CHAPTER FOUR—
TRANSMUTATION SYSTEMS

The previous chapter focuses on the chemical technologies for separating radionuclides in spent fuel and defense high-level radioactive waste (HLW). This chapter considers proposals for attaining the neutron fluxes that could transmute such radionuclides, focusing on the prospects for benefiting the disposition of spent fuel from civilian reactors.1 For achieving transmutation, any practical system must integrate transmutation with separations of radionuclides to be transmuted by reprocessing of spent light-water reactor (LWR) fuel assemblies. In addition, all the proposed transmutation processes have a complex fuel cycle and would involve not one new facility but an interdependent system of many. Such systems require many components: spent-fuel reprocessing plants, waste transmutation facilities with systems to remove heat and (optionally) convert it into electricity, fuel fabrication plants, processing plants for the wastes from the separations and transmutation processes, sites for eventual disposal of the low-level radioactive waste (LLW) and any remaining HLW, as well as transportation systems for moving fuel and radioactive material between sites.

This chapter assesses the basis for transmutation by taking a systems point of view on the following proposals made by various organizations to the U.S. Department of Energy (DOE).

  • Advanced liquid-metal reactor (ALMR) as part of an integral fast reactor (IFR) system, proposed by General Electric (GE) and the Argonne National Laboratory (ANL).

  • Particle bed reactor (PBR), proposed by the Brookhaven National Laboratory (BNL).

  • Accelerator transmutation of waste (ATW) system, proposed by the Los Alamos National Laboratory (LANL).

  • Phoenix accelerator-driven fast reactor concept, proposed by the Brookhaven National Laboratory.

  • Clean Use of Reactor Energy (CURE), a study2 by Westinghouse Hanford Co. and Battelle Pacific Northwest Laboratories, involving modified liquid-metal fast reactors (LMFR) as waste transmuters in a system with LWRs.

In addition to these systems, the LWR is evaluated as a transmuter of transuranics (TRUs) and selected fission products from reprocessed LWR spent fuel. An LWR transmuter is a feasible approach in its own right, assuming the development of a fuel cycle to support such transmutation. In the analysis presented in this chapter, an LWR transmuter system serves as a reference for evaluating the advantages, disadvantages, and development requirements of the other proposals.

The chapter presents some judgments on the current level of technological maturity of the proposed systems and the prospects for their further development, using information from foreign programs for perspective. The fuel cycle required to support transmutation is considered, including separations, waste treatment, technical issues, and costs. The impacts on the repository are covered. Proposal-specific aspects of crosscutting issues from fuel-cycle economics, impacts on the repository, health and safety, licensing, and the institutional context are reviewed, but a comprehensive discussion of such issues, and of nuclear proliferation, is deferred to Chapter 6. A detailed evaluation of the concepts is provided in the appendices to this report.

1  

 Due to the low concentration of transuranics in defense HLW, it is not considered in this chapter.

2  

 The CURE study did not propose the development of a specific reactor system. Rather, the study assessed the issues involved in transmutation and proposed a follow-on study to define a specific system and its technology base. The CURE study results and the issues it raises are covered in the evaluation of the other four proposals plus the LWR transmuter; a "CURE system" as such does not appear as a distinct transmutation option in this report.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 49
Nuclear Wastes: Technologies for Separations and Transmutation CHAPTER FOUR— TRANSMUTATION SYSTEMS The previous chapter focuses on the chemical technologies for separating radionuclides in spent fuel and defense high-level radioactive waste (HLW). This chapter considers proposals for attaining the neutron fluxes that could transmute such radionuclides, focusing on the prospects for benefiting the disposition of spent fuel from civilian reactors.1 For achieving transmutation, any practical system must integrate transmutation with separations of radionuclides to be transmuted by reprocessing of spent light-water reactor (LWR) fuel assemblies. In addition, all the proposed transmutation processes have a complex fuel cycle and would involve not one new facility but an interdependent system of many. Such systems require many components: spent-fuel reprocessing plants, waste transmutation facilities with systems to remove heat and (optionally) convert it into electricity, fuel fabrication plants, processing plants for the wastes from the separations and transmutation processes, sites for eventual disposal of the low-level radioactive waste (LLW) and any remaining HLW, as well as transportation systems for moving fuel and radioactive material between sites. This chapter assesses the basis for transmutation by taking a systems point of view on the following proposals made by various organizations to the U.S. Department of Energy (DOE). Advanced liquid-metal reactor (ALMR) as part of an integral fast reactor (IFR) system, proposed by General Electric (GE) and the Argonne National Laboratory (ANL). Particle bed reactor (PBR), proposed by the Brookhaven National Laboratory (BNL). Accelerator transmutation of waste (ATW) system, proposed by the Los Alamos National Laboratory (LANL). Phoenix accelerator-driven fast reactor concept, proposed by the Brookhaven National Laboratory. Clean Use of Reactor Energy (CURE), a study2 by Westinghouse Hanford Co. and Battelle Pacific Northwest Laboratories, involving modified liquid-metal fast reactors (LMFR) as waste transmuters in a system with LWRs. In addition to these systems, the LWR is evaluated as a transmuter of transuranics (TRUs) and selected fission products from reprocessed LWR spent fuel. An LWR transmuter is a feasible approach in its own right, assuming the development of a fuel cycle to support such transmutation. In the analysis presented in this chapter, an LWR transmuter system serves as a reference for evaluating the advantages, disadvantages, and development requirements of the other proposals. The chapter presents some judgments on the current level of technological maturity of the proposed systems and the prospects for their further development, using information from foreign programs for perspective. The fuel cycle required to support transmutation is considered, including separations, waste treatment, technical issues, and costs. The impacts on the repository are covered. Proposal-specific aspects of crosscutting issues from fuel-cycle economics, impacts on the repository, health and safety, licensing, and the institutional context are reviewed, but a comprehensive discussion of such issues, and of nuclear proliferation, is deferred to Chapter 6. A detailed evaluation of the concepts is provided in the appendices to this report. 1    Due to the low concentration of transuranics in defense HLW, it is not considered in this chapter. 2    The CURE study did not propose the development of a specific reactor system. Rather, the study assessed the issues involved in transmutation and proposed a follow-on study to define a specific system and its technology base. The CURE study results and the issues it raises are covered in the evaluation of the other four proposals plus the LWR transmuter; a "CURE system" as such does not appear as a distinct transmutation option in this report.

OCR for page 49
Nuclear Wastes: Technologies for Separations and Transmutation The sections listed below summarize the evaluation of the three primary transmutation concepts, namely, the ALMR/IFR, ATW, and LWR systems, plus an initial assessment of the PBR and Phoenix systems, on which less information is available. The evaluation that follows in the sections to this chapter covers the following topics: transmutation processes and concepts; reduction of transuranic inventories; reduction of key fission product inventories; safety issues for the reactor, fuel materials, and supporting fuel cycle; development time, cost, feasibility, and risk through system demonstration; time scale and costs for complete system deployment; and comparative thermal and electrical efficiencies. TRANSMUTATION PROCESSES AND CONCEPTS The subsection below, Transmutation Processes, reviews the principal nuclear processes for transmuting the key fission products and transuranics (TRUs). The subsections Transmutation Reactor Approaches, Overview of Critical Reactor Concepts, and Overview of Accelerator-Driven Reactor Concepts introduce the transmutation proposals and the LWR as a waste transmuter and reference system for the proposals. The subsection Separations with Transmutation of Plutonium (Only) introduces a simplified transmutation concept in which the only TRU burned is plutonium, possibly with key long-lived fission products. Transmutation Processes Transmutation of Key Fission Products The main transmutation processes for fission products are neutron capture (producing a higher mass nuclide) and beta decay. For many fission products the neutron capture cross sections in a thermal (or epithermal) spectrum can give substantial transmutation rates. The corresponding capture cross sections in a fast neutron spectrum are typically orders of magnitude smaller. Transmutation of the long-lived fission products 99Tc and 129I is feasible in a thermal reactor. The ruthenium and xenon transmutation products are stable under neutron capture processes; that is, and The 129I is produced with stable 127I, which also transmutes to stable isotopes. For a thermal neutron spectrum typical of a uranium-fueled PWR, the spectrum-averaged capture cross sections for dilute 99Tc and 129I from the ORIGEN-2 code library are 13.8 barns and 3.2 barns, respectively (Wachter and Croff, 1980).3 An ORIGEN-2 calculation for a PWR gives in-reactor transmutation rates of about 11 percent per year for 99Tc and 3 percent per year for 129I (Wachter and Croff, 1980). Practical transmutation rates will be low because the external inventory of 99Tc and 129I in reprocessing and target fabrication must be taken into account. Transmutation of 90Sr (29 yr) and 137Cs (30 yr) is possible in principle but not in practice. Their transmutation would make little improvement in the calculated radiological risk for a geologic repository because of the relatively short-term nature of their radioactivity, although the reduction in repository heat loading might be marginally advantageous. Transmutation of the long-lived radionuclide 135Cs would also be possible in principle but not in practice. This radionuclide would be transmuted to stable 136Ba through short-lived 136Cs. Unfortunately, 135Cs is produced with a larger quantity of stable 133Cs, which would undergo neutron capture to 134Cs and to more 135Cs. To remove the stable 133Cs beforehand by isotopic separation would be formidable in the presence of the intense 137Cs gamma radiation field. Rather than attempting to transmute 135Cs, its radiological risk in a geologic repository could be reduced by separating cesium in the reprocessing operation and incorporating it into waste form in which cesium would have a significantly reduced solubility in groundwater. Transmutation of Transuranics Characteristics of Thermal and Fast Neutron Processes. Transmuting the TRUs is more complex (Benedict at al., 1981). In a neutron flux, several competing processes determine the concentrations of the transmuted isotopes as a function of time. Three processes are important here: (1) neutron-induced fission, (2) neutron capture to produce a higher-mass nuclide, and (3) radioactive decay. Table 4-1 lists the half-lives of selected uranium and TRU isotopes that occur during transmutation and the ratios of their spectral-averaged capture and fission cross sections for several kinds of thermal and fast neutron spectra. A number of characteristic features may be seen by a study of the properties. 3    1 Bam = 10-28m2.

OCR for page 49
Nuclear Wastes: Technologies for Separations and Transmutation First, the "capture-to-fission" ratios for thermal neutron spectra show marked differences for odd-versus even-neutron isotopes. Odd-neutron isotopes fission well in a thermal spectrum; in fact, such fission is favored over capture by factors of 2 to 10. In contrast, most even-neutron isotopes are not fissioned by neutrons below several hundred kilovolts in energy, due largely to the extra nuclear stability conferred by neutron pairing. Hence, in a thermal reactor, such nuclei are fissioned only by the fission spectrum neutrons. Second, the thermal neutron "capture-to-fission" ratios are typically higher than those for fast spectra. This effect is exaggerated for even-neutron isotopes; indeed, 240Pu and 241Am have large thermal neutron capture cross sections and are parasitic absorbers in thermal reactors. However, odd-neutron isotopes also show the effect by factors of 1.5 to 3, mainly because capture cross sections are typically higher for thermal spectra. Most importantly, the high thermal capture-to-fission ratio of 239Pu and its capture products result in relatively large amounts of higher-mass actinides in a thermal-neutron transmuter, as compared to a fast reactor. Transuranic Production and Transmutation. The main source of the principal TRU, 239Pu, is neutron capture in 238U followed by two beta decays (i.e., in the initial spent fuel plus possible additional production during transmutation). The 239Pu fissions well with neutrons of any energy. Alternatively, it produces 240Pu by neutron capture. Table 4-1 lists a capture-to-fission ratio of 0.55 in a thermal spectrum compared to 0.17 and 0.26 in a fast spectrum with metallic and oxide fuel, respectively. Successive neutron captures, starting with 239 Pu, produce higher mass isotopes of Pu, Am, and Cm. The minor actinides and other plutonium isotopes, in particular 240Pu and 242Pu, fission well in a fast spectrum. That isotope 237Np could undergo neutron capture and furnish additional 239Pu. That is, However, in a very high neutron flux, the intermediate nucleus 238Np could attain a high probability of capturing a second neutron and fissioning before the beta decay could take place. Thus, in principle, a high-flux transmutation scheme could achieve a higher fissioning rate of various isotopes than a scheme that operates at ordinary thermal flux levels. Two even-neutron isotopes in the plutonium chain, 240Pu and 242Pu, are key precursors of higher-mass actinides by neutron capture to 241Pu and 243Pu, which can undergo beta decay to 241Am and 243Am, respectively (see Table 4-1). The two americium isotopes fission well in a fast spectrum, but can produce 242Cm (163 d) and 244Cm (18.1 yr) by neutron capture in a fast or thermal spectrum. However, the higher capture-to-fission ratio characteristic of a thermal spectrum, discussed above, results in a build-up of higher-mass actinides during thermal neutron transmutation. A chain of long-lived curium isotopesis produced by successive neutron captures—245Cm (8,500 yr), 246Cm (4,820 yr), 247Cm (1.56 × 107 yr), and 248Cm (3.7 × 105 yr), even some 250Cm (9.7 × 103 y). The chain branches at 249Cm with the 64-minute beta decay to 249Bk, which leads in steps to production of californium isotopes, in particular 252Cf. Even higher-mass actinides can be produced in a thermal flux >1015 neutrons/cm2-s, such as in the high-flux isotope reactor (HFIR). Thus, a range of higher-mass isotopes is produced in either a fast or thermal spectrum, although the relative proportions are quite different. The possible effects of the higher actinides are discussed in the sections on several of the transmutation options. In target waste material, 252Cf would be a potent source of neutrons by spontaneous fission. In addition to alpha emission, spontaneous fission also occurs in curium isotopes, becoming more probable with increasing curium mass, i.e., 242Cm (8 × 10-6%), 244Cm (1.3 × 10-4%), 246Cm (0.027%), 248Cm (0.83%), and 250Cm (~99%) (Hoffman et al., 1992). In addition, unburned 238Pu can be an important source of neutrons by (α, n) processes. Hence, for thermal neutron transmutation concepts, significant neutron emission could present problems during fuel reprocessing and refabrication, quality assurance, and performance verification. To a lesser extent, the issue could arise with fuel rods for transmutation in a fast reactor. Transmutation Reactor Approaches Classes of Reactor Concepts To take possible advantage of nuclear processes for transmutation, two quite different classes of reactors have been studied for neutron generation. The work to date consists primarily of a conceptual analysis of the effects of the neutrons generated in either approach in order to estimate the benefits and hazards that would result from changing the character of the waste to be disposed. The two classes of reactor concepts are: Critical nuclear reactors: The nuclear assembly, containing the waste and possibly additional fissile material, operates with a net neutron multiplication factor of unity. This class includes thermal reactors, such as the LWR and PBR, and fast reactors, such as the ALMR. Accelerator-driven nuclear reactors: The nuclear assembly operates with a neutron multiplication factor less than unity, i.e., subcritical, so that neutrons must be added from a source external to the nuclear assembly. Intense

OCR for page 49
Nuclear Wastes: Technologies for Separations and Transmutation TABLE 4-1 One Group Capture-to-Fission Cross-Section Ratios of Transmutation Isotopes for Thermal and Fast Neutron Spectra in Various Reactor Cores Isotope Half-Life U-PWRa Pu-PWRb LMFBRc ALMRd Main Source 235U 7.04 × 108 yr 0.22 0.28 0.27 0.25 natural uranium 236U 2.34 × 107 yr — — — — neutron capture by 235U 237U 6.75 d — — — — neutron capture by 236U & 238U 238U 4.47 × 109 yr 9.02 7.93 5.44 4.34 natural uranium 239U 23.5 min — — — — neutron capture by 238U 237Np 2.14 × 106 yr 61.3 42.43 4.22 2.71 6.8 d beta decay of 237U 238Np 2.1 d 0.10 0.10 — — neutron capture by 237Np 239Np 2.35 d — — — — 23.5 min beta decay of 239U 238Pu 87.7 yr 14.06 7.51 0.60 0.44 2.1 d beta decay of 238Np 239Pu 2.44 × 104 yr 0.55 0.56 0.26 0.17 2.35 d beta decay of 239Np 240Pu 6,560 yr 178.1 70.55 1.25 0.82 neutron capture by 239Pu 241Pu 14.4 yr 0.33 0.31 0.18 0.15 neutron capture by 240Pu 242Pu 3.75 × 105 yr 76.43 47.78 1.42 0.93 neutron capture by 241Pu 243Pu 4.98 h 0.49 0.49 — — neutron capture by 242Pu 241Am 433 yr 94.2 63.96 4.95 3.56 14.4 yr beta decay of 241Pu 242Am 16 h — — — — neutron capture by 241Am 242Am 141 yr 0.21 0.20 0.10 0.07 neutron capture by 241Am 243Am 7,370 yr 106 78 3.84 2.29 4.98 h beta decay of 243Pu and neutron capture by 242Am 242Cm 163 d 10.38 11.4 1.59 0.85 16 h beta decay of 242Am 243Cm 28.5 yr 0.12 0.13 0.09 0.07 neutron capture by 242Cm 244Cm 18.1 yr 15.79 14.83 1.71 1.21 10 h beta decay of 244Am 245Cm 8,500 y 0.17 0.17 0.12 0.10 neutron capture by 244Cm 246Cm 4,820 y 5.04 4.71 0.72 0.47 neutron capture by 245Cm 247Cm 1.56 × 107 yr 0.65 0.63 0.16 0.12 neutron capture by 246Cm 248Cm 3.4 × 105 yr 8.36 7.88 0.68 0.44 neutron capture by 247Cm 249Cm 64 min 0.03 0.03 — — neutron capture by 248Cm 250Cm 9,700 yr — — — — neutron capture by 249Cm 249Bk 523 d 807.0 460.3 4.63 1.92 64 min beta decay of 249Cm 250Cf 13.1 yr 513.6 292.8 0.37 0.20 fast beta decay of 250Cf 251Cf 800 yr 0.48 0.45 0.12 0.10 neutron capture by 250Cf 252Cf 2.65 yr 0.47 0.41 0.39 0.32 neutron capture by 251Cf a Normal low-enriched uranium in a pressurized water reactor. b Self-generated plutonium recycle in a pressurized water reactor. c Mixed oxide fuel in a liquid-metal fast reactor. d Metallic fuel in an advanced liquid-metal (fast) reactor. SOURCES: U-PWR and Pu-PWR from ORIGEN 2.1 (1991). LMRBR and ALMR from GEFR-00898 (1991).4 beams of very high-energy protons would be focused on targets such as lithium, tungsten, molten lead, or even the fuel itself. This would generate large numbers of spallation neutrons that would be multiplied by the subcritical assembly to transmute waste material surrounding the target. The various transmutation concepts differ widely in the many technical, programmatic, economic, and other dimensions by which they may be characterized. However, there are no universal figures of merit for the evaluation of the different transmutation approaches. The remaining sections of this chapter summarize the results of the evaluation using a variety of "measures" for comparison: the rate and time for various percentage reductions in the TRU inventory: the flexibility and rate for reducing key fission product inventories: 4    Other models are available for calculations and may give different values.

OCR for page 49
Nuclear Wastes: Technologies for Separations and Transmutation the safety issues for the reactor, fuel materials, and supporting fuel cycle; the development time, cost, feasibility, and risk through complete demonstration; the estimated time scale and costs for complete deployment, including overall fuel cycle economics; and the comparative thermal and electrical efficiencies per net amount of waste transmuted. The first type of measure, the reduction in TRU inventory, raises a methodological issue. Over the life cycle of a reactor, approximately 30 years, most of its initial load of TRUs is transmuted; however, additional TRUs are generated. Thus, only a partial net reduction occurs over each reactor life cycle, the untransmuted TRUs being passed to the next generation of reactors for further reduction. Some groups and authors emphasize the time to reprocess and burn the TRUs originally in the LWR spent fuel, which is much shorter than the time required for a net reduction including the TRUs created during transmutation. Others emphasize the TRUs in waste sent to a repository from reprocessing, discounting surface-stored waste at fuel-cycle facilities on the grounds that the fuel-cycle material still contains fissionable resources. However, for the major part of the transmutation period, most of the inventory is in surface facilities. This results in increased risk to nearby populations, compared to the direct disposal of spent fuel in the repository, due to the additional operations to burn the waste. As one figure of merit for each type of transmutation system, therefore, this report evaluates the net decrease in total TRUs versus time for multiple reactor life cycles of that system, compared to an untransmuted reference case. That is, the net TRU ratio as a function of time is defined to be the TRUs from a once-through, uranium-fueled LWR cycle for a given electrical power level, divided by the total TRUs for the system in question (in the reactor, the associated fuel-cycle facilities and the waste) for that power level. The time scale for achieving various percentage reductions, the asymptotic reduction factors, and the neutron efficiencies for the concepts are useful for comparisons. The calculated TRU ratios and the corresponding times for various fractional changes in net inventory range widely among the proposals. For evaluation purposes, this report considers two nuclear power scenarios, under the assumption that the transmutation systems would produce electrical power for distribution to the electrical grid. For example, for a continuing "steady nuclear power" scenario with a mix of breeding and nonbreeding transmuters, achieving a net TRU ratio of a factor of 100 requires from a few hundred years for some proposals to several thousand years for others. A TRU ratio of 10 requires from about 7 years to over 400 years for the various proposals at constant nuclear power. The time scales are significantly shorter for a "declining nuclear power" scenario, in which nuclear power is phased out as rapidly as possible consistent with transmuting the most nuclear waste. For this type of scenario, for example, the calculated time to achieve a TRU ratio of 100 ranges from about 40 years to several hundred years. Engineering and Materials Issues In both the critical reactor and accelerator-driven subcritical reactor approaches, there are variations in the chemical and physical properties of the fuel containing the waste to be transmuted. For this reason, all the proposals entail fuel development and verification issues. Moreover, even for the systems operating with subcritical reactors, the heat generation rates would require careful analysis of the provisions for safe removal of residual heat generated from the decay of fission products and minor actinides after the termination of the fission process. Also, the transmutation systems that employ a high fissile specific power (i.e., the thermal-reactor concepts) are likely to raise safety issues associated with the control of reactivity and power distributions in the reactor core. Detailed analysis of such issues is important for the licensability of the system designs. All the concepts require development and testing of at least some aspects in a pilot-scale system. Indeed, most of the systems will require demonstration at full-scale prototype before such systems could be implemented confidently for reliable operation at high-capacity factor. Several proposals aim at producing and sustaining a very large fractional destruction rate—the product of neutron flux and cross section—since that determines the rate of transmutation of the waste. A large destruction rate, in turn, entails high heat generation rates from the high fission rate and decay heat of nuclei in the reactor proper and, in the accelerator-driven case, from proton deposition in the target. For these proposals the engineering designs pose difficult problems. In some proposals the materials must withstand conditions beyond engineering experience with present-day reactors. These conditions include neutron (and in the case of accelerator-based proposals, proton) fluxes and fluences, neutron spectra, temperatures, and chemical environments, singly or in combination. High-energy neutron and proton fluxes impart atomic displacement damage that generally degrades structural materials and affects basic properties (e.g., ductility, fracture toughness) and imparts dimensional instabilities (e.g., swelling and creep). Such flux levels would also increase the level of activation products in the nuclear waste. Indeed, some of the proposed systems envision neutron fluences for certain components that are beyond the levels achieved for the corresponding components in existing systems,

OCR for page 49
Nuclear Wastes: Technologies for Separations and Transmutation e.g., more than 100 displacements per atom for some structural components or roughly an order of magnitude more than experienced in the Fast Flux Test Facility (FFTF). The displacement damage can be aggravated by nuclear transmutation that converts existing alloying elements to impurities. Two such processes are (1) transmutant gas production (e.g., helium from (n, α) production) arising from the high-energy portion of the neutron spectra in accelerator-based sources, and (2) transmutation of existing alloying elements to deleterious activation products by neutron absorption in approaches with a thermal spectrum. These types of induced impurities compound the difficulties imposed on structural materials required to operate at the high temperatures due to the high flux/power and density/thermal efficiency requirements of several of the transmuter concepts. Additional deterioration can be anticipated with those concepts that employ nonaqueous chemical environments (e.g., molten salts, liquid metals, aqueous slurries) to achieve the system performance requirements. Thus, several proposals entail considerable research and development before the chances of success can be ascertained. Program Funding As yet, there is no focused overall program in the United States for development of S&T technologies. The objectives and time table for such an effort are difficult to work out in the present climate of uncertainty about the future of nuclear power in the United States—in particular, uncertainty concerning the prospective geologic repository at Yucca Mountain and the possible future commercialization of a fast breeder reactor sometime in the next century. This committee, therefore, is able to estimate only an approximate total expenditure level for the U.S. effort. For FY 1992, DOE's Office of Nuclear Energy (DOE/NE) identified direct expenditures of about $19 million specifically for transmutation objectives: about $15 million for the LANL's ATW effort and about $4 million for transmutation applications of the ALMR and IFR concept. In addition, the national laboratories used discretionary funding to extend their efforts. A few million dollars was spent at LANL and BNL in FY 1992 on accelerator-based concepts for transmutation. Smaller efforts were conducted by Westinghouse Hanford and Oak Ridge National Laboratory (ORNL). However, DOE was spending a much larger amount for research and development indirectly related to transmutation objectives. For example, the ALMR effort at GE and the IFR at ANL were aimed primarily at breeder reactor development for future power production which are both adaptable to transmutation. This support included the support for ANL-West located at the Idaho National Engineering Laboratory. In addition, the development of accelerators for tritium production and other military objectives was indirectly relevant. For FY 1992 then, the efforts directly or indirectly applicable to transmutation systems amounted to about $75 million. The total funding allocated to separation and transmutation in FY 1993 was is $133.5 million. DOE/NE identified direct expenditures of $41.4 million specifically for transmutation objectives: $8.9 million for ALMR development at GE, $26 million for metal fuel and IFR development at ANL, and $6.5 million for LWR actinide burning. The related ANL-West supporting facilities would add $76.1 million to the FY 1993 transmutation commitment. In addition, the DOE defense budget includes $4 million for ATW-related efforts at LANL, and the waste management budget includes $12 million for efficient separations and processing. In FY1994 the administration began a phase-down or phase-out policy of the DOE/NE programs related to separations and transmutation. In FY1995, a total of $104.8 million has been identified for these programs including IFR, but these funds are exclusively designated for close out and termination costs. International Activities International activities in S&T as a method of ameliorating high-level waste management are being conducted by individual countries as well as under the coordination of international organizations (the Organization for Economic Cooperation and Development's Nuclear Energy Agency, the International Atomic Energy Agency, and the Commission of the European Communities). Most of this activity is in Japan and Western Europe, with the majority of the European activity in France. Unlike the U.S. S&T program, however, the interest in Japan, France, and the United Kingdom stems from a larger interest in commercially reprocessing fuel, both in the near-term for LWRs using mixed-oxide (MOX) fuel and in the long-term for a breeder reactor economy. As a result, these countries are working on the technologies necessary for any S&T waste management scheme, including enhanced reprocessing techniques, remote fuel fabrication, and reprocessed waste packaging. The interest in these countries in S&T for waste management has increased recently, largely as a result of increasing public resistance to high-level waste repository siting. Japan has perhaps the largest financial commitment to this effort through its OMEGA Project, which is a research and development effort aimed at partitioning actinides from HLW and transmuting actinides in either critical reactors or accelerator-driven subcritical assemblies. French and U.K. efforts have looked at actinide burning in fast reactors, but more from a perspective of controlling reactivity swings. Coordinated European programs have considered actinide

OCR for page 49
Nuclear Wastes: Technologies for Separations and Transmutation burning in fast reactor blankets, and preliminary work at the Paul Scherrer Institute in Switzerland has considered accelerator-driven spallation for actinide burning. Finally, both Japanese and European programs are examining nonaqueous separations techniques as well as enhanced aqueous reprocessing. Overview of Critical Reactor Concepts This subsection gives an overview of a fast reactor system and two types of thermal reactor systems that are proposed or are being considered for transmutation of nuclear waste, namely, transmutation of TRUs in an ALMR as part of an IFR proposed by GE and ANL; transmutation of TRUs and fission products in LWRs, either existing designs adapted for the purpose or more advanced designs currently in certification review by the Nuclear Regulatory Commission (NRC); and transmutation of minor actinides with a PBR, as proposed by BNL. The concept of transmutation using critical reactors has been studied for several decades; indeed, the United States was a leader in the field in the 1970s. Early in the commercialization of LWRs in the United States, the U.S. Atomic Energy Commission (AEC) and the nuclear industry expected that the cost of mining and enriching uranium would increase as high-grade uranium ore supplies were used up. In the 1960s and early 1970s, the AEC projected that this would occur by the 1980s and become increasingly serious in the 1990s. The parties visualized that spent LWR fuel would be reprocessed commercially and the plutonium and unburned uranium separated and recycled, first into LWRs and later into breeder reactors when they became commercialized. Thus, the total plutonium in the fuel cycle would be limited. Fast Critical Reactor Concepts The ALMR/IFR program had been sponsored by DOE/NE during the previous administration under joint development by GE and ANL.5 It emphasized the power reactor, innovative, small module (PRISM) fast reactor concept, which has been aimed primarily at the development of a breeder reactor for power production that would produce more fissile material than it consumes. It has long been recognized that a fast reactor can be modified to operate at a conversion (breeding) ratio, ß, less than 1.0 and thus could be configured to operate as a net burner of transuranics.6 For that matter, the neutron spectrum of a fast reactor can be softened, as proposed, for example, in the CURE concept, and gain capability to transmute fission products while retaining good characteristics for transmuting transuranics.7 This subsection discusses the GE and ANL proposal for an IFR for burning the TRUs. An IFR would comprise an ALMR for the purpose of burning TRUs together with its own self-contained capability for processing and fabricating the metallic fuel/waste alloy. For such reprocessing, ANL proposes to use pyrometallurgical technology. ANL is also exploring pyroprocessing of LWR spent fuel to provide the TRUs to start the ALMRs; this capability might be collocated with each IFR or configured in a larger, centralized facility to support many IFRs. As an alternative, GE has considered a centralized capability for LWR spent-fuel reprocessing based on aqueous plutonium and uranium extraction and recovery by transuranic extraction (PUREX-TRUEX) technologies. More than a dozen fast breeder reactors have been operated in the United States, the Soviet Union, the United Kingdom, France, and Japan, although none are yet on-line as breeders for reliable power production. Further development would be needed to burn minor actinides (MA), including operation of a prototype integrated system. Designs with a breeding ratio, β, of 0.22 to 1.25 appear in the ALMR/IFR project literature (Chang, 1991a, 1991b; M.L. Thompson, private communication, 1991). Recent GE studies (Thompson, 1992) involve elimination of external blankets and shortening of core height, without changes in fuel rod diameter or number of fuel rods. Preliminary ANL designs (Johnson et al., 1990) include IFRs fueled entirely with TRUs. One design involves a mixture of plutonium and MA with β = 0.22; another involves mostly MA with β = 0.85. However, with reduced fissile breeding and heavy metal inventory in the core, the TRU burner designs entail increased specific power and higher reactivity swing over a fuel cycle. GE suggests that for a TRU burner, β = 0.60 is the minimum 5    In 1994, following recommendations by DOE, the ALMR/IFR was canceled by Congress. Limited funds were appropriated for phase out and for continuation of research on separations. 6    One should note that a fast reactor with ß=1, a breeder, burns the original TRUs with which it is loaded and can properly be thought of as a waste burner. 7    The CURE concept (Rawlins et al., 1990) assessed a modified sodium advanced fast reactor (SAFR), which has a 900 MWt (400 MWe) capacity and a homogeneous core with both radial and internal uranium blanket assemblies. The radial blanket assemblies were assumed to be replaced by hydride-modified assemblies containing the target wastes to be transmuted. The hydride would moderate the fast neutrons leaking from the core into the epithermal energy range. The spectrum is soft enough to achieve annual destruction rates for the fission products 99Tc and 127I of about 5 percent/yr and 10 percent/yr, respectively. Yet, the spectrum is still hard enough to burn actinides quite efficiently. The resulting breeding ratio would be about 0.8, compared to about 1.05 for the normal SAFR; thus the design is a net consumer of plutonium.

OCR for page 49
Nuclear Wastes: Technologies for Separations and Transmutation value with acceptable safety features. This limits the waste inventory reduction capability of the system. The consequences of this limitation for the inventory reduction factor, and the time required to burn various fractions of the actinides, are explored in detail later in this chapter (see Appendix F). Thermal Critical Reactor Concepts Several earlier studies of transmutation (Croff et al., 1977; Croff, 1980; Wachter and Croff, 1980) included LWRs as transmutation devices. Those studies focused on transmuting the MA, assuming that the recovered uranium and plutonium from LWR spent fuel would be recycled to the LWRs. Wachter and Croff also showed that an LWR could transmute 99Tc and 129I. However, the reprocessing of commercial reactor fuel did not go forward in the United States and is not currently considered viable for recycle to LWRs or breeder reactors, for both economic and public policy reasons. Consequently, transmutation schemes now proposed to assist the geologic disposal of LWR spent fuel (and/or accomplish other energy objectives) must deal with the transmutation of all the TRUs, the principal one being plutonium. The section, The Light Water Reactor in Appendix F, discusses the LWR as a reference case for burning plutonium, the MAs, and the key fission products 99Tc and 129I. The section, The Particle Bed Reactor Concept in Appendix F, evaluates the BNL proposal of a PBR for transmutation (Van Tuyle, 1992). The PBR is a thermal reactor concept based on a gas-cooled space reactor design. The PBR would have a compact core and operate with a power density of 5 MWt/liter, which is a factor of 102 greater than that of a typical LWR and even higher than the 1.13 MW/liter attained by the high flux isotope reactor (HFIR). The fuel material is embedded in small graphite particles (0.8 mm in diameter) packed in cylindrical pressure tubes. The tubes accommodate the packed particle bed within two coaxial porous plugs (frits) so that the particles could be cooled by forced circulation of helium. Heavy water and beryllium carbide moderators are being studied. The BNL proposal envisions burning the MAs, with the unburned plutonium and perhaps the unused fissile uranium to be directed for recycle to LWRs. A power reactor with such a high power density would be without precedent or experience as a commercial power reactor. Overview of Accelerator-Driven Reactor Concepts An alternative to using fast or thermal critical reactors is to generate a fraction of the neutrons by spallation with high-energy protons from an accelerator. LANL and BNL have made conceptual studies for achieving subcritical accelerator-based transmutation (see Appendix F). These concepts envision the production of electricity, part of it to power the accelerator and part available for sale to offset the costs of such an approach. The accelerator-based concepts are relatively new and are less well characterized and in a less mature state of technical development than the approaches using critical reactors. For these reasons, after introducing the specific proposals, the general features of accelerator-based approaches compared to those of the more familiar critical-reactor-based transmutation will be highlighted. Appendix F treats four concepts for ATW under review by LANL for commercial LWR waste (Arthur, 1992a).8 Spallation neutrons are produced when the proton beam strikes a lithium (or heavy metal) target inside the assembly, which moderates the neutrons to a thermal spectrum and multiplies them by a large factor, namely, 20 in the current designs (corresponding to a neutron reproduction factor Keff = 0.95 for the subcritical assembly). The concepts differ in the type of fluidized waste/fuel (either a heavy-water slurry or a molten salt solution that circulates in a blanket), the type of processing used (aqueous or nonaqueous), the target material, and whether thorium is used in the molten-salt cases to provide production of fissile 233U. Each facility would have its own on-line fuel-processing capability, various storage and waste treatment facilities, and heat conversion and electrical power generation equipment. The four concepts are as follows: transmutation of TRUs and some fission products recovered by aqueous reprocessing of LWR spent fuel, using an aqueous neutron multiplying system that generates electrical power (Case ATW-1); transmutation of TRUs and some fission products recovered by nonaqueous reprocessing of LWR spent fuel, using a nonaqueous neutron multiplying system that generates electrical power (Case ATW-2); transmutation of TRUs and more fission products recovered by nonaqueous reprocessing of LWR spent fuel, using a nonaqueous neutron multiplying system, fueled partly with thorium and generating electric power (Case ATW-3); and transmutation of TRUs and some fission products in a nonaqueous thorium-breeder system for electrical power generation (Case ATW-4). The LANL effort is in an early conceptual design stage. The baseline design, referred to as Case ATW-1 in this report, employs known technology to the extent possible and 8    LANL has proposed a fifth concept for application to defense wastes, using an aqueous multiplying system similar to that of Case ATW-1. The scheme is marginal because of the concentration of TRU in the defense wastes. The concept is not further considered in this chapter.

OCR for page 49
Nuclear Wastes: Technologies for Separations and Transmutation furnishes a reference materials balance. LANL makes no claim that this design is optimized. LANL is also exploring three nonaqueous approaches with improved performance and economy. Referred to in this report as Cases ATW-2, ATW-3, and ATW-4, they attempt to optimize both the system neutron economy and overall system electric power production. In contrast, the Phoenix concept proposed by BNL is envisioned to be part of a larger scheme for separating and transmuting waste from LWRs (Van Tuyle, 1992). As discussed in the section, The Accelerator-Based Phoenix Concept (see Appendix F), the Phoenix proposal would transmute the MAs and the radio-iodine from the reprocessed LWR spent fuel. The separated plutonium and uranium would be recycled to LWRs. Like the LANL ATW concept, Phoenix generates neutrons by spallation with a high-energy proton beam from an accelerator and multiplies the neutrons in a subcritical assembly that contains the waste to be transmuted. Unlike the LANL ATW, the proton target in Phoenix is composed of the MA waste itself. The neutrons have a hard spectrum, in contrast with the thermalized spectrum of the ATW. The Phoenix concept builds on existing oxide fuel technology developed under the FFTF program, together with aqueous separations technologies to reprocess both LWR and Phoenix spent fuel. Since there is no plutonium in the Phoenix initial fuel charge, there is insufficient neutron generation for the MAs to burn both the radio-iodine and the technetium fission products. The discharged HLW containing technetium and the other fission products would be packaged for surface storage and, ultimately, geologic disposal. Thus, the scope of HLW transmutation would not be as extensive as in the ATW concept and is more like that of a fast spectrum, critical reactor approach, such as the ALMR. The ATW and Phoenix concepts have several general features in common that can be contrasted with critical reactor-based concepts. Both laboratories point out that high fissile specific power can be attained that can result in fast burn-up rates of the actinides and certain fission products as well. BNL points out that the combination of high flux and hard spectrum in Phoenix results in shorter times to achieve given reductions in actinide inventories than possible with an ALMR. Indeed, LANL believes that a neutron spallation source coupled to a fission blanket provides a fundamental enhancement in the effective number of neutrons per fission. Their calculations project a high-intensity thermal flux of 2 to 3 × 1015 neutrons/cm2-s, which is about an order of magnitude higher than is typical of thermal power reactor systems (Bowman et al., 1991). For such a system, LANL projects two key advantages (Arthur, 1992b). First, a high neutron flux could provide more efficient destruction of certain actinides by enabling a nucleus to undergo two sequential (n, α) reactions, increasing the probability of the nucleus capturing a second neutron before the state created by capture of the first neutron would decay. Thus, an ATW system could, in principle, efficiently transmute some actinides that normally cannot be transmuted well with a thermal neutron flux but require fast neutrons. However, the high thermal flux of the ATW would lead to a build-up of higher MAs in the residual inventory as transmutation proceeds, in particular, 242Cm, 244Cm, 246Cm, and 248Cm as well as 252Cf, all of which are significant neutron emitters. The level of radioactivity would be considerably higher than for the LWR and, of course, orders of magnitude higher than with a fast neutron spectrum. As discussed in Appendix F, the level of radioactivity may affect the quality assurance of the slurry/liquid target fuel during transmutation. Second, the separation of accelerator and target/blanket assembly allows for rapid transmutation and entails relatively low inventories. Indeed, a main feature of the ATW, discussed in detail in Appendix F, is that it can transmute a given fraction of its TRU inventory far more rapidly than a fast reactor. This is a consequence of the lower critical concentrations characteristic of thermal reactors and the lower inventory in the on-line reprocessing system, as discussed later in this chapter. Finally, both laboratories assert that an accelerator-based waste transmuter has potentially improved safety features vis-à-vis critical reactor concepts because, at least in theory, the assembly can be maintained subcritical, and shutdown can be accomplished rapidly by shutting off the accelerator. However, the discussion in Appendix F on the ATW and later in this chapter brings to light several types of transients that pose potentially significant safety issues for the accelerator-based concepts, in addition to the issues posed by decay-heat removal and target-heat dissipation. Separations with Transmutation of Plutonium (only)9 An LWR or an ALMR would operate much more efficiently with 239Pu as the recycle fuel than with recycle of all the TRU isotopes contained in the LWR spent fuel. This suggests a waste management concept intermediate between the once-through LWR fuel cycle and the full transmutation of all TRUs and selected fission products. That is, the accumulated LWR spent fuel could be reprocessed and only the separated plutonium recycled to either an LWR or ALMR. In addition, 14C and 129I could be captured, e.g., as off-gasses, and packaged in low-solubility waste forms that 9    The National Academy of Sciences had conducted another study on weapons plutonium and a report ''Management and Disposition of Excess Weapons Plutonium" was published in 1994.

OCR for page 49
Nuclear Wastes: Technologies for Separations and Transmutation could be developed. The recovered uranium would be packaged for geologic disposal or possibly for recycle to an enrichment facility. An LWR could also transmute the long-lived 99Tc and 129I fission products if they were separated and recycled. An ALMR could operate with the types of fuel previously developed for the fast breeder; the fast neutron spectrum would minimize higher plutonium build-up. With a LWR, the essential difference between this plutonium recycle concept and the "classic" thermal recycle of plutonium, demonstrated in the late 1960s, is that here the plutonium would be multiply recycled, resulting in a build-up of higher plutonium isotopes in the LWR thermal neutron spectrum. In particular, 240Pu and 242Pu are significant absorbers of thermal neutrons and reduce reactivity. Reactivity of the multiple recycle LWR fuel, however, can be maintained satisfactorily with make-up 239Pu or uranium of moderate enrichment. For such plutonium-only transmutation, present-day separations technology using PUREX may be sufficient, having process losses on the order of 1% for plutonium (see Chapter 3). The other TRUs would go to the HLW for geologic disposal, although some of the neptunium might also be recycled if that proved easy to do. The PUREX process waste might be contained in a glass waste form, as planned for the defense HLW. As noted above, a waste form would need to be developed for the14C. By also separating certain long-lived fission products, one could possibly devise other improved waste forms, e.g., for 99Tc and 135 Cs. If pyroprocessing were used with an ALMR plutonium burner, additional new waste forms would be required. Of course, the big costs are those of LWR spent-fuel reprocessing and the all-remote fabrication of plutonium-bearing recycle fuel. Moreover, this waste management concept opens the door to the major public policy issues entailed in the commercial use of plutonium, in particular the relationship to U.S. nonproliferation policy (see Chapter 6). The main benefits to waste management would come from (1) the improved waste forms and/or transmutation of the long-lived fission products and (2) waste form for 14C, which is an apparent issue at present for the special case of the Yucca Mountain repository. However, it is problematic whether plutonium-only transmutation could capture other prospective benefits to waste management claimed for the full transmutation of the TRUs. First, transmuting the plutonium could reduce somewhat the calculated repository hazards from human intrusion scenarios in which waste is brought directly to the surface. The effect would depend on the time after repository closure at which the human intrusion event occurred. For postclosure times of a hundred years or so, 238Pu with an 87.7-year half-life is an important contributor to the TRU activity; this radionuclide would be transmuted. However, for postclosure times to several thousand years, the untransmuted 241Am with an 433-year half-life would dominate the TRU activity. This would be especially true for the older spent fuel in which the 241Pu had already decayed with a 14.4-year half-life to 241Am. For postclosure times of 10,000 years or more, 239Pu and 240Pu would dominate the TRU activity and would be transmuted. However, when the 239Pu is eliminated by transmutation, or has decayed, 237 Np would dominate the TRU activity. In addition, plutonium is quite insoluble in groundwater under planned repository conditions. Thus, plutonium transmutation would have little direct effect on the dissolution-and-migration scenarios important for long-term repository risk (except by eliminating one of the precursors of 231Pa). Also, removal of the plutonium would reduce the long-term heat load on the repository, although the untransmuted 241Am would still be a major contributor. For the Yucca Mountain repository, however, decay heat is being considered as a means of keeping the waste containers dry while increasing the capacity of the repository, so that transmuting the plutonium (or other TRUs) may actually be a disadvantage for that site. REDUCTION OF TRANSURANIC INVENTORIES Introduction This section examines the extent to which the amounts of TRUs in wastes from the various proposed transmuters can be made significantly smaller than the amount of TRUs in spent fuel in the reference once-through LWR fuel cycle. Inventories of TRUs in wastes from fuel reprocessing and TRU recycle are considered, together with inventories of untransmuted TRUs in the transmuter and in the associated facilities for reprocessing and fabrication of recycled material. To defray the cost of transmutation, all of the transmutation concepts propose generating and selling electrical energy resulting from the fission of TRUs. In this section the inventories of TRUs in the transmutation fuel cycle and in the reference once-through fuel cycle are compared for the same amount of electricity generated. Typically, only a small fraction of the TRUs in a transmuter would be transmuted while exposed to reactor neutrons during an irradiation cycle. The fuel discharged from the reactor would be reprocessed to remove fission products and the recovered TRUs would be recycled. Special techniques for reprocessing are specified in order to reduce the amount of TRUs lost to the reprocessing wastes. Process-loss fractions of the order of 0.001 to 0.000110 or less have 10    ANL is said to be developing pyrochemical separations with process-loss fractions of 0.001 to 0.0001 for each processing cycle. Developers of the TRUEX aqueous separation, to be used in conjunction with PUREX aqueous reprocessing, expect that process-loss fractions of 0.001 to 0.0001 can be achieved. LANL, the developer of the ATW, expects process-loss fractions of 0.0002 or better for neptunium and plutonium and to 0.000003 or better for americium and curium, sufficient so that the resulting TRU waste can be treated as low-level waste (Arthur, 1993).

OCR for page 49
Nuclear Wastes: Technologies for Separations and Transmutation been proposed. However, TRUs must be recycled many times through the reactor and its external fuel cycle before being transmuted, so the actual fraction lost to the reprocessing waste is greater than the fraction lost in each reprocessing cycle. Also, for safety and fuel-cycle economics, several of the transmutation concepts convert fertile 238U or 232Th to additional TRUs that must also be recycled and transmuted. Consequently, even with new separations that would reduce the losses to waste to 0.001 to 0.00001 of the amount processed, the inventories of TRUs in the waste would be far greater than what might be expected by multiplying the TRU inventories in LWR spent fuel by the process-loss fraction. For many of the transmuter concepts, such as the ALMR, the inventories of TRUs in the reactor and its external reprocessing-refabrication fuel cycle are large. In some transmuter concepts much of the TRUs supplied to the transmuter during its life would remain in the reactor and its fuel cycle. This residual inventory must be accounted for as potential high-level waste. Consequently, as long as nuclear power from transmuters is to be continued, the TRU inventory at the end of life of each transmuter would be transferred to a succeeding transmuter. To achieve the transmutation goals, it would be necessary to operate the transmuters long enough so that the above-ground inventory in the transmuter and its fuel cycle is no longer much larger in magnitude than the inventory in the total waste produced. Therefore, the time-dependent inventories of TRUs in the transmuter and its fuel cycle should be included when analyzing the extent to which transmutation can reduce TRU inventories below the inventories in the reference once-through LWR fuel cycle. For a given number of first-generation transmuters, constant electric power from transmuters would result if each succeeding transmuter were of the same electric power as the first-generation transmuter. Alternatively, each generation of follow-on nonbreeding transmuters11 could be of smaller electrical power as the total TRU inventory is consumed, resulting in the phase-out of nuclear power at a rate consistent with the necessary reduction in total TRUs that will become radioactive waste. For nonbreeding transmuters, additional LWRs could eventually be required to furnish make-up TRUs to fuel the transmuters, once the stockpile of stored LWR fuel has been exhausted. Calculation for the various transmutation concepts of the time-dependent ratio of the amount of TRUs in waste from the reference LWR once-through fuel cycle to the amount of TRUs in the transmutation fuel cycle is the subject of this section. The Transuranic (TRU) Ratio The following analysis (Hebel et al., 1978; Pigford and Choi, 1991) describes the extent of TRU inventory reduction for the various transmuter concepts as a function of time. It is important to use an index that measures the extent to which TRU inventories in the transmuter, in its waste, and in its fuel cycle would be smaller than the inventory of TRUs for the reference fuel cycle of once-through LWRs. This index is called the "transuranic (TRU) ratio" ψ(t), defined as total inventory of transuranics sent to waste disposal in time for the reference once-through LWR fuel cycle, if no fuel reprocessing, no recycle, and no transmutation ψ (t) = —————————————————————————— total inventory of transuranics at time t in the transmuter, in its fuel cycle, and in process wastes TRUs supplied to the transmuter system for start-up and for make-up fuel can be obtained by reprocessing existing LWR spent fuel and/or by reprocessing spent fuel from future LWRs. In calculating quantities for the equivalent reference fuel cycle, i.e., for the numerator of ψ(t), that same amount of TRUs must be assumed to go directly to waste disposal. Also, in the trnasmutation fuel cycle, transmuters generate a specified electric power P(t) and a specified total electrical energy by time t. To maintain equivalence, the TRU inventory in the nontransmutation fuel cycle considered in the numerator of ψ(t) must also include the TRUs that would be produced by reference LWRs producing the same electric power P(t) and the same electrical energy as the transmuters. All of these TRUs calculated for the non-transmutation fuel cycle are assumed to be sent to waste disposal. A more direct measure of the extent of depletion of TRUs by transmutation is the ratio of the total amount of TRUs supplied to the transmuter to the inventory of TRUs in the transmuter system and in its waste. This ratio is defined as the "depletion ratio" χ(t). It is lower in magnitude than the TRU ratio ψ(t). To be meaningful, however, the comparison of waste-disposal benefits for alternative electric power systems must be made on the basis of the same electric power and the same electric energy. Therefore, the transmutation concepts should have the benefit incorporated into the TRU ratio ψ(t). 11    A nonbreeding transmutor has a breeding ratio less than unity.

OCR for page 49
Nuclear Wastes: Technologies for Separations and Transmutation However, only limited reactivity swing is available for control by shutting down the accelerator, about 5% for the present ATW concepts. This report notes several possible reactivity transients not yet studied by LANL that may well require provision for additional reactivity control. Yet, none of the ATW concepts include control absorbers. Further study of the reactivity control issues by LANL may lead to system design changes with additional development and testing requirements. Indeed, for the ATW the possible need for control rods and the related reactivity control system raises questions about the benefit of using an accelerator and subcritical neutron-multiplying-moderating assembly, compared to a high-performance, just-critical thermal reactor with conventional reactivity control. Moreover, there is an economic penalty for an accelerator, which is substantially greater for the aqueous ATW-1 than for the other three ATW concepts. Problems due to materials degradation in the high neutron fluence and to the fluid fuels (summarized below) would be essentially the same, with or without the accelerator. Thus, LANL may find that eliminating the accelerator and beam target and going to just-critical reactor concepts could result in simpler core designs and more economical systems. The ATW concepts present materials degradation issues, largely unanalyzed at present, that are likely to require extensive materials development and testing (see Appendix F). Two major types of such degradation are (1) radiation damage in the target and blanket assembly structure, especially from neutrons; and (2) alteration of the mechanical properties and dimensional stability of structural materials caused by transmutants, in particular, by helium from alpha particle release. These issues present significant problems in conventional reactors and would be more severe under the ATW operating conditions. The subcritical reactor assembly and proton beam target for any of the four ATW concepts pose major engineering and materials development challenges (see Appendix F). The beam targets are to be cooled with molten lead (ATW-1) or molten lithium (ATW-2, -3, and -4) that pose substantial heat removal and materials issues under the anticipated conditions. In addition, the slurry fuel for ATW-1 and the molten salt fuel for the other ATW concepts raise significant materials issues not yet studied by LANL, as described in an earlier section of this chapter. Both the slurry fuel and molten salt concepts were the subject of development efforts two decades ago, and major materials problems associated with various fluid-materials interactions, including corrosion were found. In addition, the system performance depends critically on the high recovery separations-reprocessing system, many aspects of which have not yet been demonstrated beyond the laboratory scale. Extensive reactor, target-fuel, and fuel-cycle development and performance verification would be required for any of the ATW proposals before the chances of success could be determined. Given success in the development stage, both pilot-scale and full-scale test and demonstration would be required. ATW concepts (aqueous and nonaqueous) require the longest time to complete full-scale demonstration—an orderly program may take more than two decades through complete demonstration at a cost of perhaps $1 billion or more. They also carry a high development risk (see Appendix F). Other (Secondary) Transmutation Concepts The PBR and the Phoenix transmutation systems are still at the conceptual study and design stage. Like the ATW, these systems entail very high neutron flux levels and associated heat generation and nuclide destruction rates that are beyond the operational experience with conventional present-day reactors. Because only a scant amount of information to evaluate the two concepts in available, only a limited evaluation is presented here. PBR Transmutation System The PBR is a critical reactor concept with a very high thermal flux that could achieve rapid burn-up of TRUs and fission products, assuming a suitable fuel form could be developed. However, PBR is based on a large extrapolation of the technology being developed for a space reactor. PBR has a much higher fissile-specific power, power density in the core, and levels of radioactive decay than are typical of current thermal reactor designs. In fact, the PBR conceptual design has almost two orders of magnitude higher power density than a conventional LWR. The advanced PBR operating conditions raise a host of major materials and engineering issues, for the reactor itself and for target fuels (see Appendix F). The following list of PBR development challenges are illustrative: (1) a highly reliable helium coolant system for the packed particle bed, to cope with the extraordinary heat generation rate and core power density; (2) requirements for the Be2C moderator blocks and the structural materials to withstand fluences of the order of 1023 neutrons/cm2; (3) target fuel comprising TRUs and fission products embedded in pyrographite particles, reminiscent of HTGR fuel particles but much smaller in diameter, and able to withstand a limiting temperature of 2,200°C while retaining fission products (considerably higher than the 1,600°C limiting temperature for HTGR fuel); and (4) a modified PUREX-TRUEX process plus a lanthanide separations stage, with an electrodissolver at the front end to remove the carbon; a fluoride volatility process is being considered as an alternative, but that process has been plagued with problems in past applications.

OCR for page 49
Nuclear Wastes: Technologies for Separations and Transmutation Extensive reactor, target-fuel, and fuel-cycle development and performance verification would be needed before even the chances of success could be assessed. Because of the substantial amount of technology development, testing, and pilot-and full-scale demonstration required to meet its ambitious goals, the PBR concept has high development risk. An orderly program may require about two decades through full-scale demonstration at a comparatively high cost, totaling perhaps $1 billion or more (see Appendix F). Phoenix Transmutation System The Phoenix fast neutron transmutation concept has accelerator requirements much like those discussed above for the ATW. The accelerator current-energy requirements look achievable with further development, provided high performance and reliability can be attained in continuous operation. However, only limited reactivity swing is available for control of the subcritical reactor by shutting down the accelerator, about 2% in the present Phoenix concept, with no ancillary control absorbers. This report notes several possible reactivity transients not yet studied that may well require provision for additional reactivity control. This could lead to system design changes with additional development and testing requirements. Moreover, the Phoenix subcritical reactor would operate at conditions beyond the experience with present-day fast reactors. Thus, Phoenix presents significant materials degradation issues, unanalyzed at present, from radiation damage and from alteration of the mechanical properties and dimensional stability of structural materials caused by transmutants. These issues would require extensive materials development and testing (see Appendix F). Phoenix proposes to use the MOX fuel as the beam target, as distinct from the ATW concepts that use a separate target. The fuel would be similar to that tested in FFTF. However, there is neither analysis nor experimental results as yet to support the extension of the previous work with MOX fuel to the requirements for long fuel life while burning MAs in the Phoenix spectrum and flux levels. This also raises similar fuel qualification and performance verification issues as discussed above for LWR transmuter fuel, albeit with much lower levels of neutron emission in the Phoenix target fuel. In addition, cooling the Phoenix fuel with molten sodium raises significant materials issues under the anticipated conditions. Finally, the Phoenix proposal assumes a developed PUREX-TRUEX separations process similar to that discussed for the LWR transmuter but with a very high process decontamination factor, i.e., 105, stated as a goal by BNL. Like the ATW, extensive reactor, target-fuel, and fuel-cycle development and performance verification would be required before even the chances of success for Phoenix could be determined. Given success in the development stage, both pilot-and full-scale demonstrations would be required. Although Phoenix draws on oxide fuel experience, its development risks are high, and the time and cost requirements through complete full-scale demonstration may be comparable to developing one of the ATW concepts. TIME SCALES AND COSTS FOR MODEL SYSTEM DEPLOYMENTS Basis for Transmutation System Deployments This section summarizes deployments of selected S&T systems in terms of material quantities, costs, and time scales. For illustration, we discuss the scenarios of declining nuclear power and of constant nuclear power generation (see section on Reduction of Transuranic Inventories). Material Flows. Table 4-2 presents inventories of TRUs in the reactor and external fuel cycle for nuclear power plants operating at 1,395 MWe. It presents ratios and other data from which annual material flow quantities can be calculated. The inventories and flow quantities can be assumed to be proportional to rate of electrical generation. For example, a 1,395-MWe LWR operating as a TRU burner would have a transuranic inventory of 21,600 Mg and an annual discharge rate of 38 Mg HM/year. Reprocessing Costs. Appendix J analyzes cost information available on several foreign facilities that use a PUREX-like separations technology to reprocess LWR spent fuel, for a once-through fuel cycle. Results are summarized in Chapter 6. The facilities analyzed include the French UP3 plant, the British THORP plant, and the new Japanese facility under construction at Rokkashomura. These data are used to estimate capital and operating costs of a technically comparable facility of 900 Mg/yr nominal annual operating capacity, if constructed in the United States. The projected capital cost in 1992 dollars is shown in Table 4-5. The capital cost includes facilities for temporary storage of spent fuel and a facility to convert the HLW to a suitable form for geologic disposal. The operating cost is projected at $380 million/yr. The resulting 30-year levelized unit reprocessing cost for U.S. construction and financing is shown in Table 4-5. TABLE 4-5 Projected capital reporcessing costs in 1992 dollars Financing Capital cost, $ billion Unit cost, $/kg HM Government 6.2 800 Utility 6.7 1,300 Private 7.3 2,100

OCR for page 49
Nuclear Wastes: Technologies for Separations and Transmutation The above cost estimates are for a facility processing LWR discharge fuel from a once-through fuel cycle, with process losses characteristic of current PUREX technology. Additional technology, such as the TRUEX process, would be required to reduce process losses to the level of about 0.1%, as sought by DOE contractors for transmutation. For illustration, we adopt a 10% increase in the above costs for a high-recovery PUREX-TRUEX facility. Costs for reprocessing MOX fuel are certain to be much higher, but we have no estimates of the magnitude. Consequently, we adopt the above costs increased by 10%. The reference reprocessing facility with a throughput of 900 Mg HM/year could reprocess spent fuel discharged from 24 1395-MWe LWRs. The total cost to reprocess the 62,000 Mg HM of spent LWR fuel now destined for Yucca Mountain would be from $50 billion to $130 billion for a U.S. reprocessing facility, depending on the method of financing. In addition to the reprocessing of the spent fuel, reprocessing of the fuel discharged from the transmuting reactor would be required. The amount of such fuel being reprocessed is about a factor of 10 less than the LWR spent fuel. Further, all the LWR fuel would have to be reprocessed by the end of the first 30-year cycle. The processing of the transmuting reactor fuel is required over a much longer time scale. If we discount the cost of this reprocessing of transmuting reactor fuel, its cost becomes small compared to the cost of reprocessing the spent LWR fuel. For the ATW, LANL plans fluid fuel systems close-coupled to on-line reprocessing plants. The proposed technology for the ATW-1 is a slurry fuel of small TRU particles suspended in a heavy-water carrier, circulating in small pressurized tubes inside the reactor. For the ATW-2, LANL proposes a molten salt carrier with dissolved TRU radionuclides, again circulating in tubes. However, the reprocessing system is not yet defined for either version—indeed, major technical development is required—so no reliable basis exists to estimate ATW reprocessing costs. Reactor Costs. The approach to reactor costs is adapted from the National Research Council report, Nuclear Power (1992). This report evaluates capital costs for current LWRs and describes the Electric Power Research Institute (EPRI) goals for capital cost and operating costs for advanced LWRs of 1.2- to 1.3- GWe capacity and passively safe LWRs of 600 MWe capacity (see National Research Council, 1992: Table 3-2, p. 94). The goal for the advanced 1.2- GWe LWR is an overnight capital cost of $1,300 per rated kWe, with goals for fixed and incremental operating cost of $61.1/kWe-yr and 0.11 cents/kWh, respectively. For the 600-MWe passively safe LWR, the corresponding goals are $1,475 per rated kWe, $72.7/kWe-yr and 0.15 cents/kWh, respectively. In addition, the report lists cost goals for ALMR-type reactors based on an estimate by EPRI (see National Research Council, 1992: Table 3-4, p. 139). The overnight capital cost goal is $1,725 per rated kWe with fixed and incremental operating cost goals of $75.5/kWe-yr and 0.15 cents/kWh, respectively. For the three types of reactors, the report quotes an EPRI uncertainty estimate of -30 to +80 percent, presumably with greater uncertainty for the ALMR than for the LWRs.22 There is no reliable basis as yet for estimating the costs of any of the ATW transmutors. Strictly speaking, these costs for reactors and the costs of reprocessing need to be put on the same basis. In practice, this would not only be difficult but pointless in view of the large uncertainty in the costs of reprocessing (see Chapter 6). Thus, the distinction is ignored for the costs of the model deployments. For an advanced LWR this evaluation uses a 30-year cost of $3.36 billion, comprising $1.30 billion in capital cost, $1.83 billion in fixed operating cost, and $0.23 billion in incremental operating cost (assuming 80% capacity factor). Similarly, for a 1-GWe ALMR, this section uses a 30-year cost of $4.31 billion, comprising $1.725 billion in capital cost, $2.265 billion in fixed operating cost, and $0.316 billion in incremental operating cost. Utility financing would entail additional cost. Deployments for Declining Nuclear Power Reprocessing Requirements and Deployment Rate The deployments for declining nuclear power are designed to transmute the 612 Mg of TRUs in the 62,000-Mg HM stockpile of spent LWR fuel. This section compares deployments for the LWR, ALMR (β = 0.65), and the ATW systems. For a rapid reduction in total transuranic inventory, the ALMR with β = 0.98 is not a suitable candidate. The first-generation transmutors are to operate at constant power for 30 years. There are to be a sufficient number so that all of the 612 Mg of TRU in the 62,000 Mg of LWR spent fuel will be used for start-up inventory and for TRU make-up during the first generation. The reactor and fuel- 22    The National Research Council Energy Engineering Board (1992: Table 2-4) lists the overnight construction costs in 1988 dollars for selected U.S. nuclear power plants, by year of commercial operation. For 10 LWRs that came into operation during 1987-1988 at an average capacity of 1,070 MWe, the cost per kWe varied from a high of $4,596 to a low of $1,342, with an average of $3,133. In the previous period of 1985-1986, 15 plants came into operation with similar high and low values and an average of $2,620 per kWe. The wide variance in cost in a given time period and the high average cost (in constant dollars) compared to the 1970s are presumably due to the time and cost to meet changing regulatory requirements following the accident at Three-Mile Island, together with the high interest rates of the early and mid-1980 period. This evaluation assumes that such factors do not continue for future reactors.

OCR for page 49
Nuclear Wastes: Technologies for Separations and Transmutation cycle inventory remaining at the end of the first generation is to be used to start and fuel a smaller number of second-generation transmutors. Similar reductions in number of transmutors are made in each subsequent generation until the desired net reduction in transuranic inventory is achieved. Using the data in Table 4-2, an LWR transmutor generating 1.395 GWe requires a start-up inventory of 21.6 Mg of TRUs and a TRU makeup rate of 0.350 Mg/yr. A total of 21.6 such first-generation transmutors would use all of the 612-Mg TRU from the stockpile during a 30-year operating life. To supply the yearly makeup of TRU, 1,097 Mg of LWR spent fuel from the stockpile must be reprocessed TABLE 4-6 Calculated Number of Transmutors and Reprocessing Rates, to utilize Total TRU in 62,000 Mg of LWR Spent Fuela TRANSMUTOR GENERATION AND MODE OF OPERATION NUMBER OF TRANSMUTORSb NUMBER OF LWRsc TOTAL SYSTEM ELECTRICAL POWER, GWe FRACTION OF INITIAL TRU INVENTORY REMAININGd LWR SPENT FUEL REPROCESSING RATE, Mg/yre TRANSMUTOR RECYCLE REPROCESSING RATE, Mg TRU/yrf Start-up Make-up LWR First Generation 13.3   18.6 0.47 5,820 1,097 46.5 Declining Power, Second Generation 6.2   8.7 0.22     21.9 Continuous Power, Second Generation 7.6 5.7 18.6 0.27   208 26.5 ALMR (β=0.65) First Generation 21.8   30.4 0.50 6,180 1,040 69.9 Declining Power, Second Generation 10.8   15.1 0.25     34.7 Continuous Power, Second Generation 13.9 7.9 30.4 0.32   287 44.6 ATW-1 AQUEOUS First Generation 7.32   10.2 0.042 2,300 230 780 Declining Power, Second Generation 1.7   2.4 0.0018     107 Continuous Power, Second Generation 2.0 5.3 10.2 0.052   192 217 ATW-2 NON-AQUEOUS First Generation 16.3   22.7 0.016 1.98 2,030 117 Declining Power, Second Generation 0.26   0.36 0.0002     1.90 Continuous Power, Second Generation 3.7 12.6 22.7 0.0036   454 27.7 a Contains 612 Mg of transuranics (TRU). b Each transmutor operates for 30 years at 1.395 GWe (see Table 4-2). c The number of LWRs required to supply TRU make-up at constant total electric power. Each LWR operates at 1.395 GWe. d Ratio of the TRU inventory in the transmutors and fuel cycles, at the end of the generation, to the original 612 Mg of TRU. e The rate is calculated for total ''heavy metal" (HM). A 5-year reprocessing campaign is assumed for the start-up inventory. f Rate at which TRU in spent fuel is discharged from transmutors, Mg TRU/yr. yearly. Reprocessing MOX spent fuel discharged from the transmutors would require an additional reprocessing capacity of 46.5 Mg HM/yr. Technology for reprocessing multiply recycled MOX fuel has not been implemented commercially. However, supplying the start-up inventory places a much greater demand on reprocessing capacity. A total of 29,100 Mg of spent fuel must be reprocessed. This would take 32 years for a single 900-Mg/yr reprocessing facility. Seven such reprocessing plants could reduce the reprocessing time to less than five years. These results are shown in Table 4-6. By the end of the first generation, 53% of the initial transuranics will have been consumed. The remaining 287

OCR for page 49
Nuclear Wastes: Technologies for Separations and Transmutation Mg of transuranics in reactor and fuel-cycle inventory could be used to start and fuel 6.2 second-generation transmutors. No further LWR-fuel reprocessing would be required in this declining power scenario. In this manner, each subsequent generation would consume 53% of the TRUs remaining after the previous generation. Finally, after a little more than seven generations and after more than 180 years, about 1% of the original transuranic inventory would remain. Subsequent transmutation would require only a portion of the one reference transmuting reactor or smaller reactors. The unit cost of reprocessing will be increasingly expensive during later generations because of the smaller throughput. An ALMR (breeding ratio = 0.65) generating 1.395 GWe requires a start-up inventory of 14 Mg of transuranics and a yearly make-up of 0.47 Mg/yr. Twenty-two such transmutors could utilize the entire stockpile inventory of 612 Mg TRU during their 30-year operating life. About half of the original stockpile inventory of TRU would remain. To supply the yearly makeup would require reprocessing 1040 Mg/yr of LWR spent fuel. To supply the start-up inventory, 30,900 Mg of LWR spent fuel would be reprocessed, requiring 3.4 plants, each with capacity of 900 Mg/yr of LWR spent fuel. Reprocessing ALMR spent fuel would require a special facility capable of reprocessing fast-reactor spent fuel, with an annual total capacity of about 200 Mg of ALMR spent fuel. This is less than the required capacity for a LWR transmutor because of the higher enrichment of fast-reactor spent fuel. ANL proposes to construct facilities for reprocessing and recycling ALMR spent fuel that are located at each reactor site. Each such facility would have an annual throughput of about 10 to 20 Mg of ALMR spent fuel, depending on the number of transmutors at each site. Second-generation transmutors would consist of 10.8 similar ALMRs. The nonaqueous ATW generating 1.395 GWe has much lower transuranic inventory than the LWR and ALMR of the same generating capacity (cf. Table 4-2). The transuranic make-up rate of 1.23 Mg/yr for each transmutor, for the nonaqueous ATW without thorium, is much higher than for the LWR and ALMR, because this ATW has no internal breeding of fissile material. A much higher fuel-cycle cost can be expected. During a 30-year operating life a total of 16.3 such ATW's could utilize all of the 612 Mg of TRU in the LWR spent fuel stockpile. A total of 2030 Mg of LWR spent fuel would be reprocessed yearly to supply the make-up TRU. The reprocessing capacity to supply the start-up TRU would be more modest. It could be obtained by operating the 2,030 Mg/yr facility for about 6 months. Each of the ATW's would have its own integrated on-line reprocessing facility, capable of reprocessing 7.2 Mg/yr of transuranics, with a total capacity of 118 Mg TRU/yr for all transmutors. At the end of the first generation of transmutors only 1.6% of the original TRU inventory will remain. For the second generation a single ATW scaled down to 360 GWe would suffice. The performance of the aqueous ATW is more modest. The large start-up inventory of 15.5 Mg of TRU for a 1.395-GWe transmutor is a consequence of the very low thermal efficiency. The transuranic make-up rate is 2.27 Mg TRU/yr for each transmutor. During a 30-year operating life a total of 7.3 such transmutors could utilize all of the 612 Mg of TRU in the LWR spent-fuel stockpile. The TRU make-up would require reprocessing 230 Mg/yr of LWR spent fuel. A total of 11,500 Mg of LWR spent fuel would be reprocessed to supply the start-up inventory. To process this in 5 years would require 2.6 reprocessing facilities, each of capacity 900 Mg/yr of LWR spent fuel. At the end of 30 years, 23 percent of the initial TRU inventory would remain. This could start and fuel 1.7 similar transmutors for the second generation. Deployments for Steady Nuclear Power Generation Using transmutors in a system of constant nuclear power generation has been described earlier (cf. "Reduction in Transuranic Inventory"). First, we summarize the material quantities based on a system of ALMR transmutors, with breeding ratio of 0.65, and once-through LWR reactors to supply the TRU make-up for the transmutors after the stockpile of LWR discharge fuel is consumed. This system could begin with first-generation transmutors in the same way as the first-generation transmutors for the declining power scenario. The parameters for that first generation are given above. In the second generation, rather than decreasing the total nuclear power as in the declining power scenario, additional ALMRs and LWRs will be installed so that the total electrical power is unchanged from the first generation. The ratio of additional ALMRs to additional LWRs will be calculated so that no additional TRU will be needed from an outside source. Using the data in Table 4-2, we calculate that in the second generation there will be 13.9 ALMRs and 7.9 LWRs, each generating 1.395 GWe. The reprocessing requirements for the first generation would be the same as calculated for ALMRs in declining power. During the second generation, newly created LWR discharge fuel must be reprocessed at a rate of 287 Mg HM/yr, a rate far below current commercial scale of 900 Mg HM/yr. At the end of the second generation the remaining TRU inventory will be 194 Mg, which is 36% of the inventory remaining at the end of the first generation. Thus, continuing at constant power reduces the burnup fraction per generation well below the 50% burnup per generation that can be achieved in declining power. In each subsequent generation the ratio of LWR power to ALMR power increases. It reaches an asymptotic value

OCR for page 49
Nuclear Wastes: Technologies for Separations and Transmutation of 1.3. At this ratio, LWR spent fuel would be reprocessed at a rate of 450 Mg/yr. The reprocessing requirements for the first-generation ATWs have been described in the previous section. During the second generation newly created LWR discharge fuel must be reprocessed at a rate of 192 Mg HM/yr. The asymptotic ratio of LWR power to aqueous ATW power is 6.3, corresponding to a reprocessing rate of 230 Mg HM/yr of LWR spent fuel. Similarly, for the second-generation nonaqueous ATW transmutors, 454 Mg HM/yr of spent fuel from 17.4 GWe of LWR's must be reprocessed to fuel the transmutors. The asymptotic ratio of LWR power to nonaqueous ATW power is 3.42, corresponding to a reprocessing rate of 192 Mg HM/yr of LWR spent fuel. The aqueous ATW transmutor can transmute transuranics at almost twice the rate of that of the nonaqueous ATW, both at the same electrical power, because of the much lower thermal efficiency of the aqueous ATW. Table 4-6 lists characteristics for second-generation systems of various transmutors and once-through LWRS, with total system power the same as in the first generation. COMPARATIVE THERMAL AND ELECTRICAL EFFICIENCIES All of the transmutation proposals except the particle bed reactor presently envision producing and selling electrical energy as a means of paying for the cost of the transmutation system. That cost includes the capital cost of the transmutor facility, the cost of reprocessing LWR spent fuel to be transmuted, and the cost of reprocessing and refabricating the transmutor target fuel itself. Setting aside Phoenix, which is a hybrid concept using LWRs to burn the plutonium, it is useful to compare the net thermal-to-electrical efficiencies of the various transmutation concepts assuming that power is delivered to the electrical grid to offset part of the cost of the system. In the present LANL design for the aqueous ATW-1 concept, in which the accelerator consumes a significant fraction of the electrical energy produced, the net thermal-to-electrical efficiency is only 19.3%. This compares to 33-36% for the nonaqueous ATW concepts and to a similar range of efficiencies for the LWR and the ALMR/IFR concepts. If the ATW-1 concept were to operate as a just-critical reactor without an accelerator, the net thermal efficiency would be about 30%, 50% higher than for the accelerator-driven ATW-1 concept. From discussions with ATW designers, it appears that the aqueous ATW-1 as proposed could be modified by eliminating the accelerator and increasing the effective neutron multiplication factor from 0.95 to 1.0 to achieve criticality (see Appendix F). One option is to add another blanket module. Another is to increase the density of TRUs in the slurry. Yet, to achieve the 19.3% efficiency of the present ATW-1 design, LANL has already increased the fluid density and pressure to the maximum possible extent; further increase of efficiency may not be practically achievable. For a rough estimate of the savings in total cost, Appendix F considers a campaign to process and transmute all the TRUs in the 62,000 MgHM of LWR spent fuel accumulated by the year 2011, otherwise destined for a geologic repository. Assuming that a just-critical reactor for the ATW-1 would cost roughly the same as the subcritical reactor plus target assembly in the present concept, and that electrical energy can be sold to the grid for about $0.02 per kWe, the cost differential (savings) in changing from an accelerator-driven to just-critical ATW-1 would be about $2.8 billion for the accumulated LWR spent fuel. Additional savings would accrue from not having the accelerator. Clearly, the ATW-1 pays heavily for using an accelerator, if the sale of electrical energy is expected to contribute significantly to funding the construction and operation. To realize the potential advantages of using aqueous reprocessing technology, which is much more highly developed than the technology used in the nonaqueous ATW, the features of a just-critical aqueous ATW-1 without accelerator should be carefully examined by LANL. PRINCIPAL FINDINGS AND CONCLUSIONS 1. The S&T of TRUs and certain long-lived fission products in spent reactor fuel is technically feasible and could, in principle, provide benefits to radioactive waste disposal in a geologic repository. However, to begin to have a significant benefit for waste disposal, an entire S&T system consisting of many facilities would have to operate in a highly integrated manner from several decades to hundreds of years. The deployment of an S&T system that is extensive enough to have a significant effect on the disposition of the accumulated LWR spent fuel would require many tens to hundreds of billions of dollars and take several decades to implement. 1a. An S&T system using reactors having a thermal neutron spectrum, such as the LWR or ATW, could transmute the 99Tc and 129I in spent fuel. This in turn could reduce the calculated long-term radiological hazard of a repository in waste dissolution-and-migration scenarios. The ALMR, with its fast neutron spectrum, would not be suitable for such transmutation. As an alternative, the technetium and iodine could be separated and packaged in low-solubility waste forms for geologic disposal. Separation and transmutation of 99Tc and 129I would not be feasible unless a full-scale reprocessing program at a commercial level were in place.

OCR for page 49
Nuclear Wastes: Technologies for Separations and Transmutation 1b. Transmutation of the TRUs could be accomplished in an LWR, an ALMR, or an ATW, assuming the latter system concept proves to be technically feasible (see 3c below). Such transmutation is one of several means (some of which do not involve S&T, of increasing the waste disposal capacity of a repository, deferring the need for a second repository (see Chapter 6). Reduction of the TRUs could also reduce the calculated radiological hazard of human intrusion scenarios in which nuclear waste is brought directly to the surface. However, transmutation of the TRUs would have little effect on the calculated long-term hazards in waste dissolution-and-migration scenarios, except possibly for 237Np in some scenarios. 1c. Alternatively, the spent LWR fuel could be reprocessed and the plutonium recycled to LWRs or ALMRs (see 2h below). Compared to the once-through LWR fuel cycle, the main benefit to waste management would be the use of improved, low-loss waste forms for the separated MAs and other key constituents, such as 14C, 129I, and 99Tc. This assumes that the improved waste forms could be developed and shown to be of greater integrity to spent fuel under the regulatory requirements for geologic disposal. In principle, transmuting the plutonium would reduce somewhat the radiological risks of the human intrusion scenarios in which nuclear waste is brought directly to the surface. The untransmuted 433-year 241Am, however, would dominate the TRU activity for postclosure times from a few hundred years to several thousand years. Plutonium transmutation, therefore, may have limited impact on repository risks. 1d. The initial commitment to reprocess the 63,000 MgHM stockpile of LWR spent fuel accumulated by 2011—which would require a reprocessing throughput of 2,100 MgHM spent-fuel per year over 30 years—would require an early decision to organize and carry out the development and demonstration of a prototype S&T system. Such a system would comprise many interdependent components, including waste transmutation reactors, spent-fuel reprocessing plants, recycle fuel fabrication plants, facilities to package the residual waste for ultimate disposal, and mechanisms for transportation between the facilities. Successful demonstration would be followed by the construction and operation (including maintenance and retirement) of tens of reactors and their associated fuel-cycle facilities, entailing an ongoing commitment for several generations (about 30 years each) of facilities. 1e. Merely developing, building, and operating the individual components of the system would give little or no benefit. To have a real effect, an entire system of many facilities would be needed in which all the components operate with high reliability in a synchronized fashion for many decades or centuries (see 2 below). System viability could be maintained only if the right facilities were built and put into operation at the right times. The magnitude of the concerted effort and the institutional complexity (involving long-term linkages among many private and governmental organizations) are comparable to large military initiatives that endure for much shorter periods than would be required for an S&T system. 1f. An estimate can be made of the time scale and the cost to deploy an S&T system based on LWRs or ALMRs, assuming that development can be completed (see 3 below) and that favorable institutional arrangements can be achieved. The licensing, construction, and initial operation of an S&T system of sufficient scale to begin to affect spent-fuel emplacement in a geologic repository would require one to two decades after a system feasibility demonstration and an expenditure of $20 to $40 billion beyond the costs of development and demonstration. 1g. Additional time and a much larger investment of funds would be necessary for an S&T system of sufficient scale to reduce repository hazards significantly or to affect the need for a second repository. For example, to reprocess the 62,000 MgHM of LWR spent fuel and transmute its 612 Mg of TRUs under a declining nuclear power (phase-out) scenario, an ALMR/IFR system operating with a 0.65 breeding ratio or an LWR system with aqueous reprocessing would cost some $500 billion and require approximately 150 years to accomplish the transmutation (see 2d and 2e below for details). Such a system, whether federally or privately owned, would not be economical in the United States at the current costs of uranium ore without a large subsidy (between $30 to $100 billion). For the same declining nuclear power scenario, an ATW deployment would require about 50 years for the transmutation (see 2f below for details). Information is insufficient to reliably estimate the cost of ATW deployments. 2. The proposed S&T systems require decades to centuries to achieve a significant net reduction in the total TRU inventory relative to that of a once-through LWR fuel cycle of the same electrical production capacity. Comparing S&T systems using LWRs, ALMRs, and ATWs of the same electrical production capacity, the ATW projects the highest TRU consumption rates. TRU consumption rates for an ALMR increase significantly as its breeding ratio is lowered to 0.65 from the conventional value near unity. The LWR would have a net TRU consumption rate similar to an ALMR of 0.65 breeding ratio. As an alternative to S&T of all TRUs, reprocessing the LWR spent fuel and recycling only the plutonium to LWRs or ALMRs—using improved low-loss waste forms for the other HLW constituents—

OCR for page 49
Nuclear Wastes: Technologies for Separations and Transmutation would be simpler and somewhat less expensive to implement (see 1c above). 2a. With an S&T system, each succeeding generation would transmute the TRUs left over from the previous generation plus make-up. In a continuing nuclear power economy with steady electrical production, the total TRU inventory (i.e., in the transmutors, fuel-cycle facilities, and waste) would gradually increase due to process losses. In contrast, with declining nuclear power the total inventory of TRUs in the S&T system could be reduced if each succeeding generation operated with less total transmutor power (i.e., fewer transmutors and/or lower power per transmutor). For example, each generation over its lifetime (around 30 years) could transmute the inventory remaining from the previous generation until insufficient TRUs remained to fuel one transmutor for its lifetime. The TRUs left over at the end would require disposal with other HLW. 2b. The ALMR transmutor has been proposed to reduce the amount of TRUs in waste below the limiting amount that could be released to the environment according to 40CFR191, the EPA environmental protection standard for management and disposal of spent nuclear fuel, high-level and TRU wastes.23 For 239Pu, the most abundant radionuclide to be transmuted, the corresponding amount in the waste would be less than 0.03% of that in LWR spent fuel. As a result of multiple recycle in the transmutation systems, process losses to waste in each processing cycle would have to be less than about 0.007%. However, the ALMR/IRF project has not defined a development goal to attain such low process losses. 2c. During the entire life (ca. 30 years) of an ALMR transmutor, the amount of TRUs transmuted would be about half of the nearly constant inventory of TRUs in the reactor and its fuel cycle. To obtain further transmutation, the residual inventory would be transferred to subsequent generations of ALMR transmutors, as outlined in 2a. However, for ALMRs with a 0.65 breeding ratio and fueled with TRUs from LWRs (as proposed by DOE) the constant-power operating time required to reduce the inventory of residual TRUs to even 1% of the inventory of the referenced LWR once-through fuel cycle would be unrealistically long, on the order of many millennia (see 2b and Figure 4-3). The first century of constant-power transmutation could only reduce the inventory fraction to about 14%, too great to meet the ALMR's waste-disposal objective stated in 2b. More extensive transmutation would be required. A break-even, breeding ALMR would require even longer operation. 2d. In the declining nuclear power scenario outlined in 2a, the ALMR with a 0.65 breeding ratio could achieve a TRU "inventory fraction" as low as 9% in 100 years.24 However, the lowest fraction attainable would be about 0.5%, corresponding to final operation of one module of a nine-module ALMR. The residual inventory would be too great to meet the ALMR's goal of waste reduction (see 2b) unless smaller terminal ALMRs could be justified. 2e. To achieve the same goals for meeting the EPA standard 40CFR191 as in 2b, the LWR transmutor could allow slightly greater process losses of plutonium and neptunium per cycle and would require slightly less time than the ALMR with a 0.65 breeding ratio. However, a lower per-cycle processing loss of americium and curium would be required. In the declining nuclear power scenario outlined in 2a, an S&T system using the LWR transmutor could achieve a similar inventory reduction in 100 years as that achieved using the ALMR of 0.65 breeding ratio (see 2d). 2f. Even though the ATW expects to achieve very low per-cycle process losses to waste, the frequent processing of the fluid fuel results in inventories of plutonium and neptunium in the ATW waste as high as 0.4% of that in the spent-fuel waste of the reference LWR once-through fuel cycle. For hypothetical constant-power operation over extended time it would take about 1,400 years for the inventories of TRUs in the aqueous ATW reactor to equal those in the ATW waste. Declining nuclear power operation (phase-out) would take about 90 years. For the nonaqueous ATW the corresponding times would be 350 years at constant power and about 50 years with declining power. This is shorter than for the ALMRs and LWR because of the very high fissile-specific power of the ATW concepts. 2g. The inventories of residual plutonium, neptunium, americium, and curium in the ATW reactor systems would require disposal in a geologic repository, as would other ATW waste such as Zircaloy pressure tubes from the aqueous ATW. It would be impossible for ATWs to effectively eliminate all long-lived radionuclides so that important residual radioactivity would persist "no longer than a human lifetime," as has been asserted by the ATW proponents. 23   Although the EPA 40CFR191 standard no longer applies to the proposed Yucca Mountain repository, it is possible that such release limits, or the equivalent thereof, could still become part of the new standard. Also, the ALMR/IFR project makes the assumption that EPA release limits applicable to LWR spent fuel could be used as such for HLW generated in transmutation. 24   The "inventory fraction" is the ratio of the TRU inventory in the transmutor, fuel cycle, and waste to that in the spent fuel of a reference LWR once-through fuel cycle of the same electrical production.

OCR for page 49
Nuclear Wastes: Technologies for Separations and Transmutation     2h. A plutonium-burning LWR is a simple and potentially useful transmutor system that could be implemented with present technology to decrease plutonium build up (it does not eliminate plutonium). Recovery of 14C in PUREX reprocessing and qualification of new waste forms for separated 14C and 129I would still be required (see 1c). TRUs other than 239Pu and 241Pu would go to the waste. The times to reduce overall TRU inventories relative to the once-through LWR fuel cycle would be comparable to, but slightly shorter than, the times for the nonbreeding ALMR. During the first few hundred years of transmutor operation, the extent of TRU reduction would be little affected by the higher process losses of existing PUREX separations. Such reprocessing and the fabrication of recycle fuel would be simpler and somewhat less expensive than for an LWR transmutor with full recycle of all TRUs. However, reprocessing of even multiple-recycled plutonium, without recycling other radioelements, would reduce higher concentrations of transplutonics in discharge fuel and would make reprocessing difficult. At the present price of natural uranium, the cost of new U.S. facilities to reprocess LWR spent fuel and to fabricate recycled plutonium as MOX fuel would be prohibitive and the benefit to waste disposal doubtful. 2i. For a given electrical production capacity, some concepts for transmuting TRUs require less fresh uranium than the base-case LWR once-through fuel cycle. However, the transmutation systems that achieve the highest net consumption rate of TRUs also have a significant need for fissile make-up. Thus, their operation would still require an appreciable amount of fresh uranium. Developing and deploying TRU-burning ALMRs could lead eventually to ALMRs operating as self-sustained breeder reactors requiring no fresh uranium ore, but with much lower net consumption rate of TRUs. 3. The S&T systems differ widely in their state of technological maturity and present a broad spectrum of development issues, risks, costs, and schedules. The most mature system concept for transmuting TRUs, based on the use of LWRs, needs fuel-cycle development and would require about a decade and significant financial resources to reach the point of deployment. Compared with the LWR system, an ALMR/IFR system for transmuting TRUs would require more financial resources and take longer (perhaps a decade and a half) to reach the point of deployment. Beyond this, the ATW concepts would require major development before even the chances of success can be realistically assessed. An LWR- or ALMR-based system would require less development for transmuting only the plutonium from reprocessed spent fuel than for full TRU recycle. 3a. The most mature system approach would use the LWR together with aqueous reprocessing. Current LWRs would be suitable although advanced LWRs could offer cost and safety advantages when licensed. This thermal reactor system has the shortest time and lowest cost to complete development and full-scale demonstration of technical performance and system costs, about 8 to 10 years at a level of effort estimated at perhaps $50 to $100 million per year, exclusive of the cost of construction of any major test facilities. It also has the highest chance of successful development. Significant issues for reprocessing and for recycle fuel are posed by the multiple recycle of TRUs, in which higher MAs would tend to build up due to the LWR's thermal neutron spectrum. For transmuting actinides and selected fission products in addition to plutonium, an LWR system needs (1) further development, scale-up, and cost assessment of improved aqueous separations technology; (2) refinement of reactivity control for operation with TRUs; and (3) technology and procedures for licensable quality assurance and performance verification of refabricated recycle fuel containing significant amounts of MAs. For recycle of plutonium only, the required development in the three areas would be significantly simplified and could be accomplished sooner at lower cost than for full TRU recycle. 3b. The ALMR has been under development for decades as a fast breeder reactor based on plutonium recycle. However, it could be modified to operate as a TRU transmutor. No transmutation of fission products is envisaged with an ALMR. For burning TRUs, DOE proposes an ALMR/IFR with integral pyroprocessing of its spent fuel, which has not yet been demonstrated at pilot scale. The EBR-II fast reactor operating with a breeding ratio near unity has demonstrated a capability for passive safe shutdown. However, considerable experimental and calculational effort at breeding ratios of 0.65 or lower would be necessary to develop a database to support licensing with a reactor core that is fueled for full recycle (as proposed by the ALMR/IFR project) and contains higher-mass TRUs and fission products such as the lanthanides. However, the fast neutron spectrum of the ALMR would generate much fewer higher-mass actinides than the thermal reactors. Compared with an LWR and an ATW, an ALMR/IFR has an intermediate time and cost to complete development and full-scale demonstration of technical performance and system costs (about 15 to 20 years at a level of effort estimated at perhaps $100 to $150 million per year, exclusive of the cost of construction of any major test facilities) and an intermediate chance of successful development. Whether for full TRU recycle or plutonium-only recycle from spent LWR fuel, a low threshold cost for spent LWR fuel reprocessing is required for a viable ALMR/IFR

OCR for page 49
Nuclear Wastes: Technologies for Separations and Transmutation     burner system. Thus, its chance of successful development is lower than that of an ALMR system optimized for power production. 3c. The ATW concepts employ a subcritical thermal reactor using fluid fuel with integrated reprocessing. The proposals have high performance objectives for burning TRUs and key fission products. The ATW systems pose major engineering and materials challenges due to the extraordinary operating conditions in the reactor and beam target. The high power proton accelerator, which provides spallation neutrons for multiplication by the subcritical assembly, may be the best understood part of the system. Heat removal is a major issue for development as well as for safety and licensing. Also, the possibility of reactor transients, unevaluated in any detail at present, may require means of reactivity control beyond that afforded by merely turning off the accelerator. Moreover, the concepts use fluid fuels that involve unproved technologies for fuel fabrication and reprocessing, raising a host of operational and safety issues. The onsite reprocessing system has unique requirements for high reliability and maintainability, but a detailed concept has not yet been completed. The high thermal flux of the ATW would produce the greatest fraction of higher mass actinides of any of the principal concepts evaluated. Indeed, the high level of alpha and neutron radioactivity during onsite reprocessing and recycle-fuel circulation poses severe problems for the aqueous-based ATW-1 concept. For all the ATW concepts, the overall system economics are uncertain and are more sensitive than the other primary S&T concepts to the economics of feed material from reprocessing LWR spent fuel. Thus, the ATW concepts are in a qualitatively different position than the ALMR- and LWR-based concepts—extensive research and development would be required even to ascertain whether an ATW is technically feasible. For these reasons, the committee regards the ATW as a much less certain approach, with a questionable chance of successful development. If feasible, the ATW may require more than 20 years of high cost for development and full-scale demonstration of technical performance and system costs. REFERENCES Arthur, E. D. 1992a. Summary, Issues, and Development Plan. The Los Alamos Accelerator Transmutation of Nuclear Waste (ATW) Concept. ATW-92-60. Los Alamos, N. Mex.: Los Alamos National Laboratory. Arthur, E. D. 1992b. Nuclear Systems Requirements for Long-Lived Radionuclide Transmutation. LA-UR-92-2542. Los Alamos, N. Mex.: Los Alamos National Laboratory. Baetsle, L. H. 1993. Role and influence of partitioning and transmutation on the management of nuclear waste systems. Pp. 1235-1241 in Proceedings of the Symposium on Waste Management, Tucson, Ariz., February 28-March 4, 1993. R.G. Post, ed. Tucson: Arizona Board of Regents. Benedict, M., T. H. Pigford, and H. W. Levi. 1981. Nuclear Chemical Engineering, 2nd ed. New York: McGraw Hill. Bowman, C. D., E. D. Arthur, P. W. Lisowski, G. P. Lawrence, R. J. Jansen, J. L. Anderson, B. Blind, M. Cappiello, J. W. Davidson, T. R. England, L. N. Engel, R. C. Haight, H. G. Hughes, J. R. Ireland, R. A. Krakowski, R. J. La Bauve, B. C. Letellier, R. T. Perry, G. J. Russell, K. P. Staudhammer, and W. B. Wilson. 1991. Nuclear Energy Generation and Waste Transmutation Using an Acceleration Driven Intense Thermal Neutron Source. LA-UR-91-2601. Los Alamos, N. Mex.: Los Alamos National Laboratory. Chang, Y. I. 1991a. Actinide recycle potential in the IFR. Paper presented at STATS Subcommittee on Transmutation visit to University of California, Berkeley, November 1992. Chang, Y. I. 1991b. The integral fast reactor program. Paper presented to the Board on Radioactive Waste Management, March 1991. Croff, A. G., D. W. Tedder, J. P. Drago, J. O. Blomeke, and J. J. Perona. 1977. A Preliminary Assessment of Partitioning and Transmutation as a Radioactive Waste Management Concept. ORNL/TM-5808. Oak Ridge, Tenn.: Oak Ridge National Laboratory. Croff, A. G. 1980. ORIGEN2: A Revised and Updated Version of the Oak Ridge Isotope Generation and Depletion Code. ORNL-5621. Oak Ridge, Tenn.: Oak Ridge National Laboratory. Davidson, J. W. 1992. Adaptation of the Pigford-Choi Transuranic Actinide Reduction Model to an ATW Transmutation System. LAUR-92-1789. Los Alamos, N. Mex.: Los Alamos National Laboratory. Gorrell, T. C. 1979. Transmutation of Waste Actinides in LWRs. Report DP-1518. Aiken, S.C.: Savannah River Site. Hebel, L. C., E. L. Christensen, F. A. Donath, W. E. Falconer, L. J. Lidofsky, E. J. Moniz, T. H. Moss, R. L. Pigford, T. H. Pigford, G. I. Rochlin, R. H. Silsbee, and M. E. Wrenn. 1978. Report to the American Physical Society by the Study Group on Nuclear Fuel Cycles and Waste Management. Rev. Mod. Phys. 50(1) Part II. pp. S114-S117. Hoffman, D. C., T. M. Hamilton, and M. R. Lane. 1992. Spontaneous Fission. LBL-33001, UC-413. Berkeley, Calif.: Lawrence Berkeley Laboratory. Hugon, M. 1994. "Research Activities on Partitioning and Transmutation Supported by the European Commission," Proceedings of the 3rd International Information Exchange Meeting on Actinide and Fission Product Partitioning and Transmutation held in Cadarache, France on December 12-14, 1994. OECD Nuclear Energy Agency report (NEA/PNT Report No. 13). Johnson, T. R., L. Burris, N. M. Levitz, and R. N. Hill. 1990. Use of Transuranic Elements from LWR Fuel in Integral Fast Reactors. ANL-IFR-127. Argonne, Ill.: Argonne National Laboratory. Kang, C. H., and M. S. Kazimi. 1993. Transmutation Effects on Transuranic Waste Inventory and Its Repository Risk. Report MIT-ANP-TR-015. Cambridge: Massachusetts Institute of Technology. Kloosterman, J.L. and J.M. Li. 1994. "Transmutation of 99Tc in Fission Reactors," Proceedings of the 3rd International Information Exchange Meeting on Actinide and Fission Product Partitioning and Transmutation held in Cadarache, France on December 12-14, 1994. OECD Nuclear Energy Agency report (NEA/PNT Report No. 13). Lee, J. C., and J. Du. 1994. Equilibrium cycle calculations for transuranics recycling in pressurized water reactors. Nuclear Technology 108:61-79. National Research Council. 1992. Nuclear Power: Technical and Institutional Options for the Future. Energy Engineering Board. Washington, D.C.: National Academy Press. Pigford, T. H., and C. S. Yang. 1977. Thorium Fuel Cycles. UCB-NE-3227, EPA 68-01-1962. Berkeley: University of California. Pigford, T. H., and J. S. Choi. 1991. Inventory reduction factors for actinide-burning liquid-metal reactors. Trans. Am. Nucl. Soc. 64:123ff.

OCR for page 49
Nuclear Wastes: Technologies for Separations and Transmutation Rawlins, J. A., S. E. Binney, C. H. Bloomster, H. R. Brager, C. A. Burgess, W. J. Gruber, G. F. Howden, A. J. Naser, L. G. Niccoli, A. W. Prichard, G. W. Reddick, W. W. Shulz, J. P. Sloughter, J. L. Swanson , J. W. Thornton, C. N. Wilson, D. E. Wood. 1990. CURE: Clean Use of Reactor Energy. WHC-EP-0268. Richland, Wash.: Westinghouse Hanford Co. Thompson, M. L. 1992. The advanced liquid metal reactor (ALMR)/fuel cycle. Paper presented to STATS Symposium, Washington, D.C., January 13-14. U.S. Department of Energy (DOE). 1987. Characteristics of Spent Fuel, High-Level Waste, and Other Radioactive Wastes Which May Require Long-Term Isolation. DOE/RW-0184. Washington, D.C.: U.S. Department of Energy. Van Tuyle, G. J. 1992. Proposed transmutation of long-lived radioactive wastes. Paper presented at the STATS Symposium, BRWM, NAS, Washington, D.C., January 13, 1992. Wachter, J. W., and A. G. Croff. 1980. Actinide Partitioning-Transmutation Program Final Report. III Transmutation Studies. ORNL/TM-6983. Oak Ridge, Tenn.: Oak Ridge National Laboratory. Young, W. H., and U.S. Department of Energy. 1991. Summary of Nuclear Energy Review of Pigford Paper. Washington, D.C.: U.S. Department of Energy.