WO2019163548A1

WO2019163548A1 - Electrode catalyst for fuel cells and method for producing same

Info

Publication number: WO2019163548A1
Application number: PCT/JP2019/004509
Authority: WO
Inventors: 東山　和寿; 敏広宮尾; 内田　裕之; 明裕飯山
Original assignee: 国立大学法人山梨大学
Priority date: 2018-02-26
Filing date: 2019-02-07
Publication date: 2019-08-29
Also published as: JPWO2019163548A1; JP7214944B2

Abstract

Provided are: an electrode catalyst for fuel cells, which is suppressed in movement/aggregation of catalyst metal fine particles when in use, said catalyst metal fine particles being supported by carbon, and which has a long service life; and a method for producing this electrode catalyst for fuel cells. The present invention provides an electrode catalyst for fuel cells, which is provided with a carbon carrier, nanopits and/or nanochannels, transition metal fine particles and a noble metal layer, and which is configured such that: the nanopits and/or the nanochannels are formed on the carbon carrier; the transition metal fine particles are in contact with the carbon carrier within the nanopits and/or the nanochannels; and the noble metal layer is formed on the transition metal fine particles.

Description

Fuel cell electrode catalyst and method for producing the same

The present invention relates to an electrode catalyst for a fuel cell, and more specifically to a long-life electrode catalyst for a fuel cell in which migration and aggregation of catalyst metal fine particles supported on carbon during use is suppressed and a method for producing the same.

As an electrode catalyst for a polymer electrolyte fuel cell, a catalyst in which an active metal mainly composed of a noble metal such as Pt is supported on a conductive carrier such as carbon is generally used. The performance of this electrocatalyst increases as the active metal loading is the same, the greater the surface area of the active metal, that is, the smaller the particle size and the higher the dispersion on the support. In addition, since platinum is expensive, it is required that the active metal be finely divided, alloyed and uniformly supported on the support in order to reduce the amount of platinum used.

Even if an electrocatalyst with high initial activity that ensures active metal atomization and high dispersion support is applied repeatedly, for example, when a potential change simulating a load variation of a fuel cell vehicle (FCV) is applied, the particle diameter of the active metal is The catalyst performance gradually decreases with the increase. It is known that the increase of the active metal fine particles proceeds through two paths of “dissolution / reprecipitation (Ostwald ripening)” and “migration / aggregation” on the carbon support.

Recently, when the active metal particle size was precisely controlled and a fine catalyst with a small particle size distribution was synthesized, “dissolution / re-precipitation” due to potential change was suppressed, and it was not possible to greatly reduce catalyst performance degradation. It was reported by patent document 1. This means that if another “movement / aggregation” can be suppressed, an electrode catalyst having an extremely high durability without increasing the particle diameter can be obtained, and its realization is expected.

As a catalyst aiming at the movement / aggregation suppression of the active metal fine particles, for example, as disclosed in Patent Document 1, it is surrounded by a graphite crystal edge surface possessed by a special carbon material such as a carbon nanotube, a carbon nanohorn, or a cup stack type carbon. There is known an alcohol dehydrogenation catalyst in which Ru and Pt particles are held in the depression due to the adsorption effect of the oxygen-containing functional group on the edge surface. Non-Patent Document 2 reports a fuel cell electrode catalyst in which durability is improved by coating a carbon nanotube carrying Pt with a thin silica layer.

JP2004-82007

However, in the catalyst disclosed in Patent Document 1, the oxygen-containing functional group is easily detached at a high potential, and it is difficult to expect long-term stability in the harsh environment of the fuel cell. Moreover, in the catalyst which has a coating layer like a nonpatent literature 2, the fall of the catalyst activity by gas diffusion resistance becomes a problem.

The present invention has been made in view of such circumstances, and provides a long-life electrode catalyst for a fuel cell in which movement and aggregation during use of catalytic metal fine particles supported on carbon are suppressed, and a method for producing the same. It is.

According to the present invention, it comprises a carbon support, nanopits and / or nanochannels, transition metal fine particles, and a noble metal layer, and the nanopits and / or nanochannels are formed on the carbon support, and the transition metal There is provided a fuel cell electrode catalyst in which fine particles are in contact with the carbon support in the nanopits and / or nanochannels, and the noble metal layer is formed on the transition metal fine particles.

As a result of intensive studies by the present inventors, it is possible to suppress the movement / aggregation of catalyst metal fine particles supported on carbon by using the catalyst configuration as described above, and to extend the life of the electrode catalyst for fuel cells. As a result, the present invention has been completed.

According to another aspect of the present invention, the method includes a supporting step, a heat treatment step, and a noble metal layer forming step. In the supporting step, a transition metal fine particle is supported on a carbon support. There is provided a method for producing an electrode catalyst for a fuel cell, in which the carbon carrier supporting transition metal fine particles is heated, and in the noble metal layer forming step, a noble metal layer is formed on the transition metal fine particles.

FIG. 1A is a perspective view of catalyst particles 2 supported on a graphite base surface by a conventional method. 1B and 1C are perspective views of catalytic metal fine particles 4 carried on nanopits 3 and nanochannels 5 according to the present invention. 2A is an end view showing the internal structure of the catalyst metal fine particles 4 held in the nanopits 3, and FIG. 2B is an end view showing the internal structure of the catalyst metal fine particles 4 held in the nanochannels 5. 3A to 3D are explanatory views showing an example of the formation process of the catalytic metal fine particles 4 held in the nanopits 3 or nanochannels 5 of the present invention. It is a block diagram of the arc plasma vapor deposition apparatus used as an example for forming the noble metal or transition metal fine particles on the carbon support in the production example. FIG. 5A is a transmission electron microscope (TEM) image after vacuum heating of Fe particles supported on graphene. FIG. 5B is a schematic cross-sectional view for explaining the structure of the catalyst, and is a schematic view for forming nanopits. The STEM image (a) and STEM image (b) in FIG. 6 are scanning transmission electron microscope (STEM) images of Fe nanopits on graphene by ultrahigh vacuum heating. The SEM image (a) and SEM image (b) in FIG. 7 are scanning electron microscope images (SEM) of Fe particles carried on the HOPG substrate of the production example. The SEM image (a) in FIG. 8 is an SEM image after the Fe particles supported on the HOPG substrate are heated in hydrogen. The SEM image (b) is an enlarged image of the same location. The TEM image (a) and TEM image (b) in FIG. 9 are TEM images of Fe particles supported on graphitized carbon black (GCB) powder after heating in hydrogen. The SEM image (a) in FIG. 10 is an SEM image before the acid treatment of Fe particles on the HOPG substrate heated in Ar (after Ar heating), and the SEM image (b) is on the HOPG substrate heated in Ar. It is a SEM image which shows the SEM image after the acid treatment of Fe particle | grains. The SEM image (a) and SEM image (b) in FIG. 11 are SEM images after the Fe / HOPG vacuum heat-treated product is subjected to acid dissolution treatment. An SEM image (a) in FIG. 12A is an SEM image of the catalyst of the production example, and an SEM image (b) is an SEM image of the same field after further acid dissolution treatment. The SEM image (c) and SEM image (d) in FIG. 12B are high-magnification images of the SEM image (a) and SEM image (b) in FIG. 12A, respectively. FIG. 13A and FIG. 13B (b) are apparatus diagrams used for selective Pt chemical plating in a manufacturing example. FIG. 13C is an enlarged view of the substrate sample. It is a SEM image before and behind Pt chemical plating of the catalyst of a manufacture example. It is a figure which shows the change of the Pt4f peak by the X-ray photoelectron spectrometer (XPS) of the catalyst of FIG. 14 with a graph (a), a graph (b), and a graph (c). The SEM images (a) and SEM images (b) in FIG. 16 are SEM images before and after applying a potential step cycle of Pt / HOPG. The SEM images (a) and SEM images (b) in FIG. 17 are SEM images before and after applying a potential step cycle of a commercially available electrode catalyst. The SEM images (a) and SEM images (b) in FIG. 18 are SEM images before and after applying a potential step cycle of Pt / KB. It is a SEM image after heating of FeNi of various compositions carried on a HOPG substrate. It is a figure which shows the FeNi average particle diameter and carbon solid solution amount of the 900 degreeC heat processing sample of FIG. It is a figure which shows the temperature dependence of the average particle diameter of Fe85Ni15 / HOPG of FIG. It is a SEM image which shows the influence of 850 degreeC heat processing time with respect to the particle diameter of Fe100 and Fe85Ni15 / HOPG. It is a SEM image before and behind the acid treatment of the 900 degreeC heat processing sample of FIG. The SEM images (a) and SEM images (b) in FIG. 24 are SEM images before and after acid treatment of Fe85Ni15 / HOPG heat-treated at 850 ° C. for 5 seconds. It is a figure which shows the supporting pattern (a), supporting pattern (b), and supporting pattern (c) of Fe and Ni used by the arc plasma vapor deposition method. It is an SEM image after heat treatment of Fe53Ni47 / HOPG carried synchronously at 900 ° C. for 5 seconds. The SEM images (a) and SEM images (b) in FIG. 27 are SEM images before and after acid treatment of Fe70Ni21Pt9 / HOPG heat-treated at 700 ° C. for 10 seconds.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. Various characteristic items shown in the following embodiments can be combined with each other. In addition, the invention is independently established for each feature.

1. Electrocatalyst 100
As shown in FIG. 1 (FIGS. 1A to 1C) and FIG. 2 (FIGS. 2A to 2B), the electrode catalyst 100 includes a carbon support 1, nanopits 3 and / or nanochannels 5, transition metal fine particles 8, and a noble metal layer. The nanopits 3 and / or nanochannels 5 are formed on the carbon support 1, and the transition metal fine particles 8 are in contact with the carbon support 1 in the nanopits 3 and / or nanochannels 5, and the noble metal layer 7 Is formed on the transition metal fine particles 8.

Here, the structural features of a typical example of an electrode catalyst according to an embodiment of the present invention will be described with reference to FIG. 1 (FIGS. 1A to 1C). FIG. 1A illustrates an electrode catalyst composed of catalyst particles 2 such as Pt particles when supported on a carbon carrier 1 by a generally known method such as an impregnation method. For example, when the carbon support 1 is graphite or graphitized carbon black, the basal plane of the graphite crystal is exposed on the surface and is smooth and has low chemical reactivity and low affinity. 2 is very easy to move and aggregate.
On the other hand, FIG. 1B shows a structure in which nano-level depressions, that is, nanopits 3 are formed on the bottom surface of the graphite when the carbon support 1 is graphite or graphitized carbon black. As shown in FIG. Includes transition metal fine particles 8 which are nanoanchors in which a transition metal having a strong carbon affinity is in contact with the carbon support 1 with a strong bonding strength. The catalytic metal fine particle 4 in FIG. 1B has the transition metal fine particle 8 as a core, and the outer peripheral portion of the transition metal fine particle 8 other than the transition metal fine particle 8 in contact with the carbon support 1 is completely covered with a noble metal layer 7 with Pt or the like. It has a structure. FIG. 1C shows a case where a nano-sized groove, that is, a nanochannel 5 is formed on the bottom surface of the graphite base when the carbon support 1 is graphite or graphitized carbon black, as shown in FIG. 2B. In addition, in the interior of the catalyst metal fine particle 4 at the tip of the nanochannel, there is a transition metal fine particle 8 that is a nano-anchor of a transition metal having a strong carbon affinity and strongly bonded to the carbon support 1, as in the case of the nanopit.
Each configuration will be described below.

1-1. Carbon carrier 1
The carbon carrier 1 is not particularly limited as long as it is a carbon material for electrodes that can carry metal fine particles. However, in the fuel cell, since the electrode catalyst is continuously exposed to low pH and high potential, and the carbon support 1 having low corrosion resistance is likely to gradually corrode, a graphitized carbon material having high corrosion resistance to acid and high potential is used. It is preferable to use it. Therefore, as the carbon carrier 1, for example, graphite and graphitized carbon black are preferable.
In addition, when graphite and graphitized carbon black are used as the carbon support 1, the movement and aggregation of catalyst metal fine particles generally composed of Pt or the like are likely to proceed, and the catalytic activity is decreased more rapidly than other carbon supports. . However, in the embodiment according to the present invention, the activity of the catalyst is maintained even when a graphitized carbon material having high corrosion resistance to acid or high potential is used because the movement and aggregation during use of the catalyst metal fine particles are suppressed. Can be kept high for a long time.

1-2. Nanopit 3 and / or nanochannel 5
The nanopits 3 and nanochannels 5 are recesses formed on the surface of the carbon support. As will be described later, the concave portion is formed by performing a heat treatment after supporting the transition metal fine particles 8. The shape of the recess formed is a pit shape and a channel shape. The pit shape is, for example, a circular depression such as a circle and an ellipse shown as conceptual diagrams in FIGS. 1B and 2A. The channel shape is, for example, a groove shown as a conceptual diagram in FIGS. 1C and 2B. Note that the shape of the concave portion is not constant because it is formed by the interaction between the transition metal fine particles 8 and the carbon support in the heat treatment process and changes depending on the behavior of the transition metal fine particles 8 that can move on the carbon support at this time. Further, the nanopits 3 and the nanochannels 5 may be present on the surface of the carbon carrier 1 alone or in a mixture.

Further, the average diameter of the nanopits 3 is not particularly limited, but is preferably 0.3 to 14 nm, and more preferably 0.3 to 7 nm, for example. Specifically, the average diameter of the nanopits 3 is, for example, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1, 2, 3, 4, 5 6, 7, 8, 9, 10, 11, 12, 13, 14 nm, and may be in the range between any two of the numerical values exemplified here. In this specification, “average diameter” means an arithmetic average obtained by measuring the diameter of a circumscribed circle of nanopits in a TEM image. The number of measurement samples is, for example, 500 or more.
Further, the average width of the nanochannel 5 is not particularly limited, but is preferably 0.3 to 14 nm, and more preferably 0.3 to 7 nm, for example. Specifically, the average width of the nanochannel 5 is, for example, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 nm, and may be within a range between any two of the numerical values exemplified here. In the present specification, the “average width” means an arithmetic average obtained by measuring the width of the groove in the TEM image. The number of measurement samples is, for example, 500 or more.

Further, when graphite and graphitized carbon black are used as the carbon support 1, nanopits and / or nanochannels can be formed on the basal plane of the graphite or graphitized carbon black.

As a study according to an embodiment of the present invention, the present inventors used transition metal nanoparticles forming interstitial carbides such as Fe as the basis of graphene and highly oriented graphite (HOPG). It was formed on the surface and heat-treated under various conditions including air, and pits and grooves formed on the surface were observed in detail with an electron microscope. As a result, it was found that in the heat treatment under an inert gas and in a vacuum, the transition metal particles slightly sink into the basal plane due to carbon solid solution in the metal lattice. Transition metal and carbon existed in this part, and when these were completely dissolved and removed by acid treatment, pits could be confirmed as spaces. In this specification, the subsidized region and the pit generated by the acid treatment are not particularly distinguished and are called nanopits. Even if the pit generated by the acid treatment is buried with the noble metal layer, the region is continuously called nanopit. That is, even if the subsidized region is completely filled with the noble metal layer (and the transition metal particles), and the subsidence due to nanopits cannot be confirmed directly from the electrode catalyst surface, the subsidized region on the carbon support actually The (recessed portion) remains formed, and the region can still be referred to as nanopits. From the electron microscope image and the calculation of the carbon solid solution amount, the nanopit depth was estimated to be about 15% of the particle diameter, and the nanopit diameter was 60% of the particle diameter.
On the other hand, when heated in the presence of carbon gasifying agents such as oxygen, hydrogen, water vapor, and carbon dioxide, in addition to nanopits, some particles identify the base surface while gasifying solute carbon on the metal surface. It moves in the azimuth (n x 60 deg), leaving a groove where carbon has disappeared in the trace. Hereinafter, this is particularly called a nanochannel. Although the moving length of the particles varies depending on the conditions, the depth and width of the grooves were found to be about 15% and about 60% of the particle diameter, as in the nanopits. The depth of nanopits and nanochannels formed by transition metal particles having a diameter of 10 nm is only 1.5 nm at most. In order to completely suppress the migration aggregation of Pt nanoparticles having a diameter of 3 nm, a mechanism for further supplementing the low barrier of nanopits or nanochannels is necessary.

1-3. Transition metal fine particles 8
As shown in FIG. 2 (FIGS. 2A to 2B), the transition metal fine particles 8 are in contact with the carbon support 1 in the nanopits 3 and / or the nanochannels 5. With such a catalyst structure, movement / aggregation of the catalyst metal fine particles 4 composed of the transition metal fine particles 8 and the noble metal layer 7 is suppressed. This is because the catalyst metal fine particles 4 are supported in the nanopits 3 or the nanochannels 5. In addition, the transition metal fine particles 8 come into contact with the carbon support 1 and have an effect as an anchor for fixing the catalyst metal fine particles to the carbon support 1. Therefore, in the present invention, the transition metal fine particles 8 can be said to be nano-sized anchors for fixing the catalyst metal fine particles, that is, “nano-anchors”.

Even if there is no such nanoanchor, a certain effect of suppressing migration / aggregation can be expected by supporting catalytic metal fine particles in pits and channels. However, in the first place, it is difficult to carry in the pit, channel or the like unless there is a core such as the nano anchor. For example, when Pt particles are carried by a generally known method, a plurality of Pt particles are carried in each pit or channel, or a large number of Pt particles are carried in regions other than pits and grooves. Unavoidable. In that case, the expected aggregation suppressing effect can hardly be expected due to the movement and aggregation of particles in each region. In addition, a sufficient suppression effect can be obtained only when the pits and channels have a sufficient depth and can serve as a barrier for movement and aggregation of the catalytic metal fine particles existing in the pits and channels.

The position where the transition metal fine particle 8 is in contact with the carbon support 1 is not particularly limited as long as it is in the nanopit 3 and / or the nanochannel 5, and even on the bottom surface in the nanopit 3 and / or nanochannel 5, There may be. In the case of the nanochannel 5, when the transition metal fine particle 8 is formed by the movement and forms the nanochannel 5 and stops at the end of the nanochannel 5, the transition metal fine particle 8 is inclined to the step portion of the nanochannel 5. There are many.

As will be described later, the transition metal fine particles 8 are supported on the carbon support 1 by a support process, and a recess (nanopit 3 and / or nanochannel 5) is formed by a subsequent heat treatment process. It is necessary to contact and fix the carbon carrier 1 in the nanochannel 5. One reason for the formation and immobilization of the recesses is presumed to be the result of the solid dissolution of the carbon support in the supported metal fine particles by heating in the heat treatment step. Therefore, the transition metal fine particles used as the transition metal fine particles 8 preferably contain a metal element having a strong carbon affinity. In the present invention, transition metals such as Fe are possible as the nano-anchor material. In general chemical terms, these are easier to understand as "transition metals that form interstitial carbides", but, as will be described later, in the system of the present invention, it is considered that they form a solid solution but do not form carbides. Therefore, the term “transition metal with strong carbon affinity” is used to avoid misunderstanding.

Examples of metal elements having strong carbon affinity include Fe, Ni, Co, Mn, Cr, Mo, V, Ta, and W. Carbon easily dissolves in metals such as Fe, Ni, Co, Mn, Cr, Mo, V, Ta, and W, and easily forms interstitial carbides in which carbon atoms enter the crystal lattice. Since these metals having a strong affinity for carbon diffuse with graphite, the metal particles 4 can act as a strong anchor that stops the movement of the catalyst metal fine particles 4 and stays at a specific position of the graphite. The same applies to alloys containing these metals. The content of the metal element having strong carbon affinity in the transition metal fine particles 8 is preferably 1 to 100 atm%, more preferably 10 to 100 atm%, and further preferably 50 to 100 atm%. Specifically, the content of the metal element having strong carbon affinity in the transition metal fine particles 8 is, for example, 1, 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 100 atm%, and may be within a range between any two of the numerical values exemplified here.

Ordinary iron does not have a large amount of carbon in the low-temperature ferrite phase, but when it changes to the austenite phase above 900 ° C., it dissolves a maximum of 1 to 2 wt% of carbon. The more the amount of solid solution, the deeper the transition metal particles can sink into the bottom surface of the graphite base. However, after cooling, the solid solution of carbon precipitates on the surface of the transition metal particles, which may hinder manufacturing. There is also. For example, austenitic stainless steel is preferable because it can prevent precipitation of solute carbon from the particles all at once because the austenite phase is stably present at room temperature even after cooling. From such a viewpoint, it is preferable that the transition metal fine particles 8 contain Fe, Ni, and Cr. That is, the transition metal fine particles 8 are preferably austenitic stainless steel containing Fe, Ni, and Cr.

In addition, the Fe—B binary alloy forms an eutectic of Fe and Fe ₂ B, so that a fine grain boundary is formed inside the nanoparticle, and solid solution carbon can be precipitated there. It may be desirable as well as steel. From such a viewpoint, the transition metal fine particles 8 preferably contain Fe and B. That is, the transition metal fine particles 8 are preferably an Fe—B binary alloy.

Moreover, the amount of carbon solid solution in the FeNi alloy in which Ni is added to Fe in a large number of temperature ranges is larger than that in the case of pure Fe. Therefore, by applying the Fe—Ni alloy as the transition metal, larger nanopits 3 and / or nanochannels 5 can be formed even under the same heat treatment conditions, and the transition metal particle surface that becomes an obstacle in the acid treatment or noble metal layer formation process can be formed. In some cases, the deposition of a dense and thick carbon layer can be suppressed. Further, the nanopits 3 and / or the nanochannels 5 can be formed even at a relatively low temperature, which contributes to the formation of fine particles. From such a viewpoint, it may be preferable to include Fe and Ni as a metal element having a strong carbon affinity. The Ni content in the transition metal fine particles 8 is, for example, preferably 1 to 90 atm%, more preferably 5 to 80 atm%, and further preferably 10 to 50 atm%. Specifically, the Ni content is, for example, 1, 2, 3, 4, 5, 10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80. , 85, 90 atm%, and may be within a range between any two of the numerical values exemplified here.

Further, from the viewpoint of coexistence of formation of the nanopits 3 and / or nanochannels 5 and dissolution of the transition metal fine particles 8 by acid treatment, the carbon solid solution amount of the metal (pure metal / alloy) constituting the transition metal fine particles 8 Is preferably 1.2 to 4.5 atm%.

Moreover, if the noble metal similar to the noble metal layer 7 such as Pt covering the surface is mixed in advance with the metal (pure metal / alloy) constituting the transition metal fine particles 8, the mutual affinity is increased, and the catalytic activity is increased. It is preferable in the manufacturing process. Further, it may be possible to form the noble metal layer 7 by acid dissolution treatment regardless of plating treatment or the like.

Further, from the viewpoint of formation of nanopits 3 and / or nanochannels 5 at a relatively low temperature and ease of acid treatment, the alloy constituting the transition metal fine particles 8 is preferably a ternary alloy of Fe—Ni—Pt. There is a case. *

In the present invention, the transition metal fine particles 8 are fixed on the carbon support 1 by contacting the transition metal fine particles 8 with the carbon support 1.

For example, when nanopit 3 is formed on graphite, nanopit 3 is considered to be a region in which a transition metal such as Fe diffuses on the bottom surface of the graphite, and the bonding strength between the metal and carbon is expected to be high. Therefore, if these transition metal particles, or particles obtained by dissolving and removing a part of them, are used as the core to form a noble metal layer such as Pt and form catalytic metal fine particles, the transition metal such as Fe in the nanopit region has just hit the ground. It functions like an anchor bolt and can firmly hold the catalytic metal fine particles coated with Pt.

Here, the average particle diameter of the transition metal fine particles 8 is not particularly limited, but is, for example, 0.5 to 10 nm, and particularly preferably 0.5 to 5 nm. Transition metal fine particles having an average particle size that is too small are difficult to produce and may not function sufficiently as nanoanchors. If the average particle size is too large, the average particle size of the catalytic metal fine particles obtained by forming the noble metal layer also increases accordingly, and the mass activity (catalytic activity per unit mass) tends to decrease. Specifically, the average particle diameter of the transition metal fine particles 8 is, for example, 0.5, 0.6, 0.7, 0.8, 0.9, 1, 2, 3, 4, 5, 6, 7, 8, 9, and 10 nm, and may be within a range between any two of the numerical values exemplified here. In the present specification, the “average particle diameter” means an arithmetic average obtained by measuring the diameter of the circumscribed circle of each particle in the TEM image. The number of measurement samples is, for example, 500 or more.

Ratio of average particle diameter of transition metal fine particles 8 as nano-anchors to average diameter of nanopits 3 (average particle diameter of transition metal fine particles / average diameter of nanopits) or ratio of average width of nanochar 5 (average particle diameter of transition metal fine particles / nano The average channel width is preferably from 0.3 to 1.4, more preferably from 0.5 to 1.0.

In one embodiment of the present invention, for example, considering that the transition metal fine particles 8 are formed of Fe as a representative transition metal, when this is heated to a maximum of 1000 ° C. on a graphite substrate, it slightly varies depending on the amount of Fe supported. However, the average particle size is around 10 nm. Moreover, the carbon solid solution amount at that time is 1.5 wt%. Taking these into consideration, the size of the subduction of the Fe particles (the size of the nanopits 3) can be calculated. As a result of the estimation, the diameter of the nanopit is about 7 nm, which is about 69% of the Fe particle diameter, and its maximum depth is 1.4 nm, which is 14% of the diameter. When the noble metal layer 7 made of Pt is formed on Fe, the Pt particle diameter of the catalytic metal fine particles 4 is usually preferably 2 to 3 nm, so that the size of the nanoanchor serving as the core is preferably suppressed to about 2 nm. As a result, the ratio of the average particle diameter of the transition metal fine particles 8 which are nano-anchors to the average diameter of the nanopits 3 (transition metal fine particle average particle diameter / nanopit average diameter) is 2 nm / 7 nm≈0.3. The production of the catalyst according to the present invention can include an acid treatment step in which the nanoanchor is acid-dissolved and reduced in volume as will be described later. The process may be omitted. In that case, the ratio of the average particle diameter of the transition metal fine particles 8 which are nano-anchors to the average diameter of the nanopits 3 (transition metal fine particle average particle diameter / nanopit average diameter) is 10 nm / 7 nm≈1.4.

1-4. Precious metal layer 7
The noble metal layer 7 is a part that directly acts as a catalyst, and is formed on the transition metal fine particles 8. The noble metal layer 7 preferably has a structure in which the surface of the transition metal fine particle 8 other than the portion in contact with the carbon support 1 is completely covered. The noble metal layer 7 is not particularly limited as long as it is a metal layer containing a noble metal having catalytic activity, but preferably contains at least one metal selected from, for example, Pt, Pd, Rh, In, Ru, and Au. The noble metal layer 7 particularly preferably contains Pt. More preferably, the noble metal layer 7 is preferably made of only Pt.

The thickness of the noble metal layer 7 is not particularly limited, but is, for example, 0.5 to 2 nm. In order to suppress the elution of the transition metal, the thickness is preferably 0.5 nm or more. Further, the thickness of the noble metal layer 7 is preferably 2 nm or less so that mass activity (catalytic activity per unit mass) is not reduced.

1-5. Others The average particle diameter of the catalytic metal fine particles 4 is not particularly limited, but is preferably 1 to 5 nm, and more preferably 2 to 3 nm. This is because catalyst metal fine particles having an average particle size that is too small are not easy to produce stably, and if the average particle size is too large, the mass activity (catalytic activity per unit mass) decreases. Specifically, the average particle diameter of the catalyst metal fine particles 4 is, for example, 1.0, 1.5, 2.0, 2.5, 3.0, 3.5, 4.0, 4.5, 5. 0 nm, which may be in the range between any two of the numerical values exemplified here. In the present specification, the “average particle diameter” means an arithmetic average obtained by measuring the diameter of the circumscribed circle of each particle in the TEM image. The number of measurement samples is, for example, 500 or more.

2. Method for Producing Electrocatalyst As shown in FIG. 3 (FIGS. 3A to 3D), a method for producing an electrode catalyst for a fuel cell according to an embodiment of the present invention includes a supporting step, a heat treatment step, and a noble metal layer forming step. In addition, in the supporting step, the transition metal fine particles 8 are supported on the carbon carrier 1, in the heat treatment step, the carbon carrier 1 supporting the transition metal fine particles 8 is heated, and in the noble metal layer forming step, the noble metal layer is formed on the transition metal fine particles 8. 7 is formed. Hereinafter, each step will be described.

2-1. Supporting Step In the supporting step, fine transition metal fine particles 8 are supported on the carbon support 1 as shown in FIG. 3A. The supporting method is not limited as long as the transition metal fine particles 8 can be supported. For example, vapor deposition, impregnation, reverse micelle method and the like are performed. In the case of vapor deposition, specifically, carrying is performed by vapor deposition using an arc plasma vapor deposition APD apparatus as shown in FIG. The vapor deposition is preferably performed under vacuum exhaust.

The transition metal used in the supporting step is the same as in “1-3. Transition metal fine particles 8”.

2-2. Heat Treatment Step In the heat treatment step, as shown in FIG. 3B, the carbon carrier 1 on which fine transition metal fine particles 8 are supported is heated. By the heat treatment, fine transition metal fine particles 8 are aggregated to form nanopits 3 and / or nanochannels 5 on the carbon support 1, and the aggregated transition metal fine particles 8 are carbonized in the nanopits 3 and / or nanochannels 5. It will be in the state which contacted the support | carrier 1. FIG.

When the Fe nanoparticles as transition metal fine particles 8 are heated and cooled on graphite, a layered structure of ferrite and Fe ₃ C (cementite) is formed in the nanoparticles as shown in the Fe—C phase diagram. It was. Since any of these can be dissolved with an acid, it was expected that after the volume is reduced by acid treatment, a Pt layer, which is a noble metal layer, is selectively formed using the remaining Fe as a core to construct the above-described catalyst structure. However, even if the obtained heated sample was refluxed with hot dilute sulfuric acid for a long time, no decrease in the Fe particle diameter could be confirmed. As a result of high-resolution TEM, STEM, and XPS analysis, Fe nanoparticles are single crystal grains, so there are no grain boundaries where Fe ₃ C is naturally stably deposited, and all solid solution carbon is dense on the Fe particle surface. It was found that it was deposited as a carbon coating layer. That is, depending on the composition of the transition metal fine particles 8 and the heat treatment conditions, the carbon coating layer may be a barrier for noble metal layer formation and acid dissolution. Therefore, it is preferable to appropriately adjust the treatment temperature and the composition of the metal particles.

The heat treatment process is not particularly limited as long as the nanopits 3 and / or nanochannels 5 can be formed. For example, at least one selected from oxygen gas, hydrogen, water vapor, and carbon dioxide is selected under an inert gas flow or vacuum exhaust. It is preferable to carry out under the gas distribution containing.

When the heat treatment step is performed under an inert gas flow or evacuation, a pit-shaped recess is easily formed, so that a state in which the transition metal fine particles 8 are submerged in the nanopit 3 can be formed.
When the heat treatment step is performed under an inert gas flow or under vacuum exhaust, the heating temperature is preferably 500 to 1140 ° C, more preferably 550 to 1000 ° C, and further preferably 580 to 800 ° C. Specifically, the heating temperature is, for example, 500, 520, 540, 560, 580, 600, 620, 640, 660, 680, 700, 720, 740, 760, 780, 800, 820, 840, 860, 880, 900, 911, 920, 940, 960, 980, 1000, 1020, 1040, 1060, 1080, 1100, 1120, 1140 ° C., and may be within a range between any two of the numerical values exemplified here. .

When the heat treatment step is performed under a gas flow including at least one selected from oxygen, hydrogen, water vapor, and carbon dioxide, channel-shaped recesses are easily formed, and the transition metal fine particles 8 are the step portions of the nanochannel 5. A state of being inclined to can be formed.
In the case where the heat treatment step is performed under a gas flow containing at least one selected from oxygen, hydrogen, water vapor, and carbon dioxide, the heating temperature is preferably 300 to 950 ° C., more preferably 350 to 850 ° C., 380 to 820 ° C. is more preferable. Specifically, the heating temperature is, for example, 300, 320, 340, 360, 380, 400, 420, 440, 460, 480, 500, 520, 540, 560, 580, 600, 620, 640, 660, 680, 700, 720, 740, 760, 780, 800, 820, 840, 860, 880, 900, 920, 940, 950 ° C., and may be within the range between any two of the numerical values exemplified here. .

In the heat treatment step, if the heating temperature is too high, the carbon film formed on the transition metal fine particles 8 at the time of cooling as described above may be thick and have a dense crystal structure. It is not preferable because it may be difficult to peel off. On the other hand, if the temperature is too low, solid dissolution of the carbon carrier in the transition metal fine particles does not occur, nanopits or nanochannels are not formed, and further, the effect of the transition metal fine particles as nanoanchors cannot be expected. Therefore, the above temperature range is preferable.

2-3. Noble Metal Layer Formation Step In the noble metal layer formation step, as shown in FIG. 3D, the noble metal layer 7 is formed on the transition metal fine particles 8.

The method for forming the noble metal layer 7 is not particularly limited as long as the noble metal layer 7 can be formed. For example, the noble metal layer 7 is formed by plating. Plating treatment includes, for example, chemical plating, bubbling of a reaction gas containing hydrogen gas in a state where a sample in a dispersed state and a water-soluble noble metal precursor coexist in a water-containing solvent (eg, water) (hereinafter referred to as “hydrogen bubbling”). Or by placing a droplet of a noble metal solution on the sample and circulating hydrogen on the surface of the droplet, the noble metal precursor is reduced on the surface of the transition metal, and the transition metal fine particles 8 are converted into noble metal. With an atomic layer of

Further, when the transition metal material used in the supporting process includes a noble metal, the noble metal layer 7 can be formed also by an acid treatment process described later. That is, the acid treatment process can be regarded as a noble metal layer forming process. Note that the noble metal layer 7 can be formed by the acid treatment process because transition metals such as Fe are more easily dissolved in acid than noble metals, and selective dissolution progresses, and the remaining noble metal atoms aggregate together and become more stable. This is to form a simple layered structure.

2-4. Acid Treatment Step The method for producing a fuel cell electrode catalyst according to an embodiment of the present invention may further include an acid treatment step after the heat treatment step. The acid treatment has several purposes.

One purpose of the acid treatment step is to remove the carbon film that can be formed on the transition metal fine particles 8. In the heat treatment step, carbon may be deposited on the surface of the transition metal that is being cooled after heating to form a carbon film. If such a carbon film is left behind, it may cause a problem in the formation of the noble metal layer. Therefore, it is preferably removed.

Another purpose of the acid treatment step is to dissolve the transition metal fine particles 8 and control the particle size of the transition metal fine particles 8 to be nano-anchors. In the heat treatment step, fine transition metal fine particles are aggregated to form transition metal fine particles 8, and at this time, the particle diameter of the transition metal fine particles 8 increases. If the average particle size of the transition metal fine particles serving as nano-anchors is too large, the average particle size of the catalytic metal fine particles obtained by forming the noble metal layer also increases accordingly, and the mass activity (catalytic activity per unit mass) decreases. Therefore, it is preferable to dissolve to an appropriate size.
For the purpose of dissolution, the method for producing a fuel cell electrode catalyst according to the embodiment of the present invention includes, for example, a supporting step (FIG. 3A), a heat treatment step, as shown in FIG. 3 (FIGS. 3A to 3D). (FIG. 3B), an acid treatment process (FIG. 3C), and a noble metal layer formation process (FIG. 3D).

Another purpose of the acid treatment process is the formation of a noble metal layer. As described above, when the transition metal material used in the supporting step includes a noble metal, the noble metal layer can be formed also by an acid treatment step described later. Therefore, it may be preferable because the noble metal layer can be formed without performing a treatment such as plating.

The acid treatment method is not particularly limited as long as the purpose of the acid treatment can be achieved, and examples thereof include treatment with sulfuric acid, nitric acid, hydrochloric acid and the like. There are no particular restrictions on other conditions such as concentration, temperature, and time, but the conditions should be such that at least the transition metal fine particles 8 serving as nanoanchors remain. This is because the transition metal fine particles 8 serve as a core for forming the catalyst metal fine particles 4 in the nanopits 3 or nanochannels 5 and serve as anchors for retaining the catalyst metal fine particles 4 in the nanopits 3 or nanochannels 5.

A fuel cell electrode catalyst was produced by the following method and subjected to various evaluations.

1. Production Example 1 (Supporting and heat treatment of transition metal fine particles having a strong carbon affinity on the graphite substrate)
In order to grasp in detail the behavior of transition metal nanoparticles such as Fe on the bottom surface of the graphite base, it is suitable to directly observe with a SEM, TEM, or STEM using a substrate sample. Here, a method for producing a substrate sample used for observing nanopits, nanochannels, nanoanchors formed on the graphite base surface by transition metal fine particles will be described.

(1) Fe support on HOPG substrate For SEM observation, a 5 mm square -2 mm thick highly oriented graphite substrate (HOPG) is cleaved to 0.5 mm thickness, and the arc plasma deposition APD of FIG. Fe was vapor-deposited using an apparatus (manufactured by Advance Riko Co., Ltd.). The cleaved HOPG 14 is placed in the vacuum vessel 13 and waits for the internal pressure to become 3 × 10 ⁻⁴ Pa or less. Then, the Fe plasma is charged with 2 Fe from the arc plasma gun 11 equipped with an Fe target under the conditions of an applied voltage of 70 V and a capacitor capacity of 360 μF. Pulse deposition was performed in the range of ~ 20 times. The temperature was room temperature and the substrate was not rotated. When the sample taken out from the apparatus is observed with FE-SEM (manufactured by Hitachi High-Technologies Corporation, SU9000), Fe particles are not observed, but in XPS, Fe2p3 / 2 peaks are observed at binding energies 711 and 707 eV. It is presumed that the partially oxidized Fe clusters of 0.5 nm or less are dispersed on the graphite base surface of the HOPG substrate surface.

(2) Supporting other transition metals on the HOPG substrate Samples were prepared by changing the Fe target of the arc plasma gun 11 to Ni, Co, austenitic stainless steel (SUS304), and hastelloy steel. Moreover, the arc plasma gun 11 and the arc plasma gun 12 were used at the same time to prepare a sample in which Ni, Co, Cr, Mo, V, Ta, and W were added to Fe. Since the amount of vapor deposition per time varies depending on the material, the number of vapor depositions was appropriately changed depending on the observation purpose and composition.

(3) Fe loading on graphene In order to observe the Fe / C interface by TEM and STEM, particularly the nanopit region, Fe particles were formed using graphene, which is a graphite thin film, as a carrier. There are two types of graphene samples used. One is a commercially available graphene nanoplatelet (manufactured by Tokyo Chemical Industry Co., Ltd.) having a thickness of 6 to 8 nm and a width of about 25 μm. After adding this powder to ethyl alcohol and dispersing with an ultrasonic vibrator, It was dripped and dried on the carbon mesh pasting TEM carbon film. The other is a graphene thin layer (6-8 sheets) that is thinner than graphene nanoplatelets, and this is a sample (manufactured by EM Japan Co., Ltd.) that is formed in advance on a perforated silicon nitride support film of a meshed TEM grid. GN-6-10) was purchased and used. The TEM grid diameter is 3 mmΦ.
The APD apparatus was used for the formation of Fe particles as in (1). A dedicated sample holder on which the above two types of graphene samples were fixed was placed in a vacuum vessel, and after vacuum evacuation, Fe was pulse-deposited 20 times under the conditions of 70 V and 360 μF.

(4) Heat treatment for forming nanopits In order to diffuse carbon into Fe particles and form a nanopit region, the Fe / HOPG samples of (1) to (3) were heat-treated in a vacuum or in an Ar stream. A high frequency induction heating furnace (Miwa Seisakusho, MU-αIV-YUNFO2) was used for the heat treatment. This apparatus can raise the temperature of the sample to 2000 ° C. in one minute in a vacuum or Ar flow. The Fe / HOPG sample was put into a graphite crucible, and a lid with a hole was placed, and then placed in a quartz vacuum vessel. The inside of the vacuum vessel was evacuated with a turbo molecular pump, and the temperature was raised after the pressure became 5 × 10 ⁻² Pa or less. The temperature was raised to 600 to 1100 ° C. in 1 to 2 minutes, held for a predetermined time, and then allowed to cool naturally while continuing to exhaust. The cooling rate during natural cooling was approximately 50 to 100 ° C./min. After confirming cooling to room temperature, Ar was introduced into the container and a sample was taken out.
When heating the sample in an Ar stream, after the pressure in the container is reduced to 5 × 10 ⁻² Pa or less by evacuation, the evacuation is stopped and the temperature is raised after flowing Ar of purity 5N at 1 L / min. Started. Although there is no particular difference between the vacuum and the Ar atmosphere, in the case of a system or temperature in which a liquid phase is generated by the reaction between carbon and a transition metal, Ar, which is an inert gas, was used to prevent scattering.

(5) Heat treatment for nanochannel formation If a gasifying agent such as H ₂ , O ₂ , CO ₂ , H ₂ O (water vapor) is present in the atmosphere during the heat treatment, the solid solution carbon in Fe is Fe It reacts with these at the surface and is released into the gas phase. As a result, the Fe particles always move on the bottom surface of the graphite while dissolving carbon, and leave nanochannels as traces of movement.
The substrate sample (1) was heat-treated in a 100% hydrogen stream to form nanochannels. A horizontal fixed flow type quartz reaction tube having a mechanism capable of inserting a quartz boat carrying a sample into the heating unit from the outside was used. The substrate sample of (1) above is set in the reaction tube cooling section, and after purging the reaction tube with N ₂ , H ₂ is circulated at 500 cc / min. When the electric furnace reaches a predetermined heat treatment temperature, the sample boat is inserted into the heating unit at a stretch, held for a predetermined time, and then quickly pulled back to the cooling unit. The rate of temperature rise measured by a thermocouple at the tip of the boat is about 100 ° C./min, and the rate of temperature drop is 50-100 ° C./min. After cooling to room temperature, H ₂ was sufficiently purged with N ₂ and a sample was taken out.
The reason why hydrogen gas was used in this example is that the gasification rate is smaller than that of other gases, so that the nanochannel is not too long, and the Fe particles are not oxidized. For example, H ₂ , O ₂ , CO ₂ , H ₂ O and a mixed gas thereof can be used as well.

2. Production Example 2 (Supporting and heat treatment of transition metal fine particles having a strong carbon affinity in carbon powder)
In Production Example 1, in order to understand the basic phenomenon, model graphite materials such as HOPG and graphene were used, and the APD method was also used for supporting transition metals with a simple operation and less carbon contamination from the raw materials. Here, it is shown that the production of the electrode catalyst can be carried out using a normal carbon powder raw material or a general metal loading method.

(1) Fe support on GCB powder by impregnation method (Preparation of 10 wt% Fe / GCB powder)
5.29 g of iron (III) nitrate nonahydrate (Kanto Chemical) was added to 44.56 g of dehydrated ethanol and stirred at room temperature to completely dissolve it. A suspension was prepared by adding 5.04 g of GCB powder to 150.11 g of dehydrated ethanol, and 37.77 g of the above iron nitrate solution was added, followed by stirring at room temperature for 1 hour. The obtained mixed liquid was transferred to an eggplant-shaped flask and distilled under reduced pressure at 40 ° C. using an evaporator to remove ethanol and obtain a black powder. This powder was dried in air at 110 ° C. for 5 hours using an electric furnace and then calcined in air at 200 ° C. for 2 hours to obtain an Fe oxide / GCB powder.

(2) Fe support on KB powder by impregnation method (preparation of 10 wt% Fe / KB powder)
12.19 g of iron (III) nitrate nonahydrate (Kanto Chemical) was added to 100.37 g of dehydrated ethanol and stirred at room temperature to completely dissolve it. A suspension was prepared by adding 5.01 g of KB powder to 149.95 g of dehydrated ethanol, and 37.53 g of the above iron nitrate solution was added and stirred at room temperature for 1 hour. The obtained mixed liquid was transferred to an eggplant-shaped flask and distilled under reduced pressure at 40 ° C. using an evaporator to remove ethanol and obtain a black powder. This powder was dried in air at 110 ° C. for 5 hours using an electric furnace and then calcined in air at 200 ° C. for 2 hours to obtain Fe oxide / KB powder.

(3) Fe loading on GCB powder by reverse micelle method (Preparation of 20 wt% Fe / GCB powder)
Fe / GCB was also produced by the reverse micelle method characterized by the formation of a uniform particle size. The synthesis was performed using a 100 mL glass reaction tube with stirring in an N ₂ atmosphere. Iron acetylacetonate, Fe (acac) _3, (Ardrich), 1,2-hexadecanediol (260 mg, Tokyo Kasei) were dissolved in diphenyl ether (12.5 mL, Kanto Chemical). After the solution was stirred at 110 ^° C. for 30 min, oleic acid (Kanto Chemical) and oleylamine (ACROS) were added and stirred for 30 min, GCB powder was added, and the temperature was raised to 220 ^° C. and stirred for 30 min. Lithium triethylborohydride (LiBEt ₃ H, Kanto Chemical) was added dropwise to the mixed solution as a reducing agent, and then the temperature was raised to 270 ^° C. and refluxed for 30 minutes to reduce and carry. It was allowed to cool to room temperature and filtered. The obtained catalyst powder was vacuum-dried at 60 ^° C.

(4) Heat treatment for forming nanopits The obtained Fe oxide / GCB powder or 2.00 g of Fe oxide / KB powder was placed on a quartz boat and placed in a water-cooled section in a fixed bed flow type horizontal quartz reaction tube. It inserted in the center part heated at 750 degreeC under nitrogen 200cc / min circulation, hold | maintained for 15 minutes, pulled back, and stood to cool in nitrogen to room temperature. In order to prevent heat generation and ignition of the powder sample due to contact with air, a sufficient amount of 0.25% O ₂ / N ₂ was passed at room temperature and the sample was taken out after being passivated.

(5) Heat treatment for nanochannel formation The obtained Fe oxide / GCB powder or 2.00 g of Fe oxide / KB powder was placed on a quartz boat and placed in a water-cooled section in a fixed bed flow type horizontal quartz reaction tube. It was inserted into the central part heated to 800 ° C. under a flow of 200 cc / min of hydrogen, held for 15 minutes, pulled back, and allowed to cool to room temperature in nitrogen. In order to prevent heat generation and ignition of the powder sample due to contact with air, a sufficient amount of 0.25% O ₂ / N ₂ was passed at room temperature and the sample was taken out after being passivated.

3. Production Example 3 (acid dissolution)
The acid treatment after forming nanopits and nanochannels in Example 1 or Example 2 was performed by the following method.

(1) Acid dissolution of substrate sample The HOPG sample after the nanopit / nanochannel formation heat treatment in (4) and (5) of Production Example 1 was placed in the Erlenmeyer flask with the metal particle support surface facing upward. . Soaked in 80 ml of 1M-H ₂ SO ₄ . While gently stirring H ₂ SO ₄ with a magnetic stirrer, the mixture was heated to 95 ° C. with an oil bath and dissolved for 3 hours. Since the dissolution rate is not constant depending on the preparation conditions of the acid-dissolving sample, the acid concentration, treatment temperature, and time were appropriately changed. The Erlenmeyer flask was equipped with a reflux condenser to keep the H ₂ SO ₄ concentration constant. Bubbles adhering to the sample surface were removed each time. After completion, the HOPG sample was thoroughly washed with distilled water and dried in a desiccator.

(2) Acid dissolution of powder sample Fe / GCB and Fe / KB powder samples subjected to heat treatment for forming nanopits and nanochannels in (4) and (5) of Production Example 2 were also heated by H ₂ SO ₄ in the same manner as the substrate samples. Processing was carried out. In a round bottom flask equipped with reflux cooling, 100 ml of 0.1M-H ₂ SO ₄ was put and mixed with 1.00 g of Fe / GCB powder or Fe / KB powder while stirring with a magnetic stirrer. Thereafter, the temperature was raised to 95 ° C. with stirring in an oil bath, and Fe / GCB was refluxed for 3 hours and Fe / KB was heated for 2 hours. After the round bottom flask was sufficiently cooled with ice water, the treated powder sample was collected by filtration. The powder sample was dried under reduced pressure after repeated washing with distilled water and filtration.
Since the acid treatment time, temperature, and sulfuric acid concentration vary depending on the sample to be treated and the purpose, it is desirable to determine appropriate conditions by preliminary tests. In this example, during the acid treatment, a small amount of H ₂ SO ₄ suspension was extracted with a pipette, and the particle dissolution state of the washed powder was observed with an SEM to determine the acid treatment end point.

4). Production Example 4 (Selective Pt Chemical Plating on Nano Anchor)
As a method of selectively forming Pt only on the anchor using the nano-anchor formed in Production Example 3 as a core, the following chemical plating was performed.

(1) Pt selective chemical plating on powder sample As a method of selective Pt chemical plating on a powder sample, the method described in Patent WO2014 / 178283A1 was applied. Tetraammine platinum hydroxide aqueous solution (Tanaka Kikinzoku Co., Ltd., Pt concentration 20.97 g / L) 6.00 ml (equivalent to Pt 0.126 g) and distilled water 20 ml were acid-treated in Production Example 3 (2) Each of Fe / GCB powder and Fe / KB powder of 0.5 g was put into a sealed glass container with a nozzle shown in FIG. 13A. The amount of Pt supported when all of the charged Pt is deposited on the catalyst is 20 wt%. At room temperature of 20 ° C., the suspension was boiled for 2 minutes with a heater while stirring with a magnetic stirrer and bubbling with N ₂ gas at 200 cc / min. After cooling to 20 ° C. with ice water, the N ₂ gas was stopped and bubbling was performed for 3 hours with 100% H ₂ gas at 100 cc / min. In the course of 3 hours, tetraammineplatinum hydrate is reduced by hydrogen, and Pt selective precipitation using nanoanchors as seed crystals occurs. When the treatment was completed, the powder was collected by filtration and washed with warm water sufficiently, and then dried under reduced pressure for 1 hour to obtain a Pt-plated sample.

(2) Pt selective chemical plating on a substrate sample In the case of a substrate sample, the number of nano-anchors on the substrate is only about 1/1000 that of a powder sample. In the method (1), the solution concentration becomes extremely thin and reduction deposition by hydrogen gas is performed. There is a concern that the selectivity of this will also decline. Further, when an aqueous chloroplatinic acid solution was used, hydrolysis occurred due to high dilution, and Pt plating itself could not be performed. In order to prevent hydrolysis, the pH was adjusted to the acidic side with hydrochloric acid. As a result, precipitation of Pt due to dissolution of Fe occurred in the entire substrate, and selective plating of Pt could not be realized.
Therefore, for the substrate sample, as shown in FIG. 13C, a method of placing a droplet of a Pt salt solution on the surface of the substrate and circulating hydrogen on the surface of the droplet was newly devised and implemented.
0.2 ml of tetraammine platinum hydroxide aqueous solution (Tanaka Kikinzoku Kogyo Co., Ltd., Pt concentration 20.97 g / L) was collected with a micropipette, diluted to 200 ml with a volumetric flask, and Pt 2.097 × 10 ⁻⁵ g / ml. An aqueous solution of was prepared. 20 μl of the diluted Pt solution is dropped with a micropipette onto the heat-treated Fe / HOPG sample prepared in Production Example 1 (4) and (5) or the acid-dissolved Fe / HOPG sample in Example 3 (1). It hold | maintained and it installed in the container like FIG. 13B. Water 25 is placed in the lower part of the bottle so that the droplets on the substrate do not evaporate due to the gas flow, and the upper part is brought into contact with gas or bubbled to maintain the internal humidity. After placing the sample and droplets, 200 cc / min of N ₂ was circulated for 10 minutes, then switching to 100 cc / min H ₂ to start reduction, holding for 3 hours, and then switching to N ₂ again to complete the reduction I let you. After completion, the substrate was sufficiently washed in boiling water and vacuum dried.

5. Comparative Example A sample in which Pt is singly supported on a HOPG or KB powder carrier without nanopits, nanochannels and nanoanchors was prepared according to the method of Production Example 1 or a commercial product was obtained and compared with the catalyst of the present invention. Used for. The manufacturing method is shown below.

(1) Pt / HOPG substrate sample (arc plasma deposition method)
A HOPG having a 5 mm square and a 0.5 mm thickness immediately after cleaving was installed in an APD apparatus in the same manner as in Production Example 1 (1), and applied with an applied voltage of 100 V and a capacitor capacity of 1080 μF from an arc plasma gun 12 on which a Pt target was installed. Pt was pulsed twice. The pressure was 3 × 10 ⁻⁴ Pa or less, the temperature was room temperature, and the substrate was not rotated. Also in the case of Pt, Pt particles are not observed by SEM in this state. Since Pt hardly dissolves carbon and does not form carbides, it is considered that no pits are formed regardless of the heating temperature. Heating was performed at 600 ° C. for 1 hour in an ultra high vacuum of 10 ⁻⁵ Pa so that the Pt particle size became an observable particle size in an infrared image furnace.

(2) Pt / KB powder sample (commercially available)
Similar to the procedure of Production Example 1 (3), a commercially available Pt / KB powder (manufactured by Tanaka Kikinzoku Kogyo Co., Ltd., TEC10E50E) was added to ethyl alcohol and dispersed with an ultrasonic vibrator, and then a carbon film for TEM was attached. It was dripped and dried on the attached Mo mesh.

(3) Pt / KB powder sample (arc plasma deposition method)
In the same manner as in Production Example 1 (3), KB powder, which is the same carbon carrier as in Comparative Example (2), was added to ethyl alcohol and dispersed with an ultrasonic vibrator, and then applied to a Mo mesh with TEM carbon film. Dropped and dried. A TEM grid to which KB was fixed was placed in a vacuum vessel with a dedicated sample holder, and after vacuum evacuation, Pt was pulse-deposited 5 times at 100 V, 1080 μF and room temperature. Heating was performed at 600 ° C. for 1 hour in an ultrahigh vacuum of 10 ⁻⁵ Pa using an infrared image furnace.

6). Catalyst Observation and Analysis Specific examples of the present invention are shown in Production Examples 1 to 4 for each production process. Here, the results of direct observation of the transition metal fine particles formed on the fabricated sample and the nanopits, nanochannels, and nanoanchors formed by the electron microscope are shown.

(1) Fe nanopits on graphene by vacuum heating The Fe-supported graphene thin layer (APD, 70V-360 μF, 20shot) prepared in Production Example 1 (3) was subjected to 600 ° C. in the vacuum exhaust of Production Example 1 (4). Heating for nanopit formation for 1 hour was applied, and observation was performed with a transmission electron microscope (TEM, manufactured by Hitachi High-Technologies Corporation, H-9500). A typical result is shown in FIG. 5A. The Fe particles are dispersed on the graphene base bottom as spherical particles (black) of 5 to 10 nm. A substantially circular white region can be confirmed around some black particles. This is a nanopit formed on the graphene base bottom surface by Fe particles. FIG. 5B is an end view schematically showing the nanopit structure. In this sample, the Fe particles 9 not only sink immediately below the graphene, but also rotate or vibrate to form nanopits 3 that are slightly larger than their own particle diameter. This is because the atmosphere during the heat treatment is not always as clean as 5 × 10 ⁻² Pa, and it seems that a very small amount of oxygen present in the atmosphere partially gasifies carbon in which Fe is dissolved. In fact, in the sample heat-treated under a high vacuum of the order of 10 ⁻⁵ Pa, no clear white color was observed around the particles. An example is given below.

(2) Fe nanopits on graphene by ultra-high vacuum heating Compared to another Fe-supported graphene nanoplatelet (APD, 70V-360 μF, 20shot) produced in Production Example 1 (3), it exceeds 10 ⁻⁵ Pa Under high vacuum evacuation, heating for nanopit formation at 600 ° C. for 5 hours was applied. FIG. 6A and FIG. 6B show the results of observation with a scanning transmission electron microscope (STEM, manufactured by Hitachi High-Technologies, HD-2700). Each shows an approximately 10 nm Fe particle and its graphene interface. In FIG. 6A, an Fe diffusion layer formed in the graphene layer from the Fe particles is observed. In FIG. 6B, it can be clearly seen that several layers on the bottom surface of the graphite are eroded and Fe particles are submerged. From the diffraction grating images of the Fe particles in FIGS. 6A and 6B, these particles are composed of a single crystal. It can be confirmed that a layer clearly different from Fe is formed on the Fe particle surface. This is a graphene carbon diffused in the Fe crystal lattice during heating, which is precipitated with cooling.

(3) Fe nanopits on HOPG by vacuum heating The Fe-supported HOPG sample (APD, 70V-360 μF, 2shot) produced in Production Example 1 (1) was 750 ° C. under the vacuum exhaust of Production Example 1 (4). The sample was observed with a scanning electron microscope (SEM, manufactured by Hitachi High-Technologies Corporation, SU-9000) after heating for nano-pit formation for 15 minutes. The surface of the HOPG substrate is composed of a graphite basal plane, but unlike the graphene sample described above, it is a polycrystalline body, and therefore there are grain boundaries and steps on the surface. Representative surface images are shown in FIGS. 7A and 7B. In FIG. 7B, the grain boundaries run diagonally. In each case, Fe particles of 10 to 15 nm are uniformly dispersed on the HOPG surface. Although it cannot be determined from the surface whether each Fe particle sinks into the HOPG and forms pits, the particle diameter around a part of the particles in FIG. 7B is slightly larger than the particle diameter seen in FIG. 5A. A pit can be observed as well. From this, it is considered that each Fe particle is also submerged due to the solid solution of carbon. The details will be described later in the acid-dissolved sample.

(4) Nanopits of other transition metals on HOPG by vacuum heating Each sample of transition metal supported on HOPG in Production Example 1 (2) was 750 ° C., 15 minutes under the vacuum of Production Example 1 (4) The nanopit formation heating was added and the surface was observed with SEM. In either case, the same results as in FIG. 7A were obtained. Among them, in Ni, most of the particles formed on the surface were not spherical particles such as Fe, but had an angular shape.

(5) Fe nanochannel on HOPG by heating in hydrogen For Fe-supported HOPG sample (APD, 70V-360 μF, 20shot) produced in Production Example 1 (1), under the hydrogen gas flow of Production Example 1 (5) The heat treatment was applied and observed by SEM. The heat treatment conditions under hydrogen gas flow were as follows: first, heat treatment was performed at 500 ° C. for 1 hour, and then heat treatment was further performed at 800 ° C. for 15 minutes. In the SEM observation, the sample was tilted by 40 ° in order to grasp the state of the Fe particle and HOPG interface. The results are shown in FIGS. 8A and 8B. In FIG. 8A, nanochannels are clearly recognized, and Fe particles A forming the channels can be confirmed at the head of the channels. Not all Fe particles form channels, and there are many particles that remain in one place, as in the case of heating in vacuum. FIG. 8B is the high-magnification image of FIG. Both Fe particles A and Fe particles B appear to sink their bottoms on the graphite base surface of HOPG. The only difference between them is that there is a nanochannel behind the particle A. When heated in the presence of not only hydrogen but also a gasification gas, when the temperature is relatively low, Fe particles in contact with the graphite crystal step form nanochannels. When the heating temperature is increased, not only the step particles but also the particles existing on the bottom surface of the smooth base such as the particles B form nanochannels. In the case of FIG. 8, the Fe particles in the step mainly form nanochannels.

(6) Fe / GCB heated in hydrogen
The Fe-supported GCB powder catalyst produced in Production Example 2 (1) was subjected to heat treatment for forming nanochannels (in a hydrogen stream at 400 ° C. for 4 hours) and observed with TEM. The results are shown in FIG. 9 (a) and FIG. 9 (b). Due to the strong water-repellent carrier, the Fe particles supported by the impregnation method are slightly larger than the above examples, and FIG. 9 (a) is 15 nm and FIG. 9 (b) is 40 nm. The formation of clear nanochannels and nanopits could not be confirmed because the particles were not smooth and the Fe particles were slightly larger. However, since the formation of precipitates can be confirmed in the Fe particles (upper part) of FIG. 9B, it is suggested that solid solution of carbon has occurred from the GCB powder, suggesting the formation of nanopits in the powder sample. doing.

(7) Confirmation of HOPG nanopits by complete dissolution of Fe Nanopits formed in Fe / HOPG cannot be directly confirmed by SEM. Therefore, all the Fe diffused in the Fe particle body and HOPG was dissolved by sulfuric acid, and it was confirmed whether nanopits were really formed. As before, Fe / HOPG was loaded with Fe (100 V, 1080 μF, 2 shots) on the APD of Production Example 1 (2), and then at 600 ° C. for 1 hour and 950 ° C. for 20 minutes in the argon stream of Production Example 1 (4). The nano pit formation heating was added. FIG. 10A is an SEM image of the sample surface before the sulfuric acid treatment. This was treated with sulfuric acid according to the method of Production Example 3 (1). In order to completely dissolve Fe, the sulfuric acid concentration was increased to 0.5 M, and the mixture was further refluxed for 1.5 hours. FIG. 10B shows the same field of view observed after processing. All the white spherical Fe particles observed in FIG. 10A are dissolved and removed, and a circular trace can be clearly confirmed at the place where each particle is present. This is a nanopit formed by each Fe particle formed by dissolving carbon on the bottom surface of the graphite.

(8) Fe nanoanchor on HOPG after acid dissolution The surface of a sample in which Fe particles were not completely dissolved by the acid treatment and a part of the Fe nanoanchor remained as nanoanchors was observed by SEM. Heating at 600 ° C. for 1 hour and 950 ° C. for 10 minutes was added to the Fe-supported HOPG (100 V, 1080 μF, 2shot) produced in Production Example 1 (2) under the argon stream of Production Example 1 (4). This sample was subjected to the sulfuric acid treatment of Production Example 3 (1) (0.5 MH ₂ SO ₄ , refluxed at 85 ° C. for 12 hours). The sample surface after the acid treatment is shown in FIG. 11 (a) and FIG. 11 (b). Large Fe particles disappear, and a white portion can be confirmed at the center of the concave portion of the nanopit remaining on the HOPG surface. In particular, the ones with arrows are the portions where white particles can be clearly confirmed.

(9) Nano-anchor by acid dissolution of FePt / HOPG In the present invention, Pt is selectively deposited on the nano-anchor formed in the nano-pit in the next step. As an improved method, Pt is deposited in the next step. Instead of this, an example in which a predetermined amount of Pt is mixed and supported in Fe will be described.
When the APD apparatus of Example 1 (1) was used to deposit Fe 70V, 360 μF, 20 times, Pt was vaporized 70 V, 360 μF, once using another plasma gun. When the FePt / HOPG immediately after deposition was analyzed by XPS, the Pt concentration was 14 atm%. The nanopit formation heat treatment of Example 1 (4) (in an argon stream, 600 ° C. for 1 hour + 850 ° C. for 15 minutes) was immediately added thereto, and then immediately after the acid dissolution treatment of Example 3 (1) (0.1 MH ₂ SO ₄ , 95 ° C. for 3 hours). 12A is a SEM image of the sample surface before the acid treatment, and FIG. 12A (b) is an SEM image of the same visual field after the acid treatment. Also, (c) in FIG. 12B and (d) in FIG. 12B are high-magnification images of (a) in FIG. 12A and (b) in FIG. 12A. It can be confirmed that the diameter of each Fe particle before the acid treatment is reduced by the acid treatment, and all of the Fe particles are greatly submerged in the nanopits. The Pt concentration in (a) of FIG. 12A is 19 atm%, indicating that the Pt concentration on the particle surface was concentrated from 14 atm% at the time of precipitation by heat treatment. Furthermore, the Pt concentration after the acid treatment in FIG. 12A (b) was 100 atm%. This means that Fe is eluted from the particle surface by the acid treatment, and the surface is finally covered only with Pt.
In acid dissolution, prior examination and careful operation are necessary so that Fe as a nano-anchor is not completely dissolved. However, the operation of acid treatment can be performed by mixing a transition metal with Pt of a predetermined concentration in advance. Not only becomes easy, but it becomes possible to make the Pt selective deposition in the next process simple or unnecessary. That is, it is suggested that the acid treatment step also serves as a noble metal layer forming step.

(10) Selective chemical plating of Pt on nanoanchor Production Example 4 was carried out to selectively deposit Pt on the nanoanchor shown in FIG. 11, and Pt was deposited on the nanoanchor and its particle size increased. It was confirmed. FIG. 14 and FIG. 15 investigate the influence on chemical plating when Pt coexists in the nanoanchor separately. FIG. 15 is a graph (a), graph (b), and graph (c) showing changes in the Pt4f peak by the X-ray photoelectron spectrometer (XPS). Pt in Run A, Run B, and Run C in FIG. Changes in XPS intensity of Pt before and after selective chemical plating (“initial” and “after plating”) are shown in the graphs of FIG. 15A, FIG. 15B, and FIG. 15C, respectively. . In either case, Pt is selectively deposited on the particles, and no precipitation of Pt can be confirmed on the substrate. However, when the Pt concentration on the particle surface measured by XPS is as high as 82 atm% and 99 atm%, the increase in the Pt intensity of XPS after plating is large. Although Pt plating can be performed using a different metal as a core, it is understood that it is better to partially mix Pt in advance for practical plating in a short time. It was shown that the presence of Pt at an appropriate concentration on the transition metal surface facilitates the aforementioned acid dissolution and the subsequent Pt selective chemical plating process.

7). Catalyst migration aggregation test In order to verify the effect of the present invention, the potential of the catalyst for each sample of Comparative Examples (1), (2), and (3) and the Pt / HOPG sample with nano-anchor in FIG. The resistance to migration aggregation due to fluctuation was compared and evaluated.
Using a glass electrochemical cell, an evaluation sample fixed with an Au wire was suspended in an N ₂ saturated 0.1 M HClO ₄ electrolyte, a Pt plate was used as a counter electrode, and a reversible hydrogen electrode (RHE) was used as a reference electrode. At an electrolyte temperature of 65 ° C, using a potentiostat (HZ-5000, manufactured by Hokuto Denko), a potential step (0.6, 1.0V) conforming to the load variation simulation test method of the Fuel Cell Practical Use Promotion Council (FCCJ) (Step cycle every 3 seconds) was applied at 1000 to 3000 cycles. After the HOPG sample was washed with water and dried, the state of migration and aggregation of Pt particles before and after the cycle application was directly observed by SEM.

(1) APD method Pt / HOPG without nanoanchors
The SEM image (a) and SEM image (b) in FIG. 16 are applied to the Pt / HOPG sample of Comparative Example (1), which compares the changes in Pt particles when the potential fluctuation is applied 1000 times at room temperature. It is a SEM image before and behind. In FIG. 16A before the cycle application, the Pt particles were almost spherical, and the measured Pt particle size was 2.4 ± 0.6 nm. On the other hand, in FIG. 16 (b) after 1000 times of application, it is possible to confirm elongated particles that are apparently combined with a plurality of particles, and the measured average particle size is as large as 2.8 ± 0.8 nm. Yes. With the KB powder catalysts described later in (2) and (3), even when a potential step cycle was applied at room temperature, the growth of Pt particles was smooth, and the graphite base had little interaction with Pt particles compared to the fact that no Pt particle growth was confirmed. The results show how easily the particles move on the surface.

(2) Commercial Pt / KB powder The SEM image (a) and SEM image (b) in FIG. 17 are compared with the commercial Pt / KB catalyst powder (TEC10E50E) fixed to the TEM holder of Comparative Example (2). It is the SEM image before and behind application which applied the potential step cycle 3000 times at the temperature of 65 degreeC, and compared the Pt particle | grain change before and after that. A step cycle test at room temperature was carried out, but no significant particle change was observed even after 3000 applications, but the changes in the same field of view were compared at a higher temperature of 65 ° C. where mobile aggregation is likely to proceed. In FIG. 17A before application of the step cycle, the Pt particles were 3.2 ± 1.3 nm, but in FIG. 17B applied 3000 times, it was 5.9 ± 1.8 nm. It can be seen that the individual particle sizes are clearly increased. In the case of a KB carrier, since Pt supported also in the pores inside the particles exists, it appears that this moved to the surface and aggregated by application of a step cycle.

(3) APD method Pt / KB powder The SEM image (a) and SEM image (b) in FIG. 18 are 3000 potential steps at a temperature of 65 ° C. with respect to the Pt / KB catalyst powder of Comparative Example (3). It is a SEM image before and after the application which applied the cycle and compared the Pt particle change before and after that. The carrier of this sample is the same KB powder as the commercially available catalyst in (2), but Pt is supported by the same APD as in (1). In the commercially available catalyst, Pt particles are supported on the pores of KB. However, due to the nature of the APD method in this sample, Pt particles are not supported in the pores and exist only on the outer surface of KB. This sample was also applied with 3000 potential step cycles at a temperature of 65 ° C. In FIG. 18A before application of the step cycle, the Pt particle size is 3.1 ± 0.6 nm, which is almost the same particle size as that of the commercially available catalyst, but the particle size distribution is narrow and uniformly distributed. After 3000 times of (b) in FIG. 18, the particles clearly aggregate and increase to 4.4 ± 1.0 nm. However, since there are no Pt particles inside the pores, the increase in particle size is suppressed compared to commercially available catalysts.

(4) APD method with nano anchors / HOPG
For Pt / HOPG with Fe nanoanchor, 1000 potential step cycles at 65 ° C. were applied. No change was observed on the sample surface before and after application by SEM.

From the above results, it can be seen that in the case of catalytic metal fine particles having transition metal fine particles as nanoanchors, the migration and aggregation of the catalytic metal fine particles supported on carbon is suppressed.

8). Production Example 5 (Effect of Fe—Ni alloy and its composition)
As can be easily seen from the Fe-Ni phase diagram, when Ni is added to Fe, the austenite phase (fcc) region is greatly expanded to a low temperature. As a result, the carbon solid solution amount in the FeNi alloy shows a larger value than in the case of pure Fe in many temperature ranges. Therefore, by applying Fe-Ni alloy as the transition metal, larger pits can be formed even under the same heat treatment conditions, and a dense and thick carbon layer on the surface of the transition metal particle that becomes an obstacle in the acid treatment and noble metal layer formation process It is expected that the precipitation of can be suppressed.

Here, the above effect is taken as an example in the case where the nanopit forming heat treatment (4) is performed on the sample carrying Fe and Ni on the HOPG substrate by the two arc plasma guns in (2) of Production Example 1 respectively. explain.
(1) Fe—Ni support and heat treatment The arc plasma deposition conditions for Fe and Ni were an applied voltage of 70 V and a capacitor capacity of 360 μF in a vacuum degree of 3 × 10 ⁻⁴ Pa or more. Samples having different compositions were prepared by changing the number of pulse depositions of Fe and Ni while keeping the amount of Fe + Ni supported constant. The Ni concentration Ni / (Fe + Ni) determined from XPS after loading was 15, 26, and 44 atm%. The total supported amount of Fe + Ni was confirmed to be in the range of 0.3 to 0.4 μg / cm ² from the particle size distribution result of the SEM image after the heat treatment for forming nanopits.

The nano pit formation heat treatment of the sample was carried out in a high-frequency induction heating furnace at a temperature of 800, 850, and 900 ° C. in an Ar stream for 30 seconds, held for 5 seconds, and then naturally cooled. FIG. 19 shows a representative SEM image of the surface of each sample after heat treatment in comparison with the case of Fe 100%. At a heat treatment temperature of 900 ° C., the particles have a spherical shape regardless of the Ni concentration, but at low temperatures of 850 ° C. and 800 ° C., the particle shape tends to be slightly flat due to the addition of Ni. From the spot analysis of each particle by EDX, all the particles were FeNi alloy although there was a composition variation.

The average particle diameter of FeNi alloy particles was determined from each SEM image at a heat treatment temperature of 900 ° C., and the results plotted against the Ni concentration are shown in FIG. Here, the result of Ni 100% is also shown. In samples other than Ni15 atm%, the average particle size was 9-12 nm and the standard deviation was relatively close to ± 4-5 nm. However, in Ni15 atm%, the average particle size was nearly doubled to 19 nm and the standard deviation was ± 6 nm. It showed a large value specifically. The curve indicated by the broken line in FIG. 20 indicates the amount of carbon that can be dissolved in the austenite phase of each FeNi composition at 900 ° C. The carbon amount is obtained from the Fe—Ni—C ternary phase diagram at 900 ° C. It was. Both tendencies for the Ni concentration are in good agreement, and in the dynamic process in which carbon atoms diffuse and dissolve in the transition metal lattice, the metal particles easily move on the carbon surface and the collision frequency between the particles increases. The particle size may increase.

As described above, the average diameter of the transition metal particles is 0.5 to 10 nm, particularly preferably 0.5 to 5 nm. When the transition metal particles are increased to 19 nm, not only the step of reducing the particle size by acid treatment increases, but also noble metals. The number density of the catalyst metal fine particles after the layer formation is lowered, and the performance per unit weight of the catalyst is lowered. Such remarkable grain growth occurring in a specific Ni concentration region can be easily avoided by lowering the heat treatment temperature. FIG. 21 shows the change in the average particle diameter of the FeNi alloy particles after a sample having a Ni concentration of 15 atm% is lowered from 900 ° C. to 850 ° C. and 800 ° C. and heated in the same manner for 5 seconds. It was confirmed that both the average particle diameter and the standard deviation can be reduced to 19 ± 6 nm at 900 ° C., 11 ± 6 nm at 850 ° C., and 12 ± 4 nm at 800 ° C. The amount of carbon solid solution at both temperatures in this composition is unknown because no Fe-Ni-C ternary phase diagram has been reported so far. Along with this, it is expected to gradually decrease. Nevertheless, as shown in FIG. 22, when heated at 850 ° C. for 60 minutes, the particle size of Fe 100 atm% did not increase significantly compared to heating for 5 seconds, but with Ni 15 atm%, particle growth progressed and the average particle size was rather than Fe 100 atm%. It is increasing. As for the heating conditions in the FeNi alloy system, it is necessary to select a proper heating time at a low temperature in consideration of the amount of carbon solid solution.

(2) Solubility of Fe—Ni by acid treatment The 900 ° C. heat-treated sample of FIG. 19 was acid-treated with sulfuric acid, and the difference in solubility of FeNi alloy particles depending on the composition was compared. The acid treatment was selected under milder conditions than in the previous examples so that differences between samples could be clearly compared. Each sample after heating was immersed in 0.05M-H ₂ SO ₄ and stirred at room temperature for 30 minutes. Then, after thoroughly washing the sample in boiling water, it was heated in a hydrogen stream at 300 ° C. for 30 minutes to remove the adhering / adsorbed material on the surface. The same field of view as before the acid treatment was searched by SEM, and the shape change of FeNi alloy particles and the presence or absence of pit formation were observed.

FIG. 23 shows SEM images of the same visual field before and after acid treatment of each sample. Except for the sample having a Ni concentration of 15 atm%, most of the Fe or FeNi alloy particles present on the HOPG before the acid treatment are dissolved. Further, pits and / or fine particles remaining without being dissolved can be confirmed at the positions where the particles existed. In particular, when the Ni concentration was 26 atm%, pits could be confirmed most clearly in the sample. This is presumably because the amount of carbon solid solution is the second largest after the Ni concentration 15 atm% sample, as expected from the carbon solid solution amount curve in FIG. On the other hand, in the sample having a Ni concentration of 15 atm% with the largest amount of carbon solid solution in the sample, most of the FeNi particles remain in an almost intact size even after the acid treatment. Although the amount of carbon solid solution was greatly increased by the addition of Ni, it is expected that acid dissolution was suppressed because the carbon layer deposited during cooling was thick (or dense).
In order to satisfy both pit formation and acid dissolution, it means that there is an appropriate amount of carbon solid solution, and the value is 1.2 to 4.5 atm% from FIG.

24. SEM images (a) and SEM images (b) in FIG. 24 are obtained by examining the solubility when a sample having the same Ni concentration of 15 atm% is heat-treated at 850 ° C. Although the same field of view was not found and comparison was made between different fields of view, the FeNi particles that were flat before acid treatment have rounded corners and a hemisphere after acid treatment. Similar to the 900 ° C. heat-treated sample, the FeNi alloy particles are not completely dissolved, but the effect of lowering the amount of carbon solid solution by lowering the temperature is obtained. Moreover, when the Ni concentration of the sample after acid treatment was analyzed by XPS, it increased from 15 atm% before acid treatment to 38 atm%. It can be seen that the dissolution rate of Fe is large and Ni concentration occurs on the particle surface. Since Ni has a smaller lattice misfit with Pt than Fe, the selective Pt chemical plating step after acid treatment becomes easier. It is important for ensuring the stability of battery performance that the formation of a dense Pt layer on Ni can suppress the elution of transition metal ions during actual operation as compared with the case of Fe alone.

(3) Fe—Ni initial mixed state The heat treatment in each of the FeNi examples employs a short-time heat treatment in which the temperature is raised to a predetermined temperature in 30 seconds and held for 5 seconds. This is because the increase in the amount of carbon solid solution due to the addition of Ni promotes the surface movement of the particles, and it is thought that the long-time heat treatment results in an excessive increase in the FeNi alloy particle diameter. Here, the influence on the FeNi alloy particle diameter after the heat treatment by the support pattern of Fe and Ni by arc plasma deposition was examined.
FIG. 25 shows an outline of a transition metal support pattern by an arc plasma deposition method. (A) is the pattern used when carrying only Fe alone. An applied voltage of 70 V and a capacitor capacity of 360 μF are applied 10 times at 1 minute intervals to an arc plasma gun provided with an Fe target, and 10 shots of pulse deposition are performed. An interval of 10 to 15 minutes was provided for every 10 shots in order to cool the tip of the arc plasma gun and obtain stable plasma emission. This operation was repeated so as to meet the target Fe loading. (B) and (c) are patterns used for supporting the FeNi binary system. (B) Alternating support is a method in which, after 10 shots of Fe deposition are performed by the arc plasma gun 11, Ni deposition is performed from another arc plasma gun 12, and pulse deposition of Fe and Ni is alternately repeated without synchronization. It is. All of the above-described FeNi support by arc plasma deposition was performed in this (b) alternating support pattern.

(C) On the other hand, synchronous loading is performed by synchronizing pulse deposition of Fe and Ni from the

arc plasma guns

11 and 12 in synchronization. FIG. 26 is an SEM image after a sample having a Ni concentration of 47 atm% produced using this supporting pattern was heated in an Ar air flow at 900 ° C. for 5 seconds. In this case, since the number of shots of Fe was 30 and Ni was 20 shots, 10 shots of Fe were vapor-deposited first, and then 20 shots of Fe and Ni were synchronously carried. A characteristic point of this sample is that it can be confirmed that many FeNi alloy particles have already been buried in deep pits on the HOPG substrate in the heating stage. The average particle size is also as small as 9 ± 4 nm and aggregation is suppressed. The reason why such deep pits that can maintain the average particle diameter of 10 nm or less and that can be confirmed without acid treatment is not clear at present, but is formed by the synchronous loading of FIG. This is probably because the structure of Fe and Ni clusters (FeNi alloy particle precursors) is suitable for quickly forming deeper pits than the alternate loading shown in FIG. That is, since the initial mixed state of Fe and Ni atoms formed by (c) in FIG. 25 is suitable for quickly shifting to the Fe—Ni austenite structure in the phase diagram, it is immediately under the contact with the temperature rising process of the heat treatment. The carbon may be dissolved in the crystal and pits may be formed before the particles begin to move.

9. Production Example 6 (Low-temperature heat treatment using Fe—Ni—Pt alloy particles)
In this invention, after forming the nano anchor which consists of transition metals in a nanopit, an electrode catalyst is produced by performing the process of depositing Pt selectively on the nano anchor. Here, it has already been shown in the case where the transition metal is Fe that the selective Pt deposition process can be simplified or eliminated at all by supporting a predetermined amount of Pt when the transition metal is supported. In the present embodiment, an example in which Pt is supported together with Fe and Ni when the transition metal is Fe—Ni-based is shown.

In the arc plasma deposition, Fe, Ni, and Pt targets were installed on three arc plasma guns, respectively, and alternately supported on the HOPG substrate with an applied voltage of 70 V and a capacitor capacity of 360 μF. The Fe: Ni: Pt composition analyzed by XPS was 62:28:10 atm%.

(A) of FIG. 27 is a typical SEM image of the sample surface after being heated up to 700 ° C. in an Ar air flow for 30 seconds and held for 10 seconds by a high-frequency induction heating furnace. (B) in FIG. 27 is an SEM image after (a) is subjected to acid treatment with 0.005M-H ₂ SO ₄ for 10 minutes at room temperature, washed with boiling water and heated at 300 ° C. for 30 minutes in a hydrogen stream. is there. Both observed the same visual field, and the change before and after acid treatment can be clearly seen.

First, the FeNiPt particles are in the form of a flat plate by heat treatment at 700 ° C. for 10 seconds in FIG. 27A, but after acid treatment, all of these particles are greatly reduced in size due to dissolution. One or a plurality of white grains can be confirmed at the same location where each tabular grain is present. As a result of XPS analysis, although the intensity is small, peaks of Pt4f5 / 2 and Pt4f7 / 2 are clearly recognized. The peaks of Fe and Ni were noise levels.
Next, in FIG. 27A, it can be confirmed that pits are clearly formed in the portion where the particles exist. The portion indicated by the arrow in FIG. 27B is particularly remarkable, and it can be confirmed that small white particles of several nm have fallen into the pit. In other places, shallow pits corresponding to the shape of the tabular grains are drilled, and white grains having a greatly reduced diameter exist. In the Fe—Ni—Pt ternary system, it was confirmed that pits were formed even at a low heat treatment temperature that could not be formed only by Fe at 700 ° C., and that Pt particles were selectively formed in the pits.

Another feature is that Fe and Ni are almost completely dissolved even under mild acid treatment conditions at 0.005M-H ₂ SO ₄ at room temperature of 10 minutes. Like Ni, Pt is also known as an element having an effect of extending the austenite phase of Fe. It can be seen from the Fe—Pt phase diagram that the austenite phase is stabilized to near room temperature by the addition of Pt. There are few reports on the phase diagrams of the current Fe-Ni-Pt ternary system and Fe-Ni-Pt-C quaternary system, and the stable region of the austenite phase and the amount of carbon solid solution are unknown. From the results, it has been confirmed that the Fe—Ni—Pt system is a material system that is extremely suitable for the stable electrode catalyst to which the transition metal of the present invention is applied and the production method thereof.

1: carbon support, 2: catalyst particles, 3: nanopits, 4: catalyst metal fine particles, 5: nanochannel, 7: noble metal layer, 8: transition metal fine particles, 9: Fe particles, 11: arc plasma gun 1 (transition metal) 12: Arc plasma gun 2 (for precious metal), 13: Vacuum container, 14: Carrier sample, 15: Sample carrying container, 16: Vacuum exhaust system, 17: Glass bottle with sealing stopper, 18: Gas introduction nozzle, 19: gas exhaust pipe, 20: nitrogen / hydrogen, 21: Pt raw material salt solution, 22: rotor and magnetic stirrer, 23: sample holder, 24: substrate sample and Pt raw material salt aqueous solution, 25: water, 26: Graphite substrate, 27: Fe nano-anchor, 28: Drop of Pt raw salt aqueous solution, 29: Graphene, 100: Electrocatalyst

Claims

A carbon support, nanopits and / or nanochannels, transition metal fine particles, and a noble metal layer,
The nanopits and / or nanochannels are formed on the carbon support;
The transition metal fine particles are in contact with the carbon support in the nanopits and / or nanochannels,
The noble metal layer is formed on the transition metal fine particles,
Fuel cell electrode catalyst.
The catalyst according to claim 1, wherein the transition metal fine particles contain at least one metal selected from Fe, Ni, Co, Mn, Cr, Mo, V, Ta, and W.
The catalyst according to claim 1, wherein the transition metal fine particles are austenitic stainless steel containing Fe, Ni, and Cr.
The catalyst according to any one of claims 1 to 3, wherein the noble metal layer contains Pt.
The catalyst according to any one of claims 1 to 4, wherein the transition metal fine particles have an average particle diameter of 0.5 to 10 nm, and the noble metal layer has a thickness of 0.5 to 2 nm.
Ratio of average particle diameter of transition metal fine particles and average diameter of nanopits (average particle diameter of transition metal fine particles / average diameter of nanopits), or ratio of average particle diameter of transition metal fine particles to average width of nanochar (transition metal) The catalyst according to any one of claims 1 to 5, wherein a fine particle average particle diameter / nanochannel average width) is 0.3 to 1.4.
The carbon support is graphite or graphitized carbon black;
The catalyst according to any one of claims 1 to 6, wherein the nanopits and / or nanochannels are formed on a basal plane of the graphite or graphitized carbon black.
Including a supporting step, a heat treatment step, and a noble metal layer forming step,
In the supporting step, the transition metal fine particles are supported on the carbon support,
In the heat treatment step, the carbon support carrying the transition metal fine particles is heated,
In the noble metal layer forming step, a noble metal layer is formed on the transition metal fine particles.
A method for producing an electrode catalyst for a fuel cell.
The manufacturing method according to claim 8, wherein the heat treatment step is performed at 500 to 1140 ° C under an inert gas flow or under vacuum exhaust.
The manufacturing method according to claim 8, wherein the heat treatment step is performed at 300 to 950 ° C under a gas flow including at least one selected from oxygen, hydrogen, water vapor, and carbon dioxide.
The production method according to any one of claims 8 to 10, further comprising an acid treatment step after the heat treatment step.
The manufacturing method according to claim 11, wherein the transition metal fine particles include a noble metal.