Carbon fiber fabric for fuel cell gas diffusion layer

Abstract

[Research Topic] To provide a carbon fiber fabric for a fuel cell gas diffusion layer that maintains the conductivity required for a gas diffusion layer in a fuel cell, promotes the diffusion of generated water, and provides excellent cushioning properties that can adapt to the expansion and contraction of the polymer membrane caused by changes in the ambient temperature of the fuel cell stack, and a method for manufacturing the same. [Solution] In a carbon fiber fabric for a fuel cell gas diffusion layer, which is a worsted yarn woven from warp and weft yarns, the diameter of the single fiber of the carbon fiber fabric is 8 μm or less, and the weight per unit area is 60 g / m². 2 The following applies. Additionally, when at least one of the warp or weft yarns is a twisted yarn, the twist angle of the twisted yarn can be 35° or less; when at least one of the warp or weft yarns is a double-ply yarn, the twist angle of the double-ply yarn can be 20° or less.

Description

Technical Field

[0001] This invention relates to carbon fiber fabrics for gas diffusion layers of fuel cells installed in vehicles, ships, aircraft and other transportation vehicles. Background Technology

[0002] Solid polymer fuel cells (hereinafter referred to as fuel cells or FCs), which are mainly used in homes and vehicles, are structures composed of repeated layers of unit cells (hereinafter referred to as battery cells) including an electrolyte membrane / platinum catalyst bonding layer (hereinafter referred to as CCM layer) formed by bonding electrodes on both sides of a polymer membrane, a gas diffusion layer (hereinafter referred to as GDL) that guides fuel gas and oxidant gas to the electrode reaction area, a separator with gas inlet / outlet channels, and sealing materials.

[0003] GDL (Gas Diode Layer) is typically a thin sheet-like component less than 1 mm thick, requiring the ability to smoothly supply either hydrogen-containing fuel gas or oxygen-containing oxidant gas from external sources to the electrode catalyst layer. In addition, the basic functions of GDL include: 1) sufficiently low resistance for efficient energy extraction; 2) good diffusion properties that allow for sufficient gas permeability for extracting large currents and the drainage of water generated by the battery; 3) buffering (elasticity) to absorb uneven thickness and variations in the laminated components; 4) corrosion resistance even against strong acids and alkalis; and 5) the ability to recover its original shape even under repeated compressive stress.

[0004] Previous GDLs, being paper-making structures formed from carbon fibers (carbon paper), suffered from insufficient drainage performance (drainage capacity) over a wide current range, and low buffering capacity (elasticity) to absorb deformation caused by variations in thickness, temperature, and humidity. Therefore, many GDLs made of carbon fiber fabrics have been developed (see Patent Documents 1-3).

[0005] Existing technical documents

[0006] Patent documents

[0007] Patent Document 1: Japanese Patent Application Publication No. 2013-108188

[0008] Patent Document 2: Japanese Patent Application Publication No. 2012-202003

[0009] Patent Document 3: International Publication No. 2003-034519 Summary of the Invention

[0010] The problem the invention aims to solve

[0011] For recent fuel cell stacks, many cell units, comprising three components—CCM, gas diffusion layer, and separators—are arranged within a stack frame of a certain size. In this stack, the polymer membrane, acting as the electrolyte, expands and contracts in the thickness direction due to changes in ambient temperature and humidity. Therefore, the gas diffusion layer also absorbs this expansion and contraction, ensuring sufficient pressure between components to maintain contact resistance. Here, the separators, made of metal or graphite, do not absorb expansion and contraction in the thickness direction. Therefore, the gas diffusion layer requires flexibility and cushioning to accommodate the expansion and contraction of the polymer membrane itself.

[0012] For carbon paper, short carbon fibers are dispersed in a planar direction and bonded with resin in a paper-like manner, but there is a lack of buffering in the thickness direction. In addition, in order to reduce the resistance of the bonding resin, it is carbonized at temperatures above 1000°C, thus becoming brittle. If subjected to repeated compressive and torsional stresses over a long period of time, tiny cracks will form in the bonded area.

[0013] As a result, the carbon paper gradually thins in the thickness direction, leading to a decrease in contact resistance due to reduced pressure, which may accelerate the battery stack's lifespan. In other words, the carbon paper lacks sufficient buffering capacity to cope with temperature and humidity changes in the battery stack, and its durability is insufficient under repeated compressive stress within the battery stack.

[0014] On the other hand, the carbon fiber fabrics disclosed in the aforementioned patent documents 1-3, compared to conventional papermaking structures (carbon paper) formed from carbon fibers, have sinusoidal interlaced yarns, resulting in superior cushioning (elasticity). However, since all of them use carbon fiber precursors (flame-resistant yarn) as starting materials to form the fabric, the thickness is also increased, and the weight per unit area must also be increased. That is, the tensile strength of the carbon fiber precursor yarn is reduced due to dry steaming treatment, and in order to form it into a fabric, it is necessary to increase the fineness of the yarn and the diameter of the single fiber. Carbon fiber fabrics with large single fiber diameters for fuel cell gas diffusion layers present practical problems.

[0015] Therefore, the object of the present invention is to provide a carbon fiber fabric for a fuel cell gas diffusion layer that maintains the conductivity that a gas diffusion layer of a fuel cell should have, while having excellent thickness-direction buffering properties that can follow the expansion and contraction of the polymer membrane due to changes in ambient temperature within the fuel cell stack, and excellent power generation performance over a large current range due to its thinness and low resistance.

[0016] Solution for solving the problem

[0017] To address the aforementioned problems, this invention provides a carbon fiber fabric for a fuel cell gas diffusion layer, which is composed of interwoven warp and weft yarns (often classified as worsted yarn), in which the diameter of a single fiber is less than 8 μm and the weight per unit area is 60 g / m². 2The following carbon fiber fabrics. When at least one of the warp or weft yarns is twisted yarn, the twist angle of the twisted yarn can be 35° or less. When at least one of the warp or weft yarns is double-ply yarn, the twist angle of the double-ply yarn can be 20° or less.

[0018] The effects of the invention

[0019] The carbon fiber fabric used in the gas diffusion layer of the fuel cell of the present invention, through the carbon fiber fabric structure with a reduced single fiber diameter, reduces the resistivity in the thickness direction, and promotes the diffusion of generated water in the yarn bundle direction by utilizing capillary phenomenon, resulting in good power generation performance over a large current range. Furthermore, by keeping the twist angle of the interlaced yarns below 35°, strength can be ensured. Here, the polymer membrane expands and contracts due to repeated changes in temperature and humidity within the fuel cell stack, but by ensuring the buffering capacity of the mechanism that adjusts the pressure, even with increased cycles, the reduction in thickness and the lack of decrease in contact resistance are maintained. Attached Figure Description

[0020] Figure 1 The tensile test piece used in Example 1.

[0021] Figure 2 This is a microscope photograph of comparative material 2 in Example 2.

[0022] Figure 3 The results are from the power generation performance test of Example 4. Detailed Implementation

[0023] An embodiment of the carbon fiber fabric used as the gas diffusion layer for fuel cells according to the present invention will be described. The carbon fiber fabric of the present invention is a fabric formed by interlacing warp and weft yarns, which are worsted yarns. The diameter of the individual fibers constituting the warp and weft yarns is 8 μm or less, and the weight per unit area of ​​the fabric is 60 g / m². 2 The following applies: When at least one of the warp or weft yarns is a twisted yarn, the twist angle of the twisted yarn is 35° or less. Furthermore, when at least one of the warp or weft yarns is a double-ply yarn, the twist angle of the double-ply yarn is 20° or less. Here, the diameter of a single fiber in the carbon fiber fabric is measured at any five points under magnification (2500x) using a microscope or similar equipment, and the average value is calculated. The smaller this diameter, the smaller the equivalent radius of the space ensured between these single fibers, the more the capillary effect is promoted, and the further the diffusivity of the water generated during power generation is improved.

[0024] Example

[0025] (Example 1)

[0026] Tensile tests were conducted to determine the impact of the various morphologies of the individual fibers constituting the carbon fiber fabric on tensile strength. The tensile strength of carbon fiber fabric varies considerably depending on its manufacturing method. To confirm whether the fabric can withstand the tensile strength under the production processes and usage conditions up to its final service state, even in electrode applications where compressive stress is primarily exerted, the relationship between the carbon fiber fabric composition conditions and the tensile strength of the individual yarns was determined.

[0027] The test pieces used in this tensile test were prepared using the following methods (1) to (5) in a total of 12 types, including 8 levels of the invention materials 1 to 8 and 4 levels of the comparative materials 1 to 4 as comparative samples.

[0028] (1) Select single yarns of 0.4 to 1.0 dtex from commercially available acrylic textile raw cotton and make them into worsted yarn.

[0029] (2) After forming a combination of warp and weft yarns with 36 to 90 yarns per inch from the worsted yarn, a refined white cloth is obtained as a fabric with a fabric coverage factor in the range of 930 to 1100. In addition, a fabric with a twist angle of 0 degrees refers to a white cloth that is essentially untwisted by untwisting lost fibers with a twist number approximately the same as that of acrylic worsted yarn, forming a woven fabric, and then washing away the lost fibers.

[0030] (3) The white cloth was subjected to flame-resistant baking in an oxidizing atmosphere at 200-255°C for 1-15 hours.

[0031] (4) Cut the flame-resistant fabric to the specified size, keep it in a nitrogen atmosphere at 1250°C for 10 minutes and then cool it.

[0032] (5) After cooling, the carbon fiber fabric is demolded with the warp or weft yarns of the invention and comparative articles aligned in the same direction, thereby producing... Figure 1 Tensile test pieces of the dimensions shown are presented. The fabric type, single fiber diameter, unit area weight, yarn type (twisted or biply), twist angle, and yarn width of these carbon fiber fabrics are shown in Tables 1 to 3. It should be noted that the yarn bundle spacing, yarn width, and twist angle were calculated as average values ​​obtained from five measurements of each of the warp and weft yarns under a microscope at 250x magnification.

[0033] [Table 1]

[0034]

[0035] [Table 2]

[0036]

[0037] [Table 3]

[0038]

[0039] For the tensile test, a tensile testing machine (LTS series load tester manufactured by MinebeaMitsumi Inc.) was used to stretch the aforementioned tensile test piece at a tensile speed of 0.2 mm per second until the test piece broke, and its peak strength was measured. Furthermore, the single yarn strength was calculated from the measured peak strength using the following formula. The peak strength and tensile strength of the fabrics of Inventive Materials 1-8 and Comparative Materials 1-4 are shown in Tables 4-6.

[0040] • Tensile stress σ (unit: N / mm) 2 )=T / (0.785×d 2 ×Nm×Ns)

[0041] • T: Peak intensity (unit: N / cm width)

[0042] • Nm: Number of yarn bundles per unit width of 1 cm

[0043] • Ns: Number of single yarns per yarn bundle (measured using a VHX7000 from KEYENCE CORPORATION, magnified to 2500 times the cross-section of the yarn bundle).

[0044] ·d (unit: mm): Diameter of a single yarn (measured using a KEYENCE CORPORATION VHX7000 magnified 2500 times).

[0045] [Table 4]

[0046]

[0047] [Table 5]

[0048]

[0049] [Table 6]

[0050]

[0051] The results of this tensile test show that the single yarn strength of the invented materials 1-8 (twist angle less than 35°) is higher than that of the comparative materials 1-4 (twist angle greater than 35°). This is believed to be because, for the invented materials, the frequency of internal defects caused by thermal shrinkage during firing is reduced compared to the comparative materials. In particular, although the single yarn diameter of invented materials 1 and 2 is smaller than that of comparative material 4, the single yarn strength is also increased.

[0052] (Example 2)

[0053] When using carbon fiber fabric as a gas diffusion layer, the compression conditions caused by the separator were considered, and the compression test confirmed that the fabric could ensure practical compressive strength. The test pieces in this test were made using a total of 12 types of carbon fiber fabrics, with 8 levels of inventive materials 1 to 8 and 4 levels of comparative materials 1 to 4 as the comparative materials.

[0054] Test pieces with a diameter greater than 1 cm were clamped between two parallel abrasive plates (1 cm in diameter) and pressed at 3 MPa. After pressing, the abrasive plates were separated from the test pieces, and the condition of the single yarn was observed using a magnifying glass to examine the indentations caused by the abrasive plates. Since this is not a destructive test like the tensile test in Example 1, it is advantageous to treat test pieces with guaranteed compressive strength as qualified products.

[0055] Even under pressure of 3 MPa, the inventive materials 1-8 did not exhibit single-yarn breakage at the pressure mark (contact area). In contrast, comparative materials 1-3 showed single-yarn breakage at the pressure mark under 3 MPa pressure. A microscopic photograph (magnification: 500x) of comparative material 4, where single-yarn breakage was confirmed on the surface, is shown below. Figure 2 As shown above, the test results indicate that the installation pressure of the gas diffusion layer made of carbon fiber as an electrode is usually 1 MPa. However, even under a higher load of 3 MPa, the invented materials 1 to 8 will not be locally damaged by compression and can be used for electrode applications.

[0056] (Example 3)

[0057] Carbon fiber gas diffusion layers are used in fuel cell stacks with multiple battery cells stacked in a specific frame size. In this case, the gas diffusion layer absorbs the deformation caused by the expansion and contraction of the polymer membrane due to temperature and humidity changes. Therefore, the gas diffusion layer must possess the resilience to return to its original shape even after repeated compressive stress over many years. A certain contact pressure must be ensured to prevent an increase in contact resistance. To determine whether deformation occurs in the thickness direction even under repeated compressive stress, repeated pressing tests are conducted.

[0058] That is, assuming the compression caused by the metal separator during actual operation, the fabric was investigated to determine whether it could ensure practical compressive strength. The test pieces used in this experiment were similar to those in Example 1, using commercially available carbon fiber fabrics of a total of five types and carbon paper of Comparative Material 5, with the three levels of Inventive Materials 4, 6, and 8 as the test pieces and the two levels of Comparative Materials 3 and 4 as the comparative materials. The thickness of the test pieces was measured after 20, 40, 60, 80, and 100 cycles of compression and recovery. The reduction in thickness (in μm) of the test pieces compared to the initial thickness is shown in Table 7.

[0059] [Table 7]

[0060]

[0061] Test pieces with a diameter greater than 1 cm were clamped between two parallel grinding plates (1 cm in diameter) and repeatedly pressed from no load up to 2 MPa. The thickness reduction was measured relative to the thickness after two cycles at no load, and then at no load for every 20 cycles thereafter, as shown in Table 4. It should be noted that Invention Material 8 is based on Invention Material 6 with a unit area weight of 12 g / m². 2 The coating consists of an MPL (microporous layer) containing PTFE and carbon black. Furthermore, Comparative Material 3 is the same material as Comparative Material 3 in Table 3.

[0062] Inventive materials 4, 6, and 8, through repeated 20 cycles of pressing up to 2 MPa, achieve a stable arrangement of warp and weft yarns, resulting in an initial reduction in thickness. However, the reduction in thickness does not increase with each subsequent 20 cycles, maintaining a stable thickness. On the other hand, comparative material 3, with a twist angle exceeding 38°, suffers from micro-cracks caused by the tightening during firing, which slowly lead to single-yarn breakage due to the repeated pressing test. This results in a gradual increase in thickness reduction and an increase in electrical resistance due to reduced pressing pressure, making it unsuitable for practical use.

[0063] Furthermore, the comparative material 4 of the commercially available carbon fiber fabric is the same as that used in Examples 1 and 2, such as... Figure 2 As shown, single yarn breaks occur in the pressed portion. This is believed to be due to the single yarn diameter of 8.4 μm, the number of single yarns of 350, the yarn width of 472 μm, and the weight per unit area of ​​130 g / m². 2 Because the gaps between the individual yarns are small due to the pressure, they break sequentially starting from the yarn to which the maximum breaking stress is applied.

[0064] Furthermore, the commercially available comparative material 5 is carbon paper with an MPL (microporous layer). This commercially available product also exhibits a gradual decrease in thickness with repeated pressing, leading to long-term residue issues. This is believed to be because while the carbon fiber core material of this comparative material has sufficient strength, the thermosetting resin used to bond its joints carbonizes and becomes brittle, slowly generating internal microcracks due to repeated pressing, thus increasing the thickness reduction. It is known that the pressure (surface pressure) generated when using a gas diffusion layer as an electrode is typically 1 MPa, and inventive materials 4, 6, and 8 can be used even under repeated pressing with a load of 2 MPa.

[0065] (Example 4)

[0066] Imagine that the fabricated carbon fiber fabric is installed in a fuel cell of an actual vehicle to conduct a power generation test. Test pieces of the inventive material 7 used in Example 1 and the comparative material 5 used in Example 3 are each cut into two 1cm square pieces. For the power generation test, the test pieces are assembled with the positive and negative electrodes of the 1D battery cell adjusted to a thickness of 1MPa, with the inventive material 7 forming a thickness of 65μm and the comparative material 5 forming a thickness of 200μm. Here, the electrolyte membrane sandwiched between the test pieces and the gas diffusion layers at the two electrodes is a 25μm thick Nafion membrane, on both sides of which platinum microparticles, used as catalysts, are supported on carbon black at a concentration of 0.5mg / cm². 2 .

[0067] The assembled 1D battery cell was connected to the fuel cell test device. Under overhumidification, the voltage was increased and decreased to 0.9–0.2V to complete the trial run. Then, 500cc of oxygen (1% nitrogen) and 99% nitrogen gas were introduced per minute on the hydrogen side and 1000cc per minute on the air side to change the voltage to 0.9–0.1V. The actual extraction current was measured at this point. The current measurement results of Invention Material 7 and Comparative Material 5 are as follows: Figure 3 As shown. It should be noted that its evaluation benchmark is based on the implementation of "Analysis of Gas Diffusion Resistance of Fuel Cell Using 1D Battery Unit (Automotive Technology Association Academic Lecture Proceedings 20133647)", which makes the power generation performance of the large current range more significant.

[0068] The results of comparing power generation performance are as follows Figure 3 As shown, for example, the power generation at 0.4V increases by approximately 2 times. Comparing material 5 with carbon paper used in existing fuel cell vehicles, the fabric structure of the invented material 7 exhibits good power generation performance over a wide current range, also demonstrates capillary action, and provides good drainage of generated water, significantly improving efficiency. This indicates that a lightweight and compact fuel cell unit can be formed. In this application, "weight per unit area" is defined in Japanese Industrial Standard (JIS) L02028 as "a unit for expressing the mass per unit area of ​​woolen fabrics, etc., and is defined as per 1m²". 2 Synonyms for "grams".

Claims

1. A carbon fiber fabric for a fuel cell gas diffusion layer, characterized in that, It is a carbon fiber fabric used in the gas diffusion layer of fuel cells, which is made of interwoven warp and weft yarns of worsted yarn. The carbon fiber fabric has a single fiber diameter of less than 8 μm and a unit area weight of 60 g / m². 2 the following, At least one of the warp or weft yarns is a twisted yarn, and the twist angle of the twisted yarn is less than 35°. Carbon fiber fabric test pieces with a diameter greater than 1 cm were sandwiched between two parallel grinding plates and repeatedly pressed from no load up to 2 MPa. The thickness was measured relative to the thickness at no load after two cycles. The thickness reduction at no load was then measured every 20 cycles thereafter. The initial 20 cycles of pressing produced the initial thickness reduction, and the thickness reduction did not increase every subsequent 20 cycles.

2. The carbon fiber fabric for the gas diffusion layer of a fuel cell according to claim 1, characterized in that, At least one of the warp or weft yarns is a double-ply yarn, and the twist angle of the double-ply yarn is less than 20°.

Images