The submitted manuscript has been authored DISCLAIMER by a contractor of the U. S. Government under contract No. W-3M09-ENG-38. Accordingly, the U. S. Government retains a nonexclusive, royalty-free license to publish This report was prepared as an account of work sponsored by an agency of the United Stales or reproduce the published form of this Go ernment. Neither the United States Government nor any agency thereof nor any of then- contribution, or allow others to do so, for U. S. Government purposes. employs, makes any warranty, express or implied, or assumes any legal Ii.bd.ty or respons.- bilitv 'for the accuracy, completeness, or usefulness of any informal apparatus, produc or process disclosed, or represents that its use would not infringe pnvately owned nghu. Refer- ence herein to any specific commercial product, process, or semce by trade name, trad ma*, manufacturer, or otherwise does not necessarily constitute or .mply Us endorser^ recom- mendation, or favoring by the United States Government or any agency ".J^6!16*8 CONP-901101—75 and opinions of authors expressed herein do not n<=cessar.ly stale or reflect those of the United States Government o: any agency thereof. DE91 006504

JHI1 ACTINIDE RECYCLE POTENTIAL IN THE INTEGRAL FAST REACTOR (IFR) FUEL CYCLE

Y. I. Chang and C. E. Till Argonne National Laboratory Argonne, Illinois 60439 [708] 972-4856

ABSTRACT (Laboratory.1 The conclusions of this study were that only rather weak economic and safety In the Integral Fast Reactor (IFR) development incentives existed for partitioning and trans- program, the entire reactor system--reactor, fuel muting the actinides for waste management pur- cycle, and waste process is being developed and poses, due to the facts that (1) partitioning optimized at the same time as a single integral iprocesses were complicated and expensive, and entity. The use of metallic fuel in the IFR j(2) the geologic repository was assumed to con- I allows a radically improved fuel cycle tech- tain actinides for hundreds of thousands of nology. Pyroprocessing, which utilizes high jvears. temperatures and molten salt and molten metal {Solvents, can be advantageously utilized for Much has changed in the few years since (processing metal fuels because the product is then. A variety of developments now combine to metal suitable for fabrication into new fuel warrant a renewed assessment of the actinide elements. The key step in the IFR process is recycle. First of all, it has become increas- electrorefining, vhich provides for recovery of ingly difficult to provide to all parties the the valuable fuel constituents, uranium and jnecessary assurance that the repository will plutonium, and for removal of fission products. [contain essentially all radioactive materials In the electrorefining operation, uranium and juntil they have decayed. Assurance can almost plutonium are selectively transported from an icertainly be provided to regulatory agencies by anode to a cathode, leaving impurity elements, isound technical arguments, but it is difficult to mainly fission products, either in the anode convince the general public that the behavior of (compartment or in a molten salt electrolyte. A wastes stored in the ground can be modeled and inotable feature of the IFR process is that the jpredicted for even a few thousand years. From lactinide elements accompany plutonium through ithls point of view alone there would seem to be a {the process. This results in a major advantage (clear benefit in reducing the long-term toxicity I in the high-level waste management, because these lof the high-level wastes placed in the actinides are automatically recycled back into repository. the reactor for in-situ burning. Based on the recent IFR process development, a preliminary Secondly, the Nuclear Waste Policy Act of assessment has also been made to investigate the 1982 mandated that (1) EPA promulgate standards feasibility of further adapting the pyrochemical for protection of the general environment from processes to directly extract aclinides from LWR off-site releases from radioactive materials in spent fuel. The results of this assessment repositories and (2) NRC promulgate technical indicate very promising potential and two most requirements and criteria, consistent with EPA promising flowsheet options have been identified standards, that NRC will apply in licensing oi for further research md development. This paper repositories. The place of actinide recycle also summarizes current thinking on the rationale 'today, therefore, must be reevaluated in the for actinide recycle, 'ts ramifications on the /light of the technical performance requirements geologic repository and the current high-level placed on the repository by these newly estab- waste management plans, and the necessary Ilished NRC regulations2 and EPA standards.3 j development programs. I Finally, IFR pyroprocessing has been' I. BACKGROUND developed only in recent years and it appears to (have potential as a relatively uncomplicated, A multitude of studies and assessments of effective actinide recovery process. In fact, actinide partitioning and transmutation were actinide recycling occurs naturally in the IFR carried out in late 1970's and r<>rly 1930's. fuel cycle. Although still very much develop- Probably the most comprehensive ox these was mental, the entire IFR fuel cycle will be demon- a study coordinated by Oak Ridge National - strated on prototype-scale in conjunction with *Work performed under the auspices of the US Department of Energy under Con- tract W-31-109-ENG-38.

MASTER DISTRIBUTION OF TKiG CC^UiVi IS UNLIMITED the EBR-II and its refurbished Fuel Cycle Facility starting in late 1991. A logical exten- 1 1 1 1 1 ision to this work, therefore, is to establish fchether this IFR pyrochemical processing can be r—^ applied to extracting actinides from LWR spent fuel. \ \w ACTINIDES - II. RATIONALE FOR ACTINIDE RECYCLE

The LWR spent fuel waste consists of both fission product and e'.tinide elements. Fission products comprise hundreds of various isotopes, which, along with energy, are the products of fissioning. Actinides (or more precisely trans- 1 NATURAL URANIUM 1 ORE uranic elements) are produced from neutron cap- ture, as opposed to neutron fission, reactions in the fuel. 10' - — \ FISSION \PRODUCTS The relative radiological toxicities of actinides and fission products contained in once- through LWR spent fuel are compared in Fig. 1, 'normalized to the toxicity of the natural uranium ore. Figure 1 was derived in the following 1 1 1 1 manner: 10 years The International Commission on Radiological Protection (ICRP) Publication 30 gives estimates ,of the cancer risk to various body organs result- i ing from ingestion of a radionuclide.* When this FIGURE 1. Relative Radiological Toxicity of i is multiplied by the number of curies of each Fission Products and Actinides Contained in the iisotope in the LWR spent fuel, the result is an LWR Spent Fuel, Normalized to the Original estimate of the number of cancer doses from each Uranium Ore. radionuclide in the spent fuel. These cancer doses are then summed over radionuclides in two categories, fission products and actini^es, for 1 metric ton of the spent fuel. This gives the instead of "risk" because the word "risk" has an total potential radiological hazard if all implicit connotation that the radioactive iradionuclides in 1 metric ton of the spent fuel nuclides in the repository are released to the tare ingested. This potential hazard is then lenvironment causing ingestion by humans and hence normalized relative to the cancer hazard, cal- linvolves cancer risks. The term "toxicity" is :culated in the same manner, associated with Imeant to indicate the potential hazard contained | natural uranium ore from which the spent fuel in the materials in the repository, without originated. In the natural uranium ore the addressing the issue of actual pathways from uranium is assumed to be in equilibrium with its repository to human body. daughter products. i Another, more concrete, way of evaluating As shown in Fig. 1, most of fission products ithe benefits of actinide recycle is in terms of have relatively short half-lives, and their the ability to satisfy the technical performance radiological toxicity drops below that of their requirements placed on the geologic repository. original uranium ore in a tlmespan of the order The EPA standards establish containment (of 200 yea-.-. Actinides, on the other hand, have requirements that cumulative releases of I long half-lives and their radiological toxicity radionuclides to the accessible environment for 'remains orders of magnitude higher than that due 10,000 years after disposal have a likelihood of ito fission products for millions of years. In lless than one chance in 10 of exceeding the ithis lies the incentive to separate actinides quantities specified as "cumulative release from the waste stream and recycle them back into limits." a reactor for in-situ burning, leaving a waste free of actinides. A proposed NRC rulemaking involves direct incorporation of the EPA cumulative release It should be noted at this point, however, limits within 10 CFR Part 60. Even before this that Fig. 1 does not represent actual radio- Idirect incorporation, 10 CFR Part 60 requires the logical risk fj-om a repository, for example, jrepository to conform to such generally applic- because it does not take any account of pathways able environmental standards for radioactivity as to the human body, and in effect assumes total may have been established by EPA. The EPA cumu- release. The term "toxicity" is used in Fig. 1 lative release limit is therefore a very important technical requirement placed on the If actinides are removed from the waste geologic repository. The EPA cumulative release scream, then the EPA standards on cumulative limits given in terms of curies per metric ton irelease limits can be met more easily. Table 2 heavy metal (MTHM) of spent fuel are listed in lalso indicates the desirable level of actinide Table 1. decontamination. If 99.991 of actinides are removed from the spent fuel, then their activity The LWR spent fuel activities for various will be in the order of unity after 1000 years. nuclides normalized to the EPA limits in Table 1 This is also consistent with the toxicity level are summarized in Table 2. If this value is less considerations. Referring back to Fig. 1, if the than 1, it means that even if the entire inven- actinides in che waste stream are reduced to 10"*, tory is released the EPA limit is not exceeded. their toxicity is in the same range as that of If this value is 100, then only II of the fission products. inventory can be released before the limit is exceeded, and so on. III. IMPLICATIONS ON REPOSITORY Following the 300-1000 years of containment period required by the NRC regulations, the long- As discussed above, actinide recycle allows lived fission products have decayed to the same the technical performance requirements placed on magnitude as the cumulative release limit. the repository to be met more easily, and seems However, the actinide (including plutonium) i highly desirabl3 from the point of view of public activity is aboV" 10* times the cumulative release acceptance. Actinide recycle should make the limit. The allowable release is 10"\ or 0.01X entire geologic repository concept more viable. over the 10,000-year period. Allowable annual At the same time, however, actinide recycle is release rates then would be 10"8 per year. This not a requirement for the high-level waste is a very stringent requirement to meet. : management, and actinide recycle does not replace the need for a geologic repository.

1TABLE 1. Release Limits for Containment Require- The geologic repository is needed, inde- ments. (Cumulative releases to the accessible pendent of actinide recycle, for the following •environment for 10,000 years after disposal) reasons.

The goal decontamination factor for the Release Limit removal of actinides from the waste per 1000 MTHM stream is 10', that is, 99.991 removal or Other Unit of actinides. The residual 0.01X of of Waste actinides means waste that is far from Radionuclide (Curies) being qualified as a non-TRU waste. A decontamination factor of the order of 10' would be needed for non-TRU Americium-241 or -243 100 qualification. Therefore, a geologic repository is needed to store even the Carbon-14 100 residual actinides. Cesium-135 or -137 1000 In addition to residual actinides, there Iodine-129 100 are a number of long-lived fission Neptunium-237 100 | products, such as Tc-99, 1-129, Cs-135, • etc., that will always need to be Plutonium-238, -239, -240, stored. or -242 100 Radium-226 100 There are solidified defense high-level wastes and other civilian high-level Strontium-90 1000 wastes right now that need to be Technetium-99 10,000 disposed in a geologic repository. Thorium-230 or -232 10 Therefore, actinide recycle must not be Tin-126 1000 allowed to interfere with the present high-level waste management plans leading to the first Uranium-233, -234, -235, geologic repository. Actinide recycle is a long- -236, or -238 100 term option. It requires much further R&D to establish the technical, economic, and Any other alpha-emitting institutional practicality. radionuclide with a half-life greater than 20 years 100 Although actinide recycle does not replace ;Any other radionuclide with a the need for a geologic repository, if successful half-life greater than 20 years it should simplify the long-term waste management that does not emit alpha particles 1000 strategy dealing with the need and timing of the second repository. TABLE 2. LWR Spent Fuel Activity (Normalized to the EPA "Cumulative Release Limit" Listed in TABLE 1)

Activity Activity Activity at 10 yrs at 103 yrs at 10* yrs

Sr-90 60,000 0.0 0.0 Cs-137 90,000 0.0 0.0 1-129 0.3 0.3 0..3 Tc-99 1.4 1.4 1.4 Other Long-lived FP 1050 5.1 4.4

Actinides 76,000 19,000 4,000

The Nuclear Waste Policy Act (NWPA) of 1982 in interim on the surface, the effective capacity prohibits the emplacement of spent fuel in the of the first repository could be extended to the first repository containing in excess of 70,000 point where a second repository is not needed for Imetric tons of heavy metal until such time as a a very long time. After their interim storage, second repository is in operation. The intent the waste containing Cs and Sr can be put into was to prevent the first repository being the I the repository for permanent disposal. only one and hence expanding its capacity forever. However, if the radiological toxicity Even after most of the long-lived actinides lis reduced by orders of magnitude through =>re separated from the waste stream, other long- actinide recycle, and the decay heat burden could lived fission products remain. A logical ques- be reduced through an interim storage of cesium tion is whether incentive exists to partition and land strontium, it is reasonable to suggest that transmute these fission products as well. As (the current NWPA be amended to allow more summarized in Table 2, the concentrations of lefficient use of the first repository capacity. these long-lived fission products are low and the activities are in the same range as the EPA The characteristics of the geological cumulative release limits. Furthermore, their formation in which the repository is to be combined toxicity level is at least two orders of located limits the quantity and configuration of magnitude less than that of the natural uranium ithe spent fuel or process wastes that are to be ore as shown earlier in Fig. 1. Therefore, there replaced therein. Analysis of the Yucca Mountain seems to be no strong incentive to partition and Site, for example, indicates that the maximum transmute these long-lived fission products. heat burden that the site can handle without damage to the rock (chemistry, structure, and j However, it should also be noted that mechanical strength) is about 57 kW/acre. For I previous repository risk assessment has shown 10-year cooled spent fuel, the heat load con- 'that Tc-99 and 1-129 dominate the long-term risk. siderations limit the boreholes to only 16 per This is because during an assumed leach incident acre. If the heat load is not a constraint, then at the repository the models used for risk more boreholes can be drilled. The structural analysis predict that technetium and iodine considerations during borehole drilling then migrate through the geosphere rapidly enough to determine the maximum number of boreholes, which |reach the biosphere within one million years, is conservatively estimated to be 13C per acre. 'while actinides are sorbed in the geosphere and This is a factor of 8 increase in t;.e effective do not reach biosphere within one million years. capacity. Furthermore, if heat load considera- That is why the risk is dominated by Tc-99 (921) ations were no longer limiting, the waste form land 1-129 (8X). In perspective, however, the could be compacted for each borehole to fairly health effects of Tc-99 and 1-129 that arise from a repository leach incident are about a factor of easily gain a further factor of about 2.5. So 20 less than the expected health effects from the overall potential for the first repository :natural background. capacity increase is of the order of a factor of 20, if heat loads were eliminated.

As shown in Table 3, the heat load is IV. ACTINIDE RECYCLE IN THE IFR dominated by Cs and Sr in the short term, and by actinides in the long term. Therefore, if A distinguishing element of the IFR concept actinides are recycled and Cs and Sr are stored is its unique fuel cycle based on metallic fuel TABLE 3. Decay Heat of LWR Spent Fuel (Watts/MTHM)

Year Actinides Sr and Cs Others F.P. Total

1 610 8270 3430 12310 5 280 1550 430 2260 10 280 940 60 1300 20 270 650 30 950 50 250 320 2 572 100 215 97 0 312 200 174 9 0 183 500 110 0 0 110

and pyroprocessing.s As shown in Fig. 2, the IFR loxidation state is such that the salt contains a fuel cycle involves only a few steps. The key mixture of actinides. When electrotransport is step in the IFR process is electrorefining. A jused simultaneously to oxidize the metals to cadmium pool in the bottom of the electrorefiner :their chlorides at an anode and to reduce an vessel serves as one electrode. The electrolyte I equivalent amount of chloride to metal at a solid is LiCl-KCl eutectic with about 2 mol X heavy [metal cathode, uranium is preferentially de- metal chlorides, rare earth chlorides, and active Iposited and the other actinides are preferen- metal (sodium, cesium, strontium) chlorides. tially oxidized. The product is essentially pure Suitable electrodes for anodic dissolution of uranium, contaminated with the salt. fuel and for product recovery are placed in the electrolyte. When a crucible containing liquid cadmium is used as the cathode, uranium, plutonium, minor Spent fuel pins are chopped and put in a actinides, and some of rare earths are all basket for dissolution in the electrorefiner at stabilized as metals by interaction with the 500°C. Cadmium chloride is then added to oxidize cadmium. The cadmium electrode product is a alkali, alkaline earth, and most rare earth mixture of actinides with small amounts of rare metals to their chlorides, which become a part of earths. Thus, the recovery and recycle of the the molten chloride electrolyte. Essentially actinides occur naturally in the IFR pyroprocess. pure uranium is electrotransported to a solid cathode and mixed U-Pu product is electro- The hard spectrum reactor characteristic of transported to a liquid cadmium cathode. These the IFR makes an ideal actinide burner. Of cathodes are removed from the electrorefinsr cell course, actinides in a given target element can and retorted to vaporize the cadmium and any be burned in any reactor type. However, large- occluded salt and to consolidate the product by scale actinide burning involving the cumulated aelting. iactinides from the LWR spent fuel inventory icannot be done efficiently in the thermal The chemistry of the pyroprocess is based on spectrum reactor. Because of high capture to die relative ease of oxidation of the elements jfission cross section ratios, substantial amounts that make up the metal fuel. This is determined iof the actinides evolve into higher mass isotopes by the free energies of formation of chlorides of iin the thermal spectrum reactor. This degrades these elements. Alkali and alkaline earth metals reactivity of the actinide mixture. Therefore, ,are readily oxidized into the salt, and less if the reactor is fueled largely with actinides, easily oxidized noble elements remain as metals. there is a limit beyond which they will be The amount of oxidant can be adjusted so that the iunlikely to be placed in a thermal reactor actinides and rare earths are found both as because of reactivity poisoning. As shown in metals and in the salt, although actinides will Table 4, after four cycle generations, the mostly be found as metal and rare earths will reactivity value of the actinide mixture has been mostly be chlorides. Thus, oxidation effects reduced to only about 7X of standard enriched most of the separation of actinides from fission uranium fuel. This requires a 43.6% actinide products. content for the fuel to have equivalent reactivity value. By this time, less than 30X of Separate collection of the uranium product the initial actinide have burned. Thermal and the mixed actinide product is possible reactors are not feasible as true actinide because uranium is slightly less easily oxidized burners. than the other actinides, and because the POWER

METAL-FUELED REACTOR

SPENT FUEL CATHODE INJECTION NEW FUEL & BLANKET ELECTROREFINJNG PROCESSING CASTING & BLANKET

FIGURE 2. Schematic Illustration of the Integral Fast Reactor (IFR) Fuel Cycle.

TABLE 4. Actinide (Including Pu) Recycle in LWR

Actinide Actinide Fraction of Recycle Reactivity Worth Loading Initial Actinide Generation Relative to U-235 Percent HM Remaining

1 0.45 7.1 1.0 2 0.19 16.7 0.84 3 0.12 26.2 0.78 4 0.09 35.2 0.73 5 0.07 43.6 0.71 The LWR-generated actinides (piutonium, V. EXTENSION OF PYROPROCESSING TO LWR SPENT neptunium, amerlcium, curium, etc.) can be used FUEL efficiently in the 1FR spectrum because in this spectrum they have a considerable reactivity IFR pyroprocessing appears to promise value. Further, because of the high fission to improvements in long-term waste management for capture cross section ratios, they tend to burn the IFR itself. The next question that arises rather than transmute and an equilibrium naturally is whether the approach can be extended concentration of actinide is quickly established s.;.- process LWR spent fuel. And, in fact, it with continuous recycling. The IFR can be turns out that there is an extensive experience designed to operate in an actinide self- base at Argonne in applying pyrochemical sustaining mode or as a net actinide burner. 'processes to oxide fuel.

The principal high-level wastes from the IFR In the late 19v0's,.the EBR-II fuel cycle pyroprocess fuel cycle are the metals and salts facility operated for about five years using a discharged from the electrorefiner. The metal .simple drossing process, known as melt refining. waste consists of fuel cladding and perhaps a 'In this process the electropositive fission small amount of cadmium that Is not recycled. products were separated from the fuel by reaction This waste will also contain the noble metal with a zirconium oxide crucible. The volatile fission products, most of the alloy zirconium, elements were released and collected on a fume and small amounts of actinides. The electro- trap or condensed cryogenically from the cell refiner salt will contain the halide, alkali- atmosphere. metal, alkaline earth and rare earth fission products along with some actinides. These waste A pyrochemical process for recovering the streams are then converted to forms that are actinides occluded with the dross, or crucible acceptable for disposal, as described below. skull, was developed and demonstrated on an engineering scale with simulated fuel. This The salt waste is treated to reduce actinide skull reclamation process employed liquid zinc- content to less than 1 ppm by contact with a Ll- magnesium and molten chloride salt as process Cd alloy, followed by filtration to remove solvents. Also a blanket process for recovery of insoluble impurities. After this treatment, the piutonium from metallic uranium blankets was salt contains only the fission products cesium, demonstrated on a bench scale. This process strontium, and iodine. Its alpha activity will involved the selective extraction of piutonium be below 100 nCi/g. The salt waste will then be from molten uranium intr an immiscible magnesium immobilized in a suitable matrix and dispersed in phase. metal or sintered ceramic matrix sealed in containers. I The techniques developed to process EBR-II skulls and blankets were then extended to The metallic wastes from electrorefining processing uranium oxide and mi.ed oxide fast will be combined with the Cd-Li used in treating reactor fuels. Rapid methods for reducing tnese the salt and the excess cadmium will be removed dense oxide fuels were demonstrated, and a liquid by retorting. The residue will be dispersed and metal-fused salt extraction step was developed immobilized in a corrosion-resistant metal matrix for isolating the uranium, plutonl'jj, and fission such as copper. This mixture will then be sealed products. This "salt transport process" was in corrosion-resistant containers for disposal as I demonstrated on a laboratory scale, but funding high-level waste. terminated before a planned pilot plant demonstration could be completed. In the IFR process, cesium and strontium are already separated in the form of the salt waste, , The earlier pyrochemical process development and therefore, there is no need to have addi- 1 efforts were discontinued because there was no tional separation processes developed if it were clear advantage to producing a pure piutonium deemed desirable to implement the alternative product stream over the traditional Purex waste management discussed earlier. The salt process. Today, however, the processing goal has waste packages can be stored for an appropriate changed. In traditional reprocessing based on period to allow the decay heat reduction before Furex, the goal was to produce a highly they are permanently disposed in the geologic decontaninated, pure Pu product stream. However, repository. when LWR processing is viewed as a waste management strategy, the goal is quite different. The status of the IFR process development Neither a pure Pu product stream nor a high has reached the point where a prototype demon- decontamination factor is required. In fact, stration of the entire IFR fuel cycle will be just the opposite is desirable. The new process conducted in conjunction with EBR-II and its goals, when LWR spent fuel processing is viewed refurbished Fuel Cycle Facility, beginning in as a waste management strategy, are as follows: late 1991. Direct extraction of all actinides (Pu, that even $1000/kgHM processing cost is likely to Np, Am, Cm, etc.) from the spent fuel be optimistic. But even this translates to as a single product stream. J4.3 mills/kWhr incremental cost to the LWRs. [Accounting adjustments such as transferring this An actinide recovery target of 99.931 to cost increment to the future LMRs is no solution 99.99X. because in the end utilities will operate both the LWRs and LMRs. The process should be incapable of producing pure Pu product. It is economically essential to develop a simple process that can extract actinides The process should be incapable of directly from the spent fuel. However, to be achieving a high decontamination factor fair, the economic cost/benefit analysis for for fission products. actinide recycle must be done for the entire system including the effects on the long term The process should be simple enough Co repository requirements. achieve acceptable economics.

A preliminary assessment has been made to VI. DEVELOPMENT NEEDS investigate the feasibility of using pyrochemical processes for directly extracting actinides from An IFR economy can be justiiied and LWR spent fuel, satisfying the new process goals establish d without calling upon justification discussed above. It appears that the pyro- from LWR actinide recycle. However, LWR/IFR chemical processes are exactly compatible with synergistic fuel cycle do provide advantages, and the new process goals and two promising flowshest without the IFR fuel cycle, actinide recycling of |opeions have been identified: (1) a salt trans- the LWR spent fuel does not seem practical. Iport process and (2) a magnesium extraction process. These two processes are presented in Specific policy decisions regarding LWR/IFR Figs. 3 and 4. synergist.ic fuel cycle implementation are not possible at this time. The first repository is The pyrochemical processes fit naturally to needed in any event. The present repository the LWR actinide extraction application and development program must continue and the IFR should provide significant advantages over the development program can proceed in parallel. IFR traditional Purex-based processes. development and Its associated actir.td

If actinide extraction is based on the traditional Purex-based technology, it appears REFERENCES Spent Nuclear Fuel; High-level and Trans- uranic Radioactive Wastes." 1. A. G. Croff, J. 0. Bloneke, and B. C. Finney, "Actinide Partitioning-Transmutation Program 4. International Commission on Radiological Final Report. I. Overall Assessment," ORNL- Protection (ICRP) Publication 2, "Permissible 5566 (1980). Dose for Internal Radiation," 1959; Publication 30, "Limits for Intakes of 2. U.S. Nuclear Regulatory Commission, 10 CFR Radionuclides by Workers," 1979; Publication Part 60, "Disposal of High-level Radioactive 30, Part 2, 1980; Publication 30, Part t, Wastes in Geologic Repositories Licensing 1988. Procedures." 15. C. E. Till and Y. I. Chang, "The Integral 3. Environmental Protection Agency, 40 CFR Part Fast Reactor," Advanaes in Nuclear Science 191, "Environmental Radiation Protection and Technology. 20, 127-154 (1988). Standards for Management and Disposal of

FIGURE 3. Salt Transport Conceptual Process for Extracting Actinides from LWR Spent Fuel.

FIGURE 4. Magnesium Extraction Conceptual Process for Extracting Actinides from LWR Spent Fuel.