5

Management and Disposal of Nuclear Waste from Advanced Reactors

This chapter responds to the charges in the statement of task that call for evaluating the waste management and disposal options for the various proposed advanced nuclear reactors. The committee conducted this evaluation by accounting for typical volumes and physical, chemical, and isotopic characteristics of waste streams, including from possible reprocessing, from these advanced nuclear reactor technologies, and examining transportation, storage, and ultimate disposal requirements for these wastes.

The committee first provides the summary, findings, and recommendations (Section 5.1), and then describes the U.S. nuclear waste management and disposal program (Section 5.2), discusses the concept of geologic disposal of highly radioactive nuclear waste (Section 5.3), describes what the committee learned from experts’ briefings (Section 5.4), delves into the specific waste issues that arise from advanced nuclear reactors and fuel cycles (Section 5.5), assesses the potential impacts of advanced nuclear fuel cycle wastes on storage and transportation operations (Section 5.6), and provides an overview of decontamination and decommissioning of nuclear power plants relevant for this study (Section 5.7). Throughout the chapter, the committee provides observations on waste management and disposal for advanced nuclear reactors and fuel cycles.

5.1 CHAPTER 5 SUMMARY, FINDINGS, AND RECOMMENDATIONS

As described in Chapters 3 and 4, the development of advanced nuclear reactors would require significant efforts to establish a supporting fuel cycle. Broadly, there are two fuel cycles: open and closed. A closed fuel cycle involves reprocessing and reuse of fissile material, mainly uranium and plutonium, in order to harvest some of the remaining energy content of spent fuel. Chapter 4 deals extensively with different strategies for closed fuel cycles. In contrast, an open fuel cycle is based on the direct, permanent disposal of spent fuel after irradiation, as well as the high-level waste that may result from chemical reprocessing. Hence, for an open fuel cycle (for the purposes of this report), spent fuel is waste.¹ In the United States, the present strategy is an open fuel cycle with direct disposal of spent fuel in a geologic repository. Indeed, the proposals for advanced reactors are based on an open fuel cycle and disposal of the many different types of fuels that are now proposed or considered. As concluded in the 1996 National Research Council report Nuclear Wastes: Technologies for Separations and Transmutation, regardless of whether a country adopts an open or closed fuel cycle, a geologic repository is required.

___________________

¹ Here, a monorecycling fuel cycle that results in spent mixed oxide fuel that will be disposed of is considered an open cycle.

Section 5.2 sets the stage with a historical review of U.S. nuclear waste management, highlighting the poor planning, continued programmatic delays, and failure to create a geologic repository. It then outlines key concepts, strategies, and challenges for geologic disposal to give readers an understanding of the relevant factors and problems that advanced reactors should target if their goal is to mitigate nuclear waste management and disposal challenges. Advanced reactors that change the inventory of actinide isotopes by using new fuel types, increased levels of enrichment, higher burnup, or actinides from reprocessing will not be solving a problem that could not be addressed by selection of a repository site with appropriate geochemical conditions and geologic environment and may only contribute to waste disposal problems by introducing new, complex waste streams that require application of additional technology before disposal.

The subsequent sections highlight these unique waste considerations for different reactor types during all stages of the fuel cycle and for all levels of waste. For advanced reactors, the different types of fuel, burnups, enrichment levels, and processing steps for the fuels being proposed raise new questions about waste management and disposal requirements that are not well understood or well studied by reactor developers. Per the committee’s statement of task, this chapter evaluates the various management and disposal options for the fuel from advanced reactors; accounts for typical volumes and characteristics of waste streams from advanced reactors; and examines the transportation, storage, and disposal requirements for these wastes. Key issues include managing large volumes of irradiated graphite, handling and developing stable waste forms for sodium-laden and molten salt waste streams, and developing storage and transport packaging for fresh and irradiated high-assay low-enriched uranium (HALEU) fuels. Overall, the implementation of advanced reactors and fuel cycles will not solve problems associated with nuclear waste disposal, and may even exacerbate them depending on the nature and quantity of the waste streams generated.

From these analyses, the committee made the following findings and recommendations:

Finding 11: As the United States nears the 40th anniversary of the Nuclear Waste Policy Act (NWPA) (Public Law 97-425) and its Amendments (Public Law 100-203, Part E), there is no clear path forward for the siting, licensing, and construction of a geologic repository for the disposal of highly radioactive waste (mainly commercial spent nuclear fuel). The United States finds itself in this difficult situation for many reasons, including (1) changes to the original NWPA of 1982 that moved the process of site selection from a consideration of multiple sites to a single site, Yucca Mountain, Nevada; (2) a slowly developing and changing regulatory framework that provided late guidance in the site selection process and the evaluation and comparison of multiple sites; (3) ineffective management of the Nuclear Waste Fund ($45 billion) by Congress, which treated what was to have been a ratepayer escrow account as if it were taxpayer monies; (4) consequential policy changes occurring with changing administrations; (5) conflicting congressional and executive policies; and (6) insufficient public engagement in decisions concerning the basic strategy for the storage and disposal of the waste. The continued delay in planning and progress has only made the situation more complicated, as the present legal and regulatory frameworks have become outdated and even more limiting. Numerous assessments during the past decade, notably the Blue Ribbon Commission on America’s Nuclear Future (2012) and Reset of America’s Nuclear Waste Management Strategy and Policy (2018), have outlined a way forward. The committee agrees with common recommendations of these studies to establish a single-mission nuclear waste management and disposal entity, for which models have been proposed that deserve consideration by Congress. The entity could be governmental, partially governmental, or private; as to the latter option, the committee notes that two successful programs are being led by fully private entities: Posiva in Finland and SKB in Sweden. Important attributes of the entity are described in Recommendation G.

• Such an entity should be responsible for “cradle-to-grave” care and disposition of spent nuclear fuel—that is, from its discharge from a reactor plant to its final disposal in a repository. This entity should have continuity of leadership and funding, as well as a consistent disposal strategy. It should also have high technical and scientific competence, be able to organize and lead research programs and large

construction projects, and, importantly, be able to engage the public in a way that engenders trust. Finally, the entity should operate effectively over the many decades that will be required to manage the present inventory of nuclear waste, as well as waste generated by future advanced reactors.

• Congress should ensure that funds collected from ratepayers that use electricity from nuclear power plants, now over $45 billion, are applied to the disposal of the spent fuel generated by nuclear power plants and that collection of funds from all commercial generators of nuclear power resumes. Moreover, funding for the entity should be held in a true escrow account and not be subject to the annual appropriations process.
• The entity should immediately initiate steps to begin the process of site selection. Before sites are considered, a decision-making process with appropriate technical criteria and an acceptable method of public engagement, such as consent-based siting, needs to be defined in collaboration with impacted communities, tribes, and states. Congress should make a decision on what to do with Yucca Mountain, which could include keeping it as a possible site for consideration, depending on the plans of the new entity.

Finding 12: The advanced reactor developers’ presentations to the committee focused on the reactors themselves, with little or no attention to nuclear waste management or disposal of the nuclear waste generated because there is no incentive for them to do so. In the absence of a final geologic disposal strategy in the United States, the expansion of nuclear power using advanced reactors will add to the amount of spent nuclear fuel and associated waste that requires disposal and increase the complexity of this challenge because of the need to dispose of new types of fuels and waste streams.

Finding 13: Presently proposed advanced reactor technologies will initially use a once-through fuel cycle; however, compared with those currently in use, the fuels will have a higher uranium enrichment (e.g., high-assay low-enriched uranium [HALEU]) and a higher burnup; also, they will use new types of fuel materials and designs (e.g., TRistructural ISOtropic [TRISO] fuels). As compared with the disposal of the present uranium oxide spent fuel, these new fuel types may result in changes of (1) the amounts (either in mass or volume), chemical compositions, and radionuclide inventories of the waste to be disposed; (2) the thermal power of fuel assemblies; and (3) the durability of the spent fuel in a disposal environment. More specifically, from the waste management and disposal perspective, it is important to note the following:

• Radiological risks from disposed waste are dominated by the mobility of long-lived radionuclides and not by the radiotoxicity inventory. Therefore, radiotoxicity itself is a poor metric for repository performance and risk to the public from waste disposal. The long-term safety of disposal of actinides in appropriate geologic settings is largely independent of the actinide inventory of the repository, except in the off-normal situation where the geologic barrier is bypassed—for instance, by human intrusion. Because the amount of mobile long-lived fission products generated is independent of reactor type, most advanced reactor technologies will have little impact on estimates of long-term repository performance. Key factors for long-term repository performance are the redox conditions of the geochemical environment, waste form stability, groundwater flow rates, and solubility/sorption of radionuclides. A reducing environment is preferred. Advanced reactor technologies will have little or no impact on these factors.
• The total quantities of fission products generated are generally related to fission rate and are largely independent of reactor technologies, although the distributions of different isotopes may differ. Both short- and long-lived fission products are important on the timescales relevant to geologic disposal. Short-lived fission products (e.g., strontium-90 and cesium-137) produce significant heat, while long-lived fission products (e.g., iodine-129 and technetium-99) are extremely mobile in a repository environment. Advanced reactor technologies will, in general, generate a higher amount of fission products in each spent nuclear fuel package because of their higher burnups, resulting in a higher thermal load. Increased thermal loads of waste containers will impact a number of repository design features, such as the size and spacing of waste packages, the size of the repository footprint, and engineering designs, thereby impacting the cost of repository construction.

• Enhanced stability and durability of waste forms in a repository environment can be beneficial to the performance of a repository by limiting the release of radionuclides from the spent fuel. Some advanced reactor technologies propose using advanced fuel designs with the potential to contain radionuclides (e.g., TRISO fuel), but this potential must first be demonstrated by experimental programs that examine the fuel’s long-term integrity in intense radiation fields and at high temperatures.

Finding 14: Conceptually, advanced reactors could be used to reduce the current inventory of transuranics in the approximately 86,000 tonnes of legacy spent fuel to date; this would require considerable resources and time to design, develop, prototype, build, and make operational the required infrastructure. Creating this infrastructure is not practicable in the near future, as long as uranium and enrichment services are readily available.

Finding 15: Most of the advanced reactor types proposed would generate waste streams for which there is little experience or mature technical ability to manage. All additional waste treatment options would entail additional costs not encountered in the management and disposal of spent light water reactor (LWR) fuel. High-temperature gas reactors will produce much larger volumes of spent fuel compared with equivalent energy production from LWRs. It may be possible to reduce the volume by removing graphite from the spent fuel, but those technologies are immature. Dust production from pebble-bed reactors would pose waste and decommissioning challenges. Sodium-cooled fast reactors would produce large volumes of irradiated sodium waste that would require treatment and disposal; sodium-bonded spent fuel is not suitable for direct disposal and would require treatment by methods not yet technically mature at the industrial scale. Molten salt reactors produce two waste streams, radioactive off-gases and the spent fuel salt waste, which would require processing into waste forms suitable for disposal. These treatment methods and suitable wastes forms are in early stages of exploration. Most of these advanced reactors would produce large quantities of irradiated graphite waste—from use as moderators or reflectors—and this material would prove challenging to manage as well. While European researchers have analyzed graphite waste disposal extensively, researchers in the United States generally lack this expertise.

Finding 16: Similar to issues with waste management, advanced reactor developers have not adequately examined the back-end operational management (i.e., storage and transportation) of advanced nuclear spent fuel. Consequently, the stability of waste forms and potential issues related to needed processing prior to storage, as well as repackaging that may be required for transportation and final disposal, have not been studied sufficiently.

Finding 17: Secondary waste streams—such as lead, sodium, molten salts, and irradiated graphite (moderators and/or from TRistructural ISOtropic [TRISO]–particle fuel disassembly)—from advanced reactor and fuel cycle operations will need to be stabilized and packaged for storage prior to downstream operations to support disposal. Waste forms for these secondary wastes can be developed to be compatible with storage regulations by the U.S. Nuclear Regulatory Commission; however, some still require research and development to properly characterize performance envelopes.

Finding 18: Because of the higher enrichments of fresh high-assay low-enriched uranium (HALEU) and potential higher burnups of irradiated HALEU fuels, maintaining subcriticality margins and having adequate thermal and shielding protection during transport and storage would most likely require at least some of the following:

• criticality experiments for enrichments above 5 percent to support benchmarking analyses;
• assessment of the feasibility of using type 30B containers for transport of enriched uranium hexafluoride, if needed; and
• criticality, thermal, and shielding assessments for storage and transportation.

5.2 THE U.S. NUCLEAR WASTE MANAGEMENT AND DISPOSAL PROGRAM

Though the United States has reached the 40th anniversary of the Nuclear Waste Policy Act (NWPA) of 1982, there remains no clear path forward for the siting, licensing, and construction of a geologic repository for the disposal of highly radioactive waste, mainly the spent nuclear fuel from commercial nuclear power plants and the high-level waste generated by chemical reprocessing associated with defense programs. Instead, as of 2021, the United States has accumulated about 86,000 MTHM (metric tons of heavy metal) of spent nuclear fuel generated by commercial nuclear power plants and hundreds of thousands of cubic meters of liquid/sludge waste at the Hanford and Savannah River sites where irradiated fuel was reprocessed to obtain plutonium for nuclear weapons (GAO, 2021a,b). The inventory of spent fuel from the 93 operating commercial reactors in the United States continues to increase at the steady rate of approximately 2,000 MTHM/year.

The distribution of both commercial and defense waste is shown in Figure 5.1(a). On the defense side, most of the wastes are located at facilities originally associated with the Manhattan Project and the Cold War production of fissile material for nuclear weapons. These facilities and laboratories are noted on the map, but many contaminated

areas related to defense activities are not included, as they generally have lower-activity waste, such as at the Mound and Fernald plants in Ohio, the Portsmouth uranium enrichment plant in Ohio, and the Paducah uranium enrichment plant in Kentucky. Figure 5.1(a) does show the Hanford and Savannah River sites, as well as Idaho National Laboratory, which house hundreds of large underground tanks of radioactive waste from reprocessing of spent fuel. Despite the extent of nuclear contamination related to defense sites (DOE-EM, 1997), the most important problem in terms of total radioactivity is the 86,000 MTHM of commercially generated spent nuclear fuel stored on the sites of nuclear power plants at 75 locations in 33 states, as depicted in Figure 5.1(b) (GAO, 2021a). A 2013 analysis concluded that all light water reactor (LWR) spent nuclear fuel generated to date can be disposed of, as it would not be needed as a source of plutonium for potential future fast reactors (Worrall, 2013). Clearly, the greatest source of radioactivity is in the spent nuclear fuel from commercial nuclear power plants (see Table 5.1).

For high-level nuclear waste generated by weapons programs, the plan remains to vitrify some of the high-level waste prior to geologic disposal. Throughout the 23-year history of the Savannah River Site, some 4,100 containers of vitrified waste have been generated at the Defense Waste Processing Facility; the “vitrified logs” remain on site, as there is no geologic repository to receive them (SRS, 2020). At the Hanford site, the vitrification plant remains under construction as the schedule slips by decades and the cost escalates by tens of billions of dollars (GAO, 2020), as further discussed in Section 5.2.3. At both the Savannah River and Hanford sites, the most recent strategy involves chemical processing of the highly radioactive waste into low and highly radioactive waste streams. The low-level activity waste streams are being or will be disposed of in near-surface disposal sites. The inventory of U.S. nuclear wastes is summarized in Table 5.1.

5.2.1 Waste Classification

Any advanced reactor or advanced nuclear fuel cycle will generate nuclear waste at each step of the cycle: uranium mining and milling, uranium enrichment, fuel fabrication, reactor operation, chemical processing, reprocessing and recycled fuel fabrication, and finally decontamination and decommissioning. In the United States, highly radioactive wastes, such as spent nuclear fuel and high-level waste, are classified based on their origin. Lower-activity wastes also are indirectly source-based, as they are defined by excluding the source-based classes of fuel (BRC, 2012). At times, this chapter uses the international category of “intermediate-level waste” (ILW), which is closest to the U.S. category of Greater-than-Class C (GTCC) waste, because the available literature used such terminology. Appendix D provides more information on U.S. waste classifications, including a discussion of ILW and GTCC.

The classification of the waste determines the type of disposal environment. Spent nuclear fuel, solidified high-level waste, and transuranic waste from defense programs must be disposed of in a geologic repository. Lower-activity waste can generally be stored or disposed of in near-surface facilities. The U.S. Nuclear Regulatory Commission (U.S. NRC) has been evaluating disposal options for GTCC waste, which contains higher concentrations of radioactivity than Class C waste and may require disposal in a deep geologic repository (DOE, 2017). In 2020, the U.S. NRC released an options memo for the commissioners that supported the conclusion that most GTCC waste streams could be suitable for near-surface disposal, pending site-specific analyses, though no rulemaking has been made as of yet (U.S. NRC, 2020h).² More information on disposal of different waste classes can be found in Appendix D. In order to estimate the cost, determine environmental impact, and complete a safety assessment of any advanced fuel cycle, it is necessary to know the type, amounts, and heat production of waste that will be generated.

TABLE 5.1 Nuclear Waste in the United States as of 2020

^a Radioactivity values provided in Ci as well as the international standard Bq, 1 Ci = 3.7 × 10¹⁰ Bq.

^b From DOE (2016).

NOTE: LLW = low-level waste; MCi = megacurie; MT = metric ton; TBq = terabecquerel; TRU = transuranic elements.

SOURCE: DOE-EM (1997), with updated calculation of total MCi for spent nuclear fuel, unless otherwise noted.

___________________

² This sentence was altered after release of a prepublication version of the report to align the text with information from the associated reference.

5.2.2 Interim to Indefinite Storage

In the absence of a geologic repository, the de facto U.S. strategy for dealing with commercially generated spent nuclear fuel is the possibility of consolidated, interim to indefinite storage. This strategy accommodates the present situation, but it is not a strategy by design. In fact, legislation will be required in order to eliminate the limitations imposed by the Nuclear Waste Policy Act of 1982, as amended in 1987, that prevents the interim storage of spent nuclear fuel prior to the opening of a geologic repository, which at that time had been designated to be at Yucca Mountain (DOE-RW, 2008).

Additionally, in 2014, the U.S. NRC replaced its revised “waste confidence rule” of 2010 with a “continued storage” rule for spent nuclear fuel. The waste confidence rule had previously been used as a statement that the U.S. NRC has “reasonable assurance that sufficient mined geologic repository capacity will be available to dispose of the commercial high level radioactive waste and spent fuel generated in any reactor when necessary” (U.S. NRC, 2010). However, in 2012, the U.S. NRC had to institute a moratorium on the issuance of licenses for new reactors and renewals of existing licenses because a federal court in the District of Columbia Circuit determined that, in the absence of progress in developing a national geologic repository, there was little basis for confidence in the federal government’s ability to permanently dispose of spent nuclear fuel. The U.S. NRC then developed a new continued storage rule concluding that spent nuclear fuel could be stored indefinitely, as long as institutional controls that ensure the safety of the waste remain in place indefinitely (U.S. NRC, 2014a). In the absence of a geologic repository program, the United States has turned to the possibility of extended storage of spent nuclear fuel. Two commercial sites in New Mexico and Texas (see Figure 5.1(b)) are under licensing consideration by the U.S. NRC (2020j). However, both states’ governors have already registered their objections to this proposal. At the direction of Congress, on December 1, 2021, the U.S. Department of Energy (DOE) initiated a request for information on the use of consent-based siting of a consolidated storage facility for spent nuclear fuel (DOE-NE, 2021h).

5.2.2.1 Evolution of Spent Fuel Over Time

Notably, spent nuclear fuel is not a passive material, even as it might be stored at centrally located facilities over long periods; rather, its composition and properties evolve over time because of radioactive decay (Peterson and Wagner, 2014). The thermal output and the type and intensity of the radiation field change significantly over time (Hedin, 1997). As an example, because of changes in isotopic composition, the activity level (i.e., radioactive decay rate) of the fuel increases for a short time after discharge (from decay of short-lived fission product neutron absorbers [e.g., xenon-135]) and then decreases out to ~100 years. There is a second peak in reactivity at about 30,000 years caused by radioactive decay of actinide neutron absorbers (e.g., americium-241 and plutonium-240). Hence, the activity or decay rate of the fuel is important over periods associated with interim storage, as well as those for geologic disposal.

The high level of radioactivity of spent fuel is important to the concept of the fuel being “self-protecting”—that is, too radioactive to handle by a bad actor intent on processing the fuel to reclaim fissile material. The high level of radioactivity is a barrier to handling the fuel because radiation shielding would be required. However, the decrease in the level of radioactivity affects this “self-protecting” strategy. The self-protection is significantly decreased between 70 and 120 years after discharge from the reactor. Because many spent fuel assemblies have already been in storage for decades, this self-protection could be lost for some fuel assemblies in as little as 30 years (Peterson and Wagner, 2014).

Finally, the evolution of the fuel composition, temperature, and radiation fields over time (the latter of which is depicted in Figure 5.2) will have a huge effect on the strategy for geologic disposal. In general, cooler fuel will be more easily handled in a geologic repository; however, the evolving radiation field (e.g., the persistence of alpha decay) may lead to radiolytic decomposition of water and accelerate the corrosion rate of the uranium dioxide in spent nuclear fuel. The complex evolution of spent fuel in a disposal environment is described by Ewing (2015) (see Figure 5.3). Hence, the changes in the properties of spent fuel have an impact on both interim storage and geologic disposal strategies.

While Figures 5.2 and 5.3 illustrate these changes for spent uranium dioxide fuel, as one envisions the behavior of new types of fuels in the disposal environment, the same processes must be considered. New fuel types, such as TRistructural ISOtropic (TRISO) fuel, may increase the complexity of the analysis but also may show better performance, such as containment of fission products, because of the silicon carbide layer surrounding the TRISO kernels.

Although other countries also have struggled with geologic disposal, and almost all have had to stop and reorient their approach, some are making good progress, including Sweden, Finland, France, Canada, and Switzerland. In Sweden, Finland, and France, which have selected repository sites, implementers have moved forward in their applications for approval from regulatory authorities. In Finland, the geologic repository is already under construction, with the expectation of receiving the first spent fuel for disposal around 2025. In Sweden, as this report was being written, the government approved the construction of a repository near Östhammar. In a review of international experience in site selection and the development of a mined, geologic repository, the Nuclear Waste Technical Review Board in 2015 analyzed the history of 24 efforts to site geologic repositories in 10 countries (NWTRB, 2015). Of these national programs, only 6 have remained on track. Most importantly, all programs have experienced a point at which it was necessary to completely reorient their program and essentially begin again. The waste management and disposal program of the United States is at such a point of reckoning.

5.2.3 Costs of Waste Management

For some 20 years, the United States investigated the Yucca Mountain site as a potential geologic repository. As of 2009, this effort had cost $15 billion (GAO, 2011), and in 2008, DOE estimated the total system life-cycle cost of the Yucca Mountain project to be $96.2 billion (2007$) (DOE, 2008). In addition, the cost of stranded waste and unfinished facilities has been substantial. Two notable examples are the mixed oxide fuel fabrication facility at the Savannah River Site and the vitrification plant at Hanford. Any plan to reprocess spent fuel from advanced reactors is likely to entail (1) the fabrication of nuclear fuels that are a mixture of uranium and plutonium and (2) the solidification of high-level waste from reprocessing prior to disposal. In 2000, a contract was awarded for the Hanford Waste Treatment Plant with a projected cost of $4.3 billion and a startup in 2007. In 2006, the cost estimate was revised to $12.3 billion, and the startup was delayed to 2019; it has been further delayed to the 2030s. In 2020, the U.S. Government Accountability Office (GAO) documented this history and noted DOE’s “2019 Hanford Lifecycle Scope, Schedule and Cost Report that completing the WTP [waste treatment plant] would cost between $19 billion and $30 billion, in addition to the more than $11 billion already spent” (GAO, 2020).

In 2007, at the Savannah River Site, work began on the mixed oxide fuel fabrication facility for the disposition of 34 MT (metric tons) of plutonium from dismantled nuclear weapons, with an initial cost estimate of $4.9 billion. In 2012, GAO estimated that the cost had risen to $7.7 billion with the earliest start-up date being 2019 (GAO, 2013b). Estimates of the life-cycle cost reached as high as $30 billion (Holt and Nikitin, 2017). Despite early interest in using the mixed oxide fuel, no utility finally agreed to use it. The construction of the mixed oxide fuel fabrication facility ended in 2018 even though major portions had already been completed (GAO, 2019). To the extent that advanced reactors are heralded as a first step toward a closed fuel cycle in the United States, one needs to carefully consider the cost and experience with the construction of facilities that will be required to support the advanced fuel cycles proposed.

Finally, because of the failure of DOE to take ownership in January 1998 of the spent nuclear fuel generated and stored at reactor sites across the nation, utilities have successfully sued DOE and collected several hundred million dollars annually from the Judgment Fund (i.e., the U.S. taxpayer) for the costs related to continued storage of spent nuclear fuel at reactor sites. Through fiscal year (FY) 2020, payments to utilities from this fund totaled $8.6 billion. DOE has estimated “that its potential liabilities for waste program delays could total as much as $39.2 billion” (Holt, 2021a).

These examples illustrate the high costs and long delays that result from failing to plan for all the necessary facilities to support the back end of the nuclear fuel cycle—whether open or closed.

5.2.4 U.S. Nuclear Waste Management Today

At present, the back end of the U.S. nuclear fuel cycle is a jumble of different facilities and activities that have resulted from poor planning, continued programmatic delays, and the failure to create a geologic repository.

How Did This Happen?

Broadly speaking, the causes were (1) changes to the original NWPA of 1982; (2) a slowly developing and changing regulatory framework; (3) erratic funding; (4) consequential changes in policy with changing administrations; (5) conflicting congressional and federal agency policies; and (6) insufficient public engagement in decisions about fundamental strategies for the storage and disposal of highly radioactive nuclear waste (Reset, 2018).

More specifically, the grand covenant hammered out by the interagency process prior to and incorporated into the NWPA of 1982 has unraveled. In particular,

• sites for a repository were to have been chosen by a technically driven evaluation of at least three candidate sites—the result of a down selection process from a greater number of sites;
• in order to promote geographic equity, two repositories were to have been developed;
• a fee levied on the use of electricity from nuclear power plants was to have funded the development of a geologic repository so that the federal government could take ownership of the spent fuel by January 31, 1998; and
• states were to have been given a meaningful role in the approval of the selected site.

In contrast to expectations, Congress designated the Yucca Mountain site as the only site for characterization in the Nuclear Policy Act Amendments of 1987; the goal of a second repository, presumably in the eastern United States, was not pursued; the Nuclear Waste Fund ratepayer money has grown to more than $40 billion, but has not been generally available for the development of a geologic repository; states have had a limited role in site selection; and most notably, the state of Nevada has waged a tenacious battle against the Yucca Mountain site as the nation’s geologic repository for more than 35 years. Finally, the Obama administration eliminated the Office of Civilian Radioactive Waste Management, and Congress has not funded the Yucca Mountain project since 2010. Presently, the United States continues to have no strategy for the management and disposal of spent fuel generated by commercial nuclear power plants during the past or into the future.

In addition to this broad backdrop of major challenges, the United States also faces important limitations in its ability to address nuclear issues. As an example, the Idaho National Laboratory (INL) is the lead laboratory for nuclear energy research. However, in 1995 the state of Idaho reached an agreement with DOE that all spent fuel on site would be removed by 2035 (IDEQ, 1995). This agreement has interfered with research that requires that spent fuel be brought to INL for research purposes. In the absence of a geologic repository, the prospects of moving spent fuel out of Idaho by 2035 dim with each passing year.

Another major issue is one of human capital and research support required for researchers to develop appropriate skills related to all aspects of nuclear fuel. More specifically, the Yucca Mountain project has been on hold or dead for more than a decade, and during that time, many researchers have moved to other subjects or retired. The loss of this knowledge base will certainly delay the success of programs that support the back end of the nuclear fuel cycle.

Why is this story important for the potential development of alternative fuel cycles? As the United States looks forward to any expansion of nuclear power, there should be a plan for the disposal of the spent nuclear fuel and other radioactive wastes that result from the use of advanced reactors and small modular reactors. The 40-year history of delay, rising costs, and finally failure to deal with highly radioactive waste is a legacy that this generation should not pass on to future generations.

5.2.4.1 The Way Forward

During the past decade, a number of groups have reviewed the U.S. situation and made recommendations that, if followed, would provide a basis for a new U.S. program for nuclear waste management and disposal. The most prominent effort was undertaken by the Blue Ribbon Commission on America’s Nuclear Future (BRC, 2012).

Their final report was submitted to Secretary of Energy Steven Chu in January 2012, and its recommendations focused on nuclear waste management issues. Relevant, principal recommendations included

• The urgent need for one or more geologic repositories;
• The formation of a single-purpose organization to implement an integrated program of transportation, storage, and disposal of nuclear waste;
• assured and continued access to the Nuclear Waste Fund, which now has a balance of more than $40 billion; and
• the use of an adaptive, staged, transparent, and particularly consent-based process in the siting of a geologic repository.

In 2018, a 3-year study, Reset of America’s Nuclear Waste Management Strategy and Policy, was completed at Stanford University and The George Washington University (Reset, 2018). The study, guided by a steering committee of nuclear waste experts, sought to identify the systemic issues that have prevented the success of the U.S. program. Key issues identified include

• the need for a new, independent, single-purpose, not-for-profit national radioactive waste management and disposal organization, and reform of the funding process;
• integration of the back end of the nuclear fuel cycle, such that all decisions are focused on the common, primary goal of geologic disposal of nuclear waste;
• a new approach to public engagement that develops a consent-based siting process based on trust in the implementer, redistribution of power among affected parties, and defined roles in the decision-making process; and
• a reexamination of the regulatory framework with a focus on a safety-case approach using both quantitative and qualitative criteria.

The linchpin of both sets of recommendations is the creation of a new waste management organization. The Blue Ribbon Commission recommended a federal corporation similar to the Tennessee Valley Authority, while the Reset committee pointed to the unique advantages of a not-for-profit, utility-owned waste management organization. A utility-owned organization would provide a “cradle-to-grave” approach from the moment the spent fuel is removed from a reactor until it is disposed of in a geologic repository.

Importantly, the recommendation for a new organization is not new. Willrich and Lester in 1977 first recommended the establishment of a “Radioactive Waste Authority,” as they had concluded “the existing organization for radioactive waste management is likely to be unworkable if left unchanged” (Willrich and Lester, 1977). And it was. The successful nuclear waste management and disposal programs—those in Finland, Sweden, Canada, and Switzerland, in particular—are managed by a utility-owned organization, not a government agency. A new organization, as the implementer of a waste management and disposal strategy, would have to be a top priority in any plan to expand nuclear power in the United States.

The establishment of a new organization would also require careful attention to a new funding approach so that funds could be used in a timely manner over the many decades required to select a site and design, license, and construct a new geologic repository. In the case of a utility-owned organization, this would require the transfer of the Nuclear Waste Fund, over an extended time, to that organization.

In addition to creating a new implementing organization, the funding of the implementer must be fixed, the regulatory framework reexamined, and a robust approach to siting that engages and addresses the concerns of the public must be followed to ensure success in repository siting. Creating a new organization and funding scheme, and developing the structure and values of a new organization would take time, but the basic roadmap has been available for a decade. In fact, DOE is on record as having accepted all of the recommendations of the Blue Ribbon Commission (DOE, 2013). Hence, there is a way forward, but this will require congressional action.

5.3 THE CONCEPT OF GEOLOGIC DISPOSAL OF HIGHLY RADIOACTIVE NUCLEAR WASTE

Some advocates for advanced reactors and small modular reactors claim that one benefit of these technologies would be to ease the burden on nuclear waste disposal strategies. A key question for this committee is whether these reactors could be a “silver bullet” for what has been a vexing, national problem. In order to evaluate the veracity of such claims, key aspects of the concept of geologic disposal³ are first summarized.

5.3.1 The Concept of Geologic Disposal

Many variations in the strategies for geologic disposal become evident in the comparison of repositories in different geologic settings; however, all approaches share the common strategy of using multiple barriers in order to provide defense in depth. In the face of the uncertainty in the extrapolated behavior of a repository over long periods, a system of multiple barriers, also known as the “Russian-doll” or “belt-and-suspenders” approach, provides a robust basis for confidence in the ability of a geologic repository to contain or reduce the mobility of key radionuclides, thereby lowering radionuclide access to the environment and subsequent exposure to humans (Ewing and Park, 2021; Hedin, 1997). Descriptions of the two types of barriers follow.

Engineered barriers include the spent fuel or nuclear waste form, the waste package, backfill and/or overpack, and structures to prevent the ingress of water to the waste package and/or the egress of radionuclides from corroded canisters. Most engineered barriers are physical, but in some cases, they are chemical additions designed to reduce radionuclide concentrations, such as the magnesium oxide emplaced with the transuranic waste in the disposal panels at the Waste Isolation Pilot Plant, a geologic repository for defense transuranic waste located in southeastern New Mexico.

Geologic barriers rely on the properties of the rock and the hydrologic system of the repository. The petrophysical, geochemical, and hydrologic properties of the rock into which a repository is constructed can have major effects on the mobility and concentrations of radionuclides. Judgment of whether a site is acceptable depends on whether the geologic setting and hydrologic system meet the regulatory requirements of the safety case.

As simple as this approach appears to be, its application is complicated when developing the safety assessment, which can have hundreds of subsystem models and many hundreds of input parameters. Most importantly, the subsystems that comprise individual barriers are highly coupled and nonlinear. Larger-scale coupling leads to the development of codes to capture the thermal-mechanical-chemical-hydrologic regimes, while models of near-field interactions capture the long-term geochemical evolution of waste form–waste package–backfill materials using reactive transport modeling (Arcos et al., 2008). As an example, in the KBS-3 waste disposal concept developed in Sweden, experiments and modeled data show the impact of accessory minerals and clay surfaces on pore water chemistry. The pore water chemistry affects the long-term stability of the bentonite barrier.

Thus, any aspect of the analysis that can be simplified by, say, a very durable waste form or very long-lived waste package can make for a simpler and more compelling safety case. As an example, if an advanced reactor uses fuel that is unusually durable in the disposal environment, this could represent an important contribution to safety; however, the variation in hydrogeochemical conditions present in a repository will necessarily cause differences in the expected behavior of the fuel due to coupled chemical reactions in an evolving thermal and radiation field.

5.3.2 Different Strategies for Geologic Disposal, or Not Every Hole in the Ground Is the Same

During the past 30 years, considerable research has been conducted on scientific and engineering issues that support geologic disposal. Importantly, different countries have pursued the development of geologic repositories in different geologic settings that can be described most simply in terms of their rock type (Ewing and Park, 2021): salt, clay, crystalline rock, and volcanic tuff. Even a cursory examination of the different national programs in a variety of geologic settings shows that the fundamental safety strategies vary as they are adapted to the specific waste types and geologic settings. In some instances, the engineered barriers are most important (e.g., corrosion

___________________

³ For a tutorial on the concept of geologic disposal, see Piet Zuidema’s presentation to the committee (Zuidema, 2021), particularly the figures on slides 28, 35, 36, and 46.

resistant copper canisters), and in other cases, the geologic barrier plays the critical role (e.g., plastic deformation and sealing by salt or clay). Ideally, all of the barriers will play an important role in containing radionuclides, generally over different time frames as the composition of the waste and the conditions in the repository evolve with time (Ewing, 1995).

This report does not provide a detailed review of the different approaches for geologic disposal; rather, the committee refers the reader to Yardley et al. (2016) and NWTRB (2015) for summaries of the approaches taken for each of the different geologic settings. Importantly, major differences in basic strategies result from the following:

• The thermal pulse from short-lived fission products (e.g., aging the waste prior to disposal versus the use of smaller waste packages). The longer-term thermal pulse from actinides is of less importance to repository performance. However, the alpha decay of actinides is important because of the effect of the radiolysis of water at the surface of the fuel and the potential for changing the redox conditions from reducing to oxidizing.
• The release of long-lived, highly mobile fission and activation product elements (e.g., by relying on containment in packages of exceptional durability or dilution during release and transport).
• The release of long-lived actinides that make up the bulk of the chemistry for spent fuel disposal (e.g., by disposal in a reducing environment versus an oxidizing environment). The redox state of the repository can profoundly affect the mobility of the multivalent actinides.

These three examples, listed in order of increasing half-life, highlight the major challenges associated with the composition of the waste. For the short-lived radionuclides, aging, ventilation on the emplacement, or changes in waste package size enable effective engineering solutions. For long-lived, highly mobile fission and activation products, the challenge is greater because a number of these elements travel in groundwater as negatively charged species; hence, they are not sorbed onto mineral surfaces, which are also negatively charged. Possible solutions to these problems can be as simple as dilution, particularly isotopic dilution of iodine, or incorporation into very durable waste forms, such as iodine into apatite (Ewing and Wang, 2002). For long-lived actinides, with half-lives on the scale of many hundreds of million years, ensuring geochemical (e.g., reducing conditions) and hydrologic conditions (e.g., transport by diffusion) that drastically reduce their mobility in the repository environment is important. As an example, disposal of actinides in the reducing conditions of an organic clay can essentially limit the mobility of actinides to within some tens of meters from the point of emplacement for a million years (Grambow, 2008).

Recently, there has been increased interest in the potential for deep borehole disposal for specific waste types, including spent nuclear fuel and high-level waste generated by advanced reactors. Present efforts in the United States are supported by two private companies, Deep Isolation, Inc., and NuclearSAFE (Deep Isolation, 2022; NuclearSAFE, 2018). Both companies propose to use deep boreholes combined with directional drilling for horizontal emplacement of nuclear waste–containing packages. The deep borehole disposal concept was first considered in the 1970s, but the concept was not pursued because drilling technology was not sufficiently advanced at the time. Recent renewed interest in the concept reflects the rapid advancement of drilling technologies from the oil and gas industry. Deep borehole disposal concepts rely on the isolation capacity of the geosphere and the deep hydrologic environment. A recent study by the Electric Power Research Institute (EPRI) (2020) on the feasibility of deep boreholes collocated with advanced reactors examined key elements such as site characterization, regulatory considerations, spent fuel and waste package characteristics, and operations and safety analysis; this report did not identify any technical showstoppers.⁴ Another study (Krall et al., 2020) examined and identified technological challenges of siting deep boreholes. Since the details of potential waste characteristics of many advanced reactors are not yet available, the committee decided not to examine the concept of deep borehole disposal at this point. But as advanced reactors progress and mature, an assessment of the deep borehole concept may be warranted.⁵

___________________

⁴ The committee notes that Deep Isolation, Inc., was a contractor for the EPRI report.

⁵ One committee member disagreed with this statement, given the technical challenges that have been identified for deep borehole disposal in independent analyses such as Krall et al. (2020) and references therein.

5.3.3 Characteristics of a Geologic Repository

Although the basic physics of all fission-based nuclear power plants is the same, a wide variety of designs is possible, depending on the type and enrichment of the fuel, the composition of the moderator, and the type of coolant, to name a few salient features. Thus, advanced reactor designs reveal a broad array of strategies for producing energy from fission reactions.

In much the same way, the basic physiochemical processes in the natural and engineered environments of a geologic repository are similar, but the basic strategies for containment of radionuclides may differ in important ways, depending on the type of fuel, level of burnup, type and degree of reliance on engineered barriers, and properties of the geologic setting. This brief section characterizes some of the major differences in repository strategies. The goal of geologic disposal should be to match the different types of waste generated by different reactor types with the most appropriate disposal strategy.

The simplest characterization of the geologic setting is usually given by noting the rock type: salt (e.g., dome or bedded formation), crystalline rock (e.g., igneous granite or metamorphic gneiss), clay and shale, and volcanic tuff. However, categorization by rock type fails to capture important differences in geologic settings. As an example, salt domes have very low water content (only fluid inclusions); bedded salt deposits have a greater abundance of fluid inclusions and the possibility of large, pressurized brine pockets; shales contain water in the atomic structure of clays, and water flow is hindered so much that in “tight” shales, radionuclides migrate mainly by diffusion; crystalline rock contains water in fracture systems that move radionuclides by advective flow; and in volcanic tuff, substantial amounts of water are typically held within the relatively high porosity (10 to 15 percent), and this water is mobilized at elevated temperatures. Hence, the properties specific to each geologic setting will drive the development of different waste disposal strategies. These might include upper limits on the heat load to reduce chemical reaction rates, preserve the chemical properties of clay rock or bentonite backfill, or ensure redox conditions that chemically stabilize key radionuclides (mainly the actinides). Reducing redox conditions are generally considered important; hence, most repositories are designed to maintain reducing conditions. Importantly, for uranium, reducing conditions lower its solubility by nearly four orders of magnitude (Shoesmith, 2000).

Thus, the final strategies will be driven by not only the type of waste disposed of but also the petrophysical, petrochemical, and hydrologic properties of the specific geologic setting (see Table 5.2). Because in some countries it is difficult to change the geologic setting, there will be greater emphasis on the engineered barriers. Even in strategies that rely mainly on engineered barriers, geologic barriers remain an important aspect of the safety case.

The strategy also depends on the regulatory framework. The goal may be complete containment within a short distance of the repository, or the point of compliance may be tens of kilometers away from the repository. In the latter case, sorption and dilution are important factors in the demonstration of compliance, even though they are well beyond the repository itself. Thus, it is not always appropriate to compare the safety assessment of one repository type with another, as the determination of safety may be based on very different compliance distances and time frames.

Finally, although strategies may vary, initially most of the radionuclides are contained within the spent fuel or, in the case of reprocessing, the waste form selected for reprocessed waste. Hence, the durability of the spent fuel or nuclear waste form is the first and last barrier to radionuclide release. This means that any selected strategy may focus attention on the properties of the fuel. New types of fuel will require research programs that substantiate their compatibility and durability in the expected disposal environment. Such research programs require considerable attention to a wide variety of processes, including redox conditions, radiolysis of water, solubility limits, and colloid formation, to name a few (Ewing, 2015).

5.3.4 Grand Challenges of Geologic Disposal

In the discussion of advanced reactors and advanced fuel cycles, there are often claims that the new reactors, perhaps combined with chemical processing, will improve the ability of the United States to manage and dispose of its nuclear waste. In order to evaluate such claims, it is necessary to understand both the progress that has been made and the challenges remaining for nuclear waste management and disposal of highly radioactive waste.

The grand challenges are as follows:

1. The investigation of multiple sites and the final selection of a single site for a geologic repository. There are multiple approaches to repository siting, but successful siting depends on finding sites that are both technically and societally acceptable. On the technical side, a major issue has been establishing appropriate criteria against which sites can be evaluated and compared. The criteria are of three types: (a) exclusionary, (b) host rock specific, or (c) generic (Ewing and Park, 2021; NWTRB, 2015). Arriving at an early consensus on the appropriate criteria greatly simplifies the evaluation and comparison of sites. However, coming to agreement on these criteria, particularly when comparing different geologic settings, remains difficult.
2. The social process of siting a repository. In many countries, the institutions responsible for developing a geologic repository have appreciated that the process of siting a geologic repository is not only a technical process, but equally a social process that requires extensive and prolonged public engagement and empowerment (Metlay, 2016, 2021). The United States began to accept this understanding with the report of the Blue Ribbon Commission, which emphasized the need for consent-based siting processes, but the country is still far from ready to implement such a process.
3. The chemical and radiological complexity of the nuclear waste streams. The fission process creates the following types of waste radionuclides: (a) short-lived fission products (e.g., iodine-131, cesium-137, and strontium-90), which generate intense radiation fields and considerable heat; (b) long-lived fission products (e.g., iodine-129, technetium-99, selenium-79); (c) long-lived actinides (e.g., uranium-235, uranium-238, plutonium-239); (d) minor actinides (e.g., isotopes of americium and curium); and (e) activation products (e.g., cobalt-60, nickel-63). Each of these radionuclides has its own chemistry and hence different mobilities in the natural environment.

   Site selection can play an important role in determining the mobility of radionuclides, as the geologic setting and related fluid compositions and hydrologic flow determine radionuclide concentrations in solution and the distances over which they may be transported. In addition, the thermal output and radiation fields (type and strength) will change with radioactive decay (Hedin, 1997). Hence, the composition of the waste will evolve because of radioactive decay, and the changing conditions induced by the changing radiation and thermal fields will impact conditions within the repository. Also, the operational parameters of nuclear power plants or chemical processing facilities and related off-gassing will generate considerable low-level waste, including GTCC waste, some of which will require geologic disposal or near-surface disposal, depending on the radionuclides contained. In the broadest sense, advanced reactors and their associated fuel cycles (e.g., reprocessing versus no reprocessing) may impact the compositional complexity of the highly radioactive waste streams.

4. Modeling behavior of radionuclides in a geologic repository over extended periods that stretch to hundreds of thousands of years and in some cases to a few million years. From a technical perspective, this is a “wicked” problem further complicated by public fear of radiation and the need to make credible and compelling extrapolations of long-term behavior.
5. The proper use of models for the extrapolated behavior of the repository. A safety assessment can include models ranging from atomic-scale behavior of corroding spent fuel to crustal scale models of tectonics and hydrology, extending over many tens of square kilometers. At every step of a safety assessment, there are models of physical, chemical, and biological processes, as well as human behavior that affect exposure rates (Swift and Sassani, 2019). The role of modeled performance and its relation to safety assessment remains much debated.
6. Safety considerations. This entails developing a clear understanding of the meaning of safety in the context of calculated risk and regulations. The issue here is that there are many metrics for the evaluation of the safety of a repository. Typical metrics are the composition of the waste, amount and types of radioactivity, radiotoxicity, thermal output, distribution of fissile materials, volume of disposed waste, mobility of released radionuclides, and calculations of exposure to groups and individuals over extended timescales that stretch to hundreds of thousands of years. These metrics provide a variety of ways to compare advanced

fuel cycles with “traditional” LWR fuel cycles; however, these metrics are not equally appropriate or useful when comparing different geologic disposal strategies.

7. Regulatory framework. Today, the United States has no viable regulatory basis for licensing a geologic repository for spent nuclear fuel, except at the Yucca Mountain site, because the regulatory framework in the United States is site specific. Creating the framework for licensing a geologic repository began in 1980 (10 CFR 60) and evolved over 40 years to the present predicament. The coordinated efforts of three federal agencies; extensive engagement with the public, states, and tribal governments; and a constant stream of litigation, often requiring federal agencies to amend rulemaking and requirements, were responsible for the long time required (Ewing, 2011). Appendix D provides information on the development of waste classifications and their jurisdictional bases.

   Any new repository site, even if sited in a willing community, will require licensing according to standards and regulations. The new standards and regulations may be site specific or generic. In any case, one can expect the process to take decades. While it is possible to develop general guidelines and standards to facilitate the siting of a new repository, there is no federal effort or organization in place to address this, casting doubt on the country’s ability to fully dispose of nuclear waste from closed or open fuel cycles.

Of these seven grand challenges for nuclear waste disposal, only the third may be impacted directly by advanced reactors and their fuel cycles. Two contributions could be made by advanced reactor fuel cycles to the safety case for a geologic repository: (1) changing the composition of waste streams for disposal or (2) improving the durability of nuclear fuels and nuclear waste forms in the disposal environment (Peters and Ewing, 2007). As one looks to advanced reactors and their fuel cycles, particularly with reprocessing, proponents need to clearly articulate the benefits to addressing the challenges of geologic disposal.

What can already be concluded is that all advanced fuel cycles will require a geologic repository. A new U.S. repository will require many decades of effort before the repository can accept nuclear waste for disposal.

Additionally, advanced reactors that change the inventory of actinide isotopes by using new fuel types, increased levels of enrichment, higher burnup, or use of actinides from reprocessing will not be solving an intractable waste disposal problem because the issue of the mobility of actinides can be addressed by the thoughtful selection of a site that has reducing conditions.

5.4 WHAT MATTERS?: WHAT THE COMMITTEE LEARNED FROM THE EXPERTS

During its deliberations, the committee was privileged to hear from a number of national and international experts on nuclear waste management and disposal. (See Appendix B for a listing of the committee’s public information-gathering sessions.) The following discussion outlines some important, specific points made during these presentations.

5.4.1 Important Factors for Evaluating the Safety of Geologic Repositories

The most common safety indicators for evaluating the impact of advanced reactors on the safety of a geologic repository include radiotoxicity inventory, radiotoxicity isolation, mobile fraction of radionuclide inventory, maximum endpoint dose, and comparison of released concentrations with natural occurrences of uranium (Grambow, 2021). For new fuel types being proposed that will use HALEU, another factor to examine in relation to repository safety is criticality. A careful analysis of each of these parameters reveals their limitations and emphasizes that the main issue is the interplay of toxicity and mobility in the geologic environment (Grambow, 2021). As an example, repositories in Sweden and Finland, under reducing conditions, show a peak dose that is governed by mobile radium-226, rather than plutonium, which has a much lower mobility in the granitic geologic setting. In contrast, the dose estimates for the proposed repository at Yucca Mountain, under oxidizing conditions, are dominated by plutonium contribution to risk assessments because of the higher mobility of plutonium under oxidizing conditions. Because the assessment of the fate of plutonium is estimated by solubility limits and sorption of the radionuclides at Yucca Mountain, the inventory reduction of actinides has little or no effect on disposal risk (Grambow, 2021).

Alternative fuel cycles may lead to the development of new waste forms, but the final analysis evaluating

the safety of a geologic repository depends on the repository concept and the efficacy of the multibarriers system. For example, for spent uranium oxide fuel on a mass basis of an individual fuel assembly,⁶ higher burnups lead to both higher inventories of fission products and faster release of iodine-129 (Grambow, 2021). However, the release of the most radiotoxic nuclides, such as isotopes of plutonium, depends on whether the geologic setting is oxidizing or reducing. The performance of mixed oxide fuel is similar to that of uranium oxide fuel, but the release of iodine-129 is faster than that of uranium oxide fuel, and the heat load will exceed that of spent uranium oxide fuel by as much as 33 percent. For vitrified waste from reprocessing, borosilicate glass is not an effective waste form for the retention of iodine-129. In the evaluation of repository performance and safety, the fuel cycle is less important than the redox state of the geologic repository.

5.4.2 Impact of Wastes from Advanced Reactors and Fuel Cycles on Geologic Disposal

The amounts and types of waste that will be generated by advanced reactors are difficult to estimate at this early stage of the development of advanced reactors; yet, this type of information is required in order to determine the impact of advanced reactors and advanced fuel cycles on the back end of the fuel cycle. Actual advanced fuel cycle options may reduce resource consumption or increase the energy output, but they are unlikely to reduce the long-term risks of waste management and disposal (Grambow, 2021). In most geologic settings, the dominant contributors to long-term dose estimates are mobile species (e.g., iodine-129), so the impact of alternative fuel cycles on estimates of long-term repository performance will be minimal (Swift, 2021; Swift and Sassani, 2019). Additionally, disposal volumes depend less on the waste volume generated for disposal and more on the heat generation from the wastes (Grambow, 2021), and without century-scale aging of fission products, alternative fuel cycles will not significantly impact thermal load management (Swift, 2021; Swift and Sassani, 2019). And, regardless, regulatory criteria may necessitate the disposal of fission products in a geologic repository (Swift, 2021; Swift and Sassani, 2019).

5.4.3 Additional Factors Influencing Fuel Cycle Decisions

As discussed in Chapter 2, the French nuclear program has established a long-term strategy for research, demonstration, and deployment of advanced reactors and fuel cycles. Therefore, insights and decision-making strategies from the French program provided useful information to this committee. First, because natural uranium resources are still available at low cost, the French nuclear industry is focused, in the short and medium time frames, on the success of third-generation reactors and does not plan to demonstrate and deploy fast breeder reactors until at least the second half of the 21st century (Landais, 2021). The public’s impression of nuclear power is “mostly determined by the accidents that have marked its history,” and emerging technologies, such as small modular reactors and fast-neutron reactors, do not at this stage change the public image of the nuclear industry in France (Landais, 2021). Furthermore, the role of nuclear energy in reducing greenhouse gas emissions continues to be highly debated within the European Community. As an example, the European Commission (EC) will issue a €250 billion “green bond,” but until 2021 the European Union excluded the use of green bonds for nuclear power plant projects (Landais, 2021). In a recent decision, the EC has included nuclear and natural gas as “green bond” projects, and in July 2022 the EU parliament voted in support of labeling nuclear and natural gas as green investments, with regulations expected in early 2023 (Ainger, 2022). Finally, the French emphasize that a key element of a successful fuel cycle, even one that includes reprocessing, is the implementation of a deep geologic repository program (Landais, 2021).

Based on the presentations and committee member expertise, the committee has concluded that the introduction and use of advanced reactors and small modular reactors in and of themselves will do little, if anything, to eliminate the need to manage and dispose of nuclear waste. Even if advanced fuel cycles were developed and implemented that reduced the actinide inventory 1,000-fold or more, the short- and long-lived fission products would still require management and isolation from the biosphere (geosphere) for long periods of time. Indeed, in some cases the amounts of certain categories of waste generated may increase, and certainly the types and characteristics of

___________________

⁶ This sentence has been modified from a prepublication version of the report to clarify the statement on spent uranium oxide fuel.

the highly radioactive waste and other waste categories—such as TRISO-based fuels, more highly enriched fuels, and higher-burnup fuels—may require focused research programs prior to their disposal in a geologic repository. That said, careful attention to the types of fuels and waste that require disposal and the selection of appropriate geologic settings should lead to the safe disposal of these fuels.

Also deserving careful attention is the combination of advanced reactors with different reprocessing schemes for actinides in fuel cycles that involve recycling: monorecycling, multiple recycling to close the fuel cycle, or different recycling schemes for managing the fate of minor actinides (Poinssot and Boullis, 2012). These factors materially impact the challenges of geologic disposal. Although each of these recycling schemes may reduce the actinide content of the waste, particularly for transuranic elements, they may actually increase the actinide content in the fuel cycle. Still, the long-term safety of disposal of actinides in appropriate geologic settings is assured in most geologic disposal projects largely independent of the actinide inventory of the repository (Grambow, 2008). The main challenge is the containment of long-lived fission product elements that, due to their chemistry and mobility in most geologic environments, can be major contributors to radiation dose over long periods. The main advantage of actinide recycling is the conservation of uranium resources at the front end of the fuel cycle and the associated reduction in the environmental impact of uranium mining.

5.5 SPECIFIC WASTE ISSUES THAT ARISE FROM ADVANCED NUCLEAR REACTORS AND FUEL CYCLES

Many advanced nuclear reactor concepts have been proposed in recent years, most of which use nonwater coolants and innovative fuels, with graphite moderators for thermal reactors. Many of these advanced reactors also use uranium fuels with higher enrichment than what is used in the current pressurized water reactor (PWR) fleet and produce high-burnup spent fuels. These advanced reactors are currently at various stages of design and development. Notably, none has reached the operating demonstration plant stage yet. As a result, it is difficult to be precise about the chemical and physical nature of the irradiated fuels or the details of operational and decommissioning wastes. Nevertheless, it is possible to describe these waste streams in broad chemical and physical terms and identify possible processing, treatment, and disposition pathways based on available information and existing experiences in waste management technologies.

Key characteristics of spent nuclear fuel from these advanced reactors, such as amount, chemical composition, radionuclide inventory, thermal power, and waste form behavior, can in general be estimated using the key design features of the reactor and fuel cycle (Wigeland et al., 2014). However, without detailed designs and operational information, it is more difficult to estimate the types, amounts, and characteristics of low-level waste, including GTCC, generated during operations. As a result, to assess waste generation and management from advanced nuclear reactors, the committee focuses on new and or/unique waste streams generated from the advanced reactors and analyzes their potential advantages and disadvantages to waste management concerns.

Chapter 3 includes detailed descriptions of advanced reactors. As can be seen from Table 3.1, a variety of advanced reactor types are proposed, with different coolants, neutron spectra, fuel types, reactor sizes, and design details. Given the large number of reactor types and the immature stages of development of these reactors, the committee has decided to select a subset of the advanced reactors to analyze their waste management issues. The analyses below are only representative of potential waste management issues.

5.5.1 Waste from iPWRs

Integral pressurized water reactors (iPWRs) are designed with more passive safety features than existing large PWRs and are sized much smaller (notionally less than 300 MWe [megawatts electric]), and they can be used in tandem to create a larger power plant, depending on the needs of the owner. Although iPWRs may experience more neutron leakage than large PWRs, which can negatively impact fuel utilization, they can be designed to produce comparable spent nuclear fuel inventories on a per thermal power basis. For example, NuScale indicates that its iPWR, a 12-module, 924-MWe plant (3,000 MWth [megawatts thermal]), plans to use standard PWR fuel assemblies but a somewhat shortened fuel assembly length, with an expected discharge burnup of 41–60 GWd

(gigawatt-day)/MTHM, and to discharge a spent nuclear fuel inventory of 20–29 MTHM/GWe-yr (gigawatt electrical year) (NuScale Power, 2022a). The NuScale data are comparable to a discharge burnup of 50 GWd/MTHM and spent nuclear fuel inventory of 22 MTHM/year for a 1.0-GWe PWR considered as a reference in DOE’s 2014 Nuclear Fuel Cycle Evaluation and Screening (NFCE&S) report (Wigeland et al., 2014) (see Appendix E). NuScale has no plans to reprocess the spent fuel, and as a result, it would go to a deep geologic repository for disposal. Heat production from NuScale spent fuel should be similar to that of existing PWRs at an equivalent level of burnup.

iPWRs may depart from existing large PWRs in their production of GTCC waste streams. Because iPWR cores are smaller, ex-core components, including the reactor vessel, baffles, reflectors, and the steam generator, will be exposed to higher neutron fluxes than in a large PWR in which this equipment is further from the core (Brown et al., 2017). With the proper use of core baffles and reflectors, however, the core power distribution could be designed similarly to that of large PWR cores, with reduced ex-core neutron leakage, and extended fuel cycle length (Suk et al., 2021). In fact, the NuScale design certification application (NuScale Power, 2020) presents the total heat flux hot channel factor of 1.86 (compared with 2.60 in the AP1000 design control document) (Westinghouse Electric Company, 2003), indicating that the core power distribution for the small NuScale core could be flatter than that for the large AP1000 core. NuScale provided to the committee an update on its newest 250-MWth design NuScale Power Module (NPM) (NuScale, 2022b). Regarding low-level wastes (including GTCC), the update stated that the NuScale design has fewer components that can become long-lived intermediate-level waste from neutron activation compared with currently operating boiling water reactors and PWRs. At the end of life, after fuel is removed, about 95 percent of the radioactivity is associated with major components (e.g., reactor pressure vessel), which after an appropriate cool-down time (~10–15 years) are removed to support shorter decommissioning times and are classified as low-level waste. A portion of materials to be handled during decommissioning never gets contaminated or activated and can be released from nuclear control as nonradioactive waste (NuScale, 2022b). The same NuScale update mentioned that it plans to submit its standard design application amendment for the 250-MWth NPM for formal U.S. NRC review by the end of 2022 (NuScale, 2022b).

5.5.2 Waste from High-Temperature Gas-Cooled Reactors

Operation of all high-temperature gas-cooled reactors (HTGRs) will result in a unique set of waste streams compared with those of LWRs. HTGRs (and one molten salt–cooled reactor being developed by Kairos) propose to use TRISO particles as the basic building block for the fuel. See Section 4.2.4.1 for discussion of TRISO fuel development and production and Figure 4.1 for an image of a TRISO particle. Here, the focus is on the waste aspects of TRISO. The multiple pyrolytic layers making up the TRISO particles include a semiporous buffer layer that provides for long-term retention of gaseous fission products generated during reactor operation at high temperatures (Ball and Fisher, 2008), possibly in excess of 1200 K for the helium-cooled Xe-100 pebble-bed reactor design from X-energy.⁷ The entire fuel assembly has radionuclides associated with fission events within the kernels and can contain uranium impurities and fission products in the matrix because of impurities or defects in the coated particles, as well as activation products from other such impurities as lithium and chlorine, and radiocarbon produced by activation of nitrogen-14, carbon-13, and oxygen-17 (Grambow et al., 2010). Typical effects observed in coated particle fuel from neutron irradiation include kernel swelling, pressure buildup, densification of the buffer layer, formation of gaps between layers, and reduction in the tensile strength of silicon carbonide. All of these phenomena may increase the probability of particle failure, although significant coating failure has not yet been observed (Grambow et al., 2010).

Although the helium coolant used in HTGRs will not become radioactive itself, it will become contaminated by radionuclides from failed fuel particles and by radioactive carbonaceous dust from the graphite moderator (Xu et al., 2020b). Radioactive carbonaceous dust production is expected to be especially high in pebble-bed reactors from friction between the pebbles as they move past each other in the reactor core, as shown in the German pebble-bed

___________________

⁷ As discussed in Box 5.2, the multiple pyrolytic layers may also be able to act as primary barriers to the release of fission products during geologic disposal, although additional research is required to determine their effectiveness.

test reactor AVR, which operated from 1967 to 1988 (Humrickhouse, 2011; von Lensa et al., 2020). Estimated dust production rates range from 15 kg/yr to 100 kg/yr (Humrickhouse, 2011). Some radioactive contaminants in the helium coolant and dust will plate out or infiltrate pores on core components, making these components low-level or GTCC waste. Radioactive carbonaceous dust also presents an additional worker ingestion hazard. Furthermore, it can complicate reactor decommissioning, as it did with the AVR: the heavily contaminated core also had spent fuel pebbles stuck in crevices and cracked graphite reflectors, requiring the entire core to be filled with concrete to stabilize these materials before being transported to storage (von Lensa et al., 2020).

By far the largest volume of waste from HTGRs is graphite. Some of this graphite is in the form of neutron reflectors and other core components, which would be treated as GTCC or low-level waste. Strategies for the treatment, storage, and disposal of this type of graphite waste are discussed in Box 5.1. The vast majority of graphite waste from HTGRs, however, is from the TRISO-based fuel forms. Direct disposal of TRISO fuel (i.e., if the particles are not extracted from their graphite prisms or pebbles) would result in very large volumes of waste to manage. One estimate of waste produced from a 200-MWe pebble-bed HTGR is 90,000 spent fuel pebbles,

including about 17 MT of carbonaceous waste annually (Fuks et al., 2020). On an energy-normalized basis, X-energy’s Xe-100 reactor would produce 160 m³/yr of pebbles, compared with 6.8 m³/yr of spent fuel from a large PWR (Mulder, 2021). Terrestrial Energy’s graphite moderator has a 7-year lifetime and will generate 80 MT of graphite waste (LeBlanc, 2021). A 1-GW ThorCon molten salt reactor requires 780 MT of high-quality graphite, and an estimated 100 m³ of irradiated graphite will be generated per GWe-yr (Jorgensen, 2021). Direct disposal is also being suggested for BWXT’s Advanced Nuclear Reactor (BANR) and Framatome’s Steam-Cooled HTGR (SC-HTGR). Moreover, the disposal volume must be further expanded to prevent overheating or fire from the release of stored energy from radiation damage. Prior research into the feasibility of direct disposal of TRISO fuel, which has mostly been performed in Europe, is summarized in Box 5.2 and discussed in more detail in Appendix G, and demonstrates that much research is still required to qualify TRISO fuel for direct geologic disposal.

Because of the very large volumes of radioactive graphite wastes that would need to be managed for direct disposal of used prismatic fuel blocks or graphite pebbles, processes are being developed to separate TRISO fuel kernels from the graphite matrix to substantially reduce the volume of waste for disposal (Grambow et al., 2006;

Guittonneau et al., 2010). However, the removal of graphite from TRISO particles is still an immature technology. Somewhat better-established methods to remove the TRISO particles from the graphite matrix include combustion and mechanical separation, whereas experimental methods include electrochemical processing (Fuks et al., 2020). Combustion results in the carbon in the graphite being burned off in the form of radioactive carbon dioxide (containing carbon-14) (Forsberg and Peterson, 2015; Fuks et al., 2020). In past experiments with this method, the radioactive carbon dioxide was released to the atmosphere, an undesirable outcome currently because of the potential radiation doses to the public and the negative impact on climate change from the carbon dioxide. The

radioactive gases containing carbon dioxide could be captured and converted into solid calcium carbonate for disposal as GTCC or low-level waste (Forsberg and Peterson, 2015; Fuks et al., 2020), but this would add cost. Alternatively, graphite-containing spent fuel can be mechanically crushed to remove much of the graphite from the TRISO particles (Forsberg and Peterson, 2015; Fuks et al., 2020). The resulting graphite waste may contain significant radionuclides, depending on the process used, and will need to be treated as either GTCC or low-level waste. The graphite waste treatment methods described in Box 5.1 could also be applied to any graphite that is separated from the TRISO fuel particles. Additionally, any of these treatment methods would likely create additional GTCC and low-level waste streams that would need to be accounted for.

5.5.3 Waste from Sodium-Cooled Fast Reactors

Two developers of sodium-cooled fast reactors (SFRs) presented their reactor designs to the committee. ARC-100 (Advanced Reactor Concepts, LLC) proposes to use metal alloy (U/Zr) sodium-bonded fuel with 3-zone HALEU enrichment. ARC plans to treat the sodium-bonded spent fuel by pyroprocessing and to recover transuranic elements for fabrication into new fuel (Sackett and Arthur, 2021). Natrium (TerraPower) will be developed as a once-through fuel cycle in a two-phased approach (Hejzlar, 2021). In the first demonstration phase, a sodium-bonded metallic fuel will be employed (HALEU enrichment with 10 weight percent Zr [U-10Zr]) sodium-bonded to high-tensile-9 (HT9) cladding. The second phase, which would start 6–8 years after the demonstration start-up, involves the transition to a full core of an advanced metallic fuel without sodium bonding to the cladding.

While the specifics of waste treatment and disposal may vary based on reactor design, this section focuses on waste issues common to all SFRs: (1) radioactive sodium waste streams during the operation and decommissioning phases and (2) management of the spent nuclear fuel/high-level waste (SNF/HLW) for both sodium-bonded spent fuel and non–sodium-bonded spent metallic fuel.

5.5.3.1 Unique Waste Streams from Sodium-Cooled Fast Reactors During Operation and Decommissioning

Multiple waste streams are expected from operating and decommissioning an SFR. There will be routine operational waste, including solid, liquid, and gaseous waste, for which technology has been developed. All fast-spectrum reactors, including lead-cooled reactors (e.g., LeadCold’s SEALER-55 and Westinghouse’s LFR) and molten salt reactors (e.g., TerraPower’s MCFR and Moltex’s Stable Salt Reactor) use reflectors and shielding, which become irradiated and contaminated and will require disposal as GTCC in the United States or intermediate-level waste for operation outside the United States. In addition to the low-level liquid and solid waste generated during normal operations, TerraPower suggests that its Natrium SFR will produce about 80 MT of reflectors and other reactor-vessel irradiated internals as low-level (Class C) waste, 70 MT of irradiated control rods also as low-level Class C waste, 4–8 m³ of activated primary cold traps (containing cesium and tritium) that will require storage with decay and treatment before disposal as GTCC (TerraPower, 2021b). The ARC-100 Advanced SMR will produce 67 MT of reflectors and shielding and 4.6 MT of control assemblies, as well as GTCC and low-level waste (Sackett and Arthur, 2021).

The most unique and challenging waste streams generated from SFRs are those associated with the liquid sodium coolant (IAEA, 2019e). Volumes of sodium coolant will vary depending on the reactor design (e.g., pool type versus loop type) and reactor power. As an example, TerraPower indicates that its Natrium reactor will produce 800 m³ of activated sodium (Hejzlar, 2021). Part of the challenge of dealing with this material is that sodium reacts with moisture and air to pose a significant fire hazard. It should also be noted that sodium wastes are Resource Conservation and Recovery Act wastes subject to state regulation. Although these challenges are unique to SFRs, they are not new, and there is some experience from decommissioning of several sodium-cooled reactors. Special processes have been developed to manage activated sodium and sodium-contaminated materials (IAEA, 2007b). Sodium coolant from decommissioned SFRs may be used in new reactors (IAEA, 2007b).

A key difference between sodium waste treatment during operations and decommissioning is the waste volume that must be handled. The waste quantities are considerably larger during decommissioning because all sodium coolant will finally be drained and become waste. For example, upon decommissioning the Experimental Breeder Reactor II (EBR-II) produced a total of 327 MT of sodium coolant, and the Fast Flux Test Facility

(FFTF) generated 900 MT of sodium coolant (IAEA, 2007b). The bulk sodium coolant contains several sources of radioactivity, including sodium activation products and contamination from actinides and fission products. The main radionuclides include sodium-22, tritium, cobalt-60, and cesium-137.

During decommissioning of an SFR, in addition to the large quantity of sodium coolant that needs to be drained, a significant amount of residual sodium usually remains. For example, residual sodium in EBR-II amounted to about 1300 kg in the primary vessel and 400 kg in the secondary circuit (IAEA, 2007b). There are several sources of “residual sodium on and within components removed from sodium circuits. Residual sodium may consist of a thin layer wetting the surfaces or may be larger quantities retained in pockets that cannot be drained during component removal” (IAEA, 2007b). The residual sodium needs to be removed because of the high chemical reactivity of sodium. Unloaded fuel assemblies would also have to be cleaned of residual sodium before storage in air or water. Additionally, “any maintenance, repair, or modification carried out on a circuit or a component wetted with sodium will generate sodium-contaminated gloves, tools, etc.” (IAEA, 2007b). Importantly, while storing such sodium-contaminated items, it is necessary to avoid contact with air due to the potential fire hazard.

Many sodium treatment technologies have been developed over the years, in both the nuclear and industrial sectors, where sodium metal is used in many processes and applications. Some techniques involve pretreatment to remove cesium and tritium, in order to reduce the waste classification for disposal. The majority of treatment processes involve phases of neutralization and separation of liquid sodium from the metal surface. There are two proven technologies for the treatment of bulk sodium: (1) the NOAH process, involving a highly controlled reaction of sodium with water to produce liquid sodium hydroxide that is subsequently neutralized, and (2) the Argonne process used for EBR-II and Fermi-1, where a caustic process is used to react sodium with aqueous sodium hydroxide solutions to generate a concrete-like sodium hydroxide monohydrate crystal that can be packed in drums (IAEA, 2007b).

A variety of technologies are being considered for conditioning the sodium waste for storage and disposal, including drying solidification, cementation, conversion into dry carbonates, and conversion into dry sodium salt. Although many of these technologies are at a high level of technical readiness, few are currently ready to be used for commercial-scale applications.

5.5.3.2 Spent Nuclear Fuel and High-Level Waste

Most fast reactors, including SFRs, use metallic fuels, which consist of a heavy metal (typically U) that is alloyed with various other metals. As described in Chapter 4, metallic fuel is usually bonded to cladding with liquid sodium to improve heat transfer. Typical fuel materials include U-Zr and U-Pu-Zr, and cladding materials are austenitic stainless steels or ferritic-martensitic steels. TerraPower has proposed a non-sodium-bonded metallic fuel comprised of U metal annular fuel with He-filled central pore and advanced ferritic-martensitic steel cladding to reduce void swelling.

Traditionally, sodium-bonded spent nuclear fuel falls into two categories: driver fuel and blanket fuel. After irradiation of a driver fuel, some of the metallic sodium enters the metallic fuel and becomes inseparable from it. In addition, spent driver fuel and cladding components interdiffuse to such an extent that removal of sodium is not practical by mechanical stripping of the fuel cladding. On the other hand, irradiation of blanket fuel results in little metallic sodium diffusion into the fuel, thus allowing mechanical stripping of the spent blanket fuel cladding (DOE, 2000).

The Natrium reactor concept proposed by TerraPower includes the development of an advanced metallic fuel (Type 1B fuel) that does not use sodium to bond the fuel to the cladding. Instead, Type 1B fuel uses annular fuel slugs with an inert gas–filled central hole, with the fuel having direct thermal contact with an advanced ferritic-martensitic steel cladding material with very low void swelling (Hejzlar, 2021).

Preparing for direct disposal of the Natrium Demo spent fuel to a geologic repository will require significant effort, including gathering the necessary technical information and verifications, because the chemical composition, radionuclide composition, and material properties of the Natrium Demo spent fuel (both Types 1 and 1B) are different from U.S. LWR spent fuel.

5.5.3.3 Treatment of Sodium-Bonded Spent Metallic Fuel

Due to the exothermic reaction of sodium with water, spent sodium-bonded metallic fuel has not been considered acceptable for direct disposal in any geologic repository and therefore must be processed prior to disposal to develop acceptable waste forms. DOE has been conducting research and development (R&D) to treat driver and blanket spent sodium-bonded metallic fuel from EBR-II using electrometallurgical technology (EMT) or pyroprocessing. As discussed in Chapter 4, pyroprocessing consists of electrorefining the spent fuel in an electrochemical cell and allows for the separation of the fuel into (1) metallic uranium, (2) a metallic waste form, and (3) a highly radioactive salt mixture (National Research Council, 2000). In DOE’s 2000 Record of Decision (DOE, 2000) choosing EMT for the treatment of sodium-bonded spent metallic fuels, it also assessed other treatment options such as plutonium uranium reduction extraction (PUREX) and “melt and dilute.” More recently, DOE is considering a “melt, distill, and dilute” process to deal with the sodium-bonded spent fuel from the Versatile Test Reactor planned for Idaho National Laboratory (Crawford, 2020). The resulting waste streams would include highly radioactive ingots, contaminated sodium hydroxide, and the condensable fission products, as well as nonrecyclable fuel production scrap (Crawford, 2020).

5.5.3.4 Wastes from Electrometallurgical Technology (Pyroprocessing)

As detailed in Sections 4.3.6.3 and 4.3.6.5, a number of waste streams are generated from pyroprocessing. The chopped spent metallic fuels are placed in an anode metal basket and immersed in a 500°C molten LiCl and KCl salt. When current is passed through the metal baskets, fission products and actinides are oxidized and dissolved in the salt bath. The U is reduced to its metallic form and accumulates on the cathode. Cladding and noble metal fission products remain in the anode and can be cast into metal ingots and become metal high-level waste forms. Fission products in the salt bath are first passed through zeolite columns, then mixed with glass and pressed into a glass-bonded sodalite, a ceramic form of high-level waste. Theoretically, U and actinides in the cathode can be recycled.

5.5.3.5 Waste Forms for Sodium-Bonded Spent Metallic Fuel

Because of the high sodium concentration, salt waste cannot be processed into glass waste form. Instead, it is blended with zeolite and glass into a glass-bonded sodalite composite ceramic high-level waste form (Hall et al., 2019a). This ceramic waste form can immobilize radionuclides in the ceramic structure and dilute the actinide concentrations. Furthermore, the chloride salts are incorporated into the cages of the zeolite structure, forming salt-loaded sodalite, which makes the salt much less soluble and less corrosive when in contact with water (Hall et al., 2019a). The EMT salt waste will generate a large mass of ceramic waste form—it is estimated that 1.72 MT of EBR-II salt waste will produce 50.95 MT of ceramic waste form (Frank and Paterson, 2014).

The predominant composition (about 90 percent) of the metal high-level waste form is irradiated stainless-steel cladding. The metal high-level waste form has two main phases interspersed in the microscopic scale: an intermetallic Zr (Fe, Cr, Ni) phase and an Fe solid solution phase. ²³⁴U and ⁹⁹Tc are the primary dose contributors, and the amount of heat-generating radionuclides is very small, so the thermal output of the metallic high-level waste form is negligible.

DOE research has concluded that ceramic and metallic waste forms converted from spent metallic fuel are acceptable for repository disposal (Hall et al., 2019a). For these waste forms, radionuclide release rates would depend largely on repository and fuel conditions, which include temperature, fluid chemistry, and radiation level. Rechard et al. (2017) also proposed direct disposal options for the salt waste: (1) deep borehole in crystalline basement rock and (2) the Waste Isolation Pilot Plant (WIPP) in bedded salt in New Mexico. In the deep borehole disposal concept, the salt waste would be placed in containers at the bottom 1–2 km of an approximately 5-km-deep borehole. According to Rechard et al. (2017), the advantages of the deep borehole disposal concept include (1) restricted migration of radionuclides due to low permeability in deep crystalline rocks, (2) little interaction between deep and shallow fluids, (3) limited solubility and enhanced sorption of radionuclides in a deep reducing

environment, and (4) further restricted mobility of radionuclides by the high salt content of the waste. The study performed thermal-hydrologic analysis of the disposal concept, demonstrating its feasibility. Rechard et al. (2017) also described a roadmap for the disposition of salt waste as remote-handled transuranic waste at WIPP, which may be technically feasible but would encounter many regulatory and legal obstacles, such as modification of the WIPP permit and an environmental impact assessment.

A 2019 assessment performed by the Center for Nuclear Waste Regulatory Analysis, prepared for the U.S. NRC, found:

For these waste forms, radionuclide release rates would depend largely on repository and fuel conditions, which include temperature, fluid chemistry, and radiation level.

5.5.4 Waste from Molten Salt Reactors

Molten salt reactors (MSRs) come in a variety of design concepts and are adaptable to a wide range of fuel cycles, as discussed in Sections 3.2.3.5 and 4.3.6.5. As a result of these different designs, the specific waste streams will vary, but there are general similarities across all MSR design concepts. MSR waste streams can include off-gases, fuel-related waste streams, graphite or carbon components, metallic reactor components, and operational and decommissioning wastes. Although many of the MSR wastes are different from LWR-type wastes, researchers have identified potential pathways for management and disposition of each potential MSR waste (Riley et al., 2018, 2019). However, spent fuel storage on-site at MSRs will differ significantly from that of LWRs. Because of the air- and water-sensitivity of the salt, it may require new interventions to stabilize it for storage or before transfer to a waste repository (Riley et al., 2018).

Two waste streams will carry most of the radionuclides generated during MSR operation: off-gas wastes and the fuel salt itself. Because MSR fuel is liquid, the volatile radionuclides are not contained by fuel cladding, and as a result, the noble gas fission products (such as Xe and Kr, which are also precursors to decay products such as Cs, Ba, Rb, and Sr), as well as aerosolized salts, particulates, reactive gases (I₂, Cl₂, F₂, HF), halides, O₂, N₂, and ³H must be captured by an off-gas system (McFarlane et al., 2020; Riley et al., 2019). Components of the off-gas system vary depending on the design but can comprise decay tanks and piping and filter banks to confine the radionuclides as they decay. Each of these components needs to be maintained during operation and treated separately for final disposal depending on the nature of the radioactive species trapped within the unit (McFarlane, 2021; Riley et al., 2019). The particulates, aerosols, and reactive gases (I₂, Cl₂, F₂) can be immobilized in a ceramic waste form; the ³H waste is more difficult to treat, as are the residual halides (McFarlane et al., 2020; Riley et al., 2019). The noble gases, such as Xe and Kr, can be stored in the off-gas system and allowed to decay (Riley et al., 2019). Development of an off-gas system continues to be an integral part of reactor design. Research into both systems and individual components is ongoing. For instance, a molten hydroxide scrubber is being developed for aerosol and acidic gas removal, and metal organic frameworks have been developed for noble gas separations. Both are being studied for MSR applications (Riley et al., 2019).

Carbon-related waste steams will be generated in some MSR designs where graphite is used as a moderator or reflector. Neutron radiation damage affects the lifetime of the carbon-based components, and for many of the reactor designs may result in the need to replace the graphite moderator approximately every 7 years, depending on such factors as reactor flux and salt type (Riley et al., 2019). These graphite components will experience both surface contamination and diffusion of radionuclide species into graphite pore space (Forsberg et al., 2017; Riley

et al., 2019) (see Box 5.1). Thus, the relatively large volume of carbon-based waste streams might pose disposal challenges unless they can be sufficiently decontaminated to low-level waste (Riley et al., 2019). To reduce the amount of high-level waste, these materials may first require processing (instead of direct disposal) to remove fission products, salt, and ³H, followed by compaction prior to disposal. It is also possible that the graphite could be recycled once most of the radionuclides have been removed, although this is unlikely because of residual ¹⁴C (Riley et al., 2019).

In addition to graphite reactor components, metal reactor components will become activated and require disposal as either GTCC or low-level waste, which could be an option for other wastes from decommissioning and operating streams if they can be reduced in size and/or decontaminated (Riley et al., 2019). Salts and the reactive gases (e.g., I₂, HF, Cl₂) are very corrosive to metallic materials, such as those found in the reactor vessel, although their effects will be minimized by redox control. Resulting corrosion products, such as Cr and Fe, will tend to transport from the hot zone to the cooler areas in the reactor primary circuit, where they are expected to plate out on the surfaces. Metal surfaces after decommissioning will be coated with salt and insoluble noble metal fission products (e.g., Mo, Pd, Rh, Tu, Tc, Nb, Sb, Ag) that tend to plate out on reactor surfaces (Riley et al., 2019). These materials may either be decontaminated and recycled or disposed of directly in a repository. While decontamination processes will result in additional waste streams, this is not unique to MSR wastes. Such an effort will require a cost-benefit analysis.

The MSR fuel salts have high water solubility, unlike ceramic wastes generated from LWRs. Thus, managing and disposing of the fuel salt itself and containment of the radionuclides poses a challenge. One option would be to simply allow the fuel salt to solidify and dispose of it directly in a deep geologic repository, especially in bedded salt or a salt dome. However, water movement through a salt repository or pressurized brines will transport radionuclides. Geologic repository environments other than salt would be inappropriate for direct disposal because of the potential for water to dissolve the fuel salt waste, readily releasing radionuclides into the subsurface. Even in salt-based repositories, concerns about interaction with a brine producing either a soluble U phase from the hydration of crystalline UF₄ or the production of hydrofluoric acid from UF₄ would exist (Grenthe et al., 1992; Kozak et al., 1992; Tracy et al., 2016). As a result, most analysts considering fuel salt disposal are planning to first process the salt into more durable waste forms (McFarlane, 2021; Riley et al., 2019; Terrestrial Energy, 2021).

Thus, salt-based waste must either be dehalogenated or the halide forms must be stabilized before disposal. Treating MSR fuel salts to produce waste forms acceptable for disposal in a geologic repository will have to be capable of containing all the different species present in the salts, including halides, alkalis, alkaline earths, rare earths, and actinides, although it is likely that the actinides will be removed before disposal. As a result, a combination of possible waste forms, including ceramics or mineral waste forms, ceramic metals (cermet), halide metals (halmet), and possibly glass waste forms are being considered (Riley et al., 2018, 2019). Numerous separation technologies can be used to process the salt, including reductive extraction, oxidative precipitation, distillation, melt crystallization, dehalogenation, phosphorylation, ion exchange, and a glass material oxidation and dissolution system (Riley et al., 2019). However, as noted by Riley et al. (2019), no single technology will provide the solution, but “management strategies exist for all types of waste based on demonstrated technologies. These management strategies range from simple to complex and from direct disposal to separating, partitioning, and recycling fractions of the waste for future MSRs.”⁸

Immobilizing MSR fuel salts may have a significant impact on the volume of high-level waste. Previous experience with production of ceramic waste forms from the salt wastes resulting from the pyroprocessing of EBR-II spent fuel resulted in a 30-fold increase in waste mass, with 1.72 MT of EBR-II waste producing 50.95 MT of ceramic waste form (Rechard et al., 2017). In addition, all of these salt waste processing technologies will generate additional GTCC and low-level waste streams, which will have to be accounted for and disposed of. Conversely, unlike pyroprocessing, additional salt does not need to be added to solubilize the spent nuclear MSR fuel. As most carrier salts for both fluoride and chloride reactors require isotopic enrichment, there are schemes for recycling the carrier salts. Finally, the MSR technology is being considered for “waste burning,” in order to

___________________

⁸ One committee member does not agree that management strategies have been demonstrated for all types of wastes from MSRs at this point in time. If that were the case, the Oak Ridge Molten Salt Reactor Experiment wastes would already have been dealt with; they have not.

reduce the inventory of actinides that will comprise high-level waste. These factors must all be considered in the determination of the masses, volumes, or compositions of these additional waste streams.

As there is no operating repository, the fuel salt will need to be stored at the reactor site prior to disposal. Experience from the Molten Salt Reactor Experiment shows that solid or frozen fuel salt in storage tanks can generate fluorine and volatile U species via radiolysis (Haghighi et al., 2002; McMillan, 2019). This situation arose because of residual actinides remaining in the salt even after most of the U had been removed by fluoride volatility. A 1997 study assessed that “criticality becomes a concern when more than 5 kg of uranium concentrates to over 8 weight percent of the salt in a favorable geometry” and advised avoiding that situation (Hollenbach and Hopper, 1997).⁹ To mitigate the accumulation of hazardous UF₆ gas, some designers plan to heat spent fuel in storage tanks to more than 200°C to increase the recombination rate of the products of radiolysis (Terrestrial Energy, 2021). Until the salt is processed into a stable waste form, storage tanks will need monitored off-gas collection systems to mitigate this risk.

Additional waste streams, common to all reactor designs, will include operational wastes from equipment, personal protective equipment, high-efficiency particle air filters, and other operating materials (Riley et al., 2019; Terrestrial Energy, 2021), as well as waste streams dominated by GTCC and low-level wastes from the decommissioning of these reactor facilities.

5.5.5 Characteristics of Spent Nuclear Fuel and High-Level Waste from Advanced Reactors and Their Impacts on Geologic Disposal

As noted in the earlier discussion, the spent nuclear fuel isotopic compositions and evolution as a function of time is important information for geologic disposal. Other than for the Natrium SFR from TerraPower, there is little detailed information provided by most advanced reactor vendors on the characteristics of spent nuclear fuel isotopic compositions. The committee therefore decided to use the information from DOE’s 2014 Nuclear Fuel Cycle Evaluation and Screening (NFCE&S) report (Wigeland et al., 2014) to supplement the assessment of potential impacts of advanced reactors on geologic repository disposal. Although the information from NFCE&S may not exactly represent the advanced reactors being developed by the vendors, it does capture key features of the designs that are similar to some of the advanced reactors. Three metrics used by NFCE&S in its assessment of waste management impacts were selected by the study committee: “(1) mass of SNF/HLW disposed per energy generated, (2) activity of SNF/HLW at 100 years per energy generated, and (3) activity of SNF/HLW at 100,000 years per energy generated” (Wigeland et al., 2014). In particular, the committee selected data from NFCE&S for three reactor designs: the high-temperature gas-cooled reactor (HTGR), the molten salt reactor (MSR), and the pressurized water reactor (PWR) (as the base case for comparison). The corresponding data for the TerraPower SFR are also compared. A two-tier coupled SFR-PWR system is also analyzed in Appendix E to indicate the potential impact of recycling the Pu generated in both the PWR and SFR cores. The symbiotic fuel cycle illustrates the synergistic arrangement where the Pu produced, together with the remaining U, is recycled, and minor actinides and fission products are sent to a repository. The only feed material required is a small amount of natural U for a combined system replenishing the fission products and minor actinides to be sent to a geologic repository. Appendix E provides details of data and information from TerraPower and NFCE&S and shows the following results for each type of advanced reactor (see Tables E.3–E.5 in Appendix E):

SNF/HLW mass per energy generated: For the three advanced reactor designs, reduction in the masses of SNF/HLW requiring geologic disposal per unit of electricity generation compared with the base case PWR ranges from a factor of 3 to about one order of magnitude. These results are to be expected for unit energy generation,

___________________

⁹ The Molten Salt Reactor Experiment Remediation Project was initiated by Oak Ridge National Laboratory to stabilize conditions in the facility by treating and removing UF₆ from fuel and flush salts, and converting it to the stable oxide form for storage, pending ultimate disposal. As of 1998, “All significant quantities of fissile material are under appropriate criticality safety controls. No portion of the off-gas system is at a positive pressure with respect to its surroundings. The potential for an energetic chemical reaction in the charcoal bed has been eliminated” (Peretz et al., 1998). Currently, surveillance and maintenance at this facility is ongoing, and decommissioning concepts are being evaluated (McMillan, 2019).

since higher enrichment, higher burnup, and any recycling will reduce the SNF/HLW mass. This mass reduction indicates the somewhat better utilization of resources for advanced reactors. However, the reduced mass does not necessarily translate into advantages to disposal in a geologic repository, because the heat load of the waste (most significant) and volume of the waste (less significant) will typically impact the repository capacity management.

SNF/HLW radioactivity per energy generated at 100 years after discharge: With the fission process releasing two fission products and approximately 200 MeV (mega electron volts) of energy per fission, a 1.0-GWe (gigawatt electrical) power plant consumes approximately 1.0 Mg (megagrams)/year of nuclear fuel and produces 1.0 Mg/year of fission products, making the fission product inventory in spent nuclear fuel proportional to the thermal energy generated. Therefore, on a per-unit-energy-generation basis, the four reactor designs produce similar radioactivity from fission products, and fission product inventories in the spent fuel are reduced for designs with a higher thermal efficiency.

The contributions from short-lived fission products (e.g., ⁹⁰Sr, ¹³⁷Cs) dominate the radioactivity of SNF/HLW at 100 years and produce significant heat. Advanced reactors will in general produce higher amounts of heat in each spent nuclear fuel package because of the higher burnup resulting in the higher heat load. The thermal load of waste, in addition to the mass, volume, and other factors,¹⁰ will impact the spacing of waste packages, repository footprints, and engineering designs.

SNF/HLW radioactivity per energy generated at 100,000 years after discharge: On a per-unit-energy-generation basis, the three advanced reactor designs, as well as the two-tier coupled PWR-SFR system, can reduce the amount of long-term radioactivity from a few percent to a factor of 2 as compared with the PWR system. These reductions, however, do not have significant impacts on the performance of the repository if geochemical and geologic conditions of the repository are carefully chosen to limit the mobility and accessibility of the radionuclides.

From the analysis summarized here and detailed in Appendix E, plus the discussion in Section 5.4.1 on the important factors for evaluating the safety of geologic repositories, the committee emphasizes the observation that the current advanced reactor technologies do not appear to offer significant beneficial impacts to the disposal of SNF/HLW in a geologic repository.

5.6 POTENTIAL IMPACTS OF ADVANCED NUCLEAR FUEL CYCLE WASTES ON STORAGE AND TRANSPORTATION OPERATIONS

5.6.1 Current Status

Storage and transportation management of radioactive wastes from LWR power generation has evolved significantly over the past 50 years of operations. The regulatory environment is stable, and the technology is mature. Operational experience further points to an understanding of how best to safely manage a hazardous material within the context of strict regulatory requirements. This is not to say, however, that there are no technical or regulatory issues left to be addressed. Higher burnups, increased enrichments, new fuel forms, extended dry storage, and potential storage-transportation-storage operational scenarios generate new issues not necessarily envisioned during the initial licensing of current storage and transportation systems. Proposed new fuel designs stemming from the advanced reactor initiatives will introduce issues that will need to be addressed from both regulatory and technical perspectives. Work continues on both the licensing and R&D fronts to address issues as they are identified to assure safety and regulatory compliance through the entire fuel cycle.

Given this backdrop, it is important to provide a brief background of the LWR spent fuel regulatory and technical status, as it will provide a benchmark by which storage and transportation management of advanced fuel cycle wastes will be evaluated. The Code of Federal Regulations (CFR) Title 10, Parts 71 and 72 (10 CFR

___________________

¹⁰ The sentence was modified following a prepublication version of the report to indicate factors other than thermal load that impact the spacing of waste packages, repository footprints, and engineering designs.

71, 72) provide the primary regulatory criteria for the design of transportation and storage systems for high-level radioactive and fissile materials.

For transportation, 10 CFR 71 safety criteria address fuel containment, criticality, shielding, and thermal management. These criteria include the 9-m drop test and 30-minute fire-loading requirements for assessing a packaging’s ability to maintain containment during extreme mechanical and thermal loading conditions. Part 71 also includes a nonmechanistic criticality criterion that assumes breach of containment, full moderation, and fuel spatial reconfiguration to afford the optimal potential of creating a possible criticality event. Shielding and cask surface temperature thresholds are also defined 10 CFR 71 criteria. To support the licensing process, the U.S. NRC has recently published a review guide, NUREG-2216, to assist both the licensee and U.S. NRC review team with a consistent approach for assessing the safety of a transportation packaging design against the regulatory criteria (U.S. NRC, 2020d).

For storage, 10 CFR 72 criteria provide design requirements for both wet pool and dry storage operations not associated with Part 50 (10 CFR 50), which covers pool storage requirements necessary for removal of fuel immediately from the reactor. 10 CFR 72 safety criteria address fuel confinement, shielding, criticality, and fuel retrieval, and provide guidance to environmental loadings such as earthquakes, tornadoes, floods, and temperature extremes that the storage system may be subjected to over its lifetime. Similar to the transportation regulations, the U.S. NRC has recently published a technical review guide for the design of dry storage systems, NUREG-2215 (U.S. NRC, 2020e).

Safety assessments for both storage and transportation, in principle, focus on fuel containment and confinement, criticality, shielding, and thermal management. While most licensing applications historically have been associated with LWR spent fuel, the limited body of licenses for advanced fuel cycle materials suggests that future license applications will be judged against these same regulatory criteria (DOE, n.d.-a; Hall et al., 2019b, 2020). Therefore, for the advanced fuel cycle materials addressed in this study, assessment of storage and transportation systems envisioned for these materials will be based on the current U.S. NRC regulations.

As a policy in the United States, the once-through fuel cycle depends on a geologic/engineered repository for final disposition of spent nuclear fuel. The Yucca Mountain project was the planned geologic disposal site; however, as noted above, development of this site has not been pursued, and the future of repository planning is uncertain. This has created a spent fuel management issue that looms large for the country. Figure 5.4 illustrates current (as of December 2019) and projected quantities of LWR spent fuel in storage (Freeze et al., 2021). The red line in Figure 5.4 shows the entire U.S. inventory of spent fuel as it is off-loaded from reactor operations over time and indicates that the spent fuel inventory has already exceeded the statutory capacity of the Yucca Mountain repository (70,000 MTHM, per the NWPA). The total sum of fuel in storage is divided between pool and dry storage. As the figure shows, current storage is split fairly evenly between wet and dry storage, with a continuing increase in dry storage to the point where virtually all spent fuel will be in dry storage systems. The Commercial Spent Fuel Projections (CSFP—solid lines) and Next Generation Systems Analysis Model (NGSAM—dashed lines) projections represent two different models with different input assumptions. While there are some differences in the plots, the general trends and results are similar.

Utilities have moved to dry storage systems because they ran out of space in spent fuel pools and because the capital and operational costs are well understood and predictable. Operations and maintenance are also straightforward and predictable. Passive cooling means that there are no pumps or coolants that need to be maintained, and thermal management using natural convection has been shown to be effective (Fort et al., 2019).

There are currently just over 3,200 loaded canisters in dry cask storage systems in the United States. Incremental design changes over the years have resulted in large canistered systems with payloads of up to 37 PWR assemblies or 89 boiling water reactor assemblies, loaded weights of up to 50 MT, and heat generation of up to 40 kW (Freeze et al., 2021). These canistered systems are designed for transportation as well as storage; however, they are licensed for transport at much a lower heat generation than during storage. Currently, the utilities do not intend to repackage the spent fuel for transportation after storage.

While these dry systems meet all the safety criteria for storage, downstream issues may arise from their canistered, high-payload configuration. High heat loads may require longer-term storage prior to transportation to meet the cask surface temperature limit of 50°C. Some of the licensed storage systems do not, as yet, have compatible licensed

transportation casks. Work is ongoing to design, test, and deploy a railcar to the American Association of Railroad’s S-2043 standard. In January 2022, DOE issued a request for proposals for “the fabrication and testing of a prototype eight-axle railcar” that meets this standard (DOE, 2022b). This railcar will be needed once transport campaigns begin. The sheer size of these canisters may create operational issues when moving from storage to transportation operations. Issues such as crane capacity, facility clearances, and rail access at shutdown sites need to be addressed. The quantity of canistered systems may render repackaging unrealistic. The current inventory of 3,200 canisters is expected to top out at approximately 10,000. Costs to repackage spent fuel from existing canisters to a disposal canister have been estimated to be ~$20 billion in 2018 dollars (Freeze et al., 2021). In addition to the operational and material costs associated with repackaging, consideration must be given to the fact that the United States currently does not have the infrastructure necessary to conduct these operations at scale. Repackaging would require a significant capital investment and an accountancy for the time needed to design, permit, and build the infrastructure.

As a result, while the United States has licensed storage systems for spent fuel that meet all current storage regulatory criteria, these systems may create issues for downstream extended storage and transportation and may not be suitable for disposal. The current lack of a repository site, coupled with the lack of established design criteria for the waste packaging, creates a dilemma for how spent fuel is currently packaged for storage. It is worth noting that DOE’s Office of Civilian Radioactive Waste Management (OCRWM) was established as part of the NWPA to develop a repository, as well as the infrastructure required to move spent fuel from the reactor sites to the repository. While OCRWM was engaged with the development of the Yucca Mountain site, it also had an active program to develop the transportation infrastructure to deliver commercial spent fuel to the repository. With the cessation of licensing activities for the Yucca Mountain repository, OCRWM was disbanded, resulting in an organizational lack of attention to linking important storage and transportation infrastructure issues to eventual disposal requirements. Nonetheless, in recent years, DOE’s Office of Integrated Waste Management has been examining integrated approaches to storage, transportation, and disposal of spent nuclear fuel and other high-level radioactive wastes (Nutt, 2021).

Regulatory accommodation for current LWR spent nuclear fuel has been addressed to allow for long-term storage through the Continued Storage Rule (U.S. NRC, 2014a). This rule provides a pathway for continued storage of spent nuclear fuel based on three storage scenarios analyzed in the U.S. NRC’s associated General Environmental Impact Statement (GEIS) (U.S. NRC, 2014b): a repository becomes available 60 years after the licensed life of a reactor (short-term storage), a repository becomes available 100 years after the short-term storage scenario (long-term storage), and a repository never becomes available. It is important to note that the U.S. NRC GEIS specifically excludes the evaluation of advanced reactors, GTCC waste,¹¹ high-level waste from reprocessing, and associated wastes from reprocessing. This points to a regulatory gap for long-term storage of those wastes excluded in the U.S. NRC GEIS. Arguably, this issue will not arise until decades into the future. However, it is important to recognize these gaps early and to understand the potential impacts on costs, schedules, and operations as the issues are addressed.

The lesson for advanced fuel cycle materials is this: while plans for packaging advanced fuels and materials for storage may meet regulatory criteria, their waste forms and packaging may not be compatible with downstream extended storage, transportation, and disposal regulations and needs. Developers of advanced reactors and fuel cycles need to be transparent and clear about development of storage systems for new materials and how their designs may impact downstream operations. This applies not only to spent fuel, but also to GTCC and low-level waste, as well as hazardous materials that are generated from reactor operations. As part of its Extended Storage Collaboration Program within its Used Fuel and High-Level Waste Management program, EPRI has established a task force in collaboration with industry to evaluate potential issues related to the back end of the fuel cycle for advanced reactors (EPRI, 2021b, 2022).

5.6.2 Advanced Reactor Fuels and Materials

Storage and transportation assessments for advanced fuel cycle materials may best be addressed by categorizing the fuel proposed for advanced reactors. However, given its unique impact on the advanced fuel cycles being proposed, HALEU used as a fissile feedstock, through to the back end as a component of spent fuel, will be discussed first, followed by advanced fuels.

5.6.2.1 HALEU-Based Fuels

HALEU is being proposed as source fissile feedstock for most of the advanced fuels that will feed the advanced reactors. From the presentations given to the committee regarding HALEU, there is a general concern with regard to the availability of supply in the time frames needed to support deployment of demonstration and prototype advanced reactors (see Section 4.2.3). The most likely near-term source of HALEU to supply DOE initiatives is Russia. Current world events have exposed the vulnerability of relying on nondomestic sources.

Beyond this general concern, transportation details associated with HALEU supply need to be considered with regard to schedules, cost, and regulatory gaps. For example, the International Atomic Energy Agency (IAEA) recently published an issues assessment of LWR fuel enriched to >5 percent (IAEA, 2020d). One cited Russian study in the IAEA publication analyzed the current 30B transport container for shipment of enriched UF₆. Results of the analyses show that the 30B would be feasible for UF₆ transport up to 7 percent–enriched UF₆. For enrichments greater than 7 percent, additional work would be needed to build a safety case for transport using the 30B, or the packaging may need redesigning to accommodate more highly enriched UF₆. Depending on the supply chain needed to acquire HALEU, if the enrichment and fuel fabrication facilities are not collocated, the enriched UF₆ will have to be transported to the fuel fabrication facility, thereby potentially impacting the ability to ship UF₆ in currently licensed containers. Orano has submitted a license application to the U.S. NRC for a modified 30B container, designated the DN30-X, to ship UF₆ enriched up to 10 or 20 percent. The capacity of this container will be approximately 50 percent of the currently licensed 30B (Redmond and Ashkeboussi, 2022). An alternative to gas shipment is reconversion of the UF₆ gas to an oxide or metallic form for shipment to the fuel fabrication facility. Discussion with DOE indicates that this would probably be the plan for augmenting their HALEU supply

___________________

¹¹ See Appendix D for the definition of GTCC waste.

(Griffith, 2021; Regalbuto, 2021). There are currently three packages licensed for feedstock HALEU in the oxide or metallic form; however, the payloads of these packages are very limited, considering the quantities of HALEU feedstock that may need to be shipped (Redmond and Ashkeboussi, 2022). The supply chain aspect of acquiring sufficient HALEU to meet DOE’s needs in the advanced reactor programs requires assessment with respect to where the HALEU is sourced to determine potential impacts to program schedules.

From a domestic supply perspective, if shipment of enriched UF₆ in the 30B is required, current regulatory requirements (10 CFR 71.55[g]) limit exemption to the subcriticality requirements (10 CFR 71.55[b]) to contents enriched to a maximum of 5 percent. For feed material enriched to greater than 5 percent, the package would have to be evaluated with water inside the cask, or the applicant would need to request an exemption (U.S. NRC, 2020f). Furthermore, U.S. Department of Transportation (DOT) regulations (49 CFR 173.420) require that UF₆ packages be designed according to American National Standards Institute Standard N14.1, which only applies to enrichments up to 5 percent. Use of the 30B cask to transport enriched UF₆ in the United States will require advanced planning to assure transport capability when needed.

As mentioned above, feedstock material in either oxide or metallic form may need to be shipped from the enrichment facility to the fuel fabrication facility. The three currently available packages for these shipments, the ES-3100, TN-BGC1, and Versa-Pac VP-55, have limited capacities. It is anticipated that packages in the hundreds of kilograms will be needed to meet the demand for HALEU (Redmond and Ashkeboussi, 2022). Once fresh fuel with initial enrichments greater than 5 percent has been fabricated and needs to be shipped to the reactor in a fresh fuel packaging, criticality assessments will need to be made for the transport package. The U.S. NRC has stated that there is a lack of critical experiments¹² providing data to support criticality benchmarking analyses (U.S. NRC, 2020f) for transport packages containing material enriched above 5 percent. The U.S. NRC recommends addressing this issue by

• performing new critical experiments,
• extrapolating existing critical experiment data using uncertainty or sensitivity analyses,
• increasing safety margins due to lack of validation analyses, or
• a combination of the above three methods.

In support of the DOE advanced reactor initiatives, assessment of transport challenges was recognized at the INL Workshop on HALEU discussed in Chapter 4. As part of this workshop, industry feedback indicated the need for DOE support in development and licensing of a HALEU transport container (Caponiti, 2020a).

For storage of spent fuel containing HALEU, technical issues associated with criticality and higher temperatures will need to be evaluated. Criticality design generally relies on neutron poisons in either soluble or solid forms (or both) in the spent fuel pool or fixed poisons in the basket structure of dry storage containers (U.S. NRC, 2020e). Pool and dry storage of spent fuel with enrichments of greater than 5 percent will need to be assessed to quantify the impact on current designs and operations as they relate to maintaining subcriticality during storage. Thermally, the higher burnups expected for these fuels will result in higher temperatures in the spent fuel. Management of these heat loads may require longer in-pool storage before transferring to dry storage. In dry storage and drying operations during packaging, assessments will be needed to ensure peak temperatures of the fuel remain below designated limits.

For the back-end transport, all forms of spent fuel containing HALEU will require assessments prior to transport. All transport systems must be evaluated under hypothetical accident conditions to ensure that the used fuel payload will remain subcritical. With initial enrichments of up to 19.75 percent, transport packages may need to be redesigned to assure that subcriticality thresholds are maintained. HALEU spent fuel will also have higher burnups, which will result in higher heat and radiation loads. Thermal and radiation shielding analyses will be required and also may result in package redesign. In particular, for thermal reactors,¹³ neutron dose rates may increase significantly due to curium-242 or -244 content (IAEA, 2020d).

___________________

¹² Critical experiments use combinations of different geometries, fissile material (fissile uranium or plutonium), and moderator (water, plastic, etc.) to achieve a k_eff (effective neutron multiplication factor) of exactly 1.0. These data are then used to evaluate the accuracy of the code used to evaluate criticality safety by modeling a large number of critical experiments to determine statistically whether there is a bias for the code (i.e., whether the code consistently calculates a k_eff higher or lower than 1.0) (U.S. NRC, 2020f).

¹³ The sentence was modified following a prepublication version of the report to indicate the applicability to thermal reactors only, and not fast reactors.

5.6.2.2 Uranium Oxide Fuels for Integral Pressurized Water Reactors (iPWRs)

The NuScale iPWR Small Modular Reactor is identified as an advanced reactor concept and has progressed in U.S. NRC licensing reviews by receiving design certification in August 2020 (Reyes, 2021). The fuel technology for this small modular reactor is a standard LWR 17×17 PWR fuel design using UO₂ fuel pellets (<5 percent–enriched) housed in M5 cladding. As such, storage and transportation licensing is expected to be consistent with existing LWR applications and does not represent a change in licensing risk. Licensing experience for existing LWR storage and transportation systems suggests that review of new applications for similar small modular reactor systems should be consistent with this past experience.

Similarly, secondary waste facilities for low-level wastes, including GTCC, are envisioned at the plant site. Dry storage systems for these waste streams are envisioned to be consistent with existing licensed GTCC and low-level waste designs. As mentioned earlier, waste volumes may be a concern as multiple units per reactor island are envisioned. A careful assessment of these impacts is required to develop a clear picture of the scope of the waste stream.

5.6.2.3 TRistructural ISOtropic (TRISO) Particle Fuel and Graphite Materials

All of the advanced reactor systems proposing to use TRISO envision using HALEU source material with up to 19.75 percent–enriched uranium. From a storage perspective, the spent TRISO fuel provides a robust containment where the graphite cover acts as a protective barrier similar to the metal cladding used in LWR systems. Existing storage systems have licensed TRISO spent fuel in prismatic graphite blocks at the Fort St. Vrain Independent Spent Fuel Storage Installation (ISFSI) (Hall et al., 2019b). For TRISO in pebble form, storage can be designed using dry canisters with overpacks, similar to conventional dry storage systems for LWR fuel. However, proposed advanced designs will require additional technical assessments (Hall et al., 2019b). First, maintenance of subcriticality margins of TRISO fuel with higher enrichments (>5 percent) and higher burnups will need to be verified. Second, higher decay heat may require additional thermal analyses to verify thermal performance with the higher burnup spent fuel. Finally, fluorine salt–cooled reactors may leave residual salt on the TRISO particles, which may result in radiolysis generating fluorine gas, creating issues concerning both safety and corrosion on the containment package.

In a study of DOE-owned nuclear fuel, the Nuclear Waste Technical Review Board (NWTRB) identified a number of technical issues related to the storage and disposal of the TRISO fuel generated by the Fort St. Vrain reactor (NWTRB, 2017): The extended storage of carbide-based fuels may require careful attention to aging-management issues in order to successfully recover and treat these fuels. Degradation and corrosion processes associated with a carbide-based fuel are different enough from those of an oxide-fuel that they require special investigation. In particular, careful attention must be paid to the possibilities of (1) gas generation; (2) mechanisms of waste form degradation, dissolution, and precipitation of specific radionuclides; (3) solubility limits of radionuclides; and (4) sorption/desorption reactions of radionuclides onto degradation products of the waste package components, to name a few. These identified technical issues illustrate the importance of considerations across operational boundaries—in this case, spanning storage and disposal operations.

As discussed above, management of residual and used graphite will need to be addressed in terms of degradation characteristics in long-term storage environments. Reactors using graphite as moderators and reflectors will produce large quantities of spent graphite over the course of reactor life; therefore, the development of graphite disposal technologies needs attention.

From a transportation perspective, spent TRISO fuel does provide a robust primary containment that will likely survive normal and loadings for hypothetical accident conditions as defined in 10 CFR 71. The Fort St. Vrain transportation cask (TN-FSV) has been licensed to transport TRISO fuel (Hall et al., 2020). Structural, thermal, criticality, and shielding considerations satisfy current 10 CFR 71 regulations. However, the proposed higher initial fuel enrichments suggest that criticality analyses will need to be conducted to assure subcriticality margins in the 10 CFR 71.55(e) nonmechanistic loading requirement to ensure payload subcriticality. Alternatively, a moderator

exclusion approach could be assessed that would include a double seal to prevent water ingress. Higher burnups will likely require packaging thermal analyses to ensure threshold temperatures are not exceeded on the cask surface.

In general, design of storage and transportation systems for TRISO spent fuel and associated secondary waste streams should be achievable under the current regulatory structure. However, it will be important to identify and assess higher enrichments and burnups, as well as degradation characteristics of these wastes that may affect the functional characteristics of the designed storage and transportation systems.

5.6.2.4 Metallic Fuels and Materials

TerraPower, ARC, LeadCold, and Oklo have proposed advanced fast reactor designs that use metallic fuel with sodium or lead as the liquid coolant (DeWitte, 2021; Hejzlar, 2021; Neider, 2021; Sackett and Arthur, 2021). Metallic fuel proposed for advanced reactors requires special consideration. The storage experience at INL indicates that issues that may affect the integrity of sodium-bonded fuel have not been fully addressed (Hall et al., 2019b). Initial storage (pool or dry), followed by long-term storage in a confinement package, needs to demonstrate a long-term ability to resist moisture degradation of both the confinement package barrier and the fuel components. Long-term performance characteristics of any proposed sodium-bonded fuel will need to be developed, taking into account the storage environment (temperature, humidity, etc.) along with chemical, radiological, and mechanical interactions between the various materials (metallic fuel, sodium layer, and stainless-steel cladding) that make up the fuel and surrounding confinement boundary. Individual fuel types with specific storage system designs will need confirmation testing and analyses to validate degradation characteristics of the fuel and storage system in order to assess performance against the regulations. While still in the R&D phase, Oklo is partnering with INL through an Advanced Research Projects Agency-Energy project to develop a bondless metallic fuel (DeWitte, 2021). All sodium-bonded spent fuel will have to undergo some form of cleaning/stabilization/reprocessing in order to make it safe for long-term storage. Storage at INL includes fuel that was chemically treated to deactivate the sodium and then was converted to ceramic or metallic HLW. The INL inventory also includes untreated sodium-bonded metallic spent fuel. This indicates the difficulty in managing the spent fuel as a waste form once the fuel is removed from the reactor. The operational difficulties arising from stabilization of this form of spent fuel to prepare it for long-term storage, transportation, and eventual disposal point to the need for reliable cost estimating to treat the spent fuel. Treatment estimates then need to be factored into the total life-cycle costs in order to provide a realistic perspective of total costs associated with the back-end management of the spent fuel.

These advanced fuels also propose using HALEU to increase initial enrichments up to 19.75 percent. As with TRISO fuel, criticality assessments will need to be conducted. For designs with high burnup, thermal analyses will have to be performed to verify thermal performance.

Secondary wastes from the metal coolant need to be considered. Current planning and design is such that it is too early to provide specific assessments concerning these wastes. However, development of plans to safely deal with the storage of these wastes need to be addressed as early as possible. The manner in which residual and excess metal is processed, cleaned, and stored needs to be developed and defined for these systems. Waste forms and packaging systems then will need to be assessed in terms of degradation characteristics relative to the storage environments.

There is a limited history of transporting sodium-bonded metallic fuel in the DOE complex (Hall et al., 2020). The T-3 cask was used to ship FFTF fuel from Hanford, Washington, to Idaho, and EBR-II fuel was shipped intrasite at INL facilities using both the NAC LWT and TN-FSV casks. These previous shipments provide evidence of licensability for transport cask designed to ship metallic spent fuel. As with storage considerations, the condition of the fuel and any confinement barrier will need to be confirmed prior to shipment in order to assess its ability to withstand both normal and hypothetical conditions of transport. As with other HALEU fuel, higher initial enrichments will require assessment to assure criticality margins meet the regulations under criteria for hypothetical accident conditions. Design burnups as high as 150 MWd/MTU (megawatt-day/metric ton of uranium) are proposed, and thermal analyses will need to be conducted to verify thermal performance of the fuel and its packaging against transport conditions.

Degradation characteristics for transportation are less of an issue than for storage, in that transportation is a short-term event. However, the integrity and condition of the waste form must be verified prior to shipping in order to assure the integrity of the waste during the shipping event.

5.6.2.5 Molten Salt Liquid Fuels

Terrestrial Energy, ThorCon, Flibe, TerraPower, and Moltex all have advanced reactor designs that use fuel as part of a molten salt mixture (Jorgensen, 2021; Latkowski, 2021; LeBlanc, 2021; O’Sullivan, 2021; Sorensen, 2021). Terrestrial Energy, ThorCon, and Flibe use a fluoride salt, and TerraPower and Moltex use a chloride salt. In general, there is a limited experience base for designing, licensing, constructing, and operating the back end of MSR fuel cycles.

Planning for back-end management of spent fuel and secondary wastes has been initiated with some fairly detailed specifics on waste form characteristics and on the secondary waste stream definition. All concepts involve some form of recycling, conversion, and/or stabilization to process the wastes into forms that can maintain integrity during long-term storage, transportation, and disposal conditions.

From a storage and transportation perspective, licensing of systems that provide confinement/containment, shielding, criticality control, and thermal management will have to be assessed via the current regulatory structure. Given the state of the development of these types of advanced reactors and the lack of operational experience of commercial-scale reprocessing, R&D is needed beyond conceptual thinking to better define waste form isotopics and concentrations, hazardous constituent assessments, degradation characteristics, and overall performance relative to the storage and transportation environments to which they will be subjected. Numerous options are available for processing, from converting the spent fuel to a stable solidified block, to separating the fuel from the salt and reprocessing and cleaning of the salt for reuse. However, little practical experience exists to identify potential issues with implementation at scale. The thermal spectrum reactors also use graphite for a moderator, which will have to be managed as a secondary waste. Issues such as the presence of carbon-14 and daughter products of xenon and krypton will have to be addressed. Much work is needed to develop these concepts to a point where an assessment of the impact on storage and transportation can be properly assessed.

5.6.3 Summary: Storage and Transportation of Advanced Fuels

The current storage and transportation regulatory structure will accommodate review and licensing of advanced reactor fuel cycle materials and systems. As has been shown in the past, considerably more work may be needed in some areas to demonstrate compliance with the regulatory criteria. Because many of the proposed fuels will use HALEU with initial enrichments up to 19.75 percent, criticality becomes an especially important safety function that may require additional testing and analyses to demonstrate compliance with the regulations for these advanced material waste forms. Validation that criticality margins will be satisfied may need additional testing to develop a database of the physical, chemical, and isotopic makeup of the waste form, along with analyses to validate the criticality margins.

Thermal management of proposed high burnup fuels may require quantification of the thermal environment and response characteristics of waste materials and packages to this environment.

For metallic fuels, R&D is ongoing to define how best to prepare the spent fuel for long-term storage in a way that best stabilizes the spent fuel. The wet and dry storage experiences at INL (Hall et al., 2019b) indicate that issues associated with degradation of the fuel need to be addressed. Dissimilar metals, sodium, moisture, and temperature all act as catalysts in potential degradation processes. For transportation of these fuels, degradation is less of an issue because of the short duration of the transportation event. However, this does point to the need to know the condition of the spent fuel prior to shipment. Typical plans call for the canisterization of the spent fuel for storage. This package would then be used for transport after storage, making verification of fuel integrity difficult. The spent fuel could be in storage for many decades or conceivably a century; thus, it is essential to characterize the spent fuel’s condition over this long time.

Management of liquid spent fuel is still in a development stage, with significant issues needing to be

addressed on how best to treat the spent fuel. As this technology becomes better defined, waste forms will need complete characterization of physical form, chemical composition, and radiological characteristics. Based on these characteristics, waste form and packaging degradation processes will need to be addressed to understand how the waste form may degrade in long-term storage conditions. The waste form characteristics, along with the understanding of degradation processes, will provide the information needed for licensing storage and transportation systems.

Secondary wastes from these advanced fuel concepts that are not generated in LWR operations include graphite, salt, lead, and sodium. This inventory includes hazardous wastes, along with GTCC and low-level wastes. While hazardous wastes are not specifically covered in 10 CFR 71 and 72, DOT regulations (49 CFR 173) will need to be followed. Packaging of these materials for storage and transportation is not seen as a difficult licensing problem, but work is needed to define the waste form characteristics, along with quantities generated.

Storage and transportation operations in the context of this study represent a link between reactor power generation and disposal of resultant wastes. While important, these operations are not the primary drivers in the licensing, technical, or economic decision-making required for the development of advanced fuel cycles. No “show-stopper” storage and transportation issues have been identified that would indicate a reason to second guess the development of any of these proposed concepts. However, significant work is needed in some areas to properly characterize waste forms in a way that quantifies response characteristics to the storage and transportation environments. In some cases, this will require significant testing and R&D analyses to properly understand waste form and packaging performance in relation to the storage and transportation environment to which they are subjected. Management and storage of these new spent fuels will in some cases entail additional costs significantly above those associated with existing LWRs.

In general, the NWTRB determined regarding DOE-owned spent nuclear fuel, “Given the uncertainty in how long DOE SNF [spent nuclear fuel] will be stored prior to disposal, the Board finds that having the ability to measure and monitor conditions of the SNF inside canisters, the external surfaces of canisters, and the storage facility itself during future storage is an important consideration in designing, developing, and deploying new DOE storage systems, such as a DOE standardized canister, and for new packaging and storage facilities” [emphasis added] (NWTRB, 2017).

5.7 AN OVERVIEW OF DECONTAMINATION AND DECOMMISSIONING OF NUCLEAR POWER PLANTS

5.7.1 Background Status

Decontamination and decommissioning (D&D) of nuclear facilities around the world has a long history to support an ever-increasing demand, as the global nuclear industry is facing a fleet of aging plants and early plant shutdowns for economic reasons. The 414 reactors operating worldwide as of March 2022 have an average age of 31 years (WNISR, 2022).

The average age of the 93 currently operating reactors in the United States is 39 years (EIA, 2021c), and 21 commercial power plants have been decommissioned or are currently decommissioning (U.S. NRC, 2021d). The United States’ substantial background in D&D began with the Shippingport plant in Pennsylvania. As the world’s first full-scale commercial nuclear power plant, Shippingport began producing power in 1957 and continued producing power until its shutdown in 1982 (Crimi, 1995). The site was released for unrestricted use after D&D activities were completed in 1989. However, the Beaver Valley, Units 2 and 3, nuclear generating station located on the same site is still operating. The Shippingport reactor vessel was shipped by barge to the Hanford Site for burial. The spent nuclear fuel is stored at the Hanford Site Canister Storage Building in 18 canisters (NWTRB, 2020). Low-level wastes were treated on site, packaged, and shipped to low-level waste burial sites.

A more recent example of successful D&D is the Trojan Nuclear Power Plant. The Trojan plant, located on the Columbia River in Oregon, operated from 1975 to 1992, and D&D was completed in 2008 (Nuclear Decommissioning Collaborative, 2022). The reactor vessel was barged up the Columbia River for burial at the Hanford Site. Different from the Shippingport solution, spent nuclear fuel at Trojan is stored on site at a licensed Independent Spent Fuel

Storage Installation (ISFSI) in 34 dry storage casks (Oregon DOE, n.d.). Without a national solution to a repository at this time, this has become a common way for utilities to deal with their spent nuclear fuel at shutdown sites.

As discussed in Section 5.1, low-level waste generated from current LWR power reactor operations is treated, stored, and shipped to one of four licensed low-level waste disposal facilities in the United States.

The operational legacy of D&D provides a basis for consideration of how best to manage both high- and low-level wastes from advanced reactors and fuel cycles. This experience, with a focus on LWR operations, also provides a basis for identifying gaps that will have to be considered in management of D&D associated with advanced reactors and fuel cycles.

5.7.2 International Guidance and U.S. Regulatory Status

The IAEA, which is not a regulator, provides international guidance on D&D; it formally began work in this area in 1973, publishing its first document on D&D issues in 1975 (IAEA, 1999b). Since then, dozens of IAEA reports have been published assessing the technological feasibility of safely conducting D&D operations. These assessments include evaluating specific waste streams, treatment of wastes, and methods to conduct the actual dismantling operations and sizing of the debris for further processing, handling, and packaging. Two reports in the form of “TECDOCs” were published covering information exchange, experiences in actual D&D activities, and lessons learned from these activities (IAEA, 1989, 2018a). More recently, the IAEA published a state-of-the-art report for specific technology processes for decontamination and dismantling of radioactive facilities (IAEA, 1999b). This work reflects an ongoing effort to assess D&D of nuclear facilities in light of the continual D&D activities in the industry as plants continue to age and go into shut-down status.

The IAEA has also published a safety guideline (IAEA, 2018a) that provides guidance for meeting safety requirements applicable to decommissioning nuclear facilities. This guide is more of an administrative roadmap for regulators and licensees to follow in decommissioning planning as opposed to the technical documents identified previously. The safety guide also provides a comprehensive list of other IAEA publications relevant to decommissioning.

In the United States, the U.S. NRC continues to assess D&D activities of licensed plants that have shut down. Currently, three decommissioning strategies are available to licensees: DECON, SAFSTOR, and ENTOMB. DECON refers to dismantling and removal of all plant structures and radioactive contaminated material to a level that would permit release of the property. SAFSTOR allows for maintenance of the facility prior to decommissioning while radioactivity of spent fuel and contaminated materials decays and decommissioning funds are amassed. Once a plant is ready, it moves from SAFSTOR to DECON. The plant is then dismantled and decontaminated. ENTOMB allows for on-site burial of encased radioactive materials and components in structurally sound containment barriers, such as concrete. The property can be released for restricted use after radioactive decay reaches a certain level defined in the license. To date, no U.S. NRC–licensed nuclear power plants have used this option (U.S. NRC, 2020m). In the SAFSTOR and in the DECON strategy as currently practiced, spent nuclear fuel is stored on-site at a licensed ISFSI facility. U.S. NRC regulations associated with D&D are promulgated in 10 CFR 20(E) and 10 CFR 50.75/82/53/96.

As of December 2021, 26 power reactors were under DECON or SAFSTOR licensing protocols (U.S. NRC, 2022c). Thirteen reactors are using the DECON strategy: Humboldt Bay 3, San Onofre 1/2/3, Zion 1/2, LaCrosse, Crystal River, Fort Calhoun, Oyster Creek, Pilgrim, Vermont Yankee, and the N.S. Savannah. Thirteen reactors are using the SAFSTOR strategy: GE EVERSR, GE VBWR, Millstone, Dresden, Duane Arnold, Fermi 2, Indian Point 1/2/3, Peach Bottom, Three Mile Island 1/2, and Kewaunee. Furthermore, 10 power reactors have completed decommissioning: Rancho Seco, Fort St. Vrain, Haddam Neck, Maine Yankee, Yankee Rowe, Big Rock Point, Shoreham, Trojan, Saxton, and Pathfinder.

This level of D&D activity at U.S. NRC–licensed facilities provides reasonable assurance that D&D can be carried out properly at nuclear facilities, with the site returning to unrestricted use pending completion of the D&D. For the U.S. fleet of LWRs using low-enriched uranium oxide fuel, plants can be properly dismantled, separating the uncontaminated components and materials from the contaminated materials and spent nuclear fuel. Uncontaminated materials can be recycled or disposed of in landfills. For all U.S. sites that have fully decommissioned, there is still

no pathway to disposal for their spent fuel, and it remains on site. As a result, the largest challenge to the ability to return sites completely to unrestricted use is to establish one or more geologic repositories for spent fuel disposal.

5.7.3 Advanced Reactors and Fuel Cycles

The proposed technologies for advanced reactors include fuel and materials that are markedly different from the current LWR fleet in the United States. Fast reactors using sodium, lead, or molten salt coolants coupled with metallic, TRISO, and liquid fuels in molten salt will create challenges in terms of D&D technologies and operations. As opposed to LWR operations, this new generation of reactors will create secondary wastes that may be particularly challenging in terms of stabilizing, packaging, storing, and disposal. Sodium and lead coolants, sodium-bonded fuels, molten salts, and graphite moderators and reflectors will create wastes that will need special attention and probably additional R&D to safely process and stabilize, adding costs to storage not experienced with LWRs. Characterization, processing, and volumes of materials after reactor shutdown remains unclear at this point. These waste streams point to the gap in the operational legacy mentioned earlier.

The IAEA has published a preliminary assessment of expected types of wastes generated from advanced reactors (IAEA, 2019a). This document provides a good general assessment of the specific types and volumes of wastes that can be expected from advanced water-cooled reactors; gas-cooled reactors; liquid metal reactors; and dedicated actinide burners, such as molten salt reactors.

Noncontaminated components of advanced reactors can be disposed of via existing approved methods. As with LWR spent fuel, the spent nuclear fuel from advanced reactors will have to be stabilized, packaged, and stored until a disposal path is available for the United States. Secondary wastes generated in these proposed advanced reactors will need careful consideration as to how they can best be managed as they are generated and removed from the reactor system. In some cases, R&D will need to be performed to understand degradation processes of these wastes and how they will interact with their environments (e.g., long-term storage in licensed dry storage systems and assessment of activation products in graphite). Wastes generated from spent fuel processing will also have to be identified and addressed (e.g., cleaning of TRISO fuel used in molten salt coolants).

Lessons learned from existing experiences can provide valuable insights into specific issues associated with managing some of these materials (Hall et al., 2019b). Notably, for many of the past “advanced” designs in the United States, the spent fuel remains incompletely treated. This is true of the sodium-bonded spent fuel from various sodium-cooled fast reactor experiments, as well as HTGR spent fuel (from Fort St. Vrain) and spent fuel from the Molten Salt Reactor Experiment at Oak Ridge.

For example, INL has ongoing experience in the long-term storage of sodium-bonded metallic fuel. Issues with early pool storage have led to corrosion of some of the fuel. While this fuel is being treated and moved to dry storage, insights into early management of this fuel once it is removed from the reactor can provide important guidance for proper processing and storage to prevent such problems in advanced reactor operations.

The Fermi-1 sodium-cooled breeder reactor has been shut down and is in the SAFSTOR licensing process under the U.S. NRC, with final site closure slated for 2032. Currently, all spent nuclear fuel, bulk sodium, and the reactor vessel, as well as other major components have been removed from the Fermi site (U.S. NRC, 2021e). The spent fuel and blanket subassemblies were shipped to INL for long-term storage. The secondary nonradioactive sodium was shipped to a commercial chemical company for reuse (U.S. NRC, 2021e). The primary sodium was stored on-site in tanks and 55-gallon drums until it was shipped to INL (Argonne-West) in the early 1970s. The sodium was treated for long-term storage at INL (Sherman and Knight, 2005). INL is still in the process of cleaning and remediating these storage tanks (U.S. NRC, 2020i). This example points to the historical difficulties in dealing effectively with sodium-bearing wastes, which will need to be addressed through additional R&D.

Fort St. Vrain, an HTGR, has been fully decommissioned except for a licensed ISFSI on-site to store its spent fuel. This provides one example of a licensed power-producing advanced reactor that has been decommissioned according to U.S. NRC regulations. Many of the plant’s components and systems have remained in place as the facility was converted to a natural gas plant. With the exception of the spent fuel, components categorized as low-level and secondary low-level wastes were treated and appropriately disposed in low-level waste disposal

facilities (Fisher et al., 1996). As with LWR plants, the challenge for the United States is to operationalize a final pathway for the spent fuel for HTGRs.

As a secondary waste, the use of graphite for moderation and as a reflector is a component of several of the proposed advanced reactors; this will result in a sizable amount of waste that will need to be disposed of after reactor operations cease. The IAEA has recognized the significance of this problem and in 2006 published a TECDOC identifying the technical issues and challenges in managing graphite as a back-end product of reactor operations (IAEA, 2006). At the time of the publication, there were more than 230,000 MT of graphite needing disposition. A single full-size reactor can result in up to 3,000 MT of graphite. Tritium, carbon-14, and chlorine-36 are all generated and need to be carefully considered in any back-end management plan. The IAEA also recently designated Collaborating Centre in France for graphite reactor decommissioning (Kilochytska, 2021), which will fund R&D to advance technologies in the management of graphite disposition.

Other secondary wastes—such as molten salt used for cooling, processing of liquid fuels, and off-gassing—will need to be properly characterized and addressed in terms of back-end management in order to provide a safe, complete fuel cycle all the way to disposal. From a SAFSTOR perspective, advanced fuels using HALEU enriched up to 19.75 percent will need to be stored pending a resolution on a U.S. repository. Issues associated with criticality and thermal management due to the high burnup of most of the fuel designs will need to be considered early in the design process.

For advanced reactor concepts that are considering reprocessing to close the fuel cycle, the West Valley Demonstration Project (WVDP), classified as a Complex Decommissioning Site by the U.S. NRC, provides an example of D&D challenges for a spent fuel reprocessing facility (U.S. NRC, 2021f). The West Valley Plant operated from 1966 to 1972. It is still in an active state of D&D, with no firm date for completion (estimated completion date, 2040). Under the 1980 WVDP Act, Congress directed DOE to assume possession of the major components of the facility, including the reprocessing plant, the U.S. NRC–licensed disposal area, the high-level waste tanks and waste lagoons, and the aboveground storage areas. Furthermore, the Act authorized DOE to “solidify, transport and dispose of HLW [high-level waste] that exists at the site, dispose of LLW [low-level waste] and transuranic waste produced by the WVDP, and decontaminate and decommission facilities used for the WVDP in accordance with requirements prescribed by [the U.S.] NRC” (WVDP Act, 1980 [Public Law 96-368]). The liquid high-level waste produced from reprocessing has been vitrified, packaged, and stored on-site. While the vitrification facility has been demolished, the Main Plant Process Building is conducting deactivation activities to complete removal of fixed plutonium in order to move forward with open air demolition. WVDP costs up to 2013 were $2.25 billion. Estimated costs from 2013 to completion were an additional $2.73 billion (Rieman, 2013).

For active designers of advanced reactors and fuel cycles, it is important to address these issues up front in order to provide a complete picture of what will be required from a technical, safety, and economic standpoint in fielding these systems and in providing responsible back-end solutions to waste disposition.