Reducing the Use of Highly Enriched Uranium in Civilian Research Reactors (2016)

3

Research Reactors and Their Uses

This chapter provides background information on research reactors, including the history and performance of research reactors, discusses alternative sources of neutrons, and provides an overview of general uses of civilian research and test reactors, including specific uses of high performance research reactors (HPRRs). It concludes by discussing the future role of research reactors in supporting science, engineering, and medicine.

HISTORY OF RESEARCH REACTOR DEVELOPMENT

The first self-sustaining nuclear reactor, Chicago Pile-1 (CP-1), was assembled in 1942, producing a maximum power of 200 W. Within 20 years, the design of research reactors had progressed to the point that the average neutron flux (number of neutrons per unit area, per second) had increased by nearly nine orders of magnitude (see Figure 3.1). The availability of highly enriched uranium (HEU) fuel allowed for much of this increase. Figure 3.1 shows how the thermal neutron¹ flux in research reactors has evolved over time.

The trend is clear: A flux of thermal neutrons in the reactor core of roughly 10¹⁵ neutrons per square centimeter per second (n/cm²-s) was achieved in the mid-1960s and has not been greatly exceeded since (see Table 3.1 for the listing of maximum flux for the highest-performance existing research reactors). This limit arises because the amount of cooling required increases with the flux, or power density, of the reactor. Beyond

_________________________

¹ Thermal neutrons have an average velocity of about 2 km/s.

FIGURE 3.1 Evolution of neutron flux in a research reactor over time. For early neutron sources and for the named research reactors, the vertical axis gives the approximate maximum thermal neutron flux in or close to the reactor core, while the horizontal axis is the year in which the facility first produced neutrons. SOURCE: Based on data from multiple sources; see Table 3.1.

a certain power density (the level of which depends on the reactor design), supplying the required level of cooling may not be possible. The high neutron flux produced by the reactors shown in Figure 3.1 allows for execution of many critical missions.

USES OF RESEARCH AND TEST REACTORS

Research reactors are indispensable tools in the education and training of reactor operators and nuclear engineers, basic and applied research in a wide range of scientific areas, and the production of scientifically and technologically important materials, such as radioisotopes. Such reactors are also used for testing new types of nuclear fuel and studying the radiation resistance of new materials and electronic devices. A technical report published by the International Atomic Energy Agency (IAEA, 2014) describes in detail the typical uses of research reactors and outlines the necessary technical criteria required for each application. The IAEA report categorizes the uses of research reactors into three main areas—training and education, irradiation applications, and extracted beam applications; the same groupings are followed here.

Training and Education

Research reactors are well suited for training operators of nuclear power plants, because they provide hands-on access to reactor systems that are effectively hidden in power reactors and the capability to simulate abnormal conditions for training purposes (which cannot be done at power plants [Agasie et al., 2011]). Any functioning research reactor, irrespective of its operating power, can also be used for broader training and education, including formal education of nuclear engineers and radiological technicians, as well as educational events involving other students and the general public.

Irradiation Applications

Irradiation applications of research reactors generally involve inserting specimens into the reactor (in either the in-core or near-core regions where the neutron flux is highest) to induce radioactivity, produce isotopes, or induce radiation damage.

Research Reactors for Transmutation Applications

Transmutation is the conversion of elements and isotopes into other elements or isotopes through reactions in the nucleus.² One useful application of nuclear transmutation involves the creation of dopants within a semiconductor. For example, neutron irradiation of pure silicon transmutes silicon-30 (³⁰Si), which constitutes roughly 3 percent of natural silicon, to the dopant phosphorus, ³¹P, thereby producing an n-type semiconductor. This neutron transmutation doping (NTD) technology is valued because it provides extremely uniform doping that leads to superior performance of silicon in high-power electronic applications. Medium-flux reactors are well suited to this application, because they achieve the required very uniform and precise irradiation levels in a reasonable length of time. These constraints limit the application of NTD to a few research reactors that have been optimized for this application.

At low power levels, research reactors may be used for analytic techniques such as neutron activation analysis (NAA), wherein irradiation produces radionuclides characteristic of the elements in a sample. These radionuclides often decay by emitting gamma rays with characteristic energy and intensity. Analysis of these gamma rays allows the identity and quantity of particular elements in a specimen to be assessed. The IAEA

_________________________

² Adsorption of a neutron causes a change in atomic number, producing another element. Fission (a splitting of the nucleus) produces two nuclides having different atomic numbers and masses from the original.

estimates that NAA is the most widely used application of research reactors after education and training. Customers for NAA include industries such as mining or agriculture (e.g., for determination of trace elements in geologic matrices or the distribution of agricultural chemicals in soils), government agencies (e.g., for forensics), medical centers (e.g., for testing doses of pharmaceutical products), and research institutions (e.g., for determination of the origin of archeological specimens such as pottery). Although not all elements can be analyzed using NAA, the method has the advantage of performing compositional analysis without chemically altering a specimen. In addition, NAA is largely independent of matrix effects, is suitable for materials that are difficult to dissolve, and is relatively insensitive to sample contamination.

Radioisotopes have a wide range of applications in nuclear medicine, industry, and research. Production of these isotopes in research reactors is based on neutron absorption by a target material introduced into the reactor core. In general, the quantity of an isotope that can be produced in a given amount of time will increase as the neutron flux increases. Although the production rate may be linearly proportional to the neutron flux in some cases, for isotopes that require multiple successive neutron capture events the production rate is proportional to higher powers of the flux. More than 80 percent of diagnostic medical procedures utilize technetium-99m, which is a decay product of molybdenum-99 (⁹⁹Mo) (OECD, 2015). Many research reactors produce or are planning to produce this radioisotope for commercial purposes. However, other medically useful isotopes are also produced in research reactors, and new diagnostic procedures involving radioisotopes continue to be developed. Because the isotopes decay with half-lives measured in hours or days, they must be produced quickly and in large quantities, so a reactor with high flux (on the order of 10¹⁴ n/cm²-s) is needed to produce them. For example, the University of Missouri Research Reactor (MURR) and the Belgian Reactor-2 (BR2) are each able to produce a wide variety of radioisotopes (Butler and Foyto, 2015; Ponsard and Blowfield, 2010).

Only the highest-flux reactors (with flux on the order of 10¹⁵ n/cm²-s) can generate significant quantities of some isotopes such as Californium-252 (²⁵²Cf) (used for neutron radiography and the start-up of nuclear reactors, for example) because capture of several neutrons in very rapid succession is required and these events are relatively rare. The High Flux Isotope Reactor (HFIR) in the United States and the SM-3 reactor in Russia are the only two research reactors in the world currently producing ²⁵²Cf.

The primary alternative to research reactors for transmutation applications is the use of accelerators, either for direct irradiation with charged particles or for generation of neutrons. Accelerators are in widespread use at hospitals around the world for medical isotope production. On-site pro-

duction facilities are necessary for the shortest-lived isotopes, and direct use of charged-particle accelerators are generally more effective for producing neutron-deficient isotopes such as fluorine-18 (¹⁸F).

The rate at which transmutation products can be produced increases as the magnitude of the flux increases. An equally important parameter for most applications is the volume over which such a uniform neutron flux is achievable. While accelerator-based systems can be developed to provide reasonable irradiation volumes, they often suffer from substantial flux gradients. For accelerator-based isotope production applications, the flux gradient results in reduced throughput; for accelerator-based transmutation doping applications, such gradients are counter to a primary advantage of the transmutation doping approach—uniform production of dopants.

Research Reactors for Materials Testing

An important class of research reactors, often referred to as Materials Test Reactors (MTRs), is used to test the behavior of structural materials and nuclear fuels for the nuclear power industry under prototypical irradiation conditions. Very high-energy neutrons, having a velocity greater than about 6,000 km/s, displace atoms and cause changes in the microscopic structure of materials. These microstructural changes accumulate over long periods of time in radiation environments, with the rate of change depending on the neutron flux. Given the consequences of material failure in many applications in the nuclear power industry, it is necessary to experimentally confirm material performance after long radiation exposures. To avoid impractically long irradiation times, neutron fluxes in the experiment must be much higher than those experienced in the normal operating environment. For a material that is designed to remain in a power reactor for up to 60 years, even a flux level 20 times higher than the power reactor requires 3 years of irradiation to adequately confirm its behavior, hence the need for HPRRs for this application.

MTR studies help establish the safety of power reactors and validate fuel behavior in operational and accident situations. Examples of materials testing include the following:

• Nuclear reactor fuel, including new high-density fuels for HPRRs;
• Metals used in power reactors (such as the steel for pressure vessels) to determine the service lifetime; and
• Electronics expected to operate in high-radiation environments.

Fuel and materials testing supports exploration of advanced materials to further improve safety and performance for the existing fleet of power reactors. In addition, advanced reactor concepts often combine novel fuel,

cladding, and coolant materials, requiring irradiation experiments to confirm their performance under reactor conditions.

MTRs are used to test the behavior of new fuels—including the new low enriched uranium (LEU) fuels that will be used in conversion of research reactors. The fuels are irradiated across a set of conditions that simulate the fuels’ expected operating and accident scenario conditions. For this reason, MTR neutron fluxes need to be large enough for significant radiation doses to be delivered in a relatively short time. MTRs have multiple test positions located within and near their cores so that test samples, including full-scale fuel elements and assemblies, can be irradiated under neutron spectra and gamma-to-neutron ratios that are relevant to the fuel or structural material under test. The Advanced Test Reactor (ATR) in the United States, BR2 in Belgium, and MIR.M1 in Russia are particularly important high-flux research reactors used for materials testing.

Materials testing is focused almost exclusively on understanding the impacts of reactor radiation environments on fuels and structural materials. Obviously, reactors themselves provide an ideal environment for such testing, particularly when the tests are sensitive to neutron and gamma spectra, or the ratio between them. Nevertheless, other systems have been proposed for performing materials irradiation in reactor-like environments.

Enhancements to large accelerator-based systems (e.g., the Los Alamos Neutron Science Center and the Spallation Neutron Source [SNS]) can provide modest irradiation volumes with conditions that are adequate for some types of materials testing. These may be appropriate for the smallest-scale irradiation experiments but do not allow for full-plate or element-scale experiments. In addition, it is difficult to use accelerators to produce the same kinds of material damage that are present in reactors because the neutron spectra are significantly different.

In general, research reactors are preferred for materials testing experiments because their neutron spectra are best matched to the expected operating environments. Different neutron energy spectra may produce different numbers of atomic displacements and different amounts of transmuted material. The materials properties will depend, in general, on both.

Extracted Beam Applications

Beams of neutrons are extracted from the research reactor via an evacuated metal tube that extends through reactor shielding. This allows neutrons produced in the core to be used outside of the reactor for a number of scientific applications.

Neutron imaging applications use beams of neutrons to produce images of neutron attenuation, just as a dental radiograph produces an image using x-ray attenuation. Neutrons and x-rays interact differ-

ently with materials: x-ray transmission through a sample decreases with increasing atomic number but neutron absorption does not. This makes thermal and cold neutrons³ ideal for imaging materials containing light atoms such as hydrogen, which make up a significant fraction of both biological and organic materials, but which are essentially invisible to x-rays. One example is the study of the location of water within a working fuel cell (Satija et al., 2004).

Neutron scattering has been used to probe the structures of materials and the motions of atoms in materials since the first nuclear reactors were built. Neutrons are neutral particles that interact with nuclei in a material sample, so they penetrate materials easily, providing information about the interior rather than the surfaces of materials. The weak interaction allows neutrons to pass through containers needed to keep samples at a particular temperature or pressure. A recent example is the study of the movement of lithium ions in a working lithium-ion battery (Wang et al., 2012). Neutrons interact with atomic nuclei and are scattered differently by different isotopes of the same element. In addition, neutrons are sensitive to magnetism within a sample and can be used to produce maps of internal magnetization.

Over the past six decades, neutron scattering experiments and techniques have improved basic scientific understanding in condensed matter physics, chemistry, polymer science, life sciences, sustainable energy research, sensors, smart materials, mechanical engineering, archeology, nanotechnology, and biotechnology. Examples range from the determination of the atomic structure of the first high-temperature superconductor through verification of theories that describe the distribution and motion of molecules in a melted polymer. Several Nobel Prizes have been awarded for scientific work that rests ultimately on experimental data obtained with neutron scattering.

Neutron scattering is a signal-limited technique, meaning that the quality and quantity of information that can be obtained is limited by the number of neutrons of a particular energy that can be directed at the sample under study. Many advances in neutron instrumentation have been made over the years to improve both the number of neutrons available to the experiment and their utilization.⁴ For example, while early investigators were able to break new ground by demonstrating the simplicity of the arrangement of magnetic atoms in antiferromagnetic crystals whose sizes were measured in inches, more recent groundbreaking work has focused

_________________________

³ Cold neutrons have speeds of less than 1 km/s.

⁴ However, the in-core and near-core applications cannot benefit from improvements in beamline instrumentation for increased flux; for those applications, neutron flux depends on the reactor design and sample placement.

on phenomena occurring in much smaller samples, such as magnetism in layers only a few nanometers thick (Grutter et al., 2015). Since the signal obtained using neutron scattering scales as the amount of material exposed to a neutron beam, experiments on thin layers are much more challenging. Because the neutron-scattering signal is limited by the neutron flux that can be directed at a sample, its most advanced implementation is restricted to high-intensity neutron sources, either HPRRs or advanced pulsed spallation neutron sources.

High-power spallation sources utilize protons from a particle accelerator to bombard a target made of a heavy element such as tungsten or mercury. One advantage of spallation sources is that enriched uranium fuel is not needed to create neutrons. On the other hand, it is currently very difficult to use the spallation process to produce the neutron fluxes that can be generated by the fission of uranium-235 (²³⁵U). For many neutron-scattering investigations of materials, this limitation can be addressed by compressing the neutrons into sharp, intense pulses (“pulsing” the accelerator).

There has been a long debate within the neutron-scattering community about the relative merits of research reactors and pulsed spallation sources. The emerging consensus is that, although a rough comparison can be given, the true relative merit depends on the application being considered. For many neutron-scattering experiments, a 1-MW pulsed source such as the SNS has a performance comparable with that of a high-flux research reactor.

Present-day spallation sources do not provide the same time-averaged neutron flux as do high-flux research reactors (also called HPRRs). The neutron-scattering research community has supported the need for both types of neutron sources to provide the full range of capabilities required for modern materials research using neutron scattering. However, this may change over the coming decade because the European Spallation Source⁵ (ESS), currently under construction in Sweden, is designed to produce the same average neutron flux as current HPRRs. For now, however, there is no alternative to research reactors for some extracted beam applications.

Finding 3: Research reactors fulfill important missions ranging from education to basic scientific research and medical isotope supply. Other mechanisms for producing neutrons to fulfill all of these missions at similar energies and spectra and average fluxes do not currently exist.

_________________________

⁵ See https://europeanspallationsource.se/.

OVERVIEW OF HIGH PERFORMANCE RESEARCH REACTORS

Over many decades, several research reactors were designed and constructed to achieve very high neutron fluxes to address the purposes outlined above. These reactors tend to have very compact cores to maximize peak power densities and peak neutron fluxes. In addition to the summary in Table 3.1, a brief description of each reactor and its mission is given below.

TABLE 3.1 High Performance Research Reactors of Relevance to the Conversion Study

SOURCES: For values of thermal power, the following sources are identified for each reactor:

MITR-II: http://web.mit.edu/nrl/www/reactor/reactor.htm

SM-3: NRC (2012)

BR2: Joppen et al., (2011). NOTE: 100 MW is the maximum thermal power, but BR2 is normally operated up to 70 MW.

HFIR: Primm et al. (2009)

MURR: Foyto et al. (2012)

ATR: IAEA Research Reactor Database (RRDB), https://nucleus.iaea.org/RRDB/RR/ReactorSearch.aspx

MIR.M1: IAEA Research Reactor Database (RRDB), https://nucleus.iaea.org/RRDB/RR/ReactorSearch.aspx

RHF: http://www.igorr.com/scripts/home/publigen/content/templates/Show.asp?P=764&L=EN

NBSR: IAEA RRDB

FRM-II: Böning et al. (2004)

JHR: http://www.cad.cea.fr/rjh/general-description.html.

MIT Research Reactor

The MIT Research Reactor (MITR-II) has operated since 1958. In 1974, the MITR-I was shut down to allow conversion to MITR-II, which offered a higher neutron flux. Major upgrades included new fuel element design, reactor core tank, and core housing. In 2010, a new operating license was issued for a power upgrade from 5 to 6 MW. The primary uses of MITR-II are advanced materials, fuel, and instrumentation irradiation tests using in-core experimental facilities, although beam ports are also available for other neutron science applications. Its primary sponsor is the DOE Office of Nuclear Energy through the Nuclear Science User Facilities (NSUF) research grants, small business innovation research, and national laboratories. MITR-II is regulated by the U.S. Nuclear Regulatory Commission (USNRC).

SM-3

The reactor facility is located at the site of the Joint Stock Company “State Scientific Center—Research Institute of Atomic Reactors” (JSC “SSC RIAR,” hereafter abbreviated as “RIAR”), Dimitrovgrad, Russia. It has been in operation since 1961. It is a pressure-vessel-type reactor. The reactor facility is mainly designed for the production of heavy transuranic elements but is also used to accumulate isotopes with high specific activity and to test materials. The SM-3 uses HEU uranium oxide (UO₂, 90 percent enriched) fuel dispersed in a beryllium-copper matrix.

Belgian Reactor 2

The Belgian Reactor 2 (BR2) has been operated by the Belgian Nuclear Research Center, or SCK·CEN (equivalent to a U.S. DOE national laboratory in Belgium), since 1963. BR2’s power limit is 100 MW, but it is generally run between 60 and 70 MW. BR2’s primary mission is in-core irradiation experiments, with a particular focus on radiation damage of materials. It is used for radioisotope production and neutron transmutation doping silicon production; these activities generate commercial revenue to supplement the sponsored amounts from the Belgian Ministry of Energy (at a ratio of 60/40). It is a leading location for accelerated testing of materials for nuclear energy applications, including fuel for LEU conversion of research reactors. The primary sponsor for BR2 is the Belgian Ministry of Energy.

High Flux Isotope Reactor

The High Flux Isotope Reactor (HFIR) has been operated by the Oak Ridge National Laboratory since 1965. HFIR was originally constructed for the production of heavy transuranic isotopes requiring multiple neutron captures, for example, ²⁵²Cf. Although it is still the only reactor outside of Russia to efficiently produce such isotopes, its mission is currently dominated by neutron-scattering experiments following the installation of a cold neutron source in one of its beamlines. HFIR is sponsored and regulated by DOE.

The University of Missouri Research Reactor

The University of Missouri Research Reactor (MURR) has operated since 1966. MURR is specifically designed for in-core irradiation. Although the reactor still performs this mission, its focus has shifted to medical isotope production. In addition to revenue from commercial customers, MURR is sponsored by the DOE Office of Nuclear Energy and regulated by the USNRC.

Advanced Test Reactor

The Advanced Test Reactor (ATR) has been operated by Idaho National Laboratory since 1967. ATR is the only U.S. research reactor capable of providing large-volume, high-flux neutron irradiation in a prototype environment. ATR allows for the study of effects of intense neutron and gamma radiation on reactor materials and fuels. ATR has many uses, supporting a variety of government- and privately sponsored research. ATR was specifically designed for in-core irradiation to test the performance of materials under naval reactor conditions. In 2007 ATR became an NSUF, and approximately 50 percent of its irradiation positions have been made available in support of this wider user group. ATR is the primary U.S. location for irradiation testing of LEU conversion fuels. ATR maintains a critical facility (ATR-C) with an identical configuration.⁶ ATR is sponsored jointly by the DOE National Nuclear Security Administration (NNSA) Office of Naval Reactors and the DOE Office of Nuclear Energy. Both ATR and ATR-C are regulated by DOE.

_________________________

⁶ ATR-C’s role requires it to be identical to ATR. Although ATR-C is not technically an HPRR, it is often included in the U.S. accounting of its HPRRs. This report follows this convention.

MIR.M1

The reactor facility is located in RIAR, Dimitrovgrad, Russia. It has been in operation since 1967. It is a channel-type reactor immersed into a pool of water. The reactor facility is equipped with experimental loops having coolants of various types and is designed mainly to test materials under irradiation conditions. MIR.M1 uses HEU UO₂ (90 percent enriched) fuel. Feasibility studies for its conversion are in progress.

High Flux Reactor

The High Flux Reactor (Réactor à Haut Flux [RHF]) has been operated by the Institut Laue-Langevin (ILL) since 1971. The primary mission of RHF is to produce neutron beams to conduct neutron science supported by more than 40 instruments in a reactor hall plus two guide halls. It provides data for approximately 600 refereed scientific publications per year, making it one of the most scientifically productive neutron facilities of any kind. ILL and RHF are sponsored jointly by the governments of France, Germany, and the United Kingdom.

Neutron Beam Split-Core Reactor

The Neutron Beam Split-Core Reactor (NBSR) has been operated by the National Institute for Standards and Technology Center for Neutron Research since 1969. The primary mission of NBSR is to produce neutron beams for the purpose of scientific research, currently supported by 28 instruments on beamlines in two guide halls, producing more than 300 scientific publications per year. NBSR is sponsored by the Department of Commerce and regulated by the USNRC.

Forschungs-Neutronenquelle Heinz Maier-Leibnitz-II Reactor

The Forschungs-Neutronenquelle Heinz Maier-Leibnitz-II reactor (FRM-II) has been operated by the Technische Universität München since 2004. FRM-II is host to five irradiation facilities, a medical application facility, and more than 30 instruments, in operation or under construc-

tion, dedicated to neutron beam science. Basic scientific investigations take up about 70 percent of the available capacity, with the rest dedicated to applied science.

Jules Horowitz Reactor

The Jules Horowitz Reactor (JHR) is under construction at CEA Cadarache and is scheduled to begin operations in 2020. JHR was initially designed to operate using a very high-density LEU dispersion fuel, which is currently under development but not yet qualified and will not be commercially available for the start of JHR operation. JHR will begin by performing tests for 2 years using currently qualified LEU fuel. Afterwards, JHR will operate at nominal conditions, at a power between 70 and 100 MW, using an alternate fuel that provides neutronic equivalence (such as HEU fuel). The mission of JHR will be material and fuel testing as well as radionuclide production for medical applications.

PLANNING FOR NEXT-GENERATION RESEARCH REACTORS AROUND THE WORLD

Europe has seen regular renewal of its HPRR capability, with research reactors commissioned 48, 35, and 11 years ago, and several new reactors planned to begin operation within 5–15 years. In addition, a large spallation neutron source (ESS) is scheduled to begin operation in 2019, and the spallation source at ISIS⁷ has recently been upgraded with an additional target station.⁸ RHF at ILL has long been a leading facility for neutron science reactor applications worldwide and has been significantly upgraded at various times during its history.⁹ For materials testing, BR2 is an important international resource, including for the testing of the high-density monolithic UMo fuel being developed by the United States during periods when ATR is unavailable. Its operator acknowledges, however, that BR2 will likely reach end of life within the next 15–20 years. In light of this eventuality, SCK·CEN in Belgium is planning a new facility to replace BR2, MYRRHA (Multi-purpose hYbrid Research Reactor for High-tech Applications), which is intended to be a particle accelerator-driven nuclear reactor. MYRRHA will focus on materials testing for fissile and fusion systems and

_________________________

⁷ ISIS is a pulsed neutron and muon facility. “ISIS” is not an acronym. See http://www.isis.stfc.ac.uk/.

⁸ The situation is not so hopeful for European medium-power (less than 20 MW) reactors, with many of these (e.g., Riso, Julich, Sudsvik) shut down in recent years and others in a precarious state.

⁹ However, ILL will be approximately 65 years old when high-density LEU fuel is expected to become available.

on research into nuclear waste treatment through transmutation. The European Commission considers MYRRHA to be a “high priority international project with significant societal relevance.” The goal is for this reactor to be operational in 2025.

JHR will also add to European materials testing and irradiation capability but will not support extracted beam applications. Finally, the 45-MW High Flux Reactor at Petten, the Netherlands (previously converted to LEU fuel, and so it is not discussed above) is recognized to be reaching its end of life, and so a new reactor, PALLAS,¹⁰ is being planned for medical isotope production and nuclear technology research. A bidders’ conference was held in August 2015; design, construction, and commissioning are expected to take about 10 years. The continuing planning and construction of new HPRRs and other neutron sources in Europe positions it well to maintain its prominence in neutron science and materials testing for decades to come.

Russia, too, is planning new reactor construction, although its emphasis is on fast reactors. The multipurpose sodium-cooled fast neutron research reactor (MBIR) (which has begun construction at RIAR in Dimitrovgrad, but is not an HEU-fueled reactor) will be a 150-MW, sodium-cooled fast reactor and will have a design life of up to 50 years. It will be a multiloop research reactor capable of testing lead, lead-bismuth, and gas coolants, and running on mixed oxide (MOX) fuel. It is expected to begin operations in 2020.

South Korea is in the midst of construction of the Kijang Research Reactor (KJRR), which will be a 15-MW reactor designed for radioisotope production and semiconductor doping. Scheduled to begin operation in 2017, it will be the first reactor in the world to operate on UMo dispersion-type LEU fuel (see Chapter 4). The fuel is being developed in South Korea in parallel with reactor planning and construction.

Aging of the U.S. Fleet of High Performance Research Reactors

The average age of reactors within the U.S.-based HPRR (USHPRR) fleet is close to 50 years. Many of the USHPRRs have recently completed license renewals that last for 20 years, but they will be due for relicensing at about the same time as their conversions to LEU fuel are expected to occur. Physical limits for existing reactors include embrittlement of the containment vessel and cooling systems. A variety of refurbishment strategies and programs (DOE, 2013b) have ensured that these reactors can continue to operate safely and can meet the evolving needs of their users.

The committee gathered end-of-life analysis information from each USHPRR. The operators of these reactors are confident that they can con-

_________________________

¹⁰ See http://www.pallasreactor.com/?lang=en.

tinue safe operations beyond the next license renewal. In the absence of any plan to replace their capabilities, their roles in support of science, engineering, and medicine missions will be just as important as they are today. Two of the USHPRRs, HFIR and NBSR, have reported an expected end of life near the 2050 time frame and have begun presenting concepts for the next generation of research reactors (Beierschmitt, 2009; Wu et al., 2014). Several others (including ATR and MURR) have indicated no expected end date as long as maintenance and longevity plans are supported. Still, one might readily consider aging as an obstacle to conversion. It is natural (and expected) for operators and those involved in the conversion programs to consider the costs of conversion of a reactor that will operate for only several years before reaching its end of life versus investment in other areas such as new reactor (or other research facilities) planning and design.

No new HPRRs have been fully designed, built, or commissioned in the United States for more than 46 years. USHPRRs will be between 58 and 69 years old by the time that LEU fuel is projected to be available for their conversion. This is an unprecedented age for research reactors, and significantly longer than the projected operating lifetimes for European research reactors. For neutron science applications, ongoing improvements to existing reactor facilities (e.g., the new guide hall at NBSR) have enabled the missions of these facilities to keep pace with the evolving science needs. Changes and augmentation in the science and engineering missions of some HPRRs have been successful, even if they were originally optimized for another purpose. That said, a purpose-built reactor may offer additional benefits and capability over one that has been “repurposed.”

Within the United States, a number of reactor facilities have been closed and a few other facilities have been built to support the neutron beam science mission, most notably the SNS. There is also an ongoing effort to construct new facilities for medical isotope production, especially ⁹⁹Mo. In contrast, no new capability has been developed for other transmutation applications or materials testing.

Finding 4: The mission and capability of some high performance research reactors have evolved to accommodate changing user needs and to expand the user base, with the consequence that the reactors are sometimes not specifically designed for current missions.

Current analyses indicate that the variety of missions spanned by the USHPRRs can be accomplished with the reactors operating with a new high-density monolithic LEU fuel. Similarly, LEU conversion plans for European HPRRs, as well as plans for construction of new European HPRRs using only LEU (or alternatives not requiring HEU), indicate that the HPRR mission space can be preserved using the high-density UMo

dispersion fuel currently undergoing development. This evidence supports the contention that such missions could be accomplished with new research reactors designed specifically to use such fuel, especially if such designs were optimized for the current and expected future missions. Although a rigorous design study has not been completed, it is reasonable to expect that there would be fewer technical constraints in designing a new reactor using a new LEU fuel than there would be in developing a new LEU fuel for existing HPRRs. As noted previously, JHR, an HPRR, was designed for use with high-density LEU fuel, but it will likely begin full operations using HEU fuel because the high-density LEU fuel is not yet qualified. Although it has not been definitively established, it may also be possible to design a new research reactor to satisfy current missions with currently qualified lower-density LEU fuel.

USHPRRs serve different communities and are operated under the auspices of different government offices. Although this arrangement has ensured that each of the important and existing customer bases for research reactors has at least one HPRR that serves its needs, it also means that communication and coordination among the full research reactor community—operators, users, and sponsors—in the United States is difficult and limited. As an example, a DOE Nuclear Energy–National Nuclear Security Administration Research Reactor Working Group¹¹ in 2013 considered future options for research reactors in the United States, but explicitly recognized that it could only consider reactors and applications within DOE’s scope of responsibility (DOE, 2013b). Also discussed previously, operators from individual reactors have begun to propose next-generation designs of research reactors with similar missions (HFIR, managed by DOE, and NBSR, managed by the Department of Commerce), seemingly without coordination across the different agencies that manage these reactors.

There has been no attempt of which the committee is aware to consider and prioritize needs for research reactors in general or HPRRs in particular across all U.S. stakeholder communities. The Office of Science and Technology Policy (OSTP) was established in 1976 via legislation¹² that authorizes the office to lead interagency efforts to develop and implement sound science and technology policies. OSTP therefore both supports the advancement of basic and applied science and has the ability to gather cross-agency insights, much more so than the individual agencies by themselves. Until such broad engagement and community priority setting takes place, it is

_________________________

¹¹ While the usual acronym for the National Nuclear Security Administration is “NNSA,” the group formed by the DOE Nuclear Energy and National Nuclear Security Administration is referred to as the “NE-NA Working Group” instead of the “NE-NNSA Working Group.”

¹² See the National Science and Technology Policy, Organization, and Priorities Act of 1976 (P.L. 94-282).

likely that the USHPRRs will operate in “run-to-failure” mode without a smooth transition to the next generation of research reactors or research facilities capable of similar missions. The result would likely be that the United States will lose capability and international presence in important areas of scientific and technological research. It is worth mentioning that other technical communities have demonstrated the possibility of reaching a community consensus on priorities for major facilities, as illustrated by the recent Particle Physics Project Prioritization Panel (P5) report of the particle physics and particle astrophysics communities.¹³

Finding 5: There is no overarching, cross-agency, long-term strategy for meeting enduring U.S. need for research reactors. The nearly 20-year time line to conversion that is currently estimated for some of the U.S. fleet of high performance research reactors is much longer than was originally estimated and coincides with many of the reactors’ time lines for relicensing. At that time, these reactors will be on average 65 years old. Because of the convergence of relicensing, conversion, and aging issues of the current U.S. high performance research reactors (USHPRRs) in 2030, it is reasonable to compare the benefit of converting/retrofitting the current fleet of USHPRRs against designing and building new research reactors that use low enriched uranium fuel and address the critical missions the current reactors support.

Recommendation 1: The U.S. Office of Science and Technology Policy should take the lead in developing a 50-year interagency strategy that enumerates and evaluates the importance of anticipated U.S. civilian needs for neutrons and provides a roadmap for how these can best be provided by reactors and other sources that do not use highly enriched uranium.

_________________________

¹³ Available at http://www.usparticlephysics.org/p5/.

This page intentionally left blank.