Hydrogen gas sensor and method for manufacturing the same

Abstract

Provided is a hydrogen gas sensor capable of detecting hydrogen gas with high sensitivity, excellent response and recovery characteristics, and low power consumption. A hydrogen gas sensor 100 according to the present invention comprises: a substrate 10 having an insulating surface; first and second pad electrodes 12A and 12B formed on the insulating surface of the substrate 10; and a nanowire 14. The nanowire 14 is formed on the insulating surface of the substrate 10 to connect the first pad electrode 12A and the second pad electrode 12B, has a line width of 50-150 nm and a thickness of 10-60 nm, and comprises a hydrogen storage metal. The hydrogen gas sensor 100 causes a current to flow between the first pad electrode 12A and the second pad electrode 12B, and detects hydrogen gas on the basis of changes in the electrical signal detected between the first pad electrode 12A and the second pad electrode 12B.

Description

[Technical Field]

[0001] This invention relates to a hydrogen gas sensor using nanowires made of a specific metal and a method for manufacturing the same. [Background technology]

[0002] Various types of hydrogen gas sensors have been developed that are suitable for applications such as detecting hydrogen gas leaks from various devices that handle hydrogen gas, such as fuel cells, and measuring the hydrogen gas concentration inside such devices. Among these, a hydrogen gas sensor is known in which a hydrogen gas detection unit is sandwiched between a pair of electrodes, and hydrogen gas is detected based on the change in current or resistance detected between the electrodes while a constant voltage is applied between the electrodes.

[0003] As an example of such a hydrogen gas sensor, a hydrogen gas sensor is known that has a hydrogen gas detection section formed by creating a thin film of palladium (Pd) as a hydrogen storage metal, as described in Non-Patent Document 1. [Prior art documents] [Non-patent literature]

[0004] [Non-Patent Document 1] T. Xu and MP Zach, Self-assembled monolayer-enhanced hydrogen sensing with ultrathin palladium films, Appl. Phys. Lett. 86, 203104 (2005) [Overview of the project] [Problems that the invention aims to solve]

[0005] However, conventional hydrogen gas sensors, including those using Pd thin films, had room for improvement in important characteristics for practical application, such as sensitivity, response and recovery speed, and power consumption.

[0006] Therefore, in view of the above problems, an object of the present invention is to provide a hydrogen gas sensor that has high sensitivity, excellent response and recovery characteristics, and can detect hydrogen gas with low power consumption, and a suitable manufacturing method therefor.

Means for Solving the Problems

[0007] In order to solve the above problems, the present inventors conducted intensive studies and obtained the following findings. The present inventors conceived a hydrogen gas sensor using nanowires made of a specific hydrogen storage metal such as palladium (Pd). It was found that by forming the hydrogen gas detection part in the shape of nanowires instead of a thin film such as Pd, sensor characteristics such as sensitivity, response and recovery characteristics, and power consumption are improved. In particular, by setting the line width and thickness of the nanowires within a specific range, the cross-sectional shape perpendicular to the extending direction of the nanowires becomes appropriate. Although the estimation mechanism will be described later, it was found that hydrogen storage and desorption with respect to the nanowires occur easily, and the above sensor characteristics are remarkably improved.

[0008] Based on the above findings, the gist configuration of the present invention completed is as follows. [1] A substrate having an insulating surface, A first pad electrode and a second pad electrode formed on the insulating surface of the substrate, On the insulating surface of the substrate, a nanowire made of a hydrogen storage metal and having a line width of 50 nm or more and 150 nm or less and a thickness of 10 nm or more and 60 nm or less, formed so as to connect the first pad electrode and the second pad electrode, And a hydrogen gas sensor that has a current flowing between the first pad electrode and the second pad electrode and detects hydrogen gas 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 line width of the nanowire is 80 nm or more and 100 nm or less.

[0010] [3] The hydrogen gas sensor according to [1] or [2] above, wherein the thickness of the nanowire is 20 nm or more and 50 nm or less.

[0011] [4] The hydrogen gas sensor according to any one of [1] to [3] above, wherein the length of the nanowire is 10 μm or more and 300 mm or less.

[0012] [5] The hydrogen gas sensor according to [4] above, wherein the length of the nanowire is 0.07 mm or more.

[0013] [6] The hydrogen gas sensor according to [5] above, wherein the length of the nanowire is 0.5 mm or more.

[0014] [7] The hydrogen storage metal is (I) Palladium (Pd), rhodium (Rh), iridium (Ir), ruthenium (Ru), osmium (Os), vanadium (V), titanium (Ti), zirconium (Zr), lanthanum (La), tungsten (W), calcium (Ca), magnesium (Mg), strontium (Sr), barium (Ba), and beryllium (Be) as exothermic metals A that are likely to form stable hydrides, and solid solution alloys of these exothermic alloys A, and (II) AB5-type alloys, AB2-type alloys, AB-type alloys, and A2B-type alloys in which the exothermic metal A is combined with one or more selected from nickel (Ni), iron (Fe), cobalt (Co), manganese (Mn), and zinc (Zn) as endothermic metals B that have no affinity with hydrogen, and The hydrogen gas sensor according to any one of [1] to [6] above, which is one or more selected from.

[0015] [8] The hydrogen gas sensor according to any one of [1] to [6] above, wherein the hydrogen storage alloy is palladium (Pd).

[0016] [9] The hydrogen gas sensor according to [8] above, wherein the palladium constituting the nanowire is polycrystallized.

[0017]

[10] The hydrogen gas sensor according to [9] above, wherein the lattice constant of the palladium constituting the nanowire is 3.925 ± 0.005 Å.

[0018]

[11] 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 nanowires made of a hydrogen storage metal on the insulating surface of the substrate, having a line width of 50 nm to 150 nm and a thickness of 10 nm to 60 nm, so as to connect the first pad electrode and the second pad electrode. A method for manufacturing a hydrogen gas sensor, comprising: having a first pad electrode and a second pad electrode, passing an electric current between them, and detecting hydrogen gas based on a change in an electrical signal detected between the first pad electrode and the second pad electrode.

[0019]

[12] A step of exposing the nanowire to an atmosphere containing hydrogen and an inert gas, Subsequently, the nanowire is subjected to a heat treatment in an atmosphere containing hydrogen and an inert gas. A method for manufacturing a hydrogen gas sensor as described in

[11] above, comprising the above.

[0020]

[13] The method for manufacturing a hydrogen gas sensor according to

[12] above, wherein the heat treatment is an RTA treatment performed at a heat treatment temperature of 350°C or more and 650°C or less.

[0021]

[14] A method for manufacturing a hydrogen gas sensor according to any one of the above

[11] to

[13] , wherein the hydrogen storage alloy is palladium (Pd). [Effects of the Invention]

[0022] The hydrogen gas sensor of the present invention has high sensitivity, excellent response and recovery characteristics, and can detect hydrogen gas with low power consumption. Furthermore, according to the manufacturing method of the hydrogen gas sensor of the present invention, it is possible to manufacture a hydrogen gas sensor that has high sensitivity, excellent response and recovery characteristics, and can detect hydrogen gas with low power consumption. [Brief explanation of the drawing]

[0023] [Figure 1A] This is a schematic perspective view of a hydrogen gas sensor 100 according to one embodiment of the present invention. [Figure 1B] This is a schematic top view of the hydrogen gas sensor 100. [Figure 1C] This is a cross-sectional view II in Figure 1B. [Figure 1D] This is a cross-sectional view taken along line II-II in Figure 1B. [Figure 2] This is a circuit diagram showing an example of measuring sensor characteristics using a hydrogen gas sensor 100 (resistive voltage distribution measurement). [Figure 3] This is a cross-sectional view of the hydrogen gas sensor 100, perpendicular to the direction of extension of the nanowire 14. [Figure 4] This is a top view showing an example of the arrangement of nanowires 14 when the nanowires 14 are long in a hydrogen gas sensor 100 according to one embodiment of the present invention. [Figure 5] Figures (A) to (D) illustrate the manufacturing process of a gas sensor 100 according to one embodiment of the present invention. [Figure 6] These are cross-sectional SEM images perpendicular to the extension direction of nanowires of various line widths in Experimental Example 1. [Figure 7] The graphs above show the change in current value over time for various line widths in Experimental Example 1, and below show the change in resistance rate over time. [Figure 8] This graph shows the relationship between line width and sensitivity (sensitivity due to resistance change rate) at various operating temperatures T, as obtained in Experimental Example 1. [Figure 9] This graph shows the relationship between line width and response time tres50 at various operating temperatures T, obtained in Experimental Example 1. [Figure 10] This graph shows the relationship between line width and recovery time trec50 at various operating temperatures T, obtained in Experimental Example 1. [Figure 11] These graphs show the relationship between line width and the activation energy of the response (hydrogen absorption reaction) and recovery (hydrogen elimination reaction), obtained from the results shown in Figures 9 and 10 in Experimental Example 1. [Figure 12] This is an SEM image showing a portion of a Pd nanowire (length: 7 mm) fabricated using electron beam lithography (EBL) in Experimental Example 2. [Figure 13A] Experimental Example 2 shows a graph illustrating the change in current value over time for various wire lengths. [Figure 13B] Experimental Example 2 shows a graph illustrating the change in current value over time for various wire lengths. [Figure 14] Experimental Example 2 shows a graph illustrating the change in resistance rate over time for various wire lengths. [Figure 15] The graphs above show the relationship between wire length and response time tres50, and below show the relationship between wire length and recovery time trec50, obtained in Experimental Example 2. [Figure 16] Experimental Example 3 shows a graph illustrating the change in the rate of change of voltage over time at various hydrogen gas concentrations. [Figure 17] Experimental Example 4 shows a graph illustrating the change in the rate of voltage change over time at various applied voltages. [Figure 18] Experimental Example 5 shows a graph illustrating the change in voltage rate over time at various operating temperatures. [Figure 19] This graph shows the change in resistance over time in Experiment Example 6, when the carrier gas is dry air (top) and when it is nitrogen (bottom). [Figure 20] This graph shows the change in resistance rate over time in Experimental Example 6, when the carrier gas is dry air and when it is nitrogen. [Figure 21] Experimental Example 7 shows a graph illustrating the change in resistance rate over time for various wire thicknesses. [Figure 22] This graph shows the time-dependent change in resistance rate during repeated introduction of hydrogen gas in Experimental Example 8, with a wire thickness of 30 nm and a wire length of 31 mm. [Figure 23] This graph shows the time-dependent change in resistance rate during repeated introduction of hydrogen gas in Experimental Example 8, with a wire thickness of 10 nm and a wire length of 18 mm. [Figure 24] The images above show a cross-sectional SEM image perpendicular to the direction of nanowire extension in Experimental Example 9, and elemental mapping by EDS of the same region as the cross-sectional SEM image. [Figure 25] The graphs above show the change in resistance over time and the change in resistance rate over time, respectively, at an operating temperature of 24°C in Experiment Example 9. [Figure 26] The graphs above show the change in resistance over time and the change in resistance rate over time, respectively, at an operating temperature of 50°C in Experiment Example 9. [Figure 27] The graphs above show the change in resistance over time and the change in resistance rate over time, respectively, at an operating temperature of 100°C in Experiment Example 9. [Figure 28] The graphs above show the change in resistance over time and the change in resistance rate over time, respectively, at an operating temperature of 150°C in Experiment Example 9. [Figure 29] The graphs above show the change in resistance over time and the change in resistance rate over time, respectively, at an operating temperature of 21°C (room temperature) in Experiment Example 9. [Figure 30] The graphs above show the relationship between operating temperature and response time tres50, and below show the relationship between operating temperature and recovery time trec50, obtained in Experimental Example 9. [Figure 31] This graph shows the relationship between operating temperature and sensitivity (sensitivity based on resistance change rate) obtained in Experiment Example 9. [Figure 32]This graph shows the X-ray diffraction spectra obtained by GI-WAXS measurements under three conditions (Condition 1: No heat treatment, Condition 2: 250°C, 5 minutes annealing, Condition 3: 500°C, RTA treatment) in Experimental Example 10. [Figure 33] The images in Experimental Example 10 show the top surface SEM images of nanowires under three conditions (Condition 3: 500°C, RTA treatment; Condition 4: 400°C, RTA treatment; Condition 5: 600°C, RTA treatment). [Figure 34] This graph shows the change in resistance rate over time under conditions 1-3 of Experimental Example 10. [Figure 35] This graph shows the change in resistance rate over time under conditions 2-5 of Experimental Example 10. [Figure 36] This graph shows the change in resistance rate over time under condition 3 of experimental example 10. [Modes for carrying out the invention]

[0024] (Hydrogen gas sensor) Referring to Figures 1A, 1B, 1C, and 1D, a hydrogen gas sensor 100 according to one embodiment of the present invention comprises a substrate 10, a first pad electrode 12A and a second pad electrode 12B, and a nanowire 14 made of a specific hydrogen storage metal. The first pad electrode 12A and the second pad electrode 12B are formed on the substrate 10. The nanowire 14 is formed on the substrate 10 to connect the first pad electrode 12A and the second pad electrode 12B. The detailed effects will be described later, but in this embodiment, the action of the nanowire 14 improves sensor characteristics such as sensitivity, response and recovery characteristics, and power consumption.

[0025] [Mechanism for detecting hydrogen gas] The hydrogen gas sensor 100 detects hydrogen gas by passing an electric current between the first pad electrode 12A and the second pad electrode 12B and observing the change in the electrical signal detected between them. For example, as shown in Figure 1B, a power supply 18 and an ammeter 20 are connected in series between the first pad electrode 12A and the second pad electrode 12B, and a voltmeter 22 is connected in parallel. In this case, the power supply 18 applies a constant voltage measurable by the voltmeter 22 between the first pad electrode 12A and the second pad electrode 12B, and the ammeter 20 detects the change in current between the first and second pad electrodes. Based on the detected change in current, hydrogen gas is detected. Alternatively, the gas can be detected by passing a constant current between the first pad electrode 12A and the second pad electrode 12B and observing the change in voltage detected between them. Furthermore, instead of detecting the current or voltage changes described above, gas can also be detected based on the change in resistance detected between the first pad electrode 12A and the second pad electrode 12B. In other words, the "electrical signal" described above refers to current, voltage, or resistance. In this embodiment, since the gas sensor circuit can be configured with two terminals, a hydrogen gas sensor can be constructed without increasing the amount of wiring or circuitry.

[0026] Furthermore, when measuring sensor characteristics, resistance-based voltage measurement can also be performed as shown in Figure 2. In Figure 2, in addition to the hydrogen gas sensor 100, power supply 18, and ammeter 20, a resistor 24 with a known resistance value is connected in series, and a voltmeter 22 is connected in parallel with this resistor 24. In this case, with a constant voltage applied between the first pad electrode (not shown) and the second pad electrode (not shown) of the hydrogen gas sensor 100 by the power supply 18, the voltmeter 22 detects the change in voltage across the resistor 24, and hydrogen gas is detected based on the detected change in voltage. In this configuration as well, the change in voltage across the hydrogen gas sensor 100 (i.e., between the first pad electrode and the second pad electrode) is indirectly detected by the voltmeter 22.

[0027] Furthermore, the voltmeter 22 connected in parallel with the resistor 24 in Figure 2 may be changed to be connected in parallel with the hydrogen gas sensor 100. In this case, with a constant voltage applied between the first pad electrode (not shown) and the second pad electrode (not shown) of the hydrogen gas sensor 100 by the power supply 18, the voltmeter 22 directly detects the change in voltage across the hydrogen gas sensor 100 (i.e., between the first pad electrode and the second pad electrode), and hydrogen gas is detected based on the detected change in voltage.

[0028] [substrate] The substrate 10 supports the first pad electrode 12A and the second pad electrode 12B, as well as the nanowire 14 which serves as the hydrogen gas detection unit. The substrate 10 is not particularly limited as long as it has an insulating surface, but for example, insulating substrates such as glass substrates, alumina substrates, and zirconia substrates, or silicon substrates with a silicon oxide film or silicon nitride film formed on the surface can be used. The shape and dimensions of the substrate 10 are not particularly limited, but when using a substrate with a rectangular main surface shape, the dimensions can be, for example, in the range of length: 1 to 300 mm × width: 1 to 300 mm × thickness: 0.1 to 1.2 mm.

[0029] [First pad electrode and second pad electrode] The first pad electrode 12A and the second pad electrode 12B are a pair of electrodes necessary for supplying current to the nanowire 14 and for detecting changes in the electrical signal corresponding to changes in the hydrogen gas concentration. The shape and dimensions of the first pad electrode 12A and the second pad electrode 12B are not particularly limited as long as they are formed on the insulating surface of the substrate 10, but if the shape of the main surface is rectangular, the dimensions can be, for example, in the range of length: 30 to 1000 μm × width: 30 to 1000 μm × thickness: 5 to 500 nm. The metal constituting the first pad electrode 12A and the second pad electrode 12B is not particularly limited and may be any metal, but for example, it can be one or more selected from platinum, gold, and palladium. From the viewpoint of process simplicity, it may be the same type of metal as the nanowire 14 described later.

[0030] [Nanowire] The nanowire 14 is an element constituting the gas detection unit, formed on the insulating surface of the substrate 10 to connect the first pad electrode 12A and the second pad electrode 12B. It is essential that the nanowire 14 is made of a hydrogen storage metal. Preferably, the hydrogen storage metal is one or more selected from the elements of palladium (Pd), rhodium (Rh), iridium (Ir), ruthenium (Ru), osmium (Os), vanadium (V), titanium (Ti), zirconium (Zr), lanthanum (La), tungsten (W), calcium (Ca), magnesium (Mg), strontium (Sr), barium (Ba), and beryllium (Be), as well as solid solution alloys of these heat-generating alloys A, which readily form stable hydrides. Furthermore, the hydrogen storage metal is preferably one or more selected from AB5 type alloys, AB2 type alloys, AB type alloys, and A2B type alloys, which combine the exothermic metal A with one or more endothermic metal B selected from nickel (Ni), iron (Fe), cobalt (Co), manganese (Mn), and zinc (Zn), which do not have an affinity for hydrogen. Examples of AB5 type alloys include LaNi5 and CaNi5, examples of AB2 type alloys include MaZn2 and ZrNi2, examples of AB type alloys include TiFe and TiNi, and examples of A2B type alloys include Mg2Ni and Ca2Fe. Among these, it is more preferable that the hydrogen storage metal be palladium (Pd). As a result, when hydrogen is absorbed into the nanowire 14, the hydrogen gas sensor 100 responds (detects hydrogen gas), and when hydrogen is desorbed from the nanowire 14, the hydrogen gas sensor 100 recovers (detects the off state of hydrogen gas).

[0031] [Wire width of nanowires] Referring to Figures 1A-D and Figure 3, it is important that the line width W of the nanowire 14 is between 50 nm and 150 nm. This allows for the remarkable effect of high sensitivity and excellent response and recovery characteristics. The inventors believe that the mechanism that produces these effects is as follows. As shown in Figure 3, the nanowire 14 has curved portions 14B1 and 14B2 at both ends in the width direction, and the upper surface connecting the pair of curved portions 14B1 and 14B2 is a flat portion 14A. The internal stress ΔP acting on the nanowire 14 is given by the following equation (1). ΔP = γ(1 / R1 + 1 / R2) ... (1) Here, γ is the surface tension of the metal constituting the nanowire 14, R1 is the radius of curvature of the nanowire 14 in the width direction, and R2 is the radius of curvature of the nanowire 14 in the extension direction. Since the nanowire 14 has no curvature in its extension direction, R2 is infinite, and the 1 / R2 term in equation (1) can be considered zero. Then, the internal stress ΔP increases as the radius of curvature R1 in the width direction decreases. As shown in Figure 3, the cross-sectional shape perpendicular to the extension direction of the nanowire 14 is a so-called semicylindrical shape, so the internal stress is distributed in the lateral direction (width direction) of the cross-section. For example, the surface tension of a clean surface of palladium is approximately 1000 mN / m. If the radius of curvature R1 of the curved portions 14B1 and 14B2 of the nanowire 14 is, for example, 5 to 30 nm, the internal stress on the curved portions 14B1 and 14B2 of the nanowire 14 is estimated to be around 200 to 30 MPa. In nanowires, it is not possible to maintain a clean palladium surface, but because palladium has a strong catalytic effect, there are surfaces that adsorb and dissociate hydrogen. However, the surface tension of the surface of the nanowire with a radius of curvature is nearly two orders of magnitude smaller than the above estimate, and is generally around 0.6 to 1 MPa. Here, in the PCT (P: pressure, C: hydrogen storage amount, T: temperature) characteristic curve of a typical hydrogen storage metal, the range of hydrogen equilibrium pressure from 0.6 to 1 MPa is a region where the change in hydrogen equilibrium pressure in response to a change in hydrogen storage amount is small at room temperature, for both adsorption (exothermic reaction) and desorption (endothermic reaction). Therefore, hydrogen is easily absorbed and desorbed in the curved portions 14B1 and 14B2 of the nanowire 14, where the internal stress is approximately 0.6 to 1 MPa. In contrast, the flat portion 14A of the nanowire 14 (especially its central portion) is in a state almost equivalent to a thin film, so there is no internal stress, and only atmospheric pressure (approximately 0.1 MPa) is applied. In the PCT (P: pressure, C: hydrogen storage amount, T: temperature) characteristic curve of a typical hydrogen storage metal, the range where the hydrogen equilibrium pressure is around 0.1 MPa is a region where the change in hydrogen equilibrium pressure is large in response to a change in hydrogen storage amount at room temperature (a region where the hydrogen storage amount hardly changes in response to a change in hydrogen equilibrium pressure) for both absorption (exothermic reaction) and desorption (endothermic reaction). Therefore, hydrogen is not easily absorbed and desorbed in the flat portion 14A of the nanowire 14 (especially its central portion).By using a nanowire 14 for the hydrogen detection section, the ratio of curved portions to flat portions can be increased compared to a thin film. This contributes to improving the sensor characteristics. From another perspective, even when the curved portions 14B1 and 14B2 of the nanowire 14 are completely free of absorbed hydrogen, their shape creates an internal stress that makes it appear as if a certain amount of hydrogen is already absorbed. In other words, the amount of hydrogen that can be absorbed in the curved portions 14B1 and 14B2 is less than that in the thin film or the flat portion 14A. Therefore, it is thought that in the curved portions 14B1 and 14B2, hydrogen is absorbed in locations where it is easily absorbed and released. The hydrogen storage process in palladium involves the steps of [1] H2 adsorption, [2] dissociation of H2 into atomic hydrogen 2H, and [3] diffusion of atomic hydrogen within the palladium. The hydrogen desorption process in palladium involves the steps of [A] diffusion of atomic hydrogen stored within the palladium to the surface, [B] covalent bonding of atomic hydrogen to form H2, and [C] desorption of H2 from the surface. In both storage and desorption, the diffusion of atomic hydrogen is the rate-limiting process. Here, if the internal stress is high, the diffusion of atomic hydrogen within the palladium is suppressed. Therefore, in the curved sections 14B1 and 14B2, the diffusion of atomic hydrogen is suppressed, and as a result, hydrogen is thought to be stored in locations where it is easier to add and remove hydrogen. Now, let's return to the topic of the nanowire 14's line width W. When the line width W is between 50 nm and 150 nm, the balance between the curved portions 14B1 and 14B2 and the flat portion 14A is optimal, making it easier for hydrogen to be absorbed and desorbed from the nanowire 14, and significantly improving the sensor characteristics. If the line width W is less than 50 nm, the proportion of the curved portions 14B1 and 14B2, which have high internal stress, becomes too large compared to the flat portion 14. Therefore, hydrogen is not easily absorbed into the nanowire 14 in the first place, and sufficient sensitivity cannot be obtained. Also, if the line width W exceeds 150 nm, the proportion of the flat portion 14A, which is only subjected to atmospheric pressure, becomes too large compared to the curved portions 14B1 and 14B2. In this case, it becomes close to a thin film, so atomic hydrogen diffuses easily within the nanowire 14, and hydrogen is absorbed even in areas where it is difficult to absorb and desorb, making hydrogen absorption and desorption difficult. Therefore, sufficient response and recovery characteristics cannot be obtained. When the line width W is between 50 nm and 150 nm, the balance between the curved portions 14B1 and 14B2 and the flat portion 14A is optimized, and internal stress is applied to the curved portions 14B1 and 14B2 and their vicinity. In this case, since the diffusion of atomic hydrogen is suppressed, small amounts of hydrogen are absorbed in locations where it is easy to absorb and remove, hydrogen absorption and desorption to the nanowire 14 becomes easier, and the sensor characteristics are significantly improved. From this viewpoint, the line width W is set to 50 nm or more, preferably 80 nm or more. Also, the line width W is set to 150 nm or less, preferably 100 nm or less.

[0032] In this specification, the line width W of the nanowire 14 refers to the width at the bottom of the nanowire 14 (the interface with the substrate 10), as shown in Figure 3. The line width W of the nanowire 14 is determined by observing a cross section perpendicular to the extension direction of the nanowire 14 at 10 equally spaced locations (e.g., 0.1 to 100 μm intervals) using SEM, and adopting the average value of the line widths obtained from each SEM image. In the case of the nanowire arrangement shown in Figures 4 and 12, 10 equally spaced cross sections can be observed with a single SEM image.

[0033] [Nanowire data] Referring to Figure 3, it is important that the thickness D of the nanowire 14 is between 10 nm and 60 nm. If the thickness D is less than 10 nm, the curved portion becomes narrower due to the thinness, and the volume of palladium decreases, resulting in reduced sensitivity and making it impossible to obtain good response and recovery characteristics. Therefore, the thickness D should be 10 nm or more, preferably 20 nm or more. On the other hand, if the thickness D exceeds 60 nm, the diffusion distance of atomic hydrogen in the thickness direction becomes longer, making it impossible to obtain good response and recovery characteristics. Therefore, the thickness D should be 60 nm or less, preferably 50 nm or less.

[0034] In this specification, the thickness D of the nanowire 14 refers to the distance from the bottom (interface with the substrate 10) to the top (flat portion 14A) of the nanowire 14, as shown in Figure 3. Furthermore, the thickness D of the nanowire 14 is determined by observing a cross-section perpendicular to the extension direction of the nanowire 14 at 10 equally spaced locations (e.g., 0.1 to 100 μm intervals) using SEM, and adopting the average value of the thicknesses obtained from each SEM image. Note that in the case of the nanowire arrangement shown in Figures 4 and 12, 10 equally spaced cross-sections can be observed with a single SEM image.

[0035] [Length of nanowire] Referring to Figure 1B, the length L of the nanowire 14 is not particularly limited, but from the viewpoint of further improving the response / recovery characteristics and sensitivity to hydrogen gas, it is preferable to have a length of 10 μm or more. In order to obtain response / recovery characteristics even at room temperature, the length L is more preferably 0.07 mm or more, and even more preferably 0.5 mm or more. Since a longer length L is preferable from the viewpoint of sensor characteristics, there is no particular upper limit to the length L. However, from the viewpoint of manufacturing process constraints and sensor size, it is preferable to have a length of 300 mm or less. One of the features of the hydrogen gas sensor 100 of this embodiment is that the length L of the nanowire 14 is large. When the nanowire 14 is made longer, the size of the sensor can be reduced by arranging the nanowire 14 in a zigzag pattern on the substrate 10, as shown in Figure 4. In this specification, the length L of the nanowire 14 refers to the length of a single nanowire connecting the first pad electrode 12A and the second pad electrode 12B. The length L of the nanowire 14 can be measured by observing the upper surface of the hydrogen gas sensor 100 with an SEM.

[0036] [Crystalline state of hydrogen storage alloy constituting nanowires] For example, in nanowires 14 immediately after formation by electron beam deposition or sputtering (As deposition), the crystalline state of the hydrogen storage alloy is a mixture of amorphous and polycrystalline states. As will be explained in detail in the manufacturing method and experimental example 10 below, if the hydrogen storage alloy constituting the nanowire 14 is palladium, then by heat treatment of the nanowire 14 while hydrogen is absorbed into the palladium, crystallization of the palladium progresses, and the palladium becomes polycrystalline. Thus, it is preferable for the palladium constituting the nanowire to be polycrystalline because it further improves the response and recovery characteristics. Note that "palladium is polycrystalline" means that the crystallization of the palladium has progressed, the amorphous state has disappeared, and the crystalline state of the palladium has become polycrystalline. The polycrystalline state of palladium can be determined based on the intensity and area of ​​the peak attributable to Pd(111) in the X-ray diffraction spectrum obtained by grazing incident wide-angle X-ray scattering (GI-WAXS) measurement. Both peak intensity and peak area are significantly increased in the polycrystalline state after heat treatment compared to the amorphous state before heat treatment.

[0037] Furthermore, as will be explained in detail in the manufacturing method and experimental example 10 below, when the heat treatment performed on palladium with absorbed hydrogen is RTA (Rapid Thermal Annealing), it was found that crystallization proceeds by heat treatment with an increased palladium lattice constant, and moreover, the increased lattice constant is maintained even after heat treatment. It was also found that using nanowires 14 with such an increased palladium lattice constant significantly improves response and recovery characteristics.

[0038] The α-phase of palladium is a hydrogen solid solution phase (Pd+H), and exists when x < 0.02, where x is the atomic ratio of H to Pd (PdHx). In the α-phase, hydrogen atoms are randomly dissolved within the lattice, and the lattice constant of the α-phase is 3.90 ± 0.02 Å. In contrast, the β-phase of palladium is a hydride phase (Pd-H), and exists when x > 0.6. In the β-phase, hydrogen atoms occupy octahedral sites (O-sites) within the lattice, and the lattice constant of the β-phase is 4.02 ± 0.02 Å. That is, the lattice constant of the β-phase is 2-3% larger than that of the α-phase. Note that a region where the α-phase and β-phase coexist (0.02 ≤ x ≤ 0.6) exists near room temperature. According to the inventors' research, when RTA treatment is performed on palladium with hydrogen absorbed into it, the lattice constant of the palladium constituting the nanowire 14 becomes 3.925 ± 0.005 Å. In other words, although it cannot be said that the palladium constituting the nanowire 14 underwent a phase transition to the β phase, it reached a value close to the upper limit of the α phase lattice constant.

[0039] The inventors believe the mechanism by which response and recovery characteristics are significantly improved in this case is as follows: When RTA treatment is performed on palladium while it has absorbed hydrogen, crystallization proceeds due to heat treatment while the lattice constant of the palladium is increased. Here, the nanowire 14 is subjected to a large internal stress due to its radius of curvature. Therefore, even after heat treatment (i.e., in a situation where hydrogen is absent), the state in which the lattice constant has increased is maintained due to the large internal stress. For example, in the case of thin film palladium, even if the lattice constant increases due to hydrogen absorption, the lattice constant returns to its original state once the hydrogen is no longer present. Considering this, the fact that the nanowire 14 maintains a lattice expansion state even in a situation without a hydrogen atmosphere after heat treatment due to the large internal stress is a unique phenomenon in the present invention. And, in this lattice expansion state, atomic hydrogen can be easily absorbed and removed, which is thought to be the reason why the response and recovery characteristics are significantly improved.

[0040] The lattice constants of the palladium constituting the nanowire 14 can be calculated from the peaks attributable to Pd(111) and Pd(200) in the X-ray diffraction spectrum obtained by GI-WAXS measurement, and it will be confirmed that the lattice constants of both are the same.

[0041] [effect] As described above, the hydrogen gas sensor 100 of this embodiment has high sensitivity, excellent response and recovery characteristics, and can detect hydrogen gas with low power consumption. Furthermore, the hydrogen gas sensor 100 of this embodiment has the effect of being able to detect hydrogen gas even at low operating temperatures (for example, at room temperature). In addition, since the hydrogen gas sensor 100 of this embodiment can detect hydrogen gas even at low hydrogen gas concentrations, it also has the effect of having a wide range of detectable hydrogen gas concentrations.

[0042] (Method of manufacturing a hydrogen gas sensor) A method for manufacturing a hydrogen gas sensor 100 according to one embodiment of the present invention will be described with reference to Figures 5(A) to 5(D). The method for manufacturing the hydrogen gas sensor 100 according to this embodiment includes the steps of preparing a substrate 10, forming a first pad electrode 12A and a second pad electrode 12B, and forming nanowires 14. The hydrogen gas sensor 100 according to this embodiment described above can be manufactured by this method.

[0043] First, prepare the substrate 10. Details of the substrate 10 are as previously described. Next, an example of the process for forming the first pad electrode 12A, the second pad electrode 12B, and the nanowire 14 will be explained.

[0044] Referring to Figure 5(A), a resist film 30 is formed on the substrate 10. The resist film 30 can be formed by coating the substrate 10 with a resist composition for electron beam exposure and drying it. The coating method is not particularly limited, but a spin coating method can be suitably used. The thickness of the resist film 30 should be set appropriately so as to be thicker than the thickness of the nanowire 14 to be formed (or thicker than both the nanowire 14 and the pad electrodes 12A and 12B when they are formed together). After spin coating, the resist film 30 may be annealed under appropriate conditions to efficiently volatilize the solvent and improve the density of the resist film 30.

[0045] Next, as shown in Figure 5(B), the resist film 30 is developed to form a mask pattern 32 of a predetermined shape. The mask pattern 32 is fabricated by exposing the resist film 30 by electron beam lithography and then developing it. The shape and dimensions of the portion of the mask pattern 32 where the resist film is removed and the substrate 10 is exposed correspond to the shape and dimensions (line width W and length L) of the nanowire 14 to be formed. Here, the shape of the recesses in the mask pattern 32 can be adjusted by appropriately controlling the type of resist composition, the thickness of the resist film 30 to be formed, and the dose of the electron beam to be irradiated. This makes it possible to adjust the shape and size (i.e., radius of curvature, length in the width direction, length in the height direction, etc.) of the curved portions 14B1 and 14B2 at both ends in the width direction of the nanowire 14. As shown in Figures 5(B) to (D), when the nanowires 14 and pad electrodes 12A and 12B are formed at once, the shape of the portion of the mask pattern 32 where the resist film is removed and the substrate 10 is exposed corresponds to the shape and dimensions of the nanowires 14 and pad electrodes 12A and 12B to be formed.

[0046] Next, as shown in Figure 5(C), a metal film 34 is formed, for example, by electron beam deposition or sputtering. At this time, the first portion 34A of the metal film is formed on the mask pattern 32, and the second portion 34B of the metal film is formed on the substrate 10 that is exposed after the resist film is removed from the mask pattern 32. Preferably, the metal film 34 includes a Ti layer, Cr layer, or Ta layer (adhesive layer) with a thickness of about 1 to 5 nm, and a layer formed thereon made of a hydrogen-absorbing metal (e.g., Pd) that constitutes the nanowires 14 and pad electrodes 12A and 12B. The Ti layer, Cr layer, or Ta layer functions as an adhesive layer for bonding the layers made of the metal that constitutes the nanowires 14 and pad electrodes 12A and 12B to the substrate 10.

[0047] Next, as shown in Figure 5(D), the nanowires 14 and pad electrodes 12A and 12B are formed on the substrate 10 by a lift-off process, which involves peeling off the mask pattern 32 and removing the first portion 34A of the metal film formed thereon. In this example, the nanowires 14 and pad electrodes 12A and 12B were formed simultaneously, as shown in Figures 5(B) to (D). However, it is also possible to form only the nanowires 14 in the above process and then separately form the pad electrodes 12A and 12B using, for example, a general photolithography method. Note that the Ti layer, Cr layer, or Ta layer is formed on the substrate 10 as an adhesive layer for bonding the nanowires 14, but during the deposition process of the metal constituting the nanowires 14, Ti, Cr, or Ta diffuse into the metal layer, and most of the Ti layer, Cr layer, or Ta layer disappears. Alternatively, because the thickness of the adhesive layer is very thin, the adhesive layer is almost invisible in the SEM image. Although Ti, Cr, or Ta remains in island-like formations in some areas between the substrate 10 and the nanowire 14, the substrate 10 and the nanowire 14 are in direct contact in most areas.

[0048] After this, the nanowire 14 may be subjected to heat treatment as an optional step. Heat treatment changes the cross-sectional shape perpendicular to the direction of extension of the nanowire 14, increasing the radius of curvature and resulting in a nanowire with a so-called incomplete circular cross-section, where a part of the circle is missing. As a result, internal stress is applied over a wide area of ​​the nanowire, further improvement of sensor characteristics can be achieved. Preferably, the heat treatment conditions are an inert gas atmosphere such as hydrogen-containing Ar, a heat treatment temperature (maximum temperature of the atmosphere during heat treatment) of 250 to 400°C, and a holding time at the heat treatment temperature of 1 to 90 minutes.

[0049] Furthermore, it is preferable to perform an optional step of exposing the nanowire 14 to an atmosphere containing hydrogen and an inert gas, and then subjecting the nanowire 14 to heat treatment in the atmosphere containing hydrogen and an inert gas. This allows for polycrystallization of palladium when the hydrogen storage alloy constituting the nanowire is palladium, and as described above, the response and recovery characteristics are further improved.

[0050] [Exposure process] The exposure process involves exposing the nanowire 14 to an atmosphere containing hydrogen, with the remainder being an inert gas and optionally any unavoidable impurity gases. This process allows hydrogen to be absorbed into the palladium constituting the nanowire 14. The hydrogen content is preferably 1 to 5 volume percent, and the remainder being an inert gas can be one or more selected from argon (Ar), helium (He), and neon (Ne). The exposure time is preferably 1 minute or more from the viewpoint of allowing a sufficient amount of hydrogen to be absorbed into the palladium, and preferably 10 minutes or less from the point at which the amount of hydrogen absorbed into the palladium becomes saturated.

[0051] [Heat treatment process] The heat treatment process can be carried out under atmospheric pressure in an atmosphere containing hydrogen, with the remainder being an inert gas and optionally any unavoidable impurity gases, following the exposure process. This heat treatment process promotes the crystallization of palladium. The hydrogen content is preferably 1 to 5 volume percent, and the remainder of the inert gas can be one or more selected from argon (Ar), helium (He), and neon (Ne).

[0052] The heat treatment process may be carried out using a general heat treatment furnace, but it is preferable to perform the process using an RTA (Restoration Therapy) apparatus. As described above, the RTA process increases the lattice constant of palladium, and crystallization proceeds through heat treatment, and moreover, the increased lattice constant is maintained even after heat treatment. From the viewpoint of obtaining this effect to the fullest, the heat treatment temperature (the maximum temperature of the atmosphere during heat treatment) is preferably 350°C or higher and 650°C or lower. It is more preferable that the heat treatment temperature be 400°C or higher. It is even more preferable that the heat treatment temperature be 600°C or lower. From the viewpoint of obtaining the above effect to the fullest, it is preferable that the heating rate from the temperature at the start of heat treatment to the heat treatment temperature be 2°C / second or higher and 200°C / second or lower. Holding at the heat treatment temperature is not essential. The holding time at the heat treatment temperature can be 0 minutes or more and 10 minutes or less.

[0053] Through the above process, the gas sensor 100 can be manufactured. [Examples]

[0054] [Experimental Example 1: Effect of Line Width W] <Fabrication of a hydrogen gas sensor> A hydrogen gas sensor was fabricated using the following procedure. First, a Si substrate (length: 15 mm × width: 15 mm × thickness: 0.6 mm) with a surface layer of approximately 1 μm of SiO2 was prepared.

[0055] Pd nanowires were formed on a substrate using an electron beam lithography (EBL) system (ELS-7500EX, manufactured by Elionix). Specifically, an electron beam resist ZEP-520A was applied to the substrate by spin coating to form a resist film. Then, a mask pattern of a predetermined shape was drawn using the EBL system. Subsequently, a Ti layer (thickness: 3 nm) and a Pd layer (thickness: 30 nm) were formed on the substrate by electron beam evaporation. After a lift-off process to remove the mask pattern, Pd nanowires (with five conditions for line width W: 40 nm (comparative example), 50 nm (inventive example), 80 nm (inventive example), 100 nm (inventive example), and 200 nm (comparative example), wire thickness D: 30 nm, wire length L: 10 μm) were formed on the substrate. Figure 6 shows cross-sectional SEM images perpendicular to the extension direction of nanowires of various line widths. The Ti layer is formed on the substrate as an adhesive layer to bond the Pd nanowires. However, because the Ti layer is very thin, most of the Ti layer is lost as it diffuses into the Pd layer during the deposition process. While some Ti remains as islands between the substrate and the Pd nanowires, in most areas the substrate and Pd nanowires are in direct contact.

[0056] Next, using a general photolithography method, a first pad electrode and a second pad electrode (75 μm × 75 μm) were fabricated, each consisting of a Ti layer (thickness: 5 nm) and a Pt layer (thickness: 40 nm) on top of it. In this way, a hydrogen gas sensor using Pd nanowires was fabricated.

[0057] <Hydrogen gas detection test> The hydrogen gas sensor was placed in a measurement chamber where the hydrogen concentration could be controlled using Ar+3%H2 gas. Using a semiconductor parameter analyzer system (Keysight Technologies B1500A), a constant voltage (V=50mV) was applied between the first and second pad electrodes. Tests were conducted to detect hydrogen gas 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 1B. The operating temperature T was set to 50°C. Measurements were performed under atmospheric pressure, and N2 gas (flow rate: 1 SLM) was introduced as a carrier gas into the measurement chamber. The hydrogen concentration when the hydrogen gas was ON was set to 1% (10000 ppm). The hydrogen gas was switched ON / OFF three times. Specifically, it was turned ON at 0 seconds, OFF at 300 seconds, ON at 900 seconds, OFF at 1200 seconds, ON at 1800 seconds, and OFF at 2100 seconds, and measurements were performed up to 3000 seconds.

[0058] <Linewidth dependence of response recovery characteristics> Figure 7 (top) is a graph showing the change in current value over time for various line widths. Figure 7 (bottom) is a graph showing the change in resistance rate over time, converted from current value to resistance value. Here, with respect to ΔR / R0 on the vertical axis, R0 is the resistance value at 0 seconds, and ΔR is the difference between the resistance value at a certain time and R0. For all line widths, a decrease in current (increase in resistance) was observed when hydrogen was ON, and an increase in current (decrease in resistance) was observed when hydrogen was OFF. When hydrogen was ON, it is thought that the volume expansion due to the diffusion of atomic hydrogen into palladium increased scattering in electron conduction, resulting in an increase in resistance.

[0059] <Linewidth dependence of sensitivity at various operating temperatures> The same tests were conducted with operating temperatures T set to 60°C, 70°C, 80°C, 90°C, 100°C, and 110°C. Figure 8 shows the relationship between linewidth and sensitivity (sensitivity based on resistance change rate) at various operating temperatures T. Note that the sensitivity ΔR / R0 on the vertical axis of Figure 8 is the average value of ΔR / R0 over three OFF periods (300 seconds, 1200 seconds, and 2100 seconds) when R0 is fixed as the resistance value at 0 seconds. As shown in Figure 8, the sensitivity was maximized when the line width W was in the range of 80 to 100 nm. When the line width W is less than 50 nm, the cross-sectional area perpendicular to the direction of nanowire extension decreases, resulting in a sharp decrease in sensitivity. In particular, when the line width W is 40 nm, the proportion of the curved surface with high internal stress becomes too large compared to the flat portion of the nanowire, resulting in internal stress being applied to the entire nanowire. This makes it difficult for atomic hydrogen to penetrate the nanowire, causing a sharp drop in sensitivity. When the line width W is 200 nm, the sensitivity decreases slightly compared to when the line width W is 100 nm, even though the cross-sectional area perpendicular to the direction of extension of the nanowire increases. The inventors believe this is due to the following mechanism: The increase in cross-sectional area is due to an increase in the flat portion of the nanowire, where no internal stress is applied and only atmospheric pressure is present. Therefore, it becomes closer to a thin film state, resulting in a slight decrease in sensitivity.

[0060] <Response time t at various operating temperatures res50 Line width dependence > Figure 9 shows the line width and response time t at various operating temperatures T. res50 This shows the relationship with response time t. res50 This represents the time required for the current value to change by 50% of the difference between the current value when hydrogen gas is ON and the current value when hydrogen gas is OFF, from the current value when hydrogen gas is ON. The shorter this value, the faster the response. Note that the vertical axis in Figure 9 represents the time taken for each of the three ON / OFF cycles. res50 This is the average value. From Figure 9, response time t res50 The line width W was shortest in the range of 80-100 nm. When the line width W is less than 50 nm, the line width approaches the thickness of the nanowire. As a result, the flat portion of the nanowire's cross-sectional shape decreases, and the proportion of curved surfaces with a small radius of curvature increases. Consequently, the proportion of the cross-sectional area subjected to internal stress is larger than when the line width is wider. This large internal stress in the curved surface area is applied to the entire nanowire, and because the internal stress is close to the hydrogen equilibrium pressure during hydrogen absorption, hydrogen is not easily absorbed, and the response time becomes longer. When the line width W is 200 nm, the flat portion of the nanowire increases. In this flat portion, there is no internal stress due to the shape of the nanowire, so hydrogen diffuses and continues to be absorbed in the deeper part of the flat portion, resulting in a longer response time. The higher the operating temperature, the shorter the response time. When the operating temperature is high, the hydrogen equilibrium pressure for hydrogen storage becomes high, making it difficult for hydrogen to be stored. Since only a small amount of hydrogen can enter, the response time becomes short. This is consistent with the fact that the sensitivity decreases as the operating temperature increases.

[0061] <Recovery time t at various operating temperatures rec50 Line width dependence> Fig. 10 shows the relationship between the line width and the recovery time t at various operating temperatures T. rec50 The recovery time t rec50 is the time required for the current value to change by 50% from the "current value when hydrogen gas is off" to "the current value when hydrogen gas is on next time - the current value when hydrogen gas is off". The shorter this time, the faster the recovery. Note that the vertical axis in Fig. 10 is the average value of each t rec50 for three ON / OFF cycles. From Fig. 10, the recovery time t rec50 was the shortest in the range where the line width W was 50 - 80 nm. When the line width W is less than 50 nm, since the line width approaches the thickness of the nanowire, the flat part in the cross-sectional shape of the nanowire decreases, and the ratio of the curved surface part with a small radius of curvature increases. Therefore, the ratio of the cross-sectional area subjected to internal stress becomes larger than when the line width is wide. The large internal stress in this curved surface part acts on the entire nanowire, and since the internal stress is close to the hydrogen equilibrium pressure during hydrogen release, it is difficult for hydrogen to desorb, resulting in a long recovery time. When the operating temperature is 80°C or lower and the line width W exceeds 80 nm, the recovery time becomes long. This is thought to be because the ratio of the hydrogen storage amount in the flat part of the nanowire increases with the increase in the line width, so hydrogen is stored deep in the flat part and it is difficult to recover. Conversely, it suggests that the nanowire structure has better recovery characteristics than the flat film structure. On the other hand, when the operating temperature is 90°C or higher and the line width W exceeds 80 nm, the recovery time becomes short. This is thought to be because as the temperature rises, the hydrogen equilibrium pressure during absorption becomes high, and in the flat part of the nanowire, the hydrogen storage amount is originally small and it is easy to desorb. For line widths W of 100 nm or more, the recovery time was shortened at higher operating temperatures. This is thought to be because higher temperatures increase the hydrogen equilibrium pressure for hydrogen desorption, making it easier for hydrogen to be desorbed. Furthermore, when the line width W was 80 nm or less and the operating temperature was 100°C or higher, the recovery time increased with decreasing line width. Especially when the line width was less than 50 nm, hydrogen was difficult to absorb due to internal stresses caused by the shape, and even if it was absorbed, it was difficult to desorb. It is expected that the internal stresses caused by the shape are greater than the increase in hydrogen equilibrium pressure for desorption due to rising temperature.

[0062] <Linewidth dependence of activation energy for response and recovery> The activation energies for response and recovery were determined using Arrhenius plots. First, from the results in Figure 9, for each line width, the horizontal axis is 1000 / T(K -1 ), with Log(1 / t) on the vertical axis. res50 The plots were analyzed, and the slope of the straight line obtained from each plot using the least squares method was determined. From this slope, the activation energy Ea of the response (hydrogen storage reaction) was calculated. Similarly, from the results in Figure 10, for each line width, the horizontal axis is 1000 / T(K -1 ), with Log(1 / t) on the vertical axis. rec50 The plots were analyzed, and the slope of the straight line for each plot was determined using the least squares method. From this slope, the activation energy Ea for recovery (hydrogen elimination reaction) was calculated.

[0063] Figure 11 shows the relationship between line width and the activation energy of the response (hydrogen absorption reaction) and recovery (hydrogen detachment reaction), obtained from the results shown in Figures 9 and 10. The activation energy of the response increased as the line width increased up to a line width W of 100 nm, and remained constant beyond a line width W of 100 nm. The activation energy of the recovery increased as the line width W increased. Furthermore, the activation energy of the response was about 3-4 kcal / mol higher than the activation energy of the recovery. This is thought to be because, as the line width narrows, hydrogen is less likely to be absorbed due to internal stresses caused by the shape, resulting in a smaller amount of absorbed hydrogen and a lower activation energy for the absorbed hydrogen.

[0064] [Experimental Example 2: Effect of Wire Length L] A hydrogen gas sensor was fabricated in the same manner as in Experimental Example 1, with the line width W fixed at 80 nm and the wire thickness D at 30 nm, and the wire length L set to five conditions: 0.07 mm, 0.7 mm, 7 mm, 14 mm, and 21 mm. In this experiment, in order to increase the wire length L, the nanowire arrangement shown in Figure 4 was adopted for the four conditions where the wire length was 0.7 mm or longer. The length of each row of wires was set to 0.07 mm (70 μm), and the nanowires were arranged in a region of 70 μm × 70 μm with the longest wire length. As a representative example, an SEM image showing a portion of a Pd nanowire (length: 7 mm) is shown in Figure 12.

[0065] A hydrogen gas detection test was conducted in the same manner as in Experimental Example 1. However, in this experiment, the circuit diagram shown in Figure 2 was adopted, and resistance-based voltage measurement was performed. The operating temperature T was set to room temperature of 24°C (297K). The applied voltage V was set to 1V. Measurements were performed under atmospheric pressure, and N2 gas (flow rate: 1 SLM) was introduced as a carrier gas into the measurement chamber. The hydrogen concentration when the hydrogen gas was turned ON was set to 1% (10000 ppm). The hydrogen gas was turned ON for 100 seconds and OFF for 700 seconds, and measurements were performed up to 1300 seconds.

[0066] Figures 13A and 13B are graphs showing the change in current value over time for various wire lengths. The current value was calculated from the measured voltage across the resistive part. Figure 14 shows the change in resistance rate over time, converted from the current value to the resistance value. Here, the vertical axis is R / R base Regarding R base R is the resistance value at 100 seconds (just before hydrogen gas is turned ON), and R is the resistance value at a specific time. From Figure 14, it can be seen that the sensitivity (sensitivity due to the rate of resistance change) increases as the wire length increases. In this experiment, the operating temperature T was set to room temperature. By increasing the wire length, it was possible to observe the response and recovery characteristics at room temperature.

[0067] Figure 15 shows the relationship between wire length and response time t. res50 The graph (above) shows the relationship between wire length and recovery time t. rec50The graph below shows the relationship with response time t. res50 and recovery time t rec50 In both cases, an overall trend can be observed where the length of the wire increases as the length of the wire decreases.

[0068] [Experimental Example 3: Effect of Hydrogen Concentration] A hydrogen gas sensor was fabricated in the same manner as in Experimental Example 1, with a line width W of 80 nm, a wire thickness D of 30 nm, and a wire length L of 14 mm. In this experiment, due to the longer wire length L, the nanowire arrangement shown in Figure 4 was adopted.

[0069] A hydrogen gas detection test was conducted in the same manner as in Experimental Example 1. However, in this experiment, the circuit diagram shown in Figure 2 was adopted, and resistance-based voltage measurement was performed. The operating temperature T was set to room temperature of 21°C (294K). The applied voltage V was set to 1V. Measurements were performed under atmospheric pressure, and N2 gas (flow rate: 1 SLM) was introduced as a carrier gas into the measurement chamber. The hydrogen concentration when the hydrogen gas was turned ON was set to four conditions: 10,000 ppm (1%), 6,000 ppm (0.6%), 1,400 ppm (0.14%), and 300 ppm (0.03%). The hydrogen gas was turned ON for 100 seconds and OFF for 400 seconds, and measurements were performed up to 700 seconds.

[0070] Figure 16 is a graph showing the change in the rate of change of voltage over time at various hydrogen gas concentrations. Here, the vertical axis is V / V base Regarding V base V is the voltage value at 100 seconds (just before hydrogen gas is turned ON), and V is the voltage value at a specific time. As shown in Figure 16, the sensitivity (sensitivity due to the rate of change of voltage) increases as the hydrogen concentration increases, and sufficient sensitivity could be obtained at hydrogen concentrations of 1400 ppm or higher.

[0071] [Experimental Example 4: Effect of Applied Voltage] A hydrogen gas sensor was fabricated in the same manner as in Experimental Example 1, with a line width W of 80 nm, a wire thickness D of 30 nm, and a wire length L of 14 mm. In this experiment, due to the longer wire length L, the nanowire arrangement shown in Figure 4 was adopted.

[0072] A hydrogen gas detection test was conducted in the same manner as in Experimental Example 1. However, in this experiment, the circuit diagram shown in Figure 2 was adopted, and resistance-based voltage measurement was performed. The operating temperature T was set to room temperature of 21°C (294K). The applied voltage V was set to four conditions: 10V, 1V, 0.5V, and 0.1V. Measurements were performed under atmospheric pressure, and N2 gas (flow rate: 1 SLM) was introduced as a carrier gas into the measurement chamber. The hydrogen concentration when the hydrogen gas was turned ON was 1400 ppm (0.14%). The hydrogen gas was turned ON for 100 seconds and OFF for 400 seconds, and measurements were performed up to 700 seconds.

[0073] Figure 17 is a graph showing the change in the rate of change of voltage over time at various applied voltages. Here, the vertical axis is V / V base Regarding V base V is the voltage value at 100 seconds (just before hydrogen gas is turned ON), and V is the voltage value at a specific time. From Figure 17, the sensitivity (sensitivity due to the rate of change of voltage) increases as the applied voltage increases, and sufficient sensitivity was obtained when the applied voltage was 0.5V or higher. The power consumption corresponding to an applied voltage of 0.5V is 0.8μW. In other words, the hydrogen gas sensor in this experimental example can be operated with very low power consumption on the order of μW.

[0074] [Experimental Example 5: Effect of Operating Temperature] A hydrogen gas sensor was fabricated in the same manner as in Experimental Example 1, with a line width W of 80 nm, a wire thickness D of 30 nm, and a wire length L of 14 mm. In this experiment, due to the longer wire length L, the nanowire arrangement shown in Figure 4 was adopted.

[0075] A hydrogen gas detection test was conducted in the same manner as in Experimental Example 1. However, in this experiment, the circuit diagram shown in Figure 2 was adopted, and resistance-based voltage measurement was performed. The operating temperature T was set to four conditions: room temperature of 21°C (294K), 50°C (323K), 75°C (348K), and 100°C (373K). The applied voltage V was set to 1V. Measurements were performed under atmospheric pressure, and N2 gas (flow rate: 1 SLM) was introduced as a carrier gas into the measurement chamber. The hydrogen concentration when the hydrogen gas was turned ON was set to 6000 ppm (0.6%). The hydrogen gas was turned ON for 100 seconds and OFF for 400 seconds, and measurements were performed up to 700 seconds.

[0076] Figure 18 is a graph showing the change in voltage rate over time at various operating temperatures. Here, the vertical axis is V / V base Regarding V base V is the voltage value at 100 seconds (just before hydrogen gas is turned ON), and V is the voltage value at a specific time. From Figure 18, it can be seen that the hydrogen gas sensor in this experimental example can operate even at room temperature.

[0077] [Experimental Example 6: Effect of Carrier Gas] A hydrogen gas sensor was fabricated in the same manner as in Experimental Example 1, with a line width W of 80 nm, a wire thickness D of 30 nm, and a wire length L of 7 mm. In this experiment, due to the longer wire length L, the nanowire arrangement shown in Figure 4 was adopted.

[0078] A hydrogen gas detection test was conducted in the same manner as in Experimental Example 1. However, in this experiment, the circuit diagram shown in Figure 2 was adopted, and resistance-based voltage measurement was performed. The operating temperature T was set to room temperature of 24°C (297K). The applied voltage V was set to 1V. Measurements were performed under atmospheric pressure, and two conditions were introduced into the measurement chamber as carrier gases: N2 gas (flow rate: 500 sccm) and dry air (flow rate: 500 sccm). The hydrogen concentration when the hydrogen gas was ON was set to 10,000 ppm (1%). The hydrogen gas was switched ON / OFF four times. Specifically, under the dry air condition, it was turned ON at 100 seconds, OFF at 400 seconds, ON at 1000 seconds, OFF at 1300 seconds, ON at 1900 seconds, OFF at 2200 seconds, ON at 2800 seconds, and OFF at 3100 seconds, with measurements being taken up to 3700 seconds. Under nitrogen conditions, the system was switched ON at 150 seconds, OFF at 400 seconds, ON at 1000 seconds, OFF at 1300 seconds, ON at 1900 seconds, OFF at 2200 seconds, ON at 2800 seconds, and OFF at 3180 seconds, with measurements taken up to 3700 seconds.

[0079] Figure 19 is a graph showing the change in resistance over time when the carrier gas is dry air (top) and nitrogen (bottom). The resistance value on the vertical axis is the resistance value of the hydrogen gas sensor, calculated from the measured voltage across the resistive part. Figure 20 is a graph showing the change in resistance rate over time when the carrier gas is dry air and nitrogen. Here, the vertical axis is R / Rbase Regarding R base R is the resistance value at 100 seconds (just before the first ON of hydrogen gas), and R is the resistance value at a specific time. From Figures 19 and 20, it was found that sufficient sensitivity (sensitivity based on the rate of change of resistance) can be obtained whether the carrier gas is dry air or nitrogen.

[0080] [Experimental Example 7: Effect of Wire Thickness D] A hydrogen gas sensor was fabricated in the same manner as in Experimental Example 1, with a line width W of 80 nm, wire thickness D of 10 nm and 30 nm, and wire length L of 7 mm. In this experiment, due to the long wire length L, the nanowire arrangement shown in Figure 4 was adopted.

[0081] A hydrogen gas detection test was conducted in the same manner as in Experimental Example 1. However, in this experiment, the circuit diagram shown in Figure 2 was adopted, and resistance-based voltage measurement was performed. The operating temperature T was set to room temperature of 24°C (297K). The applied voltage V was set to 1V. Measurements were performed under atmospheric pressure, and N2 gas (flow rate: 1 SLM) was introduced as a carrier gas into the measurement chamber. The hydrogen concentration when the hydrogen gas was ON was set to 6000 pm (0.6%). The hydrogen gas was switched ON / OFF four times. Specifically, it was turned ON at 100 seconds, OFF at 400 seconds, ON at 1000 seconds, OFF at 1300 seconds, ON at 1900 seconds, OFF at 2200 seconds, ON at 2800 seconds, and OFF at 3100 seconds, with measurements being taken up to 3700 seconds.

[0082] Figure 21 is a graph showing the change in resistance rate over time for various wire thicknesses. Here, the vertical axis is R / R base Regarding R base R is the resistance value at 100 seconds (just before the first ON of hydrogen gas), and R is the resistance value at a specific time. The resistance value of the hydrogen gas sensor was calculated from the measured voltage across the resistive part. As shown in Figure 21, sufficient sensitivity (sensitivity based on the rate of resistance change) was obtained even with a wire thickness of 10 nm, and even higher sensitivity was obtained with a wire thickness of 30 nm.

[0083] [Experimental Example 8: Repeated Introduction of Hydrogen Gas] With a line width W of 80 nm, hydrogen gas sensors were fabricated in the same manner as in Experimental Example 1, using two conditions: "wire thickness D: 30 nm / wire length L: 31 mm" and "wire thickness D: 10 nm / wire length L: 18 mm". In this experiment, due to the long wire length L, the nanowire arrangement shown in Figure 4 was adopted.

[0084] A hydrogen gas detection test was conducted in the same manner as in Experimental Example 1. However, in this experiment, the circuit diagram shown in Figure 2 was adopted, and resistance-based voltage measurement was performed. The operating temperature T was set to 100°C (373K). The applied voltage V was set to 1V. The measurement was performed under atmospheric pressure, and N2 gas (flow rate: 10 SLM) was introduced as a carrier gas into the measurement chamber. The hydrogen gas was switched ON / OFF four times. Specifically, it was turned ON at 300 seconds, OFF at 600 seconds, ON at 900 seconds, OFF at 1200 seconds, ON at 1500 seconds, OFF at 1800 seconds, ON at 2100 seconds, and OFF at 2400 seconds, with measurements being taken up to 3300 seconds. The hydrogen concentration when the hydrogen gas was ON was 30 ppm for the first time, 150 ppm for the second time, 750 ppm for the third time, and 1500 ppm for the fourth time.

[0085] Figure 22 is a graph showing the change in resistance rate over time during repeated introduction of hydrogen gas with a wire thickness of 30 nm and a wire length of 31 mm. Figure 23 is a graph showing the change in resistance rate over time during repeated introduction of hydrogen gas with a wire thickness of 10 nm and a wire length of 18 mm. Here, the vertical axis is R / R base Regarding R base R is the resistance value at 300 seconds (just before the first ON of hydrogen gas), and R is the resistance value at a specific time. The resistance value of the hydrogen gas sensor was calculated from the measured voltage across the resistive part. From Figures 22 and 23, we were able to obtain sensitivity changes that depend on the hydrogen concentration. The hydrogen gas sensor shown in Figure 22 was able to detect hydrogen gas with a concentration of 30 ppm at an operating temperature of 373 K and an applied voltage of 1 V.

[0086] [Experimental Example 9: Effects of Heat Treatment on Nanowires] A hydrogen gas sensor was fabricated in the same manner as in Experimental Example 1, with a line width W of 80 nm, a wire thickness D of 30 nm, and a wire length L of 7 mm. In this experiment, due to the long wire length L, the nanowire arrangement shown in Figure 4 was adopted. After the lift-off process, the nanowires were heat-treated. The heat treatment was performed using an infrared lamp annealing apparatus (Advance Riko Co., Ltd., desktop lamp heating device MILA-5000) with an Ar / H2 (3 vol%) mixed gas flowing through it, under atmospheric pressure, at a heat treatment temperature of 250°C, and for a holding time of 5 minutes. Figure 24 shows a cross-sectional SEM image perpendicular to the direction of nanowire extension (top) and elemental mapping by dispersive X-ray spectroscopy (EDS) of the same region as the cross-sectional SEM image (bottom). As the nanowires are arranged in a zigzag pattern as shown in Figure 4, the cross-sections of the nanowires exist periodically. The EDS elemental mapping shows the intensity of the L line of Pd, with the bright areas being palladium. The nanowire has a semi-circular cross-sectional shape, consisting of curved ends and a flat section. In the relevant SEM image, the radius of curvature is larger after heat treatment compared to Figure 6, and the area proportion enclosed by the curved portion increases. Therefore, it is thought that the proportion of the cross-sectional area subjected to internal stress is larger compared to before heat treatment.

[0087] A hydrogen gas detection test was conducted in the same manner as in Experimental Example 1. However, in this experiment, the circuit diagram shown in Figure 2 was adopted, and resistance-based voltage measurement was performed. The operating temperature T was set to four conditions: 24°C, 50°C, 100°C, and 150°C. The applied voltage V was set to 1V. Measurements were performed under atmospheric pressure, and N2 gas (flow rate: 1 SLM) was introduced as a carrier gas into the measurement chamber. The hydrogen concentration when the hydrogen gas was ON was 3333 pm (0.33%). The hydrogen gas was switched ON / OFF four times. Specifically, it was turned ON at 100 seconds, OFF at 400 seconds, ON at 1000 seconds, OFF at 1300 seconds, ON at 1900 seconds, OFF at 2200 seconds, ON at 2800 seconds, and OFF at 3100 seconds, with measurements being taken up to 3700 seconds. In addition, after the hydrogen gas detection test conducted under the above four operating temperature conditions, the same hydrogen gas detection test was repeated with the operating temperature set to 21°C (room temperature).

[0088] Figures 25-29 show graphs (top) illustrating the change in resistance over time at various operating temperatures, and graphs (bottom) illustrating the change in resistance rate over time. The resistance value on the vertical axis of the top graph is the resistance value of the hydrogen gas sensor, calculated from the measured voltage across the resistive part. The R / R ratio on the vertical axis of the bottom graph... base Regarding R base is the resistance value at 100 seconds (just before the hydrogen gas is first turned ON), and R is the resistance value at a specific time.

[0089] Figure 30 shows the operating temperature and response time t. res50 The graph (above) shows the relationship between operating temperature and recovery time t. rec50 The graph (below) shows the relationship. The vertical axis of the graph (above) represents the value of t for each of the four ON / OFF cycles. res50 This is the average value. The vertical axis of the graph (below) represents the value of each t for the four ON / OFF cycles. rec50 This is the average value. Figure 31 is a graph showing the relationship between operating temperature and sensitivity (sensitivity based on resistance change rate). The vertical axis of Figure 31 is sensitivity R / R base R base When the resistance value is fixed at 100 seconds, the R / R ratio for four OFF cycles (400 seconds, 1300 seconds, 2200 seconds, 3100 seconds) is calculated. base This is the average value.

[0090] Referring to Figure 30, the hydrogen gas sensor that underwent heat treatment showed superior response and recovery characteristics compared to the case without heat treatment (Figures 9 and 10). At each operating temperature, the response time t was superior when heat treatment was performed. res50 and recovery time t rec50 Both were shortened, enabling fast response and recovery. Referring to Figure 31, sensitivity decreased with increasing operating temperature. This trend is consistent with the trend in Figure 8.

[0091] The reason why the response and recovery time is shorter when heat treatment is performed compared to when no heat treatment is performed is that although the radius of curvature increases with heat treatment, the proportion of the area enclosed by the curved surface increases. As a result, the proportion of the cross-sectional area subjected to internal stress becomes larger than when the line width is wider, and the region in which internal stress is increased due to the radius of curvature is more optimized.

[0092] [Experimental Example 10: Effects of annealing on nanowires] A hydrogen gas sensor was fabricated in the same manner as in Experimental Example 1, with a wire width W of 55 nm, a wire thickness D of 30 nm, and a wire length L of 0.7 mm. In this experiment, due to the long wire length L, the nanowire arrangement shown in Figure 4 was adopted. After the lift-off process, the nanowires were subjected to an exposure process and heat treatment under the following conditions.

[0093] <Condition 1> Exposure and heat treatment processes were not performed.

[0094] <Condition 2> The nanowires were subjected to an exposure process in which a mixed gas of Ar / H2 (3 vol%) was supplied for 3 minutes. Subsequently, heat treatment was performed using an infrared lamp annealing apparatus (Advance Riko Co., Ltd., tabletop lamp heating device MILA-5000) under atmospheric pressure, with a heat treatment temperature of 250°C, a heating rate of 10°C / min, and a holding time of 5 minutes, in an Ar / H2 (3 vol%) mixed gas atmosphere.

[0095] <Conditions 3~5> An exposure process was performed on the nanowires by supplying a mixed gas of Ar / H2 (3 vol%) for 3 minutes. Subsequently, RTA treatment was performed using an RTA apparatus (MILA-5000UHV, manufactured by Advance Riko Co., Ltd.) under atmospheric pressure in an Ar / H2 (3 vol%) mixed gas atmosphere, with the following conditions: heat treatment temperature: 500°C (condition 3), 400°C (condition 4), 600°C (condition 5), heating rate: 50°C / second, and holding time: 0 minutes.

[0096] <Crystal state of palladium that makes up nanowires> Samples under conditions 1-3 were subjected to GI-WAXS measurements at beamline BL13XU of SPring-8. The X-ray energy was set to 12.39797043082328 eV. The obtained X-ray diffraction spectra for each sample are shown in Figure 32. As is clear from Figure 32, the intensity and area of ​​the peak attributable to Pd(111) were significantly larger in conditions 2 and 3, which included exposure and heat treatment, compared to condition 1, which did not include exposure and heat treatment. This indicates that the palladium constituting the nanowires was polycrystalline in conditions 2 and 3.

[0097] The lattice constants of palladium, calculated from the peaks attributable to Pd(111) and Pd(200), were 3.909 Å for condition 1, 3.879 Å for condition 2, and 3.923 Å for condition 3. Since the lattice constant of the α-phase of palladium is 3.90 ± 0.02 Å, conditions 1 to 3 can all be considered to be the α-phase. However, in condition 3, where RTA treatment was performed with palladium containing absorbed hydrogen, the lattice expansion state was maintained, and the lattice constant was close to the upper limit of the α-phase lattice constant.

[0098] Figure 33 shows SEM images of the top surface of nanowires in samples under conditions 3 to 5. As shown in Figure 33, in conditions 3 to 5, where RTA treatment was performed after the exposure process, numerous palladium crystal grains can be seen in nanowires with a uniform line width (approximately 80 nm), indicating that the palladium is polycrystalline.

[0099] <Hydrogen gas detection test 1> A hydrogen gas detection test was conducted in the same manner as in Experimental Example 1. The operating temperature T was set to room temperature of 21°C (294K). The applied voltage V was set to 1V. Measurements were taken under atmospheric pressure, and N2 gas (flow rate: 3 SLM) was introduced as a carrier gas into the measurement chamber. The hydrogen gas was switched ON / OFF three times. Specifically, it was turned ON at 100 seconds, OFF at 250 seconds, ON at 400 seconds, OFF at 550 seconds, ON at 700 seconds, and OFF at 850 seconds, with measurements taken up to 1050 seconds. The hydrogen concentration when the hydrogen gas was ON was 500 ppm for the first time, 1000 ppm for the second time, and 5000 ppm for the third time.

[0100] Figure 34 is a graph showing the change in resistance rate over time under conditions 1 to 3. Here, the vertical axis represents R / R base Regarding R base R is the resistance value at 100 seconds (immediately before the first ON of hydrogen gas), and R is the resistance value at a specific time. From Figure 34, conditions 2 and 3, which included the exposure and heat treatment processes, allowed for faster response and recovery than condition 1 (As deposition), which did not include the exposure and heat treatment processes. In particular, condition 3, which included RTA treatment after the exposure process, achieved extremely fast response and recovery, and the response and recovery characteristics were significantly improved.

[0101] Figure 35 is a graph showing the change in resistance rate over time for conditions 2 to 5. From Figure 35, it can be seen that in all conditions 3 to 5, where the RTA heat treatment temperatures were 400°C, 500°C, and 600°C, faster response and recovery were possible compared to condition 2, but condition 3 with a heat treatment temperature of 500°C showed the best response and recovery characteristics.

[0102] <Hydrogen gas detection test 2> A hydrogen gas detection test was conducted using the hydrogen gas sensor under Condition 3, in the same manner as in Experimental Example 1. The operating temperature T was set to room temperature of 21°C (294K). The applied voltage V was set to 1V. Measurements were performed under atmospheric pressure, and N2 gas (flow rate: 3 SLM) was introduced as a carrier gas into the measurement chamber. The hydrogen gas was switched ON / OFF five times. Specifically, it was turned ON at 100 seconds, OFF at 250 seconds, ON at 400 seconds, OFF at 550 seconds, ON at 700 seconds, OFF at 850 seconds, ON at 1000 seconds, OFF at 1150 seconds, ON at 1300 seconds, and OFF at 1450 seconds, with measurements continuing until 1600 seconds. The hydrogen concentration when the hydrogen gas was ON was 297 ppm for the first time, 2727 ppm for the second time, 4286 ppm for the third time, 10000 ppm for the fourth time, and 30000 ppm for the fifth time.

[0103] Figure 36 is a graph showing the change in resistance rate over time. Here, the vertical axis is R / R base Regarding R baseR is the resistance value at 100 seconds (just before the first hydrogen gas ON), and R is the resistance value at a specific time. From Figure 36, the hydrogen gas sensor under condition 3 was able to detect even extremely low concentrations of hydrogen gas, such as 297 ppm, at high speed, and also recovered quickly. [Industrial applicability]

[0104] The hydrogen gas sensor of the present invention has high sensitivity, excellent response and recovery characteristics, and can detect hydrogen gas with low power consumption, making it potentially applicable to mobile gas sensors and the like. [Explanation of symbols]

[0105] 100 Hydrogen gas sensors 10 circuit boards 12A First Pad Electrode 12B Second pad electrode 14 nanowires Flat portion of 14A nanowire 14B1 nanowire curved portion 14B2 nanowire curved portion 18 Power supply 20 ammeter 22 Voltmeter 24 resistors 30 Resist film 32 Mask Patterns 34 Metal film 34A First part of the metal film 34B Second part of the metal film W nanowire line width D nanowire thickness L nanowire length

Claims

1. A substrate having an insulating surface, A first pad electrode and a second pad electrode formed on the insulating surface of the substrate, A nanowire made of a hydrogen-absorbing metal is formed on the insulating surface of the substrate to connect the first pad electrode and the second pad electrode, having a line width of 50 nm to 150 nm and a thickness of 10 nm to 60 nm, with both ends in the width direction being a pair of curved portions and the upper surface connecting the pair of curved portions being a flat portion. A hydrogen gas sensor having a current flow between the first pad electrode and the second pad electrode, and detecting hydrogen gas 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 line width of the nanowire is 80 nm or more and 100 nm or less.

3. The hydrogen gas sensor according to claim 1, wherein the thickness of the nanowire is 20 nm or more and 50 nm or less.

4. The hydrogen gas sensor according to claim 1, wherein the length of the nanowire is 10 μm or more and 300 mm or less.

5. The hydrogen gas sensor according to claim 4, wherein the length of the nanowire is 0.07 mm or more.

6. The hydrogen gas sensor according to claim 5, wherein the length of the nanowire is 0.5 mm or more.

7. The hydrogen storage metal is (I) Palladium (Pd), rhodium (Rh), iridium (Ir), ruthenium (Ru), osmium (Os), vanadium (V), titanium (Ti), zirconium (Zr), lanthanum (La), tungsten (W), calcium (Ca), magnesium (Mg), strontium (Sr), barium (Ba), and beryllium (Be) as elements, and solid solution alloys of these heat-generating metals A, (II) A combination of the exothermic metal A and one or more endothermic metals B selected from nickel (Ni), iron (Fe), cobalt (Co), manganese (Mn), and zinc (Zn) that do not have an affinity for hydrogen. ５ Type alloy, AB ２ Type alloy, AB type alloy, and A ２ Type B alloy and, A hydrogen gas sensor according to any one of claims 1 to 6, wherein it is one or more selected from the following.

8. The hydrogen gas sensor according to any one of claims 1 to 6, wherein the hydrogen storage metal is palladium (Pd).

9. The hydrogen gas sensor according to claim 8, wherein the palladium constituting the nanowire is polycrystalline.

10. The hydrogen gas sensor according to claim 9, wherein the lattice constant of the palladium constituting the nanowire is 3.925 ± 0.005 Å.

11. 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 nanowire made of a hydrogen storage metal on the insulating surface of the substrate, having a line width of 50 nm to 150 nm and a thickness of 10 nm to 60 nm, so as to connect the first pad electrode and the second pad electrode. A step of exposing the nanowire to an atmosphere containing hydrogen and an inert gas, Subsequently, the nanowire is subjected to a heat treatment in an atmosphere containing hydrogen and an inert gas. A method for manufacturing a hydrogen gas sensor, comprising: having a first pad electrode and a second pad electrode, passing an electric current between them, and manufacturing a hydrogen gas sensor that detects hydrogen gas based on a change in an electrical signal detected between the first pad electrode and the second pad electrode.

12. The method for manufacturing a hydrogen gas sensor according to claim 11, wherein the heat treatment is an RTA treatment performed at a heat treatment temperature of 350°C or more and 650°C or less.

13. The method for manufacturing a hydrogen gas sensor according to claim 11 or 12, wherein the hydrogen storage metal is palladium (Pd).

Images