3

Potential Merits and Viability of Advanced Nuclear Reactors and Associated Fuel Cycles

This chapter reviews the current U.S. government and Generation IV International Forum programs for the development of advanced nuclear reactors and associated fuel cycles in response to the first charge of the statement of task, which calls for an evaluation of the merits and an assessment of the viability of different nuclear fuel cycles, including fuel cycles that may use reprocessing, for both existing and advanced reactor technology options. The evaluation and assessment of existing options for operating commercial light water reactors (LWRs) were described in Chapter 2.

In this chapter, the committee provides its summary, findings, and recommendations up front (Section 3.1) and then describes international cooperative activities and the different types of advanced reactors (Section 3.2), programs to support their development (Section 3.3), and available and needed infrastructure for prototyping and testing these developing technologies (Section 3.4). Given the task to the committee to consider technologies that could be deployed by 2050, the committee provides perspectives, based on its information gathering and collective knowledge and judgment, on the status of technological development for each advanced reactor type. (Chapter 4 provides more details on the development and infrastructure needed to support fuel cycles for advanced reactors.) The committee notes that an assessment of the cost competitiveness of advanced reactor systems as compared with other energy sources is beyond the scope of this study.

3.1 CHAPTER 3 SUMMARY, FINDINGS, AND RECOMMENDATIONS

The Generation IV International Forum categorizes advanced reactor systems as including the following: (1) very-high-temperature reactor, (2) gas-cooled fast reactor, (3) sodium-cooled fast reactor, (4) lead-cooled fast reactor, (5) molten salt reactor, and (6) supercritical water-cooled reactor (SCWR). The integral pressurized water reactor, a small power reactor that leverages the technological infrastructure of the existing large-power pressurized water reactors, is also considered an advanced design according to the definition given by the Nuclear Energy Innovation and Modernization Act of 2019 (NEIMA) (Public Law 115-439). The U.S. government currently supports research and development (R&D) for all of these systems except the SCWR because of their commitments to support only those systems relevant to domestic R&D funded by the U.S. Department of Energy’s Office of Nuclear Energy (DOE-NE) or with U.S.-based industry interest. Most of the advanced reactors being developed in the United States are small modular reactors, defined as having a notional power output of less than 300 MWe (megawatts electric) and envisioned for factory construction and modular installation. Almost all developers told

the committee that they are planning on an open, once-through fuel cycle at least for the near to intermediate terms, although several developers noted that in the longer term, their technologies have the potential of recycling spent nuclear fuel.

Considerable additional work and time are required before any of the non-LWR concepts could reach commercial deployment, and this chapter outlines the challenges faced by reactor developers. DOE-NE has programs that support the development of the advanced reactor and associated fuel cycles, but it will have to make difficult decisions in the coming years based on budgetary constraints. Finally, while the advanced reactors under development have several potential merits, these have yet to be demonstrated, and there are no operating prototypes in the United States for any of these reactors.

Finding 3: Government support to help bring advanced reactor technologies to commercial deployment will take substantial financial and technical resources. Specifically, budget limitations will require the U.S. Department of Energy (DOE) to make difficult decisions about its advanced reactor research and development programs to guarantee support, via industry cost sharing, for a few promising advanced reactor technologies and associated fuel cycle infrastructure in the next several years. If the Advanced Reactor Demonstration Program is funded consistently and fully by both the government and private industry through completion, information such as costs, reliability, project management, and manufacturing feasibility gained from this program will be key to helping DOE in its decision-making process.

Finding 4: Most of the advanced reactors, especially the non–light water reactors, will confront significant challenges in meeting commercial deployment by 2050. While at least 10 advanced reactor developers currently aim to deploy their technologies by 2050 in the United States,¹ there are no currently operating fueled prototypes of any of these specific advanced reactor designs in the United States; there are, however, some demonstration and commercial units of similar reactor designs in operation internationally. Moreover, the vast majority of advanced reactors are still in the early design phase. Depending on the maturity of the technology, advanced reactor developers face a range of challenges to bringing the proposed technologies to commercialization, including little or no direct operational experience of some designs at engineering scale; the lack of adequate capabilities to develop, test, and qualify advanced fuels and materials; and as a result, the potential considerable time for regulatory approval.

___________________

¹ As of January 2022, this number of developers had submitted applications or preapplications to the U.S. Nuclear Regulatory Commission.

Finding 5: Of the advanced nuclear reactor technologies currently in development, small modular reactors based on light water reactor (LWR) technologies are furthest along toward being connected to the electrical grid. This is because they can leverage the existing LWR and fuel cycle infrastructure and because these technologies have received government and private-investor financial support for more than a decade.

Finding 6: The common perception that the thorium-232/uranium-233 fuel cycle will generate less plutonium and minor actinides (therefore reducing the radioactive hazard of its spent fuel compared with that from the uranium-235/plutonium fuel cycle) is incorrect. Overall, because of the decay of associated actinide products, thorium-based fuels have short- and long-term radiotoxicities (hazards) comparable to uranium-based fuels.

3.2 TYPES OF ADVANCED REACTORS AND ASSOCIATED FUEL CYCLES

A large number of advanced reactors are being designed that differ significantly from the current fleet of thermal LWRs deployed in the United States. Like LWRs, the advanced reactors will mostly use the fission of uranium-235 in nuclear fuel. Unlike LWRs, many advanced reactors will rely on non-water-based coolants, such as gases (e.g., helium, carbon dioxide), liquid metals (e.g., sodium, lead), and molten salts (e.g., fluoride-, chloride-based salts). Some of the advanced concepts depend on first transforming the fertile isotope thorium-232 to the fissile uranium-233, and then extracting energy from the fission of uranium-233. Advanced reactors will sustain the fission reaction with either neutrons moderated to thermal energies (less than 1 eV [electron volt])² or unmoderated systems that use fast neutrons (energies from ~0.1 to 1.0 MeV [mega electron volt]). Thermal reactors need moderators to slow down neutrons to thermal energies, and usually light water, heavy water, or graphite blocks are used as moderators.

Fast reactors were originally designed to use plutonium-239 as their primary fuel because, compared with uranium-235, plutonium-239 has a smaller capture-to-fission ratio for fast neutrons, produces more neutrons per fission regardless of the neutron energy, and generates more neutrons per neutron absorbed. Although the United States has a supply of plutonium-239 declared excess to its nuclear weapons program, it has no plans to use this supply for fast reactors and will most likely not reprocess spent nuclear fuel to produce plutonium for fast reactors for the foreseeable future. Thus, fast reactors will have to rely on uranium-235 for their fuel, except for those that could use uranium-233 produced from a thorium fuel cycle.

To sustain the chain reaction, fast reactors will require higher enrichments of uranium-235 to counteract both the much lower probability of fission of uranium-235 induced by fast neutrons and the probability of fast neutron capture by uranium-238. As a result, these reactors will need at minimum high-assay low-enriched uranium (HALEU) with uranium-235 enrichment greater than 10 percent but less than 20 percent. (See Chapter 4 for a discussion of HALEU production infrastructure and supply.) Fast reactors can operate as burners, converters, or breeders based on the conversion ratio (CR), or ratio of fissile material produced to material consumed. Burner reactors consume more fissile material than they produce (CR <1); converter reactors produce about the same amount of fissile material as they consume (CR ≈1); and breeder reactors produce more fissile material than they consume (CR >1).

Fuel options for advanced reactors include solid fuel composed of uranium, thorium, or uranium/plutonium in physical forms, such as refractory ceramics (predominantly oxides or mixed oxides and nitrides, carbides, or silicides) and metals (typically metal alloys), or as liquid fuel (fluoride or chloride salts of uranium, thorium, and plutonium) used in molten salt reactors. Solid fuel can have different shapes, such as fuel rods, pebbles, and cylindrical fuel compacts. A fuel rod consists of fuel pellets enclosed by protective cladding. Instead of fuel pellets, the fuel can be

___________________

² The ideal Maxwell–Boltzmann distribution of thermal neutron energies has its mode, or most probable energy, at 0.025 eV, which corresponds to a temperature of 20℃, or typical room temperature.

in the form of coated fuel particles (e.g., TRistructural ISOtropic [TRISO] particles). Coated fuel particles can be arranged into cylindrical fuel compacts that are inserted into prismatic graphite blocks, forming fuel elements (e.g., prismatic graphite block, as in Framatome’s Steam Cycle High Temperature Gas-Cooled Reactor [SC-HTGR]) or pebbles (as in X-energy’s Xe-100 pebble-bed reactor). (Chapter 4 provides details on the advanced reactors’ fuels and fuel fabrication.) Coolants are required to remove heat from the reactor core; common coolant materials are water, gas (helium, carbon dioxide, nitrogen), liquid metal (lead, sodium), and liquid salt (FLiBe [2⁷LiF-BeF₂]). Fuel assemblies and fuel rods are usually arranged in a square or hexagonal lattice, or randomly, as in pebble-bed reactors.

3.2.1 Generation IV International Forum’s Advanced Reactor Systems

In 2000, the U.S. government proposed the formation of the Generation IV International Forum (GIF) as a mechanism to promote international cooperation in the R&D of advanced nuclear energy systems. As of 2022, the GIF member countries are Argentina, Australia, Brazil, Canada, China, France, Japan, the Republic of Korea, Russia, South Africa, Switzerland, the United Kingdom, the United States, and Euratom (representing 27 European Union member countries).

GIF provides a framework for categorizing the major classes of advanced reactor systems. The Generation IV (Gen IV) framework was defined from 2001 to 2002 via identifying promising reactor technologies to examine, establishing technology goals, and setting a legal framework for cooperation. An expert group considered more than 100 designs and then selected the six most promising systems for further R&D. These design systems are (1) the gas-cooled fast reactor, (2) the lead-cooled fast reactor, (3) the molten salt reactor, (4) the sodium-cooled fast reactor, (5) the supercritical water-cooled reactor (SCWR),³ and (6) the very-high-temperature reactor (VHTR)⁴ (see Figure 3.1).

The goals for the Gen IV systems are

• Sustainability for the long-term fuel supply and for minimization of waste and the long-term burden of this waste;
• Safety and reliability to achieve a very low likelihood and degree of reactor core damage and to eliminate the need for off-site emergency response;
• Economics that have life cycle cost advantage as compared to non-nuclear energy systems and financial risk comparable to other energy systems; and
• Proliferation resistance and physical protection to have unattractive materials diversion pathways and enhanced physical protection against terrorist attacks (Kamide, 2021).

The U.S. government supports R&D for five of the six Gen IV systems; it does not support R&D for SCWR because DOE-NE has committed to support only those systems that are relevant to DOE-NE’s funded domestic R&D programs or have U.S.-based industry interest (Caponiti, 2022). Progress related to the development of the six reactor systems along with the cross-cutting R&D activities under the purview of GIF task forces and working groups can be found in the Gen IV International Forum Annual Reports (GIF, 2022b).

DOE’s framework for advanced reactors designs is similar to that of GIF, but DOE more highly prioritizes versatility compared with earlier reactor generations, especially the ability to provide nonelectrical services, such as desalination, process heat, and hydrogen production, as an additional goal of the advanced reactor designs. Like the Gen IV systems, DOE’s advanced designs have potential improvements that could manifest themselves in a number of ways, such as inherent or passive safety features, simplified or modular designs for ease of fabrication,

___________________

³ SCWR is a reactor concept for operating at supercritical pressures with light water as coolant and moderator. Supercritical water has indistinguishable densities for liquid and steam, and thus, SCWRs can eliminate the need for pressurizers and steam generators (as are used in pressurized water reactors [PWRs]). Consequently, an SCWR is simplified compared with a PWR and can operate at higher thermal efficiencies—approximately 45 percent as compared with 33 percent for a PWR. The SCWR concept is not considered in this report.

⁴ According to GIF, “The VHTR is a next step in the evolutionary development of high-temperature gas-cooled reactors. It is a graphite-moderated, helium-cooled reactor with thermal neutron spectrum. It can supply nuclear heat and electricity over a range of core outlet temperatures between 700 and 950°C, or more than 1,000°C in future” (GIF, 2022a).

scalability and enhanced load-following capabilities to complement sources of renewable energy, increased safety of accident-tolerant materials, and fast neutron spectrums for increased fuel utilization via closed fuel cycles (Arostegui and Holt, 2019). However, it should be noted that the Gen IV designation refers strictly to the six designs meeting the stated goals of GIF, and some DOE-supported advanced reactor designs do not fall under the Gen IV classifications. In DOE’s broader definition, advanced reactors are further subdivided into categories of microreactors, small modular reactors, and full-sized reactors, described as follows:

• Microreactors are factory fabricated and fully assembled; transportable by truck, ship, railcar, or air to residential, remote commercial, or military locations; installed and operational quickly; and self-adjusting with passive safety systems and small support staff (DOE-NE, 2021b).
• Small modular reactors are factory fabricated; can transport major components from factory fabrication locations to the plant site by rail or truck; and can include designs that provide safeguards, security, and nonproliferation advantages. Both advanced water-cooled and non-water-cooled reactors (gas, liquid metal, and molten salts) can be configured as small modular reactors (DOE-NE, 2021c).

Most of the advanced reactors being developed in the United States are small modular reactors (SMRs). In addition to the potential benefits described by DOE, motivations for SMR development include smaller plant footprint and potentially lower operation and maintenance costs, and lower up-front capital costs (excluding the to-be-expected higher costs of the first-of-a-kind unit), as well as improved safety. However, because SMRs have yet to be commercially deployed in the United States, these systems have to demonstrate their operational economic competitiveness compared with larger nuclear power plants and other nonnuclear energy systems.⁵ The non–light water SMRs will also need to demonstrate that they can meet licensing requirements that include safety assessments by the U.S. Nuclear Regulatory Commission (U.S. NRC).

All the Gen IV systems except for the SCWR, which is not supported by DOE, are described in Section 3.2.3. Also included in Section 3.2.3 is a relatively detailed description of NuScale’s light water–cooled SMR, which represents one of the most technologically mature light water SMRs currently under development.

3.2.2 International Collaboration and Partnerships

In addition to supporting the development of almost all of the Gen IV systems, DOE has cooperative agreements with other international partners through its Office of International Nuclear Energy Policy and Cooperation (INEPC). Relevant to this study, INEPC has worked with countries such as France, Japan, and Russia investigating advanced nuclear fuels. As discussed in Section 3.5, cooperative activities involve use of materials testing reactors and other facilities for evaluation of advanced reactors’ fuels and materials. To facilitate cooperation, INEPC can use bilateral technical collaboration arrangements, memorandums of understanding, technical action plans, the International Nuclear Energy Research Initiative (I-NERI), and the International Nuclear Cooperation framework. Pertinent to this study, in 2001 DOE-NE established I-NERI to perform R&D with international partners at facilities developing advanced nuclear energy systems and to support R&D linked to DOE-NE’s principal research programs. I-NERI collaborators include Canada, the European Union, and the Republic of Korea (DOE-NE, n.d.-a).

In 2015, the Nuclear Energy Agency (NEA) of the Organisation for Economic Co-operation and Development (OECD) began a wide-ranging initiative on nuclear energy innovation, called the Nuclear Innovation 2050 project, with emphasis on fostering international cooperation, especially among OECD nations. The scope of this project’s technology areas includes reactor systems design and operation, fuels and fuel cycles, waste management and decommissioning, and nonelectrical applications of nuclear power. Cooperative activities have involved U.S. national laboratories, such as Argonne National Laboratory and Idaho National Laboratory partnering with Belgium’s SCKCEN, France’s CEA, and Japan’s Atomic Energy Agency

___________________

⁵ This topic is covered in the parallel National Academies study on advanced nuclear reactors.

on a variety of topics, including the following of relevance to this study: improving the fuels qualification⁶ process to accelerate industrial deployment, advanced structural materials for Gen IV systems, advanced fuel cycle chemistry, and advanced components for Gen IV to foster safety and economics.⁷ The cooperative activities described above could help optimize financial resources for each country involved (Todd, 2021).

3.2.3 Advanced Reactor Types

The committee received presentations from representatives of all the advanced reactor developers listed in Table 3.1; in total, the committee heard from 15 advanced reactor developers on 17 different reactor designs (see listings of presentations in Appendix B). Detailed technical design considerations, safety aspects, electricity generation, and nonelectricity applications of these advanced reactors are addressed in the parallel National Academies study Laying the Foundation for New and Advanced Nuclear Reactors.

As an indicator of the number of advanced reactor designs that have made steps toward potential deployment in the United States by 2050, 10 developers have submitted for 11 designs license applications or preapplications to the U.S. NRC. These are

• “NuScale—light-water SMR [small modular reactor]
• Oklo Aurora—fast micro-reactor⁸
• GEH BWRX-300—light-water SMR
• General Atomics EM2—gas cooled fast reactor
• Holtec SMR-160—light-water SMR
• Kairos Power—molten salt-cooled reactor with TRISO fuel
• Terrestrial Energy IMSR—thermal molten salt reactor
• TerraPower Natrium—sodium-cooled fast reactor
• TerraPower MCFR—fast molten salt reactor
• Westinghouse eVinci—micro-reactor
• X-energy Xe-100—high-temperature gas reactor” (Nichol, 2021).

3.2.3.1 High-Temperature Gas-Cooled Reactor

High-temperature gas-cooled reactors (HTGRs) are graphite-moderated, helium-cooled thermal reactors that use TRistructural ISOtropic (TRISO) particle fuel and have high outlet temperatures of 700–1,000°C, which leads to higher thermal efficiencies (Arostegui and Holt, 2019). TRISO fuel consists of uranium oxycarbide or uranium dioxide (up to 20 percent uranium-235 enrichment) encased in carbon and silicon carbide; these fuel particles (1 mm in diameter) are stable up to 1,600°C (WNA, 2020b). Two primary HTGR designs have been proposed: pebble-bed and prismatic block, named for the different shape of the fuel elements used (MIT, 2018). Pebble-bed designs contain hundreds of thousands of spherical fuel elements called “pebbles” (up to 6 cm in diameter), which are cycled through the reactor core; each pebble contains thousands of TRISO fuel particles. Prismatic block designs use hexagonal graphite blocks in the reactor core, with holes in graphite blocks for insertion of cylindrical TRISO fuel compacts and holes for circulation of helium coolant.

The primary fuel cycle research needs for HTGR reactors relate to the fabrication, qualification, and disposal of TRISO fuel, as discussed in Chapters 4 and 5; the management of radioactive graphite dust, as discussed in

___________________

⁶ While there is no explicit definition or use of fuel qualification in U.S. NRC rules or regulations, the report Fuel Qualification for Advanced Reactors (NUREG-2246 [U.S. NRC, 2022a]) defines the term as “the overall process (planning, testing, analysis, etc.) used to obtain qualified fuel,” where qualified fuel is defined as “fuel for which reasonable assurance exists that the fuel, fabricated in accordance with its specification, will perform as described in the safety analysis.”

⁷ To access papers on these topics, see https://www.oecd-nea.org/jcms/pl_21829/nuclear-innovation-2050-ni2050.

⁸ In January 2022, the U.S. NRC denied the Oklo Aurora combined license application because of the company’s failure to supply sufficient information, despite repeated requests by agency staff (U.S. NRC, 2022d). Soon after the denial, Oklo announced that it will respond and resubmit its application.

TABLE 3.1 Classification of Advanced Reactor Systems by Coolant, Neutron Spectrum, Fuel Type, Uranium Enrichment, and Fuel Cycle Options

^a Primary fuel cycle as intended currently by the advanced reactor developers.

^b Optional fuel cycle that could be developed depending on policy, regulatory, and economic drivers.

NOTE: HALEU = high-assay low-enriched uranium; HTGR = high-temperature gas-cooled reactor; LEU = low-enriched uranium; LFR = lead-cooled fast reactor; LWR = light water reactor; MOX = mixed oxide; TRISO = TRistructural ISOtropic; UC = uranium carbonide; UCO = uranium oxycarbide; U_dep = depleted uranium.

SOURCE: Adapted from MIT (2018).

Chapter 5; and the development of safeguards for material accountancy and control of TRISO pebbles, as discussed in Chapter 6. Examples of HTGRs under development are X-energy’s Xe-100 (TRISO pebble-bed), Framatome’s SC-HTGR (TRISO prismatic block), and BWXT’s Advanced Nuclear Reactor (BANR) microreactor (TRISO prismatic block).

As to technology development, important lessons can be learned from experience in other countries in developing and deploying pebble-bed HTGRs. The German AVR, an experimental pebble-bed HTGR that operated from 1967 to 1988, experienced challenges with inadmissibly high core temperatures, contamination with metallic fission products and radioactive dust, and water ingress (Moorman, 2008). The operational challenges encountered with AVR prompted additional research into improving metallic fission product retention in fuel elements, modeling core temperature behavior, and improving dismantling practices to avoid radioactive dust contamination (Moorman, 2008). In 2000, China entered an agreement with South Africa for cooperation in developing this technology. Starting in 1994, South Africa had invested significant resources in pebble-bed technology and had made progress, such as building a prototype fuel fabrication facility, but the planned test reactor was not constructed. In 2010, South African government funding was cut because of lack of investor interest in commercializing the technology (WNN, 2010). China has taken more than 20 years to develop and deploy a demonstration HTGR, and its experience could provide a potential pathway for U.S. developers of this technology. Briefly, Chinese researchers first built the prototype reactor HTR-10, which has a 10-MWe power rating, to thoroughly test materials, components, and the various systems, as well as to train a cadre of engineers, technicians, and operators for all aspects of the reactor’s systems. This extensive prototyping experience led to scaling up to the HTR-PM (high-temperature [gas-cooled] reactor—pebble-bed modular), a 200-MWe commercial demonstration reactor with a 750℃ operating temperature, resulting in about a 40 percent thermal efficiency. Construction of HTR-PM began in 2012, and the reactor went critical on September 12, 2021; the relatively long lead time can be attributed to development of first-of-a-kind technologies (Adams, 2021). In the United States, as discussed in detail in Section 3.3.2, X-energy’s Xe-100 has recently received more than $1 billion of support via DOE’s Advanced Reactor Demonstration Program (ARDP) and has a timeline to deploy a demonstration unit by 2028. X-energy has invested in developing TRISO fuel fabrication, as discussed in Chapter 4.

3.2.3.2 Gas-Cooled Fast Reactor

Gas-cooled fast reactors (GFRs) are fast neutron spectrum reactors that use ceramic-clad uranium carbide or nitride fuel and helium gas as the coolant (GIF, 2021b; WNA, 2020b). These reactors are expected to operate with high outlet temperatures of around 850°C and are typically envisioned for use in a closed fuel cycle with full actinide recycling. In the GFR design, heat is transferred from the primary helium circuit into a secondary helium/nitrogen circuit, which then powers a gas turbine.

The GFR is the only GIF reactor design that does not have an operating predecessor from which an experience base can be developed, so its introduction will almost certainly be as an experimental technology demonstration reactor. Fuel cycle R&D needs include fabrication and qualification of the fuel and cladding materials and development of a reprocessing system capable of actinide recycling from uranium carbide or nitride fuels.⁹ The AIROX recycle process, a cyclic dry pyrochemical oxidation/reduction process, is being pursued to eliminate the need for enrichment after first core and leave fission products as the only waste (Gougar, 2018). According to Majumdar et al. (1992), the AIROX process applied to spent LWR fuel would generate 0.29 MT (metric tons) of cladding and hardware and 0.1 MT of semivolatiles as GTCC waste per MT of initial heavy metal of spent fuel.¹⁰ The heat load from AIROX fuel would be similar to high-burnup fuel; consequently, AIROX-reprocessed fuel would be hotter than normal once-through spent fuel (Majumdar et al., 1992). The AIROX process produces

___________________

⁹ For uranium nitride fuel, the (n,p) reaction on nitrogen-14 produces copious amounts of carbon-14, which becomes a waste management issue unless the nitrogen is heavily enriched in nitrogen-15. Because of the enrichment costs, any recycle process must recover nitrogen-15 in high yield.

¹⁰ The text was corrected after release of the prepublication version of the report to clarify that it referred to the AIROX process as applied to spent LWR fuel.

fewer, but hotter, spent fuel assemblies. This must be factored into the repository design. General Atomics’ EM² and Euratom’s Allegro are other examples of proposed GFR designs.

The path to commercial deployment of the GFR will need to traverse uncharted territory given its lack of an operational predecessor. General Atomics has performed extensive research, especially on carbide fuel development,¹¹ and has outlined to the International Atomic Energy Agency (IAEA) the following timeline toward potential deployment: in 2010, conceptual design and development started; in 2023, high-risk development completed; in 2024, prelicensing vendor design review started in the United States; in 2029, engineering design complete; in 2030, construction of a prototype started in the United States; in 2032, potential commercial operation (GA, 2019). In comparison, the European consortium developing the Allegro reactor is seeking to deploy the first prototype GFR but recognizes that many years of development are required. This consortium was formed in 2010 and needed 5 years of preliminary conceptual work to reach the beginning of the design conceptual phase in 2015. The design conceptual phase is expected to take up to 11 years, until 2026. At that point, a decision will have to be made about whether to continue the Allegro project (Bělovský, 2019). As mentioned above, a major technical challenge is qualification of the carbide and nitride fuels. Another significant challenge is that the plant’s materials would have to withstand high levels of radiation damage (such as neutron embrittlement) coming from the long-lived cores that are being proposed (MIT, 2018).

3.2.3.3 Sodium-Cooled Fast Reactor

Sodium-cooled fast reactors (SFRs) are fast neutron spectrum, high-power density reactors that are cooled with liquid sodium and use either metallic uranium/plutonium or mixed oxide fuel (GIF, 2021a; WNA, 2020b). The use of liquid sodium as a coolant allows low-pressure operation and provides more effective heat transfer than water, but an SFR requires an air- and moisture-free environment because sodium metal reacts violently with air and water, and the opaqueness of the sodium coolant presents monitoring and inspection challenges (Flanagan et al., 2015; GIF, 2021a). Incidents related to the sodium coolant—leaks, corrosion, fires, reactivity with water—have been key challenges for previous and currently operating SFRs (IAEA, 2013c). There are two primary reactor designs: pool-type, in which the primary heat exchanger remains contained in the reactor vessel, and loop-type, in which the primary heat exchanger is outside the reactor vessel (Flanagan et al., 2015). Both designs incorporate an intermediate heat exchanger to separate the primary heat exchanger from the power generation system. A range of reactor sizes is being considered, from small modular-type reactors that generate 50–150 MWe to large loop-type reactors that generate 600–1,500 MWe (GIF, 2021a).

Several fuel cycle variations are being proposed to support SFRs in the long term that will require additional development in the areas of fuel fabrication and reprocessing. These include multirecycling of plutonium and possibly also minor actinides with either metallic or mixed oxide fuel, as discussed in Chapter 4. Depending on the choice of fuel, issues requiring attention could include fuel swelling, fuel/cladding chemical and mechanical interactions, and fuel/coolant compatibility (Hill, 2021). Electrometallurgical pyroprocessing for use with uranium/plutonium metallic fuel is envisioned, while mixed oxide reprocessing would likely involve advanced aqueous processes. If uranium nitride fuel is used in an SFR, the same carbon-14 generation and waste management problem as seen with GFR fuel occurs if the uranium nitride fuel is not first highly enriched in nitrogen-15 (see footnote in Section 3.2.3.2). The pyrophoric sodium coolant (and, for some reactor designs, metallic fuel) that uses sodium to bond the fuel thermally to the cladding presents unique waste management challenges, as detailed in Chapter 5. Example SFR designs under development include TerraPower/GE-Hitachi’s Natrium and Advanced Reactor Concepts’ ARC-100.

Experimental and commercial SFRs have been operated worldwide dating back to the 1950s. At the present time, Russia has two operating SFRs, BN-600 and BN-800, and China is building two demonstration reactors (CFR-600). According to the IAEA ARIS (Advanced Reactors Information System) database,¹² other SFR designs (experimental, demonstration, and commercial) are being pursued by Japan, Russia, India, the Republic of Korea, and the United States.

___________________

¹¹ For a list of scientific papers related to General Atomics’ EM² research, see https://www.ga.com/nuclear-fission/scientific-papers.

¹² See https://aris.iaea.org.

As to future technology development, Natrium has received significant financial support—more than $1 billion—from the ARDP, as discussed in Section 3.3.2, and equal levels of funding from its private investors. In November 2021, DOE announced its intention to invest nearly $2 billion in the Natrium demonstration reactor, which is planned to be built near a to-be retired coal plant near Kemmerer, Wyoming. As of late 2021, TerraPower is working on an application to the U.S. NRC for a permit to start construction by 2024. Under the ARDP, TerraPower has committed to achieving licensing, construction, and operation by 2028. Although there have been several experimental and commercially operational SFRs since the 1950s, DOE has described Natrium as a first-of-a-kind project (DOE-NE, 2021d). TerraPower has claimed that Natrium’s “new plant architecture minimizes cost and construction time” but told the committee that a challenge is that non-LWR licensing requirements have not yet been fully defined (Neider, 2021).

3.2.3.4 Lead-Cooled Fast Reactor

Lead-cooled fast reactors (LFRs) are fast spectrum reactors that typically use metallic uranium or uranium nitride fuel and liquid lead or lead-bismuth eutectic coolant (GIF, 2021c; WNA, 2020b). A variety of LFR designs have been proposed, with outlet temperatures ranging from 300 to 750°C, though the only operational experience has been in Russia’s Alfa-class submarines from 1968 to 1995 (WNA, 2020b, 2021d). The liquid lead or lead-bismuth eutectic employed as coolants are chemically inert to air and water and have high boiling temperatures, which provide potential safety benefits relative to sodium metal coolant (GIF, 2021c; Wallenius, 2021). While the use of lead-based coolants allows for low-pressure operation, the reactor must be maintained at high temperatures to prevent the lead from solidifying (GIF, 2021c). At such high temperatures, lead becomes corrosive to steel, and further research is required to develop alternative materials with higher durability under the required operating conditions. In 2021, LeadCold announced development of an aluminum alloyed steel, Fe-10Cr-4Al-RE, which is claimed to be resistant to lead corrosion (LeadCold, 2021). Lead and lead-bismuth coolants also pose challenges related to chemical toxicity, and long-term leaching into the environment must be prevented (IAEA, 2019e). Because of their high toxicity, lead and other heavy metals are regulated in the United States by the Environmental Protection Agency under the Resource Conservation and Recovery Act, which could potentially present a hurdle for reactor designs proposing to use such lead-based coolants. Furthermore, for both LFRs and SFRs, the opacity of the coolant and high-temperature operations complicate monitoring and maintenance procedures (GIF, 2021c; Wallenius, 2021). Likewise, LFRs and SFRs can use depleted uranium or thorium fuel and burn actinides from LWR fuel. Full actinide recycle is expected at maximum reactor technology implementation, requiring the development of advanced aqueous or nonaqueous pyroprocessing of spent fuel. As mentioned above, the use of uranium nitride fuel that is not highly enriched in nitrogen-15 leads to major carbon-14 waste management issues (see footnote in Section 3.2.3.2). Examples of LFRs under development include LeadCold’s SEALER-55, Westinghouse’s Lead Fast Reactor, and Russia’s BREST-OD-300 and SVBR-100.

As to technology development, LeadCold told the committee in February 2021 that KTH (the Swedish Royal Institute of Technology and a research partner) “has developed methods for manufacture of uranium nitride permitting tailor made manufacture of this fuel at industrial scale” with use of 99.5 percent enriched nitrogen-15, although it acknowledges that this fuel type is “difficult to manufacture using conventional methods.” Moreover, an “industrial scale supplier of ¹⁵N has been identified,” and Studsvik in Sweden could be a potential fuel fabricator (Wallenius, 2021). LeadCold aims to have a demonstration reactor of its SEALER-55 design built by 2030 at Oskarshamn, Sweden (WNN, 2021b).

Westinghouse’s LFR presenters told the committee in October 2021 that their conceptual design is nearing completion, and commercialization might occur in the 2030s, according to Westinghouse’s “staged approach to development.” This approach leverages a network of test rigs, located primarily in the United Kingdom, the United States, and Italy, which are used to also support demonstration of key systems, components, and phenomena of the Westinghouse LFR. The testing involves corrosion/erosion, fuel manufacture, plant safety and reliability, lead’s effects on materials’ mechanical properties, component testing (e.g., on the fuel bundle and primary heat exchanger), and instrumentation (e.g., under-lead viewing). The start-up core is planned to use HALEU uranium

dioxide with D9-type austenitic steel cladding; future performance-enhanced cores could use uranium nitride with advanced steel or silicon carbide cladding (Ferroni et al., 2021).

3.2.3.5 Molten Salt Reactors

Molten salt reactors (MSRs) have a wide variety of design options but limited operational experience. Two experimental MSRs have operated: the 1950s-era Aircraft Reactor Experiment and the 1960s-era Molten Salt Reactor Experiment, both in the United States, and both thermal MSRs. A fast MSR has never been operated. A primary classification consideration is whether an MSR uses molten salt as coolant only or as fuel and coolant. GIF defines molten salt reactor as the latter, in which the fuel is dissolved in the salt coolant (GIF, 2021d). However, DOE, among others, has sometimes referred to the design being developed by Kairos Power as an MSR, although that concept is a fluoride salt–cooled reactor with solid TRISO fuel and is properly referred to as a fluoride salt–cooled, high-temperature reactor (DOE-NE, 2020).

MSRs offer the potential for high fuel-resource conservation and high actinide consumption via burnup. Liquid fuel MSRs can operate with a thermal, epithermal, or fast neutron spectrum and can use a variety of uranium- and thorium-based fuels. The fuel form can be as solid particle fuel in graphite similar to the core of a high-temperature gas-cooled reactor (HTGR) or dissolved in the molten salt to form a liquid “fuel salt” such as in Flibe Energy’s design. The latter has the advantage that no solid fuel fabrication is required, but because gaseous fission products separate from the liquid fuel during operations, a fully open fuel cycle is not technically possible.¹³ Molten salt coolants can be either a fluoride- (e.g., FLiBe [2⁷LiF-BeF₂]) or chloride-based salt, which can be circulating throughout the reactor during operation (GIF, 2021d; Pereira, 2020).

Thermal MSRs are typically designed with graphite as the moderator. Liquid fuels offer the possibility of refueling at power in steady-state operation and continuously removing fission products by salt processing in a chemical processing facility directly attached to the reactor. Liquid fuel thermal MSRs operating in full recycle mode would require either continuous or batch salt processing to remove parasitic fission products with large neutron capture probabilities to maintain desired reactivity. Salt processing to remove fission products could be carried out with halogenation (fluorination or chlorination) playing a key role in many process flow sheets. For example, uranium in a fluoride salt is UF₄. Upon fluorination, UF₄ reacts with fluorine gas to produce the volatile UF₆ species that enters the gas phase, effectively separating uranium from the salt. The resulting UF₆ stream undergoes purification by selective sorption/desorption reactions to remove fission products that may also form volatile fluorides (e.g., Br, I, Cs, Se, Mo, Tc, Ru, Sb, Zr, and Te). The purified UF₆ is reduced to UF₄ prior to being reintroduced as fuel into the reactor. Materials compatibility issues continue to be important challenges for both the halogenation and reduction steps in salt processing. Fuel salt cleanup is also an important process supporting MSR operations, as it addresses such impurities as oxygen to ensure the appropriate redox potential is maintained in the salt. The redox potential is an important factor for controlling corrosion and the salt chemistry.

One advantage of MSRs is that the fuel could be processed during reactor operations in an appropriately sized facility directly attached to the reactor, which would eliminate the need to transport spent fuel to an off-site processing facility. Such a facility would also be substantially smaller with flow rates orders of magnitude lower than that of a centralized processing facility like those used to support solid fuel technologies. Additional advantages to MSRs include that there are no neutron losses in structural materials; there is online refueling; there is no need for solid fuel fabrication; the reactor can operate under low pressures and could go to high temperatures; the reactor has a very low excess reactivity; and the molten salts will not react dangerously with air or water—although, as noted below, there are concerns about corrosion. Disadvantages include material degradation; need for safe management of fission product off-gases, as well as tritium, which would be produced if lithium is used in the salt; need for remote maintenance; difficulty with material inventory tracking; need for electric heaters so

___________________

¹³ Liquid-fueled MSRs will operate on a modified open fuel cycle that requires an off-gas treatment system to capture and contain volatile fission products. An open fuel cycle in the sense of once-through use of fuel (i.e., no recycle) is possible if the spent fuel salt removed from the reactor is discarded and not treated to recover actinides for reuse as fuel. Volatile fission products may be evolved and captured.

that the salt remains a liquid; and need for both the nuclear plant and online fuel processing plant to maintain high availability simultaneously.

In fast MSRs, losses due to parasitic fission products would be considerably fewer, and salt processing could be carried out less frequently. In a recent report on molten chloride fast reactors, the Electric Power Research Institute found that the fuel salt sustainability over time may be limited by the buildup of fission products, which could “potentially impact reactor performance, waste management, and downstream reuse of fuel salt to start up subsequent units” and cautioned that “an integrated understanding of fuel salt sustainability early in the liquid-fuel MSR design process may help developers and owner-operators avoid downstream design and operational challenges” (EPRI, 2021a). Thus, the online fuel processing needs for fast MSRs would primarily be fission off-gas treatment and the removal of particulates of noble metal fission products that build up in the salt. Fuel cycles that support fast MSRs depend on whether, for example, the reactors are operated as transuranic burners using spent LWR fuel as feed, uranium/plutonium breeders, or converters using natural uranium fuel feed with removal of noble metal fission products by filtration (minimal separations).

Fuel salt chemistry in MSRs is complex, and the salt composition (e.g., quantities of fissile and fertile isotopes, as well as fission, transmutation, and radiolysis products) is continually changing with time. Salt cleanup, salt processing (using, e.g., volatility, reductive extraction, or electrochemical methods), and off-gas treatments are important areas of R&D for MSRs going forward if they are to realize their potential of fully closing the fuel cycle (Fredrickson et al., 2018; Pereira, 2020). Additional R&D challenges relate to the chemical reactivity of molten salts—specifically, chemical compatibility between the molten salt and reactor structural materials, including alloying elements; the potential for enhanced corrosion in the presence of air, moisture, or other nonmetal impurities; and the effects of irradiation (McFarlane, 2021). Further R&D is also needed for waste management—in particular the processing of chloride and fluoride salts required to generate suitable waste forms (McFarlane, 2021). Other key research areas include industrial-scale, economical, and environmentally friendly methods for the isotopic separation and recovery of chlorine-37 and lithium-7 for molten chloride and fluoride FLiBe salt systems, respectively (Arm et al., 2020). Molten salts must be highly enriched in chlorine-37 and lithium-7 to maintain the neutron economy within the reactor and avoid neutron capture reactions that produce large quantities of chlorine-36 (and sulfur-35) and tritium as waste products.¹⁴ Examples of liquid fuel molten salt reactor designs under development include Terrestrial Energy’s IMSR-400, ThorCon’s ThorCon Reactor, Flibe Energy’s Liquid Fluoride Thorium Reactor, Moltex’s Stable Salt Reactor-Wasteburner, and TerraPower’s Molten Chloride Fast Reactor.

Fluoride-cooled high-temperature reactors (FHRs) are thermal spectrum reactors that are sometimes referred to as MSRs because they use a molten fluoride salt as the coolant; however, FHRs are more properly classified as high-temperature gas-cooled reactors because they use solid fuel, typically TRISO, are graphite moderated, and are designed to operate on an open fuel cycle with no reprocessing (MIT, 2018). Because of the much higher heat capacity of a molten salt compared with a gas such as helium, the use of a molten salt as the coolant provides a significant advantage for removing heat from the reactor core. Research needs for FHR fuel cycles are analogous to those described above for both TRISO fuel and molten salts, and are detailed in Chapters 4 and 5.

Kairos Power is developing an example of this reactor type, the KP-FHR. The company initiated preapplication review activities with the U.S. NRC in September 2018 and subsequently submitted a total of 11 topical reports, 8 of which had received draft or final safety evaluation reports as of May 2022. In November 2021, the U.S. NRC formally accepted its construction permit application for the Hermes low-power test reactor, which will be built at the East Tennessee Technology Park Heritage Center site in Oak Ridge (Kairos, 2021). Kairos aims to have Hermes operational by 2026. Kairos is implementing an iterative development process from engineering test unit to demonstration reactor to commercial scale. In January 2021, Kairos representatives told the committee that they are focused on iterative testing of hardware because it can demonstrate real-world performance unlike relying on computer calculations alone (Blandford and Peterson, 2021). The electrically heated Engineering Test Unit is the first major integrated hardware iteration that has allowed Kairos to demonstrate systems in a nonnuclear environment. Kairos’ “iterative strategy is supported by capabilities in material testing, tritium testing, chemistry control, mod/sim [modeling/simulation], core design and neutronics, and instrumentation and controls” (Blandford and

___________________

¹⁴ Chlorine-36 is a long-lived (t_1/2 = 301,000 years) energetic (709 keV) beta emitter that is highly soluble (mobile) in water.

Peterson, 2021). Experience with Hermes would lead to a larger-scale nonnuclear electrically heated integrated demonstration facility, whose testing results would feed into the deployment of the first commercial-scale reactor unit. This KP-X reactor is planned to be operational before 2030, enabling subsequent commercial deployment of additional KP-FHRs in the 2030s (Blandford and Peterson, 2021).

Terrestrial Energy is pursuing licensing applications in the United States for its MSR design, having been in prelicensing application discussions since late 2019 with the U.S. NRC (2022b). Additionally, Terrestrial Energy, ThorCon, and Moltex are pursuing licensing applications outside of the United States. In 2017, Terrestrial Energy completed the first phase of prelicensing review of its IMSR by the Canadian Nuclear Safety Commission (CNSC), which determined that the safety features met the threshold to move toward formal licensing review, and they expect to complete the second-phase licensing review in 2022 (NEI, 2021a). In May 2021, Moltex completed Phase 1 of CNSC prelicensing vendor design review (Moltex Energy, 2021). ThorCon is planning to first build a test reactor outside the United States to gather test data and has signed a memorandum of understanding with Indonesia (Jorgensen, 2021). A ThorCon representative told the committee in January 2021 that the licensing process in the United States is too lengthy and prohibitive for private investors. The same representative noted that, in particular, in the U.S. regulatory process, validated software is needed to license a design before one can build a test reactor, but to validate the software, one needs to use test data.

3.2.3.6 Small Modular Light Water Reactor: NuScale Power Module (iPWR)

Small modular LWRs, such as the NuScale Power Module, are derived from the current commercial large (>300 to 1,000+ MWe) pressurized LWRs. As such, they operate with a thermal spectrum and uranium dioxide fuel enriched to less than 5 percent uranium-235, using light water as both the coolant and moderator. However, in addition to the difference in power rating, these small modular LWRs have other key differences from large LWRs. A small modular reactor “can be constructed and operated in combination with similar reactors at a single site” (Infrastructure Investment and Jobs Act of 2021, Public Law 117-58). Integral designs, called “iPWRs” (integral pressurized water reactors), enclose the whole reactor primary circuit—including the pressurizer, coolant pumps, and steam generators—in a single containment vessel, which allows for the reactor to be compact.

The NuScale Power Module (NPM), an example of an iPWR, is designed to provide 77 MWe per module. The NPM’s simple design eliminates the coolant pumps, large-bore piping, and other components typically found in large, conventional reactors. As such, the NPM is small enough to be factory built, enabling easy transport and installation, which could reduce the time and cost of construction (Arostegui and Holt, 2019).¹⁵ The NPM is also scalable for flexible integration to match load requirements (4, 6, and 12 modules per plant). Given the similarity of its base technology to commercial LWRs, the fuel cycle facilities required to fuel and manage the wastes for these reactors—with the exception of a deep geologic repository—already exist. In addition, the NPM is the first small modular reactor to undergo the U.S. NRC licensing process. In September 2020, the U.S. NRC issued a standard design approval to NuScale approving the safety aspects of a 50-MWe version of the NPM design. NuScale will also have to obtain a separate standard design approval for its 77-MWe version, which it plans to deploy in a six-module reference plant to be operational by 2029 at Idaho National Laboratory (Patel, 2022). Of the reactors evaluated by this committee, NuScale’s NPM is the only technology that could likely be commercialized in this decade. Examples of other small modular LWRs under development include GE-Hitachi’s BWRX-300 (a boiling water reactor) and Holtec International’s SMR-160 (a pressurized water reactor). The ground laid by NuScale would likely aid the development and potential deployment of these reactors.

As to technology development, the NPM is nearing commercial deployment by the end of the decade because of several factors: (1) technology derived from existing LWRs; (2) use of uranium dioxide fuel with enrichment less than 5 percent uranium-235, which can be produced with the existing LWR fuel cycle infrastructure; (3) substantial financial support from government and investors; and (4) existing U.S. nuclear regulatory processes for LWR technology. NuScale has received funding for over a decade from several federal awards and especially

___________________

¹⁵ These aspects will be covered in more depth in the parallel National Academies study Laying the Foundation for New and Advanced Nuclear Reactors.

from DOE through the SMR Licensing Technical Support Program (which ran from fiscal year [FY] 2012 to FY2017), from DOE-NE’s U.S. Industry Opportunities for Advanced Nuclear Technology Development in 2018, and through the Advanced SMR R&D Program in FY2021. NuScale also benefitted from DOE’s 2020 award to Utah Associated Municipal Power Systems for the Carbon Free Power Project, which intends to use NuScale’s NPM for generation of electrical power. As of end of 2021, NuScale had received more than $1 billion in federal funding through 30 awards (Subsidy Tracker, 2021).

3.2.4 Fuel Cycle Options for Advanced Reactors

During the information-gathering meetings, almost all developers told the committee that they are planning on an open, once-through fuel cycle at least for the near to intermediate terms, and likely for the next 30 years (see Table 3.1). Yet, several developers noted that there is potential in the longer term for their technologies to recycle spent nuclear fuel.

Some developers mentioned potential reprocessing and recycling in terms of proliferation resistance. At the January 2021 meeting, the ThorCon representative said that the company would prefer to recycle the enriched uranium because HALEU is a valuable material. However, the NNSA has advised them “not to push” for recycling of uranium because it is a “touchy subject.” Thus, the representative said that ThorCon plans to store the material until the company can work through the bureaucratic issues (Jorgensen, 2021). TerraPower’s website states that “Natrium plants will not require reprocessing and will run on a once-through fuel cycle that limits the risk of weapons proliferation” (TerraPower, 2021a). In contrast, however, the General Atomics representative discussed how the company’s EM² technology can recycle in a proliferation-resistant manner, so as not to require enrichment after the first core loading, saying that “subsequent dry recycle processing eliminates that need for further enrichment” (Back and Schleicher, 2021). See Section 3.2.3.2 for more information about the EM² technology’s development.

The reactor developers considering recycling also described the purported benefits of reduced costs, reduced waste volumes, decreased radiotoxicity,¹⁶ and improved resource sustainability. TerraPower plans to use a once-through cycle and estimates that up to 30 times better uranium utilization is possible than for LWRs, allowing long-term stability without the need for reprocessing (Hejzlar, 2021). However, the TerraPower representative told the committee at the February 2021 meeting, “If recycling is desired, Natrium can support it at reduced cost due to less frequent reprocessing” (Hejzlar, 2021). While ARC plans to use the once-through cycle, the company’s representative noted at the February 2021 meeting that their reactor’s “system can be sustained indefinitely if recycling is used” (Sackett and Arthur, 2021). Reprocessing is an option with ThorCon’s technology concept to improve resource utilization, but the company does not plan to pursue it in the near to intermediate terms. Oklo is the one developer that has moved forward with plans for recycling and stated at the committee’s February 2021 meeting that fast reactors using metallic fuel clearly have potential for recycling (DeWitte, 2021). However, given Oklo’s plans for 20-year core development, the reprocessing, if it occurs, would not be in the near term. Additionally, in January 2022, the U.S. NRC denied Oklo’s license application, although the company stated that it plans to respond to U.S. NRC’s concerns and resubmit its application.

Given the stated positions of almost all developers about pursuit of the once-through cycle for the foreseeable future, the committee chose to emphasize this option for the classes of advanced reactors considered in this study. However, because, as noted above, a number of developers discussed the potential of recycling and ultimately closing the fuel cycle through multirecycling, the committee also considered the possibility of reprocessing and multirecycling. Building on the three main fuel cycle options outlined in NEA-OECD (2021), the committee analyzed for uranium-based fuels (1) the once-through cycle for LWRs; (2) monorecycling of uranium and plutonium as mixed oxide fuel in LWRs; and (3) multirecycling of uranium, plutonium, and minor actinides (americium, curium, and neptunium) in fast reactors, in order to evaluate the merits and viability of different fuel cycle options (see Chapters 2 and 4). Chapter 5 discusses waste management and disposal issues, including those

___________________

¹⁶ As measured by the decreased inventory of long-lived transuranic element radioisotopes remaining in the high-level waste.

related to reprocessing, for four representative advanced reactor designs: integral pressurized water reactors, high-temperature gas-cooled reactors, sodium-cooled fast reactors, and molten salt reactors.

The committee also heard from two reactor developers, Flibe and ThorCon, who intend to use a fuel cycle based on thorium rather than uranium. Terrestrial Energy also noted the possibility of operating their reactor as a thorium breeder in the future. In Section 3.2.5, a brief comparison of thorium-based fuel cycles (thorium/uranium) with uranium-based fuel cycles (uranium/plutonium) is provided, noting the potential advantages and challenges of switching to thorium-based fuels. More detailed information regarding the nonproliferation implications of thorium fuel cycles is discussed in Chapter 6.

In summary, whatever fuel cycle option is potentially pursued will require relatively long time periods to implement. According to Dr. Terry Todd, who has almost 40 years of experience in fuel cycle R&D and was the national technical director of DOE-NE’s Material Recovery and Waste Form Development program from 2008 to 2020, it will require “15-20 years to design, build, permit, and start up a new fuel cycle facility” (Todd, 2021). Demonstrating the start of a new fuel cycle could just require deploying the first commercial reactor using that option, which could be done in several years potentially. Fully implementing a fuel cycle, however, could require a couple of decades to a century in order to transition from a fleet using one fuel cycle, such as the once-through LWR cycle, to a fleet of fast reactors using multirecycling (Williamson and Taiwo, 2021). To keep options available for potential future fuel cycles will require sustained investments in R&D and workforce development (Zhan et al., 2021). As mentioned in Section 3.2.2, collaborations with international partners and these partnerships could help with cost-sharing in activities such as supporting joint R&D projects and building test reactors for qualifying fuels and testing materials (NEA-OECD, 2018a, n.d.). However, funding at the U.S. national level is still limited, and not all advanced reactor programs can have full government support; thus, decisions will have to be made as to what R&D support to prioritize (Todd, 2021).

3.2.5 Comparison of Thorium-Based Fuel Cycles (Th/U) with Uranium-Based Fuel Cycles (U/Pu)

In the periodic table, only two elements above bismuth (element 83) exist in nature to a significant extent: thorium (element 90) and uranium (element 92). As shown in Table 3.2, thorium has only one long-lived isotope,

while uranium has three. Of these four nuclides, only ²³⁵U is fissile; that is, ²³⁵U can readily undergo neutron-induced fission and thus be used to fuel a fission reactor. The other three isotopes are termed fertile nuclides and can be converted into fissile nuclides upon neutron capture. Figure 3.2 shows the pathways for producing ²³⁹Pu from ²³⁸U (top) and ²³³U from ²³²Th (bottom) by the capture of neutrons.

When uranium-based fuel is irradiated in a reactor, the fission of ²³⁵U generates on average between two and three neutrons. One of these neutrons reacts with another ²³⁵U nucleus to sustain the critical fission reaction, while the excess neutrons are available to react with other nuclei in the fuel. If these excess neutrons react with ²³⁸U in the fuel, they breed ²³⁹Pu, which can go on to fission or capture a neutron to become ²⁴⁰Pu. Even-numbered atomic mass nuclides of Pu¹⁷ do not fission appreciably with low-energy neutrons, so ²⁴⁰Pu tends to capture a neutron and become ²⁴¹Pu, which can either fission or capture another neutron becoming ²⁴²Pu, and so on. Over time, neptunium, plutonium, and heavier minor actinide isotopes (e.g., Am, Cm) are generated via capture and decay reactions, as illustrated in Figure 3.3.

Pure ²³²Th-based fuel in a reactor requires the addition of fissile material (i.e., ²³⁵U, ²³⁹Pu/²⁴¹Pu, or ²³³U, called seed fuels¹⁸) to enable and sustain nuclear fission. Since low-enriched U would be inefficient as a seed fuel, the seed fuel needed is higher-enriched U or Pu. When Pu is used as the seed fuel, the fuel would be similar to conventional mixed oxide fuel, but with ²³⁸U replaced by ²³²Th. The ²³⁹Pu/²³²Th fuel is referred to as “Th-MOX.” Excess neutrons from the fission of the seed fuel can then go on to breed ²³³U from ²³²Th (SNETP, 2011).

Table 3.3 shows the values of η, the number of neutrons produced per neutron absorbed in the fuel, for thermal and fast neutrons. These values indicate that breeding using ²³⁵U or ²³⁹Pu in a thermal reactor would be difficult, since expected losses to nonbreeding processes, such as absorption, by structural materials would reduce the excess neutrons to one or less. However, with a η = 2.3 the prospects for breeding ²³³U from ²³²Th with thermal neutrons are more reasonable. Comparing all values for η suggests that the most promising breeding occurs with a fast reactor and ²³⁹Pu (i.e., a U-based fuel cycle). Once any breeder reactor goes critical, it can, in principle, breed as much or more fissile material than it consumes. Because of the more favorable breeding economy in a fast reactor, once criticality is achieved, a fast breeder can make greater quantities of fissile material than a thermal breeder reactor.

___________________

¹⁷ The sentence was modified following release of a prepublication version of the report to clarify that it is referring to Pu specifically, not all even-numbered atomic mass nuclides.

¹⁸ A closed Th fuel cycle in equilibrium would utilize only ²³³U. However, ²³³U must first be produced from ²³²Th, which requires another seed fuel.

TABLE 3.3 Number of Neutrons Produced per Neutron Absorbed in the Fuel, η, for Selected Fissile Isotopes as a Function of Neutron Energy, Thermal (<100 eV) and Fast (>0.1 MeV)

SOURCE: Adapted from Murray and Holbert (2014).

3.2.5.1 Advantages and Challenges of a Th/U Fuel Cycle

Th is more abundant in the earth’s crust than U by a factor of about 3.3; however, the known reserves of economically extractable Th and U are similar (Touran, 2020). Seawater contains significantly more dissolved U than Th, but predicted costs for U recovery from seawater are not competitive with extraction of natural U deposits by mining (Dungan et al., 2017). With the vast amount of ocean water (about 300 million cubic miles) there are about 4 billion tons of uranium in the ocean at any given time (NEA and IAEA, 2020). Although advances continue to be realized for the extraction of U from seawater, at this time the technology is not commercially viable on an industrial scale (Berger, 2018). In the future, this technology could become viable with advances in U absorbent and fiber technology (Szondy, 2018; Xu et al., 2020a).

Because η is greater than 2 over a wide range of neutron energies, the ²³²Th/²³³U fuel cycle can operate as a breeder reactor with fast, epithermal, and thermal neutrons, whereas breeding is possible only with fast neutrons for the ²³⁸U/²³⁹Pu fuel cycle. The thermal neutron absorption cross section is three times higher with ²³²Th than with ²³⁸U.

Potential benefits and challenges of implementing a Th fuel cycle are discussed in IAEA (2005a) and summarized here. If the fuel is reprocessed from a thermal reactor operating on a Th/U fuel cycle, mining of U for ²³⁵U can be eliminated, thus extending nuclear fuel resources by two orders of magnitude without the need to deploy fast reactors. Breeding ²³³U from ²³²Th over time also can eliminate the need to enrich uranium as part of the fuel cycle.

The Th/U fuel cycle does not produce appreciable transuranium (TRU) elements, since ²³⁸U is not a starting material. The Th/U fuel cycle produces some ²³⁷Np, but if the Np is removed via chemical separation, no Pu can be produced during irradiation. By avoiding Pu altogether, some consider the Th/U fuel cycle more proliferation resistant than U/Pu cycles. However, since ²³³U and ²³⁹Pu have similar nuclear properties, these fuel cycles have comparable proliferation risks. One option for reducing proliferation concerns is to add sufficient U to the initial Th load so that the ²³³U is isotopically diluted (denatured) with ²³⁸U. Isotopic separation would be required to obtain special nuclear material at the expense of the production of more TRU; such an option negates the original advantage of avoiding Pu using Th-based fuels.

There has been considerably less experience with Th than with U/Pu fuel cycles. In the mid-1950s, Oak Ridge National Laboratory developed the THORium-uranium EXtraction (THOREX) process for reprocessing of spent Th–based fuels (IAEA, 2005a). Similar to PUREX, THOREX is a solvent extraction–based separation of Th from fission products by means of tributyl phosphate (TBP). The chemistry of Th for nuclear power–generation applications (the THOREX process) is much less mature than the PUREX process for U and has not been performed commercially. Reprocessing using THOREX has been carried out in only a few countries and mostly in laboratory or pilot plant scale (Balakrishna, 2012; IAEA, 2005a). ThO₂ is chemically more inert (more stable) than UO₂ and as such, ThO₂-based fuel does not easily dissolve in concentrated nitric acid without the addition of HF, which leads to corrosion of process equipment.

Because the melting temperature for ThO₂ (3,390°C) is much higher than that of UO₂ (2,865°C), producing high-quality (high-density) solid Th fuel requires higher sintering temperatures compared with solid U fuel. However, ThO₂-based fuels have a higher radiation resistance, 10 times lower fission product release rate, higher thermal conductivity, and lower coefficient of expansion compared with UO₂-based fuels. For these reasons, ThO₂based fuels have better in-pile performance than UO₂ or UO₂–mixed oxide fuels (suggesting that higher burnups can be achieved with Th-based fuels).

During the THOREX process, protactinium is normally discarded as waste with the fission products. Proliferation concerns of the Th/U fuel cycle center around the separation of ²³³Pa as an intermediate isotope.¹⁹ Since ²³³Pa has a relatively long half-life (27 days) compared with ²³⁹Np (2.35 days) in the uranium fuel cycle, a cooling time of about a year (more than 10 half-lives of ²³³Pa to allow the decay of ²³³Pa to ²³³U) is necessary prior to reprocessing to avoid loss of ²³³U fissile material (IAEA, 2005a). The diversion of separated ²³³U during fuel reprocessing poses a proliferation risk that can be mitigated by adding ²³⁸U (denaturing) (IAEA, 2005a).

In Th-based fuels, ²³²U (half-life 73.6 years) is produced by (n,2n) reactions with ²³²Th, ²³³Pa, and ²³³U. The short-lived daughter products of ²³²U, such as ²¹²Bi and ²⁰⁸Tl, emit strong gamma radiation in Th/U fuel that requires more shielding, making handling and reprocessing more difficult but not impossible, and more expensive compared with U/Pu fuel.

²³²Th blankets can be used to breed ²³³U, in a manner analogous to breeding Pu using ²³⁸U in “conventional” fast breeder reactors.

3.2.5.2 Reactors Capable of Using Th Fuel

Six different reactor types can use Th as a fuel: pressurized heavy water reactors, high-temperature gas-cooled reactors (HTGRs), boiling water reactors, pressurized light water reactors, fast neutron reactors, and molten salt reactors (WNA, 2020c). Some demonstration reactors that have used Th-based fuels to generate electricity include the pebble-bed 300-MWe Thorium High-Temperature Reactor in Germany, and the 40-MWe Peach Bottom HTGR and 330-MWe Fort St. Vrain HTGR in the United States. For all three of these reactors, the reference fuel was Th-HEU (highly enriched U). Because of proliferation concerns, the use of HEU (U enrichment to greater than 20 percent) in civilian reactors over the past several decades has been significantly curtailed worldwide, and implementing a Th fuel cycle with HEU seed fuel would not be considered today. Furthermore, replacing the reference HEU fuel with <20 percent low-enriched U considerably reduces its overall performance, making this Th fuel cycle much less attractive (NEA-OECD, 2015a).

The first reactor to use ²³³U as fuel was the Molten Salt Reactor Experiment at Oak Ridge National Laboratory in 1968. In the late 1970s, breeding ²³³U was demonstrated in a thermal LWR at the Shippingport Atomic Power Station in Pennsylvania (Olson et al., 2002).²⁰ The Shippingport reactor core consisted of 12 seed modules, containing >98 percent ²³³U, surrounded by a stationary blanket module. To achieve breeding, the fuel assemblies were moved through the core to adjust reactivity rather than using control rods. Also, the fuel had to be removed and processed at low burnup, where the conversion ratio is optimal. Consequently, breeding ²³³U in a thermal spectrum would require specially designed reactor systems that are different from the current commercial LWRs. Other reactor types are better suited to the thermal breeding of ²³³U (Wigeland et al., 2009). In the mid-1990s, there was renewed interest in Th fuel cycles in the context of advanced reactor development, and in particular for molten salt reactors.²¹ Solid fuel is not an issue in molten salt reactors, making them best suited for thermal ²³³U breeding (GIF Experts’ Group, 2010). Continuous online processing of molten salt fuel optimizes the overall breeding ratio by removing fission product poisons and separating ²³³Pa from the core to minimize parasitic neutron capture, which produces ²³⁴U, a fertile isotope, at the expense of ²³³U production (U.S. NRC, 2014d). Molten salt reactors also offer the possibility of adjusting the fertile/fissile fuel composition without shutting down the reactor (NEA-OECD, 2015a). However, the development, licensing, and construction of advanced reactor systems that might realize the full benefit of a closed Th/U fuel cycle is a long-term undertaking, as the requisite dedicated Gen IV and beyond breeder reactors, including molten salt reactors, are currently in the conceptual design phase.

___________________

¹⁹ In the Th conversion chain, ²³³Pa has to be removed from the thermal flux of the reactor; otherwise ²³³Pa will capture thermal neutrons, becoming ²³⁴Pa, and decay within a few hours to ²³⁴U, which is fertile, rather than the intended fissile ²³³U.

²⁰ Shippingport Atomic Power Station in Pennsylvania (USA) operated from 1977 to 1982.

²¹ The use of the MOlten Salt Advanced Reactor Transmuter (MOSART) “concept as a transmuting system cycle initially fed with TRU loading from LWR SNF for producing the necessary fissile ²³³U quantities for a uranium-thorium fuel cycle” (Ignatiev et al., 2014).

3.2.5.3 Relative Radiological Hazards of Key Isotopes in Th/U and U/Pu Fuel Cycles

The actinides play a dominant role in terms of total radioactivity, ingestion radiotoxicity, and potential dose to the public. A much smaller quantity of Pu and long-lived minor actinides (Np, Am, and Cm) is created in the ²³²Th/²³³U fuel cycle as compared with the ²³⁸U/²³⁹Pu fuel cycle, thereby minimizing toxicity and decay heat problems (IAEA, 2005a). The commonly stated perception that the ²³²Th/²³³U fuel cycle will generate less Pu and minor actinides, thereby reducing the radioactive hazard of the spent fuel from the ²³²Th/²³³U fuel cycle compared with that of the ²³⁵U/Pu fuel cycle, may not be correct. According to Piet (2013), “Thorium/²³³U fuel cycles do have lower amounts of TRU isotopes, but that does not necessarily mean lower ‘long-term hazard’ since focusing on TRU without adequate attention to actinide decay products can lead to incorrect conclusions.” Furthermore, Piet found that:

Croff and Krahn (2016) reached the same conclusion, noting, “When analyzed on a consistent basis (e.g., same reactor design, fuel design, burnup, specific power, cladding type and composition, and fuel matrix trace element concentration), the radiotoxicity of thorium-based fuels and wastes being disposed of in a repository are broadly similar to those for uranium-based fuel and not ‘far lower.’” Furthermore, Croff and Krahn (2016) found that “at relatively short times (less than a few centuries) and very long times (greater than a few million years), the radiotoxicities of uranium- and thorium-based SNFs [spent nuclear fuels] are essentially equal.”²² “Overall, the ingestion radiotoxicity of thorium-based fuels containing ²³³U or plutonium fissile materials is similar to the radiotoxicity of uranium-based fuels containing ²³⁵U or plutonium fissile materials for decay times ranging from 1 year to 20 million years” (Croff and Krahn, 2016). Figure 3.4 shows the ingestion radiotoxicity of 1 kg each of ²³²Th, ²³³U, ²³⁵U, ²³⁸U, and ²³⁹Pu and their progeny as a function of time. A report by the Nuclear Energy Agency of the Organisation for Economic Co-operation and Development on Th fuel cycles highlights several long-lived radionuclides generated in Th fuel cycles—²³³U, ²³¹Pa, and ²³²U—that “have a more important radiotoxicity than their counterparts in the uranium cycle,” and also concludes that “radiotoxicity of thorium-based fuels is more accurately described as being comparable to that of uranium-based spent nuclear fuel” (NEA-OECD, 2015b).

3.2.5.4 The Bottom Line: Uranium-Based (U/Pu) or Thorium-Based (Th/U) Fuel Cycles?

DOE’s Advanced Fuel Cycle Initiative Options Study reviewed and evaluated alternative fuel cycles and technology options, including a comparison of Th/U and U/Pu fuel cycles (Wigeland et al., 2009). The study concluded that “the thorium option would have lower, but probably not significantly lower, TRU inventory and disposal requirements, both having essentially equivalent proliferation risks” (Wigeland et al., 2009). It further concluded that “the choice between uranium-based fuel and thorium-based fuels is seen basically as one of preference, with no fundamental difference in addressing the nuclear power issues,” noting, however, that the lack of infrastructure for Th-based fuels in the United States and the lower technical maturity of processing Th versus U fuels would mean higher costs and R&D and demonstration requirements for the Th option. Developing technologies and processes for Th fuels to supplement U fuel technology could enable future nuclear expansion in the United States.

___________________

²² Although minor actinides (Am, Np, Cm) are not produced to any appreciable extent in Th-based fuel cycles, Th-based fuel cycles are associated with relatively short half-life isotopes, such as ²³²U and ²²⁸Th, as well as other radionuclides such as ²³¹Pa, ²²⁹Th, and ²³⁰U, which have a longer-term radiological impact (IAEA, 2005a).

Although most countries have opted for continued development of the U/Pu fuel cycle, India is the exception. India has been actively pursuing a Th fuel cycle because its natural Th resources are larger than its natural U resources. As discussed in Appendix H, India formulated a three-stage program “in the middle of the last century to exploit the full energy potential of its resources” (IAEA, 2008).

3.3 U.S. GOVERNMENT SUPPORT FOR DEVELOPMENT OF ADVANCED REACTORS AND ASSOCIATED FUEL CYCLES

Chapter 2 provides information on U.S. government efforts since the early 2000s until about 2017 to reinitiate R&D on advanced reactors and their fuel cycle options. This section focuses on U.S. government efforts since 2017 to develop these reactors and fuel cycles.

3.3.1 Congressional Actions

As introduced in Chapter 1, Congress’s enactment of the Nuclear Energy Innovation Capabilities Act in 2018 (NEICA) (Public Law 115-248) defined advanced reactors and their potential improvements in comparison with existing LWRs. NEICA also included a number of provisions regarding testing and demonstration of advanced reactor concepts. In particular, it required DOE to “determine the mission need for a versatile reactor-based fast neutron source, which shall operate as a national user facility.” The user facility would “provide at a minimum: (1) fast neutron spectrum irradiation capability, and (2) capacity for upgrades to accommodate new or expanded research needs.” In addition, NEICA required DOE to “carry out a program for enhancing the capability to develop

new reactor technologies through high-performance computation modeling and simulation techniques.” As to the roles of the private sector, NEICA authorized “a program to enable the testing and demonstration of reactor concepts to be proposed and funded by the private sector” and directed DOE to “leverage the technical expertise of relevant federal agencies and the national laboratories in order to minimize the time required to enable construction and operation of privately funded experimental reactors at national laboratories or other DOE-owned sites.”

The Nuclear Energy Innovation and Modernization Act (NEIMA) (Public Law 115-439) was signed into law in January 2019 and reiterated the attributes of advanced reactors, as defined by NEICA. NEIMA’s main purpose was “to provide a program to develop the expertise and regulatory processes necessary to allow innovation and the commercialization of advanced nuclear reactors.” NEIMA directed the U.S. NRC to develop, within the existing regulatory structure, procedures and processes for licensing of advanced commercial reactors, as well as research and test reactors. Moreover, not later than December 31, 2027, the U.S. NRC is to “complete a rulemaking to establish a technology-inclusive, regulatory framework for optional use by commercial advanced nuclear reactor applicants for new reactor license applications.”

The relevant aspects for advanced nuclear reactors of the Infrastructure Investment and Jobs Act (Public Law 117-58), enacted November 15, 2021, are described in Section 3.3.2.

3.3.2 U.S. Department of Energy’s Office of Nuclear Energy’s Programs and Goals

On January 8, 2021, DOE-NE published its Strategic Vision for supporting existing reactors, enabling deployment of advanced reactors, developing advanced fuel cycles, maintaining U.S. leadership in nuclear energy technology, and enabling a high-performing organization (DOE-NE, 2021a). For advanced reactors and fuel cycles, the Strategic Vision includes the following goals, objectives, and indicators.

A key goal is to enable advanced reactor deployment to meet these objectives:

• “Reduce risk and time needed to deploy advanced nuclear technology.
• Develop reactors that expand market opportunities for nuclear energy.
• Support a diversity of designs that improve resource utilization.”

The deployment performance indicators are

• “By 2024, demonstrate and test a fueled microreactor core fabricated by advanced manufacturing techniques.
• By 2025, enable demonstration of a commercial U.S. microreactor.
• By 2027, demonstrate operation of a nuclear-renewable hybrid energy system.
• By 2028, demonstrate two U.S. advanced reactor designs through cost-shared partnerships with industry.
• By 2029, enable operation of the first commercial U.S. small modular reactor.
• By 2035, demonstrate at least two additional advanced reactor designs through partnerships with industry.”

For developing advanced nuclear fuel cycles, the goal is to attain these objectives:

• “Address gaps in the domestic nuclear fuel supply chain.
• Address gaps in the domestic nuclear fuel cycle for advanced reactors.
• Evaluate options to establish an integrated waste management system.”

The development performance indicators are

• “By 2021, begin procurement process for establishing a uranium reserve.²³
• By 2022, demonstrate domestic [high-assay low-enriched uranium] HALEU enrichment.²⁴

___________________

²³ DOE did not achieve this objective.

²⁴ As of April 2022, it appeared that this objective’s achievement could be delayed. See Chapter 4 for more details.

• By 2023, make available up to five metric tons of HALEU from non-defense DOE material.²⁵
• By 2030, evaluate fuel cycles for advanced reactors.”

The remaining parts of this subsection describe the DOE-NE programmatic activities to support the above goals, objectives, and performance indicators.

In May 2020, DOE-NE launched the Advanced Reactor Demonstration Program (ARDP) to partner with domestic private industry to demonstrate advanced nuclear reactors. This program includes three levels of funding opportunities:

• Advanced Reactor Demonstration awards, which provided $160 million in initial funding for cost-shared reactor demonstrations to be operational within 5–7 years of the award.
• Risk Reduction for Future Demonstration awards, which provided $30 million in initial funding to support reactor designs that could be operational within 10–14 years of the award.
• Advanced Reactor Concepts 2020 awards, which provided $20 million of funding to help move early-phase reactor designs toward demonstration.

In October 2020, DOE-NE announced the selection of TerraPower and X-energy as the Advanced Reactor Demonstration award winners for demonstration of the Natrium and Xe-100 reactor designs, respectively. On December 16, 2020, DOE announced the selections of five teams to receive funding under the Risk Reduction for Future Demonstration program: Kairos Power, LLC, for the Hermes Reduced-Scale Test Reactor; Westinghouse Electric Company, LLC, for the eVinci microreactor; BWXT Advanced Technologies, LLC, for the BWXT Advanced Nuclear Reactor; Holtec Government Services, LLC, for the Holtec SMR-160 Reactor; and Southern Company Services, Inc., for the Molten Chloride Reactor Experiment. Finally, in late December 2020, DOE announced the three Advanced Reactor Concepts 2020 award winners: Advanced Reactor Concepts, LLC, for the Inherently Safe Advanced SMR for American Nuclear Leadership; General Atomics for the Fast Modular Reactor Conceptual Design; and the Massachusetts Institute of Technology for the Horizontal Compact High Temperature Gas Reactor.

On November 15, 2021, the Infrastructure Investment and Jobs Act authorized $3.2 billion through FY2027 for the Advanced Reactor Demonstration awardees. With the previous authorizations from FY2020 and FY2021, these demonstration projects are fully authorized. In addition, the Act appropriates $2.4 billion from FY2022 through FY2025 for the ARDP’s identified existing awardees. This appropriation applies to all parts of the ARDP, including the Risk Reduction and Advanced Reactor Concepts 2020 awardees. The DOE-NE portion to each of the two awardees (TerraPower’s Natrium and X-energy’s Xe-100) for the Advanced Reactor Demonstration will be $1.89 billion and $1.25 billion, respectively. With the company’s private funding included, the total financial support will be approximately $4.0 billion for the TerraPower project and $2.5 billion for the X-energy project over a 7-year period.

DOE-NE’s support for advanced reactor development and deployment involves working with and through national laboratory–led R&D, university research programs, and cost-shared private–public industry partnerships. The objectives are as follows:

• “Conduct focused research and development to reduce technical barriers to deployment of advanced nuclear energy systems.
• Develop technologies that can enable new concepts and designs to achieve enhanced affordability, safety, sustainability, and flexibility of use.
• Sustain technical expertise and capabilities within national laboratories and universities to perform needed research.
• Engage with Standards Developing Organizations (SDOs) to address gaps in codes and standards to support advanced reactor designs.

___________________

²⁵ As of April 2022, this objective also appeared unlikely to be achieved.

• Collaborate with industry to identify and conduct essential research to reduce technical risk associated with advanced reactor technologies” (Caponiti, 2020b).

For the specific advanced reactor technologies within the scope of this study, DOE-NE’s main research areas are as follows:

• Fast reactor technologies
  • Demonstration of feasibility of advanced systems and component technologies
  • Methods and code validation to support design and licensing
  • Qualification of legacy metallic fast reactor fuel performance data
• Gas reactor technologies
  • Advanced alloy and graphite materials qualification
  • Scaled integral experiments to support design and licensing
  • TRISO-coated particle fuel development and qualification
• Molten salt reactor technologies
  • Investigation of fundamental salt properties
  • Materials, models, fuels, and technologies for salt-cooled and salt-fueled reactors

DOE-NE’s Gateway for Accelerated Innovation in Nuclear (GAIN) program provides “access to technical, regulatory, and financial support” for nuclear energy developers, both existing commercial reactor technologies and advanced reactors (Caponiti, 2020b). Situated at Idaho National Laboratory, GAIN and its website provide a number of resources.²⁶ Notably, GAIN has funding opportunities intended to help industry accelerate deployment of commercial reactors. It can offer access to technical and regulatory support, and has available a legacy U.S. nuclear research database, advanced computational tools, and access to nuclear research expertise. GAIN also convenes several workshops and webinars on topics such as HALEU, fast reactor technologies, and small modular reactors (Caponiti, 2020b).

DOE-NE provides additional U.S. industry opportunities for development of advanced nuclear technologies through its U.S. Industry Opportunities for Advanced Nuclear Technology Development program. These opportunities are intended to support innovative designs and technologies that have significant potential to increase the economic prospects of nuclear power in the United States. DOE-NE expects the resulting products to be manufactured in the United States (Caponiti, 2020b). DOE-NE has two award cycles annually for these opportunities, and the awards are described on DOE-NE’s website.²⁷

DOE-NE also supports programs to ensure a supply of HALEU; to develop fast reactor fuels; to support R&D on materials recovery and waste forms; to develop materials protection, accounting, and control technology; and to build the Versatile Test Reactor (Griffith, 2020). For domestic HALEU acquisition pathways, the three approaches are (1) downblending to HALEU enrichments from the recovery of high-enriched uranium reprocessed from spent fuel, such as from Experimental Breeder Reactor II, naval reactors, and the Advanced Test Reactor; (2) contracting with Centrus Energy Corporation for the advanced centrifuge demonstration project; and (3) producing HALEU from limited amounts of material in DOE’s uranium inventory. More details on HALEU acquisition and supply chain are in Chapter 4.

For developing fast reactor fuels, three notable R&D areas are (1) metallic fuels for closed fuel cycles and actinide transmutation for waste management missions; (2) fuels for once-through fast spectrum reactors; and (3) fuels for high-temperature, fast-spectrum reactors. For R&D on materials recovery and waste forms, four topics of interest to DOE-NE are (1) investigating fundamental fuel cycle chemistry; (2) supporting transformative studies in domestic molten salt processing and chemistry; (3) developing advanced waste forms for aqueous and salt processing; and (4) conducting a joint fuel cycle study with the Republic of Korea “to assess the technical feasi-

___________________

²⁶ See https://gain.inl.gov/SitePages/Home.aspx.

²⁷ See, for example, awardees for 2021: https://www.energy.gov/ne/articles/doe-awards-85-million-advance-promising-nuclear-technologies.

bility of electrochemical recycling for managing used fuels” (Griffith, 2020). The Advanced Reactor Safeguards Program was also established by DOE-NE in 2020 as part of the appropriations for the ARDP (Cipiti, 2021a).

The Versatile Test Reactor (VTR) project was launched in February 2019 and was “proposed to be a 300 MWth sodium-cooled, fast spectrum reactor capable of testing advanced nuclear fuels and materials for the next generation of nuclear reactors” (Griffith, 2020). In summer 2021, the House and Senate Appropriations Committees zeroed out funding for the VTR facility in FY2022, giving no reason at that time for doing so. DOE has estimated the cost for completing the VTR as between $2.6 billion and $5.8 billion, which would require annual appropriations substantially above DOE-NE’s typical budgets. The Biden administration stated that it planned to delay VTR’s construction to follow the building of TerraPower’s Natrium reactor, which shares similar base technology as the VTR and already has support from DOE’s ARDP (AIP, 2021). In an October 2021 interview with Physics Today, DOE stated that a plan was being developed and “will address collaborations with industry, such as the VTR/TerraPower [combination], as one of the methods VTR will use to establish public-private partnerships to complete this critical piece of nuclear energy research and development infrastructure” (Kramer, 2021). The final environmental impact statement for the VTR was released in May 2022 (DOE, 2022c). Section 3.4 discusses testing and test reactors in greater detail.

Given budgetary constraints, DOE-NE will need to make difficult decisions in the coming years about its programs. Congressional staff from the U.S. Senate and House Appropriations Committees told the committee in December 2021 that any particular DOE-NE program could consume the entire DOE-NE budget, and there is no feasible way to fully fund, even with industry cost sharing, all the advanced reactors in the ARDP. Congressional staff also said that having a diversity of reactors in the ARDP portfolio is a useful competitive market approach, but industry cannot rely solely on government support. Government support has the important role of maintaining continuity of R&D across programs (Goldner and McKee, 2021). However, constrained budget environments are likely to persist for the coming years. At the same meeting, Dr. Kathryn Huff, then–principal deputy assistant secretary for DOE-NE, acknowledged that DOE-NE has competing priorities and provided her view that, while continued R&D in reprocessing is important for keeping U.S. fuel cycle options open, cost reductions are needed because reprocessing is presently too expensive (Huff, 2021).

3.3.3 The Nuclear Energy Advisory Committee

In 1998, the Nuclear Energy Advisory Committee (NEAC) was established to provide independent advice to DOE-NE “on complex science and technical issues that arise in the planning, managing, and implementation of DOE’s nuclear energy program” (DOE-NE, n.d.-b). Operating in accordance with the Federal Advisory Committee Act, NEAC has a diverse membership with U.S. and foreign experts from industry, national laboratories, and universities. NEAC has periodically reviewed DOE-NE’s programs and has provided advice and recommendations on DOE-NE’s plans, priorities, and strategies for addressing the scientific and engineering challenges of R&D efforts (DOE-NE, n.d.-b). In addition, the secretary of energy or the assistant secretary for DOE-NE can request NEAC to provide advice on national policy and technical aspects of pending DOE decisions on nuclear energy programs. NEAC forms subcommittees to address specific issues related to nuclear energy. The Advanced Reactor Pipeline Subcommittee’s objective is to provide an independent and expert review of efforts within DOE-NE related to advanced reactors (DOE-NE, 2019). In December 2021, NEAC’s charter was renewed by DOE for another 2 years (DOE, 2021). In February 2022, 11 new members were appointed to NEAC, and DOE-NE announced that it will revise the existing structure in order to have NEAC focus on advising “on current priorities rather than reviewing projects and initiatives that have already been completed” (DOE-NE, 2022a).

3.3.4 U.S. Nuclear Regulatory Commission’s Regulatory Programs on Advanced Reactors and Associated Fuel Cycles

As mentioned above, NEIMA directs the U.S. NRC to develop clear regulatory procedures for licensing commercial advanced reactors, as well as for test and research reactors and facilities. As discussed in Section 2.5.1 of Chapter 2, from 2008 to 2016 the U.S. NRC examined potential rulemaking for a commercial reprocessing

facility but suspended this activity in 2016 because of limited industry interest. Section 3.2.3 provides additional information about the interest expressed by advanced reactor developers in reprocessing and recycling. Furthermore, as discussed in Chapter 6, the U.S. NRC is considering options for resuming a rulemaking for “enhanced security” of nuclear materials that was suspended in 2016. If developed, this rule would update requirements and address gaps for a range of nuclear materials that may be used in advanced reactor fuel cycles.

During its September 2020 information-gathering meeting, the committee received briefings from the U.S. NRC staff on the regulatory programs relevant for advanced reactors and associated fuel cycles (Regan et al., 2020). The U.S. NRC’s preparations for regulating advanced reactors and fuel cycles include both front-end (especially fuel supply) and back-end (especially spent fuel management) considerations. Also, the U.S. NRC staff underscored that “existing risk-informed and performance-based regulatory framework for licensing and oversight has sufficient flexibility to ensure the safe and secure operation of the complete fuel cycle for advanced reactors.” Moreover, the U.S. NRC staff are continually coordinating within their agency and with other government agencies, such as DOE, to collect relevant information and gain insights as to industry developments. Potential licensee applicants are encouraged “to engage early and often in the interest of ensuring complete, high-quality license applications” (Regan et al., 2020).

The U.S. NRC has significant experience in licensing several nuclear fuel cycle facilities, notably for the existing LWR fleet. This regulatory framework also lays the foundation for the regulatory process for advanced reactors’ fuel cycles, but the advanced reactor systems could pose challenges different from those of the LWR systems. Both ARDP awardees have engaged the U.S. NRC in preapplication discussions associated with their nuclear plant designs. In addition, U.S. NRC staff have conducted preapplication meetings with one developer (X-energy) to discuss plans for fuel fabrication in a facility licensed by the U.S. NRC. In April 2022, X-energy submitted an application for a Category II license to fabricate TRISO-coated particle fuel forms using up to 19.75 percent enriched uranium. In addition, the U.S. NRC has contracted with the national laboratories to identify potential hazards associated with metallic fuel fabrication and fuel salt preparation for molten salt reactors (Regan et al., 2020). The U.S. NRC’s security and safeguards activities related to advanced reactors’ fuel cycles, especially use of HALEU, are covered in Chapter 6.

Regarding disposal of high-level waste and spent fuel from advanced reactors, the U.S. NRC staff were not “aware of any technical issues that would require changes to its [the U.S. NRC’s] disposal safety requirements to accommodate other fuel types and waste forms” (Regan et al., 2020), and the U.S. NRC is ready to support the national program when given statutory direction. In addition, the U.S. NRC has experience in approval of transportation packages and storage systems for TRISO and metallic fuels. Moreover, the U.S. NRC is “completing technical evaluations on transport, storage, and disposal activities of advanced reactor fuel designs to identify potential information needs and determine whether additional updates to safety review guidance may be warranted” (Regan et al., 2020). Chapter 5 provides more detailed information about transportation, storage, and disposal aspects of advanced reactors and their fuel cycles.

3.4 PROTOTYPING, TESTING, AND TEST REACTORS

The following sections describe the role of prototype and test reactors, as well as other types of test facilities, in supporting R&D on advanced reactors and fuels, and provide the committee’s views on this needed infrastructure. According the U.S. NRC, a prototype plant is a nuclear reactor used for testing design or safety features. Such a plant could have “additional safety features to protect the public and the plant staff from the possible consequences of accidents during the testing period” (10 CFR § 50.2; U.S. NRC, n.d.).²⁸ A test reactor could be a smaller-scale version of an advanced reactor for the purposes of providing data for U.S. NRC testing requirements, proving the concepts of the advanced reactor’s materials, components, structures, and various systems. Testing facilities for reactors or other neutron sources can be used to evaluate the behavior under irradiation of fuels and other materials intended for use in advanced reactors (U.S. NRC, n.d.).

___________________

²⁸ This is discussed in the 2007 rulemaking amending 10 CFR § 52 (U.S. NRC, 2007) and in U.S. NRC (n.d.).

3.4.1 Prototyping

Prototyping is a necessary and expensive activity of new reactor and fuel cycle development. Effective prototyping will speed development efforts; improve modeling codes for specific reactors industry wide; reduce costs through the elimination of unknowns and associated conservatism; and provide for increasing regulatory confidence, which is necessary to translate into the savings that so many developers are anticipating. Prototyping is an important part of performance validation for materials, components, reactor systems, and accident scenarios.

Much of the original material, fuel, and component testing for the LWR commercial nuclear industry was performed by government subsidized and operated reactors, including the Shippingport demonstration project in the 1950s. DOE is currently pursuing a similar approach for advanced reactors by proposing construction of the VTR and cosharing the development (including prototyping) expenses with some advanced reactor developers. See Section 3.3.2 for discussion of the VTR program.

Effective prototyping for any nuclear reactor system would meet several needs:

• integrated system performance;
• ability to achieve servicing and inspection improvements;
• proof—under a variety of accident scenarios—of innovative safety improvements, such as natural circulation heat removal; and
• validation of design models.

Notably, advanced reactors, which propose using a wider variety of coolants and fuel types, will have more comprehensive prototyping requirements. In addition to the prototyping described above, advanced reactor developers will have to do substantial prototyping of basic materials’ strength and performance when subjected to entirely different operating environments, including differing coolants, neutron flux, temperature, and pressures. Much of this basic prototyping can be accomplished in test reactors where available.

3.4.2 Testing and Test Reactors

Test reactors are a part of the critical infrastructure of the nuclear industry worldwide. They are essential for testing fuel and fuel burnup, materials and welds under a variety of operating environments, operating conditions, and the development of increasingly complicated computer codes used industry wide. They enable innovation and refinement of engineered systems and components and are key to projecting performance for advanced reactors that inherently lack operational data to validate system and component performance under the specific conditions of the advanced design.

Test and operational performance data are critical for understanding material and fuel performance, developing operating parameters and regulatory infrastructure, manufacturing higher-performance and longer-lasting components, and determining inspection regimes. These data are used by research institutions, government laboratories, regulatory agencies, and manufacturers to drive the innovation needed to make the advances necessary to overcome the wide array of challenges faced by industry in successfully deploying advanced designs. Welds, for example, are tested under compression and tension, at different temperatures and pressures, with exposure to different coolants and contaminants, and in different flux fields in order to determine short- and long-term degradations, strengths and weaknesses, potential failure modes, and resistance to shock.

Detailed analysis of performance is then used by the industry to improve designs, codes, methods of manufacturing, material composition, and fuel composition and loading. Regulatory agencies use the data to develop new standards, regulations, and operational parameters. Importantly, test reactors are designed to provide “accelerated data”—a year of data from the test environment may correspond with several years of anticipated operational history. Such data are critical in early resolution of design flaws and in developing modified methods of manufacturing. The data are used by the industry worldwide to drive the innovation necessary for successful deployment of advanced technologies. Without the necessary data, advanced reactors could be unnecessarily plagued by

development and operational problems, mandating in-service repairs, which can be costly and time consuming in the nuclear industry.

Test and operational data also confirm performance of the original design and subsequent modifications, improving online time, eliminating failure modes, and—through increasing robustness—resulting in longer-lived components and potentially decreasing or eliminating the need for servicing and inspection. Such data have a significant impact on economic viability and the extent of regulatory oversight, as well as safety and perceptions of safety both within the nuclear power community and by interested stakeholders.

The different fuel types and coolants impact every aspect of the reactor. Every component within the different neutron fluxes or exposed to differing nonwater coolants required by advanced fuel cycles will have different failure mechanisms. By changing the fuel and coolant types, a whole new set of evaluation data will be required. Reactor components exposed to the higher neutron flux of advanced fuels and differing coolants will be subject to new methods of degradation, which are not currently well understood. For example, a weld on an internal pump or a submerged heat exchanger in a pool reactor may experience different types and speeds of stress corrosion cracking. Additionally, as components come together in the reactor, they will experience different types of thermal expansion and contraction—not only in general, but also based on their position in the reactor. For example, reactor core internals, which will be exposed to both different neutron fluxes and coolants from advanced fuel cycles, will need to be tested extensively to ensure their performance in the new environment. That makes it critically important to have a test reactor with the capability of independent loops that can maintain temperature and pressure (to establish thermal hydraulic testing conditions that will mimic the reactor concept being tested) and simulate specific neutron flux. A test reactor with this capability that supports instrumented leads experiments can help speed up the extensive testing that will be necessary to support advanced fuel cycles.

Over the history of the nuclear industry, there have been a large number of test reactors, which provided the data driving the early industry innovation, improved designs, and development of manufacturing and operating methods and parameters. The early nuclear industry was heavily supported by government policy, and the government directed and funded test facilities, many of which also supported national security policy goals. The early industry was also supported substantially by a strong backbone of dual-use designers and manufacturers that focused on both commercial and national security projects. After a government-sponsored commercial test with the Shippingport demonstration breeder reactor, the nuclear industry settled into tested and established methods, focusing largely on LWRs. As the industry consolidated around LWRs with increasingly established technology—and as facilities aged, computational models matured, and more direct operational data became available—the number of test reactors declined. At the same time, the pursuit of advanced technologies and fuel cycles floundered and generally fell under the purview of government and academic institutions with no realistic path forward toward commercialization.

In the late 1970s, changing economics and the Three Mile Island accident altered the dynamics of the U.S. commercial nuclear industry. While continuing to operate existing plants, the nuclear design, manufacturing, and construction components of the industry significantly contracted, resulting in a severe reduction in the number of nuclear qualified (“N-stamp”) firms licensed to perform nuclear work. As the industry contracted, complicated supply chains for Western reactors became more globalized and dispersed, and what had been an almost entirely self-contained and self-reliant U.S. industry now has supply chain partnerships and ownership spanning the globe. With the exception of certain national security programs, this dispersion (and the declining market) has weakened the U.S. nuclear industrial base available for design, engineering, and manufacturing innovation. This weakening in active test reactors, new build expertise, and manufacturing at a commercial level will impact how quickly technology development for advanced reactors and fuel cycles can advance in the United States, especially without considering foreign supply chain expertise.

As mentioned in Section 3.3.2, a major question for development and deployment of test reactors is, Who should pay for the reactors? While Congress is requiring DOE to form public–private partnerships to support the VTR project, DOE officials are concerned that a commercial advanced reactor developer would not want to finance construction of test facilities that would also benefit competitors (Kramer, 2021). Other test reactors at Idaho National Laboratory, such as the Advanced Test Reactor (ATR), have been the government’s responsibility for

building and operating. However, one option for funding operations for the VTR or a VTR-like test reactor would be to charge an operational fee for proprietary research but make access free if the experimental results would be made publicly available, while the government would solely or largely pay the construction costs (Kramer, 2021).

The ATR is a one-of-a-kind test reactor. It began operations in 1967, and in 2021 was undergoing its sixth overhaul and upgrades. ATR has provided testing capabilities for U.S. naval reactors, other federal agencies, university researchers, and industry. Although the ATR is a type of pressurized water reactor (PWR), it operates at very low pressures and temperatures compared with commercial PWRs. It uses a beryllium reflector to concentrate neutrons in the core of the reactor in order to provide high-flux thermal neutrons. In addition, ATR has a unique clover-leaf design with lobes that can be operated at different power levels to allow for multiple simultaneous experiments. Industry partners can access the ATR via the National Research Innovation Center, the Gateway for Accelerated Innovation in Nuclear program, or the Nuclear Science User Facilities program (INL, 2021a; NSUF, n.d.). While the ATR has capabilities relevant for testing some types of advanced reactors, the major missing capability is fast neutrons. Researchers at Idaho National Laboratory reported in 2017 that “ATR irradiations performed using cadmium shrouding are sufficiently prototypic that they can be used with confidence in the development and testing of fast reactor fuels” (Harp et al., 2017).

However, the usefulness of such a solution is debatable. Cadmium shrouding does not result in an increase of the fast neutron flux, which means that the ATR is not suitable for shortening the time required for generating and investigating neutron radiation damage in materials used, or planning to be used, in today’s or future reactors. Presently, such an accelerated testing capability is only available in the BOR-60, a sodium-cooled fast test reactor that has been operating in Russia for more than 50 years. Direct access to the BOR-60 reactor by U.S. R&D organizations has been difficult at times, and a facility such as the proposed VTR would have significant roles in the development of advanced nuclear energy generation technologies. Worldwide, there is a lack of testing facilities that can produce high-energy neutron fluxes useful for research on fast reactors’ fuel and materials (WNA, 2021i).

In the European Union, Belgium has the largest-power (125-MWth) materials testing reactor, the BR2, which has “a unique adaptable core configuration” and has a memorandum of understanding with DOE-NE’s Nuclear Science User Facilities to allow U.S.-based researchers to apply to use the reactor (SCK-CEN, 2018). At Cardarache, France, the 100-MWth Jules Horowitz LWR is under construction and is designed for testing of fuels and materials for advanced reactors (CEA, n.d.). This reactor would also be available for international collaboration when completed.