Patent Application
20150064581
1. Field of the Invention
The present invention relates to electrochemical devices and, more specifically, to electrochemical devices that employ ion exchange membranes.
2. Description of the Related Art
Fuel cells have the potential to provide clean and efficient energy sources for stationary, traction, and portable applications. Among the various types of fuel cells, the proton exchange membrane fuel cell (PEMFC) has several desirable features including a high level of development.
Typical PEMFC's employ a proton exchange membrane (such as Nafion®) around which is disposed two electrodes: an anode that includes a catalyst (such as platinum) and a cathode. A fuel, such as hydrogen gas, passes along the anode where the catalyst causes hydrogen atoms to be oxidized (act as a reducing agent) and split into positively-charged ions (e.g., protons in the case where hydrogen is the fuel) and electrons. The positively-charged tend to pass through the proton exchange membrane and the electrons travel through an electrical circuit to the reunite with the positively charged-ions and other reactants, which are reduced (act as an oxidizing agent) on the cathode side of the fuel cell.
Although PEMFC's have been successfully used in numerous applications, several disadvantageous features impede wide-scale commercialization of PEMFC's. For example, PEMFC's exhibit sluggish reaction kinetics, complex water management, carbon monoxide poisoning, limited lifetime due to membrane and electrode degradation and they require expensive noble metal catalysts and perfluorinated membranes. Typical PEMFC's require water to be transported from one side of the fuel cell to the other side to maintain the reaction. If too much water is produced by the fuel cell, the cell becomes flooded; if too little water is produced, the power output can be sub-optimal in low humidity conditions.
In an attempt to reduce fuel cell costs, attempts have been made to develop anion exchange membrane (AEM) fuel cells. Many AEM fuel cells are metal-free, operate at a high pH state and avoid carbonate poisoning (i.e. precipitation of carbonate salts). Such carbonate poising presents a major obstacle in conventional alkaline fuel cells, which use sodium or potassium hydroxide as an electrolyte.
The high pH environment within AEM fuel cells addresses many of the shortfalls with PEM-based fuel cells. Advantages of AEM-based fuel cells (AEMFC's) include the following: (i) the more facile electrokinetics allow for the use of non-noble metals, such as silver and nickel as catalysts; (ii) the wide selection of catalytic metals potentially extends the opportunity for selective catalysis; (iii) the direction of ion migration is from the cathode to the fuel-anode (opposite that of a PEM-based fuel cell) which may lower fuel crossover because electro-osmotic drag is in the opposite direction; (iv) the use of hydrocarbon membranes in place of a perfluorinated membrane may lower the cost of materials; and (v) more facile CO oxidation in an alkaline environment may significantly reduce CO poisoning.
Although AEMFCs offer important potential advantages, they also exhibit a lower ionic conductivity of AEMs compared to Nafion®. This is a concern because it may lower the performance of the fuel cell. Recent efforts have resulted in ionic conductivities of 20 to 30 mS/cm, which are lower than the conductivity of Nafion® (which is about 92 mS/cm).
Therefore, there is a need for an electrochemical device that is self hydrating without requiring complex water management schemes and that does not produce large amounts of undesirable compounds.
The disadvantages of the prior art are overcome by the present invention which, in one aspect, is a membrane electrode assembly for use in a fuel cell for converting a reducing agent to electrical energy, that includes an anode electrode, a proton exchange membrane, an anion exchange membrane and a cathode electrode. The anode electrode includes a first catalyst and has a first anode surface and an opposite second anode surface. The anode electrode is configured to receive the reducing agent along the first anode surface. The first catalyst is configured to separate the reducing agent into a plurality of positively charged ions and negative charges. The proton exchange membrane has a first proton exchange membrane side disposed adjacent to the second anode surface of the anode electrode and an opposite second proton exchange membrane side. The proton exchange membrane is configured to favor transport of positively charged ions therethrough and is also configured to inhibit transport of negatively charged particles therethrough. The anion exchange membrane has a first anion exchange membrane side that is disposed adjacent to the second proton exchange membrane side of the proton exchange membrane so as to form a membrane junction therebetween. The anion exchange membrane also has an opposite second anion exchange membrane side. The anion exchange membrane is configured to favor transport of negatively charged ions therethrough and is also configured to inhibit transport of positively charged ions therethrough. The cathode electrode includes a second catalyst and has a first cathode surface and an opposite second cathode surface. The first cathode surface is disposed adjacent to the second anion exchange membrane side of the anion exchange membrane. The cathode electrode is configured to receive at least one oxidizing agent along the second cathode surface. The second catalyst is configured to react electrons with the at least one oxidizing agent so as to create reduced species.
In another aspect, the invention is an electrochemical device that includes an anode electrode, a proton exchange membrane, an anion exchange membrane, a cathode electrode and an electrical device. The anode electrode includes a first catalyst and has a first anode surface and an opposite second anode surface. The anode electrode is configured to receive the reducing agent along the first anode surface. The first catalyst is configured to separate the reducing agent into a plurality of positively charged ions and negative charges. The proton exchange membrane has a first proton exchange membrane side disposed adjacent to the second anode surface of the anode electrode and an opposite second proton exchange membrane side. The proton exchange membrane is configured to favor transport of positively charged ions therethrough and is also configured to inhibit transport of negatively charged particles therethrough. The anion exchange membrane has a first anion exchange membrane side disposed adjacent to the second proton exchange membrane side of the proton exchange membrane so as to form a membrane junction therebetween and has an opposite second anion exchange membrane side. The anion exchange membrane is configured to favor transport of negatively charged ions therethrough and is also configured to inhibit transport of positively charged particles therethrough. The cathode electrode includes a second catalyst. The cathode electrode has a first cathode surface and an opposite second cathode surface. The first cathode surface is disposed adjacent to the second anion exchange membrane side of the anion exchange membrane. The cathode electrode is configured to receive at least one oxidizing agent along the second cathode surface. The second catalyst is configured to react electrons with the at least one oxidizing agent so as to create reduced species. The electrical device is coupled between the anode electrode and the cathode electrode.
In yet another aspect, the invention is a method of generating electrical energy from a reducing agent, in which a reducing agent is introduced to an anode electrode. The anode electrode includes a first catalyst and that is coupled to a first proton exchange membrane side of a proton exchange membrane. The proton exchange membrane has a second proton exchange membrane side disposed oppositely from the first proton exchange membrane side. The anode electrode is configured to separate the reducing agent into a plurality of positively charged ions and negative charges. An oxidizing agent is introduced to a cathode electrode that includes a second catalyst and that is coupled to a second proton exchange membrane side of an anion exchange membrane. The anion exchange membrane includes a first anion exchange membrane side that is disposed oppositely from the second proton exchange membrane side and that is adjacent to the a second proton exchange membrane side of the first proton exchange membrane. The cathode electrode is configured to receive the oxidizing agent along the second cathode surface. The second catalyst is configured to react electrons with the oxidizing agent so as to create reduced species. A load is coupled between the anode electrode and the cathode electrode. The load is configured to provide an electrical path between the anode electrode and the cathode electrode. The reduced species from the cathode are reacted with the oxidized species from the anode at an interface. For example, methanol can be oxidized at the anode to produce positively charged hydrogen ions and oxygen at the cathode can be reduced to form hydroxide. When hydrogen ions and hydroxide ions react, water is produced. The cathode electrode and the anion exchange membrane are hydrated with the water generated as a result of the reacting step.
These and other aspects of the invention will become apparent from the following description of the preferred embodiments taken in conjunction with the following drawings. As would be obvious to one skilled in the art, many variations and modifications of the invention may be effected without departing from the spirit and scope of the novel concepts of the disclosure.
A preferred embodiment of the invention is now described in detail. Referring to the drawings, like numbers indicate like parts throughout the views. Unless otherwise specifically indicated in the disclosure that follows, the drawings are not necessarily drawn to scale. As used in the description herein and throughout the claims, the following terms take the meanings explicitly associated herein, unless the context clearly dictates otherwise: the meaning of “a,” “an,” and “the” includes plural reference, the meaning of “in” includes “in” and “on.” Also as used herein, “negative charges,” “negatively charged ions” and “negatively charged particles” include electrons; “positive charges,” “positively charged ions” and “positively charged particles” include protons.
As shown in
The hybrid membrane electrode assembly 102 includes an anode electrode 110 having a first anode surface 112, along which passes the reducing agent, and an opposite second anode surface 124. The anode electrode 110 includes a first catalyst includes a material, such as a catalytic metal (e.g., platinum) that is configured to separate the reducing agent (e.g., H2) into a plurality of positively charged ions (e.g., H+ ions) and negative charges (e.g., electrons).
Disposed next to the second anode surface 124 is a proton exchange membrane (PEM) 120 that includes a sulfonated tetrafluoroethylene based fluoropolymer-copolymer membrane (which could be made of a material such as Nafion®, available from DuPont). The proton exchange membrane 120 has a first proton exchange membrane side 122 and an opposite second proton exchange membrane side 124. The proton exchange membrane favors transport of positively charged ions (e.g., H+) therethrough. However, it also inhibits transport of negatively charged particles (e.g., electrons) therethrough.
As shown in
At the interface 104, the immobile, anionic sulfonate groups remaining in the PEM 120 create an electric field WAEM that electrostatically opposes the diffusion of additional protons from the PEM 120 to the PEM/AEM interface 104. Likewise, the mobile hydroxide within the AEM will continue to diffuse to the PEM/AEM interface 104 until the fixed quaternary ammonium groups within the AEM 130 generate a field WPEM opposing the diffusion of additional hydroxide to the interface 104. At equilibrium, the net flux of protons in the PEM 120 to the interface 104 is zero because the flux to the interface 104 due to diffusion is counterbalanced by the flux away from the interface due to migration. Stated another way, protons in the PEM 120 will continue to react with hydroxide in the AEM 130 until the electrochemical potential in the two phases becomes equal. That is, the difference in activity of protons is counterbalanced by the potential difference in the two phases. The neutralization of the protons and hydroxide at the interface 104 of the AEM 130 and PEM 120 leaves that region at a relatively neutral pH. An electric field is generated by the fixed charges on each side of the interface 104 resulting in a potential difference between the two materials.
While use of hydrogen as a fuel (reducing agent) in the fuel cell 100 is discussed above, other reducing agents can also be used. For example, methanol (CH3OH) could also be used. In such an embodiment, the anode side would give rise to the following reaction: (2CH3OH+H2O→CO2+6H++6e−). The H− ions would pass thorough the proton exchange membrane 120 and the electrons would pass through the load 160. Whereas the cathode side would give rise to the following reaction: (2CO2+O2+4e−−→2CO32−). The CO32 ions would migrate through the anion exchange membrane and react with the H− ions eventually to generate H2CO3, which decomposes into CO2 and H2O. Thus, a safely-transportable organic fuel could be used to generate electricity while only producing harmless substances.
Disposed next to the second proton exchange membrane side 124 is an anion exchange membrane 130 (such as an ion exchange ionomer, e.g., poly(arylene ether sulfone) functionalized with a plurality of quaternary ammonium groups) that has a first anion exchange membrane side 132 and an opposite second anion exchange membrane side 134. The first anion exchange membrane side 132 is disposed adjacent to the second proton exchange membrane side 124 of the proton exchange membrane 120 and forms a membrane junction 104 therebetween. The anion exchange membrane 130 favors transport of negatively charged ions (e.g., OH−) therethrough and also inhibits transport of positively charged ions therethrough.
A cathode electrode 140 is disposed next to the second anion exchange membrane side 134. The cathode electrode 140 includes a second catalyst, which could include catalysts such as: platinum, palladium, silver, nickel and combinations thereof. In one lower cost embodiment, nickel is used as the second catalyst. The cathode electrode 140 has a first cathode surface 142 and an opposite second cathode surface 144. The first cathode surface 142 is disposed adjacent to the second anion exchange membrane side 134 of the anion exchange membrane 130. The cathode electrode 140 receives the oxidizing agent flowing through the oxidizing agent flow field casing 152 along the second cathode surface 144 and reacts electrons received from the anode electrode 110 through the load 160 with the oxidizing agent (and water molecules received through the anion exchange membrane 130 in the case where the reducing agent includes H2) to create reduced species (such as OH− ions).
As shown in
As shown in
In an embodiment shown in
In an experimental embodiment, to exploit the electrochemical advantages of the high pH environment in the electrodes and high conductivity of existing PEM technology (e.g. Nafion), a hybrid fuel cell was used one AEM electrode at high pH, the another electrode composed of existing PEM technology (Nafion), and a high conductivity Nafion membrane separator. The AEM/PEM junction introduces an additional potential, Ej, to the Nernst voltage, according to the following equation:
For wide differences in pH, the junction potential, Ej, can exceed 0.8 V, such as in the PEM/AEM hybrid with unit activity of H+ and OH−, respectively. The junction potential at the AEM/PEM boundary balances the changes in the standard potential at the electrodes due to their difference in pH resulting in a thermodynamic cell voltage of 1.23 V. Two configurations were tested: AEM anode with PEM cathode, and PEM anode with AEM cathode. The AEM anode/PEM cathode resulted in water dissociation at the AEM/PEM junction in order to maintain ionic conductivity.
The operation at the PEM-anode/AEM-cathode configuration was successfully demonstrated resulting in water generation at the PEM/AEM junction. The half cell, PEM/AEM interface, and overall reactions are given in the following equations:
Although oxygen reduction in a high pH environment occurs at lower potentials (0.4 V) than under acidic conditions, (1.23 V), the voltage loss at the cathode is compensated by the junction potential, Ej=(RT/F)ln(aOH
Water generation at the AEM/PEM junction produces a self-hydrating effect. Water is produced at the air cathode in a PEM cell or at the fuel anode in an AEM cell. At low humidity, both of these types of cells can experience problems because water has to be transported from the water producing electrode to the second electrode. However, in the PEM/AEM hybrid, as claimed below, water is produced within the membrane allowing for greater retention and easier transport to the second electrode. Preliminary studies here have shown that the performance of hybrid fuel cell without external humidification at 65° C. was superior to the performance of conventional PEM fuel cells.
For each water molecule consumed at the cathode, two water molecules are produced at the membrane. In a traditional fuel cell, water produced at one electrode has to be recycled to the opposite side of the device for use there. In the hybrid bipolar membrane demonstrated here, the water is produced near the electrode where it is used. Because twice as much water is produced as needed, only half the water produced has to be transported to the electrode where it is needed.
In an experimental embodiment, a Nafion® solution (5% suspension by mass) was used as the ionomer in the fabrication of a low pH electrode (PEM electrode). The high pH electrode (AEM electrode) was made using an anion exchange ionomer (AEI), poly(arylene ether sulfone) functionalized with quaternary ammonium groups. The AEI was stored in the Cl form as a solution of 5% mass in dimethyl formamide (DMF). The Nafion® membranes were pretreated with 3% H2O2 and 1 M H2SO4 solutions. The catalyst was platinum supported on carbon (Pt/C, E-Tek) with Pt loading of 20%.
The catalyst ink for the PEM electrode was prepared by mixing the Nafion® solution, Pt/C catalyst, isopropyl alcohol (IPA), and water. The catalyst ink for the anionic, AEM electrode was prepared by mixing the Pt/C catalyst and the AEI with a mixture of water and DMF (2:3 by mass). The catalyst inks were sonicated for 15 minutes and then cast onto hydrophobic Toray carbon paper (TGPH-090). Resulting electrodes had high (0.5 mg cm−2) or low (0.3 mg cm−2) ionomer content. After drying at room temperature, 50 μL of AEI in DMF (1% mass) was sprayed directly onto the AEM electrode surface. the AEM electrode was immersed in aqueous 0.1 M KOH to exchange OH− for Cl−. The resulting electrodes had a surface area of 2 cm2.
Prior to assembling the electrodes onto the membrane, 100 μL of Nafion® (5% suspension):IPA mixture (1:2 by volume) was sprayed onto both the AEM and PEM electrodes. The MEA was assembled in two steps. In the first step, the PEM electrode was pressed onto Nafion® 212 at 2 MPa gauge pressure and 135° C. for 3 min. In the second step, the AEM electrode was pressed onto the PEM half-cell assembly at 2 MPa and ambient temperature for 3 min.
A fuel cell hardware assembly (available from Fuel Cell Technologies, Inc.) was made of a pair of Poco graphite blocks with a single-serpentine flow pattern. All MEAs were preconditioned by operation at a steady state at 600 mV discharge voltage before performing I-V polarization experiments. The scan rate was 1 mV/s for I-V measurements. Electrochemical measurements were performed using a PAR 2273 potentiostat/galvanostat. Fuel cell tests were conducted at ambient pressure. AC impedance spectra were measured, following the steady state discharge at 600 mV, in the constant voltage mode using frequencies from 50 mHz-10 kHz. The amplitude of the AC voltage was 10 mV.
The performance of the hybrid cells was evaluated at different relative humidity levels. The cell voltage for a cell discharged at 100 mA cm−2 constant current at 60° C. The relative humidity (RH) was increased from 0% to 100% in increments of 25% every 24 hours. At the end of each 24 hour period, current-voltage curves were collected. Two hybrid MEAs with different ionomer loadings in the AEM electrode were tested: 0.3 mg cm−2 (named MEA-L) and 0.5 mg cm−2 (MEA-H). For the 0.3 mg cm−2 ionomer content, MEA-L, the initial cell voltage was 615 mV at 0% RH. The cell voltage increased slightly at the end of 24 hour-period for the 0 and 25% RH experiments. When the relative humidity was higher than 50%, the open circuit decreased slightly after 24 hour of operation. The drop in cell voltage was more severe at higher RH conditions. The cell voltage dropped to 500 mV at the end of 24 hours of operation at 100% RH. The reproducibility of the cell performance was confirmed by lowering the RH to 0% after the fully humidified test. The voltage rapidly increased back to 670 mV.
In-situ AC impedance spectroscopy was used to help understand the change in cell performance with RH. For all conditions, the impedance spectrum was a semicircle loop. The high frequency x-intercept is predominantly the MEA resistance established by the ionic resistance of the membrane. These spectra showed that the ionic resistance of the membrane was nearly constant from 0% to 100% RH. However, the radius of the semicircle loop increased with RH. Typically, the difference between the x-intercept values of the semi-circular response at high and low frequency is mainly governed by interfacial oxygen reduction kinetics, ionic conductivity and diffusion limitations within the depletion layer. Since the decrease in ionic conductivity of the PEM at high RH is not expected, the diffusion limitation within the catalyst layer was a likely reason of the increased resistance at higher RH values. These results demonstrate that the water generated at the interface of the AEM and PEM maintains adequate hydration in the MEA when the inlet gases were dry. Hydration of the gas streams results in excess water within the membrane and flooding of the electrodes and limited oxygen diffusion in the cathode catalyst layer. This is a significant result because the performance of conventional polymer electrolyte fuel cell relies on fully humidified gas feeds. Hydration or wicking of water from one electrode to the other can cause added complexity and loss in efficiency.
Higher ionomer loadings, 0.5 mg cm−2 (MEA-H), were investigated as a function of RH to understand the effect of ionomer loading on performance of the AEM electrode. The initial cell voltage was 504 mV at 0% RH and increased with time due to hydration of the MEA from the water produced at the AEM/PEM interface. When the relative humidity was increased to 25%, the MEA-H cell voltage gradually decreased. The decrease in cell potential became steeper when the RH was increased to 50%. Operation of the MEA-H cell failed at 75% RH. The higher ionomer loading resulted in poorer performance than the low ionomer electrode. This difference between the two MEAs is more pronounced at high RH. The response can be better understood by examining the I-V behaviour of the cells at different RH.
At 0% RH and 60° C., the MEA-H showed better performance than MEA-L at low current density (<120 mA cm−2). However, the performance of MEA-H may rapidly degrade at high current. Typically, voltage polarization at high current density is a sign of mass transfer limitations. The mass transfer resistance is also seen in the AC impedance spectra for MEA-H. Regarding AC impedance data at cell voltages from 400 mV to 850 mV, at the lowest current (highest cell voltage), 850 mV, the charge transfer resistance dominates, i.e. the largest loop. When the cell voltage decreases to 800 and 700 mV, the charge transfer resistance is lower. Decreasing the voltage below 700 mV results in a increase in the size of the semicircle. This observation shows that the diffusion limitation still occurs in the electrode layer when the gas feeds are dry. Moreover, the shape of the semicircle at low frequency was distorted when the cell voltage was 400 mV, which is generally an indication of mass transfer limitations in the gas diffusion layer. Since both oxygen and water are consumed (1:2 stoichiometry) in the cathode reaction, either could be the limiting reagent and the reason for the increase in mass transfer resistance. The diffusion of water from the PEM/AEM interface to the catalyst sites are not expected to be the limiting factor at higher current density because water is produced at the PEM/AEM interface at twice the rate that it is consumed at the AEM cathode. Thus, oxygen diffusion is believed to be primary reason for the mass transport limitations mentioned above. There are two possible causes for the increased mass transport resistance at high ionomer content in the cathode; physical barrier to gas transport by the ionomer itself or increased water content.
An additional comparison was made between the high and low ionomer loadings in the catalyst layer by observing the change in the internal resistance of MEAs during dry operation at 60° C. The high frequency x-intercept, RHF, of AC the impedance spectra for the different voltages and for the high ionomer loading, MEA-H, was 212(±3) Ωcm2 at 800 mV and decreased at higher currents (lower cell voltages) reaching 174 Ωcm2 at 200 mV. RHF for the low ionomer content, MEA-L, was 629(±96) Ωcm2 at 800 mV and decrease to 175 Ωcm2 at 200 mV. At low current density (high cell voltage), the water generated at the interface was not sufficient to fully hydrate the MEA, resulting in higher electrolyte resistance. The MEA-L electrode was more sensitive to hydration since it has a lower ionomer content (i.e. dehydrated faster at low humidity conditions). In contrast, the high ionomer content electrode retains water and more easily achieves full hydration. The hydration level quickly recovered at high current density for both electrodes following dehydration. These observations are consistent with the lower performance of MEA-L than MEA-H at low current densities. The water production is greater at higher current densities resulting a flooding in the catalyst layer. The performance of MEA-L is better than MEA-H due to faster dehydration of water from the electrodes, decreasing the flooding.
The flow rate of the inlet gases plays an important role in the fuel cell performance, particularly in dry operations. The effect of flow rate on the performance of hybrid MEA-L operating at 600 mV was studied at 0% RH. The steady-state current density obtained for different flow rates at 60° C. indicates that the performance of cells increases at higher anode and cathode flow rates. When the gas flow rates are 1 sccm O2 and 2 sccm H2, the current density is 39 mA cm−2 at 600 mV, corresponding a stoichiometric ratio (the ratio of the gas flow supplied in the gas feeds to the gas consumed at the reaction at a given current density) of 4 for both gas streams. At high flow rates, 8 sccm H2 and 8 sccm O2, the current density was 143 mA cm−2 at 600 mV. The stoichiometric ratios are 8 for H2 and 4 for O2. Typically, it is undesirable to operate an MEA at dry conditions with high stoichiometric flow rates because the rapid evaporation of the water at the cathode will lead to dry-out of the MEA. However the trend is opposite for the hybrid cells disclosed above.
In more detailed analysis, the effect of flow rate on the performance was analyzed by in situ AC impedance spectroscopy. The cell temperature was 60° C., the RH was 0%, and the oxygen flow rate at the anode was kept constant at 4 sccm. The impedance spectra of hybrid MEA operating at 600 mV for cathode flow rates of 2, 4, 6, and 8 sccm indicate that the high frequency x-intercepts (i.e. electrolyte resistance) are similar. These resistances are identical to the electrolyte resistance of MEAs operated at highly humidified conditions, showing the fully hydrated state of MEA at the flow rates used here. There is a significant drop in the low frequency x-intercept corresponding to a drop in the charge transfer resistance as the flow rate increases. This change in charge transfer resistance is attributed to the increased mass transport of oxygen into the electrode layer at high flow rates. The generation of water within the membrane close to the cathode can lower the access of the oxygen to the catalyst sites. Higher feed flow rates at the cathode increases the rate of water evaporation and increases the oxygen access to the catalyst sites. To confirm this effect on flow rate, an additional test was performed. Dry nitrogen gas was added to cathode stream increasing the total flow rate. This results in a higher rate of evaporation and dilution of the oxygen partial pressure in the cathode feed. The anode flow rate was held constant at 6 sccm for all experiments. The cell was initially run with the oxygen flow rate of 4 sccm at the current density of 100 mA cm−2. The cell voltage was 510 mV. An additional flow for 2 sccm nitrogen was added to the cathode feed at constant feed pressure and the oxygen flow rate was kept at 4 sccm. After a short period of time, the cell voltage increased to above 600 mV. When the flow rate of nitrogen was increased to 4 sccm, the cell voltage was 590 mV which was significantly higher than the pure oxygen at flow rate of 4 seem, even though the partial pressure of oxygen dropped by 50%. This test clearly shows that the improved performance at higher flow rates is prominently due to faster water removal from the cathode electrode at higher gas feed rates, enhancing oxygen access to the catalyst sites through more rapid water evaporation.
The effect of anode feed on cell performance was also evaluated. AC impedance spectra were collected for different anode flow rates while the cathode flow rate remained was constant at 4 sccm dry oxygen. The low frequency x-intercept decreased, reflecting a decrease in the charge transfer resistance, when the anode feed rate of the anode stream increased from 4 to 8 sccm. This is a significant increase in the performance with the anode flow rate. Since the anode is a traditional PEM electrode, hydrogen diffusion limitations are not expected at the moderate current densities obtained here. Also, the hybrid cell generates water close to the cathode so that anode flooding is not likely to occur. This effect of the anode flow rate on cell current is attributed to lowering the flooding in the cathode electrode. Higher evaporation rate for water at anode lowers the amount of water diffusing to the cathode from PEM/AEM interface, which subsequently decreases the water diffusion to the cathode.
An additional observation was made from the anode flow rate experiments. The high-frequency x-intercept in the AC impedance spectra, corresponding to electrolyte resistance, increased when the anode flow rate was increased from 4 sccm to 8 sccm. This indicates an increase in electrolyte resistance due to membrane dry-out. This behaviour corroborates that dehydration at high flow rates causes enhanced dehydration at the anode. However, this has little effect on the performance because the lower oxygen transfer resistance at the cathode is more significant than the loss of electrolyte conductivity due to dry-out at the anode.
The data presented above consistently shows that the cell performance at 60° C. and 0% relative humidity is prominently limited by flooding within the cathode layer. In order to reduce the flooding in the electrode layer, the hybrid cell was operated at 75° C. and 80° C. without external humidification. Under a cell voltage at 200 mA cm−2 constant current and feed rates of 8 seem H2 at the anode and 6 seem O2 at the cathode, at 75° C., the voltage reached to 500 mV after a short induction time. After 7 days of constant current operation, the voltage gradually decreased to 480 mV. When the temperature was increased to 80° C., the cell voltage increased to 580 mV and steady-state performance was achieved. An increase in the oxygen cathode flow rate resulted in an increase in the cell voltage. This preliminary evaluation indicates that the dry-feed performance hybrid fuel cells at 80° C. maintains a sufficient hydration level. The optimum balance between the rate of water removal (operating temperature and flow rates) and water generation (the current density) at the AEM/PEM interface is the key parameter for further improvements in the self-humidifying hybrid fuel cells.
The self-humidifying feature of the PEM anode/AEM cathode hybrid cell was evaluated at several test conditions. As opposed to conventional polymer electrolyte fuel cells, the performance of hybrid fuel cell was shown to improve at low relative humidity. I-V and AC impedance spectroscopy results show that the cell performance without external humidification is limited by flooding in the cathode electrode layer. The ionomer fraction within the cathode electrode plays a significant role in the cell performance. The dehydration rate of the electrode layer is lower at higher ionomer loading, increasing the flooding in the cathode layer. The effect of anode and cathode flow rate on the performance with dry gas feeds was significant. High flow rates resulted in faster water removal from the electrode lessening the flooding. Steady state operation at 580 mV and 200 mA cm−2 was demonstrated using dry H2/O2 feeds at 80° C.
Typically, the dynamics of a system can change significantly in application of fuel cells depending on the fuel used (e.g., hydrogen, methanol, ethanol etc.) and the operating conditions (relative humidity, current density etc.). Furthermore, the water dynamics in a single cell and stack vary at different locations due to the changes in feed/exhaust streams. For example, the sections close to hydrogen inlets can have a higher water content than the sections close to the anode exhaust. Therefore, several layers of control systems and designs, particularly for water management, may be required to homogenize the operating conditions throughout the cell and stack. Bipolar design enable one to alter the domain where the water is needed by simply varying the thickness of AEM and PEM layer in an individual cell as well as at different locations in a fuel cell stack. This gives flexibility and the ability to adjust the different dynamics in the cells and stack designs using different fuels and conditions as needed, depending on the specific application of the fuel cell.
Fuel cell construction and reactions are discussed in detail in U.S. patent application Ser. No. 11/502,731, filed on Aug. 11, 2006 and published as US Publication No. US 2007-0259236 A1 on Nov. 8, 2007, which is hereby incorporated by reference in its entirety.
The above described embodiments, while including the preferred embodiment and the best mode of the invention known to the inventor at the time of filing, are given as illustrative examples only. It will be readily appreciated that many deviations may be made from the specific embodiments disclosed in this specification without departing from the spirit and scope of the invention. Accordingly, the scope of the invention is to be determined by the claims below rather than being limited to the specifically described embodiments above.
This application claims the benefit of U.S. Provisional Patent Application Ser. No. 61/085,631, filed Aug. 1, 2008, the entirety of which is hereby incorporated herein by reference. This application is a continuation-in-part of, and claims the benefit of, U.S. patent application Ser. No. 12/534,117, filed Aug. 1, 2009, the entirety of which is hereby incorporated herein by reference.
This invention was made with government support under agreement No. CTS-0624620, awarded by the National Science Foundation. The government has certain rights in the invention.
| Number | Date | Country | |
|---|---|---|---|
| 61085631 | Aug 2008 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 12534117 | Aug 2009 | US |
| Child | 14536836 | US |
