G s s GREEL/TH-6415 R J. R. Engel . F. Bauman J. F. Dearing W, B, Grimes RS s = el ¥ r. 2 m‘g -.l' o ‘.9' '. e W‘W e 2 ¥ e e et ot R S T o % o o e S P .; 2 = . - R e R o 2R ORI i e AN R Printed in the United States of America. Available from National Techrical Information Service ; Li.5. Department of Commerce % 5285 Port Royal Road, Springfield, Virginia 22161 ; ! . . ’ : . ! Price: Printed Copy $8.00; Microfiche $2.00 fi This report was prepared as an ascount of work sponsared by an agency of the United i Stalzs Government Neither the United Bigtes Government nor any agency thereof, nor | any of ineir emplovees, contractors, subconiractors, or their employess, makes any L warranty, express or implied, nor assumes any legal Habitity or responsitility for any | third party's use or the results of such use of any information, apparatus, product or + process disclesed in this repaort, nor represents that s use by such third oarty wouid not infringe orivately owned rights. ORNL/TM-6415 Dist. Category UC-76 Contract No. W-7405-eng~26 Engineering Technology Division DEVELOPMENT STATUS AND POTENTIAL PROGRAM FOR DEVELOPMENT OF PROLIFERATION-RESTISTANT MOLTEN-SALT REACTORS J. R. Engel H. F. Bauman W. R. Grimes J. F. Dearing H. E. McCoy, Jr. Date Published: March 1979 NOTICE: This report contains information of a preliminary nature. It therefore does not represent a final report and is subject te changes or revisions at any time. r“—'—"‘—“—— KOTICE - | This weport was prepared as m SPO‘HSUmd by the United States Government. Neither the | Prep&red by the EZ:‘T:S States ;lOr f_!h_e .L'nited States Department of » ML any of their emplovees, nor any of their OAK RIDGE NATIONAI LABORATORY :fi;”;:::’::t;u:;:fl:zctus_s, 011" their err_ipl!.vye.es, makes 5 Oak Ridge, Tennessee 37330 1mmwf£wmmw$fit$$fl§$fi§' ¥ or usefulzess of any information, apparatus, product or Operated by pr;)c.:ess di.tscloscd, :1 (epl:Bsc:i(S tha[;pits :use, ch):ld tnct UNION CARBIDE CORPORATION L Py el e i‘ for the DEPARTMENT OF ENERGY iii CONTENTS EXECUTIVE SUMMARY ..... Seecesrescososesesenene e crceane tecccoaa e ABSTRACT lllll & # £ # & 2 3 » & & & © ® & o 3 8 & & ® ¢ 0 v O 9 4 & ® & & & & 0 3 ¢ & ¢ ® ¢t o &6 & o & & ® & e & & @ INTRODUCTION ...vevevncon seesrcosaeerrescoosaensooo tetceaenannsna REf G eIICEeE vt cc oo esuonconosasccsssnnsoscconsesssoccscnsnosnsscesoee PART I. REACTOR DESIGN AND DEVELOPMENT REACTOR DESIGN, ANALYSIS, AND TECHNOLOGY DEVELOPMENT ........ Status In 1972 it evensvesccossenssssososesssenosnssssssasasss Current Development StatuS ...iiceneeetnosccocososcescoscsnscee DMSR Development Needs ....i.ceeecrenrscenscecosrsascsscssssssscs Ref ey eNCeS t i v eecvcocoranaocosssnnancsoseeoncsoasoensnoccesosos PART II. SAFETY AND SAFETY-RELATED TECHNOLOGY REACTOR SAFETY AND LICENSING ...iveenreicoenorrvoccocnsonvcsos Status and Development Needs .c.oveievccocsocervesoooansscoonns SafELY v creeeeeneerconrsrtcncassssnescccassactoconssosnes Licensing «.ceverveesnneorconontanocenseronnoanoncnncennnss Estimates of Scheduling and Costs ...cveerenciecrnnnenascccans References ..ccieoeaesnn e bt e aeset e et o s e st e s e s o ane s e aeanennns PART ITII. FUEL-COOLANT BEHAVIOR AND FUEL PROCESSING FUEL AND COOLANT CHEMISTRY ....c.iticiieinninnccannnssnenuoaneans Key Differences in Reactor Concepts ....eicercencerconnceocss Post—-1974 Technology Advances ....eeeevsovess Citecenset e Status of Fuel and Coclant Chemistry ..cciiierencecieercanans Fuel chemistry c.eevierterricnsescerctsanenconsancsnnnnsonnas Coolant chemiStry ¢evieceneecronsnrsnccs seaasas cevessvaens Fuel—-coclant interactions .....ciceueeeescrsrsacessaccsconsas Fuel-graphite Interactions ......cciceerrvcrocnonsencoones Prime RED NeedsS v.veeeeeecoonsecesecnooassoacoansesssocosssnses Fuel chemistry .......... feeecaien e ero s tecencavaeanas Coolant chemistry ...cv.iii e ieacncacnnnas sssescaas Fuel-coolant interactions ....eceeieniiececencenncs taanseea Fuel-graphite Interactions ..o.veeecccoenrccccaososasncons Estimates of Scheduling and CostsS ...vveiivecvcoonaniecoonenns 11 14 23 23 23 25 26 26 31 31 32 34 34 an 47 48 49 49 50 50 50 o1 iv ANALYTICAL CHEMISTRY . cviveeveecoeecnonoceccasonnnscsconsssssco Scope and Nature of the Task ...... o eecceosoass e e Key Differences in Reactor ConceptsS ..osveercssoseccconncaons Post—-1974 Technology Advances «...veeveccocnorsrccononnns ee e Status of Analytical Development ........... cecosssscee e Peee Key developments for MSRE .....iciiccneeneientccsncarcccnn Analvytical development for MSBR ..... cecenn ceriserecoeenes Prime Development NeedS .....ccovtvevccccnncarconns et eecenss Estimates of Scheduling and CosSts ...ovveee ceceseascasnurene MATERIALS DEVELOPMENT FOR FUEL REPROCESSING .cccocvencsne v Scope and Nature of the Task .cccocencrcerrernosrercccsoanens Key Differences in Reactor CoOnCeptsS .ceoeveeccencanscncaccsns Post=1974 Technology Advances .vecoee.s Ctccosseaseneceanaenan Present Status of Technology ..ceiciiercnneneccoonnnncacasnsnn Materials for fluorinators and UFg absorbers ....ceececeos Materials for selective exXtractionsS vecccoooreccesacnsonnas Primary R&D Needs ...uvirecieerecccconsconansans theumassanene Estimates of Scheduling and CostsS ....icieiicetenenccosnconca FUEL PROCESSING .t cvveeteveeconosnsscasnsanssssas ceeesesenoan Scope and Nature of the Task ..ieicinieeereiricnorncvcoancssn Key Differences in Reactor Concepts ..cceeeisorianicasanaercs Post—1974 AdVANCES +ceceeencnncootnnsassassaconssssnasocsosas Status of Technology teveeeeosscearsssseceonsnsennsanncasannss Chemical status ......... cisornsas cesassresacs Ceessserienos Conceptual MSBR processing flowsheet ........cccciivnan. Fngineering StAlUS +.iceereracsceoanscssessernsasssacconsss Special characteristics of DMSR processing ...cceceeviean. Possible processing alternatives .......co00. cecresscocan Primary R&D Needs «uiviiiirieivensessiivesccosseassscosssosssnss Estimates of Scheduling and CosSts vuiiveicniioronnsnecsoncons Ref TN C eSS ot v et o veneocceanoanonscsossenaonsssccesnsoe e seees s PART IV. REACTOR MATERIALS STRUCTURAL METAL ¥OR PRIMARY AND SECONDARY CIRCUITS ...cceo-.. StAtUS TN 1977 oottt enoeenececoooeeasanossescoonssesssoeconnsnss Status Iin 19748 ... et evoossnesccoses e o e o s ecct o anneneseannne Page CUrTent SEALUS «veeseosansesscecoansescasoansoscosasssnscosss 130 Further Technology Needs and Development Plan ............... 130 GRAPHITE FOR MOLTEN-SALT REACTORS ............. ceciecneonnons 135 Status In 1972 . i. et iennroonssosassersosasssacconasssascccoas 135 Status Inm 19760 «uveveoeciorosssnssnsconssanssascoossssonacnonsnso 138 Current StafUuS ..cecrvocoosssenccoosana tee e ceesecoas et e oo 138 Further Technology Needs and Development Plan .....c.veeecoons 138 vii EXECUTIVE SUMMARY INTRODUCTION Molten-salt reactors (MSRs) are of interest in possible prolifera- £33y power plants that tion-resistant systems, particularly as denatured could be widely deployed with minimal risk of proliferation. MSRs might also be used as "fuel facteries" in secure centers, burning plutonium and producing 2337, However, before they can be used, the MSR concept must be developed into a commercial reality. The purpose of this report is to review the status of molten-salt technology from the standpoint of the development required to establish an MSR industry. Following the successful operation of the Molten~Salt Reactor Ex- periment (MSRE, 1965—69), it became necessary for the government to de- cide if MSR development should be continued. To this end, a comprehensive report on MSR technolegy was published in August 1972.' Because only limited R&D has been conducted since then, most of the information in the report is still valid and will be taken as the basis for the present review. Some additional development work done in 1974—76 will be used to update the conclusions of the 1972 study. The government decided not to proceed with the further develecpment of the Molten—-Salt Breeder Reactor (MSBR), or any other MSR, for reasons other than technological ones. DEVELOPMENT STATUS, 1972 The development status of MSBRs in 1972 is covered thoroughly in Ref, 1. All aspects of reactor development, from reactor physics to materials of construction, are covered and wilil not be repeated here. Of particular interest in that review are the discussions of technologi- cal advances believed to be needed before the next MSR could be built. These needed advances are defined briefly in the introduction of that report as follows: "In the technology program several advances must be made before we can be confident that the next reactor can be built and operated success-— fully. The most important problem to which this applies is the surface viii cracking of Hastelloy N, Some other developments, such as the testing of some of the components or the work on latter stages of the processing plant development, could actually be completed while a reactor is being designed and built. The major developments that we believe should be pursued during the next several years are the following: "l1. A modified Hastelloy N, or an alternative material that is im- mune to attack by tellurium, must be selected and its compatibility with fuel salt demonstrated with out-of-pile forced-convection loops and in- pile capsule experiments; means for giving it adequate resistance to radiation damage must be feound, if needed, and commercial production of the alloy may have to be demonstrated. The mechanical properties data needed for code qualification must be acquired if they do not already exist. "2, A method of intercepting and isolating tritium to prevent its passage into the steam system must be demonstrated at realistic conditions and on a large enocugh scale to show that it is feasible for a reactor. "3. The various steps in the processing system must first be demon- strated in separate experiments; these steps must then be combined in an integrated demonstration of the complete process, including the materials of construction. Finally, after the MSBE* plant is conceptually designed, a mock-up containing components that are as close as possible in design to those which will be used in the actual process must be built and its operation and maintenance procedures demonstrated. "4, The various components and systems for the reactor must be de- veloped and demonstrated under conditions and at sizes that allow con- tident extrapolation to the MSBE itself. These include the xenon strip- ping system for the fuel salt, off-gas and cleanup systems for the coolant salt (facilities in which these could be done are already under construc- tion), tests of steam—generator modules and startup systems, and tests of prototypes of pumps that would actuaily go in the reactor. The construc- tion of an engineering mock-up of the mazjor components and systems of the reactor would be desirable, but whether or not that is done would depend Molten-Salt Breeder Experiment; an intermediate-scale developmental plant. ix on how far the development program had proceeded in testing various com-— ponents and systems individually. "5. Graphite elements that are suitable for the MSBE should be @ purchased in sizes and quantities that assure that a commercial produc- tion capability does exist, and the radiation behavior of samples of 4 the commercially preduced material should be confirmed. Exploration of methods for sealing graphite to exclude xenon should continue. "6. On-line chemical analysis devices and the various instruments that will be needed for the reactor and processing plant should be pur- chased or developed and demonstrated on lcops, processing experiments, and mock-ups." The first three objectives were considered crucial to the MSBR con- cept; the results of further development effort on them during 1974—76 are discussed in the following section. Objectives 4 to 6, while im- portant, did not appear to present any insurmountable obstacles; in any event, they could not be pursued further because of limited funding. RESULTS OF R&D — 1972 TO PRESENT At the direction of AEC/ERDA,* the MSR program was discontinued in early 1973, resumed in 1974, and finally terminated at the end of FY 1976. Although the development effort since 1972 has been severely re- stricted, some significant results were obtained from work performed mainly in 1974—76. Alloy Development for Molten-Salt Service The nickel-based alloy Hastelloy N, which was specifically developed for use in molten-salt systems, was used in construction of the MSRE. The material generally performed very well, but two deficiencies became apparent: (1) the alloy was embrittled at elevated temperatures by ex- posure to thermal neutrons and (2) it was subject to intergranular sur- face cracking when exposed to fuel salt containing fission products. "Now the U.S. Department of Energy. Recent development work indicates that solutions are available for both these problems. Details of this work are given by McCoy;2 a summary of the results follows. Irradiation experiments early in the MSR develcopment program showed that Hastelloy N was subject to high-temperature embrittlement by thermal neutrons. The MSRE was designed around this limitation (stresses were low and strain limits were not exceeded), but the development of an im- proved alloy became a prime objective of the materials program. It was found that a modified Hastelloy N containing 27 titanium had much im- proved postirradiation ductility, and extensive testing of the new alloy was under way at the close of MSRE operations. The second problem, intergranular surface cracking, was discovered at the close of the MSRE operation when surface cracks were observed after strain testing of Hastelloy N specimens that had been exposed to fuel salt. Research since that time has shown that this phenomenon is the result of attack by tellurium, a fission product in irradiated fuel salt, on the grain boundaries. As a result of research from 1974 to 1976, two likely soclutions to the problem of tellurium attack have been developed. The first involves the development of an alloy that is resistant to tellurium attack but still retains the other required properties. This development has pro- ceeded sufficiently to show that a modified Hastelloy N containing about 1% niobium has good resistance to tellurium attack and adequate resistance to thermal-neutron embrittlement at temperatures up toc 650°C. It was ‘also found that alloys containing titanium, with or without niobium, ex- hibited superior neutron resistance but were not resistant to tellurium attack. The secend likely solution involves the chemistry of the fuel salt. Recent experiments indicate that intergranular attack on Hastelloy N is much less severe when the fuel-salt oxidation potential, as measured by the ratio of utt to U3+, is less than 60.% This discovery opens up the possibility that the superior titanium-modified Hastelloy N could The inverse of this ratio, that is, the ratio of Ut o UMt now more commonly used to describe the oxidation state of the salt. xi be used for MSRs through careful control of the oxidation state of the fuel salt. Both of the above scolutions appear promising, but extensive testing under reactor conditions would be required before either could be used in the design of a future MSK. Tritium Control Large quantities of tritium are produced in MSRs from neutron reac- tions with lithium in the fuel salt. FElemental tritium can diffuse through metal walls such as heat-exchanger tubes at elevated temperatures, thus providing a potential mechanism for the transport of tritium to the reactor steam via the secondary coolant locp and the steam generator. Recent experiments indicate that tritium is oxidized in the proposed MSR secondary coolant, sodium fluoroborate, thus blocking transpert to the Steam system. In 1975 and 1976, tritium—additiocn experiments were conducted in an engineering-scale coclant salt test loop. The results are given in a report by Mays, Smith, and Engel.3 Briefly, the experiments showed that the steady-state ratio of combined to elemental tritium in the coolant salt was greater than 4000. A calculation applying this ratio to the case of an operating 1000-MW(e) MSBR indicated that the release of tritium to the steam system would be less than 400 GBg/d (10 Ci/day). The conclusion of the study was that the release of tritium from an MSR using sodium fluorcborate in the secondary coolant systfem could be readily controlled to within Nuclear Regulatory Commission (NRC) guidelines. Engineering Development of Fuel Processing By 1972, proof-of-principle experiments had been carried cut for the various steps in the reference chemical preocess, but development and demonstration of engineering-scale equipment were just getting under way. The only large-scale processing demonstrated at that time was the batch fluorination of the MSRE fuel salt and the recovery of the uranium on NaF beds. xii In the period 1974—76, efforts were begun to develop items of equip- ment which would be vital to the success of the metal-transfer process. Some progress was made in the develepment of a salt-bismuth contactor, a continuecus fluorinator, and a U¥g absorber for reconstituting the fuel salt.” Because of the pregram closeout in 1976, this work could not be continued long enough to culminate in engineering designs for the various items of equipment. The status of this work can be summed up by stating that, although no insurmountable obstacles were encountered, the major portion of process engineering development remains to be done. Other Areas of Development The development status of areas other than those discussed above is practically unchanged since the report1 of 1972, because no further R&D was funded. These include development of reactor components, moderator graphite, analytical methods, and controcl instrumentaticn. Exceptions were a design study of a mclten-salt heat exchanger and some limited work on the in-line monitoring of fuel salt. In 1971, Foster-Wheeler Corp. was awarded a contract for a study of ¥ MSR steam-generator designs. The contract was suspended in 1973 and then reinstated in 1974 for the purpose of completing the first task (in a four—-part contract), which wag the design of a steam generator to meet specifically the steam and feedwater conditions postulated for the MSBR conceptual design. This task was successfully completed and a report issued in December 1974.° A design was presented which, based on analy-~ sis, would meet gll the requirements for an MSR steam generator. How- ever, the design was not experimentally verified because the MSR project was terminated. The 1972 status report1 described the use of an in-line electro- chemical technique known as voltammetry to monitor the oxidation poten- tial of the fuel salt. The technique has since been used to monitor various corrosion test loops and other experiments and may also be used 2+ to monitor Cr in fuel salt, a good indicator cof the overall corrosion \ rate. Recently the technique has been used to measure the oxide ion in xiidi fuel salt. Oxide monitoring is very important in molten-salt fuel be- cause an increase in oxide contamination could lead te precipitation of uranium from the fuel as UQ». SPECIAL DEVELOPMENT REQUIREMENTS FOR THE DMSR Recent reexamination of the MSR concept with special attention to antiproliferation considerations has led to the identification of two preliminary design concepts for MSRs that appear to have substantially less proliferation sensitivity without incurring unacceptable perfor- mance penalties. The designation DMSR {(for denatured molten-salt reac- tor) has been applied to both of these concepts because each would be 2359 enriched to no more than 20% and would be fueled initially with cperated throughout its lifetime with denatured uranium. The simpler of these DMSR concepts6 would completely eliminate on- line chemical processing of the fuel salt for removal of fission products. (Stripping of gaseous fission products would be retained, and some batch- wise treatment to control oxide contamination probably would be required.) This reactor would require rcutine additions of denatured 235y fuel, but would not require replacement or removal of the in-plant inventory except at the end of the 30-year plant lifetime. Adding an on-line chemical processing facility to the 30-year, once-through reactor provides the second DMSR design concept.’ With this addition, the conversion ratio of the reactor would reach 1.0 (i.e., break-even breeding) so that fuel additions could be eliminated and a given fuel charge could be used in- definitely by transferring it to a new reactor plant at the decomission- ing of the old unit. The required chemical processing facility for a DMSR, shown as a pre- liminary conceptual flowsheet in Fig. S.1, would be derived largely from the MSBR but would contain some significant differences. In particular, isolation and segregation of protactinium would be avoided, provisions would be made to retain and use the plutonium produced from “*%U , and a special step would be added for removal of fission-product zirconium. Thus, the development of on-line chemical processing for a DMSR would ORNIL DWG 78 5764 l§frs 23 sz o cmm 1 e . e e 3 L e e e e . £ e £ e e e e i [T Cs AND Rb [ACCUMULATOR UF,, ThF,, BeF, g - REGCYCLE : e i ADDITION ' r g ‘.‘+ RE‘?2 < o e i . o | ; ACCUMULATION & i ¥ VALENCE ; UF, UFg i E IN Bi = i y AR&REI" | |REDUCTION ABSORFTION] 1y, ! | F«~+ r=+y é | i Bi — ; = = —ADo L ADD REZ |t — e i | REMOVAL| bl | f : LiCl+ |STRIPPER)ag . 4 | i TO & " ' T + ' i ® P ¥ ! ! ; REACTOR i __Jmmn» Lob T RARE 1y mmm———————— ——————————— B — EARTH | 7 ! { P P T L [ransrer] g i ADD L § E )\ { PARTITION g | EARTH P E : e Wy s et _ P ¢ | i. EXTRACTION i) g Lic: ! REY ACCUMULATION| g oo | | [ S ] , | — | IN Bi l P P ! —_— e DT T e T + * i ! b ’ I ¥ i P e ! ! y KT T ’ Pl o o ] s ) e o $ i EXTRACTION el EXTRACTION P coLvenT] 1 =+ 2| STRIPPER i===%1c| jorin-| F | i 4 ! ; I e b . T | ATION | 11 E i b rfm | i T ; ; FROM . T | l I e -l ‘ L] R : REACTOR ‘ Il,‘ I + t_.__.- 4 ™1 PRIMARY [ p ; n : F T F.P. AND WASTE SALT P FLUORINATOR |ag2. SECONDARY|skiie] HYDRO- L + | FOR RECOVERY OF kit ottt ] U P | FRUORINS oy PrY0RING s F ULTIMATE DISPOSAL P ATOR ' ATOR bbbt | T w i 4 — e —_ eer o oo o e e o e ¥ ! T ! i FUEL OR FUEL SOLVENT " bbb e ¥ ----- BISMUTH - = -= T e e e e e =GP UFB ~ 4 e |G b WASTE SALT Fig. S.1. Proposed DMSR fuel processing flowsheet. XV require essentially all the technology development identified for the MSBR with additions to accommodate these differences. However, since the DMSR offers a no-processing option, a large fraction of the repro- cessing development, along with its associated materials development, could Be deferred or even eliminated. Such deferral might be expected to reduce the cost (but probably not the time) for developing the first DMSRs. To provide an overall perspective, this development plan includes costs and schedules for developing the reprocessing capability in parallel with the reactor. The only other substantial difference (in terms of development needs) between the MSBR and the proposed DMSR concepts is the reactor core design, which is similar for both. Relaxing the breeding requirement and empha- sizing proliferation resistance for the DMSR led to a core design with a much lower power density to limit losses of protactinium, the 233y pre- curser which is retained in the fuel salt of the DMSR. By reducing the rate of fast-neutron damage to the core graphite, the low power density also makes possible the design of a core in which the graphite need not be replaced for the life of the reactor. A low power density also re- duces the poison fraction associated with xenon in the core graphite and thus there is less need for a low-permeability graphite. Although im- provements in graphite life and permeability would be desirable, graph- ite grades tested before 1972 would have the properties acceptable for the DMSR core. Graphite development for the DMSR would not require (but could include) much effort beyond the gspecification and testing of com- mercial-source material. POTENTIAL PLAN FOR DMSR DEVELOPMENT A major product of the reactivated MSR program in 1974 was a de- 8 for the first several years of a development effort that tailed plan would ultimately lead to a commercial MSBR. Since the program authorized in 1974 was restricted in scope, no attempt was made in that plan to include costs and schedules for reactor plants beyond a limited treat- ment of a proposed next-generation reactor — the Molten-Salt Test Reactor XVvi (MSTR). The primary function of the 1974 program plan was to define a base technology program for the MSBR. Since the technology needs for a DMSR closely parallel those of the MSBR, extensive use was made of the 1974 program plan in evolving the plan described below for DMSR develop- ment, To develop a reasonable perspective of the potential role of the “ DMSR in providing nuclear electric power, it is necessary to concep- tualize a reactor development and construction schedule that goes beyond the MSTR to at least the first commercial (or prototype) system and pos- sibly on to the first of a series of "'standard" plants. The potential schedule that was developed (Fig. S.2) has a reasonable basis for ful- fiilment in the iight of the current state of MSR technoleogy. Four generally parallel lines of effort would be pursued, including: 1. a base program of research and development (R&D): 2. a project to design, build, and operate an MSTR; 3. a project to study and eventually design and build a prototype, or first commercial, reactor plant; 4. a project to design and build the first of possibly several ''stan- dardized" plants. . If adequate guidance is to be provided for an R&D program on MSRs, it 1s essential that some design activity be started on the prototype reactor and the MSTR at the beginning of the overall program. (These initial design efforts may be relatively small, however.) A prototype concept is required to define the systems tc be tested in the MSTR, and the MSTR design is required to guide the initial phases of the R&D effort. If such a program were started in FY 1980, the development and de- sign activities could probably support authorization of a test reactor in FY 1985, and such a reactor could probably be built by 1995. The prototype commercial plant (supported by earlier design study) could be authorized approximately on completion of the MSTR, and the authorization for the first standard plant (if desired) could follow about 5 years ‘ later. Although the technology development effort is shown as only a single line on Fig. S.2, it represents a multifaceted effort in support of all o A . - ORN DWG T8 70682R Tha N0 t‘;" I"“'—'\v.‘;_ YEerRA T R&D @ TECHNOLOGY DEVELOFMENT » LOPERATE \ UPERATE FOR S DEVELOPVENT / l / AUTHORIZE l / | / aT R o N LA o P j f N J/ DETAILED DESIGN | ! : / | | / | ’ / i i / h / CONSTRUOT QFERATE w ! E / 9(45 _ ¢ (L 2005 - < | I / o ! AUTHORIZE | / VEAD | / DETALLED DESIGN (“"‘:"'\'W:f:E-fiRC‘AL PRECONZEPTUAL | {CUNCEPTUAL L » - ONPRFLIN NARY BROTOTYPE / DESIGN DESIGN - DESIGN L E’ | | | | | n iAUTHORiZE 1) +irN 4 4 STANDARD TESIGN UMSR Fig. S.2. Potential reactor construction schedule for commercialization of DMSREs. xviii the reactor construction projects throughout the program. This effort, g which is described in some detail in the body of this report, is summa- rized in Table S.l1 alcong with estimated costs (in unescalated 1978 dollars). This tabulation includes the estimated costs for development of full reprocessing capability. A substantial fraction of these costs (shown as $147 million over 32 years) might be deferred or saved if development of on~line fuel reprocessing were deferred or eliminated. The work for the first 15 vears is shown on an annual basis, with most of the effort in support of the MSTR. In general, the funds shown here are consistent with the more detailed tabulations presented in the body of this report. However, in a few areas the development plan indicates that additional, undefined costs could be expected in some vears. For purposes of this summary tabulation, funds were arbitrarily added in those areas to cover reasonable extra costs. Costs after the first 15 years are much less certain and are shown as totals only. The estimated cost of the total base program is approximately $700 million. The costs of the reactor construction projects, about $600 million®* for the MSTR and possibly $1470 million* for the prototype, bring the estimated total program cost to about $2.8 billion. Since it is impossible to foresee all needs and costs for a program, this is probably az minimum figure. A contingency allowance should be added in a subsequent planning stage, as well as allowances for cost increases due to inflation and escalation and for any development contributions provided by industry. * These figures include the costs of integral chemical processing facilities and are consistent with Nonproliferation Alternative Systems Assessment Program (NASAP) guidelines. W and development costs for MSR hase development progran (thousands of 1978 doilars} Projected researc Buvelnpmefit activity Reactor design and analysis Reacter and component technology safety and licensing Fuel and coolant chemistry Analytical chemisrry Process materials Fuel processing technology Structural alioy Moderator graphite <. b A Includes funds authorized for major development facility. Type fund Target reactel Coat by fiscal year Operating Qperating Qperating Operating Capital Operating Operating Operating Operating Capital dperating Gperating Capital Operating Capital Operating Operating Capital Operating Operating Capital Qpevrating Uperating Capital “Total funds rhrough 2011: MSTR Dema MSTR Demo Al MSTR Demo MSTR Deno All MSTR Demo ALl MSTR All MSTR Demo All MSTR Deama ALl MSTR Demo A11 . . e Tatal funds L o . e Includes cnats estimated without detailed propram anaiysis. $702, 563, —— e e Cost from 1993a through 2011 1,000 20,300 26,000 BG, 000 8,000 8,360 40,000 3,000 15,000 2,500 2,000 5,300 l,QfIQ 1,420 g J L ¥ wn o oo 3,000 8,000 1,980 331,685 XTIX XX REFERENCES M. W. Rosenthal et al., The Development Status of Molten-Salt Breeder Reactors, ORNL-4812 (August 1972). H. E. McCoy, Jr., Status of Materials Development for Moltern-Salt Feactors, ORNL/TM-5920 (January 1978). G. T. Mays, A. N. Smith, and J. R. Engel, Distridution and Behavior of Tritium in the Coolant-Salt Technology Facility, ORNL/TM-5759 (April 1977). Molten-Salt Reactor Program, Semianmnual Progress Report for Periocd Ending February 289, 1976, ORNL-5132 (August 1976). Foster-Wheeler Energy Corporation, Task I Final Report, Design Studies of Steam Generatore for Molten-Salt Reactors, ND/74/66 (December 1974). J. R. Engel et al., Conceptual Design Characteristics of a Denatured Molten-Salt Reactor with Onece-Through Fueling, ORNL/TM (in prepara- tion). J. R. Engel rpt g e Utirlisation t al., Molten-Salt Reactors for Efficient Nuclear Fuel /. thout Plutnotum Separation, ORNL/TM-6413 {(August 1978). o =~ L. E. McNeese et al., Program Plan for Development of Molten-Salt Bregder HReactors, ORNL-5018 (December 1974). DEVELOPMENT STATUS AND POTENTTIAL PROGRAM FOR DEVELOPMENT OF PROLIFERATICN-RESISTANT MOLTEN-SALT REACTORS J. R. Engel H. F. Bauman W. R. Grimes J. F. Dearing H. E. McCoy, Jr. ABSTRACT Preliminary studies of existing and conceptual molten- salt reactor (MSR) designs have led to the identification of conceptual systems that are technologically attractive when operated with denatured uranium as the principal fissile fuel. These denatured MSRs would alsc have favorable resource-utili- zation characteristics and substantial resistance to prolifera- tion of weapons—usable nuclear materiais. This report presents a summary cof the current status of technology and a discussion of the major technical areas of a possible base program to de- velop commercial denatured MSRs. The general areas treated are (1) reactor design and development, (2) safety and safety re- lated technology, (3) fuel-coolant behavior and fuel process-— ing, and (4) reactor materials. A substantial development effort could lead to authoriza- tion for construction of a molten-salt test reactor about 5 vears after the start of the program and operation of the unit about 10 years later. A prototype commercial denatured MSR could be expected to begin operating 25 years from the start of the program. The postulated base program would extend over 32 years and would cost about $700 million (1978 dellars, unescalated). Ad- ditional costs to construct the MSTR — $600 million — and the prototype commercial plant — $1470 miilion — weould bring the total program cost to about $2.8 billion. Additional allow- ances probably should be made to cover contingencies and in- cidental technology areas not explicitly treated in this preliminary review. INTRODUCTION A concept for a proliferation-resistant molten-salt reactor (MSR) 2 2 , i 33y and/or 2°°U has been evolved in response to fueled with denatured the interest in proliferation-resistant power reactors for worldwide use. Briefly, such a reactor (1) must not provide a tempting or readily available source of weapons material; (2) must have good economics and fuel utilization and be competitive with reactors generally used or planned for use in nuclear weapons states; and (3) must provide reason- able energy independence for the nonnuclear weapons states that adopt it (i.e., an assured source of fuel and/or reprocessing capability). The proposed denatured molten-salt reactor (DMSR) concept, described in general below, meets these requirements for the feollowing reasons: 1. the fissile material is denatured and/or confined within a contained highly radicactive system; 2. the projected economic performance is competitive with other existing or proposed reactor systems; 3. uranium resources would support at least five times the electrical capacity in DMSRs as in light-water reactors {(LWRs) on a once-through cycles 4, each DMSR, as a break-even breeder {conversion ratio = 1.0) with on- line processing, once started would not need an ocutside source of fisgile fuel indefinitely. (However, fertile material and the makeup salt constituents, 'Li and beryllium fluorides, would have to he ( supplied.) At least two other MSR concepts may be attractive for proliferation- resistant systems. Their development will not be specifically considered in this report; however, they differ only in detail from the DMSR, and their development would require the solution of essentially the same prob- lems. The two concepts are a partially self-sustaining DMSR without on- line processing and a plutonium-thorium MSR designed to consume plutonium and produce 2337 for use in denatured reactors. The development of a DMSR without on-line processing would be a relatively modest extension of current technolegy and could presumably be accomplished in a shorter time and with considerably less development effort than the proposed DMSR with processing. This version could not be a break-even breeder but would still be a high-performance converter with significantly improved fuel utilization over LWRs. Addition of an ¢ on—-line fission product processing facility at some later date would transform the plant into a breeder. Preliminary results indicate that a fuel charge could last for the entire 30-year life of the reactor, at 75% capacity factor, with only routine additions of 238y and/or denatured 2353, A more detailed characterization of this concept is in progress. A plutonium-fueled MSR could be designed for use in a secure energy center as a "fuel factory" to produce U for use in denatured reactors. The outstanding advantage of an MSR for this application is the ability to remove product 233y from the circulating fuel about as fast as it is formed, so that very little is consumed by fission within the reactor it- self. Cycle times of V10 days for uranium removal are considered feasi- ble, compared with reprocessing times on the order of years for seclid- fuel reactors. An MSR on this fuel cycle has been estimated to produce 750 kg of 233y per GW(e)evear at 0.75 plant factor; this is several times more than that produced by any other type of thermal reactor and about the same as expected from a plutonium-thorium liquid-metal fast breeder reactor (LMFBR). However, the MSR would consume half again as much plu- tonium or more. More quantitative fuel cycle data are not available at this time. The nominal DMSR with processing is based on the design for the MSBR, as given in Refs. 2 and 3, with several important changes: 1. The start=-up fuel is 235y (or %°*U) denatured with *°°U rather than fully enriched uranium. Sufficient 238y is fed along with thorium to keep the fuel in the reactor denatured. 2. The process is altered so that protactinium and plutonium are not isolated from the fuel galt. Protactinium, which decays to 233 U, would otherwise be a source of undenatured fissile uranium. 3. The reactor core is larger with a lower power density to reduce parasitic neutron absorptiong in protactinium as well as in fission prod- ucts. The power density is reduced sufficiently that replacement of the moderator graphite in the core is not required due to fast neutron damage during the lifetime of the reactor. This version of the DMSR is described in greater detail in Ref. 4. The MSR research and development was conducted largely at Oak Ridge National Laboratory, but with assistance by subcontractors and others, in a nearly continuous program for more than 25 years. The effort included many large engineering experiments and the design, construction, and I~ operation of two experimental reactors as well as many small-scale ex- periments in all fields of pertinent nuclear and materials science. For nearly 20 years, that effort was directed to M5Rs for the generation of central station electricity, with the primary focus cn an MSBR breeding 2350 from 2°%Th in the thermal system. The large, varied, and impressive accomplishments of that program, along with the additional development requirements needed for demonstration of the MSBR, were described’ thoroughly as of mid-1972.* That material was updated, and a detailed description (along with a preoposed schedule and costs) of remaining R&D items was presented’ in 1974. The status of molten-salt technology as of late 1974 and the additional needs of the MSBR, accordingly, are documented fairly well, as is the base program of research and tech- nology development required to fill those needs. A large fraction of the technology developed for the MSBR is appli- cable directly, or with a minimum of additional experimentation, to the DMSR. Morecover, most of the additional technology needed for the MSBR is also needed for the DMSR. However, the two reactors differ in some important regards. Accordingly, the technology development required for a DMSR is likely to invelve significant redirection from that anticipated for the MSBR, particularly if the technclogy for on-line processing of the fuel salt is developed in conjunction with the reactor technology. In some areas (e.g., chemistry and chemical processing), this redirec- tion probably would increase the requisite development effort, while in others (e.g., safety technology and graphite development),'the required effort could decrease. An additional consideration is the fact that a small MSR technology development effort was reestablished in mid-1974 and continued for about two yvears. Although this effort was limited in scope, some significant acéomplishments were achieved that also affected the current effort to identify further technology needs. To the extent that it was applicable, the 1974 program plan was used as a basis for this review and projection of the technology needs “The MSR program was closed out in early 1973 and remained in that state for about one year. and program plan for a DMSR. The present document is focused on the major development areas, with the recognition that significant develop- ment efforts could be required in other related areas. (Such ancillary activities would add somewhat to the overall development cost but probably would not appreciably affect the total schedule.) In a number of areas where little or no technical effort has been expended since 1972 and where the perceived needs are substantially the same, the tasks, schedules, and costs were transposed directly from the 1974 plan with only adjustments of the costs to account for inflation be- tween 1974 and 1978. 1In other areas, minor adjustments were made to account for changes in either the technology status® or the apparent development needs. The areas with the greatest potential for change from the 1974 program plan are those related to the chemical processing of the fuel salt. If the once-=through version of the DMSR were developed, it might be possible to defer development of the reductive-extraction—metal- transfer process and thereby reduce the overall development cest for the DMSR. However, because the availability of this process would substantially improve the fuel utilization of a DMSR, the development needs, schedules, and costs for it were included in this plan. Thus, some latitude would exist in the implementation of the program plan. The remainder of this report consists of four major parts, each prepared by a single primary author, which deal with the following major areas of base technology development: (1) reactor design and development, (2) safety and licensing, (3) fuel-coolant behavior and fuel processing, and (4) reactor materials. The parts (which contain up to four chapters) should be regarded as units because of the high level of interdependence among the subjects treated. However, the base program needs and their projected costs and schedules are developed separately within each chapter. KLess than $10 million has been expended on MSR development since 1973, so the changes can have little effect on the overall program cost. 6 References J. R. Engel et al., Conceptual Design Characteristics of a Denatured Molten-Salt Reactor with Once-Through Fueling, ORNL/TM (in prepara- tion. R. C. Robertson, ed., Conceptual Design Study of a Single-Fluid Mol- ten-Salt Breeder Reactor, ORNL-4541 (June 1971). Ebasco Services Inc., 1000 MiW(e) Molten-Salt Breeder Reactor Concep- tual Design Study, Final Report, Task 1, Subcontract No. 3560 with Unicen Carbide Corporation, Nuclear Divisien (February 1972). J. R. Engel et al., Molten-Salt Reactors for Efficient Nuclear Fuel Utilization without Plutowium Separation, ORNL/TM-6413 (August 1978). M. W. Rosenthal et al., The Development Status of Molten-Salt Breeder Reactors, ORNL-4812 {August 1972). L. E. McNeese et al., Program Plan for Development of Molten-Salt Breeder Reactors, ORNL-5018 (December 1974). PART I. REACTOR DESIGN AND DBEVELOPMENT H. ¥. Bauman The objectives of the reactor design program are to first develop a fairly complete conceptual design for a full-scale [1000-MW(e)} DMSR in order to define more clearly the development problems that must be addressed. Then, concurrent with the technology development, a molten- salt test reactor (MSTR) would be designed, built, and operated to demonstrate all aspects of the required DMSR technology on a smaller scale. The scale of the MSTR would be decided as the program progressed. The MSER conceptual design, which the DMSR would probably follow, pro- posed four fuel heat-exchanger and steam-generator modules of 250-MW(e) capacity each. The scale suggested for the MSTR would lie in the range of 100-MW(e) power [with two 50-MW({e) steam-generator modules {i.e., 1/5 scale)] to 250-MW(e) power with a single (i.e., full-scale) steam-— generator module. The data from the component technology develcopment program would be fed into the reactor design efforts, and the experi- ence obtained in reactor construction and operation would, in turn, guide continuing effort in component development. An MSTR mockup is proposed which would permit integrated testing of most of the reactor components before the MSTR itself is built. Finally, the construction of the proto- type DMSR would influence the development of a standardized DMSR design. 1. REACTOR DESIGN, ANALYSIS, AND TECHNOLOGY DEVELOPMENT There is considerable experience in the engineering design and neu- tronic analysis of MSRs, through final design, construction, and opera- tion. The molten—-salt reactor experiment (MSRE) has been through the entire design process, and the ORNL reference concept MSBR' has reached the stage of detailed conceptual design. In addition, the various reactor components and subsystems received intensive development effort in the technology development programs. Status in 1977 At the time the development—status report2 was prepared (1972}, a detailed conceptual design had been developed for the single-fiuid MSBR. Furthermore, many alternative designs had been investigated, generally in lesser detail; one of these, also pertinent to the DMSR, was a low- power-density core design3 in which the core moderator graphite would have an expected lifetime equal to the design life of the reactor. In the area of reactor components and systems, the most important items had been identified as salt pumps, the coolant system as a whole, heat exchangers, the entire steam system, valves, control rods, fuel storage, and gas handling. Ixperience had been obtained in the MSRE in all the above areas ex-— cept the steam system. However, new developments were proposed in sev- eral areas, such as the use of godium fluoroborate as the secondary coolant (rather than lithium~-beryllium fluoride), the use of mechanical valves in addition to freeze wvalves®, and the addition of a fuel-salt gas sparging system (rather than sparging of salt in the pump bowl). Current Development Status The reactor design and analysig effort since 1972 has been minimal. However, a major study of tritium transport in molten-salt systems was ~ A freeze valve is essentially a short, flattened section of pipe which can be cooled to form an internal plug of frozen salt and subse- quently reheated to thaw the plug. 10 carried out in 197476, including experiments in a secondary-coolant—salt test loop. Large quantities of tritium are produced in MSRs from neutron reac- tions with lithium in the fuel salt. Since elemental tritium can diffuse through metal walls, such as heat-exchanger tubes, at elevated tempera- tures, a potential path exists for the transport of tritium to the reac- tor steam via the secondary coclant loop and the steam generator. Recent experiments indicate that tritium is oxidized in the proposed MSR second- ary coolant, sodium fluoroborate, thus blecking transport to the steam System. Tritium—addition experiments were conducted in an engineering-scale ccolant-salt test facility. The results are given in a recent report by Mays, Smith, and Engelf.l+ The experiments showed that the steady-state ratio of combined to elemental tritium in the coolant salt was greater than 4000. A calculation applying this ratio to the case of an operating MSER indicated that the release of tritium to the steam system would be less than 400 CBg/d (10 Ci/day). The conclusion of the study was that the release of tritium from an MSR using sodium fluoroborate in the secondary coolant system could be readily controlled within NRC guide- lines. In the area of component development, there was an effort to advance the steam system development from the conceptual design proposed for the M5BR toward hardware development. In 1971, Foster-Wheeler Corporation was awarded a contract for a study of MSR steam-generator designs. The con- tract was suspended in 1973 and then reinstated in 1974 to complete the first task (in a four-task contract) — the conceptual design of a steam generater to meet specifically the steam and feedwater conditions postu- lated for the ORNL reference-design MSBR. This task was successfully completed, and a report was issued in December 1974.° A design was pre- sented which, based on analysis, would meet all the requirements for an MSR steam generator. Because of the termination of the MSR project, the design did not receive experimental verification or further analytical study. 11 The coclant-salt test facility mentioned in connection with the tritium experiments was operated in the period 1974—76 to obtain engi- neering-scale experience with sodium fluoroborate, the proposed MSBR secondary coolant. The loop was operated successfully for >5000 hr at temperatures up to 540°C. The tests indicated that the engineering characteristics of sodium fluoroborate would be suitable for MSR second- ary coolant. 3 1“SXe3 a gaseous One of the important advantages of MSRs is that fission product with an extremely high thermal-neutron cross section, is not scluble in the fuel salt and can be rapidly removed from the system. Fission-product 135%e is the single most important parasitic absorber of neutrons in thermal reactors, and the high conversion ratio of MSRs depends on efficient 135%e removal. The xenon has twe exits from the fuel salt: by absorption into the pores of the core graphite, where it would remain as a neutron poison, and by diffusion into the fuel cover gas (helium), where it can be removed from the system. A helium-stripping system is proposed for MSRs in which fine bubbles of helium are introduced into the fuel salt to provide a sink for xenocn and are subsequently separated from the salt to remove the xenon effectively. Gas-bubble generators and strippers had been designed and were to be tested in a circulating salt loop when the program was ended in 1973. Although some additional work was done in the 1974—75 period, an integral test of the stripping system was not completed. DMSR Development Needs A1l the design and development needs described for the MSBR in the 1974 program plan6 would alsc be required for the DMSR. However, some aspects of the program should be emphasized for the DMSR. The core design and analysis for both the DMSR and the MSTR require particular attention to the effects of 238U, protactinium, and plutonium on the system fissile inventory and conversion ratio. Another important point is the core fast-flux distribution and its effect on the design life of the moderator graphite. Some prelimipnary neutronic calculations for a proposed DMSR design are given in Ref., 7. Since both thorium and 12 2380 are important neutron resonance absorbers, the lumping of fuel and the degree of thermalization of the neutron spectrum in the reactor core and reflector regions are important variables in the neutronic analysis. An extensive reactor-analysis program would be required to select optimized core designs for the MSTR and DMSR. A very important problem for all molten-salt systems is the control of the reactor power and the temperatures in the fuel salt, the coolant salt, and the steam generators and the management of thermal cycles and thermal stresses in all parts of the reactor system. Although consid- erable experience in MSR thermal problems was obtained in the operation of the MSRE, the addition of a steam system and turbine-generator to the next generation of MSRs leads to considerably more complex interac- tions between the various components of the reactor systems. The analy- sis of thermal-hydraulic dynamic behavior of the proposed reactor systems will require the development of suitable computer models. Some prelimi- nary analysis along these lines had been done for the MSBR, but a major extension of this program would be required for the DMSR. The reactor technology effort encompasses the development of all the major components required in the reactor system. For the most part, highly specialized components such as the vessels and contactors for chemical processing and specific instrument items would be covered within the development programs for the associated technologies. How- ever, some of the more general items {e.g., small salt pumps, valves, seals) might be included in the overall develepment program for reactor components. Such items probably would not greatly affect the total Program cost. A vital part of the development program for reactor components is the design of intermediate- and full-scale salt pumps. The pump preferred for molten-salt service is a vertical-shaft, centrifugal sump pump such as was used successfully in the MSRE and cother test facilities. The drive shaft for this type pump extends through the reactor shielding so that the motor is relatively accessible and is in a low-radiation field. The scaleup of these pumps is proposed to proceed by extending the line of past development with as little change in conceptual design as possi- ble. The nominal capacity of the MSRE pumps was 0.08 m°/s (1200 gpm) ; 13 the pumps proposed for the MSBR, and required for any type of full-scale MSR, would be larger by a factor of about 20. The line of development may include an intermediate-size pump rated at about 0.3 m’/sec (5000 gpm). The design of mechanical vaives for melten-salt service and the use of freeze valves in large systems are alsc very important developments. The MSRE was operated entirely without mechanical valves; freeze valves were used where flow shutoff was required. Freeze valves have been used in pipe sizes up te 1 1/2 in. IPS and have been found to be reliable, zero-leakage devices so long as the integrity of the pipe itself is maintained. The disadvantages of freeze valves are that (1) flow in a line must be stopped before a plug can be frozen, (2) they open and close relatively slowly, and (3) they cannot be used for throttling. Therefore, mechanical- type salt valves are considered essential to the operation of large MSRs. Experience with bellows-sealed mechanical valves in molten-salt experi- mental facilities has been limited. The main development problems are finding materials that will close tightly without binding in molten salt and developing reliable zero-leakage seals. Freeze valves, perhaps in- tegral with mechanical valves, may also be developed for the larger systems., Reactor control rods have been included in MSR designs, although control rods serve a limited purpose in fluid~fueled reactors. The reactivity of the core is controlled mainly by the composition of the fuel and by changes in the density of the fuel with temperature. The ultimate shutdown of the reactor is achieved by draining the fluid fuel to storage tanks. Control rods are required for short-term fine control of the temperature and reactivity of the core, as well as for rapid shutdown; so the design of control rods for melten-salt service is an important development need. The fuel drain and storage system is another vital development. The fuel storage system must have the capability of removing afterheat to the ultimate heat sink in the event that the reactor becomes inoperable or is shut down for maintenance. A drain tank with a natural-convection NaK cooling system was proposed for the MSBR. Although this system ap- pears workable, all aspects of fuel contaimment under normal and accident 14 conditions deserve further attention. Because of the relatively low power density in the fiuid fuel (compared to solid fuel), it is likely that an MSR could he develecped for which a containment melt-through accident would not be considered credible. An MSTR mock-up is proposed which would permit integrated testing of the reactor components under zero-power conditions. The mock-up would permit solving ot layout and remote maintenance problems before the reactor was built. Possible schedules for the first 15 years of the DMSR develiopment program are shown for the reactor design and analysis work in Table 1.1 and for the technology development work in Table 1.2Z. Operating fund requirements for this peried for reactor design and analysis are given in Table 1.3, and operating and capital fund require- ments for technology development are given in Tables 1.4 and 1.5. References 1. R. C. Robertson, ed., Conceptual Design Study of a Single Fluid Molten-Salt Breeder Reactor, ORNL-4541 (June 1971). 2. M. W. Rosenthal et al., The Development Status of Molten=Salt Breeder Reactors, ORNL-4812 (August 1972). 3. Molten-Salt Reactor Program, Semiannual Progress EReport for Period Imding August 31, 1970, ORNL-4622 (January 1971). b, G. T. Mays, A. N. Smith, and J. R. En%el Digtribution and Behavicr of Tritium in the Coolant-Salt Technology Facility, ORNL/TM-5759 (April 1977). 5. Foster-Wheeler Energy Corporation, Task I Final Report, Design Studies of Steam Generators for Molten Salt Feactors, ND/74/66 (December 1974). 6. L. E. McNeese et al., Program Plan for Development of Molten-Salt Breeder Reactors, ORNL-5018 (Deaembhr 19747 . 7. J. R. Engel et al., Molten-Salt Reactors for Efficient Nucleqr Fuel Liaation without Plutonium Separation, ORNL/TM-6413 (August 1978). Table 1.2, Schedule for work on reactor technology development Fiscal year Task 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 § 2 3 Fuel-salt technclogy v v v ‘ ant-salt technclogy v* v° 6 7 a8 ° 10 1 . .m system technoclogy v_ v v v v V1 Cover~ and cff- 12 3 gas system Vv Vv technology 14 5 Salt pump development v . 18 Primary heat—exchanger v development 17 {8 e Valve development v v v VZO Contrel rod development Co . . 21 22 ‘ontainment and cell heating v v development 23 Components Test Facility _....y__.— V24 v?S Milestones: 1. Gas-Systems Technolegy Facility water tests will be fin- 12. Develeopment of methods for handling gaseous effluents (in- ished and construction completed so that salt operation cluding fission products, tritium, and BF;) from the off- can start. gas systems should be complete, 2. Sufficient tests will have been completed to indicate that 13. All other problems associated with the cover- and off-gas the efficiency of the bubble generator—bubble separator systems should be resolved. is satisfactory and that mass-transfer rates are adequate 14. The design of the MSTR prototype pump and pump test stand to permit detailed design of the xencn removal system for should be complete. an MSTK. Additional development will be done to refine 15. The construction of the MSTR prototype pump and pump test the results and test the effects of other variables. stand should be completed and operational tests will be 3. All problems pertaining to the behavior of tritium in the started. fuel-salt system will be resclived. 16. All development work on the primary heat exchanger prepa- 4. Test for determining the behavior of tritium in the coel- ratory to design of the MSTR will be completed. ant system will be completed. Corrosien-product removal 17. Preliminary valve development needed to proceed with design studies will be cempleted. of the MSTR will be finished. 5. Large-scale denonstration tests of coolant-salt technology 18. Final development of specific valves for the MSTR will be should be completed. finished. - 6. The feasibility of using lower feedwater temperatures will 19, Preliminary valve development for prototype DMSR will be be determined. This may affect the subsequent design and completed. development of the steam system components. 20. All development needed for the MSTR control rods will be 7. The plan for a2 steam-generator R&D program should be com- completed. pleted. 21. Exploratory studies and preliminary development needed for 8., The small-scale steam generator work should have progressed the design of the MSTR containment and cell heating should to a stage that will permit reevaluation of the R&D program. be completed. 9, The construction of the steam-generator tube test stand, 22. Testing of the containment and cell heating design for the pressure relief system, and the 3-MW test assembly should MSTR should be completed. be complete. 23. The design of the Components Test Facility should be suf- 10. Testing in the steam—generator tube test stand should be ficiently complete to start comstruction. finished. 24, Construction of the Components Test Facility should be 11. Construction of the steam-generater model test installation, completed. the pressure relief system, and the 30-MW model steam gen— 25. Design of component test facility for prototype DMSK com- erator should be complete and operational tests will be started. ) pleted. Table 1.2. Schedule for work on reacter technology development Fiscal year Task 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 19%C 1991 1992 1993 1994 1 2 3 Fuel-salt technology v v v . 74 5 ant-salt technology v & T 8 9 10 11 .« .m system technology v v v v V'V Cover- and off-gas system ‘{7?2 VB technology 14 15 Salt pump development v Primary heat-exchanger §746 development 17 fa e Valve development v v v §720 Contrel rod development Containment and cell heating Vm V22 development 23 24 25 Compounents Test Facillry v Vv v Milestcones: 1. Gas-Systems Technology Facility water tests will be fin- 12. Development of methods for handling gaseous effluents (in- ished and construction completed so that salt operation cluding fission products, tritium, and BF3) froem the off- can start. gas systems should be complete. 2. Sufficient tests will have been completed to indicate that 13. All other problems associated with the cover- and off-gas the efficiency of the bubble generator—bubble geparator systems should be resolved. is satisfactory and that mass-transfer rates are adequate 14, The design of the MSTR prototype pump and pump test stand to permit detailed design of the xenon removal system for should be complete. an MSTR. Additional development will be done to refine 15. The construction of the MSTR prototype pump and pump test the results and test the effects of other variables. stand should be completed and operational tests will be 3. All problems pertaining to the behavior of tritium in the started. fuel-salt system will be resclved, 16. All development work on the primary heat exchanger prepa- 4, Test for determining the behavior of tritium in the cocl- ratery to design of the MSTR will be completed. ant system will be completed. Corrosion-product removal 17. Preliminary valve development needed to proceed with design studies will be completed. of the MSTR will be finished. 5. Large-scale demonstration tests cof coolant-salt technolegy 18. Final development of specific valves for the MSTR will be should be completed. finished. - 6. The feasibility of using lower feedwater temperatures will 19. Preliminary valve development for prototype DMSR will be be determined. This may affect the subsequent design and completed. development of the steam system compeonents. 20. All development needed for the MSTR control rods will be 7. The plan for a steam-generator R&D program should be com- completed. pleted. 21. Exploratory studies and preliminary development needed for 8. The small-scale steam generator work should have progressed the design of the MSTR containment and cell heating should to a stage that will permit reevaluation of the R&D program. be completed. 9. The construction of the steam-generator tube test stand, 22, Testing of the containment and cell heating design for the pressure relief system, and the 3-MW test assembly should MSTR should be completed. be complete. 23. The design of the Components Test Facility should be suf- 10. Testing in the steam-generator tube test stand should be ficiently complete to start comstruction. finished. 24, Construction of the Components Test Facility should be 11. Construction of the steam-generator model test installation, completed. the pressure relief system, and the 30-MW model steam gen— 25, Design of component test facility for prototype DMSR com- erater should be complete and cperational tests will ke started. ) pleted. Table 1.3. Operating fund requirements for reactor design and analysis Cost (thousands of 1978 dollars) for fiscal year — Task 1980 1981 1982z 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 Design studies of MSR power plants 340 930 620 230 270 520 520 540 520 360 300 300 300 300 300 Design technology 70 210 300 230 190 130 100 100 160 100 100 100 100 100 100 Codes and standards 50 90 170 250 260 170 170 260 260 260 260 260 260 Licensing of MSRs 40 170 250 330 260 260 260 260 260 260 260 260 Nuclear analysis of MSR power 20 130 130 130 130 170 100 100 100 100 100 100 100 100 1066 plants Total fundsa 430 1270 1100 720 930 1320 1170 1170 1150 1080 1020 1020 1020 1020 1020 Allocation MSTR 430 1270 1100 720 330 1120 970 970 850 880 520 520 520 100 100 Prototype DMSR 200 200 230 200 200 500 00 500 920 920 aTotal funds through 1994: $11,440. LT Table 1.4, Operating fund requirements for work on reactor technology development Cost (thousands of 1978 dollars) for fiscal year — Task 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 Fuel~salt technology 24¢ 300 300 210 330 570 260 200 330 260 100 100 100 100 160 Coolant-salt technology 220 380 130 130 260 260 200 200 200 200 200 Steam system technology 70 370 750 780 105G 1420 16206 1600 1600 1600 1,606G 500 500 5G0 500 Cover~ and off-gas system 80 160 80 100 240 140 100 100 160 100 200 200 200 technelogy Salt pump development 50 180 800 2000 1600 1400 400 400 400 400 400 500 Primary heat-exchanger 100 70 130 200 2006 200 200 200 200 development Valve development 50 130 130 130 130 130 260 200 200 200 200 300 Control rod development 80 80 80 200 200 200 200 2006 200 Containment and cell heating 80 130 80 130 100 100 100 100 100 development Components Test Facility 160 106 260 260 580 940 2100 3,000 1000 100C 10GC 1000 MSTR mock-up 820 910 1100 2000 3000 4000 4,000 3000 2000 2000 1000 Total funds® 530 1050 1260 1410 2690 4270 5920 6610 7970 9510 10,100 4000 5100 5100 4300 Allocation MSTR 530 1650 1260 1410 2690 4270 5920 6610 7970 9210 9,500 5000 3000 3000 1500 Prototype DMSR 300 600 1000 2100 2100 2800 aTotal funds through 1994: $71,820. 81 Table 1.5. promise at least a major alleviation of the problem. * : Tellurium is a member of the sulfur family of elements, and sulfur is well known to be detrimental to nickel and nickel-based alloys. TAs described in some detail in Chapter 7 under the heading Status in 1976. Subscripts g and d indicate that the species is gaseocus and dis- sclved in the molten salt, respectively. 34 First, a series of experiments in which tritium was added to the NaF-NaBF, . coolant salt in the Coolant-Salt Technology Facility {(CSTF)} showed that >907% of the tritium added under steady-state conditions appeared in the off-gas system in chemically combined (water-scluble) form and that 987 of the added tritium was removed through the off-gas system.10 These data suggest that the fluoroborate coolant system of an MSBR (or DMSR) might well diminish the leakage of tritium to the reactor steam system to acceptable limits. Continued study11 shows that the oxide film formed by the reaction of steam with steam—generator materials can greatly im- pede the permeation of the metal by tritium. Even at a steam pressure of 1 atm, where the oxidation rate is still clearly dependent on rates of diffusion from the bulk alloy through the oxide coating, tritium per- meation is impeded by factors of nearly 500 after 150 days of exposure.® Status of Fuel and Coolant Chemistry Fuel chemistry Choice of components and composition. For an MSBR, where excellent neutron economy is an abseclute requirement, acceptable fuel components are few. The careful considerations and the detailed experimentation over a pericd of many years that led to the choice of fuel constituents ? There is no doubt and composition have been completely described. '’ that (1) the major constituents of the fuel salt for an MSBR must be LiF, BeF2, ThFy, and UF4, with a composition of ~71.7, 16, 12, and C.3, respectively, and that (2) highly enriched "LiF and 2°°UF, are required. For a DMSR the requirement for excellent neutron economy might seem to be capable of slight relaxation. However, such a reactor must have a reasonably high concentration of thorium and must contain more uranium than the MSBR. There can be no reasonable doubt that the anion must be F-, and the possibility that one can find better diluent fluorides than + "LiF and BeF; is extremely unlikely. The optimum compesition of the This study was initiated under the Molten-Salt Reactor Program and has been continued, because of its obvious interest, by the fusion energy program. T . A slightly higher °Li concentration could possibly be tolerated, but the rate of tritium generation would be increased. 35 initial fuel loading for a DMSR is, of course, not yet known precisely. However, it appears likely that the optimum mixture will fall near (in mole %) 72 LiF, 16 BeFy, and 12 heavy-metal fluorides, with slightly less than 10.5 mole % ThFy and slightly more than 1.5 mole % UF, and UF;. As a consequence, much that has been learned about the MSBR fuel compo- sition is directly applicable to the DMSR. Knowledge of the behavior of MSBR fuel, although not complete, can fairly be said to be extensive and detailed.!??%»1? As an additional consequence, the DMSR (like the MSBR) will require a large-scale and reasonably economic source of lithium enriched to near 99,99% ’Li. No such enrichment facility is operating in the United States today, but the technology is well known and relatively large-scale separa- tion has been practiced in the past. Fluoride phase behavior. Phase equilibria among the pertinent MSER and DMSR fluorides have been studied in detail, and the equilibrium dia- grams, although relatively complex, are well understood. Because these reactors need a ThIFFy concentration much higher than that cof UF4, the phase behavior of the fuel is dictated by that of the LiF-BeF,-ThFy 1,12 system™ * tectic at 47 mole % LiF and 1.5 mole ¥ ThF., melting at 360°C.! The shown in Fig. 3.1. This system shows a single ternary eu- system is complicated by the fact that the compound 3LiF+ThF, can incor- porate Re?+ ions in both interstitial and substitutional sites to form solid solutions whose compositional extremes are represented by the shaded triangular region near that compound. Inspection of the phase diagraml’12 reveals that a considerable range of compositions with more than 10 mole 7 ThFy will be completely molten at or below 500°C. The maximum ThFs concentration available at this liquidus temperature is just above 14 mcle 7. As expected from the general similarity of ThFy and UFy, substitution of a moderate quantity of UFy for ThFy scarcely changes the phase behavior. Operation of the DMSR will result in production of plutenium and of smaller quantities of other transuranium isotopes. It seems likely that at equilibirum the concentration of plutonium will be near 0.05 mole %, while Np, Am, Cm, Cf, and Bk might together total an additional 0.025 QORNL-LR-DWG 37420AR7 g Thr’-; 1544 / 1400 TEMPERATURE iN °C COMPOSITION IN mole % LiF - The, - - LlF'ZTh% 3L|F-ThF4 85 -~ P 8G7 J{_ 950 - / £ 568/ 3L|F~Th§ £ 565 71N f '3- \gi;-.’ i L JO\/ \\\\ 550 S— £526 o i e e, F gag 2LIF BeF; 500%s0 200t 400 450 500 555 £ 458 £ 360 Fig. 3.1. The system LiF-BelF,-ThF,. mole %. All these species are expected to be dissolved in the fuel solu- tion as trifluorides. The golubility of PuF; in LiF-BeF;-ThF, (72-16-12 mcle %} has been measured at CRNL'? and at the Bhabba Atomic Research Center in India.'® The latter study indicated that solubility increased from 0.77 mocle % at 523°C to 2.79 mole % at 718°C. 'The ORNL measurements yielded values about 207 higher. In beth studies, more than one method was used for assay of the dissolved plutonium, and no ready explanation of the dis- crepancy is available. It is clear, however, that even the lower value far exceeds that required. The other transuranic species are known to : dissolve!® in the LiF-BeF,-ThF, solvent, but no quantitative definition of their sclubility behavior exists. Such a definition must of course be obtained, but the generally close similarity in the behavior of the 37 trivalent actinides makes it most unlikely that solubility of these individual species can be a problem. The solubility of UF3; in the fuel was known to be well in excess of that required for the MSBR, but its absolute magnitude is not well kncwn. ® with behavior in LisBeF,, the solubility of UFj3 is very By analogy1 likely to be lower than that of PuF3, but it is quite unlikely to be less than 0.4 mole 7 at 565°C. The trivalent lanthanides and actinides are known to form solid solutions so that, in effect, all the rare-earth trifluorides and the actinide trifluorides act essentially as a single element. Should it prove desirable tec operate with 107 of the uranium reduced (ca. 0.16 mole % UF3), it is possible, but highly unlikely, that the combination of all trifluorides (perhaps 0.3 mole %) might exceed this combined sclubility at a temperature somewhat below the reactor inlet temperature. A few experiments® must be performed to check this slight possibility. Oxide behaviecr. The behavior of molten fluoride systems such as this is markedly affected by appreciable concentrations of oxide ien. The solubilities of the actinide dioxides in LiF-BeF:;-ThF., are low and are known':'%7%% to decrease in the order ThO,, Pa0jy, U0,, Pul,. Solubility products and their temperature dependence have been measured 1516723 +hat these dioxides all for these oxides. Moreover, it is known have the same fluorite structure and form solid soluticons; theiy behavioer is reasonably well understood.’ Plutonium as PuF; shows little tendency to precipitate as oxide.'?” The compound Pa0s (or an addition compound of this material) is very insoluble in LiF-BeF;-ThFy (72-16-12 mole 7%). The oxide concentra- tions at which Pa;0s5 can be precipitated depend on both the protactinium concentration and the oxidation state of the fuel. The situation is indicated by the equilibrium+ 1 Pas05 + U3t = gttt 4 2 02 4 patt 2 () 2 * % Since mixtures with plutonium and beryllium are neutron scurces, the experiments are more difficult than usual. The subscript ¢ identifies the crystalline or solid state. 38 for which we estimate® the equilibrium quotient log [(X5220X, ateXud) /X 54] = 0.76 — 8590/T (+0.8) . 5 G The result is that with 100 ppm protactinium and 30 ppm oxide present, the U’1/U"" ratio must be kept above about 107™° if inadvertent precipita- tion of Pa205 is to be avoided. Such oxidizing conditions are easy to avoid in practice. There is also a dependence on the udt/utt ratio of the oxide concentration at which Pu0; precipitation cccurs. However, even stronger oxidizing conditions (U3+/Uq+ < 10™%) are required to pre- cipitate PuOp from fuel for MSBRs or DMSRs. The solubility of the oxides of neptunium, americium, or curium has not been examined. Some attenticn to this problem will be required, but it is not obvious that such studies have a high priority. It is clear that the DMSR fuel must be protected from oxide contami- nation to avoid inadvertent precipitation. Because of the low oxide tolerance, this will require some care, but the successful operation of the MSRE over a 3-year period lends cenfidence that oxide contaminaticn of the fuel system can be kept to adequately low levels. This confidence, when added to the prospect that the DMSR fuel will be reprocessed {(and its oxide level reduced by fluorinaticn of the uranium) on a continuous basis, suggests very strongly that problems with oxide contamination can be avoided. Physical properties. Most of the physical properties of LiF-BeFs;- ThFy, (72-16-12 mole %);'C are known with reasonable accuracy, although several have been defined by interpolation from measurements on slightly different compositions. The liquidus temperature is well known, and density and viscosity 2,24 are accurate to 13 and +107, resgpectively. The heat capacity has been derived from drop calorimetry;25 on the basis of this determination and with a simple model for predicting heat capacity of molten fluorides, one can reliably predict the heat capacity of the DMSR mixture. % o . Most physical properties will be trivially affected by variations among the heavv-metal cencentrations so long as the heavy-metal content remains at 12 mole %. 39 Thermal conductivity is the key property for predicting heat-transfer coefficients of molten fluorides. Measurements that are probably accurate to *10 to 15% have been obtained for LiF-BeF,-ThF,-UF, (67.5-20-12-0.5 mole %).%*® TFor future design considerations it will be helpful to de- velop an apparatus to measure thermal conductivities of fluorides with greater accuracy and to determine the conductivity of the fuel salt com- position. The surface physical properties (surface tension and interfacial tension between salt and graphite) are only qualitatively known. Such properties are important in assessing wetting behavior and in determining the degree of salt penetration into graphite. The wvapor pressure, as yet unmeasured, has been extrapolated from measurements of LiF-BeF,; and LiF-UFy mixtures. At the highest normail operating temperature, 704°C, the estimated vapor pressure is V1.3 Pa (107% torr). The vapor composition has not been measured, but the vapor would be considerably enriched in BeF; and perhaps in ThF,. Vapor pres- sure and vapor composition are not high-priority measurements. However, more than qualitative estimates of these properties will be required in future calculations of the amount and composition of salt that is trans- ported by gas streams used to cool portions of the off-gas system in the primary circuit. A transpiration experiment would provide firm values of vapor composition and improved values of vapor pressure. Manometric measurements combined with mass-spectrographic determination would pro- vide more precise information on both. Fission-product chemistry. Much attention was given to behavior of the fission products in the MSRE?’ because of their effect on reactor operation and performance, afterheat, and reactor maintenance. More experimentation will clearly be required in future DMSR (or MSBR) develop- ment. The noble gases are only slightly scluble in molten fluorides®?™ 3! and can be removed by sparging with helium. More than 80% of the *35%%e was removed by the relatively simple sparging system of the MSRE.® The more efficient sparging proposed for the MSBR should also be applicable to the DMSR. Most of the '*°Xe (the worst of the fission-product poisons) 40 g 135 T is formed indirectly by decay of 6.7-hr , and the use of rapid side- stream stripping of 1357 by the reaction HF +~ 7, .. > F7 + HI (g) (d) (d) (g) ’ was considered a remote possibiiity for the MSBR.? Such stripping seems a most unlikely need for the DMSR; if it were necessary, it woeuld preclude operation at high UF3;/UF, ratios. The rare earths and other stable soluble fluorides (e.g., Zr, Ce, Sr, Cs, Y, Ba, and Rb) are all expected to be found principally in the fuel salt* and can be removed by the fuel processing operation.T The chemical behavior of these fission products is fairly well understood and, like the noble-gas behavior, can be predicted confidently for operating DMSRs. *7 The chemical behavior of the so-called noble-metal fissien products (Nb, Mo, Te, Ru, Ag, Sb, and Te)t is considerably less predictable — as 27 — and warrants further study. has been berne out in MSRE operations According to available thermodynamic data, they are expected to appear in a reduced form at UF3/UF, ratios greater than 1077, However, in the : reduced and presumed metallic state, these fission products can disperse via many mechanisms. Analyses of MSRE salt samples for five noble-metal nuclides (°7Mo, 1DgRu, IOERug and '?°713%Te) showed that the fuel salt contained up to a few tens of percent of the nominal calculated inventory. All these species have volatile high-valence fluorides that could form under suf- ficiently oxidizing conditions. On the basis of thermodynamic comnsid- erations and a correlation of their behavior with that of lllAg, for which no stable fluoride exists under fuel-salt conditions, it has been tentatively concluded that they are metallic species that occur as finely divided particles suspended in the salt. P Some of these have noble~gas precursors; a fraction of these will escape from the fuel and appear in the off-gas system. 'See Chapter 6. And several other species of lower yield. 41 The noble-metal fission products were also found deposited on graphite and Hastelloy N specimens (on surveillance specimens as well as on post- operation specimens). However, their distribution on both sets of speci- mens varied widely and allowed only very tenuocus conclusions to be drawn. It was evident from these studies that net deposition was generally more intense on metal than on graphite, and deposition on the metal was more intense under turbulent flow. 235y operation of the Gas samples taken from the pump bowl during MSRE indicated that substantial concentrations of noble metals were pres- ent in the gas phase, but improved sampling techniques (used during h 233UF4) showed that previous samples had operation of the reactor wit been contaminated by salt mist and that only a small fraction of the noble-metal fission products escape to the cover gas. 0Of the noble metals, nicbium is the most susceptible to oxidation: it was found appreciably in salt samples at the start of the 233y MSRE operation because of the low initial U3+/U“+ ratio. It could be made to disappear by lowering the redox potential of the fuel,1 but it sub- sequently reappeared in the salt several times for reasons that were not always explainable. The %°Nb data did not correlate with the Mo-Ru-Te data mentioned previously, nor was there any observable correlation of . g . . . 1,27 niobium behavior with amounts found in gas samples. The actual state of these noble-metal fission products is important to the effectiveness of MSBR operations. 1If the products exist as metals and if they plate out on the Hastelloy-N portions of the reactor, they will be of little consequence as poisons; however, they can be of im- portance in determining the level of fission-product afterheat after reactor shutdown and will complicate maintenance operations and post-— operation decontamination. They will contribute to neutron poisoning if they form carbides or adhere in some other way to the graphite modera- tor;* however, examination of the MSRE graphite moderator indicated that the extent of such adherence was limited.!s?s?7 Niobium is the only element of this series with a carbide that is thermodynamically stable in this temperature range. It showed the largest tendency to associate with the moderator graphite. Operation with a UF3/UFy, ratio near 0.1 will apparently change the behavior of those noble metals capable of reduction to an anionic state. 57 and may be safely presumed This seems certain to include tellurium to include selenium. Antimony may exist (and may be dissolved) as Sb’~ in such melts, and other of the noble-metal fission products may be dis- solved. If so, they, along with the tellurium and selenium, would — as was not expected for the MSBR — be transported to the fuel processing circuit, where their removal should be possible. Should a decision be made to operate with strongly reduced fuel, some study of such possi- bilities will be necessary. Clearly, most of the future fiésionwproduct chemical research should be directed toward increasing cur understanding of noble-metal—fission- product behavior to a level comparable to that of the other fission prod- ucts., Factors of importance to future reactors include the redox poten- tial of the system,* the possible agglomeration of metals onto gas and bubble interfaces in the absence of colloidal (metallic, graphite, oxide, etc.) particles, the deposition of noble metals onto colloidal particles, and the deposition and resuspension of particles bearing noble metals. Tritium behavior. A 1000-MW(e) MSBR has been estimated®® to produce about 2420 Ci (v0.25 g) of °H per day; the DMSR must accordingly be ex- pected to generate H at this rate. Since metals at high temperatures are permeable to the isctopes of hydrogen, the pathways for tritium flow from the reactor to the environment are numerous. Although many of the pathways do not present serious difficulty,+ the flow to the steam genera- tor, if not inhibited, could result in tritium contamination of the steam system and release of tritium to the environment via blowdown and leakage to the condenser ceolant. As noted above (and described in more detail in a subsequent sec- tion), the NaF-NaBF, coclant appears to be the major defense against *H escape to the steam system;’’ however, the behavior of °H in the fuel ale See further discussion under Fuel-Graphite Interactions. T . . . . Since the tritium can readily be trapped and retained for disposal as waste. el 32 is generated system is obvicusly important. Most of the tritium'?> by neutron reactions on 574 and ‘Li in the fuel. Such °H is, in princi- ple, generated in an oxidized state (as 3HF). However, upon equilibra- tion with a fuel containing a UF3/UFy ratio of 0.01 by the reaction HF + UF g-é-H + UF (@ 2 " (g) (a) the HF should be almost completely reduced to Hz; reduction would of course be even more complete at UF3/UFy = 0.1. The data for this equi- . \ \ 8 librium reaction are well known. s 9 ?% to determine the solubility of Hz and Dj Attempts have been made (and, by analogy, °H») in molten LirBeFs. Plausible (and very small) solubilities were measured, but the solubility is not precisely known. Further study of the solubility relationships is needed, and measurements of diffusivity in the fuel would alsoc be valuable. It seems likely that an efficient sparging system (as for '*°Xe removal)} will strip a consid- erable fraction of the °Hz to the reactor off-gas system, where it could be coliected for disposal. Basic studies of molten fluorides. A comprehensive knowledge of the formation free energies (AGf) of solutes in molten Li»;BeF., has been gained over the year533 from measurements of heterogeneous equilibria involving various gases (e.g., BF or H»0) and solids (e.g., metals or oxides). The list of dissolved components for which formation free energies have been estimated includes LiF, Be¥,, Th¥,, several rare- earth trifluorides, Zr¥,, UFs3, UFy4, PaFy, PaFs, PuF;, CrF,, FeF,, NiFy, NbFy, NbFs, MoF3, HF, BeO, BeS, Be(OH),, and Bel,. Some of these AGf values, however, are presently insufficiently accurate for the needs of the MSBR and DMSR programs (e.g., those for PaFs, PaFs, ThF,, MoFj, NbFy, and NbFs) and additional equilibrium measurements involving these solutes are needed. Moreover, there is a need for the AGf values of certain other fission-product compounds such as the lower fluorides of technetium and ruthenium and various dissclved compounds of tellurium. A more urgent need is an increased knowledge of how activity coefficients (which have been defined as unity in Li»BeFy) vary as the melt composition changes. Such knowledge is required to predict how the numerous chemical equilibrium bty constants that may be derived from AGf values in LioBeFy will change as the melt composition is changed to that of a DMSR fuel. Coolant chemistry It has never appeared feasible to raise steam directly from the fuel (primary) heat exchanger; accordingly, a secondary ccolant must be provided to link the fuel circuit to the steam generator. The demands imposed upon this coolant fluid differ in obvicus ways from those imposed upon the fuel system. Radiation intensities will be markedly less in the coolant system, and the consequences of uranium fission will be absent. However, the coolant salt must be compatible with the construction metals that are also compatible with the fuel and the steam; it must not undergo violent reactions with fuel or steam should leaks develop in either cir- cuit. The coolant should be inexpensive, it must possess good heat trans- fer properties, and it must melt at temperatures suitable for steam cycle start-up. An ideal coclant would consist of compounds that are tolerable in the fuel or are easy to separate from the valuable fuel mixture should the fluids mix as a consequence of a leak. Choice of coolant compesition. Many types of coolant materials were carefully considered before the choice was made. The coclant which served admirably in the MSRE, "Li,BeF,, was rejected for the MSBR be- cause of economics and because its liquidus temperature is higher than desirable. No substitute with ideal characteristics was found. After consideration of molten metals and molten chloride and molten flucoride mixtures, the best material overall appeared to be a mixture of sodium fluoride and sodium fluoroborate.’ These compounds are readily avail- able and inexpensive and appear to be sufficiently stable in the radia- tion field within the primary heat exchanger. The mixture of NaF-NaBFy with 8 mole % NaF melts at the acceptably low temperature of 385°C (725°F), and its physical properties seem adequate for its service as a heat trans- fer agent. These compounds are not ideally compatible with either steam or the MSBR fuel, but the reacticns are neither violent nor even particu- larly energetic. The fact that fluoroborates show an appreciable equilibrium pressure of gaseous BF3 at elevated temperatures presents minor difficulties. The BF3 pressures are moderate; they may be calculated from log P = 11.149 — 5920/T , when pressure is in pascals and temperature is in kelvin [yielding 23 kPa (175 torr) at 600°C], and clearly present no dangerous situations. However, it is necessary to maintain the appropriate partial pressures cf BF3 in any flowing cover—-gas stream to avoid composition changes in the melt. The appropriateness of that choice for the MSBR (and for the DMSR) has been confirmed by several findings in recent years. First, a care- ful and detailed reconsideration of secondary {(and even secondary plus tertiary) coolants®” ranked the NaF-NaBF, coolant very high on the list of alternatives. After these deliberations,’® experimental information'® became available to show that the NaF-NaBF; mixture was genuinely effec- tive in trapping H. The reconsideration, accordingly, concluded: " While the information that is currently available is in- adequate for accurate extrapolation to the rate of tritium re- lease teo the steam system of an MSBR, it appears that the sodium fluorcborate salt mixture would have a substantial in- hibiting effect on such release and that environmentally ac- ceptable rates (<10 Ci/d) could be achieved with reasonable effort. Additional study needed.®* A considerable study of many aspects of fiuvoroborate chemistry has been conducted during the past few years. Neverthelesg, our understanding of the chemistry of the NaF-NaBF., system is less complete, and our knowledge of its behavior rests on a less se- cure foundation than that of the MSBR fuel system. Thus there are sev- eral areas where further or additional work is needed, although it seems unlikely that the findings will threaten the feasibility of NaF-NaBFy in the MSBR (or DMSR) concept. "It should be obvious that the additional study of NaF-NaBF, cool- ants needed for an MSBR is essentially identical to that needed for a DMSR. Accordingly, the previous documentation,'’? except as modified by more recent findings discussed later, adequately describes the neaded R&D pregram. 46 Phase behavior of the simple NaF-NaBF, system and the equilibrium wass’ pressure of BF3 over the pertinent temperature interval are well under- stood. If the NaF-NaBF, eutectic (or some near variant of it) is the final coolant choice, little effort need be spent in these areas. Additional information is needed, however, on the behavior of oxide and hydroxide ions in the fluoroborate melts. For example, the solubility of NasBoFg0 in the mixture is not well known; data on equilibria (in inert containers) among H,0, HF, NaBF30H, and Na:B»Fe0 are still needed; and rates of reaction of dilute NaBF30H solutions with metals need defi- nition. Investigation of NaBFy melts by x-ray powder diffraction, infra- red spectroscopy, and Raman spectroscopy have identified the stable ring compound Na3zB3Fg03; as the probable oxygen-containing species in ccolant 2 meits. " Measurements of condensates trapped from the coolant salt tech- nology facility (CSTF), a development loop, show a tritium concentration of 10° relative teo the salt, suggesting that a volatile species may be selectively transporting tritium from the loop through the vapor.lO Re- cent results indicate that BF3°*2H,0 may exist as a molecular compound in the vapor and could be regsponsible for the tritium trapping,3£+ However, the mechanism by which tritium diffusing from the fuel system can be trapped needs additional study. As indicated above, several of the physical property values have been estimated. These estimates are almost certainly adequate for the present, but the program needs to provide for measurement of these quan- tities. Compatibility of the NaF-NaBF, with Hastelloy N under normal opera- ting conditions seems assured. Additional study, in realistic flowing systems, of the corrosive effects of steam inleakage is necessary. This study, closely allied with the study of equilibria and the kinetics of reactions involving the hydroxides and oxides described above, would re- quire the long-term operation of a demonstration loop that could simulate steam inleakage and coolant repurification. Purification procedures for the coolant mixture are adequate for ’ the present and can be used to provide material for the many necessary tests. However, they are not adequate for ultimate on-line processing of the coolant mixture during operation. TFluorination of the coolant 47 on a reasonable cycle time would almost certainly suffice but it has not been demonstrated. A process using a less aggressive reagent is clearly desirable.’ A fluoroborate mixture has shown completely adequate radiation sta- bility in a single (but realistically severe) test run. Additional radiation testing cf this material in a flowing system would seem de- sirable and should ultimately be done but does not rate a high priority 1,2 at present. ? Fuel-coolant interactions A rupture of a tube (or tubes) in the primary heat exchanger would unavoidably lead to mixing of some coeolant salt with the fuel. The pos- sibility of a nuclear incident would seem highly unlikely because of the consequent addition of the efficient nuclear poison boron to the fuel. However, since BFj3; is volatile, mixing might result in a pressure surge, and the NaF-NaBF, mixture contains some oxygenated species. The 1972 review, accordingly, concluded:?® Mixing of coolant and fuel clearly requires additional study. The situation which results from equilibration of these fluids 1is reascnably well understood, and, even where large leakages of coolant into the fuel are assumed, the ultimate "equilibrium' seems to pose no real danger. However, the real situation may well not approximate an equilibrium condition. Studies of such mixing under realistic conditions in flowing systems are lacking and necessary. The 1974 program plan2 included a very considerable program for such study. More recent experiments’®" have thrown additional light upen such mixing. Although additional experiments are needed, it now appears likely that such mixing would not pose drastic problems. These experi- ments revealed that BF; gas was slowly evolved when the salts were mixed; some 30 min were required to complete the BF3 evolution. Furthermore, the ThF4 and UFy4 showed no tendency to redistribute, to form more con- centrated solutions, or to precipitate. Moreover, no UG, precipitated even when the molten fuel-coolant combination was agitated for several hours while exposed to air. 48 Additional confirmatory experiments will be required and the dif- ferent species in DMSR fuel must be tested. Such experiments may re- ceive less attention and lower priority than previously helieved.!s? Fuel-graphite interactions Graphite does not react with, and is not wetted by, molten fluoride mixtures of the type to be used in the MSBR. Available thermodynamic 35 data suggest that the most likely reaction, 4UFy + C @ T Cey T CFe + 4UF 4 (g) (d) ~’ should come to equilibrium at CF, pressures below 107% atm. At least one source°® lists chromium carbide (Crs3Cs) as stable at MSBR tempera- tures. If Crs:Cy is stable, it should be possible to transfer chromium from the bulk alley to the graphite. No evidence of such behavior has been observed with Hastelloy N in the MSRE or other experimental assem- blies. Although such migration may be possible with alloys of higher chromium content, it should not prove greatly deleterious, since its rate would be controlled by the rate at which chromium could diffuse . to the alloy surface and should be limited by a film of Cri3C; formed on the graphite. This consideration, taken with the wealth of favorable experience, suggests that no problems are likely from this source in the reference MSBR or in a DMSR. However, some additional examination of this unlikely problem area must be done for the DMSR, particularly if operation at UF3;/UFy ratios near 0.1 is to be attempted. The upper limit on that ratio will most likely be set by the equilibrium AUFg(d) + 2C - 3UF, + UCy (c) (d) (c) at the lower end of the operating temperature range.37 Toth and Gilpatrick,38 who used a spectrometric technique in which molten salts were contained in cells of graphite with diamond windows, made a careful study of equilibria among UFs: and UF4 in molten solution with solid graphite and uranium carbides. Their data show that the ratio 49 of UF3 to UFy in LiF-BeF;-ThFy (72-16-12 meole %) in equilibrium with graphite and UC, at 565°C lies in the range 0.11 to 0.16.% All equilibria .f. studied were found to be very sensitive to temperature' and to the free fluoride concentration of the solvent. It would seem likely that UF;/UF, ratios as high as 0.1 can be tolerated for a DMSR (though slight adjust- ments in fuel composition or fuel-inlet temperature might be required), but confirmatory experiments are needed. Since similar systems appear to be sensitive to oxide ion concentration, some experimental study of this parameter will also be required. Even at relatively high temperatures, graphite has been shown to 9 adsorb H» and its isotopes to an appreciable extent. - Further informa- tion about this phenomenon should be cbtained. Prime R&D Needs Weaknesses in the existing technological base and requirements for additional technical information have been identified throughout the preceding discussion of the technology status. These needs are consoli- dated and presented below, in cutline form, to provide a concise tabula- tion for defining and scheduling possible R&D activities. Fuel chemistry 1. Demonstrate the feasibility of operation at UF3/UF, ratio of 0.07 to .1. a. Verify the individual and collective solubilities of trivalent actinide fluorides. b. Verify the interaction of uranium with graphite. c. Verify the behavior of noble and seminoble fission products (i.e., Se, Sb, Te, and Ru) along with tellurium. 2. Define the limits on tolerable oxide concentration to avoid precipi- tation of oxides from DMSR fuel at UF3/UF,4 ratio of 0.1 and to avoid interaction of that fuel with graphite. % . - . s . s . The UC2 so formed may be stabilized by inclusion of some oxide ion in the lattice. +At 600°C the UF3/UFy ratio lies in the range 0.23 to 0.32. 50 3. Provide sound measurements of those physical properties (surface ten- g sion, interfacial tension, thermal conductivity, vapor pressure, and vapor composition) that are not known with precision. 4. Improve the knowledge of fission~product behavior, particularly of key noble metals, in reduced DMSR fuel. 5. Determine the solubility and diffusion kinetics of H» and its iso- topes in DMSR fuel. 6. Perform the basic studies to bring knowledge of solute behavior in LiF-BeF,-ThF, (70-16-12 mole %) to at least the level of current knowledge of behavior in LisBeFy. Coolant chemistry 1. Identify and characterize oxygenated and protonated species in NaF-NaBF, as functions of "contamination"™ level. 2., IBlucidate the mechanisms by which the coclant salt takes up B and determine identity of the products and tfie key reaction rates. 3. Determine {(in cocperation with the materials development effort de- scribed in Part IV of this document) the effect of the "contaminants" mentioned above, and the effect of steam inleakage, on corrosivity of coolant salt. 4. Refine the measurements of physical properties of NaF-NaBF, as re- quired. 5. Confirm the adequacy of radiation resistance of the realistic mixture (i.e., with the desired "contaminant’ level). IF'uel—-coolant interactions 1. Continue, and scale up, mixing studies to demonstrate that no hazard- ous interactions exist and define the limits of behavior. 2. Consider the problems of fuel (and graphite) cleanup consequent to a fuel-coolant leak. Fuel-graphite interactions 1. The major need identified above is to demonstrate tolerable UF3/UF, ratios and 0%~ concentration limits to avoid formation of uranium carbides. 2. Define the extent to which °H will be adsorbed by moderator graphite. 3. Verify the interaction of noble-metal fission-products with graphite at usable UF3/UFy ratios. Estimates of Scheduling and Costs Preliminary estimates of the necessary schedule and of its operating and capital funding requirements are presented below for the fuel and coolant chemistry program described above. As elsewhere in this document, it has been assumed that (1) the program would begin at the start of FY 1980, (2) it would lead to an operating DMSR in 1995, and (3) the R&D program will produce no great surprises and no major changes in program direction will be required. The schedule, along with the dates on which key developments must be finished and major decisions;made, is shown in Table 3.1. It seems virtually certain that the R&D programs (including those described else- where in this document) will provide some minor surprises and that some changes in the chemistry program will be required. No specific provi- sions for this are included; but, unless major revisions become necessary in the middle eighties, it appears likely that suitable fuel and cooclant compositions could be confidently recommended on this schedule. The operating funds (Table 3.2) and the capital equipment require- ments (Table 3.3) are shown on a year-by-year basis in thousands of 1978 dollars. No allowance for contingencies, for major program changes, and for inflation during the interval have been provided. Table 3.1. Schedule for chemical research and development Fiscal year Task 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 Fuel chemistry i 2 3 4 5 Phase equilibria v v v v v 6 7 8 g Oxide behavior v v v v fo it Physical properties v v i2 13 Tritium behavior v v 4 15 16 17 18 ig 20 Fission-product chemistry vv v v Vv v v 24 22 23 Basic studies v v v C t i cclant chemistry V24 v25 (726 \727 vaa Basic studies st Physical properties 30 3 32 Tritium chemistry v v v 33 34 35 Fuel-coolant interactions v v v 36 37 5 38 Fuel-graphite interacticns v v v v Milestones: 1. Determine solubility of UF; over reasonable fuel compositiom 20. Make final evaluation of fission-product behavior for DMSR, range. 21. Define activity coefficients for Te?t fand other Te species) Z. Determine solubility of AmF3;, NpFi:, and CmFj;. in reduced fuel. 3. Define solubility limit of total metal trifluvorides over rea- 22. Complete evaluation of porous electrode studies. sonable range of compositions. 23. Complete definition of activity coefficients for sclutes in 4, Conclude phase equilibrium investigations, including effect fuel. of small concentrations of C17. 24. Complete evaluation of boride formation on Hastelley N in 5. Make final decision as to feasibility of operation at UF3/UFy coolant salt. ratio near 0.1. 25. Finish investigation of oxide species in coolants. 6. Redetermine solubility of Paz0s. 26. Finish measurement of free energy and activity coeff1c1ents 7. Establish selubilities of Am;03, Np20i3, and Cm»0;. of corrosion products in coolant salt. 8. Determine scliid solution behavior as a function of oxide con- 27. Make final decision as to coolant composition. tamination level. 28. Finish measurements of effect of steam inlezkage into coolant. %, Set limits on tolerable oxide ion concentration in fuel and 29. Finish physical property measurements on ceoolant. assess possible separations procedures based on oxide pre- 30. Identify mechanisms for trapping of tritium in fluworcoborate. cipitation. 31. Complete evaluation of reaction rates of tritium with fluoro~ ! 1¢. Determine surface physical properties of realistic composition borate species. * range and assess wettability of metal and graphite. 32, Complete evaluation of tritium removal from coolant and re- 11. Complete physical property determinations. conditioning of coolant. 12, Determine solubility of H; and HT in fuel. 33. Complete dynamic studies of fuel-coolant mixing. 13. Determine diffusivity of H, and HT in fuel. 34. Determine precipitation behavior of fuel with realistic oxide 14. Determine possibility of removal of ZrF, from reduced fuel as and protonic contaminants. intermetallic compound. 35. Complete evaluation of methods for recovery from fuel-coclant 15, Establish sclubility of Te, Tezm, and other Te species in fuel. mixing. 16. Determine oxidation states of noble and semiroble fission- 36. Define limits om UF3/UF, ratio and oxide contamination level product metals as a function of pertinent UF;/UF, ratios. in fuyel to avoid uranium carbide formation. 17. Make final conclusions as to Se, Te, I~, and Br~ behavior at 37. Complete evaluation of fission-product—graphite interact1on . pertinent UF;/UF, ratios. with maximally reduced fuel. 18. Establish feasibility of removal of noble metals by washing 38. Complete investigation of tritium uptzke by moderator graphite with Bi (with no reductant). and activated carbons. 19, Establisgh extent of serption of I;, SeFg, and TeFg in oxidized fuel {uranium valence 4.5). o, Table 3.2. Task Fuel chemistry Operating fund requirements for chemical research and development Cost {thousands of 1978 dollars) for fiscal year — 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1593 1994 Phase equilibria 100 1G0 100 160 1060 100 75 >0 25 0 0 0 0 0 0 0 Oxide behavior 150 150 150 150 175 150 100 50 50 0 0 0 0 0 o O Physical properties g 0 60 60 &0 60 120 120 120 60 00 40 40 0 0 0 Tritium behavior 0 120 120 150 150 100 100 50 50 o 0 0 0 0 0 0 Fissicn-product chemistry 80 120 120 120 120 120 200 250 250 250 150 150 150 100 50 0 Basic studies 120 120 150 150 150 150 100 100 75 0 0 0 ¢ 0 0 0 Coolant chemistry Basic studies 50 100 100 125 125 150 150 150 150 150 100 100 75 50 0 G Physical properties 0 0 G 50 50 50 75 75 75 75 75 75 50 50 0 0 Tritium chemistry 70 125 150 150 215 230 260 330 330 300 150 100 100 50 0 0 Fuel-coolant interactions 75 75 75 75 100 100 106G 100 50 25 0 Q 0 0 0 0 Fuel~-graphite interactions 50 80 100 100 100 130 150 200 125 75 25 0 o 0 0 0 Total funds® 695 990 1125 1230 1345 1360 1430 1475 1300 935 560 465 465 250 50 0 Total funds through 1994: $13,675. €S Table 3.3. Capital equipment fund requirements for chemical research and development Cost (thousands of 1978 dollars) for fiscal year — Task ——— e et 1980 1981 1982 1983 1984 1985 1686 1687 1988 1989 1990 1991 1992 1993 1994 Fuel chemistry 40 90 145 130 44 160 100 100 95 20 10 10 D 0 Coclant chemistry 25 85 150 100 95 165 163 210 85 35 10 0 0 0 Fuel-coolant interactions 30 10 15 40 30 50 60 15 0 0 0 ¢ 0 Fuel~graphite interactions o 20 45 40 15 30 22 0 5 0 0 0 0 0 0 Total funds” 95 205 355 310 0 “Total funds through 1991: $2500. 180 410 325 350 185 a5 20 10 0 1895 0 0 wn o~ 0 55 4. ANALYTICAL CHEMISTRY Scope and Nature of the Task With conventional solid-fueled reactors, there is neither the op- portunity nor an cbvious need for chemical analyses of the fuel during its sojourn in the reactor. On the other hand, for a fluid-fueled reac- tor, particularly cne that includes a fuel processing plant, there is a pressing need to know the precise composition (particularly the concen- trations of fissile materials) in several process streams. Moreover, such information is needed at frequent intervals (if not continuously) and on a real-time basis. As a consequence, such reactors are much more dependent on analytical chemistry than are, for example, LWRs. The MSRE was indeed coperated successfully with chemical analyses performed on discrete samples of fuel removed from the reactor for hot- cell study. However, it was recognized1 early that the MSBR and its as- sociated reprocessing plant would require in-line analyses. The DMSR would be equally dependent on successful development of such analytical techniques. These requirements are several in number. It will be necessary to determine on a virtually continuous basis the redox potential, the con- centrations of uranium, protactinium, and other fissionable materials and of bismuth and specific corrosion products (notably chromium) in the stream entering the reactor from the processing plant. Such information must also be available for the fuel within the reactor circuit and in the stream to the reprocessing plant. In addition, it would be highly de- sirable to know the oxide ion concentration in the fuel within the reactor and in the stream from the processing plant and tc know the states and concentrations of selected fission products in the fuel within the reactor. It will be essential to know the concentrations of uranium and fissile isotopes in the processing streams from which they could be lost (tc the waste system) from the complex. These streams will include the small stream of released off-gas, the LiCl system®* for rare-earth transfer, the ZrFy removal system® and perhaps a system® for removal of metallic ol “These systems are described in Chapter 6. 56 noble-metal fission products. 1In addition, it will be necessary to g monitor corrosion products, oxygenated compounds, and protonated (and tritiated) products in the coolant system and in the off-gas system as well as in the system for hold-up and recovery of tritium. Accordingly, on~line analyses will be required in at least three kinds of molten salts, in gases, and perhaps in molten bismuth alloys® within the processing plant. Key Differences in Reactor Concepts% Insofar as the analytical chemistry requirements (and the R&D needs) are concerned, the DMSR and the MSBR are quite similar. The DMSR will require determination of plutonium {(and to some extent of Am, Cm, and Np) to a degree markedly different frem the MSBR. Concentrations of CF3, UFy, and PaF, in the fuel stream will be higher in a DMSR (as will that of ZrF.) than in an MSBR. Complexities of the fuel processing plants for the two reactors are very similar, and, except for the presence of transuranics, so are the analytical requirements. In principle, the emphasis on proliferation resistance would seem to place additional demands on surveillance and precise determination of plutonium, protactinium, and uranium within the DMSR system. In fact, the requirements already imposed by the demands for safe and reliable continuous coperation of the complex are almost certainly at least as stringent, Pogt-1974 Technology Advances Several advances, a few amounting to breakthroughs, were made in the 1971-1974 interval.'»>® The post-1974 studies in analytical chemistry consisted mainly of (particularly valuable) service functions and, ex- cept for studies of tellurium behavior, contained little exploratory ¥ It is clearly desirable, and may be necessary, to have on-line de- terminations o¢f lithium and of some other metallic species in bismuth streams. See Chapter 3, Fuel and Cecolant Chemistry, for discussion of key chemical differences. 57 development cof the kind ultimately needed. Primary accomplishments in the post-1974 period were therefore relatively few. They did include the following: 1. On-line voltammetric techniques for determination of the UF3/UF, ratio were refined and were applied successfully and routinely in many corrosion test 100?8#0,@1 and engineering eX]_Je:r:'Lments.6’7’“0”“2 Such techniques can now be said to be well established in the absence of radiation {which should prove of little consequence) and of fission prod- ucts. 2. Measurements of protonated (and tritiated) species in the NaF- NaBFy coolant salt were applied successfully in engineering test equip- ment, ' %s%? 3. Voltammetric and chronopotentiometric techniques have been suc- cessfully applied to measurements of Fe?t in LiF-BeF,-ThF, (72-16-12 > using anodic voltammetry show prom- moie %),"" and preliminary studies® ise for in-line monitering of oxide level in this molten salt, at least under favorable conditions. Status of Analytical Development MSRE operation was mainly conducted with analyses performed on dis- crete samples removed from the reactor. The reactor off-gas was analyzed by in-line methods, and remote gamma spectroscopy was used to study fis- sion products; all other determinations were made using hot-cell techniques on batch samples. Of course, a major program of R&D preceded that opera- 1,2 tion. ? Substantial experience had been gained in the handling and analysis of nonradioactive fluoride salts prior to the MSRE program. Ionic or instrumental methods had been developed for most metallic constituents. For MSRE application it was necessary to develop additional techniques and to adapt all the methcods to hot-cell operations. A nonselective measurement of "reducing power" of adequate sensitivity had been developed L7 . \ . in the radiochemical (hydrogen evolution method).*® A general expertise separation and measurement of fission products was available from earlier reactor programs at ORNL, and useful experience with in-line gas analysis, particularly process chromatography,l+8 was available from other programs. 58 During the operation of the MSRE and in the subsequent technology program, development of methods for discrete samples was continued, and the Laboratory has acquired instrumentation for newer analytical tech- niquea—:.h"9 Instrumental methods expected to contribute to the program include x-ray absorptiocon, diffraction, and fluorescence techniques; nuclear magnetic resonance (NMR); spark source mass spectrometry; elec-— . tron spectroscopy for chemical analysis (ESCA) and Auger spectrometry; electron microprobe measurements; scanning electron microscopy; Raman spectrometry; Fourier transform spectrometry; neutron activation analy- sis; delayed neutron methods; photon activation analysis; and scanning with high-energy particles, e.g., protons. Key developments for MSRE Homogenized and free-flowing powdered samples of radioactive fuels taken from the MSRE were routinely produced in the hot cell within 2 hr of receipt. Salt samples were taken in small copper ladles that were sealed under helium in a transport container in the sampler—enricher50 for delivery to the hot cell. Atmospheric exposure was sufficient to compromise the determination of oxide and Ut but did not affect other measurements. lechniques for taking and handling of such samples (for those analyses for which they will suffice) are well demonstrated. Oxide concentration could not be reliably determined on the pul- verized salt samples because of unavoidable atmospheric contamination. Instead, 50-g samples of salt were treated with anhydrous HF gas and the evolved water was collected and determined.”' Oxide concentrations of about 50 ppm were determined with better than *10 ppm precisiocn.* Uranium analyses by coulometric titration showed good reproduc- ibility and precision (0.57), but on-line reactivity balance data estab- lished changes in uranium concentration within the circuit with about ten times that sensitivity.l Fluorination of the uranium from 50-g sam- 52 . and was used to separate uranium for ples was shown to be quantitative precise isotopic determination. If necessary, it could also have served as the basis for a more accurate uranium analysis. The rate of production of HF upon sparging of the fuel with Hy; is a function of the UF3/UFy ratio. This transpiration method, modified 59 to allow for other ions in the fue1,53 gave values in reasonable agree- ment with "book' values during operation with 2°°U (0.9 mole % U). The method proved inadequate at the lower concentrations during operation of the MSRE with 2%3U. Attempts tec determine UF3/UF, ratios by a voltam-— metric method using remelted salt samples was not generally successfulls>® because of prior UF: oxidation via atmospheric contamination. However, it was possible to follow UF3 generation via H, sparging upon such sam- ples.1 The radiation level of the samples does not appear to affect the method. A facility for spectrophotometry of highly irradiated fuel samples from the MSRE was designed and constructed.”" The system design included devices for remelting large salt samples under inert atmosphere and dis- pensing portions to spectrophotometric cells. The entire system could not be completed in time to give much useful data for MSRE.* It has since been used to observe spectra of transuranium elements and of pre- tactinium in molten salts. Feasibility of the general technique appears to be established. Equipment was installed at the MSRE to perform limited in-line analyses of the reactor off-gases, using a thermal conductivity cell as a transducer. An oxidation and absorption train®? permitted measurement of total impurities and hydrocarbens in the off-gas. The sampling staticn also included a system for the cryogenic ccllectien of xXenon and krypton on molecular sieves to provide concentrated samples for the precise de- termination of the isotopic ratios of krypton and xenon by mass spec- trometry. During the last two runs of the MSRE, equipment was installed’”® at the reactor to convert the tritium in various gas streams to water for measurement by scintillation counting. By means of a precise collimation system mounted on a maintenance shield, radiation from deposited fission products on components was di- rected to a high-resolution, lithium-drifted, germanium diode.”® From the gamma spectra obtained, specific isotopes such as noble-metal fission products were identified and their distribution was mapped by moving ® Observations with a somewhat makeshift sampling system showed no adverse effects from radiocactivity of the fuel. 60 the collimating system. During the latter runs of the reactor, such s . . 57 measurements were made during power operation. Analytical development for MSBR#* At present, it appears that the measurement of the concentration of major fuel constituents such as lithium, beryllium, thorium, and fluo- ride ion by in-line methods may not be practical in an MSBR. Fortunately, continuous monitering of these constituents will not be critical to the operation of a reactor. The more critical determinations, which were briefly described above, are generally amenable to in-line measurement. The ultimate need for an MSBR is an analytical system that includes all needed in-line analytical measurements that are feasible, backed up by adequate hot-cell and analytical laboratories. In the interim, ca- pabilities must be developed and analytical support provided for the tech- nology development activities in the program. Electrochemical studies. For the analysis of molten-salt streams, electroanalytical techniques such as voltammetry and potentiometry ap- pear to offer the most convenient transducers for remote in-line measure- ments. Voltammetry is based on the principle that when an inert elec- trode is inserted into a molten salt and subjected to a changing voltage relative to the salt potential, negligible current flows until a criti- cal potential is reached at which one or more of the ions undergo an electrochemical reduction or oxidation. The potential at which this reaction takes place is characteristic of the particular ion or ions. If the potential is varied linearly with time, the resulting current- voltage curve follows a predictable pattern in which the current reaches a diffusion-limited maximum value that is directly proportionai to the concentration of the electrocactive ion or ioms. Basic voltammetric studies have been made on corrosion-product ions in the MSRE fuel solvent LiF-BeF,;-ThF,>%7%2 and in the proposed coolant salt Tl.\“haBFq—l\IaF.61"63 Most of this work is concerned with the determina- tion of the oxidation states of the elements, the most suitable electrode ~ It seems clear that all items described under this heading would be of value to DMSR with only minor modification at most. 61 materials for their analysis, and the basic electrochemical characteris- tics of each element. It has been shown that relatively high concentra- tions (typically 20 ppm) can be estimated directly from the height of the voltammetric waves. Lower concentrations can be measured using the technique of stripping voltammetry through observation of the current produced when a corrosion product is oxidized from an electrode on which it has previously been plat,ed.6LP A voltammetric method has been developed for the determiration of the U3t/U*t ratio in the MSRE fuel.®® This method involves the measure- ment of the potential difference between the equilibrium potential of the melt, measured by a noble electrode, and the voltammetric equivalent 3+/U4+’ of the standard potential of the U couple. The reliability of the method was verified by comparison with values cobtained spectrophoto- ®3 This determination has been completely automated with a 66 metrically. PDP-8 computer, which operates the voltammeter, analyzes the data, and computes the U3+/U“+ ratic. Recently, the method was used to determine U3+/E“+ ratios in a thorium-bearing fuel solvent, LiF-BeF,-ThF., (68-20- 12 mole %). Ratios covering the range of 107° to >107% were measured during the reduction of the fuel in a forced-convection loop.2 The data support the reliability of the method in this medium. Because the fuel-processing operation presents the possibility for introducing bismuth into the fuel, a method for bismuth determination is 3* was characterized in LiF-BeF,- required. The reductive behavior of Bi ThFy,®? and it was found to be rather easily reduced to the metal. As an impurity in the fuel salt, bismuth will probably be present in the metallic state; so some oxidative pretreatment of the melt will be nec- essary before a voltammetric determination of bismuth can be performed. The measurement of the concentration of protonated species in the proposed MSBR coolant salt is of interest because of the potential use of the coolant for the containment of tritium. The measurement could also be used to evaluate the effect of proton concentrations on corrosion rates and as a possible detection technique for steam-generator leaks. A rather unique electroanalytical technique that is specific for hydro- 62,67 gen was investigated. The method is based on the diffusicn of hy- drogen into an evacuated palladium—-tube electrode when NaBFy melts are 62 electrolyzed at a controlled potential. The pressure generated in the s electreode is a sensitive measure of protons at parts-per-billion concen- trations. The technique offers the advantages of specificity, applica- bility to in-line analysis, and the possibility of a measurement of tritium-to~hydrogen ratios in the coolant by counting the sample col- lected from the evacuated tube. Measurements by this technique have led to the discovery that at least two forms of combined hydrogen are present in NaBFy4 melts. The availability of an invariant reference potential to which other electrochemical reactions may be referred on a relative potential scale is a distinct advantage in all electroanalytical measurements. The major problem was te find nonconducting materials that would be compati- ble with fluoride melts. Successful measurements were performed with a Ni/NiF, electrode in which the reference sclution (LiF-BeF; saturated 68,869 Standard with NiF;) is contained within a single-crystal LaF3 cup. electrode potentials were determined for several metal/metal-ion couples which will be present in the reactor salt streams.®® These electrode potentials provide a direct measure of the relative thermodynamic stabil- ity of electroactive species in the melts. This information can be used in equilibrium calculations tc determine which ions would be expected to be present at different melt potentials. As noted above, preliminary studies have indicated that, in at least some of the salt streams, an electrecanalytic method for oxide con-~ centration may be feasible. Determination of C17 in the fluoride meits (as may be necessary since "LiCl from the fission-product transfer system could contaminate the fuel) can probably be accomplished by voltammetric techniques. The MSBR regquired little effort on transuranic elements other than to determine whether traces of plutonium and higher transuranics inter- fered with determination of other pertinent species. Interference pos- sibilities will be intensified in the DMSR and thus quantitative deter- mintations of Pu, Am, Np, and Cm at various points must be provided. . Spectrophoteometric studies. Because molten fluorides react with the light-transmitting glasses usually emplcved, special cell designs have been developed for the spectrophoteometric examination of MSBR melts. 03 The pendant-drop technique70 that was first developed was later repiaced with the captive-liquid cell’? in which molten salts are contained by virtue of their surface tension so that no window material is required. A concept has been proposed for the use of this cell in an in-line sys- tem. ? The light path length through a salt in a captive-liquid cell is determinable but is not fixed. The need for a fixed path length pro- moted the design and fabrication of a graphite cell having small diamond- plate windows ® which has been used successfully in a number of research applications. Another fixed-path-length cell which is still in the de- velopment stage makes use of a porous metal £f0il’" that contains a number of small irregular pits formed electrochemically; many of the pits are etched completely through the foil so that light can be transmitted through the metal. Porous metal made from Hastelloy N has been pur- chased to test its use for cell construction. The latest innovation in cell design is an optical probe® which lends itself to a sealable insertien into a molten-salt stream. > The probe makes use of multiple internal reflections within a slot of ap- propriate width cut through some portion of the internally reflected > During measurements the slot would be below the surface light beam. ’ of the molten salt and would provide a known path length for abscrbance measurements. It is believed that the probe could be made of LaF: for measurements in NaBFy streams. Spectrophotometric studies of uranium in the 3t oxidation state have shown that this method is a likely candidate for in-line determina- 76,7 f . 577 An extremely sensitive absorption tion of U'Y in the reactor fuel. peak for ytt may be useful for menitoring residual uranium in depleted processing streams. ® Quantitative characterizations, including ab- sorption peak positions, peak intensities, and the assignment of spectra, have been made for Niz+, Fe2+, Cr2+, Cr3+, U5+, U022+, Cu2+, Mn2+, Mn3+, Co?t, Mo®t, cro,?~, Pa*t, Pudt, prit, Nadt, sm®t, Er’t, and Ho’T. Semi- quantitative characterizations, including absorption peak positions, ap- proximate peak intensities, and possible assignment of spectra, have also been made for Ti3+, V2+, V3+, Eu2+, Sm2+, Cm3+, and 0%~ ot "U.S. Pat. No. 3,733,130. 64 Evidence for the existence of hydrogen-containing impurities in NaBFy was first obtained from near-infrared spectra of the molten salt and in mid-infrared spectra of pressed pellets of the crystalline mate- rial.’? 1In deuterium—exchange experiments attempted in flucroborate melts, twe sensitive absorption peaks corresponding to BF30H™ and BF3;0D7 were identified. There was no evidence thatr deuterium would exchange with BF30H™; rather, BF:0D” was generated via a redox reaction with im- ® The absorption spectra of several other species 1 purities in the melt.® have been observed in fluoroborate melts.® Work on spectrophotometric methods is also providing data fer the identification and determination of solute species in the various melts of interest for the fuel-salt processing system.62 Gas analysis. Some determinations on MSRE samples {see preceding section) were done by treatment of the salt to produce gases for analy- sis. Little development of such devices has been attempted since the MSRE ceased operations. The electrolytic moisture monitor was demon-— strated to provide more than adequate sensitivity for the measurement of water from the hydroflucrination method for oxide and to have ade- quate tolerance for operation at the anticipated radiation levels.®? A method has been developed for the remote measurement of micromolar quan- tities of HF generated by hydrogenation of fuel samples using a thermal- conductivity method after preconcentration by trapping on NaF.®3 Commercial gas chromatographic components for high-sensitivity measurement of permanent gas contaminants are not expected to be accept- able at the radiation levels of the MSBR off-gas. Valves contain elas- tomers that are subject to radiation damage and whose radiolysis products would contaminate the carrier gas. The more sensitive detectors generally depend on ionization by weak radiation sources and would obviously be affected by sample activity. A prototype of an all-metal sampling valve®" has been constructed to effect six-way, double-throw switching of gas streams with closure provided by a pressure-actuated metal diaphragm. A helium breakdown detector was found to be capable of measuring ~ Processing cycle time 1is the time required for processing a2 volume of fuel salt equal te that contained in the reactor system; for the refer- ence MSBR, the optimum appeared to be about ten days. Removal time for a particular species is an effective cycle time equal to the processing cycle ¢ time divided by the fraction of that species removed during a pass through the processing system. Lo ‘A DMSR that is routinely fed fissile material of nonweapons grade (i.e., 20% “°°yU in ?°®U) could be built without chemical processing for removal of fission products. e 89 alloy more readily than do the rare earths. Prior separation and re- covery* of these materials must, accordingly, be part of the processing scheme. The processing system must, in addition, (1) maintain the fuel at an acceptable redox potential (UF3/UF, ratio), (2) keep oxide and cor- rosion products to tolerable levels in the fuel, (3) remove radioactive species from any {gaseous) effluent streams, and {(4) place the recovered fission products into waste forms suitable for at least temporary storage at the reactor site. Fuel processing for the MSBR was far from a demonstrated reality at the termination of that effort,l’2 but all key separations had been re- peatedly demonstrated individually on a small scale. Overall feasibility seemed assured from a chemical viewpoint, but much work remained beafore engineering feasibility could be assured.T The status and the remaining pressing needs in processing R&D — and their relationships to those of DMSR — are described in the following section. Key Differences in Reactor Concepts The fuel mixtures for the MSBR and DMSR are similar in many regards,* and the overall processing concepts share many features. However, both the fuel chemistryi and the processing differ in several important ways. 233Pa A very important feature of MSBR processing was the removal of from the fuel on a short cyele time (ten days) and its isolation in a molten~-salt reservoir outside the reactor for decay there to essentially 2 pure 331 . This product was recovered by fluorination; that needed by the reactor was reintroduced into the fuel, and the excess was stored for sale and use elsewhere. The proliferation resistance imposed on a DMSR for this study obviously necessitates abandonment of that portion of the % Uranium can be separated by flucrination to volatile UFg; protac— tinium, plutonium, and transuranics other than neptunium cannot; however, they can be recovered by prior extraction into diiuvte lithium-bismuth alloy. TMaterials of construction of the several equipment items pose sub- stantial problems (see Chapter 5). + See Chapter 3. 20 MSBR system. As a nontrivial consequence, such abandonment requires that the DMSR have an alternative scheme for removal of fission-product zirconium, which was sequestered (on the ten-day cycle time) with “°*Pa in the MSBR processing plant and was discarded as waste after decay of : the 2%3pa, The MSBER produced very little Pu, Am, Np, and Cm; and since what was produced was sequestered with the 233pa and discarded to waste with the zirconium, the equilibrium fuel contained very small quantities of these materials. The DMSR is a prolific producer of plutonium (though at near equilibrium the plutonium is of relatively poor quality), and it needs to burn the fissile plutonium isotopes as fuel. The DMSR process- ing plant, therefore, needs to recover plutonium quantitatively and to return it immediately (along with #33Pa) to the reactor. As a conse- quence, DMSR fuel will contain much larger quantities of transuranic isotopes than did the MSBR. The DMSR will ncot breed and will not have excess fissile material for removal and sale for use elsewhere. However, it is likely that in 235U its early clean cperation (given start-up on at 207 enrichment) it 233 . . . . . U; this will, of ceourse, exist in a suitably wilil generate an excess of denatured state but some storage of it (on NaF beds within the reactor containment) may be required until it is needed by the reactor. Conceptual processes for the MSBR and DMSR, accordingly, employ very similar unit processes. Uranium is largely recovered by fluorina- tion to UF¢ and is returned immediately to the reactor fuel. In the DMSR, protactinium, plutonium, and the transuranium nuclides are recovered by selective extraction into dilute lithium-bismuth alloy and are immediately returned to the reactor. In both concepts the rare-earth, alkaline-earth, and alkali-metal fission products are selectively extracted into bismuth- lithium alloy and subsequently transferred to molten LiCl for recovery as waste. As in the case of fuel chemistry discussed above, the process- ) ing for the two reactors shows far more similarities than differences. w 91 Post-1974 Advances Both the proposed MSBR'!*? and the DMSR processes require removal of uranium by fluorination from the fuel salt and from a waste salt before discard. Removal from the MSBR fuel salt virtually requires* continuocus fluorination, and such fluorination from the waste stream is desirable. The salt streams in the reactor processing plant contain much radiocac- tivity and are appreciable volumetric heat sources; cooling of the ves- sel wall to form a frozen salt film without freezing the vessel contents would certainly seem feasible. Tests with normal (nonirradiated) salt mixtures must introduce this volumetric heat source artificially, and 110,111 this has proven to be difficult.T Repeated attempts to demon- strate adequate frozen walls with nonradiocactive salt have been virtual failures because of malfunction of the resistance heating systems. Early studies'*® had shown that the sorption of UFg¢ in molten LiF- BeF;-ThF, containing UFy by the reaction UFs(g) + UFq(d) -+ ZUFS(d) is rapid but that the reaction 1 = H + UF -+ UF + HF 2 " (g) °(d) ! (d) (g) is slow in the absence of a catalyst. The resaction, however, proceeds rapidly when platinum black,112 platinum alloyed with the gold con- y 112 113 tainer, or even a limited area of smooth platinum serves as the catalyst. A facility to study this step on an engineering scale was 11% with molten salt and inert gas, but it was not built and checked out operated with UFs. Two engineering assemblies to study the reductive-extraction—metal- transfer processes were completed and successfully operated in the post- % A DMSR with a processing cycle time of 100 days or more could pos- sibly use batch fluerination; the economic penalty might be acceptable. Successful frozen wzlls have been obtained with Calrod heaters in the salt, but such apparatus is hardly suitable for use in fluorination of uranium. 92 10¢C 1974 pericd. These have been documented in some detail. %> These assemblies both used mechanically agitated, nondispersing contactors® to equilibrate molten salt and molten bismuth alloy. Both assemblies were built of carbon steel, though a graphite liner was used in a por- tion of one of the apparatuses. #% nine runs were made to establish rates of In the first of these, mass transfer of °°’U and °’Zr between LiF-BeF,-ThF, and bismuth as a function of agitator speed and salt and metal flow rates. These studies showed that the system could be readily operated, that mass—transfer rates increased with agitator speed, and that mass-transfer rates in- creased markedly with only minor phase dispersal. Although confirmation is needed, these studies also suggested that some phase dispersal might be tolerated without undue contaminatiocn of the fuel solvent with bismuth. The more ambitious experimentIOG demonstrated (primarily with neo- '*7Xd as tracer) the metal-transfer process for removal of dymium, using rare—earth fission products from melten LiF-BeF;-Thy into dilute bismuth- lithium and their subsequent transfer to molten LiCl and then to concen- trated bismuth~1ithium alloy. This process was demonstrated on a small engineering scale [about 1% of the flows required for a ten-day processing cycle on a 1000-MW(e) MSBR]. Separation of the rare earths from thorium was demonstrated to be essentially that projected from laboratory-scale studies.’” However, overall mass-transfer coefficients were lower than would be required for full-scale metal-transfer process equipment of reasonable size; this was particularly true of the two Bi-LiCl interfaces. There, however, it is possible that some phase dispersion can be tolerated. A considerable study of the characteristics of mechanically agitated 5 . . 1 . nondispersing salt-metal contactors was concluded''® using mercury and H;0 to simulate the bismuth-salt system. The code for computer calculation of the MSBR processing plant per- A - . . . 16 formance was further refined and its use described in detail.!'!® ol o~ Such contactors, in which the phases are not dispersed, permit opera- tion with higher ratios of one phase to the other. They should lead to less entrainment of bismuth in salt, and they are simpler to fabricate than are extraction columns. They also must be expected to show poorer mass transfer characteristics. i 93 Status of Technology The chemical basis on which the processing system is founded is well understood; however, only small engineering experiments have been carried out to date and a considerable engineering development effort remains. Chemical status Fluorination and fuel reconstitution. Removal of uranium from mol- ten fluoride mixtures by treatment with F» is well understood. Initial studies at ORNL led to the Fused Salt Fluoride Volatility Program, and batch fluorination of the irradiated Adircraft Reactor Experiment fuel was successfully demonstrated.’'’ The studies culminated in the highly suc- cessful recovery of uranium from various irradiated zirconium-, aluminum-, 118 which in some cases were processed as 119 and stainless steel-based fuels, early as 30 days after fuel discharge. Uranium recoveries greater than 997 and uranium decontamination factors in excess of 10° were consistently demonstrated. More recently the Fused Sait Flueride Volatility Process was used for removal of the “°°U-?3%U mixture from the MSRE fuel salt at 1260 the MSRE site after the reactor had operated for about 1.5 years. This operation was also highly successful, and the fuel carrier salt was sub- 233y and returned to the MSRE for an additional sequently combined with vear of operation. There is no doubt that essentially quantitative re- covery of uranium can be accomplished 1f necessary and that many details of fission-product behavior are well understood. Such flucrination also serves to remove oxide and oxygenated compounds (via their conversion to fluorides and 02) from the melt. However, for the MSBR a continuous fluocr- inator {probably of nickel protected by a layer of frozen salt) is essen- tial, and such a device is, at least, highly desirable for the DMSR. Additional'study is needed to develop and demonstrate such a device. Gas phase reduction of UFg to UFy by hydrogenation is a well-known operation in the nuclear industry, and this process was initially consid- ered applicable for the MSBR.!»? However, consideration of the difficul- ties associated with equipment scale-down, UF4 product collection and holdup, and remote operation prompted a search for a more direct means for recombining UFs with molten fluoride mixtures. The known chemical 94 behavior suggested that UFg could be absorbed directly into molten salt that contained UFy. Subsequent experiments verified that the absorption reaction is rapid and that UFg can be combined quantitatively with mol- ten fluorides containing UF4 with the simultaneous formation of inter- 12 , \ ! Tn the fuel reconstitution mediate fluorides having a low volatility. step, a gas stream containing UFe¢ and Fp can be reacted with a recircu- lating salt stream containing dissolved UFy according to the reactions - 1 = T PFuay 72 Faqgy T Usqgy and UFL&(d) + UFs(g) = ZUFS(d) . The dissoclved UFs can be reduced in a separate chamber according to the reaction ° \ 1 _ UEs gy ¥ 2 Haqgy T UFegqy T HF . The final reaction is relatively slow in equipment of goldl’z’121 but, as noted above, can be effectively catalyzed by platinum. Additiomal study is needed to establish whether, for example, icdine fluorides, TeFg, SeFg, etc., are absorbed by the strongly oxidized UFs solution. Selective reductive extraction. Selective extracticn from molten fluoride mixtures and from molten LiCl into lithium-bismuth alloys has 2,158 ¥ been studied in detail for essentially all the pertinent elements. '’ Bismuth is a low-melting-point (271°C) metal that is essentially immiscible with molten halide mixtures consisting of fluorides, chlorides, and bromides. The vapor pressure of bigmuth in the temperature range of interest (500 to 700°C) is negligible, and the solubilities of Li, Th, U, Pa, and most of the fission precducts are adequate for processing ap- plications. Under the conditions of interest, reductive extraction reac- tions between materials in salt and metal phases can be represented by the following reaction: MXn(salt) + nLi(Bi) = M(Bi)} + nLiX(salt) , in which the metal halide MXn in the salt reacts with lithium from the bismuth phase to produce M in the bismuth phase and the respective lithium halide in the salt phase. The valence of M in the salt is +n, 515 thatr the and X represents fluorine or chlorine. It has been found'»? distribution coefficient D for metal M depends on the lithium concentra- tion in the metal phase (mole fraction, X Li) as follows: log D = n log Xii + log K; The quantity K; is dependent only on temperature, and the distribution coefficient is defined by the relation _ mole fraction of M in metal phase mole fraction of MXn in salt phase The ease with which one component can be separated from another is indi- cated by the ratio of the respective distribution coefficients, that is, the separation factor. As the separation factor approaches unity, sepa- ration of the components becomes increasingly difficult. On the other hand, the greater the deviation from unity, the easier the separation. Distribution data have been obtained for many elements!*?21%91° he- tween LiF-BeF,~ThF, (72-16-12 mole %) and bismuth-lithium and between LiCl and bismuth-lithium. As Fig. 6.1 indicates, extraction from the molten fluoride affords excellent separation of Zr, U, and Pa from Th and the rare earths but relatively poor separation of the rare earths from thorium. Plutonium, neptunium, and americium are slightly more ex- tractable from the fluoride than is protactinium; curium is slightly less extractable than protactinium. Figure 6.2 shows the quite different be- havior when molten LiCl is used. Excellent separations of thorium from the rare-earth and alkaline-earth elements can be made by use of LiCl. The distribution coefficient for thorium is decreased sharply by the addition of fluoride to the LiCl, although the distribution coefficients 96 ORNL DWG 70-1250i E02h T T EIIIIT’ T T T T T T 1] T T - Salt: 72-16-12 Mole % Zrt4 43 - ] LiF-BeFz-ThFg - - Temp: 640 °C i 10+ ] s ] - 3 2z w i O E I.O'__ — b - B m - - o i 2 o ] > 4 o ] [ i 4 o @ or - lOm|_— — 2 - ; IO"E;“ — 50’3 L by ) oot el A b i L) g3 1074 10°3 10"2 MOLE FRACTION Li IN BISMUTH Fig. 6.1. Distribution data between fuel salt and bismuth. . 97 ORNL-DWG-70-12502 |03 [ Y T T b+ Th¢ 640°C U+3 Ce+3 Nd+3 2 10°}- ] - Lat3 = Lii [®] z o' - e L O |9 5 = 5 @ o 10°}- sm*2 - v [an] +2 o1 / o | Sr+2 Q7S 1074 1073 102 10! 1.0 MOLE FRACTION Li IN BISMUTH Fig. 6.2. Distribution data between lithium chloride and bismuth. for the rare earths are affected by only a minor amount. Thus, contami- nation of the LiCl with several mole percent fluoride will not affect the removal of the rare earths but will cause a sharp increase in the thorium removal rate. Data with LiBr are similar to those with LiCl, and the dis- tribution behavior with LiCIl-LiBr mixtures would not be likely to differ appreciably from the data with the pure materials.!?? Conceptual MSBR processing flowsheet 23 is shown in Fig. 6.3. The reference MSBR processing flowsheet?»! Fuel salt is withdrawn from the reactor on a ten-day cycle; for a 1000- MW(e) reactor, this represents a flow rate of 55 em®/s (0.88 gpm). The 98 ORHL DWG 71-7852 {05 MOLE FRAC Lé} UF, EXTRACTOR B":z Ry ThF, SALT DISCARD SALT UFg & FROCESSED SALT PURIFICATION REDUCTION } é"“" 1 % 1 | | 1 H i : > EXTRACTOR | ' (Bi [ EBl [ l'—-,-j ; : L ______________________________________________ ~ g : | ! 1 i | [ uF i i EXTRACTOR | 1 1 1 REACTOR ) Lmj ! . | EXTRACTOR ] 1 —_— : LiCl —===Bi-Li | ! 1 I I I T i ? ¥ == EARTHS | HF Fz ; } ; —--Bi-Li L PA DECAY =~ , (0.05 MOLE FRAC. I 1 £ | | | | | | 1 | ! | ! 1 6 o ! | L«E HYDROFLUOF;H FLUORINATORJ-«-{ PA DECAY ] | Bi-Li 5 L — o= +DIVALENT RARE I : t I i I | | ! 1 [ 95 mole %) of U0z, by deliberate addition of oxide ion to the MSBR 5+, its oxide would, of fuel.’® If the protactinium were oxidized to Pa course, also be precipitated. 1If, in addition, plutonium were oxidized to Pu'Tt (this oxidation is more difficult® and a small concentration of dissolved F2 might be required), Pu0; would also be included in the solid solution. Whether americium, neptunium, and curium could be made to pre- cipitate as oxides is not vyet known. Engineering studies of uranium oxide precipitation have been car- ried out;l38 the studies involved the contact of 2 liters of MSBR fuel salt with Hy0~Ar gas mixtures in a 100-mm-diam nickel precipitator. Ex- periments were conducted at temperatures ranging from 540 to 630°C, and the composition of the HyO0-Ar mixture was varied from 10 to 35% water. The values for the water utilization were uniformly low (about 10 to 15%) and did not vary with the composition of the gas stream. Samples of the oxide contained about 90% UQ; even though, at the lower uranium Jo w5 See Chapter 3. TO REACTOR REACTOR -——-—w—fi PRIMARY I J [——— o 2 e sl o e e o e } HF TG UF4 ThF,, BeF, | RECYCLE ADDITION i === l i Ve VALENCE| [ UF UF ADJUST-- orian Y ) REDUCTION ABSORPTION = MENT - ‘ 1 REMOVAL b e ? A T T __ e TEmETmTmTTETEETETTTmTmmET e 9 ¢ ot Zr *—l RARE pe—— ¢ [ PARTITION fgm—me EARTH ! | ‘L EXTRACTION i o e e emn ® § 1 i T T T T T T I T ISR T T et P i ' i Pa | ? I Zr i Pu t g EXTRACTION l——-; EXTRACTION 1 7 I Y |- e i FROM ¢ | e mmesa—————— FLUORINATOR g2 SECONDARY psibiid HYDRO- FUEL OR FUEL SOLVENT ————— BISMUTH ~——e— (f —+ =+~ LiCl =+ WASTE SALT FLUORIN- HE 1 FLUORIN- i) ATOR . ATIOR C [, irT L S el M e St e i v Bl e . e e e REZ ACCUMULATION ¢OT Cs AND Rb | LACCUMULATORS® 1 ! 1 + ' ACCUMULATION Ah+l+-+1 ADD REZ [ = — Licl STRIPPER g+ 4 i + I RARE |+q L---—-—I —————————— EARTH | § + TRANSFER] § I T.O -ll‘- —+—-+1+—+-—*): LiC: ; | oo e ¥ ! + Y ! ) EXCESS | 1 TeT fm e Y FUEL | STRIPPER SOLVENT SRR i } + e T FP.AND WASTE SALT I FOR RECOVERY OR .. T ULTIMATE DISPOSAL v Fig. 6.4, Preliminary flowsheet for fuel processing in a DMSR. 106 concentrations in the salt, the solid in equilibrium with the salt would contain 50% U0, or less. This enhancement of the uranium concentration in the sclid phase appears to allow precipitation of 997 of the uranium as a solid containing 857% U02 in a single-stage batch precipitator. The oxide precipitate was observed to settle rapidly, and more than 907% of the salt could be separated from the oxide by simple decantation.' Thus ¢ an oxide precipitation scheme could possibly recover together the U, Pa, and Pu, along with a small amount of Th, as oxides. These could then be returned to the purified fuel solvent (by hydrofluorination) for return to the reactor. It would be necessary, of course, to remove residual 0* from the fuel solvent before the rare-earth removal step. If processing of a DMSR on a cycle time of 100 days or more is practicable (processing rate of <1 m® of salt per day), such an oxide precipitation might be used as a batch operation.¥ We know of no method for rare-earth removal that is comparable to the reductive—extraction—metal-transfer process. Many attempts have been made to find lon-exchange systems capable of removal of rare-earth ions from LiF-BeF; and Li¥F-BeF,-Th¥, mixtures. All but a very few such materials are unstable in contact with the salts. The only stable one known to have ion-exchange capabilities is Cels3, which will exchange Cet for other rare-earth ions,zQ but it is too soluble in LiF-BeF,-ThF, to be genuinely usefu1.+ The fuel solvent, partially freed from other rare earths but saturated with CeFs, could possibly be freed from cerium * and precipitation of Ce0;. If so, the resulting by oxidation to Ce" LiF-BeF,-ThFy would again have to be treated to remove excess 0*~ before its return (with the valuable fissile and fertile constituents) to the reactor, Several combinations of the preferred processes with some of the alternatives are possible. Their attractiveness increases as the per- missible processing cycle time lengthens. It seems certain, however, that all are less attractive than that represented by the reference MSBR unit processes. el Solids handling is axiomatically more difficult than fluids hand- . ling, especially in continucus operations. Cerium is an appreciable neutron absorber. 1G7 Primary R&D Needs The primary needs, with relatively few exceptions, are for sound engineering tests of individual process steps and ultimately for rela- tively long-term and near full-scale integrated tests of the system as a whole. For the former tests the facilities needed are relatively mod- est, though test equipment and instrumentation are quite complex. For the integrated tests a special engineering laboratory will be required and an integrated process test facility must be provided with appropriate consideration of the need to demonstrate remote maintenance. Specific needs include 1. Determination by computer calculation, in close coordination with reactor neutronics studies, the permissible range of (and the op- timum) processing cycle times and removal times. Demonstration of frozen-wall fluorination of uranium on a batch, and, hopefully, on a continucus, basis. Determination of behaviocr of neptunium in this fluorinatiom. Demonstration of adequate UFg¢ abscrption in LiF-BeF,;-ThF,;-UFy; mix- ture and suitable reduction to UFy and to 7 to 10% UF3;. Determina- tion of behavior of an NpFg in absorption system. Development and demenstration of a method for removal (by selective extraction or otherwise) of fission-product zirconium from the fuel. Demonstration of quantitative recovery of protactinium and plutonium by selective extraction on a scale at least 25% of that required for a DMSR. Determination of efficiency of recovery of americium, neptunium, and curium in that extraction. Betermination of behavior of TeFy, SeFg, IFs5, etc., in the UFg ab- sorption step. Demonstration of retention of TeFg, SeFg, Iz, etc., from the UFg absorption off-gas. Demonstration of adequate removal of rare-earth, alkaline-earth, and alkali-metal fission products in a complete metal-transfer system. Demonstration of hydrofluorination of zirconium, rare-earth, etc., fission products into waste salt for storage. 108 10. Demonstration of the application of bismuth containing U, Zr, Pa, Pu, Th, Li, etc., for valence adijustment of fuel salt. 11. Demonstration of adequate removal of bismuth (by absorption on nickel or gold weol) from salt for return to the reactor. 12. Operation of the entire integrated system reliably for moderate to long times with realistic construction materials and reasonable con- centrations of species at tracer level activity (where possible). Assessment of overall performance, achievable oxide concentration, effect of system upset on fissile losses to waste, etc. Estimates of Scheduling and Costs Preliminary estimates of the necessary schedule and of its operating and capital funding requirements are presented below for the fuel pro- cegssing development described above. As elsewhere in this document, it has been assumed that (1) the program would begin at start of ¥FY 1980, (2) it would lead to an operating DMSR in 1995, and (3) the R&D program will produce no great surprises and no major changes in program direc- tion will be required. The schedule, along with the dates on which key developments must be finished and major decisions made, is shown in Table 6.1. It seems certain that the overall R&D programs (including those described else- where in this document) will provide some minor surprises and that some changes in this development effort will be required. No specific pro- visions for this are included; but, unless major revisions become neces-— sary in the middle 1980s, it appears likely that suitable chemical pro- cesses and processing equipment could be recommended onr this schedule. The operating funds (Table 6.2) and the capital equipment require- ments {Table 6.3) are shown on a year-by-year basis in thousands of 1978 dollars. ©No allowance for contingencies, major program changes, and inflation during the interval have been provided. As with the develop- ment of materials for chemical processing (Chap. 5), much of this effort could be deferred if a decision were made to delay the development of a break-even fuel cycle for the DMSR. Table 6.1. Schedule for fuel processing development Fiscal wvear Task 1680 1981 1982 1983 1984 1985 1986 1987 1983 1949 1990 1991 1992 1993 1994 1995 q 2 3 flewsheet develonment v Y Y4 v 5 B 7 Fluorinator development % v v 8 g 7 Fuel reconstitution v v v 10 ‘71$ 12 Pa, Pu, etc., recovery 13 i4 Rare-earth removal v v Valence adjustment and v’s vm purification ‘717 MSR Process Laboratory - N 18 18 Integrated Process Test v Y Facility Milestones: 1. Define range of possible values for processing cycle time 10, Complete engineering studies of reductive extraction in a and removal times. mild-steel {low-through system. 2. Define optimum processing time. 11. Demonstrate recovery of protactinium by reductive extraction 3. Decide on system for removal of zirconium for engineering using gram quantities of “3ipg, tests. 12. Demonstrate recovery of Pu, Am, Cm, etc., using gram quanti- 4. Complete [lowsheets for conceptual DMSR. ties of Pu. 2. Test batch frozen-wall fluorinator. 13. Extend experiments in mild-steel system at 1% of DMSR scale. 6. Complete studies of continuous fluorination in engineering 14, Complete engineering experiment 5 to 10% DMSR scale, facility. 15. Demonstrate removal of trace quantities of bismuth. 7. Complete studies of combined fluorination-recombination in 16. Demonstrate continuous adjustment of uranium valence. engineering system on 25 to 507 DMSR scale. 17. Complete construction of MSR Processing Engineering Labora- 8. Complete engineering studies of fuel recenstitution neces- tory. sary for design of Fluorination-Reccnstitution Engineering 18. Complete installation of Integrated Process Test Facility. Experiment. 19. Complete operations and tests with Integrated Process Test 9. Complete engineering studies of reductive extraction in Re- ductive Extraction Precess Facility. Pacility. 60T Table 6.2. Operating fund requirements for fuel processing development Cost (thousands of 1978 dollars) for fiscal year — sk 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1692 1993 1994 1995 Flowsheet development 40 150 250 300 150 75 40 0 0 0 0 O 0 0 0 0 Fluorinator development 290 390 490 315 100 50 0 0 0 0 0 0 0 0 0 0 Fuel reconstitution 200 340 220 310 200 100 0 0 0 0 0 0 g 0 0 0 Pa, Pu, etc., recovery 330 550 610 540 200 75 0 0 0 0 0 0 0 0 0 0 Rare-earth removal 270 200 235 195 100 50 0 0 0 0 0 0 0 0 0 0 Valence adjustment and 50 60 100 100 75 20 0 0 0 0 0 0 0 o 0 0 purification MSR Process Laberatory 105 65 325 385 200 250 160 O 0 0 0 0 0 @ 0 0 Integrated Process Test Q 215 250 310 455 360 770 3200 3670 3670 3510 2000 500 0 0 0 Facility ‘ - — ; ) e —— Total funds” 1285 2170 2480 2455 1480 1010 910 3200 3670 3670 3510 2000 500 0 0 0 “Total funds through 1992: $28,340. b Additional funds related to fuel reprocessing will be required during these years in support of test reactor and test reactor mock-up. The variation in overall support level, therefore, will be considerably less abrupt. OTIl Table 6.3, Capital equipment fund requirements for fuel processing development Cost (thousands of 1978 dollars) for fiscal year — Task - o —— 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 Capital equipment facilities Flowsheet development 0 0 0 0 0 0 0 O 0 0 0 0 0 0 0 0 Flueorinator development 65 260 50 0 0 g 0 0 0 0 0 g 0 0 0 0 Fuel receonstitution 10 165 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Pa, Pu, etc., recovery 0 285 600 0 0 0 0 0 0 0 0 0 0 0 0 0 Rare-earth removal 0 300 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Valence adjustment and 0 50 100 0 0 0 0 0 0 0 0 0 0 0 0 purification IPTF:” data processing 0 0 0 0 0 510 0 260 400 515 400 200 0 0 0 0 Total funds 75 1060 750 0 G 510 0 260 400 515 200 200 0 o G G Capital projects MSR Process Laboratory lB,OOOfi 1 Integrated Process Test 7000”7 Facility - . IPTF = Integrated Process Test Facility. ®Total funds through 1991: $23,170. 17T o Ui ~ " 10. 11. 12. 13. 14, 15. 112 References M. W. Rosenthal et al., The Development Status of Molten-Salt Breeder Reactors, ORNL-4812 (August 1972). L. E. McNeese et al., Program Plan for Development of Molten-Salt Breeder Reactors, ORNL-5018 (December 1974). Molten-Salt Reactor Program, Semiavnual Progress Report, ORNL-5047 (September 1975). Molten-Salt Reactor Program, Semionnual Progress Report, ORNL-5078 (February 1978). Molten-Salt Reactor Program, Semiannual Progress FReport, ORNL-5132 (August 1976). H. E. McCoy, Jr., Status of Materials Development for Molten-Salt Reactors, ORNL/TM-5920 (January 1978). J. R. Keiser, Status cof TelluriumHastelloy N Studies in Molten Flucride Salte, ORNL/TM-6002 (October 1977). L. 0. Gilpatrick and L. M. Toth, '"The Hydrogen Reduction cof Uranium Tetrafluoride in Molten Fluoride Solutiomns," J. Inorg. Nucl. Chem. 39(10), 18171822 (1977). G. Long and F. F. Blankenship, The Stability of Uranium Trifluoride, ORNL/TM-2065 (November 1969). G. T. Mays, A. N. Smith, and J. R. Engel, Distribution and Behavior of Tritium in the Coolant-Salt Technology Faeility, ORNL/TM-5759 (April 1977). J. T. Bell, J. D. Redman, and F. J. Smith, "Permeation of Construc- tion Materials by Hydrogen Isotopes,' Chemistiry Division Annual Progress Report, CORNL-5297 (September 1977). W. R. Grimes, '"Molten-Salt Reactor Chemistry," Nucl. Appl. Technol., 8(2), 137-155 (February 1970). C. E. Bamberger et al., J. Inorg. Nucl. Chem. 33, 3591 (1971). Molten-Salt Breeder Reactor Concept, Quarterly Report for Period Ending July 31, 1971, NP-19145, Atomic Energy Commission, Bombay, India. L. M. Ferris et al., "Equilibrium Distribution of Actinide and Lan- thanide Elements between Molten Fluoride Salts and Liquid Bismuth Solutions," J. Inorg. lucl. Chem. 32, 2019 (1970). 16. 17. 18. 19. 20. 21. 22. 23. 24, 25. 26. 27. 28. 29. 30. 31. 113 C. F. Baes, Jr. , "The Chemistry and Thermodynamics of Molten-Salt Reactor Fuels," p. 617 in Nuclear Metallurgy, Vol. 15, Symposium on Reprocessing of Nuclear Fuels, edited by P. Chiotti, CONF 690801, U.S. Atomic Energy Commission, Divisien of Technical Information, (August 1969). C. E. Bamberger and C. ¥. Baes, Jr., J. Nucl. Mat. 35, 117 (1970). C. E. Bamberger, R. G. Ross, and C. F. Baes, Jr., J. Inorg. Nucl. Chem. 33, 767 (1971). C. E. Bamberger, R. G. Ross, and C. F. Baes, Jr., in Molten=Salt Reactor Program Semiannual Progress Report, Febrvuary 29, 1972, ORNL-4782, pp. 70—76. C. E. Bamberger et al., in Molten-Salt Feactor Program Semiannual Progress Repovrt, February 28, 1971, ORNL-4728, p. 62. C. E. Bamberger, R. G. Ross, and C. F. Baes, Jr., in Molten-Salt Reactor Program Semionnual Progress Report, February 28, 1971, ORNL-4676, p. 119. C. E. Bamberger, R. G. Ross, and C. F. Baes, Jr., in Molten-Salt Reactor Program Semiannual Progress Report, August 31, 1971, ORNL= 4622, p. 92. C. F. Baes, Jr., in Molten-Salt Reactor Program Semiavmual Progress Report, February 28, 1970, ORNL-4548, p. 152. S. Cantor, Density and Viscosity of Several Molten Fluoride Mix- tures, ORNL/TM-4308 (March 1973). S. Cantor (Ed.), Physical Properties of Molten-Salt Reactor Fuel, Coolant, and Flush Salts, ORNL/TM-2316 (August 1968). J. W. Cooke, Development of the Variable-Gap Technique for Measuring the Thermal Conductivity of Fluoride Salt Mixtures, ORNL-4831 (Febru- ary 1973). E. L. Compere et al., Figsion-Product Behavior in the Molten-Salt Reactor Experiment, ORNL-4865 (October 1975). A. P. Malinauskas and D. M. Richardson, "The Solubilities of Hydro- gen, Deuterium, and Helium in Molten Li.BeF,," Ind. Eng. Chem. Fund. 13, 242 (1974). W. R. Grimes, N. V. Smith, and G. M. Watson, J. Phys. Chem. 62, 862 (1958). M. Blander et al., J. Phys. Chem. 63, 1164 (1959). G. M. Watson et al., J. Chem. Eng. Data 7, 285 (1962). 32. 33. 34, 36. 37. 38. 39. 49, 41. 42. 43, b4, 45. 46. 114 R. B. Briggs and R. B. Korsmeyer, in Molten-Salt Reactor Program g Semiannual Progress Report, February 28, 1370, ORNL-4584, pp. 53—54. See review of many determinations by C. F. Baes, Jr., "The Chemistry and Thermodynamics of Molten-Salt Reactor Fuels," Nuclear Metallurgy, Vol. 15, CONF-690801, U.S. Atomic Energy Commission, Division of Technical Information, August 1969. A. D. Kelmers et al., Evaluation of Alternate Secondary (and Terti- 5 ary) Coolants for the Molten-Salt Breeder FReactor, ORNL/TM-5325 (April 1976). L. Brewer et al., "The Thermodynamic Properties of the Halides," in The Chemistry and Metallurgy of Miscellaneous Materials: Thermo- dynamics, pp. 76—192, L. L. Quill, ed., McGraw-Hill, New York, 1950. J. F. Elliott and Molly Glieser, Thermochemisiry for Steelmaking, Vol. 1, Addison-Wesley, Reading, Mass., 1960. L. M. Toth and L. 0. Gilpatrick, The Equilibrium of Dilute UF; Solutions Contained in Graphite, ORNL/TM-4056 (December 1972). 7 L. M. Toth and L. 0. Gilpatrick, J. Phys. Chem. 77, 2799 (1973). R. A. Strehlow, in Molten-Salt Reactor Program Semiannual Progress Report for Period Ending February 28, 1970, ORNL-4548, p. 167. J. R. Keiser, Compatibility Studies of Potential Molten-Salt Breeder Reactor Materials in Molten Fluoride Salts, ORNL/TM-5783 (May 1977). J. R. Keiser, J. H. DeVan, and D. L. Manning, The Corrosion FKesis- tance of Type 316 Stainless Steel to Li;BeF,, ORNL/TM-5782 (April 1977). M. R. Bennett and A. D. Kelmers, in Molten-Salt Reactor Program Semiannual Progress Report for Period Ending February 29, 1976, ORNL-5132, pp. 27-136 (August 1976). R. F. Apple, in Molten-Salt Reactor Program Semiannual Progress Report for Period Ending February 29, 1976, ORNL-5132, pp. 36—37. D. L. Manning and G. Mamantov, in Molten-Salt Reactor FProgram Semiannual Progress Report for Period Ending Lugust 31, 1976, ORNL- 5078, pp. 4748 (February 1976). D. L. Manning and G. Mamantov, in Molten-Salt Reactor Program Semiannual Progress Report for Period Ending February 25, 1978, ORNL-5132, pp. 3940 (August 1976). D. L. Manning, W. K. Miller, and R. Rowan, Methods of Determination of Uranium Trifluoride, ORNL-1279 (May 25, 1952). 115 S 47. ORNL Master Analytical Manual, Section 2, Radiochemical Methods, TID-7015 (1857}. 48. "Automatic Continuous Analysis of Helium,'" GCR Quarterly Progress Report, December 31, 1959, ORNL-2888, pp. 178—184. 49. A. S. Meyer, personal communication, July 12, 1972. 50. R. B. Gallaher, Operation of the Sampler-Enricher in the Molten- Salt Reactoyr Experiment, ORNL/TM-3524 (October 1971). 51. Molten-Salt Reactor Frogram Semiannual Progress Report, February 28, 1966, ORNL-3936, p. 154. 52. G. Goldberg and L. T. Corbin, in Molten-Salt Reactor Program Semi- annual Progress Report, August 31, 1968, ORNL-4344, pp. 198-196. 53. J. M. Dale, R. F. Apple, and A. S. Meyer, in Molten-Salt Reactor Program Semiannual Progress Report, February 28, 1967, ORNL-4119, p. 158. 54. J. P. Young, in Molten-Salt Reactor Program Semianviual Progress Report, February 28, 1869, ORNL-4396, pp. 202—204. 55. J. M. Dale, R. F. Apple, and A. S. Meyer, in Molten-Salt Reactor Program Semiannual Progress Report, February 28, 1970, ORNL-4548, p. 183. 56. R. Blumberg and T. H. Manney, in Molten-Salt Reactor Program Semi- annual Progress Report, August 31, 1968, ORNL-4344, p. 36. 57. A. Houtzeel and F. F. Dyer, Gamma Spectrometric Studies of Fission Products im the MSRE, ORNL-3151 (August 1972). 58. W. R. Laing et al., in Molten-Salt Reactor Program Semiannual Prog- ress Report, February 28, 1969, ORNL-4396, p. 208. 59. D. L. Manning and J. M. Dale, in Molten-Salt Reactor Program Semi- annual Progress Report, August 31, 1968, ORNL-4344, p. 192. 60. D. L. Manning and G. Mamantov, '"Rapid Scan Voltammetry and Chrono- poetentiometric Studies of Ironm in Molten Fluorides,"” J. Electroanal. Chem. 7, 102—108 (1964). 61. D. L. Manning, F. R. Clayton, and G. Mamantov, in Molten-Salt Reactor Program Semiannual Frogress Report, August 31, 1971, ORNL-4728, p. 75. 62. J. M. Dale and A. S. Meyer, in Molten-Salt Reactor Program Semiannual Progress Report, Februavy 29, 1972, ORNL-4782, p. 77. 116 63. D. L. Manning, in Molten-Salt Reactor FProgram Semiannual Progress e Report, February 28, 1971, ORNL-4676, p. 135. 64. J. M. Dale and A. S. Meyer, in Molten-Salt Reactor Program Semi- annual Progress Report, February 29, 1972, ORNL-4782, p. 79. 65. H. W. Jenkins et al., in Molten-Salt Reactor Program Semiannual Progress Report, February 28, 1969, ORNL-4396, p. 201. 66. M. T. Kelley, R. W. Stelzmer, and D. L. Manning, in Molten-Salt Reactor Program Semiannual Progress Report, August 31, 1969, ORNL- 4449, p. 157. 67. D. L. Manning and A. S. Meyer, in Molten-S5alt Reactor Frogram Semi- annual Progress Report, August 31, 1971, ORNL-4728, p. 74. 68. D. L. Manning and H. R. Bronstein, in Melten-Salt Keactor Program Semtannual Progress Report, February 28, 1970, ORNL-4548, p. 184. 69. D. L. Manning and F. L. Clayton, in Molten-Salt Reactor Program Semiannual Progress Report, August 1, 1970, ORNL-4622, p. 115. 70. J. P. Young and J. C. White, "A High-Temperature Cell Assembly for Spectrophotometric Studies in Molten Fluorides,' Anal. Chem. 31, 1892 (1959). 71. J. P. Young, ''Windowless Spectrophotometric Cell for Use with Cor- rosive Liquids," A4Anal. Chem. 36, 390 (1964). 72. Molten-Salt Reactor Program Semiannual Progress Report, Lugust 31, 1965, ORNL-3872, p. 145, 73. L. M. Toth, J. P. Young, and G. P. Smith, in Molten-Salt Reactor Program Semiannual FProgress Report, August 31, 1968, ORNL-4344, p. 168. 74. J. P. Young, in Molten-Salt Reactor Program Semiannual Progress Report, August 31, 1969, ORNL-4449, p. 16l. 75. J. P. Young, in Molten-Salt Reactor Program Semiarnnual Progress Report, August 31, 1871, ORNL-4728, p. 71. 76. J. P. Young, in Molten-Salt Reactor Program Semiannual Progress Report, February 28, 1867, ORNL-4119, p. 163. 77. J. P. Young, in Molten-Salt Reactor Program Semiannual Progress Report, August 31, 1966, ORNL-4037, p. 193. 78. J. P. Young, in Molten-Salt Reactor Program Semiannual Progress Report, February 28, 1967, ORNL-4119, p. 1l64. . 79. John B. Bates et al., in Molten-Salt Reactor Frogram Semionnual 1 Progress Report, February 28, 1971, ORNL-4676, p. 94. o 80. 81. 82, 83. 84, 86. 87. 88. 89. 990. 91. 92. 93. 94. 95. 96. 117 J. P. Young et al., in Molten-Salt Reactor Program Semiavnual Progress Report, August 31, 1971, ORNL-4728, p. 73. J. P. Young, in Molten-Salt Reactor Program Semiannual Progress Report, February 28, 1571, ORNL-4676, p. 136. J. M. Dale et al., in Molten-Salt Reactor Program Semicnnual Prog- ress Report, August 31, 1968, ORNL-4344, p. 188. J. M. Dale et al., in Molten-Salt Reactor Program Semiannual FProg- reggs Report, August 31, 1968, ORNL-4344, p. 189. C. M., Boyd and A. S. Meyer, in Molten-Salt Reactor Progrom Semi- annual Progress Report, August 31, 1967, ORNL-4191, p. 173. J. C. White, in Molten-Salt Reactor Program Semiannual Progress Report, July 31, 1964, ORNL-3708, p. 328. R. F. Apple and A. S. Meyer, in Molten-Salt Reactor Program Semi- annual Progress Report, February 28, 1969, ORNI-4396, p. 207. R. F. Apple and A. S. Meyer, in Molten-Salt Reactor Program Semi- annual Progress Report, Auguet 31, 1971, ORNL-4728, p. 72. J. M. Dale and A. S. Meyer, in Molten-Salt Reactor Program Semi- annual Progregs FReport, August 31, 1871, ORNL-4728, p. 69. J. M. Dale and T. R. Mueller, in Molten-5alt Reactor Program Semi- annual Progress Report, February 28, 1971, ORNL-4676, p. 138. J. R. DiStefano, in Molten-Salt Reactor Program Semiannual Report for Period Ending August 1, 1975, ORNL-5078, pp. 132—139. J. R. DiStefano, in Molten-Salt Reactor Program Semiannual Report for Period Ending February &9, 1976, ORNL-5132, p. 163. Personal communication, H. E. McCoy, ORNL, to W. R. Grimes, ORNL. A. P. Litman and A. E. Geoldman, Corrcsion Assceiated with Fluorina- tion in the Oak Ridge National Laboratory Fluorvide Volatility Pro- cess, ORKNL-2832 (June 5, 1961). L. Hays, R. Breyne, and W. Seefeldt, "Comparative Tests of L Nickel, D Nickel, Hastelloy B, and INOR-1," Chemical Engineering Divisiown Summayy Report, July, August, September, 1968, ANL-5924, pp. 49-52. M. R. Bennett, in Molten-Salt Reactor Program Semiannual Progress Report for Period Ending August 31, 1971, ORNL-4728, p. 190. M. R. Bennett and A. D. Kelmers, in Molten-Salt Reactor Program Semiannual Progress Report for Period IEnding February 29, 1976, ORNL-5132, p. 170. 97. 98. 99. 100, 101. 103. 104, 107. 108. 109. 110. 111. 118 R. M. Counce, in Molten-Salt Reactor Program Semiannual Progress s Report for Period Ending August 31, 1975, ORNL-5078, p. 157. D. H. Gurinsky et al., "LMFR Materials,'" Chap. 21, p. 743, in Flutd Fuel Reactors, J. A. Lance, H. G. McPherson, and F. Maslam, Eds., Addison-~Weslev, Reading, Mass., 1958, . C. H. Brown, Jr., J. R. Hightower, Jr., and J. A. Klein, Measurement of Mass Transfer Coefficients in a Mechanically Agitated, Nowndis- - persing Contactor Operating with a Molten Mixture of LiF-BeF,-ThF, and Mollen Bismuth, ORNL-5143 {(November 1976). H. C. Savage and J. R. Hightower, Jr., Engineering Tests of Metal Transfer Process for Extraction of Rare-Earth Fission Products from a Molten-Salt Breeder Reactor Fuel Salt, ORNL-5176 (February 1977). H. Shimotake, N. R. Stalica, and J. C. Hesson, "'Corrosion of Re- fractory Metals by Liquid Bismuth, Tin, and Lead at 1000°C," Trans. ANS 10, 141142 (June 1967). J. W. Siefert and A. L. Lower, Jr., "Evaluation of Tantalum, Molyb- denum, and Berylilium for Liquid Bismuth Service," Corrosion 17(10), 475¢—78t (October 1961). J. R. DiStefano and A. J. Moorhead, Development and Construction of a Molybdenum Test Stand, ORNL-4874 (December 1972). B. Fleischer, in Metals and Ceramics Divisiown Awnual Progress Report, June 30, 1870, ORNL-4570, pp. 1034, 0. B. Cavin et al., in Molten-Salt Reactor Program Semiannual Prog- ress Report for Period Ending February 29, 1972, ORNL-4782, p. 198. R. G. Donnelly and G. M. Slaughter, "The Brazing of Graphite," Welding J. 41(5), 46169 (1962). J. P. Hammond and G. M. Slaughter, "Bonding Graphite to Metals with Transition Pieces,' Welding J. 50(1), 33-40 (1%970). W. J. Hallett and T. A. Coultas, Dynamic Corrosion of Graphite by Liquid Bismuth, NAA-SR-188 (Sept. 22, 1952). A. L. Lowe, Jr. (Compiler), Liquid Metal Fuel Reactor FExperiment Graphite Evaluation Program, BAW-1197 (June 1960). R. B. Lindauer, in Molten-Salt Reactor Program Semiannual Progress Report for Period Ending February 28, 1975, ORNL-5047, p. 162. R. B. Lindauer, in Molten-Salt Reactor Program Semiannual Progress Report for Period Ewnding February 29, 1976, ORNL-5132, p. 184. S 112, 113. 114. 115. 116. 117. 118. 119. 120. 123. 122, 123. 124, 125, 126. 127. 119 M. R. Bemnett and A. D. Kelmers, in Molten-Salt Reactor Program Semiannual Progress Report for Period FEnding February 28, 1975, ORNL=-5047, p. 150. M. R. Bennett and A. D. Kelmers, in Molten-Salt Reactor Program Semiarmual Progresg Report for Period Ending February 28, 1976, ORNL-5132, p. 170. R. M. Counce, in Molten-Salt Reactor Program Semiannual Frogress Report for Period Ending February 59, 1576, ORNL-5132, p. 189. C. BH. Brown, Jr., and J. R. Hightower, Jr., in Molten-Salt PReactor Progpram Semiannual Frogress Report for Period Ending February 29, 1978, ORNL-5132, p. 177. C. W. Kee and L. E. NcNeese, MPRR — Multiregion Processing Plant Code, ORNL/TM-4210 (September 1976). W. B. Carr, "Volatility Processing of the ARE Fuel," Chem. ing. Prog. Symp. Ser. 56, 57 (1860). Chemical Technology Division Annual Progress Report, May 31, 1964, ORNL-3627, pp. 29—35. W. H. Carr et al., Molten=Salt Volatility Pilot Plant: Recovery of Enriched Uranivum from Aluminum-Clad Fuel Elements, ORNL-4574 (April 1971). R. B. Lindauer, Processing of the MSRE Flush and Fuel Salts, ORNL/ TM-2578 (August 1969). M. R. Bennett and L. M. Ferris, J. Inorg. Nucl. Chem. 36, 1285 (1574). L. M. Ferris et al., J. Tnorg. Nucl. Chem. 34, 31320 (1972). L. E. McNeese, Engineering Development Studies for Mclten-Salt Breeder Reactor Processing No. 8, ORNL/TM-3258 (May 1972). W. L. Carter and E. L. Nicholson, Design and Cost Study of a Fluorination—-Reductive Extraction-Metal Transfer Process Plant for the MSBE, ORNL/TM-3579 (May 1972). L. E. McNeese, IEngineering Development Studies for Molien-Salt Breeder Reactor Processing No. 6, ORNL/TM-3141 (December 1971). F. A. Doss, W. R. Grimes, and J. H. Shaffer, in Molten-Salt Reactor Program Semiannual Progress Report, August 31, 1970, ORNL-4622, p. 106. Chemical Technology Division innual Progress Report, May 31, 1967, ORNL-4145, pp. 95-97. 128. 129. 131. 132. 133, },_a (% I~ 135, 136, 138. 120 J. S. Watson and L. E. McNeese, Engineering Development Studies for Molten-Salt Breeder Reactor Processing No. 9, CRNL/TM-3259, p. 52. B. A. Hannaford, W. M. Woods, and D. D. Sood, in Molten-Salt Reac- tor Program Se#zanmual Progress Report, Februavy 29, 1972, ORNL- 4782, pp. 227-30. L. E. McNeese, Engineering Development Studies for Molten-Salt Breedeyr Reactor Frocessging No. 5, ORNL/TM-3140 (Cctober 1971). \"- 4 ng Development Studies for Molten-Salt Breeder Reactor Frocessing No. 7, ORNL/TM-3257 (February 1972). L. E. McNeese, Engineer J. S. Watson and L. E. McNeese, "Axial Dispersion in Packed Columns During Countercurrent Flew, Liquids of High Density Difference,' Ind. Eng. Chem. Process Des. Develop. 11, 120-21 (1972). J. S. Watson and H. D. Cochran, "A Simple Method for Estimating the Effect of Axial Backmixing on Countercurrent Column Performance,' ind. Eng. Chem. Process Deg. Develop. 10, 83-85 (1971). L. Brewer, Science 161, 115 (1968). D. M. Moulton et al., in Molten-Salt Reactor Program Semiarmual Progress Report for Period Ending August 31, 1968, ORNL-4449, p. 151. el et al., Molten-Salt Reactors for Efficient Nuclear Fuel on Without Plutonium Separation, ORNL/TM-6413 (August 1978). R. G. Ross, C. E. Bamberger, and C. F. Baes, Jr. in Mcolten-Salt Reaclor Program Semiannucl FProgress Report, August 31, 1970, ORNL- 4622, p. 92, M. J. Bell, D. D. Sood, and L. E. McNeese, in Molten-Salt Reactor Program Semiannual Progress Report, February 29, 1972, ORNL-4782, PpP. 234—7 =3 PART IV. REACTOR MATERIALS H. E. McCoy The material for the primary circult will be exposed at temperatures up to 700°C to fuel salt containing fission products and to irradiation by primarily thermal neutrons. A nickel-base alloy, Hastelloy N, has been demonstrated to be reasonably serviceable under these conditions, but it was embrittled by irradiation and suffered shallow intergranular embrittlement by the fission product tellurium. There is considerable experimental evidence that small modifications to the chemical composi- tion of Hastelloy N results in improved resistance to neutron and fis- sion-product embrittlement, and the materials program described in this plan is directed toward developing and commercializing a modified com- positicn of Hastelloy N with impreved properties. The graphite moderator for an MSR must be capable of withstanding neutron fluences of at least 3 X 10%? neutrons/cm®. Commercial graphites exist which are likely to meet this goal, but further testing will be re- guired to fully characterize these materials. There is also considerable evidence that graphite with improved dimensional stability can be de- veleped. Methods for manufacturing these impreoved materials must be developed and the products irradiated and characterized. The improved materials must be scaled up by a commercial wvendor. 123 7. STRUCTURAL METAL FOR PRIMARY AND SECONDARY CIRCUITS The material used in constructing the primary circuit of an MSR will operate at temperatures up to 700°C. The inside of the circuit will be exposed to salt containing fission preducts and will receive a maximum thermal fluence of about 1 X 10%1 neutrons/cm2 over the operat- ing lifetime of about 30 years. This fluence will cause embrittlement due to helium formed by transmutation but will not cause swelling such as is noted at higher fast fluences, The outside of the primary circuit will be exposed to nitrogen containing sufficient gir from inleakage to make it oxidizing to the metal. Thus the metal must have moderate oxi- dation resistance, must resist corrosion by the salt, and must not be subject to severe embrittlement by thermal neutrons. In the secondary circuit the metal will be exposed to the coclant salt under much the same conditions described for the primary circuit. The main difference will be the absence of fission preoducts and uranium in the coolant salt and the much lower neutron fluences. This material must have moderate oxidation resistance and must resist cerrosion by a salt not ccntaining fission products or uranium. | The primary and secondary circuits involve numerous structural shapes ranging from a few inches thick to tubing having wall thicknesses of only a few thousandths of an inch. These shapes must be fabricated and joined, primarily by welding, into an integral engineering structure. The struc- ture must be designed and built by techniques approved by the ASME Boiler and Pressure Vessel Code. Status in 1972 Early materials studies led to the development of a nickel-base alloy, Hastelloy N, for use with fluoride salts. As shown in Table 7.1, the alloy contained 167 molybdenum for strengthening and chromium suf- ficient to impart moderate oxidation resistance in air but not enough to lead to high corrosion rates in salt. This alloy was the sole structural material used in the MSRE and contributed significantly to the success of the experiment. However, two problems were noted with Hastelloy N 124 Table 7.1. Chemical composition of Hastelloy N Content (% by weight)a Element e o Standard alloy Medlfiggzalloy, Mod1f13§6alloy, Nickel Basge Base Base Molybdenum 15~18 11-13 11-13 Chromium 6—8 6—8b 6—8b Iron 5 .1 b 0.1 5 Manganese 1 ¢.15-0.25 0.15-0.25 Siiicon 1 C.1 0.1 Phosphorus 3.015 C.C1 0.01 Sulfur 0.020 C.0L 0.0L Boron 0.01 C.C01 0.00L Titanium 2 Niobium 02 1—2 & ) . Single values are maximum amcunts allowed. The actual concentrations of these elements in an alioy can be much lower. These elements are not felt to be very important. Alloys are now being purchased with the small concentrations specified, but the specification may be changed in the future tc allow a higher concentration. which needed further attention before more advanced reactors could be built. First, it was found that Hastelloy N was embrittled by helium produced from '°B and directly from nickel by a twc-step reaction. This type of radiation embrittlement is commen to most iron- and nickel-base alloys. The second problem arose from the fission-product tellurium dif- fusing a short distance into the metal along the grain boundaries and embrittling the boundaries. When cur studies were terminated in early 1973, considerable prog- ress had been made in finding sclutions to both problems. Since the two problems were discovered a few years apart, the research on the two prob- lem areas appears to have proceeded independently. However, the work must be breught together for the production of a single material that would be resistant to both problems. It was found that the carbide pre- . cipitate that normally occcurs in Hastelloy N could be modified to obtain resistance to the embrittlement by helium. The presence of 167 molybde- num and 0.5% silicon led to the formation of a8 coarse carbide that was 125 of little benefit. Reduction of the molybdenum concentration teo 127 and the silicon content to 0.17 and the addition of a reactive carbide former such as titanium led to the formation of a fine carbide precipitate and an alloy with good resistance to embrittlement by helium. The desired level of titanium was about 2%, and the phenomenon had been checked out through numercus small laboratory and commercial melts by 1972. Because the intergranular embrittlement of Hastellcy N by tellurium was noted in 1970, our understanding of the phenomenon was not wvery ad- vanced at the conclusion of the program in 1973. Numerous parts of the MSRE were examined, and all surfaces exposed to fuel salt formed shallow intergranular cracks when strained. Some laboratory experiments had been performed in which Hastelloy N specimens had been exposed to low partial pressures of tellurium metal wvapor and, when strained, formed intergranular cracks very similar to those noted in parts from the MSRE. Several findings indicated that tellurium was the likely cause of the intergranular embrittlement, and the selective diffusion of tellurium along the grain boundaries of Hastelloy N was demonstrated experimentally. One in-reactor fuel capsule was operated in which the grain boundaries of Hastelloy N were embrittled and those of Inconel 601 (Ni, 227% Cr, 12% Fe) were not. These findings were in agreement with laboratory experiments in which these same metals were exposed to low partial pressures of tel- lurium metal vapor. Thus, at the close of the program in early 1973, tellurium had been identified as the likely cause of the intergranular embrittlement, and several laboratory and in-reactor methods were de- vised for studying the phenomenon. Experimental results had been ob- tained which showed variations in sensitivity to embrittlement of various metals and offered encouragement that a structural material could be found which resisted embrittlement by tellurium. The alloy composition favored at the close of the program in 1973 is given in Table 7.1 with the composition of standard Hastelloy N. The reasoning at that time was that the 27 titanium addition would impart good resistance to irradiation embrittlement and that the 0 to 27% nio- bium addition would impart good resistance to intergranular tellurium embrittlement. Neither of these chemical additions was expected to cause problems with respect to fabrication. 126 Status in 1976 ,\", When the program was restarted in 1974, top priority was given to the tellurium-embrittlement problem. A small piece of Hastelloy N foil from the MSRE had been preserved for further study. The foil was broken inside an Auger spectrometer and the fresh surface analyzed. Tellurium was found in abundance, and no other fission product was present in de- tectable quantities. This showed even more positively that tellurium was responsible for the embrittlement. Considerable effort was spent in seeking better methods of exposing test specimens to tellurium. In the MSRE the flux of the tellurium atoms reaching the metal was 10° atoms cm™? sec™ and this value would be 10!° atoms cm™° sec”! for a high-performance breeder. Even the value for a high-performance breeder is very small from the experimental standpoint. For example, this flux would result in a total of 7.6 X% 10”8 g of tellurium transferred to a sample having a surface area of 10 em? in 1000 hry. Electrochemical probes were immersed directly in salt melts known to contain tellurium, and there was never any evidence of a soluble telluride species. However, there was considerable evidence that tel- lurium "moved" through salt from cne point to another in a salt system. - It was hypothesized that the tellurium actually moved as a low-pressure, pure-metal vapor and not as a reacted species. The most representative experimental system developed for exposing metal specimens to tellurium involved suspending the specimens in a stirred vessel of salt with gran- ules of CriTey and CrsTeg lying on the bottom of the salt. Tellurium, at a very low partial pressure, was in equilibrium with the CriTe, and CrsTeg, and exposure of Hastelloy N specimens to this mixture resulted in crack severities similar to those noted in samples from the MSRE, Numerous samples were exposed to salt containing tellurium, and the most important finding was that modified Hastelloy N containing 1 to 2% nicbium had good resistance to embrittlement by tellurium (Fig. 7.1). An almost equally important finding was that the presence of ti- tanium negated the beneficial effects of niobium. Thus, an alloy con- taining titanium, to impart resistance to irradiation embrittlement, and niobium, to impart resistance to tellurium embrittlement, did not have 127 ORNL-DWG 77-922 I Y \«’? 8000 |— 8926 - , 2500 hr / 7000 +— / ] 6000 5000 CRACK DEPTH fem) 4C00 CRACK FREQUENCY x { number /cm) o O O O 2000 1000 Nb CONTENT (%) Fig. 7.1. Variations of severity of cracking with niobium content. Samples were exposed for indicated times to salt containing Cr3Te, and CrsTeg at 700°C. 128 acceptable resistance to tellurium embrittlement, even though the mech- e anical properties in the irradiated condition were excellent. As a re- sult, it became necessary to determine whether alloys containing niobium (without titanium) had adequate resistance to irradiation embrittlement. There was time only to obtain the alloys and run one irradiation experi- ment, but the results loocked very promising. An alloy containing 2% . niobium and irradiated at 704°C was about 30% stronger than standard Hastelley N and had a fracture strain of about 3% compared with <1% for standard Hastelloy N. Even though alloys modified sclely with nicbium do not have as good postirradiation properties as alloys modified with titanium or titanium plus niobium, their properties are probably adequate. The niobium-modified alloys were not made in melts larger than 50 1b, but no problems were encountered in this size with niobium concen- trations up to and including 4.4%. Test welds made in the 1/2-in.-thick plate passed the bend and tensile tests required by the ASME Boiler and Pressure Vessel Code. From the chemical analysis of the niobium-modified alloy, no scaleup problems are anticipated. One series of experiments was carried out to investigate the ef- fects of oxidation state on the tendency for cracks to be formed in tellurium-containing salt, on the supposition that the salt might be made reducing enough to tie the teilurium up in some innocuous metal complex. The salt was made more oxidizing by adding NiF,; and more re- ducing by adding beryliium. The experiment had electrochemical probes for determining the ratio of uranium in the +4 state (UF4) to that in the +3 state (UF3). Tensile specimens of standard Hastelloy K were suspended in the salt for about 260 hr at 700°C. The oxidation state of the salt was stabilized, and the specimens were inserted so that each set of specimens was exposed to one condition. After exposure, the spec- imens were strained to failure and were examined metallographically to determine the extent of cracking. The results of measurements at several oxidation states are shown in Fig. 7.2. At U*T/U3* ratios of 60 or less, there was very little cracking, and at ratios above 80 the cracking was very extensive. These observations offer encouragement that a reactor could be operated in a chemical regime where the tellurium would not be embrittling even to standard Hastelloy N. At least 1.6% of the uranium # % 129 ORNL-DWG 77-4680A I I ! I [ r 900 - RECUCING OXIDIZING J e —_— : /f”"“fi‘°—4 CRACKING 600 |- : PARAMETER _ [FREQUENCY (cm™%) X AVG. DEPTH (um)] — fi 300 - @ 0 "-9—'—""“‘"1—"‘“‘7/ Lo | o i0 20 40 70 100 200 400 SALT OXIDATION POTENTIAL fU(IVI/U(IID)] Fig. 7.2. Cracking behavior of Hastelloy N exposed 260 hr at 700°C to MSBR fuel salt containing CrsTe, and CrsTeg. would need to be in the +3 oxidation state (UF:), and this condition seems quite reasonable from chemical and practical considerations. One further accomplishment during the period 1974—76 was the use of available data to predict the helium yvield from interaction of nickel with thermal neutrons. It has been known for some time that iron- and nickel-base alloys can be embrittled in a thermal neutren flux by the transmutations of "tramp" *°B to helium and lithium. This process gen- erally results in the transmutation of most of the 10g by fluences of thermal neutrons on the order of 10°%°/cm® and usually yields from 1 to 10 at. ppm of helium. With nickel there is a further thermal two-step transmutation involving these reactions: 5 * SNi 4+ m > PINi Ige t “9Ni + n > “He + °%Fe . This sequence of reactions does not saturate, and although the cross sections are still in question, it would produce a maximum of 40 at. ppm of helium in the vessel over a 30-year MSBR lifetime. This is not an unreasonable amount of helium to accommodate in the type of microstruc- ture being developed in modified Hastelloy N. 130 Current Status At the close of the program in 1976 (and at the present time), the third alloy composition shown in Table 7.1 was favored. Considerable progress had been made in establishing test methods for evaluating a ma- terial's resistance to embrittlement by tellurium. Modified Hastelloy N containing from 1 to 37 niobium was found to offer improved resistance to embrittlement by tellurium, but the test conditions were not sufficiently long or diversified to show that the alloy totally resists embrittlement. One irradiation experiment showed that the niobium-modified ailoy offered adequate resistance to irradiation embrittlement, but more detailed tests are needed. Several small melts containing up to 4.4% niobium were found to fabricate and weld well; so products containing 1 to 2% niobium can probably be produced with a minimum of scale-up difficulties. Technology Needs and Development Plan The coverall development needs were described previously, but the new findings shift the emphasis from alloys modified with titanium and rare earths to those modified with niobium. The specific techneclogy needs are identified in Table 7.2, along with a potential schedule for their development. The first task will involve irradiation, corrosion, tellurium exposure, mechanical property, and fabrication tests to final- ize the composition for scale-up. The techniques for doing most of these tests have already been established. The second task will inveclve procuring large commercial heats of the reference alloy. The material would be procured in structural shapes ranging from plate to thin-wall tubing, typical of the products to be used in a reactor. The third task consists in evaluating these materi- als by mechanical property and cerrosion tests of at least 10,000-hr duration. The two main purpeses of these tests would be to confirm the adequacy of the new alloy for reactor applications and to gather the data needed for reactor design. The fourth task would be to develop the design methods and rules needed to design a reactor tc be built of the modified Hastelloy N. This task will have already been partially completed by ASME Pressure Vessel Code work currently in progress. The Table 7.2. Schedule for development of structural metal for primary and secondary circuits Fiscal vyear Task 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 Determination of alloy compesition ..iECEfiE[EiEZ5 67 Procurement of commercial heats V'V Evaluation of commercial heats ve vs v10 v” Development of analytical design 12V ‘{713 V14 methods — ASME Code vflS Long-term material tests 16 Alloy optimization v Milestones: 1. Receipt of small commercial heats containing 1 to 8. Begin comstruction and checkout of equipment re- 2% Nb. Begin mechanical property and compatibil- quired for mechanical property tests on four large ity tests on heats. heats. 2. Receipt of products of 10,000-1b heat of 2% Nb- 9. Begin evaluation of four large heats by weldability, modified llastelloy N. Begin mechanical property mechanical property, and compatibility tests. and compatibility tests on 10,000-1b heat. 10. Begin operation of forced-circulation loops (FCL-6 3. Start forced-convection corrosion loop constructed and 7) constructed of modified alloy and circulat- of 10,000~1b heat for basic fuel salt corrosion ing fuel salt. ‘ studies. Begin “Vl-vear irradiation of fuel 11. Begin operation of forced-circulation loops (FCL-8 Pins made of most desirable alloy. and 9) constructed of modified alloy and circulating 4. Start forced-convection corrosion loop constructed coolant salt. of 10,000~1b heat for fuel salt-Te corrosion stud- 12. Begin detailed analysis of mechanical property data. ies. 13. Begin development of design methods for modified 5. Start forced-convection corrosion loop (FCL-5) con- alloy. structed of 10,000-1b heat for coolant salt corrosion 14. Submit data package for ASME Code Approval. studies. 15. Begin studies tco raise allowable temperature for use 6. Prepare specifications and solicit bids from poten- of modified alloy. tial vendors for four heats of desired composition. 16. Begin long-term mechanical property and compatibil- 7. Begin receipt of products from four large heats. ity tests on modified alloy. 1¢T 132 final product of this task would be inclusion of modified Hastelloy N into the high-temperature Code. Although the data gathered in the third task (tests of 10,000-hr duration) will probably be adequate for Code approval, it will be de- sirable to continue some of the mechanical property and corrosion tests for longer times. The continuation of these tests in the fifth task will & improve confidence in design rules and will allew last-minute changes in reactor operating parameters if necessary. Although the work in the first five tasks should result in an al- loy adequate for construction of MSRs, it is likely that further alloy development would lead to materials having improved characteristics which may allow a higher reactor-outlet salt temperature or significant relaxation of design and operating constraints. It is this further al- loy optimization which will comprise the sixth task. The operating and capital costs for these activities are summarized in Tables 7.3 and 7.4, respectively. Table 7.3. Operating fund requirements for development of structural metal for primary and secondary circuits Cost (thousands of 1978 dollars) for fiscal year — Fask 1980 1981 1982 1983 1984 1885 1986 1987 1988 1989 1990 1691 Determination of alloy composition 2200 2200 Procurement of commercial heats 520 Evaluation of commercial heats €00 2225 2660 Development of analytical design 280 930 220 methods — ASME Code Case Sub- mission Long=term material tests 1300 1300 1040 1040 910 2910 650 400 Alloy optimization 390 455 572 520 624 650 676 400 Total funds® 2200 2800 3025 3590 1910 1755 1612 1560 1534 1560 1326 800 aTotal funds through 1991: $23,672. tetl Table 7.4. Summary of capital equipment funds required for development of structural metal for primary and secondary circuits Cost {(thousands of 1978 dollars) for fiscal year -- Task i . S 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 Determination of allovy composition 955 732 Procurement of commercial heats 13 Evaluation of commercial heats 438 1424 377 Development of analytical design 65 130 methods — Code Case Submission Long-term material tests 52 78 46 72 39 52 39 40 Alloy optimization 46 91 104 104 98 98 98 40 Total funds™ 955 1170 1502 507 98 169 150 176 137 150 137 80 “Total funds through 1991: $523L. el 135 8. GRAPHITE FOR MOLTEN-SALT REACTORS The graphite in a single-fluid MSR serves no structural purpose® cther than to define the flow patterns of the salt and, of course, to ! support its own weight. The requirements on the material are dictated most strongly by nuclear considerations, namely stability of the mate- rial against radiation-induced distortion and nonpenetrability by the fuel-bearing molten salt. The practical limitations of meeting these requirements, in turn, impose conditions on the core design, specifi- cally the necessity to limit the cross-~sectional area of the graphite prisms. The requirements of purity and impermeability te salt are easily met by several high-quality, fine-grained graphites, and the main problems arise from the requirement of stability against radiation- induced distortion. Status in 1972 By the time the MSBR Program was cancelled in early 1973, the di- mensional changes of graphite during irradiation had been studied for a number of years. These changes depend largely on the degree of crys- talline isotropy, but the volume changes fall into a rather consistent pattern. As shown in Fig. 8.1, there is first a period of densification during which the volume decreases and then a period of swelling in which the volume increases. The first period is of concern only because of the dimensional changes that occur, and the second period is of concern be- cause of the dimensional changes and the formation of cracks. The forma- tion of cracks would eventually allow salt to penetrate the graphite, The data shown in Fig. 8.1 are for 715°C, and the damage rate increases with increasing temperature. Thus the graphite section size should be kept small enough to prevent temperatures in the graphite from exceeding those in the salt by a wide margin. In the breeder concept the neutron flux is sufficiently high in the [ central region of the core to require that the graphite be replaced about every four years. It was further required that the graphite be surface afa 5 Its primary function is, of course, to provide neutron moderation. 136 ORNL-DWG 71-6910R g o O VOLUME CHANGE, 100 In (1+AV/V) | N -8 . j | N ] 0 10 20 (x102% FLUENCE [neutrons/cm? (£ >50 keV)] Fig. 8.1. Volume changes for conventional graphites irradiated at 715°C. sealed to prevent penetraticn of xenon intoc the graphite. Since re- placement of the graphite would require considerable downtime, there was strong incentive to increase the fluence limit of the graphite. A considerable part of the ORNL graphite program was spent in irradiating commercial graphites and samples of special graphites with potentially improved irradiation resistance. The approach taken te sealing the graphite was surface sealing with pyrocarbon. Because of the neutronic 137 requirements, other substances could not be introduced in sufficient quantity to seal the surface. The irradiation studies with several grades of graphite revealed that the so-called binderless graphites, e.g., POCCO AXF, had improved dimensional stability over most of the conventional graphites (Fig. 8.2). The POCC graphites are presently available only in small sections, but the GLCC B-364 grade is available in large sectiomns. The GLCC H-364 grade has almost as high an allowable fluence as POCO AXF. Further work on several special grades of graphite made at ORNL showed that graphites could be developed with fluence limits even greater than those of the POCO grades. | The pyrolytic sealing work was only partially successful. It was found that extreme care had tc be taken to seal the material before ir- radiation. During irradiation the injected pyrocarbon actually caused expansion to begin at lower fluences than those at which it would occur in the absence of the coating. Thus the coating task was faced with a number of challenges. ORNL-DWG 71-6915R2 T “J © 4 ; A B} N ; 3 : ,,//H—364 , 1 5 o d -~ H A / £ 0 _5’3 ; Y - i Ry S A-A—&:—’ I a AXF-UFG ] o AXM A P-03 4 ! o HL 18 : i + H-395 | s -8 | i 0 10 20 30 (x1c2h FLUENCE [neutrons/tm?2 (£ >50 keV)] Fig. 8.2. Volume changes for monolithic graphite irradiated at 715°C. 138 Status in 1976 R No work was undertaken on graphite during the last segment of the program. Thus the status in 1976 was the same as that in 1972. Current Status With the relaxed requirements® for breeding performance in nonpro- liferating MSRs relative to the MSBR, the requirements for the graphite have diminished. First, the peak neutron flux in the core can be re- duced to levels such that the graphite will last for the lifetime of the reactor plant. Secondly, the salt flow rate through the core is reduced from the turbulent regime, and the salt film at the graphite surface may offer sufficient registance to xenon diffusicon sc that it will net be necessary to seal the graphite. The lessened gas permea- bility requirements also mean that the graphite damage limits can be raised (Figs. 8.1 and 8.2). The lifetime criterion adopted for the breeder was that the allowable fluence would be about 3 % 10?% neutrons/ em’. This was estimated to be the fluence at which the structure in advanced graphites would contain sufficient cracks to be permeable to g xenon. Experience has shown that even at volume changes of about 10% the graphite is not cracked but is uniformly dilated. For nomprolifer- ating devices where xenon permeability will not be of concern, the limit will be established by the formation of cracks sufficiently large for salt intrusion. It is likely that current technology graphites like GLCC+ E-364 could be used to 3 X 10°? neutrons/cm® and that improved graphites with a limit of 4 X 10°?? neutrons/cm? could be developed. Further Technology Needs and Development Plan The near-term goal of the future development program (see Table 8.1) will be to evaluate current commercial graphites for MSR use (Task 1). % These are manifested as lower core power density and higher fissile specific inventory in denatured MSRs. s Great Lakes Carbon Company. & Table 8.1. Schedule for graphite development Fiscal vear Task 1980 1981 1982 1983 1984 1685 1986 1987 1988 1989 1990 1991 1992 1993 1994 . . 1 2 Evaluation of commercial v v graphites . 3 Development of improved graphites Procurement of commercial §74 lots of improved graphites Evaluation of commercial 75 lots of improved graphites Milestones: Establish program for development of improved graphites. Define variables to be investigated. Complete evaluation of commercial graphites. document specification. Develop procurement specification for improved commercial graphites. Prepare 4, Begin procurement of production lots of improved commer- cial graphites. Begin long-term evaluation of improved commercial graph- ites. Evaluation to include mechanical and physical properties before and after irradiation. 6¢cT This will involve irradiation of promising commercial graphites with subsequent measurements of dimensional stability and thermal and elec- trical conductivity. A longer—range goal will be the development of a graphite with an improved fluence I1imit. Efforts to date show that graphites can be tailored to have improved dimensional stability. In Task 2 this work will be continued te obtain several improved products, which will be irradiated and evaluated. The technology for making the most desirable products will be passed on to commercial vendors, and large lots of these graphites will be obtained (Task 3). The commercial graphites will be irradiated to high fluences, and the changes in dimensions, pore spectra, thermal conductivity, thermal expansion, and electrical conduc~ tivity will be measured (Task 4). The operating and capital equipment costs fer this work are sum- marized in Tables 8.2 and 8.3, respectively. 3 Table 8.2. Operating fund requirements for graphite development Cost (in thousands of 1978 dollars) for fiscal vear — Task - 1980 1981 1982 1983 1984 1985 1986 1987 1983 1989 1990 1991 1992 1993 1994 Evaluation of c¢ommercial 300 300 300 300 300 graphites Development of improved 159 300 300 500 500 150 graphites Procurement of commercial 100 200 150 lots of improved graphite Evaluation of commercial 300 500 500 400 400 300 300 300 lots of improved graphites Total funds® 300 306 450 600 600 500 600 650 550 500 400 400 300 300 “Total funds through 1994: $6750. 300 7t Table 8.3.