Patent Application
20050282049
The present application is based on Japanese patent application Nos. 2004-183070 and 2005-035288, the entire contents of which are incorporated herein by reference.
1. Field of the Invention
The present invention relates to a fuel cell aging method and apparatus, and particularly, an aging method and apparatus for direct methanol fuel cells used in mobile and portable power supplies, electric automobile power supplies, home cogeneration systems, etc.
2. Description of the Related Art
From the point of view of global environmental protection, and the like, expectations for fuel cells have recently been rapidly raised. Fuel cells are generally classified according to kinds of electrolytes used into five kinds of solid oxide fuel cells (SOFCs), molten carbonate fuel cells (MCFCs), alkaline fuel cells (AFCs), phosphoric acid fuel cells (PAFCs), polymer electrolyte fuel cells (PEFCs), etc.
Among others, polymer electrolyte fuel cells (hereinafter, “PEFCs”) with a polymer electrolyte membrane sandwiched between two electrodes, in which these members are further sandwiched between separators, are remarkable because of their compact structure, excellent power generation efficiency and relatively low-temperature operation, thereby having a wide range of applications.
Also, among PEFCs, particularly, direct methanol fuel cells (hereinafter, “DMFCs”) are recently remarkable, which, instead of a hydrogen gas, uses a methanol solution directly as fuel. DMFCs generate power by causing an electrochemical reaction of a fuel containing methanol and water, and an oxygen-containing oxidizer gas such as air. DMFCs have various application fields. For instance, because they operate at room temperature and can be made small and sealed, they can be used in pollution-free automobiles, home power generation systems, mobile communication equipment, medical equipment, etc.
A DMFC comprises, basically, as a unit cell (hereinafter, “cell”), a stacked body having conductive separators stacked on both sides of a membrane electrode assembly (hereinafter, “MEA”). The MEA consists of three layers in which an electrolyte membrane comprising an ion exchange resin or like is sandwiched between a pair of electrodes constituting anode and cathode electrodes. The pair of electrodes each consists of an electrode catalyst layer in contact with the electrolyte membrane, and an outer fuel or oxidizer gas diffusion layer (dispersion layer) of the electrode catalyst layer. The conductive separators are stacked so as to come into contact with the diffusion layer (dispersion layer) of the MEA, and are formed with manifold apertures which serve as passages for a fuel or oxidizer gas to flow into the diffusion layer (dispersion layer), separator temperature adjustment, waste removal, etc. Such fuel cells generate power by causing an electrochemical reaction, for example, when a mixture of methanol and water is caused to flow through the manifold apertures in contact with the diffusion layer (dispersion layer) of the anode electrode, while an oxidizing gas such as oxygen, air, or the like is caused to flow through the manifold apertures in contact with the diffusion layer (dispersion layer) of the cathode electrode.
In DMFCs, because its power generation characteristic is significantly low and unstable immediately after fuel cell assembling, about 3-40 hour-power generation at a higher temperature (typically, about 60-80° C.) than room temperature is required as initial preconditioning interim operation (hereinafter, “aging”) after DMFC assembling. This allows higher cell output than the power generation characteristic immediately after the assembling.
The problems in this aging are that, firstly, the aging requires lengthy power generation, which results in an increase in cost in mass production of fuel cells, and secondly, there are restricted aging conditions after the fuel cells are incorporated into equipment.
As a method for reducing the aging time of a PEFC, there is known an aging method described in Japanese patent application laid-open No. 2003-217622, for example.
Japanese patent application laid-open No. 2003-217622 discloses a method for operating a fuel cell having a polymer electrolyte membrane which exhibits fuel ion conductivity by retaining water, wherein the utilization ratio of a consumed gas is improved so as to cause flooding within the fuel cell during preliminary operation of the fuel cell.
As described therein, flooding caused in accordance with this configuration provides the polymer electrolyte membrane with water to increase its water content so as to form three-layer interfaces properly, whereby the preliminary operation is facilitated, and its required time reduced.
By the fuel cell aging method of Japanese patent application laid-open No. 2003-217622, however, the above-mentioned aging problems cannot be obviated sufficiently.
Accordingly, it is an object of the present invention to provide a fuel cell aging method and apparatus, which are easily and conveniently capable of reducing aging time and manufacturing cost, and of aging after the fuel cell is incorporated into equipment.
To achieve the above object, the present invention provides a fuel cell aging method, comprising the steps of: feeding an anode electrode of a fuel cell with pure water or a solution; feeding a cathode electrode of the fuel cell with an oxygen-containing gas; and forcing current to flow between the electrodes in the same direction as that of current flow during power generation of the fuel cell.
The features in the preferred embodiments of the invention are as follows:
To achieve the above object, the present invention provides a fuel cell aging apparatus, comprising: a fuel cell having anode and cathode electrodes and requiring aging; aging medium feed means for feeding the anode and cathode electrodes with pure water or a solution and an oxygen-containing gas, respectively; a voltage-applying means for forcing current to flow between the electrodes in the same direction as that of current flow during power generation of the fuel cell; and a control means for controlling the aging medium feed means and the voltage-applying means.
The features in the preferred embodiments of the invention are the same as above (1)-(6).
The aging method and apparatus according to the invention can realize making aging convenient, reducing aging time, relaxing aging conditions, and thereby making fuel cells low-cost. It is also easily capable of aging after fuel cells are incorporated into equipment.
The preferred embodiments according to the invention will be explained below referring to the drawings, wherein:
The DMFC 1 may be a well-known DMFC, and that cell comprises anode- and cathode-side separators 2A and 2B, anode and cathode electrodes 3A and 3B; and an electrolyte membrane 4. The anode and cathode electrodes 3A and 3B and electrolyte membrane 4 constitute an MEA 5 which is sandwiched between the anode- and cathode-side separators 2A and 2B. The DMFC 1 is generally used by connecting a plurality of cells in series according to target electromotive force.
The anode and cathode electrodes 3A and 3B each comprises a support layer for feeding and diffusing (dispersing) a fuel or an oxidizer gas, and a catalyst layer for oxidization or reduction reactions to occur. In the anode electrode 3A, by an oxidization reaction of methanol and water fed, hydrogen ions, electrons, and carbon dioxide are produced, and the hydrogen ions produced are conducted through the electrolyte membrane 4 to the cathode electrode 3B, while the electrons produced are conducted through an external circuit to the cathode electrode 3B. In the cathode electrode 3B, water is produced by a reduction reaction of hydrogen and oxygen ions.
For a polymer electrolyte membrane of the electrolyte membrane 4, a thin membrane (thickness: the order of 50-100 μm) may be used having a perfluorocarbon sulfonic acid structure having a sulfonic acid group as an ion exchange membrane, for example, although not limited thereto, to allow fabricating a compact cell.
The anode-side separator 2A is formed with a fuel feed groove for feeding a fuel to the adjacent anode electrode 3A, while the cathode-side separator 2B is formed with an oxidizer gas feed groove for feeding an oxidizer gas to the adjacent cathode electrode 3B, so that a fuel and oxidizer gas are fed along the surfaces of the separators 2A and 2B, respectively.
For the separators 2A and 2B, there may be suitably used a carbon separator, a carbon compound-molded separator with carbon kneaded into resin, a metallic separator having on its surface a corrosion-resistant layer formed of titanium, stainless steel or noble metals, etc, although not limited thereto.
Based on a command from the control means 12, the voltage-applying means 11 applies voltage to the DMFC 1 to thereby force current to flow therein. Using a DC (direct current) power supply is preferred, but an AC (alternating current) power supply may be used. Also, the control means 12 is provided with a CPU, etc., which controls aging, as described later.
An aging method according to an embodiment of the invention will be explained next. As the aging method, its procedure is as follows:
The aging method according to the embodiment of the invention will be explained in more detail below.
(1) Anode Medium 6A
For the anode medium 6A, water or a methanol solution is used. In actual DMFC power generation, since the methanol concentration is used on the order of 0.1-10 mol/l, filling a methanol solution of this concentration range is preferred. On the other hand, even in the case of use of pure water, if after current forcing for aging, it is replaced with a methanol solution for DMFC power generation, pure water may be used in current forcing, but in order to save labor and time in the replacement, it is preferred to use a methanol solution which is typically used as the fuel. Also, the essential point of the invention is considered to lie in the electrolysis of water, and any solution in which the MEA 5 is not damaged by the anode medium 6A may therefore be used without being limited to water and a methanol solution. For instance, there may be used an ethanol solution, isopropyl alcohol solution, etc. As its feed method, there are a method for attaching a solution tank to the anode electrode 3A, a method for feeding a solution by forced circulation, and so on, without a particular limitation. Apart from the electrolysis of water, the essential point of the invention is also considered to be related to the synthesis of water, and the effects of the invention are exhibited by a combined action of both.
(2) Cathode Medium 6B
For the cathode medium 6B, an oxygen-containing gas such as air (an oxygen-containing nitrogen gas) is fed. The oxygen content may be as high as the concentration of a pure oxygen gas, or be as low as a concentration of the order of 0.001-1%, without a particular limitation. Any oxygen-containing gas may be used, and the oxygen concentration and the kinds and concentrations of other gases contained may be selected appropriately from the point of view of convenience, economy, etc. As its feed method, there are a feed method by a natural breathing DMFC structure in which the cathode electrode 3B is left unattended in atmosphere, a method for feeding a gas by forced circulation, and soon, without a particular limitation.
(3) Current Density Je in Current Forcing
In actual DMFC power generation, power is generated typically in a range of the order of 0-200 mA/cm2. Current forcing requires a current flow of more than a load current density assumed in actual DMFC power generation, and it is therefore preferred to force constant current whose current density is within a range of 150-3000 mA/cm2, more preferably, 250-2000 mA/cm2, still more preferably, 350-1500 mA/cm2, and most preferably, 400-1400 mA/cm2. Too small a current density would have no effect, while too large a current density would cause thermal destruction to the MEA 5. At a current density of 2500 mA/cm2 or higher, in order to prevent thermal destruction of the MEA 5, it is preferred to cool the cell, or reduce current flow time (e.g. to a few seconds) in current forcing. From the point of view of convenience, etc., it is preferred to force constant current whose value is within the above range.
In the case of use of an AC power supply, it is preferred to force current whose current density is within a range of ±3000 MA/cm2, more preferably, ±2000 MA/cm2, still more preferably, ±1500 mA/cm2, and most preferably, ±1400 mA/cm2.
Even in case of too small a current density, the effects of the invention can be enhanced by increasing current flow time or the number of times of repeating current flow.
(4) Applied Voltage V in Current Forcing
During current flow, it is preferred to apply, as the voltage between the electrodes per cell, a voltage of 0.3-3 V, more preferably, 0.6-2.7 V, and still more preferably, 0.9-2.5 V. Too small a voltage would cause no electrolysis, which therefore has almost no aging effect. Too large a voltage can result in an undesirable thermal or electrical damage.
(5) Current Flow Time t and the Number of Times of Current Forcing in Current Forcing
The current flow time is preferably a few seconds to a few minutes. Too short a current flow time would have no effects. Too long a current flow time would undesirably result in an increase of the temperature of the MEA 5 due to heat generation, and consequently, in an electrochemical reaction (e.g. corrosion) of the separators progressing. The proper current flow time is a current flow time until before the temperature of the MEA 5 reaches 100° C., or until before the maximum voltage per cell during constant current forcing reaches 3 V, at which the current flow is zero. It is preferred to repeat such operation 2-6 times, more preferably, 3-5 times. Repeating current flow would make voltage at the start of current flow lower than voltage at the preceding start of current flow, but it is preferred to repeat current flow until almost no voltage reduction is caused, and more preferably, until before no voltage reduction is caused. In other words, it is preferred to repeat current flow until the cell internal resistance stabilizes. For example, if the voltage at the fourth start of current flow is substantially the same as the voltage at the third start of current flow, it is preferred to complete the aging process at the third current forcing.
(6) The conditions for the applied voltage V, forced current I (current density Je) and current flow time t in (3)-(5) above are determined by the following steps of:
When the temperature T1 of the MEA 5 is on the order of room temperature (25-30° C.), it is preferred that ΔT<60-70.
The reason for ΔT [° C.]<100 is because, at T1>0, the maximum temperature of the MEA 5 does not exceed 100° C. The reason for q [W/cm2]<100 is because q [W/cm2]>100 would cause an interface transition from nuclear boiling to membrane boiling, which would result in poor heat dissipation, and further produce a boiling membrane in an anode interface which would undesirably inhibit electrolysis of water.
(8) dV/dI Characteristics During Current Forcing
It is preferred to cause large current to flow by applying as small a voltage as possible. To this end, it is preferred to cause constant current flow whose value is below the moderate range.
The above aging method allows quick aging, which serves conveniently even as ancillary facilities, compared to conventional methods. Because of the specified current flow, aging is also made possible after the DMFC 1 is incorporated into equipment.
(1) Making a Fuel Cell for Experiment
As separators for a DMFC, metal cladding sheet materials having corrosion resistance and surface conductivity is fabricated. Using a composite metallic member of Ti/SUS/Ti in which stainless steel (SUS 304) is used as the core metal and metal titanium is used as the cladding metal, surface treatment is performed on this member for having combined conductivity and corrosion resistance, by the method disclosed in Japanese patent application laid-open No. 2004-158437. Using this metallic member and an MEA (Nafion (registered trademark) used as the electrolyte membrane), a cell with an electrode surface area S=8.4 cm2) is assembled.
(2) Current Forcing Experiment 1
Using a cell assembled, this current forcing experiment is performed. In this experiment, the anode injection solution amount is taken to be va=1 [cm3/sec]. Also, the specific heat of the anode injection solution is taken to be the value of water as a representative value, C2=4.2 [J/(g·K)] and ρ2=1.0 [g/cm3]. Accordingly, since I=electrode surface area S×forced current density Je, substituting into equation (4) above, it follows that ΔT=(V1+V2)/Je×t.
A DC power supply is prepared. The anode electrode of the DMFC is connected to a positive terminal of the DC power supply output, while its cathode electrode is connected to a negative terminal of the DC power supply output. Current is forced to flow in the same direction as that during normal power generation, so that the current density is constant. In each sample, the number of times of current forcing is three. Using pure water (sample 8) or a methanol solution (0.1 mol/l, 1.0 mol/l, 3.0 mol/l, 8.0 mol/l, 10.0 mol/l) (samples 1-7, 9-16) as the anode feed solution and air (samples 1-12), a nitrogen gas containing 0.001-1% of oxygen (samples 13-16) or pure oxygen (oxygen left after removing inevitable impurities) (sample 17) as the cathode feed gas, the experiment is implemented by forcing each of them to circulate.
For each sample cell, current forcing is performed in various conditions and subsequently, DMFC power generation characteristic assessment is performed at room temperature 25° C., using air (samples 1-16) and pure oxygen (sample 17) as the cathode feed gas, and taking the cathode feed gas utilization ratio as 10%. The assessment results are shown in Table 1. Table 1 shows forced current conditions, and maximum outputs obtained for the DMFC power generation characteristic. The maximum outputs are converted into values per MEA electrode surface area. Table 1 also shows voltage V1 immediately after the start of the constant current forcing and voltage V2 immediately before the end of the current flow. In Table 1, ΔT is the calculated values, and T1 and T2 are the measured values.
Next, for the following cells as comparison examples, DMFC power generation characteristic assessment is performed in the same manner as the above embodiments: cells on which no aging is performed (Samples X1 and X2 use air and pure oxygen, respectively, as the cathode feed gas in the DMFC power generation characteristic assessment); and cells on which DMFC power generation is performed at 60° C. for 8 hrs as aging (Samples Y1-Y4 and Y5 use air and pure oxygen, respectively, as the cathode feed gas in the aging and DMFC power generation characteristic assessment). The assessment results are shown in Table 2.
*1: Sample 13 uses a mixture of (1% oxygen + remaining nitrogen) as the cathod feed gas in current forcing (aging), and air as the cathod feed gas in assessment of the DMFC power generation characteristic.
*2: Sample 14 uses a mixture of (0.1% + oxygen + remaining nitrogen) as the cathod feed gas in current forcing (aging), and air as the cathod feed gas in assessment of the DMFC power generation characteristic.
*3: Sample 15 uses a mixture of (0.01% oxygen + remaining nitrogen) as the cathod feed gas in current forcing (aging), and air as the cathod feed gas in assessment of the DMFC power generation characteristic.
*4: Sample 16 uses a mixture of (0.001% oxygen + remaining nitrogen) as the cathod feed gas in current forcing (aging), and air as the cathod feed gas in assessment of the DMFC power generation characteristic.
*5: Sample 17 uses pure oxygen as the cathod feed gas in current forcing (aging), and assessment of the DMFC power generation characteristic.
*6Pure oxygen is used as the cathod feed gas in assessment of the DMFC power generation characteristic.
For samples 1-5, this current forcing experiment is performed in the condition of ΔT=30-40. For sample 1, no distinct hydrogen generation can be detected during the current forcing, whereas, for samples 2-5, hydrogen generation is detected.
Samples 2-5 exhibit substantially the same DMFC power generation characteristic as that of comparison sample Y1, so the current forcing has the aging effect on samples 2-5. Although sample 1 exhibits the lower DMFC power generation characteristic than that of comparison sample Y1, it exhibits the higher DMFC power generation characteristic than that of comparison sample X1 on which no aging is performed, and the current forcing is therefore considered to have the aging effect on sample 1 as well. In other words, for sample 1, the current forcing conditions are considered to indicate a lower limit at which the aging effect develops.
Also, for sample 6, the current forcing experiment is performed in the condition of ΔT=58. Sample 6 exhibits the same DMFC power generation characteristic as that of comparison sample Y1, so the current forcing has the aging effect on sample 6. On the other hand, for sample 7, the current forcing experiment is performed in the condition of ΔT=73 (T2-105). Sample 7 exhibits the lower DMFC power generation characteristic than that of comparison sample Y1. From this, in the current forcing conditions of ΔT>70 and T1=30, i.e., T2>100, the MEA is considered to degrade due to water or solution boiling, which results in a decrease of the cell function. In other words, the conditions in which the water (solution) boils inside the cell are considered to indicate an upper limit of effective aging.
In addition to the cases of 1.0 mol/l-methanol solutions in samples 1-7, as shown in the results of samples 8, 9 and 10, when the anode injection solution is pure water, 0.1 and 10.0 mol/l-methanol solutions, the DMFC power generation characteristic after current forcing (aging) exhibits the suitable values (the same value as that of comparison sample Y2, in the case of the 10.0 mol/l-methanol solution), so the current forcing has the aging effect.
Also, both when the anode injection solution is 3 and 8 mol/l-methanol solutions (samples 11 and 12), and when the cathode feed gas is a nitrogen gas containing 0.001-1% of oxygen (samples 13-16), the DMFC power generation characteristic after current forcing is the same as or higher than those of comparison samples Y3 and Y4, so the current forcing has the aging effect.
Further, when the cathode feed gas is pure oxygen (sample 17), the DMFC power generation characteristic after current forcing is higher than that of comparison sample X2 on which no aging is performed, and exhibits the same value as that of comparison sample Y5, so the current forcing has the aging effect.
From the above implementation results, current forcing is considered to have the aging effect in various combinations of an anode injection solution range at least from pure water to 10.0 mol/l-methanol solutions, and a cathode feed gas range at least from a nitrogen gas containing 0.001% of oxygen to pure oxygen.
(3) Current Forcing Experiment 2
Using DC and AC power supplies, this current forcing experiment is performed. First, the anode electrode of the DMFC is connected to a positive terminal of the DC power supply output, while its cathode electrode is connected to a negative terminal of the DC power supply output. Current is forced to flow in the same direction as that during normal power generation (current forcing order: 1). Subsequently, by reverse connection, current is forced to flow in the reverse direction to a direction during normal power generation (current forcing order: 2). Next, by causing the connection to return to the first connection state, current forcing is performed (current forcing order: 3), and then by replacing the DC power supply to connect the AC power supply, current forcing is performed (current forcing order: 4). Further, for one cell (sample 17), by changing the conditions in order shown in Table 3, current forcing is performed continuously (current forcing order: 5-8). The number of times of current forcing is three in only the first conditions (current forcing order: 1; +450 mA/cm2; 30 sec), and one in the subsequent current forcing conditions. Using a methanol solution (1.0 mol/l) as the anode feed solution and air as the cathode feed gas, the experiment is implemented by forcing each of them to circulate. Subsequently, the DMFC power generation characteristic is assessed in the same manner as in “current forcing experiment 1”. The assessment results are shown in Table 3.
+: the same direction as a direction in normal power generation
−: the reverse direction to a direction in normal power generation
±: AC current flow
*7Cooling is being performed during current forcing.
From Table 3, itis seen that current forcing in the reverse direction to a direction during normal power generation (current forcing order: 2 and 5) exhibits no effect of the invention, and that current forcing by the AC power supply (current forcing order: 4) exhibits the effect of the invention since current forcing is also performed in the same direction as a direction during normal power generation. It is also seen that even the current density in the overcurrent range (current forcing order: 7 and 8) is made applicable by current-flow time adjustment or cell cooling during current forcing.
Although the invention has been described with respect to the specific embodiments for complete and clear disclosure, the appended claims are not to be thus limited but are to be construed as embodying all modifications and alternative constructions that may occur to one skilled in the art which fairly fall within the basic teaching herein set forth.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2004-183070 | Jun 2004 | JP | national |
| 2005-035288 | Feb 2005 | JP | national |
