JP2009283213A - Catalyst for fuel cell, method for manufacturing catalyst for fuel cell, and electrode for fuel cell - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide the method for manufacturing a catalyst enhancing performance per surface area by making small the particle size of metal clusters in which metal is covered with ligands as an electrode catalyst for a fuel cell, and to provide an electrode for the fuel cell manufactured by using this. <P>SOLUTION: The catalyst uses metal clusters containing non-platinum elements as a catalytic activity component, the metal cluster includes metals having different valences, and performance per catalyst surface area is enhanced by making small particle size of the metal clusters. The electrode for achieving high performance without using carbon carriers is realized by applying the catalyst to the electrode catalyst of the fuel cell. Since the electrode is made thin, resistance is decreased, and the electrode is applied to a DMFC and a PEFC as a high performance electrode. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a method for producing a catalyst for a fuel cell comprising a metal cluster, and a fuel cell electrode utilizing the result.

  Polymer electrolyte fuel cells (hereinafter abbreviated as PEFC) have been developed as distributed fuel cells for automobiles and general households. In addition, a direct methanol fuel cell (hereinafter abbreviated as DMFC) using methanol as a fuel has been developed as a power source for portable electronic devices.

  The operation principle of the fuel cell will be described below. The fuel cell mainly includes an anode electrode that oxidizes fuel, a cathode electrode that reduces oxygen, and an electrolyte membrane that transmits hydrogen ions provided between the two electrodes. An outline of the operation principle of the fuel cell will be described by taking DMFC as an example.

The DMFC is composed of an anode electrode (fuel electrode) and a cathode electrode (air electrode). At the anode electrode, as shown in the equation (1), methanol, which is a fuel, reacts with water to generate hydrogen ions (hereinafter abbreviated as H ⁺ ), electrons (hereinafter abbreviated as e ⁻ ), and CO ₂ . On the other hand, as shown in the formula (2), H ⁺ that has passed through the electrolyte membrane reacts with O _{2 in} the air supplied from the outside to generate water as shown in the formula (2).

(Equation 1)
Reaction at anode electrode: CH ₃ OH + H ₂ O → 6H ⁺ + CO ₂ + 6e ⁻ (1)

(Equation 2)
Reaction at cathode electrode: 6H ⁺ + 3 / 2O ₂ + 6e ⁻ → 3H ₂ O (2)

  A current can be obtained by connecting the anode and cathode with an external circuit. As described above, the electrode composed of the anode and cathode is the heart of the fuel cell. In DMFC, platinum is generally used as the cathode electrode material and platinum-ruthenium is used as the anode electrode material. In PEFC, platinum is used for both the anode and cathode.

  Platinum is an important component in fuel cells, but it is very expensive. For this reason, the Chemical Society of Japan, 1988, (8), p. 1426-1432 (Non-patent Document 1) has been studied to improve the performance of the catalyst and reduce the amount of platinum used.

  In general, the catalyst performance depends on the amount (surface area) and quality (performance per surface area) of the catalyst active site. As the surface area of the catalyst increases, the number of fields for activating the reaction increases, so the performance improves. In Electrochimica Acta 50, 2004, p817-821 (Non-patent Document 2), in order to increase the amount of catalyst active sites, that is, the surface area of the catalyst, the particle size of the metal cluster serving as the catalyst active sites is reduced and increased. It has been studied to improve the performance by increasing the surface area per unit weight of the catalyst by dispersing it on a carbon support having a surface area. Specifically, for the purpose of improving the dispersion of the metal clusters on the carbon support, a dispersant or the like has been added to the catalyst production process.

The Chemical Society of Japan, 1988, (8), pp. 1426-1432. Electrochimica Acta 50, 2004, p817-821

  Although the development of a catalyst with a reduced amount of platinum has been promoted as described above, the development of a catalyst having a performance exceeding that of platinum has not yet been achieved. In the method of Non-Patent Document 2 described above, a carbon support is used in order to improve the dispersion of the catalyst. As the electrode thickness increases, the electrical resistance increases and the electrode catalyst performance decreases. Further, although the catalyst performance can be improved by increasing the amount of the catalyst active site, the catalyst performance has been limited because the quality of the catalyst has not been improved.

  Accordingly, an object of the present invention is to provide a fuel cell catalyst capable of exhibiting sufficient cell performance without using platinum. In particular, an object of the present invention is to provide a fuel cell catalyst that activates the reaction between the anode and cathode.

  In order to solve the above problems, it is necessary to improve the specific catalyst activity which is the quality of the catalyst. In the present invention, a metal cluster catalyst having a particle size of 200 mm or less composed of a metal other than platinum is used as an active component of the electrode catalyst. The fuel cell electrode is composed of the metal cluster particles and an electrolyte membrane. As a result, the catalyst performance per surface area of the active metal can be improved.

  A feature of the present invention that solves the above-described problem resides in a method for producing a catalyst using a dispersing agent that disperses metal clusters serving as adsorption points of oxygen atoms while ensuring an optimum distance. Metal clusters serve as adsorption points for oxygen atoms. By ensuring that the metal clusters are dispersed at an optimum distance, the supplied oxygen atoms can be taken into oxygen lattice defects formed on the metal clusters without excess or deficiency, and the oxygen reduction reaction can be easily reacted. The dispersant preferably contains a nitrogen atom, and includes polyvinylpyrrolidine (hereinafter abbreviated as PVP), 1-methylimidazole, and the like.

  When the metal clusters are too close to each other, the metal clusters are aggregated. As a result, the particle size of the metal clusters increases, and the surface area effective for the reaction per unit weight of the active metal decreases. The optimum distance between the metal clusters is a distance larger than the particle size of the metal clusters, and it is important to secure the distance to disperse the metal clusters.

  According to the above configuration, the performance per catalyst surface area, that is, the quality of the catalyst can be greatly improved by atomizing the metal cluster as the catalytically active component, and for fuel cells made of highly active metal cluster particles. A catalyst can be provided. In addition, a highly efficient fuel cell electrode can be provided.

  The metal cluster is preferably used by directly supporting it on the electrolyte membrane. Usually, metal clusters are supported on a carbon support for the purpose of increasing the amount of catalyst active sites, that is, the surface area. However, when the carbon support is used, the electrode catalyst becomes thick, so that the electric resistance is increased, and the gas diffusion becomes worse, so that the electrode reaction is difficult to proceed. In the present invention, since the electrode reaction activity of the metal cluster which is the active metal of the catalyst is extremely high, it is possible to provide a highly active electrode catalyst without being dispersed on the support.

  A feature of metal clusters that have been particularly noted, studied and evaluated in the present invention is that the clusters have different valence metals. According to the examination so far, it is preferable that the valence is a mixture of a valence of 0 and a valence of 2 or more. For example, it is preferable that the metal cluster includes palladium, and the palladium valence includes zero valence and valence higher than divalence. Further, at that time, it is desirable that the proportion of palladium atoms having a valence greater than divalent is larger than that of zero-valent palladium atoms.

  The catalyst of the present application can be used for any fuel cell such as PEFC or DMFC, and any electrode of an anode electrode and a cathode electrode. The metal cluster catalyst can be used by directly supporting it on the electrolyte membrane. By providing the electrode catalyst without using platinum or by reducing the amount of platinum by using the catalyst of the present application, it is possible to provide a catalyst with improved electrode performance per price of the metal used than the catalyst using platinum. It is possible to reduce the cost of the fuel cell.

  Further, the present invention is characterized in a portable electronic device or a fuel cell system provided with the above fuel cell.

  In the present invention, it is preferable that a carbon carrier is not used for dispersing the clusters, or a small amount is used even if it is used. Specifically, the carbon weight ratio in the catalyst is preferably 20% or less. In a conventional catalyst for an electrode for a fuel cell, about 50 to 60 wt% of a carbon carrier is used for 40 to 50 wt% of platinum as an active component. Even a highly conductive carbon support (Ketjen Black EC or Vulcan XC-72R) has the effect of increasing the electrical resistance. By reducing the amount of the carbon support as much as possible, or by applying a catalyst composed only of an active metal without using a support, the electrical resistance at the catalyst can be reduced.

  In addition, the metal cluster of the present invention is a fine particle of 50 mm or less, and can exhibit high performance only with the metal cluster, so that the amount of carbon support in the catalyst can be reduced as much as possible.

  According to the present invention, it is possible to provide a catalyst with improved electrode performance per price of metal used compared to conventional platinum catalysts for fuel cells. Therefore, a low-cost fuel cell system can be realized by applying this catalyst to an electrode catalyst of a fuel cell.

  The catalyst manufacturing method of the present invention and a fuel cell electrode produced using the same improve the catalyst performance per surface area of the active metal by reducing the particle size of the metal cluster that becomes the active metal of the catalyst. The metal cluster produced by the catalyst production method is directly dispersed on the electrolyte membrane. In this fuel cell electrode, it is not necessary to use a carbon support that has been used for the purpose of dispersing metal clusters. In the catalyst production, a dispersing agent is used to disperse the metal clusters that serve as the adsorption points of oxygen atoms while ensuring an optimum distance, so that the oxygen lattices formed on the metal clusters without excess or deficiency are supplied. By being taken into the defect, the oxygen reduction reaction can be easily reacted.

The catalyst for a fuel cell of the present invention is characterized by using metal clusters having metal atoms with different valences. A metal cluster is defined as a molecule consisting of a group of three or more metal atoms having a bond between metals and surrounded by a ligand. Metal clusters are a group of unique compounds that are positioned between bulk metals (a simple metal) and metal complexes. A presumed reaction mechanism at the cathode of the fuel cell will be described. First, Pd ⁰ in the palladium cluster as a catalyst adsorbs hydrogen ions in the electrolytic solution. Next, oxidized metal ions of Pd ²⁺ or more adjacent to the adsorption site react with the adsorbed hydrogen to generate H ₂ O. Oxygen in the air supplied to the cathode electrode is taken into oxygen defect sites formed by desorption of oxygen, and hydrogen ion adsorption sites are formed again. By repeating the reaction route described above, the oxidation reaction shown in the formula (2) can be maintained.

  The valence is preferably a mixture of zero and two or more valences. In particular, the number of moles of divalent or higher valent metal ions is preferably larger than the number of moles of zero-valent metal. In particular, for palladium, the ratio of the number of moles of tetravalent palladium to the number of moles of zero-valent palladium is preferably 0.38 or more. The ratio of the number of moles of the zero-valent metal is preferably in the range of 20 to 50%, the ratio of the divalent metal is 20 to 50%, and the ratio of the tetravalent metal is in the range of 10 to 50%.

  The particle size of the metal cluster is preferably 160 mm or less. In particular, in the case of palladium clusters, the particle size is preferably in the range of 40 to 160 inches.

  The metal cluster may be any of gold, tungsten, copper, cobalt, nickel, iron, manganese, palladium, rhenium, osmium, iridium, rhodium, ruthenium and platinum. It is particularly preferable to use palladium. Among precious metals, platinum has the highest activity of single metal, followed by rhodium and palladium in the order of activity. On the other hand, the opposite is true for prices. Therefore, palladium has the lowest price and may have a high activity. The content of palladium in the catalyst is preferably in the range of 5 to 50% by weight.

  The metal cluster is preferably used by directly supporting it on the electrolyte membrane. In order to increase the amount of catalytically active sites, that is, the surface area, metal clusters are usually supported on a carbon support. However, when the carbon support is used, the electrode catalyst becomes thick, so that the electric resistance increases, and the gas diffusion becomes worse, so that the electrode reaction does not proceed easily. This is because the present invention eliminates the need to use a carrier because it is a metal cluster having high activity without being dispersed in the carrier.

  The amount of electricity is calculated from the result of measuring the hydrogen desorption peak by cyclic voltammetry.

(Equation 3)
Catalyst adsorption point -H → H ⁺ + e ⁻ (3)

  As shown in the equation (3), hydrogen adsorbed on the catalyst active site becomes hydrogen ions and emits electrons. The quantity of electricity at this time can be quantified from the area of the hydrogen desorption peak depicted at this time by measuring the current that changes when the potential is changed.

  In this example, a method for producing a catalyst in which a Pd cluster is supported on a conductive carbon support will be described.

(Pd cluster synthesis method)
In the synthesis of Pd clusters, the method of Kaneda et al. (Langmuir, 2002, p. 1849 to 1855) was used as a reference. In this manufacturing method, two types of Pd ₄ clusters are synthesized, and finally, Pd ₂₀₆₀ (NO ₃ ) ₃₆₀ (CH ₃ COO) ₃₆₀ O ₈₀ clusters are synthesized. In particular, this is an example in which the dispersant PVP is applied to cluster synthesis.

0.93 g of palladium acetate and 92.7 cc of acetic acid were placed in a flask, which was placed in an oil bath and heated to 50 ° C. with stirring. 10% -CO / N ₂ gas was bubbled into this solution using a glass pipette as a nozzle. The gas flow rate was 500 cc / min and the ventilation time was 6 hours. When CO was bubbled for a predetermined time, a yellow precipitate was formed at the bottom of the flask. The remaining acetic acid was decanted so that the yellow precipitate was not mixed in, and then a slight amount of remaining acetic acid was evacuated, and evacuated for 30 minutes from the point when the acetic acid ceased to be obtained, thereby obtaining a dried yellow precipitate. The yellow precipitate is a Pd ₄ (CO) ₄ (CH ₃ COO) ₄ .2CH ₃ COOH (abbreviation: PCA) cluster.

  0.39 g of PCA, 0.175 g of 1,10-phenanthrolene monohydrate, 0.386 g of polyvinylpyrrolidine as a dispersant, 7 cc of acetic acid, 7 cc of acetic acid, and put in a two-necked flask at room temperature-atmosphere Stir for minutes.

After adding 0.012 g of Cu (NO ₃ ) ₂ .3H ₂ O to the flask, the inside of the flask was evacuated, and O ₂ gas was supplied into the flask from a tetra bag equipped with an injection needle. An oxygen atmosphere flask was placed in an oil bath and stirred at 90 ° C. for 39 minutes to obtain a Pd1 cluster as a final product as a black precipitate. In the above catalyst production method, a cluster catalyst was also produced by a method using no dispersant, and the performances of the two were compared.

(Carbon support method of Pd1 cluster)
A Schlenk tube was charged with a carbon support and a Pd cluster, and 1-methylimidazole was added thereto. The amount of Pd1 cluster added was such that the Pd loading was about 45 wt%. The carbon support used here is a conductive carbon black support (hereinafter C1). These mixtures were stirred at room temperature for 1 hour. After stirring, the solvent was filtered, washed with distilled water, and air-dried overnight. Thereafter, the remaining dispersant was decomposed by firing in a hydrogen atmosphere or an air atmosphere to finally produce a catalyst having Pd1 clusters supported on a carbon black support (hereinafter referred to as Pd1 clusters / C1). Thus, the dispersant can be applied to impregnate the cluster into the carrier.

  Details of the dispersant will be described below. As the dispersant, one having nitrogen oxide is used. The molecular structure of PVP and 1-methylimidazole is shown in the figure (FIG. 1). PVP is a polymer formed from the monomer units shown. Both of them contain a nitrogen atom in the molecule, and when the lone pair in the nitrogen atom coordinates to the metal in the cluster, the affinity between the cluster and the solvent increases, and the cluster becomes easier to disperse in the solvent. Conceivable.

  Next, the catalyst performance (●) of the Pd1 cluster produced using the dispersant produced in Example 1 was compared with the Pd cluster catalyst (◯) produced without using the dispersant.

  FIG. 2 shows the result of evaluating the oxygen reduction activity per catalyst surface area, which is a performance index of the cathode electrode catalyst, for the Pd cluster / C1 catalyst produced in Example 1 and the Pd cluster catalyst produced without using a dispersant. Indicates. As for the components of the Pt catalyst, about 45 w% of both catalysts is palladium metal and the rest is carbon black.

A method for measuring the oxygen reduction activity of the produced electrode catalyst is shown below. The oxygen reduction activity was performed by the rotating disk electrode method. This method is characterized in that the activity can be evaluated by eliminating the influence of diffusion by utilizing the fact that the supply amount of the reactant is proportional to the 1/2 power of the angular velocity ω (rad / s) of the disk electrode. The electrolyte was an H ₂ SO ₄ solution, and O ₂ bubbling was performed for 1 hour or more before measurement. The measurement temperature is 35 ° C. The measurement of oxygen reduction activity was performed at a stamping speed of 10 mV / s and a stamping range of 0.2 to 1.1 V vs. NHE. When measuring the oxygen reduction activity, the disk electrode of the working electrode was rotated at various rotational speeds. The rotation speed is 400, 625, 900, 1600, 2500 rpm. The reduction current of oxygen increases as the number of reactants supplied increases as the rotational speed increases. The relationship between the reciprocal of the measured current value I (mA) at 0.7 V vs. NHE and the -1/2 power of the angular velocity ω (rad / s) of the electrode is the Koutecky-Levich equation shown in equation (4). expressed.

Where i _K : activity-dominated current (mA), n: number of reaction electrons, F: Faraday constant (c / mol), A: geometric area of disk electrode (cm ² ), c: activity of reactant (mol / ml) ), D: diffusion coefficient of reactant (cm ² / s), v: kinematic viscosity coefficient (cm ² / s) of the solution, and relationship between electrode rotation speed f (rpm) and angular velocity ω (rad / s) Is ω = 2πf / 60. In equation (4), the reciprocal of i _K can be obtained from the intercept of ω ^−1/2 = 0 (ω = ∞, that is, the amount of reactant supplied is infinite). The i _K thus obtained is the net activity of the catalyst without the influence of reactant diffusion. The oxygen reduction current i _R that is the performance of the air electrode is obtained from the equation (5),

(Equation 5)
i _R = i _K / w _M (5)

Here, w _M is the weight (mg) of active metal in the evaluated catalyst.

  Next, a method for calculating the performance per catalyst surface area will be described below.

  The catalyst performance depends on the amount (surface area) and quality (performance per surface area) of the catalyst active site. As the surface area of the catalyst increases, the number of fields for activating the reaction increases, so the performance improves. In this example, the performance per surface area of the catalyst active point necessary for expressing higher performance was evaluated.

  The amount of electricity per unit weight of active metal corresponding to the area of the H desorption peak measured for each catalyst is an index representing the size of the surface area per unit weight of active metal. Table 2 shows the amount of electricity per unit weight of the active metal corresponding to the H desorption peak area measured for each catalyst by cyclic voltammetry. The unit “c” is a quantity of electricity determined by the area of the H desorption peak in cyclic voltammetry (the larger the surface area of the active site). Therefore, c / g is an index representing the size of the surface area per unit weight of the catalyst.

(Equation 6)
S _W = S _H / w _M (6)

  Therefore, the performance per surface area is expressed by equation (7).

(Equation 7)
i _R / S _W (7)

  The horizontal axis represents the potential, and the vertical axis represents the relative value of the oxygen reduction current when the maximum performance value in the evaluated range is 1.0. The larger the relative value of the oxygen reduction current at the same potential, the higher the cathode electrode catalyst performance. When the oxygen reduction current value at 0.5 V of the Pd1 cluster / C1 catalyst produced using the dispersant was set to 1.0, the performance of the Pd cluster / C1 catalyst was dispersed in a voltage region of 0.7 V or less. Compared to the Pd cluster catalyst without using the agent, it showed higher performance. The performance was, for example, 4.8 times at 0.5 V, 2.1 times at 0.6 V, and 1.4 times at 0.7 V.

  Next, the difference in catalyst performance by the production method of Pd clusters was examined.

The method up to the production of PCA is the same as in Example 1, but a dispersant was added in the following steps. After the addition of 0.0156 g of Cu (NO ₃ ) ₂ .3H ₂ O and 0.521 g of polyvinyl pyrrolidine as a dispersant, the flask was evacuated and tetra O ₂ gas was supplied from the back into the flask. A flask in an oxygen atmosphere was placed in an oil bath and stirred at 90 ° C. for 52 minutes to obtain a Pd2 cluster as a final product as a black precipitate.

(Pd2 cluster carbon support method)
A Schlenk tube was charged with a carbon support and a Pd2 cluster, and 1-methylimidazole was added thereto. The amount of Pd2 cluster added was such that the Pd loading was about 45 wt%. The carbon support used here is a conductive carbon black support (hereinafter C1). These mixtures were stirred at room temperature for 1 hour. After stirring, the solvent was filtered, washed with distilled water, and air-dried overnight. Thereafter, the remaining dispersant was decomposed by firing in a hydrogen atmosphere or an air atmosphere to prepare a catalyst (hereinafter referred to as Pd2 cluster / C1) having Pd2 clusters supported on a carbon black carrier.

  Next, the catalyst performance (■) of the Pd2 cluster produced using the dispersant produced in Example 3 was compared with the Pd cluster catalyst (◯) produced without using the dispersant.

  The horizontal axis in FIG. 3 is the potential, and the vertical axis is the relative value of the oxygen reduction current when the maximum performance value in the evaluated range is 1.0. The larger the relative value of the oxygen reduction current at the same potential, the higher the cathode electrode catalyst performance. When the oxygen reduction current value at 0.4 V of the Pd2 cluster / C1 catalyst produced using the dispersant was set to 1.0, the performance of the Pd2 cluster / C1 catalyst was dispersed in a voltage region of 0.7 V or less. Compared to the Pd cluster catalyst without using the agent, it showed higher performance. The performance is, for example, 1.18 times at 0.4 V, 1.16 times at 0.5 V, 1.1 times at 0.6 V, and 1.1 times at 0.7 V compared to the Pd cluster catalyst that does not use a dispersant. It was twice. The performance of the Pd2 cluster / C1 catalyst is lower than that of the Pd1 cluster / C1 catalyst prepared in Example 2. This is because the dispersant is used in a higher temperature condition than in Example 2 in the catalyst production process. Since it was added at ℃, the active metal particle size was increased as a result of partial decomposition of the dispersant or sintering of the synthesized Pd cluster metal, and as a result, the effect of improving the catalyst specific activity by reducing the particle size was reduced. Is expected.

FIG. 4 is a graph showing the X-ray photoelectron spectroscopy (XPS) on the horizontal axis of a catalyst produced by adding a dispersant (●) and a catalyst (◯) produced without adding a dispersant in the process of producing a Pd cluster catalyst. ) The ratio of the Pd ⁴⁺ mole number and the ratio of the Pd ⁰ mole ratio obtained from the measurement, and the vertical axis represents the relative value of the oxygen reduction current. The relative value (specific activity) when the oxygen reduction current value of the catalyst prepared without adding the dispersant and having the largest Pd ⁴⁺ / Pd ⁰ was 1.0 was shown. The potential is 0.6V. ● 1 is a catalyst prepared by adding a dispersing agent in the process of producing Pd ₄ clusters from PCA, as in Example 1. ● 2 is a Pd ₄ to Pd ₂₀₆₀ cluster produced in Example 3. This is a catalyst prepared by adding a dispersant in the process.

The analytical instrument used for the analysis is Shimadzu / KRATOS (model AXIS-HS). Regarding the measurement conditions, the X-ray source was monochrome Al (tube voltage: 15 kV, tube current: 15 mA), the lens conditions were HYBRID (analysis area: 600 × 1000 μm ² ), the resolution was Pass Energy 40, and the scanning speed was 20 eV / min ( 0.1 eV step).

As the ratio of the Pd ⁴⁺ mole ratio to the Pd ⁰ mole ratio increased, that is, as Pd ⁴⁺ increased with respect to Pd ⁰ , the oxygen reduction current increased. Therefore, it is considered that the catalyst performance can be improved if the catalyst is produced so that the ratio of the Pd ⁴⁺ mole ratio and the Pd ⁰ mole ratio in the catalyst is increased.

The Pd ₂₀₆₀ cluster / C1 includes palladium having a plurality of valences. The relationship between metal valence and catalytic activity was investigated. As stated in the specification of patents previously filed in connection with the present invention (application number: 2008-015752), performance with Pd ₂₀₆₀ cluster catalysts supported on carbon supports exists within Pd clusters. As the ratio (Pd ⁴⁺ / Pd ⁰ ) of the existence ratio of Pd ⁴⁺ and Pd ⁰ valence increased, it improved. Using this correlation, the catalyst specific activity and Pd ⁴⁺ / Pd ⁰ of two types of catalysts prepared by adding a dispersant according to the present invention were plotted. As can be seen from the figure, when compared with the same value of Pd ⁴⁺ / Pd ⁰ , the catalyst performance increased with the catalyst prepared by adding the dispersant, the highest in the Pd1 cluster / C1 catalyst, and then the Pd2 cluster / C1. The catalyst became expensive. Next, in the catalyst using the dispersant of the present invention, the factors that improved the specific catalyst activity at the same Pd ⁴⁺ / Pd ⁰ were examined.

Here, how the Pd particles having different valences present in the Pd cluster function in the oxygen reduction reaction will be described in the reaction mechanism of FIG. Here, the black circle indicates Pd ⁰ , the light gray indicates Pd ²⁺ , and the dark gray indicates Pd ⁴⁺ . H ⁺ in the electrolyte is adsorbed on Pd ^0, adjacent to the Pd ^0, Pd ^+, and Pd ²⁺ on oxygen and H to form a reacted H ₂ O. Next, as a result, oxygen lattice defects that are generated take in oxygen in the air supplied to the cathode. By repeating this cycle, the oxygen reduction reaction at the cathode of the fuel cell proceeds.

In consideration of the correlation between Pd ⁴⁺ / Pd ^{0 in} the catalyst obtained from the XPS result of FIG. 4 and the catalyst performance and the reaction mechanism of FIG. 5, in other words, the result of FIG. To achieve this, Pd ^{0 that} is the adsorption point of H ^{+ in} the electrolyte present in the catalyst prepared using the dispersant was adsorbed when it was present in the same amount as the catalyst without added dispersant. Pd ⁴⁺ capable of oxidizing H, that is, less PdO ₂ is required.

  Next, in view of the reaction mechanism described in FIG. 5, the reason why the catalyst specific activity is improved when the dispersion of Pd clusters is improved by using a dispersant will be described with reference to FIG. As shown in FIG. 6 (a), when the Pd cluster particles are poorly dispersed to form a colony, a part of oxygen as a reaction gas cannot be taken into the oxygen lattice defects, and as a result, the reaction proceeds. It becomes difficult. On the other hand, as shown in b), when the dispersion of the Pd clusters is good, all of the oxygen lattice defects on the clusters can effectively take in oxygen.

As described with reference to FIG. 5, oxygen necessary for oxidizing the hydrogen adsorbed on Pd ⁰ is replenished onto Pd as oxygen in the supplied air is taken into oxygen lattice defects, and the reaction cycle continues. Therefore, the amount of oxygen lattice defects that can contribute to the reaction is proportional to the amount of Pd ⁴⁺ capable of oxidation. Therefore, in a catalyst having good dispersion of Pd clusters as shown in FIG. 6B, a catalyst having poor dispersion by forming colonies shown in FIG. ^6A with a small amount of oxygen lattice defects, that is, a small amount of Pd ⁴⁺ . Compared to, it is possible to express the same reaction activity.

  Since the performance per specific surface area and specific activity of Pd1 cluster / C1 catalyst and Pd2 cluster / C1 catalyst prepared by adding a dispersant as described above are extremely high compared to a catalyst not using a conventional dispersant, High performance can be exhibited without using a carbon support that has been used for the purpose of dispersing clusters in conventional catalysts. If only the metal cluster as the active component can be supported without using a carbon carrier in the electrode catalyst, the resistance can be reduced by reducing the electrode thickness compared to the conventional one, and as a result, high electrode performance can be obtained. .

  FIG. 7 shows the dispersion state of Pd particles, which was clarified from the results of scanning transmission electron microscope (STEM) observation of Pd1 cluster / C1 catalyst. It can be seen that the Pd particles indicated by black dots in a) -1 which are 200,000-fold observation results are dispersed on the carbon support. Moreover, it can be seen from the photograph a) -2 magnified 600,000 times that the fine particles are about 200 mm or less. Therefore, in the Pd1 cluster / C1 catalyst, since the dispersion of the Pd cluster is good, the effect shown in FIG. 6B can be expected.

  For comparison, FIG. 8 shows a STEM photograph of a catalyst prepared without adding a dispersant.

  As can be seen from a) -1, which is a 70,000-fold observation result, it was observed that there were many colonies (secondary particles) of about 3,000 cm which are considered to be Pd particles. Further, from the photograph a) -2 magnified 600,000 times, it was found that particles of 50 cm or less considered to be Pd were aggregated in the colony. In other observations, many colonies similar to a) -2 were observed. When Pd particles form colonies in this way, as shown in FIG. 6A, oxygen in the reaction gas, which cannot reach oxygen lattice defects on the Pd cluster surface, cannot be obtained. It will be low.

  FIG. 9 compares the performance of the catalyst (◯) prepared by adding the dispersant PVP from PCA to the Pd cluster step and the catalyst (Δ) prepared by adding PVP to the step of impregnating the Pd cluster with the carbon support (Δ). Both were fired in an air atmosphere to remove residual PVP. As can be seen, the performance of the catalyst made by adding the dispersant PVP from the PCA to the Pd cluster process was found to be higher.

Molecular structure of polyvinylpyrrolidine and 1-methylimidazole. Relationship between potential of Pd1 cluster / C1 catalyst and relative value of oxygen reduction current. Relationship between potential of Pd2 cluster / C1 catalyst and relative value of oxygen reduction current. Relationship between valence ratio and specific activity relative value of Pd1 cluster catalyst. Reaction mechanism in Pd clusters. Improved performance due to improved dispersibility of Pd clusters. Scanning transmission electron microscope observation result of Pd1 cluster / C1 catalyst. Scanning transmission electron microscope observation result of a catalyst with poor Pd cluster dispersion. The relationship between the potential of the catalyst and the relative value of the oxygen reduction current when the dispersant addition method is changed.

Claims (8)

An electrode for a fuel cell having an electrolyte membrane and a catalyst directly dispersed on the electrolyte membrane,
The electrode for a fuel cell, wherein the catalyst contains a metal cluster having a particle size of 200 mm or less. A fuel cell electrode comprising an electrolyte membrane and a catalyst dispersed on the electrolyte membrane,
The catalyst includes a metal cluster having a particle size of 200 mm or less and a carbon material that assists the dispersion of the metal cluster, and a weight ratio of the carbon material in the catalyst is 20 wt% or less. Electrode. A fuel cell electrode according to claim 1 or 2,
The electrode for a fuel cell, wherein the metal cluster includes a metal having a valence of 0 and a metal having a valence of 2 or more. A fuel cell electrode according to claim 1 or 2,
The electrode for a fuel cell, wherein the metal cluster contains palladium having a valence of 0 and palladium having a valence of 2 or more. A fuel cell electrode according to claim 1 or 2,
The electrode for a fuel cell, wherein the metal cluster contains at least one of gold, tungsten, copper, cobalt, nickel, iron, manganese, palladium, rhenium, osmium, iridium, rhodium, ruthenium and platinum. The fuel cell electrode according to claim 3, wherein
An electrode for a fuel cell, wherein the number of moles of a metal having a valence of 2 or more is larger than the number of moles of a metal having a valence of 0. A method for producing an electrode for a fuel cell, comprising a step of synthesizing metal clusters from a metal compound in an oxygen atmosphere, and a step of dispersing the metal clusters in an electrolyte membrane,
A method for producing an electrode for a fuel cell, wherein the metal cluster is dispersed in the electrolyte membrane using a dispersant made of an organic substance having nitrogen atoms.   8. The method for manufacturing a fuel cell electrode according to claim 7, wherein the metal cluster is supported on a carrier and dispersed in the electrolyte membrane.

Images