molten salt reactor chemistry

MOLTEN.SALT REACTOR CH EM ISTRY

R. GRIMES Oak Rid.ge National Laboratory,Oak RMge, Tennessee 37830

Received August 4, 1969Revised October 7, 1969

This document summarizes the large programof chemical research and deuelopment which ledto selection of fuel and coolant compositions forthe Molten-Salt Reactor Experiment MSRD and

for subsequent reactors of this ty|e. Chemicalbehauior of the LiF-BeF2-ZrFa-UFa fuel mixh,treand behaaior of fission products during Power op-eration of MSRE are fresented. A discussion ofthe chemical reactions which shout promise forrecoaery of bred 233pq and, for remoual of fissionproduct poisons from a molten-salt breeder reac-tor is included.

INTRODUCTION

A single-fluid molten- salt thermal breeder(MSBR) of the type described by Rosenthal et &1. rtBettis and Robertsonrz and Perry and Baumantmakes very stringent demands upon its fluidfuel.a-6 This fuel must consist of elements havinglow capture cross sections for neutrons typical ofthe energy spectrum of the chosen design. Thefuel must dissolve more than the critical concen-tration of f issionable material (" uU, "tfl, or"nPo), and high concentrations of fertile material("'Th) at temperatures safely below the tempera-ture at which the salt leaves the MSBR heatexchanger. The mixture must be thermally stable,and its vapor pressure needs to be low over anoperating temperature range ( 1100 to 1400"F)suff iciently high to permit generation of highquality steam f or portrer production. The f uelmixture must possess heat transfer and hydro-dynamic properties adequate for its service as aheat- exchange fluid. It must be relatively non-

KEYWORDS: fused solt fuel,chemicol reocfions, reocfors,beryllium (luorides, zirconiumfluorides, I i t h i u m (luorides,uronium hexolluoride, repro'cessing, protoctinium, sePoro-tion p roc e s s e s, breeding,fission products, MSRE

aggressive toward some othenvise suitable ma-terial of construction and toward some suitablemoderator material. The fuel must be stabletoward reactor radiation, must be able to survivefission of the uranium (or other fissionable ma-terial) and must tolerate fission product accumu-Iation without serious deterioration of its usefulproperties. We must also be assured of a gen-uinely low f uel- cycle cost; this presupposes alow- cost f uel associated with ine>rpensive turn-around of the unburned f issile material, andeffective and economical schemes for recovery ofbred fissile material and for removal of fission-product poisons from the fuel.

A suitable secondary coolant must be providedto link the fuel circuit with the steam-generatingequipment. The demands imposed upon this cool-ant fluid differ in obvious ways from those im-posed upon the fuel system. Radiation intensitieswiII be markedly less in the coolant system, andthe consequences of uranium fission wiII be ab-sent. The coolant salt must, however, be com-patible with metals of construction which willhandle the fuel and the steam; it must not undergoviolent reactions with fuel or steam should leaksdevelop in either circuit. The coolant should beinexpensive, possessed of good heat-transfer prop-erties, and it should melt at temperatures suit-able for steam cycle start-up. An ideal coolantwould consist of compounds which would be easyto separate from the valuable fuel mixture shouldthey mix as a consequence of a leak.

This report presents, in brief , the basis f orchoice of fuel and coolant systems which seemoptimum in light of these numerous-and to someextent conflicting- requirements.

CHOICE OF FUEL COilIPOSITION

Compounds which are permissible major con-stituents of fuels f or single-fluid thermal breeders

NUCLEAR APPLICATIONS & TECHNOLOGY VOL. 8 FEBRUARY 1970 137

Grimes MSBR CHEMISTRY

are those which can be prepared from beryIlium,bismuth, boron- 11, carbon, deuterium, fluoritr€,lithium-?, nitrogen- 15, oxygen, and the fissionableand fertile materials. As minor constituents onemight tolerate compounds containing the otherelements in Table I.

Many chemical compounds can be preparedfrom the several t'major constituents" listedabove, Most of these, however, can be eliminatedaf ter elementary consideration of the fuel re-quirements.n-u No hydrogen- (or deuterium-)bearing compounds possess overall propertiesthat are practical in such melts. Carbon, nitro-g€D, and oxygen form high melting binary com-pounds with the f issionable and f ertile metals;these compounds are quite unsuitable as constit-uents of liquid systems. The oxygenated anionseither lack _ the required thermal stabitity (i.e .,NOr- or NOr- ) or fail as solvents for high concen-trations of thorium compounds (i.e . , COr=; . Itquickly develops, therefore, that fluorides are theonly suitable salts indicated in this list of ele-ments.

Fluoride ion is capable of appreciable neutronmoderation, but this moderation is by itseU in-suff icient f or good neutron thermalization. Anadditional moderator is, accordingly, r€euired.

TABLE IElements or Isotopes Which may be Tolerable

in High-Temperahrre Reactor Fuels

The only good moderator material truly compati-ble with molten-fluoride fuel mixtures is graph-ite.a- 6

Phase Behavior Among Fluorides

Uranium tetraf luoride and uranium trif luorideare the only fluorides (or oxyfluorides) of uraniumwhich appear usef ul as constituents of molten-fluoride f uels. Uranium tetrafluoride (UF+) isrelatively stable, nonvolatile, and largely non-hygroscopic. It melts at 1035"C (1895"F), but thisfreezing point is markedly depressed by usefuldiluent fluorides. Uranium trifluoride dispropor-tionates at temperatures above - 1000'C by thereaction

4UFs= 3UFe + Uo (1)

It is unstableB" at lower temperatures in mostmolten-fluoride solutions and is tolerable in reac-tor fuels only with a large excess of UF+ so thatthe activity of tf is so low as to avoid appreciablereaction with moderator graphite or containermetal.

Thorium tetrafluoride (ThF4) is the only knownfluoride of thorium . It melts at 11 1 1oC (203 2"F)but fortunately its freezing point is markedly de-pressed by fluoride diluents which are also usefulwith UFe.

Consideration of nuclear properties alone leads_one to prefer as diluents the fluorides of B€, Bi,'Li, Mg, Pb, and Zr in that order. Equally simpleconsideration of the stability of these fluoridest'ntoward reduction by structural metals, however,eliminates the bismuth fluorides from considera-tion. This leaves BeFz ?.nd tLiF as the preferreddiluent fluorides. Phase behavior of systemsbased upon LiF and BeFe &s the major constitu-ents, has, accordingly, been examined in detail. r0

Fortunately for the molten fluoride reactor con-cept, the phase diagrams of LiF-BeFz -UFa andLiF-BeFz-ThFe are such as to make these ma-terials useful as fuels.

The binary system LiF-BeFz has melting pointsbelow 500'C over the concentration range from 33to 80 moLefio BeFr. to"t The phase diagram, pre-sented in Fig. L, is characterized by a singleeutectic (SZ moLeflo BeFz, melting at 360"C) be-tween BeFz and 2LiF.BeFz. The compound2LiF.BeFz melts incongruently to LiF and liquidat 458f. LiF.BeFz is formed by the reaction ofsolid BeFz and solid 2LiF'BeFz below 280'C.

The phase behavior of the BeFz - UFe 10' 11 and

BeFz-ThF41' systems are very similar. Both sys-tems show simple single eutectics containing verysmall concentrations of the heavy metal fluoride.ThFe and UF+ are isostructural; they form a con-tinuous series of solid solutions with neithermaximum nor minimum.

MaterialAbsorption Cross Section

(barns at 2200 m/sec)

Nitrogen-l5OxygenDeuteriumCarbonFluorine

BerylIiumBismuthLithium-7Boron-11Magnesium

SiliconLeadZitconiumPhosphorusAluminum

HydrogenCalciumSulfurSodiumChlorine-37

TinCeriumRubidium

0.0000240.00020.000 570.00 330.009

0.0100.0 320.0330.0 50.0 63

0.130.L70,180.2L0.23

0.330.430.490.530.56

0.60.70.7

r38 NUCLEAR APPLICATIONS & TECHNOLOGY VOL. 8 FEBRUARY 1970

900

848

800

\ 700

400

300

200LiF 40 50 60

BeF" (mole7")

Fig. 1. The system LiF-BeFz.

The binary diagrams LiF-UFno and LiF-ThFelaare generally similar and much more complexthan the binary diagrams discussed immediatelyabove. The LiF-UF+ system shows three com-pounds (none are congruently melting) and a singleeutectic, at 27 moLeflo gF*, melting at 490"C. TheLiF-ThF4 system contains four binary compounds,one of which (3LiF'ThF4) melts congruently, withtwo eutectics at 570"C and 22 mole%o ThF4 and at560'C and 29 moLe%o ThF+.

The ternary system LiF-ThF+-UF4,15 shown inFig. 2, shows no ternary compounds and a singleeutectic freezing at 488'C with 1.5 mole%o ThF4and 26.5 mole%o UF+. Most of the area on the dia-gram is occupied by the primary phase fields ofthe solid solutions UFe- ThFa, LiF'4UF+- LiF'4ThFa,and LiF.UF+- LiF.ThF+. L i q u i d u s temperaturesdecrease generally to the LiF- UF+ edge of thediagram.

The single-fluid molten-salt breeder fuel willneed a concentration of ThF4 much higher thanthat of UF+. Accordingly, the phase behavior ofthe fuel will be dictated by that of the LiF-BeFz-ThF4 system. Figure 3 gives the ternary systemLiF-BeFz-ThFei this system shows a single ter-nary eutectic at 4? mohe%o LiF and 1.5 moLe%o ThFa,melting at 360"C.10'rr The system is complicatedto some extent by the fact that the compound


BeFa

3LiF'ThF4 can incorporate Be'* ions in bothinterstitial and substitutional sites to form solidsolutions whose compositional extremes are rep-resented by the shaded triangular region nearthat compound. Liquidus temperatures( 1022"F) are available at ThF4 concentrations ashigh as 22 mole%o. The maximum ThF4 concen-tration available at liquidus temperatures of 500'C(932"F) is seen to be just above 14 moLeflo. Inspec-tion of the diagram reveals. that a considerablerange of compositions with > 10 mole%o ThF4 willbe completely molten at or below 500oC.

As expected f rom the general similarity ofThF4 and UF+-and especially from the substitu-tional behavior shown by the LiF-UF+-ThFe sys-tem (Fig. z)-substitution of a small quantity ofUF+ for ThFe scarcely changes the phase be-havior. Accordingly, and to a very good approxi-mation, Fig. 3 represents the behavior of theLiF- BeFz - ThFe - UF+ system over concentrationregions such that the mole fraction of ThF4 ismuch greater than that of UFe.

Oxide Fluoride Equilibria

Phase behavior of the pure fluoride systemLiF- BeFz - ThF+- UFa, &s indicated above, is suchthat adequate fuel mixtures seem assured. The

()o- 600lrjE.lF(rlrl

3 sooirjF

LiF LIQUID

\

\

555\

458

|- \ /

llF2 (Hrnl

Lo,rdfirz rYPE)

tl

I

LiF+ LitBeFo

I

\ J 360

Li2E eFo+3.F2 (HIGH QUARTZ TYPE)I

t-$C'cl(\J

280

.i.?.Fo

Li BeFo

L(l)(D:

LiBeFa +UiAer, + BeF. ( HIGH QUARTZ rVPe)

BeF2 (LOW,QUARTZ TYPE)\ | ZZO\

NUCLEAR APPLICATIONS & TECHNOI.OGY VOL. 8 FEBRUARY 1970 r39


ThF4,l

{,1,1

TEMPERATURE IN "CCOMPOSITION tN mote 7o

LiF.4ThF4

LiF.zThF4

LiF.rh%

P 5976 568:

3LiF.ThF46 565-

4LiF. UF4

Fig. 2. The system LiF-ThF4-UF+

behavior of systems such as this, however, ismarkedly affected by appreciable cbncentrationsof oxide ion.

when a melt containing only LiF, BeF2, anduFe is treated with a reactive oxide (such as H, o)precipitation of transparent ruby crystals of uoz.ooresults.u,t If the mert contaitrs, in addition, anappreciable concentration of ZrF+ the situation ismarkedly altered. zroz is less soluble than isUoz in such melts, and the monoclinic zroz (theform stable below ,\, 112b t) includes very litileuoz in solid solution. Thus, inadvertent oxidecontamination of a LiF-BeFz-ZrEa-UFe meltyields monoclinic zroz containing 2b0 ppm ofuoe. tu Precipitation of cubic uoz (containing asmall concentration of ZrOz) begins only af terprecipitation of Zroz had dropped the ZrF+ eon-centration to near that of the UFe.

slow precipitation of uoz followed by a suddenentrance of this material into the reactor corecould result in undesired increased in reactivity.This possibility was assumed to represent a dan-ger to the Molten- salt Reactor Experiment. Ac-cordinglv, the MSRE fuel was chosen to contain bmolefr of ZnFE to eliminate such a possibility.

PRIMARY-PHASE AREAS

(o) uro-rhFo (ss)

( b) lir. 4uF4-LiF-4ThFo (ss)

(c) r-iF.zrh(U)Fo (ss)

(d ) zr-i F. 6uF4-7LiF. 6Th[ (ss)(e) slir.Th (u) ro (ss)

(f) LiF

LiF.4u%uFq

,1035

When a mixture of LiF and BeFz containingThF4 and uF+ is treated with a reactive oxide ahomogeneous cubic phase is produced; this phaseis a solid solution of Uoz and Thoz which is veryrich in uor. tt careful studies have shown that thereaction

ThOz(ss) * t{ri= UOz(ss) * Thiil

/ Prrs\ lir.uro

(2)

[where the subscripts (ss) and (f) indicate thesolid-phase solid solution and the molten-fluoridesolution, respectively] approaches equilibrium withreasonable speed. values f or the equilibriumquotient Q for this reaction

Nuo 2(sq)' Nin*to(3)

Q) =Nr6o z(ss)' t'u*rO

increase with UOz concentration of the oxide phaseand decrease markedly with temperature. Sincevalues of e for mixtures similar to those chosenas fuel compositions are typicalty 800 to 1000, itis clear that oxide contamination of such salts willselectively precipitate the uranium.

APPLICATIONS & TECHNOLOGY VOL. 8 FEBRUARY 19?O

B %r bbo

140 NUCLEAR

zLrF' BeF,

L iF.ThF4

P 597E 5681

3LiF-ThF4

6 565:

?LtF'BeF; 400 450

Fig. 3. The system LiF-BeF2-ThFa.

It is likely, though not certain, that addition ofsome ZrF+would afford protection of the sort ob-tained with the MSRE fuel. Such addition is un-desirable, however, since the presence of ZrFqwould certainly complicate the separation pro-cesses described later in this paper and else-where in this series.

The successful operation of the MSRE over athree year period (discussed later) Iends con-fidence that oxide contamination of the fuel systemcan be kept to adequately low levels. This confi-dence, when added to the prospect that the breederfuel witl be reprocessed (and its oxide level re-duced) at regular intervals, suggests very stronglythat successful operation can be achieved withoutadded "oxide protection. "

Tolerance levels f or oxide concentration inLiF- BeFz -UFa and LiF- BeF z-ZrF*-UFe systemshave been studied in detail and are relatively wellunderstood.

16' t8- 20 Analogous values f or the LiF-BeFr-ThFa-UFe system are still largely lacking.It is known, however, that processing of thesequaternary melts with anhydrous HF and Hz s€rvesto remove oxide to a level below that required forprecipitation of the solid solutions. There seemsIittle doubt, theref ore, that initial processing of


Th% {t,tl

TEMPERATURE IN "C

COMPOSITION lN mole 7o

the type used for MSRE fuels can be successfullyapplied on a large scale to the LiF-BeFz-ThFe-UFe system.

I|ISRE and MSBR Fuel Gompositions

The fuel chosen for operation of MSRE (with a2s 5U- 238U isotopic mixture containing_ 33Vo of thefissionable isotope) was a mixture of ?LiF, BeFe,ZrFa, and UFe consisting of 65, 29.t, 5, and 0.9moLevo, f€spectively. The uranium concentrationwas fixed at '-, LVo so that there was less possibilityof fissile U precipitating (- 0.3 moLeflo

235U wasnecessary to achieve criticality and to provide a

small excess of f issionable material f or po\ileroperation of the machine). The ZtFq was added,as indicated above, to preclude possible inad-vertent precipitation of UOz. Beryllium fluorideis an extremely viscous material; its viscosity ismarkedly lowered by addition of LiF. The ratio ofLiF to BeFz in the MSRE fuel was chosen to opti-mize the confticting demands for low viscosity anda low liquidus temperature for the molten fuel.

The single-fluid breeder requires a high con-centration of ThFe; concentrations near 12

moheflo seem to be reasonable for good reactor

LiF.4ThF4

L iF.rh%3LiF'ThF4 ss.-

LiF.2ThF4

LiF.zrhF4

141

$N

NUCLEAR APPLICATIONS & TECHNOLOGY VOL. 8 FEBRUARY 1970


performance. Criticality estimates suggest thatsuch a fuel could be made critical in a practicablereactor with somewhat < 0.8 moLe%o ,ttuFn.

Theratio of tLiF to BeFz should, to d".""ase vis_cosity, be kept at a value as high as is practicable.If the liquidus temperature is to be kept at orbelow 500t (932'F) for a melt with L2 moheZo ofThF4, the beryIlium concentration limits rangefrom 16 to 25 moLe%o. The most likely choice forthe MSBR fuel-and the p_resent design composi-tion-is, ac co r dingIy, tLiF-BeFz-ThF4_UF+

at7 L.7 - 16 - L2- 0.3 moLeflo, r€spectively.

Choice of Coolant

The secondary coolant is required to removeheat from the fuel in the primary heat exchangerand to transport this heat to the power generatingsystem. In the MSBR the coolant must transportheat to supercritical steam at minimum tempe ra-tures only modesily above ?00oF. In the MSRE theheat was rejected to an air cooled radiator atmarkedty higher salt temperatures.

The coolant mixture chosen for the MSRE andshown to be satisfacto_ry in that apptication isBeFz with 66 moLe%o of tLiF. use of this mixturewould pose some difficulties in design of equip-ment for the MSBR since its liquidus temperatureis 851"F; moreover, it is an expensive material.The eutectic mixture of LiF with BeFz @amoLe%o LiF) melts at near ?00"F (see Fig. 1) but itis relatively viscous and is expensive, especiallyif t LiF is used.

The alkali metals, excellent coolants with realpromise in other systems, are undesirable heresince they react vigorously with both fuel andsteam. Less noble metal coolants such as pbo orBio undergo no violent reactions, but they

""" 1191compatible with Hastelloy-N, the Ni-based alloyused in MsRE and intended for use in MSBRrs.

several binary chloride systems are known tohave eutectics melting below (in some cases muchbelour) 7 00"F.2r These binary systems do not,however, appear especially attractive since theycontain high concentrations of chlorides (Tlcl,ZnCLz, BiCh, CdClz, or SnClr), which are easilyreduced and, accordingly, corlosive; or chlorides(AlCh , ZrCla, HfCla, or BeCtz ) which are veryvolatile. The only low-melting binary systems ofstable, non-volatile chlorides are those containingLict; Licl-cscl (330t at 45 moLe%o csct), Lict-KCI (3ss"c at 42 moLe%o KCt), LicI-RbcI (3 12t at45 moLe%o Rbcl). such sy_stems would be relativelyexpensive if made f rom t LiCl, and they could leadto serious contamination of the fuel if normal Liclwere used.

very few fluorides or mixtures of fluorides areknown to melt at temperatures below B?O"c. stan-

nous fluoride (snFz ) melts at zLz"c. This ma-terial is probably not stable during long termservice in Hastelloy-N; moreover, its phase dia-grams with stable fluorides (such as NaF or LiF)show high melting points at relatively low alkalif I u o r id e concentratiorls. coolant compositionswhich will meet the low liquidus temperature spec-ification may be chosen from the NaF-BeFz orNaF-LiF-BeFz system. These materials are al_most certainly compatible with Hastelloy-N, andthey possess adequate specif ic heats and lowvapor pressures (discussed in the next section).They (especially those including LiF) are moder-ately expensive, and their viscosities at lowtemperature are certainry higher than desirable.It is possible that substitution of zrE+ or evenAlFs for some of the BeFz would provide liquidsof lower viscosity at no rear expense in tiquidustemperature.

It now appears that the best choiee f or theMSBR secondary coolant is the eutectic mixture ofsodium fluoride and sodium fluoroborate. Thebinary system NaF-NaBFa has been described asshowing a eutectic at 60 moLe%o NaBF+ melting at304t (bB0oF).21'22 studies here have shown thatthis publication is seriously in error. Boric oxidesubstantially lowers the freezing point of NaF-NaBFe mixtures; the original authors may wellhave used quite impure materials. use of pureNaF and NaBFe leads to the conclusion that thesystem shows a single eutectic at g moLe%o NaFand a melting point of 3Bbt (7zbT')," as shown inFig. 4.

At elevated temperatures the f luoroboratesshow an appreciable equilibrium pressure of gas-eous BFs. The equilibrium pressure'n auove a

rooo

800

300

200NoF 20 40 60

NoBFo (mote%)80 NoBF-4

Fig. 4, The system NaF-NaBFn.

c)ov

700trJ(r:)k 600E.lrj(L

; sooF

\\ \

\\

142 NUCLEAR APPLICATIONS & TECHNOLOGY VOL. 8 FEBRUARY 19?O

melt maintained at the eutectic compositionmoLe%o NaF , 92 moLe%o NaBFe) is given by

Iog PTor, = 9.024 }?2| .

PHYSICAL PROPERTIES OF FUELS

AND COOLANTS

(B

(4)

Tables II and III list some of the pertinent phys-

ical propertiesza for MSRE and MSBR fuels and

secondary coolants.Many of the properties shown are estimates

rather than measured values. These estimateshave been carefully prepared from the best avail-able measurements on several salt mixtures of

similar composition. The values given are un-Iikely to be in error sufficient to remove the fluidfrom consideration. It is clear, however, from the

fact that estimates rather than measured valuesare shown that an experimental program must be

devoted to firming up the physical properties of

these materials.The densities were calculated from the molar

volumes of the pure components by assuming the

volumes to be additive. The heat capacities wereestimated by assuming that each gram atom in themixture contributes B calories per degree centi-grade. The value of 8 is the approximate averagef rom a set of similar f luoride melts. The vis-cosity of the MSBR fuel and the BeFz -based cool-ants were estimated from other measuredLiF-BeFz and NaF-BeFz mixtures.

The vapor pressure of the fuels and the BeFz

based coolants are considered negligible; extrapo-lation of measurements from similar mixtures

TABLE II

Composition and Properties of MSRE


TABLE III

Composition and Properties of Possible MSBR Secondary Coolants

aMean temperature of coolant going to the primary heat exchanger-bHighest normal operating temperature of coolant.

"Represents pressure of BFg in equilibrium with this melt compo-

sition.

yielded pressures ( 0. 1 mm. The partial pressureof BFs above the fluoroborate coolant mixture was

calculated f rom measurements on very similarmixtures.

CHEiIIICAL COMPATIBILITY OF MSBR iIATERIALS

Details and specific findings of the large pro-gram of corrosion testing are presented by McCoy

et al'u as a separate article in this issue. In brief,compatibility of the MSBR materials is assured by

choosing as melt constituents only fluorides that,insofar as is possible, are thermodynamicallystable toward the moderator graphite and towardthe structural metal, Hastelloy-N," & nickel alloycontaining '\'' IZVo Mo, |Vo Cr, 4Vo F€, IVo Ti, and

small amounts of other elements. The maior fuelcomponents (lif , BeFz, UFe , and ThF+) are much

more stable than the structural metal fluorides(NiFz, FeFz, and CrFz); accordingly, the fuel and

blanket have a minimal tendency to corrode the

metal. Such selection, combined with proper pur-ification procedures, provides liquids whose cor-rosivity'i" weII within tolerable limits. The

chemical properties of the materials and the

nature of their several interactions are describedbriefty in the following.

aHastelloy-N used in MSRE was|Vo Fe. Probable compositionfor MSBE is that shown above.

Ni with LTVo Mo, 7!s Ct,of modified HastelloY-N

Composition (mole%)

Li.quidu s TemPer ahtre :

ocoF

Physical tuoPerties atgso"F (454"C )"

Density , Lbfftg

Heat capacity,Btu/(lb'F)Viscosity, cP

Vapor pressure atttz5"r (60?'c),b mm

Thermal conductivitY'w/("C cm)

Cr Cz Cs

NaFNaBFe

8

92

385725

L2L

0.36

2.5

200'

0.005

LiF 23 |

NaF 4l I NaFBeFz 36 | ner,

328622

5743

340644

136 I

13e

0.47 I o.nnI40luu

Negligible I Nesrisible

Io.o1 | O.Ot

and MSBR Fuels

Composition(mole%)

Liqtidus"CoF

Properties at 6o0oC (1112"F)Density , g/ cmtHeat capacity, caL/(g'C)

or Btu/(tu 'r)Viscosity, cPVapor pressure' TorrThermal conductivitY,

w /("C cm)

MSREFuel

MSBRFueI

LiF 65BeFz zg.LZrFa, 5

UF+ 0.9

434813

2.27

0.47I

Negligibte

0.014

LiF 7 L.7BeFz 16ThF+ LzUF* 0.3

500932

3.35

0.3312

NegLigibte

0.011

NUCLEAR APPLICATIONS & TECHNOLOGY VOL. B FEBRUARY 19?O 143


Thermodynamic Data for Molten Fluorides

A continuing program of experimentation overmany years has been devoted to definition of thethermodynamic properties of many species inmolten LiF-BeFz solutions. Many techniques havebeen used in this study. Many of the data havebeen obtained by direct measurement of equilib-rium pressures for reactions such as

Hz(g) + FeFz(d)= Fef.; + 2HF1ry

zHF(g) + BeOlc)= BeFz(l) + HzO(g)

For the case where the totat uranium in the mol-ten solution of 0.9 moLe7o (as in the MSRE) theactivity of metallic uranium is near 10- tu withIVo of the UFe converted to UFs and is near2 x 10-10 with 20Vo of. the uFe so converted.2t op-eration of the reactor with a small fraction(usually 12Vo) of the uranium present as uF3 isadvantageous insofar as corrosion and the conse-quences of fission are concerned (see subsequentsections of this report). Such operation with someuFs present should result in f ormation of anextremely dilute (and experimentally undetectable)alloy of uranium with the surface of the containermetal. Operation withuFs would lead to much more concentrated (andhighly deleterious) alloying and to formation ofuranium carbides. Fortunately, all evidence todate demonstrates that operation with relativelylittle UFs is completely satisfactory.

Oxidation (Corrosionl of Metals

The data of Tabte Iv reveal clearly that inreactions with structural metals (M)

2UF4(d)+ M1c)=2UFs(d) + MFz(a), (B)

Cr is much more readily- attacked than is iron,nickel, or molybd.enum.2t Few thermodynamicdata exist for the fluorides of titanium in the purestate and none f or dilute sorutions in moltenfluoride solvents. Estimatess,e of free energies offormation suggest that ri is somewhat more reac-tive than C r. Titanium on the surface layers ofthe metal should, therefore, be expected to reactwith the available oxidants, such as uF+. suchoxidation and selective attack follow from reac-tions such as the following:

Impurities in the melt

Cr + NiFz+CrFz + Nior

Cr + 2HF ----+ QrFz + Hz .

Oxide films on the metal

NiO + BeFzlNiFz + BeO ,

followed by reaction of NiFz with Cr.Reduction of UF+ to UFs

(e)

(101

(1t1

and

fwhere (g), (c), and (d) represent gas, crystallinesolid, and solute, respectively] using the moltenfluoride as reaction medium. Baes has reviewedaII these studies and by combining the data with thework of others has tabulated thermodynamic datafor many species in molten ZLiF.BeFz.rG Table IVbelow records pertinent data for the major com-ponents of MSRE and MSBR fuels and for cor-rosion products in molten ZLiF.BeF2.

From these data one can assess the extent towhich UFs-bearing melt will disproportionate ac-cording to the reaction

4UFs(d)=3UF+(d) + U

(5)

(6)

(7)

TABLE TV

standard Free Energies of Formation for speciesin Molten zLiF . BeFz

"The standard state for LiF and BeFz is the moltenzLiF. BeFz liquid. That for MoFo(g) is the gas at oneatmosphere. That for all species with subscript (d) isthat hypothetical solution with the solute at unit molefraction and with the activity coefficient it would haveat infinite dilution.

Cr + 2UF+ = 2UFs + CrFz (rz1

Reactions (9), (10), and (11) above will proceedessentially to completion at all temperatureswithin the MSBR circuit. Accordingly, such reac-tions can lead (if the system is poorly cleaned) toa noticeable, rapid initial corrosion rate. How-ever, these reactions do not give a sustainedcorrosive attack.

(773 to 1000'K)

Materiala-LGl

(kcaL/ mole)-AG trooo"rl(kcaL/ mole)

LiF (r)BeFz 1l)

UFs 1ayUFa 16ythFe 1a)ZrF+ 61

NiFz 16y

FeFz 14;CrFz 1a;

MoFo (g)

L4L.B - 16.6 x 10-t ?"K243.9 - 90.0 x 10-t ?oK

338.0 - 40.8 x 10-3 ?.K445.9 - 57.9 x 10-t ?.K49L.2 - 62.4 x 10 -t ?oK453.0 - 6b.1 x 10-t ?oK

L46.9 - 36.8 x 10-t ?oKL54.7 - zL.g x 10-3 ?oKL7L. B - 2L.4 x 10 -3 ?.K3?0.9 - 69.6 x 10-t?oK

L25.210 6.9

99 .397.0

L07 .297 .0

55.366.57 5.2

50.2

14 NUCLEAR APPLICATIONS & TECHNOLOGY VOL. 8 FEBRUARY I.9?O

The reaction of UFe with Cf, on the other hand,has an equilibrium constant with a small tempera-ture dependence; hence, when the salt is forced tocirculate through a temperature gradient, a pos-sible mechanism exists for mass transf er andcontinued attack.

If nickel, iron, and molybdenum are assumed tobe completely inert diluents for chromium (as isapproximately true), and if the circulation rate inthe MSBR is very rapid, the corrosion process canbe simply described. At high flow rates, uniformconcentrations of UFs and CrFz are maintainedthroughout the f luid circuit; these concentrationssatisfy (at some intermediate temperature) tfre

equilibrium constant for the reaction. Under thesesteady- state conditions, there exists some tem-perature (intermediate between the maximum and

minimum temperatures of the circuit) at which theinitial surface composition of the structural metalis at equilibrium with the fused salt. Since theequilibrium constant for the chemical reaction in-creases with increasing temperature, the chromi-um concentration in the alloy surface tends todecrease at temperatures higher than ? and tendsto increase at temperatures lower than T. Fttsome melts (Naf-LiF-KF-gF+, for example) AG

for the mass transfer reaction is quite large, and

the equilibrium constant changes sufficiently as af unction of temperature to cause f ormation ofdendritic chromium crystals in the cold zone. ]For MSBR fuel and other LiF-BeFz-UF+ mixtures'the temperature dependence of the mass-transferreaction is small, and the equilibrium is satisfiedat reactor temperature conditions without the for-mation of crystalline chromium. Thus, ir theMSBR, the rate of chromium removal from thesalt stream by deposition at cold-fluid regions iscontrolled by the slo'w rate at which chromiumdiffuses into the cold-fluid wall; the chromiumconcentration gradient tends to be small, and theresulting corrosion is weII within tolerable limits.Titaniuffi, which must be presumed to undergosimilar reactions, diffuses less readily than does

Cr in Hastelloy-N. It seems most unlikely that theIVo of. Ti in the alloy witl prove detrimental.

The results of numerous long-term tests have

shown that Hasteltoy-N has excellent corrosionresistance to molten-fluoride mixtures at temper-atures well above those anticipated in MSBR.2s Theattack from mixtures similar to the MSBR fuel athot- zone temperatures as high as 1300"F is barelyobservable in tests of as long as 12 000 h. A fig-ure of < 0.5 miVyear is expected.zI In MSRE,

attack was - 0. 1 mll/year. Thus, corrosion of thecontainer metal by the reactor fuel does not seemto be an important problem in the MSBR.25

It is likely that the NaF-NaBFe coolant mixturewilt also prove compatible with Hastelloy-N, but


experimental study of this is much less extensive.The free energT change for the chemical reaction

BFg (e) + 3/2Cr (r) - 3/2CrFz (l) + B (") ( 13)

is - +30 kcal at 800oK.* The reaction is, there-fore, quite unlikely to occur, and similar reactionswith F€, Mo, and Ni are much less so. In addition,the above reaction becomes even less likely (per-haps by 5 kcal or so) when one considers theenergetics of formation of the compound NaBFeand dilution of the NaBF+ by NaF. Howev€f, re-actions which produce metal fluorides and metalborides are those which must be anticipated. Ti-tanium, a minor constituent of Hastelloy-N, has aquite stable boride.3o The reaction

2BFs(g) + *fi(c)= TiBz(c)+ gTiF+td) (14)

probably shows a small negative f ree energychange and must be expected to proceed. The verysmall diffusion rate of Ti in Hastelloy-N would be

expected to markedly limit the reaction. Thermo-chemical data for the borides of F€, Cr, Ni, andMo do not seem to have been established. Unlessthe borides of these materials are very stable(Ac/values more negative than -25 kcal B atom)the Hastelloy-N should prove resistant to pureNaF-NaBFe coolant.

Gompatibility of Graphite with Fluorides

Graphite does not react with molten f luoridemixtures of the type to be used in the MSBR.

Available thermodynamic datas's suggest that themost likely reaction

4UFe +C=CFe +4UF, (15)

should come to equilibrium at CFe pressures< 10-8 atm. CFe concentrations over graphite-saltsystems maintained for long periods at elevatedtemperatures have been shown to be below thelimit of detection (< 1 ppm) of this compound bymass spectrometry. Moreover, graphite has been

used as a container material for many NaF-ZtFq-gF+, LiF-BeFz-UF+, and other salt mixtures withno evidence of chemical instability.

CHETIIIICAL BEHAVIOR IN MSRE

The Molten- SaIt Reactor Experiment was, aF

detailed in an article by Haubenreich and Engeliloperated during much of 1966, 196?, and early1968 with the original fuel charge in which theuranium was 33Vo

235U with the balance consistingprimarily of "tU. The reactor accumulated nearly?0 000 IWWh of operation during this interyal withthe major fraction of this accumulated at the max-imum power of - 8 MW(th).



Behavior of Fuel Components

samples of the molten salts were removedroutinely from the fuel and coolant circuits andwere analyzed for uranium, major fuel constitu-ents, possible corrosion products, and (less fre-quently) for oxide ion contamination. Early in thepower runs, standard samples of fuel were drawnthree times per week; this schedule was markedtydiminished as confidence in the system grew. Atlate stages in the power runs, analyses for theseitems was generally done on a one-per-weekbasis. The LiF-BeFz coolant system was sampledevery two weeks initially and less frequenily dur-ing 1967 and 1968.

chemical determinations of uranium content ofthe fuel salt were run by coulometric titrationss ofdilute aqueous solutions. Such analyses showedgood reproducibitity and high precision (r0. iTo).(on-site reactivity balance calculations proved tobe about ten-fold more sensitive than this in es-tablishing changes in uranium concentrations with-in the fuel circuit.) Agreement of the chemicallydetermined uranium concentrations (in a statis-tical sense) with the reactor inventory valuesdiminished by the uranium burnup during opera-tion was excellent. The inventory value at end ofoperation on 235U showed the uranium concentra-tion to be 4. b3 270 by weight.33 The difference (200g of uranium out of 220 kg) is < 0. LVo. These datastrongly indicate that uranium losses (as to thepurge gas stream) have been extremely small andthat, as anticipated, the f uel salt is chemicallystable during long periods of reactor operation. Afurther check on these numbers will be possibtewhen the 235u- 238u mixture removed by fluori-nation f rom the MSRE f uel is recovered andassayed.

While analyses for UFe and for ZrF+ agreequite well with the inventory data, the values fortLiF and BeFz have never done so; analyses forLiF have shown higher and those for BeFz haveshown lower values than the inventory sincestart-up. Table v shows a comparison of typicalanalysis with the original inventory. value. This

TABLE V

Typical and Original Compositionof MSRE FueI Mixture

discrepancy in LiF and BeFz concentrationmains a puzzLe, but there was nothing in thealyses (or in the behavior of the reactor)suggest that any changes occurred.

Oxide concentration in the radioactive fuel saltwere performed by careful evaluation of Hz o pro-duced upon treatment of the salt samples withanhydrous gaseous HF. All samples examinedshowed less than 100 ppm of o=; no perceptibleincrease in the values with time or with opera-tional details was apparent.sa These facts arequite reassuring insofar as maintaining oxidecontamination at very low levels in future reactorsystems is concerned.

MSRE maintenance operations have necessi-tated flushing the interior of the drained reactorcircuit on numerous occasions. The salt used forthis operation consisted originally of a tLiF-BeFz(66.0 to 34.0 moleVo) mixture. Analysis of this saltbefore and after each use shows that ,\' z1b ppm ofuranium is added to the flush salt in each flushingoperation; this corresponds to removal of zZ.7 kgof fuel-salt residue (- 0.5/s of the charge) fromthe reactor circuit.

Behavior of Corrosion Products

Analysis of many samples at regular intervalsduring reactor operation showed relatively highvalues for ^r-ron

(r20 ppm) and nickel (so ppm) inthe salt.tt-3u These values scattered significanilybut showed no perceptible trends. They do notseem to represent dissolved Fe2+ or Ni2*. Molyb-denum concentrations were shown, in the f ewattempts made, to be below the detectible limit(- 25 ppm) for chemical analysis.

The chromium concentration, as determined bychemical analysis, rose from an initial value near40 ppm to a final value after ?0 000 vrwh of- 85 ppm. A considerable scatter (* lb ppm) inthe numbers was apparent but the increase withtime and reactor operation is clearly real. Allevidence suggests that the analytically determinedchromium was largely, if not entirely, present asdissolved crz*. This observed increase in chro-mium concentration corresponds to removal of< 250 g of this element from the MSRE circuit. Ifthis were removed uniformly it would deplete thechromium in the ai.loy to a depth ofsuch an estimate of corrosion seems consistentwith observations (reported elsewhere in thisseries) showing very slight attack by the f uelduring MSRE operation.

This virtual absence of corrosion-though ingeneral accord with results from a variety of en-gineering corrosion tests - is mildly surprisingsince the MSRE was operated f or appreciableperiods with the fuel salt more oxidizing than

APPLICATIONS & TECHNOLOGY VOL. 8 FEBRUARY 19?O

re-an-

to

ConstituentOriginal Value

(mole%) Typical Analysis

tLiFBeFzZrFqUFa

63.40 + 0.4930.63 + 0.55

5.L4 + 0.L20.82L + 0.008

64.3529.83

5.020.803

146 NUCLEAR

necessary. The analysis f or ratio of U4+ to Us +

has proved especiatly diff icult in the radioactivesalts, but it seems certain that the MSRE fuel has,at times, contained I 0.2V0 of the total uranium asIJ. +.

The fission process (see following section of

this report) appears to be mildly oxidizing towarddissolved U3+ in the fuel. Accordingly, and aI-though the corrosion product analysis did notseem to require it, a convenient means was de-veloped for adding U3+ to the reactor circuit asdesired. Beryllium metal (as 3 - in. rods of 3 - itt.diameter encased in a perf orated capsule ofnickel) was introduced through the sampling sys-tem and suspended in the flowing salt stream inthe MSRE pump bowl. This active metal reactswith UF+ by

Bet.;+ 2UFa16)- BeFz(d)+ 2UFs(d) (16)

at a rate such that some 600 g of UFs is producedduring an 8-h treatment. The salt flowing past theBe appears to be locally over- reduced so that atIeast 6Vo of the UF+ becomes UFs; crystalline Crwas observed upon the nickel capsule from one

such reduction attempt. The over- reduced saltmixture clearly reacted and achieved equilibriumwith the large excess of unreduced salt in thepump bowl and reactor circuit.

A molten-salt reactor which included a repro-cessing circuit could use that external circuit tomaintain the UF+/UFs ratio at the desired level"Use of techniques such as the Be addition for sucha reactor would seem to be quite f easible butunnecessary.

Behavior of Fission Products

When fission of UF+ occurs in a well-mixedhigh-temperature molten-fluoride, four fluorideions associated with the U4+ are released and the

two f ission f ragments must come quickly to a

steady state as common chemical entities" Thissteady state is made very complicated by theradioactive decay of many species. The valencestates that the fission product assume in the mol-ten system are, presum&bly, defined by the re-quirements that cation-anion equivalence be

maintained in the liquid and redox equilibria be

estabtished both among the components of the meltand the surface layeri of the container metal.n-uFluorine and uranium species higher than 4 ate,in MSRE and MSBR fuels, unstable toward reduc-tion by the active constituents of Hastelloy-N; the

f ission product cations must satisfy the f ourfluoride ions plus the fission product anions.Should they prove inadequate, of if they becomeadequate only by assuming oxidation states incom-patible with Hastetloy-N, then oxidation of UFs


must occur or the container must supply the de-ficiency. Operation of MSRE has indicated that thepresence of UFs in relatively low concentrations(see preceding sections) suffices to avoid oxida-tion or corrosion from this source"

Behavior of the fission products in MSRE wasstudied36-40 during the entire period of operationwith tt oU. Samples of MSRE f uel, drawn f rom thesampling station in the MSRE pump bowl, have

been routinely analyzed for numerous fission prod-uct species" Since samples drawn in open metalcups were shown (for reasons described below) togive erroneously high values f or gas- borne spe-cies, evacuated bulbs of Ni sealed with a fusibleplug of salt were used to obtain many of the datapresented here.

Samples of the helium gas purge within theMSRE were obtained by use of identical evacuatedsample bulbs in which the fusible plug was per-mitted to melt with the bulb exposed to the gasphase.37 t38

In addition to these salt and gas samples, as-semblies of Surveillance Specimens were exposedas the central stringer within the MSRE core.These assemblies included specimens of variousgraphites and many specimens of Hastelloy-N and

Some other metals. Their removal at reactorshutdowns permitted study of fission product de-positions after 8000, 20 000, and 70 000 MWh ofoperation.

Samples of salt removed from the pump bowlwere dissolved in analytical chemistry hot cells(after careful leaching of activities from the out-side of capsules where necessary); the f issionproducts were identif ied and their quantity de-termined radiochemically. Analysis of the gas

samples was accomplished similarly after carefulleaching of the activities from within the capsules.Deposition of fission products upon the metallicsurveillance specimens was determined in thesame way after repeated leachings of the metalsurface.

Deposition on the graphite specimens was es-tablished as a function of depth of penetration.This was accomplished by careful milling of suc-cessive thin layers from each surface of the rec-tangular specimens. The removed layers wereseparately recovered and analyzed for severalfission product isotopes.tt-no

The fission gases Kr and Xe form no com-pounds under conditions existing in a molten- saltreacto t,nt and these elements are only vgry.spar-ingly soluble in the molten f luoride.a 2 -44 Thehelium purge gas introduced into the MSRE pumpbowl serves to strip these gases f rom the in-coming salt to charcoal filled traps downstream inthe exit gas system. This stripping ensures thatthe fission product daughters of these gases will

NUCLEAR APPLICATIONS & TECHNOLOGY VOL. B FEBRUARY 1970 147


appear within the reactor system at lower than thetheoretical concentration. The MSRE graphite is,of course, permeable to Kr and Xe; radioactivedaughters of these gases are , accordingly, ex-pected in the graphite specimens.

The fission products Rb, Cs, Sr, B&, the lan-thanides and yttrium, and Zr aLL form quite stableand well-known f luorides which are soluble inMSRE fuel.nu'n6 These fission products were ex-pected (except to the extent that volatile precursorKr and Xe had escaped) to be found almost com-pletely in the molten f uel. These expectationshave been confirmed. Isotopes such as nuzr, ntSr,

and r43ce which have no precursors of consequenceare typically found in the salt at g}+Vo of the cal-culated quantity. Isotopes such as tnsr and tnoBa

whose Kr and Xe precursors have appreciablehalf-lives are found in less than the calculatedquantity; these isotopes are, &s expected, found inthe graphite specimens and the gas samples aswell.

The elements (molybdenum, niobium, ruthen-ium, and tellurium) whose higher valence fluoridesare volatile and relatively unstable toward re-duction by UFss'

re' 2s,3o,47 are virtually absent fromthe salt. Typical analyses of samples (which havebeen carefully protected from the pump bowl gasby use of the evacuated bulb samples) show12Vo of the calculated concentration of these iso-topes.

Deposition of fission products on the ,\, 7 - to10- mil outer layers of graphite specimens ex-posed for 8000 n[\Mh in MSRE is shown in Table vI.It is clear that, with the assumption of unif ormdeposition in or on the moderator graphite, ap-preciable fractions of Mo, T€, and Ru and a largerfraction of the Nb are associated with the graph-ite. subsequent examinations of graphite af ter22 000 and 70 000 Mwh have shown somewhatsmaller fractions deposited for all these isotopes.The data of Table vtr described below show valuesmore typical of the longer exposures.no

A careful determination of uranium in or on thegraphite specimens led to valuesThis quantity of uranium-equivalent to a fewgrams in the moderator stack- seems completelynegligible.

Figure 5 shows the change in concentration ofthe f ission product isotopes with depth in thegraphite. Those isotopes (such as tnoB") whichpenetrated the graphite as noble gases showstraight lines on the logarithmic plot; they seemto have remained at the point where the noble gasdecayed. As expected, the gradient for tnoBa with16- sec tnoxe precursor is much steeper than thatfor tnS", which has a 3.2-min tnKr pr;cursor. Allthe others shown show a much steeper concentra-tion dependence. Generally the concentration

TABLE VIFission Product Deposition on Surface*

of MSRE Graphite

*Average of values of 7- to 10-mi1 cuts from each ofthree o(posed graphite faces.

aPercent of total in reactor deposited on graphite if eachcm' of the 2 x L06 cmz of *od"tator had the same con-centration as the specimen.

drops a factor of 100 from the top o to 10 mils tothe second layer. More recent examinations whichused cuts as thin as 1 mil show the materials tobe concentrated in an extremely thin layer. Typ-ically gSVa of the metallic species are within3 mils of the surface.

TABLE VIIFission Product Distribution for MSRE at End of "tu operation

Nuclide

Quantity Found (Vo of. Calculated Inventory)

In FuelOn

Graphitea

OnHastelloy-

Na

InPurgeGasb Total

ttMottrTetreTetotR..,totRu

nuI.Ib

n'zrtnsrtaoBatotce

t*ce13rI

0.170.470.400.0330.10

0.001 to 2.2948396

60

9.05.15.63.54.3

4l0.148.51.90.33

0.920.11

199.5

11.52.53.2

T20.060.080.050.03

0.030.3

50743149

130

11

0.43L70.480.88

2.719

78894855

138

6495

10998

80

"Calculated on assumption that average values for surveillancespecimens are representative of all graphite and metal surface inMSRE.

bThese values are the percentage of production lost assuming meanvalues found in gas samples apply to the 4-Liter/min purge.

APPLICATIONS & TECHNOLOGY VOL. 8 FEBRUARY 1970

Isotope

Graphite Location

TopPercent of

Total a

MiddlePercent of

Total a

BottomPercent of

Tota1 a

*MotrrTetorRunuNb131I

nu zrt*c.*srtnoBatntc

etttc

s

L3.413. 8LL.4L2

0. 160.330.0523.241 .380. 19

0.07

L7.213. 610,359.20.3 30.270.273.301. 850.630.25

11.5L2.0

6.362.40.250.150.L42.74L.L40. 36

0.2L2

148 NUCLEAR

tol3

40t2

\

tt\\\\ffi

-----f-T \ \\.-

89st

\\\ \\ \\ \\ \ t

\\ \\

\ r32Te

-

I I

I \

*r\-

\ \\ \\ \\ \

\ \- lO3p,\ \\-

13{ 1


precursors. A small quantity of n'2, appears inthe gas phase; this isotope, along with severalothers with stable soluble fluorides, almost cer-tainly appears in the gas as fine particles of saltmist:se

- Ho*"ver, high concentrations of nnMo,

lt'Te , *I.Ib, totRr, and touRu are found in the gas

phase. Indeed, &s Table Vtr indicates, the gas

samples represent the largest fractions observedfor each of these isotopes except tuNb. In addition,appreciable f ractions of ttoAg and of palladiumisotopes have been observed in the purge gas

sarrrples. An appreciable quantity of 13rI appearsin the gas phase; it seems very likely that itsappearance there is due to s6volatilization" of itstellurium precursor.

The mechanisms by which the Mo, Nb, T€, andRu isotopes appear in the gas phase are still notfully understood. NbF5, MoF6, RuF5, and TeFu areall volatile (although Ag and Pd have no volatilefluorides); these volatile fluorides, however, arefar too unstable, with the possible exception ofNbF5, to exist at the redox potentials in MSRE.Stability of lower fluorides of these elements ispoorly known; it is possible that some of thesemay contribute to the "volatility. "

It seems much more likely, however, that aIIthese species exist in the reduced MSRE fuel inelemental form. They originate as (or very rap-idly become) individual metal atoms. They ag-gregate in the fuel at some slow but finite rate,and become insoluble as very minute colloidalparticles which then grow at a slower rate. Thesecolloidal particles are not wetted by the fuel, theytend to collect as gas-Iiquid interfaces, and theycan readily be swept into the gas stream of thehelium purge of the pump bowl. They tend to plateupon the metal surfaces of the system, and (as

extraordinarily f ine "smoke ") to penetrate theouter layers of the moderator. While there aredifficulties with this interpretation it seems moreplausible than others suggested to date.

Deposition of f ission products on or in thegraphite is, of course, undesirable since therethey serve most efficiently as nuclear poisons. Ofthe species studied only Nb (whose carbide isstable at MSRE temperatures) seems to prefer thegraphite.

It seems likely that in a MSBR the molten fuelwitl contain virtually aII of the zirconium, thelanthanides and yttrium produced, and a largef raction of the iodine, rubidium, cesium, stron-tium, and barium. Thess species would, therefore,be available for removal in the processing plant.The salt wiII contain very little Mo, Nb, T€, Ru,and (presumably) technetium. These will reportto the MSBR metal, grsPhite, and purge gas sys-tems in f ractions which are probably not verydilferent from those indicated for MSRE.

tot'

lrJLI(L

(r(9

Lo

=E(9

(rlrJ(L

UJFfz

=(rtdo-

azIFE,(,LdFzoo

to'o

rot

,oto{o20304050

DISTANCE FROM SURFACE OF GRAPHITE (mils)

Fig.5. Concentration profile of fission products inMSRE core graphite after 8000 MWh.

Examination of the Hastelloy-N surveillancespecimens exposed in the MSRE core reveals verylow concentrations of the stable fluoride-formingfission products such as tnS", n'zr, and tntCe. Themore noble metals such as Mo, Nb, and Ru aredeposited, along with T€, in appreciable though notin large quantities. If the surveillance specimensare truly representative of all the metal surfacein MSRE then (see Table VII) only a relativelysmall fraction of these metals are so deposited.However, flow conditions in the MSRE heat ex-changer (where most of the metal surface exists)is turbulent while laminar flow occurs over thecore specimens. It seems very likely that theactual fraction deposited on MSRE metal is higherthan that shown here.

Analysis of samples of gas from MSRE yieldedSome real Sulprises. Isotopes such aS tnsr andr4oBa are found in amounts consistent with con-sideration of the haU-Iives of their Kr and Xe



SEPARATIONS CHEII,IISTRV

Molten- salt breeder reactors will require ef-fective schemes for recovery of the fissionablematerial which is produced and for removal offission product poisons from the fuel. MSRE pro-duced no significant quantity of fissionable isotopeand made no provision for on-stream removal offission products other than the noble gases Kr andXe.

Recovery of uranium from molten fluorides byvolatilization as UFo and the subsequent decon-tamination of this UFo by sorption-desorption onbeds of NaF is weII demonstrated. Recovery ofuranium from MSRE or MSBR fuel by this schemeis clearly feasible. However, perf ormance of asingle-fluid MSBR is markedty improved if thebred protactinium is removed from the circulatingf uel and permitted to decay to 233 u outside thereactor core. Fluorination is not effective in re-moval of protactinium from molten fluoride mix-tures. Development of other processes for parecovery has , LCcordingly, been necessary.

In molten-salt reactors, from which Kr and Xeare effectively removed, the most important fis-sion product poisons are among the lanthanides.These fission products, &s indicated above, formstable fluorides which are soluble in the fuel andwill report almost quantitatively with the fuel tothe separations plant. Their removal especialtyfrom fuels with high ThF4 concentrations is diffi-cult. Rare earths have , accordingly, receivedmore attention than other fission products.

Processes f or recovery of pa and the lan-thanide f ission products f rom MSBR f uels aredescribed in some detail by whailey et al.a' rhechemical basis f or such processes is describedvery briefly below.

Separation of Protactinium

Protactinium is produced in the MSBR fuel byreaction of a neutron with " ' Th to vield 'tt Thwhich transmutes to "l_8"; the "tpa aelays with a27.!:day half -life to "tu. clearly, a separationof "tPa from the fuel needs, if it is to be effec-tive, to be accomplished on a cycle time shortcompared with this half-life. It is desirable toprocess the fuel on a cycle of three to five days.The process needs, theref ore, to be relativelysimple.

Removal of Pa from LiF-BeFz-ThF* mixturesby addition of Beo, Thoz, or even uoz has beendemonstrated in laboratory- scale eliperimentsjn' m

Addition of these oxides to MSBR fuel compositionwould, 3s indicated in a previous section of thisreport, result in a uranium-oxide-rich solid solu-tion of Uoz - Thoz. Whether the protactinium oxide

was adsorbed on the added oxide or participated inthe solid solution f ormation is not known. Theprocess was shown to be reversible; treatment ofthe oxide-fluoride system with anhydrous HF dis-solves the oxide and returns the pa to the fluoridemelt from which it can readily be reprecipitated.It seems likely that Pa might be removed fromthe MSBR fuel by effective oontact of a fuel mix-ture side stream with Uoz - Thoz solid solution.Such a possibility-which is still under considera-tion in laboratory- scale equipment-would addappreciably to the 233u inventory through holdup ofuoz in the solid solution. It would also probablyrequire that the oxide ion concentration in the fuelafter contact be diminished (as by treatment withHF) bef ore the melt could be returned to thereactor.

separations based upon reduction of the pa(present in the fuel as PaF+) to metal and extrac-tion into liquid metals appears at present to bemore attractive.as Such separations have the fot-lowing basis.

when thorium metal, dissolved as a dilute alloyin bismuth (fnu,1, is equilibrated with a moltenmixture of LiF, BeF2, and ThF4 several equilibriamust be satisfied simultaneously. These include

LiF16y + *ttr(Bi) = *fnfn(d) + Litei)

BeF2(d)+ *fn(Bi) =*ThFe + Betnil

* tnfn,u, * *Thtei) = ThFs(d) ,

(17)

( 18)

( 1e)

where (d) and (Bi) indicate that the species are influoride solution and alloyed with the Bi, respec-tively. Reaction ( 19) seems to be of litile im-portance since only a small quantity (it any) ThFgforms. Reaction (18) does not occur perceptibly;Be is 6'insoluble" in Biand forms no intermetalliccompounds with that liquid metal. Reaction (12;occurs to an appreciable extent even though LiF(see Table Iv) is much more stable than ThFa. le

uranium forms stable dilute alloys in Bi, andits f luorides are less stable than ThFa.le Accord-ingly, the reactions

UF+(d) + *fn(Bi) = *fnfn(d) + UFs(d)

UFs(d) + * ffr(Bi) = U(ni) * * ThFelay

and

(20;

(21)

proceed very far to the right with low concentra-tions of Th (and Li) in the Bi at equilibrium. uran-ium is, theref ore, very readily extracted f romMSBR f uel melts .5lrs2

If the LiF- BeFz - ThF4 melt contains paFa, thereaction

PaFe (d) + Thlsiy = ThF4(d) + Pa(si) (22)

r50 NUCLEAR APPLICATIONS & TECHNOLOGY VOL. 8 FEBRUARY 1970

proceeds to quite a convenient extents2 - 55 aS shownitt Fig. 6 although protactinium is reduced lesseasily than is uranium. Accordingly, if a meltcontaining both UF+ and PaF* is contacted withbismuth and thorium metal added in small incre-ments the uranium and the protactinium are ex-tracted"

Preliminary experimentsso disclosed that Pa

could indeed be removed from the salts by reduc-tion with Th but material balance data were verypoor and little Pa appeared in the Bi. More recentexperiments have used equipment of molybdenum,which seems essentially inert to the salt and the

alloy, and very careful _PulSication of the ma-terials and the cover gas.53-55 By these techniquestwo sets of investigators$'# have shown values inexcess of 1500 for the separation factor

Np. NrhFe(23)

Np.rn Nrn '

where N indicates mole fraction of the indicatedspecies either in metal or in LiF-BeFz-ThF4 so-lution. The separation factor for U from Pa has

been shown to be above 25. These separation fac-tors are, of course, quite adequate for proc€ssdesign, &s discussed in detail by Whatley et aI.aB

Fig. 6, Distribution of protactinium and material bal-ance for equilibrium of LiF-BeFr-ThF4 $2-L6-12 moleflo) with Bi-Th alloy (0.93 'fi70 Th).


Separation of Rare Earths

There is no real doubt that the rare- earthelements, which form fluorides that are morestable than ThF4le wiII be present in MSBR fuelsent to the processing Plant.

The limited solubility of these trif luorides(though suff icient to prevent their precipitationtmder normal MSBR conditions) has suggested a

possible recovery scheme. When a LiF- BeFz -ThFe-UF+ melt (in the MSBR concentration range)

that is saturated with a single rare-earth fluoride(LaFs, for example) is cooled slowly, the precipi-tate is the pure simple trifluoride. When the meltcontains more than one rare-earth fluoride theprecipitate is a (nearly ideal) solid solution of thetrif luorider.n u,nu Accordingly, addition of an ex-cess of CeFs or LaFs to the melt followed by

heating to effect dissolution of the added tri-fluoride and cooling to effect crystallization ef-fectivety removes the fission product rare earthsfrom solution. It is likely that effective removalof the rare earths and yttrium (along with UFs andPuFs) could be obtained by passage of the fuelthrough a heated bed of solid CeFs or LaFs. How-ever, the price is almost certainly too high; theresulting fuel solution is saturated with the sca-venger fluoride (LaFs or CeFs, whose cross sec-tion is far from negligible) at the temperature ofcontact.

Similar solid solutions are formed by the rare-earth trifluorides with UFs. It might' accordingly,be possible to send a side-stream (with the "tPaand 233U removed by methods described above)through a bed of UFs to remove these fissionproduct poisons; '"tUF, would presumably be usedfor economic reasolls. The resulting LiF-BeFz-ThF+ solution would be saturated with UFs afterits passage through the bed. This 238U would have

to be removed (for example, by electrolytic re-ductionas into molten Bi or Pb) bef ore the saltcould be returned to the cycle. The UFs bed couldbe recovered by fluorination to separate the ura-nium and rare earths. While this process probablydeserves f urther study, the instability of UFs inmelts with high UF3/UF4 ratios and the ease withwhich uranium alloys with most structural metalswould tend to make application of such a processunattractive.

Removal of rare-earth ions (and other ionicfission-product species) by use of cation ex-changers has always seemed an appealing possi-bility. The ion exchanger would, of course, need(a) to be quite insoluble, (b) to be extremely un-reactive (in a gross sense) with the melt, and(c) to take up rare-earth cations in exchange forions of low-neutron cross section. The bed ofCeFs described above f unctions as an ion ex-

Paorn =

?. 90og(l)CLv80

=lzF70(J

Fotr60JFPscE.oour 4Coo

d3cFUJ

=

=zc)tr.o-F t(

r5l

o PROTACTINIUM IN BISMUTHo PROTACTINIUM IN SALTA THORIUM METAL IN BISMUTH

FLOWING ARGON

STATIC ARGON FLOWING ARGON HELIUM

rllittti

85 105

TIME ( h)

NUCLEAR APPLICATIONS & TECHNOLOGY VOL. 8 FEBRUARY 1970


changer; it fails to be truly beneficial because itis too soluble in the melt.

unfortunately, not many materials are trulystable to the LiF-BeFz -ThF4-UF+ fuel mixture.Ztrconium oxide is stable (in its low temperatureform) to melts whose zrn*/If* ratio is in excessof -3, and Uoz-Thoz solid solutions are stable atequilibrium u/rn ratios. It is conceivable thatsufficiently dilute solid solutions of Cez o, inthese oxides would be stable and would exchangecet+ for other rare earth species. Intermetalliccompounds of rare earths with moderately noblemetals (or rare earths in very dilute alloys withsuch metals) seem unlikely to be of use becausethey are unlikely to be stable toward oxidation byuFe. compounds with oxygenated anions (such assilicates and molybdates) are decomposed by thefluoride melt; they precipitate uoz from the fuelmixture. It is possible that refractory compounds(such as carbides or nitrides) of the rare earths,either alone or in solid dilute solution with an-alogous uranium compounds, ffi&v prove useful.

By Reduction

The rare-earth fluorides are very stabletoward reduction to the metal. For example, at1000'f tne reaction

z/3laFs(l) + Be(c) * Z/gf,a(.) + BeFz(t), (24)

where (c) and (l) indicate crystaltine solid and liq-uid, respectively, shows - +30 kcal for the free en-ergy of reaction.le with the LaFs in dilute solutionand BeFz in concentrated solution in LiF BeFz mix-ture the situation is, of course, even more un-favorable. Howev€r, the rare-earth metals formextremely stable solutionsut in molten metals suchas bismuth; the activity coefficient for La at lowconcentrations in Bi is near 10-tn. Therefore, thereaction

3t atr'. (d) + Be1.; = Sl"tuil + BeF2(d) , (2b)

where (d) indicates that the species is dissolved inzLiF'BeFz, and (c) indicates crysta[ine solid canbe made to proceed essentially to completion.Accordingly, LaF3 can be reduced and extractedinto molten Bi from LiF-BeFz mixtures. Figure 7shows the behavior of several rare-earth ele-ments$ upon extraction into bismuth using

-

Llt(which is more convenient to use) as the reductant.such a process would seem to be useful.

However, when the melt is complicated, os isnecessary for the single-fluid breeder, by theaddition of large quantities of ThFr the situationbecomes considerably less favorable. use of Lioconcentrations as high as those shown in Fig. 7 isprohibited since the reaction

,lO3

402

,lo I

too

{o'l

4c-2lo-2 ,lo l {oo ,lol

LlrHluM FOUND tN METAL pHAsE (or.%)

Fis' 7' :itil: ;1,111ffi"":i:""?'"":?:1,iiff::'"'1ru:BeF, (66-34 moLels) and bismuth at 60trC.

*Bitrt + *rnrn(d) + LirBi) = LiF(d) + *rrrgiz(c) (26)

proceeds at lower concentrations than these toyield solid ThBis. Accordingly, the most reducingmetallic phase which is tolerable is that repre-sented by the (barely) saturated soluflon of ThBizin Bi. when such solutions are used the rare-earth fluoride is much less completely extractedand relatively small separation factors from tho-rium are obtain"4. 51-53, 55 Figure g shorps dataobtainedss in extraction of ce it 600t from sev-eral LiF- BeFz - ThFe mixtures (the points arelabeled to indicate mole percentages of LiF, BeF2,and ThFa, respectively) in this way. separationfactors appreciably above unity are obtained evenin the worst case, and compositions containingLZ moLeVo ThF4 withsomewhat more attractive. unfortunately, ce3+ isone of the easiest of the rare earths to reduce.

oo.cc.9OoL-

ooE

E(l)

E.gc.9C)oL-

(l)

oE

IFELd

hJE.

ElrozoFfeE.FIo

LANTHANUM

t52 NUCLEAR APPLICATIONS & TECHNOLOGY VOL. 8 FEBRUARY 1970

EF

oo4(J '.

otlgE.o53t!zIkE.

f2lda

Others in the series show even smaller separationfactors" Another paper in this series presents in-formation showing how processes with separatio.lfactors of this magnituAe could be operated.asThere is little doubt, however, that improved(Iarger) separation f actors would be desirable;there is reason to believe that improvement canbe obtained by modification of the alloy phase.

o 20 30

BeFr) -3(mole %ThFa)


properties of these fluids, estimated in most casesfrom data on similar materials, Seem adequatefor their use in MSBR.

Compatibitity of the LiF-BeFz-ThFr-UFe meltwith Hastettoy-N and with moderator graphiteseems assured, and operation of LiF-BeFz-ZrEa-UFe melts in MSRE indicate very strongly that thecompatibility will not be adversely affected by theconsequences of radiation and fission of the ura-nium. Compatibility of the secondary coolant, whileIess satisfactory, &ppears to be adequate withHasteltoy-N if salt purity is maintained.

Fission product Kr and Xe are virtually insolu-ble in the fuel and can be removed, if the moder-ator graphite is suff iciently impermeable, bysimple equilibration with an inert gas (helium).Fission products with stable fluorides (Rb, Cs,Sr, B&, the lanthanides, and Zr) appear almostentirely in the fuel as fluorides except as they arelost through volatilization of precursors. Morenoble fission products (Nb, Mo, Ru, and Te) arevirtually absent from the fuel, but plate (in ap-preciable amounts) on the graphite and the Hastel-Ioy-N, and appear to a surprisingly large extent inthe helium purge gas.

Chemical separations, of which reductive ex-traction appears most attractive, f or removinguranium and protactinium from the fuel salt andfrom each other have been demonstrated in smallscale laboratory equipment. Separations of lan-thanides from the fuel are markedly more diffi-cult, but reductive extraction of these elementsinto molten bismuth appears possible.

While much research and development remainsto be accomplished there seems to be no funda-mental chemical difficulty with design and opera-tion of a single-fluid molten-salt breeder system.

ACINOTI."EDGTENTS

This research was sponsored by the (J.S. AtomicEnergy Commission under contract with the UnionCarbide Corporation.

REFERENCES

1. M. 'W. ROSENTHAL, P. R. KASTEN, and R. B.BRIGGS, "Molten Salt Reactors-History, Status, andPotential ,' ' Nucl . ApPl . Tech., 8, 10 7 (1 970).

2. E. S. BETTIS and R. C. ROBERTSON, "MSBR De-sign and Performance Features," Nucl. Appl. Tech.,'8,190 (1970).

3. A. M. PBRRY and H. F. BAUMAN, 'cMSBR ReactorPhysics," Nncl. APPI. Tech., 8, 208 (1970).

-20EXCESS

-lo o {o

FREE FLU.RIDE te oloLiF - z(mote olo

Fis' 8' :ifi:L"|,:*' ;:frlff 'f#,J*"ff"ff?1':ll?Tjtion of cerium from LiF-BeFr-ThFn mixturesinto bismuth at 600oC.

SUMMARY

Phase behavior of LiF- BeFz - ThF4- UFe mix-tures appears suitable to permit use of a highconcentration of ThF4 in melts whose freezingpoint wiII prove acceptable for single-fluidmolten- salt breeder reactors. Oxide tolerancesof such mixtures are not entirely def ined, butIaboratory experience and the MSRE operatingeliperience combine to suggest that no oxidescavenger (such as ZrFa) need be added. Thedesign basis f uel, accordingly, consists of 71.7

mole% t LiF, 16 mole% BeFz , 12 moLeflo ThF4, and0.3 mo[efto "tUFn. The secondary coolant is to be

B mo\e|o NaF with 92 moLeflo NaBFe. Physical

74- 20-60

sI

(oI

Nt-o

oO)

III

lr)F-

?Ns3

l-Ft,^\ ay .a-n I -r,/l, ,t-^.-\9 /a

.o

(oI

oro

I

sf(o

'// NOTE:VALUES CALCULATEDAT THORIUM SATURATIONIN BISMUTH

-os

I

ro(\t

I

rO(o

NUCLEAR APPLICATIONS & TECHNOLOGY VOL. 8 FEBRUARY 1970 r53


4, 'W. R. GRIMES, F. F. BLANKENSHIP, G. W. KEIL-HOLTZ, H. F. POPPENDIBK, and M. T. ROBINSON,"chemical Aspects of Molten Fluoride Reactors,tt inProgress in Nuclear Energy, Seri.es IV, VoI . 2, Per-gamon Press, London (1960),

5. \iV. R. GRIMES, D. R. CUNEO, F. F. BLANKENSHIP,G. KEILHOLTZ, H. F. POPPENDIEK, and M. T.ROBINSON, "chemical Aspects of Molten Fluoride SaltReactor Fuels," in FfuM,Fuel Reactors, J. A. LANE,H. G. MacPHERSON, and FRANK MASLAN, Eds., Addi-son-Wesley, Reading, Mass. (1958).

6. 'W. R. GRIMES, "Materials Problems in Molten SaltReactors," in Materials and Fuels for Hi.gh TempeTa-ture Nuclear Energy Applications, M. T. SIMNAD andL. R. ZUV. IWALT, Eds., The M.I.T. Press, Mass. (1 964).

7. 'w. R. GRIMES, 6'MSR Program semiannual prog-ress Report, July 31, L964,r' ORNL-8208, p. ZL4, OakRidge National Laboratory.

8. ALVIN GLASSNER, '3The Thermochemical proper-ties of the oxides, Fluorides, and Chlorides to 2b001(,',ANL-5750, Argonne National Laboratory.

9. L. BREWER, L. A. BROMLEY, p. 'W. GILLES, andN. L. LOFGREN, MDDC-1553 (1945) and L. BREWER inThe Chemistry and, Metallurgy of Mi.scellnneous Ma-terials; Thermodynnmics, L. L. QUILL, Ed., McGraw-HiU, New York, pp. 76-192 (1950).

10. R. E. THOMA, Ed., "Phase Diagrams of NuclearReactor Materials," ORNL-254A, Oak Ridge NationalLaboratory (November 6, L95g),

11. L. V. JONES, D. E. ETTER, C. R. HUDGENS, A. A.HUFFMAN, T. B. RHINEHAMMER, N. E. ROGERS,P. A. TUCKER, and L. J. WITTENBERG, ,,phase Equi-libria in the Ternary Fused-salt system LiF-BeF z-UF4,t' J. Am. Ceram. Soc., 45, 79 (LgG2r.

L2. R. E. THOMA, H. INSLEY, H. A. FRIEDMAN, andc. F. WEAVER, "Phase Equilibria in the systems BeF2-ThF4 and LiF-BeFz-ThFa" J. phys. Chem., 64, g6b(1 e6 0).

13. C. J. BARTON, H. A. FRIEDMAN, 'W. R. GRIMES,H. INSLEY, and R. E. THOMA, "Phase Bquilibria in theA1kali Fluoride-Uranium Tetrafluoride Fused Salt Sys-tems: L. The Systems LiF-UF+ and NaF-UF*', J. Am.Ceram. 9oc ., 41, 63 (i.gb8).

L4. R. E. THOMA, H. INSLEY, B. S. LANDAU, H. A.FRIEDMAN, and 'w. R. GRIMES, 6'phase Equilibria inthe Fused salt systems LiF-ThF+ and NaF-ThF4,'' J.Phys. Chem., 63, L266 (1959).

15. C. F. WEAVER, R. E. THOMA, H. INSLEY, andH. A. FRIEDMAN, " Phase Equilibria in the SystemsUF4-ThF+ and LiF-UFa-ThFa,t' J. Am. Ceram. Soc ., 45,2L3 (1 96 0).

16. K. A. ROMBERGER, C. F. BAES, and H. H. STONE,"Phase Equilibrium studies in the uranium(rv) oxide-

zrrconium oxide System,tt presented at 151st NationalMeeting of American chemical society, Pittsburgh, Pa.,March 2L-3L, L966.

L7. C. F. BAES and C. E. BAMBERGER, ,,MSR pro-gram Semiannual Progress Report, August 81 , 1968,t t

ORNL-4344, Oak Ridge National Laboratory.

18. A. L. MATHEWS and c. F. BAEs, "oxidechemistryand Thermodynamics of Molten LiF-BeFz by Equilibri-um with Gaseous Water-Hydrogen Fluoride Mixtures,t'ORNL-TM-1129, oak Ridge National Laboratory (May1965).

19. c. F. BAES, "The chemistry and rhermodynamicsof Molten salt Reactor Fluoride solutions,r, proc. IAEAsymf . on Thermodynnmins with Emphasis on NuclearMaterials and Atomic Transport i.n Solids, vienna, Aus-tria (July 196b).

20. c. F. BAES and B. F. HITCH, "Reactor chemistryDivision Annual Progress Report, January 81, 1g6b,r,ORNL-3789, pp. 61-5, oak Ridge National Laboratory.

2L, E. M. LBVIN, C. R. ROBBINS, and H. F. McMURDIE,Plmse Diagrams for Ceramists, The American Ceram-ic Society, Inc., Ohio (1964).

22. v. G. sELrvANov and v. v. STENDER, zhur.Neorg. Khim., 3, 2, 448 (19b8).

23. C. J. BARTON, L. O. GILPATRICK, H. INSLEY,and r. N. McvAY, "MSR Program semiannual prog-ress Report, February Zg , 1968,t, ORNL- 4254, p. 166,Oak Ridge National Laboratory.

24' S. CANTOR, J. 'w. CooKE, A. S. DWORKIN, G. D.ROBBINS, R. E. THOMA, and G. M. wATSoN,',physicalProperties of Molten Salt Reactor FueI," ORNL-TM-23L6, oak Ridge National Laboratory (August 196g).

25. H. E. McCOY, W. H. COOK, R. E. GEHLBACK,J. R. WEIR, C. R. KENNEDY, C. E. SESSIONS, R. L.BEATTY, A. P. LITMAN, and J. 'W. KOGER, ,.NewDevelopments in Materials for Molten-Salt Reactors,ttNucI. AppI. Tech., 8, 1bO (1920).

26. C. F. BAES, Jr., ,.The Chemistry and Thermody-namics of Molten salt Reactor Fuels,r' presented atAIME Nuclear Fuel Reprocessing symposium at AmesLaboratory, Ames, Iowa, AuSrst 2b, 1969. published inthe 1969 Nuclear Metallurgy symp., vol. ls by theUSAEC Division of Technical Information Extension.

27. G. LONG, "Reactor chemistry Division AnnualProgress Report, January 31, 1g6b,t' ORNL-3T89, p. 6b,Oak Ridge National Laboratory.

28. "MSR Program semiannual progress Report, Au-gust 31, 1961," ORNL-92L5, p. gB, Oak Ridge NationalLaboratory.

29. C. E. WICKS and F. E. BLOCK, "ThermodynamicProperties of 65 Elements-Their oxides, Halides, car-bides, and Nitrides,t'Bureau of Mines Bulletin 60b(1963).

APPLICATIONS & TECHNOLOGY VOL. 8 FEBRUARY 1970t54 NUCLEAR

30. JANAF (Joint Army-Navy-Air Force) Interim Ther-mochemical Tables, Thermal Research LaboratorY, DowChemical Co., Midland, Mich.

31. P. N. HAUBENREICH and J. R. ENGEL, "Operationof the Molten-Sa1t Reactor Experiment," Nucl' Appl.Tech ., 8, 1L 8 (1 970).

32. R. F. APPLE, Method No. 902L206 (March16'L965)'"ORNL Master Analytical Manual," TID-7015 (Suppl. 8).

33. R. E. THOMA, "MSR Program Semiannual Prog-ress Report, Au$ust 31, 1968,tt ORNL-4344, p. 110-112'Oak Ridge National Laboratory.

34. R. E. THOMA, "MSR Program Semiannual Prog-ress Report, February 29, 1968," ORNL-4354, PP' 88-93, Oak Ridge National Laboratory.

35. R. E. THOMA, "MSR Program Semiannual Prog-ress Report, August 31, L967," ORNL-419L, PP. 102-108, Oak Ridge National Laboratory.

36. S. S. KIRSLIS and F. F. BLANKENSHIP, "ReactorChemistry Division Annual Progress Report, December31 , 1966 ," ORNL- 407 6, PP. 45-54, Oak Ridge NationalLaboratory.

37. S. S. KIRSLIS, "MSR Program Semiannual Prog-ress Report, August 31 , 1966,' ' ORNL- 4037 , pp. 165-166, Oak Ridge National Laboratory.

38. S. S. KIRSLIS and F. F. BLANKENSHIP, "MSRProgram Semiannual Progress Report, August 31 ,

L967 ,'' ORNL- 4LgL, pp. 116 -L20 , Oak Ridge NationalLaboratory.

39. S. S. KIRSLIS and F. F. BLANKENSHIP, "MSR Pro-gram Semiannual Progress Report, Februaty 28, 1968,t'ORNL-4254, pp. 94-LI4, Oak Ridge National Laboratory.

40. S. S. KIRSLIS and F. F. BLANKENSHIP, "MSRPro-gram Semiannual Progress Report, August 31, 1968,t'ORNL- 4344, pp. 115-141 , Oak Ridge National Laboratory.

4L, F. F. BLANKBNSHIP, S. S. KIRSLIS, J. B. SAVO-LAINEN, and W. R. GRIMES, "Reactor Chemistry Divi-sion Annual Progress Report, January 31,1963,tt ORNL-3417 , p. L7, Oak Ridge National Laboratory.

42. 'W. R. GRIMES, N. V. SMITH, and G. M. WATSON,"solubility of Noble Gases in Molten Fluorides. I. InMixtures of NaF-ZrFn 63-47 mole 7d and NaF-ZrF4-UFc @0-46-4 mole Vd," J. Phys. Chem., 62, 862 (1958).

43. M. BLANDER, W. R. GRIMES, N. V. SMITH, andG. M. WATSON, "solubility of Noble Gases in MoltenFluorides. II. In the LiF-NaF-KF Eutectic Mixtures,"J. Phys. Chem.,63, L164 (1959).

44. G. M. WATSON, 'W. R. GRIMES, and N. V. SMITH,' 'Solubility of Noble Gases in Molten Fluorides. InLithium-Beryllium Fluoride," J . Chem . Eng. Data, 7 ,285 (Le62).

45. R. GRIMBS, J. H. SHAFFBR, N. V. SMITH, R. A.STRBHLOW, W. T. WARD, ANd G. M. WATSON, ..High-


Temperature Processing of Molten Fluoride NuclearReactor FueIs,tt Nucl. Eng.-Part VII,55, 27,65 (1959).

46. W. T. WARD, W. R. GRIMES, R. A. STREHLOW,and G. M. WATSON, "Rare Earth and Yttrium Fluor-ides. Solubility Relations in Various Molten NaF-ZrFa-UF+ Solvents,t' J. Chern. Eng. Data, 5, L37 (1960).

47 . C. F. BABS, Jr. , " Reactor Chemistry DivisionAnnual Progress Report, Januaty 31, L967 ," ORNL-4076, pp. 49-50, Oak Ridge National Laboratory.

48. M. E. WHATLEY, L. E. McNEESB, W. L. CARTER,L. M. FERRIS, and E. L. NICHOLSON, "EngineeringDevelopment of MSBR Fuel Recycle," N?tcl. Appl. Tech.,8, 170 (1970).

49, C. J. BARTON, " Reactor Chemistry Division An-nual Progress Report, December 31 , 1965,t' ORNL-3913, p. 45, Oak Ridge National Laboratory.

50. J. H. SHAFFER, W. R. GRIMES, G. M. WATSON,D. R. CUNEO, J. E. STRAIN, and M. J. KELLY, "TheRecovery of Protactinium and Uranium from MoltenFluoride Systems by Precipitation as Oxides,t' Nucl.Sci. Eng., 18, I77 (1964).

51. J. H. SHAFFER, D. M. MOULTON, and R.

GRIMES, "MSR Program Semiannual Progress Report,February 29 , 1 96 8,' ' ORNL- 4254, PP. L52-L55 , Oak

Ridge National LaboratorY.

52. L. M. FERRIS, J. F. LAND, J. J. LAWRENCE, and

C. E. SCHIT,LING, "MSR Program Semiannual Prog-ress Report, February 29,1968,tt ORNL-4254, PP. 24L-247, Oak Ridge National Laboratory.

53. L. M. FERRIS, J. C. MAILEN, E. D. NOGUEIRA,D. E. SPANGLER, J. J. LAWRENCE, J. F. LAND, ANd

C. E. SCHILLING, "MSR Program Semiannual ProgressReport, Auzust 31 , 1968," ORNL- 4344, PP. 292-297 ,

Oak Ridge National LaboratorY.

54. C. J. BARTON and R. G. ROSS, "MSR ProgramSemiannual Progress Report, Augrst 31 , 1968,t' ORNL-4344, pp. 180-185, Oak Ridge National Laboratory.

55. J. H. SHAFFBR, D. M. MOULTON, and R'

GRIMES, "MSR Program Semiannual Progress Report,August 31, 1968," ORNL- 4344, PP. L74-t80, Oak Ridge

National LaboratorY.

56. J. H. SHAFFER, R. GRIMES, F. A. DOSS, W. K.R. FINNBLL, and W. P. TEICHBRT, 'sReactor Chem-istry Division Annual Progress Report, December 31'1965," QRNL-3913, P. 42, oak Ridge National Labora-tory.

57, J. J. EGAN and R. w. ffIS$ALL, Nucleonics, 15,

104 (1967).

58. J. H. SHAFFER, 'W. P. TEICHERT, D. M. MOUL-TON, F. F. BLANKENSHIP, W. K. R. FINNELL, ANd

W. R. GRIMES, "Reactor Chemistry Division AnnualProgress Report, December 31 , 1966,t t ORNL- 407 6,p. 34-36, Oak Ridge National Laboratory.

NUCLEAR APPLICATIONS & TECHNOLOGY VOL. B FEBRUARY 1970 155

molten salt reactor chemistry

Documents

fuel system

fuel circuit

coolant salt

ledto selection of fuel

fused solt fuel

f uelmixture

coolant system

zrfaufa fuel mixh