(This article was published in the June 2011 issue of Nuclear Engineering International, before investigation of the Fukushima Daiichi spent fuel ponds revealed that none had reduced to the extent that spent fuel was uncovered. )
It remains rather early to assess the full consequences for the nuclear industry in the aftermath of the Fukushima accident. There are a huge number of separate bodies busily preparing reports with recommendations, and this process may well last for several years. However, it seems certain that the entire used fuel management system—on-site storage, consolidated long-term storage, geological disposal—will be re-evaluated in a new light because of the Fukushima storage pool experience. Highly-packed pools suffered loss of coolant, caught fire, and released significant quantities of radioactive material. This suggests that used nuclear fuel should be removed from reactor storage pools as early as possible.
The fact that used fuel storage has undoubtedly been something of an afterthought in many national fuel cycle policies has been brought into sharper relief. There may now be a renewed impetus to move used fuel into dry cask storage more quickly and/or away from reactor sites to consolidated storage and disposal. To many observers, this would be a good thing, although it may impose additional costs on currently-operating and future nuclear power plants. There have been continuous delays in most countries in deciding what to do with used fuel; any additional costs will have to be borne, and in some cases may be significant.
Used nuclear fuel storage is a required step in all open and closed fuel cycles, as a consequence of the nuclear characteristics of the used fuel. The radioactive decay heat and gamma radiation decreases rapidly with time. There are large safety and economic incentives to allow the radioactivity to lessen before transport, processing, or disposal. Upon reactor shutdown, radioactive decay heat decreases very rapidly, to as little as 0.5% in one week. The refuelling strategy in light water reactors (LWRs) is to transfer the used fuel from the reactor core to the used fuel storage pool, where the water provides cooling and radiation shielding. Over the following decade, the radioactivity will decrease by another factor of 100. Reactor used fuel storage is therefore designed as a safety function to provide time for the used fuel decay heat to decrease sufficiently so that a serious accident can no longer happen.
If used fuel is to be shipped elsewhere, typically the minimum time before shipment is two to three years. However, there are large economic incentives to store used fuel for a decade before transport. Spent fuel has to be shipped in heavy steel casks. With short-cooled used fuel, thicker walls are required to provide radiation shielding, resulting in less fuel per cask. Cask capacity is also restricted by the requirement to limit used fuel temperatures to avoid degradation. (The radioactive decay heat must be conducted out through the cask walls.) A decade of storage enables the use of more economic large-capacity casks that minimize the number of shipments.
Used nuclear fuel can also be transferred from the spent fuel pool to dry cask storage. This is a preferred option for long-term storage of used fuel because the cask has no moving parts (natural circulation air-cooling for decay heat removal) and requires very little maintenance. Like transport casks, there are economic incentives to store the fuel in the pool for a decade before transfer to dry cask storage.
Used fuel sent to a reprocessing plant may also be stored for long periods of time before reprocessing. Used nuclear fuel is a significant potential source of energy; however, we do not know today if LWR used fuel is a waste or a valuable national resource. Because of this uncertainty, the best option is surely a policy that maintains fuel cycle options—in other words, long-term storage of used fuel.
In the 1980s, the United States passed laws requiring disposal of used nuclear fuel on a specific schedule without considering storage; however, those legal requirements did not change the need for storage. The technical solution at the proposed Yucca Mountain repository was to place the used fuel in waste packages, put the waste packages in the repository, and cool the repository with airflow for 50 years after the repository was filled. In effect, the proposed Yucca Mountain repository would have functioned as a used fuel storage facility until conversion into a geological repository after 50 years.
The desire to reduce the decay heat of used nuclear fuel resulted in several countries building centralized storage facilities in the 1980s to age the wastes before disposal. Centralized storage has since become the preferred option for many countries (such as France, Japan and Sweden) with significant nuclear power programmes. All LWRs use short-term pool storage of used fuel. In Sweden, pool storage is used for centralized long-term storage at the CLAB facility in Oskarshamn. This facility, located 30 metres underground, has a capacity of 8000 tonnes of used fuel with a current inventory of 5000 tonnes. It opened in 1985 with the specific goal to store used fuel until the decay heat decreased sufficiently for disposal in the planned Swedish repository at Forsmark. When CLAB was built, pool storage was the only technology available for long-term storage of used fuel. Pool storage is also used at reprocessing plants because it allows easy retrieval of specific fuel assemblies to be reprocessed as a batch. France, Russia, Great Britain and Japan have centralized pool storage of used fuel to support their associated reprocessing plant operations.
Dry cask storage can additionally be used for short- and long-term storage of used fuel. It is a modular storage technology and is the chosen long-term used fuel storage technology in the United States and elsewhere. Dry cask storage is also used for centralized storage in Germany at Gorleben.
Long-term storage is only a viable option because the quantities of used fuel are small and the costs of storage are small relative to the value of electricity produced. A typical reactor produces 20 tonnes of used fuel per year, with total waste management costs (including used fuel storage) of between only 1 and 2% of the cost of nuclear electricity.
These used fuel issues have also been brought into focus by the recent publication of a study by an interdisciplinary group at MIT, namely “The Future of the Nuclear Fuel Cycle” (see also NEI January 2011, pp. 22-7). In 2003 MIT had published “The Future of Nuclear Power,” whose underlying motivation was that nuclear energy, which today provides about 70% of the ‘zero’-carbon electricity in the United States, is an important option for the marketplace in a low-carbon world. This report was well-reviewed internationally and has since been very well-referenced, particularly as a credible source on all the issues surrounding new nuclear plant economics. Major changes in the energy landscape of the United States and the rest of the world have occurred since 2003, and these were reflected in a 2009 update. The new study has been carried out to look more closely at the available fuel cycle options for nuclear power, given that to enable an expansion of nuclear power production, the industry must overcome critical challenges in cost, waste disposal, and proliferation concerns while maintaining its currently excellent safety and reliability record. In the relatively near term, important decisions may need to be taken with far-reaching, long-term implications about the evolution of the nuclear fuel cycle: about the type of fuel to be used, the types of reactors, what happens to irradiated fuel, and the method to be utilized for the disposal of long-term nuclear wastes.
The latest MIT report reminds us that for decades, the discussion about future nuclear fuel cycles has been dominated by the expectation that a closed fuel cycle based on plutonium start-up of fast reactors would eventually be deployed. However, this expectation is rooted in an out-of-date understanding about uranium scarcity. Now we know that there is clearly no shortage of uranium resources that might constrain future commitments to build new nuclear plants for much of this century at least. This is a point still occasionally questioned by some in the anti-nuclear lobby, despite the fact that it is demonstrably misguided.
Also, limited recycling in LWRs using mixed oxide fuel (as is being done in some countries such as France) offers only minimal benefits to resource extension and to waste management, at least on a global scale. It can be strongly argued that scientifically-sound methods currently exist to manage spent nuclear fuel. The report says that for the next several decades, a once-through fuel cycle using LWRs is the preferred economic option for the United States and is likely to be the dominant feature of the nuclear energy system both there and elsewhere for much of this century. Improvements in light-water reactor designs to increase the efficiency of fuel resource utilization and to reduce the cost of future reactor plants should continue to be a major research and development focus.
The report’s re-examination of fuel cycles suggests that there are many alternative viable fuel cycle options and that the optimum choice among them faces great uncertainties—some economic, such as the cost of advanced reactors, some technical such as implications for waste management, and some societal, such as the scale of nuclear power deployment and the management of nuclear proliferation risks. Greater clarity should emerge over the next few decades; assuming that the needed research is carried out for technological alternatives and that the global response to climate change risk mitigation comes together. A key message from the report is that we can and should preserve our options for fuel cycle choices by continuing with the open fuel cycle, implementing a system for managed LWR spent fuel storage, developing a geological repository, and researching technology alternatives appropriate to a range of nuclear energy futures.
Long-term managed storage preserves future options for spent fuel utilization at little relative cost. Maintaining options is important because the resolution of major uncertainties over time (notably the trajectory of worldwide nuclear power deployment, the availability and cost of new reactor and fuel cycle technologies) will determine whether LWR spent nuclear fuel is to be considered as a waste destined for direct geological disposal or a valuable fuel resource for a future closed fuel cycle. Preservation of options for future fuel cycle choices has been undervalued in the debate about fuel cycle policy, which, in the United States at least, was until recently dominated by a rush towards a large geological repository at Yucca Mountain that avoided serious consideration of reprocessing options. Yet managed storage is capable of being performed safely at operating reactor sites, centralized storage facilities, or geological repositories designed for retrievability (essentially an extended form of centralized storage).
The report concludes that planning for long-term managed storage of spent nuclear fuel—for about a century—should be an integral part of nuclear fuel cycle design. While managed storage is believed to be safe for these periods, an R&D programme should be devoted to confirm and extend the safe storage and transport period. The possibility of storage for a century, which is longer than the anticipated operating lifetimes of nuclear reactors, suggests that the United States and other countries should move toward centralized storage sites—starting with used fuel from decommissioned reactor sites and in support of a long-term used fuel management strategy. In the United States, this would have the additional benefits of resolving the federal government’s liability for its failure to start moving used fuel from reactor sites starting in 1998, a nonexistent service for which the utilities concerned have continued to pay (and have commenced legal actions seeking redress).
Permanent geological isolation will be required for at least some long-lived components of used nuclear fuel, and so systematic development of a geological repository still needs to be undertaken. The 2003 MIT report concluded that the science underpinning long term geological isolation is sound, and the new report asserts that this remains valid. But the siting of a geological repository for spent nuclear fuel and high-level waste has been a major challenge, as indicated by the termination of the Yucca Mountain project.
Steve Kidd is deputy director general of the World Nuclear Association, where he has worked since 1995 (when it was still the Uranium Institute). Any views expressed are not necessarily those of the World Nuclear Association and/or its members.Related ArticlesOutages at US nuclear plants hit four-year-high in Spring 2011 NRC approves North Anna restart