JP2018133271A - Gas diffusion layer for fuel cell, method for producing the same, and fuel cell - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a fuel cell superior in power generation performance.SOLUTION: A fuel cell includes: an electrolyte film; catalyst layers which are individually disposed on both principal surfaces of the electrolyte film; and a pair of gas diffusion layers each of which is disposed at an opposite side to the electrolyte film of the catalyst layer. At least one of the pair of gas diffusion layers has an average thickness of 150 μm or less, and an oxygen diffusion coefficient Dt in a thickness direction of each gas diffusion layer is 1.2×10cm/sec or more, 6×10cm/sec or less.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a gas diffusion layer designed using a simulation result of gas diffusion behavior in a fuel cell, a manufacturing method thereof, and a fuel cell.

  A fuel cell includes a membrane electrode assembly (MEA) composed of an electrolyte membrane and a pair of gas diffusion electrodes (anode and cathode), a gasket, a conductive anode separator in contact with the anode, And a conductive cathode separator in contact with the cathode. The gas diffusion electrode includes a catalyst layer disposed on the electrolyte membrane side and a gas diffusion layer disposed on the opposite side of the catalyst layer from the electrolyte membrane.

  Patent document 1 is disclosing the gas diffusion layer which formed the water-repellent layer in the electroconductive base material. Carbon fiber woven fabric or carbon paper is used for the conductive substrate. The water repellent layer is composed of a material such as carbon black and a fluororesin.

  On the other hand, Patent Documents 2 to 4 disclose gas diffusion layers that do not have a conductive substrate. For example, the gas diffusion layer of Patent Document 2 is mainly composed of conductive particles and a polymer resin, and contains a smaller amount of carbon fiber than the polymer resin.

Japanese Patent Laid-Open No. 2003-197202 International Publication No. 2010-050219 Pamphlet Japanese Patent Laid-Open No. 2003-187809 JP 2007-141783 A

  In order to increase the gas diffusibility of the gas diffusion layer, it is desirable that the gas diffusion layer be as porous as possible and maintain the required mechanical strength. Therefore, various techniques for controlling the porosity and pore size distribution of the gas diffusion layer are being developed. However, there is a limit to increasing the gas diffusibility of the gas diffusion layer by controlling the pores.

In view of the above, one aspect of the present invention is a gas diffusion layer for a fuel cell, wherein the average thickness of the gas diffusion layer is 150 μm or less, and an oxygen diffusion coefficient Dt in the thickness direction of the gas diffusion layer is 1.2 × 10 ⁻³ cm ² / sec or more and 6 × 10 ⁻² cm ² / sec or less.

  Another aspect of the present invention is a gas diffusion layer for a fuel cell, wherein an oxygen diffusion coefficient Ds in a plane direction perpendicular to the thickness direction is larger than an oxygen diffusion coefficient Dt in the thickness direction of the gas diffusion layer. Relates to the gas diffusion layer.

  Still another aspect of the present invention is a gas diffusion layer for a fuel cell, wherein the gas diffusion layer includes a conductive material and a polymer resin, and the conductive material includes a particulate material and a plate-like material. And an acute angle formed between the major axis of the plate material and the thickness direction of the gas diffusion layer is 60 ° or more and 90 ° or less.

  Still another aspect of the present invention is an electrolyte membrane, a catalyst layer disposed on each of the principal surfaces of the electrolyte membrane, and a pair of gas diffusion layers disposed on the opposite side of the catalyst layer from the electrolyte membrane. And at least one of the pair of gas diffusion layers is any one of the gas diffusion layers described above.

Still another aspect of the present invention includes (i) a step of preparing a mixture of a conductive material, a polymer resin, a surfactant, and a dispersion medium, and (ii) a step of forming the mixture into a sheet. (Iii) firing the sheet to obtain a fired sheet from which the surfactant and the dispersion medium have been removed, and (iv) an average thickness of the fired sheet is 150 μm or less, and the thickness A gas diffusion layer for a fuel cell, comprising: a step of rolling so that an oxygen diffusion coefficient Dt in a direction is 1.2 × 10 ⁻³ cm ² / sec or more and 6 × 10 ⁻² cm ² / sec or less It relates to the manufacturing method.

  According to the present invention, the power generation performance of a fuel cell can be improved.

It is a cross-sectional schematic diagram which shows the structure of the cell of the fuel battery | cell which concerns on embodiment of this invention. It is a cross-sectional schematic diagram which shows the structure of the gas diffusion layer containing the plate-shaped material which concerns on embodiment of this invention. It is a graph which shows the relationship between the VI characteristic calculated | required by simulation, and the thickness of a gas diffusion layer. It is a graph which shows the result of having calculated | required the oxygen concentration change in the interface of a catalyst layer and a gas diffusion layer (100 micrometers in thickness) along the direction orthogonal to the some parallel fluid flow path by simulation. It is a graph which shows the result of having calculated | required the oxygen concentration change in the interface of a catalyst layer and a gas diffusion layer (thickness 200 micrometers) along the direction orthogonal to the some parallel fluid flow path by simulation. It is a graph which shows the relationship between the VI characteristic calculated | required by simulation, and a diffusion coefficient ratio. It is another graph which shows the relationship between the VI characteristic calculated | required by simulation, and a diffusion coefficient ratio. FIG. 5 is a graph showing a relationship between a VI characteristic obtained by simulation and an actual measurement value for a fuel cell including a gas diffusion layer having a thickness of 150 μm and an oxygen diffusion coefficient Dt = 6.0 × 10 ⁻² cm ² / sec. is there. FIG. 5 is a graph showing a relationship between a VI characteristic obtained by simulation and an actual measurement value for a fuel cell including a gas diffusion layer having a thickness of 100 μm and an oxygen diffusion coefficient Dt = 6.0 × 10 ⁻² cm ² / sec. is there. FIG. 5 is a graph showing a relationship between a VI characteristic obtained by simulation and an actual measurement value for a fuel cell including a gas diffusion layer having a thickness of 80 μm and an oxygen diffusion coefficient Dt = 6.0 × 10 ⁻² cm ² / sec. is there. It is the figure which displayed the measured VI characteristic of FIGS. FIG. 5 is a graph showing the relationship between the VI characteristics obtained by simulation and the measured values for a fuel cell having a gas diffusion layer with a thickness of 100 μm and an oxygen diffusion coefficient Dt = 1.2 × 10 ⁻³ cm ² / sec. is there. It is a graph which shows the VI characteristic calculated | required by simulation about the fuel cell which comprises 50-100 micrometers in thickness and the diffusion coefficient of oxygen Dt = 1.2 * 10 < ^-3 > cm < ² > / sec.

  First, the structure of the fuel cell will be described. A fuel cell according to an embodiment of the present invention includes a membrane electrode assembly (hereinafter referred to as an electrolyte membrane), a cathode in contact with one main surface of the electrolyte membrane, and an anode in contact with the other main surface of the electrolyte membrane. MEA), a conductive cathode separator in contact with the cathode, and a conductive anode separator in contact with the anode. The MEA and the pair of separators constitute one cell. Usually, a stack in which cells are connected in series is formed by stacking a plurality of cells so that the cathode separator and the anode separator are adjacent to each other.

  Specifically, as shown in FIG. 1, the cell 1 has a membrane electrode assembly (MEA) 5 including an anode 2, a cathode 3, and an electrolyte membrane 4 interposed between the anode 2 and the cathode 3. The MEA 5 is sandwiched between the anode separator 10 and the cathode separator 11. The anode 2 includes an anode catalyst layer 6 in contact with the electrolyte membrane 4 and an anode gas diffusion layer 7 in contact with the anode separator 10. The cathode 3 includes a cathode catalyst layer 8 in contact with the electrolyte membrane 4 and a cathode gas diffusion layer 9 in contact with the cathode separator 11. A gasket 14 is disposed on one side surface of the MEA 5 so as to seal the anode 2, and a gasket 15 is disposed on the other side surface so as to seal the cathode 3.

  The anode separator 10 has a fuel flow path 12 that supplies fuel to the anode 2. The cathode separator 11 has an oxidant channel 13 for supplying an oxidant to the cathode 3. A stack is configured by electrically stacking a plurality of cells as shown in FIG. 1 in series. In the stack, one side of one separator may be an anode separator, and the other side may be a cathode separator. Further, one side of one separator may be used as an anode separator or a cathode separator, and the other side may be used as a cooling medium flow path. The current collector plates 16 and 17 may be arranged outside the cell or the stack, respectively, or the end plates 18 and 19 may be arranged, and the end plates may be fastened to fix the fuel cell.

  In the illustrated example, the fuel flow path 12 is formed on the main surface of the anode separator 10 and the oxidant flow path 13 is formed on the main surface of the cathode separator 11, but the fuel flow path is an anode gas diffusion layer. The oxidant flow path may be formed on the main surface of the cathode gas diffusion layer.

  The electrolyte membrane 4 is not particularly limited as long as it has proton conductivity, but is preferably a polymer electrolyte membrane, and more preferably a polymer containing a perfluorocarbon sulfonic acid structure.

  The anode catalyst layer 6 and the cathode catalyst layer 8 each include catalyst particles and a polymer electrolyte having proton conductivity. The catalyst particles are preferably supported on a carbon material. For the catalyst particles, a noble metal such as platinum is preferably used. As the carbon material, a carbon material having a large specific surface area such as carbon black is preferably used.

The behavior related to the diffusion of the fuel gas (hydrogen) and the oxidant gas (for example, air, oxygen) flowing through the fuel cell can be simulated using calculation software. In the simulation, numerical analysis is performed by the difference method of Fick's diffusion equation. In the numerical analysis, the oxidant gas and proton concentrations in the catalyst layer in an equilibrium state are obtained. At this time, the concentration, diffusion coefficient, and saturation concentration of each gas at a predetermined location are used as parameters. When air is used as the oxidant gas, the saturation concentration of air in the gas diffusion layer is the maximum amount of air (mL / cm ³ · atm) contained in the gas diffusion layer, and the porosity of the gas diffusion layer (mL / cm ³ ) may be used as the saturation concentration.

  The fuel gas and the oxidant gas diffuse through the polymer electrolyte membrane covering the catalyst particles and react on the surface of the catalyst particles. The behavior of each gas at this time depends on the oxidant gas and proton concentrations in the catalyst layer. Using the diffusion coefficient and saturation concentration of the oxidant gas and proton in the polymer electrolyte, the concentration of reactive species on the surface of the catalyst particles is calculated, the reaction rate is obtained from the reaction rate equation with adsorption equilibrium, and the limiting current density Is calculated. The chemical engineering concept of mass balance is applied to the stoichiometric treatment of each reactive chemical species. The exchange current density is determined from the concentration of reactive chemical species on the surface of the catalyst particles and the surface area of the catalyst particles. From the limiting current density and the exchange current density, a diffusion overvoltage and a reaction activation overvoltage are calculated by the Butler-Volmer equation, respectively. The resistance overvoltage is calculated based on the electronic resistance of the gas diffusion layer, catalyst layer, separator, and the like. The total overvoltage is calculated from the sum of the diffusion overvoltage, reaction activation overvoltage, and resistance overvoltage, and the generated voltage is obtained.

  The gas diffusion layer according to the embodiment of the present invention simulates the behavior when gas diffuses in the order of the fluid flow path, the gas diffusion layer, and the catalyst layer using the calculation software in the above method, and based on the result. It is designed. According to the simulation results, the smaller the thickness of the gas diffusion layer, the better the VI characteristic of the fuel cell. The improvement of the VI characteristic by reducing the thickness of the gas diffusion layer is more remarkable than the improvement of the VI characteristic by increasing the gas diffusibility of the gas diffusion layer.

  However, when the thickness of the gas diffusion layer becomes smaller than a predetermined critical value, it becomes difficult to further improve the VI characteristic. Further reduction of the gas diffusion layer is necessary in order to reduce the critical value that is expected to improve the VI characteristics.

  Here, the VI characteristic is a characteristic indicating the relationship between the voltage (V) developed by the fuel cell and the current density (I). Generally, the vertical axis represents voltage (V) and the horizontal axis represents current density ( I). The higher the current density obtained at the same voltage, the better the VI characteristics.

In view of the above, the gas diffusion layer according to the first embodiment of the present invention has an average thickness of 150 μm or less, and an oxygen diffusion coefficient Dt in the thickness direction of 1.2 × 10 ⁻³ cm ² / sec or more. 6 × 10 ⁻² cm ² / sec or less.

  The gas diffusion layer according to the second embodiment of the present invention has an oxygen diffusion coefficient Ds in the plane direction perpendicular to the thickness direction larger than the oxygen diffusion coefficient Dt in the thickness direction.

  Moreover, the gas diffusion layer according to the third embodiment of the present invention includes a conductive material and a polymer resin, and the conductive material includes a particulate material and a plate-like material, The acute angle formed by the thickness direction of the gas diffusion layer is controlled to be 60 ° or more and 90 ° or less.

The gas diffusion layers according to the first to third embodiments are all expressed from different aspects, but are designed based on the same concept. Since the gas diffusion layer is porous, if the average thickness of the gas diffusion layer is limited to 150 μm or less as in the first embodiment, gas permeation in the thickness direction tends to be promoted. Therefore, by limiting the oxygen diffusion coefficient Dt, specifically to 6 × 10 ⁻² cm ² / sec or less, the gas permeation in the thickness direction is positively restricted, and the gas diffusion in the surface direction is restricted. There is a need to prompt.

  However, generally, the diffusion coefficient of oxygen in the gas diffusion layer is approximately the same in the thickness direction and in the plane direction. On the other hand, in order to positively limit the gas permeation in the thickness direction, for example, the gas diffusion layer is compressed and densified in the thickness direction, and a barrier against gas movement in the thickness direction is provided in the gas diffusion layer. There is a need. As a result, the oxygen diffusion coefficient Ds in the plane direction is relatively larger than the oxygen diffusion coefficient Dt in the thickness direction. Therefore, the gas diffusion layer according to the second embodiment is normally included in the category of the gas diffusion layer according to the first embodiment.

  In addition, if the gas diffusion layer includes a plate-like material and the acute angle formed by the major axis of the plate-like material and the thickness direction is increased, a barrier against gas movement in the thickness direction is formed in the gas diffusion layer. The oxygen diffusion coefficient Ds in the plane direction is larger than the oxygen diffusion coefficient Dt in the thickness direction. That is, the gas diffusion layer according to the third embodiment is included in the category of the gas diffusion layer according to the first and second embodiments.

  In any of the first to third embodiments, the gas diffusion layer may be composed mainly of a conductive material and a polymer resin. The polymer resin is preferably 2 to 20 parts by mass and more preferably 5 to 12 parts by mass with respect to 100 parts by mass of the conductive material.

  The conductive material may contain a particulate material and a plate-like material, and may further contain a fibrous material. 2-20 mass% is preferable and, as for the ratio of the plate-shaped material to an electroconductive material, 5-10 mass% is more preferable. The proportion of the fibrous material in the conductive material is preferably 0 to 25% by mass, and more preferably 5 to 20% by mass.

  The plate-like material is particulate when viewed macroscopically, but is composed of plate-like particles when viewed microscopically. When the minimum width (thickness) of the plate-like particles is T, the maximum diameter (major axis) L of the plate-like particles viewed from the direction parallel to the direction of the thickness T and the maximum diameter (minor axis) orthogonal to the maximum diameter ) W satisfies L ≧ W, and T is sufficiently smaller than W. The aspect ratio L / T of the plate material preferably satisfies 10 <L / T, and more preferably satisfies 20 ≦ L / T.

  Specific examples of the plate-like material include flaky graphite, graphitized polyimide film pulverized material, graphene and the like. Among them, pulverized graphitized polyimide film and graphene are easy to be oriented in the plane direction of the gas diffusion layer, advantageous for forming a thin gas diffusion layer, and having a gas diffusivity Dt in the thickness direction of the gas diffusion layer. Suitable for aggressively limiting small.

  When viewed microscopically, the particulate material is composed of spherical particles or amorphous particles. However, the spherical particles do not usually exist as spheres having no directivity, and have a major axis and a minor axis. The maximum diameter (major axis) L of spherical particles or irregular particles and the maximum diameter (minor axis) W orthogonal to the maximum diameter satisfy L ≧ W. The aspect ratio of the particulate material: L / W preferably satisfies 1 ≦ L / W ≦ 5, and more preferably satisfies 1 ≦ L / W <2.

  Examples of the particulate material include carbon black, spherical graphite, activated carbon and the like. Among them, it is preferable to use carbon black in terms of high conductivity and a large pore volume. As carbon black, acetylene black, ketjen black, furnace black and the like can be used.

  The polymer resin functions as a binder that binds the conductive materials. From the viewpoint of suppressing the retention of water in the pores in the gas diffusion layer, it is preferable that 50% by mass or more, more preferably 90% by mass or more of the polymer resin is a fluororesin having water repellency. Fluororesin includes PTFE (polytetrafluoroethylene), FEP (tetrafluoroethylene-hexafluoropropylene copolymer), PVdF (polyvinylidene fluoride), ETFE (tetrafluoroethylene-ethylene copolymer), PCTFE (polyethylene). Chlorotrifluoroethylene), PFA (polyfluoroethylene-perfluoroalkyl vinyl ether copolymer) and the like. Among these, from the viewpoint of heat resistance, water repellency, and chemical resistance, the fluororesin is preferably PTFE.

  In any of the first to third embodiments, the average thickness of the gas diffusion layer is preferably 150 μm or less, more preferably 100 μm or less, and even more preferably 80 μm or less or 50 μm or less. When the average thickness of the gas diffusion layer designed as described above is too large, the gas diffusion path length in the thickness direction becomes long, and the power generation performance tends to be difficult to improve. However, the average thickness of the gas diffusion layer is preferably 30 μm or more. Thereby, it is easy to ensure a balanced path length of gas diffusion in the thickness direction and the surface direction.

  When the first main surface of the gas diffusion layer and the second main surface opposite to the first main surface do not have irregularities due to cutting or pressing and are flat visually, the average thickness of the gas diffusion layer is, for example, an arbitrary 10 The thickness of the gas diffusion layer is measured at a location, and the thickness of the gas diffusion layer is obtained as an average value. On the other hand, when the gas diffusion layer has an uneven pattern, the maximum thickness and the minimum thickness of the gas diffusion layer may be obtained and multiplied by the area ratio. The maximum thickness of the gas diffusion layer is obtained, for example, as an average value of the thicknesses of the tenth thickest portions of the convex portions, and the minimum thickness of the gas diffusion layer is the thinnest of the arbitrary ten portions of the concave portions, for example. It is obtained as an average value of the thickness of the part. When the thickness of the gas diffusion layer is 150 μm or less, and further 100 μm or less, it is preferable that the two main surfaces of the gas diffusion layer have no irregularities as described above and are flat visually.

  A fluid flow path may be formed on the first main surface side or the second main surface side of the gas diffusion layer. Thereby, it is not necessary to form a fluid flow path in the separator, and the manufacturing cost of the fuel cell can be easily reduced. In addition, the contact area between the gas flowing in the fluid flow path and the gas diffusion layer is increased, and the gas is easily supplied uniformly by the catalyst layer. The gas diffusion layer having a fluid flow path has a groove that forms the fluid flow path and a rib between the grooves. The groove corresponds to the concave portion, and the rib corresponds to the convex portion.

The oxygen diffusion coefficient Dt in the thickness direction of the gas diffusion layer may be 1.2 × 10 ⁻³ cm ² / sec or more, and the oxygen diffusion coefficient Dt is 6 × 10 ⁻² cm ² / sec or less. It is limited and is preferably 1.2 × 10 ⁻² cm ² / sec or less.

<Measurement of oxygen diffusion coefficient>
The oxygen diffusion coefficient Dt can be measured by the following method. When disassembling an already assembled fuel cell and taking out the gas diffusion layer, the oxygen diffusion coefficient Dt may be measured after the gas diffusion layer is washed with pure water to remove the catalyst layer.

  The gas diffusion coefficient can be obtained from a transmission coefficient obtained by measuring a gas amount per unit time by providing a pressure difference between the front and back of the gas diffusion layer, a transmission area, a thickness of the gas diffusion layer, and a saturation concentration. However, since the gas diffusion layer is quite porous, it is close to the molecular diffusion region and it is difficult to obtain the diffusion coefficient using a gas permeability coefficient measuring device. Therefore, the air permeability of Gurley determined based on JIS P8117: 2009 is measured, and the diffusion coefficient is obtained based on this. The air permeability of Gurley is the time required for 100 mL of air to permeate. The permeability coefficient can be obtained by using the pressure applied to the gas supply side by the permeability of Gurley, the thickness of the gas diffusion layer, the permeation area based on the JIS standard and the weight. The diffusion coefficient can be obtained from the permeability coefficient and the porosity of the gas diffusion layer (that is, the saturation concentration). Specifically, the diffusion coefficient can be obtained from the calculation formula: diffusion coefficient = transmission coefficient / saturation concentration.

  The oxygen diffusion coefficient Ds in the plane direction of the gas diffusion layer is determined by, for example, cutting one gas diffusion layer and dividing it into a plurality of small pieces. On the other hand, the measurement may be performed by diffusing gas in the vertical direction.

  The oxygen diffusion coefficient Dt and the oxygen diffusion coefficient Ds may satisfy Dt <Ds, but the oxygen diffusion coefficient ratio: Ds / Dt is preferably 2 ≦ Ds / Dt, and more preferably 3 ≦ Ds / Dt. Since it is difficult to increase Ds / Dt excessively, the upper limit of Ds / Dt is about 5.

  The acute angle θ formed by the long axis of the plate material and the direction of the thickness T of the gas diffusion layer may be 60 ° or more, but is preferably 80 ° or more, and most preferably 88 to 90 °. The angle θ may be obtained as an average value of acute angles formed by taking a cross-sectional photograph of the gas diffusion layer, arbitrarily selecting 10 plate-like materials on the image, and having their long axes formed in the direction of the thickness T.

  FIG. 2 schematically shows a cross section of a gas diffusion layer containing a plate-like material. A plurality of plate-like particles 20 constituting the plate-like material are observed in the cross-sectional photograph. If ten plate-like particles 20 are arbitrarily selected from these, and a straight line (long-axis line) connecting the end portions 20t on both sides of the long-axis is drawn for each plate-like particle 20, the direction of the long-axis line and the thickness T Can be obtained.

  1-500 micrometers is preferable and, as for the magnitude | size of the major axis L of plate-shaped material, 5-30 micrometers is more preferable. The size of the long axis L may be obtained as an average value of the sizes of the long axes L by taking a cross-sectional photograph of the gas diffusion layer, arbitrarily selecting 100 plate-like materials.

<Method for producing gas diffusion layer>
Process (i)
First, a mixture containing a conductive material, a polymer resin, a surfactant, and a dispersion medium is prepared. A kneader or a mixer may be used for the mixing device. At this time, it is preferable to add a conductive material, a surfactant, and a dispersion medium to the mixing device to uniformly disperse the conductive material in the dispersion medium, and then add and disperse the polymer resin. The polymer resin is preferably imparted with an appropriate shearing force to fibrillate the polymer resin. Examples of the dispersion medium include water, alcohol, and glycols. Examples of the surfactant include polyoxyethylene alkyl ether and alkylamine oxide.

Step (ii)
Next, the obtained mixture is formed into a sheet by a forming method such as extrusion. The obtained sheet may be further rolled. A roll press machine can be used for rolling. The conditions of the roll press are not particularly limited, but it becomes easy to obtain a gas diffusion layer having high strength by rolling at a linear pressure of 0.001 ton / cm to 4 ton / cm.

Step (iii)
Next, the sheet is fired to obtain a fired sheet from which the surfactant and the dispersion medium have been removed. The firing temperature may be any temperature at which the polymer resin does not deteriorate and the surfactant or dispersion medium decomposes or volatilizes. When PTFE is used as the polymer resin, the firing temperature is preferably 310 to 340 ° C. The firing atmosphere may be an inert atmosphere, and for example, an atmosphere such as nitrogen or argon or a reduced pressure atmosphere is preferable. Most of the surfactant and the dispersion medium need only be removed from the sheet, and need not be completely removed.

Step (iv)
Next, the fired sheet has an average thickness of 150 μm or less, and an oxygen diffusion coefficient Dt in the thickness direction of 1.2 × 10 ⁻³ cm ² / sec or more, 6 × 10 ⁻² cm ² / sec. Roll to the following. The higher the pressing pressure in rolling, the smaller the oxygen diffusion coefficient Dt. Also, the oxygen diffusion coefficient ratio: Ds / Dt increases, and the acute angle θ formed by the long axis of the plate-like material and the direction of the thickness T approaches 90 °.

  By providing ribs of a predetermined fluid flow path pattern in a mold used for rolling, grooves serving as fluid flow paths may be formed on at least one of the two main surfaces of the fired sheet. Further, the method of forming the fluid flow path is not limited to the press work as described above, and the fluid flow path may be formed by cutting the main surface of the rolled sheet.

  Next, the simulation result of the influence of the thickness of the gas diffusion layer and the diffusion coefficient of oxygen on the VI characteristics of the fuel cell will be specifically described. In addition, a gas diffusion layer designed based on the simulation result is manufactured, a fuel cell including the gas diffusion layer is assembled, and a result of a verification test is also described.

<< First simulation >>
In a fuel cell having a gas diffusion layer having a general oxygen diffusion coefficient, the relationship between the VI characteristics and the thickness of the gas diffusion layer when the thickness of the gas diffusion layer was changed was investigated. Here, assuming that the oxygen diffusion coefficient Dt in the thickness direction and the oxygen diffusion coefficient Ds in the plane direction are Dt = Ds = 2 × 10 ⁻¹ cm ² / sec, the thickness of the gas diffusion layer is set as follows. It was changed in the range of 200 μm to 80 μm. The gas diffusion layer was assumed to be composed of 100 parts by mass of carbon black having a primary particle diameter of 50 nm and 10 parts by mass of PTFE.

Here, a Nafion film (made by Gore) having a thickness of 15 μm is used as the electrolyte film, and a cathode catalyst layer (catalyst amount: 0) containing carbon black carrying Pt fine particles and Nafion (made by Gore) which is a polymer electrolyte. .1 mg / cm ² ), an anode catalyst layer containing carbon black supporting Pt fine particles and Nafion (manufactured by Gore) was assumed. Hydrogen (dew point 55 ° C., no pressure) is supplied to the anode as a fuel gas so that the utilization rate is 70%, and air (dew point 65 ° C., pressure) is used as the oxidant gas so that the utilization rate is 50% to the cathode. 70 kPa), controlling the load control device so that the current flows constantly, and applying a load with a constant current so that the current density with respect to the electrode area of the anode and the cathode is 2.5 A / cm ² at maximum. Assumed. The results are shown in FIG.

  It can be understood from FIG. 3 that superior VI characteristics can be obtained when a gas diffusion layer having a thickness of 150 μm or less is used as compared with a gas diffusion layer having a thickness of 200 μm. On the other hand, there is no difference in VI characteristics between the case of 100 μm thickness and the case of 80 μm thickness. This is presumably because the oxygen diffusion coefficient Dt in the thickness direction is large and the gas diffusibility in the plane direction is insufficient.

<< Second simulation >>
FIG. 4 shows the results of determining the change in oxygen concentration at the interface between the catalyst layer and the gas diffusion layer (thickness: 100 μm) along the direction orthogonal to the plurality of fluid flow paths running in parallel. Graph A is the result when assuming the case of Dt = Ds = 2 × 10 ⁻¹ cm ² / sec, similarly to the condition of FIG. 3, and graph B shows Dt = Ds / 5 = 4 × 10 ^−2. It is a result when the case of cm ² / sec is assumed. In graph A of FIG. 4, there is a large difference in oxygen concentration index between the region facing the fluid flow path and the region not facing, whereas in graph B, the difference in oxygen concentration index is small. On the other hand, FIG. 5 shows the result of determining the change in the oxygen concentration index at the interface between the catalyst layer and the gas diffusion layer (thickness: 200 μm) along the direction orthogonal to the plurality of parallel fluid flow paths. Graph C in FIG. 5 shows the results when assuming the case of Dt = Ds = 2 × 10 ⁻¹ cm ² / sec, similarly to the condition in FIG. 3. The difference in the oxygen concentration index in graph C is about the same as that in graph B.

  The tendency of FIGS. 4 and 5 suggests that the smaller the thickness, the greater the variation in oxygen concentration between the area facing the fluid flow path and the area not facing. In FIG. 3, there is no difference in the VI characteristics between the case where the thickness of the gas diffusion layer is 100 μm and the case where the thickness is 80 μm. This is probably because it is large. The reason why such a tendency is obtained is that, in the region of the gas diffusion layer facing the fluid flow path, diffusion in the thickness direction occurs preferentially, so that the oxygen concentration in the catalyst layer tends to be non-uniform. It is thought to be a cause. If the oxygen concentration in the catalyst layer becomes uneven, it is considered that the power generation performance is lowered. In order to improve the power generation performance, it is considered important to improve the diffusibility in the plane direction rather than the thickness direction and to diffuse oxygen uniformly in the catalyst layer.

<< Third simulation >>
FIG. 6 shows the VI characteristics obtained in the same manner as in the first simulation except that the thickness of the gas diffusion layer is 80 μm and Dt = Ds / 5 = 4 × 10 ⁻³ cm ² / sec. FIG. 6 also shows the VI characteristic (Ref) when Dt = Ds = 2 × 10 ⁻¹ cm ² / sec. From FIG. 6, even if the oxygen diffusion coefficient Dt is set to 1/50 of the conventional general numerical value, if Dt <Ds (Dt = Ds / 5), more excellent VI characteristics can be obtained. Is shown.

<< 4th simulation >>
FIG. 7 shows the VI characteristics obtained in the same manner as in the first simulation except that the thickness of the gas diffusion layer was 80 μm and Dt = Ds / 5 = 2 × 10 ⁻² cm ² / sec. FIG. 7 also shows the VI characteristic (Ref) when Dt = Ds = 2 × 10 ⁻¹ cm ² / sec. FIG. 7 shows that in order to obtain good VI characteristics, it is important to increase the oxygen diffusion coefficient ratio: Ds / Dt, or if the oxygen diffusion coefficient ratio: Ds / Dt can be appropriately controlled. This shows that even if the oxygen diffusion coefficient Dt changes about 10 times, there is no significant difference in the VI characteristics.

<< 5th simulation >>
In FIG. 8, the thickness of the gas diffusion layer is 150 μm, Dt = Ds = 6.0 × 10 ⁻² cm ² / sec, and the amount of catalyst per area is four times (catalyst amount: 0.4 mg / cm ² ). Except for the above, the VI characteristic (solid line) obtained in the same manner as in the first simulation is shown. FIG. 8 shows a V-I when a gas diffusion layer having the same thickness and oxygen diffusion coefficient Dt = Ds as the simulation model is manufactured, and a fuel cell is manufactured and operated under the same conditions as the model. The characteristic is indicated by a broken line.

<< Sixth simulation >>
In FIG. 9, the thickness of the gas diffusion layer is 100 μm, Dt = Ds = 6.0 × 10 ⁻² cm ² / sec, and the amount of catalyst per area is four times (catalyst amount: 0.4 mg / cm ² ). Except for the above, the VI characteristic (solid line) obtained in the same manner as in the first simulation is shown. FIG. 9 shows a V-I when a gas diffusion layer having the same thickness and oxygen diffusion coefficient Dt = Ds as the simulation model is manufactured, and a fuel cell is manufactured and operated under the same conditions as the model. The characteristic is indicated by a broken line.

<< Seventh simulation >>
In FIG. 10, the thickness of the gas diffusion layer is 80 μm, Dt = Ds = 6.0 × 10 ⁻² cm ² / sec, and the amount of catalyst per area is 4 times (catalyst amount: 0.4 mg / cm ² ). Except for the above, the VI characteristic (solid line) obtained in the same manner as in the first simulation is shown. FIG. 10 shows a V-I when a gas diffusion layer having the same thickness and oxygen diffusion coefficient Dt = Ds as in the simulation model is manufactured, and a fuel cell is manufactured and operated under the same conditions as the model. The characteristic is indicated by a broken line.

  8 to 10 show that the VI characteristic obtained by the simulation and the actually measured VI characteristic coincide with each other with high accuracy. This indicates the high reliability of the calculation software used for the simulation. The calculation software is uniquely programmed according to the algorithm already described by the present inventors. In FIG. 11, the VI characteristics (measured data) indicated by broken lines in FIGS.

<< Eighth simulation >>
In FIG. 12, the thickness of the gas diffusion layer is 100 μm, Dt = Ds / 5 = 1.2 × 10 ⁻³ cm ² / sec, and the amount of catalyst per area is 4 times (catalyst amount: 0.4 mg / cm ² ) The VI characteristic (solid line) obtained in the same manner as in the first simulation is shown except for the above. FIG. 12 shows a V-I when a gas diffusion layer having the same thickness and oxygen diffusion coefficients Dt and Ds as the simulation model is manufactured, and a fuel cell is manufactured and operated under the same conditions as the model. The characteristic is indicated by a broken line. Here, Dt and Ds were controlled to desired values by containing 7.5% by mass of scaly graphite in the gas diffusion layer for actually measuring the VI characteristic.

<< 9th simulation >>
FIG. 13 shows that the thickness of the gas diffusion layer is 100 μm, 80 μm, or 50 μm, Dt = Ds / 5 = 1.2 × 10 ⁻³ cm ² / sec, and the amount of catalyst per area is four times (catalyst amount: 0). The V-I characteristic obtained in the same manner as in the first simulation is shown except that it is 4 mg / cm ² ). Here, the case where 7.5 mass% of scaly graphite was contained in the gas diffusion layer was assumed. From FIG. 13, it can be understood that the VI characteristic is greatly improved by reducing Dt and reducing the thickness of the gas diffusion layer.

  According to the present invention, a fuel cell having excellent power generation performance can be obtained.

1 Cell 2 Anode 3 Cathode 4 Electrolyte Membrane 5 Membrane Electrode Assembly (MEA)
6 Anode catalyst layer 10 Anode separator 7 Anode gas diffusion layer 8 Cathode catalyst layer 11 Cathode separator 9 Cathode gas diffusion layer 14, 15 Gasket 12 Fuel flow path 13 Oxidant flow path 16, 17 Current collector plate 18, 19 End plate 20 Plate Shaped particles 20t End of long axis

Claims (9)

A gas diffusion layer for a fuel cell,
The gas diffusion layer has an average thickness of 150 μm or less;
A gas diffusion layer having an oxygen diffusion coefficient Dt of 1.2 × 10 ⁻³ cm ² / sec or more and 6 × 10 ⁻² cm ² / sec or less in the thickness direction of the gas diffusion layer. The gas diffusion layer includes a conductive material and a polymer resin,
The gas diffusion layer according to claim 1, wherein the conductive material includes a particulate material and a plate-like material.   The gas diffusion layer according to claim 2, wherein an acute angle formed by the major axis of the plate-like material and the thickness direction is 60 ° or more and 90 ° or less.   The gas diffusion layer according to claim 2 or 3, wherein the plate-like material contains graphene.   The gas diffusion layer according to claim 1, wherein an average thickness of the gas diffusion layer is 30 μm or more. A gas diffusion layer for a fuel cell,
A gas diffusion layer in which an oxygen diffusion coefficient Ds in a plane direction perpendicular to the thickness direction is larger than an oxygen diffusion coefficient Dt in the thickness direction of the gas diffusion layer. A gas diffusion layer for a fuel cell,
The gas diffusion layer includes a conductive material and a polymer resin,
The conductive material includes a particulate material and a plate material,
A gas diffusion layer, wherein an acute angle formed by a major axis of the plate-like material and a thickness direction of the gas diffusion layer is 60 ° or more and 90 ° or less. An electrolyte membrane, a catalyst layer disposed on each main surface of the electrolyte membrane, and a pair of gas diffusion layers disposed on the opposite side of the catalyst layer from the electrolyte membrane,
A fuel cell in which at least one of the pair of gas diffusion layers is the gas diffusion layer according to any one of claims 1 to 7. (I) preparing a mixture including a conductive material, a polymer resin, a surfactant, and a dispersion medium;
(Ii) forming the mixture into a sheet;
(Iii) firing the sheet to obtain a fired sheet from which the surfactant and the dispersion medium have been removed;
(Iv) The fired sheet has an average thickness of 150 μm or less, and an oxygen diffusion coefficient Dt in the thickness direction of 1.2 × 10 ⁻³ cm ² / sec or more, 6 × 10 ⁻² cm ² / and a step of rolling so as to be equal to or less than sec.

Images