Carbon material for catalyst support of solid polymer type fuel cell, catalyst layer for solid polymer type fuel cell, and fuel cell

Abstract

A carbon material for a solid polymer fuel cell catalyst carrier, in which a nitrogen adsorption-desorption isotherm shows two hysteresis loops of a first hysteresis loop and a second hysteresis loop in a range of a relative pressure P / P0 of 0.4 or more.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to a carbon material for a catalyst carrier of a solid polymer fuel cell, a catalyst layer for a solid polymer fuel cell, and a fuel cell. BACKGROUND

[0002] A solid polymer fuel cell, which is one type of fuel cell, has a pair of catalyst layers disposed on both sides of a solid polymer electrolyte membrane, a gas diffusion layer disposed on the outer side of each catalyst layer, and a separator disposed on the outer side of each gas diffusion layer. One of the pair of catalyst layers is an anode of the solid polymer fuel cell, and the other is a cathode of the solid polymer fuel cell. In addition, in a typical solid polymer fuel cell, a plurality of unit cells having the above-described components are stacked in order to obtain a desired output.

[0003] A reducing gas such as hydrogen is introduced to the separator on the anode side. The gas diffusion layer on the anode side diffuses the reducing gas and then introduces it to the anode. The anode contains a catalyst component, a catalyst carrier that supports the catalyst component, and an electrolyte material having proton conductivity. The catalyst carrier is mostly composed of a carbon material. On the catalyst component, an oxidation reaction of the reducing gas occurs, and protons and electrons are generated. For example, in the case where the reducing gas is hydrogen, the following oxidation reaction occurs.

[0004] H2→ 2H + + 2e - (E0 = 0 V)

[0005] The protons generated in this oxidation reaction are introduced to the cathode via the electrolyte material in the anode and the solid polymer electrolyte membrane. In addition, the electrons are introduced to the external circuit via the catalyst carrier, the gas diffusion layer, and the separator. The electrons do work (generate electricity) in the external circuit and are then introduced to the separator on the cathode side. Then, the electrons are introduced to the cathode via the separator on the cathode side and the gas diffusion layer on the cathode side.

[0006] The solid polymer electrolyte membrane is composed of an electrolyte material having proton conductivity. The solid polymer electrolyte membrane introduces the protons generated in the above-described oxidation reaction to the cathode.

[0007] An oxidizing gas such as oxygen or air is introduced to the separator on the cathode side. The gas diffusion layer on the cathode side diffuses the oxidizing gas and then introduces it to the cathode. The cathode contains a catalyst component, a catalyst carrier that supports the catalyst component, and an electrolyte material having proton conductivity. The catalyst carrier is mostly composed of a carbon material. On the catalyst component, a reduction reaction of the oxidizing gas occurs, and water is generated. For example, in the case where the oxidizing gas is oxygen or air, the following reduction reaction occurs.

[0008] O2+ 4H+ +4e - → 2H2O (E0= 1.23 V)

[0009] The water produced in the reduction reaction is discharged to the outside of the fuel cell together with the unreacted oxidizing gas. In this way, in the solid polymer fuel cell, power generation is performed using the energy difference (potential difference) between the oxidation reaction and the reduction reaction. In other words, the electrons produced in the oxidation reaction do work in the external circuit.

[0010] In addition, in recent years, as disclosed in Patent Documents 1 to 5, a technology of using a porous carbon material as a catalyst support has been proposed. As disclosed in Patent Documents 1 to 4, for a porous carbon material having a three-dimensional dendritic structure in the porous carbon material (hereinafter also referred to as a dendritic carbon nanostructure), a characteristic structure that cannot be seen in other carbon materials is possessed. Specifically, the dendritic carbon nanostructure has a structure that takes into account both a very developed fine pore structure (porous structure) and a large-scale dendritic structure. That is, by utilizing the dendritic carbon nanostructure as a catalyst support, the catalyst layer can have a large number of fine pores that can load a catalyst component inside the catalyst support. Therefore, the catalyst layer can improve the loading concentration of the catalyst. In addition, the gas diffusion in the fine pores is also excellent, and the gas diffusion in the catalyst layer is excellent. Thus, the dendritic carbon nanostructure has a characteristic of excellent large-current power generation compared to the conventional carbon black-based porous carbon material such as Ketjen black.

[0011] In addition, in Patent Document 5, a technology of using a porous carbon as a catalyst support is proposed, which is obtained by using, as a raw material, a MgO cast carbon material in which magnesium oxide nanoparticles are made into a cast shape, and further performing a heat treatment in air. The MgO cast carbon generally has a characteristic of having uniform fine pores even inside the material. Moreover, by performing a heat treatment of the MgO cast carbon in air, the carbon wall forming the fine pores is oxidized and consumed, and the spacing between the fine pores is reduced. That is, the MgO cast carbon is proposed as a material that attempts to improve the connectivity.

[0012] Prior Art Documents

[0013] Patent Documents

[0014] Patent Document 1: International Publication No. 2014 / 129597

[0015] Patent Document 2: International Publication No. 2015 / 088025

[0016] Patent Document 3: International Publication No. 2015 / 141810

[0017] Patent Document 4: International Publication No. 2016 / 133132

[0018] Patent Literature 5: Japanese Patent Application Laid-Open No. 2015-164889 SUMMARY

[0019] PROBLEMS TO BE SOLVED BY THE INVENTION

[0020] However, in the case where a large current flows in a solid polymer fuel cell, the overvoltage tends to become large. Therefore, for example, in a fuel cell vehicle in which a fuel cell is used as a power source, in order to emphasize the maximum output, there is a strong desire to reduce the overvoltage at the time of a large current. Moreover, in order to further popularize the solid polymer fuel cell, it is necessary to reduce the overvoltage at the time of a large current more than ever.

[0021] Therefore, an object of the present disclosure is to provide a novel and improved carbon material for a catalyst carrier of a solid polymer fuel cell, a catalyst layer for a solid polymer fuel cell, and a fuel cell, which can further reduce the overvoltage at the time of a large current.

[0022] MEANS FOR SOLVING THE PROBLEMS

[0023] The carbon material for a catalyst carrier of a solid polymer fuel cell, the catalyst layer for a solid polymer fuel cell, and the fuel cell of the present disclosure include the following means.

[0024] <1>

[0025] A carbon material for a catalyst carrier of a solid polymer fuel cell, in which a nitrogen adsorption-desorption isotherm shows two hysteresis loops of a first hysteresis loop and a second hysteresis loop in a range where the relative pressure P / P0 is 0.4 or more.

[0026] <2>

[0027] The carbon material for a catalyst carrier of a solid polymer fuel cell according to <1>, as the two hysteresis loops shown by the nitrogen adsorption-desorption isotherm, has a first hysteresis loop present in a range where the relative pressure P / P0 is 0.87 or more, and a second hysteresis loop present in a range where the relative pressure P / P0 is 0.4 to 0.87.

[0028] <3>

[0029] The carbon material for a catalyst carrier of a solid polymer fuel cell according to <2> satisfies the following conditions (A), (B), and (C):

[0030] (A) The specific surface area obtained by BET analysis of the nitrogen adsorption isotherm is 450 m 2 / g to 1500 m 2 / g;

[0031] (B) the first hysteresis loop has a difference ΔV in adsorption amount in the range of relative pressure P / P0 of 0.87 ± 0.03 0.87 0.87min is 20 mL / g or less;

[0032] (C) the second hysteresis loop has an area ΔS 0.4-0.87 of 5 mL / g to 50 mL / g.

[0033] (4)

[0034] The carbon material for a solid polymer fuel cell catalyst support according to (3) further satisfies the following condition (D) :

[0035] (D) in a nitrogen adsorption isotherm, a difference ΔV in adsorption amount between an adsorption amount at a relative pressure P / P0 of 0.99 and an adsorption amount at a relative pressure P / P0 of 0.95 0.95-0.99 is 500 mL / g to 1100 mL / g.

[0036] (5)

[0037] The carbon material for a solid polymer fuel cell catalyst support according to (3) or (4) further satisfies the following condition (E) :

[0038] (E) in a Raman spectrum obtained by Raman spectroscopy, a half-value width of a G band detected in the range of 1500 cm -1 to 1700 cm -1 is 45 cm -1 to 70 cm -1 .

[0039] (6)

[0040] A catalyst layer for a solid polymer fuel cell, comprising the carbon material for a solid polymer fuel cell catalyst support according to any one of (1) to (5).

[0041] (7)

[0042] A fuel cell comprising the catalyst layer for a solid polymer fuel cell according to (6).

[0043] (8)

[0044] The fuel cell according to (7), wherein the catalyst layer for a solid polymer fuel cell is a cathode-side catalyst layer.

[0045] Effects of the Invention

[0046] ​According to the present disclosure, there is provided a novel and improved carbon material for a catalyst support of a solid polymer type fuel cell, a catalyst layer for a solid polymer type fuel cell, and a fuel cell, which are capable of further reducing overvoltage at a large current. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1 is a graph schematically showing an example of a nitrogen adsorption-desorption isotherm of the carbon material for a catalyst support of a solid polymer type fuel cell of the present disclosure.

[0048] Figure 2 is an explanatory diagram schematically showing an example of a bottle neck type pore.

[0049] Figure 3 is a schematic diagram showing an example of a general configuration of the fuel cell of the present disclosure.

[0050] Figure 4 is a photograph showing a measurement method for measuring a branch diameter in the carbon material for a catalyst support of the present disclosure when SEM observation is performed.

[0051] Figure 5 is an explanatory diagram showing a measurement method for a branch diameter of the carbon material for a catalyst support of the present disclosure. DETAILED DESCRIPTION

[0052] In the present disclosure, a numerical range indicated using “~” means a range including the numerical values written before and after “~” as lower limit values and upper limit values. In addition, a numerical range in which the numerical values written before and after “~” are annotated with “more than” or “less than” means a range not including these numerical values as lower limit values or upper limit values.

[0053] In the present disclosure, the term “process” includes not only an independent process, but also a range in which the process cannot be clearly distinguished from other processes as long as the desired purpose of the process can be achieved.

[0054] <1. Carbon material for a catalyst support of a solid polymer type fuel cell>

[0055] For the carbon material for a catalyst support of a solid polymer type fuel cell of the present disclosure (hereinafter also referred to as “carbon material for a catalyst support”), a nitrogen adsorption-desorption isotherm shows two hysteresis loops including a first hysteresis loop and a second hysteresis loop in a range where the relative pressure P / P0 is 0.4 or more. Furthermore, the carbon material for a catalyst support of the present disclosure is capable of further reducing overvoltage at a large current.

[0056] The dominant factor that affects the overvoltage at a large current is the resistance to movement (diffusion resistance) of the substances involved in the reaction at the cathode side. Among these, the substances involved in the reaction at the cathode side are electrons, protons, oxygen, and water (water vapor) that is generated in the reaction at the cathode side. It is considered that the electrons and the protons among these substances exhibit Ohmic behavior. That is, the resistance values thereof remain constant regardless of the magnitude of the current. Therefore, it is considered that the dominant factor of the overvoltage at a large current is not the electrons and the protons, but the oxygen and the water vapor. That is, it is considered that the diffusion resistance of the gas flowing within the cathode (so-called gas diffusion resistance) is the dominant factor of the overvoltage at a large current.

[0057] In the past, carbon materials have been evaluated on the basis of nitrogen adsorption isotherms. However, the characteristics of carbon materials at a large current cannot be sufficiently evaluated by nitrogen adsorption isotherms alone. For example, Ketjen black is also widely used as a catalyst support for solid polymer fuel cells. Ketjen black has a dendritic structure similar to that of dendritic carbon nanostructures, and is porous within the branches. However, the branches that form the dendritic structure are thin, and the pore diameters of the interbranch pores are small. Therefore, Ketjen black does not exhibit good gas diffusivity like dendritic carbon nanostructures. Thus, by activating Ketjen black, the total pore volume can be increased. In the case where the Ketjen black that has been subjected to activation treatment is used as a catalyst support, there is a tendency for the gas diffusion resistance to decrease and the overvoltage at a large current to become smaller compared to the Ketjen black before the activation treatment. However, the overvoltage at a large current is not sufficiently reduced. Therefore, the characteristics of carbon materials, particularly the characteristics at a large current, cannot be sufficiently evaluated by nitrogen adsorption isotherms that focus on only the adsorption process.

[0058] Therefore, the present inventors focused on nitrogen adsorption isotherms and nitrogen desorption isotherms of carbon materials for catalyst supports in order to evaluate the diffusion resistance of the gas flowing within the cathode. The nitrogen adsorption isotherms and the nitrogen desorption isotherms are obtained by nitrogen gas adsorption measurement. The nitrogen adsorption isotherm is an isotherm on the nitrogen adsorption side, and the nitrogen desorption isotherm is an isotherm on the nitrogen desorption side. Hereinafter, the isotherm obtained by combining the nitrogen adsorption isotherm and the nitrogen desorption isotherm (i.e., the isotherm obtained by connecting these isotherms) is also referred to as a nitrogen adsorption-desorption isotherm.

[0059] For example, in the support of a carbon black raw material that has been made porous by activation treatment and has been used in the past, pores are formed in the catalyst layer by the gaps between the carbon black raw materials. The pores formed in the catalyst layer reflect the primary particle diameters of the carbon black. The pore diameters are generally 30 nm to 40 nm (diameter). For example, the Ketjen black after the activation treatment has a tendency for the area of the above-mentioned hysteresis loop to increase. That is, the volume ratio of the following-described bottleneck-type pores increases. When the nitrogen adsorption-desorption isotherm in the catalyst layer that uses the carbon black raw material made porous by such activation treatment as a support is measured, the nitrogen adsorption-desorption isotherm exhibits one hysteresis loop.

[0060] In addition, the porous carbon material disclosed in Patent Literature 5 has a shape of particles. Even if the gas flowability within the particles is improved, the pore volume of the catalyst layer formed of the particles is small, and it is difficult to control the pore size. As a result, not the gas diffusion within the particles but the gas diffusion within the catalyst layer becomes a constraint, and the level required for the high current characteristics is not achieved. Thus, when measuring the nitrogen adsorption / desorption isotherm in the catalyst layer using MgO molded carbon as the carrier, the nitrogen adsorption / desorption isotherm shows a hysteresis loop.

[0061] For the catalyst carrier carbon material showing the hysteresis loop as described above, the present inventors investigated the relationship between the ideal image and the hysteresis for the catalyst carrier. In order to reduce the overvoltage at a high current, it is sufficient to reduce the resistance to the movement of the substance at a high current. For example, in the case of the dendritic carbon nanostructure, the branch diameter is about 50 nm to 100 nm, and the branch length is about 50 nm to 200 nm. Thus, the pores of the catalyst layer are on the order of about 100 nm (width), and the diffusivity of the gas in the catalyst layer is overwhelmingly higher than that of the carrier of the carbon black raw material. From this viewpoint, in the case where the dendritic carbon nanostructure is used as the catalyst carrier on the cathode side, it can be understood that the gas diffusion resistance is smaller than that of the catalyst carrier of the carbon black raw material based on the three-dimensional form of the dendritic carbon nanostructure.

[0062] Focusing on the dendritic carbon nanostructure, there are two types of movement of the substance, the movement of the substance in the pores existing within the branches themselves (hereinafter also referred to as "intra-branch pores") and the movement of the substance in the pores formed in the gaps between the branches (hereinafter also referred to as "inter-branch pores"). For example, if the gas diffusion to the surface of the carrier is insufficient, the high current characteristics will not be high even if the diffusion of the substance in the intra-branch pores is high. That is, in the catalyst layer, it is important to reduce the resistance to the gas diffusion and to improve the supply of the gas to the surface of the carrier. The pores formed in the catalyst layer are mainly the inter-branch pores of the carrier. Therefore, it is important that the pore diameter of the inter-branch pores is large, and the pore volume is also large, and the distribution width is narrower.

[0063] It is considered that in the dendritic carbon nanostructure, if the pore diameter and the pore volume in the inter-branch pores are large, the nitrogen adsorption-desorption isotherm shows a hysteresis loop in the region on the high relative pressure side. If the dendritic carbon nanostructure can show such a hysteresis loop, the substance movement resistance at a large current can be made small, and a decrease in the overvoltage at a large current can be predicted. In addition, the nitrogen adsorption-desorption isotherm of the dendritic carbon nanostructure shows a hysteresis loop due to the intra-branch pores. That is, the dendritic carbon nanostructure of the present disclosure shows two hysteresis loops in the range of a relative pressure P / P0 of 0.4 or more, and it is considered that the substance movement resistance at a large current can be reduced, and the overvoltage at a large current can be decreased. Furthermore, the conventional catalyst layer using carbon black as a carrier, the catalyst layer using MgO molded carbon as a carrier, and the like cannot show such a hysteresis loop, and it is considered that the overvoltage at a large current cannot be sufficiently decreased.

[0064] The above is described using an example of a dendritic carbon nanostructure, but it is considered that as long as the carbon material for a catalyst carrier shows two hysteresis loops in the range of a relative pressure P / P0 of 0.4 or more, the overvoltage at a large current can be reduced, regardless of the dendritic carbon nanostructure.

[0065] As described above, the carbon material for a catalyst carrier of the present disclosure shows two hysteresis loops in the range of a relative pressure P / P0 of 0.4 or more. The carbon material for a catalyst carrier is not particularly limited. From the viewpoint of easily decreasing the overvoltage at a large current, it is preferable that the carbon material for a catalyst carrier be a dendritic carbon nanostructure. Hereinafter, a dendritic carbon nanostructure of a preferable mode will be described.

[0066] In the present disclosure, the dendritic carbon nanostructure has a structure in which a very developed pore structure and a large-scale dendritic structure are combined. That is, the dendritic carbon nanostructure has a dendritic structure formed of branches composed of porous carbon. In addition, the branches constituting the dendritic structure branch diversely. Each branch is long and thick (for example, the branch diameter is about 50 nm to 100 nm (width), and the branch length is about 50 nm to 200 nm). Moreover, the deviation of the branch diameter is small. In other words, the distribution width of the branch diameter is narrow. The structure of the dendritic carbon nanostructure is that thick and long cylindrical branches branch diversely. In the inter-branch pores, thick and long cylindrical pores corresponding to the branch diameter and the branch length branch diversely.

[0067] In addition, the intra-branch pores also have the same characteristics as the inter-branch pores. That is, the intra-branch pores have a thick and long cylindrical shape and branch diversely within the branches. Therefore, it is considered that the gas diffusion resistance in the catalyst layer of the dendritic carbon nanostructure of the present disclosure is easily reduced.

[0068] Further, in the present disclosure, the dendritic carbon nanostructure shows a dendritic carbon structure having a branch diameter of 10 nm or less to 200 nm. For the branch diameter, by a scanning electron microscope (SEM; SU-9000 manufactured by Hitachi High-Technologies Corporation), at a magnification of 100,000 times (2.5 μm x 2 μm), SEM images of 5 fields of view are observed, and the branch diameter is measured at 20 places on each of the images, and the average of the measured values of 100 places in total is taken as the value of the branch diameter. Further, for the measured branch diameter, the thickness of the central portion between the two branch points adjacent to the branch of interest (the central portion of the branched branch) is measured as the branch diameter (refer to Figure 4 . Figure 4 In the present disclosure, D shows the branch diameter at one place. Figure 5 The measurement method of the branch diameter is described with reference to Figure 5 In the present disclosure, one branch of interest is shown. For the branch of interest, the branch point BP1 and the branch point BP2 are determined. Then, the determined branch point BP1 and the branch point BP2 are connected, and the thickness (width) of the branch is measured at the position of the perpendicular bisector BC of the line connecting the branch point BP1 and the branch point BP2. The measured thickness of the branch is the branch diameter D at one place. Further, the branch length is represented by the distance from the branch point BP1 to the branch point BP2 (refer to Figure 5 For example, in the present disclosure, the branch diameter of the dendritic carbon nanostructure is not particularly limited as long as it is in the above range, and for example, it can be about 50 nm to 100 nm. In addition, the branch length is not particularly limited, and for example, it can be about 50 nm to 200 nm.

[0069] Hereinafter, one example of a preferred mode of the present disclosure will be described in detail with reference to the drawings. Further, in the present disclosure, for the constituent elements having substantially the same function and configuration, the same reference numerals are annotated, and thus repeated description is omitted.

[0070] Figure 1 is a graph schematically showing the nitrogen adsorption-desorption isotherm in the carbon material for catalyst support of the present disclosure. As a preferred mode, Figure 1 indicates a mode when the carbon material for catalyst support is a dendritic carbon nanostructure. Figure 1 The nitrogen adsorption-desorption isotherm shown in Figure 1 In the nitrogen adsorption-desorption isotherm shown in

[0071] Generally, the nitrogen adsorption isotherm at liquid nitrogen temperature (e.g., 77 K) corresponds to the fact that the lower the relative pressure, the more nitrogen is adsorbed into pores with smaller diameters. Therefore, the first hysteresis loop A1 is formed based on the characteristics of interbranch pores, and the second hysteresis loop A2 is formed based on the characteristics of intrabranch pores.

[0072] Here, the mechanism for generating hysteresis rings and the policy for carrier carbon materials based on this mechanism, using hysteresis rings as an indicator, are described in detail.

[0073] like Figure 1 As shown, in the desorption curve (nitrogen desorption isotherm), the relative pressure required to reach the same adsorption volume as the adsorption curve (nitrogen adsorption isotherm) is lower than that for adsorption. This indicates that the adsorbed nitrogen is difficult to remove. During desorption, nitrogen escapes from the path most easily accessible among multiple desorption paths (i.e., the path with the largest pores). However, during desorption, the adsorbed nitrogen does not escape until the relative pressure decreases to a level comparable to the pore diameter. This is the hysteresis phenomenon. The formation of a hysteresis loop refers to the consistency of adsorption and desorption at a certain relative pressure, equivalent to the disappearance of the pore effect. In this disclosure, for example, the first hysteresis loop A1 disappears near a relative pressure of 0.87. The first hysteresis loop A1 represents the hysteresis phenomenon caused by interbranch pores. Therefore, the disappearance of the hysteresis caused by interbranch pores near a relative pressure of 0.87 manifests as the disappearance of pores generated by branches near 0.87. That is, for dendritic carbon nanostructures, the pore diameter formed by the dendritic structure also produces a corresponding distribution width depending on the size of the distribution width of the branch diameter and branch length. As a result, it becomes a carbon material that is hysteretic and non-closed, and does not form two hysteretic rings. Therefore, in the case where the carbon material used as a carrier in this disclosure is a dendritic carbon nanostructure, the branch structure of the dendritic carbon nanostructure is required to be a structure with consistent branch diameter and branch length, forming two rings.

[0074] Specifically, it was proposed that a hysteresis loop would occur in the presence of bottleneck-type fine pores. Figure 2 A bottleneck-type pore 30 is shown as an example of a bottleneck-type pore formed in a carbon material. The bottleneck-type pore 30 consists 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. In addition, the neck portion 30B communicates with the outer surface of the carbon material or other pores within the carbon material.

[0075] In this regard, as explained below, a hysteresis loop is generated depending on the difference between the adsorption process and the desorption process. That is, in the adsorption process, the adsorption thickness of nitrogen gradually becomes thick according to the relative pressure of nitrogen. In this process, first, the neck portion of the bottle neck type fine pore is occluded by adsorption. At this time, even if the bottle portion adsorbs nitrogen at a thickness equivalent to the neck portion, since the diameter of the bottle portion is larger than that of the neck portion, the bottle portion is not yet completely filled. In this state, for example, even if the neck portion is occluded, the external pressure and the pressure inside the bottle portion are in equilibrium through the nitrogen adsorption layer. Therefore, further, if the relative pressure of nitrogen increases (i.e., the external pressure increases), the pressure inside the bottle portion also increases. Therefore, in the bottle portion, adsorption equivalent to the external pressure is also performed, and even if the neck portion is still occluded, the nitrogen adsorption layer of the bottle portion becomes thick. If the external pressure further increases, eventually, the bottle portion is also completely occluded by the nitrogen adsorption layer. That is, even in the bottle neck type fine pore, in the adsorption process, adsorption is performed similarly to the usual fine pore structure without a neck portion. However, strictly speaking, when the neck portion is occluded by the nitrogen adsorption layer, at least a part of the nitrogen adsorption layer adsorbed in the neck portion becomes a liquid phase. Also, it is known that due to the surface tension of such a liquid phase, the pressure inside the liquid phase (i.e., the pressure inside the bottle portion) is slightly smaller than the external pressure. This phenomenon is called the so-called capillary phenomenon. Therefore, in the adsorption isotherm of the bottle neck type fine pore, the relative pressure required to obtain the same adsorption amount as that of the fine pore without a neck portion is shifted to a slightly high value.

[0076] Next, a desorption process of the state where nitrogen is filled from all of the bottle-neck type pores is considered. In the desorption process, until the external pressure is reduced to the pressure at which the occlusion of the neck portion is released, the neck portion is not opened. Note that, in the process where the external pressure is reduced to the pressure at which the occlusion of the neck portion is released, the external pressure becomes the desorption releasing pressure of the adsorption of the bottle portion. However, even in this state, due to the occlusion of the neck portion, the nitrogen adsorbed inside the bottle portion remains in this state (so-called blocking phenomenon). In a case where the desorption isotherm is measured with nitrogen as the adsorbate and at a liquid nitrogen temperature (for example, 77 K), at a relative pressure higher than the pressure at which the neck portion is opened, the adsorbed layer of nitrogen in the neck portion starts to boil (cavitation). Therefore, at the pressure at which the adsorbed layer of nitrogen starts to boil, the nitrogen adsorbed to the bottle portion and the neck portion is released in one breath. In a case where the bottle-neck type pores constitute the pores in the carbon particles (the pores in the branches in the dendritic carbon nanostructure), the pressure is known to be around a relative pressure P / P0 = 0.4. The relative pressure P / P0 around 0.4 is a relative pressure inherent to the adsorption of nitrogen, which is not affected by the pore structure of the adsorbate. That is, the relative pressure P / P0 at which the second hysteresis loop A2 is closed (that is, the pressure at which cavitation starts to occur) is determined by the adsorbate (nitrogen molecules) and the measurement temperature, and is independent of the pore structure. Thus, in the present disclosure, the presence of the bottle-neck type pores and the size of the volume of the bottle portion are determined from the amount of nitrogen released by cavitation. Moreover, in the present disclosure, even if the number of bottle-neck type pores is reduced, the volume of the bottle portion can be reduced even if the bottle-neck type pores are present. Furthermore, the phenomenon of hysteresis is described in detail in, for example, the following document.

[0077] (Document): Adsorption Hysteresis of Nitrogen and Argon in Pore Networks and Characterization of Novel Micro- and Mesoporous Silicas, Langmuir 2006, 22, 756-764.

[0078] As Figure 2 indicated, because the diameter of the neck portion 30B is smaller than the diameter of the bottle portion 30A, it is difficult for the gas to flow into the bottle portion 30A. In addition, it is difficult for the gas inside the bottle portion 30A to flow out to the outside. Therefore, it can be considered that the greater the volume ratio of the bottle-neck type pores in the catalyst carrier to the volume of the entire pores, the higher the gas diffusion resistance.

[0079] Moreover, the volume ratio of the bottle-neck type pores corresponds to the area of the hysteresis loop. That is, the smaller the area of the hysteresis loop, the smaller the volume ratio of the bottle-neck type pores.

[0080] As described above, in order to reduce the overvoltage at a large current, it is important to make the pore diameter between the branches large and to make the pore volume large in the dendritic carbon nanostructure. If this is re-expressed as hysteresis of nitrogen adsorption, it is as follows.

[0081] "Large pore diameter" corresponds to an increase in the adsorption curve in a region of high relative pressure. Specifically, in BJH (Barrett-Joyner-Halender) analysis, a relative pressure of 0.87 corresponds to a diameter of 20 nm, a relative pressure of 0.95 corresponds to a diameter of 38 nm, a relative pressure of 0.97 corresponds to a diameter of 64 nm, and a relative pressure of 0.98 corresponds to a diameter of 93 nm. In the case where the carbon material for a catalyst support of the present disclosure is a dendritic carbon nanostructure, since the branch diameter is, for example, approximately 50 nm to 100 nm, the relative pressure at which adsorption increases is 0.97 to 0.98. Moreover, since the branch diameters are uniform, the increase in adsorption is sharp. For example, in Ketjen black EC300, which is a typical porous carbon of the carbon black type having a dendritic structure, the particle diameter is approximately 40 nm, the increase in adsorption is near a relative pressure of 0.95, and since there is a deviation in the pore diameter, the slope of the adsorption is smaller than that of the dendritic carbon nanostructure.

[0082] On the other hand, "large pore volume" corresponds to the amount of gas adsorbed. Specifically, in the case where the carbon material for a catalyst support of the present disclosure is a dendritic carbon nanostructure, the adsorption between the branches corresponds to the amount of adsorption at a relative pressure of 0.95 to 0.99. The present inventors have successfully quantitatively defined the structure corresponding to the space between the branches by correlating the large current characteristics with the values thereof in the present disclosure.

[0083] In addition, in the present disclosure, it is preferable that the areas of the first hysteresis loop Al and the second hysteresis loop A2 be small. As described above, in the present disclosure, the inter-branch pores branch variously between the branches and have a relatively thick cylindrical shape. Similarly, the intra-branch pores branch variously within the branches and have a relatively thick cylindrical shape. Therefore, there are substantially no bottleneck-type pores between the branches and within the branches, and as a result, the areas of the first hysteresis loop Al and the second hysteresis loop A2 are small. Moreover, since the inter-branch pores are relatively thick, the first hysteresis loop Al closes at a relatively high relative pressure P / P0, i.e., approximately 0.87.

[0084] Moreover, the first hysteresis loop Al and the second hysteresis loop A2 described above are characteristics observed in the nitrogen adsorption-desorption isotherm of the dendritic carbon nanostructure as the carbon material for a catalyst support in the present disclosure. Two hysteresis loops are not observed in the nitrogen adsorption-desorption isotherms of other conventional carbon materials. Therefore, based on the nitrogen adsorption-desorption isotherm, the pore structure unique to the dendritic carbon nanostructure in the present disclosure can be evaluated.

[0085] As described above, in the dendritic carbon nanostructure in the present disclosure, the intrabranch pores have a thick and long cylindrical shape, and branch variously within the branch. In terms of further reducing the overvoltage at a large current, it is preferable to make the intrabranch pores larger and increase the branching of the intrabranch pores. Making the intrabranch pores larger is embodied as the second hysteresis loop A2. Specifically, because the diffusivity of the gas within the pores is improved, the area of the second hysteresis loop A2 becomes smaller.

[0086] Also, in the dendritic carbon nanostructure, the interbranch pores have a thick and long cylindrical shape, and branch variously between the branches. In terms of further reducing the overvoltage at a large current, it is preferable to thin the branches. Also, it is preferable that the distribution of the branch diameters be narrow. Thereby, the gas easily flows within the interbranch pores. Making the interbranch pores larger and making the distribution of the branch diameters narrow is embodied as the first hysteresis loop Al. Specifically, because the distribution of the diameters of the interbranch pores is narrow, and the distribution of the diameters of the interbranch pores is more concentrated on the long diameter side, in the nitrogen adsorption isotherm, the adsorption amount difference AV 0.95-0.99 becomes larger.

[0087] The present inventors have researched a method of manufacturing the dendritic carbon nanostructure in order to produce the dendritic carbon nanostructure satisfying the above-described conditions. Roughly speaking, the dendritic carbon nanostructure is produced by ejecting a metal from an acetylene metal. Among them, the acetylene metal is produced by adding acetylene to a metal salt solution while irradiating ultrasonic waves thereto. In this process, a nucleus of the acetylene metal is generated in the solution, and the acetylene metal grows from the nucleus as a starting point. The metal salt solution is produced by dissolving a metal salt in an ammonia solution. Then, the present inventors have greatly reduced the concentration of the metal salt and increased the concentration ratio of ammonia to the anion component of the metal salt in comparison with the conventional procedure in order to reduce the concentration of the nucleus of the acetylene metal generated in the metal salt solution. Also, the addition speed of acetylene is reduced in order to make the generation (growth) reaction of the acetylene metal proceed slowly. Also, the concentration of the dissolved acetylene gas is reduced in order to make the generation reaction of the acetylene metal proceed slowly. Specifically, the metal salt solution is subjected to heat retention. That is, the temperature at the time of the generation reaction of the acetylene metal is maintained high.

[0088] Moreover, the present inventors have caused the metal to be ejected from the acetylene group after compression molding of the acetylene metal. That is, by heating the acetylene metal, explosive self-decomposition reaction of the acetylene metal occurs, and the metal is ejected from the acetylene metal. Thereby, a mixture of a metal layer and a carbon layer, that is, an intermediate (hereinafter also referred to as "carbon nanometer intermediate") is generated. Thus, the present inventors have caused the reaction heat to be difficult to escape from the acetylene metal by compression molding of the acetylene metal. Thereby, the reaction heat can be efficiently supplied to the unreacted portion of the acetylene metal. That is, in principle, the reaction heat of the acetylene metal itself does not change. Thus, by causing such reaction heat to be difficult to escape to the outside, the amount of heat supplied to the unreacted portion of the acetylene metal is relatively increased, and the temperature increase of the acetylene metal is made more smooth. Moreover, the present inventors have increased the temperature increase rate of the acetylene metal in order to make the temperature increase of the acetylene metal more smooth. Thereby, the reaction of the acetylene metal is promoted.

[0089] This result is that, in the dendritic carbon nanometer structure as a final product, the intrabranch fine pores can be made coarser, and the branching of the intrabranch fine pores is further increased. Specifically, the area ΔS 0.4-0.87 is 5 mL / g to 50 mL / g. Moreover, the skeleton of the dendritic structure can be maintained, and the branches can be thinned. The distribution of the branch diameter also becomes narrow. Specifically, in the nitrogen adsorption isotherm, the adsorption amount difference ΔV 0.95-0.99 of the adsorption amount at P / P0 = 0.99 and the adsorption amount at P / P0 = 0.95 becomes 500 mL / g to 1100 mL / g. It is also possible to confirm, using SEM (scanning electron microscope), that the skeleton of the dendritic structure of the dendritic carbon nanometer structure is maintained, and the branches are thinned.

[0090] Moreover, the present inventors have increased the durability of the dendritic carbon nanometer structure by treating the dendritic carbon nanometer structure at a high temperature. The present inventors have obtained the carbon material for a catalyst support of the present disclosure based on the above knowledge.

[0091] Here, in the present disclosure, showing two hysteresis loops means that it is shown that there are two closed hysteresis loops. The state in which the hysteresis loop is closed indicates that, in the start point and the end point of the hysteresis loop, the value obtained by subtracting the adsorption amount indicated by the adsorption isotherm from the adsorption amount indicated by the desorption isotherm (ΔV) is 20 mL / g or less.

[0092] In addition, showing two hysteresis loops includes the following modes.

[0093] (1) A mode in which the end point of the first hysteresis loop and the start point of the second hysteresis loop are the same. That is, a mode in which the relative pressure of the end point of the first hysteresis loop and the start point of the second hysteresis loop is the same.

[0094] (2) A mode in which the end point of the first hysteresis loop and the start point of the second hysteresis loop are separated. That is, a mode in which the relative pressure of the end point of the first hysteresis loop and the start point of the second hysteresis loop is different.

[0095] For the carbon material for catalyst support of the present disclosure, it is preferable that, in two hysteresis loops, a first hysteresis loop exists in a range of relative pressure P / P0 of 0.87 or more, and a second hysteresis loop exists in a range of relative pressure P / P0 of 0.4 to 0.87. By the existence of the two hysteresis loops in the ranges, it is easy to reduce overvoltage at a large current. In the same aspect, for the carbon material for catalyst support of the present disclosure, it is more preferable to satisfy the following conditions.

[0096] A more preferable mode of the carbon material for catalyst support of the present disclosure is to satisfy the following conditions (A), (B), and (C). Furthermore, in the present disclosure, the nitrogen adsorption-desorption isotherm is measured at a liquid nitrogen temperature (for example, 77 K).

[0097] (A) The specific surface area obtained by the BET analysis of the nitrogen adsorption isotherm is 450 m 2 / g to 1500 m 2 / g.

[0098] (B) In terms of the nitrogen adsorption-desorption isotherm, a first hysteresis loop is formed in a range of relative pressure P / P0 of 0.87 or more, and in a range of relative pressure P / P0 of 0.87 ± 0.03, a minimum value ΔV 0.87 of the adsorption amount difference ΔV 0.87min of the first hysteresis loop is 20 mL / g or less.

[0099] (C) In terms of the nitrogen adsorption-desorption isotherm, a second hysteresis loop is formed in a range of relative pressure P / P0 of 0.4 to 0.87, and an area ΔS 0.4-0.87 of the second hysteresis loop is 5 to 50 mL / g.

[0100] (1-1. Condition (A))

[0101] The specific surface area obtained by the BET analysis of the nitrogen adsorption isotherm is 450 m 2 / g to 1500 m 2 / g. By this, it is possible to support more catalyst components. Among them, if the BET specific surface area is less than 450 m 2 / g, there occurs a case where the supportability of the catalyst components is reduced. In a case where the BET specific surface area exceeds 1500 m 2 / g, there is a case where it is difficult to simultaneously achieve the physical strength and the oxidation consumption durability of the carbon material for catalyst support. The preferable lower limit value of the BET specific surface area is 500 m 2 / g or more, and the preferable upper limit value of the BET specific surface area is 1450 m 2 / g or less. Furthermore, the BET specific surface area can be measured by the method described in the Examples described later.

[0102] (1-2. Condition (B))

[0103] In terms of the nitrogen adsorption / desorption isotherm, a first hysteresis loop is shown in a range where the relative pressure P / P0 is 0.87 or more, and the minimum value AV 0.87 of the adsorption amount difference AV 0.87min of the first hysteresis loop in a range where the relative pressure P / P0 is 0.87 ± 0.03 is 20 mL / g or less. Furthermore, AV 0.87 is measured by the method described in the Examples described later.

[0104] The "minimum value AV 0.87 of the adsorption amount difference AV 0.87min of the first hysteresis loop in a range where the relative pressure P / P0 is 0.87 ± 0.03 is 20 mL / g or less" means that the first hysteresis loop is closed in a range where the relative pressure P / P0 = 0.87 ± 0.03. Here, the adsorption amount difference AV 0.87 of the first hysteresis loop is a value obtained by subtracting the adsorption amount shown by the adsorption isotherm from the desorption amount shown by the desorption isotherm. Furthermore, the minimum value AV 0.87 of the adsorption amount difference AV 0.87min in a range where the relative pressure P / P0 is 0.87 ± 0.03 is 20 mL / g or less. That is, the minimum value AV 0.87 of the adsorption amount difference AV 0.87min is 0 to 20 mL / g. Figure 1 In the present disclosure, the adsorption amount difference AV 0.87 is the smallest at the relative pressure P / P0 of 0.87 (see Figure 1 ).

[0105] (1-3. Condition (C))

[0106] In the nitrogen adsorption / desorption isotherm at the liquid nitrogen temperature, a second hysteresis loop is shown in a range where the relative pressure P / P0 is 0.4 to 0.87, and the area AS 0.4-0.87 of the second hysteresis loop is 5 mL / g to 50 mL / g (see Figure 1 ). The preferable lower limit value of AS 0.4-0.87 is 7 mL / g or more, and the preferable upper limit value of AS 0.4-0.87 is 45 mL / g or less. Furthermore, AS 0.4-0.87 is measured by the method described in the Examples described later.

[0107] In the conventional dendritic carbon nanostructure, the intradendritic pores have a thick and long cylindrical shape, and branch diversely within the dendrite. In the carbon material for a catalyst support of the present disclosure, the characteristics of such interdendritic pores are maintained, and the intradendritic pores become thicker, and the branching property of the intradendritic pores is further increased. Therefore, the second hysteresis loop becomes very small. Specifically, the area AS 0.4-0.87The concentration is 5–50 mL / g. This reduces gas diffusion resistance and decreases overvoltage at high currents. Furthermore, the area ΔS of the second hysteresis loop... 0.4-0.87 When the concentration is below 50 mL / g, the volume of the bottleneck-type micropores relative to the total volume of the micropores within the branch decreases. Therefore, a reduction in gas diffusion resistance is preferred. The area ΔS of the second hysteresis ring... 0.4-0.87 When the concentration is 5 mL / g or higher, the reduction in strength of the carbon material used as the catalyst support is suppressed. As a result, the overvoltage at high currents is reduced, which is preferable.

[0108] (1-4. Condition (D))

[0109] For carbon materials used as catalyst supports, in addition to satisfying conditions (A) to (C), it is preferable to also satisfy the following condition (D). Condition (D) is the difference ΔV between the adsorption amount at relative pressure P / P0 = 0.99 and the adsorption amount at relative pressure P / P0 = 0.95 in the nitrogen adsorption isotherm. 0.95-0.99 500 mL / g to 1100 mL / g (reference) Figure 1 ). ΔV 0.95-0.99 The preferred lower limit is above 550 mL / g, ΔV 0.95-0.99 The preferred upper limit is below 950 mL / g. Furthermore, ΔS 0.4-0.87 The determination can be performed using the methods described in the examples below.

[0110] As described above, in dendritic carbon nanostructures, the interbranch pores have a coarse and elongated cylindrical shape and branch in various ways between the branches. In the carbon material for catalyst support disclosed herein, such interbranch pore characteristics are maintained, and the interbranch pores become coarser, with a smaller diameter deviation. Therefore, in the nitrogen adsorption isotherm, the adsorption amount difference ΔV between the relative pressure P / P0 = 0.99 and the relative pressure P / P0 = 0.95 is significantly reduced. 0.95-0.99 The diameter of the interbranch pores increases. Specifically, the distribution of pore diameters becomes more concentrated on the longer axis. In other words, the adsorption difference ΔV... 0.95-0.99 The concentration changes to 500 mL / g to 1100 mL / g. Consequently, the gas diffusion resistance decreases, and the overvoltage at high currents decreases. This is due to the adsorption capacity difference ΔV. 0.95-0.99 At concentrations above 500 mL / g, the diameter distribution of the interbranch pores narrows, significantly reducing diffusion resistance. This makes it extremely difficult to create an adsorption capacity difference ΔV. 0.95-0.99 Carbon materials used as catalyst supports exceeding 1100 mL / g.

[0111] (1-5. Condition (E))

[0112] For carbon materials used as catalyst supports, in addition to satisfying conditions (A) to (C) or conditions (A) to (D), it is preferable to also satisfy the requirement that the catalyst exhibits a light output at 1500 cm⁻¹ as determined by Raman spectroscopy.-1 ~ 1700 cm -1 The half width of the G band detected in the range of 1600 cm -1 ~ 70 cm -1 Condition (E). In this case, the oxidation resistance of the carbon material for a catalyst support is increased, and the durability is improved. That is, even if the start-stop of the solid polymer fuel cell using the carbon material for a catalyst support is repeated, the carbon material for a catalyst support is hardly oxidized and consumed.

[0113] wherein, in the case where the half width of the G band is 45 cm -1 ~ 70 cm -1 The degree of graphitization (crystallinity) of the carbon material for a catalyst support is increased. Therefore, the oxidation resistance is improved. In the case where the half width of the G band exceeds 70 cm -1 The oxidation resistance of the carbon material for a catalyst support is decreased, and as a result, the overvoltage at a large current is increased. In the case where the half width of the G band is less than 45 cm -1 The shape of the intrabranch pores and the interbranch pores is easily maintained, and as a result, it is possible that the conditions (B) and (C) are not satisfied. In terms of the half width of the G band, the preferable lower limit value is 48 cm -1 The preferable upper limit value is 68 cm -1 or less. Further, the half width of the G band is measured by the method described in the following examples.

[0114] <2. Method for producing carbon material for catalyst support>

[0115] Next, an example of the preferable method for producing the carbon material for a catalyst support will be described. According to the preferable method for producing the carbon material for a catalyst support, the nitrogen adsorption isotherm shows two hysteresis loops in the range of a relative pressure of 0.4 or more, and the conditions (A) to (E) can be satisfied. The example of the preferable method for producing the carbon material for a catalyst support includes an acetylene metal generation step, a first heat treatment step, and a second heat treatment step. Further, a cleaning treatment step can be included between the first heat treatment step and the second heat treatment step.

[0116] (2-1. Acetylene metal generation step)

[0117] In this step, acetylene is added to the metal salt solution while irradiating ultrasonic waves, thereby producing acetylene metal. According to this step, a precipitate of acetylene metal (for example, acetylene silver, acetylene copper, or the like) is generated in the metal salt solution.

[0118] As the metal salt, for example, silver nitrate, copper chloride, etc. can be given. Among them, silver nitrate is preferred. Further, the metal salt solution can be obtained by dissolving the metal salt in an aqueous ammonia solution. Further, as the method of adding acetylene, for example, a method of blowing acetylene gas into the metal salt solution can be given. The ultrasonic wave irradiation can be performed, for example, using an ultrasonic vibrator or an ultrasonic cleaner, etc. Further, before the addition of the acetylene gas, it is preferred to replace oxygen in the metal salt solution with an inert gas. This is to suppress an explosive decomposition reaction of the acetylene metal unexpectedly. The time of blowing the inert gas into the metal salt solution is about 40 to 60 minutes. As the inert gas, for example, argon, nitrogen, etc. can be given.

[0119] Further, in the present disclosure, in order to reduce the concentration of the nucleus of the acetylene metal generated in the metal salt solution, the concentration of the metal salt is greatly reduced compared to the conventional process, and the concentration ratio of ammonia to the anion component of the metal salt is increased. The concentration of the metal salt differs depending on the kind of the metal used, and thus it can be determined by experiment. The same applies to the concentration ratio of ammonia. That is, in the case where the concentration of the metal salt is high or in the case where the concentration ratio of ammonia is low, the condition (B) cannot be satisfied, and furthermore, any one of the conditions (A), (C), and (D) cannot be satisfied.

[0120] For example, in the case where the metal salt is silver nitrate, the concentration of the metal salt is less than 0.1 to 5 mass% with respect to the mass of the metal salt solution. It is preferred that the upper limit of the concentration of the metal salt is 2 mass% or less. Further, the concentration ratio of ammonia to the anion component of the metal salt is 1 or more, and it is preferred to be 3 or more. Thus, a carbon material for a catalyst support satisfying the above conditions (A) to (D) can be produced. In the case where the concentration of the metal salt is 0.1 mass% or more, the strength of the carbon material for a catalyst support as the final product is ensured. On the other hand, in the case where the concentration of the metal salt is less than 5 mass% or in the case where the concentration ratio of ammonia to the anion component of the metal salt is 1 or more, the concentration of the nucleus of the acetylene metal generated in the metal salt solution is moderate. As a result, the difference in the adsorption amount ΔV 0.95-0.99 becomes 500 to 1100 mL / g. That is, the distribution of the diameter of the interbranch fine pores becomes narrow, and the gas diffusion resistance becomes low. Further, the first hysteresis loop is easily closed around the relative pressure P / P0 = 0.87.

[0121] Further, in the present disclosure, in order to make the generation (growth) reaction of acetylene metal slow, the addition rate of acetylene is reduced. Specifically, for example, acetylene gas is slowly blown into the metal salt solution. The specific blowing rate varies depending on the kind of metal used, and thus is determined according to experiments. That is, in the case where the blowing rate of acetylene is high, the condition (B) cannot be satisfied, and further, any one of the conditions (A), (C), and (D) is not satisfied. For example, in the case where the metal salt is silver nitrate, the upper limit of the blowing rate of acetylene gas is less than 20 mL / min. The blowing rate of acetylene gas is preferably 15 mL / min or less, and more preferably 10 mL / min or less. In the case where the blowing rate of acetylene gas is less than 20 mL / min, the area ΔS 0.4-0.87 of the second hysteresis loop is 50 mL / g or less. That is, the thickness and the branching of the intradendritic pores are sufficient. Further, the volume of the bottle-neck type pores decreases in the total volume of the intradendritic pores, and the first hysteresis loop easily closes around the relative pressure of 0.87. The lower limit of the blowing rate of acetylene gas is not particularly limited, and can be 0.5 mL / min or more.

[0122] Further, in the present disclosure, in order to reduce the concentration of the dissolved acetylene gas, the reaction rate of acetylene metal is reduced, and the metal salt solution is heat-insulated. The specific temperature varies depending on the kind of metal used, and thus is determined according to experiments. That is, in the case where the temperature is high, the condition (B) cannot be satisfied, and further, any one of the conditions (A), (C), and (D) is not satisfied. For example, in the case where the metal salt is silver nitrate, the metal salt solution is heat-insulated at more than 10°C. The temperature of the metal salt solution is preferably 20°C or more, and more preferably 60°C or more. In the case where the temperature of the metal salt solution is more than 10°C, the adsorption difference ΔV 0.95-0.99 becomes 500 mL / g to 1100 mL / g. That is, the distribution of the diameters of the interdendritic pores becomes narrow, and the gas diffusion resistance becomes low. Further, the area ΔS 0.4-0.87 of the second hysteresis loop is 50 mL / g or less. That is, the thickness and the branching of the intradendritic pores are sufficient. Further, the volume of the bottle-neck type pores decreases in the total volume of the intradendritic pores, and the first hysteresis loop easily closes around the relative pressure of 0.87. The upper limit of the temperature is not particularly limited, and can be 70°C or less.

[0123] (2-2. First heat treatment step)

[0124] In the first heat treatment step, the metal is ejected from the acetylene metal by heating the acetylene metal. In the present disclosure, the acetylene metal is heated after being compression-molded. That is, by heating the acetylene metal, the acetylene metal generates an explosive self-decomposition reaction, and the metal is ejected from the acetylene metal. Thereby, a carbon nanometer intermediate is generated. In the present disclosure, by compression-molding the acetylene metal, the reaction heat is difficult to escape from the acetylene metal. Thereby, the reaction heat can be effectively supplied to the unreacted portion of the acetylene metal. That is, the reaction heat of the acetylene metal itself does not change in principle. Thus, in the present disclosure, by making such reaction heat difficult to escape to the outside, the heat supplied to the unreacted portion of the acetylene metal is relatively high, and the acetylene metal is warmed up more smoothly. In the present disclosure, in order to make the warming up of the acetylene metal more smooth, the warming up speed of the acetylene metal is increased. That is, in the present disclosure, by performing these processes, the reaction of the acetylene metal is promoted.

[0125] The molding pressure at the time of compression-molding varies depending on the kind of metal constituting the acetylene metal, and thus can be determined based on experiments. That is, in the case where the molding pressure is insufficient, the reaction heat cannot be effectively supplied to the unreacted portion of the acetylene metal. As a result, the condition (B) cannot be satisfied, and any one of the conditions (A), (C), and (D) is not satisfied. For example, in the case where the acetylene metal is acetylene silver, the molding pressure is greater than 0.5 kg / cm 2 . Thereby, the reaction heat can be effectively supplied to the unreacted portion of the acetylene metal. The molding pressure is preferably 1.0 kg / cm 2 or more, and more preferably 2.0 kg / cm 2 or more. Further preferably, the molding pressure is 2.5 kg / cm 2 or more. In the case where the molding pressure is less than 0.5 kg / cm 2 , the reaction heat cannot be effectively supplied to the unreacted portion of the acetylene metal. In addition, the first hysteresis loop is easily closed around the relative pressure of 0.87. The upper limit value of the molding pressure is not particularly limited, and can be 3 kg / cm 2 or less.

[0126] Further, the temperature increase rate can vary depending on the kind of metal constituting the acetylene metal, and therefore is determined based on experiments. That is, in the case where the temperature increase rate is low, the reaction heat cannot be effectively supplied to the unreacted portion of the acetylene metal. As a result, the condition (B) cannot be satisfied, and any one of the conditions (A), (C), and (D) is not satisfied. For example, in the case where the acetylene metal is silver acetylide, the temperature increase rate is greater than 2°C / min. The temperature increase rate is preferably 5°C / min or greater, and more preferably 100°C / min or greater. In the case where the temperature increase rate is greater than 2°C / min, the reaction heat can be effectively supplied to the unreacted portion of the acetylene metal. As a result, for example, the condition (C) can be satisfied. The upper limit of the temperature increase rate is not particularly limited, and can be 300°C / min or less.

[0127] Further, the holding temperature after the temperature increase varies depending on the kind of metal constituting the acetylene metal. For example, in the case where the silver acetylide is heated, the holding temperature can be 160°C to 200°C. Further, in the case where the copper acetylide is heated, the holding temperature can be 210°C to 250°C.

[0128] (2-3. Washing treatment step)

[0129] In the washing treatment step, the carbon material for catalyst support is produced by removing the metal component from the carbon nanointermediate. Specifically, for example, the carbon nanointermediate is washed with nitric acid (or nitric acid aqueous solution). Thereby, the metal component and the unstable carbon compound in the carbon nanointermediate are dissolved into the nitric acid aqueous solution. Thereby, not only the metal component but also the unstable carbon compound can be removed from the carbon nanointermediate. Further, by this step, the position where the metal component existed becomes a void, and therefore the carbon material for catalyst support has a three-dimensional structure with a large specific surface area.

[0130] (2-4. Second heating treatment step)

[0131] In the second heating treatment step, the carbon material for catalyst support is heated at 1400°C to 2500°C in a vacuum or an inert gas atmosphere. The heating temperature is preferably 1600°C to 2100°C. If the heating temperature in this second heating step is 1400°C or greater, the crystallinity of the carbon material for catalyst support is improved, and the durability under the fuel cell use environment is easily ensured. As a result, the condition (E) is easily satisfied. Further, if the heating temperature is 2500°C or less, the shape of the intradendritic fine pores and the interdendritic fine pores is easily maintained. As a result, the conditions (B) and (C) are easily satisfied. From the viewpoint of improving the durability of the carbon material for catalyst support, the second heating treatment is preferably performed.

[0132] Furthermore, the heat treatment time also affects the properties of carbon materials used as catalyst supports. A heat treatment time of 20 to 200 minutes is generally acceptable. It is believed that the heat treatment time is related to the uniformity of crystallinity. If the heat treatment time is too short, only the outer surface of the carbon material exhibits high crystallinity, while the internal crystallinity is not fully developed, leading to a decrease in durability. If the heat treatment time is 20 minutes or more, the internal crystallinity increases. On the other hand, if the heat treatment time is too long, it is associated with a decrease in industrial productivity. If the heat treatment time is substantially less than 200 minutes, any carbon material can achieve uniform internal crystallization.

[0133] The heating method for the carbon material used as a catalyst support is not particularly limited as long as it can heat the carbon material to the aforementioned temperature. Examples of heating methods include resistance heating, microwave heating, high-frequency heating, and furnace-type heating methods. Regarding the furnace type, examples include batch furnaces and tunnel furnaces. There are no restrictions on the furnace type if it can achieve an inactive environment or a reduced-pressure environment.

[0134] In summary, through the preferred manufacturing method described above, the carbon material for catalyst support disclosed herein satisfies conditions (A) to (C). Furthermore, by selecting preferred conditions, at least one of conditions (D) and (E) is satisfied. Therefore, by using this carbon material for catalyst support as a catalyst support in a solid polymer fuel cell, the gas diffusion resistance in the catalyst layer is reduced, and the overvoltage at high currents is decreased.

[0135] <3. Catalyst layer for solid polymer fuel cells and solid polymer fuel cells>

[0136] Solid polymer fuel cells are described together with catalyst layers for solid polymer fuel cells. The carbon material used as the catalyst support in this disclosure is, for example, applicable to... Figure 3 Catalyst layers 150 and 160 are provided in the solid polymer fuel cell 100 shown. Figure 3 This is a schematic diagram illustrating an example of the general configuration of a fuel cell according to the present disclosure. The solid polymer fuel cell 100 has membranes 110 and 120, gas diffusion layers 130 and 140, catalyst layers 150 and 160, and an electrolyte membrane 170.

[0137] The diaphragm 110 is the anode-side diaphragm, which introduces reducing gases such as hydrogen into the gas diffusion layer 130. The diaphragm 120 is the cathode-side diaphragm, which introduces oxidizing gases such as oxygen and air into the gas diffusion condensate phase. There are no particular restrictions on the types of diaphragms 110 and 120; any diaphragm used in conventional fuel cells (such as solid polymer fuel cells) is acceptable.

[0138] The gas diffusion layer 130 is a gas diffusion layer on the anode side, and diffuses the reducing gas supplied from the separator 110 to the catalyst layer 150. The gas diffusion layer 140 is a gas diffusion layer on the cathode side, and diffuses the oxidizing gas supplied from the separator 120 to the catalyst layer 160. The kind of the gas diffusion layers 130, 140 is not particularly limited, and can be the same as the gas diffusion layer used in the conventional fuel cell (for example, a solid polymer fuel cell). As examples of the gas diffusion layers 130, 140, a porous carbon material (carbon cloth, carbon paper, etc.), a porous metal material (metal mesh, metal wool, etc.), and the like can be given. Further, as a preferable example of the gas diffusion layers 130 and 140, a two-layer structure gas diffusion layer can be given. Specifically, in the gas diffusion layers 130, 140, the layer on the side of the separators 110, 120 can be a gas diffusion fiber layer in which a fibrous carbon material is a main component, and the layer on the side of the catalyst layers 150, 160 can be a microporous layer in which carbon black is a main component.

[0139] The catalyst layer 150 is a so-called anode. In the catalyst layer 150, an oxidation reaction of the reducing gas occurs, and protons and electrons are generated. For example, in the case where the reducing gas is hydrogen, the following oxidation reaction occurs.

[0140] H2→ 2H + + 2e - (E0= 0 V)

[0141] The protons generated by the oxidation reaction reach the catalyst layer 160 via the catalyst layer 150 and the electrolyte membrane 170. The electrons generated by the oxidation reaction reach the outside circuit via the catalyst layer 150, the gas diffusion layer 130, and the separator 110. The electrons do work (generate electricity) in the outside circuit and are then introduced into the separator 120. Thereafter, the electrons reach the catalyst layer 160 via the separator 120 and the gas diffusion layer 140.

[0142] The constitution of the catalyst layer 150 that becomes the anode is not particularly limited. The constitution of the catalyst layer 150 can be the same constitution as the conventional anode, can be the same constitution as the catalyst layer 160, or can be a constitution in which the hydrophilicity is higher than that of the catalyst layer 160.

[0143] The catalyst layer 160 is a so-called cathode. In the catalyst layer 160, a reduction reaction of the oxidizing gas occurs, and water is generated. For example, in the case where the oxidizing gas is oxygen or air, the following reduction reaction occurs. The water generated by the oxidation reaction is discharged to the outside of the solid polymer fuel cell 100 together with the unreacted oxidizing gas.

[0144] O2+ 4H + + 4e - → 2H2O (E0= 1.23 V)

[0145] As such, in the solid polymer fuel cell 100, power generation is performed using the energy difference (potential difference) between the oxidation reaction and the reduction reaction. In other words, the electrons generated by the oxidation reaction do work in the external circuit.

[0146] It is preferable that the catalyst layer 160 contain the carbon material for catalyst support according to the present disclosure. That is, the catalyst layer 160 contains the carbon material for catalyst support according to the present disclosure, the electrolyte material, and the catalyst component. Thereby, the catalyst utilization rate in the catalyst layer 160 can be improved. Also, the catalyst utilization rate of the solid polymer fuel cell 100 can be improved.

[0147] Further, the catalyst loading rate in the catalyst layer 160 is not particularly limited, and is preferably 30 mass% or more and less than 80 mass%. If the catalyst loading rate is in this range, the catalyst utilization rate is further increased. Here, the catalyst loading rate is expressed as mass% of the catalyst component with respect to the total mass of the catalyst support particles (particles on which the catalyst component is supported on the carbon material for catalyst support). In the case where the catalyst loading rate is less than 30 mass%, in order for the solid polymer fuel cell 100 to be tolerable in practice, it can be necessary to thicken the catalyst layer 160. On the other hand, in the case where the catalyst loading rate is 80 mass% or more, catalyst agglomeration is easily caused. In addition, there is a possibility that the catalyst layer 160 becomes too thin, causing water flooding.

[0148] The mass ratio I / C of the mass I of the electrolyte material to the mass C of the carbon material for catalyst support in the catalyst layer 160 is not particularly limited, and is preferably more than 0.5 and less than 5.0. In this case, the fine pore network and the electrolyte material network can be taken into account, and the catalyst utilization rate is increased. On the other hand, in the case where the mass ratio I / C is 0.5 or less, there is a tendency that the electrolyte material network becomes weak and the proton conduction resistance is increased. In the case where the mass ratio I / C is 5.0 or more, there is a possibility that the fine pore network is cut off by the electrolyte material. In either case, there is a possibility that the catalyst utilization rate is decreased.

[0149] In addition, 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 in the catalyst layer 160, and water flooding is not easily caused. In the case where the thickness of the catalyst layer 160 is 5 μm or less, water flooding is easily caused. In the case where the thickness of the catalyst layer 160 is 20 μm or more, the oxidizing gas does not easily diffuse in the catalyst layer 160, and the catalyst component near the electrolyte membrane 170 does not easily function. That is, there is a possibility that the catalyst utilization rate is decreased.

[0150] The electrolyte membrane 170 is composed of an electrolyte material having proton conductivity. The electrolyte membrane 170 introduces protons generated in the above-described oxidation reaction to the catalyst layer 160 (cathode). The kind of electrolyte material is not limited, and any electrolyte material used in a conventional fuel cell, such as a solid polymer fuel cell, can be used. As examples of suitable electrolyte materials, electrolyte resins can be given. As electrolyte resins, for example, polymers into which phosphoric acid groups, sulfonic acid groups, or the like are introduced can be given. Specifically, for example, perfluorosulfonic acid polymers, polymers into which benzenesulfonic acid is introduced, or the like can be given. Of course, the electrolyte material can be another kind of electrolyte material. As such electrolyte materials, for example, inorganic electrolyte materials, inorganic-organic hybrid electrolyte materials, or the like can be given. Note that the solid polymer fuel cell 100 can also be a fuel cell that operates in a range of normal temperature (25°C) to 150°C.

[0151] <4. Method for manufacturing solid polymer fuel cell>

[0152] The method for manufacturing the solid polymer fuel cell 100 is not particularly limited, and a conventional manufacturing method can be used. However, the catalyst carrier uses the carbon material for catalyst carrier according to the present disclosure. In the catalyst layers 150, 160, the catalyst carrier in at least the catalyst layer 160 as the cathode is preferably the carbon material for catalyst carrier according to the present disclosure. Of course, the catalyst carrier in both of the catalyst layers of the catalyst layer 150 as the anode and the catalyst layer 160 as the cathode can also be the carbon material for catalyst carrier according to the present disclosure.

[0153] Example

[0154] <1. Measurement method of each parameter>

[0155] Next, experimental examples of the carbon material for catalyst carrier according to the present disclosure will be described. First, the measurement method of each parameter will be described.

[0156] (1-1. Measurement method of nitrogen adsorption-desorption isotherm)

[0157] About 30 mg of the sample was weighed, and vacuum drying was performed at 120°C for 2 hours. Next, the sample was placed in an automatic specific surface area measurement device (BELSORP MAX manufactured by Microtrac BEL Corporation). Then, nitrogen was used as the adsorbate, and the nitrogen adsorption-desorption isotherm was measured at a measurement temperature of 77K (liquid nitrogen temperature). The BET analysis was performed on the nitrogen adsorption isotherm in a range of relative pressure P / P0 of 0.05 to 0.15, and thus the BET specific surface area was calculated.

[0158] Furthermore, the number of hysteresis loops present in the range of relative pressure P / P0 above 0.4 was measured. Then, the adsorption difference ΔV in the first hysteresis loop within the range of relative pressure P / P0 0.87 ± 0.03 was determined. 0.87 Determine the adsorption difference ΔV 0.87 The minimum value ΔV 0.87min Furthermore, the area ΔS of the second hysteresis loop was measured within the relative pressure P / P0 range of 0.4 to 0.87. 0.4-0.87 For this ΔV 0.87 The determination of ΔS 0.4-0.87 The purpose of this measurement is to make the P / P0 measurement interval smaller than that of a typical measurement. Specifically, the P / P0 measurement interval is set using a 0.005 scale as the fixed point. That is, the measurement accuracy of P / P0 is 0.005.

[0159] The area ΔS of the hysteresis loop 0.4-0.87 The calculation uses the following method. At each relative pressure P / P0 measurement point, the difference ΔV between the adsorption amount V1 during the adsorption process and the adsorption amount V2 during the desorption process is calculated and multiplied by 0.005 of the P / P0 measurement interval. Then, the value obtained by integrating over all measurement points in the range of P / P0 = 0.4 to 0.87 is calculated, and this value is taken as ΔS. 0.4-0.87 .

[0160] Similarly, within a relative pressure range of P / P0 of 0.87 ± 0.03, the adsorption difference ΔV was calculated with a precision of 0.005. 0.87 .

[0161] Furthermore, the adsorption capacity difference ΔV between the adsorption capacity at relative pressure P / P0 = 0.99 and the adsorption capacity at relative pressure P / P0 = 0.95 was measured. 0.95-0.99 The results are shown in Table 2.

[0162] Furthermore, if the first hysteresis loop exists in the range where the relative pressure P / P0 is 0.87 or higher, and the second hysteresis loop exists in the range where the relative pressure P / P0 is 0.4 to 0.87, then "2" is recorded in the "Number of Hysteresis Loops" column of Table 2. If only one hysteresis loop exists, then "1" is recorded in the "Number of Hysteresis Loops" column of Table 2.

[0163] (1-2. Methods for Raman Spectroscopy)

[0164] Approximately 3 mg of the sample was measured, and the Raman spectra were determined using a laser Raman spectrophotometer (Nippon Spectrophotometer Co., Ltd., NRS-3100 model). The 1500–1700 cm⁻¹ band, referred to as the G band, was extracted from the Raman spectra obtained under the following measurement conditions. -1The peak width at half height (ΔG) of the peak was measured. The results are shown in Table 2.

[0165] - Measurement conditions -

[0166] Excitation laser 532 nm, laser power 10 mW (sample irradiation power: 1.1 mW), microscopic configuration: Backscattering, objective lens: x 100, spot diameter: 1 μm, exposure time: 30 sec, observation wave number: 2000 cm -1 ~ 300 cm -1 , cumulative number: 6 times.

[0167] <2. Preparation of carbon material for catalyst support>

[0168] By performing each operation in the following procedures, the carbon materials for catalyst support E1-1 to E6-8 and C1-1 to C6-2 were produced.

[0169] (2-1. Ethyne metal generation step)

[0170] First, silver nitrate was dissolved in an aqueous ammonia solution to produce a silver nitrate solution. The ammonia concentration in the silver nitrate solution was 2.0 mass%. In addition, the specific concentration of silver nitrate is shown in Table 1.

[0171] To investigate the effect of the silver nitrate concentration, the silver nitrate concentration was varied for each carbon material when producing the carbon materials for catalyst support E1-1 to E1-6 and C1-1 to C1-2.

[0172] Next, argon gas was blown into the silver nitrate solution for 60 minutes to replace the oxygen dissolved in the silver nitrate solution with an inert gas. Next, while stirring the silver nitrate solution with a stirrer, acetylene gas was blown into the solution at a predetermined flow rate. In addition, the silver nitrate solution was kept at a predetermined temperature while blowing in the acetylene gas. The specific values of the temperature of the silver nitrate solution and the blowing flow rate of the acetylene gas are shown in Table 1.

[0173] To investigate the effect of the silver nitrate solution temperature, the temperature of the silver nitrate solution was varied for each carbon material when producing the carbon materials for catalyst support E2-1 to E2-5 and C2-1. In addition, to investigate the effect of the blowing flow rate of the acetylene gas, the blowing flow rate was varied for each carbon material when producing the carbon materials for catalyst support E3-1 to E3-5 and C3-1 to C3-2.

[0174] As a result, white solid of silver acetylide is precipitated in the silver nitrate solution. If the acetylene gas blown in is released in the form of bubbles directly in the solution, it is judged that the silver nitrate in the silver nitrate solution has reacted with the acetylene gas completely, and the blowing is ended. Next, the precipitate in the silver nitrate solution is filtered with a septum filter, and the obtained precipitate is dispersed again in methanol. Next, the methanol dispersion of the precipitate is filtered again. The precipitate is taken out to a petri dish, and the precipitate is immersed in a small amount of methanol.

[0175] (2-2. First heating treatment step)

[0176] Next, about 0.5 g (mass including methanol) of the precipitate (i.e., silver acetylide) immersed in methanol is weighed. Next, the weighed precipitate is put into a particle molding mold made of stainless steel with a diameter of 5 mm, and is slowly pressurized at a predetermined pressure (molding pressure). Thus, a particle of silver acetylide is produced. The specific molding pressure is shown in Table 1.

[0177] Again, in order to investigate the influence of the molding pressure, when producing the carbon materials E4-1 to E4-4 and C4-1 to C4-3 for catalyst supports, the molding pressure is changed for each carbon material. Further, the molding pressure of the carbon material C4-1 for catalyst support is "none" meaning that compression is not performed. For E4-3 to E4-4, the acetylene silver synthesis conditions are also changed in addition to the molding pressure.

[0178] Next, the particle of silver acetylide is moved to a cylindrical container made of stainless steel with a diameter of about 5 cm, and the cylindrical container is put into a vacuum heating electric furnace, and vacuum drying is performed at 60°C for about 15 minutes to 30 minutes.

[0179] Next, the silver acetylide in the cylindrical container is warmed up to 200°C at a predetermined warming-up rate in the current state thereof (i.e., without taking out the silver acetylide from the cylindrical container). In this process, the silver acetylide generates an explosive self-decomposition reaction, and silver is spouted from the silver acetylide. Thus, a mixture of a metal layer and a carbon layer, i.e., a carbon nanointermediate is generated. The specific warming-up rate is shown in Table 1. Here, in order to investigate the influence of the warming-up rate, when producing the carbon materials E5-1 to E5-6 and C5-1 for catalyst supports, the warming-up rate is changed for each carbon material.

[0180] (2-3. Cleaning treatment step)

[0181] The carbon nanointermediate obtained in the above first heating treatment step is cleaned with concentrated nitric acid with a concentration of 60 mass%. Thus, silver particles and other unstable carbon compounds remaining on the surface of the carbon nanointermediate are removed. Thereafter, the carbon material for catalyst support is washed with water.

[0182] (2-4. Second heating treatment step)

[0183] The catalyst support washed by the washing treatment step was heated with the carbon material at a predetermined temperature for 2 hours, whereby the graphitization degree of the catalyst support carbon material was increased. The heating temperature and the heating treatment time are shown in Table 1.

[0184] Here, in order to show the influence of the heating temperature and the heating treatment time, when the catalyst support carbon materials E6-1 to E6-8 and C6-1 to C6-2 were produced, the combination of the heating temperature and the heating treatment time was changed for each carbon material.

[0185] According to the above procedure, the catalyst support carbon materials E1-1 to E6-8 and C1-1 to C6-2 were produced. The physical property values of these catalyst support carbon materials are shown in Table 2.

[0186] In addition, as a commercially available catalyst support carbon material, Lion Corpn. EC600JD was prepared. In addition, EC600JD-1800 obtained by heat treating EC600JD at 1800°C for 2 hours under argon flow was prepared. These EC600JDs have a dendritic structure like the dendritic carbon nanostructure, and also have developed intra-branch pores and have a large BET specific surface area.

[0187] In addition, CNovel-MH (Toyo Tanso Co., Ltd.) was prepared. In addition, MH-2000 obtained by heat treating CNovel-MH at 2000°C for 2 hours under argon flow was prepared. These carbon materials are porous carbon materials without a dendritic structure. These carbon materials have a large number of pores formed in carbon particles.

[0188] Furthermore, if these pores are linked to each other, the gas diffusivity is increased. For this purpose, by heat treating MH-2000 in air, carbon oxidation consumption can be made. Specifically, using a horizontal tubular electric furnace, the flowing gas was treated as dry air at 560°C for 1 hour (this carbon material is MH-2000-Air560). By this air oxidation treatment, a weight reduction of 46 mass% was produced. Furthermore, by this air oxidation treatment, the area of the hysteresis loop was greatly reduced. The calculated value of the area of the hysteresis loop in MH-2000 before the air oxidation treatment was 61, and in contrast, the calculated value of the area of the hysteresis loop in MH-2000-Air560 subjected to the air oxidation treatment was 0.6. That is, by the air oxidation treatment, the hysteresis loop substantially disappeared, and the adsorption and desorption curves substantially coincided.

[0189] In addition, acetylene black AB (manufactured by Nippon Carbon Co., Ltd.) and #4500 (manufactured by Tokai Carbon Co., Ltd.) of an electrically conductive grade were prepared. These carbon materials had a developed dendritic structure and substantially no intraparticle pores. The physical property values of these commercially available catalyst support carbon materials are shown in Table 2.

[0190] [Table 1]

[0191]

[0192] [Table 2]

[0193]

[0194] <3. Production of MEA>

[0195] Using the catalyst support carbon materials E1-1 to E6-8, C1-1 to C6-2, and the commercially available catalyst support carbon materials, MEAs (membrane electrode composite) were produced by the following procedure.

[0196] (3-1. Platinum supporting treatment)

[0197] A carbon material dispersion liquid was produced by dispersing the catalyst support carbon material in distilled water. Next, formaldehyde was added to the carbon material dispersion liquid, and the carbon material dispersion liquid was left to stand in a water bath set to 40°C. Next, the carbon material dispersion liquid was left to stand until the temperature of the carbon material dispersion liquid reached 40°C, which was the same as the water bath. Thereafter, while stirring the carbon material dispersion liquid, a nitric acid aqueous solution of a dinitro diamine Pt complex was slowly injected into the carbon material dispersion liquid. After stirring for about 2 hours, the carbon material dispersion liquid was filtered, and the obtained solid was washed. The solid obtained by this operation was vacuum dried at 90°C, and then pulverized using a mortar to produce a pulverized solid. Next, the pulverized solid was heat treated at 200°C for 1 hour in an argon atmosphere containing 5% by volume of hydrogen. Thus, a platinum-supported carbon material was produced.

[0198] In addition, the platinum supporting amount of the platinum-supported carbon material was set to 40% by mass with respect to the total mass of the catalyst support carbon material and the platinum particles. The platinum supporting amount was confirmed by inductively coupled plasma atomic emission spectrometry (ICP-AES).

[0199] (3-2. Production of ink for coating)

[0200] A Nafion solution (Nafion, registered trademark: Nafion, perfluorosulfonic acid-based ion exchange resin) that dissolves as an electrolyte resin was prepared. Next, a platinum-supported carbon material (Pt catalyst) and the Nafion solution were mixed in an argon atmosphere. Here, the mass ratio of the solid content of the electrolyte resin was set to 1.0 times relative to the platinum-supported carbon material. However, in the case of using AB or #4500, which is non-porous, as the carbon material for the catalyst support, the mass ratio of the solid content of the electrolyte resin was set to 0.5 times. Next, the mixed solution was gently stirred, and the platinum-supported carbon material in the mixed solution was broken up with ultrasonic waves. Next, ethanol was further added to the mixed solution, and the adjustment was performed so that the total solid content concentration of the platinum-supported carbon material and the electrolyte resin would be 1.0 mass% relative to the total mass of the mixed solution. In this way, a coating ink containing the platinum-supported carbon material and the electrolyte resin was prepared.

[0201] (3-3. Preparation of the catalyst layer)

[0202] Next, the coating ink was sprayed onto a Teflon (registered trademark) sheet in such a way that the mass of platinum per unit area of the catalyst layer (hereinafter referred to as "platinum unit area weight") would be 0.2 mg / cm 2 . Next, a catalyst layer was prepared on the Teflon (registered trademark) sheet by performing a drying process at 120°C for 60 minutes in an argon atmosphere. Two identical catalyst layers were prepared, and one was set as the cathode and the other was set as the anode.

[0203] (3-4. Preparation of the MEA)

[0204] A square electrolyte membrane with one side of 6 cm was cut out from a Nafion membrane (NR211 manufactured by Dupont). In addition, the Teflon (registered trademark) sheet provided with each catalyst layer of the anode and the cathode was cut into a square shape with one side of 2.5 cm using a cutter. The electrolyte membrane was sandwiched between each catalyst layer of the anode and the cathode on the sheet cut in this way, and was punched and formed into a laminate. Specifically, the electrolyte membrane was sandwiched so that the center portion of the electrolyte membrane would be held by each catalyst layer, and each catalyst layer and the electrolyte membrane would be in contact with each other without being misaligned, and the electrolyte membrane was punched to obtain a laminate. The punching was performed at 120°C, 100 kg / cm 2 for 10 minutes. Next, the laminate was cooled to room temperature. Next, only the Teflon (registered trademark) sheet was carefully peeled off from each catalyst layer of the anode and the cathode. Each catalyst layer of the anode and the cathode was fixed to the electrolyte membrane by the above procedure.

[0205] Next, two carbon papers of 2.5 cm square were cut out from a carbon paper (35BC manufactured by SGL Carbon) to be a gas diffusion layer. Next, the cut-out carbon papers were stacked in a manner not to be misaligned with each of the catalyst layers of the anode and the cathode, and a stack was prepared. Next, the stack was pressed at 120°C, 50 kg / cm 2 for 10 minutes, and an MEA was prepared. Further, the mass of the catalyst layer fixed to the Nafion membrane was calculated from the difference in mass of the Teflon (registered trademark) sheet with the catalyst layer before pressing and the Teflon (registered trademark) sheet peeled off after pressing. Then, the platinum area weight, the carbon material for catalyst carrier area weight, and the electrolyte resin area weight were calculated from the mass ratio of the components of the catalyst layer. By this method, it was confirmed that the platinum area weight was 0.2 mg / cm 2 .

[0206] <4. Performance evaluation test>

[0207] The prepared MEA was assembled into a unit cell, and installed in a fuel cell measuring device, and a performance evaluation of the fuel cell was performed.

[0208] Air was supplied to the cathode and pure hydrogen was supplied to the anode at an atmospheric pressure, respectively at a utilization of 40% and 70%. The unit cell temperature was set to 80°C. In addition, the air and the pure hydrogen supplied to the fuel cell were humidified by passing through distilled water kept at 65°C in a humidifier, that is, bubbling. That is, water vapor corresponding to modified hydrogen was contained in these gases. Then, the humidified gases were supplied to the unit cell. After the gases were supplied to the unit cell under such conditions, the load was slowly increased, and the voltage between the terminals of the unit cell at 200 mA / cm 2 and 1000 mA / cm 2 was recorded as the output voltage, and a performance evaluation of the fuel cell was performed.

[0209] With respect to the performance evaluation results of the obtained fuel cell, the voltage at each current density was evaluated based on the criteria of A, B, and C. With respect to the pass level, the pass was set to B, and the higher performance was set to A.

[0210] With respect to the criteria of the pass level, the following classification of the levels was used. The evaluation results are also shown in Table 3.

[0211] (Classification of levels)

[0212] - Output voltage at 200 mA / cm 2 - Output voltage at 1000 mA / cm

[0213] A: 0.86 V or more.

[0214] B: 0.84 V or more and less than 0.86 V.

[0215] C: does not satisfy B.

[0216] - 1000 mA / cm 2 of output voltage

[0217] A: 0.65 V or more

[0218] B: 0.60 V or more and less than 0.65 V.

[0219] C: does not satisfy B.

[0220] [Table 3]

[0221]

[0222] <5. Evaluation>

[0223] The carbon material for a catalyst support satisfying the conditions of the present disclosure has high output voltage at low current (200 mA / cm 2 ) as well as at high current (1000 mA / cm 2 ). Thus, the overvoltage is small. In particular, with respect to the carbon materials for a catalyst support E2-5, E3-1, E4-3, E4-4, E5-5, E5-6, and E6-2 to E6-4, the overvoltage at high current is particularly low. These carbon materials for a catalyst support are produced in a manner satisfying any one of the conditions of temperature of silver nitrate solution, molding pressure, temperature- increasing rate at the first heat treatment, and heating temperature at the second heat treatment, particularly preferably. The carbon materials for a catalyst support E4-1 and E4-2 do not satisfy the condition (D), but the results are good. Of course, in the case where all of the conditions (A) to (E) are satisfied, particularly good results can be obtained.

[0224] On the contrary, the nitrogen adsorption-desorption isotherms of the carbon materials for a catalyst support C1-1, C1-2, C2-1, C3-1, C3-2, C4-1 to C4-3, C5-1, C6-1, and C6-2 do not show two hysteresis loops. Thus, the overvoltage at high current becomes high. The nitrogen adsorption-desorption isotherms of the commercially available carbon materials for a catalyst support do not show two hysteresis loops. Not only the overvoltage at high current becomes high, but also the overvoltage at low current becomes high.

[0225] Further, the reference numerals attached to each drawing are described below.

[0226] 100 solid polymer fuel cell,

[0227] 110, 120 separator,

[0228] 130, 140 gas diffusion layer,

[0229] 150, 160 catalyst layers

[0230] 170 electrolyte membrane.

[0231] Furthermore, the entire disclosure of Japanese Patent Application No. 2020-175410 is incorporated herein by reference.

[0232] All documents, patent applications and technical standards described in this specification are incorporated herein by reference to the same extent as the specific documents, patent applications and technical standards described separately and incorporated by reference.

Claims

1. A carbon material for a solid polymer fuel cell catalyst carrier, wherein, the nitrogen adsorption-desorption isotherm shows two hysteresis loops of a first hysteresis loop and a second hysteresis loop in a range of relative pressure P / P0 of 0.4 or more, as the two hysteresis loops shown by the nitrogen adsorption-desorption isotherm, there are the first hysteresis loop present in a range of relative pressure P / P0 of 0.87 or more, and the second hysteresis loop present in a range of relative pressure P / P0 of 0.4 to 0.87, and the following conditions (A), (B), (C), and (D) are satisfied: (A) The specific surface area obtained by BET analysis of the nitrogen adsorption isotherm is 450 m 2 / g to 1500 m 2 / g; (B) the first hysteresis loop has a difference in adsorption amount ΔV 0.87 of 20 mL / g or less in a range of relative pressure P / P0 of 0.87 ± 0.03 0.87min ; (C) the area ΔS of the second hysteresis loop 0.4-0.87 from 5 mL / g to 50 mL / g; (D) The difference AV in adsorption amount at a relative pressure P / P0 of 0.99 and at a relative pressure P / P0 of 0.95 in the nitrogen adsorption isotherm 0.95-0.99 is 500 mL / g to 1100 mL / g.

2. A carbon material for a solid polymer fuel cell catalyst carrier, wherein, the nitrogen adsorption-desorption isotherm shows two hysteresis loops of a first hysteresis loop and a second hysteresis loop in a range of relative pressure P / P0 of 0.4 or more, as the two hysteresis loops shown by the nitrogen adsorption-desorption isotherm, there are the first hysteresis loop present in a range of relative pressure P / P0 of 0.87 or more, and the second hysteresis loop present in a range of relative pressure P / P0 of 0.4 to 0.87, and the following conditions (A), (B), (C), and (E) are satisfied: (A) The specific surface area obtained by BET analysis of the nitrogen adsorption isotherm is 450 m 2 / g to 1500 m 2 / g; (B) the first hysteresis loop has a difference in adsorption amount ΔV 0.87 of 20 mL / g or less in a range of relative pressure P / P0 of 0.87 ± 0.03 0.87min ; (C) the area ΔS of the second hysteresis loop 0.4-0.87 from 5 mL / g to 50 mL / g; (E) the half-value width of the G band detected in the Raman spectrum obtained by Raman spectroscopy in the range of 1500 cm -1 - 1700 cm -1 -1 is 45 cm -1 - 70 cm -1 .

3. The carbon material for a solid polymer fuel cell catalyst support according to claim 2, wherein the following condition (D) is also satisfied: (D) The difference AV in adsorption amount at a relative pressure P / P0 of 0.99 and at a relative pressure P / P0 of 0.95 in the nitrogen adsorption isotherm 0.95-0.99 is 500 mL / g to 1100 mL / g.

4. A catalyst layer for a solid polymer type fuel cell, wherein, a carbon material for a solid polymer fuel cell catalyst carrier according to any one of claims 1 to 3.

5. A fuel cell, wherein, a catalyst layer for a solid polymer fuel cell according to claim 4.

6. The fuel cell of claim 5, wherein, the catalyst layer for a solid polymer fuel cell is a cathode-side catalyst layer.

Images