JP6964448B2 - Mold carbon material for fuel cell catalyst carrier, catalyst layer for fuel cell, and fuel cell - Google Patents

Description

The present invention relates to a template carbon material for a fuel cell catalyst carrier , a fuel cell catalyst layer, and a fuel cell.

A polymer electrolyte fuel cell, which is a type of fuel cell, has a pair of catalyst layers arranged on both sides of a solid polymer electrolyte membrane, a gas diffusion layer arranged outside each catalyst layer, and each gas diffusion layer. It is provided with a separator arranged on the outside. Of the pair of catalyst layers, one catalyst layer serves as the anode of the polymer electrolyte fuel cell, and the other catalyst layer serves as the cathode of the polymer electrolyte fuel cell. In a normal polymer electrolyte fuel cell, a plurality of unit cells having the above components are stacked in order to obtain a desired output.

A reducing gas such as hydrogen is introduced into the separator on the anode side. The gas diffusion layer on the anode side is introduced into the anode after diffusing the reducing gas. The anode includes a catalyst component, a catalyst carrier supporting the catalyst component, and an electrolyte material having proton conductivity. On the catalyst component, an oxidation reaction of the reducing gas occurs to generate protons and electrons. For example, when the reducing gas becomes hydrogen gas, the following oxidation reaction occurs.
H ₂ → 2H ⁺ + 2e ⁻ (E ₀ = 0V)

The protons generated in this oxidation reaction are introduced into the cathode through the electrolyte material in the anode and the solid polyelectrolyte membrane. In addition, electrons are introduced into an external circuit through a catalyst carrier, a gas diffusion layer, and a separator. This electron is introduced into the separator on the cathode side after working in an external circuit. Then, these electrons are introduced into the cathode through the separator on the cathode side and the gas diffusion layer on the cathode side.

The polymer electrolyte membrane is composed of an electrolyte material having proton conductivity. In the solid polymer electrolyte membrane, the protons generated by the above oxidation reaction are introduced into the cathode.

Oxidizing gas such as oxygen gas or air is introduced into the separator on the cathode side. The gas diffusion layer on the cathode side is introduced into the cathode after diffusing the oxidizing gas. The cathode includes a catalyst component, a catalyst carrier supporting the catalyst component, and an electrolyte material having proton conductivity. On the catalyst component, a reduction reaction of the oxidizing gas occurs to generate water. For example, when the oxidizing gas becomes oxygen gas or air, the following reduction reaction occurs.
_{^{O 2 + 4H + + 4e -}} → 2H 2 O (E 0 = 1.23V)

The water generated in the reduction reaction is discharged to the outside of the fuel cell together with the unreacted oxidizing gas. As described above, in the polymer electrolyte fuel cell, power is generated by utilizing the energy difference (potential difference) between the oxidation reaction and the reduction reaction. In other words, the electrons generated by the oxidation reaction do the work in the external circuit.

Patent Documents 1 and 2 disclose a method for producing porous carbon (template carbon) prepared by using an oxide as a template as activated carbon that can be used as a catalyst carrier of a fuel cell. Specifically, Patent Document 1 discloses a method for producing activated carbon, which comprises a step of heating and firing a mixture of an organic resin and an alkaline earth metal compound in a non-oxidizing atmosphere. In this technique, the pore size of activated carbon is adjusted by the crystallite size of the alkaline earth metal oxide produced after the mixture is calcined and carbonized.

Patent Document 2 discloses a method for producing activated carbon using an alkaline earth metal compound as an activator. In the technique disclosed in Patent Document 2, the organic resin is mixed with an alkaline earth metal compound and fired by heating in a non-oxidizing atmosphere to activate the organic resin while carbonizing it. After that, activated carbon is produced by removing the activator. The mesopores contained in this activated carbon have an average diameter of crystallite size of the activator.

Further, Patent Document 1 and Patent Document 2 describe that the lower limit temperature of a preferable heating and firing temperature is 500 ° C. or higher, more preferably 700 ° C. or higher, and the preferable upper limit temperature is 1500 ° C. or lower, more preferably 1200 ° C. or lower. By heating and firing, "the reason has not been clarified, but the porosity of the carbides produced by the action of the coexisting alkaline earth metal compounds is significantly enhanced, and the specific surface area is significantly higher than that of ordinary activated carbon. A large activated carbon can be obtained. " Further, in Patent Document 2, as a result of microscopic observation of a carbide containing an alkali metal oxide that has been carbonized, the surface of the oxide particles is coated with carbon having a large specific surface area. It is estimated that the carbonization of organic carbon progresses on the surface of the oxide powder and at the same time, the porosity is promoted.

Patent Document 3 discloses a catalyst for a polymer electrolyte fuel cell, which comprises a catalyst component and a catalyst carrier. This catalyst has mesopores with a radius of 1 to 10 nm derived from porous carbon (template carbon) prepared by using an oxide as a template, and the mode radius of the mesopores is 2.5 to 10 nm.

Japanese Unexamined Patent Publication No. 2006-62954 Japanese Unexamined Patent Publication No. 2014-122158 International Publication No. 2014/175106

The carbon material used in the techniques disclosed in Patent Documents 1 to 3 is porous carbon (template carbon) made by using an oxide as a template, but the template carbon obtained by the disclosed techniques is used as a fuel cell. When used as a fuel cell, the power generation performance is particularly high during power generation in a high-humidity environment or during constant-current power generation immediately after high-power power generation (hereinafter, also referred to as "power generation in a high-humidity environment"). Couldn't be high enough.

Therefore, the present invention has been made in view of the above problems, and an object of the present invention is a new and improved mold carbon material for a fuel cell catalyst carrier capable of further improving power generation performance. , A method for producing a template carbon material for a fuel cell catalyst carrier, a fuel cell catalyst layer, and a fuel cell.

The present inventor considered that the cause of the decrease in the power generation performance of the fuel cell during power generation in a highly humidified environment is the decrease in the catalyst utilization rate due to flooding. Here, the flooding means the blockage of the pores by water. That is, a large amount of water is generated at the cathode during power generation in a highly humidified environment. In addition, during constant current power generation immediately after high-power power generation, a large amount of water generated at the cathode during high-power power generation remains. Then, such water blocks the pores in the catalyst carrier. That is, flooding occurs in the pores. In this case, it becomes difficult for the oxidizing gas to reach the catalyst component supported in the back of the pores. That is, the gas diffusibility is reduced. Therefore, the catalyst component supported in the back of the pores is not used. As a result, the catalyst utilization rate decreases.

Next, the present inventor examined a technique for suppressing flooding. In the conventional technique, the catalyst utilization rate has been improved by adjusting the average pore diameter of the catalyst carrier or the peak position (mode diameter) of the pore diameter distribution. Although a certain effect was obtained by this method as well, flooding in the pores could not be suppressed as described above.

Therefore, the present inventor paid attention to the shape of the pores. Specifically, the present inventor considered that flooding is likely to occur particularly in bottleneck type pores. FIG. 5 shows a bottleneck type pore 30 as an example of the bottleneck type pore. The bottleneck type pore 30 is composed of a bottle portion 30a and a neck portion 30b communicating with the bottle portion 30a. The diameter of the bottle portion 30a is larger than the diameter of the neck portion 30b. Further, the neck portion 30b communicates with the outer surface of the carbon material or other pores in the carbon material. Then, flooding is likely to occur at the neck portion 30b. Then, when flooding occurs in the neck portion 30b, it becomes difficult for the oxidizing gas to reach the catalyst component supported on the bottle portion 30a. Therefore, the catalyst component supported on the bottle portion 30a is not used.

Therefore, the present inventor increases the minimum diameter of the neck portion (so-called "neck portion of pores") and narrows the distribution of the diameter of the neck portion (in other words, reduces the variation in diameter). It was investigated. When the minimum diameter of the neck portion is increased, the neck portion of each bottleneck type pore is less likely to be flooded, so that the catalyst component supported on the bottle portion is more likely to be used. Further, if the distribution of the diameter of the neck portion is narrowed, the oxidizing gas can be evenly supplied to each bottle portion. As a result, the catalyst components supported on each bottle can be used evenly. As a result, the catalyst utilization rate becomes high. That is, when the distribution of the diameter of the neck portion is wide, there are a plurality of bottleneck type pores having different diameters of the neck portion. In this case, the amount of oxidizing gas supplied to the bottle portion of each bottleneck type pore will vary. That is, the gas diffusibility is reduced. Then, in the bottle portion where the supply amount of the oxidizing gas is small, a catalyst component that cannot be sufficiently contacted with the oxidizing gas (that is, the contribution to the power generation reaction is small) may be generated. Therefore, the catalyst utilization rate also decreases in this case. As a result of the above examination, the present inventor has obtained the following findings.

<Knowledge 1>
It was found that the relative pressure at which the hysteresis loop closes in the nitrogen adsorption / desorption isotherm and the slope of the desorption isotherm can be used as an index of the distribution of the minimum diameter of the neck portion and the diameter of the neck portion.

Hysteresis loops may occur on nitrogen adsorption isotherms of porous materials. That is, the isotherm on the adsorption side (that is, the nitrogen adsorption isotherm) and the isotherm on the desorption side (that is, the desorption isotherm) do not match, and as a result, a hysteresis loop occurs. Nitrogen adsorption isotherm is a general term for nitrogen adsorption isotherm and nitrogen desorption isotherm. FIG. 4 shows a hysteresis loop A as an example of the hysteresis loop. (P / P ₀ ) c in FIG. 4 shows the closing relative pressure of the hysteresis loop. The amount of nitrogen desorption dV between (P / P ₀ ) c + 0.1 to (P / P ₀ ) c corresponds to the slope of the desorption isotherm. There are various theories as to the cause of the hysteresis loop, but the theory that the hysteresis loop occurs when the bottleneck type pores are present has been proposed.

In this theory, as explained next, a hysteresis loop occurs due to the difference between the adsorption process and the desorption process. That is, in the adsorption process, the adsorption thickness of nitrogen gradually increases according to the relative pressure of the nitrogen gas. In the process, the neck part of the bottleneck type pore is first closed by adsorption, but at that time, even if nitrogen is adsorbed to the bottle part with a thickness equivalent to the neck part, the diameter of the bottle part is larger than the diameter of the neck part. , The bottle part is not completely filled yet. In this state, even if the neck portion is closed, the external pressure and the pressure inside the bottle portion are in equilibrium through the nitrogen adsorption layer. Therefore, if the relative pressure of the nitrogen gas, that is, the external pressure is further increased, the pressure inside the bottle portion is also increased. Therefore, the adsorption corresponding to the external pressure also proceeds in the bottle portion, and the nitrogen adsorption layer in the bottle portion becomes thick. When the external pressure is further increased, the bottle portion is finally completely closed by the nitrogen adsorption layer. That is, even in the case of bottleneck type pores, adsorption proceeds in the same manner as in a normal pore structure in the adsorption process. However, strictly speaking, when the neck portion is closed by the nitrogen adsorption layer, at least a part of the nitrogen adsorption layer adsorbed on the neck portion becomes a liquid phase. It is known that the pressure inside the liquid phase (that is, the pressure inside the bottle portion) is slightly smaller than the external pressure due to the surface tension of the liquid phase. This phenomenon is the so-called capillary phenomenon. Therefore, in the adsorption isotherm of the bottleneck type pores, the relative pressure required to obtain the same adsorption amount as the non-bottleneck type pores shifts to a slightly higher value. Next, consider the elimination process of nitrogen from the state where all the pores of the bottleneck type pores are filled. In the desorption process, the neck portion does not open until the external pressure drops to the pressure at which the neck portion is released. Therefore, the desorption isotherm falls at a relative pressure corresponding to the diameter of the neck portion. In the process of reducing the external pressure to the pressure at which the neck portion is released, the external pressure becomes the pressure at which the suction of the bottle portion is released. However, even in this state, the nitrogen adsorbed inside the bottle portion due to the blockage of the neck portion is retained as it is (so-called blocking phenomenon). Therefore, as the external pressure, that is, the relative pressure, is lowered, the hysteresis loop is eventually closed. It is known that when the minimum diameter of the neck portion is small, the relative pressure when the hysteresis loop is closed is determined by the adsorbate and the measurement temperature regardless of the minimum diameter of the neck portion. That is, when the minimum diameter of the neck portion is small, the adsorbent starts to boil (cavitation) as the relative pressure is lowered. At this time, the adsorbent adsorbed on the bottle and neck is desorbed and released at once. At this time, the hysteresis loop is closed. The relative pressure at which the adsorbate begins to boil is determined by the adsorbate and the measurement temperature, and does not depend on the minimum diameter of the neck portion. The relative pressure when the adsorbent starts to boil is about 0.4 when the adsorbate is nitrogen gas and the measurement temperature is 77K. The diameter of the neck portion at this time is about 3.4 nm. Therefore, when the minimum diameter of the neck portion is 3.4 nm or less, the hysteresis loop is closed by cavitation regardless of the minimum diameter of the neck portion.

The present inventor further verified this theory. As a result, the present inventor has found that when the minimum diameter of the neck portion is made larger than the diameter at which cavitation occurs, the hysteresis loop closes with a relative pressure corresponding to the minimum diameter of the neck portion. Such findings have not existed in the past, and the carbon materials obtained by such findings have never existed in the past. Therefore, when the minimum diameter of the neck portion is large (specifically, it is larger than the diameter at which cavitation occurs), the desorption isotherm falls at a relative pressure corresponding to the diameter of the neck portion, and corresponds to the minimum diameter of the neck portion. The relative pressure closes the hysteresis loop. That is, in this case, the hysteresis loop is closed not by the fall due to cavitation but by the fall at the relative pressure corresponding to the diameter of the neck portion. Therefore, the higher the relative closing pressure of the hysteresis loop, the larger the minimum diameter of the neck portion.

Furthermore, when the distribution of the diameter of the neck portion is narrow, the desorption isotherm drops sharply at a specific relative pressure (that is, the slope of the desorption isotherm is large). On the other hand, when the diameter distribution of the neck portion is wide, the desorption isotherm is observed to fall in a wide range of relative pressure, so that the desorption isotherm falls gently. That is, the slope of the desorption isotherm becomes smaller. Therefore, as an index of the distribution of the minimum diameter of the neck portion and the diameter of the neck portion, the relative pressure at which the hysteresis loop is closed in the nitrogen adsorption / desorption isotherm and the slope of the desorption isotherm can be used.

Here, a concept in the case where a plurality of bottleneck type pores X1 to X3 are connected in the carbon material 50 will be described with reference to FIG. 7. The bottleneck type pore X1 is composed of a bottle portion B1 and neck portions N1 and N1'. The neck portions N1 and N1'connect the bottle portion B1 and the outside of the carbon material 50. The bottleneck type pore X2 is formed inside the carbon material 50 more than the bottleneck type pore X1, and is composed of a bottle portion B2 and a neck portion N2. The neck portion N2 connects the bottle portion B2 and the bottle portion B1. The bottleneck type pore X3 is formed inside the carbon material 50 more than the bottleneck type pore X2, and is composed of a bottle portion B3 and a neck portion N3. The neck portion N3 connects the bottle portion B3 and the bottle portion B2. The diameter d1 of the neck portion N1 is larger than the diameter d1'of the neck portion N1'. The diameter d2 of the neck portion N2 is larger than the diameter d1 of the neck portion N1. The diameter d3 of the neck portion N3 is smaller than the diameter d1 of the neck portion N1.

In this case, the adsorbent (that is, nitrogen molecule) adsorbed on the bottle portion B1 starts desorption at a relative pressure corresponding to the diameter d1 of the neck portion N1 having the largest diameter among the neck portions N1 and N1'. Therefore, the desorption of the adsorbent adsorbed on the bottle portion B1 is observed as the fall of the desorption isotherm at the relative pressure corresponding to the diameter d1.

The relative pressure when the adsorbent adsorbed on the bottle portion B2 is desorbed is the neck portion having the largest diameter among the diameter d2 of the neck portion N2 and the neck portions N1 and N1'directly connected to the outside of the carbon material 50. It is determined by comparison with the diameter d1 of N1. In the example of FIG. 7, the diameter d2 of the neck portion N2 is equal to or larger than the diameter d1 of the neck portion N1. In this case, the adsorbent adsorbed on the bottle portion B2 starts desorption at a relative pressure corresponding to the diameter d1 of the neck portion N1. Therefore, the desorption of the adsorbent adsorbed on the bottle portion B2 is observed as the fall of the desorption isotherm at the relative pressure corresponding to the diameter d1. On the other hand, when the diameter d2 of the neck portion N2 is smaller than the diameter d1 of the neck portion N1, the adsorbent adsorbed on the bottle portion B2 starts desorption at a relative pressure corresponding to the diameter d2 of the neck portion N2. Therefore, the desorption of the adsorbent adsorbed on the bottle portion B2 is observed as the fall of the desorption isotherm at the relative pressure corresponding to the diameter d2.

The relative pressure when the adsorbent adsorbed on the bottle portion B3 is desorbed is the neck portion having the largest diameter among the diameter d3 of the neck portion N3 and the neck portions N1 and N1'directly connected to the outside of the carbon material 50. It is determined by comparison with the diameter d1 of N1. In the example of FIG. 7, the diameter d3 of the neck portion N3 is smaller than the diameter d1 of the neck portion N1. In this case, the adsorbent adsorbed on the bottle portion B3 starts desorption at a relative pressure corresponding to the diameter d3 of the neck portion N3. Therefore, the desorption of the adsorbent adsorbed on the bottle portion B3 is observed as the fall of the desorption isotherm at the relative pressure corresponding to the diameter d3. On the other hand, when the diameter d3 of the neck portion N3 is equal to or larger than the diameter d1 of the neck portion N1, the adsorbent adsorbed on the bottle portion B3 starts desorption at a relative pressure corresponding to the diameter d1 of the neck portion N1. Therefore, the desorption of the adsorbent adsorbed on the bottle portion B3 is observed as the fall of the desorption isotherm at the relative pressure corresponding to the diameter d1.

There are various modes of bottleneck type pores formed in the carbon material other than the example shown in FIG. 7, but all the modes can be described by the same concept as described above. A plurality of bottleneck-type pores X _{n connected to the bottleneck-type pores X (n-1)} that have already been desorbed from the condensed adsorbent on the outside or in the bottle portion B _(n-1) are a plurality of neck portions. When N _n is provided, the adsorbent adsorbed on the _{bottle portion B n} is desorbed at a relative pressure corresponding to the diameter d _{na of the neck portion N na} having the largest diameter among the _{neck portions N n.} At this time, there is a bottle part _{B n} in concatenated next bottleneck type pore _{X (n + 1),} having a plurality of neck _{N (n + 1)} for connecting the bottle part _B and _{(n + 1)} of the bottle portion _{B n} When the diameter of the neck portion N _{(n + 1)} a, which is the largest of the neck portions N _{(n + 1) , is d (n + 1) a,} and the diameter d _{(n + 1) a} is larger than the diameter d _na , the bottle portion B ₍ The adsorbent adsorbed on _{n + 1)} is desorbed at a relative pressure corresponding to the _{diameter d na.} That is, _{the adsorbent adsorbed on the bottle portion B n} and the bottle portion B _{(n + 1)} is desorbed at the same relative pressure. When the diameter d _{(n + 1) a} is smaller than the diameter d _na , the adsorbent adsorbed on the _{bottle portion B (n + 1)} is desorbed at a relative pressure corresponding to the _{diameter d (n + 1) a.} Moreover, until the bottle-neck pore _{X n} and linked x-th bottleneck type pore _{X (n + x),} the maximum diameter _{d (n} + 1) of the neck of each bottle neck mold pores _{a, d (n + 2)} a When d _{(n + x) a} is all larger than _{d na and d (n + x + 1) a} is _{smaller than d na} , the bottle portions B _n , B _{(n + 1)} , B _{(n + 2)} ... B _{(n + x)} All the adsorbents adsorbed on the surface are desorbed at a relative pressure corresponding to the _{diameter d na.} In FIG. 7, for convenience of drawing, all the neck portions have a length and a volume, but in the actual template carbon, the length and volume of the neck portion are substantially 0. I can think. That is, the pores formed inside the template carbon vary in diameter (diameter) along the length direction. Therefore, the pores can be considered to be a combination of a bottle portion having a certain length and volume and a bottleneck type pore composed of a neck portion having a length of 0 (that is, a volume of 0).

In addition, even in the conventional technique, a nitrogen adsorption isotherm was used. However, the nitrogen adsorption isotherm was only used to adjust the average pore diameter of the carbon material or the peak position (mode diameter) of the pore diameter distribution. That is, there was no finding to use the relative pressure at which the hysteresis loop closes in the nitrogen adsorption / desorption isotherm and the slope of the nitrogen desorption isotherm.

<Knowledge 2>
We have found MgO activation treatment as a technique for increasing the minimum diameter of the neck portion. That is, the carbon material according to the present invention is generally produced by the following steps. That is, the mixture of the MgO mold and the carbon source is calcined (first heating step). As a result, a part of the carbon source (specifically, a highly volatile component) is dispersed in the MgO mold while being decomposed, and adheres to the surface of the crystallite. Then, the carbon source attached to the surface of the crystallite is carbonized. FIG. 1 schematically shows an MgO-carbon complex in which a carbon layer 20 is attached to the surface of the crystallite 10a. In FIG. 1, MgO template particles 10 are formed by a plurality of crystallites 10a. Then, a carbon layer 20 made of a carbonized carbon source is formed on the surface of the crystallite 10a. Then, the MgO-carbon composite is heat-treated at a high temperature (second heating step). This heat treatment is the MgO activation treatment. As shown in FIG. 2, the carbon layer 20 is thinned by this MgO activation treatment. Then, by removing the MgO mold, as shown in FIG. 3, the carbon material 50 according to the present embodiment is produced. In the carbon material 50, the portion where the crystallites 10a are present is mainly the bottle portion 20a or other types of pores, and the portion where the crystallites are connected to each other is the neck portion 20b. Since the carbon layer 20 is thinned, the diameters of the bottle portion 20a and the neck portion 20b are increased. That is, the minimum diameter of the neck portion 20b is expanded.

Here, the present inventor considers the reason why the carbon layer 20 is thinned by the MgO activation treatment as follows.
(1) Carbon adhering to the crystallite 10a reacts with MgO to _{generate CO 2} gas.
(2) The generated CO ₂ gas oxidizes (that is, activates) the inner wall of the carbon layer 20 (CO ₂ + C → 2CO).
(3) The carbon layer 20 is thinned by the synergistic effect of the generation of _{CO 2 gas and the activation of the inner wall by the generated CO 2 gas.}

Although details will be described later, the specific heat treatment temperature for obtaining the above-mentioned MgO activation effect is 1100 ° C to 1400 ° C. The MgO activation effect can be obtained for the first time by heat-treating the MgO-carbon composite in the temperature range of 1100 ° C to 1400 ° C.

The MgO activation effect according to this embodiment is presumed to be a phenomenon completely different from the "carbon porosification effect" described in Patent Document 1 and Patent Document 2. The reason is as follows.
(1) The examples described in Patent Document 1 and Patent Document 2 are 900 ° C., which are examples in a temperature range in which the “MgO activation effect” cannot be obtained. The "MgO activation effect" of the present invention has not been implemented.
(2) Further, in Document 3 (T. Morishita et al., CARBON, 48 (2010) 2690-2707) including the inventors of Patent Document 1 and Patent Document 2, carbon similar to that of Patent Document 1 and Patent Document 2 is reported. doing. On page 2702, it is described that the "carbonization effect of carbon" is the effect of micropores generated by shrinkage when PVA is carbonized on the surface of MgO, and the "MgO activation effect" of the present invention. The "pore enlargement effect" and the "neck diameter enlargement effect" are different. As described above, it is the first finding by the present inventor that the minimum diameter of the neck portion is increased and the diameter of the bottle portion is increased due to the MgO activation effect.

In addition, as a conventional activation treatment, a gas activation treatment can be mentioned. However, in the gas activation treatment, since the outside of the pores and the inlet of the pores are preferentially activated, it is difficult to activate the bottle portion and the neck portion. Furthermore, when the present inventor tried the gas activation treatment, it was found that the distribution of the diameter of the neck portion tended to widen.

<Knowledge 3>
It has been found that in order to narrow the distribution of the diameter of the neck portion, it is an effective means to use an MgO mold having a uniform crystallite diameter. In FIGS. 1 to 3, the crystallite diameters are uniform (uniform). In this case, as is clear from FIGS. 1 to 3, the diameters of the pores (mesopores) formed by the carbon layer 20 are uniform before the MgO activation treatment. Further, since the diameters of the connecting portions of the crystallites 10a are also the same, the diameters of the pores (corresponding to the neck portion) formed by the carbon layer 20 covering this portion are also the same. Then, by subjecting the carbon layer 20 to the MgO activation treatment, the diameters of the bottle portions 20a are aligned and the diameters of the neck portions 20b are also aligned. That is, the distribution of the diameter of the neck portion 20b becomes narrow. When MgO molds having irregular crystallite diameters are used, the diameters of the pores (mesopores) formed by the carbon layer 20 are not uniform before the MgO activation treatment. Further, since the diameter of the connecting portion between the crystallites 10a is also uneven, the diameter of the pores (corresponding to the neck portion) formed by the carbon layer 20 covering this portion is also uneven. When such a carbon layer 20 is subjected to MgO activation treatment, the inner wall of the carbon layer 20 is non-uniformly activated, so that not only the diameter of the bottle portion 20a becomes uneven, but also the diameter of the neck portion 20b becomes uneven. Become.

As a result of combining the above findings, the present inventor has succeeded in newly synthesizing a carbon material having a large minimum diameter of the neck portion and a narrow distribution of the diameter of the neck portion. The method for producing template carbon adopted in the present invention is different from the conventional method in that MgO having the same crystallite diameter is used for the template and MgO activation treatment for enlarging the neck portion and the bottle portion is performed. .. When the power generation performance of the fuel cell using the newly synthesized carbon as a carrier was investigated, the catalyst utilization rate was improved to a extent that could not be obtained with the conventional MgO template carbon.

Specifically, according to a certain viewpoint of the present invention, the relative pressure (P / P ₀ ) c at which the hysteresis loop formed by the nitrogen adsorption / desorption isotherm closes is more than 0.55 and less than 0.8, and ( _{P / P 0) c + 0.1~} (P / P 0) of nitrogen desorption amount between c dV is 65cm ³ / g greater, less than 200 cm ³ / g, BET specific surface area of 700 meters ² / g, greater than 2000 m ² / Provided is a carbon material for a fuel cell catalyst carrier, characterized in that it is g or less.

Here, the half-value width △ G of G band derived from the Raman spectrum is 35 cm ^-1 greater, may be less than 70cm ^-1.

Further, BET specific surface area of 900 meters ² / g greater, may be less than 1500 m ² / g.

Further, the average particle size of the carbon material may be more than 200 nm and less than 1000 nm.

According to another aspect of the present invention, there is provided a fuel cell catalyst layer comprising the above-mentioned fuel cell catalyst carrier template carbon material.

According to another aspect of the present invention, there is provided a fuel cell comprising the above-mentioned catalyst layer for a fuel cell.

Here, the catalyst layer for the fuel cell may be the catalyst layer on the cathode side.

As described above, according to the present invention, since the relative pressure (P / P ₀ ) c at which the hysteresis loop closes is more than 0.55 and less than 0.8, the minimum diameter of the neck portion becomes sufficiently large. Therefore, flooding in the bottleneck type pores is suppressed during power generation in a highly humidified environment. _{Further, (P / P 0) c} + 0.1~ (P / P 0) of nitrogen desorption amount dV between c is 65cm ³ / g greater, since it is less than 200 cm ³ / g, is sufficiently narrow distribution of the diameter of the neck portion Become. As a result, the oxidizing gas is evenly supplied to each bottleneck type pore, so that the catalyst component in each bottleneck type pore, particularly in the bottle portion, can be used evenly. Therefore, since the catalyst utilization rate is improved, the power generation performance of the fuel cell, particularly the power generation performance at the time of power generation in a highly humidified environment, is improved.

It is explanatory drawing which shows typically the structure of the MgO-carbon complex in which the surface of the crystallite of the MgO mold is coated with the carbon layer. It is explanatory drawing which shows typically the structure of the MgO-carbon complex after the MgO activation treatment. It is explanatory drawing which shows typically an example of the carbon material which concerns on this invention. It is a graph which shows an example of a hysteresis loop. It is explanatory drawing which shows an example of the bottleneck type pore typically. It is a schematic diagram which shows the schematic structure of the fuel cell which concerns on embodiment of this invention. It is explanatory drawing which shows typically the example which a plurality of bottleneck type pores are connected.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the present specification and the drawings, components having substantially the same functional configuration are designated by the same reference numerals, so that duplicate description will be omitted. The diameter of the pores in this embodiment is an index of the size of the pores. The pores are not necessarily cylindrical, but the diameter of the pores in this embodiment is the diameter when the pores are assumed to be cylindrical. Gas adsorption is preferably used to measure the diameter of such pores. When the nitrogen adsorption isotherm obtained by using nitrogen gas is analyzed by the BJH method, the diameter when the pores are assumed to be cylindrical can be calculated. Further, the particle diameter (that is, the particle diameter) is an index of the particle size. The particles are not necessarily spheres, but the particle diameter in this embodiment is the diameter when the particles are assumed to be spheres. For the measurement of the particle size of the present invention, it is preferable to use the particle size calculated by the disk type particle size distribution measurement. In the case of the disk type particle size distribution measurement, the particle size is calculated as the diameter when the particles are assumed to be spheres.

<1. Composition of carbon material ＞
First, the composition of the carbon material according to the present embodiment will be described. The carbon material according to this embodiment is suitably used as a catalyst carrier for a fuel cell. The carbon material according to the present embodiment has a large number of bottleneck type pores, but the minimum diameter of the neck portion is large and the distribution of the diameter of the neck portion is narrow. Hereinafter, the composition of the carbon material will be described in detail.

(1-1. Minimum diameter of neck part)
_{The closing relative pressure (P / P 0} ) c of the hysteresis loop in the nitrogen adsorption / desorption isotherm is more than 0.55 and less than 0.8. As described above, the closing relative pressure (P / P ₀ ) c of the hysteresis loop corresponds to the minimum diameter of the neck portion. That is, in the present embodiment, the minimum diameter of the neck portion is larger than the diameter when cavitation occurs. Therefore, the relative pressure (P / P ₀ ) c at which the hysteresis loop closes depends on the minimum diameter of the neck portion. When the relative pressure (P / P ₀ ) c is in the range of more than 0.55 and less than 0.8, the minimum diameter of the neck portion is sufficiently large and flooding is suppressed. In other words, it has excellent gas diffusivity.

On the other hand, when the closing relative pressure (P / P ₀ ) c of the hysteresis loop is 0.55 or less, many neck portions having a small diameter are present, and flooding is likely to occur in such neck portions. Therefore, the diffusivity of the gas is low. Further, in order to obtain a carbon material having a relative pressure (P / P ₀ ) c for closing the hysteresis loop of 0.8 or more, excessive MgO activation is often required. Then, excessive MgO activation increases the possibility that a new neck portion is formed on the inner wall of the mesopore. Therefore, it causes an increase in the non-uniformity of the diameter of the neck portion.

(1-2. Distribution of neck diameter)
In this embodiment, the slope of the nitrogen desorption isotherm immediately before the hysteresis loop closes is adopted as an index of the uniformity of the diameter distribution of the neck portion. Specifically, the amount of nitrogen desorption dV between _{(P / P 0} ) c + 0.1 to (P / P ₀ ) c, which corresponds to the slope of the nitrogen desorption isotherm just before the hysteresis loop closes, ^{exceeds 65 m 3} / g. Is. When the amount of nitrogen desorption dV exceeds 65 cm ³ / g, the distribution of the diameter of the neck portion becomes sufficiently narrow. As a result, the gas diffusibility into the bottleneck type pores becomes uniform, and many catalytic components existing inside the bottleneck type pores can contribute to the power generation reaction.

When the amount of nitrogen desorption dV is 65 cm ³ / g or less, the distribution of the diameter of the neck portion spreads widely, and the gas diffusion into the bottleneck type pores becomes non-uniform. As a result, for example, the catalyst component in the bottleneck type pores to which the oxidizing gas is not sufficiently supplied can hardly contribute to the power generation reaction. Therefore, the catalyst utilization rate is reduced.

The upper limit of the amount of nitrogen desorption dV is not particularly limited, but since a carbon material production method described later ^{can achieve about 200 cm 3} / g, this value is a substantial upper limit.

(1-3. BET specific surface area)
BET specific surface area of carbon materials according to the present embodiment is a 700 meters ² / g Ultra 2000m less than ² / g. If the BET specific surface area is ^{more than 700 m 2} / g, the catalyst component can be supported as fine particles having a diameter of about 2 to 4 nm in a highly dispersed state. Here, if the BET specific surface area is 700 m ² / g or less, the catalyst component may not be supported as fine particles in a highly dispersed manner. The upper limit of the BET specific surface area is not limited, but it is an upper limit that can realize about ^{2000 m 2 / g, which is compatible with graphite property for obtaining physical strength and oxidative consumption durability.}

BET specific surface area is preferably, 900 meters ² / g Ultra 1500m less than ² / g. In this case, the catalyst component can be supported in a more highly dispersible state. The BET specific surface area can be obtained by analyzing the nitrogen adsorption isotherm by the BET method.

(1-4. Graphite)
The carbon material according to the present embodiment preferably has a structure having high graphitability (in other words, high crystallinity) as a fuel cell from the viewpoint of oxidation consumption resistance. Specifically, it is preferable half width △ G of G band derived from the Raman spectrum is less than 35 cm ^-1 super 70cm ^-1. Here, the G band means a peak in the range of ^{1500 to 1700 cm -1.} If the half-value width ΔG of the G band is 35 cm ^-1 or less, the crystallinity may develop too much and the specific surface area of ^{more than 700 m 2} / g may not be compatible. On the other hand, if the half width ΔG of the G band is 70 cm ^-1 or more, the oxidation wear resistance is low, and it may be difficult to secure the durability.

Half width △ G is more preferably less than 40 cm ^-1 super 65cm ^-1. In this case, it is possible to obtain an excellent carrier carbon for a fuel cell in which the electron conductivity is improved and the catalyst utilization rate and the durability are highly compatible.

(1-5. Average particle size)
The average particle size of the carbon material is preferably more than 200 nm and less than 1000 nm. In this case, when the carbon material is used as the fuel cell catalyst carrier in the catalyst layer on the cathode side, particle gaps suitable for gas diffusion can be formed in the catalyst layer.

When the average particle size is 200 nm or less, when a carbon material is used as a fuel cell catalyst carrier in the catalyst layer, the interparticle voids become small and the gas diffusibility may decrease. Further, when the average particle size is 1000 nm or more, the conduction path between the particles per unit volume of the catalyst layer is reduced, and the electron conduction resistance of the catalyst layer may be increased. In addition, the gas diffusion path into the particles becomes too long, and as a result, the catalyst utilization rate may decrease. Here, the average pore size of the carbon material is the median size of the particle size distribution of the carbon material, and the particle size of the carbon material can be measured by, for example, a disk-type particle size distribution measurement.

<2. Manufacturing method of carbon material ＞
Next, a method for producing a carbon material will be described. The method for producing the carbon material is preferably an MgO mold selection step, a carbon source selection step, a carbon source and MgO mold mixing step, a first heating step, a second heating step, an MgO removing step, and a third. Including the heating step of. Hereinafter, each step will be described in detail.

(2-1. MgO mold selection process)
In this step, the MgO mold is selected. Here, the minimum constituent particles of the MgO mold are crystallites as described above. Agglomeration of a plurality of crystallites forms a mass of MgO template particles. When the MgO template is observed with an SEM (scanning electron microscope), it can be seen that the crystallites are connected everywhere to form a mass of MgO template particles.

The particle size of the MgO template particles and the particle size of the carbon material, which is the final product, are almost the same. Further, the MgO-carbon composite is formed by the first heating step described later. The diameter of the pores formed by the carbon layer in the MgO-carbon composite (so-called pore diameter) is substantially the same as the diameter of the crystallites (so-called crystallite diameter).

The above reasons are as follows. That is, when the mixture of the MgO mold and the carbon source is fired in the first heating step, the carbon source is dispersed in the MgO mold while being partially decomposed and adheres to the surface of the crystallite. Then, the carbon source attached to the surface of the crystallite is carbonized. As a result, an MgO-carbon composite is formed (see FIGS. 1 to 3). Therefore, the MgO-carbon complex has a particle size almost the same as that of the MgO template particles. Since the carbon material as the final product is obtained by removing the MgO template from the MgO-carbon composite, the particle size of the carbon material is almost the same as the particle size of the MgO-carbon composite. Therefore, the particle size of the MgO template particles and the particle size of the carbon material which is the final product are almost the same. Further, since the carbon layer is attached to the surface of the crystallite, the diameter of the pores formed by the carbon layer is substantially the same as the diameter of the crystallite.

The crystallite diameter is preferably more than 5 nm and less than 50 nm. The crystallite diameter is measured by, for example, XRD (X-ray diffraction). When the crystallite diameter is within this range, the pores are large enough for the carbon material (that is, the pores large enough to diffuse the oxidizing gas required for the power generation reaction into the carbon material). ) Can be formed. On the other hand, when the crystallite diameter is 5 nm or less, the pores in the carbon material which is the final product become small, and the gas diffusibility may decrease. Further, when the crystallite diameter is 50 nm or more, the pores in the carbon material become large, but there is a possibility that both high specific surface area and physical strength cannot be achieved at the same time.

Further, it is preferable that the crystallite diameters are the same. This is because, as described above, the distribution of the diameter of the neck portion can be narrowed by making the crystallite diameters uniform. Here, the uniformity of the particle size can be evaluated based on the peak of the pore size distribution of the MgO template. Specifically, the nitrogen adsorption isotherm of the MgO template is analyzed by the BJH method to obtain a pore size distribution curve. Next, in the pore size distribution curve, a peak of the pore volume was observed at a diameter of 1 to 50 nm, and the peak of the pore volume of the MgO template had a 3/4 value width R _3/4 and a half width R _1/2 . If the ratio (R _3/4 ) / (R _1/2 ) is less than 0.7, the peak is evaluated as sharp. In this case, it can be evaluated that the crystallite diameter is uniform. On the other hand, if (R _3/4 ) / (R _1/2 ) is 0.7 or more, it can be evaluated that the peak is broad and the crystallite diameter is non-uniform.

Further, it is preferable that the average particle size of the MgO template particles is more than 200 nm and less than 1000 nm. In this case, a carbon material having an average particle size of more than 200 nm and less than 1000 nm (that is, a preferable average pore size) can be produced. Here, the average particle size of the MgO template particles is the median size of the particle size distribution of the MgO template particles, and the particle size of the MgO template particles can be measured by, for example, a disk-type particle size distribution measurement.

It is preferable that the particle size distribution of the MgO template particles is narrow. Specifically, the ratio of the peak 3/4 width to the half width in the particle size distribution "3/4 width of the particle diameter / half width of the particle diameter" is preferably less than 0.7. In this case, a carbon material having a narrower particle size distribution can be produced, and the gas diffusibility of the carbon material is improved.

As the MgO mold, a commercially available MgO may be used. If the commercially available MgO satisfies the above-mentioned requirements, the commercially available MgO may be used as it is as the MgO mold. When the average pore size of commercially available MgO is large, MgO having a large particle size may be pulverized to use the average particle size as a value within the above-mentioned range as an MgO template. In this case, the crushing method is not particularly limited, and examples thereof include crushing with a ball mill or a jet mill. After pulverization, it is preferable to classify the particles to have the same particle size.

Further, the MgO used in the MgO mold may be a synthesized one. In this case, it is preferable to synthesize MgO so as to satisfy the above-mentioned requirements. That is, the synthesis method is preferably a method capable of controlling at least the crystallite diameter, and further preferably a method capable of controlling both the crystallite diameter and the particle size of the MgO template particles. Specific examples of the synthesis method include a method of heat-decomposing magnesium hydroxide and magnesium carbonate, and a method of heat-treating the Mg complex in an oxidizing atmosphere. In these methods, the crystallite diameter of MgO can be adjusted by the heat treatment temperature. Further, the particle size of MgO whose crystallite diameter has been adjusted may be adjusted by the above-mentioned pulverization method.

The main component of the component constituting the MgO mold is MgO. Specifically, the MgO mold preferably contains MgO in a proportion of 95% by mass or more. When the content of the MgO component is less than 95% by mass, impurities other than the MgO component increase. Then, such impurities may remain in the particles of the carbon material. When a carbon material is used as a fuel cell catalyst carrier, the electron conduction resistance may increase due to impurities. When the content of the MgO component is 95% by mass or more, such a problem is less likely to occur, so that a high-performance carbon material can be obtained.

(2-2. Carbon source selection process)
In this process, the carbon source is selected. The carbon source is not particularly limited, and may be, for example, a carbon source used in the conventional molding method. For example, the carbon source may be various organic substances. Specific examples of carbon sources include polyvinyl alcohol, aliphatic or aromatic polyester resins, polyolefin resins, acrylic resins, styrene resins, polyamide resins, polyacrylonitrile resins, and polybutadienes. Examples thereof include elastomers mainly composed of polyisoprene, natural rubbers, petroleum resins, phenolic resins, furan resins, epoxy resins, alkyd resins, polyimides and the like. Among these examples, the carbon source is preferably one that is substantially composed of carbon, hydrogen, and oxygen, that is, polyvinyl alcohol, polyester resin, styrene resin, petroleum resin, and the like. These carbon precursors may have any shape such as powder, pellets, and lumps, and may be dissolved or dispersed in an organic solvent. Further, although the examples given above are organic substances that are solid at room temperature, the carbon source may be organic substances that are liquid at room temperature. Examples of such examples include furfuryl alcohol, acrylonitrile, vinyl acetate and the like. Of the resins listed above, the carbon source is more preferably a thermoplastic resin. Examples of the thermoplastic resin include the above-mentioned polyvinyl alcohol and the like. When a thermoplastic resin is used as the carbon source, the carbon source is melted during the first heating step, so that the surface of the crystallite can be easily uniformly coated with the carbon source. As a result, a carbon layer is uniformly formed on the surface of the crystallite, so that a more homogeneous carbon material can be obtained.

(2-3. Mixing process of carbon source and MgO mold)
In this step, the carbon source and the MgO mold are mixed. From the viewpoint of homogenizing the carbon material, it is preferable that the carbon source and the MgO template are uniformly mixed (that is, so that the bias of the components is as small as possible). The mixing method is preferably a method capable of uniformly mixing the carbon source and the MgO template, but is not particularly limited. Examples of mixing methods include mixing with a dry or wet ball mill. According to this mixing method, the carbon source and the MgO mold can be uniformly mixed.

Here, when the carbon source and the MgO template are mixed in a wet manner, it is preferable that the solvent does not dissolve the MgO template. Examples of such a solvent include alcohols such as ethanol. Further, after the carbon source and the MgO mold are wet-mixed, the mixture is sufficiently dried, and then the next step is performed.

Here, the mixing ratio of the carbon source and the MgO template is preferably such that the mass ratio C / MgO of the carbon content C and the MgO component in the carbon source is more than 0.1 and less than 5. When the mass ratio C / MgO is a value within this range, the MgO mold can be sufficiently coated with the carbon source, so that the shape of the crystallites of the MgO mold is reflected, that is, the diameters of the neck portions are uniform. Moreover, a carbon material having a high specific surface area can be obtained.

On the other hand, if the mass ratio C / MgO is 0.1 or less, the carbon source may not be able to completely cover the MgO mold. In this case, pores reflecting the crystallite size of the MgO template cannot be obtained, and as a result, the distribution of the diameter of the neck portion may become large. When the mass ratio C / MgO is 5 or more, the carbon source coated on the MgO mold becomes excessively thick, so that the carbon layer also becomes thick. As a result, the specific surface area of the carbon material may be low.

(2-4. First heating step)
In the first heating step, the crystallites of the MgO mold are coated with a carbon source and the carbon source is carbonized. Specifically, in the first heating step, the mixture of the carbon source and the MgO mold is fired in an inert atmosphere. As a result, a part of the carbon source (specifically, a highly volatile component) is dispersed in the MgO mold while being decomposed, and adheres to the surface of the crystallite. Then, the carbon source (specifically, a component having low volatility) adhering to the surface of the crystallite is carbonized. As a result, an MgO-carbon composite is produced.

The carbon content in the carbonaceous material in the MgO-carbon composite is preferably 70% by mass or more with respect to the total mass of the carbonaceous material. The carbonaceous substance is a component other than MgO in the MgO-carbon composite, and the carbonaceous substance may contain hydrogen, oxygen, nitrogen, a boron atom and the like in addition to the carbon content. The carbon content in the carbonaceous substance can be quantitatively measured by organic element analysis such as a combustion infrared absorption method for the obtained carbonaceous substance after dissolving and removing MgO by acid treatment of the MgO-carbon composite. When the carbon content is 70% by mass or more, the target carbon material of the present embodiment can be easily produced. On the other hand, when the carbon content is less than 70% by mass, the MgO-carbon complex tends to contain a large amount of highly volatile carbon. Therefore, when the MgO-carbon composite is heat-treated at a higher temperature in the next step (second heating step), highly volatile carbon content can be volatilized to form extra pores in the carbon layer. There is sex. Such pores can be a neck. Therefore, the distribution of the diameter of the neck portion may be widened.

In the first heating step, specifically, it is preferable that the mixture of the carbon source and the MgO mold is fired (heated) at 500 to 1000 ° C. for 1 hour or more in a heating furnace under an inert atmosphere. In this case, the carbon content in the carbonaceous substance in the MgO-carbon complex can be 70% by mass or more. The rate of temperature rise is not limited, but may be about 10 to 30 ° C./min. If the heating temperature is less than 500 ° C., carbonization may not proceed sufficiently. As a result, the carbon content in the MgO-carbon complex may be less than 70% by mass. Further, when the heating temperature exceeds 1000 ° C., the decomposition products of the carbon source are violently volatilized. Then, a part of the carbon layer may be oxidized by the volatilized decomposition product. As a result, extra pores may be formed in the carbon layer. Such pores can be a neck. Therefore, the distribution of the diameter of the neck portion may be widened.

(2-5. Second heating step)
In the second heating step, the MgO-carbon composite is heat-treated at a high temperature. That is, the above-mentioned "MgO activation treatment" is performed. By this MgO activation treatment, the carbon layer in the MgO-carbon complex is thinned. As a result, the bottle portion and the neck portion of the bottleneck type pores are enlarged. That is, the minimum diameter of the neck portion becomes large, and the diameter of the bottle portion becomes large.

Specifically, the carbon-MgO composite obtained in the first heating step is heat-treated at a temperature of 1100 to 1400 ° C. in a heating furnace under an inert atmosphere. When the heating temperature is less than 1100 ° C., the MgO activation reaction does not occur. If the heating temperature is higher than 1400 ° C., the carbon layer may be oxidized too much and a new small neck may be formed.

The heat treatment time of the second heating step is not limited, but is usually about 0.5 to 5 hours. When the second heating step is not performed, that is, when the MgO activation treatment is not performed, the diameter of the neck portion depends on the shape of the MgO mold. Therefore, the minimum diameter of the neck portion becomes small.

Further, even if another activation treatment, for example, a gas activation treatment is performed on the carbon material containing no MgO instead of the MgO activation treatment, the carbon material according to the present embodiment cannot be obtained. For example, when the carbon material obtained by acid-treating the MgO-carbon composite to dissolve and remove MgO is heated at 900 ° C. or higher in the flow of CO gas or steam, the outside of the MgO-carbon composite particles is oxidized. The neck part cannot be enlarged. Therefore, the gas activation treatment is not suitable for this embodiment.

In the second heating step, instead of the above-mentioned treatment, the MgO-carbon composite may be heat-treated at 500 to 900 ° C. in an atmosphere in which 5 to 20% by volume of the oxidizing gas is mixed. In this case, the oxidizing gas diffuses inside the MgO-carbon composite particles, and the neck portion can be expanded from the inside of the MgO-carbon composite particles due to the activation effect of MgO. That is, the carbon layer can be thinned. If the temperature is lower than 500 ° C., the thinning of the carbon layer does not start, and if the temperature exceeds 900 ° C., the outside of the MgO-carbon composite particles is oxidized and the distribution of the diameter of the neck portion becomes wide. Further, if the content of the oxidizing gas is less than 5% by volume, the thinning of the carbon layer does not start, and if it exceeds 20% by volume, the outside of the MgO-carbon composite particles is oxidized, and the neck portion The distribution of diameter becomes wide.

(2-6. MgO removal step)
In this step, the carbon-MgO composite is pickled to dissolve the MgO mold in the pickling solution. This removes the MgO template from the carbon-MgO complex. The MgO component remaining in the carbon material after the step is preferably 0.5% by mass or less with respect to the total mass of the carbon material. For the residual amount of MgO component in the carbon material, after burning the carbon material in an oxidizing atmosphere, the remaining ash is dissolved with an acid such as aqua regia, and the solution is subjected to a component analysis method such as ICP emission spectroscopy. By doing so, it can be quantified. When the residual amount of the MgO component remaining in the carbon material is more than 0.5% by mass, when the carbon material is applied to the fuel cell, the MgO dissolved during the power generation reaction reacts with the electrolyte resin, and the proton conductivity becomes high. As a result, the effect of lowering the fuel cell performance becomes greater. The acid used for pickling may be any soluble in MgO, and preferred examples include sulfuric acid. After pickling, the carbon material is washed with water and dried. By the above steps, the carbon material according to the present embodiment can be obtained. In this carbon material, the minimum diameter of the neck portion is large, and the distribution of the diameter of the neck portion is narrow. Specifically, the relative pressure (P / P ₀ ) c at which the hysteresis loop closes is more than 0.55 and less than 0.8, and is between (P / P ₀ ) c + 0.1 to (P / P ₀ ) c. The amount of nitrogen desorption dV is ^{more than 65 cm 3} / g. Therefore, when this carbon material is used for a fuel cell, flooding is less likely to occur at the neck portion. Further, the oxidizing gas is evenly supplied to each bottleneck type pore. Therefore, the gas diffusivity is improved and the power generation performance is improved. In particular, the power generation performance during power generation in a highly humidified environment is improved.

(2-7. Third heating step)
The carbon material at the stage of performing the MgO removing step also exhibits sufficient power generation performance, but it is preferable to carry out the third heating step in order to further enhance the durability of the carbon material. In the third heating step, by improving the crystallinity of the carbon material, the carbon material can be provided with the durability required as a catalyst carrier for a fuel cell.

Specifically, in the third heating step, the carbon material is heat-treated at 1500 to 2500 ° C. in a heating furnace in an inert atmosphere. Thereby, the crystallinity (in other words, the degree of graphitization) of the carbon material can be increased. Specifically, it becomes the half value width △ G of G band derived from the Raman spectrum is 35 cm ^-1, greater than less than 70cm ^-1. If the heating temperature is less than 1500 ° C., the crystallinity may not be sufficiently high and the oxidation / consumption resistance may be lowered. If the heating temperature exceeds 2500 ° C., the specific surface area may decrease and the dispersibility of the catalyst component may deteriorate.

By performing all the steps described above, it is possible to obtain a carbon material having high gas diffusibility and high durability during power generation in a highly humidified environment.

<3. Composition of polymer electrolyte fuel cell ＞
The carbon material for a catalyst carrier according to this embodiment can be applied to, for example, the polymer electrolyte fuel cell 100 shown in FIG. The polymer electrolyte fuel cell 100 includes separators 110 and 120, gas diffusion layers 130 and 140, catalyst layers 150 and 160, and an electrolyte membrane 170.

The separator 110 is a separator on the anode side, and a reducing gas such as hydrogen is introduced into the gas diffusion layer 130. The separator 120 is a separator on the cathode side, and introduces an oxidizing gas such as oxygen gas and air into the gas diffusion-aggregated phase. The types of the separators 110 and 120 are not particularly limited, and any separator used in a conventional fuel cell, for example, a polymer electrolyte fuel cell may be used.

The gas diffusion layer 130 is a gas diffusion layer on the anode side, and after diffusing the reducing gas supplied from the separator 110, it is supplied to the catalyst layer 150. The gas diffusion layer 140 is a gas diffusion layer on the cathode side, and after diffusing the oxidizing gas supplied from the separator 120, it is supplied to the catalyst layer 160. The types of the gas diffusion layers 130 and 40 are not particularly limited, and any gas diffusion layer used in a conventional fuel cell, for example, a polymer electrolyte fuel cell may be used. Examples of the gas diffusion layers 130 and 40 include porous carbon materials such as carbon cloth and carbon paper, and porous metal materials such as metal mesh and metal wool. As a preferable example of the gas diffusion layers 130 and 140, the layer on the separator side of the gas diffusion layer becomes a gas diffusion fiber layer containing a fibrous carbon material as a main component, and the layer on the catalyst layer side contains carbon black as a main component. An example is a gas diffusion layer having a two-layer structure as a micropore layer.

The catalyst layer 150 is a so-called anode. In the catalyst layer 150, an oxidation reaction of the reducing gas occurs to generate protons and electrons. For example, when the reducing gas becomes hydrogen gas, the following oxidation reaction occurs.
H ₂ → 2H ⁺ + 2e ⁻ (E ₀ = 0V)

Protons generated by the oxidation reaction reach the catalyst layer 160 through the catalyst layer 150 and the electrolyte membrane 170. The electrons generated by the oxidation reaction reach the external circuit through the catalyst layer 150, the gas diffusion layer 130, and the separator 110. The electrons are introduced into the separator 120 after working in an external circuit. After that, the electrons reach the catalyst layer 160 through the separator 120 and the gas diffusion layer 140.

The configuration of the catalyst layer 150 serving as the anode is not particularly limited. That is, the structure of the catalyst layer 150 may be the same as that of the conventional anode, the same as that of the catalyst layer 160, or the structure having higher hydrophilicity than that of the catalyst layer 160. You may.

The catalyst layer 160 is a so-called cathode. In the catalyst layer 160, a reduction reaction of the oxidizing gas occurs to generate water. For example, when the oxidizing gas becomes oxygen gas or air, the following reduction reaction occurs. The water generated in the oxidation reaction is discharged to the outside of the polymer electrolyte fuel cell 100 together with the unreacted oxidizing gas.
_{^{O 2 + 4H + + 4e -}} → 2H 2 O (E 0 = 1.23V)

As described above, in the polymer electrolyte fuel cell 100, power is generated by utilizing the energy difference (potential difference) between the oxidation reaction and the reduction reaction. In other words, the electrons generated by the oxidation reaction do the work in the external circuit.

The catalyst layer 160 preferably contains the carbon material for the catalyst carrier according to the present embodiment. That is, the catalyst layer 160 includes a carbon material for a catalyst carrier according to the present embodiment, an electrolyte material, and a catalyst component. As a result, the catalyst utilization rate in the catalyst layer 160 can be increased, and by extension, the catalyst utilization rate of the polymer electrolyte fuel cell 100 can be increased.

The catalyst loading ratio in the catalyst layer 160 is not particularly limited, but is preferably 30% by mass or more and less than 80% by mass. Here, the catalyst-supporting ratio is preferably the mass% of the catalyst component with respect to the total mass of the catalyst-supporting particles (particles in which the catalyst component is supported on the carbon material for the catalyst carrier). In this case, the catalyst utilization rate becomes even higher. When the catalyst loading ratio is less than 30% by mass, it may be necessary to thicken the catalyst layer 160 in order to make the polymer electrolyte fuel cell 100 practically usable. On the other hand, when the catalyst loading ratio is 80% by mass or more, catalyst agglutination is likely to occur. In addition, the catalyst layer 160 becomes too thin, which may cause flooding.

The mass ratio I / C of the mass I (g) of the electrolyte material and the mass C (g) of the carbon material for the catalyst carrier in the catalyst layer 160 is not particularly limited, but is preferably more than 0.5 and less than 5.0. .. In this case, the pore network and the electrolyte material network can be compatible with each other, and the catalyst utilization rate becomes high. On the other hand, when the mass ratio I / C is 0.5 or less, the electrolyte material network tends to be poor and the proton conduction resistance tends to be high. When the mass ratio I / C is 5.0 or more, the pore network may be disrupted by the electrolyte material. In either case, the catalyst utilization rate may decrease.

The thickness of the catalyst layer 160 is not particularly limited, but is preferably more than 5 μm and less than 20 μm. In this case, the oxidizing gas easily diffuses into the catalyst layer 160, and flooding is less likely to occur. When the thickness of the catalyst layer 160 is 5 μm or less, flooding is likely to occur. When the thickness of the catalyst layer 160 is 20 μm or more, the oxidizing gas is less likely to diffuse in the catalyst layer 160, and the catalyst component in the vicinity of the electrolyte membrane 170 is less likely to work. That is, the catalyst utilization rate may decrease.

The electrolyte membrane 170 is made of an electrolyte material having proton conductivity. The electrolyte membrane 170 introduces the protons generated in the oxidation reaction into the catalyst layer 160, which is a cathode. Here, the type of the electrolyte material is not particularly limited, and any electrolyte material used in a conventional fuel cell, for example, a polymer electrolyte fuel cell may be used. A good example is an electrolyte material used in polymer electrolyte fuel cells, i.e., an electrolyte resin. Examples of the electrolyte resin include polymers into which a phosphoric acid group, a sulfonic acid group, etc. have been introduced, for example, a perfluorosulfonic acid polymer, a polymer into which a benzenesulfonic acid has been introduced, and the like. Of course, the electrolyte material according to this embodiment may be another type of electrolyte material. Examples of such an electrolyte material include an inorganic material, an inorganic-organic hybrid type, and the like. The polymer electrolyte fuel cell 100 may be a fuel cell that operates in the range of room temperature to 150 ° C.

<4. Manufacturing method of polymer electrolyte fuel cell ＞
The manufacturing method of the polymer electrolyte fuel cell 100 is not particularly limited, and any manufacturing method similar to the conventional method may be used. However, it is preferable to use the carbon material for the catalyst carrier according to the present embodiment as the catalyst carrier on the cathode side.

<1. Evaluation method of carbon material ＞
Next, an embodiment of the present embodiment will be described. First, a method of measuring each parameter will be described.

(1-1. Measurement method of nitrogen adsorption isotherm and nitrogen desorption isotherm)
Nitrogen gas adsorption measurement was performed using BELSORPmini manufactured by Microtrac Bell, and nitrogen adsorption isotherms and nitrogen desorption isotherms were obtained. The measurement temperature was 77K. The nitrogen adsorption isotherm was measured in the range of relative pressure 0 to 0.995, and the nitrogen desorption isotherm was measured continuously from the nitrogen adsorption isotherm measurement. The nitrogen desorption isotherm was measured in a relative pressure range of 0.995 to 0.3.

(1-2. Method for measuring nitrogen desorption amount dV between relative pressure (P / P ₀ ) c and (P / P ₀ ) c + 0.1 to (P / P _{0) c for closing the hysteresis loop)}
In the process of measuring the nitrogen desorption isotherm by lowering the relative pressure from 0.995, the difference (dD) between the adsorption amount of the nitrogen desorption isotherm and the adsorption amount of the nitrogen adsorption isotherm ^{became 10 cm 3} / g or less for the first time. The relative pressure was defined as the relative pressure (P / P ₀ ) c at which the hysteresis loop was closed.

Further, the difference between the adsorption amount of the relative pressure (P / P ₀ ) c + 0.1 and the adsorption amount _{of the relative pressure (P / P 0) c in the nitrogen desorption isotherm is (P / P 0} ) c + 0.1 to (P / P). ₀ ) The amount of nitrogen desorption between c was defined as dV.

(1-3. Method for measuring BET specific surface area)
The nitrogen adsorption isotherm was analyzed by the BET method using the analysis software attached to BELSORPmini, and the BET specific surface area was calculated.

(1-4. Method for measuring average particle size of carbon material and MgO template particles)
The particle size distribution of the carbon material and the MgO template particles was measured using a disk-type particle size distribution measuring device (BI-DCP, manufactured by Brookhaven). Water was used as the solvent, and the rotation speed of the centrifuge was 3000 rpm. Then, the median diameter was measured by the dedicated analysis software of the apparatus, and this median diameter was used as the average particle size.

(1-5. Method for measuring crystallite diameter of MgO mold)
The crystallite diameter of the MgO mold was measured using an X-ray diffractometer (SmartLab, manufactured by Rigaku Co., Ltd.). Specifically, the measurement result of the X-ray diffractometer was converted into an X-ray diffraction pattern by analysis software, and the crystallite diameter of the MgO template was calculated from the diffraction peak near 42 degrees.

(1-6. Uniformity of MgO crystallite diameter distribution)
The distribution of the crystallite diameter of the MgO template was based on the result of analyzing the nitrogen adsorption isotherm at the liquid nitrogen temperature by the BJH method. Specifically, in the pore size distribution obtained by analyzing the nitrogen adsorption isotherm of the MgO template by the BJH method, the peak has a diameter of 1 to 50 nm, and the half width R _{1 / of the peak of the pore size distribution of MgO. When} the ratio (R _1/2 ) / (R _3/4 ) of 2 and the 3/4 width R _3/4 was less than 0.7, the peak was evaluated as sharp. In this case, it was evaluated that the crystallite diameter was uniform (◯). On the other hand, when (R _3/4 ) / (R _1/2 ) was 0.7 or more, it was evaluated that the peak was broad and the crystallite diameter was non-uniform (x).

(1-7. Measurement of Raman spectroscopic spectrum of carbon material)
Approximately 3 mg of carbon material is measured, and a laser Raman spectrophotometer (NRS-7100 type manufactured by JASCO Corporation) is used. Excitation laser 532 nm, laser power 100 mW (sample irradiation power: 0.1 mW), microscopic arrangement: Backscattering , Slit: 100 μm x 100 μm, Objective lens: x 100, Spot diameter: 1 μm, Exposure time: 30 sec, ^{Number of observed waves: 320 to 750 cm -1} , Accumulation frequency: Raman spectroscopic spectrum of carbon material was measured under the measurement conditions of 2 times. .. ^{Then, a peak in the range of 1500 to 1700 cm -1} called the G band was extracted from the Raman spectroscopic spectrum obtained in each measurement, and the half width (ΔG) of this peak was measured.

(1-8. Evaluation of residual amount of MgO component)
Any of the carbon materials related to Examples 1 to 9 and Comparative Examples 1 to 17 was selected. The selected carbon material was then dried in an oven at 85 ° C. for 4 hours. Then, 1 g of a dry carbon material was placed in a ceramic boat and burned in an electric furnace at 600 ° C. for 5 hours. After the ash remaining on the boat has cooled to room temperature, put the ash in a Teflon beaker, add 6 mL of aqua regia (concentrated nitric acid: concentrated hydrochloric acid = 3: 1), and decompose on a hot plate at 150 ° C for 2 hours. went. The decomposition solution was transferred to a 50 mL volumetric flask and diluted with water to the grade. A part of the obtained diluted solution was sprayed into argon plasma of an ICP emission spectrophotometer (CIROS-120 manufactured by AMETEK) to quantify MgO. The mass% of MgO in 1 g of the carbon material calculated by quantification is shown in Table 2 as the residual amount of the MgO component.

<2. Adjustment of carbon material ＞
(2-1. Preparation of MgO mold (preparation of carbon material. Common))
MgO-A to F shown in Table 1 were prepared as MgO molds. MgO-A to F were obtained from the distributors shown in Table 1 (manufactured by Konoshima Chemical Industries, Ltd., Kanto Denka Kogyo Co., Ltd., IoLiTec Co., Ltd., Tateho Chemical Industries Co., Ltd.). MgO-D to MgO-F were pulverized with a ball mill, and the particle size of the MgO template particles was adjusted to the size shown in Table 1. Table 1 shows the determination results of the particle size, crystallite size, (R _1/2 ) / (R _3/4 ), and crystallite size distribution uniformity of the MgO template particles.

(2-2. Carbon material not subjected to the third heating step (Examples 1 to 5, Comparative Examples 1 to 11)
In Examples 1 to 5 and Comparative Examples 1 to 11, the following steps were performed using the MgO mold shown in Table 2. As a result, the carbon materials according to Examples 1 to 5 and Comparative Examples 1 to 11 were produced.

(2-2-1. First heating step)
Polyvinyl alcohol powder (manufactured by Denki Kagaku Kogyo Co., Ltd., Denka Poval) was prepared as a carbon source. Then, the MgO mold and the carbon source were mixed at a mass ratio of 1: 2, and the mixture was placed in a quartz crucible. The crucible was inserted into an electric furnace, and argon gas was circulated in the electric furnace at 100 mL / min. Then, the temperature inside the electric furnace was raised at 10 ° C./min, and the temperature was maintained at the first heating temperature shown in Table 2 for 1 hour. After allowing to cool naturally, the MgO-carbon complex was taken out.

(2-2-2. Second heating step)
2-3 g of MgO-carbon composite was weighed on an alumina boat and set in a horizontal tubular electric furnace. Then, nitrogen gas was circulated in the electric furnace at 100 ml / min, and the temperature in the electric furnace was maintained at the second heating temperature shown in Table 2 for 1 hour.

(2-2-3. MgO removal step)
The MgO-carbon composite after the second heating step was put into the eggplant flask, and 1M dilute sulfuric acid having a mass ratio of 100 times the mass ratio of the MgO-carbon composite was put into the eggplant flask. Then, the mixed solution in the eggplant flask was stirred for 3 hours in an oil bath maintained at a temperature of 60 ° C. As a result, the MgO mold was eluted in the mixed solution. Then, the mixed solution was cooled to room temperature. The carbon material was then isolated by filtering the mixture. Then, the carbon material was washed with pure water, filtered, and dried. A carbon material was obtained by the above steps.

(2-3. Carbon material subjected to the third heating step (Examples 6 to 9, Comparative Examples 12 to 13))
In Examples 6 to 9 and Comparative Examples 12 to 13, the following steps were performed using the MgO mold shown in Table 2. As a result, the carbon materials according to Examples 6 to 9 and Comparative Examples 12 to 13 were produced.

(2-3-1. First heating step)
Polyvinyl alcohol powder (manufactured by Denki Kagaku Kogyo Co., Ltd., Denka Poval) was prepared as a carbon source. Then, the MgO mold and the carbon source were mixed at a mass ratio of 1: 2, and the mixture was placed in a quartz crucible. The crucible was inserted into an electric furnace, and argon gas was circulated in the electric furnace at 100 mL / min. Then, the temperature inside the electric furnace was raised at 10 ° C./min, and the temperature was maintained at 800 ° C. for 1 hour. After allowing to cool naturally, the MgO-carbon complex was taken out.

(2-3-2. Second heating step)
2-3 g of MgO-carbon composite was weighed on an alumina boat and set in a horizontal tubular electric furnace. Then, nitrogen gas was circulated in the electric furnace at 100 ml / min, and the temperature in the electric furnace was maintained at the second heating temperature shown in Table 2 for 1 hour.

(2-3-3. MgO removal step)
The MgO-carbon composite after the second heating step was put into the eggplant flask, and 1M dilute sulfuric acid having a mass ratio of 100 times the mass ratio of the MgO-carbon composite was put into the eggplant flask. Then, the mixed solution in the eggplant flask was stirred for 3 hours in an oil bath maintained at a temperature of 60 ° C. As a result, the MgO mold was eluted in the mixed solution. Then, the mixed solution was cooled to room temperature. The carbon material was then isolated by filtering the mixture. Then, the carbon material was washed with pure water, filtered, and dried.

(2-3-4. Third heating step)
The carbon material from which the MgO mold had been removed was placed in a graphite crucible, and the crucible was set in a Tanman furnace. Then, 100 mL of argon gas was circulated in the Tanman furnace. Then, the temperature in the Tanman furnace was raised to the third heating temperature shown in Table 2 at 10 ° C./min, and the temperature was maintained for 1 hour. After that, the temperature inside the Tanman furnace was cooled to room temperature. A carbon material was obtained by the above steps.

(2-4. Carbon material not subjected to the second and third heating steps (Comparative Examples 14 to 16))
In Comparative Examples 14 to 16, the following steps were performed using the MgO mold shown in Table 2. As a result, the carbon materials according to Comparative Examples 14 to 16 were produced.

(2-4-1. First heating step)
Polyvinyl alcohol powder (manufactured by Denki Kagaku Kogyo Co., Ltd., Denka Poval) was prepared as a carbon source. Then, the MgO mold and the carbon source were mixed at a mass ratio of 1: 2, and the mixture was placed in a quartz crucible. The crucible was inserted into an electric furnace, and argon gas was circulated in the electric furnace at 100 mL / min. Then, the temperature inside the electric furnace was raised at 10 ° C./min, and the temperature was maintained at 800 ° C. for 1 hour. After allowing to cool naturally, the MgO-carbon complex was taken out.

(2-4-2. MgO removal step)
The MgO-carbon composite after the first heating step was put into the eggplant flask, and 1M dilute sulfuric acid in a mass ratio of 100 times the mass ratio of the MgO-carbon composite was put into the eggplant flask. Then, the mixed solution in the eggplant flask was stirred for 3 hours in an oil bath maintained at a temperature of 60 ° C. As a result, the MgO mold was eluted in the mixed solution. Then, the mixed solution was cooled to room temperature. The carbon material was then isolated by filtering the mixture. Then, the carbon material was washed with pure water, filtered, and dried. A carbon material was obtained by the above steps.

(2-5. Carbon material that does not complete the MgO removal step (Comparative Example 17))
In Comparative Example 17, the following steps were performed using the MgO mold shown in Table 2. As a result, the carbon material according to Comparative Example 17 was produced.

(2-5-1. First heating step)
Polyvinyl alcohol powder (manufactured by Denki Kagaku Kogyo Co., Ltd., Denka Poval) was prepared as a carbon source. Then, the MgO mold and the carbon source were mixed at a mass ratio of 1: 2, and the mixture was placed in a quartz crucible. The crucible was inserted into an electric furnace, and argon gas was circulated in the electric furnace at 100 mL / min. Then, the temperature inside the electric furnace was raised at 10 ° C./min, and the temperature was maintained at the first heating temperature shown in Table 1 for 1 hour. After allowing to cool naturally, the MgO-carbon complex was taken out.

(2-5-2. Second heating step)
2-3 g of MgO-carbon composite was weighed on an alumina boat and set in a horizontal tubular electric furnace. Then, nitrogen gas was circulated in the electric furnace at 100 ml / min, and the temperature in the electric furnace was maintained at the second heating temperature shown in Table 2 for 1 hour.

(2-5-3. MgO removal step)
The MgO-carbon composite after the second heating step was put into the eggplant flask, and 1M dilute sulfuric acid in a mass ratio of 1.5 times the mass ratio of the MgO-carbon composite was put into the eggplant flask. Then, the mixed solution in the eggplant flask was stirred at room temperature for 0.5 hours. As a result, the MgO mold was eluted in the mixed solution. Then, the carbon material was isolated by filtering the mixed solution. Then, the carbon material was washed with pure water, filtered, and dried. A carbon material was obtained by the above steps.

(2-6. Commercially available porous carbon material: carbon black, activated carbon (Comparative Examples 18 to 20))
Commercially available carbon black (manufactured by Lion, Ketjen Black), EC300J, EC600JD and activated carbon YP80F were prepared as carbon materials according to Comparative Examples 18 to 20.

<3. Adjustment of fuel cells and evaluation of their battery performance>
(3-1. Preparation of catalyst)
Any of the carbon materials according to Examples 1 to 9 and Comparative Examples 1 to 20 was selected. Then, the selected carbon material was dispersed in a mixed solution containing chloroplatinic acid, water, and ethanol in a predetermined ratio, and then degassed. Next, aqueous ammonia was slowly added dropwise to the mixed solution as a precipitating agent (reducing agent), and the mixed solution was stirred for 1 hour. Then, the precipitate obtained was washed and passed with aqueous ammonia. The obtained solid content was calcined in He gas at 350 ° C. for 3 hours to obtain a platinum-supported carbon material having a platinum-supported amount of 40% by mass.

The platinum-supported carbon material was placed in a container under an Ar atmosphere, Nafion (registered trademark: Nafion) manufactured by DuPont was added thereto, and the mixture was lightly stirred, and then the platinum-supported carbon material was crushed by ultrasonic waves. Further, the mixture was stirred, ethanol was added to the mixture, and the solid content concentration of the platinum-supported carbon material and Nafion was adjusted to 1% by mass. Through the above steps, a catalyst metal ink in which a platinum-supported carbon material and an electrolyte resin were mixed was prepared. Then, ethanol was further added while stirring a predetermined amount of the catalyst metal ink prepared as described above to prepare a coated ink having a platinum concentration of 0.5% by mass.

(3-2. Preparation of catalyst layer)
The mass of the catalyst metal component (platinum) per unit area of the catalyst layer (hereinafter referred to as the amount of the catalyst metal component) ^{was set to 0.2 mg / cm 2} , and the spray conditions were adjusted so as to obtain the desired amount of the catalyst layer. Then, the coating ink was sprayed onto a Teflon (registered trademark) sheet, and then dried in argon at 120 ° C. for 60 minutes. A catalyst layer was prepared by the above steps.

(3-3. Preparation of MEA)
Using the prepared catalyst layer, MEA (membrane electrode complex) was prepared by the following method. A square electrolyte membrane having a side of 6 cm was cut out from the Nafion membrane (NR211 manufactured by DuPont). Further, each catalyst layer of the anode and the cathode coated on the Teflon sheet was cut out into a square shape having a side of 2.5 cm with a cutter knife.

The electrolyte membrane was sandwiched between the anode and cathode catalyst layers cut out in this manner so that the catalyst layers were in contact with each other with the central portion of the electrolyte membrane interposed therebetween and were not displaced from each other. Then, the prepared laminate was pressed at 120 ° C. and 100 kg / cm ² for 10 minutes. After cooling the pressed laminate to room temperature, only the Teflon sheet was carefully peeled off for both the anode and the cathode. As a result, a catalyst layer-electrolyte membrane conjugate in which the anode and cathode catalyst layers were fixed to the electrolyte membrane was prepared.

Next, as a gas diffusion layer, a pair of square carbon papers having a side size of 2.5 cm were cut out from carbon paper (39BC manufactured by SGL Carbon Co., Ltd.). Then, the catalyst layer-electrolyte membrane junction was sandwiched between these carbon papers so that the anode and cathode catalyst layers were aligned and did not shift. Then, the laminate was pressed at 120 ° C. and 50 kg / cm ² for 10 minutes to prepare MEA.

The basis weight of each component of the catalyst metal component, the carbon material, and the electrolyte material in each of the produced MEAs was calculated as follows. That is, the mass of the catalyst layer fixed on the Nafion membrane (electrolyte membrane) was obtained from the difference between the mass of the Teflon sheet with the catalyst layer before pressing and the mass of the Teflon sheet peeled off after pressing, and was obtained from the mass ratio of the composition of the catalyst layer. The amount of each component of the catalyst metal component, carbon material, and electrolyte material was calculated.

(3-4. Fuel cell evaluation test)
(3-4-1. Evaluation of power generation performance with low current density)
In line with the object of this embodiment, the power generation performance of low current density immediately after high current output using a highly humidified gas that easily floods was evaluated.

Specifically, each of the produced MEAs was incorporated into a cell, and the performance of the fuel cell was evaluated using a fuel cell measuring device. Here, air was supplied to the cathode and pure hydrogen was supplied to the anode so that the utilization rates were 40% and 70%, respectively. Each gas pressure was set to 0.1 MPa with a back pressure valve provided downstream of the cell. The cell temperature was set to 80 ° C. In addition, by passing the gas (air, pure hydrogen) supplied to the fuel cell through distilled water kept at 80 ° C. in a humidifier (that is, bubbling), water vapor equivalent to reformed hydrogen is added to these gases. Included. After that, these gases were supplied to the fuel cell.

Under such settings, the cell load was increased and ^{held at a current density of 1000 mA / cm 2} for 1 hour. Current density 1000 mA / cm ^{2 and} for what cell voltage after one hour hold is equal to or greater than 0.3V, set current density of 100 mA / cm ^2, after setting the current density to 100 mA / cm ² The voltage between the cell terminals after 1 minute was recorded as the output voltage. This output voltage was used as the power generation performance with low current density.

The performance evaluation results of the obtained fuel cells were evaluated based on the pass ranks of ◎, ○, and ● and the fail ranks of ×. As for the passing rank, ◎ has the highest performance, followed by ○, and then ●. Specifically, when the output voltage is 0.860 V or higher, the power generation performance is rated as ⊚. When the output voltage was 0.840V or more and less than 0.860V, the power generation performance was rated as pass rank ◯. When the output voltage is 0.820V or more and less than 0.840V, the power generation performance is rated as acceptable. If the output voltage is less than 0.820V, or if the cell voltage is less than 0.3V after holding for 1 hour at ^{a current density of 1000mA / cm 2, the power generation performance is marked with x.} Table 2 shows the values of the output voltage (output voltage after holding at ^{100 mA / cm 2} for 1 minute) and the power generation performance evaluation results obtained in Examples 1 to 9 and Comparative Examples 1 to 20 in Table 2. .. Incidentally, what current density 1000 mA / cm ² at a cell voltage after one hour hold is less than 0.3V, because not measure the output voltage at 100 mA / cm ^2, Table 2, - was described as "".

(3-4-2. Fuel cell durability test)
Next, in the above test, a durability test was evaluated for a fuel cell having a cell voltage of 0.3 V or more after being held ^{at 1000 mA / cm 2 for 1 hour.}

Specifically, the cycle of holding the cell terminal voltage at 1.0 V for 1 second, then increasing the cell terminal voltage to 1.5 V and holding it for 1 second, and then returning the cell terminal voltage to 1.0 V is 4000. Repeated times. After that, the battery performance was measured in the same manner as in the evaluation test of the low current density characteristics.

How decrease the output voltage after endurance test (output voltage after held for 1 minute at 100 mA / cm ²⁾ is compared with (output voltage after held for 1 minute at 100 mA / cm ²⁾ Output voltage before the durability test It was calculated as the rate of decrease. When the reduction rate is less than 15%, the durability is evaluated as pass rank ◯, and when the reduction rate is 15% or more, the durability is rejected. The results are shown in Table 2. If the cell voltage is less than 0.3 V after holding for 1 hour at a current density of 1000 mA / cm ² , the durability test has not been performed, and therefore, it is described as “−” in Table 2.

Further, in Table 2, for the carrier carbon materials of Examples 1 to 9 and Comparative Examples 1 to 20, the relative pressures (P / P ₀ ) c and (P / P ₀ ) c + 0.1 to 1 (P / P 0) c for closing the hysteresis loop are shown. Desorption amount dV between P / P ₀ ) c, BET specific surface area (m ² / g), half width of G band detected in the range of 1550 to 1650 cm ^{-1 of Raman spectroscopic spectrum ΔG (cm -1} ) The measurement results are also described. Here, in Comparative Examples 18 and 20, since the hysteresis loop was not closed, "*" was added _{to the relative pressure (P / P 0) c.}

As is clear from Table 2, in Examples 1 to 9 that satisfy the requirements of the present embodiment, excellent power generation performance was obtained. In particular, in Examples 6 to 9 in which the third heating step was performed, the durability was also excellent. On the other hand, in Comparative Examples 1 to 20 which did not satisfy the requirements of the present embodiment, the power generation performance was inferior. In addition, the durability was also inferior in most of the comparative examples. Comparative Examples 14 to 16 correspond to Patent Documents 1 and 2, but were inferior in both power generation performance and durability.

Further, BET specific surface area of 900 meters ² / g greater, in an exemplary less than 1500 m ² / g, or, in the embodiment the average particle size of the carbon material is less than 200nm ultra 1000 nm, there is a tendency that the power generation performance is improved rice field.

Although the preferred embodiments of the present invention have been described in detail with reference to the accompanying drawings, the present invention is not limited to such examples. It is clear that a person having ordinary knowledge in the field of technology to which the present invention belongs can come up with various modifications or modifications within the scope of the technical ideas described in the claims. , These are also naturally understood to belong to the technical scope of the present invention.

10 MgO template particles 10a Crystallet 20 Carbon layer A Hysteresis loop 30 Bottleneck type pores 20a, 30a Bottle part 20b, 30b Neck part 100 Polymer electrolyte fuel cell 110, 120 Separator 130, 140 Gas diffusion layer 150, 160 Catalyst Layer 170 electrolyte membrane

Claims (7)

_{The relative pressure (P / P 0} ) c at which the hysteresis loop formed by the nitrogen adsorption / desorption isotherm closes is more than 0.55 and less than 0.8.
The amount of nitrogen desorption dV between (P / P ₀ ) c + 0.1 to (P / P ₀ ) c is more than ^{65 cm 3 / g.}
A template carbon material for a fuel cell catalyst carrier, characterized by having a BET specific surface area of ^{more than 700 m 2 / g.} Wherein the half-width △ G of G band derived from the Raman spectrum is 35 cm ^-1 greater, less than 70cm ^-1, a fuel cell catalyst support mold carbon material of claim 1. BET specific surface area of 900 meters ² / g greater, and less than 1500 m ² / g, the fuel cell catalyst support mold carbon material according to claim 1 or 2. The template carbon material for a fuel cell catalyst carrier according to any one of claims 1 to 3, wherein the average particle size of the carbon material is more than 200 nm and less than 1000 nm. A fuel cell catalyst layer comprising the template carbon material for a fuel cell catalyst carrier according to any one of claims 1 to 4. A fuel cell comprising the catalyst layer for a fuel cell according to claim 5. The fuel cell according to claim 6 , wherein the fuel cell catalyst layer is a cathode-side catalyst layer.

Images