Fuel cells in the test laboratory: A researcher at Siemens Corporate Technology in Erlangen measures the electrical characteristics of new solid oxide fuel cells; detail (left)
If you want to know what the future of electric power will be like, look to the past. Electrification did not begin its triumphant advance with large, centralized power plants like the ones that power the globe today, but with electric generators driven by mill wheels, steam engines and combustion engines. Anyone who needed electric power produced it on the spot. But a national electric power network has advantages: Power peaks are averaged out and less overall generating capacity is required. So it made sense for public agencies to take on the task and establish power generation and transmission companies with countrywide networks. In the process, power plants became ever larger, because the higher the power output, the lower the per-unit cost. Today, however, the watchword is decentralization. RWE, Germany's largest electric power company, is even showing ads that envision a fuel cell bought in a supermarket. RWE predicts that by 2015 about 10 % of electricity in Germany will be generated by fuel cells (see interview). The company plans to market compact power plants for private households by 2010. Its competitor, Düsseldorf-based E.ON, is more conservative. It projects that by 2025 fuel cells will supply about 20 TWh per year in Germanyi.e. 4 % of Germany's total power demand, but 15 % of the power consumed in households. Companies around the globe can provide the required equipment. In North America, companies like Ballard Power, Plug Power, H-Power and Fuel Cell Technologies are crowding into the market. Siemens Westinghouse is building a plant in Pittsburgh, Pennsylvania that will produce fuel cells. In Switzerland, Sulzer Hexis has embraced the concept of distributed power generation with fuel cells. And German heating system specialist Vaillant has teamed up with New Jersey-based Plug Power to develop a fuel cell for power and heat generation in everyone's basement. The two companies regard such individualization as a megatrend.
What's behind this paradigm shift? "The political and economic environments have changed dramatically," says Dr. Georg Rosenbauer, an energy expert in Strategic Marketing at Siemens Corporate Technology. To begin with, the electricity markets in Europe have been extensively liberalized, which has fundamentally changed the relationship between the electric utility companies and their customers. Political factors that vary from country to country are also exerting pressure to reduce emissions of greenhouse gases such as carbon dioxide (CO2). Rosenbauer is convinced that in the future a substantial proportion of all primary energy will be generated in a distributed fashionbut that there will still be room for large power plants.
"All the players understand that energy is the key to continuing economic and, above all, sustainable development, especially in the long term," says Dr. Thomas Hamacher, energy expert at the Max Planck Institute for Plasma Physics in Garching near Munich. Energy demand is expected to continue to increase. Realistically, up to ten billion people will require electricity in 2050. The required investments should, however, be kept within manageable limits. And what's more, "The risks must be manageable and acceptable."
In its World Energy Outlook 2000, the International Energy Agency (IEA) predicts that primary energy demand will grow by 2 % per year through 2020. Power trading will also increase. On the other hand, alternative power sources are expected to continue to play a subordinate role. The proportion of primary energy contributed by solar, wind and biomass-generated power, which amounts to 2 % worldwide today, will rise to nearly 3 % by 2020. Fossil energy carriers contribute 90 %, nuclear power plants 7 %. "We are currently at the peak of the fossil fuel era, and this will continue to be the case for another 20 to 30 years," projects Hamacher.
With a 3 % annual rate of growth, natural gas consumption is expected to continue overproportionally, according to the IEA, which is part of the Organization for Economic Cooperation and Development (OECD). However, energy intensitythe consumption of energy relative to productivityamong OECD countries will continue to decline by 1.1 % annually, as it has been doing since 1970. This reflects improvements in the utilization of energy carriers.
Nevertheless, these trends will not necessarily lead to decentralization. But liberalization, which, unlike in Europe, is still in its early stages in the U.S., will increase the competitive drive to maximize cost efficiency, and consequently energy efficiency. State-of-the-art combined-cycle (gas and steam turbine) power plants meet this requirement: They are economical to build and attain an efficiency of nearly 60 %, which represents the relation of useful power to input power (see article The Power of Small Steps). Furthermore, their energy yield can be increased if waste heat is utilized directly in buildings or industrial processes. As a result, their overall efficiency (electric and thermal) can reach 80 %. Germany will have invested some 4.5 billion in combined heat and power generation (CHP) by 2010.
Transporting heat over long distances, however, is complex and expensive. On the other hand, many private homes and nearly all housing developments in Germany are connected to the gas network. Instead of generating electricity in large power plants and then transporting it several hundred kilometers, it can be more economical to decentralize power generation and to use the generated heat locally. This is a significant driver for distributed power plants, especially in the European market. In the North American market, demand for more reliable power sources is expected to promote the technology. Decentralized power generation from underground gas lines is substantially more reliable than overhead power lines, which are trouble-prone. This fact also generates competition between electric and gas utilities (see illustration). Natural gas has the additional advantage that it can be stored and that power peaks are therefore easier to smooth out. And gas-fueled power plants have an additional environmental advantage: Gas emits only 350 g of CO2 per kilowatt-hourless than half the amount emitted by oil and coal-fueled power plants (see illustration).
But combined-cycle power plants are not the solution for decentralization because they can't be scaled down indefinitely. At some point, the cost advantage of a large number of units again becomes dominant. A gas turbine in a Siemens power plant that produces several hundred megawatts is more than ten meters long. But the investment would remain relatively large even if the turbine were scaled down. The bottom line looks much better for fuel cell power plants. These systems are modular right down to the individual membrane electrode units. In other words, the smaller the plant, the lower the investment.
Starting in 2004, Vaillant is planning to market affordable fuel cells even for private users. "Our residential fuel cell delivers up to 4.6 kW of electric power, and in combination with a conventional gas burner, up to 50 kW of thermal power," explains Vaillant's Stefan Jakubik. Overall efficiency of such units surpasses 80 %. Field trials currently under way in Germany will later be extended to other countries. Vaillant is basing its strategy on proton-conducting PEM (polymer electrolyte membrane) fuel cells. Larger power plants based on solid oxide fuel cells (SOFCs) can supply buildings or entire communities with a very high level of energy utilization and produce power outputs in the megawatt range. While PEM cells are operated at less than 100 °C and fueled by hydrogen derived from natural gas using reforming techniques, SOFC power plants operate optimally at nearly 1,000 °C and can be fueled directly with natural gas that has been internally converted into hydrogen.
In both cases, CO2 is produced in addition to water. However, thanks to the fuel cell's higher efficiency, the quantities are relatively small. No other emissions, such as nitrogen oxides, sulfur dioxide or particulates, are produced. The main drawback of all types of fuel cells, however, is their still exorbitant cost. In current demonstration systems, each kilowatt of electric power costs up to $20,000. To compete with present-day power plants, an investment of $1,500 per kilowatt output would be required. That's the benchmark value all manufacturers go byincluding Siemens.
Siemens is the worldwide leader in SOFC technology. A 110-kW system from Siemens Westinghouse Power Generation has been in continual operation for more than three years. It is currently used in the RWE network, and will be replaced this summer by a 300-kW power plant. The latter is equipped with an additional micro gas turbine powered by hot waste gas. As a result, its electrical efficiency has increased to more than 55 % (see box, for an overview of systems now being planned). In Pittsburgh, Siemens Westinghouse is building a factory to produce power plants based on the high-temperature fuel cell. "Starting in 2004, the first systems using these components are slated for series production," says Dr. Thomas Voigt, who is in charge of SOFC fuel cell activities at Siemens. Pittsburgh was a good choice as a location for this plant. According to Georg Rosenbauer, the fuel cell will enter the world market in the U.S. for reasons particular to this country. Unlike Europe, the U.S. does not have excess power generating capacity, and the network is trouble-prone due to its wide geographic coverage and high level of utilization. For example, after an ice storm lasting several days during the winter of 1998, over three million people in the northeastern U.S. and Canada were without electric power, some of them for weeks. But even without natural catastrophes, power failures are not uncommon in the U.S. The power crisis in California, which was the result of a failed liberalization program, has made the problem more acute. "With this level of stress and inconvenience, many people won't mind spending more for a power plant in their own home," notes Rosenbauer. What's more, Americans traditionally like independenceand that's a point in favor of a power plant in the private home or housing subdivision.
The initial systems to be produced in volume by Siemens will cost between $4,500 and $5,000 per generated kilowatt. "We believe there are customers who are willing to pay higher prices, especially if they value higher reliability," notes Voigt. Starting in 2004, about one fuel cell power plant will be produced in Pittsburgh every week. As experience accrues, costs should decline. But more decisive measures will be needed before costs decline to $1,500 per kilowatt.
New design for solid oxide fuel cells: Flat sheets containing several tubes deliver three times the power density
One way to optimize power yield is an innovative type of fuel cell produced on the basis of a design by Siemens Westinghouse. In their present configuration, these fuel cells are shaped like a tube that's open at one end and looks like a broomstick (see illustration). Air heated to 950 °C is blown into these tubes (in some systems under pressure), which are surrounded by a flow of natural gas. With a surface of nearly 850 cm², each tube generates a potential difference of slightly over 0.5 V and a current of 160 A. To generate higher voltages, the tubes are electrically connected by conductive struts, though this results in some ohmic losses due to the struts' resistance. The next generation of such fuel cells will thus be assembled as flat sheets, within which several tubes are integrated side by side. Due to the shorter distances traveled by the current, this design lowers the resistance per square centimeter to 0.45 ohms, compared to 1 ohm in the round fuel cell. With additonal improvements, a threefold increase in power density is possible, says Voigt. For a given power output this would mean massive savings in materials, at present the largest cost item.
Siemens Westinghouse is also exploring the potential uses of PEM cells. These fuel cells are particularly suitable for use in automobiles and even as mini cells in mobile phones and laptops (see article Filling Stations for Cell Phones). Prototypes already exist in Siemens laboratories, but cost, size and weight are even more critical in such applications than in stationary use. However, PEM cells have an interesting niche market in submarines. Weighing some eight tons, these cells are the world's largest PEM cells. This year the German navy will begin to operate the first submarine equipped with a Siemens fuel cell. A number of other countries have also placed order. Here, the principal advantage of fuel cell drives is that submarines can stay underwater much longer than with lead batteries.
A Siemens researcher is checking the new, extremely flat solid oxide fuel cells
Whether mobile or stationary, the fuel cell willin the not-too-distant futurechange the way power is supplied. But this will tend to be a gentle revolution because of high initial production costs. Rosenbauer envisions that domestic fuel cells, smaller fuel cell power plants and renewable energy sources will supply power in an intelligent alliance with large power plants. These distributed power systems will feed any excess production into the power grida feature that will necessitate a sophisticated management system.
Many smaller fuel cells, interconnected to form a virtual power plant, would even provide grid companies with the opportunity to use a selective control system to stabilize the network and thus minimize losses (see article The Virtual Power Plant). Operating costs can also be reduced, as it will be possible to monitor mini power plants from a central control room and coordinate maintenance activities.
In view of all this, power companies have therefore abandoned their initial reservations about distributed power generation and are now actually spearheading the movement. Eventually, they will be able to rent out fuel cells to customers, while reaping the advantages of this technology themselves. "There will be no palace revolt," says Rosenbauer, "but fuel cells will irreversibly alter the energy landscape."
Norbert Aschenbrenner
How PEM and SOFC Fuel Cells Work
In fuel cells, a reaction that would otherwise be explosivethe combination of hydrogen and oxygen to form watertakes place in a controlled manner. This is made possible by having the hydrogen or oxygen ions migrate slowly through a membrane. The electrons that are "left behind" flow through an external conductor as a direct current. A fuel cell thus acts like a battery. In PEM cells (polymer electrolyte membrane, above), hydrogen atoms surrender their electrons, diffuse through a plastic membrane and react with atmospheric oxygen on the other side. In solid oxide fuel cells (left), the oxygen ions migrate. This type of fuel cell does not require pure hydrogen to operate, and can thus be fueled directly with natural gas, which dissociates into hydrogen and carbon monoxide at the fuel cell's operating temperature of almost 1,000 °C. This high temperature is necessary in order to make the ceramic membranea yttrium-doped zirconium oxidepermeable to oxygen atoms.