WO2005078846A1

WO2005078846A1 - Dual electroreformer and fuel cell

Info

Publication number: WO2005078846A1
Application number: PCT/US2005/002631
Authority: WO
Inventors: Mahlon S. Wilson
Original assignee: The Regents Of The University Of California
Priority date: 2004-02-03
Filing date: 2005-01-31
Publication date: 2005-08-25

Abstract

A fuel cell system using electrochemical dual-cells. Each dual-cell includes a catalyzed fuel oxidation electrode (40) on one end, followed by an electrolyte separator (20), a hydrogen permselective barrier layer (10), another electrolyte separator (25), and finally a catalyzed oxidant reduction electrode (60) on the other end. A power supply (80) is electrically connected to the hydrogen permselective barrier layer (10) and the catalyzed fuel oxidation electrode (40) that provides a supplemental current to the hydrogen permselective barrier layer.

Description

DUAL ELECTROREFORMER AND FUEL CELL

STATEMENT REGARDING FEDERAL RIGHTS This invention was made with government support under Contract No. W- 7405-ENG-36 awarded by the U.S. Department of Energy. The government has certain rights in the invention. FIELD OF THE INVENTION The present invention relates generally to fuel cell technology, and, more particularly, to the use of an electroreformer with a fuel cell. BACKGROUND OF THE INVENTION A reformer is a device for converting hydrocarbon fuels, such as methanol, into free hydrogen, carbon dioxide, and water. The present invention comprises back-to-back electrochemical cells separated by an electronically conductive hydrogen permselective barrier layer that altogether form a "dual" cell: Cell One and Cell Two. (A hydrogen permselective barrier layer allows hydrogen to pass through, while at the same time preventing the passage of water and fuel.) Cell One is employed as an electrochemical reformer (electroreformer) that oxidizes a fuel (such as methanol) at the anode and provides hydrogen to the permselective barrier layer in the cathode reaction. The hydrogen then traverses the permselective barrier layer, the other side of which serves as the anode of Cell Two. Cell Two is effectively a hydrogen fuel cell that provides power for the electroreformer and an attached load. The present invention provides a supplemental current to the electroreformer that assists in maintaining a hydrogen content of the permselective barrier layer that is sufficiently high to sustain performance of the overall dual-cell. The use of the permselective barrier layer allows for a fuel cell system that reduces the difficulties with water balance and fuel cross-over to Cell Two cathode, both of which cause losses in efficiency, stability, and cathode performance. The benefits provided by the efficiency, stability, and improve performance are particularly relevant for low-power (e.g., cellular phone) types of methanol fuel cell power systems. One of the earliest uses of a permselective membrane in an electrochemical cell was taught in U.S. Patent no. 3,669,750, Fuel Cell System, issued on June 13, 1972, by Walter Juda. The 750 patent teaches an in-situ "hydrogen-permeable" layer to separate hydrogen from reformate formed over copper and zinc reforming catalysts located within the fuel cell. However, the reforming and hydrogen injection processes are chemical, not electrochemical, and, therefore, the system is essentially a hybrid between a membrane reactor and a fuel cell. U.S. Patent no. 5,141 ,604, Dehydrogenation Reaction Utilizing Mobile Atom Transmissive Membrane, issued on August 25, 1992, by William Ayers, is similar to Juda in that a chemical reaction (not electrochemical) is used to remove a "mobile atom" (hydrogen) from a reactant utilizing an electrically conductive, "atom permeable" membrane. Juda does teach the use of a biasing means (an induced electric field) to control the surface potential of the membrane, and, hence, control the sorption of the reactant on the membrane and the intermolecular bond strength within the reactant, thereby facilitating the dehydrogenation reaction. Ayers further describes fuel cell examples, whereby the dehydrogenation of a reactant on one side of the hydrogen permeable barrier layer supplies hydrogen that diffuses through the barrier layer to the other side to supply a fuel cell. A "potentiostat" (power source) is used to apply a bias to the barrier layer to facilitate the dehydrogenation reaction per above and a fuel cell power load is connected between the barrier layer and the oxygen reduction electrode. While the present invention also uses a power source and a power load, the electrical connections and current flows are different, and the power sources serve different functions. In Ayers' case, the power source induces a chemical dehydrogenation reaction to supply the permselective barrier layer with hydrogen, whereas, in the present invention, the power source drives an electrochemical process to supply supplemental hydrogen. Note that even when induced with an external bias, the chemical dehydrogenation reaction is lethargic. Ayers reports a current through the fuel cell circuit of 40 microamps/cm² when the cell is operated at 80°C. Using an electrochemical process, the dual-cell of the present invention achieves current densities a thousand times higher at 80°C. A methanol dual-cell with a permselective barrier layer was investigated in "A Methanol Impermeable Proton Conducting Composite Electrolyte System," J. Electrochem. Soc, Vol. 142, No. 7, pp L119-120, July 1995 by Pu, C, W. Huang, K. L. Ley and E. S. Smotkin. The authors concluded that the permselective barrier layer eliminated methanol cross-over but permitted hydrogen transport. However, the only results shown are with hydrogen present in the fuel stream. The addition of methanol to the fuel stream was only used to investigate the cross-over effects. Thus, with only gaseous hydrogen present, the cell taught in Pu et al., is essentially a hydrogen gas supplied fuel cell, as the potentials in the polarization curves indicate that the methanol is not contributing to the fuel cell current. No results are shown using methanol as the fuel, i.e., operation as a direct methanol fuel cell. Note that the issue of the loss of barrier layer permeability due to hydrogen discharge was not addressed. U.S. Patent 5,846,669, Hybrid Electrolyte System, issued on December 8, 1998, by Smotkin et al., builds upon the work described in the paper above, and teaches the use of a permselective barrier layer that is used to separate acidic and alkaline electrolytes in a hydrogen-fuelled fuel cell. However, although it is again stated that the use of the permselective membrane prevents the crossover of hydrocarbon fuels to the fuel cell cathode, only hydrogen fueled results are provided. Again, the issue of the loss of barrier layer permeability due to hydrogen discharge was not addressed. Experiments have shown that the fuel cell configuration as used by Pu et al. does not operate on methanol (or similar fuels) because the hydrogen transport through the permselective membrane is eventually lost if hydrogen is not in the fuel feed. This is because hydrogen transport through a palladium permselective barrier layer is a non-ionic hydrogen diffusion process. As such, hydrogen permeability is primarily a function of hydrogen content (permeability = diffusivity x solubility). Unless replenished, hydrogen content within the permselective membrane decreases due to loss by transport inefficiencies, diffusion to the reactant electrodes, or to scavenging by oxygen diffusing in from the fuel cell cathode. Hydrogen permeability, and, thus, cell performance then steadily decreases until no more hydrogen is available to the fuel cell side of the dual cell. In cells operating with hydrogen in the fuel feed as per Pu et al. and Smotkin et al., the barrier layer is replenished by hydrogen diffusing in through the electrolyte from the fuel anode. When the hydrogen is removed from the fuel supply, the permselective barrier layer becomes depleted. Even though hydrogen is continuously added to the permselective barrier layer during the process by reduction of the protons, conservation of charge dictates that for every proton that becomes hydrogen on one side of the barrier layer, a hydrogen must leave the opposite side of the barrier layer. Consequently, the net addition is zero and performance dies as hydrogen is lost to scavenging mechanisms. The solution provided by the present invention is to provide supplemental hydrogen directly to the permselective barrier layer, while keeping the supplemental hydrogen separate from the fuel feed. The separation is important, as supplemental hydrogen in the fuel feed would preferably oxidize at the anode instead of traversing the electrolyte to the barrier layer. Additional objects, advantages and novel features of the invention will be set forth in part in the description which follows, and in part will become apparent to those skilled in the art upon examination of the following or may be learned by practice of the invention. The objects and advantages of the invention may be realized and attained by means of the instrumentalities and combinations particularly pointed out in the appended claims.

SUMMARY OF THE INVENTION In accordance with the purposes of the present invention, as embodied and broadly described herein, the present invention includes a fuel cell system using an electrochemical dual-cell configuration. Each dual-cell includes a catalyzed fuel oxidation electrode on one end, followed by an electrolyte separator, a hydrogen permselective barrier layer, another electrolyte separator, and finally a catalyzed oxidant reduction electrode on the other end. A power supply is electrically connected to the hydrogen permselective barrier layer and the catalyzed fuel oxidation electrode to provide a supplemental current to the hydrogen permselective barrier layer. BRIEF DESCRIPTION OF THE DRAWINGS The accompanying drawings, which are incorporated in and form a part of the specification, illustrate the embodiments of the present invention and, together with the description, serve to explain the principles of the invention. In the drawings: Figure 1 is a pictorial illustration of a hydrogen permselective barrier layer dual cell. Figure 2 is a pictorial illustration displaying the reactions and transport species within a typical dual-cell configuration using acidic electrolytes. Figure 3 is a pictorial illustration of one embodiment of the present invention dual-cell electroreformer. Figure 4 graphically displays the present invention total dual-cell polarization and resistance curves using hydrogen gas as a fuel. Figure 5 graphically displays the present invention total dual-cell polarization and resistance curves using methanol as a fuel. Figure 6 graphically displays the Cell Two polarization curves constructed from the methanol and hydrogen test runs displayed in Figures 4 and 5. Figure 7 is a schematic of a bipolar circuit used with the present invention. Figure 8 is a schematic of a monopolar circuit configuration used with the present invention. Figure 9 is a schematic of a monopolar circuit configuration, including MOSFET switches, used with the present invention.

DETAILED DESCRIPTION Direct Methanol Fuel Cells (DMFCs) are of particular interest because of the greater ease of storing the liquid fuel when compared to the hydrogen gas typically used in conventional hydrogen fuel cells. DMFCs are not without their difficulties, however, particularly if based on low temperature technologies such as the polymer electrolyte fuel cell (PEFC). Foremost of the identified difficulties is the permeation of the fuel across the polymer electrolyte membrane (PEM), also referred to as fuel cross-over, which results in both fuel loss (with corresponding decrease in efficiency) and negative effects on the air electrode, such as mixed electrochemical potential and electrode flooding. To address these concerns, a popular strategy to minimize fuel cross-over is to provide a lean feed of methanol fuel (e.g., 0.5 M methanol in water), so that the majority of fuel is consumed and/or the amount that permeates the PEM is minimized because of the dilute source. However, even when employing this strategy in passive systems, the cross-over flux can be as great as the fuel usage at the anode, resulting in a substantial decrease in overall efficiency. Furthermore, the cross-over methanol causes problems with cathode stability, as the presence of the methanol at the cathode causes a mixed potential (i.e., competes with the oxygen reduction reaction), and longevity, as the methanol causes problems by wetting out otherwise hydrophobic structures and thus interfering with oxygen access and transport (i.e., "flooding" of the electrode). An alternative approach, provided by the present invention, is to "electrochemically reform" or "electroreform" the fuel in a separate "first" cell (Cell One) to form protons at the anode, that are then evolved as hydrogen at the cathode of Cell One. The evolved hydrogen is separated from the fuel and introduced to the anode of a "conventional" hydrogen fuel cell (Cell Two). The two separate electrochemical cells are combined to form a "dual-cell" by placing a hydrogen permeable separator, hereinafter referred to as a permselective barrier layer, between the two separate cells that allows hydrogen to pass through while blocking other chemical species. The surfaces of the permselective barrier layer, preferably treated with a catalyst, serves the electrochemical functions of directly converting protons from Cell One to (monatomic) hydrogen dissolved within the permselective barrier layer on one side and at the other side forming protons for the second cell from the monatomic hydrogen within the barrier layer. Transport of the monatomic hydrogen through the barrier layer proceeds by diffusion of the uncharged species, and, hence, is a function of hydrogen content, which can be increased or maintained by the addition of a supplemental electric current. A portion of the power produced by Cell Two may be used to provide the supplemental current that drives the methanol oxidation/ hydrogen formation in Cell One. The supplemental current drives the electrochemical oxidation of a small amount of fuel to generate a slight excess of protons that conduct across the Cell One electrolyte to introduce the supplemental hydrogen to the permselective barrier layer. Consequently, the net effect is to increase the total current flow through Cell One, somewhat higher than the current provided by Cell Two. The benefit of the extra current to Cell One is that more hydrogen is electrochemically supplied to the permselective barrier layer than is electrochemically removed. This supplemental hydrogen overcomes losses and inefficiencies in the barrier layer hydrogen transport process that would otherwise make it impossible to operate on fuels other than hydrogen. Referring now to Figure 1 , permselective barrier layer 10 was "blackened" (electrochemically coated with high surface area platinum to facilitate the hydrogen interfacial reactions) pure palladium and was placed between two electrolyte separators 20, 25 in a fuel cell configuration that comprised: fuel flow field 30, catalyzed fuel oxidation electrode (Cell One anode) 40, air flow field 50, catalyzed oxidant reduction electrode (Cell Two cathode) 60, and electric load 70. Figure 2 portrays the reactions and transport processes within an ideal dual-cell fuel cell. Note that permselective barrier layer 10 must be electrically conductive to shuttle electrons between the interfacial processes since the hydrogen transported through permselective barrier layer 10 is as an uncharged species. Fuel reacts with water at Cell One anode 40 releasing protons that travel through first electrolyte separator 20, creating an ionic current to Cell One cathode 45, which is located on the Cell One side of permselective barrier layer 10. At Cell One cathode 45, the protons and combine with electrons, creating non-charged hydrogen that travel through permselective barrier layer 10 to Cell Two anode 65 (the opposite side of barrier layer 10). Protons are again formed with the release of an electron and travel through second electrolyte separator 25, creating an ionic current to Cell Two oxidant reduction cathode 60, where the protons combine with oxygen and electrons to form water. The processes described in Figure 2 are for acidic electrolytes, however, alkaline electrolytes may also be used. When the dual-cell was operated with hydrogen as the feed in Cell One, the configuration behaved as a regular fuel cell, but at somewhat diminished currents, due to the hydrogen transport and interfacial limitations of permselective barrier layer 10. Introducing methanol along with the hydrogen had little or no effect on dual-cell performance because permselective barrier layer 10 prevented any methanol access to Cell Two cathode 60. However, when the hydrogen feed was removed, forcing methanol oxidation to provide the protons for hydrogen formation and injection into permselective barrier layer 10, the power output of the dual-cell diminished and eventually ceased. The reason for the loss of power output with using methanol as a fuel hinges on the need to maintain a high hydrogen content within permselective barrier layer 10. Since permselective barrier layer 10 was electrically isolated, the charge transferred between Cell One and Cell Two must be conserved. Consequently, the hydrogen generated at Cell One cathode 40 on one side of permselective barrier layer 10 must equal the hydrogen utilized at Cell Two anode 65 on the other side. However, any inefficiencies with the hydrogen transport into permselective barrier layer 10 or the presence of any mechanism that scavenges hydrogen from permselective barrier layer 10, decreases the amount of hydrogen within permselective barrier layer 10. Since lowering hydrogen content within permselective barrier layer 10 lowers the hydrogen flux (permeability = concentration x diffusivity), less hydrogen is available to Cell Two, resulting in an increased overpotential condition at Cell Two anode 65 causing overall fuel cell performance to drop. One mechanism that results in the loss of hydrogen from permselective barrier layer 10 is the escape of hydrogen gas through electrolyte separators 20, 25. Another mechanism that results in the loss of hydrogen is that oxygen can diffuse in from Cell Two cathode 60 to permselective barrier layer 10, scavenging hydrogen to form water. Referring now to Figure 3, one embodiment of the present invention, in order to replace hydrogen lost due to the various mechanisms, uses power supply 80, attached to permselective barrier layer 10, to electrochemically charge permselective barrier layer 10 with hydrogen. In this embodiment, Cell One anode 40 is used as the necessary counter-electrode for power supply 80. In another embodiment, a 4 cm² active area dual-cell utilizes a 50 micron thick Ta foil vacuum-coated on either side with 5,000 A of Pd as permselective barrier layer 10, with polymer electrolyte and catalyst inks painted onto both sides. A preferred polymer electrolyte is Nation™. The Pd-coated Ta foil is highly permeable to hydrogen. However, hydrogen embrittlement may crack the tantalum under certain conditions. Therefore, the cell components are tightly swaged together, helping to keep any cracks formed in permselective barrier layer 10 closed. Because methanol seepage through the cracks could occur, the fuel cell cathode performances were compared with and without methanol in the fuel feed in order to determine if any evidence of methanol cross-over was apparent. 6.45 cm² square permselective barrier layer 10 was sandwiched between two electrolyte separators 20, 25. Cell One anode 40 comprised 2.4 mg/cm² of Pt- RuOx mixed with a solid polymer electrolyte solution to form an ink and painted onto an uncatalyzed gas diffusion backing. A preferred gas diffusion backing is ELAT®, manufactured by E-TEK, Inc. Cell Two cathode 60 comprised a 0.5 mg Pt/cm² catalyzed ELAT® gas diffusion electrode impregnated with a solid polymer electrolyte. Included between first electrolyte separator 20 in Cell One and Pd/Ta permselective barrier layer 10 was polyester film 90 with a 4 cm² window that defined the active area. A preferred polyester film is Mylar®. Polyester film 90 served to prevent ionic conductivity and fuel transport between membrane 20 in the area not occluded by Pd/Ta permselective barrier layer 10. The dual-cell assembly was inserted into standard single-cell testing hardware that included machined graphite flow-fields. Aside from the electrical connection to Pd/Ta permselective barrier layer 10, all other elements and connections were identical to any conventional fuel cell. The hydrogen and air feeds used in the tests were pressurized to 3 bar (30 psig) and sparged through humidifier bottles at or above the fuel cell temperature of 80°C. Methanol/ water mixtures were introduced with a liquid feed pump at near ambient pressure. The dual-cell was connected to load 70 to control total-cell voltages or current. When supplemental charging of the barrier layer was performed, power supply 80 was connected between Cell One anode 40 (the methanol oxidation anode) and Pd/Ta permselective barrier layer 10. Charging of permselective barrier layer 10 was performed at constant current. Total dual-cell voltage, current and high frequency resistance were measured as well as the Cell One voltage (between Cell One anode 40 and barrier layer 10). Figure 4 graphically displays the dual-cell performance while operating with hydrogen and air reactants. The curves depict down-and-back voltage scans, consequently any hysteresis in the processes is evident. The total cell polarization curve indicates substantial hysteresis, with performance dropping significantly on the upward sweep. Most likely, the barrier layer becomes depleted of hydrogen on the fuel cell side as current density increases. The polarization curves show that the current densities increase unambiguously with the barrier layer charging current, particularly with increasing current. This suggests that some of the inefficiencies of the hydrogen injection process can be partially overcome by additional charging current. Despite the overall higher performance, the extraneous charging current increases polarization of Cell One anode 40, which is shown in the anode - barrier layer curves, the lowest set of curves in the figure. A roughly 10 mV increase is obtained per 5 mA/cm² of extraneous charging current at any given total cell current. The penalty is more severe than a simple superimposition of currents would anticipate. The last set of curves depict the high frequency resistances for the total cell, which are quite low, much as expected for the equivalent combined thicknesses of electrolyte separators 20, 25. This indicates that the interfaces are well-coupled ionically/ electronically and both electrolyte separators 20, 25 are well hyd rated. Figure 5 graphically displays operation of the dual-cell on a water/ methanol mixture. Curves are not depicted for the zero supplemental current charge case, as performance continuously degrades as the barrier layer discharges, as discussed previously. The total cell polarization curves are lower than with the hydrogen feed, because the polarization of Cell One anode 40 is so much greater with methanol oxidation compared to hydrogen. Indeed, the anode-barrier layer electroreformer voltage curves suggest that Cell One anode 40 is polarized 300 - 400 mV more than when hydrogen is used, as expected. Once again, the supplemental charging current had a greater than anticipated effect on voltage between Cell One anode 40 and cathode 45. The 10 mA/cm² anode barrier layer curve was about 30 mV higher than the 5 mA/cm². Correspondingly, the total dual-cell polarization curves were not much different at lower current densities, although the 10 mA/cm² run attained the higher current densities of the two curves. In general, the results indicated that the operation of a dual-cell using methanol can be sustained by supplemental charging of permselective barrier layer 10. Figure 6 graphically portrays five test runs, three with hydrogen feed and two with methanol feed. The Cell Two (fuel cell side) polarization curves were constructed for both the hydrogen and methanol experiments by adding the anode-barrier layer voltages to the total dual cell voltages in the polarization curves in Figures 4 and 5. Review of the results indicates that at the higher currents, the hydrogen-fed cells provided higher performances. Evidently, permselective barrier layer 10 in the methanol feed configuration is somewhat more hydrogen starved than permselective barrier layer 10 in the hydrogen feed configuration. This is a result of the pressurized hydrogen gas feed augmenting the barrier layer charging. Note that all the curves coincide at the lower current densities. This factor suggests that no methanol is leaking into Cell Two, otherwise the cathode performance would drop in this region. Thus, operation with methanol as the fuel in Cell One was achieved without any cross-over of the methanol to the fuel cell cathode in Cell Two. Although methanol cross-over and the attendant loss in cathode performance and fuel efficiency are avoided with this scheme, extraneous power needs to be expended to keep the barrier layer charged. For example, if the direct methanol barrier layer cell is operated at 60 mA/cm², 0.4 V/cell and a 5 mA/cm² extraneous charging current per Figure 5, about 1.7 mW/cm² is expended in the supplemental charge to yield a gross power of 24 mW/cm2. Thus a roughly 10% parasitic power loss is incurred, allowing for some inefficiencies in the charging electronics. While unfortunate, the losses are quite low compared to fuel efficiency losses in conventional cross-over prone cells. Referring now to Figure 7, a two cell section of a dual-cell stack is pictorially illustrated. Conventional bipolar plates 95 are used to separate and electronically link the dual-cells. All of the dual-cell components are labeled as in Figure 1. Load 70 is placed in parallel with energy storage device 100. In various embodiments, energy storage device 100 may be a battery or a capacitor. Thus, during time periods where the load on the fuel cell decreases, the excess power produced is used to charge energy storage device 100. DC-DC converter 110 is also connected in parallel with both load 70 and energy storage device 100, and supplies regulated power to multiplexer 120. Multiplexer 120 provides supplemental current to each dual-cell barrier layer 10 in turn. Note that DC-DC converter 110 must include electrically floating outputs that are multiplexed to each individual barrier layer 10, in order to account for the step changes in voltages across the dual-cell stack. The dual-cell voltages may be monitored through logic circuitry (not shown) that provides extra current to dual-cells that are underperforming. Referring to Figure 8, in another embodiment, a dual-cell system utilizes power switches 130 between dual-cells 5, and supplemental current switches 140 between barrier layer 10 and Cell One anode 40 of adjacent dual-cells 5. Here, the dual-cells cannot be stacked in the conventional sense using bipolar plates, and, thus, a monopolar configuration is used. In a monopolar configuration, the cells do not share components as in a bipolar plate stack (Figure 7). While the dual-cell system is providing a power output to load 70 or energy storage device 100, power switches 130 are closed and supplemental current switches 140 are open. When barrier layers 10 performance degrades, supplemental current switches 140 close as power switches 130 open, allowing barrier layers 10 to recharge using power from energy storage device 100 that also supplies load 70 while in this mode. When using methanol as the fuel, the total dual-cell voltage is roughly equal to that of the electroreformer, although opposite in sign. Consequently, a series of cells provides a charging voltage for a battery that is roughly equal to the voltage necessary to charge the permselective barrier layers, although at a slightly higher current density. A charge duty cycle of about 5% suffices in keeping the permselective barrier layers charged to the same degree as the constant current experiments. For n dual-cells, 2n+2 switches/ relays are then needed to accomplish both modes. While Figure 8 shows a simple configuration to achieve the power/ charge circuits, some of the power switches may see a reverse voltage equivalent to the open-circuit voltage of Cell Two (fuel cell) side during the charge phase. While this is not an issue with an electromechanical switch or relay, too high of a reverse bias may cause back current in solid-state switches based on reverse p-n junction diodes. This concern is resolved by reversing the cascade direction for the barrier layer charge circuit such that none of the power switches see a reverse bias any greater than about 0.1 V; this embodiment is shown in Figure 9. MOSFET (metal-oxide-semiconductor field-effect transistor) switches are commercially available (e.g., International Rectifier, Inc) with low resistances, on the order of 6 milli-Ohm, that consume very little actuation power and are rated for millions of cycles. The solid state switches can be used to cycle the power/ charge modes at relatively high frequencies to minimize the amount of energy that needs to be stored. This makes the use of capacitors more attractive for this function, which further increases the full-cycle efficiency compared to a battery. In the case of MOSFETs, the gate (actuation) voltages need to be about 1 V or greater than the drain voltage. Likewise, the use of electromechanical relays typically requires higher voltages. Therefore, a DC-DC boost converter is used to provide the higher voltages. The use of power/ charge cycling circuitry in tandem with a battery / capacitor is the preferred approach due to simplicity, low cost, and efficiency. The foregoing description of the invention has been presented for purposes of illustration and description and is not intended to be exhaustive or to limit the invention to the precise form disclosed, and obviously many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and its practical application to thereby enable others skilled in the art to best utilize the invention in various embodiments and with various modifications as are suited to the particular use contemplated. It is intended that the scope of the invention be defined by the claims appended hereto.

Claims

WHAT IS CLAIMED IS: 1. A fuel cell system comprising at least one electrochemical dual-cell, said at least one dual-cell comprising: (a) a catalyzed fuel oxidation electrode, (b) a first electrolyte separator and a hydrogen permselective barrier layer, said first electrolyte separator placed between said catalyzed fuel oxidation electrode and said hydrogen permselective barrier layer, (c) a second electrolyte separator and a catalyzed oxidant reduction electrode, said second electrolyte separator placed between said hydrogen permselective barrier layer and said catalyzed oxidant reduction electrode, and (d) a power supply electrically connected to said hydrogen permselective barrier layer and said catalyzed fuel oxidation electrode. 2. The apparatus of claim 1 where said hydrogen permselective barrier layer comprises palladium. 3. The apparatus of claim 1 where said hydrogen permselective barrier layer comprises a tantalum foil coated with palladium. 4. The apparatus of claim 1 where said hydrogen permselective barrier comprises a catalyst and a polymer electrolyte. 5. The apparatus of claim 1 where said catalyzed fuel oxidation electrode comprises a first gas diffusion backing and a Pt-RuOx layer comprising a polymer electrolyte. 6. The apparatus of claim 1 where said catalyzed oxidant reduction electrode comprises a second gas diffusion backing and a catalyst layer comprising platinum and a polymer electrolyte. 7. The apparatus of claim 1 where a polyester film is located between said first polymer electrolyte membrane and said hydrogen permselective barrier layer to prevent ionic conductivity and fuel transport between an area not occluded by said hydrogen permselective barrier layer.

8. The apparatus of claim 1 comprising a plurality of said at least one electrochemical dual-cells.

9. The apparatus of claim 8 where said plurality of said at least one electrochemical dual-cells are electrically connected in series.

10. The apparatus of claim 1 where said power supply is an electrical output of said at least one dual-cell fuel cells.

11. The apparatus of claim 1 where said power supply is a power storage device electrically connected to an electrical output of said at least one electrochemical dual-cell.

12. The apparatus of claim 1 where said power supply further comprises a DC-DC converter.

13. The apparatus of claim 12 further comprising a multiplexer to distribute power to said hydrogen permselective barrier layer.

14. A method for providing a power output from an electrochemical dual- cell, comprising: (a) supplying a fuel to a first cell anode, where said fuel is oxidized to provide a first ionic current, (b) allowing ionic conduction across a first electrolyte separator to a first cell cathode where hydrogen is formed, (c) allowing said hydrogen to traverse a hydrogen permselective barrier layer, (d) supplying a supplemental current from said first cell anode to said hydrogen permselective barrier layer, (e) oxidizing said hydrogen on a second cell anode to provide a second ionic current, and

5. The method of claim 14, further comprising the step of regulating said power output to provide said supplemental current to said hydrogen permselective barrier layer.