JP4589376B2 - Gas sensor - Google Patents

Description

  The present invention relates to a gas sensor having a gas-sensitive film, and more particularly to a gas sensor in which the gas-sensitive film is made of an oxide semiconductor and a protective layer having gas permeability is formed on the gas-sensitive film.

  2. Description of the Related Art Conventionally, a gas sensor that includes a gas-sensitive film whose electrical characteristics change depending on the concentration change of a gas to be detected and detects the presence or concentration of the gas to be detected is known. In some gas sensors, the gas sensitive film is made of an oxide semiconductor such as tin oxide, and a protective layer having gas permeability is formed on the gas sensitive film. For example, Patent Literature 1 and Patent Literature 2 disclose such a gas sensor.

  When such a gas sensor comes into contact with a poisoning gas component such as silicon, the gas-sensitive film made of an oxide semiconductor may be poisoned and the characteristics of the gas sensor may be deteriorated. May not be possible. In order to solve such a problem, there is a case in which a technique for suppressing characteristic changes during actual use is used by changing the characteristics of the gas sensor to some extent by using an expected poisoning substance. However, since there are various situations in the usage environment of the gas sensor, it is difficult to prescribe the degree of poisoning to be performed in advance, and the effect is only to suppress a large characteristic change at the initial stage of poisoning. It was.

  On the other hand, in the gas sensor of Patent Document 1, it is proposed to form a filter layer (protective layer) made of a mixture of tin oxide and amorphous alumina so as to cover a gas-sensitive film of an oxide semiconductor made of tin oxide. (Refer to the claims of Patent Document 1). It is described that by providing such a filter layer, durability against poisoning can be enhanced, and excellent effects can be obtained with respect to stability over time, shortening of the initial stabilization time, and the like.

  On the other hand, Patent Document 2 describes a gas sensor in which a coating layer (protective layer) made of tungsten oxide, titanium oxide, zinc oxide, or the like is provided on a gas-sensitive film of an oxide semiconductor made of tin oxide (quotation). (Refer to claim 2 of literature 2). Such a gas sensor can increase the sensitivity to various odors to be detected and reduce the sensitivity to odorless gas, thereby improving the S / N ratio and accurately measuring odors at a lower concentration. Is described.

Japanese Patent Laid-Open No. 9-166567 Japanese Patent No. 3171745

  However, neither the gas sensor described in Patent Document 1 nor the gas sensor described in Patent Document 2 has sufficient poisoning resistance against poisonous substances such as silicon, and suppresses a change in gas sensitivity over time over a long period of time. I could not.

  The present invention has been made in view of the present situation, and provides a gas sensor that has high poisoning resistance to poisoning substances such as silicon and can suppress a change in gas sensitivity over time over a long period of time. Objective.

The solution is a gas sensor comprising a gas-sensitive film made of an oxide semiconductor and a gas-permeable protective layer formed on the gas-sensitive film, wherein the protective layer has an average particle diameter of 500 nm. Ri Na consists of the following oxide particles, gas sensor occupying the volume of the whole pores contained in the protective layer, the proportion of pores having a pore diameter below the mean particle size or less is formed by more than 50% It is.

As is clear from the description in Patent Documents 1 and 2 described above, conventionally, by examining what kind of material is used to form the protective layer, the poisoning resistance and temporal stability of the gas sensor can be improved. It was just a technical idea to improve. However, by simply selecting the material forming the protective layer, sufficient poisoning resistance cannot be obtained, and the change over time in gas sensitivity cannot be suppressed over a long period of time.
On the other hand, in the present invention, the protective layer is formed from oxide particles, and the average particle diameter is further taken into consideration. By specifying the average particle diameter of the oxide particles, it is possible to obtain a protective layer in which the oxide particles are densely packed while having gas permeability. By providing such a protective layer on the gas-sensitive film, it is possible to enhance the trapping effect of the poisoning substance such as silicon without impairing the gas permeability of the gas to be detected, and the poisoning substance permeates and senses. The contact with the gas film can be suppressed. Therefore, the poisoning resistance against the poisoning substance can be increased, and the change in gas sensitivity over time can be reduced.
Such a protective layer having many small pores is densely filled with oxide particles while having gas permeability. For this reason, it is possible to improve the trapping effect of the poisoning substance without inhibiting the gas permeability of the gas to be detected, and more effectively suppress the poisoning substance from contacting the gas sensitive film. The resistance can be further increased, and the change in gas sensitivity over time can be further reduced.

In order to effectively trap poisonous substances such as silicon in the protective layer, the oxide particles having an average particle diameter of 150 nm or less, more preferably 100 nm or less, because the specific surface area of the oxide particles is increased. It is preferable to form a protective layer using
Further, if the gas flow rate is excessively limited, the gas detection responsiveness may be lowered. Therefore, it is preferable to form the protective layer using oxide particles having an average particle diameter of 5 nm or more.

  “Gas sensors” include reducing gases such as carbon monoxide gas, hydrocarbon gas (LPG, city gas, natural gas, methane gas, halogenated hydrocarbon gas, etc.), alcohol gas, aldehyde gas, hydrogen gas, etc. And those capable of detecting an oxidizing gas such as nitrogen dioxide. Further, examples of the “gas sensor” include those capable of detecting a strong odor gas such as ammonia, hydrogen sulfide gas, and ethyl acetate.

  The “protective layer” may be composed of oxide particles having an average particle diameter of 500 nm or less, and may be composed of one layer or a plurality of layers. The “protective layer” may be composed of one type of oxide particles, or may be composed of a plurality of types of oxide particles having different average particle diameters or different materials. Furthermore, when there are a plurality of types of oxide particles, the “protective layer” may be one in which these particles are uniformly mixed and distributed, for example, a portion on the gas sensitive film side of the protective layer. A plurality of types of oxide particles may be distributed in different states depending on the location, for example, different types of oxide particles may be distributed in the surface side portion.

  In addition, the “protective layer” only needs to be mainly composed of oxide particles having an average particle diameter of 500 nm or less, and may be composed only of such oxide particles. For example, the catalyst particles may be slightly mixed. However, the “protective layer” preferably does not contain catalyst particles therein, that is, does not contain catalyst particles that are not attached to the gas-sensitive film. If catalyst particles are contained inside the protective layer, the catalytic reaction of the gas to be detected can also occur inside the protective layer. As a result, the catalytic reaction at the catalyst particles on the gas-sensitive film is reduced, resulting in a decrease in gas sensitivity. Because there is a fear.

Examples of the “oxide particles” constituting the protective layer include titanium oxide particles, tungsten oxide particles, alumina particles, silica particles, and tin oxide particles.
The “average particle size” of the oxide particles forming the protective layer is, for example, at least using an apparatus such as Hitachi HD-2000 after performing FIB processing using an apparatus such as Hitachi FB-2100. This can be determined by TEM observation in a visual field where 30 oxide particles are observed.

Examples of the “gas sensitive film” include those composed of an oxide semiconductor such as tin oxide and indium oxide. Further, the gas sensitive film may be a thin film or a thick film. In addition, catalyst particles may be combined with the gas-sensitive film, and as a specific form thereof, a form in which the catalyst particles are dispersed on the gas-sensitive film may be adopted, or the catalyst particles may be included in the gas-sensitive film. You may take the form which distributed.
The distribution of “pore diameter” can be determined by a mercury intrusion method (JIS R1655) or a nitrogen gas adsorption method. When the pore size distribution is measured by the mercury intrusion method, for example, the pore size distribution can be obtained by using an autopore IV9510 manufactured by Shimadzu Corporation and a pressure input of 0.5 psia to 36000 psia. In addition, when measuring the pore size distribution by the nitrogen gas adsorption method, for example, the pore size distribution can be obtained by measuring an adsorption curve using a high-precision fully automatic gas adsorption device manufactured by Nippon Bell Co., Ltd. it can.
Further, the mercury intrusion method is more suitable for measuring large pore diameters, and the nitrogen gas adsorption method is more suitable for measuring small pore diameters. For example, while measuring pore diameters of 100 nm or more by mercury intrusion method, The measuring method can be appropriately changed according to the size of the pore diameter, such as measuring the pore diameter of less than 100 nm by a nitrogen gas adsorption method.
Furthermore, in the gas sensor according to any one of the above, the ratio of pores having a pore diameter of three times or more of the average particle diameter in the volume of the entire pores included in the protective layer, It is preferable that the gas sensor be 10% or less (including 0%).
Such a protective layer with few large pores is densely filled with oxide particles while having gas permeability. For this reason, it is possible to improve the trapping effect of the poisoning substance without inhibiting the gas permeability of the gas to be detected, and more effectively suppress the poisoning substance from contacting the gas sensitive film. The poisoning resistance to the substance can be further increased, and the change in gas sensitivity over time can be further reduced.
The gas sensor further comprises a gas-sensitive film made of an oxide semiconductor and a gas-permeable protective layer formed on the gas-sensitive film, wherein the protective layer has an average particle diameter of 500 nm or less. 10% or less (including 0%) of the pores having a pore diameter of 3 times or more of the average particle diameter in the total pore volume contained in the protective layer. It is preferable to use a gas sensor.
In the present invention, the protective layer is formed from oxide particles, and the average particle diameter is further taken into consideration. By specifying the average particle diameter of the oxide particles, it is possible to obtain a protective layer in which the oxide particles are densely packed while having gas permeability. By providing such a protective layer on the gas-sensitive film, it is possible to enhance the trapping effect of the poisoning substance such as silicon without impairing the gas permeability of the gas to be detected, and the poisoning substance permeates and senses. The contact with the gas film can be suppressed. Therefore, the poisoning resistance against the poisoning substance can be increased, and the change in gas sensitivity over time can be reduced.
In addition, the protective layer having few large pores is densely filled with oxide particles while having gas permeability. For this reason, it is possible to improve the trapping effect of the poisoning substance without inhibiting the gas permeability of the gas to be detected, and more effectively suppress the poisoning substance from contacting the gas sensitive film. The poisoning resistance to the substance can be further increased, and the change in gas sensitivity over time can be further reduced.

  Further, in the gas sensor described above, the protective layer is a gas sensor in which 20 ≦ TH / D ≦ 500 when the average particle diameter is D (nm) and the thickness of the protective layer is TH (nm). And good.

By making the relationship between the average particle diameter D and the thickness TH of the protective layer TH / D ≧ 20, the capturing effect of the poisoning substance is further improved, and the poisoning substance comes into contact with the gas sensitive film more effectively. Therefore, the resistance to poisoning can be further increased, and the change in gas sensitivity over time can be further reduced.
On the other hand, if the TH / D value exceeds 500, gas permeability may be hindered. Therefore, it is preferable to satisfy TH / D ≦ 500. In order to obtain good gas detection response, It is preferable that TH / D ≦ 300.
The thickness TH of the protective layer refers to the dimension from the surface of the gas sensitive film to the surface of the protective layer.

  Furthermore, the gas sensor according to any one of the above, comprising catalyst particles dispersed on the gas sensitive film, wherein the protective layer is a gas sensor formed on the gas sensitive film and the catalyst particles. good.

  Thus, by dispersing the catalyst particles on the gas-sensitive film, the gas that has permeated the protective layer effectively contacts both the catalyst particles and the gas-sensitive film, so that the gas sensitivity can be increased.

  Furthermore, in the gas sensor described above, the protective layer may be a gas sensor having a thickness that does not expose the catalyst particles.

  Since the protective layer has a film thickness at which the catalyst particles are not exposed, the effect of capturing the poisonous substance of the protective layer can be further improved, and the poisoning substance can be more effectively suppressed from coming into contact with the gas-sensitive film. . Therefore, the poisoning resistance against the poisoning substance can be further increased, and the change in gas sensitivity with time can be further reduced.

  Furthermore, the gas sensor according to any one of the above, wherein the protective layer occupies the entire volume of the pores contained therein, and the proportion of pores having a pore diameter of 200 nm or more when measured by a mercury intrusion method However, the gas sensor is preferably 5% or less (including 0%).

  Such a protective layer with few large pores is densely filled with oxide particles while having gas permeability. For this reason, it is possible to improve the trapping effect of the poisoning substance without inhibiting the gas permeability of the gas to be detected, and more effectively suppress the poisoning substance from contacting the gas sensitive film. The poisoning resistance to the substance can be further increased, and the change in gas sensitivity over time can be further reduced.

  Furthermore, in the gas sensor according to any one of the above, the protective layer is configured such that pores having an opening diameter of 10 times or more the average particle diameter are not observed when viewed from the surface. A gas sensor is recommended.

  The protective layer in which such large pores are not observed is densely filled with oxide particles while being in a gas permeable state. For this reason, it is possible to improve the trapping effect of the poisoning substance without inhibiting the gas permeability of the gas to be detected, and more effectively suppress the poisoning substance from contacting the gas sensitive film. The poisoning resistance to the substance can be further increased, and the change in gas sensitivity over time can be further reduced.

  The gas sensor according to any one of the above, wherein the protective layer is formed by depositing the oxide particles on the gas-sensitive film and heat-treating the oxide particles in a temperature range in which the oxide particles do not sinter. And good.

  Such a protective layer is heat-treated in a temperature range in which the oxide particles do not sinter, so that the oxide particles are physically bonded to each other and have gas permeability. Is closely packed. For this reason, it is possible to improve the trapping effect of the poisoning substance without inhibiting the gas permeability of the gas to be detected, and more effectively suppress the poisoning substance from contacting the gas sensitive film. The poisoning resistance to the substance can be further increased, and the change in gas sensitivity over time can be further reduced.

  Furthermore, the gas sensor according to any one of the above may be a gas sensor using titanium oxide particles as the oxide particles.

  By using metal oxide particles, particularly titanium oxide particles as oxide particles, the ability to capture poisonous substances such as silicon is higher than when other oxide particles are used. It can suppress that a toxic substance contacts a gas-sensitive film. For this reason, the poisoning tolerance with respect to a poisoning substance can be made still higher, and the change with time of gas sensitivity can further be reduced.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. 1 to 4 show a gas sensor 100 according to the present embodiment. 2 corresponds to a partially enlarged view showing a part of the AA cross section of the gas sensor 100 shown in FIG. 1, and FIG. 3 is a partially enlarged view showing a part of the BB cross section of the gas sensor 100 shown in FIG. It corresponds to. FIG. 4 is a partially enlarged cross-sectional view of the gas sensor 100 in the vicinity of the gas sensitive film 131, the catalyst particles 133, and the protective layer 140. In FIG. 2, the catalyst particles 133 are shown in a layered form for convenience, but actually, the catalyst particles 133 are dispersed on the gas-sensitive film 131 as shown in FIG. 4.

  This gas sensor 100 utilizes the fact that an electric characteristic (specifically, an electric resistance value) of a gas-sensitive film 131, which will be described later, changes due to the presence of an oxidizing gas such as nitrogen dioxide. Detection (contamination detection). This gas sensor 100 has a rectangular shape in plan view as shown in FIG. 1, and as shown in FIGS. 2 to 4, a front-side insulating coating 110 and a back-side are formed on the front surface 101b and the rear surface 101c of a 400 μm thick silicon substrate 101 An insulating coating 120 is formed, and a gas-sensitive film 131, catalyst particles 133, and a protective layer 140 are further formed on the surface-side insulating coating 110.

  Among these, the front-side insulating coating 110 includes four insulating layers 111, 112, 113, and 114 formed on the front surface 101 b of the silicon substrate 101, and one back-side insulating coating 120 serves as the back surface of the silicon substrate 101. It consists of two insulating layers 121 and 122 formed on 101c. The insulating layer 111 and the insulating layer 121 are made of silicon oxide, and are formed on the front surface 101b and the back surface 101c of the silicon substrate 101, respectively. The insulating layer 112 and the insulating layer 122 are made of silicon nitride, and are formed on the insulating layer 111 and the insulating layer 121, respectively. The insulating layer 113 is made of silicon oxide and is formed on the insulating layer 112. Further, the insulating layer 114 is made of silicon nitride and is formed on the insulating layer 113.

  A heating resistor 151 is embedded in the insulating layer 113 of the surface-side insulating coating 110. The heat generating resistor 151 generates heat when energized, and heats the gas sensitive film 131 made of an oxide semiconductor to 200 ° C. or more to activate the gas sensitive film 131. In addition, a pair of lead portions 153 that are electrically connected to the heating resistor 151 and energize the heating resistor 151 are embedded in the insulating layer 113. As shown in FIGS. 1 and 3, the lead portion 153 is electrically connected to a pair of heating resistor contact portions 160 for connection to an external circuit at the end thereof. The heating resistor contact portion 160 has a rectangular shape in plan view. The heating resistor 151 and the lead portion 153 have a two-layer structure including a platinum layer and a tantalum layer. Further, the heating resistor contact portion 160 has a structure in which a heating element contact pad 163 made of gold is formed on an extraction electrode 161 made of a titanium layer and a platinum layer (see FIG. 3).

  Further, as shown in FIG. 2, the gas sensor 100 has a recess 103 in which the back-side insulating coating 120 and the silicon substrate 101 are partially removed in a concave shape, and the insulating layer 111 of the front-side insulating coating 110 is exposed on the bottom surface. Is formed. The recess 103 is provided at a position corresponding to the heating resistor 151 described above.

  As shown in FIG. 2, an electrode 155 is formed at a position corresponding to the heating resistor 151 on the surface-side insulating coating 110 (on the insulating layer 114). A pair of lead portions 157 (see FIG. 3) for electrically connecting to the electrode 155 and energizing the electrode 155 are formed on the surface-side insulating coating 110 (on the insulating layer 114). Similar to the lead electrode 161 of the heating resistor contact portion 160, the electrode 155 and the lead portion 157 are composed of a titanium layer formed on the insulating layer 114 and a platinum layer formed thereon. Further, as shown in FIGS. 1 and 3, the lead portion 157 is electrically connected to a pair of detector contact pads 170 for connection to an external circuit at the end thereof. The detector contact pad 170 is made of gold and has a rectangular shape in plan view.

  As shown in FIG. 2, the gas-sensitive film 131 is made of tin oxide, which is an oxide semiconductor, and is formed so as to cover the electrode 155 on the surface-side insulating coating 110 (on the insulating layer 114). On the gas sensitive film 131, gold catalyst particles 133 are arranged in a dispersed state (see also FIG. 4).

  Further, a protective layer 140 is formed on the gas sensitive film 131 and the catalyst particles 133 as a whole. The protective layer 140 is configured by densely filling oxide particles having an average particle diameter of 500 nm or less, specifically, titanium oxide particles 141 having an average particle diameter of 20 nm, and gas along the thickness direction. It is configured to have transparency. Further, the protective layer 140 has a form satisfying 20 ≦ TH / D ≦ 500, where the average particle diameter of the titanium oxide particles 141 is D (nm) and the thickness of the protective layer 140 is TH (nm). In this case, since the thickness TH of the protective layer 140 is 1000 nm, TH / D = 50. The thickness TH is also a thickness at which the catalyst particles 133 are not exposed to the outside.

As will be described later, the protective layer 140 further has a ratio of pores having a pore diameter equal to or smaller than the average particle diameter D to 50% of the total pore volume (total pore volume) contained in the protective layer 140. % Or more. Specifically, the average particle diameter D is 20 nm as described above, and the ratio of pores having a pore diameter of 20 nm or less is about 88%.
Further, the protective layer 140 has a ratio of pores having a pore diameter that is three times or more of the average particle diameter D in the total pore volume (total pore volume) contained in the protective layer 140 to 10% or less. ing. Specifically, the above-mentioned ratio of pores having a pore diameter that is 3 times (60 nm) or more of the average particle diameter D = 20 nm is about 1%.
In addition, the protective layer 140 has a ratio of pores having a pore diameter of 200 nm or more when measured by a mercury intrusion method to a total volume of pores (total pore volume) contained therein of 5% or less. (Including 0%). Specifically, there are no pores having a pore diameter of 200 nm or more as measured by mercury porosimetry.

Further, the protective layer 140 is configured such that pores having an opening diameter of 10 times or more the average particle diameter D are not observed when viewed from the surface. Specifically, when an SEM image of the surface is viewed, pores having an opening diameter of 10 times (200 nm) or more of the average particle diameter D = 20 nm are not observed at all.
The protective layer 140 is formed by depositing titanium oxide particles 141 on the gas-sensitive film 131 and the catalyst particles 133 and performing a heat treatment in a temperature range in which the titanium oxide particles 141 do not sinter.

  As described above, in the gas sensor 100 of the present embodiment, the protective layer 140 is composed of the oxide particles 141 having an average particle diameter of 500 nm or less. It is possible to effectively capture the poisonous substance and prevent the poisonous substance from passing through the protective layer 140 and coming into contact with the gas sensitive film 131. Therefore, the poisoning resistance against the poisoning substance can be increased, and the change in gas sensitivity over time can be reduced.

  Furthermore, in the present embodiment, as described above, the protective layer 140 has a relationship between the average particle diameter D (nm) of the titanium oxide particles 141 and the thickness TH (nm) of the protective layer 140 as TH / D = 50. Therefore, the ability to capture poisonous substances such as silicon can be improved, the poisoning resistance can be further increased, and the change in gas sensitivity over time can be further reduced.

  Further, the protective layer 140 has a ratio of pores having a pore diameter of an average particle diameter D = 20 nm or less to a volume of the whole pores included therein of about 88%. Further, the protective layer 140 has a ratio of pores having a pore diameter that is three times the average particle diameter D = 20 nm (60 nm) or more in the entire pore volume contained in the protective layer 140 at about 1%. Further, the protective layer 140 has no pores having a pore diameter of 200 nm or more as measured by the mercury intrusion method. Further, when the surface of the protective layer 140 is observed with an SEM, pores having an opening diameter of 10 times (200 nm) or more of the average particle diameter D = 20 nm are not observed at all. Further, as will be described later, the protective layer 140 is heat-treated in a temperature range in which titanium oxide particles 141 are deposited on the gas-sensitive film 131 and the catalyst particles 133 and the titanium oxide particles 141 are not sintered together. Such a protective layer 140 is densely filled with oxide particles 141 while having gas permeability. For this reason, it is possible to improve the trapping effect of the poisoning substance without inhibiting the gas permeability of the gas to be detected, and more effectively suppress the poisoning substance from contacting the gas sensitive film. The poisoning resistance to the substance can be further increased, and the change in gas sensitivity over time can be further reduced.

  Further, since the titanium oxide particles 141 are used as the oxide particles constituting the protective layer 140, the poisoning resistance can be further increased, and the change in gas sensitivity over time can be further reduced. Further, since the protective layer 140 has a film thickness (thickness TH) at which the catalyst particles 133 are not exposed, the poisoning resistance can be further increased and the change in the gas sensitivity over time can be further reduced.

Next, a method for manufacturing the gas sensor 100 will be described.
First, a silicon substrate 101 having a thickness of 400 μm is prepared and immersed in a cleaning solution to perform a cleaning process.
Next, the cleaned silicon substrate 101 is placed in a heat treatment furnace, and a silicon oxide layer (insulating layers 111 and 121) having a thickness of 100 nm is formed on the entire surface 101b and back surface 101c of the silicon substrate 101 by thermal oxidation.
Thereafter, a silicon nitride layer (insulating layer 112) having a thickness of 200 nm is formed on the surface 101b side (on the insulating layer 111) of the silicon substrate 101 using SiH ₂ Cl ₂ and NH ₃ as source gases by LP-CVD. Similarly, a silicon nitride layer (insulating layer 122) having a thickness of 100 nm is formed on the back surface 101c side (on the insulating layer 121) of the silicon substrate 101. Thereby, the back surface side insulating film 120 which consists of the insulating layer 11 and the insulating layer 122 is formed.

Next, a silicon oxide layer having a thickness of 100 nm (an insulating layer to be a part of the lower side of the insulating layer 113) is formed over the insulating layer 112 on the surface 101b side using TEOS and O ₂ as source gases by plasma CVD.
Thereafter, using a DC sputtering apparatus, a tantalum layer having a thickness of 20 nm is formed on the insulating layer 113, and a platinum layer having a thickness of 220 nm is further formed thereon. After this sputtering, a resist is formed by photolithography, patterning is performed, and a wet etching process is performed to form the heating resistor 151 and the lead portion 153.

Then, a silicon oxide layer having a thickness of 100 nm is formed on the silicon oxide layer (a part of the lower side of the insulating layer 113), the heating resistor 151, and the seed portion 153 using TEOS and O ₂ as source gases by plasma CVD. An insulating layer that is a part of the upper side of the insulating layer 113 is formed. Thus, the insulating layer 113 having a thickness of 200 nm is formed, and the heating resistor 151 and the lead portion 153 are embedded therein.
Thereafter, a silicon nitride layer (insulating layer 114) having a thickness of 200 nm is formed on the insulating layer 113 by LP-CVD using SiH ₂ Cl ₂ and NH ₃ as source gases. As a result, the surface-side insulating coating 110 composed of the insulating layer 111, the insulating layer 112, the insulating layer 113, and the insulating layer 114 is formed.

Next, a resist is formed by photolithography, patterning is performed, the insulating layer 113 and the insulating layer 114 are etched by a dry etching method, holes are formed in portions where the heating resistor contact portions 160 are formed, and lead portions 153 are formed. To expose a part of
Next, using a DC sputtering apparatus, a titanium layer having a thickness of 20 nm is formed on the insulating layer 114, and a platinum layer having a thickness of 40 nm is further formed thereon. After this sputtering, a resist is formed by photolithography and patterning is performed, and electrodes 151, lead portions 157, and extraction electrodes 161 having a predetermined pattern are formed by wet etching.
Then, using a DC sputtering apparatus, a gold layer having a thickness of 400 nm is formed on the insulating layer 114 side of the substrate. After this sputtering, a resist is formed by photolithography, patterning is performed, and wet etching is performed to form the heating element contact pad 163 and the sensing element contact pad 170.

  Next, a resist is formed by photolithography, patterning is performed, and the back-side insulating film 120 corresponding to the heating resistor 151 is etched by dry etching. Then, the substrate is immersed in the TMAH solution and anisotropic etching of the silicon substrate 101 is performed, so that the surface-side insulating coating 110 (specifically, the insulating layer 111) corresponding to the heating resistor 151 is formed on the bottom surface. An exposed recess 103 is formed.

Next, a gas-sensitive film 131 made of tin oxide is formed on the insulating layer 114 of the surface-side insulating coating 110 at a position corresponding to the heating resistor 151 and the recess 103 using an RF sputtering apparatus. Specifically, Ar / (Ar + O ₂₎ is used in an RF sputtering apparatus with argon gas and oxygen gas. ) Is 28%, the gas pressure is 2.0 Pa, the power density is 4.34 kW / cm ² , the deposition rate is 4.8 nm / min, and the substrate temperature is 240 ° C. for 42 minutes. The gas sensitive film 131 having a thickness of 200 nm is formed. Note that the target area for this sputtering process is 5 inches × 10 inches (12.7 cm × 25.4 cm).

Next, the catalyst particles 133 are formed on the gas-sensitive film 131 by depositing gold using a DC sputtering apparatus. Specifically, argon gas is introduced into a DC sputtering apparatus, the gas pressure is 2.0 Pa, the power density is 0.62 kW / cm ² , the film formation rate is 11.4 nm / min, and the substrate is not heated. Sputtering is performed for 2 seconds to form catalyst particles 133. The target area for this sputtering process is also 5 inches × 10 inches (12.7 cm × 25.4 cm).
Thereafter, heat treatment was performed at 360 ° C. for 3 hours in an atmosphere having an oxygen concentration of 10 ppm or less using an RF sputtering apparatus. For this heat treatment, a vacuum heat treatment furnace may be used instead of using an RF sputtering apparatus. The oxygen concentration is preferably 0.2 ppm to 5 ppm.

Next, the protective layer 140 was formed on the gas sensitive film 131 and the catalyst particles 133. Specifically, first, the heating resistor contact portion 160 and the detector contact pad 170 are masked. Thereafter, the titanium oxide powder having a desired size (in this embodiment, the average particle diameter D is 20 nm) has a predetermined viscosity obtained by dispersing it in water together with ammonium polyacrylate as a dispersant (in this embodiment, a viscosity of 50 mPa). S) A sol is prepared and spin-coated on the gas-sensitive film 131 and the catalyst particles 133. The protective layer 140 can be adjusted to a desired thickness by adjusting the conditions during spin coating. In this embodiment, the rotational speed during spin coating is 5000 rpm, the substrate temperature is 25 ° C., and the ambient relative humidity. Was adjusted to 40% RH, and spin coating was performed to form a protective layer 140 having a thickness TH of 1000 nm. Then, after natural drying, in a heat treatment furnace, the protective layer 140 is dried by heating in a temperature range where the titanium oxide particles 141 do not sinter (350 ° C. in the present embodiment) for 1 hour, and further the titanium oxide particles 141. To form a protective layer 140 in which the titanium oxide particles 141 are closely packed.
Thereafter, the substrate was cut into a size of 2.6 mm × 2 mm in plan view using a dicing saw to complete the gas sensor 100.

(Example)
In order to verify the effects of the present invention, as gas sensors of Examples 1 to 4 according to the present invention, 4 gas sensors in which the material and the particle diameter of the oxide particles constituting the protective layer 140 in the gas sensor 100 of the above embodiment are changed are shown. Kind prepared. Specifically, as Example 1, a gas sensor 100 having a protective layer 140 formed of titanium oxide particles having an average particle diameter D = 7 nm was prepared. Further, as Example 2, a gas sensor 100 having a protective layer 140 formed from titanium oxide particles having an average particle diameter D = 20 nm was prepared. Further, as Example 3, a gas sensor 100 having a protective layer 140 formed of titanium oxide particles having an average particle diameter D = 100 nm was prepared. Further, as Example 4, a gas sensor 100 having a protective layer 140 formed of tungsten oxide particles having an average particle diameter D = 300 nm was prepared. The average particle diameter D of the oxide particles constituting the protective layer 140 was determined by performing TEM observation using Hitachi HD-2000 after performing FIB processing using Hitachi FB-2100.

  On the other hand, as Comparative Examples 1 to 4, four types of gas sensors were prepared in which the configuration of the protective layer 140 of the gas sensor 100 according to the above embodiment was changed. Specifically, as Comparative Example 1, a gas sensor having a protective layer formed from aluminum oxide particles having an average particle diameter D = 700 nm was prepared. Further, as Comparative Example 2, a gas sensor having a protective layer formed of titanium oxide particles having an average particle diameter D = 1000 nm was prepared. Further, as Comparative Example 3, a gas sensor having a protective layer formed of silicon oxide particles having an average particle diameter D = 3000 nm was prepared. As Comparative Example 4, a gas sensor having no protective layer was prepared.

Next, the gas sensor 100 or the like of Examples 1 to 4 and Comparative Examples 1 to 4, in a situation where silicon as a poisoning substance is present, a durability test was carried out for sensitivity to nitrogen dioxide (NO ₂₎ . Specifically, each gas sensor 100 etc. was electrically connected to the measurement jig using a gold wire. Then, each gas sensor 100 or the like is arranged in the test gas, and the heating resistor 151 is energized so that the temperature of the heating resistor 151 is 250 ° C., and is held for 400 hours to change the sensitivity to nitrogen dioxide gas. Asked. During this durability test, the gas detection voltage was continuously applied to the gas sensitive film 131 and the like. As a test gas, a mixed gas with nitrogen having a relative humidity of 40% RH and an oxygen content of 20.9 Vol% is used as a base gas, this base gas is set at 25 ° C., and hexamethyldisilazane which is organic silicon The one with 3 ppm added was used.

  In this durability test, the sensitivity to nitrogen dioxide gas was determined for each gas sensor 100 and the like before the start of energization, after 100 hours of energization, after 200 hours of energization, and after 400 hours of energization. Here, this gas sensitivity was calculated | required based on the resistance value of the gas sensor 100 measured with the following method. First, the resistance value Ra of the gas sensor 100 is measured in an atmosphere exposed to the base gas set at 25 ° C. Thereafter, the atmosphere around the gas sensor 100 is set to an atmosphere in which 1 ppm of nitrogen dioxide gas as an oxidizing gas is mixed with the base gas, and the resistance value Rg of the gas sensor 100 or the like is measured after 5 seconds. And the ratio (Rg / Ra) of both resistance values was calculated | required and this was made into gas sensitivity Rg / Ra. Note that the sensitivity at which nitrogen dioxide gas can be sufficiently detected is 1.1 or more. The results are shown in Table 1 and further shown graphically in FIG.

  As is clear from Table 1 and FIG. 5, the gas sensors 100 of Examples 1 to 4 have gas sensitivity Rg / E even when energized for 400 hours in a gas containing a poisoning substance (organic silicon). Ra was sufficiently higher than the threshold (1.1), that is, sufficiently high gas sensitivity was maintained. On the other hand, in the gas sensors of Comparative Examples 1 to 4, the gas sensitivity Rg / Ra gradually decreases when energization is performed for 400 hours in a gas containing a poisoning substance (organic silicon), and the gas sensitivity Rg. / Ra was as low as close to the threshold (1.1).

Further, the rate of change in gas sensitivity was also determined from the above durability test. That is, the gas sensitivity change rate Sc was obtained by the following equation, where A is the gas sensitivity before the durability test and B is the gas sensitivity after the predetermined time of the durability test.
Sc = logB / logA
The results are shown in Table 2 and further shown graphically in FIG. In addition, the quality criterion for the gas sensitivity change rate Sc was set to Sc ≧ 0.5.

  As is clear from Table 2 and FIG. 6, the gas sensors 100 of Examples 1 to 4 all have a gas sensitivity change rate even when energized for 400 hours in a gas containing a poisoning substance (organic silicon). Sc was sufficiently higher than the quality criterion (Sc ≧ 0.5), that is, sufficiently high gas sensitivity was maintained. On the other hand, in all of the gas sensors of Comparative Examples 1 to 4, the gas sensitivity change rate Sc is lower than the pass / fail judgment criterion (Sc ≧ 0.5) when the energization is performed for 400 hours or before, that is, the gas sensitivity. Was greatly reduced.

  From the above results, by forming the protective layer 140 using oxide particles having a small average particle diameter D (500 nm or less), the poisoning resistance to silicon (organic silicon) is improved, and the gas sensitivity over time is improved. It can be seen that there is less change.

  Next, gas sensors 100 of other Examples 5 to 7 according to the present invention were prepared. In the gas sensors 100 of Examples 5 to 7, the average particle diameter D (nm) of the titanium oxide particles 141 constituting the protective layer 140 in the gas sensor 100 of the above embodiment is constant (D = 20 nm), and the protective layer for this This is a gas sensor in which the ratio (TH / D) of the thickness TH (nm) is changed. Specifically, as Example 5, a gas sensor 100 having a protective layer 140 of TH / D = 25 was prepared. As Example 6, a gas sensor 100 having a protective layer 140 of TH / D = 50 was prepared. As Example 7, a gas sensor 100 having a protective layer 140 of TH / D = 100 was prepared.

  On the other hand, as Comparative Examples 5 and 6, gas sensors in which the configuration of the protective layer 140 of the gas sensor 100 according to the above embodiment was changed were prepared. Specifically, as Comparative Example 5, a gas sensor having a protective layer of TH / D = 15 was prepared. As Comparative Example 6, a gas sensor having no protective layer was prepared. The gas sensor of Comparative Example 6 is the same as the gas sensor of Comparative Example 4 described above.

  Next, for the gas sensors 100 and the like of Examples 5 to 7 and Comparative Examples 5 and 6, as in Examples 1 to 4 and Comparative Examples 1 to 4 described above, organic silicon as a poisonous substance is present. Then, a durability test for sensitivity to nitrogen dioxide gas was performed. The results are shown in Table 3 and FIG.

  As is apparent from Table 3 and FIG. 7, the gas sensors 100 of Examples 5 to 7 all have gas sensitivities Rg / E even when energized for 400 hours in a gas containing a poisoning substance (organic silicon). Ra was sufficiently higher than the threshold (1.1), that is, sufficiently high gas sensitivity was maintained. On the other hand, in the gas sensors of Comparative Examples 5 and 6, when energization was performed for 400 hours in a gas containing a poisoning substance (organic silicon), the gas sensitivity Rg / Ra gradually decreased, and the gas sensitivity Rg / Ra was as low as close to the threshold (1.1).

  Further, the gas sensitivity change rate Sc was also obtained from the above durability test in the same manner as in Examples 1 to 4 and Comparative Examples 1 to 4 described above. The calculation formula (Sc = logB / logA) and pass / fail judgment criteria (Sc ≧ 0.5) are the same as those in the above-described durability test. The results are shown in Table 4 and FIG.

  As is apparent from Table 3 and FIG. 7, the gas sensors 100 of Examples 5 to 7 have a gas sensitivity change rate Sc that is higher than the pass / fail judgment criterion (Sc ≧ 0.5) even when energized for 400 hours. Sufficiently high, that is, sufficiently high sensitivity was maintained. On the other hand, in both of the gas sensors of Comparative Examples 5 and 6, the gas sensitivity change rate Sc was lower than the pass / fail judgment criterion (Sc ≧ 0.5) when the energization was performed for 400 hours or before that, that is, the gas sensitivity was low. It was greatly reduced.

  From the above results, by forming the protective layer 140 having a large ratio (TH / D) of the thickness TH (nm) of the protective layer 140 to the average particle diameter D (nm) of the titanium oxide particles 141 (20 or more), silicon is formed. It can be seen that the poisoning resistance to (organic silicon) is improved and the change in gas sensitivity with time is reduced.

  Next, another gas sensor 100 of Example 8 according to the present invention was prepared. The gas sensor 100 of Example 8 is a gas sensor in which the particle size of the titanium oxide particles 141 constituting the protective layer 140 in the gas sensor 100 of the above embodiment is changed, and specifically, titanium oxide having an average particle size D = 150 nm. It has a protective layer 140 formed from particles. About the gas sensor 100 of this Example 8, the durability test about the sensitivity with respect to nitrogen dioxide gas is carried out similarly to Examples 1-7 and Comparative Examples 1-6 mentioned above in the condition where the organic silicon as a poisoning substance exists. went. The results are shown in Table 5 and FIG.

  As is apparent from Table 5 and FIG. 9, the gas sensor 100 of Example 8 has a gas sensitivity Rg / Ra of a threshold value (even when energized for 400 hours in a gas containing a poisoning substance (organic silicon). It was much higher than 1.1), that is, it maintained a sufficiently high gas sensitivity.

  In addition, the gas sensitivity change rate Sc was also obtained from the durability test in the same manner as in Examples 1 to 7 and Comparative Examples 1 to 6 described above. The calculation formula (Sc = logB / logA) and pass / fail judgment criteria (Sc ≧ 0.5) are the same as those in the above-described durability test. The results are shown in Table 6 and FIG.

As is apparent from Table 6 and FIG. 9, the gas sensor 100 of Example 8 has a gas sensitivity change rate Sc sufficiently higher than the pass / fail judgment criterion (Sc ≧ 0.5) even when energization is performed for 400 hours. That is, a sufficiently high sensitivity was maintained.
From this result, by forming the protective layer 140 using oxide particles having a small average particle diameter D (150 nm), the resistance to poisoning against silicon (organic silicon) is improved, and the gas sensitivity changes over time. You can see that it can be reduced.

Next, the pore diameter distribution of the pores formed in the protective layer 140 of the gas sensor 100 of Examples 1, 2, 4, and 8 was examined.
Specifically, first, bulk bodies of the protective layer 140 having the same porosity as the protective layer 140 of each of Examples 1, 2, 4, and 8 were prepared. That is, as a bulk body corresponding to Example 1, a bulk body composed of titanium oxide particles having an average particle diameter D = 7 nm was prepared. Moreover, as a bulk body corresponding to Example 2, a bulk body composed of titanium oxide particles having an average particle diameter D = 20 nm was prepared. Moreover, as a bulk body corresponding to Example 4, a bulk body composed of tungsten oxide particles having an average particle diameter D = 300 nm was prepared. Moreover, as a bulk body corresponding to Example 8, a bulk body composed of titanium oxide particles having an average particle diameter D = 150 nm was prepared. Each bulk body has a diameter of 10 mm and a thickness of 15 mm.

  These bulk bodies were manufactured in the above dimensions through the same spin coat process and heat treatment process as the process of manufacturing the protective layer 140 of the gas sensor 100. That is, a sol (viscosity 50 mPa · s) in which the oxide particles 141 were dispersed in water together with a dispersant (ammonium polyacrylate) was prepared and spin-coated on a substrate to a desired thickness. Thereafter, it is naturally dried, and further heated in a heat treatment furnace in a temperature range (350 ° C.) at which the oxide particles 141 do not sinter together to physically bond the oxide particles 141, so that the oxide particles 141 are densely packed. The bulk body filled in was obtained.

  And about each bulk body corresponding to Example 1, 2, while measuring the pore diameter of 100 nm or more by the mercury intrusion method, the pore diameter of less than 100 nm is measured by the nitrogen gas adsorption method, and pore diameter distribution is calculated | required. It was. The reason is that the measurement of small pore diameters of less than 100 nm can be measured more accurately by the nitrogen gas adsorption method than by the mercury intrusion method. On the other hand, for each bulk body corresponding to Examples 4 and 8, the pore diameter was measured only by mercury porosimetry to obtain the pore diameter distribution.

  In the mercury intrusion method, the pore size distribution was obtained by measuring with a pressure input of 0.5 psia to 36000 psia using Shimadzu Autopore IV9510 in accordance with JIS R1655. In the nitrogen gas adsorption method, the pore diameter distribution was determined by measuring an adsorption curve using a high-precision fully automatic gas adsorption device manufactured by Nippon Bell Co., Ltd. The results are shown in FIG. 10, and further integrated for each graph of FIG. 10 to obtain the cumulative ratio on the larger pore diameter side, as shown in FIG.

In the bulk body corresponding to Example 1 (average particle diameter D = 7 nm), as can be seen from FIG. 11, the proportion of pores having an average particle diameter D = 7 nm or less in the total pore volume is as follows. 50% or more. Specifically, according to FIG. 11, since the cumulative ratio of pores having a pore diameter exceeding 7 nm is 37%, the cumulative ratio of the volume of pores having a pore diameter of 7 nm or less is 63%.
Moreover, in this bulk body (average particle diameter D = 7 nm), the proportion of pores having a pore diameter three times (21 nm) or more of the average particle diameter D = 7 nm is 10% or less. Specifically, the cumulative ratio of the volume of pores having a pore diameter of 21 nm or more is 2%.
Further, in this bulk material (average particle diameter D = 7 nm), pore diameters of 100 nm or more are measured by the mercury intrusion method. However, pores having a pore diameter of 200 nm or more when measured by this mercury intrusion method are used. Was not present at all (0%).

In the bulk body corresponding to Example 2 (average particle diameter D = 20 nm), as can be seen from FIG. 11, the proportion of pores having a pore diameter of average particle diameter D = 20 nm or less in the total pore volume is as follows. 50% or more. Specifically, according to FIG. 11, since the ratio of pores having a pore diameter exceeding 20 nm is about 12%, the ratio of pores having a pore diameter of 20 nm or less is 88%.
Moreover, in this bulk body (average particle diameter D = 20 nm), the proportion of pores having a pore diameter three times (60 nm) or more of the average particle diameter D = 20 nm is 10% or less. Specifically, the proportion of pores having a pore diameter of 60 nm or more is 1%.
Further, in this bulk body (average particle diameter D = 20 nm), pore diameters of 100 nm or more are measured by the mercury intrusion method, but pores having a pore diameter of 200 nm or more when measured by this mercury intrusion method are present. Was not present at all (0%).

In the bulk body corresponding to Example 4 (average particle diameter D = 300 nm), as can be seen from FIG. 11, the proportion of pores having an average particle diameter D = 300 nm or less in the total pore volume is as follows. 50% or more. Specifically, according to FIG. 11, since the ratio of pores having a pore diameter exceeding 300 nm is 4%, the ratio of pores having a pore diameter of 300 nm or less is 96%.
Moreover, in this bulk body (average particle diameter D = 300 nm), the ratio of the pores having a pore diameter three times (900 nm) or more of the average particle diameter D = 300 nm is 10% or less. Specifically, the proportion of pores having a pore diameter of 900 nm or more is 3%.

In the bulk body corresponding to Example 8 (average particle diameter D = 150 nm), as can be seen from FIG. 11, the proportion of pores having an average particle diameter D = 150 nm or less in the total pore volume is as follows. 50% or more. Specifically, according to FIG. 11, since the ratio of pores having a pore diameter exceeding 150 nm is 1%, the ratio of pores having a pore diameter of 150 nm or less is 99%.
Moreover, in this bulk body (average particle diameter D = 150 nm), the proportion of pores having a pore diameter three times (450 nm) or more of the average particle diameter D = 150 nm is 10% or less. Specifically, there are no pores having a pore diameter of 450 nm or more (ratio 0%).
Further, in this bulk material (average particle diameter D = 150 nm), the pore diameter is measured by the mercury intrusion method, but there are absolutely no pores having a pore diameter of 200 nm or more when measured by this mercury intrusion method. Not (0%).

Next, for each of the gas sensors 100 of Examples 2, 4, and 8, the state of the pores of these protective layers 140 was observed from the surface by SEM.
In FIG. 12, the SEM image which looked at the protective layer 140 of the gas sensor 100 of Example 2 (average particle diameter D = 20 nm, titanium oxide particle) from the surface is shown. 13 and 14 show SEM images obtained by enlarging a part of FIG. From these, it can be seen that the protective layer 140 of the gas sensor 100 of Example 2 has an opening diameter of several tens of nanometers at most even for the large pores observed. Further, even when the entire surface of the protective layer 140 was observed, no pores having a pore diameter of 10 times (200 nm) or more of the average particle diameter D = 20 nm were observed.

  FIG. 15 shows an SEM image of the protective layer 140 of the gas sensor 100 of Example 8 (average particle diameter D = 150 nm, titanium oxide particles) viewed from the surface. From this SEM image, it can be seen that the protective layer 140 of the gas sensor 100 of Example 8 has an opening diameter of several hundred nm at most even for the large pores observed. Furthermore, even when the entire surface of the protective layer 140 was observed, pores having a pore diameter of 10 times (1500 nm) or more of the average particle diameter D = 150 nm were not observed at all.

  FIG. 16 shows an SEM image of the protective layer 140 of the gas sensor 100 of Example 4 (average particle diameter D = 300 nm, tungsten oxide particles) viewed from the surface. From this SEM image, it can be seen that the protective layer 140 of the gas sensor 100 of Example 4 has an opening diameter of about several hundred nm even for large pores observed. Further, even when the entire surface of the protective layer 140 was observed, no pores having a pore diameter 10 times (3000 nm) or more than the average particle diameter D = 300 nm were observed.

In the above, the present invention has been described with reference to the embodiment. However, the present invention is not limited to the above-described embodiment, and it is needless to say that the present invention can be appropriately modified and applied without departing from the gist thereof. .
For example, in the embodiment, the gas sensor 100 used for detecting an oxidizing gas such as nitrogen dioxide is illustrated, but the present invention may be applied to a gas sensor used for detecting a reducing gas such as carbon monoxide gas. it can. In that case, the reducing gas can be detected by forming the catalyst particles 133 with, for example, palladium instead of gold.
In the embodiment, the silicon substrate 101 is used for the gas sensor 100. However, for example, an alumina substrate can be used instead of the silicon substrate 101.
In the embodiment, when forming the protective layer 140, a sol having a predetermined viscosity is prepared by dispersing titanium oxide powder in water together with ammonium polyacrylate as a dispersing agent. Spin coated on top. However, the protective layer may be formed by dipping the sol on the gas sensitive film 131 and then performing heat treatment.

Furthermore, in the embodiment, after the sputtering process is performed to form the gas sensitive film 131 as a thin film, the catalyst particles 133 are dispersedly formed on the gas sensitive film 131 by the sputtering process. It is not limited to.
Specifically, for example, tin oxide powder is appropriately mixed with catalyst particles 133, a binder, and a viscosity modifier to create a paste, and this paste is a heating resistor on the insulating layer 114 of the surface-side insulating coating 110. There is a modified form in which the gas sensitive film 231 is thickly formed by printing at a position corresponding to 151 and the recess 103 and firing this (see FIG. 17). Thus, the gas sensor 200 having the gas-sensitive film 231 formed thick has a configuration in which the catalyst particles 133 are dispersed inside the gas-sensitive film 231.

Regarding the gas sensor according to the embodiment, (a) is a plan view as seen from the gas sensitive film side, and (b) is a side view as seen from the side in the vicinity of the heating resistor contact portion. It is the elements on larger scale near a gas sensitive film among the gas sensors concerning an embodiment. It is the elements on larger scale near the heating resistor contact part among the gas sensors concerning an embodiment. It is a partial expanded sectional view near a gas sensitive film, catalyst particles, and a protective layer among gas sensors concerning an embodiment. It is a graph which shows the change of the gas sensitivity with respect to silicon poisoning about the gas sensor which concerns on Examples 1-4 and Comparative Examples 1-4. It is a graph which shows the change of the gas sensitivity change rate with respect to silicon poisoning about the gas sensor which concerns on Examples 1-4 and Comparative Examples 1-4. It is a graph which shows the change of the gas sensitivity with respect to silicon poisoning about the gas sensor which concerns on Examples 5-7 and Comparative Examples 5 and 6. FIG. It is a graph which shows the change of the gas sensitivity change rate with respect to silicon poisoning about the gas sensor which concerns on Examples 5-7 and Comparative Examples 5 and 6. FIG. It is a graph which shows the change of the gas sensitivity with respect to silicon poisoning, and the change of a gas sensitivity change rate about the gas sensor which concerns on Example 8. FIG. It is a graph which shows the relationship between the pore diameter of the pore contained in a protective layer, and pore volume about the gas sensor from which the oxide particle which comprises a protective layer differs. It is a graph which shows the relationship between the pore diameter of the pore contained in a protective layer, and a cumulative ratio about the gas sensor from which the oxide particle which comprises a protective layer differs. It is the SEM image which observed the protective layer of the gas sensor which concerns on Example 2 from the surface. FIG. 13 is an SEM image obtained by magnifying a part of FIG. 12 for the protective layer of the gas sensor according to Example 2. FIG. It is another SEM image which expanded and observed another part of FIG. 12 about the protective layer of the gas sensor which concerns on Example 2. FIG. It is the SEM image which observed the protective layer of the gas sensor which concerns on Example 8 from the surface. It is the SEM image which observed the protective layer of the gas sensor which concerns on Example 4 from the surface. It is a partial expanded sectional view of the gas sensor film which contains a catalyst particle inside, and the vicinity of a protective layer among the gas sensors which concern on a deformation | transformation form.

100, 200 Gas sensor 101 Silicon substrate 110 Front side insulating coating 120 Back side insulating coating 131, 231 Gas sensitive film 133 Catalyst particle 140 Protective layer 141 Titanium oxide particles (oxide particles)
151 Heating resistor 155 Electrode

Claims (10)

A gas sensor comprising a gas-sensitive film made of an oxide semiconductor, and a gas-permeable protective layer formed on the gas-sensitive film,
The protective layer is
Ri Na average particle size is composed of the following oxide particles 500 nm,
A gas sensor , wherein a ratio of pores having a pore diameter equal to or smaller than the average particle diameter to a total volume of pores included in the protective layer is 50% or more . The gas sensor according to claim 1,
  The protective layer is
    The ratio of the pores having a pore diameter of 3 times or more of the average particle diameter in the total volume of the pores contained therein is 10% or less (including 0%).
Gas sensor. A gas sensor comprising a gas-sensitive film made of an oxide semiconductor, and a gas-permeable protective layer formed on the gas-sensitive film,
  The protective layer is
    It is composed of oxide particles having an average particle size of 500 nm or less,
    The proportion of pores having a pore diameter of 3 times or more of the average particle diameter in the total pore volume contained in the protective layer is 10% or less (including 0%).
Gas sensor. The gas sensor according to any one of claims 1 to 3 ,
The protective layer is
When the average particle diameter is D (nm) and the thickness of the protective layer is TH (nm),
A gas sensor in which 20 ≦ TH / D ≦ 500. The gas sensor according to any one of claims 1 to 4 , wherein:
Comprising catalyst particles dispersed on the gas sensitive membrane;
The protective layer is a gas sensor formed on the gas-sensitive film and catalyst particles. The gas sensor according to claim 5 ,
The protective layer is
A gas sensor having a film thickness at which the catalyst particles are not exposed. It is a gas sensor as described in any one of Claims 1-6, Comprising:
The protective layer is
A gas sensor in which a ratio of pores having a pore diameter of 200 nm or more when measured by a mercury intrusion method is 5% or less (including 0%) in the volume of the entire pores contained therein. It is a gas sensor as described in any one of Claims 1-7,
The protective layer is
A gas sensor in which pores having an opening diameter of 10 times or more of the average particle diameter are not observed when viewed from the surface. It is a gas sensor as described in any one of Claims 1-8, Comprising:
The protective layer is
A gas sensor obtained by depositing the oxide particles on the gas-sensitive film and performing a heat treatment in a temperature range in which the oxide particles do not sinter. It is a gas sensor as described in any one of Claims 1-9, Comprising:
A gas sensor using titanium oxide particles as the oxide particles.