CN1167544A - Organic fuel cell, method of operation thereof, and electrode fabrication method - Google Patents

Abstract

A liquid organic fuel cell (10) using a solid electrolyte membrane (18) is provided. An organic fuel such as a methanol/water mixture is circulated through the anode (14) of the cell, while oxygen or air is circulated through the cathode (16). The cell electrolyte membrane is preferably made of Nafion^TMAnd (4) preparing. Also, a method of improving the structural properties of a carbon electrode is provided, wherein a large surface carbon particle/Teflon^TMThe adhesive structure is dipped into Nafion^TM/methanol bath with Nafion^TMElectrons are impregnated. A method of making the fuel cell anode is described in which a metal alloy is deposited onto the electrode from a solution containing perfluorooctylsulfonic acid. A fuel additive and a novel organic fuel containing such an acid are also described.

Description

Organic fuel cell, method of operation thereof, and electrode fabrication method

Background of the Invention Origin of the Invention

The invention described herein was made in the performance of contract work by NASA and in accordance with the terms of Public Law 96-517 (35 USC 202), where the contractors have elected to retain title.

technical field

The present invention relates generally to organic fuel cells and in particular to liquid fed organic fuel cells.

technical background

Fuel cells are electrochemical cells in which free energy from fuel oxidation reactions is converted into electrical energy. In an organic/air fuel cell, an organic fuel such as methanol, formaldehyde or formic acid is oxidized to carbon dioxide at the anode, while air or oxygen is reduced to water at the cathode. Fuel cells employing organic fuels are particularly attractive for stationary and portable applications, partly because of the high specific energy of organic fuels, such as 6232 Wh/kg for methanol.

Two types of organic/air fuel cells are generally known:

CLAIMS 1. An "indirect" or "reformer" fuel cell in which an organic fuel is catalytically reformed and processed into carbon monoxide-free hydrogen, the hydrogen so obtained being oxidized at the anode of the fuel cell.

2. A "direct oxidation" fuel cell in which the organic fuel is added directly to the fuel cell without any pre-chemical modification and is oxidized at the anode.

Direct oxidation fuel cells do not require a fuel treatment step. As such, direct oxidation fuel cells offer considerable weight and volume advantages over indirect fuel cells. Direct oxidation fuel cells employ vapor or liquid feeds of organic fuels. Promising prior art direct oxidation fuel cells typically use a liquid feed design in which a liquid mixture of organic fuel and sulfuric acid electrolyte is circulated through the anode of the fuel cell.

There are several problems with the use of sulfuric acid electrolyte in prior art direct methanol fuel cells. The application of highly corrosive sulfuric acid has greatly limited the structural materials of fuel cells. Typically, expensive corrosion resistant materials are required. Sulfate anions generated in fuel cells have a strong tendency to adsorb on electrocatalysts, thus hindering the kinetics of electrooxidation of fuels and leading to poor performance of fuel electrodes. Meanwhile, sulfuric acid begins to degrade at temperatures above 80 °C and the degradation products usually contain sulfur that can poison the electrocatalyst. In multi-battery stacks, the application of sulfuric acid electrolytes can lead to parasitic bypass currents.

Examples of direct and indirect type fuel cells are described in US Pat.

For example, US Patents 3,013,908 and 3,113,049 describe liquid-fed direct methanol fuel cells using a sulfuric acid electrolyte. U.S. Patents 4,262,063, 4,390,603, 4,478,917, and 4,629,664 describe improvements in sulfuric acid-based methanol fuel cells in which a high molecular weight electrolyte or a solid proton-conducting membrane is inserted between the cathode and anode as an ion-conducting layer to reduce the Crossover of organic fuel to the cathode. While the application of ionically conductive layers helps reduce crossover, ionically conductive layers can only be used in conjunction with sulfuric acid electrolytes. Thus, fuel cells suffer from the above-mentioned disadvantages of using sulfuric acid as an electrolyte.

In view of the above problems associated with the use of sulfuric acid as an electrolyte, it would be desirable to provide a liquid fed fuel cell which does not require sulfuric acid as an electrolyte.

In addition to improvements in the operating characteristics of liquid-fed fuel cells, conventional fabrication methods for large-surface electrocatalytic electrodes in such fuel cells also need to be improved. The existing method of manufacturing fuel cell electrodes is a rather time consuming and expensive process. Specifically, firstly, electrode fabrication needs to prepare a large-surface carbon-supported alloy powder by a chemical method, which usually takes about 24 hours. Once prepared, the carbon-supported alloy powder was combined with a Teflon ^™ binder and coated onto a carbon fiber-based support to obtain a gas diffusion electrode. In order to volatilize impurities from the Teflon ^™ binder and obtain a Teflon ^™ matrix of fibrous bodies, the electrodes are heated to 200-300°C. During the heating step, oxidation and sintering of the electrocatalyst may occur, leading to a decrease in the activity of the electrode surface. As such, electrodes often require reactivation before use.

Electrodes fabricated by conventional methods are also generally gas diffusion type and cannot be effectively applied to liquid fed fuel cells because the electrodes cannot be completely wetted by liquid fuel. In general, the structure and performance of fuel oxidation electrodes (anodes) used in liquid-fed fuel cells are completely different from gas/vapour-fed fuel cells such as hydrogen/oxygen fuel cells. Electrode structures applied to liquid-fed fuel cells should be porous and the liquid fuel solution should wet all pores. Carbon dioxide evolved on the fuel electrode should be efficiently released from the reaction region. Complete wetting of the electrodes is a major problem for liquid-fed fuel cells, even for liquid-fed fuel cells using sulfuric acid electrolytes.

It will be appreciated that it would be desirable to provide improved methods of making electrodes, particularly for applications in liquid-fed fuel cells. It would also be desirable to devise methods for modifying electrodes originally suitable for use in gas-fed fuel cells for use in liquid-fed fuel cells.

In addition to improving the liquid-fed fuel cells themselves and providing improved methods of making fuel cell electrodes, it is also desirable to provide new efficient fuels. In general, it is desirable to provide liquid fuels that can be electrochemically oxidized cleanly and efficiently in fuel cells. In general, in direct oxidation fuel cells, the efficient use of organic fuels depends on the ease of anodic oxidation of organic compounds in the fuel cell. Electrooxidation of conventional organic fuels such as methanol is quite difficult. In particular, the electrooxidation of organic compounds such as methanol involves the transfer of multiple electrons and is a strongly hindered process with several intermediate steps. These steps involve dissociative adsorption of fuel molecules to form active surface species that are more easily oxidized. Ease of dissociation adsorption and surface reactions generally determine the ease of electrooxidation. Other conventional fuels such as formaldehyde are more susceptible to oxidation, but have other disadvantages as well. For example, formaldehyde is highly toxic. At the same time, formaldehyde is particularly soluble in water, so it can pass through to the cathode of the fuel cell, which reduces the performance of the fuel cell. Other conventional organic fuels such as formic acid are corrosive. Further, many conventional organic fuels poison the electrodes of the fuel cell during the electro-oxidation process, thus preventing continuous operation. It will be appreciated that it would be desirable to provide improved fuels, particularly for use in liquid-fed fuel cells, which overcome the disadvantages of conventional organic fuels such as methanol, formaldehyde and formic acid.

Summary of the invention

A general object of the present invention is to provide an improved direct liquid fed fuel cell. A specific object of the present invention is to provide a direct type liquid fed fuel cell which does not require a sulfuric acid electrolyte. Another specific object of the present invention is to obtain fully wetted electrodes for use in liquid-fed fuel cells. Yet another specific object of the present invention is to provide an improved method of wetting applied to fuel cell electrodes having a sulfuric acid electrolyte. Another specific object of the present invention is to provide improved fuels for use in liquid-fed fuel cells.

The object of providing an improved liquid-fed direct fuel cell that does not require a sulfuric acid electrolyte is achieved in part through the use of a solid polymer electrolyte membrane in combination with a cell-type anode that is porous and capable of wetting the fuel. In an improved liquid-fed fuel cell, a cell-type anode structure and cathode are bonded to either side of a solid polymer proton conducting membrane that forms a membrane electrode assembly. A solution of methanol and water substantially free of sulfuric acid is circulated through the anode side of the component.

A solid polymer membrane was used in part because of its excellent electrochemical and mechanical stability, high ionic conductivity, and ability to function as both electrolyte and separator. Meanwhile, the kinetics of methanol electrooxidation and air or oxygen electroreduction are easier at the electrode/membrane electrolyte interface than at the electrode/membrane electrolyte interface. The application of membranes allows fuel cells to operate at temperatures as high as 120°C. Since the fuel and water solution is substantially free of sulfuric acid, it is not necessary to employ expensive corrosion-resistant components in the fuel cell and its auxiliaries. At the same time, the absence of conducting ions in the fuel and aqueous solution essentially eliminates any parasitic bypass currents of the multi-cell stack, which are present when sulfuric acid electrolytes are used.

The solid polymer electrolyte is preferably a proton-conducting cation exchange membrane, such as a perfluorinated sulfonic acid polymer membrane, Nafion ^™ . Nafion ^TM is a copolymer of tetrafluoroethylene and perfluorovinyl ether sulfonic acid. Modified perfluorinated sulfonic acid polymer membranes, polyhydrocarbyl sulfonic acid membranes and composites of two or more proton exchange membranes may also be used.

The anode is preferably formed of large surface particles of platinum-based alloys of noble and non-noble metals. Binary and ternary compositions can be used in the electrooxidation of organic fuels. Platinum-ruthenium alloys with a composition of 10-90% platinum atoms are preferred anode electrocatalysts for the electrooxidation of methanol. The alloy particles are either in the form of finely divided metal powders, "unsupported", or supported on large surface carbon materials.

Conventional fuel cell anode structures (gas diffusion type) are not suitable for liquid feed type organic/air fuel cells. These conventional electrodes have poor fuel wetting properties. These conventional electrodes are modified for use in liquid-fed fuel cells by coating them with substances that improve their wetting properties. Nafion ^(TM) having an equivalent weight of 1000 or higher is the preferred material. The additive lowers the interfacial tension of the liquid/catalyst interface and enables uniform wetting of the electrode pores and particles by the fuel and water solution, resulting in enhanced utilization of the electrocatalyst. In addition to improving wetting properties, Nafion ^™ additives also provide ionic continuity to the solid electrolyte membrane and allow efficient transport of protons or hydronium ions generated by fuel oxidation reactions. Further, the additive makes it easier for carbon dioxide to escape from the pores of the electrode. By using perfluorinated sulfonic acid as an additive, anionic groups cannot be strongly adsorbed on the electrode/electrolyte interface. As a result, methanol electrooxidation kinetics are easier than in sulfuric acid electrolytes. Other hydrophilic proton-conducting additives with desirable properties include montmorillonites, alkoxycelluloses, cyclodextrins, zeolite mixtures, and zirconium hydrogen phosphate.

The goal of improving electrodes for operation in liquid-fed fuel cells was achieved in part through the use of PFOS as an additive in the electrodeposition baths used in the fabrication of the electrodes. A method of electrodeposition using PFOS additives involves installing a large surface carbon electrode structure in a bath containing a metal salt, installing an anode in the bath and applying a voltage between the anode and cathode up to a desired amount The step of metal deposition on the electrode. After the metal has been deposited on the electrode, the electrode is removed from the bath and washed in deionized water.

Preferably, metal salts include hydrogen hexachloroplatinate and potassium pentachlororuthenate. The anode is composed of platinum. The carbon electrode structure consists of large surface carbon particles bonded to polytetrafluoroethylene, sold under the trademark Teflon ^™ .

The goal of providing fully wetted electrodes in liquid-fed fuel cells containing sulfuric acid electrolytes is achieved by using PFOS as an additive to the fuel mixture of the fuel cell. Preferably, perfluorooctane sulfonic acid is added to the mixture of organic fuel and water at a concentration of 0.001-0.1M.

The general aim of providing new fuels for organic fuel cells is achieved by using trimethoxymethane, dimethoxymethane or trioxane. In fuel cells, all three new fuels can be oxidized to carbon dioxide and water at high rates without poisoning the electrodes. Further, neither trimethoxymethane, dimethoxymethane, nor trioxane are corrosive. The oxidation rates of the three new fuels were substantially the same or better than those of conventional organic fuels. For example, the oxidation rate of dimethoxymethane is higher than that of methanol at the same temperature. The oxidation rate of trioxane is substantially the same as that of formaldehyde. However, trioxane has a much higher molecular weight than formaldehyde, and therefore, trioxane molecules cannot travel across to the fuel cell's cathode as easily as formaldehyde molecules.

Trimethoxymethane, dimethoxymethane and trioxane can be used in fuel cells with any of the improvements described above. However, the improved fuel may also be advantageously used in other organic fuel cells, including entirely conventional fuel cells.

In this way the various general objects of the invention described above are achieved. Other objects and advantages of the invention will be apparent from the following detailed description.

Brief description of the drawings

Objects and advantages of the present invention will become more apparent upon examination of the following detailed description and accompanying drawings, in which:

Figure 1 provides a schematic representation of an improved liquid-fed organic fuel cell having a solid polymer membrane embodied in accordance with a preferred embodiment of the present invention.

FIG. 2 provides a schematic representation of a multi-cell fuel system using the improved liquid-fed organic fuel cell of FIG. 1 .

Fig. 3 is a graph showing the performance of a solid polymer membrane electrolyte and a sulfuric acid electrolyte in a liquid organic fuel.

4 is a graph showing the performance of the liquid-fed fuel cell of FIG. 1 for methanol/air and methanol/oxygen compositions.

FIG. 5 is a graph showing the effect of fuel concentration on the performance of the liquid-fed fuel cell of FIG. 1 .

FIG. 6 is a graph showing polarization behavior of a fuel electrode and a cathode in the fuel cell of FIG. 1 .

Figure 7 is a block diagram showing a method of making electrodes containing hydrophilic proton conducting ionomer additives for use in liquid fed cells.

FIG. 8 is a graph showing the polarization characteristics of methanol oxidation on an electrode containing an ionomer additive and prepared according to the method shown in FIG. 7 .

Figure 9 is a block diagram showing a method of using PFOS in an electrodeposition bath to make an electrode.

Figure 10 schematically represents an electrochemical cell for carrying out the process of Figure 9 .

FIG. 11 shows polarization curves for electrodes fabricated using the method of FIG. 9 .

Fig. 12 is a graph showing polarization curves of a fuel cell using a sulfuric acid electrolyte and using perfluorooctane sulfonic acid as a fuel additive.

13 is a graph showing polarization curves for different fuel concentrations for a fuel cell using dimethoxymethane as fuel in a half-cell containing a sulfuric acid electrolyte.

Figure 14 is a graph showing the polarization curves for various temperatures and concentrations for a fuel cell using dimethoxymethane as fuel in a half cell containing a sulfuric acid electrolyte.

FIG. 15 is a graph showing cell voltage as a function of current density for the fuel cell of FIG. 1 using dimethoxymethane as a fuel.

16 is a graph showing polarization curves for different fuel concentrations for a fuel cell using trimethoxymethane as fuel in a half-cell containing a sulfuric acid electrolyte.

Figure 17 is a graph showing the polarization curves for various temperatures and concentrations for a fuel cell using trimethoxymethane as fuel in a half cell containing a sulfuric acid electrolyte.

FIG. 18 is a graph showing cell voltage as a function of current density for the fuel cell of FIG. 1 using trimethoxymethane or methanol as a fuel.

Figure 19 is a graph showing polarization curves for different fuel concentrations for a fuel cell using trioxane as fuel in a half cell containing a two molar sulfuric acid electrolyte.

Figure 20 is a graph showing the polarization curves of a fuel cell using trioxane as fuel in a half cell for various temperatures and concentrations of sulfuric acid electrolyte.

FIG. 21 is a graph showing cell voltage as a function of current density for the fuel cell of FIG. 1 using trioxane as a fuel.

Detailed description of the invention

Referring to these figures, a preferred embodiment of the present invention will now be described. First, referring primarily to FIGS. 1-6, an improved liquid-fed organic fuel cell employing a solid polyelectrolyte membrane and an ionomer anode additive is described. Then, with reference to Figures 7-8, a method of making an anode with an ionomer additive is described. Referring to Figures 9-11, a method for achieving improved wetting by fabricating electrodes in a bath containing PFOS is described. Referring to FIG. 12, a fuel cell using perfluorooctane sulfonic acid as a fuel additive is described. Referring to FIGS. 13-21 , a fuel cell using dimethoxymethane, trimethoxymethane, and trioxane as fuel will be described. Fuel cells using solid proton-conducting electrolyte membranes

FIG. 1 shows a liquid-fed organic fuel cell 10 having a housing 12, an anode 14, a cathode 16, and a solid polymer proton-conducting cation exchange electrolyte membrane 18. As shown in FIG. As described in more detail below, anode 14, cathode 16 and solid polymer electrolyte membrane 18 are preferably a single multilayer composite structure, referred to herein as a membrane electrode assembly. The pump 20 is used to pump the organic fuel and aqueous solution into the anode chamber 22 of the tank 12 . The mixture of organic fuel and water is discharged through the outlet pipe 23 and recycled through a recirculation system including a methanol tank 19, which will be described below with reference to FIG. The carbon dioxide formed in the anode chamber is discharged through the outlet 24 in the tank 19 . An oxygen or air compressor 26 is used to force oxygen or air into the cathode chamber 28 in the tank 12 . FIG. 2 described below shows a fuel cell system in which individual fuel cells including a recirculation system are combined into a stack. The following detailed description of the fuel cell of FIG. 1 focuses primarily on the structure and function of the anode 14 , cathode 16 and membrane 18 .

Before use, the anode compartment 22 is filled with a mixture of organic fuel and water, and the cathode compartment 28 is filled with air or oxygen. During operation, organic fuel is circulated through the anode 14 and oxygen or air is pumped into the chamber 28 and circulated through the cathode 16 . Electrooxidation of the organic fuel occurs at the anode 14 and electroreduction of oxygen occurs at the cathode 16 when an electrical load (not shown) is connected between the anode 14 and the cathode 16 . The different reactions that take place at the anode and cathode result in different voltages across the two electrodes. Electrons generated by electro-oxidation at the anode 14 are conducted through an external load (not shown) and eventually captured at the cathode 16 . Hydrogen ions or protons produced at the anode 14 are transferred directly through the membrane electrolyte 18 to the cathode 16 . The current is then maintained by the flow of ions in the cell and electrons through the external load.

As noted above, the anode 14, cathode 16 and membrane 18 form a single composite layered structure. In a preferred embodiment, membrane 18 is formed from Nafion ^™ , a perfluorinated proton exchange membrane material. Nafion ^TM is a copolymer of tetrafluoroethylene and perfluoroethylene ether sulfonic acid. Other membrane materials can also be used. For example, membranes of modified perfluorosulfonic acid polymers and polyhydrocarbylsulfonic acids and composites of two or more proton exchange membranes can be used.

Anode 14 is formed from platinum-ruthenium alloy particles either as a finely divided metal powder, "unsupported" or dispersed on large surface carbon, "supported". The large surface carbon may be, for example, Vulcan XC-72A, a material supplied by Cabot Inc., USA. A carbon fiber sheet backing (not shown) is used to make electrical contact with the electrocatalyst particles. Commercially available Toray ^™ paper was used as the electrode mount. Alloy electrocatalysts supported on Toray ^™ paper backings were obtained from E-Tek, Inc. of Framingham, Massachusetts. Alternatively, unsupported and supported electrocatalysts can be prepared chemically, combined with a Teflon ^™ binder and dispersed on a Toray ^™ paper backing to make the anode. An efficient and time-saving electrocatalytic electrode fabrication method is described in detail below.

Platinum-based alloys in which the second metal is tin, iridium, osmium or rhenium can also be used instead of platinum-ruthenium. In general, the choice of alloy depends on the fuel used in the fuel cell. Platinum-ruthenium is preferred for the electrooxidation of methanol. For platinum-ruthenium, the loading amount of alloy particles in the electrocatalyst layer is preferably 0.5-4.0 mg/cm ² . More efficient electrooxidation was achieved at higher loadings rather than lower loadings.

Cathode 16 is a gas diffusion electrode in which platinum particles are bound to one side of membrane 18 . Cathode 16 is preferably formed of unsupported or supported platinum bonded to the side of membrane 18 opposite anode 14 . Unsupported platinum black (fuel cell grade) from Johnson Matthey Inc., USA or supported platinum material from E-Tek Inc., USA was suitable for the cathode. Like the anode, the cathode metal particles are preferably immobilized on the carbon backing material. The supported amount of electrocatalyst particles on the carbon backing material is preferably 0.54.0 mg/cm ² . The electrocatalyst alloy and carbon fiber backing contain 10-50% by weight of Teflon ^™ to provide the hydrophobicity required to form a three-phase boundary and effectively remove water produced by the electroreduction of oxygen.

During operation, a mixture of fuel and water (without acidic or alkaline electrolyte) at a concentration of 0.5-3.0 moles per liter is circulated in the anode chamber 22 past the anode 14 . Preferably, a flow rate of 10-500 ml/min is used. When the mixture of fuel and water circulates through the anode 14, the following electrochemical reaction occurs and electrons are released, taking methanol cell as an example:

anode: (1)

The carbon dioxide produced by the above reaction is discharged through the outlet 23 along with the fuel and the aqueous solution, and is separated from the solution in the gas-liquid separator (described below with reference to FIG. 2 ). The fuel and water solution is then recirculated by pump 20 into the pool.

Along with the electrochemical reaction described in Equation 1 above, another electrochemical reaction occurs at the cathode 16 involving the electroreduction of oxygen, which can capture electrons, and the reaction is given by

cathode: (2)

Taking methanol fuel cells as an example, the single-electrode reactions described by Equations 1 and 2 lead to an overall reaction given by:

Battery: (3)

In the case of sufficiently high fuel concentration, a current density greater than 500mA/cm ² can be maintained. However, at these concentrations, the rate of fuel crossover across the membrane 18 to the cathode 16 increases to such an extent that the efficiency and electrical performance of the fuel cell is greatly reduced. Concentrations below 0.5 mol/L limit cell operation to current densities below 100 mA/cm ² . We have found that at low current densities low flow rates can be used. High flow rates are required when operating at high current densities that increase the rate of mass transfer of the organic fuel to the anode while removing carbon dioxide produced by the electrochemical reaction. Low flow rates also reduce fuel crossover through the membrane from anode to cathode.

Preferably, oxygen or air is circulated through cathode 16 at a pressure of 10-30 psig. Above-ambient pressure improves the mass transfer of oxygen to the electrochemical reaction region, especially at high current densities. Water produced by the electrochemical reaction at the cathode is expelled from the cathode chamber 28 through the outlet 30 by the flow of oxygen.

In addition to undergoing electro-oxidation at the anode, liquid fuel dissolved in water permeates the solid polymer electrolyte membrane 18 and combines with oxygen on the surface of the cathode electrocatalyst. This process is described in Equation 3 using methanol as an example. This phenomenon is called "fuel crossing". Fuel crossover reduces the operating potential of the oxygen electrode and results in fuel consumption without generating useful electrical power. Generally, fuel crossover is an additional reaction that reduces efficiency, reduces performance and generates heat in a fuel cell. Therefore, it is desirable to minimize the fuel crossover rate. The crossover rate is proportional to the permeability of the fuel through the solid electrolyte membrane and increases with concentration and temperature. By selecting a solid electrolyte membrane with a low water content, the permeability of the membrane to liquid fuels can be reduced. For fuels, the reduction in permeability results in low crossover rates. Meanwhile, a fuel with a large molecular size has a small diffusion coefficient compared to a fuel with a small molecular size. Therefore, the permeability can be reduced by selecting a fuel with a large molecular size. When water soluble fuels are desired, fuels with moderate water solubility exhibit low permeability. Fuels with high boiling points do not vaporize and they pass across the membrane in the liquid phase. Fuels with high boiling points generally have low crossover rates because vapors are more permeable than liquids. Crossover rates can also be reduced by reducing the concentration of liquid fuel. With an optimal distribution of hydrophobic and hydrophilic moieties, the anode structure is fully wetted by the liquid fuel to sustain the electrochemical reaction and prevent excess fuel from accessing the membrane electrolyte. Therefore, proper selection of the anode structure can lead to high performance and desirably low crossover rates.

Because the solid electrolyte membrane is permeable to water at temperatures above 60°C, a considerable amount of water is transported across the membrane by osmosis and evaporation. Water passing through the membrane is condensed in the water recovery system and conveyed into the water tank (both parts are described below with reference to FIG. 2 ) so that the water can re-enter the anode chamber 22 .

Protons produced by the anode 14 and water produced by the cathode 16 are transferred between the two electrodes through the proton-conducting solid electrolyte membrane 18 . Maintaining high proton conductivity of the membrane 18 is important for efficient operation of the organic/air fuel cell. The water content of the membrane is maintained by direct contact with the mixture of liquid fuel and water. The thickness of the proton-conducting solid polymer electrolyte membrane should be a stable dimension of 0.05-0.5 mm. Membranes thinner than 0.05 mm may result in membrane-electrode components with poor mechanical strength, while membranes thicker than 0.5 mm swell the polymer by a solution of liquid fuel and water, causing drastic destructive dimensional changes, and displaying excessive electrical resistance . The ion conductance of the membrane should be greater than 1 ohm ^-1 cm ^-1 so that the fuel cell has an allowable internal resistance. As mentioned above, the membrane should have low liquid fuel permeability. Although the Nafion ^™ membrane has been found to be an effective proton-conducting solid polymer electrolyte membrane, perfluorosulfonic acid polymer membranes such as the Aciplex ^™ membrane (manufactured by Asahi Glass Co., Japan) and those produced by Dow Chemical Co., USA Polymer membranes such as XUS 13204.10 which have similar properties to Nafion ^™ are also suitable. Depending on the temperature and operating durability of the fuel cell, membranes of polyethylene and polypropylene sulfonic acid, polystyrene sulfonic acid and other polyhydrocarbyl sulfonic acids (such as those produced by RAI Corporation, USA) can also be used. Different acid equivalent weights or different chemical compositions (such as modified acid groups or polymer backbones) or different water contents or different types and degrees of crosslinking (such as through multivalent cations such as Al ³⁺ , Mg ²⁺ ) can be used. A composite membrane composed of two or more types of proton-conducting cation exchange polymers such as cross-linking) to obtain low fuel permeability. Such composite membranes can be fabricated to achieve high ionic conductivity, low liquid fuel permeability, and good electrochemical stability.

As understood from the above description, a proton-conducting solid polymer membrane is used as the electrolyte without the need for free soluble acid or base electrolytes to obtain a liquid feed direct oxidation organic fuel cell. The only electrolyte is a proton conducting solid polymer membrane. There are no acids present in free form in liquid fuel and water mixtures. By not using free acid, acid-induced corrosion of cell components, which can occur in prior art acid-based organic/air fuel cells, is avoided. This provides considerable flexibility in the choice of materials for the fuel cell and associated subsystems. Further, unlike a fuel cell containing potassium hydroxide as a liquid electrolyte, since soluble carbonate cannot be formed, the performance of the cell will not be degraded. By applying a solid electrolyte membrane, parasitic bypass currents can also be avoided.

Referring now to FIG. 2, a fuel cell system using a set of fuel cells like that shown in FIG. 1 will be described. The fuel cell system includes a fuel cell stack 25, each fuel cell having the membrane/electrode assembly described above for FIG. 1 . Oxygen or air is supplied via an oxidant supply 26 which may be, for example, a bottled oxygen supply, an air blower fan or an air compressor. A mixture of air and water or oxygen and water is exhausted from the battery pack 25 through the outlet nozzle 30 and sent to the water recovery unit 27 . The water recovery unit 27 operates to separate air or oxygen from water. A portion of the air or oxygen separated by means 27 returns to the oxidant supplier 26 and re-enters the battery pack 25 . Fresh air or oxygen is added to the supplier 27 . Water separated by unit 27 is fed to fuel and water injection unit 29 which may also receive organic fuel such as methanol from storage tank 33 . The injection unit 29 mixes the water from the recovery unit 27 with the organic fuel from the tank 33 to obtain a fuel and water solution in which the fuel is dissolved in water.

The fuel and aqueous solution supplied by the injection device 29 are fed into the circulation tank 35 . The fuel and water mixture containing carbon dioxide is discharged from the stack 25 through the outlet 23 and sent to the circulation tank 35 through the heat exchanger 37 . The circulation tank 35 thus receives the fuel and water solution from the injection device 29 and the fuel and water solution containing carbon dioxide gas from the heat exchanger 37 . Circulation tank 35 extracts carbon dioxide from the fuel and water mixture and releases the carbon dioxide through outlet 39 . The resulting fuel and water solution is fed by pump 20 into battery pack 25 . A circulation tank 35 may also be placed between the battery pack 25 and the heat exchanger 34 to remove carbon dioxide prior to the heat exchanger and thus improve the performance of the heat exchanger.

The operation of the different elements represented in Figure 2 will now be described in more detail. The circulation tank 35 is a column having a large space above the liquid level. A mixture of liquid fuel and water received from injection unit 29 is fed to the top of the column. A fuel and water mixture containing carbon dioxide is fed into the bottom part of the tower. Carbon dioxide gas evolved from the fuel and water mixture accumulates in the headspace and is eventually vented. Alternatively, the carbon dioxide containing fuel and water mixture can be passed through a set of tubes of a microporous material such as Celgard ^™ or GoreTex ^™ which allows the gas to be released through the tube walls of the microporous material. The liquid fuel flows along the tube axis. Celgard ^™ and GoreTex ^™ are registered trademarks of Celanese Corp. and Gore Association, USA.

A fixed recirculation system (not shown) can be used in the anode compartment of the stack 25 to separate carbon dioxide from the fuel and water mixture, thus requiring no external recirculation tanks. Using such a system, carbon dioxide bubbles can rise vertically in the anode chamber due to inherent buoyancy. The interaction with the viscosity of the liquid fuel mixture surrounding the gas bubbles pulls the liquid fuel upward toward the outlet 23 . Once out of the anode compartment, the liquid outgasses, exchanges heat with the surrounding environment and cools, thus becoming denser than the liquid in the cell. Through an inlet, the denser liquid is fed into the bottom of the anode chamber. Instead of consuming electrical energy at the pump, the stationary recirculation system utilizes the heat and gas generated in the battery. The foregoing process forms the basis of a stationary recirculation system which will not be described in further detail. It should be noted that the application of stationary recirculation systems may limit the orientation of fuel cell operation and may only be feasible for stationary applications. Test results of fuel cells with Nafion ^TM electrolyte membrane

The electrooxidation kinetics of methanol have been studied by measurements of galvanostatic polarization in electrochemical cells (not illustrated but similar to the electrodeposition cell illustrated in Figure 10 below) for sulfuric acid electrolytes and Nafion ^™ electrolytes. The cell consists of a working electrode, a platinum counter electrode, and a reference electrode. The working electrode is polarized in a solution containing the electrolyte of choice and a liquid fuel. The potential of the working electrode is detected relative to the reference electrode.

Figure 3 shows polarization curves, i.e., polarization versus current density in milliamperes/square centimeter (mA/cm ² ), for the kinetics of methanol oxidation in Nafion ^™ and sulfuric acid electrolytes, represented by curve 41 The polarization for the 0.5M sulfuric acid electrolyte and the polarization for the Nafion ^(TM) electrolyte are represented by curve 43. Polarization is expressed as potential versus NHE, where NHE stands for standard hydrogen electrode. The curves represent measured data for a cell consisting of a 1 M mixture of methanol in water at 60°C. As shown in Figure 3, there was less loss of polarization when the electrodes were in contact with Nafion ^™ than with sulfuric acid. Thus, it can be concluded that the electro-oxidation kinetics of methanol are easier when the electrolyte is Nafion ^™ . These findings can be explained by the fact that strong adsorption of sulfate ions occurs at the electrode/sulfuric acid interface under positive potential conditions that hinder the kinetics of electrooxidation. When ^NafionTM is used as the electrolyte, since such ions are not generated, such adsorption cannot occur. At the same time, it is believed that the electroreduction kinetics of oxygen or air are enhanced at the electrode/Nafion ^™ interface compared to the electrode/sulfuric acid interface. This latter effect may be due to the higher solubility of oxygen in Nafion ^™ and the lack of strongly adsorbed anions. Therefore, the application of proton-conducting solid polymer membranes as electrolytes is beneficial to the two-electrode reaction kinetics and can overcome the disadvantages of sulfuric acid electrolytes.

Meanwhile, the sulfuric acid electrolyte degrades at temperatures above 80 °C. Degradation products can degrade the performance of individual electrodes. The electrochemical and thermal stability of solid polymer electrolytes such as Nafion™ is quite high compared to that of sulfuric acid, and solid polymer electrolytes can be used at temperatures up to 120°C. Therefore, the application of proton-conducting solid polymer membranes allows fuel cells to operate at temperatures up to 120 °C for long periods of time, since the kinetics of electro-oxidation of fuel and electro-reduction of oxygen occur more easily when the temperature rises, which has additional advantages. The advantages.

Figure 4 shows the performance of the fuel cell shown in Figure 2 operating at 65°C for the methanol/oxygen combination and the methanol/air combination. In FIG. 4 , the voltage of the fuel cell is indicated along axis 32 and the current density in mA/cm ² is indicated along axis 34 . Curve 36 represents the performance of the methanol/oxygen combination and curve 38 represents the performance of the methanol/air combination. As can be seen, application of pure oxygen provides slightly better performance than air.

Figure 5 shows the effect of fuel concentration on cell performance. The potential of the fuel cell is represented along axis 40 , while the current density in mA/cm ² is represented along axis 42 . Curve 44 represents the performance of the cell for a 2.0M methanol solution at 150 degrees Fahrenheit (F). Curve 46 represents the performance of the cell for a 0.5M methanol mixture at 140°F. Curve 48 represents the performance of the cell for a 4.0M methanol mixture at 160 degrees Fahrenheit. As can be seen, the 2.0M methanol mixture provided the best overall performance. Also, Figure 5 shows that while the fuel cell maintains a suitably high voltage, it is possible to maintain a current density as high as 300 mA/cm ² . In particular, the 2.0M methanol mixture provides voltages greater than 0.4 volts at approximately 300mA/ ^cm2 . The performance shown in Figure 5 shows a vast improvement over previous organic fuel cell performance.

The polarization behavior of the anode and cathode of a fuel cell as a function of current density in mA/cm ² is shown in FIG. 6 , with voltage along axis 50 and current density along axis 52 . Curve 54 shows the polarization behavior of the 2.0M mixture at 150°F. Curve 56 represents the polarization behavior for fuel and curve 58 represents the polarization behavior for oxygen at 150 degrees Fahrenheit. Anode structures for liquid-fed fuel cells

Anode structures for liquid-fed fuel cells must be very different from conventional fuel cells. Conventional fuel cells use a gas diffusion type electrode structure to provide a balance of gases, liquids and solids. However, a liquid-fed fuel cell requires an anode structure similar to that of a battery. The anode structure must be porous and must be wettable by the liquid fuel. In addition, the structure must be electronically and ionically conductive to efficiently transport electrons to the anode current collector (carbon paper) and hydrogen/hydronium ions to the Nafion ^™ electrolyte membrane. Further, the anode structure must facilitate favorable gas evolution characteristics at the anode.

The electrodes required for liquid-fed fuel cells can be specially fabricated, or conventional commercially available gas diffusion electrodes can be modified with appropriate additives. Electrode impregnation with ionomer additives

The electrocatalyst layer and carbon fiber support of anode 14 (FIG. 1) are preferably impregnated with a hydrophilic proton-conducting polymer additive such as Nafion ^™ . Additives are provided to the anode in part to enable efficient transport of protons and hydronium ions generated by the electro-oxidation reaction. The ionomer additive can also improve the uniformity of wetting of the electrode pores by the liquid fuel/water solution and provide better utilization of the electrocatalyst. The methanol electrooxidation kinetics are improved by reducing the adsorption of anions. Further, the use of ionomer additives helps to achieve favorable gas evolution characteristics at the anode.

For an effective anode additive, the additive should be hydrophilic, proton conductive, electrochemically stable, and not hinder the oxidation kinetics of liquid fuels. Nafion ^™ meets these criteria and is the preferred anode additive. Other hydrophilic proton-conducting additives that have the same effect as Nafion ^™ are montmorillonites, zeolites, alkoxycelluloses, cyclodextrins, and zirconium hydrogen phosphate.

Figure 7 is a block diagram showing the steps involved in anode impregnation with an ionomeric additive such as Nafion( ^TM) . First, a carbon electrode structure is obtained or prepared. A commercially available large surface carbon electrode structure using a mixture of large surface electrocatalyst and Teflon ^™ binder coated on Toray ^™ carbon fiber paper can be used. An electrocatalytic electrode may also use an emulsion of polytetrafluoroethylene, TFE-30 ^™ , prepared from large surface catalyst particles and Toray ^™ paper, both commercially available from E-Tek, Inc. While these structures can be fabricated from the above constituent materials, prefabricated structures of any desired size are also available directly from E-Tek.

In step 302, by immersing the electrocatalyst particles in a solution containing 0.5-5% ionomer additive (with methanol or isopropanol, by diluting as appropriate from a solution provided by Aldrich Chemical Co., or by Solution Technologies Inc.) In solution, the electrodes are impregnated with an ionomer additive such as Nafion( ^TM) for 5-10 minutes. Then in step 304, the electrode is removed from the solution and dried in air or vacuum at 20-60°C to volatilize any residue of higher alcohols associated with the Teflon ^™ solution. The impregnation steps 302-304 are repeated until the desired composition (2-10% by weight electrocatalyst) is obtained. A typical loading amount is 0.1 to 0.5 mg/cm ² . Electrode compositions with more than 10% of additives can lead to increased internal resistance of the fuel cell and poor adhesion to the solid polymer electrolyte membrane. Compositions with less than 2% additives typically do not result in improved electrode performance.

To form impregnated electrodes from electrocatalyst particles, the electrocatalyst particles were mixed with a Nafion ^™ solution diluted to 1% with isopropanol. Then, the solvent was evaporated until a thick mixture was obtained. Next, the thick mixture was spread onto Toray ^™ paper to form a thin layer of electrocatalyst. Typically a mixture of approximately 200 ^m2 /g large surface particles coated on Toray ^™ paper. It should be noted here that the thus formed electrocatalyst layer contains only Nafion ^™ and no Teflon ^™ . The electrodes thus prepared were then dried under vacuum at 60° C. for 1 hour to remove higher alcohol residues, after which they were ready for use in liquid-fed cells.

A commercially available high surface area platinum-tin electrode was impregnated with Nafion( ^TM) according to the method described above. Figure 8 compares the performance of a Nafion ^™ impregnated electrode with that of a non-impregnated electrode, measured in a half-cell similar to that of Figure 10 (below) but containing a sulfuric acid electrolyte. In particular, Figure 8 shows the measurement of polarization in liquid formaldehyde fuel (1M) with sulfuric acid electrolyte (0.5M). Current density in mA/cm ² is represented along axis 306 and voltage in volts is represented along axis 308 . Curve 310 is the galvanostatic polarization curve for a platinum-tin electrode excluding Nafion ^™ . Curve 312 is the galvanostatic polarization curve of a platinum-tin electrode not impregnated with Nafion ^™ .

It can be seen from Figure 8 that much higher current densities were obtained with Nafion ^™ impregnated electrodes than with non-impregnated electrodes. In fact, with non-impregnated electrodes, little oxidation of formaldehyde occurs. Thus, the addition of Nafion ^™ provides a huge improvement. Furthermore, the absence of any hysteresis in the galvanostatic polarization curves suggests that these coatings are stable.

Thus far, improved liquid-fed fuel cell anodes impregnated with ionomer additives have been described. Anode fabrication methods including ionomer additives have also been described. The remainder of the detailed description is the use of PFOS as an additive in electrodeposition baths used to make electrodes and as a direct additive in fuels. New fuels will also be described. Electrodeposition of Electrode Using PFOS Additive

Referring to FIGS. 9-11 , an electrode fabrication method for an organic fuel cell will now be described in detail. This method is advantageously used in the manufacture of cathodes for liquid organic fuel cells as described above. However, electrodes prepared by the methods of FIGS. 9-11 can alternatively be used in a variety of organic fuel cells.

Referring first to Figure 9, the steps of the anode manufacturing method will now be described. First, at 200, a carbon electrode structure is prepared by coating a mixture of large surface carbon particles and a Teflon ^™ binder onto fiber-based carbon paper. Preferably, the carbon particles have a surface area of 200 meters ² /gram (m ² /g). A suitable carbon particle matrix known as Vulan XC-72 is commercially available from E-Tek Inc. Teflon ^™ binder is preferably added to obtain a percentage by weight of 15%. The fiber-based carbon paper is preferably Toray ^(TM) paper, also available from E-Tek Incorporated. A carbon structure can be prepared from the above constituent materials. Alternatively, 2 inch by 2 inch commercial prefabricated structures are available directly from E-Tek Inc.

In step 202, an electrodeposition bath is prepared by dissolving chloroplatinic acid (IV) and potassium (III) pentachlororuthenate hydrate in sulfuric acid. Preferably, the resulting metal ion concentration is 0.01-0.05M. Meanwhile, preferably, the concentration of sulfuric acid is 1M. The above compounds were used to obtain platinum-ruthenium deposition on carbon electrode structures. Alternative solutions can be used. For example, to obtain platinum-tin deposits, tin tetrachloride compounds are alternatively dissolved in sulfuric acid.

The metal ion salt is dissolved in sulfuric acid, mainly to prevent the hydrolysis of the solution. For ruthenium deposition, the resulting solution is preferably degassed to prevent the formation of high oxidation states.

In step 204, high purity PFOS (C-8 acid) is added to the bath. The C-8 acid is preferably added to a concentration of 0.1-1.0 g/l. The addition of C-8 acid facilitates complete wetting of the carbon particles. C-8 acids are not electroactive and do not specifically adsorb at metal sites in the structure. Therefore, the C-8 acid is not harmful for the subsequent electrodeposition process. The addition of C-8 acid has been found to be quite beneficial and may be necessary for successful electrodeposition on electrodes.

At 206 , the carbon electrode structure obtained from step 200 is placed in the electrodeposition bath obtained from step 204 . A platinum anode is also placed in the bath. For deposition of other metal ions, alternative anode materials can be used.

Then in step 208, a voltage is applied between the carbon electrode structure and the platinum anode. The voltage was maintained for about 5-10 minutes to obtain platinum-ruthenium electrodeposition with a loading of about 5 mg/cm ² on the carbon electrode. Preferably, a voltage close to -0.8 V is applied relative to the mercury sulphate reference electrode.

After the desired amount of metal is deposited on the carbon electrode, in step 210, the electrode is removed and washed with deionized water. Preferably, the electrodes are washed at least three times in circulating deionized water for 15 minutes each. The washing step is mainly to remove chlorine and sulfate ions adsorbed on the carbon electrode surface. A washing step has been found to be quite desirable and may be necessary to obtain effective electrodes for use in organic fuel cells.

It has been found that the electrodes resulting from the fabrication method in step 206 have very uniform "cotton ball" shaped particles with a considerable amount of fine structure. The average particle size has been found to be about 0.1 micron.

The deposition apparatus used to implement the dipping method of FIG. 9 is represented in FIG. 10 . Specifically, FIG. 10 shows a three-electrode cell 212 that includes a single carbon structural electrode 214 , a pair of platinum counter electrodes (or anodes) 216 and a reference electrode 218 . All electrodes were placed in bath 220 formed from the metal/C-8 acid solution mentioned above. Electrical contacts 222 and 224 are placed on the inner side of battery 212 , above bath 220 . A magnetic stirrer 226 is placed in the bath 220 to facilitate agitation and circulation of the bath. A circulating water jacket is provided around the cell 212 for regulating the temperature within the cell. The platinum anode is placed in a finely divided glass frit 230, the glass frit being provided to separate the anode from the cathode to prevent diffusion of anodic oxidation products into the cathode.

Reference electrode 214 is a mercury/mercuric sulfate reference electrode. A reference electrode is provided to monitor and control the potential of the carbon electrode structure 214 . Preferably constant voltage and constant current control methods are used. The composition of the alloy deposit is controlled by selection of the bath composition outlined above and by conducting electrode deposition at a current density well above any limiting metal deposition. When selecting an appropriate bath composition, it is important to regulate the electrochemical equivalent weight of the metals in the composition.

During operation, the amount of charge passing from the anode to the cathode is measured and used to detect the amount of deposited material. In this case, the amount of charge used for any hydrogen-evolving reactions must be subtracted from the total charge for each measurement.

Satisfactory electrodeposition typically occurs within 5 to 10 minutes, depending on operating conditions and the ideal catalyst loading.

The detection equipment used to detect and control the electrode potentials is not shown in Figure 10, since the function and operation of such devices will be familiar to those skilled in the art.

FIG. 11 shows typical electrode performance deposited by the method of FIG. 9 in the electrodeposition cell of FIG. 7. FIG. In FIG. 11 , the potential in voltage relative to the NHE is represented along axis 240 , while the current density in mA/cm ² is represented along axis 242 . Curve 246 represents the galvanostatic polarization curve of the carbon-supported platinum-ruthenium alloy electrode prepared according to the previous 5 mg/cm ² loading. Curve 246 represents the galvanostatic polarization of an electrode with a loading of 1 mg/cm ² . In each case, the electrodes are used in the sulfuric acid electrolyte of the half-cell. The fuel cell comprised an organic fuel consisting of 1M methanol and 0.5M sulfuric acid, which operated at 60°C. With a loading of 5 mg/ ^cm2 , the electrode can maintain a continuous current density of 100 mA/ ^cm2 at a voltage of 0.45 V vs. NHE.

The results presented in FIG. 11 are typical performances that can be obtained using electrodes fabricated according to the method of FIG. 9 . Performance can be further enhanced by applying appropriate optimum electrodeposition conditions and alloy compositions. Thus, the specific conditions and concentrations described above are not necessarily optimal, but merely represent the best way known today to make electrodes. PFOS as a fuel additive (C-8 acid)

The use of C-8 acids as additives in electrodeposition baths has been described above. It has also been established that C-8 acids can be beneficially used as additives in fuels for liquid-fed fuel cells employing sulfuric acid electrolytes. In particular, straight carbon chain C _{- 8} acids having the molecular formula _C8F17SO3H at a concentration of 0.001-0.1 M have been found to be excellent wetting agents in liquid-fed fuel cells.

Figure 12 shows the comparative experimental results of fuel cells using C-8 acid as additive and without additive. In particular, Figure 12 shows the results of half-cell experiments using the device's Teflon ^™ -coated large-surface carbon-supported platinum and platinum alloy electrodes in a sulfuric acid electrolyte. The results were obtained using a half-cell similar to the cell shown in FIG. 10 . FIG. 12 shows the potential relative to NHE along the vertical axis 400 and the current density in mA/cm ² along the horizontal axis 402 . Four curves are provided representing polarization for fuels with no additive (curve 404 ), with 0.0001 M additive (curve 406 ), with 0.001 M additive (curve 408 ), and with 0.01 M additive (curve 412 ).

It can be seen from Figure 12 that the addition of the C-8 additive greatly reduces the polarization. Although not shown, the oxidation of methanol was studied using a 0.1M solution of pure C-8 acid without any sulfuric acid. Polarization curves (not shown) show that the kinetics are not affected by the presence of PFOS ions.

Thus, Figure 12 shows that using a C8 acid at a concentration of 0.001 M or higher as an additive is beneficial for liquid fuel solutions when using commercially available Teflon ^™ coated fuel cell electrodes, at least for those using sulfuric acid as the electrolyte. Fuel cells are beneficial.

Referring to the remaining figures, three new fuels for liquid-fed fuel cells are described. These fuels are dimethoxymethane, trimethoxymethane and trioxane. Dimethoxymethane as fuel in liquid-fed fuel cells

Figures 13-15 show experimental results for organic direct liquid fed fuel cells using dimethoxymethane (DMM) as fuel. In use, DMM is mixed with water to a concentration of about 0.1-2M and fed to the fuel cell. Other concentrations are also effective. The fuel cell may be of conventional design or may include one or more of the modifications described above. In a fuel cell, DMM is electrooxidized on the anode of the cell. The electrooxidation of DMM involves a series of dissociation steps followed by surface reactions to form carbon dioxide and water. The electrochemical reaction is given by:

(4)

The electro-oxidation experiments for the determination of DMM were completed in a half-cell similar to the cell shown in Figure 10 using a 0.5 M sulfuric acid electrolyte and a Pt-Sn or Pt-Ru electrocatalyst electrode under temperature control. The galvanostatic polarization curves shown in Figure 13 represent the electro-oxidative properties of DMM for several different fuel concentrations using platinum-tin electrodes. The platinum-tin electrode was of the gas diffusion type consisting of 0.5 mg/ ^cm2 total metal supported on Vulcan XC-72 obtained from Etek, Inc., Framingham, MA. In FIG. 13 , current density is represented along axis 500 and polarization (in terms of potential versus NHE) is represented along axis 502 . Curves 504, 506, 508, and 510 represent polarization for DMM concentrations of 0.1M, 0.5M, 1M, and 2M, respectively. Figure 13 shows that increasing the concentration improves the kinetics of DMM oxidation. The graph of Figure 13 was measured in a half cell using 0.5M sulfuric acid as electrolyte and 0.1M C-8 acid. Measurements are performed at room temperature.

DMM has been found to oxidize at a much negative potential than methanol. At the same time, it has been found that temperature has a great influence on the oxidation rate. However, DMM has a low boiling point of 41 °C. Difficulties may therefore arise in attempting to use DMM in liquid-fed fuel cells at temperatures above boiling point.

Figure 14 shows the polarization for two different concentrations at two different temperatures. Current density is represented along axis 512 and polarization (in terms of potential versus NHE) is represented along axis 514 . Curve 516 represents the polarization for a 1M concentration of DMM at room temperature. Curve 518 represents the polarization for a 2M concentration of DMM at 55°C. As can be seen, improved polarization can be obtained using higher concentrations at higher temperatures. Also, a comparison of curve 510 in FIG. 13 with curve 518 in FIG. 14 shows that for the same concentration level, an increase in temperature results in improved polarization. Therefore, it can be said that an increase in temperature leads to improved electrooxidation kinetics.

In addition to the half-cell experiments shown in Figures 13 and 14, fuel cell experiments can also verify the effectiveness of DMM in fuel cells. In fuel cells, direct oxidation of DMM is performed in liquid fed fuel cells as represented in Figures 1 and 2 above. Thus, the fuel cell uses a proton-conducting solid polymer membrane (Nafion ^™ 117) as the electrolyte. The membrane electrode assembly consisted of a fuel oxidation electrode made of a non-supported platinum-ruthenium catalyst layer (4 mg/cm ² ) and a gas diffusion type non-supported platinum electrode (4 mg/cm ² ) for oxygen reduction. On the fuel oxidation side, the fuel cell uses 1M DMM solution and 20psi oxygen on the cathode.

Analysis of the DMM oxidation products showed only methanol. Methanol is considered a possible intermediate in the oxidation of DMM to carbon dioxide and water. However, since the fuel cell system is compatible with methanol, the presence of methanol as an intermediate product is not important since methanol is also eventually oxidized to carbon dioxide and water.

The current-voltage characteristics of a direct oxidation fuel cell using DMM as fuel with a liquid feed are shown in FIG. 15 . The fuel cell operates at 37°C. In FIG. 15 , current density in mA/cm ² is represented along axis 520 . Battery voltage in volts is represented along axis 522 . Curve 524 represents the cell voltage as a function of current density for the 1M DMM solution described above. As can be seen from Figure 15, using DMM, the cell voltage reaches 0.25V at 50mA/ ^cm2 , which is as high as the cell voltage reached when using methanol. Even better performance can be obtained by operating at higher temperatures and applying a Pt-Sn catalyst. The low boiling point of DMM also makes it a candidate for a gas feed type operation.

Thus, it has been found from half-cell and full-cell measurements that DMM can be oxidized at a very high rate. Therefore, it can be said that DMM is an excellent fuel for direct oxidation fuel cells. At the same time, DMM is non-toxic, low vapor pressure liquid, and easy to handle. Furthermore, DMM can be synthesized from natural gas (methane) by conventional techniques. Trimethoxymethane as fuel in liquid-fed fuel cells

Figures 16-18 show the results of experiments using trimethoxymethane (TMM) as fuel for organic direct liquid fed fuel cells. As with DMM described above, in use, TMM is mixed with water to a concentration of about 0.1-2M and fed to the fuel cell. Other concentrations are also effective. The fuel cell may be of conventional design or may include one or more of the modifications described above. In a fuel cell, TMM is electrooxidized on the anode of the cell. The electrochemical oxidation of TMM is represented by the following reaction:

(5)

Experiments demonstrating TMM electro-oxidation were carried out in a half-cell similar to the cell shown in Fig. 10 by controlling the temperature, applying a 0.5 M sulfuric acid electrolyte including 0.01 MC-8 acid, and a Pt-Sn electrode. The results of these half-cell experiments are shown in FIGS. 16 and 17 .

Figure 16 provides the galvanostatic polarization curves for several different TMM concentrations for the Pt-Sn electrodes described above. The Pt-Sn electrodes were gas diffusion type and consisted of 0.5 mg/ ^cm2 total metal supported on Vulcan XC-72 obtained from Etek, Inc., Framingham, MA. In FIG. 16 , current density in mA/cm ² is represented along axis 600 and polarization (in terms of potential versus NHE) is represented along axis 602 . Curves 604, 606, 608, and 610 represent polarization for TMM concentrations of 0.1M, 0.5M, 1M, and 2M, respectively. Figure 16 shows the improved polarization obtained at higher concentrations. All measurements shown in Figure 16 were taken at room temperature.

It has been found that TMM is capable of oxidation at a much negative potential than methanol. Meanwhile, it has been found that temperature has an effect on the oxidation rate of TMM. Figure 17 shows the polarization at two different concentrations and at two different temperatures. In FIG. 17 , current density in mA/cm ² is represented along axis 612 and polarization (potential versus NHE) is represented along axis 614 . Curve 616 represents the polarization for a 1M concentration of TMM at room temperature and curve 618 represents the polarization for a 2M concentration of TMM at 55°C. The curve in Figure 17 was obtained using a Pt-Sn electrode in a 0.5M sulfuric acid electrolyte including 0.01MC-8 acid. As can be seen, the use of higher concentrations results in improved polarization at higher temperatures. Also, a comparison of curve 618 in FIG. 17 with curve 610 in FIG. 16 illustrates that, for the same concentration level, an increase in temperature results in an improvement in performance. Although not shown, it has been found that at temperatures as high as 60°C, the oxidation rate of TMM is twice that at 25°C.

In addition to the half cell experiments represented in Figures 16 and 17, full fuel cell experiments were also performed to demonstrate the effectiveness of TMM in fuel cells. The direct oxidation of TMM in the fuel cell was carried out in a liquid-fed fuel cell as represented in FIGS. 1 and 2 above. Thus, the fuel cell employs a proton-conducting solid polymer membrane (Nafion ^™ 117) as the electrolyte. The membrane electrode part of the fuel cell includes a non-supported platinum-ruthenium catalyst layer (4 mg/cm ² ) and an oxygen-reducing gas diffusion type non-supported platinum electrode (4 mg/cm ² ). On the fuel oxidation side, the fuel cell uses a 2M TMM solution and 20 psi of oxygen on the cathode.

When TMM was used, analysis of TMM oxidation products showed only methanol, and methanol was considered as a possible intermediate product in the oxidation of TMM to carbon dioxide and water. For fuel cells that are compatible with methanol, the presence of methanol as an intermediate is not a critical matter because methanol is also eventually oxidized to carbon dioxide and water.

The current-voltage characteristics of the liquid-fed direct oxidation fuel cell described above are shown in FIG. 18 for TMM and methanol. Current density in mA/cm ² is represented along axis 620 , while cell voltage is represented along axis 622 . Curve 624 represents the cell voltage as a function of current density using a 1M concentration of TMM. Curve 626 represents the cell voltage as a function of current density using methanol at a concentration of 1M. At 65°C, the measurement results shown in Fig. 18 were obtained. Although not shown, at 90°C, using TMM, the cell voltage can reach 0.52V at 300mA/ ^cm2 , which is higher than that obtained using methanol.

Thus, it has been found from half-cell and full-cell measurements that TMM, like DMM, can be oxidized at a very high rate. Also like DMM, TMM is a non-toxic, low vapor pressure liquid that is easy to handle and can be synthesized from natural gas (methane) by conventional methods. Trioxane as fuel in liquid-fed fuel cells

Figures 19-21 show the experimental results of trioxane used as fuel in organic direct liquid fed fuel cells. As with DMM and TMM described above, in use, trioxane is mixed with water to a concentration of about 0.1-2M and fed to the fuel cell. Other concentrations are also effective. The fuel cell may be of conventional design or may include one or more of the modifications described above. In a fuel cell, trioxane is electrooxidized at the anode of the cell. The electrochemical oxidation of trioxane is represented by the following reaction:

(6)

Experiments demonstrating electro-oxidation of trioxane were carried out in a half-cell similar to the cell shown in Fig. 10 by controlling temperature, applying 0.5M-2.0M sulfuric acid electrolyte including 0.01MC-8 acid and Pt-Sn electrodes. The results of these half-cell experiments are shown in FIGS. 19 and 20 .

Figure 19 provides the galvanostatic polarization curves for several different trioxane concentrations using the Pt-Sn electrode described above. The Pt-Sn electrodes were gas diffusion type and consisted of 0.5 mg/ ^cm2 total noble metal supported on Vulcan XC-72 obtained from Etek, Inc., Framingham, MA. In FIG. 19 , current density in mA/cm ² is represented along axis 700 and polarization (in terms of potential versus NHE) is represented along axis 702 . Curves 704, 706, 708, and 710 represent the polarization for trioxane concentrations of 0.1M, 0.5M, 1M, and 2M, respectively. Figure 19 shows the improvement in polarization obtained at higher concentrations. All measurements shown in Figure 19 were obtained at 55°C.

Thus, for trioxane, increasing the fuel concentration leads to an increase in the oxidation rate. Meanwhile, as seen from FIG. 19 , a current density of 100 mA/cm ² was obtained at a potential of 0.4 V versus NHE. This performance is comparable to that of the application of formaldehyde. Although not shown, cyclic voltammetry studies have determined that the oxidation mechanism of trioxane does not involve cleavage to formaldehyde prior to electrooxidation.

It was also found that an increase in the electrolyte acid concentration also resulted in an increase in the electro-oxidation rate. Figure 20 shows the polarization at four different electrolyte concentrations and at two different temperatures. In FIG. 20 , current density in mA/cm ² is represented along axis 712 and polarization (potential versus NHE) is represented along axis 714 . Curve 716 represents the polarization for a 0.5M electrolyte concentration at room temperature. Curve 718 represents the polarization for 0.5M electrolyte concentration at 65°C. Curve 720 represents polarization for 1M electrolyte concentration at 65°C. Finally, curve 722 represents the polarization effect for 2M electrolyte concentration at 65°C. For all curves 716-722, the concentration of trioxane was 2M.

The curve in Figure 20 was obtained using a Pt-Sn electrode in a sulfuric acid electrolyte containing 0.01 MC-8 acid. As can be seen, improved polarization can be obtained using higher concentrations of electrolyte at higher temperatures. Since the acidity exhibited by Nafion ^TM is equivalent to 10M sulfuric acid, it is planned to use Nafion ^TM as an electrolyte in order to obtain a high electro-oxidation rate.

In addition to the half cell experiments shown in Figures 19 and 20, full fuel cell experiments demonstrating the effectiveness of trioxane were also performed in fuel cells. The direct oxidation of trioxane in the fuel cell was carried out in a liquid fed fuel cell as represented in Figures 1 and 2 above. Thus, the fuel cell employs a proton-conducting solid polymer membrane (Nafion ^™ 117) as the electrolyte. On the fuel oxidation side, the fuel cell uses 1M trioxane solution and 20psi oxygen on the cathode.

As with DMM and TMM, analysis of the trioxane oxidation products revealed only methanol, and methanol was considered a possible intermediate in the oxidation of trioxane to carbon dioxide and water. For fuel cells that are compatible with methanol, the presence of methanol as an intermediate is not a critical matter because methanol is also eventually oxidized to carbon dioxide and water.

The current-voltage characteristics of the liquid-fed direct oxidation fuel cell described above are shown in FIG. 21 for trioxane. Current density in mA/cm ² is represented along axis 724 , while cell voltage is represented along axis 726 . Curve 728 represents the cell voltage as a function of current density using a 1M concentration of trioxane. The measurement results shown in Fig. 21 were obtained at 60°C. The performance shown in Figure 21 can be considerably improved by using a platinum-tin electrode instead of a Pt-Ru electrode.

In the trioxane/oxygen fuel cell, crossover measurements (not shown) showed that the crossover rate was at least 5 times lower than in the methanol fuel cell. A reduction in the crossover rate is particularly desirable since crossover affects the efficiency and performance of the fuel cell, as previously stated.

Thus, it has been found from half-cell and full-cell measurements that trioxane, like DMM and TMM, can be oxidized at a very high rate. in conclusion

Described are a series of improvements to liquid-fed fuel cells including improved electrolyte and electrode structures, improved methods of making electrodes, additives to improve fuel performance, and a set of three new fuels. Different improvements can be applied separately or most can be combined to obtain even more enhanced performance. However, it should be noted that the use of C-8 acids as additives in fuels described above is only expected to be effective for fuel cells using an acid electrolyte such as sulfuric acid, and may not be effective if using fuel cells containing proton exchange membranes.

The methods, embodiments and experimental results shown here are only illustrative and exemplary of the invention, and should not be considered as limiting the scope of the invention.

Claims (53)

1. In a liquid-fed direct fuel cell having an anode, a cathode, an electrolyte, means for circulating an organic fuel through the anode, and means for circulating oxygen through the cathode, the improvement comprising: using a solid polymer electrolyte membrane as the electrolyte; and An organic fuel that is substantially free of acid electrolytes is provided. 2. The improvement of claim 1 wherein said membrane is a solid proton exchange membrane. 3. The improvement of claim 2 wherein said membrane is comprised of Nafion ^(TM) . 4. The improvement of claim 2 wherein said membrane is comprised of a modified perfluorinated sulfonic acid polymer. 5. The improvement of claim 2 wherein said membrane is comprised of a polyhydrocarbyl sulfonic acid polymer. 6. The improvement of claim 2 wherein said membrane is comprised of a composite of two or more proton exchange membranes. 7. The improvement of claim 1 wherein said organic fuel is selected from the group consisting of methanol, formaldehyde and formic acid. 8. The improvement of claim 1 wherein said organic fuel is selected from the group consisting of dimethoxymethane, trimethoxymethane and trioxane. 9. The improvement of claim 1 wherein said anode is a commercial electrode impregnated with a hydrophilic, proton conductive, water insoluble ionomer. 10. The improvement of claim 9 wherein said ionomer is Nafion ^(TM) . 11. The improvement of claim 9 wherein said ionomer is montmorillonite. 12. The improvement of claim 9 wherein said ionomer is an alkoxy cellulose. 13. The improvement of claim 9 wherein said ionomer is a cyclodextrin. 14. The improvement of claim 9 wherein said ionomer is a zeolite mixture. 15. The fuel cell of claim 9, wherein said ionomer is zirconium hydrogen phosphate. 16. A liquid-fed fuel cell comprising: an anode; a cathode; a solid polymer electrolyte membrane positioned between said anode and said cathode; means for circulating a solution of liquid organic fuel and water through said anode, said solution being substantially free of sulfuric acid; and means for circulating oxygen through said cathode. 17. A liquid-fed fuel cell comprising: an anode impregnated with a hydrophilic, proton-conducting, water-insoluble ionomer; a cathode; a polymer electrolyte membrane positioned between said anode and said cathode; means for circulating a solution of liquid organic fuel and water through said anode; and means for circulating oxygen through said cathode. 18. The fuel cell of claim 17, wherein said membrane is a solid proton exchange membrane. 19. The fuel cell of claim 18, wherein said membrane is comprised of Nafion. 20. The fuel cell of claim 18, wherein said membrane is comprised of a modified perfluorinated sulfonic acid polymer. 21. The fuel cell of claim 18, wherein said membrane is comprised of a polyhydrocarbylsulfonic acid polymer. 22. The fuel cell of claim 18, wherein said membrane is composed of two or more proton exchange membrane composite materials. 23. The fuel cell of claim 17, wherein said organic fuel is selected from the group consisting of methanol, formaldehyde and formic acid. 24. The fuel cell of claim 17, wherein said organic fuel is selected from the group consisting of dimethoxymethane, trimethoxymethane and trioxane. 25. The fuel cell of claim 17, wherein said ionomer is Nafion ^(TM) . 26. The fuel cell of claim 17, wherein said ionomer is montmorillonite. 27. The fuel cell of claim 17, wherein said ionomer is an alkoxy cellulose. 28. The fuel cell of claim 17, wherein said ionomer is a cyclodextrin. 29. The fuel cell of claim 17, wherein said ionomer is a mixture of zeolites. 30. The fuel cell of claim 17, wherein said ionomer is zirconium hydrogen phosphate. 31. A liquid-fed fuel cell comprising: a housing with an anode compartment and a cathode compartment; A Nafion ^™ polymer electrolyte membrane mounted within said tank and separating said anode and cathode compartments. a cathode formed on the side of the membrane facing the cathode compartment; an anode formed opposite the membrane facing the anode compartment, said anode being impregnated with Nafion ^™ ; means for circulating a solution of liquid organic fuel and water through said anode; means for circulating oxygen through said cathode; Means for removing carbon dioxide from the anode compartment; and Means for removing oxygen and water from the cathode compartment. 32. An electrode comprising a metal alloy impregnated with a hydrophilic water insoluble proton conducting ionomer. 33. A method of fabricating a carbon structure composed of binder-supported large surface carbon particles, said method comprising the steps of: immersing the carbon structure in a bath containing a liquid perfluorosulfonic acid polymer; and The carbon structure is removed and dried. 34. The method of claim 33, wherein said polymer is a solution of Nafion ^™ in 1% methanol. 35. The method of claim 33, wherein said step of immersing said carbon structure in a bath containing a liquid polymer is performed for 5-10 minutes. 36. A structure fabricated according to the method of claim 33. 37. In an electrodeposition bath for use in the manufacture of an electrode for a liquid-fed fuel cell, the improvement comprising adding an amount of perfluorooctane sulfonic acid to said bath. 38. A method of making an electrode for a fuel cell, said method comprising the steps of: providing a bath containing a solution of a metal salt dissolved in sulfuric acid; adding PFOS to said bath; disposing a large surface carbon electrode structure in said bath; placing the anode in said bath; and A voltage is applied between said anode and said electrode until the desired amount of metal is deposited on said electrode. 39. The method of claim 38, wherein said metal salt comprises chloroplatinic acid and potassium pentachlororuthenate. 40. The method of claim 38, wherein said anode is composed of platinum. 41. The method of claim 38, wherein said carbon electrode structure comprises carbon bonded to a Teflon ^(TM) binder. 42. The method of claim 38, wherein said carbon electrode comprises large surface carbon bonded by 15% by weight Teflon ^(TM) binder and coated to a carbon matrix fiber layer. 43. The method of claim 38, wherein the concentration of said acid is 0.01-0.05M. 44. The method of claim 38, further comprising the step of removing said electrode from said bath and washing said electrode with deionized water. 45. A method of making an electrode for use in a liquid organic fuel cell having metal ions deposited thereon, said method comprising the steps of: A bath solution is provided, which includes hydrogenhexachloroplatinate and potassium pentachlororuthenate solution dissolved in sulfuric acid, wherein the concentration of said chloroplatinic acid and potassium pentachlororuthenate is 0.01-0.05M; Adding PFOS to said bath at a concentration of 0.1-1.0 g/l; A large surface carbon electrode is placed in the bath, wherein the carbon electrode structure has a surface area of about 200 square meters per gram of carbon particles and a Teflon ^™ binder coated on a fiber-based carbon on paper placing a platinum anode in said bath; and A voltage is applied between said anode and said electrode until desired amounts of platinum and ruthenium are deposited on said electrode. 46. An electrode manufactured according to the method of claim 45. 47. In a liquid-fed fuel cell, the improvement comprises adding an amount of perfluorooctane sulfonic acid to the fuel of the fuel cell. 48. The improvement of claim 47, wherein said concentration of PFOS is at least 0.0001M. 49. The improvement of claim 48, wherein said perfluorooctane sulfonic acid is 0.0001M to 0.01M. 50. A liquid-fed fuel cell comprising: an anode; a cathode; means for circulating liquid organic fuel, water, acid electrolyte and PFOS additive solution through said anode; and means for circulating oxygen through said cathode. 51. A liquid-fed fuel cell comprising: an anode; a cathode; an electrolyte; means for circulating a liquid organic fuel selected from trioxane, dimethoxymethane and trimethoxymethane through said anode; and means for circulating oxygen through said cathode. 52. The fuel cell of claim 51, wherein said fuel is dissolved in water to a concentration of 0.1-2.0M. 53. A method of generating energy comprising the steps of: providing a liquid-fed fuel cell; and The liquid-fed fuel cell is operated using an organic fuel selected from trioxane, dimethoxymethane and trimethoxymethane.

Images