GAO st~p1 t’llllwl’ l!)!N NUCLEAR RESiQiRCH AND DEVELOPMENT Shippingport Decommissioning- How Applicable Are the Lessons Learned? (;,A20 KWD-W-2i)Sf --- United States General Accounting Office Washington, D.C. 20548 Resources, Community, and Economic Development Division B-239632.2 September 4, 1990 The Honorable Robert A. Roe Chairman, Committee on Science, Space, and Technology House of Representatives The Honorable Robert S. Walker Ranking Minority Member, Committee on Science, Space, and Technology House of Representatives In July 1989 you asked for information about the Department of Energy’s (DOE) decommissioning of the Shippingport, Pennsylvania, nuclear power plant. You wanted to know whether DOE had met the goals described at the July 1986 hearings before your Committee. DOE’S goals were to . demonstrate that a large nuclear plant can be decommissioned safely and within the costs ($98.3 million) and time frame (April 1990) established, . optimize contractor involvement to help transfer information to the pri- vate sector, and . develop information to assist the nuclear industry with future decom- missioning projects. Specifically, you asked us to answer nine questions relating to these goals (app. II provides detailed answers to your questions). We are also providing some perspective on additional information that could assist the commercial nuclear power industry that was not addressed by Ship- pingport. We previously addressed some of the lessons learned from Shippingport in a June 1990 report to Representative Fazio who was interested in the relationship between Shippingport and the Ranch0 Seco, California, plant1 This report provides more details about Ship pingport’s decommissioning and addresses the usefulness of Ship- pingport’s decommissioning to such commercial plants as Pathfinder, Peach Bottom, and Fort St. Vrain located in South Dakota, Penn- sylvania, and Colorado, respectively. ‘Nuclear R&D: UsefuIness of Information From Shippingport Decommissioning for Ranch0 Seco (GAO/KcEb-90-171, June 7,199O). Page 1 GAO/RCED-W208 Shippingport Decommissioning B233332.2 DOE generally met the goals that it had established for Shippingport. It Results in Brief completed all decommissioning activities in December 1989-4 months ahead of schedule-at a cost of $91.3 million-$7 million under its 1986 estimated cost. According to some utility representatives, the most sig- nificant benefit of Shippingport was that DOE demonstrated that tech- nology existed to decommission a plant within the costs and time frame established. In addition, DOE used over eight contractors on the project and produced numerous annual or topical reports that officials believe will be useful to the commercial nuclear industry. Although Shippingport increased the knowledge for decommissioning nuclear power plants, the benefits of the lessons learned will vary depending upon the timing and the decommissioning approaches selected by utilities. Very few utilities will be able to decommission their plants the way DOE decommissioned Shippingport, and it is possible that newer technology may be available by the time utilities do so. To illus- trate, Shippingport was much smaller and less radioactively contami- nated than other plants, and DOE removed the most highly radioactive component, the reactor pressure vessel, in one piece. Utilities operating commercial plants will probably have to disassemble (cut-up) the reactor pressure vessels because of their much larger size. Also, DOE dis- posed of all the low-level radioactive waste from the decommissioning activities at its Hanford, Washington, facility. Utilities will have to dis- pose of waste at commercial sites at substantially higher costs. In the mid-1950s, DOE and the Duquesne Light Company entered into a Overview of the contract to build Shippingport and demonstrate electricity generation Shippingport Project using nuclear power. On December 2, 1957, Shippingport, a 72-megawatt pressurized water reactor,z became the nation’s first operating nuclear power plant. Over its 25-year life, the plant operated for about 80,324 hours and produced about 7.4-billion kilowatt-hours of electricity. Under the contract, DOE owned the reactor and steam-generating por- tions of the plant while the utility owned the electricity-generating por- tion. According to a DOE official, the contract required DOE to return the site to safe conditions on or before 1994. Accordingly, in September 1985 DOE began the physical decommissioning of Shippingport. DOE com- pleted these activities in July 1989, including dismantlement of the %-essurized water reactors are those cooM by water that is kept at hiih pressure to prevent it from boiling. The water passes through the nuclear fuel and is heated. The heat is then transferred to a secondary system where steam is pmduced. Page 2 GAO/llCED~208 Shippineport Decommissioning B-239632.2 nonradioactive structures; certified in October 1989 that the site was radiologically safe; and issued a final report on the project in December 1989. In addition, DOE removed the fuel and sent it to its Idaho National Engi- neering Laboratory and disposed of about 216,000 cubic feet of low- level radioactive or mixed (radioactive and hazardous) waste at its Han- ford, Washington, facility. Also, DOE removed the reactor pressure vessel intact and shipped it by barge to Hanford for disposal. the decommissioning activities in December 1989-4 months ahead of schedule-and $7 million under the estimated $98.3 million cost. Also, DOE used over eight contractors for various decommissioning activities to optimize contractor involvement and developed an extensive amount of data that officials believe will help future decommissioning projects. Between Shippingport significant differences between Shippingport and other reactors and the and Commercial manner in which Shippingport was decommissioned, it is questionable whether the lessons learned can be extensively applied to larger, more Plants contaminated nuclear power plants that will be decommissioned in the future. For example, Shippingport was different from commercial plants because (1) the plant was more radiologically clean than other plants at the time of its shutdown, (2) DOE disposed of the pressure vessel in one piece instead of cutting it up or letting the radiation decay over many years before starting decommissioning, (3) DOE had predetermined sites to dispose of the spent fuel and low-level3 and mixed waste, and (4) DOE had an elaborate management structure to conduct and oversee the decommissioning activities. 3Low-level waste is waste that is not classified as uranium mill tailin@, high-level waste, or spent fuel and consists of discarded tools, rags, machinery, paper, protective clothing, and other items. Page 3 GAO/~BO-2OS Shippingport Decommissioning E239632.2 Low Radioactive Over the plant’s lifetime, some decontamination activities had been con- ducted; therefore, Shippingport-including the reactor pressure Contamination vessel4-was more radiologically clean than might be expected for a commercial plant. DOE estimates that at the time of shutdown the reactor pressure vessel contained about 30,000 curies” of radioactive material. Table 1 shows the estimated curie content for four plants that have been shut down, are awaiting the start of decommissioning, or have been partially decommissioned. Table 1: Comprrison of Shipph’qport to Four 0th~ Plant@ ElOCtdC enentod7 n Amount of Hours bi I ion kUowatt- Plants Maaawatts TEtZ! curios OCWOtOd houm1 Shippingport 72 PWRb >30,000c 80,324 >7.4* Fort St. Vrain 330 HTGC* 900,ooo 21,360 4.3 Pathfinder 62 BWR’ 30,ooo 12,000 .l Peach Bottom 40 HTGC >3.ooo.oooQ 32,375 1.4 Ranch0 Seco 913 PWR >9,ooo,ooo” 51,595 44.0 aAlthough the plants shown in table 1 differ in design and size from Shippingport, the information is useful for illustrative purposes. bPressurized water reactor. CGreaterthan 30,000 curies. *Greater than 7.4 billion kilowatt-hours eHigh-temperature gas-cooled reactor. ‘Boiling water reactor. gGreater than 3 million curies hGreater than 9 million curies. The utilities that own Fort St. Vrain and Ranch0 Seco expect to decom- mission or convert the plants to use other fuel; the utility that owns Pathfinder partially dismantled the plant in 1968 after shipping the fuel off-site. In 1990 the utility expects to start dismantling other parts of Pathfinder and complete these activities by the end of 1991. In addition, utilities owning seven small nuclear plants, including Peach Bottom, 4Generdly, reactor vessels are large, steel cylindrical vessels that can weigh almost 1,000 tons and vary from about 46 to 70 feet in height. The walls of the vessels range from about 7- to 1l-inches thick. Shippingport’s vessel weighed about 163 tons and was about 26 feet high. “A curie is a measure of the rate of radioactive decay. P8ge 4 GAO/RcEDscrzoe Shippingport Decommisdoning 5239632.2 have started to decontaminate them and put them into “safe storage” until a site is available to dispose of the high-level waste.” Pressure Vessel DOE removed the pressure vessel from Shippingport in one piece. The pressure vessel is the most highly contaminated part of a nuclear power Decommissioning plant. Although removing the vessel in one piece minimized worker exposure to radiation and reduced costs by about $7 million, DOE’S approach did not provide the nuclear industry with information on the problems that may be encountered if utilities must cut-up this compo- nent. According to an official, other DOE decommissioning projects will cut-up reactor pressure vessels. Three of the four utilities that we visited could not dispose of the pres- sure vessel in the same manner that DOE used at Shippingport. According to utility executives from Fort St. Vrain, Peach Bottom, and Ranch0 Seco, the size of the pressure vessels and the radioactive contamination they contain will preclude their shipping and disposing of this compo- nent in one piece. Only the utility that owns Pathfinder, whose reactor pressure vessel (32 feet by 12 feet) was about the same size as Ship- pingport (25 feet by 10 feet), said it expected to dispose of the vessel in one piece at a commercial site operated by US. Ecology in Richland, Washington. Waste Management and DOE had predetermined sites to dispose of the spent (used) fuel from Shippingport as well as the low-level and mixed waste generated from Disposal decommissioning activities. DOE sent the spent fuel to its Idaho National Engineering Laboratory. Currently, no disposal site exists for the spent fuel from commercial plants; DOE expects that the earliest a permanent site would be available is 2010. Also, according to DOE officials, Shippingport did not generate any of the most highly radioactive low-level waste that can remain hazardous for a few hundred to tens of thousands of years (greater-than-Class C).i According to utility officials, Fort St. Vrain is expected to generate about 142 cubic feet of greater-than-Class C waste, and Ranch0 Seco will “Humboldt Bay 3, California; Fermi 1, Michigan; Indian Point 1, New York; Vallecitos Boiling Water Reactor, California; Dresden 1, Illinois; Lacrosse, Wisconsin, and Peach Bottom 1, Pennsylvania. ‘About 3 percent of low-level waste-greater-than-Class C-ii contaminated with long-lived radioac- tive elements having concentrations greater than those specified in 10 C.F.R. Part 61 of NRC’s regulations. Page 5 GAO/RCED-9MOE Shippingport Decommissioning E239832.2 also generate such waste although officials could not estimate the volume. Currently, no disposal site exists for such waste generated from commercial operations. Further, DOE disposed of other low-level waste at Hanford at signifi- cantly lower costs than utilities will experience. DOE disposed of about 214,000 cubic feet of waste for about $2.4 million (including the reactor pressure vessel). In 1986 low-level waste disposal costs at Hanford were $3.96 per cubic foot; by 1989 the cost had increased to about $27.60 per cubic foot, excluding packaging, transportation, labor, materials, taxes, or surcharges allowed by the Low-Level Radioactive Waste Policy Act, as amended.* After January 1993 low-level waste disposal costs could range from $50 to $590 or more per cubic foot as a result of the new facilities-possibly as many as 16-that will be built by states or inter- state compacts to comply with the act and dispose of low-level waste. For example, utility officials estimate that decommissioning Ranch0 Seco will generate about 775,000 cubic feet of low-level waste and dis- posal costs could total about $24 million. Management Structure Because of the research and demonstration nature of Shippingport, DOE used an elaborate management structure to decommission the plant. DOE used over eight contractors to conduct the physical activities and three other contractors to oversee the activities conducted. According to DOE’S Program Manager, DOE recognizes that utilities may not be able to insti- tute the same type of management structure to decommission commer- cial plants. Further, at least 30 percent of DOE’S costs related to physical decommissioning activities; the remaining 70 percent included engi- neering, oversight, management, and other activities, such as waste dis- posal. Utilities, faced with setting aside funds to decommission their plants and subject to scrutiny by public service commissions when doing so, most likely will not incur as high a level of oversight and manage- ment costs relative to physical decommissioning costs that occurred with Shippingport. “To encoursge the development of new low-level wsste disposal sites, the act established surcharges ranging from S10 to $40 per cubic foot of waste disposed of between July lQEl6and December 1992 and penalties of up to $120 per cubic foot of waste during calendar year 1992. In 1989 the surcharge was $20 per cubic foot. Page 6 GAO/ItCED~2OS Shippingport Demmnisaioaing B-229622.2 The applicability of the lessons learned from Shippingport will vary Application of Lessons depending upon when utilities start to decommission their plants and Learned Is the similarity of their efforts to the way that Shippingport was decom- Questionable missioned. Thus, the lessons learned from Shippingport may diminish by the time a large number of utilities decommission their plants, The Nuclear Regulatory Commission (NRC), which issues licenses to, and oversees the safe operation of, commercial plants, estimates that by the year 2015 about one-half of the existing 113 operating licenses will ter- minate, and most of the remaining licenses will expire by about 2030. However, utilities can apply to NRC to extend the plants’ operating licenses, and because of the high cost of building new plants, a strong likelihood exists that utilities will do so. NRC currently expects that the license extension will be for 20 additional years. Also, under NRC'S regu- lations, utilities can take as long as 60 years to complete decommis- sioning activities. Therefore, utilities may not decommission a large number of plants until well into the 21st century and new technology, such as remotely operated equipment and robotics, may lessen the use- fulness of the technology used at Shippingport. Utility officials that we contacted expressed various opinions about the usefulness of Shippingport to their circumstances. An official from Peach Bottom said that Shippingport provided useful information on constructing platforms to dismantle the plant and using a plasma arc torch to cut-up some components, but most other information would not be useful to decommissioning Peach Bottom. Ranch0 Seco officials said that DOE briefed them on dismantling Shippingport, and the report on asbestos removal will be applicable to their activities.” In addition, Fort St. Vrain officials said that they will use Shippingport’s information to develop a decommissioning plan for the plant. Further, according to Humboldt Bay officials, they will not need specifics on decommissioning for at least 20 years, and until then, they could not determine whether Shippingport will be useful. One objective of the Shippingport project was to demonstrate that a Shippingport Did Not nuclear power plant could be safely and economically decommissioned Increase Basic using existing technology, such as manually dismantling radioactive ResearchKnowledge piping systems and components. Thus, DOE did not design the project to increase the basic research and development knowledge on methods or “Shippingport Station Decommissioning Project, Asbestos Removal Topical Report, Mar. 18. 1988. Page7 GAO/RCED9W208 Shippingport Decommissioning equipment needed to decommission a large plant. According to DOE offi- cials, they relied on technology the nuclear industry used for the last 30 years to construct, maintain, or demolish plant systems and components. As a result, DOE did not need, nor was it required, to develop new tech- nology, such as robotics, to decommission Shippingport. Nevertheless, DOE recognizes that the nuclear industry could benefit from information in such areas as (1) decontamination methods, (2) waste reduction and minimization techniques, (3) methods to determine the specific radioactive materials and levels of contamination in waste, and (4) robotics for facility and plant equipment disassembly. In this regard, DOE has embarked on a 5-year program to develop new tech- nology for the clean up of uranium enrichment, fabrication, and reprocessing facilities as well as plutonium production reactors. As part of this effort, DOE recognizes that the development of robotics and/or remote systems may be the only means to safely conduct decommis- sioning activities in highly radioactively contaminated facilities and minimize worker exposures. In addition, DOE is exchanging information with the United Kingdom and Japan. The United Kingdom is decommissioning two gas-cooled reac- tors-Windscale and Berkeley-and is assessing the need to use robotics. Japan is funding a major demonstration project to decommis- sion a nuclear reactor about 100 miles northeast of Tokyo. As part of the project, Japan has stimulated private sector research and develop- ment by inviting most of the country’s largest industrial firms to develop advanced techniques for future decommissioning projects. In particular, the Japan Atomic Energy Research Institute in conjunction with Mitsubishi Heavy Industries developed a robotic arm to cut-up the reactor’s internal parts, which were then put into a pool of water where they were further cut-up by a robotic saw developed by Hitachi. In addition to international efforts, General Public Utilities has been using robotics to remove fuel and conduct other activities at the dam- aged Three Mile Island, Pennsylvania, plant. Although the accident cre- ated cleanup problems that should not be experienced at other plants, utilities may derive more useful information from the activities con- ducted at Three Mile Island than from Shippingport, particularly efforts to minimize worker exposures that were minimal at Shippingport. Conclusions pingport by completing these activities 4 months ahead of schedule and Page 8 GAO/ItCEB2tS208 Shippingpor& Decommissioning B-239632.2 $7 million under the estimated cost for the project. It also utilized con- tractors to conduct various decommissioning activities to extend the experience and knowledge to the private sector. However, because Ship- pingport was a smaller reactor and less radioactive than other reactors, lessons learned from its decommissioning are limited. For example, because Shippingport had a small pressure vessel, DOE was able to dis- pose of it whole whereas future pressure vessels will probably have to be cut up, thus increasing worker exposure to hazardous nuclear materials. Further, many years may elapse before utilities dismantle a large number of plants. In the interim, decommissioning activities being con- ducted on higher radioactivity contaminated pressure vessels here and abroad will advance the state-of-the-art beyond the lessons learned at Shippingport, and information that will be developed under DOE'S recently announced initiatives and by the United Kingdom, Japan, and Three Mile Island may be more useful than Shippingport. These efforts are directed at identifying new technology to reduce worker exposures, To obtain this information, we contacted NRC, DOE, utility, and industry officials and reviewed numerous reports that DOE had prepared on the project. Our objectives, scope, and methodology are discussed in detail in appendix I. Appendix II contains responses to the questions raised in your request. Appendix III summarizes the decommissioning efforts of the four utilities that we visited. We discussed the facts in the report with NRC, DOE, and utility officials and incorporated their views where appropriate. As requested, we did not obtain official agency comments on a draft of this report. Our work was conducted between August 1989 and June 1990 in accordance with generally accepted government auditing standards. As arranged with your office, unless you publicly announce its contents earlier, we plan no further distribution of this report until 30 days from the date of this letter. At that time we will provide copies to Representa- tive Fazio; the Secretary of Energy; and the Chairman, NRC. We will also make copies available to others upon request. P8ge 9 GAO/WED-90-208 Shippingport Decommissioning B-229622.2 Please call me at (202) 275-1441 if you have any questions. Other major contributors to this report are listed in appendix IV. Victor S. Rezendes Director, Energy Issues Page 10 Page 11 Contents Letter 1 Appendix I 14 Objectives, Scope,and Methodology Appendix II 16 Responsesto Questions Asked Concerning the Decommissioning of Shippingport Appendix III 22 CaseStudies on Four Pathfinder 22 23 Nuclear Power Plants ~~~~~~ Unit1 23 Ranch0 Seco 24 Appendix IV 26 Major Contributors to This Report Tables Table 1: Comparison of Shippingport to Four Other Plants 4 Table II. 1: Breakdown of Project Costs and Forecasted 17 Labor Hours Used to Decommission Shippingport Abbreviations DOE Department of Energy EPA Environmental Protection Agency GAO General Accounting Office NRC Nuclear Regulatory Commission ORAU Oak Ridge Associated Universities SMUD Sacramento Municipal Utility District Pa@ 12 GAO/lUXD8@208 Shippingport Decommissioning P8ge 13 GAO/RCED&bS2Xb Shippingport Deeommiasioning Appendix I Objectives, Scope,and Methodology On July 21, 1989, the Chairman and the Ranking Minority Member, House Committee on Science, Space, and Technology, asked us for infor- mation about the Department of Energy’s (DOE) decommissioning of the Shippingport, Pennsylvania, nuclear power plant. Specifically, we were asked to (1) determine whether DOE had met the goals described at July 1986 hearings before the House Committee on Science, Space, and Tech- nology and (2) answer nine questions, some with several parts to them. To obtain the information needed, we reviewed the Atomic Energy Act, the Energy Reorganization Act, and the Nuclear Regulatory Commission (NRC) regulations and guidelines. We also reviewed numerous DOE reports related to decommissioning Shippingport, such as the final envi- ronmental impact statement, annual progress reports, and topical reports relating to asbestos, contaminated concrete, and pressure vessel removal. We also reviewed DOE’S final report on the project and the results of a survey conducted by Oak Ridge Associated Universities to confirm DOE’S assessment of the radiological condition of the site. We also used information from five of our reports1 In addition, we met or spoke with NRC staff in the Office of Nuclear Reg- ulatory Research, DOE officials from the Office of Remedial Action and Waste Technology, and officials from the National Academy of Sciences, the Electric Power Research Institute, Oak Ridge Associated Universi- ties, General Electric, French and British embassies, TLG Engineering, Inc., Nuclear Management Resources Council, American Nuclear Society, Pacific Gas and Electric Company, Worldwatch Institute, and Arkansas Power and Light. We also contacted utilities that own the Peach Bottom, Pathfinder, Fort St. Vrain, Ranch0 Seco, and Humboldt Bay plants. We selected these plants because they have been shut down awaiting decommissioning or have been partially decommiss’loned. For example, in a June 1989 refer- endum, Californians voted to shut down Ranch0 Seco. Also in 1989, Fort St. Vrain officials decided to shut down the plant after years of oper- ating problems. The utility that owns one Peach Bottom unit has started to decommission the plant. In addition, Northern States Power, which Page 14 GAO/lKXD-W206 Shippingport Decommissioning Appendix I Objectives, Scope, and Methodohgy owns Pathfinder, had partially decommissioned the plant in 1968. Fur- ther, we spoke to a Humboldt Bay official because we were told that the plant was similar in design to Shippingport and has been partially decommissioned. Page 16 GAO/RCED9MOS ShippIngport Decommissioning Appendix II Responsesto QuestionsAsked concerning the Decommissioningof Shippingport 1. For how long and at what power levels did Shippingport operate over its lifetime? What were the total operating hours? How much time elapsed between the shutdown and start of decommissioning activities? What was the radiation level of the reactor at the outset of dismantlement? Over its 25-year life, Shippingport operated for about 80,324 hours, pro- duced about 7.4-billion kilowatt-hours of electricity, and operated at power levels of 60,150, and 72 megawatts. The plant was shut down in October 1982 and physical decommissioning activities began in Sep- tember 1985, almost 3 years later. At the time of shutdown, the radioac- tivity in the pressure vessel was about 30,000 curies; at the outset of decommissioning, about 16,000 curies. 2. Was the cost of the project consistent with the $98.3 million estimate that DOE presented at the July 1986 hearings? What were the actual costs and hours for labor, materials, equipment, and waste disposal? DOE completed the Shippingport project for $91.3 million, or $7 million less than estimated. Because DOE kept costs and labor hours for the pro- ject by activity, such as engineering, decommissioning, and site manage- ment and support, we could not determine the costs or labor hours in the format requested. However, table II. 1 shows the project’s costs and labor hours based on DOE’s work breakdown structure. Page 16 GAO/IKXD-W2@3 Shippingport Decommissioning &qonses to Questions Asked Concerning the Decomndseioning of Shippingport Table 11.1:Breakdown of Project Costs and Forecasted Labor Hours Used to Total costs Labor (in Decommission Shippingport Description (Millions) months) Engineenng (Phase I) $6.1 NAa Project management 10.6 719 Site management and services 6.8 584 Support and services 23.5 3,589 Engineering 1.1 19 Procurement 0.7 0 Solid waste management 2.1 185 System operations support 1.7 116 Utilities 1.6 0 Liquid waste management 1.3 87 Subtotal: Site management and support 38.8 4,580 Site modifications and services 5.2 482 Reactor pressure vessel preparation, removal, and transport 6.5 222 Remove piping and equipment 6.6 601 Remove primary components 1.3 105 Remove power and control systems 0.5 111 Remove structures 5.9 285 Remove containment chambers 0.4 35 Decontamination 2.3 457 Subtotal: DecommissioninQ aCtiVitie8 28.7 2,298 Home off ice support 1.6 40 Decommissioning operations fee 5.4 . Other 0.2 . Total 91.3b 7,63Sb aDOE did not have information showrng the labor hours used during the early engrneering phase bTotals do not add due to rounding. At least 30 percent of the costs shown in table II.1 directly relate to the physical decommissioning of the plant; the remaining 70 percent includes engineering, oversight, management, and other activities. 3. Did DOE and General Electric maintain the project schedule’? Can any lessons learned lead to shorter schedules in future decommissioning projects? DOE completed Shippingport in December 1989,4 months earlier than the expected April 1990 date. The only significant delay occurred early in the project when the contractor decided to remove asbestos all at once rather than throughout the project. Of the 36 control milestones for the Page 17 GAO/RCED-9@208 Shippingport Decommissioning Responees to Questions Asked Concerning the Decommisaiining of Shippingport project, General Electric completed 16 earlier and 20 later than planned. According to DOE and utility officials that we contacted, Shippingport did not provide any specific lessons learned that could reduce schedules for future decommissioning projects. However, some utility officials rec- ognize the need for effective up-front planning as occurred with Shippingport. 4. How has the project benefited the nuclear power industry? Has Ship- pingport identified areas for cost reductions in future decommissioning projects? The transferability of the lessons learned from Shippingport to the com- mercial nuclear power industry varies depending on the needs of the individual contacted. The most significant benefit, according to some utility executives, is DOE demonstrated that technology exists to decom- mission a plant within the costs and time frame established. Others believe that only minimal benefits have been derived because DOE removed the reactor pressure vessel in one piece and did not use any new technology, such as robotics. Still others indicated that the value of Shippingport will decrease over time. Currently, 11 commercial plants have been shut down, and the licenses for about one-half of the 113 operating plants will most likely not expire until the year 2015. Most of the remaining licenses will expire by about 2030. Prior to that, utilities will decide whether to shut the plants down or seek a license extension from NRC. NRC currently expects that the license extension period will be 20 years. Also, under NRC'S regulations, utilities can take as long as 60 years to complete decommissioning activi- ties. Therefore, many years may elapse before utilities begin to decom- mission a large number of plants. Utility officials identified a few areas in which Shippingport may allow them to reduce future decommissioning costs. The areas cited include planning and scheduling, removing asbestos, and removing hazardous waste. The officials could not, however, estimate the savings that could be realized. 5. As a result of the Shippingport project, can the site be released for unrestricted use? In October 1989 DOE certified that the site met the release criteria that had been established for Shippingport. DOE had required that public Page18 GAO/lKEDW208 Shippingport Decommissioning --- Reqtonsem to Questions Aaked Concerning the DeeolNniMi~ of Shippingport exposures from the remaining contamination should not exceed 100 mil- lirem’ a year, and the level should be reduced if reasonably achievable to do so. DOE documentation indicates that public exposures will be less than 2 millirem annually. DOE contracted with the Oak Ridge Associated Universities (ORAU)to confirm its analyses. ORAUfound some contami- nated areas, and General Electric conducted additional cleanup activities to ORAU’Ssatisfaction. In its November 1989 report, ORAUindicated that DOE had effectively decontaminated and decommissioned the site. 6. Has Shippingport validated NRC’Sdecommissioning regulations? Little relationship exists between NRC’Sdecommissioning regulations and the Shippingport project. NRC’Sregulations primarily address decommis- sioning planning needs, timing, funding methods, and environmental review. The intent of NRC’Sregulations is to ensure that utilities decom- mission nuclear power plants in a safe and timely manner and that ade- quate funds will be available to conduct the needed activities. Shippingport’s costs cannot be used to validate NRC’Sgeneric estimate- $105 million for a pressurized water reactor-because utilities will not be able to use the same methods and management structure that DOE Used. For example, most utilities will not be able to dispose of their reactor pressure vessels in one piece and will experience significantly higher low-level waste disposal costs than the $2.4 million (including the pres- sure vessel) that DOEincurred for Shippingport. Further, utilities will probably have to dispose of some greater-than-Class C waste. Ship- pingport, according to DOEofficials, did not generate such waste. Because of these and other differences, Shippingport cannot be used to validate some of the costs that utilities will experience when decommis- sioning large plants. In addition, Shippingport was not licensed by NRC;therefore, DOE did not have to obtain NRC’Sapproval for the decommissioning activities con- ducted at the plant. Further, no federal agency or utility official that we contacted could identify any changes that should be made to NRC’Sregu- lations as a result of Shippingport. We noted, however, that DOE set a 100 millirem per person per year residual contamination limit for Shippingport. NRChas been suggesting lAmilliremisa thousandth of a rem (Roentgen Equivalent Man), which is the measurement used to quantify the effects of radiation on man. GAO/RCED-9lKUX3 Shippingport Decommissioning Appendix JJ Responsea to Questions Asked Concerning the Decommissioning of Shippingport that utilities decontaminate to a level that would limit public exposures to 10 millirem a year-10 times less than DOE required. Ultimately, the Environmental Protection Agency (EPA) is responsible for setting the limits of residual contamination that can remain on-site. EPA has been developing such standards for several years but does not expect to make them final until 1993 at the earliest. 7. What changes should be incorporated in planning future decommis- sioning projects as a result of Shippingport? How should decommis- sioning plans incorporate improved methods to reduce worker exposures? DOE spent over $6 million developing a decommissioning plan for the project. According to a DOE official, the planning conducted helped the project to be completed on time and under cost and allowed them to keep worker exposures within established limits. In the decommissioning plan, DOE'S contractor proposed a worker exposure limit of about 1,010 person-rem for the project; the actual exposure was 155 person-rem. In addition, DOE completed the project without any serious radiological incidents, according to officials. Utility executives that we contacted said the lessons learned from DOE'S planning efforts could facilitate their planning for future decommissioning projects. However, Shippingport provided only limited information to reduce worker exposures on future projects where the pressure vessel would be cut-up. Shippingport was more radiologically clean at the start of decommissioning than could be expected for a much larger commercial plant (1,000 megawatts or greater). Also, DOE removed the most highly radioactive component-the reactor pressure vessel-in one piece. 8. What specific examples exist showing that Shippingport’s technology transfer activities influenced other decommissioning projects, such as Three Mile Island? What special equipment did DOE use at Shippingport? With the exception of Northern States Power that plans to remove the pressure vessel from Pathfinder in one piece, specific examples showing that Shippingport influenced other decommissioning projects do not exist. DOE developed extensive information on Shippingport, but the use- fulness of the data will diminish the longer utilities wait to decommis- sion their plants. In addition, DOE did not develop any new technology, such as remotely operated equipment or robotics, to decommission Shippingport because Page 20 GAO/RCJ3DCllX2O8 Shippingport Decommissioning Appe*n, itespoMestoQuationsAsked~ncerning the Deconuniaaio~ of Skippingport one of the project’s objectives was to demonstrate that a nuclear plant could be safely and economically decommissioned using existing tech- nology. Further, some of the lessons learned from the cleanup of the damaged Three Mile Island plant may be more useful to utilities than Shippingport. 9. Did DOE use information developed overseas to plan for, and decom- mission, Shippingport? According to DOE officials, foreign countries did not provide any infor- mation that was used to help with decommissioning Shippingport. Although Japanese officials discussed their research and development projects with DOE, the agency determined that these activities were not cost-effective because of the exotic robotic techniques that were involved. A DOE official further said that he was aware of other interna- tional decommissioning projects, but the information was not applicable because the purpose of Shlppingport was to demonstrate decommis- sioning techniques using present technology. However, foreign nationals from Japan, United Kingdom, and other countries participated in some of Shippingport’s decommissioning activities. In addition, through its technology transfer program, DOE has provided both domestic and inter- national utilities a great deal of information about the project. Page 21 GAO/RCESWX208 Shippinoport Decommidoning Appendix III CaseStudies on Four Nuclear Power Plants Pathfinder, a 62-megawatt boiling water reactor’ owned by Northern Pathfinder States Power Company, operated from 1964 to 1967. The plant, located about 6 miles from Sioux Falls, South Dakota, was shut down after investigations disclosed serious flaws with some components within the reactor pressure vessel. During its limited life, Pathfinder operated for about 12,000 hours, generating about 0. l-billion kilowatt-hours of elec- tricity. At the time the plant was shut down, the radioactivity in the pressure vessel was about 30,000 curies. Northern States started to decontaminate the plant in 1968 after removing the fuel and shipping it off-site. The utility also removed almost all contaminated pipe outside the reactor and fuel handling buildings and drained and filled the reactor pressure vessel with gravel. The utility did not decontaminate the piping system inside the reactor building or remove any of the pipe. After partially decontaminating the reactor and fuel handling buildings, Northern States sealed the areas to prevent unauthorized access. In 1990 Northern States expects to begin decontaminating the previ- ously sealed areas. The utility plans to dispose of most low-level radio- active waste, including the reactor pressure vessel and the shipping package, at a commercial site operated by US. Ecology in Richland, Washington. Because of the weight (78 tons) and size (12 feet x 32 feet) of the pressure vessel and the shipping package, the utility plans to rent a special rail car and train to transport it. Company officials said that the decontamination activities completed in the 1960s represented about 25 to 35 percent of the plant’s decommis- sioning and estimate that the total decommissioning costs will be about $20 million. According to these officials, Shippingport is more relevant to Pathfinder than other plants because the size and radioactivity levels are very comparable. They said that they will use the lessons learned from Shippingport to decontaminate and remove Pathfinder’s pressure vessel. Northern States officials said that very little new knowledge was gained from Shippingport, but the knowledge gained through topical reports, seminars, feedback from contractors that participated in the project, and other information they requested confirmed that the method they selected to decommission Pathfinder is valid. They also stated that the nuclear industry, in general, could not use this same method to decommission other plants. ‘Boiling water reactors are cooled by water that is allowed to boil as it passes through the nuclear fuel. The water is used directly to produce the steam that generates electricity. Page 22 GAO/lKXD-00-208 Shippingport Decommissioning Cue !%ndlm on Four Ndear Power Planta Fort St. Vrain, a 330-megawatt high-temperature gas-cooled reactor, is Fort St. Vrain owned and operated by the Public Service Company of Colorado. The plant, located about 36 miles north of Denver, began commercial opera- tions in 1979. In August 1989 the utility shut the plant down after years of operating problems. During its life, Fort St. Vrain operated for about 21,360 hours, generating about 4.3-billion kilowatt-hours of electricity. At the time the plant was shut down, company officials estimate that the reactor contained about 900,000 curies of radioactive contamination. Fort St. Vrain is different from Shippingport and the other 112 domestic nuclear power plants. For example, the plant used graphite to control the rate of fission inside the reactor pressure vessel whereas Ship- pingport and the other plants generally use water. Also, the fuel used in Fort St. Vrain differed from that used in Shippingport and other plants. In November 1989 the utility began removing the spent fuel and had planned to send it to DOE’s Idaho National Engineering Laboratory. How- ever, the governor of Idaho ordered a halt to the shipments, and the company is now storing the fuel pending consideration of other such options as building a spent fuel facility. Public Service has not selected its final decommissioning option, but the company has requested pro- posals to determine the methods to be used and costs to dismantle the plant. According to the company’s 1989 preliminary decommissioning plan, the costs for safe&ore would be around $81 million because the plant is relatively radiologically clean. Currently, the utility expects to convert Fort St. Vrain to a gas-fired plant. These officials also stated that Shippingport provided useful informa. tion to plan, manage, and dismantle Fort St. Vrain as well as methods to control the spread of contamination during the physical decommis- sioning of the plant. Nevertheless, they also noted several significant differences between Shippingport and their plant. Of foremost impor- tance was the small size of Shippingport and the removal of the reactor pressure vessel in one piece-an option that cannot be used for Fort St. Vrain. Peach Bottom Unit 1, a 40-megawatt prototype high-temperature gas- Peach Bottom Unit 1 cooled reactor, is located about 80 miles southwest of Philadelphia. The plant, owned by the Philadelphia Electric Company, operated from June 1967 until October 1974. During the 7-year period, the plant operated for about 32,375 hours, generating about 1.4billion kilowatt-hours of Page 28 GAO/llCED~208 Shippingport Decommimioning Case Studies on Four Nuclear Power Plants electricity. At the time the plant was shut down, the radioactivity in the pressure vessel was more than 3 million curies. Philadelphia Electric decided to safestore the facility and started to decontaminate the site in January 1976. The company completed these activities in February 1978, using about 179 person-months of labor, at a cost of about $3.5 million. The utility removed all radioactive liquids, drained refrigerants and cooling water,’ and sent the spent fuel to DOE’S Idaho National Engineering Laboratory. The company left the reactor vessel, piping systems, and steam generators in the plant, and officials estimate that they will not start to remove these components or other- wise decommission the plant for about 20 more years. At that time, company officials said they would review the Shippingport information to determine whether they could apply it to Peach Bottom. These officials also noted that the usefulness of the Shippingport infor- mation will most likely decrease over time as new technology is developed. Ranch0 Seco, a 913-megawatt pressurized water reactor, located about Ranch0 Seco 25 miles southeast of Sacramento, California, is owned and operated by the Sacramento Municipal Utility District (SMUD). On June 7, 1989, SMUD shut down the plant in response to a voter referendum to close the plant. During its lifetime, Ranch0 Seco operated for about 51,595 hours and generated about 44-billion kilowatt-hours of electricity. Company officials estimate that the amount of radioactivity in the plant at shut down exceeded 9 million curies. In 1987 SMUD completed a generic cost study for the plant and expects to submit a revised decommissioning plan to NRC by July 1991. The 1987 study showed that immediate dismantlement of both radioactive and nonradioactive structures ($210 million) would be less costly than mothballing the plant ($265 million). SMUD began to remove the fuel from the plant on November 28, 1989, and place it in an on-site storage pool. In addition, SMUD tried to sell the plant but was not successful in finding a buyer. Company officials told us that they may wait up to 50 years before dismantling the plant. “Liquid or gas circulated through a nuclear reactor to remove or transfer heat. Some coolants are water, heavy water, carbon dioxide, liquid sodium, sodium-potassium alloy, and helium. Page 24 GAO/RCEDO@200 Shippingport Decommissioning Case Studies on Four Nuclear Power Phtr Nevertheless, they believe that the Shippingport experience will be helpful to them, particularly the asbestos removal report. According to SMUD officials, although about 60 percent of the information learned from Shippingport was either new or useful to them, they also cited numerous differences between Shippingport and Ranch0 Seco. For example, decommissioning Ranch0 Seco will generate greater-than-Class C waste-even if the plant is mothballed for 60 years; Shippingport gen- erated no such waste, according to DOE officials. Also, low-level waste disposal costs for Shippingport were about $2.4 million (1986 dollars), whereas SMUD estimates that such costs for Ranch0 Seco will be about $24 million (1986 dollars). Furthermore, SMUD cannot remove, ship, and dispose of the Ranch0 Seco pressure vessel in one piece as DOE did at Shippingport. SMUD officials did not believe that Shippingport demonstrated the cost- effective decommissioning of a large commercial nuclear plant because Shippingport was smaller, was relatively radiologically clean, and did not have to comply with NRC’Srequirements. Also, all waste from Ship- pingport was sent to a DOE facility, and the disposal costs were so low that the situation is not comparable to the nuclear industry. SMUD offi- cials said that today low-level radioactive waste disposal costs for com- mercial nuclear plants are at least 10 times higher than those incurred by Shippingport. Page 26 GAO/RCED-W2OgSMppingportDecommisaionlng Appendix IV Major Contributors to This Report Judy England-Joseph, Associate Director, Energy Issues Resources, Mary Ann Kruslicky, Assistant Director Community, and Philip A. Olson, Evaluator-In-Charge Economic Development Division, Washington, D.C. (301884) Page 28 GAO/RCEDOO-208 Shippineport Decommhioning *e.. --.- ~- __--__-__________-_.--__ _-.-. _..----___----.--_._I_.- ___ Ordering Information The first five cwpies of t*ach (;A() report iire free. Additional copitbs are $2 each. Orders should he wnt to the following address, accorn- panied by a check or mowy order made out to the Superintendent of Documtwts. when newssaq. Orders for 100 or morv copies to be mailed to a single address ilrt’ diwounttvi 2.i percent.
Nuclear Research and Development: Shippingport Decommissioning--How Applicable Are the Lessons Learned?
Published by the Government Accountability Office on 1990-09-04.
Below is a raw (and likely hideous) rendition of the original report. (PDF)