reference for detecting a specific gas concentration in the measurement gas; a measurement voltage detection sensor cell for detecting a measurement voltage between the reference electrode and the inner measurement electrode; A sensor element having a pump cell control unit that performs a normal measurement pump control process for pumping oxygen from the measurement chamber by controlling the measurement pump cell so that the measurement voltage becomes a normal target value during normal operation of the sensor element, and performs a startup measurement pump control process for pumping oxygen from the measurement chamber by controlling the measurement pump cell so that the measurement voltage becomes a startup target value higher than the normal target value during startup of the sensor element prior to the normal operation; a specific gas concentration detection unit that detects a specific gas concentration in the measurement target gas based on a measurement pump current flowing through the measurement pump cell by the normal measurement pump control process; It is equipped with the following:

[0135] In this gas sensor, during normal operation of the sensor element, a normal measurement pump control process is performed to pump oxygen from the measurement chamber by controlling the measurement pump cell so that the measurement voltage becomes a normal target value. Then, the specific gas concentration in the measurement gas is detected based on the measurement pump current flowing through the measurement pump cell by the normal measurement pump control process. In addition, in this gas sensor, during startup of the sensor element before normal operation, a startup measurement pump control process is performed to pump oxygen from the measurement chamber by controlling the measurement pump cell so that the measurement voltage becomes a startup target value higher than the normal target value. That is, in the startup measurement pump control process, the target value of the oxygen concentration in the measurement chamber is set to a value lower than that during normal operation, and oxygen is pumped from the measurement chamber. As a result, oxygen present in the measurement chamber before the startup of the sensor element can be quickly removed from the measurement chamber compared to the case where the normal measurement pump control process is performed from the startup of the sensor element. Therefore, the light-off time of the sensor element is shortened. Here, the oxygen present in the measurement chamber before the sensor element is activated includes, for example, oxygen molecules (O2) present in the space within the measurement chamber, oxygen molecules (O2) attached to the surface of the inner measurement electrode, and oxygen bonded to the constituent material of the inner measurement electrode (oxidizing the constituent material).

[0136] The gas sensor of the present invention includes a heater for heating the element body, and a heater control unit for conducting a heater control process to energize the heater to generate heat so that a sensor element temperature, which is the temperature of the heater or the element body, becomes a predetermined target temperature. The pump cell control unit may start the pump control process for start-up measurement when the heater control process is started and the sensor element temperature reaches a predetermined threshold value that is equal to or lower than the target temperature. In this way, the start timing of the pump control process for start-up measurement can be appropriately determined based on the sensor element temperature. The predetermined threshold value may be a value lower than the target temperature. Here, "conducting the heater to generate heat so that the sensor element temperature becomes a predetermined target temperature" includes a case where the heater is controlled based on the sensor element temperature itself and a case where the heater is controlled based on a value that can be converted to the sensor element temperature (e.g., the resistance value of the heater, the resistance value of an electric circuit including a solid electrolyte, etc.). Examples of the resistance value of the electric circuit including a solid electrolyte include the resistance value of a measurement pump cell and the resistance value of a measurement voltage detection sensor cell.

[0137] In the gas sensor of the present invention, the sensor element has an adjustment pump cell that adjusts the oxygen concentration in the oxygen concentration adjustment chamber provided upstream of the measurement chamber in the measurement gas flow section, and the pump cell control unit may perform an adjustment pump control process to operate the adjustment pump cell at the start-up of the sensor element, and when it is determined that the oxygen concentration in the oxygen concentration adjustment chamber is stable based on the operation of the adjustment pump cell, switch from the start-up measurement pump control process to the normal measurement pump control process. Here, before the start-up of the sensor element, not only the measurement chamber but also the oxygen concentration adjustment chamber upstream of the measurement chamber is in a state where there is a large amount of oxygen (oxygen not derived from a specific gas) (for example, air atmosphere, etc.). By performing the adjustment pump cell control process by the pump cell control unit at the start-up of the sensor element, the oxygen that exists in the oxygen concentration adjustment chamber before the start-up of the sensor element can also be removed. In addition, since the oxygen concentration adjustment chamber and the measurement chamber are in communication with each other, the oxygen in the measurement chamber can also be removed by the operation of the adjustment pump cell. Furthermore, since the oxygen concentration adjustment chamber and the measurement chamber are in communication with each other as described above, when the oxygen concentration in the oxygen concentration adjustment chamber is stabilized, the oxygen in the measurement chamber is also sufficiently pumped out. Therefore, when it is determined that the oxygen concentration in the oxygen concentration adjustment chamber is stabilized based on the operation of the adjustment pump cell, the startup measurement pump control process is switched to the normal measurement pump control process, so that the startup measurement pump control process can be appropriately switched to the normal measurement pump control process. Here, "when it is determined that the oxygen concentration in the oxygen concentration adjustment chamber is stabilized, the startup measurement pump control process is switched to the normal measurement pump control process" includes a case where the switching is performed immediately when it is determined that the oxygen concentration in the oxygen concentration adjustment chamber is stabilized, and a case where the switching is performed after the timing at which it is determined that the oxygen concentration in the oxygen concentration adjustment chamber is stabilized. An example of the latter is a case where the switching is performed when other conditions are satisfied in addition to the determination that the oxygen concentration in the oxygen concentration adjustment chamber is stabilized, such as a case where the switching is performed after a predetermined time has elapsed since it is determined that the oxygen concentration in the oxygen concentration adjustment chamber is stabilized.

[0138] In this case, the adjustment pump cell may include an inner adjustment pump electrode arranged in the oxygen concentration adjustment chamber, and an outer adjustment pump electrode arranged in a portion exposed to the measured gas on the outside of the element body.

[0139] In the gas sensor of the present invention having an adjustment pump cell, the oxygen concentration adjustment chamber has a first internal space and a second internal space provided downstream of the first internal space and upstream of the measurement chamber, the adjustment pump cell has a main pump cell that adjusts the oxygen concentration in the first internal space and an auxiliary pump cell that adjusts the oxygen concentration in the second internal space, the adjustment pump control process includes an auxiliary pump control process that controls the auxiliary pump cell so that the oxygen concentration in the second internal space becomes a target concentration, and a main pump control process that controls the main pump cell so that the auxiliary pump current flowing through the auxiliary pump cell by the auxiliary pump control process becomes a target current, and the pump cell control unit may switch from the startup measurement pump control process to the normal measurement pump control process when it is determined that the auxiliary pump current has stabilized near the target current. That is, when it is determined that the auxiliary pump current has stabilized near the target current, it may be determined that the oxygen concentration in the oxygen concentration adjustment chamber has stabilized. In this way, switching from the startup-time measurement pump control process to the normal-time measurement pump control process can be appropriately performed based on the auxiliary pump current.

[0140] In this case, the main pump cell may include an inner main pump electrode disposed in the first internal space and an outer main pump electrode disposed in a portion of the element body exposed to the measurement gas, and the auxiliary pump cell may include an inner auxiliary pump electrode disposed in the second internal space and an outer auxiliary pump electrode disposed in a portion of the element body exposed to the measurement gas.

[0141] In the gas sensor of the present invention, the pump cell control unit may switch from the startup measurement pump control process to the normal measurement pump control process when an open measurement voltage, which is the measurement voltage in a state where no control is performed to cause a current to flow to the inner measurement electrode and the reference electrode, reaches a predetermined threshold value or more. Since the open measurement voltage is a value corresponding to the oxygen concentration in the measurement chamber, it is possible to appropriately determine whether or not the oxygen in the measurement chamber has been sufficiently pumped out by comparing the open measurement voltage with the threshold value. Therefore, the startup measurement pump control process can be appropriately switched to the normal measurement pump control process.

[0142] This application claims priority from Japanese Patent Application No. 2020-204045, filed on December 9, 2020, the entire contents of which are incorporated herein by reference. [Industrial Applicability]

[0143] INDUSTRIAL APPLICABILITY The present invention can be used in a gas sensor for detecting the concentration of a specific gas, such as NOx, in a measurement gas, such as an exhaust gas from an automobile. [Explanation of symbols]

[0144] 1 first substrate layer, 2 second substrate layer, 3 third substrate layer, 4 first solid electrolyte layer, 5 spacer layer, 6 second solid electrolyte layer, 10 gas inlet, 11 first diffusion rate limiting portion, 12 buffer space, 13 second diffusion rate limiting portion, 20 first internal space, 21 main pump cell, 22 inner pump electrode, 22a ceiling electrode portion, 22b bottom electrode portion, 23 outer pump electrode, 24 variable power supply, 30 third diffusion rate limiting portion, 40 second internal space, 41 measurement pump cell, 42 reference electrode, 43 reference gas introduction space, 44 measurement electrode, 45 fourth diffusion rate limiting portion, 46 variable power supply, 48 air introduction layer, 50 auxiliary pump cell, 51 auxiliary pump electrode, 51a ceiling electrode portion, 51b bottom electrode portion, 52 variable power supply, 60 fourth diffusion rate limiting portion, 61 third internal space, 70 Heater section, 71 connector electrode, 72 heater, 73 through hole, 74 heater insulating layer, 75 pressure relief hole, 76 heater power supply, 80 oxygen partial pressure detection sensor cell for controlling main pump, 81 oxygen partial pressure detection sensor cell for controlling auxiliary pump, 82 oxygen partial pressure detection sensor cell for controlling measurement pump, 83 sensor cell, 90 control device, 91 control section, 92 CPU, 94 memory section, 100 gas sensor, 101, 201 sensor element.

Claims

1. an element body including an oxygen ion conductive solid electrolyte layer and having a measurement gas flow section therein through which a measurement gas is introduced and flows; a measurement pump cell including an outer measurement electrode provided on the outer side of the element body so as to come into contact with the measurement gas, and an inner measurement electrode disposed in a measurement chamber of the measurement gas flow portion, the measurement pump cell pumping oxygen from a periphery of the inner measurement electrode to a periphery of the outer measurement electrode; a reference electrode disposed inside the element body so as to come into contact with a reference gas serving as a reference for detecting a specific gas concentration in the measurement gas; a measurement voltage detection sensor cell for detecting a measurement voltage between the reference electrode and the inner measurement electrode; a main pump cell for adjusting an oxygen concentration in a first internal space provided in the measurement gas flow portion on the upstream side of the measurement chamber; an auxiliary pump cell including an inner auxiliary pump electrode disposed in a second internal space provided in the measurement gas flow portion downstream of the first internal space and upstream of the measurement chamber, the auxiliary pump cell adjusting the oxygen concentration in the second internal space; an auxiliary pump voltage detection sensor cell for detecting an auxiliary pump voltage between the reference electrode and the inner auxiliary pump electrode; A heater for heating the element body; A sensor element having a heater control unit that performs a heater control process to energize the heater to generate heat; a pump cell control unit that performs normal control processes after the heater control process is started, the normal control processes including a main pump control process that controls the main pump cell, an auxiliary pump control process that controls the auxiliary pump cell so that the auxiliary pump voltage becomes a target value, and a normal measurement pump control process that controls the measurement pump cell so that the measurement voltage becomes a normal target value to pump out oxygen from the measurement chamber; a specific gas concentration detection unit that detects a specific gas concentration in the measurement target gas based on a measurement pump current flowing through the measurement pump cell by the normal measurement pump control process; Equipped with the pump cell control unit performs a correction process to correct the target value of the auxiliary pump voltage to a higher value in an initial period of the normal control process as compared with a period after the initial period. Gas sensor.

2. the pump cell control unit corrects the target value of the auxiliary pump voltage in the correction process based on a predetermined correction pattern such that an amount of correction tends to become smaller as time passes.

2. The gas sensor according to claim 1.

3. 3. The gas sensor according to claim 1, a reference voltage detection sensor cell for detecting a reference voltage between a measurement gas side electrode provided on the outside of the element body so as to be in contact with the measurement gas and the reference electrode; Equipped with the main pump control process is a process for controlling the main pump cell so that an auxiliary pump current flowing through the auxiliary pump cell by the auxiliary pump control process becomes a target current, and when the main pump current flowing through the main pump cell during the main pump control process is deemed stable, the pump cell control unit corrects the target value of the auxiliary pump voltage in the correction process in such a manner that the amount of correction increases as the deviation between a calculated reference voltage derived based on the main pump current and corresponding to the oxygen concentration in the measurement gas around the sensor element and an actual reference voltage detected by the reference voltage detection sensor cell increases. Gas sensor.

4. the pump cell control unit performs the normal control process during normal operation of the sensor element, and performs a startup measurement pump control process at startup of the sensor element prior to the normal operation, by controlling the measurement pump cell so that the measurement voltage becomes a startup target value higher than the normal target value, thereby pumping out oxygen from the measurement chamber. The gas sensor according to any one of claims 1 to 3.

5. the heater control process is a process of energizing the heater to generate heat so that a sensor element temperature, which is a temperature of the heater or the element body, becomes a predetermined target temperature; the pump cell control unit starts the startup measurement pump control process when the heater control process is started and the sensor element temperature reaches or exceeds a predetermined threshold value that is equal to or lower than the target temperature.

5. The gas sensor according to claim 4.

6. the pump cell control unit performs the auxiliary pump control process and the main pump control process when the sensor element is started up, the main pump control process is a process for controlling the main pump cell so that an auxiliary pump current flowing through the auxiliary pump cell by the auxiliary pump control process becomes a target current, when it is determined that the auxiliary pump current has stabilized near the target current, the pump cell control unit switches from the startup-time measurement pump control process to the normal-time measurement pump control process and starts the normal-time control process.

6. The gas sensor according to claim 4 or 5.

7. the pump cell control unit switches from the startup-time measurement pump control process to the normal-time measurement pump control process and starts the normal-time control process when an open-state measurement voltage, which is the measurement voltage in a state in which no control is performed to cause a current to flow to the inner measurement electrode and the reference electrode, reaches a predetermined threshold value or more.

6. The gas sensor according to claim 4 or 5.

8. The inner measurement electrode includes at least one of Pt and Rh. The gas sensor according to any one of claims 4 to 7.

9. The inner measurement electrode has a volume C of 8×10 -3 mm 3 Above 32 x 10 -3 mm 3 is as follows: The difference ΔV between the startup target value and the normal target value is 120 mV or more and 200 mV or less. The gas sensor according to any one of claims 4 to 8.

10. when it is determined that the oxygen concentration in the measurement chamber is stable based on the operation of the measurement pump cell, the pump cell control unit switches from the startup measurement pump control process to the normal measurement pump control process and starts the normal control process. The gas sensor according to any one of claims 4 to 9.

11. the pump cell control unit determines that the oxygen concentration in the measurement chamber is stable when the measurement pump current flowing through the measurement pump cell is stabilized by the startup measurement pump control process. The gas sensor according to claim 10.

12. the pump cell control unit determines whether the measurement pump current has stabilized based on a first-order differential value of the measurement pump current flowing through the measurement pump cell by the startup measurement pump control process. The gas sensor according to claim 11.

13. the pump cell control unit determines whether the measurement pump current has stabilized based on a second-order differential value of the measurement pump current flowing through the measurement pump cell by the startup measurement pump control process. The gas sensor according to claim 11.

Images