A new type of fuel cell has been developed at Lawrence Livermore National Laboratory (LLNL). Instead of using hydrogen to react with oxygen and produce electricity, the new fuel cell uses a process called direct carbon conversion. This new type of fuel cell is called a Direct Carbon Conversion Fuel Cell (DCCFC). In direct carbon conversion, extremely fine carbon particles are joined in an electrochemical process with oxygen molecules to produce pure carbon dioxide (CO2) and electricity. The carbon for this process can be obtained from any type of fossil fuel (coal, lignite, natural gas, petroleum, petroleum, coke), or even biomass.
Coal is America’s most abundant energy resource. At current needs, we have at least 250 years’ of proven reserves. Coal produces over 50% of America’s electrical power needs. It is a stable, domestic source of energy unaffected by international politics, war or terrorism. However, traditional coal-burning plants that boil water to turn electrical turbines operate at only 40% efficiency. A DCCFC operates at better than 80% and can be scaled up to facilities producing several gigawatts of electricity. Traditional coal-burning plants produce significant pollutants (though coal is vastly cleaner than it used to be). A carbon fuel cell produces pure carbon dioxide, which can be used for industrial applications or sequestered for clean disposal.
Standard Design
A fuel cell is different from a battery in two ways. First, a fuel cell does not normally store energy. Second, the reactants generating the electricity are continuously fed into the fuel cell and produce a byproduct from their interaction. The two “fuels” are fed into the fuel cell at high pressure to force their interaction across a separation medium. The following description is for a traditional hydrogen fuel cell, but the basic design and operation is the same for all types of fuel cells.
The first fuel is hydrogen, either in its pure form (H2) or as part of another compound such as methanol (CH3OH). In some fuel cells, such as a direct sodium borohydride (NaBH4) fuel cell, the hydrogen is the byproduct of the electricity-producing reaction. The hydrogen so produced can be further used in a conventional fuel cell. The hydrogen source fuel can be pure or a diffused solution. The second fuel is oxygen (O2) taken from the surrounding air.
A fuel cell has five basic parts: the anode (the negative post of the fuel cell), the cathode (the positive post of the fuel cell), the electrolyte, an anode catalyst, and a cathode catalyst. A typical fuel cell resembles a sandwich with each of these parts as a layer. At the center of the fuel cell is the electrolyte. This acts to both block electrons and allow positively charged particles (in this case, the protons of the hydrogen atoms) to pass through to the other side. For this reason, the electrolyte is usually referred to as a “proton exchange membrane” (PEM). The PEM is a thin polymer sheet, typically an ionomer-impregnated pertetrafluoroethylene PTFE film. Ionomers are thermoplastic polymers with ionic properties, such as Dupont’s Nafion, that allow only positively charged particles to pass through. Naflon’s ionic properties were created by introducing sulfonic acid groups into the bulk polymer matrix of Teflon.
On either side of the electrolyte (PEM) is an electrode, typically carbon impregnated sheets coated with a platinum powder catalyst. The catalysts are designed to facilitate the splitting of the hydrogen and oxygen molecules and their later recombination as water molecules. Together, the three layers form a membrane electrode assembly (MEA) that separates the anode and cathode sides of the fuel cell. The ionomer matrix serves as the foundation for carbon particles. The carbon particles in turn support a layer of even smaller platinum catalyst particles. This catalyst-coated membrane is the MEA. The structure and distribution of the ionomer and its interface with the catalyst layers largely determine the efficiency and performance of the fuel cell.
Both layers work together to promote the electricity-producing reaction. The catalyst on the anode side splits the hydrogen molecules and strips them of their electrons. The protons (positive hydrogen ions, H+) pass through the ionomer. Meanwhile, the two electrons from the hydrogen molecule are forced up the external power circuit since they cannot pass through the ionomer. On the other side, the platinum catalyst is splitting the oxygen molecules into negatively charged oxygen atoms. These then combine with the protons and electrons returning from the external power circuit and the oxygen atoms catalyzed by the platinum particles. The catalyst layer must be as rough and porous as possible to maximize surface area at the micro and nanoscopic level so that the maximum surface area of the platinum can be exposed to the hydrogen or oxygen.
As mentioned above, the anode conducts the electrons that are freed from the hydrogen molecules so that they can be used in an external circuit. Physically, the anode’s surface has microscopic channels etched into it surface that ensure that the hydrogen is evenly dispersed over the anode’s catalyst layer. The cathode also has microscopic channels etched into its surface that evenly distribute the oxygen to the surface of the catalyst. It also conducts the electrons returning back to the fuel cell from the external circuit to the catalyst layer. Along the cathode’s catalyst layer, the electrons, protons, and oxygen atoms recombine to form water.
The hydrogen “fuel” is forced into the anode and through the catalyst under pressure. Meanwhile, on the cathode side of the fuel cell, oxygen gas is also being forced through its catalyst. The recombination of hydrogen and oxygen into water molecules occurs on the cathode side of the barrier. Either liquid water or water vapor then exits the fuel cell. The loss of this byproduct requires continuous replenishment of the fuel cell by additional oxygen and hydrogen.
In addition to the standard hydrogen and hydrogen-derived fuel cells described above, there are several other commercially viable types of fuel cells. While all use a hydrogen + oxygen reaction to produce electricity, each differs in the kind of electrolyte used to separate the anode and the cathode sides of the fuel cell. Solid oxide fuel cells use a hard, nonporous ceramic compound as the electrolyte and operate at extremely high temperatures (1,000 degrees C). Molten carbonate fuel cells operate at high temperatures (600 degrees C) and use a liquid electrolyte composed of a molten carbonate (CO3) salt mixture suspended in a porous, chemically inert ceramic lithium aluminum oxide (LiAlO2) matrix. Phosphoric acid fuel cells use liquid phosphoric acid as an electrolyte—the acid is contained in a Teflon-bonded silicon carbide matrix—and porous carbon electrodes containing a platinum catalyst. Alkaline fuel cells use a solution of potassium hydroxide in water as the electrolyte and can use a variety of nonprecious metals as a catalyst
Producing Hydrogen
Though oxygen is easily obtained from the surrounding air, hydrogen must be produced either by team methane reforming, pyrolysis, or electrolysis. Nearly all hydrogen production today is based on fossil fuel raw materials. Worldwide, 48% of hydrogen is produced from natural gas, 30% from oil, 18% from coal, and the remaining via water electrolysis. Steam methane reforming is the primary thermo chemical means of producing hydrogen. In this method, the methane in natural gas reacts with water vapor at high temperatures to form carbon monoxide and hydrogen gases. Hydrogen can also be produced via pyrolysis or gasification of biomass resources such as agricultural residues.
Electrolysis currently comprises 4% of the world’s hydrogen production and is used mainly in areas with very cheap electricity, such as those rich in hydro or geothermal resources, or in applications requiring high purity hydrogen, such as semiconductor manufacturing. An electric charge breaks the hydrogen and oxygen bond in the water. Hydrogen then collects at the cathode and forms hydrogen gas. High temperature electrolysis (also known as steam electrolysis) is a relatively effective means of extracting hydrogen from water. It is more efficient than traditional room temperature electrolysis and it has the further advantage of being powered mostly by heat instead of electricity. Therefore, waste heat from significant power sources (such as a nuclear power plant) can be used be used to produce hydrogen.
When deciding on the source of hydrogen to run a fuel cell, it must always be remembered that hydrogen is a carrier of energy, not a source of energy. It takes considerable energy to produce the hydrogen necessary to run a fuel cell (and at high inefficiencies compared to the direct energy use of the fossil fuels used to make the hydrogen in the first place). It takes energy to produce the hydrogen which is later used to generate energy from fuel cells. For large-scale production, hydrogen costs $0.32/lb if it is consumed onsite. When hydrogen is sold on the market, the cost of liquefying the hydrogen and transporting it can increase the selling price to $1.00-1.40/lb for delivered liquid hydrogen.
According to the Bush administration’s National Hydrogen Energy Roadmap, up to 90% of all hydrogen will be refined from oil, natural gas, and other fossil fuels—in a process using energy generated by burning oil, coal, and natural gas. Hydrogen is neither cheap nor efficient compared to the direct use of fossil fuels. So long as methane is the primary source of hydrogen, it will make more sense to fill specialized car tanks with compressed methane and run the fuel cells directly off that. The resulting system uses the methane energy more efficiently, produces less total CO2, and requires little new infrastructure. Furthermore, methane is much easier to transport and handle than hydrogen.
What is needed is a fuel cell that operates directly from a fossil fuel source.
Design of a DCCFC
DCCFC technology utilizes a non-hydrogen source of fuel in its anode, aggregates of extremely fine carbon particles, from 10 to 1,000 nanometers in diameter, evenly distributed in a mixture of molten lithium, sodium, or potassium carbonate at an operating temperature of at least 750 degrees Celsius (over 1,380 degrees Fahrenheit). This new type of fuel cell pushes the efficiency of using fossil fuels for the generation of electricity to their theoretical limits. Such increases in efficiency radically reduce the amount of carbon dioxide pollutants produced per kWh. Its rate of electrical production is 1 kW per square meter (about 0.8 kW per square yard) of fuel cell surface area. Instead of hydrogen and oxygen reacting to make electricity and form water, the carbon and oxygen react to generate electricity and form carbon dioxide. While this is the same chemical reaction that occurs when carbon-based fossil fuels are burned, no combustion occurs and there is no build up of ash in the fuel source.
The carbon used by the fuel cell consists of “turbostratic” carbon particles. Turbostratic particles have a high degree of structural disorder on the nanometer scale. These particles operate like rigid electrodes when mixed in molten carbonate slurry. Using well-established industrial processes, the carbon particles can be evenly distributed through the slurry pneumatically. The gas used for this pneumatic transport can be some of the very same carbon dioxide produced by the fuel cell reactions.
The fuel on the cathode side remains oxygen molecules from the surrounding air, same as in a hydrogen fuel cell. At the anode catalyst, the carbon is converted to positively charged carbonate ions as they lose their electrons. The cathode catalyst splits oxygen molecules into negatively charged oxygen atoms. The molten salt electrolyte is held in place by a porous ceramic container that allows positive carbon ions to pass through.
The high temperature of the carbon fuel cell is necessary since the chemical reaction of carbon with oxygen to form carbon dioxide and electricity is too slow at normal temperatures. Given its high operating temperatures, a thin polymer sheet is unsuitable as an electrolyte. The electrolyte in a DCCFC is typically a molten salt, such as sodium carbonate (2Na2CO3). Further research has shown that greater efficiencies at lower operating temperatures can be achieved using a hydroxide electrolyte. Hydroxide electrolytes have higher ionic conductivity rates, require less expensive steel alloys for container walls, and produce four electrons per carbon atom versus two electrons in a standard molten salt electrolyte.
Because this is a high temperature cell, it would be best suited for stationary power grid applications. A DCCFC would not be used to power an automobile. The DCCFC could theoretically operate at near 100% efficiency under ideal conditions. This is due to the fact that no significant entropy change occurs in the overall cell reaction. Entropy is defined as the level of disorder in a system, and the turbostatic carbon particles used for fuel are already in a highly disordered state. Furthermore, the voltage across the system does not decline with the consumption of carbon fuel. This allows carbon to be consumed as it passes through the fuel cell at a maximum voltage. In contrast, standard hydrogen fuel cells operate at no more than 70% due to entropy effects and further reduced to below 50% due to voltage degradation.
Turbostratic Carbon
The more disordered the carbon atoms, the more easily they yield electrons. This is intuitively obvious since a more rigid matrix of carbon atoms will be less able to release free electrons. Turbostratic carbon, unlike carbon arranged in neatly stacked hexagonal lattices, has disordered stacking. It has the opposite nano-structure of highly graphitized carbon. This disordered stacking is a result of random rotation or displacement of ordered layers. As such, turbostratic carbon can be considered as a structure with high concentrations of planar defects. Ironically, it’s these defects that make turbostratic carbon so useful for DCCFCs.
If turbostratic carbon is the only fuel that can power a DCCFC, it can be derived from multiple sources. The easiest to obtain, and the most useful source of turbostratic carbon is carbon black. Carbon blacks have the highest electrochemical reactivity of any carbon fuel tested in a DCCFC. A large commercial industry already produces four and a half billion kilograms per year of carbon black. This material is used for automobile tires, pigments, plastics fillers, wire insulation, and other products. Carbon black does contain some residual ash (0.02 to 0.05 percent residual ash). However, this has no significant effect on system performance, as it would take the ash over 50 years to clog the cell. This is five times the operating lifetime of the other hardware that makes up the cell.
The carbon fuel particles can be produced through thermal decomposition pyrolysis of hydrocarbons. And these hydrocarbons can be a wide range of sources including coal, lignite, petroleum, natural gas, and even biomass. Pyrolysis is a method that turns hydrocarbons into hydrogen and tiny pure carbon particles. It consumes less energy at less cost than the electrolysis or steam processes required to produce hydrogen. In fact, hydrogen is a useful byproduct of the process along with the carbon used in the DCCFC. Pyrolysis at different temperatures yields different states of disorder. The degree of disorder increase as the pyrolitic temperature of formation decreases. Maximum disorder occurs at temperatures between 700 degrees Celsius and 1,000 degrees Celsius.
Given America’s vast reserves of coal and the need for a technology that uses it as cleanly as possible, coal is the preferred choice as feedstock for turbostratic carbon. However, most coal has a high sulfur and ash content. Therefore, the carbon must be extracted from coal using hydropyrolysis.
Hydropyrolysis is a two-stage process. First, the coal is pretreated with hydrogen under pressure to produce free hydrocarbons. These hydrocarbons are in turn pyrolyzed to extract carbon fuel and hydrogen byproduct. This same hydrogen can be heated and reintroduced into the hydropyrolysis stream.
The potentially cheapest source of turbostratic carbon is petroleum coke. Petroleum coke is a heavy residue, carbonization product of high-boiling hydrocarbon fractions obtained in standard petroleum processing. It is the general term for all special petroleum coke products such as green, calcined and needle petroleum coke.
A simpler method is the hydraulic cleaning of coal. The coal is first pulverized and then hydraulically separated from ash and pyrite. A bake-out process expels moderate BTU value gases, leaving behind carbon that is usable in a DCCFC. The total cost of the process is $60 per ton, or $0.008 per kWh.
However, this process leaves a greater percentage of ash in the carbon. This will require either secondary cleaning or more frequent change-out of the electrolyte in the fuel cell. At an ash content of 0.5%, the electrolyte should be changed out twice a year.
Carbon vs. Hydrogen
Most electrical power plants are fueled by coal or natural gas. These facilities operate at efficiencies of only 35% to 40%. More advanced plants operate at nearly 60% efficiency. “Efficiency” is determined by the percentage total amount of heat released when the fuel is burned completely compared to its theoretical high heat value (HHV).
The more advanced fuel cell types described above (phosphoric acid, solid oxide, molten carbonate, etc.) operate at efficiencies equal to 35% to 55% of their high heat value. Since many of these fuel cells also give off significant waste heat, this can be used to cogenerate electricity in combined heat and power (CHP) systems.
The waste heat is used to heat the steam that drives secondary turbines. Such hybrid systems push the overall efficiency to about 70%.
The carbon-air fuel cell gives off a pure stream of carbon dioxide. This is not a problem, as it may first appear. Since it is a pure stream, the CO2 can be easily captured without the additional costs of scrubbing, collection, and separation from smokestack exhausts.
Nor are there any costs associated with other pollutant common to coal combustion such as particulates. The CO2 produced by DCCFCs is only a fraction (per kilowatt hour) of what is produced by current energy facilities; and it can be sequestered in depleted underground oil reservoirs or used for industrial purposes. Transport of CO2 for sequestering or industrial use can be accomplished using existing pipelines.
Commercial Viability
LLNL foresees a scaled-up version of the DCCFC as large as a 3-gigawatt power plant. This would provide enough energy for 3 million homes with about 1 kW each. The size of this proposed plant would only be that of a two-story office building.
In addition to the use of DCCFC for primary power, these fuels cells can also be used to augment existing power plants. Such augmentation units would derive their fuel from hydrocarbon waste products from the refinery or the power plant itself. Large-scale DCCFCs are modular by nature and can be plugged into local grids as emergency backup power for businesses, hospitals, and residential facilities.
In remoter locations subject to significant voltage line losses, DCCFCs can serve as the primary power source. Though transportation application would be limited due to the cell’s operating temperature, they can still provide the energy needed for local rail lines.
Suppliers and Researchers
The University of Kentucky’s Center for Applied Energy Research (CAER) has developed a solvent extraction method for the production of carbon black from coal. The method uses no catalysts or exotic solvents, and operates at low temperatures in a self-generated atmosphere. Crushed coal, anthracene oil, and the solvent enter an extraction reactor followed by a solids separation system.
Once off-gases are removed, the solids are fed into a vacuum distillation chamber to produce carbon pitch. The process produces carbon yields (depending on the type of coal) from 45% to 75%. The total cost of the process is $200 per ton or $0.024 per kilowatt hour.
West Virginia University has also developed a solvent extraction process. In this process, an organic solvent (NMP) is used to dissolve the organic matter in the coal, leaving behind only the mineral coal. Solvent extraction removes all inorganic sulfur and most inorganic sulfur. The raw extract contains approximately 25% volatile matter in the form of turbostratic carbon.
This is further extracted by standard coking or calcining processes. Initial cost estimates for this process appear to be very favorable with a cost to produce of $174 per ton of extract (which is equivalent to $6 per million BTUs). The total cost of the process is $140 per ton or $0.02 per kWh.
Scientific Applications Research Associates (SARA) Inc. is a leader in the development of DCCFC utilizing metal hydroxide electrolytes. Its breakthrough involves a method of preventing the absorption of CO2 by hydrates.
SARA discovered that absorption can be prevented by a high water concentration in the molten hydroxide. Its Mark II and Mark III fuel cells utilize a nonstandard configuration with the carbon source cathode in a center cylinder surrounded by the molten salt electrolyte. Oxygen is provided by an air bubbler feed at the bottom of the cell.
Conclusions
Direct carbon conversion is cutting-edge fuel cell technology. Further research is required to simplify the fuel preparation process (carbon black pyrolysis, electrolyte reprocessing), to scale up the process for power grid applications efficiently, and to determine the most cost-effective means of carbon capture and storage. DE
DANIEL P. DUFFY is an environmental engineer in Cincinnati, OH.
DE - May/June 2006
|
