ORNL-TM-6415.txt

   
  
    

 
 

  

G s

  

 
   

s

GREEL/TH-6415

R

 

 

J. R. Engel
. F. Bauman
J. F. Dearing

W, B, Grimes

RS s

=
el

¥
r.

2

 

m‘g -.l' o ‘.9' '. e W‘W e 2 ¥ e e et ot
R S
T o

%
o

 

o

e

S
P

.; 2 = .
- R

e
R

o
2R

ORI
i

e

AN

 

R
 

Printed in the United States of America. Available from
National Techrical Information Service

 

 

; Li.5. Department of Commerce
% 5285 Port Royal Road, Springfield, Virginia 22161 ;
! . . ’ : . !
Price: Printed Copy $8.00; Microfiche $2.00

fi This report was prepared as an ascount of work sponsared by an agency of the United

i Stalzs Government Neither the United Bigtes Government nor any agency thereof, nor

| any of ineir emplovees, contractors, subconiractors, or their employess, makes any

L warranty, express or implied, nor assumes any legal Habitity or responsitility for any

| third party's use or the results of such use of any information, apparatus, product or

+

process disclesed in this repaort, nor represents that s use by such third oarty wouid
not infringe orivately owned rights.

 

 

 
ORNL/TM-6415
Dist. Category UC-76

 

Contract No. W-7405-eng~26

 

 

 

 

 

Engineering Technology Division
DEVELOPMENT STATUS AND POTENTIAL PROGRAM FOR
DEVELOPMENT OF PROLIFERATION-RESTISTANT
MOLTEN-SALT REACTORS
J. R. Engel
H. F. Bauman W. R. Grimes
J. F. Dearing H. E. McCoy, Jr.
Date Published: March 1979
NOTICE: This report contains information of a preliminary
nature. It therefore does not represent a final report and
is subject te changes or revisions at any time.
r“—'—"‘—“—— KOTICE -
| This weport was prepared as m
SPO‘HSUmd by the United States Government. Neither the |
Prep&red by the EZ:‘T:S States ;lOr f_!h_e .L'nited States Department of
» ML any of their emplovees, nor any of their
OAK RIDGE NATIONAI LABORATORY :fi;”;:::’::t;u:;:fl:zctus_s, 011" their err_ipl!.vye.es, makes
5 Oak Ridge, Tennessee 37330 1mmwf£wmmw$fit$$fl§$fi§'
¥ or usefulzess of any information, apparatus, product or
Operated by pr;)c.:ess di.tscloscd, :1 (epl:Bsc:i(S tha[;pits :use, ch):ld tnct
UNION CARBIDE CORPORATION L Py el e i‘
for the

DEPARTMENT OF ENERGY

 
 

 

 

 

 

 
 

iii

CONTENTS
EXECUTIVE SUMMARY ..... Seecesrescososesesenene e crceane tecccoaa e
ABSTRACT lllll & # £ # & 2 3 » & & & © ® & o 3 8 & & ® ¢ 0 v O 9 4 & ® & & & & 0 3 ¢ & ¢ ® ¢t o &6 & o & & ® & e & & @
INTRODUCTION ...vevevncon seesrcosaeerrescoosaensooo tetceaenannsna

REf G eIICEeE vt cc oo esuonconosasccsssnnsoscconsesssoccscnsnosnsscesoee

PART I. REACTOR DESIGN AND DEVELOPMENT

REACTOR DESIGN, ANALYSIS, AND TECHNOLOGY DEVELOPMENT ........
Status In 1972 it evensvesccossenssssososesssenosnssssssasasss
Current Development StatuS ...iiceneeetnosccocososcescoscsnscee
DMSR Development Needs ....i.ceeecrenrscenscecosrsascsscssssssscs

Ref ey eNCeS t i v eecvcocoranaocosssnnancsoseeoncsoasoensnoccesosos

PART II. SAFETY AND SAFETY-RELATED TECHNOLOGY

REACTOR SAFETY AND LICENSING ...iveenreicoenorrvoccocnsonvcsos
Status and Development Needs .c.oveievccocsocervesoooansscoonns
SafELY v creeeeeneerconrsrtcncassssnescccassactoconssosnes
Licensing «.ceverveesnneorconontanocenseronnoanoncnncennnss
Estimates of Scheduling and Costs ...cveerenciecrnnnenascccans
References ..ccieoeaesnn e bt e aeset e et o s e st e s e s o ane s e aeanennns

PART ITII. FUEL-COOLANT BEHAVIOR AND FUEL PROCESSING

FUEL AND COOLANT CHEMISTRY ....c.iticiieinninnccannnssnenuoaneans
Key Differences in Reactor Concepts ....eicercencerconnceocss
Post—-1974 Technology Advances ....eeeevsovess Citecenset e
Status of Fuel and Coclant Chemistry ..cciiierencecieercanans

Fuel chemistry c.eevierterricnsescerctsanenconsancsnnnnsonnas
Coolant chemiStry ¢evieceneecronsnrsnccs seaasas cevessvaens
Fuel—-coclant interactions .....ciceueeeescrsrsacessaccsconsas
Fuel-graphite Interactions ......cciceerrvcrocnonsencoones

Prime RED NeedsS v.veeeeeecoonsecesecnooassoacoansesssocosssnses

Fuel chemistry .......... feeecaien e ero s tecencavaeanas
Coolant chemistry ...cv.iii e ieacncacnnnas sssescaas
Fuel-coolant interactions ....eceeieniiececencenncs taanseea
Fuel-graphite Interactions ..o.veeecccoenrccccaososasncons

Estimates of Scheduling and CostsS ...vveiivecvcoonaniecoonenns

11
14

23
23

23
25

26
26

31
31
32
34

34
an
47
48

49

49
50
50
50

o1
iv

ANALYTICAL CHEMISTRY . cviveeveecoeecnonoceccasonnnscsconsssssco
Scope and Nature of the Task ...... o eecceosoass e e
Key Differences in Reactor ConceptsS ..osveercssoseccconncaons
Post—-1974 Technology Advances «...veeveccocnorsrccononnns ee e
Status of Analytical Development ........... cecosssscee e Peee

Key developments for MSRE .....iciiccneeneientccsncarcccnn
Analvytical development for MSBR ..... cecenn ceriserecoeenes

Prime Development NeedS .....ccovtvevccccnncarconns et eecenss
Estimates of Scheduling and CosSts ...ovveee ceceseascasnurene
MATERIALS DEVELOPMENT FOR FUEL REPROCESSING .cccocvencsne v
Scope and Nature of the Task .cccocencrcerrernosrercccsoanens
Key Differences in Reactor CoOnCeptsS .ceoeveeccencanscncaccsns
Post=1974 Technology Advances .vecoee.s Ctccosseaseneceanaenan
Present Status of Technology ..ceiciiercnneneccoonnnncacasnsnn

Materials for fluorinators and UFg absorbers ....ceececeos
Materials for selective exXtractionsS vecccoooreccesacnsonnas

Primary R&D Needs ...uvirecieerecccconsconansans theumassanene
Estimates of Scheduling and CostsS ....icieiicetenenccosnconca
FUEL PROCESSING .t cvveeteveeconosnsscasnsanssssas ceeesesenoan
Scope and Nature of the Task ..ieicinieeereiricnorncvcoancssn
Key Differences in Reactor Concepts ..cceeeisorianicasanaercs
Post—1974 AdVANCES +ceceeencnncootnnsassassaconssssnasocsosas
Status of Technology teveeeeosscearsssseceonsnsennsanncasannss

Chemical status ......... cisornsas cesassresacs Ceessserienos
Conceptual MSBR processing flowsheet ........cccciivnan.
Fngineering StAlUS +.iceereracsceoanscssessernsasssacconsss
Special characteristics of DMSR processing ...cceceeviean.
Possible processing alternatives .......co00. cecresscocan

Primary R&D Needs «uiviiiirieivensessiivesccosseassscosssosssnss
Estimates of Scheduling and CosSts vuiiveicniioronnsnecsoncons

Ref TN C eSS ot v et o veneocceanoanonscsossenaonsssccesnsoe e seees s

PART IV. REACTOR MATERIALS

STRUCTURAL METAL ¥OR PRIMARY AND SECONDARY CIRCUITS ...cceo-..
StAtUS TN 1977 oottt enoeenececoooeeasanossescoonssesssoeconnsnss
Status Iin 19748 ... et evoossnesccoses e o e o s ecct o anneneseannne
 

Page
CUrTent SEALUS «veeseosansesscecoansescasoansoscosasssnscosss 130
Further Technology Needs and Development Plan ............... 130
GRAPHITE FOR MOLTEN-SALT REACTORS ............. ceciecneonnons 135
Status In 1972 . i. et iennroonssosassersosasssacconasssascccoas 135
Status Inm 19760 «uveveoeciorosssnssnsconssanssascoossssonacnonsnso 138
Current StafUuS ..cecrvocoosssenccoosana tee e ceesecoas et e oo 138
Further Technology Needs and Development Plan .....c.veeecoons 138
 

vii
EXECUTIVE SUMMARY

INTRODUCTION

Molten-salt reactors (MSRs) are of interest in possible prolifera-

£33y power plants that

tion-resistant systems, particularly as denatured
could be widely deployed with minimal risk of proliferation. MSRs might
also be used as "fuel facteries" in secure centers, burning plutonium and
producing 2337, However, before they can be used, the MSR concept must

be developed into a commercial reality. The purpose of this report is

to review the status of molten-salt technology from the standpoint of

the development required to establish an MSR industry.

Following the successful operation of the Molten~Salt Reactor Ex-
periment (MSRE, 1965—69), it became necessary for the government to de-
cide if MSR development should be continued. To this end, a comprehensive
report on MSR technolegy was published in August 1972.' Because only
limited R&D has been conducted since then, most of the information in
the report is still valid and will be taken as the basis for the present
review. Some additional development work done in 1974—76 will be used to
update the conclusions of the 1972 study. The government decided not to
proceed with the further develecpment of the Molten—-Salt Breeder Reactor

(MSBR), or any other MSR, for reasons other than technological ones.

DEVELOPMENT STATUS, 1972

The development status of MSBRs in 1972 is covered thoroughly in
Ref, 1. All aspects of reactor development, from reactor physics to
materials of construction, are covered and wilil not be repeated here.
Of particular interest in that review are the discussions of technologi-
cal advances believed to be needed before the next MSR could be built.
These needed advances are defined briefly in the introduction of that

report as follows:

"In the technology program several advances must be made before we
can be confident that the next reactor can be built and operated success-—

fully. The most important problem to which this applies is the surface
viii

cracking of Hastelloy N, Some other developments, such as the testing
of some of the components or the work on latter stages of the processing
plant development, could actually be completed while a reactor is being
designed and built. The major developments that we believe should be

pursued during the next several years are the following:

"l1. A modified Hastelloy N, or an alternative material that is im-
mune to attack by tellurium, must be selected and its compatibility with
fuel salt demonstrated with out-of-pile forced-convection loops and in-
pile capsule experiments; means for giving it adequate resistance to
radiation damage must be feound, if needed, and commercial production of
the alloy may have to be demonstrated. The mechanical properties data
needed for code qualification must be acquired if they do not already
exist.

"2, A method of intercepting and isolating tritium to prevent its
passage into the steam system must be demonstrated at realistic conditions
and on a large enocugh scale to show that it is feasible for a reactor.

"3. The various steps in the processing system must first be demon-
strated in separate experiments; these steps must then be combined in an
integrated demonstration of the complete process, including the materials
of construction. Finally, after the MSBE* plant is conceptually designed,
a mock-up containing components that are as close as possible in design
to those which will be used in the actual process must be built and its
operation and maintenance procedures demonstrated.

"4, The various components and systems for the reactor must be de-
veloped and demonstrated under conditions and at sizes that allow con-
tident extrapolation to the MSBE itself. These include the xenon strip-
ping system for the fuel salt, off-gas and cleanup systems for the coolant
salt (facilities in which these could be done are already under construc-
tion), tests of steam—generator modules and startup systems, and tests of
prototypes of pumps that would actuaily go in the reactor. The construc-
tion of an engineering mock-up of the mazjor components and systems of the

reactor would be desirable, but whether or not that is done would depend

 

Molten-Salt Breeder Experiment; an intermediate-scale developmental
plant.

 
ix

 

on how far the development program had proceeded in testing various com-—
ponents and systems individually.

"5. Graphite elements that are suitable for the MSBE should be
@ purchased in sizes and quantities that assure that a commercial produc-

tion capability does exist, and the radiation behavior of samples of

4 the commercially preduced material should be confirmed. Exploration
of methods for sealing graphite to exclude xenon should continue.

"6. On-line chemical analysis devices and the various instruments
that will be needed for the reactor and processing plant should be pur-
chased or developed and demonstrated on lcops, processing experiments,

and mock-ups."

The first three objectives were considered crucial to the MSBR con-
cept; the results of further development effort on them during 1974—76
are discussed in the following section. Objectives 4 to 6, while im-
portant, did not appear to present any insurmountable obstacles; in any

event, they could not be pursued further because of limited funding.

RESULTS OF R&D — 1972 TO PRESENT

At the direction of AEC/ERDA,* the MSR program was discontinued in
early 1973, resumed in 1974, and finally terminated at the end of FY
1976. Although the development effort since 1972 has been severely re-

stricted, some significant results were obtained from work performed

mainly in 1974—76.

Alloy Development for Molten-Salt Service

 

The nickel-based alloy Hastelloy N, which was specifically developed
for use in molten-salt systems, was used in construction of the MSRE.
The material generally performed very well, but two deficiencies became
apparent: (1) the alloy was embrittled at elevated temperatures by ex-
posure to thermal neutrons and (2) it was subject to intergranular sur-

face cracking when exposed to fuel salt containing fission products.

 

"Now the U.S. Department of Energy.
Recent development work indicates that solutions are available for both
these problems. Details of this work are given by McCoy;2 a summary of
the results follows.

Irradiation experiments early in the MSR develcopment program showed
that Hastelloy N was subject to high-temperature embrittlement by thermal
neutrons. The MSRE was designed around this limitation (stresses were
low and strain limits were not exceeded), but the development of an im-
proved alloy became a prime objective of the materials program. It was
found that a modified Hastelloy N containing 27 titanium had much im-
proved postirradiation ductility, and extensive testing of the new alloy
was under way at the close of MSRE operations.

The second problem, intergranular surface cracking, was discovered
at the close of the MSRE operation when surface cracks were observed
after strain testing of Hastelloy N specimens that had been exposed to
fuel salt. Research since that time has shown that this phenomenon is
the result of attack by tellurium, a fission product in irradiated fuel
salt, on the grain boundaries.

As a result of research from 1974 to 1976, two likely soclutions to
the problem of tellurium attack have been developed. The first involves
the development of an alloy that is resistant to tellurium attack but
still retains the other required properties. This development has pro-
ceeded sufficiently to show that a modified Hastelloy N containing about
1% niobium has good resistance to tellurium attack and adequate resistance
to thermal-neutron embrittlement at temperatures up toc 650°C. It was
‘also found that alloys containing titanium, with or without niobium, ex-
hibited superior neutron resistance but were not resistant to tellurium
attack.

The secend likely solution involves the chemistry of the fuel salt.
Recent experiments indicate that intergranular attack on Hastelloy N
is much less severe when the fuel-salt oxidation potential, as measured
by the ratio of utt to U3+, is less than 60.% This discovery opens up

the possibility that the superior titanium-modified Hastelloy N could

 

The inverse of this ratio, that is, the ratio of Ut o UMt

now more commonly used to describe the oxidation state of the salt.

 
 

xi

be used for MSRs through careful control of the oxidation state of the
fuel salt.

Both of the above scolutions appear promising, but extensive testing
under reactor conditions would be required before either could be used

in the design of a future MSK.

Tritium Control

 

Large quantities of tritium are produced in MSRs from neutron reac-
tions with lithium in the fuel salt. FElemental tritium can diffuse
through metal walls such as heat-exchanger tubes at elevated temperatures,
thus providing a potential mechanism for the transport of tritium to the
reactor steam via the secondary coolant locp and the steam generator.
Recent experiments indicate that tritium is oxidized in the proposed MSR
secondary coolant, sodium fluoroborate, thus blocking transpert to the
Steam system.

In 1975 and 1976, tritium—additiocn experiments were conducted in an
engineering-scale coclant salt test loop. The results are given in a
report by Mays, Smith, and Engel.3 Briefly, the experiments showed that
the steady-state ratio of combined to elemental tritium in the coolant
salt was greater than 4000. A calculation applying this ratio to the
case of an operating 1000-MW(e) MSBR indicated that the release of
tritium to the steam system would be less than 400 GBg/d (10 Ci/day).

The conclusion of the study was that the release of tritium from an MSR
using sodium fluorcborate in the secondary coolant systfem could be readily

controlled to within Nuclear Regulatory Commission (NRC) guidelines.

Engineering Development of Fuel Processing

By 1972, proof-of-principle experiments had been carried cut for
the various steps in the reference chemical preocess, but development and
demonstration of engineering-scale equipment were just getting under way.
The only large-scale processing demonstrated at that time was the batch
fluorination of the MSRE fuel salt and the recovery of the uranium on

NaF beds.
xii

In the period 1974—76, efforts were begun to develop items of equip-
ment which would be vital to the success of the metal-transfer process.
Some progress was made in the develepment of a salt-bismuth contactor,

a continuecus fluorinator, and a U¥g absorber for reconstituting the fuel
salt.” Because of the pregram closeout in 1976, this work could not be
continued long enough to culminate in engineering designs for the various
items of equipment. The status of this work can be summed up by stating
that, although no insurmountable obstacles were encountered, the major

portion of process engineering development remains to be done.

Other Areas of Development

 

The development status of areas other than those discussed above is
practically unchanged since the report1 of 1972, because no further R&D
was funded. These include development of reactor components, moderator
graphite, analytical methods, and controcl instrumentaticn. Exceptions
were a design study of a mclten-salt heat exchanger and some limited
work on the in-line monitoring of fuel salt.

In 1971, Foster-Wheeler Corp. was awarded a contract for a study of ¥
MSR steam-generator designs. The contract was suspended in 1973 and then
reinstated in 1974 for the purpose of completing the first task (in a
four—-part contract), which wag the design of a steam generator to meet
specifically the steam and feedwater conditions postulated for the MSBR
conceptual design. This task was successfully completed and a report
issued in December 1974.° A design was presented which, based on analy-~
sis, would meet gll the requirements for an MSR steam generator. How-
ever, the design was not experimentally verified because the MSR project
was terminated.

The 1972 status report1 described the use of an in-line electro-
chemical technique known as voltammetry to monitor the oxidation poten-
tial of the fuel salt. The technique has since been used to monitor
various corrosion test loops and other experiments and may also be used

2+

to monitor Cr in fuel salt, a good indicator cof the overall corrosion \

rate. Recently the technique has been used to measure the oxide ion in

 
 

 

xiidi

fuel salt. Oxide monitoring is very important in molten-salt fuel be-
cause an increase in oxide contamination could lead te precipitation of

uranium from the fuel as UQ».

SPECIAL DEVELOPMENT REQUIREMENTS FOR THE DMSR

Recent reexamination of the MSR concept with special attention to
antiproliferation considerations has led to the identification of two
preliminary design concepts for MSRs that appear to have substantially
less proliferation sensitivity without incurring unacceptable perfor-
mance penalties. The designation DMSR {(for denatured molten-salt reac-
tor) has been applied to both of these concepts because each would be

2359 enriched to no more than 20% and would be

fueled initially with
cperated throughout its lifetime with denatured uranium.

The simpler of these DMSR concepts6 would completely eliminate on-
line chemical processing of the fuel salt for removal of fission products.
(Stripping of gaseous fission products would be retained, and some batch-
wise treatment to control oxide contamination probably would be required.)
This reactor would require rcutine additions of denatured 235y fuel, but
would not require replacement or removal of the in-plant inventory except
at the end of the 30-year plant lifetime. Adding an on-line chemical
processing facility to the 30-year, once-through reactor provides the
second DMSR design concept.’ With this addition, the conversion ratio
of the reactor would reach 1.0 (i.e., break-even breeding) so that fuel
additions could be eliminated and a given fuel charge could be used in-
definitely by transferring it to a new reactor plant at the decomission-
ing of the old unit.

The required chemical processing facility for a DMSR, shown as a pre-
liminary conceptual flowsheet in Fig. S.1, would be derived largely from
the MSBR but would contain some significant differences. In particular,
isolation and segregation of protactinium would be avoided, provisions

would be made to retain and use the plutonium produced from “*%U

, and a
special step would be added for removal of fission-product zirconium.

Thus, the development of on-line chemical processing for a DMSR would
ORNIL DWG 78 5764

   
    

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

   

 

 

   

 

 

 

 

 

 

 

 

 

 

 

   
    

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

    
   

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
 

 

 

l§frs 23 sz o cmm 1 e . e e 3 L e e e e . £ e £ e e e e
i [T Cs AND Rb
[ACCUMULATOR
UF,, ThF,, BeF, g - REGCYCLE : e i
ADDITION ' r g ‘.‘+ RE‘?2 < o e
i . o | ; ACCUMULATION & i
¥ VALENCE ; UF, UFg i E IN Bi = i
y AR&REI" | |REDUCTION ABSORFTION] 1y, ! | F«~+ r=+y é | i
Bi — ; = = —ADo L ADD REZ |t — e i |
REMOVAL| bl | f : LiCl+  |STRIPPER)ag . 4 | i
TO & " ' T + ' i
® P ¥ ! ! ;
REACTOR i __Jmmn» Lob T RARE 1y mmm———————— ——————————— B
— EARTH | 7 !
{ P P T L [ransrer] g i ADD L § E
)\ { PARTITION g | EARTH P E : e Wy s et _ P
¢ | i. EXTRACTION i) g Lic: ! REY ACCUMULATION| g oo | |
[ S ] , | — | IN Bi l P
P ! —_— e DT T e T + * i ! b
’ I ¥ i P e ! ! y KT T ’ Pl
o o ] s ) e o
$ i EXTRACTION el EXTRACTION P coLvenT] 1 =+ 2| STRIPPER i===%1c| jorin-| F | i
4 ! ; I e b . T | ATION | 11 E i
b rfm | i T ; ;
FROM . T | l I e -l ‘ L]
R :
REACTOR ‘ Il,‘ I + t_.__.- 4
™1  PRIMARY [ p ; n
: F T F.P. AND WASTE SALT P
FLUORINATOR |ag2. SECONDARY|skiie] HYDRO- L + | FOR RECOVERY OF kit ottt ] U
P | FRUORINS oy PrY0RING s F ULTIMATE DISPOSAL P
ATOR ' ATOR bbbt |
T w i 4 — e —_ eer o oo o e e o e ¥ !
T ! i
FUEL OR FUEL SOLVENT " bbb e ¥
----- BISMUTH - = -= T e e e e e
=GP UFB
~ 4 e |G

b WASTE SALT

Fig. S.1. Proposed DMSR fuel processing flowsheet.

   
XV

require essentially all the technology development identified for the
MSBR with additions to accommodate these differences. However, since

the DMSR offers a no-processing option, a large fraction of the repro-
cessing development, along with its associated materials development,
could Be deferred or even eliminated. Such deferral might be expected

to reduce the cost (but probably not the time) for developing the first
DMSRs. To provide an overall perspective, this development plan includes
costs and schedules for developing the reprocessing capability in parallel
with the reactor.

The only other substantial difference (in terms of development needs)
between the MSBR and the proposed DMSR concepts is the reactor core design,
which is similar for both. Relaxing the breeding requirement and empha-
sizing proliferation resistance for the DMSR led to a core design with a
much lower power density to limit losses of protactinium, the 233y pre-
curser which is retained in the fuel salt of the DMSR. By reducing the
rate of fast-neutron damage to the core graphite, the low power density
also makes possible the design of a core in which the graphite need not
be replaced for the life of the reactor. A low power density also re-
duces the poison fraction associated with xenon in the core graphite and
thus there is less need for a low-permeability graphite. Although im-
provements in graphite life and permeability would be desirable, graph-
ite grades tested before 1972 would have the properties acceptable for
the DMSR core. Graphite development for the DMSR would not require (but
could include) much effort beyond the gspecification and testing of com-

mercial-source material.

POTENTIAL PLAN FOR DMSR DEVELOPMENT

A major product of the reactivated MSR program in 1974 was a de-

8 for the first several years of a development effort that

tailed plan
would ultimately lead to a commercial MSBR. Since the program authorized
in 1974 was restricted in scope, no attempt was made in that plan to
include costs and schedules for reactor plants beyond a limited treat-

ment of a proposed next-generation reactor — the Molten-Salt Test Reactor
XVvi

(MSTR). The primary function of the 1974 program plan was to define a
base technology program for the MSBR. Since the technology needs for a
DMSR closely parallel those of the MSBR, extensive use was made of the
1974 program plan in evolving the plan described below for DMSR develop-
ment,

To develop a reasonable perspective of the potential role of the “
DMSR in providing nuclear electric power, it is necessary to concep-
tualize a reactor development and construction schedule that goes beyond
the MSTR to at least the first commercial (or prototype) system and pos-
sibly on to the first of a series of "'standard" plants. The potential
schedule that was developed (Fig. S.2) has a reasonable basis for ful-
fiilment in the iight of the current state of MSR technoleogy. Four

generally parallel lines of effort would be pursued, including:

1. a base program of research and development (R&D):
2. a project to design, build, and operate an MSTR;
3. a project to study and eventually design and build a prototype,
or first commercial, reactor plant;
4. a project to design and build the first of possibly several ''stan-

dardized" plants. .

If adequate guidance is to be provided for an R&D program on MSRs,
it 1s essential that some design activity be started on the prototype
reactor and the MSTR at the beginning of the overall program. (These
initial design efforts may be relatively small, however.) A prototype
concept is required to define the systems tc be tested in the MSTR, and
the MSTR design is required to guide the initial phases of the R&D effort.

If such a program were started in FY 1980, the development and de-
sign activities could probably support authorization of a test reactor
in FY 1985, and such a reactor could probably be built by 1995. The
prototype commercial plant (supported by earlier design study) could be
authorized approximately on completion of the MSTR, and the authorization
for the first standard plant (if desired) could follow about 5 years ‘
later.

Although the technology development effort is shown as only a single

line on Fig. S.2, it represents a multifaceted effort in support of all

 
o A . -

   

 

 
   

         

  

 

 

 

 

 

 

 

ORN DWG T8 70682R
Tha N0 t‘;" I"“'—'\v.‘;_ YEerRA T
R&D @ TECHNOLOGY DEVELOFMENT »
LOPERATE \ UPERATE FOR
S DEVELOPVENT
/
l /
AUTHORIZE l /
| /
aT R o N LA o P j f
N J/ DETAILED DESIGN | !
: /
| | /
| ’ /
i i /
h / CONSTRUOT QFERATE w
! E / 9(45 _ ¢ (L 2005 - <
| I / o
! AUTHORIZE | /
VEAD | /

 

  
 
    

DETALLED DESIGN

(“"‘:"'\'W:f:E-fiRC‘AL PRECONZEPTUAL | {CUNCEPTUAL L » - ONPRFLIN NARY
BROTOTYPE / DESIGN DESIGN - DESIGN

L
E’
|
|
|
|
|
n

 

iAUTHORiZE

1) +irN
4 4

STANDARD
TESIGN

UMSR

 
 

Fig. S.2. Potential reactor construction schedule for commercialization of DMSREs.
xviii

the reactor construction projects throughout the program. This effort, g
which is described in some detail in the body of this report, is summa-
rized in Table S.l1 alcong with estimated costs (in unescalated 1978
dollars). This tabulation includes the estimated costs for development
of full reprocessing capability. A substantial fraction of these costs
(shown as $147 million over 32 years) might be deferred or saved if
development of on~line fuel reprocessing were deferred or eliminated.
The work for the first 15 vears is shown on an annual basis, with most
of the effort in support of the MSTR. In general, the funds shown here
are consistent with the more detailed tabulations presented in the body
of this report. However, in a few areas the development plan indicates
that additional, undefined costs could be expected in some vears. For
purposes of this summary tabulation, funds were arbitrarily added in
those areas to cover reasonable extra costs. Costs after the first 15
years are much less certain and are shown as totals only. The estimated
cost of the total base program is approximately $700 million. The costs
of the reactor construction projects, about $600 million®* for the MSTR
and possibly $1470 million* for the prototype, bring the estimated total
program cost to about $2.8 billion. Since it is impossible to foresee
all needs and costs for a program, this is probably az minimum figure. A
contingency allowance should be added in a subsequent planning stage, as
well as allowances for cost increases due to inflation and escalation and

for any development contributions provided by industry.

 

*

These figures include the costs of integral chemical processing
facilities and are consistent with Nonproliferation Alternative Systems
Assessment Program (NASAP) guidelines.

 
 

W and development costs for MSR hase development progran

(thousands of 1978 doilars}

Projected researc

 

 

Buvelnpmefit activity

 

Reactor design and
analysis

Reacter and component
technology

safety and licensing

Fuel and coolant
chemistry

Analytical chemisrry

Process materials

Fuel processing
technology

Structural alioy

Moderator graphite

<.

b A
Includes funds authorized for major development facility.

Type fund

Target
reactel

 

Coat by fiscal year

 

 

 

 

 

Operating
Qperating

Qperating
Operating
Capital

Operating
Operating

Operating
Operating
Capital

dperating
Gperating
Capital

Operating
Capital

Operating
Operating
Capital

Operating
Operating
Capital
Qpevrating
Uperating
Capital

“Total funds rhrough 2011:

 

 

MSTR
Dema
MSTR
Demo
Al

MSTR
Demo
MSTR
Deno
All

MSTR
Demo
ALl

MSTR
All

MSTR
Demo
All

MSTR
Deama
ALl

MSTR
Demo
A11

. . e
Tatal funds

 

L o . e
Includes cnats estimated without detailed propram anaiysis.

$702, 563,

—— e e

 

 

 

 

 

 

 

 

 

 

 

Cost from 1993a
through 2011

 

1,000

20,300

26,000
BG, 000

8,000
8,360

40,000

3,000

15,000

2,500
2,000
5,300
l,QfIQ
1,420

g

J

L
¥
wn

o

oo

3,000
8,000
1,980

331,685

 

 

XTIX

 
XX

REFERENCES

M. W. Rosenthal et al., The Development Status of Molten-Salt Breeder
Reactors, ORNL-4812 (August 1972).

H. E. McCoy, Jr., Status of Materials Development for Moltern-Salt
Feactors, ORNL/TM-5920 (January 1978).

G. T. Mays, A. N. Smith, and J. R. Engel, Distridution and Behavior
of Tritium in the Coolant-Salt Technology Facility, ORNL/TM-5759
(April 1977).

Molten-Salt Reactor Program, Semianmnual Progress Report for Periocd
Ending February 289, 1976, ORNL-5132 (August 1976).

Foster-Wheeler Energy Corporation, Task I Final Report, Design Studies
of Steam Generatore for Molten-Salt Reactors, ND/74/66 (December 1974).

J. R. Engel et al., Conceptual Design Characteristics of a Denatured
Molten-Salt Reactor with Onece-Through Fueling, ORNL/TM (in prepara-
tion).

J. R. Engel

rpt g e
Utirlisation

t al., Molten-Salt Reactors for Efficient Nuclear Fuel
/. thout Plutnotum Separation, ORNL/TM-6413 {(August 1978).

o

=~

L. E. McNeese et al., Program Plan for Development of Molten-Salt
Bregder HReactors, ORNL-5018 (December 1974).
 

 

DEVELOPMENT STATUS AND POTENTTIAL PROGRAM FOR
DEVELOPMENT OF PROLIFERATICN-RESISTANT
MOLTEN-SALT REACTORS

J. R. Engel

H. F. Bauman W. R. Grimes

J. F. Dearing H. E. McCoy, Jr.
ABSTRACT

Preliminary studies of existing and conceptual molten-
salt reactor (MSR) designs have led to the identification of
conceptual systems that are technologically attractive when
operated with denatured uranium as the principal fissile fuel.
These denatured MSRs would alsc have favorable resource-utili-
zation characteristics and substantial resistance to prolifera-
tion of weapons—usable nuclear materiais. This report presents
a summary cof the current status of technology and a discussion
of the major technical areas of a possible base program to de-
velop commercial denatured MSRs. The general areas treated are
(1) reactor design and development, (2) safety and safety re-
lated technology, (3) fuel-coolant behavior and fuel process-—
ing, and (4) reactor materials.

A substantial development effort could lead to authoriza-
tion for construction of a molten-salt test reactor about 5
vears after the start of the program and operation of the unit
about 10 years later. A prototype commercial denatured MSR
could be expected to begin operating 25 years from the start
of the program.

The postulated base program would extend over 32 years and
would cost about $700 million (1978 dellars, unescalated). Ad-
ditional costs to construct the MSTR — $600 million — and the
prototype commercial plant — $1470 miilion — weould bring the
total program cost to about $2.8 billion. Additional allow-
ances probably should be made to cover contingencies and in-
cidental technology areas not explicitly treated in this
preliminary review.

INTRODUCTION

A concept for a proliferation-resistant molten-salt reactor (MSR)

2 2 , i
33y and/or 2°°U has been evolved in response to

fueled with denatured
the interest in proliferation-resistant power reactors for worldwide
use. Briefly, such a reactor (1) must not provide a tempting or readily

available source of weapons material; (2) must have good economics and
fuel utilization and be competitive with reactors generally used or
planned for use in nuclear weapons states; and (3) must provide reason-
able energy independence for the nonnuclear weapons states that adopt
it (i.e., an assured source of fuel and/or reprocessing capability).

The proposed denatured molten-salt reactor (DMSR) concept, described

in general below, meets these requirements for the feollowing reasons:

1. the fissile material is denatured and/or confined within a contained
highly radicactive system;

2. the projected economic performance is competitive with other existing
or proposed reactor systems;

3. uranium resources would support at least five times the electrical
capacity in DMSRs as in light-water reactors {(LWRs) on a once-through
cycles

4, each DMSR, as a break-even breeder {conversion ratio = 1.0) with on-
line processing, once started would not need an ocutside source of
fisgile fuel indefinitely. (However, fertile material and the makeup
salt constituents, 'Li and beryllium fluorides, would have to he (

supplied.)

At least two other MSR concepts may be attractive for proliferation-
resistant systems. Their development will not be specifically considered
in this report; however, they differ only in detail from the DMSR, and
their development would require the solution of essentially the same prob-
lems. The two concepts are a partially self-sustaining DMSR without on-
line processing and a plutonium-thorium MSR designed to consume plutonium
and produce 2337 for use in denatured reactors.

The development of a DMSR without on-line processing would be a
relatively modest extension of current technolegy and could presumably
be accomplished in a shorter time and with considerably less development
effort than the proposed DMSR with processing. This version could not
be a break-even breeder but would still be a high-performance converter
with significantly improved fuel utilization over LWRs. Addition of an ¢
on—-line fission product processing facility at some later date would
transform the plant into a breeder. Preliminary results indicate that

a fuel charge could last for the entire 30-year life of the reactor, at
 

 

75% capacity factor, with only routine additions of 238y and/or denatured
2353, A more detailed characterization of this concept is in progress.

A plutonium-fueled MSR could be designed for use in a secure energy
center as a "fuel factory" to produce U for use in denatured reactors.
The outstanding advantage of an MSR for this application is the ability
to remove product 233y from the circulating fuel about as fast as it is
formed, so that very little is consumed by fission within the reactor it-
self. Cycle times of V10 days for uranium removal are considered feasi-
ble, compared with reprocessing times on the order of years for seclid-
fuel reactors. An MSR on this fuel cycle has been estimated to produce
750 kg of 233y per GW(e)evear at 0.75 plant factor; this is several times
more than that produced by any other type of thermal reactor and about
the same as expected from a plutonium-thorium liquid-metal fast breeder
reactor (LMFBR). However, the MSR would consume half again as much plu-
tonium or more. More quantitative fuel cycle data are not available at
this time.

The nominal DMSR with processing is based on the design for the
MSBR, as given in Refs. 2 and 3, with several important changes:

1. The start=-up fuel is 235y (or %°*U) denatured with *°°U rather
than fully enriched uranium. Sufficient 238y is fed along with thorium
to keep the fuel in the reactor denatured.

2. The process is altered so that protactinium and plutonium are

not isolated from the fuel galt. Protactinium, which decays to 233

U,
would otherwise be a source of undenatured fissile uranium.

3. The reactor core is larger with a lower power density to reduce
parasitic neutron absorptiong in protactinium as well as in fission prod-
ucts. The power density is reduced sufficiently that replacement of the
moderator graphite in the core is not required due to fast neutron damage

during the lifetime of the reactor.

This version of the DMSR is described in greater detail in Ref. 4.

The MSR research and development was conducted largely at Oak Ridge
National Laboratory, but with assistance by subcontractors and others, in
a nearly continuous program for more than 25 years. The effort included

many large engineering experiments and the design, construction, and
I~

operation of two experimental reactors as well as many small-scale ex-
periments in all fields of pertinent nuclear and materials science. For
nearly 20 years, that effort was directed to M5Rs for the generation of
central station electricity, with the primary focus cn an MSBR breeding
2350 from 2°%Th in the thermal system. The large, varied, and impressive
accomplishments of that program, along with the additional development
requirements needed for demonstration of the MSBR, were described’
thoroughly as of mid-1972.* That material was updated, and a detailed
description (along with a preoposed schedule and costs) of remaining R&D
items was presented’ in 1974. The status of molten-salt technology as

of late 1974 and the additional needs of the MSBR, accordingly, are
documented fairly well, as is the base program of research and tech-
nology development required to fill those needs.

A large fraction of the technology developed for the MSBR is appli-
cable directly, or with a minimum of additional experimentation, to the
DMSR. Morecover, most of the additional technology needed for the MSBR
is also needed for the DMSR. However, the two reactors differ in some
important regards. Accordingly, the technology development required for
a DMSR is likely to invelve significant redirection from that anticipated
for the MSBR, particularly if the technclogy for on-line processing of
the fuel salt is developed in conjunction with the reactor technology.
In some areas (e.g., chemistry and chemical processing), this redirec-
tion probably would increase the requisite development effort, while in
others (e.g., safety technology and graphite development),'the required
effort could decrease.

An additional consideration is the fact that a small MSR technology
development effort was reestablished in mid-1974 and continued for about
two yvears. Although this effort was limited in scope, some significant
acéomplishments were achieved that also affected the current effort to
identify further technology needs.

To the extent that it was applicable, the 1974 program plan was

used as a basis for this review and projection of the technology needs

 

“The MSR program was closed out in early 1973 and remained in that
state for about one year.

 
 

 

and program plan for a DMSR. The present document is focused on the
major development areas, with the recognition that significant develop-
ment efforts could be required in other related areas. (Such ancillary
activities would add somewhat to the overall development cost but
probably would not appreciably affect the total schedule.) In a number
of areas where little or no technical effort has been expended since
1972 and where the perceived needs are substantially the same, the
tasks, schedules, and costs were transposed directly from the 1974

plan with only adjustments of the costs to account for inflation be-
tween 1974 and 1978. 1In other areas, minor adjustments were made to
account for changes in either the technology status® or the apparent
development needs.

The areas with the greatest potential for change from the 1974
program plan are those related to the chemical processing of the fuel
salt. If the once-=through version of the DMSR were developed, it might
be possible to defer development of the reductive-extraction—metal-
transfer process and thereby reduce the overall development cest for
the DMSR. However, because the availability of this process would
substantially improve the fuel utilization of a DMSR, the development
needs, schedules, and costs for it were included in this plan. Thus,
some latitude would exist in the implementation of the program plan.

The remainder of this report consists of four major parts, each
prepared by a single primary author, which deal with the following major
areas of base technology development: (1) reactor design and development,
(2) safety and licensing, (3) fuel-coolant behavior and fuel processing,
and (4) reactor materials. The parts (which contain up to four chapters)
should be regarded as units because of the high level of interdependence
among the subjects treated. However, the base program needs and their
projected costs and schedules are developed separately within each

chapter.

 

KLess than $10 million has been expended on MSR development since
1973, so the changes can have little effect on the overall program cost.
6
References

J. R. Engel et al., Conceptual Design Characteristics of a Denatured
Molten-Salt Reactor with Once-Through Fueling, ORNL/TM (in prepara-
tion.

R. C. Robertson, ed., Conceptual Design Study of a Single-Fluid Mol-
ten-Salt Breeder Reactor, ORNL-4541 (June 1971).

Ebasco Services Inc., 1000 MiW(e) Molten-Salt Breeder Reactor Concep-
tual Design Study, Final Report, Task 1, Subcontract No. 3560 with
Unicen Carbide Corporation, Nuclear Divisien (February 1972).

J. R. Engel et al., Molten-Salt Reactors for Efficient Nuclear Fuel
Utilization without Plutowium Separation, ORNL/TM-6413 (August 1978).

M. W. Rosenthal et al., The Development Status of Molten-Salt Breeder
Reactors, ORNL-4812 {August 1972).

L. E. McNeese et al., Program Plan for Development of Molten-Salt
Breeder Reactors, ORNL-5018 (December 1974).
 

PART I. REACTOR DESIGN AND DBEVELOPMENT

H. ¥. Bauman

The objectives of the reactor design program are to first develop
a fairly complete conceptual design for a full-scale [1000-MW(e)} DMSR
in order to define more clearly the development problems that must be
addressed. Then, concurrent with the technology development, a molten-
salt test reactor (MSTR) would be designed, built, and operated to
demonstrate all aspects of the required DMSR technology on a smaller
scale. The scale of the MSTR would be decided as the program progressed.
The MSER conceptual design, which the DMSR would probably follow, pro-
posed four fuel heat-exchanger and steam-generator modules of 250-MW(e)
capacity each. The scale suggested for the MSTR would lie in the range
of 100-MW(e) power [with two 50-MW({e) steam-generator modules {i.e.,
1/5 scale)] to 250-MW(e) power with a single (i.e., full-scale) steam-—
generator module. The data from the component technology develcopment
program would be fed into the reactor design efforts, and the experi-
ence obtained in reactor construction and operation would, in turn, guide
continuing effort in component development. An MSTR mockup is proposed
which would permit integrated testing of most of the reactor components
before the MSTR itself is built. Finally, the construction of the proto-

type DMSR would influence the development of a standardized DMSR design.
 

 
 

 

1. REACTOR DESIGN, ANALYSIS, AND TECHNOLOGY DEVELOPMENT

There is considerable experience in the engineering design and neu-
tronic analysis of MSRs, through final design, construction, and opera-
tion. The molten—-salt reactor experiment (MSRE) has been through the
entire design process, and the ORNL reference concept MSBR' has reached
the stage of detailed conceptual design. In addition, the various reactor
components and subsystems received intensive development effort in the

technology development programs.

Status in 1977

At the time the development—status report2 was prepared (1972}, a
detailed conceptual design had been developed for the single-fiuid MSBR.
Furthermore, many alternative designs had been investigated, generally
in lesser detail; one of these, also pertinent to the DMSR, was a low-
power-density core design3 in which the core moderator graphite would
have an expected lifetime equal to the design life of the reactor.

In the area of reactor components and systems, the most important
items had been identified as salt pumps, the coolant system as a whole,
heat exchangers, the entire steam system, valves, control rods, fuel
storage, and gas handling.

Ixperience had been obtained in the MSRE in all the above areas ex-—
cept the steam system. However, new developments were proposed in sev-
eral areas, such as the use of godium fluoroborate as the secondary
coolant (rather than lithium~-beryllium fluoride), the use of mechanical
valves in addition to freeze wvalves®, and the addition of a fuel-salt

gas sparging system (rather than sparging of salt in the pump bowl).

Current Development Status

 

The reactor design and analysig effort since 1972 has been minimal.

However, a major study of tritium transport in molten-salt systems was

 

~

A freeze valve is essentially a short, flattened section of pipe
which can be cooled to form an internal plug of frozen salt and subse-
quently reheated to thaw the plug.
10

carried out in 197476, including experiments in a secondary-coolant—salt

test loop.

Large quantities of tritium are produced in MSRs from neutron reac-
tions with lithium in the fuel salt. Since elemental tritium can diffuse
through metal walls, such as heat-exchanger tubes, at elevated tempera-
tures, a potential path exists for the transport of tritium to the reac-
tor steam via the secondary coclant loop and the steam generator. Recent
experiments indicate that tritium is oxidized in the proposed MSR second-
ary coolant, sodium fluoroborate, thus blecking transport to the steam

System.

Tritium—addition experiments were conducted in an engineering-scale
ccolant-salt test facility. The results are given in a recent report by
Mays, Smith, and Engelf.l+ The experiments showed that the steady-state
ratio of combined to elemental tritium in the coolant salt was greater
than 4000. A calculation applying this ratio to the case of an operating
MSER indicated that the release of tritium to the steam system would be
less than 400 CBg/d (10 Ci/day). The conclusion of the study was that
the release of tritium from an MSR using sodium fluoroborate in the
secondary coolant system could be readily controlled within NRC guide-

lines.

In the area of component development, there was an effort to advance
the steam system development from the conceptual design proposed for the
M5BR toward hardware development. In 1971, Foster-Wheeler Corporation was
awarded a contract for a study of MSR steam-generator designs. The con-
tract was suspended in 1973 and then reinstated in 1974 to complete the
first task (in a four-task contract) — the conceptual design of a steam
generater to meet specifically the steam and feedwater conditions postu-
lated for the ORNL reference-design MSBR. This task was successfully
completed, and a report was issued in December 1974.° A design was pre-
sented which, based on analysis, would meet all the requirements for an
MSR steam generator. Because of the termination of the MSR project, the
design did not receive experimental verification or further analytical

study.

 
 

11

The coclant-salt test facility mentioned in connection with the
tritium experiments was operated in the period 1974—76 to obtain engi-
neering-scale experience with sodium fluoroborate, the proposed MSBR
secondary coolant. The loop was operated successfully for >5000 hr at
temperatures up to 540°C. The tests indicated that the engineering
characteristics of sodium fluoroborate would be suitable for MSR second-
ary coolant.

3
1“SXe3 a gaseous

One of the important advantages of MSRs is that
fission product with an extremely high thermal-neutron cross section,
is not scluble in the fuel salt and can be rapidly removed from the
system. Fission-product 135%e is the single most important parasitic
absorber of neutrons in thermal reactors, and the high conversion ratio
of MSRs depends on efficient 135%e removal. The xenon has twe exits
from the fuel salt: by absorption into the pores of the core graphite,
where it would remain as a neutron poison, and by diffusion into the
fuel cover gas (helium), where it can be removed from the system. A
helium-stripping system is proposed for MSRs in which fine bubbles of
helium are introduced into the fuel salt to provide a sink for xenocn and
are subsequently separated from the salt to remove the xenon effectively.
Gas-bubble generators and strippers had been designed and were to be
tested in a circulating salt loop when the program was ended in 1973.
Although some additional work was done in the 1974—75 period, an integral

test of the stripping system was not completed.

DMSR Development Needs

 

A1l the design and development needs described for the MSBR in the
1974 program plan6 would alsc be required for the DMSR. However, some
aspects of the program should be emphasized for the DMSR.

The core design and analysis for both the DMSR and the MSTR require
particular attention to the effects of 238U, protactinium, and plutonium
on the system fissile inventory and conversion ratio. Another important
point is the core fast-flux distribution and its effect on the design
life of the moderator graphite. Some prelimipnary neutronic calculations

for a proposed DMSR design are given in Ref., 7. Since both thorium and
12

2380 are important neutron resonance absorbers, the lumping of fuel and
the degree of thermalization of the neutron spectrum in the reactor core
and reflector regions are important variables in the neutronic analysis.
An extensive reactor-analysis program would be required to select optimized
core designs for the MSTR and DMSR.

A very important problem for all molten-salt systems is the control
of the reactor power and the temperatures in the fuel salt, the coolant
salt, and the steam generators and the management of thermal cycles and
thermal stresses in all parts of the reactor system. Although consid-
erable experience in MSR thermal problems was obtained in the operation
of the MSRE, the addition of a steam system and turbine-generator to
the next generation of MSRs leads to considerably more complex interac-
tions between the various components of the reactor systems. The analy-
sis of thermal-hydraulic dynamic behavior of the proposed reactor systems
will require the development of suitable computer models. Some prelimi-
nary analysis along these lines had been done for the MSBR, but a major
extension of this program would be required for the DMSR.

The reactor technology effort encompasses the development of all
the major components required in the reactor system. For the most part,
highly specialized components such as the vessels and contactors for
chemical processing and specific instrument items would be covered
within the development programs for the associated technologies. How-
ever, some of the more general items {e.g., small salt pumps, valves,
seals) might be included in the overall develepment program for reactor
components. Such items probably would not greatly affect the total
Program cost.

A vital part of the development program for reactor components is
the design of intermediate- and full-scale salt pumps. The pump preferred
for molten-salt service is a vertical-shaft, centrifugal sump pump such
as was used successfully in the MSRE and cother test facilities. The
drive shaft for this type pump extends through the reactor shielding so
that the motor is relatively accessible and is in a low-radiation field.
The scaleup of these pumps is proposed to proceed by extending the line
of past development with as little change in conceptual design as possi-

ble. The nominal capacity of the MSRE pumps was 0.08 m°/s (1200 gpm) ;

 
 

 

13

the pumps proposed for the MSBR, and required for any type of full-scale
MSR, would be larger by a factor of about 20. The line of development
may include an intermediate-size pump rated at about 0.3 m’/sec (5000 gpm).

The design of mechanical vaives for melten-salt service and the use
of freeze valves in large systems are alsc very important developments.
The MSRE was operated entirely without mechanical valves; freeze valves
were used where flow shutoff was required. Freeze valves have been used
in pipe sizes up te 1 1/2 in. IPS and have been found to be reliable,
zero-leakage devices so long as the integrity of the pipe itself is
maintained.

The disadvantages of freeze valves are that (1) flow in a line must
be stopped before a plug can be frozen, (2) they open and close relatively
slowly, and (3) they cannot be used for throttling. Therefore, mechanical-
type salt valves are considered essential to the operation of large MSRs.
Experience with bellows-sealed mechanical valves in molten-salt experi-
mental facilities has been limited. The main development problems are
finding materials that will close tightly without binding in molten salt
and developing reliable zero-leakage seals. Freeze valves, perhaps in-
tegral with mechanical valves, may also be developed for the larger
systems.,

Reactor control rods have been included in MSR designs, although
control rods serve a limited purpose in fluid~fueled reactors. The
reactivity of the core is controlled mainly by the composition of the
fuel and by changes in the density of the fuel with temperature. The
ultimate shutdown of the reactor is achieved by draining the fluid fuel
to storage tanks. Control rods are required for short-term fine control
of the temperature and reactivity of the core, as well as for rapid
shutdown; so the design of control rods for melten-salt service is an
important development need.

The fuel drain and storage system is another vital development. The
fuel storage system must have the capability of removing afterheat to
the ultimate heat sink in the event that the reactor becomes inoperable
or is shut down for maintenance. A drain tank with a natural-convection
NaK cooling system was proposed for the MSBR. Although this system ap-

pears workable, all aspects of fuel contaimment under normal and accident
14

conditions deserve further attention. Because of the relatively low power
density in the fiuid fuel (compared to solid fuel), it is likely that an
MSR could he develecped for which a containment melt-through accident

would not be considered credible.

An MSTR mock-up is proposed which would permit integrated testing
of the reactor components under zero-power conditions. The mock-up
would permit solving ot layout and remote maintenance problems before
the reactor was built.

Possible schedules for the first 15 years of the DMSR develiopment
program are shown for the reactor design and analysis work in Table 1.1
and for the technology development work in Table 1.2Z.

Operating fund requirements for this peried for reactor design and
analysis are given in Table 1.3, and operating and capital fund require-

ments for technology development are given in Tables 1.4 and 1.5.

References

1. R. C. Robertson, ed., Conceptual Design Study of a Single Fluid
Molten-Salt Breeder Reactor, ORNL-4541 (June 1971).

2. M. W. Rosenthal et al., The Development Status of Molten=Salt Breeder
Reactors, ORNL-4812 (August 1972).

3. Molten-Salt Reactor Program, Semiannual Progress EReport for Period
Imding August 31, 1970, ORNL-4622 (January 1971).

b, G. T. Mays, A. N. Smith, and J. R. En%el Digtribution and Behavicr
of Tritium in the Coolant-Salt Technology Facility, ORNL/TM-5759

(April 1977).

5. Foster-Wheeler Energy Corporation, Task I Final Report, Design Studies
of Steam Generators for Molten Salt Feactors, ND/74/66 (December 1974).

6. L. E. McNeese et al., Program Plan for Development of Molten-Salt
Breeder Reactors, ORNL-5018 (Deaembhr 19747 .

7. J. R. Engel et al., Molten-Salt Reactors for Efficient Nucleqr Fuel
Liaation without Plutonium Separation, ORNL/TM-6413 (August 1978).

 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 1.2, Schedule for work on reactor technology development
Fiscal year
Task
1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
§ 2 3
Fuel-salt technclogy v v v
‘ ant-salt technclogy v* v°
6 7 a8 ° 10 1
. .m system technoclogy v_ v v v v V1
Cover~ and cff- 12 3
gas system Vv Vv
technology
14 5
Salt pump development v
. 18
Primary heat—exchanger v
development
17 {8 e
Valve development v v v
VZO
Contrel rod development
Co . . 21 22
‘ontainment and cell heating v v
development
23
Components Test Facility _....y__.— V24 v?S
Milestones:
1. Gas-Systems Technolegy Facility water tests will be fin- 12. Develeopment of methods for handling gaseous effluents (in-
ished and construction completed so that salt operation cluding fission products, tritium, and BF;) from the off-
can start. gas systems should be complete,
2. Sufficient tests will have been completed to indicate that 13. All other problems associated with the cover- and off-gas
the efficiency of the bubble generator—bubble separator systems should be resolved.
is satisfactory and that mass-transfer rates are adequate 14. The design of the MSTR prototype pump and pump test stand
to permit detailed design of the xencn removal system for should be complete.
an MSTK. Additional development will be done to refine 15. The construction of the MSTR prototype pump and pump test
the results and test the effects of other variables. stand should be completed and operational tests will be
3. All problems pertaining to the behavior of tritium in the started.
fuel-salt system will be resclived. 16. All development work on the primary heat exchanger prepa-
4. Test for determining the behavior of tritium in the coel- ratory to design of the MSTR will be completed.
ant system will be completed. Corrosien-product removal 17. Preliminary valve development needed to proceed with design
studies will be cempleted. of the MSTR will be finished.
5. Large-scale denonstration tests of coolant-salt technology 18. Final development of specific valves for the MSTR will be
should be completed. finished. -
6. The feasibility of using lower feedwater temperatures will 19, Preliminary valve development for prototype DMSR will be
be determined. This may affect the subsequent design and completed.
development of the steam system components. 20. All development needed for the MSTR control rods will be
7. The plan for a2 steam-generator R&D program should be com- completed.
pleted. 21. Exploratory studies and preliminary development needed for
8., The small-scale steam generator work should have progressed the design of the MSTR containment and cell heating should
to a stage that will permit reevaluation of the R&D program. be completed.
9, The construction of the steam-generator tube test stand, 22. Testing of the containment and cell heating design for the
pressure relief system, and the 3-MW test assembly should MSTR should be completed.
be complete. 23. The design of the Components Test Facility should be suf-
10. Testing in the steam—generator tube test stand should be ficiently complete to start comstruction.

finished. 24, Construction of the Components Test Facility should be
11. Construction of the steam-generater model test installation, completed.

the pressure relief system, and the 30-MW model steam gen— 25. Design of component test facility for prototype DMSK com-

erator should be complete and operational tests will be
started.

)

pleted.
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 1.2. Schedule for work on reacter technology development
Fiscal year
Task
1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 19%C 1991 1992 1993 1994
1 2 3
Fuel-salt technology v v v
. 74 5
ant-salt technology v
& T 8 9 10 11
.« .m system technology v v v v V'V
Cover- and off-gas system ‘{7?2 VB
technology
14 15
Salt pump development v
Primary heat-exchanger §746
development
17 fa e
Valve development v v v
§720
Contrel rod development
Containment and cell heating Vm V22
development
23 24 25
Compounents Test Facillry v Vv v
Milestcones:
1. Gas-Systems Technology Facility water tests will be fin- 12. Development of methods for handling gaseous effluents (in-
ished and construction completed so that salt operation cluding fission products, tritium, and BF3) froem the off-
can start. gas systems should be complete.
2. Sufficient tests will have been completed to indicate that 13. All other problems associated with the cover- and off-gas
the efficiency of the bubble generator—bubble geparator systems should be resolved.
is satisfactory and that mass-transfer rates are adequate 14, The design of the MSTR prototype pump and pump test stand
to permit detailed design of the xenon removal system for should be complete.
an MSTR. Additional development will be done to refine 15. The construction of the MSTR prototype pump and pump test
the results and test the effects of other variables. stand should be completed and operational tests will be
3. All problems pertaining to the behavior of tritium in the started.
fuel-salt system will be resclved, 16. All development work on the primary heat exchanger prepa-
4, Test for determining the behavior of tritium in the cocl- ratery to design of the MSTR will be completed.
ant system will be completed. Corrosion-product removal 17. Preliminary valve development needed to proceed with design
studies will be completed. of the MSTR will be finished.
5. Large-scale demonstration tests cof coolant-salt technolegy 18. Final development of specific valves for the MSTR will be
should be completed. finished. -
6. The feasibility of using lower feedwater temperatures will 19. Preliminary valve development for prototype DMSR will be
be determined. This may affect the subsequent design and completed.
development of the steam system compeonents. 20. All development needed for the MSTR control rods will be
7. The plan for a steam-generator R&D program should be com- completed.
pleted. 21. Exploratory studies and preliminary development needed for
8. The small-scale steam generator work should have progressed the design of the MSTR containment and cell heating should
to a stage that will permit reevaluation of the R&D program. be completed.
9. The construction of the steam-generator tube test stand, 22, Testing of the containment and cell heating design for the
pressure relief system, and the 3-MW test assembly should MSTR should be completed.
be complete. 23. The design of the Components Test Facility should be suf-
10. Testing in the steam-generator tube test stand should be ficiently complete to start comstruction.

finished. 24, Construction of the Components Test Facility should be
11. Construction of the steam-generator model test installation, completed.

the pressure relief system, and the 30-MW model steam gen— 25, Design of component test facility for prototype DMSR com-

erater should be complete and cperational tests will ke
started.

)

pleted.

 
 

Table 1.3. Operating fund requirements for reactor design and analysis

 

 

Cost (thousands of 1978 dollars) for fiscal year —
Task

 

1980 1981 1982z 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994

 

Design studies of MSR power plants 340 930 620 230 270 520 520 540 520 360 300 300 300 300 300
Design technology 70 210 300 230 190 130 100 100 160 100 100 100 100 100 100

 

 

 

 

Codes and standards 50 90 170 250 260 170 170 260 260 260 260 260 260
Licensing of MSRs 40 170 250 330 260 260 260 260 260 260 260 260
Nuclear analysis of MSR power 20 130 130 130 130 170 100 100 100 100 100 100 100 100 1066
plants
Total fundsa 430 1270 1100 720 930 1320 1170 1170 1150 1080 1020 1020 1020 1020 1020
Allocation
MSTR 430 1270 1100 720 330 1120 970 970 850 880 520 520 520 100 100

Prototype DMSR 200 200 230 200 200 500 00 500 920 920

 

 

aTotal funds through 1994: $11,440.

LT
 

Table

1.4, Operating fund requirements for work on reactor technology development

 

Cost (thousands of 1978 dollars) for fiscal year —

 

 

 

 

 

 

Task

1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
Fuel~salt technology 24¢ 300 300 210 330 570 260 200 330 260 100 100 100 100 160
Coolant-salt technology 220 380 130 130 260 260 200 200 200 200 200
Steam system technology 70 370 750 780 105G 1420 16206 1600 1600 1600 1,606G 500 500 5G0 500
Cover~ and off-gas system 80 160 80 100 240 140 100 100 160 100 200 200 200

technelogy
Salt pump development 50 180 800 2000 1600 1400 400 400 400 400 400 500
Primary heat-exchanger 100 70 130 200 2006 200 200 200 200

development
Valve development 50 130 130 130 130 130 260 200 200 200 200 300
Control rod development 80 80 80 200 200 200 200 2006 200
Containment and cell heating 80 130 80 130 100 100 100 100 100

development
Components Test Facility 160 106 260 260 580 940 2100 3,000 1000 100C  10GC 1000
MSTR mock-up 820 910 1100 2000 3000 4000 4,000 3000 2000 2000 1000
Total funds® 530 1050 1260 1410 2690 4270 5920 6610 7970 9510 10,100 4000 5100 5100 4300

Allocation

MSTR 530 1650 1260 1410 2690 4270 5920 6610 7970 9210 9,500 5000 3000 3000 1500
Prototype DMSR 300 600 1000 2100 2100 2800

 

aTotal funds through 1994:

$71,820.

81
 

Table 1.5. <Capital fund requirements for work on reactor techmology development

 

 

Task

Cost (thousands of 1978 dollars) for fiscal year —

 

198¢ 1981

1982

1983

1984

1985

1586 1987

1988

1989

1990

1991

1992

1993

1994

 

Fuel-salt technology 40 90
Coolant—-salt technolegy
Steam system technology

Cover— and off-gas system
technology

Salt pump development

Primary heat-exchanger
development

Valve development
Control rod development

Containment and cell heating
development

Components Test Facility

MSTR mock-up

Total funds® 40 90

100
50

30
50

5000

330

70

1,330

13,000
65,000

70

26,000

70

330
100

100

80

260

500

330

800

 

5330

79,400

26,100 570

840

1130

14090

1400

100

500

600

300

500

800

100

300

500

900

100

 

Allocation
MSTR 40 90
Prototype DMSR

150

80

5330

79,400

26,100 570

840

830
300

800
600

600

800

900

1100

 

“Total funds through 1994: $118,530.

6T
 

&

 
 

21

PART TI. SAFETY AND SAFETY-RELATED TECHNOLOGY

J. F. Dearing

Substantial differences between the safety considerations for solid-
fueled nuclear reactors and those for liquid-fueled systems, such as the
MSR, have long been recognized. Consequently, safety studies have been
included in all MSR design and technology development pregrams. How-
ever, comprehensive studies of the safety attributes of commercial-scale
MSRs have been hampered by the lack of a reasonably complete conceptual
design for a large MSR. Thus the area of MSR safety has been subjected
to a great deal of generalization, with very little detailed system-spe-
cific analysis of the type required to define fully the safety charac—
teristics and questions of the MSR concept. Tt is presumed that any
future MSR development program would include significant effort in the
area of safety technology, along with sufficient reactor design effort

to support it adequately.

 

 
 

 

 

 
 

23

2. REACTOR SAFETY AND LICENSING

The MSR concept poses problems in safety and licensing that are
significantly different from those encountered in the present-day gen-
eration of solid-fueled reactors. The successful cperation of the MSRE
and the safety studies of the MSBR concept, however, have already estab-
lished a firm basis for identifying and solving these problems. A com-
prehensive summary of safety and safety-related technology status and
development needs of the MSBR as of 1972 is included in Ref. 1. An up-
date as of 1974 that presents a detailed plan for future work is included
in Ref. 2. Although many of the safety analyses contained in these two
documents will have to be carefully evaluated for applicability to the
DMSR (especially reactor core neutronics and thermal hydraulics), the
overall assessment of technoleogical status and future development needs

is expected to be applicable.

Status and Development Needs

 

Safety

The main feature of the DMSR which sets it apart from the other
solid fuel reactor types is that the nuclear fuel is in fluid form (mecl-
ten fluoride salt) and is circulated throughout the primary coolant sys-
tem, becoming critical only in the graphite-moderated core. The possible
problems and engineered safety features associated with this type of
reactor will be quite different from those of the present day LWR and
LMFBR designs. In the DMSR, the primary system coolant serves the dual
role of being the medium in which heat i1s generated within the reactor
core and the medium which transfers heat from the core to the primary
heat exchangers. Thus the entire primary system will be subjected to
both high temperatures (700°C at core exit) and high levels of radiation
by a fluid containing most of the daughter products of the fission pro-
cess. Because of the low fuel salt vapor pressure, however, the primary
system design pressure will be low, as in an IMFBR. The entire primary
coolant system is analogous, in terms of level of confinement, to the

cladding in a solid fuel reactor. Although much larger, it will not be
24

subjected to the rapid thermal transients (with melting) associated with
LWR and LMFBR accident scenarios. Two additional levels of confinement
will be provided in the DMSR, in accordance with present practice. The
problem of develcoping a primary coolant system which will be reliable,
maintainable (under remote conditicns), inspectable, and structurally
scund over the plant’s 30-year lifetime will probably be the key factor
in demonstrating ultimate safety and licensability.

It ig the breach of the primary coolant system boundary, resulting
in a large spill of highly radicactive salt into the primary containment,
which will probably provide the design basis accident. The analogous
level of occurrence in a solid fuel reactor would be major cladding fail-
ure. Possible initiators of this accident include pipe failure, missiles,
and pressure or temperature transients in the primary coolant system.
Failure of the boundary between the primary and secondary salt in the
IHX could be especially damaging. In the event of a salt spill, a possi-
bly redundant system of drains would be activated to channel the salt to
the cooled drain tank. The primary system containment, defined as the
set of hermetically sealed, concrete-shielded equipment cells, would
probably not be threatened by such a spill, but cleanup operations would
be difficult.

A unique safety feature of the DMSR is that, under accident shutdown
conditions, the fuel material would be led to the "ECCS" (represented by
drain tank cooling), rather than vice versa. The reactor and containment
must be designed so that the decay heated fuel salt reaches the drain
tank under any credible accident conditions. In any case, the decay
heat is associated with a very large mass of fuel salt, so that melt
through, or "CHINA SYNDOME," does not appear to be a problem.

The safety philosophy for accidents involving the reactor core is
very different for fluid-fueled reactors than for solid-fueled ones be-
cause the heat source is {(mainly) in the coolant and not in a solid,
which requires continuous cooling to avoid melting. An LMFBR, for ex-
ample, has a tremendous amount of stored energy in the fuel pins which
much be removed under any accident conditions. Dryout, which leads to
almost immediate meltdewn in an IMFBR, would not be nearly as severe in
the DMSR because the heat source would be removed along with the cooling

capability.

 

 
 

 

25

Under normal conditions, the reactor power is stable to all fre-
quecies of oscillation, because the negative prompt component of the
temperature coefficient of reactivity dominates the poesitive delayed
component. There appear to be no safety problems in the area of reac-
tivity control. Because the fuel salt will be channeled to the cooled
drain tank under many off-normal conditions, that tank probably must
have redundant systems for decay heat removal. There is no credible
means for achieving recriticality once the fuel salt hag left the

graphite~moderated core.

Licensing

Although two experimental MSRs have been built and operated in the
United States under government ownership, none has ever been subjected
to formal licensing or even detailed review by the Nuclear Regulatory
Commission {(NRC}. As a consequence, the question of licensability of
MSRs remains largely open; the NRC has not yet identified major licens-
ing issues, and the concept has not been considered by various public-
interest organizations who are often involved in nuclear-plant licensing
procedures. Furthermore, the licensing experience of solid-fueled
reactors can be used as only a general guide because of significant
fundamental differences between those systems and MSRs. Presumably
MSRs would be required to comply with the intent, rather than the letter,
of NRC requirements, particularly where methods of compliance are concept-
specific.

Before any MSR ig licensed, it probably will be necessary to define
a complete new spectrum of potential transients and accidents, along
with their applicable initiating events, to be treated in safety analysis
reports. Some of the more important safety-significant events for an
MSR were mentioned earlier, but even routine operatiocnal events may have
a different order of importance for this reactor concept. For example,
mcderate reactor power disturbances would not be very important because
one of the principal consequences, fuel cladding failure, is a nonevent
in an MSR. However, a small leak of reactor coolant would be an important

event because of the high level of radicactivity in the MSR coolant. The
26

previous examples of significant differences between MSRs and other 1i-
censed reactors illustrate why a substantial design and analysis effort
would be required first. The reasons are (1) to establish licensing
criteria for MSRs in general and a DMSR in particular and (2) to evaluate

MSR licensability in relation to that of other reactor types.

Estimates of Scheduling and Costs

 

A detailed plan for developing the technology necessary to ensure
the safe operation of the MSBR under normal conditions and safe shutdown
under accident conditions is presented in Ref. 2. There are no signifi-
cant differences between the MSBR and DSMR concepts in terms of necessary
safety and safety-related technology development; thus the program sched-
ule and operating fund requirements for the MSBR safety technology de-
velopment program found in Ref. 2 will serve as a first estimate of the
requirements for a DMSR program. Table 2.1 shows the operating fund
requirements (in 1978 dollars) for a safety techneclogy development program

beginning in 1980. ‘

References

1. M. W. Rosenthal et al., The Development Status of Molten-Salt Breeder
Reactors, CRNL-4812 {(August 1972).

2. L. E. McNeese et al., Program Plan for Development of Molten-Salt
Breedeyr Reactors, ORNL-5018 (December 1974).

 
 

Table 2.1. Summary of

operating fund requirements for work on reactor safety

 

 

Cost (thousands of 1978 dollars) for fiscal year —

 

 

 

 

 

 

Task e
1980 1981 1982 1983 1984 1985 1986 1987 19838 1989 1990 1991

Guidance studies 117 303 312 312 247 312 468 507 631 780 845
Fission-product behavior 39 59 78 59 29
Primary systems material 39 117 91 130 39
Component and systems technology 156 208 215 273 325 325
(accident conditions)
Safety instrumentation and controls 65 65 65 65 65
technology
Maintenance technology (postaccident) 65 65 65 65
Safety technology of processing and 65 65

waste storage and handling
Total funds” 117 303 351 468 397 676 839 975 1100 1235 1300

“Total funds through 1990: $7761.
 

 
 

 

29

PART TIf. FUEL-COOLANT BEHAVIOR AND FUEL PROCESSING

W. R. Grimes

The development effort on MSRs prior to 1972 produced a large body
of technical information related to the behavior of potential fuel and
coolant salts and to the fuel processing concept. The primary focus
for most of this development was the MSBR, but much of the resultant
technology is also applicable to the DMSR. However, the two reactor
concepts differ in important regards, and some of the chemistry and ana-
lytical chemistry programs thought to be virtually complete for the MSBR
will require some additional effort for the DMSR. Differences in fuel
processing, necessitated by both the differing fuels and fihe differing
philosophies, will require some additional R&D (and possibly a different
unit process in one area) as well as the engineering demonstration of
the processing steps individually and in integrated operation which the
MSBR also required.

The MSBR program continued after mid-=1974, and the subsequent two
years brought a variety of findings in the behavior of fuel and coolant
salts, in reprocessing and fuel reconstitution, and in fuel-material
interactions. All the findings, of course, contribute changes, some of
which are significant to the required development effort. These three
general topics, along with the relevant analytical chemistry, are treated
here.

The folleowing discussion attempts to briefly define the areas in
which DMSR technology requirements differ substantially from those for
the MSBR; in particular, differences in the reactors or recent advances
in molten-salt technology that require or permit significant changes in
the 1974 MSBR program plan are discussed. The general format for dis-
cussion of the four broad areas of technology includes (1} the key dif-
ferences between the MSBR and the DMSR; (2) post-1974 advances in MSBR
technology; (3) the status of molten-salt technoclogy, with special em-
phasis on additional requirements for demonstration of the DMSR; and (4)
estimates of scheduling and funding requirements of the overall DMSR

program, drawn from the 1974 MSER program plan.
 

 
 

 

31

3. FUE]L AND COOLANT CHEMISTRY

Because both the fuel (primary coolant) and the coolant (secondary
coolant) in an MSR are complex fluids, they are subject to a wide variety
of chemical effects and interactions that have a vital influence on
the overall behavior of the reactor system in which they are used.
Therefore, a detailed understanding of the chemical behavior of these
fluids under both normal and off-normal conditions is essential to the
successful development of the MSR concept. This chapter discusses that

chemical behavior as it applies specifically to the DMSR.

Key Differences in Reactor Concepts

 

Fuel proposed for the MSBR contained thorium (as ThF4)} as the fer-
tile material. The fissile material for the initial loading was highly
enriched 2%°U (essentially as UF4).* At equilibrium, the reactor (whose
fuel was to be continuously processed on a 10-day cycle to remove fission
233

products and to isclate 233pa for decay to U outside the neutron flux)

operated with 2331 45 the fizsile material. Because little *°°U was pres—
ent and the small quantity of plutonium was removed with the protactinium
and was not returned to the circuit, the amount of transuranium elements
in the system was trivial. Fission-product zirconium (present in the
fuel as ZrFy) was kept to a very low concentration since it was removed
with the 2°%Pa and was not returned to the fuel. The fission process in
UFy appears to be inherently slightly oxidizing (and, accordingly, cor-
rosive) to nickel-based alloys such as Hastelloy N; the total valence of
fission-product cations in the melt in equilibrium with Hastelloy N is
slightly less than four per fission event. This tendency toward enhanced
corrosion as a direct consequence of fission was contreclled in the MSRE*

and was to be controlled in the MSBR'»? by keeping a small fraction (ca.

1%) of the uranium present as UFs.

 

® 9

g . . 1
However, PuF3 was considered and given considerable study.

+Benefits from a somewhat higher proportion of UFs; were recognized,
but earlier they appeared to be marginal. Recent work (to be described)
suggests strongly that a higher fraction of UF3 will be very desirable.
In contrast to the MSBR (and since uranium is recoverable with
relative ease by fluorination), the DMSR must never contain fuel in
which the *%°U and 2°*°U are at weapons—usable concentrations; the over-
all uranium concentration must be markedly higher for DMSR fuel since a
large quantity of 2387 must always be present. The reactor is, accord-
ingly, like an LWR — a producer and cconsumer of plutonium.* Isolation

2330 must obviously be

of ?3%Pa outside the reactor for decay to pure
abandoned. Fission-product removal from the fuel necessitates prior re-
moval of protactinium and plutonium (as well as uranium), but the DMSR
is obviously constrained to reintroduce these materials directly to the
reactor. As a consequence, the fuel will contain an appreciable concen-
tration of transuranium isotopes. The constraint also eliminates the
easy removal of fission-product zirconium conceived for the MSBR, and,
unless some invention is realized, the DMSR must accept a low but appre-
ciable concentration of ZrF, in the fuel. Each of the differences above,
of course, implies a nontrivial (though relatively small) change in the
fuel chemistry.% A larger change in fuel chemistry {(and in R&D needs)
would result from a decision to operate the reactor with 5 to 107 of
the uranium present as UF3. A discussion of this issuec 1s presented
in a subsequent section.

Tt should be noted, even in a section such as this, that chemical
behavior and associated R&D needs of the fuels for the MSBR and DMSR

show far more similarities than differences.

Post-1974 Technology Advances

 

The appearance of numerous shallow cracks when Hastelloy N specimens
were tensile tested after exposure to the fuel was a key cbservation from
the otherwise very successful four-year operation of the MSRE. This

cracking behavior, which caused no particular problems in the MSRE but

 

“At equilibrium, about 307 of the fissions in the conceptual DMSR .
will occur in plutonium.

T o c , o
Yields of specific fission products differ among the three fissile .
species. However, the difference in needs for R&D on fission-product be-
havior is probably trivial.

 
 

 

33

which would probably prove intolerable upon extrapolation to an MSBR or
DMSR, was shown to be a consequence of fission-product tellurium.® At
the redox potential (set essentially by the ratio of UF3; to UF4)} in an
MSRE, fission-product tellurium would exist primarily as elemental tel-
lurium and could react with the alloy. Post=1974 R&D served to confirm
this hypothesis, and progress was made in understanding some details of
the behavior of telliurium in molten fuel mixtures *":° and in defining%
alloys {including slight modifications of Hastelloy N) far more resistant
to tellurium embrittlement.® Moreover, a considerable experimental pro-
gram has shown that if the fuel has a sufficiently high ratio of UF: to
UFy (0.02 to 0.1), the impact of added tellurium on ordinary Hastelloy N

is very markedly diminished. ®s”’

It seems clear that at high UF;/UF,
ratios, tellurium is constrained to exist in the fuel as Te?™ (and almost
certainly in solution) and is unable to mount an appreciable attack on
the alloy. This cbservation appears to open many possibilities and may
amount to a real breakthrough.

8 .
A careful remeasurement as a function of temperature and of solvent

composition of the equilibrium,*

has given a reasonable confirmation of the previous measurements  in LiF-
BeFs (66-34 mole %) and has extended them to other LiF-Bel; mixtures and
to LiF-BeF;~ThF, (72-16-12 mele 7%). Stability of UF3; relative to UFu,
increases rapidly with temperature and with a decrease in the free fluo-
ride ion concentration of the fuel solvent.

Adeguate retention of tritium that is generated within the fuel salt
has been known to be a problem for MSRs for vears. Two post-1974 devel-

11

opments at ORNL' %> promise at least a major alleviation of the problem.

 

* :
Tellurium is a member of the sulfur family of elements, and sulfur
is well known to be detrimental to nickel and nickel-based alloys.

TAs described in some detail in Chapter 7 under the heading Status

in 1976.

Subscripts g and d indicate that the species is gaseocus and dis-
sclved in the molten salt, respectively.
34

First, a series of experiments in which tritium was added to the NaF-NaBF, .
coolant salt in the Coolant-Salt Technology Facility {(CSTF)} showed that
>907% of the tritium added under steady-state conditions appeared in the
off-gas system in chemically combined (water-scluble) form and that 987
of the added tritium was removed through the off-gas system.10 These
data suggest that the fluoroborate coolant system of an MSBR (or DMSR)
might well diminish the leakage of tritium to the reactor steam system

to acceptable limits. Continued study11 shows that the oxide film formed
by the reaction of steam with steam—generator materials can greatly im-
pede the permeation of the metal by tritium. Even at a steam pressure

of 1 atm, where the oxidation rate is still clearly dependent on rates
of diffusion from the bulk alloy through the oxide coating, tritium per-

meation is impeded by factors of nearly 500 after 150 days of exposure.®

Status of Fuel and Coolant Chemistry

 

Fuel chemistry

Choice of components and composition. For an MSBR, where excellent

 

neutron economy is an abseclute requirement, acceptable fuel components
are few. The careful considerations and the detailed experimentation
over a pericd of many years that led to the choice of fuel constituents

?  There is no doubt

and composition have been completely described. '’
that (1) the major constituents of the fuel salt for an MSBR must be
LiF, BeF2, ThFy, and UF4, with a composition of ~71.7, 16, 12, and C.3,
respectively, and that (2) highly enriched "LiF and 2°°UF, are required.
For a DMSR the requirement for excellent neutron economy might seem
to be capable of slight relaxation. However, such a reactor must have
a reasonably high concentration of thorium and must contain more uranium
than the MSBR. There can be no reasonable doubt that the anion must be
F-, and the possibility that one can find better diluent fluorides than
+

"LiF and BeF; is extremely unlikely. The optimum compesition of the

 

This study was initiated under the Molten-Salt Reactor Program and
has been continued, because of its obvious interest, by the fusion energy
program.

T .
A slightly higher °Li concentration could possibly be tolerated,
but the rate of tritium generation would be increased.
 

 

35

initial fuel loading for a DMSR is, of course, not yet known precisely.
However, it appears likely that the optimum mixture will fall near (in
mole %) 72 LiF, 16 BeFy, and 12 heavy-metal fluorides, with slightly
less than 10.5 mole % ThFy and slightly more than 1.5 mole % UF, and UF;.
As a consequence, much that has been learned about the MSBR fuel compo-
sition is directly applicable to the DMSR. Knowledge of the behavior of
MSBR fuel, although not complete, can fairly be said to be extensive

and detailed.!??%»1?

As an additional consequence, the DMSR (like the MSBR) will require
a large-scale and reasonably economic source of lithium enriched to near
99,99% ’Li. No such enrichment facility is operating in the United States
today, but the technology is well known and relatively large-scale separa-

tion has been practiced in the past.

Fluoride phase behavior. Phase equilibria among the pertinent MSER

 

and DMSR fluorides have been studied in detail, and the equilibrium dia-
grams, although relatively complex, are well understood. Because these
reactors need a ThIFFy concentration much higher than that cof UF4, the
phase behavior of the fuel is dictated by that of the LiF-BeF,-ThFy

1,12

system™ *

tectic at 47 mole % LiF and 1.5 mole ¥ ThF., melting at 360°C.! The

shown in Fig. 3.1. This system shows a single ternary eu-

system is complicated by the fact that the compound 3LiF+ThF, can incor-
porate Re?+ ions in both interstitial and substitutional sites to form
solid solutions whose compositional extremes are represented by the
shaded triangular region near that compound. Inspection of the phase

diagraml’12

reveals that a considerable range of compositions with more
than 10 mole 7 ThFy will be completely molten at or below 500°C. The
maximum ThFs concentration available at this liquidus temperature is
just above 14 mcle 7. As expected from the general similarity of ThFy
and UFy, substitution of a moderate quantity of UFy for ThFy scarcely

changes the phase behavior.

Operation of the DMSR will result in production of plutenium and of
smaller quantities of other transuranium isotopes. It seems likely that
at equilibirum the concentration of plutonium will be near 0.05 mole %,

while Np, Am, Cm, Cf, and Bk might together total an additional 0.025
QORNL-LR-DWG 37420AR7 g
Thr’-; 1544

/ 1400

TEMPERATURE iN °C
COMPOSITION IN mole %

   
 
  
 

 
 
  
 
     

LiF - The, -

- LlF'ZTh%
3L|F-ThF4 85 -~

P 8G7 J{_ 950
- /

 
  

£ 568/
3L|F~Th§

£ 565 71N
f '3- \gi;-.’
i L JO\/ \\\\

 

 

550 S— £526

o i e e, F

gag 2LIF BeF; 500%s0 200t 400 450 500 555
£ 458 £ 360

Fig. 3.1. The system LiF-BelF,-ThF,.

mole %. All these species are expected to be dissolved in the fuel solu-
tion as trifluorides.

The golubility of PuF; in LiF-BeF;-ThF, (72-16-12 mcle %} has been
measured at CRNL'? and at the Bhabba Atomic Research Center in India.'®
The latter study indicated that solubility increased from 0.77 mocle % at
523°C to 2.79 mole % at 718°C. 'The ORNL measurements yielded values

about 207 higher. In beth studies, more than one method was used for

assay of the dissolved plutonium, and no ready explanation of the dis-

crepancy is available. It is clear, however, that even the lower value

far exceeds that required. The other transuranic species are known to :
dissolve!® in the LiF-BeF,-ThF, solvent, but no quantitative definition

of their sclubility behavior exists. Such a definition must of course

be obtained, but the generally close similarity in the behavior of the

 
 

 

37

trivalent actinides makes it most unlikely that solubility of these
individual species can be a problem.

The solubility of UF3; in the fuel was known to be well in excess of
that required for the MSBR, but its absolute magnitude is not well kncwn.

® with behavior in LisBeF,, the solubility of UFj3 is very

By analogy1
likely to be lower than that of PuF3, but it is quite unlikely to be less
than 0.4 mole 7 at 565°C.

The trivalent lanthanides and actinides are known to form solid
solutions so that, in effect, all the rare-earth trifluorides and the
actinide trifluorides act essentially as a single element. Should it
prove desirable tec operate with 107 of the uranium reduced (ca. 0.16
mole % UF3), it is possible, but highly unlikely, that the combination
of all trifluorides (perhaps 0.3 mole %) might exceed this combined
sclubility at a temperature somewhat below the reactor inlet temperature.
A few experiments® must be performed to check this slight possibility.

Oxide behaviecr. The behavior of molten fluoride systems such as
this is markedly affected by appreciable concentrations of oxide ien.

The solubilities of the actinide dioxides in LiF-BeF:;-ThF., are low
and are known':'%7%% to decrease in the order ThO,, Pa0jy, U0,, Pul,.
Solubility products and their temperature dependence have been measured

1516723 +hat these dioxides all

for these oxides. Moreover, it is known
have the same fluorite structure and form solid soluticons; theiy behavioer
is reasonably well understood.’ Plutonium as PuF; shows little tendency
to precipitate as oxide.'?”

The compound Pa0s (or an addition compound of this material) is
very insoluble in LiF-BeF;-ThFy (72-16-12 mole 7%). The oxide concentra-

tions at which Pa;0s5 can be precipitated depend on both the protactinium

concentration and the oxidation state of the fuel. The situation is
indicated by the equilibrium+

1 Pas05 + U3t = gttt 4 2 02 4 patt

2 () 2 *

 

%
Since mixtures with plutonium and beryllium are neutron scurces, the

experiments are more difficult than usual.

The subscript ¢ identifies the crystalline or solid state.
38
for which we estimate® the equilibrium quotient

log [(X5220X, ateXud) /X 54] = 0.76 — 8590/T (+0.8) .

5
G
The result is that with 100 ppm protactinium and 30 ppm oxide present,

the U’1/U"" ratio must be kept above about 107™° if inadvertent precipita-
tion of Pa205 is to be avoided. Such oxidizing conditions are easy to
avoid in practice. There is also a dependence on the udt/utt ratio of

the oxide concentration at which Pu0; precipitation cccurs. However,

even stronger oxidizing conditions (U3+/Uq+ < 10™%) are required to pre-
cipitate PuOp from fuel for MSBRs or DMSRs.

The solubility of the oxides of neptunium, americium, or curium has
not been examined. Some attenticn to this problem will be required, but
it is not obvious that such studies have a high priority.

It is clear that the DMSR fuel must be protected from oxide contami-
nation to avoid inadvertent precipitation. Because of the low oxide
tolerance, this will require some care, but the successful operation of
the MSRE over a 3-year period lends cenfidence that oxide contaminaticn
of the fuel system can be kept to adequately low levels. This confidence,
when added to the prospect that the DMSR fuel will be reprocessed {(and
its oxide level reduced by fluorinaticn of the uranium) on a continuous
basis, suggests very strongly that problems with oxide contamination can
be avoided.

Physical properties. Most of the physical properties of LiF-BeFs;-

 

ThFy, (72-16-12 mole %);'C are known with reasonable accuracy, although
several have been defined by interpolation from measurements on slightly
different compositions.

The liquidus temperature is well known, and density and viscosity

2,24

are accurate to 13 and +107, resgpectively. The heat capacity has

been derived from drop calorimetry;25 on the basis of this determination

and with a simple model for predicting heat capacity of molten fluorides,

one can reliably predict the heat capacity of the DMSR mixture.

 

% o .

Most physical properties will be trivially affected by variations
among the heavv-metal cencentrations so long as the heavy-metal content
remains at 12 mole %.

 
 

 

39

Thermal conductivity is the key property for predicting heat-transfer
coefficients of molten fluorides. Measurements that are probably accurate
to *10 to 15% have been obtained for LiF-BeF,-ThF,-UF, (67.5-20-12-0.5
mole %).%*® TFor future design considerations it will be helpful to de-
velop an apparatus to measure thermal conductivities of fluorides with
greater accuracy and to determine the conductivity of the fuel salt com-
position.

The surface physical properties (surface tension and interfacial
tension between salt and graphite) are only qualitatively known. Such
properties are important in assessing wetting behavior and in determining
the degree of salt penetration into graphite.

The wvapor pressure, as yet unmeasured, has been extrapolated from
measurements of LiF-BeF,; and LiF-UFy mixtures. At the highest normail
operating temperature, 704°C, the estimated vapor pressure is V1.3 Pa
(107% torr). The vapor composition has not been measured, but the vapor
would be considerably enriched in BeF; and perhaps in ThF,. Vapor pres-
sure and vapor composition are not high-priority measurements. However,
more than qualitative estimates of these properties will be required in
future calculations of the amount and composition of salt that is trans-
ported by gas streams used to cool portions of the off-gas system in the
primary circuit. A transpiration experiment would provide firm values
of vapor composition and improved values of vapor pressure. Manometric
measurements combined with mass-spectrographic determination would pro-

vide more precise information on both.

Fission-product chemistry. Much attention was given to behavior of

 

the fission products in the MSRE?’ because of their effect on reactor
operation and performance, afterheat, and reactor maintenance. More
experimentation will clearly be required in future DMSR (or MSBR) develop-
ment.

The noble gases are only slightly scluble in molten fluorides®?™ 3!

and can be removed by sparging with helium. More than 80% of the *35%%e
was removed by the relatively simple sparging system of the MSRE.® The
more efficient sparging proposed for the MSBR should also be applicable

to the DMSR. Most of the '*°Xe (the worst of the fission-product poisons)
40

g
135 T

is formed indirectly by decay of 6.7-hr , and the use of rapid side-

stream stripping of 1357 by the reaction

HF +~ 7, .. > F7 + HI

(g) (d) (d) (g) ’
was considered a remote possibiiity for the MSBR.? Such stripping seems
a most unlikely need for the DMSR; if it were necessary, it woeuld preclude
operation at high UF3;/UF, ratios.

The rare earths and other stable soluble fluorides (e.g., Zr, Ce,
Sr, Cs, Y, Ba, and Rb) are all expected to be found principally in the
fuel salt* and can be removed by the fuel processing operation.T The
chemical behavior of these fission products is fairly well understood
and, like the noble-gas behavior, can be predicted confidently for
operating DMSRs. *7

The chemical behavior of the so-called noble-metal fissien products
(Nb, Mo, Te, Ru, Ag, Sb, and Te)t is considerably less predictable — as

27 — and warrants further study.

has been berne out in MSRE operations
According to available thermodynamic data, they are expected to appear
in a reduced form at UF3/UF, ratios greater than 1077, However, in the :
reduced and presumed metallic state, these fission products can disperse
via many mechanisms.

Analyses of MSRE salt samples for five noble-metal nuclides (°7Mo,
1DgRu, IOERug and '?°713%Te) showed that the fuel salt contained up to
a few tens of percent of the nominal calculated inventory. All these
species have volatile high-valence fluorides that could form under suf-
ficiently oxidizing conditions. On the basis of thermodynamic comnsid-
erations and a correlation of their behavior with that of lllAg, for
which no stable fluoride exists under fuel-salt conditions, it has been
tentatively concluded that they are metallic species that occur as finely

divided particles suspended in the salt.

 

P

Some of these have noble~gas precursors; a fraction of these will
escape from the fuel and appear in the off-gas system.

'See Chapter 6.

And several other species of lower yield.

 
 

 

41

The noble-metal fission products were also found deposited on graphite
and Hastelloy N specimens (on surveillance specimens as well as on post-
operation specimens). However, their distribution on both sets of speci-
mens varied widely and allowed only very tenuocus conclusions to be drawn.
It was evident from these studies that net deposition was generally more
intense on metal than on graphite, and deposition on the metal was more
intense under turbulent flow.

235y operation of the

Gas samples taken from the pump bowl during
MSRE indicated that substantial concentrations of noble metals were pres-
ent in the gas phase, but improved sampling techniques (used during

h 233UF4) showed that previous samples had

operation of the reactor wit
been contaminated by salt mist and that only a small fraction of the

noble-metal fission products escape to the cover gas.

0Of the noble metals, nicbium is the most susceptible to oxidation:
it was found appreciably in salt samples at the start of the 233y MSRE

operation because of the low initial U3+/U“+

ratio. It could be made
to disappear by lowering the redox potential of the fuel,1 but it sub-
sequently reappeared in the salt several times for reasons that were not
always explainable. The %°Nb data did not correlate with the Mo-Ru-Te
data mentioned previously, nor was there any observable correlation of

. g . . . 1,27
niobium behavior with amounts found in gas samples.

The actual state of these noble-metal fission products is important
to the effectiveness of MSBR operations. 1If the products exist as metals
and if they plate out on the Hastelloy-N portions of the reactor, they
will be of little consequence as poisons; however, they can be of im-
portance in determining the level of fission-product afterheat after
reactor shutdown and will complicate maintenance operations and post-—
operation decontamination. They will contribute to neutron poisoning
if they form carbides or adhere in some other way to the graphite modera-
tor;* however, examination of the MSRE graphite moderator indicated that

the extent of such adherence was limited.!s?s?7

 

Niobium is the only element of this series with a carbide that is
thermodynamically stable in this temperature range. It showed the largest
tendency to associate with the moderator graphite.
Operation with a UF3/UFy, ratio near 0.1 will apparently change the
behavior of those noble metals capable of reduction to an anionic state.

57 and may be safely presumed

This seems certain to include tellurium
to include selenium. Antimony may exist (and may be dissolved) as Sb’~
in such melts, and other of the noble-metal fission products may be dis-
solved. If so, they, along with the tellurium and selenium, would — as
was not expected for the MSBR — be transported to the fuel processing
circuit, where their removal should be possible. Should a decision be

made to operate with strongly reduced fuel, some study of such possi-

bilities will be necessary.

Clearly, most of the future fiésionwproduct chemical research should
be directed toward increasing cur understanding of noble-metal—fission-
product behavior to a level comparable to that of the other fission prod-
ucts., Factors of importance to future reactors include the redox poten-
tial of the system,* the possible agglomeration of metals onto gas and
bubble interfaces in the absence of colloidal (metallic, graphite, oxide,
etc.) particles, the deposition of noble metals onto colloidal particles,

and the deposition and resuspension of particles bearing noble metals.

Tritium behavior. A 1000-MW(e) MSBR has been estimated®® to produce

 

about 2420 Ci (v0.25 g) of °H per day; the DMSR must accordingly be ex-
pected to generate H at this rate. Since metals at high temperatures

are permeable to the isctopes of hydrogen, the pathways for tritium flow
from the reactor to the environment are numerous. Although many of the
pathways do not present serious difficulty,+ the flow to the steam genera-
tor, if not inhibited, could result in tritium contamination of the steam
system and release of tritium to the environment via blowdown and leakage

to the condenser ceolant.

As noted above (and described in more detail in a subsequent sec-
tion), the NaF-NaBF, coclant appears to be the major defense against *H

escape to the steam system;’’ however, the behavior of °H in the fuel

 

ale

See further discussion under Fuel-Graphite Interactions.

T . . . .
Since the tritium can readily be trapped and retained for disposal
as waste.

el

 
 

 

32 is generated

system is obvicusly important. Most of the tritium'?>
by neutron reactions on 574 and ‘Li in the fuel. Such °H is, in princi-
ple, generated in an oxidized state (as 3HF). However, upon equilibra-

tion with a fuel containing a UF3/UFy ratio of 0.01 by the reaction
HF + UF g-é-H + UF
(@ 2 " (g) (a)

the HF should be almost completely reduced to Hz; reduction would of
course be even more complete at UF3/UFy = 0.1. The data for this equi-

. \ \ 8
librium reaction are well known. s 9

?% to determine the solubility of Hz and Dj

Attempts have been made
(and, by analogy, °H») in molten LirBeFs. Plausible (and very small)
solubilities were measured, but the solubility is not precisely known.
Further study of the solubility relationships is needed, and measurements
of diffusivity in the fuel would alsoc be valuable. It seems likely that
an efficient sparging system (as for '*°Xe removal)} will strip a consid-
erable fraction of the °Hz to the reactor off-gas system, where it could

be coliected for disposal.

Basic studies of molten fluorides. A comprehensive knowledge of

 

the formation free energies (AGf) of solutes in molten Li»;BeF., has been
gained over the year533 from measurements of heterogeneous equilibria
involving various gases (e.g., BF or H»0) and solids (e.g., metals or
oxides). The list of dissolved components for which formation free
energies have been estimated includes LiF, Be¥,, Th¥,, several rare-
earth trifluorides, Zr¥,, UFs3, UFy4, PaFy, PaFs, PuF;, CrF,, FeF,, NiFy,
NbFy, NbFs, MoF3, HF, BeO, BeS, Be(OH),, and Bel,. Some of these AGf
values, however, are presently insufficiently accurate for the needs of
the MSBR and DMSR programs (e.g., those for PaFs, PaFs, ThF,, MoFj, NbFy,
and NbFs) and additional equilibrium measurements involving these solutes
are needed. Moreover, there is a need for the AGf values of certain other
fission-product compounds such as the lower fluorides of technetium and
ruthenium and various dissclved compounds of tellurium. A more urgent
need is an increased knowledge of how activity coefficients (which have
been defined as unity in Li»BeFy) vary as the melt composition changes.

Such knowledge is required to predict how the numerous chemical equilibrium
bty

constants that may be derived from AGf values in LioBeFy will change as

the melt composition is changed to that of a DMSR fuel.

Coolant chemistry

 

It has never appeared feasible to raise steam directly from the
fuel (primary) heat exchanger; accordingly, a secondary ccolant must be
provided to link the fuel circuit to the steam generator. The demands
imposed upon this coolant fluid differ in obvicus ways from those imposed
upon the fuel system. Radiation intensities will be markedly less in the
coolant system, and the consequences of uranium fission will be absent.
However, the coolant salt must be compatible with the construction metals
that are also compatible with the fuel and the steam; it must not undergo
violent reactions with fuel or steam should leaks develop in either cir-
cuit. The coolant should be inexpensive, it must possess good heat trans-
fer properties, and it must melt at temperatures suitable for steam cycle
start-up. An ideal coclant would consist of compounds that are tolerable
in the fuel or are easy to separate from the valuable fuel mixture should
the fluids mix as a consequence of a leak.

Choice of coolant compesition. Many types of coolant materials

 

were carefully considered before the choice was made. The coclant which
served admirably in the MSRE, "Li,BeF,, was rejected for the MSBR be-
cause of economics and because its liquidus temperature is higher than
desirable. No substitute with ideal characteristics was found. After
consideration of molten metals and molten chloride and molten flucoride
mixtures, the best material overall appeared to be a mixture of sodium
fluoride and sodium fluoroborate.’ These compounds are readily avail-
able and inexpensive and appear to be sufficiently stable in the radia-
tion field within the primary heat exchanger. The mixture of NaF-NaBFy
with 8 mole % NaF melts at the acceptably low temperature of 385°C (725°F),
and its physical properties seem adequate for its service as a heat trans-
fer agent. These compounds are not ideally compatible with either steam
or the MSBR fuel, but the reacticns are neither violent nor even particu-
larly energetic.

The fact that fluoroborates show an appreciable equilibrium pressure

of gaseous BF3 at elevated temperatures presents minor difficulties. The

 
 

 

BF3 pressures are moderate; they may be calculated from

log P = 11.149 — 5920/T ,

when pressure is in pascals and temperature is in kelvin [yielding 23
kPa (175 torr) at 600°C], and clearly present no dangerous situations.
However, it is necessary to maintain the appropriate partial pressures
cf BF3 in any flowing cover—-gas stream to avoid composition changes in
the melt.

The appropriateness of that choice for the MSBR (and for the DMSR)
has been confirmed by several findings in recent years. First, a care-
ful and detailed reconsideration of secondary {(and even secondary plus
tertiary) coolants®” ranked the NaF-NaBF, coolant very high on the list
of alternatives. After these deliberations,’® experimental information'®
became available to show that the NaF-NaBF; mixture was genuinely effec-
tive in trapping H. The reconsideration, accordingly, concluded: "

While the information that is currently available is in-
adequate for accurate extrapolation to the rate of tritium re-
lease teo the steam system of an MSBR, it appears that the

sodium fluorcborate salt mixture would have a substantial in-

hibiting effect on such release and that environmentally ac-

ceptable rates (<10 Ci/d) could be achieved with reasonable
effort.

Additional study needed.®* A considerable study of many aspects of
fiuvoroborate chemistry has been conducted during the past few years.
Neverthelesg, our understanding of the chemistry of the NaF-NaBF., system
is less complete, and our knowledge of its behavior rests on a less se-
cure foundation than that of the MSBR fuel system. Thus there are sev-
eral areas where further or additional work is needed, although it seems
unlikely that the findings will threaten the feasibility of NaF-NaBFy
in the MSBR (or DMSR) concept.

 

"It should be obvious that the additional study of NaF-NaBF, cool-
ants needed for an MSBR is essentially identical to that needed for a
DMSR. Accordingly, the previous documentation,'’? except as modified by
more recent findings discussed later, adequately describes the neaded R&D
pregram.
46

Phase behavior of the simple NaF-NaBF, system and the equilibrium wass’
pressure of BF3 over the pertinent temperature interval are well under-
stood. If the NaF-NaBF, eutectic (or some near variant of it) is the
final coolant choice, little effort need be spent in these areas.

Additional information is needed, however, on the behavior of oxide
and hydroxide ions in the fluoroborate melts. For example, the solubility
of NasBoFg0 in the mixture is not well known; data on equilibria (in
inert containers) among H,0, HF, NaBF30H, and Na:B»Fe0 are still needed;
and rates of reaction of dilute NaBF30H solutions with metals need defi-
nition. Investigation of NaBFy melts by x-ray powder diffraction, infra-
red spectroscopy, and Raman spectroscopy have identified the stable ring
compound Na3zB3Fg03; as the probable oxygen-containing species in ccolant

2
meits. "

Measurements of condensates trapped from the coolant salt tech-
nology facility (CSTF), a development loop, show a tritium concentration
of 10° relative teo the salt, suggesting that a volatile species may be
selectively transporting tritium from the loop through the vapor.lO Re-
cent results indicate that BF3°*2H,0 may exist as a molecular compound in
the vapor and could be regsponsible for the tritium trapping,3£+ However,
the mechanism by which tritium diffusing from the fuel system can be
trapped needs additional study.

As indicated above, several of the physical property values have
been estimated. These estimates are almost certainly adequate for the
present, but the program needs to provide for measurement of these quan-
tities.

Compatibility of the NaF-NaBF, with Hastelloy N under normal opera-
ting conditions seems assured. Additional study, in realistic flowing
systems, of the corrosive effects of steam inleakage is necessary. This
study, closely allied with the study of equilibria and the kinetics of
reactions involving the hydroxides and oxides described above, would re-
quire the long-term operation of a demonstration loop that could simulate
steam inleakage and coolant repurification.

Purification procedures for the coolant mixture are adequate for ’
the present and can be used to provide material for the many necessary
tests. However, they are not adequate for ultimate on-line processing

of the coolant mixture during operation. TFluorination of the coolant

 
 

 

47

on a reasonable cycle time would almost certainly suffice but it has not
been demonstrated. A process using a less aggressive reagent is clearly
desirable.’

A fluoroborate mixture has shown completely adequate radiation sta-
bility in a single (but realistically severe) test run.  Additional
radiation testing cf this material in a flowing system would seem de-
sirable and should ultimately be done but does not rate a high priority

1,2
at present. ?

Fuel-coolant interactions

 

A rupture of a tube (or tubes) in the primary heat exchanger would
unavoidably lead to mixing of some coeolant salt with the fuel. The pos-
sibility of a nuclear incident would seem highly unlikely because of the
consequent addition of the efficient nuclear poison boron to the fuel.
However, since BFj3; is volatile, mixing might result in a pressure surge,
and the NaF-NaBF, mixture contains some oxygenated species. The 1972
review, accordingly, concluded:?®

Mixing of coolant and fuel clearly requires additional

study. The situation which results from equilibration of these

fluids 1is reascnably well understood, and, even where large

leakages of coolant into the fuel are assumed, the ultimate

"equilibrium' seems to pose no real danger. However, the real

situation may well not approximate an equilibrium condition.

Studies of such mixing under realistic conditions in flowing

systems are lacking and necessary.

The 1974 program plan2 included a very considerable program for such
study.

More recent experiments’®" have thrown additional light upen such
mixing. Although additional experiments are needed, it now appears
likely that such mixing would not pose drastic problems. These experi-
ments revealed that BF; gas was slowly evolved when the salts were mixed;
some 30 min were required to complete the BF3 evolution. Furthermore,
the ThF4 and UFy4 showed no tendency to redistribute, to form more con-
centrated solutions, or to precipitate. Moreover, no UG, precipitated
even when the molten fuel-coolant combination was agitated for several

hours while exposed to air.
48

Additional confirmatory experiments will be required and the dif-
ferent species in DMSR fuel must be tested. Such experiments may re-

ceive less attention and lower priority than previously helieved.!s?

Fuel-graphite interactions

 

Graphite does not react with, and is not wetted by, molten fluoride
mixtures of the type to be used in the MSBR. Available thermodynamic

35

data suggest that the most likely reaction,

4UFy + C

@ T Cey T CFe

+ 4UF 4

(g) (d) ~’

should come to equilibrium at CF, pressures below 107% atm. At least
one source°® lists chromium carbide (Crs3Cs) as stable at MSBR tempera-
tures. If Crs:Cy is stable, it should be possible to transfer chromium
from the bulk alley to the graphite. No evidence of such behavior has
been observed with Hastelloy N in the MSRE or other experimental assem-
blies. Although such migration may be possible with alloys of higher
chromium content, it should not prove greatly deleterious, since its
rate would be controlled by the rate at which chromium could diffuse .
to the alloy surface and should be limited by a film of Cri3C; formed on
the graphite. This consideration, taken with the wealth of favorable
experience, suggests that no problems are likely from this source in
the reference MSBR or in a DMSR.

However, some additional examination of this unlikely problem area
must be done for the DMSR, particularly if operation at UF3;/UFy ratios
near 0.1 is to be attempted. The upper limit on that ratio will most

likely be set by the equilibrium

AUFg(d) + 2C - 3UF, + UCy

(c) (d) (c)

at the lower end of the operating temperature range.37

Toth and Gilpatrick,38 who used a spectrometric technique in which
molten salts were contained in cells of graphite with diamond windows,
made a careful study of equilibria among UFs: and UF4 in molten solution

with solid graphite and uranium carbides. Their data show that the ratio

 
 

 

49

of UF3 to UFy in LiF-BeF;-ThFy (72-16-12 meole %) in equilibrium with
graphite and UC, at 565°C lies in the range 0.11 to 0.16.% All equilibria
.f.

studied were found to be very sensitive to temperature' and to the free
fluoride concentration of the solvent. It would seem likely that UF;/UF,
ratios as high as 0.1 can be tolerated for a DMSR (though slight adjust-
ments in fuel composition or fuel-inlet temperature might be required),
but confirmatory experiments are needed. Since similar systems appear

to be sensitive to oxide ion concentration, some experimental study of
this parameter will also be required.

Even at relatively high temperatures, graphite has been shown to

9

adsorb H» and its isotopes to an appreciable extent. - Further informa-

tion about this phenomenon should be cbtained.

Prime R&D Needs

 

Weaknesses in the existing technological base and requirements for
additional technical information have been identified throughout the
preceding discussion of the technology status. These needs are consoli-
dated and presented below, in cutline form, to provide a concise tabula-

tion for defining and scheduling possible R&D activities.

Fuel chemistry
1. Demonstrate the feasibility of operation at UF3/UF, ratio of 0.07
to .1.
a. Verify the individual and collective solubilities of trivalent
actinide fluorides.
b. Verify the interaction of uranium with graphite.
c. Verify the behavior of noble and seminoble fission products
(i.e., Se, Sb, Te, and Ru) along with tellurium.
2. Define the limits on tolerable oxide concentration to avoid precipi-
tation of oxides from DMSR fuel at UF3/UF,4 ratio of 0.1 and to avoid

interaction of that fuel with graphite.

 

% . - . s . s .
The UC2 so formed may be stabilized by inclusion of some oxide ion
in the lattice.

+At 600°C the UF3/UFy ratio lies in the range 0.23 to 0.32.
50

3. Provide sound measurements of those physical properties (surface ten- g
sion, interfacial tension, thermal conductivity, vapor pressure, and
vapor composition) that are not known with precision.
4. Improve the knowledge of fission~product behavior, particularly of
key noble metals, in reduced DMSR fuel.
5. Determine the solubility and diffusion kinetics of H» and its iso-
topes in DMSR fuel.
6. Perform the basic studies to bring knowledge of solute behavior in
LiF-BeF,-ThF, (70-16-12 mole %) to at least the level of current

knowledge of behavior in LisBeFy.

Coolant chemistry

 

1. Identify and characterize oxygenated and protonated species in
NaF-NaBF, as functions of "contamination"™ level.

2., IBlucidate the mechanisms by which the coclant salt takes up B and
determine identity of the products and tfie key reaction rates.

3. Determine {(in cocperation with the materials development effort de-
scribed in Part IV of this document) the effect of the "contaminants"
mentioned above, and the effect of steam inleakage, on corrosivity
of coolant salt.

4. Refine the measurements of physical properties of NaF-NaBF, as re-
quired.

5. Confirm the adequacy of radiation resistance of the realistic mixture

(i.e., with the desired "contaminant’ level).

IF'uel—-coolant interactions

 

1. Continue, and scale up, mixing studies to demonstrate that no hazard-
ous interactions exist and define the limits of behavior.
2. Consider the problems of fuel (and graphite) cleanup consequent to

a fuel-coolant leak.

Fuel-graphite interactions

 

1. The major need identified above is to demonstrate tolerable UF3/UF,
ratios and 0%~ concentration limits to avoid formation of uranium

carbides.

 
2. Define the extent to which °H will be adsorbed by moderator graphite.
3. Verify the interaction of noble-metal fission-products with graphite

at usable UF3/UFy ratios.

Estimates of Scheduling and Costs

Preliminary estimates of the necessary schedule and of its operating
and capital funding requirements are presented below for the fuel and
coolant chemistry program described above. As elsewhere in this document,
it has been assumed that (1) the program would begin at the start of FY
1980, (2) it would lead to an operating DMSR in 1995, and (3) the R&D
program will produce no great surprises and no major changes in program
direction will be required.

The schedule, along with the dates on which key developments must
be finished and major decisions;made, is shown in Table 3.1. It seems
virtually certain that the R&D programs (including those described else-
where in this document) will provide some minor surprises and that some
changes in the chemistry program will be required. No specific provi-
sions for this are included; but, unless major revisions become necessary
in the middle eighties, it appears likely that suitable fuel and cooclant
compositions could be confidently recommended on this schedule.

The operating funds (Table 3.2) and the capital equipment require-
ments (Table 3.3) are shown on a year-by-year basis in thousands of 1978
dollars. No allowance for contingencies, for major program changes, and

for inflation during the interval have been provided.
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 3.1. Schedule for chemical research and development
Fiscal year
Task
1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995
Fuel chemistry
i 2 3 4 5
Phase equilibria v v v v v
6 7 8 g
Oxide behavior v v v v
fo it
Physical properties v v
i2 13
Tritium behavior v v
4 15 16 17 18 ig 20
Fission-product chemistry vv v v Vv v v
24 22 23
Basic studies v v v
C t i
cclant chemistry V24 v25 (726 \727 vaa
Basic studies
st
Physical properties
30 3 32
Tritium chemistry v v v
33 34 35
Fuel-coolant interactions v v v
36 37 5 38
Fuel-graphite interacticns v v v v
Milestones:
1. Determine solubility of UF; over reasonable fuel compositiom 20. Make final evaluation of fission-product behavior for DMSR,
range. 21. Define activity coefficients for Te?t fand other Te species)
Z. Determine solubility of AmF3;, NpFi:, and CmFj;. in reduced fuel.
3. Define solubility limit of total metal trifluvorides over rea- 22. Complete evaluation of porous electrode studies.
sonable range of compositions. 23. Complete definition of activity coefficients for sclutes in
4, Conclude phase equilibrium investigations, including effect fuel.
of small concentrations of C17. 24. Complete evaluation of boride formation on Hastelley N in
5. Make final decision as to feasibility of operation at UF3/UFy coolant salt.
ratio near 0.1. 25. Finish investigation of oxide species in coolants.
6. Redetermine solubility of Paz0s. 26. Finish measurement of free energy and activity coeff1c1ents
7. Establish selubilities of Am;03, Np20i3, and Cm»0;. of corrosion products in coolant salt.
8. Determine scliid solution behavior as a function of oxide con- 27. Make final decision as to coolant composition.
tamination level. 28. Finish measurements of effect of steam inlezkage into coolant.
%, Set limits on tolerable oxide ion concentration in fuel and 29. Finish physical property measurements on ceoolant.
assess possible separations procedures based on oxide pre- 30. Identify mechanisms for trapping of tritium in fluworcoborate.
cipitation. 31. Complete evaluation of reaction rates of tritium with fluoro~ !
1¢. Determine surface physical properties of realistic composition borate species. *
range and assess wettability of metal and graphite. 32, Complete evaluation of tritium removal from coolant and re-
11. Complete physical property determinations. conditioning of coolant.
12, Determine solubility of H; and HT in fuel. 33. Complete dynamic studies of fuel-coolant mixing.
13. Determine diffusivity of H, and HT in fuel. 34. Determine precipitation behavior of fuel with realistic oxide
14. Determine possibility of removal of ZrF, from reduced fuel as and protonic contaminants.
intermetallic compound. 35. Complete evaluation of methods for recovery from fuel-coclant
15, Establish sclubility of Te, Tezm, and other Te species in fuel. mixing.
16. Determine oxidation states of noble and semiroble fission- 36. Define limits om UF3/UF, ratio and oxide contamination level
product metals as a function of pertinent UF;/UF, ratios. in fuyel to avoid uranium carbide formation.
17. Make final conclusions as to Se, Te, I~, and Br~ behavior at 37. Complete evaluation of fission-product—graphite interact1on
. pertinent UF;/UF, ratios. with maximally reduced fuel.
18. Establish feasibility of removal of noble metals by washing 38. Complete investigation of tritium uptzke by moderator graphite
with Bi (with no reductant). and activated carbons.
19, Establisgh extent of serption of I;, SeFg, and TeFg in oxidized

fuel {uranium valence 4.5).

o,
 

Table 3.2.

 

Task

 

Fuel chemistry

 

 

 

 

 

 

Operating fund requirements for chemical research and development

 

Cost {thousands of 1978 dollars) for fiscal year —

 

 

 

 

 

1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1593 1994

Phase equilibria 100 1G0 100 160 1060 100 75 >0 25 0 0 0 0 0 0 0
Oxide behavior 150 150 150 150 175 150 100 50 50 0 0 0 0 0 o O
Physical properties g 0 60 60 &0 60 120 120 120 60 00 40 40 0 0 0
Tritium behavior 0 120 120 150 150 100 100 50 50 o 0 0 0 0 0 0
Fissicn-product chemistry 80 120 120 120 120 120 200 250 250 250 150 150 150 100 50 0
Basic studies 120 120 150 150 150 150 100 100 75 0 0 0 ¢ 0 0 0

Coolant chemistry
Basic studies 50 100 100 125 125 150 150 150 150 150 100 100 75 50 0 G
Physical properties 0 0 G 50 50 50 75 75 75 75 75 75 50 50 0 0
Tritium chemistry 70 125 150 150 215 230 260 330 330 300 150 100 100 50 0 0
Fuel-coolant interactions 75 75 75 75 100 100 106G 100 50 25 0 Q 0 0 0 0
Fuel~-graphite interactions 50 80 100 100 100 130 150 200 125 75 25 0 o 0 0 0
Total funds® 695 990 1125 1230 1345 1360 1430 1475 1300 935 560 465 465 250 50 0

Total funds through 1994: $13,675.

€S
Table 3.3. Capital equipment fund requirements for chemical research and development

 

 

 

Cost (thousands of 1978 dollars) for fiscal year —

 

 

 

 

Task ——— e et
1980 1981 1982 1983 1984 1985 1686 1687 1988 1989 1990 1991 1992 1993 1994
Fuel chemistry 40 90 145 130 44 160 100 100 95 20 10 10 D 0
Coclant chemistry 25 85 150 100 95 165 163 210 85 35 10 0 0 0
Fuel-coolant interactions 30 10 15 40 30 50 60 15 0 0 0 ¢ 0
Fuel~graphite interactions o 20 45 40 15 30 22 0 5 0 0 0 0 0 0
Total funds” 95 205 355 310 0

 

“Total funds through 1991: $2500.

 

 

 

 

 

180 410 325 350 185 a5 20 10 0

 

 

1895
0
0
wn
o~
0

 
 

 

55

4. ANALYTICAL CHEMISTRY

Scope and Nature of the Task

 

With conventional solid-fueled reactors, there is neither the op-
portunity nor an cbvious need for chemical analyses of the fuel during
its sojourn in the reactor. On the other hand, for a fluid-fueled reac-
tor, particularly cne that includes a fuel processing plant, there is a
pressing need to know the precise composition (particularly the concen-
trations of fissile materials) in several process streams. Moreover,
such information is needed at frequent intervals (if not continuously)
and on a real-time basis. As a consequence, such reactors are much more
dependent on analytical chemistry than are, for example, LWRs.

The MSRE was indeed coperated successfully with chemical analyses
performed on discrete samples of fuel removed from the reactor for hot-
cell study. However, it was recognized1 early that the MSBR and its as-
sociated reprocessing plant would require in-line analyses. The DMSR
would be equally dependent on successful development of such analytical
techniques.

These requirements are several in number. It will be necessary to
determine on a virtually continuous basis the redox potential, the con-
centrations of uranium, protactinium, and other fissionable materials
and of bismuth and specific corrosion products (notably chromium) in the
stream entering the reactor from the processing plant. Such information
must also be available for the fuel within the reactor circuit and in the
stream to the reprocessing plant. In addition, it would be highly de-
sirable to know the oxide ion concentration in the fuel within the reactor
and in the stream from the processing plant and tc know the states and
concentrations of selected fission products in the fuel within the reactor.
It will be essential to know the concentrations of uranium and fissile
isotopes in the processing streams from which they could be lost (tc the
waste system) from the complex. These streams will include the small
stream of released off-gas, the LiCl system®* for rare-earth transfer,

the ZrFy removal system® and perhaps a system® for removal of metallic

 

ol

“These systems are described in Chapter 6.
56

noble-metal fission products. 1In addition, it will be necessary to g
monitor corrosion products, oxygenated compounds, and protonated (and

tritiated) products in the coolant system and in the off-gas system as

well as in the system for hold-up and recovery of tritium. Accordingly,

on~line analyses will be required in at least three kinds of molten

salts, in gases, and perhaps in molten bismuth alloys® within the

processing plant.

Key Differences in Reactor Concepts%

 

Insofar as the analytical chemistry requirements (and the R&D needs)
are concerned, the DMSR and the MSBR are quite similar. The DMSR will
require determination of plutonium {(and to some extent of Am, Cm, and
Np) to a degree markedly different frem the MSBR. Concentrations of
CF3, UFy, and PaF, in the fuel stream will be higher in a DMSR (as will
that of ZrF.) than in an MSBR. Complexities of the fuel processing plants
for the two reactors are very similar, and, except for the presence of
transuranics, so are the analytical requirements.

In principle, the emphasis on proliferation resistance would seem
to place additional demands on surveillance and precise determination of
plutonium, protactinium, and uranium within the DMSR system. In fact,
the requirements already imposed by the demands for safe and reliable
continuous coperation of the complex are almost certainly at least as

stringent,

Pogt-1974 Technology Advances

 

Several advances, a few amounting to breakthroughs, were made in
the 1971-1974 interval.'»>® The post-1974 studies in analytical chemistry
consisted mainly of (particularly valuable) service functions and, ex-

cept for studies of tellurium behavior, contained little exploratory

 

¥
It is clearly desirable, and may be necessary, to have on-line de-
terminations o¢f lithium and of some other metallic species in bismuth

streams.

See Chapter 3, Fuel and Cecolant Chemistry, for discussion of key
chemical differences.

 
 

 

57

development cof the kind ultimately needed. Primary accomplishments in
the post-1974 period were therefore relatively few. They did include
the following:

1. On-line voltammetric techniques for determination of the UF3/UF,

ratio were refined and were applied successfully and routinely in many
corrosion test 100?8#0,@1 and engineering eX]_Je:r:'Lments.6’7’“0”“2 Such
techniques can now be said to be well established in the absence of
radiation {which should prove of little consequence) and of fission prod-
ucts.

2. Measurements of protonated (and tritiated) species in the NaF-
NaBFy coolant salt were applied successfully in engineering test equip-
ment, ' %s%?

3. Voltammetric and chronopotentiometric techniques have been suc-
cessfully applied to measurements of Fe?t in LiF-BeF,-ThF, (72-16-12

> using anodic voltammetry show prom-

moie %),"" and preliminary studies®
ise for in-line monitering of oxide level in this molten salt, at least

under favorable conditions.

Status of Analytical Development

 

MSRE operation was mainly conducted with analyses performed on dis-
crete samples removed from the reactor. The reactor off-gas was analyzed
by in-line methods, and remote gamma spectroscopy was used to study fis-
sion products; all other determinations were made using hot-cell techniques

on batch samples. Of course, a major program of R&D preceded that opera-
1,2

tion. ?

Substantial experience had been gained in the handling and analysis
of nonradioactive fluoride salts prior to the MSRE program. Ionic or
instrumental methods had been developed for most metallic constituents.
For MSRE application it was necessary to develop additional techniques
and to adapt all the methcods to hot-cell operations. A nonselective
measurement of "reducing power" of adequate sensitivity had been developed

L7 . \ .
in the radiochemical

(hydrogen evolution method).*® A general expertise
separation and measurement of fission products was available from earlier
reactor programs at ORNL, and useful experience with in-line gas analysis,

particularly process chromatography,l+8 was available from other programs.
58

During the operation of the MSRE and in the subsequent technology
program, development of methods for discrete samples was continued, and
the Laboratory has acquired instrumentation for newer analytical tech-
niquea—:.h"9 Instrumental methods expected to contribute to the program
include x-ray absorptiocon, diffraction, and fluorescence techniques;
nuclear magnetic resonance (NMR); spark source mass spectrometry; elec-— .
tron spectroscopy for chemical analysis (ESCA) and Auger spectrometry;
electron microprobe measurements; scanning electron microscopy; Raman
spectrometry; Fourier transform spectrometry; neutron activation analy-
sis; delayed neutron methods; photon activation analysis; and scanning

with high-energy particles, e.g., protons.

Key developments for MSRE

 

Homogenized and free-flowing powdered samples of radioactive fuels
taken from the MSRE were routinely produced in the hot cell within 2 hr
of receipt. Salt samples were taken in small copper ladles that were
sealed under helium in a transport container in the sampler—enricher50
for delivery to the hot cell. Atmospheric exposure was sufficient to
compromise the determination of oxide and Ut but did not affect other
measurements. lechniques for taking and handling of such samples (for
those analyses for which they will suffice) are well demonstrated.

Oxide concentration could not be reliably determined on the pul-
verized salt samples because of unavoidable atmospheric contamination.
Instead, 50-g samples of salt were treated with anhydrous HF gas and
the evolved water was collected and determined.”' Oxide concentrations
of about 50 ppm were determined with better than *10 ppm precisiocn.*

Uranium analyses by coulometric titration showed good reproduc-
ibility and precision (0.57), but on-line reactivity balance data estab-
lished changes in uranium concentration within the circuit with about
ten times that sensitivity.l Fluorination of the uranium from 50-g sam-

52 .
and was used to separate uranium for

ples was shown to be quantitative
precise isotopic determination. If necessary, it could also have served
as the basis for a more accurate uranium analysis.

The rate of production of HF upon sparging of the fuel with Hy; is

a function of the UF3/UFy ratio. This transpiration method, modified

 
 

 

59

to allow for other ions in the fue1,53 gave values in reasonable agree-
ment with "book' values during operation with 2°°U (0.9 mole % U). The
method proved inadequate at the lower concentrations during operation of
the MSRE with 2%3U. Attempts tec determine UF3/UF, ratios by a voltam-—
metric method using remelted salt samples was not generally successfulls>®
because of prior UF: oxidation via atmospheric contamination. However,

it was possible to follow UF3 generation via H, sparging upon such sam-
ples.1 The radiation level of the samples does not appear to affect the
method.

A facility for spectrophotometry of highly irradiated fuel samples
from the MSRE was designed and constructed.”" The system design included
devices for remelting large salt samples under inert atmosphere and dis-
pensing portions to spectrophotometric cells. The entire system could
not be completed in time to give much useful data for MSRE.* It has
since been used to observe spectra of transuranium elements and of pre-
tactinium in molten salts. Feasibility of the general technique appears
to be established.

Equipment was installed at the MSRE to perform limited in-line
analyses of the reactor off-gases, using a thermal conductivity cell as
a transducer. An oxidation and absorption train®? permitted measurement
of total impurities and hydrocarbens in the off-gas. The sampling staticn
also included a system for the cryogenic ccllectien of xXenon and krypton
on molecular sieves to provide concentrated samples for the precise de-
termination of the isotopic ratios of krypton and xenon by mass spec-
trometry. During the last two runs of the MSRE, equipment was installed’”®
at the reactor to convert the tritium in various gas streams to water
for measurement by scintillation counting.

By means of a precise collimation system mounted on a maintenance
shield, radiation from deposited fission products on components was di-
rected to a high-resolution, lithium-drifted, germanium diode.”® From
the gamma spectra obtained, specific isotopes such as noble-metal fission

products were identified and their distribution was mapped by moving

 

®
Observations with a somewhat makeshift sampling system showed no
adverse effects from radiocactivity of the fuel.
60

the collimating system. During the latter runs of the reactor, such s

. . 57
measurements were made during power operation.

Analytical development for MSBR#*

 

At present, it appears that the measurement of the concentration of
major fuel constituents such as lithium, beryllium, thorium, and fluo-
ride ion by in-line methods may not be practical in an MSBR. Fortunately,
continuous monitering of these constituents will not be critical to the
operation of a reactor. The more critical determinations, which were
briefly described above, are generally amenable to in-line measurement.

The ultimate need for an MSBR is an analytical system that includes
all needed in-line analytical measurements that are feasible, backed up
by adequate hot-cell and analytical laboratories. In the interim, ca-
pabilities must be developed and analytical support provided for the tech-
nology development activities in the program.

Electrochemical studies. For the analysis of molten-salt streams,

 

electroanalytical techniques such as voltammetry and potentiometry ap-
pear to offer the most convenient transducers for remote in-line measure-
ments. Voltammetry is based on the principle that when an inert elec-
trode is inserted into a molten salt and subjected to a changing voltage
relative to the salt potential, negligible current flows until a criti-
cal potential is reached at which one or more of the ions undergo an
electrochemical reduction or oxidation. The potential at which this
reaction takes place is characteristic of the particular ion or ions.
If the potential is varied linearly with time, the resulting current-
voltage curve follows a predictable pattern in which the current reaches
a diffusion-limited maximum value that is directly proportionai to the
concentration of the electrocactive ion or ioms.

Basic voltammetric studies have been made on corrosion-product ions
in the MSRE fuel solvent LiF-BeF,;-ThF,>%7%2 and in the proposed coolant
salt Tl.\“haBFq—l\IaF.61"63 Most of this work is concerned with the determina-

tion of the oxidation states of the elements, the most suitable electrode

 

<t
>~

It seems clear that all items described under this heading would be
of value to DMSR with only minor modification at most.

 
 

 

61

materials for their analysis, and the basic electrochemical characteris-
tics of each element. It has been shown that relatively high concentra-
tions (typically 20 ppm) can be estimated directly from the height of
the voltammetric waves. Lower concentrations can be measured using the
technique of stripping voltammetry through observation of the current
produced when a corrosion product is oxidized from an electrode on which
it has previously been plat,ed.6LP

A voltammetric method has been developed for the determiration of
the U3t/U*t ratio in the MSRE fuel.®® This method involves the measure-
ment of the potential difference between the equilibrium potential of
the melt, measured by a noble electrode, and the voltammetric equivalent

3+/U4+’

of the standard potential of the U couple. The reliability of the

method was verified by comparison with values cobtained spectrophoto-

®3  This determination has been completely automated with a

66

metrically.

PDP-8 computer, which operates the voltammeter, analyzes the data, and

computes the U3+/U“+ ratic. Recently, the method was used to determine

U3+/E“+ ratios in a thorium-bearing fuel solvent, LiF-BeF,-ThF., (68-20-
12 mole %). Ratios covering the range of 107° to >107% were measured
during the reduction of the fuel in a forced-convection loop.2 The data
support the reliability of the method in this medium.

Because the fuel-processing operation presents the possibility for
introducing bismuth into the fuel, a method for bismuth determination is

3* was characterized in LiF-BeF,-

required. The reductive behavior of Bi

ThFy,®? and it was found to be rather easily reduced to the metal. As

an impurity in the fuel salt, bismuth will probably be present in the

metallic state; so some oxidative pretreatment of the melt will be nec-

essary before a voltammetric determination of bismuth can be performed.
The measurement of the concentration of protonated species in the

proposed MSBR coolant salt is of interest because of the potential use

of the coolant for the containment of tritium. The measurement could

also be used to evaluate the effect of proton concentrations on corrosion

rates and as a possible detection technique for steam-generator leaks.

A rather unique electroanalytical technique that is specific for hydro-

62,67

gen was investigated. The method is based on the diffusicn of hy-

drogen into an evacuated palladium—-tube electrode when NaBFy melts are
62

electrolyzed at a controlled potential. The pressure generated in the s
electreode is a sensitive measure of protons at parts-per-billion concen-
trations. The technique offers the advantages of specificity, applica-
bility to in-line analysis, and the possibility of a measurement of
tritium-to~hydrogen ratios in the coolant by counting the sample col-
lected from the evacuated tube. Measurements by this technique have led
to the discovery that at least two forms of combined hydrogen are present
in NaBFy4 melts.

The availability of an invariant reference potential to which other
electrochemical reactions may be referred on a relative potential scale
is a distinct advantage in all electroanalytical measurements. The
major problem was te find nonconducting materials that would be compati-
ble with fluoride melts. Successful measurements were performed with a
Ni/NiF, electrode in which the reference sclution (LiF-BeF; saturated

68,869 Standard

with NiF;) is contained within a single-crystal LaF3 cup.
electrode potentials were determined for several metal/metal-ion couples
which will be present in the reactor salt streams.®® These electrode
potentials provide a direct measure of the relative thermodynamic stabil-
ity of electroactive species in the melts. This information can be used
in equilibrium calculations tc determine which ions would be expected to
be present at different melt potentials.

As noted above, preliminary studies have indicated that, in at
least some of the salt streams, an electrecanalytic method for oxide con-~
centration may be feasible. Determination of C17 in the fluoride meits
(as may be necessary since "LiCl from the fission-product transfer system
could contaminate the fuel) can probably be accomplished by voltammetric
techniques.

The MSBR regquired little effort on transuranic elements other than
to determine whether traces of plutonium and higher transuranics inter-
fered with determination of other pertinent species. Interference pos-
sibilities will be intensified in the DMSR and thus quantitative deter-
mintations of Pu, Am, Np, and Cm at various points must be provided. .

Spectrophoteometric studies. Because molten fluorides react with

 

the light-transmitting glasses usually emplcved, special cell designs

have been developed for the spectrophoteometric examination of MSBR melts.

 
 

 

03

The pendant-drop technique70 that was first developed was later repiaced
with the captive-liquid cell’? in which molten salts are contained by
virtue of their surface tension so that no window material is required.

A concept has been proposed for the use of this cell in an in-line sys-
tem. ? The light path length through a salt in a captive-liquid cell

is determinable but is not fixed. The need for a fixed path length pro-
moted the design and fabrication of a graphite cell having small diamond-
plate windows ® which has been used successfully in a number of research
applications. Another fixed-path-length cell which is still in the de-
velopment stage makes use of a porous metal £f0il’" that contains a number
of small irregular pits formed electrochemically; many of the pits are
etched completely through the foil so that light can be transmitted
through the metal. Porous metal made from Hastelloy N has been pur-
chased to test its use for cell construction.

The latest innovation in cell design is an optical probe® which
lends itself to a sealable insertien into a molten-salt stream. > The
probe makes use of multiple internal reflections within a slot of ap-
propriate width cut through some portion of the internally reflected

> During measurements the slot would be below the surface

light beam. ’
of the molten salt and would provide a known path length for abscrbance
measurements. It is believed that the probe could be made of LaF: for
measurements in NaBFy streams.

Spectrophotometric studies of uranium in the 3t oxidation state
have shown that this method is a likely candidate for in-line determina-

76,7 f .
577 An extremely sensitive absorption

tion of U'Y in the reactor fuel.
peak for ytt may be useful for menitoring residual uranium in depleted
processing streams.  ® Quantitative characterizations, including ab-
sorption peak positions, peak intensities, and the assignment of spectra,
have been made for Niz+, Fe2+, Cr2+, Cr3+, U5+, U022+, Cu2+, Mn2+, Mn3+,
Co?t, Mo®t, cro,?~, Pa*t, Pudt, prit, Nadt, sm®t, Er’t, and Ho’T. Semi-
quantitative characterizations, including absorption peak positions, ap-
proximate peak intensities, and possible assignment of spectra, have

also been made for Ti3+, V2+, V3+, Eu2+, Sm2+, Cm3+, and 0%~

 

ot

"U.S. Pat. No. 3,733,130.
64

Evidence for the existence of hydrogen-containing impurities in
NaBFy was first obtained from near-infrared spectra of the molten salt
and in mid-infrared spectra of pressed pellets of the crystalline mate-
rial.’? 1In deuterium—exchange experiments attempted in flucroborate
melts, twe sensitive absorption peaks corresponding to BF30H™ and BF3;0D7
were identified. There was no evidence thatr deuterium would exchange
with BF30H™; rather, BF:0D” was generated via a redox reaction with im-

®  The absorption spectra of several other species

1

purities in the melt.®
have been observed in fluoroborate melts.® Work on spectrophotometric
methods is also providing data fer the identification and determination
of solute species in the various melts of interest for the fuel-salt
processing system.62
Gas analysis. Some determinations on MSRE samples {see preceding
section) were done by treatment of the salt to produce gases for analy-
sis. Little development of such devices has been attempted since the
MSRE ceased operations. The electrolytic moisture monitor was demon-—
strated to provide more than adequate sensitivity for the measurement
of water from the hydroflucrination method for oxide and to have ade-
quate tolerance for operation at the anticipated radiation levels.®? A
method has been developed for the remote measurement of micromolar quan-
tities of HF generated by hydrogenation of fuel samples using a thermal-
conductivity method after preconcentration by trapping on NaF.®3
Commercial gas chromatographic components for high-sensitivity
measurement of permanent gas contaminants are not expected to be accept-
able at the radiation levels of the MSBR off-gas. Valves contain elas-
tomers that are subject to radiation damage and whose radiolysis products
would contaminate the carrier gas. The more sensitive detectors generally
depend on ionization by weak radiation sources and would obviously be
affected by sample activity. A prototype of an all-metal sampling valve®"
has been constructed to effect six-way, double-throw switching of gas
streams with closure provided by a pressure-actuated metal diaphragm. A
helium breakdown detector was found to be capable of measuring <l-ppm
concentrations of permanent gas impurities in helium. Use of this de-
tector in a simple chromatograph on the purge gas of an in-reactor cap-

sule test demonstrated that it was not affected by radioactivity.85

 
 

 

65

The analysis of the coolant cover gas involves less radioactivity
but more complex chemical problems. Methods are being investigated for
the determination of condensable material tentatively didentified as BF3
hydrates and hydrolysis products86 and for other forms of hydrogen and
tritium. ''Dew=-point" and diffusion methods offer promise for such mea-

87
surements.

In-line applications. The first successful chemical analysis of a
88 3+/

 

filowing molten fluoride salt stream = was demonstrated by measuring U

U*" ratios in a loop being operated to determine the effect of salt on
Hastelloy N under both oxidizing and reducing cenditions. The test fa-
cility was a Hastelloy N thermal-convection loop in which LiF-BeFy-ZrF,-
UFy circulated at V25 mm/s (5 lin ft/min). The analytical transducers

were platinum and iridium electrodes that were installed in a surge tank

where the temperature was controlled at 650°C.

The Ug'“’“/UL:’+ ratio was monitored intermittently for several months
on a completely automated basis. A new cyclic voltammeter, which pro-
vides several new capabilities for electrochemical studies on molten-
salt systems, was designed for use with this system. The voltammeter

89 A PDP-8I computer was used

can be directly operated by a computer.
to contrel the analysis system, analyze the experimental output, make

the necessary calculations, and print out the results.

3 L . ~_ 2 .
+/U T ratios and Cr?% concentrations have

In-line measurements of U
been made in fuel salt in a forced-convection loop. Severe vibration
problems distort the waves and reduce the accuracy of the measurements
when the fuel is pumped at high velocity, but excellent voltammograms

are obtained when the pump is stopped.

In-line instrumentation has been satisfactorily demonstrated in

1922 Reduction waves for Fe3+, Fe2+, cr?* and

operation of the CSTF.
possibly Mo®T were observed at concentrations from 20 to 100 ppm. First-
order decay of active protons was observed to concentrations as low as

a few parts per billion, and hydrogen and tritium both in free and chem-

ically combined form were successfully determined.
66

Prime Development Needs¥ e

 

The following list is a consolidation of the analytical chemistry
development needs identified in the preceding section.

1. Continued on-line demonstration of UFs/UF, ratio and total
uranium concentration in operating loops, in radiation fields, and in
presence of fission products and transuranics.

2. Demonstration of satisfactory on-line methods for determination
of plutonium and pertinent transuranics in realistic fuel and/or process-
ing streams.

3. Continued demonstration of methods for determination of corro-

2+ and Fez+) in fuel, coolant, and some

sion products {particularly Cr
process streams.
4. Methods for on-line determination of bismuth and C17 in fuel
salt from processing plant.
5. Sound methods for determination of protactinium in at least
some of the processing streams and, if possible, in the fuel within
the reactor.
6. In-line determination of valence state and concentration of
some important fission products (Te, Nb, Zr, Nd, and Eu) in fuel or .
pertinent process streams.
7. On-line methods for estimaticn of 0°7 content of fuel.
8. Demonstration of methods for determination of oxygenated and
protonated {(and tritiated) species in coolant salt and, if possible, of
tritiated species in the fuel.
9. Development of methods for determination of U, Pu, Pa, Th, and
0°~ in molten LiCL.
10. Development of methods for Cs+, Rb+, F7, and corrosion prod-
tcts in LiCl.
11. Methods for determination cof Lio, total reducing power, and
hopefully specific metals (especially plutonium and protactinium) in

molten bismuth.

 

"See Ref. 2 for additional details. It is expected that spectro-
photometric methods can be used for some of these, but it is anticipated “
that on-line electroanalytical techniques will carry the major load.
Both should be sufficiently developed that a choice can be made.

 
S 12. Development of reference electrodes suitable for use in the
pertinent systems.

13. Methods for determination of UFg, Fp, HF-H; mixtures, I, TeFg,
and SeFg in process gas streams.

14. Monitor for rare-gas fission products, and HT, I, etc., in
fuel cover gas and in small gas releases from that system.

15. Methods for analysis of cover gas over coclant for HT, HTO,
other tritiated compounds, BFj3, etc.

It is clear that most if not all of the above studies must be demon-

strated to be applicable in the presence of intense radiation fields and
in realistic solutions containing a spectrum of materials that might in-
terfere with the determinations.

Initial testing (most of which has been done) requires simple labor-
atory facilities. Generally {(as has been done in the past), demonstra-
tion should use engineering-scale facilities that test other facets of
the program at the same time. However, special facilities in which alpha-
emitting isotopes (plutonium, etc.) can be safely handled and hot cells
where high levels of activity can be used will clearly be required after
the initial development stage has passed.

In addition, an in-line tegt facility will be required for testing

the complex array of analytical devices in an integrated way.

Estimates of Scheduling and Costs

 

Preliminary estimates of the necessary schedule and of its operating
and capital funding requirements are given here for the analytical chem-
istry program described above. As elsewhere in this document, it has
been assumed (1) that the program would begin at start of FY 1980, (2)
it would lead to an operating DMSR in 1995, and (3) the R&D program will
produce no great surprises and no major changes in program direction will
be required.

The schedule, along with the dates on which key developments must
be finished and major decisions made, is shown in Table 4.1. It seems
certain that the overall R&D programs (including those described else-

where in this deocument) will provide some minor surprises, and scme

 
Table 4.1.

Schedule for analytical research and development

 

Fiscal year

 

 

 

 

 

 

 

 

 

 

methods for Li, Th, Pu, and Pa in molten bismuth alloys.

10,
to 1% level.

 

Demonstrate feasibility of precise spectrophotometric methods

Task
1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995
A 2 3 4 5 & 7 8 9
Electrochemical methods v v v v v v v v v
2 4 5 510 7 2 1
Spectrophotometry v VvV v v v v
2 12 7 9
Chemical methods \Y% v v v
13 2 T g
Gas stream analysis v° v v - v
Yfl4
Gamma spectrometry
i5 16 17
In-line test facility v v v
iB i 20
Special studies \Y v v
Milestones:
L. Complete basic evaluation of electrochemical bismuth methoeds. 11. Establish ultimate precision of spectral methods for
2. Complete development of methods for corrosion products and total uranium.
protonated species for NaBF,. 12. Develop practical in-line transpiration system for
3. Establish feasibility and accuracy of UF3/UFy ratio determi- testing.
nation. 13. Evaluate gas—chromatographic and mass-spectrometric
4. Demonstrate method for protactinium in fluoride streams. methods for testing.
3. Demonstrate methed for plutonium and transuranics in fluoride l4. Complete evaluation of y-spectrometry capabilities,
streams, 15. Complete construction of Analytical Test Facility.
6. Establish methods for Cl17 in fluoride streams and for uranium 16, Demonstrate in-line oxide method for fuel streams.
and thorium in LiCl. 17. Complete tests of in-line transpiration measurements
7. Start evaluation of radiation effects on methods. (includes bismuth, oxide in fuel, ete.).
8. Demonstrate methods for protactinium, plutonium, cesium, and 18. Complete basic studies of radiolytic oxide removal.
oxides in LiCl. 19, Start in-line applications of analytical methods.
9. Complete essential methods for processing system, including 20, Submit recommendations for complete chemical analysis

for reactors.

89
69

 

changes in the analytical chemistry program will be required. No spe-
cific provisions for this are included; but, unless major revisions
become necessary in the middie 1980s, it appears that suitable analy-
tical techniques and instrumentation could be confidently recommended
as shown in this schedule.

The operating funds (Table 4.2) and the capital equipment require-
ments (Table 4.3) are shown on a year-by-year basis in thousands of
1978 dolliars. No allowance for contingencies, for major program changes,

and inflation during the interval has been provided.
Table 4.

 

2. Operating fund requirements for analytical chemistry research and development

 

 

Cost (thousands of 1978 dellars) for fiscal year —

 

 

 

 

 

 

 

 

 

 

 

Task e i e i - ane e —_— — W e e o o — —_—

1980 1981 1982 1983 1984 1985 1986 1¢87 1988 1589 1990 1991 1992 1993 1994 1995

Electrochemical methods 135 152 160 196 225 240 250 240 207 135 100 100 100 75 50 0
Spectrophotometry 30 56 72 85 115 104 99 a1 104 65 52 40 40 30 20 Q
Chemical methods 30 57 70 78 87 85 97 78 78 105 78 50 35 35 30 0
Gas stream analvsis 21 46 57 58 63 78 85 a1 78 78 52 50 25 0 0 0
GGamma spectrometry 0 0 0 0 0 20 38 59 59 65 38 35 35 35 0 0
In-line test facility 16 39 52 65 78 85 91 g1 52 0 0 0 0 Q 0 0
Special studies 28 55 74 88 102 103 104 110 117 167 160 160 150 100 100 0
Total funds” 260 405 485 570 670 715 765 760 695 615 480 435 385 275 200 0

“Total funds through 1994: $7715.

 

 

0L
 

 

 

 

 

 

 

 

 

 

Table 4.3. Capital equipment fund requirements for analytical chemistry research and development
Cost (thousands of 1978 dollars) for fiscal year —

Ta Sk B . .
1980 1681 1982 1983 1984 1985 1986 1987 1588 1989 1990 1991 1992 1993 1994 19385
Electrochemical methods 12 8 70 20 8 0 10 a 0 0 0 0 0 0 0 0
Spectrophotometry 8 35 35 70 6 100 40 30 0 0 0 0 0 0 0 0
Chemical methods 3 17 5 40 12 0 0 0 0 0 0 0 0 0 0 0
Gas stream analysis 0 35 35 0 7 50 7 Q 0 0 0 0 0 0 0
Gamma spectrometry 0 e 0 0 0 0 65 0 0 40 0 o 0 0 0 Q
In-line test facility 0 150 75 0 0 0 0 0 0 0 0 0 0 0 0 0
Special studies 12 50 70 80 152 1G5 0 0 0 g 0 0 0 0 0
Total funds” 35 295 290 210 183 2535 120 30 0 40 0 0 0 0 0 0

 

“Total funds through 1990:

51460,

 
72

5. MATERIALS DEVELOPMENT FOR FUEL REPROCESSING -

Scope and Nature of the Task

 

The materials required for molten-salt fuel reprocessing systems
depend, of course, upon the nature of the chosen process and upon the
design of the equipment to implement the process. For the MSBR#* the
key operations in fuel reprocessingl’2 are (1) removal of uranium from
the fuel stream for immediate return to the reactor, (2) removal of
23%pa and fission-product zirconium from the fuel for isclation and de-
cay of 233pa outgide the neutron flux, and {(3) removal of rare-earth,
alkali-metal, and alkaline-earth fission products from the fuel solvent
(L.iF-BeF;-ThF,) before its return, along with the uranium, to the reactor.

Such a processing plant will present a variety of corrosive environ-

. 1,2
ments. Those of greatest severity are as follows: »°

1. the presence of molten salt along with gaseous mixtures of F» and UFg

at 500 to 550°C;

B

the presence of molten salts with absorbed UFg so that average val-
ence of uranium is near 4.3 (UFy.s} at temperatures near 550°C;

3. the presence of molten salts (either molten fluorides or molten LiCl)
and molten alloys containing bismuth, lithium, thorium, and other

metals at temperatures near 650°C: and
P

4‘_\

the presence of HF-H; mixtures and molten fluorides, along with bis-
muth in some cases, at 550 to 6€50°C.

5. the presence of interstitial impurities on the outside of the system
at temperatures to 650°C, particularly if graphite or refractory

metals are used.

The sizes and shapes of the components of the processing plant will
evolve as additional pilot-plant work is performed on the various process-
ing steps. The flow from the reactor is of the order of 60 em®/s (1
gpm}, so the piping sizes will be quite small. The crucial process in
most of the processing vessels is that liquids be contacted te transfer

selected materials from one stream to the other. This contacting could

 

afa
EaY

See Chapter 6 for a detailed description of fuel reprocessing.

 
 

 

73

be carried ocut in columns with countercurrent flow or it could be ac-
complished in rather simple mixer-settler vessels with some stirring of
the fluids at their interface. Thus the processing vessels may simply
be pipe sections a few meters long and a few centimeters in diameter or
they may be cells (about 1/2-m cubes) intercomnnected with small-diameter
tubing. The high radiation and contamination levels will require that
the processing plant be contained and fiave strict environmental control.
If the components are constructed of reactive materials such as molybdenum,
tantalum, or graphite, the environment must be an inert gas or a vacuum
to prevent deterioration of the structural material.

Obviously, materials capable of long~term service under these con-
ditions must be provided. The development program necessary to do this

is described below.

Key Differences in Reactor Concepts

 

The fuel reprocessing plant envisioned for a DMSR* will differ in
several nontrivial regards from that of the reference MSBR. However,
in the conceptual DMSR plant® the same unit operations and processes
will be used as for the MSBR. Accordingly, it seems certain that mate-
rials satisfactory for construction of the reference MSBR reprocessing
plant would suffice for a DMSR and that development needs identified

for that MSBRI’2

plant differ trivially, if at all, from those of the
generally similar DMSR plant.

Alternative processes (oxide precipitation of protactinium and,
perhaps, of uranium) have been identified as being less satisfactory
but probably feasible fall-back positions for some (not all) of the
MSBR processing plant operations. It is less clear (see Chapter 6)
that these would be feasible for the DMSR;* should they prove so, the

materials effort to support them would differ little from that for an

MSBR.

 

“See Chapter 6 for a detailed description cf fuel reprocessing.

The presence of plutonium and transuranics is a complicating factor
for a DMSR.
Pogt—1974 Technology Advances

 

After 1974 the MSBR Program had severely limited funding and could
afford little effort on development of materials for reprocessing equip-
ment. Accordingly, the situation for this important development program

is not markedly different from that described in the mid-1974 survey.?

Static tests for 3000 hr at 650°C showed that concentrated bismuth-
lithium alloy (48 at. % lithium) contained in graphite crucibles® per-
meated graphite specimens nearly uniformly and to a depth (0.13 to 0.4 mm,
or 5 to 15 mils) that depended on graphite density.+ Less-concentrated
alloy (4.8 at. 7 lithium) showed little evidence of penetration except

30 However, ATJ graphite speci-

in low—density regions cof the specimens.,
mens, tested in a molybdenum thermal convection loop for 3000 hr at hot-
and cold-leg temperatures of 700 and 600°C showed very large weight gains
of up to 65%, virtually all due te bismuth, though some molybdenum was
present. ° It seems possible, though it has not been confirmed, that the
molybdenum (perhaps by formation of a carbide) greatly promoted wetting

and permeation of the graphite by the alloy.

Tantalum and its alloys (particularly Ta—l107 W) are known to be
stable to bismuth-lithium alloysfil’2 but the effect of molten fluorides
on these alloys is not known. A thermal-convection loop of Ta—107 W was
started in February 1976, with Li¥-BeF;-ThFy-UFy (72-16-11.7-0.3 mole %)
and with hot- and coid-leg temperatures of 690 and 585°C, respectively’,91
The loop was kept in operation when the MSR Program was terminated; it
is still in operation after more than 18,000 hr with no appreciable change
in flow characteristics.?? Therefcre, no marked mass transfer of the
alloy appears to have occurred. Since reprocessing equipment probably
does not need to have temperature gradients as high as 100°C, the Ta—10%

W alloy shows real promise for use in reprocessing equipment.

 

o

“
Contained in sealed capsules of stainless steel to prevent atmo-
spheric contamination.

+These concentrated alloys would be used to strip rare-earth fission
products from LiCl. BRismuth-lithium alloys for selective extraction of
uranium, protactinium, plutonium, etc., would be much more dilute.

 
 

 

~J
N

Present Status of Technology

 

Materials for fluorinators and UFs absorbers

 

Nickel or nickel-base alleys can be used for construction of fluori-
nators and for containment of ¥, UFg, and HF, though these metals would
require protection by a frozen layer of fuel solvent over areas where
contamination of the molten stream by the otherwise inevitable corrosion
products would be severe. Many years of experience in fabrication and
joining of such alloys have been accumulated'*” in the construction of
reactors and asscociated engineering hardware.

The corrosion of nickel and its alloys in the severe environment
represented by fluorination of UFs from molten salts has been studied
in some detail. Most cof the data were obtained during operation93 of
two plant-scale fluorinators constructed of L nickel at temperatures
ranging from 540 to 730°C. A number of corrosion specimens (20 differ-
ent materials) were located in the fluorinators. Several specimens,
including Hymu 80 and INOR-1, had lower rates of maximum corrosive at-

33,94 .
359 Nevertheless, L nickel, protected where neces-

tack than L nickel.
sary Ey frozen salt, is the preferred material for the fluorination—UFg-
absorption system since the other alloys would centribute volatile fluo-
rides of chromium and molybdenum to the gaseous UFg.

Absorption of UFg in molten salts containing UFy is proposed (see
Chapter 6) as the initial step in the fuel reconstitution for the MSBR
and DMSR. The resulting solution, containing a significant concentra-
tion of UFs, is quite corrosive. In principle, and perhaps in practice,
the frozen salt protective layer could be used with vessels of nickel.

73598 that gold is a satisfactory container in small-

It has been shown
scale experiments, and plans to use this expensive, but easily fabricable,
metal in engineering-~scale tests have been described. ?’

Should continuocus fluorination and UFg absorption prove incapable
0of development because of materials or engineering design problems, it

seems 1ikely that alternatives may exist.* Therefore, success with the

 

However, they are considered much less desirable for technical or
economic reasons (see Chapter 6).
76

materials problems for this segment of the processing system is highly S

desirable but perhaps not absolutely essential.

Materials for selective extractions

 

Most of the essential separations required of the processing plant
are accomplished by selectively extracting species from salt streams into
bismuth-lithium alloys or vice versa.* These extractions pose difficult
materials problems. Materials for containment of bismuth and its alloys
are known, as are materials for containment of molten salts. Unfortu-

nately, the two groups have few common members.

Iron and nickel alloys. Carbon steels and low-chromium steels were

 

5 to be satisfactory for long service in bismuth contain-

shown long ag09
ing uranium at temperatures up to about 550°C, but such service required
additions of magpnesium or zirconium to the bismuth. Such additions

would not survive the separations steps in the processing cycle. More-
over, the carbon steels are not really satisfactory long-term containers

100 45 materials

for molten fluorides. Carbon steels have been used’°:’
of construction for engineering tests of selective extraction processes
pending development of better materials. However, it has never seemed
likely that carbon or low-alloy steels could be used satisfactorily as
key components of the processing system. Nickel-based alloys are known ' 2

not to be adequate containers fer bismuth.

101,102 showed

Molybdenum. Corrosion studies at ORNL! and elsewhere
molybdenum to resist attack by bismuth and to show no appreciable mass
transfer at 500 to 700°C for periods up to 10,000 hr. Moreover, molyb-
denum i{s known to have excellent resistance to molten fluorides.'»® It
is reactive with oxygen, but a purged argon or helium atmosphere con-
taining up to 10 ppm oxygen would be acceptable. These facts make it

quite attractive as a material for the processing plant; however, there

are major difficulties associated with its use. .

 

e

Moreover, no satisfactory alternative to the selective extraction-
metal transfer process for removal of rare-earth fission products has
been identified.

 
 

=
%

 

77

Molybdenum is a particularly structure-sensitive material; its
mechanical properties vary widely depending upon how it has been metal-
lurgically processed. The ductile-brittle transition temperature for
molybdenum varies from below room temperature to 200-300°C, depending
both upon strain rate and the microstructure of the metal. Maximum duc-
tility is provided in the cold-worked, fine-grained condition. The arc-
melted molybdenum now available commercially affords relatively good
control of grain size and interstitial impurity level. Nevertheless,
the use of molybdenum as a structural material requires highly spe-
cialized assembly procedures and imposes stringent limitations on system

design from the standpoint of geometry and rigidity.l’2

Many advances in the fabrication technology of molybdenum were made
at ORNL during attempted censtruction of a melybdenum system in which
bismuth and molten salt could be countercurrently contacted in a 25-mm-
ID, 1.5~m~high packed column having 90-mm-ID upper and lower disengaging
sections, ®® Techniques were developed for the production of closed-end
molybdenum vessels by back extrusion. Parts that were free from cracks
and had high-quality surfaces were produced consistently with this tech-
nique by the use of ZrO;-ccated plungers and dies and extrusion tempera-
tures of 1600 to 1700°C. The 1.7-m-long molybdenum pipe for the extraction
column, having an outside diameter of 29.5 mm and an inside diameter of

25 mm, was produced by floating-mandrel extrusion at 1600°C.

It was found that commercial molybdenum tubing can be made ductile
at room temperature by etching 0.025 to 0.08 mm of material from the
tube interior.'s?

03 (either by

Complex components have been fabricated by welding1
gas—tungsten-arc or by electron-beam techniques). Two of the most im-
portant factors found to minimize molybdenum weldment cracking have been
stress relieving of components and preheating prior to welding. Mechanical
tube-to-header joints have also‘been produced by pressure bonding by
use of commercial tube expanders. An iron-base alloy of the composi-
tion Fe-Mo-Ge-C-B (75-15-5-4-1 wt %) has been found to have good wetting
and flow properties, a moderately low brazing temperature (<1200°C), and

adeguate resistance to corrosion by bismuth at 650°C.»?
78

Although molybdenum welds that are helium leaktight have been pro- s
duced consistently using both the electron-beam and tungsten-arc tech-
niques, the ductile-brittle transition of the resulting welds was above
room temperature, and it was necessary to design each joint to support
the welds mechanically. The joints were also back-brazed or vapor plated
with tungsten to provide a secondary barrier against leakage. &

Both previous survey‘sl’2 concluded that

The results of work to date on molybdenum fabrication tech-

niques have been quite encouraging, and it is believed that

the material can be used in constructing components for pro-

cessing systems if proper attention is given to its fabrica-

tion characteristics.

That statement still appears to define a tenable position. However,
the test assembly described above was not completed.* Each of the fabri-
cation steps constituted a special R&D effort, and each welding and
brazing operation was an adventure. There is little doubt that further
advances in molybdenum metallurgy will be made, and it seems certain that
molybdenum fabrication research should continue as a part of the DMSR.
However, such fabrication will be slow and expensive for some time, and
the products are likely to be of uncertain reliability,+ Since it can
hardly be considered likely that complex engineering equipment of molyb-
denum can be provided on a short-term schedule, fabrication of molybdenum
could be given a lower priority than previously suggested.l’z

Cn the other hand, the coating of conventional materials (such as
iron- or nickel-based alloys) with melybdenum should probably be con-
sidered for higher priority. Two types of coating processes have been
investigated:2 chemical-vapor deposition by hydrogen reduction of MoFg
and deposition from molten-salt mixtures containing MoFg¢ by chemical
reaction with the substrate. The latter method looks especially promising

because more complicated components could probably be coated using this

approach.

 

ol
"

It was still incomplete when the program terminated in 1973 and .
was not revived, partly because of funding limitations, when the program
was revived in 1974.

These remarks clearly do not apply to simple items such as crucibles,
stirrers, impellers, or transfer lines.

 
 

 

79

Tungsten and tantalum alloys. Pure tungsten is resistant tec molten

 

bismuth. Because of its high ductile-brittle transition temperature,

it is not at all amenable to the fabrication and joining operations re-
guired for complex equipment, but crucibles, stirrers, etc., of tungsten
could be used. Its use as a surface coating (by chemical vapor deposi-
tion) on meolybdenum was noted above and could perhaps be extended to more
conventional metals. Atmospheric protection equivalent to that required

for molybdenum would be necessary for tungsten.

Pure tantalum and some of its ailoys with tungsten (in particular,
T-111: 8% W, 2% Hf, balance Ta) have been shown to be usefully compati-
ble with molten bismuth and bismuth-lithium alloys. In quartz thermal-
convection loops at 700°C, the mass transfer rate of pure tantalum in
these liquid metals was greater than that of molybdenum, although the
rate was still less than 0.08 mm/year. Mass transfer rates of the alloy
T-111 were comparable to those for molybdenum, but the mechanical proper-
ties of the former alloy were strongly affected by interaction with in-
terstitial impurities, primarily oxygen, in the experiments with pure
bismuth in quartz loops. A more recent test carried out at 700°C with
the bismuth—2.3 wt % lithium mixture in a loop constructed of T-111
tubing did not measurably affect the mechanical properties of the T-111,

2
There seems

and the mass transfer rate again was insignificant.l’
little reason to expect that an alloy of tantalum and tungsten alone
(Ta—10% W, for example) would behave badiy in bismuth-lithium alloys

at 650°C.

Tantalum and its alloys have the very great virtue cof relatively
easy fabricability. Several complex assemblies have been fabricated at
ORNL using the T-111 alloy, the largest of which was a forced-convection

10% A thermal-

loop which circulated liquid lithium for 3000 hr at 1370°C.
convection loop®! of Ta—107% W was constructed in 1976 and is still in
operation. In contrast to molybdenum, the alloy is quite ductile in the
as-welded condition; thus it appears promising for complex geometries.
The tantalum alloy, however, would require a higher degree of protection

from interstitial impurities (oxygen, carbon, nitrogen) than would molyb-

denum. It is likely that a tantalum system would require operation in a
80

high vacuum [107° Pa (V10”7 torr)] and that sufficient purity could not
be maintained in an inert gas purge system.

The resistance of tantalum and its alioys to molten fluorides has
long been questioned, but no definitive tests had been made when previ-
OUS surveys were written.'s? Further tests are obviously necessary, but
the continued satisfactory operation® of the Ta—10% W loop with LiF-BeF,-
ThF,-UF, must be considered encouraging. Use of tantalum in contact with
LiCl and bismuth-lithium alloys {(as in the rare-earth transfer system)
has previously been considered a likely possibility.2

A high priority should be given to tests of tantalum alloys (par-
ticularly the Ta—10% W variety) in combinations of fluoride salts and
bismuth~lithium alloys. Should they succeed, it seems likely that the
fabrication problems could be readily managed.

Graphite. Graphite, which has excellent compatibility with fuel
salt, also shows promise for the containment of bismuth. Compatibility
tests to date have shown no evidence of chemical interaction between
graphite and bismuth containing up to 3 wt 7 (50 at. %) lithium. However,
the largest open pores of meost commercially available polycrystalline
graphites are penetrated to some extent by liquid bismuth. Static cap-

1
sule testg!?®

of three commercial graphites (ATJ, AXF-5QBG, and Graphi-
tite A) were conducted for 500 hr at 700°C using both high-purity bismuth
and bismuth—3 wt % lithium. Although penetration by pure bismuth was
negligible, the addition of lithium to the bismuth appeared to increase
the depth of permeation and presumably altered the wetting characteristics
of the bismuth. Results (see above) obtained recently in a thermal-con-
vection loop of mclybdenum containing graphite specimens at 600 to 700°C
in bismuth—3.8 at. Z lithium were considerably more pessimistic.

Limited penetration of graphite by bismuth sclutions may be toler-
able. 1If not, several approaches have the potential for decreasing the
extent to which a porous graphite is penetrated by bismuth and bismuth-
lithium alloys. Two well-established apprcaches are multiple impregna-
tions with liquid hydrocarbons, which are then carbonized and/or

graphitized, and pyrocarbon coatings. Other approaches are based on

 

L.

KSee above, under Post~-1974 Technology Advances.
 

 

81

vapor-deposited melybdenum coatings and the use of carbide-forming

sealants.

Fabrication of a processing plant from graphite would necessitate
graphite—-to~graphite and graphite~to-metal joints. Development studies
have been conducted on both types of joints using high-temperature brazes

106,107 Several of these

and also metals which bond by carbide formation.
experimental joining techniques show promise for the chemical processing
application. Graphite-to-graphite joints can also be made with plastic

108,

cements. Other workers 103 have pioneered mechanical joints that

may be satisfactory for the proposed application.

It seems very likely that graphite can be used successfuily in at
least some portions of the plant. Graphite crucibles, complete with
relatively simple piping connections, as liners within vessels of con-
ventional ailoys would seem to be clearly feasible. Whether truly com-
plex and interconnecting assemblies can be reliably fabricated from

graphite must, however, remain somewhat speculative at present.

Summary. Molybdenum exhibits excellent compatibility with the
working fluids, and the extermnal environment could be inert gas, but
the problems in fabricating molybdenum are great. Tantalum is easy to
fabricate and is likely compatible with the working fluids, but the ex-
ternal environment must be a hard vacuum. High-density graphite is
likely compatible with the working fluids and can be adequately protected
on the outside with an inert gas, but it is difficult to fabricate into
complex shapes. As the chemistry of the processing system is engineered
further through pilot plants, the precise type of hardware needed will
be better defined. The approach taken to materials development will be
to initially emphasize definition of the basic material capabilities with
respect to salt, bismuth-lithium, lithium chloride, and interstitial im-
purities, and then to develop a knowledge of fabrication capabilities.
As process equipment becomes better defined, this information will be
used to engineer the necessary components. This will involve the de-

sign, construction, and testing of prototypic units.
82

Primary R&D Needs gz

 

The R&D program will necessarily be concerned with detailed tests
of materials compatibility and with studies of welding, brazing, and ‘
other jeining techniques as well as joint design. Facilities for static
testing, for operation of thermal-convection loop assemblies, and for
fabrication and operation of forced-convecticn (pumped) loops will be
required along with sophisticated equipment for welding, brazing, etc.,
under carefully controlled atmospheres. Such facilities have been used
routinely in the past and involve little, if any, additionzl development.

The foliowing key R&D items are required:*

1. Further demonstrate compatibility of tantalum alloys at realistic
temperatures
a. with molten fluorides,
b. with molten chlorides,
¢. with salt-bismuth-lithium, etc., combinations, and
d. at controlled contamination levels.
2. Prepare sound molybdenum, tungsten, and tantalum coatings on conven-
tional substrate metals in realistic geometries.
3. Test such molybdenum and tungsten coatings
a. with pertinent salt-bismuth alloy combinations,
b. wunder thermal stress and thermal shock, and
c. at controlled contamination levels,
4. Conduct compatibility survey with a thermal-convection loop contain-
ing specimens of all likely materials
a. with molten LiCil,
b. with molten fluoride, and
c. with moiten bismuth alloy containing "lithium, thorium, etc.
5. Further test graphite compatibility with bismuth-lithium alloys con-
taining other pertinent metals (i.e., thorium, uranium)} and extend

to "impervious'" graphites. .

 

*

Not all of these items must be successful, but all need to be car-
ried until a material (or materials) for all portions of the plant has
been demonstrated.

 
 

 

33

6. Demonstrate suitable methods for joining of graphite and for obtain-

ing "impervious' joints.

-~

Continue studies of fabricaticn of molybdenum at a reasonable pace.

e

Perform studies as needed on fabrication of tantalum alloys.

O

Test, preferably in forced-convection loops with reasonable tempera-
ture gradient (100°C?), overall compatibility of the salt-alloy com-
bination with a combination of all materials to be included in the

final system.*

Estimates of Scheduling and Costs

 

Preliminary estimates of the necessary schedule and of its operating
and capital funding reguirements are presented below for the development
of materials for fuel processing described above. As elsewhere in this
document, it has been assumed that (1) the program would begin at the
start of FY 1980, (2) it would lead to an operating DMSR in 1995, and
(3) the R&D program will produce no great surprises and no major changes
in program direction will be required.

The schedule, along with the dates on which key developments must
be finished and major decisions made, is shown in Table 5.1. Tt seems
certain that the overall R&D programs (including those described else-
where in this document) will previde some minor surprises and that some
changes in this development will be required. No specific provisions
for this are included; but, unless major revisions become necessary in
the middle 1980s, it appears likely that suitable materials for the
processing cperation could be recommended on this schedule.

The operating funds (Table 5.2) and the capital equipment require-
ments (Table 5.3) are shown on a year-by-year basis in thousands of 1978
dollars. Ko aliowance for contingencies, major pregram changes, and in-
flation during the interval has been provided. The first task group
involves the development of resistant coatings for use on graphite and/or
other more conventional structural materials, while the second group of

tasks includes the simultaneous development of refractory materials

 

o -
Clearly, this proof test is required for both the fluoride-alloy
and LiCl-alloy systems.
 

Table 5.1,

Schedule for fuel processing materials development

 

Fiscal vyear

 

 

 

 

 

 

 

Task
1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991
1 2 3 4
Coating development Vv \ v v
5 &6 7 8 9 10 {1
Compatibility demonstration vV V¥ v v v v
412 13 14 15 i
Fabrication development v Y vV v v
17 18 19
Engineering experiments v v v
Milestones:

1. Demonstration of sound coating of simple speci- 9. Start of tests of coated complex geometries in
mens with Me, W, and Ta. salt—-alloy combination with forced circulation.

2. Preliminary assessment of feasibility of coating 10. Assessment of feasibility of coatings for pro-
concept. cessing plant use.

3. Demonstration of sound coatings on complex geome- 11. Demonstration of applicability of Ta—107 W for
tries. processing plant use and assessment feasibility

4, Final assessment of suitability of coating tech- of Mo from compatibility standpoint.
niques. * 12. Completion of fabrication studies with graphite.

5. Preliminary assessment of compatibility of Ta—10% 13. Completion of fabrication studies on Ta alloys.
W in fluorides. 14. Completion of fabrication studies on Mo.

6. Preliminary assessment of compatibility of Ta—10% 15. Completion of joining studies on graphite.

W in LiCl. 6. Completion of joining studies on Mo.

7. Start of thermal-convection tests of coated speci- 17. Completion of surveillance program on specimens
mens in Bi~Li-Th and of all candidate materials from Reductive Extraction Process Facility.
together in Bi-Li-Th, molten fluoride (0.1 mole % 18. Completion of surveillance program on samples
UFy), and LiCl. from Metal Transfer Facility.

8. Selection of materials for use in Integrated Pro- 19. Final decisions on materials for demonstration

cess Test Facility, completion of thermal-convec-

tion tests of Mo in molten salts, and start of

tests of graphite in Bi-Li-Th in forced-convection

loops.

reactor processing plant.

v8
1

Table 5.2. Operating fund requirements for fuel processing materials development

Cost (thousands of 1978 dollars) for fiscal yvear —

 

 

 

 

 

 

 

 

Task e —

1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991

Coating development 60 85 100 100 100 50 25 0 0 0 0 0
Compatibility demonstration 245 325 400 470 560 560 440 300 150 80 80 80
Fabrication development 100 160 240 260 270 200 200 200 150 75 75 75
Engineering experiments 20 40 80 120 120 120 100 100 100 50 50 25
Total funds® 425 610 820 950 1050 330 765 600 400 205 205 180

“Total funds through 1991: $7140,

$8
Table 5.3. Capital

fund requirements for fuel processing materials development

 

Cost (thousands of 1978 dellars) for fiscal year —

 

 

 

 

 

 

 

 

 

Task
1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991
Coating development 0 50 125 50 50 0 0 0 0 0 0
Compatibility demonstration 100 1125 1675 1300 1200 600 360 250 150 0 0
Fabrication development 0 0 220 160 30 0 0 0 0 0 0
Engineering experiments 0 0 50 50 100 100 100 100 100 100 0 0
Total funds® 100 1175 2070 1560 1380 700 400 350 250 100 0 0

 

“Total funds through 1991: $8085.

98
 

 

87

through construction and operation of both natural- and forced-circula-
tion corrosion loops. The third task group invelves development of
fabrication techniques for the materials under development, while the
fourth group is concerned with the evaluation of material performance

in gctual processing test equipment. It is, of course, conceivable that
a high level of success with one material might eliminate much of the
effort and asscciated costs on other materials.

This entire activity is time-scaled for completion in parallel with
the development of the reactor concept. However, since the reactor could
operate as a converter without on-line processing, the maior part of this
work could be deferred with the expectation of back-fitting a processing
plant to then-existing reactor facilities if the economic incentive were

sufficiently great when development was completed.
88

6. FUEL PROCESSING -

Scope and Nature of the Task

 

Fluid-fueled reactors, unlike more conventional reactor types, offer
the opportunity for continuocus processing of the fuel in a facility lo-
cated at the site and directly coupled to the reactor. Continuous pro-
cessing of its mclten fluoride fuel on a relatively short cycle time* is

233y through

esgential if an MSR is to be a high-performance breeder of
the intermediate 2°%°Pa from 2°2Th. For a "hold your own'' converter (con-
version ratio = 1.0 throughout reactor life), a longer processing cycle
time almost certainly suffices, but there is no doubt that an integrated
and essentially continuous processing system is required. A DMSR that

is to be fed no (or even little) fissile material after its initial
charge is an example of such a reactor.

Removal of fission products that are appreciable neutronic poisons
is, of course, the primary purpcse of fuel processing. In an MSR (see
Chapter 3), not all of these species remain in the salt that flows to
the reprocessing plant. The worst of the lot is *°°Xe, which is vir-
tually inscluble in the melt and is stripped by helium sparging within
the reactor itself. Noble and seminoble metals are largely deposited
within the primary heat exchanger and on other metal within the reactor
system. The serious neutron poisons that are delivered to the process-
ing plant, therefore, are primarily rare-earth isotopes dissolved as
trifivorides in the molten fuel. These, along with other less important
fission-product species, can be removed by extraction inte bismuth~lithium
alloy and subsequently transferred from that alloy into molten LiCl in a

separate processing circuit. Unfortunately, the wvaluable fuel constitu-

ents (uranium, protactinium, plutonium) all extract into bismuth-lithium

 

oJa

>~

Processing cycle time 1is the time required for processing a2 volume
of fuel salt equal te that contained in the reactor system; for the refer-
ence MSBR, the optimum appeared to be about ten days. Removal time for a
particular species is an effective cycle time equal to the processing cycle ¢
time divided by the fraction of that species removed during a pass through
the processing system.
Lo

‘A DMSR that is routinely fed fissile material of nonweapons grade
(i.e., 20% “°°yU in ?°®U) could be built without chemical processing for
removal of fission products. e
 

 

89

alloy more readily than do the rare earths. Prior separation and re-
covery* of these materials must, accordingly, be part of the processing
scheme.

The processing system must, in addition, (1) maintain the fuel at
an acceptable redox potential (UF3/UF, ratio), (2) keep oxide and cor-
rosion products to tolerable levels in the fuel, (3) remove radioactive
species from any {gaseous) effluent streams, and {(4) place the recovered
fission products into waste forms suitable for at least temporary storage
at the reactor site.

Fuel processing for the MSBR was far from a demonstrated reality at
the termination of that effort,l’2 but all key separations had been re-
peatedly demonstrated individually on a small scale. Overall feasibility
seemed assured from a chemical viewpoint, but much work remained beafore
engineering feasibility could be assured.T The status and the remaining
pressing needs in processing R&D — and their relationships to those of

DMSR — are described in the following section.

Key Differences in Reactor Concepts

 

The fuel mixtures for the MSBR and DMSR are similar in many regards,*

and the overall processing concepts share many features. However, both
the fuel chemistryi and the processing differ in several important ways.

233Pa

A very important feature of MSBR processing was the removal of
from the fuel on a short cyele time (ten days) and its isolation in a
molten~-salt reservoir outside the reactor for decay there to essentially

2
pure 331

. This product was recovered by fluorination; that needed by
the reactor was reintroduced into the fuel, and the excess was stored for
sale and use elsewhere. The proliferation resistance imposed on a DMSR

for this study obviously necessitates abandonment of that portion of the

 

%
Uranium can be separated by flucrination to volatile UFg; protac—

tinium, plutonium, and transuranics other than neptunium cannot; however,
they can be recovered by prior extraction into diiuvte lithium-bismuth alloy.

TMaterials of construction of the several equipment items pose sub-
stantial problems (see Chapter 5).

+
See Chapter 3.
20

MSBR system. As a nontrivial consequence, such abandonment requires

that the DMSR have an alternative scheme for removal of fission-product

zirconium, which was sequestered (on the ten-day cycle time) with “°*Pa

in the MSBR processing plant and was discarded as waste after decay of :

the 2%3pa,

The MSBER produced very little Pu, Am, Np, and Cm; and since what
was produced was sequestered with the 233pa and discarded to waste with
the zirconium, the equilibrium fuel contained very small quantities of
these materials. The DMSR is a prolific producer of plutonium (though
at near equilibrium the plutonium is of relatively poor quality), and it
needs to burn the fissile plutonium isotopes as fuel. The DMSR process-
ing plant, therefore, needs to recover plutonium quantitatively and to
return it immediately (along with #33Pa) to the reactor. As a conse-

quence, DMSR fuel will contain much larger quantities of transuranic

isotopes than did the MSBR.

The DMSR will ncot breed and will not have excess fissile material
for removal and sale for use elsewhere. However, it is likely that in

235U

its early clean cperation (given start-up on at 207 enrichment) it

233 . . . . .
U; this will, of ceourse, exist in a suitably

wilil generate an excess of
denatured state but some storage of it (on NaF beds within the reactor

containment) may be required until it is needed by the reactor.

Conceptual processes for the MSBR and DMSR, accordingly, employ
very similar unit processes. Uranium is largely recovered by fluorina-
tion to UF¢ and is returned immediately to the reactor fuel. In the DMSR,
protactinium, plutonium, and the transuranium nuclides are recovered by
selective extraction into dilute lithium-bismuth alloy and are immediately
returned to the reactor. In both concepts the rare-earth, alkaline-earth,
and alkali-metal fission products are selectively extracted into bismuth-
lithium alloy and subsequently transferred to molten LiCl for recovery
as waste. As in the case of fuel chemistry discussed above, the process- )

ing for the two reactors shows far more similarities than differences.

 
 

w

 

91

Post-1974 Advances

 

Both the proposed MSBR'!*? and the DMSR processes require removal of
uranium by fluorination from the fuel salt and from a waste salt before
discard. Removal from the MSBR fuel salt virtually requires* continuocus
fluorination, and such fluorination from the waste stream is desirable.
The salt streams in the reactor processing plant contain much radiocac-
tivity and are appreciable volumetric heat sources; cooling of the ves-
sel wall to form a frozen salt film without freezing the vessel contents
would certainly seem feasible. Tests with normal (nonirradiated) salt
mixtures must introduce this volumetric heat source artificially, and

110,111

this has proven to be difficult.T Repeated attempts to demon-

strate adequate frozen walls with nonradiocactive salt have been virtual

failures because of malfunction of the resistance heating systems.

Early studies'*® had shown that the sorption of UFg¢ in molten LiF-

BeF;-ThF, containing UFy by the reaction

UFs(g) + UFq(d) -+ ZUFS(d)

is rapid but that the reaction

1

= H + UF -+ UF + HF
2 " (g) °(d) !

(d) (g)

is slow in the absence of a catalyst. The resaction, however, proceeds

rapidly when platinum black,112 platinum alloyed with the gold con-
y

112 113

tainer, or even a limited area of smooth platinum serves as the

catalyst. A facility to study this step on an engineering scale was

11% with molten salt and inert gas, but it was not

built and checked out
operated with UFs.
Two engineering assemblies to study the reductive-extraction—metal-

transfer processes were completed and successfully operated in the post-

 

%
A DMSR with a processing cycle time of 100 days or more could pos-
sibly use batch fluerination; the economic penalty might be acceptable.

Successful frozen wzlls have been obtained with Calrod heaters in
the salt, but such apparatus is hardly suitable for use in fluorination
of uranium.
92

10¢C

1974 pericd. These have been documented in some detail. %> These

assemblies both used mechanically agitated, nondispersing contactors®
to equilibrate molten salt and molten bismuth alloy. Both assemblies
were built of carbon steel, though a graphite liner was used in a por-
tion of one of the apparatuses.

#% nine runs were made to establish rates of

In the first of these,
mass transfer of °°’U and °’Zr between LiF-BeF,-ThF, and bismuth as a
function of agitator speed and salt and metal flow rates. These studies
showed that the system could be readily operated, that mass—transfer
rates increased with agitator speed, and that mass-transfer rates in-
creased markedly with only minor phase dispersal. Although confirmation
is needed, these studies also suggested that some phase dispersal might
be tolerated without undue contaminatiocn of the fuel solvent with bismuth.

The more ambitious experimentIOG demonstrated (primarily with neo-

'*7Xd as tracer) the metal-transfer process for removal of

dymium, using
rare—earth fission products from melten LiF-BeF;-Thy into dilute bismuth-
lithium and their subsequent transfer to molten LiCl and then to concen-
trated bismuth~1ithium alloy. This process was demonstrated on a small
engineering scale [about 1% of the flows required for a ten-day processing
cycle on a 1000-MW(e) MSBR]. Separation of the rare earths from thorium
was demonstrated to be essentially that projected from laboratory-scale
studies.’” However, overall mass-transfer coefficients were lower than
would be required for full-scale metal-transfer process equipment of
reasonable size; this was particularly true of the two Bi-LiCl interfaces.
There, however, it is possible that some phase dispersion can be tolerated.
A considerable study of the characteristics of mechanically agitated

5

. . 1 .
nondispersing salt-metal contactors was concluded''® using mercury and

H;0 to simulate the bismuth-salt system.
The code for computer calculation of the MSBR processing plant per-

A - . . . 16
formance was further refined and its use described in detail.!'!®

 

ol

o~

Such contactors, in which the phases are not dispersed, permit opera-
tion with higher ratios of one phase to the other. They should lead to
less entrainment of bismuth in salt, and they are simpler to fabricate than
are extraction columns. They also must be expected to show poorer mass
transfer characteristics.

i

 
 

 

93

Status of Technology

 

The chemical basis on which the processing system is founded is well
understood; however, only small engineering experiments have been carried

out to date and a considerable engineering development effort remains.

Chemical status

 

Fluorination and fuel reconstitution. Removal of uranium from mol-

 

ten fluoride mixtures by treatment with F» is well understood. Initial
studies at ORNL led to the Fused Salt Fluoride Volatility Program, and
batch fluorination of the irradiated Adircraft Reactor Experiment fuel was
successfully demonstrated.’'’ The studies culminated in the highly suc-
cessful recovery of uranium from various irradiated zirconium-, aluminum-,

118 which in some cases were processed as

119

and stainless steel-based fuels,
early as 30 days after fuel discharge. Uranium recoveries greater than
997 and uranium decontamination factors in excess of 10° were consistently
demonstrated. More recently the Fused Sait Flueride Volatility Process
was used for removal of the “°°U-?3%U mixture from the MSRE fuel salt at

1260

the MSRE site after the reactor had operated for about 1.5 years. This

operation was also highly successful, and the fuel carrier salt was sub-

233y and returned to the MSRE for an additional

sequently combined with
vear of operation. There is no doubt that essentially quantitative re-
covery of uranium can be accomplished 1f necessary and that many details

of fission-product behavior are well understood. Such flucrination also
serves to remove oxide and oxygenated compounds (via their conversion to
fluorides and 02) from the melt. However, for the MSBR a continuous fluocr-
inator {probably of nickel protected by a layer of frozen salt) is essen-
tial, and such a device is, at least, highly desirable for the DMSR.
Additional'study is needed to develop and demonstrate such a device.

Gas phase reduction of UFg to UFy by hydrogenation is a well-known
operation in the nuclear industry, and this process was initially consid-
ered applicable for the MSBR.!»? However, consideration of the difficul-
ties associated with equipment scale-down, UF4 product collection and
holdup, and remote operation prompted a search for a more direct means

for recombining UFs with molten fluoride mixtures. The known chemical
94

behavior suggested that UFg could be absorbed directly into molten salt
that contained UFy. Subsequent experiments verified that the absorption
reaction is rapid and that UFg can be combined quantitatively with mol-
ten fluorides containing UF4 with the simultaneous formation of inter-

12 , \
! Tn the fuel reconstitution

mediate fluorides having a low volatility.
step, a gas stream containing UFe¢ and Fp can be reacted with a recircu-

lating salt stream containing dissolved UFy according to the reactions

- 1 = T

PFuay 72 Faqgy T Usqgy
and

UFL&(d) + UFs(g) = ZUFS(d) .

The dissoclved UFs can be reduced in a separate chamber according to the

reaction

°

\ 1 _
UEs gy ¥ 2 Haqgy T UFegqy T HF .

The final reaction is relatively slow in equipment of goldl’z’121

but, as noted above, can be effectively catalyzed by platinum. Additiomal
study is needed to establish whether, for example, icdine fluorides, TeFg,

SeFg, etc., are absorbed by the strongly oxidized UFs solution.

Selective reductive extraction. Selective extracticn from molten

 

 

fluoride mixtures and from molten LiCl into lithium-bismuth alloys has

2,158

¥

been studied in detail for essentially all the pertinent elements. '’
Bismuth is a low-melting-point (271°C) metal that is essentially

immiscible with molten halide mixtures consisting of fluorides, chlorides,

and bromides. The vapor pressure of bigmuth in the temperature range of

interest (500 to 700°C) is negligible, and the solubilities of Li, Th,

U, Pa, and most of the fission precducts are adequate for processing ap-

plications. Under the conditions of interest, reductive extraction reac-

tions between materials in salt and metal phases can be represented by
 

 

the following reaction:
MXn(salt) + nLi(Bi) = M(Bi)} + nLiX(salt) ,

in which the metal halide MXn in the salt reacts with lithium from the
bismuth phase to produce M in the bismuth phase and the respective
lithium halide in the salt phase. The valence of M in the salt is +n,

515 thatr the

and X represents fluorine or chlorine. It has been found'»?
distribution coefficient D for metal M depends on the lithium concentra-

tion in the metal phase (mole fraction, X

Li) as follows:

log D = n log Xii + log K;

The quantity K; is dependent only on temperature, and the distribution

coefficient is defined by the relation

_ mole fraction of M in metal phase
mole fraction of MXn in salt phase

 

The ease with which one component can be separated from another is indi-
cated by the ratio of the respective distribution coefficients, that is,
the separation factor. As the separation factor approaches unity, sepa-
ration of the components becomes increasingly difficult. On the other
hand, the greater the deviation from unity, the easier the separation.
Distribution data have been obtained for many elements!*?21%91° he-
tween LiF-BeF,~ThF, (72-16-12 mole %) and bismuth-lithium and between
LiCl and bismuth-lithium. As Fig. 6.1 indicates, extraction from the
molten fluoride affords excellent separation of Zr, U, and Pa from Th
and the rare earths but relatively poor separation of the rare earths
from thorium. Plutonium, neptunium, and americium are slightly more ex-
tractable from the fluoride than is protactinium; curium is slightly less
extractable than protactinium. Figure 6.2 shows the quite different be-
havior when molten LiCl is used. Excellent separations of thorium from
the rare-earth and alkaline-earth elements can be made by use of LiCl.
The distribution coefficient for thorium is decreased sharply by the

addition of fluoride to the LiCl, although the distribution coefficients
96

ORNL DWG 70-1250i

 

 

 

 

 

E02h T T EIIIIT’ T T T T T T 1] T T
- Salt: 72-16-12 Mole % Zrt4 43 -
] LiF-BeFz-ThFg -
- Temp: 640 °C i
10+ ]
s ]
- 3
2z
w i
O
E I.O'__ —
b - B
m - -
o i 2
o ]
> 4
o ]
[ i 4
o
@
or
- lOm|_— —
2 - ;
IO"E;“ —
50’3 L by ) oot el A b i L)
g3 1074 10°3 10"2
MOLE FRACTION Li IN BISMUTH
Fig. 6.1. Distribution data between fuel salt and bismuth. .
 

97

ORNL-DWG-70-12502

 

 

 

 

 

 

|03 [ Y T T
b+ Th¢ 640°C
U+3
Ce+3
Nd+3
2

10°}- ]
- Lat3
=
Lii
[®]
z o' -
e
L
O
|9
5
=
5
@
o 10°}- sm*2 -
v
[an]

+2
o1 / o |
Sr+2
Q7S 1074 1073 102 10! 1.0

MOLE FRACTION Li IN BISMUTH

Fig. 6.2. Distribution data between lithium chloride and bismuth.

for the rare earths are affected by only a minor amount. Thus, contami-
nation of the LiCl with several mole percent fluoride will not affect the
removal of the rare earths but will cause a sharp increase in the thorium
removal rate. Data with LiBr are similar to those with LiCl, and the dis-
tribution behavior with LiCIl-LiBr mixtures would not be likely to differ

appreciably from the data with the pure materials.!??

Conceptual MSBR processing flowsheet

 

23 is shown in Fig. 6.3.

The reference MSBR processing flowsheet?»!
Fuel salt is withdrawn from the reactor on a ten-day cycle; for a 1000-

MW(e) reactor, this represents a flow rate of 55 em®/s (0.88 gpm). The
98

ORHL DWG 71-7852

 

 

 

 

 

 

 

 

 

 

{05 MOLE FRAC

 

Lé}
UF, EXTRACTOR

B":z Ry
ThF, SALT DISCARD
SALT UFg & FROCESSED SALT
PURIFICATION REDUCTION } é"“"
1
% 1
| |
1 H i
: > EXTRACTOR |
' (Bi
[ EBl
[ l'—-,-j ;
: L
______________________________________________ ~ g
: | !
1 i
| [
uF i
i EXTRACTOR |
1
1 1
REACTOR ) Lmj !
. | EXTRACTOR ]
1 —_—
: LiCl —===Bi-Li
|
!
1
I
I
I

 

 

 

 

 

 

 

 

T

i ? ¥ == EARTHS
| HF Fz ; }

; —--Bi-Li
L PA DECAY =~ ,

(0.05 MOLE FRAC.

 

 

 

 

I

1

£

|

|

|

|

|

|

1

|

!

|

! 1
6

o ! |

L«E HYDROFLUOF;H FLUORINATORJ-«-{ PA DECAY ] | Bi-Li 5

L — o= +DIVALENT RARE I

:

t

I

i

I

|

|

!

1

 

 

 

 

[ <l -
jai P2 rLuortwator Y | extracTor ! Li)
1
1
: ! Bi-Li
| SALT . b — o ¢ TRIVALENT RARE
i TO bmend EARTHS :
L REDUCTANT | WASTE "
——— e P e e e e e m s e m e . — = -
ADDITION

Li

Fig. 6.3. Conceptual flowsheet for fuel processing in a single-
fluid MSBR. .

fluorinator removes 99% of the uranium. The protactinium extraction con-
tactor is equivalent to five equilibrium stages. The bismuth flow rate
through the contactor is 8.2 em?/s (0.13 gpm), and the inlet thorium con-
centration in the stream is 907 of the thorium solubility at the operating
temperature of 640°C. The protactinium decay tank has a volume of 4.5 m®
(160 ft?®). The uranium inventory in the tank is less than 0.2% of that
in the reactor. Fluorides of Li, Th, Zr, and Ni accumulate in the tank
at a total rate of about 0.003 m®/d (0.1 ftg/day)fi These materials are
removed by periodic withdrawal of salt to a final protactinium decay and
fluorination operation. The bismuth flow rate through the two upper con-
tactors in the rare-earth removal system is 790 em®/s (12.5 gpm), and the
LiCl flow rate is 2080 ecm’/s (33 gpm). Each contactor is equilivalent to
three equilibrium stages.

The trivalent and divalent rare earths are removed in separate con-

tactors in order to minimize the amount of lithium required. Only 27 of

 
 

 

99

the LiCl, or 42 ecm®/s (0.66 gpm), is fed to the two-stage divalent rare-
earth removal contactor, where it is contacted with a 2200-cm®/d (0.58-
gal/day) bismuth stream containing 50 at. % lithium. The trivalent strip-
per, where the LiCl is contacted with bismuth containing 5 at. 7 lithium,
is equivalent to one equilibrium stage.

The remaining steps in the flowsheet consist in combining the pro-
cessed salt with uranium and purifying the resulting fuel salt. The
uranium addition is accomplished by absorbing the UF¢-F; stream from the
fluorinators into fuel salt containing UFy, which results in the formation
of soluble UFs. The UFs is then reduced to UFy by contact with hydrogen.
The HF resulting from reduction of UFs is electrolyzed in order to recy-
cle the contained fluorine and hydrogen. These materials are recycled
to avoid waste disposal charges on the material that would be preduced
if the HF were absorbed in an aqueous solution of KOH.'?" The salt will
be contacted with nickel wool in the purification step in order to en-
sure that the final bismuth concentration is acceptably low.

The protactinium removal time obtained with the flowsheet is 10
days, and the rare-earth removal times range from 17 to 51 days, with
the rare earths of most importance being remcved on 27- to 30-day cycles.

123,125 indicate that the flowsheet is relatively insensi-

Calculations
tive to minor variations in operating conditions, such as changes in
temperature, flow rates, reductant concentrations, etc. However, as
noted earlier, contamination of the molten LiCl by fluoride markedly
increases extraction of thorium by the LiCl. It appears that up to 2
mole % of F~ in the LiCl (which would lead to loss of 7.7 g-moles of
thorium per day} may be tolerable.? It has been shown that treatment
of LiCl contaminated with F~ by BCli serves to volatalize BF3;; the F~
contamination should be easily removable,?»126

The reliable removal of decay heat from the processing plant is
an important consideration because of the relatively short decay time
before the salt enters the processing plant. A total of about 6 MW of
heat would be produced in the processing plant for a 1000-MW(e) MSBR.
Since molten bismuth, fuel salt, and LiCl are not subject to radiolytic

degradation, there is not the usual concern encountered with processing

of short-decayved fuel.
100

Engineering status

 

Continuous fluorinator. As noted above, molten-salt fluorinations

 

-* » 127
have been conducted in several cases. A countercurrent fluorinator

(25 mm diam, 1.8 m long, constructed of nickel) has been operated satis-
factorily. Correlations for gas holdup and axial dispersion have been

developed from studies of air-water sclutions for application to larger .

fluorinators.'?®

Frozen salt layers are believed to be essential for continuous

Z The feasibility of maintaining frozen salt layers in

125

fluorinators.'?
gas—salt contactors was demonstrated previously in tests in a 0.12-
m-diam, 2.4-m-high simulated fluorinator in which molten salt, LiF-ZrF,
(66-34 mole %), and argon were countercurrently contacted. An internal
heat source in the molten region was provided by Calrod heaters contained
in a 3/4-in. IPS pipe along the centerline of the vessel. A frozen salt
layer was maintained in the system with equivalent volumetric heat gen-
eration rates of 353 to 1940 kW/m3° For comparison, the heat generation
rates in fuel salt immediately after removal from the reactor and after
passing through vessels having heldup times of 5 and 30 min are 2000,
950, and 420 kW/m®, respectively.?’

However, as noted previocusly, recent attempts to use autoresistance
heating of the nonradioactive test salt (so that no unprotected metal

would be present in the fluorinator) have proved disappointing.lloslll

 

Fuel reconstitution. As noted earlier, engineering experiments
have been designed, built, and tested in a preliminary way but no en-
. . X . ) 7
gineering studies of fuel reconstitution have been run.

Selective reductive extractions. Both countercurrent extraction

 

columns and mechanically agitated, nondispersing contactors have been
tested on a small engineering scale. Tests of the latter contactors’ 72100
were described above.

A salt-bismuth reductive extraction facility has been operated suc-
cesfully in which uranium and zirconium were extracted from salt by

123,129 o0

countercurrent contact with bismuth containing reductant.
than 957 of the uranium was extracted from the salt by a 2l-mm~diam,
610~mm-long packed column. The inlet uranium concentration in the salt

was about 257 of the uranium concentration in the reference MSBR. These

 
101

 

experiments represent the first demonstration of reductive extraction of
uranium in a flowing system. Information on the rate of mass transfer
of uranium and zirconium has also been obtained in the system using an
isotopic dilution method, and HTU* values of about 1.4 m were obtained.?

21285130 £o4 flooding and dis-

Correlations have been developed2

. persed-phase holdup in packed columns during countercurrent flow of
liquids having high densities and a large difference in density, such
as salt and bismuth. These correlations, which have been verified!??3

by studies with molten salt and bismuth, were developed by study of
countercurrent flow of mercury and water or high-density organics and

water. Data on axial dispersion in the continuous phase during the
countercurrent flow of high-density liquid in packed columns has also

131,132 and a simple relation for predicting the effects

been obtained,
of axial dispersion on column performance’®? has beer developed. An

eddy-current detector'??® for location of the salt-bismuth interface or
bismuth level in an extraction circuit has been suécessfully demonstrated.

All aspects of the metal transfer process for rare-earth removal

have been tested at two different engineering scales,!®?s128

N Interest in mildly agitated, nondispersing extractors has developed
because (1) they should be much simpler to build of difficult-to-fabri-
cate materials such as molybdenum or graphite and (2) they might be less
likely to entrain bismuth in salt or salt in bismuth. Whether these or
mecre conventional extraction columns are to be preferred is not yet
established.

Design and development work has progressed on a Reductive Extraction

sqr2
Process Facility 5129

that would allow operation of the important steps
for the reductive extraction process for protactinium isolation. The
facility would allow countercurrent contact of salt and bismuth streams
in various types of contactors at flow rates as high as about 25% of
those required for processing a 1000-MW(e) MSBR. The facility would

r operate continuously and would allow measurement of mass transfer and

hydrodynamic data under steady-state conditioms.

 

“HTU = height of transfer unit.
102

- Bismuth removal and urapium valence adjustment. Entrained or dis- e
solved bismuth will have to be removed from the salt before it is re-
turned to the reactor, since nickel is quite soluble in bismuth (about
10 wt %) at the reactor operating temperature. Efforts tc measure the
solubility of bismuth in salt have indicated that the solubility is lower
than about 1 ppm, and the expected solubility of bismuth in the salt is
very low under the highly reducing conditions that will be used. It
appears that bismuth can be present at significant concentrations in
the salt only as entrained metallic bismuth. Sampling of salt from en-
gineering experiments indicated that the bismuth concentration in the
salt ranges from 10 to 100 ppm after countercurrent contact of the salt
and bismuth in a packed-column contactor; however, concentrations below
1 ppm are observed in salt leaving a stirred-interface, salt-metal con-
tactor in which the salt and metal phases are not dispersed.2

A natural-circulation loop constructed of Hastelloy N and filled
with fuel salt has been operated for about two yvears; a molybdenum cup
containing bismuth was placed near the bottom of the loop. To date,
the reported concentrations of bismuth in salt from the loop (<5 ppm)
are essentially the same as those reported for salt from a loop contain-
ing no bismuth. No degradation of metallurgical properties has been
noted on corrosion specimens removed from the loop containing bismuth.

Operating a MSR with a small fraction of the uranium present as
UF3 is advantageous in order to minimize corrosion reactions and the
oxidizing tendency of the fission process. The U3+/L’L++ ratio in the
MSRE was maintained at the desired level by reduction of vt with beryl-
lium metal, and a voltammetric method for the determination of this
ratio in the MSRE fuel was developed. The final step in the processing
plant will consist in continuousiy measuring and adjusting the U3+/U”+

ratio of the fuel salt returned to the reactor.

Special characteristics of DMSR fuel processing

The DMSR core will be larger than that of the MSBR, and the inven-
tory of fuel will be larger, probably by about twofold. The optimum
processing cycle time — or even the tolerable limits on it — is not yet .

well defined; however, the limits probably lie between 30 and 150 days.

 
 

 

103

If so, the DMSR will process a smaller volume of fuel per day than would
the MSBR. This is not, in itself, a real advantage since the equipment
was relatively small for the MSBR and will not be much cheaper in smaller
sizes. However, if long cycle times are tolerable to reactor neutronics,
batch operations might be possible at a few points (the fluorinators, for
example}.

Several of the special characteristics of the DMSR will affect the
processing operation to only a minor extent. The presence of plutonium
and the transuranic isotopes is such a characteristic. In the MSBR,
protactinium (plus zircoenium and what plutonium and transuranics were
present) was extracted into bismuth and immediately removed by hydro-
fluorination into the protactinium—isolation salt. In the DMSR these
species will be recovered by reduction into bismuth* and immediate re-
moval therefrom by oxidation+ into the reconstituted and purified fuel.

The higher uranium concentration (by about fivefold) will probably
mean that the quantity of UFg produced per unit time will be larger in
the DMSR than in the MSBR:; the quantity produced per unit of salt will
obviously be higher, and this may make the fuel reconstitution step more
difficult in practice.

A major difference from the conceptual MSBR process may well be
necessary to effect a reasonable removal of fission-product zirconium
from the DMSR. Zirconium is not an important nuclear poison, and ZrFy
at low concentrations should affect the fuel properties trivially. How-
ever, ZrFu in the fuel must be reduced and oxidized each time the fuel
is processed, and the reduction requires expensive "Li. Zirconium is
the most readily reduced of all the pertinent elements (see Fig. 6.1)
and it can probably be separated from plutenium and protactinium with
some difficulty by selective extraction. If sc, it can be removed —
on some reasonable c¢ycle time — by an additional extraction circuit.
Zirconium cannot be reasonably separated from uranium by selective ex-

traction, but uranium lost with the zirconium can readily be recovered

 

Plutonium and the transuranics (except curium) are more readily
extracted than is protactinium. Quantitative recovery cf protactinium

will ensure quantitative recovery of plutonium.

If the DMSR is to operate with “10Z of the uranium reduced, it seems
reasonable to use a part of the bismuth alloy to reduce UFy to UFj.
104

by fluorination. There is little doubt that a better method for removal g
of zirconium is desirable for the DMSR, and some R&D on this should be

done. Possibly, the very stable intermetallic compounds that zirconium

forms with platinum-group metals may offer such a possi,b;i,l:i;ty.“L”135
All other separations processes — and the associated R&D — would

seem very similar to those required for the MSBR. A preliminary concep-

tual flowsheet for the DMSR'*® is shown as Fig. 6.4.

Possible processing alternatives

 

137 appeared to show that protactinium could be

Small-scale studies
oxidized to pa’t by treatment of the LiF-BeF,-ThFy melt with HF and that
the protactinium could then be precipitated as very insoluble Pa»Cs.

Thus it was possible that protactinium could be selectively removed from
the salt without prior removal of the uranium. Removal of protactinium
as Pa»0s might have been a viable alternative for the MSBR, where isola-
tion of this nuclide was a major requirement, but precipitation of only
Pa;0s5 would be of no value to the DMSR. However, it is possible to pre-
cipitate a solid solution of (Th,U)0., containing a very high concentra-
tion (>95 mole %) of U0z, by deliberate addition of oxide ion to the MSBR

5+, its oxide would, of

fuel.’® If the protactinium were oxidized to Pa
course, also be precipitated. 1If, in addition, plutonium were oxidized
to Pu'Tt (this oxidation is more difficult® and a small concentration of
dissolved F2 might be required), Pu0; would also be included in the solid
solution. Whether americium, neptunium, and curium could be made to pre-
cipitate as oxides is not vyet known.

Engineering studies of uranium oxide precipitation have been car-
ried out;l38 the studies involved the contact of 2 liters of MSBR fuel
salt with Hy0~Ar gas mixtures in a 100-mm-diam nickel precipitator. Ex-
periments were conducted at temperatures ranging from 540 to 630°C, and
the composition of the HyO0-Ar mixture was varied from 10 to 35% water.
The values for the water utilization were uniformly low (about 10 to

15%) and did not vary with the composition of the gas stream. Samples

of the oxide contained about 90% UQ; even though, at the lower uranium

 

Jo

w5
See Chapter 3.
 

 

 

 

 

TO
REACTOR

REACTOR
-——-—w—fi

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

PRIMARY

I
J

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

[——— o 2 e sl o e e o e
} HF TG
UF4 ThF,, BeF, | RECYCLE
ADDITION i ===
l i
Ve VALENCE| [ UF UF
ADJUST-- orian
Y ) REDUCTION ABSORPTION
= MENT - ‘
1
REMOVAL b e ?
A T T __
e TEmETmTmTTETEETETTTmTmmET e
9
¢ ot Zr *—l RARE  pe——
¢ [ PARTITION fgm—me EARTH
! | ‘L EXTRACTION
i o e e emn
® § 1
i T T T T T T I T ISR T T et
P i ' i Pa |
? I Zr i Pu
t g EXTRACTION l——-; EXTRACTION
1 7 I Y
|- e i
FROM ¢ | e mmesa——————

 

FLUORINATOR g2

 

 

 

 

 

 

SECONDARY psibiid HYDRO-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

FUEL OR FUEL SOLVENT

————— BISMUTH
~——e— (f
—+ =+~ LiCl

=+ WASTE SALT

FLUORIN- HE 1 FLUORIN-
i) ATOR . ATIOR
C [,

 

irT

L S el M e St e i v Bl e . e e e

REZ ACCUMULATION

 

 

¢OT

 

 

 

 

 

Cs AND Rb |
LACCUMULATORS® 1
!
1
+
' ACCUMULATION
Ah+l+-+1
ADD REZ [ = —
Licl STRIPPER g+ 4
i +
I
RARE  |+q L---—-—I ——————————
EARTH | § +
TRANSFER] § I
T.O -ll‘- —+—-+1+—+-—*):
LiC: ; |
oo e ¥ ! +
Y ! )
EXCESS | 1 TeT fm e Y
FUEL | STRIPPER
SOLVENT SRR
i }
+ e
T FP.AND WASTE SALT
I FOR RECOVERY OR
.. T ULTIMATE DISPOSAL
v

 

 

 

 

Fig. 6.4, Preliminary flowsheet for fuel processing in a DMSR.
106

concentrations in the salt, the solid in equilibrium with the salt would
contain 50% U0, or less. This enhancement of the uranium concentration
in the sclid phase appears to allow precipitation of 997 of the uranium
as a solid containing 857% U02 in a single-stage batch precipitator. The
oxide precipitate was observed to settle rapidly, and more than 907% of
the salt could be separated from the oxide by simple decantation.' Thus ¢
an oxide precipitation scheme could possibly recover together the U, Pa,
and Pu, along with a small amount of Th, as oxides. These could then be
returned to the purified fuel solvent (by hydrofluorination) for return
to the reactor. It would be necessary, of course, to remove residual 0*
from the fuel solvent before the rare-earth removal step. If processing
of a DMSR on a cycle time of 100 days or more is practicable (processing
rate of <1 m® of salt per day), such an oxide precipitation might be
used as a batch operation.¥
We know of no method for rare-earth removal that is comparable to
the reductive—extraction—metal-transfer process. Many attempts have
been made to find lon-exchange systems capable of removal of rare-earth
ions from LiF-BeF; and Li¥F-BeF,-Th¥, mixtures. All but a very few such
materials are unstable in contact with the salts. The only stable one
known to have ion-exchange capabilities is Cels3, which will exchange
Cet for other rare-earth ions,zQ but it is too soluble in LiF-BeF,-ThF,
to be genuinely usefu1.+ The fuel solvent, partially freed from other
rare earths but saturated with CeFs, could possibly be freed from cerium

* and precipitation of Ce0;. If so, the resulting

by oxidation to Ce"
LiF-BeF,-ThFy would again have to be treated to remove excess 0*~ before
its return (with the valuable fissile and fertile constituents) to the
reactor,

Several combinations of the preferred processes with some of the
alternatives are possible. Their attractiveness increases as the per-
missible processing cycle time lengthens. It seems certain, however,

that all are less attractive than that represented by the reference MSBR

unit processes.

 

el

Solids handling is axiomatically more difficult than fluids hand- .
ling, especially in continucus operations.

Cerium is an appreciable neutron absorber.

 
 

 

1G7

Primary R&D Needs

 

The primary needs, with relatively few exceptions, are for sound

engineering tests of individual process steps and ultimately for rela-

tively long-term and near full-scale integrated tests of the system as

a whole. For the former tests the facilities needed are relatively mod-

est, though test equipment and instrumentation are quite complex. For

the integrated tests a special engineering laboratory will be required

and an integrated process test facility must be provided with appropriate

consideration of the need to demonstrate remote maintenance. Specific

needs include

1.

Determination by computer calculation, in close coordination with
reactor neutronics studies, the permissible range of (and the op-
timum) processing cycle times and removal times.

Demonstration of frozen-wall fluorination of uranium on a batch,
and, hopefully, on a continucus, basis. Determination of behaviocr
of neptunium in this fluorinatiom.

Demonstration of adequate UFg¢ abscrption in LiF-BeF,;-ThF,;-UFy; mix-
ture and suitable reduction to UFy and to 7 to 10% UF3;. Determina-
tion of behavior of an NpFg in absorption system.

Development and demenstration of a method for removal (by selective
extraction or otherwise) of fission-product zirconium from the fuel.
Demonstration of quantitative recovery of protactinium and plutonium
by selective extraction on a scale at least 25% of that required
for a DMSR. Determination of efficiency of recovery of americium,
neptunium, and curium in that extraction.

Betermination of behavior of TeFy, SeFg, IFs5, etc., in the UFg ab-
sorption step.

Demonstration of retention of TeFg, SeFg, Iz, etc., from the UFg
absorption off-gas.

Demonstration of adequate removal of rare-earth, alkaline-earth, and
alkali-metal fission products in a complete metal-transfer system.
Demonstration of hydrofluorination of zirconium, rare-earth, etc.,

fission products into waste salt for storage.
108

10. Demonstration of the application of bismuth containing U, Zr, Pa,
Pu, Th, Li, etc., for valence adijustment of fuel salt.

11. Demonstration of adequate removal of bismuth (by absorption on
nickel or gold weol) from salt for return to the reactor.

12. Operation of the entire integrated system reliably for moderate to
long times with realistic construction materials and reasonable con-
centrations of species at tracer level activity (where possible).
Assessment of overall performance, achievable oxide concentration,

effect of system upset on fissile losses to waste, etc.

Estimates of Scheduling and Costs

 

Preliminary estimates of the necessary schedule and of its operating
and capital funding requirements are presented below for the fuel pro-
cegssing development described above. As elsewhere in this document, it
has been assumed that (1) the program would begin at start of ¥FY 1980,
(2) it would lead to an operating DMSR in 1995, and (3) the R&D program
will produce no great surprises and no major changes in program direc-
tion will be required.

The schedule, along with the dates on which key developments must
be finished and major decisions made, is shown in Table 6.1. It seems
certain that the overall R&D programs (including those described else-
where in this document) will provide some minor surprises and that some
changes in this development effort will be required. No specific pro-
visions for this are included; but, unless major revisions become neces-—
sary in the middle 1980s, it appears likely that suitable chemical pro-
cesses and processing equipment could be recommended onr this schedule.

The operating funds (Table 6.2) and the capital equipment require-
ments {Table 6.3) are shown on a year-by-year basis in thousands of 1978
dollars. ©No allowance for contingencies, major program changes, and
inflation during the interval have been provided. As with the develop-
ment of materials for chemical processing (Chap. 5), much of this effort
could be deferred if a decision were made to delay the development of a

break-even fuel cycle for the DMSR.
 

Table 6.1.

 

Schedule for fuel processing development

 

 

Fiscal wvear

 

 

 

 

 

 

 

 

 

 

 

Task
1680 1981 1982 1983 1984 1985 1986 1987 1983 1949 1990 1991 1992 1993 1994 1995
q 2 3
flewsheet develonment v Y Y4 v
5 B 7
Fluorinator development % v v
8 g 7
Fuel reconstitution v v v
10 ‘71$ 12
Pa, Pu, etc., recovery
13 i4
Rare-earth removal v v
Valence adjustment and v’s vm
purification
‘717
MSR Process Laboratory
- N 18 18
Integrated Process Test v Y
Facility
Milestones:
1. Define range of possible values for processing cycle time 10, Complete engineering studies of reductive extraction in a
and removal times. mild-steel {low-through system.
2. Define optimum processing time. 11. Demonstrate recovery of protactinium by reductive extraction
3. Decide on system for removal of zirconium for engineering using gram quantities of “3ipg,
tests. 12. Demonstrate recovery of Pu, Am, Cm, etc., using gram quanti-
4. Complete [lowsheets for conceptual DMSR. ties of Pu.
2. Test batch frozen-wall fluorinator. 13. Extend experiments in mild-steel system at 1% of DMSR scale.
6. Complete studies of continuous fluorination in engineering 14, Complete engineering experiment 5 to 10% DMSR scale,
facility. 15. Demonstrate removal of trace quantities of bismuth.
7. Complete studies of combined fluorination-recombination in 16. Demonstrate continuous adjustment of uranium valence.
engineering system on 25 to 507 DMSR scale. 17. Complete construction of MSR Processing Engineering Labora-
8. Complete engineering studies of fuel recenstitution neces- tory.
sary for design of Fluorination-Reccnstitution Engineering 18. Complete installation of Integrated Process Test Facility.
Experiment. 19. Complete operations and tests with Integrated Process Test

9. Complete engineering studies of reductive extraction in Re-

ductive Extraction Precess Facility.

Pacility.

60T
Table 6.2. Operating fund requirements for fuel processing development

 

 

 

Cost (thousands of 1978 dollars) for fiscal year —

 

 

 

 

 

 

 

 

 

 

 

sk 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1692 1993 1994 1995
Flowsheet development 40 150 250 300 150 75 40 0 0 0 0 O 0 0 0 0
Fluorinator development 290 390 490 315 100 50 0 0 0 0 0 0 0 0 0 0
Fuel reconstitution 200 340 220 310 200 100 0 0 0 0 0 0 g 0 0 0
Pa, Pu, etc., recovery 330 550 610 540 200 75 0 0 0 0 0 0 0 0 0 0
Rare-earth removal 270 200 235 195 100 50 0 0 0 0 0 0 0 0 0 0
Valence adjustment and 50 60 100 100 75 20 0 0 0 0 0 0 0 o 0 0

purification
MSR Process Laberatory 105 65 325 385 200 250 160 O 0 0 0 0 0 @ 0 0
Integrated Process Test Q 215 250 310 455 360 770 3200 3670 3670 3510 2000 500 0 0 0
Facility ‘
- — ; ) e ——

Total funds” 1285 2170 2480 2455 1480 1010 910 3200 3670 3670 3510 2000 500 0 0 0

 

 

“Total funds through 1992: $28,340.

b
Additional funds related to fuel reprocessing will be required during these years in support of test reactor and test reactor
mock-up. The variation in overall support level, therefore, will be considerably less abrupt.

 

OTIl
 

Table 6.3,

Capital equipment fund

requirements for fuel processing development

 

 

 

 

 

 

 

 

 

 

 

 

 

Cost (thousands of 1978 dollars) for fiscal year —
Task - o ——
1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995
Capital equipment facilities
Flowsheet development 0 0 0 0 0 0 0 O 0 0 0 0 0 0 0 0
Flueorinator development 65 260 50 0 0 g 0 0 0 0 0 g 0 0 0 0
Fuel receonstitution 10 165 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Pa, Pu, etc., recovery 0 285 600 0 0 0 0 0 0 0 0 0 0 0 0 0
Rare-earth removal 0 300 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Valence adjustment and 0 50 100 0 0 0 0 0 0 0 0 0 0 0 0
purification
IPTF:” data processing 0 0 0 0 0 510 0 260 400 515 400 200 0 0 0 0
Total funds 75 1060 750 0 G 510 0 260 400 515 200 200 0 o G G
Capital projects
MSR Process Laboratory lB,OOOfi
1
Integrated Process Test 7000”7

Facility

 

- .
IPTF = Integrated Process Test Facility.

®Total funds through 1991: $23,170.

 

17T
o

Ui

~
"

10.

11.

12.

13.

14,

15.

112
References

M. W. Rosenthal et al., The Development Status of Molten-Salt
Breeder Reactors, ORNL-4812 (August 1972).

L. E. McNeese et al., Program Plan for Development of Molten-Salt
Breeder Reactors, ORNL-5018 (December 1974).

Molten-Salt Reactor Program, Semiavnual Progress Report, ORNL-5047
(September 1975).

Molten-Salt Reactor Program, Semionnual Progress Report, ORNL-5078
(February 1978).

Molten-Salt Reactor Program, Semiannual Progress FReport, ORNL-5132
(August 1976).

H. E. McCoy, Jr., Status of Materials Development for Molten-Salt
Reactors, ORNL/TM-5920 (January 1978).

J. R. Keiser, Status cof TelluriumHastelloy N Studies in Molten
Flucride Salte, ORNL/TM-6002 (October 1977).

L. 0. Gilpatrick and L. M. Toth, '"The Hydrogen Reduction cof Uranium
Tetrafluoride in Molten Fluoride Solutiomns," J. Inorg. Nucl. Chem.
39(10), 18171822 (1977).

G. Long and F. F. Blankenship, The Stability of Uranium Trifluoride,
ORNL/TM-2065 (November 1969).

G. T. Mays, A. N. Smith, and J. R. Engel, Distribution and Behavior
of Tritium in the Coolant-Salt Technology Faeility, ORNL/TM-5759
(April 1977).

J. T. Bell, J. D. Redman, and F. J. Smith, "Permeation of Construc-
tion Materials by Hydrogen Isotopes,' Chemistiry Division Annual
Progress Report, CORNL-5297 (September 1977).

W. R. Grimes, '"Molten-Salt Reactor Chemistry," Nucl. Appl. Technol.,
8(2), 137-155 (February 1970).

C. E. Bamberger et al., J. Inorg. Nucl. Chem. 33, 3591 (1971).

Molten-Salt Breeder Reactor Concept, Quarterly Report for Period
Ending July 31, 1971, NP-19145, Atomic Energy Commission, Bombay,
India.

L. M. Ferris et al., "Equilibrium Distribution of Actinide and Lan-
thanide Elements between Molten Fluoride Salts and Liquid Bismuth
Solutions," J. Inorg. lucl. Chem. 32, 2019 (1970).
 

 

16.

17.

18.

19.

20.

21.

22.

23.

24,

25.

26.

27.

28.

29.

30.

31.

113

C. F. Baes, Jr. , "The Chemistry and Thermodynamics of Molten-Salt
Reactor Fuels," p. 617 in Nuclear Metallurgy, Vol. 15, Symposium

on Reprocessing of Nuclear Fuels, edited by P. Chiotti, CONF 690801,
U.S. Atomic Energy Commission, Divisien of Technical Information,
(August 1969).

C. E. Bamberger and C. ¥. Baes, Jr., J. Nucl. Mat. 35, 117 (1970).

C. E. Bamberger, R. G. Ross, and C. F. Baes, Jr., J. Inorg. Nucl.
Chem. 33, 767 (1971).

C. E. Bamberger, R. G. Ross, and C. F. Baes, Jr., in Molten=Salt
Reactor Program Semiannual Progress Report, Febrvuary 29, 1972,
ORNL-4782, pp. 70—76.

C. E. Bamberger et al., in Molten-Salt Feactor Program Semiannual
Progress Repovrt, February 28, 1971, ORNL-4728, p. 62.

C. E. Bamberger, R. G. Ross, and C. F. Baes, Jr., in Molten-Salt
Reactor Program Semionnual Progress Report, February 28, 1971,
ORNL-4676, p. 119.

C. E. Bamberger, R. G. Ross, and C. F. Baes, Jr., in Molten-Salt
Reactor Program Semiannual Progress Report, August 31, 1971, ORNL=
4622, p. 92.

C. F. Baes, Jr., in Molten-Salt Reactor Program Semiavmual Progress
Report, February 28, 1970, ORNL-4548, p. 152.

S. Cantor, Density and Viscosity of Several Molten Fluoride Mix-
tures, ORNL/TM-4308 (March 1973).

S. Cantor (Ed.), Physical Properties of Molten-Salt Reactor Fuel,
Coolant, and Flush Salts, ORNL/TM-2316 (August 1968).

J. W. Cooke, Development of the Variable-Gap Technique for Measuring
the Thermal Conductivity of Fluoride Salt Mixtures, ORNL-4831 (Febru-
ary 1973).

E. L. Compere et al., Figsion-Product Behavior in the Molten-Salt
Reactor Experiment, ORNL-4865 (October 1975).

A. P. Malinauskas and D. M. Richardson, "The Solubilities of Hydro-
gen, Deuterium, and Helium in Molten Li.BeF,," Ind. Eng. Chem. Fund.
13, 242 (1974).

W. R. Grimes, N. V. Smith, and G. M. Watson, J. Phys. Chem. 62, 862
(1958).

M. Blander et al., J. Phys. Chem. 63, 1164 (1959).

G. M. Watson et al., J. Chem. Eng. Data 7, 285 (1962).
32.

33.

34,

36.

37.

38.

39.

49,

41.

42.

43,

b4,

45.

46.

114

R. B. Briggs and R. B. Korsmeyer, in Molten-Salt Reactor Program g
Semiannual Progress Report, February 28, 1370, ORNL-4584, pp. 53—54.

See review of many determinations by C. F. Baes, Jr., "The Chemistry
and Thermodynamics of Molten-Salt Reactor Fuels," Nuclear Metallurgy,
Vol. 15, CONF-690801, U.S. Atomic Energy Commission, Division of
Technical Information, August 1969.

A. D. Kelmers et al., Evaluation of Alternate Secondary (and Terti- 5
ary) Coolants for the Molten-Salt Breeder FReactor, ORNL/TM-5325
(April 1976).

L. Brewer et al., "The Thermodynamic Properties of the Halides," in
The Chemistry and Metallurgy of Miscellaneous Materials: Thermo-
dynamics, pp. 76—192, L. L. Quill, ed., McGraw-Hill, New York, 1950.

J. F. Elliott and Molly Glieser, Thermochemisiry for Steelmaking,
Vol. 1, Addison-Wesley, Reading, Mass., 1960.

L. M. Toth and L. 0. Gilpatrick, The Equilibrium of Dilute UF;
Solutions Contained in Graphite, ORNL/TM-4056 (December 1972).

7

L. M. Toth and L. 0. Gilpatrick, J. Phys. Chem. 77, 2799 (1973).

R. A. Strehlow, in Molten-Salt Reactor Program Semiannual Progress
Report for Period Ending February 28, 1970, ORNL-4548, p. 167.

J. R. Keiser, Compatibility Studies of Potential Molten-Salt Breeder
Reactor Materials in Molten Fluoride Salts, ORNL/TM-5783 (May 1977).

J. R. Keiser, J. H. DeVan, and D. L. Manning, The Corrosion FKesis-
tance of Type 316 Stainless Steel to Li;BeF,, ORNL/TM-5782 (April
1977).

M. R. Bennett and A. D. Kelmers, in Molten-Salt Reactor Program
Semiannual Progress Report for Period Ending February 29, 1976,
ORNL-5132, pp. 27-136 (August 1976).

R. F. Apple, in Molten-Salt Reactor Program Semiannual Progress
Report for Period Ending February 29, 1976, ORNL-5132, pp. 36—37.

D. L. Manning and G. Mamantov, in Molten-Salt Reactor FProgram
Semiannual Progress Report for Period Ending Lugust 31, 1976, ORNL-
5078, pp. 4748 (February 1976).

D. L. Manning and G. Mamantov, in Molten-Salt Reactor Program
Semiannual Progress Report for Period Ending February 25, 1978,
ORNL-5132, pp. 3940 (August 1976).

D. L. Manning, W. K. Miller, and R. Rowan, Methods of Determination
of Uranium Trifluoride, ORNL-1279 (May 25, 1952).
115

S 47. ORNL Master Analytical Manual, Section 2, Radiochemical Methods,
TID-7015 (1857}.

48. "Automatic Continuous Analysis of Helium,'" GCR Quarterly Progress
Report, December 31, 1959, ORNL-2888, pp. 178—184.

49. A. S. Meyer, personal communication, July 12, 1972.

50. R. B. Gallaher, Operation of the Sampler-Enricher in the Molten-
Salt Reactoyr Experiment, ORNL/TM-3524 (October 1971).

51. Molten-Salt Reactor Frogram Semiannual Progress Report, February 28,
1966, ORNL-3936, p. 154.

52. G. Goldberg and L. T. Corbin, in Molten-Salt Reactor Program Semi-
annual Progress Report, August 31, 1968, ORNL-4344, pp. 198-196.

53. J. M. Dale, R. F. Apple, and A. S. Meyer, in Molten-Salt Reactor
Program Semiannual Progress Report, February 28, 1967, ORNL-4119,
p. 158.

54. J. P. Young, in Molten-Salt Reactor Program Semianviual Progress
Report, February 28, 1869, ORNL-4396, pp. 202—204.

55. J. M. Dale, R. F. Apple, and A. S. Meyer, in Molten-Salt Reactor
Program Semiannual Progress Report, February 28, 1970, ORNL-4548,
p. 183.

56. R. Blumberg and T. H. Manney, in Molten-Salt Reactor Program Semi-
annual Progress Report, August 31, 1968, ORNL-4344, p. 36.

57. A. Houtzeel and F. F. Dyer, Gamma Spectrometric Studies of Fission
Products im the MSRE, ORNL-3151 (August 1972).

58. W. R. Laing et al., in Molten-Salt Reactor Program Semiannual Prog-
ress Report, February 28, 1969, ORNL-4396, p. 208.

59. D. L. Manning and J. M. Dale, in Molten-Salt Reactor Program Semi-
annual Progress Report, August 31, 1968, ORNL-4344, p. 192.

60. D. L. Manning and G. Mamantov, '"Rapid Scan Voltammetry and Chrono-
poetentiometric Studies of Ironm in Molten Fluorides,"” J. Electroanal.
Chem. 7, 102—108 (1964).

61. D. L. Manning, F. R. Clayton, and G. Mamantov, in Molten-Salt Reactor
Program Semiannual Frogress Report, August 31, 1971, ORNL-4728, p. 75.

62. J. M. Dale and A. S. Meyer, in Molten-Salt Reactor Program Semiannual
Progress Report, Februavy 29, 1972, ORNL-4782, p. 77.

 
116

63. D. L. Manning, in Molten-Salt Reactor FProgram Semiannual Progress e
Report, February 28, 1971, ORNL-4676, p. 135.

64. J. M. Dale and A. S. Meyer, in Molten-Salt Reactor Program Semi-
annual Progress Report, February 29, 1972, ORNL-4782, p. 79.

65. H. W. Jenkins et al., in Molten-Salt Reactor Program Semiannual
Progress Report, February 28, 1969, ORNL-4396, p. 201.

66. M. T. Kelley, R. W. Stelzmer, and D. L. Manning, in Molten-Salt
Reactor Program Semiannual Progress Report, August 31, 1969, ORNL-
4449, p. 157.

67. D. L. Manning and A. S. Meyer, in Molten-S5alt Reactor Frogram Semi-
annual Progress Report, August 31, 1971, ORNL-4728, p. 74.

68. D. L. Manning and H. R. Bronstein, in Melten-Salt Keactor Program
Semtannual Progress Report, February 28, 1970, ORNL-4548, p. 184.

69. D. L. Manning and F. L. Clayton, in Molten-Salt Reactor Program
Semiannual Progress Report, August 1, 1970, ORNL-4622, p. 115.

70. J. P. Young and J. C. White, "A High-Temperature Cell Assembly for
Spectrophotometric Studies in Molten Fluorides,' Anal. Chem. 31,
1892 (1959).

71. J. P. Young, ''Windowless Spectrophotometric Cell for Use with Cor-
rosive Liquids," A4Anal. Chem. 36, 390 (1964).

72. Molten-Salt Reactor Program Semiannual Progress Report, Lugust 31,
1965, ORNL-3872, p. 145,

73. L. M. Toth, J. P. Young, and G. P. Smith, in Molten-Salt Reactor
Program Semiannual FProgress Report, August 31, 1968, ORNL-4344,
p. 168.

74. J. P. Young, in Molten-Salt Reactor Program Semiannual Progress
Report, August 31, 1969, ORNL-4449, p. 16l.

75. J. P. Young, in Molten-Salt Reactor Program Semiarnnual Progress
Report, August 31, 1871, ORNL-4728, p. 71.

76. J. P. Young, in Molten-Salt Reactor Program Semiannual Progress
Report, February 28, 1867, ORNL-4119, p. 163.

77. J. P. Young, in Molten-Salt Reactor Program Semiannual Progress
Report, August 31, 1966, ORNL-4037, p. 193.

78. J. P. Young, in Molten-Salt Reactor Program Semiannual Progress
Report, February 28, 1967, ORNL-4119, p. 1l64. .

79. John B. Bates et al., in Molten-Salt Reactor Frogram Semionnual 1
Progress Report, February 28, 1971, ORNL-4676, p. 94. o
 

 

80.

81.

82,

83.

84,

86.

87.

88.

89.

990.

91.

92.

93.

94.

95.

96.

117

J. P. Young et al., in Molten-Salt Reactor Program Semiavnual
Progress Report, August 31, 1971, ORNL-4728, p. 73.

J. P. Young, in Molten-Salt Reactor Program Semiannual Progress
Report, February 28, 1571, ORNL-4676, p. 136.

J. M. Dale et al., in Molten-Salt Reactor Program Semicnnual Prog-
ress Report, August 31, 1968, ORNL-4344, p. 188.

J. M. Dale et al., in Molten-Salt Reactor Program Semiannual FProg-
reggs Report, August 31, 1968, ORNL-4344, p. 189.

C. M., Boyd and A. S. Meyer, in Molten-Salt Reactor Progrom Semi-
annual Progress Report, August 31, 1967, ORNL-4191, p. 173.

J. C. White, in Molten-Salt Reactor Program Semiannual Progress
Report, July 31, 1964, ORNL-3708, p. 328.

R. F. Apple and A. S. Meyer, in Molten-Salt Reactor Program Semi-
annual Progress Report, February 28, 1969, ORNI-4396, p. 207.

R. F. Apple and A. S. Meyer, in Molten-Salt Reactor Program Semi-
annual Progress Report, Auguet 31, 1971, ORNL-4728, p. 72.

J. M. Dale and A. S. Meyer, in Molten-Salt Reactor Program Semi-
annual Progregs FReport, August 31, 1871, ORNL-4728, p. 69.

J. M. Dale and T. R. Mueller, in Molten-5alt Reactor Program Semi-
annual Progress Report, February 28, 1971, ORNL-4676, p. 138.

J. R. DiStefano, in Molten-Salt Reactor Program Semiannual Report
for Period Ending August 1, 1975, ORNL-5078, pp. 132—139.

J. R. DiStefano, in Molten-Salt Reactor Program Semiannual Report
for Period Ending February &9, 1976, ORNL-5132, p. 163.

Personal communication, H. E. McCoy, ORNL, to W. R. Grimes, ORNL.

A. P. Litman and A. E. Geoldman, Corrcsion Assceiated with Fluorina-
tion in the Oak Ridge National Laboratory Fluorvide Volatility Pro-
cess, ORKNL-2832 (June 5, 1961).

L. Hays, R. Breyne, and W. Seefeldt, "Comparative Tests of L Nickel,
D Nickel, Hastelloy B, and INOR-1," Chemical Engineering Divisiown
Summayy Report, July, August, September, 1968, ANL-5924, pp. 49-52.

M. R. Bennett, in Molten-Salt Reactor Program Semiannual Progress
Report for Period Ending August 31, 1971, ORNL-4728, p. 190.

M. R. Bennett and A. D. Kelmers, in Molten-Salt Reactor Program
Semiannual Progress Report for Period IEnding February 29, 1976,
ORNL-5132, p. 170.
97.

98.

99.

100,

101.

103.

104,

107.

108.

109.

110.

111.

118

R. M. Counce, in Molten-Salt Reactor Program Semiannual Progress s
Report for Period Ending August 31, 1975, ORNL-5078, p. 157.

D. H. Gurinsky et al., "LMFR Materials,'" Chap. 21, p. 743, in
Flutd Fuel Reactors, J. A. Lance, H. G. McPherson, and F. Maslam,
Eds., Addison-~Weslev, Reading, Mass., 1958, .

C. H. Brown, Jr., J. R. Hightower, Jr., and J. A. Klein, Measurement

of Mass Transfer Coefficients in a Mechanically Agitated, Nowndis- -
persing Contactor Operating with a Molten Mixture of LiF-BeF,-ThF,

and Mollen Bismuth, ORNL-5143 {(November 1976).

H. C. Savage and J. R. Hightower, Jr., Engineering Tests of Metal
Transfer Process for Extraction of Rare-Earth Fission Products from
a Molten-Salt Breeder Reactor Fuel Salt, ORNL-5176 (February 1977).

H. Shimotake, N. R. Stalica, and J. C. Hesson, "'Corrosion of Re-
fractory Metals by Liquid Bismuth, Tin, and Lead at 1000°C," Trans.
ANS 10, 141142 (June 1967).

J. W. Siefert and A. L. Lower, Jr., "Evaluation of Tantalum, Molyb-
denum, and Berylilium for Liquid Bismuth Service," Corrosion 17(10),

475¢—78t (October 1961).

J. R. DiStefano and A. J. Moorhead, Development and Construction of
a Molybdenum Test Stand, ORNL-4874 (December 1972).

B. Fleischer, in Metals and Ceramics Divisiown Awnual Progress Report,
June 30, 1870, ORNL-4570, pp. 1034,

0. B. Cavin et al., in Molten-Salt Reactor Program Semiannual Prog-
ress Report for Period Ending February 29, 1972, ORNL-4782, p. 198.

R. G. Donnelly and G. M. Slaughter, "The Brazing of Graphite,"
Welding J. 41(5), 46169 (1962).

J. P. Hammond and G. M. Slaughter, "Bonding Graphite to Metals with
Transition Pieces,' Welding J. 50(1), 33-40 (1%970).

W. J. Hallett and T. A. Coultas, Dynamic Corrosion of Graphite by
Liquid Bismuth, NAA-SR-188 (Sept. 22, 1952).

A. L. Lowe, Jr. (Compiler), Liquid Metal Fuel Reactor FExperiment
Graphite Evaluation Program, BAW-1197 (June 1960).

R. B. Lindauer, in Molten-Salt Reactor Program Semiannual Progress
Report for Period Ending February 28, 1975, ORNL-5047, p. 162.

R. B. Lindauer, in Molten-Salt Reactor Program Semiannual Progress
Report for Period Ewnding February 29, 1976, ORNL-5132, p. 184.

S
 

 

112,

113.

114.

115.

116.

117.

118.

119.

120.

123.

122,

123.

124,

125,

126.

127.

119

M. R. Bemnett and A. D. Kelmers, in Molten-Salt Reactor Program
Semiannual Progress Report for Period FEnding February 28, 1975,
ORNL=-5047, p. 150.

M. R. Bennett and A. D. Kelmers, in Molten-Salt Reactor Program
Semiarmual Progresg Report for Period Ending February 28, 1976,
ORNL-5132, p. 170.

R. M. Counce, in Molten-Salt Reactor Program Semiannual Frogress
Report for Period Ending February 59, 1576, ORNL-5132, p. 189.

C. BH. Brown, Jr., and J. R. Hightower, Jr., in Molten-Salt PReactor
Progpram Semiannual Frogress Report for Period Ending February 29,
1978, ORNL-5132, p. 177.

C. W. Kee and L. E. NcNeese, MPRR — Multiregion Processing Plant
Code, ORNL/TM-4210 (September 1976).

W. B. Carr, "Volatility Processing of the ARE Fuel," Chem. ing.
Prog. Symp. Ser. 56, 57 (1860).

Chemical Technology Division Annual Progress Report, May 31, 1964,
ORNL-3627, pp. 29—35.

W. H. Carr et al., Molten=Salt Volatility Pilot Plant: Recovery
of Enriched Uranivum from Aluminum-Clad Fuel Elements, ORNL-4574
(April 1971).

R. B. Lindauer, Processing of the MSRE Flush and Fuel Salts, ORNL/
TM-2578 (August 1969).

M. R. Bennett and L. M. Ferris, J. Inorg. Nucl. Chem. 36, 1285 (1574).
L. M. Ferris et al., J. Tnorg. Nucl. Chem. 34, 31320 (1972).

L. E. McNeese, Engineering Development Studies for Mclten-Salt
Breeder Reactor Processing No. 8, ORNL/TM-3258 (May 1972).

W. L. Carter and E. L. Nicholson, Design and Cost Study of a
Fluorination—-Reductive Extraction-Metal Transfer Process Plant
for the MSBE, ORNL/TM-3579 (May 1972).

L. E. McNeese, IEngineering Development Studies for Molien-Salt
Breeder Reactor Processing No. 6, ORNL/TM-3141 (December 1971).

F. A. Doss, W. R. Grimes, and J. H. Shaffer, in Molten-Salt Reactor
Program Semiannual Progress Report, August 31, 1970, ORNL-4622,
p. 106.

Chemical Technology Division innual Progress Report, May 31, 1967,
ORNL-4145, pp. 95-97.
128.

129.

131.

132.

133,

},_a
(%
I~

135,

136,

138.

120

J. S. Watson and L. E. McNeese, Engineering Development Studies
for Molten-Salt Breeder Reactor Processing No. 9, CRNL/TM-3259,
p. 52.

B. A. Hannaford, W. M. Woods, and D. D. Sood, in Molten-Salt Reac-
tor Program Se#zanmual Progress Report, Februavy 29, 1972, ORNL-
4782, pp. 227-30.

L. E. McNeese, Engineering Development Studies for Molten-Salt
Breedeyr Reactor Frocessging No. 5, ORNL/TM-3140 (Cctober 1971).

\"-

4 ng Development Studies for Molten-Salt
Breeder Reactor Frocessing No. 7, ORNL/TM-3257 (February 1972).

L. E. McNeese, Engineer

J. S. Watson and L. E. McNeese, "Axial Dispersion in Packed Columns
During Countercurrent Flew, Liquids of High Density Difference,'
Ind. Eng. Chem. Process Des. Develop. 11, 120-21 (1972).

J. S. Watson and H. D. Cochran, "A Simple Method for Estimating
the Effect of Axial Backmixing on Countercurrent Column Performance,'
ind. Eng. Chem. Process Deg. Develop. 10, 83-85 (1971).

L. Brewer, Science 161, 115 (1968).

D. M. Moulton et al., in Molten-Salt Reactor Program Semiarmual
Progress Report for Period Ending August 31, 1968, ORNL-4449, p. 151.
el et al., Molten-Salt Reactors for Efficient Nuclear Fuel
on Without Plutonium Separation, ORNL/TM-6413 (August 1978).

R. G. Ross, C. E. Bamberger, and C. F. Baes, Jr. in Mcolten-Salt
Reaclor Program Semiannucl FProgress Report, August 31, 1970, ORNL-
4622, p. 92,

M. J. Bell, D. D. Sood, and L. E. McNeese, in Molten-Salt Reactor
Program Semiannual Progress Report, February 29, 1972, ORNL-4782,
PpP. 234—7

 

 
 

=3

 

PART IV. REACTOR MATERIALS

H. E. McCoy

The material for the primary circult will be exposed at temperatures
up to 700°C to fuel salt containing fission products and to irradiation
by primarily thermal neutrons. A nickel-base alloy, Hastelloy N, has
been demonstrated to be reasonably serviceable under these conditions,
but it was embrittled by irradiation and suffered shallow intergranular
embrittlement by the fission product tellurium. There is considerable
experimental evidence that small modifications to the chemical composi-
tion of Hastelloy N results in improved resistance to neutron and fis-
sion-product embrittlement, and the materials program described in this
plan is directed toward developing and commercializing a modified com-
positicn of Hastelloy N with impreved properties.

The graphite moderator for an MSR must be capable of withstanding
neutron fluences of at least 3 X 10%? neutrons/cm®. Commercial graphites
exist which are likely to meet this goal, but further testing will be re-
guired to fully characterize these materials. There is also considerable
evidence that graphite with improved dimensional stability can be de-
veleped. Methods for manufacturing these impreoved materials must be
developed and the products irradiated and characterized. The improved

materials must be scaled up by a commercial wvendor.
 

 
 

 

123
7. STRUCTURAL METAL FOR PRIMARY AND SECONDARY CIRCUITS

The material used in constructing the primary circuit of an MSR
will operate at temperatures up to 700°C. The inside of the circuit
will be exposed to salt containing fission preducts and will receive a
maximum thermal fluence of about 1 X 10%1 neutrons/cm2 over the operat-
ing lifetime of about 30 years. This fluence will cause embrittlement
due to helium formed by transmutation but will not cause swelling such
as is noted at higher fast fluences, The outside of the primary circuit
will be exposed to nitrogen containing sufficient gir from inleakage to
make it oxidizing to the metal. Thus the metal must have moderate oxi-
dation resistance, must resist corrosion by the salt, and must not be
subject to severe embrittlement by thermal neutrons.

In the secondary circuit the metal will be exposed to the coclant
salt under much the same conditions described for the primary circuit.
The main difference will be the absence of fission preoducts and uranium
in the coolant salt and the much lower neutron fluences. This material
must have moderate oxidation resistance and must resist cerrosion by a
salt not ccntaining fission products or uranium. |

The primary and secondary circuits involve numerous structural shapes
ranging from a few inches thick to tubing having wall thicknesses of only
a few thousandths of an inch. These shapes must be fabricated and joined,
primarily by welding, into an integral engineering structure. The struc-
ture must be designed and built by techniques approved by the ASME Boiler

and Pressure Vessel Code.

Status in 1972

Early materials studies led to the development of a nickel-base
alloy, Hastelloy N, for use with fluoride salts. As shown in Table 7.1,
the alloy contained 167 molybdenum for strengthening and chromium suf-
ficient to impart moderate oxidation resistance in air but not enough to
lead to high corrosion rates in salt. This alloy was the sole structural
material used in the MSRE and contributed significantly to the success

of the experiment. However, two problems were noted with Hastelloy N
124

Table 7.1. Chemical composition of Hastelloy N

 

Content (% by weight)a

 

 

Element e o
Standard alloy Medlfiggzalloy, Mod1f13§6alloy,

Nickel Basge Base Base
Molybdenum 15~18 11-13 11-13
Chromium 6—8 6—8b 6—8b
Iron 5 .1 b 0.1 5
Manganese 1 ¢.15-0.25 0.15-0.25
Siiicon 1 C.1 0.1
Phosphorus 3.015 C.C1 0.01
Sulfur 0.020 C.0L 0.0L
Boron 0.01 C.C01 0.00L
Titanium 2
Niobium 02 1—2

 

& ) .
Single values are maximum amcunts allowed. The actual
concentrations of these elements in an alioy can be much lower.

These elements are not felt to be very important. Alloys
are now being purchased with the small concentrations specified,
but the specification may be changed in the future tc allow a
higher concentration.

which needed further attention before more advanced reactors could be
built. First, it was found that Hastelloy N was embrittled by helium
produced from '°B and directly from nickel by a twc-step reaction. This
type of radiation embrittlement is commen to most iron- and nickel-base
alloys. The second problem arose from the fission-product tellurium dif-
fusing a short distance into the metal along the grain boundaries and
embrittling the boundaries.

When cur studies were terminated in early 1973, considerable prog-
ress had been made in finding sclutions to both problems. Since the two
problems were discovered a few years apart, the research on the two prob-
lem areas appears to have proceeded independently. However, the work
must be breught together for the production of a single material that
would be resistant to both problems. It was found that the carbide pre- .
cipitate that normally occcurs in Hastelloy N could be modified to obtain
resistance to the embrittlement by helium. The presence of 167 molybde-

num and 0.5% silicon led to the formation of a8 coarse carbide that was
 

 

125

of little benefit. Reduction of the molybdenum concentration teo 127 and
the silicon content to 0.17 and the addition of a reactive carbide former
such as titanium led to the formation of a fine carbide precipitate and
an alloy with good resistance to embrittlement by helium. The desired
level of titanium was about 2%, and the phenomenon had been checked out
through numercus small laboratory and commercial melts by 1972.

Because the intergranular embrittlement of Hastellcy N by tellurium
was noted in 1970, our understanding of the phenomenon was not wvery ad-
vanced at the conclusion of the program in 1973. Numerous parts of the
MSRE were examined, and all surfaces exposed to fuel salt formed shallow
intergranular cracks when strained. Some laboratory experiments had
been performed in which Hastelloy N specimens had been exposed to low
partial pressures of tellurium metal wvapor and, when strained, formed
intergranular cracks very similar to those noted in parts from the MSRE.
Several findings indicated that tellurium was the likely cause of the
intergranular embrittlement, and the selective diffusion of tellurium
along the grain boundaries of Hastelloy N was demonstrated experimentally.
One in-reactor fuel capsule was operated in which the grain boundaries of
Hastelloy N were embrittled and those of Inconel 601 (Ni, 227% Cr, 12% Fe)
were not. These findings were in agreement with laboratory experiments
in which these same metals were exposed to low partial pressures of tel-
lurium metal vapor. Thus, at the close of the program in early 1973,
tellurium had been identified as the likely cause of the intergranular
embrittlement, and several laboratory and in-reactor methods were de-
vised for studying the phenomenon. Experimental results had been ob-
tained which showed variations in sensitivity to embrittlement of
various metals and offered encouragement that a structural material
could be found which resisted embrittlement by tellurium.

The alloy composition favored at the close of the program in 1973
is given in Table 7.1 with the composition of standard Hastelloy N. The
reasoning at that time was that the 27 titanium addition would impart
good resistance to irradiation embrittlement and that the 0 to 27% nio-
bium addition would impart good resistance to intergranular tellurium
embrittlement. Neither of these chemical additions was expected to

cause problems with respect to fabrication.
126
Status in 1976 ,\",

When the program was restarted in 1974, top priority was given to
the tellurium-embrittlement problem. A small piece of Hastelloy N foil
from the MSRE had been preserved for further study. The foil was broken
inside an Auger spectrometer and the fresh surface analyzed. Tellurium
was found in abundance, and no other fission product was present in de-
tectable quantities. This showed even more positively that tellurium
was responsible for the embrittlement.

Considerable effort was spent in seeking better methods of exposing
test specimens to tellurium. In the MSRE the flux of the tellurium atoms

reaching the metal was 10° atoms cm™? sec™ and this value would be 10!°

atoms cm™° sec”! for a high-performance breeder. Even the value for a
high-performance breeder is very small from the experimental standpoint.
For example, this flux would result in a total of 7.6 X% 10”8 g of
tellurium transferred to a sample having a surface area of 10 em? in
1000 hry. Electrochemical probes were immersed directly in salt melts
known to contain tellurium, and there was never any evidence of a soluble
telluride species. However, there was considerable evidence that tel-
lurium "moved" through salt from cne point to another in a salt system. -
It was hypothesized that the tellurium actually moved as a low-pressure,
pure-metal vapor and not as a reacted species. The most representative
experimental system developed for exposing metal specimens to tellurium
involved suspending the specimens in a stirred vessel of salt with gran-
ules of CriTey and CrsTeg lying on the bottom of the salt. Tellurium,
at a very low partial pressure, was in equilibrium with the CriTe, and
CrsTeg, and exposure of Hastelloy N specimens to this mixture resulted
in crack severities similar to those noted in samples from the MSRE,
Numerous samples were exposed to salt containing tellurium, and
the most important finding was that modified Hastelloy N containing 1
to 2% nicbium had good resistance to embrittlement by tellurium (Fig.
7.1). An almost equally important finding was that the presence of ti-
tanium negated the beneficial effects of niobium. Thus, an alloy con-

taining titanium, to impart resistance to irradiation embrittlement, and

niobium, to impart resistance to tellurium embrittlement, did not have
127

ORNL-DWG 77-922
I Y \«’?
8000 |— 8926 -
, 2500 hr
/
7000 +— / ]

6000

 

 

 

5000

CRACK DEPTH
fem)

4C00

 

CRACK FREQUENCY x
{ number /cm)
o
O
O
O

2000

1000

 

 

 

 

 

Nb CONTENT (%)

Fig. 7.1. Variations of severity of cracking with niobium content.
Samples were exposed for indicated times to salt containing Cr3Te, and
CrsTeg at 700°C.
128

acceptable resistance to tellurium embrittlement, even though the mech- e
anical properties in the irradiated condition were excellent. As a re-

sult, it became necessary to determine whether alloys containing niobium

(without titanium) had adequate resistance to irradiation embrittlement.

There was time only to obtain the alloys and run one irradiation experi-

ment, but the results loocked very promising. An alloy containing 2% .
niobium and irradiated at 704°C was about 30% stronger than standard

Hastelley N and had a fracture strain of about 3% compared with <1% for

standard Hastelloy N. Even though alloys modified sclely with nicbium

do not have as good postirradiation properties as alloys modified with

titanium or titanium plus niobium, their properties are probably adequate.

The niobium-modified alloys were not made in melts larger than 50
1b, but no problems were encountered in this size with niobium concen-
trations up to and including 4.4%. Test welds made in the 1/2-in.-thick
plate passed the bend and tensile tests required by the ASME Boiler and
Pressure Vessel Code. From the chemical analysis of the niobium-modified
alloy, no scaleup problems are anticipated.

One series of experiments was carried out to investigate the ef-
fects of oxidation state on the tendency for cracks to be formed in
tellurium-containing salt, on the supposition that the salt might be
made reducing enough to tie the teilurium up in some innocuous metal
complex. The salt was made more oxidizing by adding NiF,; and more re-
ducing by adding beryliium. The experiment had electrochemical probes
for determining the ratio of uranium in the +4 state (UF4) to that in
the +3 state (UF3). Tensile specimens of standard Hastelloy K were
suspended in the salt for about 260 hr at 700°C. The oxidation state
of the salt was stabilized, and the specimens were inserted so that each
set of specimens was exposed to one condition. After exposure, the spec-
imens were strained to failure and were examined metallographically to
determine the extent of cracking. The results of measurements at several
oxidation states are shown in Fig. 7.2. At U*T/U3* ratios of 60 or less,
there was very little cracking, and at ratios above 80 the cracking was
very extensive. These observations offer encouragement that a reactor
could be operated in a chemical regime where the tellurium would not be

embrittling even to standard Hastelloy N. At least 1.6% of the uranium
#
%

 

 

129

ORNL-DWG 77-4680A

 

I I ! I [ r
900 - RECUCING OXIDIZING J
e —_— :

/f”"“fi‘°—4

CRACKING 600 |- :
PARAMETER _

[FREQUENCY (cm™%)

X AVG. DEPTH (um)] — fi

300 -

 

@
0 "-9—'—""“‘"1—"‘“‘7/ Lo | o
i0 20 40 70 100 200 400
SALT OXIDATION POTENTIAL fU(IVI/U(IID)]

Fig. 7.2. Cracking behavior of Hastelloy N exposed 260 hr at 700°C
to MSBR fuel salt containing CrsTe, and CrsTeg.

would need to be in the +3 oxidation state (UF:), and this condition seems
quite reasonable from chemical and practical considerations.

One further accomplishment during the period 1974—76 was the use of
available data to predict the helium yvield from interaction of nickel
with thermal neutrons. It has been known for some time that iron- and
nickel-base alloys can be embrittled in a thermal neutren flux by the

transmutations of "tramp" *°B to helium and lithium. This process gen-

erally results in the transmutation of most of the 10g by fluences of
thermal neutrons on the order of 10°%°/cm® and usually yields from 1 to
10 at. ppm of helium. With nickel there is a further thermal two-step

transmutation involving these reactions:
5 *
SNi 4+ m > PINi
Ige t
“9Ni + n > “He + °%Fe .

This sequence of reactions does not saturate, and although the cross
sections are still in question, it would produce a maximum of 40 at. ppm
of helium in the vessel over a 30-year MSBR lifetime. This is not an
unreasonable amount of helium to accommodate in the type of microstruc-

ture being developed in modified Hastelloy N.
130
Current Status

At the close of the program in 1976 (and at the present time), the
third alloy composition shown in Table 7.1 was favored. Considerable
progress had been made in establishing test methods for evaluating a ma-
terial's resistance to embrittlement by tellurium. Modified Hastelloy N
containing from 1 to 37 niobium was found to offer improved resistance to
embrittlement by tellurium, but the test conditions were not sufficiently
long or diversified to show that the alloy totally resists embrittlement.
One irradiation experiment showed that the niobium-modified ailoy offered
adequate resistance to irradiation embrittlement, but more detailed tests
are needed. Several small melts containing up to 4.4% niobium were found
to fabricate and weld well; so products containing 1 to 2% niobium can

probably be produced with a minimum of scale-up difficulties.

Technology Needs and Development Plan

 

The coverall development needs were described previously, but the
new findings shift the emphasis from alloys modified with titanium and
rare earths to those modified with niobium. The specific techneclogy
needs are identified in Table 7.2, along with a potential schedule for
their development. The first task will involve irradiation, corrosion,
tellurium exposure, mechanical property, and fabrication tests to final-
ize the composition for scale-up. The techniques for doing most of these
tests have already been established.

The second task will inveclve procuring large commercial heats of the
reference alloy. The material would be procured in structural shapes
ranging from plate to thin-wall tubing, typical of the products to be
used in a reactor. The third task consists in evaluating these materi-
als by mechanical property and cerrosion tests of at least 10,000-hr
duration. The two main purpeses of these tests would be to confirm the
adequacy of the new alloy for reactor applications and to gather the
data needed for reactor design. The fourth task would be to develop
the design methods and rules needed to design a reactor tc be built of
the modified Hastelloy N. This task will have already been partially

completed by ASME Pressure Vessel Code work currently in progress. The

 
 

Table 7.2.

Schedule for development of structural metal for primary and secondary circuits

 

Fiscal vyear

 

 

 

 

 

 

 

 

Task
1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991
Determination of alloy compesition ..iECEfiE[EiEZ5
67
Procurement of commercial heats V'V
Evaluation of commercial heats ve vs v10 v”
Development of analytical design 12V ‘{713 V14
methods — ASME Code
vflS
Long-term material tests
16
Alloy optimization v
Milestones:
1. Receipt of small commercial heats containing 1 to 8. Begin comstruction and checkout of equipment re-
2% Nb. Begin mechanical property and compatibil- quired for mechanical property tests on four large
ity tests on heats. heats.
2. Receipt of products of 10,000-1b heat of 2% Nb- 9. Begin evaluation of four large heats by weldability,
modified llastelloy N. Begin mechanical property mechanical property, and compatibility tests.
and compatibility tests on 10,000-1b heat. 10. Begin operation of forced-circulation loops (FCL-6
3. Start forced-convection corrosion loop constructed and 7) constructed of modified alloy and circulat-
of 10,000~1b heat for basic fuel salt corrosion ing fuel salt. ‘
studies. Begin “Vl-vear irradiation of fuel 11. Begin operation of forced-circulation loops (FCL-8
Pins made of most desirable alloy. and 9) constructed of modified alloy and circulating
4. Start forced-convection corrosion loop constructed coolant salt.
of 10,000~1b heat for fuel salt-Te corrosion stud- 12. Begin detailed analysis of mechanical property data.
ies. 13. Begin development of design methods for modified
5. Start forced-convection corrosion loop (FCL-5) con- alloy.
structed of 10,000-1b heat for coolant salt corrosion 14. Submit data package for ASME Code Approval.
studies. 15. Begin studies tco raise allowable temperature for use
6. Prepare specifications and solicit bids from poten- of modified alloy.
tial vendors for four heats of desired composition. 16. Begin long-term mechanical property and compatibil-
7. Begin receipt of products from four large heats. ity tests on modified alloy.

 

1¢T
132

final product of this task would be inclusion of modified Hastelloy N
into the high-temperature Code.

Although the data gathered in the third task (tests of 10,000-hr
duration) will probably be adequate for Code approval, it will be de-
sirable to continue some of the mechanical property and corrosion tests
for longer times. The continuation of these tests in the fifth task will &
improve confidence in design rules and will allew last-minute changes in
reactor operating parameters if necessary.

Although the work in the first five tasks should result in an al-
loy adequate for construction of MSRs, it is likely that further alloy
development would lead to materials having improved characteristics
which may allow a higher reactor-outlet salt temperature or significant
relaxation of design and operating constraints. It is this further al-
loy optimization which will comprise the sixth task.

The operating and capital costs for these activities are summarized

in Tables 7.3 and 7.4, respectively.
 

Table 7.3.

Operating fund requirements for development of structural
metal for primary and secondary circuits

Cost (thousands of 1978 dollars) for fiscal year —

 

 

 

 

 

 

Fask 1980 1981 1982 1983 1984 1885 1986 1987 1988 1989 1990 1691
Determination of alloy composition 2200 2200
Procurement of commercial heats 520
Evaluation of commercial heats €00 2225 2660
Development of analytical design 280 930 220
methods — ASME Code Case Sub-
mission
Long=term material tests 1300 1300 1040 1040 910 2910 650 400
Alloy optimization 390 455 572 520 624 650 676 400
Total funds® 2200 2800 3025 3590 1910 1755 1612 1560 1534 1560 1326 800

 

aTotal funds through 1991:

$23,672.

tetl
 

Table 7.4. Summary of capital equipment funds required for development
of structural metal for primary and secondary circuits

 

Cost {(thousands of 1978 dollars) for fiscal year --

 

 

 

Task i . S
1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991
Determination of allovy composition 955 732
Procurement of commercial heats 13
Evaluation of commercial heats 438 1424 377
Development of analytical design 65 130
methods — Code Case Submission

Long-term material tests 52 78 46 72 39 52 39 40
Alloy optimization 46 91 104 104 98 98 98 40

Total funds™ 955 1170 1502 507 98 169 150 176 137 150 137 80

“Total funds through 1991: $523L.

el
135

 

8. GRAPHITE FOR MOLTEN-SALT REACTORS

The graphite in a single-fluid MSR serves no structural purpose®
cther than to define the flow patterns of the salt and, of course, to
! support its own weight. The requirements on the material are dictated
most strongly by nuclear considerations, namely stability of the mate-
rial against radiation-induced distortion and nonpenetrability by the
fuel-bearing molten salt. The practical limitations of meeting these
requirements, in turn, impose conditions on the core design, specifi-
cally the necessity to limit the cross-~sectional area of the graphite
prisms. The requirements of purity and impermeability te salt are
easily met by several high-quality, fine-grained graphites, and the
main problems arise from the requirement of stability against radiation-

induced distortion.

Status in 1972

By the time the MSBR Program was cancelled in early 1973, the di-
mensional changes of graphite during irradiation had been studied for
a number of years. These changes depend largely on the degree of crys-
talline isotropy, but the volume changes fall into a rather consistent
pattern. As shown in Fig. 8.1, there is first a period of densification
during which the volume decreases and then a period of swelling in which
the volume increases. The first period is of concern only because of the
dimensional changes that occur, and the second period is of concern be-
cause of the dimensional changes and the formation of cracks. The forma-
tion of cracks would eventually allow salt to penetrate the graphite,
The data shown in Fig. 8.1 are for 715°C, and the damage rate increases
with increasing temperature. Thus the graphite section size should be
kept small enough to prevent temperatures in the graphite from exceeding
those in the salt by a wide margin.

In the breeder concept the neutron flux is sufficiently high in the

[

central region of the core to require that the graphite be replaced about

every four years. It was further required that the graphite be surface

 

afa
5

Its primary function is, of course, to provide neutron moderation.
136

ORNL-DWG 71-6910R g

 

 

 

o

O

 

 

 

VOLUME CHANGE, 100 In (1+AV/V)

|
N

 

 

 

 

 

 

 

-8 . j | N ]
0 10 20 (x102%

FLUENCE [neutrons/cm? (£ >50 keV)]

Fig. 8.1. Volume changes for conventional graphites irradiated at
715°C.

sealed to prevent penetraticn of xenon intoc the graphite. Since re-
placement of the graphite would require considerable downtime, there
was strong incentive to increase the fluence limit of the graphite. A
considerable part of the ORNL graphite program was spent in irradiating

commercial graphites and samples of special graphites with potentially

improved irradiation resistance. The approach taken te sealing the

graphite was surface sealing with pyrocarbon. Because of the neutronic

 
137

requirements, other substances could not be introduced in sufficient

 

quantity to seal the surface.

The irradiation studies with several grades of graphite revealed
that the so-called binderless graphites, e.g., POCCO AXF, had improved
dimensional stability over most of the conventional graphites (Fig. 8.2).
The POCC graphites are presently available only in small sections, but
the GLCC B-364 grade is available in large sectiomns. The GLCC H-364 grade
has almost as high an allowable fluence as POCO AXF. Further work on
several special grades of graphite made at ORNL showed that graphites
could be developed with fluence limits even greater than those of the
POCO grades.
| The pyrolytic sealing work was only partially successful. It was
found that extreme care had tc be taken to seal the material before ir-
radiation. During irradiation the injected pyrocarbon actually caused
expansion to begin at lower fluences than those at which it would occur

in the absence of the coating. Thus the coating task was faced with a

number of challenges.

ORNL-DWG 71-6915R2

T

“J

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

© 4 ; A B}
N ;
3 : ,,//H—364 ,
1 5 o d
-~ H A /
£ 0 _5’3 ; Y -
i Ry
S A-A—&:—’
I a AXF-UFG ]
o AXM
A P-03
4 ! o HL 18
: i + H-395
| s
-8 | i
0 10 20 30 (x1c2h

FLUENCE [neutrons/tm?2 (£ >50 keV)]

Fig. 8.2. Volume changes for monolithic graphite irradiated at 715°C.

 
138
Status in 1976 R

No work was undertaken on graphite during the last segment of the

program. Thus the status in 1976 was the same as that in 1972.

Current Status

With the relaxed requirements® for breeding performance in nonpro-
liferating MSRs relative to the MSBR, the requirements for the graphite
have diminished. First, the peak neutron flux in the core can be re-
duced to levels such that the graphite will last for the lifetime of
the reactor plant. Secondly, the salt flow rate through the core is
reduced from the turbulent regime, and the salt film at the graphite
surface may offer sufficient registance to xenon diffusicon sc that it
will net be necessary to seal the graphite. The lessened gas permea-
bility requirements also mean that the graphite damage limits can be
raised (Figs. 8.1 and 8.2). The lifetime criterion adopted for the
breeder was that the allowable fluence would be about 3 % 10?% neutrons/
em’. This was estimated to be the fluence at which the structure in
advanced graphites would contain sufficient cracks to be permeable to g
xenon. Experience has shown that even at volume changes of about 10%
the graphite is not cracked but is uniformly dilated. For nomprolifer-
ating devices where xenon permeability will not be of concern, the limit
will be established by the formation of cracks sufficiently large for
salt intrusion. It is likely that current technology graphites like
GLCC+ E-364 could be used to 3 X 10°? neutrons/cm® and that improved

graphites with a limit of 4 X 10°?? neutrons/cm? could be developed.

Further Technology Needs and Development Plan

 

The near-term goal of the future development program (see Table 8.1)

will be to evaluate current commercial graphites for MSR use (Task 1).

 

%
These are manifested as lower core power density and higher fissile

specific inventory in denatured MSRs. s

Great Lakes Carbon Company.
 

 

&

 

 

 

 

 

 

Table 8.1. Schedule for graphite development
Fiscal vear
Task
1980 1981 1982 1983 1984 1685 1986 1987 1988 1989 1990 1991 1992 1993 1994
. . 1 2
Evaluation of commercial v v
graphites
. 3
Development of improved
graphites
Procurement of commercial §74
lots of improved
graphites
Evaluation of commercial 75

lots of improved
graphites

 

 

Milestones:

Establish program for development of improved graphites.
Define variables to be investigated.
Complete evaluation of commercial graphites.
document specification.

Develop procurement specification for improved commercial
graphites.

Prepare

4,

Begin procurement of production lots of improved commer-
cial graphites.

Begin long-term evaluation of improved commercial graph-
ites. Evaluation to include mechanical and physical
properties before and after irradiation.

6¢cT
This will involve irradiation of promising commercial graphites with
subsequent measurements of dimensional stability and thermal and elec-
trical conductivity.

A longer—range goal will be the development of a graphite with an
improved fluence I1imit. Efforts to date show that graphites can be
tailored to have improved dimensional stability. In Task 2 this work
will be continued te obtain several improved products, which will be
irradiated and evaluated. The technology for making the most desirable
products will be passed on to commercial vendors, and large lots of
these graphites will be obtained (Task 3). The commercial graphites
will be irradiated to high fluences, and the changes in dimensions, pore
spectra, thermal conductivity, thermal expansion, and electrical conduc~
tivity will be measured (Task 4).

The operating and capital equipment costs fer this work are sum-

marized in Tables 8.2 and 8.3, respectively.

 
 

 

3

 

 

 

 

 

 

 

 

Table 8.2. Operating fund requirements for graphite development
Cost (in thousands of 1978 dollars) for fiscal vear —
Task -
1980 1981 1982 1983 1984 1985 1986 1987 1983 1989 1990 1991 1992 1993 1994
Evaluation of c¢ommercial 300 300 300 300 300
graphites
Development of improved 159 300 300 500 500 150
graphites
Procurement of commercial 100 200 150
lots of improved
graphite
Evaluation of commercial 300 500 500 400 400 300 300 300
lots of improved
graphites
Total funds® 300 306 450 600 600 500 600 650 550 500 400 400 300 300

 

“Total funds through 1994: $6750.

300

7t
 

 

 

 

 

 

 

 

 

 

Table 8.3. <Capital equipment fund requirements for graphite development
Cost (in thousands of 1978 dollars) for fiscal year —
Task
1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
Evaluation of commercial 100 75 50 50 50
graphites
Develeopment of improved 30 100 100 10Q 100 50
graphites
Procurement of commercial lots
of improved graphites
Evaluation of commercial lots 50 100 75 75 75 50 50 50
of improved graphites
Total funds® 100 75 100 150 150 100 100 100 100 75 75 75 50 50 50
%Total funds through 1994: $1350.

<1
143

ORNL/TM-6415
Dist. Category UC~76

 

Internal Distribution

 

1. T. D. Anderson 45. R. E. MacPherson
) 2. Seymour Baron 46-50. H. E. McCoy
T 3. D. E. Bartine 51. L. E. McNeese
4-8, H. F. Bauman 52, H. Postma
9. H. W. Bertini 53-57. W. A. Rhoades
10. E. S. Bettis 58. P. S. Rohwer
11. H. I. Bowers 59. M. W. Rosenthal
12. J. C. Cleveland 60. R. T. Santoro
13. T. E. Cole 61. Dunlap Scott
14, §S. Cantor 62. M. R. Sheldon
15-19. J. F. Dearing 63. R. L. Shoup
20-29. J. R. Engel 64. M. J. Skinner
30. D. E. Fergusocn 65. 1. Spiewak
31. M. H. Fontana 66. H. E. Trammell
32-36. W. R. Grimes 67. D. B. Trauger
37. R. H. Guymon 68. J. R. Weir
38. W. 0. Harms 69. J. E. Vath
39. J. F. Harvey 70. R. G. Wymer
40. H. W. Hoffman 71-72. Central Research Library
41. J. D. Jenkins 73. Document Reference Section
42, P. R, Kasten 74-76. Laboratory Records
43, Milton Levenson 77. Laboratory Records (RC)
i 44, R. S. Lowrie

External Distribution

 

/8. Director, Nuclear Alternative Systems Assessment Division, De-
partment of Energy, Washington, D.C. 205345
79. E. G. Delaney, Nuclear Alternative Systems Assessment Division,
Department of Energy, Washington, D.C. 20545
80. C. Sege, Nuclear Alternative Systems Assessment Division, Depart-
ment of Energy, Washington, D.C. 20545
81. §S. Strauch, Nuclear Alternative Systems Assessment Division, De-
partment of Energy, Washington, D.C. 20545
82. K. A. Trickett, Office of the Director, Division of Nuclear Power
Development, Department of Energy, Washington, D.C. 20545
83. H. Kendrick, Nuclear Alternative Systems Assessment Division, De-
partment of Energy, Washington, D.C. 20545
84. NASAP Control Office, Department of Energy, Washington, D.C. 20545
85. Assistant Manager, Energy Research and Development, DOE-CRO
86. Director, Nuclear Research and Development Division, DOE-ORO
# 87. G. Nugent, Burns and Roe, Inc., 185 Crossways Park Drive, Wood-
i bury 3 NY 11797
' 88. R. A. Edelman, Burns and Roe, Inc., 426 Kinderkamack Road, Ora-
dell, NJ 07649

SR
144

........

89. W. R. Harris, Rand Corporation, 1700 Main Street, Santa Monica, s
CA 90406

96. S. Jaye, S. M. Stoller Corp., Suite 815, Colorado Building,
Boulder, CO 80302

91. W. Lipimski, Argonne National Labeoratory, 9700 South Cass Ave.,
Argonne, IL 60439

92. H. G. MacPherscn, Institute for Energy Analysis, Oak Ridge
Associated Universities, Oak Ridge, TN 37830

93. R. Omberg, Hanford Engineering Development Laboratory, P.0. Box
1970, Richland, WASH 99352

94. E. Straker, Science Applications, 8400 Westpark Drive, McLean,
VA 22101

95. A. M. Weinberg, Institute for Energy Analysis, Oak Ridge
Associated Universities, Oak Ridge, TN 37830

96. B. A. Pasternak, Booz, Allen Applied Research, 4330 East-West
Highway, Bethesda, MD 20014

97. M. C. J. Carlson, Manager, Alternative Fuel Cycle Evaluation Pro-
gram, Hanford Engineering Development Laboratory, P.0. Box 1970,
Richland, WA 99352

98-202. For distribution as shown in TID-4500 under category UC-76,

Mclten-Salt Reactoxr Technology

lmi

*4).S. GOVERNMENT PRINTING OFFICE: 1979-748-188/4235