The Current Status, Safety, and Transportation of Spent Nuclear Fuel

Strong evidence shows that spent nuclear fuel can be stored and transported safely.

Spent nuclear fuel, nuclear fuel that has been in an operating nuclear reactor, is listed by the U.S. Nuclear Regulatory Commission (USNRC) as one of the three constituents of high-level waste- the other two are reprocessed (nuclear fuel) waste and "other highly radioactive material that the USNRC, consistent with existing law, determines by rule requires permanent isolation" (USNRC, 10 CFR Part 63). The current status of spent nuclear fuel is addressed by three acts of Congress: (1) Nuclear Nonproliferation Act of 1978; (2) Nuclear Waste Policy Act of 1982; and (3) Nuclear Waste Policy Amendments Act of 1987.

The Nuclear Nonproliferation Act of 1978 (P.L. 95-242), passed during the Carter administration, had the biggest impact on the disposition of spent nuclear fuel because it deferred indefinitely the commercial reprocessing and recycling of plutonium produced in the U.S. commercial nuclear power program.¹ The Nuclear Waste Policy Act of 1982 (P.L. 97-425) cleared the way for the federal government to select a site for a geologic repository for the disposal of high-level waste and directed utilities to levy a tax of 1 mil per kilowatt-hour on electricity generated by nuclear power to be paid into the Federal Nuclear Waste Fund, which would be used to develop and operate a repository. In return, the federal government agreed to accept ownership of spent fuel when a repository becomes available.

The 1987 Nuclear Waste Policy Amendments Act of 1987 (P.L. 100-203) altered the siting process and mandated that a single site, Yucca Mountain, Nevada, be characterized as a possible location for a geologic repository. The 1987 decision had a major impact on the nuclear fuel cycle, because pools of spent fuel had to be expanded and upgraded to accommodate the new interim storage requirement. The expansion involved installing high-density fuel racks so that more assemblies could be safely stored and required extensive safety analyses and licensing amendments. Today, some modified pools have reached their capacity, and on-site storage capabilities at some nuclear power plants have had to be augmented with dry storage casks.

On July 9, 2002, the U.S. Senate voted to allow the Yucca Mountain repository project to move into the licensing phase. On July 23, 2002, President George W. Bush signed the congressional resolution on Yucca Mountain allowing the U.S. Department of Energy (DOE) to prepare a license application for the repository. DOE has indicated that a license application will be submitted to USNRC by the end of 2004. Operation of the repository is scheduled for 2010.

Characteristics and Properties of Spent Nuclear Fuel
Spent nuclear fuel is produced by many types of reactors - light-water reactors, liquid-metal reactors, gas-cooled reactors, military reactors, test reactors, research reactors, and developmental reactors. Spent fuels involve a plethora of materials, including oxides of fissile materials, alloys of zirconium (zircaloy), stainless steel, aluminum, and for developmental reactors, exotic materials such as molten mixtures of several fluoride compounds. Fuel assemblies for commercial light-water reactors are bundles of zircaloy tubes filled with uranium dioxide (UO₂) pellets that are slightly enriched in uranium 235 (²³⁵U), pressurized with helium, and closed with welded zircaloy end plugs (Figure 1; see complete issue). For a pressurized-water reactor, the tubes have outside diameters of approximately 0.4 inch and overall lengths of approximately 14 feet; in a large reactor, some 200 tubes make up an assembly, and approximately 200 assemblies comprise the reactor core. The dimensions and fuel enrichments are different for boiling -water reactors, but the materials and neutronics are similar. Thus, they have essentially the same requirements for management of spent fuel.

The extent to which the fuel is irradiated (burned up) in the reactor determines the amount of radioactive waste created. The units of burn-up are usually taken to be the amount of energy produced per initial unit weight of the fuel (megawatt days per metric tons of heavy metal [Mwd/MTHM]); in nuclear reactors, the heavy metal is essentially uranium. When a nuclear reactor has achieved equilibrium in the production of radioactive species, there are some 51 new actinides and 250 fission product species, all radioactive, that were not there originally. Fortunately, only a few of these are important in spent fuel disposal; most of the other species are in very small quantities, have short half-lives, and have minor biological consequences.

There are three important categories of radioactive species for the design of a geologic repository. The first (⁹⁰Sr and ¹³⁷Cs) is not considered a health risk because of the relatively short half-lives of these species; they are, however, the main contributors to the heat released by spent fuel during the first several decades. The heat load is a major issue in repository design. In addition, ¹³⁷Cs is of concern during preclosure operations because of its shielding requirements.

The second category of radioactive species important for repository design comprises the fission products ⁹⁹Tc and ¹²⁹I. These products are very long-lived (half-lives of 2.12 x 10⁵ and 1.7 x 10⁷ years, respectively), and they are present in abundance in the inventory. In addition, they are generally soluble under geologic conditions and thus can migrate relatively quickly under ordinary groundwater conditions. The third category, from the actinide group of radioactive species, includes uranium, plutonium, neptunium, americium, and curium. Figure 2 gives some indication of the toxicity levels of these actinides as a function of time. The DOE Supplemental Science and Performance Analyses indicates that only ²³⁷Np poses a long-term risk (more than 100,000 years); the peak dose is ~35 mrem/yr at ~1 million years (DOE, 2001). The annual doses between 10,000 and 100,000 years are dominated by ⁹⁹Tc, with lesser contributions from ²³⁷Np and ¹²⁹I. Annual doses during the first 10,000 years are dominated by groundwater transport of carbon 14 (¹⁴C) and ⁹⁹Tc from waste packages that have undergone early failure. In case of an igneous disruption, the major contributors to the dose would be the actinides of americium and isotopes of plutonium.

Current Disposition of Spent Nuclear Fuel
The total estimated amount of spent nuclear fuel in the United States is ~50,000 MTHM. Some 47,000 MTHM is commercial spent fuel; ~2,500 MTHM is owned by the government and managed by DOE. If and when the Yucca Mountain geologic repository goes into operation, DOE will take ownership of all spent nuclear fuel. Table 1 is an estimate of the current disposition of all spent nuclear fuel in the United States based on projections from 2001 data (Holt, 2002).

Commercial Fuel
The combination of the continued operation of 103 U.S. nuclear power plants since 1972 (~20 percent of the nation's electric energy supply) and no reprocessing of spent fuel has resulted in an estimated total of approximately 47,000 MTHM. About 46,000 MTHM, or 98 percent of it, is stored at 72 commercial nuclear power plant sites in 33 states. Approximately 43,000 MTHM is stored in fuel pools, and 3,000 MTHM is in dry storage. Thus, more than 90 percent of commercial spent fuel is still in pools at nuclear plant sites. To date, 16 sites also have dry storage facilities, which were allowed under a general license issued to all operating nuclear power plants in the early 1990s. Fuel must be stored in casks that have been preapproved through a USNRC rule-making process. Another nine dry spent-fuel storage facilities were approved under site-specific licenses, a more complicated process that usually involves site-specific hearings. Two of the site-specific sites are owned and operated by DOE, at the Fort St. Vrain site in Colorado and at Idaho National Engineering and Environmental Laboratory (INEEL). Additional dry, site-specific storage facilities are in the planning stage or are going through the licensing process.

Beyond the nuclear plant sites, a small amount of spent nuclear fuel (less than 1,000 MTHM) is stored at other locations, including the General Electric Morris Operation in Illinois, the West Valley Demonstration Project in New York, and INEEL. Storage at the Morris and Idaho sites accounts for almost 95 percent of the approximately 1,000 MTHM.

Fuel Managed by DOE
DOE currently manages approximately 2,500 MTHM from the N-Reactor; experimental power reactors; material-production reactors; naval reactors; and test, research, and educational reactors. DOE also manages some 280 MTHM of commercial fuel submitted for examination and development purposes (DOE, 2002a). The majority of the fuel, about 2,100 MTHM, is from the N-Reactor and is stored in the K-East and K-West basins at Hanford, Washington. The spent fuel from the other reactor types represents a little more than 100 MTHM. DOE also manages about 1,000 MTHM of unirradiated fuel, the disposition of which has not been determined, but one possibility is that it will be treated as waste. Thus, the total amount of spent and unirradiated nuclear fuel managed by DOE is approximately 3,500 MTHM (NRC, 2003). A considerable amount of DOE-managed fuel (about 85 MTHM) will require some level of treatment to meet the requirements of the proposed Yucca Mountain geologic repository.

Interim storage for all of DOE’s spent nuclear fuel is to be provided at three national laboratories - Hanford in Washington, INEEL, and the Savannah River Site in South Carolina. New storage facilities will be necessary at all three sites. Waste from naval fuel is shipped to INEEL for storage at the Idaho Nuclear Technology and Engineering Center. Fuel from test, research, and educational reactors is stored at the Savannah River Site and INEEL. Table 2 is an estimate of the amount of commercial and DOE waste forms for disposal in the Yucca Mountain repository (DOE, 2002b).

Worker and Public Safety in Shipping

Experiential Evidence
The worldwide experience of storing, handling, and shipping spent nuclear fuel and high-level wastes is based on more than 50 years of operating nuclear reactors. Thirty thousand to 50,000 canisters have been shipped by all surface modes of transport (i.e., road, rail, and sea) involving an estimated 100,000 MTMH (Pope et al., 2000). U.S. experience is based on an inventory of approximately 50,000 MTHM. In the United States, between 1964 and 1997, 829 MTHM were shipped by road and 1,445 MTHM by rail; a total of 3,025 shipments. Although there were many more shipments by road, the tonnage of rail shipments exceeded the tonnage of shipments by road by a factor of about 2 (USNRC, 2002). Included in the rail shipments are naval spent fuel, which has been shipped for more than 40 years by rail in shielded shipping containers from naval shipyards, and prototypes, which are shipped to the Expended Core Facility at the Naval Reactors Facility in Idaho, where the fuel is removed from the containers and placed into water pools. U.S. experience includes both commercial and DOE-managed fuels. The data indicate that from 1979 to 1995 the commercial nuclear industry completed about 1,300 shipments of spent fuel - 1,045 by highway and 261 by rail.

The U.S. Department of Transportation reports that four highway shipments and four rail shipments were involved in accidents between 1971 and 1995, only one of which resulted in detectable damage to the cask. Although the driver was killed in the accident, radiation surveys at the scene indicated that the structural integrity of the cask was not compromised, and there was no release of radioactive contents (Weiner and Tenn, 1999). No injuries, deaths, or nonroutine exposures to radioactive material have resulted from transportation accidents.

Scientific Investigations
Numerous analytic studies and field tests have been done on the safety of transporting spent nuclear fuel by manufacturers of shipping containers as part of the licensing process, as well as by national laboratories, private contractors and consultants, DOE, and USNRC. USNRC studies include an environmental study on transport by air and other modes of transport, a study in 1980 on transporting radionuclides through urban areas, a study in 1987 on the response of shipping containers to severe highway and railway accidents, and in 2000 a reexamination of risk estimates for spent fuel shipments (USNRC, 1977, 1980, 1987, 2000a). As these analytical studies increasingly relied on risk assessments, the estimated safety levels increased.

Field tests have also been performed, and more are planned, to subject shipping casks to severe accidents. Sandia conducted crash tests sponsored by DOE in the mid-1970s: (1) a flatbed truck loaded with a full-scale cask crashed into a 700-ton concrete wall at 80 miles an hour; (2) a cask was broadsided by a 120-ton locomotive traveling 80 miles per hour; and (3) a transportation container was dropped 2,000 feet onto soil as hard as concrete (the container was traveling 235 miles per hour at impact) (Jefferson and Yoshimura, 1977).

Other tests were conducted by the Central Electricity Generating Board of Great Britain. Known as Operation Smash Hit, these tests included a live television demonstration of the integrity of a fuel cask. The test involved ramming an unmanned locomotive at 100 miles an hour into a cask used for shipping spent fuel from the United Kingdom Magnox nuclear power stations.

In no test, either in the United States or the United Kingdom, was a cask damaged to the point that radioactive material was released. The test results indicated that at the time of the tests analytical and scale-modeling techniques could predict vehicular and cask damage in extremely severe accidents with reasonable accuracy. They also indicated that spent fuel casks are capable of surviving very severe accidents.

Currently, USNRC is engaged in a program (referred to as the Package Performance Study [PPS]) to confirm the safety of full-scale casks licensed for rail and truck shipment (USNRC, 2000b). The program involves extensive public participation in the design of the tests. PPS will reexamine the level of protection provided by USNRC-certified transportation package designs under severe accident conditions. The program has two major objectives: (1) to demonstrate to the public through full-scale testing the safety of the spent fuel casks to be used to ship fuel to the proposed Yucca Mountain repository; and (2) to validate the methods used to assess the risk of transportation accidents involving shipments of spent nuclear fuel. The program is in the study and planning phase and is expected to continue through 2005.

The evidence showing the safety of the management and transport of spent nuclear fuel is impressive. This reflects both strict standards for shipping casks (e.g., impact, fire, and water-immersion tests), but also the relatively benign forms of the spent fuel. Unlike most hazardous materials, spent nuclear fuel is not a gas, liquid, or powder. In addition, neither mechanical or thermal energy is present to serve as a dispersion mechanism in the event the casks are penetrated or engulfed in fire. On the whole, undamaged fuel assemblies are very rugged and represent the first containment barrier for radionuclides.

There are some safety issues to be addressed, however, primarily because of differences between past and future shipments: the greater magnitude and increased complexity of the planned shipping campaign; the larger inventories of fuel assemblies that will be handled at any one time at multiple locations; new handling operations; and finally, subsurface emplacement operations.

The United States has limited experience in transporting spent nuclear fuel on the scale expected to support operations at Yucca Mountain. The proposed shipping campaign for Yucca Mountain is expected to last for 24 years and include shipments from 72 commercial sites and five DOE sites. If the shipments are by rail (something yet to be decided), 450 shipments will be required annually, a total of 10,700 shipments. If the shipments are by truck, the estimated number will be 2,200 annually, a total of 53,000 shipments (DOE, 2002b).

The risk-assessment studies performed to date have several limitations. The most significant limitation is that the studies are mostly generic, rather than operation-specific. Future studies should be performed for specific routes with specific human and mechanical resources. Alternative routes should also be considered to determine the most advantageous route. The public is entitled to have choices based on assessments of the risks, costs, and benefits of different routes and support systems.

Finally, there is the risk of terrorist attacks. USNRC, shipping cask manufacturers, and licensees should analyze the risk of terrorist attacks on spent nuclear fuel as rigorously as they analyze nuclear power plant safety. Specific scenarios should be developed and analyzed to pinpoint the vulnerabilities in the spent nuclear fuel cycle. Studies should include the likelihood and consequences of specific types of terrorist attacks under specific conditions.

Conclusion
There is strong evidence that operations involving spent nuclear fuel can be done safely. The experience base is solid in terms of the types of operations but limited in terms of the magnitude and repository-specific activities expected for future operations. Clearly, more emphasis must be put on repository-specific operations for the public to feel confident of the safety of geologic repository operations. For example, more risk-informed evidence will have to be developed on the safety of specific routes of shipments to the proposed Yucca Mountain repository.

I consider two actions very important. First, the record of experience with the shipping of spent nuclear fuel should be made available in a factual, understandable, and comprehensive form. The absence of a centralized, independent organization to collect, analyze, and disseminate the data has compromised the value of studies done to date. The case for the safety of operations involving spent nuclear fuel has not been well represented in the public domain. A centralized information collection and processing system would help.

Second, realistic risk studies should be done of specific alternate routes and means of transporting spent nuclear fuel to the repository site. Even if it turns out, as expected, that all of the routes can be made safe, the quantification of the risk, including the uncertainties of specific shipping routes, could reassure the public.

References
DOE (U.S. Department of Energy). 2001. Supplemental Science and Performance Analyses, Vol. 2, Performance Analyses. Washington, D.C.: DOE.
DOE. 2002a. Office of Spent Fuel Management Spent Fuel Data Base. Version 4.2.0. Washington, D.C.: DOE.
DOE. 2002b. Final Environmental Impact Statement for a Geologic Repository for the Disposal of Spent Nuclear Fuel and High-Level Radioactive Waste at Yucca Mountain. Nye County, Nevada. Washington, D.C.: DOE.
DOE. 2003. Fuel assembly for a light-water reactor.
Holt, M. 2002. Civilian Nuclear Waste Disposal. Order Code IB92059. Washington, D.C.: Congressional Research Service, Library of Congress.
Jefferson, R.M., and H.R. Yoshimura. 1977. Crash Testing of Nuclear Fuel Shipping Containers. SAND77-1462C. Albuquerque, N.M.: Sandia National Laboratories.
NRC (National Research Council). 1996. Nuclear Wastes: Technologies for Separation and Transmutation. Washington, D.C.: National Academy Press.
NRC. 2003. End Points for Spent Nuclear Fuel and High-Level Radioactive Waste in Russia and the United States. Washington, D.C.: National Academies Press.
Pope, R.B., X. Bernard-Bruls, and M.T.M. Brittinger. 2000. A Worldwide Assessment of the Transport of Irradiated Nuclear Fuel and High-Level Waste. Vienna, Austria: International Atomic Energy Agency.
USNRC (U.S. Nuclear Regulatory Commission). 1977. Final Environmental Statement on the Transportation of Radioactive Material by Air and Other Modes. NUREG-0170. Washington, D.C.: USNRC.
USNRC. 1980. Transportation of Radionuclides in Urban Environs: Draft Environmental Impact Assessment, Vols. 1 and 2, edited by N. Finley et al. NUREG/CR-4829. Washington, D.C.: USNRC.
USNRC. 1987. Shipping Container Response to Severe Highway and Railway Accident Conditions, edited by L.E. Fischer et al. NUREG/CR-4829. Livermore, Calif.: Lawrence Livermore National Laboratory.
USNRC. 2000a. Reexamination of Spent Fuel Shipment Risk Estimates, edited by J.L. Sprung et al. NUREG/CR-6672. Albuquerque, N.M.: Sandia National Laboratories.
USNRC. 2000b. Spent Nuclear Fuel Transportation Package Performance Study Issues Report, edited by D.J. Ammerman et al. Albuquerque, N.M.: Sandia National Laboratories.
USNRC. 2002. Summary of Department of Transportation Experience with Spent Nuclear Fuel Shipments. Presentation, Advisory Committee on Nuclear Waste, Working Group Session, Washington, D.C., November 19, 2002.
Weiner, R.F., and H.F. Tenn. 1999. Transportation Accidents and Incidents Involving Radioactive Materials (1971- 1998). Albuquerque, N.M.: Sandia National Laboratories.



Footnote
¹The Reagan administration lifted the ban on the reprocessing of domestic fuel in the early 1980s, but by then the economic situation, increased reserves of uranium ore, and a declining nuclear power industry provided little incentive for industry to resume reprocessing.


TABLE 1 The Disposition of Spent Nuclear Fuel in the United States (estimated through 2002)