Gas sensor

Abstract

To provide a gas sensor which can measure the concentration of impurities included in gas with high sensitivity and which is excellent in refreshment characteristics.SOLUTION: A gas sensor comprises: a gas sensor element measuring the concentration of impurities included in atmosphere to be detected; a chamber including the gas sensor element in the inside and a gas passage connecting the inside space and the outside space; an electrochemical pump provided on the wall surface of the chamber and capable of selectively transmitting main component gas included in the atmosphere to be detected from the inside of the chamber to the outside or from outside to the inside; a first heater heating the gas sensor element; and a second heater heating the electrochemical pump. The electrochemical pump comprises a first electrode positioned inside the chamber, a second electrode positioned outside the chamber and an ion conductor layer contacting the first and second electrodes, and the electrochemical pump causes the potential of the first electrode to be positive or negative on the basis of the potential of the second electrode, consequently the electrochemical pump selectively transmits the main component gas.SELECTED DRAWING: Figure 4A

Description

[Technical Field]

[0001] The present invention relates to a gas sensor, and more particularly to a gas sensor for measuring the concentration of impurity components contained in a gas. [Background technology]

[0002] Toward the realization of a decarbonized society, technologies that use hydrogen or carbon-neutral biofuels instead of fossil fuels to generate electricity are attracting attention. A well-known example of a power generation device that uses hydrogen is the polymer electrolyte fuel cell (PEFC). PEFCs deteriorate due to toxic impurities contained in the fuel hydrogen, so a technology to detect impurities contained in the fuel hydrogen is required.

[0003] Patent Document 1 discloses a technique for measuring impurities in hydrogen using a sensor that utilizes a polymer electrolyte fuel cell.

[0004] Patent Document 2 discloses an example of a sensor used for measuring gases. Patent Document 2 discloses a sensor that measures hydrogen concentration by utilizing the phenomenon in which the threshold voltage of an element changes depending on the hydrogen concentration in the atmosphere, for example, a work function type sensor that uses a field effect transistor element, a capacitance element, and a diode element.

[0005] Patent Documents 3, 4, and 5 disclose technologies for detecting specific components in the atmosphere. These disclose technologies used in, for example, NOx sensors, which are one type of exhaust gas sensor for automobiles, and disclose a method in which carbon monoxide, hydrocarbons, and other interfering gases in sensing nitrogen oxides (NOx) are inactivated by reacting them with oxygen in the atmosphere to prevent them from interfering, and the remaining interfering oxygen is removed, after which NOx is sensed. An oxygen ion pump using zirconia, an oxygen ion conductor, is used to remove the oxygen. For NOx sensing, Patent Document 3 uses an element similar to an oxygen ion pump, while Patent Documents 4 and 5 use a work function type gas sensor.

[0006] Patent Document 6 discloses technology relating to a device using a conductor of hydrogen ions, i.e., protons, similar to Patent Document 1. However, the device disclosed in Patent Document 6 is an optical device that utilizes the phenomenon in which the refractive index changes depending on the amount of protons contained in a solid electrolyte layer, and is not a gas sensor. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] Japanese Patent Application Laid-Open No. 2011-198729 [Patent Document 2] Japanese Patent Application Publication No. 2018-72146 [Patent Document 3] Japanese Patent Application Publication No. 8-271476 [Patent Document 4] Japanese Patent Application Publication No. 2018-173322 [Patent Document 5] Japanese Patent Application Publication No. 2019-90743 [Patent Document 6] Special Publication No. 2020-524307 Summary of the Invention [Problem to be solved by the invention]

[0008] High sensitivity is required for sensors that measure the concentration of impurity components contained in base gases such as hydrogen. If the sensor is poisoned by impurities, the sensor elements will deteriorate, making it difficult to perform high-sensitivity measurements. Therefore, if the sensor is poisoned by impurities, a refresh operation must be performed as necessary to remove the impurities from the sensor elements.

[0009] With the techniques described in Patent Documents 1 to 6, it is difficult to achieve both highly sensitive impurity measurement and refresh operation.

[0010] The technology described in Patent Document 1 uses a PEFC as a sensor, which deteriorates due to toxic impurities, so depending on the type of impurity, there is a risk that the sensor will deteriorate after measurement and not recover. The technologies described in Patent Documents 2 and 6 do not take into consideration the sensing of impurities in gas.

[0011] The techniques described in Patent Documents 3, 4, and 5 remove oxygen as an interfering gas and measure NOx contained in non-interfering gases such as carbon dioxide, nitrogen, and water vapor. Therefore, they are suitable for separating oxygen, which is not a major component in a gas, and measuring NOx. However, when the interfering gas is the main component, i.e., when it accounts for nearly 100%, the techniques described in Patent Documents 3, 4, and 5 have difficulty measuring the concentration of impurities contained in trace amounts in the interfering gas. In other words, when all the interfering gas is removed, the concentration of the impurity gas contained in trace amounts in the interfering gas becomes 100%, making it difficult to estimate the concentration of the impurity gas contained in the atmosphere before the interfering gas was removed from the measurement result of the impurity gas concentration. Furthermore, when the impurity gas to be measured is toxic, it is difficult to perform a refresh operation.

[0012] The present invention has been made in view of the above-mentioned problems, and aims to provide a gas sensor that can measure the concentration of impurities contained in a gas with high sensitivity and has good refresh characteristics. [Means for solving the problem]

[0013] A gas sensor according to the present invention comprises a gas sensor element for measuring the concentration of impurities contained in a detection target atmosphere, a chamber containing the gas sensor element and having a gas flow path connecting the interior space of the chamber with an exterior space, an electrochemical pump provided on a wall of the chamber for selectively transmitting a main component gas contained in the detection target atmosphere from the interior to the exterior of the chamber or from the exterior to the interior of the chamber, a first heater for heating the gas sensor element to control the temperature of the gas sensor element, and a second heater for heating the electrochemical pump to control the temperature of the electrochemical pump. The electrochemical pump comprises a first electrode located inside the chamber and exposed to the interior space of the chamber, a second electrode located outside the chamber and exposed to the exterior space of the chamber, and an ion conductor layer located between the first and second electrodes and in contact with the first and second electrodes, and the electrochemical pump selectively transmits the main component gas by setting the potential of the first electrode to a positive or negative potential relative to the potential of the second electrode.

[0014] The gas sensor according to the present invention may also have the following configuration. Specifically, the gas sensor according to the present invention includes a gas sensor element for measuring the concentration of impurities contained in a target atmosphere; a chamber having the gas sensor element therein and a gas flow path connecting an internal space with an external space; a gas separation membrane provided inside the chamber and selectively allowing permeation of a main component gas contained in the target atmosphere; a physical pump provided on a wall surface of the chamber and transporting the target atmosphere gas from the inside to the outside of the chamber or from the outside to the inside of the chamber; and a heater for heating the gas sensor element to control the temperature of the gas sensor element. The physical pump releases the main component gas that has permeated the gas separation membrane to the outside of the chamber. The gas separation membrane allows permeation of the main component gas of the target atmosphere gas introduced into the chamber by the physical pump. [Effects of the Invention]

[0015] According to the present invention, it is possible to provide a gas sensor that can measure the concentration of impurities contained in a gas with high sensitivity and has good refresh characteristics. [Brief explanation of the drawings]

[0016] [Figure 1] 1 is a block diagram showing an example of the configuration of a gas sensor according to a first embodiment of the present invention. [Figure 2] FIG. 2 is a plan view of a gas sensor chip included in the gas sensor according to the first embodiment. [Figure 3A] FIG. 1 is a diagram showing a cross section of a sensor FET (SFET). [Figure 3B] FIG. 1 shows a cross section of a reference FET (RFET). [Figure 4A] 1A and 1B are diagrams showing an example of the configuration of a gas sensing unit of the gas sensor according to Example 1, illustrating an operation of discharging hydrogen in the internal space of a chamber into an atmosphere to be detected. [Figure 4B] 1A and 1B are diagrams showing an example of the configuration of a gas sensing unit of the gas sensor according to Example 1, illustrating an operation of introducing hydrogen in a detection target atmosphere into the internal space of a chamber. [Figure 5A] FIG. 10 is a diagram showing an example of a method of supplying power to a sensor FET, a reference FET, a heater A, and a diode. [Figure 5B] 10 is a diagram showing an example of a method of supplying power to a hydrogen pump and a heater B. FIG. [Figure 6A] FIG. 10 is a diagram showing the gate voltage-drain current characteristics of a sensor FET. [Figure 6B] FIG. 10 is a graph showing the gate voltage-drain current characteristics of a reference FET. [Figure 7] FIG. 10 is a diagram showing the gate voltage-drain current characteristics of a sensor FET in an environment where poisonous impurities are present. [Figure 8A] FIG. 10 is a graph showing the dependence of ΔVg(X, 0%, 100%−X) on the hydrogen concentration X. [Figure 8B] FIG. 10 is a diagram showing the dependence of ΔVg(100%−Y,Y,0%) on the concentration Y of a poisoning impurity. [Figure 9A]FIG. 10 is a diagram showing the difference in how impurities are concentrated in the space inside the chamber due to the difference in gas temperature. [Figure 9B] FIG. 10 is a diagram showing a gas flow path equipped with a heater C. [Figure 10] FIG. 10 is a diagram showing the difference in how impurities are concentrated in the space inside the chamber depending on the current flowing through the hydrogen pump. [Figure 11A] 10A and 10B are diagrams showing differences in how impurities are concentrated in the space inside the chamber depending on differences in the length of the gas flow path. [Figure 11B] 10A and 10B are diagrams showing differences in how impurities are concentrated in the space inside the chamber depending on differences in the cross-sectional area of ​​the gas flow path. [Figure 12A] 10 is a diagram showing differences in how impurities are concentrated in the space inside the chamber depending on differences in the volume of the space inside the chamber. FIG. [Figure 12B] 10 is a diagram showing differences in how impurities are concentrated in the space inside a chamber due to differences in the impurity concentration of the atmosphere to be detected. FIG. [Figure 13] FIG. 10 is a diagram showing differences in how impurities are concentrated in the space inside the chamber due to differences in the impurity concentration in the space inside the chamber at the initial stage (time 0). [Figure 14] 10A and 10B are diagrams for explaining the operation of diluting impurities in the space inside the chamber when a refresh operation of the sensor FET is performed. [Figure 15A] FIG. 10 is a diagram showing the gate voltage-drain current characteristics of a sensor FET using a P-type FET. [Figure 15B] FIG. 10 is a graph showing the gate voltage-drain current characteristics of a reference FET using a P-type FET. [Figure 16A] 10A and 10B are diagrams showing an example of the configuration of a gas sensing unit including a chamber made of laminated ceramic, and illustrating the operation of discharging hydrogen in the internal space of the chamber into the atmosphere to be detected. [Figure 16B] 10A and 10B are diagrams showing an example of the configuration of a gas sensing unit including a chamber made of laminated ceramic, and illustrating the operation of introducing hydrogen in the atmosphere to be detected into the internal space of the chamber. [Figure 17A]10A and 10B are diagrams showing an example of the configuration of a gas sensing unit including a chamber made of MEMS, and show an operation of discharging hydrogen in the internal space of the chamber into an atmosphere to be detected. [Figure 17B] 10A and 10B are diagrams showing an example of the configuration of a gas sensing unit including a chamber configured by MEMS, and show the operation of introducing hydrogen, which is an atmosphere to be detected, into the space inside the chamber. [Figure 18] 2 is a diagram showing an example of installation of a gas sensing unit included in the gas sensor according to the first embodiment. FIG. [Figure 19A] FIG. 1 is a cross-sectional view of a sensor capacitor (SCAP). [Figure 19B] FIG. 1 shows a cross section of a reference capacitor (RCAP). [Figure 20A] FIG. 10 is a diagram showing an example of a method of supplying power to a sensor capacitor, a reference capacitor, a heater A, and a diode. [Figure 20B] 10 is a diagram showing an example of a method of supplying power to a hydrogen pump and a heater B. FIG. [Figure 21A] FIG. 10 is a diagram showing the capacitance-gate voltage characteristics of a sensor capacitor. [Figure 21B] FIG. 10 is a diagram showing the capacitance-gate voltage characteristics of a reference capacitor. [Figure 22A] FIG. 2 is a diagram showing a cross section of a sensor diode (SDIODE). [Figure 22B] FIG. 1 is a cross-sectional view of a reference diode (RDIODE). [Figure 23A] FIG. 10 is a diagram showing an example of a method of supplying power to a sensor diode, a reference diode, a heater A, and a diode. [Figure 23B] 10 is a diagram showing an example of a method of supplying power to a hydrogen pump and a heater B. [Figure 24A] FIG. 10 is a diagram showing the current-gate voltage characteristics of a sensor diode. [Figure 24B] FIG. 10 is a diagram showing the current-gate voltage characteristics of a reference diode. [Figure 25A] FIG. 10 is a diagram showing an example of the response of a sensor FET, a sensor capacitor, and a sensor diode to the concentration of poisoning impurities. [Figure 25B] FIG. 10 is a diagram showing the operation of a hydrogen pump according to the concentration of poisonous impurities in an atmosphere to be detected. [Figure 26] FIG. 10 is a block diagram showing an example of the configuration of a gas sensor according to a second embodiment of the present invention. [Figure 27] 10 is a flowchart showing a process for measuring the concentration of impurities contained in a detection target atmosphere using a plurality of gas sensing units in the second embodiment. [Figure 28A] 10A and 10B are diagrams illustrating an example of the operation of the gas sensing unit when the operation of the hydrogen pump is required to measure the concentration of poisoning impurities in hydrogen. [Figure 28B] 10A and 10B are diagrams illustrating an example of the operation of the gas sensing unit when the operation of the hydrogen pump is not required to measure the concentration of poisoning impurities in hydrogen. [Figure 29A] FIG. 10 is a diagram showing an example of the configuration of a gas sensing unit of a gas sensor equipped with an oxygen pump in Example 3 of the present invention, illustrating the operation of discharging oxygen in the internal space of a chamber into an atmosphere to be detected. [Figure 29B] FIG. 10 is a diagram showing an example of the configuration of a gas sensing unit of a gas sensor equipped with an oxygen pump in Example 3, illustrating the operation of introducing oxygen in a detection target atmosphere into the internal space of a chamber. [Figure 30A] FIG. 10 is a diagram showing an example of the configuration of a gas sensing section of a gas sensor equipped with a gas separation membrane and a physical pump, illustrating the operation of discharging the main component of the gas in the internal space of a chamber into the atmosphere to be detected. [Figure 30B] FIG. 10 is a diagram showing an example of the configuration of a gas sensing section of a gas sensor equipped with a gas separation membrane and a physical pump, illustrating the operation of introducing the main component of the gas in the atmosphere to be detected into the internal space of the chamber. DETAILED DESCRIPTION OF THE INVENTION

[0017] In the gas sensor according to the present invention, when detecting impurities contained in the gas atmosphere to be measured, highly sensitive sensing can be achieved by concentrating the impurities to be detected, and the gas sensor can be refreshed by diluting the impurities to be detected and heating the gas sensor. In the present invention, the main component gas (e.g., hydrogen gas) contained in the atmosphere to be detected can be discharged to the outside of the chamber to increase the concentration of impurities in the chamber, and the main component gas can also be introduced into the chamber to decrease the concentration of impurities in the chamber.

[0018] The gas sensor according to the present invention increases the impurity concentration when measuring the impurity concentration and decreases the impurity concentration when refreshing the gas sensor, thereby enabling high-sensitivity measurement of impurities and effective refreshing at low temperatures in a short time. As a result, the present invention can provide a small, inexpensive, high-performance, and highly reliable gas sensor.

[0019] A gas sensor according to an embodiment of the present invention will be described in detail below with reference to the drawings. The gas sensor according to the embodiment of the present invention measures the concentration of impurities contained in a detection target atmosphere. The detection target atmosphere is, for example, a gas containing hydrogen, oxygen, or carbon dioxide as a main component and containing impurities.

[0020] In the drawings referred to in this specification, identical or corresponding components are designated by the same reference numerals, and repeated explanations of these components may be omitted. Furthermore, when there are multiple identical or corresponding components and they are to be described separately, a symbol is added to the end of the reference numeral, but when it is not necessary to describe them separately, they may be described without a symbol added to the end of the reference numeral.

[0021] In the drawings referred to in this specification, hatching may be omitted even in cross-sectional views to make the drawings easier to read, and hatching may be added even in plan views to make the drawings easier to read.

[0022] In the drawings referred to in this specification, the relative sizes of elements do not necessarily represent their actual sizes, and some elements or parts included in elements may be shown relatively large to make the drawings easier to understand. Also, even when a cross-sectional view corresponds to a plan view, some elements or parts may be shown relatively large to make the drawings easier to understand. [Example]

[0023] A gas sensor according to a first embodiment of the present invention will be described. In this embodiment, the detection target atmosphere is a gas containing hydrogen as a main component gas and impurities. The gas sensor according to this embodiment measures the concentrations of toxic impurities such as carbon monoxide and hydrogen sulfide contained in the detection target atmosphere using a gas sensor element.

[0024] 1 is a block diagram showing an example of the configuration of a gas sensor according to this embodiment. The gas sensor according to this embodiment includes a gas sensing unit 1001 and a circuit unit 1008, and the gas sensor element is a field effect transistor (FET), for example, a catalytic metal gate FET type gas sensor.

[0025] The gas sensing unit 1001 includes a sensor FET (SFET) 21 and a reference FET (RFET) 22 which are gas sensor elements, a heater A 24, a thermometer unit 23, a chamber 25, a hydrogen pump 26, and a heater B 27.

[0026] The sensor FET 21 and the reference FET 22 are arranged in the space inside the chamber 25. The sensor FET 21 and the reference FET 22 can be formed, for example, by integrating them on the same silicon chip. In this embodiment, the sensor FET 21 and the reference FET 22 are formed on the gas sensor chip 20. The threshold voltage of the sensor FET 21 changes depending on the concentration of toxic impurities such as carbon monoxide and hydrogen sulfide contained in the atmosphere to be detected. The threshold voltage of the reference FET (RFET) 22 does not depend on the composition or concentration of hydrogen or components other than hydrogen contained in the atmosphere to be detected.

[0027] The heater A24 is provided on the gas sensor chip 20 and is configured to heat the gas sensing unit 1001, particularly the sensor FET 21 and the reference FET 22, by applying a voltage to the heater A24 and causing a current to flow therethrough, thereby controlling the temperatures of the sensor FET 21 and the reference FET 22. The heater A24 is provided on the gas sensor chip 20.

[0028] The thermometer unit 23 measures the temperature of the gas sensor chip 20, which includes the sensor FET 21 and the reference FET 22. The thermometer unit 23 is provided on the gas sensor chip 20. The thermometer unit 23 may include, for example, a diode, and can measure the temperatures of the sensor FET 21 and the reference FET 22, i.e., the temperature of the gas sensor chip 20, by using the diode to detect a current that flows when a certain voltage is applied.

[0029] Chamber 25 is provided therein with sensor FET 21 and reference FET 22. Chamber 25 is provided with a gas flow path that connects the internal space of chamber 25 with the external space of chamber 25, which is the detection target atmosphere. Gas can flow in and out between the internal space of chamber 25 and the detection target atmosphere via this gas flow path.

[0030] The hydrogen pump 26 is provided on a part of the wall surface of the chamber 25. The hydrogen pump 26 is a type of electrochemical pump, and is a gas permeable portion that selectively allows hydrogen to permeate. The hydrogen pump 26 allows hydrogen to permeate from the inside to the outside of the chamber 25 or from the outside to the inside of the chamber 25.

[0031] The hydrogen pump 26 includes two electrodes and a proton conductor layer located between the electrodes. One electrode is located inside the chamber 25, and the other electrode is located outside the chamber 25. The proton conductor layer is a hydrogen ion conductor layer and is sandwiched between the two electrodes.

[0032] The heater B27 is provided adjacent to the hydrogen pump 26 and controls the temperature of the hydrogen pump 26 by applying a voltage to the heater B27 and passing a current through the heater B27.

[0033] The circuit section 1008 includes a gas concentration estimation section 1002 , a current detection section 1004 , a temperature detection section 1005 , a hydrogen pump current detection section 1006 , a power supply 1007 , a control section 1003 , an I / O section 1000 , and a parameter recording section 1009 .

[0034] The gas concentration estimation unit 1002 estimates the gas concentration from the signal of the gas sensing unit 1001 .

[0035] The current detection unit 1004 measures the current flowing through the sensor FET 21 and the reference FET 22 .

[0036] The temperature detection unit 1005 estimates the temperature of the gas sensor chip 20 including the sensor FET 21 and the reference FET 22. For example, the temperature detection unit 1005 determines the temperature of the gas sensor chip 20 measured by the thermometer unit 23 (the temperature of the sensor FET 21 and the reference FET 22) as the temperature of the gas sensor chip 20.

[0037] The hydrogen pump current detection unit 1006 measures the current flowing through the hydrogen pump 26 .

[0038] The power supply 1007 applies a voltage to the gas sensing unit 1001 .

[0039] The control unit 1003 controls the power supply 1007 .

[0040] The I / O unit 1000 inputs and outputs information and power between the control unit 1003 and the outside of the gas sensor.

[0041] The parameter recording unit 1009 can record, for example, the conditions of voltages applied to the sensor FET 21, the reference FET 22, the heater A 24, the hydrogen pump 26, and the heater B 27.

[0042] In the gas sensor of this embodiment, the hydrogen pump 26 is heated to an appropriate temperature by applying a voltage to the heater B27 to pass a current through it, and then a potential difference is applied between the electrodes of the hydrogen pump 26, allowing the hydrogen pump 26 to selectively transport hydrogen in the atmosphere inside and outside the chamber 25.

[0043] The power supply 1007 can also apply a voltage to the first electrode 26a and the second electrode 26c. In the hydrogen pump 26, by setting the potential of the electrode located inside the chamber 25 to a positive potential with respect to the potential of the electrode located outside the chamber 25 as the reference, hydrogen inside the chamber 25 can be selectively transported to the outside of the chamber 25 via the proton conductor layer. Conversely, by setting the potential of the electrode located inside the chamber 25 to a negative potential with respect to the potential of the electrode located outside the chamber 25 as the reference, hydrogen outside the chamber 25 can be selectively transported to the inside of the chamber 25 via the proton conductor layer.

[0044] When hydrogen pump 26 selectively transports hydrogen from inside chamber 25 to the outside of chamber 25, the pressure inside chamber 25 becomes lower than the pressure outside chamber 25, and so gas of the detection target atmosphere outside chamber 25 (hydrogen containing impurities) flows into chamber 25 via the gas flow path. Hydrogen containing impurities, which is gas of the detection target atmosphere, is taken into chamber 25 via the gas flow path, and only hydrogen is discharged outside chamber 25 by hydrogen pump 26, so the impurities contained in the hydrogen atmosphere inside chamber 25 are concentrated to a high concentration.

[0045] Conversely, when hydrogen pump 26 selectively transports hydrogen from outside chamber 25 into chamber 25, the pressure inside chamber 25 becomes higher than the pressure outside chamber 25, and the gas inside chamber 25 (hydrogen containing impurities) flows out of chamber 25 via the gas flow path. Since only hydrogen is taken into chamber 25 by hydrogen pump 26 and hydrogen containing impurities is released to the outside of chamber 25 via the gas flow path, the impurities contained in the hydrogen atmosphere inside chamber 25 are diluted and reduced in concentration.

[0046] In this way, the gas sensor according to this embodiment can not only increase the concentration of impurities in the hydrogen gas inside the chamber 25 by discharging hydrogen gas to the outside of the chamber 25, but also introduce hydrogen gas into the chamber 25 to decrease the concentration of impurities in the hydrogen gas inside the chamber 25.

[0047] The relationship between the concentration of the impurity that has been condensed and made highly concentrated and the concentration of the impurity in the atmosphere to be detected varies depending on the shape of the gas flow path, the shape of chamber 25, the gas temperature, the current flowing through hydrogen pump 26, etc., but can be established as a one-to-one correspondence by conducting experiments and simulations. The relationship between the concentration of the impurity that has been diluted and made less concentrated and the concentration of the impurity in the atmosphere to be detected can also be established as a one-to-one correspondence by conducting experiments and simulations in the same way.

[0048] The sensor FET can detect the concentration of poisoning impurities in an atmosphere with hydrogen as the base gas. Depending on the application, the sensor FET needs to detect very low concentrations of poisoning impurities, such as concentrations of poisoning impurities less than 1 ppm in a hydrogen atmosphere. In such a low concentration range, the signal from the sensor FET increases as the concentration of the poisoning impurity increases. In other words, the higher the concentration of the poisoning impurity, the easier it is for the sensor FET to detect.

[0049] The concentration of the poisonous impurities that have become concentrated and highly concentrated inside the chamber 25 is detected by the sensor FET, and by matching it one-to-one with the concentration of the poisonous impurities in the atmosphere to be detected, it is possible to detect the concentration of the poisonous impurities in the atmosphere to be detected even if the concentration is too low to be detected by the sensor FET as it is.

[0050] As mentioned above, the sensor FET can detect the concentration of poisonous impurities contained in a hydrogen atmosphere, but over the long term it may become poisoned by the poisonous impurities and become unable to detect them. For this reason, it is necessary to refresh the sensor FET to remove the impurities from the poisoned sensor FET.

[0051] Poisoning of the sensor FET can be refreshed by heating the sensor FET to a high temperature. The lower the concentration of poisoning impurities contained in the atmosphere, the lower the heating temperature required for refreshing can be, and the shorter the time required for refreshing. Therefore, by diluting the poisoning impurities contained in the hydrogen inside chamber 25 and heating the sensor FET, the sensor FET can be easily refreshed.

[0052] The heater A24 heats the sensor FET 21 and the reference FET 22 by applying a voltage and passing a current therethrough, thereby controlling the temperatures thereof.

[0053] 2 is a plan view of the gas sensor chip 20 included in the gas sensor according to this embodiment. The gas sensor chip 20 constitutes the gas sensing section 1001 shown in FIG.

[0054] The gas sensor chip 20 is formed with a sensor FET 21 and a reference FET 22, which are catalytic metal gate FET type gas sensor elements, a heater A 24, and a diode 29. The diode 29 constitutes a thermometer section 23.

[0055] The sensor FET 21 has a gas detection material, which is made up of a metal oxide layer and a catalytic metal layer, mounted on the gate.

[0056] The heater A24 is formed of wiring 28 made of metal such as aluminum, tungsten, or platinum, and can raise the temperature of the gas sensing unit 1001 above the ambient temperature by Joule heat generated by passing a current between both ends of the heater A24 when power is supplied from a power source 1007. For example, the heater A24 can raise the temperature of the gas sensing unit 1001 above the ambient temperature by 100° C. or more.

[0057] The thermometer unit 23 (FIG. 1) can measure the current flowing through the diode 29 or the heater A24 to measure the temperature of the gas sensor chip 20. For example, as described above, the temperature of the gas sensor chip 20 can be estimated from the value of the current flowing when a constant voltage is applied to the diode 29. Alternatively, the temperature of the gas sensor chip 20 can be estimated by measuring the resistance between both ends of the heater A24.

[0058] The parameter recording unit 1009 (FIG. 1) can further record the voltage conditions to be applied to the diode 29.

[0059] In Figure 2, Nsub represents the N-type substrate, PWell represents the P-type well, PDif represents the P-type diffusion layer, NDif represents the N-type diffusion layer, Gate represents the gate, Drain represents the drain, Source represents the source, and Heater represents the heater. Pad represents the terminal.

[0060] 3A is a diagram showing a cross section of a sensor FET (SFET) 21. The sensor FET 21 includes a semiconductor substrate 1, a well 2, a source diffusion layer 3, a drain diffusion layer 4, a gate insulating film 5, a metal oxide layer 6 serving as a gas detection material (gate A), a catalytic metal gate layer 7, and an interlayer insulating film 8. The surface of the catalytic metal gate layer 7 is exposed to the atmosphere to be detected (the atmosphere inside the chamber 25).

[0061] The semiconductor substrate 1 can be made of, for example, silicon or silicon carbide (SiC). When a silicon substrate is used, the operating temperature of the sensor FET and reference FET is limited to a maximum of about 250°C. On the other hand, when a silicon carbide substrate is used, the cost of the substrate increases compared to a silicon substrate, but the operating temperature of the sensor FET and reference FET can be increased to 500°C or higher. The substrate material can be selected depending on the application of the sensor.

[0062] The metal oxide layer 6 can be made of, for example, titanium oxide, yttria-stabilized zirconia (YSZ), etc. The catalytic metal gate layer 7 can be made of, for example, noble metals such as platinum, palladium, and iridium, as well as nickel, etc.

[0063] 3B is a diagram showing a cross section of a reference FET (RFET) 22. The reference FET 22 includes a semiconductor substrate 1, a well 12, a source diffusion layer 13, a drain diffusion layer 14, a gate insulating film 15, a metal oxide layer 16 serving as a gas sensing material (gate B), a catalytic metal gate layer 17, and an interlayer insulating film 18. The surface of the catalytic metal gate layer 17 is covered with the interlayer insulating film 18, and the catalytic metal gate layer 17 is isolated from the atmosphere to be sensed.

[0064] The metal oxide layer 16 and the catalytic metal gate layer 17 can be made of the same material as the sensor FET 21 .

[0065] The wells 2 and 12, the source diffusion layers 3 and 13, the drain diffusion layers 4 and 14, and the catalytic metal gate layers 7 and 17, as well as the diode 29 and heater A24 shown in FIG. 2, are connected to a wiring layer made of metal such as aluminum, tungsten, or platinum, so that power can be supplied from the power supply 1007 shown in FIG. 1.

[0066] In this embodiment, either an N-type FET or a P-type FET can be used for both the sensor FET 21 and the reference FET 22. In the following, an example in which an N-type FET is used for the sensor FET 21 and the reference FET 22 will be mainly described.

[0067] 4A and 4B are diagrams showing an example of the configuration of the gas sensing portion 1001 of the gas sensor according to this embodiment.

[0068] A gas sensor chip 20 including a sensor FET 21, a reference FET 22, and a diode 29 serving as a thermometer section 23 is installed in an internal space 40 of the chamber 25. The chamber 25 is equipped with a hydrogen pump 26 and a gas flow path 30 that connects the internal space 40 of the chamber 25 with the external space of the chamber 25, which is the atmosphere to be detected.

[0069] The gas sensor chip 20 is preferably located between the gas flow path 30 and the hydrogen pump 26 in the direction of gas flow in the gas flow path 30 in the space inside the chamber 25. This is because the position between the gas flow path 30 and the hydrogen pump 26 is the position where the gas mainly flows in the space 40 inside the chamber 25, where the gas tends to flow smoothly and where the concentration of impurities tends to be constant. When the gas sensor chip 20 is located between the gas flow path 30 and the hydrogen pump 26, the concentration of impurities contained in the gas can be measured with higher sensitivity.

[0070] The gas sensor is assumed to be used in an environment where the number of moles of gas introduced into the space 40 inside the chamber 25 is negligibly small compared to the number of moles of gas present in the atmosphere to be detected.

[0071] The chamber 25 can be made of, for example, metal or metal oxide. The chamber 25 shown in FIGS. 4A and 4B is made of metal. After the gas sensor chip 20 is installed in the internal space 40 of the chamber 25, the walls of the chamber 25 are sealed by welding or using a gasket, etc., except for the areas where the gas flow path 30 and the hydrogen pump 26 are installed. Power is supplied to the gas sensor chip 20 using wiring, and by using a known sealing technique, it is possible to achieve both gas sealing and wiring access.

[0072] The cross section of the gas flow path 30 may be, for example, a circle with a diameter of 1 millimeter or less, the length of the gas flow path 30 may be, for example, 5 millimeters or less, and the volume of the gas flow path 30 may be, for example, 1 cubic centimeter or less.

[0073] The space 40 inside the chamber 25 is connected to the detection target atmosphere outside the chamber 25 via a hydrogen pump 26, separate from the gas flow path 30. The hydrogen pump 26 includes a first electrode 26a, a proton conductor layer 26b, and a second electrode 26c. The first electrode 26a is located inside the chamber 25 and exposed to the space 40 inside the chamber 25. The proton conductor layer 26b is disposed in contact with one surface of the first electrode 26a. The second electrode 26c is disposed in contact with one surface of the proton conductor layer 26b, is located outside the chamber 25, and is exposed to the space outside the chamber 25. The proton conductor layer 26b is located between the first electrode 26a and the second electrode 26c and is in contact with the first electrode 26a and the second electrode 26c. A voltage (Vp1, Vp2) can be applied from a power supply 1007 to the first electrode 26a and the second electrode 26c.

[0074] The proton conductor layer 26b can be made of materials such as titanium dioxide, yttrium-doped barium zirconate (BYZ), yttria-stabilized zirconia (YSZ), and ceria (CeO). YSZ and ceria are known as oxygen ion conductors, but when formed into nano-sized crystallites, they are also known to conduct protons. Titanium dioxide and BYZ are known as proton conductors.

[0075] The first electrode 26a and the second electrode 26c may be made of a thin film of a porous precious metal such as platinum or palladium.

[0076] It is known that the proton conductor layer 26b made of the above-mentioned metal oxide can obtain sufficient proton conductivity by raising the temperature to about 300° C. For this reason, a heater B27 is installed in addition to the heater A24 to control the temperature of the hydrogen pump 26. By applying a voltage across the heater B27, the temperature of the proton conductor layer 26b can be raised to 300° C. or higher.

[0077] The hydrogen pump 26 is heated to approximately 300°C, and a positive potential is applied to the first electrode 26a relative to the second electrode 26c in a hydrogen atmosphere. This causes hydrogen molecules to transfer two electrons to the first electrode 26a at the interface between the first electrode 26a and the proton conductor layer 26b and become two protons. As a result, a current (ionic current) caused by the protons flows in the proton conductor layer 26b. The protons that reach the second electrode 26c receive electrons from the second electrode 26c, and pair up to become hydrogen molecules again. These hydrogen molecules are released into the atmosphere from the second electrode 26c.

[0078] When the hydrogen pump 26 is heated to approximately 300°C and a positive potential is applied to the second electrode 26c relative to the first electrode 26a in a hydrogen atmosphere, hydrogen molecules are released from the first electrode 26a into the atmosphere using the same principle as above. That is, the hydrogen molecules are converted into protons at the interface between the second electrode 26c and the proton conductor layer 26b, a current due to the protons flows in the proton conductor layer 26b, and the protons that reach the first electrode 26a become hydrogen molecules again. These hydrogen molecules are released into the atmosphere from the first electrode 26a.

[0079] The hydrogen pump 26 can selectively allow only hydrogen to permeate by utilizing this phenomenon. The current flowing through the hydrogen pump 26 can be measured by the hydrogen pump current detector 1006. The current flowing through the hydrogen pump 26 is proportional to the number of protons flowing per unit time, i.e., proportional to the amount of hydrogen permeating through the hydrogen pump 26. Current flowing through the hydrogen pump = number of moles of protons passing through per second x Faraday's constant (1) This can be expressed as equation (1).

[0080] Considering that the number of moles of hydrogen molecules is half the number of moles of protons, it is possible to determine the number of moles of hydrogen molecules permeating per second through the hydrogen pump 26. For example, when a current of 1 ampere is applied to the hydrogen pump 26, 7 cc of hydrogen will permeate per minute, converted to standard conditions (0°C, 1 atom).

[0081] The voltage applied between the first electrode 26a and the second electrode 26c can be changed in polarity.

[0082] 4A, a positive voltage Vp1 is applied to first electrode 26a relative to second electrode 26c, and hydrogen in space 40 inside chamber 25 is selectively discharged into the detection target atmosphere outside chamber 25. As a result, the gas pressure in space 40 inside chamber 25 decreases, and hydrogen and impurities contained in the detection target atmosphere flow from the detection target atmosphere into space 40 inside chamber 25 through gas flow path 30.

[0083] Because the hydrogen pump 26 selectively discharges only hydrogen, the impurity concentration in the space 40 inside the chamber 25 increases. When the impurity concentration in the space 40 inside the chamber 25 increases to a certain level compared to the target atmosphere outside the chamber 25, the impurities are released from the space 40 inside the chamber 25 through the gas flow path 30 into the target atmosphere by diffusion, against the flow of gas. When the number of impurity molecules flowing into the space 40 inside the chamber 25 by the gas flow becomes equal to the number of impurity molecules being discharged from the space 40 inside the chamber 25 by diffusion, the impurity concentration in the space 40 inside the chamber 25 becomes saturated.

[0084] When the impurity concentration in the space 40 inside the chamber 25 is sufficiently higher than the impurity concentration in the detection target atmosphere outside the chamber 25, the impurity can be easily detected by the gas sensor chip 20. When measuring the impurity concentration, it is preferable to control the heater A24 to set the temperature of the gas sensor chip 20 to, for example, 80°C. The heater A24 can be one that is mounted on the gas sensor chip 20 as shown in FIG. 2, or one that is installed separately from the gas sensor chip 20.

[0085] 4B, a positive voltage Vp2 is applied to the second electrode 26c relative to the first electrode 26a, and hydrogen in the atmosphere to be detected is selectively introduced into the internal space 40 of the chamber 25. As a result, the gas pressure in the internal space 40 of the chamber 25 increases, causing the hydrogen and impurities contained in the internal space 40 of the chamber 25 to flow out from the internal space 40 of the chamber 25 through the gas flow path 30 into the atmosphere to be detected.

[0086] Because the hydrogen pump 26 selectively introduces only hydrogen, the impurity concentration in the space 40 inside the chamber 25 decreases. When the impurity concentration in the space 40 inside the chamber 25 decreases to a certain level compared to the detection target atmosphere outside the chamber 25, impurities are taken into the space 40 inside the chamber 25 from the detection target atmosphere through the gas flow path 30, against the flow of gas. When the number of impurity molecules discharged into the space 40 inside the chamber 25 by the gas flow becomes equal to the number of impurity molecules taken into the space 40 inside the chamber 25 by diffusion, the impurity concentration in the space 40 inside the chamber 25 becomes saturated.

[0087] The refresh operation of the gas sensor chip 20 can be easily performed when the impurity concentration in the space 40 inside the chamber 25 is sufficiently reduced compared to the impurity concentration in the atmosphere to be detected. During the refresh operation of the gas sensor chip 20, it is preferable to control the heater A24 to set the temperature of the gas sensor chip 20 to, for example, 200°C. The heater A24 may be mounted on the gas sensor chip 20 as shown in FIG. 2, or may be installed separately from the gas sensor chip 20. By raising the temperature of the gas sensor chip 20 to a sufficiently high temperature in an atmosphere with a reduced impurity concentration, the impurities in the hydrogen adsorbed to the gate material layer of the sensor FET can be removed by the refresh operation.

[0088] An example of an operation for detecting the concentration of impurities and an example of a refresh operation will be described with reference to FIGS. 5A and 5B.

[0089] 5A is a diagram showing an example of a method of supplying power to the sensor FET 21, the reference FET 22, the heater A 24, and the diode 29. FIG. 5B is a diagram showing an example of a method of supplying power to the hydrogen pump 26 and the heater B 27.

[0090] In detecting the concentration of impurities, a voltage Vh1_L is applied to the heater A24 to cause a current to flow, and the generated Joule heat raises the temperature of the gas sensor chip 20 (particularly the sensor FET 21 and reference FET 22) to a predetermined temperature (e.g., 80°C). The temperature of the gas sensor chip 20 is estimated by the temperature detection unit 1005, as the current flowing when a voltage is applied to the diode 29 is measured by the thermometer unit 23. In addition, a voltage Vh2 is applied to the heater B27 to cause a current to flow, and the generated Joule heat raises the temperature of the hydrogen pump 26 to a predetermined temperature (e.g., 300°C).

[0091] When detecting the impurity concentration, a positive voltage Vp1 is applied to the first electrode 26a of the hydrogen pump 26 relative to the second electrode 26c, increasing the impurity concentration in the space 40 inside the chamber 25. The wells 2, 12 and source diffusion layers 3, 13 of the sensor FET 21 and reference FET 22 are set to 0 V, and a voltage VD is applied to the drain diffusion layers 4, 14. A variable voltage VGR is applied to the gate of the reference FET 22, and a variable voltage VGS is applied to the gate of the sensor FET 221. The currents flowing through the drain terminals of the sensor FET 21 and the reference FET 22 are measured by the current detection unit 1004 (Figure 1), and the voltages VGS and VGR are controlled by the control unit 1003 so that both currents match a constant current Ic.

[0092] In the refresh operation, a voltage Vh1_H is applied to the heater A24 to cause a current to flow, and the gas sensor chip 20 is heated to a predetermined temperature (for example, 200°C) by the generated Joule heat. The temperature of the gas sensor chip 20 is estimated by the temperature detection unit 1005, as the current flowing when a voltage is applied to the diode 29 is measured by the thermometer unit 23. In addition, a voltage Vh2 is applied to the heater B27 to cause a current to flow, and the hydrogen pump 26 is heated to a predetermined temperature (for example, 300°C) by the generated Joule heat.

[0093] During the refresh operation, a positive voltage Vp2 is applied to the second electrode 26c of the hydrogen pump 26 relative to the first electrode 26a, thereby reducing the concentration of impurities in the space 40 inside the chamber 25. When the impurity concentration in the space 40 inside the chamber 25 is left in a state where it has been sufficiently reduced, the impurities adsorbed on the sensor FET 21 are desorbed, and the refresh operation is completed.

[0094] The hydrogen pump 26 is heated by the heater B27 to a predetermined temperature (e.g., 300°C), and the adsorbed impurities are desorbed and refreshed, similar to the sensor FET 21. That is, even if the hydrogen pump 26 is poisoned by impurities, a refresh operation for removing the impurities can be performed.

[0095] 6A is a diagram showing the gate voltage-drain current characteristics of the sensor FET 21. FIG. 6B is a diagram showing the gate voltage-drain current characteristics of the reference FET 22.

[0096] In the sensor FET 21, the catalytic metal gate layer 7, which is a gas detection material, is exposed to the detection target atmosphere. Therefore, the work function of the catalytic metal gate layer 7 changes depending on the gas to which the catalytic metal gate layer 7 is exposed.

[0097] Let X be the concentration of hydrogen in the space 40 inside the chamber 25, Y be the concentration of poisonous impurities (carbon monoxide, hydrogen sulfide, etc.), and Z be the concentration of inert impurities (nitrogen, etc.). X+Y+Z=100%.

[0098] First, consider the case where the concentration Y of the poisoning impurity is 0%. When the hydrogen concentration X is a finite value X1, the gate voltage-drain current characteristics shift in the negative voltage direction compared to when the hydrogen concentration X is 0%. As a result, the gate voltage when the threshold current Ic flows changes from VGS(X=0%, Y=0%, Z=100%) to VGS(X=X1, Y=0%, Z=100%-X1).

[0099] On the other hand, in the reference FET 22, the gate voltage-drain current characteristics do not change even if the gas concentration changes because the catalytic metal gate layer 17 is covered with the interlayer insulating film 18. That is, in the reference FET 22, the current that flows when a voltage VGR is applied to the gate remains constant at Ic.

[0100] The difference between VGR and VGS is VGRS. VGRS=VGR-VGS (2) This is expressed by equation (2). VGRS is a function of X, Y, and Z when the temperature and pressure are constant.

[0101] ΔVg is defined as follows: ΔVg represents the shift in threshold voltage of the gas sensor due to hydrogen. ΔVg(X1,0%,100%-X1) =VGRS(0,0,100%)-VGRS(X1,0,100%-X1) (3) As shown in Equation 2, by using VGRS, which is the difference in gate voltage between the sensor FET 21 and the reference FET 22, the current flowing through the drain terminal mentioned above can be set to an appropriate value to suppress the effects of noise caused by fluctuations in VGR and VGS due to temperature fluctuations, etc. In an environment without noise, there is no problem in defining ΔVg as shown below instead of Equation (3). Because VGR does not depend on the gas components, the definitions of ΔVg using Equation (3) and Equation (4) will match when there is no noise influence. ΔVg(X,0%,100%-X) =VGS(0,0,100%)-VGS(X,0,100%-X) (4) FIG. 7 is a diagram showing the gate voltage-drain current characteristics of the sensor FET 21 in an environment where poisoning impurities are present.

[0102] In addition, since the catalytic metal gate layer 17 of the reference FET 22 is covered with the interlayer insulating film 18, as shown in FIG. 6B, the gate voltage-drain current characteristics do not change even when the gas concentration changes, and the current that flows when a voltage VGR is applied to the gate remains constant at Ic.

[0103] In the gate voltage-drain current characteristics of the sensor FET 21, when poisoning impurities are present, i.e. when the poisoning impurity concentration Y and the inert gas concentration Z are finite, VGS shifts parallel to the positive voltage direction compared to when the hydrogen concentration X=100% (Y=Z=0%). This is self-evident. It should be noted that VGS is very sensitive to the poisoning impurity concentration Y, and can detect even a change in ppm, whereas a change of about 1% in the inert gas concentration Z is so small that it cannot be detected. Therefore, when Y, Z≦1%, VGRS(100%-YZ,Y,Z)≒VGRS(100%-Y,Y,0%) (5) Therefore, if ΔVg is defined as follows, δVg(100%-Y,Y,0%) =VGS(100%-Y,Y,0%)-VGS(100%,0%,0%) ≒VGS(100%-YZ,Y,Z)-VGS(100%,0%,0%) (6) This becomes:

[0104] ΔVg represents the shift in threshold voltage of the gas sensor due to the poisoning impurity.

[0105] In this way, the concentration Y of the poisonous impurity can be detected without being affected by the presence or absence of an inert gas of about 1% or less.

[0106] 8A is a graph showing the dependency of ΔVg(X, 0%, 100%-X) on the hydrogen concentration X. FIG. 8B is a graph showing the dependency of ΔVg(100%-Y, Y, 0%) on the poisoning impurity concentration Y.

[0107] Focusing on the vicinity of X=100% in FIG. 8A, it can be seen that the change in ΔVg is small, and when the hydrogen concentration X is 100% as described above, even a small amount of inert gas mixed in has no effect.

[0108] As shown in FIG. 8B, ΔVg is sensitive to the concentration Y of the poisoning impurity in hydrogen, but the poisoning impurity concentration Y has a detection limit (ΔVgmin), and if the poisoning impurity concentration Y is too low, only a ΔVg below the detection limit can be obtained.

[0109] In this embodiment, by using the hydrogen pump 26 to concentrate and measure the poisoning impurities at a high concentration, even if the poisoning impurity concentration Y in the atmosphere to be detected is below the detection limit, the poisoning impurity concentration Y in the space 40 inside the chamber 25 can be made higher than the detection limit, and the poisoning impurity concentration Y in the space 40 inside the chamber 25 can be measured. Then, the poisoning impurity concentration Y in the atmosphere to be detected can be estimated from the poisoning impurity concentration Y in the space 40 inside the chamber 25 that has been measured.

[0110] The properties of the hydrogen pump 26 will be described with reference to FIGS. 9A to 14. FIG.

[0111] 9A is a diagram showing differences in how impurities are concentrated in the space 40 inside the chamber 25 due to differences in gas temperature. Initially (time 0), the impurity concentration in the space 40 inside the chamber 25 is assumed to be equal to the impurity concentration in the atmosphere to be detected.

[0112] For a while from the beginning, the concentration of impurities in the space 40 inside the chamber 25 is low, and the influence of impurities discharged by diffusion through the gas flow path 30 can be ignored. As the hydrogen pump 26 selectively discharges hydrogen, hydrogen containing impurities in an amount equal to the amount of hydrogen discharged by the hydrogen pump 26 is introduced into the chamber 25 from the gas flow path 30. As a result, the impurities are concentrated in the space 40 inside the chamber 25, and the concentration becomes high.

[0113] The amount (number of molecules) of hydrogen containing impurities introduced from gas flow path 30 is determined by the current of hydrogen pump 26 and is therefore constant regardless of temperature. However, when the gas temperature is high, the gas expands, reducing the number of molecules present in space 40 inside chamber 25. Therefore, for the same amount of impurity molecules introduced, the higher the gas temperature, the faster the impurities concentrate. Therefore, at higher temperatures, the rate of increase in impurity concentration over time, i.e., the slope of the graph in Figure 9A, becomes greater.

[0114] As the concentration of impurities progresses, the diffusion of impurities through gas flow path 30 increases, and eventually the inflow of impurities due to the gas flow from the detection target atmosphere (outside chamber 25) to the inside of chamber 25 and the discharge of impurities due to diffusion from the inside of chamber 25 to the detection target atmosphere (outside chamber 25) are balanced. Then, the impurity concentration stops changing over time and becomes saturated.

[0115] When the temperature of the gas flowing through the gas flow path 30 increases, the amount of impurities released by diffusion through the gas flow path 30 increases, and therefore, the inflow of impurities from the detection target atmosphere due to the gas flow and the release of impurities due to diffusion are balanced when the impurity concentration in the internal space 40 of the chamber 25 is lower. In other words, when the gas temperature is high, the impurity concentration saturates at a low impurity concentration. In other words, the higher the gas temperature, the shorter the saturation time.

[0116] The temperature of the gas flowing through the gas flow passage 30, that is, the temperature of the gas inside the chamber 25, can be controlled by a heater C installed in the gas flow passage 30.

[0117] 9B is a diagram showing the gas flow path 30 including a heater C31. The gas flow path 30 can include the heater C31 that controls the temperature of the gas flow path 30. The heater C31 can heat the gas flow path 30 to a high temperature.

[0118] By adjusting the temperature of the gas flow path 30, it is possible to control the balance between the inflow of impurities from the detection target atmosphere due to the gas flow and the discharge of impurities due to diffusion, while maintaining a lower impurity concentration in the space 40 inside the chamber 25. When the gas flow path 30 is heated to a high temperature, the impurity concentration in the space 40 inside the chamber 25 saturates at a low concentration in a short time. When the gas flow path 30 is cooled to a low temperature, the impurity concentration in the space 40 inside the chamber 25 saturates at a high concentration in a long time.

[0119] FIG. 10 is a diagram showing the difference in the concentration of impurities in the internal space 40 of the chamber 25 depending on the current flowing through the hydrogen pump 26, i.e., the number of hydrogen molecules discharged from the internal space 40 of the chamber 25 via the hydrogen pump 26 per unit time.

[0120] Hydrogen pump 26 discharges hydrogen from space 40 inside chamber 25 to the detection target atmosphere (outside chamber 25). At this time, the same amount of hydrogen (including impurities) as the hydrogen discharged by hydrogen pump 26 flows from the detection target atmosphere through gas flow path 30 into space 40 inside chamber 25.

[0121] When the current of the hydrogen pump 26 is large, a large amount of hydrogen molecules are discharged per unit time from the space 40 inside the chamber 25 via the hydrogen pump 26 into the detection target atmosphere. As a result, the number of impurity molecules flowing into the space 40 inside the chamber 25 through the gas flow path 30 per unit time increases, and the impurities in the space 40 inside the chamber 25 are quickly concentrated and their concentration increases. On the other hand, when the current of the hydrogen pump 26 is large, the flow rate of the gas flowing in from the detection target atmosphere through the gas flow path 30 is fast, so when the impurity concentration in the space 40 inside the chamber 25 becomes higher, the inflow of impurities from the detection target atmosphere due to the gas flow and the discharge of impurities due to diffusion balance each other.

[0122] 10, when the current of hydrogen pump 26 is large, the impurity concentration in space 40 inside chamber 25 saturates at a high value and takes a long time to saturate. When the current of hydrogen pump 26 is small, the impurity concentration saturates at a low value and takes a short time to saturate. Therefore, by controlling the current of hydrogen pump 26, it is possible to control the impurity concentration and saturation time in space 40 inside chamber 25 for the same impurity concentration in the atmosphere to be detected.

[0123] 11A and 11B are diagrams showing differences in how impurities are concentrated in the space 40 inside the chamber 25 due to differences in the shape of the gas flow path 30. FIG.

[0124] 11A is a diagram showing differences in how impurities are concentrated in the space 40 inside the chamber 25 depending on the length of the gas flow path 30. Initially (time 0), the impurity concentration in the space 40 inside the chamber 25 is assumed to be equal to the impurity concentration in the atmosphere to be detected.

[0125] The amount (number of molecules) of hydrogen containing impurities introduced from the gas flow path 30 is determined by the current of the hydrogen pump 26, as explained with reference to Figure 9, and is therefore constant regardless of the cross-sectional area of ​​the gas flow path 30. For this reason, for a while from the beginning, the same amount of impurity molecules is introduced and the concentration of impurities proceeds at the same rate whether the length L of the gas flow path 30 is long or short. In other words, the slope of the graph in Figure 11A is the same regardless of the length of the gas flow path 30.

[0126] As the concentration of impurities progresses, the diffusion of impurities in the gas flow path 30 increases, and eventually the inflow of impurities from the atmosphere to be detected by the gas flow and the discharge of impurities by diffusion balance each other, and the impurity concentration stops changing over time and becomes saturated.

[0127] When the length L of the gas flow path 30 is large, the amount of impurities released by diffusion through the gas flow path 30 decreases, so that when the impurity concentration in the space 40 inside the chamber 25 is high, the inflow of impurities from the detection target atmosphere due to the gas flow and the release of impurities by diffusion are balanced. In other words, the longer the length L of the gas flow path 30, the higher the impurity concentration at which the impurity concentration saturates. That is, the longer the length L of the gas flow path 30, the longer the saturation time.

[0128] The length L of the gas flow path 30 is determined when the chamber 25 is manufactured. The gas temperature described with reference to Fig. 9 and the current of the hydrogen pump 26 described with reference to Fig. 10 are determined when the gas sensor is used.

[0129] 11B is a diagram showing differences in how impurities are concentrated in the space 40 inside the chamber 25 due to differences in the cross-sectional area of ​​the gas flow path 30. Initially (time 0), the impurity concentration in the space 40 inside the chamber 25 is assumed to be equal to the impurity concentration in the atmosphere to be detected.

[0130] 11A, the same amount of impurity molecules is introduced and the concentration of impurities proceeds at the same rate whether the cross-sectional area of ​​the gas flow path 30 is large or small. In other words, the slope of the graph in FIG. 11B is the same regardless of the cross-sectional area of ​​the gas flow path 30.

[0131] As the concentration of impurities progresses, the diffusion of impurities in the gas flow path 30 increases, and eventually the inflow of impurities from the atmosphere to be detected by the gas flow and the discharge of impurities by diffusion balance each other, and the impurity concentration stops changing over time and becomes saturated.

[0132] When the cross-sectional area of ​​the gas flow path 30 is large, the amount of impurities discharged by diffusion through the gas flow path 30 increases, so that the inflow of impurities from the detection target atmosphere due to the gas flow and the discharge of impurities due to diffusion are balanced when the impurity concentration in the internal space 40 of the chamber 25 is low. In other words, the larger the cross-sectional area of ​​the gas flow path 30, the lower the impurity concentration becomes, and the shorter the saturation time becomes.

[0133] The cross-sectional area of ​​the gas flow passage 30, like the length L of the gas flow passage 30, is determined when the chamber 25 is manufactured.

[0134] 12A is a diagram showing differences in how impurities are concentrated in the space 40 inside the chamber 25 due to differences in the volume of the space 40 inside the chamber 25. Initially (time 0), the impurity concentration in the space 40 inside the chamber 25 is assumed to be equal to the impurity concentration in the atmosphere to be detected.

[0135] The amount (number of molecules) of hydrogen containing impurities introduced from gas flow path 30 is determined by the current of hydrogen pump 26 as explained with reference to Fig. 9, and is therefore constant regardless of the cross-sectional area of ​​gas flow path 30. Therefore, if the same amount of impurity molecules is introduced when the volume of space 40 inside chamber 25 is small and when it is large, the impurities will be concentrated more quickly when the volume of space 40 inside chamber 25 is small. In other words, the slope of the graph in Fig. 12A is greater when the volume of space 40 inside chamber 25 is small.

[0136] As the concentration of impurities progresses, the diffusion of impurities in the gas flow path 30 increases, and eventually the inflow of impurities from the atmosphere to be detected by the gas flow and the discharge of impurities by diffusion balance each other, and the impurity concentration stops changing over time and becomes saturated.

[0137] If the shape of gas flow path 30 is the same, the inflow of impurities from the detection target atmosphere due to gas flow and the discharge of impurities due to diffusion will be balanced when the impurity concentration in space 40 inside chamber 25 is constant, regardless of the volume of space 40 inside chamber 25. In other words, the smaller the volume of space 40 inside chamber 25, the shorter the saturation time, but the impurity concentration when saturated is constant and does not depend on the volume of space 40 inside chamber 25.

[0138] The volume of the space 40 inside the chamber 25 is determined when the chamber 25 is manufactured, similar to the length L and cross-sectional area of ​​the gas flow path 30.

[0139] 12B is a diagram showing differences in how impurities are concentrated in the space 40 inside the chamber 25 due to differences in the impurity concentration in the atmosphere to be detected. Initially (time 0), the impurity concentration in the space 40 inside the chamber 25 is assumed to be equal to the impurity concentration in the atmosphere to be detected.

[0140] If the impurity concentration in the detection target atmosphere is high, the impurity concentration will saturate when the impurity concentration in the internal space 40 of the chamber 25 reaches a higher value. The impurity concentration that saturates in the internal space 40 of the chamber 25 is proportional to the impurity concentration in the detection target atmosphere. Therefore, if a proportionality coefficient between the impurity concentration that saturates in the internal space 40 of the chamber 25 and the impurity concentration in the detection target atmosphere is determined in advance, the impurity concentration in the detection target atmosphere can be estimated by measuring the value when the impurity concentration in the internal space 40 of the chamber 25 reaches saturation using the gas sensor chip 20. This proportionality coefficient can be changed, for example, by changing the magnitude of the proton current.

[0141] 13 is a diagram showing differences in how impurities are concentrated in the space 40 inside the chamber 25 due to differences in the impurity concentration in the space 40 inside the chamber 25 at the initial stage (time 0). The impurity concentration in the atmosphere to be detected is assumed to be constant.

[0142] It is not always possible to start concentrating impurities after the impurity concentration in the space 40 inside the chamber 25 has matched with the impurity concentration in the atmosphere to be detected. However, regardless of the value of the initial impurity concentration in the space 40 inside the chamber 25, if the impurity concentration in the atmosphere to be detected is constant, the impurity concentration in the space 40 inside the chamber 25 will saturate at the same value regardless of the initial value. This property eliminates the need to consider the initial impurity concentration in the space 40 inside the chamber 25.

[0143] Therefore, after the impurity concentration has reached saturation, the impurity concentration in the space 40 inside the chamber 25 is measured with the gas sensor chip 20, and the impurity concentration in the atmosphere to be detected can be estimated by utilizing the proportional relationship explained with reference to Figure 12B.

[0144] 14 is a diagram for explaining the dilution operation of impurities in the space 40 inside the chamber 25 when performing the refresh operation of the sensor FET 21. It should be noted that the vertical axis in FIG. 14 is a logarithmic scale.

[0145] In the impurity dilution operation, hydrogen is selectively taken from the target atmosphere into the internal space 40 of the chamber 25 by the hydrogen pump 26. At this time, the hydrogen containing impurities that was originally present in the internal space 40 of the chamber 25 is discharged into the target atmosphere through the gas flow path 30. The dilution of the impurities progresses exponentially with respect to time. In other words, the impurity concentration in the internal space 40 of the chamber 25 changes with time as shown by the straight line in FIG. 14.

[0146] When the impurity concentration is sufficiently reduced, the number of impurity molecules released from the internal space 40 of the chamber 25 into the atmosphere to be detected by the gas flow in the gas flow path 30 matches the number of impurity molecules that flow from the atmosphere to be detected into the internal space 40 of the chamber 25 through the gas flow path 30 by diffusion, and the impurity concentration in the internal space 40 of the chamber 25 becomes saturated.

[0147] Therefore, if the impurity concentration is reduced to a concentration at which the sensor FET 21 of the gas sensor chip 20 can be refreshed and the temperature of the gas sensor chip 20 is raised to about 200° C. using the heater B27, the sensor FET 21 can be refreshed.

[0148] In the case of the refresh operation, the impurity concentration in the space 40 inside the chamber 25 after saturation is proportional to the impurity concentration in the atmosphere to be detected. The proportionality coefficient of this proportional relationship can be changed, for example, by changing the magnitude of the proton current. However, in the case of the refresh operation, it is not always necessary to measure the impurity concentration.

[0149] In the above description, an example has been described in which N-type FETs are used for the sensor FET 21 and the reference FET 22. The sensor FET 21 and the reference FET 22 may also be P-type FETs.

[0150] Fig. 15A is a graph showing the gate voltage-drain current characteristics of the sensor FET 21 using a P-type FET. Fig. 15B is a graph showing the gate voltage-drain current characteristics of the reference FET 22 using a P-type FET. Figs. 15A and 15B correspond to Figs. 6A and 6B, respectively.

[0151] The threshold voltage of the sensor FET 21 using a P-type FET shifts in the same way as the sensor FET 21 using an N-type FET. That is, when the poisoning impurity concentration Y is 0% and the hydrogen concentration X is increased from a state in which the inert gas concentration Z is 100%, the threshold voltage of the sensor FET 21 shifts toward a negative voltage. When the poisoning impurity concentration Y is increased from a state in which the hydrogen concentration X is 100%, the threshold voltage shifts toward a positive voltage.

[0152] The reference FET 22 using a P-type FET is not affected by gas components, similar to the reference FET 22 using an N-type FET.

[0153] The response of the sensor FET 21 using a P-type FET to the hydrogen concentration X is the same as that shown in Fig. 8A. The response of the sensor FET 21 using a P-type FET to the concentration Y of the poisonous impurity in hydrogen is the same as that shown in Fig. 8B. The operation of the hydrogen pump 26 shown in Figs. 9A to 14 is the same regardless of the type of gas sensor chip 20.

[0154] <Variation 1> The chamber 25 shown in FIGS. 4A and 4B is made of metal. The chamber 25 can also be made of laminated insulating films. In the following, it is assumed that the laminated insulating films are made of laminated ceramic. The laminated ceramic (insulating film) can be made of alumina (Al2O3) with a thickness of approximately 100 micrometers, for example. A chamber 25 made of laminated ceramic can more easily reduce the volume of the space 40 inside the chamber 25 compared to a chamber 25 made of metal. Power can be supplied to the gas sensor chip 20 using known wiring between the ceramic layers. Therefore, even a chamber 25 made of laminated ceramic can achieve both gas sealing and wiring access.

[0155] 16A and 16B are diagrams showing an example of the configuration of a gas sensing unit 1001 including a chamber 25 made of laminated ceramics. Figures 16A and 16B correspond to Figures 4A and 4B, respectively.

[0156] The space 40 inside the chamber 25 is connected to the atmosphere to be detected via a gas flow path 30. The cross section of the gas flow path 30 can be, for example, 100 micrometers by 1 millimeter. The length of the gas flow path 30 can be, for example, 5 millimeters or less. The volume of the space 40 inside the chamber 25 can be, for example, 10 cubic millimeters or less.

[0157] The space 40 inside the chamber 25 is connected to the detection target atmosphere via a hydrogen pump 26, separate from the gas flow path 30. The hydrogen pump 26 includes a first electrode 26a exposed to the space 40 inside the chamber 25, a second electrode 26c exposed to the detection target atmosphere, and a proton conductor layer 26b located between the first electrode 26a and the second electrode 26c. A voltage can be applied from a power source 1007 to the first electrode 26a and the second electrode 26c.

[0158] The gas sensor chip 20 is preferably located between the gas flow passage 30 and the hydrogen pump 26 in the direction of gas flow in the gas flow passage 30 .

[0159] The gas sensor according to this embodiment, even if it has the chamber 25 made of laminated ceramic, can obtain the same effect as when it has the chamber 25 made of metal.

[0160] That is, when measuring the impurity concentration using the gas sensor chip 20, the impurities in the internal space 40 of the chamber 25 are concentrated using the hydrogen pump 26, the impurity concentration in the internal space 40 of the chamber 25 is saturated at a value proportional to the impurity concentration in the atmosphere to be detected, the impurity concentration in the atmosphere to be detected can be estimated by measuring the impurity concentration using the gas sensor chip 20 after the impurity concentration in the internal space 40 of the chamber 25 is saturated, and the sensor FET 21 can be refreshed by diluting the impurities in the internal space 40 of the chamber 25 using the hydrogen pump 26 and heating the gas sensor chip 20 to about 200°C using the heater B27, in the same manner as in the chamber 25 made of laminated ceramic or metal.

[0161] Compared to a chamber 25 made of metal, a chamber 25 made of laminated ceramic can easily reduce the volume of the space 40 inside the chamber 25, and therefore can easily shorten the saturation time for concentrating and diluting impurities for the same current of the hydrogen pump 26. In other words, a chamber 25 made of laminated ceramic requires less current for the hydrogen pump 26 to achieve the same saturation time as a chamber 25 made of metal, which is advantageous for reducing power consumption.

[0162] The gas sensor chip 20 is assumed to be used in an environment where the number of moles of gas introduced into the internal space 40 of the chamber 25 is negligibly small compared to the number of moles of gas present in the atmosphere to be detected.

[0163] The chamber 25 can also be configured as a MEMS (Micro Electro Mechanical Systems). The chamber 25 configured with MEMS can be formed using a MEMS process. For example, an additional process is performed on the gas sensor chip 20 to form a micro-sized chamber 25 surrounded by an insulating film layer on the sensor FET 21, thereby forming the chamber 25 configured with the MEMS process. The thickness of the space 40 inside the chamber 25 can be, for example, 1 micrometer.

[0164] The chamber 25 made of MEMS is advantageous for miniaturization because the volume of the space 40 inside the chamber 25 can be reduced more easily than in the case of a multilayer ceramic chamber 25. In the chamber 25 made of MEMS, power can be easily supplied to the elements on the gas sensor chip 20 using the electrode wiring and electrode pads on the gas sensor chip 20.

[0165] 17A and 17B are diagrams showing an example of the configuration of a gas sensing unit 1001 including a chamber 25 made of MEMS. Figures 17A and 17B correspond to Figures 4A and 4B, respectively.

[0166] The space 40 inside the chamber 25 is connected to the atmosphere to be detected via a gas flow path 30. The cross section of the gas flow path 30 may be, for example, 1 micrometer by 1 millimeter. The length of the gas flow path 30 may be, for example, 1 millimeter or less. The volume of the space 40 inside the chamber 25 may be, for example, 1 cubic millimeter or less.

[0167] The space 40 inside the chamber 25 is connected to the detection target atmosphere via a hydrogen pump 26, separate from the gas flow path 30. The hydrogen pump 26 includes a first electrode 26a exposed to the space 40 inside the chamber 25, a second electrode 26c exposed to the detection target atmosphere, and a proton conductor layer 26b located between the first electrode 26a and the second electrode 26c. A voltage can be applied from a power source 1007 to the first electrode 26a and the second electrode 26c.

[0168] The gas sensor chip 20 (sensor FET 21) is preferably located between the gas flow passage 30 and the hydrogen pump 26 in the direction of gas flow in the gas flow passage 30.

[0169] The gas sensor according to this embodiment, even if it is provided with the chamber 25 made of MEMS, can obtain the same effects as when it is provided with the chamber 25 made of metal or the chamber 25 made of laminated ceramic.

[0170] That is, when measuring the impurity concentration using the gas sensor chip 20, the impurities in the internal space 40 of the chamber 25 are concentrated using the hydrogen pump 26, the impurity concentration in the internal space 40 of the chamber 25 is saturated at a value proportional to the impurity concentration in the atmosphere to be detected, the impurity concentration in the atmosphere to be detected can be estimated by measuring the impurity concentration using the gas sensor chip 20 after the impurity concentration in the internal space 40 of the chamber 25 is saturated, and the sensor FET 21 can be refreshed by diluting the impurities in the internal space 40 of the chamber 25 using the hydrogen pump 26 and heating the gas sensor chip 20 to about 200°C using the heater B27, in the same manner as in the chamber 25 made of MEMS or the chamber 25 made of metal or laminated ceramic.

[0171] The chamber 25 made of MEMS can easily reduce the volume of the space 40 inside the chamber 25 compared to a chamber 25 made of metal or a chamber 25 made of laminated ceramic, and therefore can easily shorten the saturation time for concentrating and diluting impurities for the same current of the hydrogen pump 26. In other words, the chamber 25 made of MEMS requires less current for the hydrogen pump 26 to achieve the same saturation time as a metal chamber 25, which is advantageous for reducing power consumption.

[0172] The gas sensor chip 20 is assumed to be used in an environment where the number of moles of gas introduced into the internal space 40 of the chamber 25 is negligibly small compared to the number of moles of gas present in the atmosphere to be detected.

[0173] <Gas sensing unit installation method> The gas sensing unit 1001, which includes the chamber 25 and the gas sensor chip 20, has a configuration as shown in Figures 4, 16, and 17. In order to measure impurities in hydrogen, the gas sensing unit 1001 needs to be exposed to the hydrogen to be measured.

[0174] FIG. 18 is a diagram showing an example of installation of the gas sensing unit 1001 provided in the gas sensor according to this embodiment.

[0175] The gas sensing unit 1001 is installed in a measurement unit provided in the pipe 50. Hydrogen gas containing impurities, which is the gas of the atmosphere to be detected, flows through the pipe 50. The gas sensing unit 1001 can be exposed to the hydrogen gas flowing through the pipe 50, which is the hydrogen to be measured.

[0176] The gas sensing unit 1001 has a small volume and can be easily installed in the pipe 50. The circuit unit 1008 of the gas sensor can be installed at a location separated by the length of the wiring by connecting it to the gas sensing unit 1001.

[0177] <Variation 2> In the above description, an example has been described in which the sensor FET 21 and the reference FET 22 are used in the gas sensing unit 1001 provided in the gas sensor chip 20. The gas sensing unit 1001 may be provided with a sensor capacitor (SCAP) and a reference capacitor (RCAP) instead of the sensor FET 21 and the reference FET 22. The sensor capacitor and the reference capacitor may be configured as MOS capacitors.

[0178] 19A is a diagram showing a cross section of a sensor capacitor (SCAP) 61. The sensor capacitor 61 includes a semiconductor substrate 1, a well 2, a gate insulating film 5, a metal oxide layer 106 serving as a gas detection material, a catalytic metal gate layer 107, and an interlayer insulating film 108. The surface of the catalytic metal gate layer 107 is exposed to the atmosphere to be detected (the atmosphere in the space 40 inside the chamber 25).

[0179] 19B is a diagram showing a cross section of a reference capacitor (RCAP) 62. The reference capacitor 62 includes a semiconductor substrate 1, a well 12, a gate insulating film 15, a metal oxide layer 116 serving as a gas sensing material, a catalytic metal gate layer 117, and an interlayer insulating film 118. The surface of the catalytic metal gate layer 117 is covered with the interlayer insulating film 118, and is isolated from the atmosphere to be sensed.

[0180] For example, silicon or silicon carbide (SiC) can be used for the semiconductor substrate 1. When a silicon substrate is used, the operating temperature of the sensor capacitor 61 and the reference capacitor 62 is limited to a maximum of approximately 250°C. On the other hand, when a silicon carbide substrate is used, the cost of the substrate increases compared to a silicon substrate, but the operating temperature of the sensor capacitor 61 and the reference capacitor 62 can be increased to 500°C or more. Therefore, different substrate materials can be used depending on the application of the sensor.

[0181] For example, titanium oxide or yttria-stabilized zirconia (YSZ) can be used for the metal oxide layers 106 and 116. For the catalytic metal gate layers 107 and 117, noble metals such as platinum, palladium, and iridium, or nickel can be used.

[0182] The wells 2 and 12 and the catalytic metal gate layers 107 and 117 are connected to a wiring layer made of a metal such as aluminum, tungsten, or platinum, so that power can be supplied from the power source 1007 shown in FIG.

[0183] In the following description, the sensor capacitor 61 and the reference capacitor 62 are both described as having N-type wells 2, 12, but they may also have P-type wells 2, 12.

[0184] The gas sensing unit 1001 including the sensor capacitor 61 and the reference capacitor 62 does not include the source diffusion layers 3, 13 and the drain diffusion layers 4, 14, and therefore has a simpler structure than the gas sensing unit 1001 including the sensor FET 21 and the reference FET 22. On the other hand, the gas sensing unit 1001 including the sensor capacitor 61 and the reference capacitor 62 needs to measure the capacitance using an AC voltage, which is not necessary in the gas sensing unit 1001 including the sensor FET 21 and the reference FET 22, as will be described later.

[0185] Fig. 20A is a diagram showing an example of a method of supplying power to the sensor capacitor 61, the reference capacitor 62, the heater A 24, and the diode 29. Fig. 20B is a diagram showing an example of a method of supplying power to the hydrogen pump 26 and the heater B 27.

[0186] The sensor capacitor 61 and the reference capacitor 62 are heated to a predetermined temperature by Joule heat generated by passing a current through the heater A24. The wells 2 and 12 of the sensor capacitor 61 and the reference capacitor 62 are set to 0 V, a variable DC voltage VGR is applied to the gate of the reference capacitor 62, and a variable DC voltage VGS is applied to the gate of the sensor capacitor 61.

[0187] In addition to the DC voltages VGR and VGS, an AC voltage with an amplitude Vsig is applied to the terminals of wells 2 and 12. The AC current flowing through the gate terminals of the sensor capacitor 61 and the reference capacitor 62 is measured by the current detection unit 1004 (FIG. 1), thereby detecting the capacitance C(SCAP) of the sensor capacitor 61 and the capacitance C(RCAP) of the reference capacitor 62.

[0188] The control unit 1003 controls the voltages VGS and VGR so that the capacitance C(SCAP) of the sensor capacitor 61 and the capacitance C(RCAP) of the reference capacitor 62 both become the same constant capacitance C0. The difference VGRS between VGR and VGS at this time is the same as in the case of the gas sensing unit 1001 including the sensor FET 21 and the reference FET 22.

[0189] Fig. 21A is a diagram showing the capacitance-gate voltage characteristics of the sensor capacitor 61. Fig. 21B is a diagram showing the capacitance-gate voltage characteristics of the reference capacitor 62. Figs. 21A and 21B show an example in which the well 2 is an N-type semiconductor.

[0190] In the sensor capacitor 61, when a negative voltage is applied to the catalyst metal gate layer 107 with respect to the well 2, the surface of the well 2 through the catalyst metal gate layer 107, the metal oxide layer 106, and the gate insulating film 5 is depleted, and the capacitance decreases. When a positive voltage is applied to the catalyst metal gate layer 107 with respect to the well 2, the surface of the well 2 through the catalyst metal gate layer 107, the metal oxide layer 106, and the gate insulating film 5 is put into an accumulation state, and the capacitance increases. The gate voltage when the capacitance coincides with a value (C0) determined between the capacitance in the depletion state and the capacitance in the accumulation state can be defined as the threshold voltage.

[0191] The reference capacitor 62 is similar to the sensor capacitor 61; when a negative voltage is applied to the catalyst metal gate layer 117 with respect to the well 12, the surface of the well 12 via the catalyst metal gate layer 117, the metal oxide layer 116, and the gate insulating film 15 is depleted, and the capacitance decreases. When a positive voltage is applied to the catalyst metal gate layer 117 with respect to the well 12, the surface of the well 12 via the catalyst metal gate layer 117, the metal oxide layer 116, and the gate insulating film 15 is in an accumulation state, and the capacitance increases. The gate voltage when the capacitance matches a value (C0) determined between the capacitance in the depletion state and the capacitance in the accumulation state can be defined as the threshold voltage.

[0192] In the sensor capacitor 61, the catalytic metal gate layer 107 is exposed to the atmosphere in the space 40 inside the chamber 25, so the work function of the catalytic metal gate layer 107 changes depending on the type of gas. The way in which the threshold voltage changes due to hydrogen and the way in which the threshold voltage changes due to poisoned impurities is the same as in the case of the sensor FET 21.

[0193] Let X be the concentration of hydrogen in the space 40 inside the chamber 25, Y be the concentration of poisonous impurities (carbon monoxide, hydrogen sulfide, etc.), and Z be the concentration of inert impurities (nitrogen, etc.). X+Y+Z=100%.

[0194] First, consider the case where the poisoning impurity concentration Y is 0%. When the hydrogen concentration X is a finite value X1, the capacitance-gate voltage characteristics shift in the negative voltage direction compared to when the hydrogen concentration X is 0%. As a result, when the capacitance is C0, the gate voltage changes from VGS(0%, 0%, 100%) to VGS(X1, 0%, 100%-X1).

[0195] On the other hand, the capacitance-gate voltage characteristics of the reference capacitor 62 do not change even if the gas concentration changes because the catalytic metal gate layer 117 is covered with the interlayer insulating film 118. That is, in the reference capacitor 62, the capacitance remains constant at C0 even when a voltage VGR is applied to the gate.

[0196] In the capacitance-gate voltage characteristics of the sensor capacitor 61, when a poisoning impurity is present, i.e., when the poisoning impurity concentration Y and the inert gas concentration Z are finite, VGS shifts parallel to the positive voltage direction compared to when the hydrogen concentration X=100% (Y=Z=0%). VGS is very sensitive to the poisoning impurity concentration Y and can detect even a ppm change, whereas a change of about 1% in the inert gas concentration Z is so small that it cannot be detected. This is similar to the case when the gas sensor chip 20 includes the sensor FET 21 and the reference FET 22.

[0197] In this way, the properties of equations (2) to (6) when the sensor FET 21 and the reference FET 22 are used also hold when the sensor capacitor 61 and the reference capacitor 62 are used. As a result, the properties shown in Fig. 8 are also obtained when the sensor capacitor 61 and the reference capacitor 62 are used. The operation of the hydrogen pump 26 shown in Figs. 9 to 14 remains the same regardless of the type of gas sensor chip 20.

[0198] In the example shown in Figures 21A and 21B, the sensor capacitor 61 and the reference capacitor 62 have N-type wells 2 and 12, but the sensor capacitor 61 and the reference capacitor 62 can also have P-type wells 2 and 12.

[0199] 1 to 17 have been used to describe an example in which the gas sensing unit 1001 includes a sensor FET 21 and a reference FET 22, and FIG. 19 to 21B have been used to describe an example in which the gas sensing unit 1001 includes a sensor capacitor 61 and a reference capacitor 62. The gas sensing unit 1001 can also include a sensor diode and a reference diode instead of these. The sensor diode and the reference diode are configured as diodes.

[0200] 22A is a diagram showing a cross section of a sensor diode (SDIODE) 71. The sensor diode 71 includes a semiconductor substrate 1, a well 2, a metal oxide layer 106 serving as a gas detection material, a catalytic metal gate layer 107, and an interlayer insulating film 108. The surface of the catalytic metal gate layer 107 is exposed to the atmosphere to be detected (the atmosphere in the space 40 inside the chamber 25).

[0201] 22B is a diagram showing a cross section of a reference diode 72. The reference diode 72 includes a semiconductor substrate 1, a well 12, a metal oxide layer 116 serving as a gas sensing material, a catalytic metal gate layer 117, and an interlayer insulating film 118. The surface of the catalytic metal gate layer 117 is covered with the interlayer insulating film 118, and is isolated from the atmosphere to be sensed.

[0202] The semiconductor substrate 1 can be made of, for example, silicon or silicon carbide (SiC). When a silicon substrate is used, the operating temperature of the sensor diode 71 and the reference diode 72 is limited to a maximum of approximately 250°C. On the other hand, when a silicon carbide substrate is used, the cost of the substrate increases compared to a silicon substrate, but the operating temperature of the sensor diode 71 and the reference diode 72 can be increased to 500°C or more. Therefore, different substrate materials can be used depending on the application of the sensor.

[0203] For example, titanium oxide or yttria-stabilized zirconia (YSZ) can be used for the metal oxide layers 106 and 116. For the catalytic metal gate layers 107 and 117, noble metals such as platinum, palladium, and iridium, or nickel can be used.

[0204] The wells 2 and 12 and the catalytic metal gate layers 107 and 117 are connected to a wiring layer made of a metal such as aluminum, tungsten, or platinum, so that power can be supplied from the power source 1007 shown in FIG.

[0205] In the following description, the sensor diode 71 and the reference diode 72 are both described as having N-type wells 2, 12, but they may also have P-type wells 2, 12.

[0206] The gas sensing unit 1001 including the sensor diode 71 and the reference diode 72 does not include source diffusion layers 3 and 13 and drain diffusion layers 4 and 14, as compared with the gas sensing unit 1001 including the sensor FET 21 and the reference FET 22. Furthermore, it does not require the use of AC voltage, as compared with the gas sensing unit 1001 including the sensor capacitor 61 and the reference capacitor 62. Therefore, the gas sensing unit 1001 including the sensor diode 71 and the reference diode 72 has a simple structure and requires a simple power supply circuit.

[0207] In the gas sensing unit 1001 including the sensor diode 71 and the reference diode 72, it is necessary to pass a direct current through the gas sensing material during operation, and it is conceivable that the gas sensing material may be deteriorated by the current passing through. In the configuration in which the gas sensing unit 1001 includes the sensor FET 21 and the reference FET 22, and in the configuration in which the gas sensing unit 1001 includes the sensor capacitor 61 and the reference capacitor 62, it is not necessary to pass a direct current through the gas sensing material. For this reason, the gas sensing unit 1001 including the sensor diode 71 and the reference diode 72 may relatively raise concerns about reliability.

[0208] Fig. 23A is a diagram showing an example of a method of supplying power to the sensor diode 71, the reference diode 72, the heater A 24, and the diode 29. Fig. 23B is a diagram showing an example of a method of supplying power to the hydrogen pump 26 and the heater B 27.

[0209] The sensor diode 71 and the reference diode 72 are heated to a predetermined temperature by Joule heat generated by passing a current through the heater A24. The wells 2 and 12 of the sensor diode 71 and the reference diode 72 are set to 0 V, a variable DC voltage VGR is applied to the gate of the reference diode 72, and a variable DC voltage VGS is applied to the gate of the sensor diode 71.

[0210] The voltages VGS and VGR are controlled by the control unit 1003 so that the current of the sensor diode 71 and the current of the reference diode 72 both become the same constant current Ic. At this time, the difference VGRS between VGR and VGS is the same as in the case of the gas sensing unit 1001 including the sensor FET 21 and the reference FET 22.

[0211] Fig. 24A is a diagram showing the current-gate voltage characteristics of the sensor diode 71. Fig. 24B is a diagram showing the current-gate voltage characteristics of the reference diode 72. Figs. 24A and 24B show an example in which the well 2 is an N-type semiconductor.

[0212] In the sensor diode 71, the catalytic metal gate layer 107 is exposed to the atmosphere in the space 40 inside the chamber 25, so the work function of the catalytic metal gate layer 107 changes depending on the type of gas. The way in which the threshold voltage changes due to hydrogen and the way in which the threshold voltage changes due to poisoned impurities is the same as in the case of the sensor FET 21.

[0213] Let X be the concentration of hydrogen in the space 40 inside the chamber 25, Y be the concentration of poisonous impurities (carbon monoxide, hydrogen sulfide, etc.), and Z be the concentration of inert impurities (nitrogen, etc.). X+Y+Z=100%.

[0214] First, consider the case where the poisoning impurity concentration Y is 0%. When the hydrogen concentration X is a finite value X1, the current-gate voltage characteristics shift in the negative voltage direction compared to when the hydrogen concentration X is 0%. As a result, when the current is Ic, the gate voltage changes from VGS(0%, 0%, 100%) to VGS(X1, 0%, 100%-X1).

[0215] On the other hand, in the reference diode 72, the current-gate voltage characteristics do not change even if the gas concentration changes because the catalytic metal gate layer 117 is covered with the interlayer insulating film 118. That is, in the reference diode 72, the current remains constant at Ic even when a voltage VGR is applied to the gate.

[0216] In the current-gate voltage characteristics of the sensor diode 71, when a poisoning impurity is present, i.e., when the poisoning impurity concentration Y and the inert gas concentration Z are finite, VGS shifts parallel to the positive voltage direction compared to when the hydrogen concentration X=100% (Y=Z=0%). VGS is very sensitive to the poisoning impurity concentration Y and can detect even a change in ppm, whereas a change of about 1% in the inert gas concentration Z is so small that it cannot be detected. This is similar to the case when the gas sensor chip 20 includes the sensor FET 21 and the reference FET 22.

[0217] In this way, the properties of equations (2) to (6) when the sensor FET 21 and reference FET 22 are used also hold when the sensor diode 71 and reference diode 72 are used. As a result, the properties shown in Fig. 8 are also obtained when the sensor diode 71 and reference diode 72 are used. The operation of the hydrogen pump 26 shown in Figs. 9 to 14 remains the same regardless of the type of gas sensor chip 20.

[0218] In the example shown in Figures 24A and 24B, the sensor diode 71 and the reference diode 72 have N-type wells 2 and 12, but the sensor diode 71 and the reference diode 72 can also have P-type wells 2 and 12.

[0219] In the explanation so far, when measuring the poisonous impurities contained in hydrogen, the poisonous impurities are concentrated using the hydrogen pump 26. This operation is appropriate when it is assumed that the concentration of poisonous impurities contained in hydrogen is below ppm, such as in purified hydrogen.

[0220] On the other hand, when measuring the impurity concentration in hydrogen containing a large amount of impurities, such as unpurified hydrogen, it is desirable to control the operation of the hydrogen pump 26 and control the current of the hydrogen pump 26 so that the impurity concentration in the space 40 inside the chamber 25 falls within the concentration range in which the gas sensor chip 20 is highly sensitive.

[0221] FIG. 25A is a diagram showing an example of the response of the sensor FET 21, the sensor capacitor 61, and the sensor diode 71 to the concentration of poisoning impurities.

[0222] 25A, the sensor FET 21, the sensor capacitor 61, and the sensor diode 71 have high sensitivity in the intermediate impurity concentration region. ΔVg increases as the gas concentration increases. The change in ΔVg relative to the change in gas concentration is large in the intermediate gas concentration (impurity concentration) region.

[0223] For this reason, it is desirable to control the operation of the hydrogen pump 26 and control the current of the hydrogen pump 26 so that the impurity concentration in the space 40 inside the chamber 25 falls within the concentration range in which the gas sensor chip 20 becomes highly sensitive.

[0224] FIG. 25B is a diagram showing the operation of the hydrogen pump 26 in response to the concentration of poisonous impurities in the atmosphere to be detected.

[0225] When the concentration of the poisoning impurity is very low, a large current is passed through the hydrogen pump 26 to concentrate the impurities in the space 40 inside the chamber 25, thereby increasing the impurity concentration. By doing so, the concentration of the poisoning impurity in the space 40 inside the chamber 25 can be saturated in the high sensitivity region of the sensor element. As a result, the concentration of the poisoning impurity in the space 40 inside the chamber 25 can be detected with high accuracy. Since the ratio between the impurity concentration in the atmosphere to be detected and the impurity concentration in the space 40 inside the chamber 25 can be determined from the magnitude of the proton current, the concentration of the poisoning impurity in the atmosphere to be detected can be estimated with high accuracy.

[0226] When the concentration of poisoning impurities in the atmosphere to be detected increases, the current of the hydrogen pump 26 is reduced. By doing so, the concentration of poisoning impurities in the space 40 inside the chamber 25 can be saturated in the high sensitivity region of the sensor element. As a result, the concentration of poisoning impurities in the space 40 inside the chamber 25 can be detected with high accuracy. The ratio of the impurity concentration in the atmosphere to be detected to the impurity concentration in the space 40 inside the chamber 25 can be determined from the magnitude of the proton current, so the concentration of poisoning impurities in the atmosphere to be detected can be estimated with high accuracy.

[0227] If the concentration of the poisoning impurity in the detection target atmosphere increases further and becomes higher than the high sensitivity region of the sensor element, the voltage of the hydrogen pump 26 is reversed, and the direction of the current flowing through the hydrogen pump 26 is reversed. By doing so, the poisoning impurity in the space 40 inside the chamber 25 is diluted, and the poisoning impurity concentration can be saturated in the high sensitivity region of the sensor element. As a result, the poisoning impurity concentration in the space 40 inside the chamber 25 can be detected with high accuracy. Since the ratio of the impurity concentration in the detection target atmosphere to the impurity concentration in the space 40 inside the chamber 25 can be determined from the magnitude of the proton current, the poisoning impurity concentration in the detection target atmosphere can be estimated with high accuracy. [Example]

[0228] A gas sensor according to Example 2 of the present invention will be described. While the gas sensor according to Example 1 includes one gas sensor chip 20, the gas sensor according to this example includes a plurality of gas sensor chips 20.

[0229] FIG. 26 is a block diagram showing an example of the configuration of the gas sensor according to this embodiment.

[0230] The gas sensor according to this embodiment includes a plurality of gas sensing units 1001 (n gas sensing units 1001-1, ..., 1001-n), a plurality of gas sensing units 2002 (m gas sensing units 2002-1, ..., 2002-m), and a system control unit 2008.

[0231] Each of the gas sensing units 1001-1 to 1001-n is the gas sensing unit 1001 described in the first embodiment, and includes a gas sensor chip 20, a chamber 25, a hydrogen pump 26, and a heater B 27. The gas sensor chip 20 includes a sensor FET 21, a reference FET 22, a thermometer unit 23, and a heater A 24.

[0232] Each of the gas sensing units 2002-1 to 2002-m does not include the chamber 25, the hydrogen pump 26, and the heater B 27 of the gas sensing unit 1001 described in the first embodiment, but includes only the gas sensor chip 20. In the gas sensing unit 2002, the gas sensor chip 20 is directly exposed to the atmosphere to be detected. The gas sensor chip 20 includes a sensor FET 21, a reference FET 22, a thermometer unit 23, and a heater A 24.

[0233] Depending on the quality of hydrogen to be detected, it may be possible to sense impurities and refresh them using the gas sensor chip 20 without concentrating or diluting the toxic impurities. In such cases, measuring the impurity concentration using the gas sensing unit 2002 of the gas sensor chip 20, which is directly exposed to the atmosphere to be detected, can shorten the response time compared to measuring the impurity concentration using the gas sensing unit 1001. The gas sensing unit 1001 tends to have a long response time because the gas sensor chip 20 is installed inside the chamber 25 and measures the impurity concentration after waiting for the impurities to be concentrated by the hydrogen pump 26.

[0234] The system control unit 2008 includes a gas concentration estimation unit 1002 that estimates the gas concentration from the signals of the gas sensing unit 1001 and the gas sensing unit 2002, a current detection unit 1004, a temperature detection unit 1005, a hydrogen pump current detection unit 1006, a power supply 1007 that applies voltage to the gas sensing unit 1001 and the gas sensing unit 2002, a control unit 1003, an I / O unit 1000, and a parameter recording unit 1009.

[0235] In this embodiment, a plurality of gas sensing units 1001 and a plurality of gas sensing units 2002 are installed in the atmosphere to be detected, and these plurality of gas sensing units 1001 and 2002 can be used to measure the concentration of impurities contained in the atmosphere to be detected.

[0236] FIG. 27 is a flowchart showing a process for measuring the concentration of impurities contained in an atmosphere to be detected using a plurality of gas sensing units 1001 and 2002 in this embodiment.

[0237] S1: Power on the system.

[0238] In S2, all of the gas sensor chips 20 are refreshed.

[0239] In S3, the temperature and work function changes of the gas sensor chips 20 of the gas sensing units 1001 and 2002, and the current of the hydrogen pump 26 of the gas sensing unit 1001 are measured.

[0240] In S4, the concentration of the poisoning impurities in the hydrogen is estimated from the measurement results in S3 using a known method.

[0241] In S5, the estimated concentration of the poisoning impurity is compared with a predetermined reference range, and it is determined whether operation of the hydrogen pump 26, i.e., concentration and dilution of the poisoning impurity, is required to measure the concentration of the poisoning impurity in the hydrogen. If the estimated concentration of the poisoning impurity is not within the reference range, it is determined that operation of the hydrogen pump 26 (concentration and dilution of the poisoning impurity) is required to measure the concentration of the poisoning impurity in the hydrogen.

[0242] If operation of the hydrogen pump 26 is not required to measure the concentration of poisonous impurities in hydrogen, the processes of S6 to S8 are executed, and if operation of the hydrogen pump 26 is required, the processes of S9 to S11 are executed.

[0243] In S6, the concentration of poisoning impurities is measured using the gas sensing unit 2002 (2002-1 to 2002-m). Specifically, in S7, the gas sensing unit 2002 measures the temperature and work function change of the gas sensor chip 20, and in S8, the gas concentration estimation unit 1002 estimates the concentration of poisoning impurities and the degree of poisoning of the gas sensor chip 20. The degree of poisoning of the gas sensor chip 20 is estimated based on the value of ΔVg (FIG. 25A), i.e., a value indicating how high the concentration of impurities detected is.

[0244] At this time, in S6, the gas sensing unit 1001 (1001-1 to 1001-n) operates the hydrogen pump 26 to dilute the impurities in the hydrogen gas inside the chamber 25 to reduce the concentration, thereby suppressing deterioration of the sensor FET 21 of the gas sensor chip 20 due to impurity poisoning.

[0245] In S9, the concentration of poisoning impurities is measured using the gas sensing unit 1001 (1001-1 to 1001-n). Specifically, in S10, the gas sensing unit 1001 measures the temperature and work function change of the gas sensor chip 20 and the current of the hydrogen pump 26, and in S11, the gas concentration estimation unit 1002 estimates the concentration of poisoning impurities and the degree of poisoning of the gas sensor chip 20. The degree of poisoning of the gas sensor chip 20 is estimated based on the value of ΔVg (FIG. 25A), i.e., a value indicating how high the concentration of impurities detected is.

[0246] At this time, in S9, the gas sensing unit 2002 (2002-1 to 2002-m) can be used to measure the concentration of poisonous impurities in parallel with the gas sensing unit 1001. However, since the gas sensing unit 2002 does not concentrate impurities, the measurement accuracy may be inferior to that of the gas sensing unit 1001.

[0247] In S12, the operator determines whether or not to end the measurement. If the measurement is to be ended, the process of S13 is executed, and if the measurement is not to be ended, the process of S14 is executed.

[0248] In S13, all the gas sensor chips 20 are refreshed, and the measurement is completed.

[0249] In S14, it is determined whether or not there is a gas sensor chip 20 that needs refreshing for each gas sensing unit 1001. For example, the degree of poisoning of the gas sensor chip 20 (the value of deterioration due to poisoning) estimated in S8 is compared with a predetermined reference value, and if the value of deterioration exceeds this reference value, it is determined that this gas sensor chip 20 needs refreshing. If there is a gas sensor chip 20 that needs refreshing, the process of S15 is executed, and if there is not, the process returns to S5 and continues from there.

[0250] In S15, the gas sensor chips 20 that need to be refreshed are refreshed. At this time, the concentration of poisoning impurities is measured using the gas sensor chips 20 that are not being refreshed. Even if the gas sensor chips 20 that are being refreshed do not measure the concentration of poisoning impurities, the gas sensor chips 20 that are not being refreshed can measure the concentration of poisoning impurities.

[0251] 28A and 28B are diagrams showing an example of the operation of the gas sensing units 1001 and 2002 over time in this embodiment. For simplicity of explanation, an example will be described in which the gas sensor includes two gas sensing units 1001 (gas sensing units 1001-1 and 1001-2) and two gas sensing units 2002 (gas sensing units 2002-1 and 2002-2). Even if the gas sensor includes three or more gas sensing units 1001 and three or more gas sensing units 2002, these gas sensing units 1001 and 2002 can perform the same operations as those described below.

[0252] FIG. 28A is a diagram showing an example of the operation of the gas sensing units 1001 and 2002 when the operation of the hydrogen pump 26 is required to measure the concentration of poisoning impurities in hydrogen.

[0253] When the gas sensing unit 1001-1 is performing a refresh operation, the gas sensing unit 1001-2 measures the concentration of the poisoning impurity.When the gas sensing unit 1001-2 is performing a refresh operation, the gas sensing unit 1001-1 measures the concentration of the poisoning impurity.

[0254] When the gas sensing unit 2002-1 is refreshed, the gas sensing unit 2002-2 measures the concentration of the poisoning impurity.When the gas sensing unit 2002-2 is refreshed, the gas sensing unit 2002-1 measures the concentration of the poisoning impurity.

[0255] FIG. 28B is a diagram showing an example of the operation of the gas sensing units 1001 and 2002 when the operation of the hydrogen pump 26 is not required to measure the concentration of poisoning impurities in hydrogen.

[0256] The gas sensing units 1001-1 and 1001-2 constantly apply a positive voltage to the second electrode 26c of the hydrogen pump 26 relative to the first electrode 26a, selectively drawing in hydrogen from the target atmosphere to dilute the poisonous impurities (dilution mode). This prevents the sensor FET 21 from being poisoned by the poisonous impurities. The concentration of the poisonous impurities is detected using the gas sensing units 2002-1 and 2002-2.

[0257] When the gas sensing unit 2002-1 is refreshed, the gas sensing unit 2002-2 detects the concentration of the poisoning impurity.When the gas sensing unit 2002-2 is refreshed, the gas sensing unit 2002-1 detects the concentration of the poisoning impurity. [Example]

[0258] A gas sensor according to a third embodiment of the present invention will be described.

[0259] The gas sensors according to Examples 1 and 2 are provided with a hydrogen pump 26 as an electrochemical pump provided on a part of the wall surface of the chamber 25, and detect and refresh the poisonous impurities in the hydrogen by concentrating and diluting the poisonous impurities in the hydrogen using the hydrogen pump 26.

[0260] The gas sensor according to this embodiment can concentrate and measure impurities in gases other than hydrogen by using a separation membrane for other gases instead of the hydrogen pump 26.

[0261] First, a gas sensor will be described that includes an oxygen gas separation membrane, i.e., an oxygen pump, instead of the hydrogen pump 26. This gas sensor can be used to measure impurities in oxygen.

[0262] 29A and 29B are diagrams showing an example of the configuration of a gas sensing unit 1001 of a gas sensor including an oxygen pump 36. The chamber 25 is made of laminated ceramic. Figures 29A and 29B correspond to Figures 16A and 16B, respectively.

[0263] A gas sensor chip 20 including a sensor FET 21 that responds to oxygen and impurities in oxygen is installed inside a chamber 25 made of laminated ceramic.

[0264] The oxygen pump 36 is a type of electrochemical pump and a gas permeable portion that selectively allows oxygen to permeate. The oxygen pump 36 allows oxygen to permeate from the inside to the outside of the chamber 25 or from the outside to the inside of the chamber 25. The oxygen pump 36 has a structure similar to that of the hydrogen pump 26, but includes an oxygen ion conductor layer 36b instead of the proton conductor layer 26b of the hydrogen pump 26. The oxygen ion conductor layer 36b can be made of yttria-stabilized zirconia (YSZ), whose crystal grain size is not as small as nanometers.

[0265] 29A, oxygen in the space 40 inside the chamber 25 is selectively released into the detection target atmosphere by the oxygen pump 36, concentrating impurities (e.g., hydrogen) in the space 40 inside the chamber 25. Since the impurity concentration in the space 40 inside the chamber 25 is higher than that in the detection target atmosphere, the sensor FET 21 included in the gas sensor chip 20 can more easily detect the impurities. This phenomenon is similar to the case of the concentration of impurities in hydrogen shown in FIG. 16A.

[0266] However, a positive voltage is applied to the second electrode 26c of the oxygen pump 36 relative to the first electrode 26a. That is, the polarity of the voltage is opposite to that in the example shown in Fig. 16A. This is because protons have a positive charge, whereas oxygen ions have a negative charge.

[0267] 29B, oxygen is selectively introduced from the target atmosphere into the space 40 inside the chamber 25 by the oxygen pump 36, thereby diluting impurities (e.g., hydrogen) in the space 40 inside the chamber 25. Since the impurity concentration in the space 40 inside the chamber 25 is reduced compared to the target atmosphere, the sensor FET 21 included in the gas sensor chip 20 can be easily refreshed. This phenomenon is similar to the case of dilution of impurities in hydrogen shown in FIG. 16B.

[0268] However, a positive voltage is applied to the first electrode 26a of the oxygen pump 36 relative to the second electrode 26c. In other words, the polarity of the voltage is opposite to that shown in Fig. 16B. This is because protons have a positive charge, whereas oxygen ions have a negative charge.

[0269] Next, we will explain a gas sensor that combines a gas separation membrane and a physical pump instead of the hydrogen pump 26 or oxygen pump 36. This gas sensor can measure the concentration of impurities with high sensitivity in a gas system that contains trace amounts of impurities and is mainly composed of a component that selectively permeates the gas separation membrane, and can also refresh the gas sensor effectively. By providing this gas sensor with a carbon dioxide separation membrane, for example, i.e., a membrane that selectively permeates carbon dioxide, as the gas separation membrane, it can measure impurities in carbon dioxide with high sensitivity.

[0270] Figures 30A and 30B are diagrams showing an example of the configuration of a gas sensing portion 1001 of a gas sensor including a gas separation membrane 37 and a physical pump 38. Figures 30A and 30B correspond to Figures 29A and 29B, respectively.

[0271] Gas separation membrane 37 is provided inside chamber 25 and is a separation membrane that selectively allows permeation of gas of main component XA (e.g., carbon dioxide) contained in the detection target atmosphere. In addition to main component XA, the detection target atmosphere also contains impurity XB (e.g., oxygen).

[0272] Physical pump 38 is a mechanical pump made up of machines, and is provided on the wall surface of chamber 25. Physical pump 38 creates a pressure difference on both sides of gas separation membrane 37, and transports gas of the detection target atmosphere from the inside to the outside of chamber 25, or from the outside to the inside of chamber 25. Physical pump 38 releases gas (gas of main component XA) that has permeated gas separation membrane 37 from the inside of chamber 25 to the outside of chamber 25. Physical pump 38 also introduces gas of the detection target atmosphere from the outside of chamber 25 into the inside of chamber 25. Gas separation membrane 37 selectively allows gas of main component XA to permeate out of the gas of the detection target atmosphere introduced into chamber 25 by physical pump 38.

[0273] 30A, a physical pump 38 and a gas separation membrane 37 selectively release a main component XA of a gas present in the space 40 inside the chamber 25 into the detection target atmosphere outside the chamber 25, concentrating an impurity XB of the gas present in the space 40 inside the chamber 25. This increases the impurity concentration in the space 40 inside the chamber 25 compared to the detection target atmosphere outside the chamber 25, making it easier for the sensor FET 21 included in the gas sensor chip 20 to detect the impurity. This phenomenon is similar to the case of the concentration of impurities in oxygen shown in FIG. 29A.

[0274] 30B, a physical pump 38 and a gas separation membrane 37 are used to selectively introduce a main component XA of the gas present in the space 40 inside the chamber 25 from the atmosphere to be detected outside the chamber 25, thereby diluting the impurity XB of the gas present in the space 40 inside the chamber 25. This reduces the impurity concentration in the space 40 inside the chamber 25 compared to the atmosphere to be detected outside the chamber 25, making it easier to refresh the sensor FET 21 included in the gas sensor chip 20. This phenomenon is similar to the case of dilution of impurities in hydrogen shown in FIG. 29B.

[0275] The present invention is not limited to the above-described embodiments, and various modifications are possible. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to embodiments including all of the described configurations. Furthermore, it is possible to replace part of the configuration of one embodiment with the configuration of another embodiment. It is also possible to add the configuration of another embodiment to the configuration of one embodiment. It is also possible to delete part of the configuration of each embodiment, or to add or replace other configurations. [Explanation of symbols]

[0276] 1...semiconductor substrate, 2...well, 3...source diffusion layer, 4...drain diffusion layer, 5...gate insulating film, 6...metal oxide layer, 7...catalytic metal gate layer, 8...interlayer insulating film, 12...well, 13...source diffusion layer, 14...drain diffusion layer, 15...gate insulating film, 16...metal oxide layer, 17...catalytic metal gate layer, 18...interlayer insulating film, 20...gas sensor chip, 21...sensor FET, 22...reference FET, 23...thermometer section, 24...heater A, 25...chamber, 26...hydrogen pump, 26a...first electrode, 26b...proton conductor layer, 26c...second electrode, 27...heater B, 28...wiring, 29...diode, 30...gas flow path, 31...heater C, 36...oxygen pump, 36b...oxygen ion conductor layer, 37...gas separation membrane, 38... Physical pump, 40...internal space of chamber, 50...piping, 61...sensor capacitor, 62...reference capacitor, 71...sensor diode, 72...reference diode, 106...metal oxide layer, 107...catalytic metal gate layer, 108...interlayer insulating film, 116...metal oxide layer, 117...catalytic metal gate layer, 118...interlayer insulating film, 1000...I / O section, 1001, 1001-1 to 1001-n...gas sensing section, 1002...gas concentration estimation section, 1003...control section, 1004...current detection section, 1005...temperature detection section, 1006...hydrogen pump current detection section, 1007...power supply, 1008...circuit section, 1009...parameter recording section, 2002, 2002-1 to 2002-m...gas sensing section, 2008...system control section.

Claims

1. a gas sensor element for measuring the concentration of impurities contained in a detection target atmosphere; a chamber having the gas sensor element therein and a gas flow path connecting an internal space with an external space; an electrochemical pump provided on a wall surface of the chamber, which selectively allows a gas of a main component contained in the detection target atmosphere to permeate from the inside of the chamber to the outside or from the outside of the chamber to the inside of the chamber; a first heater for heating the gas sensor element to control the temperature of the gas sensor element; a second heater for heating the electrochemical pump to control the temperature of the electrochemical pump; Equipped with The electrochemical pump comprises: a first electrode located inside the chamber and exposed to the interior space of the chamber; a second electrode located outside the chamber and exposed to the space outside the chamber; an ion conductor layer located between the first electrode and the second electrode and in contact with the first electrode and the second electrode; Equipped with the potential of the first electrode is set to a positive potential or a negative potential with respect to the potential of the second electrode, thereby selectively allowing the gas of the main component to permeate; A gas sensor characterized by:

2. when the electrochemical pump is heated by the second heater and the main component gas is permeated from the inside of the chamber to the outside, thereby increasing the concentration of the impurities inside the chamber, the gas sensor element is heated by the first heater and measures the concentration of the impurities; when the electrochemical pump is heated by the second heater and the main component gas is permeated from the outside to the inside of the chamber to reduce the concentration of the impurities inside the chamber, the gas sensor element is heated by the first heater to perform a refresh operation to remove the impurities from the gas sensor element; 2. The gas sensor according to claim 1.

3. a thermometer unit for measuring the temperature of the gas sensor element; The thermometer unit includes a diode.

2. The gas sensor according to claim 1.

4. The chamber is made of a metal or a metal oxide.

2. The gas sensor according to claim 1.

5. The chamber is made of laminated insulating films.

2. The gas sensor according to claim 1.

6. the gas sensor element is located in the internal space of the chamber between the gas flow path and the electrochemical pump in the direction of gas flow in the gas flow path; 2. The gas sensor according to claim 1.

7. The gas sensor element includes a semiconductor substrate made of silicon or silicon carbide.

2. The gas sensor according to claim 1.

8. the gas sensor element is a gas sensor element formed of a field effect transistor; 2. The gas sensor according to claim 1.

9. the gas sensor element is a gas sensor element configured by a MOS capacitor or a diode; 2. The gas sensor according to claim 1.

10. the gas flow path includes a third heater that heats the gas flow path to control the temperature of the gas flow path; 2. The gas sensor according to claim 1.

11. a first gas sensing unit; a second gas sensing unit; Equipped with the first gas sensing unit includes the gas sensor element, the chamber, the electrochemical pump, the first heater, and the second heater; the second gas sensing unit includes the gas sensor element and the first heater; 2. The gas sensor according to claim 1.

12. a plurality of the first gas sensing units; a plurality of the second gas sensing units; Equipped with The gas sensor according to claim 11.

13. the gas sensor element of at least one of the first gas sensing units measures the concentration of the impurities when the gas sensor element of at least another of the first gas sensing units is performing a refresh operation, and performs the refresh operation when the gas sensor element of the other one of the first gas sensing units is measuring the concentration of the impurities; the refresh operation is an operation in which the gas sensor element is heated by the first heater to remove the impurities from the gas sensor element; The gas sensor according to claim 12.

14. a gas sensor element for measuring the concentration of impurities contained in a detection target atmosphere; a chamber having the gas sensor element therein and a gas flow path connecting an internal space with an external space; a gas separation membrane provided inside the chamber and allowing selective permeation of a gas containing a main component contained in the detection target atmosphere; a physical pump provided on a wall surface of the chamber for transporting gas of the detection target atmosphere from the inside to the outside of the chamber or from the outside to the inside of the chamber; a heater for heating the gas sensor element to control the temperature of the gas sensor element; Equipped with the physical pump discharges the main component gas that has permeated the gas separation membrane to the outside of the chamber; the gas separation membrane allows the main component gas of the gas to be detected, which is introduced into the chamber by the physical pump, to pass through; A gas sensor characterized by:

15. when the physical pump discharges the main component gas that has permeated the gas separation membrane to the outside of the chamber, thereby increasing the concentration of the impurities inside the chamber, the gas sensor element is heated by the heater to measure the concentration of the impurities; the physical pump introduces the gas of the target atmosphere from the outside to the inside of the chamber, and the gas separation membrane allows the main component gas of the introduced target atmosphere to permeate, thereby reducing the concentration of the impurities inside the chamber, and the gas sensor element is heated by the heater to perform a refresh operation of removing the impurities from the gas sensor element.

15. The gas sensor according to claim 14.

Images