Hydrogen gas sensor, and method for manufacturing same
The hydrogen gas sensor with polycrystalline CuO nanowires and voids addresses sensitivity, response speed, and baseline shift issues, offering enhanced performance in detecting hydrogen gas.
Patent Information
- Application Number
- PCT/JP2025/007565
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2024-03-04
- Filing Date
- 2025-03-03
- Publication Date
- 2025-09-11
AI Technical Summary
Conventional hydrogen gas sensors using CuO nanowires suffer from issues such as low sensitivity, slow response and recovery speed, limited detectable hydrogen gas concentration range, and significant baseline shift, which affect their reliability.
A hydrogen gas sensor design featuring CuO nanowires with a polycrystalline structure that includes voids, formed by oxidizing Cu nanowires with controlled heat treatments, allowing for high sensitivity, rapid response and recovery, and minimal baseline shift.
The sensor achieves improved sensitivity, rapid response and recovery characteristics, and a wide detectable hydrogen gas concentration range with reduced baseline shift, enhancing its practical usability.
Smart Images

Figure JP2025007565_12092025_PF_FP_ABST
Abstract
Description
Hydrogen gas sensor and manufacturing method thereof
[0001] The present invention relates to a hydrogen gas sensor using nanowires made of CuO and a method for manufacturing the same.
[0002] Various types of hydrogen gas sensors have been developed as hydrogen gas sensors suitable for applications such as detecting hydrogen gas leaking from various devices that handle hydrogen gas, such as fuel cells, and measuring the hydrogen gas concentration within devices that handle hydrogen gas. Among these, a hydrogen gas sensor is known that sandwiches a hydrogen gas detection unit between a pair of electrodes, applies a constant voltage between the electrodes, and detects hydrogen gas based on changes in current or resistance detected between the electrodes.
[0003] As an example of such a hydrogen gas sensor, as described in Non-Patent Document 1, a hydrogen gas sensor is known that has a hydrogen gas detection section with a CuO nanowire-CuO nanowire junction structure in which multiple CuO nanowires are grown vertically by thermally oxidizing a Cu layer, and the grown multiple CuO nanowires are entangled between adjacent pad electrodes.
[0004] Jae-Hun Kim et al., Growth and sensing properties of networked p-CuO nanowires, Sensors and Actuators B 212 (2015) p. 190-195
[0005] However, conventional hydrogen gas sensors using CuO nanowires, including those described in Non-Patent Document 1, have room for improvement in important practical characteristics such as sensitivity, response and recovery speed, detectable hydrogen gas concentration range, and baseline shift. Note that "baseline shift" refers to a phenomenon in which the current value or resistance value (baseline) when no hydrogen gas is being detected gradually changes as hydrogen gas detection is repeated. Baseline shift impairs the reliability of the hydrogen gas sensor and requires a circuit to correct the baseline shift, so a small baseline shift is required for practical use.
[0006] In view of the above problems, the present invention aims to provide a hydrogen gas sensor that has high sensitivity, excellent response and recovery characteristics, is capable of detecting hydrogen gas over a wide concentration range, and has little baseline shift, as well as a suitable method for manufacturing the same.
[0007] To solve the above problems, the present inventors conducted extensive research and discovered the following. The present inventors conducted extensive research into a hydrogen gas sensor using multiple CuO nanowires formed in parallel between electrodes as a hydrogen gas detection section. First, Cu nanowires with a substantially rectangular cross-section perpendicular to the extension direction are formed on a substrate. The Cu nanowires are composed of polycrystalline Cu containing amorphous material. Next, the Cu nanowires are heat-treated in an atmosphere containing hydrogen and an inert gas, which increases the grain size of the Cu crystal grains and improves their crystallinity. Due to internal stress acting on the Cu nanowires, the cross-sectional shape changes from rectangular to semicylindrical. Next, the Cu nanowires are heat-treated in an atmosphere containing oxygen, which oxidizes the Cu crystal grains to CuO crystal grains, forming a CuO nanowire composed of aggregated CuO crystal grains. Here, volume expansion occurs when the Cu crystal grains are oxidized to form CuO crystal grains. However, because the Cu nanowires are fixed to the substrate, they cannot expand in the direction of their extension, and their expansion in the width direction is also significantly restricted. Therefore, as a result of the volume expansion, some of the numerous grain boundaries formed by adjacent CuO crystal grains in the CuO nanowires peel off, forming voids. In other words, the CuO nanowires have a structure that retains voids internally. Although the estimated mechanism will be described later, by using CuO nanowires with such a characteristic structure as a hydrogen gas detection section, we were able to obtain a hydrogen gas sensor that has high sensitivity, excellent response and recovery characteristics, can detect hydrogen gas over a wide concentration range, and has little baseline shift.
[0008] The gist of the present invention, which was completed based on the above findings, is as follows: [1] A hydrogen gas sensor comprising: a substrate having an insulating surface, a first pad electrode and a second pad electrode formed on the insulating surface of the substrate, and a plurality of CuO nanowires made of CuO formed in parallel on the insulating surface of the substrate so as to connect the first pad electrode and the second pad electrode, each of the plurality of CuO nanowires being polycrystalline with an aggregate of crystal grains made of CuO and having a structure that holds voids therein, wherein a current is passed between the first pad electrode and the second pad electrode, and hydrogen gas is detected based on a change in an electrical signal detected between the first pad electrode and the second pad electrode.
[0009] [2] The hydrogen gas sensor according to [1] above, wherein the length of each of the plurality of CuO nanowires is 10 nm or more and 10 μm or less.
[0010] [3] The hydrogen gas sensor according to [1] or [2] above, wherein the line width of each of the plurality of CuO nanowires is 10 nm or more and 200 nm or less.
[0011] [4] The hydrogen gas sensor according to any one of [1] to [3] above, wherein the thickness of each of the plurality of CuO nanowires is 5 nm or more and 100 nm or less.
[0012] [5] The hydrogen gas sensor according to any one of [1] to [4] above, wherein the number of the CuO nanowires is 2 or more and 1000 or less.
[0013] [6] The hydrogen gas sensor according to any one of [1] to [5] above, wherein the average porosity in a cross section perpendicular to the extension direction of the CuO nanowires is 5% or more and 30% or less.
[0014] [7] The hydrogen gas sensor according to any one of [1] to [6] above, wherein some of the crystal grains constituting the CuO nanowire are in surface contact with the insulating surface of the substrate.
[0015] [8] A hydrogen gas sensor described in any one of [1] to [7] above, wherein in a cross section perpendicular to the extension direction of the CuO nanowire, the width of the CuO nanowire gradually decreases with increasing distance from the insulating surface of the substrate, or gradually increases and then gradually decreases with increasing distance from the insulating surface of the substrate.
[0016] [9] The hydrogen gas sensor according to any one of the above [1] to [8], wherein the first pad electrode and the second pad electrode are made of platinum (Pt).
[0017]
[10] The hydrogen gas sensor according to any one of [1] to [9] above, wherein the substrate is any one of a glass substrate, an alumina substrate, a zirconia substrate, and a silicon substrate having a silicon oxide film or a silicon nitride film formed on its surface.
[0018]
[11] A method for manufacturing a hydrogen gas sensor according to [1] above, comprising the steps of: preparing a substrate having an insulating surface; forming a first pad electrode and a second pad electrode on the insulating surface of the substrate; forming a plurality of Cu nanowires made of Cu in parallel on the insulating surface of the substrate so as to connect the first pad electrode and the second pad electrode; a first heat treatment step of subjecting the plurality of Cu nanowires to a heat treatment in an atmosphere containing hydrogen and an inert gas; and a second heat treatment step of subjecting the plurality of Cu nanowires to a heat treatment in an atmosphere containing oxygen to convert the plurality of Cu nanowires into a plurality of CuO nanowires made of CuO.
[0019]
[12] The method for manufacturing a hydrogen gas sensor described in
[11] above, wherein the first heat treatment step is performed by raising the ambient temperature to a first temperature of 80°C or higher and 250°C or lower at a heating rate of 1°C / min or higher and 100°C / sec or lower, and then maintaining the first temperature for 1 minute or higher and 100 minutes or lower.
[0020]
[13] The method for manufacturing a hydrogen gas sensor described in
[11] or
[12] above, wherein the second heat treatment step is performed by raising the ambient temperature to a second temperature of 250°C or higher and 500°C or lower at a heating rate of 0.5°C / min or higher and 20°C / min or lower, and then maintaining the second temperature for 10 minutes or higher and 90 minutes or lower.
[0021]
[14] The method for manufacturing a hydrogen gas sensor according to any one of
[11] to
[13] above, wherein the length of each of the plurality of Cu nanowires is 10 nm or more and 10 μm or less.
[0022]
[15] The method for manufacturing a hydrogen gas sensor according to any one of
[11] to
[14] above, wherein the line width of each of the plurality of Cu nanowires is 10 nm or more and 200 nm or less.
[0023]
[16] The method for manufacturing a hydrogen gas sensor according to any one of
[11] to
[15] above, wherein the thickness of each of the plurality of Cu nanowires is 5 nm or more and 100 nm or less.
[0024]
[17] The method for manufacturing a hydrogen gas sensor according to any one of
[11] to
[16] above, wherein the number of the Cu nanowires is 2 or more and 1000 or less.
[0025]
[18] The method for manufacturing a hydrogen gas sensor according to any one of
[11] to
[17] above, wherein the first pad electrode and the second pad electrode are made of platinum (Pt).
[0026]
[19] The method for manufacturing a hydrogen gas sensor according to any one of
[11] to
[18] above, wherein the substrate is any one of a glass substrate, an alumina substrate, a zirconia substrate, and a silicon substrate having a silicon oxide film or a silicon nitride film formed on its surface.
[0027] The hydrogen gas sensor of the present invention has high sensitivity, excellent response and recovery characteristics, can detect hydrogen gas over a wide concentration range, and has little baseline shift. Furthermore, the method for manufacturing a hydrogen gas sensor of the present invention makes it possible to manufacture a hydrogen gas sensor that has high sensitivity, excellent response and recovery characteristics, can detect hydrogen gas over a wide concentration range, and has little baseline shift.
[0028] 1 is a schematic perspective view of a hydrogen gas sensor 100 according to an embodiment of the present invention; 2 is a schematic top view of the hydrogen gas sensor 100; 3 (A) and (B) are cross-sectional views perpendicular to the extension direction of the CuO nanowires 16 in the hydrogen gas sensor 100; 4 (A) and (B) are cross-sectional views perpendicular to the extension direction of the CuO nanowires 16 in the hydrogen gas sensor 100; 5 (B) are a schematic exploded perspective view of the hydrogen gas sensor 100 used to explain a manufacturing method of the hydrogen gas sensor 100 according to an embodiment of the present invention; 6 (A) is a cross-sectional view perpendicular to the extension direction of the Cu nanowires 15; 7 (B) is a cross-sectional view perpendicular to the extension direction of the Cu nanowires 15 after a first heat treatment (hydrogen annealing); and 8 (C) is a cross-sectional view perpendicular to the extension direction of the CuO nanowires 16 formed by subjecting the Cu nanowires 15 to a second heat treatment (oxidation annealing). 9 are top-view SEM images of nanowires in Comparative Examples 1 to 3 and an inventive example of Experimental Example 1; 10 (Glitter-Induced XRD patterns) in Comparative Examples 2 and 3 of Experimental Example 1 and an inventive example. 1 is an X-ray diffraction spectrum of an inventive example of Experimental Example 1. 2 is an X-ray diffraction spectrum of Comparative Example 3 of Experimental Example 1 and an inventive example. 3 is a graph showing the change in resistance over time during a hydrogen gas detection test for Comparative Example 3 of Experimental Example 1 and an inventive example. 4 is a graph showing the change in resistance change rate over time during a hydrogen gas detection test for Comparative Example 3 of Experimental Example 2 and an inventive example. 5 is a SEM image of Experimental Example 3, where (A) is a cross-sectional SEM image perpendicular to the extension direction of the Cu nanowires before the first heat treatment (hydrogen annealing), (B) is a cross-sectional SEM image perpendicular to the extension direction of the Cu nanowires after the first heat treatment (hydrogen annealing), and (C) is a cross-sectional SEM image perpendicular to the extension direction of CuO nanowires formed by subjecting the Cu nanowires to a second heat treatment (oxidation annealing). 1A and 1B are SEM images from Experimental Example 3, where (A) is a top-view SEM image of a portion of the device before the first heat treatment (hydrogen annealing), and (B) is a top-view SEM image of a portion of the device (hydrogen gas sensor) after the second heat treatment (oxidation annealing). 1B is a top-view SEM image of CuO nanowires in the hydrogen gas sensor fabricated in Experimental Example 3. 1C is a cross-sectional SEM image perpendicular to the extension direction of the CuO nanowires in the hydrogen gas sensor fabricated in Experimental Example 3. 1D is a graph showing the change in the resistance change rate over time during a hydrogen gas detection test in Experimental Example 4.1 is a graph showing the relationship between hydrogen gas concentration and sensitivity in Experimental Example 4. 2 is a graph showing the change in resistance change rate over time during a hydrogen gas detection test at an extremely low hydrogen gas concentration in Experimental Example 4. 3 is a graph showing the change in resistance value over time during a hydrogen gas detection test at various gap lengths (lengths of CuO nanowires) in Experimental Example 5. 4 is a graph showing the gap length and response time t in Experimental Example 5. res90 and recovery time t rec90 1 is a graph showing the relationship between the resistance and the sensitivity in Experimental Example 5. FIG. 1 is a graph showing the relationship between the hydrogen gas concentration and the sensitivity in Experimental Example 6. FIG. 2 is a graph showing the change in resistance over time during a hydrogen gas detection test at various applied voltages in Experimental Example 6. FIG. 3 is a graph showing the change in resistance change rate over time during a hydrogen gas detection test conducted using dry air and humidified air as the carrier gas in Experimental Example 7. FIG. 4 is a graph showing the change in resistance over time during a hydrogen gas detection test conducted again using dry air as the carrier gas in Experimental Example 7. FIG. 5 is a graph showing the change in resistance over time during a hydrogen gas detection test at various operating temperatures in Experimental Example 8. FIG. 6 is a graph showing the change in resistance change rate over time during a hydrogen gas detection test conducted 1 day and 100 days after the fabrication of a hydrogen gas sensor in Experimental Example 9. FIG. 7 is a graph showing the change in resistance over time during a hydrogen gas detection test with various line widths in Experimental Example 10. FIG. 8 is a graph showing the change in resistance over time during a hydrogen gas detection test with various numbers of CuO nanowires in Experimental Example 11.
[0029] 1, 2, 3A, and 3B, a hydrogen gas sensor 100 according to one embodiment of the present invention includes a substrate 10, a first pad electrode 12, a second pad electrode 14, and a plurality of CuO nanowires 16. The first pad electrode 12 and the second pad electrode 14 are formed on the substrate 10. The plurality of CuO nanowires 16 are formed in parallel on the substrate 10 to connect the first pad electrode 12 and the second pad electrode 14. Although detailed effects will be described later, in this embodiment, the function of the plurality of CuO nanowires 16 realizes improvements in sensor characteristics such as sensitivity, response and recovery characteristics, detectable concentration range, and suppression of baseline shift.
[0030] [Mechanism of Hydrogen Gas Detection] In the hydrogen gas sensor 100, a current is passed between the first pad electrode 12 and the second pad electrode 14, and hydrogen gas is detected based on a change in the electrical signal detected between the first pad electrode 12 and the second pad electrode 14. For example, as shown in FIG. 2 , a power supply 18 and an ammeter 20 are connected in series between the first pad electrode 12 and the second pad electrode 14, and a voltmeter 22 is connected in parallel with the power supply 18. In this case, the power supply 18 applies a constant voltage between the first pad electrode 12 and the second pad electrode 14 that can be measured by the voltmeter 22, and the ammeter 20 detects a change in the current between the first pad electrode 12 and the second pad electrode 14, and hydrogen gas is detected based on the detected change in current. Alternatively, gas can be detected based on a change in the voltage detected between the first pad electrode 12 and the second pad electrode 14 while a constant current is passed between the first pad electrode 12 and the second pad electrode 14. Furthermore, instead of the above-described change in current or voltage, gas can also be detected based on a change in resistance detected between the first pad electrode 12 and the second pad electrode 14. That is, the above-described "electrical signal" means a current, a voltage, or a resistance. In this embodiment, the gas sensor circuit can be configured with two terminals, so that a hydrogen gas sensor can be constructed without increasing the number of wirings or circuits.
[0031] [Substrate] The substrate 10 supports the first pad electrode 12, the second pad electrode 14, and the plurality of CuO nanowires 16 that serve as the hydrogen gas detector. The substrate 10 is not particularly limited as long as it has an insulating surface. For example, insulating substrates such as glass substrates, alumina substrates, and zirconia substrates, or silicon substrates having a silicon oxide film or silicon nitride film formed on their surfaces, can be used. In this embodiment, from the viewpoint of adhesion to the CuO nanowires 16, the substrate 10 is preferably a silicon substrate having a silicon oxide film or silicon nitride film formed on its surface. The shape and dimensions of the substrate 10 are not particularly limited. However, when a substrate having a rectangular main surface is used, the dimensions can be, for example, in the range of length: 1 to 300 mm, width: 1 to 300 mm, and thickness: 0.1 to 1.2 mm.
[0032] [First Pad Electrode and Second Pad Electrode] The first pad electrode 12 and the second pad electrode 14 are a pair of electrodes necessary for supplying current to the multiple CuO nanowires 16 and detecting changes in an electrical signal corresponding to changes in the concentration of hydrogen gas. The first pad electrode 12 and the second pad electrode 14 are not particularly limited in shape and size as long as they are formed on the insulating surface of the substrate 10. However, in this embodiment, it is preferable that the first pad electrode 12 and the second pad electrode 14 have a shape that makes it easy to arrange the multiple CuO nanowires 16 in parallel with each other and with the same length.
[0033] For example, as shown in FIGS. 1 and 2 , the first pad electrode 12 includes a first island portion 12A, a first connecting portion 12B, and a first pad portion 12C, and the second pad electrode 14 includes a second island portion 14A, a second connecting portion 14B, and a second pad portion 14C. The first island portion 12A and the second island portion 14A have rectangular main surfaces and are spaced apart such that one side of the first island portion 12A and one side of the second island portion 14A are parallel to each other. A plurality of CuO nanowires 16 are formed on the substrate 10 in parallel, preferably parallel to each other, to connect the first island portion 12A and the second island portion 14A. That is, the first island portion 12A and the second island portion 14A function as gap electrodes. In this case, the lengths of the plurality of CuO nanowires 16 are equal to each other and are equal to the distance (gap length G) between the first island portion 12A and the second island portion 14A. The dimensions of the first island portion 12A and the second island portion 14A may be, for example, such that the long side connected to the CuO nanowires 16 is in the range of 5 to 250 μm and the short side is in the range of 1 to 150 μm.
[0034] 2, the first pad portion 12C and the second pad portion 14C are portions to which a power supply 18, an ammeter 20, and a voltmeter 22 are connected, and serve as pad electrodes. There are no particular limitations on the shape and dimensions of the first pad portion 12C and the second pad portion 14C, but when the main surface is rectangular, the dimensions may be, for example, in the range of length: 30 to 1000 μm×width: 30 to 1000 μm.
[0035] The first connecting portion 12B is a linear portion that connects the first island portion 12A and the first pad portion 12C, and the second connecting portion 14B is a linear portion that connects the second island portion 14A and the second pad portion 14C. The dimensions of the first connecting portion 12B and the second connecting portion 14B may be, for example, a length in the range of 1 to 500 μm and a width in the range of 1 to 50 μm.
[0036] The thickness of the first pad electrode 12 and the second pad electrode 14 is not particularly limited, but may be in the range of 5 to 500 nm.
[0037] The metal constituting the first pad electrode 12 and the second pad electrode 14 must be a metal that will not be oxidized by the second heat treatment (oxidation annealing) described below and can withstand heat, and is preferably one or more selected from platinum (Pt), iridium (Ir), gold (Au), rhodium (Rh), and palladium (Pd), and more preferably platinum (Pt).
[0038] An adhesive layer for adhering the first pad electrode 12 and the second pad electrode 14 to the substrate 10 may be disposed between the substrate 10 and the first and second pad electrodes 12 and 14. The adhesive layer may be made of, for example, one material selected from Ti, Cr, and Ta, and may have a thickness in the range of 1 to 5 nm.
[0039] [CuO Nanowires] A plurality of CuO nanowires 16 are elements constituting the hydrogen gas detection unit, formed in parallel, preferably parallel to each other, on the substrate 10 to connect the first pad electrode 12 and the second pad electrode 14. CuO is a p-type semiconductor, but has low conductivity due to its low carrier concentration. Therefore, by arranging a plurality of CuO nanowires 16 in parallel, the magnitude of the electrical signal detected between the first pad electrode 12 and the second pad electrode 14 is ensured.
[0040] The mechanism of hydrogen gas detection is as follows: When hydrogen gas comes into contact with the CuO nanowires 16 during response of the hydrogen gas sensor 100, O on the surface of the CuO nanowires 16 is converted into O. 2-As a result, the carrier concentration in the CuO nanowires 16 decreases, and the detected current value decreases (the resistance value increases). When the hydrogen gas sensor 100 recovers, O ions are again deposited on the surface of the CuO nanowires. 2- Ions are generated and an accumulation layer is formed on the surface of the CuO nanowires, increasing the current value (decreasing the resistance value).
[0041] 3A and 3B, this embodiment is characterized in that each of the multiple CuO nanowires 16 is a polycrystal formed by an aggregation of CuO crystal grains 16A and has a structure that maintains voids 16B therein. In general, polycrystals have grain boundaries formed by adjacent crystal grains in contact with each other, so no voids exist inside. In contrast, in the CuO nanowires 16, some of the grain boundaries formed by adjacent CuO crystal grains 16A are peeled off, forming voids 16B. The voids 16B increase the specific surface area of the CuO nanowires 16. The inventors speculate that this increase in specific surface area results in significant effects, such as high sensitivity, excellent response and recovery characteristics, the ability to detect hydrogen gas over a wide concentration range, and minimal baseline shift.
[0042] The average porosity in a cross section perpendicular to the extension direction of the multiple CuO nanowires 16 is preferably 5% or more and 30% or less. If the average porosity is 5% or more, the effect of the voids 16B is fully exerted, and the effects of the present invention can be fully obtained. If the average porosity is 30% or less, an appropriate ratio of the contact surface between the crystal grains 16A can be ensured, thereby reliably maintaining the structure of the CuO nanowires 16.
[0043] In this specification, the "average porosity" is determined by the following method. For example, as shown in FIG. 15, in an SEM image of a specific cross section perpendicular to the extension direction of a specific CuO nanowire, CuO crystal grains and voids are visually clearly distinguishable. Therefore, the cross-sectional SEM image is subjected to image processing such as binarization to identify the CuO crystal grain regions and the void regions, and the area A of the CuO crystal grain regions and the area B of the void regions are determined. The "porosity" of the cross section is determined by calculating B / (A+B). The porosity is determined at at least one location in each of the multiple CuO nanowires and at at least 10 locations across the multiple CuO nanowires, and the arithmetic average of the porosity values is defined as the "average porosity."
[0044] 3A and 3B, in this embodiment, it is also important that some crystal grains 16A constituting the CuO nanowires 16 are in surface contact with the insulating surface of the substrate 10. The CuO nanowires 16 are formed directly on the insulating surface of the substrate 10. The crystal grains located at the bottom of the CuO nanowires 16 (i.e., the crystal grains in contact with the insulating surface of the substrate 10) are in surface contact with the insulating surface of the substrate 10. This is because the CuO nanowires 16 are formed by hydrogen annealing and oxidation annealing Cu nanowires formed directly on the substrate 10. The effects of the present invention are preferably achieved by the CuO nanowires 16 retaining voids 16B and some crystal grains 16A being in surface contact with the insulating surface of the substrate 10.
[0045] A preferred shape of the cross section perpendicular to the extension direction of the CuO nanowires 16 will now be described. As shown in FIG. 3A, in a cross section perpendicular to the extension direction of the CuO nanowires 16, the width of the CuO nanowires 16 preferably gradually decreases with increasing distance from the insulating surface of the substrate 10. This means that the cross section has a so-called semi-cylindrical shape. Also, as shown in FIG. 3B, in a cross section perpendicular to the extension direction of the CuO nanowires 16, the width of the CuO nanowires 16 preferably gradually increases and then gradually decreases with increasing distance from the insulating surface of the substrate 10. In this case, the cross section has a generally rounded shape except for the contact portion with the substrate 10. That is, in this cross section, the contact width of the CuO nanowires 16 with the substrate 10 is smaller than the maximum width of the CuO nanowires 16. In other words, in a cross section perpendicular to the extension direction of the CuO nanowires 16, the width of the CuO nanowires 16 reaches a maximum near the center in the height direction and gradually decreases with increasing distance in the height direction from this maximum height position. The reason for these shapes is that extremely large stress (pressure) is applied within the Cu nanowires, which have a rectangular cross section perpendicular to the extension direction, during the process of hydrogen annealing and oxidation annealing. This internal stress is what enables the effects of the present invention to be effectively achieved.
[0046] [Length of CuO Nanowires (Gap Length)] Referring to FIGS. 1 and 2 , the length of the CuO nanowires 16 is equal to the gap distance (gap length G) between the first island portion 12A and the second island portion 14A. While the length of the CuO nanowires 16 is not particularly limited, the shorter the length, the higher the electric field strength even at a low applied voltage, resulting in better response and recovery characteristics. From this perspective, the length of each of the multiple CuO nanowires 16 is preferably 10 μm or less, more preferably 1000 nm or less, even more preferably 700 nm or less, even more preferably 400 nm or less, and most preferably 300 nm or less. On the other hand, the shorter the length of the CuO nanowires 16, the lower the sensitivity tends to be. Therefore, it is preferable to ensure a minimum length, especially for detecting low concentrations of hydrogen gas. From this perspective, the length of each of the multiple CuO nanowires 16 is preferably 10 nm or more, more preferably 30 nm or more.
[0047] 3A and 3B, the line width W of the CuO nanowires 16 is not particularly limited as long as it is on the order of nanometers (i.e., less than 1 μm). However, from the viewpoint of generating an appropriate internal stress within the CuO nanowires 16, the line width W of each of the multiple CuO nanowires 16 is preferably 200 nm or less, more preferably 150 nm or less, and even more preferably 100 nm or less. Furthermore, from the viewpoint of generating an appropriate internal stress within the CuO nanowires 16, the line width W of each of the multiple CuO nanowires 16 is preferably 10 nm or more, more preferably 50 nm or more, and even more preferably 80 nm or more.
[0048] 3A and 3B, the line width W of the CuO nanowire 16 refers to the maximum width in a cross section perpendicular to the extension direction of the CuO nanowire 16. The line width W of the CuO nanowire 16 is determined by SEM observation of the cross section perpendicular to the extension direction of the CuO nanowire 16 at five equally spaced locations along the length, and the arithmetic average value of the line widths obtained in the SEM images is used.
[0049] 3A and 3B, the thickness D of the CuO nanowires 16 is not particularly limited. However, from the viewpoint of generating an appropriate internal stress in the CuO nanowires 16, the thickness D of each of the plurality of CuO nanowires 16 is preferably 5 nm or more, more preferably 10 nm or more, and even more preferably 20 nm or more. Furthermore, from the viewpoint of generating an appropriate internal stress in the CuO nanowires 16, the thickness D of each of the plurality of CuO nanowires 16 is preferably 100 nm or less, more preferably 80 nm or less, and even more preferably 70 nm or less.
[0050] In this specification, the thickness D of the CuO nanowires 16 refers to the distance from the bottom surface (interface with the substrate 10) of the CuO nanowires 16 to the top surface, as shown in Figures 3(A) and 3(B). The thickness D of the CuO nanowires 16 is determined by SEM observation of a cross section perpendicular to the extension direction of the CuO nanowires 16 at five equally spaced locations along the length, and by averaging the thicknesses obtained from the SEM images.
[0051] [Number of CuO Nanowires] The number of CuO nanowires 16 arranged in parallel is not particularly limited, but from the viewpoint of ensuring the magnitude of the electrical signal detected between the first pad electrode 12 and the second pad electrode 14, it is preferably 2 or more, and more preferably 10 or more. On the other hand, from the viewpoint of sensitivity and response / recovery characteristics, the number of CuO nanowires is preferably 1000 or less, and more preferably 100 or less.
[0052] [Effects] As described above, the hydrogen gas sensor 100 of this embodiment has the advantages of being highly sensitive, having excellent response and recovery characteristics, being capable of detecting hydrogen gas over a wide concentration range, and having little baseline shift.
[0053] 4A, 4B, and 5A to 5C, a method for manufacturing the hydrogen gas sensor 100 according to one embodiment of the present invention will be described. The method for manufacturing the hydrogen gas sensor 100 according to this embodiment includes the steps of preparing a substrate 10 (step S1), forming a first pad electrode 12 and a second pad electrode 14 (step S2), forming Cu nanowires 15 (step S3), a first heat treatment step (step S4), and a second heat treatment step (step S5). By this method, the hydrogen gas sensor 100 according to this embodiment described above can be manufactured.
[0054] [Step S1] In step S1, the substrate 10 is prepared. Details of the substrate 10 are as described above.
[0055] [Step S2] In step S2, the first pad electrode 12 and the second pad electrode 14 are formed on the insulating surface of the substrate 10. The method for forming the first pad electrode 12 and the second pad electrode 14 is not particularly limited, but the following method can be used, for example.
[0056] First, electron beam lithography (EBL) is used to fabricate the first island portion 12A and the first connecting portion 12B of the first pad electrode 12, and the second island portion 14A and the second connecting portion 14B of the second pad electrode 14.
[0057] Specifically, a resist film is first formed on the substrate 10. The resist film can be formed by applying a resist composition for electron beam exposure onto the substrate 10 and drying it. There are no particular limitations on the application method, but spin coating is preferably used. The thickness of the resist film may be appropriately set so as to be thicker than the thickness of the portion to be formed. After spin coating, the resist film may be annealed under appropriate conditions to efficiently volatilize the solvent and increase the density of the resist film.
[0058] The resist film is then developed to form a mask pattern of a predetermined shape. The mask pattern is produced by exposing the resist film by electron beam lithography and developing it. The position, shape, and dimensions of the portion of the mask pattern where the resist film is removed and the substrate 10 is exposed correspond to the position, shape, and dimensions of the portion to be formed.
[0059] Next, a metal film is formed by, for example, electron beam evaporation or sputtering. At this time, a first portion of the metal film is formed on the mask pattern, and a second portion of the metal film is formed on the substrate 10 exposed by removing the resist film in the mask pattern. The metal film preferably includes a Ti layer, Cr layer, or Ta layer (adhesion layer) having a thickness of about 1 to 5 nm, and a layer made of a metal (e.g., Pt) formed thereon that constitutes the first pad electrode 12 and the second pad electrode 14. The Ti layer, Cr layer, or Ta layer functions as an adhesive layer for adhering the first pad electrode 12 and the second pad electrode 14 to the substrate 10.
[0060] Next, the mask pattern is peeled off and a lift-off process is performed to remove a first portion of the metal film formed thereon, thereby forming the above-mentioned portions (first island portion 12A and first connecting portion 12B, and second island portion 14A and second connecting portion 14B) on the substrate 10. The Ti, Cr, or Ta layer is formed as an adhesive layer to adhere the above-mentioned portions to the substrate 10. However, during the deposition process of the metal constituting the above-mentioned portions, Ti, Cr, or Ta diffuses into the metal layer, and most of the Ti, Cr, or Ta layer disappears. Alternatively, the adhesive layer is so thin that it is barely visible in SEM images. Although islands of Ti, Cr, or Ta remain in some locations between the substrate 10 and the above-mentioned portions, the substrate 10 and the above-mentioned portions are in direct contact with each other in most locations.
[0061] Next, the first pad portion 12C of the first pad electrode 12 and the second pad portion 14C of the second pad electrode 14 are fabricated using a general photolithography method.
[0062] Specifically, a photoresist film is first formed on the substrate 10, and the photoresist film is then exposed and developed to form a mask pattern of a predetermined shape. The position, shape, and dimensions of the portion of the substrate 10 exposed by removing the photoresist film in the mask pattern correspond to the position, shape, and dimensions of the first pad portion 12C and the second pad portion 14C. Next, a metal film is formed, for example, by electron beam evaporation or sputtering. At this time, a first portion of the metal film is formed on the mask pattern, and a second portion of the metal film is formed on the substrate 10 exposed by removing the resist film in the mask pattern. The metal film preferably includes a Ti, Cr, or Ta layer (adhesion layer) having a thickness of approximately 1 to 5 nm and a layer made of a metal (e.g., Pt) that constitutes the first pad portion 12C and the second pad portion 14C. Next, the mask pattern is peeled off, and the first portion of the metal film formed thereon is removed using a lift-off process to form the above-mentioned portions (the first pad portion 12C and the second pad portion 14C) on the substrate 10. As mentioned above, the adhesive layer almost completely disappears.
[0063] The first pad electrode 12 and the second pad electrode 14 thus formed are preferably subjected to a heat treatment to clean these pad electrodes. The heat treatment is preferably performed in an atmosphere containing hydrogen, with the balance being an inert gas and unavoidable impurity gases that may be optionally contained. The hydrogen content is preferably 1 to 5 volume %, and the balance of the inert gas may be one or more selected from argon (Ar), helium (He), and neon (Ne). This heat treatment is preferably performed under conditions in which the atmospheric temperature is raised to a temperature of 500°C to 650°C at a rate of 2°C / min to 100°C / min, and then the temperature is maintained for 1 minute to 60 minutes.
[0064] [Step S3] In step S3, a plurality of Cu nanowires 15 made of Cu are formed in parallel on the insulating surface of the substrate 10 so as to connect the first pad electrode 12 and the second pad electrode 14. The Cu nanowires 15 are preferably formed using electron beam lithography (EBL) as follows.
[0065] First, a resist film is formed on the substrate 10 at least in the area where the first island portion 12A and the second island portion 14A are separated from each other. The resist film can be formed by applying a resist composition for electron beam exposure to the substrate 10 and drying it. There are no particular limitations on the application method, but spin coating is preferably used. The thickness of the resist film may be appropriately set so as to be thicker than the thickness of the Cu nanowires 15 to be formed. After spin coating, the resist film may be annealed under appropriate conditions to efficiently volatilize the solvent and increase the density of the resist film.
[0066] The resist film is then developed to form a mask pattern of a predetermined shape. The mask pattern is created by exposing the resist film by electron beam lithography and developing it. The position, shape, and dimensions of the portion of the mask pattern where the resist film is removed and the substrate 10 is exposed correspond to the position, shape, and dimensions of the multiple Cu nanowires to be formed.
[0067] Next, a Cu film is formed by, for example, electron beam evaporation or sputtering. At this time, a first portion of the Cu film is formed on the mask pattern, and a second portion of the Cu film is formed on the substrate 10 exposed by removing the resist film in the mask pattern. Next, the mask pattern is peeled off, and a lift-off process is performed to remove the first portion of the Cu film formed thereon, thereby forming a plurality of Cu nanowires 15 on the substrate 10. Note that, since Cu has good adhesion to the substrate 10, an adhesive layer is not necessary, and the plurality of Cu nanowires 15 are formed in direct contact with the insulating surface of the substrate 10.
[0068] As shown in Figure 5(A), the cross-sectional shape perpendicular to the extension direction of the Cu nanowire 15 formed on the substrate 10 is approximately rectangular, and although the Cu nanowire 15 contains some amorphous material 15B, it is made of polycrystalline Cu crystal grains 15A, and there are no voids inside.
[0069] As already described, the preferred length of each of the plurality of Cu nanowires 15 is the same as the preferred length of each of the plurality of CuO nanowires 16, and therefore, the description thereof will be omitted.
[0070] Since the volume of the Cu nanowires 15 expands during the process of turning into CuO nanowires 16 through oxidation annealing, the line width of the Cu nanowires 15 increases slightly even though the Cu nanowires 15 are fixed to the substrate 10. However, since the preferred line width of each of the multiple Cu nanowires 15 is the same as the preferred length of each of the multiple CuO nanowires 16, a description thereof will be omitted.
[0071] Similarly, the thickness of the Cu nanowires 15 also increases slightly as they undergo oxidation annealing to become CuO nanowires 16. However, since the preferred thickness of each of the multiple Cu nanowires 15 is the same as the preferred thickness of each of the multiple CuO nanowires 16, a description thereof will be omitted.
[0072] As already mentioned, the suitable number of Cu nanowires 15 is the same as the suitable number of CuO nanowires 16, and therefore the explanation thereof will be omitted.
[0073] [Step S4] In step S4, a first heat treatment process (hydrogen annealing) is performed on the multiple Cu nanowires 15 in an atmosphere containing hydrogen and an inert gas. This first heat treatment increases the grain size of the Cu crystal grains 15A, improves the crystallinity of the Cu crystal grains 15A, and eliminates the amorphous portion 15B, as shown in FIG. 5B. As a result, the internal stress acting on the Cu nanowires 15 changes the cross-sectional shape of the Cu nanowires 15 perpendicular to their extension direction from a rectangular to a semi-cylindrical shape. However, even at this stage, no voids exist within the Cu nanowires 15.
[0074] The first heat treatment is preferably performed in an atmosphere containing hydrogen, with the balance being an inert gas and any unavoidable impurity gases that may be included. The hydrogen content is preferably 1 to 5 volume %, and the balance of the inert gas may be one or more selected from argon (Ar), helium (He), and neon (Ne). The first heat treatment is preferably performed under conditions in which the ambient temperature is raised to a first temperature of 80°C to 250°C at a rate of 1°C / min to 100°C / sec, and then the first temperature is maintained for 1 minute to 100 minutes. The means for the first heat treatment are not particularly limited, and a general heat treatment furnace or an RTA (Rapid Thermal Anneal) device may be used.
[0075] [Step S5] In step S5, a second heat treatment process (oxidation annealing) is performed on the multiple Cu nanowires 15 in an oxygen-containing atmosphere. As a result, as shown in FIG. 5C, the Cu crystal grains 15A are oxidized to become CuO crystal grains 16A, forming CuO nanowires 16, each composed of an aggregate of the CuO crystal grains 16A. When the Cu crystal grains 15A are oxidized to become CuO crystal grains 16A, volume expansion occurs. However, because the Cu nanowires 15 are fixed to the substrate 10, they cannot expand in the wire extension direction, and expansion in the wire width direction is also significantly restricted. Therefore, as a result of the volume expansion, some of the numerous grain boundaries formed by adjacent CuO crystal grains in the CuO nanowires 16 peel off, forming voids 16B. In other words, the CuO nanowires 16 have a structure that retains voids 16B within them.
[0076] The second heat treatment is performed in an oxygen-containing atmosphere, for example, a dry air atmosphere. The second heat treatment step is preferably performed under conditions in which the ambient temperature is raised to a second temperature of 250°C to 500°C at a heating rate of 0.5°C / min to 20°C / min, and then maintained at the second temperature for 10 minutes to 90 minutes. The heating rate is more preferably 10°C / min or less, and even more preferably 5°C / min or less. The means for the second heat treatment are not particularly limited, and a general heat treatment furnace or an RTA (Rapid Thermal Anneal) device may be used.
[0077] To obtain a polycrystalline CuO nanowire 16 with an internal void 16B composed of aggregated CuO crystal grains 16A, it is necessary to perform a first heat treatment (hydrogen annealing) prior to a second heat treatment (oxidation annealing). If only the second heat treatment is performed without the first heat treatment, CuO crystal grains are formed, but they are small in size. Furthermore, small pieces of CuO crystal grains are scattered around during the second heat treatment (see the SEM image (bottom) of Comparative Example 3 in Figure 6). This is because, without the first heat treatment, Cu crystal grains with low crystallinity are subjected to the second heat treatment (oxidation annealing), causing the crystal grains to expand in volume and explode during the rapid oxidation process. In this case, the internal stress on the nanowire is relieved, and no voids are formed inside the nanowire. In this embodiment, the size of the Cu crystal grains is increased and the crystallinity is improved by the first heat treatment (hydrogen annealing), and then the second heat treatment (oxidation annealing) is performed. As a result, the crystal grains grow while internal stress is applied to the nanowires, without small pieces of CuO crystal grains scattering around, and voids 16B are formed.
[0078] [Effects] Through the above steps, it is possible to manufacture a hydrogen gas sensor 100 that has high sensitivity, excellent response and recovery characteristics, is capable of detecting hydrogen gas over a wide concentration range, and has little baseline shift.
[0079] [Experimental Example 1] <SEM Observation and GI-XRD Measurement of Cu Nanowires / CuO Nanowires> The surface layer of about 1 μm is SiO 2 A Si substrate (length: 15 mm, width: 15 mm, thickness: 0.525 mm) made of was prepared, and samples according to the following invention example and comparative examples 1 to 3 were fabricated.
[0080] (Comparative Example 1) Cu nanowires were formed on a substrate using an electron beam lithography system (ELS-7500EX, manufactured by Elionix). Specifically, an electron beam resist (ZEP-520A, manufactured by Zeon Corporation) was applied to the substrate by spin coating and dried to form a resist film. A mask pattern of a predetermined shape was then drawn using the EBL system. A Cu layer was then formed by electron beam evaporation. After that, a lift-off process was performed to peel off the mask pattern, and Cu nanowires (line width: 68.5 nm, thickness: 50 nm) were formed on the substrate. Figure 6 shows a top-view SEM image of the Cu nanowires.
[0081] (Comparative Example 2) Cu nanowires (line width: 68.5 nm, thickness: 50 nm) were formed on a substrate using the same procedure as in Comparative Example 1. Subsequently, using an RTA apparatus (MILA-5000UHV, manufactured by Advance Riko Co., Ltd.), a first heat treatment was performed in which the ambient temperature was raised to 100°C at a heating rate of 10°C / min in an atmosphere containing 3% by volume of hydrogen and the remainder argon gas, and then the temperature was maintained at 100°C for 60 minutes. A top-view SEM image of the Cu nanowires is shown in Figure 6. The line width increased to 73.4 nm.
[0082] (Comparative Example 3) Cu nanowires (line width: 90 nm, thickness: 50 nm) were formed on a substrate using the same procedure as in Comparative Example 1. Subsequently, using an RTA apparatus (MILA-5000UHV, manufactured by Advance Riko Co., Ltd.), the ambient temperature was raised to 400°C at a heating rate of 2°C / min in a dry air atmosphere, and then a second heat treatment was performed in which the temperature was held at 400°C for 60 minutes, converting the Cu nanowires into CuO nanowires. A top-view SEM image of the CuO nanowires is shown in Figure 6. The line width increased to 96 nm.
[0083] (Invention Example) Cu nanowires (line width: 90 nm, thickness: 50 nm) were formed on a substrate using the same procedure as in Comparative Example 1. A first heat treatment was then performed under the same conditions as in Comparative Example 2, followed by a second heat treatment under the same conditions as in Comparative Example 3. Figure 6 shows a top-view SEM image of the CuO nanowires. The line width increased to 154 nm.
[0084] The samples of Comparative Example 2, Comparative Example 3, and Inventive Example, in which the Cu nanowires were subjected to either or both of the first and second heat treatments, were evaluated by grazing incidence X-ray diffraction (GI-XRD). Specifically, X-ray synchrotron radiation with an energy of 11.7 keV was incident horizontally on the sample, and measurements were performed while the sample was oscillated by 1.2 to 3.2 degrees. Diffraction images were captured using a curved imaging plate, and two-dimensional diffraction patterns were obtained.
[0085] The GI-XRD patterns obtained for each sample are shown in Figure 7. As is clear from Figure 7, in Comparative Example 2, superlattice reflections due to Cu (111) and Cu (200) were confirmed, confirming that the nanowires were made of Cu. 2 The superlattice reflection caused by O(111) is presumed to be due to partial oxidation of the nanowire surface. In Comparative Example 3 and the inventive example, superlattice reflection caused mainly by CuO(002) and CuO(111) was confirmed, and a small amount of superlattice reflection caused by CuO(-202) was also observed, confirming that the nanowires were made of CuO.
[0086] FIG. 8 shows the X-ray diffraction spectrum of the inventive example, and FIG. 9 shows a comparison of the X-ray diffraction spectra of Comparative Example 3 and the inventive example. In FIG. 8, strong peaks of CuO(002) and CuO(111) and a weak peak of CuO(-202) are observed, confirming that the nanowires in the inventive example are composed of CuO. FIG. 9 also shows that the inventive example, which underwent only the first heat treatment (hydrogen annealing) and the second heat treatment (oxygen annealing), had larger peak intensities and peak areas of CuO(002) and CuO(111) than Comparative Example 3, which underwent only the second heat treatment (oxygen annealing). This indicates that the crystallinity of the CuO constituting the nanowires was improved by performing the first heat treatment (hydrogen annealing).
[0087] <Fabrication of Hydrogen Gas Sensor> A hydrogen gas sensor was fabricated according to the conditions of Comparative Example 3 and the invention example. First, a surface layer of about 1 μm was formed of SiO 2A Si substrate (length: 15 mm, width: 15 mm, thickness: 0.525 mm) made of
[0088] Using an electron beam lithography system (ELS-7500EX manufactured by Elionix), the first island portion and first connecting portion of the first pad electrode, and the second island portion and second connecting portion of the second pad electrode were fabricated on the substrate. Specifically, an electron beam resist (ZEP-520A manufactured by Zeon Corporation) was applied to the substrate by spin coating and dried to form a resist film. A mask pattern of a predetermined shape was then drawn using the EBL system. A Ti layer (thickness: 3 nm) and a Pt layer (thickness: 10 nm) were then formed on the substrate by electron beam evaporation. The mask pattern was then removed through a lift-off process to form the above-mentioned portions on the substrate. The dimensions of the first island portion and second island portion were 120 μm in length and 1 μm in width. The dimensions of the first connecting portion and second connecting portion were 75 μm in length and 1 μm in width.
[0089] Next, a first pad portion and a second pad portion (length: 150 μm × width: 150 μm) each consisting of a Ti layer (thickness: 5 nm) and a Pt layer (thickness: 40 nm) thereon were fabricated using a general photolithography method, thus fabricating a first pad electrode and a second pad electrode on the substrate.
[0090] Next, for the purpose of cleaning the first pad electrode and the second pad electrode, an RTA apparatus (MILA-5000UHV, manufactured by Advance Riko Co., Ltd.) was used to perform a heat treatment in which the atmospheric temperature was raised to 550°C at a heating rate of 10°C / min in an atmosphere containing 3% by volume of hydrogen and the remainder being argon gas, and then the temperature was held at 550°C for 5 minutes.
[0091] Next, using an electron beam lithography system (ELS-7500EX manufactured by Elionix), 25 Cu nanowires were formed in parallel on the substrate so as to connect the first pad electrode and the second pad electrode. Specifically, an electron beam resist (ZEP-520A manufactured by Zeon Corporation) was applied to the substrate by spin coating and dried to form a resist film. A mask pattern of a predetermined shape was then drawn using the EBL system. A Cu layer was then formed by electron beam evaporation. After that, a lift-off process was performed to peel off the mask pattern, and Cu nanowires (length: 125 nm, line width: 90 nm, thickness: 50 nm) were formed on the substrate.
[0092] Next, in Comparative Example 3, only the second heat treatment (oxidation annealing) was performed, while in the inventive example, the first heat treatment (hydrogen annealing) and the second heat treatment (oxidation annealing) were performed. In the inventive example, the CuO nanowires had a length of 125 nm, a line width of 154 nm, a thickness of 67 nm, and a number of 25. In Comparative Example 3, the CuO nanowires had a length of 125 nm, a line width of 96 nm, a thickness of 53 nm, and a number of 25.
[0093] <Hydrogen gas detection test> Ar + 3% H 2 The hydrogen gas sensor was placed in a measurement chamber where the hydrogen concentration could be controlled using gas. A semiconductor parameter analyzer system (Keysight Technologies, Inc., B1500A) was used to apply a constant voltage (V = 4 V) between the first and second pad electrodes. A hydrogen gas detection test was conducted based on the change in current detected between the first and second pad electrodes, and various sensor characteristics were evaluated. The circuit diagram is shown in Figure 2. The operating temperature was 300°C. Measurements were performed under atmospheric pressure. Dry air (flow rate: 3 SLM) was introduced into the measurement chamber as a carrier gas. The hydrogen concentration with hydrogen gas ON was changed in steps to 5000 ppm, 2500 ppm, 1000 ppm, 500 ppm, 250 ppm, and 100 ppm.
[0094] FIG. 10 is a graph showing the change in resistance value over time during a hydrogen gas detection test in Comparative Example 3 and an Inventive Example. The resistance value is a value converted from the actually measured current value. In Comparative Example 3, the sensitivity is good, but the response and recovery characteristics are insufficient, and there is also a baseline shift. In contrast, in the Inventive Example, the sensitivity is at a sufficient level, the response and recovery characteristics are very excellent, and there is also no baseline shift. From the change in resistance value when the hydrogen concentration is 2500 ppm, the sensitivity S and response time t res90 , and recovery time t rec90 The results are shown in Table 1 below. From these results, it can be seen that the first heat treatment (hydrogen annealing) contributes greatly to improving the sensor characteristics.
[0095]
[0096] The response time t res90 is the time required for the current value to change from "current value when hydrogen gas is ON" to 90% of "current value when hydrogen gas is ON - saturated current value after hydrogen gas is ON", and the shorter this is, the faster the response. rec90 is the time required for the current value to change from "current value when hydrogen gas is off" to 90% of "current value when hydrogen gas is off - saturated current value after hydrogen gas is off", and the shorter this is, the faster the recovery. Also, sensitivity S = R / R base (where R is the saturated resistance value after hydrogen gas is turned on, R base : resistance value when hydrogen gas is ON).
[0097] [Experimental Example 2] A hydrogen gas sensor was fabricated under the same conditions as in Experimental Example 1, as well as in accordance with the conditions of Comparative Example 3 and the inventive example, except that the specifications of the Cu nanowires were length: 200 nm, line width: 90 nm, thickness: 50 nm, number: 100) and the temperature rise rate of the second heat treatment was 10°C / min. In the inventive example, the CuO nanowires had length: 200 nm, line width: 91 nm, thickness: 50 nm, number: 100), and in Comparative Example 3, the CuO nanowires had length: 200 nm, line width: 94 nm, thickness: 50 nm, number: 100.
[0098] A hydrogen gas detection test was conducted in the same manner as in Experimental Example 1. However, in this experiment, the hydrogen concentration when hydrogen gas was turned on was 500 ppm, and this was repeated three times. Figure 11 is a graph showing the change in the resistance change rate over time during the hydrogen gas detection test in Comparative Example 3 and the invention example. Note that the resistance change rate on the vertical axis is R base is the resistance value when hydrogen gas is first turned on, and R is the resistance value at a certain time. The results in Fig. 11 show the same tendency as in Fig. 10 of Experimental Example 1. The sensitivity S and response time t res90 , and recovery time t rec90 The results are shown in Table 2 below.
[0099]
[0100] [Experimental Example 3] A hydrogen gas sensor was fabricated according to the conditions of the invention example of Experimental Example 1, except that the specifications of the Cu nanowires were length: 125 nm, line width: 80 nm, thickness: 50 nm, and number: 25. The specifications of the CuO nanowires were length: 125 nm, line width: 144 nm, thickness: 65 nm, and number: 25.
[0101] 12A shows a cross-sectional SEM image of a Cu nanowire taken perpendicular to its extension direction before the first heat treatment (hydrogen annealing). At this stage, the cross-sectional shape is almost rectangular, and polycrystals consisting of aggregated Cu crystal grains can be seen, with no voids present inside.
[0102] 12(B) is a cross-sectional SEM image of the Cu nanowires after the first heat treatment (hydrogen annealing), taken perpendicular to the extension direction. The hydrogen annealing increases the grain size of the Cu crystal grains and improves their crystallinity. The internal stress on the Cu nanowires results in a semi-cylindrical cross-sectional shape. However, even at this stage, no voids exist within the Cu nanowires.
[0103] Figure 12(C) is a cross-sectional SEM image perpendicular to the extension direction of CuO nanowires formed by subjecting Cu nanowires to a second heat treatment (oxidation annealing). The CuO nanowires are polycrystalline, consisting of aggregates of CuO crystal grains, and contain voids. The density of Cu is 8.93 g / cm. 3 and the density of CuO is 6.3 g / cm 3Considering this, when Cu is oxidized to CuO, the volume is calculated to expand by 1.77 times, but the cross-sectional area is 1.9 times larger in Figures 12(B) to 12(C). This also suggests that voids are formed inside the CuO nanowire. This CuO nanowire has an extremely unique structure in which there are voids between the CuO crystal grains, but the crystal grains are in contact with each other and maintain the nanowire structure. This is presumably because the Cu nanowire is fixed to the substrate rather than being free-standing, and therefore cannot expand in the direction of wire extension, and expansion in the width direction of the wire is also significantly restricted.
[0104] Fig. 13(A) is a top-view SEM image of a portion of the device before the first heat treatment (hydrogen annealing), and Fig. 13(B) is a top-view SEM image of a portion of the device (hydrogen gas sensor) after the second heat treatment (oxidation annealing). Fig. 14 is a top-view SEM image of CuO nanowires. It can be seen that multiple CuO nanowires are formed in parallel to connect the first and second pad electrodes.
[0105] The average porosity of the CuO nanowires was determined using the method described above and was found to be 16%. For reference, one of the cross-sectional SEM images used to determine the average porosity is shown in Figure 15. As is clear from Figure 15, CuO crystal grains and voids are clearly distinguishable visually.
[0106] [Experimental Example 4] A hydrogen gas sensor was fabricated according to the conditions of the invention example of Experimental Example 1, except that the specifications of the Cu nanowires were length: 125 nm, line width: 90 nm, thickness: 50 nm, and number: 25. The specifications of the CuO nanowires were length: 125 nm, line width: 154 nm, thickness: 67 nm, and number: 25.
[0107] A hydrogen gas detection test was conducted in the same manner as in Experimental Example 1. Figure 16 is a graph showing the change in the rate of resistance change over time during the hydrogen gas detection test. As described above, the hydrogen gas sensor according to the present invention had excellent sensor characteristics, including high sensitivity, excellent response and recovery characteristics, and almost no baseline shift. The sensor characteristics at each hydrogen concentration were as shown in Table 3 below.
[0108]
[0109] 17 is a graph showing the relationship between hydrogen gas concentration and sensitivity. H2 Logarithm of and sensitivity {(R-R base ) / R base} has a linear relationship. When applied to the following formula, β = 0.8508 and A = 0.0014. When the first heat treatment (hydrogen annealing) was not performed, β = 0.5 to 0.6. Compared to the case without the first heat treatment, the β value is improved when the first heat treatment is performed, which shows that performing the first heat treatment significantly improves sensitivity to changes in hydrogen partial pressure.
[0110] Similar to Experimental Example 1, a hydrogen gas detection test was conducted at extremely low hydrogen gas concentrations. The hydrogen concentration when hydrogen gas was ON was changed in steps to 25 ppb, 10 ppb, and 5 ppb. Figure 18 is a graph showing the change in the resistance change rate over time during the hydrogen gas detection test. As is clear from Figure 18, the hydrogen gas sensor of the present invention was able to detect hydrogen gas at an extremely low concentration of 5 ppb.
[0111] [Experimental Example 5] A hydrogen gas sensor was fabricated according to the conditions of the invention example of Experimental Example 1, except that the specifications of the Cu nanowires were: length (4 conditions: 700 nm, 400 nm, 300 nm, 200 nm), line width (90 nm), thickness (50 nm), and number (100). The specifications of the CuO nanowires were: length (4 conditions above), line width (93 nm), thickness (52 nm), and number (100).
[0112] A hydrogen gas detection test was carried out in the same manner as in Experimental Example 1. Fig. 19 is a graph showing the change in resistance value over time during the hydrogen gas detection test. Fig. 20 is a graph showing the relationship between the gap length (wire length) and the response time t res90 and recovery time t rec90 10 is a graph showing the relationship between the wire length and the response and recovery characteristics. Although sufficient sensor characteristics were obtained for all wire lengths, it can be seen that shorter wire lengths provide better response and recovery characteristics.
[0113] 21 is a graph showing the relationship between hydrogen gas concentration and sensitivity. β and A were calculated from this graph by applying the above formula, and the results are shown in Table 4 below. A larger β means a higher S / N ratio.
[0114]
[0115] [Experimental Example 6] A hydrogen gas sensor was fabricated according to the conditions of the invention example of Experimental Example 1, except that the specifications of the Cu nanowires were length: 200 nm, line width: 90 nm, thickness: 50 nm, and number: 100. The specifications of the CuO nanowires were length: 200 nm, line width: 93 nm, thickness: 52 nm, and number: 100.
[0116] A hydrogen gas detection test was conducted in the same manner as in Experimental Example 1. However, in this experiment, the hydrogen concentration when hydrogen gas was ON was 500 ppm, and four applied voltage conditions were used: 1 V, 2 V, 3 V, and 4 V. Figure 22 is a graph showing the change in resistance value over time during the hydrogen gas detection test. From Figure 22, it can be seen that as the applied voltage is increased, the sensitivity decreases, but the response and recovery characteristics become better.
[0117] [Experimental Example 7] A hydrogen gas sensor was fabricated according to the conditions of the invention example of Experimental Example 1, except that the specifications of the Cu nanowires were: length: 125 nm, line width: 90 nm, thickness: 50 nm, number: 25). The specifications of the CuO nanowires were: length: 125 nm, line width: 154 nm, thickness: 67 nm, number: 25.
[0118] A hydrogen gas detection test was conducted in the same manner as in Experimental Example 1. However, in this experiment, an experiment was first conducted using dry air (humidity: 0%) as the carrier gas, followed by an experiment using humidified air (humidity: 50%) as the carrier gas. The hydrogen concentration when hydrogen gas was ON was changed in stages to 5000 ppm, 2500 ppm, and 1000 ppm. Figure 23 is a graph showing the change in the resistance change rate over time during the hydrogen gas detection test. As can be seen from Figure 23, although the sensitivity and response / recovery characteristics were somewhat degraded when using humidified air, the effect of humidity was less than that of other reported hydrogen gas sensors.
[0119] Furthermore, after conducting the experiment using humidified air, another experiment was conducted using dry air (humidity: 0%) as the carrier gas, followed by another experiment using humidified air (humidity: 50%) as the carrier gas. Figure 24 is a graph showing the change in resistance over time during the hydrogen gas detection test. In Figure 24, the "after humidity test" shows the results of the second test, and the "before humidity test" shows the results of the first test conducted before the experiment using humidified air (the same as the "dry air" result in Figure 23). This shows that there is no degradation in sensor characteristics due to the influence of humidified air.
[0120] [Experimental Example 8] A hydrogen gas sensor was fabricated according to the conditions of the invention example of Experimental Example 1, except that the specifications of the Cu nanowires were length: 200 nm, line width: 90 nm, thickness: 50 nm, and number: 100. The specifications of the CuO nanowires were length: 200 nm, line width: 93 nm, thickness: 52 nm, and number: 100.
[0121] A hydrogen gas detection test was conducted in the same manner as in Experimental Example 1. However, in this experimental example, the operating temperature was set to three conditions: 250°C, 300°C, and 350°C. The hydrogen concentration when hydrogen gas was turned on was set to 500 ppm. Figure 25 is a graph showing the change in resistance value over time during the hydrogen gas detection test. As is clear from Figure 25, the hydrogen gas sensor of the present invention was able to detect hydrogen gas even at an operating temperature as low as 250°C.
[0122] [Experimental Example 9] A hydrogen gas sensor was fabricated according to the conditions of the invention example of Experimental Example 1, except that the specifications of the Cu nanowires were length: 200 nm, line width: 90 nm, thickness: 50 nm, and number: 100. The specifications of the CuO nanowires were length: 200 nm, line width: 93 nm, thickness: 52 nm, and number: 100.
[0123] A hydrogen gas detection test was conducted in the same manner as in Experimental Example 1. However, in this experimental example, the test was conducted twice: on the day the hydrogen gas sensor was fabricated and after 100 days of storage in an indoor environment from that day. The hydrogen concentration when hydrogen gas was turned on was 5000 ppm. Figure 26 is a graph showing the change in the resistance change rate over time during the hydrogen gas detection test. As is clear from Figure 26, the hydrogen gas sensor of this example of the present invention behaved almost the same as on the first day, even 100 days after fabrication, demonstrating high stability.
[0124] [Experimental Example 10] A hydrogen gas sensor was fabricated according to the conditions of the invention example of Experimental Example 1, except that the specifications of the Cu nanowires were: length: 200 nm, line width: three conditions: 80 nm, 90 nm, and 100 nm, thickness: 50 nm, and number: 100. The specifications of the CuO nanowires were: length: 200 nm, line width: three conditions: 83 nm, 93 nm, and 102 nm, thickness: 52 nm, and number: 100.
[0125] A hydrogen gas detection test was conducted in the same manner as in Experimental Example 1. However, in this experiment, the hydrogen concentration when hydrogen gas was turned on was 500 ppm, and this was repeated three times. Figure 27 is a graph showing the change in resistance value over time during the hydrogen gas detection test. While sufficient sensor characteristics were obtained for all line widths, the highest sensitivity and shortest response time were obtained when the Cu nanowire line width was 90 nm.
[0126] [Experimental Example 11] A hydrogen gas sensor was fabricated according to the conditions of the invention example of Experimental Example 1, except that the specifications of the Cu nanowires were set to four conditions: length: 200 nm, line width: 90 nm, thickness: 50 nm, and number: 100, 200, 300, and 400. The specifications of the CuO nanowires were set to four conditions: length: 200 nm, line width: 150 nm, thickness: 80 nm, and number: 100, 200, 300, and 400.
[0127] A hydrogen gas detection test was conducted in the same manner as in Experimental Example 1. However, in this experiment, the hydrogen concentration when hydrogen gas was turned on was 500 ppm, and this was repeated three times. Figure 28 is a graph showing the change in resistance value over time during the hydrogen gas detection test. Although sufficient sensor characteristics were obtained under all conditions, the fewer the number of nanowires, the higher the sensitivity and the better the response and recovery characteristics.
[0128] The hydrogen gas sensor of the present invention has high sensitivity, excellent response and recovery characteristics, and little baseline shift, and therefore may be applicable to mobile gas sensors and the like.
[0129] 100 Hydrogen gas sensor 10 Substrate 12 First pad electrode 12A First island portion 12B First connecting portion 12C First pad portion 14 Second pad electrode 14A Second island portion 14B Second connecting portion 14C Second pad portion 15 Cu nanowire 15A Crystal grain 15B Amorphous 16 CuO nanowire 16A Crystal grain 16B Void 18 Power source 20 Ammeter 22 Voltmeter G Gap length (length of CuO nanowire) W Line width of CuO nanowire D Thickness of CuO nanowire
Claims
1. A hydrogen gas sensor comprising: a substrate having an insulating surface; a first pad electrode and a second pad electrode formed on the insulating surface of the substrate; and a plurality of CuO nanowires made of CuO formed in parallel on the insulating surface of the substrate so as to connect the first pad electrode and the second pad electrode, wherein each of the plurality of CuO nanowires is polycrystalline with an aggregate of crystal grains made of CuO and has a structure that holds voids internally; and wherein a current is passed between the first pad electrode and the second pad electrode, and hydrogen gas is detected based on a change in the electrical signal detected between the first pad electrode and the second pad electrode.
2. The hydrogen gas sensor according to claim 1, wherein the length of each of the plurality of CuO nanowires is 10 nm or more and 10 μm or less.
3. The hydrogen gas sensor according to claim 1, wherein the line width of each of the plurality of CuO nanowires is 10 nm or more and 200 nm or less.
4. The hydrogen gas sensor according to claim 1, wherein each of the plurality of CuO nanowires has a thickness of 5 nm or more and 100 nm or less.
5. The hydrogen gas sensor according to claim 1, wherein the number of the CuO nanowires is 2 or more and 1,000 or less.
6. A hydrogen gas sensor according to any one of claims 1 to 5, wherein the average porosity in a cross section perpendicular to the extension direction of the CuO nanowires is 5% or more and 30% or less.
7. A hydrogen gas sensor according to any one of claims 1 to 5, wherein some of the crystal grains constituting the CuO nanowire are in surface contact with the insulating surface of the substrate.
8. A hydrogen gas sensor described in any one of claims 1 to 5, wherein, in a cross section perpendicular to the extension direction of the CuO nanowires, the width of the CuO nanowires gradually decreases with increasing distance from the insulating surface of the substrate, or gradually increases and then gradually decreases with increasing distance from the insulating surface of the substrate.
9. The hydrogen gas sensor according to any one of claims 1 to 5, wherein the first pad electrode and the second pad electrode are made of platinum (Pt).
10. The hydrogen gas sensor according to any one of claims 1 to 5, wherein the substrate is any one of a glass substrate, an alumina substrate, a zirconia substrate, and a silicon substrate having a silicon oxide film or a silicon nitride film formed on the surface thereof.
11. A method for manufacturing a hydrogen gas sensor according to claim 1, comprising: a step of preparing a substrate having an insulating surface; a step of forming a first pad electrode and a second pad electrode on the insulating surface of the substrate; a step of forming a plurality of Cu nanowires made of Cu in parallel on the insulating surface of the substrate so as to connect the first pad electrode and the second pad electrode; a first heat treatment step of subjecting the plurality of Cu nanowires to a heat treatment in an atmosphere containing hydrogen and an inert gas; and a second heat treatment step of subjecting the plurality of Cu nanowires to a heat treatment in an atmosphere containing oxygen to convert the plurality of Cu nanowires into a plurality of CuO nanowires made of CuO.
12. A method for manufacturing a hydrogen gas sensor as described in claim 11, wherein the first heat treatment step is performed by raising the ambient temperature to a first temperature of 80°C or higher and 250°C or lower at a heating rate of 1°C / minute or higher and 100°C / second or lower, and then maintaining the first temperature for 1 minute or higher and 100 minutes or lower.
13. A method for manufacturing a hydrogen gas sensor according to claim 11, wherein the second heat treatment step is carried out by raising the ambient temperature to a second temperature of 250°C or higher and 500°C or lower at a heating rate of 0.5°C / min or higher and 20°C / min or lower, and then maintaining the second temperature for 10 minutes or higher and 90 minutes or lower.
14. The method for manufacturing a hydrogen gas sensor according to any one of claims 11 to 13, wherein the length of each of the plurality of Cu nanowires is 10 nm or more and 10 μm or less.
15. The method for manufacturing a hydrogen gas sensor according to any one of claims 11 to 13, wherein the line width of each of the plurality of Cu nanowires is 10 nm or more and 200 nm or less.
16. The method for manufacturing a hydrogen gas sensor according to any one of claims 11 to 13, wherein the thickness of each of the plurality of Cu nanowires is 5 nm or more and 100 nm or less.
17. The method for manufacturing a hydrogen gas sensor according to any one of claims 11 to 13, wherein the number of the Cu nanowires is 2 or more and 1,000 or less.
18. The method for manufacturing a hydrogen gas sensor according to any one of claims 11 to 13, wherein the first pad electrode and the second pad electrode are made of platinum (Pt).
19. The method for manufacturing a hydrogen gas sensor according to any one of claims 11 to 13, wherein the substrate is any one of a glass substrate, an alumina substrate, a zirconia substrate, and a silicon substrate having a silicon oxide film or a silicon nitride film formed on its surface.