F O R U M O N P H Y S I C S & S O C I E T Y
of The American Physical Society 
January 2006
Vol. 35, No. 1 



Previous Newsletters

this issue

Contact the Editors


The Status of Nuclear Waste Disposal

David Bodansky

A. Introduction

        Nuclear waste disposal is now perhaps the most visible problem facing the nuclear power enterprise, and it has become a cliché to refer to the wastes as nuclear power’s Achilles’ heel. In this paper, after briefly describing the nature of the wastes and options for their disposal, we will explore the current status of waste disposal plans and the associated controversies, with special reference to the proposed Yucca Mountain repository.[i]

B. The wastes and stages in their handling

       The fuel in all U.S. power reactors and the large majority worldwide consists of solid pellets of uranium oxide, enclosed in thousands of long cylindrical fuel rods. One-third of the fuel is removed at intervals of about 18 months. Unless this spent fuel is reprocessed, it becomes the waste, remaining in this solid form throughout all subsequent transfer and storage. During the past decade, the 103 operating U.S. reactors have annually discharged something in the rough neighborhood of 2000 tonnes of fuel. By about 2010, the cumulative discharges will reach the planned Yucca Mountain capacity of 63,000 tonnes of commercial spent fuel.

       The spent fuel’s activity at first drops precipitously with time, due to the relatively short half-lives of most of the radioactive fission products, and then drops somewhat more gradually as the actinides decay.[ii] After one year the reduction factor, compared to the activity when the reactor is turned off, is about 75. It is roughly 400 after 10 years, 50,000 after 500 years, 400,000  after 10,000 years, and 3,000,000 after 100,000 years.

       The first stage of waste handling is at the reactor site, where the spent fuel is initially put into a water-filled cooling pool. In an increasing number of cases, as these pools fill up the older fuel is transferred into air-cooled casks that remain at the reactor site. These can suffice for many decades of storage, but keeping the wastes on-site is considered a short-term measure. Some countries (Sweden and Switzerland) have already instituted a second stage, namely “interim” storage at a centralized location. There is no such facility in the United States. An ongoing effort by a utility consortium to establish one on Indian land in Utah has received Nuclear Regulatory Commission (NRC) approval, but is opposed by Utah officials.

       Whether interim storage is implemented or bypassed, current planning throughout the world is for eventual storage in underground mined repositories, although as yet only the United States and Finland have identified specific locations.

       A basic decision is whether or not to reprocess the spent fuel. The most ambitious reprocessing proposals call for extracting all actinides from the spent fuel and returning them to a fast reactor where they are “burned up,” primarily in fission. This is the goal of pyroprocessing, currently under development. A less ambitious approach, in which only plutonium and uranium are removed, has been used for spent fuel from commercial reactors in France, the United Kingdom, and Russia, as well as for plutonium weapons programs. The remaining radioactive residues then become the wastes.

       The United States decided in the 1970s against commercial reprocessing, primarily out of concern that separated plutonium might be diverted by terrorists or used by governments elsewhere to inaugurate nuclear weapons programs. Further, with uranium in ample supply it was less expensive to dispose directly of the spent fuel. However, Congress in November 2005 moved to appropriate funds for the development of technologies to recycle existing spent fuel---an initiative that, if pursued, may lead to major changes in the U.S. waste disposal program and revive proliferation concerns.

       For the longer term, a breeder reactor economy may become desirable if a proliferation-resistant fuel cycle can be implemented. Pyroprocessing, with co-located facilities, here offers the most promise.[iii] Its successful development would limit proliferation dangers and change the “long-term” waste problem into a 500-year problem, because without the actinides relatively little activity remains after 500 years. It would also make uranium resources quasi-infinite, by increasing one-hundredfold the energy extracted from a given uranium deposit and at least another hundredfold by making dilute uranium sources, such as seawater, affordable. Just such a solution is contemplated in some DOE planning, using yet-to-be-built new reactors---the so-called Generation IV reactors.

C. The U.S. solution: Yucca Mountain

       After decades of national indecision, Congress in 1987 designated Yucca Mountain, Nevada as the sole site for study as a possible waste disposal site. Yucca Mountain is about 160 km northwest of Las Vegas, in a sparsely populated, dry desert region. The repository is to be situated several hundred meters below ground and about 300 m above the water table. It is to be honeycombed with tunnels, into which “waste packages” would be moved on permanent rail tracks. In current planning, each waste package would be a double-walled corrosion-resistant container holding about 8 or 9 tonnes of fuel. The dry environment and the surrounding geological formations limit the flow of water into the repository and impede the outward movement of any radionuclides that escape if the waste package is breached.

       Although design of the repository has long been underway, the detailed plan is still being modified. Thus, on October 25, 2005 the DOE specified that spent fuel is to be shipped in standardized canisters that would not have to be opened at Yucca Mountain, making fuel handing there simpler and “cleaner.”

       Radiation protection standards for nuclear waste repositories were promulgated by the Environmental Protection Agency (EPA) in 1985, but key parts were thrown out in 1987 by a U.S. Appeals Court for adjudged inconsistencies. Congress in 1992 then asked the National Academy of Sciences (NAS) to recommend “reasonable standards” for a Yucca Mountain repository and instructed the EPA to establish standards “based on and consistent with” the NAS recommendations. The most striking of the resulting recommendations was that the regulatory horizon continue for the period of geologic stability, taken to be about 106 years, rather than stop after the previously prevailing 104 years.[iv]

       The EPA responded with new “final” standards in 2001. These set a dose limit of 15 mrem/yr for the “reasonably maximally exposed individual” (RMEI) living in the neighborhood of Yucca Mountain. The standard was to be in effect for 104 years. Although the EPA presented justifications for departing from the million-year recommendation, another Appeals Court in 2004 ruled that this departure violated Congress’s instructions.

       The EPA’s proposed solution has been to retain the dose limit of 15 mrem/yr for the first 10,000 years and establish a new limit of 350 mrem/yr for the next 990,000 years.[v],[vi] In partial explanation, the EPA compared the “protected” RMEI to today’s residents of Colorado. It estimated the average natural background dose to be 350 mrem/yr near Yucca Mountain and 700 mrem/yr in Colorado [Ref. 5, p. 49037]. The addition of 350 mrem/yr from Yucca Mountain wastes would raise the RMEI’s calculated total dose to 700 mrem/yr, the mean dose in Colorado today. The EPA’s next steps are to review comments, make revisions, and then issue a final rule, presumably in 2006.

D. Calculated Yucca Mountain performance

       The performance of the Yucca Mountain repository relies largely on decay and delay. Decay steadily reduces the radioactive inventory. The radionuclides are delayed in reaching the biosphere because little water enters the repository, the multi-layer waste package protects the spent fuel, and most radionuclides that eventually escape move only slowly through the ground. For example, plutonium is probably retarded by more than a factor of 100 (compared to the water itself) due to temporary attachment to the rock through which it travels [Ref. 1, p. 270].

        The overall effectiveness of the design has been analyzed by the Department of Energy (DOE) in a series of Total System Performance Assessments (TSPAs), which use Monte Carlo techniques to calculate anticipated radiation doses for the RMEI. In the last TSPA reported before the DOE made its recommendation to construct the repository, the mean calculated dose for the RMEI was under 2 x 10-5 mrem/yr for the first 10,000 years.[vii] This is far below the EPA limit of 15 mrem/yr.

       The calculated doses rise subsequently, as the waste packages gradually deteriorate and escaping radionuclides travel to the accessible environment. The mean calculated dose reaches about 0.1 mrem/yr after 100,000 years and a maximum of roughly 100 mrem/yr after 400,000 years, followed by a slow decline. While far in excess of the 10,000-year standard, this dose lies below the proposed long-term limit of 350 mrem/yr.

       Although the TSPA results emerge from extensive calculations they probably represent only rough guides, because the times involved are far beyond those of human experience, and many uncertainties exist in the models. The DOE’s successive TSPAs have differed substantially, and future TSPAs can be expected to differ further, especially if waste handling plans are significantly changed.[viii] A parallel TSPA effort, carried out by the industry-sponsored Electric Power Research Institute (EPRI) found lower doses than did the DOE at 100,000 years and beyond.

       The DOE and independent outside groups, especially the Nuclear Waste Technical Review Board (NWTRB), have highlighted areas that need further investigation. These include:

a.  Volcanism (not included in the calculations described above). The annual probability of the repository being impacted by a volcanic event has been estimated by a DOE panel to be 1.6 x 10-8 and by the NRC to be up to 1 x 10-7. High doses could be produced if molten rock flows through a waste package and carries radioactive material into the atmosphere.   Considering both likelihood and consequences, the DOE concluded that the “probability-weighted dose” is less than 0.2 mrem/yr at all times, but the DOE and NRC positions are not yet reconciled.

b Waste canister corrosion. The DOE has allayed NWTRB concerns about some corrosion mechanisms, but questions exist about additional mechanisms.

c. Water Infiltration. Observations in 1996 found that 36Cl from nuclear weapons tests had reached the repository, suggesting that water flow from the surface was faster than originally anticipated. Other groups have not confirmed these observations, and their validity is in question.

d. Radionuclide transport. A better understanding is needed of the rate of radionuclide movement from the repository to the biosphere, including the accelerated rates that might result if radionuclides become attached to small particles (colloids) or can find extensive pathways through fractures in the rock.

To date, the DOE has tended to address uncertainties by making conservative (pessimistic) assumptions. Revised TSPA calculations, with up-dated assumptions, will be used for the DOE’s forthcoming application to the NRC for a construction license. However, the DOE reports that it is now “unable to estimate realistically when the license application will be submitted” pending incorporation of the October 2005 changes into its overall design.[ix] For its review, the NRC will carry out independent analyses, including its own TSPA. Assuming NRC approval is granted and the repository is built, a further NRC license will be required before the repository is opened to receive wastes.

E. Intergenerational responsibility

An obligation towards future generations is universally acknowledged, but there is little agreement as to the nature of this obligation. One view is that worrying about people 10,000 years hence is excessive and that concern for one million years is absurd. An alternative view is that our responsibilities persist undiminished for the indefinite future. These differences do not yield to analytic discussion, and in practice are resolved by the perforce arbitrary setting of standards.

The standards proposed by the EPA reflect aspects of both views. Thus, the EPA has set stringent standards that remain constant for the first 10,000 years, while its proposed dose limit for later times is much less demanding. As a corollary provision, the EPA specifically rejects the projection of “increases or decreases of human knowledge or technology”  [Ref. 5, p. 49063].

In an extensive discussion in the Federal Register, the EPA indicated that it did not want to “unreasonably constrain the current and succeeding generations’ abilities to pursue achievable solutions they deem best suited to meet the interests of all generations,” as might happen if the dose limit for the far future was too restrictive [Ref. 5, p. 49040]. It also wanted to protect distant generations from radiation exposures that “pose a realistic threat of irreversible harm or catastrophic consequences” ­[Ref. 5, p. 49038]. The proposed 350 mrem/yr standard serves both goals, in that it probably can be met and the resulting total dose is below the natural radiation dose experienced by many people today. 

A modest formulation of intergenerational equity is encapsulated in the guideline: “Each generation should strive to pass on to immediately succeeding generations an improved world, including the potential to sustain such improvements for the indefinite future.”[x] This guideline points to the need to consider the near-term consequences of rejecting the Yucca Mountain repository along with the consequences of developing it. The abandonment of the repository would be a serious setback for nuclear power---even if other waste disposal solutions might suffice from a technical standpoint---making it all the harder to cope with the problems of global climate change and of limited oil and gas supplies.[xi] The dangers these problems pose for people in this century and the next appear to be much greater in magnitude than the more distant dangers from Yucca Mountain, and addressing them deserves a high priority in any weighing of intergenerational responsibilities.

F. The road ahead

        Recent steps in the Yucca Mountain process at first went as anticipated. The DOE in early 2002 recommended proceeding with the repository, the President approved, the Governor of Nevada objected, and Congress overrode the objection. The further schedule called for a DOE application in 2004 to the NRC for a construction license, to be followed by a three-year NRC review. Assuming NRC approval, waste deliveries were to begin in 2010 and continue until 2033. The repository is to be monitored and kept open for changes, including waste retrieval, for fifty to several hundred years after the first deliveries.

     This schedule was disrupted by delays in the DOE’s preparation of its license application and the court decision in July 2004 that required new EPA standards. But, if there are no radical shifts in DOE policies, it seems reasonable to surmise that within the next several years the EPA will present its final standards, the DOE will find that the repository meets them and will apply for a construction license, and the NRC will grant it.

     However, hurdles will remain from lawsuits and political opposition. Some environmental groups find in nuclear wastes an effective weapon “to drive a final stake in the heart of the nuclear power industry” and are unlikely to give up this weapon.[xii] The State of Nevada also seems determined to stop the Yucca Mountain project. In the end, the fate of the project appears more dependent on court decisions and political power than on technical evaluations or broad policy considerations.

     If the Yucca Mountain project is defeated, it may be a long time before support can develop for an alternative because any site is likely to run into similar obstacles. The wastes could be safely retained for many decades in dry storage at the reactor sites or in one or more central interim storage facilities.   But this would be viewed as a stop-gap, and the resort to it seen as evidence of a basic and perhaps insurmountable problem with nuclear power.

     If the project is approved, and additional reactors are built, the output of spent fuel will increase. Possibilities for the long-term disposal of waste produced after 2010 include expansion of the Yucca Mountain facility, new geologic repositories elsewhere, or burial in deep boreholes.[xiii] It may also eventually become feasible to revisit the now taboo option of burial in the clay of the deep seabed. As discussed above, and perhaps influencing the recent congressional actions, the demands upon any waste disposal option would be reduced if a proliferation-resistant reprocessing fuel cycle is developed and implemented.

     David Bodansky

Department of Physics

University of Washington


[i] The discussion of technical details and developments before 2004 draws from: David Bodansky, Nuclear Energy: Principles, Practices and Prospects, 2nd edition (New York: Springer/AIP Press, 2004).

[ii] The actinides are the elements with atomic numbers from 89 through 103; they are formed in reactors by neutron capture and beta decay.

[iii] See, e.g., William H. Hannum, Gerald E. Marsh, and George S. Stanford, “Purex and Pyro are Not the Same,” Physics and Society 33, No. 3, 8-11 (July 2004).

[iv] National Research Council, Technical Bases for Yucca Mountain Standards (Washington, D.C.: National Academy Press, 1995).

[v] U.S. EPA, “40 CFR Part 197, Public Health and Environmental Protection Standards for Yucca Mountain, Nevada; Proposed Rule,” Federal Register 70, no. 161, 49014-49065 (August 22, 2005).

[Available at: http://www.epa.gov/radiation/docs/yucca/70fr49013.pdf]

[vi] The 350 mrem/yr limit applies to the median of the calculated Yucca Mountain doses, as discussed in Ref. 5, pp. 49041-49046, while the comparison is made to the mean Colorado dose. The median is less than the mean in typical dose distributions, and its use may relax the demands upon repository performance, but the EPA adopted it to avoid the distortions that could be created by a few extreme estimates.

[vii] U.S. DOE, Yucca Mountain Science and Engineering Report Rev. 1, Report DOE/RW-0539-1 (North Las Vegas, NV: U.S. DOE, 2002), Fig. 4-180.

[viii] The TSPAs were carried out for the commercial wastes in the form of spent fuel. If some of these wastes are the product of reprocessing (followed by solidification) the results would be changed, presumably in the direction of lower doses.

[ix] Michael R. Shebelskie et al, The Department of Energy’s Sixth Monthly Status Report, Docket No. PAPO-00 (Nuclear Regulatory Commission, November 1, 2005).

[x] This sentence is taken from Ref. 1, p. 363; see Ref. 1 for references to its antecedents.

[xi] See, e.g., Robert W. Albrecht and David Bodansky, “Oil, CO2, and the Potential of Nuclear Energy,” Physics and Society 34, No. 1, 12-15 (January 2005).

[xii] Michael McCloskey, then-chairman of the Sierra Club, as quoted in Luther Carter, Nuclear Imperatives and Public Trust (Washington, D.C: Resources for the Future, 1987), p. 431; also, Ref. 1, pp. 358-9.

[xiii] Consideration of the borehole option is urged, for example, in The Future of Nuclear Power, An Interdisciplinary MIT Study (2003). [Available at http://web.mit.edu/nuclearpower/]



Previous Newsletters

this issue

Contact the Editors