WO2009118991A1

WO2009118991A1 - Fuel cell separator

Info

Publication number: WO2009118991A1
Application number: PCT/JP2009/000616
Authority: WO
Inventors: Tomokatsu Himeno; Atsushi Miyazawa; Motoki Yaginuma; Keisuke Yamamoto; Yasue Tanaka
Original assignee: Nissan Motor Co., Ltd.
Priority date: 2008-03-26
Filing date: 2009-02-17
Publication date: 2009-10-01
Also published as: JP2009238497A

Abstract

A fuel cell separator includes a metal substrate that has a gas passage through which a reactant gas flows; and a carbon layer that is formed on a surface of the metal substrate and contains a number of metal particles and electrical conductive carbon. Therefore, it is possible that an oxide of the metal particles is easily formed before acid water reaches the metal substrate so as to prevent the oxide film from being formed on the surface of the metal substrate.

Description

FUEL CELL SEPARATOR

The present invention relates to a fuel cell separator having a good resistance to corrosion while keeping high electrical conductivity.

A polymer electrolyte fuel cell is configured by stacking each cell (single cell) in layers. The single cell includes a solid polymer electrolyte membrane, electrode catalyst layers provided on both sides of the polymer electrolyte membrane, and gas diffusion layers provided on further both sides of the electrode catalyst layers and dispersing supply gas. Moreover, the single cell includes separators on the outer sides of the gas diffusion layers.

In the single cell, each of fuel gas (such as hydrogen gas) and oxidant gas (such as air, oxygen) is supplied to a positive electrode and a negative electrode, respectively. Then, electricity is generated by bringing about the following electrochemical reactions in each of the positive electrode and the negative electrode.

The separator used under such a condition should have both good electrical conductivity and high corrosion resistance. Conventionally, cut graphite and a carbon material such as a mixture of carbon and resin have been mainly used. However, the separator produced by the carbon material has difficulty reducing the thickness because of inferiority in mechanical strength. Thus, it is necessary to make the separator relatively thick. Therefore, the use of the separator made of carbon results in enlargement of the laminated body of the fuel cell.

As a result, the use of a separator made of metal has been considered instead of the carbon material. Also, the enhancement of a corrosion resistance to metal materials for the metal separator and a surface treatment without using high-priced metal has been also advanced. While, a metal separator of which a metal substrate is covered with a carbon layer having good electrical conductivity and a high corrosion resistance has been proposed.

For instance, a fuel cell separator, in which a film made of electrically conductive graphite is formed on a surface of a metal substrate, is disclosed (refer to Patent Citation 1). Also, a fuel cell separator, in which a certain separator substrate is coated with an electrical conductive hard carbon film having a good corrosion resistance, is disclosed (refer to Patent Citation 2). Further, a method for manufacturing a fuel cell separator in which a carbon film is formed on a surface of a separator substrate by dry film-forming method, while applying a negative high voltage to the separator substrate (refer to Patent Citation 3).
Japanese Patent Unexamined Publication No. 2004-235091 PCT International Publication WO 01/006585 Japanese Patent Unexamined Publication No. 2006-286457

However, a defect such as cracks and pinholes may be caused in the carbon layer formed on the metal substrate because of load when stacking cells. Moreover, the above-mentioned defect may be caused at press-forming for forming a gas passage and the like. In such a case, the corrosion resistance of the carbon layer itself is maintained, whereas oxidation corrosion on the surface of the metal substrate is caused due to acid water penetration. Thus, corrosion of the metal substrate caused a problem that a contact resistance of the separator was increased.

Even if the carbon layer itself has the high corrosion resistance, metal such as stainless steel and aluminum used as a metal substrate is easily corroded. Therefore, the obtained separator shows a low contact resistance in the early stage, while an oxide film is gradually formed on the metal substrate. Thus, it results in the increase of the contact resistance.

The present invention has been made focusing on the above-mentioned conventional problems. An object of the present invention is to provide a fuel cell separator having high electrical conductivity, corrosion resistance and adhesion.

According to one aspect of the present invention, there is provided a fuel cell separator including: a metal substrate that has a gas passage through which a reactant gas flows; and a carbon layer that is formed on a surface of the metal substrate and contains a number of metal particles and electrical conductive carbon.

Fig. 1 is a schematic view showing a structure of a fuel cell including a separator according to an embodiment of the present invention. Fig. 2 is a schematic view showing a cross section of the structure of the separator in Fig. 1. Fig. 3 is a graph showing a relationship between a contact resistance of a carbon layer and I_D/I_G. Fig. 4 is a schematic view showing a cross section of a structure of a separator according to another embodiment of the present invention. Fig. 5 is a schematic view showing a constitution at a measurement of a contact resistance. Fig. 6 is a graph showing a contact resistance value of a separator in Examples 1, 2 and Comparative Example 1 before immersion in acid water. Fig. 7 is a graph showing a contact resistance value of the separator in Examples 1, 2 and Comparative Example 1 after immersion in acid water. Fig. 8 is a graph showing a contact resistance value of a separator in Examples 3 to 5 and Comparative Example 2 before immersion in acid water. Fig. 9 is a graph showing a contact resistance value of the separator in Examples 3 to 5 and Comparative Example 2 after immersion in acid water.

A description will be made below in detail of embodiments of the present invention with reference to the figures.

(FUEL CELL SEPARATOR)
Fig. 1 is a schematic view showing a basic structure of a polymer electrolyte fuel cell, which is one of the fuel cells provided with a separator of the embodiment of the present invention. Catalyst layers 3 (anode catalyst layer and cathode catalyst layer) are provided on both surfaces of a solid polymer electrolyte membrane 2. Also, gas diffusion layers 4 (anode-side gas diffusion layer and cathode-side gas diffusion layer) are provided, which hold the solid polymer electrolyte membrane 2 and the catalyst layers 3 from both sides so as to form a MEA (membrane electrode assembly) 10. The MEA 10 is eventually held between a pair of separators 5 having electrical conductivity so as to compose a single cell of the polymer electrolyte fuel cell. Basically, gas seals are interposed between the separators 5 and the electrolyte membrane 2, and between a fuel cell 1 and another fuel cell adjacent to the fuel cell 1. However, the gas seals are omitted in Fig. 1.

In the separator 5, reactant gas passages 5a and coolant passages 5b are formed by pressing 0.5 mm or less of a thin plate. The separator 5 includes the gas passages 5a and manifolds through which the reactant gas (fuel gas and oxidant gas) flows, and the coolant passages 5b through which the coolant flows. Moreover, the separator 5 has a function to electrically connect each MEA 10 in series. The separator 5 also has a function to maintain mechanical strength of stacks.

Fig. 2 is a schematic view showing a cross section of a surface structure of the separator 5 in Fig. 1. The separator 5 includes a metal substrate 6 and a carbon layer 9 including metal particles 7 and an electrical conductive carbon 8. The carbon layer 9 is formed on the metal substrate 6. By including the electrical conductive carbon 8 and the metal particles 7 relatively oxidized easily in the carbon layer 9 together, the metal particles 7 are oxidized prior to a metal element composing the metal substrate 6. Thus, it is possible to prevent an oxide film from being formed on a surface 6a of the metal substrate 6 so as to suppress an increase of a contact resistance of the separator. In addition, since the metal particles 7 and the electrical conductive carbon 8 are present in mixed condition in the carbon layer 9, internal stress of the carbon layer 9 itself is reduced. As a result, high adhesion between the carbon layer 9 and the metal substrate 6 as a base can be obtained. Therefore, a resistance value of the fuel cell is lowered, and long-term stability can be ensured. The following are details of each component of the separator 5.

The material composing the metal substrate 6 is preferably selected based on the relationship with the metal particles 7 discussed below. Specifically, the material of the metal substrate 6 is preferably composed of a metal element equivalent to the metal particles 7, or composed of a metal element resistant to oxidation compared to the metal particles 7 in view of oxidizing tendency. More preferably, the material is composed of at least one of the metal element equivalent to the metal particles 7 and the metal element more noble than the metal particles 7.

As examples of the material composing the metal substrate 6, aluminum (Al) or an aluminum alloy, stainless steel, and titanium or magnesium, and the like are included. Especially, stainless steel, aluminum or the aluminum alloy, and titanium are preferable. Further, aluminum or the aluminum alloy, and stainless steel are more preferable.

Preferably, aluminum purity is 97 % or more, more preferably 99 % or more. Preferably, an element in the aluminum alloy other than aluminum is composed of at least one of the metal element equivalent to the metal particles 7 and the metal element more noble than the metal particles 7 as described above. As examples of the element other than aluminum, copper, manganese, silicon, magnesium, zinc, nickel, and the like are included.

As the aforementioned aluminum alloy, pure aluminum series, aluminum-manganese series, aluminum-magnesium series, and the like are included. As pure aluminum series, A1050P standardized according to Japanese Industrial Standards is included. As aluminum-manganese series, A3003P and A3004P are included. As aluminum-magnesium series, A5052P and A5083P are included. Especially, it is more preferable to use the aluminum alloy of pure aluminum series of A1050P.

Preferably, a content ratio of iron (Fe) in stainless steel is 60 to 84 % by mass, more preferably 65 to 72 % by mass. In addition, preferably, a content ratio of chromium (Cr) in the stainless steel is 16 to 20 % by mass, more preferably 16 to 18 % by mass.

As the aforementioned stainless steel, austenite series, martensite series, ferrite series, austenite-ferrite series, precipitation hardening series, and the like are included. As austenite series, SUS201, SUS202, SUS301, SUS302, SUS303, SUS304, SUS305, SUS316 and SUS317 standardized according to Japanese Industrial Standards are included. As austenite-ferrite series, SUS329J1 is included. As martensite series, SUS403 and SUS420 are included. As ferrite series, SUS405, SUS430 and SUS430LX are included. As precipitation hardening series, SUS630 is included. Especially, it is more preferable to use stainless steel of austenite series of SUS304, SUS316, and the like.

Preferably, a thickness of the metal substrate 6 is 50 micrometers to 500 micrometers, more preferably 80 micrometers to 200 micrometers in view of workability and the mechanical strength.

Next, the element composing the metal particles 7 is preferably selected based on the relationship with the material (metal element) composing the metal substrate 6 as discussed above. Specifically, the metal particles 7 are preferably composed of the metal element equivalent to the metal substrate 6, or composed of the metal element easier to be oxidized than the metal substrate 6 in view of oxidizing tendency. More preferably, the metal particles 7 are composed of at least one of the metal element equivalent to the metal substrate 6 and the element less noble than the metal substrate 6. In such a case, the metal particles 7 are oxidized prior to the material composing the metal substrate 6. Therefore, it is possible to prevent the oxide film from being formed on the surface 6a of the metal substrate 6 so as to suppress the increase of the contact resistance of the separator.

Further, with regard to combinations of the element composing the metal particles 7 and the element composing the metal substrate 6, aluminum and/or titanium for the metal particles 7 and stainless steel for the metal substrate 6 are preferable in view of ionization tendency. Or preferably, the metal particles 7 are composed of aluminum and/or magnesium and the metal substrate 6 is composed of aluminum or the aluminum alloy. In such a combination, it is possible to suppress the increase of the contact resistance more remarkably.

The following is a specific explanation of the above-mentioned two combinations. First, an explanation of the former combination will be made. In view of load in the fuel cell stack, and the mechanical strength and cost at press-forming, the above-mentioned stainless steel can be used for the material of the metal substrate 6 in the separator. Also, as a material less noble than stainless steel and forming a stable oxide, aluminum (Al) and/or titanium (Ti) can be used. By using aluminum (Al) and/or titanium (Ti) to form the metal particles 7, the metal particles 7 are oxidized prior to the surface 6a of the metal substrate 6. Therefore, it is possible to suppress the increase of the contact resistance, keeping the low contact resistance in the early stage so as to improve the corrosion resistance.

Next, an explanation of the latter combination will be made. When aluminum (Al) or aluminum alloy is used for the material of the metal substrate 6 as described above, aluminum (Al) and/or magnesium (Mg), which is equivalent to or less noble than the metal substrate 6 to form the stable oxide, can be used for the metal particles 7. By using aluminum (Al) and/or magnesium (Mg) to form the metal particles 7, the metal particles 7 are oxidized prior to the surface 6a of the metal substrate 6. Especially, since aluminum (Al) or the aluminum alloy is convenient because of lightness in weight and flexibility compared to other metals, and has good electrical conductivity, it is a great advantage of using aluminum (Al) or the aluminum alloy for the metal substrate 6. Therefore, it is possible to suppress the increase of the contact resistance, keeping the low contact resistance in the early stage so as to improve the corrosion resistance.

Thus, the metal particles 7 are preferably dispersed into the carbon layer 9 as less noble metal, that is, lower standard electrode potential metal, than the metal substrate 6. In such a case, the metal particles 7 are oxidized in the carbon layer 9 in advance of the oxide film formed on the surface of the metal substrate 6. Whereas, it is not preferable that the metal particles 7 are composed of metal more noble than the metal substrate 6 because noble metal promotes oxidation of metal composing the metal substrate 6. While, by dispersing and mixing the metal particles 7 easy to be oxidized into the carbon layer 9 in advance, it enables the metal particles 7 in carbon to easily form the oxide before oxygen reaches the metal substrate 6. Thus, it is possible to suppress the increase of the total resistance from the metal substrate 6 to the surface of the carbon layer 9. In addition, the metal particles 7 have an effect of lowering the internal stress of the carbon layer 9. Therefore, it is possible to improve adhesion between the carbon layer 9 and the metal substrate 6 as a base and suppress the contact resistance of the separator remarkably.

With regard to another advantage of using the metal particles 7, the internal stress of the carbon layer 9 itself is lowered when the metal particles 7 and the electrical conductive carbon 8 are present in the carbon layer 9 in a mixed state. Therefore, it is possible to obtain high adhesion between the metal substrate 6 as a base and the carbon layer 9, and also ensure the reduction of the resistance value of the fuel cell and long-term stability.

Preferably, an average particle diameter of the meal particles 7 is smaller than a thickness of the carbon layer 9 as described below. Specifically, it is 0.2 nm to 20 nm, more preferably 0.5 nm to 10 nm, further preferably 0.5 nm to 5 nm. When the average particle diameter is 0.2 nm or more, sacrificial oxidation of the metal particles 7 is easily caused and the generation of the oxide film on the surface of the metal substrate 6 is suppressed more efficiently so as to improve the durability of the fuel cell. While, when the average particle diameter is 20 nm or less, the possibility that the oxide is formed not only outside but also inside of the metal particles 7 increases when the metal particles 7 are oxidized. Therefore, it is possible to fully take advantage of the metal particles 7 with high efficiency. Note that, the "particle diameter" in the description of the present invention represents the maximum length of lines connecting two certain points on the circumference of the particle. In addition, with respect to the "average particle diameter", the value calculated as an average value of the particle diameters observed in several views to several tens of views is adopted, by use of an observation method such as a scanning electron microscope (SEM) and transmission electron microscope (TEM).

Preferably, a content ratio of the metal particles 7 in the carbon layer 9 is 1 to 50 % by mass, more preferably 5 to 20 % by mass. When the content ratio of the metal particles 7 is 1 % by mass or more, the sacrificial oxidation of the metal particles 7 is caused prior to oxidation at the interface between the carbon layer 9 and the metal substrate 6. Therefore, it is possible to efficiently suppress the increase of the contact resistance of the separator. While, the content ratio of the metal particles 7 is 50 % by mass or less, it is possible to form the oxide of the metal particles 7 in the carbon layer 9 without disturbing electrical conductivity of the carbon layer 9 itself.

The carbon material usable for the electrical conductive carbon 8 is not particularly limited as long as the contact resistance of the separator is not increased. When an intensity ratio (R value) of the carbon layer 9 measured by Raman spectroscopy is within a predetermined range, it is possible to suppress the increase of the above-mentioned contact resistance remarkably. Thus, the carbon material used as the electrical conductive carbon 8 is preferably selected so that the carbon layer 9 has the intensity ratio within the predetermined range.

Raman spectroscopy is to analyze a structure of a sample by analyzing Raman spectrum. When the carbon material is analyzed by Raman spectroscopy, peaks usually appear in the vicinity of 1350 cm^-1 and 1584 cm^-1. High crystallinity graphite has a single peak in the vicinity of 1584 cm^-1, which is usually described as G-band. While, as the crystallinity is lowered, in the other words, as the defect of the crystal structure is increased and the structure of graphite is disordered, a peak in the vicinity of 1350 cm^-1 appears, which is usually described as D-band. Precisely, the peak of diamond is 1333 cm^-1, which is distinct from D-band. The intensity ratio R (I_D/I_G) of D-band and G-band represents a size of a graphite cluster of the carbon material, a disordered level of the graphite structure (defect of the crystal structure), and an sp² bond ratio. That is, the ratio can be regarded as an index of the contact resistance of the separator, and used as a film parameter controlling electrical conductivity of the carbon layer 9.

The R value is calculated by measuring Raman spectrum of the carbon material by use of Micro-Raman spectroscopy. Specifically, the value is obtained by calculating the relative intensity ratio, that is, a peak area ratio (I_D/I_G), between the peak intensity (I_D) of 1300 to 1400 cm^-1 described as D-band and the peak intensity (I_G) of 1500 to 1600 cm^-1 described as G-band. Each I_D and I_G may be described as the D-band peak intensity and the G-band peak intensity, respectively. By increasing the R value, in other words, by increasing the ratio of D-band, the graphite structure is disordered and the sp² bond ratio of the carbon atoms is increased so as to suppress the contact resistance more remarkably. The value obtained by the following Raman spectrum measurement can be adopted as the peak area.

The following is a specific explanation of the relationship between the R value and the contact resistance. First, separators including carbon layers with various R values (I_D/I_G) were prepared. In this case, solid graphite is used as a raw material of the electrical conductive carbon 8. As described later, when the carbon layer is formed by a sputtering method, it is possible to obtain the separators with the various R values (I_D/I_G) by changing acceleration voltage (V). The R values were obtained by measuring Raman spectrum and calculating the peak area ratios of D-band and G-band. The measurements of Raman spectrum were performed five times at room temperature with 30 minutes exposure by use of Holo Lab 5000R made by Kaiser Optical System Inc. as a measuring device under the following conditions.

Excitation wavelength: second-harmonic generation of a Nd:YAG laser, 532 nm
Laser output: 3 mW
Spot size: up to 1 micrometer
Detector: Charge-coupled device detector

Next, the contact resistance of the separators including carbon layers with the various R values (I_D/I_G) was measured. With regard to the measuring method of the contact resistance, the layer of gas diffusion layer-separator-gas diffusion layer is held between a pair of copper electrode bars facing each other, the voltage value when applying certain current is read, and the penetration resistance is calculated. In this case, since there are two contact regions between the separator and the gas diffusion layers, the half of the measured value is considered as the contact resistance. The resistance includes a bulk resistance of the gas diffusion layers and the metal substrate in the separator. However, the value of the bulk resistance is so low that it can be ignored in this case.

Fig. 3 shows the relationship between the R value and the contact resistance. A vertical axis represents a ratio of the relative contact resistance value with respect to the lowest contact resistance value. According to Fig. 3, when the intensity ratio (R value) of the carbon layer 9 is within the predetermined range, the sp² bond ratio of the carbon atoms in the carbon layer 9 increases, so that electrical conductivity in the carbon layer 9 is improved and the contact resistance is further lowered. Preferably, the above-mentioned predetermined range of the R value (I_D/I_G) of the carbon layer 9 is 1.4 or more, more preferably 1.4 to 2.0. When 1.4 or more, as shown in Fig. 3, the carbon layer 9 with high electrical conductivity can be obtained. While, when 2.0 or less, it is possible to suppress the decrease of the graphite content and also suppress the increase of the internal stress of the carbon layer 9 itself so as to further improve adhesion to the metal substrate 6 as a base.

As the electrical conductive carbon 8, carbon black, graphite, fullerene, carbon nanotube, carbon nanofiber, carbon nanohorn, carbon fibril, and the like are included. As the specific examples of carbon black, Ketjen black, acetylene black, channel black, lamp black, oil furnace black or thermal black, and the like are included. A graphite treatment may be performed on carbon black. In view of the decrement of the contact resistance between the carbon layer 9 and the metal substrate 6 due to the structure and configuration of the electrical conductive carbon 8, graphite or carbon nanotube is preferable. In addition, the carbon material may be used individually, or in combination with two kinds or more.

An average particle diameter when the material of the electrical conductive carbon 8 is in a particle state is not particularly limited, but preferably 1 nm to 100 nm, more preferably 5 nm to 20 nm in view of reducing the thickness of the carbon layer 9.

A diameter when the material of the electrical conductive carbon 8 is in a fibrous state such as carbon nanotube is not particularly limited, but preferably 0.4 nm to 100 nm, more preferably 1 nm to 20 nm. A length when in the above-mentioned fibrous state is not particularly limited, but preferably 5 nm to 200 nm, more preferably 10 nm to 100 nm. An aspect ratio when in a fibrous state is not particularly limited, but preferably 1 to 500, more preferably 2 to 100. When within the above-mentioned range, it is possible to optimally reduce the thickness of the carbon layer 9. Preferably, the content ratio of the electrical conductive carbon 8 in the carbon layer 9 is 50 to 99 % by mass, more preferably 80 to 98 % by mass, most preferably 80 to 95 % by mass. When within the above-mentioned range, it is possible that the separator to be obtained has high electrical conductivity, corrosion resistance and adhesion altogether keeping a good balance each other.

Next, the following is a description of the carbon layer 9. The preferable range of the intensity ratio in the carbon layer 9 measured by Raman spectroscopy is as described above.

The carbon layer 9 may be formed at least on one surface of the metal substrate 6. However, the carbon layers are preferably formed on both surfaces of the metal substrate 6 so as to further achieve the desired effect of the present invention.

Preferably, a film thickness of the carbon layer 9 is 0.005 micrometers to 1 micrometer, more preferably 0.01 micrometers to 0.2 micrometers. When within the above-mentioned range, it is possible to maintain the low contact resistance in the early stage, and also suppress the increase of the contact resistance so as to improve the corrosion resistance.

More specifically, when the film thickness of the carbon layer 9 is 0.005 micrometers or more, a surface coverage ratio of the carbon layer 9 with respect to the metal substrate 6 is increased. Therefore, it is possible to form the carbon layer 9 as an even and continuous layer and suppress roughness on the surface of the metal substrate 6. In addition, it is possible to prevent the metal particles 7 in the carbon layer 9 from coming up to the outer surface of the carbon layer 9 and forming the oxide film on the surface of the metal substrate 6. Therefore, an electrical conduction path in the carbon layer 9 as a whole is increased, and it results in the further decrease of the contact resistance.

Also, when the film thickness of the carbon layer 9 is 1 micrometer or less, it is possible to prevent from developing the film stress (internal stress) of the carbon layer 9, and possible to improve adhesion to the metal substrate 6 remarkably. Moreover, it is capable of lowering the initial resistance itself in the thickness direction. It is also capable of preventing from causing cracks and the like without developing the internal stress of the carbon layer 9 caused by the reduced film thickness. Thus, it results in the promotion of the contact resistance reduction.

Although the carbon layer 9 is mainly formed of the metal particles 7 and the electrical conductive carbon 8 as described above, the carbon layer 9 may include other materials. For instance, a water-repellent material such as polytetrafluoroethylene (PTFE) is included as the other material. Also, as the later described Examples, the carbon layer may be formed of only the electrical conductive carbon and the metal particles. That is, the carbon layer 9 can comprise, consist essentially of, or consist of the metal particles 7 and the electrical conductive carbon 8.

Preferably, the carbon layer 9 is formed at least on a region contacting with the gas diffusion layer. More preferably, the carbon layer 9 is formed in the whole of the metal substrate 6 in order to maintain the high corrosion resistance.

Moreover, as shown in Fig. 4, a middle layer 11 composed of more than one selected from the group consisting of metal, metal nitride and metal carbide may be provided between the metal substrate 6 and the carbon layer 9. Due to the middle layer 11 provided between the metal substrate 6 and the carbon layer 9, adhesion of the carbon layer 9 can be improved. When the metal substrate 6 is directly coated with the carbon layer 9, optimum adhesion may not be obtained. Specifically, when the R value exceeds the upper limit of the above-mentioned preferable range, the metal substrate 6 as a base and the carbon layer 9 may not optimally adhere to each other. Therefore, by inserting the middle layer 11 in such a case, preferable adhesion between the metal substrate 6 and the carbon layer 9 can be ensured.

A material composing the middle layer 11 is not particularly limited as long as the material can provide for adhesion as mentioned above. For example, little ion elution metal such as chromium (Cr), tungsten (W), titanium (Ti), molybdenum (Mo), niobium (Nb) and hafnium (Hf), metal nitride and metal carbide of these, metal carbonitride of these are included. In particular, little ion elution metal such as chromium (Cr) and titanium (Ti), and metal nitride and metal carbide of these are preferable. Especially, when little ion elution metal, and metal nitride and metal carbide of these are used, the corrosion resistance of the separator 15 can be improved.

A thickness of the middle layer 11 is not particularly limited, but preferably 5 nm to 1000 nm, more preferably 10 nm to 200 nm. Then, it is possible to reduce the size of the laminated body of the fuel cell due to the reduced thickness of the separator 15.

The middle layer 11 is provided between the metal substrate 6 and the carbon layer 9. Therefore, the material (metal) composing the middle layer 11 is preferably oxidized prior to oxidation of the surface of the metal substrate 6 similar to the metal particles 7 in the carbon layer 9. Thus, the material composing the middle layer 11 is preferably composed of at least one of the metal element equivalent to the metal substrate 6 and the metal element less noble than the metal substrate 6, and also composed of at least one of the metal element equivalent to the metal particles 7 and the metal element more noble than the metal particles 7.

The following is an explanation of a preferable combination of the element composing the metal particles 7, the element composing the middle layer 11 and the element composing the metal substrate 6. Preferably, the metal particles 7 are composed of aluminum (Al) and/or titanium (Ti), the middle layer 11 is composed of chromium (Cr), titanium (Ti) and aluminum (Al), and the metal substrate 6 is composed of stainless steel. Alternatively, the metal particles 7 are preferably composed of aluminum (Al) and/or magnesium (Mg), the middle layer 11 is preferably composed of chromium (Cr), titanium (Ti), aluminum (Al) and magnesium (Mg), and the metal substrate 6 is preferably composed of aluminum (Al) or the aluminum alloy.

The middle layer 11 may be formed at least on one surface of the metal substrate 6. However, the middle layers are preferably formed on both surfaces of the metal substrate 6 so as to further achieve the desired effect of the present invention. Note that, when the middle layers 11 are provided on both surfaces of the metal substrate 6, the carbon layers 9 are also provided on each middle layer 11.

The carbon layer 9 can be formed on the metal substrate 6 targeting a target material formed of the element composing the metal particles 7 and a target material formed of the electrical conductive carbon 8. In this case, a physical vapor deposition (PVD) method such as a sputtering method or an arc ion plating method, or an ion beam vapor deposition method such as a filtered cathodic vacuum arc (FCVA) method is preferably used. As the sputtering method, a magnetron sputtering method, an unbalanced magnetron sputtering (UBMS) method, a dual magnetron sputtering method and the like are included. Due to such a method for forming the carbon layer 9, the carbon layer 9 with a small amount of hydrogen can be formed. Therefore, it is possible to increase the sp² bond ratio between the carbon atoms so as to obtain good electrical conductivity.

(MANUFACTURING METHOD OF FUEL CELL SEPARATOR)
The following is an explanation of a manufacturing method of the fuel cell separator of the present embodiment. The conditions for the materials and the like of each component of the fuel cell separator will be omitted here since the description is the same as described above. The following are examples of the treatment process of the above-mentioned manufacturing method.

The first example will be described below. First, the metal substrate 6 preliminarily provided with the reactant gas passages 5a, the manifolds, and coolant passages 5b through which the coolant flows is prepared by a well-known method. Next, a treatment to degrease and wash on the surface of the metal substrate 6 is performed by use of a proper solvent. As the solvent, ethanol, ether, acetone, isopropyl alcohol, trichloroethylene, caustic alkaline agent and the like can be used. As the degreasing and washing treatment, ultrasonic cleaning and the like are included. With regard to conditions of ultrasonic cleaning, the treatment time is approximately for 1 to 10 minutes, the frequency is approximately 30 to 50 kHz, and electricity is approximately 30 to 50 W.

Then, the oxide film formed on the (both) surfaces of the metal substrate 6 is removed. Regarding the method for removing the oxide film, an acid treatment, a dissolution treatment with applying an electric potential, an ion bombard treatment or the like may be included.

The carbon layer 9 is subsequently formed by use of the sputtering method and the like as described above. For example, the target material composed of the electrical conductive carbon such as solid graphite, and the target material composed of the element composing the metal particles such as aluminum (Al) are used for targeting. Then, the carbon layer 9 having the desired thickness formed by dispersing the aluminum particles on both surfaces of the metal substrate 6 is formed while applying a negative bias voltage to the substrate at preferably 50 to 300 V. Thus, the fuel cell separator of the present invention can be obtained.

One of the characteristics of the fuel cell separator of the present invention is the relationship between the carbon layer 9 and the metal substrate 6 of the components. After removing the oxide film and impurities from the surface of the metal substrate 6 in advance as described above, the carbon layer 9 is formed by depositing the electrical conductive carbon 8 and the metal particles 7 on the metal substrate 6 at the atomic level in vacuum. Thus, the interface between the carbon layer 9 and the metal substrate 6 directly adhered each other and the periphery of the interface can exert adhesion due to an intermolecular force and the entry of a small amount of carbon atoms, and thus adhesion can be maintained for long periods.

When the metal substrate 6 is directly coated with the carbon layer 9 with extremely high hardness compared to the metal substrate 6, the internal stress may be caused in the carbon layer 9, and exfoliation between the carbon layer 9 and the metal substrate 6 may be caused easily. Therefore, the carbon layer 9 with relatively similar hardness to the metal element composing the metal substrate 6 is preferably used for coating.

Next, as the second example, the treatments to degrease, wash, and remove the oxide film on the surface of the metal substrate 6 are performed similarly to the first example. Then, little ion elution metal such as chromium (Cr), tungsten (W), titanium (Ti), molybdenum (Mo), niobium (Nb) and hafnium (Hf), and metal nitride and metal carbide of these are targeted. The middle layer 11 is formed on the surface of the metal substrate 6 using the targets by use of the magnetron sputtering method, the unbalanced magnetron sputtering (UBMS) method and the like. Furthermore, the carbon layer 9 is formed on the middle layer 11 similarly to the first example. Thus, the fuel cell separator of the present invention provided with the middle layer can be obtained.

As the third example, first, the metal substrate that the reactant gas passages, the manifolds, coolant passages and the like are not preliminarily provided is prepared. Next, the treatments to degrease, wash, and remove the oxide film on the surface of the metal substrate 6 are performed similarly to the first example. Then, the carbon layer 9 is formed similarly to the first example, followed by forming the reactant gas passages, the manifolds, coolant passages and the like by means of a well-known press forming. Since the carbon layer 9 of the present invention has high adhesion to the metal substrate 6, the carbon layer 9 is not exfoliated even when the reactant gas passages and the like are formed after the carbon layer 9.

In addition, as the fourth example, first, the metal substrate that the reactant gas passages, the manifolds, coolant passages and the like are not preliminarily provided is prepared. the treatments to degrease, wash, and remove the oxide film on the surface of the metal substrate 6 are performed similarly to the first example. Then, the middle layer 11 and the carbon layer 9 are formed similarly to the second example, followed by forming the reactant gas passages, the manifolds, coolant passages and the like by means of a well-known press forming.

(FUEL CELL)
The following is an explanation of the fuel cell using the above-mentioned fuel cell separator. The fuel cell separator of the present invention can be used for various kinds of fuel cells such as a polymer electrolyte fuel cell (PEFC), a phosphoric acid fuel cell (PAFC), a molten carbonate fuel cell (MCFC), a solid oxide fuel cell (SOFC) or an alkaline fuel cell (AFC). The polymer electrolyte fuel cell (PEFC), which is optimally used for the after-mentioned vehicle, is preferable.

The following is a specific explanation of components of the fuel cell except the fuel cell separator. Note that, the present invention is characterized by the separator, while well-known components can be used for composing the fuel cell except for the separator.

(ELECTROLYTE MEMBRANE)
As an electrolyte membrane, well-known electrolyte membranes can be used. However, the polymer electrolyte membrane is preferable. The polymer electrolyte membrane is broadly divided into a fluorine polymer electrolyte membrane and a hydrocarbon polymer electrolyte membrane depending on the type of ion-exchange resin as a constituent material. With regard to the fluorine polymer electrolyte membrane, a perfluorocarbon sulfonic acid polymer such as Nafion (registered trademark, made by DuPont Corporation), Aciplex (registered trademark, made by Asahi Kasei Corporation), and Flemion (registered trademark, made by Asahi Glass Co., Ltd.), a perfluorocarbon phosphonic acid polymer, a trifluorostyrene sulfonic acid polymer, an ethylene tetrafluoroethylene-g-styrene sulfonic acid polymer, an ethylene-tetrafluoroethylene copolymer, a polyvinylidene fluoride-perfluorocarbon sulfonic acid polymer, and the like are included. These fluorine polymer electrolyte membranes are preferably used in view of electric generation performance such as heat resistance and chemical stability. Especially, the fluorine polymer electrolyte membrane composed of the perfluorocarbon sulfonic acid polymer is preferably used.

With regard to the hydrocarbon polymer electrolyte membrane, sulfonated polyether sulfone (S-PES), sulfonated polyaryl ether ketone, polybenzimidazole-alkyl sulfonic acid, polybenzimidazole-alkylphosphonic acid, sulfonated polystyrene, sulfonated polyether ether ketone (S-PEEK), sulfonated polyphenylene (S-PPP) and the like are included. These hydrocarbon polymer electrolyte membranes are preferably used in view of manufacturing reasons such as inexpensiveness of raw materials, ease of manufacture, and high selectivity of materials. The above-mentioned ion-exchange resin may be used individually, or in combination with two kinds or more. Moreover, the materials are not limited to the above-mentioned polymers, and other materials may be used.

A thickness of the polymer electrolyte membrane is not limited, and may be appropriately determined in view of the characteristics of the fuel cell to be obtained, however it is usually about 5 micrometers to 300 micrometers. When the thickness is within the above-mentioned range, the balance of intensity during the formation, durability in use, and output property in use may be controlled appropriately.

(CATALYST LAYERS)
The catalyst layers include a catalyst component, an electrical conductive catalyst carrier supporting the catalyst component, and an electrolyte.

The catalyst component used for the anode catalyst layer is not particularly limited as long as it has a catalytic function for an oxidizing reaction of hydrogen, and a well-known catalyst may be used. The catalyst component used for the cathode catalyst layer is also not particularly limited as long as it has a catalytic function for a reductive reaction of oxygen, and a well-known catalyst may be used. Specifically, the catalyst component may be selected from platinum, ruthenium, iridium, rhodium, palladium, osmium, tungsten, lead, iron, chromium, cobalt, nickel, manganese, vanadium, molybdenum, gallium and aluminum, alloys of these, and the like.

Among them, the component at least containing platinum is preferably used in order to enhance catalytic activity, poisoning resistance to carbon monoxide and the like, heat resistance, and the like. With regard to a composition of the alloy when used as the anode catalyst, it is preferable that platinum be 30 to 90 atom% and the alloyed metal be 10 to 70 atom%, although the composition depends on the type of metal formed into the alloy. With regard to a composition of the alloy when used as the cathode catalyst, it is preferable that platinum be 30 to 90 atom% and the alloyed metal be 10 to 70 atom%, although the composition depends on the type of metal formed into the alloy. Note that, an alloy is generally composed of a metal element, and one or more of different metal elements or nonmetal elements added thereto, and is a general term of elements having a metallic property. With regard to a constitution of the alloy, there are mentioned, for example, an eutectic alloy as a so-called mixture in which component elements become individual crystals, a solid solution in which the component elements are completely blended together, and a compound in which the component elements compose an intermetallic compound or a compound of a metal and nonmetal. The present invention may adopt any of those. In this case, the catalyst components for use in the anode catalyst layer and the catalyst component for use in the cathode catalyst layer can be appropriately selected from the above-mentioned ones. In the following description, unless otherwise specified, the similar definition between the anode catalyst layer and the cathode catalyst layer will be made with regard to the catalyst components, which are referred to as a "catalyst component" in a lump. However, it is not necessary that the catalyst component for the anode catalyst layer and the cathode catalyst layer be the same. The catalyst component may be appropriately selected so as to exert such desired functions as described above.

A shape and size of the catalyst component is not particularly limited, and can be similar to these of a well-known catalyst component. However, the catalyst component is preferably granular. Preferably, the average particle diameter of the catalyst particles is 1 nm to 30 nm in this case. When the average particle diameter of the catalyst particles is within the above-mentioned range, the balance of catalyst utilization efficiency and ease of supporting may be controlled appropriately. Note that, the "average particle diameter of catalyst particles" in the present invention may be measured as a crystallite diameter obtained from a full width at half maximum of a diffraction peak of the catalyst component in an X-ray diffraction, or measured as an average value of the particle diameters of the catalyst component investigated by the transmission electron microscope.

The catalyst carrier functions as a carrier for supporting the above-mentioned catalyst component, and functions as an electrical conduction path involved in communicating electrons with the catalyst component.

The catalyst component has only to have a sufficient surface area for supporting the catalyst component in a desired dispersed state, and have sufficient electron conductivity. Preferably, the catalyst carrier contains carbon as a main component. Specifically, carbon particles composed of carbon black, activated carbon, coke, natural graphite, artificial graphite, and the like are included. Note that, in the present invention, "containing carbon as a main component" denotes that carbon atoms are contained as a main component, and includes concepts that the carbon material is composed only of the carbon atoms, and also that the carbon material is substantially composed of the carbon atoms. Depending on the situation, elements other than the carbon atoms may be contained in order to enhance the characteristics of the fuel cell. Note that, "being substantially composed of the carbon atoms" denotes that inclusion of impurities with approximately 2 to 3 % by mass or less may be permitted.

A BET specific surface area of the catalyst carrier has only to have an area enough to support the catalyst component in a highly dispersed state. Preferably, the area is 20 to 1600 m²/g, more preferably 80 to 1200 m²/g. When the specific surface area of the catalyst carrier is within the above-mentioned range, the balance of dispersibility of the catalyst component and effective utilization efficiency of the catalyst component on the catalyst carrier may be controlled appropriately.

A size of the catalyst carrier is not particularly limited, but preferably 5 nm to 200 nm of an average particle diameter, more preferably 10 nm to 100 nm, in view of ease of supporting, catalyst utilization efficiency, controlling the thickness of the catalyst layer within an appropriate range, and the like.

In the electrode catalyst in which the catalyst component is supported on the catalyst carrier, a supported amount of the catalyst component is preferably 10 to 80 % by mass, more preferably 30 to 70 % by mass, with respect to the total amount of the electrode catalyst. When the supported amount of the catalyst component is within the above-mentioned range, the balance of dispersibility of the catalyst component and catalyst performance in the catalyst carrier may be controlled appropriately. Note that, the supported amount of the catalyst component may be measured by Inductively Coupled Plasma Atomic Emission Spectrometry (ICP-AES).

The catalyst layer includes the ion conductive polymer electrolyte in addition to the electrode catalyst. The polymer electrolyte is not particularly limited, and it is possible to appropriately refer to the conventional knowledge. For instance, the above-mentioned ion-exchange resin composing the polymer electrolyte membrane may be added to the catalyst layer as the polymer electrolyte.

(GAS DIFFUSION LAYERS)
The anode and cathode may have the gas diffusion layers in addition to the above-mentioned catalyst layers. The gas diffusion layer is provided on one side of the catalyst layer opposite to the side that the electrolyte membrane is in contact with.

In the electrode including the catalyst layer and the gas diffusion layer, the catalyst layer may be formed on the gas diffusion layer. Alternatively, the catalyst layer may be formed in a part of the gas diffusion layer. In the latter case, the electrode that the catalyst component and electrolyte are included in a desired region of the gas diffusion layer described later, and the like may be included.

The gas diffusion layer is not particularly limited, and the well-known gas diffusion layer can be used. For instance, a sheet-like gas diffusion layer having electrical conductivity and porousness, such as carbon-made fabric, finished paper, felt, and nonwoven fabric, may be included.

Although a thickness of the gas diffusion layer may be appropriately determined in view of the property of the gas diffusion layer to be obtained, it can be approximately 30 micrometers to 50 micrometers.

The gas diffusion layer may include a water-repellent for the purpose of ensuring higher water repellency so as to prevent from flooding and the like. As the water-repellent, which is not particularly limited, a fluorine polymer material such as polytetrafluoroethylene (PTFE), polyvinylidene fluoride (PVDF), polyhexafluoropropylene, and a tetrafluoroethylene-hexafluoropropylene copolymer (FEP), polypropylene, polyethylene, and the like are included.

In addition, in order to improve water repellency, the gas diffusion layer can include a carbon particle layer composed of a carbon particle assembly containing the water-repellent. The carbon particle layer may be provided on the region in contact with the catalyst layer in the gas diffusion layer.

The carbon particles, which are not particularly limited, can be conventionally common carbon such as carbon black, graphite and exfoliated graphite. Particularly, carbon black such as oil furnace black, channel black, lamp black, thermal black and acetylene black is preferable due to good electrical conductivity and largeness of a specific surface area.

A water-repellent used in the carbon particle layer may be similar to that used in the gas diffusion layer. In particular, the fluorine polymer material is preferably used due to high water repellency, corrosion resistance at the electrode reaction, and the like.

With regard to a mixture ratio between the carbon particles and the water-repellent in the carbon particle layer, it can be said that expected water repellency cannot be obtained if the carbon particles are excessive, or that sufficient electron conductivity cannot be obtained if the amount of the water-repellent is excessive. In view of these points, it is preferable that the mixture ratio between the carbon particles and the water-repellent in the carbon particle layer be approximately 90:10 to 40:60 by mass ratio.

A thickness of the carbon particle layer may be appropriately determined in view of water repellency of the gas diffusion layer to be obtained, but preferably 10 micrometers to 1000 micrometers, more preferably 50 micrometers to 500 micrometers.

A method for manufacturing the fuel cell of the present invention is not particularly limited. It is possible to manufacture the fuel cell by appropriately referring to the conventional knowledge in the field of fuel cells.

A type of a fuel for the fuel cell, which is not particularly limited, may include hydrogen, methanol, ethanol, 1-propanol, 2-propanol, n-butanol, sec-butanol, tert-butanol, dimethyl ether, diethyl ether, ethylene glycol and diethylene glycol. Particularly, hydrogen and methanol are preferable in view of an availability of a high output power.

Moreover, a stack in which a plurality of the membrane electrode assemblies is connected in series via the separators may be formed so that the fuel cell can be obtained a desired voltage and the like. A shape and the like of the fuel cell, which are not particularly limited, can be appropriately determined so as to obtain the fuel cell property such as a desired voltage.

The fuel cell using the fuel cell separator of the present invention can be installed in a vehicle. Although a purpose of the fuel cell of the present invention is not particularly limited, the fuel cell is preferably used as a driving power supply in the vehicle such as an automobile in view of excellent generating performance.

The present invention will be illustrated further in detail by the following Examples and Comparative Examples; however, the scope of the invention is not limited to these Examples.

(EXAMPLE 1)
A separator plate (metal substrate) formed of stainless steel (SUS316L) with a thickness of 100 micrometers was used. An ultrasonic cleaning was performed on the separator plate in ethanol for three minutes as a pretreatment. Then, the metal substrate was placed in a vacuum chamber, and an ion bombard treatment with argon gas was performed so as to remove an oxide film on the surface. Both of the pretreatment and the ion bombard treatment were performed on both sides of the metal substrate.

Next, by the unbalanced magnetron sputtering (UBMS) method, solid graphite and aluminum (Al) were used as a target, and a negative bias voltage with 150 V was applied to the metal substrate. While applying the voltage, carbon layers with a thickness of 0.2 micrometers into which Al particles (10 nm of average particle size) were dispersed were formed on both sides of the metal substrate. An R value (I_D/I_G) of the carbon layers was 1.5. A content ratio of the metal particles in the carbon layers was 10 % by mass, and a content ratio of electrical conductive carbon in the carbon layers was 90 % by mass. Note that, all the Examples and Comparative Examples used the same solid graphite.

(EXAMPLE 2)
The separator plate (metal substrate) formed of stainless steel (SUS316L) with the thickness of 100 micrometers was used, and the pretreatment and the ion bombard treatment were performed similarly to Example 1.

Next, by the UBMS method, chromium (Cr) was used as a target, and chromium films (middle layers) with a thickness of 0.2 micrometers were formed on both sides of the metal substrate. In addition, carbon layers with a thickness of 0.2 micrometers were formed on the middle layers similarly to Example 1. An R value (I_D/I_G) of the carbon layers was 1.5. A content ratio of the metal particles in the carbon layers was 10 % by mass, and a content ratio of electrical conductive carbon in the carbon layers was 90 % by mass.

(EXAMPLE 3)
A separator plate (metal substrate) formed of aluminum alloy (A1050) with a thickness of 200 micrometers was used, and the pretreatment and the ion bombard treatment were performed similarly to Example 1.

Next, by the UBMS method, solid graphite and magnesium (Mg) were used as a target, and a negative bias voltage with 150 V was applied to the metal substrate. While applying the voltage, carbon layers with a thickness of 0.2 micrometers into which Mg particles (15 nm of average particle size) were dispersed were formed on both sides of the metal substrate. An R value (I_D/I_G) of the carbon layers was 1.4. A content ratio of the metal particles in the carbon layers was 10 % by mass, and a content ratio of electrical conductive carbon in the carbon layers was 90 % by mass.

(EXAMPLE 4)
The separator plate (metal substrate) formed of aluminum alloy (A1050) with the thickness of 200 micrometers was used, and the pretreatment and the ion bombard treatment were performed similarly to Example 1.

Next, carbon layers with a thickness of 0.2 micrometers were formed similarly to Example 1. An R value (I_D/I_G) of the carbon layers was 1.5. A content ratio of the metal particles in the carbon layers was 10 % by mass, and a content ratio of electrical conductive carbon in the carbon layers was 90 % by mass.

(EXAMPLE 5)
The separator plate (metal substrate) formed of aluminum alloy (A1050) with the thickness of 200 micrometers was used, and the pretreatment and the ion bombard treatment were performed similarly to Example 1.

Next, chromium films (middle layers) with a thickness of 0.2 micrometers and carbon layers with a thickness of 0.2 micrometers were formed similarly to Example 2. An R value (I_D/I_G) of the carbon layers was 1.5. A content ratio of the metal particles in the carbon layers was 10 % by mass, and a content ratio of electrical conductive carbon in the carbon layers was 90 % by mass.

(COMPARATIVE EXAMPLE 1)
The separator plate (metal substrate) formed of stainless steel (SUS316L) with the thickness of 100 micrometers was used, and the pretreatment and the ion bombard treatment were performed similarly to Example 1.

Next, carbon layers with a thickness of 0.2 micrometers were formed similarly to Example 1 except that only solid graphite was used as a target. An R value (I_D/I_G) of the carbon layers was 1.5.

(COMPARATIVE EXAMPLE 2)
The separator plate (metal substrate) formed of aluminum alloy (A1050) with the thickness of 200 micrometers was used, and the pretreatment and the ion bombard treatment were performed similarly to Example 1.

Next, carbon layers with a thickness of 0.2 micrometers were formed similarly to Example 1 except that only solid graphite was used as a target.

(MEASUREMENT OF CONTACT RESISTANCE)
As shown in Fig. 5, the gas diffusion layers were provided on both sides of the separator formed according to the above-mentioned Examples and Comparative Examples. Furthermore, the gas diffusion layers and the separator were held between the electrodes. Then, the contact resistance value was calculated from the amount of electricity and the voltage value when applying load of 1MPa. The results are shown in Figs. 6 and 8.

(CORROSION TEST)
The separators obtained in the above-mentioned Examples were cut into a size of 30 mm by 30 mm, and immersed in acid water at 80 degrees Celsius for 100 hours keeping pH below 4. Then, the contact resistance values before and after the immersion test were measured. The results are shown in Figs. 5 to 8. The ratios of the contact resistance values of the separators obtained in Examples 1 and 2, when the contact resistance value of the separator obtained in Comparative Example 1 is 1.0, are shown in Figs. 6 and 7. While, the ratios of the contact resistance values of the separators obtained in Examples 3 to 5, when the contact resistance value of the separator obtained in Comparative Example 2 is 1.0, are shown in Figs. 8 and 9.

According to the above-mentioned results, it was demonstrated that the separator of the present invention containing predetermined metal particles in the carbon layer had superior at least corrosion resistance and electrical conductivity to the conventional separators. The results will be explained in detail below.

The conventional separator (for example, Comparative Examples 1 and 2) has the configuration in which the film (carbon layer) including only electric conductive carbon is directly formed on the metal substrate. While, good adhesion between the metal substrate and the electrical conductive carbon layer cannot be obtained, and the contact resistance value in the early stage is high. Furthermore, corrosion at the interface between the metal substrate and the electrical conductive carbon layer is easily developed.

Additionally, the conventional separator may be provided with the middle layer composed of the electrically conductive material such as chromium (Cr) so that adhesion between the metal substrate and the electrical conductive carbon layer is improved. However, while adhesion is improved, oxidation corrosion on the surface of the metal substrate is promoted and the contact resistance is increased because of acid water from cracks and pinholes caused by stacking load.

Consequently, the conventional separator had difficulty preventing the contact resistance from degrading (increasing) as a separator since oxidation corrosion on the surface of the metal substrate was caused by acid water in the fuel cell when the cracks and scratches were caused by stacking load and press forming.

On the other hand, the fuel cell separator of the present invention is characterized by the configuration in which the carbon layer includes predetermined metal particles. Even if the cracks and scratches are caused by stacking load and press forming due to such a configuration, the acid component in the fuel cell forms the oxide by reacting with the predetermined metal particles in the carbon layer before reaching the metal substrate. Thus, the acid component is, namely, "trapped". Therefore, according to the present invention, it is possible to effectively suppress the resistance increase between the metal substrate and the surface of the separator without causing the problem that the acid component reaches the surface of the metal substrate to form the oxide film.

The entire contents of a Japanese Patent Application No. P2008-081367 with a filing date of March 26, 2008 is herein incorporated by reference.

Although the invention has been described above by reference to certain embodiments of the invention, the invention is not limited to the embodiments described above and modifications may become apparent to these skilled in the art, in light of the teachings herein. The scope of the invention is defined with reference to the following claims.

In accordance with the separator of the present invention, the easily-oxidized metal particles are dispersed into the carbon layer in advance. Therefore, it is possible that the oxide of the metal particles is easily formed before acid water reaches the metal substrate so as to prevent the oxide film from being formed on the surface of the metal substrate. Moreover, due to the introduction of the metal particles into the carbon layer, the film stress is reduced and adhesion between the carbon layer and the metal substrate is also improved.

Claims

A fuel cell separator comprising:
a metal substrate that has a gas passage through which a reactant gas flows; and
a carbon layer that is formed on a surface of the metal substrate, and comprises a number of metal particles and electrical conductive carbon.
A fuel cell separator according to claim 1,
wherein the metal particles comprise at least one of an metal element equivalent to the metal substrate and an metal element less noble than the metal substrate.
A fuel cell separator according to claim 1 or 2,
wherein the metal particles comprise at least one of aluminum and titanium, and the metal substrate comprises stainless steel.
A fuel cell separator according to claim 1 or 2,
wherein the metal particles comprise at least one of aluminum and magnesium, and the metal substrate comprises aluminum or an aluminum alloy.
A fuel cell separator according to any one of claims 1 to 4,
wherein an intensity ratio R=I_D/I_G of D-band peak intensity I_D and G-band peak intensity I_G measured by Raman spectroscopy in the carbon layer is 1.4 or more.
A fuel cell separator according to any one of claims 1 to 5,
wherein a thickness of the carbon layer is 0.005 micrometers to 1 micrometer.
A fuel cell separator according to any one of claims 1 to 6, further comprising:
a middle layer that is formed between the metal substrate and the carbon layer, and comprises one or more selected from the group consisting of metal, metal nitride and metal carbide.
A fuel cell separator according to any one of claims 1 to 7,
wherein the carbon layer is formed on the metal substrate by a sputtering method, an arc ion plating method, or a filtered cathodic vacuum arc method targeting both an element composing the metal particles and the electrical conductive carbon.
A fuel cell separator according to any one of claims 1 to 8,
wherein an average particle diameter of the metal particles is 0.2 micrometers to 20 micrometers.
A fuel cell comprising:
a fuel cell separator according to any one of claims 1 to 9.
A vehicle comprising:
a fuel cell according to claim 10.