Analysis of a molten carbonate fuel cell: Numerical modeling and experimental validation

A detailed dynamic model incorporating geometric resolution of a molten carbonate fuel cell (MCFC) with dynamic simulation of physical and electrochemical processes in the stream-wise direction is presented. The model was developed using mass and momentum conservation, electrochemical and chemical reaction mechanisms, and heat-transfer. Results from the model are compared with data from an experimental MCFC unit. Furthermore, the model was applied to predict dynamic variations of voltage, current and temperature in an MCFC as it responds to varying load demands. The voltage was evaluated using two different approaches: one applying a model developed by Yuh and Selman [C.Y. Yuh, J.R. Selman, The polarization of molten carbonate fuel cell electrodes: I. Analysis of steady-state polarization data, J. Electrochem. Soc. 138 (1991) 3642–3648; C.Y. Yuh, J.R. Selman, The polarization of molten carbonate fuel cell electrodes: II. Characterization by AC impedance and response to current interruption, J. Electrochem. Soc. 138 (1991) 3649–3655] and another applying simplified equations using average local temperatures and pressures. The results show that both models can be used to predict voltage and dynamic response characteristics of an MCFC and the model that uses the more detailed Yuh and Selman approach can predict those accurately and consistently for a variety of operating conditions.     

Transient behavior of CO poisoning of the anode catalyst layer of a PEM fuel cell

A one-dimensional transient mathematical model is applied to simulate the carbon monoxide poisoning effect on the performance of a PEM fuel cell. Based on the CO kinetic model developed by Springer et al. [T.E. Springer, T. Rockward, T.A. Zawodzinski, S. Gottesfeld, J. Electrochem. Soc. 148 (2001) A11–A23], the transient behaviors of the CO poisoning process across the anode catalyst layer is investigated. The results show that the hydrogen coverage, θ_H, decreases with the time due to CO adsorption on the catalyst sites. A higher CO concentration results in fewer available catalyst sites for hydrogen electro-oxidation and a significant decrease in the response time to reach steady state, t_ss. Increasing the anode overpotential and the gas porosity would result in an increase in the current density, especially at low levels of CO concentration.     

Model development for a SOFC button cell using H2S as fuel

In this paper we present a hierarchy of models built to describe the overall performance of a single H₂S fuelled button cell solid oxide fuel cell (SOFC). The cell, used in the experimental studies of Liu et al. [M. Liu, G. Wei, J. Luo, A.R. Sanger, K.T. Chuang, Use of metal sulfides as anode catalysts in H₂S–air SOFCs, J. Electrochem. Soc. 150 (2003) 1025–1029], was a planar cell with a circular disc-like electrode assembly and the fuel and air flowing through a concentric cylindrical tube assembly. The goal is to model the electrochemical reaction coupled with mass transfer, fluid flow and current/voltage distribution in an yttria stabilized zirconia electrolyte fuel cell assembly operated between 750 and 850 °C. The models built range in complexity from an algebraic system of equations that calculates the activation, concentration and ohmic losses, to a two-dimensional finite element model that solves all the physics in the SOFC simultaneously. Kinetic parameters in these (progressively more comprehensive) models have been estimated and compared, leading hopefully to more accurate estimates for these parameters.     

The effect of reservoirs and fuel dilution on the dynamic behavior of a PEMFC

Data are presented to characterize the effects of reservoir size and hydrogen dilution on the dynamic behavior of a proton exchange membrane fuel cell (PEMFC) subjected to rapid changes in the voltage when the flowrates are constant. The data consist of the responses of the current density during low fuel stoichiometries in an effort to expand an understanding of the previously observed overshoot/undershoot behavior. That is, recent studies of the dynamic behavior of a PEMFC have shown pseudo-second-order dynamics of the current response to a change in voltage [J. Power Sources (2004); J. Electrochem. Soc. (2004)]. The data reported here lend further evidence that under fuel starved conditions, rapid changes in the cell voltage between 0.7 and 0.5 V yield pressure differences sufficient to create a “vacuum” effect. This vacuum effect may cause fuel to be drawn from the manifold in a stack or cause ambient air to enter a laboratory scale cell. The vacuum effect explained in our previous work [J. Power Sources (2004)] is shown here to depend on diameter and volume of fuel reservoirs and on the concentration of hydrogen in the fuel.     

A design tool for predicting the capillary transport characteristics of fuel cell diffusion media using an artificial neural network

Developing a robust, intelligent design tool for multivariate optimization of multi-phase transport in fuel cell diffusion media (DM) is of utmost importance to develop advanced DM materials. This study explores the development of a DM design algorithm based on artificial neural network (ANN) that can be used as a powerful tool for predicting the capillary transport characteristics of fuel cell DM. Direct measurements of drainage capillary pressure–saturation curves of the differently engineered DMs (5, 10 and 20 wt.% PTFE) were performed at room temperature under three compressions (0, 0.6 and 1.4 MPa) [E.C. Kumbur, K.V. Sharp, M.M. Mench, J. Electrochem. Soc. 154(12) (2007) B1295–B1304; E.C. Kumbur, K.V. Sharp, M.M. Mench, J. Electrochem. Soc. 154(12) (2007) B1305–B1314; E.C. Kumbur, K.V. Sharp, M.M. Mench, J. Electrochem. Soc. 154(12) (2007) B1315–B1324]. The generated benchmark data were utilized to systematically train a three-layered ANN framework that processes the feed-forward error back propagation methodology. The designed ANN successfully predicts the measured capillary pressures within an average uncertainty of ±5.1% of the measured data, confirming that the present ANN model can be used as a design tool within the range of tested parameters. The ANN simulations reveal that tailoring the DM with high PTFE loading and applying high compression pressure lead to a higher capillary pressure, therefore promoting the liquid water transport within the pores of the DM. Any increase in hydrophobicity of the DM is found to amplify the compression effect, thus yielding a higher capillary pressure for the same saturation level and compression.     

An algorithm for estimation of membrane water content in PEM fuel cells

Available humidity sensing techniques are often intrusive, and of limited practical interest for real-time control applications due to their cost, size, and inadequate response time and accuracy. In this study, we present a novel method for estimation of PEM fuel cell humidity by exploiting its effect on cell resistive voltage drop. This voltage loss is discerned from mass transport, concentration, activation losses and open circuit voltage by a well-known fuel cell voltage model. The proposed scheme makes use of measurements of voltage, current, temperature, and total pressure values in the anode and cathode. It also incorporates dynamic estimators for hydrogen and oxygen partial pressures, adapted from [M. Arcak, H. Gorgun, L.M. Pedersen, S. Varigonda, A nonlinear observer design for fuel cell hydrogen estimation, IEEE Trans. Control Syst. Technol. 12 (1) (2004) 101–110]. The membrane resistance thus obtained is then used to estimate membrane water content following functional characterizations presented in [T.E. Springer, T.A. Zawodzinski, S. Gottesfeld, Polymer electrolyte fuel cell model, J. Electrochem. Soc. 138 (8) (1991) 2334–2342]. Experiments with this estimation technique, performed at the Connecticut Global Fuel Cell Center, are presented and discussed.     

Modelling CO poisoning and O2 bleeding in a PEM fuel cell anode

Fuel gas containing carbon monoxide severely degrades the performance of a polymer electrolyte membrane (PEM) fuel cell. However, CO poisoning can be mitigated by introducing oxygen into the fuel (oxygen bleeding). A mathematical PEM fuel cell model is developed that simulates both CO poisoning and oxygen bleeding, and obtains excellent agreement with published, experimental data. Modelling efforts indicate that CO adsorption and desorption follow a Temkin model. Increasing operating pressure or temperature mitigates CO poisoning, while use of reformate fuel increases the severity of the poisoning effect. Although oxygen bleeding mitigates CO poisoning, an unrecoverable performance loss exists at high current densities due to competition for reaction sites between hydrogen adsorption and the heterogeneous catalysis of CO. Copyright © 2003 John Wiley & Sons, Ltd.     

Effects of the electrical resistances of the GDL in a PEM fuel cell

In most PEM fuel cell models, the electrical resistance of the gas diffusion layers (GDL) is neglected under the assumptions that the GDL electrical conductivity is orders of magnitude higher than the ionic conductivity of the membrane. Recently some modeling efforts have taken the effects of electrical resistance of the GDL into consideration [H. Meng, C.Y. Wang, Electron transport in PEFCs, J. Electrochem. Soc. 151 (2004) A358–A367; B.R. Sivertsen, N. Djilali, CFD-based modeling of proton exchange membrane fuel cells, J. Power Sources 141 (2005) 65–78] and some of the results showed that under certain conditions, this effect was significant enough to alter the characteristics of current density distributions under the gas channels and the land areas. If these results are applicable to real-life fuel cells, the present design criteria and optimization procedures must be significantly changed to incorporate the effect of GDL electrical resistance. To examine this issue closely, a three-dimensional fuel cell model incorporating electron transport in the GDL is developed and employed to investigate the effect of electrical resistance through the GDL. In this model, the anisotropic nature of the GDL is taken into consideration by using different electrical conductivities in the through-plane and in-plane directions. The modeling results show that when realistic electrical conductivities for the GDL are used, the effect of the electrical resistance of GDL is slight and can be neglected for all industrial applications. It is believed that the over-estimations of the GDL resistance were mainly caused by neglecting the anisotropic nature of the GDL and/or lumping the contact resistance indiscriminately into the GDL, thus overestimating the electrical resistance of the GDL in the in-plane direction. Besides taking into consideration of the electrical resistance of GDL, the present model also take into consideration of the electron transport in the catalyst layers. When realistic values of electrical conductivities are used for both the GDL and catalyst layers, there is no significant change in the characteristics of current density distribution across the land and channel.     

Physical degradation of membrane electrode assemblies undergoing freeze/thaw cycling: Micro-structure effects

The objective of this work is to investigate physical damage of polymer electrolyte fuel cell (PEFC) materials subjected to freeze/thaw cycling. Effects of membrane electrode assembly micro-structures (catalyst layer cracking, membrane thickness, and membrane reinforcement) and diffusion media with micro-porous layers were analyzed by comparing scanning electron microscopy images of freeze/thaw cycled samples (−40 °C/70 °C) with those of virgin material and thermal cycled samples without freezing (5 °C/70 °C). Ex situ testing performed in this study has revealed a strong direction for the material choices in the PEFC and confirmed the previous computational model in the literature [S. He, M.M. Mench, J. Electrochem. Soc., 153 (2006) A1724–A1731; S. He, S.H. Kim, M.M. Mench, J. Electrochem. Soc., in press]. Specifically, the membrane electrode assemblies were found to be a source of water that can damage the catalyst layers under freeze/thaw conditions. Damage was found to occur almost exclusively under the channel, and not under the land (the graphite that touches the diffusion media). Conceptually, the best material to mitigate freeze-damage is a crack free virgin catalyst layer on a reinforced membrane that is as thin as possible, protected by a stiff diffusion media.     

Carbon nanofibers improve both the electronic and ionic contributions of the electrochemical performance of composite electrodes

The impact of carbon nanofibers as conductive additive on the electrochemical performance of a LiFePO₄-based composite electrode for lithium battery is investigated here. We use a new method that allows discriminating between the electronic and the ionic wirings contributions to the polarization and the specific capacity at different discharge rates [C. Fongy, et al., J. Electrochem. Soc., 157 (2010) A885; C. Fongy, et al., J. Electrochem. Soc., 157 (2010) A1347]. Results show this conductive additive is not only beneficial in terms of electronic wiring but it also enables to reach better high rate performance by improving the ionic wiring as it decreases the tortuosity of the porosity within the composite electrode architecture.     

Enhancement of the performance and reliability of CO poisoned PEM fuel cells

CO poisoning is a major issue when reformate is used as a fuel in PEM fuel cells. Normally, it is necessary to reduce the CO to very low levels (∼5 ppm) and to use CO tolerant catalysts, such as Pt–Ru alloys. As an alternative approach, we have studied the use of pulsed oxidation for the regeneration of CO poisoned cells. Results are presented for the regeneration of Pt and Pt–Ru anodes in a PEM fuel cell fed with CO concentrations as high as 10,000 ppm. The results show that periodic removal of CO from the catalyst surface by pulsed oxidation can increase the average cell potential and overall efficiency.Although use of pulsed techniques has been studied before, the careful control of each cell's voltage that this approach requires has limited its use in large fuel cell stacks. When uniform pulsing is done on a stack of fuel cells in series, the variations in voltage across the cells can limit the usefulness of this approach. A novel method that allows each cell in a stack to be separately pulsed under controlled conditions has been developed to overcome this problem. Weak or defective cells in a fuel cell stack can also be supplemented to enhance the power output and reliability of fuel cells. We present the results of experiments and calculations that quantify these benefits, specifically as they relate to PEM fuel cells operating on impure hydrogen produced by reforming fuels.     

A three-dimensional thermal abuse model for lithium-ion cells

To understand further the thermal abuse behavior of large format Li-ion batteries for automotive applications, the one-dimensional modeling approach formulated by Hatchard et al. [T.D. Hatchard, D.D. MacNeil, A. Basu, J.R. Dahn, J. Electrochem. Soc. 148(7) (2001) A755–A761] was reproduced. Then it was extended to three dimensions so we could consider the geometrical features, which are critical in large cells for automotive applications. The three-dimensional model captures the shapes and dimensions of cell components and the spatial distributions of materials and temperatures, and is used to simulate oven tests, and to determine how a local hot spot can propagate through the cell. In simulations of oven abuse testing of cells with cobalt oxide cathode and graphite anode with standard LiPF₆ electrolyte, the three-dimensional model predicts that thermal runaway will occur sooner or later than the lumped model, depending on the size of the cell. The model results showed that smaller cells reject heat faster than larger cells; this may prevent them from going into thermal runaway under identical abuse conditions. In simulations of local hot spots inside a large cylindrical cell, the three-dimensional model predicts that the reactions initially propagate in the azimuthal and longitudinal directions to form a hollow cylinder-shaped reaction zone.     

Transient analysis of carbon monoxide poisoning and oxygen bleeding in a PEM fuel cell anode catalyst layer

The presence of carbon monoxide in the fuel stream hinders the performance of a polymer electrolyte membrane (PEM) fuel cell, known as carbon monoxide (CO) poisoning. Introducing oxygen in the fuel stream lessens CO poisoning. Since CO poisoning is a phenomenon that occurs over a substantial period of time, a transient model has been developed in this study, taking into account the effect of CO concentration, operating pressure and temperature, as well as oxygen bleeding on the performance of the cell. It is found that at a lower CO concentration the poisoning effect takes a much longer time to reach the steady state, even though for a better steady state anode performance. A higher operating temperature results in a better steady state performance, but the performance drops faster toward the steady state value at higher temperature. A higher operating pressure leads to an enhanced performance over the entire transient history, although the benefit diminishes as pressure is increased. Even with a small amount of oxygen (0.5%) introduced into the fuel stream, the anode performance can be improved significantly. Finally, it is observed that the use of pure hydrogen interspersed in carbon monoxide containing fuel improves the anode performance. However, performance recovery when operating on pure hydrogen is much slower than the performance degradation due to the CO poisoning.     

H2O chemisorption and H2 oxidation on yttria-stabilized zirconia: Density functional theory and temperature-programmed desorption studies

The mechanism of H₂O dissociation as well as the adsorption and oxidation reaction of H₂ on yttria-stabilized zirconia (YSZ), commonly used as part of solid oxide fuel cell (SOFC) anodes, was investigated employing temperature-programmed desorption (TPD) spectroscopy and density functional theory (DFT). In agreement with theory the experimental results show that interaction of gaseous H₂O with YSZ results in dissociative adsorption leading to strongly bound OH surface species. In the interaction of gaseous H₂ with an oxygen-enriched YSZ surface (YSZ + O) similar OH surface species are formed as reaction intermediates in the H₂ oxidation. Our experiments showed that in both the H₂O/YSZ and the H₂/YSZ + O heterogeneous reaction systems noticeable amounts of H₂O are “dissolved” in the bulk as interstitial hydrogen and hydroxyl species. The experimental H₂O desorption data is used to access the accuracy of the H₂/H₂O/YSZ adsorption/desorption and surface reaction kinetics data, employed in previous modeling studies of the electrochemical H₂ oxidation on Ni-pattern/YSZ model anodes by Vogler et al. [J. Electrochem. Soc., 156 (2009) B663] and Goodwin et al. [J. Electrochem. Soc., 156 (2009) B1004]. Finally a refined experimentally validated H₂/H₂O/YSZ adsorption/desorption and surface reaction kinetics data set is presented.     

Measurement and modelling of carbon monoxide poisoning distribution within a polymer electrolyte fuel cell

The distribution of carbon monoxide (CO) across the anode of a polymer electrolyte fuel cell with a single channel flow-field is modelled and compared with experimental results obtained using localised stripping voltammetry. Good agreement is observed between experiment and model over a wide range of CO carrier gas flow rates. The model is used to predict the effect of CO transient poisoning, expected to occur during system start-up, when high CO slippage is likely to occur from the fuel processor and the fuel cell is not electrically loaded.     

Experimental factors that influence carbon monoxide tolerance of high-temperature proton-exchange membrane fuel cells

The poisoning effect of carbon monoxide (CO) on high-temperature proton-exchange membrane fuel cells (PEMFCs) is investigated with respect to CO concentration, operating temperature, fuel feed mode, and anode Pt loading. The loss in cell voltage when CO is added to pure hydrogen anode gas is a function of fuel utilization and anode Pt loading as well as obvious factors such as CO concentration, temperature and current density. The tolerance to CO can be varied significantly using a different experimental design of fuel utilization and anode Pt loading. A difference in cell performance with CO-containing hydrogen is observed when two cells with different flow channel geometries are used, although the two cells show similar cell performance with pure hydrogen. A different combination of fuel utilization, anode Pt loading and flow channel design can cause an order of magnitude difference in CO tolerance under identical experimental conditions of temperature and current density.     

Water equilibria and management using a two-volume model of a polymer electrolyte fuel cell

In this paper, we introduce a modified interpretation of the water activity presented in Springer et al. [T.E. Springer, T.A. Zawodzinski, S. Gottesfeld, Polymer electrolyte fuel cell model, J. Electrochem. Soc. 138 (8) (1991) 2334–2342]. The modification directly affects the membrane water transport between the anode and the cathode (two electrodes) of the polymer electrolyte membrane (PEM) fuel cell in the presence of liquid water inside the stack. The modification permits calibration of a zero-dimensional isothermal model to predict the flooding and drying conditions in the two electrodes observed at various current levels [D. Spernjak, S. Advani, A.K. Prasad, Experimental investigation of liquid water formation and transport in a transparent single-serpentine PEM fuel cell, in: Proceedings of the Fourth International Conference on Fuel Cell Science, Engineering and Technology (FUELCELL2006-97271), June 2006]. Using this model the equilibria of the lumped water mass in the two electrodes are analyzed at various flow conditions of the stack to determine stable and unstable (liquid water growth) operating conditions. Two case studies of water management through modification of cathode inlet humidification and anode water removal are then evaluated using this model. The desired anode water removal and the desired cathode inlet humidification are specified based upon (i) the water balance requirements, (ii) the desired conditions in the electrodes, and (iii) the maximum membrane transport at those conditions.     

Carbon monoxide poisoning of proton exchange membrane fuel cells

Proton exchange membrane fuel cell (PEMFC) performance degrades when carbon monoxide (CO) is present in the fuel gas; this is referred to as CO poisoning. This paper investigates CO poisoning of PEMFCs by reviewing work on the electrochemistry of CO and hydrogen, the experimental performance of PEMFCs exhibiting CO poisoning, methods to mitigate CO poisoning and theoretical models of CO poisoning. It is found that CO poisons the anode reaction through preferentially adsorbing to the platinum surface and blocking active sites, and that the CO poisoning effect is slow and reversible. There exist three methods to mitigate the effect of CO poisoning: (i) the use of a platinum alloy catalyst, (ii) higher cell operating temperature and (iii) introduction of oxygen into the fuel gas flow. Of these three methods, the third is the most practical. There are several models available in the literature for the effect of CO poisoning on a PEMFC and from the modeling efforts, it is clear that small CO oxidation rates can result in much increased performance of the anode. However, none of the existing models have considered the effect of transport phenomena in a cell, nor the effect of oxygen crossover from the cathode, which may be a significant contributor to CO tolerance in a PEMFC. In addition, there is a lack of data for CO oxidation and adsorption at low temperatures, which is needed for detailed modeling of CO poisoning in PEMFCs. Copyright © 2001 John Wiley & Sons, Ltd.     

Characteristic studies of a PBI/H3PO4 high temperature membrane PEMFC under simulated reformate gases

A high temperature proton exchange membrane fuel cell is considered a solution to improve the cell performance under CO-contained hydrogen and to simplify the gas purification process of a reformate fuel cell system. In this study, polybenzimidazole-based phosphoric acid-doped fuel cells are studied under simulated reformate gases of different H₂, N₂ and CO concentrations. The experimental results show that the dilution effect of N₂ has a minor impact on the cell performance in absence of CO. However, the CO poisoning increases the charge transfer resistance and leads to a substantial performance drop. This work also reveals that increasing the operating temperature can effectively improve the CO tolerance by suppressing the Pt–CO binding reaction. In addition, the CO poisoning effect becomes more significant in diluted H₂. As a result, the CO concentration should be maintained lower than a critical level to prevent a high CO coverage on the catalyst which leads to a noteworthy voltage shut-down, especially in highly diluted H₂.     

Transient evolution of carbon monoxide poisoning effect of PBI membrane fuel cells

High temperature polybenzimidazole membrane fuel cells are the focus of attention due to high CO tolerance and overcoming water managements. This paper develops a transient, one-dimensional mathematical model to predict CO tolerance, and validates it with experiments. Experimental results are measured at different temperatures. Fuel cell performance degradations with time are also measured under various fuel compositions. Transient evolutions of current density, H₂ coverage, CO coverage, and ionic potential are shown during the CO poisoning process. The theoretical results show that hydrogen coverage decreases with time, reducing hydrogen oxidation reactions and dropping ionic potential loss. The effects of temperature, CO contents, and H₂ dilutions on fuel cell performance and the time to reach steady t_ss are all investigated. Predictions of fuel cell current density degradation also show good agreement with experimental results.