JP6492986B2 - Manufacturing method of membrane electrode assembly - Google Patents

Description

  The present invention relates to a method for producing a membrane electrode assembly used in a fuel cell.

  In recent years, in response to social demands and trends against the background of energy and environmental problems, fuel cells that can operate at room temperature and obtain high output density have attracted attention as power sources for electric vehicles and stationary power sources. A fuel cell is a clean power generation system in which the product of an electrode reaction is water in principle and has almost no adverse effect on the global environment. In particular, a polymer electrolyte fuel cell (PEFC) is expected as a power source for electric vehicles because it operates at a relatively low temperature.

  Among these, a polymer electrolyte fuel cell using a polymer electrolyte membrane operates at a lower temperature than other types of fuel cells such as a solid oxide fuel cell and a molten carbonate fuel cell. For this reason, the polymer electrolyte fuel cell is expected as a stationary power source or a power source for a moving body such as an automobile, and its practical use has been started.

  Generally, a polymer electrolyte fuel cell uses a membrane electrode assembly in which a pair of catalyst layers and a pair of gas diffusion layers are laminated on a solid polymer electrolyte membrane serving as a proton conductor. In the catalyst layer, an expensive noble metal typified by platinum (Pt) or a platinum alloy is used as a catalyst metal.

  Conventionally, in order to confirm the performance of the catalyst layer in this membrane electrode assembly, whether the catalytic activity is good or bad is determined by measuring the amount of reducing metal of the catalytic metal oxide and the catalytic metal oxide coverage on the surface of the catalytic metal. The technology is disclosed (Non-patent Document 1). The amount of reduced metal of the catalytic metal oxide and the catalytic metal oxide coverage on the catalytic metal surface can be used as an alternative index of IV characteristics for knowing the performance of the fuel cell.

T. Nagai, H. Murata, Y. Morimoto, ECS Transactions, 33 (1) 125-130 (2010), 10.1149 / 1.3484509 @ The Electrochemical Society

  However, the prior art is insufficient to determine the activity of the catalyst layer due to differences in environmental conditions, particularly humidity environmental conditions.

  Therefore, an object of the present invention is to determine whether a catalyst layer used in a membrane electrode assembly for a fuel cell has sufficient activity even when the humidity environment changes. A method for producing a joined body is provided.

  As a result of diligent research to solve the above problems, the present inventors show that the amount of reduction in the amount of reduced electricity of the catalytic metal oxide at low humidity tends to be small in a catalyst layer with low performance at low humidity. And found the present invention.

  That is, in one method of manufacturing a membrane electrode assembly of the present invention, a cathode catalyst layer, a cathode gas diffusion layer, an anode catalyst layer and an anode gas diffusion layer are formed on both surfaces of an electrolyte membrane, respectively. A membrane electrode assembly is manufactured. Then, the catalytic metal oxide reduction electricity quantity of the membrane electrode assembly produced thereon is measured at an arbitrary relative humidity of at least two points. Subsequently, the catalytic metal oxide reducing electricity at the other relative humidity point is divided by the measured catalytic metal oxide reducing electricity at the highest relative humidity point to obtain a reference for the catalytic metal oxide reducing electricity at the highest relative humidity point. The standard value of the catalytic metal oxide reducing electricity quantity at each relative humidity point is obtained. And the membrane electrode assembly whose obtained standard value is a predetermined value or more is selected as a non-defective product.

  In addition, another method of manufacturing a membrane electrode assembly of the present invention includes a cathode catalyst layer, a cathode gas diffusion layer, an anode catalyst layer, and an anode gas diffusion layer formed on both surfaces of an electrolyte membrane, respectively. A membrane electrode assembly is manufactured. The catalytic metal oxide reduction electricity quantity of the membrane electrode assembly produced thereon is measured at an arbitrary relative humidity of at least three points. Subsequently, the catalytic metal oxide reducing electricity quantity at each of the at least three relative humidity points measured is divided by the measured catalytic metal oxide reducing electricity quantity at the highest relative humidity point to obtain the catalytic metal oxide at the highest relative humidity point. The standard value of the catalytic metal oxide reducing electricity quantity at each relative humidity point based on the reducing electricity quantity is obtained. Subsequently, two points having a higher relative humidity are selected from among a plurality of relative humidity, and a linear expression passing through the selected standard values of the two points is created. Subsequently, using the linear equation, a standard value at a relative humidity lower than the relative humidity at the two points where the linear equation was created is calculated. Subsequently, the standard value calculated from the primary equation is compared with the standard value obtained from the measured amount of catalytic metal oxide reduction at the same relative humidity as the relative humidity from which the standard value was calculated from the primary equation. I take the. Then, this difference is regarded as a reduction amount of the catalyst activity at the relative humidity where the standard value is calculated from the linear equation, and the membrane electrode assembly in which the reduction amount is less than the predetermined value is selected as a non-defective product.

  According to the present invention, the active state (catalyst layer utilization rate) of the catalyst layer according to changes in the humidity environment is known by measuring the catalytic metal oxide reducing electricity quantity of the catalyst layer under high and low humidification conditions. Will be able to. For this reason, the catalyst layer activity in the actual operation humidification region can be estimated with high accuracy compared to the prior art, and defective products can be easily selected.

It is a schematic sectional drawing which shows the basic composition of a polymer electrolyte fuel cell. It is a schematic diagram for demonstrating the structure of a catalyst. It is a flowchart which shows the procedure of the manufacturing method of the membrane electrode assembly (MEA) for fuel cells. FIG. 3 is a schematic diagram for explaining the relationship of the state of water adsorbed in the pores of a catalyst layer with respect to water activity (relative humidity) shown by Tetsuya Mashio et al. It is a graph which shows the amount of water adsorb | sucked by the support | carrier (carbon) with respect to water activity (relative humidity) by the catalyst layer of the experiment example shown by the literature of Tetsuya Mashio et al. It is a graph which shows the amount of water adsorb | sucked by the support | carrier (carbon) with respect to water activity (relative humidity) by the catalyst layer from which I / C shown by the literature of Tetsuya Mashio et al. It is a graph which shows an example of the measurement result of an oxide reduction electric quantity. Respective catalytic oxide reducing electricity quantities q _{ox (90% RH)} and q _{ox (40% RH ) of} 90 and 40% relative humidity based on oxide reducing electricity quantity q _{ox (90% RH)} of _90% relative humidity ₎ Is a graph showing the relationship between the standard value and relative humidity. It is a graph which shows the relationship between the catalyst layer utilization rate calculated | required from the oxide reduction electric energy in Example 1, and the activity fall rate calculated | required from the ORR activity mentioned above. The respective catalytic oxide reducing electricity quantities q _{ox (90% RH)} , q _{ox of} relative humidity 90, 70% and 40% relative to the oxide reducing electricity quantity q _{ox (90% RH)} of _90% relative humidity. It is the graph which showed the relationship between the standard value of _{(70% RH)} and _{qox (40% RH)} , and relative humidity. It is a graph which shows the relationship between the catalyst layer utilization rate calculated | required from the oxide reduction electric energy in Example 2, and the activity fall rate calculated | required from ORR activity. Using a catalyst layer having a high low humidity activity, the catalytic oxide reducing electricity with relative humidity 90, 70%, and 40% relative to the oxide reducing electricity quantity q _{ox (90% RH)} at _90% relative humidity. It is the graph which showed the relationship between the standard value of quantity _{qox (90% RH)} , _{qox (70% RH)} , and _{qox (40% RH)} , and relative humidity.

  Hereinafter, an embodiment of the present invention will be described in detail with appropriate reference to the drawings. However, the present invention is not limited only to the following embodiments. Each drawing is exaggerated for convenience of explanation, and the dimensional ratio of each component in each drawing may be different from the actual one. Further, when embodiments of the present invention are described with reference to the drawings, the same reference numerals are given to the same elements in the description of the drawings, and redundant descriptions are omitted.

<Solid polymer fuel cell>
First, the configuration of the polymer electrolyte fuel cell according to the embodiment of the present invention will be described. A fuel cell includes a membrane electrode assembly (MEA), a pair of separators each including an anode side separator having a fuel gas flow path through which fuel gas flows and a cathode side separator having an oxidant gas flow path through which oxidant gas flows. Have.

  FIG. 1 is a schematic diagram showing a basic configuration of a polymer electrolyte fuel cell (PEFC) 1 which is a specific example of a fuel cell. The PEFC 1 first has a solid polymer electrolyte membrane 2 and a pair of catalyst layers (an anode catalyst layer 3a and a cathode catalyst layer 3c) that sandwich the membrane. The laminate of the solid polymer electrolyte membrane 2 and the catalyst layers (3a, 3c) is further sandwiched between a pair of gas diffusion layers (GDL) (anode gas diffusion layer 4a and cathode gas diffusion layer 4c). Thus, the polymer electrolyte membrane 2, the pair of catalyst layers (3a, 3c), and the pair of gas diffusion layers (4a, 4c) constitute a membrane electrode assembly (MEA) 10 in a stacked state.

  In PEFC 1, the MEA 10 is further sandwiched between a pair of separators (anode separator 5a and cathode separator 5c). In FIG. 1, the separators (5 a, 5 c) are illustrated so as to be located at both ends of the illustrated MEA 10. However, in a fuel cell stack in which a plurality of MEAs are stacked, the separator is generally used as a separator for an adjacent PEFC (not shown). In other words, in the fuel cell stack, the MEAs are sequentially stacked via the separator to form a stack. In an actual fuel cell stack, a gas seal portion is disposed between the separator (5a, 5c) and the solid polymer electrolyte membrane 2, or between the PEFC 1 and another adjacent PEFC. These descriptions are omitted in FIG.

  The separators (5a, 5c) can be obtained, for example, by forming a concavo-convex shape as shown in FIG. 1 by subjecting a thin plate having a thickness of 0.5 mm or less to a press treatment. The protrusions seen from the MEA side of the separators (5a, 5c) are in contact with the MEA 10. Thereby, the electrical connection with MEA 10 is ensured. In addition, a recess (space between the separator and the MEA generated due to the uneven shape of the separator) viewed from the MEA side of the separator (5a, 5c) is used for circulating gas during operation of the PEFC 1. Functions as a gas flow path. Specifically, a fuel gas (for example, hydrogen) is circulated through the gas flow path 6a of the anode separator 5a, and an oxidant gas (for example, air) is circulated through the gas flow path 6c of the cathode separator 5c.

  On the other hand, the recess viewed from the side opposite to the MEA side of the separators (5a, 5c) serves as a refrigerant flow path 7 for circulating a refrigerant (for example, water) for cooling the PEFC during operation of the PEFC 1. The Further, the separator is usually provided with a manifold (not shown). This manifold functions as a connection means for connecting cells when a stack is formed. With such a configuration, the mechanical strength of the fuel cell stack can be ensured.

  In addition, in embodiment shown in FIG. 1, the separator (5a, 5c) is shape | molded by the uneven | corrugated shape. However, the separator is not limited to such a concavo-convex shape, and may be any form such as a flat plate shape and a partially concavo-convex shape as long as the functions of the gas flow path and the refrigerant flow path can be exhibited. Also good.

  The type of the fuel cell is not particularly limited. In the above description, the polymer electrolyte fuel cell has been described as an example, but in addition to this, an alkaline fuel cell, a direct methanol fuel cell, a micro fuel A battery etc. are mentioned. Among them, a polymer electrolyte fuel cell (PEFC) is preferable because it is small and can achieve high density and high output. The fuel cell is useful as a stationary power source in addition to a power source for a moving body such as a vehicle in which a mounting space is limited. Among them, it is particularly preferable to use as a power source for a mobile body such as an automobile that requires a high output voltage after a relatively long time of operation stop.

  The fuel used when operating the fuel cell is not particularly limited. For example, hydrogen, methanol, ethanol, 1-propanol, 2-propanol, 1-butanol, secondary butanol, tertiary butanol, dimethyl ether, diethyl ether, ethylene glycol, diethylene glycol and the like can be used. Of these, hydrogen and methanol are preferably used in that high output is possible.

  Further, the application application of the fuel cell is not particularly limited, but it is preferably applied to a vehicle. The membrane electrode assembly described above is excellent in power generation performance and durability, and can be downsized.

(Catalyst layer)
The catalyst layer is used for a membrane electrode assembly. Generally, the catalyst layer includes a catalyst and an electrolyte. The catalyst is formed by supporting a catalyst metal on a catalyst carrier (also simply referred to as a carrier).

(catalyst)
Here, the structure of the catalyst will be described. FIG. 2 is a schematic diagram for explaining the structure of the catalyst.

  As shown in FIG. 2, the catalyst 20 includes a catalytic metal 22 and a support 23. The carrier 23 of the catalyst 20 has primary pores 24 (also referred to as mesopores). The primary holes 24 are holes having a radius of 1 nm or more. There are also micropores 25 which are pores of less than 1 nm.

  It suffices that at least a part of the catalyst metal 22 is supported in the primary pores 24, and a part of the catalyst metal 22 may be provided on the surface of the carrier 23.

  The catalyst is in contact with an ionomer (solid proton conducting material) 26. The ionomer 26 does not enter the mesopores 24 of the carrier 23. For this reason, the catalytic metal 22 on the surface of the carrier 23 comes into contact with the ionomer 26, but the catalytic metal 22 supported in the mesopores 24 is not in contact with the ionomer 26. The catalytic metal in the mesopores 24 forms a three-phase interface between oxygen gas and water in a non-contact state with the ionomer, thereby ensuring a reaction active area of the catalytic metal.

(Catalyst metal)
Although an example of a catalyst metal is introduced, the catalyst layer as an inspection target in the present embodiment is not limited to the one using such a catalyst metal.

  The catalytic metal has a function of catalyzing an electrochemical reaction. As the catalyst metal, one containing at least platinum is preferably used in order to improve catalyst activity, poisoning resistance to carbon monoxide, heat resistance and the like. That is, the catalytic metal is platinum or contains a metal component other than platinum and platinum.

  The metal component other than platinum is not particularly limited and can be used in the same manner as a known catalyst component. Specifically, ruthenium, iridium, rhodium, palladium, osmium, tungsten, lead, iron, copper, silver, chromium, Examples thereof include metals such as cobalt, nickel, manganese, vanadium, molybdenum, gallium, aluminum, and zinc. One or more metal components other than platinum may be used. Especially, it is preferable that it is a transition metal from a viewpoint of catalyst performance. Here, the transition metal atom refers to a Group 3 element to a Group 12 element, and the type of the transition metal atom is not particularly limited. From the viewpoint of catalytic activity, the transition metal atom is preferably selected from the group consisting of vanadium, chromium, manganese, iron, cobalt, copper, zinc and zirconium.

  The composition of the alloy depends on the type of metal to be alloyed. For example, the content of platinum is preferably 30 to 90 atomic%, and the content of the metal alloyed with platinum is preferably 10 to 70 atomic%. In general, an alloy is a generic term for a metal element having one or more metal elements or non-metal elements added and having metallic properties. The alloy structure consists of a eutectic alloy, which is a mixture of the component elements as separate crystals, a component element completely melted into a solid solution, and a component element composed of an intermetallic compound or a compound of a metal and a nonmetal. There is what is formed, and any may be sufficient.

  The shape and size of the catalyst metal (catalyst component) are not particularly limited, and the same shape and size as known catalyst components can be adopted. As the shape, for example, a granular shape, a scale shape, a layered shape, and the like can be used, but a granular shape is preferable. At this time, the average particle diameter of the catalyst particles is preferably 1 to 30 nm, more preferably 2 to 10 nm. When the average particle diameter of the catalyst particles is within such a range, the balance between the utilization rate of the catalyst layer related to the effective electrode area where the electrochemical reaction proceeds and the ease of loading can be appropriately controlled.

(Catalyst carrier)
The catalyst carrier functions as a carrier for supporting the above-described catalyst metal and an electron conduction path involved in the exchange of electrons between the catalyst metal and another member. The catalyst carrier is not particularly limited as long as it has a specific surface area for supporting the catalyst particles in a desired dispersed state and has sufficient electronic conductivity as a current collector. For example, the main component is preferably carbon. Specific examples include carbon particles made of carbon black, activated carbon, coke, natural graphite, artificial graphite and the like. “The main component is carbon” means that the main component contains carbon atoms, and is a concept that includes both carbon atoms and substantially carbon atoms. In some cases, elements other than carbon atoms may be included in order to improve the characteristics of the fuel cell. In addition, being substantially composed of carbon atoms means that mixing of impurities of about 2 to 3% by weight or less is allowed.

The BET specific surface area of the catalyst carrier may be sufficient as long as the catalyst component is supported in a highly dispersed manner. The BET specific surface area of the catalyst carrier precursor before supporting the catalyst component is preferably 700 m ² / g or more, more preferably 800 m ² / g or more, further preferably 1200 m ² / g or more, and 1500 m. ² / g or more is particularly preferable. If the specific surface area is in the range as described above, the catalyst component and the polymer electrolyte are sufficiently dispersed in the catalyst carrier to obtain sufficient power generation performance, and the catalyst component and the polymer electrolyte can be sufficiently effectively used. Oxygen transport resistance is low and easy to control. The upper limit of the BET specific surface area of the catalyst carrier precursor component is preferably 3000 m ² / g or less, and more preferably 2500 m ² / g or less.

  As a method for producing a catalyst carrier precursor having a high specific surface area as described above, methods described in JP 2010-208887 A, WO 2009/075264, etc. are preferably used.

In the present specification, “BET specific surface area (m ² / g catalyst support)” is measured by a nitrogen adsorption method. Specifically, about 0.04 to 0.07 g of a sample (carbon powder, catalyst powder) is precisely weighed and sealed in a sample tube. This sample tube is preliminarily dried at 90 ° C. for several hours in a vacuum dryer to obtain a measurement sample. For weighing, an electronic balance (AW220) manufactured by Shimadzu Corporation is used. In the case of a coated sheet, a net weight of about 0.03 to 0.04 g of the coated layer obtained by subtracting the weight of Teflon (registered trademark) (base material) of the same area from the total weight is used as the sample weight. . Next, the BET specific surface area is measured under the following measurement conditions. On the adsorption side of the adsorption / desorption isotherm, a BET specific surface area is calculated from the slope and intercept by creating a BET plot from the relative pressure (P / P0) range of about 0.00 to 0.45.

[Measurement condition]
-Measuring device: Nippon Bell Co., Ltd. high-precision fully automatic gas adsorption device BELSORP36,
・ Adsorption gas: N ₂ ,
-Dead volume measuring gas: He,
・ Adsorption temperature: 77K (liquid nitrogen temperature),
Pre-measurement treatment: 90 ° C. vacuum drying for several hours (set on measurement stage after He purge),
-Measurement mode: Isothermal adsorption and desorption processes,
Measurement relative pressure P / P ₀ : about 0-0.99
-Equilibrium setting time: 180 sec per relative pressure.

  Further, the size of the catalyst carrier is not particularly limited, but from the viewpoint of controlling the ease of loading, the utilization rate of the catalyst layer, the thickness of the electrode catalyst layer within an appropriate range, etc., the average particle size is 5 to 200 nm, The thickness is preferably about 10 to 100 nm.

  In the electrode catalyst in which the catalyst component is supported on the catalyst carrier, the supported amount of the catalyst component is preferably 10 to 80% by weight, more preferably 30 to 70% by weight, based on the total amount of the electrode catalyst. When the supported amount of the catalyst component is within such a range, the balance between the degree of dispersion of the catalyst component on the catalyst support and the catalyst performance can be appropriately controlled. The “catalyst loading” in the present invention is a value obtained by measuring the weight of the carrier before supporting the catalyst metal and the catalyst after supporting the catalyst metal.

(Catalyst production method)
The method for producing the catalyst (the method for supporting the metal catalyst on the catalyst carrier precursor) is not particularly limited, and a conventionally known method can be used. For example, known methods such as liquid phase reduction method, evaporation to dryness method, colloid adsorption method, spray pyrolysis method, reverse micelle (microemulsion method) can be used. Alternatively, a commercially available electrode catalyst may be used.

  Examples of the method for producing the catalyst by the liquid phase reduction method include a method in which a catalyst metal is deposited on the surface of the catalyst carrier precursor and then heat treatment is performed. Specifically, for example, a method in which a catalyst carrier is immersed in a catalyst metal precursor solution for reduction and then heat treatment is performed.

  Here, the catalyst metal precursor is not particularly limited, and is appropriately selected depending on the type of the catalyst metal used. Specific examples include chlorides, nitrates, sulfates, chlorides, acetates and amine compounds of catalyst metals such as platinum. More specifically, platinum chloride (hexachloroplatinic acid hexahydrate), palladium chloride, rhodium chloride, ruthenium chloride, cobalt chloride and other nitrates, palladium nitrate, rhodium nitrate, iridium nitrate and other nitrates, palladium sulfate, sulfuric acid Preferred examples include sulfates such as rhodium, acetates such as rhodium acetate, and ammine compounds such as dinitrodiammineplatinum nitrate and dinitrodiammine palladium. The solvent used for the preparation of the catalyst metal precursor solution is not particularly limited as long as it can dissolve the catalyst metal precursor, and is appropriately selected depending on the type of the catalyst metal precursor used. Specifically, water, an acid, an alkali, an organic solvent, etc. are mentioned. The concentration of the catalyst metal precursor in the catalyst metal precursor solution is not particularly limited, but is preferably 0.1 to 50% by weight, more preferably 0.5 to 20% by weight in terms of metal. .

  Examples of the reducing agent include hydrogen, hydrazine, sodium borohydride, sodium thiosulfate, citric acid, sodium citrate, L-ascorbic acid, sodium borohydride, formaldehyde, methanol, ethanol, ethylene, carbon monoxide and the like. . Note that a gaseous substance at room temperature such as hydrogen can be supplied by bubbling. The amount of the reducing agent is not particularly limited as long as the catalyst metal precursor can be reduced to the catalyst metal, and known amounts can be similarly applied.

  The deposition conditions are not particularly limited as long as the catalyst metal can be deposited on the catalyst carrier. For example, the deposition temperature is preferably near the boiling point of the solvent, more preferably from room temperature to 100 ° C. The deposition time is preferably 1 to 10 hours, more preferably 2 to 8 hours. In addition, you may perform the said precipitation process, stirring and mixing if necessary. Thereby, the precursor of the catalyst metal is reduced to the catalyst metal, and the catalyst metal is deposited (supported) on the catalyst carrier.

  As heat treatment conditions, for example, the heat treatment temperature is preferably 300 to 1200 ° C, more preferably 500 to 1150 ° C, and particularly preferably 700 to 1000 ° C. The heat treatment time is preferably 0.02 to 3 hours, more preferably 0.1 to 2 hours, and particularly preferably 0.2 to 1.5 hours. In consideration of the reduction promoting effect of the catalyst metal precursor, the heat treatment step is preferably performed in an atmosphere containing hydrogen gas, more preferably in a hydrogen atmosphere.

(Electrolyte (ionomer))
In addition to the electrode catalyst, the catalyst layer contains an ion conductive polymer electrolyte (ionomer). The polymer electrolyte is not particularly limited, and conventionally known knowledge can be appropriately referred to. Since the polymer electrolyte plays a role of transmitting protons generated around the catalyst active material on the fuel electrode side, it is also called a proton conductive polymer. For example, an ion exchange resin constituting the electrolyte layer can be added to the catalyst layer as a polymer electrolyte.

  The polymer electrolyte is not particularly limited, and conventionally known knowledge can be appropriately referred to. Polymer electrolytes are roughly classified into fluorine-based polymer electrolytes and hydrocarbon-based polymer electrolytes depending on the type of ion exchange resin that is a constituent material. In addition, since the specific description of the fluorine-based polymer electrolyte and the hydrocarbon-based polymer electrolyte is the same as the description of the solid polymer electrolyte membrane below, the description is omitted here. Among these, the polymer electrolyte preferably contains a fluorine atom because it is excellent in heat resistance, chemical stability, and the like. Particularly preferred are fluorine-based electrolytes such as Nafion (registered trademark, manufactured by DuPont), Aciplex (registered trademark, manufactured by Asahi Kasei Co., Ltd.), Flemion (registered trademark, manufactured by Asahi Glass Co., Ltd.), and the like.

  The content of the polymer electrolyte weight in the catalyst layer is not particularly limited, but the ratio of the polymer electrolyte weight to the catalyst carrier in the electrode catalyst is preferably 0.3 to 1.2.

  The method for forming the catalyst layer is not particularly limited. Preferably, the catalyst layer is preferably formed by applying a catalyst ink comprising the electrode catalyst, the polymer electrolyte and the solvent as described above to the surface of the polymer electrolyte membrane or the gas diffusion layer. The method of forming is mentioned. In this case, the solvent is not particularly limited, and usual solvents used for forming the catalyst layer can be used in the same manner. Specifically, lower alcohols such as water, cyclohexanol, ethanol, 1-propanol and 2-propanol can be used. Further, the amount of the solvent used is not particularly limited, and a known amount can be used.

  The preparation method of the catalyst ink is not particularly limited as long as the electrode catalyst, the electrolyte and the solvent, and if necessary, the water-repellent polymer and / or the thickener are appropriately mixed. For example, a catalyst ink can be prepared by adding an electrolyte to a polar solvent, heating and stirring the mixed solution to dissolve the electrolyte in a polar solvent, and then adding an electrode catalyst thereto. Alternatively, after the electrolyte is once dispersed / suspended in a solvent, the dispersion / suspension may be mixed with an electrode catalyst to prepare a catalyst ink. In addition, a commercially available electrolyte solution (for example, DuPont Nafion solution: Nafion dispersed / suspended at a concentration of 5 wt% in 1-propanol) in which the electrolyte is previously prepared in the above other solvent is used as it is. May be used in the method.

  Each catalyst layer is formed by applying the catalyst ink as described above onto a coating substrate, for example, a polymer electrolyte membrane. At this time, the conditions for forming the catalyst layer on the coated substrate are not particularly limited, and can be used in the same manner as known methods or with appropriate modifications. For example, the catalyst ink is applied onto a coating substrate so that the thickness after drying becomes a desired thickness, and is 25 to 150 ° C., more preferably 60 to 120 ° C. in a vacuum dryer or under reduced pressure. For 5 to 30 minutes, more preferably 10 to 20 minutes. In addition, in the said process, when the thickness of a catalyst layer is not enough, the said application | coating and drying process is repeated until it becomes desired thickness. The thickness of the catalyst layer (after drying) is preferably 0.5 to 30 μm, more preferably 1 to 20 μm, and still more preferably 1 to 10 μm.

  The polymer electrolyte is responsible for proton transmission. Therefore, the proton conductivity of the polymer electrolyte is important. Here, when the EW of the polymer electrolyte is too large, the ionic conductivity in the entire catalyst layer is lowered. Therefore, it is preferable to include a polymer electrolyte having a small EW. Specifically, it preferably includes a polymer electrolyte having an EW of 1500 g / mol or less, more preferably a polymer electrolyte having 1200 g / mol or less, and particularly preferably a polymer electrolyte having 1100 g / mol or less.

  On the other hand, if the EW is too small, the hydrophilicity is too high, making it difficult to smoothly move water. From such a viewpoint, the EW of the polymer electrolyte is preferably 600 g / mol or more. Further, the catalyst layer may include two or more types of polymer electrolytes having different EWs.

  Note that EW (Equivalent Weight) represents the equivalent weight of the exchange group having proton conductivity. The equivalent weight is the dry weight of the ion exchange membrane per equivalent of ion exchange group, and is expressed in units of “g / mol”.

(separator)
The separator has a function of electrically connecting cells in series when a plurality of single cells of a fuel cell such as a polymer electrolyte fuel cell are connected in series to form a fuel cell stack. The separator also functions as a partition wall that separates the fuel gas, the oxidant gas, and the coolant from each other. In order to secure these flow paths, as described above, each of the separators is preferably provided with a gas flow path and a cooling flow path. As a material constituting the separator, conventionally known materials such as dense carbon graphite, carbon such as a carbon plate, and metal such as stainless steel can be appropriately employed without limitation. The thickness and size of the separator and the shape and size of each channel provided are not particularly limited, and can be appropriately determined in consideration of the desired output characteristics of the resulting fuel cell.

<Method for producing membrane electrode assembly>
Hereinafter, the manufacturing method of the membrane electrode assembly for fuel cells of embodiment by this invention is demonstrated.

  FIG. 3 is a flowchart showing a procedure of a method for manufacturing a membrane electrode assembly (MEA) for a fuel cell.

  First, a catalyst ink is produced (S11). The preparation of the catalyst ink is a known method and is not particularly limited.

  Subsequently, the MEA is assembled using the resulting catalyst ink. For this, a part of the catalyst ink produced in S11 is taken out, applied (transferred) to the electrolyte membrane (S12), the gas diffusion layer (GDL) is joined (S13), and the MEA is assembled (S14). Here, a cathode catalyst layer and a cathode gas diffusion layer are formed on one surface of the electrolyte membrane, and an anode catalyst layer and an anode gas diffusion layer are formed on the other surface to manufacture a membrane electrode assembly (MEA). .

  Subsequently, the assembled MEA is inspected using the inspection method according to the present invention (S15). Subsequently, sorting based on the inspection result is performed (S16). Details of this inspection and selection will be described later.

  Thereafter, the MEA that has become non-defective (OK) by sorting is passed to the subsequent process (S17). As a result, it becomes possible to assemble and manufacture the fuel cell stack using the MEA that has become non-defective. On the other hand, if it becomes defective (NG) by sorting, the desired performance cannot be obtained even if the MEA is used, so the MEA or the catalyst ink that made it is discarded (Pt and other raw material) Part (or all) is collected and reused) (S18).

  In this procedure, after the MEA is selected as a non-defective product, the fuel cell stack is assembled using the MEA as it is. However, it may be a sampling inspection instead of a 100% inspection after manufacturing the MEA.

<Inspection method of membrane electrode assembly (MEA)>
The method for inspecting the membrane electrode assembly (MEA) shows how the activity of the catalyst layer in the above-mentioned membrane electrode assembly (MEA) changes depending on the difference in humidity environment conditions, in particular, sufficient characteristics at low humidity. This is to check whether it can be demonstrated. In this embodiment, there are two inspection methods.

[First inspection method]
The first inspection method is performed by the following steps.

(Process (1): Measure the catalytic metal oxide reduction electricity)
A single cell for testing is manufactured using the MEA on which the catalyst layer is formed. The single cell is a general one and is not particularly limited as long as the catalytic metal oxide reducing electricity can be measured.

The catalytic metal oxide reducing electricity is measured by cyclic voltammetry (CV). Specifically, hydrogen _(H 2) to the anode side of the cell, flushed with nitrogen _{(N 2)} on the cathode side. At this time, water vapor is introduced so that both the anode side and the cathode side are in the same humidified state. The humidity to be set will be described later. The cathode side is the working electrode, and the anode side is the reference electrode (SHE) and the counter electrode. Then, potential scanning is performed with a predetermined potential. During the measurement, the valve on the cathode side channel is closed.

  In the potential scan, the cell temperature is 80 ° C., and the potential of the working electrode (cathode) versus the reference electrode (anode) (hereinafter the same) is held at 0.85 V vs SHE for 1 sec. This holding time is necessary for oxide generation. If the surface area of the catalytic metal is very small, it can be held for 1 second or longer to sufficiently oxidize the catalytic metal and detect the reduction current density of the oxide. Set to a valid value.

  Then 0.85 Vvs. 0.4V vs. SHE. The potential is scanned up to SHE at 10 mV / sec, and the reduction electric quantity is estimated as the catalytic oxide reduction electric quantity on the basis of the maximum current density value obtained near the 0.4 V point. The amount of reduced electricity at this time is determined as the amount of electricity with respect to the area per Pt surface area (a specific example will be described in the examples described later). Here, 0.85 Vvs. SHE is an oxide generation potential, and the setting range is set to be equal to or higher than the potential at which the catalytic metal oxide is generated and equal to or lower than the upper limit potential of the potential range actually used. This oxide formation potential is meaningless unless it is a potential at which a catalytic metal oxide is formed. For this reason, a minimum is made more than the electric potential required in order to produce | generate an oxide in the catalyst metal currently used. On the other hand, the upper limit is not particularly limited as long as the potential is not high enough to cause breakdown due to a high potential. For this reason, for example, if the upper limit is set to the highest potential in the potential range when actually used as a fuel cell, it is sufficient for measurement of the catalytic metal oxide reducing electricity.

  In this first inspection method, such measurement of catalytic oxide reducing electricity is carried out at at least two different relative humidities. As the humidity to be selected, at least one point from the high humidity side and at least one point from the low humidity side are arbitrarily selected. For example, a region where the relative humidity is 50% or more is defined as high humidity, and a region where the relative humidity is less than 50% is defined as low humidity. Then, one point is selected from the high humidity side where the relative humidity is 50% or more, and one point is selected from the low humidity side where the relative humidity is less than 50%. Thereby, it is possible to estimate the index contributing to the activity of the catalyst layer with respect to the humidity environment more accurately. More preferably, the high humidity side may be set to 70 to 90%, and the low humidity side may be set to a minimum humidity of about 40% or less in actual use. This is because by setting the high humidity side to a higher relative humidity, the catalyst metal is spread with water serving as a proton conductor to achieve a high catalyst usage rate. On the other hand, the low-humidity side is for viewing the activity in the actual use state or at any low humidity.

  Here, the relationship between the humidity and the catalyst will be described.

  The relationship between humidity and catalyst is described in, for example, the literature, Tetsuya Mashio, Kazuyuki Sato, Atsushi Ohma, “Analysis of Water Adsorption and Condensation in Catalyst Layers for Polymer Electrolyte Fuel Cells”, Electrochimica Acta, 140 (2014), 238-249. Have been described.

  In short, the relationship between humidity and the catalyst appears as a difference in catalyst activity due to the amount of “water” that becomes the proton conductor in the primary pores formed in the catalyst carrier. It is.

  According to this document, as a typical catalyst layer (CL), Pt / KB (Ketjen Black), relative ratio of an ionomer to carbon weight ratio (I (ionomer) / C (carbon)) of 0.9 The humidity and catalytic activity are described as examples. Here, the ionomer (electrolyte) is Nafion (registered trademark).

  FIG. 4 is a schematic diagram for explaining the relationship between the water activity (relative humidity) and the existence state of water adsorbed in the pores of the catalyst layer, which is shown by Tetsuya Mashio et al.

  As already described (FIG. 2), the catalyst metal 22 Pt is supported on the surface of the carrier 23 of the catalyst 20 and also in the primary pores 24. On the other hand, the surface of the carrier 23 is covered with an ionomer, but no ionomer enters the primary hole 24.

  FIG. 4A shows a low humidified state. At a low steam activity up to a water activity of 0.4 (corresponding to a relative humidity of 40%), the adsorption of water 30 to Pt (catalyst metal 22, hereinafter the same) occurs on the surface of the catalyst carrier 23. However, only a very small amount of water 30 enters the primary air holes 24 and is hardly adsorbed. At such low humidity, the sulfonic acid group of the ionomer 26 becomes a major source for water adsorption. This is desirable from the viewpoint of proton conduction. For this reason, even when the water activity is low, Pt itself adsorbs the water 30.

  Next, FIG.4 (b) has shown the state of water activity 0.4-0.9 (equivalent to 90% of relative humidity). As water activity increases from 0.4 to 0.9, capillary condensation is activated and begins to increase exponentially. In the water activity from 0.4 to 0.9, the gas phase still remains in a part of the secondary holes 32, but the hydrophilic part of the primary holes 24 is filled with water 30. That is, the secondary holes 32 are not filled with water in which water vapor is condensed at 0.4 to 0.9, but the primary holes 24 are filled with water. For this reason, a proton path from the ionomer 26 to the inside of the primary hole 24 is formed. Thus, the optimal state for electrochemical conversion is achieved with a water activity of 0.9. Aggregated water in the primary cavities 24 facilitates the supply of reactant gases and protons from the ionomer 26. In addition, the secondary void | hole 32 means the void | hole between the support | carriers in a catalyst layer.

  Next, FIG.4 (c) has shown the state from water activity 0.9 to 1 (equivalent to 100% of relative humidity). In this state, the condensed water 30 also fills the secondary holes 32.

  FIG. 5 is a graph showing the amount of water adsorbed on the carrier (carbon) with respect to the water activity (relative humidity) by the catalyst layer of the experimental example shown by Tetsuya Mashio et al. The vertical axis represents the amount of water adsorbed per gram of carrier <carbon>, and the horizontal axis represents the water activity (corresponding to the relative humidity). In the graph, the “model” is a mathematical model assuming a catalyst layer in which water adsorption occurs in the above-mentioned literature.

  From the graph of FIG. 5, in both the experimental example and the mathematical model, the water adsorption amount in the catalyst layer increases as the water activity increases. Further, the condensed water having fine pores (including primary vacancies) increases from around 0.4 in the water activity. On the other hand, the water adsorption amount of the ionomer, the water adsorption amount of the carrier, and the water adsorption amount to Pt all increase slightly as the water activity increases. Also from this, as shown in the schematic diagrams shown in FIGS. 4A to 4C, the primary catalyst carrier constituting the catalyst layer from when the water activity exceeds 0.4 (that is, relative humidity 40%). It can be seen that water also enters the pores, and a proton path is formed from the ionomer present on the surface of the support to the catalytic metal inside the primary pores.

  Thus, referring to the above-mentioned literature, it is considered that a certain degree of catalytic activity is usually obtained when the relative humidity is 40% or more. Therefore, in the present embodiment, as already described, a relative humidity of 50% is set as a high humidity side with a margin, and a sufficient catalytic activity is obtained, and at least one point is selected from this region. is there.

(Step (2): Obtain activity of catalytic metal against humidity)
Next, the activity of the catalytic metal with respect to humidity is estimated using the value of the catalytic oxide reducing electricity quantity obtained from two different relative humidities in the step (1). For this purpose, the value of the catalytic oxide reducing electricity quantity at the low relative humidity is normalized with the value of the catalytic oxide reducing electricity quantity at the maximum relative humidity as a reference. Specifically, (value of catalytic oxide reducing electricity at low humidity) / (value of catalytic oxide reducing electricity at maximum relative humidity) may be calculated. The value thus calculated is referred to as a standard value. If the measured relative humidity is two points of low humidity and high humidity, the maximum relative humidity is the value of the catalytic oxide reducing electricity quantity at one point on the high humidity side.

  The standard value obtained is an index representing the activity at the humidity. For example, a catalyst layer having a low standard value at low humidity indicates that the utilization rate of the catalyst metal is low when the moisture is low. On the other hand, a catalyst layer having a high standard value at low humidity indicates that the utilization factor of the catalyst metal is high even if the moisture is low. Therefore, it can be said that the higher the standard value at lower humidity, the better the catalyst layer.

  More specific inspection methods and standard values obtained thereby will be described in the following examples.

[Second inspection method]
Next, the second verification method will be described. The second verification method is performed by the following steps.

(Process (1): Measure the catalytic metal oxide reduction electricity)
Similarly to the first verification method, a single cell is manufactured using the MEA having the catalyst layer formed, and the catalytic metal oxide reducing electricity quantity is measured so that an arbitrary temperature and humidity can be applied. Since the measurement method may be the same as the first inspection method, description thereof is omitted.

  In the second verification method, the catalytic oxide reducing electricity quantity is measured at at least three different relative humidities. As the humidity to be selected, for example, a region where the relative humidity is 50% or more is set as high humidity, and a relative humidity of less than 50% is set as low humidity. Then, at least two points from the high humidity side and at least one point from the low humidity side are arbitrarily selected. Here again, as in the first inspection method, the high humidity side is usually set as a region where catalytic activity can be obtained.

(Step (2): Obtain activity of catalytic metal against humidity)
Next, the activity of the catalytic metal with respect to humidity is obtained using the value of the catalytic oxide reducing electricity obtained at three different relative humidity.

  For this purpose, first, the value of the catalytic oxide reducing electric quantity at all the relative humidity is normalized based on the value of the catalytic oxide reducing electric quantity at the maximum relative humidity (a standard value is obtained).

  Then, a linear expression obtained from two standard values on the high humidity side is created. In the linear expression (y = ax + b), y is a standard value and x is humidity.

  The standard value of the low humidity point is calculated (referred to as a calculated standard value) using this linear expression. Then, the difference between this calculated standard value and the standard value obtained from the measured amount of catalytic oxide reducing electricity at the low humidity point is taken. The difference value obtained is the amount of decrease in the catalyst activity. This is expressed as a percentage, and the catalyst layer utilization can be obtained by setting it as 100-decrease amount (percentage value). Then, the quality of the catalyst layer under low humidification is judged using the catalyst layer utilization rate as an index. The amount of decrease may be used as a direct index instead of the catalyst layer utilization rate.

  The primary expression can also be obtained from a graph (see Example 2 described later). For this, a graph using humidity on the horizontal axis and standard values on the vertical axis is used (see FIG. 10). Then, a line passing through the standard value of the catalytic oxide reducing electricity quantity at two points on the high humidity side (that is, the line of the primary expression) is drawn. An extrapolated line is obtained by extending this line. The value of the graph at the low humidity point on this extrapolation line is the standard value of the low humidity point calculated from the linear expression. After that, the difference between the standard value obtained from the graph in the same manner as described above and the standard value of the low humidity point obtained by the measurement is taken, and the activity of the catalyst layer at low humidification is determined as good.

  Here, the reason why the quality is determined using a straight line (primary expression or extrapolation line) from the value of the catalytic oxide reducing electricity quantity at two points on the high humidity side will be described.

  FIG. 6 is a graph showing the amount of water adsorbed on the carrier (carbon) with respect to the water activity (relative humidity) by the catalyst layers having different I / C, as shown by the above-mentioned Tetsuya Mashio et al. The vertical axis represents the amount of water adsorbed per 1 g of the carrier, and the horizontal axis represents the water activity (corresponding to the relative humidity). In the graph, the “model” is a mathematical model assuming a catalyst layer in which water adsorption occurs in the above-mentioned literature. I / C is the mass ratio of ionomer to carbon support.

  From the graph of FIG. 6, even if I / C is different, the water adsorption amount on the catalyst layer is almost equal when the water activity is 0.4-0.9 (that is, the relative humidity is 40% to 90%). It turns out that it is a linear behavior. In addition, since the typical oxide formation reaction is the following equation (1), a theoretical decrease in the amount of oxide formation due to a decrease in water activity is approximated by a straight line (primary equation) with respect to relative humidity. Is.

H ₂ O + Pt → Pt—OH + H ⁺ + e ⁻ (1)
From this, the calculated value (extrapolated value) from the primary equation at low humidity obtained by a straight line passing through two points on the high humidity side (primary equation) is the water activity lost due to the decrease in humidity. It shows that the activity decreased by the amount. For this reason, it can be said that the calculated value (extrapolated value) at low humidity represents the catalyst utilization rate close to the reduction behavior of the oxide generation amount due to the decrease in water activity at the humidity. The difference between the calculated value (extrapolated value) and the actually measured value indicates how low the actual activity of the catalyst layer is from the theoretical decrease in activity at low humidity. Therefore, the smaller the difference value, the higher the activity even in a low humidity state. Thereby, the utilization rate of the catalyst layer contributing to the activity of the catalyst layer with respect to the humidity environment can be verified with higher accuracy.

  Specific extrapolation lines and standard values will be described with reference to examples described later.

<Selection of inspection method>
The proper use of the above-described two inspection methods will be described.

  In the case of a catalyst layer that is highly active even at low humidity, the first verification method may estimate that the utilization rate of the catalyst layer is decreased despite high activity at low humidity. This is because, in the first verification method, the value of oxide reduction electricity at low humidity is simply divided by the value of oxide reduction electricity at high humidity. It is because it will show.

  On the other hand, in the second verification method, the utilization rate of the catalyst layer at the time of low humidification is estimated from a value obtained by subtracting the amount of activity that has decreased due to the small amount of water serving as a proton conductor from the fine pore structure of the catalyst. It will be. For this reason, it is possible to estimate the catalyst layer utilization rate with higher accuracy.

  For this reason, if a catalyst having high activity even at low humidity is obtained, it is preferable to verify by the second verification method. On the other hand, in the first verification method, it is only necessary to measure at least two oxide reduction electric quantities and divide those values, and therefore, there is less measurement and calculation processing than the second verification method. Thus, the utilization rate of the catalyst layer can be easily obtained.

  The effect of this invention is demonstrated using a following example.

  The effects of the present invention will be described using the following examples and comparative examples. However, the technical scope of the present invention is not limited only to the following examples.

(Catalyst synthesis)
Carrier A was prepared by the method described in WO2009 / 75264. The carrier A thus obtained was measured for average particle diameter (diameter), micropore and mesopore void volume, micropore and mesopore mode diameter, and BET specific surface area. As a result, the average particle diameter (diameter) is 100 nm; the pore volume of micropores is 1.06 cc / g; the pore volume of mesopores is 0.92 cc / g; the mode diameter of micropores is 0.65 nm; Mode diameter was 1.2 nm; and the BET specific surface area was 1753 m ² / g.

  Using the carrier A prepared as described above, platinum (Pt) having an average particle diameter of 3.5 nm was supported on the carrier metal so that the supporting rate was 50% by weight, and catalyst powder A was obtained. Specifically, 46 g of carrier A was immersed in 1000 g (platinum content: 46 g) of a dinitrodiammine platinum nitric acid solution having a platinum concentration of 4.6% by weight, and 100 ml of 100% ethanol was added as a reducing agent. This solution was stirred and mixed at the boiling point for 7 hours, and platinum was supported on the carrier A. The catalyst powder having a loading rate of 30% by weight was obtained by filtration and drying. Thereafter, in a hydrogen atmosphere, the temperature was maintained at 900 ° C. for 1 hour to obtain catalyst powder A having an average particle diameter (diameter) of 100 nm.

[Production of MEA]
(Preparation of cathode catalyst ink)
The produced catalyst powder A and an ionomer dispersion (Nafion (registered trademark) D2020, EW = 1100 g / mol, manufactured by DuPont) as a polymer electrolyte are mixed so that the weight ratio of the carrier A and the ionomer is 0.9. (Mixture 1). Separately, a mixed solvent 1 having a mixing weight ratio of water and n-propyl alcohol (NPA) of 60/40 was prepared. The mixed solvent 1 was added to the mixture 1 so that the solid content (Pt + carbon carrier + ionomer) was 7% by weight to prepare a cathode catalyst ink.

  Here, the catalyst ink was prepared in advance using a catalyst of a dry performance unimproved product having low activity even at low humidity and a dry performance improved product having high activity even at low humidity.

  Each adjustment method of the Dry performance unimproved product and the Dry performance improved product will be described.

  About the product whose Dry performance is not improved, the catalyst powder A is used. On the other hand, for the Dry performance improved product, an ionomer having an EW value of 650 g / mol was used, and the weight ratio of the carrier A to the ionomer was set to 1.3.

  Thereby, the EW value of the electrolyte used for the catalyst layer is about 0.65 times lower in the Dry performance improved product than in the Dry performance unimproved product. Further, the weight of the electrolyte used in the catalyst layer is about 1.4 times higher in the dry performance improved product than in the dry performance unimproved product.

(Preparation of anode catalyst ink)
Ketjen black (particle size: 30 to 60 nm) was used as a carrier, and platinum (Pt) having an average particle size of 2.5 nm as a catalyst metal was supported on the catalyst so as to have a loading ratio of 50% by weight. Got. This catalyst powder was mixed with an ionomer dispersion (Nafion (registered trademark) D2020, EW = 1100 g / mol, manufactured by DuPont) as a polymer electrolyte so that the weight ratio of the carbon support to the ionomer was 0.9 ( Mixture 2). Separately, a mixed solvent 2 having a mixing weight ratio of water and n-propyl alcohol of 50/50 was prepared. The mixed solvent 2 was added to the mixture 2 so that the solid content (Pt + carbon carrier + ionomer) was 7% by weight to prepare an anode catalyst ink.

(Assembly of MEA)
Next, MEAs having these catalysts as catalyst layers were prepared using the Dry performance-improved product and the Dry performance-improved product prepared as described above.

  For this, a gasket (manufactured by Teijin Dupont, Teonex, film thickness: 25 μm (adhesive layer: 10 μm)) was disposed around both sides of a polymer electrolyte membrane (manufactured by Dupont, NAFION NR211, film thickness: 25 μm). Next, the cathode catalyst ink was applied to the exposed portion on one side of the polymer electrolyte membrane to a size of 5 cm × 2 cm by spray coating. The catalyst ink was dried by maintaining the stage for spray coating at 60 ° C. for 1 minute to obtain a cathode catalyst layer. The amount of platinum supported at this time is 0.15 mg / cm 2. Next, similarly to the cathode catalyst layer, an anode catalyst layer was formed by spray coating and heat treatment on the electrolyte membrane.

  Both surfaces of the obtained laminate were sandwiched between gas diffusion layers (24BC, manufactured by SGL Carbon Co.) to obtain a membrane electrode assembly (MEA).

  Here, the MEA using the Dry performance unimproved product is MEA (1), and the MEA using the Dry performance improved product is MEA (2).

<Measurement of oxide reduction electricity quantity>
A plurality of test single cells each using the manufactured MEA (1) and MEA (2) were manufactured. The structure of the single cell is as follows.

  A single cell is another one in which separators having gas flow paths are arranged and sandwiched on both sides of the MEA, a current collector plate is arranged on the outside (both sides) of the separator, and an end plate is arranged on the outside. And each gas piping is installed in this cell for each measurement mentioned later. The gas piping can be adjusted for pressure, temperature and humidity. Further, the current collector plate is wired from the potentiostat so as to be on the anode side and the cathode side.

  Measurement of the oxide reduction electric quantity was performed by cyclic voltammetry (CV). For the measurement, constant temperature and humidity injection was used, the cell temperature was kept at 80 ° C., the relative humidity was 90%, the relative humidity was 70%, the relative humidity was 70%, and the relative humidity was 40%.

  The anode side was a hydrogen atmosphere, the cathode side was a nitrogen atmosphere, and the anode side was a reference electrode (SHE) (the same applies to other measurements). CV was held at 0.85 Vvs SHE for 1 sec, then 0.85 Vvs. 0.4V vs. SHE. The potential was scanned up to SHE at 10 mV / sec, and the amount of reduced electricity was estimated as the amount of reduced catalyst oxide electricity based on the maximum current density value obtained near the 0.4 V point.

  FIG. 7 is a graph showing an example of the measurement result of the oxide reducing electricity quantity.

The oxide reduction electricity quantity q _ox is obtained by the following equation (2), with the electricity quantity obtained as shown in FIG. 7 as the quantity of electricity with respect to the area per Pt surface area.
q _ox [C / cm ^{2 (Pt)} ] = Q [C / cm ^{2 (geo.)} ] / S [cm ^{2 (Pt)} / cm ^{2 (geo.)} ] (2)
In the formula, S [cm ^{2 (Pt)} / cm ^{2 (geo.)} ] Is a Pt actual effective surface area [cm ^{2 (Pt)} ] estimated by ECA (hereinafter simply referred to as ECA) obtained from the hydrogen adsorption / desorption electricity quantity, It is the ratio of the active area (geometric area) [cm ^{2 (geo.)} ] To which the catalyst layer is applied.

  ECA may be a general technique and is not particularly limited, but here, measurement was performed as follows.

This ECA measures the amount of charge transfer associated with the hydrogen adsorption and desorption phenomenon on Pt atoms, and the measured value, the amount of Pt supported, the active area, and the amount of hydrogen adsorption per area on Pt are 210 μC / It is the Pt surface area per mass estimated when it is assumed to be cm ² .

  Specifically, the cell temperature is 80 ° C., hydrogen is supplied to the anode side, and nitrogen is supplied to the cathode side. The pressure at the cell outlet was atmospheric pressure. The gas humidity on both the anode side and the cathode side was measured at 90%, 70% and 40% relative humidity, respectively. The measurement start potential is 0.5 Vvs SHE, and then the potential is 0.02 Vvs. 0.9Vvs. The potential scan was performed three times at 10 mV / sec until SHE, and a current-potential curve of the three potential scans was obtained. From this current-potential curve, the hydrogen generation current and the electric double layer capacitance current are separated from the current value when the potential is scanned from 0.5 V vs SHE to 0.02 V vs SHE, and the hydrogen adsorption current on the Pt surface is taken out. The amount of hydrogen adsorption current on the Pt surface was obtained.

ECA is the Pt surface area per Pt mass as described above (unit: m ² / g). Multiplying this by the value of the amount of Pt supported per active area (geometric area) of the target catalyst layer (the amount charged, the unit is mg / cm ² ), it is possible to obtain “S [cm ^{2 (Pt)} / cm ^{2 (geo.)} ].

Then, using the measurement results of oxide reduction electricity quantity q _ox of the humidity, the catalyst oxide at each relative humidity 90% RH oxide reduction electricity quantity q _ox a _{(90% RH)} as the reference reducing electric The standard value of the quantity was obtained.

Table 1 shows the measurement results of the catalyst oxide reduction electricity quantity q _ox by humidity for each of the MEA (1) using the Dry performance unimproved product and the MEA (2) using the Dry performance improved product. Indicates the value. Table 1 also shows the standard value of the catalytic oxide reducing electric quantity at each relative humidity based on the oxide reducing electric quantity q _{ox (90% RH)} at a relative humidity of _90% .

  Next, the ORR activity was measured for comparison. The measuring method of ORR activity is as follows.

  (1) Hydrogen is supplied to the anode and nitrogen is supplied to the cathode and left for 15 minutes at a natural potential to remove oxide on the Pt surface of the cathode electrode.

  (2) The cathode gas flow is bypassed the cell, and the inside of the cathode gas apparatus is replaced with an oxygen-containing gas while the inside of the cell is held in nitrogen gas.

  (3) The cathode potential is set to 0.4 V with respect to the anode potential.

  (4) The cathode gas is caused to flow from the bypass to the cell, and an oxygen-containing gas is introduced into the cathode electrode.

  (5) The cathode potential is set to 0.85 V with respect to the anode potential and held for 1 second.

  (6) The current response when the cathode potential is scanned from the anode potential to 0.85 V to 0.4 V at a speed of 10 mV / sec is measured.

  By measuring according to such a procedure, the oxide on the Pt surface of the cathode electrode is removed in (1), and then the cathode potential is controlled to 0.4 V or less with respect to the anode potential until (4). Thus, it is possible to control so that an oxide is not generated on the Pt surface of the cathode electrode as much as possible. In (5), the amount of Pt oxide generated is controlled by the potential and the potential holding time. In (6), the ORR is measured with the amount of Pt oxide generated on the Pt surface of the cathode electrode controlled. can do.

  The conditions in this example are: the cell temperature is 80 ° C., the gas humidity is 90% relative humidity on both the anode side and the cathode side, and the flowing gas is anode gas: 10% hydrogen (dilution gas is nitrogen) 2 NL / min Cathode gas: 10% oxygen (diluting gas is nitrogen) 4 NL / min, gas pressure was the outlet atmospheric pressure on both the anode side and the cathode side.

  Note that the current value measured in the ORR measurement includes current components other than the pure ORR-derived current. For example, hydrogen cross leak from the anode side to the cathode side through the electrolyte membrane, internal current due to internal short circuit, adsorption / desorption phenomenon on the electrode surface, current due to charge / discharge of electric double layer capacity, etc. may be included . Thus, the following operation was performed in order to extract a current derived from pure ORR. First, separately from the ORR measurement, after performing the above (1), the current when a potential history similar to (3) to (6) was given with hydrogen flowing on the anode side and nitrogen flowing on the cathode side Measure the response. Thereafter, the current measured in this way is subtracted from the ORR measurement current. Thus, a pure ORR current can be obtained. Here, the current value at the 0.84 V point at the time of potential scanning from 0.85 V to 0.4 V is defined as the ORR activity, and the ORR activity is the actual effective surface area of Pt obtained by the ECA described above. The current density excluding.

<Example 1>
In Example 1, according to the first inspection method in the embodiment described above, the above-prepared MEA (1) and MEA (2) were inspected.

Here, according to the standard value of the catalytic oxide reducing electricity quantity q _{ox (40% RH)} at _40% relative humidity based on the oxide reducing electricity quantity q _{ox (90% RH)} at 90% relative humidity in Table 1. The quality of the MEA was judged.

FIG. 8 shows the respective amounts of catalytic oxide reducing electricity q _{ox (90% RH)} and q _{ox at} 90 and 40% relative humidity based on the oxide reducing electricity quantity q _{ox (90% RH)} at _90% relative humidity. It is the graph which showed the relationship between the standard value of _{(40% RH)} , and relative humidity.

  From FIG. 8, it can be seen that when the humidity is low, the standard value of MEA (1) (Dry performance unimproved product) is lower than MEA (2) (Dry performance improved product).

  Here, a standard value 1.0 is defined as a catalyst layer utilization rate of 100%, and a value expressing each standard value as a percentage is defined as a catalyst layer utilization rate. If it does so, MEA (1) (Dry performance unimproved product) will be 41%, and MEA (2) (Dry performance improved product) will be 76%. That is, in the catalyst layer utilization rate in the low humidity state, MEA (1) (Dry performance unimproved product) was reduced by 59%, and MEA (2) (Dry performance improved product) was reduced by 24%.

  The predetermined value of the catalyst layer utilization rate used as the criterion for determining the MEA quality is, for example, 50%. That is, the MEA having a catalyst layer utilization rate of 50% or more in the MEA inspection is passed as a non-defective product to the next step and used for stack assembly. On the other hand, if the utilization rate of the catalyst layer is less than 50%, NG is not allowed to flow to the subsequent assembly process at that time. For example, if the utilization rate of the catalyst layer is less than 50%, there is a possibility that sufficient performance as a fuel cell cannot be exhibited even if the stack is assembled thereafter. For this reason, the MEA having a catalyst layer utilization rate of less than 50% in this MEA inspection may be prevented from flowing to the subsequent assembly process as NG at that time.

  When such a catalyst layer utilization rate of 50% or more is a non-defective product, the MEA (2), that is, the Dry performance improved product is a non-defective product. As a result, only MEAs that are expected to exhibit the desired performance can be advanced to the next fuel cell assembly process. Therefore, the production yield in the fuel cell assembly process can be improved.

  Furthermore, the criteria for determining whether or not the MEA is good may be made stricter, and for example, a catalyst layer utilization rate of 90% or more may be determined as a good product. Thus, by making the judgment criteria stricter, the yield in the fuel cell assembly process can be further improved.

  Here, in order to select the membrane electrode assembly, the utilization rate of the catalyst layer showing the standard value as a percentage is used. However, the standard value may be selected without being expressed as a percentage.

  FIG. 9 is a graph showing the relationship between the catalyst layer utilization rate determined from the amount of reduced oxide electricity in Example 1 and the activity decrease rate determined from the ORR activity described above. The vertical axis represents the rate of decrease in ORR activity. This is to obtain a ratio of 40% relative humidity (ORR value of 40% relative humidity / ORR value of 90% relative humidity) for ORR activity obtained by ECA for each humidity, based on 90% relative humidity. Decrease rate = (1− (ORR value of 40% relative humidity / ORR value of 90% relative humidity)) × 100. The horizontal axis represents the catalyst layer utilization rate described above.

  This graph also shows the utilization rate of the catalyst layer obtained from ECA for comparison. The catalyst layer utilization rate obtained from ECA is simply obtained by obtaining the ratio of the ECA value when the relative humidity is 40%, assuming that the utilization rate is 100%. That is, the catalyst layer utilization obtained from ECA = (ECA value when relative humidity is 40%) / (ECA value when relative humidity is 90%) × 100.

  The activity decrease rate obtained from the ORR activity and the catalyst layer utilization rate are directly proportional. In FIG. 9, this direct proportional relationship is indicated by a dotted line.

  As shown in FIG. 9, the utilization rate of the catalyst layer determined from the amount of reduced oxide electricity is along a directly proportional dotted line. That is, MEA (1) (Dry performance unimproved product) having low ORR activity at low humidity has a low catalyst layer utilization rate obtained from the amount of oxide reduction, and MEA (2) (Dry performance) having high ORR activity. The improved product) also has a high utilization rate of the catalyst layer determined from the amount of reduced oxide electricity.

  On the other hand, as for the catalyst layer utilization rate by ECA, MEA (2) (Dry performance improved product) with high ORR activity is near the dotted line, but MEA (1) (Dry performance unimproved product) has almost no catalyst layer utilization rate. The result was not reduced. In other words, this result shows that the catalyst layer utilization rate by ECA may be estimated higher than the actual activity (ORR activity) as an index indicating the catalyst activity when the humidity is low. .

  Therefore, it can be seen that the utilization rate of the catalyst layer determined from the amount of oxide reduction electricity according to this example is a more accurate index than ECA in order to see the activity at low humidity.

  The reason why the utilization rate of the catalyst layer determined from the amount of reduced oxide electricity as compared with ECA as described above is a result showing the actual state of catalytic activity.

  The difference between the ECA value reflecting the amount of hydrogen adsorbed on Pt and the utilization rate of the catalyst layer due to oxide adsorption arises from the following matters.

  First, proton sources include sulfonic acid groups other than water and functional groups (—OH). On the other hand, the oxide supplier hardly thinks of substances other than water, and reflects the amount of water around the catalyst well.

  Therefore, when the amount of water (relative humidity) is the same, the ECA value reflecting the amount of hydrogen adsorbed by a reactant supply source other than water is higher than the utilization rate of the catalyst layer determined from the amount of oxide reduction electricity. Conceivable.

<Example 2>
In Example 2, the produced MEA (1) and MEA (2) were inspected according to the second inspection method in the above-described embodiment.

Here, the catalytic oxide reducing electricity quantity q _{ox (90 in} each of relative humidity 90% and relative humidity 70% in Table 1 with reference to the oxide reducing electricity quantity q _{ox (90% RH)} of 90% relative humidity. _{% RH)} and q _{ox (70% RH)} standard values were obtained.

A standard value of 40% relative humidity was calculated using this linear equation. And the difference with the standard value of the measured amount of catalyst oxide reduction q _{ox (40% RH)} at a relative humidity of 40% was regarded as a decrease in the catalyst activity. Then, the MEA quality was determined by using the value obtained by subtracting the amount of decrease from 100% as the catalyst layer utilization.

Figure 10 is a relative humidity of 90% of the oxide reduction electricity quantity _{q ox (90% RH)} relative humidity 90,70% relative to the, and 40% of the respective catalyst oxide reduction electricity quantity _{q ox (90% RH )} , Q _{ox (70% RH)} , and q _{ox (40% RH)} standard values, and a graph showing the relationship between relative humidity.

In this graph, the standard values of the catalytic oxide reducing electricity quantity at 90% relative humidity and 70% relative humidity based on the oxide reducing electricity quantity q _{ox (90% RH)} at 90% relative humidity are used. The line of the primary expression (extrapolated line) is drawn. The extrapolated line of MEA (1) (Dry performance unimproved product) is a one-dot chain line, and the extrapolated line of MEA (2) (Dry performance improved product) is a dotted line.

  From the figure, the decrease rate at 40% relative humidity is 56% for MEA (1) (Dry performance unimproved product) and 7.8% for MEA (2) (Dry performance improved product). Therefore, the utilization rate of the catalyst layer at 40% relative humidity is 1-56% = 44% for MEA (1) (Dry performance unimproved product), and 1-7.8% for MEA (2) (Dry performance improved product). = 92.2%.

  The predetermined value of the catalyst layer utilization rate used as a criterion for the MEA pass / fail judgment is, for example, 90%. That is, the MEA having a catalyst layer utilization rate of 90% or higher in the MEA inspection is regarded as a non-defective product and is used for assembly of the stack by passing it to the next process. On the other hand, if the utilization rate of the catalyst layer is less than 90%, the NG is not allowed to flow to the subsequent assembly process at that time. As a result, it is possible to prevent the production of a fuel cell with low performance itself and to improve the yield in the fuel cell assembly process. Further, the criteria for determining whether or not the MEA is good may be made stricter, and for example, a catalyst layer utilization rate of 95% or more may be determined as a non-defective product. By tightening the criteria, MEAs that can produce fuel cells with better performance can be selected and flowed to the stack assembly process or the like.

  In Example 2, after obtaining the amount of decrease in the standard value of the catalyst activity, the amount of reduction is subtracted from 100% to convert it to the catalyst layer utilization rate. This is a mathematical process for matching with the catalyst layer utilization in Example 1. Therefore, in the selection of the MEA using Example 2 (that is, the second inspection method), the MEA may be selected using the decrease amount of the standard value of the catalyst activity itself. In this case, the predetermined value to be selected is, of course, a value obtained by subtracting from 100 the predetermined value for selecting the catalyst layer utilization rate described above. If a predetermined value of 90% or more is selected as a non-defective product using the catalyst layer utilization rate, when a decrease amount is used, an MEA having a standard value decrease amount of less than 0.1 is determined to be a non-defective product. Of course, in the case of using a stricter criterion, an MEA in which the amount of decrease in the standard value is less than 0.05 is regarded as a non-defective product.

  FIG. 11 is a graph showing the relationship between the catalyst layer utilization rate determined from the oxide reduction electricity quantity in Example 2 and the activity decrease rate determined from the ORR activity. This graph also shows the utilization rate of the catalyst layer obtained from ECA for comparison. ECA is as described in Example 1.

  The activity decrease rate obtained from the ORR activity and the catalyst layer utilization rate are directly proportional. In FIG. 11, this direct proportional relationship is indicated by a dotted line.

  As shown in FIG. 11, the utilization rate of the catalyst layer determined from the amount of oxide reduction electricity of Example 2 is a value very close to a directly proportional dotted line. That is, it can be seen that the inspection method of Example 2 can be an index very close to the actual activity (ORR activity).

<Example 3>
Example 3 is an experimental example for properly using the first inspection method and the second inspection method.

  The proper use of the above-described two inspection methods will be described.

  In this Example 3, a catalyst layer having high activity even in a low humidity state is selected from a plurality of prepared catalysts in advance, and the first inspection method and the second The difference in the inspection method was verified. The catalyst layer with high low humidity activity has a high ORR activity determined from ECA when the relative humidity is 40% for a plurality of prepared catalysts.

FIG. 12 shows a catalyst having a relative humidity of 90, 70%, and 40% based on an oxide reducing electricity quantity q _{ox (90% RH)} of 90% relative humidity using a catalyst layer having a low humidity activity. It is the graph which showed the relationship between the standard value of oxide reduction electric quantity _{qox (90% RH)} , _{qox (70% RH)} , and _{qox (40% RH)} , and relative humidity. The vertical axis is the standard value, and the horizontal axis is the relative humidity.

  For the MEA using such a low humidity activity high catalyst layer, the standard value obtained from the catalytic metal oxide reducing electricity measured at a relative humidity of 40% by the first inspection method is 0.56. (Catalyst layer utilization rate 56%) (that is, a decrease of 44% from the standard value 1.0).

  On the other hand, in the second inspection method, a linear expression is obtained from the standard values at two points on the high humidity side. This linear expression is a dotted straight line shown in FIG.

  The calculated standard value when the relative humidity is 40% obtained from this linear expression (dotted line) is 0.49 (49%). Then, the standard value 0.56 obtained from the catalytic metal oxide reducing electricity measured at a relative humidity of 40% is 7% higher than this calculated standard value.

  From this result, the catalyst utilization rate of the catalyst layer with high low humidity activity is estimated to be low in the first inspection method. On the other hand, in the second inspection method, since it is higher than the line of the primary equation obtained from two points of high humidity, it is not estimated lower than the actual one. Therefore, in the catalyst layer having a high low humidity activity, the use of the second inspection method is an index closer to the activity performance of the catalyst layer.

  Therefore, in the catalyst layer having high low humidity activity, it is preferable to use the second inspection method because the catalyst utilization rate at low humidity can be expressed more accurately. On the other hand, the first inspection method may be used for simple inspection. For example, when it is known in advance that the catalyst layer has a high low humidity activity, it is preferable to use the second inspection method from the beginning. However, if it is unclear what kind of catalyst layer, for example, when the catalyst layer utilization rate is lower than the design target by the first inspection method, or when the defective rate is high, the second inspection method is used. You may make it change and test | inspect.

  The effects of the embodiments and examples using the present invention will be described below.

  (1) In the first inspection method, it is possible to obtain an index that contributes to the activity as a catalyst layer with respect to the humidity environment by measuring the catalytic metal oxide reducing electricity quantity while changing the humidity. And the quality of MEA is determined from the obtained index. Thereby, the active state of the catalyst layer due to the difference in humidity environment can be verified with higher accuracy than the index estimated by the ECA measurement method as in the past. Also, good products can be selected in a short time without conducting a power generation test.

  (2) selecting a relative humidity of at least one point from the high humidity side and at least one point from the low humidity side, as the humidity range when measuring the catalytic metal oxide reducing electricity quantity, by dividing the relative humidity by 50%; did. Thereby, the index of the catalyst layer utilization rate contributing to the activity as the catalyst layer against the humidity environment can be verified with higher accuracy.

  (3) In the second inspection method, after measuring the catalytic metal oxide reducing electricity quantity by changing the humidity, the standard value on the low humidity side is calculated from the standard value of the measurement result on the two points on the high humidity side. It was decided to obtain the difference between the values of the catalytic metal oxide reduction electricity amount at low humidity measured from the calculated value. For this reason, verification can be performed with higher accuracy than the index estimated by the first inspection method, and the determination accuracy of MEA quality can be improved.

  (4) Select a relative humidity of at least two points from the high humidity side and at least one point from the low humidity side as the humidity range when measuring the catalytic metal oxide reducing electricity quantity by dividing the relative humidity by 50%. did. Thereby, the index of the catalyst layer utilization rate contributing to the activity as the catalyst layer against the humidity environment can be verified with higher accuracy.

  (5) The range of the potential point held in the catalytic metal oxide reduction electric quantity measurement is set to be equal to or higher than the potential at which the catalytic metal oxide is generated and lower than the upper limit potential of the potential range actually used. . As a result, measurement under conditions closer to the actual vehicle operating conditions is possible, and the index of the catalyst layer utilization rate contributing to the activity as the catalyst layer can be verified more accurately.

  Although the embodiments and examples to which the present invention is applied have been described above, the present invention is not limited to the above-described embodiments and examples. The present invention can be implemented in various forms based on the technical idea described in the claims, and these are also within the scope of the present invention.

1 Polymer electrolyte fuel cell (PEFC),
2 solid polymer electrolyte membrane,
3 catalyst layer,
3a anode catalyst layer,
3c cathode catalyst layer,
4a Anode gas diffusion layer,
4c cathode gas diffusion layer,
5a anode separator,
5c cathode separator,
6a Anode gas flow path,
6c cathode gas flow path,
7 Refrigerant flow path,
10 Membrane electrode assembly (MEA),
20 catalyst,
22 catalytic metal,
23 carrier,
24 mesopores,
25 micropores,
26 Ionomer (solid proton conductor).

Claims (5)

A cathode electrode layer and a cathode gas diffusion layer and an anode catalyst layer and an anode gas diffusion layer are formed on both surfaces of the electrolyte membrane, respectively, to manufacture a membrane electrode assembly for a fuel cell,
The catalytic metal oxide reducing electricity quantity of the membrane electrode assembly is measured at an arbitrary relative humidity of at least two points,
The catalytic metal oxide reducing electricity quantity at the other relative humidity point is divided by the measured catalytic metal oxide reducing electricity quantity at the highest relative humidity point, thereby determining the catalytic metal oxide reducing electricity quantity at the highest relative humidity point. Determine the standard value of the catalytic metal oxide reducing electricity at each relative humidity point,
A method for producing a membrane electrode assembly, wherein a membrane electrode assembly in which the standard value is a predetermined value or more is selected as a non-defective product.   The point which becomes the highest relative humidity point among the at least two relative humidity points when measuring the catalytic metal oxide reducing electricity quantity is arbitrarily selected from a region having a relative humidity of 50% or more. 2. A method for producing a membrane electrode assembly according to 1. A cathode electrode layer and a cathode gas diffusion layer and an anode catalyst layer and an anode gas diffusion layer are formed on both surfaces of the electrolyte membrane, respectively, to manufacture a membrane electrode assembly for a fuel cell,
The catalytic metal oxide reducing electricity quantity of the membrane electrode assembly is measured at an arbitrary relative humidity of at least three points,
The catalytic metal oxide reducing electricity at each of the at least three measured relative humidity points is divided by the measured catalytic metal oxide reducing electricity at the highest relative humidity point to obtain the catalytic metal oxidation at the highest relative humidity point. Find the standard value of catalytic metal oxide reducing electricity at each relative humidity point based on the amount of reduced electricity
Select two points from the higher relative humidity among the plurality of relative humidity, and create a linear expression that passes the standard value of the selected two points,
Calculate a standard value at a relative humidity lower than the relative humidity of the two points at which the primary formula was created using the primary formula,
The standard value calculated from the primary equation is compared with the standard value obtained from the measured catalytic metal oxide reducing electric quantity at the same relative humidity as the relative humidity calculated from the primary equation. A method for producing a membrane electrode assembly, wherein the difference is regarded as a reduction amount of the catalytic activity at the relative humidity obtained by calculating the standard value from the linear equation, and a membrane electrode assembly in which the reduction amount is less than a predetermined value is selected as a non-defective product.   Of the at least three relative humidity points when measuring the catalytic metal oxide reducing electricity quantity, two points for creating the primary equation are arbitrarily selected from a region having a relative humidity of 50% or more. The manufacturing method of the membrane electrode assembly of Claim 3.   The setting range of the oxide generation potential point when measuring the catalytic metal oxide reducing electricity is set to be equal to or higher than the potential at which the catalytic metal oxide is generated and lower than the upper limit potential of the potential range actually used. Item 5. A method for producing a membrane electrode assembly according to any one of Items 1 to 4.