Facebook Twitter YouTube
  News  Expand   News  
    About    Expand     About    
  Projects  Expand   Projects  
  Members  Expand   Members  
  Publications  Expand   Publications  
  Meetings  Expand   Meetings  
  Fellowships  Expand   Fellowships &nbsp
  Member Login
Home > Publications > Research Papers > > Appendix III: Civil Back-End Fuel Technologies—Pursuit of the Closed Fuel Cycle
The Back-End of the Nuclear Fuel Cycle: An Innovative Storage Concept

Appendix III: Civil Back-End Fuel Technologies—Pursuit of the Closed Fuel Cycle

There is constant debate about whether conventional reprocessing and recycling (that is, the closed fuel cycle) should be commercially deployed, given the sensitive processes inherent in these aspects of the fuel cycle. As technologies become more widely available, the debate increasingly circulates around which states are already incorporating these aspects of the fuel cycle, which states intend to do so, and whether these policies pose problems for the international community. Numerous reports have compared conventional reprocessing technologies, but as each one points out, it is tremendously difficult to create a uniform system of analysis or baseline for comparison. In hopes of contributing an additional perspective to the conversation, Appendix III approaches the civil back-end fuel cycle issue by outlining national policies and providing commentary on the leaders in conventional reprocessing as well as advanced chemical partitioning technology; Appendix IV presents a more detailed discussion of the program in China.

The motivations behind conventional reprocessing—that is, PUREX (plutonium-uranium extraction) and fabrication of the uranium-plutonium product streams into MOX fresh fuel—have evolved along with the technology. What was initially a process intended to separate plutonium from the used reactor fuel for its use in weapons and breeder reactors has become a method for using plutonium stockpiles as fuel in reactors. In addition to modifying plutonium for fuel and thereby lowering the proliferation risks of idle stockpiles,66 conventional reprocessing and recycling the fissile material fuel from a nuclear reactor also allow approximately 25 percent more energy to be extracted from the uranium or other fuel resource.67

Advanced chemical partitioning concepts enable various waste streams to be separated from one another. When fissile material is separated from all the actinides, the collective heat load can be removed from the used fuel, thereby potentially reducing the amount of repository space that is needed. Given that the price of uranium is not volatile in the near term and that there is enough uranium to support the present rate of worldwide consumption for many decades, the arguments for pursuing a closed fuel cycle are best framed either as a waste management tool and/or as a hedging strategy in the case of energy-security concerns, including temporary or long-term disruptions in uranium supply or price spikes for individual states.68 Also, because of the expense of conventional reprocessing, as well as the likely additional expense of more chemical partitioning techniques, the recovered uranium will have to be enriched and, therefore, is generally not economically competitive with mined uranium.69

Some states, such as France, view the opportunity to use a resource more completely and efficiently as worth the pursuit of chemical partitioning, recycling, and burn-up.70 Based on very optimistic scenarios in which heat load controls the amount of real estate at the repository, proponents of fully closing the fuel cycle assert that repository space could be reduced to approximately 800 cubic meters, compared with 16,000 cubic meters if the fuel cycle is not fully closed.71 For states facing particularly difficult storage constraints, such as China and India, advanced chemical partitioning technologies may become a critical component of their respective fuel cycles.

The decision calculus is different for each state, and these calculations inevitably change in line with technology advancements and the increasing abundance of used fuel accumulation. As Table 2 illustrates, given the varying inventories and used fuel policies of key nuclear consumer states, national policies on conventional reprocessing are diverse and dynamic. Below is a summary of policies and technologies (as of the end of 2007) of states that are still pursuing recycling today.

Table 2. Inventories of Dry and Wet Stored Used Fuels, as of the End of 2007

Country Used Fuel Inventory
(tons of heavy metal)
Used Fuel Policy
Finland   1,600 Direct disposal
France 13,500
Conventional reprocessing
Germany   5,850 Direct disposal (now)
Japan 19,000 Conventional reprocessing
Russia 13,000 Conventional reprocessing
Sweden   5,400 Direct disposal
United Kingdom   5,850 Conventional reprocessing
United States 61,000 Direct disposal

Source: Modified from Harold Feiveson, Zia Main, M.V. Ramana, and Frank von Hippel, "Spent Fuel from Nuclear Power Reactors: An Overview Study by the Internationla Panel on Fizzile Materials,” International Panel on Fissile Materials, June 2011. Table reprinted under the Creative Commons Attribution-Noncommercial License, http://creativecommons.org/licenses/bync/3.0/.


Presently, treating used nuclear fuel is the national policy of France, as it has been since the beginning of the country’s nuclear power industry. France sets the standard for conventional reprocessing: it is the only state actively performing conventional reprocessing on a large scale. AREVA estimates that 17 percent of all the electricity generated in France is from recycled fuels.72 Using the PUREX technology developed by the United States as a method to generate plutonium for nuclear weapons during World War II, La Hague, the sole plant in France, is capable of treating 1,700 MT of used fuel every year.73 During the PUREX process, the reusable uranium and plutonium are extracted from the other fissile material using an acid solvent. The plutonium and uranium are then sent to other plants in southern France and Russia where MOX fuels are made from both the plutonium and the uranium; in addition, other uranium is enriched for reuse.74 France is also in the planning phase for constructing a national deep geological storage facility in Bure, where storage experiments are under way. This storage facility will operate in conjunction with the fast-breeder program that France is also committed to launching as the next generation of reactors begins to become available; however, there are no power-scale breeder reactors presently in operation in France.


The existing infrastructure and capacity for conventional reprocessing in Russia is less mature than that of France. At Ozersk, the Mayak facility processes used fuel using the PUREX technology and has a yearly capacity of 400 tons.75 Currently, it employs only 25 percent of its capacity.76 The site produces uranium that is used in Russia’s nuclear reactors, that is, the WER-440-MWe LWR plants (two of which are located in Ukraine); in its nuclear icebreaking ships; and in the fast-neutron reactor, Beloyarsk (560 MWe). The Mayak facility does not function on a scale large enough to deal with the used fuel from the thirtyone operating nuclear power plants in Russia or those in Ukraine; however, Russia intends to expand its PUREX process, use built-up stores of plutonium for MOX fuel production, and incorporate fast breeder reactors back into the nuclear program.77 According to a recent calculation, there are upwards of 80 tons of plutonium stored in Russia for reuse (50 tons of reactor-grade and 34 tons of weapons-grade).78 The old site of a second, never-completed conventional reprocessing plant in Zheleznogorsk has become a storage location, in addition to the Mayak facility, for most of the used fuel in Russia; typically, fuel is transported there after an initial on-site cooling period in pool storage. A dry storage facility is under construction on the Mayak site as well and will increase the available storage capacity by 8,600 tons.79 To support the increase in nuclear energy production, geological repository siting is now under way in Russia.

United Kingdom

The United Kingdom has been treating used fuel from both its advanced gas-cooled reactors and its Magnox reactors to make MOX fuel. While there are no plans to pursue a breeder reactor program in the state, there is significant interest in processing the 100 tons of stored plutonium into MOX fuel for later-generation reactors. Facing pending closure, the fuel fabrication plant at Sellafield is currently used for export fuel, and until recently, U.K. authorities held that it was not economical to make MOX for domestic use.80The United Kingdom already has two types of conventional reprocessing plants located at Sellafield—Magnox and THORP—with capacities of 1,500 and 900 tons of fuel per year, respectively.81 Magnox fuel is uranium metal fuel (as opposed to uranium oxide) contained in magnesium alloy.82 The processes used for THORP and Magnox recycling differ because of the composition of the used nuclear fuel, but they result in a similar separation of uranium, plutonium, and the remaining fissile waste.83 To create a long-term fuel cycle strategy, the United Kingdom has tried to engage the public as much as possible. It has no plans for geological storage repositories but is looking into siting intermediate storage facilities for the current used fuel and the future used fuel.


The national policy in Japan, even though it is not operating a conventional reprocessing plant of its own, has always been to extract the maximum amount of energy from the purchased fuel; this approach is understandable (to the authors) for a state that does not have any uranium resources and must import the vast majority of its fuel for electricity.84 Historically, Japan has shipped its fuel abroad to Europe for treatment, although it had planned to operate a slightly advanced chemical partitioning plant at the storage facility Rokkasho. Presently, the storage facilities at Rokkasho are open and being utilized; however, like so many other storage sites in Japan, even before the plant has opened, its 20,400ton storage pool capacity has already been reached. All construction and planning for additional power reactors has been suspended following the unfortunate events at Fukushima Daiichi. Japan is reevaluating all its national energy policies, including operations at Rokkasho.


Presently, China has a small pilot PUREX plant that maximizes capacity at 100 tons per year. In order to be less dependent on uranium from external sources, China’s national policy is to add more capacity soon to harmonize with the rapid pace of new reactor builds in the state: twenty-five reactors are under construction—more than doubling China’s present nuclear energy infrastructure of fourteen reactors.85 AREVA has been collaborating with China to build a coextraction (COEX) facility for MOX fuels (with a capacity of 800 tons/year), which is scheduled to become operational in 2020. Additionally, the Gansu Province has been identified as the location for a geological repository. China intends to have a completely closed nuclear fuel cycle that includes the use of breeder reactors, and it appears to be equipped with both the resources and the political will to accomplish that goal. The collaboration between China and states such as Russia and France is also greatly contributing to China’s infrastructure development. See Appendix IV for a detailed discussion of China’s program.

United States, Finland, and Sweden

Each of these three states has chosen an open fuel cycle. In the open cycle process, used fuel rods are removed from the reactor core; stored in used fuel pools to cool; and then stored either in dry storage (robust casks cooled by the air) or in deep geological storage using clay or salt materials. The United States is less committed to this route than Sweden and Finland. The economic and social hurdles are significant in deep geological storage, as the United States is acutely aware. Used fuel pool and dry cask storage are both relatively inexpensive—approximately 0.3–0.4 mills/kWh.86 However, in the absence of a definitive national repository estimate in the United States, the U.S. disposal costs are unknown. In the case of Sweden and Finland, there is more knowledge, but these countries, too, lack final cost estimates of their disposal programs.87 Both Nordic states are actively engaged in the deployment of deep geological waste repositories, while the United States has struggled with siting, licensing, and building a single repository.88


66. Matthew Bunn, Steve Fetter, John P. Holdren, and Bob van de Zwaan, “The Economics of Reprocessing vs. Direct Disposal of Spent Nuclear Fuel,” Project on Managing the Atom, Harvard University, December 2003, http://www.publicpolicy.umd.edu/files.php/faculty/fetter/2003-Bunn-repro.pdf.

67. World Nuclear Association, “Processing of Used Nuclear Fuel,” January 2011, http://worldnuclear.org/info/inf69.html.

68.China is a clear outlier; for it to be less dependent on foreign sources of uranium and meet its projected nuclear power targets, China views reprocessing and recycling as its so-called strategic fissile material reserve.

69. Bunn et al., “The Economics of Reprocessing vs. Direct Disposal of Spent Nuclear Fuel.”

70. Based on 373 typical 1,000 MWe-pressurized water reactor (PWR) power plants with 34 percent efficiency; a capacity factor of 90 percent; 4,500 MW days/ton burn-up; and 21.5 MT of fuel annually, the approximate amount of opencycle, used fuel to be disposed of is 8,000 MT. However, utilizing a PUREX recycling scheme, projected estimates could be reduced to about 400 MT. See Marilyn Waite, “Cradle to Cradle: Turning Nuclear ‘Waste’ into Nuclear Fuel,” 2009, http://energy.sigmaxi.org/wp-content/uploads/2009/10/waite_recycling.pdf.

71. Ibid.

72. Ibid.

73. AREVA, “Recycling Used Fuel from Reactors,” http://www.areva.com/EN/operations1092/areva-la-hague-recycling-used-fuel.html. (accessed July 10, 2011).

74. World Nuclear Association, “Nuclear Power in France,” http://www.world-nuclear.org/info/inf40.html.

75. World Nuclear Association, “Processing of Nuclear Fuel,” http://www.world-nuclear.org/info/inf69.html.

76. World Nuclear Association, “Russia’s Nuclear Fuel Cycle,” http://world-nuclear.org/info/inf45a_Russia_nuclear_fuel_cycle.html.

77. Ibid.

78. Harold Feiveson, Zia Mian, M. V. Ramana, and Frank von Hippel, “Spent Fuel from Nuclear Power Reactors: An Overview of a New Study by the International Panel on Fissile Materials,” International Panel on Fissile Materials, June 2011, http://www.princeton.edu/sgs/publications/Managing-nuclear-spent-fuel-BAC-June-27-2011.pdf.

79. World Nuclear Association, “Russia’s Nuclear Fuel Cycle.”

80. As noted above, Sellafield’s continued reprocessing operations are at risk as a result of the NDA’s announcement on August 3, 2011, that it would close the Sellafield MOX plant; the NDA cited the significant negative impact to the Japanese nuclear industry in the aftermath of Fukushima.

81. World Nuclear Association, “Nuclear Power in the United Kingdom,” June 2011, http://www.world-nuclear.org/info/inf84.html; World Nuclear Association, “Processing of Used Fuel.”

82. The Institution of Electrical Engineers, “Nuclear Reactor Types” (London: Institution of Electrical Engineers, 1993), http://www.carnegieendowment.org/static/npp/reports/nuclear_reactors.pdf.

83. Nuclear Decommissioning Authority, “Magnox Fuel,” July 14, 2011, http://www.nda.gov.uk/strategy/spentfuelsmgmt/magnoxfuel/index.cfm; Sellafield Ltd., “Reprocessing Overview: Fuel Storage, Reprocessing, Uranium and Plutonium Recycling,” http://www.sellafieldsites.com/UserFiles/File/new_brochures/Reprocessing.pdf (accessed July 2011).

84. World Nuclear Association, “Nuclear Power in Japan,” June 2011,http://www.world-nuclear.org/info/inf79.html.

85. World Nuclear Association, “Nuclear Power in China,” June 2011, http://www.world-nuclear.org/info/inf63.html.

86.  According to an interdisciplinary study from MIT, “The cost of uranium today is 2 to 4% of the cost of electricity. Our analysis of uranium mining costs versus cumulative production in a world with ten times as many LWRs and each LWR operating for 60 years indicates a probable 50% increase in uranium costs. Such a modest increase in uranium costs would not significantly impact nuclear power economics”; see The Future of the Nuclear Fuel Cycle.

87. Finland has adopted a robust system to adjust fees on waste generators that can certify a full cost-recovery program.

88. Feiveson et al., “Spent Fuel from Nuclear Power Reactors.”