Today, the National Nuclear Security Administration (NNSA) released its long-awaited study of alternative options for disposition of excess weapons plutonium. The report confirms what I anticipated in a paper I presented in July 2013: the option of down-blending the plutonium with inert materials and emplacing it in the Waste Isolation Pilot Plant (WIPP) is feasible, has the least technical risk, and is (by far) the least expensive alternative. Read more >
April 29, 2014 3:46 PM EDT
March 25, 2014 3:14 PM EDT
The communiqués issued at the previous two Nuclear Security Summits said almost nothing about the dangers of separated plutonium. That was a problem. The 2014 Nuclear Security Summit communiqué does say something about plutonium—but the world would have been better off if it had remained silent on the issue. Read more >
January 11, 2011 4:52 PM EDT
Last week several U.S. media organizations published reports claiming China made a “breakthrough in spent fuel reprocessing technology.” The original BBC story (it has since been replaced with a less sensational version) described the advance as “a new method of reprocessing irradiated fuel;” the current version says that China has “now perfected a procedure that will allow them to reprocess spent nuclear fuel.”
The stories suggest that this was a new process that differs from the process used by other countries that reprocess spent fuel. “Reprocessing” refers to extracting plutonium and uranium from fuel rods that have been used in a nuclear reactor. These recovered materials can be used to make new nuclear fuel. The stories reported that this breakthrough would allow China to produce sufficient amounts of nuclear energy for thousands of years using China’s domestic reserves of uranium, suggesting that this was due to China’s new process.
All of the reports were based on a three-minute Chinese Central Television (CCTV) news broadcast that aired on January 3rd. Stories touting progress are standard fare on Chinese television news in the first week of the new year.
This introductory CCTV report on Chinese reprocessing, which compared spent nuclear fuel to coal ash as a way of explaining the concept of spent fuel to its audience, was nothing out of the ordinary. The statements about thousands of years of nuclear energy were hyperbolic extrapolations based on estimates of China’s proven uranium reserves and the imagined theoretical efficiencies of fast breeder reactors that no country has succeeded in commercializing, despite decades of effort and billions of dollars of government investments.
The broadcast described the progress made during 2010 at plant 404—a nuclear facility in Gansu province that China is building as a pilot-scale commercial reprocessing plant. While the broadcast described recent progress at the plant using the Chinese word for “major breakthrough” (重大突破) it was clear that it was using this term to describe something that was a first for China, not a breakthrough in reprocessing technology. This is emphasized repeatedly in the short broadcast. The Chinese engineers who were interviewed explained that what was important is they did this completely on their own, with their own technology, and that their ability to extract uranium and plutonium from spent nuclear fuel in a commercial pilot plant allows China to join several other nations that already have this capability.
And even this “first” has to be qualified. It was the first time spent fuel had been reprocessed at this commercial reprocessing plant, which experts believe China completed in 2004. (It is not clear why it took six years for reprocessing to start at the facility, but it seems to be typical of the rate at which China’s nuclear energy program has been progressing.) But China began reprocessing as part of its nuclear weapons program in 1968, so it is not a first for Chinese reprocessing.
There was only one sentence in the CCTV broadcast about what China actually did, which is “the production of up-to-standard uranium and plutonium products” from spent fuel. This appears to mean that it produced separated plutionium and uranium that were pure enough to be used to make nuclear fuel.
Besides the technical difficulty and high cost of reprocessing and using plutonium to fuel reactors, commercial reprocessing creates large amounts of separated plutonium—which can be used to make nuclear weapons—and therefore increases the risk of nuclear proliferation and nuclear terrorism. For this reason, the U.S. government officially abandoned reprocessing in the 1970s.
October 15, 2010 12:16 PM EDT
On October 12, I was on a panel testifying to the Reactor and Fuel Cycle Technology Subcommittee of the Blue Ribbon Commission on America’s Nuclear Future about proliferation and security risks associated with new fuel cycle technologies. The other panel members were James Acton, Robert Bari, Richard Garwin, and Robert Gallucci. Here is the summary of my comments (my slides are also available):
#1. A major obstacle to reducing proliferation risk is the continued commercial production of plutonium. The U.S. can help to put the brakes on this practice by getting the geologic repository program back on track, demonstrating that safe and secure direct disposal of spent fuel is both technically and politically feasible. Technical approaches to reduce the risks of reprocessing, whether by improving safeguards technologies or by developing advanced closed fuel cycles, are unlikely to be effective, and it would be helpful if the U.S. government were to discontinue conveying the impression that transformational technologies could save the day. The U.S. also could apply significant influence on the fuel cycle policy of other nations through its bilateral nuclear cooperation agreements, but it has failed to take full advantage of these authorities.
#2. The assessment of proliferation risk is a complex undertaking that involves a host of technical and political factors, and is rife with subjectivity and uncertainty. To address these uncertainties, it is essential to base these assessments on conservative judgments of the capabilities of both state and non-state actors. One should not limit the scope of analysis by mirror-imaging the adversary or by fighting yesterday’s battles.
#3. The credibility of the international safeguards system can be strengthened by minimizing the disparity in obligations between nuclear-weapon and non-nuclear weapon states. To this end, the U.S. should lead by example and place all proliferation-sensitive fuel cycle facilities on the list of those eligible for IAEA safeguards. For new plants, this should entail providing design information to the IAEA upon the decision to construct, affording IAEA the opportunity to develop a verification approach early in the design and construction process. The U.S. should also take the lead in developing a financing mechanism that will provide the IAEA with the resources necessary to apply safeguards in weapon states. An alternative approach to eliminating disparities would be to place all fuel cycle facilities, whether in weapon- or non-weapon states, under the control of a new international entity with its own safeguards agreement.
#4. For control of proliferation and terrorism risks, the “polluter pays” principle should apply. That is, facilities that handle proliferation-sensitive materials should be assessed a tax commensurate with the danger posed by the materials and the cost of the appropriate level of safeguards and security. However, cost is not the only factor: it may be impossible to render such facilities sufficiently secure even if funds were unlimited.
#5. In the strictest sense, physical protection cannot be “risk-informed” because the probability of an event is a fundamental component of risk, and probabilities cannot be calculated for scenarios that involve deliberate actions, such as theft of weapon-usable material. Some speak of “risk-informing” physical protection as a way to reduce security requirements on certain items containing weapon-usable material, such as mixed-oxide fuel, based on the perception that such items are less attractive to terrorists. These proposals are misguided because they make tacit assumptions about the limited capabilities of adversaries that are likely to be unrealistic today, and are bound to become even more unrealistic as the tactical skills, technical knowledge and weaponry of terrorist groups continue to grow more sophisticated. We believe a truly risk-informed analysis of the threat of nuclear terrorism would lead to significant increases in security requirements for weapon-usable materials across the board and would ultimately discourage the continued production of such materials.
#6. NRC regulations should require that safeguards and security be fundamental considerations in the design of all new nuclear facilities, which is not the case today. NRC regulations do require an evaluation of the “safety-security” interface when licensees make changes to either the safety or security configuration of a nuclear power plant; such evaluations should be extended to fuel-cycle facilities.
September 17, 2010 3:39 PM EDT
My testimony before the Reactor and Fuel Cycle Technology Subcommittee of the Blue Ribbon Commission on America’s Nuclear Future (the “BRC”) elicited a predictable and depressing reaction from certain corners of the blogosphere. I informed the Subcommittee that although UCS does not oppose nuclear energy per se, we do oppose reprocessing spent nuclear fuel because of the security, safety and proliferation risks that it poses. I then presented the Subcommittee with a summary of the rationale behind our position, complete with numerous technical references. The UCS position was in direct opposition to that of four of the six members of the panel I was on (representing AREVA, General Electric-Hitachi, Westinghouse and Energy Solutions), who all supported spent fuel reprocessing and “recycling” strategies of one sort or another.
My testimony appears to have given certain bloggers heartburn. Rod Adams of Atomic Insights saw fit to criticize my competence, my understanding of technology and my use of what he called “unsubstantiated statements and vague references.” Yet he was unable to actually point to anything specific in my testimony that he could contradict. Instead, he posted a video clip of my presentation and invited his loyal readers to defend the faith by “dissecting” my testimony.
I would be more than happy to engage Mr. Adams’ readers in a technical debate on these issues, so I thought, frankly, that this was a fine idea. However, two weeks later, it appears that Mr. Adams’ gambit has backfired. Out of twenty comments, only one actually professes any knowledge of any of the references that I cited. Most simply repeat unsubstantiated assertions themselves. Some claim that I must have misinterpreted the references but did not actually bother to look them up. Several are ad hominem attacks on me or UCS. A couple actually agree with some of the points I made. All in all, not a very impressive showing. In fact, I found only two statements that merit a response. Below, I respond to those statements.
Mr. Adams and his readers should rest assured that every statement I make is supported by direct references and transparent analysis. In the future, I’d appreciate that observers interested in a technical debate actually consult my written works and references before throwing darts.
There are three main points to my testimony.
1. Reprocessing spent nuclear fuel has only a marginal impact on the volume of high-level waste requiring disposal in a geologic repository, while significantly increasing the volume of other forms of nuclear waste also requiring secure disposal.
Some reprocessing advocates argue that the technology can reduce the volume of high-level nuclear waste requiring disposal in a geologic repository. On the Atomic Insights blog, Lars Jorgensen says that it is “easy” for any recycling system to significantly reduce waste volume.
However, reprocessing actually increases, not decreases, the total volume of long-lived nuclear waste that must be stored and eventually buried in a geologic repository It only slightly reduces the volume of high-level nuclear waste that must be disposed of in a repository, as required by the Nuclear Waste Policy Act. At the same time, it significantly increases the volume of “greater-than-class-C” low-level waste, which cannot be legally disposed of in near-surface low-level waste facilities and would therefore need to be buried in a geologic repository as well. In addition, reprocessing increases the volume of low-level waste that must be disposed of in NRC-licensed near-surface facilities. Only one new low-level waste facility has been licensed in the United States in decades, and no policy (not to mention a repository) exists for disposal of GTCC LLW.
According to Argonne National Laboratory data cited by the Energy Department’s 2008 Global Nuclear Energy Draft Programmatic Environmental Impact Statement, compared to the once-through cycle, a fast-reactor-based reprocessing and recycle system would increase the total volume of radioactive waste by a factor of about seven. In particular, reprocessing would generate, in terms of volume,
- Seven times as much Class A, B and C low-level waste (LLW)
- 166 times as much greater-than-class-C LLW (over 8,000 cubic meters annually on average)
- only 25% less high-level waste (HLW)
The last data point conflicts with public statements being made by AREVA, which continues to claim that “recycling reduces by 75% the volume of high-level waste that must be sent to a repository.” However, this assertion is not even supported by AREVA’s own data. According to a 2009 presentation to the Nuclear Waste Technical Review Board, Dorothy Davidson of AREVA indicated that the volume of reprocessing waste requiring geologic disposal (vitrified high-level waste and compacted hulls and end pieces) was 10 cubic feet per metric ton heavy metal of initial spent fuel reprocessed (10 ft3/MTHM). In the same presentation, Davidson claimed that this should be compared a spent fuel volume of 45 ft3/MTHM, so that reprocessing results in a more than four-fold decrease in waste volume.
However, this latter figure is incorrect. As Table S.3-1 shows, the volume of light-water reactor spent fuel is actually closer to 15.8 ft3/MTHM (0.45 m3/MTHM). Therefore, the HLW volume per MTHM according to AREVA’s own data is only about 37% less than the initial volume of spent fuel – a much less impressive reduction than the 75% cited by AREVA, and one much closer to the Argonne/DOE estimate. Apparently, this discrepancy stems from the fact that AREVA did not directly compare the volume of HLW to the volume of spent fuel, but compared the volume of HLW to the volume of the spent fuel waste disposal package that was originally considered for the Yucca Mountain repository, which had a significant amount of empty space. This is not an apples-to-apples comparison.
But in any event, heat load, not volume, is typically the limiting factor in a geologic repository. If plutonium and other transuranic elements such as americium are removed with very high separation factors, the heat load of the residual waste will be reduced. However, unless the actinides that are removed from the spent fuel are actually destroyed through fission in a reasonable period of time, they will have to be stored for an indefinite period (posing many of the same concerns as indefinite interim storage of spent fuel), and eventually will have to be disposed of in a repository. Yet as the discussion in the next section indicates, reprocessing and transuranic recycle systems are not capable of significantly reducing total actinide inventories in a reasonable period of time.
2. Reprocessing and recycling spent nuclear fuel, whether in thermal or fast reactors, cannot effectively reduce the total quantity of hazardous radionuclides like plutonium and other transuranic elements that would require disposal in a repository.
Numerous studies have shown that fast reactor (FR) recycle systems are very slow and inefficient in actually fissioning transuranic elements, even if they operate in burner mode with very low conversion ratios. A recent study by the Electric Power Research Institute (EPRI) and Electricité de France (EdF) examined the impact of phasing in a fast reactor system operating in tandem with light-water reactors (LWRs) (35 percent FRs and 65 percent LWRs) and operating at a conversion ratio of 0.5, while keeping constant the total U.S. nuclear generating capacity.  The study found that the total inventory of plutonium and other transuranics would increase to over 1500 metric tons – roughly three times today’s inventory – and would remain essentially constant after that. Thus the system is simply not capable of reducing the total transuranic inventory, and the popular image that such a system can “burn up” nuclear waste is simply not accurate.
The EPRI study also compares the transuranic inventory in the system to the inventory that would accumulate in spent fuel if the U.S. continued with the once-through cycle. The analysis finds that the system would have to operate for 70 years just to reduce the total inventory of transuranics in the system by 50 percent relative to the once-through inventory. To reduce the inventory by 90 percent would require continuous operation for 632 years. Thus the system of reprocessing plants, fuel fabrication plants, fast reactors and associated facilities would have to operate over a period spanning many generations – and be rebuilt many times – before it could achieve a significant reduction in actinide inventory and a significant decrease in repository heat load compared to the once-through cycle. The paper concludes that “the analysis for the specific [recycling] scenario considered shows that it would take many decades, even centuries, for significant waste management benefits to materialize.” This is consistent with the one of the conclusions of the MIT study “The Future of the Nuclear Fuel Cycle,” released earlier this week – namely, that the choice of fuel cycle would make “little difference” in the total transuranic inventory in this century.
Proposals that require the essentially indefinite reprocessing and recycling of spent fuel do not provide a suitable foundation of nuclear waste management because they are inconsistent with the “intergenerational equity” principle. This principle, which underlies the rationale for a geologic repository for nuclear waste, includes the provisions that (1) those who generate the wastes should take responsibility, and provide the resources, for the management of these materials in a way which will not impose undue burdens on future generations; and (2) that a waste management strategy should not be based on a presumption of a stable societal structure for the indefinite future, nor of technological advance; rather it should aim at bequeathing a passively safe situation which places no reliance on active institutional controls.
A system that would require hundreds of years of costly and complex operations to achieve only a modest reduction in repository space requirements is not consistent with these principles. Some reprocessing advocates argue that nuclear materials that are in the fuel cycle – that is, in reactors, fuel fabrication plants, and above-ground storage facilities – need not be counted as wastes. This is only true, however, as long as the system is running. If it shuts down for any reason, these materials will require secure disposal. Thus our generation would be bequeathing to future generations the obligation of keeping the system going, without regard to cost or risk. This is clearly inconsistent with intergenerational equity.
3. Advanced reprocessing technologies do not significantly reduce nuclear proliferation and nuclear terrorism risks relative to current reprocessing technologies that produce separated plutonium.
Energy Secretary Chu has spoken of the proliferation risks associated with conventional reprocessing technology, as practiced in France and Japan, and has expressed confidence that the U.S. can develop alternatives that are “proliferation-resistant.” One of the goals of the Bush Administration’s Global Nuclear Energy Partnership program (GNEP) was also to develop so-called proliferation-resistant reprocessing technologies that did not produce “separated” plutonium.
However, a study conducted by the nuclear weapons labs reviewed the entire suite of technologies that were under study, including modified aqueous reprocessing and pyrometallurgical processing (“pyroprocessing”), with regard to their potential for reducing proliferation concerns.  The study found that the products of these processes, mixtures of plutonium and other actinides such as neptunium, americium and curium, are attractive for use in nuclear weapons or nuclear explosive devices. It concluded that there is no “silver bullet” technology that would eliminate the safeguards and security issues associated with reprocessing, and also that “none of the proposed flowsheets examined to date justify reducing international safeguards or physical security protection levels. All of the reprocessing or recycling technologies evaluated to date still need rigorous safeguards and high levels of physical protection.”
It should be noted that this study only analyzed the direct usability of these materials in nuclear weapons without further processing. It did not consider the potential for theft and off-site purification of these materials. As we and our colleagues have noted at length elsewhere, alternative reprocessing technologies under consideration do not confer significant self-protection against theft.
If stolen, these materials could be readily processed to produce even more attractive materials for weapons use.
One of the Atomic Insights blog post replies (Steve Skutnick, September 3), claimed that I misrepresented Bathke’s study. Skutnickasserted that the statements I made regarding the study’s conclusions with regard to subnational groups (terrorists) actually applied only to state-level actors. However, Skutnickis wrong. If he had actually read any of Bathke’s reports, he would have learned that the so-called Figure of Merit (FOM1) that I used “is applicable for evaluating the attractiveness of SNM or ANM for a sub-national group, for most of the less advanced proliferant nations, or for a technically advanced proliferant state.”
The other Figure of Merit (FOM2), which Mr. Skutnickasserts is designed for a “sub-state level actor,” is actually intended only “for a very few relatively unadvanced proliferant nations that desire a reliably high yield.” Thus Mr. Skutnickis wrong in attacking me for “abusing such metrics.”
 A. Machiels, S. Massara and C. Garzenne, “Dynamic Analysis of a Deployment Scenario of Fast Burner Reactors in the U.S. Nuclear Fleet,” Proceedings of the Global 2009 Conference, Paris, France, September 8-11, 2009, pp. 2567-2574. Also cited in the testimony of J. Kessler, EPRI, to the July meeting of the Reactor and Fuel Cycle Technology Subcomittee of the Blue-Ribbon Commission.
 C. Bathke et al., “An Assessment of the Attractiveness of Material Associated with a MOX Fuel Cycle From as Safeguards Perspective,” 50th Annual Meeting of the Institute of Nuclear Materials Management, Tucson, AZ, July 2009.