Abstract
Since the dawn of the nuclear era, more than two dozen nuclear power reactors have been permanently shut down in the United States. At some point, the remaining 100 nuclear power reactors currently operating in the United States also must be permanently shut down. But after a reactor is no longer generating power, how will the next step, known as decommissioning, be accomplished? In theory, the process is straightforward. The Nuclear Regulatory Commission requires that after a nuclear facility is retired, it be decontaminated, with all its nuclear fuel, coolants, and radioactive wastes removed. Once this is completed, the owner’s license is terminated, and the site is considered to be freely available for other uses—with some restrictions, as needed, on a case-by-case basis. But if recent history is any guide, owners of a nuclear power plant that is no longer operating can take a range of approaches to accomplishing the end goal of “greenfield” status, in which the site is in the same pristine condition it was before the plant was built. There are three main approaches in real life: “do it yourself,” “wait and see,” and “calling in the decommissioning cavalry.” A few case studies give an idea of what to expect based on the experiences so far. They do not cover every factor in decommissioning decision making but reveal some of the potential pitfalls, problems, and dangers that could accompany the upcoming wave of decommissionings.
Keywords
US nuclear power reactors were originally licensed to operate for 40 years. Like a car, a nuclear power reactor has parts that wear out and must be either replaced or refurbished periodically. For example, workers monitor the erosion and corrosion of pipes in order to replace the pipes before wall thinning compromises safety. Nuclear power reactors typically shut down every 18 to 24 months for refueling, where one-quarter to one-third of the nuclear core is replaced with new fuel. During these shutdowns, workers also test, inspect, repair, and replace components and structures.
Also like a car, a nuclear power reactor has other parts that might be too costly to replace. For example, the metal reactor vessel that houses the nuclear core could someday wear out and require replacement if operation is to continue. But just as car owners face decisions about whether to pay for a costly transmission repair on an older vehicle or put that money toward the purchase of a new car, nuclear plant owners must decide whether the cost of maintaining aging nuclear power reactors makes good business sense.
More than three-quarters of the United States’ nuclear power reactors have been relicensed by the Nuclear Regulatory Commission (NRC) for 20 years of operation beyond their original estimated lifetimes. The relicensed reactors will operate as long as they can do so economically. But it is inevitable that all nuclear power reactors will someday shut down permanently.
In 2013, four US nuclear power reactors were permanently shut down: Crystal River 3 in Florida, Kewaunee in Wisconsin, and San Onofre Units 2 and 3 in California. While they and the other 22 US nuclear power reactors that have been permanently shut down each operated for an average of 20.5 years, the cleanup after them, known as decommissioning, could take nearly three times as long, for an average of 60 years apiece—if they meet the target goals of the NRC. In addition, the commission requires that the cleanup work for each plant be finished before the dedicated funds set aside for that purpose by the plant’s owners are depleted. As of this writing, each facility is at a different stage in the planned decommissioning process.
Decommissioning: A timeline
Within 30 days after a reactor is permanently shut down—meaning that a nuclear reaction is no longer occurring and electricity is no longer being generated—or the decision is made to shut it down, the plant’s owner must certify to the NRC that operation has ceased. Before major decommissioning activities begin, the owner must also certify to the commission that all fuel has been permanently removed from the reactor vessel (Nuclear Regulatory Commission, 2013d). The owners have plenty of incentive to remove fuel sooner rather than later—as a fee-based agency, the NRC charges a nuclear power plant owner $4,390,000 each year for an operating reactor, while charging only $231,000 for one that is formally in the process of being decommissioned (Nuclear Regulatory Commission, 2013a). So, getting that last fuel assembly out of the reactor vessel is worth more than $4 million to the plant’s owners.
Paying for the process of decommissioning a nuclear power plant can be costly, but it is supposed to come out of the pockets of the plant owners, not from the general public. To make sure that the money is indeed where it is supposed to be, the NRC’s regulations require that the owners of a nuclear plant periodically set aside funds for decommissioning, and reassess their funding plans every two years (Nuclear Regulatory Commission, 2013c).
These reassessments consider financial-market conditions that may have reduced the value of past investments, as well as site-specific conditions that may have increased the cost of a decommissioning—such as an on-site leak, or the spill of radioactively contaminated water that must be remediated.
The commission independently reassesses the decommissioning funding levels. In the most recent NRC reassessment report (Leeds, 2013), the commission determined that the accrued funding for Pennsylvania’s Beaver Valley Unit 1, Michigan’s Palisades, Ohio’s Perry, and Wisconsin’s Point Beach Unit 2 reactors each fell below their required minimum. In each case, the agency resolved the potential shortfall through financial mechanisms, such as a financial guarantee from the parent company of the owners of the individual power plant. The NRC’s minimum decommissioning cost estimates range from $438,293,852 for Fort Calhoun in Nebraska to $1,070,497,506 for Fermi Unit 2 in Michigan. A half-billion to a billion dollars seems to cover it.
Permanently shut down nuclear power reactors in the United States
In practice, however, keeping radioactive materials on-site and encasing them for several years is really not available to power reactors, due to the large amounts involved and the lengthy half-lives of the radioactive materials present. Consequently, ENTOMB is an option for smaller research reactors that did not operate long enough to build up significant inventories of radioactive materials.
Of the two remaining decommissioning options, the owners of six of the 14 reactors scheduled to be shut down in the near future elected to go the DECON route and start decommissioning immediately, while the owners of eight other reactors scheduled for shutdown decided to defer as much as possible via the SAFSTOR option (see Table 1).
Who is right? According to the way the rules are written, no one can be considered wrong at this point, because the commission defines “right” as terminating the license within 60 years and before the decommissioning fund empties. Three reactors, Pathfinder in South Dakota, Saxton in Pennsylvania, and Shoreham in New York, have successfully reached that goal and can say that they are officially “license terminated.” Technically speaking, the other reactors still have the time and the remaining funds to be considered successful, too. Of course, there’s still time to mess it up as well.
The theories behind the different approaches to decommissioning
The advantage of a SAFSTOR-style deferral of as much decommissioning work as possible is that it allows for years or even decades of radioactive decay after the plant has stopped operating, reducing both the potential radiation-exposure risk to workers and the cost to plant owners. A disadvantage is that things can change during that time—the NRC can change its decommissioning regulations, while storage sites that are presently licensed to dispose of low-level radioactive waste can close or increase their prices.
In contrast, the advantage of DECON’s immediate dismantling approach is certainty. This is even truer now that a number of plants have been successfully decommissioned in this manner, because owners now have rough examples to go by, with well-defined tasks. In addition, with DECON, the amount and the level of radioactive debris can be inventoried, and the plant owners do not have to contend with any long-term doubts about the future availability of disposal sites.
A disadvantage to immediate decommissioning is that there is no place for the highly radioactive spent fuel to go as of this writing, because the federal government did not construct a long-term geologic repository for the waste as planned. Therefore, it is impossible to achieve the goal of a truly 100 percent “greenfield” status in which the slate is wiped completely clean and the reactor site has been restored to the conditions that existed before the nuclear power plant was built.
Regardless of which decommissioning approach the owners choose, they are allowed to spend up to three percent of their decommissioning funds for planning, such as preparing the formal Post-Shutdown Decommissioning Activities Report. They can spend up to another 20 percent on non-major decommissioning activities such as surveying the plant for radiation levels.
Such surveys must be made of the nuclear plant’s grounds, structures, and components to establish the location and the nature of any radioactive contamination. (The surveyors also look for any non-radioactive materials that require special handling, such as asbestos.) By using these detailed surveys, engineers can develop plans as to how to dismantle and dispose of structures and components.
Major decommissioning activities cannot begin until at least 90 days have passed since the decommissioning report was submitted. While the NRC officially does not formally review and approve the report, the agency publishes a notice in the Federal Register about the existence of the plant’s Post-Shutdown Decommissioning Activities Report and makes the document publicly available, and then conducts a public meeting in the plant’s vicinity about the decommissioning plans.
Materials are then separated out and handled according to how radioactive they are. Clean materials can be recycled or sent to municipal landfills. Materials with minor contamination can also be recycled or sent to landfills, after they have been treated to remove the radioactivity. For example, if a concrete wall has surface contamination, it may be pressure-washed or the outer surface may be physically removed, enabling the remainder of the wall to be handled as non-contaminated industrial waste.
Materials that are lightly contaminated with radioactivity are placed in certified containers and shipped to licensed low-level waste sites.
Highly contaminated materials are placed in certified containers and stored with the spent fuel casks on a small parcel of land on the original nuclear plant site, due to the lack of a centralized storage site for such waste at this time. These on-site, stand-alone facilities, called “independent spent fuel storage installations” or ISFSI, remain under license and NRC regulation, and must be patrolled by armed guards to prevent theft or vandalism. Licensees are responsible for their security and for maintaining insurance and funding for their eventual disposal, but other than that, the decommissioning process is considered complete for the entire nuclear reactor complex.
The ultimate goal throughout is to achieve something approaching greenfield status, without first running out of funds. Once this has been achieved, the license for operating the power plant can be considered formally terminated. The owners’ obligations and responsibilities for the site end when the license is terminated, subject to any conditions the NRC may impose. The objective is to have no such conditions, allowing the owners to walk away and the sites to be used without restrictions or limitations.
But before the owner can even apply to the NRC to terminate the reactor’s license and the obligations that come with it, the owner must submit a License Termination Plan (LTP) to the agency at least two years beforehand. The LTP describes the final decommissioning tasks as well as the final surveys to be conducted of the site for radiation levels. While most plans will likely seek release of the site for unrestricted use after license termination, they may ask for limited, restricted use of some or all of the site. If so, the plan must describe the controls to be applied to ensure that the restrictions remain in place. When satisfied that all aspects of the plan have been successfully completed, the commission issues a letter terminating the reactor license.
Case studies
Three case studies illustrate the two decommissioning approaches outlined above. First, there was the “do it yourself” DECON approach taken at Yankee Rowe in Massachusetts. Then, there’s the “wait and see” SAFSTOR approach being taken by the owner of the Millstone Unit 1 reactor in Connecticut. Finally, a new version of DECON has now come into play, in which the “decommissioning cavalry” was called in to handle the DECON of the two reactors at the Zion nuclear plant in Illinois. These case studies do not cover every factor in decommissioning decision making, but they reveal some of the considerations involved.
Yankee Rowe and do-it-yourself DECON
The experiences of the Yankee Nuclear Power Station in Rowe, Massachusetts—commonly known as Yankee Rowe—might hold some lessons for future decommissionings, particularly one scheduled to begin in late 2014 for the Vermont Yankee nuclear power plant in nearby Vernon, Vermont, only 43 miles away. (The Vermont facility is even run by some of the same engineers, technicians, and other staff who once operated Yankee Rowe (Davis, 2014).)
It took longer to decommission Yankee Rowe—15 years—than to build it, which took only three years. And decommissioning the Rowe plant cost $608 million (Broncaccio, 2013), while it cost only $39 million to build—although admittedly the latter figure is in 1960s dollars.
These figures do not include the $8 million per year that will be spent for an unknown number of years to monitor the groundwater, even after the nominal end of the decommissioning process (Yankee Rowe, undated). This monitoring is done in the hope that no radioactive contaminants from the used fuel and the high-level waste kept in the 2-acre independent spent fuel storage installation will spill into the adjacent Deerfield River, the state’s highest-quality trout river—part of a region whose rugged natural beauty wowed Nathaniel Hawthorne during a journey in the early 1800s. “I have never driven through such romantic scenery,” he wrote in his journal, later published in his American Note-Books in 1868.
More-recent authors found the same. Essayist Catherine Reid wrote in 2003: “Where it enters Massachusetts, the Deerfield has carved a valley through some of the most craggy of all the state’s lands, with slopes so steep that old-growth forests can still be found, growing well out of reach of loggers’ axes and saws. Some of the best whitewater in the northeast is found here as well, just below the site of the third nuclear power plant to be built in the U.S., which generated electricity for 31 years.” The river is still home to whitewater canoe and kayak competitions every June.
The Yankee Rowe case also reveals some of the regulatory challenges faced by owners when estimating decommissioning costs and developing associated plans. The NRC revises its decommissioning regulations, policies, and practices about every five years or so. Even the short duration of the immediate dismantling approach of DECON failed to protect Yankee Rowe from a few regulatory iterations en route to semi-greenfield status.
Yankee Rowe was an early pressurized water reactor that began commercial operation in 1961 (Wood, 1998). The reactor shut down in October 1991, amid concerns about embrittlement of its metal reactor vessel. On February 26, 1992, its board of directors decided to permanently close the plant (Yankee Atomic Electric Company, 2013). The decommissioning plan, a precursor to the Post-Shutdown Decommissioning Activities Report required by current NRC regulations, was submitted to the commission on March 29, 1994.
Workers removed the reactor vessel, four steam generators, and the pressurizer from the plant in 1993 and 1994. They placed the reactor vessel inside a steel transport cask; altogether, the reactor vessel package weighed 365 tons. A dedicated heavy-haul transporter carried it about 6.5 miles to the nearest railway on April 27, 1997, at the modest pace of about one mile per hour. A rail car, weighing 1,122,700 pounds loaded, transported the reactor vessel package to Barnwell, South Carolina, home of a private, low-level waste storage facility. Similar methods were applied to the other large radioactive components (Wood, 1998).
After all this was done, the owners submitted their License Termination Plan to the NRC in May 1997, only to withdraw it two years later. Their original plan relied on radiological survey methods outlined in a draft commission report (Berger, 1992). But before the draft report on site surveys was finalized and issued, the commission abandoned that survey request in favor of a new method, known as the Multi-Agency Radiation Survey and Site Investigation Manual (Nuclear Regulatory Commission, 2000). The commission developed this new survey method in conjunction with the Department of Defense, the Energy Department, and the Environmental Protection Agency; Yankee Rowe’s owners redid its termination plan based on the new guidelines. The owners also included a Post-Shutdown Decommissioning Activities Report, something which was not required at the time that they first started dismantling Yankee Rowe.
On November 21, 2005 and on August 10, 2007, the NRC approved the release of the majority of the 1,800-acre site from the reactor license. The only area still under the license is the 2-acre plot used for housing spent fuel.
Unexpectedly, the whole process became something of a model for others to follow for the do-it-yourself method—rightly or wrongly. “Ironically, since Yankee Atomic was the first plant to go through the formal decommissioning process, the NRC has invited scientists and researchers from all over the world to study the aged reactor, and the visitors contribute to the local economy,” wrote the provost of the University of Massachusetts, Amherst, in a study of the closure of the plant (Mullin and Kotval, 1997).
There is still a high level of interest in the former reactor site and its still-existing independent spent fuel storage installation. In 2010, a delegation from Russia—including a nuclear engineer who had inspected Chernobyl a few months after its accident and founded a St. Petersburg-based organization devoted to nuclear safety (Cone, 2010)—had to be turned away by security when the Russians showed up at the gate. Citing NRC security-related regulations regarding protection of the spent fuel and the highly radioactive waste stored at the facility, a Yankee Rowe spokesman later explained that access is only allowed to storage personnel (Capstick, 2014).
Millstone Unit 1 and SAFSTOR
Millstone Unit 1 was an early boiling water reactor in Waterford, Connecticut, that was issued an operating license on October 7, 1970. Workers shut down the reactor on November 11, 1995, when it entered its 15th refueling outage. In July 1998, the company announced its decision not to restart the reactor and certified to the NRC that it had permanently ceased operations and removed all fuel from the reactor vessel on July 21, 1998 (Dominion Energy, 2011).
All fuel may have been removed from the reactor vessel, but not all fuel remained in the spent fuel pool as intended. On January 11, 2001, the owner informed the commission that two spent fuel rods were missing and presumed lost. It seemed that a spent fuel assembly had been disassembled in October 1972, to obtain components for testing and analysis. When the spent fuel assembly was reassembled in May 1974, it was without these two fuel rods. The two spent fuel rods were stored in a canister within the spent fuel pool as verified by visual inspection in May 1979. The canister and the fuel rods it contained were last seen in the Millstone Unit 1 spent fuel pool on April 30, 1980. Then they disappeared.
They probably were not stolen, because they emitted about 8,000 rem per hour when last seen and about 1,000 rem per hour when reported missing (yielding a lethal dose in four to 30 minutes). They probably were not stuck in a worker’s back pocket either, because they were more than 13 feet long. The owner thought the two rods may have been mistakenly shipped to a burial site in the state of Washington for low-level waste, or perhaps to a burial site in South Carolina (Northeast Nuclear Energy, 2001). The NRC fined the owner $288,000 for the lost fuel rods (Nuclear Regulatory Commission, 2002).
For the remaining spent fuel and for the rest of Unit 1, the owner opted for a modified SAFSTOR approach to decommissioning. The bulk of the decommissioning activities will be deferred until Millstone Units 2 and 3 permanently shut down. But due to uncertainty about continued access to burial sites for radioactive waste, some components were removed, cut up, and shipped off-site for disposal. For example, the drywell head was cut up and shipped to a metal processor. (The drywell head is the removable upper portion of the primary containment surrounding the reactor vessel. During refueling outages, the drywell head is removed to allow workers to access the reactor vessel to offload spent fuel bundles and replace them with new fuel bundles.)
The more highly radioactive reactor vessel internals were cut up and placed into the spent fuel pool to allow the reactor cavity and reactor vessel to be drained of water. If opportunities present themselves for the safe and economical removal of components or pieces, the company may opt to do so. Otherwise, the remaining work will be deferred until the adjacent reactor units are shut down and decommissioned (Dominion Energy, 2011).
Millstone Unit 1 lost two spent fuel rods. But to be fair, the plant’s owners disposed of more spent fuel by accident than the federal government has during the past three decades by intent.
Zion Units 1 and 2, and third-party DECON
Zion, which served the Chicago area, featured two of the first large pressurized water reactors to operate in the United States. Unit 1 received an operating license in October 1973, and Unit 2 received its operating license the following month. The Unit 2 reactor last operated on September 19, 1996, while the Unit 1 reactor ceased operating on February 21, 1997. By 1998, all the fuel had been off-loaded from the reactor vessels into the spent fuel pools (Nuclear Regulatory Commission, 2012).
The owner opted for DECON, the immediate dismantling option, but elected to outsource the project to a third party. In 2008, the owner applied to the NRC for permission to transfer the reactors’ licenses and decommissioning funds to ZionSolutions, a subsidiary of EnergySolutions, a private, for-profit, Salt Lake City-based multinational corporation devoted to recycling, processing, and disposing of nuclear material (EnergySolutions, undated). The NRC sroved the transfers in 2010. ZionSolutions’ plans include constructing an independent spent-fuel storage installation with two concrete pads for 61 dry casks containing spent-fuel assemblies (with no fuel rods yet missing) and four other casks for highly radioactive reactor components (ZionSolutions, undated). Once DECON is completed, the plan is to transfer the licenses and responsibility for the spent-fuel storage facility back to the owner.
Zion marks the first time in the United States that the entire decommissioning effort has been outsourced to a third party. In the past, the owners handled their own decommissionings, relying on subcontractors only for certain tasks. If the “decommissioning cavalry” is successful at Zion, it may serve as a template for at least some future decommissionings.
Decommissioning’s elephant
Federal regulations require nuclear power reactor decommissioning to be completed within 60 years. The NRC defines decommissioning as the process of safely closing a nuclear power plant (or other facility where nuclear materials are handled) to retire it from service after its useful life has ended. This process primarily involves decontaminating the facility to reduce residual radioactivity and then releasing the property for unrestricted or (under certain conditions) restricted use. This often includes dismantling the facility or dedicating it to other purposes. Decommissioning begins after the nuclear fuel, coolant, and radioactive waste are removed.
As shown in Table 1, seven reactors have been quasi-decommissioned: Michigan’s Big Rock Point, Colorado’s Fort St. Vrain, Connecticut’s Haddam Neck, Maine Yankee, California’s Rancho Seco, Oregon’s Trojan, and Massachusetts’s Yankee Rowe. Their reactor vessels, cooling towers, control rooms, containment structures, big pipes, small pipes, tiny pipes, long pipes, short pipes, and wiring—miles and miles of wiring—are all gone. But the spent fuel remains behind as a reminder of the federal government’s decades-spanning ineptitude.
Since the 1950s, the federal government has licensed more than 100 nuclear power reactors, without licensing a single site to store the tens of thousands of tons of highly radioactive waste they produce.
The federal government enacted regulations requiring nuclear power reactors to be decommissioned to greenfield status within 60 years, without acknowledging that tons of highly radioactive waste must remain smack in the middle of those green fields, solely because of its failure to open a long-term geological repository.
Footnotes
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
