Fullscreen HTML5Med Res (??MB)HTML4

Last month, we introduced the subject of fuel cells and outlined how they are being researched by many major car manufacturers around the world. In this issue, we look more closely at the main types of fuel cells and how they work. Fuel Cells Explode! By GERRY NOLAN T here’s been an explosion in the number and type of fuel cells – either in production, in testing or in design. Fuel cell and vehicle manufacturers around the world are confidently predicting virtually zero-polluting, fuel-cell powered models entering the mainstream market perhaps as early as 2005 – and certainly by 2010. (See SILICON CHIP May 2002). This month, we’re looking at the various types of fuel cells, how they work and how they differ from one another. We even look at some which are still very much in the “concept” stage but which show great promise. Main types The main fuel cell types are alkaline fuel cell (AFC), polymer electrolyte membrane (PEMFC), also known as the proton exchange membrane, direct methanol (DMFC), molten carbonate (MCFC), phosphoric acid (PAFC), solid oxide (SOFC) and protonic ceramic fuel cell (PCFC). Although we indicated last month that there were five main types of fuel cells, we’ll treat the direct methanol fuel cells, which are quite similar to polymer electrolytic membrane fuel cells, separately. We’ll also look briefly at regenerative fuel cells (RFC) and zinc-air fuel cells (ZAFC). Fuel cells are classified by the type of electrolyte they use. This may be acidic or alkaline and is either liquid, generally in a porous matrix, or a high temperature solid state electrolyte present as a ceramic material in the solid oxide (SOFC) and proton ceramic fuel cells (PCFC). A circulating liquid electrolyte has the advantage that it can be used to manage heat removal and adjust the electrolyte concentration and water balancing while it is in operation. Sloshing of the electrolyte can be prevent80  Silicon Chip ed by using a micro-porous matrix or by crystallising or gelling the electrolyte as in a PAFC. In the PEMFC, the polymer electrolyte membrane functions as a fixed acidic electrolyte. General overview First, let’s discuss how a generic fuel cell works before we move on to specific types and their operation. As shown in the diagram of Fig.1, hydrogen is fed into the anode and oxygen enters through the cathode. Under the influence of a catalyst, each hydrogen atom splits into a proton and an electron which are forced to take different paths to the cathode. The protons pass through the electrolyte while the electrons return to the cathode, where they rejoin with the hydrogen and oxygen to form a molecule of water. The electron flow can be used in any way that an electric current from a generator or battery could be used, for example, to power a car, appliance or anything you like. Since fuel cells rely on a controlled chemical reaction and not the relatively uncontrolled combustion of an internal combustion engine, emissions from fuel cells are much lower. In fuel cells with an acidic electrolyte, positively charged hydrogen ions (protons) migrate from the anode, also called the fuel electrode, to the cathode, also called the air electrode, where water is produced. In alkaline fuel cells, the charge is carried by negatively-charged ions and the water is produced at the hydrogen electrode (anode). In principle, any exothermic chemical reaction (ie, where heat energy is released) can be used to generate electricity. All fuel cells convert chemical energy into www.siliconchip.com.au This one-kilowatt portable Ballard fuel cell generator demonstration unit is a fully automated power system which converts hydrogen fuel and oxygen from air directly into DC electricity. Water is the only byproduct of the reaction. It operates at low pressures, provides reliable, clean, quiet and efficient power and is small enough to be carried to wherever power is needed. (Photo courtesy Ballard Power Systems). electric energy and if suitable electrodes and an electrolyte to support the reaction can be provided, a fuel cell system can utilise the hydrogen from any hydrocarbon fuel, such as natural gas, methanol and even petrol. In the old way, fuels such as propane, petrol, diesel or hydrogen are burnt in an internal combustion engine or in a furnace, with the heat energy being converted to mechanical energy in a piston engine or a turbine, which drives a generator to produce electricity. In general, these thermodynamic processes are quite inefficient and this is made worse by the moving parts in a reciprocating engine, so that typically the efficiency is 20-30% at best. Hydrogen-oxygen fuel cells are far better and can achieve efficiencies in the range of 60-70%. Fig.1a (left) shows the components and chemical reactions occuring in a generic hydrogen fuel cell. In a typical stationary power generation unit (Fig.1b, right), the fuel cell hydrogen is derived from natural gas, using some of the byproduct heat energy from the fuel cell itself. (Courtesy Ballard Power Systems). www.siliconchip.com.au June 2002  81 Single Cylinder Internal Combustion Engine versus Ballard Single Fuel Cell Engine PEM (Proton Exchange Membrane) Fuel Flow Field Plate Spark Plug Oxidant Flow Field Plate Fuel & Air Mixture MISING FIGS 3A High Temperature Combustion Process (2500°C) Exhaust NOx HC Smog CO SOx Exhaust Water Vapour (No Pollution) Heat (125°C) Water Cooled Heat (90°C) Water cooled Fuel to recirculate Low Temperature Electrochemical Process (90°C) Air Fig.2: this comparison between an internal combustion engine and a fuel cell engine clearly demonstrates why engineers Output are getting so excited! (Courtesy Rotary Mechanical Power (20% Efficiency) Ballard Power Systems). Fuel (Hydrogen) To transmission (C) Ballard Power Systems Electric Motor Output Rotary Mechanical Power (45% Efficiency) Unfortunately, hydrogen is not a readily available fuel so efforts have to be made to convert hydrocarbon fuels into pure hydrogen and carbon dioxide. proton membrane exchange (PEM) and other acid types, stating that an alkaline fuel cell with a circulating liquid electrolyte would be a better choice than PEM fuel cells for electric vehicles and on-site power systems. One of Alkaline fuel cells the reasons given is that AFCs are much less expensive to build than PEMs because they contain less noble metAs discussed in last month’s issue, Francis T. Bacon al catalyst material – platinum and palladium are very developed the first successful fuel cell in 1932, using expensive. hydrogen, oxygen, potassium hydroxide as the electrolyte The cost of the AFC is becoming as low as US$200 to and nickel electrodes. So alkaline fuel cells were the first $300 per kilowatt without accessories and US$400 to $600 to be used successfully. with accessories, while the cost of the PEM is a factor of 10 Thirty years later, Bacon and a co-worker produced a higher with or without accessories, partly because AFCs 5kW fuel cell system and it is history that the Bacon design require less accessory equipment. was chosen by NASA over nuclear power and solar energy, Some of the accessory equipment that is required for as the power supply for the Apollo and Gemini missions PEMs and not for AFCs are air-compressors and humidifiand the shuttle orbiters – incidentally providing water as ers. This accessory equipment uses power, which reduces well as electricity. These cells can now achieve electrical the overall efficiency of the PEM system, as well as making generating efficiencies of up to 70% with outputs that it less convenient to use. range from 300W to 5kW. Another advantage is that AFCs produce a higher voltage Alkaline fuel cells, (AFCs) generally use solutions of than PEMs. The cell operating voltage of sodium hydroxide (NaOH) or potassium an AFC is 0.8V while the PEM is 0.6V; hydroxide (KOH) – see Fig.3. The cathode 100 AFC cells produce 80V, while 100 reaction is faster in the alkaline electrolyte, PEM cells produce 60V. resulting in higher performance. However, a major disadvantage of AFCs is that the alkaWhile PEM cells cannot be convenline electrolytes react with carbon dioxide iently shut down for extended periods, to precipitate carbonates. AFCs can be shut down for as long as required for maintenance or rest, which If there is any carbon dioxide present, is quite important. Instead of separators it will quickly degrade the electrolyte which must be kept moist at all times, and reduce the efficiency of the cell. As a AFCs have a built-in circulating electroresult, AFCs are typically restricted to spelyte system so there is no water-buildcialised applications where pure hydrogen up problem and humidifiers and air and oxygen are used, such as low power compressors are unnecessary. Shutting aerospace and defence applications. They down an AFC is as easy as turning off are considered too costly for commercial the switch, after which the electrolyte is applications but several companies are automatically removed from the stacks, working to reduce costs and improve opermaking the AFC inactive. ating flexibility. AFCs operate on hydrogen derived Alkaline fuel cell manufacturers still Fig.3: chemical reactions within an alkaline fuel cell. from ammonia and, being rich in hyclaim advantages for their cells over the 82  Silicon Chip www.siliconchip.com.au supplied. While the electrons are taking the long way around, the protons diffuse through the electrolyte directly to the cathode. Here the hydrogen ion recombines with its electron and reacts with oxygen to produce water, thus completing the overall process. PEM fuel cell output is generally in the range from 50W to 250kW. Direct methanol fuel cells Fig.4a: the components and chemical reaction in a PEMFC. drogen, anhydrous ammonia (NH3) is one of the best carriers of hydrogen. As it is not a hydrocarbon, it does not produce any harmful emissions. AFCs can use hydrogen produced by an ammonia cracker but PEM fuel cells cannot. This is because this hydrogen carries with it a trace of ammonia gas which the PEM fuel cell, being acidic, cannot tolerate. What do we conclude from this? Although most vehicles on the verge of production are using acid-type cells (quite often PEMFC), manufacturers of AFCs have not given up. But it’s early in the story yet. Let’s go on and see what the others have to offer. Proton exchange membrane fuel cells Proton exchange membrane fuel cells (PEM), (also known as polymer electrolytic fuel cells) are currently the most common type of fuel cell being developed for use in vehicles. The reasons for this are mainly that they use inexpensive manufacturing materials, ie, plastic membrane, they react quickly to changes in electrical demand and do not leak or corrode. They also operate at relatively low temperatures, 80°C, for greater efficiency and have high power density. Because their power output can change quickly to meet shifts in power demand, they are suited for motor vehicles where quick startup is required. The proton exchange membrane, which allows hydrogen ions to pass through it, is a plastic sheet, typically 0.2mm thick, coated on both sides with highly dispersed metal alloy particles, mostly platinum, that are active catalysts. The electrolyte used is a solid organic polymer, poly-perflourosulfonic acid. Using a solid electrolyte has the advantage of reducing corrosion and management problems. Hydrogen is fed to the anode side of the fuel cell where the catalyst promotes the separation into hydrogen ions and electrons – see Fig.4. The electrons are passed through an electric load (eg, electric motor) before returning to the cathode side of the fuel cell to which oxygen has been www.siliconchip.com.au These are like PEM cells but instead of pure hydrogen they use a methanol-water solution. This is introduced to the fuel electrode, where the anode catalyst extracts the hydrogen in a spontaneous reaction which splits the methanol molecules, freeing the hydrogen and allowing the carbon atom to combine with the oxygen atoms from the methanol to form carbon dioxide. Because methanol readily frees its hydrogen to react in the fuel cell, it is an ideal carrier, eliminating the need for a fuel reformer or to have a fuel tank of pure hydrogen – see Fig.4. In the process of splitting the methanol molecules to free hydrogen, the catalyst at the anode promotes the electrochemical oxidisation of the released hydrogen to produce electrons which travel through the external circuit back to the cathode electro-catalyst. This promotes the reduction reaction to combine the electrons with oxygen. As in the PEM fuel cell, the circuit is completed within the cell by protons passing through the electrolyte. Operating temperatures of direct methanol fuel cells are in the same range as PEM cells, 50-100°C, which achieves efficiencies of about 40%. The low temperature range makes this type of fuel cell a possibility for use in small to mid-sized applications such as mobile phones and laptop computers. Due to their simplicity, direct methanol fuel cells are also being considered for use by the transportation industry. Fig.3b (above) reveals detail of a Ballard fuel cell stack showing the flow field plates which supply the bodies of fuel and air to either side of the proton exchange membrane. Stacking more cells together increases the voltage produced; increasing the cell’s surface area increases the current produced. The first commercial PEM fuel cell module, designed for integration into a range of stationary and portable power generation applications. (Courtesy Ballard Power Corp). June 2002  83 Figs.5, 6 & 7: the chemical reactions in direct methanol, phosphoric acid and solid oxide fuel cells. Molten carbonate fuel cells These are second-generation fuel cells designed to operate at higher temperatures than phosphoric acid or PEM cells. Because molten carbonate technology is specifically designed to operate at the higher temperatures it is able to achieve higher fuel-to-electrical output and overall energy use efficiencies than lower temperature cells. At these temperatures, the electrolyte solution of lithium, sodium and/or potassium carbonates soaked in a matrix becomes molten and able to conduct charged particles (ions) between the two porous electrodes. Molten carbonate fuel cells are at the high power end, with units achieving outputs of up to 2MW while there are designs on the drawing board for units up to 100MW! The nickel electrode catalysts of molten carbonate fuel cells are inexpensive when compared with other catalysts and they promise high fuel-to-electrical output efficiencies – about 60% normally or 85% with co-generation. However, the high operating temperatures, typically 650°C, limit the practicality of these cells for many applications. However, the high operating temperature is not all bad news. It allows much greater flexibility in types of fuels and inexpensive catalysts because the reactions involved in breaking the carbon bonds in larger molecule hydrocarbon fuels occur much faster as the temperature is increased. Molten carbonate fuel cells have been run on hydrogen, natural gas, propane, landfill gas, marine diesel and simulated coal gasification products. These cells are mainly intended for use in electric utility applications and have been successfully demonstrated in this role in Japan and Italy. When natural gas is used as the fuel, methane and steam are converted into a hydrogen-rich gas inside the fuel cell stack in a process called ‘internal reforming’. The hydrogen produced reacts with the carbonate ions (CO3) at the anode to produce water, carbon dioxide and electrons. As with all cells, the electrons travel through an external circuit before returning to the cathode. At the cathode, oxygen from the air and carbon dioxide recycled from the anode react with the electrons to form CO3 ions that replenish 84  Silicon Chip the electrolyte and flow through the fuel cell, completing the circuit. Molten carbonate fuel cells eliminate the external fuel processors that other fuel cells need to extract hydrogen from the fuel. In reaching efficiencies approaching 60%, molten carbonate cells are considerably more efficient than the 3742% of a phosphoric acid fuel cell plant. Further, when the heat produced is used for space or water heating, the overall efficiency can be as high as 85%. Phosphoric acid fuel cells-PAFC These were the first fuel cells to become commercially available in the electric power industry. More than 200 of these ‘first generation’ phosphoric acid fuel cell systems have been installed all over the world, in hospitals, nursing homes, hotels and so on, including one that powers a police station in New York City’s Central Park. From this, it is apparent they are more suited to a stationary type of application. Efficiency ranges from 40-80% and the operating temperature is 1500-2000°C. At lower temperatures, phosphoric acid is a poor ionic conductor and carbon monoxide (CO) poisoning of the platinum (Pt) electro-catalyst in the anode becomes severe. Existing PAFCs have outputs up to 200kW and 11MW units have been tested. As already indicated, PAFCs generate electricity at more than 40% efficiency and, when the steam it produces is used for cogeneration, efficiency rises to nearly 85%. This compares to about 35% efficiency for a typical electrical power grid. Apart from the nearly 85% cogeneration efficiency, one of the main advantages is that it can use impure hydrogen as fuel. Operating at the right temperature, PAFCs can tolerate a CO concentration of about 1.5%, which increases the range of fuels they can use. However, if petrol is to be used, any sulphur content must be first removed. Now what are the problems with phosphoric acid fuel cells that make the molten carbonate fuel cells so much more attractive? They use expensive platinum as a catalyst www.siliconchip.com.au and only generate low current and power per cell, making them generally much larger and heavier than other types of fuel cells for the same total power output. However, PAFCs are the most mature fuel cell technology and for the present, that means tried and tested reliability. joining the anodes and cathodes of adjacent cells. Advanced SOFCs coupled with small gas turbines with a combined rating in the range of 250kW to 25MW could eventually compete with wholesale power rates. Solid oxide fuel cells This new type of fuel cell uses a ceramic electrolyte material that has high protonic conductivity at high temperatures. Because of the high operating temperatures, PCFCs can electrochemically oxidise fossil fuels directly to the anode, thereby eliminating the intermediate step of producing hydrogen through the expensive reforming process. Gaseous molecules of the hydrocarbon fuel are absorbed onto the surface of the anode in the presence of water vapor, where the hydrogen atoms are stripped off and absorbed into the electrolyte, with carbon dioxide being the primary reaction product. Because PCFCs have a solid electrolyte, the membrane cannot dry out as with PEM fuel cells and there is no liquid electrolyte to leak as with PAFCs. This is a promising new fuel cell which an Australian company, Ceramic Fuel Cells Ltd, with the collaboration of the CSIRO, has concentrated on. It has the potential to be used in high-power distributed generation applications, including large-scale electricity generating stations. Some developers are promoting SOFCs for motor vehicles and are developing auxiliary power units using SOFCs. Solid oxide fuel cells are a different branch altogether of fuel cell technology – see Fig.7. The anode, cathode and electrolyte are all made from ceramics, which enables the cells to operate at temperatures significantly higher than any other mainline fuel cell. They also produce exhaust gases at temperatures ideal for cogeneration for use in combined-cycle electric power plants. The fact that the cells can be produced as rolled tubes or flat plates enables them to be manufactured using many of the techniques presently used by the electronics industry. Although a variety of oxide combinations have been used for solid oxide electrolytes, the most common so far has been a mixture of zirconium oxide and calcium oxide formed as a crystal lattice and stabilised with yttria – usually called YSZ or yttria stabilised zirconium. At the high operating temperatures, oxygen ions are formed at the ‘air electrode’, a ceramic cathode conducting perovskite, lanthanum manganate (LaMnO3). A fuel gas containing hydrogen is passed over the ‘fuel electrode’, the anode, typically formed from a nickel/yttria-stabilized zirconia cermet. A cermet is a material consisting of a metal matrix with ceramic particles disseminated through it. The oxygen ions migrate through the yttria-stabilised zirconia crystal lattice of the electrolyte to oxidise the fuel. Electrons liberated at the anode pass through an external circuit to create an electrical current. Because of the high temperatures, natural gas or other hydrocarbon fuels are reformed internally to extract the hydrogen, eliminating the need for an external reformer. At present, fuel-to-electricity efficiencies of solid oxide fuel cells are around 50%. However, as indicated earlier, if the hot exhaust of the cells is used in a hybrid combination with gas turbines, this is likely to approach 60%. Where the waste heat of the system is able to be used as well, overall fuel efficiencies could exceed 80-85%. Several features of SOFC make it attractive for utility and industrial applications: high tolerance to fuel contaminants, no expensive catalysts and direct fuel processing in the fuel cells. SOFCs also have very low emissions. Because sulphur is removed from the fuel, no SOx is emitted and since the gas-impervious electrolyte does not allow nitrogen to pass from the air electrode to the fuel electrode, the fuel is oxidised in a nitrogen-free environment, removing the possibility of NOx emissions. As with all fuel cells, a series array of individual cells is operated in what is known as a ‘stack’ (much the same as batteries) with a doped lanthanum chromite interconnect www.siliconchip.com.au Protonic ceramic fuel cell-PCFC Regenerative fuel cells These are a very new member of the fuel cell family, which could be attractive as a closed-loop form of power generation, as in the Helios solar plane featured elsewhere in this issue. Using a solar-powered electrolyser, regenerative fuel cells separate water into hydrogen and oxygen which are then fed into regenerative fuel cells, to generate electricity, heat and water. Water is then re-circulated back to the electrolyser of the regenerative fuel cell and the process repeats. These types of fuel cells are currently being researched by NASA and others worldwide. Zinc-air fuel cells In a typical zinc/air fuel cell, a gas diffusion electrode-cathode and a zinc anode are separated by an electrolyte and some form of mechanical separator. The gas diffusion electrode is a permeable membrane that allows atmospheric oxygen to pass through and be converted into hydroxyl ions and water. The hydroxyl ions travel through Need power on sites without mains access? Here is Ballard Generation System’s 250-kilowatt field trial stationary fuel cell power generator. June 2002  85 Type Electrolyte Ions Operating Power temp. generating efficiency Reaction Fuel gas Features stage Development Alkaline alkali metal hydrogen” approx up to H2 anhydrous low emissions, AFC hydroxides 60o C 70% ammonia able to use hydrogen from anhydrous ammonia Polymer polymer ion hydrogen+ approx 35-45% H2 hydrogen, exchange exchange 80o C (max 10 natural gas, membrane film ppm CO) methanol, PEMFC naptha Direct methanol DMFC polymer ion hydrogen+ approx about 40% H2 exchange 80o C film methanolwater solution mature, used by NASA operates at low temp, high I density used in Evs and homes fuel stack to 10s of kW and peripherals being developed no need for external reformer early development but promising Molten carbonate CO32650o C 60% H2CO natural gas, can reform carbonate but up to methanol, fuel internally, MCFC 85% with coal gas, exhaust heat used cogeneration naptha for cogeneration second generation fuel cells: 100kW cell under development and 1MW pilot plant performance testing underway and up to 100MW planned Phosphoric phosphoric hydrogen+ approx 35-45% H2 natural gas can use exhaust acid acid 200o C and more (max 1% methanol heat for space PAFC with CO) and water heating cogeneration mature technology, over 200 units in operation, test runs completed on 11MW plants Solid stabilised O22approx 45-60% H2CO natural gas, high density, oxide zirconium 1000o C with the methanol, reforms fuel SOGC possibility coal gas, internally, exhaust of up to naptha heat used for 85%overall cogeneration and turbines cell stack to 100 kW and peripherals under development Proton ceramic hydrogen+ 700o C N/A yet H2 fossil fuels ceramic material PCFC the electrolyte to the zinc anode and react with the zinc to form zinc oxide while the electrons can be used as a source of electric power. Although the electrochemical process is similar to the PEM fuel cell, refueling is very different and is more similar to batteries. Once the zinc fuel is depleted, the system is connected to the grid and the process is reversed, leaving pure zinc fuel pellets. This reversing process takes only about five minutes to complete, so the battery recharging time is comparable to filling your fuel tank at the service station. Tests have also been carried out on a process to regenerate the zinc oxide so that it may be reused as fuel, creating 86  Silicon Chip electrochemical oxidization of fossil fuels at anode, solid electrolyte still early stages a closed-loop system in which electricity is created as zinc and oxygen are mixed in the presence of an electrolyte, creating zinc oxide. The main advantage of zinc-air technology over batteries is its high specific energy, the key factor that determines the power potential of a battery relative to its weight. ZAFCs have been used to power electric vehicles and have delivered greater driving range than any other EV batteries of similar weight. In addition, material costs for ZAFCs and zinc-air batteries are low. Next month, we’ll look at applications and what accessories are needed to put all this wonderful potential into practical use. SC www.siliconchip.com.au