o Sy | AQ 2 /d‘é " ORNL.-5388 MASTER Interim Assessment of the Denatured 233U Fuel Cycle: Feasibility and Nonproliferation Characteristics BN E S T Ao T 'LABORATORY | ,':V:‘.f'."'}r OPERATED BY UNION CAR_BIDE (ORPORATION __FOR THE DEPARTMENT UF ENERGY - s wm » Printed in the United States of America. Available from National Technical Information Service U.S. Department of Commerce 5285 Port Royal Road, Springfield, Virginia 22161 Price: Printed Copy $11.75; Microfiche $3.00 This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, contractors, subcontractors, or their employees, makes any warranty, express or implied, nor assumes any legal liability or responsibility for any third party’s use or the results of such use of any information, apparatus, product or process disclosed in this report, nor represents that its use by such third party would _ not infringe privately owned rights. e e e o e o e e I[f#, r Contract No. W—74Q5-eng-26 Engineering Physics Division ORNL-5388 Distribution Category UC-80 INTERIM ASSESSMENT OF THE DENATURED 233U FUEL CYCLE: FEASIBILITY AND NONPROLIFERATION CHARACTERISTICS Edited by L. S. Abbott, D. E. Bartine, T. J. Burns With Contributions from Argonne National Laboratory Brookhaven National Laboratory Combustion Engineering, Inc. Hanford Engineering Development Laboratory Oak Ridge Gaseous Diffusion Plant Oak Ridge National Laboratory C. Sege DOE Program Manager D. E. Bartine ORNL Program Manager I. Spiewak ORNL Program Director Detember, 1978 -0OAK RIDGE NATIONAL LABORATORY Oak Ridge, Tennessee 37830 operated by UNION CARBIDE CORPORATION for the DEPARTMENT OF ENERGY NOTICE This report was prepared as an account of work .| sponsored by the United States Government. Neither the -AUnited States nor the United States Department of Energy, nor any of their employees, nor any of their - | contractors, subcontractors, or their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness or usefulness of any information, apparatus, product or process disclosed, or represents that its use would not infringe privately owned rights. DISTRIBUTION OF TRIS DOCUMENT I3 UNLLMXlX e rr o r r o !f”f'. - 1:' | m o r 20 ME O - T 0 =X M m o L .. T 114 PRINCIPAL AUTHORS T. J. Burns Oak Ridge National Laboratory J. C. Cleveland Oak Ridge National Laboratory E. H. Gift Oak Ridge Gaseous Diffusion Plant R. W. Hardie Hanford Engineering Development Laboratory C. M. Newstead Brookhaven National Laboratory R. P. Omberg A Hanford Engineering Development Laboratory N. L. Shapiro Combustion Engineering I. Spiewak Oak Ridge National Laboratory CONTRIBUTING AUTHORS . Arthur, 0Oak Ridge Gaseous Diffusion Plant Black, Hanford Engineering Development Laboratory Brooksbank, oOak Ridge National Laboratory Chang, Argonne National Laboratory Haas, Oak Ridge National Laboratory Haffner, Hanford Engineering Dévélopment Laboratory Helm, Hanfora Engineering Development Laboratory -Ingersoll, Oak Ridge National Laboratory Jenkins, Oak Ridge National Laboratory Jolly, Hanford Engineering Development Laboratory Knee, oak Ridge National Laboratory Meyer, Oak Ridge National Laboratory Selby, Oak Ridge National Laboratory Shay, Hanford Engineering Development Laboratory Till, oak Ridge National Laboratory J q L ' L PREFACE AND ACKNOWLEDGMENTS This report describing a study of the feasibility of the denatured 233U fuel cycle integrates the data and contributions of a number of national laboratories and government contractors. Those of us at ORNL who have been responsible for compiling and editing the report wish to acknowledge the assistance of many individuals who actively participated in the study throughout the many iterations leading to -this final draft. In particular, we wish to thank Carol Sege and Saul Strauch of the U.S. Department of Energy for their guidance during the entire program; C. M. Newstead of Brookhaven National Laboratory for the proliferation-risk assessment; E. H. Gift of the Qak Ridge Gaseous Diffusion Plant for the analysis of the potential circumvention of the fuel isotope barrier; Y. Chang of Argonne National Laboratory, J. C. Cleveland and P. R. Kasten of Qak Ridge National Laboratory, R. P. Omberg of the Hanford Engineering Development Laboratory, and N. L. Shapiro of Combustion Engineering, Inc. for the characterizations of reactor and fuel cycle technologies; and R. P. Omberg and R. W. Hardie of Hanford Engineering Development Laboratory for the system economics-resources analysis. These, in turn, were assisted by several contributing authors as listed on page iii. Many others have provided data or participated in reviews of the various chapters, and to each of them we express our appreciation. Finally, we wish to thank the many secretaries and report production staff members who have so patiently prepared numerous drafts of this report. Irving Spiewak David E. Bartine Thomas J. Burns Lorraine S. Abbott . o l[i;g -IEiT . r— o - . i; ié vii CONTENTS PREFACE AND ACKNOWLEDGMENTS .............. Ceeetereceecnrareneeeans enereeeeees cees ABSTRACT ........ et teeeeeteeieeereeteeneeaaeneeeaeerernneas eeeeaaeans eenes 1. INTRODUCTION: BACKGROUND ......... tecestesenesasaassansans ceesensrense cesesses 2. RATIONALE FOR DENATURED FUEL CYCLES ........... Ceerrreraneeanans ereeeeeeeenaan 2.0. Introduction .......... casene tesesasareecsoss cerecenas easreseanas creneans 2.1. International Plutonium ECONOMY ..cvoeveevsereroscacocnsosnnnenne ceresanns 2.2. The Denatured 233U Fuel Cycle ..viveeeesvreeacssosvssccncscssnnnsasse cenes 2.3. Some Institutional Considerations for the Denatured Fuel Cycle .......... 3. ISOTOPIC CHARACTERISTICS OF DENATURED 233U FUEL ..ucvvvevneenennncnenns cerosnne 3.0. Introduction .....cieevevvenvneenns Ceeeteneeenaan evereans Ceeereereaeeas 3.1. Estimated 232U Concentrations in Denatured 233U Fuels ........... Ceeeeees 3.1.1. Light-Water Reactor Fuels .......cevvuvunn cesesssessanses tessenss 3.1.2. High-Temperature Gas-Cooled Reactor Fuels .....cenveee ceeresnseas 3.1.3. Liquid-Metal Fast Breeder Reactor Fuels ........... tretestecranes 3.7.4. ConClUSTONS . .ivverencenrreesocacoanssnsnnsoransincsossascsssossns 3.2. "Radiological Hazards of Denatured Fuel Isotopes .....ecevievnveneenncenns 3.2.1. Toxicity of 233y and 232y ,........ Creeseeos Crereereanacnennenns 3.2.2. Toxicity of 232Th ...viviverieeerenacasans Cerees Cereseeeasanasnns 3.2.3. Hazards Related to Gamma-Ray Em1ss1ons cessincass tecresesrsesanan 3.2.4. Conclusions ..eeevvececanne Crssenestessestesasaseines s rasrssaasnna 3.3. Isotopics Impacting Fuel Safeguards Considerations .....ccceeevvescnnenne 3.3.1. Enrichment Criteria for Denatured Fuel .......... Cetrescsasenanns 3.3.2. Fabrication and Handling of Denatured Fuel ......ecvveveevenencas 3.3.3. Detection and Assay of Denatured Fuel ......cciviveirenneinnnsnass 3.3.4. Potential Circumvention of Isotopic Barrier - of Denatured Fuel .......covvivevnnces teesvesesatssanarnsaranna 3.3.5. Deterrence Value of 232U Contam1nation 1n Denatured Fue] ........ 4. IMPACT OF DENATURED 233y FUEL ON REACTOR PERFORMANCE ........ sececscsssseracnce 4.0. Introduction .....c.ccniiiiiiiinan.n, cerersasan Sessvestiscnssesantaranns . 4.1, Light-Water ReaCtOrS «...ueveerennrsrennsereeseeieseiieesoisseseannenans 4,1.1. Pressurized Water REACLOTS ...cuveuieiuriuoresssnsoccsssoassorsas 4.1.2. Boiling Water Reactors ......ciivvieeiieiieniencarnecioncnnssnnes 4.2. Spectral-Shift-Controlled ReaCtOrS .....eiveeuiiisuniiisioriineenneeannss 4.3. Heavy-Water Reactors ......ecuiveemeevsinncniiiiatassasnocoscansns cesenee 4.4, Gas-Cooled Thermal REACEONS +.vuveiiununeersnnnanssunnesernsassonansaonns 4.4.1. High-Temperature Gas-Cooled Reactors ............ e eraeneanes 4.4.2. Pebble-Bed High-Temperature Reactors ........cciveevinecnennnnnn. 4.5. Liquid-Metal Fast Breeder Reactors .........ceieeiiinivenicniioneerennens © 4.6, Alternate Fast Reactors ......civeeviiovoiaanenennnen Ceeesrenes cererene | 4.6.1. Advanced Oxide~Fueled LMFBRS .....cceveuueunransnasoncnnssncns cee - 4.6.2. Carbide- and Metal-Fueled LMFBRs ceerereisessenseitiisainantnens 4,6.3 .3. Gas-Cooled Fast Breeder ReaCtors ....eeeereeeceerceceasansansaces viii Page IMPLEMENTATION OF DENATURED FUEL CYCLES ........... et eseceevesretssasssensaans 5-1 5.0, Introduction ...cieeiiiiiineiieiirensesrocterssnsesssssorssascnssssansnns 5-3 5.1. Reactor Research and Development Requiremehts ............... sesseerseans 5-4 5.1.1. Light-Water Reactors ...civeceerrrrenirencnncracnccnsnnenns erees 5-8 5.1.2. High-Temperature Gas- Coo]ed Reactors ........................ wsee 5-11 5.1.3. Heavy-Water Reactors .......ccceceiveneens e teseseccvscsassssasanan 5-13 5.1.4. Spectral-Shift-Controlled Reactors teteeessssecsscnsrseasesenonas 5-14 5.1.5. R,D&D Schedules ....civiierivienensosasnsosorcsassncacnsonsasanns 5-17 5.1.6 Summary and Conclusions ...eececececscnnncssnscnnaans eeansessennes 5-17 5.2. Fuel Recycle Research and Development Requ1rements ...................... 5-21 5.2.1. Technology Status SUMMAry ....e.evieenaencesocnacsaceanosvonnnnss 5-21 5.2.2. Research, Development, and Demonstration Cost Ranges - and SChEdU]ES ® 8 2 0 B LS F O EBE NPT EPAEEE SIS EBPEETREITOIESEAtEEID BN 5-24 '5.2.3. Conclusions ..cieeeviiiniieneonresarsenannnss teeieseesssesraaaaaas 5-26 . EVALUATION OF NUCLEAR POWER SYSTEMS UTILIZING DENATURED FUEL ........ e 6-1 6-0. IntY'OdUCtion ----------- A ss e w .-oo-.--o.---oOQoo.oolln|||-o§oo- ----------- 6-3 6.1. Basic Assumptions and Analysis Technique ........cvveivennnnas veeecessses BB 6.1.7. The U30g SUPPTY cevivenraeroscanrscsonsescenassesnsessscsvacennssns 6-5 6.1.2. ‘Reactor Options ....evevuvns e teseressaensesterantasatanesosanaas 6-6 6.1.3. Nuclear Policy Options ....... Nesesncescseseacsacaarasecasranonnn 6-10 . 6.1.4. The Analytical Method ........c......s. cerenne revensencecsensaes 6-11 6.2. Discussion of Results for Selected Nuclear Policy Options ............... 6-23 6.2.1. The Throwaway/Stowaway Option ....ieeriiiiennincenrnenerennennnns - 6-23 6.2.2. Converter System with Plutonium Recycle ....covvviiviinrncnnennn. 6-30 6.2.3. Converter System with Plutonium Throwaway ...........covcene. vees 6-33 6.2.4. Converter System with Plutonium Production Minimized; Pu-t0-2330 "Transmutation" ....eeeiieneerececnsesavaeecvaorsaanes 6-35 6.2.5. Converter System with Plutonium Production Not Minimized; Pu-t0-233U "Transmutation" ...c.ueeieereroecnnesoarorsoscscsoanss 6-38 6.2.6. Converter-Breeder System with Light Plutonium : Transmutation” ....cciieiieiienneeiscevavsoresssaranssnsscraanes 6-41 6.2.7. Converter-Breeder System with Heavy P]uton1um Transmutation" ...c.cerieriiiierittrtitntsrisiassansnonnsasannes 6-44 6.3. ConcluSions ....covvevnennsss Ceeatderevatecsuratst et tasreeratatsneesannnns 6-47 OVERALL ANALYSIS OF DENATURED FUEL SYSTEMS ...iieieinirieinnetecnnenasonnnnnans 7-1 7-0- IntY‘OdUCtion S 2t P e PR T SR IR B R Tt E st R TEE S ST RS AR s s seoen .. 7-3 7.1. Proliferation-Resistant Characteristics of Denatured 233U Fuel .......... 7-4 7.1.1. Isotopic Barrier of Fresh Fuel ...... Cesesenaen hiavsesecsssansane 7-4 7.1.2. Gamma-Radiation Barrier of Fresh Fuel .......ccciierieiiirrencenns 7-6 7.1.3. Spent Fuel Fissile Content ......... S hesssssesssanasasaastnasanns 7-7 7.1.4. Conclusions ......... eessasseeerssesrannatoseenns teveesecceacns 7-9 7.2. Impact of Denatured 233U Fuel on Reactor Performance and Selection: Comparison with Other Fuel Cycles ...vvnieeieiinressisreerereoanssnnonnes 7-10 7.2.1. Thermal Reactors ...eeeveereseesesnacesnnasnsenes e iereeereeens 7-10 Once-Through Systems ......ievierierecsrecsenononcncaronas .. 1=10 Recycle Systems ..ceeevecaconnsensanne Chrseecsasescssinanans 7-13 7.2.2, Fast Reactors ....eeieennciceronnecncnnns teeeeriseasennsan cevenee 7-14 - 7.2.3. Symbiotic Reactor Systems ....................................... 7-16 7 2 4 CUnC]US10nS ------------------ ..-c-o;l-oonoo..oaannn-.oo--o. ------- 7"]9 7.3. Pr ospects for Implementation and Commercialization of Denatured ' ' 233 FUET CYCTR wuirvererreneecvnncnocssoassssnnnns eeetesssresenaneenenanea 7-21 1 C | | — - C O o oo ( T .. 0 v ix Page 7.3.1. Poss1b1£ Procedure for Implementing and Commercializing the Denatured Fuel Cycle ....... et arecencecsneriesasasaresnsanse 7-23 7.3.2. Considerations in Commercializing Reactors Operating on Alternate Fuels .....cvivvvunvenen Ceteciecsasasestetsetrasannoe 7-25 7.3.3. Conclusions ........... Ceeetaecans teseeann Cetesaserteennsseseanans 7-27 7.4. Adequacy of Nuclear Power Systems Utilizing Denatured 233y Fuel for Meeting Electrical Power Demands ....... Ceetreeeetrecaeneratartananns 7-29 7.4.1. The Analytical Method .......ccvvevenianns Ceerseaenes teesssteenae 7-30 7.4.2. Data Base .......... Cieteereeriseseetartrstesnnran Cereeersesarens 7-31 7.4.3. Results for Price-Limited Uranium Supplies .....cvvvivinneneannss 7-31 Non-FBR Systems, Options 1, 2, 4, and 5 .....cvvveivnnrnnenns 7-33 FBR Systems, Options 3, 6, 7, and 8 ....cveeeenvnnennonannas 7-35 7.4.4. Results for Unconstrained Resource Availability ................. 7-35 7.4.5, Systems Employing Improved LWRs and Enrichment Technology ....... 7-38 7.4.6. ConClUSTONS tuvvviinniieernoseoesroscacnnsarossncannss Ceesecananas 7-40 7.5. Tradeoff Analysis and Overall Strategy Considerations ........c.evveeeces 7-42 7.5.1. No-Recycle Options ........c... heenas Chereevretaanserenrcrnsaanen 7-43 7.5.2. Recycle OpLioNns ciuuevieeeereenerioeeesooeceonnasscanansonsonnnns 7-44 7.5.3. Overall Conclusions and Recommendations ........cccceeeveeccsennes 7-48 APPENDICES ....ccvevvenn. teenee Ceeeae veeseennens crrereenaann cetetieetetanrrananras A-1 App. A. ISOTOPE SEPARATION TECHNOLOGIES ....ciivieeeereesearoasassoaronannsss A-3 A.1. Current Separation Capability .......... Cereeseriaceacstrananne A-3 The Gaseous Diffusion Process ....vevevveveees Ceeeanseenanas A-3 The Gas Centrifuge ProCess «.iiiiisiiircinsionnnssnnsssscnsens A-3 The Becker Separation Nozzle ......cciiiieirininninnenannanse A-7 The South Helikon ProCess ....veveeieeeneserecsesessasnnncas A-9 Current and Projected Enrichment Capac1ty .................. A-10 A.2. New Separation Technologies ....ccvevvervnnnnns Ceerererenceanns A-10 Photoexcitation (Laser) Methods .....vveevveveonennrenannnas A-13 Chemical Exchange Methods ...cvuveenrnnsirnniernesennsonases A-15 Aerodynamic Methods ....... Cetrertessaesasasaus Ceseseeananes A-17 Plasma-Based ProCesSSes v.eveiiercrcsscsossasnscanenssannenes A-17 Comparison of Advance Separation Processes ................. A-18 App. B. ECONOMIC DATA BASE USED FOR EVALUATIONS OF NUCLEAR POWER SYSTEMS L itriiiiiiiitrecenennsesocnsroncnssonns theetssecerienoananns B-1 App. C. DETAILED RESULTS FROM EVALUATIONS OF VARIOUS NUCLEAR POWER SYSTEMS UTILIZING DENATURED FUEL ..e.vveevvenannnnes ereseeanaenans eea C-1 App. D. CALCULATIONS OF NUCLEAR AND FOSSIL PLANT COMPETITION BASED ON ECONOMICS vivvvvrvnnvenanns teeeese fhesenecsesannrans Cedeenesanae eee D-1 . J r .o { " r— € r . & r- ¥ L xi ABSTRACT A fuel cycle that employs 233 denatured with 238U and mixed with thorium fertile material is examined with respect to its proliferation-resistance characteristics and its technical and economic feasibility. The rationale for considering the denatured 233U fuel cycle is presented, and the impact of the denatured fuel on the performance of Light-Water Reactors, Spectral-Shift-Controlied Reactors, Gas-Cooled Reactors, Heavy-Water Reactors, and Fast Breeder Reactors is discussed. The scope of the R,D&D programs to commercialize these reactors and their associated fuel cycles is also summarized and the resource require- ments and economics of denatured 233U cycles are compared to those of the conventional Pu/U cycle. In addition, several nuclear power systems that employ denatured 233 fuel and are based on the energy center concept are evaluated. Under this concept, dispersed power reactors fueled with denatured or low-enriched uranium fuel are supported by secure energy centers in which sensitive activities of the nuclear cycle are performed. These activities include 233 production by Pu-fueled "transmuters" (thermal or fast reactors) and reprocessing. A summary chapter presents the most significant conclusions from the study and recommends areas for future work. - _H.HH B J T T, o - T . T CHAPTER 1 INTRODUCTION: BACKGROUND D. E. Bartine, L. S. Abbott, and T. J. Burns Oak Ridge National Laboratory - J - 1 r—. . U r" | r- rC . C'“""’“Y r". — . " 1. INTRODUCTION: BACKGROUND In the mid-1940s, as the nuclear era was just beginning, a prestigiofis group includ- ing Robert Oppenheimer and led by David Lilienthal, the first chairman of the U.S. Atomic Energy Commission, was commissioned by Under Secretary of State Dean Acheson to recommend ways that the benefits of nuclear energy could be shared with the world without the dangers of what we now refer to as "nuclear proliferation": that is, the creation of numerous nuclear weapons states. The report! they submitted states that "the proposed solution is an international institution and framework of treaties and agreements for cooperative operation of sensitive nuclear technology." At the same time, the committee proposed several possible technological deve]opmehts to help implement an international system, including the denaturing of reactor fuels. They also suggested the restriction of the most sensitive activities within a nuclear cycle to ‘nuclear energy arenas. In the subsequeht years several steps have been taken toward international coopera- tion in the political control of the potential for making nuclear weapons. In 1953 the Atoms for Peace Program was initiated by the U.S. and in 1957 the International Atomic Energy Adency was formed, one of its chartered responsibilities being the safeguarding of fissile material and the reduction of the potential for the production of nuclear weapons. In 1970 these efforts resulted in a nonproliferation treaty that was drafted by the U.S. and the U.S.S.R. and subscribed to by 116 nations. As the diaTog has continued, inevit- ably all serious studies of the problem, including the most recent s;udies, have arrived at the same conclusion as the Acheson committee: international cooperation and safeguards with technological supports are mandatory -- or to state it another way, no purely tech- nological fix to prevent nuclear proliferation is possible. It was against this background and largely through the initiatives of President Carter that an International Nuclear Fuel Cycle Evaluation Program (INFCE) was established in the Fall of 1977 to study how proliferation-resistant nuclear fuel cycles could be developed for world-wide nuclear generation of electrical power. - At the same time a U.S. Nonproliferation Alternative Systems Assessment Program (NASAP) was formed to.carry out intensive studies that would both provide input to INFCE and recommend technical and fnstitutional approaches that could be implemented with various nuc]ear fuel cycles proposed for the U.S. | ' The principa].proiiferatioh concern in civilian riuclear power fuel cycles is the pos- sible diversion of fissile material to the fabrication of nuclear weapons. If obtained in sufficient quantities, the fissile.materigl employed in any nuclear fuel cycle can be pro- cessed into weapons-usable material,-but fuel cycles that are considered to offer the least resistance to diversion are those that include weapons usable material that can be chemi- cally separated from all the other materials in the cycle. The 235U in the low-enriched uranjum (LEU) fuel used by currently operating Light-Water Reactors (LWRs) cannot be chemi- cally separated because it is embedded in a matrix of 238U, To extract the 235U from the 238U T e e B L w i e e e ot . 1-4 would require isotopic separation which is technologically difficult and for which few facilities in the world currently exist. The uranium mixture itself could not be used for weapons fabrication because the concentration of the fissile component is too low. By contrast, the plutonium in the Pu/U mixed oxide fuel cycle developed for fast ~ breeder reactors such as the Liquid Metal Fast Breeder (LFMBR) can be chemically separated from the other materials in the cycle. Thus, as presently developed, the Pu/U fuel cycle is perceiyed to be less proliferation resistant than the LEU cycle. This facet of the . FBR-Pu/U fuel cycle was obviously a major factor in the Administration’s decision in April, 1977, to defer commercialization of the LMFBR in the United States. Another concern about plutonium centers on its presence in the “"back end" of the LEU fuel cycle. While it does not exist in the "front end" of the cycle (that is, in the fresh fuel), plutonium is produced in the 238U of the fuel elements during reactor opera- tions. Thus the spent LWR elements contain fissile plutonium that is chemica11y extract- able. The fuel cycle technology includes steps for reprocessing the elements to recover and recycle the plutonium, together with other unburned fissile material in the elements, but to date this has not been done in the U.S. and currently a moratorium on U.S. commercial reprocessing is in effect. As a result, the spent fuel elements now being removed from LWRs are being stored on site. Because initially they are highly radioactive due to a fission-product buildup, the spent elements must be heavily shielded, but as their radio- activity decays with time less shielding will be required. Various nuclear “alternatives" are being proposed by the U.S. and other countries for international consideration in Tieu of the classical Pu/U cycle. One proposal is that nations continue marketing LWRs and other types of thermal reactors fueled with natural or low-enriched uranium. A moratorium on reprocessing would be adopted, and the spent fuel would be stored in secure national or international centers such as has recently been proposed by the United States, the security of the fuel being transported to the centers being provided by its fission-product radioactivity. This scenario assumes a guarantee to the nuclear-power-consuming nations of a fuel supply for the approximately 30-year economic life of their nuclear plants. Other proposals that assume the absence of reprocessing (and thus do not include recycle of uranium and/or plutonium) are aimed at improving the in-sity utilization of fissile material within the framework of current light-water technology. Light-water reactor options such as improved refueling patterns and cycle "coastdown" procedures, as well as more extensive modifications (such as increasing the design burnup), are being studied. Significant gains in resource utilization also appear possible with the intro- duction of "advanced converter" designs based on Heavy-Water Reactors (HWRs), Spectral- Shift-Controlled Reactors (SSCRs), or High-Temperature Gas-Cooled Reactors (HTGRs). i o a[v”‘ r . r . 1 r- i ¥ r .. r°. ( ! 1-5 While these various proposals could be useful for increasing the energy generated from the uranium resource base while recycling is disallowed, they will not provide the "inexhaustible" supply of nuclear fuel that has been anticipated from the commercialization of fuel recycle and breeder reactors. To provide such a supply would require the separation and reuse of the "artificial" fissile isotopes 23%Py and 233U, It was under the assumption that recycle would occur, initially in LWRs, that the'technology for the Pu/U mixed-oxide fuel cycle, in which 23%Py is bred from 238], was developed. However, for the reasons stated above, the proliferation resistance of the cycle as currently developed is perceived as being inadequate. Its proliferation resistance could be increased by deliberately "spiking" the fresh fuel elements with radioactive contaminants or allowing them to retain some of the fission products from thé previous cycle, either of which would discourage seizure by unauthorized groups or states. The feasibility of these and other possible modifications to the cycle are currently under study. In addition, the employment of full-scope safeguards, including extensive fissile monitoring proCedures, is being investigated for use with the Pu/U cycle. Also under study are several "alternate" fuel cycles based on the use of the artificial fissile isotope 233 which is bred in 232Th, One such cycle is the 233y/238y/232Th cyc]e'proposed by Feiveson and Taylor,2 and it is this cycle that is the subject of this report. In the 233/238y/232Th fyel cycle the 233 is mixed with 238) which serves as a denaturant. The fertile isotope 232Th is included to breed additional 233U. The addition of the 238U denaturant makes the proposed fuel cycle similar to the 235y/238y cycle currently employed in LWRs in that extracting the 233U for weapons fabrication would require isotope separation facilities. Since 233 does not occur in nature, the cycle is also similar to the 23%Pu/238 cycle in that reprocessing will be necessary to utilize the bred fuel. However, as suggested by the Acheson Conmittee and again by Feiveson and Taylor, reprocessing and other sensitive activities could be restricted to secure energy centers and still allow power to be generated outside the centers. It is the purpose of this report to assess in the 1light of today's knowledge the potential of the denatured 233y fuel cycle for meeting the requirements for electrical power growth while at the same time redhcing proliferation risks. Chapter 2 examines fhe rationale for utilizing the denatured fuel cycle as a reduced proliferation measure, and Chapter 3 attempts to assess the impact of the isotopics of the cycle, especially with respect to an implied tradeoff between chemical inseparability and isotopic separability of the fuel components. Chapter 4 examines the neutronic performance of various reactor types utilizing denatured 233U'fuel,-and Chapter 5 discusses the require- ments and projections for impiementing the cycle. Chapter 6 then evaluates various nucl- ear power systems uti]izing'denatured fuel. Finally, Chapter 7 gives sumnations of the . safeguards considerations and reactor neutronic and symbjotic aspects and discusses the prospects for deploying denatured reactor systems.i Chapter 7 also presents the overall conclusions and recommendations resulting from this study. e The reader will note that throughout the study the U.S. has been used as the base case. This was necessary because the available input data -- that is, resource base ' estimates, projected reactor and fuel cycle development schedules, and assumed power growth rates --‘are all of U.S. origin. However, with access to corresponding data for an international base, the study could be scaled upward to cover an interdependent world model. References for Chapter 1 1. "A Report on the International Control of Atomic Energy," prepared for the Secretary of State's Committee on Atomic Energy by a Board of Consultants: Chester 1. Barnard, Dr. J. R. Oppenheimer, Dr. Charles A. Thomas, Harry Winne, and David E. Lilienthal (Chairman), Washington, D C., March 16, 1946, pp. 127-213, Department of State Publi- cation 2493 2. H. A. Fe1veson and T. B, Taylor, "Security Implications of Alternative Fission Futures, Bull. Atomic Scientists, p. 14 (December 1976) : _ sy . I — s — Ty r r Lty r an 3 L ———— — . r ey I ) ) | S A r T 2.0, 2.1. 2.2. 2.3. CHAPTER 2 RATIONALE FOR DENATURED FUEL CYCLES T. J. Burns Oak Ridge National Laboratory Chapter Qutline Introduction International Plutonium Economy The Denatured 2331 Fuel Cycle Some Institutional Considerations of the Denatured Fuel Cycle J " " ' r ri | i ! % -ve 1 i 4 -y ri. 1 g 2.0. INTRODUCTION The primary rationale for considering the proliferation potential of the nuclear fuel cycles associated with civilian power reactors derives from two opposing concerns: the possibility of nuclear weapons proliferation versus a need for and the perceived economic/resource benefits of a nuclear-based generating capacity. At the outset it should be emphasized that a civilian nuclear power program is not the only proliferation route available to nonnuclear weapons states, The countries that have developed nuclear explosives to date have not relied on a civilian nuclear power program to obtain the fissile material. Rather, they have utilized enrichment facilities, plutonium-production. reactors, and, more recently, a research reactor, Moreover, as opposed to a deliberate (and possibly clande- stine) weapons-development program based upon a national decision, nuclear power programs are currently subject to international monitoring and influence in most cases. Thus while civilian nuclear power does represent one conceivable proliferation route, if it is made Tess attractive than other possible routes, proliferation concerns should not inhibit the development of commercial nuclear power. Proliferation concerns regarding civilian nuclear power programs center on two intrinsic characteristics of the nuclear fuel cycle. First, nuclear reactor fuel inherently provides a potential source of fissile material from which production of weapons-grade material is possible. Second, certain fuel cycle components, particularly enrichment and reprocessing facilities, exacerbate the proliferation problem since they provide a technological capability which could be directed towards weapons development. The term "latent proliferation" has been coined by Feiveson and Taylor! to cover these characteristics of the nuclear fuel cycle which, although not pertaining directly to weapdns development, by their existence facilitate a possible future decision to establish such a capability. It should be noted that the problem of latent proliferation impacts even the “once- - through" low-enriched uranium (LEU) cycle currently employed in light-water reactors (LWRs) and also the natural-uranium cycle utilized in the Canadian heavy-water systems (CANDUs). The technology requiréd to enrich natural uranium to LWR fuel represents a technological capability which could be redirected from peaceful purposes. In addition, the plutonium- containing spent fuel, albeit dilute and contaminated with highly radioactive fission products, represents a source of potential weapons material. Thus the possibility of proliferation exists even for the fuel cycles now in use. This has already been recog- nized and it has been proposed!»2 that internationally controlled fuel cycle service centers be established whose purpose would be ‘to preclude subversion of sensitive techno]ogy (such as ‘enrichment technology) and to provide facilities for the assay and secure storage of spent once-through reactor fuel. S 2-4 The establishment of such fuel cycle service centers is currently receiving serious consideration. As the costs of U303 production increase (and as it is preceived that long- term reliance on nuclear power is necessary), the expansion of the fuel cycle service center to include reprocessing activities will become attractivé.'.The expansion:would allow the 235y remaining in the spent fuel to be utilized. It would also allow the artificial (that is, "manufactured") fissile isotopes produced as a direct result of the power production process to be recycled. Of the latter, only two possible candidate isotopes exist: 233py and 233, In considering these two isotopes, it appears that the proliferation aspects of their possible recycle scenarios are considerably different. In fact, the rationale for the present study is the need to determine whether 233|)-based recycle scenarios have significant proliferation-resistant advantages compared with plutonium-based recycle scenarios. 2.1. INTERNATIONAL PLUTONIUM ECONOMY Prior to President Carter's April 7, 1977, nuclear policy statement, the reference recycle fuel scenario had been based on plutonium, referred to by Feiveson andrTaylor'1 as the "plutonium economy." In this scenaric the plutonium generated iz the LEU cycle would be recycled as feed material first into thermal reactors and later into fast breeders, these reactors then operating on mixed Pu/U oxides instead of on uranium oxide alone. As with any recycle scenario, the plutonium-based nuclear power economy would require the | operation of spent fuel reprocessing facilities. If dispersed throughout the world, such reprocessing technology, 1ike uranium enrichment technology, would markedly increase the latent proliferation potential inherent in the nuclear fuel cycle. Of course, such facili- ties could also be restricted to the fuel cycle service centers, However, the plutonium recycle scenario introduces a far greater concern regarding nuclear pro1iferation since weapons~-usable material can be produced from the fresh mixed oxide fuel through chemical separation of the plutonium from the uranium, whereas to obtain weapons-usable material from LEU fuel requires isotopic enrichment in 23350, Since the fresh mixed oxide (Pu/U) fuel of the reference cycle is vulnerable to chemical separation, not only are the fuel fabrication facilities of the cycle potential sources of directly usable weapons material, but also the reactors themselves. While restriction of mixed oxide fabrication facilities to safeguarded centers is both feasible and advisable, it is unlikely that the reactors can be centralized into a few such internationally con- trolled centers. Rather they will be dispersed outside the centers, which will necessitate that fresh fuel containing plutonium be shipped and stockpiled on a global scale and that it be safeguarded at all points. Thus, as pointed out by Feiveson and Taylor,! the plu- tonium recycle scenario significantly increases the number of nuclear fuel cycle facilities which must be safeguarded. The prospect of such widespread use of plutonium and its as- sociated problems of security have led to an examination of possible alternative fuel cycles aimed at reducing the proliferation risk inherent in recycle scenarios. One such alternative fuel cycle is the denatured 233U fuel cycle which comprises the subject of this report. —_—d € T oy oty ) o % e —— 4 4 e g ' } .7 i 2.2. THE DENATURED 233y FUEL CYCLE In the denatured 233U cycle, the fresh fuel would consist of a mixture of fissile 233y diluted with 238U (the denaturant) and combined with the fertile isotope thorium. The pre- sence of a significant quantity of 238U denaturant would preclude direct use of the fissile material for weapons purposes even if the uranium and thorium were chemically separated. As in the LEU cycle, an additional step, that of isotopic enrichment of the uranium, this time to increase its 233y concentration, would be necessary to produce weapons-grade material, and the development of an enrichment capability would require a significant decision and com- mitment well in advance of the actual diversion of fissile material from the fresh fuel, This is in contrast to the reference Pu/U fresh fuel for which only chemical separation would be required. Moreover, even if such an enrichment capability were developed, it would ap- pear that enriching clandestinely obtained natural uranium would be preferable to diverting and enriching reactor fuel, whether it be denatured 233U or some other type, since the reactor fuel would be more 1nternat1ona11y "accountable.” The primary advantage of the denatured fuel cycle is the inclusion of this "isotopic barrier" in the fuel. Whereas in the plutonium cycle no denaturant comparable to 238U exists and the fresh fuel safeguards (that is, physical security, international monitoring, etc.) would all be external to the fuel, the denatured 233U fuel cycle would incorporate an in- herent safeguard advantage as a physical property of the fuel itself. Like the plutonium cycle, the denatured fuel cycle would require the development of fuel cycle centers to . safeguard sensitive fuel cycle activities'such as reprocessing (but not necessarily refabri- cation). However, unlike the plutonium cycle, the denatured fuel cycle would not require the extension of such stringent safeguard pfoeedures to the reactors themselves, and they are the most numerous component of the nuclear fuel cycle. (As noted above, LEU fuel is also “denatured" in the sense that a low concentration of 235U is included in a 238U matrix. Similarly, natural uranjum fuel is denatured. Thus, these fuels also have the proliferation- resfstance advantages‘of the isotopic barrier.) The eoncept of dehetured 233y fuel as a proliferatioh-resistant step is addressed principa11y at the front end of the nuclear fuel cycle, that is, the fresh fuel charged to reactors. The 238y denaturant will, " of course, produce p1utonium under irradiation. Thus, as in the LEU and mixed oxide cycles, the spent fuel from the denatured cycle is a potential source of plutonium, However. also as in the LEU and mixed oxide cycles, the plutonium generated in the sbeht fuel is contaminated with high1y radioactive fission products. Moreover, the quantity of pluton1um generated via the denatured fuel cycle will be signif- ~ {cantly less than that of the other two cyc]es. Further, the decision to use ‘spent reactor fuel as a source of weapons mater1a1 requires a previous commitment to the deve]op- ‘ment of shielded extraction facilities. In summary, the use of a denatured fuel as a source of weapons material implies one of two strategic decisions: the development of an isotopic enrichment capability to Process‘diverted fresh fuel, or the development of a, fis- sile extraction capability (chemical or isotopic) to process diverted spent fuel. In 2-6 contrast, while the plutonium cycle also would require a strategic decision concerning the spent fuel, the decision to utilize the fresh mixed oxide fuel would be easier and thus would be more tactical in nature. A subsidiary proliferation-related advantage of the denatured fuel cycle is the presence of 232y (and its highly radioactive decay daughters) in the fresh fuel. The 232, an unavoidable byproduct in the production of 233U from 232Th, constitutes a chemically inseparable radioactive contaminant in the fresh fuel, which would be alfurther deterrent to proliferation. Similar contamination of mixed Pu/U oxide fuel has been proposed via "spiking" the fuel with fission products or preirradiating it to produce the fission products in situ, but both these options would involve signifitant perturbations to the Pu/238U fuel cycle as opposed to the "natural" contamination of thdrium-based fuels, Additiona]ly; the artificial spike of mixed oxide fuel would be subject to chemical elimination, albeit re- quiring heavily shielded facilities. The natural spike of the denatured fuel (that is, the 232y decay daughters) would also be subject to chemical elimination, but the continuing decay of the 232U would replace the natural spike within a limited period of time. 233y also has the advantage of a higher fissile worth in thermal reactors than 239y, both in terms of the energy release per atom destroyed and in terms of the conversion ratio (see Section 4.0). Commercial thermal reactors are currently available and are projected to enjoy a capital cost advantage over proposed fast breeder reactors. Additionally, the technological base required for installation and operation of a thermal system'is Tess sophisticated than that for fast systems such as LMFBRs. Thus it appears likely that near- term scenarios will be dominated by current and proposed therma) systems. In considering possible replacement fissile materials for the limited 235U base, the worth of the replace- ment fuels in the thermal systems is of some importance. One important factor which must be considered in discussing the dénatured fuel cycle is the potential source of the required fissile material, 233U, It appears likely that current-generation nuclear power reactors operating on the denatured cycle will require an external source of 233U to provide makeup requirements. Moreover, even if future de~ natured reactors could be designed to be self-sufficient in terms of 233U, there would still remain the question of the initial 233U loading. One possible source of the required 233U is a 233y production reactor located in the fuel cycle service center (now perhaps more accurately termed an energy center). This system would be fueled with plutonium and would both produce power and transmute 232Th into 233U, which could then be denatured for use out- side the secure energy center. Loosely termed a transmuter, such a reactor would be con- strained to the energy center because of its utilization of plutonium fuel. The required plutonium for the transmuters is envisioned as coming initially from reprocessed LEU fuel, and later, in the more mature system, from plutonium produced in energy-center reactors or via the 238 denaturant in dispersed reactors. Thus, in mature form a symbiotic system such as that depicted in Fig. 2.2-1 will evolve in which the energy center transmuters produce fuel (233U) for the dispersed reactors and consume the plutonium produced by the dispersed - ‘é = e ool - et d’j r ) 4[ ” ! | rcL T Ty et ool K ' — T . 1 £ —— ri. - £ 2-7 ‘denatured reactors or by energy-center reactors. The dispersed reactors in turn are provided a source of 233U for initial loading and makeup requirements, as well as a means for disposing of the non-recyclable (in the dispersed reactors) plutonium. The significant point of such a system is that no plutonium-containing fresh fuel circulates outside the energy center. The plutonium contained in the spent fuel 1s returned to the center for ultimate destruction. ORNL-DWG TT-1C0T1 ‘Makeyr | In, U238 DenaTurep FueL AssemBLies (NO Pu) NERGY \ CENTER (U-233, U-238, Tu-232) Oxipe ' BounDarY Dispersep REACTORS i LWR - E IRRADIATED FueL U ] SPENT FuEL FU_,1 Hot FueL o Pu To U-233 E 22:C:fi$gn " (Pu+U-233) ProcessING —LB—st FaBRICATION gu TRANSMUTER | § . - iH " . F1ssioN E FBR , RODUCTS Pu, T ,U-233 g MASTE IRRADIATED FUEL E k| FIXATION i EE TERMINAL E ¢ | STORAGE g Fig. 2.2-1, Schematic Fuel Flow for Symbiotic System Consisting of an Energy Center and Dispersed Reactors Operating on Denatured 233U Fuel. One obvious concern regarding such a coupled system is the amount of power produced by the dispersed systems: relative to that produced in the energy center reactors. The power ratio,* defined as dispersed power generated relative to centralized power, can be viewed as a parameter characterizing the practicality of the system. While the power "ratio depends on the characteristics of the reactors actually utilized for the various components and is considered in detail later in this report, certain generic statements - can be made. In a mature "safeguarded" plutonium cycle, the ratio would be zero since all reactors would, of necessity, be located in enefgy'centers. In the current open-ended LEU cycles, this ratio 1S'eSSéntia11y infinite since current nuclear generating capacity is dispersed via "naturally denatured" thermal systems. The denatured 233U cycle will fall [4 *Also called "energy support ratio." 2-8 between these two extremes, and thus the proposed system's power ratio will be a crucial evaluation parameter. The symbiotic System depicted by Fig. 2.2-1 can also be characterized by the type of reactors utilized inside and outside the center. In general, systems consisting of thermal (converter) reactors only, systems consisting of both thermal converters and fast breeder reactors, and systems consisting solely of fast breeder reactors can be en- visioned.* One important characteristic of each system is the extent to which it must rely on an external fuel supply to meet the demand for nuclear-based generating capacity. The thermal-thermal system would be the most resource-dependent. The breeder-thermal system could be fuel-self-sufficient for a given power Tevel and possibly also provide for moderate nuclear capacity growth. The breeder-breeder scenario, if economically competitiVe with alternative energy sources, would permit the maximum resource-independent nuclear contribu- tion to energy production. While such considerations serve to categorize the symbiotic systems themselves, the transitidn‘from the current once-through LEU cycles to the symbiotic systems is_of more immediate concern. Although all-breeder systems would be resohrce-indepehdent, commercial deployment of such systems is uncertain. The transition to the denatured cycle could be initiated relatively soon, however, by using moderately enriched 235U/238U mixed with thorium (sometimes referred to as the "denatured 233y fuel cycle") in existing and pro- jected thermal systems. The addition of thorium (and the corresponding reduction of 238y over the LEU cycle) would serve a dual purpose: the quantity of plutonium generated would be significantly reduced, and an initial stockpile of 233 would be produced. It should be noted that this rationale holds even if commercial fuel reprocessing is deferred for some time. Use of denatured 235U fuel would reduce the amount of ptutonium contained in the stored spent fuel. In addition, the spent fuel would represent a readily accessible source of denatured 233U should the need to shift from 235U arise. However, substituting 232Th for some of the 238U in the LEU cycle would require higher fissile loadings and thus more 235() would be committed in a shorter time frame than would be necessary with the LEU cycle. An alternative would be to utilize energy-center Pu-burning transmuters to provide the initial source of 2330 for dispersed 233y-based reactors. From these starting points, various scenarios which employ thermal or fast energy-center reactors coupled with denatured thermal or fast dispersed reactors can be developed. On the basis of the above, eight general scenarios have been postulated for this study, with two sets of constraints on Pu'utiliiation considered: either plutonium will not be al- lowed as a recycle fuel but recycle of denatured 233U will be bermitted; or plutonium will be allowed within secure energy centers with only denatured fuels being acceptable for use at dispersed site reactors. The eight scenarios can be summarized as follows: B ' *See Section 4.0 for discussion of reactor terminology as‘app1ied in this study. oty — = e e Lé = ¥= ! ) ) ey | — i ., e - s - 2-9 1. Nuclear power is limited to low-enriched uranium-fueled {LEU) thermal reactors operat- ‘ing on a stowaway cycle (included to allow comparisons with current policy). 2. LEU reactors with uranium recyc1eiare operated outside secure energy centers and thermal Vreactors with plutonium recycle are operated inside the centers. 3. Same as Scenario 2 plus fast breeder reactors (FBRs) operating on the Pu/U cycle are - deployed within the centers. 4. LEU reactors and denatured 235U and denatured 233U reactors are operated with uranium recycle, all in dispersed areas; no plutonium recycle is permitted. 5. Same as Scenario 4 plus thermal reactors operat1ng on the Pu/Th cycle are perm1tted within secure energy centers. 6. Same as Scenario 5 plus FBRs with Pu/U cores and thorium blankets ("1ight" transmuta- tion reactors) are permitted within secure energy centers. ' 7. Same as Scenario 6 plus denatured FBRs w1th 233)1/238y cores and thorium blankets are perm1tted in d1spersed areas, ' 8. The "light" transmutation FBRs of Scenario 7 are replaced with "heavy" transmutation -reactors with Pu/Th cores and thorium blankets. 2.3. SOME INSTITUTIONAL CONSIDERATIONS OF THE DENATURED FUEL CYCLE As stated above, the implementation of the denatured fuel cycle will entail the . creation of fuel cycle/energy centers, which will require institutional arrangements to manage and control such facilities. The édvantages and disadvantages of such centers, whether they be regional, multinational, or international, as well as the mechanisms re- quired for their implementation, have been reported.3’* Although a detailed enumeration of the conclusions of such studies are beyond the scope of this particular discussion, certain aspects of the energy center concept as it relates to the denatured fuel cycle are relevant. Since only a few thousand kilograms of 233U currently exist, it is clear that production of 233y will be required prior to full-scale deployment of the denatured 233y cycle. If the reserves of economically recoverable natural uranium are allowed to become extremely limited before the denatured cycle is 1mp1emented, most if not all power pro- duced at that time would be from energy-center transmuters. Such a situation is clearly inconsistent with the principle that the number of such centers and the percentage of total power produced in them be minimized. A gradual transition in which 235U-based dispersed reactors are replaced with denatured 23%-based dispersed reactors and their accompanying energy-center transmuter systemsris thus desirable. The proposed denatured fueT-cyc1e/energy_center scenario also presents an additional ~ dimension in the formulation of the energy policies of national states - that of nuclear interdependence. By the very nature of the proposed symbiotic relationship inherent in 2-10 the denatured cycle, a condition of,mutua],dependence between the dispersed reactors and the energy-centef reactors is created. Thus while natjons choosing to operate only denatured (i.e., dispersed) reactors must obtain their fuel from nations that have energy-center trans- muters, the nations operating the transmuters will in turn rely on the nations operating dispersed réactors*for their transmuter fuel requirements (Pu). Hence, in addition to the possible nonproliferation advantages of the denatured fuel cycle, the concept also intro- “duces a greater flexibility in national energy policies. References for Chapter 2 1. H. A. Feiveson and T. B. Taylor, "Security Implications of Alternative Fission Futures," Bull. Atomic Scientdsts, p. 14 (Dec. 1976). : ' S 2. "A Report on the International Control of Atomic Energy," prepared for the Secretary of State's Committee on Atomic Energy by a Board of Consultants: Chester I. Barnard, Dr. J. R. Oppenheimer, Dr. Charles A. Thomas, Harry A. Winne, and David E. Lilienthal (Chairman), Washington, D.C., March 16, 1946, pp. 127-213, Department of State Pub- lication 2493. o 3. "Nuclear Energy Center Site Survey - 1975," Volumes 1-5, NUREG-0001, Nuclear Regulatory Commission, January, 1976. 4. "Regional Nuclear Fuel Cycle Centers," 1977 Report of the IAEA Study Project, STI/TUB-445, International Atomic Energy Agency, 1977. e - ¥y {”]: K- ! CHAPTER 3 ISOTOPIC CHARACTERISTICS OF DENATURED 233y FUEL r Chapter Qutline 3.0. Introduction, T. J. Burns and L. S. Abbott, ORNL 3.1. Estimated 232y Concentrations in Denatured 233U Fuels, D. T. Ingersoll, ORNL 3.2. Radiological Hazards of Denatured Fuel Isotopes, H. R. Meyer and J. E. Till, ORNL 3.3. 1sotopics Impacting Fuel Safeguards Considerations 3.3.1. Enrichment Criteria of Denatured Fuel, Cc. M. Newstead, BNL .3.2. Fabrication and Handling of Denatured Fuel, J. D. Jenkins and R. E. Brooksbank, ORNL ‘ . Detection and Assay of Denatured Fuel, p. T. Ingersoll, ORNL . Potential Circumvention of the Isotopic Barrier of Denatured Fuel, E. H. Gift and W. B. Arthur, ORGDP . Deterrence Value of 232U Contamination in Denatured Fuel, c. M. Newstead, ORNL r I 3 £ 2 3 4 .5 .3 T 1 C . - o J € .- rTi \ | i 3.0. INTRODUCTION T. J. Burns and L. S. Abbott Oak Ridge National Laboratory , An assessment of the denatured 233U fuel cycle - both for meeting the requirements for electrical power growth and for reducing the risks of nuclear weapons proliferation - invariably must include an examination of the isotopics of the cycle. It has been pointed out in Chapters 1 and 2 that the concept of the denatured 233U cycle is an attempt to retain the isotopic barrier inherent in the currently used LWR low-enriched 235U (LEU) cycle but at the same time to allow the production and recycling of new fuel. In both the denatured and the LEU cycles the isotopic barrier is created by diluting the fissile isotope with 238U, so that the concentration of the fissile nuclide in any uranium chemical- 1y extracted from fresh fuel would be sufficiently low that the material would not be directly usable for weapons purposes. This is in contrast to the two reference fuel cycles, the Pu/U cycle, and the HEU/Th cycle. In both of these cycles, weapons-usable material could be extracted from the fresh fuel via chemical separation. Of course, as shown in Table 3.0-1, chemically extractable fissile material is present in the spent fuel elements of all these cycles; however, the spent elements are not considered to be particularly vulnerable because of the high radioactivity emitted by the fission products - at least initially. ~In this assessment of denatured 233U fuel, the implications of substituting the denatured fuel for the reference cycles of various reactors are examined. In addition to the obvious advantage of the isotopic barrier in the fresh fuel, denatured 233U fuel has an additional proteétion factor against diversion in that its fresh fuel is radioactive to a much greater extent than any of the other fuels listed in Table 3.0-1. This characteristic is due to the presence of the contaminant 232U, which is generated as a ‘byproduct of the 233U production prbcess and which spawns a highly radioactive decay chain. As shown in Fig. 3.0-1, 232U decays through 228Th to stable 208Pb, emitting numerous gamma rays in the process, the most prominent being a 2.6-MeV gamma ray associated with the decay of 20871, | IR o o ‘Table 3.0-1. 'Comparison of Principal Fissile and Fertile Nutlides,in Some .Reactor Fuels Fuel Fresh Fue] Nuclides : | ',Spent Fuel Nuctides Denatured 233U fuel 233y, 238y, 2321, 233y, pyf, 238y, 23271). (with recycle) S : - . LEU (no recycle) 235y, 238y - | 235y, puf, 238y LEU (with recycle) 235y, pyf, 238y 235y, puf, 238y ~ Pu/U (with recy¢1e)' puf, 238y - Puf, 238y 'HEU/Th (no recycle) 235y, 232Th 233y, 235y, 2327 ORNL-DWG 65-550R3 232, al2y ZZUTh o ol9 y B 6.43n 2327, 228, 224, #1.39210° y BS.TSy e364d 2283, 220g, _ 212p, a%45s ac/l- 30412107 z»c 212g; 208py, 606m {stoble} «01583 B %% am 06 h 1 ' ' : 2izpy, 2087) (2.6-MeV y) Fig. 3.0-1. Decay of 232y, In assessing the safeguard features of denatured 233U fuel, While the 232U contamination will be essentially must be examined from several viewpoints. an inherent property of the.denatured fuel cycle, the concentration of the isotopic denaturant, 238y, is controllable. the denatured fuel cycle. venting the intrinsic isotopic barrier is increased. The radioactivity associated with the 233y significantly impacts the associated fuel cycle.. The fabrication, shipping, and handling of the fresh denatured fuel is expected to “differ markedly from the other cycles, primari]y. due to the fact that remote procedures will have to be employed throughout. To design the " necessary facilities will require a knowledge of the concentrations of 232U (and its daughter products) in the fuel as a function of time. To date, insufficient data are available on this subject, but on the basis of some pre- liminary investigations some estimates are given in Section 3.71:on the 232U concentrations that could be expected in the recycled fuel of LWRs, HTGRs, and FBRs operatlng on denatured 233y, The radiological hazards associated with the use of denatured 233U fuel represent another aspect of the cycle demanding attention. Again Tittle information is available, but Section 3.2 discusses the toxicity of the various isotopes present in the fuel and also in thorium ore, as well as the effects of exposure to the gamma rays emitted from the fresh fuel. the isotopics of the,cycle The presente of both isotopes affects the proliferation potential of As the 238U concentration is increased, the difficulty of circum- However, increasing the 238U fraction also increases the 23%Pu concentration in the spent fuel so that an obvious trade-off of proliferation concerns exists between the front and back ends of the denatured fuel cycle. As pointed out in Section 3.3.1, being formulated. the enfichment criteria for denatured 233U fuel are still The requirement for remote operat1ons throughout the fuel cyc]e will 1n 1tse1f constitute a safeguard feature in that access to fissile material will be difficult at all | But this requ1rement will also be a complicating factor in the des1gn stages of the cycle. of the fuel recyc]ing steps and operations. Chapter 5, but Section 3.3.2 of this chapter points out that the remote operation requirement could dictate the se]ect1on of techn1ques, as, for example, for ‘the fuel fabrication process. This subject is treated in more detail in r— v' L; \ 1 b r— v ‘.l " £ . 4 ¥ ] ¥ ! " =y ¢ The radioactivity of the 232U chain would also make it easier to detect diverted de- natured fuel and would complicate both the production of weapons-grade 233y from fresh denatured fuel and its subsequent use in an expioéive‘device. On the other hand, as discussed in Section 3.3.3, the radioactivity will inhibit passive, nondestructive assays for fissile accountability. Finally, the possible circumvention of the isotopic barrier must be addressed. In Section 3.3.4 it is postulated that a gas centrifuge isotope separation facility is avail- able for isotopically enriching diverted fresh denatured 233y fuel, and estimates are made of the amounts of weapons-grade material that could be so obtained. Conclusions are then drawn as to the relative attractiveness of denatured 233U fuel and other fuels to would-be diverters. 3.1. ESTIMATED 2°2U CONCENTRATIONS IN DENATURED *°°U FUELS D. T. Ingersoll Oak Ridge National Laboratory Although it is mandatory that the concentrations of 232|) at each stage of the fuel cycle be predictable for the various reactors operating on thorium-based fuels, little ihformation on the subject is available at this time. This is attributable to the fact that the interest in thorium fuel cycles is relatively recent and therefore the nuclear data requiréd for calculating the production of **?U have not been adequately developed. Of primary importance are the (n,Y) cross sections of 23'Pa, 23°Th, and #3*2Th and the (n,2n) cross sections of 233U and %32Th, all of which are intermediate interactions that can lead to the formation of 232U as is illustrated by the reaction chain given in Fig. 3.1-1. These cross sections are under current evaluation® and should appear in the Version V release of the Evaluated Nuclear Data File (ENDF/B-V). ORNL-DWG 77-15745 ) C . In spite of the nuclear data deficien- B (22 m) 87 (27d) i ] “5or Z53u cies, some results for 232U concentrations ‘ ‘ are available from calculations for denatured (ny) (n2n) fuels in light-water reactors (LWRs) and in - fast breeder reactors {FBRs). Although no 232 232 B3 d 232 o [:EEE:] [:E;EE:}-—-—-—-—*{_EQL_] results for denatured high-temperature gas- ' cooled reactors (HTGRs) are currently available, (n,2n) (n,7) 232y concentrations can be roughly inferred from existing HTGR fuel data. Moreover, the 1 py B(255h) : L [ 2 2pa analysis of 232U concentrations in standard A HTGR designs (HEU/Th) serves as an upper (n.y) bound for the denatured systems. A compila- tion of the available results is given below. |Zg@“ I The current state of the related 232U nuclear. data is amply reflected in the large variances Fig. 3.1-1 Important Reaction Chains of the calculated concentrations. Leading to the Production of 232y, 3.1.1. nght-water Reactor Fue1s Existing data on 232U concentrations in denatured LWR fuels are pr1mar1]y from cal- ' culations based on the Combustion Engineering System 80 ™ reactor de51gn.2“Resu1ts,from CE3 for a denatured 235U cycle (20% 235U-enriched uranium in 78% thorium) show the 232U concentration after the zeroth generation to be 146 ppm 232U in uranium, while after . five generations of recycle uranium, the concentration is increased to 251 ppm. These levels are in godd agréement with ORNL ca]cu]ations,“ which indicate 130 ppm 232U in uranium for the zeroth generation. The discharge uranium isotopics are summaruzed in Table 3.1-1. Also shown are the resu]ts from an ORNL calculation for a denatured 233U cyc]e = — { i A t b | ! . r e 3-7 (10% 232U-enriched uranium in 78% Th). The slight contribution from 233 peactions in- creases the 22U content to 157 ppm after the zeroth generation. Table 3;1e1. Discharge Isotopics for LWRs Operéting on Denatured Fuels Isotopic Fraction 2326 in U Cycle 232 233y 234y 235y 236 ‘éasu 232Th (ppm) 235/Th Fuel® | | cE(0)? 0.0029 1.07 0.11 1.5 0.50 16.81 76.21 146 ORNL(0) 0.0026 1.00 0.09 1.59 0.49 16.85 76.23 130 CE(5) 0.0061 1.60 0.69 1.27 1.86 18.78 75.79 251 233y/Th Fuel® | | ORNL(0) ~ ©0.0031 1.16 0.29 0.056 0.0052 18.32 75.99 . 157 IInitial isotopics: 4.4% 235U, 17.6% 2%°U, 78% 232Th. PThe number in parentheses represents the fuel ganeration number, “Initial isotopics: 2.8% 233y, 19.2% 238y, 78% 232Th, 3.1.2. High-Temperature Gas-Cooled Reactor Fuels _Although calculations for 232y concentrations in denatured HTGR fuels are not avail- able, it is possible to roughly infer this information from existing HTGR calculations if the expected changes in the thorium content are known. The conventional HTGR cycle begins with 93% 235U-enriched uranium fuel and thorium fertile material. On successive cycles, the 233U produced in the thorium is recycled, thus reducing.the required amount of 235U makeup. The 232y content of the recycled fuel becomes appreciable after only a few genera- tions. Table 3.1-2 gives the uranium isotopics of the recycle fuel batches at the beginning of recycle and at equilibrium recycle,> the latter showing a maximum 232U concentration of 362 ppm in uranium, = SRR S ‘ Table 3.1-2., Uranium Isotopics for Commercial HTGR Recycled Fuel (HEU/Th) Isotopic Fraction ) 3 233 234 235 236 % iny 232 - u - 233y u U | ._u (ppm) Beginning 0.000126 0.921 - 0.0735% 0.00568 0.000245 126 of recycte - ' i ' _ , . Equilibrium 0,000362 0.614 . 0,243 - 0.0802 0.0630 362 recycle 3-8 The values in Table 3. 1-2 are a result of a standard HTGR fuel composition which has an average Th/233U ratio of about 20, Preliminary estimates have been made of dena- tured HTGR fuels which assume a 20% denatured 235U, leading to a 15% denatured 233U,6 Because of the added 238y fertile material, the amount of thorium is correspondingly re- duced by about 30%, resultifig in a similar reduction in the 232U production. The con- centration of 232U in total uranium would also be reduced by the mere presence of the diluting 238U, so that it can be estimated that a 15% denatured 233U HTGR would contain approximately 40 ppm 232y in uranium after equilibrium recycle. The lower 232U levels in the HTGR are primarily due to a softening of the neutron energy spectrum compared with that of the LWR. This results in a marked reduction in the 232Th(n,2n") reaction rate, which is a prime source of 232y, | | : ' ' 3.1.3. Fast Breeder Reactor Fuels _ 232y concentrations calculated by Mann and Schenter? and by Burns® for various commercial-sized FBR fuel cycles are given in Table 3.1-3. Except for Case 2, these values were determined from reaction-rate calculations using 42 energy groups and one- dimensional geometry; the Case 2 results were determined from a coarse nine-group two- dimensional depletion calculation. It is important to note that Cases 1 and 2 represent the "transmuter" concept. All the discharged uranium (232y, 233y, 23%, and 235U) is bred from the 232Th initially charged and consists principally of 233U, This accounts for the high 232U/U ratio, which will be reduced by a factor of 5 to 8 in the denatured fuel manufactured from this mate- rial. Thus, denatured fuel generated via the fast Pu/Th transmuter is expected to have approximately 150-750 ppm 232U in uranium. Table 3.1-3. FBR Core Region 232U Discharge Concentrationsa 2324 in U (ppm) Case No. Fuel t=1yrP t=2yr t=3yr t=5yr No recycle 1 10% 239y in Th 982 1710 2380 3270 2 11% 239y in Th 1106 2376 - 3670 3 102 233y in Th 288 830 - 1330 2210 4 10% 233U in 238y 6.6 0.7 125 13.3 With recycle : ‘ 5 10% 233y in Th 1820 2760 3260 6 10% 233U in 238Q 35 35 35 %Cases 1, 3-6 are from ref. 7; Case 2 is from ref. 8. bt = fuel residence time for no recycle cases; t = burning time before recycle for recycle cases. , . — . o T e e r— - o T T e . x r . T | - "o ety (‘s oD oo ) r .7 KT T 3-9 The last two cases in Table 3.1-3 give the equilibrium 232y concentrations assum- ing recycle of the 233U and the associated 232y, It should be noted that these two cases represent the extremes regardingfaliowab1e enrichment (233U/U). For a 20% denatured fuel in which approximately half the heavy metal is 232Th, the expected 232U equilibrium con- centration would be ~ 1600 ppm (232U/U) for a 3-yr cycle residence time. 3.1.4. Conclusions The results presented in this section are, for the most part, preliminary and/or approximate. This is largely a consequence of the uncertainties in the anticipated fuel compositions, denaturing limits, recycle modes, etc., as well as the basic nuclear data. Also, the results assumed zero or near-zero 230Th concentrations, which can approach signi- ficant levels depending cn the source of the thorium stock, particularly in thermal sys- tems. Because of the relevant cross sections, the presence of even small amounts of 230Th can result in considerably higher 232U concentrations. It is possible to conclude, how- ever, that 232 concentrations will be highest for 233U-producing FBRs, increase with fuel recycle, and decrease with fissile denaturing. References for Section 3.1 1. Summary Minutes of the Cross Sect1on Evaluation Working Group Meeting, May 25-26, 1977. 2. N. L. Shapiro, J. R. Rec, R. A. Matzie, “AsSessment of Thorium Fuel Cycles in Pressurized-Water Reactors," ERRI NP-359, Combustion Engineering (1977). 3. Private communication from Combustion Engineering to A. Frankel, Qak Ridge National Laboratory, 1977. 4, Private communication from W. B. Arthur to J. W. Parks, Oak Ridge National Laboratory, August 11, 1977. 5. J. E. Rushton5 D. Jenkins, and S. R. McNeany, "Nondestructive Assay Techniques for Recycled 330 Fuel for H1gh -Temperature Gas- Coo1ed Reactors,' J._Inst;tute Nuclear Materials Management IV(]) (1975). 6. M. H. Merrill and R. K. Lane, "Alternate Fuel Cycles in HTGRs,“ Draft for Joint Power Generation Conference, Long Beach, Ca11f., 1977. 7. F. M. Mann and R. E. Schenter, "Production of 232U in a 1200 MWe LMFBR," Hanford Engineering Development Laboratory (June 10, 1977). 8. T. J. Burns, private conmunication. 3-10 3.2. RADIOLOGICAL HAZARDS OF DENATURED FUEL ISOTOPES H. R. Meyer and J. E. Till Qak Ridge Nationa! Laboratory - Consideration of the denatured 233U cycle has cfeated the need to determine the radiological hazards associated with extensive use of 233 as a nuclear fuel. These hazards will be determined by the toxicity of the various isotopes present in the fuel and in thorium ore, which in turn is infiuenced'by‘the path through which the iSotopes enter the body--that is, by inhalation or ingestion. In addition, the gamma rays emitted from the denatured fuel present a potential hazard. 3.2.1. Toxicity of 22U and 232y Only limited experimental data are available'on the toxicity of high specific activ- ity uranium isotopes such as 2% and 2%2U. Chemical toxicity, as opposed to radiclogical hazard, is the Timiting criterion for the long-Tived isotopes of uranium (?°°U and 238y) which are of primary concern in the light-water reactdr uranium fuel cycle.! In order to establish the relative radiotoxicity of denatured ?°°U fuel, it is helpful to consider specific metabolic and dosimetric parameters of uranium and plutonium isotopes. Table 3.2-1 lists several important parameters used in fadio]ogical dose calculations. The effective half 1ife for 23%Pu in bone is approximately 240 times that of uranium. How- ever, the effective energy per disintegration for 232y is about three times greater than that for any of the plutonium isotopes. In general, ‘the time-integrated dose from‘ plutonium isotopes would be significantly greater than the dose from uranium isotopes for the inhalation pathway, assuming inhalation of eqUal‘act1Vities of each radionuclide. Doses via the ingestion pathway, again on a per nCi basis, are much lower than those esti- mated for the inhalation pathway. It is currently assumed that all bone-seeking radionuclides are five times more- effective in inducing bone tumors than 22°Ra. However, the Timited numbef of studies that have been conducted with 233U (ref. 2) and 232U (refs. 3-5) suggest a reduced effectiveness in inducing bone tumors for these isotopes and may result in use of exposure limits that are less restrictive than current limits. - The last two columns in Table 3.2-1 represent'dose'conversion_factors (DCFs) for uranium and plutonium isotopes calculated on the basis of mass rather than activity. It may be seen that the #32U "Mass DCFs" are more than four orders of magnitude greater than those for fissionable 2°%U, due largely to the high specific activity of 2*2U. This factor contributes to the overriding importance of #32U content when considering the radiotoxicity of denatured uranium fuels. Figure 3.2-1 illustrates the’importance of 232y content.withfreépectfito‘potential toxicity of 233U fuel. This figure presents the estimated dose commitment to bone calcu-- | S 2, ol oo = o s ) h E: o = Table 3.2-1. Metabolic Data and Dose Conversion'FactOrs (DCFs) for Bone for Selected Uranium and Plutonium Isotope Effective Half Activity Dose Conversion Factor Mass Dose Conversion Factor éM? Isotope Speciggfig?ctivity Life in Bone? Tnhalation® Ingestiond Tnhalationc Ingestionc [ (Days) (rems/uCi) (rems/uCi) (rems/ug) (rems/ug) - 232y 21.42 3.00 x 102 1.1 x 102 4.1 x 10° 2.4 x 103 8.8 x 10! L 233 9.48 x 10-° 3.00 x 102 2.2 x 101 8.6 x 10! 2.1 x 1071 8.2 x 10°3 235y 2.14 x 1076 3.00 x 102 2.0x 10! 8.0 x 10-1 4.3 x 1075 1.7 x 10-6 1; 238y 3.33 x 1077 3.00 x 102 1.9 x 10! 7.6 x 10! 6.3 x 1076 2.5Ix 10-7 _ 238py 17.4 2.3 x 10% 5.7 x 103 6.8 x 1071 9.9 x 10% 1.2 x 10} L 239py 6.13 x 10-2 7.2 x 104 6.6 x 10 7.9 x 10-! 4.0 x 102 4.8 x 10-2 240py 2.27 x 1071 7.1 x 10% 6.6 x 103 7.9x 1071 1.5x 103 1.8x 107! “International Commission.on Radiological Protection, "Report'of Committee II on Permissible Dose for Internal Radiation," ICRP Publication 2, Pergamon Press, New York, 1959. b Ki]!ough, g. G., and L. R. McKay, "A Methodology for Calculating Radiation Doses from Radioactivity Released to the Environment," ORNL-4992, 1976, “Product of specific activity and activity dose conversion factor. lated for inhalatioh'of 107!2 g of unirradiated 233U HTGR fuel (n93% 2%3U/U) as a function of the 232U impurity content for two different times following separation at a reprocessing facility. The upper'curve is the dose commitment at 10 years after separation. Two basic conclusions can be drawn from these data. First as recycle progresses and concentrations of 232y become greater, the overall radiotoxicity of 233U fuel will increase significantly. Second, the ingrowth of 232U daughters in 223U fuel increases fuel radiotoxicity signifi- cantly for a given concentration of 232y, Although the data graphically illustrated in Fig. 3.2-1 were not specifically calculated for denatured 233 fuel, the required data not being available, the relative shape of_the_curves'would remain the same. Al]_e]Se being equal; the estimatéd radiotoxicity of denatured fuel would be reduced due to dilution of 233y and 2320 with 22%U, which has a low radiological hazard. A comparison of the dose commitment to bone resulting from inhalation of 107'% g of three types of fuel, HTGR 233U fuel, LWR 2%5U fuel, and FBR plutonium fuel, is given in Fig. 3.2-2. This analysis evaluates unirradiated HTGR fuel containing 1000 ppm 232y and does not consider fission products; activation products, transplutonium radionuclides, or environmental transport. As shown in Table 3.2-1, the inhalation pathway would be by far the most signifiéant for environmentally dispersed fuels. Therefore, other potential pathways of exposure are not considered in this brief analysis. 3-12 f 3 ORNL -=DWG 75-3172R3 SO T T T T T T T TITEE L T | | FBR PLUTONIUM FUEL (Reference Cycle) e i | | 400 — P — — MAXIMUM ANTICIPATED -] — 232 CONCENTRATION \ — “’“ N 107 = — 10 years FOLLOWING SEPARATION ] | / 90 days FOLLOWING SEPARATION - / | L rm C o / / RECYCLED FUEL WITH NO (0 ppm) 2320 . | L RN L 10! 10% 10° | 232\, \N RECYCLED HTGR FUEL (ppml DOSE COMMITMENT TO BONE FROM INHALATION OF 10'2 g OF FUEL (mrem) \\I T LT LU | ol | - 0 o Fig. 3.2-1. Effect of 232U Concentrations in HTGR Fuel (93% 233y/) on Dose Commi tment to Bone. . : ( | S o ( T ooy Ty €y L0 1 - Tl 3-13 1 ORNL-DWG 75-2938R3 10 SRR RN R R Ifllg : . ] — € | FBR PLUTONIUM FUEL ,1 _ E 100 — — e — s - ——— . < ] I - ] < \ T 107 = . = g E = @ — 93% 23y FUEL — W | O — - o . /,//””’—d__*‘\\\\\\\ \ % =~ \ = o — = w — _ a - —_ —_ Z 10 = = 0 — = 2 — — e . — & 104 N //\ = \ —— = — = s — — g L —_ O _ ;;:‘ = 107 = LWR URANIUM FUEL — — - ] . _ _ , | — 108 L |»Hl||JI l | LU -J OO L e 1 o' 10° 10 102 o> 1t 10° 108 Fig. 3.2-2. TIME AFTER SEPARATION (years) Relative Radiotox1c1ty of FBR Plutonium Fuel, HTGR Fuel (93% 233u/u) and LWR Uranium Fuel as a Function of the Time after Separation at Reprocess1ng Plant. It is noted that Fig. 3.2-2 applies to fresh fuel as a function of time after separation, presuming it has been released ‘to the environment. from the resuspens1on of rad1oact1ve materia]s deposited on terrestrial surfaces A commitment curve for denatured 2°° fuel would be expected to lie slightly below the given curves for HTGR fuel; however. the denatured fuel would rema1n s1gn1f1cantly more hazardous from a radiological standpoint than LWR uran1um fuel. Inhalation Tong after release could result dose 3.2.2 Toxicity of 232Th Given the potential for radio]ogiéai hazard via the mining of western U.S. thorium deposits as a result of implementation of ?32Th-based fuel cycles, current difficulties in estimation of 2*2Th DCFs must also be considered here. As is evident in Fig. 3.0-1 (see Section 3.0), both 2°2U and 2°2Th decay to 22%Th, and then through the remainder of the decay chain to stable 2°%Pb. 232 decays to 2%2Th via a single 5.3-MeV alpha emission; 232Th decays via three steps, a 4.01-MeV alpha emission to ??%Ra, followed by serial beta decays to ?2°Th. The total energy released in the convergent decay chains is obviously nearly equal. The ICRP7 lists effective energies (to bone, per disintegration) as 270 MeV for 232Th and 1200 MeV for 232U; these effective energies are critical in the determination of dose conversion factors to be used in estimation of long-term dose commitments. The large difference between the effective energies calculated for the two radionuclides is based on the ICRP assumption (ref. 7) that radium atoms produced by decay in bone of a thorium parent should be assumed to be released from bone to blood, and then redistributed as though the radium were injected intravenously. As a result, the presence of 228Ra in the 232Th decay chain implies, under this ICRP assumption, that 90% of the 22°Ra created within bone is eliminated from the body. Therefore, most of the potential dose from the remaining chain alpha decay events is not accrued within the body, and the total effective energy for the 232Th chain is a factor of 4.4 lower than that for 232U, as noted. Continuation and reevaluation of the early research®’® leading to the above dis- similarity indicated that the presumption of a major translocation of 22®Ra out of bone was suspect (refs. 10-14), and that sufficient evidence existed to substantiate retention of 97% of 228Ra in bone. Recalculation of effective energies for the 232Th chain on this basis results in a value of 1681 MeV as listed in ERDA 1451 (ref. 15}, a substantial increase implying the need for more restrictive 1imits with respect to 232Th exposures. In con- trast to this argument, the 1972 report of an ICRP Task Group of Committee 2 (ref. 16) presents a newly developed whole-body retention function for elements including_radium which effectively relaxes 2®2Th exposure 1limits. 3.2.3 Hazards Related to Gamma-Ray Emissions While fuel fabricated from freshly separated 23°U emits no significant gamma radia- tion, ingrowth of 232y daughters Teads to buildup of 2°®T1 2.6-MeV gamma radiation, as well as other gamma and x-ray emissions. As discussed elsewhere in this report, it is anticipated that occupational gamma exposures during fuel fabrication can be minimized by such techniques as remote handling and increased shielding. | | oy 4 i v | i ' o ool e ‘{j"“ r— - o ] i L Gamma exposure resulting from the transportation of irradiated fuel elements con- taining 232U will not be significantly different from that due to other fuels. Shielded casks would be used in shipment to control exposures to the public along transportation routes. Gamma exposure from 222U daughters would be insignificant compared to exposure from fission products in the spent fuel. Refabricated fuel assemblies containing ?°%U would require greater radiation shielding than LWR fuel. However, this problem can be minimized by shipping fresh assem- blies in a container similar in design to a spent fuel cask. Gamma doses to workers and to the general public due to transport of fuel materials between facilities are therefore expected to be easily controlled, and have been estimated to be low, perhaps one man-rem per 1000 Mi(e)} reactor-plant-year.?'® The estimated -gamma hazard of environmentally dispersed 2%2U, while a significant contributor to externally derived doses, is overshadowed as a hazard by the efficiencies of internally deposited alpha emitters in delivering radiological doses to sensitive tissues. ' 3.2.4. Conclusions = - Several conclusions can be made from this assessnefit. It appears that additional metabolic and toxicological data, both human and anima]-defived, focusing on high specific activity uranium, would be helpful in assessing the radiological hazards associated with denatured 233V fuel. Specifically, data on the biologica) effectiveness of 232y and 233y could modify exposure standards for these radionuclides. In terms of relative toxicities based on the dose commitment resulting from inhala- tion of equal masses of fuel, plutonium fuel is significantly more hazardous than HTGR 233 fuel or denatured 2°%U fuel. However, denatured 2°°U fuel would be significantly more hazardous than LWR uranium fuel. - As ‘the range of fuel cycle options is narrowed, more comprehensive research should be directed at derivation of toxicity data specific to facil- ities and fuel compositions of choice. | Research investigating potehtial enVirqnmental'hazards resulting from deliberate introduction (for safeguards purposes) of gamma emitters into fuels prior to refabrication is necessary, as is a thorough investigation_pf the ha;ards related tb repeated irradiation of recycle materials, with consequent buildup of low cross-section transmutation products. 1. 10. 1. 12. 13. 14. 15. 16. Reférences fok Sectidn_B;Z M. R. Ford, "Comments on Intake Guides for Var1ous Isotopes and Isotopic Mixtures of Uran1um,“ ORNL 3697, 1964. H. C. Hodge, J. N. Stannard, and J. B. Hursh, Uranium, Plutonium, and the Trans- plutonium Elements, Springer-Verlag, Heidelberg, 1973. M. P. Finkel, "Relative Biological Effectiveness of Radium and Other Alpha Emitters in CF No. 1 Female Mice," Proceedings of the Society for Exp. Biol. and Med. No. 3, p. 83, July 1953. J. E. Ballou and R. A. Gies, "Early disposition of inhaled uranyl nitrate (232U and 233()) in rats," p. 91 in BNWL-2100 (Partrl): Annual Report for 1976. J. E. Ballou and N..A. Wogman, "Nondestructive Analysis for 232U and Decay Progeny in Animal Tissues," BNWL-2100 (Part 1): Annual Report for 1976. J. E. Till, "Assessment of the Radiological Impact of 232U and Daughters in Recycled 2330 HTGR Fuel," ORNL-TM-5049, February, 1976. International Commission on Radiological Protection, "Report of Committee II on Permissible Dose for Internal Radiation,” ICRP Publication 2, Pergamon Press, New York, 1959. J. C. Reynolds, P. F. Gustafson, and L. D. Marinelli, "Retention and Elimination of Radium Isotopes Produced by Decay of Thorium Parents within the Body," USAEC Report ANL-5689, November 1957, p. 4. M. A. Van Dilla and B. J. Stover, "On the Role of Radlothortum (Th228) in Radium P01son1ng,“ Radiobiology 66: 400-401, 1956. B. J. Stover, D. R. Atherton, D. S. Buster, and N. Keller, "The Th228 Decay Series in Adult Beagles: RaZ2%, Pb212 and Bi212 in Blood and Excreta," Radiation Research 26: 226-243, 1965. o C. W. Mays letter to G. C. Butler, 25 August 1967. A. Kaul, "Tissue Distribution and Steady State Activity Ratios of Th232 and Daughters in Man Following Intravascular Injection of Thorotrast," in: Proceedings of the Third International Meeting on the Toxieity of Thorotrast, Copenhagen, April 1973 M. Faber, ed., RISO Report No. 294. S. R. Bernard and W. S. Snyder, private communication entitled "Memorandum on Thor1um Daughters," April 1968. B. J. Stover, D. R. Atherton, D. S. Buster, and F. W. Bruenger, "The Thorium Decay - Series in Adult Beagles: Ra22%, Pb2l2, and Bi212 in Selected Bones and Soft T1ssues," Radiation Research 26: 132-145, 1965. U.S. Energy Research and Development Administration, Final Envirommental Statement, Light Water Breeder Reactor Program, ERDA-1541, Vols. 1-5, 1976. ‘International Council on Radiological Protéctipn Report 20, 1972. — w - el - . = 3 3 t e ! " - r ) 1 | r*t o ol 3-17 3.3. [ISOTOPICS IMPACTING FUEL SAFEGUARDS CONSIDERATIONS 3.3.1. Enrichment Criteria of Denatured Fuel C. M, Newstead Brookhaven National Laboratory A very important problem in the determination of the characteristics of denatured fuel is the isotopic compositfon of the uranium, that is to say, the percent of 233y present in the mixture of 233y plus 238y, The guidelines provided by current regulations concerning the distinction between low-enriched uranium (LEU) and high-enriched uranium (HEU) are applicable to 23°U, the 1imit being set at 20% 235 in 238, Anything above that constitutes HEU and anything below that constitutes LEU. | LEU is considered to be unsuitable for constructing a nuclear explosive device. The rationale for making this statement is based upon the fact that the critical mass of 20% 235U-enriched uranium is 850 kg, and in a weapon this amount of material must be brought together sufficiently rapidly to achieve an explosive effect. Theoretically the enrichment could be lower and still achieve prompt criticality. However, the amount of material becomes so enormous and the difficulty of bringing it together so great that it would be impractical to attempt to produce an explosive device with less than 20% enrich- ment. It is clear that the distinction is somewhat of a gray area and the enrichment could be changed a few percent, but this should.be done extremely cautiously since the 235y enrichment vs. critical mass curve is rather steep and increasing the enrichment only slightly could reduce the critical mass substantially. Also, it is necessary to consider institutional arrangements. A number of domestic and international regulations revolve about the 20% figure and it would be no easy matter to change all these stipula- tions. This sets the background against which the enrichment considerations for denatured fuel must be addressed. The matter of arriving at a practical criterion is complicated and is currently under study by the Speéial Projects Division of Lawrence Livermore Laboratory, where an in-depth analysis of the weapons utility of fissile material (including 233U with various enrichments)rfor the Non-Proliferation Alternate Systems Assessment Program (NASAP) is being conducted in accordance with a work scope‘developed by the International Security Affairs Division (ISA) and the management of the NASAP Prografi. Unfortunately, the results of the LLL study are not yet available. Because of the considerable impact of enrichment cpnsiderétions_on the_uti]ity of particular reactors and particular symbiotic systems, it seems best at this point to discuss the several approaches for determining the gdide- 1ines for the enrichment of 233U-238y mixtures and to make a determination based on the LLL study at a later time. 3.18 There are three approaches which can be employed to estimate allowable enrichment criteria for 233y in 238Y corresponding to the statutory 20% limit set for 235U in 238y, These three criteria are: (1) critical mass, (2) infinite multiplication factor, and (3) yield. These can be employed singularly or in combination as discussed below. Critical Mass As stated above, the bare-sphere critical mass of metallic 20% 235U and 80% 238U is about 850 kg. This amount can be reduced by a factor of two to three by the use of a neutron reflector. However, the size and weight of the combination of reflector and fissile material will not be substantially less than that of the bare sphere, and may even be greater. In addition, for a nuclear explosive, an assembly scheme must be added which will increase the size and weight substantially. Concentrations of 235§, 233y, or plutonium in mixtures with 2380 such that they have bare-sphere metallic critical masses of about 850 kg represent one possible reasonably conservative criterion for arriving at concentrations below which the material 1is not usable in practica1 nuclear weapons. This 850 kg bare-sphere critical mass criterion can also be used for other materials which are or might be in nuclear fuel cycles. Although this criterion provides a basis for con- sistent safeguards requirements for 233U or 235U embedded in 238U, it leans to rather - lTow Timits. Infinite Multiplication Factor Another possible criterion is the one associated with the infinite multiplication factor k_. For a weapon to be successful, a certain degree of supercriticality must be attained. D. P. Smith of Los Alamos Scientific Laboratory has adopted this approach. He takes k_ = 1.658 for 20% 2®°U-enriched uranium, which implies k_ = 1.5346 for the oxide. He then performs a search calculation on enrichment for the other systems so as to obtain the same k_ value. His results are shown in Table 3.3-1. We note that for 233y the limits are 11.65% 233U for the oxide and 11.12% 233U for the metal. Table 3.3-1 Equivalent Enrichment Limits Fuel Material k_ Metal 20% 235y, 80y 238y 1.658 11.12% 233y, 88.887 238y 1.658 11.11% 239%uy, 88.89% 238y I Oxide (20% 235y, 80% 238U)0, 1.5346 (11.65% 233y, 88.35% 238y)0, 1.5346 - (13.76% 23%py, 86.24% 2384)0, 1.5346 (14.5% 239y, 1.5% 2%0py, 85% 238))0, 1.5344 These numbers were obtained by D. P. Smith of Los Alamos Scientific Laboratory from DTF IV calculations using Hansen-Roach cross sections. r r A we vt . £t o o o | 1 o '( oD 3-19 It may also be possible to set a minimum yield for a practical nuclear explosive device. An obvious consideration here is that in attempting to achieve supercriticality with increasing amounts of fissile material of decreasing enrichment, a point is reached where the yield of an equivalent mass of chemical high explosive exceeds the nuclear explosive yield. The LLL‘Specia] Projects Division is currently investigating the possibility of establishing such a limit. 3-20 3.3.2, Fabrication and Handling of Denatured Fuel J. D. Jenkins R. E. Brooksbank Oak Ridge National Laboratory The techniques required for fabricating and handling 233U-containing fuels encount- ered in the denatured fuel cycle differ from those employed for 235U fuels because of the high gamma-ray and alpha-particle activities present in the 233U fuels. Some jdea of the radiation IeVe]s that will be encountered can be deduced from recent radiation measure- ments for a can that contains 500 g of 233U with a 232U content of 250 ppm and has been aged 12 years since pufification. The results were as follows: Distance Radiation (mr/hr! Contact 250,000 1 ft 20,000 3 ft 2.000 These radiation levels are equivalent to those that could be expected at the same distances from 500 g of 233y containing ~ 1250 ppm 232U and aged six months, which is comparable with 233 that has'undergone several cycles in a fast breeder reactor. With such high activities, complete alpha containment of the fuel will be required, and all personnel must be protected from the fuel with thick biological shielding (several feet of concrete or the equivalent). This, of course, necessitates remote-handling operations, which constitutes an inherent safeguard against the diversion of the fuel while it is being fabricated and/or handled. The requirement for remote operation is further borne out by experience gained in two eariier programs in which 233U-containing fuels were fabricated. In these two pro- grams, the "Kilorod" program! and the Light Water Breeder Reactor (LWBR) program,? (233U,Th)0, pellets could be fabricated in glove boxes, but only because the 233y used contained extremely low (<10 ppm) amounts of 232y, Even so, the time frame for fuel fab- rication was severely restricted and extraordinary efforts were required to keep the con- tamination level of éged 233y sufficiently low to permit continued glove box operation. Based on experience at ORNL in the preparation of nearly two tons of 23300, for the LWBR program, it was determined that the handling of kilogram quantities of 233U containing 10 ppm of 232U and processed in unshielded glove boxes 25 days after purification (complete daughter removal) to produce 233U0, powder resulted in personnel radiation exposures of 50 mr/man-week. The techniques used in preparing Kilorod and LWBR fuel would not be feasi- ble in a large-scale fabrication plant using 233U containing the 100 to 2000 ppm 232y -expected in recycled 233U._ Therefore, one must conclude that remote fabrication, behind several feet of concrete shielding, will be required for 233y-bearing LWR and FBR fuels. Remote operation will impact the fabrication process and the fuel form. For ex- ampie, LWR and LMFBR fuels can be manufactured either as oxide pellets or as sol-gel .microspheres. The many powder-handling operations required in fabricating pellets with S 3 | ey | o - e o | ooy [ 3-21 their inherent dusting problems and the many mechanical operations required in blending powder, pressing, sintering, and grinding pellets make remotely operating and maintaining a 2%3y- bearing pellet fabrication line difficult. Alternatively, the relative ease of handling liquids and microspheres remotely makes the sol-gel spherepac process appear more amenable to remote operation and'maintenance:thanpowderpreparation and pelletizing processes, although the process is less fully developed. -‘ | Detailed analyses of specific flow sheets and process layouts for a particular fuel form would be required to quantitatively determine the relative safeguards merits of one process versus another. In general, however, batch processes where control of special nuclear materials can be effected by item accountability are easier than continuous pro- cesses in which the material is contained in liquid form. Thus, in our example above, an assessment might conclude that some sacrifices must be made in material accountability in order to achieve remote fuel fabrication. The overriding safeguards consideration in denatured fuel fabrication however is the remote nature of the process i1tself, which }imits personnel access to the fissile material, Access is not impossible, however, for two reasons.,. First, for material and equipment transfer, the processing cells will be Tinked to other cells or to out-of-cell mechanisms. Second, some portinns of the processing equipment may be maintained by persons who enter the cells after appropriate source shielding or source removal. Thus, some cells may be designed for personnel access, but all access points will be controlled because of the requirement for alpha-activity containment., Health physics radiation monitors would provide an indication of breach of containment and of possible diversion. Because the ingress points from the cells will be limited, portal monitors may also provide additional safeguards assurance. It should be noted that although kilogram quantities of material represent high- radiation Tevels from the standpoint of occupational exposures, the levels of recently purified 233U are low enough that direct handling of the material for several days would not result in noticeable health effects. ' ' - The remote nature of the refabrication process requires highly automated machinery _for'most of the fabrication. Elaborate control and monitoring instrumentation will be required for automatic operation and process control and can provide additional data for material accountabi]ity and material balance consistency checks. The remote nature of the process has the potential of substantially improving the safequarding of the recycle fuel during refabrication. The extent of this improvement will depend on the specific facility ‘design and on the degree to which the additional real- t1me process information can enhance the safeguards system, 3-22 3.3.3 Detection and Assay of Denatured Fuel D. T. Ingersoll Oak Ridge National Laboratory The reiative]y high gamma-ray activity of 233U fuels, enriched or denatured,'has opposite effects on detection and assayf it increases the detectability of the fuels but it also increases the difficulty of passive gamma assay. That this situation exists is appé}ent from Fig. 3.3-2, which presents a Ge(Li)fmeasuréd_gamma-ray spe;trum3_from a 233y sample containing 250 ppm 2%2U. Al major peaks in the spectrum are from the decay prbducts of 232U, which is near secular equi]ibrium with the products. The presence of the 2.6-MeV gamma ray emitted by 2°8T1 provides a useful handle for the detection of materials that contain even small quantities of 232y, thus providing a basis for preventing fuel diversion and/or for recovering diverted fuel. On the other hand, the presence of nUMETrous gamma rays in the spectrum eliminates the possibility of direct gamma-ray assay of the fissile isotope. Indirect assay using the 232U gamma rays would be impractical, since it would require a detailed knowledge of the history of the sample. Detection systems are already available. A Los Alamos Scientific Laboratory (LASL) report describes a doorway monitor system“ that employs a 12.7- x 2.5-cm NaI(T1) detector and has been used to measure a dose rate of about 2.5 mr/hr at a distance of 30 ¢m from a 20-g sample of PuO,. Approximately the same dose rate would be measured for a similar sample of 233U containing 100 ppm of 232U only 12 days following the separation of daugh- ter products. The dose rate would increase by a factor of IOIafter 90 days and by an additional factor of 4 after one year.> Also, the gamma-ray dose rate scales linearly with 232U content and is nearly independent of the type of bulk material, i.e., 233U, 235y, or 238y, The net counting rate for the Pu0, sample (shielded with 0.635 cm of lead) was 1000 cps. The observed background was 1800 cps, resulting in a signal-to-noise ratio of only 0.6. Similar samples of 232U-contaminated uranium not only would yield higher count- ing rates, but could also yield considerably better signal-to-noise ratios if the detector window were set to cover only the 2,6-MeV gamma ray present in the spectrum. Although the denaturing of uranium fuels tends to dilute the 232U content, the anticipated 232U levels in most denatured fuels is still sufficiently high for relatively easy detection, except immediately after complete daughter removal. ' ' _ The'difficulty in performing nondestructive assays (NDA) of denatured fuels relative to highly enriched fuels is attributable to two effects: {a) the desired signa]-(emitted neutrons or gamma rays, heat generatfon, etc.) is reduced because of the material dilu- tion, and (b) the signal is mostly obscured by the presence of 232U, The latter problem exists because although denaturing reduces the total concentration of 232U, the relative -proportion of 232y to fissile material remains the same. This is an especially signifi- cant problem with passive NDA techniques. As is shown in Fig. 3.3-2, the gamma-ray E {; .} |- 3-23 &[:'; spectrum from a 232) sample containing 250'bpm of 232| is totally dominated by the 232U decay gamma rays, thus eliminating the possibility of direct gamma-ray assay. Passive techniques employing calorimetry are also complicated since 232U decay particles can con- tribute significantly to the heat generation in a fuel sample. It has been calculated,3,6 that for a fresh sample of 233U containing 400 ppm 232U, nearly 50% of the thermal heat generation can be attributed to 232U decay, which increases to 75% after only one year. i | It is, therefore, apparent that fissile content assay for denatured uranium fuels will require more sophisticated active NDA techniques which must overcome the obstacles of material dilution and 232U-activity contamination. —y ! Tl : : ’ ‘ CRNLDWG M-10880 K COUNTS PER MINUTE 9220 960 1000 040 {080 ueom 43680 {600 m.mo, m_z«zo 2180 V mm&oz&wm o CHANNEL NUMBER 7Fig. 3.3-2. Gamma-Ray—Sgectrum from a 233U Sample Containing 250 ppm 232y, Al Tajor peaks age attributed to 232U decay products. Gamwma-ray energies indicated in MeV. - (From ref, 3,) S S , 3-24 '3.3.4. Potential Circumvention of the Isotopic Barrier of Denatured Fuel E. H. Gift and W. B. Arthur Oak Ridge Gaseous Diffusion Plant If a large-scale denatured-uranium recycle program is fully implemented (with secure energy centers), many types of both fresh (unirradiated) and spent fuel may be in transit throughout the world. In order to ensure that these fuels are proliferation resistant, they must meet the basic criterion that a sufficient quantity of fissile material cannot be chemically extracted from seized elements for direct use in the fabrication of a nuclear weapon. As pointed out in previous sections of this report, the addition of the denaturant 238 to the fissile isotope 233U will prevent the direct use of the uranium in weapons manufacture providing the 233U content of the uranium remains below a specified limit, which for this study has been set at 12% (see Section 3.3.1). Thus, even if the uranium were chem1ca11y separated from the thorium fertile mater1a1 included in the elements, it could not be used for a weapon. Similarly, if the 235U content of uranium is kept below 20%, the uranium would not be directly usable. For the discussion presented here, it is further assumed that fuels containing both 233U and 235U will meet this criterion if their weighted average lies between these limits. | "With the chemical isolation of the primary fissile isotopes thus precluded, two poten- tial means exist for extracting fissionable .material for the denatured fuel: (1) isotopic separation of the fresh fuel into its 233U (or 235U) and 238U components; and (2) chemical extraction from the spent fuel of the 23%u bred in the 238 denaturant or chemical extraction of the intermediate isotope 233Pa that would subsequently decay to 233UQ In this examination of the potential circumvention of the isotopic barrier of denatured fuel both these poSsibili- ties are discussed; however, the probability of the second one actually being carried out is essentially discounted. Thus the emphasis here is on the possibility that would-be proliferators would opt for producing weapons-grade uranium through the cliandestine operation of an isotope separation facility. For the purposes of this Study it is assumed that the seized fuel is in the form of fresh LWR elements of one of the following fuel types: A. Approximately 3% 2350 enriched uranium (same as current]y used LWR fuel). B. Recycle uranium from a thorium breeder blanket, denatured to m12% 233)) with dep1eted uranium. ' : ’ €. Fifth-generation recycle of fuel type B with 233y fissile makeup from a thorium breeder blanket. . D. First cycle of 235)-238y-Th fuel assuming no 233U is.available from an externa] source. In this fuel scheme the 235U concentration in uranium can be as high as 20% (see above), ’ ‘ | E. First recyc]e of fuel type D with 93% 235 in uranium makeup. In this fueling option, + not all of the fuel in a reload batch will contain recycle uranium.” Some portion of the reload batch will contain fuel type D. This option is ana1ogous to the "tradi- tional" concept envisioned for plutonium recycle fuels. It allows some of the fuel o ol w et L T ‘[jr r o o T r— - &:: r 3-25 to be fabricated in nonradioactive facilities.. This fueling Option will be referred to in the remainder of the text as fuel recycle Option 1. F. Fifth-generation recycle of fuel types D and E with 93% 235U makeup (Option 1). G. First recycle of fuel type D, with recycle uranium in all fuel assemblies of a reload batch. Makeup uranium is 20% and 93% 2350 as needed to maintain reactivity. In this option a}l fuel would probably require remote fabrication facilities. This fueling "option will be referred to in the remainder of the text as fuel recycle Option 2. H. Fifth recycle of fuel type G with 235 makeup (Option 2). The uranium compositions of thése fuels are shown in Table 3.3-2. In addition to these, it should be assumed that natural uranium is also available. Table 3.3-2. Uranium Fuel Mixtures That May Be Available (Weight Fraction in Uranium) Isotope A B ¢ D E ' F & H 232y 0 5.02 x 10-% 6.565 x 10~ 0 ©1.2363 x 10~% 2.445 x 10-% 1.134 x 10~* 2.331 x 10~% 233y 0 0.118611 0.11438 0 0.047004 0.05914 0.04310 0.05638 2.:'“‘U 1.2 x 10~% 0.008523 0.035108 0.001754 0.005430 0.02115 0.005125 0.020245 235y 0.032 0.002317 0.01255 0.2000 0.]3201- ' 0.1]3457" 0.13765 0.11749 238y 0o 0.000036 0.005327 0 : 0.02303 .l0.056496 0.021119 0.05386 238y 0.96788 0.870011 0t831228 0.798246 0.792389 0.749522 0.793021 0.75188 ‘Description of Fuel Type: - 3.2 wt ¥ 2350 from natural -uranium. - Thorium breeder blanket fuel denatured with depleted uranium. ~ Fifth generation recycle of B with thorium breeder blanket makeup. - 20 wt % 235 from natural vranium. - First recycle of D with 93 wt % 2350 {n uranium makeup (Option 1, see note). Fifth generation recycle of D with 93 wt % 235U in uranium makeup {Option 1, see note). First recycle of D with 93 wt ¥ 235U makeup (Option 2, see note} Fifth recycle of D with 93 wt % 235U makeup (Option 2, see note TOMMOOoE NOTE: Fuel types E and F are designed so that not all of the fuel in a reload batch is recycle fuel; some of the reload batch will contain fuel type D. This situation is analogous to the "traditional" concept envisioned for plutonium recycle fuels. This concept allows some of the fuel to be fabricated in non-radioactive “factlities, and is referred to in the text as fuel recycte Option 1. Fuel types G and H result if every assembly in the reload batch contains recyc1e fuel. - The fueling mode is referred to as Qption 2. ' Isotopic Separation of,Freéh.Fuel, Selection of Separation-Facility. Of the'various,uraniUm isotope separation processes _which have been conceived, only the current technology processes (i.e., gaseous diffusion, gas centrifuge, the Becker nozzle and the South African fixed wall centrifuge) and possibly the calutron process could be considered as near-term candidates for a Clandestine facility capable of enrichihg'divered reactor fuel. Of these, the gas centrifuge may be the preferred technology. This conclusion is directly related to the proven advantages of the process, which include a high separation factor per chhiné. Tow electrica]lpower needs, and the adaptability to sma1] low-capacity but high-enrichment plants. Further, more national groups (i.e., the U.S., England, Hoi]and, Germany, Japan, Australia, and France) have operated 3-26 eit;ér large centrifuge pilot plants or small commercial-sized plants, more so than for any other enrichment process, so it is apparent that this technology is widely understood and applied. A brief description'of'the centrifuge process, as well as descriptions of other current and future separation technologies, is given in Appendix A. The app11cat1on of centrifuge technology to a sma]l plant capable of produc1ng a _couple of hundred kilograms of uranium enriched to 90% 235U has not proved to be inordinately expensive. Two examples can be prov1ded, An article appearing in two Journa1s7’3 presents information on a proposed Japanese centrifuge plant. This piant, which could be operational in 1980, is designed to produce 50 MT SWU/yr in a 7000-machine facility. The total cost of the facility was estimated by the Japanese to be $166. 7 million. Simple arithmetic yields the individual centr1fuge separation capac1ty of 7 kg SWU/yr and a centrifuge cost of ap- proximately $24,000 (wh1ch includes its share of all plant facilities). An upper limit for the cost of developing a small gas centrifuge enrichment facility can be estimated from published costs from the United States uranium gas centrifuge program. A paper by Kiser? provides a convenient summary of the status and cumulative costs for the ‘U.S. program, The Component Test Facility, a plant which is expected to have a separative capacity of 50 MT SWU/yr (see Appendix A), was operational in January of 1977, To that date, the cumulative cost of the entire U.S. gas centrifuge program was given as about $310 million. Of this total, about $190 million was identified as development costs. The remain- ing $120 mi1lion was identified as equipment and facility expense. Further, only about $30 million was identified as being technology investigation. Even more intriguing is that within the initial 3-year development program (beginning in 1960 and budgeted at $6 million), the following accomplishments were recorded. a. The operating performance of the gas centrifuge was greatly improved. b. Small machines were successfully cascaded in 1961 (one year after initiation of the contract). c. When the last of these units was shut down in 1972, some machines had run continu- ously for about eight years. That these centrifuges were not commercially competitive with gaseous diffusion may be ir- relevant when they are considered as a candidate for a clandestine enrichment facility. Thus, as stated above, of the current technologies, the centrifuge process would probably be | selected. The utilization of the developing technologies (laser, plasma, etc,) for a clandestine enrichment facility is not currently feasible. Successful development of these technologies by any of the numerous national research groups would make them candidates' for such a facility, however, and they would offer the decided advantages of a high separa- tion factor, Iow-powér requirement and modular construction. ' Effect of 232U on the Enrichment Process'and'Product. A1l fuels containing 233U also " contain substantial amounts of'23?U. As mentioned éar]ier in this report, the daughter pro- ducts from 232y (t!‘5 = 72 yr) release highly energetic gamma rays and alpha particles that can complicate both the enrichment process and the subsequent weapon fabrication. - o r i r— r— b —.r— - r— 3-27 As a first step in evaluating the effect of 232U on the enrichment process and the en- riched product, consider fuel types B and C from Table 3.3-2 as feed to an enrichment plant. For making an acceptable weapon a fissile content of 90% 233U + 235U in the product should be satisfactory. An acceptable product flow rate from such a plant might be 100 kg U/yr. Based on these assumptions, the product concentrations shown in Table 3.3-3 were ob- tained from multicomponent enrichment calculational methods,l0 This table illustrates that while a suff1cient]y fissile uranium is produced, at a relat1vely low feed rate, the product has also concentrated the hxghly gamma. active (through its decay daughters) 232y by about a factor of 10. Greater than 99% of the 232U in the enrichment plant feed will be present in the product. In the enrichment plant the 232U concentration gradient from the feed point will drop rapidly in the stripping section. In the tails the 232U concentration will be reduced by about a factor of 150 from the feed concentration. As a result, the gamma radiation levels in the enrichment plant can be expected to vary by a factor of greater than 1000 from the tails to the product. Calculations have been made for a typical centrifuge enrichment plant to illustrate the gamma radiation level that could be expected at equilibrium as a function of the 232y concentration.,1l These results are shown in Table 3.3-4. Implicit in these estimates is the assumption that the daughter products of 232U are all deposited within the enrich- ment facility. This assumption seems justified since the fluoride compound of the first daughter product, 228Th (t15 = 1,9 years), is nonvoTat11e. With the exception of 22%Ra (t% = 3,6 d}, all of the other daughters have very short lives. Experimentally, little evidence exists to determine the true fractional déposition “of 232y daughters. Current evidence is incorporated in the existing specifications for UFg feed to the gaseous diffusion plants.12 These specifications call for a maximum 232y concentration of 110 parts bf”232U per billion pérts of 2350 in the feed. At this concentra- tion, the radiation levels would be s1gnif1cant in a hignly enr1ched product (m270 mr/hy at 1 ft and 3 mr/hr on the plant equipment).. ' Based on Tables 3.3-3 and 3.3-4, the maximum gamma radiation level.in a plant enriching 233U to 90% would be about 2 r/hr at equilibrium. At this radiation level, little decomposition of either lubrication.oils or the UFg gas would occur. Some evidence!l exists to show that at this radiation level the viscosity of the lubricating oils would be unaffected over a 20-year plant life. Thus, there should be no bearing problem. It is also t expected that the UFg would be fairly stable to the combined alpha and gamma radiation levels. At the'2-r/hr level, 1essrthan,one-tenth of the mean inventory of the machine would be decomposed per year. This material would be expected to be‘diStributed fairly uniformly throughout the machine with perhaps slightly higher accumulation on the withdrawal scoops. Since the individual machine inventory would be very low, this should not be a significant loss of material. 3-28 Table 3,3-3. Enriched Product Compositions (Weight Fraction in Uranium) Fuel Type B o Fuel Type ¢ Isotope Feed Product : Feed Product | 232) 5,02 x 10 4.1545 x 10~ 6.565 x 10-4 5.626 x 10-3 % 233y 0.118611 0.90 ©0.11498 0.90 234y 0.008523 0.03757 10.035108 - 0.0901 235y 0.002317 - 0.00376 ‘ 0.01255 0.00379 236 3.6 %1075 1.98x 1075 0.005327 1.73 x 107 238y 0.870011 0.05450 0.831228 3.124 x 10-* 233y in Tails 0.01 , 0.01 Feed Flow, 832 O esg kg U/yr Product Flow, 100 ' 100 kg U/yr - _ When removed from the plant, the UF; product would be condensed and probably stored in monel cylinders, If it is assumed that the cylinders were sized to hold 16 kg of UFg, the gamma dose rates that could be expected from the unshielded cylinders are as shown in Table 3.3-5. To reduce these product dose rates to acceptable levels would require substan- tial shielding. As an example, Table 3.3-6 shows the shielding required to reduce the dose rate at 1 ft to 1.0 and 50 mr/hr, Table 3,3-4., Gamma Radiation Level in an Enrichment Plant as a Function of 232y Concentration 232y Concehtration Radiation Level (r/hr) : (wt %) | at Equilibrium* | | 2.0 | - ' 6.8 1.0 | W 0.5 | 1.7 0.1 | | S L3 0.001 | | . | .0034 0.0001 o ' .00032 *Within an infinite array of centrifuges. r— r— r—= &£ rr— - S — [i 3-29 Table 3.3-5, 232y-Induced Gamma-Ray Dose Rates from Unshielded Monel Cylinders Containing 16 kg of ‘UF Dose Rate {r/hr) . Distance from Decay Time* Cylinder (day;) 0.1 wt § 232y 0.6 wt % 232y Contact 10 ' - 40.2 242 30 194 " 1,166 90 654 3,922 Equil. 7,046 42,300 1 Foot ' 10 . 4.2 ~ 25.4 30 20.4 122 90 ' 68.6 ' 412 | Equil. 740 | 4,440 1 Meter ld , 0.85 5.1 30 4.1 24.6 90 13.8 82.9 Equil. 149 894 *Time measured from chemical separation from thorium, Table 3.3-6. Shielding Required to Reduce 232U-Induced Gamma-Ray Dose Rates from Monel Cylinders Containing 16 kg of UFg* 5 : ; - Concrete Thickness (cm) Design Dose Rate Decay Time** (mr/hr) (days) 0.Twt @ 23207 0.6 wt % 2370 1.0 | 30 101 R P2 90 114 132 a Equil. 338 157 0 30 62 80 90 B @ Equil. ' 98 16 *Distance from source to shield = 1 ft. **Time measured from chemical separation from thorium. - lrmmi[i; r - .lffvf — o D ot l[fw‘ . 3-30 The high alpha activity of uranium'containing 232§ will present two problems: 1. In the UFg there will be a strong (a,n) reaction. A crude estimate of the neutron emission from a 16-kg UFg product cylinder containing 0.6 wtZ 232U is 5.7 x 107 neutrons/sec at 10 days decay, 2.5 x 108 at 30 days decay, and 8.7 x 10% at 90 days decay. 2. The 232 will provide a strong heat source in the UF¢ and the metal products. A crude estimate of the heat generation rate from pure 232U as a function of time after purification is: 0.03 W/g at 10 days, 0.13 W/g at 30 days, and_0.46 W/g at 90 days. The degree to which these properties will affect weapon manufacture or delivery is unknown, Alternative Enrichment Arrangements to Reduce 232U Content in the Product. In con- sidering the complications introduced to the final uranium metal product, i.e., the radia- tion level and heat generation resulting from 232U, it is apparent that removal of the 232y would be beneficial. Enrichment cascades can be designed to accomplish this. The most ef- ficient arrangement would be to first design a cascade to strip 232U from all other uranium Product Contai n1ng Nearly A1l the 23%y Product Containing ~90 Fissile Content and Very Low 232y Concentration Waste Fig. 3.3-2, Illustration of Enrichment Arrangement to Produce Low 232U Content Uranium, isotopes and then to feed the tails from the first cascade to a second cascade where the fissile isotopes can be enriched. This is illustrated in Fig. 3.3-2. Such an enrichment arrangement can be independent of the specific enriching device. Based on the discussion of the gas centrifuge‘ process in Appendix A and at the beginning of this section, a small, low separative work capacity machine may be within the technical capabilities of a would-be diverter (see Appendix A). Although no information exists on the separative work capacity of a Zippe machine in a cascade, a reasonable estimate of its separétive capacity is about 0.3 kg SWU/yr when separating 235U from 238y, e e e s e rs o '{j Z g r r r . © r - lfffq,uliff r- — ¥y 3-31 To further specify the'plant, it can be assumed that the diverter would like to: 1. Minimize the feed'and waste stream flows in the first and second cascades consistent with 1imiting the number of centrifuges required. 2. Achieve a significant weapons-grade product flow rate. (A flow rate of 100 kg U/yr having a fissile content of 90% 233U + 235y was chosen.) 3. Reduce the 232U content in the metal product so that contact manufacture can be achieved without serious radiation hazard, Based on these assumptions and considering the fuel types listed.in Table 3.3-2, a series of enrichment cascades, flows and selected isotopic parameters are presented in Table 3.3-7. The basic criterion chosen for the final uranium product was that the 232y concentration was about 1 ppm 232U in total uranium. At this level the gamma emission rate from the final metal product is sufficiently low that most fabrication and subsequent handling operations can be carried out in unshielded facilities using contact methods. | The first enrichment cascade to perform the separation of 232y from the remaining uranium will be very radioactive. But it will be only slightly more radioactive than if only -one cascade were used and the 232U not separated from the final product. The table shows that a factor of two increase in 232U product concentration will provide sufficient decontamination without a prohibitive increase in the number of centrifuges. If much greater (by a factor of 20) concentrations of 232U can be tolerated in the cascade, some reduction (~20 to 30%) can be made in the neceSsary number of'centrifuges. Table 3,3-7 also shows a striking difference in the number of centrifuges required to decontaminate the uranium product when the uranium makeup to the thorium cycles is 93% 235) pather than 233y from the thorium breeder blanket. This results because with the’ 235y recycle fuel it is more advantageous, both in centrifuges and in annual feed require- ments, to design the separation to throw away'in the first cascade waste stream much of the 233U and 234U in addition to the 232y, Thus, the fissile content in the final product from these fuel mixtures is nearly all 235, | B As a better ‘means of measur1ng the pro1iferat10n potential of the d1fferent fuel mixtures, the data- presented in Table 3,3-7 have been recast in Table 3,3-8 as a ' function of three parameters: (1) the number of centrifuges needed, (2) the uranium feed requirements to produce 100 kg/yr of 90% fissiIe uranium and (3) the number of ‘standard westlnghouse PWR fuel assemblies that must ‘be diverted. ~ Based on these criteria, the fol1ow1ng conclus1ons can be drawn -with respect to desirability of fuels for diversion: 3-32 Table 3,3-7, Summary of Results of Centrifuge Enrichment Survey of Potential Fuel Mixture® Fissile Content "Number of Centrifuges Required *Feed and centrifuges needed to produqe 100 kg U/yr of 90% fissile pfoduét. . 232 Content (wt. Fraction) (wt. Fraction) {0.3 kg SWU/yr Zippes) . ‘Of 1st 0f 2nd Of 2nd Of 2nd Annual ~In 232y In Fissile : Fu:; Cascade Cascade Cascard Cascade Feed Stripping Enriching Typ Initial Product Product Tails - Product (kg U/yr) Cascade Cascade Total " 3 A 0 NAd 0 . 0,002 0.90 2993 0 29220 29220 li; 8 5.02(-4)° KA 4,15(-3) 0,005 0.90 832 0 5468 5468 5.02{-4; 0.005 2.72-6 0.01 0.90 3180 82410 10880 93290 5.02(-4} 0.01 1.3(-6) 0.005 0.90 1302 50600 9981 60581 . 5.02(-4) 0.10 8.1(-7) 0,005 . 0.90 817 41653 7257 48510 : c 6.564{-4) NA 5,626(-3) 0.005 0.90 860 0 9191 9191 ; 6.564{-8) 0.0065 2.68(~6) 0.01 0.90 3000 86227 18302 104529 6.564(-4) 0.0 1.63(-6) 0.005 0,90 1749 61277 18802 80079 6.564(-4) 0.1 8.5(-7) 0.005 0.90 853 45483 11277 56760 . D 0 NA 0 0.01 © 0.90 468 0 4991 4991 L E - 1.236{-4) 0.001236 2.4{-6) 0,065 " 6.90 3000 25244 7002 32246 1.236{-4) 0.00235 1.14(-6) 0,06 0.90 1210 15459 5921 21380 1.236(-4) 0.00235 6.67(-7) 0.0 0.90 704 9292 13635 22927 F 2,445(-4) 0.002445 2.63(-6} 0.. 15 0,9 3001 33033 14398 47431 T 2.445(-4) 0,003 7.87(-6) 0.005 0.90 860 narze 20982 32854 G 1.134{-4) 0,003 6.42(-7) 0.005 .90 664 8758 13033 21791 H 2.331(-4) 0.0023 2.5[-6) 0.0715 0.90 3000 32136 12419 44555 S 2.331(-4) 0.003 7.44(-7) 0.005 0.90 805 11889 19477 31366 ! | Natural Uranium 0 NA 0 0.002 0.90 17575 0 17918 77918 %Feed and centrifuges needed to produce 100 kg U/yr of 90% fissile product. 1 bSee Table 3,3-2 for des¢ription of fuel types. | ®Read: 5,02 x 107%, dfiA = not applicable, i;: Table 3.3-8. Enrichment Resistance of Fuel Mixtures Investigated* _ ! _ 1 Number of _ feed Approximate Number of - Fuel Type Centri fuges Requirements PWR Fuel Assemblies (kg U/yr) Needed to Supply Feed R 3.2 wt %235y 29,220 2,993 6.7 - D 20 wt % 235U with thorium . 4,991 468 - 4.8 L Natural uranium (0.711 wt % 2350) 77,918 17,575 Not Applicable . 7 B st generation 233U recycle with thorium 5 No 232y removal 5,469 832 7.1 5 With 232y removal 48,910 817 6.9 ‘ Sth generation 233U recycle with thorium ) No 232y removal 9,191 860 7.0 : With 232y removal 80,079 1,750 4.2 1st generation 235U recycle with thorium {Option 1) With 232y removal o 22,927 704 6.8 B 5th generation 235U recycle with thorium (Option 1) . : l-i With 2324 removal . 32,854 860 7.4 1st generation 235U recycle with thorium (Option 2) . < With 2320 removal 21,791 664 6.6 L 5th generation 235U recycle with thorium (Option 2) . With 232U removal ' 31,366 805 7.0 - - r- r- o r o o - r - z[im - - T .. 3-33 1. Of the fuel mixtures that may be in commerce in a thorium-based fuel cycle, 20% 235y mixed with thorium is the most desirable both in ease of enrichment and because it requires diversion of the fewest fuel assemblies to produce a given quantity of highly enriched urantum. 2. Enrichment of 233U recycle fuels, without 232y remova], is an enrichment task com- parable (with respect to the number of centrifuges) to enriching 20% 235U, The product, however, will be highly radioactive, 3. If would-be proliferators must remove the 23zu;'the'2?5U makeup fuels are less prolifera- tion resistant than the 233U makeup fuels. 4, The 2350 recycle fuels with thorium and 232U removal are equivalent to 3.2 wt% slightly enriched uranium fuels with respect to both the number of centrifuges and the number of fuel assemblies to be diverted. 5. The 233y recycle fuels with thorium and 232U removal are Equiva1ent to natural uranium enrichment with respect to the number of centrifuges. 6. 1If 232y removal is necessary for ease of weapon manufacture and reliability of delivery, then a diverter would probably prefer to divert either slightly enriched uranium fuel or enrich natural uranium than to enrich either 235U or 233y recycle fuel from thorium .cycles. "This conclusion results from the fact that for each recycle fuel, the cor- responding slightly enriched or natural uranium fuel enrichment plant requires approximately the same number of centrifuges but has the decided advantage of a nonradioactive facility. Reliability of Centrifuge Enrichment Plants. As a final item, the average centrifuge failure rate and its impact on the maintainability and production rate of a centrifuge en- richment plant must be considered. Information on the reliability and operating life of centrifuges is scarce. The URENCO-CENTEC organization has over the years made claims of very long average operating 1ife and. correSpondingly Tow failure rates. Typical'examples of these claims can be found in some of their sales brochures.13 These claim an average 10-year operating Tife and a failure rate of Tess than 0. 5%/year. It is not clear how'much periodic ma1ntenance (e. 9es 0il changes and bear1ng 1nspect1on) is requ1red to ach1eve these low failure rates, o If these claims are accepted as a goal of a longwterm development proJect, then 1t can be assumed that in the early part of the deve]opment somewhat higher failure - rates would occur, perhaps greater by a factor of 10. This factor m1ght_be further justified 1n a highly radioactive plant‘sinoe periodic_maintenance would not be practical. ' The effect of centr1fuge fa11ures on the production rate 1n a radioact1ve p]ant has not been determined however, some qualitative statements can be made. All centri- fuge plants must be designed 50 that fa11ed units or groups of units can be immediately isolated from the rest of the plant. It should also be possible, for a specific cascade layout,;an‘assumed failure rate, and a specified plant operating 1ife, to provide - 3-34 statistical redundancy throughout the plant, so that as units fail a new unit is avail- able to be started. Thus, the production rate could be maintained fok_the chosen time period within the assumed statistical re]iabi]ify.‘ In order to achieve this reliability, greater numbers of centrifuges than listed in Table 3.3-9 would be required. . The exact number would be determinable when the ahove parameters are specified. Chemical Extractions from Spent Fuel As pointed out in the 1ntroduct1on to this section, another poss1b1]1ty for obta1n1ng fissionable material from diverted denatured 233U fuel is through the chemical extraction of protactinium or plutonium from spent fuel elements. 233pa is an intermediate isotope in the decay chain leading from 232Th to 233U that would'be chemically separable from the uranium prior to its decay. The plutonium available in the fuel elements wou]d be that produced in the 2380 denaturant of the fuel elements. The technical possibility of producing pure 233y yia chemical extraction of 233pa (t%'= 27.4 days) from spent denatured fuel was suggested by Wymer.l* Subsequent decay of the protactinium would produce pure 233U, While such a process is technically feasible, certain practical constraints must be considered. It is estimated!S that the equilibrium cycle discharge of a denatured LWR would contain ~34 kg of 233Pa [approximately 1 kg/metric ton of heavy metal]. However, due to its 27.4-day half-life, a 1-MT/day reprocesSing cap-- ability could recover only ~23 kg of 233pa (beginning immediately updn discharge with a 100% 233pa efficiency). Presumably a diverter group/nation choosing this route would have access to a re- processing faci]ity. Under routine operations, spent fuel elements are usually allowed a cool-down per1od of at least 120 days to perm1t the decay of short 11ved fission products, but in order to obtain the maximum quant1ty of 233pa from the denatured fuels it would be necessary to process the fuel shortly after its discharge from the reactor, This would involve handling materials giving off intense radiations and would probably involve an upgrading of the reprocessing facility, especially its shielding. On the other hand, con- ventional reprocessing plants in general already have high-performance shields and incre- mental increases ih the dose rates would not be unmangeable, especially for dedicated groups who were not averse to receiving relatively high exposures, Other problems requiring attention but nevertheless solvable would be associated with upgrading the system for controlling radioactive off-gases, making allowances for some degradation of the organic so]vefit due to the high radiation_]evel,'and obtaining shipping casks with provisions for recirculation of thercoo1ant to a radiator. - While from the above it would appear that extraction of 233Pa would be possible, considerably more fissile material could be obtained by extracting plUtonium from the spent denatured elements. Moreover, the usual coo]-dbwn'period‘prdbab1y could be allowed, which would require less upgrading of the reprocessing facility. On the other hand, the amount of plutonium obtained from the denatured elements would be considerably less (approximately a factor of 3 less) than the amount that could be obtained'by'seiZing and reprocessing spent LEU elements which are already-stdred in numerous countries. Thus it seems unlikely that a nation/ group would choose to extract either 233Pa or Pu from seized spent denatured fuel elements. f' T r~ o . O T oo ( - 3-35 3.3.5. Deterrence Value of 232\ Contamination in Denatured Fuel C. M. Newstead Brookhaven National Laboratory The preceding sections have emphasized that unless 232U is jsotopically separated from 233U, both it and its daughter products will always exist as a contaminant of the fissile fuel. And since as 232U decays to stable 208Pb :the daughter products emit several high-intensity gamma rays (see Fig, 3.0-1), all 233U fuel, except that which has undergone recent purification, will be highiy.radioactive. While the gamma rays, and to a lesser extent the decay alpha and beta particles and the neutrons from a,n reactions, will intro- duce complications into the fuel cycle, they will aiso serve as a deterrent to the seizure of the fuel and its subsequent use in the fabrication of a clandestine nuclear explosive. Consider, for example, the steps that would have to be followed in producing and using such a device: 1. Diverting or seizing the fissile materia] (as reactor fuel elements or as bulk material). ' 2. a. Chemically reprocessing the spent fuel to separate out the bred fissile plu- tonium (or 233pa) or b. Isotopically enriching the fresh fuel or bulk material to increase the 233U con- céntrafion in uranium sufficiently for its use 1n a weapon. 3. Fabricating the fissile material into a configuration suitable for an explosive device. 4, Arming and delivering the device. As indicated, at Step 2 a decision must be made as to which fissile material is to be employed, 23%Pu or 233y, Extracting the plutonium present in spent denatured fuel would require a chemical separation capability analogous to that required for current LEU spent fuel; however, the quantity of spent dénatured.fuel (i.e., kilograms of heavy metal) that “would have to be processed to obtain a sufficient amount of 23%Pu would be increased by a factor of 2 to 3 over the amount of LEU fuel that'would have to‘be'proceSsed. Moreover, for some reactor systems, the quality {i.e., the fraction of the material which is fissile) of the plutonium recovered from denatured fuel would be somewhat degraded relative to the LEU cycle. : The selection.of 233U as the weapons fissile material méans,_of_goufse,‘that the material being processed through all the operations 1isted above would be radioactive. While both national and subnational groups would-be,inhibited.fo some degree by the radiation field, it is clear that a national group would be more 1ikely to have the resources and technological base'necessaryAto.overCQme theAradiation hazard via remote handling, shielding, and various cleanup.techniques.. Thus, the radiation field due to the ?320 contamination would be effective in 11miting proliferation by a nation to the extent that it would com- plicate the procedures which the nation would have to follow in employing this path and 3-36 introduce time, cost and visibility considerations, These factors would force a trade-off between the desirability of utilizing material from the denatured fuel cycle and obtaining fissile material by some other means, such as isotopically enriching natural uranium or producing plutonium in a research reactor. A subnational group, on the other hand, would not in general possess the requisite technological capability. In addition, while a nation could, if they chose to, carry out these processes overtly, a subnational group would have to function covertly. Thus the radiation barrier interposed by the self-spiking effect of the 232y contaminant in the de- natured fuel would contribute in some measure to the safeguardab111ty of the denatured fuel cycle insofar as the subnational threat is concerned. The degree of protection provided by the self-spiking of denatured fuel varies accord- ing to the radiation level. The radiation level in turn depends on both the 232y concentra- tion and the time elapsed after the decay daughters have been chemically separated. As indicated in other sections of this chapter, in denatured fuel the expected concentrations of 232 in uranium are expected to range from ~100 to 300 ppm for thermal systems up to ~1600 ppm for recycled fast reactor fuel. It should be noted that if the latter denatured fuel (typically 10-20% 233U in 238U) is processed in an enrichment facility to obtain highly enriched (~90%) uranium, the resulting material would have a 232U content that is propor- tionally higher, in this case ~7000 to 8000 ppm maximum. Table 3.3-9 shows the radiation levels to be expected from various concentrations of 232y at a number of times after the uranium has been separated from other elements in a chemical processing plant. For a 5-kg sphere of 233U with 5000 ppm of 232y the radia- tion level 232 days after chemical separation is 67 r per hour at 1 m. The highest level of deterrence, of course, is provided when the radiation level is incapacitating. Table 3.3-10 describes the effects on individuals of various total body doses of gamma rays. Complete incapacitation requires at least 10,000 rem. Beginning at about 5000 rem the dose is sufficient to cause death within about 48 hr. In the 1000-rem range, death is practically certain within a week or two. A dose causing 50% of those exposed to die within several weeks (an LD-50) is around 500 rem. Below 100 rem it is unlikely that any side effects will appear in the short term but delayed effects may occur in the long term. In general, the gamma-ray total dose levels required to ensure that an individual is dis- abled within an hour or so are at least on the order of a magnitude higher than those likely to’ cause eventuaf death. There may be individuals who are willing to acqept doses in excess of several hundred rem and thus'eventuaIIy sacrifice their lives. As indicated above, to stop persons of suicidal dedication from‘compTeting the operations would require doses in the 10,000-rem range. Apart from the dedicated few, however, most individuals would be deterred by the prospect of long-term effects from 100-rem levels. However, it is also important to note that the individuals involved in the actual physical operations may not be informed as to the presence of or the effects of the radiation field. T . d T 3-37 Tabie 3,3-9 Gamma-Ray Dose Rates at a Distance of 1 m from a 5-kg Sphere f 233y Containing Var1ous Concentrations of 2322 t ; - 4 : Dose Rate at 1 m (mr/hr) (o Time® (days) 100 ppm" 500_ppm 1000 ppm 5000 ppm i . . -~ o 0 0 0 . 0.116 1.6x107% 8x10™% 1.6x10"3 8x10™° LJ 3.5 4.3x10° 2. 1x101 g.3x0! 2.1x10° | 10 3.5x10 1.8x102 3.5x102 1.8x103 - 23 1.1x10 5.7x10° 1.1x10° 5.7x10° - 46 2.6x10 1.3x10§ 2f6x10§ 1.3x1oz 93 5.5x10 2.8x10 5.5x10 2.8x10 232 1.3x10 6.7x10° 1.0t 6.7x10" 4 rom Ref. 16. C— Time after separation. CConcentration of 232y, Table 3.3-10, Effects of Various Total Body‘DoSes of Gamma Rays on Individuals® r Total Body Dose rem LJ < 25 25-100 LJ 100-200 ' 200-600 . 600-1,000 o 5,000~10,000 Effects No 1ikely acute health effects. No acute effects other than temporary blood changes. Some discomfort and fatigue, but no maJor disabling effects; chances of recovery excellent. Entering lethal range (LD-50 ~ 500 rads); death may occur within several weeks; some sporadic, perhaps temporary dis- - abling effects will occur (nausea, vomiting, diarrhea) with- in hour or two after exposure; however, effects are unlikely © to be comp]etely disabling in first few hours. Same as above, except that death w:thin 4 6 weeks is hzghly probable. : _ Death within week or two is practica]ly certain; disabling effects within few hours of exposure will be more severe than above, but only sporadically disabling. Death will occur within about 48 hr; even if delivered in less than one hour, dose will not cause high disability for several hours, except for sporadic. intense vomiting and diarrhea; convulsing and atax1a wil? be liker after several hours. : - Death will occur within a few hours.or less with complete ~incapacitation within minutes if dose 1s delivered within that short period A rom Ref. 17. r . -r" 3-38 An additioné] factor relative to the deterrent effect is the time required to carry out the necessary operations. This is illustrated by Table 3,3-11, which gives the dose rates (in rem/hr) required to acquire each of three total doses within various times, varying from a totally incapacitating 20,000 rem to a prudent individual's dose of 100 rem. Thus, to divert a small amount of fissile material to a portable, shielded container might take less than 10 seconds, in which case a dose rate of 107 rem/hr would be required to prevent completion of the transfer. Only 200 rem/hr would be required, on the other hand, to deliver a lethal dose to someone who spends five hodrs close to unshielded 233y while performing the complex operations required to fabricate components for an explosive ‘device. The maximum anticipated concentratioh of 232y as projected for denatured fuel does not provide sufficient intensity to reach totally disabling levels. Fast-reactor bred material (depending on time after separation and quantity as well as 232y concentra- tion) can come within the 100-rem/hr range. Table 3.3-11. Gamma-Ray Dose Rates for Three Levels of Total Dose vs. Exposure Time?: Dose Rate (rem/hr) Required to Deliver Total Dose of Time of Exposure 100 rem 1000 rem -~ 20,000 rem 10 sec 36,000 - 360,000 7,400,000 1 min 6,000 60,000 1,200,000 5 min 1,200 12,000 : - 240,000 30 min 200 2,000 40,000 1 hr 100 1,000 20,000 5 hr 20 200 4,000 12 hr 8.3 : . 83 1,660 aFr’om Ref. 18. The fact that the level of radiation of 232U-contaminated 233U increases with time is a major disadvantage for a 233U-based nuclear explosive .device. There is a window of 10 to 20 days immediately following chemical separation when the material is comparatively inactive due to the removal of 228Th and its daughters. Having to deliver a device less than ten days after fabricating it would be undesirable. While the tamper would provide some shielding, this short time schedule would compIicate the situation considerably. For a national program it is likely that the military would want a clean 233U weapon. This could be'accomplished'to'a large degree by separating the 232U from the 233y using gas centrifugation. However, because the masses are oh]y'] amu apart this requires several thousand centrifuges to make 100 kg of clean material per year (see Sec- tion 3.4.4). A nation possessing this isotObic separation capability would therefore prob- ably choose to enrich natural uranium rather than to utilize denatured fuel, thus eliminat- ing the 232U-induced complications. - emme., 3-39 In summary, for the case of national proliferation, the intense gamma-ray field as- sociated with the 232U impurity would not provide any absolute protection. However, the presence of 233y and its decay daughters would complicate weapons production sufficiently so that the nation might well prefer an alternate source of fissile material. For the case of subnational proliferation, the intense gamma-ray field is expected to be a major deter- rent, References for Section 3.3 1. R. E. Brooksbank, J. P. Nichols, and A. L. Lotts, "The Impact of Kilorod Operational Experience on the Design of Fabrication Plants for 233U-Th Fuels," pp. 321-340 in Proceedings of Second International Thorium Fuel Cycle Symposium, Gatlinburg, Tennessee, May 3-6, 1966. 2. Draft Environmental Statement, “Light-watek Breeder Reactor Program" ERDA-1541. 3. J. E. Rushtoni J. D. Jenkins, and S. R.‘McNeany, "Nondestructive Assay Techniques for Recycled 233U Fuel for High-Temperature Gas-Cooled Reactors," J. Institute Nuclear Materials Management IV(1) (1975). 4. T. E. Sampson and P. E. Fehlan, "Sodium Iodide and Plastic Scintillator Doorway Monitor Response to Shielded Reactor Grade Plutonium" UC-15, Los Alamos Scientific Laboratory (1976). : 5. N. L. Shapiro, J. R. Rec, R. A. Matzie, "Assessment of Thorjum Fuel Cycles in Pressurized-Water Reactors," ERRI NP-359, Combustion Engineering (1977). 6. Private communication from T. J. Burns to P. R. Kasten, Oak Ridge National Labora- tory, September 2, 1977. 7. Nuclear Engineering International, p. 10 (June 1977). 8. Nucleonics Week, p. 10 (June 16, 1977). 9. E. B. Kiser, Jr., "Review of U.S. Gas Centrifuge Program," AIF Fuel Cycle Conf. ‘77, Kansas City, Mo. (April 1977). - 10. A. de la Garza, "A Generalization of the Matched Abundance-Ratio Cascade for Multi- component Isotope Separation," Chemical Eng. Sci. 18, pp. 73-83 (1963). 11. E. D. Arnold, ORGDP, private communication (August 5, 1977). 12. USAEC, "Uranium Hexafluoride Specification Studies," OR0-656 {July 12, 1967). 13. URENCO-CENTEC, "Organization and Services"” (June 1976). | ~14. "Report to the LMFBR Steering Committee on Resources Fuel, and Fuel Cycles, and Proliferation Aspects," ERDA-72-60 (April 1977). 15. N. L. Shapiro, J. R. Rec, and R. A. Matzie'(Combustion Engineering), "Assessment of Thorium Fuel Cycles in Pressurized-water_Reactors.“ EPRI NP-359 (February 1977). 16. From calculations by E. D. Arnold in early 19?0’s'at O0ak Ridge National Laboratony. 17. From The Effects of Nuclear Weapons, Reviséd Edition, p. 592, Samuel GlasStone, Editor, prepared by U.S. Department of Defense, published by U.S. Atomic Energy Commission (April, 1962; reprinted February, 1964). 18. E. 0. Weinstock, "A Study on the Effect of Spiking on Special Nuclear Materials,” Submitted to Nuclear Regulatory Commission by Brookhaven National Laboratory (1976). J Wt r | . - ( 4.0. 4.1. 4.3. 4.4, 4.5, 4.6. CHAPTER 4 IMPACT OF DENATURED 233U FUEL ON REACTOR PERFORMANCE Chapter Outline Introduction, L. 5. Abbott, T. J. Burns, and J. C. Cleveland, ORNL Light-Water Reactors, J. C. Cleveland, ORNL 4.1.1. Pressurized Water Reactors 4,1.2. Boiling Water Reactors 4.2, Spectral-Shift-Controlled Reactors, N. L. Shapiro, CE Heavy-Water Reactors, Y. I. chang, ANL Gas-Cooled Thermal Reactors, J. C. Cleveland, ORNL 4.4.1. High-Temperature Gas-Cooled Reactors 4.4,2. Pebble-Bed High~Temperature Reactors Liquid-Metal Fast Breeder Reactors, r. J. Burns, ORNL Alternate Fast Reactors 4.6.1. Advanced Oxide-Fueled LMFBRs, T. J. Burns, ORNL _ 4.6.2. Carbide- and Metal-Fueled LMFBRS, p. L. Selby, P. M. Haas, and H. E. Knee, ORNL 4.6.3. Gas-Cooled Fast Breeder Reactors, T. J. Burns, ORNL J r- 0. { X Tk . r . ! e e r o r-or 4-3 4.0. INTRODUCTION L. S. Abbott, T. J. Burns, and J. C. Cleveland Oak Ridge National Laboratory The three preceding chapters have introduced the concept of 233U fuel and its use in nuclear power systems that include secure (guarded) energy centers supporting dispersed power reactors, the rationale for such systems being that they would allow for the production and use of fissile material in a manner that would reduce weapons proliferation risks relative to power systems that are increasingly based on plutonium-fueled reactors. Throughout the discussion it has been assumed that the use of denatured 233U fuel in power reactors is feasible; however, up to this point the Va1idity of that assumption has not been addressed. A number of calculations have been performed by various organizations to estimate the impact that conversion to the denatured cycle (and also to other "alternate" fuel cycles) would have on power reactors, using as models both existing reactors and reactors whose designs have progressed to the extent that they could be deployed before or shortly after the turn of the century. This thapter presents pertinent results from these calculations which, together with the predictions given in Chapfer 5 on the availability of the various reactors and their associated fuel cycles, have been used to postulate specific symbiotic nuclear power'systems utilizing denatured fuel. The adequacy of such systems for meeting projected electrical energy demands is then the subject of Chapter 6. The impact of an alternate fuel cycle on the performance of a reactor will, of course, be reactor specific and will largely be determined by the differences between the neutronic properties of the fissile and fertile nuclides included in the alternate cycle _and those included in the reactor's reference cycle. In the case of the proposed denatured fuel, the fissile nuclide is 233U and the primary fertile nuclide is 232Th, with fertile 238y included as the 233U denaturant. If LWRs such as those currently providing nuclear power in the United States were to be the reactors in which the denatured fuel is deployed, then the performance of the reactors using the denatured fuel must be compared with their performance using a fuel comprised of the fissile nuclide 235U and the fertile isotope 238y, And since the use of 233U assumes recycle, then the performance of the LWRs using denatured fuel must also be compared with LWRs in which Pu is recycled. Similarly, if FBRs were to be the reactors in which the denatured fuel is deployed, then the performance of FBRs operating on 233U/238y or 233y/238y/232Th and including 232Th in their blankets must-be compared with the performahce of FBRs operating on Pu/238U surrounded by a 238U btanket. A significant point in these two examples is that they represent the two generic types of power reactors -- thermal and fast -- and that the neutronic properties of the fissile and fertile nuclides in a thermal-neutron environment differ from their properties in a fast-neutron environment. Thus while one fissile material may be the optimum fuel in a reactor operating on thermal neutrons (e.g., LWRs) it may be the least desirable fuel for a reactor operating on fast neutrons (e.g., FBRs). 4-4 Table 4.0-1 gives some of the pertinent neutronic properties of the different fis- sile nuclides for a specific thermal-neutron energy. In discussing these properties,* it is necessary to distinguish between the two functions of a fissile material: the production of energy (i.e.; power) and the production of excess neutrons which when absorbed by fertile material will produce additional fissile fuel. Table 4.0-1, Nuclear Parameters of the Prinéipa1 Fissile Nuclides . 233y, 235y, 239y, and 241Pua:b at Thermal Energy (Neutron Energy = 0.0252 eV, velocity = 2200 m/sec) Cross Section (barns) Nuclide o a Of Uc . a v n 233y 578+2 531 +2 47+1 0,089 +0.002 2.487 + 0.007 2.284 + 0.006 235 - 678 +2 580+2 98+ 1 0.169 + 0,002 2.423 + 0.007 2.072 + 0,006 239py 1013+4 742+3 271 +3 0,366 + 0,004 2.880 + 0.009 2.109 + 0.007 241py 1375+9 1007 +7 368+8 0,365+ 0,009 2,934 + 0.012 2.149 + 0.014 “G. C. Hanna et al., Atomic Energ. Rev, 7, 3-92 (1969); figures in the referenced article were all given to one additional significant figure. o, = 0¢ + Og3 @ = Uclof; v = neutrons produced per fission; n = neutrons produced per atom destroyed = v/(1 + a). The energy-production efficiency of a fissile material is directly related to its neutron capture-to-fission ratio («), the smaller the ratio the greater the fraction of neutron-nuclide interactions that are energy-producing fissions. As indicated by'Table 4.0-1, at thermal energy the value of a is significantly smaller for 233U than for the other isotopes, and thus 233U has a greater energy-production efficiency than the other isotopes. (The energy released per fission differs only slightly for the above isotopes.) The neutron-production efficiency of a fissile material is determined by the number of neutrons produced per atom of fissile material destroyed (n)}, the higher the number the more the neutrons that will be available for absorption in fertile material. Table 4.0-1 shows that the n value for 233U is higher than that for any of the other nuclides, although plutonium would at first appear to be superior since it produces more neutrons per fission (v). The superiority of 233U results from the fact that o is lower for 233U and n = v/(1 + a). Thus at thermal energies 233U both yie]ds more energy and produces more neutrons per atom destroyed than any of the other fissile nuclides. In the energy range of interest for fast reactors (~0.05 - 1.0 MeV), the situation is not quite so straightforward. Here again, the « value for 233U is significantly lower than the values for the other fissile nuclides, and, moreover, the microscopic cross sec- tion for fission is higher (see Fig. 4.0-1). The energy release per fission of 233U is somewhat less than that of the plutonium nuclides, but the energy release per atom of 233U destroyed is significantly higher than for thc other nuclides. Thus, from the standpoint *Much of this discussion on the neutronic properties of nuclides is based on refs. 1 - 3. 3 J T a-5 ORNL-DWG 76-17705 r 1: £ | S ) 4}:3 { L4 - 1 i A b —— o ' C 001 040 100 1000 L; E(MeV) 3 Li b ORNL-DWG 7617704 &fi 04 / 03 . < a\% 0.2 . Y \\\\5::::?\*~\\\ 0.4 , 7 ' - P 004 ' 040 ; 100 iJ L E(MeV) , e e ¥ o FISSION {b) ORNL-DWG 78-13630 1 3.0~ Pu239 | 2.8 - ORNL-DWG 7647702 {0 —— /[ 8 \“?rs & ~ s 02 }/J/fi”——_________ 0 2 : 4 8 ‘ 8 {0 E(MeV) . Fig. 4.0-1. Nuclear Parameters of the Principal Fissi1e and Fert11e Nuclides at High Neutron Energies. a = ¢ /of, n= u/(] +a). . 4-6 of energy-production efficiency, 233U is clearly superior for fast systems as well as for thermal systems. However, with the historical emphasis on fissile production in fast systems, the overriding consideration is the neutron-production efficiency of the system, and for neutron production 239Pu is superior. This can be deduced from the values for n given in Fig. 4.0-1. The n value for 23°Pu is much higher than that for the other nuclides, es- pecially at the higher neutron energies, owing to the fact that 23%Pu produces more neutrons per fission than the other disotopes; that is, it has a higher v value, and that value is es- sentially energy-independent. As a result, more neutrons are available for absorption in fertile materials and 23%Pu was originally chosen as the fissile fuel for fast breeder reactors. The fission properties of the fertile nuclides are also important since fissions in the fertile elements increase both the energy production and the excess neutron production and thereby reduce fuel demands. At higher energies, fertile fissions contribute signifi- cantly, the degree of the contribution depending greatly on the nuclide being used. As shown in Fig. 4.0-1, the fission cross section for 232Th is significantly lower (by a factor of approximately 4) than the fission cross section of 238U. In a fast reactor, this means that while 15 to 20% of the fissions in the system would occur in 23U, only 4 to 5% would occur in 232Th. Thus the paired use of 233U and 232Th in a fast system would incur a double penalty with respect to its breeding performance. It should be noted, however, that since denatured 233U fuel would also contain 238U (and eventually 239py), the penalty would be somewhat mitigated as compared with a system operating on a nondenatured 233U/232Th fuel. In a thermal system, the fast fission effect is less significant due to the smaller fraction of neutrons above the fertile fast fission threshold. In considering the impact of the fertile nuclides on reactor performance, it is also necessary to compare their nuclide production chains. Figure 4.0-2 shows that the chains are very similar in structure. The fertile species 232Th and 23“U in the thorium chain corresponding to 238U and 240Py in the uranium chain, while the fissile components 233U and 2359 are paired with 23%Pu and 2 1Pu, and finally, the parasitic nuclides 236U and 2*2Py complete the respective chains. A significant difference in the two chains Ties in the nuclear characteristics of the intermediate nuclides 233Pa and 237Np. Because 233Pa has a longer half-1ife (i.e., a smaller decay constant), intermediate-nuclide captures are more probable in thé thorium cyclte. Such captures are doubly significant since they not only utilize a neutron that could be used for breeding, but in addition eliminate a potential fissile atom. A further consideration associated with the different intermediate nuclides is the reactivity addition associated with their decay to fissile isotopes following reactor shutdown. Owing to the longer half-life (and correspondingly higher equilibrium isotopic concentration) of 233Pa, the reactivity addition following reactor shutdown is higher for thorium-based fuels. Proper consideration of this effect is required in the design of the reactivity control and shutdown systems. The actual effect of all these factors, of course, depends on the neutron energy spectrum of the particular reactor type and must be addressed on an individual reactor basis. Significant differences also exist in the fission-product yields of 233y versus 235, and these, too, must be addressed on an individual reactor basis. ' W el T et r— r . [.._.u,m 4 - = i nm g - ¥ C 4-7 237\ B 1 6.75d 233U(n, v ) 3 23%U(n, ¥) —5m 235U (1, ¥) — 236U (1, y ) —o 237U g~ 1.27.4d g™ | ' 233pa(n,y) —a—234Pa e Noom 232 233 Th(fl,"{)—-—fil— Th 'Fig. 4.0-2a. Nuclide Production Chain for 232Th. 243pm oo 239y (n, y)—=—2%0Pu(n, v) —>=241Pu(n, vy} —=—242Pu(n, v ) —=—243py 8~ {2.35d B 239Np(n, y) ——240Np B~ 123.5m 238)(y, y) —a—239Y Fig. 4.0-2b., Nuclide Production Chain for 238y, Consideration of many of the above factors is inherent in the “mass balance" calcula- tions presented in this chapter for the various reactors operating on alternate fuel cycles. It is emphasized, however, that if a definite decision were made to employ a specific alternate fuel cycle in a specific reactdr. the next step would be to optimize the reactor design for ‘that particu]af cycle, as is discussed in Chapter S.A'Optimization of each reactor for the riany fuels considered was beyond the scope of this study, however, and instead the design used for each reactor was the design for that reactor's reference fuel, regardless of the fuel cycle'under consideration. ~ The reactors analyzed in the'éalculétions are light-water thermal reactors; spectral- shift-controlled thermal reactors; heavy-water thermal reactors; high-temperature gas- cooled thermal reactors; liquid-metal fast breeder reactors; and fast breeder reactors of advanced or alternate designs. 4-8 Since with the exception of the Fort St. Vrain HTGR, the existing power reactors in the United States are LWRs, initial studies of alternate fuel cycles have assumed that they would first be implemented in LWRs.* Thus the calculations for LWRs, summarized in Sec- tion 4.1 have considered a number of fuels. For the purposes of the present study the fuels have been categorized according to their potential usefulness in the envisioned power system scenarios. Those fuel types that meet the nonproliferation requirements stated earlier in this report are classified as "dispersible" fuels that could be used in LWRs operating out- side a secure energy center. The dispersible fuels are further divided into denatured 233y fuels and 235U~-based fuels. The remaining fuels in the power systems are then categorized as "energy-center-constrained" fuels. Finally, a fourth category is used to identify "reference" fuels., Reference fuels, which are not to be confused with an individual reactor's reference fuel, are fuels that would have no apparent usefulness in the energy-center, dispersed-reactor scenarios but are included as 1imiting cases against which the other fuels can be compared. (Note: The reactor's reference fuel may or may not be appropriate for use in the reduced proliferation risk scenarios.) To the extent that they apply, these four categories have been used to classify all the fuels presented here for the various reactors. Although the contributing authors have used different notations, the fuels included are in general as follows: Dispersible Resource-Based Fuels A. Natural uranium fuel (containing approximately 0.7% 235U), as currently used in CANDU heavy-water reactors. Notation: US5(NAT)/U. B. Low-enriched 235U fuel (containing approximately 3% 235U), as currently used in LWRs. Notation: LEU; U5(LE)/U. C. Medium-enriched 235U fuel (containing approximately 20% 23°U) mixed with thorium fertile material; could serve as a transition fuel prior to full-scale implementa- tion of the denatured 233U cycle. Notation: MEU(235)/Th; DUTH(235). Dispersible Denatured Fuel D. Denatured 233U fuel (nominally approximately 12% 233U in U). Notation: Denatured 233(); ‘ denatured uranium/thorium; denatured 233U0,/Th0,; MEU(233)/Th; 233y/238y; DUTH(233); U3(DE)/U/Th. ' *NOTE: The results presented in this chapter do not consider the potential improvements in the once-through LWR that are currently under study. In general, this is also true for the resource-constrained nuclear power systems evaluated in Chapter 6; however, Chapter 6 does include results from a few calculations for an extended exposure (43,000-MWD/MTU) once-through LEU-LWR. The particular extended exposure design con- sidered regquires 6% less U30g over the reactor's lifetime. 4 r } e r- r. r— . £ “§ 4 ri —\ 1 — ' ! o o E‘ e b - a7 4-9 Energy-Center-Constrained Fuels E. LEU fuel with plutonium recycle. F. Pu-232Th mixed-oxide fuel. Notation: Pu0,/Th0,; (Pu-Th}0,; Pu/Th. G. Pu-238U mixed-oxide fuel, as proposed for currently designed LMFBRs. Notation: Pu0,/U0,; Pu/23U; Pu/uU. Reference Fuels ~ H. Highly enriched 2350 fuel (containing approximately 93% 235U) mixed with thorium fertile material, as currently used in HTGRs. Notation: HEU(235)/Th;_U5(HE)/Th. I. Highly enriched 233U fuyel (containing approximately 90% 233y mixed with thorium fertile material. Notation: HE(233)/Th; U3/Th; U3(HE)/Th. | Including plutonium-fueled reactors within the energy centers serves a two-fold purpose: It provides a means for disposing of the plutonium produced in the dispersed reactors, and it provides for an exogeneous source of 233U. The discussion of LWRs operating on these various fuel cycles presented in Section 4.1 is followed by similar treatments of the other reactors in Sections 4.2 - 4.6. The first, the Spectral-Shift-Controlled Reagtor (SSCR), is a modified PWHR whose operation on a LEU cycle has been under study by both the United States and Belgium for more than a decade. The pr1mary goal of the system is to improve fuel ut111zat10n through the in- creased production and in-situ consumption of fissile plutonium (Pu ). The capture of neu- trons in the 238U included in the fuel elements is increased by mixing heavy water with the light-water moderator-coolant, thereby shifting the neutron spectrum within the core to energies at which neutron absorption in 238y is more 1ikely to occur. The heavy water content in the moderator is decreased during the cycle as fuel reactivity is depleted. The increased capfure is also used as the reactor control mechanism. The SSCR is one of a class of reactors that are increasingly being referred to as advanced converters, a term app11ed to a thermal reactor whose design has been modified to increase its production of fissile material. , . _ Heavy-water-mod1f1ed therma] reactors are represented here by Canada's natural- uranium-fueled CANDUs. L1ke the SSCR, the CANDU has been under study in the U. S. as an advanced converter, and scoping calculations have been performed for several fuel cycles, including a slightly enriched 235U‘-c‘yc'|e that is_considered_to be the reactor's reference cycle for implementation in the United States. The high-temperature gas-cooled thermél reactors considered are the U.S. HTGR and the West German Pebble Bed Reactor (PBR), the PBR differing from the HTGR in that it utilizes spherical fuel elements rather than prismatic fuel elements and employs on-line re- fueling. For both reactors the reference cycle [HEU(233U)/Th] includes thorium, and shifting 4-10 to the denatured cycle would consist initially in replacing the 93% 235U in 238y with 15 to 20% 235U in 238y, The HTGR has reached the prototype stage at the Fort Vrain plant in Colorado and a PBR-type reactor has been generating eTectricity in West Germany since 1967. While the above thermal reactors show promise as power-producing advanced converters, they will not be self-sufficient on any of the proposed alternate fuel cycles and will re- quire an exogenous source of 233U, An early but limited quantity of 233U could be provided by introducing thorium within the cores of 235U-fueled LWRs, but, as has already been pointed out in this report, for the long-term, reactors dedicated to 233U production will be required. In the envisioned scenarios those reactors primarily will be fueled with Puf. In the calculations presented here a principal 233 production reactor is the mixed-oxide-fueled LMFBR containing thorium in its blanket. In addition, "advanced LMFBRs" that have blanket assemblies intermixed with fuel assemblies are examined. The possible advantages and disadvantages of using metal- or carbide-based LMFBR fuel assemblies are also discussed. Finally, some preliminary calculations for a helium-cooled fast breeder reactor (GCFBR) are presented. The consideration of fast reactors that burn one fissile material to prodqce another has introduced considerable confusion in reactor terminology which, unfortunately, has not been resolved in this report. In the past, the term fast breeder has been applied to a fast reactor that breeds enough of its own fuel to sustain itself. Thus, the fast reactors that burn 23%Pu to produce 233U are not "breeders" in the traditional sense. They are, however, producing fuel at a rate in excess 6f consumption, which is to be contrasted with the advanced thermal converters whose primary function is to stretch but not increase the fuel supply. In order to distinguish the Pu-to-233U fast reactors from others, the term transmuters was coined at ORNL. Immediately, however, the word began to‘be'applied to any reactor that burns one fuel and produces another. Moreover, it soon became obvious that the words fast and breeder are used synonymously. Thus in this report and elsewhere we find various combinations of terms, such as LMFBR transmuter and converter transmuter. The situation becomes even more complicated when the fast reactor design uses both 238 and 232Th in the blanket, so that in effect it takes on the characteristics of both a transmuter and a breeder. Finally, the reader is cautioned not to infer that only those reactors discussed in this chaptek are candidates for the energy-center, dispersed-reactor scenarios. In fact, the scenarios discussed in Chapter & do not even use all these reactors and they could - easily consider other reactor types. The selection of reactors for this preliminary assessment of the denatured 233U fuel cycle was based primarily on the availability of data at the time the study was initiated (December, 1977). ' e.: 1 5 L -y { - 'Referénces for Section 4.0 1. P. R. Kasten, F. J. Homan et al., "Assessment of the Thorium Fuel Cycle in Power - Reactors," ORNL/TM-5665, Oak Ridge National Laboratory (January 1977). V! : R _ - 2. P. R. Kasten, "The Role of Thorium in Power-Reactor Development," Atomic Energy Re- view, Vol. III, No. 3. [ 3. "The Use of Thorium in Nuclear Power Reactors," prepared by Division of Reactor &‘ Development and Technology, U.S. Atomic Energy Commission, with ass1stance of ANL, B&W, BNL, GEA, ORNL, and PNL, WASH 1097 (June, 1969). 7 . £ -1 oy £t r ! i ' | i ! | i ! i i i i 4.1. LIGHT-WATER REACTORS .Jd. C, Cleveland Qak Ridge National Laboratory If an alternate cycle such as the denatured cytle is to. have a significant early impact, it must be implemented in LWRs already operating in the United States or soon to be operating. The current national LWR capacity is about 48 GWe and LWRs that will provide a total capacity of 150 to 200 GWe by 1990 are either under construction or on order. Much of the initial analyses of the denatured 233U fuel cycle has therefore been performed for current LWR core and fuel assembly designs under the assumption that subsequent to the required fuels development and demonstration phase for thoria fuels these fuels could be used as reload fuels for operating LWRs. It should be noted, however, that these current LWR designs were optimized to minimize power costs with LEU fuels and plutonium recycle, and therefore they do not represent optimum désigns for the denatured cycle. Also excluded from this study are any improvements in reac- tor design and operating strategies that would improve in-situ utilization of bred fuel and reduce the nonproductive loss of neutrons in LWRs operating on the once-through cycle. Studies to consider such improvements have recently been undertaken as part of NASAP (Nonproliferation Alternative Systems Assessment Program). 4.1.1. Pressurized Water Reactors Mass flow calculations for PWRs presented in this chapter were performed primarily by Combustion Engineering, with some addltional results presented from ORNL calculations. The Combust1on Engineering System 80 (PWR) design was used in all of these analyses. A description of the core and fuel assembly design is presented in the Combustion Engineering Standard Safety Analysis Report (CESSAR). The following cases have been analyzed:1-6 Dispersible Resource-Based Fuels A. LEU (i.e., low enriched uranium, ~3% 235U in 238U), no recycle. B. MEU/Th (i.e., medium-enriched uranium, 20% 235U in 238y, mixed with 232Th), no recycle, C. LEU, recycle of uranium only, 235U makeup. MEU/Th, recycle of uranium (235U + 233y), 20% 2350 makeup.* Dispersible Denatured Fuel E. Denatured 233U (i.e., ~12% 233U in 238U, mixed with 232Th), recycle of uranium, 233y makeup. *An alternate case utilizing 93% 235U as a fissile topping for recovered recycle uranium and utilizing 20% 235U as fresh makeup is also discussed by Combustion Engineering. g + ¢ ¥ | s r | o 1 e .y et ~ Energy-Center-Constrained Fuels F. LEU, recycle of uranium and self-generated plutonium, 235U makeup. G. Pu/238Y, recycle of plutonium, plutonium makeup., H. Pu/232Th, recycle of plutonium, plutonium makeup. I. Pu/232Th, one-pass plutonium, plutonium makeup. Reference Fuel J. HEU/Th (i.e., highly enriched uranium, 93.15 w/o 235U in 238U, mixed with 232Th), recycle of uranium (235U + 233U), 235U makeup. Case A represents the current mode of LWR operation in the absence of reprocessing. Case B involves the use of MEU/Th fuel in which the initial uranium enrichment is limited to 20% 235U/238y, With reprocessing again disallowed, Case B reflects a "stowaway" option in which the 233U bred in the fuel and the unburned 235U are reserved for future utilization. Case C represents one logical extension of Case A for the cases where the recycle of certain materiels is allowed. However, consistent with the reduced proliferation risk ground rule, only the uranium component is recycled back into the dispersed reactors. Case D similarly reflects the extension of Case B to the recycle scenario. In this case, the bred plutonium is assumed to be separated from the spent fuel but is not recycled. MEU(20Z 235U/U)/Th fuel is used as makeup material and is assumed to be fabricated in separate assemblies from the recycle material. Thus, only the assemblies containing recycle material require remote fabrication due to the presence of 232U, (It is assumed that the presence of the 232U pre- cludes the recovered uranium being reenriched by isotopic separation.) The recovered uranium from both the recycle and the makeup fuel fractions are mixed together prior to the next recycle. This addition of a relatively high quality fissile material (uranium recovered from the makeup fuel) to the recycle fuel stream slows the decrease in the fissile content of the recycle uranium. As in the LEU cycle, the fissile component of the recycle fuel in this fuel cycle scheme is diluted with 238U which provides a potential safeguards advantage over the conventional concept of plutonium recycle in LWRs with about the same U30g utilization. o ' Case E is the denatured 233y fuel, It utilizes an exogenous source of 233U for both the initial core fissile requirements and the fissile makeup requirements. Cases F - I represent possible fissile/fertile fuel cycle systems allowable for use in secure energy centers. Case F represents an extension of Case C in which all the fissile material present in the spent fuel, including the plutonium, is recycled. Under equilibrium conditions, about 1/3 of each reload fuel batch consists of mixed oxide {M0;) fuel assemblies which contain:the recycled p]dtonidm in a uranium diluent. The remaining 2/3 of each reload consists of fresh or recycled uranium (235U) oxide fuel. 4-14 Case G allows one possible means for utilizing the plutonium bred in the dispersed reactors. Plutonium discharged from LEU-LWRs is usec to provide the initial core fissile requirements as well as the fissile makeup requifements. This p]utonium is blended in a U0, diluent consisting of natural or depleted uranium. .The plutonium‘discharged from the UQ,/Pub, reactor is continually recycled - with two years for reprocessing and refabrica- tion - through the reactor. In the equilibrium condition, plutonium discharged from about 2,7 LEU-fueled LWRs can provide the makeup fissile Pu requirement for one UQ,/Pu0, LHR. In Case H the Pu0,/ThO, LWR also utilizes plutonium discharged from LEU-LWRs to provide the initial core fissile requirements and the fissile makeup requirements. This plutonium is blended in a ThO, diluent. The isotopically degraded plutonium recovered from the Pu0,/ThO, LWR is blended with LEU-LWR discharge plutonium (of a higher fissile content) and recycled back into the Pu0,/ThO, LWR. Not only does this case provide a means of ~eliminating the Pu bred in the dispersed reactors but, in addition, also provides for the production of 233U that can be denatured and used to fuel dispersed reactors. The Pu0,/Th0, LWR of Case I is similar to that in Case H in that plutenium discharged . from LEU-LWRs is used to provide the fissile requirements. However, the isotopically degraded plutonium recovered from the PuQ,/Th0, LWR is not recycled into an LWR but is stored for Jater use in a breeder reactor. Case J involves the use of highly enriched uranium blended with ThO, to the desired fuel enrichment. The uranium enrichment in HEU fuels was selected as 93.15 w/o on the basis of information in Ref. 7. [Initially all fuel consists of fresh HEU/Th fuel assemblies. Once equilibrium recycle conditions are achieved, about 35% of the fuel consists of this fresh makeup fuel, the remaining fuel assemblies in each reload batch containing the recycled (but not re-enriched) uranium oxide blended with fresh ThO,. Table 4.1-1 provides a summary, obtained from the detailed mass balance information, of initial loading, equilibrium cycle loading, equilibrium cycle discharge, and 30-year cumulative U305 and separative work requirements. All recycle cases involve a two-year ex-reactor delay for reprocessing and refabrication. It is important to point out that for cases which involve recycle of recovered fissile material back into the same LWR, in "equilibrium" conditions the makeup requirement for a given recycle generation is greater than the difference between the charge and discharge quantities for the previous recycle generation because of the degradation of the isotopics. This is especially important in Case H whére, for example, the fissile content of the plutonium drops from about 71% to about 47% over an equilibrium cycle. _ Comparing Cases_A and B of Table 4.1-1 indicates the penalties associated with im- plementation of the MEU/Th cycle relative to the LEU cycle under the restriction of no re- cycle, The MEU/Th case requires 40% more U305 and 214% more separative work than the LEU { P case, Clearly the MEU/Th cycle would be prohibitive for "throwaway" options. A second signi- ficant result from Table 4.1-1 is given by the comparison of Case D, MEU/Th with uranium recycle and Case F, LEU with uranium and self-generated plutonium recycle. The U305 demand in each case is the same, although the MEU/Th cycle requires increased separative work. Additionally er i | a; it should be noted that in Case D the MEU/Th fuel also produces significant quantities of plutonium, an additional fissile material stockpile which is not recycled in this case. E 4 _Table 4.1-1, Fuel Utilization Characteristics for PWRs Under Various Fuel Cycle Options"'b = Separative Work Initial Equilibrium Cycle U;0g Requirement Requirement -, Fissile - Fissile Fissile (ST/Gwe) (103 kg SHU/GN@l ! ! Inventory Charge Discharge Conversion Burnup rd e ~=J Case Fuel Type kg/GWe-yr) {kg/GWe- Ratio Initial® To al 291nit 151__19521__ Dispersible Resource-Based Fuels | A LEU, no recycle 1693 235y 794 235y 215 23:U 0.60 30.4f 392 5989/ 203 3555 ‘ 174 Pu B MEU/Th, no recycle 2538 235y 1079 235y ggg g::u 0.63 32.6 638 8360 580 7595 U T 71 Puf u C LEU, U recycle 1693 235y - - 0.60 30.4 392 4946 203 3452 D MEU/Th, self- 2538 235y 313 2339 282 2339 .66 32,6 638 4090 580 3632 generated U recycle 675 23539 - 257 23549 — 95 pufé il o tsi Dispersible Denatured Fuel E Denatured 233U0,/Th0,, 1841 233y 750 233y 446 233y 33.4 —_ U recycle (exogenous 27 235y 29 235 43 235y : 133U makeup) 63 Puf &., Energy-Center-Constrained Fuels — F LEU, recycle of U + 1693 235y 612 235y 193 235y 0.61 30.4 392 4089 233 2690 i self-generated Pu 258 puf 288 puf ih& 6 Pu0Dy/U0z, Pu recycle 1568 Puf 1153 Puf 858 puf 0.63 30.4 100 1053 0 0 546 235y 173 235y 108 235y H PuD,/Th0,, Pu recycle 2407 Puf 1385 Puf 696 puf 33.0 - . ‘ 272 23y ? I Pu0,/ThO,, single Pu 2407 Puf 1140 Puf 410 puf 33.0 g pass 284 233y Reference Fuel A J HEU/Th, seff—generated 2375 235 388 za:f’U 377 23y 0.67 33.4 597 3453 596 3436 U recycle ) . 504 235y . 172 23%y gAl1 cases assume 0.2 w/o tails and 75% capacity factor. b A1l calculations were performed for the 3800-Mit, 1300-MWe Combustion Engineering System 80 reactor design, ! CAssumes 1.0% fabrication loss and 0.5% conversion loss, | o credit taken for end of reactor 1ife fissile inventory. ffissumes 1.0% fabrication loss. ' An additional case 1s considered in Chapter 6 in which an extended exposure (43 MiD/kg HM) LEU-PWR on a once-through cycle results in a 6% reductfon in the 30-yr total U;0g requirements, while still requiring essentially the same enrichment (SHU) — requirements. Somewhat less plutonium is discfiarged from the reactor because of a reduced conversion ratio. , } ; gVatues provided are representative of years 19-23, _ . b Reference fuels are considered only as limiting cases. - ' Differences in the nuclide concentrations of fertile isotopes from case to case result k; in differences in the resonance ‘integrals of each fertile isotope due to self-shielding effects, thus signlficantly affecting the conversion of fertile material to fissile material. Table {f 4,1-2 gives the resonance integrals at core operating temperatures for various fuel combina- & tions. Although the value of the 238U resonance integral for an infinitely dilute medium P is much larger than the corresponding value for 232Th, the resonance integral for 2380 in LEU {: fuel is only 25% larger than that for 232Th in HEU/Th fuel, indicating the much larger amount % of self-shielding occurring for 238U in LEU fuel. These two cases represent extreme values, 4-16 L } since in each case the one fertile isotope is not significantly diluted by the presence of &iiii the other, -For_MEU(ZO% 235(3/0))/Th fuel, the 238U density is reduced by a factor of ~6 ; (relative to LEU fuel), causing the 238U resonance integral to increase due to the reduced L self-shielding. The decrease in the 232Th density for the MEU/Th fuel (relative to the HEU/Th) fuel is only a factor of ~0.8 - resulting in a much smaller increase in the 232Th : . resonance integral. Thus, although the 238U number density is roughly six times less in l; MEU/Th fuel than in LEU fuel, the fissile Pu production in the MEU/Th fuel is still 40% of L L L ~that for the LEU fuel as shown in Table 4,1-1 (Cases A and B) due to the increase in the 238Y resonance integral. The'preéence‘in denatured uranium-thorium fuels of two fertile isotopes having resonances at different energy levels has a significant effect on the initial loading requirement. The initial 233U requirement for the HEU/Th and MEU/Th cases is 2375 and 2538 kg/GWe, respectively, reflecting the penalty associated with the presence of the two fertile isotopes in the MEU/Th fuel. The large increase in initial 235 requirements shown in Table 4.1-1 for the thorium- based HEU/Th and MEU/Th fuels compared to the LEU fuel results primarily from the larger thermal-absorption cross section of 232Th relative to 238U as shown in Table 4.,1-2, Also contributing to the increased 235U requirements is the lower value of n of 235U which re- sults from the harder neutron energy spectrum in thorium-based fuels. ' Table 4.1-2. Thermal Absorption Cross Sections and Resonance Integrals for 232Th and 238( in PWRs Resonance IntegraTa (barns) L; Isotope % }gégfig)e‘” Infinitely In LEU In HEU/Th In MEU(2350/U)/Th Dilute Fuel Fuel Fuel [ 232Th 7.40 85.8 — 17 19 L,, 238 2.73 273.6 21-22 — 50-54 L ~the recovered uranium could be reenriched to an allowed denaturing limit prior to recycle, %for absorption from 0.625 eV to 10 MeV; oxide fuels. A further consideration regarding MEU(233U/U)/Th fuel with uranium recycle must also be noted, Since the fissile enrichment of the recovered uranium decreases with each genera- 1 tion of recycle fuel, the thorium Toadings must continually decrease. {(As pointed out above, [fi it is assumed that the recovered uranium is not reenriched by isotdpic separation techniques.) The initial core 232Th/238) ratio is ~5.8 and the first reload 232Th/238y ratio is 4.4, but L; by the fourth recycle generation the 232Th/238U ratio has declined to ~1.4.5 An alternative is to use HEU (93.15 w/o 235U) as a fissile topping for the recovered uranium. In this way thus minimizing the core 238U component and therefore minimizing the production of plutonium. ' !v el T 1 l[;”! £ r- £ " o r r—i - i | ri - 1 ). £} ) 4-17 The use of HEU as a fissile topping could be achieved by first transporting uranium recovered from the discharged fuel to a secure enrichment facility capable of producing HEU., Next, the HEU fissile topping would be added to the recovered uranium to raise the fissile content of the product to an allowable limit for denatured uranium. The product (denatured) would then be returned to the fabrication plant. MEU{20% 235U)/Th would be used to supply the remainder of the makeup requirements., Mass flows for this option in which HEU is used as a fissile topping are reported in refs. 2 and 6. For Case D, in which the recycle fuel is not reenriched by addition of HEU fissi]e topping, about 35% more plutonium is bred over 30 yr (~60% more in equilibrium) than when the HEU is used as a fissile topping. The 30-yr cumulative U30g and SWU requirements for the case in which HEU is used as a fissile topping are 4120 ST U;05/GWe and 3940 x 103 SWU/GWe respectively at a 75% capacity factor and 0.20 w/o tails.? Table 4.1-3. Isotopic Fractions of In addition to the uranium fuel cycles Plutonium in Pu0,/ThO, PURs discussed above, two different Pu/Th cases were analyzed, As indicated in Table 4.1-3, the degradation of the fissile percentage of the Equilibrium Once-Through Cycle Charged Discharged plutonium which occurs in a single pass (i.e., 3Py 0.5680 0.2482 once-through) is rather severe. Thus, in addi- 240py - 0.2384 0.3742 tion to the plutonium recycle case (Case H) a 241py 0.1428 0.2207 case was considered in which the discharged 242py 0.0508 - 0.1568 plutonium (degraded isotopically by the burnup) Fissile 0.7108 - 0.4689 is not recycled but rather is stockpiled for Plutonium later use in breeder reactors (Case I). Only limited analyses of safety parameters have been performed thus far for the al- ternate fuel types. Combustion Engineering has reported some core physics parameters for thorium-based (Pu0,/Th0,) and uranium-based (Pu0,/238U0,) APRs,* and the remaining discus- sion in this section is taken from their analysis:3 In general, the safety-related core physics parameters (Table 4.1-4) of the two burner reactors are quite similér; indicating comparable behavior to postulated accidents and plant transients. Nevertheless, the following differences are noted. The effective delayed neutron fraction_(seff) and the prompt neutron lifetime (t*) are smaller for the thorium APR. These are the controlling parameters in the reactor's response to short-term (vseconds) power transients. However, the most 1imiting accident for this type transient is ‘usually the rod ejection accident and since the ejected rod worth is less for the thorium APR, the consequences of the smaller values of these kinetics parameters are largely mitigated.' o 5 | : ~ The moderator and fuel temperature coefficients are parameters which affect the inherent safety of the core. In the power operating range, the combined_responses of these reactivity feedback mechanisms to an increase in reactor thermal power must be a decrease in core reactivity. Since both coefficients are negative, this requirement is easily satisfied. The fuel temperature coefficient is about 25% more negative for the *All-plutonium reactors. 4-18 thorium- APR, while the moderator temperature coefficient is approximately 20% less nega- tive. These differences compensate, to a large extent, such that the consequences of accidents which-involve a core temperature transient would be comparable. For some accidents, however, individual temperature coefficients are the controlling parameters,- and for these cases the consequences must be evaluated on a case-by-case basis. Control rod and soluble boron worths are strongly dependent on the thermal-neutron diffusion length. Because of the Tlarger thermal absorption cross section of 232Th and ~ the higher plutonium loadings of the thorijum APR, the diffusion length and, consequently, the control rod and soluble boron worths are smaller. Of primary concern is the mainte- Table 4.1-4. Safety-Related Core Physics Parameters for Pu-Fueled PWRs Third-Cycle - Third-Cycle Uranium APR Thorium APR Effective Delayed Neutron Fraction BOC ) ~.00430 0.00344 EOC ' . .00438 0.00367 Prompt Neutron Lifetime (x 100 Sec) BOC 10.54 9.03 EQC 12.53 11.30 Inverse Soluble Boron Worth (PPM/% Ap) BOC 221 270 EOC 180 217 Fuel Temperature Coefficient (x 10~5ap/°F) BOC -1.13 -1.40 EOC , -1.15 —1.42 Moderator Temperature Coefficient (x 10-%ap/°F) BOC -1.65 —1.31 EOC -3.32 -2.60 Control Rod Worth (% of U0, APR) ' BOC _ - 90 EOC - 96 nance of adequate shutdown margin to compensate for the reactivity defects during postu- lated accidents, e.g., for the reactivity increase associated with moderator ccoldown in the steam-line-break accidert. The analysis of individual accidents of this type would have to be performed to fully assess the consequences of the 10% reduction in control-rod worth at the beginning of cycle. The overall results of the above comparison of core physics parameters indicate ~ that the consequences of postulated accidents for the thorium APR are comparable to those of the uranium APR. Furthermore, this comparison indicates that other than the possi- bility of requiring additional control rods, a thorium-based plutonium burner is feasible and major modifications to a PWR (already designed to accommodate a plutonium-fueled core) are probab]y not required, although some modifications might be desirable if reactors were spec1f1ca11y designed for operation with high-Th content fue]s. ' - Fephriorer iy e -y T T l[ifl -~ £ £ o) ey D) o L o e © u & g 4-19 4.1.2. Boiling Water Reactors Mass flow calculations for BWRs presented in this chapter were performed by General Electric. A description of the fuel assembly designs developed by General Electric for the utilization of thorium js presented in Ref. 8. The following cases have been analyzed: 8 10 | Dispersible Resource-Based Fuels A. LEU, no recycle. B. MEU/Th, ro recycle. B'. LEU/Th mixed lattice (LEU and ThO, rods), no recycle. B". LEU/MEU/Th mixed lattice (LEU/Th, MEU/Th, and ThO, rods}, no recycle. D. LEU/MEU/Th mixed lattice, recycle of uranium, 235U makeup. Dispersible Denatured Fuel E. Denatured 233y, recycle of uranium, 233U makeup. Energy-Center-Constrained Fuels F. LEU, recycle of uranium and self-generated plutonium, 235U makeup. G. Pu/238U, recycle of plutonium, plutonium makeup. H. Pu/?32Th, recycle of plutonium, plutonium makeup. Case A represents the current mode of BWR operation. Case B involves the replacement of the current LEU fuel with MEU/Th fuel in which the initial uranium enrichment is 1imited to 20% 23547238y, Cases B' and B" represent partial thorium loadings that could be utilized as alternative stowaway options. In Case B' a few of the LEU pins in a conventional LEU lattice are replaced with pure ThO, pins, while in Case B" some LEU pins in a conventional Tattice are replaced by MEU/Th pins and a few others are replaced with the pure ThO, pins. These cases are in contrast with Case B in which a "full" thorium lToading is used (U0,/ThO, in every pin). Case D represents the extension of Case B" to the recycle mode; however, only the uranium recovered from the Th-bearing pins is recycled. Cases F-H represent possible fissile/fertile combinations for use in secure energy centers. Table 4.1-5 provides a summary of certain mass ba1ance information for BWRs operating on these fuel cycles. All recycle cases involve a two-year ex-reactor delay for repro- cessing and refabrication. S ' ' o _ As was shown in Table 4.1-1 for PWRs, the'intrdduction of thorium into a BWR core inflicts a penalty with respect to the resource:rEQUirements of the reactor (compére U30g and SWU requirements of Cases A and B). However, as pointed out above, Case B is for a full thorium loading. In the two General Electric fuel assembly designs® repfesented by Cases B' and B" a much smaller fissile inventory penalty results from the introduction of thorium in the core. {Similar schemes may also be feasible for PWRsS. ) ' 4-20 Table 4.1-5, Fuel Utilization Characteristics for BWRs Under Various Fuel Cycle Options® Separative Work U30g Requirement Requirement Initial Equilibrium Cycle : {ST/GWe) (103 kg SWU/GWe} Fissile Fissile Fissile ’ : . Inventory Charge Discharge - Burnup 30-yrb 30-yrb Case Fuel Type (kg/GWe) (kg/GWe-yr) (kg/GWe-yr) (MWD/kg HM) Initial Total Initial Total Dispefsible Resource-Based Fuels . A LEU, no recycle 22005 799 2355 235 235y 28.4 aged pos1d 2359 3499% 150 Puf B MEU/Th, no recyclee - 1132 235 244 zg:u 31.6 i - 8680f - - 7763f 428 U : 83 Puf B' LEU/Th mixed lattice, - 854 235y 24 23y 28.7 - 620]f. - 3836 S no recyclte? 243 235y ‘ 138 puf ‘ B* LEU/MEU/Th mixed lattice, - 917 2350 125 233y 30.0 - gas2’ - 51007 no recycle® 277 233y 92 puf D LEU/MEU/Th mixed - 147 233 152 233y 30.5 - 5503f - 3895f lattice, self-generated 742 2359 245 235y U recyclie? 98 Puf Dispersible'oenatured Fuel E Denatured 233U0,/ThO,, - 770 233y a52 233y 31.6 0 0 0 0 - U recycle (exogeneous - 15 235y 17 23?0 233)) makeup)@ 55 Pu Energy-Center-Constrained Fuels F LEU, recycle of U + - - e self-generated Pu 2200¢ 28.4 496 38698 235" 19809 6" Pu0,/U0,, Pu recycle - 71 235 38 235y 27.7 i i i i n7s puf 808 Puf H Pub,/ThD,, Pu recycle® - 1705 puf 275 233y 29.8 0 0 0 0 954 Puf a All cases assume 0.2 w/o tails and 75% capacity factor; blank columns included to show no data correspording to that gi fo PMRs (Table 4.1-1) are available. PoRaing given tor bNo credit taken for end-of-reactor-life fissile inventory. CInitial cycle is 1.47 yr in length at 75% capacity factor. Frgm ref. 9.; Based on three-enrichment-zone initial core, axjal blankets and improved refueling patterns which are currently pe1ng retrofitted inte many BWRs. 30-yr U0g and SWU requirements supplied to INFCE for a reference BWR not employing these improvements are 6443 ST U0g/GWe and 3887 x 103 SWU/GWe respectively. eAna'lyses performed for equilibrium cycle only. 6Approximated from equilibrium cycle requirements. 9From ref. 8. fiFrom ref. 10; adjusted from 80% capacity factor to 75%. “Tails uranium used for plutonium diluent. Case B' is a perturbation to the reference UQ, BWR assembly design in that the four UO, corner pins in each fuel assembly are replaced with four pure ThO, pins. The remaining U0, pins are adjusted in enrichment to obtain a desirable local power distribution and to achieve reactivity lifetime. In the once-through mode this deSign increases U30g require- ments by only 2% relative to the reference design. This option could be extended by removing the ThO, corner pins from the spent fuel assemblies, reassembling them into new assemb)ieé, and reinserting them into the reactor. This would permit the ThO; pins to achieve increased burnups (and also increased 233y production) without reprocessing. U305 requirements for this scheme (i.e., re-use of the ThO, rods coupled with U0, stowaway) are approximately 1.3% higher than for the reference U0, cycle.® R 2 - 1 o ot T ‘.-.‘i e‘:} e e R a7 r | i | o ) 1 ~ | v — A o -’ r i-rc”) ) 4-21 Case B" is a modification of Case B' in that in addition to the four ThO, corner pins, the other peripheral pins in the assembly are composed of MEU(235)/Th. The remainder of the pins contain LEU. In the once-through mode this design increases U30g requirements by 12% relative to the reference BWR U0, design. Both Case B' and Case B" would offer ‘operational benefits to the BWR since they have & less negative dynamic void coefficient than the reference U0, design.® This is desirable since the sensitivity to pressure transients is reduced. As shown in Table © 4,1~&, in equilibrium conditions a BWR employing the ThO, corner pin once-through de- sign would discharge 24 kg 233U/GHe annually while the BWR employing the peripheral ThO, mixed lattice design would discharge 125 kg 233U/GWe annually. Use of these options in the once-through mode not only could improve the operational performance of the BWR but also would build up a supply of 233U, This supply would then be available if a denatured 233U cycle (together with reprocessing) were adopted at a later time. Furthermore, use of the mixed lattice designs could be used to acquire experience on the performance of thorium-based fuels in BWRs, Similar schemes for the use of thorium in the once-through mode may also be feasible in PWRs, Although only limited scoping analysis of the safety parameters involved in the use of alternate fuels in BWRs has been performed,® the BWR thorium fuel designs appear - to offer some advantageous trends over UQ, designs relative to BWR operations and safety. Uranium/thorium fuels have a less negative steam void reactivity coefficient than the U0, reference design at equilibrium. This effect tends to reduce the severity of overpressurization accidents and improve the reactor stability. The less negative void ‘reactivity coefficient for the denatured 233/Th fuel indicates that the core will have a flatter axial power shape than the reference U0, design. This could result in an increase in kW/ft margin and increase the maximum average planar heat generation ratio {MAPLHGR). Alternatively, if current margins are maintained, the flatter axial power shape could be utilized to increase the power density or to allow refueling patterns aimed at improved fuel utilization. ' ' References for Section 4.1 1. N. L. Shapiro, J. R Rec, and P A. Hatzie (Combustion Eng1neering), "Assessment of Thorium Fuel Cycles in Pressur1zed Water Reactors,” EPRI NP-359 (Feb. 1977). 2. "Thorium Assessment Study Quarter]y Progress Report for Second Quarter Fiscal 1977," ORHL/TM-5949 (June 1977). 3. R. A. Matzie, J. R. Rec, and A, N. Terney, ”An'Evaluation of Denatured Thorium Fuel Cycles in Pressurized Water Reactors," paper presented at the Annual Meeting of the American tluclear Society, June 12-16, 1977, New York, Hew York. 10. 4-22 Letter from R. A. Matzie (Combustion Engineering) te H. B. Stewart (Muclear Technology Evaluations Company), "U304 Requirements in LHRs and SSCRs,” July 29, 1977. Letter from R, A. Matzie (Combustion Engineering) to J. C. Cleveland (ORNL), "Mass _Ba1ances for Various LWR Fuel Cycles," May 1977. "Quarterly Progress Report for Fourth Quarter FY 77, Thorium Assessment Program," Combustion Engineering. , "Nuclear Power Growth 1974-2000,“ Office of Planning and Analys1s, U.S. Atemic Energy. Commission, WASH 1139(74), (February 1674). " "Assessment of Utilization of Thorium in BWRs " ORNL/SUB 4380/5, NEGD-24073 (January 1978). "Monthly Progress Report for August 1978, NASAP Preliminary BWR Uranium Utilization Improvement Evaluations,” General Electric Co. '"Appraisal of BWR Plutonium Burners for Energy Centers," GEAP-11367 (January 1976). = ¥ D T ror— I(t:; T "% £ L e r— | ot )y ) Ty £ t o ] fih et 7 Y | . i 4-23 4.2. SPECTRAL-SHIFT-CONTROLLED REACTORS N. L. Shapiro Combustion Engineering, Inc. The Spectral-Shift-Controlled Reactor (SSCR) is an advanced thermal converter reactor that is based on PWR technology and offers improved resource utilization, partic- ularly on the denatured fuel cycle. The SSCR differs from the conventional PWR in that it is designed to minimize the number of reactions in control materials throughout the plant 1life, utilizing to the extent possible captures of excess neutrons in fertile material as a method of reactivity control. The resulting increase in the production of fissile material serves to reduce fuel makeup requirements. ‘ In the conventional PWR, long-term reactivity control is achieved by varying the concentratidn of soluble boron in the coolant to capture the excess neutrons generated throughout plant life. The soluble boron concentration is relatively high at beginning of cycle, about 700 to 1500 ppm, and is gradually reduced during the operating cycle by the introduction of pure water to compensate for the depletion of fissile inventory and the buildup of fission products. ' The SSCR consists basically of the standard PWR with the conventional soluble boron reactivity control system replaced with spectral-shift control. Spectral-shift control is achieved by the addition of heavy water to the reactor coolant, in a manner analogous to the use of soluble boron in the conventional PWR. Since heavy water is a poorer moderator of neutrons than light water, the introduction of heavy water shifts the neutron spectrum in the reactor to higher energies and results in the preferential absorption of neutrons in fertile materials. In contrast to the conventional PWR, where absorption in control absorbers is unproductive, the absorption of excess neutrons in fertile material breeds additional fissile material, increasing the conversion ratio of the system and decreasing the annual makeup requirements. At beginning of cycle, a high {approximately 50-70 mole %) D,0 concentration is employed in order to increase the abSorption of neutrons in fertile material sufficiently to control excess reactivity. Over the cycle, the spectrum is thermalized by decreasing the D,0/H,0 ratio in the coolant to compensate for fissile material depletion and fission-product buildup, until at end of cycle essentially pure light water (approximately 2 mole % D,0) is present in the coolant. The basic changes required to implement spectral-shift control in a conventional ~ PWR-are illustrated in a simplified and.SChematic formrin Fig. 4.2-1. In the conventional PNR.'pure water is added and borated water is removed duhing the cycle to compensate for the depletion of fissile material and buildup of fission-product poisons. The borated water removed from the reactor is processed by the boron contentratpr which separates the discharged coolant into two streams, one containing pure unborated water and the second 4.24 ORNL-DWG 78-15056 Conventional ' Spectral Shift Poison Control Control E L ‘. & 1200 | Depletion-Reactivity = Control = 60 o - Fr = [¥ L = =z = - f 5 o S od - & e Ly b1t BOC Burnup EOC H»0 H,0 H,0 |Hs0 2 Makeup S— Makgup 2 - Tank ‘ Tank 1 Charging ' Charging Pump / , Pump Ho0 Borated ' HZO ' D,0/H,0 Boron L H20 0,0 _ Dzo !fixtur‘e Boron Concentrator 2 Upgrader / \ Storage of Highly Storage of Highly Concentrated Concentrated Boric Acid (80%) D,0 Fig., 4.2-1, Basic Spectral Shift Control Modifications, containing boron at high concentrations. The latter stream is stored until the beginning of the subsequent cycle where it is used to provide the boron necessary to hold down the excess reactivity introduced by the loading of fresh fuel. The SSCR can consist of the identical nuclear steam supply system as employed in a conventional poison-controlled PWR, except that the boron concentrator is replaced with a D,0 upgrader. The function of this upgrader is to separate heavy and light water, so that concentrated heavy water is available for the next refueling. The upgrader consists of a series of vacuum distil- lation columns which utilize the differences in volatility between 1ight and heavy water to effect the separation. Although the boron concentrator and the upgrader perform analogous functions and operate using similar processes, the D,0 upgrader is much larger and more sophisticated, consisting of three or four towers each about 10 ft in diameter and 190 ft tall. Although Fig. 4.2-1 illustrates the basic changes required to implement the shift—control concept, numerous additional changes will be required to realize spec- tral-shift control in practice. These include modifications to minimize and recover D,0 leakage, to facilitate refueling, and to remove boron from the coolant after refueling. A - r (”! f - r- | S ! T L . i | 4-25 Initial analyses of spec¢tral-shift-controlled reactors were carried out in the U.S. by M. C. Edlund in the early 1960s and an experimental verification program was performed by Babcock & Wilcox both for LEU fuels. and for HEU/Th fuels.! Edlund's studies, which were performed for reactors designed specifically for spectral-shift control, indicated that the inventory and cbnsumption of fissile material could be reduced by 25 and 50%, respectively, relative to poison control in reactors fueled with highly enriched 235U and thorium oxide, and that a 25% reduction in uranium ore requirements could be realized with spectral shift control using the LEU cycle.2 The spectral-shift-control concept has been demonstrated by the Vulcain reactor experiment in the BR3 nuciear p]aht at Mol, Belgium.3 The BR3 plant after two years of operation as a conventional PWR was modified for spectral-shift-control operation and successfully operated with this mode of control between 1966 and 1968. The Vulcain core operated to a core average burnup of 23,000 MWdA/T (a peak burnup of around 50,000 Md/T) and achieved an average load factor and primary plant availability factor of 91.2 and 98.6, respective]y.“ The leakage rate of primary water from the high—pressure reactor system to the atmosphere was found to be negligible, about 30 kg of D,0-H,0 mixture per year.3 After the Vulcain experiment was completed, the BR3 was subsequently returned to conventional PHR operation. In addition to demonstrating the technical feasibility of spectral-shift control, the Vulcain experiment served to identify the potential engineering problems inherent in converting existing plants to the spectral-shift mode of control. At the time of the major development work on the SSCR concept,; fuel resource con- servation was not recognized as having the importance that it has today. Both uranium ore and separative work were relatively inexpensive and the technology for D0 concen- tration was not as fu11y developed as it is now. With the expectation that the plutonium- fueled breeder reactor would be deployed in the not too distant fdture, there appeared to be 1ittle incentive to pursue the spectral-shift-controlled reactor concept. The dec1s1on to defer the commerc1a1 use of plutonium and the commercial plutonium- fueled breeder reactor is, of course the primary motivation for reeva1uat1ng advanced converters, and the principal incentive for cons1dering,the spectral-shift-controlled reactor is that the potential gains in resource_utilization possible with the SSCR con- cept'may be obtainable fiith changes 1arge1y Timited. to aneillary components and subsystems in ex1st1ng PWR systems. The’ prospects of rapid acceptance and dep10yment of the SSCR are also enhanced by the Tow risk inherent 1n the concept. Since the SSCR can always be operated in the convent1ona1 poison control mode, there would be a reduced risk to station ' generat1ng capacity if the SSCR were deployed and financial risk would be limited to the cost of the additional equipment required to realize spectral-shift control, which is estimated to be only a few percent of the total cost of the plant., The risk, with respect both to capital and generating capac1ty, is thus much lower than for other alternate reactor systems. 4-26 It may also prove feasible to backfit existing pressurized water reactors with spectral-shift control. Such backfitting might\possib]y be performed in some completed plants where the layout favors modifications. waever,,even when judged feasib1e, the benefits of backfitting would have to be gréat to jdstify the cost of replacement power during plant modification. A second and potentially more attractive alternative is thé . possibility of modifying plants still in the early stage of construction for spectral- shift control, or of incorporating features into these plants which would allow conversion to spectral-shift control to be easily accomplished at a later date. | In order to establish the potential gains in resource utilization which might be realized with spectral-shift control, scoping mass balance calculations have been performed by Combustion Engineering for SSCRs operating on both the LEU cycle and on thorium-based cycles, including the denatured 233U cycle.> The calculations were performed for the C-E system SOTM core and lattice design, with the intent of updating the earlier analyses re- ported by Edlund to the reactor design and operating conditions of modern PWRs using state- of-the-art analytic methods and cross sections. Preliminary results from this evaluation are presented in Table 4.2-1. Note that these results were obtained using the standard System 80 design and operating procedures, and no attempt has been made to optimize either the lattice design or mode of operation to fully take advantage of spectral-shift control. For the LEU throwaway mode, Table 4,2-1 indicates a reduction of roughly 10% both in ore reguirements and in separative work requirements relative to the conventional PWR (compare with Case A of Table 4.1-1). If uranium recycle is allowed, the SSCR also reduces " the ore demand (and separative work) for the MEU/Th case by about 20% {compare with Case D in Table 4.1-1). Of particular interest to this study is the reduced equilibriim cycle makeup re- quirements for the spectral-shift reactor fueled with 233U, As indicated, the equilibrium cycle makeup requirement is 236 ' = 233U/GWe-yr as opposed to 304 kg 233U/GWe-yr for the ;onventional PWR (see Case E in Table 4.1-1). The reduced 233y requirements, coupled with the slightly higher fissile plutonium production, would allow a given complement of energy- center breeder reactors to provide makeup fissile material for roughly 40% more dispersed denatured SSCRs than conventional denatured PWRs. A comparison of the Pu/Th case with Case H in Table 4.1-1 shows that the SSCR and PHR are comparable as transmuters. These results are, of course, preliminary and are limited to the performance of otherwise un- modified PWR systems, A more accurate assessment of SSCR performarce, inc]uding the performance of systems optimized for spectral-shift control, will be performed as part of the NASAP program.® The preliminary studies performed to date and the demonstration of spectral- shift control in the Vulcain core have served to demonstrate the feasibility of the concept_and to identify the resource utilization and economic incentives for this | e | T [l £ r—- i ‘ " i - A| 4-27 Table 4.2-1. Fuel Utilization Characteristicg for SSCRs - Under Various Fuel Cycle Options% Equilibrium Cycle 30-Yr Cumulative , Inftial Fissile Fissile 30-yr Cumulative Se . Fissile Inventory Makeup Discharge U305 Requirement R:;fi%:;;efiggk Fuel Type (kg/GWe) (kg/GWe-yr) (kg/GWe-yr) (ST/GWe) (103 kg Swu/GWe) Dispersible Resource-Based Fuels LEU, no recycle 1577 235y 713 235y 182 235y 5320 3010 . 196 Puf ' MEU/Th, 2350 feed, 2580 25U 235y 228 235 3220 3077 U recycle . 371 23y . - 65 Puf Dispersible Denatured Fuel Denatured 233U0,/Th0,, 1663 233y 236 23y 449 233y - - U recycle . 57 23y 72 pPuf Energy-Center-Constrafned Fuel Pu0,/ThO,, Pu recycle 2354 Puf 9 puf 780 Pyf .- » . 273 23y 3 235 :1290-Mwe SSCR; 10-MWe additional power required to run reactor coolant pumps and D,0 upgrader facility. Assumes 75% capacity factor, annual refueling, and 0.2 w/o tails assay. mode of operation. Because the basic PWR NSSS* is used, the utilization of the denatured thorium fuel cycles will pose no additional problems or R&D needs beyond ‘those required to implement this type of fuel in the conventional PWR. Although the general feasibility of spectral-shift control appears relatively well established, nevertheless there are a number of aspects of SSCR design which must be evaluated in order to fully assess the commercial practicality of spectral-shift-controlled reactors. The more significant of these are briefly discussed below. 1. Resource Utilization - A more accurate assessment of resource utilization is required to more definitively establish the economic incentives for spectral-shift control on the LEU cycle. If the concept is to be economically competitive with conventional water reactors, the savings in U;05 and separative work for 235U-based systems must be demonstrated to be sufficiently large to compensate for the additional capital cost of equipment required to implement spectral-shift control. A similar assessment for denatured 233U fuel is also required. | ' | ’ 2. Plant Modifications - The plant modifications necessary to realize spectral- shift control must be identified, and the cost of these modifications established. The practicality and cost of these modifications; of course, bear directly on the economics and commercial feasibility of the concept. Of particular concern are modifications which may be required to 1imit the leakage of primary coolant (from valve stems, seals, etc.) and thé‘equipment required to recover unavoidable primary coolant leakage. Primary coolant leakage is important both from the standpp1nt of economics, because of the high cost of D0, and from the standpoint of radiation hazard, because of the problem of occu- pational eXposures to tritium duking'fioutine maintenance, Other possible modifications to current designs which result from the presence of D,0, such as the increased fast fluence on the reactor vessel and possible changes in pumping power, will also have to be addressed. NSSS = Nuclear Steam Supply System, 4-28 3. Refueling System Modifications - At the end of each operating cycle, spent fuel must be discharged and fresh fuel inserted into the reactor (typically 1/3.of the core loading is replaced each year), and the 1ight water present at end of cycle must be replaced with a D,0-H,0 mixture before the reactor can be returned to power operation. Refueling procedures and equipment must be developed which will allow these operations to be performed with minimum D20 inventory requirements. Minimizing the D,0 inventory is important to the economics and commercial feasibility of the SSCR, since the cost of Dy0 represents roughly 75% of the additional capital expenditures required to realize spectral- shift control. Care must also be taken tc ensure that refueling does not increase outage times because of the adverse effect on capacity factor and the resulting increase in. power cost. The exposure of personnel to tritium generated in the coolant must alsc be mini- mized during refueling operations. 4. D,0 Upgrader Design - Although D,0 upgraders have yet to be employed in con- junction with spectral-shift control, similar units have operated on CANDU reactors, and vacuum distillation columns are also utilized in heavy-water production facilities. Thus, the technical feasibility of the D,0 upgrader can be considered as demonstrated. However, a conceptual upgrader design optimized for the specific demands of the SSCR must be developed so that its cost can be determined. The upgrader is probably the single most significant and costly piece of equipment which must be added to realize spectral-shift control. 5. Licensability and Safety - Although the spectral-shift-controlled reactor is not expected to raise any new safety, licensing or environmental issues except the basic issue of tritium production and containment, a number of core physics parameters are changed sufficiently that the response to postulated accidents must be evaluated. The most significant of these appears to be the somewhat different moderator temperature co- efficient of reactivity, which could lead to a number of potentially more severe accidents early in cycle when the D,0 concentration is relatively high. The D,0 dilution accident must also be addressed; this accident is analogous to the boron dilution accident in the poison-controlled PWR, but the response to D,0 dilution may be more rapid and hence the accident may be potentially more severe than its counterpart in the PWR. Finally, it should be pointed out that while the relationship of the SSCR to the LWR gives it market advantages, it also gives it some disadvantages relative to other alternatives. Although the SSCR demand for U;0g will be less than that of the conventional LWR, the basic properties of light water and the LWR design characteristics inherent in the SSCR will 1imit its fuel utilization efficiency to lower levels than those achievable with other alternatives such as the HWR. On the other hand, the prospect for early and widespread deployment may mean that it could effect a more significant reduction in over- all system U305 demand than might be achievable with other alternatives, even though the inherent resource utilization of an individual SSCR plant may be less than that of other systems. Employing denatured SSCRs would allow additional time to develop effective b et et — g A o e _— T Ty o (| gy \ i —) ) el - 4-29 safeguards for breeder reactors which will eventua]ly‘be required. These breeders might produce 233U, which, as pointed out above, could then be denatured and used in SSCRs. 1. References for Section 4.2 T. C. Engelder, et al., “Spectra] Shift Control Reactor Bas1c Physics Program, BAW- 1253 (November 1963). M. C. Edlund, "Developments in Spectral Shift Reactors," Proceed1ngs of the Third International Conference on the Peaceful Uses of Atomic Energy," Vol. 6, p. 314-320, Geneva, Switzeriand (1964). J. Storrer, "Experience with the Construction and Operation of the BR3/Vulcain Experiment," Symposium on Heavy Water Power Reactors, IAEA Vienna (September 11-15, 1967). J. Storrer, et al., "Belgonucleaire and Siemens PWRs for Small and Medium Power Reactors," Proceedings of a Symposium on Small and Medium Power Reactors, IAEA Oslo (October 12-16, 1970? Letter, R. A. Matzie (Combustion Eng1neer1ng) to J. C. Cleveiand (ORNL), "Mass Balances for SSCRs and LWRs," {May 10, 1977). - Nonproliferation Alternative Systems Assessment Program -~ Preliminary Plan (draft) May 1977. ' 4-30 4.3. HEAVY-WATER REACTORS Y. I. Chang Argonne National Laboratory Due to the low neutron absorption cross section of deuterium, reactors utilizing heavy water as the moderator theoretically can attain higher conversion ratios than reactors using other moderators. As a practical matter, however, differences in the neutron absorption in the structural materials and fission products in the different reactor types make the con- version efficiency more dependent on reactor design than on moderator type. In the study reported here, a current-géneration 1200-MWwe CANDU design'was chosen as the model for ex- amining the effects of various fuel cycle options, including the denatured 233U cycle, on heavy-watér—moderated reactors. The CANDU design differs from the LWR design primarily in three areas: its reference fuel is natural uranium rather than enriched uranium; its coolant and moderator are separated by a pressure tube; and its fuel management scheme employs continuous on-1ine refueling rather than periodic refueling. In the development of the CANDU reactor concept, neutron economy was stressed, trying in effect to take maximum advantage of the D,0 properties. The on-line refueling scheme was introduced to minimize the excess reactivity requirements, Unlike in most other reactor systems, in the natural-uranium D,0 system the payoff in re- ducing parasitic absorption and excess reactivity requirements is direct and substantial in the amount of burnup achievable. These same considerations also make the CANDU an efficient converter when the natural uranium restriction is removed and/or fueling schemes based on recycle materials are introduced, Penalties associated with the improved neutron economy in the naturaleuranius- fueled CANDU include a large inventory of the moderator (the D,0 being a significant por- tion of the plant capital cost), a large fuel mass flow through the fuel cycle and a lower thermal efficiency. 1In enriched fuel cycles, with the reactivity constraint removed, the CANDU design can be reoptimized for the prevailing economic and rescurce conditions. The reoptimization of the current CANDU design involves tradeoffs between economic considerations and the neutron economy (and hence the fuel utilization). For example, the D,0 inventory can be reduced by a smaller lattice pitch, but this results in a poorer fuel utilization. Also, the lattice pitch is constrained by the practical lTimitations placed on it by the refug]ing machine operations. The fuel mass flow rate (and hence the fabrication/reprocessing costs) can te re- duced by increasing the discharge burnup, but the increased burnup also results in a poorer fuel utilization. In addition, the burnup has an impact on the fuel irradiation perform- ance reliability. The fuel failure rate is a strong function of the burnup history, and C ( . o 0ot 1 el e 4 e I 4-31 a significant increase in burnup over the current design would require mechanical design modifications., The thermal efficiency,éan be improved by increasing the coolant pressure. This would require stronger pressure tubes and thus penalize the neutron economy. The use of enriched fueling could result in a higher power peaking factor, which would require a re- duced linear power rating, unless an improved fuel management scheme is developed to re- duce the power peaking factor, Scoping calculations have been performed to address possible design modifications for CANDU fuel cycles other than natural uranium,!™ and detailed design tradeoff and optimization studies associated with the enriched fuel cycles in CANDUs are being carried out by Combustion Engineering as a part of the NASAP program. In the study reported here, in which only the relative performance of the denatured 233y cycle is addressed, the current- genération 1200-MWe CANDU.fuel design presented in Table 4,3-1 was assumed for all except the natural-uranium-fueled reactor. A discharge burnup of 16,000 MWD/T (which is believed to be achievable with the current design) and the on-line refueling capability were also assumed. The fuel utilization charaéteris;ics for various fuel cycle options, including the denatured 233y cycle option, were analyzed at Argonne National Laboratory> and the results are summarized in Table 4,3-2. Some observations are as follows: 1. Natural-Uranium Once-Through Cycle: In the reference natural uranium cycle, the 30-yr U30g requirement is about 4,700 ST/GWe, which is approximately 20% less than the requirement for thé_LWR once-through cycle. Even though the fissile plutonium concentration in the spent fuel is low (~0.27%), the total quantity of fissile plutonium discharged annually is twice that from the LWR, 2. Slightly-Enriched-Uranium Once-through Cycle: With slightly-enriched uranium (1% 235y), a 16,000-MWD/T burnup can be achieved and the U304 consumption is reduced by 25% from the natural-uranium cycle. As shown in Fig. 4.3-1, the dptimum enrichment is in the area of 1,22, which corresponds to a burnup of about 20,000 MWD/T. 3. Pu/U, Pu Recycle: In this option, the natural uranium fuel is "topped” with 0.3% fissile plutonium. A discharge burnup of 16,000 MWD/T can be achieved and the plu- tonium content in the discharge is sufficient to keep the system going with only the natural-uranium makeup, The U30g requirement is reduced to about one half of that for the natura]~ur&n1um_gycle.i (Smaller plutohium toppings decrease the burnup and make the system a net p]utoniufi producer; larger toppings increase the burnup and make the system a net plutonium burner,) ' 4-32 - Table 4.3-1. CANDU-PHW Design Parameters Natural. Uranium Thorium System System Fuel Element Sheath o.d, mm 13.075 13.081 Sheath i.d, mm 12.237 12.244 Sheath material Ir-4 ir-4 Pellet o.d, mm 12.154 12.154 Fuel density, g/cc 190.36 9.4 Fuel material UO2 ThO2 Bundle Number of elements/bundle 37 37 Length, mm 495.3 495.3 Active fuel length, mm : 476,82 475.4 Volume of end plugs, etc., ccC 54.29 65.68 Yoid in end region, cc 24.14 34.99 Coolant in end region, cc 76.69 66.43 Ring 1€No.lradius, mimn) 1/0.0 1/0.0 Ring 2(No./radius, mm) 6/14.885 6/14.884 Ring 3(No./radius, mm 12/28.755 12/28.753 Ring 4(MNo./radius, mm 18/43.305 18/43.307 Channel Number of bundles 12 12 Pressure tube material Zr-Hb Zr-Nb Pressure tube i.d, mm 103.378 103.400 Pressure tube o.d, mm 111.498 111.782 Calandria tube material ir-2 Ir-2 Calandria tube i.d, mm 128.956 129.200 Calandria tube o.d, mm 131.750 131.740 Pitch, mm 285.75 285.75 Core Number of channels 380 728 Net Mie 633 1229 Net thermal efficiency, % 29.0 29.7 Operating Conditions , 020 purity, % 99.75 99.75 Average pin linear power, W/cm 271.3 269.3 Average temperature, C Fuel 936 850 Sheath 290 293 Coolant 290 293 Moderator 68 57 b | ] K= lIZT:c[Z‘”TZ Table 4.3-2. Fuel Utilization Characteristics for CANDUs Under Various Fuel Cycle Options® . Equilibrium Cycle Net Fissile Consumption U,0, Requirement Initial Fissile Fissile Fissile Fissile ‘ Initial: Inventory ‘Charge Discharge Enrichment Burnup - Annual Lifetime? Loading Annual Lifetime Fuel Type - (kg/GWe) (kg/GWe-yr) (kg/Ge-yr) - (% HM) (MWD/kg HM) (kg/GWe-yr) (kg/GWe) (ST/GWe)} (ST/GWe) (ST/GWe) S Dispersible Resource-Based Fuels B Natural U, 897 235y 852 235y . 249 235y° 0.71 7.5 603 235y 25605 235y 164 156 4688 no recycle ‘ - 340 puf -340 puf .=10200 Puf Slightly enriched 1261 235y '~ 561 235y 59 235y° 1.0 16 502 235y 17530 235y 257 14 3563 U, no recycle _ - 183 puf . -183 puf -5490 Puf _ MEU/Th, 2121 235y 1052 235y 336 2350° 1.88 16 - 716 235y 32629 235 538 267 8281 no recycle ' _ _ | 25 puf (20% in V) -25 pyf -750 puf _ 476 233y | -476 233y . .14280 233y MEU/Th, 2121 2350 250 2359 g9 235y 1.65 16 151 235y 6500 235y 538 387 1640° - U recycle , - 30 puf (13% in v) -30 Puf -900 puf - ‘ 6§85 233y 685 233y 0 233y 0 233 x . : Denatured Dispersible Fuel _ _Denatured 1648 233y 831 233y 729 233y 1.46 16 102 233y 4606 233y 0 0 0 233)0,/ThO,, ' ' 32 puf (12% in U). =32 puf -960 Puf U recycle ' ' Energy-Center Constrained Fuel LEY, S 897 235y 399 235y 61 235° NU containing 16 338 235y 10699 235y 164 73 2281 U + Pu recycle 378 puf 168 puf’ 197 puf 0.3% Pu -29 puf -870 puf . ' ' ' Reference Fuel HEU/Th, U recycle 2159 235y 191 235pf 86 235yF 1.91 16 105 235y 5204 235y 548 27t 1331¢ | - 2 puf (93% in U) -2 puf -60 puf ' 750 233y 750 233y ' 0 233y 0 233y gAII cases assume 75% capacity factor. For fresh fuel. No credit, 250 kg minus 99 kg 235U/GWe-yr is equivalent to 63 ST minus 25 ST U;05/GWe-yr; thus annual U305 requirement is 63 - 25=38 ST/GWe. ®excludes transition requirements and out-of-core inventories. 191 kg minus 86 kg 235U/GWe-yr is equivalent to 48 ST minus 21 ST U,05/GWe-yr; thus annual U;04 requirement is 48 - 21=27 ST/GWe. INo credit for end-of-life core. 07 08 09 10 II 12 13 14 15 16 INITIAL 2°°U ENRICHMENT, % Fig. 4.3-1. Fuel Utilization Characteristics for Enriched-Uranium-Fueled CANDU. 4-34 L - 1.0 30 ) T T T T T T T = g U ~ E 0.9 [— 9 u S — 20 5 o 0.3 % TAILS ASSAY 5 , @ > LJ © 08— % O , m 3 0.2 % TAILS ASSAY Ll L i =10 g > 7 ; 0.7 % F < O | | w b o 0.6 | | | | | | | I A L L 4. HEU/Th, U Recycle: With 93% 235U-enriched uranium startup and makeup, the annual U305 makeup requirements at near-equilibrium are about 27 ST/GWe for the 16,000-MWD/T burnup case. This net consumption of U;0g is only 14% of the LWR once-through cycle and 28% of the LWR thorium cycle (see Cases A and J in Table 4.1-1), However, the initial core U30g requirement is more than douB]e that of the CANDU slightly enriched uranium cycle, In addition, the transition to equilibrium and the out-of-core inventory requirements, de- pending on the recycle turn-around time, can be very significant. | : . 5. Denatured U/Th, U Recycle (233U Makeup): The initial core 233U inventory require- ment is about 1,650 kg/GWe, with an annual net requirement of about 100 kg 233U/GWe. 6. MEU/Th, U Recycle (235U Makeup): The initial core requirement is about the same as that for the standard thorium cycle (i.e., HEU/Th cycle); however, the equilibrium net U305 consumption is slighily increased. 7. MEU/Th, No Recycle: This cycle option is included to indicate that recycle of the self-generated 233U is advisable for the MEU/Th cycle. The lifetime U305 requirement for the once-through MEU/Th cycle is about 8,300 ST, which is a factor of 2.3 higher than that for the once-through enriched-uranium cycle in CANDU reactors. " hi o ) r ;\..u.i 1 " .;n.' e ncn o | . ) ]o 4-35 References for Section 4,3 J. S. Foster and E. Critoph, "The Status of the Canadian Nuclear Power Program and Possible Future Strategies," Annals of Nuclear Energy 2, 689 (1975). S. Banerjee, S. R. Hatcher, A. D. Lane, H. Tamm and J. I. Veeder, "Some Aspects of the Thorium Fuel Cycle in Heavy-Pater—Moderated Pressure Tubes Reactors," Nucl. Tech. 34, 58 (1977). E. Cr1toph S. Banerjee, F. \. Bérc]ay, L. Hame1' M. S. Milgram and J. I. Veeder, "Prospects for Self-Sufficient Equliibrlun Thorium Cycles in CANDU Reactors," AECL-5501 (1°7€¢). c. E. Ti11, E. M, Bohn, Y. I. Chang and J. B. van Erp, "A Survey of Considerations Involved in Introducing CANDU Reactors into the U.S.," ANL-76-132 (January 1977). Y. I. Chang, C. E. Ti1l, R. R. Rudolph, J. R. Deen, and M. J. King, "Alternative Fuel Cycle Options: Performance Characteristics and Impact on Huclear Power Growth Potential,” ANL-77-70 (Sept. 1977). , 4-36 4.4, GAS-COOLED THERMAL REACTORS J. C. Cleveland Oak Ridge National Laboratory 4.4,1. High-Temperature Gas-Cooled Reactors The High-Temperature Gas-Cooled Reactor (HTGR) is another candidate for implementing alternate fuel cycle options, particularly the denatured 233U cycle. Unlike other reactor types that generaily have been optimized for either LEU or mixed oxide (Pu/238Y) fuel, the HTGR has a design based on utilization of a thorium fuel cycle, and although current- design HTGRs may not meet potential proliferation-based fuel cycle restrictions, the refer- ence design involves both 232Th and 233U, which are the primary materials in the denatured fuel cycle. In contrast to the fuel for water-cooled reactors and fast breeder reactors, the fuel for HTGRs is not in the form of metal-clad rods but rather is composed of coated fuel particles bonded together by a graphite matrix into a fuel stick. The coatings on the in- dividual fuel particles provide fission-product containment. The fuel sticks are loaded in fuel holes in hexagonal graphite fuel blocks., These blocks also contain hexagonal arrays of coolant channels through which the helium flows. In the conventional HTGR the fuel particles are of two types: fissile particies consisting of UC, kernels coated with layers of pyrocarbon and silicon carbide; and fertile particles consisting of ThO, kernels coated only with pyrocarbon. The pyrocarbon coating on the fertile particles can be burned off while the SiC coating on the fissile particles cannot. Therefore the two particle types can be physically separated prior to any chemical reprocessing. As indicated in Chapter 5, hot demonstrations of the hcad-end processing operations unique to this reactor fuel, the crushing and burning of the fuel elements, the mechanical particle separation, and the particle crushing and burning are needed to ensure that low-loss reprocessing can take place. An inherent feature of the HTGR which results in uranium resource conservation is its high (~ 40%) thermal efficiency. All else being equal, this fact alore results in a 15% reduction in uranium resource requirements compared to LWRs, which achieve a 34% thermal efficiency. This larger thermal efficiency also Teads to reduced thermai discharges that provide significant siting advantages for HTGRs, especially if many reac- tors are to be deployed in central locations such as energy centers. Other factors inherent in HTGR design that lead to improved Ui0g utilization due to the improved neutrcn economy are: _ 1. Absorption of only ~ 1.6% of the neutrons by HTGR particle ccatings, graphite moderator, and helium coolant, compared to an absorption of ~ 5.€% of the neu- 4-37 trons in the Zirca]by cladding and the coolant of conventional PWRs (~4% of all neutron absorptions in PWRs result from hydrogen absorption). 2. Low 233pa burnout due to the low (7-8 W/cm3) power density, The combination of low power dersity and large core heat capacity associated with the graphite moderator and the ceramic fuel largely mitigate the consequences of HTGR loss- of-coolant accidents. Loss of cooling does not lead to severe conditions nearly as quickly as in conventional LWRs or FBRs since the heat capacity of the core is maintained, there- fore allowing considerable time to initiate actions designed to provide auxiliary core cooling. The HTGR offers a near-term potential for realization of improved U305 utilization. The 330-MWe Fort St. Vrain plant has been under start-up for several years with a current Ticensed power level of 70% and the plant has operated at the 70% power level for limited oo el ‘[fi”? | oo periods. A data collection program is providing feedback on problem areas that are becoming apparent during this start-up period and will serve as the basis for improvements in the commercial plant design. An advantage of the HTGR steam cycle is that its commercialization could lead to later commercialization of advanced gas-cooled systems based on the HTGR technology. These include the HTGR gas turbine system which has a high thermal efficiency of 45 to 50% and the VHTR (Very High Temperature Reactor) system for high-temperature process heat applica- tion. ‘ | Mass balance calculatibns have been performed by General Atomic for several alternate HTGR fuel cycles,! and some additional calculations carried out at ORNL have verified certain GA r_esu]ts.2 Their results for the following fuel cycles are presented here: £ Dispersible Resource-Based Fuels 1. LEU, no recycle. a. Carbon/uranium ratio (C/U) = 350. b. C/U = 400, optimized for no recycle. | 2. MEU/Th (20% 235U/Urmixed with 232Th), C/Th = 650, no recycle. 3. MEU/Th (20% 235U/0), C/Th = 306 for initial core, C/Th = 400 for reload segments, 233y recycle, . : S — Dispersible Denatured Fuel 4, MEU/Th (15% 233U/U), C/Th = 2747300 (initial core/reload segments), optimized for uranium recycle (233U + 23%y), ' = T — o Energy-Center-Constrained Fuel - L | 5. Pu/Th, C/Th = 650 (batch-loaded core). _ Reference Fuels s | t; 6. HEU(235U)/Th, C/Th = 214/238 (initial core/reload segments), no recycle, E_J : 7. HEU(233Y)/Th, C/Th = 150, high-gain design, uranium recycle. 8. HEU(235U)/Th, C/Th (from ref. 3). ' 180/180 (initia] cdre/re]oad segments), uranium recycle 4-38 A1l of the above fuel cycles are for a 3360-MWt, 1344-MWe HTGR with a core power den- sity of 7.1 W¢/cm3. Table 4.4-1 provides a summary, obtained from the detailed mass balance information in ref. 1, of the conversion ratio, fissile requirements, fissile discharge, and U30g and separative work requirements, Cases 1-a and 1-b involve the use of LEU fuel with an equilibrium cycle enrichment of 7.4 w/o and 8.0 w/o, respectively. Case 1-b would be preferred for no-recycle conditions. In Case 2 thorium is used with 20% 235U/U (MEU/Th) for no-recycle conditions. Note that while the initial U305 and fissile loading requirements are higher for the MEU/Th case than for the LEU cases, due to the larger thermal absorption cross section of thorium and the partial unshielding of the 238U resonances resulting from its reduced density, the cumulative U305 requirements are slightly less for the MEU/Th case., This results from the high burnup attainable in HTGRs and the resultant large amount of bred 233U which is burned in situ. Other converter and advanced converter reactors (LWRs, SSCRs, and HWRs) typically require less U30g for the LEU case than for the MEU/Th case with no recycle. Case 3 also uses the MEU/Th feed but with recycle of 233U. The unburned 235U and plutonium discharged in the denatured 235U particles is not recycled. The bred 233U re- covered from the fertile particle, however, is denatured, combined with thorium, and recycled. In the calculations for all cases involving recycle of denatured 233U, GA assumed that an isotopic mix of 15% 233U and 85% 238U provided adequate denaturing. Due to the high burnup and the fact that the thermal-neutron spectrum in HTGRs peaks near the 23%Py and 2%1Py resonances, a large amount of the fissile plutonium bred in the denatured fuel is burned in situ, thus resulting in the low fissile plutonium content of the fuel at discharge. Con- siderable 238U self-shielding is obtained by the lumping of the 238U in the coated particle kernels. Studies are currently underway at GA concerning the use of larger diameter fissile particles, thereby lowering the 238U resonance integral and, conseguently, the amount of bred plutonium discharged." Case 4 employs a denatured 233U feed and includes uranium recycle., It represents a feasible successor to Case 3 once an exogenous source of 233U is available. Case 5 involves Pu/Th Fuel. Since no 238U is present in the core, no plutonium is bred; only 233U is bred. This reactor has greatly reduced requirements for control poison, resulting in enhanced neutron economy. This results from the fact that this Pu/Th HTGR essentially achieves the "Phoenix" fuel cycle effect, i.e., the decrease in 23%Pu content is largely compensated for by buildup of 2%lPy from 240Py capture and by buildup of 233y from 232Th capture, resulting in a nearly constant ratio of fissile concentration to 24%0Pu concentratiofi}_ Therefore the fuel reactivity is relatively constant over a long burhup period, reducing the need for control poison. This allows the core to be batch loaded; i.e., the entire core is reloaded at approximately 5-yr intervals. This reload scheme minimizes down time for refueling and eliminates problems of power sharing between fuel elements of different ages. Furthermore, it allows easy conversion to a U/Th HTGR after any cycle. It is important to note that the Pu/Th caée presented in Table 4.4-1'15 not I 4 g . oo Table 4.4-1, Fuel Utilization Characteristics for HTGRs Under Various Fuel Cycle Options Us0g Requirement® Separative Work Requirement® Initial Core Requirements® Equilibrium Cyc\eb (ST/GMe) (10 kg SWi/GWe) o . Discharge of ‘ ‘ Fissile HM Fissile Nonrecyclable 30-yr Total 30-yr Total : Conversion Ratto Inventory - Loading Makeup Fissile Material for CF og for CF og Case, Fuel Type (1st Cy./Eq. Cy.} (kg/GWe) (MT/GWe) {kg/GWe~yr} {kg/GWe-yr) Initial 65.9%/7539¢ Initial 65.9%/75% ‘ Dispersible Resource-Based Fuels 1-a, LEU, 0.580/0.553 901 235y 24.6 U 608 235y 113 235y . 217 427274860 142 331973781 no recycle, - . . 69 Puf C/U = 350 1-b, LEU, '0.557/0.526 - 819 235 21,6 U 576 235y 77 235y 197 4040/4594 130 3188/3629 no recycle, w 52 puf C/U = 400 . . 2, MEU(20% 2350)/Th, 0.630/0.541 1077 235y 5.4 U 551 235 a7 235y 274 3967/8515 249 3640/4143 no recycle, ' : 20.2 Th 74 233y ' C/Th = 650 o - _ - 22 puf 35 MEU(20% 235U)/Th,f ‘ 0.682/0.631' © 1474 235y 7.4 U 397 235y .65 235y n 3229/ 3666 340 293373361 233y pecycle, : o 27.5 Th 36 puf C/Th = 30674009 ' : ‘ ‘ : Dispersible Denatured Fuel : 4, MEU{15% 233U)/Th,f‘ 0.824/0.764 1168 233y 7.9 U 246 233y _ 35 puf 0 0 0 0 U recycle, | L | 30,7 Th C/Th = 274/300 ‘ Energy-Center-Constrained Fuel 5, Pu/Th, 0.617/0.617 3s3 pf® 12,2 Th 630 Puf 102 Puf 0 0 0 0 C/Th = 650 : _ 97 233y ‘ . o Reference Fuels?i ‘ 6, HEU(235U)/Th, 0.723/0.668 - 1358 235y 1.5 U 508 235y 49 235y 345 3864/4395 344 3858/4387 no recycle, , : 37.2 Th 183 233y C/Th = 214/238 ‘ ‘ 1 puf 7, HEU(233U)/Th, 0.915/0.859 1395 233 2,0 U 120 233y - 0 0 a 0 hi/gain, U recycle, ' o139 23y '53.0 Th 12 235y C/Th = 150 | 8, KEU(2350)/Th, | /0.75 1087 2350%5% 44 6 Thdsk 239 235K 1Py’ 50572 /2280 5057+ /2278 hi/gain, 2.1 Wsk 6 235y : U recycle, C/Th = 180/180 glnitial cycle lasts one calendar year at 60% capacity factor, Equilibrium cycle capacity factor 1s 72%. ssumes 0.2 w/o tails. g/alue preceding slash is for an average 30-yr capacity factor of 65.9; value following slash is for a constant capacity factor of 75%. fNo credft taken for end of life core. No 2354 from MEY particle or Pu recycled in Case 3; all U recycled in Case 4, but no Pu recycled. ZInitial core/reload segment, ;Core is batch Toaded; tnitial load provides fissile material for ~5 yr of operation. :Reference fuels are considered only as limiting cases. YInitial cycle length is 1.6 yr. Numbers shown are for a capacity factor of 75%. ¢ 6E-v 4-40 optimized for high conversion; rather it is a Pu burner designed for low fuel cycle costs. A Pu/Th case designed for high 233U production would have a C/Th ratio for the equilibrium cycle of ~430 rather than 650 as in Case 5 (ref., 5). ' In Case 6 the feed is fully enriched {93%) uranium and thorium and no recycle is allowed. Such a system would provide the means for generating a potential stockpile of 233 in the absence of reprocessing capability. If 233U recycle is not contemplated, the economical optimum once-through cycle would have a lower thorium loading (C/Th = 330). ' Case 7 involves the use of highly enriched 233U and uranium recycle. The heavy fer- tile loading (C/Th = 150) results in the high conversion ratio (and high initial fissile loading fequirement) shown in Table 4.4-1. Case 8 involves the use of fully enriched (93%) uranium and thorium designed for recycle conditions. This is included as the pre-1977 reference high-gain HEU(235U)/Th recycle case for comparison with the other above cases. Both GA and ORNL have performed mass balance calculations for an HEU{235U)/Th fuel cycle with uranium recycle.?»% These calculations were for a 1160-MWe plant with a power density of 8.4 wt/cm3 and a C/Th ratio for the first core and reload cycles of 214 and 238 respectively. The GA results indicate cumulative U304 and separative work requirements (for a capacity factor of 75% and an assumed tails enrichment of 0.2 w/o) of 2783 ST U30g/ GWe and 2778 kg SWU/GWe, respectively. The corresponding results for the ORNL calculations are 2690 ST U;04/GWe and 2684 kg SWU/GWe. As can be seen, the agreement is fairly good. Comparison of these resuits with the same case without recycle (Case 6, Table 4.441) shows a U504 savings of ~38% if uranium is recycled. It is conventional to compare 30-yr cumulative U30g and separative work réquirements for different reactor types on a per GWe basis with an assumed constant capacity factor. The results reported in Table 4.4-1 were generated for an assumed variable capacity factor which averaged 65.9% over the 30-yr life. To facilitate comparison with U30g requirements in other sections of Chapter 4, estimated 30-yr requirements for a constant capacfty factor of 75% have also been included in the table, These values were obtained by applying a factor of 0.750/0.659 to the calculated requirements for the variable capacity factor. Obviously this technique is an approximation but it is fairly accurate. The 30-yr require- ments for a 75% capacity factor for Case 8 were explicitly calculated and not obtained by the above estimatihg procedure, As is indicated in Table 4.4-1, the MEU(20% 233U)/Th no-recycle case is more re- source efficient than the LEU no-recycle case. This results from the high exposure attain- able in HTGR fuels and the high in situ utilization of 233U. In water reactors, the once- through MEU{20% 235U)/Th cycle requires significantly more U305 than the once-through LEU cycle. Thus MEU(20% 2350)/Th fuels in HTGRs are an attractive option for stowaway cycles in which 233U is bred for later use. 2 ; il i i i — o e 1 e r . k Radial reflector: 4-41 4.4.2. Pebble-Bed High-Temperature Reactors A second high-temperature gas-cooled thermal reactor that is a possible candidate for the denatured 233U fuel cycle is the Pebble-Bed Reactor (PBR). Experience with PBRs began in August, 1966, in Julich, West Germany, with the criticality of the Arbeitgemeinshaft Versuch Reaktor (AVR), a 46-M{t reactor that was developed to gain knowledge and experience in the construction and operation of a high-temperature helium-cooled reactor fueled with spherical elements comprised of carbon-coated fuel particles. This expérience was intended to serve as a basis'for further development of this concept in West Germany. Generation of electricity with the AVR began in 1967. In addition to generating electric power, the AVR is a test facility for investigat- ing the behavior of spherical fuel elements. It also is a supplier of high-burnup high- temberature reactor fuel e1ements for the West German fuel reprocessing development work. The continuation of the PBR development initiated by the AVR is represented by the THTR at Schmehausen, a reactor designed for 750 MWt with a net electrical output of 300 MW, Startup of the THTR fs expected about 1980. | ' Table 4.4-2. PBR Core Design " The PBR concept offers favorable : ' conservation of uranium resources due to Power, Q, " 3000 th'- its low fissile inventory requirements and Power density 5 M /m3 to the high burnup that is achievable in Heating of helium 250+985 °C PBR elements. .This has been demonstrated Helium inlet pressure '-' 40 atm by the ana]y;is of‘several once-through Plant effi , ‘ . cycles calculated for the PBR by a physics Heig:te:fIEZ::cii1?e/Qt - | ‘553 :z design group? at KFA Julich, West Germany, Radius 539 cm and summarized here. The reactor core de- Ball packing . 5394 balls/m3 sign used for the study is described in Inner fueling zone: _ Tab'le 4.4-2. Various fuel element types OQuter radius | 505 cm were considered, differing by the coated Number of ball flow channels ' 4 particle types uSed_and by the heavy meta) Relative residence time ""9/9/9/9-" loading. .The,basic fuel élement‘design is Ogter fueling zone: : EE “shown in Table 4.4-3, the coated particle uter radius - . 589 cm are decrribed in 4. _ Number of ball flow channels M designs are descr1bgd in Table 4.4-4, and Relative residence time o 13 the compositions of the various fuel ele- Top reflector: - - - ‘'ment types are given in Table 4.4-5. The - Thickness | 7 - 200 once~through cycles considered are de- Graphite,density ' ' 0'32'_' ;scribed below, with the core compositions bottom reflector: - - " - . . . of each given in Table 4.4-6. Thickness r T 150 o ' ' Graphite density . : 1.60: i : : ' Case 1, LEU. Low-enriched uranium Thickness L 100 " is loaded into the coated fuel particles. Graphite density , 1.60 " The radial power profile is flattened by varying the enrichment in the inner and 4-42 Table 4.4-3. PBR Fuel Element Design _outer.radial core zones. The enrichment | | of the inner zone is 7.9 at.% and that of the outer zone is 11.1 at.%. Ball diameter 6 cm Thickness of graphite she)l 0.5 cm L U+ Th fuel Graphite density - 1.70 g/em3 Case 2, MEU/Th. (U + Th)0y fue with 20% enriched uranium is loaded into ‘the coated fuel particles. The heavy metal loadIng in the MEU/Th fuel element is between that of the THTR and AVR elements As in Case 1, the radial power is flattened by the choice of fissile loading of the elements in the inner and outer radial core zones, 6.85 and 11. 4% respectively. The coated particles would require some development and testing. S | ' ' Case 3, Seed and Breed MEU/Th. (U + Th)0, fuel with 20% enriched uranium is loaded into seed elements and Th02 js loaded into breed elements. By thus separat1ng the seed and breed elements, 238U bred into the seed elenents will not .have contaminated the 233U pro- duced in the breed elements in case recycle is opted for later. Graphite balls are added to the inner core zone to adjust the carbon/heavy metal ratio (C/HM) to that of the outer zone. The heavy metal loading of 6 g HM/ball in the seed elements is essentially the | same as in the AVR. The feasibility of a considerably heav1er load1ng of the breed ele- ments, 16. 54 g HM/ball, is currently being tested. Case 4, HEU/Th. (U + Th)O, fuel with 93% enriched uranium is loaded into the coated fuel particles. The coated particle and fuel e]emeht designs are essentially identical to those of THTR fuel e]émehts, which have been licensed and are being manufactured. The only modification is the fissile loading. Again the fissile loading of the elements in the inner and outer radial core zones is varied to flatten the radial power distribution, the inner zone fissile loading being 6.23% of the heavy metal and the outer zone fissile loading being 10.9%. Case 5, Seed and Breed HEU/Th. (U + Thj0, fuel with 93% enriched uranium is loaded into seed elements and breed elements contain ThO, only. The radial power profile is flat- tened by the choice of the mixing fraction of seed and breed balls 1n the inner and outer radial core zones, and graphite balls are added to the inner zone to adapt the C/HM ratio to that of the outer zone. In the seed elements the HEU is mixed with some Th0, in order . to achieve a prompt negative Doppler coefficient. Again the heavy metal 10ad1ng of the balds is essentially the same as that in the AVR and the feas1b1]1ty of the ]oad1ng of the breed elements is being tested. The mass flow data for the equilibrium cycle of each of the five cases are pre- sented in Table 4.4-7. The high thermal cross sections of 23%Pu, 240Py and 241Pu, the soft spectrum, and the'Tow self-shielding of the fuel element design Tead to a very high in-situ utilization of the fissile plutonium (95% for the MEU/Th cycles). In addition, the high burnup results in the low discharge plutonium fissile fractions shown in Table 4.4-7. The bui1dup_of piutohium,isotepes in the MEU/Th cycle is shown in Fig. 4.4-1, " r— - 2 K= e — - « - = rT — e/ ;e — | . 4-43 Table 4.4-4. PBR Coated Particle Design L . r- " Kernel Carbon Coatings Type Dia . . . X idmeter Density Thicknesses . Densities Haterial (um) (g/cn’) (um) (g/cnd) I U/ThO, 400 9.50 85/30/80 1;0/1.6/}.85 II U/ThO, 400 9.50 50/80 1.0/1.85 111 uo, 800 9.50 110/80 1.0/1.85 Table 4.4-5. Composition of PBR Fuel Elements : Heavy Metal Moderation Identification Type of a Leading Ratio Coated Particle (g/ball) (NC/NHM) H1 I 11.24 325 M2 I 8.07 458 $1 11 €.0 617 s I1 6.0 629 1 IT 20,13 180 B2 II 16.54 220 L1 II1 G.88 380 L2 111 11.70 320 G Carbon %See Table 4.4-4. Table 4.4-6. Composition of PBR Core Regions Used in Mass Flow Calculations Inner Core Outer Core Case _ Fuel . ' . - Fuel s Element Type? Fiss('iN]e %fidl)ng " Element Type F1ss(iN1e L/(;‘adi)ng (Fractional Mixing) Vifis’ HM (Fractional Mixing) fis/ "HM 1, LEU 11 (1.0) 0.079 L2 (1.0) 0.111 2, MEU/Th M2 (I.O) 0.0685 M2 (1.0) 0.114 3, Seed and S2 (0.485 0.20 $2 (0.765) 0.20 Breed MEU/Th B2 (0.305 - B2 (0.235) G (0.210) 4, HEU/Th M1 (1.0) 0.0623 M1 (1.0) 0.109 5, Seed and S1 (0.40) 0.27 S1 (0.69) 0.27 Breed HEU/Th B1 (0.39) BT (0.31) G (0.21) %5ee Table 4.4-5. Table 4.4-7. Fuel Utilization Characteristigs for Equilibrium Cycles of PBRs Under Various Fuel Options™ with No Recycle ‘Fuel Conversion Fuel b Loading Discharge Isotopic Fraction Burnup Case Type Ratio Elements {kg/GWe-yr) (kg/GHe-yr) of Discharge Pu {MWD/kg HM) Dispersible Resource-Based Fuels 1 LEU 0.58 L1 + L2 575 235y 93 235y 100 - 80 236U 6168 238y 5719 238y 6743 ytot- 5892 ytot. 42 239(Pu,Np) 0.37 239(Pu,Np) 26 240py 0.23 240py 21 241py 0,19 241py 24 242py 0,21 242py 113 pytot- puf/putots = 0.56 2 MEU/Th 0.58 M2 4158 Th 3881 Th : 100 - 91 233(y,Pa) _ 22 23y 534 235y 39 239 - 79 236U 2163 238y 1965 238y 2697 ytot- 2195 ytot- g 239(py,Np) 0.25 239(Py,Np) 9 240py 0.26 240py 5 241py 0.14 2%lpy 13 242py 0.34 242py 36 pytots puf/putots = 0,39 3 Seed & Breed 0.56 2 | 540 235y 30 235y _ | 201 MEU/Th - 81 236y ' 2190 238y 1982 238y 2730 ytot 2093 ytot: 9 239(py,Np) 0.24 23%(Pu,Np) g 240py 0.25 240py- 5 241py 0.14 241py 14 242py ‘ " 0.38 2h2py 37 pytot- puf/Putot: = 0.38 v B2 4170 Th 3881 Th | 35 82 233(u,Pa) » 22 234U 4 235U 1 236U 108 utot Reference Fuels® 4 HEU/Th 0.59 M 6302 Th 5794 Th ' - 128 233(y,Pa) - : 38 23uu 495 235 ' 23 235 - _ 73 236y a8 238y 30 238y 533 ytot- 292 ytot- 0.263 239(Pu Np) 0.244 2 0.148 2"1Pu 0.512 2k2py 1.166 putot 5 Seed & Breed 0.58 s1 1287 Th 1185 Th - HEW/Th - 25 233(y,pa) - g 234y 496 235y 16 235y - 76 236y 38 238 30 238y 534 ytot. 155 ytot- 239 : 0.227 zqog:u,flp) 0.257 0.120 241py 0.500 242py 1.106 pytot- B1 4983 Th 4594 Th 91 233(y,Pa) 292'0 5 235y 1 238y 126 utot. 2calculated for 1000-Mie plant operating at 75% capacity. See Tables 4.4-3 through 4.4-6 for descriptions of cases and fuel elements. “Reference fuels are considered only as limiting cases, r - - . o puf/pu 0.23 239(Pu,Np) 0.21 24 _ 0.13 2"lPl.:l 0.44 242py Puf/PutOt' = 0.36 243 0.21 23%(Pu,Np) 0.23 240py 0.11 24ipy 0.45 242py tot. < 0.32 r— 100 48 Sy-¥ 4-46 As can be seen, the 23%Pu content peaks at ~ 30 MWD/kg, decreasing thereafter, The higher Pu isotopes tend to peak at higher burnupé so that at discharge 242pu domi- nates. Compared to an LWR with LEU fuel, e Lhe PBR with MEU/Th fuel discharges only &% Haas- Teuenerr-R verren.xra as much fissile plutonium. Furthermore, the Fig. 4.4-1. Buildup of the Plutonium fissile fraction of the discharged plutonium Isotopic Compositien in the MEU/Th Fuel. GR/BRALL is only 39% compared to 71% for an LWR. Table 4.4-8 presents U305 requirements of the various once-through cycles.?.8 The 30-yr cumulative U;0g demands for the MEU/Th once-through cycle and the HEU/Th once- through cycle were determined by explicit 30-yr calculations.® The 30-yr cumulative U;04 demands for the LEU, the seed-and-breed MEU/Th and the seed-and-breed HEU/Th cycles were determined from the U0y demand for the equilibrium cycles and estimates of the inventory of the startup core and of the requirements for the approach to equilibrium.® As can be seen from Table 4.4-8, from the viewpoint of U305 utilization for once- through cycles in the PBR, LEU fuel is the least favorable and HEU/Th fuel is the most favorable with MEU/Th fuel having a U305 utilization between HEU/Th and LEU fuel. It should be noted that the cases presented in Table 4.4-8 do not include recycle of the_bfed'fissi1e material. Under these no-recycle constraints the MEU/Th cases have a 30-yr U;0g demand com- parable to a PWR operating with uranium and self-generated Pu recycle (see Case F, Téble 4.1-3). Thus if recycle were performed with the MEU/Th PBR cases, significantly less U30g would be required than for the PWR with U and Pu recycle. One option for the recycle in the seed~and-breed MEU/Th PBR case would be to cycle the fertile balls back into the feed stream (without reprocessing) for an additional pass through the pebble bed if the irradiation behavior of the fertile balls permits. Table 4.4-8. U305 Requirements for Once-Through PBR Cycles? Case 1, Case 2, Case 3, Case 4, Case 5, - Seed and Breed Seed and Dreed LEU MEU/Th MEU/Th HEU/Th HEU/Th Eaquilibrium cycle 143 135 137 126 12¢ U30g demand, ST/GWE-yr 30-year cumulative d o d . U30g demand,? ST/GWE 4500¢ 4184 4200 4007 4000 %The basis for these requirements is a 1000-Mle plant operating at 75% capacity factor for 30 vears; tails composition is assumed to be 0.2 w/o. Prssumes no recycle. “Estimated value; could differ from an explicit 30-yr calculation by + 3%. dExplicit 30-yr calculation. r— M, e o e, rr e = e 1ti;J | [j errrnenlh = r o T o D T oo l[j - L .y 4-47 References for Section 4.4 Reactor Design Characteristics and FueZ Tnventory Data, compiled by HEDL, September 1977. Thorium Assessment Study Quarterly -Progress Report for Second Quarter Fiscal 1977, ORNL/TM-5949, Oak Ridge National Laboratory (June 1977). Letter, A. J. Neylah, Manager, HTGR Generic Technology Program, to K. O. Laughon, Jr., Chief, Thermal Gas-Cooled Reactor Branch, DOE/NRA, "Technical Data Package for NASAP," March 3. 1978. M. H. Merrill and R. K. Lane, "Medium Enriched Uranium/Thorium Fuel Cycle rarametric Studies for the HTGR," General Atomic Report GA-A14659 (December 1977). Letter, R. F. Turner, Manager, Fuel Cycles and Systems Department, Genera] Atomic Company, to T. Collins, DOE/NPD, May 8, 1978. Thorium Assessment Study Quarterly Progress Report for Third Quarter Fiscal 1977, ORNL/TM-6025, Oak Ridge Nationa] Laboratory. E. Teuchert, et al., "Once-Through Cycles in the Pebble=Bed HTR," Trans. Am. Nucl. Soc., 27, 460 (1977). (Also published as a KFA-Julich report Jul-1470, December 1977T Letter, E. Teuchert (Institut Fur Reaktorenturcklung, Der Kernforschungsanlage Jilich GmbH) to J. C. Cleveland (ORNL), "Once~-Through Cyc]es in the Pebble- Bed HTR," May 19, 1978. 4-48 4,5, LIQUID-METAL FAST BREEDER REACTORS T. J. Burns Oak Ridge National Laboratory A preliminary analysis of the'impact of denatured fuel on breeder reactors was performed by Argonne National Laboratory,! Hanford Engineering Development Laboratory,?2 and Oak Ridge National Laboratory? for a variety of fissile/fertile fuel options. The ana]ys1s concentrated principally on oxide-fueled LMFBRs due to the1r advanced state of deve]opment relative to other potent1a] breeder concepts ' ‘ Table 4.5-1 summarizes some of the significant design and performanbe‘parameters for the various LMFBR designs considered. The procedure followed by each analysis group " in assessing the impact of alternate fuel cycles was essentially the same. A reference design (for the Pu/238U cycle) was selected and analyzed, and then the performance para- meters of alternate fissile/fertile combinations were calculated by replacing the refer- ence core and blanket material by the appropriate alternative material(s). As indicated by Case 1 in Table 4.5-1, a different reference design was selected by each group, emphasizing different design chéracteristics. The three basic designs do share certain characteristics, however. Each is a “"classical™ LMFBR design consisting of two core zones of different fissile enrichments surrounded by blankets (axial and'radia]) of fertile material. In assessing the performance impact of various fissile/fertile com- binations, no attempt was made to modify or optimize any of the designs to account for the better thermophysical properties (e.g., melting point, thermal conductivity, etc.) of the alternate materials relative to the reference system. (Note: The question of selection and subsequent optimization of proliferation-resistant LMFBR core designs is currently being addressed as part of the more detailed Proliferation-Resistant Core Design study being carried out by DOE and its contractors.)* In all cases ENDF/B-IV nuclear data® were utilized in the calculations. The ade- quacy of these nuclear data relative to detailed evaluation of the denatured fuel cycle in fast systems is open to some question. Recent measurements of the capture cross section of 232Th,6 the primary fertile material in the denatured fuel cycle, indicate significant discrepancies between the measured and tabulated ENDF/B-IV cross sections for the energy range of interest. Additionally, the adequacy of the nuclear data for the primary de- natured fissile species, 233U, for the LMFBR spectral range has also been questioned.’ Due to these possible nuclear data uncertainties and also to the lack of design optimiza- tion of the reactors themselves, it is prudent to regard the results tabulated in Table 4.5-1 as preliminary evaluations, subject to revision as more data become available. The compound system fissile doubling time given in Table 4.5-1 was calculated using the simple approximation that C.S.D.T = 0.693 s (Initial Core + Eq. Cycle Charge) ToeE (RF x Eq. Cycle Discharge - Eq. Cycle Charge) g - worell | Reference fuel for LMFBR. CReference fuels are considered only as limiting cases.. 4-49 Table 4.5-1. Fuel Utilization Characteristics and Performance Parameters for LMFBRs Under Various Oxide-Fuel Options . LY \ . Apparent i ' Cgre Specific g?‘“l’?“‘"d Equilidbrium Cycle ‘ ower ssile Initial Net Fissile — Reactor Materials - (Mith per kg Breeding Doubling Fissile Fissile Production Calculation | Axial - Radial Core Vol, Fractions, Capacity Thermal Fissile Ratio, Time Inventory Charge (kg/GHe-¥r) Burnu Parameters, Data Case Core . Blanket. Blanket Fuel/Na/SS/Control Factor Efficiency Material) MOEC {yr) {kg/Gwe) (kg/GWe) 233y,Py (MiD/kg [HM) Dim./Gr./Cy. Contributor - Enérgy-Center-Constrnined Fuels 1 Pus238yd 239y 238y 42/38/20/0 0.75 0.36 1.27 17.2 3424 1647 0,+242 51 2/ ANL ~ -41/44/15/0 0.72 0.32 1.36 9.6 3072 1453 0,+363 2/4/2 HEDL 43/40/15/2 . 0,75 0.39 1,10 1.27 12.7 2270 804 0,+187 88 2/9/12 ORNL 2 Pu/238y 238y . 232Th 42/38/20/0 0.75 0.36 1,27 17.5 3443 1523 +122,+110 51 2mn ANL _ Co : 41/44/15/0 0.72 0.32 1,35 10.4 3077 1540 +150,+197 2/4/2 HEDL o . ' 43/40/15/2 0.75 0.39 1.1 | Co1.27 13.1 2291 804 +154,+30 88 2/912 ORNL 3 Pus238y 2321h 232Th 42/38/20/0 0.75 0.36 1.27 19.5 3480 1674 +298,-77 51 21/ ANL - 41/44/15/0 0.72 0.32 1.34 - 10.8 3093 1545 +299,+35 2/4/2 HEDL =" 4 Pu/Th - 2321 2321h 42/38/20/0 0.75 0.36 1.20 40.2 4016 S mz +798,~662 57 2117 ANL . 41/44/15/0 0.72 - 0.32 . 3 1.19 27.9 3641 1806 +898,-723 2/472 HEDL - 4374071572 0.75 0.39 0.94 ; 1.14 36.1 212 920 +583,-493 95 2/9/12 ORNL . Dispersible Denatured Fuels 5 233y 238y -~ 234 238y 41/44/15/0 0.72 0.32 | 1.20 16.1 2937 1483 -698,923 2/4/2 HEDL 23538 . 2381 232-.‘ — ARV n.-vn 0‘32 ‘ i '!.'!9 !?'3 2955 1488 -566'+778 27472 HEDL ° e o 43/40/15/2 0.75 - 0.39 1.2 . 143 24.2 2038 " 795 -354,+453 92 2/9/12 ORNL 233,238 232 2321 . 42/38/20/0 0.75 0.36 ' 1.16 27.5 3135 1330 -348,+490 51 2/11/1 ANL 7 ks U/? _U Th - 41f44515’/o 0.72 0.32 1.18 19.2 2973 1498 -243,+638 2/4/2 HEDL 43/40/15/2 0.75 0.39 1.25 1,12 26.4 2056 801 -254,+347 92 2/9/12 ORNL 233y/238y 2327y 2327h 43/40/15/2° 0.75 0.39 1.16 1.09 43,0 2208 834 -136,+203 95 | 21912 ORNL +232Th(20%) M r zaéU/zaeu 2ézTh 2321H 43/40/15/2 0.75 0.39 1.10 ! 1.05 118.1 2322 875 -41,+78 98 2/9N2 ORNL +232Th(40%)" ' ' : 3 - Reference Fuels® 10 233y/Th 232t 2327 42/38/20/0 0,75 0.36 | 1.04 -- 3822 1673 +31,0 57 | ML _ 41/44/15/0 0.72 0.32 ; 1.06 154.0 3452 1726 +59,0 2/4/2 HEDL 43/40/15/2 0.75 0.39 1.06 g 1.02 - 2419 m +15,0 99 2/9/12 ORML : sDimensions/Groups/Cycles. i 4-50 where RF is the reprocessing recovery factor (0.98). While such an expression is not absolutely correct, it does provide a measure of the relative growth capability of each reactor. Since the data summarized in Table 4.5-1 are based on three separate reference LMFBRs operating with a variety of design differences and fuel management schemes, the above expression was used simply to prdvide relative values for each system. It should also be noted that some reactor configurations 1isted have dissimilar core and axial blanket materials and thus would probably require modifications to standard reprocessing procedures, The data presented in Table 4.5-1, although preliminary, do serve to indicate cer- tain generic characteristics regarding the impact of the alternate LMFBR fuel options. By considering those cases in which similar core materials but different blanket materials are utilized it is clear that the choice of the blanket material has only a rather small effect on the reactor physics parameters. On the other hand, the impact of changes in the core fissile and fertile materials is considerable, particularly on the breeding ratio. Utilizing 233U as the fissile material results in a significant decrease in the breeding ratio fe1ative to the corresponding Pu-fueled case (ranging from ~ 0.10 to 0.15, depending on the system). This decrease is due primarily to the Tower value of v {neutrons produced per fission) of 233U relative to 23%u and 2“1Pu. Somewhat compensating for the difference in v is the fact that the capture-to-fission ratio of 233U is significantly less than that of the two plutonium isotopes. The differences in breeding ratios given in Table 4,5-1 reflect the net result of these two effects, the decrease in-v clearly dominating. Use of 233 as the fissile material also results in a slight decrease in the fissile inventory required for criticality. This is due to two effects, the lower capture-to-fission ratio of 233y prelative to the plutonium isotopes, and the obvious decrease in the atomic weight of 233y relative to Pu {» 2.5%). . The replacement of 238U by 232Th as the core fertile material also has a significant impact on the overall breeding ratio regardless of the fissile material utilized. As the data in Table 4.5-1 indicate, there is a substantial breeding ratio penalty associated with the use of 232Th as a core material in an LMFBR. This penalty is due to the much lower fast fission effect in 232Th relative to that in 238U (roughly a factor of 4 lower). . The fertile fast fission effect is reflected in the breeding ratio in two ways. First, although the excess neutrons generated by the fission of a fertile nucleus can be sub- sequently captured by fertile material, their production is not at the expense of a fissile nucleus. Moreover, the fertile fission efféct‘produces energy, thereby reducing the fission rate required of the fissile material to maintain a given power level. Since both these effects act to improve the breeding ratio, it is not surprising that use of Th-based fuels result in significant degradation in the bfeeding ratio. A further consequence of the reduced fast fission effect of 232Th is a marked increase in fissile inventory required for criticality, evident from the values given in Table 4.5-1 for the required initial loadings. ' 4-51 The calculations for LMFBRs operating on denatured 233U fuel cover a range of enrich- ments. Cases 5, 6, and 7 assume‘an ~12% enrichment, Case 8 a 20% enkichment, and Case 9 a 40% enrichment, A1l these reactors are, of course, subject to the breeding ratio penalty inherent in replacing plutonium with 233U as the fuel material, The less denatured cases (8 and 9) also reflect the effect of thorium in the LMFBR core spectrum. (These higher enrichment cases were calculated in an attempt to parameterize the effect of varying the amount of-denaturing.) A further point which must be addressed regarding the denatured - reactors is their self-sufficiency in terms of the fuel material 233U, Since the denatured LMFBRs typically contain both 232Th and 238) as potential fissile materials, both 233U and 239py are produced via neutron capture. Thus in evaluating the self-sufficiency of a fast breéder reactor, the 233U component of the overall breeding ratio is of primary importance since the bred plutonium cannot be recycled Back into the denatured system. As illustrated schematically by Fig. 4.5-1, the 233U component of the breeding ratio increases as the allowable denatured enrichment is increased (which allows the amount of thorium in the fuel material to be increased). More importantly, the magnitude of the 233U component of the breeding ratio is very sensitive to the allowable degree of denaturing at the lower enrich- ments (i.e., between 12% and 20%). The overall breeding ratio decreases as the allowable enrichment is raised, but a concomitant and significant decrease in the required 233y makeup presents a strong incentive from a performance viewpoint to set the enrichment as high as is permitted by nonpro]iferatioh constraints. In fact, based on the data summarized in Table 4.5-1, the lowest enrichment 1imit feasible for the conventional LMFBR type systems analyzed.lies in the 11-14% (inner-outer core) range. Such a system would utilize all U0, fuel and would requfre significant amounts of 233U as makeup. (It should be noted that the 233U/Th system is not denatured. It is included in Fig. 4.5-1 because it represents an upper bound on the 233U enrichment.) Since all denatured reactors require an initial inventory of 233U, as well as varying amounts of 233U as makeup material, a second class of reactors must be considered when evaluating the denatured fuel cycle. ,The_pufpose of these systems would be to produce the 233 required by‘the denatured reactors. Possible LMFBR candidates for this role are the Pu/238 reactor with thorium blankets (Cases 2 and 3), a Pu/Th reactor with thorium blankets (Case 4),and a 233U/Th breeder (Case 10).* In the reduced-proliferation risk scenario, all three of these systems, since they are not denatured, would be subject to rigorous safe- guards and operated only in nuclear weapon states or in internationally controlled energy centers. Performance parameters for these three types of systems are included in Table 4,5-1, and the isotopic fissile production (or destruction) obtained from the ORNL calcu- lations is schematically depicted by Fig. 4,5-2. Clearly, each system has its own unique | properties. From the standpoint of 233U7productioh capability, the hybrid Pu/Th system is *See discussion on "transmuters" on p.4-10. LMFBR BREEDING RATIO COMPONENTS 1-2 S // """" 7 ! 0 7 1 V// / ' / g % //A 7 /% 4-52 77-16949 N 12% 20% 40% 2331 /Th Denatured U Denatured U Denatured U Fig. 4.5-1. Mid-Equilibrium Cycle Breeding Ratio Isotopics for , Denatured Oxide-Fueled LMFBRs. (ORNL Cases 7. 8. 9. and 10 from Table 4,5-1) clearly superior. However, it does require a large quantity of fissile plutonium as makeup since it essentially "transmutes" plutonium into 233U, The Pu/238U system with the thorium radial blanket generates significantly less 233U but also markedly reduces the required In fact, for the case illustrated, this system actually produces a slight The 233U/Th breeder, characterized by a very small excess 233U pro- plutonium feed. excess of plutonium, duction, does not provide a means for utilizing the plutonium bred in the denatured systems, and thus it does not appear to have a place in the symbiotic systems utilizing energy-center reactors paired with dispersed reactors. (The coup]ing of each type of fissile production reactor with a particular denatured system is considered in Section 7.2.) _ As a final point, preliminary estimates have been made of the safety characteristics of some of the alternate fuel cycle LMFBRs relative to those of the Pu/238U reference cycle. Initial calculations have indicated that the reactivity change4due to sodium voiding of a 233U-fueled system is significantly smaller than that of the corresponding Pu-fueled system.® Thus, the denatured reactors, since they are fueled with 233U, would have better sodium voiding characteristics relative to the reference system. However, for oxide fuels the reported results indicate that the Doppler coefficient for ThO,-based fuels is com- parable to that of the corresponding 238U0,-based fuels, — - T D - — e Lj - 'NET FISSILE PRODUCTION (KG/GH(E)-YR) L 4-53 Q 77-16948 < 3 | | fffi:’// . / o Q | 7 ///// c? ,Qér—'“‘fjj /522 Pu o ' 233 Pu | ' 233y Pu 233y O o A 1 < § 233/Th - Pu/U + ThO,RB | Pu/Th 1 Fig. 4.5-2. Equilibrium Cycle Net Fissile Production for Possible Oxide-Fueled 233)) production Reactors. (ORNL Cases 10, 2, and 4 from Table 4.5-1) References for Section 4.5 D. R. Hoffner, R. W. Hondie, and R. P. Omberg, "Reactor Physics Parameters of ?1ternate gueled LMFBR Core Designs,” Hanford Engineering Development Laboratory, June 1977). ' Y. I. Chang, C. E. Til1l, R. R, Rudolph, J. R. Deen, and M. J. Ring, "Alternative Fuel Cycle Options: Performance Characteristics of and Impact on Muclear Power Grqwth Potent1a1," RSS-TM-4, Argonne Mational Laboratory, (July 1977). T. J. Burns and J. R, wfiite, ”Pre1ih1nary Evaluation of A1ternatiVe LMFBR -Fuel Cycle Options,"vORNL-5389, (978). _ "The Proliferation-Resistant Preconceptual Core Design Study," J. C. Chandler, D. R. Marr, D. C. Curry, M. B. Parker, and R. P. Omberg, Hanford Engineering Development Labo;atory; and V. W. Lowery, DOE Division of Reactor Research and Technology (March, 1978). . o o . BNL-17541 (ENDF-201), 2nd Edition, “ENDF/B Summary Documentation,” compiled by D. Garber (October 1975). . - R o S w R. L. Macklin and J. Halperin, "232Th(n,v) Cross Section from 2.6 to 800 keV," Nucl. Sei. Eng. 64, No. 4, pp. 849-858 (1977). L S L.'westbn, private communication, March 1971. B. R. Seghal, J. A. Naser, C. Lin, and W. B. Loewenstein, "Thorijum-Based Fuels in Fast Breeder Reactors," Nusl. Tech. 35, No. 3, p. 635 (October 1977). ! 4-54 4,6, ALTERNATE FAST REACTORS 4.6.1. Advanced Oxide-Fueled LMFBRs T. J. Burns Oak Ridge National Laboratory One method of improving the breeding performance of the LMFBRs discussed in the previous section is to increase the core fertile loadings. Typically, this goal is accomplished by one of two means: redesign of the pins to accommodate larger pellet diameters or the use of a heterogeneous design (i.e., intermixed core and blanket assemblies). To maintain consistency with the "classical" designs considered in the ‘ ‘previous section, using the same fuel elements for both concepts, the latter option was pursued to assess the impact of possible redes%gn options. Table 4.6-1 summarizes some preliminary results from calculations for a heterogeneous reactor core model consisting of alternating concentric fissile and fertile annuli (primed cases) and compares them with results from calculations for corresponding homogeneous cores {unprimed cases). As‘the data in Table 4.6-1 indicate, the heterogeneous configuration results in a significant increase in the overall breeding rafio relative to the corresponding homo- geneous calculation. The heterogeneous reactors also require a much greater fissile Toading for criti¢a1ity due to the increase in the core fertile loading. However, the increase in the breeding gain more than compensates for the increased fissile require- ments, resulting in an overall improvement in the fissile doubling time, On the other hand, because of the high fissile loading requirements, it appears that a heterogeneous model for the denatured cases with 12% enrichment (cases 6 or 7 of the previous section) is unfeasible; therefore, an enrichment of ~ 20% was considered as the minimum for the denatured heterogene- ous configuration. While the denatured heterogeneous «configurations result in an increase in the overall breeding ratio, it is significant that the 233y component of the breeding ratio also improves., Figure 4.6-1 depicts the breeding ratio components for both the homo- geneous and heterogeneous denatured configurations. (Again, the 233U/Th LMFBR is included as the upper limit.) As Fig. 4.6-1 indicates, the heterogeneous configurations are clearly superior from the standpoint of 233y seTf-sufficiency (i.e., requiring less makeup requirements). Moreover, if enrichments in the range of 30% - 40% are allowed, it appears possible for a denatured heterogeneous reactor to produce enough 233y to satisfy its own equilibrium cycle fuel requirements. Production reactors would therefore be reguired only to supply the initial inventory plus the additional makeup consumed before the equilibrium cycle is reached. tj oS e e o S e r—C v el C . - " — - O T . Tablé\4.6—l. Comparison of Fuel Utilization Characteristics and Performance Parameters for Homogeneous and Heterogeneous LMFBRs Under Various Oxide-Fuel QOptions - , Equilibrium Cycle Reactor Materials : Fissile Initial ' o Breeding . Doubling Fissile - Fissile Fissile Discharge Axial Internal Radial Ratio, Time (yr) Inventory Charge kg/GWe-yr Case® Driver Blanket ~ Blanket Blanket MOEC (RF=0,98) {kg/GWe) . (kg/GHe-yr) U Pu . Energy-Center-Constrained Fuels 1 Pu/U u - U 1.27 12.7 2270 804 - 991 1 Pu/U U u U 1.50 10.2 3450 1173 - 1517 2 PusU U - | Th 1.27 13.1 2291 . 804 - 154 834 2! ~ Pusu u Th. . Th 1.44 12.9 3725 1250 536 1013 4 Pu/Th ' Th' - Th 1.14 36,1 22 R 920 583 427 4 Pu/Th - Th . Th Th 1.35 18.2 4159 1365 800 808 _ _ Dispersible Denatured Fuels ‘ g? 233y/(U+Th) Th - Th 1.09 43.0 2208 834 698 203 8 233y U Th Th 1.29 18.0 3338 1624 1548 306 g¢ -233U/(U+Th) Th - Th 1.05 112.3 2322 875 835 78 9+d 233y v Th Th 1.29 20.8 4062 1354 1457 108 | S Reference Fuels® | 10 233y/Th Th - Th 1.02 - 2419 911 926 -0 10' 233U/T_h Th Th Th 1.20 30.1 3718 : 1309 1454 0 4Capacity factor is 75%; unprimed cases are for homogeneous cores, primed cases for heterogeneous cores; bsee Table 4.5-1 for case description. : 20% 233ysy, ‘ €40% 233y7U. o Included for illustrative purposes only; exceeds design constraints. “Reference fuels are considered only as limiting cases, — oo [ g5-v 4-56 78-2612 -6 1.2 ------- -y -----{--fl--fi—---‘—-—p———- -k o eh W W 0- 4 LMFBR BREEDINE RATIO COMPONENTS -8 - _OO]’]) AMMHANN 0-0 8 4! 9 9! 10 10 CASE NUMBER 4.6-1. Breeding Ratio Components for LMFBRs Operating on 233y, (Cases 8,8"' for 20% 233U/U and gases 9,9' for 40% 23gU/U Cases 10,10' for 233U?Th with no g38U see Tables 4,5-1 and 4.6-1. The heterogeneous designs also can be employed for the energy-center production reactors recuired by the denatured fuel cycles. As indicated in Table 4.6-1, the three possible production reactors all show significant increases in the quantity of 233U produced. The net production rates are illustrated schematically by Fig. 4.6-2. More importantly, however, use of a heterogeneous core design will allow the isotopics of the fissile material bred in the internal blankets to be adjusted for changing demand requirements without modifying the driver assemblies. For example the internal blankets of the Pu/Th LMFBR could be either ThO, or 2380,, depending on the demand requirements for 233y and Pu. 4 &[fi o T r . C . r-. . T o, T r " K o ¥ . 800-0 4-57 78-2613 | 40?-6 233, N ML AN I 0-0 -400-0 4 -800-0 NET FISSILE PROBUCTION (KG/BW(E)-YR) 10 10* Fig. 4].6"2. with no 238U, Cases 2,2' for Pu and 4,6-1.) 2 2! CASE NUMBER 4 4' issi i ' 233y/Th core Fissile Production Rates for LMFBRs. (Cases 10,10" for .3 /238y core, and Cases 4,4' for Pu/Th core; see Tables 4.5-1 4-58 4.6.2. Carbide- and Metal-Fueled LMFBRS - D. L. Selby P. M. Haas H. E. Knee 0ak Ridge National Laboratory Another method that is being considered for improving the breeding ratios of LMFBRs and is currently under development! is one that uses carbide- or metal-based fuels. - The major advantages of the metal- and carbide-based'fuels are that they will require lower initial fissile inventories than comparable oxide-based fuels and will result in shorter doubling times. This is especially true for metal-based fuels, for which doubling times "as Tow as 6 years have been calculated.2 Since for fast reactors the denatured fuel cycle -would have an inherently lower breeding gain than the reference plutonium-uranium cycle, these advantages would be especially important; however, as discussed below, before either- carbide- or metal-based fuels can be fully evaluated, many additional studies are needed. Carbide-Based Fuels Carbide-based fuels have been considered for use as advanced fuels in conventional Pu/U LMFBRs. Burnup levels as high as 120,000 MWD/T appear feasible, and the fission gas release is less than that for mixed oxide fuels.? Carbide fuels also have a higher thermal_conduc- tivity, which allows higher linear power rates with a lower center—]ine'temperature. In general, the breeding ratio for carbide fuels is higher than the breeding ratio for oxide fuels but lower than that for metal fuels. Both helium and sodium bonds are being considered for carbide pins. At present 247 carbide pins with both types of bonds are being irradiated in EBR-II. Qther differences in the pins include fuel density, cladding type, cladding thickness, type of shroud for the sodium-bonded pin, and various power and temperature conditions. The lead pins have already achieved a burnup level of 10 at.%, and interim examinations have revealed no major problems. Thus there appears to be no reason why the goal of 12 at.% burnup cannot be achieved. In terms of safety, irradiated carbide fuel releases greater quantities of fission gas upon melting than does oxide fuel. Depending upon the accident scenario, this could be - either an advantage or a disadvantage. Another problem associated with carbide fuels may be the potential for large-scale thermal interaction between the fuel and the coolant [see discussion of potential FCIs (Fuel-Coolant Interactions) below]. Metal-Based Fuels Reactors with metal-based fuels have been operating in this country since 1951 (Fermi-1, EBR-I, and EBR-II). Relative to oxide- and carbide-fueled systems, the metal- fueled systems are characterized by higher breeding ratios, lower doubling times, higher heat conductivity, and lower fissile mass. These advantages are somewhat offset, however, by several disadvantages, including fuel swelling problems that necessitate operation at Tower fuel temperatures. r [z L — - = r - - ( ¥ (. = r- r—. B ' r S E; O st r ,'um* e L b 4-59 Most of the information available on metal fuels is for uranium-fissium (U-Fs) fuel. (Fissium consists of extracted fission products, principally zirconium, niobium, molyb- denum, technetium, ruthenium, rhodium, and palladium.) Some information is available for the Pu/U-Zr and U/Th alloy fuels but none exists on Pu/Th metal fuels. (The U/Th fuels do not require the addition of another metal for stability.) In terms of irradiation experience, approximately 700 U-Fs driver fuel elements have achieved burnups of 10 at.% without failure. Less irradiation information is available for the Pu/U-Zr alloy, with only 16 Pu/U-Zr encapculated elements having been irradiated to 4.6 at .% burnup.* Fast reactor experience with U/Th fuels is also quite limited; however, a recent study at Argonne National Laboratory has shown that the irradiation performance of U/Th fuels should be at least as good as that of U-Fs fuels.S ‘With respect to safety, one concern with metal fuels is the possibility of thermal interactions between the fuel and the cladding. For most metal alloys, the fuel will swell to contact the cladding between 3 and 5 at.% burnup. This effect has been observed in jirradiation experiments; however, for burnups up to 10 at.% , no more than 4% of the cladding has been affected. Thus whether or not fuel-cladding interactions will be a ]imiting factor for fuel burnup remains to be determined. For transient overpower (TOP) ana1ysis; the behavior of U/Th elements has been shown to be superior to the behavior of the present EBR-1I fuel (uranium with 5% fissium), the U/Th elements having a 1360°C failure threshold versus 1000%C for the EBR-II elements. Thus U/Th metal pins would have a higher reliability during transfents than the fuel pins already in use in fast reactors. On the other hand, fuel-coolant interaction (FCI) accidents may pre- sent a major problem, more so than for carbide fuels (see below). Potential for Large-Scale FCIs The potential for a large-scale FCI that would be capable of producing mechanical work sufficient to breach the reactor vessel and thereby release radioactivity from the primary containment has been an important safety concern for LMFBRs for a number of years. The assumed scenario for a large-scale FCI is that a large mass of molten fuel (a major portion of the core) present as the result of an hypothetical core disruptive accident (HCDA) contacts and "intimately mixes with" about the same mass of liquid sodium, The extremely rapid heat transfer from the molten fuel (with temperatures perhaps 3000 fo 4000%K) to the much cooler sodium {~1000%K) produces rapid vaporization of the sodium. If the mixing and thermal conditions are ideal, the potential exists for the vaboriza- tion to be extremely rapid, i.e., for a vapor "explosion" to occur with the sodium vapor active as the working fluid to produce mechanical work. A great deal of laboratory experimentation, modeling effort, and some "in-pile" testing has been carried out in this country and elsewhere to define the mechanisms for and the necessary-and-sufficient conditions for an energetic FCI or vapor explosion for 4-60 given materials, particularly for oxide LMFBR fuel and sodium. Although there is no con- clusive theoretical and/or experimental evidence, the most widely accepted theory is that for an energetic vapor explosion to occur, there must be intimate liquid-liquid contact of the fragmented molten fuel particles and the contact temperature at the fuel-sodium surface must exceed the temperature required for homogéneous nucleation of the sodium. A considerable amount of evidence exists to suggest that for oxide fuel in the reactor environment, the potential for a large-scale vapor explosion is extremely remote. The key factor is the relatively low thermal conductivity of the oxide fueI; which does not permit rapid endugh heat transfer from the fuel to cause the fuel-sodium contact tempera- ture to exceed the sodium homogeneous nucleation femperature. The primary difference between carbide and/or metal fuels as opposed to oxide fuels is their relatively higher thermal conductivity. Under typical assumed accident conditions, it is possible to calculate coolant temperatures which exceed the sodium homogeneous nuclea- tion temperature, This does not mean, however, that a large-scale FCI will necessarily occur for carbide-sodium or metal-sodium systems, As noted above, these theories as mecha- nisms for vapor explosion have not been completely substantiated. However, insofar as the homogeneous hucleation criterion is adequate, it is clear that the potential for large- scale vapor explosion, at least in clean laboratory systems, is greater for carbide or metal in sodium than for oxide in sodium. Continued theoretical and experimental study is necessary to gain a thorough understanding of the details of the mechanisms involved and to estimate the 1ikelihood for vapor explosion under reactor accident conditions for any breeder system. Breeding Performance of Alternate Fuel Schemes Table 4.6-2 shows that in terms of fissile production, the reference Pu/U core with U blankets gives the best breeding performance regardless of fuel type {oxide, car- bide, or metal). For the carbide systems considered, a heterogeneous core design using Pu/U carbide fuel with a U carbide blanket gives a breeding ratio of 1.550. For the metal systems considered, a nominal two-zone homogeneous_core design using U-Pu-Zr alloy fuel gives a breeding ratio of 1.614. The increased fissile production capability of the carbide and metal fuels is especially advantageous for the denatured cycles. A breeding ratio as high as 1.4 has been calculated for a metal denatured system, and the breeding ratio for a carbide de- natured system is not expected to be substantially smaller. However, a good part of the fissile production of any denatured system is plutonium. Thus the denatured system is not a good producer of 233U, However, when used with the energy park concept, where the plutonium produced by the denatured systems can be used as a fuel, the denatured carbide “and metal uranium systems are viable concepts. Metal and carbide concepts may also prove to be valuable as transmuter systems for producing 233U from 232Th, ' a (‘. et ) r. o i — c-. . ) r_., ri | 4-61 Table 4.6-2. Beginning-of-Life Breeding Ratios for Various LMFBR Fuel Concepts Breeding Ratio Oxide Carbide Metal Fuel? Blanket Fuels = Fuels Fuels Pu/238y (reference) 238)) 1.44> 71,5507 1.629° 233Y,238y /py-ZIr | 238y 1.614 2337238y /py-Zr Th 1.537 233y/238Yy/py/Th 238y 1.532 233y/238j/py/Th ~ Th 1.406 Pu/Th Th 1.307 1.353° 1.381€ 233Y/Th | Th 1.041 1.044 1.105° 235)/Th ' - Th 0.786 0.817 0.906° 233y/238y-7y (denatured) Th ' .40 2A11 Pu is LWR discharge Pu, PRadial heterogeneous_design. “From ref. 2. Of the thorium metal systems'considered, the U/Pu/Th ternary metal system was found to to be the best 233U producer, Irradiation experiments have shown that the U/Pu/Th alloy can be irradiated at temperatures up to 700°C with burnups of up to 5.6 at.%.° Beginning-of- cycle breeding ratios around 1.4 have been caTculated for this system, and it appears that optimization of core and blanket geometry may increase the breeding ratio to as high as 1.5, It is also clear that the equilibrium cycle breeding ratio may be as much as 10% higher due to the flux increase in the blankets from the 233 production. This system not only is a pure 233y producer (no plutonium is produced), but also acts as a plutonium sink by burning plu- tonium produced in light-water reactors. | Summary and Conclusions Both carbide- and meta]-based fuels have larger breeding gains and potentially Tower doubling times than the oxide-based fuéjs. When the prqliferatiOn issue is considered in the design aspect (especially for 233U/Th cohcepts:with their inherently lower breeding gains), these advantages are enhanced even more. In light of the emphasis on proliferation- - resistant nuc]ear design, the carbide— and metal-fueled reactors have the potent1a1 to contr1bute extens1ve1y to the energy requwrements of this country in the future., However, the first step is to establish carbide and metal fuel data bases similar. to‘the.present data base for oxide fuels, particularly for safety analyses. Present development plans for carbide and metal fuels call for a lead concept se1ect1on for the carbide fuels by ~1981, w1th the metal fue] selection com1ng in m1984 4-62 4,6.3 Gas-Cooled Fast Breeder Reactors T. J. Burns Oak Ridge National Laboratory In addition to the sodium-cooled fast reactors discussed above; the impact of the various alternate fissile/fertile fuel combinations on the Gas-Cooled Fast Breeder Reactor (GCFR) has also been addressed (although not to the degree that it has for the LMFBR). A 1200-MWe Pu/U GCFR design with four enrichment zones was selected as the reference case.?’-8 The various alternative fissile/fertile fuel combinations were then substituted for the reference fuel. No design modifications or optimizations based on the alternate fuel propefties were per- formed. It should also be emphasized that the results of this scoping evaluation for alternate-fueled GCFRs are not comparable to the results given in Section 4.5 for LMFBRs due to markedly different design assumptions for the reference cases. The results of the preliminary calculations for the alternate-fueled GCFRs, sum- marized in Table 4.6.3, reflect trends similar to those shown by LMFBRs; i.e., relative to the reference case, a significant breeding ratio penalty occurs when 233U is used as the fissile material and 232Th as the core fertile material. Moreover, the magnitude ~ of the penalty (aBR) is larger for the GCFR than for the LMFBR. Owing to the helium coolant, the characteristic spectrum of the GCFR is significantly harder than that of a comparably sized LMFBR, In light of the relative nuclear properties of the various fissile and fertile species discussed in Section 4.5, this increased penalty due to the harder spectrum is not surprising. The number of neutrons produced per fission (v) of the fissile Pu isotopes in the GCFR is significantly higher than the number produced in the softer spectrum of an LMFBR. The value of v for 233U, on the other hand, is rela- tively insensitive to spectral changes. Hence, the larger penalty associated with 233).based fuels in the GCFR is due to the better performance of the Pu reference system rather than to any marked changes in 233U performance. A similar argument can be made for the rep1acement of core fertile material. Owing to the harder spectrum, the fertile fast-fission effect is more pronounced in the GCFR than in an LMFBR. Thus, the reduction in the fertile fission cross section resulting from replacement of 238U by 232Th results in a larger decrease in the breeding ratio. It should also be noted that as in the LMFBR case, 233U-fueled GCFRs require smaller fissile inventories than do the corresponding Pu-fueled cases, The better breeding performance of Pu in the harder spectrum of the GCFR, on the other hand, indicates that the GCFR would be a viable candidate for the role of energy center “transmuter," either as a Pu/Th system or as a Pu/U + ThO, radial blanket system. It must be emphasized, however, that these conclusions are tentative as they are based — i e ) e E ;| r- - r n b + ¥ r; o~y r e b - r. r. o i — r.. T oy ':Tm_' ol r r-i e . T 4-63 on only the preliminary data presented in Table 4.6-3. The possibility of employing heterogeneous designs and/or carbide- or metal-based fuels has not been addressed. It should also be noted that evaluation of which type of reactor is best suited for a given role in the denatured fuel cycle must also reflect nonneutronic considerations such as capital cost, possible introduction date, etc. Table 4.6-3. Fuel Utilization Characteristics and Performance Parameters for GCFRs Under Various Fuel Options? (2% losses assumed in reprocessing) Injtial Fissile Equilibrium Cycle Reactor Mat?rials Fissile Breeding Doubling Fissiie Fissile Discharge Axial Radial Inventory Ratio, Time (yr) Charge kg/GWe-yr Core Blanket Blanket (kg/GwWe) MOEC {RF=0.98) {kg/GWe-yr) U Pu Energy—Center~éonstrained Fuels Pu/U U U 2641 1.301 14.3 965 - 1163 Pu/U U Th 2693 1.276 15.4 987 224 941 Pu/Th Th Th 3170 1,150 48,3 1158 626 619 Dispersible Denatured Fuels 233yy® u Th 2538 1.088 50.5 - 1001 671 400 233y° Th. Th 2587 1,074 66.8 1019 822 256 233y7y + ThY Th Th 2720 1.060 98.4 1031 871 208 233/ + Th® Th : Th 2956 1.004 naNn 1054 81 Reference Fuels f , 233y/Th Th Th 3108 0,970 1192 1169 = - - Capacity factor is 75%. 317.9% 23370, 17.7% 233000, 520% 233u/u, 0% 233ysy. Reference fuels are considered only as 1imiting cases. References for Section 4.6 J. M. Simmons, J. A. Leary, J. H. Kittel and C. M. Cox, "The U.S. Advanced LMFBR Fuels Development Program," Advanced LMFBR Fuels, pp. 2-14, ERDA 4455 (1977). Y. I. Chang, R. R. Rudolph and C. E. Till, "“Alternate Fuel Cycle Options (Performance Characteristics and Impact on Power Growth Potential)," June 1977. A. Strasser and C. Wheelock, "Uranium-Plutonium Carbide Fuels for Fast Reactors," Fast Reactor Technology National Topical Meeting, Detroit, Michigan (April 26-28, 1965). W. F. Murphy, W. N. Beck, F. L. Brown, B. J.'Koprbwshi, aner. A. Meimark, “Post- irradiation Examination of U-Pu-Zr Fuel Elements Irradiated in EBR-II to 4.5 Atomic Percent Burnup,” ANL-7602, Argonne National Laboratory (November 1969). B. R. Seidel, R. E. Einziger, and C. M. Walter, “Th-U Metallic Fuel: LMFBR Potential Based Upon EBR-II Driver-Fuel Performance," Trans. Am. Nucl. Soc. 27, 282 (1977). B. Blumenthal, J. E. Sanecki, D. E. Busch, and D. R. 0'Boyle, “Thorium-Uranium- Plutonium Alloys as Potential Fast Power-Reactor Fuels, Part II. Properties and Irradiation Behavior of Thorium-Uranium-Plutonium Alloys," ANL-7259, Argonne National Laboratory (October 1969). Letter to D. E. Bartine from R. J. Cerbone, April 22, 1976, 760422032 GCFR. Letter to D. E. Bartine from R. J. Cerbone, April.22, 1976, 760422032, Subject: 1200~ MWe GCFR Data. ' . { . o i > ; : = ¢ CHAPTER 5 b IMPLEMENTATION OF DENATURED FUEL CYCLES - : Chapter Qutline - 5.0. Introduction, T. J. Burns, ORNL L- 5.1. Reactor Research and Development Requirements, N. L. Shapiro, CE 5.1.1. Light-Water Reactors ;- 5.1.2. High-Temperature Gas-Cooled Reactors 5.1.3. Heavy-Water Reactors b 5.1.4. Spectral-Shift-Controlled Reactors 5.1.5. R,D&D Schedules - 5.1.6. Summary and Conclusions Li 5.2. Fuel Recycle Research and Development Requirements, r. Spiewak, ORNL 5.2.1. Technology Status Summary (T . 5.2.2. Research, Development, and Demonstration Cost Ranges o and Schedules 5.2.3. Conclusions 3} v ¥ P g .5 J —— b r-: A - { i r—i r . . r C7i 5-3 5.0. INTRODUCTION T. J. Burns Oak Ridge National Laboratory Currently, a major portion of the nuclear generating capacity in the U.S. consists of LWRs operating on the LEU once-through cycle. Implementation of the denatured 233U fuel cycle will require that the nuclear fuel cycle be closed; thus research and development efforts directed at nuclear fuel cycle activities, that is, reprocessing, fabrication of fuel assemblies containing recycle material, etc., will be necessary, as well as research and development of specific reactor systems designed to utilize these alternate fuels. To date, most fuel cycle R&D has been directed at closing the Pu/U fuel cycle under the assumption that plutonium would eventually be recycled in the existing LWRs. With the exception of the HTGR (for which a 330-MwWe prototype reactor is undergoing testing at Fort St. Vrain), and the Light Water Breeder Reactor (LWBR) at Shippingport, Pa., U.S. reactors have not been designed to operate on thorium-based fuels, and thus the R&D for thorium- based fuel cycles has not received as much attention as the R&D for the Pu/U cycle. As a result, any strategy for implementation of the denatured fuel cycle on a timely basis must be concerned with fuel cycle research and development. It must also be concerned with reactor-specific research and development since the implementation of the denatured 233U cycle in any reactor will necessitate‘design changes in the reactor. ~ The following two sections of this chapter contain estimates of the research and development costs and possible schedules for the reactor-related research and development and the fuel-cycle-related research and development required for implementation of the denatured fuel cycle in the various types of reactors that have been considered in earlier chapters of this report. It should be noted that these two sections are intrinsically connected: the implementation of a reactor operating on recycle fuel hecessitates the prior imp1ementat%on of the reprocessing and fabrication facilities attendant to that fuel, and conversely, the decision to construct a reprocessing facility for a specific recycle fuel type is dictated by the existence (or projected existence) of a reactor discharging the fuel. 5-4 5.1. REACTOR RESEARCH AND DEVELOPMENT REQUIREMENTS N. L. Shapiro Combustion Engineering Power Systems The discussions in the preceding'chapters, and also the discussion that follows in Chapter 6, all aésume that LWRs and advanced conyerters based on the HTGR, HWR, and 3SCR con- cepts will be available for commercial dperation on denatured uranium-thorium (DUTH) fuels on a re]ative]y_near-term time scale. If this commercialization schedule is to be achieved, substantial reactor-related research and deve]ophent will be required. The purpose of this éection is to delineate to the degree pdssib]e at this preliminary stage of deveiopment:the magnitude and scope of the reactor R,D&D requirements necessary for implementationAof the reactors on DUTH fuels and, further, to determine whether there are significant R,D&D cost differences between the reactor systems.' The refiuireménts listed are those believed to be necessary to resolve the technical issues. that currentiy preclude the deployment of the various reactor cdncepts on DUTH fuels, and no attempt is made to prejudge or to indicate a preferred_fiystem. It is to be emphasized that the proper development of reactor R,D&D costs and schedules would require a comprehensive identification of design and licensing prob]ems, the development of detailed programs to address these problems, and the subsequent deve]obment of costs and schedules based upon these programs. Unfortunately, the assessment of alternate converter concepts has not as yet progressed to the point that problem areas can be fully identified, and so detailed development of R,D&D progfams is generally impractical at this stage. Con- sequently, we have had to rely on somewhat subjective evaluations of the technological status of each concept, and upon rather approximate and somewhat intuitive estimates of the costs required to resolve the still undefined problem areas. A more detailed development of the requirements for many of the candidate systems will be performed as part of the characteriza- tion and assessment programs currently under way in the Nonproliferation Alternative Systems Assessment Program (NASAP). In general, reactor R,D&D requirements can be divided into two major categories: (1) the R,D&D pertaining to the development of the reactor concept on its reference fuel cycle; and (2) the R,D&D necessary for the deployment of the reactor operating on an altern- ate fuel cycle such as a DUTH fuel cycle. In the discussion presented here it is assumed that, with the exception of the HTGR (whose reference fuel cycle already includes thorium), the reference cycles of. the advanced converters would initially be the uranium cycle (i.e., 235y/238y) and that no reactor would employ DUTH fuel until after its satisfactory per- formance had been assured in'a large-plant demonstration. Although it is possible to consider the development of advanced converters using DUTH fuel as their reference fuel cycle, such simultaneous development could be a potential impediment to commercialization since surveys of the utility and manufacturing sectors! indicate a near universal reluctance to embark on either a new reactor technology or a new fue cycle technology, largely because b, ‘ ol ——- r ‘[: | g i € o £ — — r " ¢ ] L 5-5 of the uncertainties with respect to reactor or fuel cycle performance, economics, licens- ability, and the stability of government policies. Thus attempts to introduce a new re- actor technology conditional upon the successful development of an untried fuel cycle tech- nology would only compound these‘Concerns and complicate the a]ready difficult problem of commercialization. The development of advanced converter cencepts intended initially for- uranium fueling would allow research and development, design, and the eventual demonstra- tion of the concept to proceed s1mu1taneous]y with the separate deveiopment of the DUTH cycle. The R,D&D related to the reactor concept itself typically can be divided into three components (1) Proof of principle (operating test reactor of small size). (2) Design, construction, and operation of prototype plant (intermediate size). (3) Design, construction, and operation of commercial-size demonstration plant (about 1000 MWe). | Each stage typically involves some degree of basic research, component design and testing, and Ticensing development. In certain instances, various stages of the development can be bypassed. This is particularly true of technologies representing only a modest departure from the present reactor technology, in which case prototype reactor construction may be bypassed completely and demonstrations performed on commerc1a1 -size units. If a decision is made to do this, the time required to introduce commercial-size units can be shortened, but financial risks are increased because of the larger capita] commi tment required for full- scale units. On the other hand, total R&D costs are somewhat reduced, since some fraction of the R&D required for prototype design usually proves not to be applicable to large-plant - design. It is also possible in certain instances to perform component R&D and design for the prototypes in such a fashion that identical components can.be used directly in the demon- stration units.. Thus, by employing components of the same design-and size in both systems the R&D necessary to scale up components could be avoided. ’ Each of the fhree.advanced converter reactors discussed in this section has already proceeded through the proof-of-principle stage. Of these, the HTGR is the most highly develop- ~ed within the United States, with a 330-MWe prototype currently operating (the Fort St. Vrain plant). HWRs have received much less development within the United States,:but reactors of this type have been commercialized in the Canadian CANDU reactor. However, due to differences in design between the CANDU and the HWR postulated for U.S. siting \ for example, the ex- pected use of slightly enriched fuel. in a U.S. HWR)'and also to differences in licensing 'cr1ter1a, it would still be des1rable to construct a U.S. prototype p]ant before proceeding to the commercial-size demonstration plant phase.. The SSCR represents onTy a modest departure from the design of PWRs already operating, but even so, the construction and operation of a prototype plant would also be the logical next stage in the evolution of this concept. o | 5-6 As has been pointed out above, relatively rapid introduction schedules for the various reactors have been postulated in the nuclear power scenarios described in Chapter 6. This is because one of the objectives of this report is to establish the degree to which advanced converters and the denatured uranium-thorium (DUTH} cycle can contribute to improved uranium resource utilization so as to defer the need for plutonium-fueled breeder reactors and to eliminate from further consideration those concepts which cannot contribute significantly to this goal even if rapidly introduced. The SSCR is assumed to be intro- duced in 1991 and HWRs and HTGRs in 1995. In view of the time requirements for plant construction and licensing, it is clear that the prototype piant stage will have to be bypassed if these introduction dates are to be achieved. Consequently, for the discussion below it has been assumed that the program for each reactor will be directed toward the construction of the demonstration plant. This reactor/fuel cycle demonstration is in turn divided into two parts: one consisting of the generic reactor R&D required to provide the basic information necessary for the design and licensing of a commercial-size demonstration facility; and another consisting of the final design, construction, and operation of the facility. For this demonstration program, continued government funding has been assumed because of the substantial R&D and first-of-a-kind engineering costs that will be incurred and because of the increased risks associated with bypassing the prototype stage. In considering fuel-cycle-related reactor R,D&D, it is assumed that the demonstration of the reactor concept on its reference cycle has beén accomplished and only that R,D&D re- quired to shift to an alternate cycle (specifically a DUTH cycle) need be addressed.* The basic\pypes of fuel-cycle-related reactbr R,D&D are: | | (1) Data-base development. (2} Reactor components development. (3) Reactor/fuel cycle demonstration. The purpose of the data base development R&D is to provide physics verification and fuel performance information necessary for the design and licensing of reactors operating on the subject fuel cycle; the intent here is to provide information similar to that which has been developed for the use of mixed-oxide fuels in LWRs. Physics verification experiments have typically consisted of critical experiments to provide a basis to demonstrate the ability of analytical models to predict such important safety-related parameters as reactivity 1eve], coefficients of reactivity, and poison worths. Safety-related fuel performance R&D might consist of such aspects as fuel rod irradiations to establish in-reactor performance and discharge isotopics; special reactor experiments to establish such parameters as in-reactor swelling, densification, center-line temperature and fission gas release; and tests of the *Note that the R,D&D requirements included are those related to the design, licensing and operation of the reactor only. The requirements for developing the fuel cycle itself are considered separately (see Section 5.2). The prime example of such fuel-cycle-related reactor R,D&D is that already performed for plutonium recycle. Here, fairly extensive R,D&D was performed both by the government and by the private sector to develop reactor design changes and/or reactor-related constraints, Ticensing information, and in-reactor demonstrations to support the eventual utilization of mixed-oxide fuels. % f ' a r i -y { 5-7 performance of the fuel during ant1c1pated operat1ona1 trans1ents Since sueh safety-related fuel performance information would be developed as part of ‘the fuel recycle program dis- cussed in Section 5.2, the R&D costs for this aspect are mentioned here only for completeness. Reactor components development has been included since, in principle, the use of alternate fuels might change the bases for reactor design sufficiently that additional com- ponents development could be required The extent of the reactor design modifications re- quired to accommodate a change from a reactor s reference fuel to denatured fuel would, of course, vary with the reactor type The third aspect of fueT-cyc]e-re]dted R&D is the reactor/fuel cycle demonstration. This demonstretion includes the core physics design and safety analysis, which identifies any changes in design basis events or in reactor design necessitated by the denatured uranium-thorium fuel cycles, the preparation of an analysis report (SAR), and the subse- quent in-reactor demonstration of substantial quantities of denatured fuels. In summary, a number of_aSsumptiqns have been made to arrive at a point of refer- ence for evaluating the research and development required for reactors to be commercialized on a DUTH fuel cycle within the postulated schedule. In particular, it has been assumed that the prototype pTant stage either has been completed or can be bypassed for HTGRs, HWRs, and SSCRs, and thus the remaining R,D&D related to the reactor concept itself 1is that required to operate a commercial-size demonstration plant. The demonstration plants are based on each reactor's reference fuel rather than on a DUTH fuel; to convert the reactors to a DUTH fuel will require additional R,D&D that will be fuel-cycle-related. For the LWRs, which have long passed the demonstration stage on their reference fuel, all the reactor R,D&D required to operate the reactors on a DUTH fuel is fuel-cycle-related. The demonstration program in this case would be the demonstration of DUTH fuel in a current-generation LWR. (Note: This discussion does not consider reactor R,D&b to substantiaTiy improve the resource uti]ization!oerWRs, which, as is pointed out in Section 4.1 and Chapters 6 and .7, 1is currently,being,studied as” one approach for increas- ing the power production from a fixed resource base.) / . : p This evaluation has also required that asSumptions'be made regarding the degree of financial support that'cou1d_be expected from the government. These assumptions, and the criteria on which they are based, are presented in the discussions below on .each reactor type. While the assumptions regarding government participation are unavoidably arbitrary and may be subject_to debate, it is to be pointed out that basically the same assumptions have been made for all reactor types. Thus the reader may scale the costs presented to correspond to other sets of assumpt1ons F1na11y, it is to be noted that while the nuclear power systems included in this study of the denatured 233U fuel cycle include fast breeder reactors, no estimates are included in this section for FBRS Estimated research and development cost schedules. for 5-8 the LMFBR on its reference cycle are currently being revised, and a sfudy of the denatured fast breeder fuel cycle, which includes fast transmuters and denatured breeders, is included as part of the INFCE program (International Nuclear Fuel Cycle Evaluation). The results from the INFCE study should be available in the near future. 5.1.1. Light-Water Reactors ~ Preliminary evaluations of design and safety-related considerations for LWRs operat- ing on the conventional thorium cycle indicate thorium-based fuels can be employed in LWRs with 1ittle or no modification. Consequently, the R&D costs given here have been estimated under the assumption that denatured fuel will be employed in LWRs of essentially present design. This assumption is not meant to exclude minor changes to reactor design (for example, changes in the number of control drives, shim loadings, or fuel management, etc.) but rather reflects our current belief that design changes necessitated by DUTH fuels will be sufficiently straightforward so as to be accommodated within the engineering design typically performed for new plants. ’As has been described in the discussion above, the first phase of such fuel-cycle- related research consists of the development of a data base from which safety-related parameters and fuel performance can be predicted in subsequent core physics design and safety analysis programs. First, existing thorium materials and fuel performance infor- mation should be thoroughly reviewed, and a preliminary evaluation of safety and licensing issues should be made in order to identify missing information and guide the subsequent development program. Although this initial phase is required to fully define the required data base R&D, it is possible to anticipate in advance the need to establish information in the areas of physics verification and safety-related fuel performance, As shown in Table 5.1-1, the physics verification program under data base develop- ment is estimated to cost ~$10 million. This program should be designed both to provide the information required to predict important safety-related physics parameters and to demonstrate the'accurdcy of such predictions as part of the safety analysis. Improved values must be obtained for cross sections of thorium and of isotopes in the thorium depletion chains, such as 233U and protactinium, all of which have been largely neglected in the past. Resonance integral measurements should also be performed for denatured fuels both at room temperature and at elevated temperatures, such experiments being very im- portant for accurately calculating safety-related physics characteristics and also for establishing the quantities of plutonium produced during irradiation. Finally, an LWR physics verification program should include a series of critical experiments, preferably both at room temperature and at elevated moderator temperatures, for each of the fuel types under consideration (i.e., for thorium-based fuels utilizing denatured 235U, denatured 233y, or plutonium). These experiments would serve as a basis for demonstrating the adequacy of the cross-section data sets and of the ability of analytical models to predict such safety-related parameters as reactivity, poWer'distributions, moderator temperature reactivity coefficients, boron worth, and control rod worth. i r { 5-9 ‘Table 5,1-1, Government Research and Development Required to Convert Light-Water Reactors to Denatured Uranium-Thorium Fuel Cycles (20% 235y/238y-Th or 20% 233y/238y-Th)- Assumptions: A1l basic reactor R&D required for commercialization of LWRs operating on their reference fuel cycle (LEU) has been complieted. Use of denatured fuel can be demonstrated in a current-generation LWR. Because utility sponsoring demonstration will be taking some risk of decreased reactor avilability, a 25% government subsidy is assumed for a 3-year demonstration program, Note: LWRs can be operated on the denatured 235)/238y-Th fuel cycle before any other reactor system; however, they cannot be economically competitive with LWRs operaping on the LEU once-through cycle because higher U30g requirements are associated with thorium fuel. Any commercial LWRs operating on a denatured cycle before the year 2000 must be subsidized. . : Cost Research and Development ($M) A. Data base development Al. Physics verification program 10 Improve cross sections for Th, 233U, Pa, etc. Measure resonance integrals for denatured uranium- , thorium fuels at room temperature and at elevated ‘ temperatures. | Perform and analyze critical experiments for each fuel. A2. Fuel-performance program (30 -~ 150)% Perform in-reactor properties experiments Perform power ramp experiments Perform fuel-rod irradiation experiments Perform transient tests B. Reactor components development (deve]bp handlihg | 5 -25 equipment/procedures for radioactive 232U-con- taining fresh fuel elements). ' C. Demonstration design and licensing ( 20 - 100 Cl. Develop core design changes as required for denatured fuels ' ' C2, Perform safety analysis of modified core C3. Prepare safety analysis report (SAR); carry _ through ticensing = | D. Demonstration of LWR operating on denatured fuel | , : 50b - 200 (probably 235U/238y-Th) “Would be included in fuel recycle RA&D costs (see Section 5.2). bpotential government subsidy; i.e., total cost of demonstration is $200M. 5-10 The fuel performance program under LWR data-base development would consist of the establishment of safety-related fuel performance information such as transient fuel damage limits, thermal performance both for normal operation and with respect to LOCA* margins on stored heat, dimensional stability (dénsification and swelling), gas absorption and release behavior, and fuel cladding interaction. The initial phase of this program should consist of in-reactor properties experiments, power ramp tests, transient fuel damage tests, and fuel rod irradiations. The in-reactor properties experiments would be similar to the program currently undefway in Norway's Halden HWR and would be designed to provide informa- tion on such parameters as center-line temperature, swelling and densification, and fission- gas release during operation. The power ramp experiments would consist of preirradiation of the fuel rod segments in existing LWRs and the subsequent power ramping of these segments in special test reactors to establish anticipated fuel performance during power changes typically encountered in the operation of LWRs. Examples of such programs are the inter- national inter-ramp and over-ramp programs currently being undertaken at Studsvik. The transient fuel damage experiments would be designed to provide information on the performance of the denatured fuels under the more rapid transients possible during operation and in postulated accidents. Lastly, the fuel rod irradiation experiments would provide informa- tion on the irradiation performance of prototypical thorjum-based fuel rods, and, with subsequent post-irradiation isotopic analyses, would also provide information on burnup and plutonium production. (As noted previously, the fuel performance program costs are included, though not specifically delineated, under the fuel cycle R,D&D discussed in Section 5.2.) In addition to the data base development, some as yet unidentified reactor components development could be expected. To cover this aspect of the program, an estimated cost of $5 - $25 milljon is included in Table 5.1-1. The remaining fuel-cycle-related R&D for LWRs would be devoted to developing core design changes and safety analysis information in preparation for a reactor/fuel cycle demonstration. In this phase of the program, safety-related behavior of alternate fuel would be determined using the specific design attributes of the demonstration reactor. The effects of alternate fuel cycles on plant safety and licensing would require examina- tion of safety criteria and the dynamic analyses of design basis events. Appropriate éafety'criteria, such as acceptable fuel design limits and limits on maximum energy deposi- tion in the fuel, would have to be determined. Changes in core physics parameters that result from alternate fuel loadings and the implication of these changes on reactor design and safety would also have to be identified and accommodated within the design. For example, changes in fuel and moderator témperature reactivity coefficients, boron worth, control-rod worth, prompt-neutron lifetime and delayed-neutron fraction must be addressed since they can have a large impact on the performance and saféty of the system. The ef- fects of alternate fuel cycles on the dynamic system responses should be determined for all transients required by Regulatory Guide 1.70, Revision 2. It would also be necessary to determine the imp]ications of denatured fuel cycles on plant operation and load change performance to determine whether the response of plant control and protection systems is *LOCA = Loss—of—Coo]ant Accident. o — (. - v r T ] el - b b el (:’_ £ r—i | N ] | A i 1 - B an N el o altered. A safety analysis Eéport for denatured thorium fuels would be prepared as part of this development task and pursued with licensing authorities through approval. The reactor development cost associated with commercializing the LWR on the DUTH fuel cycle is thought to be about $200 million. This relatively low cost results from the com- mercial status of the LWR and from the relatively small risk associated with deploying a new fuel type, since if the demonstration program is unsuccessful, the reactor can always be returned to uranium fueling. The estimated cost for the light-water reactor is based on an-assumed 25% government subsidy for a three-year in-reactor demonstration. The 25% subsidy is intended primarily to ensure the sponsoring utility against the potential for _decreased reactor avaiiability which might result from unsatisfactory performance of the DUTH fuel. (The cost of the fuel itself is included in the fuel recycle development costs d1scussed in Section 5.2.) 5.1.2. High-Temperature Gas-Cooled Reactors Although a number of alternate high-temperature gas-cooled reactor technologies have been or are being developed by various countries, this discussion considers the reactor con- cept developed by the General Atomic Company. U. S. experience with high-temperature gas- cooled reactors dates from March 3, 1966, when the 40-MWe Peach Bottom Atomic Power Station became operabTe. More recently, the 330-MWe Fort St. Vrain HTGR plant has been completed and is currently undergoing initial rise-to-power testing. Consequently, HTGR status in the U. S. is considered to be at the prototype stage and the basic reactor development still required is that associated with the demonstration of a large plant design. Al- though the success of the Fort St. Vrain prototype cannot be fully assessed until after several years of dperation, in this discussion satisfactory performance of the Fort St. Vrain plant has been assumed. | * Cost estimates for the R&D requirements for the development of a large commercial HTGR on its reference HEU/Th cycle are shown in Table 5.1-2. These estimates include only that R&D required relative to the Fort St. Vrain plant. As these tables indicate, the majority of the R&D expenditures would be directed toward component R&D and component - design, specifically for the development of the PCRV (prestressed concrete reactor vessel), steam generator, instrumentation and contro1,-matérja1s and methods, and the main helium circulators and service systems. In addition, an estimated $30 mi}lion to $60 million would be required for licensing and preparing a safety ana]ys1s report for the initial power reactor demonstratxon program The cost of_a power reactor demonstration plant fbr the HTGR on its reference cycle would be significantly higher than the cost given earlier for an LWR on a DUTH cycle, reflecting the increased cost and risk associated with deploying new concepts. In developing the potential reactor demonstration costs for the HTGR, we have assumed that a substantial government subsidy (50%) would be required for the first unit. Since it will be necessary to commit at least the second through fifth of a kind prior to the successful operation of this initial demonstration unit if the postulated deployment 5.1.2. Government Research and Development Required to Demonstrate HTGRS, HWRs, and SSCRs on Their Reference Cycles Assumptions 1. A1l reactors except LWRs still require basic reactor research and development for operation on their reference fuel cycles. 2. Llogical progression of basic reactor R&D (excluding fuel performance and recycle R&D} is: A. Proof of principle with small test reactor. _B. Design, construction, and operation of prototype reactor and/or component testing facility. C. Design, construction, and operation of demonstration plant. 3. Substantial government subsidies are required for rapid commercialization of reactors since unfavorable near-term economics and/or high-risk factors make early commitment on concepts by private sector unattractive. - m W = m oW W = = m om m m m om m o= ow om o om o= A oEm om oW Em OE s m oE oW W o moEm OE W om oW R oW W m om wm = omom o @ W e o ®mom o= meoeeom m o= o oa o= e e o= Heavy-Water Reactors.'b’c Spectral-Shift-Controlled Reactors b {Reference Fuel Cycle: SEY) {Reference Fuel Cycle:. LEY) High-Temperature Gas-Cooled Reactors {Reference Fuel Cycle: HEU/Th)4 Cost Cost ' Cost Research and Development Research and Development ($M) Research and Development {3M) A. Proof of principle accomplished' Proof of principle accomplished Proof of principle accomplished in in Peach Bottom Reactor by Canada -- BR3 reactor in Belgium -- B. Prototype reactor operation in Prototypes of natural-uranium " Prototype operation not believed to progress (Ft. St. Vrain plant) fueled reactors already operated be necessary - -- at <1000 MWe by Canada C. Larée plant design and licensing Largé plant design and licensing targe plant design and licensing C1. Component RAD Cl. Technology transfer and 120 C1. Component RED 30-€0 PCRY; steam generators; manufacturing license fee Develop D.0 upgrader technology; control and instrumentation; C2. Component R&D 60-150 perform thermal-hydraulic tests; materials; main helium cir- . valve, seal, and pump development culators and sérvice systems Core modifications; develop- to minimize leakage; develop ment and modification for . refueting techniques U,s. siting C2. Component design 3. Licensing and SAR development 30-100 €2. Licensing and SAR development 20-50 C3. Licensing and SAR development 30-60 D. Large plant demonstration Large plant demonstration Large plant demonstration (in modified PWR) 50% subsidy of first unit 50% subsidy of first unit 400 100% subsidy of extra equipment (plus other costs) for first unit 140 25% subsidy of next four units 25% subsidy of next four unfts 700 100% subsidy of extra equipment for next four units . 100 Apctimates based on those from Arthur D. Little, Inc, the Ft. $t, Vrain plant. study, "Gas Cooled Reactor Assessment," August, 1976, plus subsequent experience at bDemonstration plant may require reactivation of U.5. heavy-water facilities; commercialization of these reactors will necessitate development of D20 production industry, ®pcsumed to be CANDU-PHWR-based design deployed under Canadian license; this assumption, a U.S. prototype is not thought necessary, although it may still be desirable. in Canadian plants, while other design modifications such as demonstration program after completion of component R&D. R&D costs would be significantly higher for U.S.-originated design. Under The use of SEU/higher burnups can be demonstrated higher operating pressures can be demonstrated in the lead plant of the large plant ZL-9 | rorrocoo oot i 5-13 schedule is to be maintained, our costs presume further governmental support will be nec- essary {a 25% subsidy is assumed) for the second through fifth units. As noted in Table 5.1- 2, a 50% subsidy of the first unit is expected to be about $400 million, and a 25% subs1dy of the next four units is expected to total $700 million. Since the assumptions “underlylng government subsidies of the reactor demonstrat1on program shown in Table 5.1-2 have been defined, these costs can be adjusted to reflect either different levels of govern- ment support or a change in the overall cost of the demonstration program. As has been stated above, it has been assumed that the advanced converters such as the HTGR would all be successfully demonstreted on their reference cycles before they are converted to DUTH cycles. However, since the reference cycle for the HTGR is already a thorium-based cycle, it is likely that a denatured Cycle could be designated as the reference cycle for this reactor and thus that the lead plant demonstration program would be for a DUTH-fueled HTGR. If this were dbne, the additibha] costs required to convert the HTGR to a denatured fuel might be smaller than those associated with converting LWRs from their uranium-based fuel cycle to a thorium-based cycle. 5.1.3. Heavy-Water Reactors Although a number of alternate heavy-water reactor concepts have been developed by various nations, only the CANDU pressurized heavy-water reactor has been deployed in sig- nificant numbers. Therefbre,-as noted previouslv, the CANDU reactor is taken as the reference reactor for deployment in the United States. The R&D cost can vary cdnsiderab]y, depending on whether developed Canadian technology is utilized or whether the U,S..e1ects to independently develop a heavy-water-reactor concept. It is assumed here that the U.S. HWR will be based on the CANDU-PHWR and deployed under Canadian license and with Canadian cooperation. Thus, our costs address only those aspects reQuired to extend'the present CANDU design to that of a large plant (1,000-MWe) for U.S. siting. An order of magnitude higher R&D commitment would be required if it were necessary to reproduce the development and demonstrations which the Canad1ans have performed to date. Research and deve1opment requ1rements for the HWR are included 1n Table 5.1-2. In- herent in these requirements is the assumption that although the U.S. design would be based on the CANDU PHNR sign1f1cant changes would have to be made in order to realize a com- mercial offer1ng in the U.S. These modifications consist of the development of a large plant design {1,000- MWe), the use of s]1ght1y.enr1ched fuel both to improve resource utilization and to reduce power costs, modifications of the HWR design to reduce capital cost (the pract1ca11ty of which is genera]ly re]ated to the use of s]ight]y enriched fue]), and mod1f1cat1ons requ1red for U S. 11cens1ng The rather large range of pOtentia1 R&D costs shown in Table 5.1-2, particu]ar]y for 1icensing and SAR development, is indicative of the uncertainty introduced by lTicensing, i.e., to the degree to which the HWR will be forced to conform to licensing criteria developed for the LNR.. | | | | The first aspect of large plant design and‘11censing R&D, identified as component R&D, is related primarily to the extension of the CANDU to 1,000 MWe, the use of slightly enriched fuel, and possible increases in system pressure so as to reduce effective capital ~cost. In general, increasing the power output of the HWR to 1,000 MWe should be more readi- 1y accomplished than with other concepts such as the LWR, since it can be accomplished simply by adding additional fuel channels and an additional coolant Toop. The use of slightly enriched fuel and higher operating pressures should result in no fundamental changes to CANDU design, but nevertheless will necessitate some development in order to accommodate the higher interchannel peaking expected with slightly enriched fuels and the effect of higher system pressures on pressure-tube design and performance. Modifications for U.S. siting are somewhat difficult to quantify since a thorough licensing review of the HWR has yet to be completed. Althohgh there is no doubt of the fundamental safety'of the CANDU, modifications for'U.S. siting and licensing are nevertheless anticipated for such reasons at differing seismic c¢riteria (due to the differing geology between the U.S. and Canada) and because of differing licensing traditions. Additional experimental informa- tion on the performance of slightly enriched uranium fuel should also be developed by ir- radiating such fuel in existing HWRs (such as in Canada's NPD plant near Chalk River) to the discharge burnupsxhnticipated for the reference design (about 21,000 MWe/TeM). Methods of analyzing the response of the HWR to anticipated operational occurrences and other postulated accidents will have to be developed and approved by the Nuclear Regulatory Commission, and a safety analysis report in conformance with NRC criteria will have to be developed and defended. As is the case for the HTGR, the cost for a power demonstration plant for the HWR would be significantly higher than the cost for a DUTH-fueled LWR. The large plant demon- stration costs shown in Table 5.1-2 have been estimated under the same set of assumptions used for estimating the HTGR plant. The cost of a program to convert an HWR from its reference uranium cycle to denatured fuel would be approximately equal to that previously described for the LWR. 5.1.4. Spectral-Shift-Controlled Reactors As was noted in Chapter 4, the SSCR consists basically of a PWR whose reactivity control system utilizes heavy water instead of soluble boron to compensate for reactivity changes during the operating cycle. Since the SSCR proof-of- principle has already been demonstrated by the operation of the BR3 reactor in Belgium, and since various components required for heavy-water handling and reconcentration are well established by heavy-water reactor operating experience, the SSCR is considered to be at a stage where either a prototype or a large power plant demonstration is required. v For most alternative reactor concepts at this stage of development, a prototype program would be necessary because of the capital cost and high risk associated with I A S ] e i ! i r— ! | ! i r— T ~] ] e “t 1 .. - 5-15 bypassing the prqtbtype stage and constructing a large power reactor demonstration. Such a prototype program may also be desirable for the SSCR, particularly if the prototype pro- gram involved the modification of an existing PWR for spectral-shift control rather than the construction of a wholly new plant for this purpose. However, the estimates of the reactor R&D requirements given for the SSCR in Table 5.1-2 are based on the assumption that this prototype stage is bypassed. This can be justified on the basis that the SSCR is rather unique among the various alternatives because of its close re]aiionship to present PWR technology. In particular, no reactor develophent would be required and the reactor could be designed so that the plant would be operated in either the conventional poison control mode or in the spectral-shift control mode. As a result, a great majority of the capital investment in the plant and the power output of the plant itself is not at risk. Likewise, the potential for serious licensing delays is largely mitigated, since the reac- tor could initially be operated as a poison-controlled PHR and easily reconfigured for the spectral-shift control once the licensing approvals were obtained. 'Consequently, the capital at risk is limited to the additional expenditures required to realize spectral- shift control, roughly $30 - $60 million for component R&D, plus rental charges on the heavy water inventory. The additional expenditures for design and licensing, $20 - $50 million, would have also been necessary for the prototype. The component R&D would consist of a thermal-hydraulic development task; valves and seal development; development of D,0 upgrader'techno]ogy; and refueling methods development, design and testing. The thermal-hydraulic tests would be designed to produce a departure from nucleate boiling correlation for the SSCR moderator similar to that which has been developed for the PWR light-water moderator. The correlations are expected to be very similar, but tests to demonstrate this assumption for the various mixtures of heavy and light water will be required. Valves and seal development will be necessary in order to minimize leakage of the heavy-water mixture; reduction of coolant leakage is important both from an economic ~ standpoint (because of the cost of D,0) and because of the potential radiological hazard from tritium which is produced in the coolant. Methods of reducing coolant leakage from valves and seals have been extensively eXp]dred as part of the design'effort on heavy- water reactors and utilization of heavy-water reactor experience is assumed. The R&D program would address the application of the technologies developed for the heavy-water reactor to the'larger'size'components and higher pressures'encountered in the SSCR. The D,0 upgrader employed in the SSCR is identical in concept to the upgraders used on heavy-water reactors and in the last stage (finishing stage) of D,0 production facilities. The sizing of various components in the»upgrader'would, however, be somewhat different for SSCR application becausé of the range of D,0 concentration feeds (resulting from the changing D,0 concentration during a reactor operating cycle), and because of the large volume of low D,0 concentration coolant which must be upgraded toward the end of each operating cycle. The upgrader R&D program would consider the sizing of the upgrader, and should also address methods of minimizing the D,0 inventory in the upgrader so as to minimize DéO inventory charges. Lastly, component R&D should address methods for refueling and for coolant exchange during refueling. Refueling should be performed with pure light water present in the reac- tor (so as to avoid the radiological hazard of tritium); the 1ight water must subsequently be replaced with the light-water/heavy-water mixture prior to initiating the next operating cycle. In order to accomplish this refueling/coolant exchange without necessitating large volumes of heavy water for this pdrpose, a modified bleed-and-feed procedure is being ex- plored in which the differences in density between the warm water in the core and the cool makeup water is exploited in order to minimize coolant mixing and the amount of excess D,0 inventories required. Scale tests of this refueling procedure (or any other refueling/ coolant exchange procedure selected) will be required. The R&D related to safety and licensing should consist first of data development for the SSCR operating on the uranium fuel cycle. This data base has been partially developed in the initial SSCR development work performed by the USAEC in the 1960s. However, additional work, primarily in the area of physics verification of safety-related parameters (i.e., critical experiments which establish reactivity predictions, power distributions, D,0 worths, and con- trol rod wbrths) are required for uranium fuel. The second aspect of the safety and licens- ing R3D should consist of a preliminary system design, the performance of a safety analysis for the SSCR, and the development of a safety analysis report for spectral-shift-control operation. At this stage, component design and development would be limited to those areas in which some design changes would be required in order to ensure that the consequences of postulated accidents and anticipated operational occurrences with the SSCR would be comparable to those for the conventional PWR. ' The main areas thought to require attention are the implications of coefficients of reactivity on accidents that result in a cool-down of the primary coolant, the D,0 dilution accident, and tritium production. The implications of ‘the spectral-shift mode of control on plant operation and load change performance should also be addressed as part of the preliminary design evaluation. ~ With respect to the large plant demonstration of the SSCR, the financial risk to utilities would be Timited to the extra capital equipment required to realize spectral-shift control. Because the proposed schedule for commercialization is more rapid for the SSCR than for any of the other advanced converters, it has been assumed here that the government would essentially purchase the extra equipment réquired for the first five units (at $25 mil- lion per unit). In the case of the first unit, additional funding to mitigate the lower capacity factors anticipated for an experimental unit have been added. Also the cost for the first unit includes the carrying charges on the D,0 inventory. D,0 carrying charges are not included for the second through fifth units since it should be possible to demonstrate the spectral-shift control on the first unit before the D,0 for the remaining units needs to be purchased, so that a decision to employ spectral-shift control in sub- sequent units would be one which is purely commercial in nature. - 1 ] C ¥ r r,.—f» \ { wrmnmi oA | ] E ; b } \ r N ~ - | It is unlikely that an-SSCR would be converted to the denatured fuel cycle unless a similar change had previously occurred in the LWR. In this case, only a demonstration of the performance of denatured fuel in the spectral-shift mode of control would be needed. These incremental costs are estimated to be $10 - $60 million. 5.1.5. R,D&D Schedules Schedules for completing the R,D&D effort delineated above are summarized in Fig. 5.1-1. Although it can be argued that, givén strong governméntal support both in funding and in helping usher the various concepts through the licensing process, these schedules could be accelerated, the schedules shown are thought to be on the optimistic side of what can reasonably be expected to be achieved. In particular, a nine-year period has been as- sumed for the design, licensing and construction of a new reactor type; this would appear somewhat optimistic since it is currently taking longer to bring conventional LWRs on line. It should also be noted that in general the time sca]e required to develop alternate fuel cycle technologies (cf. Section 5.2) is estimated to be at least as long, and sometimes longer, than that required to develop reactor-related aspects. In general, this is because test facilities (for example, to perform demonstration irradiation) are available either in the U.S. or in Canada, so that R&D work prior to the design, licensing, and construction of a large demonstration plant could be rapidly initiated. 5.1.6. Summary and Conclusions It has been the purpose of this section to delineate the magnitude and scope of reac- tor R,D&D expenditures associated with the use of DUTH fuel in converter reactors and to determine if there are significant R,D&D cost differences between‘reactor systems. Recom- mendations for the further development of specific denatured reactors are provided in Section 7.5 where the R&D requirements discussed here are weighed against the potential benefits of various nuclear power syStems utilizing denatured fuels, as presented in Chapter 6. - In developing the nuclear power scenarios examined in Chapter 6, it was recognized that the bénefits of operating‘Lsz and alternate reactor types on DUTH fuels are dependent upon the speed and extent to which the systems can be dep1oyed S1nce the pr1mary goal of this interim report is to establish whether there is an 1ncentive for DUTH- fueled systems, a rather rapid deployment schedule was assumed so that the maximum benefits that could be anticipated from each reactor/fuel cycle system could be determined. Systems for which there is insufficient incentive for further deve]opment could thus be identified and eliminated from further consideration. Trade-offs between the prospects for commercialization, R&D costs, and deployment schedules and econom1c/resource incentives could then be evaluated in greater detail for the remaining options. 5-18 LWRs on Denatured Cycle® o ESTIMATED CALENDAR YEAR COSTS 1978 1980 1985 190 1995 2000 2005] ($M) DATA BASE DEVELOPMENT mem— 40 - 160° DEMO DESIGN AND LICENSING — 25 - 125 DEMONSTRATION R 50 - 200 Andicates minimum time from standpoint of reactor development; start time would be delayed for interfacing with fuel cycle development, Includes $30-150 million for fuel performance program (see Table 5, 1 2). ©¢50 million is potential government subsidy. HTGRs, HWRs, and SSCRs on Reference Cycles DEMO OPERATION ESTIMATED CALENDAR YEAR CusTS 1978 1980 1985 1990 1995 2000 2005 (8 HIGH-TEMPERATURE. GAS- (DOLED REACTORS (HEU/Th CYCLE) PROTOTYPE CONSTRUCTION ;1 PROTOTYPE IN AND OPERATION OPERATION DEMO DESIGN AND LICENSING} 160 - 250 DEMO CONSTRUCTION ) DEMO OPERATION h t 400_d HEAVY-HATER REACTORS (SEU CYCLE) PROTOTYPE CONSTRUCTION PROTOTYPE NOT AND OPERATION NECESSARY DEMO DESIGN AND LICENSING 210 - 370 DEMD CONSTRUCTION DEMO OPERATION 200%* SPECTRAL-SHIFT-QONTROLLED REACTORS (LEU CYCLE) DEMO ConSTRUCTION © . | bl s0-10 l S ————— T ——— 1409-€ dFlrst demonstration unit only. fExcludes cost of D,0 plant facilities. Incremental costs above PWR costs. Fig. 5.1-1. R&D Schedules and Costs for Government-Supported Demonstration of Various Reactor Systems = | " i bl T i o U 5-19 The most rapid dep1byment schedule considered tdube feasible was one in which time was allowed to resolve technical problems but one that was largely unimpeded by commercializa- tion considerations. The R,D&D schedules that have been presented in this section are consistent with this approach. However, it is recognized that the high-risk factors and potentially unfavorable near-term economics of such a schedule would make it unattractive to the private sector, espec1a]1y for those systems requiring large-plant demonstration. Demonstration program costs are viewed as highly uncertain and dependent upon the specific economic incentives for each reactor/cycle concept and on such factofs as the licensing climate and general health of the industry prevailing at the time of deployment. Thus the costs associated with the R,D&D schedules are assumed to be largely government financed. A comparison of the total estimated costs to the government for the various reactor systems discussed above is presented in Table 5.1-3. As noted, the R,D&D costs are lowest Table 5.1-3. Estimated Total Government Support Required for Demonstration of LWRs on DUTH Fuels and Advanced Converters on Various Fuels Total Costs System (4M) Comments LWR; DUTH Fuels 85 - 2152 In current-generation LWR; no demon- stration plant required. Advanced Converters; Reference Fuels HTGR; HEU/Th Fuel 560 ~ 750 If DUTH fuel selected as reference fuel, additional incremental cost probably Tess than cost of convert- ing LWRs to DUTH fuels. HWR; SEU Fuel 610 - 770b50 Additional incremental cost to con- vert to DUTH fuels approximately equal to that for LWR conversion, SSCR; LEU Fuel 190 - 250P:¢ Could be converted to DUTH fuel for $10M - $60M 1f LWRs already con- verted. : %Includes 25% subsidy for demonstration of LWR on DUTH fuel; excludes fuel bperformance program (see Table 5.1-2). Covers first demonstration unit only; 25% subsidy of four additional units anticipated (see Table 5.1-2). “Excludes costs of heavy-water plant fac111t1es. for the LWR on denatured fuel because of the already widespread deployment of this reactor concept. It is assumed that all basic R&D required for commercialization of LWRs operat- ing on their reference fuel cycle (LEU) has been'completed, and that the use of denatured fuel can be demonstrated in current-generation LWRs. Thus, an LWR demonstration plant, as such, will not be required.. The commitment of an LWR to DUTH fuels will entail some risks, however, and a 25% government subsidy is assumed to be necessary for a three-year demonstration program. 5-20 The R,D&D costs are highest for the HTGR and HWR, which are yet to be demonstrated on their reference cycles for the large unit size (1000-MWe) postulated in this report. The cost of these demonstration units constitutes the largest fraction of the total esti- mated R,D&D costs, although substantial costs will also be incurred for large plant design and licensing, which includes component R&D, component design, and ]icenéing and SAR deve]opMént. The R,D&D requirements for the HTGR and HWR are judged to be similar under the assumption that experience equiva]ént to that of the Fort St. Vrain HTGR prototype can be obtained from Cénadian technology. The SSCR is viewed as having R,D&D costs intermediate between those of the LWR and those of the HTGR because of the heavy reliance of the SSCR on LWR technology. As has been discussed in the text, once theée reactors have been demonstrated on their reference cycles, additional R,D&D will be required to convert them to DUTH fuels. ' Section 5.1'References 1. "The Economics and Utilization of Thorium in Nuclear Power Reactors.," Resource Planning Associates, Inc., January 16, 1968 (draft). N T . D o ke l[j . ™ — i =y ] 1 T r ¢ - - 5-21 5.2. FUEL RECYCLE RESEARCH AND DEVELOPMENT REQUIREMENTS I. Spiewak Oak Ridge National Laboratory The purpose of this section is to summarize the technica] problems that must be ad- dressed by a fuel recycle research and development program before reactor systems pfoducing and using denatured uranium-thorium (DUTH) fuels can be deployed commercially. Preliminary estimates of the schedule and costs for such a program are also inciuded to provide some perspective on the commitments that will be required with the introduction of reactors operating on denatured fuels. Wide ranges in the estimates reflect the current uncertain- ties in the program. However, detailed studies of the research and development requirements for the recycle of DUTH fuels are now being conducted by the DOE Nuclear Power Division's . Advanced Fuel Cycle Evaluation Program (AFCEP), and when the results from these studies be- come available, the uncertainties in costs and schedules should be reduced. ' 5.2.1. Technology Status Summary The technological areas in a fuel recycle program cover fuel fabrication/refabrication (fuel material preparation, rod fabrication, element assembly); fuel qualificétion (irradia- tion performance testing and evaluation}; fuel reprocessing (headend treatment, solvent extraction, product conversion, off-gas treatment); and waste treatment (concentration, cal- cination, vitrification, and radioactive-gas treatment). Fuel Fabrication/Refabrication and Qualification In general, the basic technology for the fabrication of uranium oxide pellet fuels is established, with the fabrication of both LWR and HWR uranium fuels being conducted on a commercial scale. In contrast, Pu/U oxide pellet fuels have been fabricated only on a small pilot-plant scale, and a significant amount of research and development is still required. Areas requiring further study include demonstration of: (1) a pelletizing process to ensure uniform product chafacteristics and‘performance; (2) methods for verifying and'chtrql1ing the characterisfics of the Pu/U fuels; (3) processes for the recovery of contaminated scrap; | (4) a reliable nondéstructive assay system for powders, fuel rods, and wastes; (5) the ability to operate a large-scale plant remotely, but with hands-on maintenance (in the case where Pu/U oxides containing high quality plutonium are being fabricated); and ' ' ’ : (6) satisfactory irradiation performance of Pu/U fuels produced in commercial-scale processes and equipment. 5-22 In the case of metal-clad oxide fuels that are thorium based, the areas requiring further study are essentially the same as those listed above for the Pu/U oxide fuels; how- ever, in contrast to Pu/U-oxide fuels, where significant effort has already been devoted toward resolving this list of areas, relatively ]ittlé R&D has been performed to date for thorium-based fuels and consequently a larger amount of research and development would be required. The intense radioactivity of the decay daughters of 232U (which is produced in the thorium along with the 233U) requires that the refabrication processes all be remotely operated and maintained. This requirement will necessitate additional development of the refabrication processes and may require the development of new fabrication methods. The qualification of U/Th and Pu/Th oxidé fuels will also require additional R&D efforts. HTGR fuels are coated uranium oxide or carbide microspheres embedded in a graphite fuel element. The process and equipment concepts for refabricating HTGR fuel remotely have been identified; however, additional R&D prior to construction of a hot demonstra- tion facility is needed. This should cover: ' (1) the scaleup of refabrication equipment, (2) the recycle of scrap material, (3} the control of effluents, and (4) the assay of fuel-containing materials. Additional R&D will also be required for qualification of the recycle fuel. While the reference HTGR fuel cycie already includes thorium, further development work will be required to fabricate DUTH fuels for HTGRs because of the requirement of a higher uranium content of the fissile particle and the increased production of plutonium during irradiation. Fuel Reprocessing The basic technology for reprocessing of uranium and uranium/plutonium oxide pellet fuels with low burnup exists in the Purex process. This technology is based on many years of government reprocessing experience with military-related fuels; however, a commercial reprocessing plant for mixed oxide power reactor fuels that conforms to current U.S. federal and state requirements has not yet”been operated. Additionally, while engineering or pilot-scale work has been successfully carried out on all important processes and components of the reprocessing plant, operability, reliability, and costs of an integrated plant have not been demonstrated in all cases at fuel exposures expected in commercial reactors. Specific areas that still require development work include the following: (1) operation and maintenance of the mechanical headend equipment; (2) 'methods for handling highly radioactive residues that remain after the dissolution of high-burnup fuel; (3) the technology for reducing radioactive off-gas releases (e.g., Kr-85, iodine and tritium) to conform to anticipated regulations; - r i el M 2 e e r— r— r— M S - r— rr EfffL T Ii” "l L r .- r 5-23 (4) remotely operated éhd:directly maintained conversion processes for plutonium from power reactor fuels; and \ (5) high-level waste solidification and vitrification to prepare for terminal storage. The technology for reprocessing thorium-based oxide pellet fuels is less advanced than that for uranium-based fuels. The Thorex process has been used to process irradiated thori- um oxide fuels of low burnup in government plants and in Timited quantities in a small-scale industrial plant. Thorium oxide fuels have not been processed in a large-scale plant specif- ically designed for thorium processing, nor has highly irradiated thorium oxide fuel been processed by the Thorex process in engineering-scale equipment. ' The principal differences between the reprocessing development required to reprocess metal-clad thorium-based oxide fuels and graphite-based HTGR fuel occur in the headend treatment. Partitioning of fuel materials from both classes of reactor fuel can then be accomplished by a Thorex-type solvent extraction process. In the case of metal-clad oxide fuels, additional headend process R&D is required to determine how zirconium cladding can be removed and the ThU, fuel dissolved. Significant waste handling problems may be encountered if fluoride is required to.disso]ve ThO,. In the case of the headend process development for graphite-based HTGR fuels, deve]op— ment work is needed with irradiated materials in the crushing, burning and particle separation operations, and in the treatment of 1%C-containing off-gases associated with the headend ~ of the reprocessing plant. Specific areas of solvent extraction process development work required to reprocess all thorium-containing reactor fuel include: (1) fuel dissolution, feed adjustment, and clarification; (2) technology development for containing 229Rn and other radioactive gases to conform to regulations; R o ' ' (3) recovery of fully irradiated thorium in large-scale facilities; (4) partitioning of fuel solutions containing U, Pu, and Th; (5) recovery and handling of highly radioactfve product streams; (6) " process and equipment design integration; and (7) high-level waste concentration and vitrification. Waste Treatment Waste treatment R&D requirements common to all fuel cycles involve development of the techho]ogy needed for immobilizing high-level and intermediate-level solid and gaseous wastes. Processes for concentration, calcination, and vitrification of these are needed. The waste treatment requirements for the various fuel cycles are similar, but they would be more complex for the thorium-based cycles if fluorides were present in the wastes. 5-24 5.2.2. Research, Development, and Demonstration Cost Ranges and Schedules While fuel recycle R&D needs can be identified for a variety of alternate fuel cycles and Systems, the launching of a major developmental effort to integrate these activities into a specific integrated fuel cycle must await a U.S. decision on the fuel cycle and reactor development strategy that would best support.our nonproliferation objec- tives and our energy needs. -whether it would be more expeditious to develop individual 'cycles independently in separate facilities or to plan for an integrated recycle develop- ment facility will depend on the nature and timing of that decision. If a number of related cycles were devéloped in the same facilities, the total costs would be only moderately higher than the costs associated with any one cycle. Since the denatured 233U cycle implies a system of symbiotic reactors (233U producers and 233U consumers), such an approach is likely to be attractive if a decision were made to develop the denatured 233U cycle. The existence of major uncertainties in.the fuel recycle development and demonstration - programs make cost projections highly uncertain. There are, first, difficulties inherent in projecting the costs of process and equipment development programs which address the resolu- tion of technical problems associated with particular reactors and fuel cycles. In addition, there are uncertainties common to projecting costs and scheduies for all fuel recycle develop- ment programs; specifically, uncertainties in the future size of the commercial nuclear in- dUstry cause problems in program definition. It is necessary to identify the reactor growth scenario associated with the fuel cycle system so that fuel loads can be projected and typical plant sizes estimated. This is critical from the standpoint of establishing the scale of the technology to be developed and the principal steps to be covered in the development. For example, if the end use of a fuel cycle is in a secure energy center, smaller plants are involved and the development could conceivably be terminated with a plant that would be considered a prototype in a large (1500 MT/yr) commercial reprocessing facility development sequence. Similarly, growth rates for particular reactor types may be much. smaller than others, or the fuel loads may be smaller because of higher fuel burnup. Thus, smaller fuel cycle plants would be required. The problem is further complicated by the fact that the fuel recycle industry has for a number of years been confronted with uncertain and escalating regulatory requirements. Permissible radiation exposure levels for operating personnel, acceptable safeguards - systems, and environmental and safety requirements, all of which affect costs, have not been specified. Nevertheless, based upon experience with previous fuel recycle develop- ment programs, typical fuel recycle R,D&D costs for the fuel cycles of interest can be pre- sented'in broad ranges. In the past, reprocessing costs had been developed for the U/Pu systems with partitioned and decontaminated product streams. These have been used here to provide base-line costs. Any institutional consideration, such as a secure fuel service center, that would permit conventional Purex and Thorex reprocessing to take place would give more credence to the base-line technology development costis used here. Lo ] — r— T r . _l{j' | r r—— 1 | T 5-25 Estimated cost ranges and times for the development and commercialization of a new reprocessing technology and a new refabrication technology are presented in Tables 5.1 and 5.2 respectively. From these tables, it can be seen that the total cost to the federal government to develop a new reprocessing technology would range between $0.8 billion and $2.0 billion. The corresponding cost for a new refabrication téchno]ogy would be / A Table 5.2.1. Estimated Cost Range for Development and - Commercialization of a Typicg] New keprocessing Technology Unescalated Billions of Dollars Base technology R&D 5.1 - 0.5 Hot pilot plant testing 0.5 - 1.0 Subtotal | 0.6 - 1.5 Large-scale ¢old prototype testingb_ 0.2 - 0.5 ' Total | 0.8 - 2.0 Large-scale demonstration plant® (1.0 - 3.0) ‘Uestimated lapsed time requirements from initial devel- opment through demonstration ranges from 12 years for established techno]ogy to 20 years for new tech- nology Government might incur costs of this magnitude as nart of demonstration program. ®Commercial facility - extent of government participa- tion difficult to define at this time. Table 5.2-2, Estimated Cost Range for Development and Demonstration of a Typi&al New Refabrication Technology Unescalated - Eillions of Dollars - Base technology ‘0,1 ~ 0.3 Cold component testinn 0.2 - 0.4 Irradiation performance testing - 0.1~ 0.4 Total 0.4 - 1.1 Large-sca]evdemonstrationb 7 o (O.Zj— ].4) %Estimated lapsed time requirements from initial development through demonstration ranges from about 8 - 10 years for technology near that established to about 15 years for new technology. Commercial facility - extent of government part1c1pation d1ff1cu1t to define at this time._ between $0.4 billion and $1.1 bil- lion. For fuel recycle deve]opment, the costs traditionally borne by “the government include basic R&D, construction and operation of pilot plants, development of large- scale prototype equipment, and sup- port for initial demonstration facilities. To these costs should be added the costs of the waste freatment technology development needed to close the fuel cycle. The capital costs estimated for a commercial demonstration facility are listed separately in Tables 5.1 and 5.2 because the extent that the government might support these facilities is un- known. Since they will be commercial facilities, costs ihcurred either by the government or by a private owner could be recovered in fees. The total ~ capital costs'might range between '$1.0 billion and $3.0 billion for a large reprocessing demonstration ~ facility and between $0.7 billion and ' $1.4 billion for a refabrication demonstration fac1]1ty ',Tables 5.1 and 5.2 show that “the ‘major costs associated with - commercialization of fuel cycles lie at the far end of the R&D progress1on, name]y, in the steps involving pilot plants, large-scale prototype equipment development, and demonstration p1ants, if required. The rate and sequencing of R&D expenditures can be inferred from Tables 5.2-1 and 5.2-2. Base technology R&D to identify process and equipment concepts may require 2-6 years. The engineering phase of the development ~ 5-26 program, including hot testing, may fequire 5-12 years. Reference facility design'ahd con- struction might require 8-12 years. There can be consfderab]e overlapping of phases so that for a given fuel cycle the total lapsed time from initial development to commercialization of fuel recycle ranges from about 12-20 years. The total time would depend,upon the initial technology status, the degree to which the-R&D'program-stepS‘are telescoped to save time, and the stage to which the development program must be carried.' The thgrium cycles would be at the far end of the development time range. ' - o " Table 5.2-3 presents the R&D cost ranges in terms of reactor types and fuel recycle - systems. For all fuel cycles, the uncertainiy'in'the R&D costs should be emphasized. Thus, in water reactors, the estimated range of R&D costs is $1.3-2.3 billion for U/Pu recycle development, and $1.8-3.3 billion for DUTH recycle development. For HTGRs, the correspond- ing ranges are $1.4-2.6 billion and $1.8-3.3 billion for U/Pu and DUTH recycle development, respectively; for FBRs, the corresponding ranges are $1.6-3.0 billion and $2.0-3.6 billion, respectively. Although there is a significant cost uncertainty for each reactor type and fuel cycle, for a given reactor type the trend in costs as a function of fuel cycle is significant. Generally, the reference U/Pu cycle would be least expensive and the DUTH " cycle the most expensive, with the Pu/Th and HEU/Th cycles intermediate. Table 5.2-3. Estimated Range of Fuel Recycle R&D Costs* Billions of Dollars Reactor Type U/Pu Pu/Th DUTH HEU/Th Water Reactors 1.3-2.3 1.6-3.0 1.8-3.3 1.6-2.9 HTGRs 1.4-2.6 1.6-3.0. 1.8-3.3 1.6-2.9 FBRs 1.6-3.0 1.8-3.2 2.0-3.6 1.7-3.1 *Includes costs for developing reprocessing and refabrication technologies and a portion of the waste treatment techno1ogy development costs. 5.2.3.‘ Conclusions A decision to develop reactor systems‘operating on denatured fuel cycles requires a government commitment to spend $0.5 billion to $2 billion more on a fuel recycle develop- ment program than would be required to develop reactors operat1ng on the reference_- (partitioned, uncontaminated products) U/Pu cyc]es. ‘The d1fferent1a] is even larger when reactors - dperating on DUTH cyc]es are cempared'with reactors operat1ng on once-through cycles. No comparison has been made with the costs of deve]op1ng d1vers1on res1stant U/Pu cyc]es (using co- process1ng, sp1k1ng, etc.). ' ' Expend1tures to deve?op recyc]e systems for DUTH fuels would span a per1od of 20 years from initial development to commerc1a11zat1on The pr1nc1pa1 expend1tures would occur in the second ha1f of this period, when 1arge fac111t1es with h1gh operat1ng costs are needed = d k A i e (.». - . 6.0. 6.1. 6.2. - 6.3. CHAPTER 6 EVALUATION OF NUCLEAR POWER SYSTEMS UTILIZING DENATURED FUEL M. R. Shay, D. R. Haffner, W. E. Black, T. M. Helm, W. G. Jolly, R. W. Hardie, and R. P. Omberg Hanford Engineering Development Laboratory Chapter Qutline Introduction Basic Assumptions and Analysis Technique 6.1.1. The U303 Supply 6.1.2. Reactor Options . 6.1.3. Nuclear Policy Options 6.1.4. The Analytical Method Discussion of Results for Selected Nuclear Policy Options 6.2.1. The Throwaway/Stowaway Option 6.2.2. Converter System with Plutonium Recycle 6.2.3. Converter System with Plutonium Throwaway 6.2.4. Converter System with Plutonium Production M1nim1zed Pu-t0-233y "Transmutation" 6.2.5. Converter System with Plutonium Production Not MInimized _ Pu-t0-233y “Transmutation"” 6.2.6. Converter-Breeder System with Light Plutonium "Transmutation" 6.2.7. Converter-Breeder System with Heavy Plutonium "Transmutation" Conclusions (- b 6-3 6.0. INTRODUCTION In this chapter civilian nuclear power systems that utilize denatured 233U fuel to various degrees are analyzed to determine whether they could meet projected nuclear power demands with the ore resources assumed to be available. The reactors employed in the systems are those discussed in earlier chapters of this report as being the reactors most likely to be developed sufficiently for commercial deployment within the planning horizon, which is assumed to extend to the year 2050. The reactors included are Light Water Reactors (LWRs), Spectral-Shift-Controlted Reactors (SSCRs), Heavy Water Reactors (HWRs), High:-Temperature Gas-Cooled Reactors (HTGRs), and Fast Breeder Reactors (FBRs). In each case, the nuclear power system is initiated with currently used LWRs operating on the low-enriched 235U fuel cycle, and other converter reactors and/or fuel cycies are added as they become available. On the basis of information provided by the reactor designers, it is assumed that 235U-fueled LWRs alone will be utilized through the 1980s and that LWRs operating on denatured 233U and 239%u will become available in the early 1990s. It is also assumed that SSCRs operating on the various fuel cycles will become available in the early 1990s. Thus nuclear power systems consisting of LWRs alone or of LWRs and SSCRs in combination, with several fuel cycle options being available, could be introduced in the early 1990s. LWR-HWR_and_LWR-HTGR systems could be expected in the mid 1990s, and FBRs could be added to any of the systems after the year 2000. The nuclear power systems utilizing denatured 233U fuel were *ivided into two major categories: those consisting of thermal converter reactors only aru those consisting of both thermal converters and fast breeders. Three "nuclear policy options" were examined under each category, the individual options differing primarily in the extent to which plutonium is produced and used to breed additional fissile material. For comparison, a throwaway/stowaway opt1on employing LEU converters was also analyzed, and two options utilizing the classical plutonium-uranium cycle were studied, one using converters only and the other using both converters and breeders. A11 of the options studied were based on the concept of secure energy centers and dispersed reactors discussed in previous chapters. Thus, all enrichment, reprocessing, and fuel fabrication/refabrication.activities, as well as fuel and/or waste stordge, were assumed to be confined to the energy centers. In‘addition, all reactors operating on'plutohium or highly enriched uranium were assigned to the centers, while reactors operating on low-enriched or denatured uranium were permitted to be outside the centers. Determining the precise nature and structure of the energy center ~was not within the scope of.this study Presumably it could be a relatively small Tocalized area or a large geograph1ca1 region cover1ng an entire nuclear state, or even a collection of nuclear states. If more than one country were involved, the sensitive fac1]1t1es could be nationally owned but operated under international safeguards. But whatever the character of the center an 1mportant cons1derat1on for any nuclear policy op- ~tion is its "energy support ratio," which is defined as the ratio of the nuclear capacity installed outside the center to the capac1ty installed inside the center, Only as the sup- port ratio increases above unity is the capability of the system to deliver power to dis- 6-4 persed areas ensured - a fact which is particularly important if nuclear states are planning to provide nuclear fuel assurances to nonnuclear states. The philosophy used in this study is illustrated in Fig. 6.0-1. Given a specified VN U305 supply and a specified set of reactor SPECIFIED U0y SUPPLY Ly development options, the potential role of nuclear power, the resources required to SPECIFIED REACTOR : : ) . : DEVELOPMENT OPTIONS éi;) éi;) éfi;) achieve this role, and the composition and movement of fissile material were calculated. The deployment of the individual reactors and e - their associated fuel cycle facilities were | in all cases consistent with the nuclear CALCILATE NUCLEAR GROWTH POTENTIAL INSTALLED CAPACITY " RESOURCE REQUIREMENTS AND r- policy option under consideration. The intro- FISSILE MATERIAL LOCATION —-— e 7 . : . s = : l duction date for each individual reactor con- cept and fuel cycle facility was assumed to be © HEDL 7802-98.1 ' Fig. 6.0-1. The Philosophy of the Nuclear Systems Assessment Study. This allows an evaluation of the maximum im- pact of the system on any particular nuclear option. The effect of delaying the deployment of a reactor/cycle because it produces undesirable consequences was determined'simply by e]iminating'it from the option. the earliest technologically feasible date. It was assumed that a nuclear power system was adequate if its installed nuclear capacity was 350 GWe in the year 2000 and a net increase of 15 GWe/yr was realized each year thereafter, with the increase sustained by the U30g supply. Two different‘optimizing patterns were used in the study. A few runs were made assuming economic competition between nuclear fuel and coal, the plants being selected to minimize the levelized cost of power over time. These runs, described in Appendix D, indicated that for the assumptions used in this analysis nuclear power did not compete well at Uj0g prices above $160/1b; therefore, in the remaining runs an attempt was made to satiéfy the demand for nuclear power with U;0g available for 1ess than $160/1b U30g. It is these runs that are described in this chapter. The specific assumptions regarding the Uj0g Supp]y'are presented in Section 6.1 below, which also includes descriptions of the operating characteristics of the individual reactors utilized, the various nuclear policy options chosen for ana]yses; and the analytical method “applied. Section 6.2 then comparaes the results obtained for a selected set of nuclear policy options, and Section 6.3 summarizes the conclusions reached on the basis of those comparisons, The economic data base used for these studies is given in Appendix B, and detailed resu]ts for all the nuclear policy options are presented in Appendix C. L r— r” ©r— ©—~ M \ i r “mining and milling rate was -less than 60,000 ST of U30g per year. 6-5 1.. BASIC ASSUMPTIONS AND ANALYSIS TECHNIQUE 6:1.1. The U30g Supply The most recent estimates of the supply of U0, available in the United States as re- ported by DOE's Division of Uranium Resources and Enrichment (URE)} are summarized in Table, 6.1-1 (from ref. 1). On the basis of a maximum forward cost of $50/1b, the known reserves plus probable potential resources total 2,325 x 103 ST. URE estimates that an additional 140 x 103 ST is available from byproducts (phosphates and copper), so that the amount of U305 probably available totals 2.465 x 103 ST (br approximately 2.5 million). If the "possible"” and "speculative" resources are also considered, the URE estimates are increased to approximately 4.5 million ST. Neither of these estimates include U;0g which may be available from other U.S. sources, such as the Tennessee shales, or from other nations.* The actual U30g supply curves used in the analysis were based on the long-run marginal costs of extracting U30g rather than the forward costs. The long-run marginal costs con- tain the capital costs of facilities currently in operation plus a normal profit for the industry; thus they are probably more appropriate for use in a nuclear strategy analysis. The actual long-run marginal costs used in this analysis are shown in Table B-7 of Appendix B and are plotted in Fig. 7.4-1 in Chapter 7. These sources show that if the recoverability of the U30g supply is such that large quantities can be extracted only at high costs, then the supply available at a cost of less than $160/1b is probab]y no more than 3 million ST. If, however, the recoverability is such that the extraction costs fall in what is cons1dered to be an intermediate-cost range, then as much as 6 million ST U30g could be available at a cost of less than $160/1b. In the remainder of this study, these two assumptions are referred to as "high-cost" and "intermediate-cost” Us0g supply assumptions. The rate at which the U30g resource is extracted is at least as important as the size of the resource base. URE has estimated that it would be difficult for the U.S. to mine and mill mofe than 60,000 ST of U30g .per year in the 1990's (ref. 3). (Note: This estimate was based on developing reserves and potent1a1 resources at forward costs of less than $30/1b These costs do not 1nc1ude cap1ta1 costs of faC111t1es or industry profits.) Although the combined maximum capab1]1ty of. a coalition of states may exceed this, it is not poss1b1e to- spec1fy a definite upper 11m1t until more is known about the Jocations of the sources of U30g and the d1ff1cu1t1es encountered in recover1ng it. Recogn1z1ng th1s, and also recognizing that the annuaT capacity .is still an 1mportant variable, the nuclear policy_opt1ons ana]yzed in this study were considered to be more feasible if their annual ‘ : , , : *Editor's Note: 1In 1977 the U.S. produced 15,000 ST of Us0g concentrate (ref. 2). TEditor's Note: In 1977 the U.S. gaseous diffusion plants produced 15.1 million kg SWU per year (ref. 4). After completion of the cascade improvement program (CIP} and cascade up-. dating program (CUP) in the 1980's, the U.S. capacity will be 27.4 million kg SWU per year (refs. 5 and 6). A gas centrifuge add-on of 8.8 miliion SWU has been proposed for the government-owned enrichment facility at Portsmouth, Ohio. Considerable enrichment capacity also exists abroad; therefore, enrichment capacity is inherently a less rigid constraint than uranium requirements or production capab111t1es 6-6 6.1.2. Reactor Options The reactor designs included in this study have not been aptimized to cover every con- ceivable nuclear policy option. Such a task is clearly impossible until the options have been reduced to a more manageable number. However, the designs selected have been developed by using detailed design procedures and they are more than adequate for a reactor strategy study such as is described here. | e Table 6.1-1. Estimates of Us0g Supply Available in U.S.A.2 Resources (103 ST) Forward Cost ($/1b) Known Probable Possible Speculative Total 15 360 560 485 165 _ 1,570 30 690 1,065 1,120 415 3,290 50° 875 1,450 1,470 570 4,365 % rom ref. 1. bAt $50/1b, the known reserves of 875 x 103 ST plus the probable reserves of 1,450 x 103 ST plus 140 x 103 ST from byproducts (phosphates and copper) total 2,465 x 103 ST (or o 2.5 million ST). If the possible and speculative resources are included, the total is increased to 4,505 x 103 ST (or ~ 4.5 million ST). Four general types of reactors are included: LWRs, represented by Pressurized Water Reactors (PWRs); HWRs, represented by Canadian Deuterijum Uranium Reactors (CANDUs); High Temperature Gas Cooled Reactors (HTGRs); and Fast Breeder Reactors (FBRs). The data for the PWRs were provided by Combustion Engineering (CE)} and Hanford Engineering Development Lab- oratory (HEDL); the data for the CANDUs by Argonne National Laboratory (ANL); the data for the HTGRs by General Atomic (GA); and the data for the FBRs by HEDL. In addition to the standard LWRs (PWRS); spectral-shift-controlled PWRs (SSCRs) are also included in the study, the data for the SSCRs being provided by CE. Descriptions of the individual reactors used in the study are given in Tables 6.1-2 and 6.1-3 (ref. 7), and the economic data base for each is given in Appendix B. ‘ The LWR designs include reactors fueled with low-enriched and denatured 235U, denatured 233y, and plutonium, the diluent for the denatured designs consisting of either 238 or ' thorium, or both. In addition, a low-enriched LWR design optimized for throwaway has been studied, and also three SSCRs fueled with low-enriched 235U, denatured 233, and Pu/Th. The HWRs are represented by three 235U-fueled reactors (natural, slightly enriched, and denatured), a denatured 233U reactor, a Pu/238U reactor, and a Pu/Th reactor. The HTGR designs consist of low-enriched, denatured, and highly enriched 235U reactors; denatured* ‘and highly enriched 233U reactors; and a Pu/Th reactor. The FBR designs consist of two Pu/238U core designs (one with a 238y blanket and one with a thorium blanket) and one Pu/Th core design (with a thorium blanket). In addition, a 233238 core design with a thorium blanket has been studied. The 233U enrichment is less than 12%, and thus this FBR is a denatured design. *In contrast to the other reactor types, the denatured 233y HTGR design is assumed to contain 15% 233U in 238y instead of 12%. r— \ (. r— - U~ . " A r— ] ,T~n r—. - e ifi 6-7 Introduction dates for each reactor type are included in Table 6.1-2. A slight modifica- tion to an existing PWR fuel design, such as a thicker fuel pin cladding to extend the dis- charge exposure, was introduced in 1981. A more extensive modification, such as a denatured 235 PWR fuel pin, was delayed until 1987, The remaining PWR designs, including the SSCRs, were introduced in 1991. The HWRs and HTGRs were all introduced in 1995 while the FBRs were not introduced until 2001. The 1lifetime-averaged 233U, 235U, and fissile plutonium flows given in Table 6,1-3 show that for the throwaway cycle, low-enriched HTGRs offer significant (atmost 20%) uranium ore savings compared to lTow-enriched PWRs, Slightly enriched HWRs reduce uranium ore require- ments by an additional 20% over HTGRs and more than 35% over LWRs. Although low-enriched LWRs and HTGRs have roughly the same enrichment requirements, the slightly enriched HWRs require 5 to 6 times less enrichment. The Tow-enriched SSCR offers about a 22% savings in enrichment. Core discharge exposures for FBRs are approximately twice the exposures for LWRs, while exposures for HWRs are about half those for LWRs, An exception is the natural- uranium HWR, which has a discharge exposure of one-fourth that for the LWR. HTGR dis- charge exposures are extremely large - nearly 200 MWd/kg for the Pu/Th fuel design. The two FBRs with Pu-U cores have breeding ratios of 1.34 to 1.36. Replacing the uranium in the core with thorium reduces the breeding ratio by 0.15, while replacing the plutonium with 233U reduces the breeding ratio by 0.16. Finally, comparing 235U-fueled thermal reactors with 233U-fueled reactors shows that the 233U-fueled reactors have con- version ratios about 0.10 to 0.15 higher. The most striking observation that can be made from the total fissile fuel requirements shown in Table 6,1-3 is the significantly lower fissile requirements for the denatured 233U- fueled SSCRs and HWRs and for the highly enriched 233U/Th-fueled HTGR. Finally, a few comments should be made about the relative uncertainties of the per- formance characteristics for the reactor deSignslifi this”study. Clearly, the low-enriched 235y-fyeled LWR (PWR) has low performance uncertainties. ‘Numerous PWRs that have been designed using these methods are currently in operation. The highly enriched 235U-fueled HTGR also would be expected to be quite accurate since Fort St Vrain started up in 1977. For the same reason, the successful operation of HWRs in Canada gives a high leveT of confidence in the natural uranium fueled CANDUs. The Pu-U-fueled FBRs have had a great dea] of critical experiment backup, and a few FBRs have been built in the U. S and abroad, giving assurance in the calculated performance parameters of these reactors., Most of the remaining reactors, however, have rather large uncertainties associated with their performance characteristics. This is because these reactors have not been built, and most have not even had critical experiments to verify the designs., The uncertainty for the alternate-fueled reactor designs is even greater since the effort in developing nuclear data for 233U and thorium has been modest compared to that expended in developing data for 235y, 238y, and plutonium, Tabte 6.1-2. Characteristics of Various Reactors Equilibrium Conditions _ Lifetime Requirements - Us0s TrvTchiment ‘ Heavy Metal _Core Breeding a Introduction E:::; (tons U303/GNe)b _ (108 kg SWU/GHe)® Rzafigigggigg 2;3323:3e ‘tonvg:sion Reactor/Cycle Date © {MWe) Charge Discharge Net . Charge Discharge Net (MT/GWe-yr) (MWD/kg) Ratio " LMR-US(LE)/U-S 1969 1150 5236 1157 4078 3.1 0.17 2.94 25.8 30 0.60 LWR-US(LE)/U-EE 1981 1150 4904 0 4904 . 0 3.1 18.2 43 0.54 LWR-US (DE)/U/Th 1987 1150 8841 3803 5038 . 8,03 3.20 4.83 24.1 3 0.66 LWR-U3{DE)/U/Th 199) 1150 0 0 0 0 0 0 24.1 32 0.80 LWR-Pu/U : 1991 1150 950 - 0 950 L0 0 0 25.7 30 0.70 LWR-Pu/Th 1991 1150 0 0 0 0 0 0 22.6 33 - SSCR-U5(LE)/U 1991 1300 4396 908 3489 2.42 0.05 2,37 25.3 30 - SSCR-U3(DE)/U/Th 1991 1300 0 0 0 0 0 0 23.0 33 - SSCR-Pu/Th . 1991 1300 0 0 0 0. 0 0 23,0 33 - HWR-US{NAT) /U 1995 1000 4156 0 4156 0 0 0 1149 7.5 - HWR-US(SEU)/U 1995 1000 3187 0 3187 0.59 0 - 0.59 53.9 16 - HWR-US5(DE )/U/Th 1995 1000 7337 2402 4935 6.66 1.94 4,73 53.9 16 - HWR-U3(DE )/U/Th 1995 1000 0 0o 0 0 0 0o - 53.9 16 - HWR-Pu/U 1995 1000 2030 0 2030 0 0 0 53,9 16 - HWR-Pu/Th 1995 1000 0 0 0 0 0 0- 53.9 16 - HTGR-US(LE)/U-T 1995 1344 4017 0 4017 3.23 0 3.23 8.2 80 0.50 "HTGR-U5(LE)/U 1995 1384 3017 - 431 3586 3,23 © 0.12 3.1 7.2 91 ' 0.50 HTGR-US(DE ) /U/Th 1995 1344 3875 465 3410 3.52 0.30 3.22 6.3 104 0.54 HTGR-US (HE)/U/Th 1995 1344 3903 © 558 3345 3.90 0.55 3.35 8.9 74 0.67 HTGR-U3(DE)/U/Th 1995 1344 0 - 0 0 0 0 0 10,4 63 0.65 HTGR-U3/Th | 1995 1344 0 0 0 0 0 0 14,0 47 0.86 HTGR-Pu/Th 1995 1344 0 0 0 0 0 0o 3.4 196 0,62 FBR-Pu-U/U 2001 1200 0 0 0 0 0 0 12.7/5.1/7.0° 62 1.36 FBR-Pu-U/Th 2001 1200 0 0 0 0 0 0 12.7/4.6/6.4 62 1.34 FBR-Pu-Th/Th , 2001 1200 0 0 0 0 0 0 11.6/4.6/6.4 68 1.19 FBR-U3-U/Th 2001 - 1200 0 0 0 -0 0 0 12.7/4.6/6.4 63 1.18 %LE = low enriched; DE = denatured; NAT = natural; SEU = slightly enriched; HE = highly enriched; U5 = 235U; U3 = 233y; S = standard LWR; EE = LWR with bextended discharge exposure; T = optimized for throwaway. ‘ With 1% fabrication and 1% reprocessing losses; enrichment tails assay 0.2%. “Core/Radial Blanket/Axial Blanket. c e — 8-9 Table 6.1-3. Average Fissile Mass Flows* for Various Reactors . 233y (kg/GWe-yr) 235 (kg/GWe-yr) Pu (kg/GWe-yr) Total (kg/GWe-yr) Reactor/Cycle Charge Discharge Net Charge Discharge Net Charge Discharge - Net Charge Discharge Net LWR-US(LE)/H-S . ‘ 0 -0 .0 736.9 213.4 523.5 0 146.8 -146.8 736.9 360.2 376.7 LWR-U5(LE)/U-EE 0 o . 0 683.3 0 683.3 0 0 0 683.3 0 683.3 LWR-US(DE}/U/Th 0 256.2 -256,2 1169.7 507.9 661.8 0 77.8 -77.8 1169.7 841.9 327.8 LWR-U3(DE}/U/Th . 807.0 530.4 276.6 13.5 16.8 -3,3 0 88.2 -88.2 820.5 635.4 - 185,1 LWR-Pu/U _ : 0 0 0 173.1 9%.2 82.0 700.6 472.2 228.5 873.7 563.4 310.5 LWR-Pu/Th 0 239.0 -239,0 0 2.3 -2.3 1294.1 620.2 673.9 1294,1 861,5 432.6 SSCR-US(LE)/U ' 0 ‘ 0. 0 626.6 169.3 457.3 0 185.0 -185.0 626.6 354.3 272.3 SSCR-U3(DE)/U/Th ©619,9 426,2 193.7 26.8 31.2 -4.4 0 72,9 -72.9 646.7 530.3 116.4 SSCR-Pu/Th 0 281.2 - -281.2 0 4,3 -4.3 1202.3 556.4 645.9 1202.3 841.9 360.4 HHR—US(NAT)/U | 0 0 0 757.4 227.8 529.6 0 290.4 -290.4 757.4 518.2 239.2 HWR-US(SEU)/U 0 0 0 . 521,8 72.2 449.7 0 159.8 -159.8 521.8 232.0 289.9 HNR-Ufi(DE)/U/Th 0 418.2 -418,2 970.8 322.8 648.0 0 22.5 -22.5 970.8 - 763.5 207.3 HWR-U3{DE }/U/Th 765.8 664.7 101.1 33.6 37.0 -3.4 0 26.9 -26.9 799.4 728.6 70.8 HWR=-Pu/U : o o 0 0 -369.9 67.2 302.7 156.6 177.7 -21.1 526.5 244.9 281.6 HWR-Pu/Th _ 0 391.9 - -391.9 0 2.8 -2.8 895.5 234.4 661.2 895.5 629.1 266.4 HTGR-US(LE)/U-T’ 0 0 0 540, 1 0 540.1 0 0 0 540,1 0 540.1 HTGR-US(LE)/U 0 ¢ 0 540.1 £9.1 471.0 0 43.1 -43,1 540.1 12,2 427.9 HTGR-US({DE)/U/Th 0 68.9 '-68.9 © 689.0 64.8 624.2 ] 27.3 -27.3 689.0 161.0 528.0 HTGR-US({HE)/Th 0 186.9 -186.9 512,3 73.3 439.0 0 1.0 -1.0 512.3 261.2 251.1 HTGR-U3(DE)/U/Th - 411.0 108.4 302.5 13.2 21,0 -7.7 0 27.9 -27.9 424.2 157.3 266.9 HTGR-U3/Th 501.5 389.0 112.5 73.8 69.9 3.9 0 0 0 575.3 458.9 116.4 HTGR-Pu/Th 0 94.1 -94.1 0 2.9 -2.9 637.0 126.7 510.3 637.0 223.7 413.3 FBR-Pu-U/U - \ 0 0 0 69.7 48,1 21.6 1253 1526 -273.3 1322.7 1574,1 -251.7 FBR-Pu-U/Th -0 237.5 -237.5 31.8 17.8 14,0 1261 1283 -21.9 1292.8 1538,3 -245,4 FBR-Pu-Th/Th 0 743.2 -743,2 0 0 0 1484 853.7 630.7 1484 1596.9 -112.9 FBR-U3-U/Th 1212.5 844.,5 368.0 33.3 19.4 13.9 0 499.8 -499,8 1245.8 1363.7 -117.9 6-9 *Lifetime average with 1% fabrication and 1% reprocessing losses, 6-10 6.1.3. Nuclear Policy Options Under the assumption that the reactor/fuel cycles listed in Tables 6.1-2 and 6.1-3 could be deployed, a set of nuclear policy options were developed for studying the relative capabilities of the various reactors to produce civilian nuclear power during the period from 1980 to 2050. As was pointed out above, it was assumed that for a system to be adequate, it should have an installed nuclear capacity of 350 GWe by the year 2000 and a net increase of 15 GWe thereafter, with each plant having a 30-yr lifetime. (Note: 1In order to determine the effect of a lTower growth rate, a few cases were also run for an installed capacity of 200 GWe in the year 2000 and 10 GWe/yr thereafter.) It was also assumed that reactors fueled with natural, low-enriched, slightly enriched, or denatured uranium could be dispersed outside the secure energy centers and those fueled with highly enriched uranium or with plutonium would be confined within the centers. A1l enrichment, reprocessing, and fabricating facilities would also be confined within the centers. The nuclear policy options fell under four major categories: (1) the throwaway/ stowaway option; (2) classical plutonium-uranium options; (3) denatured uranium options employing thermal converters only; and (3) denatured uranium options employing both converters and breeders. The various options under these categories are described in Table 6.1-4, and the specific reactors utilized in each option are indicated in Table 6.1-5. Schematic repre- sentations of the options are presented in Figs. 6.1-1 through 6.1-4. Runs were made for both intermediate-cost and high-cost 2308 supply assumptions. These nuclear options cannot be viewed as predictions of the future insofar as nuclear power is concerned; however, they can provide a logic framework by which the future implica- tion of current nuclear policy decisions can be understood. Suppose, for example, a group of nations agree to supply nuclear fuel to another group of nations providing the latter agree to forego reprocessing. A careful analysis of the nuclear system options outlined above can illustrate the logical consequences of such a decision upon the civilian nuclear power systems in both groups of nations. Only those nations providing the fuel would main- tain secure energy centers, since the nations receiving the fuel would be operating dispersed reactors only. (Note: The analysis presented here considers only the U.S. ore supply. A similar analysis for a group of nations would begin with different assumptions regarding the ore supply and nuclear energy demand.) For the purposes of this analysis, all the nuclear system options were assumed to be mutually exclusive. That is, it was assumed that any option selected would be pursued to jts ultimate end. In actuality, a nation would have the ability to change policies if con- sequences of the policy in effect were determined to be undesirable. However, the ability to successfully change a policy at a future date would be quite limited if the necessity of changing has not been identified and incorporated into the current program. The purpose of the study contained in this report was to identify the basic nuclear system options, and to determine the consequences of pursuing them to their ultimate end. (Note: A study of the consequences of changing policies at a future date - and thereby the implication of current programs - will be analyzed in a later study.) ™ + reeerilir 4 ¥ c.. €. - r- o 0 - 6.1.4. The Analytical Method The principal components of the analytical method used in this study are illustrated in Fig. 6.1-5 and are based on the following assumptions: (1) Given a specified demand for nuclear energy as a function of time, nuciear units are constructed to meet this demand consistent with the nuclear policy option under consideration. (2) As nuclear units requiring U;0g are constructed, the supply of U30g is continuously depleted. The depletion rate is based on both the first core load and the annual reloads required throughout the 1ife of the nuclear unit. The long-run marginal cost of U30g is assumed to be an increasing function of the cumulative amount mined. This is indicative of a continuous transition from highér grade to lower grade resources. (3) If the nuclear policy option under consideration assumes reprocessing, the fuel is stored after discharge until reprocessing is available. After reprocessing, the fissile plutonium and 233U are available for refabrication and reloading. (4) A nuclear unit which requires 23%Pu or 233U cannot be constructed unless the supply of fissile material is sufficient to provide the first core load plus the reloads on an annual basis throughout the unit's 1ife. (5) The number of nuclear units specified fof operation through the 1980's is exogenously consistent with the current construction plans of utilities. (6) A nuclear plant design which differs from established technology can be intro- duced only at a limited maximum rate. A typical maximum introduction rate is one plant during the first biennium, two plants during the second biennium, four during the third, eight during the fourth, etc. {7) 1If the mahufactufihg capability to prod0ce a particular reactor type is well established, the rate at which this reactor type will lose its share of the new construction market'is limited to a specified fraction per year. A typical maximum construction market loss rate is-]O%/yr This reflects the fact that some utilities will continue to purchase plants of an established and reliable techno?ogy, even though a new techno]ogy may offer an 1mprovement The acqu1s1t1on of fiss1le mater1a1 w111 be the principa] goa] of any nation embarked upon a nuclear weapons program. Therefore, any ana1ysis of a d1vers1on resistant civ111an nuc]ear power strategy must include a detailed analysis of the nuc]ear fuel cycle. The steps in the nuclear fuel cycle which were explicitly modeled in this analysis are shown in Fig. 6.1-6. They include: the mining of U30g; the conversion of U305 to UFg; the enrichment of the uranium by either the gaseous diffusion technique or the centrifuge r Table 6.1-4, Nuclear Po1icy_0ptionsa Throwaway/Stowaway Option (see Fig, 6.1-1) Option 1: LEU (235u7238y) converters’ operating on the throwaway/stowaway cycle are permitted outside the'enefgy centers and no reac- tors are operated inside the centers. Spent fuel is returned to the secure energy centers for ultimate disposal. . Plutonfum-Uranium Options (see Fig, 6.1.2) Option 2: LEU (2350/238y) converters are operated outside the secure energy centers and Pu/U converters and 235(HE)Th, 233u/Th, and Pu/Th HTGR's are permitted inside the centers, Uranium is recycled in all reactors, and plutonium is recycled in energy-center reactors, Option 3: LEU (235U/238y) converters are operated outside the secure energy centers and Pu/U converters, Pu-U)U breeders, and 235)(HE)/Th, 233Y/Th, and Pu/Th HTGRs are permitted inside the centers. .Uranium is recycled in all the reactors, and plutonium is re- cycled in the energy-center reactors, Denatured Uranium Options with Converters Only (see Fig. 6.1-3) ] , Option 4: LEU (235U/2380) converters and denatured 235U and 233U converters are operated outside the energy centers and no reactors are operated inside the centers. The fissile uranium is recycled into the converters, but the plutonium is stored inside the centers either for . . ultimate disposal or for future use at an unspecified date, * - : option 5U: LEU (2350/238) converters and denatured 2350 and 233U converters are operéted outside the energy centers and Pu/Th con- verters are permitted inside the centers. The fissile uranium is recycled into the outside reactors and the plutonium into the inside reac- tors. The goal in this case is to minimize the amount of plutonium produced and to "transmute" all that is produced into 233U in the energy- center reactors, : : ' ‘ . Option 5T: LEU (2350/238y) converters and denatured 233 converters are operated outside the energy centers and Pu/Th converters are permitted inside the centers. The fissile uranium is recycled into the outside reactors and the plutonium into the inside reactors, The goal in this case is not to minimize the amount of plutonium produced but "transmute” all that is produced to 233U in the energy-center reactors. . 21-9 Denatured Uranium Options with Converters and Breeders (see Fig, 6.1-4) Option 6: LEU (233U/238Y) converters and denatured 235U and 233U converters are operated outside the energy centers and Pu/Th con-' verters and Pu-t)/Th breeders (Pu-Y cores, Th blankets) are permitted inside the centers, - The fissile uranium is recycled into the outside reactors and the inside breeders and plutonium is recycled into the inside converters and breeders, With the reactors used, only a light "Pu-to-2330" transmutation rate is realized. : Option 7: LEU (235U/238y) converters, denatured 235U and 233U converters, and denatured 233U breeders are operated outside the energy centers and Pu/Th converters and Pu-U/Th breeders (Pu-U cores, Th blankets) are permitted inside the centers. The fissile uranium is re- cycled into the outside reactors and the inside breeders and plutonium is recycled in the inside converters and breeders. With the reactors used, only a light "Pu-to-233y" transmutation rate is realized. This case represents the first time a denatured breeder is introduced in the system, . ‘ . _ Option 8: LEU (235U/238y) converters, denatured 2350 and 233U converters, and denatured 233U breeders are operated outside the energy centers and Pu/Th converters and Pu-Th/Th breeders {Pu~Th cores, Th blankets) are permitted inside the centers, The f1§ggle uranjum is_ recycled into the outside reactors and the plutonium into the inside reactors. With the reactors used, & heavy "Pu-to-<3°U" transmutation rate is realized, Again a denatured breeder ie utilized in the system. %In all options except Option 1, spent fuel is returned to the secure energy centers for reprocessing. For Option 1, the spent fuel is returned to the center for ultimate disposal. - waRs that are fueled with natural or slightly enriched uranium are fncluded in this category. e e Table 6.1-5. Reactors Available in Secure {S) Centers or Dispersed (D) Areas for Various Nuclear Policy Options ‘78;g¥gx?y _Pu-U Options opt?fiflitflfiifl %g%clfigers Opf1onSDSQ%EUEgfivg:%glg?Breeders ) ‘ Option 1 Option 2 Option 3 QOption 4 Option SU Qption 5T Option 6 Qption 7 Option 8 Reactor/Cycle’ LSHGE LSHGE LSHGE LSHGE LSHE LSHGE LSHE LSHGE LSHE LWR-US (LE)/U-S DDODDD DDDD DDODOD DDDOD D DDOD DDDD DDDD DODDD DDODOD LWR-U5{LE )/U-EE D - - - “« 2 e e e e - - - - - - - - - - - - - - - - - - - - - - - - LWR-US(DE)/U/Th e e e e e e e DD - - DD - - - - - - D D - - P D - - DD - - LWR-U3{DE}/U/Th = - - - - - - - - - - - D - - - D - - - D - - - D - - - D - - - D - - LWR-Pu/U 2 e S S - - 58S - - - - - - - - - - . e . - - - - - - - - - - - LWR-Pu/Th - - - - - me e o o o - - - - S - - - S - - S - - - S - - - S - - SSCR-US{LE)/U - D - - =D -~ -D - - - - - - D - - - D - - - D - - - D - - - - - SSCR-U3(DE)/U/Th = = = = . = = = = = = = =« E 24 P S| - B @ 2 - 5 I o 2AF w z b Q 20 - ¥ Q e 9 ’ o 8V ] — g i8 JL 5 o 17T— [ : j bl O 16+ U308 PRICE OF 40 $/4 | H = L N YEAR OF STARTUP 5 REACTOR OPTIONS: LWR SSCR HWR HTGR FBR . " NN H ' I ) i I ( i t 1 I I 1 ) l | I I : I w - Lt - wi - w - - S w = w - < w = wi « w = w us S w - w -y ! . £,V xw Y ww &,V 2y axw 7 e Y ®y xw Z o U 2y, n:u..% wy t FUEL CYCLE 926 85 80 99 &,» O%C Og Q0 8, 030 Og Q9 &,w 036 80 Og 8.4 8% O & ¢ ! OPTIONS: nad> ni o _,:b quwy npe> -nC _CC oy & > ab _cc oLy e m(‘j _ct; MY 3 - 32 b HZx Rz g2 €2 REg 8Bs 8% g2 §%¢ ¥Ex fo g@ §%c HEg &2 <3 BEg 3% &% D1k D1 &€+ &1 DU Sid€ o £ S0 O/ 2 L& DOUE Sig D £ 208 2 &3 HEDL 7B05~090.53 - Fig. 6.1-7. Total Levelized Power Cost Sensitivity to Capital, Fabrication, and Reprocessing Cost Uncertainties. Since the valuation of the bred fissile material is related to the cumulative U304 price structure, the rate at which the U30g is consumed during a particular scenario also affects the time-dependent price calculated for the bred fissile material. Rapid consump- tion of the resource base (i.e., a high energy demand) yields a rapidly rising shadow r r— E ; price. Such an effect is readily noticeable in the calculation of the power costs.of b breeder reactors since it is possible for the credit calculated for the bred material to f = exceed the period’s charges for the reactor's inventory., Thus, the net fuel expense for " i; certain systems producing highly valued fissile material can be negative, resuiting in ’ significant power cost differences when compared to the reactor systems operating with ; f ; high-cost natural resources, This type of phenomenon is illustrated schematically by , &; Fig. 6.1-8 in which the power costs of a fast breeder and of an LEU-LWR are plotted as a | function of U30g price. The rising power cost of the LWR is directly attributable to the increasing fuel expense caused by the U;0g price. ‘The declining feSt reactor power cost reflects the increasing value of (and hence larger credit-fdr) the bred material when compared to U308 -derived fissile material, | - L "~ | The situation is still complicated even 1f one cons1ders on]y the conceptually 1 simple case of the throwaway cycle. From Fig, 6.1-9, where for SImp11c1ty the price of ! ‘U305 was assumed to be constant over the 1ife of the plant, it appears that the LWR is | the Teast expensive reactor when the U;0g price is less than $60/1b, and that the HWR r r will be less expensive than the LWR when the U303 pr1ce is greater than $160/1b. However, t; | an examination of the uncerta1nt1es leads one again to the conclusion that they dominate the problem, and that conclusions based on economic arguments are tenuous at best. Thus, P i‘i,} the decision was made to construct or not construct a nuclear unit on the basis of its LJ ability to extend the U30g supply rather than on its relative cost. 6-22 50 : ORNL-DWG. 78-14830 T T T 1 1 L 1 sl i o : '? 7 LWR-LEU ] Total Power Costs, mills/kwhr 20 - FBR 10 - 0 i l i | { | i o0 100 150 200 250 300 ] 350 . 400 U30g Price, $/1b Fig. 6.1-8. Influence of U304 Prices on Total Power Costs. 24 T — T T I T . ONCE-THROUGH OPTION & > np E _ ,v;; LWR-USQLE) 0 O o 20 '3 2 Q & HWR-USSE)A) o« 5 ——$SCR-US(LE 0 |gb ey _ o Al N d . } § : HTGR-USQLE)A i & T_\ 1 I L | L | ' 0 40 0 80 100 120 . 140 160 'U3°a PRICE, SAB ' HEDL 7805-090.39 Fig. 6.1-9. Total Power Cost of Various Reactor Systems as a Function of U304 Price {Constant U30g Price with Time; Once-Through Option). C - {- o ol rm——— o —— —d e Yt it o b ~ET ‘h;“"".\w-"\-. o e ) -y r t * - i l[?fl-n! €. r7 o o . " - r 6-23 6.2. DISCUSSION OF RESULTS FOR SELECTED NUCLEAR POLICY OPTIONS This section discusses results obtained in this study for a selected set of nuclear system options that typify the role of nuclear power under different nuclear policy deci- sions. The intent is to identify the basic issues, to determine the logical consequences of decisions made in accordance with those issues, and to display the consequences in an illustrative manner. Detailed results for a1l the nuclear system options outlined in Section 6.1 are presented in Appendix C, 6.2.1. The Throwaway/Stowaway Option The throwaway/stowaway cycle (see Fig. 6.1-1) is a conceptually simple nuclear system option and therefore has been selected as the reference cycle against which all other op- Avg. Capacity Factor = 0.67 Tails Composition = 0.0020 — : : U30g REQUIREMENTS (ST/GWe) 43 MuD/kg 30 Mub/kg = 80 MWD/kg = 16 MWD/kg g T Exposure = 30 MWbD/kg 1000 - Core Discharge LWR-S LWR-EE SSCR HTGR HWR-SEU HEDL 7805-090.41 Fig. 6.2-1. - Lifetime U305 Requirements for Various Reactors on the‘Throwaway‘Cycle.: tions are compared. In order to thorough- 1y understand the implications of the throw- away cycle, the effect of several deployment dptions utilizing the various advanced con- verters on the throwaway cycle was analyzed in detail. In general, the analysis assumed a nuclear growth rate of 350 GWe in the year 2000 followed by a net increase of 15 GWe/yr, but the consequences of a significant reduc- tion in the nuclear growth rate were also considered. In addition, the effect of both the high-cost and the intermediate-cost U30g supplies was determined. A summary of the 30-yr U;0g requirements for several reactors on the throwaway cycle, including an LWR with a fuel system designed for an extended.discharge exposure, is shown in Fig. 6.2.1., In each case, the average capacity factor of the reactor was _ S _ | ‘assumed to be 0.67, and the tai]s composi- tion of the enrichment plant was_assumed to be 0.0020. As the figure indicates, all the reactors haVe,]Qwer U304 requiréments than the standard LWR, the extendededischarge LWR being 6% lbwer,_the SSCR 16% lower, the HTGR 23%'10wer, and the SIightly enrfched HWR 39% ]ower. These U30g requirements were calculated for essentially standard designs without elaborate design optimization. ‘It is recognized that design optimization could improve the reactor , performance charatteristiqs; however, the goal of_this'analysis was hdt to delineate the _ ultimate role of any particular reactor concept based on current performance characteristics, but rather to identify the probable role of each reactor concept and the incentive for improving its performance characteristics. 6-24 The potential nuclear contribution with LWRs on the throwaway cycle, both with and without a fuel system designed for extended exposure'being included, is shown in Fig. 6.2-2 for the high-cost U30g supply. The nuclear contribution passes through a maximum of approximately 420 Gwe'insta1]ed'capacity in about 2010 ahd dec1ines'continuously thereafter, the system with the LWR-EE providing a slightly greater capacityaover most of the period.* The cumulative capacity constructed throughout the planning. hor1zon is approximately 600 GWe. The maximum installed capac1ty is less than the cumulative capac1ty because new units must be con- structed to replace those ret1red during the period. The maximum annual U;0g requirement is 72,000 ST/yr and the maximum annual enrichment requ1rement is 45 m1111on SWU/yr, neither of which can be regarded as excessive. Thus, the principal limitation, in this case, is simply the size of the economic Us0g supply. A more costly U;0g sUpb]y would, of course, imply a smaller maximum installed capacity occurring earlier in time, while the converse would be true for a cheaper U30g supply. As is shown in Fig. 6.2.3, if the U305 supply were a factor of two larger, the maximum nuclear contribution would ihcrease from approximately 420 GWe to approximately 730 GWe and would occur at about the year 2030. If, on the other hand, the supply were a factor of two smaller, the maximum huclear contribution would decrease to approximately 250 GWe and would occur in about the year 2000. A cross-plot of the effect of the -U30g supply on the maximum installed nuclear capacity for the LWR on the throwaway cycle is shown in Fig. 6.2-4. It is noted in Fig. 6.2-3 that if the U;05 supply should be as large as 6.0 ' ~ million ST, the maximum annual U30g requirement would be 120,000 ST/yr and the maximum | annual enrichment requirement would be 77 million SWU/yr. Given the probable limitation on the amount of U304 that could be mined and milled annually, these annual U30g requirements could be the Timiting factor. The effect of adding an advanced converter {SSCR, HTGR, or HWR) to a nuclear power system operating on the throwaway cycle with the high-cost U304 supply is shown in Fig. 6.2-5. The increase in the nuclear contribution for each of the advanced converter options is relatively small, At most the maximum installed nuclear capacity increases by approximately 30 GWe and the year in which the maximum occurs by approx1mate]y three years., Adding the SSCR to an LWR produces a sTightly greater nuclear contribution than adding an HTGR. This may at first appear to be a paradox since the lifetime U305 require- ment for the HTGR is less than that for the SSCR (see Fig. 6-2.1), but the 4-yr difference in introduction dates is sufficient to offset the difference in U30g requirements. (The dif- ference is not large enough to be significant, howéver.) The reason that so small an increase in nuclear'capacity is realized by intrOducing thé'varidus converters is.that by the time they dominate the‘nuclear system a very significant fraCtiQn of the U30g supply has already been committed to the standard LWR. This is illustrated in Fig. 6.2-6, where an HWR intro- duced in 1995 does not become dominant until 2010. It follows that if the U304 supply were larger with the same nuclear growth rate, or if the nuclear growth rate were smaller with the same U303'supp1y, the addition of an advanced converter would have a greater impact. This is illustrated in Fig. 6.2-7, for which the intermediate-cost U303 supply was assumed, and *Note: In general, unless a system cons1st1ng of the standard LWR alone is des1gnated, it 1s the LWR system including an LWR-EE that is denoted as 1L and compared with other systems in later sections of this chapter. However, as pointed out here, the installed capacities of the two LWR systems differ only slightly. r Ay - - ¥ r—-r- iy i N t * 1 o g -0} -) e B V40y COMMTMENT, MILLION TONS : 6-25 1000 T T T T T T 1000 T T T T Y T IHE LWK ON THE THROWAWAY CYCLE THE LWR ON THE THROWAWAY CYCLE s O b . 800 [~ T £ £ | i i MAXIMUM ANNUAL REQUIREMENTS ; E MAXIMUM ANNUAL REQUIREMENTS: E Us0y - 120 - 10 ST/ £ sl UP, - 72 - W i ‘ 4 ¥ o0 | ENHCHMENT - 77 - 10° swu/pr ., 3 ENRICHMENT - 45 + 10° swyuAm g % % INTERMEDIATE - COST U0g SUPPLY g 16.0 - 108 ST BELOW 5160 L8+ s &o - 3 40 b a IL-EE « System Including a g ' léun with Extended 5 HIGH COST U30q SUPPLY Xposure = IL-5 - System Utilizing X 13.0 - 105 51 BE LOW §160 LBy z Standard MR Only g = 200 - = 200 15 - 100 ST U3Ug BELOW $160 LB 0 1 1 1 1 ! i 0 i 1 1 1980 1990 2000 010 020 2030 2040 2030 1980 1990 2000 2010 2020 2030 2040 2050 YEAR Fig. 6.2-2. The Nuclear Contribution of LWRs on the Throwaway Cycle (High- Cost Us0g Supply). - 1000 T Y T 1 T T : THE LWR ON THE THROWAWAY CYCLE 5 O 800} g & S ~ 5 g s00 |- . 2 5 b=l g g g wl- - i e f or £ MAXIMUM 5 ¥ 200 - 0 1 1 1 1 1 1 0 1 2 1 4 5 6 7 U0, SUPRY, 10° TONS Fig. 6.2-4. The Effect of U305 Supply on the Maximum LWR Installed Nuclear _ Capacity. 5 T T T T ™ 7T . THE THROWAWAY CYCLE - ' SR -w s N - LWR FOLLOWED - Y HWR YEAR Fig. 6.2-6. The'U308 Commi tment = versus Time for an LWR-HWR System on the Throwaway Cycle (High-Cost U;0g Supply). aa e 2 15 « LWR FOLLOWED BY S5CK ADVANCED CONVERTER INTRODUCTION é‘w ™ DATES ' $sCr - 199 é : MGk - 1o 1G - LWR FOLLOWED 8Y HTGR B0} a L~ STANDARD L -2 2 Fig. 6.2-3. The Effect of U305 Supply on the Nuclear Contribution of LWRs on the Throwaway Cycle. 1000 I ¥ - 1 T 1 I THE LWR FOLLOWED Y AN ADVANCED CONVERTER ON THE THROWAWAY CYCLE : 800 |- - o 2_.: ADVANCED CONVERTER INTRODUCTION DATES v . . < wof ssck - 1991 - 3 HTGR - 1993 = HWR = 1995 g IH = LWR FOLLOWED §Y 2 wof / HWR i & 15 - LWR FOLLOWED 3 §Y SSCR X 2 200 - 1G = LWR FOLLOWED - BY HIGE WL = STANDARD LWR ‘ 0 1 1 1 1 1 1 1980 1990 2000 2010 2120 2000 2040 2050 YEAR Fig. 6.2-5. The Effect on the Nuclear Contribution of Adding Advanced Converters on the Throwaway Cycle (High-Cost U30g Supply). 1000 == T T T T T T THE LWk FOLLOWED §Y AN AD‘VANCED CONVERTER - . ON THE THROWAWAY CYCLE 1K ~ LWR FOLLOWED BY HWR 0 - 1980 1990 2000 000 2020 2000 2040 250 YEAR r'Fig.-6.2-7. The Effect on the Nuclear Contribution of Adding Advanced Converters on the Throwaway Cycle (Intermediate-Cost . U305 Supply). 6-26 1000 T T T T T T I 1000 T T T T T THE LWR FOLLOWED BY AN ADVANCED CONVERTER THE LWR ON THE THROWAWAY CYCLE ON THE THROWAWAY CYCLE J . 2% ;'m B ILTM - LWR WITH IMPROVING | o INSTALLED NUCLEAR CAPACITY - o MAXIMUM ANNUAL REQLIREMENTS: ENRICHMENT TAILS E mgxlfiir 2000 PLUS 10 GWo/yr E‘ u30, - ST C?MPOS"IW g o = 1H-LWR FOLLOWED BY HWR | § &0 - ENHICHMENT - n_ - swf ii‘%fi'! e 1 3 s 3 et g’ 400 § o - ~+-0.0005) . z 2 " 7 ENRICHMENT TAILS = 0 =00 COMPOSITION {0-0020) - 0 1 1 1 ! ! 1 0 1 1 1 1 1 1 1980 19%0 2000 200 2020 200 2040 2050 1980 1990 2000 2010 2020 2030 2040 2050 YEAR YEAR Fig. 6.2-8. The Effect on the Nuclear - Fig. 6.2-9. The Effect of Enrichment Contribution of Adding Advanced Converters Tails Composition on the Nuclear Contri- on the Throwaway Cycle (200 GWe in 2000 bution with the LWR on the Throwaway Cyc]e plus 10 GWe/yr Thereafter) (High~Cost U304 (High-Cost U305 Supply). Supply) 0 0025 T 1 T T T T 5 T T T t T T .- THE LWR ON THE THROWAWAY CYCLE o 0020 REFERENCE ENRICHMENT TAILS COMPOSIFION (0.0020) < ' é ©.0015 - - é i ] g : % 0. 00101 IMPROVED ENRICHMENT TARS = é 2 - COMPOSITION (0,0020-~0,0005) 2 = g 0.0006 — = 1+ ?970 1;00 1;90 2(:@ 20|ID 2".:20 2(':30 2048 !1,9?0 1980 1990 2000 20 2020 2330 2040 YEAR 1 YEAR Fig. 6.2-10. The Enrichment Tails Fig. 6.2-11. The Amounts of U304 Composition as a Function of Time for the Processed Through the Enrichment Plants Reference Case and for an Improving Tails as a Function of Time for the LWR on the Strategy. Throwaway Cycle (High-Cost U30g Supply). in Fig. 6.2-8, for which a reduced growth rate was assumed. With the intermediate-cost supply, the effect of the 4-yr difference in introduction dates between the SSCR and the HTGR is no longer significant, and the HTGR makes the greater contribution. The effect of changing the enrichment tails composition upon the nuclear contribution with the LWR on the throwaway cycle is shown in Fig. 6.2-9 in which the reference case with a constant enrichment tails composition of 0.0020 is compared with two other cases: one in which the enrichment tails composition decreases linearly from 0.0020 in 1980 to 0.0005 in 2010 and remains constant thereafter; and another in which the tails composition similarly decreases and in addition the tails stockpile accumulated prior to 2010 is mined at a later date with a tails composition of 0.0005. The decreasing enrichment tails composition, shown in Fig. 6.2-10, is the industry average, and hence the improving tails strategy implies Tow- ering the tails composition of the gasequs diffusion plants beginning in 1980. In addition, the'strategy implies a continual fransition toward an industry based upon an enrichment process capable of operating at an average tails composition of 0,0005. L z & — M ‘JJ e o 3 : A \ - a - 6-27 The effect of applying the improving tails strategy to a nuclear system based on the throwaway cycle is to increase the maximum installed nuclear capacity by approximately 60 GWe and to delay the maximum by approximately five years (see Fig. 6.2-9). Mining the tails stockpile accumulated prior to 2010 does not significantly change the result. The reason that mining the past tails stockpile does not produce a significantly larger nuclear contri- bution is explained by Fig. £.2-11, which shows the cumulative amount of U30g processed through the enrichment plants as a function of time. The amount is considerably less than the amount of U 05 comnmitted at any given time, as shown in Fig. 6.2-6. It is important to note that the amount of U305 actually processed through the enrichment plants prior to 1990 is relatively small, and at this time the tails composition for the 1mproving'tails strategy has been decreasing linearly for 10 yr. Thus, most of the U30g in the improving tails case is processed at lower tails compositions, and mining the past stockpile does not produce a significant improvement. The most dramatic effect associated with the improving tails option is the increase in the maximum annual enrichment requirement. As indicated in Fig. 6.2-9, the maximum annual U;0g requirement for this option is 67,000 ST/yr, while the maximum annual enrichment requirement is 92 million SWU/yr. Thus, the principal limitation in this case would be the availability of enrichment capacity. The utilization and movement of fissile material per GWe of installed capacity in the year 2035 for each of the converter options is shown in Fig. 6.2-12a-d, assuming the high-cost U305 supply. These figures represent a snapshot of the system in time and include the first core loadings for units Starting up in the year 2036. As can be seen, the U30g con- sumption for Case 1L in the year 2035 is approximately 142 ST U30g/GWe, with the LWRs having an extended discharge exposure comprising 92% of the installed capacity. When the LWRs are followed by SSCRs (Case 1S), the annual U30g consumption is 135 ST U;0g, with the SSCR com- prising 74% of the installed capacity. The fractional installed capacity of the SSCR is less than that of the extended-exposure LWR in Case 1L because the extended-exposure LWR is intro- duced'in 1981 while the SSCR is not introduced until 1991. In general, the fractional installed capacity of a reactor concept in the year 2035 will decrease monotonically as the intro- duction date for the concept increases. Similarly, the fractional installed nuclear capacity of a reactor concept will increase mohotonical]y as its U305 requirement decreases. When the LWRs are followed by HTGRs (Gase 1G), the U304 consumption in the year 2035 is 133 ST U404/GWe, with the HTGR comprising 54% of the installed capacity. The annual U304 consumpt1on is lower than in Case 1S because . the U303 requ1rement of the HTGR is less than that of the SSCR (see Table 6.1-2 and Fig. 6.2~ 1). The: fract1ona1 installed capacity of the HTGR is less than that of the SSCR in the Case 1S because the SSCR is introduced in 1991 while the HTGR is not introduced unt11 1995 then HWRs folldw the Lsz (Case'lH), U303 consumption in year 2035 is approximately 106 ST Uj0g/GWe and the HWR comprises 79% of.thé,installed,capacity; -The HWR in this case and the HTGR in Case 1G have'the‘same introduction date, The HWR, however, has a lower U30g requirement and hence the total installed nuclear capacity is greater with this 6-28 HEDL 7805-090.34b 935 LWR/U ’ : 3 L JUSQE) 2,907 Kg HM 51 Kgu : v 9 i CF = 60.3 7.2 x 10° SWU | S _ o THROWAWAY — 141.6 5T -—[m%n: 80.1 x 10° SWU | - LWR 235 5QE)AE 28,730 Kg HM 543 Kg U . - =] 7,020 K:HM*-EO-W GWe ’ CF = 60.3 _ (a) Case 1L: LWRs Only; High-Cost U305 Supply. 25 [ LR HEDL 7805~090, 342 165 Kg U ~—US(LE) /U 9,327 Kg HM 5,650 kg bn o [T]0.26 GHe’ - CF = 60.3 23.1 x 107 sW | i THROWAWAY . 135,01 ST uaoaq_rfiTc_m‘ 53.7 x 10° SWU SSCR | 235 us{LE)/u 23,750 Kg HM 810 kg “M-..fio 74_GW | 15,870 Kg H .74 Gie | CF = 60.3 (b) Case 1S: LWRs Followed by SSCRs; High-Cost U30g Supply. Fig. 6.2-12. Utilization and Movement of Fissile Material in Nuclear Systems Consisting of Converters Operating on Throwaway/Stowaway Cycle (year 2035). (Note: Except for Case 1L, which utilizes the extended exposure LWR all LWRs included here and in subsequent systems are standard LWRs.) ecorralg. - —— e e s - et r — 1T [! - r ,-fi! — ) r r - + - r | ! * N - ¥ ™ 6-29 Fig. 6.2-12 (cont.) 235 LWR O\ _ 291 Kg U ~—1 US(LE)YU 16,450 Kg HM ~ 9,967 Kg HM o[ ]0.46 GWe | CF=60.3 40.9 X 10° Swu 8 A — 133.3 5T U,0 e {fomen] THROWAWAY 51.2 X 16° swu 5,146 Kg HM HTG 260 Kg UZS L_3,495 Kg HM . : 7 HEDL 7805-090.31 (c) Case 1G: LWRs Followed by HTGRs; High-Cost U30g4 Supply. F 4,545 Kg HMetel___ | CF=60.3 18,6 x 10° swu ) o THROWAWAY «105.6 ST Us0g 13,9 x 10° swu l365 Kg 4235 . 43,810 Kg HM o 356,100 Kg HM o : Lo CF=60.6 - : 'HEDL 7805-090.58 (d) Case TH: LWRs Foflowe_d by ‘HWRs; 'High—Cost U305 Supply. 6-30 reactor, Since this increase is due simply to the construction of additional HWRs, the fractional installed capacity of the HWR is incréased commensurately. In summary, using the assumptions contained in this study, the following conclusions can be drawn about the behavior of a nuclear power system operating on the throwaway option: (1) The effect of deploying an advanced converter in 1995, under the assumption of 350 GWe 1n the year 2000 and 15 GWe/yr thereafter with the h1gh cost U305 supply, would be small. (2) If the U305 supply available below $160/1b should be larger than 3 million ST, or if the nuclear growth should be smaller than assumed above, then the effect of deploying the advanced converter would be larger. (3) The effect of reducing the enrichment tails composition is somewhat larger than -that of deploying an advanced converter under the assumeq conditions. (4) The dominant variable for the nuclear powef system on the throwaway cycle is the U305 supply; a Us0g supply either twice as large or twice as small is of greater consequence than any of the effects discussed above. 6.2.2. Converter System with Plutonium Recycle In order to assess the option of plutonium reéyc1e in converters it was assumed that a reprocessing capability would be available in 1991. (This assumption does not argue that the reprocessing capacity would be economically attractive or diversion-resistant, but merely that it would be technologically feasible by this date.) In this option the classi- cal plutonium recycle was modified somewhat by rejecting converters with self-generated recycle in favor of converters with complete plutonium 1oads This has the advantage of reducing the number of reactors that must be placed in the energy centers and commensurate- ly increases the number of reactors that can be placed outside the centers. The individual reactor concepts and their locations are shown in Fig. 6.1-2 (Option 2). A comparison of the nuclear contribution of the LWR with plutonium recycle to that of the LWR on the throwaway cycle (Fig. 6.2-13) shows that with recycle the maximum in- stalled nuclear capacity is increased from approx1mate1y 420 GWe to approximately 600 GWe and the time at which the maximum occurs is increased from about year 2010 to about year 2020 (high-cost U30g supply). The maximum annual U30g requirement for this case is 67,000 ST/yr and the maximum annua1 enrichment requirement is 46 million SWU/yr. These [ ey et g 3 C R 2y r r! il ‘ § ) ¢ r—i - e 1 ! B -0 L 6-31 1000 T T T T T T . 1900 T T T T T T THE LWR WiTH PLUTONIUM RECYCLE CASE 2L ~ THE EWR WITH PLUTONIUM RECYCLE 200 - - a0 |- - 3 5 t_ MAXIMUM ANMNUAL RECUIREMENTS : N - :‘E”(_Ryg"” PLUTONIUM ‘.:- E . . LE = g | 0,0, - &7 * 16° ST/ g 3 600 ENRICHMENT - 46 * 10° SWU/Arr 7 3 - - 3 3 g Y 2 40 | - Z 400 8 g - -d 7 3 2 0 IL - LWR ON THE THROWAWAY 2 - CYCLE - 200 0 t 1 i 1 ] 1 9 1980 1990 2000 2010 2020 2030 2040 2050 1 YEAR Fig. 6.2-13. The Effect on the Nuclear Fig. 6.2-14. Relative Nuclear Contri- Contribution of Recycling Plutonium in LWRs butions of LWRs Located Inside (LWR-Pu) and (High-Cost U305 Supply). Outside (LWR-U) Energy Centers (High-Cost U30g Supply). T T 7 . T ) requirements do not differ significantly from HE LR WITH PLUTONiUM RecvcLt those of the LWR on the throwaway cycle (see Fig. 6.2-2) because the nuclear growth pro- o &y seuow St e | Jection was specified to be 350 GWe in the year 2000 p]us 15 GWe/yr thereafter. Thus, the primary effect of reprocessing is to allow the nuclear system to grow beyond the 400-GWe Tevel even though a scarcity of U305 exists at costs below $160/1b. Viewed differently, om0 xm 0 200 2050 the primary effect of reprocessing is not @ = I MAXIMUM ANMNUAL REQUIREMENTS Uy0q - 110 - 10% 5T/ ENRICHMENT - 72 - 10° swyyr & HIGH COST U50g SUPPLY 130 - 108 $¥ BELOW $160 La & INSTALLED NUCLEAR CAPACITY, GWe to support the construction of additional Fig. 6.2-15. The Effect of U30g Supply on the Nuclear Contribution of the LWR with Plutonium Recycle (Case 2L). U30g is in plentiful supply. nuclear units in the earlier years when The installed nuclear capacity that must be located in the energy centers as a function of time is shown by the lower curve in Fig. 6. 2-14, the difference between the two curves indicating the nuclear capac1ty that can be made available outside the centers. The maximum capacity which must be located in the energy centers is approx1mate1y 260 GWe, while a maximum of 400 GWe can be available outswde the center. For approx1mate1y three decades (from the year 2000 to the year 2030), -over 300 GWe can be available outside the centers The use of pluton1um recycle to a]]ow ‘the nuclear system to grow beyond the 400-GWe level as the U303 supply becomes scarce is v1v1d1y 111ustrated in F1g 6.2- ]4 Note that the number of un1ts loaded with plutonium increases sugnif1cantly as the in- sta]]ed capac1ty exceeds the 400- GWe Tevel and that they compr1se an increas1ng fraction of the total 1nsta11ed capac1ty in 1ater years ' ' 6-32 LWR ' 87 Kg fis Pu 288 Kg U2 ] US(LEJY \ e lgg Kg U235 77 IR AN R AN NN 368 Kg fis Pu 35.5 X 10° SWU g7 gy 20 12,670 Kg HM —59.1 ST usoa-_@ Ij \ REPROCESSING & REFABRICATION SN NN NN N A NN RNONRNNNY SN OB - //I/I/7//////1///7//”/////[///)7[////////////////////II//II//////?//I//////////I ' HEDL 7805-090.30 Fig. 6.2-16. Utilization and Movement of Fissiie Material in a Nuclear System Consisting of LWRs Operating with Plutonium and/or Uranium Recycle (Case 2L, High- Cost U30g supply) {Year 2035). The effect of the intermediate-cost U;0g supply on the LWR plutonium recycle case is shown in Fig. 6.2-15. With 6.0 million ST U305 below $160/1b, the maximum nuclear contribution would increase from approximately 600 GWe in the year 2020 to approximately 960 GWe in the year 2045. Thus, the U30g supply is again\the dominant variable. The maximum annual U30g requirement would be 110,000 ST/yr and the maximum annual enrichment requirement would be 72 million SWU/yr. These annual requirements would constitute the principal limitation of the system. o - ~ The utilization and movement of fissile material per GWe of installed capacity for the LWR with plutonium recycle is shown in Fig. 6.2-16. Again this figure represents a snépshot of the system in time (in the year 2035) and includes both the first core loading for those reactors that are starting up and the last core discharge for those reactors that are shutting down. The annual U304 consumbtidn in 2035 is 59 ST U305/GWe, and the LWR utilizing plutonium comprises 54% of the installed capacity. Approximately 368 kg of fissile plutonium in fresh fuel per GWe of installed capacity per year'mfist be handled within the energy centers for this case. (Note: Simply identifying the amount of fissile plutonium in fresh fuel that must be handled is not analogous to determining the diversion resistance of the system. While the amount of fissile p]utohium being handled may be important, the state and location of the fissile plutonium and the procedures used to handle it are more important in assessing the diversion resistance of a system.) { - o o am «fl.‘, ] r r ri l — € ey o - r ’ ' C | ) - o lt BT 1 - e . i “wpr01iferétion resistance. If either of these items is desirable, however, - this option minimizing the production and use of plutonium does offer a significant increase in the energy support ratio and a significant decrease in the amount of fresh-fuel plutonium that must be handled. ' It is important to note that the deployment of the p]utonium minimization and utilization option would require the development of a nuclear industry capable of reprocessing fuel containing thorium and-befabricating fuel containing 232U. As Fig. 6.2-24 indicates, only one reactor providing 3% of the installed capacity in year 2035 does not utilize thorium. Thus, inrorder to successfully implement this option, 97% of the reprocessing capacity in year 2035 must be capable of handling fuel containing thorium, and 51% of the fabrication capacity must be capable of handling fuel containing 232U, In summary, a converter strategy based on the LWR which minimizes the amount of plutonium produced, but uses that which is produced, could supply a maximum nuclear con- tribution of 700 GWe with the high-cost U30g 5upp1y. This is approximately 100 GWe greater than the maximum nuclear contribution obtained in the case of plutonium throwaway | and fissile uranium recycle. The strategy does, howevef, reduire that approximately 100 GWe be located in an energy center. With the intermediate-cost U30g supply, the system could make a maximum nuclear contribution of more than 1000 GWe. In either case, the development of fuel designs capable of minimizing the amount of b]utonium produced and also the development of a nuclear industry capable of handling thorium-based fuels must be developed. 6.2.5. Converter System with Plutonium Production Not Minimized; Pu-t0-233U "Transmutation" This option'diffgrs from the preceding option .in that the dispersed reactors are not designed to minimize the amount of plutonium produced, Thus more plutonium is handled as fresh fuel and more is “transmuted" into 233U, Again a converter with a plutonium-thorium core is located in the energy center, and other reactors'are located outside the center (see Fig. 6.1-3, Option 5T). S Figure 6.2-25 shows that the nuclear contribution for this option using LWRs only (Case 5TL) reaches a maximum of approximately 640 GWe shortly before year 2025. The maximum contribution is less than the 700-GWe maximum in the preceding case primarily because of the different amounts of fissile plutonium utilized in the two sysfems. Since 23%uy is worth less in a thermal reactor than either 235( or'233U, the system which minimizes the amount of plu- tonium should (and does) make a slightly larger nuclear contribution. ] — = -t e e o 4 waaer —; (l ~} ' [_% + * ¥ o Tt ) ¥ $ - C ) 6-39 The fraction of the installed nuclear capacity which for this case must be located in energy centers is shown in Fig. 6,2-26 as a function of time. The maximum is approximately 120 GWe, which is slightly greater than that for the previous case. The amount of nuclear capacity available for tocation outside energy centers ranges from approximately 300 GWe in the year 2000 to approximately 500 GWe in the year 2025. The maximum annual U304 and enrich- ment requirements are 65,000 ST/yr and 45 million SWU/yr, respectively. These are quite similar to the maximum annual requirements for the case of the LWR with classical plutonium recycle (see Fig. 6.2-13). The disadvantage of this option is that the energy support ratio decreases continu- ously as the end of the U305 supply is approached. Figure 6.2-27 indicates that if a U304 supply of 6.0 million ST below $160/1b were available, the system would continue to grow 1000 T T T T T T THE LWR WITH PLUTONIUM TRANSMUTATION oo T T Y T T T . 800 |- ] CASE STL - THE LWR WITH PLUTONIUM TRANSMUTATION (;J STL -~ LWR WITH PLUTONIUM o 800 |- — ?_-‘ MAXIMUM ANMNUAL REQUIREMENT: TRANSMUTATION 3 o -85+ 100 s, : gwo - mmcr:agm -625 ~||o°ss{vyru/yr 7] E o - | S | 1 3 3 3 400 DENATURED LWR's : / g 2wl o RowAWAY CYCLE §m LWR = Pu/Th % 80 wlm zu'uu zolm a0 szaT P 2050 % w0 1990 2000 zo;a zn;o t zo:;o zo:o x50 YEAR YEAR Fig. 6.2-25. The Effect on the Nuclear Fig. 6.2-26. Relative Nuclear Contri- Contr1but3on of "Transmuting“ Plutonium butions of LWRs Located Inside (Pu/Th) and Produced in LWRs to 233y (High-Cost U304 Outside (Denatured LWRs) Energy Centers Supply). (PTutonium "Transmuted” to 233U) (High-Cost U30g Supply). 1000 T T - T —7 T : s _ until approximately year 2050, and thus the ol high energy support ratio associated with 5 | i AL B o this option could be maintained much longer. t LOQUIREMENT ; INTE RMEDIA‘I.E CcusT UJU!’ SUPPLY . . S sl upsflm-rE;Ar go-fstenowsioiel - The maximum annual U305 and enrichment 3 ENRICHMENT - 77 - 107 SWU/yr . ‘ ’. - . 3 \\, B requirements in this case are 109,000 ST/yr §4w O\ o ST uygg suecy and 77 million SWU/yr, respectively. Thus, 2 again we have an option for which the = - principal limitation would be the annual o . . L 1 ore and enrichment requirements. memeo e m LG ™ e e The utilization and movement of fissile ] N . materia j i Fig. 6.2-27. The Effect of Us0 Supply al per GWe of installed capacity for on the Nuclear Contribution of LWRs in Case 5TL in the year 2035 are shown in Fig. 6.2-28. System with Plutonium "Transmutation" (Case 5TL) . The annual U30g consumption is approximately 68 ST U;04/GWe, and the LWR utilizing plutonium comprises 18% of the installed capacity. Approximately 260 kg of fissile plutonium per GWe of instal]ed_capacity must be handlied as fresh fuel each year within the energy centers. This can be compared to the classical case of plutonium recycle in which 56% of the installed capacity is located in the energy centers and 368 kg of fissile plutonium is handled as fresh . fuel each year. Thus, using the plutonium to produce 233U results in a significant reduction in the amount of installed capacity that must be Tocated in secure regions, and it also reduces the amount of fissile plutonium that must be handled as fresh fuel each year. ' 104 Kg fis Pu 25 LWR 2 404 Kg U ¢~ USQE)U 154 Kg U ~13,710 I(g HM=={ 10.62 GWe 17,260 Kg HM = 61,9 ////////////////////l’//////////////.///.//////////////////I/////‘// \ 144 Kg fi s Pu 53 Kg U 4161 Kg Th 4598 Kg HM 261 Kg fis Pu . 52.4x 10° SWU 3848 Kg Th o 4250 Kg HM / - — 7.8 ST U0 -—@ | = kefeh u ko U Z : 3K U g 4157 Kg HM Z WR Z 141Kg U233 3PE / Al Z 2Kg U2 10.20 GWe 3304 Ka Th e / 4437 Kg HM P/ CF = 67.2 2 7777777 ] HEDL 7805-0%0.56 Fig. 6.2-28. Utilization and Movement of Fissile Material in an LWR Nuclear System "Transmuting" Plutonium to 233y (Case 5TL, High-Cost U30g Supply) (Year 2035). As for the preceding option, the high energy support ratio associated with this case requires the development of a nuclear industry capable of reprocessing significant amounts of fuel containing thorium and refabricating significant amounts of fuel containing 232, although these amounts are considerably smaller, As Fig. 6.2-28 -indicates, the LWR loaded ~with approximately 3% enriched 235U comprises 62% of the installed capacity in year 2035, the LWR loaded with Pu in Th comprises 18%, and the LWR loaded with 12% 233U in 2380 comprises 20%. Thus approximately 34% of the reprocessing capacity must be capable of handling fuel containing thorium and 20% of the fabrication capacity must be capable of handling fuel con- taining 232y, In summary, a converter strategy based on the LWR which "transmutes® all pl utonium to 233y could supply a maximum nuclear contribution of 640 GWe with the high-cost U304 - supply, of which about 120 GWe would be located in energy centers. While the nuclear con- tribution for this case is somewhat less than for the case in which the production of plutonium is minimized, it does not require the development of new reactor concepts and it will require handling smaller amounts of 233U, r-—-—-i: O oy ward - r T ! £ — i ri } e | Vl! ) 6-41 6.2.6 Converter-Breeder System with Light Plutonium "Transmutation" The results presentéd-%n the preceding sections'fiave demonstrated that nuclear power systems based on converter reactors will ultimately be limited by the quantity of economically recoverable uranium. While a larger U30g resource base will allow larger systems to develop, the converse is also true. Since the U30g resource base has always been somewhat uncertain, the deployment of fast breeder reactors has traditionally been considered as the method by which the consequences of this uncertainty would be minimized. Thus, it has historically been assumed that by deploying-FBRs nuclear power systems would outgrow the constraints naturally imposed by the U305 resource base. In the option discussed here (Option 6), an FBR with a plutonium-uranium core and a thorium blanket is located in the energy center to produce 233U which is then used in de- natured converter reactors outside the center. Because a higher plutonium "transmutation" rate could be obtained with a plutonium-thorium core in the FBR, this option is referred to as having a Zight "Pu-to-233U" transmutation rate. The individual reactor concepts contained in this option are shown in Fig. 6.1-4, The nuclear contribution associated with this option when all the converters utilized are LWRs (Case 6L) is shown in Fig. 6.2-29. In this case, even with the high-cost U30g supply, the system is capable of maintaining a net addition rate of 15 GWe/yr throughout the planning horizon - i.e., from 1980 through 2050. The ability of the nuclear system to maintain this net addition rate is a direct consequence of the compound system doubling time of the FBR, which, in this case, is 13 yr. This doubling time in turn is a direct consequence of the FBR having a Pu-U core. In this option the installed nuclear capacity which must be located in energy centers increases as a function of time to approximately 560 GWe in year 2050 (see Fig. 6.2-30). The most rapid increase occurs between 2010 and 2020 as the number of FBRs on line in- ~ creases significantly. The amount of nuclear capacity available for installation outside the centers increases from approximately 300 GWe in year 2000 to over 500 GWe in year 2050. Initially, the LWR loaded with approximately 3% enriched 235U is the principal reactor available, but as the U30g is depleted, it is replaced by the LWR loaded with 11% 233y in 238y, This is illustrated in Fig. 6.2-31, which also indicates that this option is capable of maintaining an energy support'ratio greater than unity throdghouf the planning horizon. ' ' ' The maximum annual U308'and enrichment requirements for this case are 62,000'5T/yr and 44 million SNU/yr, respectively. These annual requirements do not differ significantly from those obtained with the LWR on the throwaway cycle, the reason being that in either case, the goal of the nuclear power system is to maintain a net addition rate of 15 GWe/yr provided this increase can be sustained by the Us0g supply. The maximum installed capacity - 1000 T T T T T THE FBR WITH LIGHT PLUTONIUM TRANSMUTATION o 800 [~ - x MAXIMUM ANNUA:P REGUIREMENT; _ . ' U0, - 82 « 107 ST/ e - FBR WITH LIGHT PLU~ % ENRICHMENT - 44 + 10* SWui/r L TONIUM TIANSMUTATION & o0 |- - 3 a ! -t < & 400 - 3 § 0 1L - LWR ON THE THROW. T AWAY CYCLE 0 1 1 I 1 ] L. 1980 1990 2000 2010 220 2000 2040 2050 YeEAR Fig. 6.2-29. The Nuclear Contribution of an LWR-FBR System with Light Plutonium “Transmutation" (High-Cost U305 Supply). 1000 "CASE 6L = THE FAR WITH LIGHT PLUTONIUM TRANSMUTATION & g g WNSTALLED NUCLEAR CAPACITY, GWe g Fig. 6.2-31. Relative Nuclear Contri- butions of Each Reactor Type in LWR-FBR System with Light Plutonium "Transmutation" (High-Cost U304 Supply). 6-42 1600 T 1 T —T T CASE 6L = THE FBR WITH LIGHT PLUTONIUM TRANSMUTATICN, : a0 |- - o E DENATURED LWR's § 00 - s g 3 &0 o - 3 . = g 200 FBR-Pu/U/Th AND LWRPy/Th 1 1 1 1980 1990 2000 2010 2020 200 2040 260 YEAR Fig. 6.2-30. Relative Nuclear Contri- butions of Reactors Located Inside (Pu-Fueled) and Outside (Denatured LWRs) Energy Centers (High-Cost U305 Supply). o for the LWR loaded with approximately‘3% en- riched 235U in either case is approximately 420 GWe. However, in this option, as the in- stalled capacity of the 235U-loaded LWRs decreases, the energy center FBRs produce in- creasing amounts of 233U for the denatured LWRs, and thus the total installed nuclear capacity con- tinues to increase at a net rate of 15 GWe/yr. The amount of fissile plutonium that must be handled in the energy centers as fresh fuel each year is shown in Fig. 6.2-32, Approxi- mately 620 kg of fissile plutonium per GWe must be handled in this case, as pompared to approxi- mately 170 kg of fissile plutonium in fresh fuel per GWe each year for the case of plutonium minimization and utilization. Thus, it appears that the ability to maintain an energy support ratio greater than unity while simultaneously adding 15 GWe/yr will necessitate handling more fissile plutonium in fresh fuel in the energy centers, ' As pointed out “in previous cases, the ability to maintain a high energy support ratio requires the development of a nuclear industry capable of reprocessing.fuel containing thorium and refabricating fuel containing 232y. In this option in the year 2035, the LWR loaded with approximately 3% enriched 233 comprises approximately 28% of the installed capacity, the FBR comprises 48%, and the LWR Toaded with 11% 233U in 238y comprises 24%. Upon examining the flow of thorium and uranium metal associated with these reactors, it can be seen that 38% of the reprocessing capacity must be capable of handliné fuel con- taining thorjum and 27% of the fabrication industry must be capable of handling fuel ‘containing 232y, B ] # & r( - 4 . I [1 - e g - 1 i i - i £ ) e 1 i ) 6-43 The annual consumption of U30g in 2035 was found to be approximately 32 ST U;0g/GWe. This consumption rate will decrease continuously as the 235U-loaded LWR isreplaced with the 233))~1paded LWR, : 49 Kg fis Pu 235 LWR 6,679 Kg HM____ —0.28 Gwe 8,451 Kg HM ' TR T T 20 20 o 2o 27 ' LWR /Th Py = 0.0 GWe 24.7 X 10° swu 31.8 5T U,O Efl- — 378 | enRICH, 616 Ko Fis Pu : 16 Kg U 5,546 Kg Th 11,480 Kg HM 4,469 Kg Th 9,927 Kg HM (7 /‘ | 552 Kg fig Pu k O 102 kg U z b N\ 6,573 (Ll L HEDL 7805-090.32 Fig. 6.2-32. Utilization and Movement of Fissile Material in an LWR-FBR Nuclear System with a Light "Pu-to-233U" Transmutation Rate (Case 6L, High-Cost U30g Supply) (Year 2035). In summary, a strategy based on an FBR with a Pq-U core and a thorium blanket could supply a net addition rate of 15 GWe/yr to the year 2050 and beyond with a Ui05 supply of 3 million ST below $160/1b. The installed nuclear capacity in 2050 would be 1100 GWe, with 560 GWe, or approximately 50% of the installed capacity, Tocated in secure energy centers. Approximately 27% of the:fabricatfon capécity must be capable of handling fuel containing 232y, Thus, while a nuclear system based on an FBR with a Pu-U core and a thorium blanket can supply 15 GWe/yr for an indefinite period of time, it simu]taneoué]y requires that a significant amount of nuclear capacity be located in secure regions. 6-44 6.2.7. Converter-Breeder System with Heavy Plutonium "TranSmutation“ The preceding discussion indicates that a nuclear power system that includes an FBR having a Pu-U core and producing 233U in a thorium blanket can maintain an energy support ratio greater than unity while simultaneously adding 15 GWe/yr to the installed capacity throughout the planning horizon. The possibility exists, however, that a nuclear power system that includes an FBR having a Pu-Th core and a thorium blanket would result in a heavy Pu-to-233Y transmutation rate which would maintain an‘energy'subport ratio signi- ficantly greater than unity over the same period of time. The principal problem associated with a nuclear system based on an FBR with a Pu-Th core is that the breeding ratio of the breeder, and hence the breeding ratio of the entire system, tends to be low. Therefore, the effect of adding to the system an FBR operating on denatured 233U to augment the 233y production was also investigated, The individual reactor concepts contained 1n this system are shown in Fig. 6.1-4 (Option 8). ' “The nuclear contribution associated with.this option (Case 8L, with denatured breeder) is compared to that of the LWR on the throwaway cycle for the high-cost Uj0q supply in Fig. 6.2-33. The system is capable of ma1nta1n1ng a net addition rate of 15 GWe/yr throughout the plann1ng horizon. The installed nuclear capacity which for Case 8L must be located in energy centers is shown in Fig. 6.2-34 as a function of time., The maximum is less than 300 GWe through- out the planning horizon., The amount available for location outside the energy centers ranges from approximately 300 GWe in the year 2000 to approximately 800 GWe in the year 2050. This can be compared to Option 6 for which the nuclear capacity that must be located in secure regions increases continuously to approximetely 560 GWe in 2050. Thus, a nuclear system containing FBRs with Pu-Th cores plus FBRs with denatured 233U cores is capable of maintaining a very high energy support ratio for an indefinite period of time. It does require, however, that reactors that are net producers of fissile material be located in energy centers. | ' ~ The utilization and movement of fissile material in year 2035 for_Case 8L and the small U30g supply are shown in Fig. 6.2-35. The LWR loaded with approximate]y‘B% enriched ' 235y comprises approximately 13% of the 1nsta1]ed capacity, the denatured 235U LWR compr1ses approximately 12%, the energy center FBR compr1ses approximately 29%, the denatured 233y LWR comprises 8%, and the denatured FBR comprises 38%. The denatured 235U LWR is being rapidly phases out of the nuclear system in year 2035, while the denatured 233U LWR is being rapidly phased in. This is indicated in Fig. 6.2-35 by the fact that the heavy metal dis- charge for the denatured 233U LWR is considerably greater than the heavy metal charge, while the heavy metal charge for the denatured 233U LWR is considerably greater than the heavy metal discharge. The former is indicative of final core discharges, while the latter is indicative of first core loadings. vl & T o g B .,‘.i 1 (’"‘! ' ' (i i ) - - 1 — R 6-45 1000 T Y Y Y Y — 1000 T T T T T CASE BL - THE FBR WITH HEAVY PLUTONIUM TRANSMUTATION CASE 8L - THE FBR WITH HEAVY PLUTONILM TRANSMUTATION (3 800 - -l osm [ -l x = o ° MAXIMUM ANNUAL REQUIREMENTS ; Z Bt - FBR WITH HEAVY PLU- T U0, - 68 © 10° ST/ E, o0 |- TONIUM TRANSMUTATION# 9 0 ENRICHMENT - 55 « 0% SWU/pr i 3 3 DENATURED LWR's AND FBR's 3 Sl 1 fw 2 2 < 2 2 0 - 2 200 - m: \?r& TCHLEE THROW- FBR-Pu/Th/Th 0 1 1 1 4 L 1 0 ¥ " 1980 1990 2000 2010 2026 2030 2040 2050 1980 1990 2000 2000 2020 2030 2040 250 YEAR ‘ ‘ ’ YEAR Fig. 6.2-33. .The Nuclear Contributions Fig. 6.2-34. Relative Contributions of an LWR-FBR System with Heavy Plutonium of Reactors Located Inside (Pu-Fueled) and "Transmutation" (High-Cost U30g Supply). Outside (Denatured LWRs and FBRs) Energy Centers (High-Cost U30g Supply). In this option the annual consumption of U30g is approximately 25 ST U30g in year 2035, decreasing thereafter as the LWRs loaded with 235U are replaced by the LWRs loaded with 233y, Approximately 430 kg of fissile plutonium per GWe of installed capacity must be handled as fresh fuel each year within energy centers,'§omewhat less than the 620 kg that must be handled in Option 6. The ability to maintain a high energy support ratio while simultaneously adding 15 GWe/yr again requires the development of a nuclear industry capable of reprocessing fuel containing thorium and refabricating'fue1 containing 232y, Figure 6.2-35 shows thap 65% of the reprocessing capacity in year 2025 must be capable of handling fuel containing thorium and that 31% of the refabrication capacity must be capable of handling fuel containing 232y, | The effect of deleting the denatured FBR from the system is shown in Figs. 6.2-36 and 6.2-37. Figure 6.2-36 shows that without the denatured FBR the installed nuclear capacity reaches a maximum of approximately 840 GWe in about 2035 and declines continuously there- after. The reason for this, of course, is that without the denatured FBR the system has a net breeding ratio of less than unity. Therefore, while the system can multiply the fissile supply significantly, 1t cannot continue to grow indefinitely.' The nuc]éar capacity that must be located in energy centers for the modified Case 8L is shown in Fig. 6.2-37. ‘This capacity does not exceed 140 GWe throughout the planning horizon. The amount of capacity available for location outside the secure regions ranges from approximately 300 GWe in the year 2000 to approximately 700 GWe in year 2035.. " In summary, a strategy based on an FBR with a Pu-Th core and a thorium blanket can supply a net addition rate of 15 GWe/yr to year 2050 and beyond provided a denatured'breeder is included in the system. If the denatured breeder is not included, then the maximum nuclear contribution would be approximately 840 GWe. The amount of nuclear capacity that must be located in secure regions does not exceed 140 GWe in this case. LWR 19 Kg fi;al’su 97 kg UPS L USILEWU 29 Kg U ' 3320 Kg A [__]0.13 GWe 3312 Kg HM 6-46 CF=67.6 ST S S LSS SSLS LSS ST S SS S S LSS LSS LSS LSS SIS A LSS LSS LI ISV, EBR 233 U3-U/Th 466 Kg U™ e | 8926 Kg HM CF=68.2 | 180 Kg fis Pu 304 Kg U 3758 Kg Th 12,7 % 107 swy 8309 Kg HM — 3 ' ' B :fi&fififlfik uzzsruo——@ ' 2 Kg fis Pu ‘ 3% 430KgfisPuaKU 13 Ky U 83X 103 SWU 1 Kg fls Pu 637 Kg HM - 217 Kg s | Pu 'ég ‘d — I:s(ms)/u 188 Kg UZ 235 kg 2| o B0 owe 4509 Kg Th . 2821 Kg HM 2 27 1 5624 Kg HM _’ . TWRN 13 Kg fis Py 128 kg UBS ISRV 44 Kkg U . 2309 Kg Th [ 10.12 GWe 3589 Kg Th 2917 Kg HM CF=60.3 ‘4522 Kg HM HEDL 7805-090.29 Fig. 6.2-35. Utilization and Movement of Fissile Material in an LWR-FBR Nuclear System with Heavy "Pu-to-233U". Transmutation Rate (Case 8L, High-Cost U30g Supply) (Year 2035). 1000 T T T T T 1 THE FBR WITH MEAVY. PLUTONIUM TRANSMUTATION D WITHOUT A DENATURED BREEDER . 800 - U0y SUPPRLY - 3.0 - 16° 5T N 3 a FBR WITH HEAVY z PLUTONIUM TRANSMU- E TATION AND WITHOUT A X &0 |- DENATURED BREEDER — 3 - 5 3 %00 |- - a IL - LWR ON THE THROW- 3 L/ AWAY CYCLE 3 £ 1 1 1 1 L 1 1980 1990 2000 2010 2020 2030 2040 2050 Fig. 6.2-36. Effect on Nuclear Contri- bution of Eliminating Denatured Breeder from LWR-FBR S_ystem with Heavy Plutonium "Trans- mutation." {(Case 8L Minus Denatured Breeder) (High-Cost U30g Supply). Y000 T T ™1 T T THE FBR WITH HEAVY PLUTONIUM TRANSMUTATION AND WITHOUT A DENATURED BREEDER , 80 U,Op SUPPLY - 3.0 - 10° 57 » £ & < o0 . § DENATURED LWR™ X 3 i § 400 -1 o w S < 2 200 - m-n./n-./?h-"'——' L ! L 1 1 1 Fig. 6.2-37. Relative Nuclear Contri- butions of Reactors Located Inside (Pu-Fueled) and Qutside (Denatured LWRs) Energy Centers (Case 8L Minus Denatured Breeder-) (High- Cost U30s Supply). ! -t ..a...., S — oo s e — o« — 1 . d"‘ e ;_l 1 e 4 £ | 1 e Y o { e 6-47 6.3. CONCLUSIONS The principal conclusions developed during the course of this study are summarized in Tables 6.3-1, 6.3-2, and 6.3.3. From the preceding discussion and Table 6.3-1, the following conclusions are drawn for various nuclear systems operating on the throwaway cycle: (1) With a U305 supply of 3.0 million ST below $160/1b, the maximum installed capacity with the standard LWR on the throwaway cycle would be approximately 420 GWe, and this would occur in about year 2006. (2) A reduction in the U;0g requirement of all LWRs commencing operation in 1981 and thereafter by 6% would not significantly increase the maximum installed capacity. Thus, for the case of the LWR on the throwaway cycle, the effort should be on improvements in U30g util- ization significantly greater than 6% for LWRs commencing operation after 1981 or on improve- ments which can be retrofitted into existing LWRs. Table 6.3-1. Summary of Results for Nuclear Power Systems Operating on the Throwaway/Stowaway Cycle Technology Maximum Nuclear Development Contribution Year of Maximum Option Requirement (GWe) Contribution High-Cost U304 Supply Standard LWR None 420 2006 Improved LWR LWR with extended dis- 430 2010 charge exposure LWR plus advanced - SSCR, HTGR, or HWR . 450 2012* converter _ . . ' o _ o LWR with improved Advanced enrichment 500 2015 tails composition process : S - Intermediate- Cost U308 Supply Standard LWR ~~ Successful U;04 explora- 730" | 2030 o B tion program - - LWR plus advanced - SSCR, HTGR, or HNR also 850 2035 ‘converter L successful U308 exp]orat1on D - : program *Depends on advancéa‘converter concept and 1ts introduction Hate. 6-48 Table 6.3-2. Summary of Results for Nuclear Power Systems Utilizing LWR Converters with and without FBRs (with Recycle) Maximum Nuclear Contribution Option ‘Technology Development Year Total Fraction of GWE ‘ Requirement Glle in Energy Center High-Cost U304 Supply . Py recytle (2L) ’ Reprocessing, refabrication 2020.. 600 ~0.40 Pu throwaway {(4L) Advanced fuel design, repro- 2020 590 - cessing Pu production minimized, Advanced fuel design, repro- . 2030 700 0.15 Pu-to-2330 "transmutation” (SUL) cessing ) 7 Pu production not minimized, Advanced fuel design, repro- 2025 640 0.21 Pu-to-233U “transmutation" (STL) cessing . . S : : FBRs added, light Pu Advanced fuel design, repro- ~2050 >1100 ~0,56* transmutation ({6L) cessing, FBR (w/o denat. FBR) FBRs added, heavy Pu Advanced fuel design, repro- »2050 ~1100 ~0.27* transmutation: (7L) ~ cessing, FBR {with denat. FBR) : 22035 ‘850 ~0.16 . {w/o denat. FBR) Intermediate-Cost U304 Supply Pu recycle (2L) Reprocessing, refabrication 2045 960 | - Pu throwaway (4L) Advanced fuel design, repro- 2045 980 - cessing _ Pu production minimized, Advanced fuel design, repro- >2050 >1000 - Pu-to-233y "transmutation" {5uL) cessing Pu production not minimized, n2050 1020 - Pu-to-233y “transmutation" (5TL) " Advanced fuel des1gn, repro- cessing *In year 2050. (3) The deployment of an advanced converter beginning in 1995 will not signifi- cantly increase the maximum installed capacity if thg U504 supply is limited to 3.0 million ST below $160/1b. This is primarily due to the‘fact that a significant amount of the U305 supply has been committed to the standard LWR prior to the advanced converter attaining a large fraction of the instalied capacity. large as 6.0 million ST below '$160/1b, then the effect of the advanced converter is cons1derab1y larger. If'fhe U305 supply should be as (4) An advanced enrichment process capable of eCQnomica]ly reducing the tafi]s compo- sition to 0.0005 could-have a greater effect than improvements in LWR U30g utilization or the deployment of an advanced converter. e el e =il r / { il 4 = e £ r ol - — 1 6y e ( 1w ' r—i £ £ — - | t i w ! 1 '.»' s g 6-49 Table 6.3-3. Summéry of Fuel Cycle Requirements for Nuclear Power Systems Utilizing LWR Converters with and without FBRs (with Recycle; High-Cost U30g Supply) Fraction of Installed Nuclear Capacity Permitted Outside Fraction of Reprocessing Fraction of Refabrication Energy Center in Capacity to Handle Th Capacity to Handle Option Year 2025 in Year 2035 232y in Year 2035 Pu recycle 0.61 -0 0 Pu throwaway 1.00 _ 0.95 0.57 Pu production minimized; 0.85 0.97 0.53 Pu-to-233U "transmutation” Pu production not minimized; 0.79 0.34 0.20 Pu-to~233J "transmutation” FBRs added, light Pu 0.56 0.38 0.27 transmutation FBRs added, heavy Pu 0.76 0.65 0.31 transmutation (5) The effect of an exploration program successful enough to reliably increase the U305 resource base to 6.0 million ST below $160/1b would be considerably greater than any of the above. Thus, when analyzing the throwaway option, the size of the U;05 resource base and the uncertainty associated with it dominate the analysis. From the discussion in Section 6.2 and Tables 6.3-2 and 6.3-3, the following conclu- sions are drawn for LWR and LWR-FBR systems operating with recycle: (1) With the high-cost U305 supply, the effect of plutonium recycle in LWRs would be to increase the installed nuclear capacity to 600 GWe, and this would occur in about year 2020. This would require, however, that as much as 40% of the nuclear capacity be located in the energy centers. If the U30g supply should be as large as 6.0 million ST below $160/1b, the maximum installed nuclear capacity would be 960 GWe, and this would occur in about year 2045. (2) .If all plutonium were thrown away but f1ss11e uranium were refabricated and reloaded the maximum installed nuclear capacity could be as large as 590 GWe with the high-cost U305 supply. Attaining 590 GWe, however, requires the development of fuel designs which minimize the amount of plutdnium produced. In addition, it requires the development of an 1ndustry in which as much as 95% of the reprocess1ng capacity is devoted to fuel containing thorium and as much as 57% of the refabrication capac1ty is devoted to fuel containing 232y, (3) If the p]utoniUm produced in the Systém*described'immediately'above were re- fabricated and reloaded, the'maximum installed nuclear capacity would increase to approxi- mately 700 GWe, which is an increase in the maximum of approximately 110 GWe. 6-50 (4) If all plutonium produced were transmuted to 233U but no attempt was made to minimize the amount of plutonium produced, the maximum installed nuclear capacity could be as large as 640 GWe with the high-cost U30g supply. As much as 21% of the installed nuclear capacity would have to be located in secure energy centers, however, and it would require that 34% of the reprocessing capacity be devoted to fuel containing thorium and 20% of the refabrication capacity be devoted to fuel containing 233y, (5) If a nuclear system utilizing an FBR with a Pu-U core and a thorium blanket were developed, the system could maintain a net addition rate of 15 GWe/yr indefinitely. The installed nuclear capacity, in this case, could be as high as 1100 GWe in year 2050; however, 56% of this capacity would have to be located in secure energy centers. Also, approximately 38% of the reprocessing capacity would have to be devoted to fuel containing thorium and 27% of the refabrication capacity would have to be devoted to fuel containing 232y. (6) If a nuclear system utilizing an FBR with a Pu-Th core and a thorium blanket were developed, the maximum installed capacity would depend upon the performance characteristics of the denatured design receiving fuel from the FBR. If this design were a denatured breeder, the nuclear system would be capable of adding 15 GWe/yr indefinitely. If, however, the design were a denatured LWR; then the installed nuclear capacity would increase to approxi- mately 850 GWe in about year 2035 and decrease thereafter. ‘ In addition to the results and conclusions presented in this chapter, detailed results for all the nuclear policy options calculated are tabulated in Appendix C. Also, as men- tioned earlier, a separate analysis performed under the assumption of an un1imited'U303 supply but with the nuclear power systems in competition with coal-fired plants is described in Appendix D. ' Chapter 6 References 1. R. D. Nininger, "Remarks on Uranium Resources and Supply," Fuel Cycle 78, Atomic Industrial Forum, New York, March 7, 1978. 2. John Klemenic, Director, Supply Analysis Division, Grand Junction Office, DOE Uranium and Enrichment Division, "Production Capability," October 1978, 3. John Klemenic and David Blanchfield, Mineral Economist, Grand Junction Office, "Produc- tion Capability and Supply," paper presented at Uranium Industry Seminar, October 26-27, 1977, Grand Junction, Colorado; proceedings published as GJ0-108(77). _4. "Uranium Enrichment Serv1ces Activity F1nanc1a1 Statements for Peraod End1ng September 30, 1977," p. 13, Schedule C, ORO-759. 5. “AEC Gaseous Diffusion Plant Operations," OR0-684, USAEC (January 1972). 6. "Data on New Gaseous Diffusion Plants," OR0-685, USAEC (April .1972). 7. See also, T. M. Helm, M. R. Shay, R. W. Hardie, and R. P. Omberg, "Reactor Design Characteristics and Fuel Inventory Data," TC-971, Hanford Engineering Development Laboratory (September 1977). ' ol 4 OO i - ~ 1 dzji w 4 o o KT ey Y [* | - - [ -»--u»,,' | | . - = s, iy i Nl L ¥ 7.0. 7.1. 7.2, 7.3. 7.4. 7.5. CHAPTER 7 OVERALL ANALYSIS OF DENATURED FUEL SYSTEMS Chapter Qutline Introduction, T. J. Burns, ORNL Proliferation-Resistant Characteristics of Denatured 233U Fuel, ¢. M. Newstead, BNL, and T. J. Burns, ORNL Isotopic Barrier of Fresh Fuel Gamma-Radiation Barrier of Fresh Fuel Spent Fuel Fissile Content Conclusions s =l —t o et . - * * =W P Impact of Denatured 233U Fuel on Reactor Performance and Selection: Comparison with Other Fuel Cycles, T. J. Burns, ORNL 7.2.1. Thermal Reactors 7.2.2. Fast Reactors 7.2.3. Symbiotic Reactor Systems 7.2.4. Conclusions Prospects for Implementation and Commercialization of Denatured 233U Fuel Cycle, J. C. Cleveland and T. J. Burns, ORNL . 7.3.1. Possible Procedure for Implementing and Commercializing the Denatured Fuel Cycle 7.3.2. Considerations in Commercializing Reactors Operating on Alternate Fuels 7.3.3. Conclusions Adequacy of Nuclear Power Systems Utilizing Denatured 233U Fuel for Meeting Electrical Power Demands, M. R. Shay, D. R. Haffner, W. E. Black, T. M. Helm, R. W. Hardie, and R. P. Omberg, HEDL . The Analytical Method Data Base . Results for Price-Limited Uranium Supplies Results for Unconstrained Resource Availability Systems Employing Improved LWRs and Enrichment Technology . Conclusions e e Y S~ BRSNS NN - Tradeoff Analysis and Overall Strategy Considerations, 7. J. Burns and I. Spiewak, ORNL .1. No-Recycle Options .2. Recycle Options .3. OQverall Conclusions and Recommendat1ons '\.I\I\.l U'IU"IU'I (e ' r— by e d 1 | i:"":( 7.0. INTRODUCTION T. J. Burns O0ak Ridge National Laboratory The assessment of any proposed fuel cycle must of necessity consider various topics that affect the feasibility and viabi]ity of the particular cycle. Moreover, an assessment of a particular fuel cycle must consider the relative merits of the fuel cycle compared to other potentially available fuel cycle options. This study of the denatured 233U fuel cycle has addressed various aspects of the cycle in the preceding chapters: the proliferation- resistant characteristics of the cycle (in Chapter 3); the impact of denatured 233y fuel on the performance of several types of reactors (in Chapter 4}; the implementation and com- mercialization aspects of the denatured fuel cycle (in Chapter 5); and the economic/resource implications of the cycle (in Chapter 6). In each of these chapters, the assessment of the denatured 233U cycle was limited primarily to the specific aspect under consideration. In this chapter the detailed results of the assessment are summarized and integrated, and the potential tradeoffs possible between the various considerations are addressed. In addition, recommendations for further study of crucial aspects of the denatured 233U fuel cycle are made. 7-4 7.1. PROLIFERATION-RESISTANT CHARACTERISTICS OF DENATURED 233U FUEL . C. M. Newstead Brookhaven National Laboratory " and T. J. Burns - 0ak Ridge National Laboratory - As has been stated in earlier chapters, the priméry goal of the denatured fuel cycle js to permit the recycle of fissile fuels in dispersed reactors in a manner consistent with nonproliferation considerations. In this section the proliferation-resistant character- istics of the denatured 23 %: fuel cycle that have been described in detail in Chapter 3 are summarized, and their significance with respect to both national proliferation and subnational terrorism is -noted. In general, these characteristics derive from three - distinguishing features of the denatured fuel cycle: (1) the intrinsic. isotopic barrier of the fresh denatured fuel, (2) the gamma radiation barrier associated with the 232y impurity present in thorium-derived fuel, and (3) the low chemically separable fissile content of the spent denatured fuel. 7.1.1. Isotopic Barrier of Fresh Fuel The isotopic barrier of the fresh fuel is created by the addition of the 238y denaturant to the 233y fissile fuel, its purpose being to preclude the use of the 233y directly in a nuclear weapons program. Although the thorium present in most proposed denatured fuels could be chemically removed, the separated uranium would have too low a fissile content for it to be directly usable in a practical nuclear device. By contrast,. the other potential fuel cycle relying on recycled material, the Pu/U cycle, would require only a chemical separation to extract weapons-usable material directly from power reactor fuel. The isotopic barrier in denatured fuel is not an absolute barrier, however, since any isotope separation (i.e., enrichment) technique can be used to circumvent it. Depending upon its technological resources, a nation may have or may develop separation facilities. On the other hand, it is unlikely that a subnational group would possess jsotopic separation capabilities and thus the isotopic barrier inherent in denatured fuel would provide considerable protection against terrorist nuclear activities. As is pointed out in Section 3.3.4 and Appendix A, enrichment technology has made great strides in recent years and is presently undergoing rapid further development. Ten years ago the only operational enrichment facilities were based on the gaséous diffusion technique, a method requiring a large expenditure of energy and a large plant to be economic. Today the gas centrifugation technique, which requires a significantly Tower energy consumption than the gaseous diffusion method, is available and is practical with small-scale plants. For example, the URENCO consortium is currently operating centri- fuge enrichment plants of 50 tonnes per year capacity at Capenhurst in the United Kingdom and at Almelo in The Netherlands. The URENCO centrifuge represents an economic design built by technologically advanced countries (England, The Nether]énds, Germany) t r — o= e L e e — S . r— - iy 4 | | r— L 7-5 without benefit of U.S. experience. For a military program, economics would not be an overriding criterion and could be sacrificed in favor of a more moderate level of technology. Moreover, the oben literature contains sufficient information concerning the centrifuge designs to guide mechanically competent engineers with access to adequate facilities. Replication of an economic design would require a somewhat higher level of technology than prototype construction. The following particular points regarding the enrichment of denatured 233U fuel should be noted: (1) Because of the lower mass of 233, separating 233 from 238U would require only 9/25 of the effort required to separate 235y from 238U, assuming equal feed enrichments. (2) Since the fast critical mass of 233 is less than that of 235U, less enrichment capacity would be required to produce a 233 ‘weapon from 233\/238y feed than would be required to produce a 235U weapon from 235U/238y feed, again assuming equal enrichments of the feed material. | | {3) The higher the énrichment of the source material, the less separative work that would have to be done to upgrade the material to 90% enrichment. For example, enriching natural uranium to a 10% level consumes 90% of the separative work required to achieve a 90% level. It is to be noted that the enrichment of denatured 233 fuel is approximately 12%, whereas the enrichment of currently used LWR 235U fuel is around 3-4%. With respect to items (2) and (3), a rough comparison can be made of the feed requirements and the number of centrifuges that would be necessary to produce 90% enriched material from various fueis in one year (normalized to 1 kg of product): Number of Centrifuges Required Feed Required 0.3 kg SWU/yr 5 kg SWU/yr Fuel _ (kg) Capacity Capacity 124 233%y . 8 o | 55 3 20% 235y | 5 50 3 3.24 235y 30 292 7 Natural Uranium 178 ST 46 The above values do not consider measures to eliminate the 232U contamination and they assume that a reasonable tails assay will be maintained (~0.2% 235y), If a higher tails assay were acceptable, the number of centrifuges could be reduced but the feed material required would be increased. | | ' One year, of course, is a long time when compared to a period of weeks that would be needed to obtain approximately 10 kg of plutonium by chemically reprocessing two to three spent LWR-LEU fuel elements. It would be possible to speed up the process time for the centrifuge method either by increasing the individual machine capacity, by adding additional centrifuges, or by operating at a higher tails assay. Increasing the capacity 7-6 would be quite difficult and would require increasing technological sophistication; how- ever, adding centrifuges would require only that the same device be duplicated as many times as necessary. Increasing the tails assay would require more feed material. Finally, in considering the hotentia] circumVention of the isotopic barrier, it is important to anticipate the enrichment techno]oéies that could exist in 20 to 25 years - the time when the denatured fuel cycle could be deployed. Technologically advanced countries already have the necessary technological base to design and construct centri- fuges, and many presently developing countries may have acquired the technoTogy base by that time. Countries with a primitive technology are unlikely to use this route, since even with the financial assets and technically competent personnel they would have the difficult task of developing the requisite support facilities. Other potential isotope separation techn1ques are under development in many countries. Laser isotope separation (LIS), plasma techniques, aerodynamic methods, chem1ca] techniques, and electromagnetic separation methods currently show varying degrees of promise. The current status of these methods is discussed in Appendix A. It is 1mpossib]e to predict the ultimate success or failure of these aTternatlve methods, and hence the isotopic separation capability which m1ght exist in 25 years is even more d1ff1cu1t to estimate. Current estimates for the U.S. development program in LIS and plasma methods suggest that it will be at least ten years before such methods could be operative on a working industrial basis, even with a highly sophisticated R&D effort. 7.1.2. Gamma-Radiation Barrier of Fresh Fuel The production of 233U results in the concomitant production of a small but radio- actively significant quantity of 232U through the 232Th(n,2n) reaction [and the 230Th(n,y) " reaction if 230Th is present in the thorium]. As the 232U decays through 228Th and its daughter products, the gamma activity of the 233y-containing fuels increases, thus providing a radiation barrier much more intense than is found‘in other fresh fuels. While chemical processing could be employed to remove the 232U decay products, such a procedure would provide a relatively low radioactivity for only 10-20 days, since further decay of the 232U present in the fuel would provide a new population of 228Th and its daughters, the activity of which would con- tinue to increase in intensity for several years. | The concentration of 232)) in the recycle fuel is usually characterized as so many parts per million (ppm) of 232U in total uranium. Due to the threshold nature of the 232Tp(n,2n) reaction, the 232 concentration varies with the neutron spectrum of the reactor in which it is produced. It also varies with the amount of recycle. For 12% 233 denatured fuel, the 232U concentration (in ppm U) rangés from 250 ppm for LWR- produced 233U to a maximum of 1600 ppm for certain LMFBR-derived denatured fuels (see Section 3.1.3). If the latter material were enriched to produce weapons-grade material, the 232( concentration would be approximately 8000 ppm, and thus the material would be highly radioactive. r— 3 ey 4 4 . - K £ While the radiation field would introduce complications in the manufacture of a weapon, particularly for a terrorist group, the resulting dose rates would not provide an absolute barrier (see Section_3.3.5). As mentioned above, it would be possible to clean up the fissile material so that it was relatively free of radiation for a period of 10 to 20 days. Alternatively, providing shielding and remote handling would allow the radiation barrier to be circumventedy however, construction and/or acquisition of the shielding, remote handling equipment, etc., could increase the risk of detection of a covert pro- gram before its compietion. Non-fissile material included in the weapon would also provide some shielding during delivery, and additional shadow shielding to protect the operator of the delivery vehicle and to facilitate the loading operations could be developed. In another approach, the 232U could be separated from the 233U by investing in a rather large cascade of over some 3000 centrifuges, possibly including 228Th cleanup to 1imit the radiation contamination of the centrifuges. A willingness to accept certain operational disadvantages would permit the radiation-contaminated material to be processed in the cen- trifuges provided they were shielded and some provision was made for remote operation. By comparison, clean mixed oxide Pu/U fuel would have a much less significant radiation problem and the currently employed fresh LEU fuel would have essentially none at all. 7.1.3. Spent Fuel Fissile Content Spent denatured fuel contains three possible sources of fissile material: unburned 233Y; 233pa which decays to 233y; and Py produced from the 238 denaturant. Use of the uranium contained in the spent denatured fuel is subject to all the considerations out- Tined above and would also be hindered by the fission-product contamination (and resultant radiation) inherent in spent reactor fuel. As was noted in Section 3.3.4, the relatively Tong half-1ife of 233pa (27.4 days) could permit the production of weapons-grade material via chemical separation of the 233%a; however, such a procedure would require that chemical separation be initiated shortly Upon discharge from the reactor (while radiation levels are very high) to minimize the amount of 233Pa which decayS'to 233y while still contained in the 238U denaturant. Moreover, since the d1scharge concentration of 233Pa is typically 5% of that of 233y, a cons1derab]e heavy metal processing rate would be required to recover a significant quantity of 233, (and hence 23%)) within the time frame avail- able. The pluton1um concentrat1on 1s comparable to that of 235$a, but very Tittle s lost by decay. Hence, the spent fuel can be allowed to cool for some time before reprocessing. It would seem, therefore, that if denatured 233y spent fuel were d1verted it wou]d be primarily for 1ts p1uton1um content. ~ Any fuel cycle utilizing 233U inevitably leads to some plutonium production. Compared to the LEU cycle and the Pu/U cycle, the denatured 233U fuel cycle reduces the plutonium production by (1) employing as little 238U as necessary to achieve the denaturing objective, and (2)*repldcing'the displaced 2381 with 232Th to enhance the i 7-8 production of "denaturable" 233, The plutonium production rates fpr'variOus reactors operating on conventional and denatured fuel cycles are discussed in Chapter 4 and summarized in Table 7.1-1, where the Light-Water Reactor (LWR) is represented by the pressurized-water reactor (PWR); the SSCR {Spectral-Shift-Controlled Reactor) is a modified PWR; the heavy-water reactor (HWR) is assumed to be a slightly enriched CANDU; the'High-Temperature Gas-Cooled Reactor (HTGR) is taken to be the Fort St. Vrain plant; and the High-Temperature Reactor (HTR) of the Pebble-Bed Reactor (PBR).type is'represented by the West German design. Plutonium discharge data for Fast Breeder Reactors (FBRs) represented by the Liquid-Metal Fast Breeder Reactor (LMFBR) are included for comparison. It is quite clear from Table 7.1-1 that the denatured fuel cycle for the HWR gives the greatest reduction in plutonium production between the regular and denatured cycles. The HTGR has about the same absolute plutonium production for the denatured fuel cycle as the HWR and in both cases the plutonium amounts are rather small., The HTR-PBR is best in absolute minimum plutonium production, yielding only 14 kg/GWe-yr and even less in a highly optimized design. o o - Table 7.1-1. Fissile Plutonium Discharge for Various Reactor and Fuel Cycle Combinations (Capacity Factor = 0.75) Fissile Pu Discharge (kg/GHe-yr) LEU Cycle Pu/U Cycle Denatured Cycle LWR 174 858% 63 SSCR 196 - 72 HWR (CANDU) 183? - 32 HTGR 72 - 36 HTR-PBR 63 - 14 LMFBR - 991 347 gP]utonium burner. S1ightly enriched CANDU. For the LWR, SSCR and HWR the percentage of the discharge plutonium that is fissile plutonium is approximately the same for the denatured cycle as for the LEU cycle. For the HTGR and PBR, the fissile pTutonium percentage is only ~39% for the denatured cycle (comparéd to 56% for the LEU cycle). Further, the discharge plutonium from the HTGR and PBR, and also from the HWR, is more diluted with other heavy material by a factor of three to four than that from the LWR or SSCR. Thus, more material must be processed in the HTGR, HTR, and HWR to obtain a given amount of plutohium, which provides an additional prolifera- tion restraint associated with spent fuel discharged from these reactors. However, the on-Tine. refueling feature of the CANDU, and also of the PBR, may be a disadVantage from a proliferation viewpoint since low-burnup fuel could be removed and weapons-grade plutonium extracted from it. On the other hand, premature discharge of low-burnup fuel from the reactors would incur economic penalties. 1 ¥ e ik r— ] L3 o sl oy 4 r— ( e ] . T Emt_ i ! = o ol e C 7-9 Viewed solely from the plutonium production viewpoint, the order of preference in terms of higher proliferation resistance for the various denatured reactor candidates to be employed at dispersed sites is as follows: HTR-PBR, HWR, HTGR, LWR, and SSCR. However, other factors must aiso be addressed in evaluating the candidate reactors, one of which is that their plutonium production maintains the symbiosis of a system that includes plutonium-fueled 233U producers in secure energy centers. This plutonium being consumed within the center as it is recovered from the spent fuel would limit the amount of plutonium available for possible diversion. While such an energy center could also be impTeménted for the Pu/U cycle, the denatured cycle would permit the dispersal of a larger fraction of the recycle-based power generation capability. Hence, the number and/or size of the required energy centers might be markedly reduced relative to the number required by the Pu/U cycle. 7.1.4. Conclusions ~ The proliferation-resistant characteristics of the denatured 233U fuel cycle derive from its intrinsic isotopic barrier, its gamma radiation barrier, and its relatively low content of chemically separable fissile material in spent fuel: e The isotopic denaturing of the denatured 233U cycle would provide a significant technical barrier (although not an absolute one) that would decrease with time at a rate which is country-specific. Teéhno]ogical]y primitive countries will find it an imposing barrier relative to other routes. Countries that have the technological expertise to develop isotope separation capabilities will have the technology required to circumvent this barrier; however, they will also have the option of utilizing possible indigeneous natural uranium or low enriched 235y fue] as alternate feed materials. e The denatured 233U cycle imposes a significant radiation barrier due to the 232y daughter products in the fresh fuel as an inherent property of the cycle. Such a radiation field increases the effort requ1red to obtain weapons -usable material from fresh denatured reactor fuel. e MWhile the ambunt of plutonium discharged'in the denatured 233 fuel cycle is significantly less than in either the Pu/U cycle or the LEU cycle, the presence of plutonium in the cycle (even though it is in the spent fuel) does represent a proliferation concern. Conversely, it also represents a resourcé potentially useful in a symbjotic power system employing denatured fuel. The'concept_of a ~ safeguarded energy centek provides a means of addressing this duality in that the fissile plutonlum can be burned 1n ‘the center to produce ‘a proliferation- resistant fuel. In summary, the denatured 233U fuel cycle offers a technical contribution to pro- liferation resistance. However, the fuel cycle must be supplemented with political and institutional arrangements also designed to_discourage preliferation. 7-10 7.2. IMPACT OF DENATURED 233U FUEL ON REACTOR PERFORMANCE AND SELECTION COMPARISON WITH OTHER FUEL CYCLES T. J. Burns Oak Rldge National Laboratory The discussion in Chapter 4 has shown that the impact of the denatured 233U fuel cycle on the performance of the various reaétors considered in this study is largely due to differences in the nuclear properties of 233U and 232Th relative to those of 239y (and 235U) and 238y, respectively. For thermal systems, 233U is a significantly better fuel than either 239Py or 235y, both in terms of energy production and in terms of the conversion ratio* that can be attained. For fast systems, however, the substitution of 233(-based fuels for 239Pu-based fuels results in a somewhat poorer reactor performance, particularly with respect to the breeding ratio.* In this section the performance of the various reactors operating on the denaiured 233y fuel cycle is compared with their per- formance on other fuel cycles. In addition, the dependence of the denatured 233y fuel cycle on auxiliary fuel cycles for an adequate supply of 233U is discussed. Because of this dependence, reactors fueled with denatured 233U must be operated in symbiosis with reactors that produce 233U. These latter reactors, referred to as transmuters, may be either thermal reactors or fast reactors. The particular reactors selected for operation as transmuters and those chosen to operate on denatured 233U fuel will depend on several factors, two of the most jmportant being the resource requirements of the individual reactors and the energy growth capability required of the symbiotic system. The influence of these various factors is pointed out in the discussion below. ' ' 7.2.1 Thermal Reactors In comparing the performance of thermal reactors operating on denatured 233U fuel with their performance on other fuels, it is useful to distinguish between two generic fuel cycle types: those that do not require concurrent reprocessing (that is, once-through systems) and those that do. Although the denatured 233U fuel cycle cannot itself be employed as a once-through system, the implementation of the MEU{235)/Th once-through cycle is a logical first step to the implementation of the denatured 233U cycle. Thus both once-through and recycle scenarios are considered here for thermal reactors. Once-Through Systems Two fuel cycles of interest to this study can be implemented without concurrent reprocess- ing capability: the LEU cycle and the MEU(235)/Th cycle. The LEU cycle is, of course, already used % : The conversion ratio and breeding ratio are both defined as the ratio of the rate at which fissile material is produced to the rate at which fissile material is destroyed at a specific point in time (for example, at the midpoint of the equilibrium cycle). The term conversion ratio is applied to those reactors for which this ratio is less than 1, which is usually the case for thermal reactors, while the term breeding ratio is applied to those reactors for which this ratio is greater than 1, which is usually the case for fast reactors (i.e., breeders). ~ o] -4 = ey b ' r—-r— r r= - S - =t st — r -t - I C routinely in LWRs and small-scale fabrication of MEU(235)/Th fuels for LWRs might be attain- able within 2 - 3 years. However, it is pointed out that the ohce-through cycia has two variants - throwaway and stowaway - and in certain systems (for example, the PWR, as noted below), the MEU(235)/Th cycle might be economic only from a stowaway standpoint - that is, only if a reprocessing capability is eventually envisioned. Table 7.2-1 summarizes the U, 0g and separative work requirements estimated for PWRs HWRs, HTGRs, PBRs, and SSCRs operating as once-through systems on both the LEU and the MEU(235)/Th cycles. Several interesting points are evident from these data. The LEU-HWR requires the smallest resource commitment (as well as the smallest SWU requirément). The conventional PHR requires a significantly greater resource commitment and larger SWU requirements for the MEU/Th once-through cycle than for the LEU once-through cycle and hence no incentive exists for the MEU/Th cycle on PWRs if on1y the throwaway option is considered, Significantly, however, both of the gas-cooled graphite-moderated reactors, the HTGR and the PBR, require smaller U303 commitments for the MEU/Th once-through cycle than for the LEU case. Moreover, for both of these reactors, the SWU requirements for the MEU/Th cycle are not significantly different from those for the LEU cyclé; in'fact. for the PBR, the MEU/Th cycle is slightly less demanding than the LEU cycle. These effects are pri- marily due to the high burnup design of both the HTGR and the PBR. At the higher burnup levels of the gas-cooled reactors, most of the 233y produced in the MEU/Th cycle is burned in situ and contributes significantly to both the power and the conversion ratio. It is also interesting to note that, while not considered in Table 7.2-1, the unique design of the - PBR would permit recycle of the fertile elements without intervening reprocessing and thus would further reduce both the ore and SWU requirements for the MEU/Th cycie. [Note: The data given in Table 7.2-1 for PWRs considers only current commertia]]y deployed designs. Studies now underway in the DOE-sponsored Nonproliferation Alternative Systems Assessment Program (NASAP) indicate that LWR modifications to reduce urdnium requirements are feasible. Similarly, much of the other reactor data are subject to design refinement and uncertain- ties, as well as to future optimization for specific roles.] Tab]é 7.2-1. 30-Year Uranium and Separative Work Requirgments for Once-Through LEU and MEU(235)/Th Fuel Cycles® Uranium Requirement Separative Work Requirement (ST U304/6Ne) ~ (MT" SWU/GWe) Reactor LEU MEU/Th LEU MEU/Th PR 5989 8360 3555 7595 HWR 3563 8281 666 7521 HTGR 4860 4515 3781 4143 PBR 4289 4184 3891 3663° SSCR 5320 7920 3010 7160 a&75%rcapacity.factor; no credit for end-of-1ife core inventories; 0.2% tails. ' ' ' | The data presented in this table are consistent with the data submitted by the U.S. to INFCE (International Nuclear Fuel Cycle Evaluation) for the cases in which corresponding reactors are considered. Does not include recycle of fertile elements without intervening re- processing. 7-12 If these once-through systems_are operating oa,the throwaway option, the fissile material discharged in their spent fuel elements is deemed unusable; in fact, no value is assigned to the spent fuel in once-through fuel cycle accounting. Thus, in this case the most resource-efficient once-through fuel cycle is the one that requires the lowest fissile charge per unit power. If, however, a capability for reprocessing the spent fuel is eventually envisioned (i.e., if the throwaway option becomes a stowaway option), then the quantity of fissile material in the spent fuel becomes an important consideration. Esti- mates of the amounts of the various fissile materials discharged by each reactor type operating on both the LEU cycle and the MEU(235)/Th cycle are given in Table 7.2-2. ~ Table 7.2.2. 30-Year Charge and D1scharge Quantities - for Once-Through Fuel Cycles® MT/GWe Fissile Dischargeb | 'Cumulative 235y Total Net Fissile Reactor .Charge 233y 235y Pu’ Fissile Consumption L .LEU Cycle _ PHR 24,72 - 6.45 5.22 11.67 13.05 HUR 17.53 - 1.77 5.49 7.26 10,37 HTGR 19.49 - 3.25 2.16 5.3 14.08 PBR 18.09 - 279 1.89 4,68 . 13.4 SSCR 22.25 - 5.46 5.88 11.34 10.91 MEU(235)/Th Cycle PKR 33.83 7.80 11.52 2,13 21,45 12.38 HWR 32.63 14,28 10,08 0.75 25.11 7.52 HTGR 17.99 2,31 1.35 0.69 4,35 13.64 PBR 16.55 2.73 1.17 0.42 4.32 12.23 SSCR - - - - - - At 75% capacity factor. Estimated from equi]1br1um cycle. For the PWR and HNR, the use of the MEU/Th fuel cycle rather than the LEU fuel cycle results in a significant increase in the amount of fissile material contained in the spent fuel. It should be noted, however, that this increase is primarily the result of higher feed requirements (i.e., 235U commitment). In contrast, converting from the LEU cycle to the MEU/Th cycle does not materially affect the net consumption of the gas- - cooled HTGR and PBR (although it dramatically affects the types of fissile material pre- sent in their spent fuel). The relatively low values for the discharge quantities for the gas-cooled reactors is the result of two effects: a lower initial loading; and a design that is apparently based on higher burnup, which in turn reduces the amount of fissile material discharged. Finally, it is to be remembered that the resources represented by the spent fuel inventory are recoverable only when the spent fuel is reprocessed, whereas the U30g commitment is necessary throughout the operating lifetime of the reactor. Thus, in a sense, the spent fuel resource must be discounted in t1me to order to assess the best system from a resource utilization basis. ‘ - i - _;”s &ri — - = [ r— ra ra r— = r el e bm: = The isotopic composition of the spent fuel inventories is also of interest from a proliferation standpoint. For both the LEU and the MEU/Th once-through fuel cycles, the fissile uranium content of the spent fuel is denatured (diluted with 238U) and hence is protected by the inherent isotopic barrier. Thus the plutonium in the fuel would be the fissile material most subject to diversion. The use of the MEU/Th cycle in place of the LEU cycle sharply reduces the amount of plutonium produced (by 60-80%, depending on reactor type), and for both cycles the quantity of plutonium produced in the gas-cooled reactors is substantially less than that produced in the other reactor types. ' Recycle Systems If recyling of the fissile material in the thermal reactors is permitted, then 233U {and plutonium) produced in the MEU(235)/Th is recoverable on a schedule dictated by the production rate of the system. Table 7.2-3 gives estimates of the net lifetime consumption and production of various fissile materials for the MEU(235)/Th fuel cycle under the as- sumption that the capability for uranium recycle is available. (The 235U consumption tabul- ated does not reflect the 233U lost to the enrichment tailings.) For comparison purposes, the MEU{233)/Th fuel cycle estimates are also provided. The most striking aspect of Table 7.2-3 is the apparent 30% reduction of fissile consumption achieved with the 233U system, indicating the higher value of 233y as a thermal reactor fuel. In fact, the true extent of this effect is masked somewhat since a large fraction of the recycled fuel for the 235U makeup case is in fact 233U. It should also be noted that the MEU(233)/Th cycle generally results in a smaller net plutonium production, even though the degree of denaturing is less (i.e., the 238 fraction of uranium charged is higher). Table 7.2-3. Estimated Net 30-Year Fissile Consumption and Production for MEU/Th Cycles with Uranium Recycle? MT/GWe With 235U Loading and Makeup - With 233U Loading and Makeup - - . Fissile Pu - Fissile Pu Reactor 235U Consumption Production 233y Consumption Production PHR 2.8 2.85 91 189 HWR . 45 0.90 3.7 0.96 HTGR 10.4 1.3 1T 09 PBR - o - - . SSCR 8.7 . 2.56 5.9 - 2.44 At 75% capacity factor. As has been stated earlier, the consideration of an MEU/Th cycle that utilizes 233y makeup presumes the existence of a source of the requisite 233U. Although the 233U in the - spent fuel elements would be recovered, the amount would be inadequate to maintain the system and an exogenous source must be developed. One means for generating 233y is by using a Pu/Th-oxide-fueled thermal reactor. Tablé_7i2-4 summarizes some pertinent results for the various thermal reactors operating on the Pu/Th cycle. It should be noted that the HTGR case given in Table 7.2-4 is for a case in which the full core is refueled every 5 yr and is not optimized for 233U production. Thus, much of the 233U bred during this period is consumed in providing power, and the transmutation efficiency (tons of plutonium "transmuted" into tons of 233y) is significantly reduced relative to the PWR and SSCR. The transmutation efficiency of 0.40 for the PWR and SSCR is also rather poor, however, compared to the’1.20 value for a Pu/Th-fueled FBR (see Section 4.5). Production of 233U via plutonium-consuming " transmuters is more suited to fast reactors. On the other hand, it is recognized that Pu/Th- fueled thermal reactors could provide an interim source of 233y, Table 7.2-4, Net 30-Year Fissile Consamption and Production for Pu/Th Cycles MT/GWe o Fissile Pu 233y Transmutation Reactor Consumption Output Efficiency PWR ' 20.7 8.16 0.394 HWRD 19.84 11.76 © 0.593 " HTGR 16.5 3.03 0.184 PBR - - - SSCR 23.8 9.63 0.405 At 75% capacity factor, using equilibrium cycle bvalues. ' From data in Table 6.1-3. 7.2.2. Fast Reactors In this study fast reactors have been considered as possible candidates for two roles: as power reactors operating on denatured 233y fuel; and as transmuters burning plutonium to produce 233U. With LMFBRs used as the model, the denatured FBRs were analyzed for a range of 233U/U enrichments to parameterize the impact of the fuel on the reactor performance (see Section 4.5), and the transmutér FBRs were analyzed both for a Pu/238y core driving a ThO, blanket and for a Pu/Th system in which the thorium was included in both the core and the blanket. | The speC1fied 233y/U enrichment is a crucial parameter for the denatured fast reactors. Increasing the allowable enrichment permits more thorium to be used in the fuel material and hence allows the reactors to be more self-sufficient (i.e., reduces the P N | {j o1 rm—— - = - [ r r- - - — r = — V 7-15 required 233U makeup). Increasing the 233U enrichment also reduces the amount of fissile plutonium contained in the discharged fuel, which is obviously desirable from a safeguards viewpoint. However, increasing the 233U fraction also increases the vulnerability of the denatured fuel to isotopic enrichment, effectively forcing,a compromise between prolifera~ tion concerns regarding the fresh fuel versus proliferation concerns regarding the spent fuel. The lowest enrichment feasible for the denatured LMFBR systems analyzed iiec in the range of 11-14%. Such a system would utilize UD, as fuel and would require significant amounts of 233y as makeup since the plutonium it produced could not be recycled into it. The "breeding" ratio components of certain denatured LMFBRs as a function of 233U enrichment are shown in Table 7.2-5. The ratio of 233U produced to Puf produced is very sensitive to the specified degree of denaturing in the range of 12-20% 233U/U. This sug- gests that significant performance improvements may be possible (i.e., increased 233U produc- tion and decreased 239Py production) for re]at1ve1y small increases in the denatur1ng criteria. Of course, the overall "breeding" ratio of the denatured LMFBR is significantly degraded below that for the reference Pu/238U cycle (see Table 4.5-1 in Chapter 4). Table 7.2-5., Denatured LMFBR M1d-Equ111br1um Cycle Breeding Ratio Components* 233y 233y “Bpeeding" Pu "Breeding" Overall "Breeding" Enrichment Component Component Ratio 2% 0.41 0.71 L 1.12 20% 0.70 0.39 - 1.09 40% 0.90 0.15 1.05 100% 1.02 - 1.02 *Using values from Section 4.5-1. A more recent study [Prolifera~ tion Resistant Large Core Design Study (PRLCDS)] indicates that substantial improvements in the FBR performance is possible. Because of the superior breeding potent1a1 of a 239Pu fue]ed system relative to a 233y.fyeled system in a fast neutron spectrum, the fast reactor is ideally suited to the role of a pluton1um-fue]ed transmuter. Moreover, in contrast to the thermal transmuters, the fast reactors resu1t in a net overall f1ssi1e material gain.* Two types of FBR transmuters have been analyzed for the c]assjcai homogeneous FBR core configuration (a central_homogeneous_core surrounded by fertile blankets). In the first, the usual Pu/238U-fueled core was assumed with a ThO, radial blanket (also a ThO, axial ~blanket in one case). 1In the second type a Pu/Th core was assumed. Table 7.2-6 summarizes the net production data for typical fast transmuters of each type. The overall fissile gain/cycle with the Pu/238 core is s1gn1f1cant1y h1gher than that with the Pu/Th core, the result be1ng that the "breed1ng" ratio is not noticeably reduced from the breeding ratio for the reference Pu/238| cycle. . The production of 233y in the Pu/Th reactor is approximately a factor of 4 higher, but this is achieved as a result of “sacrificial” consumption of plutonium. Thus, these two reactor types reflect a tradeoff between 233U As noted in Chapter 4.5, significant uncertainties are associated with the fast neutron cross sections for 233y and Th. 7-16 and overall fissile production (i.e., potential growth rate). Table 7.2-6. Equilibrium Cycle Net Fissile Production for _ o Potential LMFBR Transmuters* ‘Net Fissile _ . Production: Reactor ' . (kg/GWe-yp) Core Axial Blanket Radial Blanket - Material Material Material Pu 233y Fissile (Pu/238y)0, v, ThO, +30 +157 +184 (Pu/Th)0, ThO, | Tho, -493 4583 +90 *Using values from Section 4.5-1 (~75% capacity factor). A more recent study [Proliferation Resistant Large Core Design Study (PRLCDS)] indi- cates that substantial improvements‘in the FBR performance is possible. . In addition to the systems utilizing the classical homogeneous core configuration, systems utilizing a heterogeneous core configuration (i.e., interspersed fissile and ferti]e‘regions) were examined as a possible means of improving the performance of fast reactors operating on alternate fuel cycles. The substitution of different coolants and fuel forms (i.e., carbides and metals versus oxides) were also considered. The net effect of these changes is to increase the fuel volume fraction in the reactor core, harden the spectrum, or, in some cases, both. The advanced fast reactor concepts show significant improvement regarding the breeding ratio (and doubling time) relative to the classical design when operating on alternate fuel cycles; however, the performance of the alternate fuel cycles is still degraded over that of the same reactor type operating on the Pu/238y cycle. 7.2.3. Symbiotic Reactor Systems As has been stated throughout this report, in considering denatured 233y reactor systems it is assumed that the denatured reactors will operate as dispersed power systems supported by fuel cycle services and reactor transmuters located in secure energy centers. When the system is in full operation no external source of fissile material is supplied; that is, the system is self-contained. Initially the resource base (i.e., natural uranium) can be used to provide a source of 233U for implementing the denatured 233U fuel cycle [via the MEU(235)/Th cycle]; however, a shift to plutonium-fueled transmuters will eventually be required. During this transition period, the system can be characterized by the rate at which the resource base is consumed (see Chapter 6). In order to compare the long-term potential of various reactor systems under the restrictions imposed by the denatured fuel cycle, two system parameters have been developed: (1) the energy support ratio, defined as the ratio of dispersed reactor power relative to the energy center (or centralized) power, and (2) the inherent growth potential of the system, Since both.the growth rate and the energy support ratio involve fissile mass flows, they are interre}ated. In order to unambig- uously determine both parameters, the inherent system growth rate is determined at the asymptotic value of the support ratio, a value which can be viewed as the "natural" operat- ing ratio of the system. a— \ E 7-17 Three generic types of symbiotic reactor systems can be envisioned by considering various combinations of thermal converters and fast breeders for the dispersed (D) and energy center (S) reactors: .thermal(D)/thermal(S), thermal(D)/fast(S), and fast(D)/fast(S). In order for the generating-capa¢ity of a system to increase with time without an external supply of fissile material, a net gain of fissile material (of some type) must occur. Thus, the growth potential of the thermal(D)/thermal(S) system is inherently negative; that is, the installed nuclear capacity must decay as a function ofi,time since the overall conversion ratio is less than 1. The thermal(D)/fast(S) system, however, does have the potential for growth since the net fissile gain of the fast component can be used to offset the fissile loss of the thermal reactors. However, a tradeoff between the support ratio [thermal(D)/ fast(S)] and the growth rate clearly exists for this system, since maximizing the support ratio will mean that net fissile-consuming reactors will constitute the major fraction of the system and the growth rate will be detrimentally affected. The fast{(D)/fast(S) system provides a great deal more flexibility in terms of the allowable energy support ratio and inherent growth rate. To illustrate the tradeoff between the growth potential and the support ratio, the "operating envelopes" shown in Fig. 7.2-1 have been generated using denatured PWR data from Section 4.1 and LMFBR transmuter data from Section 4.5.1. .Each envelope represents the locus of permissible symtiotic parameters (growth rate, support ratio) for the system considered,! i.e., the permissible combinations of growth rate and support ratio for each specific reactor combinations., At points A, B, and C on the curves, the transmuter used is, respectively, the classical (Pu/U)0, reference system with a U0, radial blanket, a (Pu/U)0, system with a ThO, radial blanket, and a (Pu/Th)0, system with a ThO, radial blanket. At each point along the curves connecting points A, B, and C, the transmuter is a combination of the two reactors defined by the end points of each curve segment (see key in upper right- hand corner). Points within the envelope correspond to-combinations of the three trans- muters in different proportions. ' The‘lower envelope in Fig. 7.2-1a (repeated in Fig. 7.2-1b) illustrates the tradeoff for the denatured PWRs and LMFBR transmuters, and the upper envelope depicts the fast/fast analogue in which the denatured PWR is replaced by an ~12% denatured LMFBR, As indicated, the fast{D)/fast(S) symbiotic system prbvides a higher growth rate for a given energy sup- port ratio, and, moreover, the growth rate is. éfiways positive. The upper envelope in Fig. 7.2-1b represents the correspond1ng case using 20% denatured LMFBRs. In all caées the fast reactor data uti]ized'were taken from Section 4.5.1; that is, homogeneous LMFBR cores were assumed, The use of a heterogeneous core for the transmuter reactor would have the effect of displacing the curves in F1g. 7.2-1 upwards and to the right. The ‘employment of an advanced converter (a high conversion ratio therma] reactor) would have a similar effect on the therma]/fast curve.- 7-18 o ORNL-DWG 78-19808 « . ; (2} A SEGMENT AB: TRANSMUTER = x*A + (1-x)+B g_ SEGMENT BC: TRANSMUTER = X.B + (1-X)-C SEGMENT AC: TRANSMUTER = x-A + (1-x).C REACTOR B: (Pu/U)Oz + ThO, R8 REACTOR C* f_PulTh)(]2 + Th02 RB -8.0 — 0sx<1l o +— LMFBR Transmuters é - + 12% Denatured LMFBRs B : o Z Gl £ | - o ¥ ¥ 1 1 1 " ho 2.0 25 - 3.0 3.5 1.0 4s X SE ENERGY SUPPORT RATIO o & o _ % ! «— LMFBR Transmuters = + Denatured PWRs @ = N «7 > i o o . ‘fl- REACTOR A: _ (Pu/U)Oz + UO2 RB c 8.0 ORNL-DWG 78-19809 (b) SEGMENT AB: TRANSMUTER = X w < 7 REACTOR A: (Pu/u)0, + VO, RB REACTOR B: (Pu/U)0, + ThO, RS o | REACTOR C: (Pu/Th)O, + ThO, RB o 1 Fig. 7.2-1. Operating Envelopes for Symbiotic Systems Utilizing LMFBR Transmuters. | r s o] 7.2.4. Conclusions Since optimization of the various reactors for the particular fuel cycle considered was beyond the scope of this study, the results presented above are subject to several uncer- tainties, Nevertheless, certain general conclusions on the impact of the various fuel ‘qyc1es on reactor performance are believed to be valid: e For once-through throwaway systems, the various systems studied are ranked in order of resource utilization as follows: the HWR on the LEU cycle; the HTGR and HTR-PBR on either the LEU cycle or on the MEU/Th Cyc1e; and the SSCR and PWR on the LEU cycle. On the MEU/Th cycle the SSCR and PWR require more uranium than they do on the LEU cycle and hence do not merit further consideration for once-through operation. r~ - o ( - e For once-through stowaway systems, in which the fissile material in the spent fuel is expected to be recovered at some future date, the relative ranking of the systems would depend on the ultimate destination of the fissile material. If future nuclear power systems are to be thermal recycle systems, then early emphasis should be placed on reactors and fuel cycles that have a high 233U discharge. If the future systems are to be fast recycle systems, then emphasis should be placed on reactors and fuel cycles that will provide a plutonium inventory. e For recycle systems utilizing thermal reactors, the preferred basic fissile material is 233U. However, implementation of a 233U fuel cycle will require an exogenous source of the fissile material; therefore, it is likely that the MEU(235)/Th cycle would be implemented first to initiate the produc- tion of 233U. Both the unburned 235U and the 233U would be recycled; thus the system would evolve towards the MEU(233)/Th cycle, which is the denatured 233U cycle as defined in this study. However, it is to be emphasized that these reac- tors will not produce enough 233| to sustain themselves and separate 233U production facilities must be operated. A Pu/Th-fueled thermal reactor has been considered r r ‘as a 233y production facility. . For recycle systems uti1izing fast reactors,'the preférred basic fissile - material is 23%u, Using 233U as the primary fissile material or placing “thorium in the core sharply reduces the bfeeding_pefformance-of fast reactors. However, fast reactors using plutonium fuel and thorium b]ahkets would be efficient 233U production facilities. r i f“ h .ki 1. 7-20 e The inherent symbiotic nature of the denatured 233U fuel cycle (i.e., dispersed reactors fueled with denatured 233U and supported by energy-center reactors fueled with Pu) mandates a tfiadeoff analysis of growth potential versus energy support ratio (ratio of power produced outside the energy center to the power produced inside the center), assuming no external source of fissile material. For thermal/thermal systems, the growth potential is negative. Fast/thermal systems would permit some of the net fissile gain (i.e., growth potential) of the fast reactors to be sacrificed for a higher energy support ratio.',Fast/fast systems would provide the highest growth potential. Factors other than those affecting reactor performance would also influence the choice of reactors for the system, as has been discussed in Chapters 5 and 6. Section 7.2. Reference T. J. Burns and J. R. White, "Preliminary Evaluation of Alternate Fuel Cycle Options Options Utilizing Fast Breeders," ORNL-5389 (1978). - r l[“'%‘z'lffi T - e v { - K r r— 7-21 7.3. PROSPECTS FOR IMPLEMENTATION AND COMMERCIALIZATION OF DENATURED 233y FUEL CYCLE J. C. Cleveland and T. J. Burns Qak Ridge National Laboratory Chapter 5 has discussed the reactors in which denatured 233 might be deployed, as well as the accompanying fuel recycle facility requirements, and has presented schedules of deployment that are based solely on the minimum time estimated to be required to solve technical problems. These schedules, which have been used in the nuclear power system evaluations presented in Chapter 6, were developed in discussions between Hanford Engi- neering Development Laboratory (HEDL), Argonne National Laboratory (ANL), Oak Ridge National Laboratory (ORNL), Combustion Engineering (CE), and the Department of Energy (DOE) specifically as a bounding case for assessing the maximum benefits that could be obtained by employing denatured 233 fuel. As a result, the schedules are not entirely consistent with those that have been developed subsequently in the Nonproliferation Alternative Systems Assessment Program (NASAP). While the introduction dates of the lead plants do not differ significantly, the NASAP scenarios predict a much slower deployment of commercial reactors. ' The reactor introduction dates and deployment schedules used in this study were based on the following assumptions: ~10 yr to develop/commercialize new fuel design w14 yr to develop/commercialize modified reactor design ~18 yr to develop/commercialize new advanced converter design 24 yr to develop/commercialize new breeder design The resulting introduction dates for the various reactors are as listed below, where the introduction date is defined as the date of startup of the first unit, reactor deployment thereafter being limited to a maximum introduction rate*' by biennium of 1, 2, 4,... reactors: 1969 - LWRs operating on LEU fuel 1987 - LWRs operating on "denatured 235U" fuel (i.e., MEU(235)/Th) 1991 - LWRs operating on denatured 233U, Pu/U, and Pu/Th fuels 1991 - SSCRs operating on LEU, denatured 233y, or Pu/Th fuels 1995 - HWRs operating on any qf several proposed fuels 1995 ~ HTGRs operéting on'any of several proposed fuels 2001 - FBRs operating on Pu/U, Pu/Th, or denatured 233U fuels Since the above introduction dates are those estimated to be the earliest possible dates that technical problems could be resolved, it is clear that they cannot be achieved without substantial initiatiVes and strong financial support from the U.S. Government. ‘i}he introduction rate of any new technology is likely to be less than the maximum rate noted above, since the construction market loss rate of an established technology is Timited to 10% per year and total nuclear capacity additions cannot exceed 15 GWeyyr. 233y systems are further constrained because the number of 233U-burning plants that can be operated is limited by the 233 production rate. 7-22 Even with government support, achieving the postulated schedules would be a difficult undertaking and would entail considerable risk since it would be impossible to fully demonstrate an alternate reactor concept before constructijon on the jnitial commercial size units has to begin. A minimum of six years would be required to construct a nuclear unit, and a minimum of three years would be required prior to construction for R&D and licensing approval. (It currently takes 10 to 12 yr to license and construct LWRs in the U.S.) At least two additional years of operation of the demonstration unit would be necessary to establish satisfactory reactor performance. Thus the earliest time a new reactor concept could be demonstrated is in the 1991-1995 period indicated, and that assumes that a commitment to proceed has been made by 1980. Because of design, licensing, and construction schedules, the first commercial units would have to be ordered well in advance of the operation of the initial demonstration reactor to achieve the buildup rates assumed in this study. In order to achieve such commitments prior to the first successful demonstration, government suppdrt would have to extend through the initial commercial units in addition to the lead plant. The new reactor cycle would also have to be perceived as economically advantageous to attract the postulated number of customers. Although several of these reactor/fuel options (e.g., Pu/Th LWRs, denatured advanced converters, etc.) are based on the use of recycled fissile material, it should be emphasized that commercial-scale reprocessing is not necessarily required on the same time scale as the introduction of the recycle fuel types because the demand for recycle fissile material may be quite modest during the initial introduction phase. In the analysis presented in Chapter 6, many of the new fuel types are, in fact, introduced before the associated fuel reprocessing is fully developed, it being assumed that pilot or prototype-plant scale reprocessing would be adequate to support the initial phase of deployment of fuel recycle. Hence, although commercial reprocessing of 233y-containing fuels is not projected until around the turn of the century, limited introduction of denatured 233U fuel is permitted ©as ear]y‘as 1991. A further argument is that commercial-scale reprocessing for the alternate fuels would not be feasible until the backlog of spent fuel required for plant startup had accumulated and the number of reactors utilizing recycled fuel could assure continued operation of commercial-scale facilities. On the other hand, for 233U-containing spent fuel elements to be available even for pilot-plant processing, it is essential that early irradiation of thorium in reactors be implemented. In Section 7.3.1 a possible procedure for implementing and eventually commercializing the denatured 233U cycle is discussed. Included is a scenario which would provide for the early introduction of thorium fuel into current light-water reactors and allow an orderly progression to the utilization of denatured 233y fuel in breeders. The major considera- tions in commercializing these various reactors operating on alternate fuels, and in particular on denatured 233U fuel, are summarized in Section 7.3.2. % . o o ( T - r- r— o C 7.3.1. Possible Procedure for Implementing and Commercializing the Denatured Fuel Cycle On the basis of the above assumptions, and the discussion in Section 5.1, it is ob- vious that the only reactors that could operate on denatured 233U fuel in the near term (by 1991) would be LWRs. Two possibilities exist for producing 233U for LWRs prior to the introduction of commercial fuel reprocessing. One involves the use of "denatured 235y fuel (i.e., MEU(235)/Th) in LWRs, thereby initiating the production of 233U, However, this scheme suffers from very high fissile inventory requirements associated with full thorium loadings in LWRs (see Section 4.1). A second option involves the use of partial thorium loadings in LWRs. In this option ThO, is introduced in certain lattice locations and/or MEU(235)/Th fuel is used in only a fraction of the fuel rods, the remaining fuel rods being conventional LEU fuel rods. This scheme significantly reduces the fissile inventory penalty associated with full thorium loadings in LWRs and - for BWRs may offer operational benefits as well (see Section 4.1). Also, the partial thorium loadings would allow experience to be gained on the performance of thorium-based fuels while generating significant quantities of 233U. Either of the above options for producing 233y will probably require some form of government incentive since the U30s and separative work requirements (and associated costs) will increase with the amount of Th utilized in the once-through throwaway/stowaway modes in LWRs. Although a reproceésing capability would be required to recover the bred 233U from thorium fuels, such a capability would not be required for the qualification and demonstration of thorium-based fuel, which initially would employ 235U rather than 233y, As has been pointed out above, the operation of LWRs with MEU(235)/Th or with partial thorium loadings could be accomplished during the next decade while the development and demonstration of the needed fuel cycle facilities for the implementation of the denatured 233) cycle are pursued. Initially the spent fuel could be stored in repositories in secure fuel storage centers which would represent a growing stockpile of 233y and plutonium. Additional fuel cycle service facilities, such as isotopic separation, reprocessing, fuel refabrication and possibly waste isolation, could be introduced into these centers as the need develops. As pointed out above, these could inftially be pilot-plant-scale facilities followed by larger prototypes and then commercial-scale plants. It has been estimated (in Section 5.2) that commercialization of a new reprocessing technology would require 12 to 20 yr and the commercialization of a new refabrication technology would require 8 to 15 yr. With'the depioyment of the pi]ot-sca1eikeprocessing and refabrication facilities, recovery of Pu and U from spent fuel and the subsequent refabrication of Pu/Th and denatured 233(/Th fuels could be demonstrated within the center. Pu/Th LWRs* could then *That is, thermal transmuters of an LWR design (see Section 4.0). As used in this report, a transmuter is a reactor (thermal or fast) which burns one fuel and produces another (specifically, a reactor that burns Pu to produce 233U from Th). 7-24 introduced within the centers to provide an additional means for‘233U'production, as well as additional power production. Concurrently, 233U (and unburned 235U) recovered from MEU(235)/Th or from partial thorium loadings could be utilized in denatured 233U fueled LWRs introduced at dispersed sites. Later, 233y recovered from the Pu/Th fueled LWRs could also be utilized to fuel dispersed reactors. At this point the first phase of a nuclear power system that includes reactors operating both in energy centers and at dis- persed locations outside the centers would be in effect. PUREX REPROCESSING ,('—-‘\ \ we-Ltev ) / \\ - -1 THOREX REPROCESSING | - LWR/Pu/Th & INITIAL PHASE PUREX REPROCESSING Fig. 7.3-1. Evolving Energy Center. THOREX REPROCESSING FBR Pu/Th b. INTERMEDIATE PHASE PUREX REPROCESSING ORNL-DWG M-21118 DENATURED LWR DENATURED FUEL DENATURED LWR ADVANCED CONVERTER DENATURED FUEL DENATURED FBR DENATURED THOREX REPROCESSING FBR Pu/Th t. FINAL PHASE FUEL Three Phases for an During this phase, which is repfesented in Fig. 7.3-1a, the research and development that will be required to deploy Pu-fueled FBR transmuters with thorium blankets in the energy centers could be pursued, With these advance preparations having been made, by the time conventional LEU fueling in LWRs begins to phase out (due to increasing depletion of an eco- nomical resource base), the power system would evolve into a fast/thermal combination employihg FBR transmuters and 233U-fueled converters, which by then might include denatured LWRs and advanced converters (SSCRs, HTGRs, or HWRs), depending on the reactor(s) selected for development (see Fig. 7.3-1b). Such a system could proVide adequate capacity expansion for modest energy demand growth; however, if .the energy demand is such that the fast/thermal system is inadequate, an all-fast system including denatured FBRs could be substituted as shown in Fig. 7.3-lc. The necessity of the third phase of the energy center development is uncertain at this time, reflecting as it does assumptions concerning the supply of economically recoverable U305 and energy demand. It is noted that this proposed scheme for imp]ementing the denatured fuel cycle and instituting the energy center concept relies heavily on two strong technical bases: currently employed LWR technology, and the research and development already expended on LMFBRs, which includes the Purex and, to a lesser extent, the Thorex reprocessing technologies. While alternative fuel cycle technologies or other types of reactors will be involved if they can'be demonstrated to have resource or economic advahtages, the LWR- LMFBR scenario has been selected as representative of the type of activity that would be required. 7-25 7.3.2. Considerations in Commercializing Reactors Operating on Alternate Fuels Although the introduction dates cited above for commercial operation of the various reactors on alternate fuels are considered to be attainable, they can be realized only if the first steps toward commercialization are initiated in the near future under strong and sustained government support. Currently, there is little economic: incentive for the private sector to proceed with such development alone. For example, while recent evaluationsl:? of LWRs have indicated the feasibility of using thorium-based fuels with current core and lattice designs, either as reload fuels for reactors already in operation or as both initial and reload fuels for future LWRs, the resource-savings benefit of such fuels relative to once-through LEU fuel cannot be realized in the absence of fuel repro- cessing and refabrication services. Moreover, the introduction of thorium into the core will require high initial uranium loadings, so that the fuel costs for the core would increase. Obviously, the lack of strong evidence that fuel recycle services would be available as soon as they were needed would discourage a transition to thorium-based fuels. Alternatively, such services could not be expected to be available commercially until utilization of thorium has been_established and a market for these services exists. Thus commercialization of the denatured fuel cycle in LWRs, especially within the time frame postu]ated in this study, is unlikely unless major government incentives are provided. The government incentives could be in-the form of guarantees for investment in the fuel cycle services and/or subsidies for the costs associated with the additional U304 and separative work required for thorium-based fuels or for partial thorium loadings on the rro - oo T . o [ T once-through cycle. This would also encourage the development of the fuel cycle services by establishing widespread use of thorium-based fuels. The commercial introduction of the required new LWR fuel cycle services could probably be accomplished by allowing a 7-yr lead time for construction of demonstration reprocessing and refabrication plants and an additional 7 yr to construct commercial-size plants. In the meantime, fabrication of MEU(235)/Th fuel or fuel designs invo1ving partial thorium loadings for LWRs could probably be accomplished with existing LEU facilities within 2 to 3 yr (Ref. 3) with an additiona] 5 to 7 yr:required for fue]-qua]ification and/or—demonstration. The R&D costs for demonstrating denatured uranium fuel in commercial reactors would be borne by the U government. The .commercial 1ntrdduction in’the U. S. of the advanced converter Concepts {SSCRs, HTGRs, and HWRs) would be more difficult today than was the past commerc1a1 introduction of the LWR. Although the 1ntroduct10n in 1958 of the first LWR, the Shlpplngport reactor, - did involve government support, a re]at1ve1y small investment was requ1red due to its size (~68 MWe). The largest base-load power plants were about 300 MWe when LWRs initially pene- trated the commercial market. Also, during the initial years of deplpyment'of nuclear power, delays due to licensing procedures were considerably shorter, allowing plants to be construc- ted and brought on-line more rapidly than the current 10- to 12-yr lead time. The longer time causes much larger interest payments and much greater risk of Ticensing difficulties. C i | | I i i 7-26 Prior to commercial introduction, a demonstration phase of a new advanced converter concept will be required, and, as has been pointed out in Chapter 5.1, it is assumed here that the demonstration will be on the reactor's reference cycie, which except for the HTGR, does not involve thorium. Utilities are unwilling to risk the large investment for | commercial-size plants of 1000 MWe to 1300 MWe on untried concepts. With the large investments necessary for demonstration units, significant government support would be required: i.e., a demonstration program involving government construction of the initial. unit with government financial support of the first commercial-size plant (1000 MWe to 1300 MWe). For commercial sales to occur, a vendor would have to market it and make the necessary investment to establish the manufacturing infrastructure. The SSCR is expected to draw heavily on existing LWR technology, and it may even be feasible to operate a conventional PWR in the spectra]-shift-cohtro] mode by addition of certain equipment. The feasibility of spectral-shift-control has already been demonstrated | in the Belgian VULCAIN experiment (see Section 4.2). While the possibility of retrofitting existing large PWRs to the SSC mode exists, for reactors going into operation after the late-1980s, designing PWRs to accept SSC control at some later date is a more likely possibility. A major_impedimedt-to commercial introduction of the SSCR in the U.S. is likely to be the supply of D,0 and government incentive would probably also be required in this area, as it will be for the deployment of the CANDU reactor (see below). The technology for HTGRs is already well under way, with a prototype reactor currently undergoing startup testing at Fort St. Vrain. Prior to commercial deployment, however, successful operation of a demonstration HTGR in the 1000-MWe to 1300-MWe range would be required. Initially, HTGRs could operate on the stowaway MEU(235)/Th or LEU cycle. Again, commercial-scale reprocessing and refabrication facilities would not be expected until a demonstrated market for such services is present. The technology for HWRs is also well advanced, with the CANDU reactors fueled with natural uranium already commercialized in Canada. It would be necessary, however, to demonstrate that the CANDU with appropriate modifications for slightly enriched fuel could be licensed in the U.S. and produce power at an atceptable cost. Commercialization'of_ the CANDU in the U.S. would probably require government action in three areas: 1. Transfer of technology from Canada to take advantage of CANDU reactor development and demonstrated performance. Alternatively, a demonstration unit designed to U.S. Tlicensing standards would be required. 2. Government financial support of a large (1000-MWe to 1300-MWe) CANDU in the U.S. 3. Development of D,0 production facilities in the U.S, on a larger scale than currently exists. ' | CANDUs operating on thorium-based fuels could possibly be introduced simultaneously with the deployment in the U.S. of the CANDU reactor concept itself. Assuming Canadian participation, thorium-based fuel could be demonstrated in Canadian reactors prior to the operation of a CANDU reactor in the U.S. Furthermore, if by then the LWR thorium fuel T O o e Tt i | 71-27 cycle services of reprocessing and refabrication had been commercially developed, the extension of these services to CANDU reactors could be built on the existing LWR facility base. Otherwise, the commercial introduction of these services could not be expected until some time after it becomes clear that CANDU reactors will be commercially deployed in the U.S. with thorium fuel, thereby indicating the existence of a market for associated fuel cycle services. The introduction dates postulated for the alternate fuel cycle CANDUs assume that requisite fuel cycle services have already been developed for thorium- fueled LWRs. As pointed out in Section 5.1, no attempt has been made here to consider the com- mercialization prospects of FBRs since the INFCE program (International Nuclear Fuel Cycle Eva]uatibn) is currently studying the role of FBRs in nuclear power scenarios and their results should be available in the near future. In summary, it is apparent that significant barriers exist for the private sector either to convert LWRs to thorium-based fuels or to develop advanced reactor concepts. While U30q is still re]ative]y inexpensive, the economics of alternate reactor and fuel cycle concepts at best show marginal savings relative to the LWR and consequently their development and deployment would have to be heavily subsidized by the government. In the longer term, as the price of uranium increases due to depletion of lower-cost uranium deposits, these alternate concepts could achieve superior economic performance compared to the LWR. The most optimistic introduction dates for advanced converters result in a relatively small installed capacity by the year 2000, and, as shown in Chapter 6, the impact of advanced converters on the cumulative U;0g consumption by the year 2000 would be small. However, deployment of alternate reactor concepts in the time from 1995-2000 could have significant impact on resource use in the period 2000-2025. Except for HTGRs, none of the alternate reactor concepts that promise improved resource utilization has undergone licensing review by the government. Due to these factors, conversion to the denatured fuel cycle and/or introduction of alternate reactor concepts on a time scale which can dissuade international tendencies toward conventional plutonium recycle will require very significant government involvement and financial incentives in the near future. 7.3.3. _Conclusions From the above discussion the folloWing conc1usiohs can be summarized: ¢ The production of 233U for the denatured 233U fuel cycle could be initiated by introducing Th into the LWRs currently operating on the once-through cycle. However, there is an economic disincentive within the private sector to convert LWRs to thorium-based fuels because of the increased costs associated with the higher U305 and separative work requirements. Thus commercialization of the denatured fuel cycle is not plausible unless government incentives are provided. Initial production of 233U 1. 7-28 for later recycle could be initiated by the mid-1980's if such incentives were forthcoming. Recycle of 233y on a commercial scale is not plausible prior to the year 2000, however. ¢ The introduction of advanced reactor concepts that would provide significant resource savings beyond the year 2000 will require very large government support for R&D, for demonstration facilities, and for lead commercial plants. If a rapid deployment schedule were required, additional resources would have to be committed to cover the risks of early commercial plants. e Fuel service/energy centers whoseAultimate_purpose is to utilize plutonium both for energy production and for 233y production would progress through various phases. Initially these centers would be‘fuel_storage facilities. With the introduction of reprocessing and retabrication in the center, LWRs located at dispersed sites would be fueled with denatured 233U. Concurrently Pu-fueled thermal transmuters would be deployed within the center. Ultimately,'tokmeet long-term energy demands, Pu-fueled fast transmuters would be introduced within the centers. e It is desirable that a fuel recycle R&D program be initiated for denatured fuels at the same time a decision is made to fabricate thorium-containing fuel for large-scale irradiation in existing LWRs, Pilot-scale recycle facilities could be required within seven years after the initiation of a thorium irradiation program. Section 7.3 References N L. Shap1ro, J. R. Rec, R. A. Matzie (Combustion Engineering), "Assessment of Fuel Cycles in Pressurized Nater Reactors," EPRI-NP-359 (February. 1977). "Assessment of Utilization of Thorium in BWRs," ORNL/SUB- 4380/5 (NEDG 24073), prepared by General Electric Company (January 1978) "The Economics and Utilization of Thorium in Nuclear Reactors," ORNL-TM-6331 (also Technical Annexes 1 and 2, ORNL-TM- 6332) prepared by Resource Planning Associates, Inc. (May 1978). sty v A 1{;' - r— € 7-29 7.4. ADEQUACY OF NUCLEAR POWER SYSTEMS UTILIZING DENATURED 233y FUEL FOR MEETING ELECTRICAL POWER DEMANDS M. R. Shay, D. R. Haffner, W. E. Black, T. M. Helm, R. W. Hardie, and R. P. Omberg Hanford Engineering Development Laboratory An important measure for evaluating a nuclear power system is whether it can meet projected power demands with the uranium resources estimated to be available at an accept- able cost. This section summarizes the results of analyses performed in this study to . determine whether various nuclear power systems utilizing denatured 233U fuel could meet a projected power demand of 350 GWe installed capacity by the year 2000 and a net increase of 15 GWe/year through the year 2049, the total capacity in the year 2050 being 1100 GWe. r supply as a function of cost. r— r— ORNL-DWG 78-21747 O T T T T T T T T 240 — r— HIGH-COST U30g 220 — SUPPLY r._—., ar aan ) 200 — INTERMEDIATE-COST U304 {180 SUPPLY 160 [ o MARGINAL €OST ( §/pound ) e ornny o o - o o ¢+ 2 3 a4 5 6 7 ‘U30q QUANTITY (0% tons) and Intermediate-Cost U3O8 Supply Curves. L LJ Fig. 7.4-1. Marginal Costs for High- The analyses were based on a uranium supply model shown in Fig. 7.4-1 and in Table B-7 (Appendix B), which provides both conservative and optimistic predictions of the uranium The power systems analyzed are described in detail in Chapter 6. They are comprised of LEU-LWRs operating in conjunction with LWRs on other fuel cycles or in conjunction with one of the three types of advanced converters (SSCR, HWR, or HTGR) considered in the study. In some cases, FBRs are included in the system. Since the maintenance of proliferation- resistant power systems was one of the primary concerns, the concept of a secure energy center supporting dispersed reactors was used, with the fuel utilized in the dis- persed reactors restricted to LEU (or SEU) and denatured fuels. A reactor operating on the denatured 233U fuel cycle is not self- sustaining, however, and therefore it requires an exogenous source of 233U, In the power systems studied, the 233y is ~ provided by MEU/Th-fueled thermal reactors _or plutonium-fueled thermal and/or fast transmuters. These reactors, of course, also contribute to the power generation, Because the transmuters have plutonium cores, ~ however, they must be located within the secure "Energy centers. (Note: With this restriction - the "energy support ratio" of a nuclear . system becomes a second important measure of evaluation, as is discussed in Section 7.2.3. The energy”support ratios for the systems described here are given in Appendix C, along with other detailed results from the analyses.) 7-30 A nuclear power systems evaluation such as the one performed in this study requires three basic components. First, the various nuclear power systems to be analyzed must be identified. Second, there must be an analytical model capable of modeling each system in sufficient detail that differences betwéen'the systems can be accurately calculated. And finally, a data base that contains both reactor performance data and economic data must be developed. Sections 7.4.1 and 7.4.2 below give brief descriptions of the model and data base as they were applied to this evaluation. The results of the analyses for specified nuclear power systems are then summarized in Sections 7.4.3, 7.4.4 and 7.4.5, with the detailed results presented in Appendix C. | | 7.4.1. The Analytical Method Two fundamental aspects of the model used in the analyses relate to the nuclear energy demand and the U30g supply, both of which have been specified above. The nuclear energy demand assumed in the model is consistent with the current construction plans of utilities through the 1980's. As more nuclear units were required, with the supply of low-cost U30g progressively depleted, it was assumed that more expensive lower-grade uranium resources would be mined. This was modeled by assuming that the long-run marginal cost of U30g was an increasing function of the cumulative amount mined. For a particular nuclear policy option, the plant construction pattern was therefore governed by economics and/or uranium utilization. ‘ Two different optimizing patterns were used in the study. In the first runs economic competition between nuclear fuels and coal was assumed, and the plants were selected to minimize the levelized cost of power over time. These runs, which are pre- sented in Appehdix D, indicated that nuclear power did not compete well at U30g prices above $160/1b for the assumptions used in this study. Thus for the runs of all-nuclear power systems, described in Chapter 6 and summarized here, an attempt was made to satisfy the demand for nuclear power with the U305 available at a price less than $160/1b U30g. Other considerations also affected the selection of the nuclear power plants to be constructed. For example, a reactor that required Pu or 233U could not be constructed unless the projected supply of fissile material was sufficient throughout the reactor's lifetime. In addition, a nuclear plant design that differed from established technology could be introduced only at a limited rate. Furthermore, once the manufacturing capability to produce a particular reactor type was well established, the maximum rate at which that reactor type could lose its share of the new construction market was limited to a speci- fied rate. ' Both the total power cost of each nuclear policy option and the total power cost of each reactor type available in each option were calculated. For each reactor type, the total power cost was calculated for four components -- capital, operation and maintenance, J g 5 I ¥y v - ey ) i ! ’ S ! r ! f £ ¥ i r— r— - e o T D oo T T =} ' - 7-31 taxes, and fuel cycle. The fuel cycle costs were, in turn, divided into seven components -- 233, yranium, thorium, enrichment, plutonium, fabrication, and reprocessing. It is to be noted that the power systems calculated were all assumed to be U.S. based, the input data all being of U.S. origin. With appropriate input modifications, however, the model could be used for other scenarios. For example, it could be used to analyze the potential for the deployment of transmuters both to produce power in secure states and to produce 233 for export to states wishing to base their own power systems on thermal reactors without national reprocessing. ~7.4.2. Data Base The data required by the model for each reactor type include power level, annual isotopic charge and discharge, annual fabrication requirements, introduction dates, etc. These data are presented in Tables 6.1-2 and 6.1-3 in Chapter 6. It is to be pointed out, however, that the data are for reactors of essentially conventional designs, and that the U0g requirements for the various reactor types could be reduced through design optimiza- tion. {(Note: The effect of optimizing LWRs has been considered in a separate analysis and is discussed in Section 7.4.3 below.). The major parameters in the economic data base used for this study are capital costs, uranium costs, fabrication costs, spent fuel disposal costs, reprocessing costs, and money costs. The entire data base, which was developed in a joint effort involving government and industry representatives, is presented in Appendix B. 7.4.3. Results for Price-Limited Uranium Supplies As noted above, the denatured nuclear power systems utilized various combinations of thermal converters and fast reactors. These in turn were examined under six fuel cycle options, which are summarized in Table 7.4-1 (Options 4-8). In addition, the same reactor types were examined under three reference fuel cycle options -- a throwaway/stowaway option (Option 1) and tWo_plutonium—uranium options (Optioné 2 and 3). Four cases were considered under each option, each case being distinguished by the type of converter being emphasized -- LWRs, SSCRs, HWRs, or HTGRs. Thus a total of 36 different nuclear power systems were analyzed. : _ The maximum nuclear capacity and the year in which the maximum occurs for each nuclear system Studied is shown in Table 7.4-2 for the two uranium supply asSumptions (see Fig. 7.4-1). As stated earlier, with the intermediate-cost supply it was assumed that 6 million ST of U30g could be recovered at costs less than $160/1b, while with the high-cost supply it was assumed that 3 million ST of Uj0g would be available. 7-32 Table 7.4-1. Description of Fuel Cycle Options* Throwaway/Stowaway Option (see Fig. 6.1-1): % Option 1. LEU converters oh once-through cycle. Plutonium-Uranium Options (see Fig. 6.1-2): Option 2. Pu/U recycle option; LEU converters outs1de center, Pu/U converters inside center; HTGRs inside center operate on 235U/Th 233U/Th, and Pu/Th. : Option 3. Pu/U recycle option; LEU converters outside center; Pu/U converters and breeders inside center; HTGRs inside center operate on 235y/Th, 233y/Th, and Pu/U. Denatured Uranium QOptions Using Converters Only (see Fig. 6.1-3): Option 4. Plutonium throwaway option; LEU and denatured 235U and 233U converters outside center; no reactors inside center; U only recycled. % , . Option 5U. Plutonium minimization option; LEU and denatured 235U and 233U con- i verters outside center; Pu/Th converters inside center; U and Pu recycled. Option 5T. Same as 5U without denatured 235U converters. Denatured Uranium Options Using Both Converters and Breeders (see Fig. 6.1-4): Option 6. Light transmutation option; LEU and denatured 235U and 233U conver- ters outside center; Pu/Th converters and Pu-U/Th breeders 1ns1de center. Option 7. Light transmutation option with denatured breeder; LEU converters, denatured 235U converters, and denatured 233U converters and breeders outside center; Pu/Th converters and Pu-U/Th breeders inside center. Option 8. Heavy transmutation option; same as Option 7 except inside breeder is a Pu-Th/Th breeder. *Four cases considered under each option, identified by letters L; S, H, and G to denote type of converter empioyed in addition to LEU-LWRs (L = LWR, S = SSCR, H = HWR, G = HTGR). The effect of varying the fuel cycle system can be seen by reading across Table 7.4-2 and the effect of changing the converter reactor option can be deduced by reading down a column. An installed nuclear canacity_of 1100 GWe in year 2050 indicates that the projected energy demand is fully met by tne reactors in a given nuclear fuel cyc1e system. i et o I' 1 P ' rt 7 T 1[; N . - w b N r— i - v e T ra—s‘. £ o - & . - C 7-33 Table 7.4-2. Maximum Nuclear Capacity of Varijous Nuclear Power Options and Year in Which Maximum Occurs (Note: A capacity of 1100 GWe in year 2049 meets demand.) Maximum Installed Nuclear Capacity (GWe)/Year maximum occurs Converter Reactor : _ ' _ Option 1 2 3 4 - BUY 5T 6 7 8 With High-Cost U30g Supply LWRs 433 611 1100 585 716 637 1100 1100 1087 (L) 2008 2021 2049 2019 2027 2021 2049 2049 2049 SSCRs 440 '561 1100 660 820 764 1100 1100 1084 (S) 2009 2023 2049 2023 2033 = 2029 2049 2049 2049 HWRs 444 630 1100 756 915 856 1100 1100 1100 (H) 2011 2081 2049 2031 2041 2035 2043 2049 2049 HTGRs 437 818 1100 545 671 638 1091 1100 958 (G) 2009 2033 2049 2019 2023 2021 2049 2049 2041 With Intermediate~Cost U30g Supply LWRs 729 968 1100 1002 - 1062 1012 1100 1100 1097 {L) 2027 2041 2049 2047 2049 2047 2049 2049 2049 SSCRs 763 1078 1100 1084 1100 1100 1100 1100 1100 (S) 2029 2049 2049 2049 2049 2049 2049 2049 2049 HWRs 852 1062 1100 1084 : 1100 1100 1100 1100 1100 (H) 2035 2049 2049 2049 © 2049 2049 2049 2049 2049 HTGRs 783 1100 . 1100 - 971 1065 996 - 1100 1100 1100 (G) 2031 - 2049 2049 2041 . 2049 - 2045 2049 2049 2049 ‘Non-FBR Systems, bptions 1, 2, 4, and 5 ; For the high-cost U0 supply case (3 million ST Us0g below $160/1b), it is evident that introducing advanéed(converters on the throwaway/stowaway fuel cycle (Option 1) has little effect on the maximum attainable nuclear capacity. This is directly due to the introduction dates assumed for the advanced converter reactors. By the time the converters 7-34 dominate the new capacity being built, a very significant fraction of the U30g supply has already been committed to the standard LWR. It follows that if the intermediate-cost U30q were used (6 million ST U30g below $160/1b), together with the same nuclear growth rate, the addition of an advanced converter would have a much larger impact. For example, in this case the system including HWRs has a maximum attainable installed nuclear capacity for the throwaway/ stowaway option that is approximately'17% greater than the installed capacity of the system comprised of LWRs alone, while for the high-cost supply case it is only 3% greater. In Option 2 converter reactors are operated on the LEU fuel cycle outside the energy center and Pu/U converters and 235U(HE)/Th, 23%/Th, and Pu/Th HTGRs are operated inside the center. As expeéted, the thermal recycle systems all support nuclear power growth with less U305 consumption than the once-through systems of Option 1, and, in general, the options including advanced converter reactors (SSCRs, HWRs, and HTGRs) provide for increased 'maximum installed capacity relative to the LWR option for both the high-cost and the intermediate-cost U30g supply assumptions. - The HTGR option (2G) provides for the greatest level of installed nuclear capacity for both Uj0g supplies. The resource efficiency of these scenarios is largely due to the fact that they include the nondenatured 23 3J/Th fuel ~cycle which is used_on]y by HTGRs in this study. Option 4 utilizes only denatured 235U and 233U fuels and LEU fuel, all outside the energy center, and none of the plutonium produced is recycled. Here it is interesting to observe that for both uranium supply assumptions the HWR converter option (4H) has installed capacity levels that are greater than or equal to those of any other converter reactor option, while the HTGR option (4G) has the lowest installed capacities. It appears that the HTGRs used in this study do not operate efficiently on denatured fuel cycles relative to the other converters available (see also Options 5UG and 5TG). This can be partially attributed to the fact that the reactors used in these evaluations were not optimized for the roles in which they were employed, and for the HTGR this has a greater impact than for the other reactor types. Option 5 uses denatured and LEU-fueled reactors outside the center and Pu/Th-fueled converters within the center. This option is divided into two suboptions: Option 5U, in which both denatured 235U and denatured 233 units are used; and Option 5T, in which the denatured 235U units are excluded.. In both cases, 233 is produced in the Pu/Th converters. In these cases the HWR options producé the greatest maximum installed nuclear capacity with the high-cost ore supply, and both the HWR options and SSCR options meet the power demand with the intermediate-cost ore supply. Again, the HTGRs do not appear to operate as | efficiently as the other converters for the reasons cited above. " In summary, non-FBR power systemé using denatured fue]_but discarding p]utonium' require about the same amount of U0g as thermal systems on the classical Pu/U cycle and offer potential nuclear growth rates that are roughly the same. If the plutonium is re- r—- . ; 4, ¢ b ) »i . - o T 1[;“'i . ? I =1 — . € T e oo ' i C » i i 7-35 cycied in Pu/Th converters, the systems have potentia].nucleaf'growth rates that exceed those of analogous reactors operating on the Pu/U fuel cycle. If the intermediate-cost U305 supply assumption proves to be correct, advanced converters in the recycle mode can satisfy the postulated nuclear energy demand through year 2050 at competitive costs. This analysis therefore indicates that, at least under optimistic resource conditions, advanced converters using denatured fuels can defer the need for commercial use of an "inexhaustible" energy source (such as FBRs) beyond the year 2050. FBR Systems, Options 3, 6, 7, and 8 Table 7.4-2 shows that almost all of the nuclear system'options using FBR fuel cycles (Options 3, 6, 7, and 8) are able to meet the projected nuclear energy demand without mining U30g costing more than $160/1b., The only exception is Option 8 for the case of the high-cost ore supply, and even this option, which includes the Pu-Th/Th breeder and the denatured 233U breeder, would satisfy the demand if slightly improved FBR reactor design parameters were used. Thus, as was expected, this analysis indicates that FBR-containing systems will potentially support much larger nuclear capacities than thermal recycle systems and/or will require less mining. The Th-containing FBR cycles supporting dispersed denatured converters perform as well as the analogous Pu/U cycles within the framework of this analysis. Of the Th-containing cycles, the FBR with a Pu/U core and Th blanket is particularly resource-efficient. 7.4.4. Results for Unconstrained Resource Availability The preceding results represent a somewhat artificial situation because of the $160/1b limitation on the U305 availability. That is, the failure to meet the projected power demand in many of the scenarios investigated is a direct result of the system's inability to utilize U305 costing more than $160/1b. In order to address the potential of the various fuel cycle/reactor options under the condition that the projected demand for nuclear power must be satisfied, the $160/1b constraint was removed. The cumulative quantity of U30g required to completely satisfy the demand for nuclear generating capacity was then estimated for each of the nuclear pbwer options; these results are presented in Table 7.4-3. The rate at which U304 is required to support the projected nuclear capacity represents an important additiona] constraint on a system. An overall maximum U304 production rate is difficult to specify because of the possibility of importing U;0g and because any prediction of the production of U308 from uncertain resources in the next century'is highly Speculative{l Recognizing this, and also reéognizing that the required U30g production rate is still an important variable, the maximum required Us0y production rates for each scenario were estimated and are tabulated in Table 7.4-4. As a point of reference, note that DOE has estimated that domestic mining and mi1ling could sustain a production rate of 60,000 ST of U30g per year in the 1990s by developing U305 reserves and potential resources at forward costs* of less than $30 per pound. * Forward costs do not include the capital costs of facilities or industry profits, which are included in the long run marginal costs used in this study. 7-36 Table 7.4-3. Cumulative U30g Consumption of Various Nuclear Policy Options Cumulative U30g Consumption (millions of tons) Converter Through year 2025/Through year 2049 Reactor Option 1 ' 2 3 4 5U 5T 6 7 8 With High-Cost U304 Supply LWRs 3.41 2.39 2.14 2.87 2.36 2.36 2.18 2.14 2.29 (L) 7.05 5.28 2.73 5.41 4.83 4.94 2.82 2.83 2.86 SSCRs 3.26 2.23 1.9 270 2.35 2.4 1.93 1.93 2.07 (s) 6.52 4.35 2.70 4.65 3.86 3.86 2.69 2.69 2.83 HWRs 3.10 2.72 2.29 2.50 2.16 2.14 2.25 2.2 2.29 (H) 5.58 4.64 2.70 4.36 3.27 3.77 2.61 2.55 2.87 HTGRs 3.23 2.19 1.97 2.58 2.32 2.3¢ 2.15 2.12 2.32 (6) 6.26 4.04 2,75 5.13 4.43 4.94 2.70 2.68 3.18 With Intermediate-Cost U30g Supply LWRs 3.41 2.39 2.28 2.87 2.36 2.36 2.37 2.37 2.37 (L) 7.05 5.23 4.40 5.41 4.91 4.94 4.38 4.38 4.48 SSCRs 3.26 2.23 2.20 2.70 2,14 2.14 2.14 2.14 2.14 (S) 6.52 4.35 4.14 4.65 3.86 3.86 3.86 3.86 3.86 HWRs 3.10 2.72 2.31 2.94 2.52 2,51 2.32 2.30 2.38 (H) 5.58 4.64 2.71 5.40 4.32 4.37 3.66 2.70 3.37 W HTGRs 3.23 2.32 2.30 2.58 3.32 2.34 2.23 2.23 2.26 (G) 6.26 4.23 4.22 5.13 4.43 4.9¢ 4.19 4.19 4.24 1 | i r— The results presented in Tables 7.4-3 and 7.4-4 indicate the relative resource efficiencies of the various nuclear power systems since the energy produced was held constant. It should be noted that although the U30g cost limitation of $160/1b was removed, the uranium requirements were estimated.for both the intermediate- and high-cost U305 supplies. Hence, the differences in the cumulative U30g requirements and annual U305 production rates for similar fuel cycle/reactor combinations are due to different reactor mixes associated with each uranium price structure. r— - maeven— ok T X _-r o T 1{; RN S ey [ r—-. o r r— = b 7-37 Table 7.4-4. Maximum U305 Requirements of Various Nuclear Policy Options Converter Maximum U0, Consumption (thousands of tons per year) Reactor , . Option 7 1 2 3 4 5U 5T 6 7 8 With High-Cost U305 Supply LWRs 183 120 60 11 115 115 62 60 68 SSCRs 160 115 52 83 83 83 50 50 55 HWRs 120 83 66 78 62 69 64 63 65 HTGRs 140 82 53 105 96 115 61 60 65 With Intermediate-Cost U304 Supply LWRs 183 120 92 111 117 115 86 86 92 SSCRs - 160 115 93 83 83 83 83 83 83 HWRs 120 - 83 66 110 89 - 90 66 66 66 HTGRs - 140 86 86 105 96 115 87 87 87 Satisfying the demand for 1100 GWe in year 2050 with the standard LWR once-through cycle (Option 1L) would require that about 183,000 ST U30g be produced in year 2049, with a cumulative consumption of 7.1 million ST through that date. Introducing advanced converters (Options 1S, 1H, and 16) would reduce both the cumulative Us04 consumption and the maximum prdductioh rate requirements on the once-thrbugh cycle — in the case of the HWR as Tow as 5.6 mitlion ST and 120,000 ST/yr, respectively. | Thermal recycle modes (Options 2, 4, 5U, and 5T) would reduce U404 consumption ‘through year 2049 to within the range of 3.3 to 5.4 million ST U30g, depending on the policy option chosen and to a lesser extent on the uranium cost level. The maximum U0 ~consumption would vary from 62,000 to 120,000 ST/yr. The resource consumption is sensi- tive to the uranium price level to,therextent that high-cost uranium favors the choice of efficient high-capitdl—cost-sy#tems such as the HWR, whereas lower-cost uranium favors continued use of LWRs even if other reactors are available. It should be noted that when plutonium is recycled in thermal power systems includ- ing denatured reactors_(ODtions 50 and 5T7) the total resource requirements (including Pu) 7-38 are generally less than those for-therma] systems in the Pu-U recycle mode (Option 2). Discarding Pu from the recycle of denatured thermal systems (Option 4) reduces the efficiency of the denatured cycle. ' The nuclear power sy%tems that include fast breeders (Options 3, 6, 7, and 8) have cumulative Us0g requirements through year 2049 within the range of 2.71 to 4.41 million ST Ui0g in the case of the intermediate-cost Ui0g supply and within 2.6 to 3.2 million ST ag in the case of the high-cost supply. The maximum U30g consumption varies from 66,000 to 93,000 ST/yr for the intermediate-cost supply and from 52,000 to 68,000 ST/yr for the high-cost supp?y. The breeder-containing options are able to.adjust the reactor mix effectively to reduce U{g consumption in the event U40g costs are high. The larger the fraction of breeders in the reactor mix, the lTower the Ui0g requirements. It should be noted that the U0z requirements for the systems containing breeders with Pu/U cores and Th blankets (Options 6 and 7) are similar to the U30g requirements for the system containing the classical Pu/U breeder (Option 3). The systems containing breeders with Pu/Th cores and Th blankets require somewhat more Udg on an integrated basis. ' The U40g requirements presented in Table 7.4-4 qualitatively support the ranking of cycles in the cost-constrained runs. Specificd]ly, the power systems operating on once- through cycles require 5.6 to 7.1 million ST U30g to satisfy the demand for nuclear power through 2050, the therma]-recyclé systems require 3.3 to 5.4 million ST U30g, and the breeder-containing systems require 2.6 to 4.4 million ST U30g. The systems including denatured 233} reactors require approximately the same cumulative amount of U30g as their Pu/U counterparts. The results presented in Table 7.4-5 also support these statements: the required production rates are highest for the once—througfi systems; they are reduced somewhat for the thermal recycle cases; and they are lowest for the breeder-containing scenarios. | f 7.4.5.‘ Systems Employing Improved LWRs and Enrichment Technology While not considered in the analysis summarized above, it is possible to optimize LWR designs to greatly enhance their utilization of U305 per unit energy produced. These optimized designs may result in reduced U305 requirements of up to 30% relative to more conventional LWR designs. The 30% improvement in LWR U305 requirements assumes no spent fuel reprocessing, the improvements be1ng the result of increased d1scharge exposure fue]s and/or reconfigured reactor cores. The effect of developing these LWR cores optimized for throwaway/stowaway operation was examined by assuming that the U305 utilization would be improved in séquential incre- ments U30g requirements equal to 90% of the standard LWR. It was also assumed that th1s 1mprovement would be retrofitted into ex1st1ng reactors. T Similarly, reactors starting up TNeither the down time required for retrofitting nor the associated costs were addressed in this analysis. r—= ek r= et e F L: | r ‘{b T € v t | r - 5 i r~ ! ' ' G ! ; [ _, ey ¥ ) ¥ £ 1 + | oy oty b N A » 1 ' £ T 7-39 between 1991 and 2001 were assumed to have Ui0g requirements equal to 80% of the standard LWR, with the improvements retrofitted to all existing reactors at that time. Finally, those plants beginning operation after 2001 were assumed to have U30g requirements equal to 70% of the standard LWR, again with the improvements retrofitted to existing plants. In addition, the effect of a lower enrichment tails assay was examined for both the standard and the optimized LWR designs. The standard enrichment tails schedule assumed that the assay fraction was a constant 0.0020. The reduced tails schedule began at 0.0020 but decreased to 0.0005 between 1980 and 2010 and remained constant thereafter. The latter tails schedule was assumed to represent a changeover to an improved enrichment technology. The effects of considering both the improved LWR design and the improved tails technology are summarized in Table 7.4-5. The results show that with tails improvements alone the U30g requirements may be reduced by 16% by year 2029. This reduced level of U305 consumption translates to an increase in the maximum installed capacity of approxi- mately 60 GWe for standard LWRs on the throwaway/stowaway fuel cycle. Table 7.45. Comparison of Ui;0g Utilization of Standard and Improved LWRs Operating on Throwaway/Stowaway Option With and Without Improved Tails ST U,0g /GMe Standard LWR Technology Improved LWR Technology Normal Improved Normal Improved Year Tails - Tails Tails Tails 1989 5236 4759 4649 4224 2009 6236 4508 4079 3560 2029 5236 4398 3923 3346 *Normal tails assume 0.2 w/o 235U in 238|; improved tails as- sumed 0.05 w/o 235 in 238y; 75% capacity factor. ' ~ With improved LWR technologies (no tails improvements) the U30g consumption levels could be reduced ~25% in year 2029. This translates to an increase of 100 GWe in the ~ maximum ihstal]ed capacity for optimized LWRs. 'If both reduced tails and advanced LWR technologies were used, the maximum achievable installed nuclear capacity would increase by about 144 GWe. - | It is important to place these results within the pefspective of the results re- ~ ported in Table 7.4-2. The maximum 1nstai]ed_nUc1ear capacities obtained with these improvements are comparable to those for standard LWRs operating on the classical Pu/238y recycle mode or on the denatured 233 cycle. Obviously, if both improved LWRs and Pu recycle were available, the nuclear capacity could be even greater. | 7-40 1.4.6_ Conclusions —-From the preceding discussion and the results presented in Chapter 6 and Appendix C, the following conclusions may be drawn concerning the reactor options, the fuel cycle options, and the U30g supply cases analyzed for this study. It should be emphasized that the conclusions are tentative and may be changed as a result of different demand growth projections or more accurate or improved reactor characterizations. o If nuclear power systems were limited to the once-through cycle, it would be necessary to utilize U30y sources at above $160/1b sometime between year 2009 and year 2035 in order to satisfy the projected nuclear power capacity demand. o If nuclear power systems were limited to the once-through cycle and to U30g ~ supplies below $160/1b, the U.S. nuclear power capacity would peak some time between 2009 and 2035. Nuclear power would fail to satisfy the projected nuclear demand during the 10-year period preceding the peak. If improved LWR designs'and improved tails stripping techniques were implemented, the peaks would occur 10 to 15 years later. | e If the high-cost U30g supply is assumed (3 million ST below $160/1b), all once-through systems, regardless of the converter type employed, result in approximately the same maximum installed nuclear capacity. For less-restrictive U0 supply assumptions, advanced converters have time to increase the total nuclear power supply on the once-through cycle. ¢ Thermal recycle systems have the capability of substantially reducing requirements for U30g or of increasing the maximum installed capacity over the capacity of the once-through cycle. The best thermal recycie systems can support over twice the max imum 1nstélled'capacity of the ohce-through cycle, and, under the intermediate- cost U30g supply assumption (6 million ST below $160/1b), they can fully support the assumed nuclear power growth through year 2050. ¢ The systems including breeders have the capability of substantially reducing the mining reqdirements and/or increasing the maximum installed capacity beyond thermal systems with recycle. This capability is needed to satisfy the nuclear capacity- demand through year 2050 under the high-cost uranium supply assumption (3 million ST below $160/1b). ¢ Thermal recycle systems including denatured 233U reactors have the capability of supporting more nuclear capacity than the thermal Pu/238U recycle systems. However, achieving this capability would usually require Pu utilization. ¢ From a resource utilization point of view, nuclear power systems utilizing denatured 2331 reactors can be started equally well with MEU(235)/Th or Pu/Th fuels, providing the eventual use of the plutonium generated in the MEU(235)/Th cycle is assumed. S 4 — — { = o - X"~ -1 | (“"l - | r— ) - €~ £ i £ 4 ¥ 3y 3 i, C + ' g 7-41 e Systems that use breeders (i.e., fast transmuters) to produce 233y for LWRs or advanced converters operating on denatured 233U fuel have a capability comparable to systems employing the classical Pu/U breeder cycle to satisfy the assumed demand through 2050 with the U30g resource base assumed in this study. Section 7.4. References John Klemenic, Director, and David Blanchfield, Mineral Economist, Supply Analysis Division, Grand Junction Office, DOE Uranium and Enrichment Division, in paper entitled "Production Capability and Supply," paper presented at Uranium Industry Seminar, October 26-27, 1977, Grand Junction, Colorado; proceedings published as GJ0-108(77). 7-42 7.5. TRADE-OFF ANALYSIS AND OVERALL STRATEGY CONSIDERATIONS T. J. Burns and I. Spiewak O0ak Ridge Nationa] Laboratory One of the principal concerns about c1v111an nuclear power centers on the possible d1vers1on of recycled fissile material to weapons fabrication, in particular, the diver- sion of plutonium. Depending on the degree to which this concern is addressed, various nuclear -power strategies can be developed between the current ho-reprocessing dption (and " hence no recycle) and options that would permit the unconstrained recycle of plutonium. The denatured 233U fuel cycle that is the subject of this report provides a middie ground within which nuclear power strategies may be developed. Although the denatured cycle does employ recycled fissile material, it can be structured so that it has more proliferation- resistant characteristics than the plutonium cycle. Before any proposed new fuel cycle "~ can be implemented, however, it must be addressed in the light of practical considerations such as the supply of U30q available, the projected nuclear power demand, the reactors and fuel cycles available, and the technological and implementation constraints imposed on the nuclear power system. These various aspects of nuclear power systems utilizing denatured 233U fuel have been discussed at length throughout this report. It is the purpose of this final section of the report to restate the most important conclusions of the study and to address trade-offs inherent in developing nuclear policy strategies that include the denatured 233U fuel cycle as opposed to strategies that do not. The nuclear power systems that have been examined can be classified as (a) no- recycle options, (b) classical reference recycle options, and (c) denatured recycle options. An integrated assessment of options in these three categories is presented in matrix form in Table 7.5-1, which also serves as a basis for the discussion that follows. In evaluating the systems, each option was characterized on the basis of the following criteria: (1) Nuclear proliferation resistance relative to other nuclear power systems. (2) Potential for commercialization of the reactor/fuel cycle components. (3) Technical feasibility on a reasonable schedule (and cost) for research, development and demonstration of the reactor/fuel cycle components. (4) Capability of the system for'meeting long-term nuclear energy demands. (5) Economic feasibility. As has been pointed out in earlier sections of this report, throughout this study the United States has been used as a base case since the available input data (that is, reactor design data, nuclear growth projections, etc.) required for the analytical model are all of U.S. origin. However, with appropriate data bases, the same model could apply to other individual nations. Moreover, it could apply to cooperating nations, in which case the nuclear strategy would include a mutual nuclear interdependence of the participat- ing nations. P et r— [ - A— ot i i - e ~eemamg tw b - e b (”F | I" ey | ¥ i € ! ’ 1 4 & e s b 7-43 7.5.1. No-Recycle Options Since commercial-scale reprocessing is not envisioned for some time, the currently employed once-through low-enriched uranium cycle (LEU) represents the only significant commercial possibility in the near term. At current ore and separative work prices, power generated via the once-through LEU cycle in LWRs is economically competitive with other energy sources. The once-through fuel cycle also has favorable proliferation- resistant characteristics: . its fresh fuel contains an inherent isotopic barrier; and while its spent fuel contains plutonium, the fuel is contaminated with highly radioactive fission products and thus .has a radiation barrier. On the basis of these and other advantages (see Case A in Table 7.5-1), the continued near-term use of the once-through LEU fuel cycle for nuclear-based electrical generation is desirable. The principal drawback of the once-through fuel cycle lies in the fact that it is tied to resources that will become increasingly more expensive{ Satisfying the nuclear demand postulated in this study to year 2050 would require the consumption of 5.6 to 7.1 million tons U0g. An equally important consideration is that it would also require an annual U40g production capacity of 90,000 to 130,000 tons of Uy0g by the year 2030. As the price of uranium increases, there will be incentives to reduce both these requirements by using uranium more efficiently. For example, improved LWR technology could potentially reduce U;0g consumption levels up to about 25% in the year 2030. A reduction in enrichment tails assay could result in an additional reduction in the uranium requirements of about 16%; however, this would require about 80% additional SWU capacity to maintain a constant production level of enriched uranium. But even with these gains the viability of the once- through cycle would be Timited by the avaijlability and producibility of U;0g from uncertain resources in the next century. A second once-through option (Case B in Table 7.5-1) would involve the addition of advanced converters to the power system either on the LEU cycle or on the MEU(235)/Th cycle. The implementation of the MEU(235)/Th once-through cycle in LWRs is uneconomic relative to the LEU cycle primarily because it would require higher fissile loadings and hence higher U30g commitments. And even if incentives were provided; the use of thorium-based fuels in LWRs would necessitate additional fuel R,D&D. To use either the LEU cycle or the MEU/Th cycle in other reactor types w0u]d entail significant expenditures to commercialize the reactors in the U.S. Moreover, the generic drawback of once-through cycles — that is, the uncertainty in the size of the economically recoverable resource base - would remain. On the other hand, as costs for extracting the resource base increase (to above $100/1b UJ0g, for example), commercialization of the alternate reactors will become more attractive. CIf continded re]iande'on'once-through fuel cyc]eS'is'to-be a Tong-term policy, it - would be desirable to make provisions for kestricting-the'spread of enrichment facilities. Also, safeguarding the spent fuel elements is necessary since the'p1utonium bred in the spent fuel represents a potential source of weapons-usable material which becomes increas- ingly accessible as.its radioactivity decays with time. Near-term resolution of the storage 7-44 question'cou]d be accomplished via international facilities chartered for just such a pur- pose. Such centers (and the institutional arrangements attendant to them) cod]d also serve as forerunners of the full-scale fuel cycle service/energy center concept considered for the recycle-based options. | 7.5.2. Recycle Options The inherent limitations of the resource base would regquire the use of recycled material to supplement the LEU cycle if the growth of a nuc]ear-baseq electrical generation capacity were to be sustained. Table 7.5-1 compares three recycle options utilizing de- natured fuel (Cases E-G) with two reference recycle options utilizing the classical Pu/U cycle (Cases C and D). The two reference cycles differ in that Case D employs FBRs while Case C does not. The denatured cases differ in that Cases E and F are all-thermal systems and Case G empToys FBRs in addition to thermal_reactors. Case E uses only LWRs as dis- persed reactors while Case F uses both LWRs and advanced converters (HWRs, HTGRs, or SSCRs). It has been assumed that, given a strong government mandate and financial support, all the fuel cycles and reactor types that have been considered in this report could be developed by the time they would be needed - by the year 2000 or later. However, the Pu/U cycle is much closer to being commercialized than the Th-based cycles, and, as noted in Chapter 5, the research, development, and demonstration costs for 1mp1ementing the denatured 233U fuel cycle in LWRs would be between $0.5 and $2 billion higher than the costs for implementing the reference Pu/U cycle in LWRs. If the HWR or HTGR were the reactor of choice, an additional $2 billion would be'required for reactor research, development, and demonstration. | A system in which reactors consuming Pu and producing 233 (transmuters) are combined with reactors operating on denatured 233U fuel appears to have somewhat better proliferation-resistant characteristics than a system based solely on the Pu/U cycle. The "fresh" 233 fuel is denatured with 238U, and thus some of the proliferation-resistant features of the front end of the LEU cycle would be extended to the recycle mode. That is, both chemical and isotopic éeparation of the fresh fuel would be necessary to obtain weapons-usable material. Additionally, the fresh denatured fuel is contaminated with radioactivity due to the décay daughters of a 232U impurity that is unavoidably produced along with the 233y, and the associated complications introduced into the isotope separation procedure would be severe. By contrast, weapons material could be obtained from Pu/U or 233/Th fuel through chemical separation alone, although the 233 obtained would also be radioactive due to the 232U daughters. (The Pu/U fuel would also be radioactive but much less so.) The spent denatured fuel represents a somewhat lower proliferation risk than the spent fuel from other options would. The recovery of a given quantity of Pu bred in the 238} denaturant would require the processing of more material than would be necessary in e ] i i ! ' e “r.-—-m-n i ot r— - F!!!? — e ey e o R g ey N [ O P 7-45 | | | | | Table 7.5-1. Integrated Assessment of Various Nutlear Policy Options for Meeting U.S. Nuclear Power Grpwth Demand Reactor/Fuel Cycle Combination A LWRs on LEU cycle LEU-LWRs followed by advanced converters on LEU {SEU) cycle or on MEU(235)/Th cycle Once-through LEU-LWRs followed by LWRs with Pu recycle . ' - 'Once-thFOugh LEU-LWRs followed by LWRs and FBRs with‘Pu_recycle : Dispersed LWRs operating on LEU and denatured 233U fuel with U recycle; energy- _ center thermal transmuters (LWRs) with Pu recycle Dispersed LWRs and advanced converters operating on LEU and denatured 233U fuel with U recycle; energy-center thermal transmuters (LWRs - and advanced converters) with Pu recycle . DiSperSed LWRs and advanced converters operating on LEU and denatured 233y fuel U recycle; energy-center fast transmuters with Pu recycle ‘ Proliferation Resistance 'Probably. best to the extent that non-nuclear weapons states continue to forego natfonal fuel recycle Fresh fuel has isotopic barrier; spent fuel contains radioactive fission products Spent fuel stockpile containing Pu is a ~ risk; requires institutional barriers Similar to above : HTGRs on. MEU/Th cycle would reduce Pu pro- duction by factor of 5 over LEU-LWRs but - fresh fuel would have higher 235U content (20%) | HWRs on SEU cycle about equal to LWRs on LEU cycle in Pu production Recycled Pu in fresh fuel chemically sepa- “rable; probably acceptable if Pu can be. 1imited to nuclear weapons states and to secure international fuel service centers " Option requires technical and institutional ‘barriers for Pu-fueled reactors (~30%) Spent fuel contains radioactive fission products -Increased risk over Case C because system tends .to become Pu dominated . Leads to significant Pu invertories and requires extensive Pu.transpor- tation for dispersed reactors - . Requires technical and institutional barriers ' "Fresh" denatured fuel has isotopic and radioactive barriers; spent fuel contains radfoactive fission products Spent denatured fuel contains less Pu than spent LEU fuel (factor of 2.5 Tess) Requires technical and institutional barriers to 1imit Pu to secure energy -centers Reduces Pu-fueled reactors by factor of 2 compared with Case C Fresh and spent denatured fuel advantages same as for Case E : Requires technical and institutional barriers Use &f HWRs or HTGRs substantially reduces Pu production relative to Cases C and E Pu.produced in denatured HWRs and HTGRs may be discarded with minor loss of fuel efficiency Very similar to Case E except that 15 to 50% of reactors may be Pu-fueled FBRs, depending on choice of cycles Implementation/Commercialization In wide commercial use Concern exists about fuel supply ' Emphasis on improved LWRs and U30g resource development needed Little commercial incentive to e Up to $2 billion for advanced converter introduce advanced converter Known to be technically feasible R,D&D Cost and Time of Commercial Introduction ~ No-Recycle Options " e Low cost e Gradual improvements introduced from year 1980 to year 2000 R,D&D e Advanced converters introduced in 1990's Concern exists about long-term: fuel supply Classical Reference Recycle Options Acceptable to private sector e About $1 billion, mainly for fuel cycle Requires completion of Generic R&D Environmental Impact Statement e Introduction in late 1980's on Mixed Oxide Fuel Preferred by private sector FBR 1icensing and commercial- . jzation may be difficult Uncertain public acceptance Fuel cycle somewhat more com- ! e FBR R,R&D up to $10 billion ' Fuel cycle R,D&D $1.6 to $3 billion fBRs not available before 2000 Dénatured Recycle Options e Up to $0.5 billion, PWRs and BWRs plex than Pu/U cycle, but func- ® Fuel cycle R,D&D $1.8 to $3.3 billion tionally equivalent /e Introduction in 1990's Requires government incentive | Same as Case E Advanced converters likely to to be attractive if FBRs are unavailtable Same as Case E Private sector 1ikely to accept government mandate Should be structured for maximum e Up to $2.5 billion for advanced converters ® Fuel cycle same as in Case E e Introduction in late 1990's Up to $10 billion for FBRs Converter R,D&D as in Cases E and F Fuel cycle $2 to $3.6 billion Introduction after year 2000 thermal-to-fast reactor ratio to allow siting flexibility Ability to Meet Ppwer Demands nt Peaks out between. years 2010 and 2030 and declines thereafter unless large amounts of low-grade P30y are exploited Peak could be increaspd and delayed 10 to 15 years with reactor improvements and reduced tails asshy Advanced converters qould extend usefulness of once-through cycle up to 10 years over standard LWRs Least resource efficiE Gains 10-15 years relative to Case A; somewhat less relative to improved A Superior ability to Jespond to power growth greater than that considered in this study Divorce from mining possible Somewhat better than Case C due to :up$r10r1ty of 233U as thermal reactor ue Can fully satisfy asJumed demand through year 2050 for plentiful Us0g supply; especially true if HQR converters used As good as Case D abo%e for assumed power demand Divorce from U mining Jess likely than for Case D above Economics Economics closely linked to U304 price Very favorable at current U;04 prices Uncertain capital costs cloud near-term interest Advanced converters favored at high U305 prices (>$100/1b) Preferred over Case A at high U40, (>$100/1b) Economics uncertain because of FBR costs, but probably acceptable Close to Case C Possibly lowest cost for U;0g price range of $100-$200/1b, especially for HTGR converter Economics similar to Case D above If FBR costs are high, can compen- sate by reducing the fraction of FBRs in the mix and increasing the mining rate 7-46 either the Pu/U cycle or the LEU cycle (about 2.5 times more than the LEU cycle). It must be noted, however, that the presence of chemically separable fissile material at any point in a fuel cycle represents a proliferation risk, and thus these points must be subject to stringent safequards. Also, the potential spread of enrichment facilities and improve- ments in enrichment technology (and hence greater ease in obtaining fissile material) may make such differences between the various fuel cycles less important. As is evident from Table 7.5-1, the private sector prefers the Pu/U cycle to the denatured fuel cycle, and a government mandate would probably be required to induce commercialization of denatured recycle in preference to Pu/U recycle. Private investors have developed recycle technology for mixed-oxide Pu fuels extensively, while putting little effort into recycle technology for thorium-based fuels. Because reprocessing is inherent in the denatured 233y cycle, 1mp1ementation of the cycle is Tikely to require the development of "fuel service centers," safeguarded facilities whose purpose would be to protect sensitive fuel cycle activities. Such centers could evolve from the safequarded spent fuel storage facilities required for the once-through fuel cycles. For the recycle scenarios, the center would first contain sensitive fuel cycle facilities to produce denatured 233U fuels from stored 233U-containing spent fuel; later it would include those reactors that operate on fuel from which the fissile component could be chemically separated. Under the assumption that no weapons-usable fuel that is chemically separable can be used in dispersed reactors, a power system utiiizing denatured 233y fuel has a significant advantage over one based on the Pu/U cycle alone. The Pu/U cycle would necessitate that all reactors be constrained to the enefgy center, which will result in a penalty for electric power transmission since energy centers could not be sited as conveniently as dispersed reactors. With a denatured system, a significant fraction (up ~ to 85%) of the power could be dispersed since only the Pu-fueled transmuters would be oper- ated in such centers and thus the system could maintain a relatively high energy-support ratio (ratio of nuclear capacity installed outside center to nuclear capacity installed inside center). - Evaluation of the denatured 233U fuel cycle on the basis of economics and/or energy supply is difficult due to the uncertainties in unit cost factors and potential energy demand. With the economic and energy demand assumptions employed in the analtysis pre- sented in Chapter 6, however, the economics of the denatured cycle appear to be equjvalgnt to, or slightly better than, the economics of the classical Pu/U cycle for moderate growth-rate scenarios (i.e., those employing combinations of fast and thermal systems). Although the fuel cycle unit costs of the denatured cycle were assumed to be higher than those of the Pu/U cycle, power systems utilizing denatured 233U fuel typically allow a larger fraction of the reactors constructed to be thermal reactors, which have lower capital costs. This is possible because the nuclear properties of 233U are such that it can be used in thermal reactors more efficiently than in fast reactors. | - ( bt onsd e b e (- o) e v - ith. - Yo i 1 i o T ! P} 7-47 ~ Although the strategy analyses presented in Chapter 6 considered various advanced convefters as potential dispersed denatured reactors, the selection of an optimum advanced converter is precluded at this time due to cost and performance uncertainties and the failure of this study to identify a single advanced converter for further development on the basis of commonly accepted selection criteria. For example, at high U305 prices, the HTGR appears to generate the lowest-cost power of the thermal reactors, while an HWR | appears to be the most resource-efficient and to have the best energy-support ratio on the denatured cycle. The SSCR might be developed most quickly and cheaply. A1l the advanced converters, but particularly the HWR and the HTGR, appear to have certain superior fuel utilization characteristics relative to standard LWRs due to their higher conversion ratios (i.e., lower 233U makeup requirements), lower fissile inventories, and lower Pu broduction. Denatured advanced converters also can be sustained at higher supbort ratios than can denatured LWRs. [Cycles with potentially higher thermal efficiencies (such as the direct cycle) and potential siting advantages were not considered in the comparisons of the advanced converters.] The introduction of denatured advanced converters, however, is estimated to require up to $2 billion more research, development, and demonstration expenditures than would the introduction of a denatured LWR. Moreover, a denatured LWR could be commercialized up to 10 years sooner than a denatured advanced converter. Developing a denatured LWR would be less difficult due to the backlog of LWR experience and the reduced risk associated with a previously demonstrated reactor system. The capital cost of an advanced converter, although generally lower than the cost of a fast reactor, is estimated to be- somewhat higher than that of an LWR. Thus, the improved performance must be weighed against the increased capital costs, the delay in introduction, and the research and deveTOpmént costs in any decision relative to the use of advanced converters in con- junction with the denatured cycle. The analysis of Chapter 6 indicates that, as 233 producers, fast transmuters would have more favorable resource characteristics than thermal transmuters. For the energy demand assumed in this study, the most satisfactory denatured power system would consist of denatured thermal reactors coupled to fast transmuters in a symbiotic relationship, the logical transmuter candidate being a fast reactor with. (Pu-U)0, drivers and ThO, blankets. It should be noted, however, that a more rapid_growth-ih energy demand could dictate that Pu/U breeders be constructed to meet the demand or that some combination of Pu cycle breeders containing thorium and dispersed denatured breeders be used. In these cases the nuclear power capacity could grow independent of the resource base. .~ Although the aenatured cycle appears to possess advantages relative to the Pu/U cycle, several importaht areas require further study. In particular, the refinement of the denatured advanced converter characterization is of prime importance, both to evaluate various reactor Options and to study the overall use of advanced converters as opposed to LWRs. As the potential for improving the performance of LWRs, both on the once-through 7-48 and recycle modes, is better defined and as advanced cdnverter desighs are optimized for denatured systems, the analysis will become more useful for R,D&D planning. Also, system interaction studies for the dispersed denatured reactors and centralized transmuters require refinement based on improved reactor designs and updated mass balances. Finally, the question of implementing the energy-center concept, togethéf with the use of specially designed transmuters as a source of denatured fuel, deserves more detailed study.” The Nonproliferation Alternative Systems Assessment Program (NASAP) is currently developing characterizations of improved fast transmuters, improved LWRs, and reoptimized advanced converters and LMFBRs. Light Water Breeder Reactors (LWBRs) wilt also be included in these characterization studies. - - | 7.5.3. Overall Conclusions and Recommendations The denatured 233U cycle emerges from this assessment as a potential alternative to the conventional Pu/U cycle. Its advantages may be characterized as follows: e The denatured 233 cycle offers proliferation-resistanée advantages relative to the Pu/U cycle in that the "fresh" denatured fuel has an isotopic barrier; that is, it does not contain chemically separable Pu or highly enriched uranium. By contrast, the Pu/U cycle together with fast breeder reactors tends toward an equilibrium with all reactors using Pu fuels. Also, fresh denatured fuel has a much more intense radioactive barrier than does the fresh fuel of the classical Pu/U cycle. e For moderate growth rate scenarios, deployment of power systems that include reactors operating on denatured 233U fuel would allow a larger fraction of the reactors in a power system to be thermal reactors. This would tend to minimize the overall capital costs of the system compared to fast/thermal power systems based on the Pu/U cycle. e If in addition to LWRs, the denatured thermal reactors of the power system were to include denatured advanced converters, the‘dépendence of the power system on a fast reactor component (i.e., fast transmuters)'could be further minimized due to the improved resource utilization of denatured advanced converters compared to denatured LWRs. Although the advanced converters would have higher capital costs than the LWRs, this might be offset by reduced requirements for FBRs. The disadvantages of the cycle are the following: o The denatured 233 fuel cycle is more complex than the Pu/U cycle, and since 233 must be produced in transmuter reactors, the rate at which denatured 233y reactors can be introduced will be inherently limited. Because the Pu/U cycle } { —y e o, il +oaline A [ i Y-t l[": N i (= ) 1 i 7-49 technology is closer to commercialization, there is a reluctance both by U.S. industry and by foreign governments to embrace an alternative which is less developed and which is considered primarily on the basis of its rnonpro1iferation advantages, and this would have to be overcome. | The R,D&D costs for'deveIOping the denatured 233U fuel cycle are significantly higher than those for the Pu/U cycle. If advanced converters must also be developed, significant additional costs would be incurred. Other important conclusions from this study are as follows: On the The once-through cycle based on LWRs is likely to dominate niuclear power production through the year 2000. This provides time to develop either the denatured cycle or the Pu/U cycle for the recycle mode. The denatured 233U fuel cycle can be used in LWRs, SSCRs, HWRs, HTGRs, and FBRs without major changes from the present conceptual reactor designs based on their reference fuels. After the necessary R,D&D is completed, the denatured 233 fuel cycle appears to be economically competitive with the Pu/U fuel cycle in LWRs, advanced converters, and in symbiotic fast-thermal recycle systems. With the fuel resources assumed, the nuclear power demand postulated in this study (350 GWe in the year 2000 and a net increase of 15 GWe/yr thereafter) can be met as well by the denatured fuel cycle as it can by the Pu/U cycle. However, the Pu/U-FBR cycle has an inherent ability to grow at a faster rate than the other cycles. basis of this study, it is recommended that: Optimized designs of alternate breeders, improved LWRs, HWRs, SSCRs, and HTGRs be examined to refine the characteristics of the denatured cycle relative to fuel utilization, economics and energy-support ratio. The study should also be expanded to include LWBRs and the fast breeder designs developed by DOE in the Proliferation Resistant Large Core Design Study (PRLCDS). More detailed assessments of the proliferation risks and the economics of the denatured cycles compared to other recycle options (Pu/U and HEU/Th) should also be pursued. 7-50 These studies could provfde guidance for the following R&D programs: Thorium fuel cycle R&D to investigate the use of MEU(235)/Th, MEU(233)/Th (denatured 233U), and Pu/Th fuels in LWRs and HWRs (the latter in cooperation with Canada). This program might also include the LWBR fuel cycle. Studies to consider denatured 233U or 235y fuels as candidates for the HTGR reference fuel cycle. Thorium technology studies, particularly for blanket assemblies, as an integral part of the FBR programs (LMFBRs and GCFBRs). ' Exploratory work with uti]ities and PWR and BWR vendors for qualification and use of MEU/Th and Th fuel rods in commercial reactors. An example of the beneficial use of Th would be in corner rods of the BWR fuel assembly. ¥ APPENDICES . e ihv_....u . £ ( £ - o Ty e | o ' g s r—l i | o —C A-3 Appendix A. ISOTOPE SEPARATION TECHNOLOGIES E. H. Gift Oak Ridge Gaseous Diffusion Plant A.1. Current Separation Capability . Three enrichment technologies exist that are sufficiently advanced to be classi- fied as current separation technology. These are: a. The Gaseous Diffusion process. b. The Gas Centrifuge process. ; c. The Becker Separation Nozzle process (and its variant, the South African Helikon process). Both the centrifuge and the Becker processes are expected to provide enrichment services that are competitive with gaseous diffusion. The centrifuge process, in parti- cular, is projected to provide a 30%! saving in separative work cost when fully imple- mented in a large scale plant. A brief description of each of these processes and their current productive capacity follows. The Gaseous Diffusion Process? The gaseous diffusion process is based upon the physical fact that in a gas made up of molecules of different masses, molecules containing the lighter mass isotopes will, as a result of the distribution of kinetic energies, have average velocities slightly faster than those which contain the heavier isotopes. As a result, these lighter isotopes will reach the walls or pores in the walls of a containment vessel more frequently and at higher velocities. « In the gaseous diffusion process, the container wall is a porous tube (barrier) through which diffusion is accomplished. The maximum theoretical separation that can be achieved is a function of the square root of the ratio of the masses of the gas molecules. In the diffusion process, utilizing uranium hexafluoride, the square root of the ratio is 1.00429. Because this “number is so close to unity, the degree of enrichment which can be achieved in a single diffusion stage is very small, but the effect can be multiplied by making use of a cascade consisting of a number of stages. Production of 90 weight percent 235U from 0.711 weight percent 235y material, as found in natural ore, requires about 3,000 diffusion stages in series. A'pldnt constructed for the purpose of producing material of up to 4.0 weight percent 235U, as might be required for typical light water power reactors, would contain about 1200 stages. To take advantage of the small separation factor discussed above, diffusive flow must be ensured, not just simple gas flow. Diffusive flow requires not only small pores, i.e., less than two-millionths of an inch in diameter, but also uniformity of pore size. Because of the small pore size, literally acres of barrier surface are required in a large production plant. . Complexity of plant design is increased by the difficulties arising from the nature of fhe diffusing gas itself. A volatile compound of uranium must be used, and the hexafluoride (UFg) is the only known suitable compound. It is a solid at room temperature; consequently, the diffusion plants must be operated at temperatures and pressures necessary to maintain the UFg in gaseous form. Although it is a stable com- pound, UFg; is extremely reactive with water, very corrosive to most common metals, and not compatiblie with orgahics such as lubricating oils. This chemical activity dictates the use of metals such as nickel and aluminum and means that the entire cascade must be leak-tight and clean. The corrosiveness of the process gas also imposes added diffi- culties in the fabrication of a barrier which must maintain its sépardtive-qua}ity bvér long periods of time. ‘ ‘ The enrichment stage is the basic unit of the gaseous diffusion process. In all stages gas is introduced as UFg and made to flow along the inside of the barrier tube. In the standard case about one-half the gas diffuses through the barrier and is fed to the next higher stage; the remaining undiffused portion is recycled to the next lower stage. The diffused stream is slightly enriched with respect to 235U, and the stream which has not been diffused is depleted to the same degree. The basic equipment components vital to the process are the axial flow compressors, the converter shell and the barrier tubes. Axial flow compressors are used to compress the UFg; gas to maintain the interstage flow, and electric motors are used to drive the compressors. A gas cooler is provided in the converter since gas compression unavoidably generates heat which must be removed at each stage. The diffuser, or converter, is the large cylindrical vessel which contains the barrier material. It is arranged in such a fashion that the diffused stream and the stream that has not diffused are kept separate. Groups of stages are coupled td make up operating units and such groups, in turn, make up the cascade. - Gaseous diffusion plants are in operation in the Uni ted States, England, Frahce, and Russia. e vl 1 ~y ¥ ey iy S [.._‘ Lo r:::(t i ot v ety ? o d g ot i o o oo -t C " A-5 The Gas Centrifuge Process The countercurrent gas centrifuge separafion of uranium isotopes is based on processes developed more or less independently in the U.S. at the University of Virginia, in Germany,“ and in Russia® during World War II. Much of this work was reported at the 1958 Geneva Conference. In the U.S. this work was continued at the University of Virginia and reported in 1960.% The machine developed is shown in Fig. A-1. 3 The theory“:7 for operation of the gas centrifuge shows that the maximum separative capacity of a gas centrifuge is proportional to: a. The fourth power of the periphera]-speed, b. the length, and c. the square of the difference in molecular weights. Thus, it is evident that one should make the peripheral speed and the length of the centrifuge as large as possible. The peripheral speed is limited by the bursting strength of the material of the rotor wall. A long rotor of small diameter is comparatively flexible and will pass through a series of resonant mechanical vibration frequencies while being accelerated to high peripheral speed. Unless provided with special damping bearings, a centrifuge would destroy itself while passing through one of these resonant speeds. Much of the world's effort in advanced centrifuge development has been designed to keep below the first resonant frequency. As a result, they are comparatively short and have relatively low separative capacity. Some of the differences between gas centrifuge and gaseous diffusion technologies should perhaps be noted. Gaseous diffusion requires fabrication of permeable barriers with a very small pore size; the manufacture of these barriers is a difficult process and a closely guarded secret. Gas centrifugation requires manufacture of high-speed rotating equipment. While such ‘manufacture is certa1n1y not trivial, it basically requires a we]]-equ1pped precision mach1ne shop that may well be within the technical capabilities of many nations. The technology of rotating mach1nery is widespread and designs for gas centrifuges are in the openi1itereture. The power requirements for a centr1fuge fac111ty are much less than for a diffusion facility of the same size. For U.S. plants of economic scale and of the same separative capacity, gas centr1fugat1on requires about 7% of the power needed for gaseous diffusion.® : Fo1lowing the early work ih the U.S., further research on the centr1fuge process was undertaken for the USAEC by the University of Virginia, Union Carbide Corporat1on Nuclear D1v1saon and Garrett Corporation-AiResearch Manufactur1ng Co. , and Dr. Lars Onsager. The current status of the U.S. program can best be indicated by a brief description of the operating and planned facilities:! Heavy Fraction » Light Fraction UF,. Feed 6 Magnet "/’,,a?”"” Suspension << Statiohary e N Magnet e /\ Vacuum —_ . Pump — e \, Spinning Magnet A Damper SElE=S 5| | X e : . Molecular / ] ' * N Pump N\ N N \\*' Scoop - \ g N\ N | _ Rotor ‘ } }/- c —Y N Baffle asing : , Plate \ />/ Y1 Yo s \ N T N\ N I~ =) o I b Scoop N C N\ | | J"‘———-h Drive I —————— K er N Motor NN N N NN NN . Bearing Fig. A-1. ZIPPE Centrifuge (Simplified). ~ The Equipment Test Facility (ETF) was conceived to provide for the reliability testing of "high Capacity" centrifuges. This facility, which began operation in 1971 ,- has been the source of reliability testing for two generations of machine designs. Many of the first generation high capacity machines are still operating in this facility. £t . ‘C;“f T 4 1 1 | ik T | i § ) C The Component Preparation Laboratories (CPL) in Oak Ridge, Tennessee and Torrance, California, were built to evaluate, improve and demonstrate techniques amenable to the mass production for manufacturing centrifuges. This facility became operational in early 1974. The Component Test Facility (CTF) was designed to demonstrate the machine reli- ability and operability testing of substantial numbers of centrifuges in a cascade operation. Construction was begun in 1972 and the first phase of startup of the facility was completed in January 1977 with cascade operation of about one-haif of the machines operating. The remaining machines were operable within a few weeks later. The capacity of the CTF is significant, about 50,000 SWU/yr, or about the annual enriching requirement for a 500 MW power reactor. | The Advanced Equipment Test Facility (AETF), in addition to being a reliability test facility will also test the plant subsystems which 5upport the machines. The machines to be installed in this facility will have significént]y greater separative work capability than those in the CTF. The AETF is expected to be operable in the spring of 1978. In Europe, the URENCO organization, consisting of participants from England, Germany, and Holland, has a program that so far has been directed toward machine reli- ability and long lifetime. URENCO is currently producing about 200 MTSWU/yr from plants at Almelo, Holland and Capenhurst, England. Ekpansion of these facilities is planned by 1982. The URENCO group expects to have 2000 MTSWU/yr in operation, 1300 MTSWU/yr at Almelo, and the remaining 700 MTSWU/yr at Capenhurst. The Becker Separation Nozzle The Becker process,? being developed in Germany by Dr. E. W. Becker and his associates, utilizes the pressure gradient developed in a curved expanding supersonic jet to achieve separation in a gas mixture. The separation nozzle stage is shown schematically in Fig. A-2. A light gas, helium or hydrogen, is added to the UFg in order to increase the velocity of the jet. As the expanding jet traverses the curved path, the heavier component is enriched in the vicinity of the wall. A knife edge divides the jet into two fractions--one enriched in the light'component,'and the other enriched in the heavy ,component—fwhich are then pumped off sepérate]y from the'stage. Although the separation obtained per stage is relatively high (~1.025), many separation nozzle stages are needed to obtain an appreciable enrichment, This process avoids the problems associated with the fine-pored membrane required for gaseous diffusion, and those associated with the high-speed rotating parts of the gas centrifuge. It does suffer, however, from the disadvantage of a relatively high power requirement, primarily because a great deal of light gas must be recompressed between stages along with the UFg process gas. A-8 DWG. NO. G-65-848 P, = 48mm Hg P = l4mm Hg Ng = 0.05 Ng = 0.15 LIGHT HEAVY FRACTION _ FRACTION FEED GAS _ NOZZLE , DEFLECTING WALL Imm — P = total pressure; N = mole fraction of'UFf in the UFg/He mixture. Subscripts o, M, and K refer to feed gas, light and heavy fractions, respectively. Fig. A-2. Cross Section of the Separation Nozzle System of the Becker Process. A small 10-stage pilot plant was operated in 1967 to prove the technical feasi- bility of the process. Following that, a single large prototype stage suitable for use in a practical cascade was fabricated. | A prototype separation stage contains 81 separating elements and is reported to | i have a separative capacity of approximately 2000 kg U SW/yr. A plant producing a product | enriched to 3% 235y and with tails at 0.26% 235U is expected to require about 450 such stages. ' ' ‘ - Figure A-3 shows the individual separating elements, each containing 10 separation nozzle slits on its periphery. The fabrication of these units is not as simple as one might at first expect. In order to obtain the desired separation performance at reasonable pressures, it is necessary to employ very sma11'geometries. The spacing between the knife edge and the curved wall in the prototype separating unit should be about 0.0005 of an inch. In order to obtain good performance, it is necessary that this spacing not deviate by more than 210% over the 6-foot length of slit. The power requirement for the Becker proceSs is turrent]y estimated to be about one and one-third times as great as that required for gaseous diffusion. Dr. Becker believes that further process improvement is still posszble and that the power reguire- ment can be substantially reduced. r i ! | et . o r A-9 DWG. ND. G-69-643 Light fraction Feed Hea vy gas fraction Detail Fig. A-3. Becker Separating Element With Ten Slits The South African Helikon Process The South Africanl® (or UCOR) process is of an aerodynamic type whose separating element is described by the developers as a high-performance stationary-walled centri- fuge using UFg in hydrogen as process fluid. All process pressures throughout the system will be above atmospheric and, depending on the type of "centrifuge" used, the maximum process pressure will be in a range of up to 6 bar. The UFg partial pressure will, however, be sufficiently low to eliminate the need for process heating during plant operation, and the maximum temperature at the compressor delivery will not exceed 75°C. The process is characterized by a high separation factor over the element, namely from 1.025 to 1;030,depending on economic considerations.‘_Furthermore, it has a high degree of asymmetry with respect to the UFg flow in the enriched and depleted streams, which emerge at different pressures. The feed-to-enriched streams pressure ratio is typically 1.5, whereas the feed-to-depleted streams pressure ratio is typically dniy 1.12. To deal with the sma]l'UFs_cut, a new cascade technique was deveToped-che so-called "helikon" technique, based on the principle that an axial flow compressor can simul- tanebusly'transmitrseverai streams of differeht isotopic composition without there being significénf mixing between them. The UCOR process must, therefore, be regarded as a combination of the separation element and this technique, which makes it possible to achieve the desired enrichment with a relatively small number of large separation units by fully utilizing the high separation factor available. A further feature of the helikon A-10 technique is that a module, defined as a separation unit consisting of one set of com- pressors and one set of separation elements, does not as in the classic case, produce only one separation factor of enrichment in one pass but can produce for a constant separative work capacity various degrees of enrichment up to a maximum of several times the separation factor over the element. 4 Full sca]é modules of this type are nearing the.prototype stage. Recent design jmprovements are expected to result in a nominal capacity of 80 to 90 kg SWU/yr!l! per separation module. A valuable feature of a plant based on this process is its very low uranium inven- ~tory, which results in a short cascade equilibrium time, of the order of 16 hours for a commercial plant enriching uranium to 3% 235y, The theoretical lower 1imit to the specific energy consumption of the separation element can be shown to be about 0.30 MW.h/kg SW. The minimum figure observed by the developers with laboratory separating elements is about 1.80 MW.h/kg SW, based on adiabatic compression and ignoring all system inefficiencies. This difference is a measure of the improvement potential expected by the South Africans. Current and Projected Enrichment Capacity Most of the known installed enrichment capacity is based upon gaseous diffusion technology. Only small increments of centrifuge technology are in operation (i.e., URENCO, Japan and U.S.), and one plant utilizing modified nozzle technology (the South African Helikon plant) may be operating. Indicative of the status of other isotope separation methods, all planned additions to the world enrichment capacity are based on ejther diffusion, centrifuge or nozzle technology. | ' The existing worldwide capacity and pTanned additions to capacity are shown in Table A-1 by country and technology type. In the table the groups identified as Eurodif and Coredif are multinational organizations building gaseous diffusion plants in France. ' ' A.2. New Separation Technologies In addition to the more developed technologies (gaseous diffusion, gas centri- fuge, and the Becker nozzle), there.are several other separation methods that either have been utilized in the past or are currently being developed. These technologies . are listed in Table A-2. e el ) 4 3 i C = - o - Table A-1. Approximate Schedule. of World Enrichment Capacitya ) World's " Capacity Cumulative Nation Technology Increment i Capacity Year or_Group . Type (MT SWU) Present Status of Increment __(MT _SWu) 1977 u.s.b Diffusion 15,400 - Existing 15,400 UK-France Diffusion 800-1000 Existing, but dedicated to 16,400 military use Russia® Diffusion 800 Existing, actua) total 17,200 capacity unknown China Diffusion Unknown Existing, mostly military URENCO Centrifuge 200 Existing 17,400 U.S. Centrifuge 50 Existing 17,450 S. Africa Helikon-Fixed Unknown Existing pilot plant or in wall centrifuge process of coming on-line 1978 v.s.? Diffusion 3,300 From CIP/CUP plus added 20,750 power purchase URENCO Centrifuge 200 Facilities at Almelo & Capen- 20,950 hurst now in construction Japan Centrifuge 20 Currently under construction 20,970 Russia® Diffusion 200 21,170 1979 U.S.b Diffusion 2,200 From CIP/CUP 23,370 Russia® Diffusion 500 23,870 URENCO ~ Centrifuge 400 Under construction 24,270 Eurodif Diffusion 2,600 26,870 1980 U.S.b Biffusion 1,600 From CIP/CUP 28,470 URENCO Centrifuge 400 Planned 28,870 Eurodif Diffusion 3,700 Under construction 32,570 Japan Centrifuge 20 Under construction 32,600 Russia® Diffusion 500 33,100 1981 U.S.b Diffusion 700 From CIP/CUP 33,800 URERCO ~ Centrifuge 400 Planned 34,200 Eurodif Diffusion 2,100 Under construct1on 36,300 Russia® Diffusion 500 36,800 1982 u.s.? Diffusion 300 Incr. Power Implementing CUP 37,000 URENCO Centrifuge 400 - Planned 37,500 Eurodif Diffusion 2,400 Under construction 39,900 Russia Diffusion 500 40,400 Brazil Becker nozzle 180 Planned 40,580 1983 URENCO ~ Centrifuge 1,300 Planned 41,880 Coredif Diffusion 1,800 Planned 43,680 1984 u. .b Diffusion 2,000 Incr. Power Implementing CUP 45,680 URENCO Centrifuge 1,300 Planned 46,980 S. Africa Fixed wall 1,600 Planned 48,580 centrifuge Coredif Diffusion 1,800 Planned 50,380 1985 u.s.? Diffusion 2,000 Incr. Power Implementing CUP 52,380 URENCO Centrifuge 1,400 Planned 53,780 S. Africa Fixed wall 1,600 Planned 55,380 . centrifuge Coredif Diffusion 1,800 Planned 57,180 . Japan Centrifuge 6,000 Planned, but should be 63,180 , considered conditional 1986 u,s.? . Centrifuge . . 650 Planned 63,730 S. Africa Fixed wall 1,800 Planned 65,530 ' centrifuge - : . URENCO Centrifuge 2,000 Planned 67,530 1987 - u.8.° Centrifuge 2,750 Planned 70,280 : . URENCO Centrifuge 2,000 . .Planned 72,280 1988 v.s.2 _Centrifuge 3,300 Planned 75,580 1989 v.s.? Centrifuge 2,200 . - Planned 77,780 . . Coredif Diffusion 5,400 Planned, but should be 83,180 considered conditional ®Information from references 12 and 13. Byot included in this schedule are possible additions to the U.S, enrichment capacity by private corporations. such as Exxon Nuclear, Garrett and Centar; these may amount to as much as 10, 000 MT SWU by 1990. °For Russia, this is a scheduie of growth in enrichment sales ayailability and not necessarily of capacity expansion. A-12 Table A-2. Other Isotope Separation Technologies A. Discarded Technologies Thermal Diffusion Electromagnetic (the Calutron Process) B. Developing Technologies Photo-Excitatioh Methods (Laser) Chemical Exchange Methods Aerodynamic Methods (Other Than the Becker Nozzle and the Fixed Wall Centrifuge) Plasma Based Processes The discarded technologies listed in Table A-2 have been used to produce enriched uranium. ' - A large-scale, liquid-phase, thermal-diffusion plant was constructed in 1945 by the Manhattan Project.l% This plant produced very slightly enriched uranium (0.86%). Thermal diffusion is impractical for commercial enrichment of uranium isotopes because of its very high energy requirements. Compared to gaseous diffusion, the energy requirement is over 200 times greater. - The electromagnetic or Calutron methods were used dufiing the Manhattan Project to produce highly enriched uranium.l* The process was discarded shortly after the more economical gaseous diffusion plant began operation. A brief description of the process follows. The Calutron Process involved the vaporization of a salt feed material, typically UC1,, from an electrically heated charge bottle through slots into an arc chamber where the salt was ionized by an electron beam which trayels along the lines of flux of the magnet.. The ionized uranium, as the ut ion for the most part, passed through another slot where it was accelerated by other slotted electrodes into the vacuum tank which filled the pole area of a large electromagnet. The ions from the accelerating electrodes diverged several degrees from the slots and at the 90° point passed by some baffles as a rather thick beam. This beam wasgbrpught to a focus at the slots of a receiver system as curved lines by the shimmed magnetic field. 1In the large units, 96-in. beam diameter, there were up to four of these beams in a given tank. The divergent trajectories of the ions from the four sources intersected some few degrees from the accelerating electrodes and separated as distinct beams, again a similar distance from the receivers. There were various side beams of UC1+, U++, and other jons which hit the baffles and the walls of the tank“a; a series of locations. The uranium content of these beams condensed as various compbunds‘of uranium. The product was, for the most part, converted to UC by interaction of the very high voltage uranium ions with the graphite of the receivers. Since, in even the most | r i -~ !n——_—-n-g ] cr e = i — Pty 1 ! A g - vy 1 . t A e i N . r e . A-13 efficient of the units developed, only about 22% of the feed was collected as product in a vaporization cycle of the feed, there were large amounts of uranium compounds to be recovered and recycled through the system. The chemical operations required were complex, but the amount of space and the number of workers required in the chemical function were always small compared to the requirements of the rest of the process. The processing of the receivers to recover the product uranium was a small scale but very demanding series of chemical procedures, The developing technologies listed in Table A-2 offer no current capability forproducing kilogram gquantities of enriched uranium. If any of them approaches commercial feasibility, they may provide enhanced opportunities for a clandestine enrichment operation. A brief description of each of these processes follows. Photoexcitation (Laser) Methods The development of high intensity narrow-frequency' tunable lasers has raised the possibility of nearly complete isotopic separation in a single step. Thus, reactor grade and perhaps even weapons grade uranium could be produced in one pass through the apparatus. Such a single—stage process would allow for a much more compact enrichment plant, saving land area, capitaT investment and power consumption. These hopes have led to active research and development programs in the Un1ted States, the Soviet Union, Israel, France and possibly other countries. In the U.S. the development of laser enrichment is being puréued d]ong,two distinct lines. One line of development uses atomic uranium vapor as the source material for the laser excitation whereas the other line of development is pursuing excitation of molecular uranium hexafluoride. Each method has its virtues and defects. Laser Enrichment with Atoms.!> In the atomic enrichment process most often discussed, molten uranium js heated in an oven to about 2500°K. The atomic vapor emerges in the form of a long, thin ribbon into a h1gh1y evacuated region where it is illuminated by two visible or near-ultrav1o1et lasers. One laser is tuned to a transition from the ground state of uranium to"an exc1ted state roughiy halfway up the ladder to 1on1zat1on This is the tsotop1ca11y se]ect1ve step, and it 1s hoped | that very h1gh seTect1v1t1es will be ach1eved here, The purpose of the seCond laser is to boost the excited 235U atoms to a level just below the jonization limit. This step need not be isotopica]]y'se1ective, and in principle the second laser could be used to ionize the atom d1rect1y But ioniza- tion cross sections are genera]Ty about 1000 times smaller than resonant exc1tat1on cross sections, and so it is far more efficient to use a resonant transition to excite the atom to a state just below the ionization level and then to uée either a static A-14 electric field or an infrared laser pulse to pull the electrons off the atoms. Once the atoms are ionized, they can be separated from the neutral atoms in the beam by the use of electric or magnetic fields, or both. The major 1imiting factor in the above process is the density of atoms in the uranium "ribbon." There is an upper limit on the density and therefore on the rate of production of enriched uranium, because both excitation energy and ionic charge are very easily transferred to other atoms in collisions. Such collisions must be kept to ‘a minimum if a high selectivity is to be obtained. Other technical difficulties in the development of the process are: a.. The corrosiveness of the uranium vapor. The presencé of thermally excited or ionized atoms of 235U in the uranium vapor (at 2500°K, ~55% of 2350 atoms are not in the ground state). c. The potential for self lasing of the uranium vapor. d. Thermal ionization of 238y will ser1ously degrade the se]ect1v1ty and thus limit the enrichment. - , e. Lasers combining high energy density, rapid pulse repetition rate, high tuning precision, and long-term stability and reliability must be developed. Laser Enrichment with Molecules.l5 Gaseous UFg is used in all proposed schemes for molecular enrichment, since this is the only compound of uranium with a sizable vapof pressure at reasonable temperatures. Because the molecule contains seven atoms and exhibits a high degree of symmetry, it produces a complicated spectrum of vibrational and rotational excitations. The most interesting vibrational modes from the point of view of laser excitations are those which involve motion of the uranium atom and which therefore produce an oscillating electric dipole moment. Only these modes are\likely to produce transitions from the ground state when excited by elec- tromagnetic energy. The low energies associated with these transitions lead to two serious problems for laser enrichment in UF;. The first problem is the creation of an infrared laser with the correct frequency. The second problem is related to the high occupation numbers of the low-energy vibrational states at temperatures where UFeg has a high vapor pressure. Because so many low-lying states are occupied, it is impossible to find a single excitation frequency that wiil be absorbed by most of the molecules. The presence of these so-called "hot bands" reduces the efficiency of the process very drastically. The second problem is easily solved, at least in principle, if warm UF; gas is passed through a supersonic nozzle. The effect of the expansion is to convert most of the kinetic energy of random motion of the gas in the reservoir into kinetic energy of translational motion of the gas in the nozzle. As the gas accelerates r . ‘[- . — r . s o o et oL e o oY, C through the nozzle, it becomes colder and the energy stored in the vibrational and rotational degrees of freedom of the molecules is reduced by intermolecular collisions in the narrow region just downstream of the slit. The molecules can now be illuminated by a laser beam which has been tuned to excite selectively molecules containing 235U. This teéhniqueAyields the first step in the molecular isotope separation process; however, this selective excitation does not provide a way of segregating the excited molecules. To do this, considerably more laser energy must be absorbed by the molecules to get them to dissociate to 235UF; and fluorine. 1In theory, this energy can be provided by either an infrared or an ultraviolet laser. Since it is not necessary for either of these secondary processes to be isotopically selective, the primary demands on the ultraviolet or infrared lasers are related to their energy output and pulse repetition rates. In both cases considerably higher powers are required for the molecular than for the atomic processes because much larger numbers of molecules can be processed in the same period of time. This high power requirement follows because the density restrictions apparently are less severe for molecules than for atoms. The dissociated product must still be physically separated frem the undissociated material and substantial recombination could occur if the recombination probabilities for UFs and F are high. ' As with the atomic process, the molecular process must also overcome formidable technical difficulties before it becomes a feasible production process. Some of these obstacles are: a. The high probabi]1ty of resonant vibrational energy exchange between the 235yFg and the 238yFg, b. The recombination of dissociated molecules. ¢. An infrared high-powered 1aser tunab]e to the requ1red wave 1ength for the primary -excitation must be invented. _ o . d. The secondary. laser must satlsfy the comb1ned demand of h1gh pu]se energy, rapid repet1t1on rate and high efficiency. ' ' : e. The rapid and efficient separat1on of the d1ssoc1ated product from the depleted tails. ' ' ' ‘ChemicallExchange Methods The use of a chem1ca1 exchange system to separate meta] 1sotopes has been under investigation in the U.S. for several years In addltton to work in the U.S,, the French recently have made allusions to similar research. It has been shown that calcium A-16 isotope enrichment can be accomplished using a simpTe extraction process involving the relatively new class of compounds known as polyethers. Work is underway to determine whether a similar process could be used for uranium isotope enrichment. The electron exchange equilibrium between U(IV) and U(VI) may result in a significant isotope enrichment. The extraction of a single uranium cation without a valence change yields a small 1sotope effect which by itself wou1d have no practical use. Combining the two processes leads to a potent1a11y economic process for uranium isotope enrichment. C The electron exchange reaction which occurs in the aqueous phase can be described by Equation 1: 2354+ 4 233U022+ & 2384+ 4 235UO22+ _ _ ('[) This reaction was reported to have an « = 1.0014 with 238U concentrating on the U(IV) ion. The solvent extraction exchange reaction of the U(VI) ion can be described by Equat1on 2: ‘ - 2?5U°22+(aq) . 238U02L(0r9) Fa_.zaauo22+(aq) + 2350051 109 (2) Although the o for Equation 2 is unknown, theory and experience predict that 238y will concentrate in the aqueous phase. The constructive nature of the two processes might, therefore, be expected to result in an « suitably large to be the basis of a uranium isotope enrichment process. ' From a chemical standpoint, several problems immediately appear as critical ones. Obviously, one needs an extractant which will separate U(IV) and U(VI). It - must operate under some very specific conditions set by other portions of the system. - In order to form the basis of a useful process, the electron exchange reaction in Equation 1 must have a half-time, t,, on the order of a few seconds. Also, the exchange reaction shown in Equation 2 must be rapid. Both these reactions must, therefore, be well understood. Finally, it must ‘be demonstrated that a sufficiently large a ex1sts under these conditions. ' Based on these exchange reactions and based on a reasonable value of a (between 1.0014 and 1.002), countercurrent liquid extractors can be set up into a cascade arrangement. Further assuming that the exchange reactions and the o are independent of the relative concentrations of 235U and 238y, estimates of the equilibrium time to achieve 3% enrichment range from approximately three months to one year. To achieve 90% enrichment, the equilibrium time may range from 3 to 30 years. - } ik ey i ™ 3 s (- — — | Sy o o o S 2 e el ! Aerodynamic Methods Both the separation nozzle and the stationary-walled centrifuge can be classed as aerodynamic processes. These are considered to be competitive processes by their proponents and plans for their implementation are well advanced. Research efforts have been directed at several other aerodynamic methods such as the vortex tube, the separation probe, crossed beams, velocity slip and the jet membrane. None of these appear at the present time to offer the promise of the two aforementioned aerodynamic processes, although an expanded effort is proceeding on the jet membrane process. Commonly known as the Muntz-Hamel process, it involves the penetration of a stream of UFy gas into an expanding jet of easily condensible carrier gas. The Tighter 235UF; molecules penetrate the jet more easily than the heavier 238UF, molecules. A tube placed on the axis of the jet collects the enriched UFg. The depleted UFg flows out of the other end of the scattering chamber, after the carrier gas is separated from it by condensation. Plasma-Based Processes Since a plasma can be made to rotate at speeds greater than that of an ultra- centrifuge, it occurred to various investigators that such high speed gas rotation without the use of revolving equipment might possibly be developed into a more efficient isotope separation process than that based on a mechanical centrifuge. Five papers on this topic were presented at the International Conference on Uranium Isotope Separation in London in March 1975. The authors' assessment of the prospects for such a process ran the gamut from highly optimistic-—technoTogy is simple and well known so that minimal development will be required--to pessimistic--a rotating plasma process cannot possibly be economically competitive. To our knowledge, no one has separated uranium isotopes by means of the plasma centrifuge. Since that time, several other plasma-based processes have been proposed. Of all these processes, the currently most feasible seems to be the Plasma Ion Enrichment process (the Dawson separation process). In this process a plasma of UFg (or of uranium atoms) within a strong uniform magnetic field is exposed to a low energy radio-frequency wave resonant with the cyclotron frequency of the 235UF¢ ions. The rotation thereby imparted preferentially to the 235UF. jons enables the 235U to be separated from the 238U by properly placed collection plates. ~ This method has been used successfully to enrich macroscopic samples of po- tassium. 16 The collector was a cooled tungsten ribbon having_a voltage bias to collect selectively the excited ions. The potassium vapor was contact ionized at the entrance to the mass spectrometer. To eliminate spurious effects, samples were collected under three conditiohs of rf excitation: (1) no rf; (2} excitation at the 39K cyclotron frequency; and (3) excitation at the “*1K cyclotron frequency. The A-18 resulting ratios of %1K/39K abundance as measured by the mass spectrometer were, respectively, 0.07 (the natural abundance), 0.02 and 4. The abundance ratio of 4 corresponds to a more than tenfold enrichment of “1K. In addition to potassium ipns, work has been done on neon, argon, xenon and uranium toward resolving the ion cyclotron resonances for individual positive ions. The work with uranium is proceeding toward estimates of realistic operating parameters (ion densities, magnetic field strength, isotopic excitation energies, device length, jon temperatures, and collector types). A second process involves the achievement of a UFg plasma by chemi-ionization. UFg molecules are accelerated by expansion with an inert carrier gas through a supersonic jet:. A cross beam of alkali metal molecules results in the formation of NA* or Cs* and UFg~. A radio-frequency quadrupole mass filter deflects the 238UF, out of the plasma beam, permitting the separation of the two isotopes by collection of the two beams on separate baffles cooled by liquid nitrogen. This process seems to have less potential than the first. | " Comparison of Advanced Separation Processes The estimated costs of the processes mentioned are compared in Table A-3 with that of gaseous diffusion. With two exceptions, the table is based on process evaluations made by the Nuclear Division of the Union Carbide Corporation!’ for ERDA. For the exceptions, which are the FRG's separation nozzle and South Africa's stationary-walled centrifuge, the comparison is based on published statements by the developers of the process. Of all the processes listed, only the costs for the centrifuge, and possibly for the separation nozzle, are known with any degree of certainty. e L [ O K = e r- o i | | b o ™ Table A-3. ‘Comparison of Process Economics ’ Operating Specific Costs Capital Power Other Than Investment Cost Power Centrifuge ' > < > Separation Nozzle* < > = Stationary-Walled Centrifuge* = = ? LIS-Atomic < < > LIS-Molech]ar ' | | < < > Ch. Exchange: UIV(aq)-UVI(org) | = < > Other Aerodynamic Processes > > = Plasma: Chemi-ionization > < > Plasma Ion Enrichment (Dawson Process) o< < > *Based on estimates made by the process developers. DEFINITION OF SYMBOLS: = Approximately equal to the diffusion process. >,< Greater than or ‘less than the diffusion process, respectively. ? Unknown. References for Appendix A E. B. Kiser, Jr., "Review of U.S. Gas Centrifuge Program," AIF Fuel Cycle Conf. '77, Kansas City, Mo. (April 1977). . USAEC, "AEC Gaseous Diffusion Plant Operations," OR0-684 (Jan. 1972). H. D. Smythe, Atomic Energy for Military Purposes, Princeton University Press, Princeton, N.J., Aug. 1945, W. E. Groth,'K. Beyerle, E. Nann, and K. H. Welge, "Enrichment of Uranium Isotopes by the Gas Centrifuge Method," 2nd Int'1l. Conf. on the Peaceful Uses of Atomic Energy, Geneva, Switzerland (Sept. 1958). J. Los Kistemaker and E. J. Z. Veld Huyzen, "The Enrichment of Uranium Isotopes with Ultra-Centrifuges," 2nd Int'l. Conf, of the Peaceful Uses of Atomic Energy, Geneva, Switzerland (Sept. 1958). ?. Z;ppe, "The Development of Short Bowl Ultra Centrifuges," OR0-315 (June 15, 960). o K. Cohen, The Theory of Isotope Separation, McGraw-Hi1l Book Company, New York, 1951. 8. 10. 11. 12, 13. 14. 15. 16. 17. A-20 W. R. Voigt, “Enr1chment Po11c1es,“ AIF Fue] Cyc]e Cnnf '77, Kansas City, Mo. (Apr11 1977). E. Von Halle, "Summary Review of Uranium Isotope Separation Methods Other Than Gaseous Diffusion and Gas Centr1fugat10n," 0R0-690 (Feb. 23, 1972). A. J. A. Roux and W. L. Grant, "Uranium Enrichment in South Africa," presented to European Nuclear Conference, Paris, France_(Aprll 1975). Nuclear News, p. 80 (June 1977). S. Blumkin, “Survey of Foreign Enrichment Capacity, Contracting and Technology: Jan. 1976-Dec. 1976," K/OA-2547, Pt. 4 (April 18, 1977). Letter, R. J. Hart, ERDA-ORO, to W. R. Voigt, ERDA-HQ, "Interim Uran1um Enrichment Long-Range Operating Plan," (August 9, 1977). M. Benedict, et al., "Report of Uranium Isotope Separatlon Rev1ew Ad Hoc Committee,® OR0-694 (June 2, 1972). _ Allas S. Krass, "Laser Enrichment of Uranium: The Proliferation Connection,” Science, Vol. 196, No. 4291, p. 721-731 (May 13, 1977). M. Dawson, et al., "Isotope Separation in Plasmas Using ITon Cyclotron Resonance, TRw Defense & Space Systems, Redondo Beach, Calif. pP. R. Vanstrum and S. A. Levin, "New Processes for Uranium Isotope Separation," TAEA-CN-36/12 (11.3), Vienna, Austria (1977). L: Tom o . it [ oy K —C - — e e £ ror— T — K Appendix B. ECONOMIC DATA BASE USED FOR EVALUATIONS OF NUCLEAR POWER SYSTEMS M. R. Shay, D. R. Haffner, W. E. Black, T. M. Helm, W. G. Jolly, R. W. Hardie, and R. P. Omberg Hanford Engineering Development Laboratory The economic data base used in the assessment of the impact of denatured fuel cycles in the various nuclear systems options described in Chapter 6 was jointly developed by Combdstion Engineering, Oak Ridge National Laboratory, United Engineers and Constructors, Argonne National;Laboratory, Resource Planning Associates, Hanford Engineering Development Laboratory, DOE Division of Uranium Resources and Enrichment, and DOE Division of Nuclear Research and Applications. The data base includes capital costs, operation and maintenance costs, fuel fabrication and reprocessing costs, capacity factors, money costs, and uncertainties. The deflated and present-valued capital costs for LWRs, SSCRs, HTGRs, CANDUs, and FBRs, excluding interest during construction, are shown in Table B-1. The same capital costs | including interest during construction are shown in Table B-2. In either case, the stream of expenses incurred during the construction of the plant is discounted to the date of startup and is measured in dollars of constant purchasing power. The uncertainty ranges included in Table B-2 represent current best estimates of the most probable variations in capital costs. For flexibility, the uncertainties are expressed relative to the reference LWR capital cost. Table B-1. Capita] Costs of Power Plants The operation and maintenance costs Excluding Interest During Construction for the same power plants are shown in Power Plant Type Costs ($/kile)” Table B-3. The higher costs for the SSCR and the CANDU over the standard LWR are due LWR | 500 | to the heavy water replacement requirement SSCR 520 + 39 (for D,0) = 558 and the necessjty for performing some HWR 605 + 156 (for D,0) = 761 maintenance in atmospheres containing HTGR 560 to 580 tritium. Additional minor reactor costs FBR 625 to 875 are given in Table B-4, *Based on 7/1/76 dollars. Table B-2. Captial Costé of Power Plants Including Interest During Construction Power Plant _ Cost Cost Relative © Cost . Type = - ($/kWe)* ' - to LWR Cost " Uncertainty LWR o 6285 95% to 105% reference cost SSCR 650 + 40 (heavy water) = 690 #103 -~ 105% to 120% of LWR cost HWR 755 + 160 (heavy water) = 915 +46% ~120% to 150% of LWR cost HTGR 715 | +148 105% to 125% of LWR cost FBR 800 +28% 125% to 175% of LWR cost *Based on 1/1/77 dollars. B-2 The fuel fabrication costs for the various reactor types are shown in Table B-5 as a function of time’beginning with the expected introduction date for a particular reactor and fuel design. If a particular reactor and fuel design should prove successful, fabrication costs should decrease as larger plants with higher throughput rates are constructed. The decrease in fabrication costs over the first decade after introduction is simply indicative of a transition from small fabrication plants with high unit costs to larger fabrication piants with lower unit costs. These costs are a strong function of the fissile isotope and a weak function of the fertile isotope. The sensitivity to the fissile isotope is caused either by the spontaneous fission associated with high-exposure'fiSSilé plutonium or by the gamma activity associated with high-exposure 233, The costs are based on the assumption that fuels containing 235U are fabricated on a line with contact operation and contact maintenance, fuels containing fissile'plutonium are fabricated on a line with remote operation and contact maintenance, and fuels containing 233 are fabricated on a line with both remote operation and remote maintenance. The expected variations in fuel fabrication costs (cost uncertainties given in footnote b of Table B-5) represent the hpper and lower - cost boundaries anticipated for fabrication costs and are'expressed as percentages. For example, the expected fabrication cost for plutonium-bearing LWR fuel with uncertainties applied ranges from $306 per kg HM (-10% of Feference).to $510 per kg HM (+50% of reference) for year 2001 and beyond. Table B-3. Power Plant Operation and Maintenance Costs {=.[F1'xed +(Variable x Capacity Factora)]xPower}' ' _ Fixed Cost : Power Plant Type ($/kWe-yr)? Variable LWR 3.6 1.9 SSCR 4.8 1.9 ' HR 8.4 1.9 HTGR 3.6 1.9 FBR 4.1 - 2.3 “See Table B-9 for capacity factors.\ Pgased on 1/1/77 dollars. Table B-4. Minor Reactor Costs Property Insurance Rate 0.0025 Capital Replacement Rate 0.0035 Nuclear Liability 58 x 10% $/yr The expebted reprocessing costs are shown in Table B-6. These costs were obtained by estimating the capital and operating costs associated with each of five stages of the - reprocessing process. The stages were: headend, solvent extraction, product conver- sion, off-gas treatment, and waste treatment. The costs are shown as a function of time reflecting the transition from a new industry consisting of small plants with high unit costs to a mature industry consisting of larger plants with lower unit costs. The expected costs for spent fuel shipping, waste shipping, . and waste storage are also included in Table B-6, as well as the total costs for all these processes. The total cost uncer- tainty factor for all fuel types is estimated to be a 50% increase for the reference values. Thus, the total reprocessing cost for LWR fuel with the uncertainty included ranges from $220 to $330 per kg HM for year 2001 and beyond. It should be noted that it is assumed here that a policy decision will have been made in tjme for the first reproceséing plant to be in operation by 1991. Al1 fuel discharged from the reactor prior to this date is - ' b i r— r— F— = - e e — o r-" 2 r o . r B-3 Table B-5. Reactor Fuel Fabrication Costs? Reactor Type LWR-US(LE) /U LWR-US(DE)/U/Th LWR-U3(DE)/U/Th LWR-Pu/U LWR-Pu/Th SSCR-US{LE)/U SSCR-U3(DE)/U/Th SSCR-Pu/Th HWR-US{NAT) /U HWR-U5(SEU) /U Cost ($/kg HM)? Over First Decade After Introduction 100 (1969 - 2089)° 230 (1987) -~ 140 (1997) 880 (1991) ~ 550 (2001) 550 (1991) -~ 340 (2001) 550 (1991) ~» 340 (2001) 100 (1991 - 2089)¢ 880 (1991) -+ 550 550 (1991) » 340 (2001) (2001) 60 {1995 - 2089)¢ 60 (1995 - 2089)° HWR-U5{DE ) /U/Th 140 (1995) » 85 (2005) HWR-U3(DE)/U/Th 560 (1995) + 350 (2005) HWR-Pu/U 320 (1995) + 200 (2005) HWR-Pu/Th 320 (1995) > 200 (2005) HTGR-US(LE /U HTGR-US5 (DE)/U/Th HTGR-U5(HE)/Th C/Th + U = 150 340 21995) + 210 {2005) C/Th + U = 238 500 {1995) + 300 (2005) C/Th + U = 335 660 (1995) » 400 (2005) C/Th + U = 400 760 (1995) ~ 470 (2005) C/Th + U = 650 1220 (1995) + 770 (2005) HTGR-U3(DE)/U/Th HTGR-U3/Th C/Th + U = 150 860 (1995) + 470 (2005) C/Th + U = 238 1220 (1995) + 670 (2005) C/Th + U = 335 1640 (1995) + 900 (2005) C/Th + U = 400 2000 (1995) - 1100 (2005) C/Th + U = 650 3200 (1995) + 1750 {2005) HTGR-Pu/Th | C/Th = 238 1220 (1995) ~ 670 (2005) © 1750 (2001) + 950 (2011) 1750 (2001) + 950 (2011) 3000 (2001) + 1650 (2011) 35 (2001) » 25 - (2011) FBR-U radial blanket - 250 (2001) -+ 150 - (2011) FBR-Th axial blanket 35 (2001) » 25 (2011) . FBR-Th radial blanket 250 {2001) -+ 150 (2011) FBR-Pu-U core FBR-Pu-Th core FBR-U3-U core FBR-U axial blanket Fabrication costs based on the following: for LWR and SSCR, a 17 x 17 pin assembly (374-mi1-0D pin}; for the HWR, a 37-pin CANDU assembly ~20 in. long ~ (531-mi1-0D pin); for the HTGR, standard carbon- ~coated uranium carbide fissile microspheres formed “into cylindrical rods located in a hexagonal gra- - phite block; and for the FBR, a 217-pin assembly in a hexagonal duct (310-mi1-0D pin). : bUncertainities on fabrication costs: 235U-bearing fuels, no uncertainty; Pu-bearing fuels, -10% to 50% -increase; 233U-bearing fuels, -10% to. 50% increase. ®Costs assumed to remain constant. assumed to have been stored, with the spent fuel stockpilie being reduced in an orderly manner after the advent of repro- cessing. After the spent fuel stockpile has beén reduced to zero, the out—of-reactor time required for reprocessing and refab- rication is assumed to be two years. The long-run marginal costs estimated for U305 ore as a function of the cumulative supply are shown in Table B-7. As noted in Chapter 6, the U30g estimates have been provided by DOE's Division of Uranium Resources and Enrichment (URE), the high- cost supply being based on the assumption that approximately 2.5 million tons of U304 will be available from conventional uranium ore resources and the intermediate-cost supply being based on the assumption that approximately 4.5 million tons of Uj04 will be available. In either case, it is assumed that shales can be mined after the conventional resources are depleted. The. cost of extracting the shales increases from $125/1b to $240/1b for the high-cost supply case and from $100/1b to $180/1b' for the intermediate-cost supply case. It is important to note that the long-run marginal costs shown in Table B-7 are larger than the forward costs shown in Table 6.1-1 of Chapter,G_because the ]ong-run marginal costs contain the capital cost of facilities currently in operation, plus a normal profit for the.industry. The long-run marginal costs are more appropriate for use in a nuclear strategy analysis. The enrichment costs and tails compositions assuming either a continuation of the gaseous diffusion technology or the deployment of an advanced enrichment tech- . nology are shown in Table B-8. It was ‘assumed that if the gaseous diffusion technology is continued the tails composi- - tion will be stabilized at 0.0020 and that the cost of enrichment will increase to $80/SWU in 1987 and remain constant there- after. If an advanced enrichment technology B-4 Table B~6. Reprocessing, Shipping, and Waste Storage Costs for Various Reactor Types Costs ($/kg HM) Reactor Reprocessing Costs Spent Fuel Waste Shipp1ng Waste Storage Total Costs Type Over First Decade? Shipp1gg Costs . Costs Over First Decade Costs After Introduction® LWR 225 (1991) » 150 (2001) 15 10 45 295 (1991) - 220 (2001) SSCR 225 (1991) -+ 150 (2001) 15 10 45 295% (1991) - 220 (2001) “HWR 225 (1995) +~ 150 (2005) 10 5 15 255 (1995) - 180 (2005) HTGR 800 (1995) + 400 (2005) 85 35 65 985 (1995) + 585 (2005) FBR 500 (2001) + 200 (2011) 80 50 115 745 (2001) + 445 (2011} “Fissile storage costs after reprocessing = $2/g-yr for ?¥%( and fissile plutonium. Tota] costs for throwaway cycle are spent fuel shipping costs plus $100/kg HM. cSO% uncertainty on total costs for all reactor types. is deployed, the tails composition would decrease continuously from 0.0020 to 0.0010 between the years 1980 and 2000 as the installed capacity of the advanced technology increased, and the cost of a unit of separative work would decrease to approximately 60% of that of the gaseous diffusion process., It was also assumed that the tails composition‘WOuld,further decrease from 0.0010 to 0.0005 between the years 2001 and 2030 due to improvements in technology, while the cost of a unit of enrichment would remain constant during this period. ‘The tails composition and enrichment cost were assumed to remain constant thereafter. The capacity factors of a plant throughout its 30-yr lifetime are shown in Table B-9. The capacity factor increases from 60% to 72% during the first 3 yr of operation and remains at 72% during the subsequent 14 yr. It then decreases continuously as the forced outage rate increases and as the plant is shifted from a base-load unit to an intermediate-load unit. The long-term real cost of money to the electric utility industry is shown in Table B-10. These costs were developed by analyzing the deflated cost of debt and equity to the industry over the past 30 yr. The long-term deflated cost of debt has been 2.5%/yr and the long-term deflated cost of equity has been 7.0%/yr. Assuming the industry to be funded at approximately 55% debt and 45% equity, the long-term real money cost is approximately 4.5%/yr. The combined effects of capital, fuel fabrication, and reprocessing (or permanent disposal) cost uncertainties on the levelized total power costs for individual'reactorrand fuel cycle options are shown in Fig. B-1. These costs represent typical nonfuel components whose uncertainties are easily quantified. Figures B-2a and B-2b show the relationship of total power costs to the Uj0g price for four reactors on the throwaway fuel cycle. The sensitivity of the total power costs to the U308 price was analyzed first by assuming that the price remained constant over the 30-yr life of the reactor, and second by assuming that ‘the price increases in relation to the rate of consumption (see Fig. B-3). Thus, the total power costs in Fig. B-2b are given for a reactor starting up with the U;30g price shown on : — r- Ko o ~— r—r/ - ! i i ! ~ | Table B-7. Marginal Costs of U30g as Table 8-8. Tails Composition and r— e - - r— sl a Function of Cumulative Supplya.b Long-Run . . Tails Quan%agytgzsu303 Marg1n?; Cost Composition - (10° tons) —(81b) Time (235( Fraction) Cost{$/SWU) Intermediate-Cost U30g Supply above 6,5' - 240 Enrichment Costs Gaseous Diffusion Technology 0.0 - 0.25 14 0.25 - 0.75 23 1969 to 1976 0.0020 50 0.75 - 1.25 33 1977 to 1986 0.0020 75 1.25 - 1.75 44 1987 to 2089 0.0020 80 1.75 - 2. 53 Advanced Technology 2.5 - 3. 61 1969 to 1976 0.0020 50 3.5 - 4.25 80 1977 to 1980 0.0020 75 4.25 - 4.75 107 1981 to 2000 0.0020 to 0.0010 75 to 55 375 = 5.25 128 2001 to 2030 0.0010 to 0,0005 55 5.25 - 5.75 143 0 ' o7 - 2031 to 2083 0.0005 55 5.75 - 6.0 165 - 6.00 - 8.5 165 above 8.5 180 Table B-9. Plant Capacity Factors High-Cost U30g Supply Year CF(%) Year CF(%) 0.0 - 0.25 14 . 0.25 - 0.75 24 1 60.0 20 65.7 0.75 - 1.25 35 2 66.0 21 64.1 1.25 - 1.75 | 54 3 72.0 22 62.6 1.75 - 2.25 84 4 72.0 23 61.0 2.25 - 2.75 128 24 59.4 2.75 - 3.00 158 : : 25 57.9 e —mm—emmcee—memeee—em—eeeee—m—ene 15 72.0 26 56.3 3.00 - 3.25 158 16 72.0 7 N 3.25 - 3.75 173 17 70.4 . 28 63.1 3.75 - 4.25 - 180 18 - 68.9 29 51.6 4.25 - 4.75 | 180 19 1 67.3 30 50.0 4.75 - 6.5 210 ' Y Table B-10. Long-Term Real Costs of Money 2For those cases in which plant selection Debt Interest 2.5% was determined by uranium utilization a limit . ‘ ‘ of 3 million tons of ore are assumed at Equity Interest 7.0% below $150/1b U305 for the high-cost U;0g ‘ supply and 6 million tons for the inter- Fraction Debt 0.55 mediate-cost supply. , : . bcost of converting Us0g to UFg = $3.50/kg Fraction Equity 0.45 of U. Effective Interest Rate 4,525% B-6 - _ $ ut %23— I T 2 - A - 3 ol O - I N @ sl g S v 5 ] 5 1} I e 15 L REACTOR . , OPTILONS: LWR SSCR HWR HTGR FBR / r v """ NN 1 o i ' ' 1 I P 1 M ( I 1 I g8 g b we 3 2 S w. 3 28w 28w, 3 AL l ,xw w 1 i o Wt 1 ad i 1 w al e o« [-*3 FUEL CYCLE 826 80 g0 9p &.+ 820 89 8o &,» 930 ©9 Oy &,« 836 08 90 8,4 82 O & ORTIONS: 8%x 89 20 8 82¥ 8%x 88 <9 8P 8%y 8 0 el 82k 8@ £ 8yl oF et -3 e a“ S“ WOy wrEz W= Sz YOy W3 e E“‘ rPOS - E e S""‘ y Ou g_g g’_fl 21 Dy &1 d: D08 D:T D £1 DUE Di 4 ‘55‘16 — ~ X o ) 0 ,]\‘ 1 } 1 I I . |. 0 40 60 80 100 120 140 160 U308 PRICE AT STARTUP, $/1b | o HEDL 7805-090.17 Fig. B-2. Effect of U305 Price on Total Power Cost for Reactors Operating on L| U Throwaway Cycle. . B-8 —— 1 1 i | 1 1T _ -1R - —43 S - e Q. o < oy - - - y Y o e 2 = « & - -2 m% -+ = Q o o= & o . 8 & . o ™ Lo o~ 1 o0 .n_.u w e .lw ] aVv$ ‘3o1d 8ofn r— . - U~ i i 1 C-1 Appendix C. DETAILED RESULTS FROM EVALUATIONS OF VARIOUS NUCLEAR POWER SYSTEMS UTILIZING DENATURED FUEL M. R. Shay, D. R, Haffner, W. E. Black, T. M. Helm, W. G, Jolly, R. W. Hardie, and R, P. Omberg Hanford Engineering Development Laboratory This appendix presents detailed results from the calculations performed for the economic/resource evaluation of denatured nuclear reactors operated in concert with other reactors to form nuclear-based power generation systems. For purposes of comparison, it also presents results for similar systems that do not utilize dehatured fuel. As pointed out in Chapter 6, nine different nuclear policy options were examined with four cases under each option. The resulting cases can be classified as shown in Table C-1, where the letters L, S, G, and H indicate the thermal converter option employed in each case. For all cases jdentified with an L, the only converters used are LWRs. For cases identified with an S, SSCR converters are used in addition to LWRs. Similarly, for cases identified with H and G, the converters used are HWRs and HTGRs respectively, both again in combination with LWRs. Under Options 3, 6, 7, and 8, FBRs are also included in the nuclear systems, In addition to these 36 caees, Case 1L was recalculated for e'standard LWR alone; that is, the LWR with an extended discharge'exposure, which is included in Case 1L, was eliminated from the system. This case is identified in this appendix as Case IE. Table C-1. Nuclear Policy Options*' Options LWR SSCR HTGR HWR Throwaway Option (1) 1L 1S 16 1H Pu/U Options w1th Converters On]y (2) 7 2L 2S - 26 2H " With Converters and Breeders (3) S . 35 36 3H Denatured Uranium Options'with ' Converters Only Plutonium Throwaway (4) _ . 4L 45 , ;.4G 4H - Plutonium Miminization (5U) =~ 5UL SUS 5U6 UM Plutonium "Transmutation" (5T) o 5TL . BTS 576 5TH Denatured Uranium Options with Converters and Breeders Light "Transmutation” Rate (6) = 6L 65 66 6H ~Light "Transmutat1on“ Rate, Denatured S e ' “ ' . Breeder (7) , o Lt - 75 16 7H Heavy "Transmutatton“ Rate, Denatured ST LT Breeder (8) o 8L 8 - 8G- 8H *See Table 6.1-5 in Chapter 6 and Tables C-2 and C-4 in this append1x for identification of specific reactor types in each case. In all cases the reactors operating on plutonium or on highly enriched uranium were assumed to be restricted to secure energy centers, while those operating on Tow-enriched, slightly enriched, natural, or denatured uranium were permitted to operate outside the centers. The specific reactors used for each case, and their locations, are given in Table 6.1-5 of Chapter 6. - | | All cases were run assum1ng 350 GWe of installed nuclear capaczty in the year 2000 and a net 1ncrease in installed capacity of 15 GWe per year thereafter. Each new p]ant was assumed to have a 30—yr lifetime. For Option 1, some addit1onal cases were run for a lower energy demand -- 200 GWe in the year 2000 and a net increase of 10 GWe per year thereafter These latter cases are identified with a C following the case number (i.e., cases ILEC, 1LC, etc.). In ;he results presented here, particular emphasis is given to uranium utilization, separative work utilization, and energy-support ratios. Two important criteria are to be ~considered when anaTyz1ng uran1um ut111zat1on of reactor systems The first is the ab111ty of the system to meet the Spec1f1ed nuclear energy demand with the availab1e Us0g supply For these calculat1ons two different supplies were assumed: 3 million and 6 million ST below $160/1b U30g, corresponding to a high-cost and an intermediate-cost supp1y, respect1ve1y (As shown in Append1x D, nuclear power p]ants do not compete well at higher U30g costs.) The second criterion is the capability of the uranium industry to discover, mine and mill the ore at a rate adequate to satisfy the demand for uranium. The specification of the overall maximum production rate is difficult to postulate because of the possibitity of importing Us0g and because of the difficulties that might be encountered in developing uncertain resources. As pointed out in Section 7.4.4 of Chapter 7, the DOE Uranium and Enrichment Division has estimated that by developing known and potential reserves domestic mining and milling could sustain 60,000 ST of U30g per year. When analyzing enrichment utilization, the same two criteria - total amount and enrich- ment capacity - were also used, the more meaningful being the capacity since enrichment is not a limited natural resource like uranium. For the cases in which 3 million ST of uranium below $160/1b U305 was assumed, the lack of low-cost U30g dominates-the plant selection because the amount of ore available is inadequate for meeting the projected nuclear energy demand. As a result, resource-efficient reactors are constructed regardless of their cost. With a U30g supply below $160/1b as large as 6 million ST, however, most systems are no longer dominated by the lack of Us30g, and the relative total power costs of the individual reactors play a more important role. In fact, if the system is not Timited in any way by the supply of Us0g, then the solution is determined solely by economics. The results in this case become more tenuous because of the uncerta1nty 1n cap1ta1 costs, fabrication costs, reprocess1ng costs, etc The cumulative nuclear capacities that could be constructed through the year 2050 for the various cases are shown in Table C-2. Only those cases totaling 1959 GWe will have r- e ro T oo —C.— " " c . r. o U r. o r - T Table C-2. Cumulative Nuclear Capacity Built Through Year 2050 with Various Nuclear Policy Options (Adequate Capacity = 1959 GWe) Advanced ' Option Capacity (GMWe) Converter Option 1E % 1 2 3 2 5U 5T 6 7 8 High-Cost U30g Supply L?E;S 572 594 953 1959 945 1205 ;027 1959 1959 1647 S%g?'s - 607 1043 1959 1071 1423 1275 1958 1959 1943 H?g;s - 667 987 1959 1334 1747 1505 1959 1959 1859 H{g?'s - 603 1417 1959 855 1064 1004 195¢ 1959 1791 Intermediate-Cost Us0g Supply L?E;s 1135 1193 1783 1959 1852 1921 t864 1959 1959 1856 S?g?'s - 1271 1937 1959 1943 1959 1959 1959 1959 1959 H?fiis { | 1497 1921 1959 1943 1559 1959 ;959 1959 1953 HTGR's - 1320 1959 1959 1794 1924 1844 1958 1959 1658 (6) *System with standard LWR only. met the projected nuclear demand under the criteria of an installed capacity of 350 GWe in year 2000 and an increase of 15 GWe per year thereafter.* With the high-cost U30g supply: some of the systems fall far short of satisfying the demand; in fact, the only nuclear systems that fully meet the demand are those including FBRs (Options 3, 6, 7, and 8). The throwaway option, in particular, builds 1ess than a third of the desired nuclear plants. Of the cases that do not 1nc1ude FBRs, those employlng HWRs come closest to meet1ng the demand. One HTGR case (ZG) is also clearly superlor to most of- the other cases. This is to be expected since Case 2G includes tradttiona} HTGRs that are fue]ed w1th h1gh1y enriched 235U and also with ‘*233U/Th. A doubling of the economic Ugoa"supply to 6 m11110n'tons allows ‘many more_nuclear system opt1ons to meet the proaected nuclear energy demand. In fact, only the throwaway opt1on ‘has cases that don't even come close to satisfying the demand. None of the- Option 4 ‘cases meet the demand e1ther, however, Cases ‘4S and 4H are w1th1n 16 GWe of ‘the demand. - Al other systems have at 1east one.advanced converter option - that builds the desired 1959 GWe of energy. It shou]d be. emphasized that for the systems where the demand was met with the high-cost USOB (i.e., the systems- with FBRs), a doubling -of the ore supply means ~ that the ore supply is no Ionger the sole constra1nt and plant selection is based on economics. *NOTE: ~ Since this is a 50-year span, some'of_the reactors built in the first few years will have been decommissioned after having operated 30 years.. Table C-3, Utilization of U30g Ore and Enrichment Through Year 2050 with c-4 Various Nuclear Policy Options ’ cfig:;gigfi : U 05 Utilization {tons USOBIGNe)/Enrichment Utilisation (million SWU/GWe) Option 1E* 1 2 3 4 5U 5T 6 7 8 High-Cost U305 Supply LWR's 5236 5042 3138 1497 3165 2480 2908 1512 1514 1525 (L) 3.08 - 3.08 2,03 0.92 2.70 2.12 2,08 1,03 1,03 1,17 SSCR's - 4931 2864 1492 2793 2098 2340 1487 1487 1528 (s) - 2.83 1.7¢ 0.87 2.38 1.78 1.59 . 0.85 0.95 1.01 . HWR's - 4489 3027 1391 2243 1707 1983 1345 1314 1520 (H) - 2.18 1,37 0.99 1.78 1.33 0.90 0.96 0.94 1,00 ' HTGR's 4963 2105 1505 3497 2807 2974 1503 1496 1666 (6) 3.10 1.71 1.1§ 2.75 2.22 2.10 1.02 1.01 1.20 - Intermediate-Cost U0y Supply IWR's 5236 4973 3188 - 2758 3103 2957 3037 2733 2733 2798 v 2,95 2.82 1.7§ 1.45 2.4¢ 1.86 1.77 1.58 1.58 1.61 SSCR's - 4657 2820 2n1 2844 2511 2511 2511 2511 2511 (s) - 2.43 1.36 1.27 2,03 1.3¢ 1.34 1,34 1,34 1.3 HWR's - 3916 2894 1398 3030 2431 2475 2195 1392 1924 (H) - 1.40 1.22 1,00 2.10 1.56 1.58 1.32 0.99 1.23 HTGR's - 4478 2683 2680 3172 2865 3055 2683 2682 2698 - 2.89 1.60 2.21 1.77 1.77 1.58 1.58 1.62 (6) 1.81 *System with standard LWR only. Uranium and enrichment utilization for the various cases are shown in Table C-3. The uranium utilization values are the total amount of uranium consumed plus the forward commit- ment per GWe of nuclear power constructed through the year 2050. values are the total amount of separative work units required through the year 2050. _ As pointed out above, for the cases for which only 3 million ST U30g was assumed to .be available below $160/1b, the ore is the limiting factor. Comparing Case 1LE with Case 1L The enrichment utilization gives the savings in ore on the throwaway cycle as a result of introducing the extended .Cases 1L, 1S, 1H, and 1G exposure LWR -- less than 4% in ore and none in enrichment. _ compare the relative ore and enrichment utilization of the various advanced converter options ~on the throwaway cycle. The HWRs clearly offer the greatest savings in both ore and enrichment. requirements by almost 30%. In contrast, the SSCRs only offer a 2% ore savings and an 8% 4 The HTGRs reduce the ore usage by less than 2%, with about the same The impact on ore utilization of the SSCR, HWR, and HTGR advanced The reason for the minimal enrichment savings. enrichment requirements. converters on the throwaway cycle is less than might be expected. Compared with LWRs, the HWRs reduce ore requirements by over 10% and SWU effect is because most of the 3 million ST of U305 has already been committed to LWRs before enough advanced converters can be built to have much influence. Allowing the recycle of fuel in thermal reactorsl(Option 2) results in significant savings in ore compared to the throwaway cycle ~- almost 60% for the HTGRs and from 30 to 40% for the other converters. For this nucleaf policy option and the high-cost U305 supply, the HTGR clearly has the best ore utilization, although the HWRs have better enrichment utiliiation. ety » e C—~ 0 r— r— - k; e T o o Y ; —ry r- - o o r-. e C-5 The introduction of the classical Pu-U/U FBR in Option 3 results in an additional ore and enrichment savings of about a factor of two from that in Option 2 except for the HTGRs. Note, however, that in Option 2 the HTGRs already had a low ore and enrichment usage. In Option 3 all the advanced converter cases have about the same usage. Recycling uranium in denatured reactors and throwing the plutonium away (Option 4) - requires enrichment about halfway between Options 1 and 2. Compared with the classical recycle of plutonium in thermal reactors (Option 2}, Option 4 consumes roughly the same quantity of uranium with LWRs and SSCRs. That is, the increased worth of 233U in LWRs and SSCRs is nearly balanced by throwing away the plutonium. The requirements for HWRs, however, are considerably reduced over those of Option 2 when 233U is recycled compared to recycling plu- tonium. The very low fissile requirements for the denatured 233U HWRs is responsible for the more favorable Us0g utilization in Option 4 compared to Option 2. In contrast, the HTGRs in Option 4 look much worse than in Option 2. This is because the HTGRs were already operating on the 233U/Th cycle in Option 2. However, in Option 2 the uranium-fueled reactors all use highly enriched fuel while in Option 4 they use denatured fuel. Options 5U and 5T allow the recycle of plutonium in plutonium/thorium transmuters, the difference between the two being that denatured 235U reactors are available in 5U whereas they are not in 5T. This forces the 5T system to initially rely on the Pu/Th-fueled reactors for 233y, Compared to Option 4, Option 5U results in 20 to 25% savings in ore usage and Option 5T in 10 to 15% savings. The HWRs are the most efficient advanced converters for uranium and enrichment utilization for Options 5U and 5T, Option 6 introduces FBRs with thorium blankets, although these FBRs have uranium as fertile material in the core. Comparing Option 6 with QOption 3 reveals that both systems have approximately the same resource utilization. Option 7 is identical to Option 6 except the denatured 233U FBR is included. The impact of this reactor on resource utilization for these cases is small. ' ‘In Option 8 the Pu-U-fueled FBRs of Option 7 are replaced with Pu-Th-fueled FBRs. The longer doubling time of this reactor type results in somewhat increased uranium and enrichment requirements. A key point for all of the systems containing FBRs (Options 3, 6, 7, and 8) is that the ore and enrichment usage is relatively independent of the advanced converter option. This is in contrast to the nonbreeder systems where the type of advanced converter available (LWR, SSCR, HWR, or HTGR) much more strongly affects the resource utilization. Another very importent point that needs emphasis is that the superior ore utilization of the HWRs relative to the other advanced converters for the alternate fueled systems (Options 4 - 8) is directly dependent on the denatured 233U-fueled HWR. Of all the reactor designs, the desigh of alternate fueled HWRs have probéb1y received the least amount of analy- sis and therefore have the largest uncertainty. Thus, before it can be concluded that the HWRs offer significant resource savings, more work needs to be performed to verify the optimistic performance characteristics of the denatured 233U-fueled HWR. C-6 Since 6 million ST of U305 below $160/1b is adequate, or nearly adequate, to satisfy the projected nuclear energy demand for most cases in the various nuclear options, the power growth patterns for these cases are strongly influenced by economics as-well as resource utilization. Thus, as mentioned earlier in this appendix, the results for the cases based on the intermediate-cost U30g supply are subject to much larger errors because of large cost uncertainties. Table C-3 shows that the advanced converters for the throw- away cycle reflect a larger U30g savings when 6 million ST is used as a base rather than 3 million ST. This is because many more nuclear plants-are built with the larger supply and therefore more advanced converters can be built, resu]ting'in a larger impact. For . - the high-cost Us0g case, most-of the economic U;0g was already committed to the LWR before the advanced. converters could have an effect. - For Option 2, the results are about the same for both U305 supplies except for the case with HTGRs (Case 2G). Ore requirements per GWe are 27% higher for this case with the intermediate-cost U305 assumed to be available. This is because 6 million ST of economic U30g is an adequate amount of ore for the system of reactors in Case 2G to satisfy the nuclear energy demand and economic considerations are also affecting the mix of reactors that are built. Thus, the fraction of low-enriched LWRs constructed is larger because this reactor is less expensive than the HTGRs, even though the HTGRs use less. uranium. The plant selection for the cases that include FBRs (Options 3, 6, 7, and 8) -is also determined by economics when 6 million ST of U30g below $160/1b is assumed to be available. Therefore, the uranium utilization for these cases has less meaning. Similarly, some of the advanced converter options for the denatured cases (Options 4, 5U, and 5T) are resource Timited and some are not, so it is difficult to draw conclusions regarding relative uranium and enrichment utilization. To summarize, there are two important and competing effects when comparing the cases for the two uranium supplies: (1) For systems that fall far short of meeting the demand with the high-cost U30g supply, the larger supply allows the advanced converters to have a greater impact and therefore better ore utilization; and (2) systems that have almost enough ore with the high-cost U30g supply have plenty of ore with the intermediate-cost- supply, and therefore plant selection with the larger supply is based on cost and ore utilization is lower. The maximum annual U30g requirements and the maximum annual enrichment requirements through the year 2050 are shown in Table C-4, The number in parentheses next to each maximum indicates the year the maximum occurs. As was mentioned above, it has been estimated that the maximum domestic mining and milling rate may be approximately 60,000 ST/yr. Table C-4 indicates that if the high-cost U30g supply is assumed, the annual U30g requirements vary ~from 50,000 ST/yr (Case 7S) to 80,000 ST/yr (Case 4L). For most.of the cases, the maximum occurs during the first decade of tile next century. Thus, most of the cases require annual ore usage within the next 25 - 30 years that exceeds the 60,000/yr criterion. ' - - ' 1 r r— o et € . ——. rT— ™ - - O - c-7 Table C-4. Max1mum Annual U405 and Enrichment Requirements Through Year 2050 for , Various Nuclear Policy Options cgg:g’:.:gg U305 Requirements (thousands tons/yr)/Enrichment Requivements (million SWU/yr) Option 1E* 1 2 3 a 50 5T 6 7 8 +High=Cost U30g Supply LWR's 73(2007) 72(2007) 67(2009) 60(2009) 80(2005) 75(2009) 65(2011) 62(2009) 60(2009) 68(2005) (L) 44(2007) 45(2007) 46(2009) 41(2008) 69(2009) 6£5(2011) 45(2011) 44(2008) 42(2009) §55(2005) SSCR's - 72(2007) 62(2011) 52(2009) 79(2009) 69(2011) S8(2017) 50(2005) 50(2005) 55(2009) (s) - 42(2007} 40(2011) 34(2009) 68 (2009) 60(2011) 39(2010) 35(2005) 35(2005) 38(2009) HWR's - 68(2009) 58(2011) 66(2009) 71(2009) 55(2003) 53(2019) 64(2009) 63(2009) 65(2009) (H) - 36(2005) 36(2003) 46(2008) §58(2011) 46(2023) 35(2003) 46(2009) 44(2008) 46(2008) HTGR's - 72(2007) S7(2019) 53(2003) 65(2009) S57(2011) 64(2011) 61(2009) 60(2009) 65(2009) (6) - 45(2009) 51(2019) 33(2005) §2(2011) 49(2017) 45(2011) 44(2008) 42(2009) 46(2009) Intermediate-Cost U305 Supply LWR's 124(2025) 120(2025) 110(2039) 92(2037) 105(2037) 115(2039) 109(2039) 86(2033) 86(2033) 92(2043) (L) 74(2025) 77(2025) 72(2039) 60(2037) 100(2037} 90(2039) 77(2039) 61(2033) 61(2033) 65(2043) 114(2027) 96(2043) 93(2047) 82(2049) 83(2049) 83(2049) 83(2049) 83(2049) 83(2049) SSCR's - (s) - 63(2029) 57(2045) §3(2047) 73(2039) 55(2049) §55(2049) 65(2049) §55(2049) 55(2048) HWR's - 98(2031) 81(2023) 66(2009) 117(2031) 89(2029) 90(2029) 66(2009) 66(2009) 66(2009) (1) - 42(2009) 53(2011) 47(2009) 96(2033) 64(2029) 64(2031) 47(2009) 47(2009) 46(2009) HTGR's - 110(2029) 86(2049) 86(2049) 96(2039) 04(2043) 108(2041) 87(2047) 87(2047) B87(2047) (6) - 84(2029) 70(2048) 70(2048) 80(2039) 86(2047) 76(2041) 74(2047) 74(2047) 75(2047) *System with standard LWR only. The maximum annual separative work requirements based on the high-cost U,;0g supply varies from 34 million SWU/yr to 69 million SWU/yr. This means that the current separa- tions capacity would have to be doubled or quadrupled to meet the demand. As expected, the year in which the maximum separative work capacity occurs is nearly the same as the year when the U30g demand is greatest. AssumIng the 1ntermed1ate cost U305 supply, the max1mum annua] ore requ1rements are greater than 60,000 ST for all cases. For most of the opt1ons the year the maximum occurs is 40 yr later than for the high-cost cases. This is because, with 6 million ST of economic U305, the nuclear industry continues to expand The breeder reactor systems that include HWRs (Cases 3H, 6H, 7H, and 8H) are the only cases that have ore requirements that are close to be1ng as Tow as 60,000 ST/yr. The maximum separative work requ1rements are also very high for this uranium supply -- from 42 to 100 million SWU/yr. Table C-5 shows the energy support ratios calculated in this study for the year 2025, the energy support ratio being the ratio of installed nuclear capacity outside the energy centers to the installed nuclear capacity inside the centers. A1l the reactor types that are available in Options 1 and 4 could be constructed outside the centers; therefore, the energy support ratio for each case in these options is ». However, it has already been shown that these systems offer the lowest uranium_uti]ization and therefore the lowest nuclear growth potential, even if it is assumed that 6 million ST of U30g is available at below $160/1b. - Table C-5. Energy Support Ratios in Year 2050 for Var1ous Nuclear Policy Options (Support Ratio = Installed Nuclear Capacity Outside Energy Center/Installed Nuclear Capacity Inside Energy Center Advanced Support Ratio Converter - ‘ Option 1E* 1 2 3 ‘ 4 Su 5T 6 7 8 High-Cost U;05 Supply ‘ Li(iE;s ‘ - ® | 1.54 -0.72 o« 5.69 3.74 | 1.27 1.46 3.09 S?gl)l's - o 1.47 0.76 w 6.33 3.86 2.13 2.13 3.27 H'v(.’s;s - | = 0.49 0,92 @ 5.79 5'.07 1.07 1.06 2.89 HTGR's - o 0.24 0.24 o 4.02 2.50 1.26 1.28 3.11 (6) Intermediate-Cost U30g Supply L\gE;s ® = 2.42 1.65 ® 5.06 5.05 5.37 5.37 5.49 e?§§-s - o 2.10 1.65 o 4.78 4.78 4.78 4.78 4.78 Htal};s - » 1.85 0.94 - 4,03 3.84 ;.03 1.04 3.07 H{gr)a's - © 177 1.82 - 3.30 5.20 2.74 2.74 . 3.62 *System with standard LWR only. As pointed out previously, with only 3 million ST of U30g available below $160/1b, the only systems that satisfy the energy demand of 350 GWe in the year 2000 and 15 GWe/yr thereafter are those with breeders. The disadvantage of the classical Pu-U breeder cycle (Option 3), of course, is the low energy support ratio since the plutonium that is produced must be used in the energy centers. One technique for increasing the energy support ratio is to load thorium in the blanket of these breeders, while retaining plutonium and uranium in the cores. The 233U that is produced in the blankets is then burned in denatured LWRs located outside the centers (Option 6). The resulting energy suppoft ratios for Option 6 vary from 1 to 2, depending upon the advanced converter option. Option 7 introduces a denatured FBR which would provide ?33U to the system and therefore should increase its nuclear growth potentiel. However, since Option 6 can meet the projected nuclear growth demand itself, the addition of the denatured breeder in Option 7 actually had a minimal impact. The energy support ratios of Options 6 and 7 could be further increased by replacing the uranium in the core of the Pu-U breeder with thorium (Option 8). With the high-cost U3;0g supply, energy support ratlos of about 3 are obtained for this system. The intro- duction of thorium in the core of a breeder lowers the breeding ratio to the point that, in contrest to Option 7, significant quantities of FBRs operating on denatured fuel must be built to meet the projected nuclear growth demand. ' ' - ! r— r— g r— r— i, i e r— - r— r— c-9 In general, the energy support ratio trends for the various options are the same if i1 6 million tons of U30g is available below $160/1b; however, they are significantly higher, Lg largely because more low-enriched LWRs can be built. i: Selected detailed results for all the cases calculated are presented in Table C-6, C~7, and C-8. While many of the numbers in these tables appear elsewhere in this report, many numbers are also shown for the first time. For example, the plant mix in year 2025 and the levelized power cost for each plant starting up in the year 2025 are shown. The purpose of these tables is to group all the data together and also to provide sufficient data to help explain the behavior of the various reactor systems. (Note: Cases ILT and 1LTM in Table C-6 are for changing enrichment compositions; see Section 6.2-1 in Chapter 6.) £ © o 9 o o ror G fi F r C-10 Summary of Results for Cases'ASEUming High-Cost U305 Supply, 350 GWe *System with standard LWR only. Table C-6. ‘Installed Capacity in Year 2000, and 15 GWe Installed Capacity Each Subsequent Year BE* 1L 2 X 4 s s & L& Cmulative Nuclear Capacity mm ) () throus 859 579 834 1029 853 1005 916 1029 1029 1029 2049 ’ 5 594 953 1959 M5 1205 1027 1959 1959 1947 witu Costs ($B) 1977 through 2050 discounted at 4.5% 359 352 440 So7 473 480 439 510 509 b3 ) 7.5% ' 185 186 208 220 h731 20 207 2 2 29 10.0% 119 119 129 132 135 133 128 12 132 135 Levelized System Power Costs Mills/Kwhr) in 2000 18, 13.0 16.1 15.5 18.2 16.2 15.7 15.7 18.7 16.0 201% 20,1 19.6 17.7 16.1 20.0 18.0 17.3 16.0 16.2 17.0 2025 20.9 20.3 18.4 17.2 20.6 19.0 18.3 16.6 16.8 18.3 2035 n.s 2.1 . 188 178 2.0 196 190 180 179 20,0 Cuulative U,0, Consumption ) (Million Ton$)3through . 2025 2,57 2.5% 2.38 2.4 2.63% 2.50 2.3 2.18 .14 .29 2049 2.97 2.9 2.95 2,73 2.98 2.97 2.94 2.82 2.83 2.86 Total U.0, Committed (Million Tons) :R:B . 2025 2.93 2.92 2.45 2.49 2.9 Z2.86 2.83 2.49 2.54 2.59 2049 .99 2.99 .99 2.93 .99 99 2.99 2.9 2.97 2.97 Maximm Annual Enrichment Require- (3) 41 69 65 45 44 42 55 mont through 2050 (Million SNU/yr) (2007)"7 (2007) {2009) (2009) (2009) (2011) " (2011) (2009) (2009) (2005) CGruilative Enrichment (Billion SWJ) through 2028 1,53 1,58 1.60 1.47 2,20 2.08 1.61 1,53 1.51 1.82 2049 1.76 1,83 1.93 1.79 2.55 2.58 2,11 2.0 2,02 2.9 U;0, Utilization (Tons Uy0,/Géc) in{) 2025 5236 5045 3228 2420 3394 2847 3086 423 2169 2513 2049 $236 5042 3138 1497 3165 2480 2908 1512 1514 1528 Enrichmct Utilization (4iL1ien Sw/oie) (2 2028 N 2,72 1.81 1.43 2,58 2.07 1.7§ 1.49 1.46 1.76 2049 3.08 J.08 2.03 92 2,70 2.12 2.06 1.03 1.03 1.17 15 = 38 s ws 58 s 18 8s Cmualative Muclear Capacity Built {(We) through 2025 591 946 1029 0944 1029 1029 1029 1029 1029 049 607 1043 1959 1071 1423 1275 1959 1959 1943 System Costs ($8) 1977 through 2050 discounted at 4.5% 369 451 S02 495 498 476 500 500 513 7.5¢ {88 21 219 226 222 213 218 08 22 10.0% 120 129 13t 136 133 129 131 131 132 Levelized System Power Costs Mills/Kwhr) in 2000 18.1 16.0 15.4 18.0 15.9 15.% 15.4 15. 15.6 2015 19.7 17.1 15.9 19.5 17,2 16.6 15.9 15.9 16.3 20258 20.4 17.6 16.6 20.1 8.1 17.4 15.9 15.9 17.0 2035 21.0 17.9 17.0 20.5 19.0 18.2 14.4 14.4 17.8 Comlative Us0, Consumption (Million Ton3)Bthrough 2025 2.54 2.27 1.99 2.62 2.35 2.14 1.93 1.93 2.07 2049 2.96 2.92 2.70 2.98 2.90 2. 2.69 2.69 2.83 Total 11,0, Committed (Million Tons) tl"fi &gh 2025 2.92 2.81 2.43 2.89 2.81 2.77 2.36 2. 2.58 2049 2.99 2.99 2.92 2.99 2.99 2.9 2.91 2. 2.9 Maximun Annual Enrichuent Require- 42 . (s 40 34 68 60 39 35 35 38 ment through 2050 (Million SWU/yr) {2007) (2011) (2009) {2009} 20n (2010) (2005} {2008) (2009) Cmulative Enrichment (lhllion SW1) through 2025 1.48 1.47 1.32 2.1% 1.94 1.45 1.33 1.33 1.42 2049 1.72 1.84 1.720 2.5 2.54 2.02 1.86 1.86 1.97 Uy0, Utilization (Tons Ug0g/(ic) in)) 025 4939 2975 2362 3066 1730 2687 229 2297 2506 2049 4931 2864 1492 2793 2098 2340 1487 1487 1528 Enrichment Utilization (Million Swy/cuc)(2) in 2025 2.50 1.55 1.28 2.32 i.88 1.41 1.29 1.29 1.38 2049 2.83 1.76 87 2.38 1.79 1.59 .95 .95 1.01 = 1 . sl et foi E m -] ™ " s osm @ o™ . Gmulative Nuclear Capacity Built ' : ((We) through 2025 o1 908 1029 1003 1020 1029 1029 1029 1029 £ 2049 667 987 1959 1334 1747 1505 1959 1959 1959 L System Costs ($B) 1977 through i 2050 discounted at 4.5% 387 494 524 551 568 549 529 523 535 7.5% 192 222 225 234 236 n2 226 225 227 o 10.0% 121 134 134 137 138 136 134 133 134 ,E ' Levelized System Power Costs L | Mills/Kwhr) in 2000 17.9 17.6 17.4 17.9 16.1 15.6 9 167 15.8 2015 20.3 20.0 18.0 20.8 19.1 18.9 7.6 171 171 2025 21.3 20.8 17.2 22.4 20,7 20.7 17.6 17.1 18§ 2035 : 21.8 21.1 15.7 23.1 21.7 22.4 7.3 17.6 206 t # Cusulative U,0, Consumption - (Million Ton2)3through H 2025 2.47 2.4 2.29 2.4 2.16 2.4 2.2 221 2.29 2049 2.95 2.91 2.70 2.97 2.92 2.90 2.6} 2.55 2.87 Total U0, Comitted (Million Tons) tRrBugh 2025 2.90 2.81 2.63 2.90 2.70 2.79 2.5 .50 2.69 2049 2.9 2.99 2.72 2.9 2.98 2.98 2.63 2.57 2.9 Maximum Annual Enrichacnt Require- 343) 36 46 58 46 35 46 44 46 mont through 2050 (Million SWU/yr) (2005) (2003) (2009) (2011) (2023) (2003) (2000) (2009) (2009) i Cumilative Enrichment (Billion SWU) ) through L 2025 1.30 1.24 1.61 1.90 1.63 1.23 1.58 1.55 1.61 ' 2049 1.45 1.35 1.94 2.37 2.33 1.35 188 1.84 1.95 U0, Utilization (Tons U;0,/Ghc) ) 2025 4524 3095 2558 2890 2620 2707 2482 2426 2608 - 2049 4489 3027 1391 2243 1707 1983 1345 1314 1520 -t . } i Enrichment Utilization (Million Swy/Gwe) (2 in ili 2025 2,02 1.37 1.57 1.90 1.58 1.19 1.54 1.51 1.57 2049 2,18 1.37 .99 1.78 1.33 .90 .96 .91 1.00 ko (- 6 x M 4 S S6 & 6 86 NT M Cmulative Nuclear Capacity Built ' (Giwe) through - 2025 588 1029 1029 803 958 917 1029 1029 1029 678 705 { ! 2048 608 1417 1959 855 1064 1004 1950 1959 1791 703 734 L System Costs {$B) 1977 through 2050 discounted at 4.5% 368 434 502 4 451 4“2 506 505 si8 3@ a7 7.5¢ 188 217 219 209 20 200 m 220 224 197 208 - 10.0% 120 1351 131 129 129 129 132 132 13 124 132 Levelized System Power Costs Mills/Xwhr) in 2000 8.2 16.0 . 17.7 159 15.4 15.8 158 153 17.4 17 2015 19,9 164 16,0 186 17.1 113 160 16.0 168 193 19.0 2025 20.5 16.8 15.8 189 177 181 16.1 16.1 188 19,7 19.6 . . 2035 2.1 173 4.2 18,9 181 18.6 16.4 164 208 2.3 . 20.2 ] l Cumulative U,0, Consumption ' o ) ) J (Million Tond)Bthrough : 2028 2,58 2.1 1.97 235 221 .M 2.15 212, 232 243 2.38 049 2.9 . 2,92 275 4 292 1.94 270 2.68 2,91 2,95 2.94 Total U0, Committed (Million ) Tons) tfi h : 202 292 .78 .41 2.85 2,80 -2.83 242 238 217 289 2.87 2049 N ) 2.99 2.9 2.95 2,99 299 2.9 293 2,985 2,98 299 2.9 Maximum Annual Enrichment Require- 55 Sl 39 - 52 . 49 . 45 “ 2 46 92 o5 ment through 2050 (Million SWU/yT) (2000 2019) (2005) o11) (217} (zo11) (2009) (2009) (2009) (2011) ' (2011) 'fi Cunslative Enrichment (Billion SWJ) ' through : .- h 2025 1.5 1.1 1.49 - 1.80 1.69 1.62 1.55 1.5 164 2.69 3.25 2049 _ 1.87 2,42 2.35 . 2.3§ 2.3 2.1t 2.00 1,98 2.15 342 . 4.06 U0, Utilization (Tons U0,/Gk) inl!) _ S : . , 2025 4973 2700 2342 3557 2920 3082 2352 2316 2692 4268 4078 { 2049 o 4963 2105 1505 3497 2807 . 20M4 1503 1496 1666 4258 4074 H Enrichwont Utilization (Million SW/aic)®) . n | . 2025 2,70 1.66 1.45 2.4 L7 LM 148 145 160 3.97 4.60 2040 300 1.1 .15 2,75 2.2 210 1.02 101 120 4.86 5.53 cmorita (1) cumulative U;05 consumed through year 2050 (including forward commitments) per cumulative nuclear capacity built through 2050, Cumulative enrichment requirements through 2050 per cumulative nuclear capacity buflt through 2050, Year in which maximum enrichment requirements occur, vrm-mr.-——} ——, Lo M et ot T W—_ T R C-12 Table C-6 {(cont.) Reactor IWNR-US(LE)/U LwR-US(DE) /U/Th LWR-U3(DE)/U/Th INR-Pu/y LWR-Pu/Th SSCR-US(LE)/U SSCR-U3 (DEY/U/Th SSCR-Pu/Th FBR-Pu-U/l] FBR-Pu-U/Th FBR-Pu-Th/Th FBR-U3-U/Th LWR-US(LE} /U IWR-US (NAT) /U HWR-US (SBI/U INR-US(DE) /U/Th H¥R-U3(DE) /U/Th HYR-Pu/U HMR-Pu/Th FBR-Pu-U/U FBR-Pu-U/Th FBR-Pu-Th/Th FBR-U3-U/Th LWR-US(LEY/U HTGR-US (LE)AJ HTGR-US (LE) /U-T HIGR-U3 HTGR-Pu/Th FBR-Pu-UN) FBR-Pu-U/Th FBR-Pu-Th/Th FBR-U3-U/Th LWR-US(LE) /U LNR-US5 (LE)/U-EE Installed Capacity (GWe)/Levelized Power Cost (Mills/Kwhr) in year 2025 1LE* 1L 2L 3L aL SUL STL 6L 0 8L 269/22.3 zgg;zz.s 360/19.6 310/18.0 S2/21.5 49/19.7 412/19.8 327/17.5 342/18.0 118/17.9 - - - - 202/23.0 296/21.4 - 0/19.0 0/19.5 187/18.8 - 220/20.4 264/20.0 82/20.7 87/17.6 60/18.4 5/20.7 234/19.0 72/19.3 - - . . 107/18.7 132/19.6 9/28.9 21/24.1 9/26.0 - 357/20.6 - . 316/19.8 280/18.0 - - 172/21.7 38/19.5 245/21.7 18 25 35 4 SUS STS 65 78 8S 101/22.2 83/19.7 83/18.0 49/21.5 45/19.2 80/18.8 80/17.4 80/17.4 79/17.9 - 289/22,3 287/19.6 - 0/18.z 0/18.2 1/18.8 . 266/17.8 123/17.3 - 200/21.0 307/18.6 237/17.2 8/20,7 4/17.9 372/17.6 251/16.4 257/16.4 318/16.6 - - 308/20.5 303/19.7 135/19.9 166/15.5 166/15.5 42/17.2 - - - - 101/19.0 152/19,1 48/14.9 48/14.9 23/22.9 . 207/17.8 - - 188/11.7 188/11.7 - - 150/19.8 0/17.2 126/19.3 A A M, M sm SH e 00 0mM 0 & 129/22.1 158/21.1 355/19.9 151/21.3 157/18.8 158/18.4 337/19.0 323/18.7 329/18.3 0/24.9 0/26.8 0/25.6 0/27.0 0/22.0 217/21.4 0/23.9 0/23.3 0/20.0 222/22.0 45/22.9 0/22.0 0/25.1 0/20.5 20/20.0° ©0/21.1 0/20.7 32/19.7 - - 22/24.2 178/22.0 - 0/28.9 0/26.5 12/21.6 - - - 350/24.0 296/22.4 163/24.3 45/17.3 0/19.4 0/22.9 - 415/21.1 0/20.5 - , - 109/21.7 182/22.7 2/20.9 11/20.8 0/26.2 . 384/14.6 - ; 356/17.4 348/17.4 - - 190/22.3 $7/17.2 176/21.1 16 26 3G e SUG 5TG 6G 76 3G 172/22.3 142/19.4 14217.5 172/21.2 142/19.1 404/19.2 294/16.9 295/17.2 SBAT/17.8 . 0/19.8 0/18.5 0/20.4 0/19.8 0/19.9 0/18.4 0/18.4 0/19.4 125,20.7 - - - - - - - - - . . 284/19.0 305/18.5 - 14/17.4 1/17.4 50/18.1 . 305/17.2 195/15.8 - . . - A - . . 56/18.5 87/18.1 43/19.0 104/15.4 91/16.0 0/20.2 - 175/17.9 127/14.2 - S o - - 117/15.8 79/16.5 - 133/18.3 179/18.5 15/22.3 30/21.0 0/27.5 . - 195/11.4 - . - . - 313/17.8 2941173 - A 180725.7 . . - 29/15.9 162/23.8 1T 1Im™ 30/21.5 30/21.5 358/20.7 385/20.7 *System with standard LWR only, S T . el li_,_w § e e T e r— vt ot — r— v - o — e e " C oy r A S v wn . E.-—n., ¥ Py b r.m i ! i r— T—_— o) ' - ¥ C 1 ey C-13 Table C-7. Summary of Results for Cases Assuming Hii;h-Cost U30g Supply, 200 GWe Installed Capacity in Year 2000, and 10 GWe Installed Capacity Each Subsequent Year s uw 15¢ e 1c Cumulative Nuclear Capacity Built ) ((¥e) through 2025 533 554 579 619 589 2049 570 638 727 654 System Costs {$B) 1977 through 2050 discounted at 4,5 269 269 279 302 81 7.5 128 128 130 138 131 10.0% 81 80 8 83 81 I.e\_rul ized System Power Costs Plills/Khe) i 16.8 16.5 16.5 16,5 16.5 2018 19,2 18.6 18.$ 19.3 18.5 205 20.1 19.5 19.4 20.5 19.3 2035 20.9 20.1 19.9 21.1 19.7 CQulative U lJs Consumpt ion Million Twé) through 2025 2.08 2.02 1.94 1.88 1.94 2049 2.90 2.89 2.87 2.82 2.8 Toral U0, Committed (Million Tons) tfirsugh 2025 2.79 2.76 2,72 2.62 2,71 2049 2.98 2.98 2.98 2.97 2.9 Maximm Annual Pawichment Require- 39 1 41 35 24 A5 ment through 2050 (Million SWU/yr) 2019f 2021) (2021) (2011) (2023) Gulative Enrichment (Billion SWJ) through 2025 1.23 1.26 1.1 .54 1.28 2049 1.73 1.8 1.62 1.20 1.99 U0 Utilization (Tons V;0,/Ghe) in(*) 2025 5236 4979 4694 22 4603 2049 5236 4974 4669 4090 4554 Fnrichment Utilization (Million Sw/Gwc) () " 2025 ' 2,31 2.28 1.91 1.52 2.18 2049 3.03 3,92 2.54 1.66 3,04 Installed Capacity ((We)/Levelized Power Cost (Mills/Kwhr) in Year 2025 Reactor 1LEC* 1LC 18C 1IHC 1GC LWR-US({LE)/V 363/21.7 11/21.6 44/21.4 144/21.2 114/21.4 LWR-US(LE)/U-EE - 374/20.8 - - - LWR-U5(DE) /U/Th - - - - - LWR-U3(DE)/U/Th - - - - - LWR-Pu/U) - - - - - L¥R-Pu/Th - - - - - SSCR-US(LEY/) - - 365/20.4 - - SSCR-U3 gE)/U/Th - e - - ‘SSCR- ‘ - - - - - HWR-US (NAT) /U - - - 0/24.2 - HWR-US(SEDH/N . - - - 305/21.5 - -~ WWR-US(DE)/U/Th - - - - - HWR-U3 (DE)/U/Th - - - - - HR-Pu/U - - - - - HWR-Pu/Th - - - - - - HTGR-US (LE)/U - - - - - . HIGR-US (LE)/U-T = - - - 304/20.1 . 1) Cumulative U305 consumed through year 2050 (including foffiard commitments) per cumulative nuclear capacity built through 2050. 2) Cumulative enrichment requirements through 2050 per cumulative nuclear capacity built through 2050. 3) Year in which maximum enrichment requirements occur. '*System wfth standard LWR only, . Table C-8. Summary of Results for Cases Assuming Intermediate-Cost U30g Supply, 350 GWe Installed Capacity in Year 2000, and 15 GWe Installed Capacity Each Subsequent Year nE* n- i aP L m oL n 1 Cumulative Nuclear Capacity Suilt ((Me) through v 2025 994 1015 1020 1029 1029 108 102 1029 1029 1020 7049 1135 103 1785 1959 1852 1921 1864 1959 1959 1956 Systen Costs ($B) 1977 through 2050 discounted at . 473 470 485 485 L2 ¥ 4E9 485 485 485 486 7.5% m m m a3 21 21 73 213 3 213 10.0% 128 27 128 120 137 129 129 129 129 129 Levelized System Power Costs (Mills/Kvhr) in 2000 16.6 16.8 150 .3 166 148 14,7 47 T 187 2015 18,5 179 155 3150 176 155 153 15.2° 15.2 15.2 2025 19,§ 187 163 153 180 160 157 154 154 155 2035 20.1 193 16,5 149 18.2 163 158 141 Ma 150 Cumulative U0, Consumption (Million Tond)*through 2025 3,53 s 2.8 2,7 287 A% 2% 2.7 Ly W 2049 5.63 ‘s6 4.76 440 ST 481 470 433 43 a8 Total U0, Committed (Mildion Tons) tfir&gh : 2025 5.20 5.06 350 328 366 337 3% 3. 33 LK 2049 5.54 595 5.68 540 574 568 5.6 5.33 535 547 Mexims Annwal Enrichaent Roquire- ny 7 7 & 100 7 61 82 65 ment through 2050 (Million SWU/yr) (2025§ (2025) (2037) (2037) (2039) (2039) (2033) (2033) (204%) Cuurlative Enricheent (Billion SWU) 2025 : 2.09 12 L5686 147 241 1.64 1.68 1.64 .64 .64 2049 3.35 3.49 8.2 88 455 555 N 3.00 309 316 0y Utilization (Tons Ug0y/Gic) in{)) ' Usfs 2025 3 ) 236 4985 306 3187 3ss2 82 3o 5206 3206 3303 2049 52% €973 sist 2756 s105 2987 3037 2733 2733 2798 farichment Utilization (4illion st/gie) ) 2025 2.1 2.09 1,51 1435 2.3 1,58 1.5 1.5 1.5 1.5 2049 .95 2092 1.7 145 246 1.8 1.7 1.8 1.58 1.6 15 & 38 as s SIS 6 » 8 Cmulative Nuclear. Capacity Built (%) through 2025 1020 1029 1029 1029 1079 1029 102 1023 1029 2049 1Z7 1837 1959 1:¢3 1959 1959 1950 1959 19%8 System Costs (3B) 1977 through 2050 discounted at 4.5% 483 484 481 536 485 485 485 485 485 7.5% 23 213 212 230 24 e 214 24 24 10.0% 128 © 129 128 136 129 129 129 129 129 Levelized System Power Costs (Mil1s/kwhr) in 2000 16. 149 147 163 T T M7 W7 KT 2015 179 152 147 17.0 149 4.9 4.9 149 149 2025 18,7 15.5 4.9 17,3 15.2 15.2 152 152 15.2 2035 19, 156 146 17.0 153 15.3 15.8 15,3 15.3 Omulative U.0, Consumption (Mil1ion Tond)Sthrough 2025 5.6 213 220 27 214 2.4 214 zae 2.M4 2049 546 430 404 4060 386 3.8 3.86 .86 3.8 Total U0, Committed (Million Tons) tRrdugh o 2025 4.85 3.18 3.10 3.5 K 2.9¢ 2.9 2.94 2.94 2049 \ 5.92 5.46 5.3 552 4.2 4.92 492 4.92 492 Maximm Annual Entichment Require- 63 1) 57 53 73 $s 55 §5 55 55 mont through 2050 (Million SWU/yr) 020§ (2045) (2047) (2039) (2049) (2049) (2049) (2049) (2049) Cumstative Enrichment (Bi)lion SWU) through - 2028 1.86 1.40 1.38 2.26 1.46 1.46 1.46 1.46 1.46 2049 3.09 263 249 8 267 .82 .62 2.62 2.62 Uy, Utilization (Tons Ug0g/ce) in(V) , 2025, ) 4TI4 3086 3010 3262 253 2858 2858 2858 2858 2049 4657 2820 711 2844 11 2511 W s 81 Frichmont Utilization (Million Swy/Gic)(?) 2025 1.81 1.3 L34 219 142 142 142 142 14 2049 w3 1% 1,27 oz 1M 1M 1.8 M LM *System with standard LWR only. i 4 - r ok 1 [. P [uwm-‘ [ L 1 s et - C-15 Table C-8 (cont.) M .| » o Sl S & M & Cumilative Nuclear Capacity Muilt ' (me)zfi'z";mgh " 1020 1020 1020 1029 1029 1020 1029 1029 1029 2045 : 1497 1921 1989 1943 1959 1959 1959 1959 1959 System Costs ($B) 1977 through 2050 discounted at B 519 a4 512 552 523 523 514 512 514 7o m 228 222 229 224 24 222 222 2 10.0% 130 134 132 134 133 133 132 132 132 Levelized System Power Costs Mills/Kwhr) in . 2000 6.5 163 158 167 16.0 16.0 15.7 15.7 15.7 2015 185 174 161 18.0 16.7 16.7 16,0 16.06 15.9 2025 19.6 18.3 158 1.8 17.0 17.0 160 158 159 2018 20,1 18.8 1.9 195 17.1 17.2 159 15.3 15.5 Cuomulative U;% Consumpt ion Million Ton3)>th (Million Ton<)”through 500 272 231 2.4 2582 251 232 2,30 2.38 2049 _ $.20 455 271 5.36 A.32 437 366 2.70 3.37 Total U0, Committed (Million Tons) thrBugh 2025 435 367 2.65 421 5.59 3.57 2.1 264 2.3 2049 S.86 5.5 2,04 S.89 4.7 .85 &30 273 .17 Maxism Annual Fnrichment Reguire- a2 s3 7 96 o4 64 a7 a7 *% ment through 2050 (Million SWU/yr) (2009§%) (2011) (2009) (2033) (2029) (2031} (2009) (2009) (2009) Cumilative Enrichmcnt (Billion SWl) ‘ . t 2025 : 1.57 1.0 1.62 2.05 1,75 1.74 " 1.65 1.62 1.67 2049 7 210 2,34 1.95 4.08 3.06 3.9 2.59 1.84 2.40 U30, Utilization (Tons U,0,/(kc) in(!) ' 2025 4225 3562 2572 4093 3490 470 263 2562 2773 2049 3096 2804 1398 3030 2431 475 2195 1392 1924 Enrichment Ueilization 0ai1lion SU/0kc) @ 2025 1,52 1.7 1.5 1.9 1.7 1.69 1.5 1.57 1.62 2049 140 1,22 1,00 2,10 1.5 158 1.3 % 1.2 16 x 3 G WG SG 8 6 5 Csmlative Nuclear Capacity Built (G¥e) through 2025 1029 1029 1029 1029 1029 1029 1029 1029 1029 2049 1320 1950 1959 1794 1924 1844 1959 1956 1959 System Costs ($B) 1977 thmug. h 2050 discounted at 4.5% 487 486 486 515 487 486 485 486 486 7,58 214 214 24 73 214 214 214 24 n4 10.0% 128 529 129 133 129 129 129 129 129 l(f;elim System Power Costs ills, r) in 2000 ) in. 16.4 150 15,0 162 15.1 15.0 . 1S.0 15,0 15.0 2015 170 149 150 16,6 15.2 15.5 4.9 14,9 149 2025 186 1.8 150 167 15.6 15.7 149 149 150 2035 1900 142 148 16, 15.8 6.0 14,7 M7 147 Camilative U o. Cmsi;rtim (Million Tond)®through 2025 3.2 232 230 258 2.2 22 L3 213 2.2 2049 541 4l 4l AT 4% 465 419 419 424 Total U0, Comsitted (Million ' _ Tons) tArBugh - . 2025 473 320 . 317 357 3,20 3.29 . 3.09 3.09 3.1S 2049 _ 591 52 .52 569 55 - 564 526 526 §29 Maximm Annual Enrichaent Roquire- B4gz) 70 70 90 . - 86 7% 7 " 75 ment through 2050 (Million SWU/yr} (20295 ) (2049) (2049) (2039) - (2047) (2041) (2047) (2047) (2047) Cumilative Enrichment (Billion SWi) ' 2025 ’ ; 2.1 1.62- - 1.60 1.99 1.64 1.63 1.5 1.58 1.57 2049 .81 16 313 397 - 3.4} 327 310 310 316 U,0, Utilization (Tons U.0,./GWe) inY - - g 378 2025 (Tons Ug0g/Gle) 4597 305 308t M7z 3108 3198 3005 3004 3057 2049 : _ 4478 2683 2680 3172 2865 3055 . 2683 2682 2698 Enrichacnt Utilization (Million Swi/ve)(®) 2025 2,05 1, 1.56 1,93 1,50 1.58 1.51 - 1.51 . 1.53 2049 2,89 . 1.61 1.60 221 177 1.77. 1.58 1,58 1.62 } Cumulative Uj0g consumed through year 2050 {including forward commitments) per cumulative nuclear capacity built through 2050. ) Cumulative enrichment requirements through 2050 per cumulative nuclear capacity built through 2050. Year in which maximum enrichment requirements occur, ‘Table C-8 {cont.) Reactor LWR-US({LE)/U INR-US(LE)/U-EE - LR-US (DE) /U/Th LWR-U3 (DE)/U/Th LWR-Pu/U LWR-Pu/Th. FBR-Pu-U/U F3R-Pu-U/Th FBR-Pu-Th/Th FBR-U3-U/Th INR-US(LEVU - LWR-US (DE) /U/Th LWR-U3 (DE) /U/Th LWR-Pu/l) LWR-Pu/Th sexmn, SSCR- thh) FBR-Pu-U/UJ FBR-Pu-U/Th FBR-Pu-Th/Th FBR-U3-U/Th LWR-US(LEY/U HWR-US (NAT) /U HWR-US (SEU) /U FBR-Pu-UAJ FBR-Pu-U/Th FBR-Pu-Th/Th FBR-U3-U/Th LWR-US(LE)/U HTGR-US (LE) U HTGR-US (LE) /U-T HTGR-US (DC)/U/Th HTGR-US (HE)/Th HTGR-U3(DE)/U/Th HIGR-U3/Th HTGR-Pu/Th FBR-Pu-U/1} FBR-Pu-U/Th FBR-Pu-Th/Th FBR-U3-U/Th " Installed Capacity (GWe)/Levelized Power Cost (Mills/Kwhr) in year 2025 1LE* 1 L 3L 4L SUL STL 6L 7 8L 703/20.6 30;20.5 523/17.0 460/15.8 57/18.8 $541/16.9 544/16.3 551/15.8 551/15.8 $53/16.0 - 695/19.8 - - - - - - - - - - - - 439/19.0 3/17.8 - 0/16.3 0/16.3 0/16.8 . - - - 243/17.8 73/17.0 72/16.4 72/13.2 72/13.2 13/14.2 - - 216/16.8 254/14.9 - - - - - - - . - - - 12216.2 122/15.8 103A12.5 103/12.5 103/14.1 - - - 25/12.5 - . - - . . . - - - . - - 13/10.7 13/10.7 - - - - - - - - - 11/12.8 = - = = - - = - 0/1‘05 011602 15 28 ‘38 4s SUS TS 6S 78 8s 109/20.3 83/16.4 115/15.4 S7/17.9 184/15.6 184/15.6 184/15.6 184/15.6 184/15.6 - - - 380/17.6 0/17.3 - 0/17.3 ©/17.3 0/17.3 . 239/15.7 279/14.4 - - - - - - 630/19.5 418/16.0 346/15.0 0/17.8 300/15.5 300/15.5 300/15.5 300/15.S 300/15.5 - - - 302/17.1 128/15.5 128/15.5 128/15.5 128/15.5 128/15.5 . - - . 128/15.5 128/15.5 128/15.5 128/15.5 128/15.5 - - 0/1401 - - - - - - - - - - - - 0/16.3 0/16.3 - - - - - - - - - 0/16.3 - - - - . . . 0/18.4 0/19.8 1H 2H 31 4 St STH 6H ™ 8H 232/19.5 480/17.8 359/16.1 666/19.5 592/17.4 587/17.4 375/16.1 357/15.8 410/15.7 0/22.9 0/23.9 0/22.2 0/25.4 0/23.0 0/23.0 ©0/21.7 0/21.7 0/21.5 507/20.4 0/20.6 0/19.3 0/21.8 0/20.1 ©0/20.1 ©0/19.1 ©0/19.0 0/18.9 - . - 63/21.7 0/26.2 - 0/25.1 0/24.4 0/24.1 - .- . 10/20.4 0/17.5 0/17.5 ©0/17.3 0/17.4 0/17.6 - 259/19.6 0/18.8 - - - - - - - . - . 147/17.2 153/17.3 0/18.8 0/17.0 0/17.3 - - 380/14.6 - . - - . - . - - . - - 364/15.9 363/15.2 - - - - - - - - - 182/15.8 - - - - - . - 19/15.1 148/15.4 16 26 3G 4G _SuG TG G 76 8 1201/20.3 472/14.6 477/14.9 193/17.9 405/16.1 518/16.2 466/1S.1 464/15.1 471/15.1 . 0/16.1 0/16.3 0/17.7 0A17.1 0/17.2 0/16.3 0/16.3 0/16.3 $39/19.2 - - - - - - - - - - - 471/16.7 109/16.2 - 5/15.3 7/15.3 14/15.4 . 28/14.5 14/15.1 - - - - - L. - - - 76/15.9 54/15.7 45/15.6 71/15.0 71/15.0 94/15.0 - 63/13.4 63/14.5 - - . - S - . 176/15.3 175/15.7 - 172/16.0 176/16.2 148/16.5 148/16.5 132/16.6 - - 10/11.1 - - . - - - - - - - - 50/12.8 50112.7 - - - - - - . - - 28/12.6 - - - - . - - 0/16.1 0/16.6 *System with standard LWR only, - r. i | r— ro T o r— I r— r— rm r—- U D-1 Appendix D, CALCULATIONS OF NUCLEAR AND FOSSIL PLANT COMPETITION ' BASED ON ECONOMICS M. R. Shay, D. R. Haffner, W. E. Black, T. M. Helm, W. G. Jolly, R. W. Hardie, and R. P. Omberg Hanford Engineering Development Laboratory In a series of calculations that preceded those reported in Chapter 6 for nuclear power systems, the same analytical model was used to evaluate power systems that include both nuclear power plants and coal-fired power plants, with the two types of plants being in economic competition, As was stated in Chapter 6, the results of these calculations indicated that at U30g prices above $160/1b, nuclear power plants do not compete well for the assumptions used in this study. Therefore, for the all-nuclear systems it was decided to 1imit the uranium resources to those available at prices below $160/1b, This appendix describes the initial set of calculations. The nuclear plants used were LWRs, with and without recycle, and they correspond to Cases 1L, 2L,....8L in Chap- ter 6. The primary differences betWeen the calculations presented in Chapter 6 (and in Appendix C) and the calculations described here are as follows: (1) Instead of a nuclear energy growth projection, a total electrical energy growth projection was used. (2) In add1t1on to nuclear plants, coal plants were available to sat1sfy the total electrical energy demand. (3) No price constraint on ore existed. Instead it was assumed that additional uranium ore was always available at increasingly higher costs. As with the all-nuclear systems, two different U30g price structures were used. (4) Power plant selection was based on economics instead of Us0g utilization. The electrical energy demand that waS'used for these calculations is shown in - Table D-1. This projected demand assumes a 5 6% per year growth rate until 1980, and a 5.1% per year growth rate from 1980 to 1990. The growth rate decreases each decade until year 2030, after which a constant 2.5% per year growth rate is assumed. | The marginal cqs# of urahium ‘as a function of the cumulative qUantity mined was shown in Table B-7 of Appendix B. In this appendix cases that use the high-cost uran1um supply are denoted as cases lL; 2L, ..., while cases that use the intermediate-cost uranium supply are denoted as cases LU, 2LU 'As has already been emphasized, it was assumed for these calcu]at1ons that the quant1ty of availab]e uranium was unlimited. The only restriction on uranium consumption was Table D-1. Projected Total based on economics - that is, the Electrical Generation . L. marginal cost of an additional pound of U30g increases as more uranium is Electrical Electrical Energy Growth consumed. ‘ Energy Rate Year (1012 kWh) (% per year) Fossil-fueled power plants were re- _ — presented by nine different coal plant 1975 1-9; | 5.6 " types which are indicative of different 1980 - | 2-?} 5.1 coal regions. The principal differences 1990 4.1} 4.1 - between coal plant types are the coal price, 2000 | fifl} 3.5 the coal energy content, and the size of 2010 8-5; 3.0 the demand that can be satisfied by each 2020 4 11-5} 2.5 coal plant type. The maximum fraction of 2030 14.9 the total electrical energy demand that can be satisfied by each regional coal plant type is shown in Table D-2. ' This table also gives the heat éontent'of the coal for each region. The capital cost associated with building a coal plant was assumed to be 12% lower than the capital cost of a LWR, or $550/kWe (in 1/1/77 dollars). Therefore, for nuclear plants to be built instead of coal plants, the fuel costs of the nuclear plants must be enough lower than the fuel cost of fossil plants to override this capital cost differential. If nuclear plants are less expensive than coal plants for all regions, then all of the new plants built will be nuclear. Figure D-1 shows how the nuclear market fraction decreases as nuclear plants become more expensive. If nuclear plants increase in price by 20% over the price where all of the market would be:nuclear, the nuclear market fraction decreases to 0.75. An increase of about 35% in the price of a nuclear unit reduces the nuclear market fraction to.about 0.34, while a 57% 1ncrease results in a1l of the new plants built being fossil- fueled p]ants Nuclear power growth projections for the LWR on the throwaway cyc]e are shown for both uranium supplies in Fig. D-Za. For the high-cost uranium supply case, nuclear power peaks at 500 GWe of installed capacity around the year 2005 and then phases out to about 100 GWe in 2040. On the other hand, if the intermediate-cost uranium supply is assumed, nuclear power continues to grow until about 2015 to almost 900 GWe, and then decreases to about 300 GWe in 2040. As a result, nuclear is more compet1t1ve with coal and captures a - larger share of the market. Figure D-2b shows that recycling plutonium in LWRs (Case 2L) increases the nuclear power market even more than the assumption of a larger uranium'sapply, and introducing the Pu/U-fueled FBR with recycle (Case 3L) further increases the nuclear market to 1300 GWe of installed nuclear capacity in the year 2040. The U 0g utilization, defined as the — r— L e T o U T r— r—- r— = £ ; Table D-2. D-3 Maximum Electrical Energy Demand Satisfied by Regional:Coal Piants Maximum % of Total Heat Content Electrical Sales (Btu/1b) New England (NE) 3.9 13,500 Middle Atlantic (MA) - 13.1 11,783 East North Central (ENC) 19.5 10,711 West North : Central (WNC) 6.6 9,408 South Atlantic (SA) 16.6 11,855 East South Central (ESC) 9.6 11,006 West South ’ . Central (WSC) 12.2 6,583 Moutain (MT) 4.9 9,637 Pacific (PA) 13.5 8,101 center. outside the center to those inside is less than unity and rapidly decreasing. total Uz0g consumed plus committed per GWe of nuclear power constructed through the year 2050, is also given for these cases. As noted, recycling plutonium in LWRs reduces U30g usage by 38% per GWe, while introducing the FBR results in a 62% reduction. With the intermediate-cost U30g supply, 1300 GWe for the FBR case becomes almost 1800 GWe in 2040 (see Case 3LU in Fig. D-2c). each of the ore supplies occurs around the year 2040, although the installed nuclear capacity is very fiat at this The nuclear power peak for point. The disadvantage of classical plutonium recycle in FBRs is demonstrated in Fig. D -2d for Case 3L. Here the two Pu-fueled reactors are inside the energy center and the LEU-LWR is outside the It can be seen that after about 2020, the ratio of reactors that can be located In fact, as o o < . . . o~ O o 1 { 1 MARKET FRACTION AVAILABLE TO NUCLEAR POWER o ) i L ST 0.9 Fig. D-1. Effect of Changing . ‘.0 1.1 1.2 o -3 .. RELATIVE POWER COST Nuclear Power I N P HEDL 7805-070.50 Costs on the Nuclear Market Fraction. 25007 Y T 2500 T T T T T = . i 1 T (b} HIGH-COST UOg SUPPLY U0, UTRIZATION T U We) - NUCLEAR-FOSSIL COMPETITION NUCLEAR-FOSSIL COMPETITION L Lo IL 5040 2000~ ‘ - - o 20001 A 3100 n F £ a 1% 5 ° E B < 1500 - toSwsol 7 S 8 3 - LWR + FBR Pu RECYCLE % 3 2 , g 1000l 1LU - LWR, INTERMEDIATE COST Uy0g SUPPLY | g 1000 - o : < % "i_' _ 2 - LWR Pu RECYCLE - 500 = 500 IL = LWR THROWAWAY 1L = LWR, HIGH-COST U30g SUPPLY , ! 1 ? l 1 L 1 i 0 1 1 { 1 1 i 960 1950 2000 2000 - 2020 2030 2040 1980 1950 2000 2000 2020 2030 2040 L YEAR YEAR ' 2500 T T T T T (<) NUCLEAR-FOSSIL COMPETITION ' : = 200 T T T T T T ‘ {d) « 2000 - LU - LWR ® FBR Pu RECYCLE, = 3 INTERMEDJATE-COST U50g SUPPLY CASE 3L -~ HIGH -COST U, SUPPLY, £ im - ~ucn.un,rossuiogowenno~ - E 2 1500 © 3 Z : 3 § ’ | g = | Z 100 3 ! e 3L - LWR + FBR Pu RECYCLE, 8 : 3 HIGH-COST U30g SUPPLY Zlow - P > ; 2 ; 2 z : 500 I 2 oo b LWA-Pu/U _ LWR-US(LE}U 0 1 ! 1 ] 1 ° 1 I ) 1 1 1 : 1980 1990 2000 2010 2020 2030 2040 1980 1990 2000 2010 2020 200 240 2050 YEAR YEAR ; 2500 2 1 T T T T te) 2500 T T T — T ” : HIGH-COST Li40g SUPPLY, T A N ST U W, NUCLEAR-FOSSiL COMPETITION w CONVERTER Pu/Th TRANSMUTER 5040 ® 2000 |- 2’ 3410 i NUCLEAR-FOSSIL COMPETITION x S 2880 2000 |- - ° & 1680 3 ' i o g z = B 5 5LU INTERMEDIATE-COST U30g SUPPLY g 1500 g 1500 N . L] . 5 & - FBR Pu/Th TRANSMUTER : > : o 100 ] g 1000 - — o 4 - LWR Py THROWAWAY, z o = SL = LWR Py/Th TRANSMUTER 3 £ & 5L HIGH COST U30g SUPPLY sl % 50 b = IL ~ 1WR THROWAWAY 0 1 ] i 1 i ) 0 1 1 1 1 1 1980 1990 2000 2010 2020 2030 2040 1980 1990 2000 2010 2020 . 2030 2040 YEAR YEAR ‘ £ Fig. D-2. Installed Nuclear Capacities During Years 1980-2040 (or 2050) for Emj Various Power Systems Including Both Nuclear and Coal Power Plants. Fig. D-2 (cont.) i 2300 1 T FBR Pu/Th TRANSMUTER : NUCLEAR-FOSSIL COMPETITION ! o 2000 - = 6LU 0 5 = U e % 1500 o < w - g > 1000 + o 2 = s 4 g ot . .t 11 ] 1980 1990 2000 2010 2020 2030 2040 INTERMEDIATE-COSY U30g SUPPLY 6L HIGH-COST U30g SUPPLY YEAR 2500 T I T CASE & - HIGH-COST U305 SUPPLY, » 2000 -NUCLEAR-FOSSIL COMPETITION x 6 > = A\ £ soof 3 o 3 -t ) 2 100} [=] wd - - I 2 o g I ; = s00 1 1 { LWRU5QEA FBR-Pu~U/Th 2010 5 3 g | 1 2020 2030 2040 <4050 YEAR INSTALLED NUCLEAR CAPACITY, GWae INSTALLED NUCLEAR CAPACITY, GWe 2500 T T T 1 T {h) HIGH-COST U50g SUPPLY, U0 UTILIZATION 5T uaoa/Gw.) NUCLEAR-FOSSIL COMPARISON [ & 1680 - 2000 A 1480 & 2500 1500 |~ - § g 1980 1990 2000 2010 2020 2030 YEAR 2500 I 1 | 1 ¥ CASE ®. HIGH-COST U0g SUPPLY, [~ NUCLEAR-FOSSIL COMPETITION g FBR-Pu-Th Th LWER-Pu Th LWR-UJ DL g g U Ti LWEK-US Lt 1) (i) 0 1950 199¢ 2000 2010 020 2030 YEAR 2040 2050 the system becomes less and' less dependenf upon uranium ore and more and more upon plutonium, the energy support ratio will approach zero. The denatured fuel cycle Cases 4L, 5L,* and 6L are compared with the throwaway cycle in Fig. D-2e. Nuclear market penetration for plutonium throwaway (Case 4L) is not sub- stantially greater than for the throwaway cycle (Case 1L). The peak penetration is about 630 GWe of installed nuclear capacity versus 500 GWe for the throwaway cycle. However, if the plutonium is utilized in an LWR Pu/Th converter (Case 5L), the maximum nuclear penetration is 1000 GWe, which is a factor of two gréater than for the throwaway cycle and, furthermore, the peak does not occur until more than 10 years later. Introduction of the FBR with a Pu-U core and thorium blankets (Case 6L) results in a peak penetration of 1250 GWe in about 2025. After 2025, the nuclear market fraction is constant because the system is essentially independent of uranium, which is becoming increasingly more expensive. | - With respect to U30g utilization, Fig. D-2e shows that the Pu/Th converter case has slightly better ore utilization (by 7%) than classical plutonium recycle in LWRs (Case 2L in Fig. D-2b). Furthermore, plutonium "transmutation" in Pu-U FBRs also has better U30g utilization (by 12%) than classical plutonium recycle in FBRs (compare Cases 3L and 6L). The reason for these trends is that the 233U fuel that is being bred is worth more as a fuel in thermal reactors than the plutonium that is being destroyed. The effect of a larger uranium supply on the market penetration for converters and FRBs that produce 233 is shown in Figs. D-2f and D-2g. For both cases (5 and 6), the targe uranium supply increased the maximum nuclear penetration by about 450 GWe. Case 7L introduced a denatured 233U-fueled FBR to the 6L case, and Case 8L is identical to Case 7L except that the FBR with a Pu-U core is replaced with an FBR with a Pu-Th core. The maximum nuclear penetration for Cases 7L and 8L are compared with 6L in Fig. D-2h. The denatured 233)-fueled FBR doesn't have any impact because this reactor is competing with less expensive 233U-fueled LWRs and therefore isn't built. The nuclear market penetration for Case 8L is seen to decrease after about 2020. This is because the neutronics properties of FBRs fueled with Pu-Th are degraded significantly from those fueled with Pu-U. As a result, the doubling time of these reactors is longer and the cost is higher. The degraded neutronics of the Pu-Th FBRs are reflected in the U30g utilization of Case’ 8L where the ore usage per GWe is almost 50% higher than for Case 6L. The objective in building FRBs with Pu-Th cores is to increase the 233 production and therefore the ratio of reactors located outside the energy center‘to those inside the * The nuclear reactors that are available in Case 5L with nuclear-fossil competition are similar to Case 5UL described in the other sections of this report. However, in 5L the denatured 235U-fueled LWR isn't built because of economics. Therefore, the solution more closely resembles Case 5TL. o 4 —C, ! 1 1 i r 1 { ‘ energy center. It can be seen from the nuclear power growth patterns for Cases 6L and 8L, shown in Figs. D-2i and D-2j, that the energy support ratio for Case 8L is higher. The degraded neutronics of the FBRs fueled with Pu~Th are reflected in the U;0g utilization of Case 8L where the ore usage per GWe is almost 50% higher than for Case 6L (see Fig. D-2h). However, for most years the total amount of energy that is available to be built in the energy centers is about the same for Case 8L as it is for Case 6L because the total amount of nuclear energy is lower. Key selected results from the nuclear-fossil competition calculations are presented in Tables D-3 and D-4 for high-cost and intermediate-cost U;0g supplies respectively. Each table presents the cumulative capacity of nuclear and fossil plants built through year 2050, the total system costs, the annual coal consumption in 2025, data on uranium and enrichment utilization, the installed capacity of each reactor type in year 2026, and the levelized power cost of each reactor type for a reactor starting up in year 2025. The most striking conclusion that can be drawn from the comparison of levelized power costs of each reactor type is that there isn't a large difference. The reason, of course, is that the total amount of uranium consumed doeén't vary much from case to case because when uranium becomes expensive, fossil plants are constructed in place of nuclear plants. This point is demonstrated in Table D-5, which shows the time behavior of the U30g price. It can be seen from this table that the differences in the price of U30g for the different nuclear systems are not large. | D-8 Table D-3. Summary of Results for Cases Assuming High-Cost U303- Supply, an Electrical Energy Growth Projection, and Power Systems Including Both Nuclear and Coal Power Plants Cumilative Capacity Built (Gwe) through 2050 : Nuclear Fossil System Costs ($B) 1977 through 2050 Discounted 8 Armnuval Coa190munmmkion in 2025 (10 tons) Cumylative U0 Consumption (10 tons)tfinnmh 2026 2050 , Total tted U0, through 2050 (gggn:m) 38 Maximun- Ammual Enrichment Rquirements through 2050 (10° SWU/yT) Cumflamise Enrichment through 2050 (10° swpn U,0g Utilization(!) Enrjchnent Utilization () (10° SWU/GWe) L 2L 3L 4L 5L 6L 7L 8L 705 1585 2663 933 1684 2597 2595 1909 4611 3731 2653 4383 3632 2719 2721 3407 1804 1733 1701 1806 1724 1703 1703 1718 787 764 758 791 761 760 760 761 479 470 468 483 468 469 469 469 5.22 3.72 3.15 4.79 3.59 2.91 2.91 3.25 2.92 3.50 3.56 2.88 3.62 3.68 3.68 3.69 3.42 4.75 4.60 3.13 4.75 4.33 4.33 4.70 3.55 4,92 5.06 3.18 4.85 4.37 4.37 4.77 54 5y 65 73 727 7 80 80 79 (2005) (2011) (2009) (2005) (2015) (2011) (2011) (2015) 2.12 .1 2.89 2.53 3.40 3.11 3.1 3,37 5.04 3.10 1.90 3.41 2.88 1.68 1.68 2.50 3.01 1.96 1.09 2.1 2.02 1.20 1.20 1.77 Installed Capacity (GWe) in Year 2026/Levelized Power Costs (Mi11/Kwhr) in Year 2025 Reactor 1L LWR-US (LE) /U 36/23.2 US{LE)/U-EE 225/22.3 Us (DE)/U/Th - U3(IE)/U/Th - Pd/u - Pu/Th - FBR-Pu-U/U y Pu-U/Th ) Pu-Th/Th - U3-U/Th - Fossil 1934 Total Nuclear 261 2 579/21.1 336/22.3 1280 915 Sk 513/20.8 196/19.5 444/18.4 1042 1153 4L SL 6L 7L 8L 113/21.6 661/21.2 594/20.7 594/20.7 668/20.8 189/22.5 0/23.5 0/23.2 0/23.2 0/23.1 157/20.0 120/20.6 190/19.6 190/19.6 230/20.8 - - - - - 181/20.1 52/22.1 52/22.1 102/23.0 408/19.4 408/19.4 - - - - - 104/22.6 - = - 0/2300 0/2510 1736 1233 951 951 1091 459 962 1244 1244 1104 (1) Cumulative U30g comsumed through 2050 (including forward commitments) per cumulative nuclear capacity built through 2050, {2) Cumulative enrichment requirements through 2050 per cumulative nuclear capacity built through 2050, (3) Year in which maximum enrichment requirements occur. PR el iy L et L e o r— oo r— T DA s ( Table D-4. D-9 Summary of Results for Cases Assuming Intermediate-Cost U304 Supply, an Electrical Energy Growth Projection, and Power Systems Inc1ud1ng Both Nuclear and Coal Power Plants Cunulative Capacity Built (GwWe} through 2050 Nuclear Fossil System Costs ($B) 1977 through 2050 Discounted € Annual CoalQOunsmpticm in 2025 {10 tons) ative U,0 Consumption (10° tons) tfi 2026 2050 Total Cngmtted U.0, through 2050 (10° tons) ° © Maximum Arnual Enrichment glrements through 2050 (10° SWU/yx) Cumulat i Ke Enrichment through 2050 (10 USOS Utilization @) gchment Utilization 2 (10 SWJ/GWe) 1LY 1257 4059 1732 759 466 4.13 4.75 6.10 6.28 93 0133 ¢ 3.80 5.00 3.02 2523 2793 1652 738 459 2.28 4.60 7.44 7.88 103 2025) 4.87 3.12 1.93 2LU SLU 3415 1501 1622 734 458 1.92 4.43 6.29 6.90 93 (2011) 3.96 2.02 1.16 41U SLY 6LU 71U sy 1815 2701 3296 3338 2727 3501 2615 2020 1978 2589 1743 1643 1624 1624 1638 770 735 735 735 736 474 458 459 459 458 3.41 2.22 1.82 1.77 2.01 4.41 4.63 4.48 4.50 4.60 5.75 7.40 5.75 5.75 6.62 5.94 7.99 5.87 5.89 6.84 119 i1l 101 102 103 (2011) (2023) (2011) (2011) (2017) 4.78 5.26 4.12 4.12 4,73 3.27 2.96 1.78 1.76 2.51 2.63 1,95 1.25 1.23 1.73 Installed Capacity (GWe) in Year 2026/Levelized Power Costs (Mills/Kwhr) in Year 2050 Reactor LWR-US (LE) /U US(LE) /U-EE US (DE)/U/Th U3(DE)/U/Th Pu/U Pu/Th FBR-Pu-U/U Pu-U/Th Pu-Th/Th U3-U/Th Fossil Total Nuclear 1LU 61/22.4 675/21.6 1458 736 2LU 3LU 1028/19.8 827/19.4 - 441/19.2 269/18.7 516/17.3 4LU, 5LU 235/20.6 1108/19.9 489/21.6 336/20.4 0/21.9 143/19.5 235/18.9 6LU 7LU 8LU 874/19.2 872/19.2 1028/19.7 - - - 725 - 1470 - 583 1612 1135 11060 710 1485 0/21.3 0/21.3 0/21.7 219/19.6 221/19.6 280/19.7 63/20.8 56/20.8 119/21.3 486/19.2 509/19.2 - - - 136/20.6 - 0/23.7 0/23.5 553 537 632 1642 1658 1563 (1) nuclear capacity built through 2050. (2) (3) through 2050, Year in which maximum enrichment requ1rements occur, Cumulative U;0g consumed through 2050 {inciuding forward commitments) per cumulative Cumulative enr1chment requ1rements through 2050 per cumu]atlve nuc]ear capac1ty built = > D-10 ! r K _. Table D-5. Variation of 0308 Price with Time for Various Nuclear Cases | U,05 Price ($/1b) Yer 1L 23 A& & & It 8L . 1987 76 81 83 73 82 83 83 82 L 1997 104 112 114 99 113 114 114 113 i 2007 136 150 153 130 150 153 153 151 L 2017 187 177 175 151 177 175 175 175 : 2027 167 185 179 158 184 180 180 180 | L 2037 172 189 180 158 186 180 180 180 | 2047 173 195 180 158 189 180 180 180 L somorers-1 — vm'nT 1 e ~Internal Distribution Reactors 1. L. S. Abbott 2. R. G. Alsmiller 3. T. D. Anderson -4, W. B, Arthur, ORGDP 5. S. Baron 6. D. E. Bartine 7. H. I. Bowers 8. J. T. Bradbury, ORGDP 9. R. E. Brooksbank 10. W. D. Burch 11. T. J. Burns 12. W. L. Carter 13. J. C. Cleveland 14. T. E. Cole 15. A. G. Croff 16. J. G. Delene 17. J. R. Engel 18. D. E. Ferguson 19. G. F. Flanagan 20. M. H. Fontana 21. E. H. Gift | 22. P. M. Greebler (Consultant) 23. P. M, Haas 24, W. 0. Harms 25, J. F. Harvey 26. R. F. Hibbs 27. D. T. Ingersoll 28. J. D. Jenkins 29. D. R. Johnson 30. P. R. Kasten 31. H. E. Knee 32. M. Levenson 33. W. B. Loewenstein (Consu]tant) 34. A. L. Lotts 35. R. S. Lowrie = 36. F. C. Maienschein 37. B. F. Maskewitz - DOE, Wash1ngton, D.C. 20545 - 135. 136. E. S. BeckJord INFCE Coordinator 137, 138. 139, ORNL-5388 Dist. Category UC-80 . R. Meyer . D. McGaugh . R. Mynatt . J. Notz R. Olsen . W. Peelle . Postma Primm . Prince Rainey . Renier . Rohwer Row Santoro Selby . Simard . Smolen piewak Stockdale Til1 . Trauger . Trammell Uhrig (Consultant) Vath Vondra Vondy . Weisbin Whi te . H J F K A R H R. B. R. J. P. T R. D. R. G. I. W. J. D. H. R. J. B. D. C. J. :D:U:ur'mmmcommm:ur'r—'——lzcm-c:l:m—-l . . * - - * - . - . “R. Wilson (Consu1tant) R. G. Wymer A. Zucker Central Research L1brary Document Ref. Section EPD Reports Office Laboratory Records Dept. . Laboratory Records, RC External D1str1but1on W. W. Ballard, Asst. Dlrector, Fue] Cyc]e Deve]opment Harold Bengelsdorf, Office of Nuclear Affairs S. T. Brewer, D1rector,_Program_P]annjng and Analysis Phillip C]ark, Assoc. Director for Reactors, Div. of Naval DOE, Washington, D.C. 20545 (contd.) 140-142, 143, 144, 145, 146, 147, 148, 149, 150. 151. 152, 153, 154, 155, 156. 157. ]58. 159, 160-180. - 181, 182. .7<.U'!OZ - E. G. DeLaney, NASAP Control Office, Office of Fuel Cycle Evaluation D. E. Erb, Division of Reactor Research and Technology H. Feinroth, Division of Nuclear Power Development Neil Go]denberg, Division of Advanced Systems and Materials Production E. J. Hanrahan, Director, Office of Fuel Cycle Evaluation J. R. Humphreys, Program Planning and Analysis Hugh Kendrick, Office of Fuel Cycle Evaluation M. W. Koehlinger, Program Planning and Analysis P. M. Lang, Asst. Director, Light Water Reactor Programs K. 0. Laughon, Asst. Director, Gas Cooled Reactor Programs D. E. Mathes, Office of Fuel Cycle Evaluation W. H. McVey, Division of Nuclear Power Development Marvin Moss, Office of Energy Research C. W. Newstead Office of Energy Research J. A. Patterson Chief, Supply Evaluation Branch, D1v1s1on of Uranium Resources and Enrichment Pressesky, Director, Division of Nuclzar Power Development . F. Savage, Division of Advanced Systems and Materials Production S. Scheib, Division of Nuclear Power Deve]opment Sege, 0ff1ce of Fuel Cycle Evaluation Strauch, Office of Fuel Cycle Evaluation A. Trickett, Office of the Director, Division of Nuclear Power Development DOE, Oak Ridge 183. 184. Asst. Manager for Energy Research and Development Director, Nuclear Research and Development Division Federal Agencies 185, 186. 187. 188. 189. 190. 191. 192, D. L. Bell, Tennessee Valley Authority, 503 Power Bldg., Chattanooga, TN 37401 John Boright, Director, Office of Energy & Technology, Department of State, Rm. 78-30, Washington, DC 20520 D. T. Bradshaw, Tennessee Valley Authority, 503 Power Bldg., Chattanooga, TN 37401 Greg Canavan, Office of Chief of Staff Air Force, Pentagon, Washington, DC 20301 John Depres, CIA Headquarters, 7E 47, wash1ngton DC 20505 H. L. Falkenberry, Tennessee Valley Author1ty, 503 Power Bldg., Chattanooga, TN 37401 _ Joseph Kearney, Office of Management and Budget, 17th and H Street, NW, Washington, DC 20036 S. N. Keeney, Arms Control and Disarmament Agency, Rm., 5934, New State Bldg., Washington, DC 2045] / — T = — r— - . rS Wit .gf;(ff;,. | g, - L r— - r- - o e = - | lf“?(f Federal Agencies (contd.) 193. 194. 195. 196. 197. 198. Louis V. Nosenzo, OES/NET 7830, Department of State, Washington, DC 20520 Joseph Nye, Department of State, Rm. T-7208, Washington, DC 20520 - Robert Rochlin, Arms Control and Disarmament Agency, Rm. 4930, New State Bldg., Washington, DC 20451 Lawrence Scheinman, Department of State, Rm. T-7208, Washington, DC 20520 James Sheaks, Arms Control and Disarmament Agency, Rm. 4933, New State Bldg., Washington, DC 20451 Charles Van Doren, Asst. Director, Arms Control and Disarmament Agency, Rm. 4930 New State Bldg., Washington, DC 20451 Outside Organizations 199. 200. 201. 202. 203. 204, 205, 206. 207. 208. 209, 210. 211. 212, 213. 214, 215. - 216. 217. o O M~ TA G =z OZT G o= » . » " - . . 'w. E. Black, Hanford Engineering Development Leboratory, P. 0. Box 1970, Richland, WA 99352 George Bunn, University of Wisconsin, Madison, WI 53706 A. Carnesale, Harvard University, #9 Divinity Avenue, Cambridge, MA 02138 Y. Chang, Argonne National Laboratory, 9700 South Cass Avenue, Argonne, IL 60439 Thomas Cochran National Resources Defense Council, 917-15th St., NW, wash1ngton, DC 20005 Gordon Corey, Commonwealth Edison Electric, P. 0. Box 767, Chicago, IL 60690 Russell Crowther, General Electric Company, 175 Curtner Ave., San Jose, CA 95125 Joe Cupo, Westinghouse Electric Corporat1on, P. 0. Box 355, Pittsburgh, PA 15230 K. Davis, Bechtel Power Corporation, 50/11/813, P. 0. Box 3965, San Francisco, CA 93119 M. de Montmollin, Sandia Laboratories, Dept. 1760-A, ‘Albuquerque, NM 87185 : . J. Driscoll, Massachusetts Inst1tute of Technology, 138 A]bany Street, Cambr1dge, MA 02139 C. Ed]und Virginia Polytechn1c Inst1tute, B]acksburg, VA 24060 | F. Foran, Resource Planning Assoc1ates, Inc., 3 Enbarcadero Center, Suite 2080, San Francisco, CA 94111 K. Glennan, 11483 Waterv1ew Cluster, Reston, VA 22090 Goldstein, Co]umb1a Un1versity, 520 W. 120th St., New York, NY 10027 : ; - Gordon Resources for the Future 1977 Massachusetts Ave., -Nash1ngton, DC 20036 R. Haffner, Hanford Eng1neer1ng Development Laboratony, P. 0. Box 1970 ‘Richland, WA 99352 : . W. Hardie, Hanford Eng1neer1ng Development Laboratory, P. 0. Box 1970, Richland, WA 99352 William Harris, RAND Corporat1on 1700 Main Street, Santa Monica, CA 90406 Qutside Organizations {contd.) 218. 219. 220. 221. 222, 223. 224. 225. 226. 227. 228. 229, 230. 231. 232, 233, 234. 235, 236. 237, 238. 239. 240. ‘2471. 242, 243. 244, ‘Chuck Hebel, Xerox Palo Alto Research Center, 3333 Coyote Hill Rd., Pa]o Alto, CA 94303 T. M. Helm, Hanford Engineering Deve]opment Laboratory, P. 0. Box 1970, Richland, WA 99352 M. Higgins,'Science App]icatibns, Inc., 8400 Westpark Dr., McLean, VA 22101 William Higinbotham, Brookhaven National Laboratory, Associated Universities, Inc., Upton, NY 11973 - Fred Hoffman, RAND Corporat1on, 1700 Ma1n Street, Santa Monica, CA 90406 William W. Hogan, Stanford University, Stanford, CA 94305 W. G. Jolly, Hanford Engineering Development Laboratory P. 0. Box 1970, Richland, WA 99352 J. M. Ka]]fe]z, Georg1a Institute of Techno]ogy, Atlanta, GA 30332 Arthur Kantrowitz, AVCO Everett Research Lab 2385 Revere Beach Parkway, Everett, MA 02149 Walter Kato, Argonne National Laboratory, 9700 South Cass Ave., Argonne, IL 60439 John Kearney, Edison E1ectr1c Institute, 1140 Connecticut Ave., Washington, DC 20036 Herbert Kouts, Brookhaven National Laboratory, Associated Universities, Inc., Upton, NY 11973 F. W. Kramer, westinghouse Nuclear Fuel Division, P. 0. Box 355, Pittsburgh, PA 15230 Myron Kratzer, International Energy.Associates, 2600 Virginia Ave., NW, Suite 200, Washington, DC 20037 . C. Lipinski Argonne National Laboratory, 9700 South Cass Ave., Argonne IL 60439 A. S. Manne, Terman Eng. 432A, Stanford Unlvers1ty, Stanford, CA 94305 G. Nugent, Burns & Roe, P. 0. Box 663, Route 17, South, Paramus, NJ 07652 R. P. Omberg, Hanford Engineering Development Laboratory, P. 0. Box 1970, Richland, WA 99352 B. Pasternak, Booz-Al]en and Hamilton, 4330 East—west H1ghway, Bethesda, MD 20014 R. G. Post, University of Arizona, Tucson, AZ 85721 Richard Richels, Electric Power Research Institute, P. 0. Box 10412, Palo Alto, CA 94303 C. L. R1ckard General Atomic Company, P. 0. Box 81608, San Diego, CA 92138 David Rossin, Commonwealth Edison E]ectr1c Co., P. 0. Box 767, Chicago, IL 60690 H. S. Rowen, Stanford University, Stanford, CA 94305 Thomas Schelling, JFK School of Government, Harvard University, Cambridge, MA 02138 N. L. Shapiro, Combustion Engineering, 1000 Prospect H11] Rd., Windsor, CT 06095 M. R. Shay, Hanford Engineering Development Laboratory, P. 0. Box 1970, Richland, WA 99352 r— r- r- - r— - - Irm v—" —ad ! r— rr— r— r— ITTT‘:'"fi r Outside Organizations (contd.) 245, 246. 247. 248. 249. 250. 251. 252, 253. 254, 255. 256. 257-396. B. I. Spinrad, Dept. of Nuclear Eng1neer1ng, Oregon State University, Corvallis, OR 97331 Chauncey Starr, Electric Power Research Institute, 3412 Hill View Ave., Palo Alto, CA 94303 H. B. Stewart, Nuclear Technology Evaluations Company, 4040 Sorrento Valley Blvd., Suite F, San Diego, CA 92121 S. M. Stoller, S. M. Stoller Corporation, 1250 Broadway, ‘New York, NY 10001 E. Straker, Science Applications, Inc., 8400 Westpark Drive, ~ MclLean, VA 22101 C. E. T111 Argonne National Laboratory, 9700 South Cass Ave., Argonne, IL 60439 James Tulenko, Babcock and Wilcox, P. 0. Box 1260, Lynchburg, VA 24505 R. F. Turner, General Atomic Company, P. 0. Box 81608, San Diego, CA 92138 Frank von Hippel, Program on Nuclear Policy Alternatives, Center for Env1ronmenta1 Studies, Princeton University, Princeton, NJ 08540 Eugene We1nstock Brookhaven National Laboratory, Associated Universities, Inc., Upton, NY 11973 A. Weitzberg, Science Applications, Inc., 8400 Westpark Drive, McLean, VA 22101 Albert Wohlstetter, 518 S. Hyde Park Blvd., Chicago, IL 60615 For Distribution as Shown in TID-4500 under UC-80, General Reactor Technology ¥.S. GOVERNMENT PRINTING OFFICE: 1979-640-079/374