A Fuel Cell
Technical Field
The present invention relates tc a solid proton exchange membrane (PEM) and a membrane-electrode assembly (MEA) of a proton exchange membrane fuel cell (?EFC) for a hydrogen fuel cell and a direct organic fuel cell in which an organic fuel can be directly supplied without use of a reformer.
Background Art
Patent literature 1: Japanese Patent Publication No. Hei 10-507572
Patent literature 2: Japanese Patent Publication No. 2000-516014
Patent literature 3: Japanese Patent Laid-Open Publication No. 2003-175340
Patent literature 4: Japanese Patent Laid-Open Publication No. 2003-217341
Patent literature 5: Japanese Patent Laid-Open Publication No. 2003-272439
Non-patent literature 1: "Technology of Fuel Cell", edited by Special Committee of Examination of Next-Generation System for Electricity Generation in Fuel Cell (The Institute of Electrical Engineers of Japan), Ohm-sha, August 30, 2002, p.55-98 (in Japanese).
Non-patent literature 2: "New Development of Electrode Catalyst Chemistry", edited by Yoshio Takasu, Akiko Aramata and Yoshio Hori, Book Publisher of Hokkaido University, February 25, 2001, p.207-230, Chapter 9 "New Development of Electrode Catalyst Science" by Masayuki Morita (in Japanese).
Non-patent literature 3: "Journal of America Chemical Society", Vol. 105, No. 3, 1983, p.658-659, Katayama-Aramata, A., and Ohnishi, R.
Non-patent literature 4: International Journal for Numerical Methods in Engineering, 2002, 54, p.1717-1749, "Molecular dynamics and multiscale homogenization analysis of seepage/diffusion problem in bentonite clay", Ichikawa, Y., Kawamura, K., Fujii, K., and Theramast, N.
Non-patent literature 5: Journal of the Clay Science Society of Japsan, Vol. 41, No. 2, 2001, p.43-47, "Microstructural modeling of compacted sodium-bentonite and application of unified molecular dynamics/ homogenization analysis for diffusion process" Suzuki, Satoru et al. (in Japanese).
Non-patent literature 6: Journal of Nuclear Science and Technology, 1992, 29, p873-882, "Effect of dry density on diffusion of some radionuclides in compacted sodium bentonite", Sato, H., Ashida, T., Kohara, Y., Yui, M., and Sasaki, N.
Non-patent literature 7: Radioactive Waste Management and Environmental Remediation, American Society of Mechanical Engineers, 1999, "Activation energies of diffusion of tritium and electrical conduction in water-saturated compacted sodium montmorillonite", Nakazawa, T., Takano, M. , Nobuhara, A., Torikai, Y., Sato, S., and Ohashi, H.
A polymer electrolyte fuel cell or proton exchange fuel cell (PEFC) is expected as a power source for fuel cell automobiles and household systems. The PEFC involves a solid polymer membrane for an electrolyte and has the following characteristics.
(1) The electrolyte is not dissipative ar.d is easy to handle.
(2) It can start up at room temperature and works at temperature less than 100 °C.
(3) It can be small because it has a simple
structure and can develop a high output density.
The working principle and manufacturing process of an organic fuel cell with the solid electrolyte membrane is disclosed in Japanese Patent Publication No. Hei 10-5C7572 (July, 1998). A further improvement of a proton exchange membrane (PEM) filled with inorganic materials is disclosed in Japanese Patent Publication No. 2000-515014 (November, 2000) .
The PEFC will be put to a practical use in the near future. Its characteristics have been widely studied and are reported in many literatures (see nonpatent literatures 1 and 2 as general reports).
FIG. 1 shows the working principle of PEFC using a solid electrolyte membrane. Water-premixed methanol is illustrated as an example of its fuel. Electrodes with a dispersed catalyst are placed on both sides of the menbrane to form a membrane-electrode assembly (MEA). An anode separator for supplying the fuel and a water separator for cooling are attached at the anode side, anc a cathode separator is attached for supplying air (oxygen) at the cathode side to form a single electrochemical cell. The structure of this electrochemical cell is shown in FIG. 2 (see non-patent literature 2). Multiply-connected electrochemical cells form a cell stack, and then a PEFC system is built up.
There are two types of PEFCs due to a kind of fuel: hydrogen and organic fuels. The PEFC of the organic fuels is further classified into a reformed gas supply model and a direct fuel supply model: The reformed gas supply model uses hycrogen through a reformer extracted from the fuel, such as natural gas, liquefied petroleum gas or methanol. The direct fuel supply model uses an organic fuel which is directly supplied to the fuel electrode. The direct fuel supply model using methanol is called the direct methanol fuel cell (DMFC) and seems to be put to practical use soon as
a power source for automobiles and portable electronic devices.
A perfluorosulfonic acid polymer, an organic polymer material, is commonly used for a solid electrolyte membrane of PEFC. The polymer membrane is commercially supplied: Nation (a trademark of DuPont Co.), Flemion (a trademark of Asahi Glass Co.) and Aciplex (a trademark of Asahi Kasei Co.) .
A carbon paper with a thickness of several tens micrometers is usually used for the electrode on which a dispersed catalyst of noble metal (mainly, platinum (Pt) and its alloy) is supported. Composite membranes in which a catalyst is directly supported are also used currently and their catalytic activity is reported to be maintained for a long time (see non-patent literature 3) .
Oxidation reaction at the anode of DMFC is known as a migration process of six electrons, which is ideally represented by the following formula:

This may involve complex six elementary steps of reaction. In this reaction process, HCHO, HCOOH, CO and others are detected in addition to CO2, but details of the reaction mechanism are not yet clear (see non-patent literature 2) . CO generated in this process is a catalyst poisoning species and results in a rapid reduction of the catalytic activity. DMFC is generally said to be excellent in energy efficiency since it has no heat loss caused by a reformer, however, currently it shows a low output density (that is, low cell voltage) due to a slow oxidation rate at the anode.
In order to increase the reaction rate, the working temperature may be increased. However, if increased, the polymer membrane will tend to deteriorate
and a methanol crossover phenomenon becomes serious, so that the output power may not be increased. Thus, it is a major challenge of DMFC to develop a heat-resisting and anti-crossover membrane.
The cost for electricity generation of PEFC is mostly determined by the price and lifetime of the membrane and catalyst. For example, the polymer membrane costs about one hundred thousand Japanese Yen/m2. Since for an automobile of 60 kW output, a membrane of about 10 m2 is needed, so the membrane costs about one million Yen for one car. A membrane material of high performance and low price is strongly expected.
Disclosure of the Invention
Instead of the polymer membrane, the present invention utilizes an inorganic porous electrolyte membrane made of one or plural kinds of layered silicate minerals (generally known as clay) which involve a nanometer scale of pores (that is, nanopores), and utilizes an electrolyte membrane made of an intercalation compound, which shows an improved performance by intercalating inorganic or organic ions into the interlayer space of the layered silicate mineral. Hereinafter, both the original and intercalated electrolyte membranes will be referred to as "layered silicate membranes" and the PEMC with the layered silicate membrane as the "layered silicate PEFC". Suitable candidates of the layered silicate minerals which are used in the present invention include montmorillonite (a clay mineral of the smectite group), beidellite (similarly a clay mineral of the smectite group), illite (a clay mineral of the illite group), and sericiee (a mica group mineral).
The layered silicate mineral is a proton conductive inorganic material under a suitable
condition. The working temperature of the layered silicate PEFC can be increased, if a proper layered silicate mineral is chosen and the working conditions such as moisture contents are adequately controlled. Then the catalyst can be chosen from a wide range of candidate materials. If the density and impregnated liquid contents are properly controlled when producing the layered silicate membrane, the fuel crossover can be prevented. Because of a molecular sieve effect of the layered silicate membrane, two inconsistent properties of high proton conductivity and low fuel crossover can be satisfies simultaneously. Accordingly, fundamental difficulties involved in PEFC (that is, increasing the working temperature and preventing the fuel crossover) can be solved.
The layered silicate minerals are widely distributed in nature and inexpensive. The layered silicate membrane can be easily produced by a molded compression method under control of the density and impregnated liquid contents. Moreover, a composite membrane in which a catalyst is directly supported may be produced easily.
Japanese Patent Publication No. Hei 10-5075^2 (Patent literature 1) shows a technique of impregnating montmorillonite into a polymer electrolyte membrane as a hydrophilic additive agent. However, the role of the montmorillonite in said patent is an additive agent in order to improve the electric conductivity of electrodes (cathode and anode) . This is evident from Claim 11 of Japanese Patent Publication No. Hei 10-507572 (Patent literature 1), which describes as follows: "the ioncmer is montmorillonite".
Japanese Patent Publication No. Hei 10-507572 contains no mention of the use of montmorillonite other than for a conductive additive agent improving zhe
activity of electrodes. In the present invention, the layered silicate minerals, including montmorillonite, are directly used for the electrolyte membrane to give an excellent performance, so the present invention clearly differs from Japanese Patent Publication No. Hei 10-507572.
Although Japanese Patent Laid-Open Publication Nos. 2003-175340 (patent literature 3), 2003-217341 (patent literature 4) and 2003-272439 (patent literature 5}, as patents similar to the present invention, , disclose clay minerals, these patents all relate to composite membranes dispersing a clay material in the polymer membrane, and the solid polymer is in charge of proton conductivity. This is an embodiment different from the present invention wherein the layered silicate mineral, particularly the interlayer space of stacks shown in FIG. 3(b), is in charge of txhe proton conductivity.
In Japanese Patent Laid-Open Publication Nos. 2003-175340, 2003-217341 and 2003-272439, the fact that the solid polymer is in charge of the proton conductivity is evident from the following descriptions of each laid-open publication.
In Japanese Patent Laid-Open Publication No. 2003-175430, Claim 1 set forth "an ion exchange composite membrane wherein clay or organic clay as a barrier material is dispersed in an ion-conductive polymer film". The publication contains no mention of the use of clay itself as a proton exchange membrane.
In Japanese Patent Laid-Open Publication No. 2003-21734, Claim 1 sets forth a solid polymer proton conductive membrane dispersing a clay mineral therein, and the specification describes: "clay mineral is dispersed in the solid polymer proton conductive membrane of the present invention". The publication also
contains no mention of the use of clay itself as a proton exchange membrane.
In Japanese Patent Laid-Open Publication No. 2003-272439, Claim 1 sets forth "a composite polymer electrolyte dispersing a solid polymer electrolyte and a clay mineral in the solid polymer electrolyte". Claim 4 sets forth "a composite polymer electrolyte dispersing a solid polymer electrolyte and a flake powder made of an inorganic insulating material in said polymer electrolyte". The publication also contains no mention of the use of clay itself as a proton exchange membrane.
The layered silicate mineral which is generally known as clay is an inorganic porous material with nanopores. The ion conductive characteristics of clays, particularly smectite clays, are now understood clearly based on the physical and chemical facts covering from the molecular properties to macroscopic behaviors (nonpatent literatures 4 and 5). In the present invention, after these knowledges, the optimum material conditions to permeate only proton H+ and to prevent fuel crossover were investigated and the molecular sieve property thereof was confirmed by an experiment. Since a proton is generally combined with one water molecule in an aqueous solution to form an oxonium ion (H3CO , the diffusion of oxonium ions was examined in this experiment.
In the experiment, a bentonite (including more than 99 wt% of montmorillonite) was provided to form thin circular specimens (20 mm in aiameter and 1 mm in thickness) by a molded compression method. The test piece was subjected to a permeation/diffusion test so as to determine the effective diffusion coefficients (De) of methanol, ethanol and oxonium ions.
One montimorilionite mineral is a lamellar layer with the size of approximately 100x100x1 ran3, and several
rr.ontmorillonite layers gathered in a group together with interlayer wa:er form a stack, and a number of these stacks are congregated into an aggregate/ped (see FIGS. 5 to 5) . In a pure smectite clay, there are interlayer spaces between layers and interparticle spaces between stacks and between aggregates. The space size of Compacted bentonite aepends on its dry density. If the dry density is sufficiently large, the size of the interparticle space becomes almost equal to that of the interlayer space. Based on the results of X-ray diffraction analysis and considering the geometrical arrangement of clay particles, it is believed that the interlayer spaces of dry densities of 1.4 and 1.8 Mg/m3 are about 0.9 nm and 0.6 nm, respectively, and the interparticle spaces are of similar sizes.
It is proved by molecular-based analyses that the water (an aqueous solution) in the interlayer space shows specific properties because of the surface charge of clay minerals (non-patent literature 4). In the compacted bentonite saturated with aqueous solution, the diffusion behavior of each chemical species is determined by the properties of the interlayer water and the size of the interlayer space. Since the size of a water molecule is about 0.3 nm which is smaller than that of methanol or ethanol, a molecular sieve effect is expected in the compacted bentonite to pass oxonium ions and to prevent methanol and ethanol.
The permeation/diffusion experiment was performed at a dry density 1.0 Mg/m3 which is lower than the above dry densities (1.4 ana 1.8 Mg/m3). To a solution, sodium chloride of concentration 0.1 mol/dm3 was added as a support electrolyte. As diffusing sources, methanol and ethancl of concentrations 7.5 wt% and 2.5 wt%, respectively were added. Also, as a diffusing source of oxonium ions, hydrochloric acid of concentration 1G~3 N
was added. Before the experiment, bentonite was saturatec with 0.1 mol/dirr of NaCl aqueous solution.
The concentrations of methanol and ethanol were determined by infrared absorption peak intensities of ATR-FTIR spectroscopy. Even though the mixture of methanol and ethanol was used, the peaks of the two organic materials could be clearly separated (see FIG. 6a). Under such test conditions, the calibration lines showed a sufficient linearity (see FIG. 6b) . Meanwhile, in the case of oxonium ions, the proton activity was measured by a pH meter.
Results after the elapse of two weeks are shown as follows. Effective diffusion coefficients calculated from diffusion fluxes are given in FIG. 7. For comparison, the results of diffusion coefficients of deuterium water (HDO) and tritium water (HTO) are also shown (non-patent literatures 6 and 7) . At the dry density 1.0 Mg/nr, the diffusion coefficient is of the greatest value in the case of oxonium ions and decreases in the order of tritium water, methanol and ethanol. The diffusion coefficient of oxonium ions is about two times greater than that of methanol and one exponent greater than that of ethanol. Also, the diffusion coefficients of methanol and ethanol are smaller than that of the tritium water which is a neutral molecule. For high-compacted bentonite, no ethanol crossover was observed, so the molecular sieve effect can be detected.
Eased on the experimental results of FIG. 7, it is understood that the layered silicate membrane provides a competent solid electrolyte of PEFC, if the density and impregnated liquid contents are properly controlled when producing.
Erief Description of the Drawings
FIG. 1 shows the working principle of PEFC.
FIG. 2 shows the structure of a single electrochemical cell.
FIG. 3 shews the microstructure of pure smectite clay (montmorillcnite) saturated with water.
FIG. 4 is a microphotograph of transmission electron microscope TEM) showing the cross-section of a stacK.
FIG. 5 is a microphotograph of scanning electron microscope (SEM) showing an aggregate/ped.
FIG. 6 shows the quantitative determination of methanol and ethanci. (a) infrared absorption spectra and (b) calibration lines.
FIG. 7 shows a relationship between the effective diffusion coefficient and the dry density.
Best Mode for Carrying Out the Invention If the solid electrolyte of the layered silicate membrane can be produced, it is possible to set up an electrochemical cell as shown in FIG. 1 on the basis of the existing technology and to build a fuel cell system by stacking the cells.
If the density and impregnated liquid contents are properly controlled when producing the layered silicate membrane or the intercalation compound membrane, the fuel crossover can be prevented. Because of a molecular sieve effect of the layered silicate membrane, two inconsistent properties of high proton conductivity anc low fuel crossover can be satisfies simultaneously.
Based on the exoerimental results shown in FIG. 7, ethanol is a preferable fuel of the direct organic fuel cell using the layered silicate membrane. Furthermore, if the density and impregnated liquid contents are strictly controlled when producing the layered silicate membrane, it is possible no produce a solid electrolyte of the layered silicate membrane which shows a molecular
sieve effect for organic fuels such as methanol, natural gas (methane), liquefied petroleum gas (propane) and gasoline.
In the case of the hydrogen PEFC it is known that a clay of layered silicate minerals saturated with water does not permeate gaseous hydrogen under a relatively low pressure condition. Thus, it is easy to fabricate a solid electrolyte membrane which does not permeate the hydrogen fuel.
Moreover, a composite membrane of the layered silicate mineral or the intercalation compound in which a catalyst is directly supported can also be easily fabricated. In this case, since the layered silicate mineral or the intercalation compound is an inorganic material as described above, a catalyst can be selected from a wide range of candidate materials. Furthermore, it is also easy to fabricate a membrane-electrode assembly (MEA) consisting of the solid electrolyte membrane of the layered silicate mineral or the intercalation compound, and electrodes in which electron conductive particles showing a high catalytic activity are dispersed.
Moreover, if diffusion layers of a porous material which also works as a support member of MEA, an anode separator, a cathode separator and a water separator are arranged on the faces of MEA, an inexpensive and energy-efficient electrochemical cell for PEFC can be fabricated easily.
Industrial Applicability
As can be seen from the foregoing, since the layered silicate mineral is an inorganic material, the layered silicate membrane is not: deteriorated even when the working temperature of PEFC is set at high in order to increase the reaction rate, and it is possible to
find the setting conditions for preventing the crossover of the fuel. This suggests a possibility to fabricate an efficient PEFC working at high temperature.
The layered silicate minerals are widely distributed in nature and inexpensive. Thus, the unit cost processing the solid electrolyte membrane is reduced, leading to a reduction in the cost of PEFC
As a result, according to the present invention, problems which are involved in the polymer membrane fuel cells can be improved. Then, an inexpensive and energy-efficient direct ethanol fuel cell (DEFC) and inexpensive and energy-efficient hydrogen PEFC can be fabricated.
As described above, the use of the layered silicate mineral or the intercalation compound thereof according to the embodiment of the present invention in the solid electrolyte membrane allows the fabrication of a PEFC which is inexpensive and energy-efficient. It will result in innovations in the following energy systems: (1) automobile engines; (2) industrial engines; and (3) household and industrial cogeneration systems.

Claims
1. A fuel cell comprising an anode, a solid electrolyte membrane and a cathode, wherein the solid electrolyte membrane is made of one or plural kinds of layered silicate minerals or intercalation compounds of the layered silicate minerals.
2. The fuel cell as claimed in claim 1, wherein the solid electrolyte membrane is made of one or plural kinds of intercalation compounds which are produced by intercalating inorganic ions into the interlayer space of the layered silicate minerals.
3. The fuel cell as claimed in claim 1, wherein the
solid electrolyte membrane is made of one or plural
kinds of intercalation compounds which are produced
by intercalating organic ions into the interlayer
space of layered silicate minerals.
4. The fuel cell as claimed in claim 1, wherein the
solid electrolyte membrane is fabricated under a
control of the density and impregnated liquid
contents because a molecular sieve effect is
brought into action.
5 . A membrane-electrode assembly for fuel cell as claimed in claim 1 (MEA) consisted of a solid electrolyte membrane made of one or plural kinds of layered silicate minerals or intercalation compounds of the layered silicate minerals, and electrodes in which electron conductive particles showing a high catalytic activity are dispersed.
6. An electrochemical cell comprising: a membrane-electrode assembly (MEA) as claimed in claim 5; and diffusion layers of porous material which also works as a support member of MEA, an anode separator, a cathode separator and a water separator, which are arranged on the faces of MEA.
7. A fuel cell which operates with an organic fuel,
comprising a combination of the electrochemical
cells as claimed in claim 6.
8. The fuel cell as claimed in claim 7, wherein the
organic fuel is ethanol.
9. The fuel cell as claimed in claim 7, wherein the
organic fuel is methanol.
10. The fuel cell as claimed in claim 7, wherein the organic fuel is natural gas (methane).
11. The fuel cell as claimed in claim 7, wherein the organic fuel is liquefied petroleum gas (propane).
12. The fuel cell as claimed in claim 7, wherein the organic fuel is gasoline.
13. A fuel cell which operates with hydrogen, comprising a combination of the electrochemical cellsas claimed in claim 6.