could cause membrane cracking and poor electrode-membrane contact. Another important challenge for >100°C operation of the baseline Nafion -base MEA is ...
MEMBRANE-ELECTRODE ASSEMBLY FOR HIGHTEMPERATURE PEMFC
in the cathode catalyst layer. Membrane improvement alone is not sufficient to achieve the PEMFC commercialization goal.
Chao-Yi Yuh, Ludwig Lipp, Pinakin Patel and Ray Kopp
Advanced Proton-Conducting Membranes Proton-conducting mechanisms have been extensively discussed by Kreuer2,3. Literature survey has showed that a useable proton conducting membrane material with desired proton conductivity comparable to fully hydrated Nafion (>0.1 S/cm) between 100150°C is not yet available4-6. Because useable 100-150°C protonconducting materials are not yet available, many new hightemperature proton-conducting materials are being actively developed. The reported approaches include: Mechanical Support to enhance high-temperature mechanical strength for ultra-thin membranes Solid Proton Conductor to enhance proton conductivity at low R.H. New High-Temperature Proton-Conducting Ionomer Substitutes for Water in Nafion to reduce humidity effect Ultra-thin membrane or Pt Doping for Self Humidification
FuelCell Energy, Inc. 3 Great Pasture Road Danbury, CT 06813 Introduction Fuel cells offer the best alternative to conventional fossil fuel combustion power generation technologies. However, for fuel cells to be commercially viable, issues such as cost, size, and functionality need to be addressed. A natural-gas fueled PEMFC system for stationary application desires near atmospheric pressure operation, >35% HHV efficiency, >100°C operation for cogeneration, simple construction, reliable >40,000h life, and low system cost (120°C) is highly desired. The ultimate MEA goals for commercial use include: 1) higher membrane proton conductivity with negligible electronic conductivity, area specific resistance less than 100 mΩcm2; 2) improved humidification properties (minimal water transport and low hydration) and dimensional stability (low swelling); 3) high mechanical strength; 4) low gas permeability (less than 0.1 percent gas crossover); 5) long life of 40,000h for stationary and 5,000h for transportation applications; 6) cell performance >0.7 V (0.8V preferred) at 400-500 mA/cm2; 7) low cost. Challenges with high-temperature operation Most developers presently use perfluorosulfonic acid (PFSA) polymer membrane such as Nafion made by DuPont. It is deficient in terms of ionic conductivity at temperatures above 100°C and at low relative humidity (R.H.). High-temperature operation dries out the membrane, drastically reducing proton conduction. To operate Nafion beyond 100°C, pressurization is needed to maintain high R.H., requiring complicated compressor system. Dried Nafion is also more permeable to gases, resulting in increased cross-leakage. Furthermore, the loss of liquid water embrittles the membrane and could cause membrane cracking and poor electrode-membrane contact. Another important challenge for >100°C operation of the baseline Nafion-base MEA is the significantly reduced cell voltage, mainly due to a large cathode polarization increase. This increase is mainly caused by a loss of proton conductivity and catalyst utilization
The solid proton conductors (superacid) under evaluation include PTA (phosphotungstic acid), ZHP (zirconium hydrogen phosphate), zeolite, silica, etc. These materials are brittle inorganics and therefore are generally incorporated into a composite structure containing flexible polymeric ionomer phases. However, although to a lesser extent than Nafion, they also lose water at high temperatures, with reduced proton conductivity. Considerable modifications of their morphologies (such as fine nano-size) may be needed to effectively enhance proton conductivity. New hightemperature iomomers are in general sulfonated or phosphonated polymers, usually containing aromatic backbone7-9, including sulfonated polyphosphazene, polysulfone, or PEEK. So far, longterm high-temperature proton conductivity and durability under low R.H. have yet to be demonstrated. These liquid acid (less volatile than water) impregnated membranes may adsorb at the Pt surface, resulting n high cathode polarization. The liquid acids may also evaporate away slowly during long-term use, limiting their hightemperature durability. Because water dry-out is the main cause of proton-conductivity loss in many materials at high temperatures, substitutes for water in the ionomers with high boiling-point protonconducting liquid such as pyrazole or imidazole have been evaluated. However, these liquid proton conductors may adsorb on the Pt catalyst surface, interfering ORR and resulting in high cathode polarization. The liquid proton conductor may also slowly evaporate away. The composite approaches are adopted by many researchers because few single-component monolithic materials have all the desired properties. So far, few composite materials have sufficient proton conductivity comparable to a fully hydrated Nafion. Many of the membrane materials reported above were studied for their proton-conducting and mechanical properties. However, very little performance data of MEA incorporating the advanced membranes at high temperatures were reported. The reported performance data were usually obtained with non-system conditions (e.g., very high R.H. or stoichs). The electrode Pt catalyst loading level was also usually too high (such as using Pt black, >1mg/cm2 per electrode), not practical for commercial use. In order for the Pt catalyst in the cathode to be utilized, its surface needs to be accessed by the proton-conducting phases and reactant oxygen. Therefore, an ionomer of high proton conductivity is not sufficient to guarantee a high cathode performance. The acid groups (ionic clusters) of the ionomers have to be situated right at the Pt for it to be active for the ORR. The water in the acidic ionic
Prepr. Pap.-Am. Chem. Soc., Div. Fuel Chem. 2003, 48(2), 893
clusters is also needed to facilitate ORR1. Furthermore, high oxygen permeability is needed to reduce mass-transfer loss. The baseline Nafion ionomer if well humidified has a very high oxygen permeability1. Any new proton-conducting phase in the cathode catalyst layer also needs to have similarly high oxygen permeability.
Acknowledgement This study was supported by the Office of Energy Efficiency and Renewable Energy, U.S. Department of Energy, under Contract DEFC02-99EE27567. And by U.S. National Science Foundation, under Contract DMI-0232574.
This paper detailed the MEA work performed at FCE focusing on improving high-temperature membrane durability and cathode catalyst layer activity.
References 1. Gottesfeld, G. and Zawodzinski, Z. in Electrochemical Science and Engineering, R. Alkire, H. Gerischer, D. Kolb and C. Tobias, Editors, Wiley-VCG, New York, 1994, 5, p.195-301, 2. Kreuer, K., Solid State Ionics, 2000, 136-137, 149. 3. Kreuer, K., Solid State Ionics, 1997, 94, 55. 4. Roziere, J. and Jones, D., in 1st European PEFC Forum, F. Buchi, G. Scherer and A. Wokaun, Editors, 2001, p. 145. 5. Zawodzinski, T., et al., in 2002 US DOE Hydrogen and Fuel Cell Joint Program Review, May 6-10, 2002. 6. Norby, N., Solid-State Ionics, 1999, 125, 1. 7. Kerres, J., J. Membrane Science, 2001, 185, 3. 8. Jones, D. and Roziere, J., J. Membrane Science, 2001, 185, 41. 9. Kreuer, K., J. Membrane Science, 2001, 185, 29. 10. Yuh, C, FY2002 Progress Report, Hydrogen, Fuel Cell and Infrastructure Technologies, Energy Efficiency and Renewable Energy, IV.B.5, 297-299.
MEA EVALUATION A number of composite membrane and MEA have been developed to improve water retention and proton conduction at 120°C. The experimental MEA are all less than 75 µm thick. The composite membranes consist of ionomer and solid superacid. Fine additives/modifiers with high-proton conductivity have been incorporated into the MEA (membrane and/or cathode). The protonconducting additives include: PTA, sulfated nano-oxide, ZHP and several types of zeolites. The ionomer phase under study in the composite MEAs include Nafion of various EW as well as low EW experimental ionomers. Cross-linking approach is also adopted to strengthen membrane mechanical strength. The fabricated MEAs were tested in laboratory-scale 25cm2 cells at 80-140°C to evaluate promising MEA formulations. FCE has achieved significant performance and endurance improvements10 under atmospheric pressure operation, hydrogen/air The cross-linking approach has atmosphere, at 400mA/cm2. enhanced membrane durability significantly, by reducing MEA OCV decay rate. The CO tolerance was significantly improved (at least to 100ppm CO) as illustrated in Figure 1, demonstrating the benefit of >100°C operation. Detailed MEA test results will be presented at the meeting.
CO Polarization curves for FCE-25-134 (120C) o
o
Stoich flow: Fuel/Air or O2 1.2/ 2.0, Humi.Temp.:Fuel/Air 95 C/ 90 C 1000
2000 Voltage H2/Air
800
1600
Voltage H2+ 100 ppm CO/Air
700
Voltage (mV)
1800
Voltage H2+ 10 ppm CO/Air
Voltage H2+1000ppm CO/Air
1400
600
Resistance H2/Air
1200
500
Resistance H2+10 ppm CO/Air
1000
Resistance H2+ 100 ppm CO/Air
800
Cell Voltage
400
Resistance H2+1000ppm CO/Air
300
Resistance
200
600 400
100
200
0 0
200
400
600
800
1000
2
Current Density (mA/cm )
Figure 1. Significant tolerance to 100ppm CO was achieved at 120°C.
Prepr. Pap.-Am. Chem. Soc., Div. Fuel Chem. 2003, 48(2), 894
0 1200
Resistance (mohms-cm 2)
900
