JP2009008476A - Hydrogen gas sensor and its manufacturing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a hydrogen gas sensor responding to a hydrogen gas at the normal temperature with high sensitivity at a high speed, enhanced in reliability, low in cost, used repeatedly and easy to manufacture, and to provide its manufacturing method. <P>SOLUTION: The hydrogen gas sensor 1 includes microelectrodes 2a and 2b, a sensor element which subjects CNT to dielectric migration to crosslink it to the microelectrodes 2a and 2b and Pd particles for modifying the surface of CNT to dissociate and adsorb the hydrogen gas and constituted so as to measure a change in the conductance or resistance across the microelectrodes 2a and 2b to detect the concentration of the hydrogen gas. The Pd particles for modifying the surface of CNT contain Pd particles precipitated by the oxidation and reduction reaction of CNT and Pd when CNT is immersed in a solution of a Pd salt or complex and Pd particles precipitated by the oxidation and reduction reaction of Pd and an electrode material superposed on the oxidation and reduction reaction in the solution to be brought about. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a hydrogen gas sensor that responds with high sensitivity and high speed at room temperature, is inexpensive and has low power consumption, and a hydrogen gas sensor manufacturing method capable of easily manufacturing the hydrogen gas sensor.

  In recent years, vigorous development of fuel cell vehicles, household fuel cell power generation devices, etc. is expected, and it is expected that infrastructure development of hydrogen gas plants, hydrogen stations, etc. for supplying hydrogen gas will proceed in the future. This hydrogen gas as a fuel has an excellent feature as clean energy with a small environmental load. On the other hand, it is also known that the reaction of hydrogen gas rapidly proceeds when the concentration becomes 4% (lower concentration) or higher in air at normal temperature and pressure. Therefore, in order to support the spread of hydrogen gas, it is indispensable to develop a hydrogen gas sensor that can detect even a very small amount of hydrogen gas with high sensitivity and speed.

  Hydrogen gas sensors have been researched for a long time, and many of them utilize a catalytic action of dissociating hydrogen molecules of palladium (Pd, hereinafter Pd) and platinum (Pt, hereinafter Pt). One of the typical hydrogen gas sensors is a catalytic combustion type that detects hydrogen based on temperature rise due to catalytic combustion (hydrogen atoms adsorbed and dissociated on the catalyst surface and oxygen atoms derived from air react to produce water molecules). There are sensors and FET type gas sensors that directly measure the interface potential (for example, Patent Document 1). The gas sensor of Patent Document 1 uses a MISFET (metal-insulated gate type FET), uses Pd as a gate electrode, and has a Pd—SiO 2 —Si structure. In this hydrogen gas sensor, the hydrogen gas to be detected is adsorbed and dissociated on the surface of Pd and becomes H in an atomic state, and reaches the SiO2 interface by diffusion and generates an interface potential. As a result, the threshold voltage that is a characteristic parameter of the FET changes. The concentration of hydrogen gas is detected by measuring the change in threshold voltage.

  Considering the expected widespread use of hydrogen gas in the future, the capacity required for hydrogen gas sensors is in the concentration range of 0.01% to 0.1%, which can be evaluated as about 1/400 to 1/40 of the above lower limit concentration. Hydrogen gas must be able to be detected selectively and quantitatively, and conventional catalytic combustion type sensors and hydrogen gas sensors such as FET type gas sensors have limitations (about 1%), and the detection results vary greatly.

  By the way, in recent years, a gas sensor and a CNT sensor for detecting gas using carbon nanotubes (hereinafter referred to as CNT) have attracted attention. In this CNT sensor, when gas molecules are adsorbed on the semiconductor CNT, charge transfer occurs between them, and the electrical characteristics (conductance, capacitance) of the semiconductor CNT change. Therefore, this phenomenon is used to detect gas. For example, in the case of a reducing gas such as ammonia gas, the adsorbed gas molecules give electrons to the CNT, the carrier (hole) density of the CNT that is a p-type semiconductor is reduced, and the conductance of the CNT is lowered. Gas can be detected by observing this change in conductance. For a CNT sensor using such a CNT, see Patent Document 3 described later.

  Here, although hydrogen gas is a reducing gas, its reducibility is low, it is difficult to detect hydrogen gas by CNT alone, and in order to improve reducibility, it is necessary to dissociate hydrogen molecules into atoms, such as Pd and Pt. It is necessary to modify the CNT with a catalytic metal. Several methods for modifying the CNT with a catalytic metal have already been proposed, and there are roughly two methods, a physical method and a chemical method. Typical physical methods include electron beam evaporation and sputtering, and typical chemical methods include electroless plating (see Non-Patent Document 1).

  Among these chemical methods, there is a method using an oxidation-reduction reaction between Pd or Pt nanoparticles and carbon (see Non-Patent Document 2). For example, it has been reported that nanoparticles of these metals are deposited on the surface of a CNT just by immersing CNT grown on a Si substrate in a metal salt solution of Au or Pt. This technique can also be applied to the production of metal nanoparticles having a small ionization tendency such as Cu, Pd, and Ag (see Non-Patent Document 3). However, these are all deposited as templates for producing this metal nanowire from Pd and Pt nanoparticles, and what kind of precipitation forms and particle distributions are modified as CNT sensors, Not disclosed. In order for the sensor to function with high sensitivity, it is necessary to sufficiently control the modification state of Pd and Pt nanoparticles so as to be suitable for the sensor. The disclosures of Non-Patent Documents 2 and 3 are not useful for sensor manufacture.

Similarly, at least one proton (H ⁺ ) dissociable group is introduced into, for example, CNT into a carbon atom of a cluster mainly composed of carbon atoms, and 2 proton conductors constituted by the obtained cluster derivative are obtained. A hydrogen gas sensor sandwiched between two electrodes has also been proposed (Patent Document 2). Provided is a fuel cell that can operate normally in a rapid manner, can be used in a wide temperature range, can achieve low power consumption and gas selectivity, and measures the hydrogen concentration using such a hydrogen gas sensor. Is.

However, it is actually difficult to introduce a proton (H ⁺ ) dissociable group (OH group) into CNTs by hydrolysis, and it is practically difficult to perform acid treatment (OSO ₃ H group) with a strong acid. Or there was only a method of plasma processing. Even if OH groups can be introduced, hydrophilic properties are imparted to refractory CNTs, and the fibers of the CNTs break apart during hydrolysis, and are used for sensor production, which is the original purpose. It has the potential to be difficult.

  The conventional method for modifying CNTs has been described above. In order to use the modified CNTs as a sensor, it is necessary to crosslink between the electrodes and obtain good electrical contact between the CNTs and the electrodes. It is also disclosed that there is the following method for manufacturing a sensor element in which CNTs are cross-linked between electrodes (see Patent Document 3). The first method is a method in which a large number of CNTs are grown on the sensor electrode by a CVD method between a pair of electrodes, and the second method is a method in which a large number of CNTs produced in advance are dispersed in a solvent and applied between the electrodes. It is a method of drying and accumulating at random. However, none of these methods is a method in which CNT can be freely controlled, and a good electrical connection cannot be realized between the electrode and the CNT, and it has not become a bridge suitable for modifying the CNT.

As a method of manipulating such a micro-sized micro object, one of the inventors of the present application applies an unequal electric field to the micro object to polarize it, and manipulates the polarized micro object with dielectrophoretic force. Then, we have studied the DEPIM (Dielectrophoretic Impedance Measurement Method) method of collecting on the microelectrode. And the CNT gas sensor is proposed as mentioned above using this DEPIM method (refer patent document 4). By using this method, nano-sized CNTs can be manipulated freely, good electrical connection becomes possible, and it is possible to accurately detect NO ₂ and NH ₃ gas at ppm level at room temperature. became.

  Although not using CNT, the present inventors have also proposed a gas sensor in which metal fine particles such as Pd and Pt are integrated by dielectrophoresis and crosslinked between electrodes by the DEPIM method (Japanese Patent Application). 2006-228656). This utilizes the hydrogen occlusion action of metal fine particles such as Pd and Pt. However, a gas sensor that bridges an expensive metal such as Pd or Pt as a hydrogen storage metal is inferior in cost performance as compared with a gas sensor using CNT. Further, once responding to hydrogen gas, even if it is refreshed with a gas such as air that does not contain hydrogen after that, it is memorized and difficult to use repeatedly. Considering the expected widespread use of hydrogen gas in the future, it should be reusable and not disposable.

JP-A-9-159633 JP 2006-156410 A JP 2006-329802 A JP 2003-227808 A M.K. Kumar et al., “Nanostructured Pt functionalized multiwalled carbon nanotube based hydrogen sensor”, J. Phys. Chem. B, 2006, vol. 110, p. 1129-p. 11298 H.C. Choi et al., “Spontaneos reduction of metal ions on the sidewalls of carbon nanotubes”, J. Am Chem. Soc., 2002, vol. 124, p. 9058-p. 9059 B. Xue et al., “Growth of Pd, Pt, Ag and Au nanoparticles on carbon nanotubes”, J. Mater Chem., 2001, vol. 11, p. 2378-p. 2381

  As explained above, contact combustion type sensors and FET type gas sensors have poor accuracy for accurate concentration measurement, and vary widely between manufactured individuals, and need to be heated to several tens to several hundreds of degrees centigrade. In addition, hydrogen selectivity and responsiveness were not sufficient. Problems remain in handling hydrogen gas.

  It is difficult for a CNT sensor to detect hydrogen gas alone, and it is absolutely necessary to modify it with a catalytic metal. In addition, it is necessary to accumulate CNTs after processing such as coating, and it has not been possible to easily manufacture a hydrogen gas sensor. As described above, it is difficult to control to improve the CNT accumulation result and the modified state, and a practical method for producing a hydrogen gas sensor has not been established so far. It has been proposed to form a proton conductor as in Patent Document 2, but the manufacture is not simple, and the cross-linking itself of the CNTs may collapse, making it difficult to manufacture the sensor and reducing the sensitivity.

  The catalyst metal modification techniques of Non-Patent Documents 3 and 4 are all depositions as templates for producing nanowires of these metals from Pd and Pt nanoparticles, and what kind of deposition is necessary as a CNT sensor. It is not considered at all how to control the manner and distribution of particles. This is because the template is eventually removed and becomes a metal wire. The problem with the hydrogen gas sensor is that the modification state by the catalytic metal is sufficiently controlled so as to be suitable for the hydrogen gas sensor, and the techniques of Non-Patent Documents 2 and 3 do not have such a purpose and provide suggestions for the production of the hydrogen gas sensor. Absent.

  In hydrogen gas plants and hydrogen gas stations that will be installed in the future, there is a high possibility that hydrogen gas sensors will be installed at a number of remote detection points. There must be. Accordingly, the hydrogen gas sensor is required to have low temperature and low power consumption. There is a demand for a hydrogen gas sensor that can be used repeatedly without throwing away expensive catalyst metals such as Pd and Pt, and that dramatically increases sensitivity, reliability, and the like.

  There is one hint in the problems described above. According to the DEPIM method already proposed by one of the present inventors, it is easy to collect and control CNTs that cross-link between electrodes. On top of this, if an integrated state suitable for modification of the hydrogen gas sensor is further developed and nanoparticles of catalytic metals such as Pd and Pt can be deposited by a suitable modification method, a highly sensitive CNT hydrogen gas sensor can be a simple device. And can be manufactured by the manufacturing method.

  Accordingly, an object of the present invention is to provide a hydrogen gas sensor that responds to hydrogen gas at room temperature with high sensitivity and high speed, has high reliability, can be used repeatedly at low cost, and is easy to manufacture.

  Another object of the present invention is to provide a method for producing a hydrogen gas sensor that is highly sensitive to hydrogen gas at room temperature, responds at high speed, has high reliability, can be used repeatedly at low cost, and is easy to manufacture. And

  A hydrogen gas sensor of the present invention comprises a sensor element obtained by dielectrophoresis of a carbon nanomaterial and cross-linking between the electrodes, and a catalytic metal body that dissociates and adsorbs hydrogen gas by modifying the surface of the carbon nanomaterial, A hydrogen gas sensor for detecting a hydrogen gas concentration by measuring a change in conductance or resistance between electrodes, wherein the catalyst metal body for modifying the surface is immersed in a solution of a salt or complex of a catalyst metal. Catalyst nanoparticle and catalyst metal particles deposited by the oxidation-reduction reaction of the catalyst metal, and catalyst deposited from the oxidation-reduction reaction of the catalyst metal and the electrode material occurring in the solution superimposed on the oxidation-reduction reaction The main feature is that it contains metal particles.

  In the method for producing a hydrogen gas sensor of the present invention, a carbon nanomaterial is dielectrophoretically cross-linked between a pair of electrodes formed on an insulating substrate, and once dried, a catalyst metal salt having a smaller ionization tendency than the electrode material or The electrode is immersed in a complex solution, and the carbon nanomaterial is modified with a particulate catalyst metal body that deposits, and a redox reaction between the electrode material and the catalyst metal is generated in an overlapping manner to generate a surface of the carbon nanomaterial. The main feature is that a particulate catalytic metal body derived from this redox reaction is precipitated.

  According to the hydrogen gas sensor and the manufacturing method thereof of the present invention, a hydrogen gas sensor that responds to hydrogen gas at room temperature with high sensitivity and high speed, has high reliability, can be used repeatedly at low cost, and is easy to manufacture. Can be provided.

  The first aspect of the present invention includes a pair of electrodes, a sensor element in which a carbon nanomaterial is dielectrophoretically crosslinked between the electrodes, and a catalyst that dissociates and adsorbs hydrogen gas by modifying the surface of the carbon nanomaterial. A hydrogen gas sensor for detecting a hydrogen gas concentration by measuring a change in conductance or resistance between the electrodes, wherein the catalyst metal body for modifying the surface is a solution of a catalyst metal salt or complex When the carbon nanomaterial is immersed in the catalyst, the catalyst metal particles precipitated by the oxidation-reduction reaction of the catalyst metal, and the oxidation-reduction reaction of the catalyst metal and the electrode material occurring in the solution in a superimposed manner with the oxidation-reduction reaction The hydrogen gas sensor is characterized in that it contains catalytic metal particles deposited from the above. With this configuration, the catalyst metal, its salt or complex, electrode material, and solvent can be selected, and the conditions such as dielectrophoresis and immersion time can be changed. In addition, it is possible to provide a hydrogen gas sensor that is highly reliable, inexpensive, can be repeatedly used, and is easy to manufacture.

  According to a second aspect of the present invention, there is provided a hydrogen gas sensor according to the first aspect, wherein the atomic ratio of the catalytic metal to carbon is 1% to 25%. With this configuration, it is possible to provide a highly reliable hydrogen gas sensor that responds to hydrogen gas at room temperature with extremely high sensitivity and high speed.

  The third aspect of the present invention is a form subordinate to the first or second aspect, in which a pair of electrodes is immersed in a solution as a cathode, and a DC voltage is applied between the separately immersed anode. The hydrogen gas sensor is characterized in that a cathode oxygen reduction reaction is performed. With this configuration, the catalytic metal body can be reliably modified by electroplating.

  According to a fourth aspect of the present invention, there is provided a pair of electrodes, a sensor element obtained by dielectrophoresis of a carbon nanomaterial and cross-linking between the electrodes, and a catalytic metal that modifies the surface of the carbon nanomaterial to dissociate and adsorb hydrogen gas. A hydrogen gas sensor for detecting a hydrogen gas concentration by measuring a change in conductance or resistance between the electrodes, wherein the ionization tendency of the electrode material of the electrode and the carbon nanomaterial is higher than that of the catalytic metal. It is a hydrogen gas sensor characterized by being selected to be large, and it is possible to provide a hydrogen gas sensor that is highly sensitive, fast response, highly reliable, inexpensive, and repeatable only by considering the ionization tendency.

  The fifth aspect of the present invention is a form subordinate to any one of the first to fourth aspects, and when the solution compares the detachment force of the carbon nanomaterial from the electrode and the physical adsorption force, The hydrogen gas sensor is a solution in which a salt or complex of a catalytic metal is dissolved in a solvent having a higher adsorption power. With this configuration, peeling from the electrode in the vicinity of the bridging end of the carbon nanomaterial can be easily prevented, and the reliability can be improved.

  A sixth aspect of the present invention is a hydrogen gas sensor according to any one of the first to fifth aspects, wherein a change in conductance between the electrodes is a time change rate of conductance. . With this configuration, an accurate hydrogen gas concentration can be quantified using a conductance time change rate proportional to the hydrogen concentration.

  According to a seventh aspect of the present invention, a carbon nanomaterial is dielectrophoretically cross-linked between a pair of electrodes formed on an insulating substrate, and once dried, a catalyst metal salt or complex having a smaller ionization tendency than the electrode material. The electrode is immersed in a solution, and the carbon nanomaterial is modified by a particulate catalyst metal body that deposits, and an oxidation-reduction reaction of the electrode material and the catalyst metal is generated in a superimposed manner on the surface of the carbon nanomaterial. A method for producing a hydrogen gas sensor, comprising depositing a particulate catalytic metal body derived from an oxidation-reduction reaction. With this configuration, just select the catalyst metal and its salt or complex, electrode material, solvent, change the conditions such as dielectrophoresis conditions, immersion time, etc., it responds with high sensitivity and high speed at room temperature, and has high reliability. Therefore, it is possible to manufacture a hydrogen gas sensor that can be used repeatedly at low cost and is easy to manufacture.

  An eighth form of the present invention is a form subordinate to the seventh form, and is a hydrogen gas sensor manufacturing method characterized in that the atomic ratio of the catalyst metal to carbon is 1% to 25%. With this configuration, it is possible to provide a highly reliable hydrogen gas sensor that responds with high sensitivity and high speed at room temperature.

  In the ninth aspect of the present invention, the carbon nanomaterial is dielectrophoretically crosslinked between a pair of electrodes formed on an insulating substrate, and once dried, the carbon nanomaterial is detached from the electrode and physically adsorbed. When a force is compared, a catalyst metal salt or complex having a smaller ionization tendency than that of the electrode material is dissolved in a solvent having a larger physical adsorption force. The carbon nanomaterial is modified with a catalytic metal body, and a redox reaction between the electrode material and the catalytic metal is generated in a superimposed manner to form a particulate catalytic metal body derived from the redox reaction on the surface of the carbon nanomaterial. It is a hydrogen gas sensor manufacturing method characterized by making it precipitate. With this configuration, by selecting the catalytic metal and its salt or complex, electrode material, and solvent, and changing the conditions of dielectrophoresis, immersion time, etc., it responds with high sensitivity and high speed at room temperature. Peeling from the electrode in the vicinity of the bridging end of the material can be easily prevented, and a highly reliable hydrogen gas sensor can be manufactured.

  A tenth aspect of the present invention is a form subordinate to the ninth aspect, wherein a cathode is formed by immersing a pair of electrodes in a solution, and a DC voltage is applied between the separately immersed anode. It is a hydrogen gas sensor manufacturing method characterized by carrying out a reduction reaction. With this configuration, the catalytic metal body can be reliably modified by electroplating.

(Embodiment 1)
Hereinafter, a hydrogen gas sensor according to Embodiment 1 of the present invention, a CNT accumulation device that accumulates CNTs on the hydrogen gas sensor, and a method for producing a hydrogen gas sensor will be described. FIG. 1 is an explanatory diagram of a hydrogen gas sensor and a CNT integration device according to Embodiment 1 of the present invention. FIG. 2 is an explanatory diagram illustrating a state of a liquid phase oxidation-reduction reaction of the hydrogen gas sensor according to Embodiment 1 of the present invention. (A) is a SEM image photograph before depositing catalyst metal particles of the hydrogen gas sensor in Embodiment 1 of the present invention, FIG. 3 (b) is an SEM image photograph after depositing catalyst metal particles of (a), FIG. 4 is an EDX analysis diagram of the CNT subjected to the reduction reaction treatment of the hydrogen gas sensor according to Embodiment 1 of the present invention, and FIG. 5A is a diagram showing catalyst metal particles on the CNT of the hydrogen gas sensor according to Embodiment 1 of the present invention. SEM image photograph when depositing at an atomic ratio of 6%, FIG. 5B is an SEM image photograph when catalytic metal particles are deposited at an atomic ratio of 80% on the electric CNT of the hydrogen gas sensor of FIG. It is.

  FIG. 1 shows a hydrogen gas sensor and a CNT integrated device according to Embodiment 1 of the present invention. In FIG. 1, 1 is a hydrogen gas sensor formed as a substrate for detecting hydrogen gas, 2a and 2b are a pair of microelectrodes such as a castle wall type and a comb type made of chromium (Cr) or aluminum (Al). (Electrode of the present invention) 3a and 3b are the tip portions (edges which become corner portions) of the microelectrodes 2a and 2b, and an electric field concentration portion which forms an unequal electric field, 4 is a glass substrate, plastic, silicon oxide, etc. 5 is a sealed chamber for accommodating the hydrogen gas sensor 1 and performing dielectrophoresis, 6 is a power source for applying an AC voltage to the electrodes 2a and 2b, and 7 is an impedance adjusting device provided with a resistance of about 1 kΩ. is there. As shown in FIG. 1, the castle wall type microelectrodes 2a and 2b have a concavo-convex shape where the electric field concentration portions 3a and 3b are closest to each other as shown in FIG. A pair of microelectrodes in which teeth (for example, 30 μm to 100 μm width) are formed like combs are inserted and combined to face each other, and an unequal electric field is mainly formed between opposing upper and lower edges in the thickness direction of the electrodes Is formed.

  When using, for example, Cr as the electrode material of the microelectrodes 2a and 2b, a Cr thin film is vacuum-deposited on the insulating substrate 4 and the electrodes are formed by photolithography. The electrode finger of the first embodiment is a castle wall type, has a length of 5 mm, and an electrode shortest gap length of 5 μm. In addition to vacuum deposition, a film may be formed by plating, sputtering, or the like, and the thickness of the thin film is preferably about 50 nm to 200 nm. The electrode material of the microelectrodes 2a and 2b needs to be a metal with a small ionization tendency so that electrolysis does not occur when an alternating voltage is applied by dielectrophoresis, and is based on a catalytic metal such as Pd or Pt that modifies CNT, which will be described later. Must be a material with a high ionization tendency.

  Next, 8 is a CNT suspension (solution of the present invention) in which CNT is forcibly dispersed in an organic solvent such as ethanol or a solvent such as water, 8a is a single-walled CNT (SWCNT), a multilayered CNT (MWCNT), or the like. CNT (carbon nanomaterial of the present invention), 9a is an introduction path for supplying the CNT suspension, 9b is a discharge path for discharging the CNT suspension, and 10 is a pump for supplying the CNT dispersion solvent to the sealed chamber 5 It is. The cross-linked portion obtained by collecting the CNTs 8a by dielectrophoresis (DEPIM method) between the microelectrodes 2a and 2b is a sensor element for hydrogen gas in the present invention.

  When collecting CNTs on the microelectrodes 2a and 2b by dielectrophoresis, the CNT suspension 8 is circulated by the pump 10 through the sealed chamber 5, and the AC voltage is supplied to the microelectrodes 2a and 2b in the sealed chamber 5 by the power source 6. Is applied. The CNTs 8a are induced and accumulated by dielectrophoresis between the electric field concentration portions (electrode gaps) where the electric field strength becomes maximum due to the unequal electric field generated thereby. After the CNTs are accumulated, the CNTs are dried by evaporating the solvent, and are physically adsorbed on the insulating substrate 4 and the electrode material by physical adsorption force such as van der Waals force.

Here, the dielectrophoretic force F _DEP of the DEPIM method will be described. The dielectrophoretic force F _DEP can be expressed in a complex number expression as F _DEP = 2πε _m · a ³ · Re [K] ▽ E ² . Where ε _{m is} the dielectric constant of the suspension, a is the radius of the microparticle when approximated by a sphere, Re [K] is a parameter depending on the complex dielectric constant of the microscopic object and the suspension, and E is the electric field strength. is there. This Re [K] changes positively and negatively with the frequency f of the electric field used for dielectrophoresis as a parameter. A positive dielectrophoretic force works in a specific frequency range, for example, 10 kHz to 1 MHz, and a negative dielectrophoretic force works otherwise. Therefore, it is necessary to effectively _collect fine particles by selecting a frequency and applying the maximum positive dielectrophoretic force F _DEP . Incidentally, CNT is much larger aspect ratio is an elongated particle, the electric field concentrated portions by receiving the action of the dielectrophoretic force in accordance with the F _DEP is only in principle coefficients shown changes in the radius a ³ of the expression Aggregated into 3a and 3b.

  When the CNTs 8a are subjected to dielectrophoresis, the CNTs 8a are accumulated at a stable position in terms of electric field. A large number of slender fibrous particles bent in an S shape and having more bent portions are oriented in the direction of the electric field formed around the tip portions of the microelectrodes 2a and 2b and overlap each other to form a mesh It becomes a shape. FIG. 3 (a) shows the cross-linking of CNTs formed in a mesh shape between the microelectrodes 2a and 2b by this dielectrophoresis. Such a mesh is never obtained by accidentally controlled coating or CVD growth. As the sensor element, a film-like element spread in a plane like graphite is suitable, but this can be realized neatly and simply by selecting the type of CNT 8a and performing dielectrophoresis. Further, by adjusting the time of dielectrophoresis, the mesh size can be adjusted coarsely or finely.

  An example of an electrode having a sensor element in which such CNTs 8a are integrated will be described as an example. Single-walled CNTs (SWCNTs, manufactured by Sigma-aldrich, average diameter 1 nm, 1-4 mm length, purity 50%) are ethanol. It is ultrasonically dispersed at a concentration of 1 g / ml and circulated in the sealed chamber 5 using a pump 10 at a flow rate of 0.5 ml / min. A high-frequency voltage of amplitude AC 10 Vpp and frequency 100 kHz is applied between the microelectrodes 2 a and 2 b. The CNTs may be produced by dielectrophoretic integration of CNTs, and after the dielectrophoretic integration is completed, ethanol is evaporated at room temperature.

  Subsequently, a process for modifying the CNT 8a cross-linked between the microelectrodes 2a and 2b of Embodiment 1 of the present invention with a catalytic metal will be described. The microelectrodes 2a and 2b of the hydrogen gas sensor of Embodiment 1 are characterized in that catalyst metal particles (catalyst metal body of the present invention) are deposited not only on the CNTs 8a but also on and near the microelectrodes 2a and 2b. Have. That is, the CNT 8a is reduced, the catalytic metal is oxidized and the catalytic metal particles are deposited on the CNT 8a, the electrode material is also reduced at the microelectrodes 2a and 2b, and the catalytic metal particles are converted into the microelectrodes 2a and 2b and this This precipitates in the vicinity, and further, this oxidation-reduction reaction assists the oxidation of the nearby catalyst metal, and in addition to the original deposition on the CNT 8a, more catalyst metal particles are deposited on the surface of the CNT 8a. That is, the catalyst metal particles that modify the surface of the CNT 8a are superposed in the solution with catalyst metal particles that are precipitated by a redox reaction between the CNT 8 and the catalyst metal when immersed in a solution of a salt or complex of the catalyst metal. Thus, two types of catalyst metal particles that are deposited due to the redox reaction between the catalyst metal and the electrode material that occur in this manner are included.

  First, the oxidation that occurs in the catalyst metal when modifying with the catalyst metal particles and the reduction reaction that occurs in the CNT 8a will be described. In the first embodiment, when the CNT 8a is modified with the catalyst metal, the CNT 8a is immersed in an aqueous solution of a salt or complex of a catalyst metal such as Pd or Pt. Since the catalytic metal in the aqueous solution and the CNT 8a are different in ionization tendency, the CNT 8a is oxidized in the aqueous solution as shown in FIG. 2, and the catalytic metal ion in the solution obtains electrons and is reduced in the form of particles in the surface of the CNT 8a. It precipitates in.

  However, the oxidation-reduction reaction of Embodiment 1 also occurs between the electrode material and the catalyst metal in a superimposed manner. In Embodiment 1, since metals such as Cr and Al, which have a higher ionization tendency than the catalyst metal, are used as the electrode material, the microelectrodes 2a and 2b are placed in an aqueous solution of a salt or complex of a catalyst metal such as Pd or Pt. When immersed, catalytic metal particles are deposited on the microelectrodes 2a and 2b.

  Therefore, in the CNT 8a and the microelectrodes 2a and 2b, redox reactions occur simultaneously between the catalytic metal and the CNT 8a and between the catalytic metal and the electrode material. The reaction between the catalyst metal and the electrode material increases the deposition of the catalyst metal on the surface of the CNT 8a. That is, the catalytic metal particles generated by the substitution with the electrode material are formed not only on the surface of the electrode but also on the surrounding insulating substrate 4 and the electrode gap (between the electric field concentration portions 3a and 3b) where the CNTs 8a are integrated. The First, small catalytic metal nanoparticles (nuclei), which cannot be confirmed by SEM, are replaced by a redox reaction on the electrode gap, and nucleated and precipitated. Therefore, on the CNT surface, in addition to the catalyst metal nanoparticles (nuclei) generated by the original substitution reaction of CNT8a, nuclei due to the influence of the electrode material are added, and two types of nuclei grow, and a large amount of CNT is produced. Will be qualified. Here, the two oxidation-reduction reactions cause nucleation on CNTs, and the modification is performed in a superimposed manner. The pH and potential of the catalyst metal salt solution change depending on the presence of the electrode material-catalyst metal oxidation-reduction reaction. It is also considered that nucleation and nucleation are promoted on uneven CNTs that have been formed into a net-like shape by dielectrophoresis. 5 (a) and 5 (b), it can be seen that the nucleus growth proceeds in a range where the electrode material Cr would have been replaced, eluted and diffused. Thus, the catalyst metal needs to have an ionization tendency in the order of (electrode material, CNT)> catalyst metal in the difference in ionization tendency. At this time, the modification of the CNT with the catalytic metal is doubled as compared with the case of the CNT alone due to the action derived from the electrode material.

  A sensor element made of an assembly of CNTs 8a produced by dielectrophoresis has a bridging end (total of both ends) having a length close to twice the electrode gap. Therefore, the influence of the electrode material is strong at the cross-linking ends of the CNTs 8a that occupy about 50% to 70% of the total length of the cross-linking portion, and the above-described catalytic metal deposition action easily proceeds. The action of the catalytic metal makes it possible to effectively capture hydrogen atoms, and when the CNT 8a touches hydrogen gas, the conductance increases sharply. If this change in conductance is detected, the hydrogen gas concentration can be detected with high sensitivity.

Therefore, the simple dielectrophoresis does not increase the deposition of the catalytic metal on the CNT 8a. However, by selecting the difference in the ionization tendency of the catalytic metal, the CNT, and the electrode material, the catalytic metal on the CNT is selected. The amount of precipitation can be increased. 5 (a) and 5 (b), when the electrode material is Cr, when the microelectrode is immersed in an aqueous solution of a catalytic metal (in this case, palladium chloride (PdCl ₂ )), fine particles of Pd which is a catalytic metal on the surface of Cr. It is shown that a large number of are deposited almost uniformly.

  However, the present inventors have obtained a new finding that the greater the amount of catalyst metal deposited, the better the sensitivity. That is, as shown in FIG. 6B, when the atomic ratio representing the atomic ratio exceeds 25%, a large amount of catalyst metal is deposited on the microelectrodes 2a and 2b (FIG. 5B). In this state, the sensitivity of the sensor element is reduced as shown in FIG. One reason for this is that excessive deposition of the catalytic metal on the electrode material is excessive, the catalytic metal that is also a hydrogen storage metal stores hydrogen, and the amount of hydrogen adsorbed by the catalytic metal on the CNT 8a is reduced. It can be relatively lowered. As shown in FIG. 6A, when this ratio is less than 1% which is the lower limit, the catalytic metal is not sufficiently deposited on the CNT 8a (see FIG. 5A), and the catalytic action is not sufficient. . Here, FIG. 6A is an EDX analysis diagram in the case where the catalyst metal particle atomic ratio of the hydrogen gas sensor in Embodiment 1 of the present invention is less than 1%, and FIG. 6B is the catalyst metal particle atomic ratio. FIG. 7 is a response diagram to hydrogen gas when the hydrogen gas sensor according to Embodiment 1 of the present invention has a sensor element with an atomic ratio of 25%. FIG. 7 is a response diagram when the atomic ratio is 25%, but the same response is obtained when the ratio exceeds 25% or less than 1%. Therefore, it is necessary to make the atomic ratio of the catalytic metal to carbon 1% to 25% in order to produce a highly sensitive CNT sensor element.

  Note that the atomic ratio of the catalytic metal to the carbon described above may be obtained by energy dispersive X-ray (hereinafter, EDX) analysis. As shown in FIG. 4, since Pd is detected together with C, Si, and O derived from the glass substrate of the electrode by EDX analysis, the atomic ratio of the catalyst metal to carbon is obtained by taking the ratio of the strength of C to the strength of this Pd. Number ratio can be calculated.

  By the way, two oxidation-reduction reactions of catalytic metal-CNT and catalytic metal-electrode material occur even when an organic solvent such as acetone is used. However, when an organic solvent such as acetone is used, the oxidation-reduction reaction is weak, and the binding and aggregation state of the CNTs accumulated between the microelectrodes in the CNT accumulating device is loosened and becomes easy to peel off. Certainly, if an organic solvent is used, the CNTs are easily dissolved in the solvent and easy to process, but the CNTs easily peel off at the cross-linked ends of the CNTs. Accordingly, a solvent having a detachment force of CNT in a solvent lower than a physical adsorption force of the CNT to the electrode material at room temperature is suitable as the solvent. Of these, water is most preferably used as the solvent. Whether or not the detachment force is lower than the physical adsorption force may be confirmed by actually immersing it in a solvent and not being peeled off, or by conducting electricity and confirming continuity. Further, as will be described later with reference to FIG. 13B, in the case of an organic solvent such as acetone, once the hydrogen gas is touched, it is not refreshed even if it is touched again, making it difficult to repeatedly use the hydrogen gas sensor. Become.

In the first embodiment, a microelectrode using palladium acetate (Pd (CH ₃ COO) ₂ ) as the catalyst metal is used. Palladium acetate is originally hardly dissolved in water, but it is sufficient to obtain a suspension by ultrasonic dispersion or the like and use water as a solvent. In this case, the cross-links of the accumulated CNTs 8a do not peel from the microelectrodes 2a and 2b. For example, in order to modify with Pd, 50 mg of palladium acetate is suspended in 175 ml of ion-exchanged water, subjected to ultrasonic dispersion for several hours to form a suspension, and insolubles are removed with a filter. Thereafter, the microelectrodes 2a and 2b in which the CNTs 8a are integrated by dielectrophoresis are immersed in an aqueous palladium acetate solution at room temperature, and after a certain period of time, the microelectrodes 2a and 2b are pulled up from the solution and dried in the air. Then, it was confirmed by EDX analysis that CNT was modified with Pd by the above treatment as shown in FIG.

  According to the EDX analysis of FIG. 4, a small Pd peak appears together with C, Si, and O derived from the glass substrate of the electrode, and it can be confirmed that CNT is modified with Pd. In this case, the atomic ratio of Pd to carbon (CNT) was about 6%. A highly sensitive sensor element can be realized with an atomic ratio in the vicinity. Therefore, it can be confirmed that the liquid phase oxidation-reduction reaction has occurred just by immersing the CNT in an aqueous solution of the catalyst metal salt at room temperature, and it can be seen that the CNT is modified by the catalyst metal.

  In addition, confirmation by a scanning electron microscope (SEM) was also performed. FIGS. 3A and 3B are SEM image photographs before and after precipitation of Pd, respectively, but no significant change is observed in the SEM image. This is presumably because the metal particles deposited on the CNT surface by the liquid phase oxidation-reduction reaction are nano-sized and cannot be observed with SEM resolution. This can be seen from the fact that the catalyst metal particles in FIG.

Next, the hydrogen response characteristics of the hydrogen gas sensor of Example 1 described above will be described. FIG. 8 is a principle diagram of sensor response when hydrogen gas is detected by the hydrogen gas sensor according to Embodiment 1 of the present invention. This shows how the hydrogen gas changes on the sensor element modified with Pd when performing hydrogen gas detection. According to this, hydrogen gas (H ₂ ) in the air causes a reaction of H ₂ → 2H ⁺ + 2e ⁻ by the action of the catalyst metal, here Pd.

That is, the hydrogen gas molecules (H ₂ ) that have reached the surface of the Pd-modified CNT dissociate on the surface of the catalyst metal, and generate and adsorb hydrogen atoms (H ⁺ ). This hydrogen atom (H ⁺ ) diffuses into Pd, and as a result, the work function of Pd decreases, and electron transfer from Pd to CNT, which is a p-type semiconductor, occurs to cause a CNT reduction reaction. As a result, it is considered that the hole carrier density in the CNT is lowered and the conductance of the CNT is lowered.

  In the case of Embodiment 1, since the catalytic metal Pd is deposited in a relatively large amount at the junction between the CNT and the microelectrode, the hole carrier density in the CNT is sufficiently reduced, and the sensor element It is thought that the conductance of the is reduced. This improves the sensitivity of the sensor element.

Here, the result of measuring the hydrogen response characteristic will be specifically described. FIGS. 9 (a) and 9 (b) show the responses at room temperature to hydrogen-diluted hydrogen (concentration 1%) of a hydrogen gas sensor provided with Pd-modified CNT immersed in an aqueous palladium acetate solution for 1 hour or 5 minutes, respectively. FIG. 9 (a) is a response diagram of the hydrogen gas sensor according to Embodiment 1 of the present invention to the air-diluted hydrogen of the sensor element subjected to the immersion treatment for 1 hour, and FIG. 9 (b) is 5 of the hydrogen gas sensor according to Embodiment 1 of the present invention. It is a response figure with respect to air dilution hydrogen of the sensor element which carried out the immersion process for minutes. Alternating exposure to air and hydrogen gas. Both the vertical axis represents a variation ΔG of conductance sensor response normalized by the initial conductance G ₀ of the sensor element (normalized conductance).

  According to this, the normalized conductance decreased immediately after hydrogen exposure for both sensors and saturated within a few minutes. When switched to air, the normalized conductance returned to its initial value exponentially, and a similar response was obtained with good reproducibility when exposed to hydrogen again. 9 (a) and 9 (b) show only responses of 1 hour and 5 minutes. However, when the immersion time (treatment time) for immersing the catalyst metal salt is 1 hour or less, the sensor response increases as the time increases. (Standardized conductance) increases.

The response amount of the hydrogen gas sensor immersed in 5 minutes shown in FIG. 9 (b) is about 50% of the sensor response of the sensor element immersed in 1 hour shown in FIG. 9 (a) (when the hydrogen gas concentration is 1%) ΔG / G ₀ is −0.04 and the latter ΔG / G ₀ is −0.09), and the response amount is small. However, even in this case, sufficient hydrogen detection is possible. The hydrogen gas sensor provided with CNTs not subjected to Pd modification hardly responded to hydrogen gas. The increase rate of the conductance of the sensor element immersed for 5 minutes from before immersion was 1.2, but the increase rate of the conductance of the sensor element immersed for 1 hour from before immersion was 2.2.

  Next, the characteristics of the hydrogen gas sensor provided with CNT modified with Pd by oxidation-reduction reaction in the catalyst metal salt solution will be described in detail. FIG. 10 is a response diagram when the hydrogen gas concentration of the hydrogen gas sensor in the first embodiment of the present invention is changed, and FIG. 11 is a hydrogen gas concentration and conductance time change rate of the hydrogen gas sensor in the first embodiment of the present invention. FIG. FIG. 10 shows the sensor response of the Pd-modified hydrogen gas sensor of FIG. 9A to hydrogen with a concentration of 0.01% -1%. FIG. 10 shows that the normalized conductance decreases even with 0.01% hydrogen, and 0.01% hydrogen can be detected at room temperature. The higher the hydrogen concentration, the greater the sensor response saturation value and the rate of change in conductance time immediately after hydrogen exposure. Among these, the sensor response saturation value does not show linearity with respect to the hydrogen concentration, but the conductance time change rate immediately after exposure to hydrogen is almost proportional to the hydrogen concentration as shown in FIG. It can be used for quantitative determination of hydrogen gas concentration. Since FIG. 10 is complicated, it is not shown in FIG. 10, but the sensor responses of 5% and 0.05% hydrogen gas show the same tendency, and the result shows a response proportional to the gas concentration. It has gained.

  Next, the gas selectivity of the hydrogen gas sensor in Embodiment 1 will be described. As can be seen from the above description, the sensor element including the Pd-modified CNT according to the first embodiment can sufficiently meet the future needs in terms of sensitivity, reproducibility, quantitativeness, operating temperature, and the like. It has the nature. Therefore, the gas selectivity important for practical use will be described.

As a gas for evaluating the selectivity, nitrogen dioxide (NO ₂ ), which is present in the atmosphere and is known as a gas to which a sensor element equipped with CNT easily responds, was used. FIGS. 12A and 12B show the response of the hydrogen gas sensor provided with CNTs before and after the Pd modification to NO ₂ (1 ppm). FIG. 12A is a response diagram to NO ₂ before the catalyst metal modification of the hydrogen gas sensor according to the first embodiment of the present invention, and FIG. 12B is a diagram after the catalyst metal modification of the hydrogen gas sensor according to the first embodiment of the present invention. it is a response diagram of NO _2. In both sensor elements, the normalized conductance increased by exposure to NO ₂ , but the response amount of the Pd-modified sensor element decreased to about 60% before the modification. This is also considered to be due to the decrease in NO ₂ adsorption sites due to Pd modification, suggesting the possibility that the response to gases other than hydrogen can be suppressed by Pd modification.

The catalyst metal salt described above was palladium acetate, and the solvent was water. Therefore, hereinafter, (1) a hydrogen gas sensor in which the catalyst metal salt is changed to palladium chloride (PdCl ₂ ) and the solvent is water, and (2) the catalyst metal salt is palladium acetate, but the solvent is acetone, which is an organic acid. The case where is used will be described. FIG. 13A is a response diagram when the catalytic metal of the hydrogen gas sensor in Embodiment 1 of the present invention is palladium chloride, and FIG. 13B is acetone as the solvent of the hydrogen gas sensor in Embodiment 1 of the present invention. It is a response diagram when doing.

FIG. 13A shows the sensor response (normalized conductance) of a hydrogen gas sensor in which the catalyst metal is palladium chloride (PdCl ₂ ) and the solvent is water. According to this, it turns out that there exists an effect equivalent to the case of the palladium acetate mentioned above. In addition, for example, a platinum salt such as potassium chloroplatinate (K ₂ [PdCl ₆ ]) can provide a similar sensor response (normalized conductance) when the solvent is water (not shown). The increase rate of the conductance of the sensor element immersed in palladium chloride for 5 minutes from before immersion was 2.8, and ΔG / G ₀ at the time of hydrogen exposure was −0.05.

Next, the case where the catalytic metal salt is palladium acetate and the solvent is acetone will be described. As shown in FIG. 13B, the hydrogen gas sensor in this case causes a problem when exposed to air and hydrogen gas alternately. It does not return to the initial value when exposed to air after being exposed to hydrogen gas. Therefore, reproducibility is poor and it is difficult to repeatedly use the hydrogen gas sensor. This hydrogen gas sensor was immersed in palladium acetate for 5 minutes, the rate of increase in conductance of the sensor element from before immersion was 1.7, and ΔG / G ₀ at the third hydrogen exposure was −0.02. there were. As described above, water for which the detachment force of CNTs in the solvent is lower than the physical adsorption force is suitable as the solvent for the immersion treatment, and an organic solvent such as acetone is not suitable as the solvent for the immersion treatment.

  From the above, it can be seen that the modification with the catalytic metal to the hydrogen gas sensor provided with CNTs by the oxidation-reduction reaction described above is extremely effective as a method for producing a sensor element of a highly sensitive hydrogen gas sensor operable at room temperature. Its greatest feature is the ease of production, and it is only necessary to immerse a hydrogen gas sensor equipped with CNTs produced in advance by dielectrophoresis in a catalytic metal salt for several minutes to several hours and dry it. By adjusting this immersion time, the atomic ratio of the catalyst metal to carbon can be easily set to 1% to 25%. Then, by using the DEPIM method when accumulating CNTs, it becomes possible to easily and inexpensively produce modified film-like CNTs.

  Next, Table 1 shows the result of comparing the initial conductance (value before hydrogen exposure) of the hydrogen gas sensor equipped with CNTs before and after the catalytic metal (Pd) modification treatment.

  According to (Table 1), the initial conductance is increased by the Pd modification treatment. This is presumably because CNT was oxidized opposite to Pd by the modification treatment, and the conductance increased. The case of cathodic reduction reaction will be described in detail in Embodiment 2. There is a possibility that the degree of Pd modification can be quantified from the increase rate of the initial conductance. For example, the initial conductance increase when the immersion time is 1 hour is larger than that when the immersion time is 5 minutes, which is qualitatively consistent with the difference in sensor response shown in FIG. And it becomes possible to control the responsiveness of a sensor element easily by adjusting immersion time. At this time, as described above, it is necessary to select an immersion time in which the atomic ratio of the catalytic metal to carbon is 1% to 25%.

  As described above, the method for manufacturing a hydrogen gas sensor according to Embodiment 1 can provide a highly sensitive hydrogen gas sensor that can operate at room temperature by catalytic metal modification by oxidation-reduction reaction, and can be manufactured easily and inexpensively. Moreover, according to this hydrogen gas sensor, hydrogen in the air having a concentration of at least 0.01% to 0.1% can be detected reversibly within 1 minute at room temperature.

  In addition, since the response of this hydrogen gas sensor increases as the hydrogen concentration increases, it is easy to fabricate a highly sensitive sensor, and the rate of change in conductance time immediately after exposure is almost proportional to the hydrogen concentration, enabling high-speed response. become. And the hydrogen gas sensor excellent in hydrogen selectivity can be provided.

  (Embodiment 2)

  Next, a hydrogen gas sensor and a hydrogen gas sensor manufacturing method according to Embodiment 2 of the present invention will be described. FIG. 14 is an explanatory diagram showing the state of the cathodic oxidation-reduction reaction of the hydrogen gas sensor according to Embodiment 2 of the present invention, and FIG. 15 is a sensor response diagram of the hydrogen gas sensor according to Embodiment 2 of the present invention. Since the description of the hydrogen gas sensor of Embodiment 1 and the CNT accumulating device is the same in Embodiment 2, the drawings are referred to and the description is omitted.

  In FIG. 13, reference numeral 21 denotes a cathode obtained by immersing the microelectrodes 2a and 2b in which CNTs are accumulated in a catalyst metal salt or complex solution, and an anode provided by immersing the anode separately. It is a power supply which applies a DC voltage between.

  In the second embodiment, a DC voltage is applied between one of the microelectrodes 2 a and 2 b cross-linked with CNTs as a cathode and an anode 22 by a DC power supply 22. Thereby, a catalytic metal, for example, Pd is deposited on the surfaces of the CNT and the microelectrodes 2a and 2b. The hydrogen gas sensor according to the first embodiment is modified by a very simple process using the difference in ionization tendency between the catalyst metal and the CNT. However, the hydrogen gas sensor according to the second embodiment electrochemically applies the catalyst metal to the catalyst metal. The only difference is that it is reliably deposited.

  FIG. 15 shows the sensor response to the air diluted hydrogen (concentration 1%) of the hydrogen gas sensor provided with the CNT modified with the catalytic metal by the cathodic reduction reaction of the second embodiment. The sensor response was almost the same as the sensor response, the response amount, and the response speed of the hydrogen gas sensor modified with the catalytic metal by the oxidation-reduction reaction only by immersion shown in FIG. There is almost no difference.

  As described above, the hydrogen gas sensor and the hydrogen gas sensor manufacturing method according to the second embodiment of the present invention can provide a highly sensitive hydrogen gas sensor capable of operating at room temperature by reliably modifying the catalytic metal body to CNT by electroplating. It can be easily and inexpensively manufactured. Moreover, according to this hydrogen gas sensor, hydrogen in the air having a concentration of at least 0.01% to 0.1% can be detected reversibly within 1 minute at room temperature.

  The present invention can be applied to a hydrogen gas sensor that detects leakage of hydrogen gas from a hydrogen gas plant, a hydrogen station, or a fuel cell.

Explanatory drawing of the hydrogen gas sensor and CNT integrated device in Embodiment 1 of this invention Explanatory drawing which shows the state of the liquid phase oxidation-reduction reaction of the hydrogen gas sensor in Embodiment 1 of this invention (A) SEM image photograph before depositing catalyst metal particles of hydrogen gas sensor in Embodiment 1 of the present invention, (b) SEM image photograph after depositing catalyst metal particles of (a) EDX analysis diagram of CNT subjected to reduction reaction treatment of hydrogen gas sensor in Embodiment 1 of the present invention (A) SEM image photograph when catalyst metal particles are deposited on the CNT of the hydrogen gas sensor in Embodiment 1 of the present invention at an atomic ratio of 6%, (b) the catalyst on the CNT of the hydrogen gas sensor of (a) SEM image when metal particles are deposited at an atomic ratio of 80% (A) EDX analysis diagram when the catalyst metal particle atomic ratio of the hydrogen gas sensor of Embodiment 1 of the present invention is less than 1%, (b) EDX when the catalytic metal particle atomic ratio exceeds 25% Analysis chart Response diagram for hydrogen gas when the hydrogen gas sensor according to Embodiment 1 of the present invention has a sensor element with an atomic ratio of 25% Principle diagram of sensor response when hydrogen gas is detected by hydrogen gas sensor according to Embodiment 1 of the present invention (A) Response diagram of the sensor element immersed in the hydrogen gas sensor for 1 hour according to Embodiment 1 of the present invention with respect to air diluted hydrogen, (b) Sensor element subjected to the immersion treatment for 5 minutes of the hydrogen gas sensor according to Embodiment 1 of the present invention Response diagram for air diluted hydrogen Response diagram when the hydrogen gas concentration of the hydrogen gas sensor in Embodiment 1 of the present invention is changed Relationship diagram between hydrogen gas concentration and conductance time change rate of hydrogen gas sensor in Embodiment 1 of the present invention (A) Response diagram for NO ₂ before catalytic metal modification of the hydrogen gas sensor in Embodiment 1 of the present invention, (b) Response diagram for NO ₂ after catalyst metal modification of the hydrogen gas sensor in Embodiment 1 of the present invention. is there. (A) Response diagram when the catalytic metal of the hydrogen gas sensor in Embodiment 1 of the present invention is palladium chloride, (b) Response diagram when the solvent of the hydrogen gas sensor in Embodiment 1 of the present invention is acetone. Explanatory drawing which shows the state of the cathode oxidation reduction reaction of the hydrogen gas sensor in Embodiment 2 of this invention Sensor response diagram of hydrogen gas sensor according to Embodiment 2 of the present invention

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Hydrogen gas sensor 2a, 2b Micro electrode 3a, 3b Electric field concentration part 4 Insulating substrate 5 Sealed chamber 6,22 Power supply 7 Impedance adjustment apparatus 8 CNT suspension 8a CNT
9a Introduction path 9b Discharge path 10 Pump 21 Anode

Claims (10)

A pair of electrodes, a sensor element in which a carbon nanomaterial is dielectrophoretically crosslinked between the electrodes, and a catalytic metal body that modifies the surface of the carbon nanomaterial to dissociate and adsorb hydrogen gas. A hydrogen gas sensor for detecting a hydrogen gas concentration by measuring a change in conductance or resistance of the carbon nano-particles when the catalytic metal body for modifying the surface is immersed in a solution of a catalyst metal salt or complex. Catalyst metal particles deposited by oxidation-reduction reaction of the material and the catalyst metal, and catalyst metal particles deposited from the oxidation-reduction reaction of the catalyst metal and the electrode material occurring in the solution in superposition with the oxidation-reduction reaction And a hydrogen gas sensor. 2. The hydrogen gas sensor according to claim 1, wherein the atomic ratio of the catalytic metal to carbon is 1% to 25%. 3. The hydrogen gas sensor according to claim 1, wherein the pair of electrodes are immersed in the solution to form a cathode, and a direct-current voltage is applied to the separately immersed anode to cause a cathode oxygen reduction reaction. A pair of electrodes, a sensor element obtained by dielectrophoresis of a carbon nanomaterial and cross-linking between the electrodes, and a catalytic metal body that modifies the surface of the carbon nanomaterial to dissociate and adsorb hydrogen gas, A hydrogen gas sensor for detecting a hydrogen gas concentration by measuring a change in conductance or resistance, wherein the ionization tendency of the electrode material of the electrode and the carbon nanomaterial is selected to be larger than that of the catalyst metal. A hydrogen gas sensor characterized by that. The solution is a solution in which the catalyst metal salt or complex is dissolved in a solvent having a larger physical adsorption force when comparing the separation force of the carbon nanomaterial from the electrode and the physical adsorption force. The hydrogen gas sensor according to any one of claims 1 to 4. 6. The hydrogen gas sensor according to claim 1, wherein the change in conductance between the electrodes is a time change rate of conductance. A carbon nanomaterial is dielectrophoretically cross-linked between a pair of electrodes formed on an insulating substrate, and once dried, the electrode is immersed in a catalyst metal salt or complex solution that has a lower ionization tendency than the electrode material and deposited. The carbon nanomaterial is modified with a particulate catalytic metal body that generates a redox reaction of the electrode material and the catalytic metal in an overlapping manner, and the particulate nanoparticle derived from the redox reaction is generated on the surface of the carbon nanomaterial. A method for producing a hydrogen gas sensor, comprising depositing a catalytic metal body. The method for producing a hydrogen gas sensor according to claim 7, wherein the atomic ratio of the catalytic metal to carbon is 1% to 25%. When the carbon nanomaterial is dielectrophoretically cross-linked between a pair of electrodes formed on an insulating substrate, and once dried, the physical adsorption force is compared with the separation force of the carbon nanomaterial from the electrode and the physical adsorption force. In a solvent having a larger size, a catalyst metal salt or complex having a smaller ionization tendency than that of the electrode material is dissolved, the electrode is immersed in the solution, and the carbon nanomaterial is dispersed by a particulate catalyst metal body that precipitates. A hydrogen gas sensor characterized by modifying and generating a redox reaction of the electrode material and a catalytic metal in an overlapping manner to deposit a particulate catalytic metal body derived from the redox reaction on the surface of the carbon nanomaterial Production method. The method for producing a hydrogen gas sensor according to claim 9, wherein the pair of electrodes are immersed in the solution to form a cathode, and a direct current voltage is applied to the anode separately immersed to cause a cathode oxygen reduction reaction.