Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
1 Introduction The Molten Salt Reactor Experiment (MSRE) at Oak Ridge National Laboratory (ORNL) was a demonstration of a novel nuclear reactor design. The reactor generated a power of ~ MW from nuclear fission reactions in operations during 1965-1969. Graphite in the MSRE reactor vessel acted as a moderator to slow down (or "moderate") the high-energy neutrons released from fission events, to reduce the neutron kinetic energies to low (or "thermal") values where the fission cross section is much larger than it is at higher neutron kinetic energies. As in any nuclear reactor, the combination of fissile filet and moderator in the core generated a spatially distributed flow of neutrons, or a neutron flux, which is the number density of neutrons passing through a unit area in a unit time as a function of position in the reactor, time, and neutron energy. The following features made the MSRE design atypical. The MSRE fuel was in a homogeneous molten fluoride salt medium, rather than in solid nuclear idle! rod assemblies that are used in all commercial nuclear power plants today. The filer was fissile uranium (first a 23su charge, later 233U) and fissile plutonium (239Pu) contained as the uranium tetrafluoride (UF4) and plutonium tetrafluoride (PuF4) salts in a molten salt medium consisting predominantly of lithium, beryllium, and zirconium fluorides (LiF, BeF2, and ZrF4), at an operating temperature of approximately 650°C. Another unusual feature was the lack of a separate coolant in the core design. The reactor vessel contained only the molten salt filet and the graphite moderator. Heat from the fission reactions elevated the iTo sustain a critical chain reaction in an assembly of fissile material, each nuclear fission event produces, for every thermalized neutron captured, approximately two energetic neutrons (which, with moderation and loss mechanisms, result in approximately one thermalized neutron to participate in a subsequent fission event). Energy is also liberated, on the order of 200 million electron volts (MeV) per fission. The fissile nuclei used in the MSRE were uranium-233 and -235 (233U, 35U) and plutonium-239 (239Pu). 10
INTRODUCTION 11 temperature of the fuel salt and was removed by circulating the fuel salt through a heat exchanger external to the reactor core. This design is to be contrasted with that of commercial reactors, which circulate the coolant (water or gas) in the core of solid fuel rods. Figures I.l and I.2 show the layout of the MSRE. The fuel salts were circulated in a loop containing the reactor vessel, a heat exchanger, and a pump. The coolant salts circulated in a loop consisting of the heat exchanger, a separate pump, and a radiator, from which the heat was dissipated to the air (up a stack) by fans. The reactor is housed in an isolated concrete building of ORNE inside the Oak Ridge Reservation. The drain tanks and reactor vessel are in a sealed containment system (Peretz, 1996c, p. I-6) surrounded by concrete walls several feet thick to provide a primary radiation shield between the radioactive materials and the environment (see Figure 1.31. The building provides a secondary containment. The nearest location for exposure to the public is the Bethel Valley Road outside ORAL (Peretz, 1996c, p. 1-36~. The MSRE demonstrated the potential for a thermal neutron breeder design. Introduction of thorium-232 (232Th) in the salt (as a fluoride compound ThF4) would permit breeding (i.e., generation by nuclear-induced transmutation) of 2 3U. Absorption of a low-energy (near thermal) neutron by a 232Th nucleus transmutes it to 233Th, which undergoes beta decay to protactinium-233 (233Pa), followed by a second beta decay to produce 23 U. Breeding is achieved in a reactor when the rate of production of fuel (233U) is greater than the rate of consumption of fuel by nuclear fission inside the core. The ratio of the rate of production and the rate of consumption defines the breeding ratio. Although actual experiments with a thorium blanket were not performed, experiments on the MSRE demonstrated the potential for a next-generation design to breed 233U with a breeding ratio greater than one. An on-line chemical processing system could have been installed (in the fuel processing cell shown in Figure 1.2) to extract 233Pa from the salt during operations to enhance the production of 233U. If the 233Pa were left in the core, its absorption of thermal neutrons would reduce both the neutron flux in the core and the breeding of 233U (Nero, 1979~. The processing system that was installed was used successfully to recover uranium by fluorination (Peretz, 1996c).
12 AN EVALUATION OF DOE ALTERNATIVES FOR MSRE ~ REMOTE MAINTENANCE REACTOR CONTROL ROOM \ \ r ~ ' 1 ~ ~ ~ ~ ~ _ CONTROL ROOM I l 1. REACTOR VESSEL 2. HEAT EXCHANGER 3. FUEL PUMP 4. FREEZE FLANGE 5. THERMAL SHIELD 6. COOLANT PUMP 7. RADIATOR 8. COOLANT DRAIN TANK 9. FANS 10. FUEL DRAIN TANKS 11. FLUSH TANK 12. CONTAINMENT VESSEL 13. FREEZE VALVE , -:... FIGURE 1.1 Arrangement of the principal components of the MSRE. SOURCE: Modified from Peretz (1996c, Figure 1.1~.
13 o X ~ Ci ~ cat ~ ~ m 4:: 0 ~ ~ i> a: ~ m U,Y 0 ~ :5 m ~ ~ y z ° ~ · ~ ~ 0 0 hi ~ ok\ ° O° . // - // \ \ 1 ~ I z z z ~ ~ \ C) it: ~: IT)= A\ \ \ \ r I ! ; \\ Y\ \ n n___~_________________ ZYJ ~ Z . . ~ _ ~ ~1 1~ ~ _, _. . . . me.' 1 Ha . ~ .' Zen wmJ~, ~ ~ or Zip ~~J Z._ ~ < _ _ at' C) 1 1"''-~ _ ,, C l Z" 0 1_ of Zig it: to - <t - 1 1 0 '_ J O me I J ~ ~ O Z Z. J Z Y z Y O ' ~ ~ ~ ~r _ O O ~ ~ 1~ I 40~` m~0-o J Cl ~ =<U} V) ~:J O Z Z ~r ~ ~J J Ct - . _ X 3 ._ C~ C~ o 3 C~ ~_ ·3 o - . _ Ct ~ . ~' a~ ._ C) Ct Ct P~ _' .. _ C: o CL c~
14 L" o o 1 - C~ w _ ~ . ~ *. . . A - or . ~ . , . . ~ lo is - a, ' ~ ~ Ct ~ Ce O so ~ 5 Ct 'are 1 1 At:< 1~ _w O ~ ~ ~ 0 >Z I tY ~ L.] O LL1 ^ ~ _' in; ~ O =) · ~ a' O ~ _ ~ C) ~ C~ Ct U7 U) O ~ a' ~ ;~ CO _ ~1 LLI ~ o CC V: . _ o ._ C) - U: _t - O ~ ._ . ~ _' ~ ._ U) ~ C) C) ~ ~ ¢N · G~ Ct S: ._ _~ - . . o U) w ~ j: W W ~ '~i. ~ ,,, ~ q7 , .,. ..... ,, . '';~..V~~; ~ ]~ ,,E, ,~'~\ ~ 1 .: ' .. ',
INTRODUCTION 15 A more comprehensive discussion of the reactor can be found in many sources (see, for example, Benedict et al., 1981, p. 10; Knief, 1992, pp.318-321; Weinberg, 1 994, pp. ~ 25-127; Peretz, ~ 996c, and references therein). CURRENT STATUS OF THE MSRE After shutdown in 1969, the molten fluoride fuel salt and a batch of flush salt (used to "flush" the system) were drained into three drain tanks, two fuel salt tanks and one flush salt tank, and were permitted to solidify. The tanks are located in a hermetically sealed (welded shut) enclosure containing both the drain tank cell and the reactor cell (see Figures I.~-~.4~. The major chemical constituents of the salts are lithium, beryllium, and zirconium fluorides; the radioactive species are actini`des, their decay products, and fission products. Radiation emitted from the decay of radioactive species within the salt has interacted with the salt medium to cause radiolytic formation of both F2 (fluorine) and UFO (uranium hexafluoride) gas. These gases have diffused out of the solid salt. This diffusion may have been aided by periodic annealing operations at elevated temperature that were performed in the 1970s and 1980s. Over time, headspace gases in the tanks have migrated into the reactor vent piping and the off-gas vent trap system to the interior of the activated charcoal bed (ACB; see Figure ~ .5) through valve openings. In recent years, the migration of UFO gas through the system apparently has resulted in the deposition of uranium fluoride solids in the ACE, and such deposits may be present on the interior walls of piping and in the drain tanks. In October 1996, the vent piping was at an overpressure of approximately one atmosphere, due to F2 and UFO gases, with solid plugs of uranium fluoride deposits inferred from pressure differences in piping runs leading from the drain tanks to the ACE at the end of the vent line. The facility's two immediate hazards gas buildup and solid uranium fluoride deposits are being addressed by work in progress at ORNL. A "reactive gas removal" action has been designed to unblock or bypass the plugged piping system and collect the chemically reactive F2 and UFO gases in traps. Pumping initiated in November 1996 has relieved the overpressure in some piping, although nonvolatile plugs have prevented ~ ~ ~ V ~ A ~
16 AN EVALUATION OF DOE ALTERNATIVES FOR MSRE STEAM OUTLET- GONDENSATE RETURN - WATER DOWNCOMER INLETS ok,! . _, N`! I BAYONET SUPPORT PI ATF _ STRIP WOUND FLEXIBLE HOSE WATER DOWNGOMER ~ V;N GAS PRESSURIZATION ,~.. AND VENT LINES ~,~' //,,:~ FUEL SALT SYSTEM FILL AND DRAIN LINE BAYONET HEAT EXCHANGER THIMBLES (32) it_ - . . act, , 1~ ' :~1 1 ~ 1 1 , I I ,, ,, 1 1 FUEL SALT SYSTEM ~ FILL AND DRAIN LINE TANK FILL LINE ~ INSPECTION, SAMPLER, AND / LEVEL PROBE AGGESS STEAM DROME I CORRUGATED FLEXIBLE HOSE STEAM FIEFS BAYONET SUPPORT PLATE HANGER GABLE _~` INSTRUMENT THIMBLE ~ FUEL SALT DRAIN TANK ,. ~ TANK FILL LINE 11 ~ THIMBLE POSITIONING RINGS FIGURE 1.4 Sketch of an MSRE fuel drain tank, made of a nickel alloy (Hastelloy N), showing bayonet thimbles attached to a headspace drome. The flush drain tank lacks the bayonet thimbles. SOURCE: Peretz (1996c, Figure 1.3~.
17 o lo lo / I ~ _ CUP 1 1 1 1 l 1 ~ _ Of'' ;1 1 \1 / I F / 1 1 /'/ 1 1 ~ I ~ \ _ _~ ~ ~ . _ 1 1 1 1 1 1 1! 1 1 1 1 1 1 1 .. 5 ~ ~ u, \ ~ ~ ~ / ~ U!) Hld3a l 1 : ~ ,,,2 ICE - ~ , _ l a: =) _ < DS to ~ ,,.,, _ Z ~ o ,0 _ _ ~,,,! ~ [,,~ ~ _ ~ Ln Z (n m _ ! ; 1 ~ 1 ~\ o ._ 3 3-> o 5, - ~ ~ _ ~ O O ~ _ ~ ~ ~ U1 o ~ V ° Ct . ~ ;> 5 o ,4 a' . ~ . a~ ° 5 ~o ~ 1~ co £ ~ C) a~ ~ ^ Ce o ._ ~ ~ ~ ·_ ._ ~ Ct X 11 s~ Ct ~ ~ .o ° .~` .= · ~ o _ .^ 3 £ °= ~ =0 ,D 11 c¢ _ <= o au =^ ~ ~_ . o C) ~ ~ o ~ o (, 1 1 o o _) 1 C) .. U: s:~ o <u ~ Z ~ . C~ ~ a' 3 ~ ~ V: l \ lOOlONOdHO 3~31 831VM .
18 AN EVALUATION OF DOE ALTERNATIVES FOR MSRE complete access to all system piping, including that closest to the drain tanks. A robotic vacuuming operation is being designed to extract the uranium deposits and charcoal from the top of the ACB trap. The uranium fluoride salts in the three drain tanks are a more long-term concern. These fluoride salts are not in a favorable configuration for long-term storage. The salts are unstable due to radiolysis that continues to occur, liberating reactive gases. The fissile materials in the tanks could form a critical configuration if a moderator such as water entered the system. The probability of this happening in the long term must be considered if the salts are to be stored indefinitely in the present location (see Figures 1. I- ~ .3~. In particular, the drain tank cell is capable of being flooded if the hermetic seal is lost because the natural water table is above the tanks (Peretz, 1996c). Waterflooding could occur in the absence of active ground water pumping operations, which are currently practiced. Because a nuclear criticality excursion cannot be ruled out absolutely under these conditions, the salts in the present mode of tank storage pose a long-term hazard, even though it is small. The Department of Energy (DOE), following the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) regulatory process, is exploring technically feasible operations designed to reduce the hazard posed by the salts in these tanks. A variety of technical alternatives have been identified to date (Peretz, 1996c) for salt removal, separation of uranium, and ultimate disposition of all waste products. Seven primary alternatives, some with subalternatives, were identified by Peretz (1996c). They are summarized as follows: (1) do nothing, (2) enhance the present storage, (3) treat the salts as spent nuclear fuel awaiting eventual storage in a federal repository, (4) treat the salts as transuranic waste bound for the Waste Isolation Pilot Plant, (5) electrorefine the salts, (6) reuse the salts elsewhere in the DOE complex, and (7) use interim storage at a DOE site. Alternatives 4 through 7 include a possible fluorination step to remove uranium. Chapter 7 lists these alternatives in more detail. These technical alternatives are also considered in the larger context of a waste management strategy with a view to ultimate disposal options, such as geologic storage in a repository. The strategic conclusion of Peretz (1996c) is that on-site interim storage is the preferred strategy for the foreseeable future, provided the materials in question cannot be
INTRODUCTION 19 reused beneficially elsewhere in the DOE complex. The particular technical alternative recommended by Peretz (1996c) is a melting and fluorination process to strip uranium from the salt as UFO gas. The uranium fluoride would be converted to an oxide and stored with other 233U inventories, while the stripped salt would be stored in a "gettered" mode in shielded cells. This alternative, and others, would involve remediation work using the MSRE facility and equipment. Project personnel have begun testing and maintenance, such as replacement of deficient pressure sensors. However, the operability and status of some components are unknown and contributes to the panel's recommendation for a stepwise remediation approach. Present DOE practices require an assessment of the integrity of all components of the MSRE that would be vital to a remediation operation. These considerations are at a stage where relevant scientific and technical information now under development on the identified alternatives will provide input to a future binding regulatory decision on the particular remediation option selected for the MSRE salts. This decision involves selection of the remediation alternative considered to be the best, as judged against all CERCLA criteria.2 ROLE OF THE NATIONAL RESEARCH COUNCIL At the request of DOE, the National Research Council (NRC), under the auspices of the Board on Radioactive Waste Management (BRWM), has undertaken a study to provide an independent technical review of the alternatives offered in Peretz (1996c~specifically, to examine the process by which these alternatives for MSRE salt treatment and disposition were identified and used in decision making. The complete Statement of Task describing the scope of this study is reproduced in Box 1.1. The study was undertaken by the Molten Salt Pane} (hereafter referred to as the panel) of the Committee on Remediation of Buried and Tank Wastes (CRBTW). The present report, written by the panel, represents consensus positions reached as a result of 2The CERCLA criteria are the following: overall protection of health and environment, compliance with applicable regulations, long-term effectiveness, reduction of hazard, short-term effectiveness, implementability, and cost.
20 AN EVALUATION OF DOE ALTERNATIVES FOR MSRE deliberations and discussions. Two meetings (in September and October 1996),3 parts of which were open to the public, were held to solicit input from and interaction with DOE and its contractors. The pane} also heard from current and retired Oak Ridge employees, a State of Tennessee representative, and interested members of the public. The panel used many available background materials. The problems and proposed solutions associated with the salt tanks have been discussed in some detail at various technical meetings (Peretz, [996a,b), in the reports of a Senior Review Board review (February 20, 1995), and in briefings on MSRE fuel salt disposition to the CRBTW during 1996. Chemical and technical aspects of the problem, including the current state of the molten salt reactor fuel and flush salts, tanks, and related equipment (e.g., associated tubing, valves, and sensors), have been discussed and summarized in Peretz (1996c) and references therein. Presentations to the pane} (Rushton et al., 1996a,b) have reviewed the history of facility operations and have outlined a targeted timeline for safe removal and disposition of the MSRE fuel and flush salts that is consistent with CERCLA requirements. The future schedule is intended to encompass actions subject to the approval of the two agencies with regulatory authority over these actions, the State of Tennessee and the U.S. Environmental Protection Agency. SCOPE AND ORGANIZATION OF THIS REPORT To address the three key questions posed in the Statement of Task (Box ~.1), this report is organized as follows. Some of the scientific challenges common to most of the alternatives are discussed in this report first. Radiolysis and nuclear reactions are treated in Chapter 2. Fluoride salt chemistry, partitioning, and system corrosion are treated in Chapter 3. The assessment of the present condition of the salts and the drain tanks that is contained in these two chapters is relevant to any remediation approach. Other challenges are discussed as they would pertain to the execution of a remediation plan. In Chapter 4, the logical process steps are presented, from which the panel develops a preferred, stepwise 3The September meeting included a tour of the MSRE facility.
22 AN EVALUATION OF DOE ALTERNATIVES FOR MSRE approach based on the understanding that some decisions are best postponed until after further information on the system is acquired. Chapter 5 presents selected technical processing approaches to strip uranium, with commentary on the relative advantages and disadvantages of each. Issues pertaining to nuclear criticality safety provide one of the most significant constraints in some of the future processing options; therefore, these considerations are explicitly treated in Chapter 6. The scientific and technical background information in Chapters 2 through 6 is important to the assessment of strategic alternatives, which are discussed in Chapter 7. Several of the various alternatives in Peretz (1996c) are compared and contrasted to provide the basis for a recommended alternative. Chapter ~ addresses the management of potential hazards associated with MSRE salt cleanup operations. This chapter also lists several specific information-gathering activities recommended to reduce hazards and provide experimental evidence to better inform the decision- making process. The final chapter (Chapter 9) contains a summary discussion with commentary on the waste management strategy appropriate for the molten salt cleanup efforts. To the extent that future experimental evidence confirms the anticipated condition of the salt and drain tanks, a preferred approach is recommended for consideration. The three questions posed in the Statement of Task are also answered explicitly.