"A Renaissance for Nuclear Energy?"
Andrew C. Kadak
The California electricity "crisis" gave national prominence, once again, to the issue of energy supply. The "crisis" apparently came and went and will likely soon be forgotten. What it did accomplish however was a more lively discussion of the importance of supply, recognizing the ever increasing demand as we electrify. In this discussion of demand came the realization that approximately 20% of the nationUs electricity was being generated by nuclear energy which was not subject to the escalating cost of natural gas that drove electricity prices far above what the public would tolerate. The net consequence of a number of factors, such as a faulty deregulation scheme that drove one of the major electric companies in California into bankruptcy, were rolling blackouts due to lack of generation at any price. Given that, in California, the construction of new plants of any type was frowned upon and the regulatory climate somewhat hostile, no company was interested in making generation investments. So, a "crisis" was created that has for the time being abated at a cost to the economy of at least $ 5 Billion.
In the rest of the world, serious people are debating the existence and implications of increasing greenhouse gases in our environment due to the burning of these same fossil fuels. While the environmental ministers of nations from around the world seek to find ways to meet the 1992 Kyoto accords which call for reductions in CO2 and other greenhouse gases to 10% below 1990 levels, the reality almost 10 years later is that CO2 emissions have not decreased at all but increased by 10%. As most know, one of the key advantages of nuclear energy is that it is essentially a greenhouse-gas-emission-free technology. Yet, at its most recent meeting of the Conference of the Parties in Bonn in July, these same environmental ministers voted to specifically exclude nuclear energy fom helping address the global warming problem. Clearly, there is something wrong here since, in the United States, nuclear energy provided over 69% of the emission-free generation, far exceeding the 30% hydroelectric power. Solar and other renewables provide the rest (~1%).
For many years, nuclear energy, while arguably a -CO2 emitting energy source, has been judged to be unacceptable for reasons of safety, unstable regulatory climate, a lack of a waste disposal solution and, more recently, economics. In recent years, however, the nuclear industry has made a remarkable turnaround. While a number of older plants have been shutdown for largely economic reasons, the 104 operating nuclear plantsU performance has increased to the point, that as an overall fleet, its capacity factor was 89% in 2000. This means that these plants were operating full power for over 89% of the year. This improvement in the last 10 years is essentially the same as building 23 new 1,000 Mwe plants in that time period, based on historical performance averages. In addition, all safety statistics, as measured by the Nuclear Regulatory Commission, have shown dramatic improvements as well. The Three Mile Island accident occurred over 22 years ago. The image of nuclear energy as an unsafe technology still persists. Yet the record is quite the opposite.
There has not been a new order for a nuclear plant since the mid 1970Us. The reason for the lack of new orders was the high capital cost. When operating in a difficult regulatory environment, utility executives simply avoided new nuclear construction and went to the cheapest and fastest way to make on-line generation available, which was natural gas. Combined cycle gas plants were the generation source of choice for many years for those companies that needed to build plants. Hence the beginning of the "crisis."
Today, utility executives still do not have new nuclear plant construction in their future plans even though the regulatory regime has stabilized. Nuclear plants are performing extremely well. Safety issues have been addressed with no new issues emerging and slow progress is being made to finally dispose of spent fuel at Yucca Mountain. What has happened is a consolidation of the utility and nuclear industry with some larger utilities purchasing existing nuclear plants from companies that do not want to be in the business. To address the inevitable problem of replacing existing generation, utilities have chosen to re-license existing plants from the current 40 years to 60 years. Several nuclear plants have applied and received Nuclear Regulatory Commission approval to do so. These extensions will allow utilities to continue to use these plants as long as they are economic and continue to be safely operated. Unfortunately, we still donUt see anybody ready to build a new nuclear plant. The reason is there is no new nuclear plant on the market that can compete with natural gas or mine mouth coal plants.
However, there are developments in two parts of the world that are aimed at changing that situation. The objective of these related efforts is to design, license, and build a nuclear power plant that can compete with natural gas. The two projects started independently but reached the same conclusions P that small modular high temperature gas reactors, that are naturally safe, can be built in two to three years and can compete in the electricity market. While the basic technology is over 20 years old, the application and concepts are quite new. The leader in this effort is ESKOM, the 5th largest utility in the world, located in South Africa. The other effort is being lead by the Massachusetts Institute of Technology with support from the Idaho Nuclear Engineering and Environmental Laboratory.
The nuclear energy plant that both groups are developing is a modular 110 Mwe high temperature, pebble bed reactor, using helium gas as a coolant and conversion fluid and gas turbine technology. The fundamental concept of the reactor is that it takes advantage of the high temperature properties of helium which permit thermal efficiencies upwards of 50%. It utilizes an online refueling system that can yield capacity factors in the range of 95%. Its modularity design concepts, in which the balance of plant can fit on a flat bed truck be shipped from the factory, allows for a 2 to 3 year construction period, with expansion capabilities to meet merchant plant or large utility demand projections.
Economic projections for the plant in South Africa indicate capital costs of between $ 800 to $ 1,000 per kilowatt. Staffing levels for an 1100 Mwe, 10 unit module are about 85, and fuel costs about 0.5 cents/kwhr. When all is combined, the total busbar cost of power ranges from 1.6 to 2 cents/kwhr. Very preliminary estimates in the US for the MIT project show higher costs but, on a relative, scale the numbers are well within the range of competitive prices with new combined cycle plants, at 3.3 cents per kilowatt hour.
What is a Pebble Bed Reactor?
Pebble bed reactors were developed in Germany over 20 years ago. At the Juelich Research Center, the AVR pebble bed research reactor rated at 40 Mwth and 15 Mwe, operated for 22 years, demonstrating that this technology works. The reactor produced heat by passing helium gas through a reactor core consisting of uranium fueled pebbles. A steam generator was used to generate electricity through a conventional steam electric plant. Germany also built a 300 Mwe version of the pebble bed reactor but it suffered some early mechanical and political problems that eventually lead to its shutdown. In December of 2000, the Institute of Nuclear Energy Technolog, of Tsinghua University in Beijing China, achieved first criticality of their 10 Mwth Mwe pebble bed research reactor. In the Netherlands, the Petten Research Institute is developing pebble bed reactors for industrial applications in the range of 15 Mwth. The attraction to this technology is its safety, simplicity in operation, modularity and economics.
Advances in basic reactor and helium gas turbine technology have produced a new version of the pebble bed reactor concept. Instead of wasting heat by using steam to produce electricity, new designs are going to direct or indirect cycle helium gas turbines to produce electricity. By avoiding the use of high temperature water all the difficulties associated with maintaining high temperature water systems is eliminated. The optimum size for a pebble bed was concluded to be about 250 Mwth thermal to allow for rapid and modular construction as well as maintaining its inherent safety features. These designs do not require expensive and complicated emergency core cooling systems since the core can not melt. These advances have lead ESKOM and the MIT team to independently conclude that the modular pebble bed reactor can meet the safety and economic requirements for a new generation. Each group is working cooperatively to develop and demonstrate the technology for commercial application. The time line for demonstration in South Africa is to have the first reference plant in startup testing in 2005 and commercial operation in 2006. Presently, a detailed design feasibility study is underway that will lead to a decision in November of 2001 as to whether the South African project will continue to build the demonstration plant. Licensing submittals are being prepared for submission to the South African nuclear regulator. In short, within the next 5 years, should this project be successful, there will be a credible nuclear alternative to fossil fuels that is projected to be competitive with natural gas even under the old, uninflated prices.
A pebble bed reactor is graphically illustrated in Figure 1. The reactor core contains approximately 360,000 uranium fueled pebbles about the size of tennis balls. Each pebble contains 9 grams of low enriched uranium in 10,000 tiny grains of sand-like microsphere coated particles, each with its own hard silicon carbide shell. These microspheres are embedded in a graphite matrix material as shown in Figure 2. The unique feature of pebble bed reactors is the online refueling capability in which the pebbles are recirculated with checks on integrity and consumption of uranium. This system allows new fuel to be inserted during operation and used fuel to be discharged and stored on site for the life of the plant. It is projected that each pebble will pass through the reactor 10 times before discharge in a three year period, on average. Due to the on-line refueling capability, plant maintenance outages are now required every 6 years.
The key reactor specifications for the modular pebble bed reactor are shown in Table 1.
Nuclear Specifications for the MIT Pebble Bed Reactor
Thermal Power 250 MW
Core Height 10.0 m
Core Diameter 3.5 m
Pressure Vessel Height 16 m
Pressure Vessel Diameter 5.6 m
Number of Fuel Pebbles 360,000
Microspheres/Fuel Pebble 11,000
Fuel Pebble Diameter 60 mm
Fuel Pebble Enrichment 8%
Uranium Mass/Fuel Pebble 7 g
Helium Mass Flow Rate 120 kg/s (100% power)
Helium entry/exit temperatures 450 C/850 C
Helium Pressure 80 bar
Mean Power Density 3.54 MW/cubic meter
Number of Control Rods 6
Number of Absorber Ball Systems 18
The pebbles are located in the reactor core structure whose cross section is shown in Figure 3. The internals are made of carbon blocks which act as a reflector and structural support for the pebble bed. A picture of the Chinese internal carbon structure is shown in Figure 4. Please note the pebbles in the bottom of the central portion of the graphite discharge section.
Balance of Plant
There are two options under development P a direct cycle helium gas turbine system being developed by ESKOM and an indirect P helium to helium intermediate heat exchanger gas turbine system being developed by MIT. Each has its advantages and disadvantages with the key being the bottom line cost as measured in cents per kilowatt-hour. The direct cycle plant configuration of the ESKOM PBMR design is shown in Figure 5. In this design there are essentially two large vessels P one containing the reactor and the other the balance of plant. The indirect cycle being developed by MIT is shown in Figure 6. Conceptually, the MIT turbomachinery module could be built in a factory and truck shipped to the site for simple assembly. If this modularity strategy is realized, it would revolutionize how nuclear energy plants are built. The MIT schematic of the thermo-hydaulic system for its indirect cycle is shown on Figure 7. The basis for this preliminary design is that all components are commercially available today. Advanced designs are being developed to simplify the plant even further.
Should there be a need for an 1,100 Mwe plant, 10 modules could be built at the same site. The concept calls for a single control operating all 10 units through an advanced control system employing many of the multi-plant lessons of modern gas fired power plants in terms on modularity and automatic operation. Construction plans and schedules were developed to refine the cost estimates and schedule expectations. The preliminary schedule calls for getting the first unit on line in slightly over 2 years with additional modules coming on line every three months. A unique feature of this modularity approach is that it allows one to generate income during construction as opposed to interest during construction.
The Safety Case
The basis for the safety of pebble bed reactors is founded on two principles. The first is the very low power density of the reactor which means that the amount of energy and heat produced is volumetrically low and that there are natural mechanisms such as conductive and radiative heat transfer that will remove the heat even if no core cooling is provided. This is significant since the temperature that is reached in a complete loss of coolant accident is far below the core melt temperature and it takes about 70 to 80 hours to reach the peak temperature. Hence the conclusion that the core will not melt is valid and supported by tests and analysis performed in Germany, Japan, South Africa and the US.
The second principle is that the silicon carbide, which forms the tiny containment for each of the 10,000 fuel particles in a pebble, needs to be of sufficient quality that it can retain the fission products. In tests performed to date on fuel reliability, it has been shown that micospheres can be routinely manufactured with initial defects of less than 1 in 10,000. In safety analyses, it is assumed therefore that 1 in 10,000 of these microspheres has a defect that would release the fission products into the coolant. Since the amount of fuel in each particle is very small, only 0.0007 grams, even with this assumption and under accident conditions, the release from the core would be so low that no offsite emergency plan would be required. In essence, it is recognized that the fuel can not be made perfectly but the plant is still safe because is has natural safety features that prevent meltdowns. Manufacturing fuel quality is a key factor in the safety of high temperature gas reactors.
The other safety issue that needs to be addressed with all graphite reactors is that of air ingress. When at high temperatures, oxygen reacts with carbon to form C0 and CO2. This oxidation and corrosion of the graphite is both an exothermic and endothermic reaction depending upon the conditions. Analyses and tests in Germany have shown that it is very difficult to "burn" the graphite in the traditional sense, but it can be corroded and consumed. Some have made references to Chernobyl as an example of the problems with graphite reactors. Fortunately, however, the Chernobyl design is radically different than the pebble bed in that the pebble bed does not contain water (steam explosions), nor zirconium that really burns in air at high temperature, and the pebble bed reactor can not reach the temperatures for melting fuel P all of which fueled the Chernobyl fire.
The key issue for the pebble bed reactor is the amount of air available in the core from the reactor cavity and whether a chimney can form allowing for a flow of air to the graphite internal structure and fuel balls. Tests and analyses have shown that at these temperatures graphite is corroded and consumed but the natural circulation required for "burning" is not likely due to the resistance of the pebble bed to natural circulation flow. The corrosion process is more of a diffusion process. MIT is now performing confirmatory analyses to understand the fundamental behavior of air flow into a pebble bed reactor under the assumption of a major break in the circulating pipes or vessels.
No matter what the environmental, public health, safety and energy security advantages that nuclear energy may offer, if the product is not competitive, it will not be used. The MIT team used a comparative analysis of energy alternatives that was performed in 1992 by the Nuclear Energy Institute. The results of this comparative analysis for capital costs for a 10 unit modular plant show that the base plant overnight construction cost was $ 1.65 Billion. Applying a contingency of 23 % and an overall cost of money of 9.47%, total capital cost estimate was $ 2.3 Billion or about $ 2,000 /kw installed. On a per unit module, for a 110 Mwe plant the capital cost is estimated to be about $ 200 million. This estimate is approximately double that of the PBMR proposed by ESKOM.
If construction costs were all that mattered, the pebble bed reactor would clearly not be economic compared to natural gas plants. However, when one includes the fuel and operating and maintenance costs, since the pebble bed plant requires far fewer staff than conventional reactors due to their simplicity, the total cost of power is estimated to be 3.3 cents per kilowatt hour, well within the competitive range for new natural gas plants.
The financial community is justifiably skittish about nuclear investments due to the huge right-offs that were required for the latest generation of nuclear plants. The beauty of this small modular plant is the initial investment for a module may range from $ 100 to $ 200 million dollars. This is not an astronomical amount of money. Also the plant should be producing electricity within two and a half years, a fairly short time to be nervous about getting a return. These two factors should provide sufficient confidence to make the required investments, as opposed to the billions and 6 years to see similar generation and returns for conventional light water reactor plants.
Nuclear Waste Disposal
The lack of a final repository for used nuclear fuel has been cited by some as a major obstacle to the building of new nuclear plants. While the need for a permanent waste disposal facility is real for existing plants and future plants, the progress being made in the US and in other parts of the world in actually siting a number of these facilities is encouraging. As most scientists and engineers realize, geological disposal is the right strategy. In the US, studies of the Yucca Mountain repository site on the grounds of the former Nevada Nuclear Weapons Test site continue to show that this location is a good site for the burial of nuclear wastes for tens of thousands of years. This year, the Department of Energy will send to President Bush a recommendation to proceed with licensing. In Sweden and Finland, underground repository experimental facilities have gotten local community support to actually build a test facility. Some nations are looking to reprocessing and long term storage since they do not feel the urgent need to have a facility in operation since the quantity of spent fuel in storage today is still relatively small. One repository could store all the spent nuclear fuel from all this nationUs operating nuclear reactors for their 40 year licensed lives. Under optimistic circumstances, a repository at Yucca Mountain could be open by 2010 according to DOE. After 15 years of study and exploration of Yucca Mountain, it is likely that the facility can be safely built at the site for long term storage of spent fuel for hundreds of thousands of years.
As the California electricity crisis reaches the consciousness of the American public, the politics of nuclear energy will surely improve. Right now, the industry does not have a product that can compete in the market place even with currently high natural gas prices. It is vital to develop that product and not wait for the price of natural gas to be so high that nuclear energy becomes competitive, since it will hurt the US and world economy. MIT and the ESKOM PBMR projects are working to provide this product that is not only competitive but also will gain the publicUs support due to its safety advantages. A lot of work is still required to demonstrate the capabilities of the pebble bed, but all work to date by both ESKOM, its US partners, and MIT continue to show positive results. While not a replacement for natural gas today, in the next five to ten years, nuclear plants powered by pebbles may not
Andrew C. Kadak, Ph.D.
Professor of the Practice, Nuclear Engineering Department
Massachusetts Institute of Technology
Room 24-207 A, Massachusetts Ave., Cambridge, MA 02139
617-253-0166, fax: 617-258-8863
- AVR P Experimental High Temperature Reactor: 21 Years of Successful Operation for a Future Energy Technology, Association of German Engineers (VDI) P the Society for Energy Technologies, Dusseldorf, 1990.
- "Advanced Design Nuclear Power Plants: Competitive, Economic Electricity, Nuclear Energy Institute, 1992.
- "Evaluation of the Gas Turbine Helium Reactor" - DOE-HTGR-90380 - Dec. 1993.
- MIT Nuclear Engineering Web site: web.mit.edu/pebble-bed/
- G. Melese, R. Katz, Thermal and Flow Design of Helium Cooled Reactors, American Nuclear Society, 1984.
- R.A. Knief, Nuclear Engineering, Theory and Technology of Commercial Nuclear Power, second edition, 1992, Hemisphere Publishing Corp.