b e e 1 S ot ‘ ‘d“:.; b : L / v T B 1 "fiii * OAK RIDGE- NA'I'IONAI. I.ABORATORY - . - operated by , umon CARBIDE CORPORATION ~ NUCLEAR DIVISION ' e o e u 's__,_,Aromc ENERGY commrssnon : - ,;ORN!. TM- 1858 | ififi. LT P }copvuo. ;ffDATE- June 9, 1967: (fifiSII REICES SAFETY PROGRAM FOR MOLTEN-SALT BREEDER REACTORS - xmsRacT Investigations required ‘in determining the safety characteristics - of MSBR power plants are outlined, and associated safety program cost - é,festimates are given. ~ in the MSBR are described; the favorable characteristics arise from the ri;;;prompt negative temperature coefficient of reactivity, the low system - pressures, the mobility of- fluid fuel, and the low excess reactiv1ty ‘available to the reactor at any time.- The safety: features of the major plant systems ‘Reactivity additions which need itfifdetailed study include those assoclated with net fuel addition to the »lifcore region, those due to graphite behavior, those caused by changes " ’in fluid flow conditions; and those due to control rod movement. Re- Jlrjactivity coefficients which: require evaluation include those assoc1ated :*7:wa1th temperature, voids, pressure, fuel concentration, and graphite con- flr;ffcentration. - conditions and also under circumstances where reactivity itself is not ‘The integrity of plant containment under’ react1v1ty inc1dent - - involved need to be evaluated; included here are evénts such as mixing - of water and steam with. coolant salt criticality in regions outside the -.core, and flow blockage Within the- fuel or .coolant streamsi_ Stability . .';siffanalysis of the reactor plant is required to determine the operating, . ...control, and/or design requirements for Obtaining satisfactory plant 'fi'f"characteristics Phy51cal behavior of materials and of equipment under . _MSBR conditions, as _they relate to reactor safety, need to be determined h_l;experimentally. In order to dellneate and resolve the basic safety prob- lems associated.w1th MSBR- systems, it is ‘estimated that ébout $1.3 million i required over = peried of ‘about eight years, with most of the effort :i“;q;($0 9 million) occurring during the first four years._f'f“; | - “UT":E Thrs documem confcms rnforrnahon of a prehmmory nalure and was prepcred primerily for internal use at the Oak Ridge National - _ Laboratory ‘It is subject 1o revision or correction and therefore does not represent a fmo! reporf T _ . THIS DOCUMENT HAS BEEN REV]EWED NO INVENTIONS OF BATEHT IHTL.RL'ST Ignnann“ ncnmmmdscsnmsaucuusns;7'"~ N » ~LEGAL NOTICE This report was proparod os an account of Government sponsored work. Neithor the United Shhs, ' nor the Commission, nor ony person acting on behalf of the Commission: A. Mckes any warranty or representation, expressed or implied, with respect te the accuracy, completeness, or usefulness of the information contained in this report, or that the use of - .any information, apparotus, mefhod or process ‘disclosad in this report may net lnfringo privately owned rights; or B. Assumes any licbilities with respect to fho use of, or for damages usulflng from the use 0‘-,-." any information, apparatus, method, or process disclosed in this report. As used inthe above, * person acting on behalf of the Commission® includes any employee or controctor of the Commission, or employee of such contractor, 1o the axtent that such employes - or contractor of the Commission, or employee of such contractor prepares, disseminates, or provides access to, any information pursuant fo his employment or contract with the Cemmnulon, o or hls -mploymom with such contractor, TABLE OF CONTENTS Page | 1. INTRODUCTION . « & v v ¢ v 4 o 4 o o v o o o o v v o o o & 3 | 2. MAJOR PLANT SYSTEMS INFLUENCING REACTOR SAFETY . . . ... 6 | 2.1 Reactor System . . . ¢ & . 4 4 ¢ v 4 4 e 4 4 e e e T 2.2 Steam System . . & ¢ v ¢ ¢ ¢ v ¢ e v 4 e e e e e .. 10 2.3 Fuel Recycle Processing System c 4 s e s s e s e s 13 2.4 Off-Gas System . . . v v v 4 4 4 e e e e e e .. 15 3. REACTOR SAFETY ASPECTS . + & « ¢ + & « « o « o o« « « » « « 18 3.1 Reactivity Coefficients and Kinetics Parsmeters . . . 18 . . . O 5 02 ContI‘Ol _ROd Func tiOIl . . * o‘ . . . . . . . . . 2 7 oA | ce .. o2l whe 5.3 Reactor Incidents . . . . . . . . ¢« « s ¢ o . ‘;‘ . . » » . - . 2 5 . 3 . l Reactiv1‘ty‘ Additlons - . . . . . » - . 3 3.3.2 Mechanical and Physical Integrity - 29 C on-tainment ® & & & s e .8 a4 @ ® e & e 9 o e t . M . . . . 5 1 5.3.3 Miscellaneous Incidents . . . . . ’ - - . . e . . * 5 2 3 . LI' Reactor Stability . '.o ., . .. e 9 . e e . _ . 7 - | : N » o ’ e » 8 a # e s s o *. = ’ . . 5 5 ,'l'l MSBR SAFETY PmeM . . L] » . - . - . | 7 ' ) ) - * o o+ e w s s e s 33 b1 Summary . . ... ... St | . | o e S - L .- L. 3 & - 7 ) . ) . o o . ¢ ¢ 0 e e o . 0 l' e 3 8 ACKNOWLEDGMENTS .« « v v v v o v v o . Trrrrrr . LEGAL NOTICE | This Teport was prepared as an sccount of Government sponsored work, Neither the United Btates, nor the Commission, nor ARy person acting on behaif of the Commission . A, Makes any Warranty or representation, expregsed or iraplied, with respect to the accu- TAcy, completeness, or usefulness of the information contained in this report, or that the use -of any mformation, Apparatus,, method, or process disclosud in this Teport may not infringe | . privately owned rights; or - : . - i B. Assumes any labliities with Tespect 1o the use of, or for damages resulting from the : : Y , ; use of any information, apparatus, method, or proceas disclosed in thia report, i ; : . b _ As used in the shove, *‘person acting on behaf of the Commisaion** fncludes any em-~ - .k ;‘ployoe or contractor of the Commissio ' n, or employee of such contractor, to the extent that i: such employes or contractor of the Commission, or employes of such contractor prepares, - disseminates, or provides access to, any fnformation pursuant 1o his employment or contract with the Commisaton, or his employment with guch contractor, ; - e OF THS DOCUHERT, 15 ONOMTED DISTRIEUTION OF THIS O ™. «) ( ,'y - » ¥ SAFETY PROGRAM FOR MOLTEN-~SALT BREEDER REACTORS Paul R. Kasten I. INTRODUCTION The purpose'of this report is to discuss important aspects of molten-salt breeder reactor-plants which are related to the opera- tional and ultimate safety of such systems, and to present a program for investigating reactor characteristics and associated cost require- ments. In order to be relatively specific, the Molten Salt Breeder Reactor plant (MSBR) described in Ref. 1 forms the basis for this dis- cussion. However, general studies which also consider other design concepts will need to be’performed' the general studies required will come into better focus as MSER safety and design information is _developed. Briefly, the MSBR design concept concerns a two-region, two-fluid system with fuel salt separated from the blanket salt by graphite tubes. Circulating-fuel temperatures are high (~1300°F), and reactor pressures are low (~100 psi)r The energy produced in the reactor fluid is trans- ferred to a secondary coolant§salt-circuit, which couples the reactor to a supercritical steam cycle. The fuel salt consists of uranium fluoride dissolved in a carrier salt containing a mixture of lithium and beryllium fluorides, while the blanket salt contains thorium fluo- ride dissolved in a similar carrier salt. The blanket salt also cir- culates through passages in_the;graphite moderator region of the core. "The coolant salt is‘a mixture'of sodium fluoride and'sodium fluoroborate. Fuel processing is performed: on-site, in a processing plant integral -with the reactor plant - Figure 1 gives a flowsheet of the 1000-Mw(e) MSBR power plant, ‘while. Figure 2 gives the associated processing flow- sheet. Details of these flowsheets are discussed in References 1l and 2. The safety of MSBR's has not as yet been investigated in detail -[however, it can be .discussed in e qualitative manner, p01nting out areas o and 1tems,which need to be investigated. The operating philosophy and ORNL-Dwg. 67-6313 REACTOR VESSEL o S48 - | _ - f . 100874 _ j_usu | | 1518.5h-540p -1000° | 1918 58 ! r i H24h-35|5p-|000‘ i B | 3600p- 000°*F | 24 E . HP GEN. H 150 1%s0¢ {2011 Ysee i N } tureine [1'° TURBINE 527.2 Mwse Fl95.7 1teee o BOILER | | oo Gross R0\ _100°F { | SUPERHEATERS | seinl 4 __, 100 g | ¢ 4 Jeso°F oot r | BLANKET SALT HEAT COOLA ! 85I7° L | EXCH. AND PUMPS | PU:LSSA'-T | P 3 GEN. ‘ FUEL SALY HEAT o TURBINE [ TURBINE 5°Té'r’m“" H25*F EXCH. AND PUMPS 13092} E | | | _ 1125 F ' o ‘ beior REMEAT STEAM | | AL PRENEATERS | (ot ¢ | i T CONDENSER 8 FEEDWATER o | 300p-T00°F| | poren| _ SYSTEMS I | J | -" | . 1 I 759.2'! | 5 ' '. 35009"55'F | 3500 . T Y "= o3ysy ! _,\__'____ _l - 2H] el - e ——- e T66Ah L——_—*"# | - SR ] | | __ Pumes S e - | (- FUEL SALT (COUNT SR . DoREoRwawce : tboeny \—ODAIN TANKS ~ \_DRAIN-TANKS | CONET OUTRUT - 1000 Mws LEGEND R | S , ~ GROSS GENERATION 10349 v CORUEL — S " ‘ . BF BOOSTER PUMPS - 92 Mwe . BUANKET mmaem - o - STATION AUXILIARIES 257 Mwe =~ " COOLANT =—eoom U DR : - .- REACTOR HEAT INPUT 2.225 ¥t | smu ,.A.'.....\.'_;:;.-g_-. AT o R o - NET HEAT RATE 7601 Bte/kwh Mo T o o " NET EFFICIENCY ML Y SR S | Pareemaeftit : eeeeeeBlu/IN O ...Fraeze Valre | , | 'Fig. 1. Flowsheet of a 1000-Mw(e) MSBR Power Plant. . ] - l (- \) ‘ © d’ > ‘ ) ¢ il > ‘e . . ! . g > . - . . . - ORNL-DWG 66-TeE8R UFg RECYCLE TO REACTOR . + ' o VAN NS AN S | - mNggof % fi sz//% )y : ] . | 4 ///4/////5_% 1 S o T I | NPT _ SN /s‘roRAGea/;, R : , R T Vel excess 7] : o NaF /MgFa/FP . . : . . S 7 ; o : F PRODUCTIO : DRt I"///gt_z//;fi.f T | oy VWASTE‘ / . V////Z S o YOLATEFP | B Narmraee ] | . MAKEUP . g S : o/ S | CLiF/BeFy/ThRg -~ - ] ':‘ sl o ‘ ‘ UFg + , ] . o : V°'rf“'_"—£ FP | MAKE UP ‘ , L | ' - | LiF/BeF; ' v e L o : : ‘ o o - ‘ Y v O e R e T ‘ FOR ./ FLUORIDE vacuuM 77| DISTILLATE —~UF, 7 [, g% ”,‘“55,”//, /| voLaniLiTy /] €ore /) sPENT _['Fe oEcaY A~ voamiLimy 47/ isticLation ] , LiFsseF, A/ Z coucrion | FILTRATION 4 T NS A LiF/BeR/URJ° FEL Vil e st /1o Assosc 7/ | ~000°C |/ | /~800C /| ), 550-600°C /17 A | % %/ ,,550.0,/% /////// [405 Bats /fi:"“__ _w//mmflg//,/ . ;7 | // 7 | //////// 7 /////'26//////////////' 2|74 7 , LiF/BeF, /ThF | ‘ N\ N (L NAN L o e A : J = . 5 ‘j - A ’ | ‘ - o ) . - LiF/éer/m;/FP- O R— g SR C fp—a ""ETR:‘SA!?; : L—nz | l - o A : ?B/R(/p’fl// - i l. - sslzt:c:;::::sm.s ' BiScARD For L ////cnmovm_ 7 ?fié{ 7 S Vs R e/ . A S S - FERTILE STREAM RECYCLE A 17 1 LiF/BeF,/UFy RECYCLE ' FERTLE STREAM RECYCLE YALT DISCARD FOR , S FP REMOVAL “ Fig. 2. Fuel- and Fertile-Stream Processing for the MSBER and MSBR(Pa). the organization for safety in MSBR power plants will have to satisfy =~ &“; the licensing and regulatory requiremenfs which exist; also, MSBR plants must satisf&ctorily pass safety feviews, inspections, and testing, Plant operations will have to be safe and efficient so that the health and safety of plant personnel and that of the general public will not ; be endangered, and so that thé plant can operate economically on a long- term basis. While it appears that the safety of MSER systems can be’ assured at costs as low or lower than the-safety-requiremsnt-cOsts of other feactor power plants, a definitive evaluation cannot be made until detailed safety studies have been performed. | In discussing MSBR safety, credible 1ncidents which would normally never occur must be considered. Plant systems involved are primarily N the reactor system, the supercritical-steam system, the fuel processing system, and the off-gas_system.' These are discussed below relative to their influence and function on reactor safety. Also, a discussion is includéd of possible events which can be described qualitatively, but | which need detailed investigation to be evaluated adequately. These . involve reactivity coefficients, control rod function, possible inci- fi# dents, and reactor stability. Finally, a summary is given of the MSBER safety program, along with estimates of the costs associated with re- solving safety design questions. 2. MAJOR PLANT SYSTEMS INFLUENCING REACTOR SAFETY The reactor system is the primary one of interest, but other systems can also influence reactor behavior. For example, rupture of the supsr- critical boiler-superheaters could lead to high pressures in the secondary coolant system, which in tumm could lead to rupture of the prifia:y heat exchanger if proper safeguards are not employed. Such a train of events would influence the reactivity of the reactor core, and need to be cen- sidered relative to the adequacy of reactor plant containment _ . Another plant system of importance is the fuel recycle system, since it is integrated with the reactor plant and operates "on-line." This operation could introduce reactivity changes into the reactor system. ) [i -~y "‘tj » ) 1) 'Also, the off-gas system is an important protective system relative to | the reference of radiocactive gases from the plant site. -~ 2.,1. Reactor System As considered here, the reactor system contains the reactor core, the primary and secondary circulating-salt loops, and associated pumps‘ 'vthe_heat transfer equipment. Important!items in this system are indi- cated in Figure*} The reactor vessel is housed in a circular cell of reinforced con- crete, about 36-ft-diam by 42- -ft-high. This volume also contains the four fuel- and blanket-salt primary heat exchangers and. their respective cir- _culating'pumps;tr The wallrseparating this cell from the adjoining cells is L-ft-thick, and the removable bolt-down roof plugs totel 8 ft in ‘thickness. The pump drive shafts pass through stepped openings in the special concrete roof plugs'tolthe drive motors which are located in sealed tanks_pressurizedlabore'the'reactor cell pressure. The control rod drive mechanisms.paSSfthrough the top shielding in a similar manner. The coolant-salt pipes passing through the cell wall have bellows seals at the penetrations. | , | N The cell is lined with l/h to 1/2-in.-thick steel plate having welded Joints, which, together with the seal pan that forms a part of the roof structure, provides a cell leak rate less than 1% (volume) per 24 hr. The cell is heated to above 1050°F by radiant heating sur- faces’loCated at the hottom'of thefcell. The liner plate and the con- - crete structure are protected from high temperatures by 6 in. or more of thermal insulation and by_a.heat removal system. The reactor and heat exchanger'support“structures are cooled as-required. Thus, there are several barriers to protect against the escape of radioactivity The first is the primary reactor piping and equipment the second is the seal-welded containment vessel, and a thlrd is the ~ reactor building proper which is maintained at & negative pressure by : ventilating fans which discharge through a stack-filter arrangement All penetrations into the reactor cell, such as those associated with instrument, electrical, and service lines, are equipped with sealing devices. ORNL-Dirg . -67 -6314 R ; ALY, - nrr-a g E G St v . o ta &se 8 of 2 -.'E :;.' " gy Fig. 3. General Arrangement of Equipment in the Reactor Cell and Coolant Cells. *) LG =) 1 The four cooling-salt-circulating.circuits are houeed-in individual - compartments heving h;ft#thick reinforced‘concrete walls and bolted-do#n; removable roof plugs. Each compartment contains four boiler-superheaters, two reheaters, one coolant-salt pump serfiing'the boiler-superheaters, and one coolant-salt pump_supplying the reheaters. All pipingrpassing into these cells from the tfifbine_plant has-sealed'penetrations and valving located outside the walls. The coolant-salt‘pufip drive_shafts extend - through the roof plugs and the cells are sealed and heated in the same manner as in the reactor cell., Normally the temperature need not be maintained above T50°F, however. | | The secondary .coolant lines are maintained at a higher pressure than the reactor system (about 200 psi, compared with ~100 psi in the reactor), so that in the event of a primary heat exchanger tube failure, leakage of. radiocactive fuel salt-ihto the secondary circuit will be minimized. Ordi- narily, the activity of_the;éoolant salt will be that due to N*° (formed from the N,a reaction on fluorine and having & half life of 7.k sec) and Na®% (formed by an n,y reectiOn'and~having”a half life of about 15 hr). .In each case the neutron source for activation is the delayed neutron emission in the primazy,heatiexchanger. The'design~pressure-for the reactor cell and the four adjoining compartments is expectedlto:be-about 45 psig.'.Pressure-suppression systems are provided, the reactor cell system being separate from the system used for the other compartments. These suppression systems would contain water storage tanks sofithat.vapors released into a cell would pass through'these.tankswendfbe7conden5ed,maintaining the.cell pressure below the design value. Noncondensable gases would be contained until - they could be'disposed“ofrby5paesage through the off-gas system. When the ~coolant salt is discharged into the water in the pressure suppression system some HF will be produced, The quantity and the effects need to | be evaluated. Studies made for the MSRE suggest that corrosion of the ‘steel llners and tanks by the HF will not be & ‘serious problem. The fuel drain tanks malntainisubcritical storage of the fuel and also remove decay heat for maintaining proper fuel temperatures. Evapora- tive cooling is provided. The coolant drain tank is similar to the fuel drain tank except no cooling is required. . An inert cover gas system is 10 O, - provided to protect the molten salt from oxygen and moisture at all times. In order to keep stresses within‘eQuipment low, normal heating end cooling of the reactor will be done slowly at rates of 100°F/hr or less, applying temperature differences less than about 100°F. However, the reactor system should withstand several severe thermal shocks (such as & rapid fuel-salt temperature rise of about 4O0°F) without'breaching. The homogenebus and fluid nafiure_of molten-salt fuels permits réady transport of material from one system to another.. From the viewpoint of safety, it is important that the fissile fuel'remainhomogeneqqsly'diétri- buted in the carrier salt. This has been demonstratedrrepedtedly‘under both nonirradiation and irradiation conditions; in additibn,'chemical stability of the fuel salts improves with increasing temperatufe,'a favorable relation. Also, the fuel salt expands with increasing tempera- ture, effectively leading to expulsion of fuel from the core region and leading to a negative temperature coefficient of reactivity. Because of the ease of fuel addition and removal, very little excess reaétivity is provided within the reactor during normal operating conditions. - o s Fission gases are continuously removed from the reactbr"core on & very short cycle time (less than one minute) by sparging the salt with - inert gas. Fuel processing takes place on about a 30-day cycler(for the fuel salt), so that the fission product content of the reactor.systém is always relatively low. Since the fuel salt does not wet the container material or the moder- ator, drainage of the fuel salt plus flushing the system with carrier salt should remove a large fraction of the fission products from the circulating-fuel system. The actual behavior will need to be studied experimentally. ' 2,2. Steam System The steam system is indicated in Figure 4 and consists of the - coolant-salt heat exchangers, boiler feed pumps, feedwater heaters, the 'turbine-generator, and associéted equipment. The steam~power system uses Steam conditions of 3500 psia --.1000°F/1000°F, which are repre- sentative of modern steam power plant practice. The feedwater enters | 'y wh dRNL-;-Dwg .-67-6315 N e . e ST T . . s wiemm. | diils SN Wiy -l - . ‘ L1 —. noRteed | ’ 1 MO 1 5 ool ¥ ' " s t . _ “ . ¢ | ._ . - - s p— .fi‘!l. ! ¥ I‘b”'lui-'l‘ — "U . $ ool s ' e ' ro bt 1 I . — | + I . — Tll_!v ' b .m i —-ru b sag-bk-——a fppeeRe— 4o q1 P JENH ¥ - — i - m .8«4.!..!....._.." Jd Laus RE-—H I [ e R ‘ . M" ““ s === “ N2 qu.-ssewnuunlaq LT e e oilmlii o i oo H ! » ' j§ 1 benc=ddc o R S iR e et 11 JOLpron =300 b mr‘_—:m. N R T X SR VTV - PO OO 1 _M gk AN i &m Mfl. A-fl"-“ 1t -2 X1 nmqrmammm ey oy At v 948 B . ST s 0N e XS e _n 700 Oniab AR FOSER SN MUK . F PRORSAR-POMTER RAPY - am. WTRY o Py WEACTOR MEAT SUNT snom Y & evaL T 7. & Ol MEY WRAT RATY l'lllll".lll- - 30OP- JRA i+ MLEN MSBR Power Cycle. Fig. 4. 12 | | | " the boiler at T00°F so that little or no freezing of the secondany f : \o/ coolant salt takes place. | | ~ The 16 boiler-superheaters consist of U-tube-U-shell hea't exchangers, | which transfer heat from the 1125 F coolant salt to the 700°F feedwater and generate steam at71000°F and 3600 psia. Variable-epeed, coolant-~ { salt pumps are used to permiticentrol of the outlet steam temperapuré. There arejeiéht shell -and-tube heat exchangers which'function as re- Jheeters and transfer heat from the coolant salt to 570 psia steam from the highfipressure turbines exhsust, raising its temperature to 1000°F. Reheat steam preheaters are used to heat this exhaust steam to. about 600°F before it enters the reheaters. - S . The heat-exehange equipment is located within containment cells which communicate with the reactor cell by means of coolant-salt lines, and with the turbine room by means of steam and water lines. In addi- tion, these eells communicate with the fuel proeessing area by'means of - small coolant-salt lines and with the control and service areas through penetrations for gas, cooling water, instrumentation lines, etc. These ¥ cells also communicate with a vapor suppression volume through a large | | conduit equipped with a rupture disc. The vapor-condensing system pro- vides pressure control of the coolant-salt cell in the event Qf'a rupture of the_steam or water circuits. Biological shielding is provided.for the cells, and a controlled inert gas atmosphere is maintained. | Molten salts do not undergo a significant chemical reaction with water; however, high-temperature steam is produced when water contacts molten salt. In order to provide for accidents producing steam, or for leakage of high-pressure steam into the coolant-salt cells, a vapor- suppression system is used to provide pressure relief, and maintain - pressures below the containment design value of about L5 psig. Auto- matic block valves are provided in the steam lines to reduce the likeli- hood of draining the water in the steam system into these cells in the event of a rupture. _ To protect against high pressures in case of failure of a super- heater tube in the heat exchanger, rupture discs are provided on the shell Side'of the superheaters‘and»reheaters for venting the coolant . O ”C " )] ) o 15 system into the vapor condensing,system. These rupture discs protect against overpressure in the coolant-salt circuit and thus protect the reactor'system; which is separated from the coolant salt by the tube walls of the primary heat‘erchanger. 2.3. Fuel Recycle Processing System \ The floWsheet for the MSBR processing system has been given pre- viously in Figure 2. The core fuel is processed by the fluoride vol- | atility process to separate the uranium from the carrier salt and fission products. The valuable carrier selt 1is separated from the rare-earth fission products by_the vacuum-distillation process. The fuel salt is reconstituted by absorbing UFs in uranium-containing‘carrier salt, followed by reduction in the'liquid phase bycbubbling hydrogen through the melt. Excess uranium fromithe‘reactor is sold as an equilibrium mixture of the fuel isotopes. Fuel salt is returned to the reactor as needed. | - ' The blanket salt is processed by the fluoride volatility process along with a Pa-removal process in which Pa is extracted by liquid bismuth containing dissolved thorium. The same process also removes uranium. Small side streams of fuel salt and blanket salt are continuously withdrawn from the reactor circulating systems and. routed to the process- ing plant located within the same building. At the same time, makeup streams. are returned to the fuel and blanket systems at the same rate they are removed. Theserratesiare:lowVenough thatino significant reac- 'tivity additions to the reactor should normally be possible. The fuel-recycle processing‘plant is located in two cells adjacent ,'to the reactor shield one contains the high radiation 1evel Qperations,r “and the other contains the lower-radiation-level qperations. Each cell '_is designed for top access through a removable biological shield having ‘& thickness equivalent to 6 ft of high-density concrete. A general plan 'of the processing plant and a partial view of the reactor: cell is shown ~in Figure 5. The highly radioactive operations in the fuel-stream proc- essing are carried out in the smaller cell (upper left). The other cell houses equipment for the fertile stream and the fuel-makeup-stream operations. 14 ORNL-DWG €6-7459 ) b PROCESSING - c CeLLs o ARNoK I .; WEAY v z EXCHANGER ! % : w p [ '_..-.ujt..l:\x[.‘vf.".v, TN S s e ‘ | ' W 35 ‘ ~hK ! SUPPLY e ; A . MAKEUP s b AREAS__ i | 1 e 2 o | .ets'f—f“ : : o @ f i . o] H @ . . swePYe o (F A O Gl ' oo N ‘. ®e 5 i :! T) o3k | W, w ©O0OL % swPrYd O Iy, oo ™ Q Oy . UFg 0O ; PRODUCT () freavi®] ¢ it §—] : f '? ‘ "o - Lo @ 4 B8 i 7 Fig. 5;' General Locatiofi of Processing Plant Eqnipment. O, » » - *) 15 The highly radioactive.cell contains only fuel-stream processing equipment consisting primarily of theffluorinator, still, waste receiver, NeF and MgF, sorbers, and associated vessels. The other cell houses the ‘blanket processing equipment,and fuel- andlfertile-stream makeup vessels. The processing plant will use hydrogen and fluorine gases in the treatmentrof the salts. Care must be taken in utilizing these gases because of the hazards assoclated with obtaining explosive mixtures of hydrogen and oxygen, or fluorine. Thus, hydrogen must be isolated from the fluorine and from the reactor cell. Also, fluorine must be isolated from the reactor system, and:organicgluhricants must not enter the fluorine system. - | | ‘ The processing plant will utilize the same off-gas disposal system as the reactor plant. This combined use should not introduce operating hazards. The integrity of the cooling systems needed for cooling of processing equipment must be assured, both during continuous processing and during storage of waste. | o Criticality considerations must be considered, such that recovery of fissionable material constitutes no criticality hazard; however, due to the relatively small quantities of fissile fuel held up in the proc- essing plant and the character of the materials handled, no difficulty is anticipated. | Reactor fuel additions will be done primarily through the return line from the processing-plant. The associated components would be of all-welded construction and would be maintained by remote maintenance ‘procedures. ‘g 2;4.s OfffGas.System ‘ Xenon and . knypton as well as tritium are stripped from the fuel ,salt in the reactor circulating system by sparging with an inert gas, such as helium. This gas along with. the gases generated are treated in the off-gas system. - | | The flowsheet for the off-gas system in shown in Figure 6. After pass1ng through a decay tank, the fission product gases are passed through water-cooled charcoal beds where xenon is retained for 48 hr. Cm OUnLETY ORNL-Dwg . -67-6316 \‘\ 49 hey MOLDUP NENON WATER ~ COOLED CHARCOAL ADSORECRS Fig. 6. Flowsheet for the Off-Gas System. » i) i 17 In addition to removing the *®BXe, this system of circulation effectively transfers arlarge fraction of the other gaseous fission products to areas where the decay heat can be removed more readily. ' About 0.1 scfm of the gas stream leaving the initial charcoal beds (or 0.4 scfm total for the four fuel-salt circulating loops) is passed through additional charcoal beds and then through a molecular sieve (operated at liquid nitrogen temperature) to remove 99% or more of the 85Kr and other gaseous,products. The effluent helium can be recycled into the reactor system or passed through filters, diluted, and dis- charged into an off-gas stack. . The molecular sieves canh be regenerated, and the radioactive gases that are driven off can be sent to storage tanks. Concentration and storage of the tritium will probably require additional equipment; this operation needs additional study. A helium system provides cover gas for the blanket pump bowls, the drain tanks, fuel-handlingtand processing systems, etc. Essentially all helium will be recycled to the cover-gas system. Any discharged | cover gas passes through charcoal adsorbers and absolute filters, is diluted with air, and discharged through the off-gas stack. Relative to the off-gas processing of the fuel recycle system, most of the facilities are located 1n the processing plant proper. In the processing plant, off-gas comes primarily'from the continuous fluorinators, while smaller amounts are formed in various other processing vessels. The gases are processed to prevent'the'release of any contained fission prod- ucts to the atmosphere. . Excess fluorlne used in the fluorinators is re- 'cycled through a. surge chamber by a. pos1tive displacement pump, and a small side stream of the recycling fluorine is sent through a caustic scrubber to prevent gross buildup of fission products. Each of the 'processing vessels and holdup tanks has off-gas llnes which lead to the "scrubber for treating HF, fluorine, and volatile fission products. The scrubber operates as’'a. continuous, countercurrent packed bed with rec1rculating aqueous KOH. A small side stream of KOH sOlution'is ~ sent to. Waste, and the scrubber off-gas is contacted with steam to hydro- lyze flss10n products such as tellurlum. A filter removes the hydrolyzed products. The noncondensable fission products are sent to the reactor off-gas facility. '718 : The off-gas gystem must be designed to handle the very radioéctive , gases and to provide'cooling of these gases. Also, while the wvapor pressure of molten salts is very low, MSRE experience;indicates that some particulate matter can be carried into the off-gas stream. Cold ‘trapping or.filtering must be provided in the off-gas lines for removing . these mist-like'particles. Any oill leakage and -associated decomposition' ‘prbducts entering the off-gas system must be removed by a filter system. The off-gas system primarily removes fission products, recirculates sparge gases back to the reactor system, and holdslup fission products until they have decayed sufficienfly for disposal. If fission products are not held up sufficiently, radioactive gases are discharged prematurely, leading to high activity levels. 3. REACTOR SAFETY ASPECTS In operating a reactor power plant there always exists the possi- bility that reactivity can be inadvertently added to the system, lead- ing to a system disturbance. If this disturbance is very small, no ill effects result. Increasing the degree of disturbance can lead to con- ditions which affect reactor operation (operating safety) and eventually to conditions which affect the safety of the general public (ultimate' safety). In this section the MSBR operations are discussed from the viewpoint of items which need to be evaluated from a safety standpoint such as reactivity coefficients, control rod funétion,VPOSSible reactivity events that could cause reactivity additions to the reactor, and thé stability requirements of the reactor power plant. In general, the specific situations which need to be evéluated are dependent upon the design and operational features of the system. 3.1. Reactivity Coefficients and Kinetics Parameters A number of reactivity coefficients are associated with an MSER system. These include those associated with temperature, voids, pressure, fuel concentration, graphite concentration, xenon concentration, fuel »¥ . 19 burnup,‘fuelrflcw'rate,:and involve the fuel and blanket fluids separately and together. From the'viewpoint of‘reactor?safety, the most important coefficients,appear to be the temperature coefficients of reactivity for ' the fuel salt, the blanket salt, and the graphite moderator, and the fuel concentration coefficient of reactivity. There are special circumstances where others are also of importance. All of these need to be determined specifically. o | Molten-salt reactors have, in general, a relatively large negative fuel temperature coefficient of.reactivity, due to the expulsion of fuel from the core region with increasing temperature; :The'value for MSBR systems will be in the range of =1 x 107® Ake/°F to =5 x:107F Ake/°F,'the, value being a function of design and operating conditions. This coef- ficient gives inherent control and safety to molten-salt systems, since any increase in power level tends to decrease the reactivity and thus decrease the power level. Since MSBR's will normally operate with only low values offexcess'reactivity available, the temperature coefficient appears sufficientcfor controlling the reactor without excessive tempera- ture variations. This inherent control feature permits use‘of control rod mechanisms which have relatively slow action. -~ ' ' - Increasing the prompt temperature coefficient of reactivity generally improves the safety and stability margins of reactor operations, provided that-the reactivity is added by means other than the temperature coeffi- cient. HOyever, the temperature coefficient itself can add reactivity by means of '"cold slug" type occurrences. . Such an occurrence in an MSBR - would be normally associated with en increase of fluid flow rate; however, 1ncreasing the flow rate tends to- decrease reactivity due to the associated ‘increased loss in delayed neutrons. The effective value for the delayed ‘neutron fraction in 223y. fueled reactors 1is about 0.003 in fixed fuel systems; in MSBR systems, the effective value for beta durlng fuel cir- culation would be about 0. OOl. e o T | ' Reactivity coefficients need to be determined in order to prqperly "'evaluate the safety of MSBR systems.» Primary values appear ‘to be the temperature coefficients associated with the fuel and blanket- fluids and with the graphite; the void coefficients associated with both the fuel 20 and blanketrfluids; concentration coefficients associated with the fissile and fertile salts in the core; reactivity coefficients associated with loss of-fuelfflow;,effective,delayed neutron fraction as a function of - flow and power conditions; and the reactivity effects associatedrwith g&aphite shrinkage, graphite breakup, and fuel soakup by graphite. The reactivity cdefficients need to be consistent with the kinetics model used in the:safety evaluations, and time- and space-dependent criti- cality effects need to be included in such studies. These time- and’ space-dependent effects should include consideration of the'different heating and flow, rates within the reactor, afterheat generation, and the change in the effective delayed neutron fraction during a power pulse.. Other parameters needed in the kinetics“analysis include the prompt . neutron lifetime and xenon poisoning effects. 3.2. Control-Rod Function One or more control rods are provided in the MSBR in order to provide flexibility in reactor operations, and to control reactivity additions such that fuel temperatures and associated temperatures do nbt become excessive. As mentioned in Section 3.1, inherent control is provided | through the negative temperature coefficient of reactivity, which pro-- vides prompt protection ageinst reactivity additions. At the same time, if reactivity additions take place over a long-time interval, the total regctivity added may lead to undesirably high fuel temperatures if'only the temperature coefficient is utilized (hcwever, such temperatures may be permissible for relatively short times -- order of hours). Installa~ tion of control rods which are slow acting (response time of about one second) appears sufficient for controlling maximm fuel temperatures, and would pemit reactivity control independent of fuel temperature. Control rods provide an easy means of controlling reactor power at low power levels where the temperature coefficient is a poor operational control; during power operation, control rods would normally be fully withdrawn from the core. S o . - oy . » [ 1] 21 The required reactivitY'worth of control rods is a function of shim and shutdown margin requirements, and needs to be investigated in detail. Control-rod.wbrth as a function of fuel concentration, power conditionms, ~and reactor deeign'should be studied. In particular, use of "control rods" which use fertile blanket salt as absorber material need to be evaluated. | 7 The action and position'of'cantrol rods during reactor stertup need to be examined. It appears reasonable that the rods be fully inserted prior to start of fuel circulation, with eriticality achieved by with- drawal of the rods. o In general, the control’rods of the MSBR need not be used for shim 'requirements (e.g., change.in steady-state Xe level, or fuel temperature); rather, associated reactivity changes can be made by adjusting the fuel concentration. Reactor shutdown can be obtalned by insertion of a con- trol rod, or by stopping & fuel pump which leads to fuel drainage from . the core region. | _ It does not appear that control rods need to control large amounts of reactivity (probably less ‘than 1/2% in reactivity) or to have fast response times (response times of about a second are probably sufficient). chever, detailed studies need to be performed relative to specific re- gquirements as a function of eore'design. The results obtained will be used to determine general considerations concerning'control rods and MSBR safety. 3.3, Reactor Incldents 'Items'to be considered hene‘eOHCern physical~events9which influence . ,isystem reactivity, as. well as some which do not influence reactivity per_se. QperatiOnal safety, or the abllity to continue reactor Qperation after'abnormal events, 1s involved as well as ultimate safety where con- _ tainment of gross radioactivity and public safety are. the important con- icer.ns These definitions are illustrated below. 22 As normally considered, a reactor ificident involves a core reactivity ‘addition. If the reactivity addition is small enough, thére is primarily a small disturbance in reactoi power, with no deleterious effects to the reactor plant. Under these circumstances, operational safety is main- tained. If the reactivity addition is large enough, a graphite tube separating the fuel and blanket fluids may break because 6f the pressure rise, with no other untoward effects. Under these circumstances the reactor plant has produced no public hazard, but must be shut down for repairs. Under these circumstances operational safety has not been maintained, but ultimate safety has not been involved. If the reac- tivity addition is so large that the reactor vessel ruptutes and gen- erates a disruptive force which results in penetration of the reactor containment, both operational and ultimate safety may be violated. Reactor plant incidents can also occur without'the'reactof itself being involved. For example, if mechanical failures occur which permit water or supercritical steam to contact secondary coolant salt within the cell containing the steam generators, high pressures could occur in the cell and lead to rupture of this containment. Release of steam con- taining particles of radiocactive coolant salt could involve personnel hazard and ultimate safety. The design of an MSBR plant must consider both operational and ultimate safety aspects; the resulting reactor plant must have opera- tional safety assured under nearly all credible circumstances, and ultimate safety assured under all credible circumstances. Items which need to be considered in such safety design studies are discussed below and are separated into those which involve reactivity addifiions to the reactor proper, those associated with mechanical and physical'integrity, and items not covered in either of the above categories. In nearly all cases, these events require malfunction of equipment or reactor-qperation as indicated below. | | C. &} By 23 ) 5.53.1 Reactivity Additions | Reactivity can be added to the MSBR byemechanisms and events similar to those.considered for the‘MSBE;s in addition, the use of two fluid streams separated by graphite tube walls and the supercritical steam- power cycle requires that several other events be-considered. Possible reactivity additions need to be investigated in detail. The protective devices available to the MSBR are similar to those in the MSRE. “Prompt" protection is afforded by the negative temperature coefficient of reactivity and '"delayed" protection is provided by the control rods and also by drainage of fuel salt from the core region: Since all reactivity changeS-involve rates of addition rather than re- activity steps, an important factor in protection is the minimum neutron source strength which can exist_inithe core. The MSBR fuel contains an inherent neutron source oflnearly;lof-n/sec“due to the a,n reactions re- sulting from the alpha decay of *22U and ®2*U in the fuel salt. An addi- tional neutron source exists from the 7,n reaction resulting from the decay of fission products W1thin the fuel salt; the photoneutron source is greater than 10 n/sec_for_about four months after reactor shutdown following a month's operation at power. Thus a strong internal neutron source is always present if reactivity 1s added at low rates, multipli- cation of this source results in a significant increase in reactor power before large‘amounts.of,reactivity,can_be_added to the system, which in turn permits\the temperature.coefficient=to become effective after rela- tively small gross reactivity;additions. _ _ Net Fuel Addition to Core. . Probably the largest reactivity addition that can take place in-an MSBR is that associated W1th breakage of one or 'more graphite tubes with net. addition of fuel salt to the core region. ,However, spec1al c1rcumstances have to exist for this to take place since | ‘the blanket region qperates at pressures higher than the fuel region, and i.:tube breakage under normal conditions ‘would add fertile salt to the fuel region and reduce reactiv1ty -Thus, to add reactivity, ‘the fuel pressure - would have to rise higher than the blanket pressure at the time of, or "ishortly after, breakage of a graphite tube. This is possible if the high pressure of the supercritical steam system is at least partially 2y transmitted to the fuel-salt system, or if there is a decrease in the ' \Ej blanket pressure without a concufrent decrease'in‘the core-fuel pressure. Failure of the'tubing in the boller-superheater could allow the high-pressure steam to enter the coolant-galt system. - To protect against a buildup of pressure in the coolant system, rupture discs are provided in the steam generator and reheaters, and also could be provided'on the shell sides of the fuel and blanket heat;exchangers.’ If these rupture discs fail to operate, or fail to operate quickly enough, it is con- ceivable that a buildup of pressure in the coolant system could cause failure of the primary fuel heat exchanger. The likely means of failure would be rupture of the shell or éollapse of the tubes, neither event transmitting the pressure increase to the fuel fluid. However, if-there' were localized weakness in a fuel-heat-exchanger tube, due to a defect in manufacture, fretting corrosion, etc., failure of & tube could occur leading to & buildup of pressure in the fuel system. Alternatively, loss of overpressure in the blanket region could permit operation with fuel pressures higher than blanket pressure. If a graphite tube failed under ® such operating conditions, there would be a net fuel addition to the core region. The reactivity addition would depend upon the pressures and flow passages involved and their variation with time. If steam does contact coolant salt, no éxothermal reactions of any consequence are involved. Mixing of steam with coolant salt would oxidize the coolant salt, but no safety hazard would be introduced beéause:of this action. However, the corrosiveness of the mixture to the container ma- terial needs determination. There are no fission products in the coolant salt, and the induced activity present would decay (the primary activity is associated with Na®** and N'®, having half lives of about 15 hr and T sec, respectively). Cleanup of the system and repair or replacement of damaged equipment appears possible. | P The coolent salt is compatible with the fuel salt, so leakage of coolant salt into the reactor system does not involve safety; any-such leakage would reduce reactivity The BF3 added to the reactor fuel could be readily removed by heating the salt, with the BF; removed as & gas. ¢ 22 Contacting fuel salt with steam would oxidize the uranium, but probably would not cause“any;problems other than those associated'with subsequent cleanup~of,thejfuelts HOwever,_possible~reactivity effects due to fuel precipitation need to be studied specifically. At this time it appears reasonable that engineered safeguards, such as:installing rupture discs within the heat exchangers of the coolant system, and providifig*strehgtheoed.primary system heat exchanger tubes can eitherfprotect againstpsuch-sh_accideht, or-keep_the_amount'of fuel salt added to the core“regithsmall enough that ultimate”safety'is not involved. However, detailed-studies are needed to examine this situation. Reactivity Changes Due to Graphite Behavior. In addition to the case discussed'sbove‘in whichibreakagehof_graphite tubes was'sssumed to teke place, other graphitelbehavior can effect reactivity changes. For example, shrinksge of graphite during radiation exposure can effec- tively influence fuel concentrations' however, the associated reactivity changes should take place at: rates such that they can be readily com- _pensated by adding or removing fuel through normal- operations.: Reactivity can be sdded if part of the graphite inside a fuel tube were to break awayrfrom the.tube'proper'and.be swept out of the core region. ;Only.smsll amounts of reactivity,could'be involved so long as this actidn'took'place in. single_tubes,'snd noudifficulty‘for-this situa- tion would be anticipated..'Alternatively, if graphite were removed from - the blanket portion of the core: region, it would be displaced by fertile salt leading to a decrease in reactivity such that safety is not involved. - Graphite is compatible with molten salt but fuel penetration into “the- gr&phite could. take plsce with time. Here again, the time element involved'would make such events insignificant from a safety viewpoint. ';_ij on the other hand a pressure rise tock place in the ‘core which caused | the fuel to penetrate and. fill voids in the graphite, perhaps a signifi- 'rcant reactivity addition could be obtained. The actual addition is de- 'pendent upon ‘the physical properties of the graphite employed. If the ‘7pressure rise. occurs because of a previous reactiv1ty addition, the - pressure buildup itself would expel fuel salt from the core and tend to decrease reactivity 26 Fuel-salt penetration in graphite appears to present little problem during normal operation, but may present difficulties during emergency shutdowns which require fuel-salt drainage. Fuel remaining in the- graphite would generate decay heat which could lead to undesirably high temperatures (temperature distributions and levels influence thermal stresses and creep rates, which can affect the mechanical integrity of the graphite). I ebility of blanket salt to remove this decay heat needs inVestigefion. Reactivity Changes Associated With Changes in Flow Conditions. In a circulating-fuel reactor, an appreciable fraction of the delayed neutrons can be emitted external to the-core“nnder normal flow canditiqne. In- creasing flow thus tends.to:1gwer”the,contributidn of delayed neutrons t0 the fission chain and also decreases the average neutron lifetime of the reactor. While lowering the delayed neutron fraction (beta) is nor- mally considered detrimental to safety, this is in the context of systems having instrument control. Lowering the value of beta in a system having inherent control under the condition that reactivity additions take place at relatively low rates does not significantly decrease the ultimate safety of the system. Also, the effective value of beta increases during a rise in power, a favorable condition. Lo Since delayed neutrons are "lost" because of fuel circulation, stop- page of flow due to pump power failure would tend to add reactivity to the system. However, in the MSBR the reactivity addition would only be about 0.002. In addition, stoppage of flow leads to drainage of the core, which would make the reactor subcritical. The fuel temperature rise due to afterheat during drainage of the core may be.the most signifieant vari- able, and needs detailed study.' Also, time delays invfuel"drainage“frdm the core following pump stoppage needs to be investigated experimentally, and the results interpreted relative to reactor safety. Another reactivity incident possible with systems having a negative temperature coefficient of reactivity is that of the "cold slug" accident. Such an accident could occur by starting the fuel-eirculating pump at a time when the fuel external to the core has been cooled well below that . of the fuel in the core. The cooler fuel would add reactivity when it entered the core; this addition could exceed the reactivity decrease due & -y o7 to the "loss" of delayed neutrons associated with fluid transport. By going critical,only'with.thefpump.on;fmaking use of the control rod for thiS'purpose; would avoid the "cold slug" incident. - The seriousness of - the cold slug incident and the control mechanisms needed under various _ c1rcumstances needs investigation. Drainage of .the reactor fuel sy stem begins automatically due to - gravity forces when the fuel pump stops. Fuel from the core drains by gravity into the sump tank of the fuelrpumpoWhere afterheat is removed by cooling coils. Convective circulation may be assisted by flow of gas used to sparge xenon from the fuel salt. However, as pointed out above, ~ fuel and graphite temperatures also-need to be studied during fuel drain- age from the core. In general, the ability and need for afterheat re- - moval requires detailed studies. Changes in Fuel Concentration. Reactivity"can-be-added by increasing the concentration of fiss1le_material within the'fuel-fluid; examples of possible events are filling:the'fuel tubes with salt containingrabnormally high fissile eoncentrationsf"and returning salt having sbnormally high - 'fuel concentrations from the processing system to the reactor system. The reactor would initially be "filled" by adding fissile material to the'carrier-salt while the latter was circulating. If, however, follow- ing criticality and drainage of fuel salt from the reactor core, the 'fiss1le concentration'infthe drained fuel salt were increased inadvertently, refilling the core could result in a supercritical reactor. Such'an event 1s highly- unlikeLy, since fuel ‘would. not be added in large amounts to the -'dralned system,.also, partial freezing of the fuel salt does not appear ' ”to 1ead to Significant increases of fissile concentration in the fluid '*',portion of the fuel. Specific cases need to be evaluated, however. The rate of return of fuel from the processing plant is low, and it 'will be difficult to add react1v1ty at a high rate through the processing '1ines because of- the limited rate at which fuel can be added , A more :frlikely way to increase fuel concentration above the normal value'would ipfltbe to fill the core with fuel having a temperature 1ower than the critlcal temperature. A reactlvity added by this means would correspond to & low- 'mrate addition and should cause no difficulty. - 28 _If fuel were to accumulate outside the core. region, and inadvertently “return to the core, reactivity could be added rapidly to.thé reactor. Since the fuel is homogeneous and chemically stable, this event does not appear to be likely; also, eny such possibility would be indicated by a previous reactivity loss.. Nonétheless, the consequences of:urahium precipitation or accumilation outside the core and its-subséquent'additiOn to the core region requireé general evaluation. Such studies will help determine operating procedures consistent-with reactor safety. While none of the above events eppears to constitute an operational or ultimate safety hazard, all should be considered in detail. - Reactivity‘Addition by Control-Rod Movement. The presen¢e5of a con- trol rod permits reactivity addition to the reactor by rod movement. - Normally the reactor would be critical with the control rod completely removed, - but there could be conditions whg:e criticality 1s achieved _wii:h the rod partially or completely inserted. The amount and rate of reactivity addition associated'wifh control rod movement.under'tfiese.con— ditions would be limited by the control rod worth (which will probably be under 0.005 Ake) and the rate of withdrawal (which will be limited to a low value). As with the MSRE, no difficulty is foreseen, particularly if rod withdrawal does not continue after the power level reaches an initial peak value as & result of rod movement. . Reactivity Addition Due to Positive Pressure Coefficient.' The'MSBR design specifies use of helium as & sparge gas to remove xenon from the circulating fuel. As a result of this operation, some gas will undoubtedly circulate through the MSBR core, resulting in a positive préssure coef- ficient of reactivity. The importance of this coefficient on safety is a function of the gas content of the core, which in turn is related to the ease of stripping xenon from the fuel salt and the efficiency df the'gas separator used to remove sparge gas before it enters the core region. An. increase in pressure would decrease the fraction gas in the core'énd'in-' crease reacfivity. Experience'with'the MSRE indicates that the above is’ not & serious problem, but it needs to be evaluated_specifically-fbr the MSER. - 1 W C 29 33,2 -MEchanical-andsPhysical Integrity — Containment - This subject is related to the reactivity additions discussed above. Here, the discussion is concerned with containment relative to events which do not necessarily require or result in reactivity additions to the reactor systemr, Some offthe,questions which arise are: What are the'consequences of havinglwaterrand salt in a cell if these materials accidentally make contact? What are the-cooling conditions required if there is-mixingrofVSalt,and'waterfr What are the consequences of fuel- salt leakage or diffusion into;the_coolant-salt.system? ,How-practicable is it to maintain low leakage.fromra-containment-cell'at the temperatures involved (leakage of no more than 1% of the containment volume per day)? What are the consequences of a major spill of fuel salt within the reactor rcell?.. . ' The containment of the?reaC£or plant has to be assured even though there is rupture of, or‘leakageifrom, the primary and secondary salt systems. Rupture and/or leskage may result from overheating, overstress- ing, corrosion, or other unekpected'material failures. The severity of the containment problem will depend on the amount offsalt.spillage, the rate at which water mixes with hot salt, end the amount of water added to the cell. ConsequenCes'of'a'spillfaccident are heat generation, pressure buildup,-and release of fission products into the cell, and these will'need’to be evaluated:for specific cases. Problems associated with a major spill of fuel salt within the reactor cell must be considered in the detailed design of MSBR. systems and must also be studied experi- mentally If water is present' corrosion of steel by HF must be cons1dered. The effects of local.thermalhexpansion or energy deposition due to hot salt‘spiliage-needs evaluation. - Provision should also be made that oil from the: pump lubrication system does not .contact hot components, al- - though if this does occur, there normally would.not be sufficient oxygen i'to support combustion in the- cell atmosphere of inert gas (nitrogen) - In order to assure contalnment knowledge of the very long-term creep ebehavior of materials under plant operatlng conditions is needed. Infor- mation. is also needed on’ the conditions required to produce steam ex-" .plosions" upon mixing of salt and water; similar 1nformat10n is needed for the mixing of oil and salt.__' 30 The containment of fissionrpro&upts‘shquldfibe assureg,,and release of these through the off-gas system mst not constitute a safety hazard. This involves the amount of volatile materisl which is to be released and the amount of fission products carried in very small,-mistelike.Salt | particles. Ih-any case, the release of material through the off-gas system should be controlled so that exposure of individuals is not ex- cessive. This can be accomplished'by-filtration and retention systems- as required. Beryllium afid fluorine hazards, as well as radioactive iodine, must be considered relative-to_permissible release rates during normal operation as well as follcwihg & severe incident. rThe'réleasezof fission productS'upbn mixingrof fuel salt and water, or of salt and oil elso needs detérmination. A fission'product'flow and ifiventory shéet will be made as MSER design studies are made in more detail. Also, in- vestigation of the plating out of fission products throughout the reactor system is an important part of the chemical development program. The implication that fission product plating have upon reactor safety needs to be considered. In designing the reactor system containment, consideration must be given to the possibility of earthquakes. The effect of such an event on reactor containment is, of course, dependent upon its severit&, which in turn is a function of local conditions.' The possibility of.flooding and associated consequences is also dependent upon local conditioms. The most likely method of rupturing the'secondary containment is through sabotage, missile damage, acts of nature, or excessive pressuré. The generation of missiles in the reactor cell is not likely, since the reactor pressure is low. Missile damage and high pressurés'are more 1ikeLyrin.the coolant cell and steam plant, and, although massive'concrete shielding is provided, such events need further investigation. The con- tainment cells will be protected by vapor-suppression systems, whichf ~should prevent the pressure from exceeding the containment design figure (§h5 psig for present MSBR design) in case of buildup of steam pressure. In désigning the vapor-suppression systems, it is necessary.to'consider the amount of salt and water that can come together and/or the leakage of high-pressure steam into the containment volume. ‘Valves are located O « ME [ 1 A 31 in the steam lines which can be closed to prevent draining all the steam ' system into the coolant cell., The reservoir of condensing water should be adequate to keep the cell pressure below the design containment pressure. Also, the supercritical steam systems contain relatively small amounts of water in comparison with subcritical systems. 3.3.3 Miscellaneous Incidents Included here are possible incidents which are not covered in the above sections. These involve vessel criticality, heat removal, and heat addition conditions. | Studies are needed relstive to the possibility of attaining super- critical conditions in fuel drain tanks and in vessels of the processing plant, along with consequences of such occurrences. Also, criticality conditions might occur as a result of fuel spills.a In general, tanks which hold fuel should store it indefinitely in a suberitical condition. Accumulation of SPilled'fuel salt should be in regions which cannot attain criticality. - The afterheat conditions which can exist within the reactor plant particularly need to be studied in detall, and cooling and heating pro- vided and assured as needed. The temperatures occurring in the core following fuel drainage need to be evaluated as a function of fuel re- tention by the graphite. The influence of air contact on fuel salt needs study for conditions assoclated with core maintenance operations. The effects of salt freezing and melting in various parts of the primary and secondary salt circuits require evaluation, with equipment designed to minimize undesirable:effectsx(e.g.,_rupture of equipment). fTheconsequencesofflov_blockage~withthe reactorVSYStem require investigation. A partiallyrplngged‘fueljtube would normally not be deteotedqand could lead'to,ealt-bOilingrand“tenperature gradients which may:affect*the'mechanioal'integrity'of*the'fnel tube. Flow blookage”may - also lead to inability to remove all the fuel salt from the core, ‘which 'may lead to afterheat problems and/or maintenance difficulties.' 32 | s - 3.4, Reactor Stability. Although usually treated separately, reactor safety and stability- " are intimately related. Reactor safety normally considers‘reiatively large reactivity additiéns and their influénce on system behavior for small time intervals, while reactor stabiiity studies normslly consider small reactivity additions and determine whether they result in & buildup ' of oscillations to the point where reactor safety is involved. For the MSBR, investigations of stabllity are required to study the influence of inherent characteristics on instrumentation and control system }equirements. Altfiough the MSBR has a-negativé temperature éoef- ficient of reactivity, this ifi itself,is‘not sufficient to insure stability, particularly if the system has time delays. The MSBR has a number of builtin time delays which can either help or»hinder feactor stability, éuch as the tiine lags associated with heat transport from the graphite to the fuel, with fuel and fertile salt transport, and with delayed neutron _production. Because of the complexity of the three-loop system from a dynamics enalysis point of view, & preliminary linearized analysls should first be made to evaluate the current design and aid in establishing ap- propriate means of system control. ’ It is estimated that an adequate preliminary analysis for the com- plete system (reactor core to turbogenerator) would involve about 60 first-‘ order equations (about 1k for’{he fuel stream, 14 for the fertile stream, 7 for nuclear kinetics, 15 for coolent streams, and 10 for the steam system). These equations would comsider fuel and blanket nodes, transport deleys, heat exchanger nodes, and fuel leakage effects. Work is required in formulating the specific equations and in compiling and_e#aluating the physical parameters. Present computer codes could be utilized in the initiel analysis. In particular, codes are available for pérforming a dynamic enalysis utilizihg a general linear model. These can be used to give system eigenvalués, system freduency response, and/or gystem trans- ient fesponsel R - | S Some of the important items to be investigated in stability énalysis would be the significance of heat transfer lags between various parts of . bt c wh o 55 the reactor system, the, relative importance of the fuel temperature and graphite temperature coefficients of reactivity, and the influence of delayed neutron fraction and flow effects on reactor system behavior. The effect of fuel flow on_thejeffective-delayed neutron fraatibn may reduce this value from 0.003 to less than 0.00l. This reduction is not necessarily bad from either the viewpoint of inherent safety or inherent stabiliiy. 'In-fact,rlowering the delayed-neutron‘fraction can increase the degree of stability, as was the case for the Homogeneous Reactor Test (HRT). Also, the tendency (in circulating-fuel systems) of the effective delayed neutron fraction to.increase‘during-fhe power rise portion of a power pulse tends to aid stability. Plans for the MSRE call for operating that reactor with 233U fuel beginning in the fourth quarter of FY 1968. Studies will be made of the stability of the MSRE with the ®33U fuel and the results will be used where applicable in the analysis of the stability of the MSBR. , More detailed stability,analysis'studies would be dependent upon the results.obtained from the initial evaluation but presumably would include investigation of nonlinear effects and their relative influence on results. 4. MSBR SAFETY PROGRAM 4.1. Summary The studles and. investigations associated with MSBR safety are summarized here in terms of. general and detailed: studies which need to be done in order to evaluate ‘MSBR safety; these: constitute investi- gations which will be carried out in the MSBR Safety Program.. ~ * The’ favorable safety characteristics of MSBR systems arise from the 1bw excess react1v1ty-available,to,the reactor, the prompt negative tem- pefature‘doefficient*of'reaCtivity,ithe low system pressures, the low level of f1551on gases and. fission products retained w1th1n the reactor, . the mobility of fluid fuel, and the ease of fuel drainage from the re-. actor. At the same tlme,_there are a number of p0551ble incidents and 3l safety aspects which need detailed investigation; these aspects are re- lated to the specific plant design and involve both mechanical and nuclear design features. Plant systems which have a major influence on MSBR f reactor safety are the reactor system proper, the steam system, the fuel- recyclefiprocessing system, the coolant systems, and the off-gas system These are described above (Section 2), along with safety features that vere incorporated in the plant de51gn , Safety analysis requires a study of possible incidents, their conse- quences, and their avoidance.A,TypeSIOf accidents whichkcan take place include those due to reactivity additions. Reactor behavior under such circumstances. is influenced by reactivity coefficients and kinetics param-- eter values. Reactivity coefficients which will be considered include those associated with temperature, voids, pressure, fuel concentration, and graphite concentration, and involve the fuel and blanket fluids . separately and together. The function and design of control rods will be fully investigated; these studies will deteiudne the number, reactivity 1 worth, placement, and response requirements of control rods, as well as the ability to utilize blanket salt as a control rod. Possible reactor incidents will be evaluated as to their probability and their consequences, also, the influence of design changes (including alternate core designs) on safety aspects will be obtained. Under normal operating conditions, the MSBR should be load-following and self-controlling because of the prompt, negative temperature coeffi- cient of reactivity associated with the fuel salt. The temperature coef- ficient also protects against excessively high reactor temperatures and pressures in case of reactivity incidents. This situation is partially due to the large inherent neutron source strength present in-the fuel salt (nearly 10’ n/sec due to the a,n'reaction),'which permits thertem- perature coefficient to become effective as & reactivity cofitrol agent soon after initiation of rate additions of reactivity. ' Reactivity additions and their safety implications which will be considered in detail involve: breakage of graphite fuel tubes and the possible net fuel addition to the core region; other types of graphite . | behavior; changes in fluid-flow conditions; changes in fissile con- centration within the fuel fluid; abrupt changes in fission product “t3 wi alk 25 concentrations;'chahge-in control rod position; and the effect of pressure increases on reactivity. Relative to graphite tubes, & study of the cred- ibility and the consequences of single and multiple failures of graphite tubes in the reactor Wiil be made. ' The integrity of plant containment under both reactivity incident conditions and finder circumstances where reactivity itself is not involved will be evaluated for & number of physical possibilities; these include events such as mixifig of coolant salt with water or steam; spills of fuel or coolént salt and associated thermal, chemical; corrosion, -and criticality affects; temperature changes due to afterheat generation; container damage due to high temperature and/or corrosion; criticality in regions outside the core; flow blockage within the fuel or coolant streams; and blockage of flow in the off-gas system. The consequences of credible accidents will bé determined in all cases. . The application-df pressure-suppression systems. to molten-salt reactor plants will be 1nvestigated, and-problems'associated with their use will be analyzed. Additional design studies will be performed to better define such systems and their operation in detail. There are a number of areas which will be investigated experimentally in order to determine the general safety problems of molten-salt reactors. Some of these are closely related to areas studied as-part of the engineering-development and research programs of the MSBR Program. These include determination of the effects of reactor operating conditions on the physical behavior and properties of graphite, of graphite-~-to-metal joints,“and-of'Hasteiloy‘N., Tfiéilong-term creep properties of Hastelloy N and graphite need to be xnown end understood; also, the physical and éhemicalJproperties‘of'Saltsééfid‘of salt-water mixtures need to be known. . In addition, the ability to drain the fuel from the core under credible | conditions,needs study;-alsofthe'desirébility of alternative'core-designs: --relatiVE“to=afterheat-removai{will:be evaluated. 1~*Experimental-informationfwill5berobtained'on salt permeation of grephite, fission-product deposition in reactor systems, the ability to remove fission products from surfaces, and the ability to remove after- heat generated in the5fuel'salf;“fExpérimentsrwill be performed concerning 36 - the release of fission gases from solid as well as molten fuel salts, end concerning the fission-product flow and inventory throughout the - reactor.system. Retention of - fission products as well as tritium in off-gas systems will be demonstrated. - Measurements will be made concerning the conditions required to. produce "steam explosions' when molten salt and:water are mixed., The release of fission pro&ucts from fuel salt upon mixing it with water, or with oil, also will be measured. Investigations will elso be. performed concerning the reactivity effects associated with prec1pitation of uranium, rapid movements of fission products, "cold slug" accidents,'fuel-lesksge-into'thevblanket region via a plennm chamber, boiiing of blanket salt within the core | region, and buckling of a fuel-plenum.wall with associated change in graphite distribution. ' A stability analysis of the reactor plant systems will be made to determine the operating, control,,and/or design requirements for obtain- ing satisfactory plant characteristics. Items to be considered are time- lag events, spatial-distribution effects, the effect of fuel tube oscillations upon reactor behavior, and the relative importance of various parameter values upon system behavior. 4,2. Cost Estimates The planned safety studies are to resolve the basic safety problems associated with MSER plants. This means that enough information will be obtained to know which problems are the most important ones and how they can be overcome or eliminated (e.g., by chenging either the reactor de- sign or methods of operation). These studies will also provide experi- mental information which is necessary in order to resolve safety problems. On this basis, the cost estimates required to carry out the investigations indicated above are those given in Table 1. These estimates take into consideration the efforts planned in other parts of the MSBR Program . which are related to reactor safety, but do not include costs of such studies. However, the MSBR safety program depends on these other in- vestigations for major contribufions. Information which will be obtained Ty Table 1. Cost Estimates for the MSBR Safety Program%. Cost, in millions of dollars' "fInvéstigations ' — : —_— o S “FY 1968 1969 1970 1971 1972° 1973 1974 1975 Total Reactivity-related = 0.05 0.05 0.05 0.05 -- -- - - 0.20 events - ‘ Physical and chemical @~ 0.10 0.1 0.05 0.05 0.05 0.05 0.05 0.05 0.50 behavior of materials Equipment-failure | 0.10 0.1 0.10 0.05 0.05 0.05 0.05 0.05 0.60 -events : o : Total 0.25 0.30 0.20 0.05 0.10 0.10 0.10 0.10 1.30 *Does not include costs of safety studies carried out as part of other programs. Le 38 from other parts of the Program include the physical and chemical prop- erties of salts and structfiral materials, the characteristics of pressure- suppfession systems and containment structures, and the behavior for relisbility 6f reactor components. | | The safety program outlined above includes the specific mathéfiatical and physical formulation of the various problems, the compilation and evaluation of parameter values, and-determination of reactor plant be- havior under the postulated conditions. Experimental studies will be performed in conjunction with other MSBR investigations, which will in- volve modifying or initiating nevw experiments so as to give pértinent 'safety information. The objective is to détermine design and cperating conditions which are compatible with reactor safety and economic power production; The present MSBR design would serve as a starting point in these studies; however; general safety information related to molten- salt reactors will be obtained as problems become more clearly defined. ACKNOWLEDGMENTS The assistance of E. S. Bettis, T. W. Kerlin, W. B. Cottrell, S. J. Ditto, and J. 0. Kolb in providing information for this report is gratefully acknowledged. Thanks are also given to R. B. Briggs and R. C. Robertson for their review of this report. T -« ) REFERENCES P. R. Kasten, E. S. Bettis, and R. C. Robertson, Design Studies of 1000-Mw(e) Molten-Salt Breeder Reactors, USAEC Report ORNL-3996, ‘Oak Ridge National Laboratory (August 1966). W. L. Carter and M. E. Whatley, Fuel and Blanket Proce381ng Develop- ment for Molten Salt Breeder Reactors, USAEC Report ORNL-TM-1852, Osk Ridge National Laboratory (June 1967). S. E. Beall et al., MSRE Design and Operations Report, Part V, Reactor Safety Analysis Report, USAEC Report ORNL-TM-T752, Oak Ridge National Laboratory (August 1964). Q. C 1-50. 51. 52, 55. 54. 55. 56. 57 58. 59. 60. 61. 62. 63. 64, 66. 67. 69: 70, 1. 72- Th. T5. 76. T7. 78. T19. 80. 81. 82. 83. 8k. 85. 86, 87 . - 88. 89. 7 - 90. 91. 2. 9. 9L, Internal Distribution 41 MSRP Director's Office Room 325, 9204-1 R. K. Adams G. M. Adamson R. G. Affel L. G. Alexander R. F. Apple C. F. Baes - J. M. Baker S. J. Ball W. P. Barthold H. F. Bauman S. E. Beall M. Bender E. S. Bettis "F. F., Blankenship R. E. Blanco J. 0. Blomeke R. Blumberg E. G. Bohlmann C. J. Borkowski'_ G. E. Boyd ~J. Braunstein M. A. Bredig R. B. Briggs H. R. Bronstein G. D. Brunton D. A. Canonico S. Cantor W. L. Carter G. I. Cathers - ~J. M. Chandler. E. L. Compere W. H. Cook L. T. Corbin - J. L. Crowley F. L. Culler 'J. M. Dale ' D. G. Davis S. J. Ditto- A. S. Dworkin J. R. Engel E. P. Epler = D. E. Ferguson L. M. Ferris A. P. Fraas 95. 96. 97. a98. 99. 100. 101. 102. 103. 10L. 105. .106. 107. 108. 109. 110. 111. 112. 113. 11k, 115. 116-120. 121. 122. 123 124, 125. - 126, 127. 128. ..129_ 130. 131, - 132. 133, 134, 135. 136. 137. 138. 139, - 140, 1. 1k2. 143, 1hk. B' A. P. H. D. G. C. S. P. N. F. A. P. G. J. R. H. W. R. W. T. L. ORNL-TM-1858 Friedman Frye, Jr. . Gabbard . Gallsher Goeller | « Grimes . Grindell . Guymon ‘Hannaford Harley Harman Harrill Haubenreich Heddleson Herndon Hightower - Hoffman - Horton Hudson H. Inouye W. H. P. R. R. J. H. E. Ho F. C- _K- Jordan Kasten Kedl . Kelley . Kelly . Kennedy Kerlin . ‘Kerr Kirslis Krakoviak.r . Krewson . Lamb . Lane Lindauer . Litman Lundin . Lyon MacPherson MacPherson Martin " Mathews . Matthews MCClung ‘McCoy McDuffie McGlothlan ORNL-TM-1858 145. C. J. McHargue 171. O. L. Smith 146. L. E. McNeese 172. P. G. Smith 147. A. S. Meyer 173. W. F. Spencer 148. R. L. Moore 17k. 1I. Spiewak 149, J. P. Nichols 175. R. C. Steffy 150. E. L. Nicholson 176. H. H. Stone 151, L. C. Oakes 177. J. R. Tallackeon 152. P. Patriarca 178. E. H. Taylor 155. A. M. Perry 179. R. E. Thoma 154. H. B. Piper 180. J. S. Watson 155. B. E. Prince 181. C. F. Weaver 156. J. L. Redford 182. B. H. Webster 157T. M. Richardson 183. A. M. Weinberg 158. R. C. Robertson 184, J. R. Weir 159. H. C. Roller 185. W. J. Werner 160. H. C. Savage 186. K. W. West 161. C. E. Schilling 187. M. E. Whatley 162. Dunlap Scott 188. J. C. White *163. H. E. Seagren 189. L. V. Wilson 164. W. F. Schaffer 190. G. Young 165. J. H. Shaffer 191. H. C. Young 166, M. J. Skinner 192-195. Central Research Library 167. G. M. Slaughter 194-195. Document Reference Section 168. A. N. Smith 196-205. Laboratory Records (LRD) 169. F. J. Smith 206. Laboratory Records (LRD-RC) 170. G. P. Smith External Distributiomn 207-208. D. F. Cope, Atomic Energy Commission, RDT Site Office 209. A. Giambusso, Atomic Energy Commission, Washington 210. W. J. Larkin, Atomic Energy Commission,iORO 211-225. T. W. McIntosh Atomic Energy Commission, Wa.shi.ngton 226. H. M. Roth, Atomic Energy Commission, ORO. 227-228. M. Shaw, Atomic Energy Commission, Washington 229. W. L. Smalley, Atomic Energy Commission, ORO 230. R. F. Sweek, Atomic Energy Commission, Washington 231-245, Division of Technical Information Extension (DTIE) 246. Research and Development Division, ORO g 247-248. Reactor Division, ORO