Membrane electrode assembly degradation by dry/wet gas on a PEM fuel cell

The vehicle that use a Polymer Electrolyte Membrane Fuel Cell (PEMFC) as a power source frequently experiences start up and shut down. Membrane Electrode Assembly (MEA) degradation by wet/dry gas repetition was studied for vehicle start up and shut down. The time of the wet/dry equilibrium state on the PEMFC was measured with High Frequency Resistance (HFR). The gas injection time was 20 min and 5 min for dry gas and wet gas, respectively. An experiment was carried out using electrochemical methods and a cross-section of the MEA was visualized with a Field Emission Scanning Electron Microscope (FE-SEM). After 1200 wet/dry cycles, the performance of the cell decreased by 45.7% to its current density of 800 mA/cm². Ohmic and charge transfer resistances of the cell increased in the Electrochemical Impedance Spectroscopy (EIS). The crossover current of hydrogen also increased in the linear sweep voltammetry (LSV). The reduction of the electrochemical active surface area (ECSA) was confirmed through cyclic voltammetry (CV). The interface among the membrane, catalyst layer, and gas diffusion layer was separated and significantly deteriorated compared with fresh MEA.     

Performance test and degradation analysis of direct methanol fuel cell membrane electrode assembly during freeze/thaw cycles

Performance and degradation of direct methanol fuel cell (DMFC) membrane electrode assembly (MEA) are analyzed after repeated freeze/thaw cycles. Three different MEAs stored at −20 °C for 8 h with the anode side full of methanol solution are selected to test the effects of low temperatures on performance. After the cell heated to 60 °C within 30 min, they are inspected to determine the degradation mechanism. The resistance R obtained by the polarization curve is essential for identifying the main component affecting cell performance. The electrochemical impedance spectroscopy (EIS) technique is used to characterize the DMFC after freeze/thaw cycles. Thus, deterioration is assessed by measuring the high-frequency resistance (HFR) and the charge-transfer resistance (CTR). The electrochemical surface area (ECA) is employed to investigate not only the actual chemical degradation but also membrane status since sudden loss of ECA on the cathode side can result from a broken membrane. Moreover, a strategy is designed to simulate actual conditions that may prevent the membrane from being broken. A DMFC stack without any heating or heat-insulation devices shall avoid to be stored at subzero temperatures since the membrane will be useless due to frozen of methanol solution.     

Effects of reverse voltage and subzero startup on the membrane electrode assembly of a PEMFC

Effects of reverse voltage and frozen fuel cell startup on the membrane electrode assembly (MEA) were investigated for a proton exchange membrane fuel cell (PEMFC). A single cell was started from a subzero temperature by applying reverse voltage. The voltages applied to the cell were 0.8 and 1.2 V. The fuel cell performance was measured with a polarization curve and by cyclic voltammetry (CV), electrochemical impedance spectra (EIS), linear scan voltammetry (LSV) after each experiment. From the results, it was concluded that the catalyst activity, electrochemical active surface area (ECA) and the membrane were not damaged by the reverse voltage if the voltage was below 0.85 V. In contrast, a reverse voltage improved cell performance slightly. If the reverse voltage was larger than 0.85 V, the cell performance degraded. Another single cell with an active area of 128 cm² was started up at −15 °C by applying reverse voltage. The cell performance and MEA physical characteristic were tested before and after the freeze startup. From the results, the cell performance decayed MEA delamination was observed and the pore size distribution of the MEA changed.     

Multi-mode control strategy for fuel cell electric vehicles regarding fuel economy and durability

A proton electrolyte membrane (PEM) fuel cell system and a Li-ion battery (LIB) are two power sources in a fuel cell electric vehicle (FCEV). The fuel cell system is composed of a fuel cell stack and subsystems for air/hydrogen supply and cooling water. The operation procedure of the fuel cell system can be generally separated into several processes, e.g. starting up, normal/abnormal working and shutting down. In this paper, a multi-mode real-time control strategy for a FCEV is proposed. The strategy is established based on three typical processes (starting up, normal working, shutting down) of the fuel cell system, taking the fuel economy and system durability into consideration. The strategy is applied into a platform vehicle for the 12th 5-year project of “the next generation technologies of fuel cell city buses”. Experiments of the “China city bus typical cycle” on a test bench for the bus were carried out. Results show that, the fuel economy is 7.6 kg (100 km)⁻¹ in the battery charge-sustaining status. In a practical situation, a total driving mileage of more than 270 km can be achieved. Cycle testing also showed that, the degradation rate of the fuel cell was reduced to half of the original level. No performance degradation of the LIB system was observed in the cycling test.     

Kinetic studies of cathode degradation on PEM fuel cell short stack level undergoing freeze startups with different states of residual water and current draws

One of the biggest challenges for a wide spread introduction of polymer electrolyte membrane fuel cells in automotive applications is the freeze start at subzero temperatures as this poses a severe threat to fuel cell performance and overall lifetime. Therefore, the impact of current draw during stack freeze startup at various rest water contents and current densities was investigated applying state of the art in-situ testing as quasi cyclic voltammetry.The results indicate a clear dependency between number of freeze startups and performance loss, whereas higher initial water content within the stack reduced its destructive impacts.In our earlier work we were able to show the dependency between residual water and degradation for non-freeze-startup-capable systems at the unit cell level [31]. With this work we confirm that the same physical relationship also applies to freeze-startup-capable systems on the short stack level.Furthermore, reversible performance losses that were encountered during this study can be assigned to oxidizable fuel contaminants, which are believed to be CO, and an easy cleansing procedure is being suggested.     

Redistribution of phosphoric acid in membrane electrode assemblies for high-temperature polymer electrolyte fuel cells

We demonstrate that the performance of a high-temperature polymer electrolyte fuel cell with a phosphoric acid-based electrolyte is almost independent of the way of introducing the acid into the membrane electrode assembly (MEA). The same power densities were obtained with different MEAs in which the poly(2,5-benzimidazole) membrane was either pre-doped or not and in which either one or two catalyst layers were impregnated with H₃PO₄. Chemical analysis after shut down revealed that in all these MEAs the phosphoric acid distribution between the membrane and the electrodes was nearly the same. An MEA with acid impregnation via the electrodes was started up rapidly from room temperature, delivered a power density of 120 mW cm⁻² at 600 mV (H₂/air, 160 °C, ambient pressure) after only 11 min and was operated for 1000 h (degradation rate: 0.06 mV/h). Based on the analysis of the H₃PO₄ content in the MEA components, reflections on the kinetics of the redistribution of phosphoric acid within the MEA are provided.     

A preliminary study of a miniature planar 6-cell PEMFC stack combined with a small hydrogen storage canister

The fabrication and performance evaluation of a miniature 6-cell PEMFC stack based on Micro-Electronic-Mechanical-System (MEMS) technology is presented in this paper. The stack with a planar configuration consists of 6-cells in serial interconnection by spot welding one cell anode with another cell cathode. Each cell was made by sandwiching a membrane-electrode-assembly (MEA) between two flow field plates fabricated by a classical MEMS wet etching method using silicon wafer as the original material. The plates were made electrically conductive by sputtering a Ti/Pt/Au composite metal layer on their surfaces. The 6-cells lie in the same plane with a fuel buffer/distributor as their support, which was fabricated by the MEMS silicon–glass bonding technology. A small hydrogen storage canister was used as fuel source. Operating on dry H₂ at a 40 ml min⁻¹ flow rate and air-breathing conditions at room temperature and atmospheric pressure, the linear polarization experiment gave a measured peak power of 0.9 W at 250 mA cm⁻² for the stack and average power density of 104 mW cm⁻² for each cell. The results suggested that the stack has reasonable performance benefiting from an even fuel supply. But its performance tended to deteriorate with power increase, which became obvious at 600 mW. This suggests that the stack may need some power assistance, from say supercapacitors to maintain its stability when operated at higher power.     

Influence of current densities in SOFC and PEFC stacks on a SOFC–PEFC combined system

A solid oxide fuel cell (SOFC)–polymer electrolyte fuel cell (PEFC) combined system was investigated by numerical simulation. Here, the effect of the current densities in the SOFC and the PEFC stacks on the system's performance is evaluated under a constant fuel utilization condition. It is shown that the SOFC–PEFC system has an optimal combination of current densities, for which the electrical efficiency is highest. The optimal combination exists because the cell voltage in one stack increases and that of the other stack decreases when the current densities are changed. It is clarified that there is an optimal size of the PEFC stack in the parallel-fuel-feeding-type SOFC–PEFC system from the viewpoint of efficiency, although a larger PEFC stack always leads to higher electrical efficiency in the series-fuel-feeding-type SOFC–PEFC system. The 40 kW-class PEFC stack is suitable for the 110 kW-class SOFC stack in the parallel-fuel-feeding type SOFC–PEFC system.     

Low temperature operation and influence parameters on the cold start ability of portable PEMFCs

The start up behaviour of PEM fuel cells below 0 °C is one of the most challenging tasks to be solved before commercialisation. The automotive industry started to develop solutions to reduce the start up time of fuel cell systems in the middle of the nineties. The strategies varied from catalytic combustion of hydrogen on the electrode catalyst to fuel starvation or external stack heating via cooling loops to increase the stack temperature.Beside the automotive sector the cold start ability is as well important for portable PEMFC applications for outdoor use. But here the cold start issue is even more complicated, as the fuel cell system should be operated as passive as possible.Below 0 °C freezing of water inside the PEMFC could form ice layers in the electrode and in the gas diffusion layer. Therefore the cell reaction is limited or even inhibited. Product water during the start up builds additional barriers and leads to a strong decay of the output power at isothermal operating conditions.In order to find out which operational and hardware parameters affect this decay, potentiostatic experiments on single cells were performed at isothermal conditions. These experiments comprise investigations of the influence of membrane thickness and different GDL types as well as the effect of gas flow rates and humidification levels of the membrane. As pre stage to physical based models, empirical based prediction models are used to gain a better understanding of the main influence parameters during cold start. The results are analysed using the statistical software Cornerstone 4.0.The experience of single cell investigations are compared to start up behaviour of portable fuel cell stacks which are operated in a climate chamber at different ambient temperatures below 0 °C. Additional flow sharing problems in the fuel cell stack could be seen during cold start up experiments.     

Development of a passive direct methanol fuel cell (DMFC) twin-stack for long-term operation

Direct methanol fuel cell (DMFC), with benefits such as high energy efficiency, quick start capability and instantaneous refueling, is a promising power source to meet the ever-increasing power demand for portable electronic products. In this paper, a novel CO₂-driven fuel-feed device was produced and equipped in a passive 8-cell DMFC twin-stack for long-term operation. It was shown that this fuel-feed device was capable of supplying methanol solution continuously in response to the change in discharging current of the stack. Stainless steel sheet was photochemically etched as current collectors based on MEMS techniques. Series interconnections between two neighbor cells were realized in banded configuration which avoided the external connection. TiN-plated mesh was placed between current collector and membrane electrode assembly (MEA), which was used to lessen the internal resistance of the stack. A peak power density of 16.9 mW cm⁻² was achieved with 4 M methanol at ambient temperature and passive operation. The stack equipped with the fuel feed device successfully powered a sensor node for 39 h with the consumption of 80 ml of 4 M methanol.     

PEM stack test and analysis in a power system at operational load via ac impedance

This paper presents a method for collecting ac impedance data from a commercial PEFC power system at operational loads. The PEM fuel cell stack in the power system, including 47 MEAs, was operated using room air and pure hydrogen (>99.99%). For a stack test in the power system, the power source for the embedded controller board was simultaneously switched from the fuel cell stack to a similar external power source after the system reached a steady temperature. The PEM fuel cells in the stack were tested by collecting the ac impedance data at different current levels. By using ac impedance, a single fuel cell, a group of fuel cells, and a complete stack were then tested without the embedded control devices for ohmic, activation, and mass transport losses. The ohmic resistance for each cell component in the stack was obtained as 117 mΩ cm² at operational loads from 2.5 A to 35 A. The membrane thickness was further estimated as ca. 51–89 μm. Resistances from ohmic conduction, anode/cathode activation, and mass transport were measured and discussed using the Nyquist plots from the ac impedance spectra.     

A small mono-polar direct methanol fuel cell stack with passive operation

A passive direct methanol fuel cell (DMFC) stack that consists of six unit cells was designed, fabricated, and tested. The stack was tested with different methanol concentrations under ambient conditions. It was found that the stack performance increased when the methanol concentration inside the fuel tank was increased from 2.0 to 6.0 M. The improved performance is primarily due to the increased cell temperature as a result of the exothermic reaction between the permeated methanol and oxygen on the cathode. Moreover, the increased cell temperature enhanced the water evaporation rate on the air-breathing cathode, which significantly reduced water flooding on the cathode and further improved the stack performance. This passive DMFC stack, providing 350 mW at 1.8 V, was successfully applied to power a seagull display kit. The seagull display kit can continuously run for about 4 h on a single charge of 25 cm³ 4.0-M methanol solution.     

Modelling and evaluation of heating strategies for high temperature polymer electrolyte membrane fuel cell stacks

Experiments were conducted on two different cathode air cooled high temperature PEM (HTPEM) fuel cell stacks; a 30 cell 400 W prototype stack using two bipolar plates per cell, and a 65 cell 1 kW commercial stack using one bipolar plate per cell. The work seeks to examine the use of different heating strategies and find a strategy suited for fast start-up of the HTPEM fuel cell stacks. Fast start-up of these high temperature systems enables use in a wide range of applications, such as automotive and auxiliary power units, where immediate system response is needed. The development of a dynamic model to simulate the temperature development of a fuel cell stack during heating can be used for assistance in system and control design. The heating strategies analyzed and tested reduced the start-up time of one of the fuel cell stacks from 1 h to about 6 min.     

Scale-up of a high temperature polymer electrolyte membrane fuel cell based on polybenzimidazole

A high temperature PEM fuel cell stack with a total active area 150 cm² has been studied. The PEM technology is based on a polybenzimidazole (PBI) membrane. Cast from a PBI polymer synthesised in our lab, the performance of a three-cell stack was analysed in static and dynamic modes. In static mode, operating at high constant oxygen flow rate (QO₂>1105 ml O₂/min) produces a small decrease on the stack performance. High constant oxygen stoichiometry (λO₂>3) does not produce a decrease on the performance of the stack. There are not differences between operating at constant flow rate of oxygen and constant stoichiometry of oxygen in the stack performance. The effect of operating at high temperature with a pressurization system and operating at higher temperatures are beneficial since the performance of the fuel cell is enhanced. A large shut-down stage produces important performance losses due to the loss of catalyst activity and the loss of membrane conductivity. After 150 h of operation at 0.2 A cm⁻², it is observed a very high voltage drop. The phosphoric acid leached from the stack was also evaluated and did not exceed 2% (w/w). This fact suggests that the main degradation mechanism of a fuel cell stack based on polybenzimidazole is not the electrolyte loss. In dynamic test mode, it was observed a rapid response of power and current output even at the lower step-time (10 s). In the static mode at 125 °C and 1 atm, the stack reached a power density peak of 0.29 W cm⁻² (43.5 W) at 1 V.     

Cold start dynamics and temperature sliding observer design of an automotive SOFC APU

This paper presents a dynamic model for studying the cold start dynamics and observer design of an auxiliary power unit (APU) for automotive applications. The APU is embedded with a solid oxide fuel cell (SOFC) stack which is a quiet and pollutant-free electric generator; however, it suffers from slow start problem from ambient conditions. The SOFC APU system equips with an after-burner to accelerate the start-up transient in this research. The combustion chamber burns the residual fuel (and air) left from the SOFC to raise the exhaust temperature to preheat the SOFC stack through an energy recovery unit. Since thermal effect is the dominant factor that influences the SOFC transient and steady performance, a nonlinear real-time sliding observer for stack temperature was implemented into the system dynamics to monitor the temperature variation for future controller design. The simulation results show that a 100 W APU system in this research takes about 2 min (in theory) for start-up without considering the thermal limitation of the cell fracture.     

Design,fabrication and performance analysis of a 200 W PEM fuel cell short stack

A PEM fuel cell short stack of 200 W capacity, with an active area of 100 cm² has been designed and fabricated in-house. The status of unit cell performance was 0.55 W cm⁻². Based on the unit cell technology, a short stack has been developed. The proper design of uniform flow distribution, cooling plate and compressed end plate were important to achieve the best performance of the short stack. The performance of four cells stack was analyzed in static and dynamic modes. In the static mode of polarization curve, the stack has peak power density of 0.55 W cm⁻² (220 W) at 0.5 V per cell, when the voltage was scanning from low to high voltage (1.5–3.5 V), and resulted in minimum water flooding inside the stack. In this study a series of dynamic loadings were tested to simulate the vehicle acceleration. The fuel cell performances respond to dynamic loading influenced by the hydrogen/air stoichiometric, back pressure, and dynamic-loading time. It was needed high hydrogen stoichiometric and back pressure to maintain high dynamic performance. In the long-time stable power testing, the stack was difficult to maintain at high performance, due to the water flooding at high output power. An adjusting cathode back-pressure method for purging water was proposed to prevent the water flooding at flow channels and maintain the stable output power at 170 W (0.42 W cm⁻²).     

Modeling electrochemical performance in large scale proton exchange membrane fuel cell stacks

The processes, losses, and electrical characteristics of a Membrane-Electrode Assembly (MEA) of a Proton Exchange Membrane Fuel Cell (PEMFC) are described. In addition, a technique for numerically modeling the electrochemical performance of a MEA, developed specifically to be implemented as part of a numerical model of a complete fuel cell stack, is presented. The technique of calculating electrochemical performance was demonstrated by modeling the MEA of a 350 cm², 125 cell PEMFC and combining it with a dynamic fuel cell stack model developed by the authors. Results from the demonstration that pertain to the MEA sub-model are given and described. These include plots of the temperature, pressure, humidity, and oxygen partial pressure distributions for the middle MEA of the modeled stack as well as the corresponding; current produced by that MEA. The demonstration showed that models developed using this technique produce results that are reasonable when compared to established performance expectations and experimental results.     

Characterization of a novel, highly integrated tubular solid oxide fuel cell system using high-fidelity simulation tools

A novel, highly integrated tubular SOFC system intended for small-scale power is characterized through a series of sensitivity analyses and parametric studies using a previously developed high-fidelity simulation tool. The high-fidelity tubular SOFC system modeling tool is utilized to simulate system-wide performance and capture the thermofluidic coupling between system components. Stack performance prediction is based on 66 anode-supported tubular cells individually evaluated with a 1-D electrochemical cell model coupled to a 3-D computational fluid dynamics model of the cell surroundings. Radiation is the dominate stack cooling mechanism accounting for 66-92% of total heat loss at the outer surface of all cells at baseline conditions. An average temperature difference of nearly 125 °C provides a large driving force for radiation heat transfer from the stack to the cylindrical enclosure surrounding the tube bundle. Consequently, cell power and voltage disparities within the stack are largely a function of the radiation view factor from an individual tube to the surrounding stack can wall. The cells which are connected in electrical series, vary in power from 7.6 to 10.8 W (with a standard deviation, σ = 1.2 W) and cell voltage varies from 0.52 to 0.73 V (with σ = 81 mV) at the simulation baseline conditions. It is observed that high cell voltage and power outputs directly correspond to tubular cells with the smallest radiation view factor to the enclosure wall, and vice versa for tubes exhibiting low performance. Results also reveal effective control variables and operating strategies along with an improved understanding of the effect that design modifications have on system performance. By decreasing the air flowrate into the system by 10%, the stack can wall temperature increases by about 6% which increases the minimum cell voltage to 0.62 V and reduces deviations in cell power and voltage by 31%. A low baseline fuel utilization is increased by decreasing the fuel flowrate and by increasing the stack current demand. Simulation results reveal fuel flow as a poor control variable because excessive tail-gas combustor temperatures limit fuel flow to below 110% of the baseline flowrate. Additionally, system efficiency becomes inversely proportional to fuel utilization over the practical fuel flow range. Stack current is found to be an effective control variable in this type of system because system efficiency becomes directly proportional to fuel utilization. Further, the integrated system acts to dampen temperature spikes when fuel utilization is altered by varying current demand. Radiation remains the dominate heat transfer mechanism within the stack even if stack surfaces are polished lowering emissivities to 0.2. Furthermore, the sensitivity studies point to an optimal system insulation thickness that balances the overall system volume and total conductive heat loss.     

Performance of a proton exchange membrane fuel cell stack using conductive amorphous carbon-coated 304 stainless steel bipolar plates

In this study, 304 stainless steel (SS) bipolar plates are fabricated by flexible forming process and an amorphous carbon (a-C) film is coated by closed field unbalanced magnetron sputter ion plating (CFUBMSIP). The interfacial contact resistance (ICR), in-plane conductivity and surface energy of the a-C coated 304SS samples are investigated. The initial performance of the single cell with a-C coated bipolar plates is 923.9 mW cm⁻² at a cell voltage of 0.6 V, and the peak power density is 1150.6 mW cm⁻² at a current density of 2573.2 mA cm⁻². Performance comparison experiments between a-C coated and bare 304SS bipolar plates show that the single cell performance is greatly improved by the a-C coating. Lifetime test of the single cell over 200 h and contamination analysis of the tested membrane electrode assemble (MEA) indicate that the a-C coating has excellent chemical stability. A 100 W-class proton exchange membrane fuel cell (PEMFC) short stack with a-C coated bipolar plates is assembled and shows exciting initial performance. The stack also exhibits uniform voltage distribution, good short-term lifetime performance, and high volumetric power density and specific power. Therefore, a-C coated 304SS bipolar plates may be practically applied for commercialization of PEMFC technology.     

Performance of DMFC with SS 316 bipolar/end plates

This work mainly emphasizes the development of new materials and design for a bipolar/end plate in a direct methanol fuel cell (DMFC). According to the DOE requirements, preliminary studies show that SS 316 (Stainless Steel 316) is a suitable candidate. Several flow field designs were studied and a modified serpentine design was proposed. SS 316 end plates were fabricated with an intricate modified serpentine flow field design on it. The performance of a single stack DMFC with SS 316 end plates were studied with different operational parameters. A long-term test was carried out for 100 h with recycling the methanol and the contaminants in the MEA were characterized. The stack efficiency is found to be 51% and polarization losses are discussed. SS 316 with low permeability resulted in an increased pressure drop across the flow field, which increased the fuel cell performance. The use of SS 316 as bipolar plate material will reduce the machining cost as well as volume of the fuel cell stack.