T

he U. S. National Academies and Russian Academy of Sciences will organize a joint symposium to discuss progress, challenges, and op­portunities for conversion of research reactors from highly enriched uranium to low enriched uranium fuel. The symposium will address the following topics:

• Recent progress on conversion of research reactors, with a focus on U. S.- and Russian-origin reactors.

• Lessons learned for overcoming conversion challenges, increasing the effectiveness of research reactor use, and enabling new reactor missions. • Future research reactor conversion plans, challenges, and opportunities. • Actions that could be taken by U. S. and Russian organizations to promote conversion.

[1] This report was authored by committees of the National Academy of Sciences (NAS) and RAS. These committees are responsible for the report’s quality and accuracy.

[2] No effort was made in this report to attribute statements of fact to individual participants.

[3] In this report, the term “research reactors” is defined to include research, test, and training reactors, including critical and subcritical assemblies.

[4] By international agreement, HEU is defined as uranium enriched to a concentration of 20 percent uranium-235 or greater, whereas LEU is defined to be uranium enriched to a concentration of less than 20 percent uranium-235.

[5] This symposium focused on HEU-fueled reactors; however, some research reactors are also fueled with plutonium. The challenges of managing plutonium-fueled reactors—which will need to be accomplished through materials protection, control, and accounting measures—are mentioned in this report but were not the focus of this symposium.

[6] The terms “origin,” “supplied,” and “designed” are used interchangeably in this report to describe reactors that were developed by the United States and Russia for both domestic and third-country use.

[7] Important statements of opinion are attributed to individual workshop participants where appropriate, but no attempt has been made to attribute statements of fact.

[8] “The term ‘special nuclear material’ means plutonium, uranium enriched in the isotope

233 or in the isotope 235, and any other material that the [Nuclear Regulatory] Commission… determines to be special nuclear material.” (42 U. S.C. § 2014)

[9] Although LEU could, in principle, be enriched and converted into HEU for use in building a nuclear weapon, this process would require a significant technical infrastructure, and the mass of LEU required would be very large. The international community could track an effort to enrich LEU more effectively than one involving the theft of HEU.

[10] Assuming 50 kilograms of HEU per explosive device. This may be a conservative as­sumption. The IAEA defines the siqnificant quantitity of HEU to be 25 kilograms. Significant quantity is defined as “the approximate amount of nuclear material for which the possibility of manufacturing a nuclear explosive device cannot be excluded” (IAEA, 2001).

[11] Much research reactor used fuel is not considered to be “self-protecting” (formally defined as producing a dose rate greater than 100 rad per hour at 1 meter in air) because of its low radioactivity. However, irradiated fuel from virtually all of the high-performance reactors mentioned in this report would be considered to be self-protecting, as would irradiated fuel from commercial power reactors, for a period following removal from the reactor.

[12] www-naweb. iaea. org/napc/physics/ACTIVITIES/Research_Reactors_Worldwide. htm.

[13] A moderator is a material used to slow down neutrons (i. e., reduce their kinetic energies), which increases the probability of fission when the neutrons are captured by uranium nuclei.

Light materials such as water and graphite are commonly used as moderators.

[14] Control rods contain materials (e. g., boron) that absorb neutrons; they are used to control fission rates in the reactor fuel and hence the power levels in the reactor.

[15] Tank-type research reactors are similar to pool-type reactors in overall design, but they typically operate at higher power densities, requiring higher coolant flows and pressures, making it necessary to separate the coolant from the remainder of the pool contents.

[16] Power upgrades of U. S.- and Soviet-supplied research reactors were undertaken to in­crease neutron fluxes in experimental positions.

[17] More information about this program can be found at www. rertr. anl. gov.

[18] Research, test, and training reactors that have military or national security missions are outside the scope of DOE’s conversion program.

[19] DOE and GTRI assist reactor operators to perform feasibility studies and safety analyses required for regulatory approval to convert and procure LEU replacement fuels. GTRI also funds work to develop and qualify higher-density uranium-molybdenum (UMo) LEU fuel to convert high-performance research reactors (see Chapter 2).

[20] In 1996 the Bochvar All-Russian Research Institute of Inorganic Materials (VNIINM) became the lead Russian institute under the contract with ANL.

[21] These reactors are regulated by the U. S. Nuclear Regulatory Commission or the U. S. Department of Energy.

[22] The NTR General Electric Reactor in California and the Idaho National Laboratory’s TREAT reactor (Roglans, 2011b).

[23]A critical assembly contains sufficient fissionable and moderator material to sustain a fission chain reaction at a low (close to zero) level. It is designed so that fissionable and moderator materials can be easily rearranged in various geometries to mock up different reactor designs.

[24] Not including naval or other defense-related reactors.

[25] Chapter 1 (this chapter) provides the context for this study and introductory material from the symposium;

• Chapter 2 addresses challenges associated with conversion as well as potential solutions;

[26] As noted in Chapter 1, there are two additional HEU-fueled research reactors in the United States (NTR General Electric and TREAT; see Footnote 20 in Chapter 1) that appear to be convertible using current-type LEU fuels. The Department of Energy (DOE) is com­pleting studies to confirm the feasibility of converting these reactors using current-type LEU fuels. Additional research will be required to more fully develop the capability to fabricate these LEU fuels.

[27] The Reduced Enrichment for Research and Test Reactors (RERTR) program (see Chapter 1) also participated in the qualification of a fourth LEU fuel system: a uranium-zirconium hydride with an erbium burnable poison (UZrHx-Er) fuel system that is used for the conversion of TRIGA (Test, Research, Isotope production—General Atomics) reactors. General Atomics began developing a higher-density fuel (up to 3.7 gU/cm3) before the RERTR program was started in 1978. The RERTR program performed irradiation tests on 20/20 (i. e., 20 weight percent uranium, 20 percent enriched), 30/20, and 45/20 fuels. The 30/20 fuel was used to convert the Oregon State TRIGA Mark II reactor, discussed later in this chapter, and the University of Wisconsin Nuclear Reactor, discussed in Chapter 3, as well as a number of other TRIGA reactors in the United States and abroad.

[28] These high-performance reactors have high-power-density (i. e., high-flux-density) cores. Fuels having higher uranium densities than are available with existing LEU fuels are required to convert these reactors.

[29] That is, alloys consisting of 9 parts uranium to 1 part molybdenum by weight.

[30] Fission gas bubbles are formed in the fuel phase as a result of the production of gaseous fission products.

[31] At present, no LEU replacement fuel has been identified for the FRM II reactor.

[32] As the name suggests, a partial fuel assembly contains only portions of a full fuel assembly. For example, a partial assembly might contain fewer fuel plates than a full assembly.

[33] Extrusion processes are used to manufacture research reactor fuel in Russia, whereas rolling processes are used to produce research reactor fuels in the United States and Europe. Both processes produce suitable fuels, but fuel produced by extrusion generally has a lower density than fuel produced by rolling.

[34] This conclusion was reached by studying the use of existing IRT-4M LEU fuel. A feasibility study with the IRT-3M UMo fuel of higher density, currently under development, is under way at MEPhI (see Chapter 3) and may reach different conclusions when completed.

[35] In Russia, reprocessing costs are based on fuel mass.

[36] Develop or identify an LEU fuel assembly that is acceptable for conversion.

• Ensure that the ability of the reactor to perform its scientific mis­sion is not significantly diminished.

• Ensure that conversion can be achieved without requiring major changes in reactor structures or equipment.

• Demonstrate that the LEU fuel meets all safety requirements and that conversion and subsequent operations can be accomplished safely.

• Ensure that annual operating costs do not increase significantly as the result of conversion.

• Develop a conversion schedule that is based on operational require­ments, capabilities, and regulatory processes.

[37] Service lifetimes of LEU fuel assemblies can be increased if the uranium-235 loadings are higher than comparable loadings in the HEU fuel.

[38] This reactor is also discussed in another symposium presentation that is summarized in

Chapter 3.

[40] And under a Contract Service Agreement with the IAEA.

[41] The 77 reactors represented in the template responses range from less than 5 years to more than 50 years old. The average age was 37.8 years (Figure 2-7).

• The most frequently reported ageing problems were obsolescence and technology changes (92 out of 367 reports); corrosion (73 out of 367); and changes to regulatory requirements and standards (49 out of 367).

[42] Some of the Russian reactors described in this presentation are also discussed in another presentation that is summarized in Chapter 3.

[43] The USNRC does not regulate the High Flux Isotope Reactor at the Oak Ridge National Laboratory or the Advanced Test Reactor and its critical assembly at the Idaho National Laboratory. These reactors are the responsibility of DOE.

[44] The Cintichem Reactor was located in Tuxedo, New York. It was shut down in 1990.

[45] Depending on its design, HEU fuel is shipped to either the Savannah River Site in South Carolina (for aluminum-based fuels) or the Idaho National Laboratory (for other fuel designs), where it is stored.

[46] This number includes radiation sources at hospitals.

[47] This deadline slipped to 2022 while this report was being completed because of Fiscal Year 2011 federal budget reductions.

[48] In Chapter 1, it was noted that 35 conversions or shutdowns of HEU-fueled reactors have occurred since 2004. This larger number includes 10 reactors that were shut down and 2 reactors that were converted to LEU under domestic programs rather than GTRI.

[49] Option 1: Assess the possibility of changing the facility mission such that it can be accomplished with LEU fuel. However, GTRI does not advocate a change of reactor mission for the sole purpose of converting.

• Option 2: Reduce HEU enrichments. This may be technically fea­sible in some cases where LEU conversion is not. Note, however, that re­duced enrichments above 20 percent are not considered HEU minimization under international norms or GTRI policy.

• Option 3: Shut down the facility or consolidate it with similar facilities if it is underutilized.

• Option 4: If no other options exist for the facility other than to operate with HEU, remove all excess HEU and enhance physical protection measures to achieve threat reduction.

[50] As noted in Chapter 1, a critical assembly contains sufficient fissionable and moderator material to sustain a fission chain reaction at a low (close to zero) level. It is designed so that fissionable and moderator materials can be easily rearranged in various geometries to mock up different reactor designs.

[51] As noted in Chapter 1, a material is considered to be “self-protecting” if it produces a dose rate greater than 100 rad per hour at 1 meter in air. These high levels create substantial radiological barriers to illicit use.

[52] Research reactors will continue to be an essential tool for many applications. B. Myasoedov commented that he expected the role of re­search reactors to grow in the future to support the development of more complex reactor designs for nuclear power plants, including those based on fast reactor designs; for radiopharmaceutical production; and for analytical methods (such a neutron activation analysis) to support safety monitoring and control. He suggested that Russia and the United States should agree to work together and with third-party countries to design a standardized research reactor that could be produced on an industrial basis. This would eliminate the need to design individual, customized cores and fuel elements.

• Past experience suggests that successful conversion solutions can be found for most reactors. Jim Snelgrove commented that in view of the suc­cess that has occurred in converting reactors in the United States and some other countries, a key take-away message from this symposium should be that it is possible to find conversion solutions if one works hard enough to uncover them. Yu. S. Cherepnin added that some of the presentations in this session documented how enrichment levels could be reduced without degrading reactor performance. These examples should be publicized. H.-J. Roegler commented that conversion can result in improvements to reactors.

[53] “Neutronics” refers to an analysis of the neutron flux throughout the core, which entails analysis of fission and neutron capture events caused by absorption of neutrons by the reactor core, scattering of the neutrons, and losses of the neutrons from the reactor.

[54] Delayed fission neutrons are neutrons emitted spontaneously from decay of a fission product from a prior fission event, whereas prompt neutrons are neutrons emitted from the fission process directly. The delayed neutron fraction is the ratio of the mean number of de­layed fission neutrons per fission to the mean total number of neutrons per fission (prompt plus delayed).

[55] The prompt neutron lifetime is the average time between the emission of neutrons and either their absorption in or their escape from the system.

[56] The control element worth is the negative reactivity change caused by inserting a control element into the reactor. UWNR has five separate control elements.

[57] The prompt temperature coefficient is the change in reactivity per degree change in fuel temperature.

[58] This confirmatory analysis was a two-dimensional deterministic analysis coupled with a one-dimensional diffusion approximation.

[59] When preparing for conversion, the University of Wisconsin was provided with two additional LEU fuel assemblies so that the reactivity could be boosted if required.

[60] This analysis assumed no cross-flow, i. e., no exchange of coolant with adjacent fuel or reflector assemblies.

[61] DNBR is the ratio of the heat flux needed to cause departure from nucleate boiling to the actual local heat flux of a fuel rod. Departure from nucleate boiling is the point at which the heat transfer from a fuel rod rapidly decreases because of the insulating effect of a steam blanket that forms on the rod surface when the temperature continues to increase (USNRC, 2011).

[62] This is the relationship between the conditions in a heated channel and the heat flux at which the heat transfer becomes impaired as a result of the transition from nucleate boiling to film boiling. These conditions may include the mass flow rate, channel geometry, and thermal properties of the fluid (e. g., the density of liquid and vapor, heat of vaporization, specific heat). The critical heat flux correlations used in this analysis were the Groeneveld 2006 look up tables and the Bernath correlation. More information on these correlations can be found in Vitiello (2008).

[63] The fuel temperature-limiting safety setting is the temperature below which the fuel is required to be maintained to prevent fuel element failure.

[64] A “near maximum hypothetical accident” maintains an intact pool and ventilation.

[65] This is the relationship between the pressure drop and the conditions in the flow channel such as mass flow rate.

[66] This is the relationship between the conditions in the flow channel at the time when there is net vapor generation. It is a relationship between parameters such as mass flow rate, heat flux, and thermal properties of the liquid (e. g., thermal conductivity, specific heat, saturation temperature).

[67] The Bergles-Rohsenow correlation is commonly used for prediction of ONB in narrow

rectangular coolant channels and relates cladding local heat flow at ONB to local heat flux

and pressure.

[70] At specified setpoints, the reactor will shut down by opening the circuit breakers that supply electrical power to control rods.

[71] COMSOL is a multiphysics engineering software package.

[72] Descriptions of some of these reactors were provided in the presentations that are summarized Chapter 2.

[73] In addition, Dr. Starkov stated during the symposium that he believed there have been some problems in reprocessing silicide fuels. As was noted previously, Russia reprocesses its research reactor fuel unlike in the United States.

[74] The other reactors are IR-8, discussed in the next section, and OR (OP-M in Table 1-2), a 0.3 MW pool-type reactor. See Chapter 2.

[75] Tomsk Polytechnic Institute has recently requested that this reactor be licensed to operate at 11 MW.

[76] Aluminum is now being used rather than beryllium in a number of locations because of swelling of some of the beryllium blocks.

[77] Nuclear physics of the interaction of radiation with matter.

[78] Radiation damage of metallic and nonmetallic reactor materials.

[79] For example, one facility might require only very few radiation protection measures to isolate nuclear materials, whereas another facility might require more sophisticated measures. These operational characteristics affect the proliferation risk of the facility.