JP5650093B2 - Platinum / metal oxide composite catalyst for fuel cell and production method thereof - Google Patents

Description

  The present invention relates to a platinum / metal oxide composite catalyst for fuel cells having both excellent catalytic activity and durability, and a method for producing the same.

  As various electrode catalysts and gas diffusion electrode catalysts that control energy conversion and substance conversion, generally, metal fine particle supported catalysts obtained by supporting a metal component on a conductive carrier such as carbon are used. In particular, in a fuel cell that is an energy conversion device that directly converts chemical energy into electrical energy by supplying fuel and an oxidant to two electrically connected electrodes and causing the fuel to be oxidized electrochemically. In addition, supported platinum fine particles and supported platinum alloy fine particles are employed as electrode catalysts for the anode and the cathode. It has high fuel oxidation activity as an anode and cathode electrode catalyst of a fuel cell, and high catalytic activity for reduction reaction of oxygen as an oxidant, and is dissolved as an electrode catalyst of an anode and a cathode. This is because an element having high durability has not been found other than noble metals represented by platinum.

  Until now, supported noble metals and alloy-based materials containing noble metals have been used as anode and cathode electrode catalysts for gas diffusion electrodes used in fuel cell gas diffusion electrodes and other electrochemical devices. However, reliance on catalysts containing precious metals, such as platinum, is due to the problem of resources such as reserves and mining, the problem of prices due to their relative scarcity, and the production area. As a result, it is easy to be affected by the international situation. Therefore, research and development has been conventionally conducted using platinum as a novel electrode catalyst in combination with a cheaper and more stable material.

  In addition to the advantage of reducing the amount of platinum, platinum was used in the shell on the particle surface from the standpoint that only the surface of the catalyst promotes the catalytic reaction and the inner core does not contribute to the reactivity of the individual catalyst particles. Electrocatalyst technology having a core / shell structure has attracted attention. As a technique to which the electrode catalyst is applied, Patent Document 1 discloses a catalyst particle having a core / shell structure and having an average diameter of 20 to 100 nm.

Special table 2010-501345

As a result of studying a catalyst having a core / shell structure (hereinafter sometimes referred to as a core-shell catalyst), the present inventors have found that there is a problem in specific activity.
The present invention has been accomplished as a result of diligent research in view of the above circumstances, and an object thereof is to provide a platinum / metal oxide composite catalyst for a fuel cell having both excellent catalytic activity and durability, and a method for producing the same. And

The platinum / metal oxide composite catalyst for a fuel cell of the present invention is a platinum / metal oxide composite catalyst for a fuel cell containing platinum fine particles, a metal oxide, and a carbon material, wherein the metal oxide is Ti Including at least one metal element selected from the group consisting of Sn, Nb, and Ta, and the carbon material includes platinum fine particles and the metal oxide on the surface thereof, and around the platinum fine particles. The metal oxide layer includes a metal oxide layer that surrounds the platinum atom in the platinum fine particles and the surface of the carbon material is in electrical contact, so that the metal oxide layer is formed on the surface of the carbon material and the composite catalyst. At least one conductive channel through the platinum fine particle is provided between the surface and the surface, and the platinum atom in the platinum fine particle and the metal oxide have a bond in the conductive channel.

  In this invention, it is preferable that the platinum atom in the said platinum fine particle and the oxygen defect of the said metal oxide have a coupling | bonding.

  In the present invention, the metal oxide layer preferably has a thickness of 0.2 to 0.6 nm.

In the present invention, the metal oxide layer includes titanium (IV) oxide (TiO ₂ ), tin (IV) oxide (SnO ₂ ), niobium oxide (V) (Nb ₂ O ₅ ), and tantalum oxide (V). It is preferable to include at least one metal oxide selected from the group consisting of (Ta ₂ O ₅ ).

  In the present invention, the platinum fine particles preferably have an average particle size of 0.25 to 2.0 nm.

  In the present invention, the thickness of the metal oxide layer is preferably less than the average particle diameter of the platinum fine particles.

  In the present invention, the carbon material may be at least one carbon material selected from the group consisting of acetylene black, furnace black, carbon black, activated carbon, mesophase carbon, and graphite.

Method for manufacturing a fuel cell platinum-metal oxide composite catalysts of the present invention, platinum particles, metal oxides, and containing carbon material, a process for the preparation of the platinum-metal oxide composite catalyst for a fuel cell, wherein The metal oxide includes at least one metal element selected from the group consisting of Ti, Sn, Nb, and Ta, a step of preparing a platinum ion solution, a step of preparing a precursor solution of the metal oxide, A step of mixing a precursor solution of a metal oxide and the carbon material to prepare a first mixed solution, mixing the first mixed solution and the platinum ion solution, so that at least the metal oxide, A step of preparing a second mixed solution containing a carbon material and platinum ions, a step of bubbling carbon monoxide in the second mixed solution, and a reducing agent added to the second mixed solution after bubbling; White By depositing fine particles on the surface of the carbon material, a metal oxide layer containing the metal oxide and surrounding the platinum fine particles is formed on the surface of the carbon material, and in the metal oxide layer At least one conductive channel via the platinum fine particles is formed between the surface of the carbon material and the surface of the composite catalyst, and the platinum atom and the metal oxide are formed in the conductive channel. It has the reduction | restoration process which forms the coupling | bonding of this.

  In the production method of the present invention, before the step of preparing the second mixed solution, a base is added to the first mixed solution, and a part or all of the precursor of the metal oxide is subjected to the metal oxidation. You may have the process of converting into a thing.

  According to the present invention, a conventional platinum alloy catalyst or a core-shell catalyst is obtained by securing a conductive channel through platinum fine particles on the surface of a carbon material and combining a platinum atom with a metal oxide in the conductive channel. It is possible to exhibit superior catalyst performance and catalyst durability.

It is sectional drawing which showed the typical example of the composite catalyst which concerns on this invention typically, and its enlarged view. In the manufacturing method which concerns on this invention, it is the schematic diagram which showed the chemical structure of the carbon material surface from the process of preparing a 2nd liquid mixture to a reduction process. It is the graph which showed the platinum particle size distribution in the catalyst of Example 1 and Example 2 obtained by the SAXS measurement. 2 is a TEM image of the catalyst of Example 1. 2 is an ESR spectrum of Example 1. It is a schematic diagram for demonstrating the coupling | bonding between a platinum atom and a titanium atom. FIG. 4 shows an XRD spectrum of a catalyst fired under conditions of a firing temperature of 500 ° C., 600 ° C. and 700 ° C. based on the manufacturing method of Comparative Example 2. FIG. It is a schematic diagram of the apparatus used for CO bubbling. It is the isometric view schematic diagram which showed the apparatus which performs an electric potential process.

1. The platinum / metal oxide composite catalyst for fuel cells The platinum / metal oxide composite catalyst for fuel cells of the present invention is a platinum / metal oxide composite catalyst for fuel cells containing platinum fine particles, a metal oxide, and a carbon material. The carbon material has, on the surface thereof, a platinum oxide fine particle and a metal oxide layer containing the metal oxide and surrounding the platinum fine particle, and the platinum atom in the platinum fine particle and the carbon material. The metal oxide layer is provided with at least one conductive channel through the platinum fine particles between the surface of the carbon material and the surface of the composite catalyst. In the conductive channel, the platinum atom in the platinum fine particle and the metal oxide have a bond.

  As described above, from the viewpoint of the high cost of platinum and the limited reserves of platinum, the platinum alloy catalyst for the purpose of improving the catalytic activity of platinum, and the core shell for the purpose of improving the utilization rate of the platinum catalyst Research and development of catalysts are actively conducted. However, since the platinum alloy catalyst and the platinum core-shell catalyst are both a combination of base metals rather than platinum, the base metals are eluted under the operating environment of the fuel cell, and the catalytic activity of platinum and the durability of the catalyst. There is a problem that the performance is lowered.

Regarding the conventional core-shell catalyst, paragraph [0027] of the specification of Patent Document 1 describes that titania (TiO ₂ ) is used as a core. As for the conventional method for producing a core-shell catalyst, paragraph [0051] of the specification of Patent Document 1 includes plasma coating, vapor deposition, physical vapor deposition (PVD), and chemical vapor deposition (CVD). ), Atomic layer deposition (ALD), and the like.
However, knowledge about the three-dimensional growth of platinum on metal oxides as described in known literature (U. Diebold. Et.al, Surf. Sci., 331, 845-854, 109 (1995)). From the knowledge that there is usually no interaction between platinum and metal oxides such as titania, it is extremely difficult to form a platinum shell with a metal oxide as a core. Even so, both the activity and durability are considered to be low.
Paragraph [0061] of the specification of Patent Document 1 discloses Ag / Pt particles having a platinum shell thickness of 3 nm (Example 1). However, at a thickness of 3 nm, it is considered that the advantage of adopting the core / shell structure is extremely small from the viewpoints of catalytic activity and manufacturing cost. Furthermore, it is extremely difficult to achieve a platinum shell with a thickness of a monoatomic layer level (monoatomic layer = 0.23 nm) by the method such as plasma coating described in Patent Document 1.

  As a result of diligent efforts, the present inventors use a platinum / metal oxide composite catalyst having a completely different structure from conventional platinum alloy catalysts and core-shell catalysts, while being able to achieve both excellent catalyst performance and catalyst durability. The present inventors have found that the amount of precious metals can be reduced and cost reduction can be achieved.

  The platinum / metal oxide composite catalyst for a fuel cell according to the present invention (hereinafter sometimes referred to as a composite catalyst) contains platinum fine particles, a metal oxide, and a carbon material. Hereinafter, the platinum fine particles, the metal oxide, and the carbon material used in the present invention will be described in order.

1-1. Platinum fine particles The platinum fine particles used in the present invention are present on the surface of the carbon material. Here, the presence of platinum fine particles on the surface of the carbon material means that the surface of the platinum fine particles and the surface of the carbon material are in contact with each other, and includes an embodiment in which the platinum fine particles are supported on the surface of the carbon material.

  Furthermore, the platinum atoms in the platinum fine particles and the carbon material surface are in electrical contact. In general, when a metal oxide described later is used alone as an electrode catalyst for a fuel cell, it is generally very difficult to exhibit sufficient conductivity. Assuming that the metal oxide layer covers the entire surface of the carbon material, the conductive path connecting the carbon material and the outside of the composite catalyst is blocked by the metal oxide layer, and between the composite catalyst and within the composite catalyst and the fuel cell. There is a possibility that any electrical conductivity between the other members may be impaired. Therefore, when the platinum atom in the platinum fine particle is in electrical contact with the surface of the carbon material, a conductive path, a so-called conductive channel is formed between the surface of the carbon material and the composite catalyst surface via the platinum fine particle, The conductivity between the composite catalysts and between the composite catalyst and other members inside the fuel cell can be ensured.

Here, the conductive channel can be rephrased as an electron conduction path formed of at least one platinum fine particle and penetrating through the metal oxide layer to connect the surface of the carbon material and the outside of the composite catalyst. At least one conductive channel is formed in the metal oxide layer. Although it is considered that there are as many conductive channels as there are platinum fine particles present on the surface of the carbon material, the number of conductive channels may not necessarily match the number of platinum fine particles.
Further, in the conductive channel, the platinum atom in the platinum fine particle and the metal oxide in the metal oxide layer have a bond. This connection will be described in detail later.

The average particle diameter of the platinum fine particles is preferably 0.25 to 2.0 nm.
According to a known document (T, Ioannides et al. Journal of Catalysis 161, 560-569 (1996)) regarding charge transfer by a metal-semiconductor junction model, if the diameter of the metal particles on the metal oxide is 2 nm or less, 0 There is a description that a charge transfer of .5 electrons / (metal atom) occurs and as a result affects the entire metal particle. Therefore, the average particle diameter of the platinum fine particles used in the present invention is preferably 2.0 nm or less from the viewpoint of containing platinum atoms bonded to the metal oxide.
On the other hand, the effective radius of platinum atoms is 1.39 mm (National Astronomical Observatory, “Science Chronology 2008”, Maruzen Co., Ltd., page 486). Therefore, in view of the effective radius, the lower limit of the average particle diameter of the platinum fine particles used in the present invention is 0.25 nm.
When the average particle diameter measured by X-ray small angle X-ray scattering (hereinafter sometimes referred to as SAXS) measurement is 1.5 nm or less, the effect of electron donation (SMSI effect) described later is obtained. From the viewpoint of being able to confirm more clearly, the average particle size of the platinum fine particles used in the present invention is more preferably 0.50 to 1.5 nm, and further preferably 1.0 to 1.5 nm.

The average particle diameter of the particles in the present invention is calculated by a conventional method.
The average particle diameter of the particles can be determined by, for example, SAXS measurement. As SAXS measurement conditions, for example, an X-ray diffractometer (manufactured by Rigaku, model number: RINT-2500, etc.), X-ray output is set to 50 kV, 300 mA, etc., copper is used as a target, and measurement is performed in the atmosphere. Is mentioned.
Another example of the method for calculating the average particle diameter of the particles is as follows. First, in a transmission electron microscope (hereinafter referred to as TEM) image at a magnification of 400,000 times or 1,000,000 times, a particle when the particle is regarded as a spherical particle. Calculate the diameter. The calculation of the particle size by such TEM observation is performed on 200 to 300 particles of the same type, and the average of these particles is defined as the average particle size.

1-2. Metal oxide The metal oxide used in the present invention exists as a metal oxide layer on the surface of the carbon material.
The metal oxide layer used in the present invention preferably covers a part of the surface of the carbon material, and preferably does not completely cover the entire surface of the carbon material. This is because, as described above, if the carbon material that controls the conductivity of the composite catalyst is completely covered with the metal oxide layer, there is a risk that electronic conduction may be hindered.
The coverage of the metal oxide layer with respect to the carbon material is preferably 20 to 99%. If the coverage of the metal oxide layer with respect to the carbon material is less than 20%, the proportion of the metal oxide layer is too small and the platinum fine particles are also small. Therefore, relatively, the proportion of the carbon material in the composite catalyst increases, and the platinum loading rate decreases. As a result, when the membrane / electrode assembly is produced, the composite catalyst is used in order to increase the amount of platinum. As a result, the catalyst layer becomes thick, oxygen diffusion in the catalyst layer decreases, and the discharge performance of the fuel cell decreases. There is a fear. On the other hand, if the coverage of the metal oxide layer on the carbon material exceeds 99%, the surface of the carbon material is completely covered with the metal oxide layer. There is a possibility that the electrical conductivity between the members cannot be maintained.

  The “coverage ratio of the metal oxide layer to the carbon material” as used herein refers to the ratio of the surface area of the carbon material covered with the metal oxide layer when the total surface area of the carbon material is 100%. Point to. An example of a method for calculating the coverage will be described below. First, the amount of metal oxide (A) in the composite catalyst is measured by inductively coupled plasma mass spectrometry (ICP-MS) or the like. On the other hand, the average particle diameter of the composite catalyst is measured by TEM or the like. From the measured average particle diameter, the number of atoms that the particle having the particle diameter has on the surface is estimated, and the amount of metal oxide (B) when one atomic layer on the particle surface is replaced with metal oxide is estimated. A value obtained by dividing the amount of metal oxide (A) by the amount of metal oxide (B) and multiplying by 100 is “the coverage ratio of the metal oxide layer to the carbon material (%)”.

  The metal oxide layer surrounds the platinum fine particles described above. Here, the state in which the metal oxide layer surrounds the platinum fine particles means that the metal oxide constituting the metal oxide layer surrounds the platinum fine particles, and preferably the metal oxide layer covers a part of the surface of the platinum fine particles. However, the metal oxide layer does not completely cover the entire platinum fine particles, and a part of the platinum fine particles appears on the composite catalyst surface. As described above, the metal oxide layer surrounds the periphery of the platinum fine particles, whereby the above-described conductive channel is formed and the electrochemical specific surface area of platinum that can react with oxygen or the like (Electrochemical Surface Area; hereinafter referred to as ECSA). ) Is also secured.

The thickness of the metal oxide layer is preferably less than the average particle size of the platinum fine particles. When the thickness of the metal oxide layer is equal to or greater than the average particle diameter of the platinum fine particles, the fine metal particles are completely buried in the metal oxide layer and contain platinum fine particles that do not appear outside the composite catalyst. This is because the catalytic activity of the oxygen reduction reaction may be impaired.
The output performance of a fuel cell using a platinum catalyst is strongly influenced by the ECSA of platinum. Fuel cells, depending on the application, typically ^are required surface area of 100m 2 ^-Pt / m 2 -MEA more platinum. Depending on the amount of platinum used and the type and amount of ionomer used, if the basis weight is 0.1 mg-Pt / cm ² -MEA, the ECSA of platinum needs to be 100 m ² / g-Pt or more. When the average particle diameter of the platinum fine particles is, for example, 1.5 nm or less, the ECSA is about 200 m ² / g-Pt. When the thickness of the metal oxide layer surrounding the platinum fine particles exceeds 0.6 nm, ECSA that can participate in the oxygen reduction reaction is less than 100 m ² / g-Pt.

From the viewpoint that the platinum fine particles can exhibit oxygen reducing ability without being buried in the metal oxide layer, the thickness of the metal oxide layer is preferably 20 to 60% of the average particle diameter of the platinum fine particles.
The thickness of the metal oxide layer is preferably 0.2 to 0.6 nm. When the thickness of the metal oxide layer exceeds 0.6 nm, the platinum fine particles are buried too much in the metal oxide layer due to the above-mentioned relationship with the average particle size of the platinum fine particles, so that sufficient ECSA of platinum can be secured. There is a risk of disappearing. Further, when the thickness of the metal oxide layer is less than 0.2 nm, the metal oxide may not be present as a compound in the first place depending on the type of the metal oxide. The thickness of the metal oxide layer is more preferably 0.3 to 0.6 nm, and further preferably 0.4 to 0.6 nm. From the same viewpoint, the thickness of the metal oxide layer is preferably a monoatomic layer or more and a tetraatomic layer or less.

  Here, the thickness of the metal oxide layer refers to, for example, the average thickness of the metal oxide layer at any two or more sites obtained from a TEM image of the composite catalyst. The measurement conditions for TEM measurement are as follows. First, energy dispersive X-ray spectroscopy (Energy Dispersion) using a field emission transmission electron microscope (manufactured by JEOL Ltd., model number: JEM-2100F, etc., with Cs correction) equipped with a UTW type Si (Li) semiconductor detector. Spot analysis by X-ray Spectrometry (EDS) is performed, and the mass ratio (M / C) of the metal atom (M) in the metal oxide layer to the carbon atom (C) in the carbon material is calculated. From the mass ratio (M / C), the thickness of the metal oxide layer at the measurement point can be calculated by geometric calculation. The average thickness of the metal oxide layer at two or more measurement points can be set as the thickness of the metal oxide layer in the composite catalyst.

The metal oxide layer includes at least one conductive channel capable of conducting electrons between the surface of the carbon material and the surface of the composite catalyst via the platinum fine particles. The conductive channel is as described above.
In the conductive channel, the platinum atom in the platinum fine particle and the metal oxide in the metal oxide layer have a bond.
As an example in which a platinum layer is formed on a metal oxide, U.S. Pat. A report by Diebold et al. Is known. However, the bond between platinum and titanium itself is not necessarily strong.
In the present invention, from the standpoint that a strong metal support interaction (hereinafter referred to as SMSI) effect is exhibited and both the specific activity and durability of platinum can be improved, oxygen is partially absorbed from a part of the metal oxide. It is preferable that the removed portion, that is, a so-called metal oxide oxygen defect and a platinum atom have a bond.

The metal oxides suitably used in the present invention will be specifically described by giving element names and compound names.
As can be seen from the pH-potential diagram at 25 ° C. of the titanium-water system, under the pH-potential conditions (pH = 0-2, potential = 0.4-1.2 V) in the normal operating environment of the fuel cell In titanium, titanium exists in a state of titanium (IV) oxide (TiO ₂ ). Therefore, for the composite catalyst produced using TiO ₂ for the metal oxide layer, the metal oxide layer is not likely to elute in the normal operating environment of the fuel cell. The same applies to the pH-potential diagrams at 25 ° C. for tin-water, niobium-water, and tantalum-water systems. That is, also about the composite catalyst manufactured using tin oxide (IV) (SnO ₂ ), niobium oxide (V) (Nb ₂ O ₅ ), or tantalum oxide (V) (Ta ₂ O ₅ ) for the metal oxide layer. In the normal operating environment of the fuel cell, there is no possibility that the metal oxide layer will elute.
As mentioned above, it is preferable that the metal oxide used for this invention contains Ti, Sn, Nb, or Ta. The metal oxide may contain only one of these elements, or may contain two or more. An example containing two or more of these elements is TiNb.
The metal oxides used in the present invention _is preferably _{TiO 2, SnO 2, Nb 2} O 5, or _Ta 2 _{O 5.} In particular, TiO ₂ , SnO ₂ , Nb ₂ O ₅ and Ta ₂ O ₅ are ionic compounds. Therefore, TiO ₂ , SnO ₂ , Nb ₂ O _5, and Ta ₂ O ₅ are high by causing ionic oxygen defects on the crystal surface and inside the crystal and bonding the generated oxygen defects and platinum atoms. Catalytic ability can be expressed. The metal oxide layer used in the present invention may include only one of these four types of metal oxides, or may include two or more types.

For _{TiO 2, SnO 2, Ta 2} O 5 and _Nb 2 _{O 5,} there is little superiority in terms of stability. However, from the viewpoint of catalytic activity, from the viewpoint of the ease of electron donation to the catalytic element arranged in the oxygen defect, and from the viewpoint of cost, TiO ₂ and SnO ₂ are compared with Ta ₂ O ₅ or Nb ₂ O ₅ . TiO ₂ and SnO are also preferred from the standpoint that reserves, yields, safety to the human body, and metal oxide dispersions (oxide sols) have been established and stable supply is possible. ₂ is more preferable.
As mentioned above, it is more preferable that the metal oxide used in the present invention contains Ti or Sn. The metal oxide used in the present invention is more preferably TiO ₂ or SnO ₂ .
In particular, the selection of TiO ₂ as the metal oxide used in the present invention is significantly more advantageous in terms of cost than conventional catalysts containing Pd (Pd: 700 to 1000 yen / g, TiO ₂ : 100 yen). / Kg).

FIG. 6 is a schematic diagram for explaining a bond between a platinum atom and a titanium atom. An arrow 11 in FIG. 6 is an arrow representing the flow of electrons from trivalent titanium to platinum.
According to the mechanism based on the SMSI theory, the following can be understood. That is, a titanium atom having a bond with a platinum atom usually exists as a partially reduced unstable metal cation (such as Ti ³⁺ ). An unstable valence cation such as Ti ³⁺ usually changes its valence to a more stable high valence cation (Ti ⁴⁺ ), and thus supplies electrons to the platinum of the bond destination. By this electron donation, platinum is easily maintained in a metal state of zero valence, and even if oxidized, the electron is supplied from a metal cation (Ti ³⁺ or the like) having a low unstable valence to zero again. It is considered to return.
From the above principle, it is presumed that the platinum atom bonded to the oxygen defect of the metal oxide has a weak bond with oxygen adsorbed from the gas phase as compared with normal bulk platinum. Therefore, oxygen adsorbed on the platinum atom tends to be oxide ion (O ²⁻ ), and the oxide ion forms a bond with proton (H ⁺ ), so that it can be easily desorbed from platinum as water (H ₂ O). Release.
Thus, the composite catalyst according to the present invention is considered to have a particularly high catalytic activity for promoting the oxygen reduction reaction that occurs at the cathode electrode of the fuel cell as compared with the conventional platinum catalyst.

In order to charge-compensate oxygen defects, it is more preferable that the metal ions in the metal oxide have a plurality of valences and the valence change of the metal ions in the metal oxide is small.
Table 1 below shows the ranking of platinum activity based on the SMSI theory (hereinafter referred to as the SMSI ranking) according to the valence change of titanium (Ti), tin (Sn), niobium (Nb), or tantalum (Ta). And it is the table | surface which put together the order | rank of platinum activity (henceforth a 1st principle calculation order | rank) based on the first principle calculation result. In addition, the SMSI ranking in the following Table 1 ranks the stability from the difference in the electron binding energy (Eb value) of X-ray photoelectron spectroscopy (XPS). Moreover, although the first principle calculation of tin was not carried out, the specific activity of platinum by experiment is larger when titanium is used than when tin is used.

From Table 1 above, it can be seen that the SMSI ranking and the first-principle calculation ranking are almost the same. Therefore, the elements contained in the metal oxide are preferably Ti, Sn, Nb, and Ta in this order. The metal oxides used in the present invention _is preferably in the order of _{TiO 2, SnO 2, Nb 2} O 5, Ta 2 O 5.

From the viewpoint that the SMSI effect is exhibited regardless of the crystallinity of the metal oxide, the metal oxide used in the present invention may be crystalline or amorphous. In particular, when crystalline TiO ₂ is used, anatase type crystals are more preferable, but rutile type or brookite type crystals may be used.

1-3. Carbon materials Specific examples of carbon materials include acetylene black, furnace black, carbon black, activated carbon, mesophase carbon, graphite, channel black, and thermal black; carbonization and activation of materials containing various carbon atoms Treated activated carbon: carbon-based materials such as graphitized carbon, carbon fibers, porous carbon fine particles, carbon nanotubes, carbon porous bodies, and the like can be used. BET specific surface area is preferably 100-2000 m ² / g, more preferably 200~1600m ² / g. Within this range, platinum fine particles can be highly dispersed and supported. Particularly in the present invention, it is preferable to use carbon black such as acetylene black, furnace black, carbon black, activated carbon, mesophase carbon, graphite, etc. as the carbon material. Since these carbon materials can carry highly dispersed platinum fine particles, a composite catalyst having high activity can be obtained.
These carbon carriers may be used alone or in combination of two or more.

With regard to platinum-supported carbon that is standardly used in fuel cells, platinum particles having a large average particle diameter cannot be used, although they are excellent in specific activity and durability in view of cost. This is because when the average particle size is increased, the surface area per 1 g of platinum is reduced, so that an attempt to secure the necessary platinum surface area requires more platinum.
Moreover, in the core / shell structure using a noble metal such as palladium as the core, platinum is only the outermost layer of 1 to 3 atoms, so that the surface area per 1 g of platinum is large. However, in terms of cost, internal noble metal content must also be taken into account, and there is a limit to increasing the average particle size as with the platinum fine particles. In the case of a core-shell catalyst using a palladium core, considering the cost, the average particle size is preferably around 6 nm. When the average particle size is 10 nm with sufficient durability, the potential of the core / shell structure is Cannot fully demonstrate.

On the other hand, the cost of the metal oxide used in the present invention is extremely low, being 1/1000 or less of the cost of the noble metal. Further, the catalytic performance of the composite catalyst according to the present invention tends to depend on the average particle diameter (platinum surface area) of the platinum fine particles, but does not depend so much on the particle diameter of the entire composite catalyst. Therefore, the composite catalyst according to the present invention is different in principle from the core-shell catalyst using a noble metal as a core, and in principle, it is possible to sufficiently exert the catalyst performance even if the overall composite catalyst has an average particle size of 10 nm or more. It is.
The average particle diameter of the composite catalyst according to the present invention is determined by the average particle diameter of the carbon material. Hereinafter, the case where the composite catalyst according to the present invention is used for a catalyst layer of a fuel cell will be described. The average particle diameter of practical carbon fuel carrier carbon (for example, ketjenEC, VulcanXC-72, etc.) is about 30 nm at the maximum. Therefore, in view of the average particle diameter of the platinum fine particles and the thickness of the metal oxide layer, the average particle diameter of the composite catalyst according to the present invention is about 30 nm. However, if a carbon material having an average particle diameter exceeding 30 nm is used, the average particle diameter of the entire composite catalyst can be further increased, but the thickness of the catalyst layer is contradictory.

FIG. 1 is a cross-sectional view schematically showing a typical example of a composite catalyst according to the present invention and an enlarged view thereof. The double wavy line means that the drawing is omitted. Moreover, FIG. 1 is not necessarily a figure that faithfully reflects the particle diameter of platinum particles actually used, the thickness of the metal oxide layer, and the size of the carbon material.
Fig.1 (a) is a cross-sectional schematic diagram of the typical example 100 of the composite catalyst based on this invention. The typical example 100 includes platinum fine particles 1, a metal oxide layer 2 made of a metal oxide, and a carbon material 3. On the surface of the carbon material 3, a platinum fine particle 1 is supported, and a metal oxide layer 2 surrounding the platinum fine particle 1 is formed.

FIG. 1B is an enlarged schematic diagram of the frame 4 surrounded by the alternate long and short dash line in FIG. White double-headed arrows indicate electron conduction paths (conduction paths).
As shown in FIG. 1B, platinum atoms (represented by Pt in a circle) in the platinum fine particles and the surface of the carbon material 3 are in electrical contact. A conductive channel that connects the surface of the carbon material 3 and the surface of the composite catalyst is formed in the metal oxide layer through platinum fine particles. Further, in the conductive channel, at least one of oxygen (represented by O in the circle) that is normally bonded to a metal ion (represented by M in the circle) in the metal oxide is removed by partial reduction, Platinum atoms in the platinum fine particles are arranged in the removed portion, and the platinum atoms and metal ions are adjacent to each other and have a bond.

  As described above, the composite catalyst according to the present invention increases the ECSA of platinum by using platinum fine particles, and combines the platinum atoms in the platinum fine particles with an inexpensive and chemically stable metal oxide, thereby improving the SMSI effect. By providing this, the specific activity and catalyst durability superior to those of conventional platinum catalysts for fuel cells are exhibited. In addition, unlike conventional platinum alloy catalysts and core-shell catalysts, there is no use of unstable metals in the fuel cell operating environment, so both excellent catalytic activity and durability can be achieved, and the amount of noble metals is further reduced. Cost can be reduced.

2. Method for Producing Platinum / Metal Oxide Composite Catalyst for Fuel Cell The method for producing a platinum / metal oxide composite catalyst for a fuel cell according to the present invention comprises platinum fine particles, a metal oxide, and a carbon material. A method for producing a metal oxide composite catalyst, comprising: preparing a platinum ion solution; preparing a precursor solution of the metal oxide; mixing the precursor solution of the metal oxide and the carbon material; Preparing a first mixed solution, and mixing the first mixed solution and the platinum ion solution to prepare a second mixed solution containing at least the metal oxide, the carbon material, and platinum ions. A step of bubbling carbon monoxide in the second mixed solution, and a reducing agent added to the second mixed solution after bubbling to precipitate platinum fine particles on the surface of the carbon material, Forming a metal oxide layer containing the metal oxide and surrounding the platinum fine particles on the surface of the carbon material; and the surface of the carbon material and the surface of the composite catalyst in the metal oxide layer And a reduction step of forming at least one conductive channel through the platinum fine particles and forming a bond between the platinum atom and the metal oxide in the conductive channel. To do.

The present invention includes (1) a step of preparing a platinum ion solution, (2) a step of preparing a precursor solution of a metal oxide, (3) a step of preparing a first mixed solution, and (4) a second mixing. A step of preparing a liquid, (5) a CO bubbling step, and (6) a reduction step. The present invention is not necessarily limited to the above six steps. In addition to the above six steps, for example, a filtration / washing step, a drying step, a pulverizing step, a heating step, an acid treatment step, and an electric potential as described below. You may have a processing process etc.
Hereinafter, the steps (1) to (6) and other steps will be described in order.

2-1. Step of Preparing Platinum Ion Solution The platinum ion solution used in this production method is not particularly limited, but the pH may be adjusted according to the liquidity of the metal oxide precursor solution described later. That is, if the precursor solution of the metal oxide is acidic, the platinum ion solution may be previously acidic (pH: about 1 to 5). If the precursor solution of the metal oxide is basic, the platinum ion solution may be basic (pH: about 9 to 13) in advance. Thus, aggregation of platinum fine particles and metal oxides can be suppressed by matching the liquidity of the metal oxide precursor solution with the liquidity of the platinum ion solution. If the liquidity of the metal oxide precursor solution and the liquidity of the platinum ion solution are different, aggregation of platinum fine particles or metal oxide may occur due to heat of neutralization. A platinum ion solution prepared in advance may be used, or a commercially available one may be used.
The platinum ion concentration in the platinum ion solution is preferably adjusted after calculating in advance the amount of platinum in the composite catalyst to be produced.
A typical example of this step is as follows. That is, the platinum complex crystal is dissolved in an aqueous solution having a desired pH to prepare a platinum ion solution. Examples of the platinum complex that can be used in this step include hexahydroxyplatinum (IV) acid (H ₂ [Pt (OH) ₆ ]), chloroplatinic acid (H ₂ PtCl ₆ .6H ₂ O), and tetrahydroxyplatinum (IV ) Acid (H ₂ [Pt (OH) ₄ ]), platinum chloride (IV) acid (H ₂ [PtCl ₄ ]) and the like.

2-2. Step of Preparing a Metal Oxide Precursor Solution The metal oxide precursor used in the production method refers to a compound that is converted into a metal oxide in any step in the production method. Examples of the metal oxide precursor include salts containing metal ions contained in the metal oxide. The salt containing the metal ion can be converted into a metal oxide by hydrolysis.
The precursor solution of the metal oxide is not particularly limited as long as it is a solution capable of generating a metal oxide by mixing with a platinum ion solution. A metal oxide precursor solution prepared in advance or a commercially available solution may be used.
The concentration of the metal oxide in the precursor solution is preferably adjusted after calculating in advance the amount of metal oxide in the composite catalyst to be produced.
A typical example of this step is as follows. That is, a metal oxide precursor solid is dissolved in an aqueous solution having a desired pH to prepare a metal oxide precursor solution. The metal oxide precursor solid that can be used in this step may be a crystal or an amorphous solid.

2-3. Step of preparing first mixed solution This step is a step of preparing the first mixed liquid by mixing the above-described metal oxide precursor solution and the carbon material.
The carbon material that can be used in this step is as described in the above section “1-3. Carbon material”.
A surfactant may be added to the first mixed solution together with the metal oxide precursor solution and the carbon material. By adding the surfactant, the dispersibility of the carbon material can be mainly improved.

  Examples of the surfactant that can be used in the present invention include octylphenol poly (ethylene glycol ether) (for example, Triton X-100 (trade name, manufactured by Roche)), di (2-ethylhexyl) sodium sulfosuccinate (AOT), poly Oxyethylene nonylphenyl ether, magnesium laurate, zinc caprate, zinc myristate, sodium phenyl stearate, aluminum dicaprylate, tetraisoamylammonium thiocyanate, n-octadecyltri n-butylammonium formate, n-amyltrin-butyl Ammonium iodide, sodium dinonylnaphthalene sulfonate, calcium cetyl sulfate, dodecylamine oleate, dodecylamine propionate, cetyltrimethylan Nitrobromide, stearyltrimethylammonium bromide, cetyltrimethylammonium chloride, stearyltrimethylammonium chloride, dodecyltrimethylammonium bromide, octadecyltrimethylammonium bromide, dodecyltrimethylammonium chloride, octadecyltrimethylammonium chloride, didodecyldimethylammonium bromide, ditetradecyldimethylammonium bromide , Didodecyldimethylammonium chloride, ditetradecyldimethylammonium chloride, (2-octyloxy-1-octyloxymethyl) polyoxyethylene ethyl ether, and the like.

  In the present invention, an ultrasonic homogenizer or the like may be used in order to further improve the dispersibility of the metal oxide precursor and the carbon material. In addition, when the liquid temperature of a 1st liquid mixture rises by a homogenization process, since there exists a possibility that a carbon material may aggregate again, it is preferable to cool after a homogenization process.

2-4. Step of preparing the second mixed solution In this step, the second mixed solution containing at least a metal oxide, a carbon material, and platinum ions is obtained by mixing the first mixed solution and the platinum ion solution described above. Is a step of preparing
In this step, the mixing speed of the first mixed solution and the platinum ion solution is preferably as slow as possible. For example, when the first mixed solution is acidic and the platinum ion solution is basic, the metal oxide precursor, platinum ions, and the carbon material may aggregate due to the heat of neutralization generated during mixing. Because there is. As a method of slowing the mixing speed of the first mixed solution and the platinum ion solution, for example, a method of slowly dropping the platinum ion solution into the first mixed solution, or vice versa, Examples include a method of slowly dropping the mixed solution.

  By mixing with the platinum ion solution, the precursor of the metal oxide is converted into the metal oxide. When the precursor of a metal oxide is a salt containing a metal ion contained in the metal oxide, the platinum ion solution becomes an acid or a base and is converted into a metal oxide by hydrolysis. In this step, part or all of the metal oxide may be deposited as a metal oxide layer on the surface of the carbon material.

Before the step of preparing the second mixed solution, a step of adding a base to the first mixed solution to convert a part or all of the precursor of the metal oxide into a metal oxide may be included. In this way, by adding a base to the first mixed solution in advance, the aggregation of the metal oxide is suppressed, and as a result, the thickness of the resulting metal oxide layer is reduced, and the conduction of the platinum fine particles is easily ensured. The ECSA of platinum can be increased.
Examples of the base that can be used include sodium hydroxide (NaOH), potassium hydroxide (NaOH) and the like, aqueous solutions thereof, sodium borohydride (NaBH ₄ ) and the like. These bases may be used alone or in combination of two or more.

2-5. CO Bubbling Step This step is a step of bubbling carbon monoxide in the second mixed liquid described above.
In this step, by coordinating carbon monoxide to platinum ions, platinum ions can be prevented from aggregating with each other, and platinum fine particles having a smaller particle diameter can be obtained.
The concentration of carbon monoxide is not particularly limited, but is preferably about 10 to 30% from the viewpoint of safety. The CO bubbling time is not particularly limited, but is preferably about 30 minutes to 2 hours from the viewpoint of productivity.

2-6. Reduction step In this step, a reducing agent is added to the second liquid mixture after bubbling, and platinum fine particles are precipitated on the surface of the carbon material, so that the surface of the carbon material contains a metal oxide and surrounds the platinum fine particles. And at least one conductive channel through platinum fine particles is formed between the surface of the carbon material and the surface of the composite catalyst in the metal oxide layer, and This is a step of forming a bond between a platinum atom and a metal oxide in a conductive channel.

In this step, by adding a reducing agent, adhesion (preferably supporting) of platinum fine particles to the surface of the carbon material and formation of a metal oxide layer on the surface of the carbon material proceed. The adhesion of the platinum fine particles and the formation of the metal oxide layer may proceed at the same time, the metal oxide layer may be formed after the platinum fine particles adhere to the surface of the carbon material, or the metal oxide layer is formed. After that, platinum fine particles may adhere to the surface of the carbon material.
When the fine platinum particles adhere after the metal oxide layer is formed on the surface of the carbon material, several modes are conceivable. For example, there are an aspect in which platinum fine particles adhere to the carbon surface in the gap between the metal oxide layers, and an aspect in which platinum fine particles are deposited so as to penetrate the metal oxide layer once formed.
When the metal oxide layer is sufficiently thin as 2 nm or less, conduction of electrons can be ensured even if the platinum fine particles are not in direct contact with the surface of the carbon material.

As shown in FIG. 1 described above, the platinum fine particles adhere to the surface of the carbon material, conductivity between the carbon material and the platinum fine particles is ensured, and the metal oxide layer is formed so as to surround the platinum fine particles. Thus, it means that a conductive channel is formed in the metal oxide layer. This conductive channel is a path that ensures conduction of electrons between the surface of the carbon material and the surface of the composite catalyst.
Furthermore, a bond between a platinum atom and a metal oxide is formed in the conductive channel. The bond, particularly the bond between the platinum atom and the oxygen defect of the metal oxide is as described in the above section “1-2. Metal oxide”.

However, in this step, not all platinum fine particles deposited on the surface of the carbon material are surrounded by the metal oxide layer and do not necessarily have a bond with the metal oxide. That is, it is considered that only platinum fine particles are deposited on the surface of the carbon material, and there is a portion where the metal oxide layer is not formed in the vicinity of the platinum fine particles.
However, the metal oxide is amorphous and has a large number of functional groups such as a peroxide group (—OOH) and a hydroxyl group (—OH) on the surface. Therefore, platinum ions are more likely to be adsorbed to the metal oxide than the carbon material, and it is considered that platinum fine particles are preferentially deposited on the surface of the carbon material on which the metal oxide layer is formed and on the surface of the carbon material in the vicinity of the metal oxide layer. .
In this step, only the platinum fine particles are precipitated, and the portion where the metal oxide layer is not formed in the vicinity of the platinum fine particles is preferably 5% or less of the total surface area of the carbon material, and preferably 1% or less. It is more preferable.

The reducing agent that can be used in this step is not particularly limited as long as it is a reducing agent having a strong reducing power. For example, sodium borohydride (NaBH ₄ ), hydrogen, hydrazine, sodium thiosulfate, citric acid, sodium citrate, Examples include L-ascorbic acid and formaldehyde. These reducing agents may be used alone or in combination of two or more.
In this step, CO bubbling may be performed from the viewpoint of enhancing the dispersibility of the precipitated platinum fine particles together with the use of a reducing agent. The concentration of carbon monoxide used for CO bubbling and the CO bubbling time are as described above.

2-7. Other Steps After the reduction step, the obtained composite catalyst may be subjected to filtration / washing, drying, pulverization, heating, acid treatment, potential treatment, and the like.
The filtration / washing of the composite catalyst is not particularly limited as long as it is a method capable of removing impurities without impairing the structure of the produced composite catalyst. Examples of the filtration / washing include a method of separating by suction filtration using a filter paper (Whatman, # 42), PTFE membrane filter (0.25 μm), etc. using distilled water as a solvent. Moreover, the obtained filtrate, further or distilled water, may be washed by NH ₄ CO ₃ aqueous solution and the like.
The drying of the composite catalyst is not particularly limited as long as the method can remove the solvent and the like. Examples of the drying include a method of vacuum drying for 10 to 24 hours under a temperature condition of 60 to 100 ° C.
The pulverization of the composite catalyst is not particularly limited as long as the solid catalyst can be pulverized. Examples of the pulverization include pulverization using a mortar and the like, and mechanical milling such as a ball mill, a turbo mill, a mechanofusion, and a disk mill.

The heating of the composite catalyst is not particularly limited as long as it can remove impurities without impairing the structure of the manufactured composite catalyst. The heating performed in this step is preferably baking.
Specific examples of the firing are as follows. The heating conditions are not limited to the following.
Atmosphere: Purge with inert gas (such as argon) for 30-120 minutes.
Temperature: 150-1000 ° C, preferably 200-900 ° C
Temperature raising condition: The temperature is raised from room temperature to the above temperature over 30 to 120 minutes.
Holding conditions: Hold at the above temperature for 30 to 120 minutes.

The composite catalyst obtained by the present invention may be subjected to acid treatment or potential treatment in order to activate the platinum fine particles.
FIG. 7 (a) is a part of the XRD spectrum of a catalyst that was calcined under the conditions of calcining temperatures of 500 ° C., 600 ° C., and 700 ° C. based on the production method of Comparative Example 2 described later. As can be seen from FIG. 7A, the peak at 2θ = 40 ° in the spectrum at the firing temperature of 500 ° C. (the peak representing Pt (111)) has almost disappeared in the spectrum at the firing temperature of 700 ° C. On the other hand, from FIG. 7A, a peak at 2θ = 30 ° (a peak representing PtS (002) (101)) which does not appear at all in the spectrum at the firing temperature of 500 ° C. is strong in the spectrum at the firing temperature of 700 ° C. You can see it. These results show that a part of platinum becomes PtS and platinum is oxidized by raising the firing temperature. Therefore, it can be seen that platinum needs to be cleaned in order to perform a correct electrochemical evaluation of the catalyst after the high-temperature calcination.

The acid that can be used for the acid treatment is not particularly limited as long as it does not impair the structure of the obtained composite catalyst. For example, perchloric acid, hydrochloric acid, nitric acid and the like can be used. The acid treatment is completed by immersing the composite catalyst in these acids for a predetermined time.
A normal electrochemical cell can be used for the potential treatment. The electrochemical cell preferably uses a rotating disk electrode. The acid used for the said acid treatment can be used for an electrochemical cell.
FIG. 7B is a cyclic voltammogram showing the result of sweeping the potential for 120 cycles at a potential sweep range of 0.05 to 1.2 V (vs RHE) and a potential sweep rate of 100 mV / sec (hereinafter sometimes referred to as CV). It is. In FIG. 7B, the outer voltammogram represents a voltammogram having a larger number of cycles. As can be seen from FIG. 7B, it can be seen that the peak of platinum becomes clear by repeating the potential treatment.
Note that if the potential treatment is extended after the platinum cleaning is completed, the platinum fine particles may be dissolved.

Hereafter, the manufacturing method of this invention is demonstrated using figures. FIG. 2 is a schematic diagram showing the structure of the surface of the carbon material from the step of preparing the second mixed solution to the reduction step in the manufacturing method according to the present invention. However, this figure shows only one aspect of this manufacturing method. Further, FIG. 2 is not necessarily a diagram that faithfully reflects the size of the carbon material, metal (or its ions), and compound that are actually used.
FIG. 2A is a schematic view showing the structure of the surface of the carbon material immediately after mixing the first mixed solution (including the metal oxide precursor solution and the carbon material) and the platinum ion solution. Metal ions (represented by M ⁴⁺ in a circle) and platinum ions (represented by Pt ⁴⁺ in a circle) are dispersed on the carbon material 3. For convenience of illustration, the surfactant and other additives are omitted.
FIG. 2B is a schematic view showing the structure of the surface of the carbon material after a predetermined time has elapsed after mixing the first mixed solution and the platinum ion solution. Depending on the pH in the mixture, metal ions are converted into metal oxides (oxygen atoms are represented by O in circles and metal atoms are represented by M in circles), while platinum remains in an ionic state.
FIG.2 (c) is the schematic diagram which showed the structure of the carbon material surface after CO bubbling. Bubbled CO (carbon atoms are represented by C in a circle) coordinates to platinum, and a carbonyl complex of platinum is temporarily generated, so that aggregation of platinum fine particles is difficult to occur.
FIG. 2D is a schematic diagram showing the structure of the carbon material surface after the reduction process. When the platinum ions are reduced to platinum, as shown in the figure, oxygen defects (-OM) and platinum atoms (Pt) of the metal oxide are bonded to produce a composite catalyst according to the present invention.

  Specific embodiments of the present invention will be described below in more detail with reference to examples. However, the present invention is not limited to these examples unless it exceeds the gist.

1. Production of platinum / metal oxide composite catalyst for fuel cell [Example 1]
First, 6.465 g of titanium (IV) chloride solution (manufactured by Wako Pure Chemical Industries, product number: 203-08955, Ti: 16 to 17% by mass) was dissolved in 200 mL of ultrapure water to prepare a titanium aqueous solution. Next, 5 g of ketjen carbon powder and a few drops of a surfactant (Triton X-100 (trade name, manufactured by Roche)) diluted 1:10 with distilled water were sequentially added to the aqueous solution. While stirring the mixture, the mixture was dispersed with an ultrasonic homogenizer at 300 W for 30 minutes, and then cooled to room temperature to prepare a titanium-carbon dispersion.
On the other hand, crystals of hexahydroxyplatinic acid (Cataler) were dissolved in a basic solution to prepare a platinum ion solution (Pt amount: equivalent to 4.214 g, pH: about 9 to 12).
While stirring the titanium / carbon dispersion, 82 g of a platinum ion solution was slowly dropped little by little with a dropping funnel to prepare a titanium oxide / platinum / carbon dispersion.

FIG. 8 is a schematic view showing a reactor used for CO bubbling.
CO was supplied from a CO gas cylinder 31 stored in the CO gas cabinet 30. The CO gas cylinder 31 was connected to a valve 34 in the draft 33 via a SUS line 32. The draft shutter was lowered to prevent gaps between the shutter and the experimental table. In addition, a CO detector 35 was installed in the draft 33, and a blow-off 36 was provided in the lower part of the shutter to prepare for CO gas leakage from the draft 33. Based on the display of the gas flow meter 37, the CO gas flow rate was appropriately adjusted by the valve 34. CO gas was introduced into the reaction vessel 39 on the stirrer 38 through a flexible hose and the like, and CO bubbling into the reaction solution was performed. The CO after the bubbling was exhausted to the outside of the draft through the bubbler 40 through the exhaust port above the draft. Note that the bubbler 40 was constantly monitored during CO bubbling, and when gas bubbles in the bubbler 40 changed, the CO gas leakage was warned.
Using the apparatus shown in FIG. 8, 20% CO was bubbled into the obtained titanium oxide / platinum / carbon dispersion for 60 minutes.
1.67 g of sodium borohydride (NaBH ₄ ; hereinafter may be referred to as SBH) was added to the dispersion after CO bubbling, and the mixture was further stirred for 60 minutes while CO bubbling. Thereafter, the dispersion was filtered through a PTFE membrane filter (0.25 μm), and the obtained filtrate was added to 200 mL of NH ₄ CO ₃ aqueous solution (concentration: 6.43 mg / L) and stirred for 30 minutes without heating. . The neutralized filtrate was repeatedly washed with ultrapure water at 80 ° C., and the resulting solid was vacuum-dried for 1 day.

  The vacuum-dried solid was added to the quartz cell and placed in a tubular furnace. The inside of the tube was replaced with 100% G1 argon gas (flow rate = 750 mL / min, replacement time: 60 minutes). After the replacement with argon, the temperature in the tube was increased to 250 ° C. over 60 minutes and maintained at 250 ° C. for 60 minutes. After firing, the mixture was cooled to room temperature, and a platinum / metal oxide composite catalyst for a fuel cell of Example 1 was produced.

[Example 2]
First, 6.465 g of titanium (IV) chloride solution (manufactured by Wako Pure Chemical Industries, product number: 203-08955, Ti: 16 to 17% by mass) was dissolved in 200 mL of ultrapure water to prepare a titanium aqueous solution. Next, 5 g of ketjen carbon powder and several drops of a surfactant (Triton X-100 (trade name, manufactured by Roche) diluted 1:10 with distilled water) were sequentially added to the aqueous solution. While stirring the mixture, the mixture was dispersed with an ultrasonic homogenizer at 300 W for 30 minutes, and then cooled to room temperature. A 1 mol / L sodium hydroxide aqueous solution was added to the mixture after cooling to adjust the pH of the mixture to 10, and 2.0 g of SBH was further added and stirred for 24 hours to prepare a titanium / carbon dispersion.
Thereafter, in the same manner as in Example 1, a platinum ion solution was dropped into the titanium / carbon dispersion liquid, and CO bubbling, filtration, washing, and firing were performed, and the platinum / metal oxide composite catalyst for the fuel cell of Example 2 was obtained. Manufactured.

2. Production of core-shell catalyst for fuel cell [Comparative Example 1]
First, carbon-supported palladium particle powder (manufactured by Basf, 20% Pd / C) was prepared.
Next, copper single atoms were coated on the palladium particles by the Cu-UPD method. Specifically, first, 0.5 g of carbon-supported palladium particle powder and 0.2 g of Nafion (trade name) were dispersed in water, and a mixture paste obtained by filtration was applied to a glassy carbon electrode.
Subsequently, 500 mL of a mixed solution of 0.05 mol / L CuSO ₄ and 0.05 mol / L H ₂ SO ₄ previously bubbled with nitrogen was added to the electrochemical cell. Next, the glassy carbon electrode (working electrode), the reference electrode and the counter electrode were immersed in the mixed solution. After sweeping from 0.8 V (vs RHE) to 0.4 V (vs RHE) at a sweep rate of 0.05 mV / sec, the potential was fixed at 0.4 V (vs RHE) for about 30 minutes, and the surface of the palladium particles was made of copper. A monoatomic layer was deposited.

Next, 1 g of K ₂ PtCl ₄ was dissolved in 500 mL of 0.1 mol / L HClO ₄ to prepare a platinum ion solution. The platinum ion solution was sufficiently stirred and nitrogen was bubbled through the solution in advance.
After the copper monoatomic layer was deposited on the surface of the palladium particles by the above method, the glassy carbon electrode was quickly immersed in a platinum ion solution in a nitrogen atmosphere. It was immersed for 2 hours to deposit a platinum monoatomic layer on the surface of the palladium particles, and a core-shell catalyst for a fuel cell of Comparative Example 1 was produced.

[Comparative Example 2]
First, 462 mL of decane and 20.0.468 g of AOT were added to a 1000 mL beaker, and the mixture was stirred with a magnetic stirrer for 1 hour. Subsequently, 10 g of anatase-type crystalline TiO ₂ sol (manufactured by Taki Chemical Co., Ltd., trade name: Tynoch M-6) was added to the stirred solution, and the mixture was further stirred for 3 hours to obtain reverse micelles containing TiO ₂ fine particles. A dispersion was obtained.

_Then, the _{H 2 PtCl 6 · 6H 2 O} 5.1778g, dissolved in purified water 99 mL, was prepared _H 2 PtCl ₆ aqueous solution of 0.1 mol / L. Next, 462 mL of decane and 20.89 g of AOT were added to a 1000 mL beaker, and the mixture was stirred with a magnetic stirrer for 1 hour. Subsequently, 10.16 mL of a 0.1 mol / L H ₂ PtCl ₆ aqueous solution was added to the stirred solution, and the mixture was further stirred for 1 hour to obtain a reverse micelle dispersion containing platinum ions.

The dispersion of reverse micelles containing TiO ₂ fine particles prepared by the above method and the dispersion of reverse micelles containing platinum ions were mixed and stirred with a magnetic stirrer for 1 hour. Next, 0.565 g of carbon black (Ketjen) was added as a carbon carrier and stirred at 10 ° C. for 30 minutes. Subsequently, 0.38 g of SBH powder was added and stirred for 5 hours. Further, 500 mL of a mixed solution of 2-propanol: ethanol = 4: 1 was added and stirred at 10 ° C. for 30 minutes. The dispersion was subjected to suction filtration to recover a solid as a catalyst precursor.
The collected solid was washed with 500 mL of a mixed solution of decane: alcohol = 4.3: 3.0 and dried under vacuum at 80 ° C. for 18 hours.

The catalyst precursor powder obtained by the above method was calcined under the following conditions.
Initial conditions Room temperature, argon purge 60 minutes (Ar: 750 mL / min, Ar purity: 99.9999%)
・ Temperature rise temperature From room temperature to 700 ° C. over 120 minutes ・ Holding condition Hold at 700 ° C. for 60 minutes. It dried and the core-shell catalyst for fuel cells of the comparative example 2 with which the platinum layer deposited on the surface of the titanium oxide particle was manufactured.

3. Analysis of Catalyst Example 1-Platinum / Metal Oxide Composite Catalyst for Fuel Cell of Example 2 and Comparative Example 1-Core Shell Catalyst for Fuel Cell of Comparative Example 2 (hereinafter, Example 1-Example 2 and Comparative Example 1) -In some cases, the average particle diameter of platinum in the catalyst and the platinum loading rate were measured. In addition, the catalyst of Example 1 to Example 2 was subjected to TEM observation and electron spin resonance (hereinafter referred to as ESR) measurement.

3-1. Measurement of average particle diameter of platinum The average particle diameter of platinum in the catalyst was measured by small-angle X-ray scattering (SAXS) measurement. Detailed measurement conditions are as follows.
-Equipment using SAXS Device: X-ray diffractometer (Rigaku, model number: RINT-2500)
Target: Cu (wavelength 1.541 mm)
X-ray output: 50 kV, 300 mA
Optical system: Multi-layer parallel beam Detector: Scintillator NaI
・ SAXS measurement conditions Step: 2θ = 0.04 °
Average time / step: 5 sec
Scan range: 2θ = 0.08-8.12 °
Fixed divergence slit: 0.03mm
Heating rate, cooling rate: None Atmosphere: In air

FIG. 3 is a graph showing the platinum particle size distribution in the catalysts of Example 1 and Example 2 obtained by SAXS measurement. FIG. 3 is a graph with frequency on the vertical axis and logarithm of particle size (nm) on the horizontal axis.
From FIG. 3, it can be seen that the platinum fine particles in the catalysts of Example 1 and Example 2 both have a very narrow particle size distribution and very little variation in particle size between particles. From FIG. 3, it can be seen that the average particle diameters of the platinum particles of Example 1 and Example 2 are both 1.3 nm.

3-2. Measurement of platinum loading rate The platinum loading rate in the catalyst was measured. First, metal components such as platinum and titanium in the catalyst were dissolved by heating in aqua regia. Next, the amount of platinum in the sample solution was quantified by ICP-emission spectroscopy using an ICP emission spectroscopy analyzer (manufactured by Jarrel Ash, IRIS ADVANTAGE).

3-3. TEM observation The mass ratio of titanium and carbon in the catalyst was measured by TEM observation. Detailed measurement conditions are as follows.
Using a field emission transmission electron microscope (manufactured by JEOL Ltd., model number: JEM-2100F, with Cs correction), at an accelerating voltage of 200 kV, a field of view of 0.25 μm × 0.25 μm (magnification 600,000 times, FIG. a)) or dark field STEM observation (Scanning Transmission Electron Microscopy, hereinafter referred to as STEM) was performed in the range of about 20 nm × 20 nm (magnification 5 million times, FIG. 4B).
Energy dispersive X-ray spectroscopy (EDS) using a field emission transmission electron microscope (manufactured by JEOL, model number: JEM-2100F, with Cs correction) equipped with a UTW type Si (Li) semiconductor detector (manufactured by JEOL) ) Was performed, and the mass ratio of titanium and carbon at the measurement point was calculated.

FIG. 4 is a TEM image of the catalyst of Example 1 obtained under the above observation conditions. FIG. 4A is a TEM image with a magnification of 600,000 times. FIG. 4B is a TEM image at a magnification of 5 million, further enlarging the part surrounded by the frame in FIG.
A white circle having a diameter of about 1 nm, which can be seen in FIG. 4B, shows platinum fine particles. Moreover, the dark part seen in FIG.4 (b) shows a titanium oxide. Moreover, the part seen thinly brightly in FIG. 4 (a) and (b) shows a carbon support | carrier. 4 (a) and 4 (b), it was confirmed that the titanium oxide layer and the platinum fine particles were in contact with each other.
At the two observation points shown in FIG. 4B, the mass ratio of titanium and carbon was measured. At the observation point 1 (white circle in FIG. 4B), the mass ratio of titanium (Ti) to carbon (C) (hereinafter referred to as Ti / C) is 0.13. At the observation point 2 (black circle in FIG. 4B), Ti / C is 0.10. The thickness of the titanium oxide layer calculated by geometric calculation from these Ti / C is 0.6 nm at the observation point 1 and 0.5 nm at the observation point 2. Therefore, the thickness of the titanium oxide layer is 0.6 nm taking the average of the two observation points.

3-4. ESR measurement Attempts were made to detect Ti ³⁺ cations in the catalyst by ESR measurement. Detailed measurement conditions are as follows.
Apparatus: Electron spin resonance apparatus (manufactured by JEOL, model number: JEOL FE-3X)
Measurement temperature (T): room temperature (25 ° C.)
Microwave frequency (Fr): 9.19 GHz
Microwave output (power) (Pw): 2 mW
Central magnetic field (Fd) of the chart: 327.5 ± 7.5 mT
Sweep width from center magnetic field (SW): 327.5 ± 50 mT
Magnetic field sweep time (ST): 2.0 min
Magnetic field modulation width (MW): 0.2 mT
Magnification (gain) (G): x1000
Time constant (TC): 0.1s
AD conversion level (AC): 1
Standard sample: Mn ²⁺ / MgO

FIG. 5 is an ESR spectrum of Example 1. As indicated by an arrow in the figure, in the catalyst of Example 1, a peak was detected at g = 2.0018. This peak is attributed to the Ti ³⁺ cation itself or traces of the Ti ³⁺ cation (for example, Ti ³⁺ cation bonded to platinum). Similarly, in the ESR spectrum of Example 2, a peak attributed to the trace of Ti ³⁺ cation itself or Ti ³⁺ cation was detected.
From the above, the presence of Ti ³⁺ cations in the catalysts of Example 1 and Example 2 was proved.
In the case of platinum / titanium alloy particles and particles that platinum is simply on TiO ₂ without binding to titanium, the peak attributed to the trace of Ti ³⁺ cation itself or Ti ³⁺ cation is usually not detected. . Therefore, from this result, it was proved that a bond between a platinum atom and a titanium atom was present in the catalysts of Example 1 and Example 2.

4). Evaluation of Catalysts Example 1-Example 2 and Comparative Example 1-Comparative Example 2 catalyst and a catalyst in which a platinum-cobalt alloy is supported on a carbon support (PtCo / C, hereinafter referred to as catalyst of Comparative Example 3) Further, a catalyst (Pt / C, hereinafter referred to as catalyst of Comparative Example 4) having platinum fine particles having an average particle diameter of 2.6 nm supported on a carbon carrier, and a platinum fine particle having an average particle diameter of 5.0 nm supported on a carbon carrier. For the catalyst (Pt / C, hereinafter referred to as the catalyst of Comparative Example 5), the catalyst performance and the catalyst durability were evaluated by measurement using a rotating disk electrode.

4-1. Potential Treatment for Catalyst The potential treatment for the catalyst of Example 1-Example 2 and Comparative Example 1-Comparative Example 5 was performed for the purpose of cleaning platinum before measurement with a rotating disk electrode.
FIG. 9 is a schematic perspective view illustrating an apparatus that performs potential processing.
An aqueous perchloric acid solution 52 was added to the glass cell 51, and a rotating disk electrode 54 coated with a catalyst slurry 53 was set. The rotating disk electrode 54 is connected to a tachometer 55. In the perchloric acid aqueous solution 52, in addition to the rotating disk electrode 54, a counter electrode 56 and a reference electrode 57 are disposed so as to be sufficiently immersed in the perchloric acid aqueous solution 52. It is electrically connected to the analyzer. Argon introduction pipe 58 is disposed so as to be immersed in perchloric acid aqueous solution 52, and argon is bubbled into perchloric acid aqueous solution 52 at room temperature for a certain period of time from an argon supply source (not shown) installed outside the cell. As a result, argon was saturated in the perchloric acid aqueous solution 52. A circle 59 indicates an argon bubble.
The details of the apparatus are as follows.
-Perchloric acid aqueous solution: 0.1 mol / L HClO ₄
・ Rotating disk electrode: electrode made of glassy carbon ・ Rotometer: Hokuto Denko, HR-201
・ Counter electrode: Platinum electrode (made by Hokuto Denko)
・ Reference electrode: Hydrogen electrode (manufactured by KM Lab)
Dual electrochemical analyzer: ALS700C manufactured by BAS
With the apparatus shown in FIG. 9, platinum was activated by sweeping the potential for 120 cycles at a potential sweep range of 0.05 to 1.2 V (vs RHE) and a potential sweep rate of 100 mV / sec.

4-2. Measurement Using Rotating Disc Electrode (a) Calculation of ECSA For the catalysts of Example 1-Example 2 and Comparative Example 1-Comparative Example 5, the ECSA of the catalyst was calculated.
With the apparatus shown in FIG. 9, the potential was swept for two cycles at a potential sweep range of 0.05 to 1.2 V (vs RHE) and a potential sweep rate of 50 mV / sec. ECSA was calculated from CV at the second cycle.

(B) Measurement of specific activity and mass activity Electrochemical measurement was performed on the catalysts of Example 1-Example 2 and Comparative Example 1-Comparative Example 5, and the oxygen reduction reaction (Oxygen reduction reaction; hereinafter referred to as ORR). .) Specific activity and mass activity, which are indicators of activity, were measured.
In the apparatus shown in FIG. 9, while bubbling oxygen into the perchloric acid aqueous solution 52 in the glass cell 51, the potential is 2 at a potential sweep range of 0.1 to 1.05 V (vs RHE) and a potential sweep rate of 10 mV / sec. A cycle sweep was performed. An activity-dominated current (hereinafter referred to as IK) was calculated from a current value of 0.9 V in the ORR curve at the second cycle. The value obtained by dividing the IK by the above-mentioned ECSA was defined as specific activity, and the value obtained by dividing the IK by the platinum mass on the glassy carbon electrode was defined as mass activity.

(C) Evaluation of durability Electrochemical measurement was performed on the catalysts of Example 1-Example 2 and Comparative Example 1-Comparative Example 5 to evaluate durability. The detailed conditions of the electrochemical measurement are as follows.
In the apparatus shown in FIG. 9, a rectangular wave potential cycle of 0.65 / 5 sec to 1.2 V / 5 sec is set to 10,000 cycles (vsRHE) while bubbling oxygen into the perchloric acid aqueous solution 52 in the glass cell 51. ) Swept. After sweeping 10,000 cycles, ECV was calculated by performing CV in the same manner as described in the section “(a) Calculation of ECSA” described above.
The ECSA calculated from the CV after 10,000 cycle sweep is divided by the ECSA calculated from the CV of the second cycle described in the section “(a) Calculation of ECSA”, and then multiplied by 100. The ECSA maintenance rate (%) of the catalyst was used.

  Table 2 below summarizes the average particle size, platinum loading, ECSA, specific activity, mass activity, and ECSA retention rate of platinum in the catalysts of Example 1-Example 2 and Comparative Example 1-Comparative Example 5. It is. In Table 2 below, “metal fine particles” refers to platinum fine particles in Example 1 to Example 2 and Comparative Example 4 to Comparative Example 5, and in Comparative Example 1, fine particles in which platinum is coated with platinum. In Comparative Example 2, the fine particles are obtained by coating titanium oxide with platinum. In Comparative Example 3, the fine particles are platinum-cobalt alloy fine particles. In Table 2 below, the platinum loading is a value obtained by dividing the mass of platinum by the mass of the entire catalyst (that is, the sum of the mass of platinum, the mass of the carbon material, and the mass of the metal oxide) and multiplying by 100. It is.

First, the catalyst of Comparative Example 1 in which fine particles in which palladium is coated with platinum (hereinafter referred to as Pt / Pd fine particles) is supported on a carbon carrier will be examined. From Table 2 above, the average particle size of the Pt / Pd fine particles of Comparative Example 1 is 5 nm, and the platinum loading is 30%.
As can be seen from Table 2 above, the ECSA of Comparative Example 1 is 154 m ² / gPt, and the mass activity is 550 A / gPt. These values are the largest values in the catalysts of Example 1-Example 2 and Comparative Example 1-Comparative Example 5. Therefore, the catalyst of Comparative Example 1 has no problem at least with respect to the surface area and mass activity of platinum that can be used as a fuel cell catalyst.
However, as can be seen from Table 2 above, the ECSA maintenance rate of Comparative Example 1 is 40%. This indicates that 60% or more of the platinum surface area is lost from the Pt / Pd fine particles when operated for 10,000 cycles or more. Therefore, it can be seen that the catalyst of Comparative Example 1 is extremely inferior in durability.

Next, the catalyst of Comparative Example 2 in which fine particles obtained by coating titanium oxide with platinum (hereinafter referred to as Pt / TiO _x / TiO ₂ fine particles) are supported on a carbon carrier will be examined. From the above Table 2, the average particle size of the Pt / TiO _x / TiO ₂ fine particles of Comparative Example 2 is 6 nm, and the platinum loading is 20%. The platinum loading of Comparative Example 2 is the lowest among the catalysts of Example 1-Example 2 and Comparative Example 1-Comparative Example 5.
As can be seen from Table 2 above, the specific activity of Comparative Example 2 is 850 μA / cm ² and the ECSA maintenance rate is 96%. These values are the largest values in the catalysts of Example 1-Example 2 and Comparative Example 1-Comparative Example 5. Therefore, the catalyst of Comparative Example 2 has no problem at least with respect to specific activity and catalyst durability.
However, as can be seen from Table 2 above, the ECSA of Comparative Example 2 is 33 m ² / gPt. This value is the smallest value among the catalysts of Example 1-Example 2 and Comparative Example 1-Comparative Example 5. Therefore, the catalyst of Comparative Example 2 has an extremely small surface area of platinum that can be used as a fuel cell catalyst, and is not suitable for practical use.

Subsequently, the catalyst of Comparative Example 3 in which platinum-cobalt alloy fine particles (hereinafter referred to as PtCo fine particles) are supported on a carbon carrier will be examined. From Table 2 above, the average particle size of the PtCo fine particles of Comparative Example 3 is 4.5 nm, and the platinum loading is 30%.
As can be seen from Table 2 above, the ECSA of Comparative Example 3 is 60 m ² / gPt, the specific activity is 450 μA / cm ² , and the mass activity is 270 A / gPt. Therefore, the catalyst of Comparative Example 3 is inferior in mass activity and is not suitable for practical use.
Moreover, the ECSA maintenance rate of Comparative Example 3 is 70%. This indicates that 30% or more of the platinum surface area is lost from the PtCo fine particles when operated for 10,000 cycles or more. Therefore, the catalyst of Comparative Example 3 has poor durability and is not suitable for practical use.

Next, the catalysts of Comparative Example 4 (average particle size: 2.6 nm) and Comparative Example 5 (average particle size: 5.0 nm) in which platinum fine particles are supported on a carbon carrier will be examined. From Table 2 above, the platinum loadings of Comparative Example 4 and Comparative Example 5 are both 30%.
As can be seen from Table 2 above, the ECSA of Comparative Example 4 is 86 m ² / gPt, the specific activity is 230 μA / cm ² , and the mass activity is 199 A / gPt. Moreover, the ECSA of Comparative Example 5 is 56 m ² / gPt, the specific activity is 330 μA / cm ² , and the mass activity is 180 A / gPt. The specific activities of Comparative Example 4 and Comparative Example 5 are extremely low compared to the specific activities of Example 1-Example 2 and Comparative Example 1-Comparative Example 3. Moreover, the mass activity of Comparative Example 4 and Comparative Example 5 is extremely low as compared with the specific activities of Example 1-Example 2 and Comparative Example 1-Comparative Example 3. Therefore, it can be seen that a catalyst in which platinum fine particles are supported on a carbon support is extremely inferior in specific activity and mass activity.
Moreover, the ECSA maintenance rate of Comparative Example 4 is 27%. This indicates that 70% or more of the platinum surface area is lost when operating 10,000 cycles or more. Therefore, it can be seen that the catalyst of Comparative Example 4 is inferior in durability.

On the other hand, from Table 2 above, the average particle diameters of Example 1 and Example 2 are both 1.3 nm and the platinum loading is 33% or 35%.
As can be seen from Table 2, ECSA of Example 1 and Example 2 is ^72m 2 / gPt or ^87m 2 / gPt, specific activity is 570μA / cm ² or 400 .mu.A / cm ^2, the mass activity is 410A / gPt or 350 A / gPt. In particular, the mass activities of Example 1 and Example 2 were as follows: Comparative Example 2 (a catalyst containing Pt / TiO _x / TiO ₂ fine particles), Comparative Example 3 (a catalyst containing PtCo fine particles), and Comparative Example 4 and Comparative Example It is higher than the mass activity of the catalyst containing 5 platinum fine particles.
Moreover, the ECSA maintenance rate of Example 1 and Example 2 is 82% or 86%, and the result is Comparative Example 1 (a catalyst containing Pt / Pd fine particles), Comparative Example 3 (a catalyst containing PtCo fine particles), In addition, the ECSA maintenance rate of the catalyst containing the platinum fine particles of Comparative Example 4 and Comparative Example 5 is higher.
From these results, in the catalysts of Example 1 and Example 2, due to the above-mentioned SMSI effect, electron donation from the trivalent titanium to platinum occurred, and the electronic state of platinum shifted to oxygen adsorption suppression. It can be seen that, unlike the catalyst and the platinum alloy catalyst, both the catalytic activity and the durability are improved.

DESCRIPTION OF SYMBOLS 1 Platinum fine particle 2 Metal oxide layer 3 Carbon material 4 The frame 5 enclosed with a dashed-dotted line 5 The double arrow 11 which shows the conduction direction of an electron The arrow 30 which shows the flow of the electron from trivalent titanium to platinum 30 CO gas cabinet 31 CO gas cylinder 32 SUS line 33 Draft 34 Valve 35 CO detector 36 Blow 37 Gas flow meter 38 Stirrer 39 Reaction vessel 40 Bubbler 51 Glass cell 52 Perchloric acid aqueous solution 53 Catalyst slurry 54 Rotating disk electrode 55 Tachometer 56 Counter electrode 57 Reference electrode 58 Argon Introduction tube 59 Argon bubble 100 Typical example of platinum / metal oxide composite catalyst for fuel cell according to the present invention

Claims (9)

A platinum / metal oxide composite catalyst for a fuel cell, comprising platinum fine particles, a metal oxide, and a carbon material,
The metal oxide includes at least one metal element selected from the group consisting of Ti, Sn, Nb, and Ta,
The carbon material includes, on the surface thereof, the platinum fine particles and the metal oxide layer that includes the metal oxide and surrounds the platinum fine particles,
Since the platinum atom in the platinum fine particle and the surface of the carbon material are in electrical contact, the metal oxide layer is formed between the surface of the carbon material and the surface of the composite catalyst. Comprising at least one conductive channel via
A platinum / metal oxide composite catalyst for a fuel cell, wherein a platinum atom in the platinum fine particle and the metal oxide have a bond in the conductive channel.   The platinum / metal oxide composite catalyst for a fuel cell according to claim 1, wherein a platinum atom in the platinum fine particle and an oxygen defect of the metal oxide have a bond.   The platinum / metal oxide composite catalyst for a fuel cell according to claim 1 or 2, wherein the metal oxide layer has a thickness of 0.2 to 0.6 nm. The metal oxide layer includes titanium (IV) oxide (TiO ₂ ), tin oxide (IV) (SnO ₂ ), niobium oxide (V) (Nb ₂ O ₅ ), and tantalum oxide (V) (Ta ₂ O ₅ ). The platinum / metal oxide composite catalyst for a fuel cell according to any one of claims 1 to 3 , comprising at least one metal oxide selected from the group consisting of: The platinum / metal oxide composite catalyst for a fuel cell according to any one of claims 1 to 4 , wherein the platinum fine particles have an average particle diameter of 0.25 to 2.0 nm. The platinum / metal oxide composite catalyst for a fuel cell according to any one of claims 1 to 5 , wherein a thickness of the metal oxide layer is less than an average particle diameter of the platinum fine particles. 7. The carbon material according to claim 1 , wherein the carbon material is at least one carbon material selected from the group consisting of acetylene black, furnace black, carbon black, activated carbon, mesophase carbon, and graphite. Platinum / metal oxide composite catalyst for fuel cells. A method for producing a platinum / metal oxide composite catalyst for a fuel cell, comprising platinum fine particles, a metal oxide, and a carbon material,
The metal oxide includes at least one metal element selected from the group consisting of Ti, Sn, Nb, and Ta,
Preparing a platinum ion solution;
Preparing a precursor solution of the metal oxide;
Mixing the metal oxide precursor solution and the carbon material to prepare a first mixed liquid;
A step of preparing a second mixed liquid containing at least the metal oxide, the carbon material, and platinum ions by mixing the first mixed liquid and the platinum ion solution;
Bubbling carbon monoxide in the second mixture, and
By adding a reducing agent to the second mixed liquid after bubbling, and precipitating platinum fine particles on the surface of the carbon material,
Forming on the surface of the carbon material a metal oxide layer containing the metal oxide and surrounding the platinum fine particles; and
Forming at least one conductive channel via the platinum fine particles between the surface of the carbon material and the surface of the composite catalyst in the metal oxide layer; and
A method for producing a platinum / metal oxide composite catalyst for a fuel cell, comprising a reduction step of forming a bond between the platinum atom and the metal oxide in the conductive channel. Before the step of preparing the second mixed solution,
The platinum / metal oxide for a fuel cell according to claim 8 , further comprising a step of converting a part or all of the precursor of the metal oxide into the metal oxide by adding a base to the first mixed solution. A method for producing a composite catalyst.