NAT_MSRintro.txt

'MOLTEN-SALT REACTORS-HISTORY,

STATUS, AND POTENTIAL

M. W. ROSENTHAL, P. R. KASTEN, and R. B. BRIGGS
Oak Ridge National Laboratory, Oak Ridge, Tennessee 37830

Received August 4, 1969
Revised October 10, 1969

 

Molten-salt breeder veactors (MSBR’s) ave
being developed by the Oak Ridge National Lab-
ovatory for genevating low-cost powevr while
extending the nation’s vesources of fissionable
fuel. The fluid fuel in these reactors, consisting
of UF, and ThF, dissolved in fluorides of beryl-
lium and lithium, is civculated through a veactor

cove moderated by graphite. Technology develop-
ments over the past 20 years have culminated in
the successful opervation of the 8-MW(th) Molten-
Salt Reactor Experiment (MSRE), and have in-
dicated that operation with a wmolten fuel is
practical, that the salt is stable undevr reactor
conditions, and that corrosion is very low. Pro-
cessing of the MSRE fuel has demonstrated the
MSR processing associated with high-pevformance
converters. New fuel processing methods undey
development should permit MSR’s to opervale as
economical breeders. These features, combined
with high thermal efficiency (44%) and low primary
system pressuvre, give MSR converters and
breeders potentially favorable economic, fuel
utilization, and safety chavacteristics. Fuvther,
these reactors can be initially fueled with U,
2850, or plutonium. The construction cost of an
MSBR power plant is estimated to be aboul the
same as that of light-water reactors. This could
lend to power costs ~0.5 to 1.0 mill/kWh less than
those for light-water reactors. Achievement of

“economic molten-salt breeder reactors requives
the construction and operation of several reactors
of increasing size and their associated processing
plants.

 

THE HISTORY OF MOLTEN-SALT REACTORS

Investigation of molten-salt reactors started in
the late 1940’s as part of the United States’ pro-

NUCLEAR APPLICATIONS & TECHNOLOGY VOL. 8

 

KEYWORDS: molten-salt re-
actors, breeder reactors, de-
sign, operation, performance,
economics, power reactors

gram to develop a nuclear powered airplane. A
liquid fuel appeared to offer several advantages,
so experiments to establish the feasibility of
molten-salt fuels were begun in 1947 on ‘‘the
initiative of V. P. Calkins, Kermit Anderson, and
E. S. Bettis. At the enthusiastic urging of Bettis
and on the recommendation of W. R. Grimes,
R. C. Briant adopted molten fluoride salts in 1950
as the main line effort of the Oak Ridge National
Laboratory’s Aircraft Nuclear Propulsion’ pro-
gram.’’ The fluorides appeared particularly ap-
propriate because they have high solubility for
uranium, are among the most stable of chemical
compounds, have very low vapor pressure even at
red heat, have reasonably good heat transfer
properties, are not damaged by radiation, do not
react violently with air or water, and are inert to
some common structural metals.

A small reactor, the Aircraft Reactor Experi-
ment, was built at Oak Ridge to investigate the use
of molten fluoride fuels for aircraft propulsion

" reactors and particularly to study the nuclear

stability of the circulating fuel system. The ARE
fuel salt was a mixture of NaF, ZrF,, and UF4, the
moderator was BeO, and all the piping was In-
conel. In 1954 the ARE was operated successfully
for 9 days at steady-state outlet temperatures
ranging up to 1580°F and at powers up to 2.5 MW
(th). No mechanical or chemical problems were
encountered, and the reactor was found to be
stable and self-regulating.®

That molten-salt reactors might be attractive
for civilian power applications was recognized
from the beginning of the ANP program, and in
1956 H. G. MacPherson formed a group to study
the technical characteristics, nuclear per-
formance, and economics of molten-salt convert-
ers and breeders.® After considering a number of
concepts over a period of several years, Mac-
Pherson and his associates concluded that graph-
ite-moderated thermal reactors operating on a

FEBRUARY 1970 ' 107
Rosenthal et al.

thorium fuel cycle would be the best molten-salt
systems for producing economic power. The tho-
rium fuel cycle with recycle of ***U was found to
give better performance in a molten-salt thermal
reactor than a uranium fuel cycle in which 2*%U is
the fertile material and plutonium is produced and
recycled. Homogeneous reactors in which the
entire core is liquid salt were rejected because
the limited moderation by the salt constituents did
not appear to make as good a thermal reactor as
one moderated by graphite, and intermediate spec-
trum reactors did not appear to have high enough
breeding ratios to compensate for their higher
inventory of fuel. Studies of fast spectrum molten-
salt reactors®® indicated that good breeding ratios
could be obtained, but very high power densities
would be required to avoid excessive fissile in-
ventories. Adequate power densities appeared dif-
ficult to achieve without going to novel and
untested heat removal methods.

Twotypes of graphite-moderated reactors were
considered by MacPherson’s group—single-fluid
reactors in which thorium and uranium are con-
tained in the same salt, and two-fluid reactors in
which a fertile salt containing thorium is kept
separate from the fissile salt which contains ur-
anium. The two-fluid reactor had the advantage
that it would operate as a breeder; however, the
single-fluid reactor appeared simpler and seemed
to offer low power costs, even though the breeding
ratio would be below 1.0 using the technology of
that time. The fluoride volatility process,® which
could remove uranium from fluoride salts, had
already been demonstrated in the recovery of
uranium from the ARE fuel and thus was available
for partial processing of salts from either type of
reactor.

The results of the ORNL studies were con-
-sidered by a U.S. Atomic Energy Commission task
force that made a comparative evaluation of fluid-
fuel reactors early in 1959. One conclusion of the
task force’ was that the molten-salt reactor, al-
though limited in potential breeding gain, had “‘the
highest probability of achieving technical feasi-
bility.”’

By 1960, more complete conceptual designs of
molten-salt reactors had emerged. Although em-
phasis was placed on the two-fluid concept be-
cause of its better nuclear performance,® the
single-fluid reactor was also studied.” ORNL
concluded that either route would lead to low-
power-cost reactors, and that proceeding to the
breeder either directly or via the converter would
achieve reactors with good fuel conservation char-
acteristics. Since many of the features of civilian
power reactors would differ from those of the
ARE, and the ARE had been operated only a short
period, another reactor experiment was needed to

108 NUCLEAR APPLICATIONS & TECHNOLOGY

MSR—HISTORY, STATUS, AND POTENTIAL

investigate some of the technology for power
reactors.

The design of the Molten-Salt Reactor Experi-
ment was begun in 1960. A single-fluid reactor
was selected that in its engineering features re-
sembled a converter, but the fuel salt did not
contain thorium and thus was similar to the fuel
salt for a two-fluid breeder. The MSRE fuel salt
1s a mixture of uranium, lithium-7, beryllium, and
zirconium fluorides. Unclad graphite serves as
the moderator (the salt does not wet graphite and
will not penetrate into its pores if the pore sizes
are small). All other parts of the system that
contact salt are made from the nickel-base alloy,
INOR-8 (also called Hastelloy-N), which was spe-
cially developed in the aircraft program for use
with molten fluorides. The maximum power is
~8000 kW, and the heat is rejected to the atmos-
phere.

Construction of the MSRE began in 1962, and
the reactor was first critical in 1965. Sustained
operation at full power began in December 1966.
Successful completion of a six-month run in
March of 1968 brought to a close the first phase
of operation during which the initial objectives
were achieved. The molten fluoride fuel was used
for many months at temperatures > 1200°F without
corrosive attack on the metal and graphite parts
of the system. The reactor equipment operated
reliably and the radioactive liquids and gases
were contained safely. The fuel was completely
stable. Xenon was removed rapidly from the salt.
When necessary, radioactive equipment was re-
paired or replaced in reasonable time and without
overexposing maintenance personnel.

The second phase of MSRE operation began in
August 1968 when a small processing facility
attached to the reactor was used to remove the
original uranium by treating the fuel salt with
fluorine gas. A charge of ***U fuel was added to
the same carrier salt, and on October 2 the MSRE
was made critical on **U. Six days later the
power was taken to 100 kW by Glenn T. Seaborg,
Chairman of the U. S. Atomic Energy Commission,
lé)éinging to power the first reactor to operate on

U

During the years when the MSRE was being
built and brought into operation, most of the de-
velopment work on molten-salt reactors was in
support of the MSRE. However, basic chemistry
studies of molten fluoride salts continued through-
out this period. One discovery during this time
was that the lithium fluoride and beryllium flu-
oride in a fuel salt can be separated from rare
earths by vacuum distillation at temperatures
near 1000°C. This was a significant discovery,
since it provided an inexpensive, on-site method
for recovering these valuable materials. As a

VOL. 8 FEBRUARY 1970
Rosenthal et al.

consequence, the study effort looking at future
reactors focused on a two-fluid breeder in which
the fuel salt would be fluorinated to recover the
uranium and distilled to separate the carrier salt
from fission products. The blanket salt would be
processed by fluorination alone, since few fis-
sion products would be generated in the blanket if
the uranium concentration were kept low. Graph-
ite tubes would be used in the core to keep the fuel
and fertile streams from mixing.

Analyses of these two-fluid systems showed
that breeding ratios in the range of 1.07 to 1.08
could be obtained, which, along with low fuel in-
ventories, would lead to good fuel utilization. In
addition, the fuel cycle cost appeared to be quite
low. Consequently, the development effort for
future reactors was aimed mainly at the features
of two-fluid breeders. A review of the technology
associated with such reactors was published in
1967.'! The major disadvantage of this two-fluid
system was recognized as being that the graphite
had to serve as a piping material in the core
where it was exposed to very high neutron fluxes.

In late 1967, new experimental information and
an advance in core design caused the molten-salt
program at ORNL to change from the two-fluid
breeder to a single-fluid breeder. Part of the
information influencing this change concerned the
behavior of graphite at higher radiation exposures
than had been achieved previously, and the other
part related to a development in chemical proces-
sing.

The irradiation data'®>>® showed that the Kkind
of graphite planned for use in an MSBR changes
dimensions more rapidly than had been antici-
pated. This made it necessary to lower the core
power density for the graphite to have an accept-
able service life, and to plan on replacement of the
core at fairly frequent intervals.” Moreover,
complexities in the assembly of the core seemed
to require that the entire core and reactor vessel
be replaced whenever a graphite element reached
its radiation limit or developed a leak. Under
such circumstances, many years of operation of a
prototype reactor would be required to prove
convincingly that the two-fluid core is practicable.

At about the time the problems associated with
long graphite exposure became evident,a chemical
processing development occurred'® that greatly
improved the prospect for a single-fluid breeder.
To obtain good breeding performance in a single-
fluid reactor, ***Pa (27.4-day half-life) must be
held up outside the core until it decays to **°U.
The processing development that showed promise
of accomplishing this was a laboratory demon-
stration of the chemical steps in a liquid-liquid
extraction process for removing protactinium and
uranium from molten fluoride salts. The tech-

NUCLEAR APPLICATIONS & TECHNOLOGY VOL. 8

MSR—HISTORY, STATUS, AND POTENTIAL

nique is to exchange thorium and lithium dissolved
in molten bismuth for the constituents to be re-
moved from the salt. The process has similari-
ties to one being developed at Argonne National
Laboratory for processing fast reactor fuels and
involves technology explored at Brookhaven Na-
tional Laboratory for removing fission products
from a liquid bismuth fuel. Additional data have
confirmed the early results and have shown that
the uranium can be selectively stripped from the
salt into bismuth, the protactinium can be trapped
in salt in a decay tank, and the uranium can be
transferred back to the salt by electrolysis for
return to the reactor. Calculations indicated that
the extraction and electrolysis could be carried
out rapidly and continuously, and that the process
equipment would be relatively small.

Laboratory experiments also indicated that
rare earths might be extracted from salt from
which the uranium had previously been removed.
Unfortunately, the chemical potentials that de-
termine which constituents transfer to the bis-
muth are relatively close for thorium and the
higher cross-section rare earths, and the separa-
tion is more difficult than it is for uranium and
protactinium. The extraction process may still be
workable, however, because with no processing
the rare earths have a much smaller effect on
breeding ratio than does protactinium and so need
be removed only relatively slowly (50- to 80-day
cycle time). Several other processes for rare
earth removal are under investigation, and
whether liquid metal extraction will be the most
attractive is still uncertain.

The advance in core design that was important
in the switch to the single-fluid breeder was the
recognition that a fertile “blanket’ can be
achieved with a salt that contains uranium as well
as thorium. The blanket is obtained by increasing
the volume fraction of salt and reducing the
volume fraction of graphite in the outer part of the
reactor. This makes the outer region under-
moderated and increases the capture of neutrons
there by the thorium. With this arrangement,
most of the neutrons are generated at some dis-
tance from the reactor boundary, and captures in
the blanket reduce the neutron leakage to an
acceptable level. H. G. MacPherson had proposed
this scheme several years before, but it was only
studied thoroughly after the discovery that pro-
tactinium could be removed from the salt. Opti-
mization calculations then showed that proper
selection of dimensions and volume fractions
could keep the inventory of uranium in the outer
region from being excessive, while producing a
blanket region.

As a result of the developments cited above,
design studies of single-fluid breeders were pur-

FEBRUARY 1970 109
Rosenthal et al.

sued. The studies indicated that the fuel utiliza-
tionin single-fluid, two-region molten-salt reactors
can be almost as good as in two-fluid reactors,
and with the present limitations on graphite life,
the economics probably can be better. Conse-
quently, in 1968 ORNL’s Molten-Salt Reactor
Program was directed toward the development of
a single-fluid breeder reactor. The immediate
effect was to bring the reactor itself to a more
advanced state, since in many respects the single-
fluid concept is a scaled-up MSRE. At the same
time, however, a number of new features were
introduced into the processing concept by adoption
of the extraction processes.

The characteristics of single-fluid reactors
and the work that is required to develop commer-
cial-size plants are discussed in the sections
which follow.

ECONOMIC AND NUCLEAR PERFORMANCE
OF MOLTEN-SALT BREEDER REACTORS

Molten-salt breeder reactors offer attractions
as power producers because of potentially favor-
able economic, fuel utilization, and safety charac-

Fig, 1.

110 NUCLEAR APPLICATIONS & TECHNOLOGY

MSR—HISTORY, STATUS, AND POTENTIAL

teristics. The avoidance of fuel fabrication, the
ease of processing, and the low fissile inventory
should result in low fuel cycle costs. Capital
costs benefit from high thermal efficiency, low
primary system pressure, and low pumping re-
quirements. Some inherent safety features assist
the designer in providing a safe plant.

The above characteristics can more than com-
pensate for the extra costs of handling radioactive
fluids and maintaining radioactive equipment, and
that MSBR’s can have low power costs. Since the
reprocessing can be done in a small on-site plant,
the attainment of low costs does not await the
development of a large fuel reprocessing industry.
The use of bare graphite in the core and the rapid
removal of fission products, when combined with
low fuel inventory, result in good fuel utilization.
In addition, molten-salt reactors can be started up
economically on **U, *°U, or plutonium, and
hence, can use the fuel that gives the lowest cost
at a given time.

A simplified flow diagram of a single-fluid
two-region MSBR having these characteristics is
shown in Fig. 1, and some of the features of such
reactors are given in Table I. An example of the

| COOLANTSALT

 

Simplified flow diagram of a single-fluid two-region molten-salt breeder reactor.,

VOL. 8 FEBRUARY 1970
 

Rosenthal et al.

TABLE 1

Characteristics of One-Fluid, Two-Region
Molten-Salt Breeder Reactors

 

72 "LiF, 16 BeF.,
12 ThF4, 0.3 UF,

Melting point, 930°F

Fuel-fertile salt, mole %

Moderator Graphite (bare)

Core, 13; blanket, 40
Inlet, 1050; outlet, 1300
1000-2000

3500 psia, 1000°F,
44% net cycle efficiency

Salt volume fractions, %
Core temperatures, °F
Reactor power, MW(e)

Steam system

Breeding ratio 1.05-1.07

Specific fissile fuel inventory,? 1.0-1.5
kg /MW(e)

Doubling time (compound interest),?| 15-25
year

Fuel cycle cost,? mills/kWh 0.6-0.7

 

 

 

(including graphite replacement)

 

aThe lower values are associated with the higher reactor powers.

fuel cycle cost breakdown for a 1000 MW (e) MSBR
is shown in Table II, to permit comparison with
the fuel cycle costs of solid fuel reactors, the
capital and operating costs of the processing plant
and the cost of replacing the graphite have been
included in the table.

We have also estimated the capital cost of
1000 MW(e) molten-salt reactor plants and com-
pared them with cost estimates of light-water
reactors made on the same bases. The estimates
assume that a molten-salt industry has advanced
to the point where development costs have largely
been absorbed, and the manufacture of materials
and the construction and licensing of plants have
become routine. Under these conditions we find
that the capital costs of molten-salt and light-
water reactors should be about the same. While
the molten-salt reactor has some features which
add cost, such as the off-gas system and the

TABLE II

Fuel Cycle Cost Breakdown for a Single-Fluid
Molten-Salt Breeder Reactor*

 

 

Mills /kWh
Fissile inventory 0.26
Thorium inventory 0.01
Carrier salt inventory 0.04
Thorium and carrier salt makeup 0.05
Processing plant fixed charges and operating cost 0.30
Credit for sale of bred material -0.09
Graphite replacement cost (4-year interval) 0.10
Net fuel cycle cost 0.67

 

 

 

*At 10% per year inventory charge on material, 13.7% })er
year fixed charge rate on processing plant, $13/g 2*°U,
$11.2/g 2°°U, $12/kg ThO:, $120/kg 'Li, $26/kg carrier
salt (including "Li). ‘

NUCLEAR APPLICATIONS & TECHNOLOGY VOL. 8

MSR—HISTORY, STATUS, AND POTENTIAL

equipment provided for remote maintenance of the
radioactive systems, it also has features that re-
duce costs; for example use of the high-tempera-
ture, high-efficiency power cycle gives a savings
of ~$17 000 000 in the turbine-generator plant.

An important factor to include when estimating
the performance of reactors is the downtime re-
quired for maintenance and refueling. Because of
the fission product activity, longer times will be
required for maintenance of some parts of molten-
salt reactors than for comparable parts of solid-
fuel reactors. There will, however, be no loss in
operating time for refueling a molten-salt system,
so the overall plant factor ought to be at least as
high as for other reactors. A molten-salt reactor
can operate for an extended period without pro-
cessing, so the reactor need not be shut down
while work is being done on the processing plant.

Considering all these factors, the power cost
from a molten-salt breeder reactor can be 0.5 to
1 mill/kWh below that of a light-water reactor
with present uranium ore costs. The differential
will increase if ore costs go up, since power costs
for molten-salt reactors are considerably less
sensitive to the cost of uranium.

Many analyses have shown that the development
of breeder reactors is necessary to avoid rapid
depletion of our uranium reserves. Similar an-
alyses have been performed here, in which the
ability of molten-salt breeders to fill this need
has been compared with the abilities of other
kinds of breeders, and Fig. 2 gives some of the
results obtained. In the analyses, the fuel utiliza-
tion characteristics used for light-water converter
reactors were based on estimates for future light-
water reactors,’ and varied with time. A single
set of fuel utilization characteristics was used
for the light-water breeder,” while several com-
binations of breeding ratio and specific inventory
were used for fast and molten-salt breeders. The
curves for the breeders are based on building only
light-water converters for a specified time, and
then phasing in a particular breeder such that
after 16 years, all new reactors are breeders. In
these studies, the introduction date for breeders
was assumed to be 1976 for the light-water
breeder and 1982 for the fast and molten-salt
breeders.

The curves in Fig. 2 indicate that the breeding
gain and doubling time in themselves are not
adequate measures of the ability of a breeder
reactor to limit the amount of uranium ore that
must be mined to fuel a large nuclear power
economy. The fissile inventory is also important
and a low specific inventory is particularly im-
portant in a rapidly growing nuclear economy. It
is their low specific inventory that makes it pos-
sible for molten-salt thermal breeder reactors to

FEBRUARY 1970 111
Rosenthal et al. MSR—HISTORY,

STATUS, AND POTENTIAL

 

 

  
 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

T z3T
o, ONLY LIGHT WATER W =
1.0l> CONVERTER REACTORS = 293 =
O3 BUILT THROUGH - =
S ENTIRE PERIOD 2 =245 2
D | I - =« o
3 5 = o QL w
> 8 [ S Bo w
—_ — w O a X %
o (Vo N U
oo oo
&5=" £8 3-0ITLIGHT WATER BREEDERS F—=FBR 15 5 1.35
=n 3 START 1976. ONLY /
35 oo | |LIGHT WATER BREEDERS J| 7
& T2 2-57TBUILT AFTER 1992 ~f/ 17
/
55 & N/ _— REFERENCE MSBR DESIGN
zZ» 25 2.0 /) [— MSR 20 1.3 1.07
S« 5% - r Z3——_FBR 10 3 1.3]
we oF s I -1 MR 15 1.0 1.07
=5 5§ |2= ,;:,/ L——T MSR 10 0.7 1.07
< r— v wv
== 20 G2 A A"T———F——FR 5 3 1.63
3=~ T /4/25> BREEDERS START 1982. ONLY
~c0 4 BREEDERS BUILT AFTER 1998.
O M
0.5 Te— D
Wb 0
00
0 ! 1
1970 1980 1990 2000 2010 2020 2030

YEARS

Fig, 2,

compete with fast breeder reactors in limiting the
resource requirements.

To demonstrate further the importance of in-
ventory, the peak ore requirements obtained from
curves like those shown in Fig. 2 have been
cross-plotted in Fig. 3 as a function of specific
inventory and doubling time. A point to note is
that values plotted in Fig. 3 depend on the assump-
tions made about the growth of nuclear power. It
was assumed that the US nuclear power capability
would reach 140 000 MW(e) in 1980, 930 000 MW (e)
in 2000, and expand at a rate of 100 000 MW(e) per
year afterwards. If nuclear power grew expo-
nentially after 2000, the ore requirements would
not reach a peak unless the doubling time of the
reactor considered was less than the doubling
time of the long term growth of the nuclear power
economy.

MOLTEN-SALT CONVERTER REACTORS

Nearly all the present ORNL effort on molten-
salt reactors is devoted to developing a breeder,
since it has the lowest potential power cost as
well as good fuel utilization characteristics.
Achieving good breeding performance is dependent
upon the successful development of advanced pro-

112

NUCLEAR APPLICATIONS & TECHNOLOGY

Uranium ore requirements for a growing nuclear power economy,

cessing schemes, but low power costs can be ob-
tained with smaller breeding ratios using
processing already demonstrated at the MSRE,
namely, xenon stripping and fluoride volatility.
The MSRE processing plant, if run in a semicon-
tinuous fashion, is large enough to process a
1000 MW(e) reactor on a three-year cycle. If
used with a single-fluid reactor to recover the
uranium, and with the fission products and other
constituents of the salt being discarded, a con-
verter having a breeding ratio between 0.8 and 0.9
would be obtained.”® The accompanying fuel cycle
cost, including graphite replacement cost and the
fixed charges and operating cost of the processing
plant, would be ~ 0.8 mills/kWh. Thus, a reactor
which is a scaled-up, higher power- density MSRE,
using an MSRE-type processing plant, is capable
of producing power at attractive costs. With the
addition of extractive processing, such a reactor,
Oor one very similar, becomes a high-performance
breeder.

SAFETY

Safety is an important factor to be considered
in evaluating civilian power reactors, but it cannot
be divorced from economics. Most reactors can

VOL. 8 FEBRUARY 1970
Rosenthal et al.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

7
6
"o DOUBLING
< TIME
= 5 35 Year /
)
. | /
: 4 v | //
é 20 Year /
L ; /15Y}ar,/
& _— —
o /// 10 qur‘/-
A S Year
Ll
0
0 ] 2 3 4 5 6 7
SPECIFIC INVENTORY [kg/MW(e)]
Fig. 3. Maximum uranium ore requirements.

probably be made adequately safe by conservative
design features and redundant engineered safe-
guards that add to the cost of power. In this
regard, molten-salt reactors have some inherent
features that may result in cost savings relative
to other systems: the primary system operates at
low pressure with fuel salt that is more than
1000°F below the boiling point, some fission prod-
ucts are removed from the reactor continuously,
iodine and strontium form stable compounds in the
salt, and the salts do not react rapidly with water
or air. The need for excess reactivity is reduced
by the continuous fuel processing, and a prompt
negative temperature coefficient is associated
with heating of the fuel salt.

A safety disadvantage of molten-salt reactors
is the accumulation of fission products in the pri-
mary system, the off-gas system, the fuel storage
tanks, and the processing plant, which requires
provisions to ensure that the fission products will
be contained and their decay heat will be removed
under all conceivable circumstances. Partially
offsetting this is the ability to drain the fuel into
a tank having always ready, redundant cooling
systems.

Preliminary evaluation of conceptual designs
suggests that the safety characteristics of molten-
salt reactors will be a net asset, but a thorough
evaluation of fairly detailed designs will be re-
quired before the economic importance of this can
be assessed.

NUCLEAR APPLICATIONS & TECHNOLOGY VOL. 8

FEBRUARY 1970

MSR—HISTORY, STATUS, AND POTENTIAL
REMAINING TECHNOLOGY DEVELOPMENTS

While molten-salt reactors offer many attrac-
tive features, there is much to be done before
full-scale economic molten-salt breeders can be
built. What is seen as the remaining primary re-
search and development in different areas is
summarized below.

Physics

No crucial physics problems are seen in the
development of molten-salt reactors, but in view
of the relatively small breeding gain of an MSBR,
particular care must be taken to ensure that the
nuclear calculations are accurate. The combined
uncertainty in breeding ratio due to uncertainties
in cross sections of all the reactor materials is
~+0.016, of which ~ 0.012 results from uncertainty
in the n value of **°U.

To reduce the uncertainty, the capture-to-
fission cross-section ratio of **°U is being de-
termined from changes in isotopic ratios in the
MSRE fuel. Additional measurements of the ab-
sorption cross sections of the salt constituent—
Li, Be, and F—could reduce the uncertainty
further. Lattice experiments with concentrations
and geometries appropriate for the core of an
MSBR should be made in facilities such as the
Physical Constants Tests Reactor and the High
Temperature Lattice Test Reactor at Battelle
Northwest Laboratories, with emphasis on verify-
ing the temperature coefficients of reactivity,
including Doppler coefficients. These experiments
would also provide an integral check on the
adequacy of cross-section values and on compu-
tational models.

Chemistry

An extensive program of molten-salt chemical
research over the past 20 years has established a
sound basis for understanding the chemistry of the
pertinent fluoride salts. Additional work is de-
sirable, however, in several areas.- A better
understanding is needed of the behavior of the
noble metal fission products in MSBR systems.
Alternative processes for rare earth removal
should be investigated, and chemical information
on the protactinium removal process needs to be
extended to a wider range of conditions. The
phase behavior of PuF; in molten salts ought to be
studied more thoroughly to provide a firmer
chemical basis for the use of plutonium. Further
measurements are needed of the physical and
chemical properties of molten salts, and particu-
larly of the sodium fluoride/sodium fluoroborate
salt mixtures which are the presently proposed
secondary coolants for MSBR’s.

113
Rosenthal et al.

Analytical Chemistry

Adequate control of the chemistry of the salts
in the reactor systems depends on development of
rapid, reliable, and economical analytical tech-
niques for the precise determination of uranium,
thorium, and other constituents of the fuel salt,
the detection of bismuth in fuel salt, the measure-
ment of the concentration of trivalent uranium in
fuel salt, and the analysis of constituents of fuel
and coolant system cover gases. At present, such
measurements are obtained through sampling and
subsequent analysis, but promising electrochem-
ical and spectrophotometric methods for con-
tinuous, or almost continuous, analyses are under
development. These methods need to be developed
to the point where dependable instruments can be
demonstrated.

Fuel Processing

Obtaining good breeding in a single-fluid reac-
tor requires rapid on-site reprocessing to keep
protactinium out of the high neutron flux as well
as to remove fission products. Investigation of
most of the steps needed to effect rapid repro-
cessing has not proceeded past preliminary
laboratory tests. Thus, development of the liquid-
bismuth extraction system will require an ex-
tensive program of engineering experiments.
Materials problems, including the inherently dif-
ficult corrosion problem of the electrolyzer for
transferring uranium from bismuth to salt and
thorium and lithium from salt to bismuth, will
have to be overcome.

Nonradioactive engineering experiments should
be sufficient to develop and prove out components
and to demonstrate the practicality of salt-bismuth
contractors and of the electrolyzer. Removal of
rare earths also can be tested using nonradio-
active material, but experiments using significant
concentrations of protactinium will have to be
performed in facilities that have provisions for
high a activity. Demonstration of a processing
system with high fission-product heating rates
will only be possible on a reactor.

Although a workable fluorination process has
already been demonstrated, advantages would ac-
crue from using a layer of frozen salt to protect
metal surfaces from corrosion. Since fluorination
can have an auxiliary role in single-fluid breeder
processing as well as the main role in processing
a converter, further work should be done on this.
Additional work is needed on vacuum distillation,
which appears applicable to recovery of lithium
and beryllium, and possibly thorium, from waste
streams.

114 NUCLEAR APPLICATIONS & TECHNOLOGY

MSR—HISTORY, STATUS, AND POTENTIAL

Materials

Graphites that would be acceptable for the
cores of early molten-salt reactors can probably
be obtained commercially in the near future, but
production techniques must be scaled up, and
commercially produced materials must be tested
for reactor use. Damage from high-energy neu-
trons eventually leads to deterioration of graphite
at temperatures of interest to molten-salt reac-
tors. However, there is some experimental evi-
dence that the resistance of graphite to radiation
damage can be significantly increased, and im-
proved grades need to be developed and tested.
Graphite development work centers on use of the
High-Flux Isotope Reactor, in which a dose of
4 X 10** nvt (E > 50 keV) can be obtained in one
year.

Helium sparging will remove '**Xe from molten
salts, and reduction of xenon poisoning by a factor
of 6 has been demonstrated in the MSRE. To
achieve the factor of 10 reduction desired for a
breeder, however, will require that the perme-
ability of the graphite, as measured with helium,
be reduced to ~107° cm?/sec. Pyrolytic carbon
impregnation of the surface has been shown to be
capable of producing graphite of such low perme-
ability, but much more testing under irradiation is
required, and the process must be scaled up for
use with reactor-size graphite bars.

The Hastelloy-N used in the past for containing
molten salts is embrittled by helium produced in
the metal when it is irradiated. Although the
present material has been satisfactory for the
MSRE, an improved alloy is needed for power
reactors. A modified form of Hastelloy-N con-
taining small amounts of titanium or hafnium
appears to meet this requirement. Further testing
under a variety of irradiation conditions must be
completed to demonstrate fully that the desired
improvement has been accomplished. Large heats
must be obtained, and the modified material must
be subjected to enough testing to have it approved
for pressure vessel use by the ASME. The
changes in composition from standard Hastelloy-N
are not great, however, and full requalifications is
probably not required. For the longer term, creep
rupture tests at low stress conditions and extend-
ing over many years should be made to verify the
extrapolation of data from the requalification
tests. Resistance of the modified alloy to corro-
sion by fuel and coolants salts—particularly the
fluoroborate—must be demonstrated at all reactor
operating conditions.

In a reductive extraction fuel processing sys-
tem, materials will be exposed to both salt and
bismuth at temperatures between 500 and 700°C.
Candidate materials will have to be tested, and

VOL. 8 FEBRUARY 1970
Rosenthal et al.

methods of protection, such as deposition of films
or frozen salt layers on the wall, may have to be
developed.

Molten-Salt lrradiation Experiments

MSRE operation has shown the compatibility of
graphite with fissioning fuel at low power densi-
ties and low burnup, and a variety of in-pile
capsules and loops have provided additional infor-
mation. However, the experience to date has not
included exposures at the power densities of
breeder reactors. In-pile loop experiments thus
need to be performed using appropriate materials
at the peak power densities of high-performance
breeders.

Engineering Development

Development of large components is probably
the most costly of all the technology needs of the
MSR. Most of the components built so far have
been of MSRE size or smaller. While the MSRE
has demonstrated the workability of small com-
ponents, much larger equipment will be needed for
MSBR’s.

Vertical shaft pumps with overhung impellers
similar to the MSRE pumps would be used, and the
basi¢ requirement is to increase the size. Heat
exchangers for MSBR’s must be of very high
performance to obtain a low fuel salt inventory
and must be capable of replacement. The develop-
ment of steam generators heated by molten salt
involves an area in which there has been almost
no experience, and the high melting point of the
salt will probably cause the design to be consider-
ably different from that of other steam generators
built to date. The control rods and drives for
large MSBR’s will be different from those on the
MSRE, because air cooling will not suffice at the
higher power densities. Rods and drives that per-
mit insertion of control elements directly into the
fuel salt will be needed.

Xenon is effectively stripped from the fuel salt
in the MSRE pump bowl, but for a large, high-
performance MSBR, it will be necessary to inject
bubbles into the salt under controlled conditions
and to remove them effectively. Reliable methods
for disposing of afterheat in the fuel salt after
reactor shutdown must be demonstrated for a
variety of reactor conditions.

The ability to maintain the reactor primary
system and the processing system of an MSBR
power plant must be assured. New techniques
need to be developed that go beyond those used on
the MSRE. The ability to cut and weld piping
remotely would be valuable, and existing automatic
welding equipment should be adapted for this use
and demonstrated.

NUCLEAR APPLICATIONS & TECHNOLOGY VOL. 8

MSR—HISTORY, STATUS, AND POTENTIAL

A number of test facilities will be needed for
use in the development and testing of components
and systems.

These and other developments must be accom-
plished on an appropriate scale before the next
molten-salt reactor is completed. Where they are
successful, only scale-up will be required for the
larger reactors that follow.

Reactor Development

Achievement of economic molten-salt breeder
reactors requires the construction and operation
of several reactors of increasing size and their
associated processing plants. Construction of
these reactors will establish the industrial capa-
bility to manufacture molten-salt reactors and
will test and demonstrate the technology which
leads toward dependable reactor operation and low
power costs. The next reactor could be a molten-
salt breeder experiment which has all the features
and systems of a full-scale power breeder and
operates at power densities and other conditions
extreme enough to provide a thorough test of
future reactor conditions. Alternatively or con-
currently, a converter reactor could be built which
uses technology largely demonstrated by the
MSRE except for equipment scale-up to higher
performance and larger size.

CONCLUSIONS

The objective of developing breeder reactors is
to obtain a source of low-cost energy for our-
selves and for future generations. Three events
of the past year make it seem likely that molten-
salt reactors can accomplish this:

1. the very successful operation of the MSRE
on both **°U and ***U

2. the processing of a complete charge of fuel
at the MSRE

3. the advances in chemical processing and
core arrangement that make it possible to
design a single-fluid breeder.

The first event gives assurance that molten salt
technology can be used dependably for power
reactors. The second event means that a con-
verter reactor having a breeding ratio of 0.8 to
0.9 and attractively low power costs can be based
on a demonstrated processing system. And the
third means that molten-salt breeders, as well as
converters, can essentially be scaled-up versions
of the MSRE.

While adoption of the single-fluid concept
brings the breeder reactor nearer to fruition, it
also increases the development needed for the

FEBRUARY 1970 115
Rosenthal et al.

breeder processing plant. Scale-up towards com-
mercial-size breeder reactors, however, need not
await the demonstration of breeder processing.
Since the converter and breeder are basically the
same except for the processing plant, the scale-up
can begin by building converters, with expectation
that they will evolve into single-fluid breeders as
the processing is improved. This flexibility—
generally a result of using a fluid fuel—extends
also to the freedom to start up the reactors with
3%y, ?*U, or Pu, depending on which is most
economic at the time.

Much remains to be done before economic
molten-salt breeder reactors can be achieved, but
the basic virtues of the molten-salt system and
the success of the MSRE offer promise that the
breeder development program will meet with
early success.

ACKNOWLEDGMENTS

This research was sponsored by the U.S. Atomic
Energy Commission under contract with the Union
Carbide Corporation.

REFERENCES

1. H. G. MACPHERSON,
Proc. Intern. Conf. Constructive Uses At. Energy,
Washington, DC, November 1968, American Nuclear
Society (March 1969).

‘““Molten-Salt Reactors,”’

2. ‘‘The Aircraft Reactor Experiment,”” Nucl. Sci.
Eng., 2, 797 (1957).

3. J. A. LANE, H. G. MACPHERSON, and F. MASLAN,
Fluid Fuel Reactors, pp. 567 ff, Addison-Wesley, Read-
ing, Mass. (1958).

4. L. G. ALEXANDER et al., Oak Ridge National
Laboratory, Personal Communication (1962-1963).

o. P. R. KASTEN, ‘‘The MOSEL Reactor Concept,”’
Thivd U.N. Intern. Conf. Peaceful Uses of At. Energy,
Geneva, A/Conf 28/P/538 (1964).

6. G. I. CATHERS, ‘‘Uranium Recovery from Spent
Fuel by Dissolution in Fused Salt and Fluorination,’’
Nucl. Sci. Eng., 2, 768 (1957); W. H. CARR, ‘‘Volatility
Processing of the ARE Fuel,”” Chem. Eng. Symp. Sevies,
96, 28 (1960).

7. US-AEC TASK FORCE, ‘“‘Report of the Fluid Fuel
Reactors Task Force,”’ TID-8507 (February 1959).

8. L. G. ALEXANDER, W. L, CARTER, R. H. CHAP-
MAN, B. W. KINYON, J. W. MILLER, and R. VAN
WINKLE, ‘‘Thorium Breeder Reactor Evaluation—Part
I, Fuel Yield and Fuel Cycle Costs in Five Thermal
Breeders,”” ORNL-CF-61-3-9, Oak Ridge National Lab-
oratory (September 1961).

116

NUCLEAR APPLICATIONS & TECHNOLOGY

MSR—HISTORY, STATUS, AND POTENTIAL

9. L. G. ALEXANDER, W, L. CARTER, C. W. CRA-

VEN, D. B. JANNEY, T. W. KERLIN, and R. VAN
WINKLE, ‘‘Molten-Salt Converter Reactor Design
Study and Power Cost Estimated for a 1000 Mwe
Station,”” ORNL-TM-1060, Oak Ridge National Lab-
oratory (September 1965).

10. M. J. KELLY, ‘““Removal of Rare Earth Fission
Products from Molten-Salt Reactor Fuels by Distilla-
tion,”” Trans. Am. Nucl. Soc., 8, 170 (1965).

11. The status of molten-salt reactor development was
surveyed in a series of ORNL reports issued in 1967,
They are:

R. B. BRIGGS, ‘‘Summary of the Objectives, the
Design, and a Program of Development of Molten-Salt
Breeder Reactors,”” ORNL-TM-1851, Oak Ridge National
Laboratory (June 12, 1967).

W. R. GRIMES, ‘‘Chemical Research and Develop-
ment for Molten-Salt Breeder Reactors,’”’ ORNL-TM-
1853, Oak Ridge National Laboratory (June 1967).

H. E. McCOY and J. R. WEIR, ‘““Materials Develop-
ment for Molten Salt Breeder Reactors,’”’ ORNL-TM-
1854, Oak Ridge National Laboratory (June 1967).

DUNLAP SCOTT and A. G. GRINDELL, “Components
and Systems Development for Molten-Salt Breeder Re-
actors,”” ORNL-TM-1855, Oak Ridge National Labora-
tory (June 30, 1967).

A. M. PERRY, ‘““Physics Program for Molten Salt
Breeder Reactors,”” ORNL-TM-1857, Oak Ridge Na-
tional Laboratory (June 1967).

PAUL R. KASTEN, ‘‘Safety Program for Molten-Salt
Breeder Reactors,”” ORNL-TM-1858, Oak Ridge Na-
tional Laboratory (June 9, 1967).

J. R. TALLACKSON, R. L. MOORE, and S. J.
DITTO, ““Instrumentation and Controls Development for
Molten-Salt Breeder Reactors,”” ORNL-TM-1856, Oak
Ridge National Laboratory (May 22, 1967).

W. L. CARTER and M, E. WHATLEY, ‘‘Fuel and
Blanket Processing Development for Molten-Salt
Breeder Reactors,”” ORNL-TM-1852, Oak Ridge Na-
tional Laboratory (June 1967),

ROBERT BLUMBERG, ‘‘Maintenance Development
for Molten-Salt Breeder Reactors,”” ORNL-TM-1859,
Oak Ridge National Laboratory (June 1967).

12, R. W. HENSON, A. S. PERKS, and J. H. W. SIM-
MONS, “‘Lattice Parameter and Dimensional Changes in
Graphite Irradiated Between 300 and 1350°C,”” AERE R
5489, United Kingdom Atomic Energy Research Estab-
lishment, Harwell (June 1967).

13. J. W. HELM, ‘“Long Term Radiation Effects on
Graphite,”” Paper MI 77, 8th Biennial Conference on
Carbon, Buffalo, New York (June 1967).

VOL. 8 FEBRUARY 1970
Rosenthal et al.

14, P. R. KASTEN, E. S. BETTIS, W. H. COOK,
W. P. EATHERLY, D. K. HOLMES, R. J. KEDL,
C. R. KENNEDY, S. S. KIRSLIS, H. E. McCOY, A. M.
PERRY, R. C. ROBERTSON, D. SCOTT, and R. A.
STREHLOW, ‘Graphite Behavior and Its Effects on
MSBR Performance,”’ Nucl. Eng. Design, 9, 2 (Feb-
ruary 1969).

15, W. R. GRIMES, ‘‘Molten-Salt Reactor Chemistry,”
Nucl. Appl. Tech., 8,137 (1970).

16. JACKSON & MORELAND and S. M. STOELLER
ASSOCIATES, ‘“Current Status and Future Technical and

NUCLEAR APPLICATIONS & TECHNOLOGY VOL. 8

MSR—HISTORY, STATUS, AND POTENTIAL

Economic Potential of Light Water Reactors,”” WASH-
1082, U.S. Atomic Energy Commission (March 1968).

17. ‘“‘Performance Potential of Large Seed-Blanket
Breeder Reactors,”’ WAPD-LSBR-300, Bettis Atomic
Power Laboratory, Westinghouse Electric Corporation
(February 1965).

18. A, M. PERRY and H. F. BAUMAN, ‘Reactor
Physics and Fuel-Cycle Analyses,’’ Nucl. Appl. Tech.,
8, 208 (1970). -

FEBRUARY 1970 117