Comparison of single-cell testing,short-stack testing and mathematical modeling methods for a direct methanol fuel cell

In this paper, a comparison between direct methanol fuel cell (DMFC) measurements performed on a single cell and a short-stack, and the results of a mathematical model for a DMFC, is presented. The testing of a short-stack, which consists of 5 cells with an active area of 315 cm², was performed at various current densities, permeation current densities, and cathode flow rates (CFR) in order to determine the voltage outputs of each cell. Methanol concentration and stack temperature results obtained from short-stack testing were then integrated into the single cell test and single cell mathematical model as the input parameters. For the mathematical modelling, transport equations originating from methanol, water, and oxygen were coupled with the electrochemical relations. Therefore, a comparison between these three methods is made in order to gain a deeper understanding of the effects of the operating parameters on DMFC performance. This study showed that the model could describe experimental results well when lower methanol concentrations (under 1.2 M) and temperature (under 60 °C) values are used as input parameters. The results also show very good agreement at lower methanol permeation rates and therefore lower temperatures. It is found that the voltage output for a given current density is higher for the theoretical model than that of the experimental studies; and the differences in the results can be up to 0.04 V for a cell.     

Real time measurements of methanol crossover in a DMFC

Methanol crossover is a serious problem in a direct methanol fuel cell (DMFC), which causes significant voltage loss and waste of fuel. Due to methanol crossover, most DMFCs must operate on a fuel with a very low methanol concentration; yet very low methanol concentration also causes a poor cell performance. Thus, it is very important to find the optimal operating conditions of methanol concentration and other operating parameters. In this research, methanol crossover rate in a DMFC is determined by measuring the carbon dioxide concentration at the cathode exit in real time. By measuring methanol crossover and cell performances at different inlet methanol concentrations and various operating conditions three types of characteristics are identified in the relationships between methanol crossover and cell current density. Further analysis of these relationships between methanol crossover and cell performances reveals the optimal methanol concentration and other operating parameters, at which the cell reaches optimal performance without incurring excessive methanol crossover. Furthermore, transient peaks of methanol crossover have been identified when the cell voltage suddenly changes. Analyses of these peaks show that they are caused by the hysteresis of methanol concentration at the interface between the anode catalyst layer and the membrane.     

One-dimensional and non-isothermal model for a passive DMFC

Passive direct methanol fuel cells (DMFCs) are promising energy sources for portable electronic devices. Different from DMFCs with active fuel feeding systems, passive DMFCs with nearly stagnant fuel and air tend to bear comparatively less power densities. A steady state, one-dimensional, multi-component and thermal model is described and applied to simulate the operation of a passive direct methanol fuel cell. The model takes into consideration the thermal and mass transfer effects, along with the electrochemical reactions occurring in the passive DMFC. The model can be used to predict the methanol, oxygen and water concentration profiles in the anode, cathode and membrane as well as to estimate the methanol and water crossover and the temperature profile across the cell. Polarization curves are numerically simulated and successfully compared with experiments for different methanol feed concentrations. The model predicts with accuracy the influence of the methanol feed concentration on the cell performance and the correct trends of the current density and methanol feed concentration, on methanol and water crossover. The model is rapidly implemented and is therefore suitable for inclusion in real-time system level DMFC calculations. Due to its simplicity the model can be used to help seek for possibilities of optimizing the cell performance of a passive DMFC by studying impacts from variations of the design parameters such as membrane thickness, catalyst loading, diffusion layers type and thicknesses.     

Modelling and analysis of a direct methanol fuel cell with under-rib mass transport and two-phase flow at the anode

For the past decade, extensive mathematical modelling has been conducted on the design and optimization of liquid-feed direct methanol fuel cells (DMFCs). Detailed modelling of DMFC operations reveals that a two-phase flow phenomenon at the anode and under-rib convection due to the pressure difference between the adjacent channels both contribute significantly to mass-transfer in a DMFC and its output performance. In practice, comprehensive simulations based on the finite volume technique for two-phase flow require a high level of numerical complexity in computation. This study presents a complexity-reduced mathematical model that is developed to cover both phenomena for a realistic, but fast, in computation for the prediction and analysis of a DMFC prototype design. The simulation results are validated against experimental data with good agreement. Analysis of the DMFC mass-transfer is made to investigate methanol distribution at anode and its crossover through the proton-exchange membrane. From a comparison of the influence of two-phase flow and under-rib mass-transfer on DMFC performance, the significance of gas-phase methanol transport is established. Simulation results suggest that both the optimization of the flow-field structure and the fuel cell operating parameters (flow rate, methanol concentration and operating temperature) are important factors for competitive DMFC performance output.     

Mass transport analysis of a passive vapor-feed direct methanol fuel cell

A two-dimensional, two-phase, non-isothermal model using the multi-fluid approach was developed for a passive vapor-feed direct methanol fuel cell (DMFC). The vapor generation through a membrane vaporizer and the vapor transport through a hydrophobic vapor transport layer were both considered in the model. The evaporation/condensation of methanol and water in the diffusion layers and catalyst layers was formulated considering non-equilibrium condition between phases. With this model, the mass transport in the passive vapor-feed DMFC, as well as the effects of various operating parameters and cell configurations on the mass transport and cell performance, were numerically investigated. The results showed that the passive vapor-feed DMFC supplied with concentrated methanol solutions or neat methanol can yield a similar performance with the liquid-feed DMFC fed with much diluted methanol solutions, while also showing a higher system energy density. It was also shown that the mass transport and cell performance of the passive vapor-feed DMFC depend highly on both the open area ratio of the vaporizer and the methanol concentration in the tank.     

Sensor-less control of methanol concentration based on estimation of methanol consumption rates for direct methanol fuel cell systems

Adequate control over the concentration of methanol is critically needed in operating direct methanol fuel cell (DMFC) systems, because performance and energy efficiency of the systems are primarily dependent on the concentration of methanol feed. For this purpose, we have built a sensor-less control logic that can operate based on the estimation of the rates of methanol consumption in a DMFC. The rates of methanol consumption are measured in a cell and the resulting data are fed as an input to the control program to calculate the amount of methanol required to maintain the concentration of methanol at a set value under the given operating conditions of a cell. The sensor-less control has been applied to a DMFC system employed with a large-size single cell and the concentration of methanol is found to be controlled stably to target concentrations even though there are some deviations from the target values.     

Water management in a passive direct methanol fuel cell

Passive direct methanol fuel cells (DMFCs) are under development for use in portable applications because of their enhanced energy density in comparison with other fuel cell types. The most significant obstacles for DMFC development are methanol and water crossover because methanol diffuses through the membrane generating heat but no power. The presence of a large amount of water floods the cathode and reduces cell performance. The present study was carried out to understand the performance of passive DMFCs, focused on the water crossover through the membrane from the anode to the cathode side. The water crossover behaviour in passive DMFCs was studied analytically with the results of a developed model for passive DMFCs. The model was validated with an in‐house designed passive DMFC. The effect of methanol concentration, membrane thickness, gas diffusion layer material and thickness and catalyst loading on fuel cell performance and water crossover is presented. Water crossover was lowered with reduction on methanol concentration, reduction of membrane thickness and increase on anode diffusion layer thickness and anode and cathode catalyst layer thickness. It was found that these conditions also reduced methanol crossover rate. A membrane electrode assembly was proposed to achieve low methanol and water crossover and high power density, operating at high methanol concentrations. The results presented provide very useful and actual information for future passive DMFC systems using high concentration or pure methanol. Copyright © 2012 John Wiley & Sons, Ltd.     

Methanol and water crossover in a passive liquid-feed direct methanol fuel cell

Methanol crossover, water crossover, and fuel efficiency for a passive liquid-feed direct methanol fuel cell (DMFC) were all experimentally determined based on the mass balance of the cell discharged under different current loads. The effects of different operating conditions such as current density and methanol concentration, as well as the addition of a hydrophobic water management layer, on the methanol and water crossover were investigated. Different from the active DMFC, the cell temperature of the passive DMFC increased with the current density, and the changes of methanol and water crossover with current density were inherently coupled with the temperature rise. When feeding with 2–4 M methanol solution, with an increase in current density, both the methanol crossover and the water crossover increased, while the fuel efficiency first increased but then decreased slightly. The results also showed that a reduction of water crossover from the anode to the cathode was always accompanied with a reduction of methanol crossover. Not only did the water management layer result in lower water crossover or achieve neutral or reverse water transport, but it also lowered the methanol crossover and increased the fuel efficiency.     

Two-dimensional two-phase thermal model for passive direct methanol fuel cells

A two-dimensional two-phase thermal model is presented for direct methanol fuel cells (DMFC), in which the fuel and oxidant are fed in a passive manner. The inherently coupled heat and mass transport, along with the electrochemical reactions occurring in the passive DMFC is modeled based on the unsaturated flow theory in porous media. The model is solved numerically using a home-written computer code to investigate the effects of various operating and geometric design parameters, including methanol concentration as well as the open ratio and channel and rib width of the current collectors, on cell performance. The numerical results show that the cell performance increases with increasing methanol concentration from 1.0 to 4.0 M, due primarily to the increased operating temperature resulting from the exothermic reaction between the permeated methanol and oxygen on the cathode and the increased mass transfer rate of methanol. It is also shown that the cell performance upgrades with increasing the open ratio and with decreasing the rib width as the result of the increased mass transfer rate on both the anode and cathode.     

1D modeling of a flowing electrolyte-direct methanol fuel cell

In this study, the performance characteristics of a flowing electrolyte-direct methanol fuel cell (FE-DMFC) and a direct methanol fuel cell (DMFC) are evaluated by computer simulations; and results are compared to experimental data found in the literature. Simulations are carried out to assess the effects of the operating parameters on the output parameters; namely, methanol concentration distribution, cell voltage, power density, and electrical efficiency of the cell. The operating parameters studied include the electrolyte flow rate, flowing electrolyte channel thickness, and methanol concentration at the feed stream. In addition, the effect of the circulation of the flowing electrolyte channel outlet stream on the performance is discussed. The results show that the maximum power densities that could be achieved do not significantly differ between these two fuel cells; however the electrical efficiency could be increased by 57% when FE-DMFC is used instead of DMFC.     

Development of an advanced MEA to use high-concentration methanol fuel in a direct methanol fuel cell system

Despite serious methanol crossover issues in Direct Methanol Fuel Cells (DMFCs), the use of high-concentration methanol fuel is highly demanded to improve the energy density of passive fuel DMFC systems for portable applications. In this paper, the effects of a hydrophobic anode micro-porous layer (MPL) and cathode air humidification are experimentally studied as a function of the methanol-feed concentration. It is found in polarization tests that the anode MPL dramatically influences cell performance, positively under high-concentration methanol-feed but negatively under low-concentration methanol-feed, which indicates that methanol transport in the anode is considerably altered by the presence of the anode MPL. In addition, the experimental data show that cathode air humidification has a beneficial effect on cell performance due to the enhanced backflow of water from the cathode to the anode and the subsequent dilution of the methanol concentration in the anode catalyst layer. Using an advanced membrane electrode assembly (MEA) with the anode MPL and cathode air humidification, we report that the maximum power density of 78 mW/cm² is achieved at a methanol-feed concentration of 8 M and cell operating temperature of 60 °C. This paper illustrates that the anode MPL and cathode air humidification are key factors to successfully operate a DMFC with high-concentration methanol fuel.     

A one-dimensional,two-phase model for direct methanol fuel cells – Part I: Model development and parametric study

A one-dimensional, steady-state, two-phase direct methanol fuel cell (DMFC) model is developed to precisely investigate complex physiochemical phenomena inside DMFCs. In this model, two-phase species transport through the porous components of a DMFC is formulated based on Maxwell–Stefan multi-component diffusion equations, while capillary-induced liquid flow in the porous media is described by Darcy's equation. In addition, the model fully accounts for water and methanol crossover through the membrane, which is driven by the effects of electro-osmotic drag, diffusion, and the hydraulic pressure gradient. The developed model is validated against readily available experimental data in the literature. Then, a parametric study is carried out to investigate the effects of the operating temperature, methanol feed concentration, and properties of the backing layer. The results of the numerical simulation clarify the detailed influence of these key designs and operating parameters on the methanol crossover rate as well as cell performance and efficiency. The results emphasize that the material properties and design of the anode backing layer play a critical role in the use of highly concentrated methanol fuel in DMFCs. The present study forms a theoretical background for optimizing the DMFC's components and operating conditions.     

A three-dimensional mathematical model for liquid-fed direct methanol fuel cells

A three-dimensional, single-phase, multi-component mathematical model has been developed for a liquid-fed direct methanol fuel cell (DMFC). The traditional continuity, momentum, and species conservation equations are coupled with electrochemical kinetics in both the anode and cathode catalyst layer. At the anode side, the liquid phase is considered, and at the cathode side only the gas phase is considered. Methanol crossover due to both diffusion and electro-osmotic drag from the anode to the cathode is taken into consideration and the effect is incorporated into the model using a mixed-potential at the cathode. A finite-volume-based CFD technique is used to develop the in-house numerical code and the code is successfully used to simulate the fuel cell performance as well as the multi-component behavior in a DMFC. The modeling results of polarization curves compare well with our experimental data. Subsequently, the model is used to study the effects of methanol crossover, the effects of porosities of the diffusion layer and the catalyst layer, the effects of methanol flow rates, and the effects of the channel shoulder widths.     

Study of pulsed-current loading of direct methanol fuel cells using a new time-domain model based on bi-functional methanol oxidation kinetics

This work describes a non-linear time-domain model of a direct methanol fuel cell (DMFC) and uses that model to show that pulsed-current loading of a direct methanol fuel cell does not improve average efficiency. Unlike previous system level models, the one presented here is capable of predicting the step response of the fuel cell over its entire voltage range. This improved model is based on bi-functional methanol oxidation reaction kinetics and is derived from a lumped, four-step reaction mechanism. In total, six states are incorporated into the model: three states for intermediate surface adsorbates on the anode electrode, two states for the anode and cathode potentials, and one state for the liquid methanol concentration in the anode compartment. Model parameters were identified using experimental data from a real DMFC. The model was applied to study the steady-state and transient performance of a DMFC with the objective to understand the possibility of improving the efficiency of the DMFC by using periodic current pulses to drive adsorbed CO from the anode catalyst. Our results indicate that the pulsed-current method does indeed boost the average potential of the DMFC by 40 mV; but on the other hand, executing that strategy reduces the overall operating efficiency and does not yield any net benefit.     

Modeling of a passive DMFC operating with neat methanol

A mathematical model is developed to simulate the fundamental transport phenomena in a passive direct methanol fuel cell (DMFC) operating with neat methanol. The neat methanol operation is realized by using a ‘pervaporation’ membrane that allows the methanol concentration from the neat methanol in the fuel reservoir to be declined to an appropriate level in the anode catalyst layer (CL). The water required by the methanol oxidation reaction on the anode is passively obtained by diffusion from the cathode through the membrane. The numerical results indicate that the methanol delivery rate from the fuel reservoir to the anode CL is predominately controlled by the pervaporation process. It is also found that under the neat methanol operating condition, water distribution across the membrane electrode assembly is greatly influenced by the membrane thickness, the cathode design, the operating temperature, and the ambient relative humidity.     

Comparison of numerical simulation results with experimental current density and methanol-crossover data for direct methanol fuel cells

The simulation results of a one-dimensional (1D) direct methanol fuel cell (DMFC) model are compared with the current density and methanol-crossover data that are experimentally measured under several different cell designs and operating conditions. No fitting parameters are employed for the comparison and model input parameters obtained from the literature are consistently used for all the cases of comparison. The numerical predictions agree well with the experimental data and the 1D DMFC model successfully captures key experimental trends that are observed in the cell current density and methanol-crossover data. This clearly illustrates that the present DMFC model can be applicable for optimizing DMFC component designs and operating conditions. In addition, the model simulations further indicate that the reduction of the methanol concentration in the anode catalyst layer is critical to simultaneously suppress both the electro-osmotic drag (EOD) and the diffusion aspects of methanol crossover.     

Modeling of a 5-cell direct methanol fuel cell using adaptive-network-based fuzzy inference systems

The methanol concentrations, temperature and current were considered as inputs, the cell voltage was taken as output, and the performance of a direct methanol fuel cell (DMFC) was modeled by adaptive-network-based fuzzy inference systems (ANFIS). The artificial neural network (ANN) and polynomial-based models were selected to be compared with the ANFIS in respect of quality and accuracy. Based on the ANFIS model obtained, the characteristics of the DMFC were studied. The results show that temperature and methanol concentration greatly affect the performance of the DMFC. Within a restricted current range, the methanol concentration does not greatly affect the stack voltage. In order to obtain higher fuel utilization efficiency, the methanol concentrations and temperatures should be adjusted according to the load on the system.     

Performance characterization of passive direct methanol fuel cells

The performance of a fuel cell is usually characterized by a polarization curve (cell voltage versus current density) under stabilized operating conditions. However, for passive direct methanol fuel cells (DMFC) that have neither fuel pumps nor gas compressors, the voltage at a given current density varies with time because methanol concentration in the fuel reservoir keeps decreasing during the discharging process. The important question brought up by this transient discharging behavior is: under what conditions should the polarization data be collected such that the performance of the passive DMFC can be objectively characterized? In this work, we found that the performance of the passive DMFC became relatively stable as the cell operating temperature rose to a relatively stable value. This finding indicates that the performance of the passive DMFC can be characterized by collecting polarization data at the instance when the cell operating temperature under the open-circuit condition rises to a relatively stable value.     

Operational characteristics of the direct methanol fuel cell stack on fuel and energy efficiency with performance and stability

This paper is presented to investigate operational characteristics of a direct methanol fuel cell (DMFC) stack with regard to fuel and energy efficiency, including its performance and stability under various operating conditions. Fuel efficiency of the DMFC stack is strongly dependent on fuel concentration, working temperature, current density, and anode channel configuration in the bipolar plates and noticeably increases due to the reduced methanol crossover through the membrane, as the current density increases and the methanol concentration, anode channel depth, and temperature decreases. It is, however, revealed that the energy efficiency of the DMFC stack is not always improved with increased fuel efficiency, since the reduced methanol crossover does not always indicate an increase in the power of the DMFC stack. Further, a lower methanol concentration and temperature sacrifice the power and operational stability of the stack with the large difference of cell voltages, even though the stack shows more than 90% of fuel efficiency in this operating condition. The energy efficiency is therefore a more important characteristic to find optimal operating conditions in the DMFC stack than fuel efficiency based on the methanol utilization and crossover, since it considers both fuel efficiency and cell electrical power. These efforts may contribute to commercialization of the highly efficient DMFC system, through reduction of the loss of energy and fuel.     

Non-isothermal dynamic modelling and optimization of a direct methanol fuel cell

A non-isothermal dynamic optimization model of direct methanol fuel cells (DMFCs) is developed to predict their performance with an effective optimum-operating strategy. After investigating the sensitivities of the transient behaviour (the outlet temperature, crossovers of methanol and water, and cell voltage) to operating conditions (the inlet flow rates into anode and cathode compartments, and feed concentration) through dynamic simulations, we find that anode feed concentration has a significantly larger impact on methanol crossover, temperature, and cell voltage than the anode and cathode flow rates. Also, optimum transient conditions to satisfy the desired fuel efficiency are obtained by dynamic optimization. In the developed model, the significant influence of temperature on DMFC behaviour is described in detail with successful estimation of its model parameters.