active electrode area) and the reversible cell voltage for the reaction is termed overvoltage (overpotential when referring to a single electrode). Prominent sources of overvoltage in a fuel cell are 1. Mixed potential at electrodes arising due to unavoidable parasitic reactions that tend to lower the equilibrium electrode potential. One particularly important cause of mixed potential is the crossover of fuel through the electrolyte from anode to cathode or vice versa. This is the dominant source of losses at open circuit, especially in direct methanol fuel cells where fuel crossover through the electrolyte membrane is high. 2. Activation losses arising Fuel Cells predominantly due to the kinetics at by Vijay Ramani the electrodes. The effects of these losses are most pronounced at low 2 Fuel cells are electrochemical devices that directly convert current densities (~1 to 100 mA/cm ). Examples include sluggish oxygen chemical energy to electrical energy. They consist of an reduction kinetics at the cathodes of electrolyte medium sandwiched between two electrodes (Fig. polymer electrolyte and phosphoric 1). One electrode (called the anode) facilitates electrochemical acid fuel cells and sluggish methanol oxidation kinetics at the anode of a oxidation of fuel, while the other (called the cathode) promotes direct methanol fuel cell electrochemical reduction of oxidant. Ions generated during 3. Ohmic losses arising due to the oxidation or reduction are transported from one electrode to resistive losses in the electrolyte and the other through the ionically conductive but electronically in the electrodes. The effects of these losses are perhaps most pronounced insulating electrolyte. The electrolyte also serves as a barrier at intermediate current densities between the fuel and oxidant. Electrons generated at the anode (~100 to 500 mA/cm2). during oxidation pass through the external circuit (hence 4. Mass transport losses arising generating electricity) on their way to the cathode, where due to nonreacting diffusion in the gas-diffusion layer and to reacting they complete the reduction reaction. The fuel and oxidant diffusion in the electrode layers. do not mix at any point, and no actual combustion occurs. The effects of these losses are most The fuel cell therefore is not limited by the Carnot effi ciency pronounced at high current densities 2 and, theoretically (although not practically), can yield 100% ( >500 mA/cm ). The combined contributions of these effi ciency. Fuel cells are primarily classifi ed according to the sources of overvoltage cause the cell electrolyte material. The choice of electrolyte material also voltage output to decrease with increasing governs the operating temperature of the fuel cell. Table I current density. A plot of cell voltage vs. lists the various types of fuel cells along with electrolyte used, current density is known as a polarization curve. A typical polarization curve for a operating temperature, and electrode reactions. hydrogen fueled polymer electrolyte fuel cell is shown in Fig. 2. The power output of the fuel cell (in mW/cm2) is given Operating Principles transferred and F is Faraday’s constant by the product of voltage and current (96,475 C/equiv). Because n, F, and E density. A comprehensive discussion of The reversible cell electromotive force o,cell are positive numbers, the standard free irreversibility and overvoltage in fuel cells (emf; E ) is defi ned as the difference 1 o,cell is presented in the literature. between the standard reduction potentials energy change of the overall reaction of the cathode and anode reactions is negative, indicating a spontaneous Challenges (E and E ). The actual number reaction. This is the thermodynamic o,cathode o,anode rationale behind fuel cell operation. Each type of fuel cell has its own may vary depending on the reactions that advantages and disadvantages. For In an ideal (reversible) fuel cell, occur at these electrodes, but is always example, alkaline fuel cells allow the use the cell voltage is independent of the positive. For example, in a hydrogen/ of nonprecious metal catalysts because current drawn. Practically, the reversible oxygen polymer electrolyte fuel cell of facile oxygen reduction kinetics at cell voltage is not realized even under operated at standard conditions, the high pH conditions, but suffer from the open-circuit (zero current) conditions reversible cell emf is 1.23 V. The standard problem of liquid electrolyte management free energy change (∆G ) of the overall due to the myriad irreversibilities that o and electrolyte degradation. Similarly, arise during fuel cell operation. The reaction of the fuel cell is given by molten carbonate fuel cells can tolerate difference between actual cell voltage at ∆G = –nFE [1] high concentrations of carbon monoxide o o,cell a given current density (current per unit where n is the number of electrons (continued on next page)

The Electrochemical Society Interface • Spring 2006 41 Fuel Cells (continued from previous page)

Table I. Classification of fuel cells Fuel Cell Type Electrolyte Used Operating Tempearture Electrode Reactions

+ - Polymer Electrolyte Polymer Membrane 60-140˚C Anode: H2 = 2H + 2e 1 + - Cathode: /2 O2 + 2H + 2e = H2O + - Direct Methanol Polymer Membrane 30-80˚C Anode: CH3OH + H2O = CO2 + 6H + 6e 3 + - Cathode: /2 O2 + 6H + 6e = 3H2O - - Alkaline Potassium Hydroxide 150-200˚C Anode: H2 + 2 OH = H2O + 2e 1 - - Cathode: /2 O2 + H2O + 2e = 2 OH + - Phosphoric Acid Phosphoric Acid 180-200˚C Anode: H2 = 2H + 2e 1 + - Cathode: /2 O2 + 2H + 2e = H2O 2- - Molten Carbonate Lithium/Potassium Carbonate 650˚C Anode: H2 + CO3 = H2O + CO2 + 2e 1 - 2- Cathode: /2 O2 + CO2 + 2e = CO3 2- - Solid Oxide Yittria Stablized Zirconia 1000˚C Anode: H2 + O = H2O + 2e 1 - 2- Cathode: /2 O2 + 2e = O

in the fuel stream (CO is a fuel for such fuel cells!), but their high operating temperature precludes rapid start-up and sealing remains an issue. Solid oxide fuel cells offer high performance, but issues such as slow start-up and interfacial thermal conductivity mismatches must be addressed. High cost is an issue that affects each type of fuel cell. It is difficult to enumerate the merits and demerits of each type of fuel cell given the limited scope of this article. Hence, the discussion below is restricted to polymer electrolyte fuel cells (PEFCs) and outlines a few of the challenges that currently impede PEFC commercialization. A PEFC is unique in that it is the FIG. 1. only kind of low temperature fuel cell that uses a solid electrolyte, usually a polymer electrolyte membrane (PEM). Perhaps the most common PEM in use today is Nafion® a well-researched2 perfluorosulfonic acid (PFSA) ionomer which is commercially available in films varying from 25 to 175 µm thick. PEFCs presently operate at relatively low temperatures (60-120°C; more toward the lower end of the range). The operating environment is kept well hydrated to maximize membrane conductivity. Hydration is typically achieved by humidifying inlet reactant gases by bubbling them through saturators containing water at a fixed temperature. By varying the saturator temperature relative to the cell operating temperature, the desired inlet relative humidity can be maintained. There has been enhanced interest in recent times to operate PEFCs at FIG. 2. temperatures >100°C for a multitude of reasons: (i) to minimize the poisoning effect of carbon monoxide impurities in the reformate hydrogen feed stream; (ii) to enable rapid heat rejection

42 The Electrochemical Society Interface • Spring 2006 with a reasonably sized radiator (for The Hydrogen Rush: Boom or Bust? transportation applications); and (iii) to enhance reaction rates and simplify by Krishnan Rajeshwar water management issues. It is diffi cult Energy has been listed as humanity’s of 2005 includes $3.7 billion for hydrogen to operate much above 100°C and number one problem for the next 50 years and fuel cell research and development. maintain 100% relative humidity by several agencies. That it surpasses other The bill also includes a fuel cell tax without incurring parasitic power issues such as water, food, environment, credit up to $1,000 per kilowatt on the losses due to pressurization. Hence, poverty, terrorism, war, disease, education, purchase of fuel cells used in residential PEMs that have high conductivities at democracy, and population, underscores or commercial applications. President elevated temperatures and low relative the criticality of energy in almost Bush’s just-announced budget for fi scal humidities are sought.3 Unfortunately, everything that a society does. The U.S. year 2007 includes increases in both solar the conductivity of Nafi on® drops with Department of Energy projects (http:// and hydrogen funding. Gold, copper, www.eia.doe.gov/oiaf/ieo) that the world’s and other metals spurred much of the decreasing relative humidity (RH), total energy consumption will rise by 54% economy in this country during the leading to unacceptable ohmic losses and between 2001 and 2025. This projected 1800s and communities sprouted up and precluding elevated temperature operation increase may be an underestimation due became ghost towns as the metal deposit sans pressurization. This is one important to the rapid economic development in veins were mined and depleted in turn. challenge facing the PEFC community heavily populated countries like India Before the 1800s, wood was the dominant today. In response to this challenge, and China. This worldwide competition energy source, later replaced by more extensive efforts have been made to for energy sources is likely to create energy-dense fossil fuels. The mining improve the properties of Nafi on® and to tensions among countries unless they can of hydrogen and other alternatives to identify alternate replacement materials. develop new energy sources. To meet our fossil fuels (e.g., ethanol and methanol) These efforts can be divided into three future energy needs, we must develop will ignite a new rush with hydrogen broad categories: (i) modifi cation of alternative energy sources because (i) the becoming the dominant energy force in Nafi on® and similar perfl uorinated fossil fuel reserve will not be suffi cient to the 21st century. meet the demand beyond 2050; and (ii) membranes by inclusion of inorganic The Energy Technology Division an over-reliance on fossil fuels (many of additives;4-10 (ii) membranes based on (ETD) within ECS has responded to this them are located in politically unstable sulfonated hydrocarbon backbones;11-17 tremendous challenge by sponsoring or regions in the Middle East, Central Asia, co-sponsoring several symposia related and (iii) acid impregnated polymer Africa, and Latin America) has both matrices.18-20 to energy conversion and storage. environmental and political implications. Examples include “Energy for Cleaner A second important issue is the poor The U.S. and much of the world Transportation” (208th Meeting, Los kinetics of the oxygen reduction reaction currently rely heavily on coal, oil, and Angeles, CA); “Photovoltaics for the 21st (ORR) in acidic media. Even with the natural gas for energy. These fossil fuels Century III”—sequel to I and II in earlier use of high loadings of expensive noble are nonrenewable, fi nite resources that meetings (208th Meeting, Los Angeles, metal catalysts such as platinum in the will eventually dwindle, becoming CA); “Proton Exchange Membrane Fuel electrode, the activation overpotential too expensive or too environmentally Cells V” (208th Meeting, Los Angeles, for the ORR is on the order of 500 mV at damaging to retrieve. To achieve a CA); “Energy Systems for the Twenty-First acceptable current densities. Using such sustainable energy future, we must Century: Opportunities for Applications develop fuel diversity. By contrast, of Solar and Conversion Technologies” high platinum loadings (0.4 mg/cm2) hydrogen can be mined from diverse (209th Meeting, Denver, CO); “Electrode lead to fuel cell costs that are too high sources, in particular water, using Materials and Processes for Energy by at least an order of magnitude (two nonfossil fuel energy such as solar, Conversion and Storage” (209th Meeting, orders of magnitude for transportation wind, and nuclear and later used for Denver, CO); “Portable Energy Sources” applications) to permit commercialization. the world’s energy needs with minimal (209th Meeting, Denver, CO); “Biological Three approaches are being pursued to environmental impact. The element is Fuel Cells” (209th Meeting, Denver, resolve this issue: (i) efforts are ongoing abundant and can be found not only CO); “Hydrogen Production, Transport, to enhance the activity of the noble metal in water but also in many organic and Storage” (209th Meeting, Denver, catalyst through alloying. By enhancing compounds. Once separated from other CO); “Direct Methanol Fuel Cells” activity, it is anticipated that the amount elements, hydrogen can be burned (209th Meeting, Denver, CO); “Organic of catalyst required can be lowered; (ii) a as a fuel or converted into electricity Photovoltaics” (210th Meeting, Cancun, variety of less expensive, non-noble metal through fuel cells (see the accompanying Mexico). article on fuel cells). Developing this catalysts such as metal porphyrins are A fuller account of the renewable marvelous clean energy carrier as a being investigated in acidic media; and hydrogen story may be found in the solution to the world’s energy appetite special issue of this magazine titled (iii) in a more indirect approach, attempts requires electrochemists working are being made to engineer the electrode “Hydrogen: Production and Storage” together with other scientists, engineers, which came out in fall 2004 (Vol. 13, in a manner designed to ensure that venture capitalists, business leaders, and No. 3).  ohmic and mass transport losses are kept government leaders to research, develop, to an absolute minimum and that the and deploy hydrogen technologies that Acknowledgment reaction is activation controlled. Gasteiger will not only be cost effective but will et al.21 provide an excellent review of generate long-term technological and Dr. Robert McConnell (National Renewable Laboratory, Golden, CO) these approaches and suggest that Pt-alloy economic benefi ts for the countries fi rst to provided many helpful comments. catalysts seem to offer the best chance of mine this particular element successfully. achieving suffi cient enhancement in mass The development of hydrogen and fuel activity over pure platinum. cell technologies has been accelerated About the Author by the $1.2 billion in research funding The third challenge relates to KRISHNAN RAJESHWAR is a Distinguished proposed by President Bush in his 2002 University Professor of Chemistry and durability. Typically, lifetimes of 40,000 State-of-the-Union Address (Freedom Biochemistry and Associate Dean (College h of continuous operation (for stationary Fuel initiative). The call for energy of Science) in the University of Texas at applications) and 5,000-10,000 h of independence was reiterated in his 2006 Arlington. His research interests are in cyclic operation (for transportation address where he specifi cally emphasized photoelectrochemistry, photocatalysis, and applications) are targeted as a precursor to we had to overcome our “addiction to solar energy conversion. He may be reached at oil.” Furthermore, the Energy Policy Act (continued on next page) [email protected].

The Electrochemical Society Interface • Spring 2006 43 Fuel Cells dissolution followed by redeposition leads 4. P. Antonucci, A. Aricò, P. Cretì, E. Ramunni, (continued from previous page) to an increase in catalyst particle size. and V. Antonucci, Solid State Ionics, 125, 431 (1999). commercialization. By inference, the 25- Precipitation in the membrane phase also 5. K. Adjemian, S. Lee, S. Srinivasan, 50 µm thick membrane and the 10-15 µm occurs if the dissolved platinum migrates J. Benziger, and A. Bocarsly, J. Electrochem. thick electrodes should be able to survive signifi cantly. Both these processes lead Soc., 149, A256 (2002). for this time frame. From a membrane to reduced surface area and mass activity 6. S. Malhotra and R. Datta, J. Electrochem. Soc., standpoint, with the possible exception of and result in lowered performance. These 144, L23 (1997). Nafi on®, few (if any) membrane materials effects tend to be exacerbated under load 7. B. Tazi and O. Savadogo, Electrochim. Acta, can claim to meet these requirements, cycling conditions, as would be common 45, 4329 (2000). 8. V. Ramani, H. R. Kunz, and J. M. Fenton, largely because of the degradation in automotive applications. Further details J. Membr. Sci., 266, 110 (2005). mechanisms in play during fuel cell about these processes, as well as factors 9. G. Alberti and M. Casciola, Solid State Ionics, operation. PEM degradation mechanisms such as catalyst support degradation and 97, 177 (1997). can be classifi ed as mechanical (pinhole membrane degradation are presented in 10. K. A. Mauritz, Mater. Sci. Eng., C, 6, 121 and crack formation), thermal (dry- an excellent overview by Mathias et al.24 (1998). out, solvolysis, and desulfonation), The interrelationships between operating 11. D. J. Jones and J. Rozière, J. Memb. Sci., 185, and chemical (peroxide initiated free temperature, operating humidity, and 41 (2001). radical degradation).22 Mechanical and degradation rates are also discussed at 12. K. D. Kreuer, J. Membr. Sci., 185, 29 (2001). 13. J. A. Kerres, J. Membr. Sci., 185, 3 (2001). thermal degradation can be minimized length in the above overview. 14. G. Alberti, M. Casciola, L. Massinelli, and by proper choice of materials, careful Concluding Remarks B. Bauer, J. Membr. Sci., 185, 73 (2001). fuel cell fabrication, and by maintaining 15. P. Jannasch, Curr. Opin. Colloid Interface Sci., reasonable operating temperatures Should the above challenges be 8, 96 (2003). and humidities. However, chemical overcome, PEFC technology promises 16. G. Alberti and M. Casciola, Annu. Rev. Mater. degradation of the membrane can be to be an attractive future proposition, Res., 33, 129 (2003). diffi cult to mitigate. One reason for this with applications in the transportation, 17. D. J. Jones and J. Rozière, Annu. Rev. Mater. Res., 33, (2003). diffi culty is the generation of hydrogen stationary power, portable electronics, 18. R. Savinell, E. Yeager, D. Tryk, U. Landau, J. peroxide at the oxygen reduction and military sectors. Typical applications Wainright, D. Weng, K. Lux, M. Litt, and C. electrode (a four electron reduction at that have been proposed include replacing Rogers, J. Electrochem. Soc., 141, L46 (1994). the cathode produces water, but a two internal combustion engines with 75-80 19. Q. Li, R. He, R. Berg, H. A. Hjuler, and N. J. electron reduction yields hydrogen kW PEFC stacks; using 1-5 kW stacks Bjerrum, Solid State Ionics, 168, 177 (2004). peroxide). Such generation has been to power individual homes; using 250 20. L. Xiao, H. Zhang, E. Scanlon, L. S. demonstrated by Schmidt et al.23 kW and larger stacks to provide highly Ramanathan, E.-W. Choe, D. Rogers, T. Apple, and B. C. Benicewicz, Chem. Mater., Metal ion impurities in the carbon reliable power to installations such as 17, 5328 (2005). based electrocatalyst support used in fuel credit card clearing centers and hospitals; 21. H. A. Gasteiger, S. S. Kocha, B. Sompalli, and cells serve to combine with the generated developing 1-100 W methanol fueled F. T. Wagner, Appl. Catal., B, 56, 9 (2005). peroxide to yield reactive oxygen species stacks to power consumer electronics 22. A. B. LaConti, M. Hamdan, and R. such as cellular phones and laptops C. McDonald, Handbook of Fuel Cells (such as free radicals) which initiate and – Fundamentals, Technology and Applications, 22 (replacing rechargeable batteries); and accelerate membrane degradation. Two Vol. 3, W. Vielstich, H. A. Gasteiger, and approaches that have been employed developing methanol fueled portable A. Lamm, Editors, p. 647, John Wiley & to minimize reactive oxygen species power for military applications. Other Sons (2003). include (i) the use of free radical fuel cell technologies also have their 23. T. J. Schmidt, U. A. Paulus, H. A. Gasteiger, and R. J. Behm, J. Electroanal. Chem., 508, scavengers, and (ii) the use of dispersed own niche applications (e.g., solid oxide fuel cell based auxiliary power units for 41 (2001). peroxide decomposition catalysts within 24. M. F. Mathias, R. Makharia, H. A. Gasteiger, 22 large vehicles, molten carbonate fuel cell a composite membrane. The former J. J. Conley, T. J. Fuller, C. J. Gittleman, approach can only delay the onset of based combined cycle power plants, etc.). S. S. Kocha, D. P. Miller, C. K. Mittelsteadt, degradation for a defi nite period due to Considerable research efforts are ongoing T. Xie, S. G. Yan, and P. T. Yu, Electrochem. Soc. Interface, 14(3) 24 (2005). steady consumption of the scavenger in academia and industry to help alleviate molecules. A third approach that suggests existing bottlenecks to permit rapid fuel About the Author itself is the design of electrocatalysts with cell commercialization. enhanced selectivity toward four electron The Electrochemical Society, VIJAY RAMANI is an assistant professor of chemical engineering in the Department of Chemical and reduction, thereby minimizing the particularly the Energy Technology Environmental Engineering at the Illinois Institute generation of hydrogen peroxide through Division, has taken a leadership role of Technology, Chicago. His interests include the two electron pathway. In addition to in PEFC technology by sponsoring developing novel materials for polymer electrolyte minimizing the concentration of reactive a series of symposia on many of the fuel cells and organic photovoltaics, fuel cell oxygen species, attempts are also ongoing topics addressed above. A Fuel Cells membrane and electrode durability, and fuel cell education and outreach. He may be reached at to render the ionomer more resistant to Subcommittee has also been constituted [email protected]. degradation by improving the synthesis [see Electrochem. Soc. Interface, 14(3), 11 methodology. (2005)] for coordinating fuel cell symposia Membrane durability alone is not within ECS and to promote ECS as the suffi cient. The electrocatalyst also must leading professional organization for the retain its activity over the entire span development of fuel cells.  of operation. Unfortunately, this is a References diffi cult task, especially at the oxygen 1. H. A. Liebhafsky and E. J. Cairns, reduction electrode, where high potentials Fuel Cells and Fuel Batteries: A Guide to Their Research are attained at low loads. Under high and Development, John Wiley & Sons, New potential conditions, given the low pH York (1969). of the acidic electrolyte, conditions are 2. K. A. Mauritz and R. B. Moore, Chem. Rev., tailor-made for platinum dissolution. 104, 4535 (2004). This is evidenced by a glance at the 3. R. K. A. M. Mallant, J. Power Sources, 118, 424 (2003). Pourbaix diagram for platinum. Such

44 The Electrochemical Society Interface • Spring 2006