Graphite and Xenon Behavior and Their Influence On

Home , Radiation damage

GRAPHITE AND XENON BEHAVIOR AND THEIR INFLUENCE ON KEYWORDS: molten-soh re- REACTOR DESIGN ocfors, breeder reocfors, rQ' MOLTEN-SALT diotion effecls, temPeroture, grophi te moderoto_r, _sfresses, DUNLAP SCOTT and P. EATHERLY oak Ri'd,ge -ing,porosity, xenon't15, Poison' bubbles, desig_n_, Nationat Laboratory, Nuclear Di,uision, Oak Ridge, Tennessee 37830 helium, ec-onotics, reoctor core, MSBR

Received August 4, 1969 Revised October 2, 1969

maintenance requirements; i.€., the downtime f€- quired for scheduled maintenance should coincide Existing data on di,mensional changes in gra- with that for graphite replacement. Hence, to graph- phite haue been to parabolic temperahrre' avoid load-factor penalties associated with fitted f €- sensiti,ue ct(r?)es. From these , the graphite life , ite replacement, we have set as a minimum two radiation-induced s tre s s € s, and, permissible quirement that the graphite have a life of geometries haae been calculated. It is concluded years; at the same time, a longer life would be Zxi,sting materials can be utitized in a molten-salt desirable and consistent with the power industry's reactor which has a core graphite life of abant objective of increasing the time interval between Thus, in the four years, without serious cost penalty. turbine maintenance operations. Fission product xenon can be remoued by reference design the reactor performance is con- sparging the salt with helium bubbles and strained to yield a graphite lif e of about four fuet ,paper basis that remoaing them after enrichment. With rea,sonable years, and this points out the for transfer coefficient and value. aalues of salt-to-btrbbte 135Xe puts graphite permeabitity , the Pernlty to breeding The removal of from the core also a graphite. The fuel salt must be ratio can be red,uced to l.. .0.5%. constraint on the excluded from the graphite to prevent local over- heating and also to decrease fission-product poisoning, and this, in turn, requires that the graphite pore diameters not exceed one micron. limitation, for the xenon INTRODUCTION This, however, is not a removal will be shown below to require a gas and One of the attractive a^speets of the molten-salt permeability of the order of 10-8 cm'/sec, 0.1 coneept is that even the most stringent of sugh a value requires pore diameters of - P. reactor graph- the present materials or process limitations per- Even though xenon is excluded from the mit reactor designs having aeeeptable economic ite, it needs to be removed from the salt stream performance. In this paper we conSider, aS if the desired neutron economy is to be attained. examples, the effects of finite graphite lifetime This removal is accomptished by iniecting helium and xenon poisoning on MSBR design. Finite bubbtes into the flowing salt, which transfers the graphite lifetime implies periodic replacement of xenon from the salt to the bubbles and effectively the graphite; neutron economy requires the f€- removes xenon from the core region. ttuxe discuss in moval of the bulk of from the core to keep In the following sections we shall existing the xenon poison fraction below the target value some detail: the method of analysis of data on radiation damage to permit prediction of of LVo' The major economic penalty associated with the graphite lifetime in MSBR cores; the applica- graphite replacement would be the load factor tion of these damage rates and the radiation- in penalty associated with taking the reactor off - induced creep to calculate induced stresses in the stream. This cost can be Circumvented by aSSUf - the graphite; the considerations involved 135Xe other noble ing that the graphite will maintain its integrity for distribution and removal of and at least the time interval between normal turbine gases; and last, the method proposed for safely 179 NUCLEAR APPLICATIONS & TECHNOLOGY VOL. B FEBRUARY 1970 Scott and Eatherly GRAPHITE AND XENON BEHAVIOR collecting and disposing of these gases. we con- and clude that graphite and the removal of xenon um -L2.0 + 8.92 x 10-3 present no questions of feasibility, but require = T Vo , only minor extensions of existing technology. where T is the temperature in oc, valid over the range 400 to 800o. At 700oC, these yield Qr - 3.1 x !022 with a goTo double-sided confidence GRAPHITE LIFETITUIE limit of *,0.2 x LO", and um = -5 .8Vo, with the limit t0 .2 . Figure 1 shows It has been recognized for several years that the behavior of the linear distor- under prolonged radiation exposure, graphite tion as a function of fluence with temperature as be- parameter. gins to swell extensively, even point a to the of As pointed cracking and breaking into fragments. However, out in the introduction, it is nec€s- sary that the graphite data at high fluences and within the operating exclude both salt and xenon. The requires graphite temperature ranges anticipated for molten-salt first the to have no pores larger than ,-,L breeder reactors were Iargely nonexistent. p in diameter; the second, as will be seen below, r€euires graphite Nevertheless, by using existing datq it was possi- the to have a permeability to xenon ble to estimate graphite behavior over the range of the order of 10-8 emz/ of MSBR conditions. For a large group bt sec. CIearIy, as the graphite expands and visible cracking (u commercial graphites - including British Gilso- occurs ) +3%ol, these requirements graphi,te, Pile Grade A, and the American grades will have been lost. Laeking definitive datq we AGOT and csF it was found that the volume have made the ad hoc assumption that the pore - size and permeability distortion, u, could be related to the fluence, O, requirements will be main- by a parabolic curve, tained during irradiation until the time when the original graphite volume is reattained, namely, u = Aa + Bez , (1) when t = r as defined above. where A and B ate functions of temperature only. To estimate the lifetime of an MSBR core, we The fit of this equation to the experimental data is musi take into aeeount the strong dependence of r excellent for the isotropic Gilso-graphite, but the relation is approached only asymptoticatly for the anisotropic graphites. we may write the fluence a^s the product of flux and time, i.€., O = Qt. The 850oc / 8000 ttsco behavior of u is to decrease to a minimal value, u*, and then increase, crossing the v = 0 axis at a defined ti,me, r, given by 0- AEr+B(W)' (21 Clearly, in terms of um and r, Eq. (1) can be rewritten as I o z0 u=4u*+(t (Zal 9 *) uF o For isotropic graphite, the linear dimensional F an changes will be given approximately by one-third 6 the volume change. If the graphite is anisotropic, u- _l the preferred c-axis direction will expand more zUJ quickly, the other directions more slowly. This f will induce a more rapid deterioration of the material in the preferred c-direction. As a consequence, we require that the graphite be isotropic, and can rewrite Eq. (zal in terms of the linear distortior5 G,

G=+u=++(t (zb) ) -3 and this relation will be used hereafter. 0r23 FLUENcE Ox rc-22 (E >50 kev) The best values for the parameters Qr and v* were found to be Fig. 1. Graphite linear distortion as a function of W = (9.36 - 8.93 x 10-3 f) x LOz'nvt (E >to keV) fluence at various temperatures. r80 NUCLEAR APPLICATIONS TECHNOLOGY VOL. 8 FEBRUARY 19?O Scott and Eatherly GRAPHITE AND XENON BEHAVIOR of the on temperafure, and the temperature of the core radius q,. Finally, since the heat capacity io keeping graphite will depend on the fuel salt temperatures, salt is a weak function of temperature, power density the heat transf er coeffieient between salt and with the sinusoidal variation of graphite, and the gamma, beta, and neutron heat along the a>ris, the salt temperature, T o, becomes generation in the graphite. In all core designs + (rr rr) cos (8) which have been analyzed, the power generation r o =*lr,aL r; T) (i.e., fission rate) has been found to vary closel.y and T7 are the entering and exiting as sin (n z/t ) along the urial centerline, wher e z/t in which Ti fractional height, and / is the effective total temperafures, respectively, of the salt. is the determine core height. Thus, the heat generation rate in the Equations (3) through (8) completely graphite. As we sha1l see graphite will varY as the temperatures in the below, the radiation-induced strain in the graphite (3) q= eo + Qt sin(nz/l), along the z-a)

Re is the ReYnolds number Thermal conductivitY, W/{cm "C) 3?.63 exp(-0.?f "K) Thermal e:rpansion, ("C)-r i.izx to-" + 1.0 x 1oe(r "c) Pr is the Prandtl number Youngts modulus, Psi 1.9 x 106 Poissonts ratio 0.27 Ko is the salt thermal conductivity. Creep constant, k (5.3-1.45x10-2z^+L.4 x 1o-5Tt; x 1o-'? the coolant channels at the outer We are assuming k in psi-t n/(cm' sec) surfaces of the cylinder will have the same effec- T in "C. tive hydraulic radius as the interior channel of r81 NUCLEAR APPLICATIONS & TECHNOLOGY VOL. B FEBRUARY 1970 Scott and Eatherly GRAPHITE AND XENON BEHAVIOR

papers. This design optimizes the fuel conser- since the salt-to-graphite ratios in the core vation coefficienta with the peak power density are determined by nuclear requirements, and the constrained to a value of 63 W/emt to prolong salt flow by cooling requirements, the only vari- graphite life. Case No. CC -24 is the identical able left at this point is the absolute value of the core scaled to a smaller geometry (half the internal radius a, which scales the size of the volume of No. CC-58) anO with the constraint on graphite prisms. Figure 2 shows the lifetime of graphite lifetime raised, In both cases the fuel the central graphite prism as it is affected by the conservation coefficient is 1b.1 [MW(th)/kg]2. The radius a; there i.s an obvious decrease in core life damaging flux in case No. cc -24 is also more as the radius a increases and the internal graphite typical of the case when the core is optimized temperatures climb. we note that the ratio of with no constraints. The pertinent core param- fltrxes in the two cases is 1.01 i at a = 0.9 cffi, the etersa are given in Table II. corresponding reciprocal ratio of lifeti.mes is L.74, the additional gain in lifetime at the lower flwr being due to decreasing graphite tempera- hlres. TABLE II The latter case has been studied in more detail Relevant core characteristics for calcutation of since corresponds Graphite Lifetime and Stresses it to the current reference design concept. For this design the equivalent Case No. Case No. radii are cc-24 cc-58 a = 0.762 cm (E Peak flux > 50 keV) , n/(enf sec) b.1b x lot{ 3.2 x 1or4 Satt flow per unit area, g/(cm' sec) 1.39 x 103 g.ZL x 102 b =5.39 cm. Heat generation, detayed, W/cmg 1.16 0.?1 The associated temperature Heat generation, promS, W/ems 7 .L7 4 .39 distributions for the Salt inlet temperature, oC 550 550 central core prisms are given in Fig. g, and the oC Sal.t outlet temperature, 700 700 local lifetimes r(n as a function ot z/l in Fig. 4. The life of .the prism, i.€.; the minimal r(Tl, occurs at z/l = 0.55 and has a value of 4.1 y""", at 80Vo plant factor. The entire prism will chattg" length as given by the integral of the right-hand side of Eq. (9) over the length of the prism; this is shown a,s a function of time in Fig. b. The radial distortion for various times is shown in Fig. 6, and gives the prism a double hourglass L o = 3.2 x I0l" o -Qn shape toward the end of life. oA,d. (J fJ- = o- obs @ F 6 zoo MAXIMUM INTERNAL H3 o TEMPERATU FI lrl E, u.J d. F= INNER AND OUTER o 5.15 u d, (J x lrl SURFACE JOt CL TEMPERATURE = tl H ooo BULK SALT TEMPERATURE

0.60 0.75 0 .90 INTERNAL RADIUS, a (in cm) 500 0 0.1 0,2 0.3 0.4 0.5 0.6 0.7 0.g 0.g l. FRACTIONAL HEIGHT (z/tl Fig. 2. Core lifetime as a function of graphite prism dimensions for cores with peak damage fluxes Fig, 3. Temperatures associated with the central gra- of 3.2 and 5.15 x 1014 nv @ > 50 keV), In all phite prism as a function of vertical position cases the ratio of the radti, b/a, is 6.67. for the reference core design.

182 NUCLEAR APPLICATIONS & TECHNOLOGY VOL. 8 FEBRUARY 19?O Scott and EatherlY GRAPHITE AND XENON BEHAVIOR

\ bs

b d. 0 o _'l o tt) \ o \ ,/ oE J 56 \ r-lx \ lJ- J Fz. o- -2 bso @ 0 1234 F TIME AT 80% PLANT FACT0R (yeor) tFs \ P Fig. 5. Axial distortion of the central graphite prism \ I as a function of time for the reference design.

\ \-, / INTERNAL STRESSES IN THE GRAPHITE preceding section we have tacitly &s- 0.3 0.4 0.5 0.6 0.7 0.8 0.9 In the 0.2 thermal and radiation-induced HEIGHT (z/1) sumed that the FRACTI0NAL stresses developed during the lifetime of the of vertical graphite are not limiting. We turn now to validate Fig. 4. Local lifetime r(T) as a function the central position for the central graphite prism of the this assumption. We again consider reference core design. prism as in the preceding section, and have for the eonstitutive equations' + uAV, - The cost of replacing graphite will include both ei = + lo; - tt(6i+ 6e)l i,ro,+ "-,] material procurement and labor. We have esti- ,i,,; mated these costs based on an industry supplying where ten or more reactors. The total direction the order of e ; = strain in the i'th operating cost associated with gr.aphite replacement would amount to ''.'0.2 mi}I/k$h for a two- o; = stress life. year 1ife, or 0.10 mill/kwh for a four-year E = Young's modulus investment would also be required for the icapttal ratio replacement equipment and is included in the p = Poisson's capital cost estimates of Bettis and Robertson.3) h = secondary creep constant the cost penalty associated with graphite Thus, flttx replacement is not a crippling one' although it is A = damaging large enough to merit considerable effort on dots = time derivatives graphite imProvement. function defined from Assumittg pyrolytic impregnation, as discussed g = dc/dt = damage rate by McCoy et a1.,5 successfully excludes xenon Eq. (2b). from the graphite, three significant material In cylindrical coordinates, f = (r, 0rzl. rn the must requirements need to be met: the graphite absence of externally apptied stresses and be- be isotropic, have entrance pore diameters cause of the vanishing of Q at the' ends of the least as < 1 F,-as and have a radiation stability at core prism, the z component of strain will not be good the Gilso-graphite. AlI of these require- a function of r and 0 (pEl" strain). We shall ments can be met with existing graphite tech- assume k is a function of 7, but will not Let k or been met in a single (10) be inte- nology, although all have not O vary with r and 0. Then Eq. can gpapnite of the dimensions required. unquestion- in closed form. The resulting dimensional in lrated aury, there will be production difficulties changes in the Prism are producing such a graphite, but the prob- initially A,z a,a process eontrol rather than in G , (11) lems will be in za,o= = * = basic technology. 183 FEBRUARY 1970 NUCLEAR APPLICATIONS & TECHNOLOGY VOL' 8 Scott and Eatherly GRAPHITE AND XENON BEHAVIOR

oz "+ pAs,) + (13 i6(ago - h"sJ . ) It can also be shown that g =g(7) with an error ba not exceeding ZVo parameter I for the values of oz, interest. F The thermal stresses have the same interr€- oE F Iationships as the radiation stresses, and are tt', o given by J l-a b s, o(b) 0, o o" o = oo@) = o(b') d. l8 mos. and 24 mos. ozo = I1 p lTLr r(r)) e-e, (14) -3 we may substitute this result into Eq. (Lzl to give 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.g 1.0 0.g the murimum stress o, (b) throughout the entire FRACTI0NAL HETGHT (z/I) life of the central prism. For the reference design concept, the maximum stresses also occur at the point z /t 0. bb Fig. = 6. Radial distortion of the central graphite prism and are shown in Fig. 1. The stresses reach a as a function of vertical height for the refer- murimum at the end of life and amount to only ence design. Times are calculated at gaqo plant 490 psi. since factor. the isotropic graphite which is presumably to be used in the reference design would have a tensile strength in the range 4000 to where d is the average value of G over the cylindrical cross section. Hence the prism behaves locally as though it were at the average radiation distortion d. 500 // The tensile stresses are murimum at the outside surface of the cylinder, and are given by o, (b) = o, os (b) = o, (b) 400 / and =o E -c / oz= e -et ee' dt + oz N i-rr I:lg - s(r)l oe-P' $zl b X 3oo / with (J b EhQ rt-, B= tt) 2(L p) uIU i zoo / we may consider the initial stress o"o to be the ltJ thermal stresses introduced a.s the reactor is U ]L brought to power, and these anneal out expon€D- u tially with time because of the radiation-induced et,:) .|00 / creep. we are interested only in large times t, and if we set / Ag = gi - g(b) / and remember g is a linear function of time dG O= V go + gtt odt =f = , -60 +('+) 1234 then TIME AT 80% PLANT FACTOR (yeor) Ag = Ag'o + Agrt Fig. 7 . Axial or tangential surface stres s at z/t = 0.5b and Eq. (Lzl thus takes the asymptotic form as a function of time.

184 NUCLEAR APPLICATIONS & TECHNOLOGY VOL. 8 FEBRUARY 19?O Scott and Eatherly GRAPHITE AND XENON BEHAVIOR

23sg 5000 psi, there appears to be no reasonable proba- vxe = o.oo24 YI = 0.0617 graphite 2339 bility that these stresses can cause vxe = o'olll Yt = 0.0505 failure. I I I I

XENON.I35 BEHAVIOR Because of its high-neutron-absorption cross G to keep xenon out of the 4 ) section, it is important 135Xe ; graphite and also out of the fuel salt. The f eomes from two sources, directly from the fission 5 23 agl of uranium and indirectly from the B decay of the ./ fission products tellurium and iodine. Tellurium 3 probably exists as a free metal which is believed = 6" to be insoluble in molten satt and tends to migrate to the available surfaces such as the core vessel d= 2 / walls, the heat exchanger, the graphite moderator, g and even any circulating gas voids which may be = present. However, 135Te has a relatively short and it is conservative for our half-life,

FEBRUARY 1970 r85 NUCLEAR APPLICATIONS & TECHNOLOGY VOL' 8 Scott and Eatherly GRAPHITE AND xENoN BEHAVIOR salt, are determined by the heat transfer condi- method, the gas is injected into the throat of an tions required to eool the graphite. However, the inverted venturi which is formed in the annulus effective diffusivity and available volume of the between a teardrop shape (installed on the pipe graphite can be controlled during manufacture by axis) and the pipe wall. when tested with water, impregnating the surface with a thin layer of bubbles of -0.5 mm were produced, and we €X- material having low permeability. The base peet similar results when used with molten salt. stock graphite which would be used in the core The bubbles in the salt are allowed to recirculate will probably have properties which are dictated through the system several times before removal. by radiation damage and lifetime considerations The solubility of xenon in molten salt is such that and not those which affect xenon concentration in with helium-to-salt volume fraetions as small as the voids. However, as witl be illustrated below, *Vo, the bubbles can be recirculated -10 times be- a sealed surface will act as an effective barrier fore the xenon concentration in the bubbles gets to the transfer of xenon into the graphite interior high enough to significantly affect xenon poisoning. if the value for the diffusivity of the sealed sur- The final step in the xenon processing of the face can be controlled to (-0.25 x 10-8 cm' /see, salt is the removal of the bubbles, and this can be and the associated void volume is l%o. As dis- done with a centrifugal gas separator of a type cussed by McCoy et al,s experiments have been which was developed for another circulating fuel performed which indicate that the application of reactor.to Kedl has proposed LLrLz a model for de- such a sealant is feasible. with the above condi- scribing the migration of the noble gases in a tions of permeability and available void volume molten-salt reactor and for estimating the poison- met, the xenon poisoning in the core becomes ing resulting from this migration. This model controlled by its concentration in the salt and was evaluated in summary tuKra experiment using a therefore by our abitity to process the salt stream gas containing a tracer on the MSRE during for xenon removal. prenuclear operation; calculated results were in The salt will be processed for xenon removal good agreement with experiment. After the MSRE by inj ecting small bubbles of helium into the fuel had been at power, the ttlKe.operating calculated r€- salt stream and then removing them after they sults for poison were not in complete agree- have taken up a portion of the xenon present. The ment with experiment results; in general the rate of transfer of xenon to the bubbles is affected poisoning was less than expected for the case by the concentration of xenon in the salt, the total where the circulating bubble fraction was very surface area of bubbles, the mass transfer coef- small and about as expected for the larger bubble ficient of the dissolved xenon from the salt to the fractions. Studies are under way to determine the entrained bubbles, and the residence time of the source of the difference. bubble in contact with the salt. we can control all Figure 9 summarizes the predicted effect on of these except the mass transfer coefficient to the poison fraction of varying the volume flow of the bubbles, and a good value for the coefficient helium into the salt stream of the MSBR for two is difficutt to determine. A reviews made of different graphites in the core. Poison fraction existing data on rates of mass transfer to gas is the neutron absorptions in ttu xe relative to bubbles points to reliable equations for predicting fissile fuel absorptions. The values of the physi- mass transfer under certain simple conditions cal constants used are considered the most likely involving stationary liquids and rising bubbles. It was also found by analysis that the mass transfer to bubbles associated with turbulent tiquid in a 5 pipe could vary by as much as a factor of 6 de- MOLTEN-SALT BREEDER REACTOR- \ SINGLE FLUID I 000 Mw(e) pending upon the characteristics assigned to the E4 t r r I r' I bubble-salt interface. An experiment is under way MASS TRANSFER COEFFICIENT TO o= BUBBLES, 17 X l0-3 cm/sec to evaluate present l-l ? \ this effect; at a conservative Fr' DIFFUSION COEFFICIENT IN (J value of L7 x 10-3 em/ see at the lower end of the -XeGRAPHITE, AS SHOhlN E rJ- FRACTION OF BUBBLES REMOVED range for the mass transfer coefficient under 2 \\ \ \ ,ER CYCLE, r0% oz, furbulent conditions is being used in the caleula- v, \ BU }BLE SIZE. 0.5 rm AVERAGE I tions to estimate the xenon poisoning in the MSBR. \ \ \L- Hr I l0-s cm2lsec The surface area of a given volume of €n- ll =0 ,25 x l0-8 cn'/sF I trained bubbles varies as the inverse of the 0 spherical diameter and therefore the void volume 0 1 ? 3 4 s 6 7 8 9 t0 needed in the circulating fuel can be kept small if He INJECTI0N RATE (scfm) the bubbles are small. Methods of injecting the bubbles into the salt are under study and one in Fig. 9. Effect of helium injection rate sn 1351e poison particular appears promising.e In this latter fraction.

186 NUCLEAR APPLICATIONS & TECHNOLOGY VOL. 8 FEBRUARY 19?O Scott and Eatherly GRAPHITE AND XENON BEHAVIOR

gas into the fuel salt for the MSBR. The upper curve is for an uncoated maj or portion of the stream resistance to xenon for recycle. The remainder of the stream is fur- base graphite where the main longer- the fluid flow boundary and not in ther processed for the removal of the transfer is in helium can The lower curve represents the ad- lived fission products so that the clean the graphite. (i.e', the gained by surface impregnation of the be used in the helium purge stream along vantage is described below. base graphite to a xenon permeability of ^.'0.25 X fuel pump shaft). This system flow diagram whi.ch illustrates 10-8 em' / see. One important difference between Figure 10 is a helium which the considerations involved in the off-gas system the two curves is in the volume of from get desired results. The design. Ten percent of the total salt flow must be handled to the the lower curve are such that only the primary salt pump is bypassed through conditions of the the gas gas flows are needed to reach the r3s;g" removal system, which includes low helium generator' About 9 scfm 0.5V0 Xe poison fraction' However, it separator and the bubble target of products is removed from should be noted that some helium processing is of helium with fission passes through the bubble separator. required even if a good coating is obtained. the salt as it This g&s, together with some salt' goes to the entrainment separator where the salt is removed OFF-GAS SYSTEM and returned to the primary circulating system The remaining The removal of t'35xe from the eirculating fuel through the pump expansion tank. also remove many g&S, [ogether with '-'2 scfm of purge gas from the by sparging with helium will particle fission products including the other noble pump expansion tank, then goes to the other pipe lines from the gases together with some of the relatively noble l""p- and 1-h decay tank. The products. Altogether these repre- bubble separator to and including the entrainment metal fission primary salt; radiation and heat source. It is separator are cooled by entrained sent a significant flowing in an purpose of the off-gas removal and disposal in addition a secondary coolant salt the the and store these materials annulus around the section of gas line from system to safely collect particle trap will to permit recycling the helium. The present v€f - entrainment separator to the - At the the off-gas system removes the helium remove the heat released to the pipe wall. sion of decay tank the entrained and certain fission products from the circulating particle trap and l-h ffid de-entraing any salt which fission-product particles are removed and allow salt stream, krypton isotopes to may be carried along; separates the noble gas the short-lived xenon and helium, ffid holds the 135 Xe outside the decay for t h before entering the absorption from the up for 4? h' reactor for a total of 48 h, and fhen reiniects a charcoal beds where the xenon is held

2 scfm CLEAN PURGE HC PRIMARY SALT PUMP 11 scfm ENTRA I NMENT HC SUPPLY SEPARATOR 18 MW ACCUMULATOR 3.2 MW

DECAY SYSTEM 47 h, Xe PR I MARY ' HOL DUP SALT CHARCOAL SEPARATOR REACTOR BUBBLE VESSEL GENERATOR

COMPRESSOR

85 Kr AND 90 day TRITIUM HOLDUP REMOVAL CHARCOAL SYSTEM BED

Fig. 10. Molten-Salt Breeder Reactor Off-Gas System' r87 NUCLEARAPPLICATIoNS&TEcHNoLoGYvol,.SFEBRUARYl9T0 scott and Eatherly GRAPHITE AND xENoN BEHAVIOR

The decay of the noble gases in the l-h volume 1. Radiation damage in graphite will timit delay tank reduces the heat load on the head end of core life, but existent materials are adequate to the charcoal bed by a factor of 6, making it easier provide a eore life of the order of four to five to keep the temperature of the charcoal 1ow years without excessively penalizing reactor per- enough to maintain an acceptable adsorption ef- formance. Although a specific grade of graphite ficiency. In the decay tank the fission-product that fulfills all our requirements is not currently particles will be removed from the gas stream available commercially, only minor extensions by washing with a liquid coolant which will also of existing technology, including the experience cool the gas and all surfaces of the tank walls. factor, are required to produce such a material. This coolant stream will be pumped through a heat exchanger where the heat wiII be transferred 2. under the conditions existing in the central to a closed steam system which in turn dumps its region of the core, stresses induced internally in heat to a natural-draft air-cooled condenser. the graphite are small and do not limit the per- About 18 Mw of energr is released in the decay missible graphite exposure based on the present design. tank. one of the primary obj ectives in the design of this system is to make the heat-rejection sys- 3. Although not specifieally discussed here, tem independent of external power needs such that we have had sufficient experience with pyrolytic a power failure will not jeopardize the system. carbon impregnations of graphite to feel confident The coolant stream will be processed to keep the the process can be developed for use with MSBR concentration of fission products below the satu- graphite. The times and process conditions used ration level to prevent deposition on the wal1s. to impregnate small graphite samples are practi- The gas goes from the decay tank to the char- cal for scale-upr and in at least one case the per- coal adsorption beds where the xenon is held up meability of the graphite was not significantly for 47 h and the krypton is held for 4 h. This affected by neutron i,rradiation. holdup is by adsorption of these hearry gases on the charcoal which does not delay the helium 4. Xenon- 13 5 can be removed successfully carrier gas. The total delay of 48 h in the two from the circulating salt by sparging with helium holdup sections is sufficient to reduce the ttlKe bubbles. However, several parameters such a"s concentration returning to the circulating salt salt-to-gas mass transfer coefficients are only system to Ti%o of its value when it left the salt. approximately known. Nevertheless, it appears parameter About 3.2 Mw are released in this adsorption that the values are in a range where holdup system. the sparging will be effecti.ve. After leaving the charcoal bed the flow splits 5. He1ium bubble insertion and extraction have into two streams. one system of ,.'.'z scfm goes to been demonstrated successfully in an aft/water the low-flow charcoal beds where the xenon is system. Although previous experience with mol- held for 90 days, after which tuKr most of the activity ten-salt systems has confirmed the ability to has deeayed. However, (10.4-year half-life) transform fluid flow results on water to a molten- and tritium (l2-year half-life) remain in the gas salt system, it remains to be demonstrated that stream and are removed by appropriate collector the helium-bubble/ saht system performance can beds using hot titanium bedstuKr for tritium removal be predicted using results from the aw /water and moleeular sieves for removal. The system. resulting clean helium is then compressed into the aceumulator for reuse in the cover-gas sys- 6. An off-gas system has been conceptually tem and the pump purge. designed which permits removal of the fission shaft products The other stream of ,--,9 scfm is simply eorn- from the helium, recirculation of the pressed and returned generator cleaned helium, ild control of the fission-product ttuxe to the bubbte for recycle to the removal system. heat. such a system appears practical, based on the capabilities of present technology.

SUTIMARY AND CONCLUSIONS

The previous sections indicate that, although ttuxe ACKNOilLEDGIIENTS radiation damage to the graphite and poisorr- ing of the core introduce design complexities into the reactor plant, accommodation of these factors This research was sponsored by the u.s. Atomic requires Energy Commission under contract with the Union neither significant new technology nor Carbide excessive penalties. Corporation. Oak Ridge National Laboratory is cost we specifically con- operated by the Union Carbide Corporation Nuclear Di- clude the following. vision for the U. S. Atomic Energy Commission.

188 NUCLEAR APPLICATIONS & TECHNOI,OGY VOL. 8 FEBRUARY 19?O Scott and Eatherly GRAPHITE AND XENON BEHAVIOR

REFERENCES 7. G. M. WATSON, R. B. EVANS, III, W. R. GRIMES' and N. V. SMITH, "solubility of Noble Gases in Molten 1. H. S. CARSLAW and J. C. JAEGER, Conl,uction of Fluorides ," J. Chem. Eng. Data,7, 2, 285 (1962). Heat in Solid's,2nd ed', Oxford, Clarendon Press (1959). 8. F. N. PEEBLBS, " Removal of Xenon-135 from 2. 'W. H. McADAMS, Heat Transmission, 3rd ed., Circulating FueI Salt of the MSBR by Mass Transfer to McGraw-Hill, New York (1954). Helium Bubbles," ORNL-TM-2245, Oak Ridge National Laboratory (1968). 3. E. S. BETTIS and R. C. ROBERTSON, "The Design and Performance Features of a Single-Fluid Molten-Salt 9. R. J. KEDL, "Bubble Generator,tt MSR Program 190 (1 9 7 0). Breeder Reactor ," Nucl . Appl . T ech. , 8, Semiannual Progress Report, Februaty 28, L969' ORNL- 4396, Oak Ridge National Laboratory (1969). 4. A. M. PE RRY , ' 'Reactor Physics and Fuel Cycle Ana1yses,' Nucl. Appl. Tech., 8, 20 8 (1970). ' L0. R. H. CHAPMAN, "HRE -2 Design Manual,t' ORNL- p. LL2, Oak Ridge National Laboratory (1964). 5. H. C. McCOY, W. H. COOK, R. E. GEHLBACK, TM-348, J. R. WEIR, C. R. KENNEDY, C. E. SESSIONS, R. L. 'W. R. J. KEDL and A. HOUTZEEL, of a BEATTY, A. P. LITMAN, and J. KOGER, "Materials 11. tsxe "Development for Molten SaIt Reactors," Nucl. ApPl. Tech.,8r, 156 Model for Computing Migration in the MSRE (1970). Graphite," ORNL-TM-4069, Oak Ridge National Labora- tory (1 96 7) . 6. C. B. BIGHAM, A. OKAZAKI, and W. H. WALKER, "The Direct Yield of Xe-135 in the Fission of U-233, L2. R. J. KEDL, "A Model for Computing the Migration U-235, Pu-23 9, and Pu-24L ," Trans . Am . Nucl . Soc ., of Very Short-Lived Nobte Gases into MSRE Graphite," 8, 11 (1965). ORNL-TM-1810, Oak Ridge National Laboratory (1967).

NUCLEAR APPLICATIONS & TECHNOLOGY VOL. B FEBRUARY 1970 r89