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.
Executive Summary The Molten Salt Reactor Experiment (MSRE) is a nuclear facility that is no longer operational and now poses a cleanup challenge to the U.S. Department of Energy (DOE). This report comments on several alternative cleanup strategies that are under consideration (Peretz, 1996a,b,c) for remediation of radioactive fluoride salts stored at the site. A brief description of the facility provides a useful context in which to understand its present status and the important issues that affect remediation plans. TEIE MSRE FACILITY AND CURRENT REMEDIATION PLANS The MSRE was built at Oak Ridge National Laboratory (ORNL) In the 1960s to explore the possibility of thermal breeding, using the thorium-232 (232Th) and uranium-233 (233U) fuel cycle. For this experiment, the fissile fuel (initially 23sUF4 [uranium tetrafluoride], and later 233UF4 and 239PuF4 [plutonium tetrafluoride]) was dissolved in a molten salt mixture and circulated at 650°C through a metal (Hastelloy N) vessel to achieve a controlled nuclear chain reaction, moderated by graphite rods. This novel nuclear reactor configuration produced a thermal power of g MW. However, after the facility was shut down in 1969, no further work was conducted on this reactor or on any similar reactor, except for design studies of a breeder reactor for which a prototype was never built (Weinberg, 19941. At the time of shutdown, the molten fused salt was drained from the reactor vessel into three metal drain tanks provided as part of the original installation. Two of the drain tanks store the now solidified fluoride salts containing most of the uranium and plutonium fuel. A third "flush" drain tank stores the fluoride salt mixture used to flush the system piping, an operation that imparted to the flush salt a small inventory of radioactive species. From the 1960s (Peretz, 1996c, p. 1-1 1; Thoma, 1971, p. 59) it was recognized that radiolysis of the solid fluoride salts produced
2 ANEVALUATION OF DOE ALTERNATIVES FOR MSRE fluorine, but uranium hexafluoride (UFO) gas was not observed. Aside from periodic maintenance checks and reheating to submersing temperatures, the facility essentially lay dormant until recent years, when migration away from the drain tanks of both fluorine and UFO (representing more than 10 percent of the total uranium) was detected in the system. Without remediation, these gases form continually due to the instability of fluoride salts in the presence of ionizing radiation and increasingly pressurize the system. The condition of the facility is such that eventual environmental or human exposures to some of the toxic materials in the facility cannot be ruled out (NRC, 19851. Health hazards may arise from radioactive materials such as actinides (especially gaseous UFO) and fission products and from chemical substances such as hydrogen fluoride (HF) and fluorine (F2) gases. Accordingly, the facility is now a cleanup priority of DOE. Both the U.S. Environmental Protection Agency (EPA) and the State of Tennessee have regulatory authority over cleanup activities. One of the cleanup objectives is to reduce the hazard associated with fluoride salts that are now stored in the three unheated drain tanks. These tanks contain 4650 kg (4.65 metric tons, or 5.13 English short tons, where one short ton is 2000 pounds) of solidified fluoride salt that was a fused solution of lithium fluoride (LiF), beryllium fluoride (BeF2), and zirconium fluoride (ZrF4~. The fuel salt contains approximately 0.7 percent (by weight) radioactive compounds specifically, uranium fluorides (the compounds UF3, UF4, UFs, and UFO are all believed to be present), plutonium tetrafluoride, fission products, and alpha decay products. The flush salt contains much smaller amounts of the same materials. This cleanup project is at a stage where DOE is assessing several technical approaches to salt removal, conditioning, and processing, as summarized by Peretz (1996c), and is beginning the regulatory approval process. Safe cleanup of the fluoride salts in this one-of-a-kind facility poses unusual challenges. More than one way to process the salts is possible in principle. One option under consideration is to remelt the salts in the drain tanks where they are now stored, redissolve any radiation-induced precipitates, and transport the fused salt (in liquid form) through the system pipes to an external vessel for further processing. A second option is to extract the
EXECUTIVE SUMMARY 3 uranium content of the fuel salt (through direct fluorination in the drain tanks after partial or complete remelting of the filet salt, in order to convert the uranium fluorides into volatile UFO gas) prior to transport of the melted salt to external storage containers. This option would leave fission products and plutonium fluorides in the residual salt. A third option is to chip or blast away solid pieces of the salt without applying heat to remelt it and then remove the salt as solid particles. In all cases, the radioactive waste would require storage at an approved site, with provision to "getter" any released fluorine and other volatile fluorides. Which technical approach should be used and why? In order to comply with regulatory requirements, several alternatives must be considered. Relevant EPA Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) criteria include the protection of human health and the environment, compliance with regulatory requirements, effectiveness (in both the long and the short terms), reduction of toxicity, implementability, and cost. The approach that satisfies these criteria most favorably should be the remediation method of choice. The challenge for project personnel is to select a remecliation approach that deals appropriately with all the potential chemical and nuclear problems that could arise in the handling of the fluoride salts. Examples of such concerns are release of UFO, exothermic chemical reactions of fluorine, and a nuclear criticality excursion due to the quantity of Missile isotopes (e.g., 233U and 239Pu) present in the system. PURPOSE OF THIS REPORT The Peretz (1996c) report, published in August, identifies various technical alternatives for the treatment and disposition of the radioactive fluoride salts presently stored in three drain tanks. At the request of DOE, the National Research Council (NRC) has undertaken the present study to review current evaluations of these different cleanup approaches. Specifically, the pane! was asked to review the proposed alternatives for removal, separation, and stabilization of the salts to determine the extent to which
4 ANEVALUATION OF DOE ALTERNATIVES FOR MSRE I. appropriate technologies and options have been identified and evaluated, 2. evaluations are sufficiently complete to form a basis for decision making, and 3. potential hazards associated with fuel and flush salt removal have been identified and addressed. Topics relevant to this determination include nuclear criticality safety, radiolysis, nuclear reactions associated with the radioactive species in the facility, and the chemistry and partitioning of the fluorine salts and fluorinated compounds that exist in various phases and compositions in the facility. A summary of the findings on these three enumerated issues, posed as responses to questions, is presented below. SUMMARY OF FINDINGS ON THREE MAJOR ISSUES To what exter'' have appropriate technologies andt options been id~enfif edt and evaluated(? The pane! finds that the relevant alternatives have been identified. The pane} does not find that any important options have been omitted, although some of the information required for a final choice is still to be developed. Nevertheless, with the present state of knowledge, the pane} considers that fluorination to extract UFO from the salt is the most promising approach for isolation of the uranium. Fluorination has been utilized successfully in the facility in the past. For in-tank fluorination to succeed, major issues need to be resolved, among them the potential for effective redissolution of all the solid phases in the drain tanks and evaluation of the extent of corrosion damage to the tank walls due to the effects of the complex interactions of ionizing radiation and fluorine compounds. Some alternative approaches exist within this favored technology. These include fluorination within the drain tanks, transfer of the salt to a new fluorination vessel, or use of the existing fluorination vessel in the process cell. Other options are to use alternative fluoridating
EXECUTIVE SUMMARY agents other than F2, such as bromine pentafluoride (BrFs), chlorine trifluoride (CIF3,) dioxygen difluoride (O2F2), krypton difluoride (KrF2), or xenon difluoride (XeF24. To what extent are the evaluations suff ciently complete fit form a basisfor decision making? The pane} finds that the final waste disposal objectives in the proposed (Peretz, 1996c) alternatives 1-6 are presently insufficient to lead to a sound remediation strategy and concludes that interim waste storage (alternative 7) is the only practical approach at present. Additional information about the system is needed to support decisions about remediation options. The final selection of an approach must be based on additional information about the system and the hazards. A crucial part of molten salt processing will be the initial melting, which will demonstrate whether the precipitated phases are manageable and the tank confinement is intact. Positive results would show that direct fluorination on a liquid system could proceed. In Peretz (1996c), the remediation strategy being used by the project is not defined. Each strategy developed for remediation of the drain tank salts should have a primary alternative and one or more backup alternatives to cover the hazard of failure of the primary alternative. A comparative cost estimate should be completed for each case. The decision maker can then optimize the choice of strategies based on probable success, initial costs and risks, and possible ultimate costs. To what extent have the potential hazards associated with fuel and flush salt removal beer' adequately identity ed arid addressed? The term hazard is used here instead of risk because the probability of occurrence of the hazard has not been defined. Assessment of a risk can be made only when the likelihood of a hazard scenario has been assessed and applied to the measure of the importance (or size) of the possible consequences. In other words, risk is quantitatively defined as the product of probability of occurrence and consequence of a hazard scenario; in the absence of numerical calculations, one may refer to hazards, not risks. Both concepts (hazards and risks) are used in this report.
6 AN EVALUATION OF DOE ALTERNATIVES FOR MSRE The Peretz (1996c) report contains a baseline hazard section that identifies the present hazards. A preliminary hazards screening has also been made (ORNE, 1995~. Evaluation of the possible hazards associated with the various process alternatives awaits the further development of those alternatives. Once the pertinent assessments of likelihood and magnitude of hazard scenarios are available, present safety standards and risk analysis methodologies, properly applied, can provide adequate delineation of hazards and their potential for becoming risks. In a risk analysis of the MSRE drain tanks, an overly conservative, upper-bound estimate of any particular risk is undesirable if the quantitative results are used to choose among various remediation alternatives, because this approach might then fail to identify the course of action with the least risk. In the absence of more complete data (e.g., realistic probability distribution functions for every uncertain parameter) and for present decisions, the panel believes that estimates of risk that provide the best basis for decision making should be on an expected value basis, bracketed by an uncertainty range. The panel believes that the probability of an inadvertent criticality hazard during mitigation of the MSRE salt is extremely low and that, even if such an event were to occur, the safety and technical consequences would be minor. However, because the public concern and political consequences could be very large, the pane! has addressed the question in some detail. Regarding radiation hazards, high radiation fields external to the drain tanks necessitate remote operations, with workers outside the shielded drain tank cell except for very brief episodic access. For example, the gamma radiation levels at the surface of the fuel salt tanks due to fission products are approximately 640 roentgens (R) per hour (Williams, 1995), primarily due to the 0.66-MeV (million electron volt) gamma rays from cesium- 137. MAJOR CONCLUSIONS The pane} concludes that cleanup of the MSRE can be accomplished with risks to the public, operations personnel, and the
EXECUTIVE SUMMARY 7 environment that can be well managed to quantitative levels at or below those values that form standard safety guidelines (such as an annual probability of occurrence of ~ o~6~. Additional information is needed to define and select the most favorable processes for the removal of fissionable materials, the associated removal of the salt, and their interim disposition. However, present information is good enough, and overall safety well enough ensured, to be used as a basis for making the decision to proceed with the overall project. The responses given above to questions posed in the Statement of Task are generally favorable judgments subject to important cautionary caveats, for example, that some of the specific parameters and procedures required for processing steps are confirmed by further information gathering, testing, and analysis. Examples and suggestions of this kind are contained in the report, particularly in Chapter ~ and Appendix E. These technical options, offered as suggestions for consideration to MSRE project personnel, are secondary to the basic recommendation that MSRE remediation work is ready to proceed, with the fluorination approach the most promising way to remove uranium, given the present state of knowledge. The panel is not chartered to recommend specific parameters or procedures, and these selections are best done by project personnel, as new information is obtained. What is missing and what could be collected in the first stages of the remediation project are current data on the fuel salts and on the condition of the piping, vent lines, and equipment of the MSRE, which has been shut down since 1969. Final selection of processes and the necessary backup systems will depend on the crucial aforementioned data and on laboratory and engineering developments. Irradiated samples from the period of MSRE operation are available for simulation and mockup of the remediation processes to be used. The project would proceed ideally in stages dete~ined in part by what is learned as the work progresses. While the drain tanks and their contents were well characterized at reactor shutdown, more than 25 years of radiolysis and corrosion have produced changes in chemistry and may have produced changes in the integrity of components. As a result, the panel recommends a cautious, stepwise approach to all operations.
8 ANEVALUATION OF DOE ALTERNATIVES FOR MSRE Fluorination procedures, subject to important caveats and to further information gathered on the system, as noted in this report, appear to be the technical approach most worthy of Further consideration. Storage of the separated fissile 233U as the uranium oxide U3Os in an existing 233U storage site at ORNE represents a suitable interim disposition. A stabilized or glittered salt residue after fluorination treatment also is a reasonable form for interim storage at ORNL. MAJOR RECOMMENDATIONS The Peretz (1996c) report on remediation alternatives is only a preliminary document related to the CERCLA process. Further analyses and development of specific plans leading to a final decision will occur over the next several years. As new information becomes available on the fuel and flush salts and on the status of the rest of the MSRE system, additional reviews of the major issues may be warranted. The pane} recommends that cleanup strategies for the MSRE project provide one or more sets of workable approaches for the safe removal, processing, and interim storage of fuels and flush salts that take into account the need for alternative and backup strategies. All relevant factors should be considered, including cost, equipment failure, criticality potential, remediation effectiveness and implementability, risk management, uncertainties, trade-offs, and duration of actions. Of equal importance is the need to fully consider possible process perturbations, failures, contingencies, resource requirements, and other factors that may warrant backup support or alternative approaches to offset them. Plans and work in progress at ORNE are addressing these issues. Sound procedures, such as stepwise processes for the acquisition of information and remediation, are a necessary strategy because not enough is known about the system (e.g., distribution of uranium in the system) to make a fully informed decision at the onset. Because the final resolution of disposal alternatives may take considerable time, the panel suggests that DOE use a phased decision strategy focusing on interim storage, with the flexibility it provides, rather than trying to make a final disposition alternative determination in the near future. It is premature at this stage to derive treatment
adhesives Mom requirements far ukim~e d~poshion in geologic repositories o~she, because these long-te~ storage options are not developed sufficiently to define Anal waste acceptance crheris~ It appears that resolution of uldm~e (geologic) disposal she criteria and characteristics probably lies beyond the time horizon of the ASH cleanup pr~ec1 The panel ~ ate 1h~ any processing of 1he Mel sags ma prevent them cl=~0c~ion a.